JOURNAL OF

COMPOSITE
Article M AT E R I A L S

Journal of Composite Materials

2021, Vol. 55(1) 145–156
Ageing effects after ozone and water ! The Author(s) 2020
Article reuse guidelines:
immersion on tensile strength at room sagepub.com/journals-permissions
DOI: 10.1177/0021998320943947
and high temperatures of carbon/epoxy journals.sagepub.com/home/jcm

F8552 laminates

JAP Cunha1,2, TC Silva3,4 , ML Costa5,6 and MC Rezende2,4

Abstract
Tensile tests were assessed to evaluate the ageing effects on carbon fiber/epoxy resin laminates exposed to ozone and
water immersion. Laminates with distinct fiber arrangements (unidirectional - UD and plain weave - PW) were tested
under transversal and longitudinal tensile strength at room (22
C) and high (88
C) temperatures after conditioning. The
transversal tensile strength tests of UD laminates performed at room temperature show that the water immersion
decreased this property in 28% when compared to non-conditioned samples. Ozone atmosphere also deteriorates the
tensile strength considerably, drop of 27%. It was expected a higher drop in water immersion conditioning than in ozone
atmosphere due to this conditioning attacks preferentially the external surface of specimen contrariwise the water
immersion has diffusion inside the epoxy matrix. Tensile strength at high temperature shows more accentuated degra-
dation for UD specimens exposed to water immersion (50% drop) than ozone (25%). Higher temperatures favor the
degradation of the polymeric matrix and fibers/matrix interface. Tensile strength of PW laminates non-conditioned and
conditioned with ozone, tested at room temperature, presented tensile strength results slightly different, 17% lower.
When submitted to water immersion the results have dropped 26%, which are similar to unidirectional composite. At
high temperature, PW laminate attacked with ozone had tensile strength reduced in 7%, while after water immersion
the drop achieved 18%. These results show that water immersion affected the composite much more than ozone.
Fractographic analyses confirmed that the combination between high temperature and conditioning promoted the
degradation of matrix and fiber/matrix interface.

Keywords
Carbon/epoxy composite, ageing, ozone conditioning, water immersion conditioning, fractography

Introduction
1
Structural materials can be divided into four basic cat- Serviço Nacional de Aprendizagem Industrial, Taubat
e, S~ao Paulo, Brazil
2
Instituto Tecnol
ogico de Aeronáutica, S~ao Jos
e dos Campos, S~ao Paulo,
egories: metals, polymers, ceramics and composites.
Brazil
Composites consist of two or more distinct materials 3
Centro Federal de Educaç~ao Tecnol
ogica Celso Suckow da Fonseca,
combined in a structural unit and they are typically Angra dos Reis, Rio de Janeiro, Brazil
4
composed by various combinations of the other three Instituto de Ciência e Tecnologia, Universidade Federal de S~ao Paulo,
materials. Composites are generally used because they S~ao Jos
e dos Campos, S~ao Paulo, Brazil
5
Departamento de Materiais e Tecnologia, Universidade Estadual Paulista
have desirable properties that cannot be achieved by
J

ulio de Mesquita Filho, Guaratinguetá, S~ao Paulo, Brazil
any of the constituent materials acting alone,1 in 6
Lightweight Structures Laboratory, LEL/IPT, S~ao Jos
e dos Campos, S~ao
other words, one material seeks to reinforce the Paulo, Brazil
other, resulting in a new material. And the reinforce-
ment is usually carried out in the form of fibers (orien- Corresponding author:
TC Silva, CEFET/RJ, R. do Areal, 522, Parque Mambucaba, Angra dos
tated in different directions), because most materials
Reis – RJ, CEP 23953-030, Brazil.
are much stronger in this shape.2 Email: [email protected]
146 Journal of Composite Materials 55(1)

The range of applications for composite materials is composites processed with carbon fibers and thermoset
very wide. They can be used in electrical area as insu- epoxy resin F8552 were submitted to ozone and water
lating material, in wind blades, civil construction in immersion. Specimens with distinct fiber configura-
chimneys and windows, road transport in chassis and tions, unidirectional and plain weave arrangements,
suspension springs, for example, rail transport such as were tested under transversal and longitudinal tensile
wagons and interior panels, maritime transport in strength at room (22
C) and high temperatures (88
C)
patrol boats and racing boats; cable transport such as after conditioning. Fractographic analyses supported
cable cars, air transport in helicopter blades, aircraft by scanning electron microscopy (SEM) were per-
brake discs, and other components; in space vehicles as formed after tensile strength tests in order to comple-
protection for re-entry into the atmosphere and rock- ment the test results.
ets, in mechanical general applications like gears and
bearings, in sports like tennis rackets and surfboards,
among other applications.3 Experimental
The reason for such versatility and range of appli-
cations of composite materials is due in most cases to Materials
the structural performance with low weight when com-
pared to metallic materials. This characteristic attracts This work used two types of prepregs, both impregnat-
mainly automotive and aerospace industries, where ed with epoxy resin code F8552. The first prepreg, code
high mechanical resistance associated with low weight T-T145-12”-F8552, was reinforced with unidirectional
is a constant challenge.4–8 carbon fibers. The second type, code FT145-12”-
Composites in their applications suffer the effects of F8552, was reinforced with plain weave carbon
temperature, humidity, ultraviolet radiation, chemical fabric, both supplied by Hexcel Composites. The rein-
environments, including oxidizing atmospheres, such forcement used in the prepregs was high resistance
as ozone.9–11 After the first oil supply crisis in 1973, carbon fiber, IM7, with Young modulus of 276 GPa.
in order to save fuel consumption, airliners fly regularly All laminates were cured in an autoclave according to
at higher altitudes, up to 45,000 feet, where the ozone Hexcel Composites instructions, as follow: heating up
concentration is high. Ozone can enter the aircraft to 116
C at a heating rate of 2.5
C/min; isotherm at
cabin together with the fresh-air supply,12 being harm- 116
C for 60 min; heating from 116
C to 177
C at a
ful to humans and some materials.13 In this case, air- heating rate of 2.5
C/min, isotherm at 177
C for
craft are under ozone atmosphere from outside and 120 min and cooling to 65
C at a cooling rate of
inside. Ozone can decrease mechanical properties of 2.5
C/min. The application of a pressure of 0.69 MPa
epoxy resin,9 which is commonly used as matrix in was carried out at 116
C.
the manufacture of structural composites. The samples were divided in three families with six
Mechanical properties of thermosetting polymer specimens each: the first one was kept at room temper-
composite, similar to any other material, depend on ature, called as dry or non-conditioned. The second
various parameters. Environmental influences such as family was conditioned in an ozone chamber and the
moisture absorption are the most crucial, because they
third exposed to water immersion until humidity
can lead to a plasticization of the matrix material.
saturation.
Fundamental investigations have been done on the
number of factors influencing the moisture uptake of
fiber reinforced composites.14,15 The authors in previ- Conditioning
ous works16,17 evaluated these same laminates through
Prior to any conditioning, all specimens were dried in
tensile and compression tests after exposure to other
an oven (60  2
C) for four hours. For this monitoring,
different ageing effects, i.e., salt spray and hygrother-
mal chamber, and show the deleterious effects of these four traveler-samples (samples without tabs of each
ageing on the studied laminates. On the other hand, family of laminates, with nominal dimensions of
although these materials have been in use for many 50 mm  50 mm) were quickly removed from the oven
years, only limited data are available concerning the to a desiccator until they reached room temperature
influence of the ozone atmosphere on epoxy composite and weighed in an analytical balance (0.1 mg of preci-
reinforced carbon fiber. sion) while the other specimens to be mechanically
Given this scenario and the few studies found in the tested remained in the oven. This procedure was
literature regarding the effect of ozone on the mechan- based on ASTM C562-1018 and was repeated until
ical behavior of thermoset polymeric composites,9–13 the dry weight stabilizes with a precision of 0.002 g.
this study was proposed. In this work, tests were Then, samples were transferred to their respective
assessed to evaluate the ageing effects when laminated conditioning.
Cunha et al. 147

Ozone chamber the tab over the specimen. After the adhesive film cure,
the specimens (transversal and longitudinal) were cut
Exposure of samples to ozone (O3) was performed
attending the dimensions used in the tensile tests. The
based on ASTM D1149-07 standard.19 This test
use of tabs guarantees a homogeneous distribution of
makes it possible to assess the material’s resistance
against weather conditions in the ozone environment. the gripping forces in the specimen fixation on the test
Basically, the equipment consists in a tube (dielectric) machine and protects the laminate surface from
which flows oxygen, where a constant electrical dis- damage, due to its low Young modulus, which deforms
charge (corona effect), generated through a transform- elastically distributing stresses over a large area and
er, transforms oxygen molecule (O2) adding an oxygen absorbing energy during the test.
atom. Then the ozone gas is sucked by a Venturi system Table 1 summarizes all tests performed with a total
and mixed with water to be treated in a mixer that feed of 108 specimens.
the ozone chamber. Samples were exposed to ozone Longitudinal and transversal tensile strengths were
environment, with 50 ppm concentration, for 72 h at calculated according to equation (1):
(40  1)
C, according standard recommendation.
Fmax
These parameters were programmed on a Gramma rmax ¼ (1)
ozone chamber, as per Figure 1(a). A

Water immersion where: r is the maximum tensile strength applied to the

specimen, in MPa, F is the breaking load, in N, and A
Water immersion conditioning was based on ASTM is the section area, in mm2.
D570-10 standard20 to evaluate sample resistances at
extreme humidity condition. Samples were fully
Scanning electron microscopy
immersed in distilled water at room temperature
(23  2)
C on a Becker properly sealed to prevent SEM was employed to analyze the cross-section view of
evaporation, as shown in Figure 1(b). The mass gain the fractured laminates, i.e., the fracture surfaces.
was monitored by weekly weighing of the traveler- The fracture surfaces of the specimens after the
samples until the wet weight to stabilize with a preci- mechanical tests were analyzed as well by LEO 435
sion of 0.002 g. Hence, for this conditioning 4 weeks Vpi microscopy. The samples were previously coated
were needed to reach saturation. with 5 nm Au layer to avoid charging during the elec-
tron irradiation.
Tensile strength tests
Specimens with the same epoxy matrix (F8552) and Results and discussions
two distinct fiber configurations, unidirectional and To assess the influence of environmental conditioning
plain weave, were prepared and tested according to
under water immersion, the variation in the uptake of
ASTM 3039-08.21 Transversal and longitudinal tensile
moisture by the laminates as a function of exposure
strength tests were performed at room (22
C) and high
time was monitored by weekly measurement of mass
temperatures (88
C), before and after conditioning, in
gain, using traveler-samples totally immersed in water,
a machine Instron mode 1332 at speed test of 1.27 mm/
until moisture equilibrium achieved, according to the
min. Transversal tensile specimens had dimensions of
Fick
s diffusion theory.22 The mass gain was measured
247.2 mm  25.4 mm  1.6 mm and the longitudinal
ones 247.2 mm  12.7 mm  1.6 mm (Figures 2 and 3). in terms of percentage in relation to the dry mass of the
Tabs were obtained from the epoxy/fiber glass lam- laminate for approximately 8 weeks, when the wet
inate with dimensions of 300 mm  300 mm  2.9 mm. weight stabilized with a accuracy of 0.002 g. The
From this laminate were cut pieces with dimensions of unidirectional samples presented an average mass
300 mm  60 mm  2.9 mm with one side chamfered gain of (0.60  0.06)% and the plain weave samples
with 16 degrees. These pieces were glued to the ends an average mass gain of (0.65  0.06)%. The standard
of the carbon fiber laminates with an adhesive film to deviation was less than 12% for all measurements. The
reduce stress concentration in the glued interface. greatest gain observed for the plain weave laminate is
In order to glue the epoxy/fiber glass laminate on the expected, considering that this family of composite
carbon fiber laminates, the contact areas of both lam- has richer regions in resin favored by the warp and
inates were sanded aiming to increase roughness. weft arrangement in the fabric. These results are in
Subsequently, the bonding was performed using 3 M agreement with the literature that shows the mass
AF191 adhesive and to ensure a good bonding, a gain varying for carbon/epoxy laminates between
press was used to apply the pressure and consolidate 0.63 – 0.68%.23,24
148 Journal of Composite Materials 55(1)

Figure 1. Gramma ozone chamber (a) and specimen being conditioned in water immersion (b).

Figure 2. Tensile strength specimens dimensions in millimeters: (a) transversal and (b) longitudinal.

Figure 3. Specimens for tensile strength tests: (a) transversal and (b) longitudinal.

These differences may occur because the mass gain parameters used in the conditionings, and type of
depends on several parameters, such as type of poly- sample used in the monitoring of mass gain (traveler-
meric matrix and reinforcement, fiber arrangement sample or the specimens themselves to be used in
style, stacking sequence, finishing of specimen edges, mechanical tests).22,25
Cunha et al. 149

Table 1. Tensile strength sample quantities.

Longitudinal tensile strength Transversal tensile strength

Unidirectional Plain weave Unidirectional

Conditioning Troom Thigh Troom Thigh Troom Thigh

Non-conditioned 6 6 6 6 6 6
Ozone chamber 6 6 6 6 6 6
Water immersion 6 6 6 6 6 6

Table 2. Tensile strength results.

Longitudinal tensile strength (MPa) Transversal tensile strength (MPa)

Unidirectional Plain weave Unidirectional

Conditioning Troom Thigh Troom Thigh Troom Thigh

Non-conditioned 2627  117 2532  40 992.1  15.7 856.5  31.7 61.4  1.9 47.6  3.6
Ozone chamber 1838  67 1812  143 819.7  70.9 799.1  33.1 44.9  2.3 35.4  6.9
Water immersion 2079  39 2012  43 732.8  58.4 706.6  21.3 44.3  5.8 22.6  3.8

Tensile tests considered that it does not occur significant change in

the value of longitudinal tensile strength for these com-
Mean tensile test results for all specimens are shown in
posite families.
Table 2, comparing non-conditioned against ozone and
Transversal tensile strength values for the unidirec-
water immersion conditionings. Standard deviation
tional specimens (Figure 4(b)) show that the strength at
was calculated to characterize data dispersion. This is
room temperature for this laminate family is mainly
fundamental to point the manufacture is consistent affected by water immersion, with a drop of 28%,
within the results. In order to give greater reliability when compared to non-conditioned samples. Ozone
to the correlation of the results, a statistical study atmosphere also deteriorated the tensile strength of
was carried out, through the measurement hypothesis unidirectional laminates considerably, with a drop of
test by the one-way ANOVA method, which tests the 27%. Comparing statistically these results it is evi-
hypothesis that the means of two or more populations denced the results are equivalents. It was expected a
are the same or different, with a level of 95% of higher drop at water immersion conditioning than at
significance. ozone atmosphere due ozone attack occurs preferen-
For longitudinal tensile strength on unidirectional tially on external surface of laminate, contrariwise the
laminates (Figure 4(a)) is expected the load be sup- water immersion whose diffusion penetrates inside the
ported by the fibers as the load is located on the bulk epoxy matrix. When tensile strength is assessed at
same direction. Water immersion conditioning caused high temperature the degradation behavior is more
a drop in the tensile strength of 21%, at room temper- accentuated, with 50% drop for the specimens condi-
ature. An interesting fact to observe is that ozone con- tioned in water immersion and a drop of 25% for speci-
ditioning was considerably affected with a drop of mens exposed to ozone atmosphere. At higher
30%. Despite the ozone attack occurring from the temperatures the degradation is more pronounced sug-
external surface to the inside of the material, the defects gesting the degradation of polymeric matrix and/or
induced on the surface ended up degrading the resis- fiber/matrix interface.
tance of the whole composite severely. At high temper- Tensile strength on plain weave laminates (Figure 5)
ature, the decreases were 20 and 28%, for the was not necessary perform as both transversal and lon-
specimens conditioned under water immersion and gitudinal configurations have fibers at 0
and 90
, and
ozone, respectively. In these cases, when applying the the strength results shall be almost the same. When
means hypothesis test by the One Way ANOVA comparing non-conditioned and conditioned under
method, it is proved that the mean values of resistance ozone atmosphere, the samples at room temperature
are not different, with a reliability of 95%, so it can be presented tensile strength results slightly different,
150 Journal of Composite Materials 55(1)

Figure 4. Longitudinal (a) and transversal (b) tensile strength for unidirectional laminates at room and high temperatures.

chamber were similar as humidity-controlled chamber

and water immersion.
Other works had also experienced a drop on tensile
strength when carbon epoxy composites were exposed
to water immersion. Murthy and collaborators23
observed that degradation increases substantially with
increase in water immersion time, in addition moisture
penetration along the fiber/matrix interfaces caused
interfacial debond, which led to rupture or degradation
of the interface. The water uptake weakened the fiber/
matrix interface exposing the fibers and as a final con-
sequence tensile strength dropped. Ramirez and collab-
orators26 noted large reductions in transverse strength
of the composites after water immersion conditioning
and concluded that matrix and fiber matrix interface
Figure 5. Longitudinal tensile strength of plain weave laminates were degraded. Mansouri and collaborators24 experi-
at room and high temperatures.
enced temperature aging as water absorption is an
important parameter in the degradation of composites,
which influences the elastic properties of the composite
17% lower, although when submitted to water immer-
laminates, and consequently the stiffness of the whole
sion the results have dropped 26%, which is similar to
composite structure. Humeau and collaborators27
unidirectional composite. At high temperature, plain
tested water immersion combined with mechanical
weave laminate attacked with ozone had tensile
loading and this coupling caused large effects on tensile
strength reduced in 7%, while after water immersion
strength which led to premature failure.
the drop achieved 18%.
Comparing our previous work16 with humidity-
controlled and salt spray chamber conditioning against Fractographic study
the studied conditioning in this work, water immersion Figure 6 presents a representative image of the speci-
and ozone chamber, it is possible to conclude when the mens of unidirectional composite tested in longitudinal
load is aligned with fibers at room temperature (longi- and transversal tensile strength. It can be seen in
tudinal tensile test for unidirectional and plain weave) Figure 6(a) that the longitudinal specimens presented
results do not differ considerably. However, when a rupture that is difficult to analyze from a fracto-
transversal tensile is applied for unidirectional compos- graphic point of view, since they fragmented, prevent-
ite and the matrix is responsible to support the load, ing a detailed morphological investigation of the
tensile strengths achieved for salt spray and ozone fractured region from being carried out.
Cunha et al. 151

Figure 6. Representative images of the unidirectional specimens tested in longitudinal (a) and transversal (b) tensile strength.

Figure 7. SEM of unidirectional specimen, non-conditioned, after transversal tensile strength at room temperature (a) and detail of
cusps in the tested specimen (b).

Fractographic analysis is hampered by the fact that the the specimens present a clear transverse rupture in the
fracture surfaces are very damaged, due to the impact load direction, which allows a detailed morphological
of fragments generated in the test with other parts of investigation of the fractured region.
the specimen itself. In addition, the cutting of these Comparing the fracture surfaces of the unidirection-
samples, for observation in a scanning electron micro- al specimens without conditioning, tested at room and
scope is jeopardized by the fragility of the composite elevated temperatures, obtained by scanning electron
and the condition they are, in the form of sticks. This microscopy, it is observed that the samples evaluated
type of occurrence, loss of the fracture surface due to at room temperature, Figure 7, shows the presence of
impacts or contamination, is common in real cases and resin adhered to the surface of the fibers, with a well-
prevents the analysis of failures of many fractured parts defined anchoring region of the matrix on the fiber. The
in service or during aircraft maintenance.28 arrows indicate pointed regions of resin located that
Figure 6(b) presents a representative picture of the strongly interconnect one fiber to another, an aspect
specimens characterized in transversal tensile strength fractographic called cusps.29–36 This aspect of fracture
test for unidirectional composites. It is observed that shows that there is no displacement of the fibers and
152 Journal of Composite Materials 55(1)

Figure 8. SEM of unidirectional composite, non-conditioned, after transversal tensile strength at high temperature (a) and details of
the marked area (b).

Figure 9. Unidirectional composite exposed to ozone conditioning and tested in transversal strength. SEM of: center region of
fracture surface tested at room temperature (a), external region of specimen tested at high temperature (b), and detail of external
region of specimen tested at high temperature (c).

reveals that the rupture of the samples during the test shows debris arising from polymeric matrix degrada-
occurs preferably in the region with resin. tion. This degradation is in agreement with Table 2
Composites tested at high temperature (Figure 8) results, which shows a 27% drop on this property.
present a thin layer of resin covering the surface of On Figure 9(b), related to high temperature trans-
the fibers, but the regions rich in resin are mostly versal tensile strength test after ozone chamber condi-
broken, leaving resin adhered to the fiber. Such obser- tioning, the upper region on the image, which is the
vation is attributed to the matrix degradation, caused external surface of specimen, reveals a rougher resin
by deleterious temperature effect. surface with exposed fiber due polymeric matrix loss,
Figure 9 shows representative SEM images of the while at the central region, where the fracture occurred,
central region of the fractured surface of the unidirec- the texture is more intact, indicating that the attack
tional laminate conditioned in ozone chamber and occurs from the outside to the inside in the specimen.
tested in transversal tensile strength at room tempera- The correlation of these observations with the property
ture. When this image is compared to the one in of transversal tensile strength shows that this property
Figure 8, non-conditioned specimen and submitted to was impaired, with a decrease of around 26%, as
the same test, it is possible to verify that the ozone shown in Figure 4(b). Figure 9(c) presents in more
chamber damaged the material integrity, as per detail outer surface texture from specimen submitted
smooth fiber surface and interfacial detachment in to ozone chamber, revealing the severe attack of this
this region during the mechanical test. Figure 9(a) conditioning on the resin region.
Cunha et al. 153

Figure 10. Representative images of plain weave based specimen tested in longitudinal tensile strength.

Figure 11. SEM of plain weave specimen, non-conditioned, tested at room temperature (a); at high temperature (b); and detail of
fracture surface obtained in high temperature tests (c).

Figure 10 presents a representative picture of com- Comparing the fracture surfaces of the specimens
posites reinforced with carbon fabric tested in tensile. It reinforced with plain weave fabric without condition-
can be observed a clear transverse rupture in the load ing, tested at room and high temperatures (Figure 11),
direction, which allows a detailed morphological inves- it is observed that the sample’s fracture surface tested
tigation of the fractured region. at room temperature (Figure 11(a)) shows the presence
154 Journal of Composite Materials 55(1)

Figure 12. SEM of plain weave specimen exposed to water immersion and tested at room temperature (a) and to ozone chamber
and tested at high temperature (b).

of resin adhered to the surface of fibers (upper right specimen, resulting from severe attack samples suffered
region, with transversal fibers), presenting a well- during conditioning in ozone environment, as also
defined anchoring region of the matrix on the fiber, observed for the samples with unidirectional compo-
with aspects of cusps. This figure also shows the pullout sites (Figure 4). This behavior was observed for the
of fibers37 in the region that shows the top fibers samples tested at room and high temperatures.
(median region on the left). Finally, the region where
the weft crosses with the warp reveals a region rich in
Conclusions
resin, which broke during the mechanical test. In this
region it is also evidenced the good anchoring of the The tensile tests performed in transverse and longitu-
resin on the fiber (bottom of the image). dinal directions of F8552-epoxy resin composites rein-
Figure 11(b) refers to laminates reinforced with forced with carbon fibers, unidirectional and plain
plain weave fabric, without environmental condition- weave fabric, are mainly affected by exposure in a
ing and tested at high temperature. These figures show humid environment, such as the direct immersion in
fiber pullout presence and more extensive discontinu- water. Ozone conditioning is also detrimental to the
ities in the regions where the weft crosses with the mechanical behavior of the material, but in lesser
warp. Regarding the fiber/resin interface, there are no extent, due to the fact that ozone attack occurs on
significant differences. Figure 11(c) shows the fracture the external surface of the material. The results
surface in greater detail with emphasis on the pullout obtained from fractographic analyses show the impor-
aspect of the fibers and the presence of residues, prob- tance of this tool in the characterization of materials.
ably arising from the degradation of the epoxy resin Fractographic analyses complement the composite
matrix. mechanical test results and provide subsidies that
Figure 12(a) shows a typical view of fracture surface could drive the processing optimization of the materials
of specimens exposed to water immersion conditioning, and its applications. On the specimens tested at room
where more pronounced discontinuities presence is temperature, fractographic analyses showed good
observed (white arrow) in the regions with the highest adhesion between fibers and matrix as designed,
resin concentration, which usually occur at the inter- which can be evidenced by a well-defined anchoring
section of weave with warp. These discontinuities prob- region of the matrix on the fiber. In contrast, specimens
ably justify the greater drop in tensile strength of this tested at high temperature show regions rich in resin
laminate family. are mostly broken and even with fiber pullout, which
With respect to samples subjected to ozone condi- led us to conclude the temperature effect by itself jeop-
tioning, the analysis of the fracture surface (internal ardize mechanical properties as could be confirmed by
part of the specimen) does not show significant changes tensile test results. Specimens conditioned with ozone
in the texture of the epoxy resin matrix or in the resin- did not demonstrate significant changes when analyzed
fiber interface region (Figure 12), as already noted in with SEM when compared to specimens conditioned
unidirectional laminates. These aspects suggest that the with water immersion, those presented more pro-
drop in tensile strength is due to the influence of nounced discontinuities. Regarding the mechanical
the damage occurred on the external surface of the behavior of unidirectional composites, it is observed
Cunha et al. 155

that transversal tensile strength property is strongly 5. Chawla KK. Composite materials: science and engineer-
influenced by water immersion conditioning, with ing. 4th ed. Berlin: Springer International Publishing,
drops of this property at room temperature (22
C) of 2019.
6. Campbell FC. Structural composite materials. Cleveland:
approximately 28%. This decrease was accentuated in
ASM international, 2010.
tests at high temperature (82
C), about 50%. These
7. Campbell FC. Manufacturing technology for aerospace
behaviors are attributed to the deleterious effects of structural materials. Amsterdam: Elsevier, 2006.
water absorption and test temperature on the polymer- 8. Vasiliev VV and Evgeny VM. Advanced mechanics of
ic matrix and the reinforcement/matrix interface. composite materials and structural elements. 4th ed.
Although the longitudinal tensile strength is dominated London: Newnes, Elsevier, 2018.
by the reinforcement, the results obtained in this eval- 9. Middleton J, Burks B, Wells T, et al. The effect of ozone
uation show that this property is also affected by the and high temperature on polymer degradation in polymer
conditioning. In the case of water immersion and ozone core composite conductors. Polym Degrad Stabil 2013;
98: 2282–2290.
exposure, there are drops of 21 and 30%, respectively,
10. Marenco A, Thouret V, N
ed
elec P, et al. Measurement of
in tests carried out at 22
C. Plain weave composites ozone and water vapor by Airbus in-service aircraft: the
when tested over tensile strength at room temperature MOZAIC airborne program, an overview. J Geophys Res
presented a decrease, the most significant being on the 1998; 103: 25631–25642.
samples conditioned with water immersion (26%). 11. Sofia TF, Zarouchas D and Poulis JA. The use of acous-
Tests carried out at 82
C also show the deleterious tic emission and composite peel tests to detect weak adhe-
effect of temperature on the mechanical behavior of sion in composite structures. J Adhes 2018; 94: 743–766.
the studied materials. 12. Lu Y, Zhao X, Wang M, et al. Feasibility analysis on
photocatalytic removal of gaseous ozone in aircraft
cabins. Build Environ 2014; 81: 42–50.
Declaration of Conflicting Interests
13. Rai AC and Chen Q. Simulations of ozone distributions
The author(s) declared no potential conflicts of interest with in an aircraft cabin using computational fluid dynamics.
respect to the research, authorship, and/or publication of this Atmos Environ 2012; 54: 348–357.
article. 14. Berges M, L
eger R, Placet V, et al. Influence of moisture
uptake on the static, cyclic and dynamic behaviour of
Funding unidirectional flax fibre-reinforced epoxy laminates.
Compos Part A: Appl Sci Manuf 2016; 88: 165–177.
The author(s) disclosed receipt of the following financial sup- 15. Wolfrum J, Eibl S and Lietch L. Rapid evaluation of
port for the research, authorship, and/or publication of this long-term thermal degradation of carbon fibre epoxy
article: The authors acknowledge the financial support composites. Compos Sci Technol 2009; 69: 523–530.
received from the Brazilian Funding Institutions FAPESP 16. Cunha JAP, Costa ML and Rezende MC. Influência de
(Fundaç~ao de Amparo à Pesquisa do Estado de S~ao Paulo) diferentes condiç~
oes higrot
ermicas na resistência à traç~ao
(2018/09531-2), CNPq (Conselho Nacional de de comp
ositos de fibra de carbono/ep
oxi modificada.
Desenvolvimento Cient
ıfico e Tecnol
ogico) (305123/2018-1 Pol
ımeros 2006; 16: 193–201.
17. Cunha JAP, Costa ML and Rezende MC. Study of the
and 305492/2017-9) and Coordenaç~ao de Aperfeiçoamento
hygrothermal effects on the compression strength of
de Pessoal de N
ıvel – Brasil (CAPES – Superior carbon tape/epoxy composites. Lat Am J Solids Struct
Coordination for the Improvement of Higher Education 2008; 5: 157–170.
Personnel) – Finance Code 001. 18. ASTM C562-10 Standard test method for moisture in a
graphite sample, ASTM International, Philadelphia, PA,
ORCID iD 2000.
TC Silva https://orcid.org/0000-0002-3695-8515 19. ASTM D1149-07 Standard test method for rubber
deterioration – cracking in an ozone controlled environ-
ment, ASTM International, West Conshohocken, PA,
References 2007.
1. Gibson RF. Principles of composite material mechanics. 20. ASTM D570-05 Standard test method for water absorp-
Boca Raton: CRC Press, 2016. tion of plastics moisture, ASTM International, West
2. Barbero EJ. Introduction to composite materials design. Conshohocken, PA, 2005.
Boca Raton: CRC Press, 2018. 21. ASTM D3039/D3039M-08 Standard test method for ten-
3. Gay D. Composite materials: design and applications. sile properties of polymer matrix composite materials,
Boca Raton: CRC Press, Taylor & Francis, 2015. ASTM International, West Conshohocken, PA, 2008.
4. Zhang J, Chaisombat K, He S, et al. Hybrid composite 22. Carlsson LA and Du E. Water uptake in polymer compo-
laminates reinforced with glass/carbon woven fabrics for sites with voids. Cham: Springer, 2018, pp. 33–57.
lightweight load bearing structures. Materials & Design 23. Murthy HN, Sreejith M, Krishna M, et al. Seawater
2012; 36: 75–80. durability of epoxy/vinyl ester reinforced with
156 Journal of Composite Materials 55(1)

glass/carbon composites. J Reinf Plast Compos 2010; 29: 31. Barbosa LCM, Souza S, Botelho EC, et al. Fractographic
1491–1499. evaluation of welded joints of PPS/glass fiber thermoplas-
24. Mansouri L, Djebbar A, Khatir S, et al. Effect of hygro- tic composites. Eng Fail Anal 2019; 102: 60–68.
thermal aging in distilled and saline water on the mechan- 32. Opelt C, Cândido G and Rezende M. Compressive fail-
ical behaviour of mixed short fibre/woven composites. ure of fiber reinforced polymer composites – a fracto-
Compos Struct 2019; 207: 816–825. graphic study of the compression failure modes. Mater
25. Cândido GM, Rezende MC and Almeida S. Today Commun 2018; 15: 218–227.
Hygrothermal effects on the tensile strength of carbon/ 33. Opelt CV, Cândido GM and Rezende MC.
epoxy laminates with molded edges. Mat Res 2000; 3: Fractographic study of damage mechanisms in fiber rein-
11–17. forced polymer composites submitted to uniaxial com-
26. Ramirez FA, Carlsson LA and Acha BA. Evaluation of pression. Eng Fail Anal 2018; 92: 520–527.
water degradation of vinylester and epoxy matrix com- 34. Reis VL, Opelt CV, Cândido GM, et al. Effect of fiber
posites by single fiber and composite tests. J Mater Sci orientation on the compressive response of plain weave
2008; 43: 5230–5242. carbon fiber/epoxy composites submitted to high strain
27. Humeau C, Davies P and Jacquemin F. An experimental rates. Compos Struct 2018; 203: 952–959.
study of water diffusion in carbon/epoxy composites 35. Opelt CV, Paiva JM, Cândido GM, et al. A fractographic
under static tensile stress. Compos Part A Appl Sci study on the effects of hygrothermal conditioning on
Manuf 2018; 107: 94–104. carbon fiber/epoxy laminates submitted to axial compres-
28. Kar RJ. Composite failure analysis handbook. vol. 2. sion. Eng Fail Anal 2017; 79: 342–350.
Technical handbook/part 2. Atlas of fractography. 36. Silveira NNA, Sales RCM, Cândido GM, et al.
Hawthorne: Northrop Corp Aircraft Div, 1992. Fractographic analysis of the hygrothermal effect in co-
29. Zimmermann N and Peng HW. A review of failure bonded and secondary bonded joints under mode II
modes and fracture analysis of aircraft composite materi- delamination loading. Int J Adhes Adhes 2020; 98: 102508.
als. Eng Fail Anal 2020; 115: 104692. 37. Nakamura T, Singh RP and Vaddadi P. Effects of envi-
30. Greenhalgh E. Failure analysis and fractography of poly- ronmental degradation on flexural failure strength of
mer composites. Amsterdam: Elsevier, 2009. fiber reinforced composites. Exp Mech 2006; 46: 257–268.