2006 - LIVRO - Epoxy Adhesive Formulations PDF
2006 - LIVRO - Epoxy Adhesive Formulations PDF
ADHESIVE
FORMULATIONS
This page intentionally left blank
EPOXY
ADHESIVE
FORMULATIONS
Edward M. Petrie
McGRAW-HILL
New York Chicago San Francisco Lisbon London Madrid
Mexico City Milan New Delhi San Juan Seoul
Singapore Sydney Toronto
Copyright © 2006 by The McGraw-Hill Companies, Inc. All rights reserved. Manufactured in the
United States of America. Except as permitted under the United States Copyright Act of 1976, no part
of this publication may be reproduced or distributed in any form or by any means, or stored in a data-
base or retrieval system, without the prior written permission of the publisher.
0-07-158908-2
The material in this eBook also appears in the print version of this title: 0-07-145544-2.
All trademarks are trademarks of their respective owners. Rather than put a trademark symbol after
every occurrence of a trademarked name, we use names in an editorial fashion only, and to the
benefit of the trademark owner, with no intention of infringement of the trademark. Where such
designations appear in this book, they have been printed with initial caps.
McGraw-Hill eBooks are available at special quantity discounts to use as premiums and sales
promotions, or for use in corporate training programs. For more information, please contact George
Hoare, Special Sales, at [email protected] or (212) 904-4069.
TERMS OF USE
This is a copyrighted work and The McGraw-Hill Companies, Inc. (“McGraw-Hill”) and its licensors
reserve all rights in and to the work. Use of this work is subject to these terms. Except as permitted
under the Copyright Act of 1976 and the right to store and retrieve one copy of the work, you may not
decompile, disassemble, reverse engineer, reproduce, modify, create derivative works based upon,
transmit, distribute, disseminate, sell, publish or sublicense the work or any part of it without
McGraw-Hill’s prior consent. You may use the work for your own noncommercial and personal use;
any other use of the work is strictly prohibited. Your right to use the work may be terminated if you
fail to comply with these terms.
THE WORK IS PROVIDED “AS IS.” McGRAW-HILL AND ITS LICENSORS MAKE NO
GUARANTEES OR WARRANTIES AS TO THE ACCURACY, ADEQUACY OR COMPLETE-
NESS OF OR RESULTS TO BE OBTAINED FROM USING THE WORK, INCLUDING ANY
INFORMATION THAT CAN BE ACCESSED THROUGH THE WORK VIA HYPERLINK OR
OTHERWISE, AND EXPRESSLY DISCLAIM ANY WARRANTY, EXPRESS OR IMPLIED,
INCLUDING BUT NOT LIMITED TO IMPLIED WARRANTIES OF MERCHANTABILITY OR
FITNESS FOR A PARTICULAR PURPOSE. McGraw-Hill and its licensors do not warrant or
guarantee that the functions contained in the work will meet your requirements or that its operation
will be uninterrupted or error free. Neither McGraw-Hill nor its licensors shall be liable to you or
anyone else for any inaccuracy, error or omission, regardless of cause, in the work or for any damages
resulting therefrom. McGraw-Hill has no responsibility for the content of any information accessed
through the work. Under no circumstances shall McGraw-Hill and/or its licensors be liable for any
indirect, incidental, special, punitive, consequential or similar damages that result from the use of or
inability to use the work, even if any of them has been advised of the possibility of such damages. This
limitation of liability shall apply to any claim or cause whatsoever whether such claim or cause
arises in contract, tort or otherwise.
DOI: 10.1036/0071455442
Professional
CONTENTS
Preface xiii
1.1 Introduction / 1
1.2 Importance of Epoxy Resins / 3
1.3 Importance of Epoxy Adhesives / 6
1.3.1 Advantages and Disadvantages of Adhesive Bonding / 6
1.3.2 The Nature of the Epoxy Adhesive Industry / 8
1.3.3 Epoxy Adhesive Markets / 9
1.4. Formulating Epoxy Adhesives / 18
1.4.1 The Job of the Adhesive Formulator / 18
1.4.2 The Basics of Adhesive Formulation / 19
References / 24
2.1 Introduction / 27
2.2 Epoxy Resin Characteristics / 27
2.3 Synthesis of Epoxy Resins / 30
2.3.1 Diglycidyl Ether of Bisphenol A (DGEBA) / 30
2.3.2 Other Epoxy Resins / 32
2.4 Epoxy Curing Mechanisms / 36
2.4.1 Polyaddition Reactions / 37
2.4.2 Homopolymerization Reactions / 38
2.5 Stoichiometry / 38
2.5.1 Amine Concentration / 39
2.5.2 Anhydride Concentration / 41
References / 41
3.1 Introduction / 43
3.2 Properties of the Uncured Epoxy System / 44
3.2.1 Vapor Pressure / 45
3.2.2 Viscosity / 45
3.2.3 Wetting / 49
3.2.4 Reactivity / 53
3.3 Properties of the Curing Epoxy System / 54
3.3.1 Preassembly Reactions / 54
3.3.2 Localized Stresses due to Gas or Air Pockets / 55
v
vi CONTENTS
3.3.3 Shrinkage / 57
3.3.4 Thermal Expansion Differences / 59
3.4 Properties of the Cured Epoxy System / 60
3.4.1 The Problem with Rigidity in Epoxy Adhesives / 61
3.4.2 Effect of Crosslink Density / 62
3.4.3 Glass Transition Temperature Tg / 64
3.4.4 Properties Resulting from Elevated versus Room Temperature Cure / 67
References / 68
4.1 Introduction / 71
4.2 Diglycidyl Ether of Bisphenol A Epoxy Resins / 72
4.2.1 Liquid DGEBA Resins / 73
4.2.2 Solid and Semisolid DGEBA Resins / 75
4.2.3 Brominated DGEBA Epoxy Resins / 76
4.3 Epoxy Novolac and Other Epoxy Resins / 77
4.4 Flexible Epoxy Resins / 78
4.5 Waterborne Epoxy Resins / 79
4.6 Epoxy Acrylate Resins / 82
References / 84
5.1 Introduction / 85
5.2 Aliphatic Amines / 88
5.2.1 Primary and Secondary Aliphatic Amines / 88
5.2.2 Modified Aliphatic Amines / 93
5.3 Polyamides and Amidoamines / 95
5.3.1 Polyamides / 95
5.3.2 Amidoamines / 96
5.4 Aromatic Amines / 96
5.4.1 Metaphenylene Diamine / 97
5.4.2 Methylene Dianiline / 98
5.4.3 Other Aromatic Amines / 98
5.5 Anhydrides / 99
5.5.1 Hexahydrophthalic Anhydride / 102
5.5.2 Nadic Methyl Anhydride / 102
5.5.3 Pyromellitic Dianhydride / 102
5.5.4 Other Anhydride Curing Agents / 103
5.6 Catalysts and Latent Curing Agents / 103
5.6.1 Tertiary Amines / 104
5.6.2 BF3-Monoethylamine / 104
5.6.3 Imidazoles / 105
5.6.4 Dicyandiamide / 106
5.6.5 Other Latent Catalysts / 107
5.7 Mercaptan and Polysulfide Curing Agents / 107
References / 109
Index 527
This page intentionally left blank
PREFACE
The adhesive industry has seen significant changes in recent years. The type and number
of these changes have been astounding. They include new substrate materials, regulatory
burdens, acquisitions and mergers, new raw materials, new application and curing processes,
and a host of volatile technical, commercial, and political issues. However, a staple in the
industry throughout this change has been the adhesive formulator.
At first glance, the job of the adhesive formulator appears to be deceptively simple.
There are many resources and tools available on the supply side of the industry for the for-
mulator to use. The problem is that there seems to be an even more staggering amount of
requests for new products on the demand side of the industry.
Epoxy technology gives the formulator an almost unlimited number of tools to employ.
The type of epoxy polymer backbone, curative, resinous modifiers, and special additives or
fillers all serve as degrees of freedom available in developing an adhesive system for a
given application. However, often the adhesive must fulfill many different requirements,
such as ease of assembly, moderate cure time, resistance to thermal cycling, and resistance
to stress at high temperature and humidity. The creation of a functional adhesive system for
a new application, therefore, is usually a lengthy undertaking, which depends on the skill
and discretion of the formulator.
Unfortunately there is not a lot of information to assist the formulator in the proper or
optimal use of the tools and resources available. There are not many forums that provide
instruction on the “art” of adhesive formulating. Only a few textbooks have concentrated
on the subject. And even though epoxy adhesives are the “workhorse” of the industry and
occupy the majority of the structural adhesives market, practical, concentrated information
on epoxy adhesive formulation is noticeably absent. This book is an attempt to correct this
situation.
I have tried not so much to present new material and the latest developments in epoxy
adhesive technology as to provide a useful and organized summary of the many fragmented
sources of information that already exist. Most of the information in this text comes from
raw materials suppliers, technical papers, conference proceedings, and other scattered
details. I have attempted to blend this with my nearly 40 years of experience in formulat-
ing, using, and consulting on epoxy adhesive applications. Hopefully, the outcome is (1) a
useful starting point and guidance for the novice in the field, (2) a handbook of readily
accessible practical information for those working everyday with epoxy adhesives, and
(3) a portal to additional technical information and research on the subject.
This book is aimed primarily at the formulator, but it is sometimes difficult to define
who the formulator actually is. Often the end user may also need to formulate an epoxy
adhesive from raw materials in order to reduce cost or because of the uniqueness of the
application. Thus, I hope that the end user as well as those involved in other aspects of the
adhesives industry (purchasing agents, designers, manufacturing engineers, etc.) find this
book interesting and worthwhile.
When undertaking the job of preparing for this book, I originally intended to provide a
collection of epoxy adhesive formulations that could be used for specific applications, and
hopefully I have accomplished that task. However, it is also necessary to discuss surface
xiii
Copyright © 2006 by The McGraw-Hill Companies, Inc. Click here for terms of use.
xiv PREFACE
preparation, testing, application processes, and curing methods to properly employ the formu-
lation information. As a result, I also provide “accessory” information that is needed to either
manufacture a successful epoxy formulation or use it in practice. For greater detail regarding
adhesives in general, bond theory, substrate preparation, test methods, and so forth, the reader
is directed to the Handbook of Adhesives and Sealants (Edward M. Petrie, McGraw-Hill,
2000).
Epoxy adhesives can be formulated to bond to a variety of substrates and perform over
a broad range of conditions. They are used as one- or two-component systems, which are
curable under ambient or elevated-temperature conditions. They are available in a variety
of forms including liquids, pastes, film, solids, solvent solutions, and water dispersions.
They are used in many industries and can be truly considered a workhorse.
The epoxy adhesives are popular with both the formulator and the end user for a
number of reasons. When compared to other possible adhesive systems, epoxies offer
the following added values:
• Nearly infinite number of ways to engineer an adhesive to provide the required applica-
tion properties and end-use properties in a joint
• A variety of forms and curing methods to optimize the assembly process
• No evolution of volatiles and low shrinkage during cure
• Good wetting properties resulting in excellent adhesion to most substrates
• Excellent cohesive strength and mechanical properties
• Good moisture, humidity, chemical, and temperature resistance
The main raw materials used in epoxy adhesive formulations (resins and curing agents)
can be synthesized in a variety of ways to create many different products. Epoxies react
readily via several polymerization mechanisms. The extent of crosslinking is an important
determinant of the final properties of the adhesive. Crosslinking can be controlled by the
choice of resin and curing agent and by the curing conditions.
Crosslinked epoxy resins are cohesively very strong, but their brittleness reduces their
usefulness in many adhesive applications, especially those requiring a high degree of peel
and impact strength or thermal cycling. Development in recent years has largely focused
on the toughening needed for epoxies to be structural adhesives. Epoxies can be toughened
by modifying the chemistry of the raw materials within the formulation or by adding par-
ticulates, elastomers, or thermoplastics to the formulation.
Since epoxy adhesive formulation represents a surprisingly broad area of technology, a
road map to the use of this book may be valuable. Chapters 1 through 3 discuss the synthesis
of raw materials, epoxy chemistry in general, and the physical and chemical properties that
are important for an epoxy adhesive. These properties are important during the three primary
phases or conditions of an adhesive: (1) uncured, (2) during cure, and (3) fully cured.
Chapters 4 through 10 describe the basic raw materials that are commonly employed in
formulating epoxy adhesives. These include the epoxy resins, curing agents and catalysts,
solvents and diluents, resinous modifiers, flexibilizers and tougheners, fillers, and adhesion
promoters.
Formulation details are then presented in Chapters 11 through 14 for the various possi-
ble forms of epoxy adhesive systems: room temperature and elevated-temperature curing
liquids, pastes, and solids. The more or less unconventional forms of epoxy adhesives are
also identified and discussed, since these are now achieving prominence in industry. These
include uv and electron beam radiation curable, waterborne systems, and epoxy adhesives
capable of curing via the indirect application of heat or energy.
Chapters 15 and 16 describe formulations that have been specifically developed for use in
certain service environments and with certain substrates. The effects of the environmental
CONTENTS xv
exposure on epoxy adhesives are described. The substrate properties that enhance or are
detrimental to bonding success are identified. Surface preparation procedures for common
substrates are identified in the text and detailed processes are defined in Appendix F.
Chapters 17 through 20 describe the various processes and equipment employed in the
formulation or end use of epoxy adhesives. Health and safety issues regarding the use of
these materials are discussed. The importance of quality control methodologies and speci-
fication preparation is also noted. Finally, test methods that are commonly used by both the
formulator and the end user are identified.
Throughout this book, specific applications and examples illustrate the concepts being
discussed. Due to the broad nature of the subject, the reader will often be directed from one
section to another relevant section. References at the end of each chapter direct the inter-
ested reader to more in-depth information and understanding regarding specific subjects.
Formulating skills cannot be easily learned in the classroom or from reading books.
These sources can provide a starting point and a road map, but that is not the end point.
Formulating is somewhat of an “art” and as such can only be effectively learned at the
bench by practicing and, unfortunately, often by trial and error. This book can be a means
of reducing the time, cost, and potential problems inherent in this process.
ACKNOWLEDGMENT
This book is dedicated to my mother, Katherine. I am sure that I would never have achieved
what I have in life without her skills at formulating and applying the real basics. She has
been both the cornerstone and the shining beacon of our family. Most of all, I am grateful
for her dedication and tireless enthusiasm in resolving all the troubles that life throws at us.
EDWARD M. PETRIE
This page intentionally left blank
CHAPTER 1
EPOXY ADHESIVES
1.1 INTRODUCTION
Epoxy adhesives are chemical compounds used to join components by providing a bond
between two surfaces. Epoxy adhesives were introduced commercially in 1946 and have wide
applications in the automotive, industrial, and aerospace markets. Epoxies are probably the
most versatile family of adhesives because they bond well to many substrates and can be eas-
ily modified to achieve widely varying properties. This modification usually takes the form of
1. Selection of the appropriate epoxy resin or combination of resins of which many are
available
2. Selection of curing agent and associated reaction mechanism
3. Simple additions of organic or inorganic fillers and components
Copyright © 2006 by The McGraw-Hill Companies, Inc. Click here for terms of use.
2 CHAPTER ONE
these materials can be applied and set as well as variations in final performance properties
within a joint.
Most commercially available epoxy structural adhesives are either single-component,
heat curing adhesives or multiple-component adhesives that cure at either room or elevated
temperature. Generally, epoxy systems that cure at elevated temperatures have a higher
crosslinking density and glass transition temperature than systems that are formulated to
cure at room temperature. This provides the elevated-temperature curing epoxies with bet-
ter shear strength, especially at elevated temperatures, and better environmental resistance.
However, they usually have poor toughness and peel strength due to their rigidity.
Depending on the epoxy resin and curing agent used, room temperature curing adhesive
formulations can harden in as little as several minutes at room temperature, but most sys-
tems require from 18 to 72 h to reach full strength. The room temperature curing epoxy
adhesives can also be cured at elevated temperatures, if faster curing times are required.
The curing time is greatly temperature-dependent, as shown in Fig. 1.1.
Once the curing agent is added to the epoxy resin, the adhesive must be used within a
time period that is dependent on the resin and curing agent in the formulation and on the
ambient conditions. With most room temperature systems, the time period is short, and the
maximum time from mixing to application is considered the pot life of the system. For most
single-component chemistries, the allowable time from mixing the hardener into the resin
to application is relatively long and dependent on the storage conditions. This time period
is defined as the adhesive’s shelf life.
An exothermic reaction occurs on curing of epoxy adhesive systems. Exotherm is the
heat that is self-generated by the curing reaction. If the exotherm is not controlled, it can
lead to greatly reduced working life and even to a dangerous situation where the exotherm
can be great enough to cause combustion. The exotherm can be minimized by reducing the
reactivity of the mixed components and by limiting the batch size.
Epoxy adhesives are often given a postcure. Room temperature curing and elevated-
temperature curing epoxy adhesives are cured to a state in which the resin is hard and the
bond is capable of handling (i.e., handling strength). This provides the opportunity of
4000
165°F
3000
Tensile shear, psi
100°F
2000
75°F
1000
0
0 2 4 6 8 21 23 25
Cure time, h
Gel and hardening Handling strength Full strength
releasing the fixturing used to hold the parts in place and moving the bonded component to
the next manufacturing stage. Once the handling strength has been reached, the joint is
postcured without fixturing by placing it in an oven or other heat source until it achieves
full strength. Often postcuring can be combined with some other postbonding process such
as paint curing or component drying. The epoxy adhesive manufacturer should be con-
sulted to determine if postcuring is a viable option.
Many epoxy adhesives are capable of being B-staged. A B-staged resin is one in which
a limited reaction between the resin and hardener has taken place so that the product is in a
semicured but solid state. In the B-staged state, the polymeric adhesive is still fusible and
soluble. On additional heat curing, the adhesive will progress from the B stage to a com-
pletely cured state. This will usually be accompanied by moderate flow.
The advantage of B-staged resins is that they permit the formulation of one-component
solid adhesives such as films, powders, and preforms. The user can then purchase an adhe-
sive product that does not require metering or mixing. There is very little waste associated
with B-staged products, and they generally provide better consistency. The B-staged epoxy
resins are formulated using curing agents, such as aromatic polyamines, along with solid
forms of DGEBA epoxy resins.
Secondary ingredients in epoxy adhesives include reactive diluents to adjust viscosity;
mineral fillers to lower cost, adjust viscosity, or modify the coefficient of thermal expan-
sion; and fibrous fillers to improve thixotropy and cohesive strength. Epoxy resins are often
modified with other resins to enhance certain properties that are necessary for the applica-
tion. Often these modifications take the form of additions of elastomeric resins to improve
toughness or peel strength.
Epoxy resins are high-performance thermosetting resins, which display a unique combina-
tion of properties. Epoxy resins have been commercially available for almost a half-century.
Epoxy resins are arguably one of the most versatile polymers with uses across an enor-
mously wide variety of industries. The outstanding physical properties exhibited by epoxy
resins include
Furthermore, epoxy resin systems are capable of curing at either ambient or elevated tem-
peratures, and they require only minimal pressure during the cure. Thus, epoxies can be
applied and cured under many adverse conditions including outdoors. These properties pro-
vide great added value in many industries engaged in product assembly. Epoxy resins have
been commercially available for almost a half-century.
Many new applications are being developed that will ensure the prominence of epoxy
resins in the future. For example, epoxies are the material of choice for the reinforcement or
consolidation of aging or damaged concrete structure including walls, ceiling beams, bridge
columns, and anchoring mechanisms. They are the main binders for graphite-reinforced
4 CHAPTER ONE
composites such as those used in manufacturing tennis rackets, fishing rods, skis, snow-
boards, and golf club shafts. They have also been used for manufacturing high-strength,
lightweight carbon fiber pultrusion as replacements for steel rebar in special concrete expan-
sion projects.
Industries using epoxy adhesives include aerospace, civil engineering, automotive, chem-
ical, electrical, marine, leisure, and many others. Table 1.1 lists some of the more common
applications for epoxy compounds. The prevailing reason for the broad acceptance of epoxy
resins in these important and diverse markets is their capacity to provide a good balance of
handling characteristics and ultimate physical properties. They adhere well to a very large
variety of substrates, and they generate tough, environmentally resistant films or matrices.
One of the major advantages of epoxy chemistry is the wide latitude it provides the for-
mulator for solving technical problems. Epoxies can be designed to be flexible or rigid; high
or low modulus; homogeneous, filled, or foamed; conductive or insulative; fire-retardant;
and resistant to heat and chemicals.
The number of raw materials that the formulator has to work with is enormous. These
include epoxy resins and modified epoxy resins of all types and forms, fillers and additives,
Chemical: Textile:
Linings for storage tanks Equipment parts
Chemical plant including coatings Glass and carbon fiber sizing agents
Pipes and pipe linings Light engineering:
Filters Adhesives
Protective and decorative coatings
Composite structures for artificial limbs
EPOXY ADHESIVES 5
modifiers, reinforcements, diluents, and solvents. The possible curing agents that can be
used also provide great latitude in formulation. Often the curing agent becomes an integral
part of the resulting compound. Its choice is a controlling influence on the curing proper-
ties of the mixture and on the performance properties of the cured adhesive. One of the
chief advantages of epoxy resins is that they generally allow easy incorporation of addi-
tives, and the resulting formulation is one that can be easily adapted to many manufactur-
ing processes, such as
• Pultrusion
• Lamination
• Filament winding
• Molding
• Casting and potting
• Coating
• Adhesive bonding
Epoxy resins are not the lowest-cost resins potentially available for most applications.
Thus, epoxy resins must provide added value to justify their additional cost. This added
value is usually realized by the incorporation of a special property or combination of prop-
erties into the final product.
The epoxy resin production value in the United States, western Europe, and Asia is esti-
mated to be over $2.5 billion. Almost 900,000 metric tons (t) of epoxy resins is consumed
in these regions with around 35 percent of the consumption in Asia. The overall global con-
sumption in 2002 was approximately 970,000 t (2.14 billion lb) with a 4 to 5 percent annual
growth rate.3 The 7 to 10 percent growth rate in the 1970s has slowed except in Asia. Future
consumption will average about 2 to 4 percent per year from 2004 to 2008 in the United
States and western Europe.4 In North America the demand for epoxy resins is forecast at
690 million lb, valued at $1.4 billion, in 2004.5
The three leading producers of epoxy resin account for approximately 75 percent of the
world’s capacity. These resin manufacturers are Resolution Performance Products (formerly
Shell), Dow Chemical, and Huntsman (formerly Ciba). The other 50+ smaller epoxy manu-
facturers primarily produce epoxy resins only regionally or produce specialty products.
The industry is currently in the midst of a major restructuring period, with producers of
epoxy resins announcing mergers, acquisitions, and divestiture of operations. Examples of
leading producers of epoxy resins, curing agents, and additives for the adhesives market are
included in Apps. A through D. These and other companies also offer a wide range of asso-
ciated products, including solvents, primers, and chemicals.
Epoxy resins are commercially available as either liquids or solids. The liquids are avail-
able as (1) solvent-free resins, ranging in viscosity from waterlike liquids to crystalline solids;
(2) waterborne emulsions; and (3) solvent-borne solutions. Generally, the higher the molec-
ular weight of the epoxy resin molecule, the higher the viscosity or melting point.
Epoxy resins are composed of polymeric molecules that are converted to a solid by a
chemical reaction. Epoxy systems physically comprise two essential components: a resin and
a curative. The curative causes the chemical reaction, which turns the epoxy resin into a solid,
crosslinked network of molecules. This polymer is called a thermoset polymer because, when
cured, it is irreversibly rigid and relatively unaffected by heat. (By contrast, thermoplastic
polymers are not crosslinked and can be made to flow with the application of heat.)
Cure of the epoxy resins is initiated once the resin is mixed with a curative. The cure of
all epoxy resins is an exothermic process where heat is generated as a natural result of the
chemical reaction. Success in using most epoxies is dependent on handling the product in
the correct way to avoid premature cure and unwanted side reactions.
6 CHAPTER ONE
Epoxy resins are seldom used in their unmodified forms. Formulation is generally a
necessity of any epoxy product. There are many reasons for this, including
Almost everything that is made by industry has component pieces, and these have to be fixed
together. There are many alternatives to adhesive bonding that a manufacturer can employ
such as screws, rivets, and spot welds. Each potential joining process must be considered
with regard to its specific requirements. There are times when adhesives are the worst pos-
sible option for joining two substrates, and there are times when adhesives may be the best
or only alternative. Usually, the choice of joining process is not all black or white. Certain
processes will have distinct advantages and disadvantages in specific applications. The
choice may involve a tradeoff in performance, production capability, cost, and reliability.
Often, one must consider the time, trouble, and expense that may be necessary to use an
adhesive. For example, certain plastics may require expensive surface preparation processes
to allow the adhesive to wet the surface. Applications requiring high-temperature service
conditions may require an adhesive that necessitates an elevated-temperature cure over a
prolonged period.
On the other hand, certain applications could not exist without adhesive bonding.
Examples of these are the joining of ceramic or elastomeric materials, the joining of very
thin substrates, the joining of surface skin to honeycomb, and numerous other applications.
There are also certain applications where adhesives are chosen because of their low cost
and easy, fast joining ability (e.g., packaging, consumer products).
Sometimes conventional welding or a mechanical joining process is just not possible.
Substrate materials may be incompatible for metallurgical welding due to their thermal
expansion coefficients, chemistry, or heat resistance. The end product may not be able to
accept the bulk or shape required by mechanical fasteners.
The science of adhesive bonding has advanced to a degree where adhesives must be con-
sidered an attractive and practical alternative to mechanical fastening for many applications.
Adhesive bonding presents several distinct advantages over other conventional methods of
fastening. There are also some disadvantages which may make adhesive bonding imprac-
tical. These pros and cons are summarized in Table 1.2.
EPOXY ADHESIVES 7
Advantages Disadvantages
The design engineer must consider and weigh these factors before deciding on a method
of fastening. However, in many applications adhesive bonding is the only logical choice.
In the aircraft industry, for example, adhesives make the use of thin metal and honeycomb
structures feasible because stresses are transmitted more effectively by adhesives than by
rivets or welds. Plastics, elastomers, and certain metals (e.g., aluminum and titanium) can
be more reliably joined with adhesives than with other methods. Welding is usually at too
high a temperature, and mechanical fastening destroys the lightness and aesthetics of the
final product. Certain examples of less obvious applications where adhesive bonding is a
practical method of assembly are shown in Table 1.3.
The design engineer will find that once he or she decides that adhesive bonding is the
optimal method of assembly, there are a significant number of adhesive families to choose
from and an enormous number of formulations within each family. However, several uni-
versal factors should be considered when one is selecting or developing a formulation for
an adhesive application. These are generally related to the materials that will be bonded, the
bonding processes that are available and practical for use, and consideration of the envi-
ronments to which the final adhesive joint will be exposed.
• Materials to be bonded:
Surface energy of the adhesive and the substrate
Surface preparation processes that are available and practical
End-use properties of the joint
• Process of bonding:
Size of gap, viscosity of adhesive formulation
Off-gassing, flammability, etc., during the curing process
Pot life: maximum time between metering/mixing and application of the adhesive
• Environmental considerations:
Service temperature limited by glass transition and thermal degradation of the adhesive
Influence of water on the adhesive and interface
Chemical degradation by ultraviolet (uv), radiation, solvents, and other environmental media
8 CHAPTER ONE
Application areas
for adhesives Examples
When considered against the criteria listed above, epoxy adhesive systems provide a well-
balanced set of properties. This solidifies their competitive position in the adhesives industry.
The industries that are most affected by epoxy adhesives consist of four main categories:
believe that it will improve a customer’s performance or production efficiency or will add
some other value. However, a minimum volume is usually necessary before formulators
will make modifications to a standard formulation or develop a new product for a specific
application. Formulators have a significant knowledge base regarding adhesive or sealant
systems and how they are to be applied in practice. The properties of several commercial
epoxy adhesive formulations appear in App. B.
The trend is for the end user to purchase an adhesive from a formulator, rather than pro-
duce it internally. It is increasingly difficult for an end user to keep up with continuing tech-
nological changes in adhesives. However, the end user must select a commercial adhesive,
substrate, joint design, and processing conditions for specific applications. Once these are
selected and verified as to performance and cost, the end user must be vigilant that none of
the processes, materials, or other relevant factors in the assembly process change.
Several other industries are also affected by adhesives and sealants. For example, equip-
ment suppliers specialize in producing machinery for application, assembly, curing, surface
preparation, etc. Also equipment suppliers specialize in developing and manufacturing test-
ing apparatus that can be used to measure joint strength and processing parameters. Then
there are testing laboratories and consultants who provide assistance and services to the for-
mulators and end users on a contractual basis.
The inception of epoxy adhesives occurred almost simultaneously with the commercialization
of epoxy resins. This is due to the fact that all the unique handling, application, and perfor-
mance properties of epoxy resins are especially important when it comes to adhesive bonding.
Epoxy adhesives have become the most recognizable structural adhesive type. They
have found commercial success in demanding industries such as aerospace, automotive,
building and construction, and electrical and electronic. Their ease of use has also encour-
aged their commercialization in the do-it-yourself markets, which has added to the exem-
plary reputation of epoxy adhesives.
Adhesive markets represent only a small percentage of the total consumption of epoxy
resins. However, epoxy adhesives provide significant value added, so that their prices and
profit margins are generally higher than those for other adhesive types. Globally in the
year 2002, epoxy adhesives represented about 31 percent (or about 0.7 billion lb) of the
total liquid epoxy demand (Fig. 1.2). The U.S. market is estimated to be just less than
200 million lb.10
Epoxy adhesives represent a significant part of the overall structural adhesives market
(about $1.8 billion). The main competitors to epoxy adhesives are polyurethanes; however,
thermosetting acrylics and cyanoacrylate adhesives are also strong challengers in certain
market segments.
Single-component epoxy adhesive formulations are the largest type of epoxy adhesives
sold, with about 55 percent of the consumption, while two-component formulations
account for another 44 percent of the volume. Radiation cure formulations represent the
remainder of the market. Epoxy adhesives can also take many forms including solids, solvent-
free liquids, solvent-borne systems, and waterborne systems.
Despite epoxy adhesives finding use in many fragmented markets, actual consumption
in volume is surprisingly concentrated in a few specific end-use market segments. For
example, automotive assembly applications account for nearly 50 percent of the total vol-
ume of epoxy adhesives consumed in the United States. The highest-value market areas
include structural automotive, aircraft, and many specialty product assembly applications.
These market areas are also expected to enjoy the highest growth rates.
Although the overall annual growth rate for epoxy adhesive is in the 3 to 5 percent
range, certain regional markets, such as China, are expected to have an annual growth rate
10 CHAPTER ONE
Other 28%
Adhesives
31%
Coatings
41%
FIGURE 1.2 Global liquid epoxy resin production.11 2002 Total
Liquid Epoxy Resin Demand: 2.14 billion pounds; 970 thousand
metric tons; $2.0 billion; AGR 4–5%. (Source: DPNA International
Inc.)
of at least 8 percent per year (Fig. 1.3). There are several new developments that are affect-
ing the growth rate for epoxy adhesive systems. These drivers are classified as “technical
developments” and “market trends” in Table 1.4.
Analysis of the epoxy adhesive markets is difficult because so many market segments
fragment the industry and because it is difficult to confine manufacturers, distributors, and
end users into specific categories. However, there are several organizations that periodi-
cally publish market reports on the adhesives industry including epoxy adhesives.14
Advances in epoxy adhesive technology have meant that an increasing number of
assembly and repair problems can be resolved without the use of mechanical fasteners,
welding, or other nonadhesive assembly processes. Several of the more important devel-
oping market segments for epoxy adhesives are described in the following sections.
Automotive. The automotive industry has a wide variety of applications for adhesives
and sealants. Nonstructural applications dominate, and they include all interior trim and
much of the exterior trim such as side molding, wheel covering emblems, front and rear
8
7
Percent growth rate
6
5
4
3
2
1
0 North West Asia- China Japan Others
America Europe Pacific
FIGURE 1.3 Epoxy adhesive regional growth rate.12 (Source: DPNA International Inc.)
EPOXY ADHESIVES 11
taillight assemblies, side-view mirrors, and ornamentation. Adhesives and sealants are
also being used in many under-the-hood applications. However, structural adhesives are
making the largest gains in this industry. After years of field experience, structural
adhesive bonding has proved that it can provide high-volume assemblies with greater
stiffness and better crash performance. Being viscoelastic, most structural organic adhe-
sives also provide improvements in vehicle noise, vibrations, and harshness (NVH).
This became possible only after the successful development of new generations of adhe-
sives and specifically with the development of new toughening additives. As a result,
adhesive bonding is now accepted in the transportation industry as a “classic” joining
technique.
To understand the popularity of adhesive bonding in the automotive industry, one needs
to take a look at the characteristics that define the current market situation for autos.
• Although steel, including high-strength types and coated types, will continue to be the
dominant body material, nonferrous metals (e.g., aluminum) and composites are increas-
ingly being used.
• Combinations of different materials (e.g., metals and composites) are being utilized for
design variation and for reduced weight.
• Increasing emphasis is being placed on multifunctional application (e.g., vibration damp-
ing plus joining) to reduce costs and improve production times.
• The need to improve crash capability and to reduce weight is more important today than
ever before to meet safety and environmental standards.
• Fatigue and durability of assembled parts are more important because of increasing war-
ranty coverage.
• Modular design concepts are being introduced to maximize design versatility and to min-
imize costs. These modular designs are conveniently assembled with adhesives.
Because adhesive formulators resonate well with these requirements, adhesives are bound
to play an increasingly important role in the future of the automotive industry.
12 CHAPTER ONE
Also, several new bonding processes and products have gained the attention of the auto-
motive industry. Weldbonding (see Chap. 14) has improved the production situation by
eliminating the need for fixturing and being easily adapted to robotic processing methods.
New structural adhesives such as toughened epoxy (Chap. 8), thermosetting acrylics, and
polyurethanes are providing a greater degree of crash protection in automobiles, and these
materials bond well to nonmetals and poorly prepared surfaces.
Just as their counterparts rejuvenated the aircraft industry decades ago, modern epoxy
structural adhesives are providing automobile designers and manufacturers with innovative
possibilities that could not have been considered only a short time ago. As a result, struc-
tural automotive adhesives will have an average annual growth rate of greater than 7 percent
over the next 5 years.15 This growth rate is additive to any growth that will occur in the auto-
motive industry itself.
Electrical and Electronic. There are many industries where adhesives must perform sev-
eral special functions in addition to their primary function of bonding or joining. These indus-
tries often have test methods, specifications, and nomenclature that are unique and somewhat
foreign to adhesive formulators. One such industry is the electrical and electronic industry.
The electrical and electronic industries are generally lumped into one since it is difficult
to place boundaries between the two, and they often have similar, if not identical, require-
ments. The electrical industry is comprised of manufacturers of motors, transformers, meters,
etc. It is a large, stable industry with demands for many types of adhesive products.
The electronic industry, on the other hand, is growing significantly, and new products are
forthcoming at a rapid rate. Demand for electronic polymer products is forecast to expand
6.4 percent per year to $3.5 billion in 2005. Adhesives used in this industry will grow at a rate
of 8.8 percent, faster than the industry itself. This will require 485 million lb of resin valued at
$1.6 billion. Of this, about 123 million lb of adhesives will be used in the electronics market.16,17
Adhesives are used in the electrical and electronic industries in a variety of different
ways, from holding microcomponents in place on a circuit board to bonding coils in large
power transformers. Reliability is always a concern, since bond failure could lead to com-
ponent failure, which in turn leads to equipment failure and then possibly to a massive sys-
tem failure. A “system” in this industry could be a commercial aircraft’s electrical system
or the power distribution system in an urban city.
In the electrical and electronic equipment market, adhesives compete with other joining
methods such as mechanical fasteners, brazing, welding, soldering, and thermocompres-
sion bonding. Major uses of adhesives in the electrical and electronic equipment market are
shown in Table 1.5. Adhesives are generally the preferred method of joining whenever
Thermosetting adhesives, such as epoxies, are the workhorses of the industry. However,
there are persistent challenges from various higher-priced specialty resins.
In general, the use of adhesives in this industry assumes that the adhesive must provide
a dielectric function in addition to providing the mechanical joining function. The dielec-
tric function is often related to some degree of electrical insulation, but could also be related
to electrical conductivity, depending on the application. An excellent example of an insu-
lating function is the bonding of conductive foil to plastic films and laminates for the fab-
rication of printed-circuit boards and flexible wiring harnesses. Here the adhesive must
EPOXY ADHESIVES 13
TABLE 1.5 Representative Adhesive Applications for the Electrical and Electronic Industries
Electrical Electronic
adhere the conductive copper to the laminate. It must also provide a degree of surface insu-
lation once conductive paths are etched from the bonded specimen.
In addition to joining, adhesives in electrical applications may be required to conduct
heat, conduct or isolate electricity, provide shock mounting, seal, protect substrates, etc.
Thermal and chemical resistance, weathering, and structural compatibility must also be
considered in diverse electrical and electronic applications. Of course, the choice of adhe-
sive will also be governed by application methods, cure temperature, processing speed, and
overall economic cost.
Epoxy resin can have varying amounts of inorganic material remaining in the product after
its synthesis. One of these is sodium chloride, which is formed by the reaction of sodium
hydroxide with epichlorohydrin molecules, as described in Chap. 2. This is important in elec-
trical applications because the sodium salt can be hydrolyzed and can degrade electrical prop-
erties. Some epoxy resin manufacturers will offer special ultrapure electrical grades or
low-chlorine grades of epoxy resins for application where this is an important factor.
TABLE 1.6 Common Applications and Polymeric Resins Used in the Construction and
Repair of the Transportation Infrastructure
weather as are slower-curing systems. Their physical properties do not change significantly
on aging in the field.
One problem with early epoxy formulations is that they cured to a relatively brittle
material. By using reactive flexibilizers, such as polysulfides, epoxy adhesive formulators
have obtained the flexibility required for many applications in this industry. Polyamides
and even coal tars have also been used to provide flexibility to epoxy base resins.
Another problem is that many two-component, room temperature curing epoxies cure
with a high exotherm. This can cause the mixed uncured epoxy to get very hot. Excessive
heat of exotherm provides a safety hazard and significantly shortens the working life of the
mixed adhesive. This problem is generally solved by mixing small batches or using con-
tinuous mixing equipment.
Cured concrete can be bonded to cured concrete, as in the installation of precast buttons
to a highway surface. Steel bridge railings can also be bonded to the concrete surface of a
bridge sidewalk. In the case of deteriorated concrete, the adhesive can be used to rebuild
the structure to its former line and grade. Epoxy adhesives are also commonly used on other
roadway materials, such as asphalt and brick; however, the predominant application is con-
crete substrates. The most frequent combinations of substrates that are bonded with adhe-
sives in this market segment are
In the bonding of cured concrete to cured concrete, the adhesive is applied directly to the
substrates. In the case of new concrete, the bonding agent may be incorporated into the con-
crete formulation, or the adhesive may be applied to the nonconcrete surface.
There are many applications for bonding cured concrete to another material. These
include bonding of plastic reflectors to road surfaces or dividers, attachment of railings or
metal structures, and bonding of floor covering materials. An interesting application that
has received much attention because of the aging road infrastructure in many regions is the
repair and strengthening of existing concrete structures. Concrete beams can be strength-
ened by attaching steel plates and composites with adhesives.
For applying new concrete over old concrete, such as in the case of a patch, epoxy is often
the best solution as an alternative for the new concrete. In these formulations an epoxy
EPOXY ADHESIVES 15
mortar is made using aggregate similar to the aggregate used in concrete mixtures. Adhesion
is excellent and the epoxy mortar has generally better properties than the concrete; however,
it is a relatively high-cost solution. The old concrete surface should be cleaned and sand-
blasted down to sound concrete before the epoxy patch is applied.
There are fewer standards for highway adhesives than there are for sealants. However,
a very useful reference, published by the American Concrete Institute (ACI), provides the
engineer, contractor, or architect with a description of the various types of polymer adhe-
sives most frequently used for bonding of concrete.18 The guide emphasizes the factors that
should be considered when one is selecting a structural adhesive, including characteristics
during installation and when in service. The benefits and limitations of adhesive bonding
are also discussed for each application.
In the future, the use of adhesives in this market is expected to grow significantly due
to the aging infrastructures. Also, the use of epoxy adhesive systems with fiber reinforce-
ment is expected to increase significantly. Such adhesives will be applied to bridge piers,
beams, and other elements requiring repair. These adhesives can be used not only to repair
deteriorating structures but also to provide added strength to new structures.19,20
Medical. Many types of medical devices rely on adhesives for assembly. Medical device
manufacturing requires that the final product exhibit maximum reliability and performance
under many conditions that are not common in other industries. As a result, manufacturers
of medical devices require significant testing and verification prior to choosing an adhe-
sive. The standards and regulations in this industry are much different from those of other
industries. The nature of the medical device market also dictates that the adhesive be eco-
nomical and amenable to high-volume manufacturing methods.
This section focuses on adhesives that are used for the assembly of medical devices. In
medical device assembly, the primary substrates are plastics, elastomers, and metals. The total
“medical adhesive” market is much larger since it encompasses a broader definition of prod-
ucts. For example, medical adhesives can be used for bonding human tissue, transdermal drug
delivery systems, dental restoration, and wound care in addition to medical device assembly.
Adhesives are currently used in a wide range of medical devices, and the market for
these specialized adhesives is expected to grow significantly over the next decade. Factors
that will contribute to this growth include longer human life expectancy and the develop-
ment of new medical technology.
Just about any medical apparatus or diagnostic device may have applications for medical-
grade adhesives. Traditionally, there are three classes of medical devices that have been
assembled with adhesives:
• Disposables (e.g., syringe, catheters, and oxygenators)
• Reusables (e.g., surgical instruments, diagnostic equipment)
• Implantables (e.g., pacemakers)
These products generally come into contact with blood or bodily fluids, and they must be
subjected to a sterilization process prior to use. Nonsterile medical devices are also used
(e.g., diagnostic devices, medical electronics). These generally do not come into direct con-
tact with the patient.
Recently two additional types of medical devices have emerged:
• Sterile reusables (e.g., endoscopes, laparoscopes)
• Resposables (devices originally intended as a disposable but now considered for reuse)
Sterile reusable devices have undergone significant market growth over the past several
years due to advances in less invasive surgery. To the adhesive manufacturers, resposables
may be considered to have the same requirements as sterile reusables.
16 CHAPTER ONE
Implantables are less likely to use adhesives because of their longevity, performance,
and toxicity requirements. With these parts, self-assembly (snap-fit or interference fit) or
solvent and heat welding processes are commonly used for assembly.
End users of adhesives and medical devices expect the device to withstand the rigors of
sterilization, exposure to fluids, and occasional abuse. Because of the criticality of the prod-
uct, the medical device manufacturer is generally more motivated than most to pay atten-
tion to adhesive selection criteria and the requirements of good bonding practice.
Choosing an adhesive for a medical application follows the same process as choosing
an adhesive for any other purpose. Criteria include the particular substrates to be joined,
strength requirements, type of loading, impact resistance, temperature resistance, fluid
resistance, and processing requirements. However, the function of many medical devices
requires at least two other important criteria: resistance to sterilization and low toxicity. As
a result, several important standards and regulations have been developed in the industry.
Adhesives used in medical devices are tested for their effect on cells (cytotoxicity), blood
constituents (hemolysis), and adjacent tissues, and for overall systemic effect. Several classes
of biocompatibility testing exist. Adhesive suppliers, however, generally test to the following
guidelines that have been established for toxicological properties and biocompatibility:
• United States Pharmacopoeia (USP), Class VI Standard
• International Standards Organization (ISO), ISO-10993
These guidelines were originally developed for testing the suitability of plastics used in
medical devices that may come into contact with bodily fluids, but they have been extended
to adhesives as well. Generally, products are tested by an independent laboratory. The
results are typically provided to device or adhesive manufacturers in the form of certifica-
tions of compliance on an as-requested basis. Meeting these standards verifies that the suc-
cessfully tested products are nontoxic and biologically inert in the cured state.
The two standards specify slightly different tests and follow different methodologies.
The USP Class VI test method consists of acute systemic (over the tissue), intracutaneous
(under the skin), and muscle implantation (in the muscle) tests. Establishing a USP Class VI
rating has little bearing on whether the product will win approval from the Food and Drug
Administration (FDA). The Class VI rating merely states that the products exhibit a low
level of toxicity under the test conditions.
The ISO-10993 standard is generally recognized as the standard of choice for compa-
nies operating globally. It is widely recognized by North American, European, and Asian
countries. It is also somewhat more extensive than the USP Class VI standard. The ISO-
10993 biocompatibility testing includes
• Intracutaneous injection tests to evaluate the irritation potential of the material
• Acute systemic injection tests to evaluate the material for potential toxic effects as a
result of single-dose systemic injections
• Cytotoxicity tests to determine the biological reactivity of monolayer cell cultures to the
material
• Hemocompatibility tests to evaluate the hemolytic potential of the material with rabbit blood
(In vitro hemocompatibility tests ensure that the test material extract does not adversely
affect the cellular components of the blood.)
Such biocompatibility tests are only a guideline, and more extensive and specific testing
may be required for certain device manufacturers. The extent and nature of the testing can
also vary from adhesive supplier to adhesive supplier.
Another important criterion to recognize for medical device adhesives is the resistance to
sterilization. Most disposable and reusable medical devices go through some type of
EPOXY ADHESIVES 17
sterilization process prior to use. Some products (endoscopes, surgical instruments, etc.) may
be subjected to multiple sterilization cycles, and the adhesive must resist these processes and
continue to perform its primary function.
The most common sterilization processes are gaseous ethylene oxide (EtO), gamma radia-
tion, and autoclave sterilization. The resistance of adhesives to these processes will be depen-
dent on the conditions of the sterilization cycle as well as on the chemistry of the adhesive.
Autoclave sterilization is one of the most difficult common sterilization environments for
a medical adhesive, and it is commonly used in hospitals and health care facilities for reusable
devices. Autoclaves sterilize with high-pressure steam. Temperatures inside the sterilization
chamber typically can reach 130°C with pressures above ambient. Certain adhesive systems,
such as polyurethanes, may show hydrolytic degradation in such environments especially after
multiple cycles. Epoxies perform the best under multiple autoclave exposures. However, on
certain substrates, light-cured acrylics and cyanoacrylates will also perform fairly well.
Epoxies cure to a thermosetting molecular structure. As a result, they provide optimum
thermal, chemical, and environmental resistance compared to the other medical adhesives.
Epoxy adhesives can be used at temperatures over 150°C, and they are resistant to most flu-
ids and chemicals. Depending on their formulation, epoxy adhesives can vary significantly
in modulus and toughness. However, most epoxy systems are relatively rigid when cured
so that their peel strength is less than those of adhesives with a greater degree of elongation
(e.g., polyurethanes).
Both room temperature and heat-curable epoxy adhesives are used in medical applica-
tions. The room temperature curing systems require metering and mixing, and the cure time
is generally slow (several hours). Elevated-temperature cures could affect temperature-
sensitive substrates.
Epoxies can cure in deep sections and are useful in potting and deep-section sealing
applications. They adhere well to different substrates and therefore are used in the general
assembly of many medical devices. A clear, medical-grade, low-viscosity epoxy adhesive
has proved useful in the fabrication of access ports that are implanted beneath the skin of
patients who require multiple infusions.21
Aircraft and Aerospace. Adhesives have always played a significant role in the aircraft
and aerospace industries primarily because they offer a low-weight, fatigue-resistant, and
aerodynamically sound method of assembly. Adhesive bonding is also less labor- and cost-
intensive when applied to large structures such as those commonly utilized in the aerospace
industry. Structural adhesives account for the greatest market share of all the adhesives
used in aerospace applications.
In the aerospace market a distinction is made between primary and secondary structural
applications. Joint failure in a primary structure will result in the loss of the aircraft, whereas
failure in a secondary structure will result in only localized damage. Structural adhesives are
used in both applications. Figure 1.4 illustrates the degree to which adhesive bonding is used
in modern aircraft.
The primary types of epoxy adhesives that are used in the aerospace industry are struc-
tural film adhesives and liquid and paste adhesives. The film applications include metal-to-
metal and honeycomb bonding. Advanced composite-to-composite bonding using film
adhesive is gaining in significance because of the rapid growth rate and use of these materi-
als. Liquid and paste epoxy adhesives are used for a variety of assembly operations including
liquid shimming and aircraft maintenance and repair.
The early aerospace adhesives were primarily based on epoxy resin chemistry. However,
unique applications requiring high temperatures and fatigue resistance have forced the devel-
opment of epoxy-phenolic, epoxy-nitrile, epoxy-nylon, and epoxy-vinyl adhesives specifically
for this industry. The aerospace industry has led in the development and utilization of these
epoxy-hybrid adhesives.
18 CHAPTER ONE
Laminated edge
cover panels
Spoiler
Tear straps
Lap splice
Flaps Wing trailing
edge cover
panels
Strut fairing
Cowl panels
Bonded area
Over the past several decades, significant advances have been made in developing epoxy-
based adhesives having improved performance over these early adhesive systems. These
improvements were made possible by (1) the incorporation of toughening additives into
epoxy resin formulations and (2) the use of multifunctional epoxy resins primarily for high-
temperature applications. These innovations are discussed in later chapters.
Structural adhesive bonding in the aerospace industry has primarily been through the
initiative and development of applications for military. For example, in 1975 the U.S. gov-
ernment sponsored a program known as the Primary Adhesively Bonded Structures
Technology (PABST) to extend the use of adhesive bonding for aerospace and to optimize
manufacturing. Furthermore, many of the standards and specifications that are used in the
industry come from military and/or government sources. These include Mil-A-25463
(bonding honeycomb structures) and MMM-A-132 (bonding metal-to-metal airframe
structures). These standards define the physical (temperature, fatigue, etc.) and chemical
environments (e.g., salt spray, humidity, jet fuel, hydraulic oil) to which aerospace adhe-
sives will be exposed and their minimum requirements.
The adhesive formulator has many challenges in meeting the demands of the marketplace,
and many of the tools that the formulator had utilized in the past are no longer acceptable.
These internal and external pressures place the adhesive formulator in a special place among
other professions.
EPOXY ADHESIVES 19
At first glance, the job of the adhesive formulator appears deceptively simple. There are
many resources on the supply side of the industry for the formulator to use. The problem is
that there seems to be an even more staggering amount of requests for new products on the
demand side of the industry.
The formulator has the unenviable task of responding to the latest issues impacting the
end users of adhesives and devising the technology and resources to come up with practi-
cal solutions. It may seem that as soon as a product is developed, tested, and commercial-
ized, the demands change. This is, of course, not necessarily the fault of the end user. There
are many outside influences that affect the end users’ processes, just as there are many
influences affecting the type of raw materials and equipment that can be utilized for for-
mulating adhesives and sealants.
This job has not gotten any easier in the last few years. Substrates continue to undergo
significant transformations. Today’s formulator is faced with the challenge of adhering to
surfaces with a broad range of surface energies and to products that have been developed and
characterized only recently. A competitive marketplace also necessitates the use of the most
economical substrate available. These substrates may contain extenders and cost-reducing
additives or processes that are not compatible with adhesives. Often substrate changes are
made without notice to the end user or adhesive formulator who must work together for an
assembly solution.
In an age of increasing environmental concerns, many of the tools that formulators had
utilized in the past are no longer acceptable. Use of solvents has been drastically curtailed.
Plasticizers must often be used in limited amounts. Regional regulations place demands
that total VOC emissions be kept to a minimum. Certain catalysts and additives are dis-
couraged because of safety or health issues.
Another megatrend in this competitive business environment is that all production
processes need to be “world-class” and “lean.” This means that processes must be automated;
lead times shortened; rework, scrap, and waste minimized or eliminated; and less product must
appear in inventory. This places a burden on the adhesive formulator in that the products not
only must perform in service (e.g., creep resistance, peel strength) but also must “perform” in
the assembly process (e.g., fast cure time, minimal scrap, low energy cure).
The job of the adhesive formulator has been made particularly difficult by the lack of
practical information on the topic. There are only several forums that provide an introduc-
tion to adhesive formulating. One such forum is the Adhesive and Sealant Council, which
offers short courses on adhesive formulating. There are also a few books that offer infor-
mation and guides on adhesive formulations.23–27 Information specific to epoxy adhesives
is usually found in a chapter or section within the work. There has been no book devoted
solely to epoxy adhesive formulations, although several have focused on the more general
topics of epoxy resins and their applications.28,29
However, valuable information can be gained from literature provided by raw materials
suppliers, who often develop a formulary using their materials. (Many of the formulations
that appear in this book are from such sources.) Several Internet web sites have recently
been established that cater specifically to the adhesive formulator. They provide informa-
tion on raw materials, formulations, and end markets. One of the best of these sites is
SpecialChem4Adhesives.com.
As shown in Table 1.7, the epoxy resin is rarely used unmodified as an adhesive system.
Rather, it is used in the form of a compound containing various modifiers and additives to
improve properties, such as strength, flow, and heat resistance, and to add or advance other
properties that are demanded by the specific application.
20 CHAPTER ONE
Resin content Epoxy resin The bisphenol A type is most common. However, there
are many other types of epoxy resins having differing
properties.
Curing agents and Curing agents and catalysts react with epoxy groups to
catalysts form a three-dimensional network structure by
crosslinking.
Modifying Flexibilizers Elasticity agents flexibilize compounds to improve
ingredients their peel strength, impact resistance, and elongation.
Tougheners Tougheners eliminate brittleness from epoxy resin to
prevent cracks and decrease distortion. Unlike
flexibilizers, they do not reduce crosslink density.
Heat-resistant additives Generally these resins provide increased heat resistance
by nature of their multifunctionality and/or
aromaticity.
Fillers Fillers increase the weight in order to decrease cost
and improve various types of properties (application
and performance).
Diluents Diluents reduce viscosity. Reactive diluents have epoxy
groups, and nonreactive diluents have no epoxy groups
available for reaction.
Thixotropic agents Thixotropic agents impart thixotropy to the adhesive to
control flow and increase viscosity.
Other agents Pigments, coupling agents, defoaming agents, film
formers, etc. are used for specific properties.
The following section offers a brief description of how certain additives and formula-
tion parameters are used to control the characteristics of adhesive systems. The roles of spe-
cific additives are defined in Chaps. 6 through 10.
The customary techniques used by the formulator to overcome several everyday problems
are also discussed below. These techniques include formulation processes and materials to
control flow, extend temperature range, improve toughness, match thermal expansion coeffi-
cients, reduce shrinkage, increase tack, and modify electrical and thermal conductivity.
This description of additives is admittedly generic and brief. There are literally thou-
sands of additives that can be used in adhesive systems. The choice depends on the com-
position of the adhesive system, how it is to be used, system cost, and the properties that
need to be obtained.
by reducing the viscosity of the formulation so that additions of other components and uniform
mixing may be more easily achieved. Solvents used with synthetic resins and elastomers are
generally organic, and often a mixture of solvents is required to achieve the desired properties.
Polar solvents are required with polar resins; nonpolar solvents with nonpolar resins.
When solvents are used in the adhesive formulation, they must be completely evaporated
from the bond line prior to cure, or otherwise bubbles could form in the bond line, causing
a weak joint. The substrate must also be tested so that the solvents that are used in the adhe-
sive formulation do not attack or degrade it.
Water is sometimes used as a solvent for water-soluble resins. Certain epoxy adhesives
are available as water-based emulsion or latex formulations. In the early 1970s, during the
time of the petroleum crisis, water-based adhesives were thought of as a possible replace-
ment for solvent-based adhesives systems. However, water-based adhesives never met the
lofty expectations primarily because of the time and energy required to remove water from
the bond line, the corrosion that the water causes in drying ovens, and the poor moisture
resistance of cured water-based adhesives.
An ingredient added to an adhesive to reduce the concentration of base resin or binder
is called a diluent. Diluents are principally used to lower the viscosity and modify the pro-
cessing conditions of some adhesives. The degree of viscosity reduction caused by various
diluent additions to a conventional epoxy adhesive is shown in Fig. 1.5. Diluents do not
evaporate as does a solvent, but they become part of the final adhesive. Reactive diluents
react with the resin base during cure, so that the final adhesive characteristics are deter-
mined by the reaction product of the binder and diluent. Nonreactive diluents do not react
with the resin or curing agent and, therefore, more seriously weaken the final properties.
Coal and pine tar are common nonreactive diluents.
Extenders are substances that usually have some adhesive properties and are added to
reduce the concentration of other adhesive components and, thereby, the cost of the over-
all formulation. Extenders also have positive value in modifying the physical properties of
800
700
C12 C14 Aliphatic glycidyl ether
600
Butanediol diglycidyl ether
Viscosity, mPa·s
500
Cresyl glycidyl ether
400
Butyl glycidyl ether
300
200
100
0
5 10 15 20 25 30
Diluent, wt%
FIGURE 1.5 Viscosity reduction of a diglycidyl ether of bisphenol A (DGEBA)
epoxy resin by reactive diluents.31
EPOXY ADHESIVES 23
the adhesive. Common extenders are coal tar, soluble lignin, and pulverized partly cured
synthetic resins.
Fillers are relatively nonadhesive substances added to the adhesive formulation to
improve its working properties, strength, permanence, or other qualities. The improve-
ments resulting from the use of fillers are listed in Table 1.8. Fillers are also used to reduce
material cost. By selective use of fillers, the properties of an adhesive can be changed sig-
nificantly. Thermal expansion, electrical and thermal conduction, shrinkage, viscosity, and
thermal resistance are only a few properties that can be modified by the use of fillers.
Common fillers are wood flour, silica, alumina, titanium oxide, metal powders, china clay
and earth, slate dust, and glass fibers. Some fillers may act as extenders.
A carrier or reinforcement is usually a thin fabric, cloth, or paper used to support the
semicured adhesive composition to provide a tape or film. In tapes, the carrier is the back-
ing on which the adhesive is applied. The backing may be used for functional or decorative
purposes. In epoxy films or structural tape, the carrier is usually porous and the adhesive
saturates the carrier. Glass, polyester, and nylon fabric are common carriers for supported
B-staged epoxy adhesive films. In these cases, the carrier provides for a method of apply-
ing the adhesive and also may act as reinforcement and a internal “shim” to control the
thickness of the adhesive.
Plasticizers and flexibilizers are incorporated into an adhesive formulation to provide it
with flexibility and/or elongation. Plasticizers may also reduce the melt viscosity of hot
melt adhesives or lower the elastic modulus of a solidified adhesive. Similar to diluents,
plasticizers are nonvolatile solvents for the base resin, and by being incorporated into the
formulation, they separate the polymer chains and enable their deformation to be more eas-
ily accomplished. Plasticizers generally affect the viscoelastic properties of the base resin
whereas diluents simply reduce the viscosity of the system. Whereas diluents result in brit-
tle, hard adhesive systems, plasticizers result in increased flexibility and lower modulus.
The temperature at which polymers exhibit rubbery properties (i.e., the glass transition tem-
perature) can also be modified by incorporating plasticizers.
Certain resinous materials that act as plasticizers are well noted for increasing the tack of
the formulation. Traditional tackifiers were based on naturally occurring resins such as pine
tar. Today, tackifiers used in modern adhesive formulations include aliphatic and aromatic
Material cost
Flow properties (viscosity, thixotropy)
Bond line thickness
Coefficient of thermal expansion
Shrinkage
Conductivity (electrical and thermal)
Electrical properties
Specific gravity
Mechanical strength properties
Heat and chemical resistance properties
Adhesion properties
Working life and exotherm
Fire resistance
Color
24 CHAPTER ONE
hydrocarbons, terpenes, and rosin esters. Tackifiers are useful in pressure-sensitive adhesives or
adhesives that require aggressive tack or “green strength” to assist in assembly of the product.
In addition to increasing tack, increased peel strength and decreased shear strength also result
from the addition of tackifiers in the adhesive formulation.
Thickeners and thixotropic agents are used to maintain reasonable thickness of the glue
line through viscosity adjustment. Thixotropic fillers are materials which when added to the
adhesive increase the viscosity when it is under rest (i.e., not undergoing stress). Therefore,
thixotropic agents provide sag resistance and the ability for an adhesive to remain in place
on a vertical substrate. However, when a slight force is applied, such as in the act of stir-
ring or extruding the adhesive, the system then acts as if it had a lower viscosity and flows
with relative ease.
Thickeners, fillers, and thixotropic agents are also commonly used to control flow and
the bond line’s thickness within a joint. Scrims, carriers, and woven reinforcements are
other methods commonly used to control bond line thickness.
Many polymers have a limited life and are subject to aging processes even before they
are used in production. To delay these aging processes to a usable time period, antioxidants,
antihydrolysis agents, and stabilizers are added to adhesive formulations. Antifungal
agents or biocides are used in many water-based adhesive systems.
Aqueous suspensions comprise a wide range of adhesives. These will contain as addi-
tives the various soaps, surfactants, and wetting agents necessary to stabilize the emulsion
or latex. Additives are also incorporated into aqueous formulations to provide system sta-
bility under repeated freeze-thaw cycles during storage.
The removal of solvent from a true solution of a polymer leaves a film of the polymer. If
the same polymer is available as an emulsion, removal of the liquid by evaporation does not
necessarily leave a coherent film. The individual globulars of polymer will merge to form a
film only if the polymer is well above its glass transition temperature. Drying of some emul-
sions will produce a powder unless its glass transition temperature has been first lowered by
the addition of a plasticizer or film former to the emulsion. Film formers are carefully chosen
to lower the glass transition temperature without appreciably lowering the strength of the film.
REFERENCES
1. Austin, J. E., and Jackson, L. C., “Management—Teach Your Engineers to Design Better with
Adhesives,” SAE Journal, October 1961.
2. The Long and Short of Epoxy Resins, Shell Chemical, Houston, TX, 1992.
3. Nick, D. P., The World of Epoxy Adhesives, TRFA Annual Meeting, Philadelphia, PA, November
11, 2003.
4. Michaud, P., “Epoxy Resins in Composites,” JEC—Composites, no. 12, October-November 2004,
pp. 24–25.
5. Freedonia Group, Cleveland, OH, “Epoxy Resins Demand to Reach 690 Million Pounds in 2004,”
Paint and Coating Industry, June 2000.
6. Lee, H., and Neville, K., Handbook of Epoxy Resins, McGraw-Hill, New York, 1967.
7. May, C. A., and Tanaka, Y., eds., Epoxy Resins: Chemistry and Technology, Marcel Dekker, New
York, 1973.
8. Oldring, P. K. T., ed., Waterborne and Solvent Based Epoxies and Their End User Applications,
2d ed., vol. 2, SITA Technology Ltd., Edinburgh, UK, 1999.
9. Schields, J., Adhesives Handbook, 3d ed., Chapter 1, Butterworths, London, 1984.
10. Smith, R., “The U.S. Epoxy Adhesives Business,” Composites News, June 30, 2004.
11. Nick, The World of Epoxy Adhesives.
12. Nick, The World of Epoxy Adhesives.
EPOXY ADHESIVES 25
2.1 INTRODUCTION
Proper formulation of epoxy adhesives requires knowledge of the chemical reactions that
lead to polymerization as well as the chemical and physical properties of both the uncured
mixture and the cured material. This chapter reviews the general principles of epoxy resin
chemistry including synthesis of the epoxy monomer itself and its possible polymerization
reactions.
The general properties of cured and uncured epoxy resins are reviewed in Chap. 3. The
chemical structures of the resin and curing agent will determine these physical properties.
They will also determine, to a great extent, the surface chemistry and adhesion properties
of the final product.
The term epoxy, epoxy resin, or epoxide (Europe) refers to a broad group of reactive com-
pounds that are characterized by the presence of an oxirane or epoxy ring, shown in Fig. 2.1.
This is represented by a three-member ring containing an oxygen atom that is bonded with
two carbon atoms already united in some other way.
An epoxy resin can be any molecule containing more than one of these epoxy groups.
The number of epoxy groups per molecule is the functionality of the resin. The groups can
be situated internally, terminally, or on cyclic structures. Epoxy groups are capable of
reacting (1) with suitable curing agents or (2) catalytically (homopolymerized) to form
higher-molecular-weight polymers. Once cured, the epoxy polymers have a densely
crosslinked, thermosetting structure with high cohesive strength and adhesion properties.
However, the term epoxy can also be used to indicate an epoxy resin in the thermoplastic
or uncured state.
A general formula for an epoxy resin can be represented by a linear polyether with
terminal epoxy groups and secondary hydroxyl groups occurring at regular intervals along
the length of the chain. The epoxy resin structure and properties influenced by the various
chemical groups are illustrated in Fig. 2.2.
Several important statements can be made relative to the structure of the epoxy molecule:
• The epoxy groups at both terminals of the molecule and the hydroxyl groups at the mid-
point of the molecule are highly reactive.
27
Copyright © 2006 by The McGraw-Hill Companies, Inc. Click here for terms of use.
28 CHAPTER TWO
Epoxy resins that are commercially produced are not necessarily completely linear or
terminated with epoxy groups. Some degree of branching occurs, with the end groups being
either epoxy or hydroxyl. The amount and degree of branching vary from resin to resin and
from supplier to supplier. Epoxy resins are not completely difunctional. Tri-, tetra-, and
polyfunctionality are possible. Various end groups can be introduced as a consequence of
the manufacturing process.
The ratio of the main ingredients used in the synthesis of epoxy resins (epichlorohydrin:
bisphenol A) determines the extent of the reaction and the molecular weight (or value of n
repeating units in the molecular chain). The addition of bisphenol A to the reaction mix will
advance the molecular weight of the resin and the value of n. As n increases, the viscosity
or melting point of the resin also increases. Also as the value of n increases, the number of
hydroxyl groups increases while the number of epoxy groups remains constant.
An important term that is used in formulating epoxy adhesive compositions is epoxy
equivalent weight (EEW). This is defined as the weight of resin in grams that contains one
equivalent of epoxy. As the resin’s molecular weight increases, the EEW will also increase.
CH3
CH3
O C O CH2 CH CH2
CH3 n O
Chemical resistance
FIGURE 2.2. The structure and properties of an epoxy resin.1
EPOXY RESIN CHEMISTRY 29
6000
4000
3000
0 1.0 2.0 3.0
Hydroxyl content, (mol/L)2/3
FIGURE 2.3 Relation between the adhesive strength
of epoxy resins and their hydroxyl content.2
If the resin chains are assumed to be linear with no side branching, and it is further assumed
that an epoxy group terminates each end of the molecule, then the epoxide equivalent
weight is one-half the average molecular weight of the epoxy resin.
The use of EEW in determining curing agent concentrations is described in Sec. 2.5.
Generally, the higher the molecular weight and EEW of the epoxy resin, the less curing
agent is required because there are less functional epoxy groups per unit weight of resin.
There are many terms used to refer to the number of epoxy groups per molecule in addi-
tion to EEW, and as a result there is a lot of confusion around this issue. Other terms that
can be used interchangeably with EEW and are numerically equivalent are weight per
epoxy (WPE) and epoxy molar mass (EMM). Sometimes the term epoxy value or epoxy
group content is also used. This is the fractional number of epoxy groups per 100 g of resin.
Dividing the epoxy value into 100 gives the epoxide equivalent weight.
The use of the expression epoxy group content (EGC), expressed in millimoles per kilo-
gram (mmol/kg), is a somewhat more recent trend. Note that as the EGC increases, the mol-
ecular weight decreases. This is in contrast to the other expressions used. In summary,
10 6
EEW = EMM = WPE =
EGC
Most of the epoxy resins that are used in the formulation of adhesives have EEWs in the
range of 180 to 3200, corresponding to a molecular weight range from 250 to 3750. This
applies equally to epoxy resins that have undergone chemical reaction or modification to
be used as a base resin in an adhesive formulation.
Similarly, the hydroxyl equivalent weight is the weight of the resin containing one
equivalent of hydroxyl group, and it may also be expressed as equivalents per 100 g.
Several studies have demonstrated that the hydroxyl content greatly influences the adhe-
sion of the epoxy resin.3,4 These results are best explained by the polar character of the
30 CHAPTER TWO
hydroxyl groups. Figure 2.3 shows the relation between adhesive strength and hydroxyl
content for a series of epoxy resins bonding stainless steel substrates.
Epoxy resins were first introduced by Ciba-Geigy, Ltd., in 1946 although their develop-
ment began decades earlier. The most common commercial epoxy resin is formed from the
reaction of bisphenol A and epichlorohydrin. This resin is known as the diglycidyl ether of
bisphenol A or DGEBA (Fig. 2.4).
The addition of NaOH catalyzes the reaction to produce the chlorohydrin intermediate,
acts as the dehydrohalogenating agent, and neutralizes the formed HCl. The reaction
generally occurs over 16 h at 110°C. The caustic is added slowly as a 30% aqueous solution.
The resulting organic is separated, dried with sodium sulfate, and fractionally distilled under
vacuum.
The reaction is always carried out with an excess of epichlorohydrin so that the result-
ing resin has terminal epoxy groups. By varying the manufacturing conditions and the con-
centration of epichlorohydrin, resins of low to high molecular weight (MW) (low to high
value of n) can be produced. If a large excess of epichlorohydrin (10 to 20 mol per mole of
bisphenol A) is used, the likelihood of producing a low-molecular-weight resin, where n is
near to 0, is considerably improved. Table 2.1 shows the effect of varying reactant ratios
on the molecular weight of epoxy resins. Very low-molecular-weight epoxy resins can be
obtained by fractionation (distillation) of the liquid epoxies.6 As n increases, the molecular
weight increases, and the resin moves from a low-viscosity liquid to a semisolid and then
to a brittle solid. When n is between 0 and 1, the product is a liquid; when n is greater than 1,
it is a solid. Solid resins generally need to be melted or dissolved in solvent prior to com-
pounding or processing.
Commercially useful grades of DGEBA are relatively low-MW weight resins in which
n ranges from 0 to 4, although n could be as high as 20. Most commercial resins are mix-
tures of polymers so that the value of n is an average for a given product. Typical proper-
ties of the DGEBA resins are shown in Table 2.2. Note that as n increases, the epoxy
content is decreased and the hydroxyl content is increased.
CH3
O
HO NaOH
C OH + 2CICH2CH CH2
CH3
CH3
−HCI
CICH2CHCH2O C OCH2CHCH2CI
OH CH3 OH
CH3 OH CH3
O
CH2 CH CH2 (O C O CH2 CH CH2 ) O C O CH2 CH CH2
n
CH3 CH3
FIGURE 2.4 Synthesis of DGEBA epoxy resin from bisphenol A and epichlorohydrin.5
EPOXY RESIN CHEMISTRY 31
TABLE 2.1 Effect of Varying Reactant Ratios on Molecular Weight of Epoxy Resins7
The low-MW, liquid epoxy resins are generally the most desirable to use in adhesive
formulations because of easier compounding capability and application via liquid dis-
pensers. They also have relatively high reactivity and crosslink density. The high concen-
tration of epoxy rings per unit of mass provides the high reactivity. This allows the resin to
be cured at low temperatures and with conventional curing agents. The crosslink density
results in a material with a high glass transition temperature having good heat, chemical,
and solvent resistance.
Solid, higher-MW epoxy resins are often used for adhesive formulations that are applied
as solids (e.g., film and powder) or a solvent solution. The higher-MW DGEBA resins are
also used where improved toughness, flexibility, and adhesion are required. These resins
have a greater number of hydroxyl groups along the chain and, thus, can provide better
adhesion and additional reaction mechanisms.
Extremely high-MW (MW equal to about 30,000 and n equal to about 100) DGEBA
resins have also been produced. These resins are known as phenoxy compounds. They are
not necessarily epoxy-terminated, and when they are epoxy-terminated, they have a very
low epoxy equivalent weight. Phenoxy resins are essentially linear molecules that are ther-
moplastics. However, they can be crosslinked through their epoxy groups and, more con-
veniently, through the hydroxyl groups (through reaction with isocyanates or melamine
formaldehyde resins) along the molecular chain. These hydroxyl groups also provide excel-
lent adhesion to many polar surfaces. In adhesive formulations, phenoxy resins can be
processed into powders, films, and solvent solutions. Phenoxy resins have considerable
mechanical strength by themselves and are often used as unformulated thermoplastic hot
melt adhesives.
In addition to the DGEBA resins, there are several other types of epoxy resins of commer-
cial significance. The most common of these are epoxy novolacs, glycidyl ether of tetraphe-
nolethane, bisphenol F–based resins, and aliphatic and cycloaliphatic resins.
Epoxy Novolac and Other Phenols. Resins of greater functionality than DGEBA can be
produced in several ways. Polyols having more than two hydroxyl groups per molecule
(e.g., phenol novolac resins) can be reacted with epichlorohydrin to produce epoxy novolac
resins with a structure shown in Fig. 2.5.
The epoxy novolac resins are synthesized by reaction of phenolic or cresol novolacs
with epichlorohydrin in the same fashion as the bisphenol A resins. The number of epoxy
groups per molecule is dependent on the number of hydroxyls in the phenol novolac mol-
ecule and to the extent to which they are reacted. Complete epoxidation can be accom-
plished, but this will lead to steric factors, which could limit the useful size of the cured
polymer. Thus, selective epoxidation is often practiced.9
The epoxy novolacs are high-viscosity liquids or semisolids, and they are often
mixed with other epoxy resins to improve handling and formulation properties. They
cure more rapidly than DGEBA epoxy resins and have higher exotherms. The epoxy
novolac resins have much greater functionality (generally 2.3 to 6.0) than conventional
DGEBA resins, and the increased crosslink density results in greater temperature and
chemical resistance.
The inherent thermal stability of the phenol formaldehyde chemistry is preserved but
with the crosslinking characteristics of the epoxy groups. However, epoxy novolacs also
form very rigid and brittle polymers when fully cured because of their high crosslink den-
sity. For this reason, they are often used as modifiers in epoxy adhesive systems rather than
as the base polymer.
O O O
CH2 CH CH2 CH2 CH CH2 CH2 CH CH2
O O O
CH2 CH2
O O
HC CH
O O
Bisphenol F Resins. Diglycidyl ether resins based on bisphenol F (DGEBF) have been
developed to provide cured epoxy resins with greater flexibility and lower softening tem-
peratures than conventional DGEBA epoxy resins. The preparation of bisphenol F resins is
from formaldehyde and phenol. Three isomers are possible because substitution can occur
at the ortho-, meta-, or para- positions. The proportion of isomers depends upon the pH of
the reaction medium.
The DGEBF resins have much lower viscosity (generally an order of magnitude) than
their DGEBA counterparts for the same n value. The lower viscosity of DGEBF resins pro-
vides for greater latitude in formulation. The major disadvantage of bisphenol F epoxy resins
is their significantly higher cost than DGEBA resins.
Bisphenol F–based epoxy resins (Fig. 2.7) are analogous to DGEBA-based epoxy resins
in most respects. They use the same curing agents and reaction mechanisms. Bisphenol F
epoxies are often used in blends with DGEBA resins to lower the viscosity or to modify
certain properties.
Aliphatic and Cycloaliphatic Resins. Aliphatic and cycloaliphatic epoxy resins have been
produced from the epoxidation of olefinic compounds. The epoxidation process involves
the use of an olefinic or polyolefinic compound and a peracid (e.g., peracetic acid) or other
H H
O O
O O
FIGURE 2.7 Chemical structure of bisphenol F resin
(4,4 isomer).
34 CHAPTER TWO
oxidizing substances such as hydrogen peroxide, molecular oxygen, or even air. Several
common cycloaliphatic epoxy resin structures are illustrated in Fig. 2.8.
Two types of epoxy resins are formed by this process: (1) cycloaliphatic resins and
(2) aliphatic resins. Of the many structures that can be synthesized by this process, the
cycloaliphatic diepoxies offer the most interesting combination of properties. However,
the aliphatic epoxy resins have the greatest utilization in epoxy adhesive formulation.
The cycloaliphatic epoxy resins are characterized by the saturated ring in their chemical
structure. They are almost water-white, very low-viscosity liquids. They provide excellent
electrical properties such as low dissipation factor and good arc-track resistance, good weath-
ering, and high heat distortion temperature. They are also free of hydrolyzable chlorine,
sometimes present in DGEBA resins, which adversely affects certain electronic applications.
Because of their low viscosity, cycloaliphatic epoxies are often used to dilute other
epoxy resins. These resins, however, have not achieved general importance in adhesive for-
mulations because of relatively low tensile strength and because they do not cure well at
room temperature. One major application for cycloaliphatic epoxies, however, is for adhe-
sives and coatings that can be cationically cured by exposure to uv light.
The aliphatic epoxy resins formed from reaction with hydrogen peroxide or peracetic
acid include epoxidized polybutadiene, epoxidized soya or linseed oil, and epoxidized
polyglycols. The resulting products have too low a functionality for use as base polymers.
They are almost always used in combination with other epoxy resins to improve properties
such as cure rate, flexibility, and heat deflection temperature. Therefore, these resins are
often considered to be reactive diluents and flexibilizers.
Glycidyl Ethers of Aliphatic Polyols. Glycidyl ether epoxy resins based on polyols pro-
vide greater flexibility and lower softening temperatures for the final cured epoxy system.
The polyol is reacted with epichlorohydrin to produce these resins. These resins are gener-
ally not used alone because of water sensitivity and overall lack of toughness. However,
they serve as modifiers for DGEBA-based epoxy resins. An idealized structure for a flexi-
ble resin based on this chemistry is shown in Fig. 2.9.
Special grades of glycidyl ether epoxies have been made with components such as glyc-
erol, polyglycols, pentaerythritol, and cashew nut oil. Other epoxy-polyglycidol resins
have been produced from the reaction of epichlorohydrin and polyester polyol based on
C O CH2
O O
O O
CH3 CH3
CH CH2
O
O R R O
CH2 CH CH2 O CH2 CH O CH2 CH O CH2 CH CH2
n
R-an organic group or hydrogen
FIGURE 2.9 An idealized structure of a flexible epoxy resin made from a polyol.11
ethylene or propylene oxide. These resins are produced by the reaction of epichlorohydrin
with the polyol in the presence of a catalyst, similar to the phenolic-based glycidyl ethers.
The resins based on glycerol and pentaerythritol are water-soluble and have low viscosity.
They can have greater functionality and reactivity than conventional DGEBA resins. Resins
based on polytaerythritol are claimed to have excellent adhesive properties including the abil-
ity to adhere to wet surfaces. They cure between 2 and 8 times faster than DGEBA epoxy resins
and reduce the viscosity of DGEBA by 50 percent when used in concentration of 20 pph.
The polyglycol diepoxies are used primarily as flexibilizers and reactive diluents.
Commercial products are available where n varies from 2 to 7. Generally, these are used in
the range of 10 to 30 percent by weight with DGEBA or epoxy novolac resins to improve
flexibility without a significant loss in physical properties.
Other products that may be epoxidized in this way include dihydric and trihydric phe-
nols, aliphatic polyols such as glycerol, and simple alcohols such as butanol or alyl alcohol.
These products, especially the monofunctional glycidyl ethers, are used at relatively low
percentages to reduce the viscosity of formulations containing DGEBA resins. In this way,
they act as reactive diluents.
Brominated Epoxy Resins. The conventional DGEBA epoxy resins are flammable when
cured. In an adhesive, flammability is generally not considered critically important because
the mass of adhesive in any one area is relatively small. However, in certain applications
(printed-circuit manufacture, aircraft interiors, furniture, etc.) nonflammability is an impor-
tant criterion. Flame-retardant additives and chlorinated curing agents have been used to
impart nonflammability to epoxy resins.
However, the optimum degree of flame resistance can only be achieved by making the
resin itself nonflammable. A halogenated form of bisphenol A, such as tetrabromobisphe-
nol A, has accomplished this. This resin is formed from the reaction product of epichloro-
hydrin and brominated bisphenol A. These resins acquire their fire-retardant characteristics
through bromine substitution on the phenyl rings of the bisphenol A.
Tetrabromobisphenol A epoxy resins are available as viscous liquids with several mol-
ecular weight ranges. As the bromine content increases, the flame resistance increases but
the viscosity of the resin generally increases as well. They are primarily used as an additive
in formulations producing epoxy laminates and adhesives that require improved resistance
to ignition.
Epoxy Adducts. The epoxy resin itself may be reacted with di- or polyfunctional reac-
tants containing more than one epoxy group. The number of potential epoxy adducts is very
large. Typical are phosphorous-containing epoxy resins prepared by reaction of diepoxies
with acid dichlorides of phosphoric acids. This is accomplished at an acid : diepoxy ratio of
1:2 in the presence of FeCl3 catalyst. Adducts may also be prepared by the reaction of chain
hydroxyls.
Epoxy-amine adducts are produced to provide curing agents that have reduced skin and
eye irritation and better flexibility than unmodified amines. These epoxy amines are formed
by the reaction of either a primary or secondary amine with an epoxy resin. They are more
fully described in Chap. 5.
36 CHAPTER TWO
The epoxy resins are capable of reacting with various curing agents or with themselves (via
a catalyst) to form solid, crosslinked materials with considerable strength and adhesion. This
transformation is generally referred to as curing or hardening. This ability to be transformed
from a low-viscosity liquid (or thermoplastic state) into a tough, hard thermoset is the most
valuable single property of epoxy resins.
This transformation or conversion is accomplished by the addition of a chemically active
compound known as a curing agent or catalyst. Depending on the particular details of the
epoxy formulation, curing may be accomplished at room temperature, with the application
of external heat, or with the application of an external source of energy other than heat such
as ultraviolet (uv) or electron beam (EB) energy.
Two primary types of epoxy curing reactions are discussed in this section:
1. Polyaddition reactions
2. Homopolymerization reactions
When the curing is based on the reaction of the epoxy molecule and other kinds of reactive
molecules with or without the help of a catalyst, the reaction is an addition reaction. The
cured structure is a heteropolymer that is composed essentially of the epoxy resin mole-
cules linked together through the reactive sites of the curing agent molecules. The curing
agent can be thought of as a comonomer in the polymerization reaction.
The compound that reacts with the epoxy resin and actually forms part of the final epoxy
network is generally known as a curing agent or hardener. It is usually a polyfunctional
coreactant, so that stoichiometric amounts need to be considered when one is developing a
formulation. The curing agent makes up a significant portion (several percent to over
50 percent by weight) of the epoxy formulation.
Significant latitude in mix ratios can be provided with most curing agents because they
are locked within the polymeric structure after cure. As a result, the type and concentration
of curing agent have a significant effect on both the curing chemistry and the final proper-
ties of the cured resin. Curing agents that have flexible molecules between reactive groups
will provide tough, flexible cured adhesives.
The polyaddition reaction is the most commonly used type of reaction for the cure of
epoxy resins. The curing agents used in this type of reaction have an active hydrogen com-
pound, and they include amines, amides, and mercaptans. With this reaction mechanism,
the most important curing agents for adhesives are primary and secondary amines contain-
ing at least three active hydrogen atoms and various di- or polyfunctional carboxylic acids
and their anhydrides.
A generalized polyaddition reaction of an active hydrogen compound and an epoxy mole-
cule is represented in Fig. 2.11. Hydroxyl groups, especially phenolic hydroxyls, and tertiary
amines catalyze this reaction. The bulk and nature of substituent groups on the epoxy and cur-
ing agent molecules also play a major role in the reaction rate. Thus, low-molecular-weight cur-
ing agents and resins are more reactive and produce more densely crosslinked structures than
high-molecular-weight curing agents and resins. This is why low-molecular-weight DGEBA
resins will react much more readily at room temperature than will epoxy novolac resins.
The exotherm caused by reactive, low-MW resins and curing agents is generally much
more severe than with high-MW systems. Although the exotherm may result in a faster reac-
tion, it could also result in an objectionably short working life and even a dangerous situation
if the exothermic temperature rises above the flame point of the resin. Thus, the formulator
will often use steric hindrance as a tool for controlling exotherm to provide systems that will
cure readily at elevated temperature but will react sluggishly at room temperature.
O
H
RNH2 + CH2 CH C RN
CH2 CH C
OH
OH
O
H CH2 CH C
RN + CH2 CH C RN
CH2 CH C CH2 CH C
OH OH
FIGURE 2.11 Polyaddition reaction of an epoxy resin with an active hydrogen compound.
38 CHAPTER TWO
O
O O CH2 CH C
BF3
n CH2 CH C n−1
R3N
CH2 C C
FIGURE 2.12 Homopolymerization reaction of an epoxy resin.
When the curing process is based on the reaction between the reactive epoxy molecules
only, the reaction is homopolymerization (see Fig. 2.12). The cured structure is essentially
made up only of the original epoxy molecules linked together through their own reactive
sites. The reactive compound that initiates the homopolymerization reactions is generally
known as a catalyst. (Note: Sometimes the term hardener is used interchangeably for cat-
alyst, although this is not the convention followed in this book.)
The catalyst does not make up part of the final epoxy network structure or have a sig-
nificant effect on the final properties of the cured resin. Thus, the final cured properties of
the epoxy system are primarily due to the nature of the epoxy resin alone. Homopolymeri-
zation normally provides better heat and environmental resistance than polyaddition reac-
tions. However, it also provides a more rigidly cured system, so that toughening agents or
flexibilizers must often be used. In adhesive systems, homopolymerization reactions are
generally utilized for heat cured, one-component formulations.
The catalytic curing agents commonly used include tertiary amines, Lewis acids and
bases, and dicyandiamide. Since their function is truly catalytic, the catalyst is added at rel-
atively low concentrations (0 to 5% by weight) to the epoxy formulation. Homopolymeri-
zation generally requires both the presence of catalysts and elevated temperatures for the
reaction to proceed. Like the polyaddition reaction, the homopolymerization reaction is
accelerated by hydroxyl groups or tertiary amines.
Reactions between resin and curing agent or between resin and catalyst are generally
much more complex than the reactions shown above. A number of side reactions can occur,
and these reactions can occur at different rates and reaction sequences depending on the
types of materials employed and the curing conditions. Therefore, a significantly different
chemistry could be achieved by curing an epoxy system at two progressive temperature
stages than by curing at only one stage. These differences in chemistry are, of course, noticeable
in the final properties provided by the cured epoxy structure.
As epoxy resins can vary enormously as suggested above, so can the means for curing
these resins. Curing reactions specific to specific types of curing agents or catalysts are
described in Chap. 5.
2.5 STOICHIOMETRY
With formulated epoxy adhesives, the supplier will generally establish the correct mix ratios
to use for specific end properties and/or production situations. Depending on the resin and
curing agent used, the mixing ratios of these ingredients could vary substantially, or tight
mixing tolerances may be required.
The most important parameters for determining the stoichiometric mixing ratio are the
reactivity (functionality) of the epoxy and the curing agents. The reactivity of the epoxy can
be determined by the epoxy equivalent weight. The amount of curing agent needed to cure
epoxy resins is the same for epoxy resins of similar equivalent weight.
The calculation of stoichiometric quantities of curing agents differs for amine and anhy-
dride curing agents, and these are discussed below. The stoichiometric calculation has no
EPOXY RESIN CHEMISTRY 39
meaning for catalytic curing agents since their action is truly catalytic. Catalytic curing
agents are generally added at low levels (0 to 5%).
Mixing ratios of resin to curing agent do not necessarily have to be stoichiometric ratios (i.e.,
the exact ratio depending on chemical functionality of resins and curing agent). With certain cur-
ing agents, especially the higher-molecular-weight flexible curing agents (e.g., polyamides,
amidoamines) and acid anhydrides, the actual stoichiometric equivalency does not always pro-
duce a cured adhesive with optimum properties. The balance between curing agent and epoxy
resin will affect the toughness and high-temperature strength of the adhesive, the former being
favored by low degrees of crosslinking and the latter by high degrees of crosslinking.
When one is using polyamide curing agents, for example mixing ratios that employ cur-
ing agents at a greater than stoichiometric concentration will provide flexibility and elon-
gation to the resulting adhesive. This is a result of the higher concentration of the flexible
polyamide molecule between epoxy linkages. Employing less than the stoichiometric con-
centration of the curing agent will result in a more brittle system, but it will have greater
tensile strength and heat and chemical resistance.
When one is using primary amines, for example diethylene triamine (DETA), the mix
ratio should be closer to stoichiometric amounts. However, there is always a percentage of
homopolymerization in practice, especially at the temperature of reaction. Smaller amounts
of DETA than stoichiometric, therefore, will cause a complete cure. But this will generally
occur at the expense of increasing brittleness.
Petterson13 has shown that the use of only 50 percent of the stoichiometric amount of
hexamethylenediamine imparts to a cured DGEBA adhesive the maximum butt-joint strength,
bulk tensile strength, and flexural moduli, whereas higher proportions of the diamine give
lower properties.
In the case of the adhesive supplier or the manufacturing engineer who finds it neces-
sary to internally formulate an epoxy adhesive, stoichiometric ratios will need to be deter-
mined so that one can find a safe mixing ratio to start formulating and to understand the
implications of various reactive proportions.
The amount of curing agent selected should be fine-tuned for the specific application.
This can be done by varying the concentration over a narrow range and then testing per-
formance to find which concentration is optimal. The optimized concentration should then
be tested in the actual product and under end-use conditions since there may be other vari-
ables that affect performance (i.e., the mass and geometry of the joint, cure conditions,
combinations of environmental conditions).
Of course, in the case of both curing agents and catalysts, suitable adjustments will have
to be made for the presence of nonreactive fillers and modifiers. Such ingredients can be
liquids such as a solvent, a hydrocarbon resin, or a plasticizer. Since they do not contribute
any epoxide functionality, they should not be considered when one is determining
stoichiometry. However, if the additives have epoxy functionality, such as in the case
of reactive diluents, the stoichiometric calculations will have to take these materials into
consideration, by calculating ratios similarly as with an epoxy resin.
Amines are one of the most important curing agents for epoxy resins. They provide fast
cures with a relatively high crosslink density. Unmodified amine cured epoxy resins are
generally too brittle for adhesive applications, and so there are many derivatives that have
been developed. More information on amine curing agents can be found in Chap. 5.
The stoichiometric quantity of amine to be used to cure an epoxy resin is easily calcu-
lated from the molecular weight of the amine, the number of active hydrogens in the amine,
and the epoxy equivalent weight of the resin. This calculation is shown in Table 2.3.
When the chemical structure of the amine is not known (as with proprietary curing
agents), the amine equivalent or active hydrogen equivalent weight is generally provided
40 CHAPTER TWO
103.2
Amine equivalent weight = = 20.6
5
2. Calculate the stoichiometric ratio of amine to use with the epoxy resin.
amine equivalent weight × 100
pph of amine =
epoxy equivalent weight of resin
20.6 × 100
= = 10.9
189
Procedure for calculating the stoichiometric quantity of amine to use with a blend of materials
For example, using the following materials in the formulation of an epoxy adhesive base resin:
100 parts by weight DER 331 Average EEW 189
100 parts by weight DER 337 Average EEW 240
30 parts by weight BGE (diluent) Average EEW 130
230 parts filler No epoxy functionality
1. Calculate the EEW of the total mix.
total weight
EEW of mix =
Wt a /EEWa + Wt b /EEWb + Wt c /EEWc + …
460
= = 391
100/189 + 100/240 + 30/130
20.6 × 100
2. pph of amine = = 5.27
391
2. Calculate the stoichiometric ratio of anhydride to use with the epoxy resin.
by the supplier. Nomographs are also often provided by the resin or curing agent supplier
to simplify the calculation.
After the primary amines, acid anhydrides are the next most important class of curing
agents, although these are not used as often in adhesive systems as they are in casting com-
pounds, encapsulants, molding compounds, etc. More information on anhydride curing
agents can be found in Chap. 5.
The stoichiometric quantity of anhydride to be used to cure an epoxy resin may be
calculated from the molecular weight of the anhydride, the number of anhydride groups per
mole, and the epoxy equivalent weight. An example of this calculation is shown in Table 2.3.
However, it must be recognized that when one is working with anhydrides, one anhydride
group provides two carboxylic acid groups, and it is these carboxylic acid groups that react
with the epoxy resin. Thus, anhydrides have a functionality of 2 in calculating the stoi-
chiometry of the epoxy-anhydride reaction.
Accelerators are often used to increase the reactivity of epoxy-anhydride systems. The
proper amount of cure accelerator is empirically determined and varies according to the
material being used as the accelerator, the anhydride and epoxy resin used, and the desired
cure speed and end properties.
As with amine cured epoxies, frequently the optimum properties of anhydride-epoxy
systems are obtained at other than the calculated stoichiometric amount. This is because
competing reactions occur, such as the reaction of the anhydride group with a hydroxyl
group. Typically, optimum anhydride concentrations occur at 85 to 95% of stoichiometric
concentrations.
REFERENCES
1. Osumi, Y., “One-Part Epoxy Resin,” Three Bond Technical News, Three Bond, October 1, 1987.
2. Dannenberg, H., and May, C., “Epoxide Adhesives,” in Treatise on Adhesion and Adhesives, vol. 2,
R. L. Patrick, ed., Marcel Dekker, New York, 1969.
3. DeBruyne, N. A., Journal of Applied Chemistry, London, vol. 6, 1956, p. 303.
4. May, C. A., “The Physical Significance of Acid Anhydride Cures on Epoxy Adhesive Properties,”
SPE Transactions, vol. 3, 1963, p. 251.
5. Schwartz, S. S., and Goodman, S. H., Plastics Materials and Processes, van Nostrand Reinhold
Company, New York, 1982, p. 351.
6. Dowd, R. T., “General Chemistry of Bisphenol A Based Epoxy Resins,” Chapter 1 in Epoxy Resin
Technology, P. F. Bruins, ed., Interscience Publishers, New York, 1968.
7. Newey, H. A., and Shokal, E. C., Glycidyl Ether Compositions and Method of Using Same, U.S.
Patent 2,575,558.
8. Coyard, H., et al., Resins for Surface Coatings, vol. 1: Acrylics and Epoxies, John Wiley & Sons,
New York, 2001, p. 142.
9. Grover, “New Developments in 100% Solids Epoxy Adhesives,” Society of Plastic Engineers,
15th ANTEC, 1957.
10. Stevens, J. J., “Cycloaliphatic Epoxy Resins,” Chapter 2 in Epoxy Resin Technology, P. F. Bruins,
ed., Interscience Publishers, New York, 1968.
11. Meath, A. R., “Epoxy Resin Adhesives,” Chapter 19 in Handbook of Adhesives, 3d ed., I. Skeist,
ed., van Nostrand Reinhold, New York, 1990.
12. Lee, H., and Neville, K., Handbook of Epoxy Resins, McGraw-Hill, New York, 1967, p. 5.2.
13. Petterson, C. M., Kolloid Z. – Z. Polymers, vol. 222, 1968, p. 148.
This page intentionally left blank
CHAPTER 3
IMPORTANT PROPERTIES
OF EPOXY ADHESIVES
3.1 INTRODUCTION
Physical chemistry is an important factor, which leads to the exceptionally good perfor-
mance properties of epoxy adhesives. Physical chemistry deals with the physiochemical
and surface chemistry aspects of the system. Here system is defined as the adhesive, its con-
stituents, the substrates, and the interface region of the joint. The physical chemistry can
determine the success or failure of an epoxy adhesive application.
These properties have a profound effect on the processing properties of the uncured
adhesive and on the end properties of the fully cured product. The properties determined by
physical chemistry affect both the cohesive strength of the adhesive film as well as the
degree of adhesion to the substrate. They also affect the permanence and durability of the
adhesive bond once it is placed into service.
To understand the overall effect of physical chemistry, one must consider the adhesive
system in three stages:
1. Uncured
2. As it cures
3. When finally cured
In all these stages the adhesive properties will be determined primarily by the molecular
structures of the epoxy resin and curing agent or catalyst employed. Additives and modi-
fiers, of course, will also affect these properties, and they are discussed in later chapters.
All three stages markedly influence overall joint performance. The degree of interfacial
adhesion is greatly determined by the first two stages. The final two stages determine the
degree of cohesive strength within the cured adhesive film. From a practical standpoint, the
last stage is where the ultimate performance of the adhesive or sealant is measured. It
includes aging of the joint “system” in the service environment as well as exposure to any
stresses required of the application.
There are processes going on in all three stages that ultimately will affect the adhesion
and the performance of the adhesive or sealant. It is important to understand these
processes and the effects that they will have on the quality of the joint. They are summarized
in Table 3.1 and described in the sections that follow.
The discussion centers on how the epoxy adhesive properties depend on the processes
that occur during each stage. Special attention is given to the three requisites for good adhe-
sion: wetting, solidification, and avoidance of internal stresses. These are fundamental and
well-known factors that affect the quality of the adhesive bond. The objective here is not to
43
Copyright © 2006 by The McGraw-Hill Companies, Inc. Click here for terms of use.
44 CHAPTER THREE
TABLE 3.1 Important Properties of Epoxy Adhesives during Various Stages of Cure
go into theory, definitions, or detailed dissertations on these subjects, since they are well
covered in many textbooks and essays on adhesive bonding. However, these properties are
the reason that epoxy adhesives have become so well accepted in many markets.
The properties of epoxy adhesives in their uncured condition will determine primarily how
easily the adhesive can be processed, applied, and cured. They will also determine, to some
extent, the performance characteristics of the cured joint. The properties of the individual
components as well as that for the mixed formulation are important.
Properties often used by epoxy resin manufacturers to specify particular grades of resin
include the epoxy content, viscosity or softening point, and color. In addition, properties
such as density, vapor pressure, flash point, refractive index, solubility characteristics, and
hydroxyl content are often reported. The important properties of uncured epoxy formula-
tions with regard to most adhesive applications are
• Vapor pressure
• Viscosity
• Wetting of the substrate (surface tension)
• Reactivity
IMPORTANT PROPERTIES OF EPOXY ADHESIVES 45
These are essential considerations for a practical adhesive system. The adhesive composi-
tion must have sufficient reactivity to cure under conditions that are convenient and prac-
tical to the end user. It must have viscosity that allows easy mixing and application. Once
on the substrate, the adhesive must be able to flow over the substrate and come into inti-
mate contact with it—a process called wetting.
The vapor pressure of an epoxy resin system will have an indirect effect on the final prop-
erties of the cured adhesive. However, vapor pressure of the epoxy adhesive and its com-
ponents may have a direct effect on the health and safety of those who manufacture or apply
these products.
Vapor pressure is defined as the pressure of the gaseous phase of a substance in equi-
librium with its liquid phase. Vapor pressure increases with temperature. When the vapor
pressure equals the surrounding or atmospheric pressure, the component will boil. Thus,
high-vapor-pressure materials will boil at relatively low temperatures, and their vapors are
likely to be present in the surroundings at lower temperatures.
Vapor pressure is largely determined by the molecular weight of the epoxy resin or cur-
ing agent. Low-molecular-weight polymers have a high vapor pressure, and high-molecular-
weight polymers have a low vapor pressure. Vapor pressures of low-molecular-weight
curing agents are important considerations because they are generally higher than those of
epoxy resins. Figure 3.1 shows the vapor pressure for several common amine curing agents
as a function of temperature.
High-vapor-pressure materials are more likely to be present in vaporous form in the sur-
roundings. Thus, health and safety issues may be a concern if these materials are toxic or
cause skin irritation. The occurrence of such events is always exacerbated by high temper-
atures because of the increase in vapor pressure.
In hot mixing or elevated-temperature curing of an epoxy system, vapor pressure could
also be of concern relative to the quality of the adhesive bond. If the components in an
epoxy system become too hot, boiling can occur, resulting in gas bubbles. If gas bubbles
become trapped in the cured adhesive film, they can lead to reduction of cohesive strength
and stress risers. For many adhesive applications, particularly those in the electrical and
electronic industries (due to possible ionization of air voids), complete removal of any gas
bubbles from the epoxy is essential.
Vapor pressure must also be taken into account if vacuum is used to remove entrapped
air from the formulation after mixing. High-vapor-pressure components, such as certain
curing agents, can be evaporated and lost during the vacuum processing operation. The
result can be a mixed formulation with lower than expected curing agent concentration.
3.2.2 Viscosity
Epoxy resins and curing agents must have a relatively low viscosity so that formulation
compounding can be accomplished easily and without a great deal of energy or degra-
dation of the components. Viscosity is defined as the resistance of a liquid material to
flow. It is usually measured in fundamental units of poise (P) or centipoise (cP). Table 3.2
shows a relationship between various common fluids and their viscosity as measured in
centipoise.
Viscosities of liquid epoxy systems are usually measured with a rotating spindle instru-
ment, such as a Brookfield viscometer. Solid resins are usually dissolved in solvent for vis-
cosity measurement by these instruments. Temperature and spindle speed are important
46 CHAPTER THREE
2,000
1,000
500
ine
m
ola
ine
han
200
lam
ine
t
Trie
no
m
yla
ha
no mine
th
let
ine
ie
Vapor pressure, mmHg
100
thy
Tr
am
dia
Die
l
e
eth ylen
ha
ine
50
op
let
lam
y
e
Pr
min
ine
uty
Dim
ola
e
am
Dib
min
an
nol
20
eth
nta
ine
tha
no
epe
TA
olam
die
Mo
DE
ylen
thyl
10
han
eth
Me
Diet
a
Tetr
0
10 20 30 40 50 60 70 80 90 100 120 140 160 180 200 240 280 320 360
Temperature, °C
FIGURE 3.1 Vapor pressure of various amines.1
Viscosity, cP Fluid
1.0 Water
400 #10 Motor oil
1,000 Castor oil
3,500 Karo syrup
4,500 #40 Motor oil
25,000 Hershey chocolate syrup
100,000 Peanut butter
IMPORTANT PROPERTIES OF EPOXY ADHESIVES 47
1. For a given resin or curing agent, viscosity increases in proportion to the high-molecular-
weight species present. Epoxy molecules having n greater than 2 are semisolid or solid.
2. Linear resins or curing agents of the same molecular weight often give higher viscosity
than do branched or cycloaliphatic resins.
3. Aromatic epoxies with three or more groups per molecule are semisolid or solid at room
temperature.
4. Epoxies based on ring structures (cycloaliphatic epoxies) give lower viscosities than
comparable resins based on aromatic rings (DGEBA).
5. In general, the effect of substitution is to decrease the viscosity compared to unsubsti-
tuted product.
The viscosity of an epoxy resin is dependent primarily on its molecular weight (MW).
Low-MW resins typically have a viscosity in excess of 6000 cP, and conventional DGEBA
epoxy resins (EEW = 190) have a viscosity around 12,000 cP. Therefore, for applications
requiring relatively low viscosity, it is necessary to include other types of epoxy resin or to
use diluents to achieve the desired properties.
The viscosity of epoxy resin systems decreases rapidly with temperature, as shown in
Fig. 3.2. Several problems, however, can occur by using heat to reduce the viscosity of the
resin system. The heating can increase the reactivity of the system and, thereby, reduce the
working life. Heating can also increase the exotherm generated by the system on curing.
48 CHAPTER THREE
100
90
80
70
60
B G
Viscosity, P
50
C
F
40
30
E
20
10
0
0 20 40 60 80 100 120
Temperature, °C
FIGURE 3.2 Variation of liquid epoxy resin viscosity with temperature (A, flexibilizing
resin based on dimerized fatty acid; B, liquid diglycidyl ether plus n-BGE reactive diluent;
C, liquid diglycidyl ether plus glycidyl ester of a tertiary carboxylic acid as a reactive dilu-
ent; D, liquid diglycidyl ether plus dibutyl phthalate as a plasticizer; E, unmodified digly-
cidyl ether; F, semisolid diglycidyl ether; G, epoxy novolac).2
IMPORTANT PROPERTIES OF EPOXY ADHESIVES 49
3.2.3 Wetting
Initially, the adhesive must be either a liquid or a readily deformed solid so that it can be
applied and formed to the required geometry within the assembled joint. It is necessary
for the adhesive to flow and conform to the surface of the adherends on both micro- and
macroscales. Small air pockets caused by the roughness of the substrate must be easily dis-
placed with adhesive.
While it is in the liquid state, the material must “wet” the substrate surface. The term
wetting refers to a liquid spreading over and intimately contacting a solid surface, as shown
in Fig. 3.3.
As explained under the adsorption theory of adhesion,3 an adhesive must first wet the
substrate and come into intimate contact with it. (A brief description of the adsorption the-
ory of adhesion is presented in the section below.) The result of good wetting is simply that
there is greater contact area between adherend and adhesive over which the forces of adhe-
sion (e.g., van der Waals type of forces) may act. For good wetting, the surface free energy
(surface tension γLV) of the liquid adhesive must be less than that (critical surface tension γC)
of the solid adherend, or
γLV < γC for good wetting
Table 3.3 provides surface tensions γLV for common adhesive liquids and critical sur-
face tension γC for various solids. A brief discussion of how these properties are mea-
sured follows.
Wetting is favored when the substrate surface tension γSV or its critical surface energy
γC is high and the surface tension of the wetting liquid γLV is low (that is, γC substrate > γadhesive).
Low-energy polymers, therefore, easily wet high-energy substrates such as metals.
Conversely, polymeric substrates having low surface energies will not be readily wet by
most other materials and are useful for applications requiring nonstick, passive surfaces.
Adhesive
Substrate
Trapped air
Adhesive
Substrate
Adhesive completely
fills irregularities
FIGURE 3.3 Illustration of poor (top) and good (bottom) wetting by an
adhesive spreading over a surface.
50 CHAPTER THREE
Epoxy resin 47
Fluorinated epoxy resin 33
Epoxy resin + DETA 44
Epoxy resin + DEAPA 33
Silicone 24
Polyolefins 31
Water 72
Substrate Critical surface tension, dyn/cm
Aluminum ∼500
Copper ∼1000
Glass ∼1000
Acetal 47
ABS 35
Epoxy 47
Nylon 46
Polycarbonate 46
PPS 38
PET 43
Polyimide 40
Polystyrene 33
Polysulfone 41
PTFE 18
PVC 39
Polyethylene 31
Silicone 24
Most common adhesive liquids readily wet clean metal surfaces, ceramic surfaces, and
many high-energy polymeric surfaces. However, epoxy adhesives do not wet low-energy sur-
faces such as polyethylene and fluorocarbons. The fact that good wetting requires the adhe-
sive to have a lower surface tension than the substrate explains why organic adhesives, such
as epoxies, have excellent adhesion to metals, but offer weak adhesion on many untreated
polymeric substrates, such as polyethylene, polypropylene, and the fluorocarbons.
Figure 3.4 provides a simple view of the relationship of wetting and surface energies.
Here the contact angle of a drop of an epoxy adhesive on a variety of surfaces is shown.
The surface energy of a typical epoxy resin is about 42 mJ/m2 (dyn/cm). The expected bond
strengths would increase as the contact angle decreased. Therefore, the bond strength of the
epoxy adhesive on an epoxy substrate would be expected to be the greatest, followed by
polyvinyl chloride, polyethylene, and polytetrafluoroethylene in that order.
The wetting of surfaces by adhesives can be described by two activities: (1) a lateral
spreading of the film and (2) a penetration of the fluid adhesive into the surface cavities that
are characteristic of the inherent surface roughness. The first activity is controlled by the rel-
ative surface energies of the adhesive and substrate as explained above. The second activity
is controlled mainly by the viscosity of the adhesive and the time it is in the liquid state.
The surface energetics that control wetting are largely related to the general chemical
composition of the epoxy polymer molecule. However, the surface tension of an epoxy
IMPORTANT PROPERTIES OF EPOXY ADHESIVES 51
Epoxy adhesive
γ = 42 mJ/m2
FIGURE 3.4 Contact angle of an uncured epoxy adhesive on four substrates of varying, critical
surface tension. Note that as the critical surface tension of the substrates decreases, the contact
angle increases, indicating less wetting of the surface by the epoxy adhesive.4
system can be modified somewhat by the choice of curing agent, as shown in Table 3.4.
Certain amines are surprisingly effective in reducing the surface tension of epoxy resins.5 The
surface tension of a specific product is also generally inversely dependent on temperature.
Epoxy resins have relatively low surface tension as well as good processing properties,
strength, and durability, and thus they make good adhesives. The good adhesion of epoxy
resins to high-energy surfaces is attributed to interfacial hydrogen bonding, which can be con-
sidered a special form of crosslinking. The relatively good wetting and adhesion properties of
epoxy resins can be explained by the pendant secondary hydroxyl groups, which occur along
the molecular chain, and which are strongly adsorbed onto oxide and hydroxyl surfaces.
Note that good wetting is necessary for bond formation. However, it is not the sole cri-
terion for a strong adhesive joint. Several other important parameters, as noted in the sec-
tions that follow, strongly affect the adhesive strength of epoxy systems.
Measurement of Surface Energy Properties. The surface tension and surface energy of
liquids are numerically equivalent. Surface energy is generally given in millijoules per
square meter (mJ/m2), while surface tension is given in units of dynes per centimeter
(dyn/cm) or newtons per meter (N/m). The surface tensions of liquids are readily deter-
mined by measuring the surface tension with a duNouy ring7 or Wilhelmy plate.8
The surface energy (critical surface tension) of solids is measured by a method devel-
oped by Zisman.9 In this method a series of contact angle measurements are made with var-
ious liquids with known surface tensions on the solid to be tested. The contact angle θ is
plotted as a function of the γLV of the test liquid. The critical surface tension is defined as
the intercept of the horizontal line cos θ = 1 (i.e., when the contact angle is 0°) with the
extrapolated straight-line plot of cos θ against γLV of the liquids. The γLV at this intersection
point (i.e., where a hypothetical test liquid would just spread over the substrate) is defined
as the critical surface tension of the solid.
The critical surface tension value for most inorganic solids is in the hundreds or thou-
sands of dynes per centimeter. For polymers and organic liquids, it is at least an order of
magnitude lower. Critical surface tension is an important concept that leads to a better
understanding of wetting and adhesion.
Adsorption Theory of Adhesion. The adsorption theory states that adhesion results from
molecular contact between two materials and the surface forces that develop. Adhesion
results from the adsorption of adhesive molecules on the substrate and the resulting attrac-
tive forces, usually designated as secondary or van der Walls forces. For these forces to
develop, the respective surfaces must not be separated more than 5 angstroms (Å) in dis-
tance. Therefore, the adhesive must make intimate, molecular contact with the substrate
surface.
The process of establishing continuous contact between an adhesive and the adherend
is known as wetting. Figure 3.3 illustrates good and poor wetting of an adhesive spread-
ing over a surface. Good wetting results when the adhesive flows into the valleys and
crevices on the substrate surface; poor wetting results when the adhesive bridges over the
valley. Obtaining intimate contact of the adhesive with the surface is essentially ensur-
ing that interfacial flaws are minimized or eliminated. At a minimum, poor wetting
causes (1) less actual area of contact between the adhesive and adherend and (2) stress
risers at the small air pockets along the interface. This results in lower overall joint
strength.
Wetting can be determined by contact angle measurements. It is governed by the Young
equation, which relates the equilibrium contact angle θ made by the wetting component on
the substrate to the appropriate interfacial tensions:
The term γSV is the interfacial tension of the solid material in equilibrium with a fluid vapor;
γLV is the surface tension of the fluid material in equilibrium with its vapor; and γSL is the
interfacial tension between the solid and liquid materials. Complete, spontaneous wetting
occurs when θ = 0° or when the material spreads uniformly over a substrate to form a thin
sheet. A contact angle of 0° occurs with pure water droplet on a clean, glass slide.
Therefore, for complete spontaneous wetting, cos θ > 1.0 or when
After intimate contact is achieved between adhesive and adherend through wetting, it is
believed that permanent adhesion results primarily through forces of molecular attrac-
tion. Four general types of chemical bonds are recognized as being involved in adhesion
and cohesion: electrostatic, covalent, and metallic, which are referred to as primary
bonds, and van der Walls forces, which are referred to as secondary bonds.
IMPORTANT PROPERTIES OF EPOXY ADHESIVES 53
3.2.4 Reactivity
Reactivity of the formulated adhesive is determined primarily by the epoxy resin and curing
agent or catalyst that is used. The structure of the molecule and the number and type of
functional groups greatly influence reactivity. The reactivity of the epoxy resin molecule is
dependent primarily on the number of reactive epoxy groups on the molecule. The number
of reactive epoxy groups can be determined by the weight per epoxy or epoxy equivalent
weight (EEW), as explained in Chap. 2. Epoxy equivalent weights can be determined by
chemical or physical methods or by the use of infrared spectroscopy. These methods are
described in Chap. 20.
The type of epoxy group and its location within the molecule influence reactivity. Some
epoxy resin structures prefer to react with acid curing agents, and others with basic curing
agents. Certain epoxy structures are extremely reactive with specific catalysts, and others
are virtually inactive. Several types of epoxy resins are reactive with almost all classes of
curing agents.
Reactive groups other than epoxy rings may be present in the molecule. Hydroxyl and
olefinic groups are the most common. These can affect reactivity as well as the course and
sequence of the polymerization reactions. Their number and location could accelerate or
retard the overall reaction rate and lead to significantly different three-dimensional cured
polymeric structures. Catalytic sites (e.g., tertiary nitrogen) on the epoxy molecule can also
influence the reactivity and favor certain reactions.
The hydroxyl content is generally expressed as hydroxyl equivalent weight, the weight
of resin in grams containing one gram equivalent of hydroxyl, or as hydroxyl equivalent
per gram. The hydroxyl equivalent weight is used similarly to the epoxy equivalent
weight, and it is also measured by physical and chemical processes as well as by infrared
spectroscopy.
Steric factors influence reactivity by blocking possible reaction sites. Bulky side chains
that are developed during the course of polymerization may also inhibit reaction. Reactivity
is also dependent on the temperature that occurs during the curing process. The tempera-
ture that the epoxy resin will experience during cure is reliant on (1) the external curing
process temperature (e.g., oven temperature for an elevated-temperature curing epoxy) and
(2) the internal exotherm generated during the process of cure.
Exotherm is defined as the increase in temperature of the mixed compound above the
external cure temperature due to energies released as the epoxy groups react. Unlike external
heating, exothermic heating originates at the center of the mass, and the epoxy is heated from
inside to outside. The degree of exotherm is very much dependent on the mass of the epoxy
due to its relatively low thermal conductivity. As mass increases, so does the exotherm.
With epoxy adhesive systems, exothermic heating can be both good and bad. The exother-
mic temperatures accelerate the reactivity of the system. Many epoxy adhesives rely on the
exotherm for complete curing and dense crosslinking under moderate external temperature
conditions. However, exothermic temperature rise can be significant. For example, a 1-lb
mass of epoxy could reach a temperature in excess of 350°C depending on the curing agent.
Amine curing agents because of their high reactivity have high exotherms, whereas anhydride
curing agents generally have low exotherms. Exotherm temperatures can degrade heat-
sensitive substrates, reduce the working life of the adhesive, and even cause smoldering or
burning of the epoxy resin itself.
Reactivity can also be increased by externally heating the epoxy formulation to a pre-
selected curing temperature. Epoxy resin reactions roughly obey Arrhenius’ law that for
every 10°C rise in temperature, the reaction rate doubles. Certain epoxy resin systems must
be heated for any reaction to take place at all. This is beneficial in that these “latent” adhe-
sive formulations are one-component products that do not require metering or mixing yet
have long, practical shelf lives.
54 CHAPTER THREE
In a heat-cured epoxy resin system, the hydroxyls generally react with either epoxy or
acid groups. Hence, heat-cured diepoxies can be regarded as being potentially tetrafunc-
tional, rather than difunctional. Conventional diepoxy resins, therefore, yield much more
highly crosslinked structures with higher heat and chemical resistance after heat cure than
after room temperature cure.
Reactivity is often measured by the working life (ASTM D 1338) of a predetermined
amount of mixed resin at a defined temperature. ASTM D 1338 uses two methods of deter-
mining the working life. One method uses the viscosity change, and the other method uses the
shear strength development as the criterion for determining when the effective working life
has expired. However, a measure of the reactivity can be made by simply picking at the react-
ing mass with the end of a toothpick and determining when penetration is no longer possible.
Cure rate of an actual adhesive film can also be determined by several useful analytical
methods. With these methods, fundamental properties of the adhesive, such as dielectric
loss, mechanical damping, or exotherm, are measured as a function of time and tempera-
ture as the adhesive cures. Several of these test methods are described in Chap. 20.
Once the adhesive is applied to the substrate, it must harden into a thermosetting film hav-
ing high cohesive and adhesive strength. Several processes occur during the curing or solid-
ification of epoxy resins that can lead to fundamental problems with adhesive systems.
These problems are generally due to internal stresses that develop within the joint as the
result of the curing processes, but they can also be due to side reactions that occur prior to
joint assembly. Such problems are primarily caused by
These internal stresses often can have a degrading effect on the adhesive properties but
little or no effect on the cohesive properties of the adhesive film. They mainly affect the
interface area of the joint.
The viscosity increase in a epoxy resin–curing agent system could result in poor wetting of
the substrate surface, resulting in suboptimal adhesion. Several reaction mechanisms can
also occur to an epoxy adhesive once it is mixed and applied to a substrate but before the
substrates are mated. These mechanisms can result in a weak boundary layer, which will
prevent optimal wetting and reduce the strength of the adhesive.
Many epoxy resin–curing agent mixtures are hygroscopic and will adsorb moisture from
the ambient humidity. This is primarily due to the properties of the curing agent. One exam-
ple is a polyamide curing agent. If the mixed adhesive is applied to the substrate and allowed
to wait until the substrates are mated, water can be adsorbed onto the surface of the adhesive
mainly through the polyamide molecules. Assemblies that are most prone to this effect
are those that have a long time period between adhesive application and joint closure and
IMPORTANT PROPERTIES OF EPOXY ADHESIVES 55
where relative humidity conditions are high. The very thin layer of water that forms on the
adhesive surface will present a weak boundary layer between the adhesive and the mating
substrate when the joint is closed. When in service, the boundary layer fails under stress. The
failure mode that is noticed in these cases is generally an adhesion failure at the interface.
Another possible preassembly reaction mechanism has been noted with regard to amine
cured epoxy resins.10 A variability and reduction in the rate of conversion of epoxy groups
in DGEBA epoxy resin cured at room temperature with diethylene triamine (DETA) was
noticed. This is due to a side reaction of the amine with air, resulting in bicarbonate for-
mation. As a result, the adhesive strength decreased drastically when the uncured epoxy
amine was exposed to ambient air for a significant period of time.
One concludes that when room temperature curing systems are to be used as adhesives,
the assemblies should be joined quickly to preclude various reaction mechanisms from tak-
ing place that could degrade the strength of the final bond.
Loss of theoretical adhesive strength can also arise from the action of internal stress con-
centrations caused by trapped gas and voids. Griffith11 showed that adhesive joints may fail
at relatively low stress if cracks, air bubbles, voids, inclusions, or other surface defects
occur as a result of the curing process.
These trapped gas pockets could be formed due to poor wetting or from evaporation of
high-vapor-pressure components. If the gas pockets or voids in the surface depressions of
the adherend are all nearly in the same plane and not far apart (as is shown in Fig. 3.5, upper
adherend), cracks can rapidly propagate from one void to the next. However, a variable
degree of roughness, such as shown in Fig. 3.5 lower adherend, provides barriers to spon-
taneous crack propagation. Therefore, not only is surface roughening important, but the
degree and type of roughness may be important as well.
Since real surfaces are not smooth or perfectly flat and most epoxy adhesives are vis-
coelastic fluids, it is necessary to understand the effects of surface roughness on joint
strength. A viscous liquid can appear to spread over a solid surface and yet leave many gas
pockets or voids in small surface pores and crevices. Even if the liquid does spread spon-
taneously over the solid, there is no certainty that it will have sufficient time to fill in all the
voids and displace the air. The gap-filling mechanism is generally competing with the set-
ting mechanism of the liquid.
This problem occurs when the liquid solidifies rapidly after being applied. An exam-
ple is a fast-curing epoxy. Very fast-reacting epoxy adhesive systems that set in several
minutes generally do not have the high adhesion strength that slower-curing epoxy sys-
tems have. One reason for this (there are others primarily related to the chemistry of these
Smooth adherend
Rough adherend
FIGURE 3.5 Effect of surface roughness on coplanarity of gas bubbles: Upper adherend
is smooth, and gas bubbles are in the same plane; lower adherend has roughness, and gas
bubbles are in several planes.
56 CHAPTER THREE
fast-acting systems) is that the curing reaction does not provide sufficient time for the
adhesive to fill the crevices on the substrate surface. This effect can also be shown by the
differences in adhesive strength that can be achieved with an epoxy system that is applied
to the substrate after varying time periods between curing agent addition and joint closure.
A lower degree of adhesion accompanies the system that is applied after a longer “induc-
tion” period, presumably due to increase in viscosity during the early stages of cure.12
Since slower-curing epoxy adhesives systems flow over and wet high-energy surfaces
very well, there is little chance for air to become trapped at the interface. As a result,
mechanical abrasion is often recommended as a substrate surface treatment prior to appli-
cation of the epoxy adhesive. The added surface area and the mechanical bonding provided
by the additional peaks and valleys on the surface will enhance adhesive strength. If the
adhesive does not wet the substrate surface well, such as in the case of epoxy resin on poly-
ethylene, mechanical abrasion is not recommended since it will only encourage the proba-
bility of gas voids being trapped at the interface.
It has also been shown13 that a concentration of stress can occur at the point on the free
meniscus surface of the adhesive (edge of the bond line). This stress concentration increases
in value as the wetting becomes poorer and the contact angle θ increases. At the same time,
the region in which the maximum stress concentration occurs will move toward the adhesive-
adherend interface. Thus, poor wetting will be associated with a weak spot at the adhesive-
adherend interface and at the edge of the adhesive joint with a consequent likelihood of
premature failure at this region.
The stress concentration factors for a lap joint in shear bonded with adhesives having a
contact angle from 0 to 90° are illustrated in Fig. 3.6. For contact angles less than 30° (i.e.,
good wetting) the maximum stress occurs in the free surface of the adhesive away from the
edges, and the stress concentration is not much greater than unity. These localized stress
effects can be reduced by improving the wetting of the adhesive, optimizing surface prepa-
ration processes, or modifying the adhesive with a toughener to stop the propagation of
cracks due to these phenomena.
Maximum Maximum
between at edges
Stress concentration factor
A and B A, A
A A 2
Hexane 20 25
Isopropanol 20.8 25
Acetone 22.6 25
n-Propanol 23.8 20
Methyl ethyl ketone 24.6 20
Trichloroethane 25.1 20
n-Butyl acetate 27.6 27
Toluene 28 NA
Xylene 28 NA
Ethylene glycol 48.4 20
Polypropylene glycol 72 25
Water (distilled) 72.8 20
Solvents can also be used to reduce the surface tension of the adhesive formulation. The
surface tensions of common solvents are shown in Table 3.5. Of course, when using solvents,
one needs to make sure that they evaporate from the bond line before cure. Solvent solu-
tions do not change the equilibrium surface energetics of the system. They only provide
lower viscosity so that wetting is established at a faster rate.
3.3.3 Shrinkage
Nearly all polymeric materials (including adhesives and sealants) shrink during solidifica-
tion. Sometimes they shrink because of escaping solvent or volatile by-products, leaving
less mass in the bond line. Even 100 percent reactive adhesives, such as epoxies, with no
formation of by-products during cure experience some shrinkage because their solid poly-
merized mass occupies less volume than the liquid reactants.
Table 3.6 shows typical percentage volumetric shrinkage for various reactive adhesive
systems during cure. One of the reasons for the great acceptance of epoxy resins as adhe-
sive materials is their low degree of shrinkage on cure relative to other reactive adhesives.
Typical linear shrinkage values are shown in Table 3.7 for various highly filled epoxy
adhesive formulations and cure conditions. Notice that the incorporation of filler signifi-
cantly reduces the shrinkage of the epoxy systems. Higher shrinkage values are generally
noticed when the cure is carried out at higher temperatures. It is proposed that this is due to
greater crosslinking density that occurs during an elevated-temperature cure.
Acrylics 5–10
Anaerobic 6–9
Epoxies 4–5
Urethanes 3–5
Polyamide hot melts 1–2
Silicones <1
58 CHAPTER THREE
TABLE 3.7 Effect of Curing Agents on the Shrinkage of Highly Filled Epoxy-
Amine Systems during Cure16
Cure schedule
Coefficient
of thermal
expansion
Adhesives Adherends
2 × 10−4
Cellulosic plastics
4 × 10−6
Pyrex glass
Quartz glass
5 × 10−7
FIGURE 3.7 Thermal expansion coefficients of common materials.18
IMPORTANT PROPERTIES OF EPOXY ADHESIVES 59
The result of such shrinkage is internal stresses at the adhesive-substrate surface and the
possible formation of cracks and voids within the bond line itself. Formulators are often
able to adjust the adhesive in several ways to minimize stress from shrinkage:
1. Elastic adhesives deform when exposed to such internal stress and are less affected by
shrinkage. Thus, flexibilizers or tougheners may be added to the formulation to distrib-
ute the stresses over a larger area (Chap. 8).
2. Inorganic fillers may be added to the adhesive composition to reduce the concentration
of epoxy resin in the formulation as the degree of polymeric shrinkage is dependent on
the volume of epoxy in the joint (Chap. 9).
3. Certain epoxy monomers17 have been developed that expand slightly during cure. These
formulations are expected to provide very high adhesive properties when blended with
conventional epoxy resins to provide zero net shrinkage; however, to date, their com-
mercialization has been limited due to cost.
One of the most common causes of internal stress is due to the difference in the thermal
expansion coefficients of the adhesive and the adherend. These stresses must especially be
considered when the adhesive solidifies at a temperature that is different from the normal
temperature that it will be exposed to in service.
Figure 3.7 shows that thermal expansion
coefficients for some common adhesives
Metal
and substrates are more than an order of
magnitude apart. This means that the bulk
Adhesive Plastic
adhesive will be required to move more
than 10 times as far as the substrate when
the temperature changes, thereby causing
stress at the interface. In general, the more
highly flexibilized the epoxy resin system,
the higher the expansion rate, and the more
highly filled the epoxy, the lower the expan-
sion rate.
The stresses produced by thermal expan-
sion differences can be significant. Take,
e.g., an annular journal bearing where a
polyamide-imide sleeve is bonded to the
FIGURE 3.8 Journal bearing application. Outer
internal diameter of a stainless steel housing cylinder (stainless steel) is bonded to inner cylinder
(Fig. 3.8). Further, assume that the epoxy (polyamide-imide) with an epoxy adhesive. Exposure
adhesive used is one that cures at 125°C. At to low temperatures causes significant stress on bond
the cure temperature, all substrates and the due to differences in coefficient of thermal expansion.
gelled adhesive are in equilibrium. However,
when the temperature begins to reduce as the
system approaches ambient conditions after cure, stresses in the adhesive develop because the
polyamide-imide substrate wants to shrink faster and to a greater extent than the steel. At
room temperature, these stresses may be significant but not high enough to cause adhesive
failure. Now further assume that the bonded bearing is to be placed in service with operating
temperatures that will vary between 125 and −40°C. At 125°C the internal stresses due to the
mismatch in thermal expansion are reduced to zero, since we are back at the equilibrium con-
dition (assuming that there was no shrinkage in the adhesive as it cured or other stresses in the
joint). However, when the service temperature reaches −40°C, the thermal expansion
60 CHAPTER THREE
Shear strength
Test temperature
FIGURE 3.9 Plot of tensile shear strength of an aluminum joint bonded
with an elevated-temperature curing epoxy adhesive as a function of test
temperature.
differences create internal stresses in addition to those already there due to curing, and a fail-
ure could result.
A similar example is evident by a typical graph (Fig. 3.9) of tensile shear strength ver-
sus temperature for an elevated-temperature cured epoxy adhesive. Notice that the bond
strength actually increases with temperature to a maximum, and then it falls off with
increasing temperature. This is similar to the case above where the internal stresses are
actually reduced by the increasing service temperature. At some elevated temperature, the
internal stresses are completely relieved, and the bond strength reaches a maximum. The
test temperature at which this occurs is usually very close to the curing temperature. At
higher test temperatures, additional stresses develop or the effects of thermal degradation
become evident, and the bond strength then decreases with increasing test temperature.
There are several possible solutions to the expansion mismatch problem. One is to use a
resilient adhesive that deforms with the substrate during temperature change. The penalty in
this case is possible creep of the adhesives, and highly deformable adhesives usually have
low cohesive strength. Another approach is to adjust the thermal expansion coefficient of the
adhesive to a value that is nearer to that of the substrate. This is generally accomplished by
selection of a different adhesive or by formulating the adhesive with specific fillers to “tai-
lor” the thermal expansion. A third possible solution is to coat one or both substrates with a
primer. This substance can provide either resiliency at the interface or an intermediate ther-
mal expansion coefficient that will help reduce the overall stress in the joint.
In addition to good adhesion, an adhesive must have satisfactory cohesive strength and
durability once it is cured. The physical and chemical properties of the cured product are
determined by
• The structure of the resin, particularly the number, location, and reactivity of the epoxy
and other groups present
• The curing agent, stoichiometry, and cure conditions
IMPORTANT PROPERTIES OF EPOXY ADHESIVES 61
These parameters establish the rigidity, thermal stability, chemical resistance, and overall
usefulness of the adhesive joint.
Factors that can affect cohesive strength include (1) the molecular weight and the nature
of the molecules between crosslinks and (2) the degree of crosslinking, Epoxy polymers are
network polymers with three-dimensional molecular linkages formed throughout the bulk
polymer. In these types of polymers, a great deal of importance falls on the functionality of
monomer units and the degree to which crosslinking takes place. These factors determine the
most important properties such as rigidity, heat resistance, strength, and chemical resistance.
The structure of a crosslinked epoxy resin can be represented by a ladder-type model
(Fig. 3.10) in which the dimensional network is formed by tying together a relatively high-MW
backbone polymer with random and relatively short crosslinking segments. The properties of
the network are then determined by the nature of the backbone polymer chain segment, crosslink
segments, and branch points. As any one, or any combination, of the structural changes shown
in the left-hand column of Fig. 3.10 occurs, all the properties shown in the right-hand column
are changed in the direction shown. Notice that certain properties can be obtained at the expense
of other properties. It becomes apparent, therefore, that the formulator must perform a balanc-
ing act to provide the correct set of properties for the application. Fortunately, the nonresinous
components in the formulation (filler, modifiers, etc.) help provide the tools to do this.
Much of the development in epoxy adhesive technology has been toward attempts to com-
bine the advantages of rigid and flexible epoxy systems and to minimize their disadvantages.
The property tradeoffs are illustrated in Fig. 3.11. Generally, the rigid structural adhesive
that displays excellent tensile or shear strength does not perform well when tested in peel,
impact, or fatigue. This reason is that the elongation of the adhesive distributes stresses
Curing agent
molecule Epoxy resin
molecule
Crosslinking
site
40 4000
20 2000
Chemical, thermal
to H2O, solvents
Permeability
stability
0 0
0 Hardness, modulus
FIGURE 3.11 Property tradeoffs for single-phase, thermosetting adhesives.19
over a much larger area (reducing stress concentration points) than if the adhesive were
brittle.
The major problem is that the attainment of properties such as peel, flexibility, and
toughness is generally accompanied by the reduction in properties such as heat resistance,
chemical resistance, and shear strength. Future chapters discuss how the epoxy adhesive
formulator can merge these properties.
It should also be emphasized that the epoxy resins are only building blocks in the devel-
opment of epoxy-based adhesives. The formulation will ultimately involve one or more
epoxy resins; curing agents; modifiers, such as plasticizers, flexibilizers, accelerators, sta-
bilizers, and flow control agents; and fillers. In general, flexibility can be controlled by the
parameters shown in Table 3.8.
Epoxy adhesives have cohesive properties that significantly depend on the crosslink den-
sity. For example, torsional butt shear strength of epoxy aluminum joints has been noticed
to decrease as Mc increased.20 However, flexibility and toughness are increased with increas-
ing Mc. Figure 3.12 shows the general physical relationship between Mc and the physical
state of epoxy resins.
The crosslink density ultimately defines the rheological and mechanical properties of
the polymer. Polymers that have a high crosslink density are thermosets and are infusible,
insoluble, and dimensionally stable under load. These properties make epoxy resin systems
useful as structural adhesives as well as important materials in other applications. Polymers
that have a low crosslink density are more flexible and show greater resistance to stress
concentration, impact, and cold.
In the DGEBA product (n = 1) shown in Fig. 3.13, the epoxy groups are separated by
seven units (the aromatic ring is counted as one unit). In other resins the groups may be
Hard-brittle
Physical state
Tough-yielding
Soft-flexible
5 10 15 20 in
130 260 390 520 mm
Distance between crosslinks × 108
FIGURE 3.12 Effect of molecular weight between crosslinks on the physical state of
epoxy resins.
64 CHAPTER THREE
CH3
O CH3 O
11
10 Glassy state
9 Glass transition (Tg)
8
Log modulus, Pa
Leathery region
7
Rubbery plateau
6 Rubbery flow
4
Liquid flow
3
Temperature, °C
FIGURE 3.14 Relationship between elastic modulus and temperature showing the
glass transition region.22
usually associated with the onset of long-range motion in the polymer backbone due to tem-
perature effects.
As shown in Fig. 3.14, the glass transition temperature is usually a narrow temperature
range rather than a sharp point, as is the freezing or boiling point. Molecular motion at this
point does not involve entire molecules, but in this region deformation begins to become
nonrecoverable as permanent set takes place.
Generally Tg is measured as the temperature at which the slope of a temperature-volume
plot undergoes a sudden upward change, as shown in Fig. 3.15. It is the temperature at
which there is significantly more molecular mobility than at lower temperatures (i.e., the
molecules have sufficient thermal energy to be considered mobile).
Tg is a property of a polymer that depends on its chemical composition and the degree
of crosslinking or molecular interaction. Often the glass transition temperature is used as a
measure of the degree of crosslinking in a specifically defined epoxy system. Adhesive
Volume
Tg
Temperature
FIGURE 3.15 The effect of temperature on the total vol-
ume of a polymer.
66 CHAPTER THREE
formulators use the glass transition temperature Tg as a practical basis for compounding
products with the appropriate amount of internal motion.
Many properties of adhesive bonds are influenced by the Tg or mobility of the molecu-
lar chain structure, as shown in Fig. 3.16. When chain segments can move easily, such as
when the temperature exceeds the Tg, they can deform under impact or assume new align-
ments under mechanical or thermal expansion stresses. This movement spreads the applied
energy over a greater number of atoms and thus gives the bond a better chance to resist
stress. Brittleness is, therefore, reduced and flexibility is increased.
Like all polymers, adhesive materials undergo constant thermally induced vibration.
The amplitude of these vibrations is determined primarily by temperature, chain flexibility,
and crosslinking, and to a lesser extent by fillers and physical stresses. A certain amount of
chain flexibility is desirable since it imparts resiliency and toughness to the adhesive film.
Too much flexibility, however, may lead to creep (i.e., plastic flow under load) or very poor
elevated-temperature resistance.
For elevated-temperature performance it is necessary for the Tg of the adhesive to be
higher than the highest temperature to be encountered in service. At temperatures above the
Tg of any polymer strength and stiffness diminish rapidly. Bond strength at elevated tem-
peratures can be increased by raising the degree of crosslinking or using more thermally
resistant (aromatic) epoxy base resins.
However, flexibility and peel strength at room temperature will be low if the Tg is high.
It is very difficult to provide an adhesive that has high peel strength with good cohesive
strength and environmental resistance. This problem and potential solutions are discussed
in detail in the following chapters. A high Tg may also limit the low-temperature properties
of the adhesive.
Typical glass transition temperatures for adhesive resins are shown in Table 3.9. Note,
however, that the Tg for epoxy adhesives can vary significantly with their formulation.
Glass transition temperatures for several epoxy formulations are shown in Table 3.10.
Creep
Adhesive properties
Impact resistance
Permeation
Density
Brittleness
Cohesive strength
Tg
Room temperature curing cannot, in principle, achieve the same degree of crosslinking as
is obtained by curing at elevated temperatures, although it does achieve sufficient proper-
ties for many adhesive applications. At elevated temperatures, the epoxy resin and curing
agent molecules are mobile, and there is a greater potential for reaction than at room tem-
perature. Figure 3.17 shows the glass transition temperatures of common epoxy formula-
tions cured at both room and elevated temperatures.
Another reason for higher Tg with elevated-temperature curing systems is that the
hydroxyls along the epoxy chain can react. At ambient temperature, reaction between an
epoxy group and a hydroxyl group proceeds very slowly. Hence, the diepoxy group, which
is formed when the epoxy group reacts with an amine, generally cannot enter into further
reaction with a second epoxy ring at room temperature. As a result, heat-cured diepoxies
can be regarded as being potentially tetrafunctional, rather than difunctional.
Postcuring at elevated temperatures after a room temperature cure is a common process
in epoxy technology, and this can moderately increase the Tg in some systems.26 Such effects
could be due to secondary reactions (irreversible) or to free volume effects (reversible).
These effects could also be realized during the normal aging of the epoxy system in service.
One should be careful, however, in assuming that a low-temperature cure followed by an
elevated-temperature postcure (cure condition 1) will provide properties equivalent to only
a high-temperature cure (cure condition 2). As explained in the previous section, the types
Curing agent
Amidoamines
Aliphatic
amines
Cycloaliphatic
amines
Aromatic
amines
REFERENCES
1. Lee, H., and Neville, K., Handbook of Epoxy Resins, McGraw-Hill, New York, 1967, p. 17.11.
2. Potter, W. G., Epoxide Resins, Springer-Verlag, New York, 1970, p. 30.
3. Petrie, E. M., Handbook of Adhesives and Sealants, McGraw-Hill, New York, 2000, pp. 59–62.
4. Pocius, A. V., Adhesion and Adhesives Technology, Chapter 6, Hanser Publishers, New York,
1997.
5. Schornhorn, H., and Sharpe, L., Journal of Polymer Science Part B: Polymer Letters, vol. 2, no. 7,
1964, p. 719.
6. Dannenberg, H., and May, C. A., “Epoxide Adhesives,” in Treatise on Adhesion and Adhesives,
R. L. Patrick, ed., Marcel Dekker, New York, 1969, p. 18.
7. DuNouy, P., and Lecounte, J., General Physiology, vol. 1, 1919, p. 521.
8. Wilhelmy, L., Annals of Physics, vol. 119, 1863, p. 177.
9. Zisman, W. A., “Relation of Equilibrium Contact Angle to Liquid and Solid Constitution,”
Chapter 1 in Contact Angle, Wettability, and Adhesion, R. F. Gould, ed., American Chemical
Society, Washington, 1964.
IMPORTANT PROPERTIES OF EPOXY ADHESIVES 69
10. Bell, J. P., et al., “Amine Cured Epoxy Resins: Adhesion Loss due to Reaction with Air,” Journal
of Applied Polymer Science, vol. 21, 1977, pp. 1095–1102.
11. Griffith, A. A., Philosophical Transactions of the Royal Society of London, ser. A., vol. 221, no. 163,
1920.
12. Chessin, N., and Taylor, G., “Working Life of Room Temperature Curing Epoxy Adhesives,”
Adhesives Age, vol. 10, no. 9, 1967, p. 29.
13. Mylonas, C., Proc. Seventh International Congress of Applied Mechanics, London, 1948.
14. Baier, R. E., et al., “Adhesion: Mechanisms That Assist and Impede It,” Science, vol. 162,
December 1968, pp. 1360–1368.
15. Schneberger, G. L., “Basic Bonding Concepts,” Adhesives Age, May 1985.
16. Dannenberg, and May, “Epoxide Adhesives,” in Treatise on Adhesion and Adhesives.
17. Sadhir, R. K., and Luck, R. M., Expanding Monomers: Synthesis, Characterization, and Applications,
CRC Press, Boca Raton, FL, 1992.
18. Perry, H. A., “Room Temperature Setting Adhesive for Metals and Plastics,” Adhesion and
Adhesive Fundamentals and Practice, J. E. Rutzler and R. L. Savage, eds., Society of Chemical
Industry, London, 1954.
19. Bolger, J., “Structural Adhesives State of the Art,” in Adhesives in Manufacturing, G. L.
Schneberger, ed., Marcel Dekker, New York, 1983, p. 143.
20. Lin, C. J., and Bell, J. P., “The Effect of Polymer Network Structure upon the Bond Strength of
Epoxy-Aluminum Joints,” Journal of Applied Polymer Science, vol. 16, 1972, p. 1721.
21. Schneberger, G. L., “Polymer Structure and Adhesive Behavior,” in Adhesives in Manufacturing.
22. Baker, A. M. M., and Mead, J., “Thermoplastics,” Modern Plastics Handbook, C. A. Harper, ed.,
McGraw-Hill, New York, 2000.
23. Schneberger, “Polymer Structure and Adhesive Behavior,” Adhesives in Manufacturing.
24. Schneberger, “Basic Bonding Concepts.”
25. Potter, Epoxide Resins, p. 93.
26. Brewis, D. M., Comyn, J., and Fowler, J. R., Polymer, vol. 20, 1979, p. 1548.
27. Air Products, “Curing Agents Center for Epoxy,” SpecialChem4Polymer.com, December 2003.
This page intentionally left blank
CHAPTER 4
EPOXY RESINS
4.1 INTRODUCTION
The syntheses of commercial epoxy resins that are commonly used in many applications
were discussed in Chap. 2. Additional information is provided in this chapter with regard
to the physical and chemical properties of certain epoxy resins relative to their use in adhe-
sive systems.
Many types of epoxy resins can be used in adhesive formulations. These are character-
ized in Table 4.1. The most commonly used type is the resin-based diglycidyl ether of
bisphenol A (DGEBA). Epoxy novolac, flexible epoxy, high-functionality, and film-forming
epoxy resins are also used in specialty applications.
The epoxy resin is a primary component in any epoxy adhesive formulation, and it is
often referred to as the base polymer. However, it is certainly not the only or even not
always the predominant component in influencing desirable end properties. Epoxy resins
by themselves are often too rigid to provide the required properties such as flexibility, peel
and impact strength, and thermal cycling resistance. As a result, they are often modified
with other components or hybridized with other types of polymeric resins to provide these
functions.
Similarly, lower-molecular-weight diluents may be added to reduce viscosity. Viscosity
reduction is often necessary to allow for easier compounding of the ingredients into the for-
mulation or to provide specific application properties. Accelerators are also frequently
added to speed the cure rate of epoxy resins at or below room temperature.
Thus, the epoxy resin is only the base foundation on which a complete formulation must
be constructed. The formulation freedom and opportunities that it provides are some of the
most valuable features of epoxy technology.
This chapter reviews the various types of epoxy resins and their characteristics in adhe-
sive systems. These resins are classified by generic type as one of the following:
Future chapters describe the other raw materials that contribute to the epoxy adhesive for-
mulation (curing agents and catalysts, Chap. 5; solvents and diluents, Chap. 6; hybrid resins,
Chap. 7; flexibilizers and tougheners, Chap. 8; fillers, Chap. 9; and adhesion promoters,
Chap. 10). Complete adhesive formulations are then discussed in subsequent chapters.
71
Copyright © 2006 by The McGraw-Hill Companies, Inc. Click here for terms of use.
72 CHAPTER FOUR
Primary curing
Epoxy resin type Characteristics requirements*
Epoxy resins based on diglycidyl ether of bisphenol A are not only the oldest type of epoxy
resins but also the most valuable in adhesive formulations. The primary reasons for their
popularity have been
1. The relatively low cost of raw materials used to synthesize these resins
2. Their ability to be cured by a large number of different crosslinking agents at both room
and elevated temperatures
3. Their relatively good mechanical and chemical properties that are exhibited by the
cured resins
The DGEBA family of epoxy resins provides superior overall performance in a variety of
end-use applications. With proper additives or modifiers, these base polymers enable for-
mulators to achieve practically any desired performance level, while optimizing process-
ability and durability in many environments.
Properties of several commercially available DGEBA epoxy resins from various sup-
pliers are listed in App. C. Generally they can be divided into the following classifications:
• Low-molecular-weight liquids
• High-molecular-weight semisolids and solids
• Brominated resins
Figure 4.1 shows a comparison of the physical and curing properties for liquid and solid
DGEBA epoxy resins.
These resins are most often characterized by their epoxy equivalent weight (EEW),
molecular weight (number of repeating units n), and viscosity. Table 4.2 shows the rela-
tionship between EEW and viscosity. These DGEBA epoxy resins can be used alone or in
blends with other DGEBA resins, other epoxy resins, or even other types of polymeric
resins. Very often commercial epoxy resin products are actually blends of resins having a
broad molecular weight distribution.
EPOXY RESINS 73
% Epoxide
0 2.5 5 10 15 20 25
EEW
Physical state
1720 860 430 287 215 172
N value
Form
Resin cure
High temperature
Cure process
Room temperature
Epoxide
Hydroxyl & Cured through epoxide
Hydroxyl
FIGURE 4.1 Comparative physical and curing properties for DGEBA epoxy resins.1
A majority of the world’s epoxy resin market consists of the DGEBA type. The liquid
DGEBA resins are typically used where low viscosity, high reactivity, and high crosslink
density are required. The low viscosity allows them to be conveniently and easily com-
pounded with fillers and other additives. The relatively high functionality allows them to
Approximate number
Number of Epoxy Viscosity, of groups
repeating equivalent poise Melting Molecular
units n weight at 25°C point, °C weight Epoxy Hydroxyl
be cured at low temperatures and with conventional curing agents. This reactivity produces
reasonably high crosslink density necessary in applications requiring high heat, chemical,
or solvent resistance.
DGEBA epoxy resins are formed by reacting bisphenol A with epichlorohydrin (see
Chap. 2). The reaction proceeds via a chlorination intermediate to produce bisphenol A
diglycidyl ether. Variation of the bisphenol A and epichlorohydrin ratios enables resins to
be produced with different molecular weights and viscosities. The DGEBA having a mol-
ecular weight of approximately 380 or an epoxy equivalent weight of about 190 is the pri-
mary liquid epoxy resin in commercial use (e.g., EPON 828, Resolution Performance
Products LLC; DER 331, Dow Plastics).
Liquid DGEBA epoxy resins can be cured by a great variety of curing mechanisms, as
is shown in Chap. 5. These are the types of epoxy resins that are generally used in formu-
lating conventional one-component adhesives that cure at elevated temperatures or two-
component adhesives that can cure at either elevated or room temperatures. The liquid
form enables convenient formulation with additives and fillers. Formulated systems can
be easily mixed prior to application and applied as uniform films to the substrates to be
joined. Because the resulting adhesives are high-viscosity liquids or pastes, they will wet
the substrate well. Yet they do not readily flow out of the joint area to cause adhesive star-
vation. Pressure is not a requisite to cure these adhesives as only contact pressure is gen-
erally applied.
The lower-viscosity grades have an EEW of about 175 and are virtually pure diglycidyl
ethers of bisphenol A. They are so pure, however, that they will crystallize on storage. The
crystals melt on warming above 40°C, and heating can be used to restore a crystallized resin
to its previous form. Special crystallization-free resins systems have been formed by blend-
ing low-viscosity DGEBA resins with more conventional bisphenol A–based epoxy resins.
These low-viscosity DGEBA resins provide all the general properties of the higher-MW
epoxies with the following additional advantages:
• Better wetting of fillers and reinforcements and better penetration of porous substrates
due to low viscosity
• Increased crosslink density (increased Tg and heat distortion temperature)
• Greater chemical uniformity (often preferred in electronic applications because of purity)
• Pale, almost water white color
• Reactivity is reduced. Thus, pot life is increased, and the exothermic temperature is
reduced. Cure may be required at elevated temperatures.
• The degree of crosslinking is generally less, leading to a lower glass transition tempera-
ture, heat distortion temperature, and chemical resistance.
• Adhesion properties are improved.
• Property increases occur in flexibility, impact, and elongation.
EPOXY RESINS 75
The higher-MW semisolid (EEW of 225 to 280) and solid epoxy resins (EEW > 450) may be
blended into lower-MW resins to improve flexibility and decrease reactivity. They also
improve adhesion due to the higher concentration of hydroxyl groups along the molecular
chain. Ten percent of a higher-MW epoxy resin blended into a conventional liquid epoxy resin
(EEW of 190) can significantly improve flexibility, but a reactive diluent frequently needs to
be added to the formulation to counteract the increased viscosity caused by the addition.
The highest-MW DGEBA epoxy resins are termed phenoxy resins. They are highly lin-
ear molecules that are used primarily as thermoplastic coating resins. However, they can be
blended with lower-MW epoxy resins for the improvement of specific properties such as
flexibility, impact and fatigue resistance, and thermal cycling. Phenoxy resins are some-
times used alone as a thermoplastic hot melt adhesive generally in film form.
The solid and semisolid DGEBA resins are often employed as coatings or adhesive sys-
tems where solvents are added to the formulation to reduce viscosity and allow for easy
application onto the substrate. Blends of ketone solvents (methyl ethyl ketone and methyl
isobutyl ketone) with aromatics (xylene and toluene) are generally suitable for thinning
these systems. Esters are also often used as solvents for epoxy-based systems. Higher-boiling-
point solvents such as glycol ethers can be used in amounts of 5 to 15 percent by weight to
improve flow and film surface properties. Solvents and diluents for epoxy resins are dis-
cussed in Chap. 6.
However, one of the greatest advantages of epoxy resin chemistry is that epoxy resins can
provide low-viscosity formulations without solvents. As a result, these 100 percent solids sys-
tems elude environmental regulations and production precautions that are required for a flam-
mable material.
Solvents are used, however, for special applications. For example, solvents may be added
to reduce viscosity and assist penetration on porous substrates. On certain polymeric sub-
strates, solvents may be added to improve adhesion by assisting the diffusion of the adhesive
molecules into the substrate. On nonporous substrates, volatile solvents must be evaporated
before cure because the solvent could interfere with the degree of crosslinking, and under
certain curing conditions, gaseous bubbles could form in the bond line and degrade joint
strength.
Solid epoxy resins are usually formulated as solvent solutions and blends with lower-
MW resins for the production of liquid adhesive systems. However, solid epoxy resins are
also often employed in the manufacture of adhesive systems having solid form. There are
several forms of solid epoxy adhesives that find application. The most common are sup-
ported or unsupported film, powder, and solder stick. Formulations for these adhesives are
detailed in Chap. 13.
Epoxy adhesives can be manufactured into a film form. This is most conveniently done
with solid epoxy resins in solution. Epoxy film adhesives can be thermoplastic (e.g., linear
ultrahigh molecular weight phenoxy resin) hot melts, but more commonly they are formu-
lated, thermosetting materials.
Latent catalysts, fillers, and additives are added to the epoxy resin solution, and the for-
mulation is cast into a free film or onto a supporting medium such as a glass or polymeric
fabric. Once the solvent evaporates in a drying oven, the film can be wound into rolls with
a separator sheet and applied directly to the joint without mixing or metering. The joint is
then assembled and placed in an elevated-temperature oven or autoclave to cure under light
pressure. At the curing temperature, the solid epoxy melts, wets the surfaces of the sub-
strate, and eventually cures to a thermosetting structure. Epoxy films are commonly used
to bond large-area parts such as aerospace, transportation, or paneling components.
Similarly prepared solid epoxy films can be ground into powder. The powder can be
pressed into preformed pellets and applied directly to the adhesive joint as annular rings or
76 CHAPTER FOUR
via another shape. This type of adhesive is conveniently employed in mass-produced arti-
cles that have concentric joint designs (e.g., flared tubing joints). The joint configuration
will hold the solid preform until it melts under the application of heat and forms a reservoir
for the cured adhesive. A powder epoxy adhesive can also be applied to a substrate by elec-
trostatic methods.
The primary advantage of solid epoxy adhesives is that they avoid the disadvantages of
working with liquids. Waste and cleanup are minimized, and health problems are reduced
because the end user handles only a solid substance. Since they are essentially one-component
adhesives, they also eliminate the need to meter and mix individual components.
Brominated epoxy resins are the reaction product of epichlorohydrin and brominated
bisphenol A. They are primarily used in applications where ignition resistance is a require-
ment, such as printed-circuit boards and other products that need to be flame-retardant.
Tetrabromobisphenol A is the largest flame retardant in terms of commercial use at present.
It is used in an estimated 95 percent of all flame-retardant printed wiring boards and is used
in many flame-retardant surface-mounted adhesives.2 It is manufactured by several pro-
ducers and is priced as a commodity product.
The synthesis of brominated epoxy resins was discussed in Chap. 2. The resulting resins
are available primarily as semisolids or solids in solvent solutions. They have properties sim-
ilar to those of other DGEBA epoxies except that the high bromine content (18 to 21 percent)
in the finished resins provides outstanding flame ignition resistance. Tetrabromo diphenylol-
propane (Fig. 4.2) is an example of a commercially brominated epoxy resin.
Standard epoxy resins are usually blended with brominated epoxy resins to provide the
concentration of bromine required to provide ignition temperature resistance. Brominated
resins have also been used as flame-retardant additives in thermoplastic compounds.
Brominated epoxy resins function by liberating acid halide gases as the product ther-
mally breaks down at the temperatures incurred in a fire. These halide gases act as extin-
guishers to significantly increase the ignition temperature of the cured epoxy resin. The
volume of gases liberated and the degree of flame resistance are dependent on the bromine
content of the cured epoxy.
Environmental activists, especially in Europe and Japan, tend to oppose the commercial
use of halogen compounds, such as bromine-containing flame retardants and tetrabromo-
bisphenol A. The German Environmental Department (UBA) has issued a position paper
in which it recommends phasing out tetrabromobisphenol A because it has been detected
in the “food chain.”3 As a result, newer epoxy formulations are employing nonbrominated
flame-retardant additives. These are discussed in Chap. 9. However, one of the greatest
advantages of using a halogenated epoxy resin rather than an additive is the ability to main-
tain physical properties.
Br CH3 Br Br Br
OH CH3
CH2 CH CH2 O C O CH2 CH CH2 O C O CH2 CH CH2
O O
Br CH3 Br Br CH3 Br
n
Epoxy novolac resins are polyglycidyl ethers of a novolac resin. They are prepared by react-
ing epichlorohydrin with a novolac resin (see Chap. 2). The most common epoxy novolacs
are based on medium-MW molecules with phenol and o-cresol novolacs. They generally
have significantly different properties from DGEBA epoxies because of the presence of the
phenolic structure.
Epoxy novolac resins also differ from standard DGEBA-based epoxy resins in their
multifunctionality, which is about 2.5 to 6.0. The multiplicity of epoxy groups allows these
resins to achieve increased crosslink density. The commercial epoxy novolac resins (e.g.,
DER 438, Dow Plastics, and EPON 164, Resolution Performance Products LLC) are semi-
solid to solid resins with EEW in the range of 170 to 230. Recently low-viscosity epoxy
novolac resins have been produced (18,000 to 28,000 cP) to provide easy processing; how-
ever, these generally have lower epoxy content.
When cured with any of the conventional epoxy curing agents, epoxy novolacs gener-
ally produce a product with better elevated-temperature performance, chemical resistance,
and adhesion than those of the bisphenol A–based resins. Thus, epoxy novolacs are used in
structural adhesive systems that require high-temperature performance. Table 4.3 shows
the thermal stability, as measured by weight loss, of epoxy novolac systems having a func-
tionality of 2.5 and 4.0. Notice that significant improvements over standard DGEBA epoxy
resins occur only when the functionality of the epoxy novolac is above 2.5.
To develop the properties of an epoxy novolac to its fullest extent, a high-temperature
cure is necessary. With room temperatures cures, the properties of the final product are sim-
ilar to those of conventional DGEBA systems. The thermal stability of most epoxy novolac
systems is affected markedly by the length of the cure cycle.
Bisphenol F epoxy resins are produced by condensing phenol with formaldehyde, result-
ing in a mixture of isomers and higher-MW condensation products. Bisphenol F epoxy resins
have lower viscosity then DGEBA epoxy resins for the same molecular weight (or number of
repeating units n). Cured bisphenol F epoxy resins also have increased solvent resistance.
Bisphenol F resins are often mixed with conventional DGEBA epoxy resins because of
the relatively high cost of the bisphenol F product. When mixed with bisphenol A resins,
the two form crystallization-free resins of moderate viscosity.
TABLE 4.3 Thermal Weight Loss, Weight Percent, of Nadic Methyl Anhydride
Cured Epoxy Novolac Resin (Cured 2 h at 90°C, 4 h at 165°C, and 16 h at 200°C)4
Epoxy adhesive formulators have generally addressed the problem of improving flexibility
by adding chemical groups to the epoxy structure—via either the base resin or the curing
agent—or by adding separate flexibilizing resins to the formulation to create an epoxy-
hybrid adhesive system. This section mainly addresses the flexibility achieved through the
epoxy resin structure. The flexibility achieved via additives, hybrid resins, and curing
agents is addressed in later chapters.
Flexibility can be provided through the epoxy resin constituent by incorporating large
groups in the molecular chain, which increases the distance between crosslinks. Gross
changes in the flexibility of a resin system can be obtained only through a major change in
the structure of the cured resin. Such changes include shifting from an aromatic structure
to a more aliphatic hydrocarbon or by reducing functionality. Insertion of long hydrocar-
bon side chains also can provide additional flexibility to the epoxy molecule.
Examples of epoxy resins modified for improved flexibility include the following:
• Epoxy resins derived from acid functional oils (dimer acid, cashew nut oil, and Castor oil)
• Epoxy resins derived from polyalkylene glycol (polyethylene or polypropylene glycol)
• Epoxy resins having increased molecular weight while maintaining the same number of
reactive sites
The reduction in crosslink density increases the flexibility (elongation) of the resulting
molecule, but at the expense of a lowering of the glass transition temperature. This, in turn,
generally results in a significant decrease in tensile and shear strength as well as a decrease
in other performance properties, such as chemical and heat resistance.
EPOXY RESINS 79
TABLE 4.4 Properties of an Epoxy Resin Formulation Based on a Blend of Polyglycol Diepoxy
Flexibilized Resin and Standard DGEBA Resin, Cured with Methylene Dianiline5
Long-chain aliphatic epoxy resins provide flexible molecules with high elongation but
little toughness. They are generally based on polyglycol or vegetable oils reacted with
epichlorohydrin. Because of their lack of hydrolytic stability and lack of strength, they are
generally not used alone but are blended as modifiers with other epoxy resins. Generally
when used in a concentration range of 10 to 30%, they improve flexibility without greatly
impairing other properties.
Glycidyl ethers of aliphatic polyols based on polyglycol, glycerin, and other polyols are
flexible epoxy resins. They are used as reactive diluents and flexibilizers for solvent-free
epoxy resin formulations. Epoxy-polyglycol resins that are produced from the reaction of
epichlorohydrin and polyester polyols based on ethylene or propylene oxide are the most
common of these types of flexible epoxy resins. Examples of typical commercial aliphatic
epoxy resins are shown in App. C.
Commercial products are available where n varies from 2 to 7. The flexible epoxy resins
based on polyglycol also make excellent reactive diluents because they have a viscosity of
100 cP at 25°C. Table 4.4 shows the effect of two polyglycol diepoxides on the physical
properties of a cured epoxy system. Another type of flexible epoxy resin is derived from
dimerized unsaturated fatty acid, cashew nut oils, and other vegetable oils. Other flexible
epoxy resins can be made with thiols, aliphatic acids, and hydroxyl-terminated compounds.
Applications where flexible epoxy resins are valued include
Waterborne epoxy dispersions have been employed effectively for many years in the coatings
market, primarily for the surface protection of concrete and metals. These products were devel-
oped in response to environmental regulations to reduce solvent levels in coatings. Since these
80 CHAPTER FOUR
dispersions adhere well to a wide variety of substrates and provide a high degree of strength
and resistance to service environments, they have naturally found applications as adhesives.
Epoxy resins are hydrophobic and consequently are not, by themselves, dispersible in
water. However, water dispersibility can be conveyed to epoxy resins by two general methods:
Both of these processes are applicable to waterborne epoxy adhesives and coatings, although
the emulsification process is generally used with adhesives.
Chemical modification of the epoxy resin includes either attaching hydrophilic groups to
the epoxy resin or attaching the epoxy resin to hydrophilic polymers. This is most often done
by grafting. For example, one of the largest volume uses for waterborne epoxy is the coat-
ing of metal cans. In this application the epoxy resin is rendered water-dispersible by the
grafting of the epoxy resins to acrylic polymer.
The emulsification method is primarily used for waterborne epoxy adhesive systems
and is the focus of this section. The epoxy resin is made water-dispersible by partitioning
the epoxy resin within a micelle, effectively separating the resin from the water. This emul-
sification can be achieved by a suitable surfactant.
The choice of surfactant and processing parameters determines the long-term mechan-
ical and chemical stability of the dispersion. The use of improper surfactants results in
epoxy dispersions of relatively large and unstable particle size, which exhibits noticeable
settling within as little as 1 day. Also, the possibility of hydrolysis of the epoxy group is
present unless the surfactant provides a “protective” function. Through the use of surfac-
tant technology, novel waterborne epoxy dispersions can be manufactured that, with few
exceptions, contain no organic cosolvents, are mechanically and chemically stable, and are
formaldehyde-free, and several are FDA-acceptable.6
The surfactant selection determines the emulsion properties, such as stability, particle
size, viscosity, and internal phase content. A correct balance between the hydrophobic and
hydrophilic character of the emulsifier is necessary for minimizing the surfactant concen-
tration at the resin-water interface. The surfactants used in resin emulsification can be ionic
(in most cases anionic), nonionic, polymeric, or a combination of these.
To emulsify liquid epoxy resins, high-MW, nonionic emulsifiers can be used to develop
high solids (typically 55 to 60 percent) epoxy resin emulsions with a particle size of less
than 1 micrometer (µm).7 A typical surfactant concentration for this application is about 8
to 10 percent by weight based on the resin. The physical and chemical characteristics of
several nonionic surfactants are given in Table 4.5. The emulsification can be done with-
out the use of any solvents, although small concentrations of solvent are often employed in
waterborne epoxy adhesives and coatings for good film-forming properties.
In addition to the surfactant and epoxy resin, the parameters of the emulsification
process will significantly influence the properties of the final emulsion. To obtain the
smallest achievable droplet size with a narrow droplet size distribution, it is essential to
optimize process parameters such as temperature of emulsification and mix ratio of surfac-
tants when more than one surfactant is used.
The emulsification process is simple but must be carefully controlled. Epoxy resin is loaded
into a high-speed disperser, and the surfactant is added. A defoamer is generally added to pre-
vent excessive aeration, and high-shear mixing is employed. Water is then slowly added to the
mixture. The system at this stage has the epoxy resin as the continuous phase and the water as
the dispersed phase. As the water addition continues, the ratio of the dispersed phase to the con-
tinuous phase increases until a phase inversion occurs. The inversion occurs at about 65 per-
cent volume ratio of dispersed to continuous phase and is accompanied by a rapid reduction in
viscosity. Water addition is then continued until the desired solids concentration is achieved.
Additional additives and modifiers can be incorporated into the formulation at this stage.
EPOXY RESINS 81
There are several epoxy resin chemistries that are appropriate for waterborne adhesives.
These can be broadly classified into three types:
The base epoxy resin can be either liquid or solid. As molecular weight increases, the epoxy
equivalent weight and the number of hydroxyl groups available for reaction increase.
Waterborne epoxy adhesives provide excellent adhesion to metals and other high-energy
substrates. Modified waterborne epoxy adhesives can also provide good adhesion to sub-
strates such as vinyl and flexible plastic film. Characteristics of these epoxy dispersions are
summarized in Table 4.6.
TABLE 4.6 Characteristics of Major Classes of Epoxy Dispersions Used in Adhesive Applications
44 5 3,000
10 2,500
20 2,000
50 1,500
50 5 25,000
10 20,000
20 12,000
50 7,500
Epoxy resin emulsions are commercially available from several sources. As a group, the
typical particle size of the dispersion is in the 0.5- to 3.0-µm range. Solids typically range
from 50 to 65 percent, and viscosity from 10,000 to 12,000 cP. The dispersions, in general,
are thixotropic as supplied. There is also a dramatic decrease in viscosity of the system with
the addition of water. Table 4.7 shows the effect of dilution and Brookfield viscosity spin-
dle speed (thixotropy) on a typical epoxy emulsion.
There are basically two types of epoxy acrylate resins used in formulating adhesive sys-
tems. One is a vinyl ester resin that is used in two-component adhesive formulations much
as a DGEBA epoxy or a polyester resin is. The other is a special type of resin that is used
in radiation cure processes. This latter type of epoxy acrylate does not have any free epoxy
groups, but reacts through its unsaturation.
Epoxy acrylate resins or “vinyl esters” are made from the esterification of epoxy resin
via their terminal group with an unsaturated acid, such as methacrylic acid derived from
epoxy resin. A typical reaction sequence is illustrated in Fig. 4.3. The resultant polymer is
usually dissolved in a reactive monomer such as styrene.
Three types of vinyl ester resins are commercially available:
These resins are primarily used as a resin matrix in reinforced plastic composites and in
structural adhesives.
O
R CH CH2 + 2CH2 CH COOH R CH CH2 O C CH CH2
O OH
2 2
10
FIGURE 4.3 Reaction of epoxy resins with acrylic to produce epoxy acrylate resin.
EPOXY RESINS 83
These resins act more as polyester resins than they do as epoxy resins. They are easily
processed, have fast cure rates at room temperature, and can be cured with peroxides. They
effectively wet out glass fiber reinforcement and cure quickly. Therefore, vinyl ester resins
are often used in the manufacture of composites such as filament- wound and pultruded
structures.
However, once crosslinked, vinyl esters have much greater strength, stiffness, and
toughness than conventional polyester resins. They have excellent chemical resistance and
mechanical properties at both room and elevated temperatures. Bisphenol A vinyl esters are
relatively tough (4 percent elongation), and heat distortion temperatures can range from 93
to 260°C depending on the cure agents and processing conditions.
In adhesive systems, vinyl esters impart low-viscosity, flexibility, and superior wetting
characteristics to DGEBA-type epoxy resins. Because of their good chemical resistance they
are often found in construction applications such as flooring and grouting. However, their
shrinkage is greater than that of any conventional epoxy resins, and often the formulator will
have to counteract this.
TABLE 4.8 Various Types of Epoxy Acrylates That Are Commonly Used for UV/EB Curing
Adhesives and Coatings11
Aromatic difunctional Members of this group have very low molecular weight, which gives
epoxy acrylates them attractive properties such as high reactivity, high gloss, and low
irritation. Common applications for these resins include overprint
varnishes for paper and board, wood coatings for furniture and flooring,
and coatings for compact discs and optical fibers. Aromatic
difunctional epoxy acrylates have limited flexibility, and they yellow
to a certain extent when exposed to sunlight. The aromatic epoxies are
viscous and need to be thinned with functional monomers. These
monomers are potentially hazardous materials.
Acrylated oil epoxy These epoxy acrylates are essentially epoxidized soybean oil acrylate.
acrylates These resins have low viscosity, low cost, and good pigment
wetting properties. They produce relatively flexible coatings.
Acrylated oil epoxy acrylates are used mainly in pigmented
coatings or to reduce cost.
Epoxy novolac These are specialty products. They are mainly used in the electrical and
acrylates electronics industries because of their excellent heat and chemical
resistance. However, they provide rigid coatings with relatively
high viscosity and high costs.
Aliphatic epoxy Several varieties are available as difunctional and trifunctional or higher.
acrylates The difunctional types have good flexibility, reactivity, adhesion, and
very low viscosity. Some difunctional types can be diluted with water.
The trifunctional or higher types have moderate viscosity and poor
flexibility but excellent reactivity. Aliphatic epoxy acrylates have
higher cost than the aromatic epoxy acrylates and are generally
used in niche applications.
Miscellaneous epoxy This group consists mainly of oligomers with fatty acid modification.
acrylates They provide good pigment wetting properties and higher molecular
weight but lower functionality than other aromatic epoxy acrylates.
They are used in printing inks and pigmented coatings.
84 CHAPTER FOUR
Epoxy acrylates are also commonly used as oligomers in radiation-curing coatings and
adhesives. However, their name often leads to confusion. In most cases, these epoxy acrylates
have no free epoxy groups left but react through their unsaturation. These resins are formu-
lated with photoinitiators to cure via uv or electron beam (EB) radiation. The reaction mech-
anism is generally initiated by free radicals or by cations in a cationic photoinitiated system.
The uv/EB cured epoxy formulations are discussed in Chap. 14.
Epoxy acrylate oligomers that are used in uv/EB curing are very low-viscosity systems
with high vapor pressures. Within this group of oligomers, there are several major subclas-
sifications: aromatic difunctional epoxy acrylates, acrylated oil epoxy acrylate, novolac
epoxy acrylate, aliphatic epoxy acrylate, and miscellaneous epoxy acrylates. Characteristics
of these various classes are summarized in Table 4.8.
REFERENCES
1. Meath, A. R., “Epoxy Resin Adhesives,” Chapter 19 in Handbook of Adhesives, 3d ed., I., Skeist,
ed., van Nostrand Reinhold, New York, 1990.
2. Weil, E. D., and Levchik, S., “A Review of Current Flame Retardant Systems for Epoxy Resins,”
Journal of Fire Sciences, vol. 22, January 2004, pp. 25–40.
3. Leisewitz, A., et al., Chapter V, Summarized Substance Evaluation: 2. Tetrabromobisphenol A
(TBBPA), UBA Report 204 08 542, Substituting Environmentally Relevant Flame Retardants:
Assessment Fundamentals, Oko-Recherche, Frankfurt am Main, Germany, 2001.
4. Meath, A. R. “Chemistry, Properties, and Applications of Epoxy Novolac, Flexible Epoxy, and
Flame Retardant Epoxy Resins,” Chapter 3 in Epoxy Resin Technology, P. F. Bruins, ed.,
Interscience Publishers, New York, 1968, p. 34.
5. Meath, “Chemistry, Properties, and Applications of Epoxy Novolac, Flexible Epoxy, and Flame
Retardant Epoxy Resins,” p. 38.
6. Buehner, R. W., and Atzinger, G. D., “Waterborne Epoxy Dispersions in Adhesive Applications,”
Epoxy Resin Formulators Conference, San Francisco, February 20–22, 1991.
7. Uniqema Synperonic Surfactants, “Resin Emulsification for Waterborne Coating and Adhesives,”
Uniqema, Wilmington, DE, 1999.
8. Uniqema Synperonic Surfactants, “Resin Emulsification.”
9. Buehner and Atzinger, G.D., “Waterborne Epoxy Dispersions in Adhesive Applications.”
10. Savia, M., “Epoxy Resin Adhesives,” Chapter 26 in Handbook of Adhesives, 2d ed., I. Skeist, ed.,
van Nostrand Reinhold, New York, 1977, p. 438.
11. Radiation Cure Center, www.SpecialChem4Coatings.com, 2004.
CHAPTER 5
EPOXY CURING AGENTS
AND CATALYSTS
5.1 INTRODUCTION
A variety of curing agents and catalysts will react with epoxy resins to provide crosslinked
adhesives. The curing agents generally react with the available epoxy or hydroxyl groups.
The catalysts initiate homopolymerization of the epoxy groups. Six main classifications of
curing agents are commonly utilized with epoxy adhesive formulations, and these can be
further divided into several subclassifications.
Some curing agents can also be used to form adducts with epoxy resins, and these
adducts, in turn, offer unique curing properties with other epoxy resins. Adducts are com-
monly used in adhesive formulations to reduce the vapor pressure of the system, to modify
the reactivity of the system, to improve mix ratios so that they are closer to equal parts of
the resin component and of the curing agent component, and to provide a certain degree of
flexibility into the end product.
The choice of a particular curing agent or catalyst depends on the processing require-
ments (e.g., viscosity, pot life, application method, curing temperature, reactivity, mix
ratio) and the end-use requirements (e.g., thermal and chemical resistance, shear strength,
toughness) of the cured adhesive. The curing agents along with the epoxy resin determine
the type of chemical bonds and the degree of crosslinking that will occur.
The advantages, disadvantages, and applications for the major types of epoxy curing
agents are summarized in Table 5.1. The required mix ratios, curing temperatures, and the
resulting heat distortion temperatures of the cured product are provided in Table 5.2.
Curing agents can be used to improve the flexibility of inherently rigid epoxy resins.
Certain common epoxy formulations can be flexibilized by altering the stoichiometric mix
ratio of resin to curing agent or by changing the type of curing agent to one that has a more
flexible molecule. For example, by changing from hexahydrophthalic anhydride to hexamine-
ethylenediamine, one can double the impact resistance of a resin system and increase its
tensile elongation at break.4
85
Copyright © 2006 by The McGraw-Hill Companies, Inc. Click here for terms of use.
TABLE 5.1 Advantages, Disadvantages, and Applications of Common Epoxy Curing Agents1,2
Depending on the number of amine groups in the molecule, the amine can be a mono-, di-,
tri-, or polyamine. Aliphatic amines can also be classified by their molecular structure as lin-
ear, branched, aliphatic, or containing aromatic groups. However, the most valuable method
of classification is by functionality.
The functionality of an amine is determined by the number of reactive hydrogens pres-
ent on the molecule. The amines typically have greater than three reactive sites per mole-
cule that facilitate the formation of the crosslinked epoxy structure. The difference
between primary, secondary, and tertiary amines is reflected by the number of hydrogens
that are bound to each nitrogen atom:
The primary and secondary amines are discussed in this section. The secondary amines are
derived from the reaction product of primary amines and epoxies. They have rates of reactiv-
ity and crosslinking characteristics that are different from those of primary amines. The sec-
ondary amines are generally more reactive toward the epoxy group than are the primary
amines, because they are stronger bases. They do not always react first, however, due to steric
hindrance. If they do react, they form tertiary amines. Tertiary amines are primarily used as cat-
alysts for homopolymerization of epoxy resins and as accelerators with other curing agents.
The chemical structures of important amines for curing epoxy resins in adhesive sys-
tems are identified in Fig. 5.1. Diethylenetriamine (DETA), triethylenetetramine (TETA),
n-aminoethylpiperazine (AEP), diethylaminopropylamine (DEAPA), m-phenylenediamine
(MPDA), and diaminodiphenyl sulfone (DDS) are the most commonly used members of this
class. They are all primary amines. They give room or elevated temperature cure at near
stoichiometric ratios. Ethylenediamine is too reactive to be used in most practical adhesive
formulations. Polyoxypropyleneamines (amine-terminated polypropylene glycols) impart
superior flexibility and adhesion.
Amine curing agents are one of the most common types of curing agents for epoxy resins
and can be primary or secondary. Aliphatic amines, such as diethylenetriamine or triethyl-
enetetramine, were the first epoxy curing agents. Although they provided systems with
acceptable properties, wider usage of epoxy adhesives has led to the synthesis of amine
derivatives, mixtures, and epoxy adducts with a wide range of improved properties to meet
specific user requirements.
Primary and secondary aliphatic amines react relatively rapidly with epoxy groups at
room or lower temperature to form three-dimensional crosslinked structures. The resulting
cured epoxies have relatively high moisture resistance and good chemical resistance, par-
ticularly to solvents. They also have moderate heat resistance with a heat distortion tem-
perature in the range of 70 to 110°C. Thus, short-term exposures of cured adhesive joints
at temperatures up to 100°C can generally be tolerated.
These aliphatic amines can also be cured at elevated temperatures to provide a more
densely crosslinked structure with better mechanical properties, elevated-temperature
performance, and chemical resistance. Table 5.3 illustrates the effect of curing temperature
on the bond strength of DGEBA epoxy with two different aliphatic amines.
EPOXY CURING AGENTS AND CATALYSTS 89
Aliphatic
NH2CH2CH2NHCH2CH2NH2 Diethylenetriamine (DETA)
CH3 CH3
Poly(oxypropylene diamine)
NH2CHCH2(OCH2CH)n NH2
CH3
CH2(OCH2CH)nNH2
CH2(OCH2CH)nNH2
CH3
Aromatic
NH2
NH2 m-phenylenediamine
Other amines, such as aromatic or cycloaliphatic, are less reactive and generally require
elevated-temperature cures that result in higher heat distortion temperatures (140 to 150°C).
However, aromatic amine adducts of liquid epoxies can be accelerated to cure at room tem-
perature. Aliphatic amines can also be accelerated.
Epoxy resin formulations containing aliphatic amines will blush and provide an oily sur-
face under very humid conditions. This is due to a reaction of the amine primary hydrogen
atoms with carbon dioxide. Resistance to blushing is more important for coatings than for
90 CHAPTER FIVE
Bond strength on
aluminum, psi
adhesives. Amines that are exposed to the ambient conditions for long times will also
absorb moisture from the air. This can result in a weak boundary layer if the moisture is
trapped at the interface before joint assembly. Modifications of aliphatic amines have been
developed to counteract such problems.
A tertiary amine is often used as an accelerator in primary amine systems to correct for
these problems and to adjust the rate of cure. Ortho-(dimethylaminomethyl) phenol, DMP-
30, and tris-(dimethylaminomethyl) phenol, DMP-10, are common commercial tertiary
amine accelerators.
The fast reaction rate of aliphatic amines at room temperature can, however, lead to var-
ious problems. Pot life is short, which can lead to an unacceptable working life. The cured
resin is relatively brittle due to its high crosslink density. High exotherms in thick sections
or large masses can lead to thermal decomposition of the resin.
In adhesive formulations, aliphatic amines are most commonly used to cure the DGEBA
type of epoxy resin. Aliphatic amines are not widely used with the non–glycidyl ether
resins, since the amine-epoxy reaction is slow at low temperatures. The reaction usually
requires heat and accelerators for an acceptable rate of cure. Aliphatic amines are primar-
ily used with lower-viscosity DGEBA resins because of the difficulty in mixing such low-
viscosity curing agents with the more viscous epoxy resins.
In the well-recognized epoxy–aliphatic amine reaction, the primary or secondary amine
adds to the epoxy ring, forming a tertiary amine, as shown in Fig. 5.2 (top). The formed
hydroxyl groups accelerate the amine curing, and with excess epoxy present, the secondary
hydroxyl groups can also add to the epoxy ring, as shown in Fig. 5.2 (bottom).
The resin and polyfunctional amine must interact in approximately stoichiometric amounts
to obtain the best balance of properties. Thus, mix ratios and their precision are considered crit-
ical for optimum performance. It is assumed that the functionality of the amino group toward
the epoxy is one active amino hydrogen for each epoxy ring. (See Sec. 2.5 for calculating the
stoichiometric mix ratios.) Since the tertiary amines weakly catalyze the reaction of epoxy
groups with each other, slightly less than stoichiometric ratios of amine curative to epoxy
groups are generally used when a tertiary amine catalyst is present.
The complicated reaction mechanism described above leads to the conclusion that cur-
ing epoxy adhesives with amines can be highly complex. Considerable study and skill must
be exercised in (1) selecting the amine curing agent, (2) determining the curing conditions,
and (3) maintaining the curing conditions from adhesive joint to joint to ensure perfor-
mance consistency.
EPOXY CURING AGENTS AND CATALYSTS 91
O OH O
RN CH2CH
CH2CH OH
OH
OH O
OCH2CH OH
FIGURE 5.2 Top: Addition of primary or secondary amine to epoxy ring. Bottom: Addition of
secondary hydroxyl groups to epoxy ring.7
Aliphatic polyamines are corrosive and may result in skin sensitization upon prolonged
exposure. The high vapor pressure is a significant disadvantage with this type of curing agent.
Evaporation of the curing agent due to heating or vacuum degassing can lead to lower than
anticipated amine concentrations. The high vapor pressure can lead to gassing in adhesive
joints that are cured at elevated temperatures. High vapor pressure also magnifies the toxicity
and skin irritation properties of the amines.
120 6
100 4
90 TETA 3
80 2
70 1
8 9 10 11 12 13 14
pph of curing agent
FIGURE 5.3 Effect of concentration of DETA and TETA curing agents on heat deflec-
tion temperature of DGEBA.8
at operating temperatures greater than 50°C. For maintenance of properties at higher tem-
peratures, the aromatic amines are preferred as curing agents.
and color stability. Cycloaliphatic amines cure well at low temperatures, even under damp
conditions. A range of working lives and cure schedules are possible depending on the type
of cycloaliphatic amine that is used in the epoxy adhesive formulation. When cured at ele-
vated temperatures, cycloaliphatic amines provide high glass transition temperatures, excel-
lent chemical resistance, and strong mechanical properties. However, they are relatively
brittle and exhibit low peel strength and poor impact characteristics in unmodified systems.
Aminoethylpiperazine (AEP) is a cycloaliphatic amine possessing primary, secondary,
and tertiary amine groups. Only the primary and secondary groups are involved in the
DGEBA curing mechanism. AEP is a clear, high-boiling-point liquid with a low vapor pres-
sure. It is often used instead of DETA or TETA when improved toughness is required. The
Izod impact strength of AEP cured DGEBA resins is about 2 to 3 times greater than that of
epoxy systems cured with DETA or TETA.
Optimum cures with AEP catalyzed DGEBA are obtained using 20 to 22 pph. However,
the crosslink density is not great, and the cured product does not attain a high degree of ther-
mal or chemical resistance. The pot life and exotherm are similar to those of DETA and
TETA, but a postcure (2 h or more at 100°C) is required to develop properties fully.
Other cycloaliphatic diamines such as isophorone diamine, bis-p-aminocyclohexylmethane
and 1,2-diaminocyclohexane are used as epoxy resin curing agents for both ambient and
heat cured epoxy resin systems. While they have advantages, such as light color and good
chemical resistance, they provide rather sluggish cure rates at low temperatures.
Dimethylaminopropylamine (DMAPA) provides similar cure characteristics and prop-
erties to DEAPA, but it has a slightly shorter pot life.
Secondary amines can be represented by piperidine and diethanolamine. Secondary
amines when used alone may be considered a special class of tertiary amine in that after the
secondary hydrogens have reacted, the resulting crosslinking is believed to occur via the
tertiary amine mechanism. These curing agents generally have limited temperature resis-
tance, poorer chemical resistance, and equivalent mechanical properties to the primary
amines. The curing temperature significantly influences the reactivity, heat generation, and
properties of the cured resin. In adhesive formulations, they are used as blends with primary
amines for providing specialized properties.
There are several reasons why unmodified aliphatic amines, such as those described above,
are less widely used than other curing agents in epoxy adhesive systems. These include
Most of these objections can be overcome through modification of the aliphatic amine
by reducing the density of active hydrogens on the molecule. As a result, amines have been
converted to adducts of higher molecular weight and viscosity by reaction with mono- or
polyfunctional glycidyl compounds. These adducts have generally decreased volatility and
irritancy, more convenient mix ratios, and lower reactivity. They also produce cured epoxy
systems with somewhat greater toughness.
The aliphatic amines can be adducted with mono- and diepoxies that are prepared from
ethylene and propylene oxide, from acrylonitrile, from aldehydes, and from a variety of
other compounds. When cured with DGEBA epoxy resins, these curing agents do not offer
significantly different end properties other than the properties mentioned above for unmod-
ified amines. However, because there are longer molecular chains between active hydrogen
groups, the flexibility and toughness of the cured resin are improved at the disadvantage of
a lower crosslinking density.
Ethylene and Propylene Oxide Amine Adducts. Polyamines, such as DETA, react read-
ily with ethylene oxide in the presence of water to yield mono- or dihydroxyalkyl derivatives,
depending on the ratio of reactants. As the extent of the reaction progresses, the resulting
compound contains fewer and fewer active
H hydrogens. The most common commercial
H2N(CH2 CH2 N)2 CH2 CH2 OH product in this class is hydroxyethyldieth-
N-Hydroxyethyldiethylenetriamine ylenetriamine. Several others are noted in
H
Fig. 5.4.
HO(CH2 CH2 N)3 CH2 CH2 OH The advantages of these products are
N,N′-Bis(hydroxyethyl)diethylenetriamine
low skin irritation potential and low viscos-
ity. Physical properties and reactivity suffer
CH3 somewhat due to the reduction of active
H amine hydrogens of DETA to four in the
H2NCH2 CH2 NCH2 CH OH
case of the monoadduct of ethylene oxide
N-(2-Hydroxypropyl)ethylenediamine
and three when the bisadduct is formed.
CH3 CH3 These compounds tend to be hydroscopic
and must be stored in tightly closed contain-
(HOCHCH2)2 NCH2 CH2N (CH2 CH OH)2 ers. High humidity will interfere with cure,
N,N,N′,N′-Tetrahydroxypropylethylenediamine particularly in thin films. The slow cure can
FIGURE 5.4 Ethylene and propylene oxide adducts be overcome by the addition of bisphenol A,
of amines. which acts as an acid accelerator.
EPOXY CURING AGENTS AND CATALYSTS 95
Two curing agents that have found their way into many epoxy adhesive formulations are
the polyamides and amidoamines. These are commonly used in the “hardware store vari-
ety” two-part epoxy resins that cure at room temperature. Both are reaction products of
aliphatic amines, such as diethylenetriamine, and should be included under the subclassifi-
cation of modified amines. However, these products have such widespread and popular use,
they are addressed here as a separate classification.
When compared to other curing agents, polyamides and amidoamines offer the follow-
ing three unique features.
1. They can be used over a broad range of noncritical concentrations including one-to-one.
2. They are less volatile, and as a result they have fewer odors and lower skin-irritating
potential than do other room temperature curing hardeners.
3. They provide a modest amount of flexibility that is directly related to the concentration
of curing agent used in the epoxy formulation.
5.3.1 Polyamides
Polyamide curing agents are the reaction products of dimerized fatty acids and aliphatic
amines such as diethylenetriamine. This introduces a bulky, oil-compatible, C36 carbon group
between the amine sites. Similar to the diglycidyl ether adducts of aliphatic amine, they are
manufactured by adding the fatty acid to an excess of amine. They are available in a range of
viscosities that can be achieved by varying the amine/acid molar ratio in the reaction.
The reactive sites are the terminal primary and internal secondary amine groups along the
backbone of the polyamide. The amide groups for all practical purposes are not reactive. As a
result of their relatively large molecular weight, the amide groups add flexibility to the final
crosslinked structure. Generally, polyamides are used as room temperature curing agents.
These products have higher molecular weight than do primary amines and, therefore,
exhibit lower vapor pressure, resulting in lower irritation potential and odor. The various
molecular weight (MW) derivatives show different degrees of compatibility with epoxy
resin. Some high-MW polyamides might show incompatibility with epoxies unless a par-
tial reaction is accomplished. This is known as an induction period, and it ensures compat-
ibility. Fortunately, the polyamide curing agents offer a long pot life and low exotherm, so
that the induction period is not usually a detriment.
Polyamide mix ratios can be very forgiving and are less critical than with aliphatic poly-
amines. They are generally used at a 50- to 100-pph level with DGEBA epoxy resins. A mix-
ture consisting of 50 parts polyamide and 50 parts DGEBA epoxy can provide moderately
good physical properties in most environments. Increasing the curing agent levels yields
increased flexibility and adhesion but reduces heat distortion temperature and chemical
resistance.
Polyamide cured DGEBA epoxies provide improved flexibility, moisture resistance,
and adhesion over aliphatic amines alone. However, polyamide cured epoxies are generally
inferior in thermal resistance and shear strength due to the reduction in crosslink density.
Polyamide cured epoxies lose structural strength rapidly with increasing temperatures and
96 CHAPTER FIVE
Amidopolyamine, pph
are generally limited to applications under 65°C. A further disadvantage of polyamide cur-
ing agents in certain applications is that they have a much darker color than polyamines.
The Versamid series of curing agents were the original and best known polyamide cur-
ing agents. Current commercial polyamide curing agents and their properties are shown in
App. D. More recently developed polyamides provide lower viscosity, better compatibility
with epoxy resins, and better cure profiles under adverse conditions.
5.3.2 Amidoamines
Amidoamine or polyamidoamine curing agents have reactivity with DGEBA epoxy resins
that is similar to the polyamides. However, they are lower-viscosity products and are also
lower in color. Amidoamines are derivatives of monobasic fatty carboxylic acids and
aliphatic polyamines. Since the amidoamines have only one amide group per molecule,
they are lower in molecular weight, viscosity, and amine functionality than the polyamides.
Amidoamines have noncritical mixing ratios with large curing agent/resin ratios simi-
lar to those of polyamide curing agents. Similarly, the physical properties of the cured prod-
uct can be varied significantly by altering the mix ratio. Table 5.5 shows properties of
DGEBA epoxy that is cured with varying concentrations of an amidopolyamine derived
from tall-oil fatty acid and TETA.
Both the polyamide and amidoamine curing agents can be accelerated by the addition
of a tertiary amine such as DMP-10, tris(dimethylamino methyl) phenol.
Some products in the amidoamine group are manufactured to contain a significant amount
of imidazoline structures. This is accomplished by a high reaction temperature that converts
the open amide structure to the cyclic imidazolin with loss of water. This conversion leads to
lower viscosities since the concentration of the polar amide group is reduced.
Amidoamines exhibit very good adhesion characteristics, particularly to porous sub-
strates such as concrete and wood. They also cure extremely well under humid conditions.
They are much less corrosive than aliphatic amines and provide less skin irritation.
Aromatic amines are widely used as curing agents for epoxy resins. However, they are not
used as widely in adhesive formulations as they are in composites, molding compounds,
and castings. They offer cured epoxy structures with good heat and acid resistance.
EPOXY CURING AGENTS AND CATALYSTS 97
The primary advantages of aromatic amines over aliphatic amines for curing epoxy resins
are the longer pot life as well as the development of higher heat resistance and greater chem-
ical resistance. The major disadvantages are that they are solids at room temperature and gen-
erally require heat for processing as well as for cure. The added heat required for mixing and
cure increases the dermatitis and toxicity potential by releasing irritating vapors. Modifications
of aromatic amines are available that cure at room temperatures.
The lower reactivity of the aromatic amines in adhesive formulations is an advantage in
that epoxy resin mixtures can be B-staged at room temperature (react to a glassy but fusible
and thermoplastic intermediate structure) and will not fully cure for months. In this way,
dry films and solid powders can be formulated as elevated-temperature curing, one-component
adhesives with long shelf life.
The color of aromatic amines is poor (dark), and they stain easily. They are generally
solid materials that require some formulating at elevated temperatures to produce a prod-
uct that can be easily handled. The vapors resulting from elevated temperatures can cause
staining, and their irritancy can be a problem. Certain aromatic amines such as diamin-
odiphenylmethane are carcinogenic.
When compared to aliphatic amines, aromatic amines generally have reduced exotherm
and reactivity. Elevated temperatures are required to achieve optimum properties. In cer-
tain cases aromatic amines can be cured at room temperature with catalysts such as phe-
nols, BF3 complexes, and anhydrides.
Examples of aromatic amines are shown in Fig. 5.5. Of these compounds the most com-
mon are meta-phenylene diamine (MPDA), methylene dianiline (MDA), and eutectics of
the two.
Metaphenylene diamine (MPDA) is one of the most common of the aromatic amines used to
cure epoxies. This product is amber to very dark in color. It is a solid that melts at 65°C and
is generally mixed with the epoxy resin at that temperature. The molten liquid or vapors from
MPDA can stain the skin and nearby structures rather badly. The para-isomer is reported to
be carcinogenic, but the meta-isomer is free from this disadvantage.
MPDA has four active hydrogens and is used stoichiometrically with DGEBA at
14.5 pph. Once it is in solution within an epoxy resin, the resulting mixture has excellent
NH2 NH2
NH2
Single-ring
NH2
NH2
NH2
MPDA IPDA 1,2 DACH
FIGURE 5.5 Chemical structures of common aromatic amine curing agents.
98 CHAPTER FIVE
handling characteristics and low viscosity. There are three methods of mixing this curing
agent with epoxy resins.
1. Heat both the MPDA and the epoxy resin to 65°C; the two components can then be
blended.
2. Heat the resin to about 80°C and then, while stirring continuously, dissolve the curing
agent into the resin.
3. Use technical-grade MPDA that can be heated to about 80°C and then supercooled slowly
to room temperature to obtain a liquid, which is stable for at least 9 months at room
temperature.
In all of the above cases, a considerable amount of objectionable staining and irritating
fumes will be given off during the mixing operation. After mixing, the pot life will be about
6 h at 25°C.
Once the pot life is exceeded at room temperature, the MPDA–epoxy mixture will B-stage.
This technique is used to produce dry filament winding materials (prepreg) and solid molding
compounds. In adhesive compounding this technique can be used to produce one-component,
dry adhesives in the form of solid stick, powder, or film. To cure the B stage, the product is
exposed to temperatures in the range of 150 to 175°C, which causes the B-staged material to
flow and then cure. The adhesive is then postcured at 175°C for optimal property formation.
The B stage can also be dissolved in solvent and used to impregnate reinforcement of a carrier.
Methylene dianiline (MDA) is also a solid diaromatic amine. Similarly to MPDA, MDA is
not often used in adhesive formulations because of the difficulty in compounding and cur-
ing practical epoxy formulations and the resulting brittleness of cured structures. MDA also
has relatively low polarity so its adhesion properties would be suspect.
MDA has a melting temperature of about 90°C. It has four active hydrogens, and its sto-
ichiometric mixing ratio with DGEBA epoxy resin is in the range of 27 to 30 pph. The
resulting viscosity is somewhat greater for MDA–epoxy resin mixtures than for MPDA.
The mixing and curing procedure is similar to that described above for MPDA except that
higher temperatures are required. The pot life of MDA mixtures with DGEBA epoxy resins
(20 h) are somewhat longer, and the reaction rates are slower than they are for MPDA.
MDA B-stages epoxy resins similarly to MPDA, and is used to manufacture stick solder and
other solid adhesive forms.
MDA does not stain skin or equipment as badly as MPDA. For this reason along with its
lower cost, MDA is often preferred to MPDA in epoxy formulations. MDA does not provide
the high-temperature strength or chemical resistance of MPDA for equivalent cure conditions.
However, these properties are significantly superior to those of epoxy resins cured with pri-
mary amines.
Aromatic Amine Eutectics. There are several curing agents available that consist of
eutectics of various aromatic amines. These perform very much as MPDA and MDA do.
However, the eutectics are liquids with viscosity of approximately 2000 cP at room tem-
perature. They are readily miscible with liquid epoxy resins at room temperature.
The aromatic amine eutectics may crystallize on storage. They can be reliquefied by
heating to 40°C with stirring. This liquefaction can be accomplished without sacrificing
either the curing properties or the final physical and chemical properties of the cured resins.
EPOXY CURING AGENTS AND CATALYSTS 99
Solvent Solutions. Certain solvent solutions of aromatic amines have been noticed to
polymerize epoxy resins at room temperature.11 The effect of the solvent is probably to allow
sufficient mobility of the polymer chains for an adequate degree of crosslinking to occur
before the viscosity becomes so high that the molecules are immobilized. The aromatic
amine solutions are usually used with a cure accelerator to achieve practical cure rates at
room temperature.
However, the incorporation of the solvent in the cured resin will significantly lower the
glass transition temperature and thermal resistance. When cured at room temperatures, these
solutions give properties more similar to those of the polyamide curing agents, but they do have
the advantage of low viscosity and adjustable cure rate.
5.5 ANHYDRIDES
After the primary amines, acid anhydrides are the next most important class of epoxy curing
agents, although these are not used as often in adhesive systems as they are in casting com-
pounds, encapsulants, molding compounds, etc. The most common types of anhydrides are
hexahydrophthalic anhydride (HHPA), phthalic anhydride (PA), nadic methyl anhydride
(NMA), and pyromellitic dianhydride (PMDA), although there are several others. Chemical
structures of several anhydrides are illustrated in Fig. 5.6.
These compounds do not readily react with epoxy resins except in the presence of water,
alcohol, or some other base, called an accelerator. Tertiary amines, metallic salts, and imi-
dazoles often act as accelerators for anhydride cured epoxy systems. The reaction between
acid anhydride and epoxy resins is illustrated in Fig. 5.7.
The reaction of anhydrides with epoxy groups is complex, with several competing reac-
tions capable of taking place. The most significant reaction mechanisms are as follows:
1. The opening of the anhydride ring with an alcoholic hydroxyl forms the monoester.
2. Subsequent to the opening of the ring, the nascent carboxylic groups react with the epoxy
to provide an ester linkage.
3. The epoxy groups react with nascent or existing hydroxyl groups, catalyzed by the acid,
producing an ether linkage.
At low elevated-temperature cures, the ether and ester reactions take place at the same fre-
quency. At higher temperatures the ester linkage is predominant. The presence of a catalyst
O
C
Phthalic anhydride O
C
C
Tetrahydrophthalic O
anhydride C
C
Methyltetrahydrophthalic O
anhydride C
CH3
O
C
Hexahydrophthalic O
anhydride
C
C
Nadic methyl anhydride O
C
CH3
O
O
Cl
Cl C
Chloroendic anhydride O
Cl C
Cl
O
Cl2
100
EPOXY CURING AGENTS AND CATALYSTS 101
CO
C
R OH O
C
CO
Alcohol Anhydride
CO O R CO O R
O
C C
R CH CH2
C C
Epoxy
CO O R CO O CH2 CH R
OH
FIGURE 5.7 Anhydride epoxy reaction.
can change the balance of ester-ether linkages. The presence of ester linkages is believed
to result in reduced elevated-temperature performance.
Liquid and solid anhydrides are used extensively to cure epoxy resins in such applica-
tions as casting, potting, and reinforced plastics. They are valued in these applications
because of their relatively high heat distortion temperature, good physical properties, low
exotherm, and long pot life when mixed in large masses. However, their use in adhesive
formulations is generally limited except for high-temperature applications because of the
rather slow reactivity, long and elevated-temperature cure requirement, and inferior adhe-
sion compared to amine cured epoxies.
Anhydrides are sometimes used in epoxy adhesives to provide specific properties or to
provide improvements in handling strengths. The most important anhydride in epoxy adhe-
sive formulations is pyromellitic dianhydride (PMDA), which provides very high temper-
ature properties.
The mix ratio of anhydride to epoxy resin is less critical than with amines and can vary
from 0.5 to 0.9 equivalent of epoxy. The specific ratio is generally determined experimen-
tally to achieve desired properties. Compared to aliphatic amine cures, the exotherm gen-
erated by anhydride cured epoxies is low. Elevated-temperature cures up to 200°C and
postcures are required to develop optimal properties.
The high elevated-temperature cures are damaging to adhesive systems due to a mismatch
in thermal expansion coefficient that can occur between the epoxy and the substrate. The dif-
ference in rate of expansion when returning to room temperature from the cure temperature can
lead to significant internal stress within the adhesive joint, which results in poor adhesion.
The reactivity of the epoxy-anhydride reaction is slow; therefore, an accelerator is often
used at 0.5 to 3 percent to speed the gel time and cure. Most often the accelerator is a ter-
tiary amine, and the optimum concentration is dependent on the anhydride, the resin used,
and the cure conditions. The accelerator concentration, like the anhydride concentration, is
usually determined experimentally based on a specific set of end properties.
Anhydrides are hygroscopic materials and should not be allowed to remain exposed to
the atmosphere for extended periods. Absorption of moisture from the air or from fillers
causes hydrolysis of the anhydride to the acid. When used to cure epoxy resins, this mois-
ture absorption results in variable pot life, reduced thermal resistance, and other problems.
As a result, drying of fillers is particularly recommended for anhydride systems.
102 CHAPTER FIVE
Nadic methyl anhydride (NMA) is the most versatile of all the anhydrides. NMA is a liq-
uid of viscosity about 200 cP at room temperature, and it is readily soluble in epoxy resins.
The mix ratio is 60 to 90 pph when used with a liquid DGEBA epoxy resin. At 60 pph and
with no catalyst, the working life is about 2 months; and with the incorporation of 0.5 pph
DMP-30 as an accelerator, the working life reduces to 4 to 5 days.
The balance of properties for the final crosslinked resin can be varied over a wide range
by altering the resin/curing agent ratio, changing the type and concentration of the acceler-
ator, and modifying the cure conditions. In general, the highest degree of crosslinking and
hardness is obtained by using stoichiometric mixtures of NMA and long cure schedules at
high temperatures. Decreasing the amounts of anhydride and cure temperature leads to an
improvement in toughness but at a reduction in heat resistance.
Wide variations in cure schedule are possible. To attain the highest heat resistance, a cure
of about 2 h at 220 to 260°C is required. A 90-pph NMA concentration plus an imidazole
accelerator results in a more practical cure of 2 h at 80 to 100°C plus 4 h at 140 to 150°C.
O
O
C COOROOC C
O
O
C COOH
COOH C
O
O
13
FIGURE 5.8 Idealized PMDA-glycol adduct.
The final two techniques described above are generally used in the preparation of PMDA
cured adhesives.
PMDA can be reacted with glycols to produce an adduct having the general structure
shown in Fig. 5.8. These adduct resins form in the presence of solvent, under dry nitrogen.
The reaction is continued until a clear solution is obtained. Such PMDA adducts are used
in the formulation of high-temperature adhesive films.
A PMDA dispersion is prepared by mixing finely powdered PMDA into liquid epoxy
resins at room or slightly elevated temperature by stirring. No noticeable settling will occur
in resins having an initial viscosity greater than 5000 cP.
Because of its high functionality, PMDA can also be used in monoepoxy resins. These
systems produce heat distortion temperatures on the order of 150°C. Cure times are rela-
tively long, but may be accelerated by the addition of glycols or acidic accelerators.
Other acid anhydride curing agents are used for optimization of specific properties such as
electrical strength. Dodecyl succinic anhydrides (DDSA) and adducts of DDSA with poly-
glycols give long pot life formulations with epoxy resins. When used to crosslink epoxy
resins, they provide good heat resistance and excellent electrical properties. DDSA cured
epoxies are also useful for bonding many plastics. They have been found to provide espe-
cially high adhesion to plastics such as polyethylene terephthalate film (e.g., Dupont’s
Mylar) as well as polycarbonate.14
Catalytic curing agents achieve crosslinking by initially opening the epoxy ring and causing
homopolymerization of the resin. The resin molecules react directly with one another, and the
cured polymer has essentially a polyether structure. The catalysts do not themselves partici-
pate in the epoxy polymerization reactions, as do the curing agents described above which
provide a polyaddition reaction mechanism.
Therefore, catalysts merely act as an initiator and promoter of epoxy resins curing reac-
tions. The amounts of catalyst used with epoxy resins are usually determined empirically and
are chosen to give the optimum balance of properties under the required processing condi-
tions. Generally, only several parts per hundred of catalyst is used with an epoxy resin. Excess
amounts of catalyst can result in poor physical properties and degraded resin.
Catalysts can be used with epoxy resins in any of two primary ways:
This section reviews the function of a catalyst as a sole crosslinker in epoxy resin systems.
Their function as accelerators is covered under the sections related to the specific curing agent.
Some catalysts, such as certain Lewis acids, are so reactive that they can provide extremely
short gel times with epoxy resins at room temperature.15 For example, BCl3, can polymerize
epoxy resins, resulting in gel times of less than 60 s. However, these reactive systems result in
a very rigid adhesive with low peel strength properties and poor impact strength. As a result,
less reactive catalysts are commonly employed in adhesive formulations.
The most popular catalysts for epoxy resins are tertiary amines, tertiary amine salts, boron
trifluoride complexes, imidazoles, and dicyandiamide. Many of these catalysts provide very
long pot lives (months) at room temperatures and require elevated temperatures for reaction
with the epoxy groups. These catalysts are often referred to as latent hardeners.
Tertiary amines are a type of Lewis base catalyst and are, perhaps, the most widely used
catalyst. Two of the most widely used tertiary amines are
• DMP-10: tris-(dimethylaminomethyl) phenol
• DMP-30: o-(dimethylaminomethyl) phenol
The DMP designation comes from the original manufacturer of these chemicals, Rohm and
Haas, and continues today with their manufacturer, Resolution Performance Polymers
LLC. There are other tertiary amine catalysts such as benzyldimethylamine (BDMA), pri-
marily salts of the above, and substituted imidazoles.
Generally, when used as a sole catalyst, tertiary amines are used only in specialty appli-
cations where short pot life can be tolerated and where maximum physical or chemical
properties are not required. DMP-10 and DMP-30 are used at concentrations of 4 to 10 pph
with liquid DGEBA epoxy resins. They achieve fairly fast cures overnight, even at room
temperatures since the hydroxyl groups present in the epoxy molecule enhance the catalytic
activity of the tertiary amine groups.
Tertiary amine salts of DMP-30 provide extended room temperature pot life (6 to 10 h at
20°C) when used at concentrations of 10 to 14 pph in liquid DGEBA epoxy resins. They cure
at moderately elevated temperatures (4 to 8 h at 60°C), or even at room temperature with a
heat bump. The acid moiety blocks the tertiary amine centers and deactivates them. The salt
then dissociates on heating, freeing the amine groups, which are then able to react with the
epoxy group.
The tertiary amine salts are claimed to provide epoxy formulations with very good adhe-
sion to metal. The cured resins also show a hydrophobic effect when in contact with water
or at high humidities. The strength, toughness, and elongation (4.7 percent) of the cured
epoxy resin are very good. However, heat distortion temperature is only in the range of 70 to
80°C, and chemical resistance is relatively poor for an epoxy. The physical properties fall
off rapidly with any rise in temperature.
Benzyldimethylamine (BDMA) is another tertiary amine that can be used as either a
sole catalyst or an accelerator with other curing agents. It is used with DGEBA epoxy resins
at 6 to 10 pph. The pot life is generally 1 to 4 h, and the cure will be complete in about 6 days
at room temperature. When used by itself, BDMA can provide epoxy adhesive formula-
tions with high-temperature resistance (Chap. 15). However, BDMA is mostly used as an
accelerator for anhydride and dicyandiamide cured epoxy resins.
5.6.2 BF3-Monoethylamine
Boron trifluoride monoethylamine (BF3-MEA) is a Lewis acid catalyst. Lewis acids are elec-
tron pair acceptors that function as curing agents by coordinating with the epoxy oxygen,
EPOXY CURING AGENTS AND CATALYSTS 105
facilitating transfer of the proton (Fig. 5.9). BF3-MEA is the only Lewis acid that has
achieved broad commercial use in epoxy resin systems. BF3-MEA is an effective catalyst
for the polymerization of linear and cycloaliphatic epoxies as well as for the glycidyl ethers.
BF3-MEA is a complex that is formed between boron trifluoride gas and monoethy-
lamine. It is a solid melting at close to its dissociation temperature (80 to 85°C). BF3-MEA
shows hydroscopic tendencies. On exposure to moist air, it hydrolyzes into a viscous liq-
uid that is unsuitable as a curing agent.
The BF3-MEA must be melted and dissolved in liquid epoxy resins. For small batches,
the procedure is to heat the resin to 85°C and stir in the curing agent. For larger batches,
about 3 parts by weight of the catalyst is stirred into about 5 parts by weight of the resin
preheated to 50°C. This produces a smooth paste, which then may be added to the remain-
der of the resin heated to 85°C. Alternatively, the BF3-MEA can be dissolved in a solvent
such as furfural alcohol, which will also dissolve the epoxy resin.
Although it is formed from a very reactive catalyst (BF3 gas), the monoethylamine
blocks the reactions sufficiently that BF3-MEA can be considered to be a latent catalyst. It
provides a pot life of 6 to 12 months at room temperature. It does not show significant cur-
ing activity until temperatures of 100 to 125°C have been reached.
In most formulations, the concentration of BF3-MEA in DGEBA epoxy resins is on the
order of 2 to 4 pph. Curing temperatures are usually 2 h at 105°C followed by a postcure at
150 to 200°C for 4 h for optimum properties. The rate of cure is very sensitive to tempera-
ture; below 100°C the rate is negligible, and at 120°C it is very rapid and accompanied by
a significant exotherm.
The epoxy product cured with BF3-MEA is densely crosslinked and has excellent phys-
ical properties at high temperatures (150 to 175°C). When reacted with an unmodified epoxy
resin, the resulting product is very hard and brittle. The chemical resistance, however, is only
fair and somewhat less than that of epoxies that are cured with aliphatic amines.
5.6.3 Imidazoles
O R CH2
H CH
The catalyzed resin has a pot life of 8 to 10 h at 25°C and a normal cure of 6 to 8 h at 60°C.
EMI cures to a densely crosslinked structure with liquid DGEBA epoxy resins. These mix-
tures can cure at relatively low temperatures (60°C) or at higher temperatures (170°C) in
very short times.
It has been recognized that the imidazole becomes permanently attached to the polymer
chain in the epoxy curing reaction. Figure 5.10 suggests a possible reaction sequence.16 The
imidazole is thus an effective crosslinking agent, operating both through the secondary and
tertiary amines.
Compared with other catalysts that homopolymerize epoxies, the imidazole offers improved
thermal properties and retention of mechanical properties at more elevated temperatures.
The cured resin has a heat distortion temperature between 85 and 130°C, which can be fur-
ther increased by a postcure to about 160°C.
Latent imidazole catalysts have also been developed to provide cure rates considerably
faster than those of dicyandiamide cured epoxy resins.18 They also exhibit excellent adhe-
sive characteristics and heat and chemical resistance.
A unique feature of these imidazole catalysts is that they do not have the high exotherm
that dicyandiamide produces when cured in epoxy resins. Thus, they do not char or burn
when exposed to high cure temperatures for fast cure. This is an important factor for adhe-
sives that are cured via induction or dielectric heating. These adhesive systems are also
much safer to ship via air freight than conventional dicyandiamide catalyzed epoxy formu-
lations due to their low exotherm.
5.6.4 Dicyandiamide
Dicyandiamide (DICY) is a solid latent catalyst that reacts with both the epoxy terminal
groups and the secondary hydroxyl groups. DICY has the advantage that it only reacts with
the epoxy resins on heating beyond an activation temperature, and once the heat is
removed, the reaction stops. It is widely used with epoxy resins where long shelf life (up to
OH
Me · C NH Me · C N · CH2 · CH
O
CH C · Et + CH2 CH CH C · Et
N N
O
+ CH2 CH
OH
Me · C N Me · C N · CH2 · CH
CH C · Et CH C · Et
NH CH · CH2 · N
+
O
_
12 months) is required prior to curing. Significantly longer shelf lives can be obtained by
storage under refrigeration until use.
As a result of the latency and excellent properties produced by DICY cured epoxies,
DICY is used in many B-staged supported film adhesives. DICY is also probably the lead-
ing catalyst used in one-component, elevated-temperature curing epoxy adhesives.
DICY is considered a catalyst and polymerizes epoxy resin through the homopolymer-
ization mechanism. But DICY has also shown behavior with epoxies that indicates some
breakdown at cure temperatures to produce a curing agent that contributes to the polyaddi-
tion reaction mechanism.
The early reaction mechanism of DICY with epoxy resin consists of the epoxy reaction
with all four hydrogen atoms on DICY and the epoxy-to-epoxy reaction that is catalyzed
by the tertiary amines. The final curing mechanism is between hydroxyl groups in the partly
cured resins and DICY cyano groups. This results in the disappearance of the cyano groups
to form amino groups. This step is also catalyzed by tertiary amines.
DICY is used at about 5 to 7 pph of liquid epoxy resins and 3 to 4 pph for solid epoxy
resins in adhesive formulations. It is generally ball-milled into the epoxy resin. DICY forms
very stable mixtures with epoxy resins at room temperature because the catalyst is not sol-
uble at low temperatures. However, on being exposed to temperatures greater than 140°C,
the DICY becomes soluble in the epoxy resin, and cure progresses rapidly.
The particle size and distribution are important for maximizing the shelf life of epoxy–
DICY systems. Generally optimum properties are produced when the particle size of the
DICY is less than 10 µm. Usually fumed silica is used to keep the DICY particles in sus-
pension and evenly distributed in the epoxy resin.
When formulated into one-component adhesive systems, the product is stable when
stored for 6 months to 1 year at room temperature. It will then cure when exposed to 145 to
160°C for about 30 to 60 min. Since the reaction rate is relatively slow at lower tempera-
tures, the addition of 0.2 to 1 percent benzyldimethylamine (BDMA) or other tertiary amine
accelerators is common to reduce cure times or cure temperatures. Other common acceler-
ators are imidazoles, substituted urea, and modified aromatic amine.
Substituted DICY derivatives have been developed to increase solubility and lower the
required activation temperatures. These techniques can reduce the activation temperatures
for DICY–epoxy resin mixtures to as low as 125°C.
Epoxy resins cured with DICY exhibit a good balance of physical properties with heat
and chemical resistance. The glass transition temperature of a DGEBA liquid epoxy resin
cured with 6 pph of DICY is on the order of 120°C, whereas an elevated-temperature cur-
ing aliphatic amine would provide a glass transition temperature of no greater than 85°C.
Tougheners can be added to the adhesive formulation to achieve relatively high levels of
peel strength and impact strength.
Polymercaptans and polysulfides are aliphatic oligomers containing sulfohydro (−SH) groups
that will react with epoxy groups at room temperature to form cured epoxy structures and
epoxy adducts. A generalized structure is shown in Fig. 5.11.
108 CHAPTER FIVE
HS ( C2H4OCH2OC2H4 S S )n C2H4OCH2OC2H4 SH
O OH
R CH CH2 + R′ SH R′ S CH2 CH R
Epoxy Mercaptan
FIGURE 5.12 Reaction of mercaptan with an epoxy resin.19
EPOXY CURING AGENTS AND CATALYSTS 109
• Activation
• Propagation
RSCH2CH(O−)CH2OR′′ + RSH
RSCH2CH(OH)CH2OR′′ + RS−
The tertiary amines and TETA are used at about 10 pph, and the liquid polysulfide is used
at about 50 to 100 pph.
Polysulfides are typically used at ratios of 1:1 or less with epoxy resins, and can be used
as co-curing agents with aliphatic amines. As the proportion of liquid polysulfide polymer to
epoxy increases, the cured compound becomes softer and has a higher degree of elongation.
Similarly, the tensile strength, heat resistance, and chemical resistance are reduced. Often it
is better to think of these materials as co-resins or co-curing agents in hybrid systems.
Epoxy-polysulfide systems do not generate significant amounts of exotherm, and their
cure rate does not rely on the exothermic reaction. Thus, epoxy-polysulfide compositions
have a cure rate that is relatively insensitive to temperature, and they can cure at very low
temperatures (below room temperature).
Stoichiometric quantities of aliphatic amine and 25 to 50 parts by weight of polysulfide
will react with 100 parts by weight of epoxy resin to yield a relatively flexible product with
good tensile strength at ambient temperatures. As a result of the excellent flexibility, epoxy-
polysulfide resin systems have been used more as sealants and coatings than as adhesives sys-
tems. Because of their excellent adhesion to metals and glass and their cold-weather curing
properties, they are often used in the construction industry as barriers to moisture penetration.
Although the properties are quite good at room temperatures, some degree of flexibility is lost
on thermal aging.
One of the advantages of having an adhesive or sealant with such a low degree of hard-
ness is that the bond can easily be broken either by high shear stress or by simply cutting with
a sharp instrument. Thus, it is easy to salvage expensive components from an assembly.
REFERENCES
1. Meath, A. R., “Epoxy Resin Adhesives,” Chapter 19 in Handbook of Adhesives, 3d ed., I. Skeist,
ed., van Nostrand Reinhold, New York, 1990, p. 350.
2. Cranley, P. E., “Epoxy Adhesives,” Paint and Coatings Industry, April 1994, pp. 42–47.
3. Burgman, H. A., “Selecting Structural Adhesive Materials,” Electrotechnology, June 1965.
4. Epon Resin Structural Resin Manual—Additive Selection, Resolution Performance Polymers
LLC, Houston, Tx, 2001, p. 8.
5. Belm, D. T., and Gannon, J., “Epoxies,” in Adhesives and Sealants, Engineered Materials
Handbook, ASM International, Materials Park, OH, 1990.
6. Houwink, R., and Salomon, G., eds., Adhesion and Adhesives, Elsevier, New York, 1965, p. 247.
110 CHAPTER FIVE
6.1 INTRODUCTION
Solvents and diluents are used to lower the viscosity of epoxy resins systems either to per-
mit easy compounding with other ingredients or to aid in application of the adhesive onto
a substrate. Both solvents and diluents are low-molecular-weight liquid compounds that are
chemically and physically compatible with epoxy resins and their curing agents. They differ
primarily by their vapor pressure.
Solvents have a relatively high vapor pressure and will evaporate given a specific set of
environmental conditions. Certain solvents will evaporate quickly at room temperature and
atmospheric pressure, and others may require heating to elevated temperatures and pres-
sures that are even lower than atmospheric.
Diluents have much lower vapor pressures and generally do not evaporate at ambient
conditions. However, they do have a finite vapor pressure, and given the right set of condi-
tions (time, temperature, and pressure) they will vaporize. Two distinct classes of diluents
are used with epoxy resins: nonreactive diluents and reactive diluents. Reactive diluents will
enter into the crosslinking reaction with the primary resin, and nonreactive diluents will
not. Nonreactive diluents primarily act as low-molecular-weight plasticizers for the epoxy
composition.
Epoxy adhesive formulations demand a great variety of solvents and diluents with a
wide range of evaporation rates, solvent strengths, and dispersion powers. These variations
are required due to (1) the many types of epoxy resins, curing agents, and possible organic
additives that can be used within a formulation and (2) the many different possible meth-
ods that can be used to apply the epoxy to the substrate (brush, spray, trowel, etc.).
The choice of solvent or diluent is made with regard to the solubility of individual com-
ponents and to the viscosity, drying times, and wetting characteristics required of the final
product. All these properties affect the bond performance of the resulting epoxy adhesive
formulation.
6.2 SOLVENTS
Solvents are employed to temporarily lower the viscosity of the epoxy system by impart-
ing a degree of added mobility to the polymeric resins used in the formulation. They are
employed for one or several of the following reasons:
1. To aid in dispersing, mixing, and wetting of components in the resin system at the for-
mulation stage
111
Copyright © 2006 by The McGraw-Hill Companies, Inc. Click here for terms of use.
112 CHAPTER SIX
The solvent used in any adhesive formulation may need to solubilize more than one com-
ponent (e.g., resin, curing agent, or polymeric additive). Solvents are generally organic, and
often a blend of solvents is necessary to achieve the required degree of solubility or to pro-
vide for certain processing conditions (e.g., drying time). Polar solvents are required with
polar resins; nonpolar solvents, with nonpolar resins.
Water is also sometimes used as a solvent for water-soluble resins. In the case of epoxy
resins, water is generally used to disperse epoxy particles in an emulsion. These waterborne
epoxy adhesives are discussed in Chaps. 4 and 14.
When used in epoxy adhesive systems, solvents are generally employed for reducing the
viscosity for formulation or application purposes. Once the adhesive is applied to the sub-
strate, the solvent must evaporate prior to cure. Otherwise, bubbles or vapor pockets could
form in the bond line, causing a physically weak joint with poor adhesion. The solvent in
the adhesive formulation must not adversely affect the substrate to which it is applied.
Plastics, elastomers, and polymeric foams are especially sensitive to certain solvents used
in epoxy adhesives.
Examples of solvents typically used in epoxy adhesive systems are acetone, methyl
ethyl ketone (MEK), toluene, xylene, glycol ethers, and alcohols. Generally, the efficiency
of the solvent declines with increasing epoxy molecular weight (MW). Table 6.1 presents
viscosity data for DGEBA (molecular weight 1000 to 3000) that is dissolved in various sol-
vent systems at a concentration of 40% by weight. The viscosity versus resin concentration
for several DGEBA solvent solutions is illustrated in Fig. 6.1.
Solvent blends are often used to provide for a specific evaporation rate and degree of
solvency. These blends normally consist of combinations of fast-evaporating and slow-
evaporating solvents. Fast-evaporating solvents are based on low-boiling-point ketones
such as acetone or MEK. Slow-evaporating solvents are based on higher-boiling-point
(lower-vapor-pressure) compounds.
“True” solvents are considered to be those solvents that provide resin solutions which
can be diluted to infinity without resin precipitation. True solvents for epoxies include
MEK, diacetone alcohol, methylcyclohexanone, and most glycol ethers and their acetates.
Acetone, not a true solvent, can be used to prepare 40% solutions of high-MW DGEBA
epoxy resin, but not a 20% solution. Aromatic solvents such as toluene and xylene, as well
as simple alcohols, such as isopropyl and n-butyl alcohol, are not active solvents for DGEBA
resins; however, they can be used in combination with other solvents to improve solubility.
The choice of solvent or solvent blend for a particular epoxy adhesive formulation will
primarily depend on
Ketones:
Acetone 200
Methyl ethyl ketone 10 10 250
Methyl isobutyl ketone 10 125 1100
Diacetone alcohol >1500
Isophorone >1500
Esters:
Ethyl acetate >1500
n-Butyl acetate 1300
Cellosolve acetate 125 450 >1500
Ether alcohols:
Methyl Cellosolve >1500
Ethyl Cellosolve 125 450 >1500
Butyl Cellosolve >1500
Ethyl carbitol >1500
Butyl carbitol >1500
Chlorinated solvents:
Trichloropropane >1500
Chloroform 220 1050 >1500
Mixed solvents:
Toluene/acetone (1/1) 300
Toluene/methyl ethyl ketone (1/1) 350
Toluene/methyl isobutyl ketone (1/1) 600–900
Toluene/diacetone alcohol (1/1) 25 200 >1500
Toluene/isophorone (1/1) >1500
Toluene/isopropyl alcohol (1/1) >1500
Toluene/Cellosolve acetate (1/1) 10 300 >1500
*
Values shown are viscosity, cP, for a 40% solvent solution.
can be formulated to a specific resin system. One typical solvent blend for a DGEBA with
molecular weight of 950 consists of the following:
Xylene 32 pph
Methyl isobutyl ketone 32 pph
Cellosolve 32 pph
Cyclohexanol 4 pph
Solutions made from this solvent blend were found to provide good leveling and flow prop-
erties and resulted in a practical evaporation rate.2
By proper combination of solvents, it is possible to reduce the viscosity of a given resin
system while preserving the solids content at a given percent. For instance, a typical formu-
lation consisting of 70% DGEBA with 30% of solvent system A (below) provides a viscosity
114 CHAPTER SIX
40,000
Molecular weight of epoxy resin
1000 1500 3000 4000 Dissolved in MIBK-MIBC-
20,000
toluene-xylene, 1-1-1-1
Dissolved in “Cellosolve”
10,000
8000 acetate-toluene, 1-1
6000
4000
2000
Viscosity, cP
1000
800
600
400
200
100
80
60
40
20
10
80 70 60 50 40 30 20
Resin concentration, % weight
FIGURE 6.1 Viscosity of DGEBA resins in solution at 25°C.1
of 5000 cP that is suitable for brush application, whereas the same resin concentration with
solvent system B provides a viscosity of 2400 cP that is suitable for spraying.3
Even with proper solvent balance and using fast-drying solvents, it is probable that some
solvent will remain entrapped in the cured resin unless the cure is preceded by an elevated-
temperature exposure to eliminate the solvent. For example, MEK retention in amine cured
DGEBA films was noticed after 9 days at room temperature.4
Some of the solvents that are commonly used in epoxy resins can present a flammabil-
ity hazard and special health hazards. Contact with solvents will cause drying of the skin,
which may result in an increased probability of skin irritation, especially when one comes
in contact with curing agents. Solvents also have the ability to dissolve epoxy resin system
components and carry them through the skin in liquid form or into the respiratory system
in vapor form. The inhalation of solvent vapors or mist may cause respiratory irritation and
SOLVENTS AND DILUENTS 115
problems with the central nervous system. Chapter 18 discusses the safety and health-
related issues of epoxy formulation components including solvents.
The solvent industry has made significant strides in developing newer grades and blends
of solvents for a variety of applications. “Safety” solvents are being developed that are low
in volatility (vapor pressure), low in toxicity, and biodegradable. However, these newer
solvents are finding commercial acceptance mainly as cleaning solvents rather than as a
dilution medium for epoxy resins. For adhesives, rather than replace the solvent, the trend
has been to develop waterborne emulsions.
With regard to cleaning solvents, the industry has developed solvents that are either
biodegradable and/or environmentally compatible. New low-volatility solvents are taking
the place of the older, less environmentally safe solvents in the adhesive and sealant indus-
tries. Substitute solvents, such as those shown below, manufactured by Inland Technology,
Inc., have been found acceptable for some industrial applications.
Solvent Low-volatility, nonchlorinated substitutes
There is also an excellent web site, SAGE (http://clean.rti.org), that provides a compre-
hensive guide to pollution prevention information on solvent and cleaning process alterna-
tives. The U.S. EPA Air Pollution Prevention and Control Division developed SAGE
(Solvent Alternatives Guide).
The effect of solvent type on the curing rate of epoxy reactions has been well defined.
Hydroxyl compounds, such as alcohols, act as catalysts and accelerate curing. However,
these solvents are not serious competitors with amines for reacting with the epoxy ring.
Water, functioning as a hydroxyl compound, also accelerates the reaction, even more than
alcohols. Aprotic solvents, such as aromatic hydrocarbons or mineral spirits, have no effect
on the amine-epoxy resin and behave as inert diluents. Carbonyl solvents, such as acetone
and methyl ethyl ketone, retard the reaction.
Acceleration by the hydroxyl groups will affect the pot life, penetration, film forma-
tion, adhesion, and other critical properties. Figure 6.2 shows that the viscosity buildup for
100
80
60
Butanol
Viscosity, P
40
Methyl
isobutyl ketone
20
10
0 1 2 3 4 5 6
Time, h
FIGURE 6.2 Effect of solvent type on the pot life of a hydrogenated bisphenol A diglycidyl
ether cured with a polyamide.5
116 CHAPTER SIX
a solubilized epoxy formulation is much more rapid when the solvent is butanol than when
it is methyl isobutyl ketone.
6.3 DILUENTS
Diluents are higher-MW components than solvents that are also added to the epoxy adhe-
sive formulation to lower the viscosity and modify processing conditions. The primary
function of a diluent in an epoxy resin formulation is to reduce its viscosity to make it eas-
ier to compound with fillers, to improve filler loading capacity, or to improve application
properties. Solvents, certain curing agents, and flexibilized epoxy resins can also lower the
viscosity of epoxy adhesive formulations, but this is not their primary function. The effect
of various diluents on the initial viscosity of a diglycidyl ether of bisphenol A (DGEBA)
epoxy resin is illustrated in Fig. 6.3.
Lower viscosity is important in applying the adhesive because it determines what type
of mixers and dispensers are required and if the epoxy can be trowled, brushed, or sprayed.
10,000
8000
6000
4000
1000
800
Styrene oxide
600
Allyl
400 glycidyl
ether
Xylene
200
100
0 5 10 15 20 25
Reactive diluent, pph
FIGURE 6.3 Effect of various diluents on the viscosity of a standard DGEBA
liquid epoxy resin.6
SOLVENTS AND DILUENTS 117
Lower viscosity is also important in achieving good adhesion in that it allows greater
penetration of porous substrates and faster wetting of the microroughness on nonporous
surfaces. Diluents also increase the working life of the catalyzed epoxy system by (1) increas-
ing the time that the mixture’s viscosity is below a workable limit and (2) decreasing the reac-
tivity of the curing agent primarily because of dilution of the resin.
Although most diluents decrease the reactivity of the epoxy system, reactive diluents
may increase the exotherm because of the heat release of the high number of epoxy groups
per gram in the diluent. Diluents containing alcoholic hydroxyl groups accelerate the cur-
ing rate in the presence of amine curing agents. However, when low percentages of diluent
are used, these effects on reactivity and exotherm are generally minor.
Diluents will also affect the performance properties of the adhesive. Diluents generally
lower the degree of crosslinking and degrade the physical properties of the cured epoxy.
This reduction in crosslink density increases the resiliency of the adhesive, but it also
reduces tensile strength as well as heat and chemical resistance. These effects are more pro-
nounced at elevated temperatures than at room temperature. The degree of these effects will
depend on whether the diluent has epoxy functionality (reactive diluents) or whether the
diluent is incapable of reacting with the epoxy system (nonreactive diluents).
Reactive diluents are generally preferred over nonreactive diluents because they are
chemically linked into the epoxy network. However, they still degrade the physical prop-
erties because their functionality is lower than that of the resin. Nonreactive diluents can be
thought of as plasticizers. Because nonreactive diluents are relatively mobile, they can be
more easily driven off on heating or vacuum degassing. If this occurs during cure, the result
is greater shrinkage and appearance of vapor bubbles. This results in reduced adhesive
strength because of the internal stresses created in the joint. Entrapped nonreactive diluents
can also migrate out of the cured adhesive during service conditions, thereby causing a
change in properties.
Both reactive and nonreactive diluents should be used sparingly if the properties of the
cured system are to be preserved. An amount of 5 to 10 pph is best and generally provides
a sharp reduction in viscosity. Concentrations greater than 20 pph are seldom used in adhe-
sive formulations.
Diluents are generally not as much a problem as solvents in causing skin irritation, but
they are low-viscosity, relatively high-vapor-pressure compounds and could lead to irri-
tation in certain applications. Many of the diluents, especially those containing epoxy
groups, are more severe skin irritants than the epoxy resins themselves. This is due to
their lower molecular weight and high vapor pressures. The viscosity reduction capabil-
ity (and skin irritation tendency) for the diluents is directly related to their molecular size.
With the larger molecules the skin hazard potential is less, but so is the viscosity-reducing-
efficiency.
Nonreactive diluents do not react with the resin or curing agent and, therefore, generally
dilute the final physical properties of the epoxy structure. In essence, they act as plasticizers
to the epoxy network. In addition to lowering viscosity, nonreactive diluents are often used
to balance the mix ratio proportions in certain epoxy systems. However, nonreactive dilu-
ents have not found wide acceptance in epoxy adhesive technology, primarily because most
are incompatible with the cured resin and because similar effects can be achieved by proper
selection of a long-chain curing agent or a reactive diluent.
High-boiling-point solvents such as xylene can be used as a nonreactive diluent; how-
ever, this is not normally done because of the high vapor pressure of the solvent and the
probability that solvent remaining after cure would degrade the physical properties and
118 CHAPTER SIX
80
14,000
12,000 60
10,000
8,000 40
6,000 TETA curing agent
4,000 20
2,000
0 0
0 10 20 30 40 0 10 20 30 40
Percent phenol Percent phenol
(a) (b)
FIGURE 6.5 (a) Viscosity versus nonyl phenol concentration in DGEBA epoxy. (b) Gel
time versus nonyl phenol concentration in catalyzed DGEBA epoxy.8
phenol and 5 pph antimony oxide to produce a self-extinguishing system. Another nonre-
active diluent, polymethyl acetal, can be used to improve tensile shear strength in room
temperature curing epoxy adhesive formulations.9
Reactive diluents enter into the polymerization reaction of the epoxy resin and the curing
agent. In this way the final adhesive characteristics are determined by the reaction product
of the binder and the diluent. The most common reactive diluents used for epoxy adhesive
formulations are shown in Table 6.2.
Most reactive diluents are mono- or difunctional. They are made by reacting epichloro-
hydrin with an alcohol, a phenol, or a polyol to produce a mono- or polyglycidyl ether resin.
However, there are some nonepoxy diluents that are used as well. These nonepoxy diluents
generally react with the curing agent or other functional groups in the epoxy chain.
In addition to viscosity reduction, the presence of a reactive diluent generally leads to a
faster rate of cure and a higher crosslink density than with an undiluted resin. This is due
to the fact that by lowering the viscosity the molecules become more mobile in the early
stages of cure. This increased mobility results in a more reactive system. Table 6.3 shows
the effectiveness of various reactive diluents as viscosity reducers for DGEBA epoxy resins
and their effect on the heat distortion temperature.
The primary reactive diluents are monoepoxy low-molecular-weight epoxy resins. These
may be used at rather high concentration with little effect on cured properties. Often they are
employed to make selective improvements on certain properties such as adhesion, thermal
cycling resistance, and impact strength. Some lower-viscosity commercial epoxy resins are
already reduced with these diluents.
The monofunctional epoxy diluents are essentially chain stoppers since they inhibit
crosslinks from forming. The extent to which the cured properties are affected is directly
dependent on the concentration of the diluent added to the epoxy resin. The general effect
is to reduce viscosity and improve the impact and thermal shock resistance while slightly
reducing the thermal resistance. The thermal expansion of the cured resin is increased, as
it is also with nonreactive diluents. This can lead to internal stress on the bond line depend-
ing on the thermal expansion of the substrate material.
Monofunctional epoxy diluents are used primarily with DGEBA epoxy blends. The most
common monofunctional diluents are butyl glycidyl ether and phenyl glycidyl ether. The effect
of butyl glycidyl ether and other reactive diluents on the viscosity of epoxy resin is shown in
Fig. 6.3. Because the monofunctional diluents reduce crosslink density, they are used at
relatively low levels to avoid degrading heat and chemical resistance or other properties of the
adhesive.
The reaction of polyols with epichlorohydrin produces polyglycidyl ethers, such as
butanediol diglycidyl ether. These are only slightly less effective viscosity reducers than
butyl glycidyl ether, but they are less volatile and contribute less to worker irritation prob-
lems. They also produce less loss in physical properties.
Difunctional epoxy diluents are low-viscosity, low-MW epoxy resins. These diluents
may be used at very high concentrations, and they do not greatly affect the properties of the
cured resin. In some instances the difunctional epoxy diluents will actually improve proper-
ties. These materials include butadiene dioxide, vinyl cyclohexane dioxide, and diglycidyl
ether of resorcinol.
Long-chain polyether polyols, such as polypropylene glycol, when reacted with
epichlorohydrin produce a diepoxy resin with an internal polyether chain that serves both to
TABLE 6.3 Viscosity Reducing Power of Reactive Diluents and Their Effect on Heat Distortion
Properties11
decrease viscosity and to increase flexibility. Other reactive diluents that provide flexibility
to the cured epoxy structure can be synthesized from reacting epichlorohydrin with an
organic acid or polybasic acid to form glycidyl esters. The reactions of a polyol and acid with
epichlorohydrin to form a reactive diluent
are illustrated in Fig. 6.6.
Since many of the difunctional epoxy
diluents are added to epoxy adhesive formu- CH2 CH2
P ·O O
lations as flexibilizers as well as viscosity
CH 2 CO
reducers, they are discussed in greater detail 3
in Chap. 8 with other flexibilizing additives. Triphenyl phosphite γ-butyrolactone
There are also several reactive diluents FIGURE 6.7 Chemical structures of triphenyl
that do not contain epoxy groups. These phosphite and γ-butyrolactone.
are represented by triphenyl phosphite and
γ-butyrolactone (Fig. 6.7). The phosphite is
a low-viscosity colorless liquid, which is
sensitive to moisture. It reacts with hydroxyl groups in the resin. The γ-butyrolactone is a
very effective viscosity reducer. It can reduce the viscosity of a liquid DGEBA from about
15,000 to 2000 cP with only 10 pph. In the curing reaction with amines (Fig. 6.8), the lac-
tone forms an amide, which can then crosslink with the polymer via the hydroxyl groups.
Other nonepoxy-containing reactive diluents may be co-curing agents, unsaturated mol-
ecules, active hydrogen-containing molecules, materials that are capable of promoting
transesterification with available hydroxyl groups, and materials capable of reacting with
the curing agents. These will affect the physical properties by influencing the crosslink den-
sity and system functionality.
It should be pointed out that diluents are not the only way to lower the viscosity of filled
epoxy resins systems. Surface active agents can also be added to the system. They provide
better wetting of the filler by the epoxy resin matrix. This can lead to substantial viscosity
reduction for systems having equivalent filler concentration. The surface active agent, in
turn, could also be used to produce formulations with higher filler loading at equivalent vis-
cosity. These surface active agents are discussed in Chap. 10.
CH2 CH2
O + R NH2 HO (CH2)3 CO NHR
CH2 CO
REFERENCES
1. Shell Chemical Company, Technical Literature. Also see Lee, H., and Neville, K., Handbook of
Epoxy Resins, McGraw-Hill, New York, 1967, p. 24.30.
2. Alter and Sooler, “Molecular Structure of Epoxy Resin Polymers as a Basis for Adhesion,” ACS
Symposium, New York, September 1957.
3. Lee and Neville, Handbook of Epoxy Resins, p. 24–32.
4. Dannenberg, H., et al., “Infrared Spectroscopy of Surface Coatings in Reflected Light,” Analytical
Chemistry, March 1960.
5. Bauer, R. S., “Formulating Weatherable Epoxy Resin for Maximum Performance,” Waterborne
and Higher Solids Coating Symposium, New Orleans, 1982.
6. Behm, D. T., and Gannon, J., “Epoxies,” in Adhesives and Sealants, vol. 3, Engineered Materials
Handbook, ASM International, Materials Park, OH, 1990.
7. Potter, W. G., Epoxide Resins, Springer-Verlag, New York, 1970, p. 110.
8. Lee and Neville, Handbook of Epoxy Resins, p. 13.6.
9. Russel, D. H., Epoxy Resin Composition, U.S. Patent 3,050,474, 1962.
10. Meath, A. R., “Epoxy Resin Adhesives,” in Handbook of Adhesives, Skeist, I., ed., 3d ed., van
Nostrand Reinhold, New York, 1990, p. 355.
11. May, C., and Tanaka, Y., Epoxy Resins: Chemistry and Technology, Marcel Dekker, New York,
1973, p. 299.
12. Bolger, J., “Structural Adhesives State of the Art,” Chapter 7 in Adhesives in Manufacturing, G. L.
Schneberger, ed., Marcel Dekker, New York, 1983, p. 172.
13. Potter, Epoxide Resins, p. 115.
CHAPTER 7
HYBRID RESINS
7.1 INTRODUCTION
A variety of polymers, both thermosets as well as thermoplastics, can be blended and co-
reacted with epoxy resins to provide for a specific set of desired properties. The most com-
mon of these are nitrile, phenolic, nylon, polysulfide, and polyurethane resins. At high
levels of additions these additives result in hybrid or alloyed systems with epoxy resins rather
than just modifiers. They differ from reactive diluents in that they are higher-molecular
weight-materials, are used at higher concentrations, and generally have less deleterious
effect on the cured properties of the epoxy resin.
These hybrid epoxy adhesives are generally used for demanding structural applications
such as in the aerospace industry where the optimal properties from each component are
desired. For example, epoxy is generally used to provide good adhesion and processing
characteristics. They are blended with the following resins to provide additional improve-
ments in the properties noted.
Phenolic––high-temperature resistance
Nylon––toughness and peel strength
Polysulfide––elongation
Vinyl––high resiliency and peel strength
Nitrile––toughness and peel strength, chemical resistance
Polyurethane––peel and impact strength
Hybrid resins have been used to improve the flexibility, thermal shock resistance, elon-
gation, heat distortion temperature, and impact strength of unmodified epoxy adhesives.
However, there can also be some sacrifice in certain physical properties due to the charac-
teristics of the additive. These alloys result in a balance of properties, and they almost never
result in the combination of only the beneficial properties from each component without
carrying along some of their downside.
These blends can take a number of different forms. The added resin may be reacted with
the epoxy resin, or it may be included as an unreacted modifier. The modifier may be
blended into a continuous phase with the epoxy resin (epoxy alloys) or precipitated out as
a discrete phase within the epoxy resin matrix (as is generally done in the case of toughen-
ing modifiers). Epoxy hybrid adhesives are often used as film (supported and unsupported)
or tape because of the ease with which formulated systems can be dissolved into solvent
and applied to a carrier or deposited as a freestanding film. Some systems, notably epoxy-
urethanes and epoxy-polysulfides, can be employed as a liquid or paste formulation because
of the low-viscosity characteristics of the components.
123
Copyright © 2006 by The McGraw-Hill Companies, Inc. Click here for terms of use.
124 CHAPTER SEVEN
Epoxy-nylon Films are blends containing 30 to 50% by weight of epoxy resin. The
nylon constituent provides high tensile shear strength as well as high
peel strength. Suitable catalysts are dicyandiamide and aromatic
polyamines. These adhesives have useful properties at low
temperatures but have only moderately good high-temperature
properties.
Epoxy-phenolic Epoxy-phenolic adhesives are generally used in aerospace applications
requiring high shear strength at temperatures in excess of 150°C.
Usually the phenolics are a resole-type, and often the epoxy is a minor
component. These adhesives are relatively brittle and have low peel
and impact strengths.
Epoxy-nitrile Nitrile-epoxy adhesives are composed of solid epoxy resin modified
with carboxyl-terminated butadiene nitrile (CTBN) copolymer. The
CBTN is introduced into the epoxy resin at elevated temperatures. The
modification provides toughness and high peel strength without
sacrificing heat and chemical resistance. The film adhesives are
widely used in the aerospace industry in the construction of jetliners.
Epoxy-vinyl Vinyl-epoxy adhesives have moderate strength at temperatures up to
150°C. Oxidation stability is excellent. Vinyl constituents increase
the toughness and peel strength. The adhesive is often used for bonding
safety glass, aerospace engine structures, and structural panels.
Epoxy-polysulfide Epoxy-polysulfide adhesives and sealants have a very high degree of
elongation but poor tensile strength compared to unmodified epoxies.
They are used primarily as sealants or coatings in the building and
construction industries.
Epoxy-polyurethane Epoxy-polyurethane adhesives have a great deal of toughness. They
have high elongation and very good peel strength and impact properties.
They also have a relatively high degree of tensile strength, given their
elongation. These adhesives bond well to flexible substrates such as
plastics. They also bond well to oily surfaces. The main market for
epoxy-polyurethane adhesives is automotive.
1. Epoxy resins that are toughened with elastomeric and nonepoxy resins
2. Alloyed blends consisting of epoxy-phenolic, epoxy-nylon, and epoxy-polysulfide
adhesives
The first group, resins that are used primarily to toughen epoxy adhesive systems, is
described in Chap. 8. This chapter focuses on the resinous modifiers that are used as alloy
blends. Characteristics of commercially available epoxy alloy adhesives are presented in
Table 7.1. Tensile shear and peel strengths that are typical of these hybrid adhesives are
compared in Table 7.2.
This section focuses on the modification of epoxy resins by blending with acrylonitrile
butadiene (nitrile) resins. These are true alloyed blends since the nitrile rubber usually con-
tains no groups that are normally reactive with epoxy groups. The nitrile molecules and the
epoxy molecules intermingle as a blend to provide a single-phase alloy. If a large elastomer
concentration is used, no phase separation will occur to form precipitates.
These single-phase hybrids are very different from the two-phase toughened epoxy-
nitrile adhesives that are discussed in Chap. 8. These two-phase adhesives have redefined
structural adhesives to a great extent and have opened the door to many applications that
were previously not possible because of the epoxy resin’s inherent rigidity. The polymer
mixtures that exist as separate phases provide significant increases in toughness but have
only a small improvement in elongation at typical use levels.
The single-phase epoxy-nitrile adhesive is made by choosing an epoxy resin–curing
agent combination in which the nitrile rubber is soluble. Therefore, after cure the adhesive
is a clear, rubbery, single-phase resin. The nitrile component is introduced into the epoxy
resin at elevated temperatures. The blend is cured with standard epoxy curing agents and
catalysts at elevated temperatures. Nitrogen-containing curing agents are generally used
with these modifiers because they catalyze the reaction of the carboxyl group with the
epoxy functionality.
The final product is generally an adhesive solution or an adhesive film. The film is
widely used in the construction of commercial jetliners where a high degree of peel strength
and good resistance to moisture is required. The cured hybrid generally has a Shore A hard-
ness of 75 and an elongation of over 100 percent.
The epoxy-nitrile adhesives were introduced commercially in the late 1960s. They con-
sisted primarily of DGEBA epoxy resin modified with carboxyl-terminated butadiene
nitrile (CTBN) rubber. These first nitrile copolymers were available from B.F. Goodrich
under the trade name of Hycar. The most convenient form of epoxy nitrile adhesive, espe-
cially when one is bonding large parts (aircraft structures), is a supported film. However,
solvent solutions of epoxy-nitrile adhesives have also been commercially available.
Table 7.3 shows the effect of adding a nitrile rubber to a rigid epoxy formulation. There
is significant improvement in toughness and elasticity in the cured bond line, which trans-
lates into high tensile shear and peel strength. As a result, nitrile-modified epoxy adhesives
have found application in bonding airframe structures and other applications where a com-
bination of high tensile strength with peel and impact resistance is required.
The improvements in peel strength compare favorably with the improvements provided by
epoxy resins that are blended with nylon (discussed below). One advantage of the nitrile-epoxy
126 CHAPTER SEVEN
TABLE 7.3 Comparison of Bond Strengths of Epoxy Adhesive Formulations with and without
Nitrile Rubber Addition1
formulations is that they provide a higher degree of peel strength retention at subzero tem-
peratures. The nitrile-epoxy adhesives also provide better resistance to moisture.
However, the single-phase epoxy nitrile adhesive achieves its high peel strength by bulk
elongation. Other properties, such as heat and chemical resistance, are generally degraded
as they are when other flexibilizers or plasticizers are added to the epoxy. The two-phase
version of this adhesive hybrid solves many of these problems.
7.3 EPOXY-PHENOLIC
Epoxy-phenolic adhesives are made by blending epoxy resins with phenolic resins to
improve the high-temperature capabilities of the standard epoxy resins. Developed in the
early 1950s, they were the first “high-temperature” epoxy adhesives to become commer-
cially available.2,3
Epoxy-phenolic adhesives were developed primarily for bonding metal joints in high-
temperature applications. Their first major application was to join major aircraft components.
They are also commonly used for bonding glass, ceramics, and phenolic composites.
Because of their relatively good flow properties, epoxy phenolics are also used for bond-
ing honeycomb sandwich composites.
These adhesives are generally based on blends of solid epoxy resins with resole-type
phenolic resin. The epoxy resin component is often not the predominant component in the
blend, depending on the end properties required. Phenolics are compatible with epoxy
resins and will react through the phenolic hydroxyl group. The amount of phenolic resin
used is generally much greater than that required to crosslink with the epoxy, so one can
debate whether (1) the epoxy toughens the phenolic adhesive or (2) the phenolic increases
the heat resistance of the epoxy.
Adhesives based on epoxy-phenolic blends are good for continuous high-temperature
service in the 175°C range or intermittent service as high as 260°C. They retain their prop-
erties over an extended temperature range, as shown in Fig. 7.1. Shear strengths of up to
3000 psi at room temperature and 1000 to 2000 psi at 260°C are possible. Resistance to
weathering, oil, solvents, and moisture is very good. However, because of the rigid nature
of the constituent materials, epoxy-phenolic adhesives have relatively low peel and impact
strength and limited thermal-shock resistance.
These adhesives are available as pastes, solvent solutions, and film supported on glass
fabric. The adhesive films generally give better strengths than do liquid systems. Cure
requires a temperature of 175°C for 1 h under 15- to 50-psi pressure. Most adhesives of this
HYBRID RESINS 127
2
Phenolic-
nitrile
1
0
−400 −200 0 200 400 600
Temperature, °F
FIGURE 7.1 The effect of temperature on the tensile shear strength
of modified epoxy-phenolic compared to other hybrid adhesives (sub-
strate is aluminum).4
type have limited storage life at room temperature and are generally stored under refriger-
ated conditions.
Typical compositions of epoxy-phenolic adhesives that have been commercially avail-
able over the years are shown in Table 7.4. Note that additives such as quinolate or gallate
act as chelating agents to inhibit the reaction with iron in metallic substrates, such as stain-
less steel, at high temperatures. The current trend in formulating epoxy-phenolic adhesive
is to increase the epoxy content for greater toughness and toward using acid accelerators
for faster cure. The epoxy-phenolic adhesives are composed with other high-temperature
epoxy adhesives in Chap. 15.
7.4 EPOXY-NYLON
Epoxy-nylon adhesives, introduced in the 1960s, were one of the first structural adhesives
designed specifically to have high shear strength and extremely high peel strength. These
characteristics are achieved by blending epoxy resins with copolymers from the polyamide
(or nylon) resin family which is noted for toughness and tensile properties.
Adhesive was prepared by dissolving nylon (Zytel 61), epoxy resin (ERL 2774), and curing agent
(2,4-dihydrozino-6-methamino-S-triazine) in methanol/water/furfural alcohol solvent solution. The adhesive
was applied to acid-etched aluminum, the solvent was allowed to evaporate, and cure was 60 min at 175°C.
HYBRID RESINS 129
TABLE 7.6 Effect of Humidity and Water Immersion on the Tensile Shear Strength of
Epoxy-Nylon and Nitrile-Phenolic Adhesives*,10
resistance. However, they do not achieve the moisture resistance associated with unmodi-
fied epoxy or nitrile-epoxy adhesives.
Epoxy-nylon adhesives are generally available as unsupported film or as solvent solu-
tions. Adhesive films are made either by solution casting or by dry mixing and calendering.
Common solvent solutions are blends of methanol, water, and furfural alcohol. A curing
agent is 2,4-dihydazino-6-methylamino-S-trazine. Other suitable catalysts are dicyandi-
amide and aromatic polyamines. Typical formulations are shown in Table 7.7.11
A pressure of 25 psi and temperature of 175°C are required for 1 h to cure the epoxy-
nylon adhesive. Because of their excellent filleting properties and high peel strength,
epoxy-nylon adhesives are often used to bond aluminum skins to honeycomb core in air-
craft structures. In these applications, climbing drum peel strengths in excess of 150 lb/in
have been achieved.
TABLE 7.7 Effect of LP-3 Polymer on the Physical Properties of Liquid Epoxy Resin13
Formulation, pbw
Component A B C D E F G
7.5 EPOXY-POLYSULFIDE
Polysulfide compounds that are compatible with epoxy resins are liquid elastomers at room
temperature. The most significant commercial resin of this type is LP-3 from Toray. The
predominant product is a mercaptan-terminated liquid polymer (LP) that contains approx-
imately 37% bound sulfur (see Fig. 7.2). It is the high concentration of sulfur linkages that
provides these products with their unique chemical properties. A sulfur odor is noticeable
during processing, making ventilation important.
Polysulfides can be cured by themselves with an oxidizing agent as a catalyst. They can
also be used to cure epoxy resins (see Chap. 5); however, the rate of cure is very slow for
a practical adhesive. Thus, polysulfide resins are generally added into epoxy formulations
as a modifier to increase flexibility. In applications where maximum flexibility is required,
the level of polysulfide may be greater than the epoxy resin present in the formulation.
For adhesive systems, the liquid epoxy resins most widely used with LP-3 polymers are
liquid unmodified and diluent-modified bisphenol A resins and liquid blends of bisphenol
A and bisphenol F resins. Solid bisphenol A, multifunctional, and aliphatic diepoxy resins
have also been used. Ratios of liquid polysulfide polymer to epoxy are in the range of 1:2
to 2:1. The effect of various degrees of polysulfide on cure properties of a DGEBA epoxy
is shown in Table 7.7. An increase in elongation and impact strength is the result of
increased amounts of the liquid polysulfide polymer.
The addition of the polysulfide resin acts as both an elastomeric modifier and a diluent
for the epoxy resin. The low viscosity of LP-3, for example, can drastically reduce the vis-
cosity of the overall formulation. This provides greater ease of mixing and application and
the ability to be applied and sprayed without a solvent.
Suitable curatives for the polysulfide-epoxy reaction include liquid aliphatic amines,
liquid aliphatic amine adducts, solid amine adducts, liquid cycloaliphatic amines, liquid
amide-amines, liquid aromatic amines, polyamides, and tertiary amines. Primary and sec-
ondary amines are preferred for thermal stability and low-temperature performance. Not all
amines are completely compatible with polysulfide resins. The incompatible amines may
require a three-part adhesive system. The liquid polysulfides are generally added to the liquid
epoxy resin component because of possible compatibility problems. Optimum elevated-
temperature performance is obtained with either an elevated-temperature cure or a postcure.
Polysulfide resins combine with epoxy resins to provide adhesives and sealants with excel-
lent flexibility and chemical resistance. These adhesives bond well to many different substrates.
Tensile shear strength and elevated-temperature properties are low. However, resistance to peel
forces and low temperatures is very good. Epoxy polysulfides have good adhesive properties
down to –100°C, and they stay flexible to –65°C. The maximum service temperature is about
50 to 85°C depending on the epoxy concentration in the formulation. Temperature resistance
increases with the epoxy content of the system. Resistance to solvents, oil and grease, and exte-
rior weathering and aging is superior to that of most thermoplastic elastomers.
Liquid polymer:
HC(C2H4OCH2H4SS)xC2H4OCH2OC2H4SH
O OH OH
Formulation, pbw
Component A B C
7.6 EPOXY-VINYL
Epoxy resins may be blended with certain vinyl polymers to improve the impact strength and
peel strength of the adhesive. Polyvinyl acetals, such as polyvinyl butyral and polyvinyl for-
mal, and polyvinyl esters are compatible with DGEBA epoxy resins when added at concen-
trations of 10 to 20% by weight. The addition improves the resulting impact resistance and
peel strength of the cured adhesive. However, temperature and chemical resistance are sac-
rificed by the addition of the low-glass-transition-temperature vinyl resins.
The solubility of the vinyl polymer in the epoxy resins will depend on the molecular
weight and grade of vinyl used. Polyvinyl esters can be dissolved in liquid epoxy resins that
are heated to about 100°C or in melted solid epoxy resins. However, most epoxy-vinyl
blends are normally prepared via solvent solution.
The formulated adhesives are generally available as films or solvent solutions. They are
commonly used as laminating adhesives for film or metallic foil because of their high peel
strength. A composition consisting of a plasticized polyvinyl chloride copolymer and an
epoxy resin can be cured with an aliphatic polyamine, which will crosslink by reacting with
both resins. This adhesive possesses excellent adhesion to metals.
7.7 EPOXY-URETHANE
Epoxy-urethane adhesives provide properties when cured that are similar to those of epoxy-
nylon adhesives except they offer a major improvement in moisture resistance. Isocyanate
monomers and prepolymer react with the hydroxyl groups on epoxy resins to give tough,
132 CHAPTER SEVEN
flexible hybrids. These adhesives have been noted for their excellent adhesion to many high-
energy plastic substrates and for their propensity to bond to oily metal substrates.
The reactivities of isocyanates and epoxy resins are similar in that both can be cured
with amines or other hydrogen-containing compounds. There have been many attempts to
marry the rigidity, heat, and chemical resistance of the epoxies with the toughness and peel
strength of the polyurethanes. However, these attempts have often resulted in adhesive sys-
tems that display the worst properties of the parent resins, rather than the best properties.
This is due primarily to the significant different reaction rates between the −NCO groups
on the isocyanate and the epoxy groups on the epoxy molecule with common curing agents.
The common reaction mechanism is shown in Fig. 7.3.
The isocyanate reaction is several orders of magnitude faster than the epoxy reaction.
Hence, attempts to cure mixtures of epoxy and isocyanate resins must provide for essen-
tially complete consumption of all –NCO groups before any of the epoxy rings can react.
Another toughened epoxy formulation comes from mixtures of epoxies and urethane
oligomers with pendant epoxy groups.15 Curing of the pendant epoxy groups unites the ure-
thane and nonurethane components through conventional epoxy reactions to give tough,
durable films. Tensile elongation of the films can vary from 15 percent to more than 100 per-
cent compared to about 2 percent for most unmodified cured epoxies. This approach,
however, results in a decrease in high-temperature strength and chemical resistance although
500- to 1000-psi shear strength is achievable at temperatures in the 80 to 100°C range.
Straight polyurethane adhesives show much poorer strength at temperatures in this range.
These reactive hybrid epoxy-urethane adhesives were developed initially for bonding to
oily cold-rolled steel, but they have given good results on other substrates as well. It is
believed that the oil contaminant on the substrate is adsorbed into the uncured adhesive and
acts as a plasticizer. Table 7.9 shows tensile shear values obtained after curing 20 min at
177°C on various untreated substrates. The epoxy-urethane adhesives resist hydrolysis and
give useful bond strengths up to about 100°C.
Another approach for formulating a high-peel-strength epoxy-urethane adhesive has
been developed by Hawkins.17 Toluene diisocyanate is first reacted with a mixture of long-
chain polyether triols plus short-chain diols to obtain a viscous liquid resin terminated
with hydroxyl groups. This resin is then blended with a diepoxy resin, dicyandiamide cur-
ing agent, fillers, and diluents to give a one-component adhesive. When cured for 60 min
at 180°C, the dicyandiamide reacts with the epoxy and catalyzes the reaction between the
epoxy and the –OH groups on the modifying resin. T-peel strength on aluminum is claimed
to increase from 5 lb/in to over 50 lb/in via the addition of the urethane modifier. The increase
in peel strength is accompanied by only a minor decrease in heat distortion temperature.
O OH
CH CH2 + H X R CH CH2 X R
N C O + H X R N C X R
FIGURE 7.3 Common amine reaction mechanisms for epoxy and polyurethane resins.
HYBRID RESINS 133
TABLE 7.9 Cured Tensile Shear Strength of Uncleaned Substrates Bonded with
Epoxy-Urethane Hybrid System16
The number of possible hybrid systems that can be manufactured with epoxy resins is
nearly infinite, and many adhesive formulations have been attempted in a quest to improve
the main disadvantages of a cured epoxy: brittleness and rigidity.
Functionalized, liquid polybutadiene derivatives have also been developed as hybrid
flexiblizers for epoxy resins. Carboxyl-terminated butadiene/acrylonitrile polymers, buta-
diene homopolymers, and maleic anhydride–amino acid grafted butadiene homopolymers
have been used as flexibilizers to impart good low-temperature strength and water resis-
tance to DGEBA-based epoxy adhesives. An epoxy system toughened by polybutadiene
with maleic anhydride is claimed to provide a hydrophobic backbone, low viscosity, soft-
ness, and high tensile strength and adhesion (Table 7.10).
134 CHAPTER SEVEN
Epoxy-PVC plastisols are a type of PVC plastisol adhesive that is used in large quanti-
ties in the automobile industry. It is used to bond sheet steel to inner stiffener panels and to
seal around the crimped panel edges. These adhesives are formulated as high-solids,
thixotropic pastes and are applied as discrete dots or droplets to the stiffener surface or
panel edge before joining or crimping. These adhesives are called Hershey drops in the
trade because of the characteristic geometry of the droplets.
The epoxy-PVC plastisol type is a mixture of a plastisol-grade PVC powder, primary
PVC plasticizers (e.g., dioctyl phthalate), a liquid DGEBA epoxy resin, thickeners, stabi-
lizers, surfactants, and other additives. The epoxy serves as a secondary plasticizer, acts as
a stabilizer (acid scavenger), and helps to fortify the plastisol by crosslinking during cure.
These adhesives require relatively high cure temperatures (above 130°C). In the past
this had not been a problem, because the adhesive cured in the paint bake cycle. However,
with newer paint systems and efforts to reduce energy consumption, lower-cure-temperature
adhesives are needed, and epoxy-urethane adhesives and other flexibilized adhesives are
beginning to replace the plastisols.
The epoxy-PVC plastisol adhesives are soft and flexible after cure. They provide 50 to
100 percent elongation and tensile shear strength of about 400 to 700 psi on steel. Higher
bond strengths are generally not needed in the auto applications because of the large bond
area. The important properties in addition to low cost are long-term toughness and adhe-
sion as well as adhesion to oily steel.
Fluoroepoxies have gained interest because of the unique adhesion properties that can
be provided by the fluorine groups. There have been several attempts to marry the proper-
ties of epoxy resins with those of fluorocarbon resins. In general these have focused on
adhesive systems that (1) have a lower surface tension than unmodified epoxy or (2) have
significant hydrophobicity to resist exposures in moist environments.
A fluoroepoxy compound was made from a fluorodiepoxy resin cured with an amine
adduct. The use of fluoroepoxies is most advantageous in bonding fluorocarbons such as
polytetrafluoroethylene. Good bond strengths are achieved on fluorocarbons without the
need to surface-treat the substrate because the surface tension of the fluoroepoxy adhesive
is reduced from about 45 to 33 dyn/cm.25
Fluoroepoxy resins have also been developed that when cured with silicone amines or
fluoroanhydrides show substantial hydrophobicity. As the resin’s fluorocarbon content is
increased, so is the hydrophobicity. Adhesives made with these compounds showed much
HYBRID RESINS 135
These hybrids generally yield improved flexibility and toughness, but they are more
commonly used in epoxy coating formulations than in adhesive systems.
CH3 CH3
O O
CH3 CH3
REFERENCES
1. Lewis, A. F., and Saxon, R., “Epoxy Resin Adhesives,” Chapter 10 in Epoxy Resins, H. Kakuichi,
ed., Marcel Dekker, New York, 1969.
2. Black, J. M., and Bloomquist, R. F., “Development of Metal Bonding Adhesives with Improved
Heat Resistance,” Report FPL-710, Forest Products Laboratory, Madison, WI, May 1974.
3. Black, J. M., and Bloomquist, R. F., “Polymer Structure and the Thermal Deterioration of
Adhesives in Metal Joints,” Adhesives Age, vol. 5, nos. 2 and 3, 1962.
4. Burgman, H. A., “Selecting Structural Adhesive Materials,” Electrotechnology, June 1965.
5. Bulletin SC-54-57, Shell Chemical Company; also in Bolger, J., “Structural Adhesives State of the
Art,” Chapter 7 in Adhesives in Manufacturing, G. L. Schneberger, ed., Marcel Dekker, New
York, 1983, p. 155.
6. Gorton, B. S., “Interaction of Nylon Polymers with Epoxy Resins in Adhesive Blends,” Journal
of Applied Polymer Science, vol. 8, 1964, p. 1287.
7. Smith, M. B., and Sussman, S. E., “Development of Adhesives for Very Low Temperature
Applications,” Narmco Research and Development, NASA Contract NAS-8-1565, May 1963.
8. Frigstad, R. A., U. S. Patent to 3M Co., 3,449,280, 1970.
9. DeLollis, N. J., and Montoya, O., “Mode of Failure in Structural Adhesive Bonds,” Journal of
Applied Polymer Science, vol. 11, 1967, pp. 983–989.
10. DeLollis and Montoya, “Mode of Failure in Structural Adhesive Bonds.”
11. Frigstad, R. A., U.S. Patent No. 3,449,280.
12. Amstock, J. S., “Polysulfide and LP Polymers,” Chapter 11 in Handbook of Adhesives and
Sealants in Construction, McGraw-Hill, New York, 2001.
13. Panek, J. R., “Polysulfide Sealants and Adhesives,” in Handbook of Adhesives, 3d ed., I. Skeist,
ed., van Nostrand Reinhold, New York, 1992.
14. Peterson, E. A., “Polysulfides,” in Adhesives and Sealants, vol. 3, Engineered Materials
Handbook, AMS International, Materials Park, OH, 1990.
15. Guthrie, J. L., and Ching Lin, S., “One-Part Modified Epoxies for Unprimed Metal, Plastics,”
Adhesives Age, July 1985.
16. Guthrie, and Lin, “One Part Modified Epoxies for Unprimed Metal, Plastics,” pp. 23–35.
17. Hawkins, J. M., U.S. Patent No. 3,525,779, 1970.
18. Grieves, R., U.S. Patent 4,623,702, 1986.
19. Lay, D. G., and Millard, T. G., Paper presented at ASC Meeting, St. Louis, MO, April 14–17, 1991.
20. Durig, J., et al., “High Performance Flexible Epoxy System for Civil Engineering,” Paper presented
at the SPI-Epoxy Resin Formulators Meeting, San Francisco, February 20–22, 1991.
21. Air Products Patent, EP 703277-A2.
22. Shah, D., and Dawdy, T. H., Lord Corporation, U.S. Patent, no. 5,232,996, 1993.
23. Asahi Chemical Co. Patent, EP 703259-A2.
24. “Ricon Toughened Epoxy System,” Sartomer Application Bulletin, Sartomer Company, Exton,
PA, 2004.
25. Sheng, Yen Le, “The Use of Fluoroepoxy Compounds as Adhesives to Bond Fluoroplastics
Without Surface Treatment,” SAMPE Quarterly, vol. 19, no. 2, 1988.
26. Goodman, S. H., “Epoxy Resins,” in Handbook of Thermoset Plastics, 2d ed., S. H. Goodman,
ed., Noyes Publishing, Westwood, NJ, 2000, p. 207.
27. Rifle, J. S., et al., “Elastomeric Polysiloxane Modifiers for Epoxy Networks,” ACS Symposium
Series, vol. 221, 1983, pp. 21–54.
28. Strasila, D., U.S. Patent No. 5,151,318, 1992.
29. Isayama, K., et al., U.S. Patent No. 4,657,986, 1987.
30. Sadhir, R. K., and Luck, R. M., Expanding Monomers: Synthesis, Characterization, and Applications,
CRC Press, Boca Raton, FL, 1992.
CHAPTER 8
FLEXIBILIZERS
AND TOUGHENERS
8.1 INTRODUCTION
Polymerized epoxy adhesives are amorphous and highly crosslinked materials. This micro-
structure results in many useful properties such as high modulus and failure strength, low
creep, and good chemical and heat resistance. However, the structure of epoxy resins also
leads to one undesirable property—they are relatively brittle materials. As such, epoxy
adhesives tend to have poor resistance to crack initiation and growth, which results in poor
impact and peel properties. In sealant formulations, epoxy resins do not often provide the
degree of elongation or movement that is required for many applications.
Thus, two qualities are often required for adhesives or sealants that unmodified epoxy
resins lack—flexibility and toughness. Formulators have overcome these problems by
either flexibilizing the epoxy structure or incorporating modifiers into the adhesive. It must
be recognized, however, that flexibility and toughness have different meanings in adhesive
technology, and flexibilizers and tougheners operate by different mechanisms.
Flexibility is primarily characterized by a material’s elongation. Flexibilizers in epoxy
systems work by allowing the material to deform under stress. In this way stresses on the
joint are distributed over a larger area. The flexibilized epoxy resin has the capability of
compensating for differences in thermal expansion or elastic moduli of the substrates.
Flexibilizers improve peel and impact strength generally by allowing the adhesive to
deform under the application of stress. Figure 8.1 shows that an adhesive that has greater
elongation distributes peel stress over a much larger area than do more rigid adhesives. The
adhesive with higher elongation generally provides a higher peel strength as measured in
pounds per inch of bond width.
Flexibilized adhesives reduce mechanical damage by lowering the modulus and
increasing elongation via plasticization, and this allows deformation of the adhesive.2,3 This
increase in flexibility is mainly due to decreasing the crosslink density of the cured adhe-
sive. Decreasing the crosslink density, unfortunately, may also result in the lowering of
certain properties such as tensile strength and glass transition temperature.
Toughness, on the other hand, is characterized by a material having both high elonga-
tion and tensile strength (i.e., maximization of the area under the stress-strain curve).
Whereas flexibilizers work by deforming, tougheners improve properties because elas-
tomeric inclusions in the epoxy matrix absorb energy and stop a crack from propagating
throughout the bond line. Toughened adhesives tolerate damage by preventing crack
growth and, thus, limiting the damage area. This results in enhanced fracture, impact, and
stress resistance with minimal change in the crosslink density. Tougheners enable formulators
137
Copyright © 2006 by The McGraw-Hill Companies, Inc. Click here for terms of use.
138 CHAPTER EIGHT
FIGURE 8.1 Tough, flexible adhesive distributes peel stress over a larger area.1
and end users to have the best of two worlds—resistance to peel, impact, and fatigue and a
high degree of temperature and chemical resistance.
Epoxy adhesive formulators have generally addressed the problem of improving flexibility
in several ways:
1. By adding flexible chemical groups to the cured epoxy structure, via either the epoxy
base resin or the curing agent
2. By adding nonepoxy flexibilizing resins to the formulation to create an epoxy-hybrid
adhesive system
3. By adding flexible diluents to the base epoxy resin formulation
Flexibilizers have been previously discussed as resins (Chap. 4), curing agents (Chap. 5),
and hybrid modifiers (Chap. 7). This chapter primarily discusses flexibilizers that are used
as additives to rigid epoxy resins either during the adhesive formulation stage or during the
epoxy resin manufacturing process. They are generally reacted into the epoxy resin or have
reactivity with various curing agents so that the cured properties are not significantly
reduced due to the addition of the flexibilizer.
TABLE 8.1 Tradeoffs in Properties That Occur due to Flexibilizing the Adhesive by Reducing
Crosslink Density
Decrease in crosslink density will generally cause Increase in crosslink density will generally cause
Flexibility can be achieved within the cured epoxy molecule in a number of ways, most
notably by
The main reason for the greater flexibility is due to long-chain difunctional materials that
upon cure become part of the epoxy matrix. The result is a single-phase, flexible system. The
disadvantage of this approach is a reduction in the crosslink density and consequently reduc-
tions in glass transition temperatures as well as the heat and chemical resistance of the system.
Both the epoxy resin and the curing agent components can be made more flexible by
increasing their molecular weight. Polysulfides, polyamides, and n-aminoethylpiperazine
curing agents are examples of long-chain difunctional curing agents that offer improved
flexibility in epoxy adhesive formulations. There are many other examples of incorporating
flexibility into an epoxy adhesive by appropriate selection of the curing agent. For exam-
ple, by changing from hexahydrophthalic anhydride to hexamethylenediamine, one can dou-
ble the impact resistance of a resin system and slightly increase its tensile elongation at break.4
Gross changes in the flexibility of a resin system can also be obtained through a major
change in the structure of the cured resin. Such changes include shifting from an aromatic
structure to a more aliphatic hydrocarbon or reducing curing agent functionality. Insertion
of long hydrocarbon side chains also can provide additional flexibility to the epoxy mole-
cule. The effect on a typical epoxy formulation, using Araldite 508 flexible epoxy resin, is
shown in Table 8.2. Cured epoxies without the addition of flexible resins show elongation
less than 2 percent.
TABLE 8.2 Typical Flexible Epoxy Formulations and Their Effect on Properties5
Formulations
Cure schedule 7 days at 25°C 7 days at 25°C Gel at 25°C plus 2 h at 100°C
Tensile strength, 103 psi 3.34 2.23 2.51
Elongation at failure, percent 88 107 91
Modulus of elasticity, 106 psi 0.1 0.013 0.029
Hardness, Barcol 70 38 56
properties. Flexible hybrid resins (Chap. 7) improve adhesive properties, especially low-
temperature performance and adhesion to flexible substrates. They also relieve stresses
within the adhesive by allowing chains to relax after cure, as well as any stresses that occur
in the bonded joint during use.
A good example of a flexible epoxy hybrid adhesive is the epoxy polysulfides. Figure 8.2
shows the improved impact resistance that can be realized by using increasing amounts of
163 120
135 100
Impact resistance, ft·lbf
108 80
Impact resistance, J
81 60
54 40
27 20
0 0
0:1 0.5:1 1:1 1.5:1 2:1
Liquid polysulfide−epoxy ratio
FIGURE 8.2 Impact resistance of varying concentrations of polysulfide
resin in an epoxy adhesive formulation.6
FLEXIBILIZERS AND TOUGHENERS 141
polysulfide in epoxy blends. The elongation properties of the cured epoxy hybrid are sim-
ilarly increased.
Soluble thermoplastic polymer additions (e.g., epoxy-nylon and epoxy-vinyl) provide
very tough and somewhat flexible systems, but they are limited because of their high vis-
cosity and high raw materials costs. Almost all these systems need to be handled at elevated
temperatures or as solvent solutions. They require elevated temperature to achieve wetting
and cure.
Diluents are primarily used in epoxy adhesive formulations to reduce viscosity. However,
diluents also provide a degree of flexibility by two mechanisms:
1. By introducing long-chain, flexible molecules into the epoxy structure that remain unre-
acted after cure
2. By introducing long-chain, flexible molecules into the epoxy structure that react during
cure
The first category consists of unreactive diluents, and they may be considered mainly as
plasticizers. Generally they do not increase flexibility, and often they result in the overall
degradation of physical properties. The cured resin has reduced strength and hardness and
could even exhibit a “cheeselike” structure. The second category consists of reactive dilu-
ents, and they can be considered as true flexibilizers.
Of the two groups, the reactive diluents are generally preferred for structural adhesives
applications because they do not degrade end properties as much as the unreactive diluents
do. Note that diluents, either reactive or unreactive, do not necessarily result in an increase
in elongation of the cured epoxy. They may only reduce modulus and hardness. Their
enhancement regarding flexibility will depend on the component’s functionality and mol-
ecular nature.
Flexibilizers of these types do not convert a rigid epoxy structure to one that has the
elongation and hardness of synthetic elastomers. Rather, they provide a degree of resiliency
to a normally rigid system, thereby improving properties such as peel strength, impact
resistance, and thermal shock. These flexibilizers are also valued in adhesive systems
because they reduce internal stresses that could occur within the joint during cure or on
thermal cycling. For example, they provide a degree of resiliency that is better suited for
counteracting differing thermal expansion rates between the adhesive and the substrate, and
they also reduce the amount of stress caused by shrinkage during cure.
Nonreactive diluents for epoxy resins are discussed in Chap. 6. Of the various plasti-
cizers for epoxy that have been investigated, dibutyl phthalate is perhaps the most popular.
However, the use of unreactive diluents or plasticizers to develop highly flexible epoxy
adhesives has generally been unsatisfactory. Plasticizers do not provide sufficiently high
levels of flexibility to offset large reductions in tear resistance, chemical and heat resistance,
and adhesion properties. Few of the typical plasticizers are fully compatible, separating from
the resin during or after cure. Liquid hydrocarbon resins and benzyl alcohol have been used
with certain epoxy-amine systems, but at only low concentrations (5 to 15% by weight).
Reactive diluents, on the other hand, have one or two reactive epoxy groups that allow
the modifier to be compatible and nonmigrating. Thus, they are more effective than nonre-
active diluents for adhesive systems.
Monofunctional epoxy diluents are long-chain molecules that have one functional group
that can react with the epoxy system. The linkage prevents the diluent from migrating out of
the cured epoxy during aging. The long chains force the molecules in the cured structure to
142 CHAPTER EIGHT
be farther apart, and this permits the crosslinked molecule freer movement under stress.
However, it also results in decreased crosslink density and thus in degradation of thermal
and chemical resistance.
In adhesive formulations, glycidyl ethers are the most common monofunctional diluent.
These include butyl glycidyl ether and allyl glycidyl ether. Other effective monofunctional
O CH3 O O
O
CH CH2
CH3(CH2)3OCH2CH CH2 O
Butyl glycidyl ether (mol. wt.130)
Cyclohexene vinyl monoxide (mol. wt. 124),
b.p. 169°C at 760 mm
O
CH3
CH CH2
C O
O R1 O O
R3
H29C15 (C12−14H22−26O3) Glycidyl ester of tert-carboxylic
3 (Pentadecyl) phenol glycidyl ether acid (mol. wt. 240−250), b.p. 135°C at 760 mm
Epoxy resins from polyols Epoxidized polyether glycol Varies in molecular weight from n = 2
to 7. Higher-molecular-weight types
provide greater flexibility.
Epoxy resins from aliphatic Cyclic diglycidyl ester When used with primary amine curing
acids agents, reactivity is reduced.
Hydroxyl-terminated Hydroxyl-terminated Used with cycloaliphthalic epoxies
compounds polyesters and polyols that do not contain hydroxyl groups.
without ester linkages
diluents are epoxidized vegetable oils, olefin oxides, styrene oxide, and pinene oxide.
Figure 8.3 shows the chemical structure of several commercial epoxy-containing mono-
functional reactive diluents.
Di- or polyfunctional epoxy resins may be considered reactive diluents that provide a
degree of resiliency, yet most preserve the inherent properties of the base epoxy resin.
When added to an epoxy formulation, they will not reduce functionality, and in some cases
an actual increase in crosslink density is noticed. Typical resins of these types are shown in
Fig. 8.4. As flexibilizers, these diluents are often blended with the more reactive epoxy
resins. They, of course, can also be used as epoxy resins in themselves.
The commercial flexibilizing reactive diluents are predominantly glycidyl derivatives
of glycols, dimerized acids, or reaction products of dicarboxylic acids with epoxy resins.
The properties of certain polyepoxy diluents are summarized in Table 8.3. Some of, but not all,
these materials provide good reactivity with aliphatic primary amines at room temperature.
The above-mentioned methods used to achieve flexibility are widely recognized and
practiced in commercial epoxy formulations. The following discussion summarizes some
of the newer methods that have been developed.
Modern flexibilizers include isocyanate monomers and prepolymers. These react with
the hydroxyl groups on epoxy resins to give tough, flexible hybrids. Silicone polymer con-
taining amine termination will also react with epoxy groups to produce more flexible epox-
ies, while still maintaining good chemical and thermal stability. Acrylic polymers with
carboxyl termination have also been investigated as epoxy flexibilizers.
Silyl-terminated polyether can be blended with epoxy resins to form elastic adhesives
at room temperature.9 In this system, small rigid epoxy particles are chemically linked by
an elastic polyether phase. A typical formulation is shown in Table 8.4. Elongation and peel
O O CH3 CH
3
O O O O
CH2 CHCH2OCH2CH CH2 CH2 CHCH2O(CH2)4OCH2CH CH2
Diglycidyl ether (mol. wt. 130) Butanediol diglycidyl ether (mol. wt. 202)
O
O O CH CH2
CH2 CHCH2O(CH2)2O(CH2)2OCH2CH CH2 O S
Diethylene glycol diglycidyl ether (mol. wt. 218)
Vinylcyclohexene dioxide (mol. wt. 140)
O O
CH3
O C
O
CH CH2
Bis (2,3-epoxycyclopentyl) Ether (mol. wt. 182)
CH2 CH2
O
CH
COCH2
CH3C CH2
O S S O
O
CH3 CH3
Limonene dioxide (mol. wt.168)
3,4-Epoxy-6-methylcyclohexylmethyl 3,4-epoxy-
6-methylcyclohexane carboxylate (mol. wt. 280)
O O
CH CH2 OCH2CH CH2
O
O
CH2CH CH2
OCH2CH CH2
O
CH2 CHCH2 CH2CH CH2
OCH2CH CH2
O O
O
Diglycidyl ether of resorcinol (mol. wt. 222) 2,6-Diglycidyl phenyl glycidyl ether (mol. wt. 262)
40
1K-Epoxies
2K-Epoxies
Bond strength, MPa
30
2K-Methacrylates
20
Structural 2K-polyurethanes
strength (20 to 22 piw*) are significantly increased with a moderate decrease in tensile
properties (tensile shear strengths are 900 to 1500 psi). Tuning of the mechanical proper-
ties (i.e., strength from the epoxy and high impact resistance from the polyether) can be
achieved by adjusting the ratio and selection of the silylated polyether and the epoxy resin.
Maximum strength (both t-peel and shear strength) properties are achieved at a silylated
polyether/epoxy ratio of approximately 100/40.
Commercial silyl-epoxy hybrid adhesives have been developed by Collano AG. These
are polymer alloys consisting of a matrix of silyl reactive polymers (elastic phase), which
host domains of epoxy reactive polymers (hard domains). With a different choice of com-
ponents, mixing ratios, and the size of respective domains, the overall properties of the
bond line can be tailored for specific purposes.
Figure 8.5 depicts representative ranges of bond strength and elongations for typical
structural adhesives such as epoxies, methacrylate, polyurethanes, and silicones. The silyl-
epoxy hybrids conveniently fill the gap between the silicones and polyurethanes.
Flexibilizers based on functionalized, liquid polybutadiene derivatives have also been
developed for epoxy resins. Carboxy-terminated butadiene-acrylonitrile polymers, butadi-
ene homopolymers, and maleic anhydride–amino acid grafted butadiene homopolymers
have been used as flexibilizers to impart good low-temperature strength and water resis-
tance to DGEBA-based epoxy adhesives. An epoxy system toughened by polybutadiene
with maleic anhydride is claimed to provide a hydrophobic backbone, low viscosity, and
softness, along with high tensile strength and adhesion (Table 7.10).
Polyoxyalkylene ethers based on propylene oxide or propylene oxide/ethylene oxide
copolymers have also been described as useful flexibilizers. DGEBA-terminated ethylene
oxide/propylene oxide polyethers (30/70) have been used to modify DGEBA-based adhe-
sives by improving room temperature shear and peel strengths.
Diglycidyl ether of ethoxylated resorcinol (DGER), a resorcinol-based epoxy monomer,
can be cured with various epoxy curing agents to yield a balance of flexibility, low viscos-
ity, and fast cure capability.12 The molecular flexibility of DGER is claimed to improve the
impact strength and fracture toughness of the cured epoxy system. These flexibilized
*
piw = pounds per inch width.
146 CHAPTER EIGHT
systems are suggested for application in high-performance coatings, adhesives, and com-
posite matrix resins.
In recent years, much attention has been paid to siloxane oligomers as a specific type of
rubber modifier for epoxy resins because of their unique properties. These liquid functionally
terminated siloxanes belong to the elastomer class of materials. Organosiloxanes exhibit
important characteristics, such as very low glass transition temperature (−120°C), moisture
resistance, good electrical properties, low stress, high flexibility, good weathering ability,
and good thermal and oxidative stabilities. In addition, because of their low surface tension,
they provide excellent adhesion to many substrates and a hydrophobic surface.
It has been shown that by incorporating siloxane oligomers with reactive organofunctional
terminal groups, such as amine, epoxy, or carboxyl, into the structure of epoxy networks, one
can achieve improvement in the fracture toughness, water absorption, and surface properties
of the resultant polymer.13 Yayun14 and Menghuo15 studied the thermal and mechanical prop-
erties of methoxyl-terminated polymethylphenylsiloxane modified epoxy resins. The thermal
stability and the fracture toughness of the modified products were improved, and the glass
transition temperature and flexural modulus of the crosslinked epoxy network were not sig-
nificantly lowered.
Flexibilizers, such as those mentioned above that work by reducing crosslink density, pro-
vide a major dilemma for both the epoxy adhesive formulator and the end user. Although
flexibility is somewhat improved, other valuable properties are reduced. Reductions in ten-
sile strength and heat and chemical resistance are characteristic of many flexibilized epoxy
systems. Until the development of toughened epoxy formulations, it was difficult, if not
impossible, to provide an epoxy adhesive having high impact and peel strength and high
heat resistance.
Toughened structural adhesives generally have two distinct phases: The larger phase is
the base resin, and the minor phase consists of small (on the order of 1 µm in diameter), dis-
tributed elastomeric entities. The addition of the second phase significantly improves frac-
ture toughness by providing crack pinning and stress distribution mechanisms within the
material. The elastomeric, particulate phase is often thought of as a “crack stopper” for
improving fracture toughness.
A variety of toughening agents have been used to modify epoxy adhesives to improve
peel strength and fracture toughness without significantly affecting other properties of the
epoxy base resin. Generally these modifiers can be classified into three types:
Early structural epoxy adhesives were based on modification of epoxy resins by the use of
liquid butadiene-acrylonitrile polymers with terminal carboxyl groups. By using low lev-
els of these liquid polymers, the fracture energy of cured DGEBA-type epoxy resins was
increased by a factor of 15. Since that time, considerable work has been done on develop-
ing toughening agents for epoxy resins using liquid polymers with reactive terminal groups.
FLEXIBILIZERS AND TOUGHENERS 147
Formulations have been developed where small rubber domains of a definite size and
shape are formed in situ during cure of the epoxy matrix. The domains cease growing at
gelation. After cure is complete, the adhesive consists of an epoxy matrix with embedded
rubber particles. The formation of a fully dispersed phase depends on a delicate balance
between the miscibility of the elastomer, or its adduct with the resin, with the resin-hardener
mixture and appropriate precipitation during the crosslinking reaction.
The advantage of this toughening process is that the high modulus and Tg of the epoxy
resin are not sacrificed. The Tg of the product is indicative of the epoxy resin used as the
matrix, although the precipitated domain has a Tg of about −40 to −50°C.
The butadiene-acrylonitrile tougheners that have been especially successful in both
DGEBA and DGEBF types of epoxy resins are
CTBN and ATBN are the most commonly used in structural epoxy adhesive formulations.
CTBN (Fig. 8.6) is generally the elastomer of choice because of its miscibility in many
epoxy resins. These tougheners were originally developed by BF Goodrich (now Noveon,
Inc.) under the tradename Hycar.
The degree of toughness is determined by the crosslink density of the matrix, the elas-
tomer particle size and size distribution, the volume fraction of the elastomeric phase, and the
degree of adhesion between the epoxy matrix and the particle. The formulating procedure was
found to have as strong an effect on the fracture toughness as the materials themselves.16
In general, the elastomer must be prereacted (adducted) with the epoxy for the toughen-
ing effect to take place. Adducts reduce the likelihood of early phase separation and maintain
the solubility of the elastomer in the uncured resin system. For CTBN the reaction is carried
out at high temperatures (150 to 160°C) and usually in the presence of a catalyst, such as tris-
dimethylamino phenol or piperidine. The resulting epoxy-CTBN adducts are available from
several suppliers, and they can be easily formulated into epoxy adhesives.
Table 8.5 shows a comparison of adhesives formulated with and without a CTBN
adduct. When compared to the control epoxy, the toughened formulation exhibits signifi-
cantly higher peel strength and moderately higher tensile shear strength. CTBN modified
epoxy adhesives are generally one-part systems, cured with dicyandiamide at elevated tem-
perature. However, two-part, room or mildly elevated-temperature curing systems are also
possible and provide similar improvements in properties (Table 8.6).
One of the disadvantages of CTBN-epoxy adhesives has been their high viscosity, which
limits additional formulation options. Recently new adducts, such as EPON 58003 and RSM-
2577 from Resolution Performance Products LLC, have been introduced which have signif-
icantly lower viscosities.19 In addition, lower concentrations of these new CBTN-epoxy
adducts are generally required to achieve equivalent adhesive performance.
ATBN liquid toughening agents are synthesized by reacting a CTBN polymer with an
amine. For example, n-aminoethylpiperazine (AEP) will give a low-molecular-weight,
CN
FIGURE 8.6 Chemical structure of carboxyl-terminated butadiene acrylonitrile
(CTBN).
148 CHAPTER EIGHT
TABLE 8.6 Formulas for Two-Part CTBN Modified Epoxy Adhesive System Cured
Two Week at 25°C18
Part A
DEGEBA epoxy resin (EEW: 190) 60
Diethylene glycol 1
Aluminum (Toyal 101) 37
Fumed silica (Cab-O-Sil TS720) 2
Part B
Polyamide (Ancamide 2482) 38
ATBN (Hycar 1300x16) 12
Aluminum (Toyal 101) 22
Talc (Microtuff 325F) 27
Fumed silica (Cab-O-Sil TS720) 1
Properties
TABLE 8.8 Typical Two-Part CTBN and ATBN Toughened Epoxy Adhesives22
Part A
DGEBA epoxy resin, pbw 100 77.5 100 100
CaCO3 filler (Atomite whiting), pbw 30 30 30 30
DGEBA-CTBN adduct,* pbw 37.5
Part B
ATBN (Hycar 1300x16), pbw 30
Polyamine (Ancamine AD, Air 60 59.3
Products and Chemicals), pbw
Amidoamine (Ancamide 501, Air 35 32.8
Products and Chemicals), pbw
Properties
Cure schedule 2 weeks at 25°C
Tensile shear strength, MPa, on
aluminum measured at
• −40°C 6.9 22.7 8.6 15.5
• 25°C 8.6 17.9 9.6 17.2
• 75°C 4.4 5.5 13.8 11.7
T-peel strength, kN/m, on 0.5 5.0 0.5 5.3
aluminum measured at 25°C
*
Adduct contains 40% CTBN (Hycar 1300x13).
149
150 CHAPTER EIGHT
epoxy systems. Unlike early epoxy-hybrid adhesives (e.g., epoxy nylons, epoxy polysul-
fides), these modern attempts use engineering thermoplastics—tough, ductile, chemically
and thermally stable polymers having a high Tg. These are chosen so as to improve tough-
ness and not to negatively affect high-temperature performance. Due to the high modulus
and Tg of the thermoplastic modifier, the modulus and Tg of thermoplastic modified epoxy
resins will reach or even exceed the corresponding values of unmodified resin. In addition
to improved fracture resistance, this modification leads to reduced shrinkage on cure.
The thermoplastic resins are usually blended with the epoxy resin in a solvent solution.
Early researchers realized that to make this approach effective, it was necessary to increase
the compatibility and interfacial adhesion of the thermoplastic modifier and the epoxy
resins. The problem of poor miscibility of the thermoplastic resins and poor processability
of the final product are the main reasons that these materials have not achieved commercial
success.
Several research efforts have been directed at making a more miscible system. In an
effort to toughen an epoxy resin by incorporating engineering thermoplastic units into the
main chain, Kun-Soo Lee and coworkers developed epoxy resins containing ether ether
ketone units.23 These resins provided toughness without phase separation at elevated tem-
peratures. Low-molecular-weight polyphenylene ether (PPE) resins have also been devel-
oped that, unlike their high-molecular-weight counterparts, are readily miscible with epoxy
resins and form low-viscosity blends with them.24 The cured thermoset product exhibits
improved thermal properties as well as enhanced fracture toughness.
Compared to the carboxylated nitrile elastomer additives, the use of thermoplastics has
primarily been focused on the aerospace industry. On a cost per pound basis, the two-phase
nitrile additives offer the best combination of property improvement without negative
impact. The thermoplastic additives, however, may offer better high-temperature performance,
but they are more difficult to formulate and to process as adhesives. As a result, the cost of these
adhesives is generally much higher than that of other toughened epoxy mechanisms.
Some inorganic particulate fillers have also been considered as toughening agents for epoxy
materials. Glass beads, fly ash, alumina trihydrate, and silica were used early on to improve
the toughness of filled epoxy resins. Various studies, however, have demonstrated that the
fracture energy of filled epoxies reaches a maximum at a specific filler concentration.
Brittle, hollow glass beads have been found to be both an extender and an impact mod-
ifier for coatings and adhesives. The extender particles have to be designed in such a way
that they actively dissipate deformation energy and hence reduce the likelihood of loss of
adhesion between the epoxy and the substrate.25
In recent years, preformed particles such as thermoplastic powders or core-shell poly-
mers made from elastomeric latexes are increasingly being used as modifiers to improve
mechanical and thermal properties of polymer systems. However, these modifiers have yet
to gain widespread use in commercial epoxy adhesive systems. Figure 8.7 shows an ideal-
ized view of an epoxy matrix toughened with a core-shell polymer. There are several
advantages from using preformed particles as toughening agents:
Rubber (Core)
tire treads or whole tires. Following the grinding process, the rubber particles are exposed
to a reactive oxidizing gas, which creates polar groups on the particle surface. The result-
ing higher-surface-energy particles are easily wetted by polar polymers.26
Multiwalled carbon nanotubes were recently discovered as a toughener for epoxy adhe-
sives used to bond graphite fiber–epoxy composite substrates.27 Significant enhancements
of the bonding performance were observed as the weight fraction of the carbon nanotubes
was increased. Nanomaterials are materials whose size is on the order of angstroms.
Nanomaterial particles dispersed in epoxy resin hold the promise for producing toughened
adhesives with properties that are even more remarkable than those present in today’s adhe-
sive products.
Recently it has also been shown that modified spherical nanosilica particles can be used
to toughen epoxy resins without the loss of other properties, such as glass transition tem-
perature or modulus.28 When such surface modified nanoparticles are added to CTBN
toughened epoxy resins, the performance of both one- and two-component epoxy adhesives
was greatly improved.
A series of adhesives has also been modified by creating interpenetrating networks, allow-
ing for extensive control of adhesive properties without sacrificing thermal stability.29
These adhesives are based on resins containing epoxy groups. Additionally, they contain
monomers that can create a network of their own using a different chemical mechanism,
without reacting with the epoxy monomer. During the curing two independent networks are
created, penetrating each other but not being covalently bonded (Fig. 8.8).
Such a mixture of networks is called an interpenetrating polymer network (IPN). Much
higher shear strength and impact resistance are claimed for IPN epoxy systems than for stand-
alone single networks, and at the same time the high-temperature strength is improved.
Core-shell structures (Fig. 8.7) have also been suggested as a method of toughening
epoxy resin formulations. The polymeric shell, which surrounds an elastomeric core, pro-
vides good adhesion to the epoxy matrix. Hyperbranched polymers with hydroxyl
152 CHAPTER EIGHT
REFERENCES
1. Rider, D. K., “Which Adhesives for Bonded Metal Assembly,” Production Engineering, May 25,
1964.
2. Farrakhov, A. G., and Khozin, V. G., “Mechanical Behavior of Plasticized Epoxy Polymers,”
Crosslinked Epoxies, B. Sadlacek and J. Kahovec, eds., Walter de Gruyter and Co., Berlin, 1987,
pp. 585–597.
3. Misra, S. C., et al., “Effect of Crosslink Density Distribution on the Engineering Behavior of
Epoxies,” Epoxy Resin Chemistry, ACS Symposium Series 114, R. S. Bauer, ed., American
Chemical Society, 1979, pp. 137–156.
4. EPON Resin Structural Reference Manual—Additive Selection, Resolution Performance Products,
Houston, TX, 2001, p. 8.
5. Buckley, W. O., and Schroeder, K. J., “Adhesive Modifiers,” in Adhesives and Sealants, vol. 3,
Engineering Materials Handbook, ASM International, Materials Park, OH, 1990.
6. Buckley and Schroeder, “Adhesive Modifiers.”
7. Lee, H., and Neville, K., Handbook of Epoxy Resins, McGraw-Hill, New York, 1968, p. 13.10.
8. Lee and Neville, Handbook of Epoxy Resins, p. 13.16.
9. Devroey, D. R. E., and Homma, M., “Blends of Silyl-Terminated Polyethers and Epoxides as
Elastic Adhesives,” International Journal of Adhesion and Adhesives, vol. 21, 2001, pp. 275–280.
10. Devroey and Homma, “Blends of Silyl-Terminated Polyethers and Epoxides as Elastic
Adhesives.”
11. Ferrand, D., and Schwotzer, W., “Elastic Structural Adhesives by Hybrid Technology,” Adhesives
and Sealants Industry, April 2004.
12. Press Release, “INDSPEC Chemical Corporation Formulates New Epoxy Monomer,” INDSPEC
Chemical Corporation, Pittsburgh, PA, March 5, 2003.
13. Ho, T., et al., “Modification of Epoxy Resins with Polysiloxane TPU for Electronic
Encapsulation,” Journal of Applied Polymer Science, vol. 60, 1996, p. 1097.
14. Yayun, L, Master’s thesis, Syracuse University, NY, 1999.
15. Menghuo, C., Master’s thesis, Syracuse University, NY, 2001.
16. Klug, J. H., “Model High Performance Adhesive Systems,” Journal of Applied Polymer Science,
vol. 66, 1997, pp. 1953–1963.
17. Buckley and Schroeder, “Adhesive Modifiers,” p. 185.
FLEXIBILIZERS AND TOUGHENERS 153
18. Technical Bulletin, “Hycar CTBN Modified Epoxy Adhesives,” B. F. Goodrich Company, AB-8,
Cleveland, OH, 1990. (Hycar products now manufactured by Noveon, Inc.)
19. Farris, R. D., and Steward, S. L., “New Epoxy Tougheners Widen the Adhesive Formulation
Window,” Adhesives and Sealants Industry, January 2002.
20. Petrie, E. M., “Improving the Toughnes of Structural Adhesives,” at www.SpecialChem4Adhesives.
com, April 7, 2004.
21. “Two Component Adhesive Starting Formulation,” General Purpose Adhesive, Air Products and
Chemicals Company, Allentown, PA, 2001.
22. “Two Component Adhesive Starting Formulation,” General Purpose Adhesive, Air Products and
Chemicals Company.
23. Lee, Kun-Soo, et al., “Preparation of Epoxy Resins Containing Ether Ether Ketone Unit and Their
Thermal Properties,” Bulletin of Korean Chemical Society, vol. 22, no. 4, 2001.
24. Yeager, G. W., et al., “Use of Low Molecular Poly(phenylene ether) Resin in Epoxy Thermosets,”
International SAMPE Symposium and Exhibit (Proceedings), vol. 44, pt. 1, 1999, pp. 1075–1089.
25. Rosler, M., and Kinke, E., “Hollow Glass Beads as Impact Absorbing Extenders,” Farbe and
Lack, vol. 103, no. 1, 1997, pp. 50–54.
26. Gerace, M. J., and Gerace, J. M., “Surface Activated Rubber Particles Improve Structural
Adhesives,” Adhesives Age, December 1995, pp. 26–31.
27. Hsiao, Kuang-Ting, et al., “Use of Epoxy/Multiwalled Carbon Nanotubes as Adhesives to Join
Graphite Fiber Reinforced Polymer Composites,” Nanotechnology, vol. 24, July 2003, pp. 791–793.
28. Bowtell, M., “Novel Adhesive Solutions,” Review of 17th International Swissbonding
Symposium, Zurich, European Adhesives & Sealants, December 2003.
29. Technical Bulletin, “IPN Technology,” Polymerics GmBH, Berlin, Germany, 2004.
30. Boogh, L., et al., “Novel Tougheners for Epoxy Based Composites,” SAMPE Conference
Proceedings, vol. 28, 1996.
31. Hillmyer, M., et al., “Nanostructured Thermosets,” Journal of American Chemical Society, vol. 119,
1997, p. 2749.
This page intentionally left blank
CHAPTER 9
FILLERS AND EXTENDERS
9.1 INTRODUCTION
Fillers and extenders are used in epoxy adhesive formulations to improve properties and to
lower cost. Properties that can be selectively improved include both the processing proper-
ties of the adhesive as well as its performance properties in a cured joint. However, the use
of fillers can also impair certain properties. Typically, the formulator has to balance the
improvements against property decline. The advantages and disadvantages of filler addi-
tion in epoxy formulations are listed in Table 9.1. Common fillers used in epoxy formula-
tions and the properties that they are used to modify are shown in Table 9.2.
Fillers generally represent one of the major components by weight in an adhesive for-
mulation. However, their concentration is quite often limited by viscosity constraints, cost,
and negative effects on certain properties. The degree of improvement provided by a filler
in an epoxy formulation will heavily depend on the type of filler and its properties (parti-
cle size, shape, size distribution, and concentration), surface chemistry, dispersion charac-
teristics, dryness, and compatibility with the other components in the formulation. Table 9.3
summarizes the properties of selected fillers.
A significant number of fillers can be used to modify the properties of an adhesive. They
can be either organic or inorganic. The most common types of fillers used in epoxy adhe-
sive formulation are inorganic solids. They are commercially available in particle sizes
greater than 0.1 µm and in various grades. Naturally occurring organic materials such as
cellulose, cotton fiber, etc., are also sometimes used as fillers and extenders in epoxy for-
mulations. Synthetic resins such as thermoplastic powders may be used as fillers; however,
they are more often considered as a resinous modifier of toughening agent.
Selection of the proper filler is based on a number of factors. The most important, of course,
is the improvements in processing and end properties that it will provide. However, other
important considerations when one is selecting a filler include availability, cost, surface
chemistry, water adsorption, oil adsorption, density, and other physical or chemical prop-
erties of the filler grade itself.
A filler should be dry, nonreactive with the uncured resin, and of a neutral or only
slightly basic pH. Adsorbed water, which is present in some degree in most fillers, inhibits
dispersion. Thus, most fillers must be dried before they are added to the adhesive formula-
tion. The drying process will drive off adsorbed moisture and gases from the surface of the
filler. The filler should generally be nonreactive with the base resins or curing agents that
155
Copyright © 2006 by The McGraw-Hill Companies, Inc. Click here for terms of use.
156 CHAPTER NINE
TABLE 9.1 Advantages and Disadvantages of Fillers in Epoxy Formulations (Depending on Type
of Fillers Employed)1
Advantage or disadvantage
Advantages Disadvantages (depending on application)
Filler Function
Aluminum Machineability
Alumina Abrasion resistance, electrical
Aluminum silicate Extender
Aluminum trioxide Flame-retardant
Barium sulfate Extender
Calcium carbonate Extender
Calcium sulfate Extender
Carbon black Pigment, reinforcement
Copper Machineability, electrical conductivity
Glass fiber Reinforcement
Graphite Lubricity
Iron Abrasion resistance
Kaolin clay Extender
Lead Radiation-shielding
Mica Electrical resistance
Phenolic or glass microspheres Decreased density
Silica sand Abrasion resistance, electrical properties
Silicon carbide Abrasion resistance
Silver Electrical conductivity
Titanium dioxide Pigment
Zinc Adhesion, corrosion resistance
Zirconium silicate Arc resistance
TABLE 9.3 Selected Properties of Fillers Used in Epoxy Adhesive Formulations3
the filler. However, these effects need to be considered, especially when the curing process
is critical. The effects of fillers on working life and exotherm are described in Sec. 9.3.12.
The filler concentration in any epoxy adhesive formulation will depend primarily on the
following three factors:
1. Handling and viscosity characteristics of the formulated adhesive
2. Wetting characteristics of the filler with regard to other components in the formulation
3. Ultimate properties desired
The particle size, density, and oil absorptivity will determine the maximum weight load-
ing that is achievable in a specific formulation. The high-oil-absorption lightweight fillers,
such as diatomaceous silicas and chopped glass, can greatly increase the viscosity at lower
filler loadings, even at concentrations of several parts per hundred. Medium-weight granu-
lar filler, such as talc, powdered aluminum, and alumina, can be used at loadings up to 200
pph. The heavier, nonporous fillers, such as aluminum oxide, silica, and the calcium car-
bonates, can be used at levels as high as 700 to 800 pph without causing the viscosity to be
unworkable. Fillers with fine particle size will tend to settle out less than those with larger
filler sizes.
Particle interactions resulting in aggregates of particles will adversely affect dispersion.
Special surface treatments are provided to reduce these aggregation forces and achieve
higher loadings and better suspension stability with less effect on viscosity. These surface
treatments can be applied directly to the filler, and many grades of treated fillers are com-
mercially available.
The filler surface treatment can also be applied as an additive to the adhesive formula-
tion. Coupling agents, such as organotitanates, zircoaluminates, or organosilanes, are
added to the formulation so that they preferentially find their way to the filler’s surface and
provide for optimal surface chemistry and adhesion between the filler and the matrix resin.
These coupling agents are discussed in Chap. 10.
Particle size and shape will affect the degree of mixing required. Particles with large
aspect ratios, such as fibrous fillers, and particles of large size are typically more difficult
to disperse. In addition, high filler loading makes sufficient wetting of the filler more dif-
ficult because of the increased viscosities encountered.
In certain circumstances, blends of different-size fillers are utilized. With heavier or
larger fillers, there is a greater tendency of the filler to settle, unless lightweight secondary
fillers are employed as antisettling agents, or unless the filler is thoroughly milled into the
system. A broad size distribution of a singular type of filler is an effective way of achiev-
ing antisettling and good dispersion properties.
To obtain the best results with fillers, they should be heated before being incorporated
into the epoxy formulation. This will drive off any moisture or adsorbed gases from the sur-
face of the fillers. Once thoroughly dry, the fillers may be stirred directly into the resin.
High-shear mixing is generally required for compounding of most formulations. The
mixing vessel must sometimes be equipped with a vacuum to eliminate the possibility of
air being beaten into the adhesive formulation. Often the formulation is warmed to a mod-
erately elevated temperature to facilitate mixing. For very viscous formulations, roll-
milling equipment may be required to achieve efficient mixing and dispersion of the filler.
In some cases it may be necessary to employ a grinding device to accomplish the mixing
with sufficient thoroughness. Unless complete and efficient mixing is accomplished, the
final formulation will not possess the desired properties.
Solvent addition or blending the epoxy resin with low-molecular-weight diluents is
another method of lowering the viscosity so that fillers can be efficiently added to the epoxy
adhesive formulation. However, in these cases the formulator must address the high vapor
pressures of the solvent or diluent (as well as various health, safety, and environmental
issues). In the case of diluent addition, the reduction in crosslinking density and thermal or
FILLERS AND EXTENDERS 159
100,000
80,000
60,000
40,000
20,000
Viscosity 25°C, cP
10,000
8000
6000
4000
Resin
2000
1000
0 5 10 15 20 25
Phenyl glycidyl ether, phr
FIGURE 9.1 Effect of reactive diluent concentration on the viscosity of a filled epoxy resin.4
chemical properties must be considered. Figure 9.1 illustrates the viscosity of a filled and
unfilled resin as a function of percentage of the diluent phenyl glycidyl ether.
In two-part epoxy adhesive systems, the filler can generally be incorporated into either the
epoxy resin or the curing agent component. Prevailing factors will be the characteristics men-
tioned above (viscosity of each component, dispersion characteristics, etc.). However, the
effect on the mix ratio and on the ease of mix-
ing the two components prior to application TABLE 9.4 Forms of Fillers and Reinforcing5
must also be considered. Fillers are often Materials Commonly Used with Epoxy Resins
used to provide practical mix ratios for two-
component adhesives. 1. Powders 9. Continuous strand
Fillers are available in many forms 2. Spheres 10. Yarn
including particles, fibers, and mats or fab- 3. Granules 11. Spun roving
4. Fibers 12. Fabric
rics. Table 9.4 lists common forms of fillers
5. Shreds 13. Woven roving
and reinforcements that are used with 6. Whiskers 14. Chopped strands
epoxy adhesives. Note that epoxy adhesive 7. Needles 15. Reinforced mats
film carriers such as fabrics or mats can be 8. Flakes
considered as a type of filler.
160 CHAPTER NINE
This section considers only the more common types of fillers for epoxy adhesive formula-
tions. They are categorized by the specific contributions that they provide to adhesive prop-
erties. Modifications to the following properties are discussed:
• Materials cost
• Flow properties
• Bond line thickness
• Coefficient of thermal expansion
• Shrinkage
• Conductivity
• Electrical properties
• Specific gravity
• Mechanical strength
• Heat and chemical resistance
• Adhesion
• Pot life and exotherm
• Fire resistance
• Color
Fillers are often used for the sole purpose of reducing the raw materials cost of the adhe-
sive system. They do this by replacing relatively expensive synthetic organic components
with inexpensive inorganic components, generally naturally occurring minerals. Fillers that
are used for this purpose are often referred to as extenders.
Extenders may also provide one or several secondary functions in addition to cost
reduction. Depending on the type and concentration of the extender, it could reduce shrink-
age that occurs on polymerization or during aging, lower the coefficient of thermal expan-
sion of the adhesive formulation, provide a higher viscosity to avoid adhesive starvation in
the joint, or perform a number of other property improvements.
Extenders have two very broad classifications: inorganic and organic. Inorganic exten-
ders are primarily naturally occurring minerals, and these are often used in epoxy formula-
tions. Organic extenders are naturally occurring resins or fibrous particles. In either case,
the most distinguishing features of these extenders are their low cost and availability.
Inorganic Extenders. There are many mineral extenders that are commonly used in
epoxy adhesive formulations. The more important ones are summarized below.
Calcium carbonate is the most commonly used extender. It is widely available and
low in cost, and it provides for improvements in certain performance properties. The
material is a mineral that is mined throughout the world. Common forms of calcium car-
bonate include limestone, marble, calcite, chalk, and dolomite. It is manufactured by pre-
cipitation processes and is commercially available from a number of sources. Calcium
carbonate is available in many different particle sizes and in various grades. To improve
dispersion in certain resins, the filler is often coated with calcium stearate or stearic acid.
Silica is also often used as an extender in adhesive formulations. Similar to calcium car-
bonate, silica is an abundant mineral found in crystalline form (quartz) and amorphous
FILLERS AND EXTENDERS 161
form (diatomaceous silica). Diatomaceous silica is used more extensively than quartz
because it is a softer material providing fewer machining and abrasive problems. There is
also concern over respiratory problems possibly associated with inhalation of finely
divided quartz. Although it is an excellent additive for increasing viscosity, diatomaceous
silica has a low oil adsorption and very large surface area so that it is not a particularly good
extender, since it cannot be easily incorporated into the formulation in high concentrations.
Kaolin, commonly called clay, is another naturally occurring mineral that is often used
as an extender in adhesive formulations. It is a hydrated aluminum silicate with a hexago-
nal platelike crystal structure. In addition to reducing cost, kaolin clay provides shrinkage
control, dimensional stability, viscosity increase, and pigmentation. Certain clays, such as
bentonite, are highly alkaline and can accelerate or even catalyze the epoxy curing reac-
tions. Therefore, their reactivity must be taken into consideration when one is calculating
the stoichiometric mix ratio of resin and curing agent.
Talc is a hydrated magnesium silicate that is composed of thin platelets primarily white
in color. Talc is useful for lowering the cost of the formulation with minimal effect on phys-
ical properties. Because of its platy structure and aspect ratio, these extenders are consid-
ered reinforcement. Polymers filled with platy talc exhibit higher stiffness, tensile strength,
and creep resistance, at ambient as well as elevated temperatures, than do polymers filled
with particulate fillers. Talc is inert to most chemical reagents and acids. The actual chem-
ical composition for commercial talc varies and is highly dependent on the location of its
mining site.
Organic Extenders. Organic extenders are primarily of two types: (1) fillers derived
from organic materials and (2) low-cost, naturally occurring or synthetic resins. Of the first
type, wood flour, shell flour, and other cellulosic fillers are the most common. They also
provide a margin of mechanical property reinforcement because of their relatively high
aspect ratio. Of the resinous types these are petroleum-based derivatives as well as soluble
lignin and scrap synthetic resins.
Coal tar pitch is the most widely used resinous extender for epoxy resins. It is primar-
ily used in surface coating formulations, but can also be used as a cost reducer and flexibi-
lizer in epoxy adhesives and sealants. In addition to the increase in flexibility (and
reduction in thermal and chemical resistance), coal tar pitch extenders provide excellent
water resistance. Their primary applications, therefore, are often in the marine, pipe, tank,
and general industrial maintenance areas.
Petroleum-derived bitumens are also used as extenders in epoxy formulations. High-
boiling-point petroleum distillates can serve as low-cost extenders, but a compatibilizer
such as an alkylphenol must be present in the mix to achieve compatibility between the
resin and the extender. Furfural resins can also be used with epoxies, to reduce cost and to
obtain increased resistance to acids.
Controlling flow is an important part of the adhesive formulation process. Control of flow
is important for several reasons.
1. It allows easy and reproducible metering and mixing of two-component systems prior
to application.
2. It provides for certain application characteristics of the adhesive (brush, spray, trowel,
penetration, etc.).
3. It can provide sag resistance (thixotropy) for adhesives that are applied to vertical surfaces.
4. It can provide for a practical and reproducible bond line thickness in the final joint.
162 CHAPTER NINE
The first three factors are generally controlled by the rheological properties of the liquid
adhesive through the application of fillers in the formulation. The final factor can be con-
trolled through the viscosity; however, other methods are also possible to control the bond
line thickness.
Control of Viscosity. To ensure that adhesives and sealants function well during their appli-
cation and end use, the formulator must be able to control the flow properties of the product.
The challenge that faces the formulator is that the adhesive or sealant may need different flow
characteristics at different times. For example, adhesives must flow readily so that they can be
evenly applied to a substrate and wet out the surface. Yet, there should not be an excess of pen-
etration into porous substrates, nor should the adhesive “run” or “bleed” to create a starved
joint. Certain adhesives and sealants must be capable of convenient flow application by trowel
or extrusion, but they must also exhibit sag and slump resistance, once applied. Therefore, the
flow properties, or rheology, of the material must fit the desired method of application.
Adhesive and sealant manufacturers employ rheological additives for thickening and to
prevent sag of their products. In practice, rheological additives may provide benefits in addi-
tion to viscosity or flow control. When properly formulated into adhesives and sealants, rheo-
logical additives can
Control of Thixotropy. Often the adhesive application will require that the product be fluid
for mixing and application, but it must not flow or sag once applied. For example, ASTM
C920 defines a nonsag sealant as “one that permits application in joints on vertical surfaces
without sagging or slumping.” This property is called thixotropy. Thixotropic materials, such
as tomato catsup, toothpaste, etc., undergo a decrease in viscosity when subject to shearing.
Although most fillers provide epoxy systems with viscosities that are unaffected by the
shear rate, certain fillers can provide thixotropy that results in an adhesive that will not flow
under low levels of stress (e.g., under its own weight when applied to vertical surfaces). Yet
the product will exhibit lower viscosities when under higher levels of stress, such as when
being dispensed or applied to a substrate.
FILLERS AND EXTENDERS 163
25,000
20,000
0
ina T-6
der
Magnesium silicate
r alum
pow
15,000
Viscosity, cP
num
Tabula
umi
n 347
l
44 a
rcon
Kaoli
MD
c zi
e
10,000
at
ami
on
te
rb
Cer
ca
ca
sili
um
m
lci
nu
Ca
mi
Alu
5,000
1,000
0 10 20 30 40 50 60 70
Percent filler by weight
FIGURE 9.2 Viscosity of DGEBA epoxy resin diluted with dibutyl
phthalate as a function of filler loading volume.7
164 CHAPTER NINE
Up curve,
increasing speed
Shear stress
Down curve,
decreasing speed
Shear rate
FIGURE 9.3 Hysteresis due to thixotropy.
Thixotropy provides a shear thinning effect; that is, viscosity decreases as the shear rate
increases, and vice versa. This not only allows easy pumping, dispensing, and mixing of the
adhesive, but also provides sag resistance once the adhesive is applied. The thixotropic fillers
work by forming a temporary “structure” in the mixture, which can be broken down at high
rates of shear. This structure is generally the result of van der Waals forces between molecules.
The thixotropic effect is shown in Fig. 9.3. The viscosity curves are for a material that
is exposed to first increasing and then decreasing shear rates. Thixotropy results from the
ability of the dispersed particles to come together and form network structures when at rest
or under low shearing forces. Viscosity decrease occurs when this structure breaks down
due to shearing stress and the resistance to flow decreases. (See Fig. 9.4.)
Thixotropy can be obtained at fairly low loading concentrations with colloidal silica,
bentonite, metallic leafing powders, and hydrated magnesium aluminum silicates. If
required, thixotropic adhesive pastes may be formulated which will not flow during cure
even at elevated temperatures and which are useful for bonding loose-fitting joints.
The addition of asbestos fibers at one time provided excellent thixotropic adhesive for-
mulations, especially at elevated temperatures. However, health and environmental regula-
tions have severely limited the use of this material. Today, fumed silica, precipitated
calcium carbonate, certain clays, and cellulose and other fibers offer thixotropic properties
at relatively low levels of loading.
Fibers. At one time asbestos fibers were used as the primary thixotrope in the
adhesive and sealant industry. Asbestos fiber filled materials are very resistant to sag
under high temperatures—an important property for many adhesives and sealants.
However, asbestos abruptly fell from the formulator’s arsenal of additives primarily due to
Shearing
Standing
health and environmental factors. Asbestos fillers also stiffen and harden the resulting
product; so it is not the ideal additive when adhesives and sealants need to remain flexible.
Although no single material is a direct replacement for asbestos, the need was eventually
filled by a number of materials.
Several types of polymeric fibers are used as thixotropes in both adhesive and sealant
formulations. These are cellulose, polyolefin, and aramid fibers. A fiber that exhibits good
thixotropy usually has several key characteristics including
• The ability to interact with the polymer matrix via dipolar and/or dispersion interactions
• The ability to achieve polymeric entanglement
• The presence of branchlike fibrils on the fiber which increase the surface area and pro-
mote rheological effects
• A high surface area with surface activity
It is interesting to note that these characteristics also are necessary to impart physical rein-
forcement and resulting cohesive strength improvement to the adhesive or sealant.
Cellulose fibers (e.g., Interfibe RT Cellulose Fiber from Interfibe) are perhaps the best
accepted thixotrope of this group. This material provides good thixotropy, reinforcement
of the bulk polymer, minimal surface texture, and low cost. Cellulose fibers are claimed to
provide greater formulation latitude than the other thixotropes at reduced cost.
Studies have been conducted on adhesive and sealant systems using cellulose fibers as
a thixotrope in epoxy adhesives. Table 9.6 shows thixotropic epoxy adhesive formulations
and resulting properties using fumed silica and reinforcing thixotropic (RT) cellulose fiber
additives. Note that a certain percentage of fumed silica is still required in the cellulose
fiber filled formulations for optimal nonsag properties. The cellulose fibers’ low specific
gravity (1.1) relative to other filler materials also contributes to weight savings and further
improves the sag resistance of these systems.
Polyolefin (Spectra from Honeywell-Allied Signal) and aramid (Kevlar from DuPont)
fibers have much lower surface areas and polarities than asbestos fibers and do not bond
Formulation A (control) B
Viscosity, P
Initial 86 91
Aged 105 120
Slump resistance (1/2 in × 1/2 in × 4 in) No movement No movement
Tensile shear strength, psi
Initial 1770 1801
250-h humidity 1579 1727
250-h salt spray 1288 1592
166 CHAPTER NINE
well with all polymeric matrix resins. However, aramid fibers have been proposed as a
replacement for asbestos or fumed silica.9 They are claimed to provide better sag resistance
than fumed silica. Disadvantages are that aramid fibers are very expensive, and they
increase the modulus as much as 2.5 times that of fumed silica.
Fumed Silica. Today colloidal silica (fumed silica) is the most common thixotropic
agent in epoxy resins. Fumed silica, an amorphous silicon dioxide, is a versatile, efficient
additive used for flow control and thixotropy. Fumed silica has long been the dominant
thixotrope employed in the adhesive and sealant industry.
Fumed silica is generally incorporated at concentrations of less than 10 pph. Epoxy sys-
tems based on lower-viscosity resins generally tend to hold thixotropic action better at ele-
vated temperature than systems based on the higher-viscosity resins.
Fumed silica is typically available with sizes of 7 to 40 nm and surface areas ranging
from 50 to 380 m2/g. Unlike precipitated silica, fumed silica has no internal surface area.
The specific gravity of fumed silica is approximately 2.2. Because of its high surface area
to weight ratio, formulations generally require only a minor amount of fumed silica (1 to
5 percent by weight) to achieve thixotropic properties.
The surface chemistry of fumed silica is extremely important because of its influence
on the rheological behavior of the formulation. Three types of chemical groups can be
formed on the surface of the particle depending on the processing procedures:
1. Isolated hydroxyl
2. Hydrogen bonded hydroxyl
3. Siloxane
The isolated and the hydrogen bonded hydroxyl groups are hydrophilic sites, whereas
the siloxane is a hydrophobic site. Thus, fumed silica grades are generally characterized by
their surface area and whether they are hydrophilic (standard grade) or hydrophobic. The
hydrophilic silica is most effective in nonpolar and medium polar media. The performance
of hydrophilic fumed silica is often improved by adding a polar substance such as ethylene
glycol, glycerin, or some secondary amines to the formulation. In medium polar to polar
media, hydrophobic fumed silica is a more efficient thickening agent and generally pre-
ferred. Hydrophobic fumed silica is also noted for providing superior moisture resistance
in adhesives, sealants, and coatings.10
The thixotropic characteristics provided by fumed silica are due to its ability to develop
a loosely woven, latticelike network by hydrogen bonding between particles. This network
raises the apparent viscosity of the system, increases the cohesive forces, and contributes
to the suspension of the solid. Because the hydrogen bonds themselves are relatively weak,
they are easily disrupted through the action of an applied stress or shearing force and
quickly reform when the stress or shearing force is removed.
Parameters that are important to the performance of fumed silica systems include
At times the results of the formulation are less than expected because these factors are not
considered or understood relative to the final rheological properties. For example, proper
dispersion can maximize the efficiency of both hydrophilic and hydrophobic fumed silica.
To ensure proper dispersion, the addition of silica to the formulation in the right sequence
and the effectiveness of the dispersion equipment become very important. High-shear mixing
FILLERS AND EXTENDERS 167
equipment will improve the efficiency of fumed silica in the formation. In most cases, sil-
ica should be added to the resin directly or to the more viscous part of the formulation with
as little solvent or diluent as possible. Incorporating the silica before adding any fillers or
pigments ensures homogeneous distribution. Dispersing the silica in a concentrated base,
or master batching, provides optimum efficiency and stability.
Fumed silica is often used in 100 percent solids, liquid polymers. With epoxy adhesives
and sealants only a few percent by weight of the additive will eliminate common problems
such as slumping and separation. The fumed silica also raises the effective viscosity of the
base resin to prevent other components from settling while the extrudability or spreadabil-
ity is unaffected. Also note that fumed silica provides a surface that is free of texture. This
is important in architectural-grade paints and sealants.
Comparative sag resistance properties of various commercial types of fumed silica are
shown in Table 9.7. In most liquid epoxy resin systems, fumed silica is only employed at a
1 to 3 wt percent basis to provide thixotropic characteristics.
Precipitated Calcium Carbonate and Other Minerals. Precipitated calcium carbonate
(CaCO3) functions as a thixotrope in sealant and adhesive formulations as well as being a
low-cost extender (often used in the 40 to 50 percent by weight range) and reinforcement.
Ultrafine (<100 nm) precipitated calcium carbonate provides the greatest efficiency. These
have surface areas from 15 to 30 m2/g. CaCO3 is hydrophilic, and as a result it has a ten-
dency to agglomerate in organic polymers and plasticizers.
Precipitated calcium carbonates are generally surface-treated to render them hydropho-
bic and to improve their dispensability in hydrophobic systems. Conventional surface treat-
ment of CaCO3 is with fatty acids. However, the use of different types of acids yields subtle
differences in filler wettability and polymer compatibility.
Other mineral additives have been used for many years as functional extenders and
fillers in adhesives and sealants. These include kaolin (hydrated aluminum silicate), ben-
tonite (hectorite clay), talc (magnesium silicate), and attapulgite (hydrated magnesium alu-
minum silicate) additives.12,13 Kaolin and talc are considered to be viscosity thickeners
whereas attapulgite is more of a conventional thixotrope. They are considered to be very
cost-effective rheological additives.
Kaolin is a common inexpensive filler used primarily as an extender in adhesive and
sealant formulations. The cost of the adhesive is reduced because the kaolin addition
increases the product’s volume. Depending on the grade, kaolin can also
The finer particle sizes build viscosity more efficiently and provide better reinforcement
than do their coarser counterparts. Fine grades of kaolin have particle size of 0.5 µm or less.
Kaolin has a hexagonal particle with the finer particles occurring as platelets and coarser
particles forming stacks or books of platelets. It is relatively soft and is a nonabrasive filler.
As with other solid rheological additives, the key parameters in choosing a particular grade
are particle size, shape, and particle distribution. Kaolin is also available in hydrous and
calcinated (where dehydroxylation is used to remove water) grades. The calcinated grades
have a more porous particle shape.
Special surface modifications are available to further improve reinforcement. The
objective of the surface treatment is to increase filler loading and/or improve physical prop-
erties without loss of rheological characteristics. A variety of surface-modified kaolins
have been introduced including clays treated with silane, titanate, polyester, and metal
hydroxide. Silane-treated kaolin is used in applications requiring maximum aging charac-
teristics in the service environment.
Bentonite is a colloidal clay that is both hydrophilic and organophilic. It is water-
swelling with some types of clay absorbing as much as 5 times their own weight in water.
It is used in emulsions, adhesives, and sealants. It is a gritty, abrasive white particle filler.
A macroscopic particle of bentonite is composed of many thousands of stacked and/or
overlapped submicroscopic flakes.
Talc is also often used as an extender in epoxy adhesives and sealants, but it also has
flow control properties. Talc is used in higher-solids, high-viscosity applications such as
caulking compounds, automotive putties, mastics, and sealants. Talc is a hydrophobic and
organophilic material.
Talcs are either platy or acicular in particle shape. Thin platelet particles have aspect
ratios varying from 20 : 1 to 5 : 1. Coarse particle sizes (10 to 75 µm) are commonly used in
these applications at loading levels of 5 to 30 percent. Fine talcs (1 to 10 µm) are more
expensive and require intensive dispersion processes. Platy grades enhance barrier proper-
ties and air, water, and chemical resistance.
Polymers filled with platy talc exhibit higher stiffness, tensile strength, and creep resis-
tance than do polymers filled with standard particulate fillers. These properties are main-
tained at both ambient and elevated temperatures. Surface treatments for talc particles
include magnesium and zinc stearates, silanes, and titanates.
Attapulgite is a very cost-effective thixotrope.14 However, unlike kaolin or talc, it is not
used at high loadings and thus is not considered as an extender. Attapulgite consists of
needlelike particles with a size of about 0.1 µm. Thickening occurs upon the mutual sepa-
ration of individual particles as the porous mineral surface is exposed to the surrounding
medium and forms stable colloids. Cationic surfactants often enhance the performance of
attapulgite thixotropes in organic systems.
Typical usage levels range from 2 to 8 percent by weight. Attapulgite particles are easy
to disperse and commonly added to the formulation with other dry components. At levels
ranging from 10 to 20 percent attapulgite is considered to be a partial asbestos replacement
in polymeric-based sealants for the construction and automotive industries.
Other Thixotropic Additives. Microcellular fillers such as glass and plastic are also
used to provide nonsag properties to adhesives and sealants. However, their mode of oper-
ation is very much different from that of the rheological modifiers mentioned above.
Microcellular fillers work because they drastically reduce the specific gravity of the prod-
uct. Thus, because the adhesive or sealant is lighter, there is less stress on the adhesive or
sealant due to the forces of gravity.
Microcellular fillers such as glass microballoons have been around for some time and
are commonly used in syntactic foam. However, they are expensive raw materials. Sibrico
Corporation, of Hodgkins, IL, has recently introduced a new, relatively inexpensive filler
called Sil-Cell. This is a microsized, multiwall hollow glass filler. Sil-Cells range in size
FILLERS AND EXTENDERS 169
from 1 to 150 µm. The particles have both spherical and irregular geometries, which is
claimed to provide advantages in improving tensile strength. The resin matrix mechanically
keys to the microglass bubble whereas spherical bubbles must rely on coupling agents for
maximum interfacial strength.
If the adhesive has a propensity to flow easily before and during cure, then one risks the
possibility of a final joint that is starved of adhesive material. If the adhesive flows only
with the application of a great amount of external pressure, then one risks the possibility of
entrapping air at the interface and too thick of a bond line. These factors could result in
localized high-stress areas within the joint and reduction of the ultimate joint strength.
Flow characteristics can be regulated by the incorporation of fillers, by the use of scrims
or woven tapes as “internal shims” within the adhesive itself, or by the careful regulation
of the cure cycle. All these options along with a few more are available to the adhesive for-
mulator and end user. Generally, fillers are incorporated to control the viscosity of the adhe-
sives as well as other properties such as thixotropy and sag resistance.
The type and amount of fillers are chosen so that a practical bond line thickness will
result after application of the necessary pressure (usually only contact pressure, approxi-
mately 5 psi) during cure. Ordinarily, the objective is a bond line thickness of 2 to 10 mils.
Consideration, of course, must be given to the curing temperature. Viscosity of the formu-
lation could drastically be reduced at elevated temperatures, and unless there is a furrow
designed into the joint to contain the adhesive, much of the adhesive could flow out of the
joint area before the adhesive is completely cured.
Glass, nylon, polyester, and cotton fabric or mat are often used as a carrier in tape or
film adhesive systems. In addition to being a carrier and a reinforcement, the strands of the
fabric offer an “internal shim” so that the bond line cannot be thinner than the thickness of
these strands. Sufficient pressure need only be applied to cause the adhesive to flow so that
the shims meet the substrate surfaces to provide a positive stop. Paper, mat, and other car-
rier materials may also be used for this purpose.
Glass or polymeric microballoons, incorporated directly into the adhesive formulation,
can also provide the shimming function. Here the diameters of the microballoons are the
positive stops that will prevent too thin of a bond line.
Another option is to design mechanical shims into the joint itself. The parts to be assem-
bled are designed with lips or stops so that the adhesive cannot flow out of the joint area or
so that a certain predetermined adhesive thickness is always maintained.
Depending on the substrate, the curing temperatures, and the service temperatures that are
expected, the adhesive formulator may want to adjust the coefficient of thermal expansion
of the adhesive system. This will lessen internal stresses that occur due to differences in
thermal expansion between the substrate and the adhesive. These stresses act to degrade the
joint strength.
There are several occasions when the difference in coefficient of thermal expansion
between the substrate and adhesive will result in internal stresses in the joint. Common
occurrences are (1) when the cured joint is taken to a temperature that is different from the
curing temperature and (2) when the joint is exposed to thermal cycling.
When a liquid adhesive solidifies, the theoretical strength of the joint is reduced because
of internal stresses and stress concentrations that usually develop. The most common cause
170 CHAPTER NINE
of internal stress is due to the difference in the thermal expansion coefficients of the adhe-
sive and the adherends. These stresses must be considered when the adhesive or sealant
solidifies at a temperature that is different from the normal temperature that it will be
exposed to in service. Figure 3.7 shows thermal expansion coefficients for some common
adhesives and substrates. The significant difference in thermal expansion coefficient
between epoxy adhesives and some substrate materials means that the bulk adhesive will
move more than 10 times as far as the substrate when the temperature changes, thereby
causing stress at the interface.
There are several possible solutions to the expansion mismatch problem. One is to use
a resilient adhesive that deforms with the substrate during temperature change. The penalty
here is possible creep of the adhesives, and highly deformable adhesives usually have low
cohesive strength. Another approach is to adjust the expansion coefficient of the adhesive
to a value that is nearer to that of the substrate. This is generally accomplished by formu-
lating the adhesive with specific fillers to “tailor” the thermal expansion coefficient.
The general effect of most fillers is to reduce the coefficient of thermal expansion of the
cured epoxy resin in proportion to the degree of filler loading. Certain fillers, such as
zirconium silicate and carbon fiber, have a negative coefficient of thermal expansion.
These are very effective in lowering the expansion rate of the epoxy, especially at ele-
vated temperatures.
Ideally, the coefficient of thermal expansion should be lowered (or raised) to match that
of the material being bonded. With two different substrate materials, the adhesive’s coeffi-
cient of thermal expansion should be adjusted to a value between those of the two sub-
strates. This is generally done by using fillers, as shown in Fig. 9.5.
It is usually not possible to employ a sufficiently large filler loading to accomplish the
degree of thermal expansion modification required to match the substrate. High loading vol-
umes increase viscosity to the point where the adhesive cannot be applied or cannot wet the
substrate. For some applications and with some fillers, loading volumes up to 200 pph may
be employed, but optimum cohesive strength values are usually obtained with lesser amounts.
6
Average coefficient of thermal expansion,
Copper powder
in/(in·°C) × 107 (−50 to +50°C)
Aluminium powder
4 Iron powder
Silicon
Calcium carbonate
Parts Aluminium oxide
2 Formulation Aluminium
by weight
Epoxy resin (EEW: 180) 100 Brass
Metaphenylenediamine 15 Steel
Filler As indicated
0
0 10 20 30 40 50 60 70 80 90 100
Filler, parts by weight
FIGURE 9.5 The coefficients of thermal expansion of filled epoxy resins compared with those of common
metals.15
FILLERS AND EXTENDERS 171
9.3.5 Shrinkage
Nearly all polymeric materials (including adhesives and sealants) shrink during solidification.
Sometimes they shrink because of escaping solvent, leaving less mass in the bond line. Even
100 percent reactive adhesives, such as epoxies and urethanes, experience some shrinkage
because their solid polymerized mass occupies less volume than the liquid reactants. Table
3.6 shows typical percentage cure shrinkage for various reactive adhesive systems.
The result of such shrinkage is internal stresses at the adhesive-substrate interface and
the possible formation of cracks and voids within the bond line itself. Depending on the pri-
mary base resin, the adhesive formulator may need to reduce the amount of shrinkage when
the adhesive hardens. This can be accomplished in several ways.
Elastic adhesives deform when exposed to such internal stress and are less affected by
shrinkage. Fillers also reduce the rate of shrinkage by bulk displacement of the resin in the
adhesive formulation. This results in an increase in the inherent bond strength of the adhe-
sive. Fillers may improve operational bond strength by 50 to 100 percent although all of
this improvement is not due to reduced shrinkage (see Sec. 9.3.11) .
In certain applications, such as in the electrical and electronic industries, adhesive systems
must have a degree of electrical and/or thermal conductivity. Electrical conductivity is, of
course, important in adhesives that must make electrical interconnection between compo-
nents and in adhesives that must provide electromagnetic or radio-frequency interference
(EMI and RFI) functions.
Thermal conductivity is also important in highly integrated electronic applications
where the heat generated by components must be transferred to a heat pipe or by some other
means outside the electronic package. Thermal conductivity within adhesive systems is
also a means of reducing exotherm and stresses that could develop during the curing cycle
or other excursions to elevated temperatures.
Increased electrical or thermal conductivity is generally accomplished by adding con-
ductive fillers to the epoxy adhesive formulations. For optimum electrical or thermal con-
ductivity, the particles must be so concentrated that they come into contact with one
another. This generally requires such a high level of filler loading that other properties such
as flexibility and tensile shear strength may be significantly degraded.
Electrical Conductivity. Unmodified polymeric resins are natural insulators and do not
exhibit electrical conductivity. There are certain applications, however, where electrically
conductive adhesives provide a significant value. One such application is the use of con-
ductive adhesives as an alternative for wire or circuit board soldering. Another application,
with less of a requirement for conductivity, is the assembly of components that are shields
or protection from electromagnetic interference.
Appropriate fillers have been used to produce adhesives with high electrical conductiv-
ity. Note that, regardless of the adhesive system itself, electrical conductivity is improved
by minimizing the adhesive bond line and by minimizing the organic or nonconductive part
of the adhesive.
Electrically conductive adhesives owe their conductivity as well as their high cost to the
incorporation of high loadings of metal powders or other special fillers of the types shown
in Table 9.8. If enough metal particles are added to form a network within the polymer
matrix, electrons can flow across the particle contact points, making the mixture electri-
cally conductive. Virtually all high-performance conductive products today are based on
flake or powdered silver. Silver offers an advantage in conductivity stability that cannot
172 CHAPTER NINE
TABLE 9.8 Volume Resistivities of Metals, Conductive Plastics, and Various Insulation Materials
at 25°C16
Thermal conductivity,
Material Btu/(h ⋅ °F ⋅ ft2/ft)
Silver 240
Copper 220
Beryllium oxide 130
Aluminum 110
Aluminum oxide 20
Best silver filled epoxy adhesives 1–4
Aluminum filled epoxy (50%) 1–2
Unfilled epoxies 0.1–0.15
4.2 100
3.8 90
er
wd
po
3.3 80
Thermal conductivity ×104, cal/(cm2 ·s·°C)
m
Thermal conductivity ×104, MW/(m2 ·K)
nu
mi
2.9 70
Alu
2.5 60
e
2.1 olit 50
Cry
a
1.7 Mic 40
ica
1.3 Sil 30
8.4 20
te
bona
m car
4.2 Calciu 10
0 0
0 10 20 30 40 50 60 70
Filler, %
FIGURE 9.6 Effect of various fillers on adhesive thermal conductivity.21
174 CHAPTER NINE
probably due to the effect of the fibers physically connecting the particulate filler in the
system.19,20
Thermal conductivity also helps to improve heat transfer during cure. This reduces
exothermic temperatures and extends pot life, particularly at high filler loadings. Shrinkage
during cure is also reduced, as explained in the sections above.
The majority of mineral fillers have a higher specific gravity than do epoxy resins and will,
therefore, increase the specific gravity of the fully cured product. The increase in specific
gravity is in proportion to the loading volume of the filler.
TABLE 9.10 Arc Resistance, s, of DGEBA as a Function of Filler, Loaded to Produce an Initial
Viscosity of 4000 to 5000 cP at Working Temperature, and Cured with a Stoichiometric Amount of
Listed Curing Agent22
Fillers with a density lower than that of epoxy can be used to provide reduced specific
gravity in cured products. These are usually gas-filled microballoons. Although they gen-
erally bring about a significant increase in viscosity, the microballoon filled epoxies (some-
times called syntactic foam adhesives) are often used in marine applications where low
density and buoyancy are important criteria.
Properties of an epoxy system filled with glass microballoons are shown in Table 9.11.
Other low-density microballoons, based on phenolic and other materials, have been devel-
oped for use in epoxy adhesive formulations.
Only certain fillers can be used to increase the cohesive strength of the cured epoxy adhe-
sive formulation. Generally fibrous or flake fillers, such as talc, glass fiber, or mica, will
bring about a certain degree of increase in tensile and flexural strength. However, to notice
a significant increase in physical strength, these fillers must be used at a relatively high
concentration. Particulate fillers, on the other hand, usually reduce room temperature ten-
sile and flexural strength, although some improvement may be experienced at elevated
temperatures.
Tensile strength and flexural modulus are generally increased and ultimate elongation
is reduced in proportion to the amount of filler, when tested at room temperature. The
reverse is generally true when the cured epoxy is tested at elevated temperatures. Impact
strength is generally adversely affected by particulate fillers.
The addition of particulate fillers generally decreases compression fatigue, but increases
ultimate compressive modulus and compressive yield strength, because of a stiffening
effect. Compressive strength as a function of filler loading is shown for a cured epoxy for-
mulation in Fig. 9.7.
The reinforcement of adhesives has typically been achieved through the use of fibrous
and flake fillers. Asbestos historically has been one of the most widely used reinforcing
fibers in adhesive systems due to its needlelike shape. However, it is now seldom used
because of health concerns.
One material that has been used to replace asbestos in certain applications is wollas-
tonite or calcium metasilicate. This is also a fibrous mineral filler but with a lower aspect
ratio than asbestos. Surface-treated versions are available to improve adhesion to the epoxy
matrix. It can be applied at relatively high loading levels to provide for high strength and
improvements in moisture resistance.25
Talc or hydrated magnesium silicate is another mineral that is used to reinforce epoxy
adhesives. It has a platelike structure that provides good stiffness and creep resistance at
elevated temperatures. It also provides good electrical and chemical resistance characteris-
tics.26 It is relatively inexpensive and disperses well in the resin.
TABLE 9.11 Properties of DGEBA Filled with Glass Microballoons Compared with Resin Filled
with Mica (Epoxy Cured with 12 pph Diethanolamine)23
20
16
Iron oxide
14
12
MD 101 Aluminum
10
8
0 10 20 30 40 50 60 70 80
Percent filler by weight
FIGURE 9.7 Effect of filler concentration on compressive strength
of DGEBA cured with 8 pph of DETA.24
Fillers improve the thermal properties of the cured epoxy formulation by bulk displacement
of organic components. This reduces long-term shrinkage and thermally induced weight
loss. However, the glass transition temperature is not significantly affected and in some
cases may be lowered.
The main effect of fillers on thermal properties is achieved through improvement in
thermal shock resistance. This is achieved via modification of the thermal expansion
coefficient. Differences in particulate fillers have been noticed with regard to thermal
shock resistance. Silica, for example, provides relatively poor thermal shock improve-
ment. Mica and fibrous fillers used in relatively high loading provide very good
improvement. However, their use is often precluded due to their handling difficulties.
Fillers having high thermal conductivity will also increase the heat transfer in the cured
FILLERS AND EXTENDERS 177
epoxy system. The improved heat dissipation will result in improved thermal cycling
properties.
Aluminum powder, in particular, is frequently employed at relatively high concentrations
in high-temperature epoxy adhesive formulations. The filler provides improvement in both
tensile strength and heat resistance, and it increases the thermal conductivity of the adhesive.
Aluminum powder fillers also reduce undercut corrosion and, hence, improve adhesion and
durability of epoxy adhesive between bare steel substrates. It is believed that this is accom-
plished by the aluminum filler providing a sacrificial electrochemical mechanism.27
Fillers often have a significant effect on the moisture resistance, the moisture-vapor trans-
mission rate, and the solvent and chemical resistance of the cured adhesive bond. The effect,
however, can be in either direction. Some fillers such as calcium carbonate tend to lower acid
resistance, whereas others, such as silica or aluminum, may tend to lower alkali resistance.
Many fibrous fillers exhibit a wicking action for moisture, and this is particularly true
of high-surface-energy glass fibers and especially when the fibers are exposed, such as
when the cured epoxy is machined or a crack is formed in the adhesive resin matrix. For
this reason glass fibers that are used in adhesives are often treated with a coupling agent
such as an organosilane to improve the bond between the fiber and the epoxy matrix.
Particulate fillers, on the other hand, are believed to extend the pathway along which the
water must diffuse, resulting in a reduction in the rate of water absorption. Silica flour (1
to 75 µm) has been reported as improving the boiling water resistance of epoxy films.
Figure 9.8 shows the water absorption rate at 40°C of water in DGEBA epoxy filled with
various types of materials.
Fillers have very little effect on the actual adhesion mechanism, since most of this is con-
trolled by intimate contact of the liquid epoxy and the substrate. However, as mentioned
4
1. Resin 4. +50% Microsil
2. +30% clay 5. +65% Zr(SiO3)2
3. +50% CaCO3
3
1
2
Weight gain, %
3
2
1
5
0 2 4 6 8 10 12
Time, days
FIGURE 9.8 Water absorption versus filler for cured DGEBA.28
178 CHAPTER NINE
Aluminum oxide
Tensile shear strength, psi × 103
Talc
Mica
2
Parts
Formulation by weight
Epoxy resin (EEW: 180) 75
Glycerol-based epoxy 25
Triethylamine 12.5
1 Filler As indicated
0
0 20 40 60 80 100
Filler, parts by weight
FIGURE 9.9 The effect of mineral extenders on the tensile shear strength of epoxy adhesive (the sub-
strate is aluminum).29
above, fillers improve the measured adhesive strength of cured epoxy joints by decreasing
internal stresses related to shrinkage and differences in the coefficient of thermal expan-
sion. Because of this, one can generally observe a significant improvement in tensile shear
strength with the addition of fillers. Fillers also provide for a method of controlling viscos-
ity so that a predetermined and regulated bond line thickness is possible.
The degree of tensile strength improvement is often in the 50 to 100 percent range. The
effect of various fillers and loading ratios on the strength properties of epoxy adhesive for-
mulation is indicated in Fig. 9.9. The effect of different fillers loaded at a constant 100 pph
is indicated in Table 9.12 for shear strength on phenolic laminate and aluminum substrates.
Although the addition of fillers often results in an improvement in tensile shear strength,
as noted above, it also often results in a decrease in peel or impact strength. This is due to
Aluminum block
Phenolic laminate tested at tested at
the reduction in elongation and toughness that is often accompanied by high filler loadings
in epoxy adhesive systems. When an adhesive joint is stressed in peel or cleavage, the more
rigid bond line results in all the stress being focused at a very small region near the edge of
the bond. Adhesives with a higher degree of elongation distribute the stress over a large
area, resulting in higher peel and cleavage strength.
Fillers generally lower the curing reaction rate and reduce the degree of exotherm. This is
primarily due to their diluting effect and the resulting increase in thermal conductivity. The
effect of various fillers on working life and peak exotherm is presented in Table 9.13. The
more filler added, the less will be the heat evolved during cure. However, even with highly
filled epoxy systems, the filled resin is still a relatively poor conductor of heat.
Char formers: Usually phosphorus compounds, which remove the carbon fuel source
and provide an insulation layer against the fire’s heat
Heat absorbers: Usually metal hydrates such as aluminum trihydrate (ATH) or magne-
sium hydroxide, which remove heat by using it to evaporate water in their structure
Flame quenchers: Usually bromine- or chlorine-based halogen systems that interfere
with the reactions in a flame
Synergists: Usually antimony compounds, which enhance performance of the flame
quencher
Flame retardants for polymers can also be classified as reactive (the flame-retarding
agent reacts chemically with the polymer to become an integral part of the molecule) or
additive (nonreactive agents that are simply blended or mixed into the compound). Flame
retardants can also be classified by their major chemical group, as shown in Table 9.14.
There are many families of flame retardants, each with advantages and disadvantages.
It is common to formulate polymers with multiple flame retardants, typically a primary
TABLE 9.13 Effect of 50 percent Filler Loading on Viscosity, Exotherm, Modulus of Rupture,
and Shrinkage of an Epoxy Resin31
Alumina trihydrate Available in various particle-size Polyesters, phenolics, and epoxies. Also Albemarle-Martinswerk, Alcan Chemicals,
grades and in surface-treated grades polypropylene and ethylene/rubber Alcoa, Aluchem, Amspec Chemical,
compounds DuPont Polymer Modifiers, J.M. Huber,
Nabaltec
Magnesium Magnesium hydroxide Synergists with ATH for smoke reduction Addenda, Albemarle-Martinswerk, Amspec
Magnesium carbonate Chemical, Dead Sea Periclase, DSBG,
J.M. Huber, Kyowa Chemical Industry,
Sumitomo Chemical, Witttaker
Chlorines Chlorinated paraffins LDPE film and flexible PVC Amspec Chemical, Asahi Denka Kogyo,
Tris (dichloropropyl) phosphate, Urethane foam and topical fabric ICC Industries-Dover, Kettlitz Chemie,
methyl pentachlorostearate, and Polypropylene and nylon Lehmann & Voss, OxyChem, Polytechs
other chlorinated phosphates Reactive intermediate in making polyester and
180
flame retardant plus a synergist such as antimony oxide, to enhance overall flame resistance
at the lowest cost. Several hundred different flame-retardant systems are used by the poly-
mer industry because of these formulation practices.
Selecting a flame retardant for an adhesive system has many ramifications, depend-
ing on the formulation being modified, the end use, how it will be processed, and the
cost/performance ratio. When one is choosing a flame retardant, characteristics such as
water extraction, particle size, viscosity, toxicity, dusting, uniformity, as well as econom-
ics must be considered. The materials chosen to perform the function of flame retardation
must not interfere with the final product’s performance. The major problem with incorpo-
rating flame retardants in adhesives is that very often a significant amount is required, and
they interfere with the other properties of the adhesive and contribute to the cost. This is
why bromo bisphenol epoxy resins are often employed in flame-retardant epoxy adhesives.
The criteria in selecting and using the right flame-retardant system for a specific adhe-
sive formulation can be broad and somewhat complex. Care must also be taken with regard
to choosing the proper test method and determining the parameters of the test. Consultation
with the flame-retardant suppliers is generally necessary to optimize this process.
9.3.14 Color
Fillers can be used as pigments to provide color to the epoxy formulation. Adhesive color-
ing is used to match the color of substrates. Different coloring is often added to each com-
ponent in a two-part epoxy as a method for determining efficient mixing of the components.
The resultant color from the mixture helps the user know that the mix ratios are close and
that the degree of mixing is sufficient.
Both pigments and dyes have been used successfully as colorants. With liquid epoxy
resins, inorganic pigments seem to provide optimal properties. Organic pigments are gen-
erally less effective with epoxy resins.
Pigments are generally dispersed into the resin formulation at relativity low percentages
(1 to 3 percent) to provide the required color. Titanium dioxide is frequently employed in con-
nection with the colorant to provide a whiting agent with the necessary hiding power suitable
for tinting. The pigments are generally dispersed directly into the epoxy formulation.
However, concentrated dispersions in liquid DGEBA and in diluents, such as dibutyl phtha-
late, are commercially available and may provide for easier incorporation of the pigment.
REFERENCES
1. Potter, W. G., Epoxide Resins, Springer-Verlag, New York, 1970, pp. 115–116.
2. Materials Engineering, May 1976, p. 73.
3. Katz, H. S., and Milewski, J. V., eds., Handbook of Fillers for Plastics, van Nostrand Reinhold,
New York, 1987.
4. Lee, H., and Neville, K., Epoxy Resins, McGraw-Hill, New York, 1957, p. 148.
5. Magee, J. E., “Formulating Techniques and Evaluations,” in Epoxy Resin Technology, P. F.
Bruins, ed., John Wiley & Sons, New York, 1968, p. 120.
6. Potter, Epoxide Resins, p. 118.
7. Lee, H., and Neville, K., Handbook of Adhesive Resins, McGraw-Hill, New York, 1982, p. 14.7.
8. Grace, M. J., and Sullivan, C., “Cellulose Fiber as a Reinforcing Thixotrope in Adhesive and
Sealants,” Adhesives Age, July 1995.
9. Technical Bulletin, “Kevlar: Applications,” Sealants and Adhesive Group, DuPont Company,
Wilmington, DE, 2001.
FILLERS AND EXTENDERS 183
10. Nargiello, M., “Hydrophobic Silicas Contribute to Consistency in Formulating,” Adhesives Age,
July 1989, pp. 30–31.
11. Katz and Milewski, Handbook of Fillers for Plastics, p. 180.
12. Ralston, H. P., “Kaolin, Talc, and Attapulgite Extend and Thicken Adhesives,” Adhesives Age,
July 1989, pp. 32–34.
13. Additives for Adhesives and Sealants, Elementis Specialties, Hightstown, NJ, 2002.
14. Carr, J. B., “Attapulgite-Based Thixotropes Offer Performance and Flexibility,” American Paint
and Coatings Journal, November 1997.
15. Lee and Neville, Epoxy Resins, p. 151.
16. Conn, R. C., “Special Effects,” Adhesives Age, February 2002, pp. 47–48.
17. Ishibashi, K., and Kimura, J., “A New Anisotropic Conductive Film with Arrayed Conductive
Particles,” AMP Journal of Technology, vol. 5, June 1996, p. 2430.
18. Bolger, J. C., and Morano, S. L., “Conductive Adhesives: How and Why They Work,” Adhesives
Age, June 1984.
19. Katz and Milewski, Handbook of Fillers for Plastics.
20. Lee and Neville, Handbook of Adhesive Resins.
21. Buckley, W. O., and Schroeder, K. J., “Adhesive Modifiers,” in Adhesives and Sealants,
Advanced Materials Handbook, ASM International, Materials Park, OH, 1990, p. 178.
22. Lee and Neville, Handbook of Adhesive Resins, p. 14.35.
23. Quaint, A., “Low Density Potting Compound,” Sandia Corp., Preprint SCR-417, 1961.
24. Technical Bulletin, The Effect of Fillers on the Physical Properties of Epoxy Resins, Shell
Chemical Corp., Houston, TX, 1960.
25. Seymour, R. B., ed., Additives for Plastics, Academic Press, New York, 1978.
26. Katz and Milewski, Handbook of Fillers for Plastics.
27. Wright, C. D., and Muggee, J. M., “Epoxy Structural Adhesives,” Chapter 3 in Structural
Adhesives, S. R. Hartshorn, ed., Plenum Press, New York, 1986.
28. Formo, J., and Bolstad, L., “Where and How to Use Epoxies,” Modern Plastics, July 1955.
29. Burgman, H. A., “Selecting Structural Adhesive Materials,” Electrotechnology, June 1965.
30. Lee and Neville, “Epoxy Adhesives,” Chapter 21 in Handbook of Adhesive Resins.
31. Lee and Neville, Epoxy Resins, p. 150.
This page intentionally left blank
CHAPTER 10
ADHESION PROMOTERS
AND PRIMERS
10.1 INTRODUCTION
Some adhesives and sealants may provide only marginal adhesion to certain substrates.
This could be due to the low surface energy of the substrate relative to the adhesive (e.g.,
epoxy bonding polyethylene) or to a boundary layer that is cohesively weak (e.g., powdery
surface on concrete). The substrate may also be pervious, allowing moisture and environ-
mental chemicals to easily pass through the substrate to the adhesive interface, thereby
degrading the bond’s permanence. Generally, attempts are made to overcome these prob-
lems through adhesive formulation and by substrate surface treatment. When these
approaches do not work, additional bond strength and permanence may possibly be pro-
vided by the use of (1) primers or (2) adhesion promoters.
Primers and adhesion promoters work in a similar fashion to improve adhesion. They
add a new, usually organic, layer at the interface. The new layer can be bifunctional and
bond well to both the substrate and the adhesive or sealant. The new layer is very thin so
that it provides improved interfacial bonding characteristics, yet it is not so thick that its
bulk properties significantly affect the overall properties of the bond.
Both primers and adhesion promoters are strongly adsorbed onto the surface of the sub-
strate. The adsorption may be so strong that instead of merely being physical adsorption, it
has the nature of a chemical bond. Such adsorption is referred to as chemisorption to dis-
tinguish it from reversible physical adsorption.
The main difference between primers and adhesion promoters is that primers are liquids
that are applied to the substrate as a relatively heavy surface coating prior to application of
the adhesive. Adhesion promoters, on the other hand, are liquids that form a very thin (usu-
ally monomolecular) layer between the substrate and the adhesive. Usually chemical bonds
are formed (1) between the adhesion promoter and the adhesive and (2) between the adhe-
sion promoter and the substrate surface. These bonds are stronger than the internal chemi-
cal bonds within the adhesive. These new bonds also provide an interface region that is
more resistant to chemical attack from the environment. Adhesion promoters are also
sometimes referred to as coupling agents.
Adhesion promoters can be applied either by incorporating them directly into the adhe-
sive formulation or by applying them directly to a substrate, similar to a primer. When
applied in situ, through the adhesive formulation, the adhesion promoter migrates to the
interface region and attaches itself between the adhesive molecule and the substrate before
the adhesive cures.
Adhesion promoters or coupling agents are also used in applications other than improv-
ing the adhesive-substrate interface. With highly filled compounds, adhesion promoters
185
Copyright © 2006 by The McGraw-Hill Companies, Inc. Click here for terms of use.
186 CHAPTER TEN
provide formulators a way of effectively improving properties and reducing the overall cost
of the compound. Adhesion promoters can be applied to particulate fillers or fibers to
improve their wetting characteristics. In this way, viscosity can be decreased or more filler
can be added while maintaining a specified viscosity. Bulk properties of the resulting adhe-
sive, such as tensile strength and modulus, can also be significantly improved. Thus, when
one is discussing adhesion promoters, the term substrate takes on the added possibilities of
fillers, reinforcements, etc., as well as conventional adherends.
Adhesion promoters are a group of specialty bifunctional compounds that can react chem-
ically with both the substrate and the adhesive. The adhesion promoter forms covalent
bonds across the interface that are both strong and durable. Adhesion promoters can be
applied directly to the substrate, similar to primers, or they can be mixed with the adhesive
itself. When mixed into the adhesive formulation, adhesion promoters are sometimes
referred to as coupling agents because they can interact with both the substrate and the other
fillers in the formulation.
When mixed with the adhesive, the adhesion promoter is capable of migrating to the
interface and reacting with the substrate surface as the adhesive cures. When applied
directly to the substrate, adhesion promoters are used as a very thin coating that ideally is
only one molecular layer thick.
Adhesion promoters can also be applied directly to mineral fillers or fibers prior to their
addition to a formulation. The adhesion promoter provides for a formulation that has sev-
eral unique benefits:
1. Enhanced processability, shown by reduced viscosity, easier flow, and lower energy
required to compound and mix
2. Increased tensile, flexural, and impact strength
3. Better retention of these properties under adverse temperature and humidity conditions
over long periods of time
Adhesion promoters usually consist of molecules with short organic chains having dif-
ferent chemical compositions on either end of the chain. On one end is an organofunctional
group that is particularly compatible with the given adhesive material. At the other end of
the chain is an inorganic functionality that is especially compatible with a given substrate.
The adhesion promoter, therefore, acts as a chemical bridge between the adhesive and the
substrate.
Adhesion promoters were first used to treat glass fibers and other fillers before they are
incorporated into liquid resin to make composite materials. In the fiber industry, adhesion
promoters are also known as finishes. Certain finishes have been specially developed to
match a fiber with a resin matrix. Without adhesion promoters, the interfacial resin–glass
fiber adhesion is weak, and water can diffuse along the interface with catastrophic results
on the end properties of the composite.
Silanes are the most common commercial adhesion promoter. They are commonly used to
enhance adhesion between polymeric and inorganic materials.1,2 They usually have the form
X3Si––R, where X is typically a chlorine or alkoxy group and R is the organofunctionality.
ADHESION PROMOTERS AND PRIMERS 187
The organofunctional portion bonds with the resin in the adhesive or the organic medium,
and the silane portion bonds to the inorganic or substrate surface. Silane coupling agents
are commonly used between the adhesive and the adherend, between resin matrix and rein-
forcing fibers in composites, and between resin matrix and mineral fillers in compounds.
The resulting interface provides
These chemicals are usually applied to fillers or to the substrate surface as an aqueous
solution, but it can also be applied via solvent solution. The solutions usually are very dilute,
only 0.01 to 2 percent by weight of silane to keep the highly reactive hydrolyzed molecules
from reacting with one another. The bond strength enhancement increases with silane con-
centration up to a maximum of about 2 percent, and then the enhancement falls off gradu-
ally with additional concentration. Silane coupling agents react with water in aqueous
solutions to form hydrolyzed silanes, which then further react with the surface of the inor-
ganic substrate. The bound silanes then polymerize, building up layers outward from the
substrate with the organic functionality oriented toward the adhesive. This process is shown
in Fig. 10.1.
Silanes form strongly adsorbed polysiloxane films on ceramic and metal surfaces. The
chemical and mechanical integrity of these films are highly dependent on application para-
meters such as solution concentration, solution pH, and drying time and temperature. The
character of the substrate may also influence the polysiloxane film structure.4 Silanes are
applied as primers to the surface of the substrate by wiping, spraying, brushing, or dipping.
The film thickness is less than 0.1 mil. The solvent in the silane system is removed by
drying at 50 to 60°C for 10 min. The advantage of the primer application method is that
it efficiently utilizes the silane material, and there are minimal stability problems. The dis-
advantages are that it is a two-step process and that it is difficult to see the clear silane coat-
ing unless it is pigmented. Therefore, complete surface coverage may be in doubt.
Tests have shown that silanes arrange themselves in layers with a high degree of order,
influenced to a great extent by the surface of the substrate. The molecules order themselves
virtually perpendicular to the surface to which they attach, and subsequent layers arrange
themselves steplike in a head-to-head fashion. A rough surface can break up the first
ordered layer, preventing the formation of the second. The thickness of the silane inter-
phase has important effects on mechanical performance. Thinner, but continuous, layers
seem to give stronger and more durable adhesive bonds.
OH R R R
OCH3 OH
3H2O 3 R Si OH O Si O Si O Si O
R Si OCH3 R Si OH OH OH OH 3H2O +
O O O
OH
CH3 OH
Hydrolized silane Inorganic Polymerized silane
surface on surface
FIGURE 10.1 Organosilane coupling agents react with water in aqueous solutions to form hydrolyzed
silanes that react with the surface of inorganic fillers or the adherend. The bound silane polymerizes, build-
ing up layers outward, with the organic functionality oriented to the resin matrix.3
188 CHAPTER TEN
Three primary mechanisms have been suggested for enhanced adhesion via silane cou-
pling agents.5 The classical explanation is that the functional group on the silane molecule
reacts with the adhesive resin. Another possibility is that the polysiloxane surface layer has
an open porous structure. The liquid adhesive penetrates the porosity and then hardens to
form an interpenetrating interphase region. The third mechanism applies only to polymeric
adherends. It is possible that the solvent used to dilute and apply the silane adhesion pro-
moter opens the molecular structure on the substrate surface, allowing the silane to pene-
trate and diffuse into the adherend.
The interphase provided by the adhesion promoter may be hard or soft and could affect
the mechanical properties. A soft interphase, for example, can significantly improve fatigue
and other properties. A soft interphase will reduce stress concentrations. A rigid interphase
improves stress transfer of resin to the filler or adherend and improves interfacial shear
strength. Adhesion promoters generally increase adhesion between the resin matrix and
substrate, thus raising the fracture energy required to initiate a crack.
There are a number of silane adhesion promoters available, and they differ in the nature
of their reactivity to the resin or adhesive. Silanes may be produced with amine, epoxy,
mercaptan, and other functionalities. Some examples are given in Table 10.1.
Many mineral fillers are also commercially available with an organofunctional silane
surface treatment. Suppliers of treated mineral fillers are shown in Table 10.2. These fillers
are used in a variety of applications having critical property requirements that must be pro-
tected from moisture. Surface treatment of the mineral fillers also provides the formulator
with a tool to reduce the viscosity of highly filled systems.
Glass fibers and fabrics are commercially available with silane treatments. The result-
ing physical properties and the resistance of the composite to deterioration by water immer-
sion are greatly enhanced by the addition of the coupling agent. Table 10.3 shows the effect
of silanes on the dry and wet flexural strength of several composite materials.
The moisture resistance properties of filled molding compounds are enhanced by the
treatment of the fillers with silane adhesion promoters prior to compounding. Silane pro-
moters on wollastonite fillers in thermoplastic polyester molding compounds (50 percent
filled) will improve the flexural strength after 16 h in 50°C water by as much as 40 percent.
Silane-treated silica fillers have been found to significantly increase the moisture resistance
of epoxy adhesives used in the electronics industry for chip, surface-mounted, and printed-
circuit processes.8
Silanes are generally effective in improving adhesion to metals, including aluminum,
steel, cadmium, copper, and zinc. Figure 10.2 shows the relative influence of the type of
substrate on the effectiveness of the silane coupling agent in improving adhesion. Note that
smooth, high-energy substrates are excellent substrates for silane attachment, and rough,
discontinuous substrates show very little benefit.
Organic
Filler functionality Treated filler Supplier
Silica
Excellent
Quartz
Glass
Aluminum
Copper
Alumina
Inorganics
Silane effectiveness
Aluminosilicates (Clays)
Good
Mica
Talc
Inorganic oxides
Steel, Iron
Asbestos
Slight
Nickel
Zinc
Lead
Chalk (Calcium carbonate)
Gypsum (Calcium sulfate)
Barytes (Barium sulfate)
None
Graphite
Carbon black
Silane adhesion promoters increase the initial bond strength and also stabilize the sur-
face to increase the durability and permanence of the bonded joint in moist aging environ-
ments.10,11 The lap shear values in Table 10.4 show the improvement in initial bond strength
when silane coupling agents are incorporated into the adhesive formulation. Silane-based
coupling agents are capable of increasing the environmental resistance of aluminum,12 tita-
nium,13 stainless steel,14 and other metal joints. The effect of a silane adhesion promoter on
the durability of mild steel joint bonded with an epoxy adhesive is shown in Fig. 10.3.
Although the best results can be obtained in using silanes as substrate primers, they can
also be added to the adhesive with some effect. The integral blend method of applying the
silane involves adding 0.1 to 2.0 percent by weight of the silane to the polymer matrix prior
to application. The advantage of this method is that it does not require a separate substrate
TABLE 10.4 Tensile Shear Strengths of Several Structural Adhesives Incorporating Silane
Adhesion Promoters15
2024T3 aluminum, 0.063-in thickness, 12 -in overlap. Tests run at 23°C; appropriate cure schedule used.
ADHESION PROMOTERS AND PRIMERS 191
60 9
1% aqueous solution
y-glycidoxypropyltrimethoxysilane
primer
Butt joint strength, MPa
20 3
0 0
0 500 1000 1500
Exposure time, h
FIGURE 10.3 Effect of silane adhesion promoter on the durability of a mild steel
joint bonded with epoxy adhesives.16
coating step. The disadvantage is that there could be stability problems in storage due to
hydrolysis and polymerization. Also, the efficiency of promoter coverage at the interface
is reduced due to the migration process.
Conventional silanes are limited to high-solids and solvent adhesive and sealant appli-
cations in which moisture is not encountered until use. However, water-stable epoxy silane
promoters have recently been developed to enhance the wet strength of waterborne adhe-
sives and sealants.17
While silanes predominate as adhesion promoters, the uses of titanate, zirconate, and other
agents are growing. These coupling agents also offer increased impact strength and other
physical properties when used with fillers or applied at the interface. Like the silanes, they
react with surface hydroxyl groups, but there is no condensation polymerization to produce
a polymer network at the interface. Similar to the silanes, organometallic adhesion pro-
moters typically provide a dual function of improving processing and improving adhesion.
Organic titanates and zirconates have been found to provide several important functions
as additives for organic adhesives and sealants. At least eight primary functions have been
proposed.18 Organic titanates and zirconates can be used as
4. Surface protection
5. Water scavengers
6. Crosslinkers
7. Catalyst for crosslinking
8. Thixotropic agents
The mechanisms and processes by which the organic titanates and zirconates provide
these important functions are summarized in Table 10.5.
The use of organic titanates and zirconates for surface modification is based on their
ability to hydrolyze to a coating that is very thin, amorphous, and primarily inorganic. The
properties of this film depend on the type and amount of organometallic coupling agent
used, the chemistry of the organometallic, and the processing properties used to apply the
coating. These coatings modify the surface of the filler or substrate to provide the follow-
ing unique properties.
• They promote adhesion of adhesives and coatings to glass, metal, and plastics.
• The organometallic interface improves dispersibility of pigments and fillers in aqueous
and nonaqueous systems and reduces viscosity.
• It can provide scratch-resistant and reflective properties to glass.
• It can modify frictional characteristics of the substrate.
Titanates The titanate structure may be tailored to provide desired properties through the
functionalities on the basic structure shown below.
(RO)m––Ti––(O––X––R2––Y)n
The (RO)m is the hydrolyzable proton that attaches to the inorganic substrate, as shown in
Fig. 10.4. It also controls dispersion, adhesion, viscosity, and hydrophobicity. The X group
enhances corrosion protection and acid resistance and may provide antioxidant effects,
depending on the chemistry. The R2 provides entanglements with long hydrocarbon chains
and bonding via van der Waals forces. The Y group provides thermoset reactivity, chemi-
cally bonding the filler to the polymer. It can be one of any number of chemical groups
reactive with a number of different matrices.
OH
OCH3
OH
x Ti (C OR′)3
+
O
OH
OTi (C OR′)3
OTi (C OR′)3
+ x CH3OH
OTi (C OR′)3
Like the silanes, the organic titanates react with surface hydroxyl groups, but there is no
condensation polymerization to produce a polymer network at the interface. Titanate cou-
pling agents are unique in that their reaction with free protons on the substrate surface
results in a monomolecular layer whether it is a filler or substrate. Thus, one of the most
common problems in using organic titanates is overconcentration. Since excess titanate
(amount greater than necessary to form a monolayer) does not result in a polymer network,
it is suspected that it can form a weak boundary layer, resulting in degraded properties.
Thus, the amount of titanate that is used is an important parameter.
Typically, titanate-treated inorganic fillers or reinforcements are hydrophobic, organophilic,
and organofunctional and, therefore, exhibit enhanced dispersibility and bonding with the poly-
mer matrix. When used in filled polymer systems, titanates claim to improve impact strength,
exhibit lower viscosity, and enhance the maintenance of mechanical properties during aging.
Major suppliers of titanate coupling agents include DuPont (Tyzor) and Kenrich
Petrochemicals (Ken-React). Typical titanate coupling agents are shown in Table 10.6.
There are many types and chemistries available. The manufacturer of the titanate is an
excellent source for information on the appropriate type, dosage, and processing require-
ments in specific applications. Titanates can be made available as liquids (solutions and
water-soluble salts). New neoalkoxy technology allows powdered and pellet coupling
agents to be processed as concentrates without pretreatment steps.21
Adhesion promoters based on organotitanium have been used to improve particulate
filler adhesion and to pretreat aluminum alloy for adhesive bonding.22 Titanates have been
used predominantly to modify the viscosity of filled systems.23 It has been shown that a
small percentage of titanate in a heavily filled resin system can reduce the viscosity signif-
icantly. Thus, titanate adhesion promoters allow higher fillings of particulate matter to
either improve properties or lower the cost of the system without having a negative effect
on the viscosity. Improved bond strength to halocarbon substrates and improved hydrolytic
stability are also claimed.24
Typical dosage of organotitanates is 0.2 percent by weight of polymer. They can also be
used to treat aluminum alloy for adhesive bonding. Researchers have investigated titanate
as a primer for adhesion of 1-2, polybutadiene modified epoxy to anodized aluminum.25
The breaking strength of the joints is strongly affected by the amount of titanate used as
well as by the drying condition used prior to adhesive application.
Zirconates Zirconate coupling agents have very similar structure to the titanates. They
also perform similar functions. Zirconium compounds exist in both water and organic sol-
vent soluble forms. Like the titanates, zirconate coupling agents are useful in improving the
dispersion characteristics of fillers in polymer systems. Examples of zirconate coupling
Titanate Application/advantages
Zirconate Application/advantage
agents and their applications are shown in Table 10.7. Manufacturers of organic zirconate
coupling agents include DuPont and Kenrich Petrochemical.
The aqueous chemistry of zirconium is complex and dominated by hydrolysis. One
aspect is that polymerization takes place when salt solutions are diluted. The polymeric
species can be cationic, anionic, or neutral. Polymers that are formed include ammonium
zirconium carbonate, zirconium acetate, and zirconium oxychloride.
Zircoaluminates, introduced in 1983, claim performance at least comparable to that of
silanes at substantial cost savings. Several functionalities are available. They are stable and
soluble in an aqueous environment and do not require the presence of water to function. The
surface reaction is irreversible. Among the fillers treated successfully are silica, clay, cal-
cium carbonate, alumina trihydrate, and titanium dioxide.
10.3 PRIMERS
Primers are resinous liquids that are applied to a substrate prior to application of an adhe-
sive or a sealant. They provide a thicker coating at the interface than do adhesion promot-
ers, and therefore often affect the bulk properties of the joint. The reasons for their use are
varied and may include, either singularly or in combination, the following:
• Protection of surfaces after treatment (primers can be used to extend the time between
preparing the adherend surface and bonding)
• Adjustment of the free surface energy by providing a surface that is more easily wettable
than the substrate
196 CHAPTER TEN
• Dissolving of low levels of organic contamination that otherwise would remain at the
interface as a weak boundary layer
• Promotion of the chemical reaction between adhesive and adherend
• Inhibition of corrosion of the substrate during service
• An intermediate layer to enhance the physical properties of the joint and improve bond
strength (e.g., adjustment of the rheological properties at the interface or strengthening
weak substrate regions)
Being lower in viscosity than the adhesive or sealant, primers can be used to penetrate
porous or rough surfaces to provide better mechanical interlocking and for sealing such sur-
faces from the environment. Primers are often applied and appear as protective surface
coatings. Application processes and equipment for applying primers are similar to those
used in applying a paint coating to a substrate.
The application of a primer is an additional step in the bonding process, and it comes
with associated costs and quality control requirements. Therefore, primers should be used
only when justified. The most likely occasions for a primer to be used are when (1) the
adhesive or sealant cannot be applied immediately after surface preparation, (2) the sub-
strate surface is weak or porous, or (3) the adhesive-adherend interface requires additional
protection from service environments such as moisture.
Epoxy-based primers are commonly used in the aerospace and automotive industries.
These primers have good chemical resistance and provide corrosion resistance to aluminum
and other common metals. Primer base resins, curing agents, and additives are much like
adhesive or sealant formulations except for the addition of solvents or low-viscosity resins
to provide a high degree of flow.
Unlike substrate surface treatments, primers always add a new organic layer to the surface
and two new interfaces to the joint structure. Most primers are developed for specific adhe-
sives, and many are developed for specific adhesive/substrate combinations.
Primers are applied quickly after surface preparation and result in a dry or slightly tacky
film. It is generally recommended that they have a dried coating thickness range of tenths
of a mil to approximately 2 mils. It is necessary to control the primer thickness, since if the
primer layer becomes too thick, its bulk properties may predominate, and the primer could
become the weakest part of the joint.
Primers usually require solvent evaporation and several curing steps before the adhe-
sive or sealant can be applied. Adhesive primers are usually not fully cured during their
initial application. They are dried at room temperature, and some are forced-air dried for
30 to 60 min at 65°C. This provides a dry, nontacky surface that can be protected from
contamination and physical damage by good housekeeping practices until the substrate is
ready to be bonded with an adhesive. Full primer cure is generally achieved during the
cure of the adhesive.
Primers developed to protect treated surfaces prior to bonding are generally proprietary
formulations manufactured by the adhesive producer to match the chemistry of the adhe-
sive. These usually consist of a diluted solution (approximately 10 percent by weight) of
the base adhesive in an organic solvent. Like the adhesive formulation, the primer may also
contain wetting agents, flow control agents, and toughening compounds. If the primer is for
a metal surface, corrosion inhibitors such as zinc and strontium chromate and other inor-
ganic chromate salts may also be added to the primer formulation.
ADHESION PROMOTERS AND PRIMERS 197
The application of corrosion-resistant primers has become standard practice for the
structural bonding of aluminum in the automotive and aerospace industries. The adhesive-
primer combinations are chosen to provide both maximum durability in severe environ-
ments and higher initial joint strength. Improved service life is typically achieved by
establishing strong and moisture-resistant interfacial bonds and protecting the substrates’
surface region from hydration and corrosion.
Primers can be used to protect both treated metal and nonmetal substrates after surface
treatment. The use of a primer as a shop protectant may increase production costs, but it
may also provide enhanced and more consistent adhesive strength. The use of a primer
greatly increases production flexibility in bonding operations. Usually primer application
can be incorporated as the final step in the surface preparation process. The primer is
applied as soon as possible after surface preparation and usually no more than a few hours
later. The actual application of the adhesive may then be delayed significantly.
With aluminum, for example, the maximum safe surface exposure time (SET) inter-
val between surface preparation and bonding is 12 h. Many other substrates have maxi-
mum SETs that are less than this. By utilizing an adhesive primer, the SET may be
extended to days and even months depending on the particular adhesive-primer system
used and the storage conditions prior to bonding. This process allows a shop to prepare
the surfaces of a large number of parts, prime them, and store them for relatively long
periods prior to bonding. It also enables an assembly shop to outsource the more haz-
ardous surface preparation processes. The primer provides protection of the treated joints
during transportation between the treating shop and the bonding shop. With primers,
scheduling of the entire assembly operation is not dependent on the nature of the surface
preparation.
As with metallic substrates, primers may be used to protect treated nonmetallic sub-
strates. After surface treatment, a high-energy substrate has an active surface that will read-
ily adsorb atmospheric contamination. The primer protects the treated surface until the time
when the adhesive or sealant is applied. Primers are especially useful in this way for the
protection of polymeric parts that are treated by flame or corona discharge. Primers also
find benefit on polymeric substrates in that their solvents will soften the surface, and some
of the primer resin will diffuse into the bulk of the substrate, thereby increasing the adhe-
sive strength by molecular diffusion.
Another production-related reason for priming is associated with bonded assemblies
having many subsections. With primers, individual subsections can be treated, primed, and
then fit into place before the bonding step without regard to time. This then allows the entire
assembly to be bonded at one time.
Sometimes primers can take the place of surface treatments. Two examples are with
porous substrates and with certain plastic substrates. With weak porous substrates, such as
wood, cement, or porous stone, the primer can be formulated to penetrate and bind weakly
adhering material to provide a new, tightly anchored surface for the adhesive. Chlorinated
polyolefin primers will increase the adhesion of coatings and adhesives to polypropylene
and to thermoplastic olefins. The chlorine atoms in the outer surface of the primer increase
surface energy and enhance adhesion of adhesives, sealants, and paints.
Low-viscosity primers can also easily fill the irregularities on the substrate surface
and displace air and fill hollows. This can improve the wetting properties of the adhe-
sive “system.” For example, if the adhesive is a hot melt and it is applied to a bare metal-
lic surface, the adhesive will gel before it gets a chance to efficiently wet the surface and
mechanically interact with any surface roughness. However, if a dilute primer were first
applied to the substrate and dried, the hot-melt adhesive could bond directly to the
primer that in turn has bonded to the interstices of the substrate, thus providing excel-
lent adhesion.
198 CHAPTER TEN
When a corrosive medium contacts the edge of a bonded joint and finds an extremely active
surface, such as that produced by a fresh acid treatment of the metal substrate to improve
adhesion, corrosion at the metal-adhesives interface can occur. This initial corrosion and
the subsequent penetration can take several forms.
Some primers will inhibit the corrosion of metal adherends during service. By protect-
ing the substrates’ surface area from hydration and corrosion, these primers suppress the
formation of weak boundary layers that could develop during exposure to wet environ-
ments. Primers that contain film-forming resins are sometimes considered interfacial water
barriers. They keep water out of the joint interface area and prevent corrosion of the metal
surfaces. By establishing strong, moisture-resistant bonds, the primer protects the adhesive-
adherend interface and lengthens the service life of the bonded joint. However, moisture
can diffuse through any polymeric primer, and eventually it will reach the interface area of
the joint. Therefore, the onset of corrosion and other degradation reactions can only be
delayed by the application of a primer, unless the primer contains corrosion inhibitors or it
chemically reacts with the substrate to provide a completely new surface layer that provides
additional protection.
Representative data are shown in Fig. 10.5 for aluminum joints bonded with an epoxy film
adhesive and a standard chromate-containing primer. Up until recently, standard corrosion-
resistant primers contained high levels of solvent, contributing to high levels of volatile
organic compounds (VOCs) and chromium compounds, which are considered to be carcino-
gens. As a result, development programs have been conducted on waterborne adhesive
primers that contain low VOC levels and little or no chrome. Data are presented on several of
these primers in Tables 10.8 and 10.9.
5000
1.2-in lap shear strength after exposure, psi
4000
250° cure film, 250° cure film,
250° cure film, standard primer corrosion resistant
without primer primer
3000 (O = Individual coupon)
2000
1000
Note: To increase test severity, 1/2-in shear coupons, l in wide with both
edges cut, were exposed to salt spray.
0
0 10 20 30
Days exposed to 5% salt spray
FIGURE 10.5 Effect of primer on lap shear strength of aluminum joints exposed to 5% salt spray.31
TABLE 10.8 Examples of Commercial Primers for Structural Adhesives32
BR 127 (a) FM 73, 94, 903 2.5–5.0 µm Air drying 30 min; curing Corrosion-inhibiting; protects metal
toughened epoxy at 120°C for 30 min oxide from hydrolysis
adhesives
EA 9203 (b) EA 9602 and other 5–10 µm Air drying 60 min Corrosion-inhibiting for use with room
epoxy adhesives temperature curing adhesives
Primer 2000 (b) Thermosetting acrylic 5–10 µm Air drying in 3–5 min at Cleans and etches aluminum and stainless
adhesives room temperature steel
Provides improved durability to salt water
EC-2320 (c) AF-111, AF-126, and 1.3–5.1 µm Air drying up to 120°C Improves shear and peel strengths as
199
Source:
(a) Cytec Industries
(b) Hysol – Loctite Div. of Henkel
(c) 3M Company
(d) Mitsui Chemicals
(e) Tramaco GmbH
200 CHAPTER TEN
TABLE 10.9 Tensile Lap Shear Strengths Using Environmentally Acceptable Adhesive Primers
for 121°C Curing Adhesives33
30-Day 60-Day
Primer (source) Adhesive 22°C 82°C* 82°C wet† salt fog‡ salt fog‡
REFERENCES
1. Plueddemann, E. P., Silane Coupling Agents, Plenum Press, New York, 1991.
2. Mittal, K. L., Silane and Other Coupling Agents, VSP, Utrecht, The Netherlands, 1992.
3. English, L. K., “Fabrication of the Future with Composite Materials, Part IV: The Interface,”
Materials Engineering, March 1987.
4. Bascom, W. D., “Primers and Coupling Agents,” Adhesives and Sealants, vol. 5, Engineered
Materials Handbook, ASM International, Materials Park, OH, 1990.
5. Kinloch, A. J., Adhesion and Adhesives, Chapman and Hall, London, 1987, pp. 156–157.
6. Berger, S. E., “Organofunctional Silanes,” in Handbook of Fillers for Plastics, H. S. Katz and
J. V. Milewski, eds., van Nostrand Reinhold, New York, 1987.
7. Marsden, J. G., and Sterman, S., “Organofunctional Silane Coupling Agents,” in Handbook of
Adhesives, 3d ed., I. Skeist, ed., van Nostrand Reinhold, New York, 1990.
8. Conn, R. C., “How Silane Fumed Silica Influences Adhesion Properties of a Model Epoxy
Adhesive System after Water Immersion,” Adhesives Age, February 2002.
9. Waldman, B. A., “Organofunctional Silanes as Adhesion Promoters and Crosslinkers,” Kent State
University, Ohio, Adhesion Principles and Practice Course, May 1997.
10. Plueddemann, E. P., Silane Coupling Agents.
11. Bascom, “Primers and Coupling Agents.”
12. Patrick, R. L., et al., Applied Polymer Symposium, vol. 16, 1981, p. 87.
ADHESION PROMOTERS AND PRIMERS 201
13. Boerio, F. J., and Dillingham, R. G., in Adhesive Joints, K. Mittal, ed., Plenum Press, New York,
1984, p. 541.
14. Chovelon, J. M., et al., “Silanization of Stainless Steel Surfaces: Influence of Application
Parameters,” Journal of Adhesion, vol. 50, 1995, pp. 43–58.
15. Kothari, V. M., “Specialty Adhesives Can Be Compounded for Applications in Many Market
Niches,” Elastomerics, October 1989.
16. Kinloch, A. J., “Predicting and Increasing the Durability of Structural Adhesive Joints,” in
Adhesion, vol. 3, K. W. Allen, ed., Elsevier Science Publisher, New York, 1978.
17. Huang, M. W., and Waldman, B. A., “Water Stable Epoxysilanes Enhance Wet Strength of
Sealants and Adhesives,” Adhesives and Sealants Industry, Oct./Nov. 1998.
18. Technical Note, Adhesives and Sealants, DuPont Tyzor Organic Titanates, Wilmington, DE, 2001.
19. Technical Note, Adhesives and Sealants.
20. Pocius, A.V., Adhesion and Adhesives Technology, Hanser Publishers, New York, 1997, p. 139.
21. Monte, S. J., and Sugerman, G., “New Neoalkoxy Titanate Coupling Agents Designed to
Eliminate Particulate Pretreatment,” 125th Meeting of the Rubber Division, Indianapolis, IN,
May 8–11, 1984.
22. Calvert, P. D., et al., “Interfacial Coupling by Alkoxytitanium and Zirconium Tricarboxylate,” in
Adhesion Aspects of Polymer Coatings, K. L. Mittal, ed., Plenum Press, New York, 1983, p. 457.
23. Waddington, S., and Briggs, D., “Adhesion Mechanism between Polymer Coatings and
Polypropylene Studies by XPS and SIMS,” Polymer Communications, vol. 32, 1991, p. 506.
24. Monte, S. J., “Titanates,” in Modern Plastics Encyclopedia, McGraw-Hill, New York, 1989, p. 177.
25. Monte, S. J., “Titanates,” Chapter 4 in Handbook of Fillers for Plastics, H. S. Katz and J. V.
Milewski, eds., van Nostrand Reinhold, New York, 2000, pp. 88–90.
26. Kinloch, Adhesion and Adhesives, pp. 156–157.
27. De Nicola, A. J., and Bell, J. P., “Synthesis and Testing of β-Diketone Coupling Agents for
Improved Durability of Epoxy Adhesion to Steel,” in Adhesion Aspects of Polymeric Coatings,
K. L. Mittal, ed., Plenum Press, New York, 1983, p. 443.
28. Yoshida, S., and Ishida, H., “A FT-IR Reflection Absorption Spectroscopic Study of an Epoxy
Coating on Imidazole Treated Copper,” Journal of Adhesion, vol. 16, 1984, p. 217.
29. Cotton, J. B., et al., U.S. Patent 3,837,964, 1974.
30. van Ooij, W. J., and Biemond, M. E. F., “A Novel Class of Rubber to Steel Tire Cord Adhesion
Promoters,” Rubber Chemistry and Technology, vol. 57, 1984, p. 688.
31. Krieger, R. B., “Advances in Corrosion Resistance of Bonded Structures,” in Proceedings of the
National SAMPE Technology Conference, vol. 2, Aerospace Adhesives and Elastomers, 1970.
32. Comyn, J., Adhesion Science, Royal Society of Chemistry, Cambridge, England, 1997, pp. 27–28.
33. Kuhbander, R. J., and Mazza, J. J., “Evaluations of Low VOC Chromated and Nonchromated
Primers for Adhesive Bonding,” 38th Annual SAMPE Symposium, May 1993, pp. 785–795.
This page intentionally left blank
CHAPTER 11
ROOM TEMPERATURE
CURING EPOXY ADHESIVES
11.1 INTRODUCTION
One of the distinct advantages of epoxy adhesives is that they can be cured at room tem-
perature or even at lower temperatures. Epoxy adhesives are often divided into room tem-
perature curing types and elevated-temperature curing types. This chapter discusses room
temperature epoxy formulations. The major advantages and disadvantages of room tem-
perature curing epoxy adhesives are shown in Table 11.1.
Room temperature curing epoxy adhesives are two-component adhesives, indicating that
the curing agent portion and the epoxy resin portion are packaged separately. These are often
referred to as “2K” adhesives, with the K derived from the German spelling of komponent.
One of the two components is commonly referred to as a base, and it contains the epoxy
resin including additives. The second component is commonly referred to as the curing
agent, hardener, catalyst, activator, or accelerator, and it contains the triggering agent for
the crosslinking reaction. The two components are also sometimes referred to as part A
(base) and part B (curing agent).
The two parts are mixed just prior to application. Metering occurs before the mixing
operation at the time of assembly, or the two components are packaged in preweighted con-
tainers of various types and sizes.
There are also one-component epoxy systems that do not require mixing and cure at
room temperature. These, however, need an energy source such as ultraviolet light or elec-
tron beam for the crosslinking reaction to proceed. There are also two-component epoxy
formulations that are premixed by the supplier, immediately frozen, and shipped to the
user in the frozen state. These, of course, must be kept in the frozen state prior to use, and
even in this situation the shelf life is relatively limited. However, to the user, this type of
adhesive has many of the application advantages of a one-component adhesive. These uv
cure epoxy adhesives and innovative packaging technologies are discussed in upcoming
chapters.
Once mixed, the two-part room temperature curing epoxy adhesives are designed to
react at ambient conditions and temperatures near room temperature (20 to 25°C) or
lower. In many cases the cure time may be accelerated with heat, but this is not a require-
ment. Once the adhesive is mixed, the crosslinking reaction begins immediately, and the
resulting working life is limited. The cure rate and working life are dependent on the spe-
cific formulation as well as the ambient temperature. Although the adhesive hardens to a
handling strength in usually a short time, it continues to develop strength for a much
longer period.
203
Copyright © 2006 by The McGraw-Hill Companies, Inc. Click here for terms of use.
204 CHAPTER ELEVEN
TABLE 11.1 Advantages and Disadvantages of Room Temperature Curing Epoxy Adhesives
Advantages Disadvantages
Very long shelf life of stored components Short working life of mixed adhesive that can
at room temperature result in waste, difficult application
No thermal energy required to heat either the Components that must be metered accurately
adhesive or the parts being bonded and mixed thoroughly
Can be accelerated by exposure to elevated Tensile strength, heat and chemical resistance
temperatures not as high as when cured at elevated
temperatures
Can be cured at room temperature to a Long cure times that may limit production and
handling strength and then exposed to require fixturing
elevated temperatures for final cure
Less shrinkage and internal stress resulting from
thermal expansion differences
Moderate strength as well as heat and chemical
resistance
Good toughness and moderate peel and impact
strength
Lower cost due to less energy and no heating
equipment
Less hazardous (less vapors, no high-temperature
equipment)
With room temperature curing epoxy adhesives, very reactive curing agents are required.
An exothermic temperature increase is generated during the cure reaction, the degree of
which depends on the specific epoxy adhesive system and the mass of adhesive that is mixed.
This exotherm may limit the working time of the adhesive before it is applied to the substrate.
However, any heat that is produced by the exothermic reaction is quickly dissipated due to
the thin film of the adhesive and the thermal conductivity of the adjoining substrates.
Depending on the type of curing agent chosen, epoxy adhesive systems can reach full
strength in minutes to several days. The most common types of room temperature curing
epoxy adhesives reach handling strength in several hours and full cure in 5 to 7 days. The
faster-curing agents can reach handling strength in as little as 5 min. It is generally good
practice to allow all room temperature epoxy adhesives to cure for 5 to 7 days for the full
strength to develop.
The most widely used epoxy adhesive of this type employs a polyamide or amidoamine
curing agent. Even the simplest of such systems, consisting of equal parts of a liquid
DGEBA epoxy resin and a polyamide hardener, provide good bond strength and durability
in most environments. However, other curing agents may also be used depending on the
application and performance properties required. These include crosslinking aliphatic
amines, such as DETA and triethylenetetramine; catalytic amines, such as DMP-30; sulfur
curing agents such as the mercaptans; and amine adducts. The curing mechanisms relative
to these various room temperature curing agents are presented in Chap. 5.
Two-part, room temperature curing adhesives are generally preferred for large parts due
to the energy required to heat the substrates. Two-part room temperature curing epoxies
eliminate ovens and heaters while offering a reasonably short time to reach handling
strength. If a large part is to be assembled or a large amount of adhesive is to be applied, a
two-component adhesive with a short pot life would be inappropriate. However, automatic
ROOM TEMPERATURE CURING EPOXY ADHESIVES 205
metering, mixing, and dispensing equipment is available to minimize the time for adhesive
preparation and application. As a convenient alternative to bulk containers, many two-part
epoxy adhesives are available in dual-chamber cartridges for economical handheld appli-
cations, which simultaneously meter, mix, and dispense the epoxy through disposable sta-
tic mixing nozzles.
Another characteristic difference between a two-component, room temperature curing
epoxy adhesive system and a one-component, heat curing system is the shelf life. The shelf
life of a one-component epoxy can be 3 to 6 months when stored at room temperature and
may even require refrigeration or freezing for a practical shelf life. The shelf life of each
component in a two-component system is generally 6 months to 1 year.
Two-part, room temperature curing epoxy adhesives are generally more flexible than
one-part epoxy adhesives. The curing mechanism is much different, and the resulting prop-
erties are more dependent on the type of curing agent used than are those of the one-
component epoxy adhesive systems. As a result of the greater flexibility, these adhesives
have generally higher bond strength to flexible substrates (e.g., thermoplastics) or sub-
strates that tend to change dimensions with the environment (e.g., wood, cardboard). The
higher degree of flexibility results in improved low-temperature performance and better
thermal shock resistance. The greater flexibility also provides these adhesives with a higher
degree of peel and cleavage strength and resistance to impact-type forces.
Table 11.2 shows typical physical property and performance data for several two-
component epoxy adhesives. As can be seen from the property data, the adhesives can be
Viscosity, Pa ⋅ s
• Resin 50 260 91
• Hardener 35 160 103
• Mixed 45 250 54
Mix ratio, weight 100 resin/ 100 resin/ 100 resin/
80 hardener 100 hardener 71 hardener
Specific gravity
• Resin 1.17 1.48 1.36
• Hardener 0.92 1.44 0.97
Gel time 2h 4 min 1h
Cure schedule 24 h at 25°C or 4 h at 25°C 5 days at 25°C or
30 min at 100°C 2 h at 88°C
Elongation at break, — — 11
percent
Aluminum tensile
shear strength, psi, at
• −60°C 2900 1500 4200
• 25°C 2600 2900 4500
• 82°C 300 1200 2600
• 121°C — — 1000
Tensile shear after
• 30 days at 60°C 2200 3000 3800
• 30 days after 1700 (40°C, 92% RH) 1000 (54°C, 95% RH) 2900 (54°C, 95% RH)
humidity aging
T-peel strength, lb/in — — 11
206 CHAPTER ELEVEN
28 28
24 24
Shear strength, psi × 10−2
20 20
16 16
12 12
8 DETA 8
4 DMP−30 4
0 0
4 8 12 0 40 80 120 160 200
Days at 25°C Hours at 120°C
FIGURE 11.1 Effect of cure on shear strength of DGEBA formulations.3
ROOM TEMPERATURE CURING EPOXY ADHESIVES 207
used in the formulation, and the length of cure. When room temperature curing epoxy
adhesives are cured at elevated temperatures or are given a postcure at elevated tempera-
tures, the heat resistance is somewhat improved. However, the heat resistance of epoxies
that are fully cured at elevated temperatures is generally far superior. The effect of cure
temperature and time on typical room temperature curing rigid epoxy adhesives is shown
in Fig. 11.1.
Chemical resistance is similarly affected by curing agent type and the temperature of the
cure. Moisture, solvent, and general chemical resistance are usually superior for epoxy adhe-
sives that are cured at elevated temperatures. Room temperature epoxy adhesives, having a
lower glass transition temperature, are more severely affected by environmental exposures.
Room temperature curing epoxy adhesives provide widely varying application and per-
formance properties depending on the formulation employed. The following sections high-
light certain formulations and commercial products that fall under this popular classification
of epoxy adhesives.
Many curing agents can be used in two-part, room temperature curable epoxy adhesives.
However, the most useful curing agent families are polyamides, amidoamines, and
aliphatic amines. These curing agents react directly with the epoxy and become part of the
crosslinked epoxy structure, once cured. The polyamides and amidoamine curing agents
provide flexibility and a degree of toughness to the cured epoxy. The aliphatic amines are
somewhat more reactive and provide moderate temperature performance and good chemi-
cal resistance. However, they are generally more brittle than the polyamides or ami-
doamines. Table 11.4 shows the effect of each of these curing agents on selected properties
of an epoxy adhesive.
Polyamides and amidoamines are perhaps the most popular type of curing agent used in
general-purpose epoxy adhesives. They are either used as the sole curing agent or blended
with other curing agents or accelerators in consumer epoxy adhesive products, such as two-
tube, ready-to-mix systems. They provide low toxicity, excellent flexibility, good toughness,
TABLE 11.4 Curing Agent Selection for General-Purpose Room Temperature Curing
Epoxy Adhesives
Tensile strength + + − +
Peel, impact strength + +
Heat resistance + + − −
Chemical resistance + + − −
Water resistance + + − +
Safety, hazardous vapors − + + +
+ Relatively good.
− Relatively poor.
208 CHAPTER ELEVEN
and good water resistance. Older types of polyamide curing agents have high viscosity that pre-
sents compatibility problems with certain liquid epoxy resins and long cure times that could be
degraded by moisture in the ambient environment. Newer polyamide and amidoamine prod-
ucts, however, have lower viscosity and better cure properties under adverse conditions.
The polyamide and amidoamine curing agents react directly with the epoxy resin and
become an integral part of the cured product. Their flexibility helps to distribute the stress
and impact loads throughout the adhesive joint. A typical liquid DGEBA cured with a
polyamide will give peel strengths of about 16 lb/in on metal. Peel strengths of only 4 to 5 lb/in
are common for the more rigid epoxy adhesives.
Higher percentages of polyamide or amidoamine curing agents in the epoxy adhesive
will improve peel strength; lower percentages will provide higher-temperature and chemi-
cal resistance at the cost of reduced peel and impact properties. The epoxy–curing agent
ratio for these systems can range from 1 : 2 to 2 : 1.
The pot life of polyamide or amidoamine cured epoxy adhesives is generally on the
order of hours at room temperature. Full cure is achieved in 5 to 7 days at room tempera-
ture, and handling strength is achieved in about 16 to 24 h. A faster cure can be achieved
in 20 min to 4 h by heating to 60 to 150°C. When room temperature cures are required, an
accelerator such as an amine is often added to the formulation.
Table 11.5 shows a typical formulation for an accelerated and unaccelerated epoxy-
amidoamine adhesive system. The accelerated formulation provides a faster-curing adhe-
sive, but it is not as flexible. Common fillers such as alumina, talc, and silica can be
incorporated up to a level of 100 pph before any deleterious effect on properties is observed.
A typical epoxy-polyamide adhesive that is useful for general metal bonding applica-
tions is described in Table 11.6. The effect of varying the filler type and concentration on
lap shear strength at room and elevated temperatures and resistance to impact is also shown.
Some fillers increase the tensile shear strength of epoxy adhesives that are cured with
polyamide or amidoamine curing agents, and other fillers will reduce the strength.
As a family of curing agents for epoxy resins, the amidoamines are lower in viscosity
than the polyamides. They exhibit very good adhesive properties due to their chemical
structure and easy penetration. Amidoamine cured epoxy adhesives have shown very good
properties on concrete and other porous substrates. They cure extremely well under humid
conditions. In fact amidoamine cured epoxy formulations have been used to cure under-
water in certain applications. A typical general-purpose room temperature curing epoxy-
amidoamine system is described in Table 11.7. This adhesive is used as a general-purpose
metal-to-metal adhesive and body solder in the automotive industry.
Aliphatic amines typically provide two-component, room temperature curing epoxy for-
mulations with fast reactivity. They are often used as accelerators for polyamide and
TABLE 11.6 Effect of Filler on Tensile Shear Strength of Polyamide Cured Epoxy Adhesives5
After mechanical
Type of filler Filler, pph At 23°C At 120°C shock of 80 in ⋅ lb
Adhesive formulation:
DGEBA epoxy (EEW: 180–200) 45 pbw
Fatty polyamide (amine value: 210–230) 50 pbw
Butyl glycidyl ether 5 pbw
Filler As specified
This formulation with aluminum filler will provide tensile shear strengths as follows on various substrates:
Aluminum 3200 psi
Cold-rolled steel 2800 psi
Naval brass 2100 psi
Hard cooper 1400 psi
amidoamine curing agents. When compared to polyamide and amidoamine curing agents,
aliphatic amines provide high shear strength; moisture insensitivity; good chemical resis-
tance, particularly to solvents; and somewhat higher operating temperatures, especially
when cured at elevated temperatures. Typical of the aliphatic amines are diethylene
tetramine (DETA), triethylenetetramine (TETA), and catalytic amines such as DMP-30.
Table 11.8 presents a typical triethylenetetramine cured epoxy adhesive formulated with
selected fillers. In this formulation the use of aluminum powder and alumina increases sub-
stantially the resistance of the adhesive to boiling water.7 This is also true when DETA is used
as the curing agent.8 A typical room temperature cured aliphatic amine cured epoxy adhesive
for general-purpose use is shown in Table 11.9. This shows the difference that is achieved in
shear strength by curing at elevated temperatures versus room temperature.
As with amidoamine and polyamide cured adhesives, epoxy resins cured with aliphatic
amines exhibit tensile shear strength that is dependent on the type of filler and concentra-
tion. Table 11.10 shows the effect of filler loading on strength of a simple general-purpose,
room temperature curing epoxy adhesive composed of liquid DGEBA epoxy mixed with
10 pph of a tertiary amine.
TABLE 11.9 Room Temperature Cure, General-Purpose Epoxy Adhesive Cured with
Triethylenetetramine10
Part A
Liquid epoxy resin (EEW: 190) 70
Aliphatic epoxy resin (e.g., DER 732, Dow) 30
Alumina T-60 50
Part B
Triethylenetetramine 13
Cure conditions 7 days at 25°C 2 h at 95°C
Tensile shear strength, psi
Aluminum, 16-gauge 1150 1600
Stainless steel, 16-gauge 1400 1730
210
ROOM TEMPERATURE CURING EPOXY ADHESIVES 211
Highly reactive two-component epoxy adhesives have several advantages. These adhesives
can come to a practical handling strength in minutes at room temperatures and can develop
useful bond strengths at ambient temperatures as low as −20°C. The various uses for these
adhesives include
These fast-curing systems also have difficulties, however, that are directly related to
their reactivity. Generally, the disadvantages must be balanced against the relative merits
of using these adhesives. The major disadvantages of using fast-curing epoxy adhesives
include the following:
Thus, fast cure is often looked upon as a liability rather than a benefit. Because the fast-
acting catalysts react so quickly with epoxy resins and because they are used in very small
proportions, homogeneous mixing is difficult and variation in crosslinking density can
occur throughout the adhesive. Another consequence of high reactivity is a high exotherm.
A high exothermic temperature can be degrading to the final epoxy adhesive properties and
even dangerous if large masses of mixed adhesives overheat.
The curing agents used with fast-setting epoxy adhesives are typically accelerated
aliphatic amines and accelerated polymercaptan-terminated curatives. Lewis acids, such as
BF3, can also provide exceptionally fast cure, frequently within seconds of introducing the
catalyst, although they are not generally used in practical adhesive systems. Figure 11.2
compares the pot lives and bond line gel times of these curing agents.
Many fast-reacting epoxy adhesives utilize a polymercaptan and an accelerator as the
main curing agent. The mixing ratio provided by the mercaptan is very practical. The
polymercaptan-epoxy formulations bond to most substrates except low-surface-energy
plastics. They are especially effective on alkaline surfaces such as concrete, gypsum board,
and rubber containing amine antioxidants. Epoxy cured with mercaptan-terminated curatives
212 CHAPTER ELEVEN
BF3 dihydrate in
polyester carrier
Accelerated
polymercaptan
(Epi−Cure 861)
Accelerated amine
(Epi−Cure 874)
0 10 20 30 40 50
Minutes to gel state
FIGURE 11.2 Gel time relationship of highly reactive adhesives (100 g mixing mass at 25°C; 0.006-in-
thick adhesive film at 25°C).13
Part A
Epoxy resin (EEW: 190–210) 60
Silica flour 60
Carbon black 0.1
Fibrous filler or colloidal silica 3
Part B
Polymercaptan (e.g., Dion 3-800LC) 75
Polyamide 12
Tertiary amine 8
Silica flour 50
Titanium dioxide 10
Fibrous filler or colloidal silica 4
and accelerated with amine catalysts is the basis of the “5-min epoxies” often found in hard-
ware stores.
A fast-curing polymercaptan-epoxy adhesive formulation is shown in Table 11.11. The
gel time for the mixed adhesive is about 8 min at room temperature. Tensile shear strength
on aluminum is reported to be 2270 psi at 23°C. Note that the different color pigments used
in parts A (carbon black) and B (titanium dioxide) are to assist by visually indicating when
the two components are well mixed.
The primary disadvantages of polymercaptans over other fast-reacting curing systems
are (1) skin irritation caused by aromatic mercaptans (aliphatic mercaptans are less der-
matitic than common polyamides) and (2) an objectionable sulfur odor, especially when the
vapor pressure of the mercaptan is relatively high.
Figures 11.3 and 11.4 show the tensile shear strength development at 23 and −18°C,
respectively, of mercaptan cured epoxy formulations relative to accelerated amine and ami-
doamine curing systems. The practical cure times at −18°C illustrate why mercaptan cured
epoxy adhesive systems are popular in the building and construction industries where the
adhesive can be applied at cold, outdoor temperatures.
300 A
B D
Tensile bond strength, psi
C E F
200
100
0
0 5 10 15
Cure time at 25°C, h
Adhesive formulations
Parts by weight
Component A B C D E F
300
A
Tensile bond strength, psi
B
200
C
100
0
0 10 20 30
Cure time at 18°C, h
Adhesive formulations
Parts by weight
Component A B C
FIGURE 11.4 Tensile bond strength development at −18°C of various adhesives (steel-to-
concrete).16
Exceptionally fast cures can also be achieved with an epoxy acrylate resin. These resins
provide very high reactivity when combined with aliphatic amine curing agents. These can
be used alone or as a highly reactive modifier in fast-setting adhesives or in adhesives that
must set at low temperatures. Substituting conventional DGEBA epoxy for the epoxy acry-
late can decelerate the cure speed. This allows the ability to “dial in” a preferred rate of
gelation and cure speed.
Table 11.12 shows two rapid-setting, room temperature cure epoxy adhesives based on
epoxy acrylate resins with aliphatic amine curing agents. These adhesives have gel times
of less than 5 min for a 100-g mass. The bond strength development is rapid with handling
strength occurring in about 1 h at room temperature.
The room temperature cured epoxy adhesives discussed thus far exhibit a general lack of
flexibility, especially when considered next to elastomeric sealants. Flexibility is generally
desired when the performance requirements include high peel strength, impact strength,
and resistance to thermal shock or thermal cycling.
ROOM TEMPERATURE CURING EPOXY ADHESIVES 215
TABLE 11.12 Starting Formulations for Rapid-Setting, Room Temperature Curing Epoxy
Adhesives17
Parts by weight
Part A
Epoxy acrylate resin (EPON 8111, Resolution 100 100
Performance Products)
Fumed silica (Cab-O-Sil M5 Silica, Cabot Corp.) 5 5
Ground quartz (Novacite 1250, Malvern Minerals Co.) 150 75
Part B
Aliphatic amine (Epi-Cure 3271, Resolution 22.1 5
Performance Products)
Aliphatic amine (Epi-Cure 3270, Resolution 80
Performance Products)
Fumed silica (Cab-O-Sil M5 Silica, Cabot Corp.) 5
Ground quartz (Novacite 1250, Malvern Minerals Co.) 77
Property
Several methods can be used by the formulator to improve the flexibility of room tem-
perature curing epoxy resin adhesive formulations:
The most common method is by using long-chain, flexible curing agents or epoxy resins
(items 1 and 2 above). Flexible polyamide and amidoamine curing agents have been used
as described to provide less brittleness in epoxy adhesives. Table 11.13 is a starting for-
mulation for a flexible, high-peel-strength epoxy adhesive using both polyamide curing
agents and reactive diluent as flexibilizers for the base epoxy molecule. This can be used
as a general purpose adhesive on metal and plastic substrates.
Figure 11.5 shows 90° peel strength and tensile shear strength for epoxy adhesives con-
sisting of various amounts of a flexibilized epoxy resin (Epi-Rez 505) in a mix with a stan-
dard DGEBA liquid epoxy resin. The peel strength increased dramatically with the
concentration of flexibilized resin. The accompanying increase in tensile shear strength is
because of better stress distribution within the lap shear joint due to the greater flexibility
rather than an increase in cohesive strength of the adhesive.
216 CHAPTER ELEVEN
TABLE 11.13 Starting Formulation for a Flexible Epoxy Adhesive Containing Polyamide Curing
Agent and Reactive Diluent18
Part A
DGEBA epoxy resin (EEW : 190) 50.5
Epodil 748 diluent (Air Products and Chemicals Inc.) 7.0
Talc (Microtuff 325F, Barretts Minerals, Inc.) 40.0
Fumed silica (Cab-O-Sil TS-720, Cabot Corp.) 2.5
Part B
Polyamide (Ancamide 910, Air Products and Chemicals, Inc.) 38.0
Modified polyamide (Ancamide 2482, Air Products and Chemicals, Inc.) 15.0
Talc (Microtuff 325F, Barretts Minerals, Inc.) 25.0
Fumed silica (Cab-O-Sil TS-720, Cabot Corp.) 2.0
Property
Flexibilized reactive epoxy compounds have also been used for providing stress relief
and properties of high peel and impact strength to an epoxy adhesive. A starting formula-
tion for such an adhesive is shown in Table 11.14. This formulation incorporates a modi-
fier (HELOXY 505, Resolution Performance Products) that is a polyglycidyl ether of castor
oil. It provides the formulation with good flexibility and relatively low viscosity. This can
be used as a general-purpose adhesive where good peel and impact strength and resistance
to thermal cycling are required.
The remainder of this section explores room temperature curing epoxy formulations
where a great deal more flexibility is required. The most important application for this type
of epoxy system is as a sealant or potting compound.
Polysulfides provide substantially greater flexibility than do the polyamide or ami-
doamine curing agents. They generally also provide lower working viscosities and are
widely used for concrete bonding. Liquid polysulfides can be utilized over a rather wide
mixing range with liquid DGEBA epoxy resins to provide controlled amounts of flexibil-
ity to adhesive joints. Two typical formulations for a two-component, epoxy-polysulfide
adhesive are described in Table 11.15.
The mixed components provide a pot life of about 20 min at room temperature. The
adhesive will reach a handling strength overnight. After 7 to 14 days of curing at room tem-
perature or 2 h at 70°C, the bond will develop a tensile shear strength of about 500 psi.
Epoxy-polysulfide adhesives are noted by their rather low degree of cohesive strength, but
high degree of elongation. By varying the concentration of DMP-30, the cure speed can be
varied according to need.
ROOM TEMPERATURE CURING EPOXY ADHESIVES 217
4000 16
3000 12
2000 8
1000 4
0 0
0 10 20 30 40 50 60 70
% Epi−Rez 505 substitution
Adhesive formulation
FIGURE 11.5 Effect of a flexibilized epoxy on tensile shear and peel strength of epoxy
adhesive formulation.19
The polysulfide additive provides excellent adhesion to concrete because of its chemi-
cal nature and low viscosity, resulting in good penetration. Table 11.16 describes a formu-
lation for a flexible epoxy-polysulfide adhesive. This adhesive is claimed to provide
excellent adhesion between new and old concrete surfaces. Table 11.17 describes several
epoxy-polysulfide formulations for general-purpose adhesives.
Epoxy-polysulfide systems are used in the construction, electrical, and transportation
industries because of their unique combination of cured flexibility, adhesion to many sub-
strates, and chemical resistance. They are typically used as adhesives for
Because of their excellent flexibility, epoxy-polysulfide adhesives are commonly used for
bonding plastic substrates where differences in thermal expansion are a concern.
218 CHAPTER ELEVEN
Part A
DGEBA epoxy resin (EPON 828, Resolution Performance Products) 75
PGE of castor oil (HELOXY 505, Resolution Performance Products) 25
Fumed silica (Cab-O-Sil TS-720, Cabot Corp.) 2
Part B
Amidoamine (Epi-Cure 3046, Resolution Performance Products) 42
Fumed silica (Cab-O-Sil TS-720, Cabot Corp.) 0.6
Property
Table 11.18 shows the effect of increasing the polysulfide-epoxy ratio on impact prop-
erties; the effect at concentrations of polysulfide polymer greater than 1 : 1 is especially sig-
nificant. The combination of increased flexibility, tensile strength, and elongation is very
desirable in certain adhesive applications.
Polysulfide-epoxy adhesives can be used in various chemical and solvent environments,
even under immersion conditions. The systems exhibit low permeability to water and water
Part A
DGEBA epoxy resin (Bakelite ERL-2774, Bakelite AG) 100
DGEBA epoxy resin (EPON 828, Resolution Performance Products) 50
Surfex M.M. (Diamond Shamrock Corp.) 50
Part B
Polysulfide LP-3 (Toray Chemicals) 100 25
Silica 80
Tertiary amine (e.g., DMP-30) 20
Aliphatic amine (Epi-Cure 3253, Resolution Performance Products) 5
Property
Part A
LP-3 Polysulfide 100
Hydrite Clay 121 140
Trimethylaminomethlyphenol 20
Toluene 65
Part B
Epoxy resin (DER 331, Dow) 200
Hydrite Clay 121 105
Toluene 5
Property*
vapor, which provides improved water and corrosion resistance. Epoxy-polysulfide adhe-
sives also cure with very low shrinkage, in contrast to unmodified epoxy resins, which may
exhibit volume shrinkage as high as 5 percent. Epoxy-polysulfide adhesives provide resis-
tance to thermal shock and cycling as well as resistance to low-temperature environments.
Typical properties of several epoxy-polysulfide adhesives are shown in Table 11.18.
Room temperature tensile shear strength ranges from about 500 to 3000 psi, depending on
the substrate and the adhesive formulation. Generally, tensile strength properties reach a
maximum at about 10 to 30 percent polysulfide concentration. Peel strength can be as high
as 20 lb/in.
The heat distortion temperature is only slightly affected by the incorporation of 20 per-
cent polysulfide polymer, but at a 1 : 1 ratio the drop becomes significant. This property
prevents the use of epoxy-polysulfide adhesives at elevated temperatures.
Parts by weight
Tough, Tough,
general-purpose general-purpose Concrete bonding
Components adhesive adhesive adhesive
Part A
Epoxy resin (Araldite 6020) 100 100
Epoxy resin (Epotuf 6140) 100
Part B
Polysulfide LP-3 50 50 50
Dimethyl aminopropylamine 10
(DMAPA)
DMP-30 6 7.5
Calcium carbonate 120
220 CHAPTER ELEVEN
TABLE 11.18 The Effect of LP-3 Polymer on Physical Properties of Liquid Epoxy Resin25
Parts by weight
Component A B C D E F G
Epoxy resin (EEW : 175–210) 100 100 100 100 100 100 100
Polysulfide LP-3 — 25 33 50 75 100 200
DMP-30 10 10 10 10 10 10 10
Property
Two-component, room temperature curing epoxy adhesives have evolved over three dis-
tinct generations, from brittle to flexible to toughened. The quest for toughening is due
to the inherent lack of peel and impact strength properties in brittle epoxy adhesives and
Parts by weight
Component A B C
the reduced cohesive strength and heat and chemical resistance in flexibilized epoxy adhe-
sives. The “toughened” epoxy adhesive claims to have the best of several worlds: improved
peel strength, greater resistance to impact and thermal shock, and the ability to absorb inter-
nal stresses in the joint without sacrificing high shear strength or temperature and chemical
resistance. Recent advances in toughened two-part epoxy adhesives have resulted in sys-
tems that can cure at room temperature in practical times.
The toughening mechanism of elastomer modified epoxy systems is different from that
of flexibilized epoxy systems and can be used in combination with them. Flexibilized
epoxy systems reduce mechanical damage by a reduction in modulus or plasticization of
the adhesive. This allows stress to be relieved through distortion of the adhesive, but it also
generally results in a lowering of the adhesive’s glass transition temperature with an
accompanying reduction in heat and chemical resistance.
By contrast, tougheners work by relieving stress through the prevention of crack growth
by incorporating elastomeric particles in an epoxy resin matrix. These adhesives generally
maintain a large percentage of the modulus and temperature resistance of the unmodified
resin system. The primary elastomers used in these toughened epoxy systems have been
functionalized butadiene acrylonitrile copolymers that are one of the following:
TABLE 11.20 Formulation for a Two-Part CTBN Modified Epoxy Adhesive System27
Part A
DGEBA epoxy resin (EPON 828, Resolution Performance Products) 80
CTBN modified epoxy resin (EPON 58034, Resolution Performance 20
Products)
Part B
Amidoamine (Epi-Cure 3072, Resolution Performance Products) 32
Property
ATBN liquid toughening agents are synthesized by reacting a CTBN polymer with an
amine. For example, n-aminoethylpiperazine (AEP) will give a low-molecular-mass
amine-terminated product at a reaction temperature of 130°C. The terminal secondary
amine groups undergo the typical reaction chemistry of secondary amines. Therefore,
Parts by weight
Component A B C
Part A
Epoxy acrylate resin (EPON 8121, Resolution 80 60 40
Performance Products)
DGEBA epoxy resin (EPON 828) 0 20 40
CTBN modified epoxy resin (EPON 58304) 20 20 20
Part B
Aliphatic amine (Epi-Cure 3273, Resolution 86 86 86
Performance Products)
Property
ATBN liquid polymers cannot be mixed directly into the epoxy resin component of a two-
part epoxy adhesive or in a one-part adhesive since crosslinking and shortened shelf life
will result. ATBN adducts are, therefore, mixed with the curing agent component of two-
component epoxy adhesives.30
ATBN tougheners increase the ultimate tensile shear and T-peel strengths and environ-
mental durability of epoxy resins with conventional room temperature curing agents such
as fatty polyamides, amidoamines, and amine ethers. However, the improvement in tough-
ness, as measured by peel tests, becomes less significant with the faster-curing hardeners.
Formulations for typical ATBN two-part epoxy adhesives are given in Tables 8.7 and 8.8.
There are several ways by which the formulator can moderately improve the heat or chem-
ical resistance of room temperature curing epoxy adhesives. Using an elevated-temperature
cure or a postcure will, of course, improve the temperature resistance by virtue of improved
crosslink density. However, this section describes formulations that have been developed
for moderately improved heat resistance after only a cure at room temperature. Optimal
(heat-curing) high-temperature and chemically resistant epoxy adhesives are discussed in
Chap. 15.
Generally aliphatic amines will provide greater heat resistance than polyamide or ami-
doamine curing agents. Higher-functionality epoxy resins such as epoxy-novolacs will also
provide better heat resistance. Table 11.22 provides a formulation for a room temperature
curing, two-component epoxy adhesive that maintains high shear strength when tested at
−55 to +150°C.
Cycloaliphatic amines can also provide high heat resistance when cured with a liquid
DGEBA epoxy at room temperature (although better high-temperature characteristics can
be achieved with an elevated-temperature cure). A range of cure schedules and working
Part A
Epoxy novolac resin (e.g., DEN 438, Dow) 40
Epoxy resin (EEW: 180–200) 40
Fibrous filler or colloidal silica 20
Part B
Triethylenetetramine (TETA) 12
Property
TABLE 11.23 H2SO4 Acid Resistance of Epoxy Adhesive Formulations Cured 1 Week at 25°C32
Parts by weight
Component A B C D
Part A
DGEBA epoxy resin (e.g., Epi-Rez 510) 100 100 100 100
Part B
Accelerated aliphatic amine (e.g., Epi-Cure 874) 18 18 18 —
Petroleum extender 73
Coal tar extender 81
Accelerated aromatic amine (e.g., Epi-Cure 8494) 45
Weight gain, %, after immersion in 33% 2.9 3.6 3.9 1.5
H2SO4 for 2 weeks at 65°C
4000
Aluminum
3000
Tensile shear strength, psi
2000
Copper
1000
0
0 6 12 18 24
Months of weathering (Louisville)
Adhesive formulation
FIGURE 11.6 Tensile shear strength retention of weathered aluminum and copper test
specimens.23
ROOM TEMPERATURE CURING EPOXY ADHESIVES 225
lives are possible with these curing agents. Cycloaliphatic amines provide good adhesion
and very good chemical resistance. These curing agents have been discussed in Sec. 5.2.1.
Generally, those factors tending to promote thermal stability also tend to improve chemical
resistance. A relatively high degree of chemical resistance can be obtained with aliphatic
amine or cycloaliphatic amine cured epoxy adhesives. Accelerated aliphatic polyamines
are frequently employed as curing agents in battery adhesive formulations. However, these
do not provide the higher degree of chemical resistance of aromatic amines.
Systems cured with DETA and TETA generally have excellent resistance to aqueous
sodium hydroxide even at 50% concentration and 80°C. Their resistance to 25 percent sul-
furic acid, 23 percent hydrochloric acid, and 25 percent chromic acid is also very good up to
80°C. Strong organic acids such as acetic acid generally attack DETA and TETA cured
epoxy resins as well as strong (40%) nitric acid. Resistance to intermittent solvent exposure
is excellent; however, long-term immersion in any aliphatic solvent such as kerosene is not
recommended.
Table 11.23 presents H2SO4 acid resistance data for epoxy adhesive formulations con-
sisting of accelerated aliphatic amines and accelerated aromatic amines. Generally a weight
gain of less than 4% is indicative of moderate chemical resistance. Aliphatic and aromatic
amines as well as polymercaptan curing agents are often used in bonding reinforced plas-
tic pipe in the chemical industry. The combination of a high degree of chemical resistance
along with rapid, on-site cure makes these adhesive systems especially valuable.
Although most epoxy adhesives have good weather resistance, optimum properties are
generally achieved when the adhesive has a combination of good water resistance and
thermal shock resistance. Figure 11.6 illustrates the retention of tensile shear strength of cop-
per and aluminum strips bonded with an amidoamine cured epoxy after 2 years of weathering
in a temperate climate.
Great caution must be exercised in exposing any adhesive joint to the simultaneous
effects of environment and stress. The stress can act to accelerate the degradation caused
by the environment, and vice versa. Joints that will be exposed to both high-humidity envi-
ronments and high load at the same time are especially vulnerable, and prototype speci-
mens need to be tested. This degradation mechanism and the performance of several epoxy
adhesive systems to combined environmental stress conditions are discussed in Chap. 15.
REFERENCES
1. Behm, D. T., and Gannon, J., “Epoxies,” in Adhesives and Sealants, vol. 3, Engineered Materials
Handbook, ASM International, Materials Park, OH, 1990, p. 101.
2. Behm, D. T., and Gannon, J.,“Epoxies,” p. 101.
3. Petronio, M. et al., “Glass to Metal Bonding,” Adhesives Age, January 1959.
4. Savia, M., “Epoxy Resin Adhesives,” Chapter 26 in Handbook of Adhesives, 2d ed., I. Skeist, ed.,
van Nostrand Reinhold, New York, 1977.
5. Lee, H., and Neville, K., Handbook of Epoxy Resins, McGraw-Hill, New York, 1967, p. 21.35.
6. Lee and Neville, Handbook of Epoxy Resins, p. 21.35.
7. Lee and Neville, Handbook of Epoxy Resins, p. 21.38.
8. Wiles, Q. T., U.S. Patents 2,528,934 and 2,528,933, 1950.
9. Lee and Neville, Handbook of Epoxy Resins, p. 21.38.
226 CHAPTER ELEVEN
10. Meath, A. R., “Epoxy Resin Adhesives,” Chapter 19 in Handbook of Adhesives, 3d ed., I. Skeist,
ed., van Nostrand Reinhold, New York, 1990, p. 356.
11. Lee and Neville, Handbook of Epoxy Resins, p. 21.39.
12. Bruins, P. F., Epoxy Resin Technology, Interscience Publishers, New York, 1968, p. 166.
13. Bruins, Epoxy Resin Technology, p. 168.
14. Savia, M., “Epoxy Resin Adhesives,” p. 443.
15. Bruins, Epoxy Resin Technology, p. 169.
16. Bruins, Epoxy Resin Technology, p. 171.
17. Resolution Performance Products, Starting Formulations 4000 and 4001, “Rapid Setting Room
Temperature Cure Adhesive,” Houston, TX, August 2003.
18. Air Products and Chemicals, Inc., “Flexible Adhesive,” Allentown, PA, 2001.
19. Bruins, Epoxy Resin Technology, p. 165.
20. Resolution Performance Products, Starting Formulation 4007, “High Peel Strength Adhesive,”
Houston, TX, August 2003.
21. Dannenberg, H., and May, C. A., “Epoxide Adhesives,” Chapter 2 in Treatise on Adhesion and
Adhesives, R. L. Patrick, ed., Marcel Dekker, New York, 1969.
22. Resolution Performance Products, Starting Formulation 4026, “Metal Adhesive,” Houston, TX,
August 2003.
23. Meath, “Epoxy Resin Adhesives,” p. 356.
24. Amstock, J. S., Handbook of Adhesives and Sealants in Construction, McGraw-Hill, New York,
2002, p. 2.11.
25. Panek, J. R., “Polysulfide Sealants and Adhesives,” in Handbook of Adhesives, 3d ed., I. Skeist,
ed., van Nostrand Reinhold, New York, 1992.
26. Peterson, E. A., “Polysulfides,” in Adhesives and Sealants, vol. 3, Engineered Materials Handbook,
AMS International, Materials Park, OH, 1990.
27. Resolution Performance Products, Starting Formulation 4018, “High Strength Room Temperature
or Heat Cure Adhesive,” Houston, Tx, August 2003.
28. Farris, R. D., and Steward, S. L., “New Epoxy Tougheners Widen the Adhesive Formulation
Window,” Adhesives and Sealants Industry, January 2002.
29. Resolution Performance Products, Starting Formulation 4008, “Adhesive for Thermoplastic
Substrate,” Houston, TX, August 2003.
30. Petrie, E. M., “Improving the Toughness of Structural Adhesives,” at www.SpecialChem4Adhesives.
com, April 7, 2004.
31. Savia, “Epoxy Resin Adhesives,” p. 443.
32. Bruins, Epoxy Resin Technology, p. 162
33. Bruins, Epoxy Resin Technology, p. 161.
CHAPTER 12
ELEVATED-TEMPERATURE
CURING LIQUID AND PASTE
EPOXY ADHESIVES
12.1 INTRODUCTION
Although room temperature curing epoxy adhesives are very convenient to use in many
applications, the short working life, long curing time, and inherently suboptimal physical
properties of these systems are often a serious disadvantage to their use. In these cases,
elevated-temperature curing epoxy adhesives are required. The major advantages and dis-
advantages of elevated-temperature curing epoxy adhesives are shown in Table 12.1.
Heat curing epoxy systems may consist of one or two components. The nomenclature
is similar to that employed with the room temperature curing adhesives (e.g., resin compo-
nent, curing agent component, 2K system, etc.). Many heat curing epoxy adhesives systems
are liquids or pastes. However, heat curing systems also can be processed into solid adhe-
sive forms (e.g., powders and films). This chapter describes the paste and liquid formula-
tions; solid adhesive forms are considered separately in Chap. 13.
Two-component, elevated-temperature curing epoxy adhesives, once mixed, are designed
to cure only at elevated temperatures. However, these systems generally will have a finite
working life at room temperature, although it is much longer than that of room temperature
curing epoxy adhesives. Generally, a minimum temperature of around 80°C is required to
fully cure these systems. Various cure temperatures and cure schedules are often provided
to offer the user flexibility in processing.
One-component (or “1K”), elevated-temperature curing epoxy adhesives have the cat-
alyst or curing agent incorporated directly with the epoxy resin. Thus, no metering or mix-
ing is required. However, since the curative is integrated with the resin, these systems have
limited shelf life, and special storage conditions such as refrigeration may be required.
Elevated-temperature curing epoxy adhesives can be desirable when the short working life
of a room temperature curing adhesive cannot be tolerated. One-component epoxy adhe-
sives are advantageous when it is necessary to eliminate measuring or mixing errors or
when dispensing equipment cannot handle multicomponent systems.
Several important factors must be considered relative to the “elevated temperatures”
that are required to cure these adhesives. The curing temperature and time specified repre-
sent the temperature and time that the adhesive film actually experiences. For example, if
a cure of 60 min at 150°C is recommended, this does not mean that the assembly should be
simply placed in a 150°C oven for 60 min. The temperature is to be measured at the adhe-
sive within the joint. A large assembly will act as a heat sink and may require substantial
time for an adhesive within the joint to reach the necessary temperature. In this example,
227
Copyright © 2006 by The McGraw-Hill Companies, Inc. Click here for terms of use.
228 CHAPTER TWELVE
Advantages Disadvantages
Fast cure time results in higher production Greater internal stresses than in room temperature
speed and less tie-up of fixturing equipment curing adhesive because of
• Higher shrinkage on polymerization
• Thermal expansion coefficient differences
Long working life More brittle than room temperature curing
adhesives (poorer peel and impact properties)
High-temperature resistance Safety and hazardous nature of high temperatures
Good chemical resistance One-component systems have shorter shelf life;
may require refrigeration
One-component systems possible Solid systems easier to apply, less of a health
(no need to meter and mix) hazard than liquids or pastes
Long shelf life for two-component systems Energy consumption
Solid systems (film, preforms, powders) Viscosity of adhesive decrease at elevated
possible temperature could result in starved joint
Viscosity decrease at elevated temperature
provides more efficient wetting of substrate
the total oven time would be 60 min in addition to the time that it takes to bring the adhe-
sive film up to 150°C.
Bond line temperatures are best measured by thermocouples placed very close to the
adhesive. In some cases it may be desirable to place the thermocouple directly in the adhe-
sive for the first few assemblies being cured.
Oven heating is the most common source of heat for curing epoxy adhesives that require
an elevated-temperature cure. Good air circulation within the oven is mandatory for uni-
form heating. Temperature distribution within an oven should always be checked before
items are placed in the oven. Many ovens will have significant temperature distribution
variation and dead spaces, especially in corners where air circulation is not uniform. The
number of items and their arrangement in the oven also affect the time for the joint to reach
the proper temperature. The geometry and size of the part may affect air circulation and
cause variations in temperature distribution.
Heated-platen presses are also commonly used for bonding flat or moderately contoured
panels. Autoclaves or vacuum bagging is used for large parts such as in the aerospace indus-
try. In addition to direct heating, epoxy adhesives can be cured by induction and dielectric
heating as well as other indirect heating sources. Formulations specifically developed for
indirect heating are discussed in Chap. 14. Chapter 17 provides additional information regard-
ing the processing and equipment used to cure epoxy adhesive at elevated temperature.
Regardless of the heating source, elevated-temperature curing epoxy adhesives neces-
sitate that the parts be held with clamps, pressure pads, or fixturing methods to apply con-
stant and uniform pressure throughout the curing cycle. Reduction in viscosity due to
elevated temperatures could result in excessive adhesive flow during cure. This could result
in joints that are starved of adhesive or movement of the parts, if not firmly fixtured.
Although room temperature curing epoxy adhesive can also be cured at elevated tem-
peratures, there are important differences in morphology and internal stresses between a
room temperature cured adhesive and a heat cured epoxy adhesive (see Chap. 3). One
should not assume that the properties would be equivalent or better if the adhesive were
cured at elevated temperature rather than room temperature. A number of factors need to
be considered in such a deliberation.
ELEVATED-TEMPERATURE CURING EPOXY ADHESIVES 229
Whereas most room temperature curing epoxy adhesives are cured with aliphatic amines,
polyamides, or amidoamines, most elevated-temperature curing epoxy adhesives are cured
with aromatic amines, modified aliphatic amines, alcoholic and phenolic hydroxyls, acid
anhydrides, Lewis acids, and a host of other curatives. Latent curing agents, such as dicyan-
diamide and imidazoles, are typically used in one-component epoxy adhesives systems.
Part A
Liquid epoxy resin (e.g., EPON 828) 100
Epoxy resin (e.g., EPON 1009) 10
Epoxidized soybean oil 12
Thixotrope As required
Part B
Diethylaminopropylamine (DEAPA) 11
Property
is added to the more conventional liquid epoxy resin to improve high-temperature perfor-
mance. The epoxidized soybean oil is used to reduce viscosity.
Cycloaliphatic amines, such as aminoethylpiperizine (AEP), isophoronediamine (IPDA),
and bis(p-aminocycloethyl)methane (PACM), provide longer pot life than the aliphatic amines
along with greater temperature resistance and toughness when cured. AEP is used to provide
higher impact strength and toughness than aliphatic amines; however, it will not cure at room
temperature in thin sections. IPDA and PACM have been considered as alternative to aromatic
amines. They provide shorter pot life than the aromatic amines, but they are colorless and non-
staining to the skin. IPDA and PACM also provide a high heat distortion temperature.
General-purpose rigid and flexible two-component, elevated-temperature cure epoxy
adhesives can easily be formulated from cycloaliphatic amine (rigid) or polyamide (flexi-
ble) curing agents. Table 12.3 describes two such formulations. Note that the flexible adhe-
sive system exhibits a T-peel strength on aluminum of 14 lb/in compared to only 4 lb/in for
the rigid formulation. However, the rigid formulation has a higher tensile shear strength
and better heat resistance.
Temperature-resistant two-part, elevated-temperature curing epoxy adhesives can be for-
mulated with aromatic amines, such as metaphenylenediamine (MPDA), methylene dianiline
(MDA), or a eutectic blend of the two. These adhesives will provide relatively high tempera-
ture strength, but they are generally brittle. When mixed with epoxy resin at concentrations of
about 15 pph for MPDA and 26 pph for MDA, they provide complete cure in about 30 min at
175°C. The aromatic amines also provide a working life of several hours at room temperature.
Starting formulations for aromatic amine cured epoxy adhesives are shown in Table 12.4.
MDA cured epoxy adhesive will provide shear strength over 4 times that of DETA at
temperatures of 100 to 150°C. Table 12.5 indicates the high-temperature performance that
Parts by weight
Part A
DGEBA epoxy resin (EEW: 190) 68 50.5
Diluent (Epodil 748) 7
Talc (Microtuff 325F) 29.5 40
Fumed silica 2.5 2.5
Part B
Ancamine 2264 (Air Products and Chemicals Inc.) 40
Ancamide 910 (Air Products and Chemicals Inc.) 38
Ancamide 2482 (Air Products and Chemicals Inc.) 15
Aluminum (Toyal 101) 24 20
Talc (Microtuff 325F) 24 25
Fumed silica 2 2
Property
TABLE 12.4 Starting Formulations for Epoxy Adhesives Cured with MPDA2
Part A
Epoxy resin (EPON 828) 100
Allyl glycidyl ether 2
Polyvinyl acetate 6
Thixotrope As required
Part B
Metaphenylenediamine (MPDA) 15.2
Property
can be achieved with several aromatic amines. The aromatic amines, however, generally
provide poorer bond strength at lower temperature due to their relatively brittle nature and
high degree of shrinkage on polymerization. Aromatic amines can be used in one-component
epoxy adhesives although their shelf life is relatively short when compared to that of the
more commonly used latent curatives.
Low-melting-point blends of aromatic amines (e.g., Shell Chemical Curing Agent Z or
Celanese Epi-Cure 841) can be used to formulate two-component epoxies, which cure satis-
factorily at temperatures below 100°C. These adhesives are often used to bond heat-sensitive
TABLE 12.5 Effect of Aromatic Amines on Tensile Shear Strength of DGEBA Adhesive
Supported on Glass Cloth Carrier3
adherends. The aromatic amines also provide B-stage cure characteristics that prove useful
in formulating solid adhesive forms.
Diaminodiphenyl sulfone (DADPS) cured epoxy adhesive also exhibits very good
elevated-temperature properties. With this curing agent and resin blends of DGEBA epoxy
and polyglycidyl of tetraphenolethane, adhesive strengths of greater than 1000 psi have
been realized on aluminium substrates after aging for 400 h at 260°C.4 The DADPS is gen-
erally mixed with the epoxy resin at a concentration of about 30 pph. Another formulation
consisting of 100 parts EPON 1031 (a tetrafunctional solid epoxy resin), 100 parts alu-
minum powder, and 30 parts DADPS was found to yield bond strength in excess of 1000 psi
when cured 30 min at 175°C and tested at 260°C.5
Anhydride cured epoxy adhesives are sometimes used where high-temperature perfor-
mance or chemical resistance is desired. These curing agents are not widely used in adhe-
sive formulations, however, because the adhesive properties are inferior to those realized
by the amine curing agents and because anhydrides require rather long cure cycles.
Anhydride curing agents are used where exceptionally long working life is required. The
exotherm that is generated with epoxy-anhydride systems is low, and the reaction rate is
slow. Elevated-temperature cures (greater than 150°C) are necessary, and long postcures
are required to develop ultimate properties. Tertiary amines are the most widely used accel-
erator for this system. Anhydride cured epoxy resins exhibit very low viscosities.
The physical and electrical characteristics of the anhydride cured systems are very good
over a wide temperature range. Compared to amines, anhydride cured epoxies exhibit better
chemical resistance to aqueous acids, but less chemical resistance to some other reagents.
When epoxy resins are cured with anhydrides, the product has relatively high heat distor-
tion temperature and low moisture sensitivity.
Phthalic anhydride (PA) and its derivatives are among the most common anhydride cur-
ing agents. Table 12.6 describes an anhydride cured epoxy adhesive that is accelerated with
Parts by weight
Component A B
Part A
Epoxy resin (EPON 828) 100 12
Condensation product of DGEBA epoxy 88
and ethylene glycol
Thixotrope As required As required
Part B
Phthalic anyhydride 43 43
Phthalic acid 20
Benzyldimethylamine 0.8
Pyromellitic dianhydride 2.5
Pyridine 0.3
Property
a tertiary amine. This formulation cures in 2 h at 150°C. At higher anhydride contents, lower
room temperature strength results owing to a reduced hydroxyl content in the cured resin.6
Other anhydrides such as dodecyl succinic anhydride (DDSA) or adducts of DDSA with
polyglycols, can also be used for formulating heat cured epoxy adhesives. These have
excellent electrical properties and good thermal shock resistance. Anhydride cured epoxies
are also useful for bonding plastics, notably polyester such as Mylar.8
In one-component epoxy adhesives, the curing agent and resin are compounded together as
a single product by the adhesive formulator. The curing agent system is chosen so that it
does not react with the resin until the appropriate processing conditions are applied.
One-component epoxy curing agents, such as dicyandiamide, BF3 complexes, aromatic
amines, and imidazoles, are commonly used since these compounds are stable at room tem-
perature but cure when heated past an activation temperature. All these curing agents are
slow-reacting curatives that provide a relatively long shelf life when mixed into the epoxy
resin. However, the greatest success has been realized by a group of curatives known as
latent curing agents.
Latent curing agents work by several mechanisms. The most common is due to the curing
agents’ insolubility in the epoxy resin and, therefore, its inactivity at room temperature. When
heated, the curative melts and dissolves into the resin thereby activating the curing process.
These types of curing agents are best represented by dicyandiamide and several other cura-
tives. They represent the most common curatives used in single-component epoxy adhesives.
A second method of achieving latency is by using a curative precursor, which is chemi-
cally inactive at room temperature but then converts to an active curative at the cure tem-
perature. Examples of this type include Monuron and di-urea adduct of toluene diisocyanate
and dimethylamine. Full cure in about 1 h at 125°C is achievable with these materials.
Dicyandiamide (sometimes referred to as “dicy”) or its derivatives are used in most
commercial one-component epoxy adhesives. This curative is a white crystalline solid and
is easy to incorporate into an epoxy formulation as a finely ground powder. When cured
with epoxy resin, dicyandiamide provides an excellent set of performance properties.
Dicyandiamide has an activation temperature of about 177°C. However, this can be
reduced to about 120°C by the use of accelerators. When mixed into liquid DGEBA resins,
dicyandiamide has very good shelf life due to its latency. The shelf life of unaccelerated
systems is about 6 months at room temperature or longer if refrigerated. The shelf life is
reduced significantly with the addition of accelerators.
Dicyandiamide cured epoxies have been known to exhibit high strength and chemical
resistance; however, they are relatively brittle. Through the utilization of toughening
agents, very resilient and tough adhesives can be formulated without sacrificing any of the
inherently good properties of the unmodified system. With toughened dicyandiamide cured
epoxies, peel strengths on the order of 30 lb/in have been achieved along with tensile shear
strength in the range of 3000 to 4500 psi. The toughened dicyandiamide cured epoxy adhe-
sive also shows very good resistance to thermocycling.
A typical dicyandiamide cured epoxy adhesive formulation is described in Table 12.7.
These formulations are unaccelerated so that they have a shelf life of greater than 6 months
at room temperatures. The adhesives will cure in 30 to 90 min at 149 to 177°C. Tensile
shear strength on aluminum is in the range of 2500 to 3000 psi.
234 CHAPTER TWELVE
Parts by weight
Component A B C D
Accelerators for dicyandiamide cured epoxy adhesive formulations include tertiary amines,
modified aliphatic amines, imidiazoles, and substituted ureas. All except the substituted
ureas can cure epoxy resins by themselves. All these materials provide good latency and
excellent adhesive applications.
Probably the most effective accelerator for dicyandiamide systems is the substituted
ureas because of their synergistic contribution to the performance properties of the adhe-
sive and their exceptionally good latency. It has been shown that adding 10 pph of a sub-
stituted urea to 10 pph of dicyandiamide will produce an adhesive system for liquid
DGEBA epoxy resins that can cure in only 90 min at 110°C. Yet this adhesive will exhibit
a shelf life of 3 to 6 weeks at room temperature. Cures can be achieved at temperature even
down to 85°C if longer cure times are acceptable.10
Table 12.8 shows the effect of three commercially available substituted ureas on shelf
life, cure rate, exotherm, and glass transition temperature of a dicyandiamide cured epoxy
adhesive. The accelerators are compared at use levels of 1, 3, and 5 pph in a one-component
adhesive consisting of 100 pph of DGEBA epoxy, 8 pph of dicyandiamide, and 3 pph of
ELEVATED-TEMPERATURE CURING EPOXY ADHESIVES 235
Control 0
U-52 (CVC Specialties)* 1 3 5
U-405 (CVC Specialties)† 1 3 5
Diuron‡ 1 3
Property
fumed silica. Usable shelf life is generally taken as the time for a twofold increase in vis-
cosity to occur.
Table 12.9 shows a formulation for an accelerated general-purpose one-component,
dicyandiamide cured epoxy adhesive compared to one with a modified aliphatic amine cur-
ing agent. Notice that the dicyandiamide cured system provides a higher glass transition tem-
TABLE 12.9 Formulations for a Dicyandiamide and Modified Aliphatic Amine Cured
Epoxy Adhesive
Parts by weight
Component A B
TABLE 12.10 Starting Formulation for an Epoxy Adhesive Cured with BF3 Amine Catalyst12
perature by virtue of its greater crosslinking density. The dicyandiamide also has greater than
3 months’ shelf stability even though it is catalyzed. The accelerated dicyandiamide cured
epoxy adhesive formulation is typical of the general-purpose one-component epoxy adhe-
sive products that are commercially available today from many adhesive suppliers.
Boron trifluoride monoethylene amine complex (BF3-MEA) is a Lewis acid that has
also been used as a latent curing agent. It is an adduct of boron trifluoride and diethylamine
existing as a solid that melts at 80 to 85°C. When mixed with a DGEBA epoxy, the for-
mulation exhibits exceptionally long shelf life of 6 months to 1 year at room temperature.
There is no significant cure that takes place until the curing temperature exceeds 100 to
125°C. The cured adhesive exhibits good physical properties when tested in the 150 to
175°C temperature range; however, the peel and impact properties are somewhat poorer
than those of dicyandiamide cured epoxies.
The BF3-MEA complex offers a slightly faster rate of cure and a reduced shelf live in
liquid epoxy systems when compared to unaccelerated dicyandiamide cured epoxy formu-
lations. However, when used as a sole curing agent, BF3-MEA has not had the commercial
success of dicyandiamide because of their lower bond strength and brittleness. The BF3-
MEA complex compounds also hydrolyze in the presence of moisture, so that mixtures
with epoxy resins must be stored in tightly closed containers to maintain shelf life.
BF3-MEA cured epoxy adhesives are very economical to produce since only a small
amount of catalyst is required. These adhesives have good elevated service temperature.
Table 12.10 is a starting formulation for a one-component epoxy adhesive cured with a BF3
MEA catalyst. It has a shelf life of approximately 4 months at room temperature.
Imidazoles, such as 2-ethyl-4-methyl imidazole (EMI), represent another family of
latent curing agents that can be used alone or as an accelerator for dicyandiamide. When
EMI is mixed with DGEBA epoxy resins, the shelf life is greater than 6 months at 40°C,
yet it will gel in a very short time (minutes) at temperature from 120 to 170°C. Cure of this
adhesive at 120°C for 30 min gives strong water-resistant bonds to stainless steel. The
cured adhesive exhibits a high degree of thermal and chemical resistance.
Several novel methods have been developed over the years to provide one-component systems
with long shelf life and good performance properties when cured at a moderate time and low
temperatures. These include frozen, microencapsulated, and molecular sieve catalyzed systems.
One method of formulating a single-component epoxy system is to blend a liquid epoxy
resin with a conventional amine curing agent and then package and immediately flash-
freeze the product to stop further reaction. In essence, the user then has a one-component
adhesive that does not require metering or mixing, but which must be kept frozen until use.
The adhesive formulations can be such that they cure at either room or slightly elevated
temperatures. While offering some of the conveniences of a one-component adhesive, these
types of adhesive systems do not provide the performance properties of conventional
elevated-temperature cure adhesive. They are expensive because of the added processing
steps and the cost of packaging in small containers to avoid waste.
Amines have also been microencapsulated within small cellulosic or polyelectrolyte
capsules. This is a method for keeping the amine separate from the epoxy resin during stor-
age. When the user decides to initiate cure, the capsules are broken, usually in the applica-
tion process, and the amine is free to react with the epoxy resin. A successful example of
this type of product is an epoxy adhesive that can be preapplied to machine screw threads.
When the screw is ultimately threaded into place, the shearing action causes the capsule to
break. Bond strengths are generally low for this type of adhesive, but this may not be impor-
tant in certain applications.
Molecular sieves such as zeolite have also been used as carriers for low-molecular-
weight amines. An amine, such as DETA, can be absorbed into the sieve prior to mixing
with an epoxy resin. The sieve protects the amine from reacting with the epoxy resin, and
a relatively long shelf life is possible. Release of the DETA is then initiated by heat or
through displacement by atmospheric moisture.
Epoxy-novolac resins and highly functional DGEBA epoxy resins have been used in combi-
nation with anhydride, phenolic, or aromatic amine curing agents to produce moderately high-
temperature-resistant adhesives. Although useful tensile shear strength at temperatures as high
as 260°C can be achieved for short periods of time (hundreds of hours), longer-term service
temperatures are limited by thermal degradation processes to 150 to 175°C. Figure 12.1 shows
tensile shear strength at test temperature for various different formulations.
A general-purpose two-component adhesive that will provide high tensile shear strength
up to 150°C is described in Table 12.11. The base epoxy resin in this formulation is a mix-
ture of an epoxy novolac and a liquid DGEBA epoxy resin.
Benzoquinone tetracarboxylic acid dianhydride (BTDA) has been found to provide epoxy
adhesives with excellent high-temperature properties, in both the short and long terms. The
formulation described in Table 12.12 provides good resistance to 260°C. This two-part
adhesive can be cured 2 h at 200°C. The disadvantage of BTDA is that relatively high cure
temperatures are required that result in a high degree of internal stress within the bond line.
238 CHAPTER TWELVE
4000
B
2000 E
D
C C
1000 D
0
0 100 200 300 400 500
Test temperature, °F
Adhesive formulations
Parts by weight
Component A B C D E
Other curing agents that are known to provide moderately good heat resistance are pen-
tamethyldiethylenetriamine (PMDA) and short-chain acid anhydrides, such as methyl
nadic anhydride. Imidazoles, such as 2-ethyl-4-methyl imidazole, and certain aromatic
amines, particularly diaminodiphenyl sulfone (DADPS), provide very good high-temperature
properties in two-component epoxy systems, as noted above. Thermal stability is particu-
larly enhanced when these curing agents are used with epoxy resins having a higher func-
tionality than the DGEBA types. The most common epoxy resins used in high-temperature
formulation, therefore, are epoxy novolacs or tetrafunctional solid epoxy.
ELEVATED-TEMPERATURE CURING EPOXY ADHESIVES 239
Part A
Epoxy novolac (DEN 438, Dow) 90
Epoxy resin (DER 736, Dow) 10
Atomized aluminum powder 40
Part B
Cycloaliphatic amine 28
PMDA or trimellitic anhydride has also been shown to provide epoxy adhesive formu-
lations with high-temperature properties. Table 12.13 shows the elevated-temperature tensile
shear strength of an epoxy adhesive cured with 4 pph of PMDA. Another specialized
formulation employing PMDA was found to provide high shear strength when tested at
260°C even after aging 1000 h at 260°C.15
The main difficulty lies in incorporating curing agents such as PMDA into the epoxy resin
and then providing for resiliency when cured. One formulation by DuPont16 was developed
for improved toughness. It requires reacting 1 mol of dialcohol with 2 mol of PMDA to yield
a more readily fusible and soluble product, which maintains two anhydride groups.
12.5.2 Toughening
For many years, the typical method of improving the toughness of high-temperature
structural adhesives was to add elastomeric resins to rigid high-temperature base polymer
to create a hybrid product such as epoxy-nitrile. However, the toughening of high-temperature
adhesives can provide a difficult challenge, since the service temperatures usually exceed
the degradation point of most rubber additives. Also, the addition of an elastomer generally
resulted in lowering of the glass transition temperature of the base polymer.
−58 3240
26 3110
121 2950
149 1040
204 570
260 380
However, newer adhesives systems having moderate temperature resistance have been
developed with improved toughness but without sacrificing other properties. When cured,
these structural adhesives have discrete elastomeric particles embedded in the matrix.
The most common toughened hybrids using this concept are acrylic and epoxy systems.
The elastomer is generally a amine- or carboxyl-terminated acrylonitrile butadiene copoly-
mer (ATBN and CTBN).
Formulations have been developed in which small rubber domains of a definite size and
shape are formed in situ during cure of the epoxy matrix. The domains cease growing at
gelation. After cure is complete, the adhesive consists of an epoxy matrix with embedded
glass rubber particles. The formation of a disperse phase depends on a delicate balance
between the miscibility of the rubber with the resin, with the resin-hardener mixture, and
appropriate precipitation during the crosslinking reactions.
Table 12.14 shows a comparison of dicyandiamide cured epoxy adhesives formulated
with and without a CTBN adduct. When compared to the control epoxy, the toughened for-
mulation exhibits significantly higher peel strength and moderately higher tensile shear
strength. CTBN modified epoxy adhesives are generally one-part systems, cured with
dicyandiamide at elevated temperature.
Parts by weight
ATBN tougheners are generally used in room temperature formulations (see Chap. 11).
ATBN liquid polymers cannot be mixed directly into the epoxy resin component of a two-
part adhesive or in a one-part adhesive, since crosslinking and shortened shelf life will result.
ATBN adducts are, therefore, mixed with the curing agent component of two-component
epoxy adhesives.
Within the past several years, improvements in the toughening of high-temperature
epoxies and other reactive thermosets, such as cyanate esters and bismaleimides, have been
accomplished through the incorporation of engineering thermoplastics. Additions of
poly(arylene ether ketone) or PEK and poly(aryl ether sulfone) or PES have been found to
improve fracture toughness. Direct addition of these thermoplastics generally improves frac-
ture toughness but results in decreased tensile properties and reduced chemical resistance.
REFERENCES
13.1 INTRODUCTION
Epoxy adhesives are most commonly used as liquids or pastes. However, certain types of
epoxies can be employed in the form of a solid. The components in these adhesives are
mixed and processed to a stage where the resulting adhesive product is in a solid but still
fusible (uncrosslinked) state. When the applied solid adhesive is heated, it melts, flows, and
wets the substrate. Additional heating time then causes the adhesive to cure completely into
a strong, thermosetting structure.
Solid epoxy adhesives can be formulated in various ways. The more common methods
are described below.
1. Latent curing agents such as dicyandiamide are dissolved into solvent solutions of solid
epoxy resins. This is then followed by evaporation of the solvent.
2. Soluble curing agents are added into liquid epoxy resins and cured until a B-stage con-
dition is reached. The B stage is a solid, thermoplastic stage. When given additional heat,
the B-stage epoxy will flow and continue to cure to a crosslinked condition or C stage.
3. Reactive powdered resins and curing agents can be combined by dry blending. These
powder blends can then be preformed into various shapes by dry compression pressing.
The main mechanism in all these methods is the physical separation and restriction of
molecular mobility of the epoxy resin and the curing agent that are imposed by the solid
state of the product. These adhesive systems generally provide a shelf life of up to 6 months
at room temperature depending on the reactivity of the curing agent and resin. All these
products require elevated temperatures to liquefy and crosslink.
The most widely recognized types of solid epoxy adhesives are tapes or films that are
commonly used to bond large substrates such as honeycomb skins and structural paneling.
These are generally manufactured using the first or second technique described above.
Solvent solutions that are used to manufacture tape or film can also be applied directly to
the substrate. Once the solvent evaporates, not only is the adhesive an integral part of the
substrate, but also it protects the surface during storage and handling.
Although tapes, films, or coatings applied from solvent solution make up the majority
of the solid epoxy adhesives, formulations in the form of a powder or shaped solids can also
be employed in certain applications. These are manufactured using the second or third tech-
nique described above.
The main advantages of solid adhesives are that they are single-component (i.e., meter-
ing and mixing are not required) and that they can be applied uniformly to a substrate with
243
Copyright © 2006 by The McGraw-Hill Companies, Inc. Click here for terms of use.
244 CHAPTER THIRTEEN
TABLE 13.1 Advantages and Disadvantages of Solid Epoxy Adhesives over Liquids and Pastes
Advantages Disadvantages
Elimination of plural component dispensing and metering More expensive by weight than liquid or
equipment and associated labor and spare parts paste adhesives
Reduction of labor required to oversee, maintain, and Not effective in filling a deep volume
clean equipment and containers in contact with liquids
or pastes
Improved raw materials procurement and flow by Not cost-effective generally for small-
replacing two containers associated with dual- volume production processes
component adhesive with a single-component system
Elimination of pot life consideration and line downtime A more limited number of formulations
resulting from adhesive advancement available than with liquids or pastes
Elimination of mix ratio tolerance concerns with Parts that must generally be designed to
resulting improvement in product consistency hold the melt before it cures to a solid
Elimination of variations from the coating process with Difficulties that can occur in application
resulting improvement in product consistency of solids to vertical or contoured surfaces
Improved worker safety and reduced hazardous Heat required to make the solid epoxy
materials exposure to liquid and paste systems adhesives flow and cure
little or no waste. Being solid, these adhesives also avoid the mess, cleanup requirements,
and health hazards often associated with liquid adhesives. As a result, solid epoxy adhe-
sives have gained a certain degree of popularity in several high-volume production
applications including the assembly of electronic, automotive, and aerospace components.
Table 13.1 summarizes the advantages and disadvantages of using solid epoxy adhesives
over more conventional liquid or paste forms.
A significant advantage of tape and film adhesives is the greater toughness that is
available compared to other adhesive types. This is primarily due to the ease with which
resinous modifiers can be added to the formulation via solvent solution. Thus, hybrid
epoxy adhesives such as epoxy-nylon, epoxy-phenolic, etc., are often found in tape or
film form.
Solid epoxy adhesive formulations can be processed to either a thermoplastic or a ther-
moset state. Solid epoxy resins of exceptionally high molecular weight (e.g., phenoxy)
can be used without any degree of cure as a hot-melt type of adhesive. However, fully
crosslinked, thermoset systems are generally employed in structural applications.
Single-component, solid epoxy adhesives are made in several ways. Generally all methods
consist of completely formulating the adhesive system, including resins, fillers, curing
agents, etc., in the liquid state and then converting it to a thermoplastic solid. This conver-
sion can be done through the cooling of a melt, or by the removal of a solvent from a solu-
tion, or by partial curing—a process sometimes known as B-staging.
A B-stage resin is one in which a limited reaction between resin and curing agent is
allowed to take place. The reaction is arrested while the product is still fusible and soluble,
SOLID EPOXY ADHESIVE SYSTEMS 245
although having a higher softening point and melt viscosity than originally. The B-
stageable formulation contains sufficient curing agent to effect crosslinking on subsequent
heating.
For tape or film adhesives, the curing agent is usually incorporated into either a liquid
epoxy resin or a high-molecular-weight solid epoxy resin solution. The hardening process
occurs by B-staging (in the case when a liquid epoxy resin is used), solvent evaporation, or
both. The hardening process is usually accompanied by extruding, calendering, or casting
the adhesive into thin films that are typically 5 to 15 mils thick.
The product is processed to a condition where it is either tack-free or slightly tacky to
aid in application to vertical or contoured substrates. These films may be in the form of
unsupported sheet, or they may be reinforced with glass fabric, paper, or another reinforc-
ing medium. These supported films are sometimes referred to as prepreg.
For powder or preformed adhesives, the incorporation of curing agent can be done via
various processes.
1. When high-molecular-weight solid epoxy resins are employed, they may be reduced to
powder and then combined with a powdered curing agent.
2. Alternatively, the high-molecular-weight solid epoxy may be melted, the curing agent
added, and the mixture cooled and pulverized.
3. With liquid epoxy resin, the procedure is to react the resin with the curing agent and
advance the mixture to a B stage.
Solid adhesive made by the third process can be (1) cast directly into a solid usable
shape or, more commonly, (2) cast and then ground to a powder form. This grinding oper-
ation is conducted with standard pulverizing equipment. With lower-melting-point blends,
cooling may be required to prevent softening and blocking of the resin during the pulver-
izing operation.
The powder can be applied directly to the substrate surface by electrostatic coating
processes or “dusting” onto a warmed surface. It can also be formed into shapes or preforms
by the application of pressure in a die mold. This process is similar to how pharmaceutical
tablets are made. In this way shaped preforms can be made that will conform to a specific
joint geometry. There are a wide variety of applications to which epoxy preforms can be
adapted. They can be used to
The incorporation of particulate fillers into the liquid epoxy formulation can be
achieved with any of the processes mentioned above by the use of roll mills or screw-type
kneaders. If the formulation is solid, fillers may be blended into the product by the use of a
pebble mill or the like. After blending, they may be rolled under pressure and then ground
if desired.
In general, solid epoxy adhesives will have a somewhat limited shelf life, and to ensure
reproducible wetting and flow, it is advisable that they be stored in a refrigerated condition
until their use. Depending on the chemistry and nature of the curing agents that are used,
these storage conditions could require temperatures of 5°C or less. With the more reactive
systems, flow will decrease on aging even with refrigeration. Thus, the storage life of the
product must be rigidly controlled and verified.
246 CHAPTER THIRTEEN
13.3 CHEMISTRY
Solid epoxy adhesives generally rely on high-molecular-weight epoxy resin for the solid
appearance of the uncured adhesive. This epoxy resin is generally formulated with
either:
1. A latent curing agent
2. A curing agent capable of developing a B stage
The cure system must be slow at room temperature to prevent storage life and flow prob-
lems, but sufficiently rapid at elevated temperatures to permit reasonably short curing
times. Even the most latent curing system in use today does not completely eliminate room
temperature reaction; thus shelf life and storage requirements must be critically controlled.
A wide range of epoxy resins as well as a wide range of curing agents and catalysts are
available for formulating solid epoxy adhesives. Resins with different viscosities, amounts
of reactive groups, and structures are available. Additives that change the uncured resin vis-
cosity, reduce brittleness, or impart some other property are also available.
Epoxy resins with aromatic backbones and high functionality give a strong, high-
temperature, highly crosslinked matrix. However, the resulting adhesive is usually brittle.
Aliphatic epoxies with lower functionality usually result in matrices with higher elongation
and toughness but lower temperature capability.
Latent curing agents, such as dicyandiamide or 4, 4′-diaminodiphenyl sulfone (DADPS)
are commonly used in producing solid epoxy adhesives. Being latent, these curing agents
react with epoxies only on heating. They are relatively insoluble in the epoxy resin at room
temperature, yet melt and become soluble at an activation temperature. Once activated,
they cure relatively quickly at elevated temperatures. Both dicyandiamide and DADPS pro-
vide excellent high-temperature properties.
Dicyandiamide is a true latent catalyst for epoxy resin curing. It is also considered to be
the workhorse of one-component adhesives due to its ease of use, excellent performance
properties, long shelf stability, and low toxicity. In certain admixtures with DGEBA, it has
demonstrated a room temperature storage life in excess of 4 years. Dicyandiamide is usu-
ally added to the solid epoxy resin in concentrations of about 3 to 6 pph. It melts at about
150°C. Cures can be conducted in the range of 120 to 175°C but are very slow at the lower
temperatures. As a result, it is common practice to add accelerators such as benzyl-
dimethylamine (BDMA) and mono- or dichlorophenyl substituted ureas to these systems.
DADPS also provides excellent high-temperature properties and chemical resistance. Of
the amine curing agents, DADPS provides the best retention of strength after prolonged expo-
sure to elevated temperatures. It melts at 135°C and can be cured with epoxy resins at 20 to
30 pph with cure temperatures ranging from 115 to 150°C. Because of the low reactivity of this
system, an accelerator, such as BF3-MEA, is usually employed at about 1 pph.
Other latent curing agents that are used in solid adhesives are dihydrazides and BF3-
MEA complexes. These compositions are also stable at room temperature but cure when
heated. Solid anhydrides can be used in one-component powder blends (e.g., 10 pph of
trimellitic anhydride accelerated with 0.5 pph of 2-methylimidazole). Solid systems with
aromatic diamines are prepared by comelting the solid epoxy with the amine. Typically 30 pph
of curing agent is used.1
For B-stageable adhesives, aromatic amines are generally used in liquid or solid epoxy
resins. They provide excellent chemical resistance and electrical properties. However,
when fully cured, they are rather rigid. The rigidity results in high tensile strength but poor
toughness, peel strength, and impact properties. The aromatic amines are blended with liquid
epoxy resin at 10 to 30 pph. A B-stage will occur in several hours at room temperature or
SOLID EPOXY ADHESIVE SYSTEMS 247
faster with mild heating. Crosslinking requires a sustained heat cure at higher temperatures.
The temperature range for full cure is 60 to 200°C depending on the type of resin and cur-
ing agent.
Metaphenylenediamine (MPDA) is the best known of these aromatic amines. Methylene
dianiline (MDA) requires somewhat longer cures and has a higher processing viscosity.
Both products are solids and generally must be melted before being blended with resin. The
difficulty in handling these materials as hot melts has led to the development of aromatic
amine eutectics, which are liquid at room temperature.
There are several common forms of solid epoxy adhesives. These include film, tape, pow-
der, and preformed shapes. Certain formulations are better suited for specific forms. For
example, casting of tape or film adhesive from solvent solutions lends itself to working
with multicomponent hybrid systems, where each resin can be solubilized and blended
together in a universal solvent. B-staged systems are generally more brittle and better suited
for powders or preformed adhesives.
Tape and film are the most common forms of solid epoxy adhesives. Tape and film are terms
that are used rather loosely for adhesives in a thin sheet form. The term tape generally refers
to an adhesive that is supported on a web of paper or nonwoven fabric, or on an open-weave
scrim of glass, cotton, or nylon. Films, on the other hand, are free of supporting material and
consist only of the adhesive. Both tape and film products may be accompanied with a release
liner depending on the tack or blocking characteristics of the adhesive.
Supporting fibers in epoxy adhesive tapes are useful in that they provide for a positive
stop under bonding pressure. This can be used to control bond line thickness and to help
distribute stresses evenly during service. The supporting fibers that are used in these adhe-
sives are primarily for the purposes of carrying the adhesive and convenient application to
the substrate. Their reinforcing function within the epoxy matrix is generally considered to
be of secondary importance.
The final thickness of epoxy tape or film adhesives is on the order of 5 to 15 mils. These
adhesives may be soft and tacky, or stiff and dry, depending on their formulation. The soft
and tacky products are valuable in products requiring application to contoured or vertical
surfaces. If the product is especially tacky, a release liner (e.g., polyethylene film, coated
paper) is generally used to keep the film from blocking. The stiff and dry products are gen-
erally used for flat surfaces where speed and ease of application are required.
The main attribute of tape or film adhesives is that they are single-component systems
requiring no need for metering or mixing. They can be easily cut to size and applied
between the substrates to be joined without waste or mess. Although tape and film adhe-
sives offer a uniformly thick adhesive product that is easily dispensed, they are poor gap
fillers, especially if the gap between the mating parts varies significantly across the bond-
ing surface. To provide better gap-filling characteristics, epoxy films are sometimes for-
mulated with chemical blowing agents that are activated during the high-temperature cure.
The resulting foamed epoxy expands to 2 to 3 times its original thickness to fill relatively
large gaps. The slight degree of foaming has minimal effect on the performance properties
of the cured adhesive.
248 CHAPTER THIRTEEN
Once the tape or film is in place between the substrates, the joint is heated under pres-
sure so that the adhesive becomes slightly fluid, flows into the microroughness on the sub-
strate, and wets the substrate. With additional time at the curing temperature, the adhesive
completely crosslinks to a thermosetting condition.
Tape and film adhesives are most often used to bond large areas, such as for applications
in the aerospace industry. For example, the joining of aluminum honeycomb structure to flat
metal skins is often accomplished with thermosetting epoxy film adhesives. These films
(Fig. 13.1) can easily be applied without the need to mix, meter, or apply a liquid coating.
Typically tape or film epoxy adhesives are modified with synthetic thermoplastic poly-
mers to improve flexibility in the uncured film and toughness in the cured adhesive. Epoxy
resins can also be blended with phenolic resins for higher heat resistance. The most com-
mon hybrid systems include epoxy-phenolics, epoxy-nylon, epoxy-nitrile, and epoxy-vinyl
hybrids. These hybrid film adhesives are summarized in Table 13.2, and structural proper-
ties are shown in Table 13.3.
Tape and film adhesives dominate the markets where the assembly of large parts is
required. They are also used in applications where greater reliability and consistency are
desired. Tape and film adhesive eliminate any variation in concentration that may occur
from metering and mixing operations. Also, the incorporation of a carrier, such as glass or
polyester fabric, into the adhesive acts to control the bond line thickness, thus avoiding thin,
adhesive-starved areas where part curvature or external pressure may be greatest. Although
these adhesives were originally developed for the aerospace markets, they are now finding
applications in automotive, building, and general industrial areas.
Guidelines for formulating solid epoxy adhesives are similar to those employed for liq-
uid or paste adhesives. Table 13.4 shows starting formulations for several epoxy adhesive
tapes and films.
Epoxy-phenolic adhesive was the first true high-temperature adhesive. It was developed
in the early 1950s as a high-temperature aircraft adhesive. An example formulation is pro-
vided in Table 13.4, but these adhesives are discussed predominantly in Chap. 15.
Epoxy-nylon adhesives were developed in the 1960s for high-peel-strength applica-
tions. The key to their development is the use of noncrystalline nylons that are soluble in
alcohols and other epoxy-compatible solvents. A commercial example of adhesive-grade
nylon includes DuPont’s Zytel 61. Standard nylons are not practical because of their
incompatibility with most other resins.
TABLE 13.2 Common Epoxy Hybrid Resins Used in the Formulation of Tape and Film Adhesives
Nylon-epoxy Films are blends containing 30 to 50% by weight of epoxy resin. The nylon
constituent provides high tensile shear strength as well as high peel strength.
Suitable catalysts are dicyandiamide and aromatic polyamines. These adhesives
have useful properties at low temperatures but have only moderately good high-
temperature properties.
Epoxy-phenolic Epoxy-phenolic adhesives are generally used in aerospace applications
requiring high shear strength at temperatures in excess of 150°C. Usually the
phenolics are a resole type, and often the epoxy is a minor component. These
adhesives are relatively brittle and have low peel and impact strengths.
Nitrile-epoxy Nitrile-epoxy adhesives are composed of solid epoxy resin modified with
carboxyl-terminated butadiene nitrile (CTBN) copolymer. The CTBN is
introduced into the epoxy resin at elevated temperatures. The modification
provides toughness and high peel strength without sacrificing heat and
chemical resistance. The film adhesives are widely used in the aerospace
industry in the construction of jetliners.
Vinyl-epoxy Vinyl-epoxy adhesives have moderate strength at temperatures up to 150°C.
Oxidation stability is excellent. Vinyl constituents increase toughness and
peel strength. The adhesive is often used for bonding safety glass, aerospace
engine structures, and structural panels.
TABLE 13.4 Starting Formulations for Epoxy Tape and Film Adhesives3
Tape adhesive is prepared by impregnating the above liquid adhesive into a glass mat and procuring
at 133°C for 35 min. The adhesive can withstand temperatures up to 287°C.
Vinyl-epoxy tape adhesive Parts by weight
The adhesive is deposited on a glass cloth and dried at 95°C for 15 min. Cure is 30 min at 240°C.
Resulting shear strength on steel is 1090 psi at 287°C after 400 h at 260°C.
Nylon-epoxy adhesive film Parts by weight
Epoxy resin 25
Nylon resin 75
Dicyandiamide 10
Cure conditions are 60 min at 163°C. Resulting adhesive will have a peel strength of 60 to 80 lb/in
of width and a shear strength of 6200 to 6600 psi at room temperature.
General-purpose epoxy adhesive tape Parts by weight
Mix resins and solvent. Ball mill curing agent into varnish. Varnish is then B-staged on desired film
backing and heat cured in place.
Unsupported general-purpose epoxy adhesive film Parts by weight
Solvent removal occurs at 83°C for 1.0 h. Dry film thickness is 0.002 in. Cure schedule is 1 h at
175°C and 120 psi.
SOLID EPOXY ADHESIVE SYSTEMS 251
epoxy-nitrile adhesives is that they require relatively low cure temperatures in the range of
125°C.
The toughness of an epoxy-nitrile adhesive is nearly equivalent to that of an epoxy-
nylon adhesive. However, the epoxy-nitrile system has much better hydrolytic stability.
Also, the low-temperature properties of an epoxy-nitrile adhesive are superior to those of
epoxy-nylon adhesive. Table 7.3 illustrates the effect of nitrile addition on tensile shear and
peel strength.
Another form of solid epoxy adhesive is powder or granules, which must be first heated to
be made liquid and capable of flowing. Being a single-component system, powdered adhe-
sives also eliminate proportioning and mixing errors. However, uniform distribution of the
powder over large areas is sometimes difficult to achieve, especially if the surface is verti-
cal or contoured. Powdered epoxy adhesives can be sifted, dusted, or sprayed onto a sub-
strate surface, but special application processes, such as electrostatic powder coating, are
often required for uniformity. Fluidized-bed coating processes have been used for applying
adhesive coatings to magnetic wire and motor lamination stacks. The fundamentals of the
powder coating process can be found in the Web site www.SpecialChem4Coatings.com in
the Powder Coating Center.4
Solid epoxy adhesives may also be fused or compacted into various shapes or preforms
(see Fig. 13.2) such as sticks, rings, and beads.5 They can then be easily applied to a specif-
ically shaped part. Similar to thermoplastic hot-melt adhesives, these thermosetting epoxy
adhesives flow on the substrate by heating. However, the product will cure to an infusible
thermoset stage with continued heating. Once cured, the product demonstrates properties
similar to those of other structural epoxy adhesives. These solid epoxy adhesives are some-
times referred to as thermosetting hot melts.
Certain types of thermosetting epoxy powders can be fused into small rings that are then
used in rod-and-tube types of joint design. The preformed epoxy ring can be slipped over
one of the substrates before assembly. When exposed to high curing temperatures, the
epoxy in the preform melts, flows, and wets the substrate. Additional heating fully cures
the adhesive. In this type of application, a depression or trough must generally be designed
into the joint to hold the liquid adhesive in place while it cures.
Preformed sticks or tubes of solid epoxy may also be applied much as a solder is.6 To
apply, the area to be bonded is heated until sufficiently hot to cause the stick solder to flow
when drawn across the surface. After application, the adhesive is cured by conventional
thermal techniques. If two substrates are to be bonded, both must be preheated prior to
application of the epoxy solder.
Solid shapes or powders are formulated from solid or liquid (when used as a B stage)
epoxy resins and curing agent. Fillers, additives, and other modifiers are often used as
they are with liquid or paste epoxy adhesive formulations. However, consideration must
be given to the flow properties of the adhesive when heated as well as the application
properties.
Guidelines for formulating solid epoxy adhesives are similar to those employed for
liquid or paste adhesives. Table 13.5 shows starting formulations for a shaped solid and
powder epoxy adhesive.
The stick solder formulation (Table 13.5) can be made in several ways. The components
can be mixed as a hot melt and cast into small-diameter tubes. The components can also be
ground to 60-mesh powder and then compressed into void-free rods. To apply, the area to
be bonded is heated so that when the stick solder comes into contact with the surface, it
melts and wets the substrate. An elevated-temperature cure is then required to crosslink the
adhesive.
TABLE 13.5 Starting Formulations for Epoxy Powder and Stick Adhesives7
Application can be by fluidized bed or electrostatic spray. Cure conditions are 3 min at 204°C or
12 min at 149°C. Shear strength on steel is 3300 psi at room temperature.
Stick solder adhesive Parts by weight
The ingredients are prepared as a hot melt and cast into small-diameter tubes or are ground to
60-mesh powders and then compressed into void-free sticks.
SOLID EPOXY ADHESIVE SYSTEMS 253
are readily soluble in ketones, esters, and mixtures of aromatic hydrocarbons, chlorinated
hydrocarbons, or alcohols with ketones.
The toughness of the resin (90 percent ultimate elongation and 9000 psi tensile strength)
provides for a high degree of tensile shear strength and peel strength at room temperature.
However, because the adhesive is not crosslinked, the elevated-temperature properties and
creep resistance are poor. The heat deflection temperature of phenoxy is about 90°C.
Applications for such adhesives include metal foil lamination, coextrusion of plastic film,
and other applications where there is a need for fast processing and where elevated-
temperature properties or solvent resistance is unimportant.
The phenoxy resins can form strong bonds to metals in seconds at temperatures from
315 to 340°C. At these bonding conditions, a tensile lap shear strength of 3500 to 4000 psi
can be obtained on aluminum. The same bond can be made in 2 to 3 min at 260°C and in
30 min at 190°C due to the thermoplastic phenoxy’s time-temperature creep characteristics
and relatively low softening temperature (100°C).8
The phenoxy resins can also be plasticized with many of the common plasticizers and
still maintain a high percentage of strength, while the bonding temperature is substantially
reduced. An effective hot-melt adhesive is one in which polyvinyl ether is blended into the
phenoxy resin.
Table 13.6 shows the characteristics and properties of supported and nonsupported high-
molecular-weight epoxy resins cast from solvent solution. A crosslinked and uncrosslinked
formulation using the same high-molecular-weight epoxy is shown for comparison.
Parts by weight
Component A B C D E
Epoxy resin, mol. wt. >20,000 100 100 374 374 374
(e.g., Epi-Rez 2287)
Methyl ethyl ketone 11 11 36 36 36
Epoxy novolac resin 15 15 45
(e.g., Epi-Rez 5155)
Dicyandiamide 1 1 2
Dimethylforamide 10 10 10
Benzyldimethylamine 0.05 0.05 0.05
Carrier, cloth None Glass None Glass None
Solvent removal 1 h at 83°C
Dry film thickness, in 0.002 0.003 0.002 0.003 0.002
Cure schedule 5 min at 170°C 1 h at 170°C
Pressure 115 psi
Property
REFERENCES
1. Savia, M., “Epoxy Resin Adhesives,” Chapter 26 in Handbook of Adhesives, 2d ed., I. Skeist, ed.,
van Nostrand Reinhold Co., New York, 1977.
2. Politi, R. E., “Structural Adhesives in the Aerospace Industry,” Chapter 44 in Handbook of
Adhesives, 3d ed., I. Skeist, ed., van Nostrand Reinhold Co., New York, 1993.
3. Savia, “Epoxy Resin Adhesives,” and Politi, “Structural Adhesives in the Aerospace Industry.”
4. “Powder Coating Center,” www.SpecialChem4Coatings.com, 2004.
5. Harvill, K., “Finding the Cure for Epoxy Dispensing Frustrations,” ECN Magazine, Dec. 15, 2001.
6. Paul, M. N., Process for Preparing Polyepoxide Solder, U.S. Patent 2,965,930, 1960.
7. Savia, “Epoxy Resin Adhesives,” and Politi, “Structural Adhesives in the Aerospace Industry.”
8. Anderson, C. C., “Adhesives,” Industrial and Engineering Chemistry, vol. 59, no. 8, August 1967,
pp. 91–94.
9. Shimp, D. A., “Epoxy Adhesives,” in Epoxy Resin Technology, P. F. Bruins, ed., Interscience
Publishers, New York, 1968, p. 175.
CHAPTER 14
UNCONVENTIONAL EPOXY
ADHESIVES
14.1 INTRODUCTION
This chapter considers epoxy adhesive formulations that are rather unconventional. These
include
• Epoxy adhesives that cure via radiation (ultraviolet light or electron beam energy)
• Epoxy adhesives that are available as waterborne emulsions
• Epoxy adhesives that cure via indirect thermal means (induction, dielectric, microwave,
and several others)
These adhesives have found their way into several niche markets where their advantages,
such as fast setting speed, are highly valued. They have not seen application in the more
ordinary markets because of their materials and equipment cost. However, it is expected
that these epoxy adhesive systems will grow at a faster annual rate than the average market
as their advantages become more widely known.
255
Copyright © 2006 by The McGraw-Hill Companies, Inc. Click here for terms of use.
256 CHAPTER FOURTEEN
One of the most popular uses of radiant curing is the advancement (viscosity increase) or
crosslinking of pressure-sensitive adhesives. These applications have been satisfied mostly
with acrylate-based adhesive systems. With epoxy-based adhesives, the main applications
are electrical and electronic components, the bonding of large aerospace structures such as
composites, and the bonding of transparent substrates such as glass and plastic.
Radiation cured epoxy adhesives significantly reduce the long cure times needed for
conventional adhesives. These adhesives develop excellent physical and mechanical prop-
erties in a nonthermal cure process that requires a cycle time of only seconds to minutes,
versus hours for conventional thermal curing.
Radiant cured epoxy systems also provide high glass transition temperatures, low
shrinkage during cure, and relatively low residual stresses. They have no volatile emissions
and excellent shelf life. The good physical and adhesive properties of conventional epoxy
adhesives systems are maintained with the radiant cured systems.
Radiation curing permits the use of low-viscosity, solvent-free adhesive and coating com-
positions. Generally, a radiant curable low-viscosity liquid resin is coated on a substrate by
conventional techniques. Viscosity increase, crosslinking, and viscoelastic properties are
then developed quickly by exposing the moving substrate to a radiant energy source, as
illustrated in Fig. 14.1. Radiant cured pressure-sensitive adhesives have replaced solvent-
based pressure-sensitive adhesives in significant quantities.
The primary advantages of radiant curing are (1) regulatory compliance through the vir-
tual elimination of volatile organic components (VOCs) and (2) fast production speeds. Major
processing advantages of radiant curing adhesives include the absence of heat, water, or sol-
vents. Performance advantages include superior heat and chemical resistance, excellent clar-
ity, and high shear strength when compared to many waterborne or solvent based adhesives.
Some of the benefits of using radiation curable adhesives are shown in Table 14.1.
The primary disadvantages of radiant curing adhesives include relatively high material
cost due to the fact that these compositions generally have no low-cost components such as
solvents, extenders, or fillers. However, radiant cured adhesives are often justified on a total-
cost basis when one is considering energy bills, reduced waste, labor cost, production time,
and factory space availability.
Adhesive
dispenser
UV
lamp
Coating
station
Unwind Winding
station station
No VOC emissions More than 100% increase in Temperature, solvent, and moisture-
line speed; web line speeds sensitive substrates that can be
of 1000 ft/min common coated
No regulatory reporting or Energy needed to cure 1 g Control of cure: degree of
permitting requirements of pressure-sensitive crosslinking that can be varied on
adhesive about 1% that of the same product
waterborne and 4% that of
solvent adhesives
Low flammability Solvent cost savings Multilayer composites that can be
constructed
Floor space savings Easy to screen-print
Savings in insurance costs Improved adhesion to substrates due
to grafting
Reduced overall costs Ease of coating due to lower viscosity
Product immediately ready Radiation curing used to crosslink hot
for testing and shipment melts as well as increase molecular
weight of liquids
Single-component system UV lamps that can usually be
installed on existing production lines
Another distinct disadvantage of radiation cured adhesives is that at least one substrate
must be transparent to the radiant energy. The radiant energy must also penetrate the depths
of the adhesive for cure to be initiated. Complex joint geometries, thick bond lines, and
shadowed areas protected from the radiant energy create problems. However, modern for-
mulations have been specifically created to reduce these disadvantages, and partial cure by
radiant energy followed by a complete cure by thermal energy is always an option.
In 1995, U.S. consumption of radiation cured products was 77 million lb, valued at $450
million.1 Growth is forecast to average about 7 percent per year—a rate about twice that of
conventional thermal cured products. Although radiation curable adhesives comprise a rel-
atively small segment of the overall adhesive market (13 percent) and epoxy adhesives rep-
resent an even smaller component, epoxy systems are a fast-growing part of the market.
Market penetration is expected to increase further due to stricter environmental regula-
tions and the availability of a greater variety of products. The first radiation curable adhe-
sives were limited to acrylate and epoxy resins. Today, many different types of radiation
curable adhesive systems are commercially available.
The radiation curing industry is often represented by RadTech International (www.
radtech.org). RadTech is a nonprofit organization that promotes the use of radiation curing.
RadTech’s goal is to provide the industry with a better understanding of the science, materi-
als, processing equipment, and applications in hopes of spreading the use of radiation curing.
The organization serves as an international forum and reliable source of education and infor-
mation for individuals and organizations involved with uv and EB processing.
Radiation is used to crosslink (Fig. 14.2) or cure organic resins into durable coatings or
adhesives having excellent physical properties with high chemical and temperature resis-
tance. Radiation curing technology involves at least four considerations: type of radiation
258 CHAPTER FOURTEEN
Energy from uv or EB
radiation source
Oligomeric
chains Polymeric
network
case of EB cure, the high-energy electrons interact directly with the atoms of the unsatu-
rated site to generate highly reactive molecules.
Many potential radiant curable epoxy systems exist, but only two have found significant
commercial use in adhesives:
The chemistry and curing mechanisms of these systems are outlined in the following sec-
tions. Further information can be found in several textbooks on radiant cured adhesives.2,3,4
A significant amount of information can also be acquired through RadTech, a professional
society focusing on radiation curable products.5
UV Cure Mechanisms. There are two fundamental mechanisms for the uv curing of coat-
ings: free radical and cationic. The more common free radical mechanism consists of a
chain reaction with four primary steps:
The reaction can be illustrated by using a photoinitiator (I) and reactive monomers (R, R′),
as shown in Fig. 14.3. Creating free radicals for uv curing requires the use of photoinitia-
tors, which decompose on exposure to uv to produce initiating free radicals, which start the
chain reaction. Quenching or deactivation of the photoactivated initiator by oxygen can
occur, and the growing polymer radicals can also react with oxygen. This oxygen inhibi-
tion causes short polymer chains to form, resulting in tacky surfaces and poor physical
properties of the coating. Fortunately with most initiator systems, the propagation rates are
high, and oxygen quenching and competing reactions are minimized.
Cationic cure mechanisms are an alternative approach to uv curing. This involves the
photogeneration of ions, which initiate ionic polymerization. This process is not subject to
oxygen inhibition, as are some of the free radical mechanisms. Cationic cure mechanisms
generally also provide less shrinkage and improved adhesion. The disadvantages are that
the photoinitiators are sensitive to moisture and other basic materials. The acidic species
can also promote corrosion. As a result, the vast majority of uv formulations are acrylate-
based and cure by a free radical mechanism.
The energy sources of uv light cured coatings are typically medium-pressure mer-
cury lamps, electrodeless vapor lamps, pulsed xenon lamps, or lasers. These generally emit
electromagnetic radiation in the region of 200 to 760 nm. They also produce a certain degree
of infrared radiation as heat, and this acts either to anneal the cured coating and relieve inter-
nal stresses and strains or to enhance cure rate in cationic cure systems. Pigmentation blocks
or diffuses the uv radiation. Thus uv coatings that are cured with these light sources are usu-
ally clear or transparent, although thin, opaque coatings are also possible.
Electron Beam Cure Mechanisms. Electron beam curing mechanisms are similar to the
uv free radical mechanism. However, the electrons are accelerated to a much higher energy
state, and the electron itself has sufficient energy to initiate polymerization. The impact of
these electrons is high enough to break chemical bonds and to generate ions. The ions then
transform themselves into free radicals, which then initiate polymerization. Thus, the EB
mechanism requires no photoinitiator. There is also greater penetration of the radiation
(greater depth of cure) with less interference from pigmentation. Clear coatings of up to 20 mils
and pigmented coatings of about 15 mils can be cured with EB equipment.
The absence of photoactive fragments in electron beam curable adhesives results in
greater stability of both the cured and uncured adhesives. However, these advantages are
offset by the relatively higher capital costs of EB curing equipment. The higher capital cost
has generally restricted the use of electron beam curing to high-volume operations. EB cur-
able adhesives are mainly used for laminating applications, such as in the production of
metallized papers and plastic or foil laminates, whereas uv curable adhesives have found
their way into a diversified range of industrial and commercial applications.
High-energy electron accelerators are used to cure EB adhesives. They can generate the
radiant energy (150 to 500 kV) capable of curing thicker, pigmented resins, as EB energy
has greater ability than uv energy to penetrate the material. The electron source is a fila-
ment that is heated inside a vacuum tube.
Free Radical Cure UV Adhesives. As with any adhesive, formulation variables are crit-
ical to the processing and performance characteristics. Variables such as oligomer selec-
tion, modifiers and additives, monomer structure, molecular weight, and glass transition
temperature directly affect application and performance properties.
Practical radiation curable materials are composed of more than a single reactive
monomer. Commercial adhesives normally contain the following:
Examples of typical uv and EB coating compositions and their applications can be found
in several sources.6,7,8
UNCONVENTIONAL EPOXY ADHESIVES 261
able to initiate crosslinking. EB accelerators belong to a fairly wide range of chemicals, but
generally they are quite polar in nature.
Other additives used to improve the performance of radiant cured adhesives are similar
to those that might be found in more conventional adhesives. These include adhesion pro-
moters, fillers, light stabilizers, antioxidants, and plasticizers.
Photoinitiators are perhaps the most important component in uv cured radiation coat-
ings. The photoinitiator is an ingredient that absorbs light and is responsible for the pro-
duction of free radicals in a free radical polymerized system or cations in a cationic
photoinitiated system. The photoinitiators are usually added to the reactive coating formu-
lations in concentration ranges from less than 1 to 20 percent by weight based on the total
formulation. The absorption bands of the photoinitiators should overlap the emission spec-
tra of the various commercial light sources.
There are two general classes of photoinitiators: (1) those that undergo direct photofrag-
mentation on exposure to uv or visible light irradiation and produce active free radical inter-
mediates and (2) those that undergo electron transfer followed by proton transfer to form a
free radical species. The choice of photoinitiator is determined by the radiation source, the
film thickness, the pigmentation, and the types of base resin employed. Examples of typical
photoinitiator systems used to cure reactive resins are shown in Table 14.2. Benzophenone
is perhaps one of the most common photoinitiators.
The only significant difference between uv cured adhesives and visible light (sunlight)
cured adhesives is the use of a photoinitiator that absorbs the appropriate wavelength of
light energy. Sunlight contains about 5 percent uv light. Adhesives developed to cure in
sunlight generally do not cure when exposed solely to incandescent or fluorescent light,
which contains only about 0.1 percent uv. Photoinitiators having absorption capabilities in the
visible light energy range are based on dyes, quinines, diketones, and heterocyclic chemical
structures.
Epoxy resins and other polymers (tetrahydrofuran, vinyl ethers, styrene, etc.) can be
cured when exposed to an acid or cation intermediate species. The photoactive catalyst sys-
tem commonly used to cure epoxy resins and multifunctional vinyl ether materials is com-
posed of salts of aryldiazonium, triarylsulfonium, and diaryliodonium. These systems are
commonly employed in coatings and adhesives for electronic products. The acid initiator
generated from the photoinitiator continues to be active even after uv curing, and so con-
version of reactants and crosslinking continue even in the absence of uv light. This phe-
nomenon is typically referred to as dark cure.
Visible light cured epoxy adhesives and coatings have been developed for architectural,
industrial, and maintenance applications and for products difficult to heat or uv/EB cure
because of their size. These are clear, one-part epoxy resins that cure by exposure to visi-
ble light for a few hours. They are formulated with cycloaliphatic epoxy compounds and a
cationic photoinitiator that generates a strong acid when exposed to sunlight.
Some of the uv curable adhesives contain a combination of uv and infrared (ir) initia-
tors to take advantage of the ir output that many uv lamps generate. At times a photoac-
tive crosslinking agent is used to improve cohesive strength without affecting tack and
peel.
Photoinitiators, sensitizers, and other radiation-sensitive agents may have an effect on
the adhesive properties and especially on processing of the adhesive. Thus, all additives
need to be tested with regard to storage and processing properties as well as with regard to
their adhesive properties. The photoinitiator package will also need to be optimized for a
given adhesive thickness and uv dosage.
Stabilizers are common additives included in uv cured adhesives to prevent premature
polymerization, resulting in viscosity increase, gelation in storage, or premature curing that
occurs due to low levels of light exposure. These compounds act as scavengers for free rad-
icals. They neutralize the free radicals before they have a chance to start the chain reaction
leading to polymerization. Light stabilizers, such as hindered amine light stabilizing
(HALS) compounds, also protect the cured film from exposure to direct sunlight. These
products are used to provide coatings with good outdoor stability, color retention, and
nonyellowing properties.
Oxygen scavengers may be required as oxygen inhibits the curing of acrylates by
quenching the photoinitiator or by scavenging free radicals. Scavenging produces stable
species that slow down the cure rate but also can degrade the properties of the cured coat-
ing. Other methods of oxygen inhibition are nitrogen blanketing (a process called inerting)
or speeding up the cure by using higher-intensity lamps and by varying the initiator type
and concentration. The specific formulation determines the propensity for oxygen inhibi-
tion and the degree of sophistication needed to counteract it.
Adhesion promoters are used to provide good adhesion to such substrates as glass, hard
plastics, and certain metals such as brass. These are specialized materials (e.g., organosi-
lanes) that have the ability to promote adhesion to a substrate material and at the same time
form a part of the polymer network.
Pigments generally inhibit uv curing to some degree since the pigments absorb and/or
scatter uv radiation. This interferes with the ability of the photoinitiator to absorb the light
energy required to initiate the polymerization reactions. Thus, the majority of commercial
radiation curable adhesives are clear or contain silica.
Other additives used to improve the processing and performance of radiation cured
adhesives are similar to those that might be found in more conventional formulations.
These include antioxidants, defoamers, flow and wetting agents, and slip aids.
Cationic Cure Epoxies. As with free radical uv chemistry, the same principles of formu-
lation component selection apply to cationic cured uv epoxy adhesives. Although different
monomers and oligomers are normally required for this type of chemistry, the main differ-
ence lies with the photoinitiator system.
A typical cationic uv adhesive formulation contains an epoxy resin, a cure-accelerating
resin, a diluent (which may or may not be reactive), and a photoinitiator. The initiation step
results in the formation of a positively charged center through which an addition polymer-
ization reaction occurs. There is no inherent termination, which may allow a significant
postcure. Once the reaction is started, it continues until all the epoxy chemistry is consumed
and complete cure of the resin has been achieved. Thus, these systems have been termed
living polymers.
264 CHAPTER FOURTEEN
The cationic photoinitiator has relatively limited absorbance spectra, and the cure is more
sluggish than a uv cured epoxy-acrylate. As a result, the uv may be used to develop handling
strength in the epoxy adhesive, and a room temperature or elevated-temperature postcure may
be necessary after the initial uv exposure for the adhesive to achieve ultimate properties.
Cationic photoinitiators are frequently found in classes of compounds such as the triaryl
sulfonium, tetraaryl phosphonium, and diaryliodonium salts of large protected anions
(hexafluorophosphates or antimonates). These compounds are soluble in most epoxy
resins, do not activate epoxy cure until exposed to uv light, are insensitive to room light-
ing, and have long storage life at room temperature. Cationic photoinitiators form an acid
catalyst when exposed to uv light and consequently start the cationic chemical reaction.
The cationic photoinitiators are sensitive to moisture, and the acid species formed can
promote corrosion. Frequently the formulation for these curing methods contains solvent,
which evaporates during curing, so that an important environmental advantage is lost. For
these reasons cationic uv cure is usually preferred over free radical cure only when the
higher-performance properties are justifiable.
Recently photoinitiators have been developed that can trigger photocuring when
exposed to visible light.9 As an industrial process, cure with visible light offers the advan-
tage of increased safety because it poses less of an eye hazard than uv light. Visible light
curing adhesives also provide for better curing in thick films and in shadowed areas. The
visible light photoinitiators include 2,4-diiodo-6-butyoxy fluorine (DIBF). Usually a sec-
ondary photoinitiator such as (4-octyloxyphenyl)phenyliodonium hexafluoroantimonate10
is required to trigger the cationic cure.
Cationic cured epoxies may also be crosslinked by electron beam radiation. A major
application for this technology is the repair of composite aerospace structures. Direct ben-
efits of EB processing include rapid cure, allowing completion of a permanent repair in the
same or less time than a traditionally temporary repair, and ease of material handling. Other
benefits include improved process control, reduced scrap due to process time limitations,
reduced inventory of various repair materials, reduced spares inventory, reduced tooling
costs, improved repair reliability, increased facility production, and the ability to do certain
repairs not currently possible using existing repair techniques.
Recently, Acsion Industries and UCB Chemicals have undertaken a development
program of cationic initiated epoxy adhesive for aluminum-to-aluminum, aluminum-to-
composite, and composite-to-composite bonding. This has led to a series of bonding adhe-
sives that have tensile shear strengths of 13 to 17 MPa on fiberglass-to-fiberglass substrates
and 35 to 52 MPa on aluminum-to-aluminum substrates.11
In a recent study, the EB curing of epoxy resins by cationic polymerization was investigated
to determine if cured materials with superior mechanical properties, high glass transition tem-
perature, and shorter cure times could be produced.12 Diaryliodonium salts were found to be
an effective initiator for the cationic polymerization of epoxy resins when a high-energy,
power electron beam accelerator was used as the source of ionizing radiation. For example,
Dow Tactix 123 (a bisphenol A epoxy) containing 3 pph (4-octyloxyphenyl)phenyliodonium
hexafluoroantimonate was irradiated at a total dosage of 100 kGy. Glass transition tempera-
ture of the cured material as determined by dynamic mechanical analysis was 182°C as com-
pared to 165°C for the thermally cured material.
Waterborne epoxy dispersions have been employed effectively for many years in the coat-
ings market primarily for the surface protection of concrete and metals. These products
were developed in response to environmental regulations to reduce solvent levels in coatings.
UNCONVENTIONAL EPOXY ADHESIVES 265
Since these dispersions adhere well to a wide variety of substrates and provide a high
degree of strength and resistance to service environments, they have also naturally found
applications as adhesives.
Epoxy dispersions also can easily be blended with other waterborne polymers to make
modified latex adhesives. The resulting hybrid adhesive produces performance properties
and application characteristics that are superior to those of the originating latex system.
This section reviews the chemistry behind waterborne epoxy adhesives and the formu-
lation possibilities. The characteristics of epoxy dispersions and the performance proper-
ties of cured adhesive films are addressed. The advantages and disadvantages of these
adhesive systems are discussed with the focus on determining whether waterborne epoxy
systems can replace traditional epoxy adhesives.
Over the last several decades, environmental restrictions on solvents used in adhesives and
coatings have become increasingly stringent. Hazardous air pollutant (HAP) limitations and
the phaseout of ozone-depleting substances are only two examples of regulatory require-
ments that are leading to increased interest in alternatives to solvent-based systems.
Another deciding factor in the development of waterborne epoxy coatings was the escalat-
ing cost of organic solvents in the mid-1970s.
Many of the attributes of solvent-borne epoxy coatings could be carried over to the
waterborne epoxy coatings. These same attributes are useful in the application of water-
borne epoxies as adhesive systems. They include good adhesion to a variety of substrates
such as metals, wood, concrete, glass, ceramics, and many plastics; chemical resistance;
low shrinkage; toughness and flexibility; and abrasion resistance.
In addition to the excellent performance properties and the reduction of solvent carriers,
waterborne epoxy adhesives were found to have processing advantages. They could be eas-
ily applied by conventional coating systems (spray, roller, etc.); they were less hazardous
to workers due to lower dermatitis potential and inflammability; ventilation equipment
costs could be reduced; and application equipment could be easily cleaned with soap and
water. In many applications, these processing advantages became the primary market drivers
for waterborne epoxy adhesives as alternatives to more conventional adhesives.
However, waterborne epoxy systems are not without certain disadvantages, which have
limited their application as adhesives. These disadvantages include increased use of energy
to evaporate the water and dry the adhesive, lower resistance of the cured film to high-
humidity environments, and storage and application limitations due to potential freezing at
low temperatures.
There are many applications for polymeric waterborne adhesives. These include pack-
aging adhesives, pressure-sensitive tape, coatings for textiles, wood adhesives, and various
industrial adhesives and coatings. The potential applications for waterborne epoxy adhe-
sives are more limited due to their lack of tack and pressure-sensitive characteristics and
the time it takes for the chemical reaction to complete cure. However, waterborne epoxy
systems have found significant markets in niche areas.
Waterborne epoxy coatings and adhesives have established the building and construc-
tion industry as their largest market. Commercial systems have been available for many
years. The following characteristics propel their use over conventional alternatives:
Typical applications in the building and construction industry that benefit from these
characteristics include
• Use as a waterproof coating for inside or outside tanks, basements, and utility substations
• Repair of spalled concrete by use as an adhesive and protective coating
• Waterborne high- and medium-gloss wall and floor coatings to resist high-pressure
cleaning in industrial plants
• Exterior textured and flexible coatings over pavement and precast concrete
Epoxy resins are hydrophobic and consequently are not, by themselves, dispersible in water.
However, water dispersibility can be conveyed to epoxy resins by two general methods:
Both processes are applicable to waterborne epoxy adhesives and coatings, although the
emulsification process is generally used with adhesives. Preparation of epoxy resin emul-
sions is covered in Chap. 4.
Several hybrid epoxy emulsions have been commercially prepared. An epoxy emul-
sion blended with waterborne aliphatic urethanes exhibited peel strength on aluminum of
10 lb/in—1.5 times greater than with the polyurethane itself. The optimum concentration
of urethane in the final emulsion was about 50 percent by weight.13 Epoxy-phenolic dis-
persions have also been developed to provide waterborne adhesive systems with high glass
transition temperature and chemical resistance.
The epoxy functional waterborne dispersions can be cured with many of the same cur-
ing agents that are used with nonaqueous liquid or solvent-borne epoxy resins. The curing
agents most conveniently employed are those that are water-soluble or dispersible and are
stable in an aqueous medium. The hardeners commonly used in waterborne systems are
polyamides or polyamidoamines. These are manufactured by dimerizing tall-oil fatty acid
and then reacting the dimer acid with an aliphatic amine, typically diethylenetriamine.
Dicyandiamide and substituted imidazoles have also been used for elevated-temperature,
latent curing epoxy systems. Water-compatible melamine and urea-formaldehyde resins
can be used to cure those epoxy dispersions, which contain sufficient hydroxyl groups.
substrates are mated. Often the drying is done in a forced-air oven at a temperature near
65°C. With the particular adhesive presented in Table 14.3, the working life is several hours
at room temperature.
Table 14.4 presents formulation information for bisphenol A and polyfunctional epoxy
resin emulsions that are cured with an aliphatic amidoamine curing agent. Adhesive perfor-
mance data are also provided for substrates common to the automotive industry. Both formu-
las are based on a 1 : 1 epoxy-amine stoichiometry and they are reduced to 45 percent
nonvolatiles with water. The working life of each system is several hours at room temperature.
Similar information is presented in Table 14.5 for starting adhesive formulations made
from an epoxy resin emulsion and dicyandiamide latent curing agent. This adhesive has
exceptionally good water resistance when cured. The adhesive was applied to the indicated
substrates in a manner similar to that described above, and it was cured for 3 min at 65°C
followed by 10 min at 175°C.
Parts by weight
Components A B C
Parts by weight
Formulation A B
Many epoxy dispersions are compatible with most types of latex emulsions including acrylic,
urethane, styrene butadiene, vinyl chloride, and polyvinyl acetate. The epoxy dispersion can
be used as a modifier for these emulsions to alter handling and application characteristics
such as emulsion rheology, foaming tendencies, pH sensitivity, wetting properties, and
coating coalescence. They can also be reacted into the latex resin either by reacting the
epoxy with a functionalized latex or by use of an epoxy with a coreactant. In this way adhe-
sive systems can be formulated that are cured at room or elevated temperatures.
UNCONVENTIONAL EPOXY ADHESIVES 269
Adhesives formulated with epoxy-modified latex retain the tack and conformability of
the original latex but show improvements in green bond strength and fully cured bond
strength. Cured epoxy latex epoxy resin systems also exhibit improved water and chemical
resistance over unmodified latex systems.
In choosing an epoxy and polymeric latex, it is important that they have compatibility.
Incompatibility usually occurs when the pH of the epoxy resin dispersion alters the pH of
the latex into a range where the ionically stabilized latex is broken, causing agglomeration
of the latex polymer. The pH of the epoxy resin’s emulsion may need to be adjusted before
blending with the polymeric latex.
Addition of small (10 to 20 percent by weight based on solids) amount of epoxy to a
latex emulsion will generally not change the latex viscosity characteristics. Addition of the
epoxy will help decrease the foaming tendencies of the system. Wetting characteristics of
the latex can also be improved by using epoxy modifications since the surfactant present in
the waterborne dispersion acts as a wetting agent in the latex system. This can provide a
benefit in bonding low-energy substrates such as polyolefins. Also the lower molecular
weight of epoxy polymers compared to latex polymers tends to make the epoxy a good
plasticizer for the cured latex.
The epoxy dispersion can be reacted into the latex polymer either by reacting the epoxy
with a functionalized latex (carboxyl or amino functional groups) or by use of an epoxy
with an epoxy coreactant or curing agent. In the latter case, the epoxy will react with the
coreactant and form a network within the latex polymer network.
It has been shown that tensile shear and peel strength for several latex polymers (ethyl-
ene vinyl acetate, polyvinyl alcohol, ethylene vinyl chloride, polyvinyl chloride, and
acrylic) can be significantly increased by the addition of 10 percent by weight of an epoxy
emulsion cured with a tertiary amine curing agent.17 The epoxy modification improves the
bond strength in all cases. The degree of improvement is dependent on the selection of the
latex type and the chemistry of the latex polymer.
Table 14.6 illustrates typical improvements noted in epoxy hybrid formulations with
vinyl chloride, acrylic, and styrene butadiene lattices. Tensile strengths of cured, latex-
saturated paper substrates are listed in absolute numbers while those of latex-epoxy hybrids
are listed as percent increases in tensile strength over that of the latex alone. The mecha-
nisms believed responsible for these improvements are (1) cocuring of the epoxy group
with carboxyl and amine functional groups present on the latex backbone and/or (2)
homopolymerization of the epoxy catalyzed by the tertiary amine included in some hybrid
formulations.
Epoxy modified polymer latex systems offer improved handling performance and mois-
ture and chemical strength advantages over unmodified formulations. The wide range of
latex polymers and the range of waterborne epoxy dispersions offer the formulator a wide
latitude in performance characteristics required by specific applications.
One of the major disadvantages of using structural epoxy adhesives is the cost and time
required for long cure cycles. The number of fixtures, energy use, and production times
(either at room temperature or in an oven) required have discouraged many from exploring
epoxy adhesives as an alternative to mechanical fastening.
One approach to solve this problem has been the development of fast-reacting room
temperature curing epoxy adhesive systems, as discussed in Chap. 11. However, they often
result in brittle joints, poor adhesion due to reduced time to achieve substrate wetting, and
TABLE 14.6 Properties of Epoxy-Latex Hybrids18
Styrene-
Properties Vinyl chloride-A Acrylic-A Butadiene Acrylic-B Vinyl chloride-B Vinyl chloride-C
waste from material that has passed its working life or parts that could not be positioned in
time. Modern metering, mixing, and dispensing equipment will help to eliminate this last
problem, but maintenance of the equipment, flushing of the unreacted adhesive, and the
expense of the equipment result in a completely new set of problems.
An effective way of speeding the cure of conventional adhesives has resulted from inno-
vative modifications of the heating process itself. These new processes employ various
methods to “indirectly” heat the adhesive through a medium other than heated air or elec-
trical resistance heaters that surround the joint. The processes that have been developed
have been successfully employed within industries where very high production speeds and
volumes are characteristic (e.g., transportation, furniture, appliance, etc.). These new heat-
ing processes include electromagnetic heating (induction and microwave), weldbonding,
and embedding resistance heaters within the adhesive.
These processes have an advantage in that the heat penetrates deeply into the joint and
into the epoxy material itself. With conventional thermal energy processes, the heat must
be conducted into the mass of the epoxy adhesive from outside the joint. This is hindered
by the presence of the substrates, the substrate geometry, and the relatively low thermal
conductivity of the epoxy itself.
This section first describes some of the disadvantages of traditional heating processes
that are used in adhesive bonding. Then the fundamentals of these newer indirect heating
processes are discussed along with the formulation requirements of adhesives that can be
used in these processes. The advantages and disadvantages of each heating method are
described, as are example applications. Starting formulations that are appropriate for each
process are identified.
Curing adhesive joints by traditional methods, such as oven heating, can be difficult. Ovens
tend to be energy-inefficient. Temperature ramping times are generally quite long and depend
on the type and geometry of the substrate and joint. The entire joint must be brought up to
temperature and maintained at the curing temperature for the prescribed period of time.
Loading and unloading specimens into the oven and into the fixturing required to hold the
substrates in place during cure can be cumbersome and expensive. Temperature gradients
within an oven can cause nonuniform curing, and the resulting adhesive properties can be
inconsistent. Even so, hot air curing and oven curing are often used for small production runs.
Large production runs often are cured in batches in large ovens, which run continuously.
Infrared radiant heaters provide an increase in the efficiency of heat transfer, exceeding
that of oven heaters. Infrared heaters are useful in rapid heating of localized areas of a sub-
strate. However, the rate of heat transfer is dependent, to some extent, on the color of the
workpiece. The darker the part, the more rapid the heating.19 Radiant heaters are also more
expensive and maintenance-dependent than air circulating ovens.
The traditional thermal curing methods mainly depend on a high rate of energy transfer.
The rate of heat transfer and the efficiency of the process are demonstrated in the follow-
ing three modes:
The traditional heating processes used to cure epoxy adhesives are fully described in
Chap. 17. These traditional curing processes include oven heating, hot presses and platens,
272 CHAPTER FOURTEEN
autoclave, infrared heating, and heating by electrical resistance heaters that are in contact
with or surrounding the joint area.
The primary disadvantage of these processes is that they are relatively inefficient,
mainly because the entire joint must be heated to cure only several mils of epoxy adhesives.
The energy consumed, the time to get up to temperature, and the time to cool down to a safe
handling temperature can be prohibitive in many production applications.
Speed is the most important feature of induction heating. Workpiece heating rates
greater than 40°C/s are possible. One supplier of induction curing epoxy adhesives claims
production speeds of 600 parts per/hour are possible.21
UNCONVENTIONAL EPOXY ADHESIVES 273
Metal Nonmetallic
adherend adherend
Adhesive with
Adhesive
metal powder
Metal Nonmetallic
adherend adherend
FIGURE 14.5 Left: induction heated metal parts heat thermosetting adhesive by conduction; right: parts
not heated by induced electric currents are bonded with adhesive that is heated because it contains electri-
cally conducting particles.22
274 CHAPTER FOURTEEN
Induction curing is generally used as a means of initiating or accelerating the initial cure
of structural adhesives and getting them to a point where they can be handled and passed
down the assembly line to a subprocess (e.g., paint oven) where the adhesive completely
cures. The induction cured adhesive is used to rapidly attain handling strength on the order
of 50 to 400 psi in a few seconds. After handling strength is attained, the substrate may be
unclamped and removed from a fixture. Typically, full strength of the adhesive is attained
later by way of a secondary heating source or, in the case of a two-part reactive adhesive,
cure at room temperature via its natural curing process.
Although speed may be the most important factor in considering induction curing, this
curing method is also characterized by a uniform heat distribution within a sharply defined
boundary and by unique, precisely controlled temperature. Thus, the quality of the joint can
be better and more consistent with induction curing than with other heating methods.
The advantages and disadvantages of using induction curing are shown in Table 14.7.
Induction curing eliminates the biggest disadvantage of bonding with structural adhesives—
the length of curing times required. Therefore, induction should be a prime consideration
when the productivity gains outweigh the additional equipment costs. As a result, induction
curing has been primarily developed for the demanding time constraints of the automobile
industry’s assembly-line environment.
The present uses of induction curing in the automobile industry include the fastening of
plastic retaining clips, the bonding of manufacturer’s identifying logo to the car’s body, and
the bonding of metallic fastening clips to the curved surfaces of glass windows and to the
automobile plastic grill.23 Also, induction curing offers a new approach to the production of
hem flange automotive sheet metal joints. In this application the adhesive is applied to the
upper surface of the lower steel sheet before the joint is formed. This lower sheet is then
folded up and over the end of the top sheet, and the resulting joint is subjected to a 4- to 8-s
period of induction heating.24 This hem flange joining process is illustrated in Fig. 14.6.
Advantages Disadvantages
Accelerates the cure time of the adhesive, Capital costs of the induction equipment
dramatically improving productivity
Improves the strength of the bond by improving May limit process flexibility
surface cleaning and wetting
Process easily adaptable to many available Risk of overcuring the adhesive especially with
adhesives adhesives that have a high exotherm
Localized heat that utilizes energy only where Fixturing difficult relative to the working coil
required and geometry restraints
Very low maintenance process
then occurs after the induction heating process). In either case, an adhesive must be chosen
that has very good hot strength, since the adhesive will be at very high temperatures once
cured and released from the fixturing. This will necessitate an adhesive with a high glass
transition temperature and one that can compensate for the stresses caused by differences
in the substrates’ thermal expansion coefficients.
Overheating of the adhesive is always a critical concern with induction curing. This is
especially true when the adhesive cures with a high exotherm, such as epoxy-dicyandiamide
systems. Several proprietary epoxy curing agents have been developed that provide lower
exotherm yet faster cure rate than typical dicyandiamide reactions.27,28 Tertiary amines and
modified polyamines29 are often used to accelerate the cure of dicyandiamide-epoxy
adhesives and to reduce the activation temperature. They exhibit rapid green strength devel-
opment at low temperatures while maintaining excellent shelf stability.
Typically a full cure will take less than 1 min and can occur with certain adhesives in
15 to 30 s. The partial cure process is usually employed on larger, more difficult to fixture
parts. Heat times for sufficient handling strength can be as low as 3 to 10 s. A secondary
heat source such as a final paint cure oven will then fully cure the adhesive.
Ideal induction cure epoxy adhesives can be one- or two-part and have a broad cure
temperature range of 135 to 220°C. A two-part reactive epoxy adhesive is typically used
where full strength must be attained and a secondary curing oven is impractical or unavail-
able. Although these adhesives generally have a working life of 20 min to 1 h at room tem-
perature and achieve full cure in 24 h, induction heating will accelerate the cure time. With
metal substrates and a typical two-part epoxy, handling strength can be attained in about
10 s.
One of the advantages of using a two-component room temperature curing adhesive
system is that induction curing can be used to spot-cure the adhesive in a joint (similar to a
spot-welding process). The adhesive that is not spot-cured by induction will then fully cure
at room temperature as the part moves down the assembly line.
If both substrates to be bonded are nonconducting, then the adhesive formulation must
contain a susceptor material. Susceptors can have a small percentage of magnetic iron
oxide, iron filings, or carbon additives. A susceptor can also be a steel screen or perforated
steel foil that is embedded in the adhesive bond line. It has been found that graphite fiber
composites used in the automotive and aerospace industries are sufficiently conductive that
they can be successfully heated with induction. Design considerations must be taken into
account in placement of the graphite reinforcement, so that the material heats uniformly.
The induction process will then heat the adhesive directly through the susceptor addi-
tives. A typical heat cycle for this type of joint is approximately 30 to 40 s because the
smaller particles are slower to react to the induction field than a metallic substrate.
Induction curing offers a special advantage for bonding plastic substrates. Since the plastic
substrate is not heated (assuming no metal fillers), the bulk material does not see high tem-
peratures, which could cause degradation and warpage. Hot dies can be eliminated, and this
will eliminate any surface read-through problems.
Short-length, electromagnetic waves also provide a fast method for curing structural adhe-
sives. This method of heating is also known as dielectric heating. These processes use fre-
quencies commonly associated with radio-frequency (RF) waves and microwaves. Like
induction curing, RF and microwave curing have been investigated as an alternative to
conventional processing. Such processing is more energy-efficient and has reduced curing
time when compared to conventional heating methods.
Pneumatic press
SMC composite
Epoxy adhesive
Ground
much simpler than the shaped coils used for induction curing. This dielectric curing process is
shown in Fig. 14.7. Microwave heating uses high-frequency (0.3- to 300-GHz) electromag-
netic radiation to heat a polar material or a material with a susceptor located at the joint inter-
face. The susceptor material (see below) adsorbs the microwave energy and transfers it to heat.
The Federal Communications Commission in the United States regulates microwave
usage, and the frequencies for heating are designated by the International Telecommunications
Union (ITU). The allowed frequencies most commonly used in the United States are 915 MHz
and 2.45 GHz.
Radio-frequency heating is often used on thermoplastics with a susceptor embedded in
the material and on water-based adhesives, such as acrylic, starch, and polyvinyl acetate.
The RF energy is used to quickly drive off the water and dry the adhesive. Microwave heat-
ing is generally used on structural adhesives, such as the epoxies, where the heating occurs
via rapid oscillation of the polar groups in the field. The feasibility of RF dielectric heating
to cure one- and two-part epoxy adhesives on steel, thermoplastic, and thermoset plastic
substrates has been shown.31 Process cycle times for RF curing are about 20 to 60 s, com-
pared with about 20 to 30 min for the same materials using conventional oven cure meth-
ods. Table 14.8 shows the bond strengths achieved on joints of SMC to steel.
RF or microwave heating methods are not preferred for bonding metal substrates
because the electromagnetic field tends to break down, causing arcing. Recently a new type
of microwave heating system has been commercialized that offers significant improvement
over conventional fixed-frequency microwave processing. This is the patented variable-
frequency microwave (VFM) process.33 VFM sweeps the microwave frequency over a pre-
determined range (dependent on average frequency and cavity size). In this way the power
distribution becomes quite uniform because of the superposition of a great many microwave
modes. This eliminates the problems of hot and cold spots and allows the use of metals in
microwaves by reducing arcing.
Dielectric heating makes use of the polar characteristics of the adhesive. The polar groups
become aligned with the electromagnetic field, which rotates at frequencies (for example,
2.45 GHz) that are much higher than in induction heating. The rotation of the polar groups
with the electric field produces internal heating within the adhesive. This internal heating
is sufficient to cure the adhesive. As with induction curing, a problem can easily occur
where the adhesive is overheated.
RF or microwave cured adhesives must be highly polar. This is generally recognized by
resins and curing agents that have a high electrical loss factor. It is this loss that causes
internal heating within the adhesive. Conductive fillers are not required, and in fact they are
discouraged because of microwave breakdown.
Epoxy adhesives should be 100 percent solids so as to eliminate the possibility of
gassing and bubbles occurring in the bond line. However, dielectric heating is often used
in the furniture industry to drive off water quickly from water-based adhesive emulsions.
Water, being a polar material, heats rapidly in a microwave field, as every cook knows.
Many commercial one- and two-part epoxy adhesives can be used for RF or microwave
heating without extensive reformulation. When susceptors are necessary, they are gener-
ally polyaniline materials doped with aqueous acid, such as hydrochloric acid. This intro-
duces polar groups and a degree of conductivity into the molecular structure of the
adhesive. It is these polar groups that preferentially generate heat when exposed to the high-
frequency fields. Such doped materials are also often used to produce thermoplastic gas-
kets, which can be used as an adhesive or sealant.
Because of the highly polar nature of common epoxy resins and their curatives, most
formulations can be cured via a microwave environment. A further advantage is that the
uncured material absorbs microwave energy more strongly than that which is cured, so the
cure occurs uniformly within the material.
Microwave curing has been found to increase the glass transition temperature of epox-
ies and in some instances improve the mechanical properties. It is also possible to control
the morphology of toughened epoxies more closely with microwaves.35
Several studies have shown that a microwave cure cycle can be developed that provides
equivalent performance properties to a thermal cure cycle. Table 14.9 shows the process-
ing and performance characteristics of three commercial one-component epoxy adhesives
cured via microwave and conventional thermal energy. Certain commercial epoxy adhe-
sives could contain a large number of bubbles due to volatiles present during the cure cycle
and the fast rate of cure. Therefore, specifically formulated adhesives for microwave cur-
ing may be necessary to optimize performance.
There are several claims in the literature regarding the maximum speed of microwave
curing of epoxy adhesives. Claims have been made of ten- to twentyfold cure time reduction
when compared to conventional thermal heating. A more practical estimate, however, is
that microwave joining using epoxy-based adhesive will significantly reduce the curing time
UNCONVENTIONAL EPOXY ADHESIVES 279
TABLE 14.9 Curing Parameters and Properties of Microwave Cured Epoxy Adhesives34
to one-third to one-quarter of the conventional cure time. This is accomplished while main-
taining equal or slightly higher values of the ultimate tensile strength obtained in a single
lap shear test.36 A threshold in the rate of energy deposition or input power level exists.
Above a certain energy level, the adhesive will outgas and form voids, which reduce both
the adhesive and the cohesive strength of the final product.
The epoxy adhesive formulation may require special additives and curing agents to avoid
the effects of overheating and reaction speed. A DGEBA epoxy resin with 4,4′-diamin-
odiphenylsulfone (DADPS), 4,4′-diaminodiphenylmethane (DDM), and metaphenylene
diamine (MPD) can be cured between 200 and 600 W of microwave power. The DADPS cur-
ing agent exhibits a slower reaction rate than DDM or MPDA.37 Coupling to the microwave
energy source is made more efficient by the use of various additives in the epoxy adhesive.
Carbon black, ZrO2, and Al2O3 have been shown to be particularly effective.38,39
14.6 WELDBONDING
Weldbonding is a hybrid method of assembly that utilizes components of both the metallic
welding and adhesive bonding processes. Weldbonding is claimed to provide the advantages
of both processes while minimizing the disadvantages. The benefits of instant strength and
high peel resistance provided by the welds supplement the adhesive bonding advantages of
uniform stress distribution, fatigue and vibration resistance, improved strength and durability,
and greater design flexibility.
Initially developed in the Soviet Union and used in the fabrication of transport aircraft,
weldbonding continues to be used in the assembly of aircraft parts and other large struc-
tures. Weldbonding has also found its way into the ground transportation markets. It is in
high-volume market segments where the potential of weldbonding seems to be greatest.
Weldbonding can be fully automated and utilized with robotic systems. It has been suc-
cessfully applied on both thin-gauge aluminum and steel substrates and to a lesser extent
on titanium.
Two general processing approaches prevail for weldbonding, and minor variations are pos-
sible within each approach. The first and most common method is to apply the adhesive and
weld through the joint. This process is sometimes referred to as the weld-through method.
The second approach, sometimes referred to as the flow-in method, is to form the welded
joint first and then allow a low-viscosity adhesive to flow into the spaces between the joint
280 CHAPTER FOURTEEN
1-Apply adhesive
1-Weld
2-Assemble
2-Apply adhesive
3-Spot weld
Heat Heat
(a) (b)
FIGURE 14.8 Schematic illustrations of the two fundamental approaches for produc-
ing weldbonded joints: (a) filling adhesive into the spaces between previous made spot
welds and (b) spot welding through preapplied adhesive.40
by capillary action. These two fundamental approaches for producing weldbonds are
shown in Fig. 14.8.
In these approaches, several different forms of welding can be utilized such as resis-
tance spot welding, gas tungsten spot welding, laser spot welding, and electron beam spot
welding. However, resistance spot welding is the most popular and well accepted of these
methods.
Weldbonding process specifications have been developed for use by both manufacturers
and government agencies. These process specifications give detailed steps to provide opti-
mum weldbonds. The descriptions below provide a generic summary of these processes.
Weld-Through Method. Since this is a hybrid process, techniques common to both weld-
ing and adhesive bonding are modified to fit the combined process. The essential steps of
the weld-through method of weldbonding are
4. Clamping the joint members into position and spot welding directly through the adhe-
sive layer
5. Spacing of spot welds in a predetermined pattern designed to increase the structural
integrity of the joint
6. Final curing of the adhesive either at room temperature or in an oven at elevated tem-
peratures
7. Nondestructive examination of the joint
The joints are generally made by first applying a paste adhesive, much as in conven-
tional adhesive bonding. However, after the adhesive is applied (and generally before it is
cured), the joint is assembled and a resistance spot weld is made through the adhesive layer
(Fig. 14.9). The welding electrode’s force displaces the adhesive to obtain electrical con-
tact between the substrates, and a weld is made in the conventional way. As the heating of
the weld is very localized, little damage occurs in the adhesive around the weld.
The adhesive is then cured at either room or elevated temperatures to complete the
assembly. Heat curing adhesives are normally used because of production requirements.
Typically such adhesives are cured in an oven at up to 180°C for 30 min. Often the cure
requirements can be achieved during another processing step farther down the assembly
line, such as during a paint baking operation. In this way a separate process for curing the
adhesive can be eliminated and costs saved. The spot weld allows the joint to be held
together until the assembly reaches the curing station, thereby eliminating the need for fix-
turing equipment and further saving time and costs.
Flow-in Method. In the flow-in method of weldbonding, spot welding precedes the adhe-
sive application. A normal substrate separation will result from the weld nugget buildup in
the joint (Fig. 14.8a). A low-viscosity adhesive is then used to infiltrate the clearance
between the substrate members, forming a sealed bond line. The adhesive may be drawn into
the joint by capillary action during cure or by other means such as with vacuum or pressure.
Once the joint is infiltrated with adhesive, it is cured at either room or elevated temperature.
Several joint designs have been specifically developed to provide for the flow-in method
of hybrid joining. Invented by TWI, AdFAST is a series of specially designed fasteners that
allow the adhesive to be introduced into the joint after the structure has been assembled using
welds or other mechanical fasteners.41 These fasteners incorporate a means of controlling the
spacing between the top and bottom substrates. This provides easy access for the adhesive,
greater bond line control, and improved process reliability and joint quality.
There is a debate among experts as to whether the flow-in method provides better prop-
erties and lower cost than the weld-through method. A relatively recent analysis indicates
that the flow-in technique is easier to implement and better preserves the microstructure
and hardness of the produced weldments.42 However, the weld-through method is the most
commonly used because of its high-speed production advantages.
Important Processing Issues. As with conventional adhesive bonding, there are several
important issues that cannot be overlooked with weldbonding. Two of the most important
issues are joint design and surface preparation.
It is important in the weld-through method that the adhesive be displaced as much as
possible from the local area where the resistance weld is to be made. This provides for opti-
mal welding and reduces the potential of the adhesive contaminating the weld site. The
welding electrode force locally brings the adherends together and displaces the adhesive to
allow electrical contact between the electrodes. A joint design where the spot weld areas
are raised (Fig. 14.10) facilitates flow of adhesive away from the weld location. Such a
design also allows adhesive to more easily enter the joint in the case of the flow-in method
of weldbonding.
Other techniques can be used to achieve displacement of the adhesive from the weld
area. This can be done through (1) use of an adhesive in tape form with a circle cut out from
the expected weld areas or (2) masking the weld area before the liquid adhesive is applied
to the substrate.
The surface preparation method must be carefully considered, especially if the com-
pleted weldbond is to have long-term durability to hostile environments. The surface prepa-
ration should provide an optimal surface for both adhesion and welding. Thus, the choice
of surface treatment is crucial, and there can be a conflict of requirements. The spot weld-
ing process requires a low electrical surface resistance, and many adhesive surface prepa-
ration processes provide a high surface resistance because of oxide layer buildup. When it
is impossible to harmonize on a surface treatment, current practice tends to favor treatments
that yield good weld nuggets at the expense of the adhesive bond.
Adhesive/sealant
FIGURE 14.10 Joint designs are preferred that allow the substrates to electrically contact and
displace the adhesive from the weld locations. (Courtesy: TWI World Center for Materials
Joining Technology.)
UNCONVENTIONAL EPOXY ADHESIVES 283
Metal surface resistance should be uniformly low to ensure good, consistent weld nuggets.
Because of its surface resistance, the spot welding of aluminum requires 2.5 to 3 times the
current required for steel.
Good process control is required with weldbonding to ensure correct joint filling of the
adhesive and to avoid weld quality problems. The process needs to be carefully controlled
so that health and safety requirements are met. Welding through the adhesive may create
hazardous fumes, and little information is available as to the organic compounds that are
produced. Suitable ventilation and fume extraction equipment should be provided.
Epoxy, modified epoxy, and polyurethane adhesives are commonly used for weldbonding
applications. These are used as one- or two-part liquids, pastes, or film forms. Generally
the adhesive system must provide high bond strength and durability, be easy to process,
provide good weldability, and afford corrosion protection to the substrate surfaces. The
most important criteria for any weldbonding adhesive are that it (1) have the capability of
flowing under pressure of the welding electrodes in order for metal-to-metal contact to
occur at the joint interface (weld-through method only) and (2) have sufficient heat resis-
tance to the welding temperatures so as not to detrimentally affect the strength of the final
joint.
Most adhesives used for weldbonding are arbitrarily selected from adhesives that were
developed for other purposes, resulting in a compromise when used for weldbonding.
Adhesives specifically developed for weldbonding should have the following characteris-
tics in addition to those mentioned above:
1. Viscosity should be low enough to be forced out of welding area by the pressure of the
electrodes yet sufficiently high that it will not flow out of the joint during the cure cycle.
For the flow-in process, the viscosity must be low enough to fill the joint via capillary
pressure or with moderate vacuum or positive pressure.
2. Metal fillers are often used to provide low electrical resistance between substrates.
3. An adhesive layer with appropriate thickness and elastic modulus is necessary to obtain
reasonable distribution of stresses in the region of a weldbond joint. A thin adhesive
layer of high elastic modulus improves the fatigue properties of weldbonded joints.43
4. The heat created by the spot welding process should not degrade the adhesive so as to
weaken the final bond strength or create thermal decomposition products, which can
contaminate and weaken the weld area.
5. The adhesive must remain pliable long enough to allow process completion.
6. The thermal expansion coefficient of the adhesive should closely match that of the
substrates.
Film adhesives can be cut out to provide for the absence of adhesive at the weld loca-
tions. However, often these adhesives do not flow sufficiently at the cure temperature with-
out pressure (as would be the case in weldbonded joints) to wet the substrates.
Many different adhesives have been tested and used in weldbonding. These include both
high-strength structural adhesives, such as modified epoxies, and relatively low-strength
adhesives, such as vinyl plastisols or epoxy-polysulfides that are commonly used for seal-
ing and vibration damping. Table 14.10 shows that a variety of adhesives and substrates are
compatible with the weldbonding process.
A commonly used adhesive in weldbonding applications is a modified epoxy, one- or
two-component paste containing conductive metal filler. Other fillers commonly used in
284 CHAPTER FOURTEEN
Curing Adhesive
Joint alloy Adhesive condition Spot weld, lb bond, lb Weldbond, lb
adhesives for weldbonding include fumed silica to provide thixotropy and prevent run-out
during cure and corrosion inhibitors for maximum durability in moist environments.
To assess the advantages of the weldbonded joint, one must look at the properties of the
spot weld alone, the adhesive bond alone, and compare these to the properties of the weld-
bonded joint. One must also be aware of the physical and environmental effects on the joint.
Studies show that weldbonded joints can be stronger than joints that are only spot welded
or only adhesively bonded. However, metal thickness, surface preparation, adhesive flow
and cohesion, and weld quality can influence the results.
Weldbonded aluminum joints (1-mm aluminum alloy 2036-T4) where compared with
spot welded aluminum (1-mm aluminum alloy 2036-T4) and spot welded steel (1-mm steel
1010) joints.45,46 Weldbonded 2036-T4 joints made with a vinyl plastisol adhesive and with
a one-part modified epoxy had a fatigue strength about twice that of the spot welded alu-
minum alloy and approximately the same as that of the spot welded steel. The fatigue
strength of the weldbonded aluminum joints with polysufide-epoxy adhesive or with high-
peel-strength epoxy adhesive is higher than that of steel spot welds alone.
The degree of acceptance of weldbonding applications has been increasing, as the
process has been understood and its mechanical properties developed.47 The principal
advantages that have been claimed for weldbonding include the following.
Aircraft applications include fuselage skin panels, wing center sections, leading and
trailing edges, wingtips, spoilers, and access or actuated floors. Other indicated aerospace
applications are for cryogenic tanks and for rocket shrouds. Weldbonding is important in
this industry to achieve the weight savings and structural performance required of aircraft
design.
In the automotive industry, spot welding can be used in numerous joints of the structural
frame of the vehicle. Currently, weldbonding is not being used to reduce the weld spacing
as much as it is being used to gain higher durability and stiffness. Weldbonding is beginning
to find use for improved vehicle NVH (noise, vibration, and harshness) and for increased
fatigue resistance. The increased stiffness of the vehicle reduces noise and provides a higher
driving performance. Weldbonding is also being investigated to improve crash performance.
Applications include the bonding of side apertures, the bonding of inner to outer rocker pan-
els, the bonding of cowling, various cross bracing, and the rear shelf panel.
The several disadvantages listed above for weldbonding are being overcome with prod-
uct and process development. Therefore, a high growth rate of nearly 20 percent annually
for weldbonding over the next 5 years is predicted.48 Future directions for structural weld-
bonding adhesives in vehicles will be to reduce metal gauge and weight for cost savings
and fuel economy.
The heating technology that can be used to cure epoxy adhesives is very broad. The processes
described above are the most widely used of these innovative, indirect heating processes.
However, the possible approaches appear to be nearly unlimited. Two less widely used
processes are ultrasonic curing and embedded resistance heating processes. Like those
described above, the advantage of these processes is that they cure the epoxy adhesive from
the inside of the bulk material. Thus, significant advantage can be gained in efficiency and
curing time.
Ultrasonic welding is a frictional process that has been well established for heat welding
thermoplastic parts. Like induction welding, it also has been adapted for curing structural
adhesives such as epoxy.
286 CHAPTER FOURTEEN
Booster
Horn
Substrate
Base or anvil
The basic parts in a standard ultrasonic welding device are shown in Fig. 14.11. During
ultrasonic welding, a high-frequency electrodynamic field is generated which resonates a
metal horn that is in contact with one substrate. The horn vibrates the substrate sufficiently
fast relative to a fixed substrate that significant heat is generated at the interface. Thus the
adhesive is heated at the interface, and a strong and efficient joint can be obtained.
A slow-curing, liquid thermosetting epoxy is not appropriate for ultrasonic bonding.
The liquid lubricates the interface and is hammered out of position with little curing action.
However, a B-stage epoxy adhesive can be used successfully with ultrasonic activation.
Table 14.11 shows that several epoxy adhesives are capable of gaining good strength dur-
ing a few seconds of ultrasonic activation, and then full-strength bonds can be accom-
plished with an oven postcure. No clamps were used during the oven cycles with these
particular adhesives.
Rapid curing of structural epoxy adhesive by ultrasonic heating has been demonstrated
successfully in recent work. The conversion of epoxy groups produced by ultrasonic cur-
ing for 50 s was almost 3 times higher than that obtained by thermal heating.51
Tensile shear
Adhesive Form Adherends Weld time, s strength, psi
REFERENCES
1. “Study Focuses on Radiation Cured Products,” Paint and Coatings Industry, vol. 12, no. 10,
October 1996.
2. Pappas, S. P., UV Curing, Science and Technology, vol. 1, Technology Marketing Corporation,
Norwalk, CT, 1978.
3. Pappas, S. P., UV Curing, Science and Technology, vol. 2, Technology Marketing Corporation,
Norwalk, CT, 1985.
4. Holman, R., UV and EB Curing Formulations for Printing Inks, Coatings, and Paints, Selective
Industrial Training Association, London, 1984.
5. RadTech N.A, Chevy Chase, MD.
6. Carder, C. H., “Radiation Curing Coatings,” Paint and Varnish Products, vol. 64, no. 8, 1974, p. 19.
7. Zwanenburg, R. C. W., “How to Formulate UV-Curing Coatings,” RadNews Highlights, Paint
Research Association, Middlesex, UK, Spring 1998.
8. Oldring, P. K. T., ed., Chemistry and Technology of UV and EB Formulations for Coating, Inks,
and Paints, vol. 4: Formulation, SITA Technology, London, 1991.
9. Shirane, K., “Radiation-Induced Free Radical Cure of Epoxy Resin,” Journal of Polymer
Science, vol. 17, 1979, p. 139.
10. H-Nu 470, Spectra Group, Ltd., Maumee, OH.
11. Valero, G., “Enlightening Developments,” Adhesives Age, June 2002, pp. 16–17.
12. Janke, C. J., et al., “Electron Beam Curing of Epoxy Resins by Cationic Polymerization,” SAMPE
International Symposium, vol. 41, 1996, p. 196.
13. Buehner, R. W., and Atzinger, G. D., “Waterborne Epoxy Dispersions in Adhesive Applications,”
Epoxy Resin Formulators Conference, San Francisco, Feb. 20–22, 1991.
288 CHAPTER FOURTEEN
14. Amstock, J., Handbook of Adhesives and Sealants in Construction, McGraw-Hill, New York, 2002.
15. Waterborne Epoxy Dispersions in Adhesive Applications, Resolution Performance Products LLC,
SC2267, Houston, TX, 2000.
16. Waterborne Epoxy Dispersions in Adhesive Applications, Resolution Performance Products LLC.
17. Young, G. C., “Modifying Latex Emulsions with Epoxy Resin Dispersions,” Adhesives Age,
September 1996, pp. 24–27.
18. Waterborne Epoxy Dispersions in Adhesive Applications, Resolution Performance Products LLC.
19. Schields, J., Adhesives Handbook, 3d ed., Butterworths, London, 1984.
20. Technical Bulletin, “EMA Bond Process,” Ashland Inc., Norwood, NJ, 2004.
21. Turi, D., “Supplier/User Cooperation Yields Beneficial Induction Cure Epoxy,” Adhesives Age,
June 1993, p. 22.
22. Bolger, J. C., and Lysaght, M. J., “New Heating Methods and Cures Expand Uses for Epoxy
Bonding,” Assembly Engineering, March 1971, pp. 46–49.
23. Mittleman, E., “Fast Bonding Cuts Auto Costs,” IEEE Spectrum, November 1977, pp. 73–75.
24. Stefanides, E. J., “Epoxy Cured by Induction Heating Gives Strong Sheet Metal Joint,” Design
News, June 22, 1987, pp. 102–103.
25. Stefanides, E. J., “Epoxy Cured by Induction Heating Gives Strong Sheet Metal Joint,” Design
News, June 22, 1987.
26. Turi, D., “Supplier/User Cooperation Yields Beneficial Induction Cure Epoxy,” pp. 22–25.
27. Bolger, J. C., “New One Component Epoxy Insulation Compounds,” Insulation, October 1969.
28. Bolger and Lysaght, “New Heating Methods and Cures Expand Uses for Epoxy Bonding.”
29. Technical Bulletin, “Ancamine 2441,” Air Products and Chemicals Company, Allentown, PA, 2004.
30. Li, C., and Dickie, R. A., “Bonding Adhesive Joints with Radio Frequency Dielectric Heating,”
International Journal of Adhesion and Adhesives, vol. 11, no. 4, October 1991, pp. 241–246.
31. Li and Dickie, “Bonding Adhesive Joints with Radio Frequency Dielectric Heating.”
32. Li and Dickie, “Bonding Adhesive Joints with Radio Frequency Dielectric Heating.”
33. Technical Bulletin, “MicroCure,” Lambda Technologies, Raleigh, NC, 2004.
34. Gaskin, G. B., et al., “Electromagnetic Curing of Epoxy Adhesive Systems,” 38th SAMPE
International Symposium, May 10–13, 1993.
35. Howell, B. F., “Modification of Epoxy Resins,” in Polymer Modification: Principles, Techniques,
and Applications, J. J. Meister, ed., Marcel Dekker, New York.
36. Gaskin, “Electromagnetic Curing of Epoxy Adhesive Systems.”
37. Boey, F. Y. C., et al., “Microwave Curing of Epoxy Amine System—Effect of Curing Agent on
Rate Enhancement,” Polymer Testing, vol. 18, no. 2, 1999.
38. Paulauskas, F. L., et al., “Adhesive Bonding/Joining Via Exposure to Microwave Radiation,” 27th
International SAMPE Technical Conference, September 9–12, 1995.
39. Soesatyo, B., et al., “Effect of Microwave Curing Carbon Doped Epoxy Adhesive–Polycarbonate
Joints,” International Journal of Adhesion and Adhesives, vol. 20, no. 6, 2000, pp. 489–495.
40. Mussler, R. W., “Weld-Bonding: The Best or Worst of Two Processes,” Industrial Robot, vol. 29,
no. 2, 2002, pp. 138–148.
41. AdFAST Joining, TWI World Center for Materials Joining Technology, Cambridge, UK, 2004.
42. Darwish, S. M. H., and Ghanya, A., “Critical Assessment of Weld-Bonded Technologies,”
Journal of Materials Processing Technology, vol. 105, 2000, pp. 221–229.
43. Chang, B. H., et al., “A Study on the Role of Adhesive in Weld-Bonded Joints,” Welding Research
Supplement, August 1999, pp. 275s–279s.
44. Anonymous, “Spot Welding Teams Up with Adhesives for Stronger Metal-to-Metal Bonds,”
Product Engineering, May 1975.
45. Anonymous, “Spot Welding Teams Up with Adhesives for Stronger Metal-to-Metal Bonds.”
46. Anonymous, “Weldbonding Joins Auto Body Sheet,” Welding Design & Fabrication, May 1979,
pp. 80–81.
UNCONVENTIONAL EPOXY ADHESIVES 289
47. Schwartz, M. M., Metals Joining Manual Book, McGraw-Hill, New York, 1979, pp. 1–32.
48. Lohman, R. J., “Weldbonding Leads Growth of Structural Adhesives in Auto Market,” Adhesives
& Sealants Industry, May 1997, pp. 30–34.
49. Grimm, R. A., “Welding Processes for Plastics,” Advanced Materials and Processes, March 1995.
50. Hauser, R. L., “Ultra Adhesives for Ultrasonic Bonding,” Adhesives Age, March 1969.
51. Kwan, K. M., and Benatar, A., “Investigation of Non-Thermal Effects Produced by Ultrasonic
Heating on Curing of Two-Part Epoxy Adhesive,” Society of Plastic Engineers Annual Technical
Conference, Dallas, TX, 2001.
This page intentionally left blank
CHAPTER 15
EFFECT OF THE SERVICE
ENVIRONMENT
15.1 INTRODUCTION
For an adhesive or sealant bond to be useful, not only must it withstand the mechanical
forces that are acting on it, but also it must resist the service environment or the chemical
forces that are applied. Thus, one of the most important characteristics of an epoxy adhe-
sive or sealant is its endurance to the operating environment.
Strength and permanence are influenced by many common environmental elements.
These include high and low temperatures, moisture or relative humidity, chemical fluids,
and outdoor weathering. Table 15.1 summarizes the relative resistance of various types of
epoxy adhesives to common operating environments.
The effect of simultaneous exposure to both mechanical stress and a chemical environ-
ment is often more severe than the sum of each factor taken separately. Mechanical stress,
elevated temperatures, and high relative humidity can be a fatal combination for certain
adhesive formulations if all occur at the same time. Add to this the possible cyclic effects
of each factor, and one can easily see why it is important to understand the effects of envi-
ronment on the joint.
In this chapter, first the importance of environmental testing is reviewed. Then the
effects of individual and combined environments on epoxy adhesives are considered. The
root cause of environmental degradation is studied so that those seeking to design, manu-
facture, and use epoxy adhesives can make better judgments about the durability of bonded
joints. This chapter points to common problem areas, suggests changes in the bonding sys-
tem to provide greater bond durability, and acts as a foundation for further, more detailed
investigation into this important area.
291
Copyright © 2006 by The McGraw-Hill Companies, Inc. Click here for terms of use.
TABLE 15.1 Relative Resistance of Epoxy Adhesives to Common Service Environments1
Shear 2 2 2 1 2 1 2
Peel 5 5 2 6 2 1 3
Heat 3 1 6 1 6 6 4
Cold 5 4 2 3 2 2 3
Water 2 3 2 2 1 1 3
Hot water 2 3 6 2 6 6 4
292
Acid 2 2 3 2 2 2 3
Alkali 2 2 6 2 2 2 2
Oil, grease 2 2 2 3 2 2 3
Fuels 3 2 2 3 2 2 5
Alcohols 1 2 1 2 2 2 5
Ketones 6 6 6 6 6 6
Esters 6 6 6 6 6 6
Aromatics 1 2 3 2 2 2 6
Chlorinated solvents 6 6
TABLE 15.2 ASTM Test Methods for Determining Environmental Resistance of Adhesives and
Sealants
If there is only one parameter that changes due to environmental exposure, then the
application of accelerated test techniques and analysis may yield useful information as to
service life. In the electrical insulation industry, for example, Arrhenius plots are often used
to predict end of life of insulating materials by simple extrapolation. This can be accom-
plished because insulation life is primarily dependent on temperature, and other factors are
relatively minor. However, there are a multiplicity of aging phenomena that can occur
within the adhesive, interface, and adherend and with each having a possible effect on the
others. The formidable task of determining the end of life is one of the most difficult chal-
lenges in adhesive science and possibly the single item that most greatly inhibits the use of
adhesives and sealants in most structural applications.
Predicting the service life of adhesives is a risky business. The most difficult question ever
put to an adhesive consultant is, How long will the adhesive joint last in service? The prob-
lem is that an adhesive joint is not made up of just one element. It contains several elements,
and some of them interact. In fact, in most adhesive joints at least five elements must be
considered: substrate A, interface A, the adhesive, interface B, and substrate B. To under-
stand and predict the rate of degradation of each of these elements is challenging, but it can
be done. The most difficult failure situations to predict are those that result from interactive
effects.
Thus, it is important to consider and evaluate the adhesive joint as a complete “system,”
but to do so we must generally separate these elements and look at their individual mecha-
nisms of degradation. The following discussion considers mainly the bulk adhesive com-
ponent of the joint. The process summarized below is the combination of work that has
been described previously in several references.2,3
Life prediction methodology embraces all aspects of the numerous processes that could
affect the function of the element—in this case the bulk adhesive. The first step is to define
the function of the adhesive clearly enough for a failure criterion to be derived. This fail-
ure criterion may be an unacceptable reduction in tensile strength, time to creep failure
under a given stress, reduction in modulus due to moisture ingression, increase in modulus
due to oxidation, unacceptable crack depth, or a variety of other possible criteria. It is also
important that the criteria be related to practical adhesive joint performance. This is where
it is difficult, and one must presume, at least for this limited analysis, that the adhesive will
fail via a bulk (cohesive) property.
After the failure criterion has been defined, the various processes that could cause this
failure must be analyzed. For example, an increase in modulus could occur by thermal oxi-
dation, increased postcure crosslinking, or the loss of plasticizer. Whatever the mechanism,
each possible process needs to be identified and its rate characterized separately. Only then
can interactions between different mechanisms be considered for life prediction.
Finally, the rate of change in the critical property must be measured relative to expected
environments of different severity and time intervals. If measurement cannot be made at the
service temperature in the time that is available (as is generally the case), then accelerated
tests may be used at elevated temperature or increased frequency. However, it is extremely
important that care be taken to match the accelerated test conditions to the service condi-
tions in as realistic a way as possible. For example, if accelerated aging by elevated tem-
peratures is being used, the temperature must not be so high as to begin a degradation
mechanism that would not normally be seen in service.
After the rate of each process is determined under the accelerated conditions, then the
rate at the actual service condition can be determined by extrapolation. The Arrhenius rela-
tionship (i.e., plotting the log failure rate against reciprocal temperature) is often used to
EFFECT OF THE SERVICE ENVIRONMENT 295
accomplish this. This can also be expressed by the following equation for the time to fail-
ure tf at absolute temperature T and constant stress σ.
log tf = C − log T + a(1/T) − b(σ/T)
The terms C and b are constants, and a = EA/2.303R, where EA is the Arrhenius constant.
All three terms can be found by graphical procedures.4
A comparison of experimentally determined failure times for different stress levels and
those predicted by the above equation for epoxy-aluminum lap shear joints aged at 60°C
and 95 percent RH is presented in Table 15.3. These results indicate that the reaction rate
method is satisfactory for predicting the effects of temperature and stress on the lifetime of
adhesive bonds, provided that failure is cohesive within the adhesive. This, of course,
should be validated by prototype testing.
After each process is separately understood, interactions can be put together to provide
a life prediction of the durability of the component under actual service conditions. One
approach is to use multiple regression analysis to develop predictive equations for failure
times in which several parameters (e.g., temperature, relative humidity, and stress) are
treated as independent variables.6
The above analysis applies to degradation processes that relate to the bulk adhesive.
Interfacial degradation processes such as corrosion can be similarly determined. Thermal
and oxidative stability, as well as corrosion and water resistance, depends on the adherend
surface as well as on the adhesive itself. Epoxy-based adhesives degrade less rapidly at ele-
vated temperatures when in contact with glass or aluminum than when in contact with cop-
per, nickel, magnesium, or zinc. The divalent metals have a more basic oxide surface than
the higher-valence metal oxides and hence serve to promote dehydrogenation reactions,
which lead to anion formation and chain scission.7
A diagram that one might use to illustrate a possible set of experimental data to repre-
sent all failure modes of an adhesive joint is presented in Fig. 15.1. When the data are
closely analyzed and the extent of ultimate service life and proper safety margins are spec-
ified, the critical failure mode and time can be defined by identifying the “weakest link”—
in this case the corrosion mechanism. If this predicted life is longer than the expected
service life of the product, then the material specified for the adhesive joint can be quali-
fied for use.
What is presented above is a very simplistic approach. Joint geometries, for example,
may have a significant effect on the rate of degradation, again depending on the environ-
ment. As a result, geometric modeling and finite element analyses have been employed
with durability studies to assist in life predictions.
Mode 3
(thermal oxidation)
+ Mode 2
(mechanical fatigue)
Margin
of safety Ultimate
_ service life
Mode 1 (corrosion)
Time
FIGURE 15.1 Determination of critical failure mode.8
One of the most degrading elements for organic adhesives, including epoxies, is heat. All
polymeric materials are degraded to some extent by exposure to elevated temperature. Not
only do elevated temperatures lower short-term physical properties, but also properties will
likely degrade with prolonged thermal aging at lower temperatures. Thus, several impor-
tant questions need to be asked of an adhesive if high service temperatures are expected.
• What is the maximum temperature that the bond will be exposed to in service?
• What is the average temperature to which the bond will be exposed?
• How long will the bond be exposed to various temperatures?
• What is the rate of temperature change?
• What is the type of stress and stress levels that the bond will be exposed to during ser-
vice? How does this relate to temperature?
Ideally, one would like to have a definition of the entire temperature-time relationship rep-
resenting the adhesive’s expected service history. These data would include time at various
temperatures, number of temperature cycles, and rates of temperature change.
EFFECT OF THE SERVICE ENVIRONMENT 297
• Split polymer molecules (chain scission), causing lower molecular weight, degraded
cohesive strength, and low-molecular-weight by-products
• Continued crosslinking, resulting in bond embrittlement and shrinkage
• Evaporation of plasticizer, resulting in bond embrittlement
• Oxidation (if oxygen or a metal oxide interface is present), resulting in lower cohesive
strength and weak boundary layers
Often thermal stability can be indicated by the weight loss that the bulk adhesive experi-
ences in a high-temperature environment. However, this is not a very good method for pre-
dicting joint strength because it neglects the effect of joint geometry, adhesion, interface
chemistry, etc. Figure 15.2 is a comparison of cured DGEBA epoxy with various other poly-
mers in terms of weight loss after 15 h at 175°C.
This moderately good thermal resistance is one of the reasons for the wide use of epoxy
adhesives in critical high-temperature environments. However, as described below, the
elevated-temperature performance of epoxy adhesives is limited by the chemical nature of
the epoxy molecule. Additives, modifiers, or improved curing agents only make minor
improvements regarding thermal resistance.
Most organic adhesives including epoxies degrade rapidly at service temperatures
greater than 150°C. However, several polymeric materials (e.g., polyimides, polybenzimi-
dazoles) have been found to withstand up to 250 to 300°C continuously and even higher
temperatures for a short-term basis. To use these materials, one must generally pay a pre-
mium in adhesive cost and be able to provide long, high-temperature cures, often with pres-
sure. Long-term temperature resistance, greater than 250 to 300°C, can only be accomplished
with inorganic or ceramic-based adhesives.
The base polymer, of course, is a key ingredient in a high-temperature epoxy adhesive sys-
tem. For an adhesive to withstand elevated temperatures, it must have a high melting or
softening point and resistance to oxidation. Materials with a low melting point, such as
many of the thermoplastic adhesives, may prove excellent adhesives at room temperature;
298 CHAPTER FIFTEEN
4
Percent weight loss
0
Phenolic Polyester Silicone Modified Epoxy
asphalt
FIGURE 15.2 Comparison of cured DGEBA epoxy with various other polymers in
terms of weight loss after 15 h at 175°C.9
however, once the service temperature approaches the glass transition temperature plastic
flow results in deformation of the bond and degradation of cohesive strength.
Thermosetting adhesives, exhibiting no melting point, consist of highly crosslinked
networks of macromolecules. Because of this dense crosslinked structure, they show rela-
tively little creep at elevated temperatures and exhibit relatively little loss of mechanical
function when exposed to either elevated temperatures or other degrading environments.
Many of these materials are suitable for moderately high-temperature applications.
When one is considering thermosets, the critical factor is the rate of strength reduction
due to thermal oxidation or pyrolysis. Thermal oxidation can result in chain scission or
crosslinking. Crosslinking causes the polymer to increase in molecular weight, leading to
brittleness and decreased elongation. Progressive chain scission of molecules results in
losses of weight, strength, elongation, and toughness within the bulk adhesive. Figure 15.3
illustrates the effect of oxidation by comparing epoxy adhesive joints that are aged in both
high-temperature air and inert gas (nitrogen) environments. The rate of bond strength
degradation in air depends on the temperature, adhesive, rate of airflow, and even the type
of adherend.
For reasons as described above, some metal adhesive interfaces are chemically capable
of accelerating the rate of oxidation. For example, it has been found that nearly all types of
structural adhesives exhibit better thermal stability when bonded to aluminum than when
bonded to stainless steel or titanium (see Fig. 15.3).
Pyrolysis is simple thermal destruction of the molecular chain of the base polymer in
the adhesive or sealant formulation. Pyrolysis causes chain scission and decreased molec-
ular weight of the bulk polymer. This results in reduced cohesive strength and increased
brittleness. Resistance to pyrolysis is predominantly a function of the intrinsic heat resis-
tance of the polymers used in the adhesive formulation. As a result, many of the aromatic
and multifunctional epoxy resins that are used as base resins in high-temperature adhesives
are rigidly crosslinked or are made of a molecular backbone referred to as a ladder struc-
ture, as shown in Fig. 15.4.
EFFECT OF THE SERVICE ENVIRONMENT 299
5000
Aged in nitrogen on 17−7 pH stainless steel
Tensile shear measured
3000
2000
1000
0
0 200 400 600 800
Aging time at 500°F, h
FIGURE 15.3 The effect of 260°C aging in air and nitrogen on an epoxy-phenolic
adhesive.10
The ladder structure is made from aromatic or heterocyclic rings in the main polymer
structure. The rigidity of the molecular chain decreases the possibility of chain scission by
preventing thermally agitated vibration of the chemical bonds. The ladder structure pro-
vides high bond dissociating energy and acts as an energy sink to its environment. Notice
in Fig. 15.4 that to have a complete chain separation (resulting in a decrease in the molec-
ular weight), two bonds must be broken in the ladder polymer; whereas only one needs to
be broken on a more conventional linear or branched chain structure.
To be considered a promising candidate for high-temperature applications, an adhesive
must provide all the usual functions necessary for good adhesion (wettability, low shrink-
age on cure, thermal expansion coefficient similar to that of the substrate, toughness, etc.),
and it must also possess
Thermal energy
R R R
The high-temperature resins described above provide the main elements in the adhesive
formulator’s recipe. However, there are also additives, fillers, etc., that can further enhance
the thermal properties of more conventional epoxy adhesives. These additional compo-
nents improve thermal resistance by providing oxidation resistance, toughening, and con-
trol of bond line stress.
TABLE 15.4 Short-Term Strength and Cure Properties of High-Temperature Structural Adhesives
5
Shear strength, 1000 psi
4
Epoxy
Polybenzimidazole (PBI)
3
Epoxy-phenolic
2
Polyimide
1
Modified epoxy
0
0 200 400 600 800
Temperature, °F
(a)
3.5
PBI at 350°F
3.0
Shear strength, 1000 psi
2.0
Epoxy phenolic at 250°F
1.5 Polyimide at 550°F
reactive free radicals caused by heat or metallic impurities. The function of an antioxi-
dant is to prevent propagation or the reaction of these free radicals with oxygen to form
unstable species.
Antioxidants should be included in high-temperature adhesive formulations in order to
achieve optimum thermal aging properties. Antioxidants used in structural adhesives differ
from those used to improve thermal stability of thermoplastic materials in that they must be
less volatile, resistant to higher temperatures, longer-acting, and of course compatible with
the base polymer. Antioxidants used in structural adhesives are generally of inorganic origin,
302 CHAPTER FIFTEEN
Toughening. For many years, the typical method of improving the toughness of high-
temperature structural adhesives was to add elastomeric resins to rigid high-temperature
base polymer to create a hybrid product such as epoxy-nitrile or epoxy-polysulfide systems.
However, the toughening of high-temperature adhesives can provide a difficult challenge,
since the service temperatures usually exceed the degradation point of most rubber addi-
tives. Also, the addition of an elastomer generally resulted in lowering of the glass transi-
tion temperature of the adhesive.
None 670
n-Propel gellate 1074
Gallic acid 820
Acetyl acetone 985
Catechu 980
Aluminum triacetonylacetonate 960
Ethylenediamine 835
*
Measured on aluminum adherends; tested at 23°C after aging 200 h
at 286°C.
EFFECT OF THE SERVICE ENVIRONMENT 303
However, newer adhesives systems having moderate temperature resistance have been
developed with improved toughness without sacrificing other properties. When cured,
these structural adhesives have discrete elastomeric particles embedded in the matrix. The
most common toughened hybrids using this concept are acrylic and epoxy systems. The
elastomer is generally a vinyl- or carboxyl-terminated acrylonitrile butadiene copolymer.
These adhesive formulations are discussed in detail in Chaps. 8 and 12.
Within the past several years, improvements in the toughening of high-temperature epox-
ies and other reactive thermosets, such as cyanate esters and bismaleimides, have been accom-
plished through the incorporation of engineering thermoplastics. Additions of poly(arylene
ether ketone), PEK, and poly(aryl ether sulfone), PES, have been found to improve fracture
toughness. Direct addition of these thermoplastics generally improves fracture toughness but
results in decreased tensile properties and reduced chemical resistance.
Chemical functionalization of the thermoplastics was found to improve toughness with-
out such detractions. High-molecular-weight resins based on amine-terminated PES
oligomers or chain extension of bismaleimide resin with the same amine-terminated PES
were found to have improved fracture resistance and reduced thermal shrinkage.14 Also a
mechanism was found to toughen cyanate esters by incorporating epoxy resins, which can
react with the ester.15
The only high-temperature resin family that retains a moderate amount of flexibility is
the polysiloxanes. A significant amount of research has been devoted to trying to marry the
properties of siloxanes with epoxy resins to obtain less brittle, high-temperature adhesives.
However, these efforts have yet to result in commercial adhesives systems.
Reducing Internal Stress. Internal stresses are common in joints made with high-
temperature adhesives. These stresses can be due to
Stresses caused by items 1 and 2 above are magnified by the mismatch in thermal
expansion coefficients between the adhesive and the substrate. Incorporating fillers into the
adhesive formulation can often reduce these stresses. Fillers also reduce the thermal shrink-
age during aging by bulk displacement of the polymeric resin.
Flexibilizers generally cannot be used to counteract internal stress in high temperature
adhesive because of their relatively low glass transition temperature and thermal endurance
properties. However, most high-temperature adhesive systems incorporate metallic fillers
(generally aluminum powder) to reduce the coefficient of thermal expansion and degree of
shrinkage.
It is usually not possible to match the adhesive’s coefficient of thermal expansion to
the substrate, because of the high filler loadings that would be required. High loading
volumes increase viscosity to the point where the adhesive could not be easily applied or
wet a substrate. For some base polymers, filler loading values up to 200 parts per hun-
dred (pph) may be employed, but optimum cohesive strength values are usually obtained
with lesser amounts.
Metal fillers for high-temperature adhesives must be carefully selected because of their
possible effect on oxidation, as indicated in the previous section. Carrier films, such as
glass cloth, are generally used to facilitate the application of the adhesive, but they also pro-
vide a degree of reinforcement and lowering of the coefficient of thermal expansion. Thus,
they reduce the degree of internal stress experienced at the joint’s interface.
304 CHAPTER FIFTEEN
Epoxy adhesives are generally limited to applications below 125°C. Figure 15.6 illustrates
the aging characteristics of a typical epoxy adhesive at elevated temperatures. The epoxy
adhesives using aliphatic polyamine hardening agents are not serviceable above 65°C. The
aromatic diamine and monoanhydride cured products are usable at temperatures of 120 to
150°C. Certain epoxy adhesive formulations, however, have been able to withstand short
terms at 260°C and long-term service at 150 to 175°C. These systems were formulated
specifically for thermal environments by incorporation of stable epoxy coreactants or high-
temperature curing agents into the adhesive.
Although conventional bisphenol A epoxy resins are limited to service temperatures
under 125°C owing to their molecular structure, a number of approaches have been inves-
tigated in the development of high-temperature adhesives with epoxylike processability.
Investigations of high-temperature epoxy adhesives have generally taken one of four
courses to development:
1. Epoxy coreactants which will increase the temperature resistance of the system (e.g.,
epoxy-phenolic adhesives)
2. High-temperature curing agents
3. Special epoxy resins
4. Combinations of the above
The number of epoxide groups per molecule and the rigidity of the molecular structure
are factors that affect the hot strength of the epoxy adhesive. Thus, epoxy novolac and gly-
cidyl ethers of tetraphenolethane have become important resins for incorporating into high-
temperature epoxy adhesives because of their multifunctionality.
Advantages of epoxy-based high-temperature adhesives relative to other adhesive types
include relatively low cure temperatures, no volatiles formed during cure, low cost, and a vari-
ety of available formulating and application possibilities. The higher-temperature aromatic
6000
Aged at 266°F (130°C)
5000
Shear strength tested at room
temperature, psi
4000
2000
0
0 4 8 12 16 20 24 28 32
Time, weeks
FIGURE 15.6 Effect of temperature aging on typical epoxy adhesive in air. Strength
is measured at room temperature after aging.16
EFFECT OF THE SERVICE ENVIRONMENT 305
adhesives, such as polyimides, lose many of these advantages in favor of improved ther-
mal-aging characteristics.
Epoxy Coreactants. One of the most successful epoxy coreactant systems developed thus
far is an epoxy-phenolic alloy. The excellent thermal stability of the phenolic resins is cou-
pled with the valuable adhesion properties of epoxies to provide an adhesive capable of
371°C short-term operation and continuous use at 175°C. The heat resistance and thermal-
aging properties of an epoxy phenolic adhesive are compared with those of other high-
temperature adhesives in Fig. 15.5. Epoxy-phenolic adhesives are generally preferred over
other high-temperature adhesives, such as the polyimides and polybenzimidazoles, because
of their lower cost and ease of processing.
Two of the first commercial epoxy-phenolic adhesive formulations are described in
Table 15.6. These adhesives generally contain more phenolic resin than epoxy resin, and as
a result, they have good high-temperature and chemical-resistant properties, but the adhe-
sive suffers from relatively poor peel and impact strengths. The first epoxy-phenolic adhe-
sives were developed at the Forest Products Lab (FPL) as a result of research in the 1950s
aimed at high-temperature military aircraft adhesives. The cure cycle to yield optimum
properties is about 1 h at 175°C. In addition to good hot strength and moderate thermal sta-
bility, epoxy-phenolic adhesives also exhibit good long-term resistance to moisture.
The formulation on the right of Table 15.6 is intended to be applied as a solution. The
adhesives solution is brushed on both substrates, precured for 30 min at 93°C, and then
pressed at 160°C for 30 min at 50 psi. The 442J formulation was designed to be applied to
a glass fabric carrier. When cured, this adhesive provides 2000- to 3000-psi shear strength
at room temperature and good strength at 260°C. Typical properties of 442J are compared
to those of polyimide and polybenzimidazole adhesives in Table 15.7.
Chelating agents are commonly used in epoxy-phenolic adhesives to stabilize the metal
interface. In the epoxy-phenolic formulations given above, quinolinate and gallate are used
Epoxy-phenolic
Epoxy-phenolic 422J solution coating
as chelating agents. They provide increased bond durability above 200°C on substrates
such as iron and steel.
The epoxy-phenolic adhesives have moderately good peel strength (10 to 15 lb/in),
which is acceptable for many structural applications. However, the demands for high-peel-
strength, high-temperature adhesives, such as aerospace honeycomb bonding, have resulted
in formulations with the addition of thermoplastic modifiers in the epoxy-phenolic formula-
tion. These materials generally sacrifice high-temperature resistance for improved peel
strength.
Epoxy Resins. Where long-term service is required, most epoxies are limited to temper-
atures no higher than 125°C. With the increased demand for high-temperature and chemi-
cal resistance, new epoxy resins have been developed which have a higher epoxide and
aromatic content. These resins are polyfunctional and are often cured with an aromatic
diamine, such as methylene dianiline, metaphenylene diamine, or diamiodiphenylsulfone.
Heat distortion temperature up to 165°C can be obtained. Several commercially available
high-temperature epoxy resins are shown in Fig. 15.7. The high-performance resins can be
crosslinked at elevated temperatures with either an aromatic amine or a catalytic curing
agent.
Certain epoxy adhesives based on new multifunctional resins were found to provide
strength retention to about 232°C. These epoxy resins have been the subject of research, but
have not as yet reached significant commercial status.
Glycidyl polyether of tetraphenylethane is a standard, multifunctional high-temperature
epoxy resin, and Fig. 15.8 illustrates the relationship between strength and temperature of
this resin cured with two different amines and two different anhydrides. This resin has also
been used with diaminodiphenylsulfone (DADPS) curing agents and polyvinyl acetal mod-
ifiers (to improve peel strength and toughness). Glycidyl ether of resorcinol has also been
shown to have excellent hot strength even when cured at room temperature.
Epoxy novolacs are a class of epoxy adhesives that are often used in high-temperature
application because of the resin’s polyfunctionality that results in a high degree of crosslink-
ing. They are made by reacting phenolic resins of the novolac type with epichlorohydrin.
The response to curing agents is similar to that of the more standard DGEBA epoxy resins.
Polyimide
Tensile shear strength, psi (PI-1101) Polybenzimidazole Epoxy-phenolic
CH HC CH CH2
307
O O O O O
O CH2 CH CH2 CH2 CH CH2 O O CH2 CH CH2 CH2 CH CH2 N CH2 CH CH2
Tactix 742 EPON 1031 Araldite MY 720
FIGURE 15.7 High-performance epoxy resins: Tacktix 742 (Dow), EPON 1031 (Resolution Performance Products LLC), and Araldite MY 720 (Ciba Geigy).19
308 CHAPTER FIFTEEN
3000
Tensile shear strength, psi
Diaminodiphenyl
sulfone
2000
Methylenedianiline
Chlorendic anhydride
1000
Pyromellitic dianhydride
0
0 50 100 150 200 250 300
Temperature, °C
FIGURE 15.8 Strength-temperature relationship of glycidyl ether of tetraphenylethane
cured with four different curing agents.20
For maximum heat resistance, pyromellitic dianhydride is often used. The composition and
properties of metal-to-metal adhesives based on a combination of epoxy novolac and a
bisphenol A epoxy resin have been described in Sec. 12.5.1. Depending on the cure tempera-
ture and the choice of amine curing agent, strength as high as 3000 psi at room temperature and
over 1000 psi in the range of −55 to + 150°C is possible.
Epoxy Curing Agents. The selection of the curing agent can have a significant effect on
the heat resistance of an epoxy adhesive. Table 15.8 shows the heat resistance of a con-
ventional liquid DGEBA epoxy resin (EEW: 200) cured with various hardeners. The ten-
sile shear strength is provided after no thermal aging and after 200-h aging at 260°C.
Anhydride curing agents give unmodified epoxy adhesives somewhat greater thermal
stability than most other epoxy curing agents. Benzophenonetetracarboxylic dianhydride
(BTDA), phthalic anhydride, pyromellitic dianhydride, and chlorendic anhydride allow
greater crosslinking and result in short-term heat resistance to 232°C. Long-term thermal
endurance, however, is limited to 150°C.
Although they are more difficult to formulate into epoxy adhesive systems, anhydride
cured epoxies have somewhat better thermal stability than amine cured systems. Aromatic
and cyclic anhydrides, such as phthalic anhydride, pyromellitic dianhydride, and chloren-
dic anhydride, provide the most stable structures.
A typical formulation for a metal-to-metal adhesive-sealant that is cured with a combi-
nation of phthalic anhydride and pyromellitic anhydride is shown in Table 12.6. Table 15.9
shows the high-temperature properties of another epoxy formulation cured with pyromel-
litic dianhydride. Epoxy formulations cured with pyromellitic dianhydride (PMDA) show
good short-term thermal stability in the temperature range of 150 to 230°C.
Benzophenonetetracarboxylic dianhydride (BTDA) is another anhydride curing agent
that provides good high-temperature epoxy adhesives formulations. Tables 12.12 and 15.10
show the effect of temperature on the tensile shear strength of a liquid DGEBA cured with
BTDA on acid-etched aluminum substrates. Table 15.11 shows the effect of long-term
thermal aging on the adhesive properties of several epoxy formulations cured with BTDA.
The BTDA-epoxy blends are relatively easy to formulate, and several high-temperature
epoxy adhesive formulations have been developed aimed at specific end properties.
TABLE 15.8 Tensile Shear Strength (Aluminum-Aluminum) of DGEBA Epoxy Resin (EEW: 180–200) Cured with Various Hardeners21
Curing agent Cure conditions Unaged Aged Unaged Aged Unaged Aged Unaged Aged
23 2630
150 1600
260 1200
315 470
TABLE 15.11 Effect of Heat Aging on the Tensile Shear Strength of BTDA-Epoxy Adhesives24
Formulation: 100 phh atomized aluminum, 3 pph Cab-O-Sil. Cure: 2 h at 200°C. Substrate: acid-etched alcad
aluminum.
EFFECT OF THE SERVICE ENVIRONMENT 311
Fillers were found to significantly increase the room temperature tensile shear strength of
BTDA cured epoxy adhesives. However, other modifiers such as phenolic or nylon resin
did not significantly improve the peel strength, flexibility, or temperature resistance of the
unmodified formulation.
Many applications for adhesives and sealants require high strength and durability at low
temperatures. Many of these same applications also require resistance to thermal cycling
between high and low operating temperatures. Unfortunately, the properties of adhesives
and sealants at low temperatures are not as well studied or documented as they are at high
temperatures.
Typical examples of adhesive and sealant applications requiring low-temperature or
thermal cycling performance include
Because of their ease of use and overall good adhesive properties, epoxy structural adhe-
sives often find themselves in these types of applications.
Many of the applications listed above are affected by more than just diurnal (day/night
cycling) or seasonal variations. Outdoor electrical equipment, such as light fixtures, trans-
formers, etc., is subjected to the normal temperature variations that occur during the day or
season; however, it is also exposed to the temperature variations that occur when energiz-
ing and deenergizing the equipment.
Applications in the electrical and electronics industries are often the most severely
stressed due to thermal cycling because of the fast energization and deenergization of the
devices employed. A housing for an outdoor lighting fixture in a shopping center parking
light is a case in point. The design engineer must design the adhesive joint not only for the
maximum temperature difference that occurs during the day and night, but also for the tem-
perature difference that the joint will see when the light is turned on and off. This latter tem-
perature cycling effect could have a degrading influence on the adhesive joint because of
the rate of temperature change and the temperatures involved.
These low-temperature environmental effects can be significant factors that con-
tribute to an adhesive system’s durability and life. This section discusses the character-
istics of epoxy adhesive joints exposed to low temperatures and to thermal cycling and
suggests formulations for improving the resistance of adhesives and sealants to these
conditions.
312 CHAPTER FIFTEEN
Basically, there are two major considerations when one is formulating or selecting adhe-
sives or sealants for low-temperature applications. The first is the effect of the low tem-
perature on the bulk properties of the polymer, and the second is the effect of thermal
cycling and resulting internal stresses on the joint interface.
In essence, for optimal properties at low temperatures, the adhesive joint
1. Must be able to absorb stresses and have a high fracture energy at the service tempera-
tures (perhaps as low as −196°C)
2. Equally important, must be able to resist the transition and cycling from high to low
temperatures
Bulk Polymer Properties. At its service temperature, the adhesive should be tough and
strong. The mechanical energy caused by loading of the joint should be readily distributed
throughout the adhesive. Generally, at very low temperatures, bonds are quite brittle and have
reduced peel and impact strength, resistance to vibrations, and tensile-compressive fatigue life.
Acceptable bonds at reduced temperatures are partially attributed to molecular transi-
tion in the polymer from the plastic to the glassy state. At these transitions the backbone of
the polymer chain can vibrate and actually move in restricted motion. These transitions are
usually given Greek letters with the higher letters representing transitions at lower and
lower temperatures. These are minor transitions compared to the well-known glass transi-
tion temperature Tg at which a polymer’s physical properties change from that of a glass-
like material to that of a tough or leathery material (see Table 3.9). In materials having
low-temperature transitions, polymer chain motion can take place at temperatures far
below the Tg.
A certain amount of molecular chain flexibility is desirable since it imparts resiliency
and toughness to the polymer. This toughness is highest at temperatures around these tran-
sition regions. Therefore, polymers that are most resistant to low temperatures are those
that have transition temperatures in the low-temperature region.
Unmodified epoxy adhesives have moderately high glass transition temperature
depending on the curing agent used and the cure cycle. Therefore, they are generally not
considered good candidates for low-temperature applications or applications where there is
a great amount of thermal cycling.
It can be expected, then, that one of the major problems in adhesives technology is the
development of adhesives that must withstand both elevated temperatures as well as peri-
odic excursions to low temperatures. Several solutions have been developed. Certain adhe-
sive systems, notably blends of epoxy resin with more elastic resins, have been formulated
with a very broad glass transition temperature range or with multiple glass transitions at
both high and low temperatures. These have found some success in the applications dis-
cussed in this chapter.
Flexibilizers and plasticizers can be used to lower the glass transition temperature and
improve the low-temperature bond strength of epoxy adhesives. These will also provide a
degree of elongation when there are differing coefficients of thermal expansion between
the substrates and/or the adhesive. For adjusting the coefficient of thermal expansion, min-
eral or metallic fillers are normally used. With these modifications, good properties often
can be obtained down to about the range of −20 to −40°C. However, optimal properties in
the cryogenic temperature range (less than −100°C) dictate the base polymers recom-
mended in Sec. 15.4.2.
Stresses at the Interface. When adhesive systems are used over a wide temperature range,
the coefficient of thermal expansion becomes quite important in determining residual
EFFECT OF THE SERVICE ENVIRONMENT 313
stresses in a joint. This is most important in low-temperature applications, since the modu-
lus of elasticity generally increases with decreasing temperature and the adhesive is likely
to be less forgiving to stress.
The most significant factors that determine the strength of an adhesive joint when used
over a wide temperature range are the following:
Residual stress resulting from thermal expansion or contraction is due to the differences
in the thermal expansion coefficient between the adhesive and adherend and to temperature
distribution in the joint due to differences in thermal conductivity.
The adhesive’s thermal conductivity is important in minimizing transient stresses dur-
ing cooling. This is why thinner bond thickness and adhesives or sealants with higher lev-
els of thermal conductivity generally have better cryogenic properties.
Other opportunities for stress concentration in bonded joints that may be aggravated by
low-temperature service include trapped gases or volatiles evolved during bonding, resid-
ual stresses in adherends as a result of the release of bonding pressure, and elevated-
temperature cure (i.e., shrinkage and thermal expansion differences).
These internal stresses are magnified when the adhesive or adherend is not capable of
deforming to help relieve the stress. Often these residual stresses are present at room tempera-
ture; however, the adhesive strength and resiliency are sufficient to resist them. On excursions
to low temperatures, the residual stresses become magnified and could lead to bond rupture.
It is necessary that the adhesive retain some resiliency if the thermal expansion coeffi-
cients of the adhesive and adherend cannot be closely matched. At room temperature, a
standard low-modulus adhesive may readily relieve stress concentration by deformation.
At cryogenic temperatures, however, the modulus of elasticity may increase to a point
where the adhesive can no longer effectively release the concentrated stresses. At low ser-
vice temperatures, the difference in thermal expansion is very important, especially since
the elastic modulus of the adhesive generally decreases with falling temperature.
Epoxy-nylon
6
Shear strength, 1000 psi
Polyurethane
Vinyl
phenolic
4 Epoxy-phenolic
Filled
2 epoxy
Epoxy polyamide
Rubber-phenolic
0
−400 −300 −200 −100 0 100 200
Temperature, °F
FIGURE 15.9 Properties of cryogenic structural adhesive systems.25
others because of their brittleness and corresponding low peel and impact strength at cryo-
genic temperatures. On a basis of lap shear strength at low temperatures (below −55°C), the
epoxy formulations are ranked in decreasing order of shear strength as follows: epoxy-
nylon, epoxy-polysulfide, epoxy-phenolic, epoxy-polyamide, and amine cured and anhy-
dride cured epoxy.28
Epoxy-nylon adhesives are among the toughest and strongest adhesives and are usually
produced as a dry B-staged film. Epoxy-nylon adhesives retain their flexibility and provide
5000 psi shear strength in the cryogenic temperature range. They also have useful impact
resistance properties down to −147°C. Peel strengths can be as high as 40 lb/in of width,
and resistance to vibration and fatigue is excellent. However, epoxy-urethane hybrid
8000
Tension
6000
Strength, psi
4000
2000
Shear
0
0 50 100 150 200 250 300
Temperature, K
FIGURE 15.10 Tensile shear strength versus temperature for
aluminum-filled DGEBA epoxy cured with DEAPA.21
TABLE 15.12 Properties of Low-Temperature Structural Adhesives
adhesives are generally noted to have higher peel strength in the cryogenic temperature
range. The epoxy-polyurethane hybrid adhesives are especially promising because of the
low-temperature properties that are provided by the polyurethane constituent.
Other useful epoxies include epoxy-polysulfide, epoxies modified with nitrile butadi-
ene rubber, and epoxy-phenolic. Epoxy-phenolic adhesives are exceptional in that they
have good adhesive properties at both elevated and low temperatures. Sandwich peel for
these systems at −55°C is as high as 12 lb/in, and tensile shear strength is retained in the
range of 3000 psi.29 However, as shown in Fig. 15.9, although the tensile shear strength of
epoxy-phenolic adhesives remains steady over a very broad temperature range, it is not as
great as that for epoxy-nylon or polyurethane adhesives at cryogenic temperatures.
Next to the epoxy-nylons and epoxy-urethanes, the epoxy-polysulfide adhesives show the
greatest lap shear tensile strength at temperatures below 0°C. Bonded etched steel substrates
show a tensile shear strength of 2900 psi at room temperature, and this increases to 3400 psi
when the temperature is reduced to −156°C. Epoxy-nylon and epoxy-polysulfide are the only
adhesives that show an increase in strength as the temperature is significantly reduced.
The good low-temperature properties of epoxy-polysulfide adhesives are one reason
why they have found significant use in the building and construction industry. Another rea-
son is their excellent flexibility and ability to absorb stress and move with the thermal
expansion and contraction of the substrates. However, in these applications the adhesive is
generally not formulated for high tensile shear strength but rather for optimum elongation.
Moisture is the substance that often causes the greatest difficulties in terms of environ-
mental stability for bonded or sealed joints. Water can be an exceptional problem because
it is very polar and permeates most polymers. Other common fluids, such as lubricants and
fuels, are of low or zero polarity and are not as likely to permeate and weaken adhesive or
sealant joints.
Moisture can degrade a cured adhesive joint in three distinctive ways.
• Moisture can degrade the properties of the bulk adhesive or sealant itself.
• Moisture can degrade the adhesion properties at the interface.
• Moisture can also degrade physical properties and cause dimension changes of certain
adherends.
The hostility of certain moisture environments can be seen in Fig. 15.11 Aluminum
joints were bonded with room temperature curing epoxy-polyamide adhesive and aged in
a hot, wet (tropical) environment and in a hot, dry (desert) environment. Excellent dura-
bility is achieved under dry conditions while significant degradation is caused by the wet
conditions.
Ambient moisture can also affect certain types of uncured adhesive, either as it is being
mixed and applied to a substrate or as it is stored in a container waiting to be applied. The
degradation mechanism before cure of the adhesive is discussed in Chap. 3.
Even though epoxy adhesives are insoluble in water, they are not immune to water attack.
Moisture can affect the strength of the epoxy adhesive joint by
EFFECT OF THE SERVICE ENVIRONMENT 317
25
50
75
100
0 1 2 3 4
Exposure time, years
FIGURE 15.11 Effect of outdoor weathering on the strength of aluminum joints
bonded with epoxy-polyamide.30
Moisture degradation of adhesive bonds occurs within the bulk adhesive material, at the
adhesive-adherend interface, and within certain substrates. These degradation mechanisms are
discussed below. Particularly insidious is the effect of the combined elements of moisture,
stress, and temperature. Unfortunately, this synergistic effect occurs at relatively low temper-
atures, and such a service environment is common to many adhesive applications. For these
reasons, this combined environment is given special focus in Sec. 15.5.2.
Effect on the Bulk Property of the Adhesive. Moisture can alter the properties of the bulk
material by changing its glass transition temperature, inducing cracks, or chemically react-
ing with the polymer—a process called hydrolysis. But before these mechanisms occur, the
moisture must first find its way into the bulk polymer.
Internal degradation within the bulk adhesive or sealant occurs primarily by absorption
of water molecules into the polymer structure. All polymers will absorb water to some
extent. Moisture can also enter by wicking along the adhesive-adherend interface or by
wicking along the interfaces caused by reinforcing fibers and the resin. Deterioration may
occur more quickly in a 100 percent relative humidity (RH) environment than in liquid
water because of more rapid permeation of the vapor.
318 CHAPTER FIFTEEN
The water ingress properties of various polymers can be assessed by values of their per-
meability coefficient and the diffusion constant of water (Table 15.13). The permeability
coefficient is defined as the amount of vapor at standard conditions permeating a sample
that is 1 cm2 and 1-cm thickness within 1 s with a pressure difference of 1 cmHg across the
polymer. The diffusion coefficient is a measure of the ease with which a water molecule
can travel within a polymer. There is a wide variation in the maximum amount of water
absorbed by polymeric materials. Certain systems have very low absorption at lower tem-
peratures, but the rate of absorption increases significantly at higher temperatures.
Epoxies, nitriles, and phenolics show relatively low diffusion rates and are less suscep-
tible to moisture attack than most other polymers. As a result, these materials are often used
in adhesive and sealant formulations where resistance to moisture is essential.
Water permeation generally lowers the glass transition temperature of the polymer by
reducing the attractive forces between molecules. Data for certain epoxy adhesives cured
with different curing agents are given in Table 15.14. The effect of absorbed water on the
mechanical properties of cured epoxy adhesives is shown in Table 15.15. Water lowers ten-
sile strength and modulus but increases elongation at break. Since no chemical linkages are
broken, these properties generally recover fully when the polymer is dried unless irre-
versible hydrolysis has taken place.
Some polymeric materials, notably certain anhydride cured epoxies and ester-based
polyurethanes, will chemically change or “revert” when exposed to humid conditions for a
DAPEE* 67 37 49
TETA† 99 86 111
DAB‡ 161 143 157
DDM§ 119 110 130
*Di-(1-aminopropyl-3-ethoxy) ether.
†
Triethylenetetramine.
‡
1,3-Diaminobenzene.
§
4,4′-Diaminodiphenylmethane.
EFFECT OF THE SERVICE ENVIRONMENT 319
*Di-(1-aminopropyl-3-ethoxy) ether.
OH OH
+ H2O
H
H
OH
H
OH
Large molecules Smaller molecules
FIGURE 15.12 The degradation of polymer chains by reaction with water is called
hydrolysis.34
320 CHAPTER FIFTEEN
G
80
F
60
A
40 E D C
B
20
0
0 10 20 30
Exposure time, days at 100°C, 95% RH
FIGURE 15.13 Hydrolytic stability of potting com-
pounds. Materials showing rapid loss of hardness in this test
soften similarly after 2 to 4 years in high-temperature, high-
humidity climate zones.35
Reversion is usually much faster in flexible materials because water permeates them
more easily. Hydrolysis has been seen in certain epoxy, polyurethane, and cyanoacrylate
adhesives. The reversion rate also depends on the type and amount of catalyst used in the
formulations and the degree of crosslinking. Best hydrolytic properties are obtained when
the proper stoichiometric ratio of base material to catalyst is used.
In the case of conventional construction sealants, the polysulfides, polyurethanes, epox-
ies, and acrylics have all shown various degrees of sensitivity to moisture. Hydrolysis
causes the breaking of bonds within the sealant. Thus, the bond strength decreases and
cohesive failure results. However, before this occurs, the sealant usually swells and may
cause deformation or bond failure before hydrolysis can completely take action.
Effect on the Interface. Water can also permeate the adhesive or sealant and preferen-
tially migrate to the interfacial region, displacing the bulk adhesive material at the interface.
This mechanism is illustrated in Fig. 15.14. It is the most common cause of adhesive strength
reduction in moist environments. Thus, even structural adhesives that are not susceptible to
the reversion phenomenon may lose adhesive strength when exposed to moisture.
The degradation curves shown in Fig. 15.15 are typical for an adhesive exposed to
moist, high-temperature environments. The mode of failure in the initial stages of aging
usually is truly cohesive. After 5 to 7 days, the failure becomes one of adhesion. It is
expected that water vapor permeates the adhesive through its exposed edges. The water
molecules are absorbed into the adhesive and preferentially concentrate on the metal
adherend, thereby displacing the adhesive at the interface. This effect is greatly dependent
on the type of adhesive and the adherend material.
Notice that the moisture resistance of the epoxy-amine adhesive is better than that of the
heat-resistant phenolic in Fig. 15.15. The higher initial bond strength of the epoxy is likely
due to better adhesion to metal and less internal stress. But the epoxy-amine adhesive
EFFECT OF THE SERVICE ENVIRONMENT 321
Substrate
External
chemicals
(a)
Substrate
(b)
FIGURE 15.14 Competition between an adhesive and other chemicals for surface sites
leading to displacement of the adhesive from the surface. (a) Adhesive adsorbed at surface
sites; (b) adhesive displaced from the surface sites.36
shows a more rapid rate of decline in bond strength on exposure to the humid environment.
This is likely due to the epoxy adhesive’s higher degree of water absorption.
Certain adhesive systems are more resistant to interfacial degradation by moist environ-
ments than are other adhesives. Table 15.16 illustrates that a nitrile-phenolic adhesive does not
succumb to failure through the mechanism of preferential displacement at the interface. Failures
occurred cohesively within the adhesive even when tested after 24 months of immersion in
water. A nylon-epoxy adhesive bond, however, degraded rapidly under the same conditioning
owing to its permeability and preferential displacement by moisture.
Shear
6000 Aromatic amine/epoxy
Tension
Shear
4000
2000
0
0 4 8 12 16 20
Exposure time, days
FIGURE 15.15 Effect of humidity on adhesion of two
structural adhesives to stainless steel.37
322 CHAPTER FIFTEEN
TABLE 15.16 Effect of Humidity and Water Immersion on the Shear Strength of Two
Structural Adhesives38
Substrate is aluminum treated with a sulfuric acid–sodium dichromate etch.
The shape of the curve showing rate of strength degradation in Fig. 15.15 is common
for most adhesives being weakened on exposure to wet surroundings. Strength falls most
rapidly at the beginning of the aging process and then slows down to a low or zero rate of
degradation. The initial rate and overall percent of degradation will vary with the adhesive
and surface treatment.
There also appears to exist a critical water concentration within the adhesive below
which water-induced damage of the joint will not occur. This also infers that there is a crit-
ical humidity for deterioration. For an epoxy system, it is estimated that the critical water con-
centration is about 1.35 to 1.45 percent and that the critical humidity is 50 to 65 percent.39,40
Any loss in joint strength by the absorbed water can be restored upon drying if the equilib-
rium moisture uptake is below the critical water concentration.
Another way moisture can degrade the strength of adhesive joints is through hydration
or corrosion of the metal oxide layer at the interface. Common metal oxides, such as aluminum
and iron, can undergo hydration. The resulting metal hydrates become gelatinous, and they
act as a weak boundary layer because they exhibit very inadequate bonding to their base
metals. Thus, the adhesive or sealant used for these materials must be compatible with the
firmly bound layer of water attached to the surface of the metal oxide layer.
Effect of the Bulk Substrate. Certain substrates, notably wood but also other materials
such as laminates and certain plastics, will change dimensions significantly when exposed
to variations in ambient relative humidity or moisture. Wood is an anisotropic material, so
dimensional change will be greater in one direction than in another. Such change with rel-
ative humidity can result in large internal stress on the joint and even warpage of the assem-
bly. The adhesive must be selected to withstand these dimensional changes.
Maximum bond performance and minimal internal stresses are sometimes achieved if
the substrate has a moisture content during bonding that is close to the average moisture
content anticipated during service—provided, of course, that the moisture retained in the
substrate does not adversely affect the initial bond strength. This may require precondi-
tioning of the substrates in a controlled environment before bonding.
The interaction of temperature and moisture causes greater degradation than can be attrib-
uted to either environment by itself. There is evidence to suggest that exposure to humid
EFFECT OF THE SERVICE ENVIRONMENT 323
environments, at temperatures above about 60°C, can produce permanent damage in epoxy
adhesives.41 This can be explained by the formation of microcavities produced by clusters
of water molecules and due to possible hydrolytic reactions.
Mechanical stress accelerates the effect of environment on the adhesive joint. A great
amount of data is not available on this phenomenon for specific adhesive systems because
of the time and expense associated with stress-aging tests. However, it is known that mois-
ture, as an environmental burden, markedly decreases the ability of an adhesive to bear pro-
longed stress, especially at slightly elevated temperatures.
This stress-induced effect was first noted in the 1960s, when stressed and nonstressed
aluminum lap shear joints were aged in a natural weathering environment in Florida.42
Stress was applied by flexural bending of lap shear samples and keeping them in that state
during the aging period. Depending on the type of adhesive, there was a significant degra-
dation after 1 to 2 years due to stress-weathering, whereas all of the joints that were aged
in the nonstressed condition showed little degradation.
Joints made with a flexible adhesive having a low glass transition temperature fail by
creep of the adhesive at relatively short service times. Figure 15.16 illustrates the effect of
stress-aging on specimens exposed to humidity cycling from 90 to 100 percent and simul-
taneous temperature cycling from 25 to 50°C. The loss of load-bearing ability of a flexibi-
lized epoxy adhesive is great. The stress on this particular adhesive had to be reduced to 13
percent of its original static strength in order for the joint to last a little more than 44 days
in the high-temperature, high-humidity environment.
6000
Joint strength at 734°F, 50% RH
5000
Applied stress, psi
4000
3000
2000
Aluminum adherends
1000 Stainless steel adherends
0
1 10 100
Time to failure, days
(a)
4000
Applied stress, psi
0
1 10 100
Time to failure, days
(b)
FIGURE 15.16 Time to failure versus stress for two
adhesives in a warm, high-humidity environment. (a) One
part, heat cured modified epoxy adhesive. (b) Flexibilized
amine cured epoxy adhesive.43
324 CHAPTER FIFTEEN
Strength retention, %
Saltwater spray
Florida (3 years) Panama (3 years) (30 days)
The 8 × 9 in panels stressed by mounting on a steel bending frame to get 0.25-in deflection at the center of a 6-in
span; 1/2-in joint overlap is at center of span.
Table 15.17 shows the adverse effect of stress and tropical climates on aluminum joints
bonded with various adhesives. If a 30-day saltwater spray were used as an accelerated test
to determine the long-term performance of these adhesives in a tropical climate, it would
be very misleading. Saltwater spray had very little effect on the strength of stressed or
unstressed joints with the exception of one acrylic adhesive. However, the stressed speci-
mens in Florida almost all completely degraded. Panama was not nearly as severe an envi-
ronment. These data illustrate the point that permanence or durability must be tested in the
specific environment.
Several sources seem in general agreement as to the relative durabilities of structural
adhesives. Results of sustained load laboratory testing and outdoor weathering studies pro-
vide the same order for the durabilities of different adhesive classes. These are summarized
in Table 15.18. However, such a ranking should be taken only as a general guide, since
other factors can also affect the performance of an adhesive or sealant joint. A comparison
of the sustained load durability of various classes of one-component epoxy adhesives is
given in Table 15.19. The results obtained for two-part room temperature curing epoxies
under the same conditions are significantly inferior.
It has also been shown in several studies that the combination of stress, temperature, and
moisture can accelerate the hydrolytic instability of certain epoxy adhesives. In an FTIR
study of the effect of moisture on DGEBA epoxy cured with nadic methyl anhydride, spectra
changes were observed in stressed specimens aged for 155 days at 80°C and 100 percent RH.47
This was attributed to the slow, stress-induced hydrolysis of ester groups. In another study,
Most durable
175°C cured film (nitrile-phenolic, vinyl-phenolic, novolac-epoxy)
120°C cured film (modified epoxy, nitrile-epoxy, nylon-epoxy)
Heat cured paste (nitrile-epoxy, nylon-epoxy, vinyl-epoxy)
Least durable
Room temperature cured paste (epoxy-polyamide, epoxy-anhydride)
EFFECT OF THE SERVICE ENVIRONMENT 325
TABLE 15.19 Effect of Adhesives on the Sustained Load Durability of FPL-Etched Aluminum
Specimens46
Epoxy novolac 1550 60°C, 100% RH All samples failed before 2 years
900 No failures after 9 years
Epoxy-nitrile 1500 60°C, 100% RH All samples failed before 2 months
900 All samples failed before 6 months
Modified epoxy 1500 60°C, 100% RH All samples failed before 2 years
600 No failures after 7 years
Epoxy 800 38°C, 100% RH Partial sample population failed
before 8 years
400 No failures after 8 years
Resistance of adhesive joints to moisture can be improved either by preventing water from
reaching the interface or by improving the durability of the interface itself. Several meth-
ods of minimizing degradation are possible.
1. Selection of a base polymer having a low water permeability and diffusion coefficient
2. Chemically modifying the adhesive or sealant to reduce water permeation
3. Incorporating inert fillers into the adhesive to lower the volume that can be influenced
by moisture
4. Coating the exposed edges of the joint with very low-permeability resins
5. Using primers or chemically treating the substrate surface to improve adhesion and thus
protect the interface from the intrusion of water
6. Chemically altering the substrate surface prior to bonding to provide for better adhesion
and corrosion protection
Both the adhesive formulator and the end user can help to mitigate the degradation
effects of moisture on the adhesive bond. The formulator can do this through proper selec-
tion of a base polymer along with additives and modifiers that will inhibit moisture ingress,
and if moisture penetration does occur, it will not result in an irreversible degradation. The
end user can prevent or delay degradation due to moisture by using the proper curing con-
ditions and taking steps to protect the substrate surface (both before and after application
of the adhesive) and interface.
Addressing the Bulk Adhesive. All polymers absorb moisture to some extent and in
doing so become plasticized by the water molecules. The bulk properties are changed: glass
transition temperature, tensile strength, and modulus are lowered, and elongation is increased.
326 CHAPTER FIFTEEN
In sealants, swelling and deformation are also noted. These properties generally recover on
drying unless hydrolysis takes place. Even if the properties completely recover, the migra-
tion of moisture into the adhesive and possible preferential accumulation at the joint inter-
face can result in loss of adhesion.
To improve moisture resistance, the formulator generally must operate on the bulk
adhesive. This will occur mainly through modification or change in the bulk polymer and
somewhat by modification or change of the fillers and additives in the formulation.
Base Polymer. The water uptake properties of polymers can be assessed by immers-
ing films in water and recording increases in weight. The diffusion coefficient can be
obtained from such data.49 Values of the diffusion coefficient were given in Table 15.13.
There is a wide variation in the maximum amount of water absorbed by polymeric resins.
Certain systems have a very low absorption at the lower temperatures, but this breaks down
at higher temperatures.
It is apparent that the first step toward formulation or selection of a moisture-resistant
adhesive is to choose a base polymer that has low diffusion coefficient and permeability to
water. This helps in two ways.
• It reduces the rate of diffusion of moisture to the critical interphase between the substrate
and the adhesive.
• It reduces the effect on the bulk properties of the adhesive.
There have been many investigations to determine the best chemical structures to pro-
vide for resistance to moisture and hydrolysis. Attempts have been made to synthesize
epoxy adhesives with improved water resistance by replacing some hydrogen atoms by
fluorine.50 However, the cost of processing of such materials has restricted commercial
development. For electronic sealants, it is highly desirable to keep moisture from penetrat-
ing into critical areas. Hydrophobic polymers have been developed to accomplish this task.
They are siloxyimides, fluorosilicones, fluoroacrylics, phenylated silicone, and silastyrene.
Other things being equal, water permeates fastest through flexible polymers. Hence the
moisture pickup is generally much faster for flexible compounds than for more rigid types.
Unfortunately those polymers that provide the best resistance to ingress of moisture tend to
be rigid, highly crosslinked systems. They form brittle adhesives with relatively low peel
and impact strengths. Microvoids can also be formed within the polymer by clusters of
water molecules, and a mechanism of damage is evident in thermoset resins in which the
rigid crosslinked structure does not allow the matrix to relax after microvoid formation.
Most structural adhesives are, therefore, formulated to provide the best compromise
between environmental resistance and the desired mechanical properties. Experience has gen-
erally revealed that although the moisture ingress of the adhesive or sealant does affect the
durability, it is seldom the dominant factor. Generally, of greater importance is how the mois-
ture influences the adhesive-adherend interface region. Table 15.16 summarizes the moisture
resistance and performance properties of some of the more common structural adhesives.
There is great variation within types of adhesives because of differences in chemical
linkages, formulation parameters, crosslinking density, etc. For example, the room tem-
perature curing two-part epoxy adhesives are usually considered to have a lower level of
performance than the heat cured counterparts. However, investigators have shown that cer-
tain two-part, room temperature curing epoxy adhesives can demonstrate excellent dura-
bility even after 12 years of tropical exposure.51,52 The performance of many two-part
systems can be improved by a heat treatment following room temperature cure to optimize
crosslinking.
Extensive information on durability of adhesive joints is more available on aluminum than
on other substrates. Figure 15.17 illustrates typical results showing the effect of adhesive vari-
ations on joint durability during marine exposure. Vinyl-phenolics and nitrile-phenolics have
EFFECT OF THE SERVICE ENVIRONMENT 327
6000
4000
4
3000
3
2000
5
1000 2
1
1 2 3 4 5 6
Exposure time, years
FIGURE 15.17 Effect of adhesive on the durability of etched 6061-T6
aluminum alloy joints exposed to a marine environment. (1) Two-part
epoxy, (2) one-part epoxy, (3) nitrile-phenolic, (4) vinyl-phenolic,
(5) vinyl-phenolic.53
an excellent history of joint durability and rank among the most resistant to environmental
deterioration. However, the current trend is to use epoxy-based adhesives, which provide
easier processing and higher initial strength.
Certain chemical linkages are susceptible to hydrolytic attack and, if present in an adhe-
sive or sealant, are potential sites for irreversible reaction with water that has diffused into
the joint. Such hydrolytic (chemical) degradation causes a permanent reduction in the cured
physical properties. The functional groups present in the chains are hydrolyzed, resulting
in both chain breaking and loss of crosslinking.
The hydrolytic attack on an ester linkage in the presence of water at high pH is an impor-
tant example of this mode of degradation. The attack initiates on the electron-deficient
atom in a highly polarized bond, as on carbonyl carbon:
O ζ− O O
OH−
R1 C O R2 [R1 C OR2] R1 C O− + R2OH
H2O
:
ζ+
HO−
Substitutions of electron withdrawing groups for the aliphatic R1 and R2 groups will
delocalize the charge on the carbonyl carbon, leading to reduced rates of hydrolysis. Thus,
hydrolytic stability increases in the order: aliphatic ester groups < aromatic ester groups <
urethane groups < aliphatic amide < urea groups < aromatic amide groups.54
Since the reaction between an epoxy resin and an acid anhydride curing agent also pro-
duces an ester linkage, anhydride cured epoxies have poorer hydrolytic stability than do
328 CHAPTER FIFTEEN
amine cured epoxies. It has also been shown that reversion rates of urethanes and anhydride
cured epoxies increase as the amount of tertiary amine or other base catalyst increases.55
Other things being equal, the rate of hydrolytic attack on any adhesive increases rapidly
as the crosslinking density decreases. Epoxy resins cured with flexibilizing anhydrides,
derived from long-chain aliphatic acids, will hydrolyze rapidly. Epoxy resins cured with
short-chain, highly functional acid anhydrides such as methyl nadic anhydride yield a rigid
network having a much lower permeability for water and a greatly reduced rate of hydrolytic
attack under alkaline conditions.
Many attempts have been made to develop epoxy adhesive systems with outstanding
hydrophobicity to improve their moisture resistance. Barrie, Johncock, and coworkers con-
sidered the halogenation of tetraglycidyl 4,4′-diaminodiphenylmethane resin cured with a
diaminodiphenylsulfone (DDS).56 Goobich and Marom57 investigated the effect of intro-
ducing bromine into an epoxy resin formulation employing a brominated coreactant into a
system similar to that investigated by Barrie. These modifications were shown to reduce
water uptake; however, glass transition temperature and the processability of the adhesive
were negatively affected.
Perhaps the greatest success at hydrophobic epoxy adhesive development was the
development of fluorinated epoxies together with compatible curing agents. These adhe-
sives resulted in equilibrium water concentrations as low as 0.2%. However, for various
reasons these adhesives have not been commercialized.
Additives. Formulation additives do not always have a positive effect on moisture
resistance. Therefore, selection of additives must go through a similar assessment process
as the base polymer with respect to hydrolytic stability. Any polymeric additive used to
modify the properties of the base resin should be considered for its possible effect on mois-
ture resistance and environmental durability.
Sometimes filled adhesives will show better resistance to moisture resistance than unfilled
adhesives simply because incorporating inert fillers into the adhesive lowers the organic vol-
ume that can be affected by moisture. Aluminum powder seems to be particularly effective,
especially on aluminum substrates. The filler can provide a reduction of shrinkage on cure, a
reduction of the thermal expansion coefficient, and a reduction of the permeability to water
and other penetrants. However, fillers do not always produce more durable bonds.
Many film adhesives have a supporting carrier or reinforcement fabric incorporated into
the adhesive to improve handling of the film and provide control of bond line thickness.
The carriers are usually glass, polyester, or nylon fabrics of knitted, woven, or nonwoven
construction. The difficulty with such carriers is that they can provide an effective way of
moisture entering the bulk of the adhesive. Moisture can wick along the fiber-adhesive
interface. Nylon carriers should especially be reviewed since they have a strong tendency
to absorb moisture.
When fibrous carriers or fillers are needed in structural adhesive formulation, coupling
agents are often used to provide for better bonding of the base polymer matrix to the fiber,
to minimize the potential for moisture wicking along the fiber surface. Coupling agents,
either formulated into the adhesive or coated onto the carrier surface, are designed to bond
to hydroxyl groups on the fiber’s surface and to be reactive toward the appropriate bulk
polymer matrix. These coupling agents are generally organosilanes.
Addressing the Interface. Methods that are available to improve the durability of the
interface itself mainly center on surface preparation of the substrate and/or the use of
primers or coupling agents. These actions are generally taken by the end user, and they are
usually most beneficial on metallic surfaces that are prone to degradation by corrosion.
However, simple steps, such as coating the exposed edges of the joint with sealants having
very low permeabilities to inhibit wicking of moisture along the interface, have also pro-
duced positive results.
EFFECT OF THE SERVICE ENVIRONMENT 329
Strong chemical bonds between the adhesive and adherend help stabilize the interface
and increase joint durability. Aluminum joints formed with phenolic adhesives generally
exhibit better durability than those with epoxy adhesives. This is partially attributable to
strongly interacting phenolic and aliphatic hydroxyl groups that form stable primary chem-
ical bonds across the interface.
Primers. Primers tend to hinder adhesive strength degradation in moist environments
by providing corrosion protection to the adherend surface. A fluid primer that easily wets
the interface presumably tends to fill in minor discontinuities on the surface. Organosilane,
organotitanate, and phenolic primers have been found to improve the bond strength of
many adhesive systems.
Silanes and other coupling agents can be applied to various substrates or incorporated
into an adhesive primer to serve as hybrid chemical bridges to increase the bonding between
organic adhesive and inorganic adherend surfaces. Such bonding increases the initial bond
strength and also stabilizes the interface to increase the durability of the resulting joint.
Some primers will improve the durability of the joint by protecting the substrate surface
area from hydration and corrosion. These primers suppress the formation of weak bound-
ary layers that could develop during exposure to wet environments. Primers that contain
film-forming resins are sometimes considered interfacial water barriers. They keep water
out of the joint interface area and prevent corrosion of the metal surfaces. By establishing
a strong, moisture-resistant bond, the primer protects the adhesive-adherend interface and
lengthens the service life of the bonded joint.
However, moisture can diffuse through any polymeric primer, and eventually it will
reach the interface region of the joint. Therefore, the onset of corrosion and other degrada-
tion reactions can only be delayed by the application of a primer unless the primer contains
corrosion inhibitors or it chemically reacts with the substrate to provide a completely new
surface layer that provides additional protection.
Representative data are shown in Fig. 10.5 for aluminum joints bonded with an epoxy
film adhesive and a standard chromate-containing primer. Until recently, standard corrosion-
resistant primers contained high levels of solvent, contributing to high levels of VOCs, and
chromium compounds, which are considered to be carcinogens. As a result, development
programs have been conducted on waterborne adhesive primers that contain low VOC
levels and low or zero levels of chromium.
Epoxy-based primers are commonly used in the aerospace and automotive industries.
These primers have good chemical resistance and provide corrosion resistance to aluminum
and other common metals. Polysulfide-based primers have been developed for applications
where a high degree of elongation is necessary. These systems are used where the joint is
expected to encounter a high degree of flexing or thermal movement.
Resins, curing agents, and additives used in primer formulation are much like adhesive
or sealant formulations except for the addition of solvents or low-viscosity resins to pro-
vide a high degree of flow.
Chemical Surface Modification. In considering the interface, one must contemplate
not only the possibility of moisture disrupting the bond but also the possibility of corrosion
of the substrate. Corrosion can quickly deteriorate the bond by providing a weak boundary
layer before the adhesive or sealant is applied. Corrosion can also occur after the joint is
made and, thereby, affect its durability. Mechanical abrasion or solvent cleaning can pro-
vide adhesive joints that are strong in the dry condition. However, this is not always the
case when joints are exposed to water or water vapor. Resistance to water is much improved
if metal surfaces can be treated with a protective coating before being bonded.
Approaches for the development of water-resistant surface treatments include applica-
tion of inhibitors to retard the hydration of oxides or the development of highly crystalline
oxides as opposed to more amorphous oxides. Standard chemical etching procedures,
which remove surface flaws, also result in improved resistance to high humidity.
330 CHAPTER FIFTEEN
50
30
20
Grit-blasted
10 Solvent-degrease
0
0 500 1,000 1,500
Time in water at 50°C, h
FIGURE 15.18 Effect of surface pretreatment on the durability of aluminum-
epoxy joints subjected to accelerated aging in water at 50°C.59
EFFECT OF THE SERVICE ENVIRONMENT 331
Epoxy adhesives and sealants are generally not significantly affected by simple outdoor
weathering. However, there are certain circumstances that could affect the permanence of
joints exposed to outdoor service. It is important that these be considered early in the design
of the adhesive joint and selection of materials.
The outdoor durability of epoxy bonded joints is very dependent on the type of epoxy
adhesive, specific formulation, nature of the surface preparation, and specific environmen-
tal conditions encountered in service. The data shown in Fig. 15.19, for a two-part room
temperature cured polyamide epoxy adhesive with a variety of fillers, illustrates the differ-
ences in performance that can occur due to formulation changes. Excellent outdoor dura-
bility is provided on aluminum adherends when chromic-sulfuric acid etch or other
chemical pretreatments are used.
Differences in performance are also noted depending on the specific nature of the envi-
ronment. Differences can be expected in joint durability when bonds are exposed to an
aggressive wet-freeze-thaw cycle, marine seacoast, or inland environments. Outdoor
weathering conditions are often classified by one of the following exposure conditions:
3000 Aluminum
Shear strength, psi
Carbonate
2000
Silica
Unfilled
1000
0 8 16 24 32
Exposure time, months
FIGURE 15.19 Durability of polyamide-epoxy two-part adhesives exposed
to marine atmosphere (fillers are indicated).60
332 CHAPTER FIFTEEN
The factors that influence the durability of epoxy adhesives in outdoor weathering con-
ditions are
Because these factors vary so widely over the earth’s surface, the weathering of adhesives
is not an exact science. Furthermore, most materials are weathered by a combination of
these factors, and the contribution of each is dependent on the adhesive formulation.
By far, the most detrimental factors influencing adhesives aged in a nonseacoast environ-
ment are heat and humidity. The reasons why warm, moist climates degrade many adhesive
joints were presented in the last section. Near the seacoast, corrosion due to salt water and salt
spray must also be considered when one is designing an adhesive joint. Thermal cycling due
to weather, oxygen, ultraviolet radiation, and cold are relatively minor factors with most
structural adhesives.
Extensive information on the outdoor durability of bonded aluminum joints is available
in the reviews of Minford.61,62,63 Hartshorn has also provided a catalog of references to out-
door weathering of structural adhesives by adhesive class and exposure conditions.64
When exposed to weather, structural adhesives may rapidly lose strength during the first 6
months to 1 year. After 2 to 3 years, however, the rate of decline usually levels off at
strength that is 25 to 30 percent of the initial joint strength, depending on the climate zone,
adherend, adhesive, and stress level. Adhesive systems that are formulated specifically for
outdoor applications show little strength degradation over time in a moderate environment.
Figure 15.20 shows the weathering characteristics of unstressed epoxy adhesives to the
Richmond, Virginia, climate.
4000 EC−2158 (3 M)
Overlap shear strength, psi
EA−907 (HYSOL)
M 5592 (TREMCO)
3000 34 (U.S.S.)
2000
1000
0
0 4 8 12 16 20 24 28 32 36
Exposure time, months
FIGURE 15.20 Effect of outdoor weathering on typical aluminum joints made with four different
two-part epoxies cured at room temperature.65
EFFECT OF THE SERVICE ENVIRONMENT 333
The following generalizations are of importance in designing a joint for outdoor service:
1. The most severe locations are those with high humidity and warm temperatures.
2. Stressed panels deteriorate more rapidly than unstressed panels.
3. Stainless steel panels are more resistant than aluminum panels because of corrosion.
4. Heat cured adhesive systems are generally more resistant than room temperature cured
systems.
5. For the better adhesives, unstressed bonds are relatively resistant to severe outdoor
weathering, although all joints will eventually exhibit some strength loss.
For most adhesive bonded metal joints that must see outdoor service, corrosive environ-
ments are a more serious problem than the influence of moisture. The degradation mecha-
nism is corrosion of the metal interface, resulting in a weak boundary layer. Surface
preparation methods and primers that make the adherend less corrosive are commonly
employed to retard the degradation of adhesive joints in these environments.
The quickest method to measure corrosion-dominated degradation in bonded metal
joints is to store the bonded parts in a salt spray chamber with a continuous 5% NaCl salt
solution spray in 95 percent RH at 35°C. This is a procedure described in ASTM B 117.
Usually only several hundred hours of exposure are needed to show significant differences
in corrosion resistance of various adhesive joint systems. Figure 15.21 shows the shear
25 3.6
Unaged
After 200-h salt spray test
Standard deviation
20 2.9
Lap shear strength, MPa
Sandblasted
10 1.5
Degreased
Oiled
Degreased
Sandblasted
Sandblasted
Sandblasted
Sandblasted
Sandblasted
5 0.7
Degreased
Degreased
Degreased
Degreased
Ground
Ground
Ground
Ground
Ground
Ground
Oiled
Oiled
Oiled
Oiled
Oiled
0 0
(a) (b) (c) (d) (e) (f)
FIGURE 15.21 Shear strength of mild steel joints. (a) two-part epoxy, (b) two-part acrylic, (c) anaero-
bic acrylic, (d) cyanoacrylate, (e) PVC plastisol, ( f ) one-part heat cured epoxy.68
strength of mild steel joints with various adhesives before and after exposure for 200 h in
a salt spray test.
Resistance of the adhesive joint to salt climates depends not only on the type of adhe-
sive but also on the method of surface preparation and on the type of primer used. The good
bond durability in saltwater exposure of anodized surface pretreated joints has been shown
by several studies.69
Room temperature curing, two-part epoxy adhesives are usually considered to have a lower
level of outdoor performance than the heat cured adhesives. Their performance, however,
can be improved by heat treatment following the room temperature cure.
Epoxy adhesive cured with diethylene triamine, a primary amine, tends to be somewhat
water-sensitive. However, under conditions of ambient laboratory storage, there is no sig-
nificant change in bond strength over a period of 11 years.70 Unfilled polyamide cured
epoxy adhesive survives 11 years of room temperature storage at more than double the
shear strength of diethylene triamine cured epoxy with no significant decrease in strength.
The initial value of 2408 psi leveled off at about 2900 psi within a few months and for the
remainder of the 11 years.
Polyamide cured epoxy adhesives were also found to survive up to 4 years’ exposure
to industrial atmosphere, high-humidity, and water immersion environments without any
appreciable bond degradation.71 However, the strength of these adhesives were noticed to
increase from 1500 to 1590 psi during aging for 3 years in a jungle environment of
Panama.72 Elevated-temperature curing, one- and two-part epoxy adhesives generally
EFFECT OF THE SERVICE ENVIRONMENT 335
have a higher level of durability when exposed to aggressive outdoor environments. These
adhesives generally show good performance when sustained loads are less than 50 percent
of the initial bond strength.
Diethylaminopropylamine (DEAPA) was studied extensively as a curing agent for
epoxy adhesives exposed to outdoor weathering. No changes were noticed during 11 years
of ambient laboratory storage, or during 24 weeks of exposure in environments of 50°C,
100 percent RH, and thermal cycling between −18 and 50°C. When exposed to more rig-
orous environments, the bond strength degraded.
Many organic adhesives tend to be susceptible to chemicals and solvents, especially at ele-
vated temperatures. Most standard tests to determine chemical resistance of adhesive joints
last only 30 days or so. Unfortunately, exposure tests lasting less than 30 days are not
applicable to many service life requirements. Practically all adhesives are resistant to these
fluids over short time periods and at room temperatures. Some epoxy adhesives even show
an increase in strength during aging in fuel or oil over these time periods. This effect is pos-
sibly due to either postcuring or plasticizing of the epoxy by the oil.
There are two properties of adhesive joints that protect them from exposure to chemical
or solvent environments: high degree of crosslinking and low exposure area.
Adhesive joint designers will take maximum advantage of this second effect by designing
the joint configuration for protection or by specifying a protective coating and/or sealant
around the exposed edges of the adhesive.
Figure 15.22 shows the long-term effect of a heat cured one-part epoxy adhesive to var-
ious chemical environments. As can be seen, the temperature of the immersion medium is
a significant factor in the aging properties of the adhesive. As the temperature increases, the
adhesive generally adsorbs more fluid, and the degradation rate increases.
Epoxy adhesives are generally more resistant to a wide variety of liquid environ-
ments than other structural adhesives. However, the resistance to a specific environment
is greatly dependent on the type of epoxy curing agent used. Aromatic amine (e.g.,
metaphenylene diamine) cured systems are frequently preferred for long-term chemical
resistance.
There is no universally “best adhesive” for all chemical environments. As an example,
maximum resistance to bases almost axiomatically means poor resistance to acids. It is rel-
atively easy to find an adhesive that is resistant to one particular chemical environment. It
becomes much more difficult to find an adhesive that will not degrade in two widely dif-
fering chemical environments. Generally, adhesives that are most resistant to high temper-
atures have good resistance to chemicals and solvents because of their dense, crosslinked
molecular structure.
336 CHAPTER FIFTEEN
4000 A
B
3000
C
2000
1000 D
E
0
0 4 8 12 16 20 24 28
Exposure time, months
FIGURE 15.22 Effect of immersion in various chemical environments
on a one-part heat curing epoxy adhesive (EA 929, Hysol Corp.): (A) gaso-
line at 23°C, (B) gear oil at 120°C, (C) distilled water at 23°C, (D) tap
water at 100°C, (E) 38% Shellzone at 120°C.73
From the information reported in the literature with regard to aging of adhesive joints
in chemical environments, it can be summarized that
1. Chemical resistance tests are not uniform, in concentrations, temperature, time, or prop-
erties measured.
2. Generally, chlorinated solvents and ketones are severe environments.
3. High-boiling-point solvents, such as dimethylforamide and dimethyl sulfoxide, are
severe environments.
4. Acetic acid is a severe environment.
5. Amine curing agents for epoxies are poor in oxidizing acids. Anhydride curing agents
are poor in caustics.
TABLE 15.21 Standard Test Fluids and Immersion Conditionals for Adhesive
Evaluation per MMM-A-132
chemicals. Standard chemical reagents are listed in ASTM D 543, and the standard oils and
fuels are given in ASTM D 471. Standard test fluids and immersion conditions used by
many adhesive suppliers and specified in MMM-A-13274 are listed in Table 15.21.
15.9 RADIATION
140
Tensile shear strength, % of unirradiated strength
120
100
80
60 Vinyl-phenolic
Nitrile-phenolic
Vinyl-phenolic
40 Neoprene-phenolic
Nylon-phenolic
Modified epoxy
20 Nitrile-phenolic
Epoxy
Epoxy-phenolic Adherends-2024−T6 aluminum
0
0 10 30 50 70 90 110 130
Radiation dosage, 107 rad
FIGURE 15.23 Percent change in initial tensile shear strength caused by nuclear radiation dosage.76
Thread locking grades of anaerobic adhesives have sustained 2 × 107 rads without molec-
ular change or loss of locking torque.77
In an early study by McCrudy and Rambosek78, radiation does not appear to have seri-
ous effects on the tensile shear strength of highly crosslinked adhesives. Nitrile-phenolic
adhesives are somewhat more resistant to radiation damage than epoxy-based adhesives.
The study also showed that thick adhesive layers retain useful strength better than thin glue
lines. Ten mils is recommended as the minimum glue line thickness when radiation is an
environmental factor.
REFERENCES
1. Waggemans, D. M., “Adhesives Charts,” in Adhesion and Adhesives, vol. 2, Elsevier, Amsterdam,
1967.
2. Hartshorn, S. R., “The Durability of Structural Adhesive Joints,” Structural Adhesives, S. R.
Hartshorn, ed., Plenum Press, New York, 1986.
3. Lewis, A. F., and Gounder, R. N., “Permanence of Structural Adhesive Joints,” Treatise on
Adhesion and Adhesives, vol. 5, R. L. Patrick, ed., Marcel Dekker, New York, 1988.
4. Levi, D. W., Journal of Applied Polymer Science: Applied Polymer Symposium, vol. 32, 1977, p. 189.
5. Levi, D. W., et al., “Effect of Titanium Surface Pretreatment on Adhesive Bonds,” SAMPE
Quarterly, April 1976.
EFFECT OF THE SERVICE ENVIRONMENT 339
6. Jones, W. C., et al., “Use Multiple Regression Analysis to Develop Preductive Models for Failure
Times of Adhesive Bonds at Constant Stress,” Journal of Applied Polymer Science, vol. 18, 1974,
p. 555.
7. Bolger, J. C., and Michaels, A. S., “Molecular Structure and Electrostatic Interactions at
Polymer–Solid Interfaces,” in Interface Conversion for Polymer Coatings, P. Weiss, ed., Elsevier,
New York, 1969.
8. Lewis and Gounder, “Permanence of Structural Adhesive Joints.”
9. Lee, H., and Neville, K., Handbook of Epoxy Resins, McGraw-Hill, New York, 1967.
10. Krieger, R. B., and Politie, R. E., “High Temperature Structural Adhesives,” in Aspects of
Adhesion, vol. 3, D. J. Alner, ed., University of London Press, London, 1967.
11. Kausen, R. C., “Adhesives for High and Low Temperatures,” Materials Engineering,
August–September 1964.
12. Krieger and Politi, “High Temperature Structural Adhesives.”
13. Black, J. M., and Bloomquist, R. F., “Metal Bonding Adhesives for High Temperature Service,”
Modern Plastics, June 1956.
14. Wilkinson, S. P., et al., “Reactive Blends of Amorphous Functionalized Engineering
Thermoplastics and Bismaleimide/Diallyl Bisphenol A Resins for High Performance Composite
Matrices,” Polymer Preprints, vol. 33, no. 1, 1992, p. 425.
15. Shimp, D. A., et al., “Co-Reaction of Epoxide and Cyanate Resins,” 33d SAMPE Symposium and
Exhibition, Anaheim, CA, Mar. 7–10, 1988.
16. Burgman, H. A., “The Trend in Structural Adhesives,” Machine Design, Nov. 21, 1963.
17. Lee and Neville, Handbook of Epoxy Resins, pp. 21.43–44.
18. Bolger, J. C., “Structural Adhesives for Metal Bonding,” in Treatise on Adhesion and Adhesives,
vol. 3.
19. Meath, E. R., “Epoxy Resin Adhesives,” Chapter 19 in Handbook of Adhesives, I. Skeist, ed., van
Nostrand Reinhold, New York, 1990, p. 348.
20. Hourwink, R., and Salmon, G., eds., Adhesion and Adhesives, 2d ed., Elsevier, Amsterdam, 1986,
p. 272.
21. Licari, J. J., “High Temperature Resistant Adhesives,” Product Engineering, December 1964, p. 104.
22. Licari, “High Temperature Resistant Adhesives,” p. 105.
23. Barie, W. P., and Franke, N. W., “High Temperature Epoxy Adhesives Based on BTDA,”
Adhesives Age, October 1971, pp. 36–39.
24. Barie and Franke, “High Temperature Epoxy Adhesives Based on BTDA.”
25. Kausen, R. C., “Adhesive for High and Low Temperatures—II,” Materials in Design
Engineering, September 1964, pp. 108–112.
26. McClintock, R. M., and Hiza, M. J., “Epoxy Resins as Cryogenic Structural Adhesives,” Modern
Plastics, June 1958.
27. “Low Temperature Strength of Epoxy Resin Insulation Adhesive,” Insulation, December 1958.
28. Miska, K. H., “Which Low Temperature Adhesive Is Best for You?” Materials Engineering, May
1975.
29. Kausen, “Adhesive for High and Low Temperatures—II.”
30. Sung, N. H., “Moisture Effect on Adhesive Joints,” Adhesives and Sealants, vol. 3, Engineered
Materials Handbook, ASM International, Materials Park, OH, 1990, p. 622.
31. Sung, “Moisture Effect on Adhesive Joints,” p. 622.
32. Comyn, J., Adhesion Science, Chapter 10, “Adhesive Joints and the Environment,” Royal Society
of Chemistry, Cambridge, 1997.
33. Comyn, “Adhesive Joints and the Environment.”
34. Schneberger, G. L., “Polymer Structure and Adhesive Behavior,” in Adhesives in Manufacturing,
G. L. Schneberger, ed., Marcel Dekker, New York, 1983.
35. Bolger, J. C., “New One Part Epoxies Are Flexible and Reversion Resistant,” Insulation, October
1969, pp. 38–44.
340 CHAPTER FIFTEEN
67. Tanner, W. C., “Adhesives and Adhesion in Structural Bonding for Military Material,” in
Structural Adhesive Bonding, M. J. Bodnar, ed., Interscience, New York, 1966.
68. Brockmann, W., “Durability and Life Assessment and Life Prediction of Adhesive Joints,” in
Adhesives and Sealants, vol. 3.
69. Minford, J. D., “Comparison of Aluminum Adhesive Joint Durability as Influenced by Etching
and Anodizing Treatments of Bonded Surfaces,” Journal of Applied Polymer Science, Applied
Polymer Symposia, vol. 32, 1977, pp. 91–103.
70. DeLollis, N. J., “Durability of Structural Adhesive Bonds: A Review,” Adhesives Age, September
1977.
71. Minford, “Durability of Adhesive Bonded Aluminum Joints,” in Treatise on Adhesion and
Adhesives, vol. 3.
72. Bodnar, M. J., and Wegman, R. F., “Effect of Outdoor Aging on Unstressed, Adhesive Bonded
Aluminum to Aluminum Lap Shear Joints,” Technical Report No. 3689, Picatinny Arsenal,
Dover, NJ, May 1968.
73. “Aerospace Adhesive,” EA 929, Hysol Division, Dexter Corp., Technical Bulletin A5-129, Olean,
NY, 1970.
74. MIL-STD-132, U.S. Department of Defense, Washington, DC.
75. Landrock, A. H., “Effect of Environment on Durability of Adhesive Joints,” Chapter 9 of
Adhesives Technology Handbook, Noyes Publications, Park Ridge, NJ, 1985.
76. Arlook, R. S., and Harvey, D. G., “Effect of Nuclear Radiation on Structural Adhesive Bonds,”
Wright Patterson Air Development Center, Report WADC-TR-456-467, August 1956.
77. Pearce, M. B., “How to Use Anaerobics Successfully,” Applied Polymer Symposia No. 19,
Symposium on Processing for Adhesive Bonded Structures, M. J. Bodnar, ed., Interscience, New
York, 1972, pp. 207–230.
78. McCrudy, R. M., and Rambosek, G. M., “The Effect of Gamma Radiation on Structural Adhesive
Joints,” SAMPE National Symposium on the Effects of Space Environment on Materials, St. Louis,
MO, May 1962.
This page intentionally left blank
CHAPTER 16
EPOXY ADHESIVES ON
SELECTED SUBSTRATES
16.1 INTRODUCTION
This chapter identifies and discusses various epoxy adhesives and the processes that have
been used to successfully bond or seal specific substrates. There are only a few materials
that epoxy adhesives will not bond well. These uncooperative substrates are most notably
low-surface-energy plastics, such as the polyolefins, fluorocarbons, and silicones. However,
even these materials can be bonded effectively with epoxy adhesives if a prebond surface
treating process is used to change the nature of the substrate surface. Of the other substrate
materials, there are some that epoxy adhesives will bond more effectively than others.
Table 16.1 lists substrates that generally provide excellent epoxy adhesive joints.
The selection of the proper epoxy adhesive formulation depends on
1. The physical nature of the bulk substrate material (porosity, modulus, thermal expan-
sion coefficient, etc.)
2. The physical and chemical nature of the surface (surface energy, weak boundary layer
attachment, resistance to corrosion, etc.)
Optimal joints can generally be fabricated by the correct combination of epoxy adhesive for-
mulation (i.e., type of resin, curing agent, fillers, modifiers) and surface treatment process.
It is impossible to avoid a discussion on prebond surface preparation since it is one of
the most important factors in the fabrication of a durable and consistent epoxy adhesive
joint. Selection of a proper surface preparation is not an easy task, and the actual imple-
mentation of the surface treating process in production is equally daunting.
Various substrate surface treatments suggested for use with a common epoxy-substrate
joint and service environment combinations are discussed in this chapter. Surface prepara-
tion processes for a range of specific substrates and detailed process specifications are pro-
vided in App. F. The reader is also directed to several excellent texts that provide prebond
surface treatment recipes and discuss the basics of surface preparation, the importance of
contamination or weak boundary layers, and specific processes for adhesive systems other
than epoxy.1,2,3
The discussion in this chapter is organized by the type of substrate material. The sub-
strates are broadly identified as
• Metals
• Plastics (thermosets and thermoplastics)
• Composites
343
Copyright © 2006 by The McGraw-Hill Companies, Inc. Click here for terms of use.
344 CHAPTER SIXTEEN
TABLE 16.1 Common Substrates that Provide for Good Epoxy Adhesion
• Foams
• Elastomers
• Wood and wood products
• Glass and ceramics
• Sandwich and honeycomb structures
Specific substrates are described under each classification. Of course, adhesive applica-
tions are not limited to joints having only one type of substrate. Metal-to-plastic, aluminum-
to-steel, metal-to-wood, glass-to-metal, and an infinite variety of other joint configurations
are all possible. In these applications the nature of each substrate needs to be understood and
considered in the overall selection of an adhesive formulation and bonding process.
This chapter certainly does not consider all possible substrates. However, the guidance
that is offered should be sufficient for the user to select “candidate” joining processes and
epoxy adhesive materials, no matter what substrate or combination of substrates is involved.
Metals, having a relatively high surface energy, are generally considered easy to bond.
However, several problems could occur when one is working with metallic substrates.
These are related to the following characteristics:
1. Durability related to environmental effects on the substrate surface and the interface of
the adhesive joint
2. Variation in the surface chemistry depending on alloy, processing, preconditioning, etc.
3. Relatively low thermal expansion coefficient and high thermal conductivity compared
to most epoxy adhesives
One difficulty in bonding metals is the durability of the joint. It is not so much a problem
of making a strong joint as one of keeping it that way throughout its expected service life. A
weld may have strength of only 600 lb, but it is likely to remain that strong for 5 to 10 years
afterward. An epoxy adhesive, on the other hand, may have 3 to 4 times the initial strength of
a weld, but it could weaken when exposed to high humidities, cycled between hot and cold
temperatures, or immersed in salt water and then dried. By definition, a structural adhesive
must be able to withstand these conditions without significant deterioration.
A second difficulty in bonding metals lies in understanding the surface. One of the
important points to consider in bonding metals is that only the surfaces are involved.
Adhesives and sealants are active only on the molecular surface layer that forms the joint
EPOXY ADHESIVES ON SELECTED SUBSTRATES 345
interface and on any surfaces contained in the porosity of the metal itself. Thus, if unprepared
steel is being bonded, it is not the bulk iron-carbon alloy that is being bonded, but the iron oxide
layer on the surface (presuming that the metal surface was cleaned of contaminants). Similarly
with aluminum, the actual bond is to aluminum oxide rather than the pure metal.
The physical and chemical nature of the metal substrate surface depends on the bulk
alloy composition, processing conditions used, and any preconditioning environments that
the substrate may be exposed to during fabrication and storage. Once the joint is assembled,
the physical and chemical nature of the adhesive-metal interface may change due to the
environment and to the chemical nature of the adhesive.
The base metal is highly reactive in most cases and forms various oxides, sulfides, and
hydrates when exposed to the atmosphere. As a result, it is necessary to consider not only
the bulk metal but also the ability to bond to its hydrated oxide. One must consider the inher-
ent nature of the adhesive force existing between the base metal and its oxide. The final joint
will be no stronger than its weakest link. The weakest link will likely be determined by:
The actual metal surface that takes part in the bonding is illustrated in Fig. 16.1.
Adhesives recommended for metal bonding are in reality used for metal-oxide bonding.
They must be compatible with the firmly bound layer of water attached to surface metal-
oxide crystals. Even materials such as stainless steel and nickel or chromium are coated
with transparent metal oxides that tenaciously bind at least one layer of water.
The nature and characteristic of these oxide layers depend on the base metal and the
conditions that were present during its formation. With steel, for example, the oxide adhe-
sion to the base metal is very weak. In the case of aluminum, however, the oxide is
extremely stable and clings tightly to the base metal. In fact, it adheres so well that it serves
as a protective coating for the aluminum, which is one reason why aluminum is a corrosion-
resistant metal. Certain metals possess surfaces that interact more effectively with one type
of adhesive than with another. This is the reason why adhesive formulators need to know
as much as possible about the surfaces being assembled.
One of the benefits of surface treatment prior to adhesive bonding is that it not only
removes weak boundary layers such as contamination, but also provides a more consistent
surface to which the adhesive can bond. Common surface treatments used for metal sub-
strates are characterized generally in Table 16.2.
A third difficulty in bonding metal surfaces is that they have a higher thermal coeffi-
cient of expansion and thermal conductivity than most epoxy adhesive systems. As
explained in other chapters of this book, the difference in rates of thermal expansion results
in internal stresses in the adhesive joint, especially when the adhesive bond is cured at ele-
vated temperatures or when it is exposed to low temperatures or repeated thermal cycling.
The sections below describe the characteristics of various metal substrates with regard
to epoxy adhesives. Only the more common substrates and those that are most reactive (to
either the adhesive or the environment) are discussed. Metal substrates not covered are gen-
erally easy to bond with epoxy adhesives, and there is little change in properties with envi-
ronmental aging.
16.2.1 Aluminum
Aluminum is an almost ideal substrate for adhesives. It has high surface energy and is very
resistant to most environments. It is also a material with good formability and high
346 CHAPTER SIXTEEN
H
H O
M O O
O M O H H
O O
M H
O O M O
O M OH
O M O M O O H
O O O
M OH
O O H
M O M H
O O
M O H O H O
O O M O
O M O H H
O M OH
O O H
M O M O H
O O M O H H
O M O M O O H O
O O
O M O M O H H H
O O M
O O H H
O H
M O M O M O O H
O O M O H H H
O M O M
O H O H
O M O H O
O M O M O H
O O H
M OH
O M O M O
O
Note:
1. The oxide layer is typically 40−80 Å thick.
2. The hydrated layer is tightly bound.
3. A pure metal surface is rarely available for bonding with adhesives.
FIGURE 16.1 Metal surfaces are actually hydrated metal oxides.4
strength-to-weight ratio that can benefit greatly from properties offered by adhesive joints.
As a result, adhesive bonded aluminum joints are commonly used in the aircraft and auto-
motive industries. Bonding of aluminum most likely comprises the majority of the appli-
cations for epoxy adhesives. Aluminum joints are also commonly used in adhesive studies
and for comparison of different adhesive materials and processes. Adhesive manufacturers’
literature generally describes the properties of bonded aluminum joints.
However, the corrosion resistance of aluminum as well as the durability of joints made
with epoxy adhesives is very dependent on the type of aluminum alloy used. Bonds made
with relatively corrosion-resistant 6061-T6 aluminum alloy will last about 4 times as long
as equivalent joints made with 2024-T3 alloy when exposed to marine environments.
However, with the proper combination of surface treatment and adhesive, these differ-
ences in durability to aggressive environments can be minimized, as shown in Table 16.3.
The initial shear strength and permanence depend on the type of alloy and the pretreatment
used. Note that the data presented here show only the relative differences in joint strength
for one specific epoxy adhesive and are not representative of other adhesive formulations.
There was no attempt to maximize any of these values through choice of the adhesive.
The oxide layer that forms on aluminum is more complex than with other metal substrates.
Aluminum is a very reactive surface, and oxide forms almost instantaneously when a freshly
machined aluminum surface is exposed to the atmosphere. Fortunately, the oxide is extremely
stable, and it adheres to the base metal with strength higher than could be provided by most
adhesives. The oxide is also cohesively strong and electrically nonconductive. These surface
characteristics make aluminum a desirable metal for adhesive bonding, and they are the rea-
sons why many adhesive comparisons and studies are done with aluminum substrates.
The strength of an epoxy adhesive aluminum joint can be improved by cleaning the sur-
face to remove contaminants or by converting the existing surface to a new surface that may
be more consistent. In the case of the aluminum surface, chemical conversion can also pro-
tect the base metal from corrosion and enhance the durability of the bonded joint to various
service environments. The most common surface preparations that have been used for
bonding aluminum can be generally segregated into three groups:
TABLE 16.3 Effect of Surface Treatment on Aluminum Alloy Joints Bonded with a Heat Cured
Epoxy Adhesive (EC-3443, 3M Company)6
A B C A B C A B C
Initial, no aging 1930 1850 2200 2550 2690 2530 2250 2270 2110
After 3 months’ aging 1390 1050 2400 1350 1550 2410 1860 1890 1910
at 23°C, 85% RH
After 3 months’ aging 420 350 1110 920 1050 1100 1260 1520 1230
at 52°C, 100% RH
After 3 weeks’ aging in 0 0 1410 80 150 2090 690 530 1210
5% salt spray solution
Included in the simple cleaning and abrading category are (1) solvent wiping, (2) vapor
degreasing, or (3) either of these methods combined with mechanical abrading. In each
instance, care must be taken to ensure that the cleaning materials themselves do not become
unknowingly contaminated, thus providing ineffectual cleaning or cross-contamination,
resulting in poor bond performance.
Mechanical abrasion or sandblasting is commonly used for treating aluminum surfaces
prior to adhesive bonding because of its simplicity and economics. However, chemical
treatments such as etchants produce higher reliability and longer service life in a bonded
assembly. If aluminum adherends are first cleaned, then sandblasted, and finally chemi-
cally treated, the surface area is increased, the contaminants are removed, and the initial and
long-term strengths are generally excellent. However, this three-step process is often not
necessary when only moderate strengths (500 to 2000 psi) are required or if the finished
adhesive joint will not be exposed to aggressive service environments (especially high-
moisture or high-humidity conditions). Useful bonds in these low- to medium-strength
applications can be achieved simply with cleaning and/or abrasion.
When one is bonding aluminum to itself or to other materials, the optimal surface prepa-
ration should be determined for the application based on the initial strength and durability
required, and then the process must be rigidly followed. Overspecifying the strength
requirements should be avoided since it could result in the selection of a surface prepara-
tion process that is time-consuming, difficult to control, and expensive. Table 16.4 serves
as a guideline for selecting the pretreatment to try first.
The parameters of the pretreating process and their control can be very important in
ensuring consistency from joint to joint. It has been reported that the temperatures used for
rinsing and drying aluminum should not exceed 70°C since different oxides are formed
above and below this temperature. The oxide (bayerite, β-Al2O3⋅H2O) formed at lower tem-
peratures supposedly provides better adhesion.8 It has also been recognized that the use of
distilled or deionized water rather than tap water for cleaning and etching baths is preferred
for greater bond strength and consistency.
In selecting a pretreatment process for aluminum or any other substrate, both the initial
strength and the permanence in a specific operating environment must be considered.
Mechanical abrasion is a useful pretreatment in that it removes the oxide and exposes bare
aluminum. When this is done, however, many of the benefits of the protective oxide layer
are lost. For example, if bare abraded aluminum is bonded, the reactive metal at the joint
interface can potentially become hydrolyzed and oxidized, which will displace the adhe-
sive. Hence, this bonded joint may initially be much stronger than one made with
unabraded metal, but it will deteriorate rapidly when exposed to a harsh environment such
as heat and humidity. This is why pretreatments that modify the oxide layer or create a new,
stable oxide layer are especially desirable when permanence is a primary consideration.
They improve bondability and maintain protection.
To protect the aluminum joint from the effects of the environment, especially water and
corrosion, an artificially thickened oxide layer is generally formed on the surface. Historically,
chemical etching as a surface preparation has provided the surest way of obtaining durable
adhesive bonds with aluminum.
While various acidic or caustic procedures can be employed with or without vapor
degreasing, the most recognized etching pretreatment for bonding aluminum has been the
sulfuric acid–dichromate solution used by the aircraft industry and described in ASTM D
2651. This process is sometimes known as the FPL etch, named after its developers, Forest
Products Laboratories.9 The first step in this process is vapor degreasing followed by alka-
line cleaning and then chemical immersion. The substrates are finally forced air dried.
There are several modifications of this treatment including a pastelike etching solution to
allow for parts that cannot be immersed in the acid solution and a chromate-free etching
process (designated PT2) for improved environmental and occupational health and safety
perspectives.10
Other important methods of pretreating aluminum for adhesive bonding include anodiz-
ing and chromate conversion coating.11 In anodizing, the aluminum is immersed in various
concentrations of acids (usually phosphoric or chromic) while an electrostatic charge is
applied. The oxide reacts with the etchant to form a compound that protects the surface and
is compatible with the adhesive. In this way the aluminum oxide is retained, but it is ren-
dered more receptive to bonding. It has been shown that an anodized surface on aluminum
alloy can constitute a very durable surface for epoxy adhesive bonding with excellent resis-
tance to seacoast or other saltwater types of exposure. Examples of widely used anodizing
processes are the Boeing phosphoric acid anodize (PAA) process12 and the chromic acid
anodize (CAA) process.13
The landmark U.S. Air Force Primary Adhesively Bonded Structure Technology
(PABST) program in the late 1970s demonstrated that properly designed and manufactured
bonded fuselage panels made from the correct aluminum materials can actually operate
safely at higher stress levels than comparable rivet-joined aluminum structures.14,15 The
results of this program show phosphoric acid anodizing as an optimal way to achieve
durable aluminum bonds.
• PAA is the most durable pretreatment for aluminum that is processable within reasonable
production tolerances.
• PAA with a corrosion-resistant primer provides the best corrosion resistance.
• The adhesive should be selected on the basis of durability as defined by slow cyclic test-
ing in a hot and humid environment.
The chemical conversion coating method is also commonly used to treat aluminum sub-
strates prior to bonding. Chemical conversion coating is an amorphous phosphatization
process wherein the aluminum is treated with a solution containing phosphoric acid, chromic
acid, and fluorides. Chromate conversion coatings on aluminum constitute an effective way
to enhance the surface bondability and also improve the corrosion resistance of the bond
line.16 The resulting durability observed with mechanical abrasion and chemical conversion
coatings is variable and depends on the particular processing conditions, whereas anodizing
and etching processes produce consistent and generally durable aluminum joints.
The durability of epoxy bonded aluminum joints that were immersed in water is shown
in Fig. 16.2. The anodized and grit-blasted surface treatments, although giving different ini-
tial joint strengths, showed no deterioration after 2 years’ exposure. Both the vapor-degreased
and conversion coating treatments were significantly degraded by the moist environment.
350 CHAPTER SIXTEEN
3000
1
2000 2
Shear strength, psi
3
1000
test method for measuring the surface treatment effectiveness. The yield factor of the
adherend is never approached because of relatively poor peel strength of the adhesive.
Once the surface considerations are taken care of, there are many types of epoxy adhe-
sives that can bond well to aluminum. The selection will depend on the strength needed, the
type of stress involved (e.g., peel or shear; static or dynamic), and the operating environ-
ment. Reynolds Metals Company21 offers some general observations in selecting an adhe-
sive for aluminum bonding.
All of the commercial epoxy adhesives presented in App. B bond well to aluminum and to
a wide variety of other materials. Sell22 has ranked a number of aluminum adhesives in order
of decreasing durability as follows: nitrile-phenolics, high-temperature epoxies, elevated-
temperature curing epoxies, elevated-temperature curing rubber-modified epoxies, vinyl epox-
ies, two-part room temperature curing epoxy paste with amine cure, and two-part urethanes.
The starting formulations presented in Table 16.5 are designed for general-purpose
bonding of aluminum where other substrates may also be involved. Note that aluminum
powder is a key ingredient in these formulations to provide for a closer match in coefficient
of thermal expansion between the adhesive and the substrate.
16.2.2 Beryllium
Beryllium and its alloys (e.g., beryllium copper) have gained interest in the aerospace
industry and specialty sports equipment industry in recent years. Brazing or riveting can be
used for joining, but these methods are expensive, and distortion or highly stressed areas
may be encountered. The metal must be handled with care when the processing produces
dust, chips, scale, slivers, mists, or fumes, since airborne particles of beryllium and beryl-
lium oxide are toxic with latent health effects. Abrasives and chemicals used with beryl-
lium must be disposed of properly.
The gain in the popularity of beryllium has been due in large part to adhesive bonding.
Bonding permits one to take advantage of the excellent combination of physical and
mechanical properties of beryllium and minimizes the inherent problems of high notch sen-
sitivity and low ductility. Adhesive bonding is generally applied where high strength-to-
weight ratios are important. Since beryllium retains significant strength at temperatures up
to 540°C, many of the adhesives and surface preparations developed for beryllium have
been used to exploit the metal’s high-temperature properties in aerospace applications.
Beryllium is a very reactive metal, and it reacts quickly with methyl alcohol, fluorosol-
vents, perchloroethylene, and blends of methyl ethyl ketone and fluorosolvents. Beryllium
352 CHAPTER SIXTEEN
TABLE 16.5 Starting Formulations for Epoxy Adhesive for Bonding Aluminum23
Parts by weight
Component A B C
Part A
DEGEB epoxy resin (EPON 828, Resolution 100 100 50
Performance Products)
Ground tubular alumina (Alumina T-60/T-64, 230
Alcoa)
Aluminum powder 100 50
Part B
Aromatic amine eutectic (60/40 blend of MDA 12
and MPDA)
Aliphatic amine (EPI-CURE 3234, Resolution 5
Performance Products)
Aliphatic amine (EPI-CURE 3245, Resolution 5
Performance Products)
Polyamide (EPI-CURE 3125) 54.5
Property
can be pitted by long exposure to tap water containing chlorides or sulfates.24 Proprietary
coatings are used to provide a corrosion-resistant barrier.25
A number of prebonding surface preparations for bonding beryllium and its alloys with
epoxy adhesives have been suggested in the literature. One procedure is to degrease the
substrate with trichloroethylene, followed by immersion in the solution listed below for
5 to 10 min at 23°C.
The substrate should then be washed in tap water and rinsed with distilled water. The final
step in the process is an oven-dry of 10 min at 121 to 177°C.26 Several other surface prepa-
ration procedures for beryllium have been reported to have merit. Epoxy and epoxy hybrid
adhesives have been found to provide high strengths on sulfuric acid–sodium dichromate
etched beryllium.27
EPOXY ADHESIVES ON SELECTED SUBSTRATES 353
16.2.3 Copper
Copper substrates are commonly bonded with epoxy adhesives in the microelectronics and
marine industries. Compared to aluminum substrates, copper when bonded with epoxy
adhesives provides lower initial strength. Depending on the adhesive and the type of test
used, this can be as much as 50 percent lower. Similar to aluminum joints, copper joints
bonded with epoxy adhesives can show poor durability in moist environments unless the
interface is protected.
Copper is used in three basic forms: pure, alloyed with zinc (brass), and alloyed with tin
(bronze). Copper and copper alloys are difficult to bond satisfactorily, especially if high
shear and peel strengths are desired. The primary reason for this difficulty is that the oxide
that forms on copper develops rapidly (although not as fast as the rate of oxide develop-
ment on aluminum). The copper oxide layer is weakly attached to the base metal under
usual conditions. Thus, if clean, bare copper substrates were bonded, the initial strength of
the joint would be relatively high, but on environmental exposure an oxide layer could
develop which will reduce the durability of the joint.
Cleaning and mechanical abrasion are often used as pretreatments for copper and its
alloys where low to medium bond strengths are acceptable. For optimum bond strength and
permanence, the oxide layer must be specifically “engineered” for adhesive bonding. This
is often done through what has been called the “black oxide” coating process or through
chromate conversion coatings.
The black oxide processes form microrough morphologies on the copper surface. One
such process uses a commercially available product28 to form a strong black oxide coating
on the copper surface. This process requires alkaline cleaning and acid immersion before
the copper surface is immersed at elevated temperatures in a solution containing the pro-
prietary product. The process steps are critical, and time and temperature must be carefully
controlled. Parts should be joined within 4 h after chemical treating. Another black oxide
coating process, Method E in ASTM D 2651, is intended for relatively pure copper alloys
containing 95 percent copper.
The sodium dichromate–sulfuric acid process has been found by some to be superior to
other ferric chloride methods for the prebond treatment of copper. This dichromate–sulfuric
acid method is also defined by ASTM D 2651. Nitric acid and ferric chloride etching
processes have also been found to be useful for copper, brass, and bronze substrates in cer-
tain applications.
Copper may also react with certain chemicals in the adhesive formulations at elevated
cure temperatures. Copper has a tendency to form brittle amine compounds with some
epoxy curing agents including certain amines and anhydrides. However, room temperature
curing epoxy adhesives will generally give very good adhesion to copper and copper alloy
substrates. Table 16.6 shows the effect of heat aging on tensile shear strength of DGEBA
epoxy cured with two curing agents.
For elevated-temperature cures, however, dicyandiamide and melamine curing reac-
tions have been shown to be beneficial in epoxy-based adhesives when used on copper sub-
strates. These epoxy adhesives perform about as well in copper joints as in aluminum or
steel joints. The formulations show significantly increased time to adhesive failure on
either bare or alkaline permanganate-treated copper.30
Brass is an alloy of copper and zinc, and bronze is an alloy of copper and tin.
Sandblasting or other mechanical means of surface preparation may be used for both of
these copper alloys. Surface treatment combining mechanical and chemical treatment with
a solution of zinc oxide, sulfuric acid, and nitric acid is recommended for maximum adhe-
sion properties. Adhesives similar to those recommended for copper may be used on brass
and bronze substrates.
354 CHAPTER SIXTEEN
TABLE 16.6 Effect of Heat Aging on Tensile Shear Strength of Copper Joints Bonded
with a DGEBA Epoxy Cured with Tri-2-ethylhexoate Salt of DMP-30 and DEAPA29
16.2.4 Magnesium
The surface preparation and bonding methods developed for magnesium and magnesium-
based alloys are closely associated with corrosion prevention because magnesium is one of
the most reactive of all metals. Numerous processes have been developed for both painting
and bonding of magnesium. The choice of a bonding process will be determined by crite-
ria such as high bond strength, high corrosion resistance, or both.
The position of magnesium in the electrochemical or galvanic series indicates that it has
a potential of reacting chemically with a great number of other metals when there is elec-
trical connection. The conductive path could be caused by direct metal-to-metal contact, an
aqueous solution in which there is an electrolyte (e.g., chloride ions in solution), or by other
ways. The key in any assembly involving magnesium is to design and assemble the parts
in such a manner that the conductive path is eliminated.
Magnesium will react with moisture to form a magnesium hydroxide that, in itself, pro-
vides some corrosion protection. However, this coating is not quite uniform and may con-
tain carbonates and sulfates or other ionic materials that are present in the surrounding
environment. Magnesium is quite resistant to alkalies, and under controlled conditions this
factor is used to advantage in surface preparation. Organic compounds, in general, do not
react with magnesium as do the aqueous electrolytes. Anhydrous methyl alcohol is an
exception—it can react readily with magnesium.
Minor constituents found in magnesium alloys can play a significant part in adhesive
bonding. Not all magnesium alloys react to protective surface treatments in the same way.
Knowledge of the alloy composition must be coupled with a careful surface preparation
selection process. Grain structure of the metal can also influence the nature of the surface
preparation. The temper of the alloys, such as strain-hardened (W), annealed (O), or fabri-
cated (F), and the various ways of treating (T1 through T10) indicate the manner of treat-
ment an alloy has received. These different surfaces have a bearing on the effectiveness of
the surface preparation process.
Removal of light oil or light chromate coatings (used to protect the magnesium during
shipment and storage), mill scale, lubricants, welding fluxes, etc., must occur before the
EPOXY ADHESIVES ON SELECTED SUBSTRATES 355
desired surface preparation is applied. Three cleaning methods, used either separately or in
combination, constitute the necessary cleaning before the final surface preparation: solvent
degreasing, mechanical cleaning (galvanic reactions from the grit must be avoided), and
chemical treatment (acid pickling using chromic, nitric, or phosphoric acid solutions). It is
not uncommon for the acid pickling or cleaning procedure to suffice as the final surface
preparation process. An alkaline detergent cleaning process is described in ASTM D 2651
Method A. A hot chromic acid cleaning process that can be combined with the alkaline
cleaning process is given by ASTM D 2651 Method B.
Light anodic treatment and various corrosion preventive treatments produce good sur-
faces on magnesium for adhesive bonding (ASTM D 2651 Method D). These treatments
were developed by magnesium alloy producers, such as Dow Chemical Company (Dow 17
and Dow 7) and others. Details are available from the ASM Metals Handbook31 and in
MIL-M-45202, Type I, Class 1, 2, and 3.32 Some surface dichromate conversion coatings
and wash primers designed for corrosion protection can also be used for adhesive bonding
of magnesium (ASTM D 2651 Method E). Details are found in ASM Metals Handbook and
MIL-M-3171.33
The thickness of the surface coating influences the relative degree of protection for the
magnesium against corrosive elements. Heavy anodic coatings of about 0.001 in are used
for maximum corrosion resistance. However, such thick coatings are generally not optimal
for adhesive bonding because cohesive failure in the coating results. Thinner coatings of
0.0001 to 0.0003 in. obtained from the Dow 17 anodic surface treatment are better for adhe-
sive bonding. Thin films also provide greater coating flexibility with less likelihood of
cracking upon flexure.
A problem unique to magnesium substrates is the formation of loose particles (termed
smut) on the surface of the metal. The source of the smut can be varied but is usually related
to the environment during the cleaning process. Such loose, gray, black, or brown smut par-
ticles are believed responsible, in some cases, for erratic adhesive strength results as well
as lower bond strengths. These loose particles generally can be rubbed off with a solvent-
dampened pad.
High-humidity and salt spray environments have been found to cause the greatest
decrease in bond strength of magnesium adhesive joints. Of the several surface treatments
that have been evaluated, the Dow 17 surface preparation (ASTM D 2651 Method C) pro-
vides the best overall performance with all adhesives under these types of environmental
conditions.34
Almost any epoxy adhesive can be used on magnesium provided that proper surface pro-
tection is maintained. In view of the corrosion potential, water-based adhesives or adhesives
that allow water permeation may be expected to cause problems with magnesium substrates.
16.2.5 Nickel
Nickel and its common alloys such as Monel (nickel-copper), Inconel (nickel-iron-
chromium), and Duranickel (primarily nickel) can be bonded with procedures that are rec-
ommended for stainless steels.35 A simple nitric acid process has also been used consisting of
solvent cleaning, immersion for 4 to 6 s at room temperature in concentrated nitric acid, rins-
ing with cold deionized water, and finally drying. Also, a chromium trioxide–hydrochloric
acid process consisting of a 60- to 80-s immersion in acid solution has been suggested. If
immersion is impossible, this latter solution may be applied with a cheesecloth after sol-
vent cleaning. The solution is applied to approximately 1 ft2 of the substrate surface at a
time, and it remains on the part for approximately 1 min.
Epoxies are commonly used to bond nickel substrates. However, the nickel alloys often
find applications at temperatures higher than most organic adhesives are capable of resisting.
356 CHAPTER SIXTEEN
Thus, there is relatively little development work done on optimizing pretreatments for
nickel alloy substrates.
One of the major problems associated with bonding plated parts is the different surface con-
ditions that can be caused by variations in plating equipment, process methods, and solu-
tion concentrations. These variables result in plated surfaces with broad conditions of
surface finish and inconsistent metallurgical and adhesion properties.
Nickel-plated parts should not be heavily etched or sanded. Roughening the surface to
obtain good bonds is unnecessary, as excellent bonds can be made to the smoothest of sur-
faces as long as it is properly cleaned. The most likely surface treatments are solvent clean-
ing, vapor degreasing, or soap cleaning. A recommended practice for surface-treating
plated parts is light scouring with a nonchlorinated commercial cleaner, rinsing with dis-
tilled water, and drying at temperatures under 50°C. The substrate is then primed or bonded
as soon as possible after surface preparation.
Chromium and chrome-plated alloys can be etched in a 50% solution of concentrated
hydrochloric acid for 2 to 5 min at 90°C. Zinc and galvanized metal parts can be similarly
immersed for 2 to 4 min at room temperature in such a solution at 15% concentration. In
both cases, the part should be primed or the adhesive applied as soon as possible after sur-
face treatment.
Another metal for which adhesive bonding is widely utilized is galvanized steel.
Adhesive bonding is a preferred method of joining this material because it is difficult to
weld. Galvanized steel is nominally zinc-coated steel, but in reality it is much more com-
plex. Aluminum, lead, tin, and magnesium are also present, and these contribute to the dif-
ficulty in bonding this material. Magnesium oxide, in particular, has virtually no bond
strength to magnesium or zinc. The most common way to prepare galvanized steel for
bonding is through repeated detergent washings. This removes the magnesium oxide pro-
gressively with the aluminum oxide, so that eventually only zinc oxide remains, which is
well attached and relatively easily bonded.
Because of their widespread use in industry, steel and iron are frequently bonded. Like the
surfaces of most metals, their surfaces actually exist as a complex mixture of hydrated
oxides and absorbed water. Unfortunately, iron oxides are often not the best surface for
adhesives because the oxides may continue to react with the atmosphere after an adhesive
has been applied, thus forming weak crystal layers. Iron oxides are more difficult to “engi-
neer” than aluminum, copper, or titanium oxides. As a result, grit blasting is used in most
applications. Although it provides adequate adhesion and durability for many applications,
grit blasting does not provide great durability in severe environments. Conversion coatings
are often used in these cases.
Bonding operations frequently require the mechanical or chemical removal of loose
oxide layers from iron and steel surfaces before adhesives are applied. To guard against
slow reaction with environmental moisture after the bond has formed, iron and steel sur-
faces are often phosphated prior to bonding. This process converts the relatively reactive
iron atoms to a more passive, “chemically stable” form that is coated with zinc or iron phos-
phate crystals. Such coatings are applied in an effort to convert a reactive and largely
unknown surface to a relatively inert one whose structure and properties are reasonably
well understood.
EPOXY ADHESIVES ON SELECTED SUBSTRATES 357
Corrosion protection is critical in bonding steel—even more critical than for many
other metallic adherends. The initial adhesion to steel is usually good but deteriorates
rapidly during environmental conditioning. Thus, corrosion-preventing primers are usu-
ally recommended because they protect the surface against changes after bonding, and the
time lapse from cleaning to bonding is not as critical. Steel alloys will form surface oxides
in a very short time. Drying cycles after cleaning can be critical. During these processes
an alcohol rinse after a water rinse tends to accelerate drying and reduce surface layers that
are undesirable.
Mild steel (carbon steel) may require no more extensive treatment than degreasing and
abrasion to give moderate-strength adhesive bonds. Tests should be carried out with the
actual adhesive to be used to determine whether a chemical etch or other treatment is essen-
tial.36,37,38 Figure 15.21 illustrates the initial and residual strengths obtained with joints of
mild steel employing different adhesives and different surface treatments. From these
results it may be concluded that the durability of steel bonds produced with a elevated-
temperature curing epoxy adhesive is better than that of specimens produced with room
temperature curing adhesive systems.
The general sequence of surface preparation for ferrous surfaces such as iron, steel, and
stainless steel consists of the following methods: degreasing, acid etch or alkaline clean,
rinse, dry, chemical surface treatment, and priming. The chemical surface treatment step is
not considered a standard procedure, but it is sometimes used when optimum quality joints
are required. It consists of the formation of a corrosion-preventing film of controlled chem-
ical composition and thickness. These films are a complex mixture of phosphates, fluo-
rides, chromates, sulfates, nitrates, etc. The composition of the film may be the important
factor that controls the strength of the bonded joint.
Stainless steel, or corrosion-resistant steel, has a high chromium content (11% or
higher) as the primary alloying element. The stable oxide film that exists on stainless steel
will tightly bind ions from prior manufacturing steps. Rinsing after treatment seems to be
especially critical in ensuring the successful bonding of stainless steel surfaces; but even
bond quality on well-rinsed surfaces may not be comparable with that on other, more eas-
ily bonded materials.
There have been a large number of surface preparation methods reported to give excel-
lent bonds with stainless steels. In addition to mechanical methods, strong acids and strong
alkalies are used. A wet abrasive blast with a 200-grit abrasive followed by thorough rins-
ing to remove the residue is an acceptable procedure for some uses but does not produce
high bond strengths. Acid treatments are usually used to produce strong bonds with most
adhesives. Passivation in nitric acid solution and concentrated sulfuric acid saturated
sodium dichromate solution both produce high bond strength but with low or marginal peel
strength. Such joints may fail under vibration stress, particularly when a thin stainless steel
sheet is bonded with a brittle adhesive.
Acid etching can be used to treat types 301 and 302 of stainless. These processes result
in a heavy black smut formation on the surface. This material must be removed if maxi-
mum adhesion is to be obtained. The acid etch process produces bonds with high peel and
shear strengths. The 400 series of straight chromium stainless steels should be handled in
the same manner as the plain carbon steels. The various types of precipitation hardening
(PH) stainless steels each present an individual problem. Processes must be adopted or
developed for each type.
Once properly treated, there are practically thousands of organic adhesive compounds
that are available for bonding steel alloys. Epoxies are the most common of structural adhe-
sive for bonding steel. Figure 15.15 shows the effect of humidity on the adhesion of two
structural epoxy adhesives used to bond stainless steel.
In high-volume industries such as automotive and appliances, there is a desire to mini-
mize or eliminate any surface preparation process for steel. Special adhesive systems have
358 CHAPTER SIXTEEN
been developed that bond well to steel coming direct from the mill. This substrate is
usually coated with light oil for protection against corrosion. Toughened epoxy, epoxy-
urethane hybrid, and thermosetting acrylic adhesives have been found suitable for this
application. These adhesives usually work with oily surfaces through either (1) thermody-
namic displacement of the oil at the surface by the adhesive or (2) absorption of the oil into
the body of the adhesive as a plasticizer. Bond strengths of an epoxy-urethane adhesive to
oily steel surfaces as well as other substrates commonly found in the auto industry are
shown in Table 7.9.
16.2.8 Titanium
Turco Product Division, Purex Corporation. When combined with vapor degreasing and
alkaline cleaning, this process offers very high-strength bonds on titanium. Plasa Jell is also
a proprietary chemical marketed by Semco Division, Products Research and Chemical
Corp. This formulation is available either as a thixotropic paste suitable for brush applica-
tion or as an immersion solution for tank treatment.
The VAST process was developed by Vought Systems of LTV Aerospace Corporation;
VAST is the acronym for Vought Abrasive Surface Treatment. In this process, the titanium
is blasted in a specially designed chamber with a slurry of fine abrasive containing fluo-
rosilicic acid under high pressure. The aluminum oxide particles are about 280-mesh, and
the acid concentration is maintained at 2 percent. The process produces a gray smut on the
surface of 6,4 titanium alloy that must be removed by a rinse of 5 percent nitric acid. The
resulting joint strength is claimed to be superior to that provided by the unmodified phos-
phate fluoride process, but is slightly lower than that provided after the Turco 5578 alka-
line etch.
Adhesives recommended for bonding titanium substrates include epoxies, nitrile-
epoxy, nitrile-phenolic, polyimide, and epoxy-phenolic. Epoxy adhesives have generally
been selected when a combination of high initial strength and durability is required.
Representative data are shown in Fig. 16.3 for an epoxy adhesive and several proprietary
and nonproprietary treating processes. Keith49 has covered all aspects of titanium adhesive
bonding, including adhesive selection.
Plastics are usually more difficult to bond with adhesives than are metal substrates. Plastic
surfaces can be unstable and thermodynamically incompatible with the adhesive. The
actual bonding surface may be far different from the expected substrate surface. The plas-
tic part can possess physical properties that will cause excessive stress in the joint. The
operating environment can change the adhesive-plastic interface, the base plastic, the adhe-
sive, or all three.
However, even with these potential difficulties, adhesive bonding can be an easy and reli-
able method of fastening one type of plastic to itself, to another plastic, or to a nonplastic sub-
strate. Pocius et al. provides an excellent treatise on the use of adhesives in joining plastics.51
The physical and chemical properties of both the solidified adhesive and the plastic sub-
strate affect the quality of the bonded joint. Major elements of concern are the thermal
expansion coefficient, modulus, and glass transition temperature of the substrate relative to
the adhesive. Special consideration is also required of polymeric surfaces that can change
during normal aging or on exposure to operating environments.
Significant differences in the thermal expansion coefficient between the substrate and
the adhesive can cause severe stress at the interface. This is common when plastics are
bonded to metals because of the difference in thermal expansion coefficients between the
substrates. Residual stresses are compounded by thermal cycling and low-temperature ser-
vice. Selection of a resilient adhesive or adjustments in the adhesive’s thermal expansion
coefficient via fillers or additives can reduce such stress.
Bonded plastic substrates are commonly exposed to peel because the part thickness is
usually small and the modulus of the plastic is low. As a result, tough adhesives with high
peel and cleavage strengths are usually recommended for bonding plastics. The require-
ment of flexibility is especially important for thermoplastics because of their lower modu-
lus and greater thermal expansion coefficient than thermosetting plastics. Table 16.7 shows
starting formulations and properties for several epoxy adhesives that are recommended for
bonding plastic substrates.
360 CHAPTER SIXTEEN
50 2.0
MPF
46 1.8
I
41 1.6
PF
Crack extension, mm
36 1.4
Crack extension, in
30 1.2
25 1.0
20 0.8
DA
15 0.6
DP
10 II 0.4
LP
TU 0.2
5 CAA-5
CAA-10 III
0 0
0 100 200 300 400 500 600 700 800
Time, h
(a)
21 3.0
DP
PF
18 2.6
MPF
CAA-5
15 2.2
PF CAA-10
Stress, MPa
Stress, ksi
12 1.8
TU LP
9.5 1.4
DA
7 1.0
4 0.6
10 100 103
Time to failure, h
(b)
FIGURE 16.3 Typical bond durability data for Ti-6 Al-4 V adherends bonded with an epoxy adhesive and
aged at 60°C and 100 percent RH. (a) Crack propagation versus time for the wedge crack propagation test.
(b) Applied stress versus time to failure for the lap shear geometry. PF—phosphate fluoride; MPF—modified
phosphate fluoride; DP—PasaJell 109 dry hone; LP—PasaJell 107 liquid hone; CAA-5—5% solution;
CAA-10—10% solution; TU—Turco 5578 etch; DA—Dapcotreat.50
EPOXY ADHESIVES ON SELECTED SUBSTRATES 361
TABLE 16.7 Starting Formulations for Several Epoxy Adhesives Recommended for Bonding
Plastic Substrates52
Parts by weight
Component A B C D E F G
Part A
Modified epoxy resin (EPON 8132, 75 100 100 100
Resolution Performance Products)
DGEBA resin (EPON 828) 25 75
CTBN modified bisphenol A resin 25
(EPON 58005)
Modified epoxy resin (EPON 862) 75
CTBN modified bisphenol F resin 25
(EPON 58003)
Elastomer modified bisphenol A resin 100
Fumed silica (Cab-O-Sil M5) 2 2 2 2 2 2 2
Part B
Amidoamine (EPI-CURE 3055, 42 42 46 42
Resolution Performance Products)
Polyamide (EPI-CURE 3125) 50
Polyamide (EPI-CURE 3163) 138
Polyamide (EPI-CURE 3164) 86
Property
Gel time, min, at 25°C, 100 g 345 165 178 142 193 113 109
Thin film setting time, h 16 11 13 13 10 20.5 34.5
Tensile shear strength, psi, after
7 days at 25°C on
• PVC 691 677 754 776 308 549 529
• Polyurethane 479 435 759 573 718 906 730
• Polyethylene terephthalate 255 254 487 506 525 800 632
Tensile shear strength, psi,
after 30-min cure at 140°C on
• PVC 1216 1281 939 951 661 969 674
• Polyurethane 319 550 385 407 307 583 429
• Polyethylene terephthalate 768 653 890 796 964 1244 848
• Nylon 434 — 414 444 650 725 466
Glass transition temperature, °C 60 86 75 70 50 45 40
Structural adhesives must have a glass transition temperature higher than the operating
temperature or preferably higher than that of the part that is being bonded, to avoid a cohe-
sively weak bond and possible creep problems at elevated temperatures. Modern engineer-
ing plastics, such as polyimide or polyphenylene sulfide, have very high glass transition
temperatures. Common adhesives have a relatively low glass transition temperature, so that
the weakest thermal link in the joint may often be the adhesive. The use of an adhesive too
far below the glass transition temperature could result in low peel or cleavage strength.
Brittleness of the adhesive at very low temperatures could also manifest itself in poor
impact strength.
Plastic substrates could be chemically active, even when isolated from the operating
environment. Many polymeric surfaces slowly undergo chemical and physical change.
362 CHAPTER SIXTEEN
The plastic surface, at the time of bonding, may be well suited to the adhesive process.
However, after aging, undesirable surface conditions may present themselves at the inter-
face, displace the adhesive, and result in bond failure. These weak boundary layers could
come from the environment or from within the plastic substrate itself. Plasticizer migration
and degradation of the interface through uv radiation are common examples of weak
boundary layers that can develop with time at the interface.
Moisture, solvent, plasticizers, and various gases and ions can compete with the cured
adhesive for bonding sites. The process by which a weak boundary layer preferentially dis-
places the adhesive at the interface is called desorption. Moisture is the most common des-
orbing substance, being present both in the environment and within many polymeric
substrates. Solutions to the desorption problem consist of eliminating the source of the weak
boundary layer or selecting an adhesive that is compatible with the desorbing material.
Excessive moisture can be eliminated from a plastic part by drying the part before bonding.
Additives that can migrate to the surface can possibly be eliminated by reformulating the
plastic resin. Also, certain adhesives are more compatible with oils and plasticizers than
others. For example, the migration of plasticizer from flexible polyvinyl chloride can be
counteracted by using a nitrile-based adhesive. Epoxy-nitrile rubber and to a lesser extent
epoxy-urethane hybrid adhesives are capable of absorbing the plasticizer without degrading.
Thermosetting plastics cannot be dissolved in solution and do not have a melting tempera-
ture since these materials are crosslinked. Therefore, they cannot be heat- or solvent-welded.
In some cases, solvent solutions or heat-welding techniques can be used to join thermo-
plastics to thermoset materials. However, most thermosetting plastics are not particularly
difficult to bond with adhesive systems. They are generally bonded with many different
types of adhesives such as epoxies, acrylics, and urethanes. Since thermosetting parts are
often highly filled and rigid, a flexible adhesive is not so important as one that can resist
the service environment and provide practical joining processes.
In general, unfilled thermosetting plastics tend to be harder, more brittle, and not as
tough as thermoplastics. Thus, it is common practice to add filler to thermosetting resins.
These fillers can affect the nature of the adhesive bond (either positively or negatively) and
are a possible source of lot-to-lot and supplier-to-supplier variability.
Thermosetting plastics, being chemically crosslinked, shrink during cure. Sometimes
the cure is not entirely complete when the part is bonded. In these cases, cure of the part
can continue during the bonding operation or even on aging in service, resulting in shrink-
age and residual stresses in the joint. Depending on the nature of the crosslinking reaction,
volatile by-products could also generate due to postcuring of the part and could provide
materials for a weak boundary layer.
The surface of thermoset materials may be of slightly different chemical character than
the material beneath the surface because of surface inhibition during cure or reaction of the
surface with oxygen and/or humidity in the surroundings. By abrading the surface, a more
consistent material is available for the adhesive to bond.
Abrasion and solvent cleaning are generally recommended as a surface treatment for
thermosetting plastics. Frequently a mold release agent is present on thermoset materials
and must be removed before adhesive bonding. Mold release agents are removed by a
detergent wash, solvent wash, or solvent wipe. Clean, lint-free cloth or paper tissue is com-
monly used, and steps must be taken to ensure that the cleaning materials themselves do not
become contaminated. Cleaning solvents used for thermosetting materials are acetone,
toluene, trichloroethylene, methyl ethyl ketone (MEK), low-boiling petroleum ether, and
isopropanol.
EPOXY ADHESIVES ON SELECTED SUBSTRATES 363
Diallyl Phthalate. Diallyl phthalates are among the best of the thermosetting plastics
with respect to high insulation resistance and low electrical losses. These properties are
maintained up to 200°C or higher and in the presence of high-humidity environments. Also,
diallyl phthalate resins are easily molded and fabricated. There are several chemical varia-
tions of diallyl phthalate resins, but the two most commonly used are diallyl phthalate
(DAP) and diallyl isophthalate (DAIP). The primary difference is that DAIP will withstand
somewhat higher temperatures than will DAP. Both DAP and DAIP have excellent dimen-
sional stability and low shrinkage after molding. Surfaces are hard and tough, and they pick
up very little moisture. DAP parts are ordinarily molded or laminated with glass fibers.
Only filled molding resins are commercially available.
Typical surface preparation calls for cleaning with acetone, MEK, or other common sol-
vent. Once clean, the substrate is then mechanically abraded with sand, grit or vapor blast,
or steel wool. The surface is again wiped clean with fresh solvent. Typical adhesives that
are employed include epoxies, urethanes, and cyanoacrylates. Polysulfides, furanes, and
polyester adhesives have also been suggested.
Epoxy. Epoxy resins are one of the most commonly used thermosetting materials. They
offer a wide variety of substrate properties depending on base resin, curing agent, modi-
fiers, fillers, and additives. Epoxies show good dimensional stability, electrical properties,
and mechanical strength. They have good creep resistance and will operate over a wide
temperature range. However, high temperatures tend to oxidize epoxies after long periods.
Loose surface particles caused by UV exposure (chalk) could provide a weak boundary
layer. Both filled and unfilled grades are available. Common fillers include minerals, glass,
silica, and glass or plastic microballoons. Epoxy composites are available with continuous
or discontinuous reinforcing materials of several types.
364 CHAPTER SIXTEEN
The common surface preparation treatment for epoxy resins is to wipe with solvent,
mechanical abrasion, and final solvent cleaning. Epoxy parts can be most easily bonded
with an epoxy adhesive similar to the material being bonded. Urethanes, cyanoacrylates,
and thermosetting acrylics have also been used when certain properties or processing param-
eters are required.
Phenolic, Melamine, and Urea. The phenolics are heavily commercialized thermoset-
ting materials that find their way into many applications. They have an excellent combina-
tion of physical strength and high-temperature resistance. They have good electrical
properties and dimensional stability. Like epoxies and diallyl phthalate, phenolic resins are
often found to contain fillers and reinforcement.
Phenolics are formed by a condensation reaction. This results in the formation of water
during cure. If a phenolic substrate is not completely cured, it may continue to cure during
processing or when exposed to elevated temperatures in service and liberate additional
water vapor. This water vapor could form a weak boundary layer and drastically reduce the
strength of a bonded joint.
The phenolics can be surface-treated by any of the standard processes for thermosetting
materials. Solvent cleaning and mechanical abrasion are commonly employed for high
joint strengths. The parts must be completely dry and cleaned of any mold release.
Phenolic parts can be bonded with a wide variety of adhesives. With many adhesives it
is possible that the bond strength of the joint will be greater than the strength of the
adherend. Phenolics are used in many high-temperature applications. Adhesives with good
high-temperature properties are required when these parts are bonded. For this reason,
high-temperature epoxy, nitrile-phenolic, and urea-formaldehyde adhesives are com-
monly used. Neoprenes and elastomeric contact adhesives are used for bonding decorative
laminate.
Melamine (melamine formaldehyde) resins and urea (urea formaldehyde) resins are
similar to the phenolics. They are hard, rigid materials that have excellent electrical and
abrasion-resistant characteristics. Melamine parts are also noted for high impact resistance
and resistance to water and solvents. Only filled melamine resins are available.
As a result of their properties, melamines are often used as decorative laminates. The
melamine resins cure via an addition reaction mechanism so no reaction by-products can
be produced on postcure as with the phenolic resins. The specific surface preparation for
adhesive bonding and the preferred adhesives for bonding melamine and urea parts are sim-
ilar to those suggested for phenolic resins.
Polyimides. Polyimide plastics have exceptional thermal stability. Some types of poly-
imides are able to withstand temperatures up to 480°C for short periods. Polyimides are
available as both thermosetting and thermoplastic materials. Certain types of thermosetting
polyimides can be cured either by a condensation reaction or by an addition reaction mech-
anism. Polyimides are available in many forms such as fabricated parts, molding resins,
films, and coatings. Polyimides are often filled for greater physical properties and dimen-
sional stability. In certain bearing products, polyimides are filled with low-friction mate-
rials such as graphite, polytetrafluoroethylene (PTFE), and molybdenum disulfide. These
low-surface-energy fillers could interfere with adhesion when they are present at the
interface.
Polyimide parts can be bonded using either a standard thermosetting substrate surface
treatment listed above or one of a number of specialty processes developed for higher adhe-
sive strength and permanence. DuPont recommends that Vespel polyimide parts be bonded
by a process consisting of solvent cleaning in trichloroethylene, perchloroethylene, or
trichloroethylene. Because of the high-temperature and solvent resistance of this plastic,
parts have also been cleaned by refluxing in the solvent and by ultrasonic agitation. Once
EPOXY ADHESIVES ON SELECTED SUBSTRATES 365
clean, the parts are abraded with a wet or dry abrasive blast, solvent-cleaned again, and
dried before bonding.
A sodium hydroxide etch process has been developed for polyimide parts that require
maximum adhesive strength.53 The parts are first degreased and then etched for 1 min at 60
to 90°C in 5 percent solution of sodium hydroxide in water. After etching, the parts are
rinsed in cold water and air-dried.
Thermosetting polyimide materials can be bonded with any structural adhesive.
Unfortunately, the high-temperature strength of the adhesives is generally not as good as
the high-temperature characteristics of the polyimide. Where the maximum strength is
needed at elevated temperatures, high-temperature curing epoxy, phenolics, or polyimide
adhesive formulations are generally applied. Where high strengths are not crucial, silicone
adhesives are commonly employed.
Polyester. Polyesters are also common thermosetting resins that are used in many appli-
cations such as automobile and boat parts, fiberglass structures, and molding compounds.
Polyesters are generally heavily filled and/or contain reinforcing fibers such as glass.
Polyester resins are somewhat lower in cost than epoxy resins and can be formulated to
have very short molding times. They have good resistance to water, good electrical prop-
erties, good resistance to oils and solvents, and high strength-to-weight ratios. Polyester
parts exhibit a high shrinkage rate.
Adhesive bonding of polyester parts could be complicated by the addition of a variety
of additives in the resin to enhance curing, mold release, and surface gloss. In certain poly-
ester resin formulations that are cured in contact with the atmosphere, waxes are included
to shield the free radical polymerization process from inhibition by oxygen. Certain poly-
ester mold release agents are formulated directly into the resin (internal mold release). For
a glossy finished part appearance, thermoplastic polyolefins are sometimes incorporated
into the polyester formulation. Surface preparation must take into consideration these pos-
sible weak boundary layers. The recommended surface preparation is simple solvent clean-
ing and mechanical abrasion. Generally, the same surface preparations as recommended for
epoxies and phenolics are recommended for polyesters.
Polyester parts are frequently bonded with epoxy, polyester, or polyurethane adhesives.
Polyester adhesives, however, form a rigid bond and exhibit a high shrinkage on curing,
resulting in internally stressed joints. A primer is usually not necessary for polyester parts.
Good results have been obtained in the automotive industry with two-part epoxies and
one- and two-part urethanes.54 Good outdoor weather resistance of polyester fiberglass-
aluminum bonds has been reported for epoxy, acrylic, and silicone adhesives.55
Silicone. Silicones are a family of unique polymers that are partly organic and partly
inorganic. Silicones have outstanding thermal stability. Silicone polymers may be filled or
unfilled. They can be cured by several different mechanisms and are available as both flex-
ible and rigid resins. They are low-surface-energy materials and are generally difficult to
bond with adhesives. In fact, the poor bonding characteristics of silicone surfaces are often
used profitably for baked-on release agents and graffiti-resistant paints. Silicone resins are
also used for molding compounds, laminates, impregnating varnishes, high-temperature
paints, and encapsulating materials.
Rigid silicone resins are used as organic coatings, electrical varnishes, laminates, and
circuit-board coatings. The noncoating products are generally filled or reinforced with min-
eral fillers or glass fibers. Thermosetting molding compounds made with silicone resins are
finding wide application in the electronic industry as encapsulants for semiconductor
devices.
Rigid silicone resins also have a low surface energy and are difficult materials to bond
with adhesives or sealants. Parts should be clean and dry, and the best bonding candidate is
366 CHAPTER SIXTEEN
another silicone resin (either rigid or flexible) plus a compatible primer. Flexible silicones
elastomers are discussed in Sec. 16.6.
Unlike the thermosetting resins, the thermoplastic resins will soften on heating or on con-
tact with solvents. They will then harden on cooling or on evaporation of the solvent from
the material. This is a result of the noncrosslinked chemical structure of thermoplastic mol-
ecules. The following are important characteristics of thermoplastic resins that can affect
their joining capability.
• Many thermoplastic compounds are alloyed and really consist of two or more resins.
• Additives and mold release agents are commonly employed in the formulation.
• Thermoplastics exhibit a relatively high degree of water absorption and mold shrinkage.
• Dimensional changes due to moisture migration, thermal expansion, etc., are generally
greater than with other materials.
• The properties of the surface, such as surface energy and crystallinity, may be different
from those of the bulk (this is especially true for thermoplastics that are molded at very
high temperatures).
• Many grades of the same material are available (high-flow, high-density, etc.).
Thermoplastic materials often have a lower surface energy than do thermosetting mate-
rials. Thus, physical or chemical modification of the surface is necessary to achieve accept-
able bonding. This is especially true of the crystalline thermoplastics such as polyolefins,
linear polyesters, and fluoropolymers. Methods used to increase the surface energy and
improve wettability and adhesion include
The reasons that these surface treatments improve the adhesion of epoxy adhesive to
plastic substrates are summarized in Table 16.8.
As with metal substrates, the effects of plastic surface treatments decrease with time, so
it is important to carry out the priming or bonding as soon as possible after surface prepa-
ration. The surface preparation methods suggested in App. F are recommended for con-
ventional adhesive bonding. Greater care must be taken in cleaning thermoplastics than
EPOXY ADHESIVES ON SELECTED SUBSTRATES 367
with thermosets. Thermoplastic parts can be attacked or can swell on contact with certain
solvents. Therefore, cleaning solvent selection must be made depending on the materials
being joined.
Unlike thermosets, many thermoplastics can be joined by solvent cementing or thermal
welding methods. Solvent cementing and thermal welding do not require abrasion or chem-
ical treatment of the plastic surfaces. The surfaces must be clean, however, and free of
impurities that could cause a weak boundary layer. Bond strengths achieved by solvent or
thermal welding are generally as high as or higher than those of adhesive bonding. Bond
strengths are often greater than 80 percent of the strength of the substrate material.
Acetal. Acetals are among the group of high-performance engineering thermoplastics that
resemble nylon somewhat in appearance but not in properties. Acetals are strong and tough
and have good moisture, heat, and chemical resistance. There are two basic types of acetals:
the homopolymers by DuPont and the copolymers by Celanese. The copolymers are more sta-
ble in long-term, suitable for high-temperature service, and are more resistant to hot water.
368 CHAPTER SIXTEEN
The homopolymers are more rigid, stronger, and have greater resistance to fatigue. Both
types of acetal resins are degraded by uv light.
Because acetals absorb a small amount of water, the dimensions of molded parts are
affected by their water content, which varies with the relative humidity of the environment.
The absorbed water should be considered when bonding because it could migrate to the
interface during exposure to elevated temperatures from the joining process or from the ser-
vice environment. This released moisture could create a weak boundary layer. Dimensional
changes must also be considered when acetal parts are bonded to other substrates and then
exposed to changing temperature and humidity conditions.
The acetal surface is generally hard, smooth, glossy, and not very easy to bond with
adhesives. The nonstick and solvent-resistant nature of acetal requires that the surfaces be
specially prepared before adhesive bonding can occur. Once prepared, the surface can then
adhere to like substrates or to others.
Because of the solvent and chemical resistance of acetal copolymer, special etching
treatments have been developed for surface preparation prior to adhesive bonding. A
chromic acid etch and a hydrochloric acid etch have been suggested. Acetal parts that have
been formed by heat treatment or machining should be stress-relieved before etching.
Acetal homopolymer can be effectively bonded with various adhesives after the surface
is sanded, etched with a chromic acid solution at elevated temperature, or “satinized.”
Satinizing is a patented process developed by DuPont for preparing Delrin acetal homopoly-
mer for painting, metallizing, and adhesive bonding. In this process, a mild acidic solution
produces uniformly distributed anchor points on the adherend surface. Adhesives bond
mechanically to these anchor points, resulting in strong adhesion to the surface of the
homopolymer. Performance of various epoxy adhesives on acetal homopolymer is shown
in Table 16.9. Oxygen plasma and corona discharge treatment have also shown to be effec-
tive on acetal substrates.59 Epoxies, nitrile, and nitrile phenolics can be used as adhesives
with acetal homopolymer substrates that are treated first.
Epoxies, isocyanate cured polyester, and cyanoacrylates are used to bond acetal
copolymer. Generally, the surface is treated with sulfuric-chromic acid. Epoxies have shown
150- to 500-psi shear strength on sanded surfaces and 500- to 1000-psi on chemically
treated surfaces. Two-component epoxies give slightly lower bond strengths. However,
they bond acetal to itself and to many other materials.
Acrylic. Acrylic resins (polymethyl methacrylate) have exceptional optical clarity and
good weather resistance, strength, electrical properties, and chemical resistance. They have
low water absorption characteristics. However, acrylics are attacked by strong solvents,
gasoline, acetone, and similar organic fluids.
Adhesive bonded acrylic joints usually give lower strength than solvent- or heat-welded
joints. Cyanoacrylate, epoxy, and thermosetting acrylic adhesives offer good adhesion but
poor resistance to thermal aging. Epoxy adhesive will generally provide tensile shear
strength on acrylic substrates of greater than 3 MPa.61
Surface preparation for bonding can be accomplished by wiping with methanol, ace-
tone, MEK, trichloroethylene, isopropanol, or detergent; abrading with fine abrasive
media; wiping with a clean, dry cloth and repeating a clean solvent wipe. Care must be
exercised, however, when one is contacting solvent with molded acrylic parts. Stress crack-
ing or crazing can occur due to a solvent attack at highly stressed areas within the molded
part. The tendency for stress cracking can be significantly reduced by annealing the part at
a temperature slightly below the heat distortion temperature.
Chlorinated Polyether. This thermoplastic resists most solvents and is attacked only by
nitric acid and fuming sulfuric acids. Thus, it is not capable of being solvent-cemented.
Chlorinated polyether parts can be bonded with epoxy, polyurethane, and polysulfide-
epoxy adhesives after treatment with a hot chromic acid solution. Tensile shear strength of
1270 psi has been achieved with an epoxy-polysulfide adhesive.
Fluorocarbons. There are eight types of common fluorocarbons. They differ primarily
by the concentration and arrangement of fluorine atoms along their molecular chain.
Chemical types, suppliers, and trade names are given in Table 16.10. Like other plastics,
each type of fluorocarbon is available in several different grades. They differ principally in
the way they are processed and formed, and their properties vary over the useful tempera-
ture range.
The original basic fluorocarbon, and perhaps the most widely known one, is tetrafluo-
roethylene (TFE). It has the optimum electrical and thermal properties and almost complete
moisture resistance and chemical inertness. However, TFE does cold-flow or creep at mod-
erate loading and temperatures. Filled modifications of TFE resins are available; these are
generally stronger than unfilled resins. Fluorinated ethylenepropylene (FEP) is similar to
TFE except that its operating temperature is limited to 200°C. FEP is more easily processed,
370 CHAPTER SIXTEEN
and it can be molded, which is not possible with TFE. Ethylene-tetrafluoroethylene copoly-
mer (ETFE) is readily processed by conventional methods including extrusion and injec-
tion molding. Perfluoroalkoxy (PFA) is a class of melt-processable fluoroplastics that
perform successfully in the 260°C area. Chlorotrifluoroethylene (CTFE) resins are also
melt-processable. They have greater tensile and compressive strengths than TFE within
their service temperature range. Ethylene-chlorotrifluoroethylene copolymer (E-CTFE) is
a strong, highly impact-resistant material. E-CTFE retains its strength and impact resistance
down to cryogenic temperatures. Vinylidene fluoride (PVF2) is another melt-processable flu-
orocarbon with 20 percent lower specific gravity compared to that of TFE and CTFE. Thus,
it is economical to use PVF2 in parts requiring a useful temperature range from −62 to
150°C. Polyvinyl fluoride (PVF) is manufactured as film and has excellent weathering and
fabrication properties. It is widely used for surfacing industrial, architectural, and decora-
tive building materials.
Because of their high thermal stability and excellent resistance to solvent, fluorocarbons
cannot be joined by solvent cementing, and they are very difficult to join by thermal weld-
ing methods. Because of their inertness and low surface energy, they also tend to be diffi-
cult materials to join by adhesive bonding. Surface treatment is necessary for any practical
bond strength to the fluorocarbon parts.
Surface preparation consists of wiping with acetone; treating with commercially avail-
able sodium naphthalene solutions; washing again with acetone and then with distilled or
deionized water; and drying in a forced-air oven at 40°C. Because of the hazardous nature
of these surface treatment chemicals, the user must exercise extreme caution and follow all
the manufacturers’ recommendations. The sodium naphthalene etching process removes
fluorine atoms from the surface of the substrate, leaving it with a higher surface energy and
a surface that is more wettable by conventional adhesives. The sodium naphthalene treated
fluorocarbon surface is degraded by uv light and should be protected from direct exposure.62
Sodium napthalene solutions in tetrahydrofuran solvent have been used to provide sur-
face treatment of fluorocarbon surfaces for several decades. A commercial surface treat-
ment based on this process has been Tetra-Etch.63 Recently the manufacturers of
Tetra-Etch introduced Fluoro-Etch, which has several advantages. Fluoro-Etch requires
no refrigeration, is easier to handle, and offers significant cost, efficiency, and safety
advantages. It offers a 1-year shelf life and is manufactured in both the United States and
Europe.
Flexible epoxy adhesives generally give good bond strengths to treated fluorocarbon
surfaces. Table 16.11 shows the effect of surface treatments on the bondability of epoxy
EPOXY ADHESIVES ON SELECTED SUBSTRATES 371
Nylon (Polyamide). Nylons, also known as polyamides, are strong, tough thermoplastics
with good impact, tensile, and flexural strengths from freezing temperatures up to 150°C.
They provide excellent low-friction properties and good electrical resistivities. Four com-
mon varieties of nylon are identified by the number of carbon atoms in the diamine and
dibasic acid that are used to produce that particular grade. They are referred to as nylon 6,
nylon 6/6, nylon 6/10, and nylon 11. These materials generally vary by their processing
characteristics and their dimensional stability.
All nylons absorb some moisture from environmental humidity. Moisture absorption
characteristics must be considered in designing and joining these materials. They absorb
from 0.5 to 2 percent by weight of moisture after 24-h water immersion. Freshly molded
objects contain less than 0.3 percent moisture since only dry molding powder can be suc-
cessfully molded. Once molded, these objects absorb moisture when they are exposed to
humid air or water. The amount of absorbed moisture increases until an equilibrium condi-
tion is reached based on the relative humidity of the environment. Equilibrium moisture
contents of two commercial nylon resins for two humidity levels are as follows:
The absorbed water can create a weak boundary layer under certain conditions.
Generally, parts are dried to less than 0.5 percent moisture before bonding. Caution must
be observed when one is mating the nylon to another substrate. The nylon part grows and
shrinks due to ingress and egress of moisture from within the substrate. This leads to
stresses at the interface that may cause warpage and degradation of the bond strength.
Various commercial adhesives have been used to provide bond strength with nylon on
the order of 250 to 1000 psi. Priming of nylon adherends with a composition based on resor-
cinol formaldehyde, isocyanate modified rubber, and cationic surfactants has been reported
to provide improved joint strength. Some epoxy, resorcinol formaldehyde, phenol-resorcinol,
and rubber-based adhesives have been found to produce satisfactory joints between nylon
and metal, wood, glass, and leather. Exposure of nylon 6 to oxygen and helium plasmas for
30 s to 1 min improved the adhesion of two-part epoxy adhesives.66
372 CHAPTER SIXTEEN
Polycarbonate. Polycarbonates are among those plastic materials that are grouped as
engineering thermoplastics because of their high-performance characteristics in engineer-
ing designs. Polycarbonates are especially outstanding in impact strength, having strengths
several times higher than those of other engineering thermoplastics. Polycarbonates are
tough, rigid, and dimensionally stable. Polycarbonate resins are available in either trans-
parent or colored grades. An important molding characteristic is the low and predictable
mold shrinkage, which sometimes gives polycarbonates an advantage over nylons and
acetals for close-tolerance parts. Polycarbonates are also alloyed with other plastics to
increase strength and rigidity. Polycarbonates are somewhat hygroscopic, like nylon, so it
is important to keep the humidity low before bonding.
Polycarbonate plastics are generally joined by solvent cementing or thermal welding
methods. However, caution needs to be used when one is solvent-cementing because these
plastics can stress-crack in the presence of certain types of solvents. When one is joining
polycarbonate parts to metal parts, a room temperature curing adhesive is suggested to
avoid stress in the interface caused by differences in thermal expansion. When adhesives
are used to join polycarbonate, the epoxies, urethanes, and cyanoacrylates are generally
chosen. Adhesive bond strengths with polycarbonate are generally 1000 to 2000 psi.
Recommended solvents for cleaning are methanol, isopropanol, petroleum ether, hep-
tane, and white kerosene. Ketones, toluene, trichloroethylene, benzol, and a number of
other solvents including paint thinners can cause crazing or cracking of the polycarbonate
surface. After cleaning, possible surface preparations include flame treatment or abrasion.
In flame treatment, the part is passed through the oxidizing portion of a propane flame.
Treatment is complete when all surfaces have a high gloss, and then the part must be cooled
before bonding.
10
Exposed to uv light
in presence of solvent Etched in
chromic acid
Single-lap shear joint strength, MPa
Flame-treated
Exposed to uv light
Untreated
0
25 30 35 40 45
Critical surface tension γc, mN/m2
FIGURE 16.4 Tensile shear strength of polyethylene pretreated by various methods before
bonding with an epoxy adhesive.67
polyethylene treated for 10 min in an oxygen plasma and bonded with an epoxy adhesive.68
Table 16.12 presents the tensile strength of epoxy-bonded polyethylene and polypropylene
joints after plasma surface treatments.
Flexibilized epoxy adhesives have moderate strength on flame and corona treated poly-
olefin substrates. Elevated cure temperature results in better adhesion because of more effi-
cient wetting of the substrate surface. Table 16.13 shows a starting formulation for an
epoxy adhesive that develops high peel strength to many difficult-to-bond substrates such
as polyethylene, thermoplastic rubber, and polyester film.
TABLE 16.13 Starting Formulation for an Epoxy Adhesive for Polyolefin, Polyester,
and Thermoplastic Substrates70
Part A
DGEBA epoxy resin (EPON 828, Resolution Performance Products) 80
Butyl glycidyl ether reactive diluent (HELOXY 61 Resolution 20
Performance Products)
Adhesion promoter (Silquest A-172) 3
Part B
Polyamide (EPI-CURE 3125, Resolution Performance Products) 180
Property
These plastics are quite inert; thus, compatibility with other substrates does not pose
major problems. The terephthalates have high tensile and tear strengths, excellent chemi-
cal resistance, good electrical properties, and an operating temperature range from −55 to
200°C. These materials are generally joined with adhesives, and surface treatments are
used to enhance adhesion, if required.
Commonly used adhesives for both PET and PBT substrates are isocyanate cured poly-
esters, epoxies, and urethanes. Surface treatments recommended specifically for PBT
include mechanical abrasion and solvent cleaning with toluene. Gas plasma surface treat-
ments and chemical etch have been used where maximum strength is necessary.
Polyethylene terephthalate cannot be solvent-cemented or heat-welded. Adhesives are the
prime way of joining PET to itself and to other substrates. Only solvent cleaning of PET sur-
faces is recommended as a surface treatment. The linear film of polyethylene terephthalate
(Mylar) provides a surface that can be pretreated by alkaline etching or plasma for maximum
adhesion, but often a special treatment such as this is not necessary. An adhesive for linear
polyester has been developed from a partially amidized acid from a secondary amine, reacted
at less than stoichiometric with a DGEBA epoxy resin, and cured with a dihydrazide.72
it is also available as film and molded parts. PEEK can be either adhesive-bonded or ther-
mally welded using ultrasonic, friction, or hot plate welding techniques. When one is weld-
ing PEEK, it must be remembered that the melting point is very high, and considerable
amounts of energy must be put into the polymer during welding to achieve a good bond.
PEEK parts can be bonded with epoxy, cyanoacrylate, silicone, and urethane adhesives.
The epoxies give the strongest bond. High-temperature epoxy adhesives may be necessary
in high-temperature applications. Surfaces must be clean, dry, and free of grease.
Isopropanol, toluene, and trichloroethylene can be used to clean PEEK surfaces prior to
mechanical abrasion. Surface roughening and flame treatment or etching may improve
bond strengths. PEEK composites that have been surface-treated by plasma activation
show excellent bond strength with epoxy adhesives.73 Polyaryletherketone and polyether-
ketone parts can also be joined with adhesives by the methods described above.
TABLE 16.14 Joint Strengths for Polystyrene-Aluminum Tensile Shear Joints Exposed
to Various Environments74
Polyvinyl Chloride. Polyvinyl chloride (PVC) is perhaps the most widely used type of
plastic in the vinyl family. PVC is a material with a wide range of flexibility. One of its
basic advantages is the way it accepts compounding ingredients. For instance, PVC can be
elasticized with a variety of plasticizers to produce soft, yielding materials to almost any
desired degree of flexibility. Without plasticizers, PVC is a strong, rigid material that
can be machined, heat-formed, or welded by solvents or heat. It is a tough material, with
378 CHAPTER SIXTEEN
Adhesive
high resistance to acids, alcohol, alkalies, oils, and many other hydrocarbons. It is available
in a wide range of forms and colors. Typical uses include profile extrusions, wire and cable
insulation, and various foam applications. It is also made into both rigid and flexible film
and sheets.
Plasticizer migration from the vinyl part into the adhesive bond line can degrade the
strength of the joint. Adhesives must be tested for their ability to resist the plasticizer.
PVC can be made with a variety of plasticizers. An adhesive suitable for a certain flexi-
ble PVC formulation may not be compatible with a PVC from another supplier. Nitrile
rubber adhesives have been found to be very resistant to plasticizers and are often the pre-
ferred adhesive for flexible PVC films. However, certain epoxy adhesive formulations
have also been found to provide excellent adhesion to flexible PVC substrates. Several
such starting formulations are presented in Table 16.7. A comparison of the performance
of several classes of adhesive when bonding PVC to itself and to various other materials
is given in Table 16.15.
16.4 COMPOSITES
Modern structural composites are a blend of two or more components. One component is
generally made of reinforcing fibers, either polymeric, carbon, or ceramic. The other com-
ponent is generally made up of a resinous binder or matrix that is polymeric in nature. The
fibers are strong and stiff relative to the matrix. Composites are generally orthotropic mate-
rials (having different properties in two different directions). When the fiber and matrix are
joined to form a composite, they both retain their individual identities and both directly
influence the composite’s final properties. The resulting composite is composed of layers
(laminates) of the fibers and matrix stacked to achieve the desired properties in one or more
directions. The reinforcing fiber can be either continuous or discontinuous in length. The
fiber’s strength and stiffness are usually much greater than those of the matrix material.
The commonly commercially available fibers are
• Glass
• Polyester
• Graphite
EPOXY ADHESIVES ON SELECTED SUBSTRATES 379
• Aramide
• Polyethylene
• Boron
• Silicon carbide
• Silicon nitride, silica, alumina, alumina silica
Glass fiber composites are the most common type of composite. However, graphite,
aramide, and other reinforcements are finding applications in demanding aerospace func-
tions and in premium sporting equipment such as fishing rods, tennis rackets, and golf clubs.
The resin matrix can be either thermosetting or thermoplastic. Thermosetting resins
such as epoxy, polyimide, polyester, and phenolic are used in applications where physical
properties are important. Polyester and epoxy composites make up the bulk of the ther-
moset composite market. Of these two, polyesters dominate by far. Reinforced with glass
fiber, these are known as fiberglass-reinforced plastics (FRPs). FRPs are molded by layup
and spray-up methods or by compression molding either a preform or sheet molding com-
pound (SMC).
Thermoplastic matrix composites are generally employed where high-volume and eco-
nomic considerations exist such as in the automotive and decorative paneling industries.
Thermoplastic resin-based composites range from high-priced polyimide, polyethersul-
fone, and polyetheretherketone to the more affordable nylon, acetal, and polycarbonate
resins. Practically all thermoplastics are available in glass-reinforced grades.
Resin-based composites are usually defined as either conventional or advanced.
Conventional composites usually contain glass or mineral fiber reinforcement, and some-
times carbon fiber, either alone or in combination with others. Conventional composites are
usually produced in stock shapes such as sheet, rod, and tube. There are many methods of
processing composite materials. These include filament winding, layup, cut fiber spraying,
resin transfer molding, and pultrusion.
Advanced composites is a term that has come to describe materials that are used for the
most demanding applications, such as aircraft, having properties considerably superior to
those of conventional composites and much like metals. These materials are “engineered”
from high-performance resins and fibers. The construction and orientation of the fibers are
predetermined to meet specific design requirements. Advanced composite structures are
usually manufactured in specific shapes. An advanced composite can be tailored so that the
directional dependence of strength and stiffness matches that of the loading environment.
Thermoset composites are joined by either adhesive bonding or mechanical fasteners.
Thermoplastic composites offer the possibility of thermal welding techniques, adhesives,
or mechanical fasteners for joining. Composites are also often joined with a combination
of mechanical fasteners and adhesives. Many manufacturers distrust adhesive bonds in
applications where joints undergo large amounts of stress (e.g., aircraft structures).
Mechanical fasteners must be sized to avoid fiber crushing and delamination; adhesives
must balance strength and flexibility. The joining of composite materials involves some
special problems not faced with other materials.
A significant advantage is that adhesive bonding does not require the composite to be
drilled or machined. These processes cut through the reinforcing fibers and drastically
weaken the composite. A cut edge will also allow moisture or other chemicals to wick deep
into the composite along the fiber-matrix interface, thereby further weakening the structure.
Much of what we know about bonding to composite materials has come through the
aerospace industries. The early studies on adhesives, surface preparation, test specimen
preparation, and design of bonded composite joints reported for the PABST Program76
gave credibility to the concept of a bonded aircraft and provide reliable methods of trans-
ferring loads between composites and metals or other composites.
380 CHAPTER SIXTEEN
Adhesives that give satisfactory results on the resin matrix alone may also be used to
bond composites. The three adhesives most often used to bond composites are epoxies,
acrylics, and urethanes.
Epoxies are especially reliable when used with epoxy-based composites because they
have similar chemical characteristics and physical properties. Room temperature curing
adhesives are often used to bond large composite structures to eliminate expensive fixtur-
ing tools and curing equipment required of higher-temperature cure adhesives. However,
room temperature epoxies require long cure times, so they are not suitable for large, high-
speed production runs. Some of the lower-temperature composite materials are sensitive to
the heat required to cure many epoxies. Epoxies are too stiff and brittle to use with flexible
composites.
Structural adhesives that are commonly used for composites are supplied in two basic
forms: semisolid B-stage film and thixotropic pastes. The film adhesives are cast or
extruded onto carrier fabrics or films and partially cured to a semisolid. They can easily be
handled, cut, and applied to the joint area. There is no need for mixing, metering, or dis-
pensing of liquid components. In use, these adhesive systems are activated by heat and
pressure. The semisolid B-stage film liquefies briefly on application of heat and then cures
to an insoluble state. Epoxy, polyimides, epoxy-nylons, epoxy-phenolic, and nitrile-phenolic
adhesives are available as B-stage film.
Paste adhesives are supplied as either one- or two-component adhesive systems. They
can be used in applications where pressure cannot be applied. Some two-part pastes cure at
room temperature after the appropriate proportions are mixed. Epoxy, urethane, and acrylic
adhesives are all available as paste adhesives.
Surface preparation of composite parts for adhesive bonding will depend on the spe-
cific adherend and adhesive. Recommended surface preparations of many composites
simply consist of a solvent wipe to remove loose dirt, oil, and mold release, followed by
a mechanical abrading operation. Abrasion should be done carefully to avoid damaging
the composite’s surface fibers. A degree of abrasion is desired so that the glaze on the
resin surface is removed but the reinforcing fibers are not exposed. Many surface abra-
sion methods have been applied to composite parts, and all these have some merit. These
abrasion processes include light sanding, grit honing, vapor honing, Scotch-Briting (3M
Company), and other methods. In some cases, a primer may be used to coat the compos-
ite before the adhesive is applied.
One surface preparation method that is unique for composites employs a “peel” or
“tear” ply.77 Utilization of the peel ply is illustrated in Fig. 16.5. With this technique, a
closely woven nylon or polyester cloth is incorporated as the outer layer of the composite
during its production layup. This outer ply is then torn or peeled away just before bonding.
The tearing or peeling process fractures the resin matrix coating and exposes a clean, fresh,
roughened surface for the adhesive. This method is fast and eliminates the need for solvent
cleaning and mechanical abrasion.
Although the fabrication and joining technology that has been developed for thermoset
composites is well advanced and established, the enabling technology for bonding ther-
moplastic composites is still in the relatively early stages of development. However, since
thermoplastic composites can be softened by the action of heat and solvent, welding tech-
niques that are unsuitable for use with thermosets can be used for joining thermoplastic
composites.
Reinforced thermoplastic parts are generally abraded and cleaned prior to adhesive
bonding. However, special surface treatment such as used on the thermoplastic resin matrix
may be necessary for optimum strength. Care must be taken so that the treatment chemi-
cals do not wick into the composite material and cause degradation. It may not be a good
idea to use chemical surface treatment without first verifying that the treatment does not
degrade the substrate.
EPOXY ADHESIVES ON SELECTED SUBSTRATES 381
Tear ply
Tear ply
Bonding
surface
Fiberglass reinforced
plastic laminate
FIGURE 16.5 Structural reinforced composite with tear ply to produce fresh bonding
surface.78
Plastic foams are manufactured from thermoplastic and thermoset resins in various forms.
The main pitfall in joining plastic foam is that of (1) causing the foam to swell or collapse by
contact with a solvent or monomer and (2) having the adhesive alter the properties of the
foam through its absorption into the foam. Adhesion and joining are usually not a serious
problem because of the porous nature of the foam.
Note that there are closed-cell and open-cell foams. Adhesives may spread or wick
deeply into the open-cell variety, thereby affecting the resulting mechanical properties of the
foam and perhaps even weakening the foam. When foam is bonded to another less porous
substrate, the adhesive could be applied to the nonfoam substrate to minimize the wicking
and ingress of the adhesive into the body of the foam. With the closed-cell variety, the adhe-
sive cannot wick deeply into the foam, but usually the foam’s skin must be machined or
abraded to allow for some surface roughness for the adhesive to mechanically attach.
There are also low-surface-energy foams, such as polyethylene, that require either sur-
face treatment or special adhesives for bonding. Fortunately, extremely strong bonds are
generally not required because the foam has a relatively low cohesive strength. Therefore,
simple cleaning is generally the only surface preparation required. The surface treatments
that are recommended are those that are described in the previous section for the parent
plastics. However, the possibility of wicking of the chemical compounds into the foam and
degradation of the foam must be considered.
382 CHAPTER SIXTEEN
16.6 ELASTOMERS
There are over 30 broad groups of chemical types of elastic polymers. These are arranged
by ASTM D 1418 into categories of materials having similar chemical chain structures.
There are several problems with joining elastomer materials.
One problem is the significant variation that can exist within a given chemical type. This
is due to differences in average molecular weight, molecular weight (MW) distribution,
polymerization processes, variation of structural arrangement, copolymer or terpolymer
ratios, etc. There also can be significant differences in the compounding recipe for a given
product. This is due to differences and latitude of choice in fillers and reinforcing agents,
liquid plasticizers, compounding agents, and the like. Thus, almost an infinite number of
compounds are possible for a given generic elastomer such as neoprene or nitrile, and each
variation can significantly affect the adhesion properties of the material.
Elastomer materials specifications usually do not focus on the adhesive properties, but
mainly address the chemical and physical properties of the rubber. Thus, the supplier has
wide latitude within the specification to make changes in the compound formulation that
could be disastrous to the adhesive bond. One solution is to qualify every new lot of elas-
tomer material for adhesion as well as the more standard properties.
Besides curing systems, fillers, and plasticizers, an elastomeric compound may contain
protective chemicals such as antioxidants, antiozone agents, waxes, and fungicides. Some
of these are purposefully designed to “bloom” or to come to the surface of the elastomer
either during processing or on aging. These weak boundary layers often cannot be removed
prior to bonding because their supply to the interface is relatively unlimited by the capac-
ity of the bulk elastomer.
Another problem in joining elastomers with adhesives is that since they are deformable
materials, it is easy to develop internal stresses at the bond interface. These stresses could
adversely affect the bond strength and permanence of the joint. Minimal pressure to achieve
close substrate contact with the adhesive is all that is necessary when bonding with elastomers.
Bonding of already vulcanized elastomers to themselves and to other materials is gener-
ally completed by using a flexible thermosetting adhesive such as epoxy-polyamide or
10
9
More bondable
8
7
6
Index
5
4
3
2
1
0
Nitrile SBR Natural Ethylene Fluoro-
rubber rubber propylene elastomer
Neoprene Hypalon Butyl Urethane Silicone
rubber
FIGURE 16.6 Bondability index of common and specialty elastomers.79
EPOXY ADHESIVES ON SELECTED SUBSTRATES 383
polyurethane. These adhesives offer excellent bond strength to most elastomers. Figure 16.6
illustrates a bondability index of common and specialty elastomers.
Surface treatments consist of washing with solvent, abrading, or, in the most demand-
ing applications, cyclizing with acid. The most common elastomers to be bonded in this
way include nitrile, neoprene, urethane, natural rubber, SBR, and butyl rubber. It is more
difficult to achieve good bonds with silicones, fluorocarbons, chlorosulfonated polyethyl-
ene, and polyacrylate.
in cutting or sawing prior to bonding. Damaged wood fibers provide a cohesively weak area
near to the adhesive, and they will become a failure mode under stress just as the weakest
link in a chain.
Thin glue lines are preferred in bonded wood joints. The prevailing theory is that pene-
tration of the adhesives into the wood is not necessary for good adhesion. Good bonding
need occur only to the outside cellulose fibers. Thick bond lines tend to give erratic joint
quality because of uneven stresses introduced in the joint. Thus, the best surface finish is
one that does not tear wood fibers and provides sufficiently close tolerances that the parts
fit together uniformly and closely. Usually, such surface preparation can be accomplished
by standard wood machining operations.
The major synthetic adhesives used for bonding wood include urea, phenol, and melamine
formaldehyde: resorcinol formaldehyde, phenol resorcinol, and polyvinyl acetate emulsions.
More recently one-component, moisture cured polyurethane adhesives have become pop-
ular for bonding wood. Natural adhesives such as casein and animal glues are also often
used for general-purpose wood bonding. Epoxies have been used for certain specialized
wood joining applications such as when wood is bonded to metal substrates.
Glass and ceramic substrates are generally high-surface-energy materials, and most adhesives
wet them readily. One problem in bonding optically clear glass is to select an adhesive that is
optically clear and does not change the optical characteristics of the glass. Another problem
is that shrinkage of the adhesive or differences in thermal expansion coefficients could pro-
vide high internal stresses that cause catastrophic failure of the brittle glass substrate.
Adhesives used for glass substrates are generally transparent, heat-setting resins that are
water- and uv-resistant to meet the requirements of outdoor applications. They are usually
flexibilized systems so as not to place stress on the glass substrate either after cure or dur-
ing thermal cycling. These adhesives include polyvinyl butyl, phenolic butyral, phenolic
nitrile, neoprene, polysulfide, silicone, vinyl acetate, and flexible epoxy adhesives. Optical
adhesives used for bonding glass lenses are usually styrene modified polyesters and styrene
monomer-based adhesives. Epoxies, especially UV cured epoxy adhesives, are also begin-
ning to be used in this application.
Adhesives used for bonding glass and glazed ceramics should have minimum VOC con-
tent, since these substrates are nonporous. Epoxy adhesives, being 100 percent solids materi-
als, are probably best for these applications. Thermosetting acrylic is good for bonding glass
and ceramics to thermoplastic polymers and to metal surfaces. Where large areas and mate-
rials with greatly differing coefficients of thermal expansion are being bonded, polysulfides
or urethanes should be considered, since these adhesives have very high elongation.
Usually solvent cleaning is the only surface preparation needed for bonding glass.
However, (1) solvent cleaning coupled with light sandblasting or (2) a chromium trioxide etch
followed by ultrasonic detergent bath has been suggested for highest bond strength. Mechanical
or chemical surface treatment can degrade the optical transmission quality of glass.
Most ceramic surfaces, like metal surfaces, have an adsorbed layer of water that may
cause difficulty in bonding. If this water is partially removed by heating prior to bond for-
mation, atmospheric moisture may diffuse into the bond during service and displace adhe-
sive molecules from the interface since they will be preferentially adsorbed on the glass
interface. To overcome this problem, coupling agents, typically silanols, have been devel-
oped to enhance the attraction between the adhesive and the glass or ceramic surface.
Coupling agents can also provide a degree of stress release at the interface and reduce the
thermal expansion-related stresses between the substrate and the adhesive.
EPOXY ADHESIVES ON SELECTED SUBSTRATES 385
Adhesives for glass- or ceramic-to-metal seals should never become fully rigid because
thermal expansion rates for adhesives are much higher than those for the bonded materials.
Temperature changes can lead to high stresses, which can cause cracking and joint failure.
Nitrile rubber epoxies, phenolic resin blends, silicones, cyanoacrylates, and anaerobics are
some of the important organic adhesives that can be used to make good glass- or ceramic-
to-metal seals. The silicones can be used from −73 to 232°C while maintaining good flex-
ibility, but they are relatively costly. The cyanoacrylates are especially useful in bonding
glass to metals, but they are less useful for joining porous ceramics because the porosity
will absorb most of the adhesive and inhibit its cure.
Structural sandwich panels represent what is probably one of the most common appli-
cations for structural adhesives. Facing materials can be made to adhere to core materi-
als, such as aluminum or paper honeycomb, to give a high composite strength-to-weight
ratio.
Adhesives with filleting properties are required for honeycomb cores. A modified phe-
nolic is often used with aluminum honeycomb for high strength, while a neoprene- or
nitrile-based organic solvent type of adhesive is often used with impregnated paper honey-
comb. Epoxy adhesives are also commonly used in the fabrication of honeycomb sandwich
panels.
Facings of metal, paper, or plastic may be combined with a variety of solid-core mate-
rials. The adhesive must transmit shear stress from the facing to the core. It must also be
able to resist the tendency of the thinner facing to buckle under design loads or under
thermal stresses caused by variations in the panel components’ coefficients of thermal
expansion.
Adhesives for sandwich structures such as honeycomb and faced cores must have dif-
ferent properties than other adhesives. To achieve a good attachment to an open-cell core,
such as honeycomb, the adhesive must have a unique combination of surface wetting and
controlled flow during its early stages of cure. This controlled flow prevents the adhesive
from flowing down the cell wall and leaving a low-strength top-skin attachment and an
overweight bottom-skin attachment. The adhesive must resist being squeezed out from
between faying (close-fitting) surfaces when excessive pressure is applied during cure to a
local area of the part. Many adhesives are formulated to achieve good core filleting and are
subsequently given controlled flow by adding an open-weave cloth or fibrous web, cast
within a thicker film of adhesive. This cloth, called a scrim cloth, then prevents the faying
surfaces from squeezing out all the adhesive, which would result in an area of low bond
strength.
The adhesive for sandwich construction must also have good toughness. It must be
resistant to loads that act to separate the facings from the core under either static or dynamic
conditions. Experience has shown that greater toughness in the bond line usually equates
to greater durability and longer service life. If the facing is flexible enough that it can be
rolled on a drum, the climbing drum peel test (ASTM D 1781) is the most common test to
measure toughness in sandwich construction. The resulting value of peel strength will vary
considerably depending on the properties of the facing, toughness of the adhesive, amount
of adhesive used, density of the core, cell size of the core, direction of peel, adequacy of
surface preparation, and degradation of the joint.
Common adhesives for sandwich construction include nitrile phenolic films, modified
epoxy pastes and films, epoxy-nylon films, polyimide films, and modified urethane liquid
386 CHAPTER SIXTEEN
and pastes. Accessory materials that are commonly used in the joining process are core-
splicing adhesives, syntactic foams, tapes, and corrosion-inhibiting primers.80
16.10 CONCRETE
Concrete provides a substrate surface that changes during its cure and also during environ-
mental exposure. The main difficulty with bonding concrete is due to the presence of mois-
ture. This moisture can be retained in fresh concrete and may be present in old concrete due
to the environment.
Concrete is a substrate whose surface characteristics are also likely to be affected by the
environment in which it cures. The surface can be different depending on the temperature
and humidity conditions during the cure of the concrete.
Once it is cured, concrete has several surface characteristics that are problems for bond-
ing or sealing. The concrete surface is extremely alkaline and will destroy any hydrolysis-
sensitive materials that are present at the interface. It often has a weak, porous surface layer
that must be penetrated or removed before being bonded. Thus, sealers and primers are
commonly used to moisture-proof and strengthen the concrete surface prior to bonding.
The substrate should be free from contaminants such as cement laitance, oils, waxes,
greases, and curing compounds. While the old concrete does not need to be thoroughly dry,
any standing water must be removed before the adhesive is applied.
Adhesives commonly used on concrete must be formulated so that they cure well in the
presence of moisture. The adhesive should have a viscosity that allows it to penetrate the
concrete for mechanical bonding. Epoxy and epoxy-polysulfide adhesives are especially
effective adhesives for concrete. The starting formulations given in Table 16.16 are for an
ambient cure 100 percent solids epoxy and a waterborne epoxy adhesive.
The waterborne emulsion is formulated so that the mixed adhesive can be easily cleaned
with water, yet it provides good water resistance properties of a conventional epoxy adhe-
sive. Bond strength of these adhesives on new concrete is greater than the strength of the
concrete itself.
Parts by weight
Component A B
Part A
DGEBA epoxy resin (EPON 828, Resolution Performance Products) 100
Epoxy resin emulsion (EPI-REZ WD-510, Resolution 100
Performance Products)
Super white silica (C.K. Williams Co.) Optional
Fumed silica (Cab-O-Sil TS 720, Cabot Corp.) 1
Part B
Amidoamine (EPI-CURE 3072, Resolution Performance Products) 35 35
Tap water (added after mixing part A and part B) 53
Property
REFERENCES
1. Petrie, E. M., Chapter 16 “Bonding and Sealing Specific Substrates,” in Handbook of Adhesives
and Sealants, McGraw-Hill, New York, 2000.
2. Snogren, R. C., Handbook of Surface Preparation, Palmerton Publishing Co., New York,
1974.
3. Wegman, R. F., Surface Preparation Techniques for Adhesive Bonding, Noyes Publications, New
York, 1989.
4. Schneberger, G. L., “Adhesives for Specific Substrates,” Chapter 21 in Adhesives in
Manufacturing, G. L. Schneberger, ed., Marcel Dekker, New York, 1983.
5. Brewis, D. M., “Pretreatment of Metals Prior to Bonding,” in Handbook of Adhesion, D. E.
Backham, ed., Longman Scientific and Technical, Essex, England, 1992.
6. Adhesive Bonding of Aluminum Automotive Body Sheet Alloys, The Aluminum Association, Inc.,
New York, NY, 1975.
7. Adhesive Bonding of Aluminum, Reynolds Metals Company, Richmond, VA, 1966.
8. Murphy, J. F., and Page, H. A., American Chemical Society Division Paint Plastics Printing Ink
Chemical Papers, vol. 15, no. 1, 1955, p. 27.
9. Eichner, H. W., and Schowalter, W. E., Forest Products Laboratory Report 1813, Madison, WI,
1950.
10. Wilson, I., et al., “Pretreatment for Bonded Aluminum Structures,” Advanced Materials and
Processes, August 1997.
11. Minford, J. D., “Surface Preparations and Their Effect of Adhesive Bonding,” Adhesives Age,
July 1974.
12. Kabayashi, G. S., and Donnelly, J. P., Boeing Company Report DG-41517, February 1974.
13. Bijlmer, P. F. A., “Influence of Chemical Pretreatments on Surface Morphology and Bondability
of Aluminum,” Journal of Adhesion, vol. 5, 1973, p. 319.
14. Thrall, E., “Bonded Joints and Preparation for Bonding,” AGARD Lecture Series, no. 102,
October 1979.
15. Shannon, R. W., et al, “Primary Adhesively Bonded Structure Technology (PABST) General
Materials Property Data,” Douglas Aircraft Co., McDonnell Douglas Corp., Air Force Flight
Dynamics Laboratory, Technical Report AFFDL-TR-77-107, September 1977.
16. Kim, G., and Ajersch, F., “Surface Energy and Chemical Characteristics of Interfaces of Adhesively
Bonded Aluminum Joints,” Journal of Materials Science, vol. 24, 1994, pp. 676–681.
17. Hartshorn, S. J., “Durability of Adhesive Joints,” in Structural Adhesives, S. R. Hartshorn, ed.,
Plenum Press, New York, 1986. Also in Minford, J. D., Treatise on Adhesion and Adhesives,
vol. 5, R. L. Patrick, ed., Marcel Dekker, New York, 1981, p. 45.
18. Hartshorn, “Durability of Adhesive Joints.”
19. Browne, J., “Aerospace Adhesives in the 90s,” 38th International SAMPE Symposium, May
10–13, 1993.
20. Wilson, I., et al., “Pretreatment for Bonded Aluminum Structures,” Advanced Materials and
Processes, August 1997.
21. Adhesive Bonding Aluminum, Reynolds Metals Company, Richmond, VA, 1966.
22. Sell, W. D., “Some Analytical Techniques for Durability Testing of Structural Adhesives,”
Proceedings 19th National SAMPE Symposium and Exhibition, vol. 19, New Industries and
Applications for Advanced Materials Technology, April 1974.
23. Resolution Performance Products, Starting Formulations 4011 and 4012, Houston, TX, 2004.
24. Magariello, E., and Hannon, M., “Bonding Beryllium Copper Alloys with High Temperature
Adhesives,” Adhesives Age, March 1971.
25. Berylcoat D available from Brush Wellman, Incorporated, Cleveland, OH.
26. Schields, J., Adhesives Handbook, 3d ed., Butterworths, London, 1984.
27. Cagle, C. V., “Surface Preparation for Bonding Beryllium and Other Adherends,” in Handbook of
Adhesive Bonding, C. V. Cagle, ed., McGraw-Hill, New York, 1973.
388 CHAPTER SIXTEEN
17.1 INTRODUCTION
Epoxy adhesives have achieved their commercial success due in no small way to their
processing capabilities. Epoxy chemistry is compatible with (1) a variety of formulating
techniques and processes accessible to the adhesive manufacturer and (2) a variety of appli-
cation and curing methods accessible to the end user.
The role of the adhesive formulator has become so important that it has been associated
with the creation of “new” or “engineered” adhesives. New products actually come both
from the development of new polymeric materials such as resins and curing agents and
from the creation of more versatile and usable adhesives through selecting, blending, fill-
ing, and modifying existing raw materials. With adhesive systems the easiest and most eco-
nomical method of producing new products is arguably through formulating.
Epoxy resins are not ideal adhesives in their natural form; so they are mixed with materials
to improve and enhance their properties and, thereby, make them more useful in a variety of
applications. This process is called compounding or formulating. Compounding is the com-
bining of a base epoxy resin with curing agents, modifiers, additives, reinforcement, fillers, and
other polymers to make the base polymer perform better, cost less, and process more easily.
Under this broad definition there are two groups that perform compounding: (1) the sup-
pliers of the base resins or adhesives and (2) the end users of the adhesive system.
Most often the compounding is done by the formulators who are specialists in com-
pounding methods and equipment. They can marry a set of specifications to a formula and
provide the product at lowest cost. These independent compounders can be subdivided into
two main groups: the proprietary compounder and the custom compounder. Proprietary
compounders compound and sell their own special formulations, and custom compounders
do compounding for someone else.
Adhesive compounding also occurs at the end-user level. This is generally done when
the end user cannot find a product to meet his or her individual needs, to reduce cost asso-
ciated with a middleman, or to make use of other materials (e.g., scrap resin, fillers, etc.)
that are generated during the course of business. Epoxy adhesives are more conducive to
the end user or in-house processor than other polymeric adhesives because of the ease with
which the liquid resins can be mixed with other ingredients. Many of these operators buy
raw materials from various suppliers and do their own blending and formulating.
The role of the epoxy resin manufacturer depends on the nature of the product and the
markets. It is often more convenient for the resin manufacturer to “modify” a base resin,
such as providing an adduct, a graft polymer, or a blend. In this case the modified resin
would likely be shipped to a formulator for incorporation of other fillers and modifiers and
391
Copyright © 2006 by The McGraw-Hill Companies, Inc. Click here for terms of use.
392 CHAPTER SEVENTEEN
for packaging with a curing agent. The resin supplier may also be interested in supplying a
completely formulated adhesive product, but generally the market is not large enough.
Because of the nature of the epoxy resin and the other raw materials used in formulating
an epoxy adhesive, compounding requires a wide range of mixtures from powders, to liquids,
to pastes. The final product may also be supplied in several forms, such as pellets, film, and
tape, and in several packaging systems, such as two-component kits, cartridges, and bubble
packs. This requires a corresponding broad range of processing knowledge and equipment.
The types of compounding machinery and equipment have increased in number and variety
at a rate that is almost equivalent to the addition of new raw materials. This machinery has
become more sophisticated and versatile and in some instances more specialized.
Compounding operations also require extensive testing from quality control checks on
incoming resins and additives through reports and testing on outgoing products. The vari-
ety of test equipment to characterize physical, chemical, environmental, thermal, and rhe-
ological properties is likewise large.
The following section looks at the processes and equipment commonly used in the com-
pounding of epoxy adhesive systems. Since they can be utilized by either the formulator or
the end user, there will be no strict division of the discussion by user. Hazards and safety
issues related to these materials and processes are discussed somewhat in this chapter, but
a more thorough discussion is found in Chap. 18. Test methods, quality control processes,
and standards that are commonly used at the formulation level are addressed in Chaps. 19
and 20.
Just as there are many formulations, there are many ways to compound and process the req-
uisite raw materials into an adhesive formulation. The methods chosen and how they are
used also influence the final physical properties of the adhesive. The important processes
to be considered in compounding and use of epoxy adhesives can be listed as follows:
Liquid epoxy resins are stable for long periods at room temperature. Resins that are cut with
diluent or solvent should be stored in tight containers to prevent loss of volatiles. Also, cer-
tain resins and curing agents (e.g., anhydride curing agents) can be degraded by exposure
PROCESSING OF EPOXY ADHESIVES 393
to ambient moisture and should be stored in tightly closed containers. Moisture absorption
by these materials may affect their rate of cure and other properties.
Storing of materials below room temperature can significantly extend the shelf life of
epoxy products. One-component adhesives are generally stored in a freezer at −40°C and
shipped in the frozen condition. Prolonged exposure to cold storage conditions, particularly
with unformulated epoxy resins, can cause crystallization of the individual components.
Formulated adhesive can also, on occasion, form crystals that look like small lumps or agglom-
erates. If this is the case, the epoxy components should be heated in their containers to 40 to
55°C until the crystals are dissolved. The cover should be loose-fitting but kept on the product.
Once they are at temperature, stir the components with a spatula until homogeneous.
Moderate elevated-temperature storage of epoxy resins to reduce viscosity or prevent
resin crystallization has little effect on most resins, even after several months of storage
time. Storage at temperatures above 55°C is satisfactory for periods of 2 or 3 days for pro-
cessing purposes.
The storage and transfer of raw chemicals requires special considerations relative to the
equipment that come into contact with epoxy resins. Carbon steel is generally satisfactory
for pipe and tank construction and will not rust beneath the epoxy resin. Rust occuring
above the level of the epoxy and accidentally introduced into the resin may be removed by
filters. Epoxy coated steel pipe, aluminum or stainless steel pipe, or epoxy fiberglass com-
posite tubing is used to prevent rusting downstream of the filter.
When one is incorporating ingredients into the epoxy resin or curing agent, the order of
addition is most important. The objective is to maintain as liquid a mix as possible at the
beginning, and then progress stepwise to a stiffer plastic mixture. The completeness of the
mixing may be limited by the most viscous or difficult to incorporate ingredient.
Fillers are normally incorporated during the final stages of the adhesive compounding
process. Fillers should be dried to remove any residual moisture before being added to
either the epoxy resins or the curing agent component. Air circulating ovens are most com-
monly used for the drying processes, but dielectric heating can also be used in certain cases.
It is usually most convenient to add the fillers to a warm resin. However, diluents and
solvents are sometimes employed to lower the viscosity of the system to allow a higher con-
centration of filler in the mixture.
After the filler is mixed in, the formulation should be allowed to return to room tem-
perature and stand for a suitable time until all bubbles have escaped the system. This can
be facilitated by a vacuum degassing process.
For some applications, filler may be incorporated as a third component at the time the
epoxy component is mixed with the curing agent at the bonding site. This is usually the case
with highly filled systems such as are commonly used in the construction industry. It is gen-
erally done to incorporate high levels of fillers or to negate the settling properties experi-
enced with some fillers.
Fillers may be added to low-viscosity resins by conventional sigma-blade mixer or the
same type of equipment used for mixing paint systems. Three roll mills are generally employed
for more viscous systems to obtain thorough wetting of the filler by the resin matrix.
In the epoxy adhesive industry, fillers are generally added as a batch process rather than
a continuous process. Batch systems are especially suited to short-run operation in which
additives and materials are often changed. They allow for greater control of residence time,
shear, and temperature, and they can accommodate a feed rate that is not free-flowing.
Mixing equipment is generally classified in terms of the shear or mixing intensity. High-
shear mixers are used for producing filled compounds. But most often, low- or medium-
shear mixers suffice for compounding epoxy adhesives.
394 CHAPTER SEVENTEEN
Medium-Shear Mixers. High-speed dispersers are designed to use the high peripheral
speed of a mixing element to produce the blend. These machines depend on a high-speed
impeller turning in a mixing vessel. The differences among the various models on the mar-
ket relate to two components: the design of the mixing elements and the design of the mix-
ing chamber.
Some mixing chambers have curved bottoms that enhance movement of materials.
Others have internal baffles or some similar design to increase turbulence and eliminate
dead spots in the chamber. Shafts can enter the chamber from the top (allowing easier
cleanup) or from the bottom (allowing easier vacuum degassing).
The basic high-speed dispersing impeller blade is a flat disk with teeth or similar serra-
tion around the diameter, as shown in Fig. 17.1. On some models, disks may be without
teeth, or several may be stacked up with small gaps between disks so that material between
the disks is sheared as the impeller turns. Sometimes two shafts are used in the same mix-
ing chamber; their disks may interlock and counterrotate, thereby enhancing shear.
Low-Intensity Mixers. Low-intensity liquid mixers are used for simple jobs involving
low-viscosity resins. Impellers are of various designs resembling marine or aircraft pro-
pellers, long slender screws, shafts with paddles, and shafts with two or more impellers.
(See Fig. 17.2.) These units are intended to generate greater motion within the mixture and
greater turbulence at the impeller than the flat blades of the high-speed disperser, which are
intended to go fast enough to shear material, break down agglomerates, and disperse solids
in liquids.
These low-intensity units are less energy-intensive, requiring lower horsepower, but
they do not achieve mixing of the same quality in comparable cycle times. They are lim-
ited in use to such jobs as blending additives and fillers into low- to medium-viscosity
resins.
17.2.3 Packaging
Often the adhesive formulator or supplier can provide added value to the end user and par-
ticipate in the sharing of this benefit. Perhaps one of the more innovative business solu-
tions has been in the ingenious packaging of adhesive systems, especially reactive epoxy
adhesives.
The adhesive user cares only about achieving a reliable assembly with minimal cost.
The end user would rather not be bothered with handling messy and sometimes hazardous
components, carefully metering them out, mixing the components together adequately
without air entrapment, and finally applying the product to the substrate. These are all nec-
essary steps in the production of an adhesive bond; however, they have been almost made
to vanish by innovative packaging. Although the cost of the packaging is certainly more
than that of conventional containers, the savings that result for the end user can be shared
between the adhesive supplier and end user by premium pricing.
The two-component epoxy adhesive, of course, can be packaged as a supply kit with a
matched set of components. In this method metering is eliminated, and the container of one
of the components can be used as a mixing vessel. There are also several well-recognized
preproportioned adhesive containers that incorporate means for mixing and dispensing—
all within the same package. These include side-by-side syringes with attachable plastic
static mixers, coaxial cartridges, and bubble or divider packs with a removable hinge that
can be used for spreading and mixing the products.
Innovative “packaging” systems that might be overlooked at first include
• Barrier and injection cartridges
• Frozen adhesives
• Solid and preformed adhesives
The last two methods are especially ingenious in their simplicity.
396 CHAPTER SEVENTEEN
Barrier cartridges generally have a foil barrier that is held in place inside a cartridge to
separate the reactive components. The barrier is broken by a plunger rod similar to the one
in a cartridge gun. The head on the plunger rod also serves as a mixer to blend the compo-
nents with several repeated back-and-forth movements.
Injection cartridges work on a similar basis. A cartridge is filled with resin, and a cura-
tive is contained within a hollow injection rod. The rod is pushed against a base within the
cartridge to cause a valve to open at the end of the rod, and then a piston is used to push the
curative out of the rod. A mixing head at the end of the injection rod mixes the components.
A plunger is attached to the other end of the cartridge to extrude the mixed adhesive.
Examples of the barrier- and injection-type packaging are shown in Fig. 17.3.
Another package is a semisolid two-component ribbon with hardener and base resin
identified by different colors. Suitable lengths of ribbon are cut off and kneaded together
by hand until the colors combine to form a third color. Once this is completed, the paste-
like material is applied to the joint in a conventional manner.
In some industries, notably the electronics industry, an often overlooked packaging
method is sometimes used—frozen adhesives. With these adhesives the supplier does the
metering and mixing. The mixed product is then placed into a syringe or other form of con-
tainer, and it is immediately brought to a temperature of −67 to −73°C. The cold delays any
chemical reaction and significantly increases the shelf life of the adhesive system. Frozen
containers are packed in dry ice and shipped overnight to the end user. Generally a warm
water bath, microwave, or hot air is used to thaw the adhesive and bring it to a workable
consistency. In this process, premetered and mixed room temperature curing adhesives can
have a shelf life of 4 to 6 weeks when stored in the frozen condition.
Another often overlooked packaging method is a B stage. The B stage is the intermedi-
ate stage in the curing of certain thermosetting resins such as epoxies cured with aromatic
amines. The components are metered, mixed, and allowed to solidify. However, at this
stage the solid is not fully cured to a thermosetting structure. This B-staged solid can be
heated and caused to flow, thereby wetting the substrate and crosslinking. The B-staged
material can be cast to form a certain shape, such as gasket, rod, or film. These preformed
shapes facilitate storage and application to the joint. The B-staged solid may also be ground
into a powder for press-forming or application via electrostatic coating devices. With a B-
staged solid adhesive, the end user avoids the need to meter and mix, safety and health
issues are minimized, waste is greatly reduced or eliminated, and with a preformed adhe-
sive even the application is almost complete.
The method by which the adhesive or sealant is stored can affect the quality of the bond
even before the containers are open. The manufacturer’s storage directions are usually
found in the technical bulletins describing the particular adhesive or sealant. As a general
rule, all adhesives and sealants should be stored in a cool, dry place until it is time for their
use. Certain adhesives are very tolerant of storage conditions, but some types may have to
be stored at low temperature or under special conditions. Some adhesive systems are
affected by light or moisture, and others will require periodic agitation to make sure that
the components do not settle irreversibly.
As organic resins age in storage due to exposure to temperature, moisture, light, or other
environments, the viscosity of the system will usually undergo noticeable change. Once the
storage life is exceeded, the system will generally be too viscous to use, or else it will be
unusable from separation of its components. Figure 17.4 shows the increase in viscosity of
a paste adhesive as a function of storage time at room temperature. Ordinarily, the shelf life
PROCESSING OF EPOXY ADHESIVES 397
Part A
Removable spline
Plunger
Ram inserted
into rod to open
valve and inject catalyst
Mixing head
Catalyst
6
Viscosity, Pa·s
0
Time
FIGURE 17.4 Increase in viscosity of a paste adhesive as a function of time at room
temperature.4
is chosen so that the majority of the viscosity change will occur just after the expiration of
the adhesive’s shelf life.
Quite often the containers of an adhesive or sealant will bear a special label stating the
safe upper limit of temperature during storage. Containers should be checked when received
to see if such a label is present. As soon as the container is received, the user should place
an additional label on it, stating the date received and the date when the shelf life will
expire. This will provide useful information when it comes to scrapping old product or
maintaining the freshest product in inventory. A record should also be kept of the manu-
facturer’s lot number for possible future reference. Containers should be kept tightly
closed, and labels should be kept clean for proper identification.
Three levels of storage temperature are common with adhesive and sealants: room tem-
perature (15 to 30°C), refrigerated (0 to 5°C), and frozen (−20°C). Generally adhesives or
sealants that are delivered as multiple-component systems have long shelf lives and are
storable for very long times. They can usually be stored for 6 to 12 months at room tem-
perature and often longer. Single-component systems, where the catalyst and base resin are
combined in the same product, generally require refrigerated or frozen storage depending
on the type of catalyst used and the reactivity of the system. These systems may have stor-
age lives of only several weeks or months at the conditions specified.
Extended exposure of the uncured material to temperatures or conditions outside those
recommended by the manufacturer will cause change in physical properties of the uncured
material and will likely reduce its resulting cohesive and adhesive strength. The reactions
that occur due to ambient storage conditions are described in Chap. 3. Figure 17.5 shows
the effect of aging conditions on the tensile shear strength of an epoxy film adhesive.
PROCESSING OF EPOXY ADHESIVES 399
49 7
42.2
Tensile shear strength, MPa 42 39.9 6
34.6
28 26.7 4
21 3
14 2
7 1
0 0
A B C D
FIGURE 17.5 Effect of aging conditions on epoxy film adhesive before
cure: (A) fresh adhesive, (B) after aging 90 days at 24°C, (C) after 90 days at
32°C, (D) after 1-h exposure to condensing humidity.5
ASTM D 1337 describes a standard test method for measuring storage life of adhesives by
consistency and bond strength.
The safe storage of adhesives and sealants that contain flammable, corrosive, or haz-
ardous substances must also be considered. Many organic solvents are flammable, and cer-
tain adhesives will generate copious amounts of heat when reacted. For this reason,
containers used for two- or three-part materials should be stored separately to prevent acci-
dental mixing of the components in the case of spillage.
The more hazardous adhesives and sealants should be stored in a small building or area
separate from the main building or storage area in the plant. An outside storage building is
recommended if several drums of flammable materials are to be stored. The storage facil-
ity should be built of fire-resistant materials and a fire extinguishing system installed. For
storage of just a few drums of flammable materials, insulated metal storage cabinets are
available that will protect the drums by closing their doors automatically in the event of a
fire. An electrical ground should be provided to all drums so that a potentially explosive
spark cannot occur. The most appropriate internal location for a flammable storage area is
a room with brick or masonry walls and fire doors that open outward. Some adhesive prod-
ucts contain corrosive compounds that require special containers and need to be separated
from other materials to avoid interaction.
When removed from refrigerated storage, the adhesive must be brought to its applica-
tion temperature, usually room temperature. The containers should remain closed until they
are at temperature. A material is most vulnerable to moisture collection when it is removed
from cold storage and the package or container is not properly sealed. Moisture condenses
on the adhesive and quickly reduces its capabilities. Containers that are not completely
used when originally opened, should be sealed immediately after use to prevent loss of
volatile ingredients or contamination from moisture in the air. Equipment used to remove
the contents from the containers, such as spatulas, flasks, and cups, should be cleaned
between each use.
400 CHAPTER SEVENTEEN
Bulk adhesives and sealants are normally supplied in containers that range in volume from
1 qt to 55-gal drums. The method used to extract the adhesive from the container should be
carefully thought out.
If a filled resin is used, it will be advantageous to first place the unopened can on a paint
shaker or roller and vibrate or roll the contents for a time. If the resin is in a drum, drum
rollers may be used to ensure that the components within the drum are uniformly dispersed
before the drum is opened. The least aggressive method should always be used so as to
obtain uniform distribution of the components with minimal addition of air into the product.
Once the contents of the container are evenly distributed, the adhesive or sealant
material can be removed. The use of pumps for transferring the adhesive or sealant has
several drawbacks. Filled compounds have an abrasive action on the moving parts. This
is particularly true with fillers such as silica. Compounds that polymerize easily will
tend to begin their polymerization due to friction within the pump. This will gum up the
moving parts and make them inoperable. Therefore, transfer is best done under some
type of pressure.
Although it may require special designs, pressure transfer does eliminate the problems
that occur with gear pumping methods. Many resinous systems may be moved from their
original drums by air pumps. Transfer distances should be short, and lines should be
smooth and noncorrosive. In some instances, barrel warmers and heated lines may be
required to reduce pressures required to transport materials. Under no circumstances should
the same transfer equipment be used on different resins without complete and thorough
cleaning. This will avoid the contamination of one resin system with the other.
Metering and mixing are important processes for the epoxy adhesive end user. Accurate
metering of components and thorough mixing of all components are required for the adhe-
sive to attain its ultimate cured properties. Metering and mixing are often done by hand,
although automatic systems are available and preferred in many circumstances. Automated
equipment is especially useful in continuous or high-volume bonding applications.
If the components require weighing, the weighing equipment should be clean and cali-
brated. Containers used for weighing should have their tare weights plainly marked on the
outside. Containers for weighing adhesive must not be interchanged with those for weigh-
ing water or any other ingredients.
Mixing. The measured components must be mixed thoroughly. Mixing should be con-
tinued until no color streaks or density stratifications are noticeable. Caution should be
taken to prevent air from being mixed into the adhesive through overagitation. This can
cause foaming of the adhesive during cure, resulting in porous bonds. If air does become
mixed into the adhesive, vacuum degassing may be necessary before application.
In mixing, cleanliness is almost as important as accuracy. Scoops, weighing pans, mix-
ing spatulas, and other accessories should be kept clean. Organic solvents are recommended
for cleaning. Adhesives that appear to be similar may be ruined if one is contaminated with
the other. Water will spoil many adhesives and sealants.
Only enough adhesive should be mixed to work with before the adhesive begins to cure.
The working life of an adhesive is defined as the period of time during which a mixed adhe-
sive remains suitable for use. Working life is decreased as the ambient temperature
increases and the batch size becomes larger. One-part and some heat curing, two-part adhe-
sives have very long working lives at room temperature, and application and assembly
speed or batch size is not critical.
The optimum batch size depends on the curing agent used in the epoxy adhesive system
and the temperature at which the system is worked. Epoxy adhesive systems generally main-
tain a workable viscosity for the majority of their pot life. However, once the exotherm
begins, the viscosity of the epoxy system increases rapidly to a point where it gels. For the
more highly exothermic curing agents, the size of the batch will establish working life.
For a given type of mixing equipment, the larger the mass, the longer will be the mixing
time. This reduces the time available for application of the adhesive. Alternatively, a system
of batch replenishment may be employed, but this is sometimes difficult because of the
exotherm of the resident batch causing acceleration of the additive batch.
For mechanical mixing often a variable-speed mixer, preferably air-driven, and a con-
tainer are all that is required. An air-driven mixer is safer, less costly, and more easily reg-
ulated than an electrically driven mixer. For the most reliable mixing operation, the
temperature of the compound in the mixing vessel should be uniform. For adhesives or
sealants that are stored at cold temperature, adequate mixing viscosity can be reached
quickly by the use of a hot liquid jacket around the vessel.
Special mixing blades have been designed to minimize air entrapment and to scour the
sides and bottoms of containers. These blades are made for standard container sizes. In this
case often a fixed blade is used along with a rotating platform. Mixing equipment is
described in Sec. 17.2.2.
For a large-scale bonding operation, hand mixing is costly, messy, and slow; repeatability
is entirely dependent on the operator. Equipment is available that can meter, mix, and dis-
pense multicomponent adhesives on a continuous or shot basis. Figure 17.6 illustrates a stan-
dard adhesive metering mixing and dispensing machine for multicomponent systems. An
important advantage for using such equipment is that it allows the use of short-working-life
materials by mixing only the smallest practical quantity continuously. This offers shorter curing
cycles, resulting in faster assembly times. A disadvantage of this type of equipment is that it
should be used continuously for best results. When the equipment must be shut down, it
requires thorough cleaning. Often the equipment is supplied with an automatic solvent flush
to make sure that all resin is purged from the lines and from internal mixing devices.
For maximum strength glue lines, it is necessary to remove all air from the adhesive
before using it. Entrapped air represents a source of voids in the cured bond line and, hence,
the possibility of a failure. Trapped air is especially noticeable when the adhesive requires
402 CHAPTER SEVENTEEN
Stroke adjustment
Mechanical linkage and Air cylinder
adjustable fulcrum block
Fulcrum adjustment
Inert fluid chamber
Catalyst material Base material
reservoir Bleeder reservoir
Displacement
chamber
Displacement
rod
Solvent Air
Shots of
mixed material
FIGURE 17.6 Operating schematic of an automatic metering, mixing, and dispensing machine for
adhesive systems.6
an elevated-temperature cure; significant amounts of trapped air can cause the adhesive to
foam when exposed to elevated temperatures.
When one is hand-mixing liquid adhesives or sealants, two methods are commonly
employed to degas the mixed material: (1) centrifuging and (2) subjecting the mixed sys-
tem to a vacuum. The latter is the more popular method and requires only a vacuum pump
and vacuum chamber. The vacuum pump should be capable of pulling a vacuum of 29 in
Hg at a high rate of speed.
Vacuum degassing consists of placing the mixed material (in the container in which it
was mixed) into a vacuum chamber. The level of the mixture in the container should be low
enough to allow for a foam head to rise and break during application of a vacuum. Usually
the volume of the chamber is 5 to 10 times the volume of the adhesive to be mixed. If the
adhesive level is too high, the foam head will not break.
It is desirable to know when the foam head has broken, so the vacuum tank should be
provided with two sight windows: one for illumination and one for observation. It is best
to repeatedly break the vacuum during the degassing operation so that multiple foam heads
rise and break. Care should be taken to avoid a “rolling boil” effect. Once all the air is
removed from the adhesives, it will become very difficult to cause a foam head to rise. It is
best to try to minimize the vacuum time, to reduce the loss of any volatile materials from
the mixture. Applied vacuum along with agitation and vibration will effectively remove all
entrapped air.
Air can also be eliminated from an adhesive mixture by using a centrifuge. This process
is generally employed only with low-viscosity, unfilled adhesives. The centrifuging
process consists of rotating the materials at 1000 to 3000 rpm for 1 to 2 min. This is gen-
erally sufficient to remove any entrapped air from liquid adhesive systems.
PROCESSING OF EPOXY ADHESIVES 403
The adhesive or sealant should be mixed and furnished for application in clean containers.
The containers should be small enough that no adhesives or sealant will be left beyond the
time when the resin becomes unusable due to its working life. Disposable waterproof card-
board cups or aluminum containers work very well. Containers may be marked to show the
hour at which the contents were mixed. Any unused adhesive should then be collected at
the end of the working life period recommended by the manufacturer.
Header systems with manual or automatic dispensing valves have been designed to
carry adhesives and sealants over long distances within the plant. Distances of as much as
300 ft are not uncommon with drops at points of application. Figure 17.7 shows a modular
header system. Adhesives and sealants are nearly ideal materials for application by robots.
Robotic application is commonly used in the automobile industry to increase quality and to
reduce labor and material cost.7,8
The epoxy adhesives are usually applied by simple extrusion from dispensing nozzles,
brushing, or troweling. However, spray equipment may be used in certain applications.
Conventional spray equipment is used with solvent-borne epoxy adhesives, but frequent
cleaning is generally required. For fast-reacting systems, dual-nozzle spray equipment is
available.
The type of pumps used to dispense epoxy adhesives depends on whether it is a one-part
or two-part adhesive. One-part epoxy adhesives are dispensed using direct metering extru-
sion pumps. An electric motor pushes a follower plate into a drum of adhesive, which is
then extruded through a hose to the dispensing valve. This technique is used for medium-
and high-volume production rates.
Two-part adhesives are generally dispensed by volumetric pumps for semiautomated
and automated medium- to high-volume assemblies. Metered dispensing systems are eas-
ily capable of processing adhesives with mix ratios of 1 : 1 up to 15 : 1.
The method used to apply the adhesive or sealant can have as much to do with the suc-
cess of the joining operation as the kind of material applied. The selection of an application
or dispensing method depends primarily on the form of the adhesive: liquid, paste, powder,
Header piping
300 ft
Manual
guns
Hose
Controls
Pumps
FIGURE 17.7 Modular header system.9
404 CHAPTER SEVENTEEN
or film. Other factors influencing the application method are the size and shape of parts to
be bonded, the areas where the adhesive is to applied, and the production volume and rate.
One of the prime functions of production equipment is to keep the production employee
removed from contact with the adhesive materials.
Liquids. Liquids are the most common form of adhesive, and they can be applied by a
variety of methods. Liquids have an advantage in that they are relatively easy to transfer,
meter, and mix. They also tend to wet the substrate easily and provide uniform bond line
thickness. However, they have the disadvantages of sometimes being messy, requiring
cleanup, and having a relatively high degree of waste.
Brushes, simple rollers, syringes, squeeze bottles, and pressurized glue guns are man-
ual methods that provide simplicity, low cost, and versatility. These methods are probably
the most widely used because of their simplicity. Manual dispensing methods allow appli-
cation of adhesive to only a small segment of a surface and are particularly effective for
small or irregular parts or low-volume production.
It is possible to mechanize brush methods by using hollow brushes through which the
liquid adhesive is fed. This improves production rates and coating uniformity and cuts
waste. Uniform films are generally difficult to achieve, but operators can become proficient
with time.
Squeeze bottles, oil cans, and pressurized glue guns permit precise and speedy adhesive
application. By adjusting the pressure, the rate of adhesive flow can be matched to the pro-
duction rate. These devices can apply the adhesive or sealant inside a blind hole or limited-
access area. The tip of the applicator can be used to deliver multiple spots of adhesive.
Silk screen application is often used when the adhesive has to be applied to specific con-
trolled areas. The liquid adhesive is forced through pores in a cloth or screen. It is possible
to coat only selected areas by masking parts of the screen so that adhesive does not pass
through in the unwanted areas. Adhesives generally must be specifically formulated for silk
screen processing. Very low-viscosity adhesives, with flow characteristics similar to those
of coatings, are best for silk screening operations.
One of the most attractive precision dispensing methods for low-viscosity adhesives is
the use of microjet printing technology.10 This technology is based on piezoelectric demand
mode ink-jet printing, which can produce droplets of polymeric resins 25 to 125 µm in
diameter, at rates up to 1000 drops per second.
Spraying, dipping, and mechanical-roll coaters are generally used on large production
runs. Dipping is also used to cover large areas quickly. Areas that do not require adhesive
can be masked off. Manual dipping is often effective; and baskets, screens, and perforated
drums are available to facilitate immersion of the part into the adhesive bath.
Spraying usually is evaluated against other methods on the basis of cost. Spraying applies
the liquid resin fast, reduces the required drying time through better evaporation of solvent, and
is capable of reaching areas that are inaccessible to other manual application tools. When auto-
mated, spraying is particularly useful for coating long runs of identical or similar products.
Several spraying methods can be used to apply adhesives, including conventional air
spray, hydraulic cold airless spray, hot spray, and hot airless spray. The main types of spray
equipment are summarized in Table 17.1.
Heating the adhesive before atomization enables heavier adhesive buildup, reduces
overspray losses, and minimizes contamination from atmospheric water vapor. Spray
methods can be used on both small and large production runs. The liquid to be sprayed is
generally in solvent solution. Sizable amounts of product may be lost from overspray. Two-
component adhesives are usually mixed prior to placement in the spray gun reservoir.
Application systems are available, however, that meter and mix the adhesive within the
spray gun barrel. This is ideal for fast-reacting systems, but guns must be thoroughly
cleaned to avoid buildup of polymerized product.
PROCESSING OF EPOXY ADHESIVES 405
Type Characteristics
Air spray Utilizes air pressure to spray a fine mist of atomized adhesive on the substrate.
Generally unsuitable for small parts. Maintenance is high due to nozzle clogging
and overspray. Several passes necessary for thick coating buildup
Hot spray Adhesive is first heated, permitting the use of high-viscosity materials. Heavier
coatings are possible than with air spray. Drying is accelerated due to evaporation
of solvent in the spray
Airless spray Uses hydraulic pressure rather than air pressure. Saves energy and overspray.
Can be applied both hot and cold
When large, flat surfaces and webs of materials are to be coated and when production
reaches rates of 200 to 300 pieces per day, machine methods of application should be con-
sidered over the manual methods described above. Mechanical-roller methods are commonly
used to apply a uniform layer of adhesive via a continuous roll. Such automated systems are
used with adhesives that have a long working life and low viscosity. Various machine meth-
ods of adhesive application are illustrated in Fig. 17.8, and several are described below.
Bench coaters are roughly 20 to 35 percent faster than hand-brushing and cut waste by
up to 20 to 35 percent. They start to be useful at rates of 200 pieces per day and can reach
12,000 pieces per day. Roller sizes range from 4 to 26 in, and they come with various sur-
faces from smooth for thin applications to increasingly coarse for heavier adhesive layers.
Floor-mounted roll coaters are noted for their extreme efficiency with large webs and
flat sheets. Their waste is also very low, less than 2 percent in some cases. Several types
are available because no single model or type of roll coater is best for all the variables of
the material surface, coating formulation, and end use. Table 17.2 summarizes the roll
coaters that are commonly used, and these are illustrated in Fig. 17.8.
Pastes and Mastics. Bulk adhesives such as pastes or mastics are the simplest and most
reproducible adhesive to apply. They produce heavy coatings that fill voids, bridge gaps, or
seal joints. They can be in the form of high-viscosity extrudable liquid or a trowelable mastic.
These systems can be troweled on or extruded through a caulking gun. Little operator skill
is required. Since the thixotropic nature of the paste prevents it from flowing excessively,
application is usually clean, and not much waste is generated. With trowel application, the
depth of notch and the spaces between them help regulate the amount of adhesive applied.
Shallow, rounded, and closely spaced notches are often used with lower-viscosity adhesives
to allow the lines of adhesive to flow together and form a continuous, unbroken film.
High-viscosity liquids, pastes, and mastics are ideal adhesives for application by robots.
Robotic application is commonly used in industries such as automotive and consumer prod-
ucts to increase quality and to reduce labor and material costs.12,13
Robots are fast gaining prominence in the field of material dispensing, especially in the
application of adhesives and sealants. The use of robots provides the end user with the fol-
lowing benefits:
Transfer
roll
Support roll
rubber covered
Support roll
Doctor knife
Engraved roll
Reservoir
Several areas are important when one is considering a robotic system for dispensing
adhesive or sealants. These are summarized in Table 17.3. Monetary justification of robot
installation has been and will continue to remain an issue for the manufacturing industries.
Several installations have seen payback periods of less than 2 years.
Powders. Powdered adhesive systems are not common, although powder resin coatings
are ordinarily used for protective and decorative coating. B-staged powder adhesives may
be applied dry to a natural or primed surface. They can be applied electrostatically as pow-
der coatings are, or they can be applied gravimetrically by being dispersed onto a hot sub-
strate surface. Heat and pressure must be applied for full curing, and considerable care
taken to obtain uniform thickness over large areas. If coating uniformity is poor, large vari-
ations of joint strength may result with powder adhesives. The main advantages of using
powder adhesives are their cleanliness and minimal waste. Spills and overspray may be
PROCESSING OF EPOXY ADHESIVES 407
Doctor knife
Coating material
Web
(c) Knife-over-roll coater
Reservoir
Transfer roll
Web
Support roll
Transfer roll
Reservoir
Type Characteristics
Kiss No backup roll; coater is in light contact with substrate; contact is controlled
by web tension. Best with adhesives having good flow characteristics and
relatively slow drying rates
Pressure or squeeze Also called nip coaters; can provide greater penetration into open-web
structures. Often used to apply coatings to both sides simultaneously
Reverse roll Popular for web precision coating because of better control over thickness
Two-roll reverse For additional control over adhesive coating; can handle wet film
thicknesses from 1 to 20 mils
Dip roll Simplest of the basic types of coaters; used mostly with low-viscosity
adhesives; works well where impregnation or saturation is required
Engraved roll Provides accurate control over coating uniformity; best for extremely low
adhesive weights
Air blade Similar to the kiss coater in that the adhesive is applied where the web
touches it; possible to apply coating uniformly at high speeds; limited use
with solvent-based adhesives
Bead Two rollers set at specific clearance; a rod between the bottom roll and the
web makes a bead of adhesive that coats the web passing above it
Knife spreader Usually used to apply high-viscosity adhesives to flexible webs; web is
drawn between a precision knife edge and a rubber roller or flat-bed plate
Flow or curtain Use a head with a slot in the bottom through which the adhesive is poured
or lowered onto the parts or surfaces as they pass beneath the head by a
conveyor; provides materials savings and ability to coat irregular
surfaces uniformly
collected and reused, if not contaminated. The main disadvantage is the cost of the appli-
cation equipment and the lack of manual methods of application.
Powder adhesives are generally one-part, epoxy-based systems that do not require
metering and mixing but often must be refrigerated for an extended shelf life. Elastomeric
types are generally not available in powder
TABLE 17.3 Considerations regarding form. Systems that need a primer require an
Robotics in Adhesives Application additional coating step prior to applying the
powder. Equipment cost for powder applica-
Type of robot tion tends to be high.
Pumping equipment Powder adhesives are generally applied in
Interface controls three ways: sifting or spraying the powder on the
Type of adhesive or sealant substrate, dipping the substrate into the pow-
Gun and nozzle holders der, or melting the powder to liquid form for
Safety fencing more conventional application. When sifted or
Bead detection device sprayed onto a substrate, the powder falls onto
Arrangement and flexibility of arrangement
the preheated substrate, melts, and adheres.
Regulating equipment
Electrostatic spraying of adhesive powder has
Area of dispensing
Size and shape of bead
also been used to apply a B-staged powder to
Weight of robot arm tooling various substrates. A preheated substrate could
Flow control also be dipped into the powder and then
extracted with an attached coating of adhesive.
PROCESSING OF EPOXY ADHESIVES 409
This method helps to ensure even powder distribution. Lastly, the powder can be melted
into a paste or liquid and applied by conventional means. Once the powder is adhered to the
substrate, the assembly is then mated and cured under heat and pressure according to rec-
ommended processes.
Films. Both structural and nonstructural adhesives are commonly available in film form.
Adhesives applied in the form of dry films offer a clean, hazard-free operation with mini-
mum waste and excellent control of film thickness. However, the method is generally lim-
ited to parts with flat surfaces or simple curves. Optimum bond strength requires curing
under heat and pressure, which may involve considerable equipment and floor space, par-
ticularly for large parts. Film material cost is high in comparison to liquids, but waste or
material loss is the lowest of any application method.
Practically any solid or B-stageable epoxy adhesive that has adequate stability at room
temperature can be supplied in the form of a film. This film may be pressure-sensitive,
thermoplastic, or, in applications where the greatest structural strength is required, a ther-
mosetting type. Epoxy, epoxy hybrid, phenolic, and other thermosetting film adhesives
are commonly used in many industrial applications. Elastomeric adhesives are available
in dry films in both solvent and heat reacting types. Both elastomer- and resin-based adhesives
are also supplied as pressure-sensitive adhesives, but they have lower strength than the
solvent or heat reacting types. Epoxy adhesives generally do not have pressure-sensitive
characteristics.
The use of dry adhesive films is expanding more rapidly than other forms because of
their following advantages:
After the adhesive is applied, the assembly must be mated as quickly as possible to prevent
contamination of the adhesive surface. The substrates are held together under pressure and
heated if necessary until cure is achieved. The equipment required to perform these func-
tions must provide adequate heat and pressure, maintain constant pressure during the entire
cure cycle, and distribute pressure uniformly over the bond area. Of course, many adhe-
sives cure with simple contact pressure at room temperature, and extensive bonding equip-
ment is not necessary.
It is important when one is selecting equipment to give consideration to life-cycle costs
which include original investment cost and operation and maintenance cost, such as energy
410 CHAPTER SEVENTEEN
• Simplicity of operation
• Compatibility with adhesive or sealant
• Ease of disassembly and cleanup
• Trouble detection devices and self-diagnostics
• Service backup relative to spare parts, local resources, etc.
• Knowledge of life expectancy of system and its components
• Clearly defined warranty coverage
• Knowledge of the equipment production rates
• Ease of pressure regulation
• A means of checking the proportions of different adhesive components
• Temperature controls and indicators.15
Pressure devices should be designed to maintain constant pressure on the bond during the
entire cure cycle. They must compensate for thickness reduction from adhesive flow-out or
thermal expansion of assembly parts. Thus, screw-actuated devices such as C-clamps and
bolted fixtures are not acceptable when constant pressure is important. However, spring
clamp fixtures perform well on assemblies machined to distribute mechanical loads. Spring
pressure can supplement the clamps during heat curing to compensate for variation in pres-
sure when the adhesive begins to flow and adherends expand. Deadweight loading may
also be applied in many instances. However, this method is sometimes impractical, espe-
cially when a heat cure within an oven is necessary.
Pneumatic and hydraulic presses are excellent tools for applying constant pressure.
Steam or electrically heated platen presses with hydraulic rams are often used for adhesive
bonding. Some units have multiple platens, thereby permitting the bonding of several
assemblies at one time.
Large bonded areas, such as on aircraft parts, are usually cured in an autoclave. The
parts are mated first and covered with a rubber or film blanket to provide uniform pressure
distribution. The assembly is then placed in an autoclave, which can be pressurized and
heated. This method requires heavy capital equipment investment.
Vacuum bagging techniques can be an inexpensive method of applying pressure to large
parts. A film or plastic bag is used to enclose the assembly, and the edges of the film are
sealed airtight. A vacuum is drawn on the bag, enabling atmospheric pressure to force the
adherends together. Vacuum bags are especially effective on large areas because the size is
not limited by pressure equipment. Pressures, of course, are limited to atmospheric pressure.
Many structural adhesives require heat as well as pressure to cure. Even with conventional
room temperature curing systems, most often the strongest bonds are achieved by an
elevated-temperature cure. With many adhesives, tradeoffs between cure times and tem-
perature are permissible. Generally, the manufacturer will recommend a certain curing
schedule for optimum properties.
PROCESSING OF EPOXY ADHESIVES 411
If a cure of 60 min at 150°C is recommended, this does not mean that the assembly
should be simply placed in a 150°C over for 60 min. The temperature is to be measured at
the adhesive bond line. A large part will act as a heat sink and may require substantial time
for the adhesive in the bond line to reach the necessary temperature. In this example, total
oven time would be 60 min in addition to whatever time is required to bring the adhesive
up to 150°C. Bond line temperatures are best measured by thermocouples placed very close
to the adhesive. In some cases, it may be desirable to place the thermocouple in the adhe-
sive joint for the first few assemblies being cured.
Oven heating is the most common source of heat for bonded parts, even though it involves
long curing cycles because of the heat sink action of large assemblies. Ovens may be heated
with gas, oil, electricity, or infrared units. Good air circulation within the oven is mandatory
for uniform heating. Temperature distribution within an oven should always be checked
before items are placed in the oven. Many ovens will have significant temperature distribu-
tions and dead spaces in corners where air circulation is not uniform. The number of items
placed in the oven will also affect the time for the bond to get to temperature. The geometry
and size of the part may affect air circulation and cause variations in temperature distribution.
Heated-platen presses are good for bonding flat or moderately contoured panels when
faster cure cycles are desired. Platens are heated with steam, hot oil, or electricity and are
easily adapted with cooling-water connections to further speed the processing cycle. Strip
or cartridge heaters are useful for localized heating in small areas. They adapt easily to flat
dies, pressure bars, specially shaped dies, and other pressure fixtures. They are often used
for tack-bonding small areas to facilitate handling prior to complete cure.
Induction and dielectric heating (Chap. 14) are the fastest heating methods because they
focus heat at or near the adhesive bond line. Workpiece heating rates greater than 55°C/s
are possible with induction heating. For induction heating to work, the adhesive must be
filled with metal particles, or the adherend must be capable of conducting electricity or
being magnetized, as shown in Fig. 14.5.
Dielectric heating (radio frequency and microwave) is also an effective way of curing adhe-
sives if at least one substrate is a nonconductor. Metal-to-metal joints tend to break down the
microwave field necessary for dielectric heating. This heating method makes use of the polar
characteristics of the adhesive materials. Dielectric heating is used in the furniture industry to
drive off the water and harden water-based adhesives. Dielectric heating is also being explored
as a rapid method of curing adhesives on substrates such as glass, plastics, and composites.15
Both induction and dielectric heating involve relatively expensive capital equipment out-
lays, and the bond area is limited. Their most important advantages are assembly speed and the
fact that an entire assembly does not have to be heated to cure only a few grams of adhesive.
An interesting development that eliminates the need for fixturing equipment and heating
equipment for specifically curing the adhesive joint is weldbonding. Weldbonding is adhe-
sive bonding combined with spot welding to make fast, strong, reliable joints. The process
produces joints that are stronger, more durable, and more fatigue-resistant than when either
method is used by itself. This process has found its way into the automotive industry for use
on aluminum and steel substrates. A variety of adhesives and substrates are compatible with
this method. Epoxy adhesives capable of weldbonding are described in Chap. 14.
REFERENCES
3. Devlin, W., “Metering and Mixing Equipment,” in Adhesives and Sealants, vol. 3, Engineered
Materials Handbook, ASM International, Materials Park, OH, 1990.
4. Van Twisk, J., and Aker, S. C., “Storing Adhesive and Sealant Materials,” in Adhesives and Sealants.
5. Van Twisk, and Aker, “Storing Adhesive and Sealant Materials.”
6. Graco Pyles Corporation, Minneapolis, MN.
7. “Robotic Dispensing of Adhesives and Sealants,” Handbook of Adhesives, 3d ed., I. Skeist, ed.,
Van Nostrand Reinhold, New York, 1990.
8. Dueweke, N., “Robotics and Adhesives—An Overview,” Adhesives Age, April 1983, pp. 11–16.
9. Killick, B. R., “Dispensing High Viscosity Adhesives and Sealants,” in Adhesives in
Manufacturing, G. L. Schneberger, ed., Marcel Dekker, New York, 1983.
10. Hayes, D. J., et al., “Development and Application by Ink Jet Printing of Advanced Packaging
Materials,” in Proceedings of the International Symposium on Advanced Materials Process,
Properties and Interfaces, 1999, pp. 88–92.
11. Petrie, E. M., Handbook of Adhesives and Sealants, McGraw-Hill, New York, 2000, pp. 744–745.
12. “Robotic Dispensing of Adhesives and Sealants,” in Handbook of Adhesives.
13. Trees, B., “Auto Adhesive Extrusion Systems,” Adhesives and Sealants Industry, May 2002.
14. Dueweke, N., “Robotics and Adhesives—An Overview,” Adhesives Age, April 1983, pp. 11–16.
15. Raulauskas, F. L., “Adhesive Bonding/Joining via Exposure to Microwave Radiation,” 27th
International SAMPE Technical Conference, October 9–12, 1995.
CHAPTER 18
HEALTH AND SAFETY ISSUES
18.1 INTRODUCTION
Completely cured epoxy adhesives are almost nontoxic, although caution must be exer-
cised when machining, grinding, etc., joints bonded with epoxies due to the proliferation of
airborne particles. Also, decomposition products may exhibit substantial vapor pressures
and pose health and safety issues. However, it is the exposure to the uncured components
that can be the most harmful and is the subject of this chapter.
Four primary factors must be considered in all adhesive bonding or sealing operations:
toxicity, flammability, hazardous incompatibility, and equipment. The adhesive or sealant
must be carefully considered because not only can it create health and safety issues within
the factory but also it can provide issues within the community relative to release of
volatiles and waste disposal.
The epoxy raw materials can be harmful and hazardous. These materials could be toxic,
flammable, and irritating to skin. It is generally the duty of the formulator to make sure that
these materials are worked within a safe environment. Once they are formulated, the end
user also has responsibilities with regard to safe handling of epoxy adhesive products.
Some materials involved in the epoxy adhesive formulations are more dangerous than
others, and some affect certain parts of the body whereas others do not. Amine curing
agents, for example, are generally the most irritating to the skin and eyes. They may also
be considered to be strong caustics, and they produce serious local injury on short expo-
sure. If they are worked in a hot condition, their fumes pose additional hazards.
In a two-component epoxy product, the epoxy resin and the curing agent are packaged
separately and must be mixed together just before use. Each component can be hazardous.
In a single-component product, the resin and curing agent are supplied in a premixed form.
Single-component systems are generally safer because the hazardous chemicals are already
combined and because their time of exposure is less.
Solid epoxy adhesives, such as powder, preforms, or film, are the least toxic because
they have very low vapor pressures. However, they still can present a hazard when they are
heated to a liquid and more volatile state to cure.
Accessory materials and equipment used to process and apply epoxy adhesives can also
be hazardous. Solvents that are used either to dilute the epoxy adhesive or to clean equip-
ment after use can be both dangerous and toxic. Mixing equipment can run at very fast
speeds, and caution needs to be exercised when one is operating such equipment. Curing
equipment often runs at high temperatures and can create a fire or explosion hazard.
Electrical hazards are also associated with most equipment.
Because of the wide variety of materials, equipment, and processes used with epoxy
adhesive systems, the following discussion provides only a general guide. For more detailed
413
Copyright © 2006 by The McGraw-Hill Companies, Inc. Click here for terms of use.
414 CHAPTER EIGHTEEN
information, the reader should contact the manufacturer of each material or equipment item
used. The discussion that follows focuses primarily on
All adhesives, solvents, chemical treatments, and like materials must be handled in a man-
ner preventing toxic exposure to the workforce. Methods and facilities must be provided to
ensure that the maximum acceptable concentrations of hazardous materials are never
exceeded. These values are prominently displayed on the material’s Material Safety Data
Sheet (MSDS), which must be maintained and available for the workforce.
There is no medical or laboratory test that can accurately measure the amount of epoxy
components in the human body, and most of the chemicals found in epoxy systems are not
stored in the body. Regional environmental regulators set and enforce workplace exposure
limits. These exposure limits are generally set for some of but not all the epoxy components
that one may contact.
Table 18.1 shows legally permissible exposure limits (PELs) for some chemicals com-
monly found in epoxy adhesive systems. These specific levels have been set by the
California Division of Occupational Safety and Health (Cal/OSHA). The maximum per-
missible exposure limits may change from region to region. Legally, an individual’s expo-
sure may be above the PEL at times, but only if it is below the PEL value at other times so
that the average exposure for any 8-h work shift is not greater than the PEL for the specific
chemical. Methods of reducing exposure are described in Sec. 18.5.
TABLE 18.1 Permissible Exposure Limits (PELs) for Some Common Epoxy
Resin System Chemicals (California/OSHA)1
Older epoxy resins were noticed to cause skin cancer in laboratory animals. This was
most likely due to the epichlorohydrin. Most newer epoxy resins, which contain less
epichlorohydrin, do not seem to cause cancer in animals. Certain curing agents, such as
metaphenylene diamine (MPDA) and diaminodiphenyl sulfone (DADPS), and certain gly-
cidyl ethers are carcinogenic in laboratory animals. It is not known if these materials cause
cancer in humans.
The chemicals in epoxy adhesive systems can affect human health when they come into
contact with the skin or if they evaporate and form a mist or vapor in the air. The main
effects of overexposure are irritation of the eyes, nose, throat, and skin; skin allergies; and
asthma. The solvent additives and other high-vapor-pressure materials can cause other
effects such has headaches, dizziness, and confusion. The effects of these chemicals on the
various parts of the human body are summarized below.2
Lungs. Vapor inhalation with most liquid resins is not considered a problem unless the
resins are heated or the epoxy component has a very high vapor pressure, such as with sol-
vents or certain diluents. Vapor and spray mists of most epoxy chemicals can irritate the
lungs. Some people develop asthma from the curing agents. Symptoms of asthma include
chest tightness, shortness of breath, wheezing, and coughing. These symptoms may occur
at times other than when in contact with the epoxy component. Once a person becomes
allergic to curing agents, even the dusts from sanding or grinding, future contact will gen-
erally always cause a reaction.
Skin. Most epoxy resins are not acutely irritating to the skin. However, they are capable
of causing skin sensitization. Susceptibility to skin irritation and sensitization varies from
person to person. The epoxy resins are considered to be milder skin sensitizers than amine-
type curing agents or epoxy functional reactive diluents.
Skin irritation symptoms can include redness, swelling, flaking, and itching on the
hands, face, or other areas of contact. Some people develop a skin allergy. Skin allergies
may develop after just a few days of contact or after many years of exposure, depending on
the individual. In the event of skin contact, thoroughly wash the affected area with copious
amounts of soap and water. Remove contaminated clothing and launder before reuse.
Eyes, Nose, and Throat. All the epoxy resins are low in acute oral toxicity. However,
most epoxy chemicals and their vapors can irritate the eyes, nose, and throat. Some people
develop headaches as a result of this irritation. Eye contact with epoxy resins could result
in slight, transient irritation. However, curing agents, solvents, and diluents can cause
severe irritation. In event of eye contact, the eyes should immediately be flushed with plenty
of water and then medical attention sought.
Nervous System. Solvents inhaled or absorbed through the skin can affect the central ner-
vous system in the same way as drinking alcohol does. Symptoms of solvent overexposure
include headache, nausea, dizziness, slurred speech, confusion, and loss of consciousness.
Reproductive System. Epoxy resins and curing agents themselves do not seem to affect
pregnancy and reproduction in humans. However, some of the diluents and solvents that
are used may affect reproduction. Two solvents sometimes found in epoxy resin systems
(ethoxyethanol and methoxyethanol) have been determined to cause birth defects in labo-
ratory animals and reduced sperm counts in men. Some diglycidyl ethers also damage the
testes and cause birth defects in test animals. It is not known whether these materials cause
the same effect in humans.
Most other solvent additives have not been adequately tested to determine if they affect
reproduction. However, it is known that solvents inhaled by a woman can reach a developing
416 CHAPTER EIGHTEEN
fetus and may contaminate the woman’s breast milk. It is generally recommended that
pregnant and nursing women minimize their exposure to solvents.
The main components of epoxy adhesive systems may contain materials that are hazardous
and/or affect the health of those who come into contact with them. In general, epoxy adhe-
sives have hazardous properties, but can be handled safely. The hazards associated with the
specific adhesive being handled depend on the hazardous nature of the components.
Because different material components in epoxy adhesives have different properties and
characteristics when it comes to health and safety, one should always try to find out what
chemicals are in the products being used. This can be done by consulting the specific man-
ufacturer’s Material Safety Data Sheet (MSDS).
An MSDS lists the hazardous chemical contents of a product, describes its health and
safety hazards, and gives methods for its safe use, storage, and disposal. The MSDS also
includes information on fire and explosion hazards, reactivity, first aid, and procedures for
handling leaks and spills.
Generally, environmental regulations require that an employer have all MSDSs for any
workplace product that contains hazardous substances, and these must be made available
to employees on request. MSDSs can also be obtained from the manufacturer of the mate-
rial, and many are available over the Internet.
Potential hazards associated with the various components of epoxy resins are reviewed
below and in Table 18.2. This information was summarized from professional societies
such as the Society of Plastics Industry (SPI) and manufacturers’ brochures.3,4 The reader
is directed to similar sources or to the manufacturers themselves for additional information.
Epoxy Resins. Liquid epoxy resins are mild to moderate irritants to the skin, eyes, and
mucous membranes. The irritant potential is increased by their sticky nature, which tends
to lead to prolonged skin contact. These resins are generally mild to moderate dermal (skin)
sensitizers in susceptible individuals.
Solid epoxy resins are not readily absorbed through the skin and present a low risk of
skin irritation. Direct contact with solutions of these resins can cause mild to moderate irri-
tation of the skin and the eyes, principally because the solvents remove the protective layer
of fat that is on the surface of the skin. When crushed into a fine powder, epoxies should be
considered an irritant dust, and inhalation and skin contact should be avoided. Solid resins
are generally considered to be low to mild sensitizers.
Resins that are modified by the addition of reactive diluents or solvents can be more
serious irritants. These resins should be handled with the same precautions as taken when
one is using chemical solvents. Their sensitizing potential tends to increase with decreas-
ing molecular weight. Epoxy components with significant volatility could cause irritation
to skin, eyes, and respiratory tract, but inhalation is normally not a hazard except under cer-
tain conditions of use, such as heating, spraying, or applications with large surface areas.
Curing Agents and Catalysts. Curing agents are generally more hazardous to health than
the unmodified epoxy resins. Polyamide-type curing agents are considered to present a low
degree of health hazard compared to other curing agents. Catalytic-type curing agents are
difficult to generalize because of the differences in chemical makeup. Specific information
on these, as well as all materials used in epoxy adhesives, should be requested from the
manufacturer prior to use.
TABLE 18.2 Typical Exposure Effects Associated with Epoxy Resin Systems5
Epoxy resin
system components Examples and types Dermal exposure Inhalation exposure Ingestion exposure
Liquid epoxy resins Based on the reaction product Mild to moderate irritants Low volatility Low toxicity
of epichlorohydrin and Mild to moderate sensitizers Exposure unlikely unless heated,
bisphenol A or bisphenol F sprayed, or spread over large
unventilated areas
Solid epoxy resins Based on the reaction product Mild to moderate irritants and mild Low volatility Low toxicity
of epichlorohydrin and sensitizers Exposure unlikely unless crushed
bisphenol A or bisphenol F Not readily absorbed through skin or ground
Modified liquid Liquid epoxy resins with Mild to moderate irritants Low volatility Low toxicity
epoxy resins added reactive diluents or Moderate to strong sensitizers Exposure unlikely unless heated,
solvents sprayed, or spread over large
unventilated areas
417
The aliphatic amines, cycloaliphatic amines, and anhydride curing agents may cause
irritation or damage to the skin, eyes, and lungs. Certain aliphatic and cycloaliphatic amines
are skin sensitizers. Solid anhydride curing agents may cause sensitization in workers
exposed to the curing agent dust. The aromatic amines are not strong irritants, but several
are skin sensitizers. Certain aromatic amines may absorb through the skin and cause dam-
age to organs such as the liver and interfere with the blood’s ability to carry oxygen.
Aromatic amines seriously stain skin and surrounding materials.
Reactive Diluents. Reactive diluents are generally considered to present a high degree of
hazard. They are capable of causing skin and eye irritation and present a significant hazard
from inhalation due to their high vapor pressure. Most reactive diluents fall into a chemi-
cal family known as the glycidyl ethers. These materials vaporize more readily than the
epoxy resins and, therefore, have an increased potential for inhalation exposure. These dilu-
ents are also likely to be much stronger sensitizers than the epoxy resins.
Fillers. Fillers vary in their degree of hazard from handling. Some are considered to be
essentially nonhazardous. However, dusts of glass, silica-bearing powders, and powdered
metals may present a serious hazard from inhalation and/or explosion. The explosion haz-
ard is present due to the high surface areas of finely divided fillers.
Fillers present a potential inhalation and dermal contact hazard. They can cause
mechanical damage to the skin, which may aggravate the irritant effect of other chemicals
and additives. When fillers are handled in a liquid epoxy matrix or in a cured epoxy, their
inhalation hazard is low. However, inhalation exposure to fillers can occur when they are
handled in the dry state or when one is machining or grinding cured epoxy products.
Inhalation exposure to fillers such as crystalline silica or fiberglass may result in delayed
lung injury. Asbestos fillers have long been abandoned from use for these reasons.
18.4 PROCESSES
Potential exposure to the chemicals used in epoxy adhesive formulations varies with the
type of process or task. Closed systems with engineering controls are used to prevent work-
ers from overexposure. However, occasionally open areas with limited controls are encoun-
tered, and the potential for exposure increases.
Dermal exposure is the most likely route of exposure, and this generally occurs from
tasks involving hand contact with the adhesive. However, if certain curing agents or sol-
vents are being used, inhalation exposure may also be a problem. Inhalation and dermal
risks are most significant with any task involving the use of solvents or curing agents. The
risk of exposure from the ingestion of epoxy materials is generally minimal.
Several of the more common processes that occur with the formulation and end use of
epoxy adhesives are described in Table 18.3. Potential exposure risks for these processes
are characterized as dermal, inhalation, or ingestion. Comments regarding the effect of the
process on potential exposure are also included.
HEALTH AND SAFETY ISSUES 419
TABLE 18.3 Exposure Potential of Epoxy Adhesive Production and Application Processes
Exposure potential
There are a number of processes that can be used to limit the exposure to chemicals that are
commonly used in formulating epoxy adhesives. In general, the employer is required to
protect his or her workers from being exposed to any hazardous chemical over the permis-
sible exposure level.
Chemical manufacturers, formulators, and distributors must make health, safety, and envi-
ronmental protection an integral part of the product. Several guides have been created to help
develop and implement policies and practices that ensure protection through the product life
cycle. One of these, developed by the Epoxy Resin Formulators Group of the Society of
Plastics Industry, has been found to be most useful for the epoxy formulator.6
In general the workplace processes that can be used to limit or prevent potential expo-
sure to hazardous chemicals are:
• Training
• Substitution
• Engineering controls
• Protective equipment and clothing
• Good housekeeping
420 CHAPTER EIGHTEEN
18.5.1 Training
Those engaged in formulating or using epoxy adhesive should conduct continuing training
programs for all personnel involved in handling of epoxy materials or those who can pos-
sibly come into contact with epoxy materials. Planning for employee and plant safety has
value only when it is interpreted and practiced by the people involved.
Continued instruction of all employees must be given concerning the consequences of
contact, as well as the precautions necessary for some operations (storage, handling, com-
pounding, packaging, and disposal). The training program should address at a minimum the
following items.7
The training of individuals should emphasize several areas that are unique to epoxy resins.
Some of these are described below.
Where flammable solvents and adhesives are used, they must be stored, handled, and
used in a manner preventing any possibility of ignition. Proper safety containers, storage
areas, and well-ventilated workplaces are required.
Certain adhesive materials are hazardous when mixed together. Epoxy and polyester
catalysts, especially, must be well understood, and the user should not depart from the
manufacturers’ recommended procedure for handling and mixing. Certain unstabilized sol-
vents, such as trichloroethylene and perchloroethylene, are subject to chemical reaction on
contact with oxygen or moisture. Only stabilized grades of solvents should be used.
Certain adhesive systems, such as heat curing epoxy and room temperature curing poly-
ester, can develop very large exothermic reactions on mixing. The temperature generated
during this exotherm is dependent on the mass of the materials being mixed. Exotherm tem-
peratures can get so high that the adhesive will catch fire and burn. Adhesive products
should always be applied in thin bond lines to minimize the exotherm until the chemistry
of the product is well understood. Never use elevated temperature curing sealants or adhe-
sives for casting or for application in excessively thick cross-sections without first consult-
ing the manufacturer. These materials are formulated to be applied and cured in thin
cross-sections so that the heat generated by exotherm can easily be dissipated.
Safe equipment and proper operation are, of course, crucial to a workplace. Sufficient
training and safety precautions must be installed in the factory before any bonding process
is established.
18.5.2 Substitution
The most effective way to reduce hazardous chemical exposure is to use a safer chemical
if one is available. Unfortunately this is also one of the most difficult methods of control.
Substitution is a difficult task because the specific ingredients will control the applica-
tion and end-use properties of the epoxy adhesive. Any substitution of material generally
requires that the complete product be reverified with production and prototype testing. Of
course, the health and safety issues of any alternative material must also be carefully con-
sidered to ensure that it is actually safer.
HEALTH AND SAFETY ISSUES 421
There are several elements to consider when looking for a substitute material or when
selecting materials for original formulation. For example, one may be able to choose an
epoxy resin system that:
In addition to training, engineering controls are the most effective way of minimizing
exposure to harmful chemicals. Engineering controls include process or equipment mod-
ifications that reduce the amount of potentially hazardous materials to which an employee
may be exposed. Isolation and ventilation are the primary methods of control. Engineering
controls also include the maintenance, policing, and changing of work practices when
necessary.
Isolation, or enclosure of a process or work operation to reduce exposure is a standard
industrial hygiene method. The use of epoxy resins, for example, could be isolated to des-
ignated areas that are separate from the remainder of the plant. Examples of isolation are
spray booths, enclosed curing and mixing rooms, and glove bag systems. Isolation is an
ideal method for use with adhesives systems since the isolation also can prevent unwanted
materials, generally contamination (e.g., moisture, mold release, dust) from getting to the
work piece and degrading the strength of the finished joint.
Ventilation is the standard method of controlling exposure to airborne vapors of epoxy
resins and solvents. Ventilation involves controlling air flow to reduce exposure. Local
exhaust ventilation systems capture the vapor at the source and either filter or remove it
from the work area. The ventilation system needs to be designed so that vapors, aerosol,
and dusts are pulled away from, and not into, the breathing zone of the workers. A constant
supply of fresh, noncontaminated air should be available at all times.
Examples of local exhaust ventilation system are drawdown exhaust tables, slot hoods,
dust extraction systems, and portable vapor and dust collectors. Care must be taken that the
ventilation systems are utilized and maintained as designed. The condition of the filters and
air flow rate should be checked periodically as well as the condition of the duct work,
motors, belts, etc.
Certain work processes, such as the heating or curing of epoxy resin systems, can be iso-
lated, enclosed, or automated to reduce exposure. Heating of epoxies during compounding
or cure can cause components to evaporate more quickly. The higher the temperature, the
greater the amount of contaminant released into the air. Therefore, the lowest possible tem-
peratures to perform the functions should be used, and adequate ventilation must be main-
tained around these areas. Electrostatic spray systems can reduce the amount of work place
contamination and waste resulting from sprayed epoxy resin systems.
422 CHAPTER EIGHTEEN
Good housekeeping also maintains that separate eating, drinking, and smoking facilities
should be maintained that are isolated from the workplace. Refrigerators and freezers used
for the storage of epoxy adhesive components must not be used to store edible items or
drinks.
First aid should be an important component of any training program. First aid materials and
facilities should be maintained in a near-by but isolated area, away from hazardous opera-
tions and materials. The specific type of first aid for various problems is generally indicated
on the MSDS.
If the skin or clothing becomes contaminated with epoxy components, remove the con-
taminated clothing and wash the exposed area with soap and water for at least 15 mins.
Seek medical attention immediately if irritation or other complications develop. If the eyes
become contaminated, they should be washed out with copious quantities of clean water
and medical advice should be sought. Eye wash facilities should be provided in areas of the
work place where such accidents have a potential to happen.
If respiratory distress is experienced, immediately remove the individual from the con-
taminated area to fresh air. If the person is not breathing, artificial respiration should be pro-
vided. Seek immediate medical attention. If breathing is difficult, transport the individual
to a medical care facility for treatment, and if available, give the individual supplemental
oxygen.
If a chemical is accidentally ingested, seek immediate medical attention. If the victim is
conscious, give water. Do not induce vomiting unless directed to do so by a physician or as
directed by the MSDS.
Emergency procedures are most generally required with epoxy adhesives with respect to
flammability and spill contamination and clean-up.
Most resinous materials used in epoxy adhesive are organic and will burn when suffi-
cient heat and oxygen are supplied. A common measure of the flammability is the flash
point temperature. This value indicates the minimum temperature at which flammable con-
ditions are produced in controlled laboratory experiment at atmospheric pressure. Solvents,
diluents, and other materials used with epoxy resins commonly increase the hazard of flam-
mability and/or explosion.
Fires involving epoxy resins can be extinguished with foam, dry powder, or carbon
dioxide. Water is not normally an effective extinguishing agent for these resins. When
burning, these resins give off toxic by-products, such as carbon monoxide gas. Therefore,
avoid breathing fumes or gases resulting from a fire. Fire fighters should use an organic
vapor respirator or self-contained breathing apparatus.
Any spills of epoxy resin or epoxy components should be cleaned up immediately. The
immediate concern is to protect personnel, prevent a possible fire hazard, and contain the
spill until it is cleaned up. Persons engaged in spill clean-up should be protected from
vapors and from skin contact by wearing appropriate protective clothing and equipment.
Persons engaged in spill clean-up should also be aware of proper disposal techniques for
the materials used in the process.
424 CHAPTER EIGHTEEN
For small spills of epoxy resin liquid or solution, apply an absorbent material or a high
surface area material such as sand to the spill. Shovel the mass with absorbed epoxy into a
suitable container. The residue that is left should then be cleaned with hot soap water or
steam. For spills of material containing solvent, keep spark producing equipment away
form the spill site and shut off or remove all potential sources of ignition.
For larger spills, employees should stay upwind of the spill to avoid inhalation of epoxy
components. Evacuate and rope off the spill area, and shut off all potential sources of igni-
tion. The spill should be contained with a dike, and excess resin should be collected in suit-
able containers for final disposal. Hot soapy water or steam may be used for cleaning up
the residue from floors or equipment. The use of solvents during clean-up should be
avoided because of the hazardous nature of the solvent.
Liquid and solutions of epoxy resins must be prevented from entering sewers or drains,
or any body of water, including rivers, streams, or lakes. Flexible emergency dikes, some-
times known as “pigs,” can be used to block entry of the spill into sewers and drains. If
spilled materials does enter drains or water ways, local authorities should be notified at
once.
REFERENCES
1. Epoxy Resin Systems, California Department of Health Services, Oakland, CA, June 1989.
2. Epoxy Resin Systems.
3. Epoxy Resin Systems Safe Handling Guide, The Society of the Plastics Industry, New York,
September 1997.
4. Hazards and Handling Precautions, Dow Plastics, Dow Chemical Company, Midland, MI.
5. Epoxy Resin Systems Safe Handling Guide.
6. Product Stewardship Guide, Epoxy Resin Formulators Group, The Society of the Plastics Industry,
New York, 2001.
7. Epoxy Resin Systems Safe Handling Guide.
CHAPTER 19
QUALITY CONTROL AND
SPECIFICATIONS
19.1 INTRODUCTION
This chapter discusses quality control and the writing and use of specifications that apply
to epoxy adhesive systems. These subjects are of interest to both the formulator and the
end user.
They are of interest to the formulator as they apply to raw materials and to the final
adhesive product. The formulator must understand not only how to control the quality of
the adhesive product but also how to use specifications as they relate to raw materials or to
the end user or customer.
Similarly, the end user must develop a plan to ensure the consistency and quality of the
entire bonding processes. Specifications that are employed by the end user will apply both
to the formulated adhesive and to the final bonded assembly that is produced.
Quality control testing processes that are commonly used with epoxy adhesives are dis-
cussed in Chap. 20. Tests included are those used by the formulator as well as those employed
by the end user of the adhesive system.
Both the formulating of adhesives and bonding with adhesives are complex, multiple-part
processes, complete with interacting and sometimes unexpected parameters that may con-
tribute to success or failure of the final product. Thus, it is important that the quality control
process consider the entire operation from receipt of materials to final product testing.
A generalized flowchart for the quality control process in formulating epoxy adhesives
is shown in Fig. 19.1. A flowchart for controlling the quality of the adhesive bonding
process is shown in Fig. 19.2. It must be realized that in both cases, the decisions made in
one phase of the process may affect the subsequent phases. Therefore, all the individual
phases must be carefully coordinated and controlled.
Although sophisticated testing and analysis equipment is available, many of these
processes do not require advanced equipment. Simple equipment, visual examination, and
common sense are the main tools in most quality control departments. They are supported
by strong specifications and proper training.
Quality control encompasses all the processes and activities that ensure adequate quality
in the final product from receipt of materials, through manufacture, to final product test.
Quality control is very important when one is using adhesives because once a product is fully
425
Copyright © 2006 by The McGraw-Hill Companies, Inc. Click here for terms of use.
426 CHAPTER NINETEEN
Mixing of materials
Thoroughness of mixing Selection and metering of
Degradation by overmixing materials
Degassing
Packaging
Storage and
shipping
bonded, joints are difficult to take apart or correct. By the time it takes to notice that one step
in the formulation or bonding process is out of control, significant costs could be incurred.
For adhesives, quality control must be defined in the broadest sense. This consists of
defining the means to prevent problems as well as to detect problems. These preventive
measures may include training, controlling manufacturing procedures, incoming inspection
of materials, and visual or physical examination of the finished product. Good quality con-
trol begins even before the receipt of materials. This usually begins with proper training of
personnel and conditioning of the manufacturing area.
The total control program monitors and provides appropriate feedback and corrective
actions. Generally, activity within each business process has its own defined quality con-
trol plan. These activities include but are not limited to
• Receiving inspection
• Manufacturing processes
• End product inspection
QUALITY CONTROL AND SPECIFICATIONS 427
Adhesives
Adherends (details)
Receiving acceptance tests
Receiving acceptance tests
Material certification
Material certification
Periodic recertification tests
Controlled storage
Controlled storage
Certified material
Surface preparation
Equipment control Certified
Process control material
Surface inspection
(test specimens)
Bond fabrication
Equipment control
Tooling control
Process control
(test specimens)
Bond cure
Process control
(test specimens)
Completed assembly
Inspection
Proof or visual test
Receiving inspection is the first step in the total control program. This generally consists of
comparison of what is received to the specification used in developing the purchase order.
Acceptance tests on raw materials should be directed toward assurance that the incoming
materials are correct and identical from lot to lot.
A receiving inspection program consists of comparison of the purchase order with what
is received. If the vendor’s test reports are required with the shipment, it is verified that they
were received and that the test values are acceptable, as defined by the original material
specification.
It may also be necessary to test the incoming bulk materials in-house. These inspections
usually consist of an evaluation of physical and chemical properties, such as color, viscos-
ity, percentage of solids, weight per gallon, and moisture content. The tests should be those
that can quickly and accurately detect deficiencies in the materials’ physical or chemical
properties or differences from lot to lot. ASTM lists various test methods that are com-
monly used for acceptance of raw materials. Several of the more common test methods are
described in Chap. 20.
Because of the cost and time required for extensive in-house testing, the trend today is
to have most of the quality control tests done by the supplier. The supplier then provides a
certified test report to the customer with shipment of the product. The test program that is
used by the supplier, internal controls, etc., are usually verified and approved by the cus-
tomer on a periodic basis, such as once per year. The approval tests are used to determine
the adequacy of the particular materials in question and the production methods that will be
employed. The tests that are done with every lot and submitted on delivery usually are made
to ascertain the consistency of the product from lot to lot.
Once identified and approved for receipt, all incoming material should be labeled with
a receipt date. This date is prominently shown on the container or material while it is in
inventory. The approved incoming material should then be stored in a controlled storage
facility. Precautions must be taken to ensure that the oldest material is used first and that
the shelf life of the product does not expire before it is used. The date stamp on the prod-
uct is the most reliable method of doing this.
The manufacturing processes that are most important in formulating are dispensing and
mixing. It is difficult to provide for specific recommendations regarding the quality control
of these processes because they vary significantly depending on the type of materials
employed and the final product required. However, it is important that all parameters in the
process, such as component proportions, temperature, and mixing parameters, be monitored
and controlled. This should be done as frequently as possible, and preferably on a continu-
ous basis. All records should be kept a sufficient time so that possible comparisons can be
made in the future regarding problems that could arise for the customer of the adhesive or
the customer’s customer (i.e., the recipient of the bonded part).
Final inspection of the adhesive products is used to satisfy the customer’s specification
and to ensure the formulator of the quality and consistency of the product. The test meth-
ods employed will vary considerably depending on the industry for which the product is
intended and the specific customer.
A flowchart for the adhesive bonding quality control system within a large aircraft com-
pany is illustrated in Fig. 19.2. This system is designed to ensure reproducible bonds and,
if a substandard bond is detected, to make suitable corrections. However, good quality con-
trol begins with proper training and conditioning of the manufacturing operation for the use
of adhesives and sealants.
QUALITY CONTROL AND SPECIFICATIONS 429
Premanufacturing Processes. The human element enters the adhesive bonding process
probably more than in any other fabrication technique. An extremely high percentage of
defects can be traced to poor workmanship or lack of understanding regarding adhesion.
This generally shows up first in the surface preparation steps but may also arise in any of
the other steps necessary to achieve a bonded assembly. This problem can be largely over-
come by proper motivation and education. All employees from design engineer to laborer
to quality control inspector should be somewhat familiar with adhesive bonding technol-
ogy and aware of the circumstances that can lead to poor joints.
In addition to the staff, the operational facilities must be well prepared prior to the use
of the adhesive. The plant’s bonding area should be as clean as possible prior to receipt of
materials. The basic approach to keeping the assembly area clean is to segregate it from the
other manufacturing operations either in a corner of the plant or in isolated rooms. The air
should be dry and filtered to prevent moisture or other contaminants from gathering at a
possible joint interface. The cleaning and bonding operations should be separated from
each other. If mold release is used to prevent adhesive flash from sticking to bonding equip-
ment, great care must be taken to ensure that the release does not contaminate the adhesive
or the adherends. Spray mold releases, especially silicone release agents, have a tendency
to migrate to undesirable areas.
As with the quality control plan for the formulator described above, acceptance tests on
adhesives should be directed toward assurance that incoming materials are identical from
lot to lot. However, this plan should also extend to the receipt and control of incoming
adherends and other materials used in the bonding process.
Important factors regarding incoming adherends are the chemical and physical proper-
ties of the material. This can be especially important with adhesive bonding because dif-
ferent metal alloys have different surface oxidation, and different elastomers and plastics
can have different additives and modifiers. With elastomeric and plastic substrates, lot-to-lot
differences should be tested. Often a supplier will change formulations but still be within
the requirements of the specification. The difference in formulation may have a profound
effect on the quality of the ultimate adhesive bond.
Another important factor with regard to adherends is the uniformity and tolerances
allowed in the joint fit-up. Improper tolerances on the geometry of the joint can lead to
adhesive-starved areas or areas where the bond is so thick that unwanted peel and cleavage
forces can easily develop.
The control over storage conditions should be similar to that described above for the for-
mulator. However, the end user must also store the substrate materials. This should be done
in an isolated area so that the substrates cannot become contaminated with the shop envi-
ronment. If stored in a cold area, the substrates must be brought up to room temperature
before the surface treatment or adhesive is applied. Moisture can condense on cold sub-
strates, leading to contamination.
In addition to tests performed on the incoming materials, test specimens may be made
to verify the strength of the adhesive joint. The quality control tests should be those that can
quickly and accurately detect deficiencies in the adhesive’s physical or chemical proper-
ties. ASTM lists various test methods that are commonly used for adhesive acceptance.
The test specimens should be stressed in directions that are representative of the forces
that the bond will see in service, i.e., shear, peel, tension, or cleavage. If possible, the spec-
imens should be prepared and cured in the same manner as actual production assemblies.
If time permits, specimens should also be tested in simulated service environments, e.g.,
high temperature, humidity.
Surface Treatment. Generally, some sort of surface preparation is required for reliable adhe-
sion. The extent of the actual surface preparation process varies according to the performance
characteristics desired, the nature of the adherend, and the time and cost considerations.
430 CHAPTER NINETEEN
The Bonding Process. Before the actual assembly operation, the cleanliness of the shop
and tools should be verified. The shop atmosphere should be controlled as closely as pos-
sible. Temperature in the range of 18 to 32°C and relative humidity from 20 to 65 percent
are best for almost all bonding operations. All parts should be fitted together without adhe-
sive or sealant to indicate possible production problems due to fit. The suitability of fit is
established by either visual inspection or direct measurement with gauge or shim. It is
desirable that the extremes in mechanical tolerances also be noted and that test specimens
be made with the worst possible fit to ensure that the bonding process will always provide
reliable joints.
The adhesive metering and mixing operation should be monitored by periodically sam-
pling the mixed adhesive and testing it for adhesive properties. Simple viscosity measure-
ments, flow tests, and visual inspection of consistency are the best methods of monitoring
conformance. A visual inspection can also be made for air entrapment and degree of mixing.
The quality control engineer should be sure that the oldest adhesive is used first and that
the specified shelf life has not been exceeded.
Innovative formulations have been developed to help the end user monitor the effi-
ciency of the mixing operation. If the epoxy resin component and the curing agent compo-
nent are different colors, the end user can easily visually determine if the products are
mixed thoroughly. Fluorescent dyes that change colors as the adhesive cures have been
added to some adhesives. Also pigments that irreversibly change color when encountering
the heat of exotherm are used in some adhesives to indicate that cure has taken place.
However, these methods of determining the state of mixing and cure lack accuracy and
should be used as guidance only.
The amount of the applied adhesive and the final bond line thickness must be monitored
because they can have a significant effect on joint strength. Curing conditions should be
monitored for pressure, heat-up rate, maximum and minimum temperatures during cure,
time at the required temperature, and cool-down rate. The primary concerns are to ensure
the following:
• The adhesive is metered and mixed as required.
• The adhesive is applied uniformly and with the correct coating thickness.
• The pot life and open time (time between adhesive application and joint assembly) of the
adhesive are not exceeded.
• The joints are assembled correctly and in the proper sequence.
QUALITY CONTROL AND SPECIFICATIONS 431
• The mixing and application of the adhesive are done under the specified environmental
conditions (temperature and relative humidity).
• The parts are held together under the specified pressure and time until cured.
• Cure time and temperature are as required and measured at the adhesive bond.
Bond Inspection. After the adhesive or sealant is cured, the joint area can be inspected to
detect gross flaws or defects. This inspection procedure can be either destructive or nonde-
structive. The nondestructive type of tests can be visual or use advanced analytical tests.
These types of bond inspections are described below.
Destructive Testing. Destructive testing generally involves placing a sample of the
production run in simulated or accelerated service and determining if it has similar
strength properties to a specimen that is known to have a good bond and adequate service
performance. It is desirable to fabricate a standard test specimen in the same production
cycle as the part actually being bonded. The test specimen could be test coupons (stan-
dard ASTM test specimens), extensions of actual parts (i.e., tabs that can be removed
from the part and tested), or special test specimens that are close to the actual part design
but amenable to mechanical testing. If a special test specimen is used, the specimen
should be designed for a test method that is easy to perform and indicative of the way the
part will be loaded in service; but it should also be designed so that the geometry and
mass do not depart too severely from those of the actual production part. The test speci-
men is then tested either immediately after bonding or after a simulated environmental
cycle.
The joint area and mode of failure should be examined closely, for this generally leads
to clues that are indicative of problems. The causes and remedies for faults revealed by such
mechanical tests and visual inspection after testing are described in Table 19.1.
Test specimens such as those mentioned above are often used to verify the quality of the
first article through the production line and then to periodically test articles for conformance.
This type of testing procedure detects discrepancies affecting the entire lot of parts, but it
cannot evaluate factors that affect individual joints or specific areas within a particular joint.
Visual, Nondestructive QC Testing of Joints. There are several forms of nondestruc-
tive testing (NDT) that can be used to determine the quality of the final joint. Those that are
most appropriate for on-line quality control are described immediately below. More sophis-
ticated nondestructive testing processes are described in the Chap. 20.
VISUAL EXAMINATION: A trained eye can detect a surprising number of faulty joints by
close inspection of the adhesive around the bonded area, even if the substrate is not trans-
parent or translucent. Close examination of the adhesive or sealant that is visually apparent
(generally around the edges of the joint) can lead to useful conclusions. Unfilled areas and
voids can sometimes be detected by noting lack of adhesive or sealant material. Misalignment
of parts is readily visible.
The texture of the adhesive around the edges of the joint can be a clue to the effective-
ness of the curing process and whether air was entrapped in the adhesive. The adhesion of
the adhesive flash to the substrate can also be qualitatively measured by attempting to pry
the flash away from the substrate. Table 19.2 lists the characteristics of faulty joints that can
be detected visually, their causes, and possible remedies.
TAP TEST: One of the first nondestructive methods used to evaluate the quality of an
adhesive joint was by tapping on the bonded joint and assessing the resulting tone. Tone
differences indicate inconsistencies in the bonded joint. This could be due to insufficient
cure, voids, or other problems. Simple tapping of a bonded joint with a coin or light ham-
mer can indicate an unbonded area. Sharp, clear tones indicate that adhesive is present and
adhering to the substrate to some degree; dull, hollow tones and inconsistencies indicate a
void or unattached area.
432 CHAPTER NINETEEN
Thick, uneven glue line Clamping pressure too low Increase pressure. Check that
clamps are seated properly
No follow-up pressure Modify clamps or check for
freedom of moving parts
Curing temperature too low Use higher curing temperature.
Check that temperature at the
bond line is above the minimum
specified throughout the curing
cycle
Adhesive exceeded its shelf life, Use fresh adhesive
resulting in increased viscosity
Adhesive residue has Excess air stirred into adhesive Vacuum-degas adhesive before
spongy appearance or application
contains bubbles
Solvent not completely dried out Increase drying time or
before bonding temperature. Make sure drying
area is properly ventilated
Adhesive material contains Seek advice from manufacturer
volatile components
A low-boiling constituent boiled Curing temperature is too high
away
Voids in bond (i.e., areas Joint surfaces not properly Check pretreatment process; use
that are not bonded), treated clean solvent and wiping rags.
clean bare metal Wiping rags must not be made
exposed, adhesive from synthetic fiber. Make sure
failure at interface cleaned parts are not touched
before bonding. Cover stored
parts to prevent dust from
settling on them
Resin may be contaminated Replace resin. Check solids
content. Clean resin tank
Substrates distorted Check for distortion; correct or
discard distorted components. If
distorted components must be
used, try adhesive with better
gap-filling properties
Adhesive can be softened Adhesive not cured properly Use higher curing temperature or
by heating or wiping extend curing time.
with solvent; adhesive Temperature and time must be
can easily be indented above minimum specified
with fingernail pressure throughout the curing cycle.
Check mixing ratios and
thoroughness of mixing. Large
parts act as a heat sink,
necessitating longer cure times
QUALITY CONTROL AND SPECIFICATIONS 433
TABLE 19.2 Faults Revealed by Visual Tests of Area around the Joint
The success of the tap test depends on the skill and experience of the operator, the back-
ground noise level, and the type of structure. Some improvement in the tap test can be
achieved by using a solenoid-operated hammer and a microphone pickup. The resulting
electric signals can be analyzed on the basis of amplitude and frequency. However, the tap
test, in its most successful mode, measures only the qualitative characteristics of the joint.
It tells whether adhesive is in the joint or not, providing an acoustical path from substrate
to substrate; or it tells if the adhesive is undercured or filled with air, thereby causing a
mechanically damped path for the acoustical signal. The tap test provides no quantitative
information and no information about the presence and/or nature of a weak boundary layer.
The success of the tap test, although limited, has led to the use of ultrasonics to deter-
mine bond quality. Ultrasonic methods are at present the most popular NDT technique for
use on adhesive joints. Ultrasonic testing measures the response of the bonded joint to load-
ing by low-power ultrasonic energy. Short pulses of ultrasonic energy can be introduced on
one side of the bonded structure and detected on the other side. This is called through-
transmission testing. An unbonded area, void, or highly damped adhesive (undercured or
filled with air) prevents the ultrasonic energy from passing efficiently through the structure.
A number of different types of ultrasonic inspection techniques have been adopted to test
adhesive bonds.1
434 CHAPTER NINETEEN
19.3 SPECIFICATIONS
Specifications are, perhaps, the cornerstone of any good quality control system.
Specifications are intended to define what is needed for all parties involved. These parties
include outside suppliers as well as internal departments within the factory. Table 19.3 is a
short list of dos and don’ts for specification writing. The language of the specification
should be clear, exact, brief, technical, and to the point.
Specifications are a necessary part of a quality control program. A specification simply
is a statement of the requirements that the adhesives or sealant must meet to be accepted for
use. A specification is an agreement between supplier and user. Conformance to a specifi-
cation does not mean that the adhesive or sealant will perform perfectly in service. It only
means that the product conforms to the specification. Specifications not only account for
the adhesive or sealant, but also define the adherends and the accessory processes for
preparing the adherends and the joint assembly.
The specification writer must try to put into the specification the requirements that, if
met, will provide the greatest likelihood of success. These requirements should be standard
tests and acceptable test limits that are agreed upon by both the supplier and the user. The
tests should be indicative of how the adhesive is used in production and how the finished
joint is to be used in service. Tests that are not directly applicable to the specific applica-
tion should not be included. Tests should not be used simply because they are standard test
methods or have been used in the past.
The language of the specification is extremely important. The specification must be
unambiguous and use direct and simple words and phases. For example, the words must
Do: Don’t:
• Use simple words. • Use words having a double meaning.
• Use the exact meaning of technical terms. • Make the specification too voluminous.
• Use short sentences. • Put in requirements that cannot be justified.
• Use the same word throughout, never synonyms. • Use trade names.
• Use the same mathematical system throughout. • Make limits too tight or too loose.
• Define requirements loosely.
QUALITY CONTROL AND SPECIFICATIONS 435
and should have very different meanings and connotations in a specification. Proprietary or
trade names should be avoided, if possible, since they may be changed with a company’s
merger or acquisition. The typical specification has the following format.3
• Title
• Scope
• General requirements
• Performance requirements
• Test methods
• Controls
• Reference documents
• Approved source list
Specifications from other sources (e.g., ASTM, military) may be used if they are applic-
able, and often test methods from ASTM, etc., are used within a customer specification.
The trend is to use well-established ASTM or ISO test methods where possible rather than
nonstandard or proprietary tests.
Specifications may require different categories of testing. For example, there may be
extensive series of tests required for initial verification or qualification of the adhesive and
supplier. These tests would be used to approve a certain product at the onset. Other receiving
tests may be used to verify the consistency of the product from lot to lot.
An adhesive or sealant specification, like all material specifications, is a document that
specifies values for all the important properties, together with limits of variability and
methods for determining these values. There are many adhesives and sealant specifications,
of which the most prominent are the industrial and government specifications. These
describe and establish the technical and physical characteristics or performance require-
ments of adhesive materials. The most common sources of standards and specifications for
the adhesives and sealants industry are the following:
TABLE 19.4 Sources of Common Specifications and Standards for Adhesives and Sealants
REFERENCES
20.1 INTRODUCTION
Adhesive tests are of interest to both the formulator and the end user. They are generally
used to do the following:
There are two categories of common tests for adhesives: fundamental property tests and
end-use tests. End-use tests, such as peel and shear, are those that try to simulate the type
of loading and service conditions to which a joint will be subjected. These tests are rela-
tively straightforward, but experience is required to establish the correct sample type and
testing procedures, judge the reliability of the resulting data, and interpret the results and
apply them to a practical application.
The determination of a fundamental or bulk property, such as viscosity, hardness, or set-
tling rate, is usually simpler and more reproducible than end-use tests. However, a correla-
tion between fundamental properties and the results of end-use testing is very difficult.
Fundamental property tests are usually employed to assess the consistency of the incoming
raw materials, adhesive formulation, or substrates. Often fundamental property testing is
done after there is a failure or unexplained occurrence to determine if a change in the
incoming material may be the possible culprit.
A number of standard tests for adhesives and sealants have been specified by the
American Society for Testing and Materials (ASTM) and other professional organizations
such as the U.S. Department of Defense and the Society of Automotive Engineers. By far,
ASTM standards are the most commonly referenced test methods. Selected ASTM stan-
dards are presented in App. G. The properties usually reported by suppliers of adhesives
and displayed in the technical literature generally reference ASTM standards.
Adhesive testing can also be classified as to the condition of the epoxy adhesive at the
time of test. Tests are conducted on the
• Raw materials, uncured epoxy mix, and cured epoxy (bulk property tests)
• Adhesive in various states of cure
• Cured adhesive and final test specimen or production joint
437
Copyright © 2006 by The McGraw-Hill Companies, Inc. Click here for terms of use.
438 CHAPTER TWENTY
The most commonly used tests for properties of the materials used in epoxy adhesive for-
mulation and uncured mixed adhesives are viscosity, shelf life, percent of solids, and mois-
ture content. These are generally applied to the resinous components.
ASTM has developed a Standard Guide for Testing Epoxy Resins, which covers several
procedures that can be conducted on the same starting sample (e.g., viscosity, color, and
density). Other contents of the standard include test methods for nonvolatile content, acid
value, epoxy content, hydrolyzable chlorine, and total chlorine. The most important of
these test methods are described below.
20.2.1 Viscosity
Viscosity is defined as the resistance of a liquid material to flow. Viscosity can be used as
an indirect determination of the epoxy resin’s molecular weight and chemical structure. It
is usually measured in fundamental units of poise or centipoise. The unit of centipoise (cP)
is sometimes confusing unless one is familiar with these particular units. Practical com-
parisons of the viscosities of common fluids may be of assistance in understanding the vis-
cosity of epoxy resins measured in centipoise.
Water 1.0
#10 Motor oil 400
Castor oil 1,000
Karo syrup 3,500
#40 Motor oil 4,500
Hershey chocolate syrup 25,000
There are also a number of specialty viscosity tests that employ their own relative units.
These tests have been developed for specific industries such as paints and coatings.
Generally, these viscosity units can be directly converted to poise.
Adhesive viscosity is an indication of how easily the product can be pumped or spread
onto a surface. It reveals information, together with the liquid’s setting rate and surface ten-
sion, that is pertinent to the wetting characteristics of the adhesive. The viscosity also
reveals information regarding the age and compounding of the adhesive. Through the rel-
atively easy measurement of viscosity, changes in density, stability, solvent content, and
molecular weight can be noticed. Viscosity measurements for free-flowing adhesives or
sealants are usually based on one of the methods described in ASTM D 1084.
The most popular viscosity test for products ranging in viscosity from 50 to 200,000 cP
uses a rotating spindle instrument such as the Brookfield viscometer. The equipment used
for this measurement is shown in Fig. 20.1. The instrument measures the resistance of the
fluid to a spindle of certain size that is rotating at a predetermined rate. The method is rel-
atively simple and quick. It can be adapted to either the laboratory or production floor.
Another test for determining the viscosity of liquid adhesives measures the time it takes
the test liquid to flow by gravity completely out of a cup with a certain size hole in the bot-
tom. These consistency cups are designed to expel 50 mL of sample in 30 to 100 s under
controlled temperature and relative humidity conditions. The number of seconds for com-
plete flow-out of the sample is determined. There are different cup volumes and hole sizes
TESTING 439
Resins melting above room temperature are often characterized by softening point rather
than by solution viscosity or elevated-temperature (melt) viscosity. Uncured, solid epoxy
resins may be qualitatively graded by determining the temperature at which they begin to
soften. Typical softening point values for DGEBA epoxy resins as a function of molecular
weight are given in Fig. 20.2.
There are several methods commonly employed to measure the softening point, such as
the ring-and-ball method and Durran’s mercury methods. These measure the temperature
at which the resin reaches an arbitrary softness.
Indentation tests, such as the ring and ball method, are the most common methods used
for determining the softening point. In Durran’s mercury method, a known weight of the
resin is melted in a standard test tube and cooled, and a known quantity of clean mercury
is placed on top of the solidified sample. The tube and contents are again heated, and the
temperature at which the molten resin rises to the top of the mercury is recorded as the melt-
ing point. Another approach for measuring the softening point is to employ a dilatometer
or thermal analysis test to measure second-order transition temperatures (i.e., sharp break
points in the thermal expansion characteristics as a function of temperatures).
Since epoxy resins are not single crystalline structures and have a broad molecular
weight distribution, sharp melting points are generally not noticeable. Any value for the
440 CHAPTER TWENTY
200
180
Softening temperature, °C
160
140
120
100
80
60
40
20
n=0 n = 1 n = 2 n = 3 n = 4 n = 5 n = 6 n = 7 n = 8 n = 9 n = 10
0 200 600 1000 1500 2000 2500 3000 3500 4000
Molecular weight
FIGURE 20.2 Typical softening point values for uncured DGEBA epoxy versus molecular weight.1
melting point is empirical and determined partly by the type of test and conditions under
which it is performed.
Next to viscosity, epoxy content and hydroxyl content are the prime properties of importance
in formulating epoxy-based adhesive systems. Epoxy content allows one to determine the
correct amount of curing agent to use. With certain curing agents it may also be important
to know the hydroxyl content. These values determine the reactivity of the system.
Epoxy content is generally determined by wet analytical techniques. ASTM D 1652 is
widely used. The most common method is based on the addition of hydrogen halide (e.g.,
hydrogen chloride, hydrogen bromide, or hydrogen iodide) to the epoxy group. The differ-
ence between the amount of acid added and the amount unconsumed, determined by titra-
tion with standard base, is a measure of the epoxy content. There are numerous epoxy assay
methods based on this technique. The specific method used depends on the resin type being
analyzed as well as on the extent to which side reactions can occur.
Epoxy content can also be determined by infrared spectroscopy. In going from lower-
to higher-molecular-weight species of the same epoxy resin, bands in the infrared spectrum
change in intensity as a percentage composition of the various groups changes. The specific
epoxy resin determines which bands should be used for the calculation of epoxy equivalent
weight.
The hydroxyl equivalent weight of epoxy resins can be determined by several methods.
The most common is esterification with acids, reaction with acetyl chloride, and reaction
with lithium aluminum hydride. Infrared spectroscopy may also be used.
The hydroxyl content is of importance in estimating average molecular weight. To an
extent, it also determines system functionality since the hydroxyl groups could be reactive
sites. Certain curing agents may preferentially react with the hydroxyl groups.
When a polymeric resin is stored for a considerable time, physical and chemical changes
may occur within the material that will affect its performance as an adhesive or sealant.
TESTING 441
The shelf life is generally defined as the time that an adhesive or sealant can be stored,
preferably under controlled conditions, and remain unchanged, at least from a fundamen-
tal property point of view. ASTM D 1337 provides a method for determining the shelf life
of an adhesive or sealant. The change in consistency (viscosity) or bond strength is mea-
sured after various storage periods at a specified temperature, as shown in Fig. 17.4. The
shelf life of the product is generally determined to be the point at which it is no longer work-
able or its viscosity has increased by a factor of 2.
All the materials to be tested for shelf life are stored in unopened containers. Storage in
containers that have been opened exposes the resin to oxygen and humidity that, depend-
ing on the type of resin, could drastically reduce the shelf life and affect final properties.
Most polymeric adhesives and sealants and their components have a shelf life greater than
6 months at room temperature. However, some one-component adhesives need to be stored
at refrigerated conditions to have a practical shelf life.
Certain adhesives or sealants may also have limited life when stored at cold tempera-
tures or when exposed to repeated freeze-thaw cycles. Generally, these are water-based
products, but freeze-thaw stability problems are not restricted to them. The resistance to
freeze-thaw cycles in low-temperature storage is measured using procedures similar to the
shelf life measurement discussed above. Even though an adhesive or sealant may have
good low-temperature storage properties, it should be brought to room temperature before
mixing or application.
The solids content of an adhesive or sealant should be checked to ensure that formulation
or dilution errors have not been made. Solids can refer to the nonvolatile component of the
adhesive or the inorganic component of the adhesive.
The nonvolatile solids content or percent of solids is usually determined by weighing a
small amount of material in a clean container, heating or curing until a constant weight is
obtained, and then weighing the container again. The percent of solids may be determined
as the ratio of the sample weights before and after curing times 100. The percent of solids
is an indirect measurement of the amount of volatile material in the sample that was driven
off during the heating cycle. The volatile material may be solvent, water, or other additives.
ASTM D 1489 offers a test method for determining the nonvolatile content of aqueous
adhesives. The solids content (or conversely the volatile content) is a measurement of the
actual nonvolatile materials that are contained within the formulation. Addition of solvent
can be used to “correct” the viscosity of adhesives or sealants to account for variability in
the manufacturing process or inconsistent base materials in the resin formulation.
The ash test is a measure of the total inorganic content in a sample. In this test, a weighed
sample is placed in a muffle furnace at approximately 540°C for 30 min or at a temperature
and time long enough to completely pyrolize any organic matter. The remaining material is
inorganic fillers, reinforcement, etc. ASTM D 5040 provides a method of measuring the ash
content of adhesives. Ash content will determine the amount of inorganic fillers in an adhe-
sive or sealant sample. The manufacturer can adjust the filler content, like the solids content,
to correct for errors or inconsistency in formulation. Higher concentrations of solvent or
filler are often used to lower the cost of the adhesives or sealant system.
The specific gravity of epoxy resins and other raw materials is the weight of the materials com-
pared with the weight of an equal volume of water. Specific gravity can be determined either
on the uncured, liquid epoxy resin or on the cured, solid epoxy. The specific gravity of the
cured specimen is generally higher than that of the uncured specimen due to cure shrinkage.
Changes in specific gravity are most noticeable when the inorganic filler content changes.
442 CHAPTER TWENTY
A variety of gravimetric methods may be used for the determination of specific gravity.
Common instruments used with the liquid resins are the hydrometer and the psychometer.
ASTM D 901 provides a method for measuring the specific gravity of low-viscosity liquids.
20.2.7 Color
Epoxy resins will contain some hydrolyzable or active chlorine depending on the comple-
tion of the epichlorohydrin reaction. It may also contain less active chlorine via chlorinated
groups along the polymeric chain. Hydrolyzable chlorine may be determined by ASTM D
1726 and total chlorine by ASTM D 1847.
The active chlorine is capable, under certain conditions, of blocking reactions with
Lewis base catalysts. The main effect of high chlorine content is to reduce system reactiv-
ity and crosslink density. Chlorine may also cause corrosion of delicate substrates. Chlorine
content is a common test for epoxy resins in the electrical and electronic industries.
Fundamental property tests are also commonly used to characterize the substrate and pro-
vide a “signature” for lot-to-lot comparisons. In all too many cases, the substrate is consid-
ered to be a constant throughout the life of a production part. In practice, there are a number
of common reasons for substrates to change.
Metallic substrates change primarily due to the nature of their surfaces. They can form
different oxide layers that are dependent on the chemistry of the surface and the way it has
been treated during fabrication of the part. The degree of surface cleaning or surface rough-
ness may also change throughout a production run due to differences in processing chemi-
cals or procedures or due to contamination of the materials that are used.
For organic substrates, the possible types of change are numerous and potentially more
significant. Polymeric substrates may change in modulus due to aging, loss of plasticizer,
or continued chemical reaction. They may change because of slight formulation variations
that take place from lot to lot. These changes may not be sufficient to cause a change in the
bulk physical properties. Therefore, the change may go unnoticed by the quality control
personnel because they are looking only at bulk properties and not at subtle formulation
alterations that may affect the adhesion properties. However, these substrate modifications
could result in drastic changes in the surface properties of the material.
A case in point is an elastomeric substrate, such as compounded nitrile rubber. Generally,
the user will specify the rubber part by compressive strength, extension, hardness, or other
TESTING 443
such standard bulk property tests. However, the supplier can formulate virtually thousands
of compositions that will meet these specified properties. In this process she or he may use
an incompatible plasticizer, or a low-molecular-weight extender, or a new filler that is more
hydroscopic. These small changes could have a very large effect on the resulting joint
strength. Therefore, any change in substrate formulation must be reverified with regard to
its adhesion characteristics.
Test methods used to determine the uniformity of substrates are numerous and vary with
the type of material. They are generally the same tests used to characterize the material or
to determine its fundamental physical properties. Tests that are commonly employed are
hardness, tensile strength, modulus, and surface characteristics such as roughness or con-
tact angle with a standard liquid. Often a test similar to the nonvolatile test mentioned above
is used to determine if there are any compounds in the substrate that are capable of out-
gassing on exposure to elevated temperatures. Moisture content of certain hydroscopic
polymers, such as nylon and polycarbonate, is also known to affect adhesion.
Many different test methods are used to measure the reactivity or cure rate of the epoxy
adhesive. Some of these, such as working life or pot life, are very practical and are used to
plan the production process. Others, such as exotherm, are used to determine reaction kinet-
ics. Still others are used to characterize the epoxy network as it cures for the purposes of
determining the degree of crosslinking and the rheological properties of the curing adhesive.
The working life of an adhesive spans the time from when the adhesive is ready for use
(i.e., mixed and ready to apply to a substrate) to the time when the system is no longer
usable because the setting mechanism has progressed to such an extent that the adhesive
is no longer workable. This characteristic is also known as the pot life of the adhesive.
The term pot life usually defines a specific mass of adhesive and a specific temperature.
Whereas working life does not denote such specifics. ASTM D 1338 establishes two pro-
cedures for determining the working life. One method uses viscosity change, and the
other uses shear strength tests as the criteria for determining when the effective working
life has expired.
Working life is usually determined on a volume of adhesive that is practical and nor-
mally used in production. The mass of the tested material must be defined in the test report
because many adhesives have a working life that is dependent on sample mass.
Structural adhesives usually require curing by the application of heat, the addition of a
catalyst, the addition of pressure, or a combination of the three. The strength developed in
the adhesive joint at various times during the curing process may be determined by lap
shear tensile specimens. This test is commonly used to determine when an adhesive or
sealant is fully cured or when the system reaches a “handling” strength so that the assem-
bled product can be moved with moderate care.
There are also several methods of determining cure rate on the bulk adhesive material.
These are generally analytical procedures that are common in most polymeric materials
444 CHAPTER TWENTY
laboratories. With these methods, such fundamental properties as dielectric loss factor,
mechanical damping characteristics, and exotherm are measured as a function of time and
temperature.
Cure time is very important for sealants as well as adhesives. Often the sealant will be
required to function as a barrier or resist the movement of substrates very soon after it is
applied. With construction sealants, for example, it may not be possible to delay the envi-
ronmental conditions until after the adhesive cures. Thus, curing time becomes a critical
parameter in selection of the sealant. ASTM C 679 covers a method for determining the
time that a mechanic can work the sealant into the joint before the sealant starts to skin or
solidify.
Simple Solvent Wipe Test. The simplest test method to determine the extent of cure is to
rub a cotton swab that has been soaked in a suitable solvent (e.g., methyl ethyl ketone)
against the surface of the cured adhesive. If the adhesive softens, it is very far from a fully
cured condition, and the degree of softness is a very gross indication of the degree of cure.
The cotton swab will remove any uncured material. The contents of the swab can be
analyzed for traces of unreacted material. Solvent extraction can be used to chemically
remove unreacted components from the swab for analytical measurement.
Direct Methods of Assessing Degree of Cure. The test methods described above are indi-
rect methods for measuring the degree of cure. They are generally used for simple quality
control tests. For more sophisticated analysis, direct methods are used to measure the
degree of cure of epoxy adhesives. These direct methods include spectroscopy, dielectric
measurements, glass transition temperature monitoring, rheological measurements, and
optical effects.
Nuclear magnetic resonance (NMR), infrared, and Raman spectroscopy are the most com-
mon methods of direct monitoring of the cure of an epoxy adhesive. Each of these techniques
produces a unique profile or spectrum of the materials being investigated. Features in the spec-
trum of the adhesive (normally peaks or shoulders) are monitored for specific changes that
relate to the formulation of new chemical bonds or the disappearance of functional groups.
These features are seen to grow, shrink, or move position on the spectrum. The distance over
which the change occurs is measured to provide a quantitative measure of the state of cure.
Dielectric test methods are used to measure the cure of epoxy adhesives between two
conducting electrodes. This method is especially appropriate for metal-to-metal joints
because the substrates themselves can be used as the electrode. The adhesive is treated as
a capacitor during the test. Its response (dielectric constant, dissipation factor, etc.) over a
range of electrical frequencies is measured as a function of curing time.
As the adhesive cures, its measured dielectric constant and dissipation factor change.
Generally the dielectric constant gradually decreases until the adhesive finally cures, at which
point the dielectric constant remains constant with cure time. The dissipation factor generally
goes through an early peak during cure, which represents the process of the liquid adhesive
becoming lower in viscosity (due to cure temperature or exotherm effects), and then increas-
ing in viscosity until gelation. The dissipation factor then gradually decreases until full cure
is achieved as ionic and polar groups are becoming restricted through crosslinking.
Dielectric monitoring has been used to determine the optimum time at which pressure can
be applied on adhesives and composites that are very pressure-dependent. This is especially
important for films and prepregs that have a high degree of retained solvent. Optimum pres-
sure should not be applied until the solvent has had a chance to escape the adhesive when it
is in the liquid state. Pressure is then applied when the solvent has escaped and the viscosity
of the adhesive is high enough to prevent squeeze-out from the joint.
Rheological measurements can also be made to determine the curing rate of an epoxy
adhesive. The viscosity will change over time. The change in viscosity is measured as a func-
tion of the resistance to rotation of a sample placed between two plates. Cone and plate
rheometry using disposable plates are employed to monitor the mechanical loss and storage
modulus of adhesive as a function of curing time. This method is more suitable for measur-
ing the early stages of cure.
The rotational viscosity method described above to measure working life or pot life is a
form of rheological measurement of cure. However, cone and plate rheometry is preferred
for accurate measurements because the specimen size and geometry are similar to those that
occur in an adhesive joint.
A number of tests on finished bonds are useful in checking the assembly processes and in
determining the adequacy of a specific adhesive for an application. These tests are gener-
ally in the form of standard test specimens or on prototype joints of the actual product.
446 CHAPTER TWENTY
A word of caution is due here. End-use data are sufficient to compare strengths of var-
ious bonding systems. The data can be used to compare the relative effectiveness of differ-
ent adhesives, surface treatments, curing schedules, and so forth. They can also be used to
separate and quantitatively define the many variables that ultimately determine the perfor-
mance of a joint. However, end-use test results cannot be readily translated into specific
strength values for an actual production joint. Actual joints generally have a complex
geometry that is significantly different from the test specimen geometry. Under the specific
environments and the kinds, frequency, and severity of the stress that the joint actually sees
in service, an adhesive or sealant may perform spectacularly better or worse than what is
represented by ASTM tests on a supplier’s data sheet.
The most noticeable differences between standard test results and the strength from an
actual joint in service are due to several factors.
1. Joint design is seldom the same.
2. The mode and application of stress loading in service are usually different and more
complex than they are in practice.
3. Environmental aging is usually less severe in service (laboratory tests tend to “acceler-
ate” aging so that the testing can be completed in a reasonable time). However, the
effects of the actual service environment are generally more complex. For example,
there may be simultaneous exposure to cyclic stress, cyclic temperature, and humid
environments.
4. Laboratory test specimens are usually made in “controlled” environments. This control
pertains to the equipment, the cleanliness (less weak boundary layer opportunity in the
lab), and the personnel (training and awareness).
5. Sample population is limited with actual production parts because of expense. Even
with laboratory specimens, a full design-of-experiment statistical process is difficult to
achieve because of the many production variables that can affect the joint strength.
Thus, the most reliable test is to measure the strength of an actual assembly under actual
operating conditions. Unfortunately, such tests are often expensive or impractical. The next
best method is to measure the strength of an actual assembly under simulated operating
conditions that do not stress the joint significantly differently than it would normally be
stressed in service.
The usefulness and limitations of standard testing methods should be understood clearly
before a testing program is established. The end user must choose the test geometry, pro-
cedure, and methodology that best serve the application. To do this, one needs to under-
stand the differences in the various test methods and the outside parameters that will affect
the data. Once the advantages and limitations of the various standard tests are understood,
the end user may find it necessary to devise his or her own methods to test specific combi-
nations of loads and environments that are anticipated.
It should be apparent by now that numerous parameters can affect the performance of a
joint, and many combinations of those parameters are possible. Therefore, a prime rule in
any adhesive or sealant testing program is to standardize and document test variables as
thoroughly as possible. The adhesive formulator, supplier, and end user should all utilize
the exact same procedures and specimen construction. One should make every effort to
ensure that similar tests that are performed at different locations in the company and by dif-
ferent personnel are identical duplicates of one another.
The physical testing of standard adhesive joints provides a standard of comparison for mate-
rials and processes that are being evaluated for a specific bonding application. Standard tests
TESTING 447
also provide a measure to control the adequacy of the bonding process, once it is estab-
lished, and of assessing its conformance to specification.
Standard test methods are useful only if they can be reproduced. It is important that the
same results be measured by both the adhesive developer and the end user. It is also impor-
tant that the results be reproducible with time and with different testing personnel. The
accuracy and reproducibility of test results depend on the conditions under which the bond-
ing process is performed. The following variables must be strictly controlled.
1. Procedures for cleaning, etching, and drying the surfaces of the substrates prior to appli-
cation of adhesive.
2. The time between surface preparation and application of adhesive and the environmen-
tal conditions present during this period. This includes the temperature and percent of
relative humidity. Usually standard atmospheric conditions are specified (25°C and
50 ± 4 percent relative humidity).
3. Complete procedures for mixing the adhesive components, if more than one.
4. Conditions and methods for application of the adhesive to the substrate surface.
5. Curing conditions, including the pressure, temperature, and time of the curing cycle. It
should be specified whether the temperature is measured within the glue line, or on
some other point on the substrate, or at some location within the curing oven. The tem-
peratures could vary significantly at these different locations depending on the weight
and size of the assembly. When an adhesive producer specifies a temperature and time
for cure, it refers to the conditions of the actual adhesive within the bond line.
6. Conditioning procedures for specimens after curing and prior to testing.
7. The rate at which the sample is loaded during test. Peel and impact tests especially are
very dependent on the speed at which the sample is tested.
A standard test report usually documents the resulting measurements, such as tensile
shear strength and peel strength. It should also indicate all the pertinent conditions that are
required to ensure reproducibility in subsequent testing. It is often very useful to describe
the failure mode of the tested specimens. An analysis of the type (or mode) of failure is an
extremely valuable tool to determine the cause of adhesive failure. The failed joint should
be visually examined to determine where and to what extent failure occurred. The percent
of the failure that is in the adhesion mode and that in the cohesion mode should be provided.
A description of the failure mode itself (location, percent coverage, uniformity, etc.) is
often quite useful. The purpose of this exercise is to establish the weak link in the joint to
better understand the mechanism of failure.
Numerous standard test methods have been developed by various government, indus-
trial, and university investigators. Many of these have been prepared or adopted under the
auspices of the ASTM Committee D 14 on Adhesives or other professional societies.
Reference to the appropriate standards will adequately equip one with the background nec-
essary to conduct the test or a version of it. Several of the more common standard tests are
described in this section. Numerous variations exist for specific applications or materials.
In these descriptions, the emphasis is on understanding of the reasons for the test, its rela-
tionship to a specific adhesive property, advantages and limitations of the test, and possi-
ble variations or extrapolations of the test method. The detailed description of the test
mechanics is kept to a minimum, since they are adequately covered in the existing stan-
dards and specifications.
Tensile Tests. The tensile strength of an adhesive joint is seldom reported in the adhesive
supplier’s literature because pure tensile stress is not often encountered in actual produc-
tion. An exception to this is the tensile test of the bonds between the skins and core of a
448 CHAPTER TWENTY
honeycomb or composite sandwich. However, not only is the tensile test useful as a qual-
ity control test for metal and sandwich adhesives, but also it can be employed to yield fun-
damental and uncomplicated tensile strain, modulus, and strength data for the adhesive.
The ASTM D 897 tensile button test is widely used to measure the tensile strength of a
butt joint made with cylindrical specimens (Fig. 20.3). The tensile strength of this bond is
defined as the maximum tensile load per unit area required to break the bond (measured in
pounds per squre inch). The cross-sectional bond area is usually specified to be equal to 1 in.2.
The specimen is loaded by means of two grips that are designed to keep the loads axially in
line. The tensile test specimen requires considerable machining to ensure parallel surfaces.
A similar specimen design uses a sandwich construction with a dissimilar material
bonded between the two cylindrical halves of the button specimen. This design is commonly
used to measure the tensile strength of adhesives between dissimilar materials or if the
adherend does not have the strength or characteristics to be machined into the shape of the
button specimen. With some modifications in the dimensions, the button tensile test has also
been adapted for testing adherence of honeycomb-cover sheets to the core (ASTM C 297).
Although developed and used primarily for testing wood joints, the cross-type tension
test, shown in Fig. 20.4, can be used to test other substrates or honeycomb specimens
(Fig. 20.4c). This simple cross-lap specimen is described in ASTM D 1344. This test
method is attractive in that it does not involve significant machining or high specimen cost,
as the button tensile test specimens do. It is very important that the specimens be thick and
rigid enough to resist bending. With only moderate bending, the loads will quickly go into
peel or cleavage stress. Because of a high degree of variation arising out of possible bend-
ing modes, a sample population of at least 10 is recommended for this test method.
Generally, the high-modulus adhesives possess the highest tensile strengths. There
appears to be an inverse relationship between thickness and tensile bond strength for par-
ticular adhesive-adherend combinations. Thin bonds, without adhesive starvation, are best.
Thicker bonds give lower strengths because cleavage forces more readily occur due to
nonaxial loading, bending of the specimens, or internal stresses in the adhesive.
Tensile Shear Tests. The lap shear or tensile shear test measures the strength of the adhe-
sive in shear. It is the most common adhesive test because the specimens are inexpensive,
To U-joint
Adhesive
To lower crosshead
FIGURE 20.3 Standard button tensile test specimen and tensile test grips.2
TESTING 449
easy to fabricate, and simple to test. However, at times it is difficult to minimize or elimi-
nate bending stresses in common shear joint specimens. Because the standard lap shear
tests introduce some degree of cleavage or peel into the adhesive joint, values obtained for
the lap shear strength of epoxy adhesives may average 4000 to 5000 psi, whereas values for
bulk tensile strengths have been reported up to 12,000 psi.4
The common lap shear test method is described in ASTM D 1002, and the standard test
specimen in shown in Fig. 20.5. This is the most commonly used shear test for structural
adhesives. However, due to the stress distribution in the adhesive arising from the joint con-
figuration, the failure strength values are of little use for engineering design purposes.
Testing is carried out by pulling the two ends of the overlap in tension, causing the adhe-
sive to be stressed in shear. Hence, this test is frequently called the tensile shear test. Since
the test calls for a sample population of 5, specimens can be made and cut from larger test
panels, illustrated in Fig. 20.5b.
450 CHAPTER TWENTY
Glue line
0.064 in
Shear area
1 in
Area
in test grips
1/2 in
1 in 1 in
21/2 in 21/2 in
Area
in test 51/2 in
grips 71/2 in
(a)
4.00 in
± 0.01 0.500 in
± 0.01
Discard
Discard
90°
±1
Typical 1.000 in
± 0.01
7.0 in ± 0.125
(b)
FIGURE 20.5 Standard lap shear test specimen design: (a) form and dimension of
lap shear specimen; (b) standard test panel of five lap shear specimens. (From ASTM
D 1002.)
The width of the lap shear specimen is generally 1 in. The recommended length of over-
lap, for metal substrates of 0.064-in thickness, is 0.5 ± 0.05 in; however, it is recommended
that the overlap length be chosen so that the yield point of the substrate is not exceeded. In
lap shear specimens, an optimum adhesive thickness exists. For maximum bond strengths,
the optimum thickness varies with adhesives of different moduli (from about 2 mils for
high-modulus adhesives to about 6 mils for low-modulus adhesives).5
The lap shear specimen can be used for determining the shear strength of dissimilar
materials in a manner similar to that described for the laminated button tension specimen.
TESTING 451
Thin or relatively weak materials such as plastics, rubber, or fabrics are sandwiched
between the adherends and tested.
Two other variations are used to avoid the bending forces that occur with simple
ASTM D 1002 specimens: the laminated lap shear specimen (ASTM D 3165) shown in
Fig. 20.6a and the double-lap specimen (ASTM D 3528) shown in Fig. 20.6b. These spec-
imens minimize the joint eccentricity and provide higher strength values than does the single-
overlap specimen. For the specimen in Fig. 20.6a, the overlap joint can be made from saw
cuts in the top and bottom substrates of a bonded laminate. This process negates the effects
of extruded adhesive at the edges of the lap and the sheared edge of the standard type of lap
shear specimen. As a result, the chances of deformation and uneven surface preparation are
lessened.
Compression shear tests are also commonly used. ASTM D 2182 describes a simple
compression specimen geometry and the compression shear test apparatus. The compres-
sion shear design also reduces bending and, therefore, peeling at the edges of the laps.
Higher and more realistic strength values are obtained with the compression shear speci-
men over the standard lap shear specimen.
Peel Test. A well-designed joint minimizes peel stress, but not all peel forces can be elim-
inated. Because epoxy adhesives are notoriously weak in peel, tests to measure peel resistance
are very important. Peel tests involve stripping away a flexible adherend from another
adherend that may be flexible or rigid. The specimen is usually peeled at an angle of 90° or
180°. The most common types of peel test are the t-peel, the floating roller peel, the 180° peel,
and the climbing drum methods. Representative test specimens are shown in Fig. 20.7. The
values resulting from each test method can be substantially different; hence it is important to
specify the test method employed.
Peel values are recorded in pounds per inch of width of the bonded specimen. They tend
to fluctuate more than any other adhesive test result because of the extremely small area
at which the stress is localized during loading. Even during the test, the peel strength val-
ues tend to fluctuate depending on the type of adhesive, adherend, and condition of the test.
In preparing the samples, care must be taken to produce void-free laminated bond lines.
FIGURE 20.7 Common types of adhesive peel test: (left) floating roller peel, (center) climbing
drum peel, (right) t-peel.7
A typical load curve of the t-peel test is shown in Fig. 20.8. Peel strength is taken as the
average value of the center portion of the curve, usually over at least a 5-in length.
The rate of peel loading is more important than in lap shear loading, and it should be
known and controlled as closely as possible. The rate at which the load is applied is usually
specified in the ASTM test procedure. Adhesive thickness also has a significant effect on
peel strength values, as does the angle of peeling.
The t-peel test is described in ASTM D 1876 and is the most popular of all peel tests.
The t-peel specimen is shown in Fig. 20.7. Generally, this test method is used when both
adherends are flexible. Because the angle of peel is controlled by the properties of the
adherends and the adhesive, the test is less reproducible than other peel tests. A sample pop-
ulation of at least 10 is required, whereas most other ASTM tests require a minimum of 5.
The floating roller peel test is used when one adherend is flexible and the other is rigid.
The flexible member is peeled through a spool arrangement to maintain a constant angle
Avg. peel
strength
0
Crosshead travel
FIGURE 20.8 Peel test record.8
TESTING 453
of peel. Thus, the values obtained are generally more reproducible. The floating roller peel
resistance test is designated ASTM D 3167.
The climbing drum peel specimen is described in ASTM D 1781. This test method is
intended primarily for determining peel strength of thin metal facings on honeycomb cores,
although it can be used for joints where at least one member is flexible. The fixtures of the
floating roller peel and drum peel tests help stabilize the angle of peel so that they gener-
ally provide higher and more reproducible peel values for a given adhesive than the t-peel
method.9,10
A variation of the t-peel test is a 180° stripping test, described in ASTM D 903. This
method is commonly used when one adherend is flexible enough to permit 180° turn near
the point of loading. This test offers more reproducible results than the t-peel test because
the angle of peel is kept constant, although it is dependent on the nature of the adherend.
Cleavage Test. Cleavage tests are conducted by prying apart one end of a rigid bonded
joint and measuring the load necessary to cause rupture. Cleavage tests are usually used in
place of peel tests when both adherends are rigid. The test is a subjective measure of the
cleavage strength of an adhesive and a qualitative measure of the fracture toughness of the
adhesive. Data obtained are adaptable to engineering design when the service stress applied
to the functional joint is in the cleavage mode.
The cleavage test utilizes a specimen where the load is intentionally placed on one edge
of the bonded area. The test method is described in ASTM D 1062. The specimen is usu-
ally loaded at a rate of 0.05 in/min until failure occurs. The failing load is reported as
breaking load per unit area in pounds per square inch. A standard test specimen is illus-
trated in Fig. 20.9. Because cleavage test specimens involve considerable machining, peel
tests are usually preferred where possible.
Fatigue Test. Fatigue testing places a given load repeatedly on a bonded joint. Lap
shear or other specimens are tested on a machine capable of inducing cyclic loading
1.000 in
1.25 in
± 0.005 in
0.625 in
0.625 in
0.25 in
Adhesive
0.25 in
(usually in tension but also in bending) on the joint. ASTM D 3166 provides procedures
for testing and measurement of the fatigue strength of lap specimens.
The fatigue strength of an adhesive is reported as the number of cycles of a known load
necessary to cause failure. Cycles to failure and the corresponding loads are plotted on
coordinates of stress versus the logarithm of the number of cycles. The point at which the
smooth curve connecting the points of minimum stress crosses the 10 million cycle line is
usually reported as the fatigue strength. Fatigue strength is dependent on the adhesive, cur-
ing conditions, joint geometry, mode of stressing, magnitude of stress, and frequency and
amplitude of load cycling.
Lap shear fatigue data are limited in engineering design because of the complex stress
distribution patterns of the lap shear specimen configuration relative to most practical joint
designs. However, fatigue testing of the actual part or of the functional joint design itself
provides useful engineering design values.
Impact Test. Impact testing is of importance because adhesives, like most polymeric
materials, are sensitive to high rates of application of force. The resistance of an adhesive
to impact can be determined by ASTM D 950. This test is analogous to the Izod impact test
method used for impact studies on materials.
The specimen is mounted in a grip, shown in Fig. 20.10, and placed in a standard impact
machine. One adherend is struck with a pendulum hammer traveling at 11 ft/s, and the energy
of impact is reported in pounds per square inch of bonded area. It is often difficult to achieve
reproducible results with impact testing, and as a result, the test is not widely used in produc-
tion situations. Impact data indicate that as the thickness of the adhesive film increases, its
apparent strength also increases.11 This suggests that a portion of the energy required to rup-
ture the bond is absorbed by the adhesive layer and is independent of adhesion.
The impact measurements of a viscoelastic adhesive are affected by the rate at which
the impact occurs. Often it is very difficult to achieve high rates of impact with laboratory
testing. One example of this is an adhesive system used to bond shock mounting pads to
electrical equipment aboard submarines. Impact testing of specimens done in a laboratory
using pendulum or drop-weight impact fixtures showed that the adhesive would not fail
under the loads expected. However, the proof test was to place the electrical equipment,
with shock mounting attached, aboard a barge and then set off explosive charges at various
depths under the barge. The proof test showed that the viscoelastic adhesive failed when
high rates of impact were experienced. The adhesive acts as a brittle polymer at high rates
of loading because the molecular chains within the material do not have time to slip by one
another and absorb a great deal of the energy. Special, high-speed impact tests have been
developed for certain applications. These tests generally use chemical explosive force or
electromagnetic energy to establish the high speeds required.
Environmental Tests. It is desirable to know the rate at which an adhesive bond will lose
strength due to environmental factors in service. Strength values determined by short-term
tests do not give an adequate indication of an adhesive’s performance during continuous envi-
ronmental exposure. Laboratory-controlled aging tests seldom last longer than a few thousand
hours. To predict the permanence of an adhesive over a 20-year product life requires acceler-
ated test procedures and extrapolation of data. Such extrapolations are extremely risky
because the causes of adhesive bond deterioration are complex (see Sec. 15.2.2). Unfortunately
no universal method has yet been established to estimate bond life accurately from short-term
aging data.
Adhesives may experience many different and exotic environments. Environmental
aging of adhesives is accomplished by exposing a stressed or unstressed joint to simulated
operating conditions. Exposure is typically to elevated temperature, water, salt spray, or
various chemical solutions that simulate the service conditions. A number of standard
chemicals that are used to soak bonded specimens for 7 days at room temperature are
TESTING 455
Impact
Testing machine
Adhesive bond
1 in 1 in
3/8 in
3/4 in
described in ASTM D 896 (see Table 20.1). The method merely outlines the environment
in which the test specimens are conditioned before testing. After suitable exposure, bonds
may be tested in whatever manner seems appropriate.
The type of stress and environment should be selected to be a close approximation of
real-life conditions. Two simultaneous environments, such as heat and moisture, may cause
the adhesive to degrade much faster than when it is exposed in any single environment
because one condition could accelerate the effect of the other.
Stress aging tests are important because they more accurately simulate service condi-
tions than simple tests where the specimens are merely hung or placed in a test environ-
ment. The Alcoa Stress Test Fixture (Fig. 20.11) is a device for applying stresses of various
magnitudes to lap shear specimens and then aging these stressed specimens in different
environments. A plot of stress to time of failure can give information about the loss of bond
strength under stressed conditions. This aging phenomenon and a typical stress-to-failure
plot are shown in Chap. 15.
456 CHAPTER TWENTY
1 −70 As conditioned
2 −30 As conditioned
3 −30 Presoaked†
4 32 As conditioned
5 73.4 50 percent relative humidity
6 73.4 Immersed in water
7 100 88 percent relative humidity
8 145 Oven, uncontrolled humidity
9 145 Over water‡
10 145 Immersed in water
11 158 Oven, uncontrolled humidity
12 158 Over water‡
13 180 Oven, uncontrolled humidity
14 180 Over water‡
15 212 Oven, uncontrolled humidity
16 212 Immersed in water
17 221 Oven, uncontrolled humidity
18 300 Oven, uncontrolled humidity
19 400 Oven, uncontrolled humidity
20 500 Oven, uncontrolled humidity
21 600 Oven, uncontrolled humidity
*
The tolerance for test temperature shall be ±1.8°F up to 180°F and ±1 percent for temperatures
over 180°F.
†
Presoaking shall consist of submerging specimen in water and applying vacuum at 20 in of
mercury until weight equilibrium is reached.
‡
The relative humidity will ordinarily be 95 to 100 percent.
Benchmark
Specimen
Bolts for applying
stress to specimen
Pin
Benchmark
9" I.D.
10" O.D.
FIGURE 20.11 Alcoa stressing fixture for measuring simultaneous effect
of stress and environmental conditions.13
TESTING 457
There are several analytical tools that provide methods of extrapolating test data. One of
these tools is the Williams, Landel, Ferry (WLF) transformation.14 This method uses the prin-
ciple that the work expended in deforming a flexible adhesive is a major component of the
overall practical work of adhesion. The materials used as flexible adhesives are usually vis-
coelastic polymers. As such, the force of separation is highly dependent on their viscoelastic
nature and is, therefore, rate- and temperature-dependent. Test data, taken as a function of rate
and temperature, can be expressed in the form of master curves obtained by WLF transfor-
mation. This offers the possibility of studying adhesive behavior over a sufficient range of
temperatures and rates for most practical applications. High rates of strain may be simulated
by testing at lower rates of strain and lower temperatures.
There are several ways that actual joints can be tested. For quality control purposes, a proof
test is commonly used, as described above. This test imposes a stress on the specimen but
limits it to a point well below where any destruction of the joint can occur. This type of test
only looks for serious flaws in the bonding processes such as interface contamination, air
entrapment in the joint, or undercured adhesive. The prototype joints can also be tested to
destruction by using similar test methods described above for standardized testing. This,
however, is generally not done at great frequency because of the cost. Generally, most
actual joint tests are performed to ascertain the cause of failure (i.e., forensic analysis of the
failed joint) or for nondestructive determination of the adequacy of the bond.
Examination of Failed Joints for Cause of Failure. Close examination of failed joints
can sometimes lead to an explanation of why the specimen failed. With visual or micro-
scopic examination it is sometimes evident that adhesive failure occurred at the original
interface due to improper wetting or at some new interface, leaving behind a thin layer of
adhesive. In comparing surface features after bond failure with the original adherend sur-
face, the maximum resolution of about 1 × 10−8 m (100 Å) for scanning electron micro-
scopes may not always be sufficient to detect a thin film of adhesive closely reproducing the
original surface profile. Optical and staining methods described by Brett15 to determine the
presence of such films are mainly applicable to fairly thick films, since optical techniques
use interference phenomenon. Films a few angstroms thick are still largely undetected.
In recent years the use of highly specialized surface characterization tools has greatly
improved the opportunity for deducing the surface composition. These tools have been
developed for the purpose of analyzing both the adherend and the adhesive. For adherends,
analytical examination generally centers on either (1) the surface chemistry for prepared
surfaces by elemental analysis, chemical species, or analysis of contaminants and bound-
ary layers or (2) analysis of failed surface for evidence of interfacial failure, failure within
the adherend (e.g., metal oxide, composite matrix or fiber), or failure within a primer or
other boundary layer. Analysis of the adhesive generally consists of characterizing the
cured film, curing agents, and failed specimen surface chemistry.
The sciences of microphotography and holography have been used for analysis of adhe-
sive bonds. Magnification and photography of the failed substrate often lead to useful clues
to the cause of failure. Holography is a method of producing photographic images of flaws
and voids using coherent light such as that produced by a laser. The major advantage of
holography is that it photographs successive “slices” through the scene volume. A true
three-dimensional image of a defect or void can then be reconstructed.
Over the last 20 years, analytical tools have become available that allow for the charac-
terization of the elemental and chemical composition of solid surfaces. The application of
these analytical tools has increased our understanding of surface properties and success-
fully characterized surface layers.
458 CHAPTER TWENTY
The most popular of these are secondary ion mass spectroscopy (SIMS), ion scattering
spectroscopy (ISS), and Auger electron spectroscopy (AES) which have been developed by
investigators such as Baun, McDevitt, and Solomon.16–20 These tools have proved practical
even when the surface films are only on the order of atomic dimensions or when the failure
occurred near the original interface and included parts of both the adhesive and the adherend.
By itself, SIMS has been shown to be a powerful tool for elemental surface characteri-
zation by Benninghoven21 and Schubert and Tracy;22 however, uncertain or rapidly chang-
ing secondary ion yield due to changes in chemical bonding makes quantitative analysis
virtually impossible using SIMS alone.23,24 SIMS is most helpful when combined with
other techniques, such as ISS and AES, which use a beam of ions of correct energy for com-
bined use with SIMS.
The greatest strength of electron spectroscopy for chemical analysis (ESCA) lies in its abil-
ity to provide information on the surface chemistry of polymers or organics. The excitation of
the surface is relatively nondestructive. AES is a useful high-spatial-resolution technique for the
analysis of metals, alloys, and inorganic materials. Polymer and organic surfaces pose problems
because of beam damage and sample charging. ISS is an elemental analysis technique with sin-
gle atomic layer resolution. ISS is often used in conjunction with other surface techniques.
Pulse echo techniques are perhaps the easiest to use in production. The sound is trans-
mitted through the part, and reflections are obtained from voids at the bond interface. The
result is generally considered only qualitative because a poorly bonded joint will show as
a good joint as long as it is acoustically coupled. A thin layer of oil or water at the interface
may also act as a coupling and disguise an unbonded area. Shear waves can be introduced
into the structure with a wedge-shaped transducer.26 This technique is effective in analyz-
ing sandwich structures.
Bonded structures that are ultrasonically tested by the immersion methods often use a
C-scan recorder to record the test. This recorder is an electric device that accepts signals
from the pulser-receiver and prints out a plan view of the part. The ultrasonic search unit is
automatically scanned over the part. The ultrasonic signals for bond or unbond are detected
from built-in reference standards. C-scan NDT techniques are used extensively by aircraft
manufacturers to inspect bonded parts.
One of the oldest and best known ultrasonic testing systems for NDT is the Fokker Bond
Tester. This method uses a sweep frequency resonance method of ultrasonic inspection.
Some degree of quantitative analysis is claimed with the Fokker Bond Tester in the aircraft
industry.
Other NDT Methods. Radiography (x-ray) inspection can be used to detect voids or
discontinuities in the adhesive bond. This method is more expensive and requires more
skilled experience than ultrasonic methods. The adhesive must contain some metal powder
or other suitable filler to create enough contrast to make defects visible. This method is
applicable to honeycomb sandwich structures as well as metal and nonmetal joints.
Thermal transmission methods are relatively new techniques for adhesive inspection.
Heat flow is determined by monitoring the surface temperature of a test piece a short time
immediately after external heating or cooling has been applied. Subsurface anomalies alter
the heat flow pattern and, thereby, affect the surface temperature. The surface temperature
difference can be detected by thermometers, thermocouples, or heat-sensitive coatings.
Liquid crystals applied to the joint can make voids visible if the substrate is heated.
Thermal transmission testing is an excellent way of detecting various types of anomalies
such as surface corrosion under paint before the corrosion becomes visually evident. Thin,
single-layer structures, such as aircraft skin panels, can be inspected for surface and subsur-
face discontinuities. This test is simple and inexpensive, although materials with poor heat-
transfer properties are difficult to test, and the joint must be accessible from both sides. For
nonmetallic materials, the defect diameter must be on the order of 4 times its depth below
the surface to obtain a reliable thermal indication. For metals, the defect diameter must be
approximately 8 times its depth. Some bright surfaces such as bare copper and aluminum do
not emit sufficient infrared radiation and may require a darkening coating on their surface.
With thermal surface impedance testing, heat is injected into the test objects’ surface
from a hot gas pulse. The resulting surface temperature transient is analyzed to determine
the bond quality in nearly real time. The surface temperature transients are sensed using an
innovative noncontacting, emissivity-independent infrared sensor.27 This method is not
adversely affected by surface blemishes or roughness.
REFERENCES
1. Lee, H., and Neville, K., Handbook of Epoxy Resins, McGraw-Hill, New York, 1968, p. 4.26.
2. Anderson, G. P., et al., “Effect of Removing Eccentricity from Button Tensile Adhesive Tests,”
Adhesively Bonded Joints: Testing , Analysis, and Design, ASTM STP 981, W. S. Johnson, ed.,
American Society of Testing and Materials, Conshohocken, PA, 1988.
3. Elliot, S. Y., “Techniques for Evaluation of Adhesives,” Chapter 31 in Handbook of Adhesive
Bonding, C. V. Cagle, ed., McGraw-Hill, New York, 1973.
460 CHAPTER TWENTY
4. Mathews, D. H., and Silver, I., Methods for Testing the Strength Properties of Adhesives and Test
Data, NEVOID Report 3923, 1962.
5. Elliot, “Techniques for Evaluation of Adhesives.”
6. Elliot, “Techniques for Evaluation of Adhesives.”
7. DeLollis, N. J., Adhesive for Metals Theory and Technology, Industrial Press, New York, 1970.
8. Elliot, “Techniques for Evaluation of Adhesives.”
9. DeLollis, N. J., Adhesives, Adherends, Adhesion, Robert E. Krieger Publishing Co., Huntington,
NY, 1980.
10. Gent, A. N., “More on the Peel Test Riddle,” Adhesives Age, February 1997, p. 59.
11. Silver, I., “Shear Impact and Shear Tensile Properties of Adhesives,” Modern Plastics, vol. 26,
no. 9, 1949, p. 95.
12. “How to Test Adhesive Properties,” Materials Engineering, March 1972, pp. 60–64.
13. Miniford, J. D., “Permanence of Adhesive-Bonded Aluminum Joints,” Chapter 23 in Adhesives in
Manufacturing, G. L. Schneberger, ed., Marcel Dekker, New York, 1983.
14. Miniford, “Permanence of Adhesive-Bonded Aluminum Joints.”
15. Brett, C. L., “The Detection of Thin Films of Epoxy Resin on Metal Surfaces,” Journal of Applied
Polymer Science, vol. 18, 1974, p. 315.
16. Baun, W. L., et al., “Chemistry of Metal and Alloy Adherends by Secondary Ion Mass
Spectroscopy, Ion Scattering Spectroscopy, and Auger Electron Spectroscopy,” ASTM STP 596,
American Society of Testing and Materials, Conshohocken, PA, March 1975.
17. McDevitt, N. T., and Baun, W. L., “Some Observations of the Relation Between Chemical Surface
Treatments and the Growth of Anodic Barrier Layer Films,” Air Force Materials Laboratory
Technical Repot 76-74, June 1976.
18. McDevitt, N. T., et al., “Surface Studies of Anodic Aluminum Oxide Layers Formed in
Phosphoric Acid Solutions,” Air Force Materials Laboratory Technical Report 77-55, May 1977.
19. McDevitt, N. T., et al., “Accelerated Corrosion of Adhesively Bonded 7075 Aluminum Using
Wedge Crack Specimens,” Air Force Materials Laboratory Technical Report 77-184, October
1977.
20. Baun, W. L., et al. “Pitting Corrosion and Surface Chemical Properties of a Thin Oxide Layer on
Anodized Aluminum,” in Air Force Materials Laboratory Technical Report 78-128, September
1978.
21. Benninghoven, A., “Statistical Method of Secondary Ion Spectroscopy,” Surface Science, vol. 28,
1971, p. 541.
22. Schubert, R., and Tracy, J. C., “Simple, Inexpensive SIMS Apparatus,” Review of Scientific
Instruments, 44:487 (1973).
23. Schubert, R., “The Analysis of 301 Stainless Steel by SIMS,” Journal of Vacuum Science
Technology, vol. 12, no. 1, 1975, p. 505.
24. Werner, R., Developments in Applied Spectroscopy, vol. 7A, 1969, p. 297.
25. Hagemaier, D. J., “End Product Nondestructive Evaluation of Adhesive Bonded Metal Joints,”
Adhesives and Sealants, vol. 3, Engineered Materials Handbook, ASM International, Materials
Park, OH, 1990.
26. Botsco, R. T., and Anderson, R. T., “Nondestructive Testing: Assuring Reliability in Critically
Bonded Structures,” Adhesives Age, May 1984, pp. 19–21.
27. Thomas, R. L., and Favro, L. D., “Thermal Wave Inspection of Adhesive Disbonding,” 41st
Annual SAMPE Symposium, March 1996.
APPENDIX A
TRADE NAMES AND
MANUFACTURERS: EPOXY
ADHESIVES, EPOXY RESINS,
CURING AGENTS AND
CATALYSTS, ADDITIVES AND
MODIFIERS
Copyright © 2006 by The McGraw-Hill Companies, Inc. Click here for terms of use.
462 APPENDIX A
TABLE A.1 Trade Names and Manufacturers of Selected Epoxy Adhesive Formulators
Copyright © 2006 by The McGraw-Hill Companies, Inc. Click here for terms of use.
TABLE B.1 Two-Component Epoxy Adhesives
Adhesive trade name Adhesive supplier Mix ratio, by weight Curing conditions Properties
Adhesives LLC
Tra Bond 2101 Tra-Con 24 h at RT Low hardness (90D); service temperature
−76 to +248°F
TABLE B.2 One-Component Epoxy Adhesives
Supreme 10HT Master Bond 60 min at 125°C High-strength adhesive with good chemical and temperature
35–45 min at 150°C resistance (to 200°C)
Supreme 10 HT/S Master Bond 60 min at 125°C Silver-filled version of Supreme 10HT for electrical
35–45 min at 150°C conductivity
EP 36 Master Bond Melts at 82°C; cures in 90–120 min B-stage epoxy; good temperature resistance (to 250°C) with
at 125–150°C flexibility and elongation. Good thermal cycling properties
Araldite AV 118 Huntsman 45 min at 125°C Good peel strength; good heat and chemical resistance
Araldite AV 8553 Huntsman 30 min at 125°C Thixotropic epoxy with good impact strength
470
Hysol EA 9602.3 Loctite-Henkel 1 h at 121°C Modified epoxy film, tacky, available supported and unsupported; exceeds
MMM-A-132 and MIL-A-25463
Hysol EA 9696 Loctite-Henkel 1 h at 121°C Moisture-resistant, toughened adhesive; service temperature to 121°C
Hysol EA 9673 Loctite-Henkel 1 h at 177°C BMI modified epoxy; superior strength to 288°C; moisture-resistant
Hysol EA 9628H Loctite-Henkel 1 h at 121°C Modified epoxy film, high peel strength; good environmental resistance
and temperature strength to 121°C
471
AF-42 3M Company 1 h at 177°C Very high peel strength (70 lb/in of width) and tensile shear strength
(4800 psi); transparent
AF-191 3M Company 1 h at 177°C High peel and shear strengths. High strength at 177°C; MIL-A-132;
supported film
AF-111 3M Company 1 h at 125°C Bonds metal to metal and metal to honeycomb; low-temperature cure
FM 73 Cytec Engineered Materials 60–90 min at 121°C Toughened epoxy film; preferred system on the PABST program for
primary bonding of aluminum
FM 94 Cytec Engineered Materials 60–90 min at 121°C Modified epoxy; provides metal-to-metal and metal-to-honeycomb bonds
FM 1000 Cytec Engineered Materials 1 h at 177°C High peel and elongation; tensile shear strength of 2900–6000 psi;
unsupported film
This page intentionally left blank
APPENDIX C
SELECTED EPOXY RESINS
Copyright © 2006 by The McGraw-Hill Companies, Inc. Click here for terms of use.
Viscosity, cP, at 25°C
Resin Epoxy equivalent weight unless indicated otherwise Comment Supplier
DER 317 192–203 16,000–25,000 Fast cure with polyamine-type curing agents Dow
DER 330 176–185 7,000–1,000 Standard undiluted DGEBA Dow
DER 331 182–192 11,000–14,000 For general purposes up to 204°C; undiluted Dow
DER 332 172–176 4,000–6,000 High-purity DGEBA Dow
DER 364 190–210 4,000–7,000 Noncrystallizing Dow
DER 383 176–183 9,000–10,500 Extended pot life Dow
EPON 825 175–180 5,000–6,500 High-purity resin with strong crystallization Resolution
tendency
474
EPON 826 178–186 6,500–9,500 Provides good chemical resistance; subject Resolution
to crystallization when cold
EPON 828 185–192 11,000–15,000 Standard undiluted DGEBA Resolution
EPON 8280 185–195 11,000–15,000 Antisettling and antifoaming resin Resolution
EPON 830 170–225 17,000–22,500 Minimal crystallization tendency Resolution
EPR 164 188 10,000 Bakelite AG
EPR 154 180 8,500 Bakelite AG
Epalloy 7200 Modified DGEBA; high reactivity for
low-temperature cures CVC
Liquid bisphenol F type
DER 324 195–204 600–800 Aliphatic glycidyl ether diluent in DGEBA Dow
DER 353 190–200 800–1,000 Aliphatic glycidyl ether diluent in bisphenol F Dow
EPR 355 167 500 Bisphenol A/bisphenol F blend Bakelite AG
EPR 362 178 500 Bisphenol A Bakelite AG
EPR 275 253 550 Bisphenol A Bakelite AG
(Continued)
(Continued)
DER 542 305–355 Semisolid Generally blended with other DGEBA resins Dow
EPON 1163 380–410 Semisolid Low-melting; containing 48% bromine Resolution
EPON 1124-A-80 425–443 Acetone solution Conventional brominated epoxy for FR-4 applications Resolution
Epoxy acrylate resins
EPON 58901 195–210 1,000–5,000 at 52°C CTBN elastomer epoxy functional adduct with Resolution
EPON 828; provides tack and early green strength to
adhesives
EPON 58135 208–220 250–500 at 52°C Elastomer modified bisphenol A specifically designed Resolution
for adhesion to vinyl
This page intentionally left blank
APPENDIX D
EPOXY CURING AGENTS
Copyright © 2006 by The McGraw-Hill Companies, Inc. Click here for terms of use.
Active hydrogen Viscosity, cP, Mix ratio, pph Curing conditions;
Curing agent Type equivalent weight at 25°C (DGEBA EEW: 185) comments Supplier
Polyamines
at room temperature
HHPA Hexahydrophthalic 60–75 Low-melting-point
anhydride (35°C) solid; soluble
in DGEBA at RT
TMA Trimellitic anhydride 60–90 Good high-temperature
properties; reacts
rapidly at high
temperatures
DDSA Dodecenyl succinic 95–130 Liquid anhydride;
anhydride imparts flexibility
to cured composition
PA Phthalic anhydride 40–65 Solid anhydride with
melting point of
128°C; low exotherm
and long pot life
BTDA Benzophenonetetracarboxylic 40–50 Good high-temperature properties;
dianhydride solid at RT
(Continued)
(Continued)
Catalysts
The following is an index to complete epoxy adhesive formulations that are presented in
this book. Generic or less complete formulations can be also found in many of the tables
and figures.
(Continued)
485
Copyright © 2006 by The McGraw-Hill Companies, Inc. Click here for terms of use.
486 APPENDIX E
(Continued)
Note: All formulations in the following tables of this appendix are presented on a
parts-by-weight basis unless otherwise indicated.
487
Copyright © 2006 by The McGraw-Hill Companies, Inc. Click here for terms of use.
TABLE F1 Surface Preparation Methods for Metal Substrates
Aluminum and Trichloroethylene 1. Sandblast or 100-grit emery cloth followed by solvent Medium- to high-strength bonds, suitable for
aluminum alloys degreasing. noncritical applications
2. Immerse for 10 min at 70–82°C in a commercial Optimum bond strength per FPL etch process.
alkaline cleaner or Specified in ASTM D 2651 and MIL-A-9067.
Sodium metasilicate 3.0 Solvent degrease may replace alkaline cleaning.
Sodium hydroxide 1.5
Sodium dodecylbenzene sulfonate, such as 1.5
488
be corrosion-protected
Brass and bronze (see Trichloroethylene Etch for 5 min at 20°C in Temperature must not exceed 65°C when washing
also copper and Zinc oxide 20 and drying.
copper alloys) Sulfuric acid (96%) 460
Nitric acid (69%) 360
(Continued)
TABLE F1 Surface Preparation Methods for Metal Substrates (Continued)
Chromium Trichloroethylene 1. Abrasion. Grit or vapor blast, or 100-grit emery cloth, Suitable for general-purpose bonding
followed by solvent degreasing.
2. Etch for 1–5 min at 90–95°C in For maximum bond strength
Hydrochloric acid (37%) 17
Water 20
(Continued)
TABLE F1 Surface Preparation Methods for Metal Substrates (Continued)
Gold Trichloroethylene Solvent or vapor degrease after light abrasion with a fine
emery cloth
Iron See steel (mild)
Lead and lead-based Trichloroethylene Abrasion. Grit or vapor blast, or 100-grit emery cloth
solders followed by solvent degreasing
492
Magnesium and Trichloroethylene 1. Abrasion with 100-grit emery cloth followed by solvent Apply the adhesive immediately after abrasion
magnesium alloys degreasing
2. Vapor-degrease. Immerse for 10 min at 60–70°C in Medium to high bond strength. ASTM D 2651
Deionized water 95 alkaline detergent treatment
Sodium metasilicate 2.5
Trisodium pyrophosphate 1.1
Sodium hydroxide 1.1
Sodium dodecylbenzene sulfonate, such as 0.3
Nacconol 90G
(Continued)
TABLE F1 Surface Preparation Methods for Metal Substrates (Continued)
Steel (stainless) Trichloroethylene 1. Abrasion with 100-grit emery cloth, grit or vapor blast, Dry grit or sandblasting tends to warp thin-sheet
followed by solvent degreasing. materials; these methods are suited only for thick
section parts.
2. Solvent-degrease and abrade with grit paper. Degrease ASTM D 2651 sulfuric, nitric, hydrofluoric etch
again. Immerse for 10 min at 65–71°C in the following
solution:
Water 90
494
(Continued)
TABLE F1 Surface Preparation Methods for Metal Substrates (Continued)
Steel (mild, iron, and Trichloroethylene 1. Abrasion. Grit or vapor blast followed by solvent Xylene or toluene is preferred to acetone and
ferrous metals other degreasing with water-free solvents. ketone, which may be moist enough to cause
than stainless) rusting
2. Etch in the following solution for 5 min at 23°C: ASTM D 2651 nitric-phosphoric acid etch.
Deionized water 64.99% by volume Bonding should follow immediately after etching
Surfactant 0.01% by volume treatments since ferrous metals are prone to
Phosphoric acid (85%) 30% by volume rusting. Abrasion is more suitable for procedure
where bonding is delayed
496
4. Etch for 10–15 min at 38–52°C in Alternative etch for alloys to be bonded with
Nitric acid (69%) 6 polyimide adhesives is nitric : hydrofluoric : water
Hydrofluoric acid (60%) 1 in a ratio of 5 : 1 : 27 by wetght. Etch 30 s
Water 20 at 20°C.
(Continued)
TABLE F1 Surface Preparation Methods for Metal Substrates (Continued)
6. Immerse for 15 min at 76°C in the following solution to ASTM D 2651 nitric hydrofluoric etch
make 1 gal (3.6 L):
Caustic cleaner, such as Vitro-Klene 6–8 oz
(Turco Purex Industries, Carson, CA)
Water Remainder
498
Acetal (copolymer) Acetone 1. Abrasion. Grit or vapor blast, or medium-grit emery For general-purpose bonding
cloth followed by solvent degreasing.
2. Etch in the following acid solution for 10 s at 25°C: For maximum bond strength. ASTM D 2093
Potassium dichromate 75
Distilled water 120
Sulfuric acid (96%) 1500
500
SAN, polysulfone,
and rigid vinyl
Methyl pentene Acetone 1. Abrasion. Grit or vapor blast or 100-grit emery cloth For general-purpose bonding
followed by solvent degreasing.
2. Immerse for 1 h at 60°C in
Potassium chromate 7.5
Water 27.5
Sulfuric acid (96%) 65.0
(Continued)
TABLE F2 Surface Preparation Methods for Plastic Substrates (Continued)
3. Prime with epoxy adhesives and fuse into the surface Bond within 15 min of treatment
by heating for 10 min at 370°C followed by 5 min at
400°C.
4. Expose to one of the following gases activated by
corona discharge:
Air (dry) for 5 min
Air (wet) for 5 min
Nitrous oxide for 10 min
Nitrogen for 5 min
5. Expose to electric discharge from a tesla coil Bond within 15 min of treatment
(50 kV ac) for 4 min.
(Continued)
TABLE F2 Surface Preparation Methods for Plastic Substrates (Continued)
Polyesters, Detergent, acetone, 1. Abrasion. Grit or vapor blast or 100-grit emery cloth For general-purpose bonding
polyethylene methyl ethyl ketone followed by solvent degreasing.
terphthalate (Mylar) 2. Immerse for 10 min at 70–95°C in For maximum bond strength. Suitable for Mylar
Sodium hydroxide 2 films
Water 8
ketone
Polyvinylchloride, Trichloroethylene, 1. Abrasion grit or vapor blast or 100-grit emery cloth Suitable for rigid plastic. For maximum strength,
polyvinylidene methyl ethyl ketone followed by solvent degreasing. prime with nitrile phenolic adhesive
chloride, polyvinyl 2. Solvent wipe with ketone. Suitable for plasticized material
fluoride
Styrene acrylonitrile Trichloroethylene Solvent degrease.
Urea formaldehyde Acetone, methyl ethyl Abrasion. Grit or vapor blast or 100-grit emery cloth
ketone followed by solvent degreasing.
TABLE F3 Surface Preparation Methods for Elastomeric Substrates
Natural rubber Methanol, isopropanol 1. Abrasion followed by brushing. Grit or vapor blast or For general-purpose bonding
280-grit emery cloth, followed by solvent wipe.
2. Treat the surface for 2–10 min with sulfuric acid (96%) Adequate pretreatment is indicated by the
at RT. Rinse thoroughly with cold water/hot water. Dry appearance of hairline surface cracks on flexing
after rinsing in distilled water. Residual acid may be the rubber. Suitable for many synthetic rubbers
neutralized by soaking for 10 min in 10% ammonium when given 10–15 min etch at RT. Unsuitable for
hydroxide after hot water washing. use on butyl, polysulfide, silicone, chlorinated
polyethylene, and polyurethane rubbers
3. Treat surface for 2–10 min with paste made from
sulfuric acid and barium sulfate. Apply paste with
stainless steel spatula and follow procedure 2 above.
4. Treat surface for 2–10 min in Suitable for those rubbers amenable to treatments
506
(Continued)
TABLE F3 Surface Preparation Methods for Elastomeric Substrates (Continued)
elastomer system; 1% by weight is usually sufficient. to adhesive eliminates need for priming and
improves adhesion to glass and metals. Silane
may be used as a surface primer
Silicone Acetone, methanol 1. Application of primer (e.g., Chemlok 607, Lord
Chemical Co.).
2. Exposure to oxygen gas activated by corona discharge
for 10 min.
TABLE F4 Surface Preparation Methods for Miscellaneous Substrates
Brick and fired Methyl ethyl ketone Abrade surface with a wire brush; remove all dust and
nonglazed building contaminants.
materials
Carbon graphite Acetone Abrasion. Abrade with 220-grit emery cloth and For general-purpose bonding
solvent-degrease after dust removal.
Glass and quartz Acetone, detergent 1. Abrasion grit blast with carborundum and water slurry, For general-purpose bonding. Drying process
(nonoptical) and solvent-degrease. Dry for 30 min at 100°C. Apply improves bond strength
the adhesive before the glass cools to RT.
2. Immerse for 10–15 min at 20°C in For maximum strength
Sodium dichromate 7
Water 7
Sulfuric acid (96%) 400
509
(Continued)
TABLE F4 Surface Preparation Methods for Miscellaneous Substrates (Continued)
Concrete, granite, Perchloroethylene 1. Abrasion. Abrade with a wire brush, degrease with For general-purpose bonding
stone detergent, and rinse with hot water before drying.
2. Etch with 15% hydrochloric acid until effervescence Applied by stiff bristle brush. Acid should be
ceases. Wash with water until surface is litmus-neutral. prepared in a plastic pail. The 10–12% hydro-
Rinse with 1% ammonia and water. Dry thoroughly chloric or sulfuric acids are alternative etchants;
before bonding. 10% w/w sodium bicarbonate solution may be
used instead of ammonia for acid neutralization
Wood, plywood Abrasion. Dry wood is smoothed with a suitable emery For general-purpose bonding
paper. Sand plywood along the direction of the grain.
Painted surface Detergent 1. Clean with detergent solution, abrade with a medium Bond generally as strong as the paint
emery cloth, final wash with detergent.
2. Remove paint by solvent or abrasion, and pretreat For maximum adhesion
510
exposed base.
APPENDIX G
SPECIFICATIONS
AND STANDARDS
ASTM standard
(Continued)
511
Copyright © 2006 by The McGraw-Hill Companies, Inc. Click here for terms of use.
512 APPENDIX G
ASTM standard
ASTM standard
(Continued)
514 APPENDIX G
ASTM standard
D 3164 Test Method for Determining the Strength of Adhesively Bonded Plastic Lap
Shear Sandwich Joints in Shear by Tension Loading
D 3165 Test Method for Strength Properties of Adhesives in Shear by Tension Loading
of Laminated Assemblies
D 3166 Test Method for Fatigue Properties of Adhesives in Shear by Tension Loading
(Metal-to-Metal)
D 3167 Test Method for Floating Roller Peel Resistance of Adhesives
D 3310 Test Method for Determining Corrosivity of Adhesive Materials
D 3418 Test Methods Relating to Glass Transition Temperature
D 3433 Practice for Fracture Strength in Cleavage of Adhesives in Bonded Joints
D 3434 Practice for Multiple Cycle Accelerated Aging Test (Automatic Boil Test) for
Exterior Wet Use Wood Adhesives
D 3482 Test Method for Determining Electrolytic Corrosion of Copper by Adhesives
D 3498 Specification for Adhesives for Field Gluing Plywood to Lumber Framing for
Wood Systems
D 3528 Test Method for Strength Properties of Double Lap Shear Adhesive Joints by
Tension Loading
D 3535 Test Method for Resistance to Deformation Under Static Loading for
Structural Wood Laminating Adhesives Used Under Exterior (Wet Use)
Exposure Conditions
D 3632 Practice for Accelerated Aging of Adhesive Joints by the Oxygen Pressure
Method
D 3658 Test Method for Determining the Torque Strength of Ultraviolet (UV) Light
Cured Glass-to-Metal Adhesive Joints
D 3706 Test Method for Hot Tack of Wax Polymer Blends by the Flat Spring Test
D 3747 Specification for Emulsified Asphalt Adhesive for Adhering Roof Insulation
D 3762 Test Method for Adhesive Bonded Surface Durability of Aluminum
(Wedge Tests)
D 3807 Test Method for Strength Properties of Adhesive in Cleavage Peel by Tension
Loading (Engineering Plastics-to-Engineering Plastics)
D 3808 Practice for Qualitative Determination of Adhesion of Adhesives to Substrate
by Spot Adhesion Test Method
D 3929 Practice for Evaluating the Stress Cracking of Plastics by Adhesives Using the
Bent-Beam Method
D 3930 Specification for Adhesives for Wood Based Materials for Construction of
Manufactured Homes
D 3931 Test Method for Determining Strength of Gap Filling Adhesive Bonds in
Shear by Compression Loading
D 3932 Practice for the Control of the Application of Structural Fasteners When
Attached by Hot Melt Adhesives
D 3933 Practice for Preparation of Aluminum Surface for Structural Adhesive Bonding
(Phosphoric Acid Anodizing)
D 3983 Test Method for Measuring the Strength and Shear Modulus of Non-rigid
Adhesives by the Thick Adherend Tensile Lap Specimen
D 4027 Test Method for Measuring Shear Properties of Structural Adhesive by the
Modified Rail Test
D 4299 Test Methods for Effect of Bacterial Contamination on Permanence of
Adhesive Preparations and Adhesive Films (Discontinued 1990; replaced by
D 4300 and D 4783)
SPECIFICATIONS AND STANDARDS 515
ASTM standard
D 4300 Test Methods for the Ability of Adhesive Films to Support or Resist the
Growth of Fungi
D 4317 Specification for Polyvinyl Acetate Based Emulsion Adhesives
D 4338 Test Method for Flexibility Determination of Supported Adhesive Films by
Mandrel Bend Test Method
D 4339 Test Method for the Determination of Odor of Adhesives
D 4426 Test Method for Determination of Percent Non-volatile Content of Liquid
Phenolic Resins Used for Wood Laminating
D 4497 Test Method for Determining the Open Time of Hot Melt Adhesives
(Manual Method)
D 4498 Test Method for Heat-Fail Temperature in Shear of Hot Melt Adhesives
D 4499 Test Method for Heat Stability of Hot Melt Adhesives
D 4500 Test Method for Determining Grit, Lumps, or Undissolved Matter in Water
Borne Adhesives
D 4501 Test Method for Shear Strength of Adhesive Bonds Between Rigid Substrates
by the Block-Shear Method
D 4502 Test Method for Heat and Moisture Resistance of Wood Adhesives Joints
D 4562 Test Method for Shear Strength of Adhesives Using Pin-and-Collar Specimen
D 4680 Test Method for Creep and Time to Failure of Adhesives in Static Shear by
Compression Loading (Wood-to-Wood)
D 4688 Test Methods for Evaluating Structural Adhesives for Fingerjointing Lumber
D 4689 Specification for Adhesives, Casein Type
D 4690 Specification for Urea-Formaldehyde Resin Adhesives
D 4783 Test Methods for Resistance of Adhesives Preparation in Container to Attack
by Bacteria, Yeast and Fungi
D 4800 Guide for Classifying and Specifying Adhesives
D 4896 Guide for the Use of Adhesive Bonded Single Lap-Joint Specimen Test
Results
D 5040 Test Methods for Ash Content of Adhesives
D 5041 Test Method for Fracture Strength in Cleavage of Adhesives in Bonded Joints
D 5113 Test Method for Determining Adhesive Attack on Rigid Cellular Polystyrene
Foam
D 5215 Test Method for Instrumental Evaluation of Staining of Vinyl Flooring by
Adhesives
D 5266 Practice for Estimating the Percentage of Wood Failure in Adhesive Bonded
Joints
D 5267 Test Method for Determination of Extrudability of Cartridge Adhesives
D 5330 Specification for Tape, Pressure Sensitive, Packaging, Filament Reinforced
D 5375 Test Methods for Liner Removal at High Speeds from Pressure Sensitive
Label Stock
D 5570 Test Method for Water Resistance of Tape and Adhesives Used as a Box
Closure
D 5574 Test Methods for Establishing Allowable Mechanical Properties of Wood
Bonding Adhesives for Design of Structural Joints
D 5677 Specification for Fiberglass (Glass Fiber Reinforced Thermosetting Resin)
Pipe and Pipe Fittings, Adhesive Bonded Joint Type, for Aviation Jet
Turbine Fuel Lines
(Continued)
516 APPENDIX G
ASTM standard
C 510 Test Method for Staining and Color Change of Single or Mulicomponent
Joint Sealants
C 603 Test Method for Extrusion Rate and Application Life of Elastomeric Sealants
C 711 Test Method for Low Temperature Flexibility and Tenacity of One Part
Elastomeric Solvent Release Type Sealants
C 712 Test Method for Bubbling of One Part Elastomeric Solvent Release Type
Sealants
SPECIFICATIONS AND STANDARDS 517
ASTM standard
C 719 Test Method for Adhesion and Cohesion of Elastomeric Joint Sealants Under
Cyclic Movement (Hockman Cycle)
C 794 Test Method for Adhesion in Peel of Elastomeric Joint Sealants
C 811 Recommended Practice for Surface Preparation of Concrete for Application of
Chemical Resistance Resin Monolithic Surfaces
C 879 Method of Testing Release Papers Used with Preformed Tape Sealants
C 906 Test Method for T-Peel Strength of Hot Applied Sealants
C 907 Test Method for Tensile Adhesive Strength of Preformed Tape Sealants by
Disk Method
C 920 Specification for Elastomeric Joint Sealants
C 972 Test Method of Compression Recovery of Tape Sealant
C 1016 Standard Test Method for Determination of Water Absorption of Sealant
Backing (Joint Filler Material)
C 1087 Standard Test Method for Determining Compatibility of Liquid Applied
Sealants with Accessories Used in Structural Glazing Systems
C 1135 Determining Tensile Adhesion Properties of Structural Sealants
C 1184 Specification for Structural Silicone Sealants
C 1247 Test Method for Durability of Sealants Exposed to Continuous Immersion in
Liquids
C 1248 Test Method for Staining of Porous Substances by Joint Sealants
C 1249 Guide for Secondary Seal for Sealed Insulating Glass Units for Structural
Sealant Glazing Applications
C 1253 Test Method for Determining the Outgassing Potential of Sealant Backing
C 1257 Test Method for Accelerated Weathering of Solvent Release Type Sealants
C 1265 Standard Test Method for Determining the Tensile Properties of an Insulating
Glass Edge Seal for Structural Glazing Applications
C 1294 Standard Test Method for Compatibility of Insulating Glass Edge Sealants
with Liquid Applied Glazing Materials
C 1311 Specification of Solvent Release Sealants
C 1330 Specification for Cylindrical Sealant Backing for Use with Cold Liquid
Applied Sealants
C 1369 Specification for Secondary Edge Sealants for Structurally Glazed Insulating
Glass Units
C 1375 Guide for Substrates Used In Testing Building Seals and Sealants
C 1392 Guide for Evaluating Failure of Structural Sealant Glazing
C 1394 Guide of In-Situ Structural Silicone Glazing Evaluation
D 1985 Practice for Preparing Concrete Blocks for Testing Sealant for Joints and
Cracks
D 2202 Test Method for Slump of Sealants
D 2203 Test Method for Staining from Sealants
D 2377 Test Method for Tack Free Time of Caulking Compounds and Sealants
D 2828 Specification for Nonbituminous Inserts for Contraction Joints in Portland
Cement Concrete Airfield Pavements, Sawable Type
D 3406 Specification for Joint Sealant, Hot Poured, Elastomeric Type, for Portland
Cement Concrete Pavements
D 3538 Standard Test Method for Strength Properties of Double Lap Shear Adhesives
Joints by Tension Loading
(Continued)
518 APPENDIX G
ASTM standard
D 3569 Specification for Joint Sealant, Hot Applied, Elastomeric, Jet Fuel Resistant
Type for Portland Cement Concrete Pavements
D 3581 Specification for Joint Sealant, Hot Applied, Jet Fuel Resistant Type, for
Portland Cement Concrete & Tar-Concrete Pavements
D 3910 Practice for Design, Testing, and Construction of Slurry Seal
D 4070 Specification for Adhesive Lubricant for Installation of Preformed Elastomeric
Bridge Compression Seals in Concrete Structures
D 4259 Practice for Abrading Concrete
D 4260 Practice for Acid Etching Concrete
D 5167 Practice for Melting of Hot Applied Joint and Crack Sealant and Filler for
Evaluation
D 5249 Specification for Backer Materials for Use with Cold and Hot Applied Joint
Sealants in Portland Cement Concrete and Asphalt Joints
D 5749 Standard Specification for Reinforced and Plain Gummed Tape for Sealing
and Securing
D 5893 Specification for Cold Applied, Single Component, Chemically Curing
Silicone Joint Sealant for Portland Cement Concrete Pavements
SPECIFICATIONS AND STANDARDS 519
Specification
(Continued)
520 APPENDIX G
Specification
MIL-C-8073 Core Material, Plastic Honeycomb, Laminated Glass Fabric Base, for Aircraft
Structural Applications
MIL-C-81986 Core, Material, Plastic Honeycomb, Nylon Paper Base: for Aircraft Structural
Applications
MIL-C-83019 Protective Coating, Integral Fuel Tank
MIL-C-83231 Coating, Polyurethane, Rain, Erosion Resistant
MIL-C-8514 Resin-Acid Metal Pretreatment Compound for Aircraft
MIL-C-897 Cement Rubber
MIL-D-17951 Sealing Compound for Use with Deck Covering
MIL-G-413 Marine Glue
MIL-P-20628 Putty, Sealing
MIL-P-23236 Paint Coating, Ship Fuel, Salt Tanks
MIL-P-46276 Primer, Bonding
MIL-R-17882 Epoxy Resin for Pipe Repair
MIL-S-11030 Non-curing, Polysulfide Base Sealing Compound
MIL-S-11031 Adhesive, Curing Compound for Bonding Metal to Metal
MIL-S-12935 Synthetic Resin Lumber Knot Sealing Compound
MIL-S-15204 High Temperature Sealing Compound
MIL-S-17377 Compound Composed of Plastics or Resinous Binders
MIL-S-20541 Liquid Rubber Cement
MIL-S-22473 Anaerobic Single Component Sealing Compound
MIL-S-23498 Sealing Compound, Bearing Preservation
MIL-S-24340 Sealing, Ship Deck Polyurethane
MIL-S-3927 Sealing Compound, Thread
MIL-S-4383 Sealing Compound Synthetic Rubber Base
MIL-S-45180 Thread and Gasket Sealing Compounds
MIL-S-46163 Sealing, Lubricating and Wicking Compounds Thread Locking
MIL-S-7502 High Adhesion Sealant
MIL-S-7916 Thread and Gasket Sealing Compound
MIL-S-81732 Sealing Compound, Electrical
MIL-S-81733 Sealing, Coating, Corrosion Inhibitive
MIL-S-83315 Sealing, Aluminum Structure
MIL-S-83318 Sealant, Quick Repair Integral Fuel Tank
MIL-S-83430 Sealing Compound, Integral Fuel Tanks and Fuel Cell Cavities
MIL-S-8516 Sealing Compound—Protects Electrical Components
MIL-S-8784 Sealing Compound for Aircraft Fuel Tank
MIL-S-8802 High Temperature Sealant
MIL-T-5542 Thread Compound, Antiseize and Sealing
MIL-T-83483 Thread Compound, Antiseize, Molybdenum Disulfide Petrolatum
MMM-A-001058 Adhesive, Rubber Base Pressurized Dispensers
MMM-A-001993 Adhesive, Epoxy, Flexible, Filled
MMM-A-100 Animal Glue
MMM-A-115 Asphalt Tile Cement
MMM-A-121 Adhesive, Binding
MMM-A-122 High Strength Adhesive
MMM-A-125 Casein Glue in Powder Form
MMM-A-132 Adhesive, Heat Resistant
MMM-A-134 Adhesive, Epoxy Resin, Metal to Metal Structural Bonding
MMM-A-137 Strong Linoleum Cement
MMM-A-138 Adhesive, Metal to Wood
SPECIFICATIONS AND STANDARDS 521
Specification
ARP 1524 Surface Preparation and Priming of Aluminum Alloy Parts for High Durability
Structural Adhesive Bonding, Phosphoric Acid Anodizing
APR 1575 Surface Preparation and Priming of Aluminum Alloy Parts for High Durability
Structural Adhesive Bonding, Hand Applied Phosphoric Acid Anodizing
ARP 1843 Surface Preparation for Structural Adhesive Bonding Titanium Alloy Parts
ARP 4069 Aerospace Recommended Practice for Sealing Integral Fuel Tanks
AMS 1320 Decal Adhesive Remover
AMS 3374 Sealing Compound, One Part Silicone, Aircraft Firewall
AMS 3375 Adhesive/Sealant, Fluorosilicone Aromatic Fuel Resistant, One Part Room
Temperature Vulcanizing
AMS 3376 Sealing Compound, Noncuring, Groove Injection, Temperature and Fuel
Resistant
AMS 3491 Surface Treatment of Polytetrafluoroethylene
AMS 3106 Primer, Adhesive, Corrosion Inhibiting, −67 to 200°F
AMS 3107 Primer, Adhesive, Corrosion-Inhibiting, for High Durability Structural
Adhesive Bonding
AMS 3681 Adhesive, Electrically Conductive, Silver Organic Base
AMS 3685 Adhesive, Synthetic Rubber, Buna N Type
AMS 3686 Adhesive Polyimide Resin, Film and Paste, High Temperature Resistant, 315°C
or 600°F
AMS 3687 Adhesive Film, Humidity Resistant, for Sandwich Panels
AMS 3688 Adhesive, Foaming, Honeycomb Core Splice, Structural, −55 to 82°C
AMS 3689 Adhesive, Foaming, Honeycomb Core Splice, Structural, −55 to 177°C
AMS 3690 Adhesive Compound, Epoxy, Room Temperature Curing Adhesive Compound,
Epoxy, Room Temperature Curing
AMS 3691 Adhesive Compound Epoxy, Medium Temperature Application
AMS 3692 Adhesive Compound, Epoxy, High Temperature Application
AMS 3693 Adhesive, Modified Epoxy, Moderate Heat Resistant, 120°C Curing, Film Type
AMS 3695 Adhesive Film, Epoxy Base for High Durability Structural Adhesive Bonding
AMS 3696 Aerodynamic Fairing Compound, −55 to 85°C
AMS 3697 Aerodynamic Fairing Compound, −55 to 150°C
AMS 3698 Adhesive Film, Hot Melt, Addition Type Polyimide, for Foam Sandwich
Structure, −67 to 450°F
AMS 3704 Adhesive, Contact Chloroprene, Resin Modified
J 1523 Recommended Practice for Metal to Metal Overlap Shear Strength Test for
Automotive Type Adhesives
J 1525 Recommended Practice for Lap Shear for Automotive Type Adhesives for
Fiber Reinforced (FRP) Bonding
American Architectural Manufacturers Association
AAMA 1407.1 Voluntary Specification for a Single Component Sealant for Residential Sheet
Products
AAMA 850 Penetration Sealants Guide Manual
AAMA CW-13 Curtain Wall Manual No. 13—Structural Sealant Glazing Systems
American Concrete Institute
Copyright © 2006 by The McGraw-Hill Companies, Inc. Click here for terms of use.
526 APPENDIX H
To convert To Multiply by
Area
ft ⋅ lbf J 1.355818
N⋅m J 1.000000
Length
lb kg 4.5359237 × 10−1
oz gm 28.35
Pressure
527
Copyright © 2006 by The McGraw-Hill Companies, Inc. Click here for terms of use.
528 INDEX
Copyright © 2006 by The McGraw-Hill Companies, Inc. Click here for terms of use.