SLE MECHANICS Basic Concepts and Engineering ApplicationsSoil Mechanics: Basic Concepts and Engineering Applications A. AYSEN The University of Southern Queensland, Australia SSO A.A. BALKEMA PUBLISHERS / LISSE / ABINGDON / EXTON (PA)/ TOKYO. FER G74Library of Congress Cataloging-in-Publication Data Applied for: $ April 2002 oust {$ Unter = ‘tice’ Ney of Cover design: Studio Jan de Boer, Amsterdam, The Netherlands. Printed by: Gorter, Steenwijk, The Netherlands. Copyright © 2002 Swets & Zeitlinger B. Lisse, The Netherlands All rights reserved. No part of this publication or the information contained herein ‘may be reproduced, stored in a retrieval system, or transmitted in any form or by ‘any means, electronic, mechanical, by photocopying, recording or otherwise, ‘without written prior permission from the publishers. Although all care is taken to ensure the integrity and quality of his publication and the information herein, no responsibility is assumed by the publishers nor the author for any damage to property or persons as a result of operation or use of this ‘publication andor the information contained herein. Published by: A.A. Balkema Publishers, a member of Swets & Zeitlinger Publishers ‘www.balkema.nl and www.szp.swets.nl ISBN 90 5809 358 1Contents Preface ix CHAPTER1 Nature of Soils, Plasticity and Compaction 1 Li INTRODUCTION 1 12. NATURE AND CHEMISTRY OF SOILS 1 13 MASS-VOLUME RELATIONSHIPS 2 14 PARTICLE SIZE DISTRIBUTION 2 15 INDEX PROPERTIES AND VOLUME CHANGE IN FINE GRAINED SOILS 26 16 SOIL CLASSIFICATION FOR GEOTECHNICAL PURPOSES 3 17 COMPACTION 4 18 PROBLEMS 51 19. REFERENCES 3 CHAPTER2 Effective Stress and Pore Pressure in Saturated Soils 55 INTRODUCTION 35 ‘STATE OF STRESS AT A POINT DUE TO SELF-WEIGHT 55 STATE OF STRESS AT A POINT DUE TO EXTERNAL FORCES 6 PROBLEMS or REFERENCES 68 CHAPTER3 The Movement of Water through Soil cy 3.1 INTRODUCTION 3.2. PRINCIPLES OF FLOW IN POROUS MEDIA o 33 PERMEABILITY n 34 FLOWNETS: 85 35 MATHEMATICS OF THE FLOW IN SOIL 92 36 SEEPAGE THROUGH EARTH DAMS 100 37 PROBLEMS 107 3.8 REFERENCES 10vi Contents CHAPTER4 Shear Strength of Soils and Failure Criteria ni 41 INTRODUCTION 1 42 MOHR-COULOMB FAILURE CRITERION 1 43 LABORATORY SHEAR STRENGTH TESTS us 44 STRESS-STRAIN BEHAVIOUR OF SANDS AND CLAYS 132 45. CRITICAL STATE THEORY 137 4.6 PROBLEMS 150 4.7 REFERENCES 152 CHAPTERS Stress Distribution and Settlement in Soils 153 5.1 INTRODUCTION 153 5.2, FUNDAMENTAL EQUATIONS OF ELASTICITY 154 5.3. STRESS DISTRIBUTION DUE TO EXTERNAL AND INTERNAL LOADING 157 5.4 ELASTIC SETTLEMENT OF FOOTINGS isa 5.5 SOIL-FOOTING INTERACTION MODELS 19 5.6 PROBLEMS. 206 5.7 REFERENCES 210 CHAPTER6 One Dimensional Consolidation 213 6.1 INTRODUCTION 213 62 CONSOLIDATION INDICES AND SETTLEMENT PREDICTION 23 63. SOLUTION OF ONE DIMENSIONAL CONSOLIDATION DIFFERENTIAL EQUATION 226 64 APPLICATION OF PARABOLIC ISOCHRONES 243 6.5. LIMITATIONS OF ONE DIMENSIONAL CONSOLIDATION THEORY 251 66 PROBLEMS 256 6.7 REFERENCES 258 CHAPTER? Application of Limit Analysis to Stability Problems in Soil Mechanics 261 7.41 INTRODUCTION 261 7.2. LOWER BOUND SOLUTION 263 7.3. UPPER BOUND SOLUTION 280 7.4 FINITE ELEMENT FORMULATION OF THE BOUND THEOREMS 291 7.3 LIMIT EQUILIBRIUM METHOD AND CONCLUDING REMARKS 298 7.6 PROBLEMS 299 7.7 REFERENCES BotContents vii CHAPTERS — Lateral Earth Pressure and Retaining Walls 303 8.1 INTRODUCTION 303 82 EARTH PRESSURE AT-REST 303 8.3 RANKINE'S THEORY FOR ACTIVE AND PASSIVE SOIL PRESSURES 305 84 COULOMB WEDGE ANALYSIS 34 8S COMMON TYPES OF RETAINING STRUCTURES AND FACTOR OF SAFETY 323 86 STATIC ANALYSIS OF CANTILEVER AND GRAVITY RETAINING WALLS 326 87. STATIC ANALYSIS OF SHEET PILE WALLS 331 88 INTERNALLY STABILIZED EARTH RETAINING WALL 354 89. THE OVERALL STABILITY OF RETAINING STRUCTURES 358 5.10 PROBLEMS 361 8.11 REFERENCE 308 cHarreR9 Stability of Earth Slopes 367 9.1 INTRODUCTION 361 92 STABILITY OF SLOPES IN cy Oy~0 SOIL - CIRCULAR FAILURE SURFACE. 368 93. STABILITY OF SLOPES IN c',9 SOIL - THE METHOD OF SLICES 372 9.4 STABILITY OF INFINITELY LONG FARTH SLOPE: 381 955. STABILITY OF REINFORCED AND NAILED EARTH SLOPES 385 9. GENERAL SLOPE STABILITY ANALYSIS 393 9.7 APPLICATION OF THE WEDGE METHOD TO UNREINFORCED SLOPES 399 9.8 CONCLUDING REMARKS 403 99. PROBLEMS 408 9.10 REFERENCES 408 CHAPTER 10 Bearing Capacity of Shallow Foundations and Piles 411 10.1 INTRODUCTION aur 10.2 ULTIMATE BEARING CAPACITY OF SHALLOW FOUNDATIONS 4 10.3 FIELD TESTS 5 104 AXIAL ULTIMATE BEARING CAPACITY OF PILES 433 105 PILE. GROUPS 446, 10.6 PROBLEMS 450 10.7 REFERENCES 431 Index 455Preface Soil Mechanics: Basic Concepts and Engineering Applications is primatily designed as 4 main text for university students taking first degree courses in civil engineering as well as environmental and agricultural engineering. Emphasis is placed on presenting fundamental behaviour before more advanced topics are introduced. The special structure of the book, embodied in each chapter, makes it possible to be used in two, three and four year undergraduate courses in soil mechanics. However, as new and advanced topics that extend ‘beyond standard undergraduate courses are included, the book will also be a valuable resource for the practicing professional engineer. A problem solving approach is adopted through all chapters and 152 worked examples demonstrate the engineering applications, simulate problem solving leaning and facilitate self-teaching. There are 113 unsolved problems with answers set for solution by students. The use of S.1, units throughout, and frequent references to current intemational codes of practice and refereed research papers, make the contents universally applicable. The main feature of the book in comparison to a traditional soil mechanics text is the inelusion of the soil reinforcement in retaining walls and slopes and tunnel stability problems. ‘This book is written for readers that have no prior knowledge in soil mechanics, however a reasonable background in engineering mechanics is essential 1 am most grateful for publishers, individuals and engineering organizations that have permitted the use oftheir materials in this book. It must be noted that all ofthe refereed figures are reconstructed and no copy has been carried out, ‘The author is grateful to the colleagues in the Faculty of Engineering and Surveying in USQ for their continuous encouragements, Special thanks are due to Dr A. Kilpatrick of La ‘Trobe University, Australia for his help and time in proofreading of the material Tam in debt to my family and my sincere thanks go to my wife Pari for her patience during this project. A. Aysen, M.Se., Ph.D. The University of Souther Queensland, Australia [email protected] February 2002CHAPTER 1 Nature of Soils, Plasticity and Compaction LI INTRODUCTION This chapter describes the basic characteristics of soil, and physical properties that affect its engineering behaviour, For a civil engineer the geological history ofa soil may nat seem to be of much importance as most of the time a civil engineer is searching for a stress-strain model justified by laboratory results, However some information is necessary in order to facilitate evelopment of the model and understand the physical and chemical behaviour of the material of interest and, consequently, a brief description of soil chemistry is included. A significant part ofthe chapter is devoted to the phase relationships that describe volume-mass related pa- rameters which control the engineering behaviour of the soil. Essemtial information about par- ticle size analysis, plasticity and soil classification is provided. More details about laboratory procedures and soil classification may be obtained from the standard codes of the relevant ‘country. The final part of the chapter discusses soil compaction and its importance to the soil 1.2. NATURE AND CHEMISTRY OF SOILS L2A Origin of soils, geological classification and primary minerals Soils are the unconsolidated layers that cover the earth's surface. For a civil engineer soil con- sists of particles of different size and shape with minor bonds between them forming a struc: ‘ure that undergoes deformation when subjected to nauural or artificial forces. For an enviroit= ‘mental engineer sol is a product of the environment, and its formation is a function of climate, ‘organisms, parent rock or material, relief and time (Jenny, 1940). Soil mechanics is the art of applying the mechanics of solids tothe soil to predict its deformation behaviour and strength, ‘This allows the design of soit structures and the investigation of its interaction with other structures built on it. Soils are derived from the weathering of parent rocks. Weathering or disintegration of rock is of two types: physical and chemical. The physical weathering isthe action of the forces as- sociated with wind, water, glaciers and successive freezing and thawing. The chemical agents are water, oxygen, and carbon dioxide. In the physical weathering the mineralogical composi- 12 Soil Mechanics: Basic Concepts and Engineering Applications tion of the parent rock does not change and the resulting soil inherits similar minerals. Inthe chemical weathering the alteration of the minerals of the parent rock creates new mineral. The process of soil formation is controlled by semi-independent factors of ime, climate, or anisms and topography. The time factor controls the equilibrium of the process with the envi ronment, a process that may take more than a million years FitzPatrick, 1983) The geotogical cassfication places rocks into three major groups, viz. igneous, metamor- phic and sedimentary rocks. Igneous rocks are those that have cooled down and crystallized from a molten material called magma. These rocks are composed of a variety of mineral erys- tals, the size of which depends on the rate of cooling. The igneous rocks within the earth's crust are formed by slow cooling and therefore have large crystals. They can be found on the surface ofthe earth ifthe overlying material is removed or eroded. Igneous rocks formed on tbe surface ofthe ensth have been subjected to rapid cooling and therefore have small crystals. Residual soils area result of the physical and chemical weathering ofthese igneous rocks and the constituent minerals are different to those of the parent rock. Sands have quartz and mus covite, similar 9 quartz and muscovite of the parent rock granite. The orthoclase feldspar and plagioclase of granite break down inco the clay minerals of Kaolinite and montmorillonite Metamorphic rocks originate from residual soils under extreme temperature and pressure. Both metamorphic rocks and residual soils are subjected to weathering and transportation processes that result ia the formation of sedimentary deposits. Compaction ofthese deposits, together with cementation due to chemical reactions, produces sedimentary rocks. The sedi- mentary teks are also produced directly from metamorphic rocks within a Weathering, trans- portation, consolidation and cementation process, These sedimentary rocks could return to their origin as either sedimentary deposits or metamorphic rocks as a result of weatbering or exeme temperature and pressure. This isan ongoing geological eyele dough time and the residual soils and sedimentary deposits so formed are known to us as soils. The soit particles axe of various sizes and are broadly categorized as clay, silt, sand, gravel and rocks. Each size fraction exhibits different mechanical properties that can be successfully predicted using the principles of soil mechanics ‘Typical depositional environments io soit formation (Fockes & Vaughan, 1986) include: alacial environment, take and marine environment and desert environment. For sols and soit mechanics the most significant recent geological events beside the depositional environments are rising and falling land and sea levels that result in further deposition and weathering. Ina geotechnical site investigation a geotechnical or civ engineer needs the cooperation of a ge= ‘ologist to identify the deposition environment and the subsequent deposition and weathering eycle (Atkinson, 1993). The effect of mineralogical history on the mechanical properties is evident in clay soils where the very smal particles exhibit colloidal properties. Minerals ar dived into two groups ~ primary and secondary ~ depending on their state in the geological cycle. Primary minerals are those who inherit te properties of parent rock and their chemical compositions remain unchanged. The common primary minerals in soils are shown jn Table {.1 (Pan, 1994) The sand and silt fraction ofthe most soils consists largely of primary minerals. Quartz (SiOz) and feldspars (MAISis0x, M= Na", K” and Ca" cations) are the most abundant primary minerals in soils (Bohn, eta, 1985). A rock of igneous origin may ave $0% to 70% quartzNature of Soils, Plasticity and Compaction ‘Table 1.1. Common primary minerals in sil. Primary mineral {Chemical composition Quant SiO 2. Feldepar: rthoclase, microcine KAISi905 Albitetplagioelase) NaAlSisOy 3. Mica Muscovite HKALSIO Biotite (HK) UMg.Fe(AL Fe), SiO 4. Ferromagnesians Homblende Ca(FeMg)a Sina Olivine (FeMen Sidx 5. Magnesium Silicate Serpentine HaMgsSioOy 6, Phosphate Apatite (CaxPODDsCatFe-Cy 7. Carbonates Calcite CaCO, Dolomite CaMgtCOs)2 3 Quartz is believed eo have been crystallized at low temperatures (Tan, 1994) and therefore it is reasonably stable at ambient temperatures and resistant co weathering but the particles event ally become rounded in nature. Other primary minerals such as micas, and a variety of acces sory minerals of primary origin, are present in soils but in smaller quantities than quartz and feldspars. Primary minerals may also occur in the clay fraction, but these are not the major constituents of the elay soils. Secondary minerals are produced by weathering of rocks and primary minerals and have different chemical compasition from their parent materials. The clay fraction of most clayey soils consists of secondary minerals. ‘The Jackson-Sherman weathering stages considers three stages in the weathering process (Sposito, 1989). These are classified as early stage, intermediate stage and advanced stage, & Classification that indicates the intensity of leaching and oxidizing conditions. The products of the early stage consist of carbonates, sulphates and primary silicates. Intermediate stage ‘weathering results in quartz, muscovite and secondary alumine-silicates of clay fraction of soil including mica and smecttes. The advanced stage creates kaolinite that is an important clay ‘mineral. In general clay minerals fall into intermediate and advanced stages as shown in Table 1.2. This Table shows the increasing order of persistence of the soil minerals and thelr oceur- rence in time. The size of the mineral, its shape and its hardness controls the rate of weathering and their engineering behaviour (Ollier, 1972). The large minerals are more resistant to the ‘weathering because of their low specific surface which reduces the area of contact against the weathering agents. The clay minerals have a fine texture, with individual particles less than 0.002 mm in diameter. As a result of the large surface area associated with the small particles and masses, clay minerals develop plasticity when mixed with water. A knowledge of the crystal structure of clay minerals is necessary in order to understand their behaviour.4 Soil Mechanics: Basic Concepts and Engineering Applications Table 1.2. Jackson-Sherman weathering stages (Sposite, 1989). ‘Clay minerals in soi fraction ‘Soil chemical and physical conditions Early sage Gypsum ‘Very low content of water and organic mater Carbonates Ver limited leaching Olivine / pyroxene / amphibole Reducing environments Fesbearing micas Limited amount of ine for weathering Feldspar {Intermediate stage Quarez Retention of Na, K, Ca, Mg. Pet}, and silica Dioctahedal mica lite Ineffective leaching and alkalinity Vermiculit/ chlorite Tgncous rock rich in Ca, Me, Fel), but no Smecttes Feil) oxides Silicates easly hydrolysed Ploceulation of silica transport of silica into the weathering zone Keoline Removal of Na, K, Ca, Mg, Fe() and silica: Gibsite Effective leaching, fesh water Iron oxides (gotte, hematite) ‘Oxidation of Fell). acidic compounds, tow pH Titanium oxides (anatase, rll, ilmenite) Dispersion of silica, Al-hydroxy polymers 1.2.2 Crystal structure of clay minerals Tn general, clay minerals are alumino-silicates made by a combination of silica and aluminium ‘oxide units with metal ions substituted within the erystal. A schematic diagram ofa silica unit is shown in Figure 1.1(a). Each silica unit consists of a teirahedron with four oxygen atoms, O, located at the vertices ofthe tetrahedron equidistant from each other. A silicon atom, Si, is lo- cated inside the tetrahedron equidistant from the oxygen atoms. These units combine to form a silfea sheet as showa in Figure 1.1(b). The arrangement could happen in variety of ways. Fig- ture £-1(b) shows an arrangement in 2 hexagonal pattem in which each basal oxygen atom is shared with the adjacent unit. This sharing results a negative charge io the basic unit which can be increased to zero if, for example, aluminium replaces silicon, Figure 1.1(c) shows the short ‘hand symbol generally used for silica. Figure 1.2(a) shows a unit of alumina (aluminium ox- ide) where six hydroxyl ions surround one aluminium atom, Al, in an octahedral arrangement ‘The combination of alumina units in staggered rows results in the formation of en alumina sheet where each hydroxyl ion is shared by three basic units (Figure 1.2(b)). This sharing ce- sults a positive charge in each basic unit which can be reduced to zero if magnesium or iron replaces aluminium. Figure 1.2(c) shows the short hand syinbol generally used for slumina. ‘The basic elemental structure of a clay mineral is formed by different combinations of silica and alumina shects. Their type and engineering characteristics depend on the strength of the ‘ponds that connect these elemental structures.Nawure of Soils, Plasticity and Compaction $ © Oxygen ©@ Silicon ‘Show Hand symbol © Figure 1.1.(@) A base silica unit, (b) silica sheet, (c) shorthand symbol for silica shee. -1 0K: Hydroxy © Aiuminiurn ‘Shor Hand symbol ©) Figure 1.2. (a) A basic alumina unt, () alumina sheet, (c) shorthand symbol for alumina sheet. Kaolinite. Kaolinates, with the general chemical composition of 2Si02Alz032H20, are pro duced from weathering of the parent rocks that have orthaclase feldspar (e. granite). An ele- ‘mentary layer of kaolinite crystal is made of one silica sheet and one alumina sheet mutually sharing the oxygen atoms between them. These layers join together to form a kaolinite particle as shown in Figure 1.3, The number of elemental layers in one stack may reach one hundred or more. The elementary layers are held together with hydrogen bonding giving a stable structure to the mineral. The hydrogen bonding is a resutt of the attraction forces between the oxygen atoms of the silica sheet and the hydroxyl ions of the alumina sheet. These forces are high enough to prevent the penetration of water between the layers and consequential expansion, The minerat is moderately plastic when itis Wetted and has some frictional resistance to shear. Hoalloysite is another type of kaolinite mineral with a generat chemical composition of 2Si02Al2034H20 and a tubular structure with water molecules between the elemental Layers of the kaolinite crystals6 Soil Mechanics: Basic Concepts and Engineering Applications AL St @ ©) Figure 13. Elemental stwctre of kaolinite. As a result, in contrast to kaolinite, it is susceptible to expansion and contraction. Upon heat- ing, halloysite is ireversibly dehydrated and is converted to metahalloysite being a precursor of kaolinite following the weathering sequence: Igneous racks —> smectite (monmorillonite) ~ halloysite > metaballoysite —> kaolinite (Tan, 1994), ite. Mite is produced from the weathering of micas with the major parent mineral of musco- vite. A schematic representation of illite struccure is shown in Figure 1.4, lite has a mice type structure in which an elementary layer is made from one alumina sheet sandwiched hetween sheets. The bond between the two elementary layers is made by potassium, K*, which joins the six oxygen atoms of the two silica sheets. This bond is not as strong asin kao- linite and, with a random staking of layers, there is more space to water 1o enter between the elemental layers. Montmorillonite. Montmorillonite, with chemical composition of 4SiO2AI2031120 + nl120, is 1 product of the weathering of volcanic ash in marine water under poor drainage conditions Figure 1.5 shows the elemental structure of montmoriltonite. Similar to illite the structure is of ‘mica type but the bond between the elemental layers is by weak oxygen-to-oxygen links com= prises of water molecules and exchangeable cations. s Al Si @ Kions ) Figure 1.4, Elemental structure of illite.Nature of Soils, Plasticity and Compaction 7 Si AL Si @ ©) Figure 1.5, Elemental structure of montmorillonite “The major characteristic of this mineral is its volume expansion or contraction due 0 an in- crease or decrease in moisture content. At high moisture contents the mineral is plastic and can he deformed easily under minor stress but hardens when dry. The type of structural units and their bonding influences the shape ofthe clay minerals and ther surface area. As a result they axe ether plate shaped or tubular. The average diameter varies from 0.1 to 4 pm, while the thickness has a range of 30 Angstrom (1 Angstom = 10"" mm) t0 2 pm. In general particle size decreases from kaolinite to lite and to montmorilonite. The surface area fora constant ass (specific surface) isa function of shape and size, The specific surface of montmorillonite is approximately 50 t 60 times tha of kaolinite and 10 times tha of lite. Due w the high ad- sorption of water by montmorillonite, the mixture ofthis group of minerals and water into the state of a viscous fluid, is used in engineering construction a8 a grouring material to reduce ‘oid ratio and permeability of soils (ASCE, 1962). 123 Soil organic matter Soil is a multiphase system containing solids, liguids, and gases in which the gas and liquid phases are essentially air and water respectively. The solid phase consists of inorganic and or- ganic material. The organic material contains most of the carbon of the soil and usually con sists of both dead and living matter. Dead organic matter is the remains of dead plant material, and animal residues, It is the main source nutrients for plants and influences the engineering properties ofthe soil. The live microbial and plant matter comprises a few percent of the total ‘organic matter, che average amount of which may change up to 5% to 10% (by mass) depend- ing on the drainage conditions, In some peat soils, however, the amount of organic matter ap- proaches 100%. According to White (1979) the amount of organic matter on the surface of soils (eg. Australia) varies up t0 16%, but decreases rapidly with depth. Organic soils and pats are recognized by their high liquid limit (the amount of water needed to turn the soil into a viscous liquid) but the range of moisture content in which the sails show plastic behaviour is small, Organic soils, especially those with a spongy structure, are also highly compressible and do not meet the Settlement imitations of mast structures that may be built upon them,8 Soil Mechanics: Basic Concepts and Engineering Applications Tale 13. Methods of tests for soils fr civil engineering purposes: chemical and electrochemical tests Text ASTM, BS aS. ‘AASHTO ‘Organic mater Loss on ignition Taraion 1377 TIO Tieton content D.2974 {Loss on ignition:1377, T1938 Sulphate Gravitational 1377 1289-421 sontent fon exchange: 1377 pil Value Elecrometric: D-1067 Electrometric: 1377 1289.43.1 Colorimetric: D-L067 Colorimetric: 1289.43.2 Electical Disc electrodes: 137 1289.4.8.1 Carbonate, Titration: 1377 Chloride Water-soluble and content Acid-sluble: 1377 ‘The determination of the organic matter in a soil and its chemical composition ~ sulphates, carbonates and chlorides is recommended in civil engineering construction works because of ‘their detrimental effects on its engineering properties (Table 1.3). 124 Electric charge in soils The solid phase of most soils caries a net negative charge whichis associated with the small colloidal particles of inorganic and organic origin. While the charge is negative on the surface of te clay particle; the edges carry positive or negative charges due to broken bonds. The sil particles that are responsible for cation exchange ae referred to as colloid particles. A colloid 4s a very fine particle of organic or inorganic origin with a maximum size of 0.2 ha and a ‘minimum sizeof 50 A. A permanent charge is @ result of isomorphous substitution inthe erys- tal structure ofthe clay minerals (eg. substitution ofthe Als” ion for Sig" or Mg’ for Als). The resulting charge can be neutralized by the cations tha are positively charged atoms pre- seat in the solution surrounding the clay particles. In lite and mica clays the potassium bond is strctural rather than exchangeable (McLaren & Cameron, 1996), and the cation exchange capacity is relatively low. The unbalanced electric charge is responsible for the Formation of clay structure. Surface-to-surface contact is not possible as both have net negative charges ‘hilt the edge-o-surface contact is common asa result of the attraction between the positive and negative charges. The electic charge does not affect the engineering behaviour of coarse particles, asthe ratio of surface to mass is Low in comparison to that of the clay particles. A ‘variable charge i a result ofa change in pH in organic soils. A decrease or increase in pH will result & decrease or increase in negative charge. In some soils that have iron and aluminium oxides, there are some areas, along the colloid particle surface, that have a positive charge and will be neutralized by the adsorption of negatively charged anions. The quantity of exchange- able anions held is much smaller than the amount of exchangeable cations. The distribution of cations and anions from the surface of the particle towards the voids created by the arrange ‘ment of the particles has been modelled by different molecular adsorption models (Sposito,Nature of Soils, Plasticity and Compaction 9 1989). The simplest model isthe Helmholtz double layer model in which it is assumed that the layer balancing the negative charges is immediately adjacent to the surface of the colloid. In tie diffuse double layer, referred to as the Gouy-Chapman model, the ion concentration de- creases gradually away from the surface of the colloid particle untt it becomes equal t the ion concentration of the bulk solution within the voids (Figure 1.6). The forces of atraction or re pulsion between electrical charges can be calculated mathematically from Coulomb's law. Combining the equation of cation atraction with the mathematicat expression of diffusion, a relationship can be found between the electric potential and the distance from the colloid sur- face (Boltzmann equation). Thus, the characteristics of the diffuse double layer may be pre= sented by three types of distribution that include the distribution of cations (Figure 1.6). the distribution of anions and the electric potential within the layer. The thickness of the diffuse layer depends om the type of the ion and its bulk concentration and for natural soils it may range from 1 t0 20x 10-® mm, This model considers only the electrostatic forces of attraction and repulsion and disregards the specific forces created by the finite size ofthe fons. The Stern model is a modified version of the Gouy-Chapman madel in which the double layer is divided into two layers of the Stem layer immediately next to the colloid surface and the Gouy. ‘Chapman model beyond this layer (Figure 1.7). Electrical potential decreases linearly within the Stern layer and exponentially across the diffuse layer. ‘The investigation af the nature of the layer of water surrounding a colloid particle is impor- tant in soil chemistry, agriculture and environmental soil science from the view of exchange- able cations available for plant intake. Furthermore, the clay fraction may intercept some pol- Tuas, carried by soil water, duc to its cation exchange capacity and thus acts like a natural purifying agent. This layer is strongly attracted to the soil particle and has different properties from the water in the pare voids. In soll mechanics the study of the nature and mechanisms of held and attracted water is important, as it is responsible for the plastic behaviour ofthe soil The term adsorbed water is used to describe this layer: whilst the water beyond the boundary ‘of adsorbed water is called free warer. The water particles within the adsorbed layer can move parallel to the surface of the colloid particle, but the movernent in the direction perpendicular (o the surface of particle is not possible or is restricted. ne ‘Thickness of Gouy- Chapman layer (ay particle lon concentration Normal concentration “SL inpore water Distance trom colloid surface Figure 1.6. The Gouy-Chapman double layer model10. Soit Mechanics: Basic Concepts and Engineering Applications Linear Exponential Clay particle Electrical potentiat “S_ Normal concentration in pore water Distance from collid surtace Figure 1.7. The Stem double layer mode! 125 Soil structure Soil structure isthe three dimensional geometrical arrangement of pores and particles of vari ‘ous sizes. Some geologists have referred to the structure as soil fabric ro make i¢ compatible with the term rock fabric which describes the arrangement of mineral grains in the volume of the rock. Brewer (1964) defines fabric as the physical constitution of a soil material as ex- pressed by the spatial arrangement of the solid particles and associated voids. The spatial ar- angement of coarse material is called the soil skeleton. Brewer & Sleeman (1988) suggest that ‘the soil structure is based on the arrangements of solids and pores in which the primary parti cles form a compound and the arrangement of compounds constitutes the soil structure. Al- ‘though in many definitions the emphasis is on the three-dimensional nature of the soil struc- ture, two-dimensional studies of stability problems and soil water movernent have been very successful in soil mechanics. In clay soils the interparticle forces described previously have a distinct effect on the arrangement of the clay particles. Ifthe resultant of the forces creates a net repulsion then a face-to-face arcangement called a dispersed structure will be created (Fig ure 1.8(a)). The net atraction results in an edge to face structure referred to as a floceulated structure (Figure 1.8(b)). Experience indicates that clay deposits that are developed in fresh or salty waters have a flocculated structure, while the transported and remoulded clays tend to have a dispersed structure. The floceufated structure in salty water is more dominant than in fresh water. In salty clays (e.g. marine clay) the concentration of cations is high, resulting in thin adsorbed layers around the surface of the particles. The attraction forces dominate the par- ticle arrangement; thas an edge to face structure is made. A clay soil with pure water moves towards the dispersed structure. In laboratory techniques based on sedimentation, a salt, com- monly sodium hexametaphosphate, is added to the suspension liguid as a dispersing agent 0 facilitate particle sedimentation, Sometimes the aggregations made by original mineral units ‘make dispersed oF flocculated type structures depending on the deposition environment (Fig- ‘ues 1.8(c) and 1,8(d)), These aggregations have face-to-face structures, as there are little or no attraction forces between mineral units. Depending on the velocity of the water in which the particles settle these aggregations, or fine silts and sands, make a honeycomb structure asNature of Soils, Plasticity and Compaction 11 showa in Figure 1,8(c). This structure contains a large amount of voids that are bridged by the assemblage of aggregations or fine silts and sand particles. A diagrammatic reconstruction of an electron microscope image of clayey sand is shown in Figure 1.8(f) (Tan, 1994), The sand particles (S) form the main skeleton of the soil structure. Clay (C) and organic material (black Particles fill the voids berween the sands and the remaining voids are occupied by air. This reconstruction cannot be generalized for sands and coarse materials as the process ofthe struc tural arrangement of particles is complex and must be axsessed by standard tests and scanning electron micrographs. An idealization of sand or a coarse-grained soil structure is shown in Figures 1-8(g) and 1.8(h) for dense and loose structures respectively. The dense model was used by Rowe (1962) to demonstrate the stress-dilatancy in sand during shear 1.2.6 Clay structural analysis: X-ray diffraction method ‘This technique was originally investigated by Whittig (1964) and Iater suggested by Moore (1970a, b) to the ASTM. The X-ray diffraction method has probably contributed more to the ‘mineralogical characterization of the elemental structure (soil layer silicates) than any other single method of analysis (Boben, etal. 1985). Optical and electron microscopy methods are @ my Figure 1.8, Different structures of clay, natural soil and idealization of dense and lose sand12 Soil Mechanics: Basic Concepts and Engineering Applications Figure 1.9, Geometric conditions forthe X-ray reflection from mineral planes. ‘wo other techniques used for the measurement of the fabric within a relatively small volume of soil. The optical microscopy is based on the optical birefringence of the clay particles sub- {ected to polarized light. In electron microscopy a relatively larger area is scanned by the elec tron beam allowing the direct observation of the material at a wide range of magnifications. In the X-ray diffraction method the material is exposed to a filtered X-ray beam as idealized by Figure 1.9. The X-ray passes into the material and causes the electrons in the atoms of the erals to vibrate and reflect the beam thtough the successive planes. The method involves increasing of incidence angle and monitoring the intensity of the diffracted X-radiation until a maximum value of the diffracted intensity is achieved. The X-ray diffraction maximum is de- tected whenever the following equation is satisfied: .dsin® aay where n =a series of imegers, 1,2, 3.....=known wavelength of the X-radiation, d = the re- peated space between diffracting planes and @ is the angle of incidence of the X-rays. Detec- tion of the maximuin diffracted radiation intensity is carried out either by a cylindrical photo- graphic film placed around the sample or by a rotating detector. This technique is valuable for ‘measuring the thickness ofthe elemental structure as well 3s monitoring the interlayer spacing with any change of the external conditions. The X-ray diffraction method is also used in soil stabilization where the soil or a manufactured fine grained material such a fly ash is stabilized by lime or cement. Development of hydration and bond formation between particles may be ‘monitored by this technique to justify the water content used for stabilization, 13 MASS-VOLUME RELATIONSHIPS 13.1 Basie physical properties ‘A soil sample contains soil particles (solids), liquid (usually water) and air. These three phases can be visualized as three Separate blocks as shown in Figure 1.10. This representation is an artificial concept and all phases are inextricably mixed. The phase diagram is dimensioned inNature of Soils, Plasticity and Compaction 13 votume Mass Vv, AIR ° y Ve WATER, vs sous |m, Figure 1.10, Dimensioned phase diagram, terms of volume and mass. A completely dry soil ora fully saturated soil is a two-phase sys- tem, Volume related symbols are defined as follows: Vi = Volume of the air within the voids between particles. Vj = Volume of the water within the voids between particles, ¥; = Volume of the solids. Vi = Volume of the voids within given sample = Va-+ Va ‘otal volume of the soil sample = Vs + Mass related symbols are: My = Mass of air M-= Total mass of soil sample = My + M,. ‘The common volume related parameters are as follows: = Mass of water, M, = Mass of solids. Void ratio e, The ratio of the volume of the voids to the volume of the solids: 1.2) Specific volume v. The volume occupied by a unit volume of solids: v_y4Y, Ye yeite 13 aA 3) Porosity n. The ratio of the volume of voids to the total volume: % Vee nett nee 14) v Vi+¥, Ive un ‘Void ratio is usually expressed as a decimal whilst porosity may be expressed as a percentage. of the volume of the water to the voluine of the voids: Degree of saturation 5. The us)14. Soll Mechanics: Basic Concepts and Engineering Applications Degree of saturation is usually expressed as a percentage. For a dry soil S; = 0. When the vol- ume of the voids is fully occupied by water, Vi, = Vp and S, = 1 or 100%. Air content Ay. The rato ofthe volume of air othe tot! volume: ee a6) ‘The common mass-volume related parameters ae: Moisture content or water content m or w. The moisture content or the water content isthe ra: tio of the mass of water tothe mass of the sotids or dry mass of soit: mor an This parameter is usually expressed as a percentage. Density of solids Ps The ratio of the mass of the solids to the volume of the solids: p= ie 8) v, In the SI system the prefered unit is Mg/m’. Numerically 1 Mel 1000 kg/m gyem’ = 1 tonne/m? Specific gravity of solids Gy, The ratio ofthe density ofthe solids tothe density of water: <9) where py = density of water = 1 Mp/m® at 4°C, This is a dimensionless parameter within the range of 2.6 to 2.8 depending on the mineralogy of the soil Density of soil p. The ratio ofthe total mass to the total volume: M v A preferred usage unit is Mg/m®, The terms bulk density or wer density are also used. For a given void ratio, this parameter changes from a minimum at the dry state to a maximum at the saturated state (1.10) ty Me+ViPw =! abinimun, py MY H ¢ataximamy aay nya Y, Pug BEATE » » “The relationship between dry density, moisture content and bulk or wet density is oe aNature of Soils, Plasticity and Compaction 15 Density index. To express the consistency states of sand and gravel, the natural void ratio is compared with the minimum and maximum void ratios obtained in the laboratory. Density in- dex (or relative density) is defined by: Ip = BS (loosest) 0 ¢ Ip $I (densest) (1.13) Usefal equations: Dry density oxpressed in terms of specific gravity and void rato Aes Se aus) ‘Void ratio expressed in terms of moisture content, specific gravity and degree of saturation: WG, “Ze ew, (fr fully saturated soit (115) By combining Equations 1.14 and 1.15, a relationship i found w express dry den ‘of moisture content, specific gravity and degree of saturation, Gp, pete 1 Pa T0615, oe py = 2H for fully saturated soil) 17) +wG, Ina coordinate system with w on the horizontal axis and py on the vertical axis, this equation represents the possible states of saturated soil and is termed zero air curve. Dry density expressed in terms of specific gravity, moisture content, and air content. en YatVe MulDe AVA, My tVADe GM ytGVADy i ,GeADe * ® i MIB, MIG, a Substituting fore in Equation 1.14 and rearranging, Gpul=A) ai 1406, For zero air curve 4y = 0 this equation reduces to Combining Equation 1.16 and 1.18 the folowing relationship between degree of saturation, air content and moisture content can be obtained. a GI Ay) A466, Relationship between ¢, Gs, is 8d Pea (ay16 Soil Mechanics: Basic Concepts and Engineering Applications 2G, ~ Pan Gte PaGe—Pat or: 9.5 = Dy 1.20 EO Bat Pw (1.20) ‘Te submerged or buoyant unit weight = Yq Satrated unit weight) —7, (unit Weight of Water) = Oga ~ using Equation 1.20: G, Ye T 5 (21) Te Example Ll A soil specimen has the following properties: G,=2.7, ¢ = 0.6 and w = 14%. Calculate: dry density, dry unit weight. wet density, wet unit weight and degree of saturation. Solution: When the void rato eis known, we calculate the volume of solids V; and the volume of voids Vp in terms of a unit total volume (V= I mr’) : (10.6) =0.625 m’, V,4V, =Le=V,IV, orV,=ld+e¥, =e +e), fore =0.6.¥, ¥, =1-0.625 =0.375 m’. Dry density is the mass of the solids that occupies 2 unit volume: Py = G,Py¥, =2:71.0%0.625 = 1.69 Mg/m’. Dry unit weight: yy =1.69x9.81=16.6kN/m. 93Mgim’. Wet density: p= py (1+ w) = 1.690 +0.14 ‘Wet unit weight: y = 1.93x9.81 = 18.9 kN/m’ Vig = Mg [Pqy = OLAM Py =0.141.69/1.0 5, =¥yp Ve =0.237/0.375 = 0.632 = 63.2% 3 237 ms Example 1.2 A soil sample has the following properties: G;=2.7, e = 0.7, S,= 80%. Calculate: dry density, dry unit weight, moisture content, wet density, wet unit weight and air content Solution: For e=0.70, V, = 1M1+0.7)= 0.588 m', V, =1-0.588=0.412m*. Py =G,PyV, = 2.7x1.0%0.588 = 1,59 Mgim’, yy =1.59x9.81=15.6KN/m- 5, =Vy/V, #08, V,,/0412=0.8, Vy =0.330 m>, V, =0.412-0.330 = 0.082 My, =Vj, XP, =0.330%1,0=0.33 Mg, w= 0.33/1.59 = 0207 «= 20.7% P= py (I+ w)=1.59(1+ 0.207) =1.92 Mpim’, 4, = V, JV = 0,082/1.0= 0.082 = 8.2%. Example 1.3 Tina field density est the following results are obtained: p = 195 Mg/m’, w = 0.16. The spe cific gravity of solids is 2.7. Find: dry density, void ratio, air content, degree of saturation, saturated unit weight and moisture content a full saturation.Nature of Sotls, Plasticity and Compaction 17 Solution: Py =PM(L+ w) =1.95/(1+0.16) = 1.68Mp/m®, Assume the total volume is 1 m’, thus V,=M,/Py =M, NG,XPy) = 1.68/(2.7X1.0)= 0.622 m°, V, =1-0.622=0.378mm° = 0:378/0.622 =0.608. Vj, = My/P = (1.95 ~1.68)/1.0 = 0.27 my A, SV, /¥ =(0.378~0:270)/1.0 = 0.108 = 10.8%. 5, =V,, IV, =0.270/0.378 =0.714= 71.4% ‘The volume of water at the saturgted state = 0.378 m and its mass is 0.378 Mg. Thus, Pyqy = 1.68 +0.378= 2.06 Mel’, Yay = 2.06%9.81 = 20.2 KNimn 5%. ‘Moisture content at full saturation = 0.378 / 1.68 132. Experimental determination of physical properties of soils Determination of volume-mass related parameters require the direct measurement of the total mass, mass of water, total volume, and specific gravity of solids. The volume of a specimen of an irregular shape is measured using water or mercury displacement techniques. Determina tion of the moisture content includes measurements of wet mass and oven-dried mass to estab- lish the masses of the water and solid particles. In most standards the temperature of the oven for drying purposes is 105°C to 110°C, and the specimen must stay inthe oven for 24 hours. The specific gravity of solids. This parameter is measured using a simple apparatus called a pyknnometer which is a glass jar with a conical top as shown in Figure 1.11. The cone has 2 hole at the top (0 allow the water to overflow. The test procedure described in ASTM D-854, AASHTO T100, BS 1377 and AS 1289,3.5.1, consists of the following steps: [. Measure the mass af empty pyknometer (1). 2. Fill the pyknometer with water until it overflows from the hole at the top of the cone and find the total mass of the pyknpmeter and water (M2) 3. Empty the pyknometer and place a sample of oven-dried soil inside the pyknometer and measure the combined mass (M3). 4. Add water until it overflows from the top of the cone and measure the tocal mass of the ap- paratus (Ms). At this stage the jar must be agitated to remove the air bubbles. The specific gravity isthe ratio of the mass of dry soil tothe mass of water displaced by soil: Ms-M, Ot, —M)- a= M5) ‘A similar method uses a narrow-necked density bottle (oF volumetric flask) of 25, 50 or 100 ‘ml volume that may be agitated by a mechanical device called an end over end shaker. 6, a2 Field density. Common methods are based on excavating a hole atthe site and measuring the volume of the hote and the mass of the excavated soil, Field density is the ratio ofthe mass of the excavated soil to the volume of the hole. The volume of the hole is measured in a variety ‘of ways using mostly sand or water replacement methods. Inthe sand replacement method two18. Soil Mechanics: Basic Concepts and Engineering Applications Figure 1.11, Pyknometer with rubber washers (Wykeham Parrance) apparatus are available, In BS 1377 and BS 1924 the apparatus is an assembly of a sand pour- ing metal cylinder of 100 of 200 mm diameter, a metal cone and a metal tray of 300 mm square with 100 or 200 mm diameter central hole (Figure 1.12(a)). The metal tray is adjusted horizontally on the ground surface and the soit is excavated through the hole in the tray to a depth of approximately 1-5 times the diameter, with the resulting excavation being preferably cylindrical in shape. wis Figure 1.12. (a) Typical sand pouring apparatus, (b) calibration cylinder, (c) sand cone apparatus ‘Wykeham Farance).Nature of Soils, Plasticity and Compaction 19 ‘The sand-pouring cylinder is placed on the tray and fitted to the central hole. Opening the shutter system allows the sand to flow through the cone, until the hole and the come are filed with the sand. The mass of the sand in the excavated hole is calculated by measuring the mass of the apparatus before and after the test, aid knowing the mass of the sand in the cone, The volume of the hole is equal to the mass af sand in the hole divided by the pouring density of sand. To determine the mass of the sand in the cone the cylinder is filled with sufficient mate- rial and the mass of the assembly is measured. The apparatus is sanded over the hole in the ttay which is placed on a flat and level surface. When the sand stops ruaning the tap is closed and the remaining mass of sand is measured. The difference between the two measurements before and after the operation gives the mass of the sand in the cone. The pouring density is ‘measured by pouring the sand into a calibration cylinder (Figure 1,12(b)) of known volume. A similar method using a sand pouring can is recommended by AS 1289-5.3.2. The sand used is clean and dry, 90% passing a 1.18 mm sieve and 90% retained on a 600 um sieve. Sometimes an upper size of 2.36 mm / 1.18 mun or a lower size of 600 um / 300 um is used. Care has to be taken in shielding the sand to prevent moistening and bulking that may invalidate the re- sults. The second apparatus based on the sand replacement method is called sand cone appara- tus (Figure 1-12(¢)) and is used by ASTM D-1556, AASHTO T191 and AS 1289.5.3.1 stan- dard systems, This apparatus is used in holes where the maximum particle size does not exceed 38 mm, The apparatus includes a plastic container of 5 litres capacity, a density plate Gay) with 165 mm diameter central hole and a cone with 152 mm diameter that has a shutter mechanism and threaded nozzle. According to ASTM D-1556-90 (1996) recommendations, the sand cone method is not suitable for organic, saturated, or highly plastic soils that would deform or compress during the excavation of the test hole. It may not be suitable for soils con- sisting of unbounded granolar materials or granular soils having bigh Void ratios that will not ‘maintain stable sides in the test hole. The balloon density apparatus used by ASTM D-2167, AASHTO T205 and AS 1289,5.3.4, utilizes an empty rubber balloon to measure the volume of excavated soil. The balloon is slowly filled with water from a calibrated glass container until the hole is completely filled by the expanded balloon. A direct water replacement method (ASI289.5.3.5) uses a density ring on the surface and a plastic sheet to retain the water in the hole. This test is particularly appropriate for soils containing large particles as defined in AS 1289.0. In BS 1377 and BS 1924 the water displacement method is recommended only for Stabilized soils. In fine-grained soils the core cutting method is used to determine the field density (ASTM D-2937, BS 1377 and AS 1289.5.3.3). The core cutter commonly bas a diame- {er of 100 mm and is 130 mm long and is driven into the soil using a driving ram of standard ‘weight. By measuring the mass of the soil in the core cutter and knowing its volume the field density can be calculated. Application of nuclear gauges for the determination of field density and in-situ moisture con- tent provides a rapid method for compaction and moisture control (ASTM D-2922, D-2950, 3017, BS 1377 and 1924). Readout in dicot units are also provided for void ratio and de gree of compaction. A nuclear moisture-density gauge operates by emitting gamma and neu- won radiation from radioactive materials sealed within a capsule in the gauge. The gauge also includes a detector of the Geiger-Mueller type to collect and read gamma radiation emitted from the cesium 137 and thereby determine the field density. For neutron radiation americium20. Soil Mechanics: Basie Concepts and Engineering Applications @ ) Figure 1.13, Transmission modes in nuclear moistre-density gauge. 241: berylium is used in the measurement of moisture content. A material of high density ab- sorts the gamma radiation and acts as a radioactive shield resulting in a Tow reading, whilst a ‘material of low density will give a high reading. The americium 241: berylium emits neutron radiation to the test material. Some of this radiation or high energy neutrons are moderated by the hydrogen atoms of the water in the material and the rest is detected by the Geiger-Mueller detector showing a reading on the gauge. A high reading in a specified period of time corre sponds to a wet material, while a low reading corresponds to a dry material. Based on the amount of radiation detected in both cases the gauge could be calibrated either by the manu facturer or by the user. Two basic techniques, backscatter and direct transmission, are used for the determination of moisture content and density as shown in Figure 1.13. Moisture content is, determined by using the backscatter mode only whilst density may be evaluated by either backscatter or direct eransmission. The backscatter mode is shown in Figure 1-13¢a) in which Doth the radiation source and detector are on the surface of the test material. Direct transmis sion is not possible because ofthe intervening shield and the detector counts only on the radia tion received by reflection or radiation scattered back to the detector. Using this method depths up 10 65 mm may be analysed. In the direct ransmission mode (Figure 1.13(b)) a small hole is Grilled in the test material to a depth of 300 mm and, whilst the radioactive source is lowered into the hole, readings are carried out at increments of 25 to 50 mm. 14 PARTICLE SIZE DISTRIBUTION distribution curve 141 Definition of soils and partcte siz Figure 1.14 shows a generally used size classification for soils as proposed by MIT (Massa cchusetts Institute of Technology). Each group is divided into three subdivisions of fine, me- dium and coarse particles. Sands and gravels are coarse-grained soils; and considered as non= cohesive (or cohesionless) materials. They are also referred to as granular materials. Clay and sile are fine-grained and are classified as cohesive materials. In practice fine-grained materials are soils passing # 200 sieve, which has square openings of 0.075 mm.Nature of Soils, Plasticity and Compaction 21 002 0.06 20 600 CLAY sit SAND GRAVEL [COBBLES 0008 ose | 92 06 | 60 200 e[m[c{r[mi[co[e[ulc Particle Size (mm) Figore 1.14, Definition of sits according to MIT. A particle size distribution curve describes the percentage by mass of particles of the different size ranges. The horizontal axis represents the particle size on a logarithmic seale. The vertical axis represents the percentage by weight of particles that are finer than a specific size on the horizontal axis. An example of this curve is shown in Figure 1.15. Consider point A on this curve with coordinates of 9.5 mm and 82%, This means that 82% by weight of particles are finer than 9.5 mun, In atypical particle size distribution test, sieves are stacked in a motorized shaker in order of decreasing aperture size and a pan at the bottom of the sieve column retains fine material that passes through all the sieves. For the material passing through sieve # 200, sedimentation testis carried out. Figure 1.16 show's three size distribution curves: soil Wis garded as a well-graded soil, soil U is @ uniformly graded soil witha size range much less than soil W, soil P is a poorly graded soil and lacks a particular range of particles. To describe the state of grading, two parameters are defined as follows: Coetcien of Uniformity: Cy = Dey! Dio a2 Coefficient of Curvature: Ce = 03s Kw 2) aa where Dio, Dyp and Deo are pasticle sizes corresponding to 10%, 30% and 60% finer respec- ‘ively and are determined from the particle size distribution curve obtained in the laboratory. 100 90 80 70 60 50 40 |- Percentage finer by woight 0001 001 ot 1 10:2. 100 Particle size (mn) Figure 1.15. Particle size distribution curve, Example | 4),22 Soil Mechanics: Basic Conceps and Engineering Aplications 100 so Tass 0 serene * \ & so Fd so 2 vo 3 door aaron tt Particle size (mm) u has this size range only Percentage finer by weight Figure 1.16, Wellgraded, uniformly greded and poorly graded sols. ‘The magnitude of Cy indicates the spread of grain size values: a larger range leads to larger ‘number. According to the Unified Soil Classification, for well-graded soils, the value of Ceis between I and 3. For a well-graded gravel, or gravel-sand mixtures with litle or no fines, the value of Cy is greater than 4, For well-graded sands, and gravelly sands with litle or no fines, the value of Cy is greater than 6, 142 Sieve and hydrometer analyses “The standard procedure for determining the distribution of particle sizes may be found i ASTM D-422, BS 1377 and AS 1289,3.6.1. Sieve requirements are specified in ASTM El, BS 410 and AS 1152 and are manufactured in three diameters of 200 mm, 300 mm, and 450 mm. Normally, sieves have a mesh of woven brass wire with square apertures ranging from 0.075 mm to 125 mm. The sieve with 0.075 mm apertures (or sieve # 200, indicating 200 ‘openings in 25.4 mm length) is the smallest practical sieve size used in soil mechanics labora- tories. Prior to sieving, soil samples must be oven-dried at 105°C to 110°C. For intermediate ‘and fine size particles the soil sample js first submerged in a dispersing solution at least for an hour. It is then washed with water on the 0.075 mm sieve and oven dried. An analysis without washing is also permitted by standard codes, but the results are not reliable if an appreciable amount of clay and silt is present in the sample. Sieving may be done by hand or by a sieving machine. A typical motorized shaker or sieving machine is activated by electromagnetic im- pulses with a control panel for time adjustment and vibrating intensity. Some advanced shak- ers permit the adjustment of pause time between vibrations. A sieve shaker holds up to ten nesting sieves of any diameter. The particle size distribution forthe fraction ofthe soil passing through the 0.075 mm sieve (clay and silt) is carried out using the hydrometer method speci fied in ASTM D-422, AASHTO TS8, BS 1377 and AS 1289.3.6.2. The methods described in these standard codes are not applicable if less than 10% of the soil passes the 0.075 min sieveNature of Soils, Plasticity and Compaction 23 ‘The hydrometer method is based on the measurement of the velocity of soil particles in a sedimentation solution and the dry mass of soil in the solution in different intervals of time. ‘The velocity of falling particles and the dry mass of soil ata specific depth are measured by a hydrometer originally manufactured to measure the specific gravity of a solution. The results are combined with the Stokes’ law, which gives the relationship between the velocity of a spherical particle and its diameter while settling within a solution. AS a result, wo relation ships are constructed between the hydrometer reading and both the diameter and percentage finer than that diameter for each reading at a specific time. Stokes’ law was originally pre- sented in the following form: fisnviey, ~¥.) (128) where D is the diameter of a spherical particle in cm, 1 is the, dynamic viscosity of water in diyne-seo/em’, is the unit weight of the particle in gforce/em Yq is the unit weight of water in g-force/em’, and vis the velocity of the spherical particle in cm/sec. The reading R of a by- drometer atthe time # gives the amount of soil particles in g/1000 mi of solution (or equivalent suspension density) atthe centre of the hydrometer volume, The distance of a reading from this centre, L (Figure 1.17(b), represents the length covered by a particle of a specific diame- ter during this time. Replacing » with L/rin Equation 1.25: b= imirer=%) (126 oe Da VTi0x1.269 0, -p x10 am where D is the diameter of the soil particle in lum, A and Pw are in Mg/ n°, is in 10° Nim? x seconds andr is in minutes, The effective length Lis found by calibrating the hydrometer. Equation 1.27 can be written in the following form: D=FXPXFy (1.28) Parameters F and Fs are calculated from the test data. Parameter 2 depends on the properties of the soil and water and can be tabulated for a specified p,¢Table 1.4) A sample calculation for T= 15°C: 9 = 0.001139 Nim?xs, Py = 0.9991 Mgia’ "y= L749, f1 1392.7 09991) = 1.432. The percentage finer than D is calculated from the following equation: Re Xai M , 100% (129) where M, is dhe mass of dry soil and Rc-is the corrected value of hydrometer reading for tem perature, density of water, and meniscus effects. The parameter a corrects the reading due to Percentage fine the specific gravity of solids as the bydrometer is made (e.g. bydrometer 152H, ASTM) to measure the density of a solution in g/1000ml for G, = 2.65: 1656, 0.6236, 30) 265G,-) G,-124 Soil Mechanics: Basic Concepts and Engineering Applications ‘Table 1.4. Values of parameter Fa versus temperature for p,= 2.7 Mgt TeepsC TempC Temp°C Fe 10 153417 2 1.282 " bin as 2 1267 2 rast 19 26 1253 a ist) 20 2 1239 13 rast 21 2 12s 5 143222 2 6 Lag 30 ‘The amount of oven-dried soil used for the testis 50 g in a clayey soil to 100 g in a sity soil and the sample may be collected from the sieve washings and then oven-dried. The test is cas- ried out in 1000 ml sedimentation glass cylinders (Figure 117(a)). The soil is mixed with wa- ter and a dispersing solution (commonly sodium hexametaphosphate) and is shaken manvally for about one minute. The cylinder is then placed on a firm level and vibration free surface and by immersing the hydrometer in the suspension the reading starts in logarithmic sequences of tiie. An alternative method of obtaining suspension density is the use of an Andreasen pipette (BS 1377). The 10 ml capacity glass pipette is used to obtain a small quantity of suspension from a prescribed depth at different imervals of time to calculate the suspension density. The pipette is fixed toa carriage assembly on a vertical stand and operates with no vibration whilst the pipette is inserted and withdrawn from the liquid suspension. The cylinder used in this ‘method is $00 ml and sampfes are taken fromm a 100 mm depth, Example 14 A sieve analysis on asoil sample of 978 g has given the following results. Plot the particle size distribution curve, determine Cy and Ce and classify the soit @ ® Figure 1.17. (a) Hydromever and sedimentation cylinder (Wykeham Farrance.(b effective depthNature of Soils, Plasticity and Compaction 25 ‘Sieve size (mm) Mass retained (g) Sieve size (num) Mass retained (g) 75 @ Lis uss 35; 285 os 331 0 528. 03, 399 95 978 os 246 475 2013 007s 23 236 298.2 Pan 30. Solution: ‘The results are tabulated below and shown in Figure 1.15, from which: Djo = 0.63 mm, D3o = 2.40 mm, Deo = 4.57 min. Cy = Deg/ Dyo = 457/0.63=7.25, < (DgqDj0)= 2.40" 4 57%0.63)= 2.0. The soil is a well-raded gravel with sad Sieve ire crmy Retained (8) Total retained (5) _ Finer than (i) 5 000 9.000 ‘on 000 ys 2607 2600 ona 9 5308 005 91.995 os 0.000 {8.005 81.995 475 21.15 35201 0.799 236 30390 oon 40:09 iis ine 51807 ings 06 £96 90.308 20 03 400 94392 sis os 2315 36897 vi03 ors rm 38.688 aan Pan 0.306 Example 15 For the following data with M, ‘The bydrometer used is a 152H ASTM, where 0 g and G, =2.7 draw the particle size distribution curve. 163 ~ Rx.49 / 30 (mm), T(C) Time(min) Reading T(C*) Time(min.) Reading r “5s 30 ns 2 Ms ge 60 260 3 300602120 284 4 350 620 250 8 0 9 iu 240 Is 200 Solution ‘The results are tabulated and shown in Figure 1.18. Sample caleulation at = 30 min. L=163.0-27,5x49/30=118.L mm, F, = yI18.1/10 =3.436, F = 1.378, Fy =10/¥300 = 1,826, D = 3.436x! 378%1.82 (finer = 27.5 / 50=0.55 = 55%. 64 4m. From Equation 1.29:26 Soil Mechanies: Basie Concepts and Engineering Applications i Tl a Percentage fi oll U 10 100 Panicle size um) Figure 1.18. Example 15 TEC) Time (ming Reading Lom Fy oy Finer ise 45 «87 ~—«2978 1378 1000 310 2 4159823085 1378 O71 830 is 3 380 009 3.176 B37 S775 760 iB 4 a5 Wso 3249 1378 5.000 no BOS 3200 «OT 3327 1378 3535, 640 1 16 20 186 3400-1378 2.500, 580 Is 30 MS 1813436. «1378 1820 Be 550 2 6. 260 105 43471 1344 1291 6m 52.0 2 10 4 12133486 1312 09S LIS 508, 2 0 250 1222 3496 1282 O64 289 500 19 aso. 40 1238 351813610263 126 480 15. INDEX PROPERTIES AND VOLUME CHANGE IN FINE GRAINED SOILS LS.1 Consistency states of ine grained soils ‘The volume change and flow behaviour of a fine grained soil both depend upon its moisture conient. Ata high level of moisture the soil las the properties ofa liquid, whilst ata low mois- ‘ure content it takes on the properties ofa solid, At moisture contents betseen these 1Wo states the soil passes from a plastic state (0a semi-solid state as the moisture content decreases. The physical condition of soil-water mixture is denoted as its consistency: Consistency isthe resis- tance to flow, which is related to the force of attraction between particles and 1s easier to feel Physically than to describe quantitatively (Yong & Warkentin, 1966). Figure 1.19 shows the Uifferent consistency states of a mixture of water and fine-grained soil. The boundaries of these states, expressed in terms of moisture content, are termed the Atterberg limits (after Al- bert Atterberg, a Swedish agricultural sciemtist), and are defined as follows:Nature of Soils, Plasticity and Compaction 27 st PL uw Sole State !Somi Sold State | Paste State {Liquid State ee epee Moisture Content Figure 1.19. Consistency sacs Liquid Limit: The iguid timic LL is the moisture coment above which the soil-water mixture passes 10 liguid state. Ar this state the mixture behaves like a viscous fluid and flows under its own weight. Below this moisture content, dhe mixture is in a plastic state. Any change in moisture content either side of the LL produces a chanige in volume of the soil, Plastic Limit: The plastie limit PL is the moisture content above which the soil-water mixture asses t0 a plastic state, At this state the mixture is deformed to any shape under minor pres- sure. Below this moisture content, the mixture is in semi-solid state. Any change in moisture, content at either side of the PL produces a change in volume of the soi Shrinkage Limit: The shrinkage limit SL is the moisture content above which the mixture of soil and water passes to a semi-solid state. Befow this moisture coment, the mixture is ima solid state, Any increase in moisture content is associated with volume change but a decrease in moisture content does not cause volume change. This is the minimum moisture content that causes full saturation of the soil-water mixture. The volume remains constant as the mixture oes through the dry state to the SL moving from zero saturation to 100% saturation (Figure 1.20). On the wet side of the SL the volume of the mixture increases linearly with increasing moisture content. On this line the mixture is fully saturated. A decrease in moisture content ‘moves the state of the mixture along the broken line CBA. Using these limits the following in- dices are defined and used in the classification and description of fine grained-sils: Plasticity Index PY = LL—PL (mixture is plastic at this range) asp Activity = PY /Clay particles (%) «2 Liquidity Index 1, =(w-PLY PL (w= moisture content in the field) (1.33) ‘The Atterberg limits are used extensively inthe classification of fine-grained soils. Finding re- lationships between these limits and the engineering properties of the soil has been a matter of research for many years. Early researchers such as Terzaghi & Peck (1967) suggesied the di rect proportion of liquid limit and soil compressibility. Sherard (1953) reported similar behav jour while investigating the effects of index properties on the performance of earth dams ‘Whyte (1982) suggested a method based on extrusion forthe determination ofthe plastic limit and found that the ratio of strength at the plastte limit to the strength atthe liquid limit is ap- proximately 70, According to Skempton & Northy (1953), however, this ratio is approxi- mately 100. 4 comprehensive collection of equations relating compressibility indices and soil plasticity is reported by Bowles (1996). These relationships may be useful in guiding the engi reer in the early stages of a feasibility study before conducting extensive soil exploration and28 Soll Mechanics: Basic Concepts and Engineering Applications 4 Volume > Ls ° SL Moisture Contant Figure 120. Volume-moistre content eelationship. strength tests. The liquid limit for clay minerals may vary from 50% for kaolinite to 60% for illite and up to 700% for montmorillonite. Kaolinite and ilite may show moderate plastic lim- its of 25% to 399, whilst in montmorillonite the plastic limit can reach 100%. Note that eX- changeable cations such as Na and Ca may influence clay activity, and this has to be invest gated if these minerals are to be subjected to salty water. 15.2 Determination of the liguid, plastic and shrinkage limits of soils The Casagrande method for determination ofthe liquid lit. Te relevant apparats is shown in Figure 1.21 and complies with ASTM D-4318, AASHTO 89, BS 1377, AS 1289.3.1.1 and [AS 1289.3.12. Tis comprised of a brass bow! (A in Figure 1.21) that is hinged to a erank, ‘which, on rotation, causes the bow! tobe lifted and dropped 10 mam onto a hard rubber base — 2 counter records the number of rotations (blows). A pat of sili placed in the cup covering approximately 2/3 of the area, and js grooved into two pieces with a standard grooving tool (B). The cup is apart of a 54 mm diameter sphere and has a maximum depth of 27 mm. The original grooving tool was later replaced by the ASTM grooving tool ©) to overcome dfficu- ties in grooving silty and sandy soils. The test specimen is made from 250 g of soil passing the 425 ym sieve. Water is added in increments anid mixed with the soil until the soil becomes a thick homogeneous paste. Ici then left at oom temperature to cure for 12 hours. par of this mixture is placed in the Casagrande eup and jevelled parallel to the base to make a depth of approximately 10 mm. Afier closure of 10 mm of the groove as a result ofthe lifting and $0% The equation for the A line which separates the clay fraction from the silt and organic matter is given by: Pe 0.73(LL 20) (1.38) ‘Table 1.6 shows the Unified Soil Classification System for engineering purposes according to ASTM D-2478. Use of this standart results in a single classification group symbol and group name except when a soil has 5 to 12% fines or the state ofthe soil inthe plasticity chart falls into the cross-hatched area. In these cases, a dual symbol is used, eg. GP-GM, and CL-ML. Any soil on the borderline of two group symbols is identified by both group symbols separated bya slash, eg. SCICL, CLICH, and GMISM. Borderline symbols are particularly useful when the LL is close to 50%. These soils can have expansive characteristics and using a borderline symbol alerts the user of this potential (ASTM D-2487-1998).Nature of Soils, Plasticity and Compaction 38 60 Ang on Mit oroH ou cu. fA or OL =a” = o Lh 0 10 20 90 40 50 60 70 80 90 100 110 Liu timit (2) Figure 1.27, Plasticity chart (ASTM D-2487). Table LS. Fits and second leer of group symbols First Jeter of group symbol Second leer of group syibol (Coarse grained soils Gr gravel, sand 1. Well graded, P; Poorly graded Fine grained sols Mf sil, C:elay E: Low plasticity. 21 High plasticity Organic soils ° 1: Low plasticity, 14 High plasticity Highly organic soils Pr [No sovand lewer 1.63 Description, identification and classification of soils (AS 1726) The Unified Soil Classification System with minor modifications is used. Coarse-grained soils tre defined as materials in which more than half of material less than 63 mm is larger than 0.075 mm. The CL group of the original system (LL < 50%) has been divided into two groups ‘of CL (inorganic clay of low plasticity) with LL < 35% and CT (inorganic clay of medium plas- Uoity) with 35% < LL < 50%. Field idemtification procedures are performed on particles finer than 0.2 mm after coarse particles are removed by hand, Dry strength (crushing characteristics). A pat of soil is made to the consistency of putty and is left in the oven or sun or air-dried until completely dry. By crushing it between the finger, its strength can then be assessed but only after developing experience from applying the method lo different soil samples with known strength characteristics. Dilacaney (reaction 10 shake). A mixture of soil and water (approximately 10 em’) is made ‘with sufficient moisture to make the mixture soft bu not sticky, The sample is then placed in36. Soil Mechanics: Basic Concepts and Engineering Applications ‘Table 1.6, Unified Soll Classification System (ASTM D-2487). Major division Group symbol | Finer” than 75 pm ow <3% Gravelly sols 50% of coarse frac- oP. <3% tion larger than sieve > 1 eet cM 12% 236mm Coarse grained oe 28 soils s Ww eae | a Sandy soils SOF of coarse frac SP. <5% tion Finer than sieve No “ aa ne sc > 12% ML Silks and clays ce LL < 50% Finegan on 50% <75 um 7 Sits and clays cH LL> 30% on Highly organic soils Be i ‘>For sols with S127 passing the No. 200 (75 wm) use a dual symbol, the open palm of one hand and is shaken rapidly, striking vigorously against the other hand, ‘Water will appear on the surface of the sample, but will disappear ifthe shaking is stopped and the sample is squeezed between the fingers. The rapidity of appearance of water during the shaking process and of its disappearance during squeezing assist in demtfying the character of the fines in the soil. For instance, very fine clean sands give the quickest and most distinct ce action whereas plastic clay has no reaction. Inorganic silts show a moderately quick reaction, Toughness (consistency near plastic limit). A specimen of soil approximately 10 cm° in vol ume is moulded to the consistency of putty and then rolled out by and on a smooth surface or bberween palms into a thread about 3 mm in diameter. The thread is then folded and re-rolied repeatedly. The sample gradually loses moisture and crumbles when the plastic limit is reached. The pieces of the crumbled thread are lumped together with a slight kneading action which should be continued until the lump crumbles. The tougher the thread near the plastic limit and the stiffer the lump when it finally crumbles, the more potent is the colloidal clayNature of Soils, Plasticity and Compaction 37 ‘Table 1.6 (continued). Unified Soil Classification System (ASTM D-2487) ‘Typical names Laboratory classification eiteria Wel graded gravels, Good gravelsand mixtures LeCe<3 Poorly graded gravels, gravel-sand mixtures [Nat meeting gradation requirements for GW. Silty gravels poorly graded yravel-sand-sik Aeterg Tinie below A-- | Above Aline rxures line or P< With4 7 dus symbols Well graded sands, gravelly sands cu>6 1ecce3 Poorly graded sds, gravelly sands Not meting gradation requirements for SW Sily sans, poorly graded sand-it miares Aueerg limits low A- | Above Astine | Tine or Pet with be Pre? Cavey sands, poorly graded sand-clay minsures | Atterberg limits above Af requires tine with PU 7 dual sinbols Tnorgani sits very Fine sands, sty or clayey fine} Plasisiy chan sands with sigh plas ock our Inorganic clays of low o medium plastic Plasiciy chart gravelly lays, sandy lays sity lays, lean lays Organic sits and oetanie sit- 12525 Canbe moulded by light finger pressure Firm 325550. Cane moulded by strong finger pressure salt >50 100 Cannot be moulded by Angers: canbe indented by thurs, Very stiff 3100200 Can he indented by thumbnail Hard 3200 Can be indented wit difficulty by thumbnail Table 1 8(¢) Consistency terms for non-cohesive soils Term Density index Very loose 20.15 Loose >0:15<0.35 Mediumdense 30.5063 Dense 30655085 Very dense > 0.85 1.64 Summary of British soit classification system for engineering purposes ‘The British soil classification system is shown in Table 1.9. The qualifying terms are simitar ‘o the Unified Soil Classification System with more subgroup symbols for both coarse and fine soils, Poorly graded gravel and sand have been divided into two group symbols by addiog Pu and Pg to represent uniform and gap-graded soils respectively. n coarse soils subgroups have been defined by adding a third letter (M or C) to well graded or poorly graded gravel and sand 70 wy 60 a ‘atina Me Cita, ot cL _ c= ° om © 10 20 30 40 50 60 70 80 90 100 110 120 Liguie limit (2%) Figure 1.29. Plasticity chan according wo British sil classification system40 Soil Mechanics: Basic Concepts and Engineering Applications “able 1.9. British so classification system for engineering purposes (BS 5930). Soi groups Subgroups and aboraory Meneaon Grav and sand nay be defined as sndy rove | Group Subs Syminl ‘Fins LL fad poely and und pop on Sigiyanyerdaey [6 GW GW os 2 GP Ghar Grvets EB os es Siy geval GF Gu Gus, sO%at | Gr 1s conseina, | Clay gel oc __Gweore teal of Vary sy grave OF GM GMi.ac. waved size | Very clayey gravel Gc Gch Se Gas ita | ca Sas sey finer than GCE 06mn Signy siyorciaey | 5 SW SW os sand Sr Srusre sands mands [Sty sand SF 5M sWa,.s™m 5. wisn | cuayey sand Sci eswesncs( coarse [Very sity sd [SF SW SME, teralisot | Very elyey sand se SL sand size 1 diner tan os mn) pe sce Gravely at Fo WG MiG. 0 Gravelly of | Gravelly clay ce CLG 3s sandy sity ae 3550 : “ha cays atte 50-70 Fine soils | “35st ac an rmovethan | 639 nes ov nn ss finerthan Sandy at MS MES. as 06 mn Sandy clay cs _ cise Sil Msi a Sitsand | Ctay c ch <35 clays a 3550 65% to cH 50-70 100% fins a 7050 & 330 Gani wil Deserve ua Oslin wo say Bwup or suberoup sya Par 7Nature of Soils, Plasticity and Compaction 4 Fine soils, and the fine fraction of coarse soils, are shown by F, and the subgroup symbols of fines and fine fraction of coarse soils are: L (low plasticity}, / (intermediate plasticity), 1 (high plasticity), V (very high plasticity), £ (extremely high plasticity) and U (upper plasticity range) as the second or third letter (Figure 1.29 & Table 1.9), Field identification is based on particle size, grading, structure, compactness, dry strength, dlatancy, consistency and weathering (BS 5930, 1981). 17 COMPACTION LL Dry density-moisture content relationship ‘Compaction is the process of reducing the air content by the application of energy to the moist soil, Compaction inreases the number of particles within a specific volume thereby increasing the shear suengtb. Consequently, any displacement due to external loading and surface setle- tment thereby reduced because of the denser structure. Compaction of soil samples in the labo- ratory iscaried out in standard cylinders. Energy is applied by a hammer of standard size and tmass dropping freely from a standard height on a layer of the sample inside the cylinder. Each specimen is made in 3 or 5 layers witha specified numberof blows depending on the codes and method of testing. Inthe Fel, the enery i applied by means of differen ypes of rollers. Standard codes specify this energy per unit volume of soil. The tem compactve effort is used to describe the energy given toa Unit volume of sil. Compaction increases the number of pat- ticles per uit volume and dry density is used to indicate the degre of compaction Dry density Moisture content Figure 30. Soil behaviour under a speifiod compactive eff.42. Soil Mechanics: Basie Concepts and Engineering Applications CCompactive effort is most effective if a uniform mixture of soil and water is used. Dry density increases progressively with increasing moisture content co a maximum under a constant com- pactive effort. The moisture content corresponding (0 the maximum dry density is called the ‘optimum moisture content (OMC). Increasing the moisture content beyoné this valve reduces the dry density. The common and narural behaviour ofa soil-water mixture under a specified compactive effort is shown in Figure 1.30. The curve representing the experimental results is termed the dry density-moisture content curve. Maximum dry density and the relevant opti- ‘mum moisture content are obtained from this plot Lambe (1958) explained the shape of the dry density-moisture content curve in ferms of the development and characteristics of the diffuse double layer with increasing moisture content. At low moisture contents and on the dry side of the optimum moisture content there is insuffi cient water to form a double layer and che structure of the fine particles is of a flocculated type. The increase in moisture content causes the formation of the double layer and randomly distributed particles become more arientated. This produces a lubrication effect between the large particles causing them to slide over each other as the compactive effort is applied. At the maximum dry density the specimen of compacted soil has a high degree of saturation. Further increase in the moisture content has a dilution effect and dry density decreases while the de gree of saturation remains approximately constant. The effects of compactive effort and mois ture content are insignificant on granular soils that have little or no fines. Test points on the wet side of the OMC may not be established by a standard compaction test due to loss of water during compaction. A method combing vertical vibration and water gives satisfactory results, ‘The theoretical relationship between dry density, air content, and moisture cantent is ex- pressed by Equation |.18. Figure 1.31 shows the plot ofthis equation for Ay = 0% (termed the Dry density Moisture content Figure 131. Theoretical curves and experimental compaction curveNature of Soils, Plasticity and Compaction 43 Dry density ‘Moisture content Figure 1.32. The effec of compactive effort on maximum dry density ad optimam moisture content sero air curve), 5% and L0% for acertain value of specific gravity of the solids, In practice the Compactive effort cannot de-air the voids completely. thus the dry density-moisture content curve is always on the left side of the zero air curve. The descending part of the compaction curve usually has a minimum air content of 3%-54% If the descending part intersects the 2er0 air curve, either the test data or the specific gravity of the sofids is wrong. In general the right side of the zero air curve is an impossible state for experimental data points. It is always con- venient to plot the air void curves of 54, 109, ec. to understand the development of compac: tion and the progress of the degree of saturation as moisture content ingreases. The standard ‘energy applied to soil speciniens in the laboratory is 592 kNm per I m” of soil and was sug- ‘gested by soil soientist RR. Proctor during the 1930s. In time, the need for having a denser and stronger soil was increased and the Proctor method was modified by increasing the com pactive effort to 2695 kNm per Lm” of soil, Increasing the compactive effort increases the dry ensity and reduces the optimum moisture content, as shown in Figure 1.32. Curve C has higher compactive effort than curve B. The inerease in maximum dry density is believed to be a a result Of greater orientation of the fine particles under the increased compactive effort Figure 1.33 provides an estimation of the dry densities and optimum moisture contents for the fine and coarse grained soils defined by the Unified Soil Classification System. 1.7.2 Standard laboratory compaction test TThe moulds used for the compaction tes, with internal diameters of 101.6 mm or 105 min and 152.4 mm, and the hand-operated rammer are shown in Figure 1.34. Each mould has a remov= able collar assembly with a decachable base plate. The mould type is selected depending on the44 Soll Mechanics: Basic Concepts and Engineering Applications 23 Fy Dry density (Mgim*) ° 10 20 20 Moisture content (4) Figute 1.33. Typical compaction curves for couse and fine grained soil size fraction of the particles in accordance to the relevant standard code. In the standard Proc- tor test compaction of the prepared sample is conducted in 3 layers with a rammer of 2.5 or 2.7 kg. The number of blows per layer is 25 or 27 for the small size mould and $6 of 60 for the Figure 1.34. (@ ) Proctor mould, (b) modified Proctor mould, ()ramumer (Wykeham Fatrance),Nature of Soils, Plasticity and Compaction 45 large size mould. This small variation in the number of blows and the mass of rammer is due to minor differences in the diameter and height of the moulds in the different codes. In the ‘modified Proctor test the Jayers are increased to 5 and a heavier rammer of 4.5 kg oF 4.9 kg is used. Mechanical versions of hand ramimers, which comply with most codes, are also avait- able. Two common types are portable, lightweight vibrating rammer and the stationary auto- matic compactor. Table 1.10 shows the adopted standards for the compaction test. Details of the preparation of samples, treatment of oversize material, moisture content intervals, and plot- ting of the experimental results are in the designated codes in the first column of Table 1.10. ‘The butk density of the soit in each test is calculated by dividing the mass of the compacted soil by the volume of the mould. After removing the compacted soil from the mould, a repre sentative sample from the whole height of the specimen is obtained for the determination ofits moistufe content. Dry density is calculated from Equation 1.11. The testis repeated 6 times, (over a range of moisture contents to establish the dry density-moisture content curve. Minimum and maximum dry density of cohesionless materials. Moisture content has little or ‘no influence on the compaction characteristics of granular soils (except when the soil is fully saturated). Their state of compaction can be obtained by relating the dry density to the mini- ‘mum and maximum dry densities obtained in the laboratory. The test technique requites a vie ‘oratory rammer or table and a pouring method. Accordingly the density index (Equation 1.13) may be calculated in terms of minimum and maximum dry densities if a specified density in- dex is required. Combining Equations 1.13 and 1.14 Cmax =€ = Pax Pa Paton 1 (139) Pena Emin Pd Pagan Penn Minimum dry density is obtained using an oven-dried material and is referred to as the dry placement method. In this method a mould with known volume and a pouring device are e- lected according t0 the maximum size of the soil particles, as recommended by the relevant Table 1.10, Adopted standards far compaction ts. ‘Mould: Diameter (oun). Rammer No of hers Standaed designation Height (ram), Drop (aim). No. of blows per layer Volume tem) Mass (ka) ASTM D-696, 1016, 1165.98 305.25 3.28 (Proctor) 13241165. 2100 457.45 525 ASTM De15S7. 101.6, 1165. 988 305.25 ‘AASHTO 799, 7180 152.4116.5,2100 5745 (waditied Prosi) BS 1377 & BS 1924 105.0,115.5.1000 300,25 (Proctor) {modified using AASHTO) 105.0.115.5.1000 45045 AS 1289.5.1.1 108.0.115:3,1000 300.27 . (sandard) 132.0. 132.5, 2400 0 AS 1289.521 195.0.115.5.1000 450,49 5.25) (wodified) 152.0, 1325, 2400 5.10046 Soil Mechanics: Basic Concepts and Engineering Applications code. To prepare a uniform sample with minimum segregation, the mould is fitted with mate- Fial, using a funnel or by means of a scoop, in a steady stream, ftee of vibration and any other disturbances. After levelling the material the mass of the mould and its contents is measured for the calculation of dry density. The testis repeated several times until a Jowest value forthe mass inside the mould is obtained. Maximum dry density is determined by the wet placement ‘method using either a vibratory rammer (BS 1377) of a vibratory table (ASTM D-4253, D- 4254, and AS 1289,5.5.1). A diagrammatic section (and view) of a vibrating table permitted by AS 1289.5.5.1 is shown in Figure 1.35. Generalty the apparatus comprises a vibratory table ‘which oscillates ata nominated rate (say 3600 vibrations per minute) and is equipped with two to five moulds with nominal volumes of I litre co 5 lites, A surcharge of 5 t0 70 kg is neces- sary to provide vertical support during vibration. The saturated material is poured into the ‘mould by means of funnel or a scoop until it is filled and overflowing. If necessary, water is added to ensure full saturation, The mould is vibrated during the filling process with smaller amplitude of vibration. The surcharge is lowered to the surface of the sample and the vibrator is set to a required amplitude for approximately 10 minutes or until the settlement of the sur- face has stopped. After removing the surcharge and levelling off the material, the contents of the mould is placed in an oven to dry wo a constant mass. The maximum dry density is then calculated by dividing the dry mass by the volume of the mould 1.7.3 Compaction in the field ‘Compaction in the fiekd may be divided into the two categories of deep and surface compac- tion depending on the thickness of the compacted layer. Deep compaction is based on vibto- compaction techniques, dynamic compaction and the use of explosives. A detailed study of deep compaction is beyond the scope ofthis book, however, due to its ineteasing applications in surface and near-surface compaction, a brief description of éynamnic compaction is included atthe end of this section Table Vibrator ‘Concrete floor Figure 1.35, Apparaus for vibratory compaction.Nature of Soils, Plasticity and Compaction 47 Surface compaction includes the compaction of layers in 0.15 m wo 0.3 m layers, however, by. using vibratory techniques and dynamic compaction, the depth of the compacted layer may be Increased to 1.0:m oF more ia granular soils The maxisnum dey density obtained in laboratory conditions as achievable inthe field by us- different types of compactors that employ different concepts inthe application of the com: pactive effort. Degree of compaction isa means af comparing te tield density with luhoratory results and is defined ay dhe ratio of the dry density inthe field to the maxitnam dry densety obtained in the laboratory. In nest construction works the degree of compaction is specified 95, 954% or more, usually without any designation of the prethod of compaction. The hulk snd dry density in the field are measured using one of the methous described in Scetion 1.3, Ifthe compacted sos] does aot expand or consolidate after the liek compacvon is completed, its dry nsity remains constant regardless of the magnitude of its moisture content, In the field, the moisture content nay vary from zero fo a maximum value whieh is the intersection of a hori zontal ling drawn trom the maximurn dry density on the experimental curve with the theoret cal 2e10 air curve. Stability studies of eomystcted soils show that a compacted Si} may undergo compression ar consolidation duc to self weight cesulting a substantial surface settlement or failure Lawton et at, 1989: Brandon ct al., 1990), In order to predict the stability of 2 com pacted fill itis necessary to perform Ishorstory strength and eonsolidatinn fests the con pacted specimens, ‘Compaction equipment commns used in the Fie are Sheep's foot roller. This soller gonsists of a steel Geum with studs or Feet 40 19 1DA crn in ares pjecting some 200 to 250 mm. Ts either pushed ov pulled by a tractor or is ylf-propelled Unit. A sheep's ft roller is conveniens for cohesive soils and applies gompactayn by 3 corny mation of tamping and kneading. Contact pressines upplied by the projected studs are high and vary from 700 to 4000 KPa The thickness of a compacted layer is wf the order of 0.15 m 10 03 n. Bond between tw compacted layers is very strong due 10 the action of the studs. The type of the roller (light or heavy) and the numher of passes must be determined hy the site engineer after conducting a suitable number of field density Preunaic gre roller. This type of sller is either wowed or isa self-propelled unit commonly equipped with an auesde fated box for weight adjustment. 1 usually has to axles with 2.0 5 Gor ore) tyres per axle and the hack avle is designed to overlap the area compacted by the Iront axle, The wheels are semetinnes equipped with a wobbling mechanism to cover the loser surfaces. The tyres have Mat trads and theselore cannot apply sideways displacements to the soil particles. This compactor iy suitable for many different soils and applies vompaction by kneading only. Compacted layers vary fanz 0.15 m to 0.5 m depending on the mass of the compactor. Usually only a few passes are necessary 19 acbieve maninur dry density. Vibratory rollers. These aré applicable to many 1ypes of souls but are primarily designed for granola materals, In the smcth drum vibrators the vibrating mechanism consists of eee cally arranged weights (Terzaghi ard Peck, 1967) which in rotation apply’ vertical vibration to the drum. In pneumatic-tyred vibratory compactors the vibeating mechanism is applied to48 Soil Mechanies: Basic Concepts and Engineering Applications the axle of the unit. In the vibrating plate compactors a plate or plates that are in contact with the ground surface are subjected to vibration: however the depth of influence is not as great as smooth drum and pneumatic tyred vibrators. These latter two machines compact the Soil to a depth of up to 1.0 m, but an effective depth of 0.3 m is generally accepted ifthe soil consists, (of silt and clay. Vibratory rollers are available in towed or self-propelled units. Dynamic compaction. In this method a weight of 100 to 200 kN is dropped from a height of 1S to 40 musing crawler cranes. Higher weights of 500 KN may also be used if a deeper compa tion is required. Establishing the number of passes and drops per pass needs a careful in-situ study. Experience has shown that granular materials require more energy than fined grained soils. In the latter the applied energy increases the pore water pressure and adequate time is needed for dissipation of any excess pore pressure. In the laboratory dynamic compaction is ‘modelled using the centrifuge method in which the force can be increased up to 100g (g = 9.81 rmisec"). Here dynamic compaction is provided by dropping a steel rod several mm in diameter and approximately 100 mm in length on the surface of the sample. The centrifuge machine for this purpose should be capable of accepting & 0.5 m cube of soil 1.74 Application of compaction test in soil stabilization rechniques ‘The standard compaction test is also used in sol stabilization to investigate the effect ofthe 195 19 Zero air voids (soi) Untreated so Vv 188 Diy density (Magic) oe 4% Quieklime 16 lab mix Ss 8 10 12 14 16 18 20 22 24 Moisture content (4) Figure 1.36, Compaction curves fo a clay sol stabilize by Time (Aysen eta. 1996).Nature of Soils, Plasticity and Compaction 49 stabilizer on the maximum dry density and optimum moisture comtent. Depending on the type of the material and the stabilizer (e.g. lime or cement) the maximum dry density may increase or decrease slightly or moderately causing variation in the optimum moisture content in the reverse direction. In these cases the strength of dhe treated material is not represented by the ‘maximum dry density and the increase in strength is mastly due tothe formation of the bonds between the particles. Investigating the optimum moisture content is vital to ensure that sufti= cient water is available for hydration, An example of lime stabilization of clay material is shown in Figure 1.37 (Aysen etal. 1996). Example 19 ‘The following results were obtained from a standard compaction test. Determine the optimum moisture content and maximum dry density. Plot the curves of 0%, 5% and 10% air content and give the valug of air content at the maximum dry density, The volume of the standard mould is 1000 em? and Gy= 27. Mass(g) 1768 1929 2014 2178 2106 2052 2007 et Gee to ae ae ie ‘Solution: Catcutate dry density for each test aq tablate the results, Sample calculation for w = 4% p=MIV 768 Mg/m’, py =l(I + w) =1.768/(.+0.04) = 1.70 Mein Pegi) 170 182 192 198 188 180 1.73 ON es 6 ee 0 ig ee To plot the constant air content curves use Equation 1.18 and tabulate the resus. The results are shown in Figure 1.37, Itis seen that the maximum dry density is 1.98 Mp/m’ and optimum moisture content is 10%, we i eet Bee 222 213 20s 196 188 181 175 (gem) 2M 202 Lod 186 179 172 166 200 192 1st 176 169 163157 Calculate the air content at maximum dey density using Equation 1.18: Py 3.) =1.98, 2.71.00 = 4, )AL+0.1x2.7) > A, = 0.0686 = 6.86% Example 1.10 For a soil with G,= 2.65, che following results wer obtained from a standard compaction test: ‘Sample number C sas Moisure content(%) 162 167 190 204 21 Dey densiy (Mgim') 1.580 1620 1.687 1605 1.50 Soil Mechanics: Basic Concepts and Engineering Applications 25 24 N=8% 23 22 24 2 19 18 Ww 16 18 Dry density (Mgin®) 2 4.6 8 10121216 182022 Moisture content (2) Figure 1.37. Example 1.9. To obtain two extra points close to the maximum dry density two standard compaction tests ‘were performed and tabulated. (a) Determine Pax and Wop and plot the 2er0 air curve, (6) & (0.3 m layer ofthis soil is compacted in the field to its maximum dry density. After some time the natural moisture content of the layer is measured to be 16%. How much water in terms of mis needed to make the tayer fully sarurated”? ‘Sample number 67 Mass of mould + compacted wet soil (kg) 8.966. 8.978 Mass of mowid (he), 70 20 ‘Volume of mould (em?) 410001900 Mass of subsample taken from mould (g) (788 1558 Mass of subsample afer drying (2) 123 118 Solution: (a) For sample number 6 ; =A IV = (8.966~7.0)/1000 = 1.966 Mem", w= (178.8~152.3)/152. 2g =Plll +) = L.966/(1+ 0.174) =1.675 Men For Ay = 0% use Equation 1.17; results are tabulated below and are shown in Figure 1.38 from which pymay = 1.68 Men? and wope = 17.74. =0.174, (©) Py = 6,0, J+ WG,) = 168, 265%10/ 1+ 265i) © L68— w 2 218%, Aw= 218160 = 5.8%, AM, =1.68X0.058=0.097 Mg/m? or AV, =0.097 mim", AV,.(0.3m layer) =0.097x0,3=0.029 m'/m?, 4, =0,029/1.0=0.029 mm’,Dry density (Mgien*) Nature of Soils, Plasticity and Compaction SL 18 Wopt 17.7% 15 AS iG 21 as Moisture content (%) Figure 1.38, Example 1.10. Tet Wo pg(Mpm) Tet Wg (Mglan) points (%) A= 0% Test poms (%)_A=0% 6 1621834 1380 3 19.01 167 1837 1620 4 = 204 1720 174 184 1673 526 1085.58 1821788 1.670. PROBLEMS 18 Ld 13 The following data are given for a specimen of clay soil: M = 221 g, Mc 27, $, = 25%. Determine the total volume and the porosity ofthe specimen, Answers: 171. em’, 72.3% Dry soil (G; = 2.71) is mixed with 16% by weight of water and compacted to produce cylindrical sample of 38 mm diameter and 76 mm long with 6% air content. Calculate the mass of the mixed soil that will be required and its void ratio. Answers: ITT.7 g, 0.525 28 g. Gs Daring afield density est 1850 g of soil was excavated from a hole having a volume of 900 em’. The oven-dried mass of the soit was 1630 g. Determine the moisture content, dry density, void ratio and degree of saturation. G, =2.71. Angwers: 13.5%, 1.81 Mgim?, 0.496, 73.8%52 Soil Mechanics: Basic Concepts and Engineering Applications ry Ls 16 18 A soil specimen has a moisture content of 21.4%, void ratio of 0.72, and G, = 2.7. De termine: (a) bulk density and degree of saturation, (b) the new bulk density and void ratio, if the specimen is compressed undrained until full saturation is obtained. Answers: 1.905 Me/m’, 80.1%, 2.08 Mg/m*, 0.578 ‘The moisture content of a specimen of a clay soil is 22.4%. The specific gravity of the solids is 2.71. (a) Plot the variation of void ratio with degree of saturation and calculate the void ratio, and the dry and wet densitics at $0% saturation, (b) a sample of this soil with initial degree of saturation of 50% is isotropically compressed to achieve a void ra- tio of 0.55, Calculate the volume change in terms of percentage of the initial volume. How much of this volume change is due to the outward flow of water ftom the sample? Answers: 1.214, 1224 Mg/m?, 1.50 Mg/m?, 30.0%, 2.6% The results of a particle size analysis are: Sieve sie (mm) Mass retained (g) Sieve size (mum) Mass retained (2) a 00 a5 30 375 26 236 1b? 190 8 Lis 46 132 18 06 3 95 2 8. 3M 67. 8 0.075 30 ‘The total mass was 469 g. Plot the particle size distribution curve and determine the co- efficient of uniformity, coefficient of curvature and soil description. Answers: 13.3, 32, GW ‘The following data were recorded in a tiquid limit test using the Casagrande apparatus, Determine the fiquid limit ofthe soil. Classify the soil assuming PL = 19.8%. Nuniberof Massofean(g) Massofwe Massof dy blows soiltcan (soil scan @® 8 1180 36.05 2918 16 1320 34.15 2860 2 14.10 3695 3116 0 © 332% Deu 4.2%, CLL ‘The recorded data in aliquid limit test using the cone penetration method are as fotlows, Determine the liquid limit ofthe soi. Cone peneuation mm) 141183. 22.4 272 eee el eel Answer: 47.0%Nature of Soils, Plasticity and Compaction $3 9. The maximum and minimum void ratios for a sand are 0.805 and 0.501 respectively. The field density est performed on the same soil has given the following results: p= 1.81 Mgim’, w = 12.7%, Assuming G,=2.65, compute the density index, Answer: 0.50 1.10 The following results are obtained from a standard compaction test Mase of compacted soil(g) 19205 20515 21385 21870 21200 20815 Moisture content (%) Ho 1 2s Be las 163 The spoyitic gravity of the solids is 2.68, and the volume of the compaction mould is 1000cm*. Plot the compaction curve and obtain the maximum dry density and aptimum ‘moisture content, Plot also the 0%, 5% and 10% air void curves. At the maximum dry density, calculate the void ratio, degree of saturation and air content. Ifthe natural mois- ture content in the field is 11.8%, what will be the possible maximum dry density ifthe soil is compacted with its natural moisture content? Answers: 1.907 Mg/m*, 13.0%, 0.407, 85.6% 4.1%, 180 Mgim* 1.9. REFERENCES ASCE. 1962, Symposium on grouting, Jowmal SMEE, ASCE 87(SM2) 1-145. ‘ASTM D-2487. 1998. Standard classficaion of soils for engineering purposes (Unified Soil Classifica. tion System. American sociery for testing and materials. West Conshohocken, PA. Atkinson J. 1993. Am introduction to the mechanics of soils and foundations. London: MeGraw-Uil Australian Standaed.1993, Geotechalcal sie investigations. 3° edition. Austalia NSW: Standard Asso- ciation of Australia Aysen, A. Ayers R., & Ricks, J. 1996. Soll stabilisation with cement and sluminium hydroxide, Re ‘search Report No CEOS. ISBN 0909756295. Toowoomba: The University of Southern Queensland. Bohen, H.L, MeNeal, BLL. & OConnor.G.A. 1985, Soil chemistry. New York: John Wiley. Bowles, JE, 1996, Foundation analysis and design. New York: McGraw-Hill Brandon, T.L., Duncan, J.M. & Gardener, WS. 1990. Hydrocompresson settlement of deep fils. Jour nal SMPED, ASCE 116(GT 10): 1536-1548 Brewer, R. 1964. Fabric and mineral analysis of sols. New York: John Wiley & Sons Brewer, R.& Sleeman,.R. 1988, Soil structure and fabric. Australia, Adelaide: CSIRO Div. Soils. British Standard $930. 1981. Code of practice for ste investigation. London: Briish Standanis Insti ton. Casagrande, A. 1948. Classifications and identifteations of soils, Translated: ASCE 113: 901-991 Craig. RP. 1997, Soil mechanics. 6" edition. London: E& EN SPON. FitzPatrick, .A. 1983. Soils, their formarion, classifications and distributions. London: Longman. ookes, PG, & Vaughan, PRR. 1986. 4 handbook of engineering geomorphology, Sumy: Surry Univer- sity Press Isbell, RF. 1996. The Australian sol clasification. Australia, Collingwood: CSIRO Publishing Jenny, H. 1941. Factors of soil formation. New York: MeGraw Hil,54 Soil Mechanics: Basic Concepts and Engincering Applications ‘Lambe, TW, 1958. The engineering behavior of compacted clay. Joumal SMFE, ASCE 84(SM2), May 1988, Lawton, F.C, Fragaszy, RJ. & Hardcastle 1H. 1990. Collapse of compacted clayey soll, Joumat SMFED. ASCE I15(GT 9): 1252-1267. MeLaren, R.G. & Cameron. K.C. 1996, Ion exchange in soils and soil aciity. Soi science: sustainable ‘production and environmental protection. Auckland: Oxford Press. Moore. C.A. 1970a. Suggested method for application of X-ray diffraction of clay srectural analysis «0 the understanding ofthe engineering behavior af soils. Special procedures for testing soil and rock for engineering purposes. Philadelphia: ASTM publication, Moore, C.A. 1970h. Suggested techniques for measuring the fabeic of engineering soils Special prace- dures for testing soil and rock for engineering purposes. Philadelphia: ASTM publication. Ollier, C.D. 1975. Weathering. London: Longman Group Lid Rowe, P.W, 1962, The stess-dilataney relation for static equibbsium of an assembly of particles in con- tack. Proe. roy. soe. A 269: $00-527. ‘Shera IL, 1953. The influence of soil properties and constrection methods on the performance of ho- smogencous earth dams. US bureau of reclamation. Tech. Memo (645). Skemplon, A.W. & Northy. R.D. 1983, The sensitivity of clays. Gearechnique 3(1}: 3038, Sposito©. 1989. The chemistry of sos. Oxford: Oxford Univesity Press. ‘Tan, K.H. 1994, Environmental soil science, New York: Marcel Dekker In “Terzaghi.K. & Peck. R. B. 1967. Sil mechanics in engincering practice. New York: John Wiley White, RE. 1979. Iuroduction 10 the principles and practice of soil science. Melbourne: Blackwell Sei- ene ‘Whitig, LID. 1966, X-ray diffraction techniques for mineral Identification and mineralogical composi- tion. In, Methods of rll analysis, American society of agranomy monograph. Part 1, Chapicr 49. ‘Whyte. LL. 1982. Soil plasicity and strength: a new approach using extrsion. Ground engineering 151) 16-24 ‘YongR.N, & Warkentin, BP. 1966. Ineroduction to soll behaviour. New York: The Macrillan Com- pany.CHAPTER 2 Effective Stress and Pore Pressure in Saturated Soils 2.1 INTRODUCTION In a saturated soil with no water movernent, equilibrium exists between stresses in the solid and liquid phases. The stress related to the internal forces acting on the contact points of the particles is termed the effective stress whilst the stress within the liquid phase ‘or water is termed pore pressure. The combination of taese two stresses represents the fo tal stress at a point. When a saturated soil is subjected to external forces, the state of equi- librium is altered and changes the effective stresses and pore pressures from their initial values to new values. This increase in the pore pressure, called the excess pore pressure, dissipates in time depending on the drainage conditions and permeability of the soil. This, pressure dissipation, in turn, causes a reduction in the volume of the soil, which is pre dicted by a mathematical-physical model called the consolidation analogy 2.2. STATE OF STRESS AT A POINT DUE TO SELF-WEIGHT 2.2.1 The effective stress concept Yerzaghi’s concept of effective stress states that: su ay) ° ‘where 6 is the total normal stress at a point in a specified direction or plane, of is the ef fective normal siress on that plane resisted by the particles and wis the pore pressure act ing on the plane. To fully understand the concept of effective stress, consider a cylinder containing coarse granular material in a dry or a wet condition subjected 10 a force N’ (Figure 2.1(a)). Assuming there is no friction between the soil particles and the inner sur face of the cylinder, the entire force N’ is transmitted completely to the base. The average vertical stress at the base of the cylinder is the ratio of WV to the internal cross-sectional area of the cylinder. This stress is much less than the stresses ereated at the contact points As there is no pore pressure in the system the total stress and effective stress at the base of the cylinder are identical. Consider, now, that the cylinder js filled with water until the soit js fully sanurated and that a force U is applied through another piston as shown in Figure 2.1(b). The loading plate that carries force N” has a hole fe for the passage of water to en- sure a uniform pore pressure through the whole system. The fraction of the base area 0c~ cupied by the contact points between the particles and the base may be ignored without 3556 Soil Mechanics: Basic Concepts and Engineering Applications oe soil, (@ Figure 2.1, Mutation ofthe concept of effective tess significant error. Ifthe setf-weight of the components are negligible (in this example) then the total vertical stress at the bottom of the cylinder is the sum of the pore pressure caused by U and the eflective vertical stress caused by 4’. In this example the loading systems for the particles and the water are independent with no apparent relationship. However, real soil, M” and U may have the same source. changing the values of effective stresses and pore pressures as time related functions. Equation 2.1 satisfies the states of stress from the initial state theough to the final state. Figure 2.2 shows the average sectional area ab that is equivalent to the area of the base of cylinder in the example above. The effective normat stress on the plane ab is equal to the sum of the components of the forces F to Fy perpen- dicular to the plane ab divided by the sectional area represented by ab, 22.2 Bffeetive stress and pore pressure within unconfined and confined aquifers AA schematic illustration of the different types of ground water conditions is shown in Fig- ure 2.3 (Bell, 1993). In an unconfined ground water the surface of the water, known as the phreatic surface, water level or water table, is in equitibrium with the atmospheric pres- sure; that is, the aquifer is not overlain by a material of lower permeability. In a confined aquifer both the upper and lower boundaries are confined with impermeable strata, and the water may have sufficient pressure to rise above the overlaying stratum (artesian condi- sions). In a teaky aquifer, @ layer of low permeability material separates the confined and unconfined aquifers. This layer slowly transmits the water and creates a water table in the unconfined region and a piezometric level (the level of water rising in a piezometric standpipe) in che confined region. Figure 22, The concept of effective stressEffective Stress and Pore Pressure in Saturated Soils 57 ‘Ground Sudace } Ground surtace ' Ss ————1 Sasa" Watertable; {| Watertable | Uncontined aquiler —~ —_{ }Unconfined aquifer i Impermeable Stratum WA 1 Aqitard id Piezometrc level Confined aquifer tf Leaky aquifer '|_Impermeable Stratum, Impermeable Stratum @ o) Figure 2.3, (a) Unconfined and confined aquifers b) leaky aquifer. Assuming that there ig no movement of water within the voids of the soil, the effective vertical stress 0’ at depth z for three possible positions of the water table (Figure 2.4) is given by: In case b the water table is below the ground surface and therefore ty 23) ©, = Yoaty # Yay By)“ Yue = Ya? Incase c where the water able is above the ground surface: = (Y par + Vorb) Vl # ye) = 7S. which is identical to Equation 2.2 and shows that o’y at a point below the water table Consantreardes ofthe position ofthe water table above the ground surface, Fora sll section composed of layers each having a thickness of hi of = Bpe-u = SD hy e4) WT (case o) Ground surface Figure 24 Effective stress ina saturated soi58 Soll Mechanics: Basic Concepts and Engineering Applications ‘where ye is the effective unit weight of each layer. In dry and partially saturated layers Yer is the dry or wet unit weight; in sacurated layers with hydrostatic water pressure, ye isthe submerged unit weight. In the presence of seepage forces due to movement of the water through the voids the hydrostatic based equations become invalid and the effective stress ‘must be evaluated using the main definition given in Equation 2.1 Example 2.1 {A soil section is comprised of two layers with the following properties: Sg 1 (0 to 3 my: Pary = 1.8 Myim’, Pyox = 2.0 Mgfm’; soil 2 3 m to 8 m): Pygr = 2.1 Mglm’. The water ta- ble is 1S m below che ground surface. Plot the total vertical stress, pore pressure, and effective vertical stress over the soit section Solution: On the ground surface, 6; = 9.0, 1 = 0.0, 0'y =0.0. AtZ= 1:5 mG, =18x9.81X15 = 26.5kPa, 1 = 0.0, o'y = 26.5 kPa Ate~30m:0, =1.8x9.81X1.5 +2.0%9.81x1.5=55.9 kPa t= L0x9.81x1.5~ 14,7 kPa, of, =55.9—14.7 = 41.2 kPa Atz=80 mio, ~2.1X9.815.0+559 =158.9kPa, w= 10x9.81%6.5 = 63.84Pa, [58.9 63.8 =95.1 kPa, The results are presented in Figure 2 Example 2.2 A soil profile is shown in Figure 2.6(a). Plot the distribution of cotal vertical stress, pore pressure, and effective vertical stress up to a depth of 12 m. Solution: Calculate the dry and saturated densities in soil I: V, +¥, 1(L+ 2) and V, =e/(L+e) . for e = 0.6: Vs = 0.625 m°, Vy eV, (Vs, Thus: 375m’ Stress (kPa) ° 0 00 | isn | | 200 ° i wr 2 ‘Soil 1 3 =| Soil 2 g5 & 6 : 7 ® PPore presstre Effective Total 8 igure 2.5. Example 2.Effective Stress and Pore Pressure in Saturated Soils 59 Suess (kPa) © 80 100 150 200 250 300 50m = Sol2 — pyg=213 Mol? = 5 80m 8 SoS pag = 2.18 Mai? 120m @ w) Figure 2.6. Example 22. Pary = GX Pw x Va 2.65.x 1.0 x 0.625 ‘M-= 1.656 Mg + 0.375 m° x 1.0 Me/m> Par = 2.031 Mes’. ‘The results are tabulated below and presented in Figure 2.6(b). 656 Mei", 2.031 Mg, Depih(m) oy (KPa) _u (KPa) 0 00 00 2 32500 5 38.4 8 1550589 2 2405 981 Example 2.3 Resolve Example 22 assuming that: (a) the water table is at the ground surface, (b) the water table is 2 m above the ground surface Solution: Calculations are summarized in the table below and are presented in Figure 2.7. @ ©) Dene G, kPap_u (kPa) ov (KPa) oy (KPa) _w (KPa) _o%y (KPa) 0 0900 (00«198GSGSSCOD 5 90 506 119.3687 5 BS 838 18198 77 1301 2674 Note: effective stresses at both cases are identical60 Soil Mechanics: Basic Concepts and Engineering Applications Stress (kPa) © 50 100 150 200 250 300 ° 1 Total 2 EO) gua 3 4 2° Es Soil 2 a7 és ob pore Soil 9 10 | pressure 11 | aN tect wes 13 Law Figure 27, Example 23 Example 24 ‘The lower sand layer inthe soil profile of Figure 2.8(a) isin an artesian condition. Caleu- late (a) of» atthe top and base of the clay layer, (b) the height of water inthe standpipe for oy = 0 a the base of the clay layer, (6) the maximum depth of the proposed excavation Figure 2.816). Solution: (2) At2= 4.0 m: 0, = (.7%2.0+2.1x2.0)9.81 = 74.6 kPa, 0x9.81%2.0=19.6 kPa, 6, = 74.6 19.6 = 55.0 kPa, Standpipe 30m 0.0 m 5} 2.0m L_| {Sand Pay™ 1.70 Moi? Wr Psy =2.10Mgim® Sand 4.0m Clay Py =210Mgim? — Clay 80m Sand Send @ co) Figure 28, Example 24Effective Stress and Pore Pressure in Saturated Soils 61 At z= 8.0 m:o, = (L.7%2.042.142.0+2.14.0)x9.81=1570KPa, {t= 1.0x9,81x11.0=107.9 Pa, ¢, =157.0~107.9= 49,1 kPa (6) Atz=8.0 mo 0)~1=0, 51 == ISTO RPS, «= pux g x= 157.0 KPa ‘i= 157.0/9.81 = 16.0 m. Height above the ground Surface = 16.0 - 8.0= 8.0 m. (6) With the arangement of Figure .8(6) 0'y = 0)~ w= 0, oy =u, assume 2> 2.0 m, 6, = [2.1(4.0~ 2) +2.1x4]x9.81 = 1.0x11.0x9.81 = 107.9 kPa, 2= 2.76 m>2.0 m, 2.2.3. Negative pore pressure due to capillary rise Capillary rise results from the combined actions of surface tension and intermolecular Forces between the liquid and solids. Noting that the pressure on the water table level is zero, any water above this level must have a negative pressure. In soils a negative pore pressure increases the effective stresses and varies with the degree of saturation and by drying or wetting processes. The phenomenon can be demonstrated by immersing a glass tube of small inner diameter into water as shown in Figure 2.9. The capillary water has a hydraulic continuity with the water table and rises 10 a certain height where equilibrium between the weight of the water column and the vertical resultant of the surface tension forces is achieved. The height ofthe capitlary column i is defined by: hy =F cosOH 4d) as) Where T is the surface tension (0.074 Nim at 20°C), @ is the angle of surface tension with the vertical, d is the diameter of the tube, and 7 is the unit weight of water. The water surface in the tube (meniscus) is assumed to be a portion of a sphere. For pure water and a lass tube of very small diameter, 8 = 0, the meniscus is hemispherical and A 247 Hrd) 26) ‘The maximum negative pore pressure is AY SAT en Any application of Equations 2.6 and 2.7 to a soil raises questions concerning the tube concept, the definition of d and the limitations of the equation. The capillary process starts as water evaporates from the surface of the soil. The capillary zone is comprised ofa fully Figure 29. Cylindrical capillary tube62 Soil Mechanics: Basic Concepts and Engineering Applications saturated layer with a height of usually less than fi, and a partially saturated layer overlain by wet or dry soil. In the partially saturated layer the water and air within the voids are both continuous. Terzaghi & Peck (1967) suggested that the pore pressure in this layer should be treated as a negative hydrostatic pressure with its maximum expressed by Equa- tion 2.7. Negative pore pressure results in an increase in the effective stress and is termed soil suction. ‘The boundary between the capillary and the gravitational water is ill defined and may not be determined accurately (Bell, 1993). In the field, a borehole that can create a well is, a simple way of determining the water level, but suction measuring devices are also used, The diameter of the equivalent tube d largely depends on the particle size distribution of the soil and the assumed value of d = eDig has been generally accepted by soil scientists Estimation for the range of capillary rise can be made by assuming 0.2Di9 < d < Dio (Powrie, 1997), In granular materials (gravels and sands) the amount of capillary rise is negligible while in silty soils che water may rise up to several metres. The rise and fall of the warer table and gravitational drainage from the ground surface can affect the height of capillary rise and may create different types of unsaturated zones above or close to the capillary zone. Immediately above the level hc. the suction in the water exceeds the tensile strength of the ‘water and generates vacuum conditions and vapour pressure (Klausner, 1991). As the ele ‘ation increases beyond the capillary rise, suction increases and the soil moisture and de- gree of saturation decrease. Eventually continuous air voids with atmospheric pressure evolve with typical behaviour shown in Figure 2.10. Locally created capillary menisci be- ‘oveen the small particles causes an increase in strength due to the increase in effective stress. Simple examples include the stability of sand castles made by children at the beach or the ability of a moist sand layer to resist shear stresses induced by vehicular tyres. The apparent increase in the strength of a fine granvlar material disappears with full saturation or full deying. In the latter instance, the effective stress increases due (0 the reduction in the radius of the meniscus and the soil shrinks. The process continues until the shrinkage it is reached and soil volume becomes constam. The strength increases as the drying process continues and reaches its maximum in full dry conditions. This behaviour is re versible in fined-grained soils (Section 1.5), that is, the soil volume increases (after the shrinkage limi, decreasing the strength of the soit Ground surface. (4) Pore pressure Capillary tise {+) Pore pressure Figure 2.10, Suction profile (Klausner, 1991)Effective Stress and Pore Pressure in Saturated Soils 63 Stress (kPa) 30 10 10 30 80 70 90 110 190160 00 STITT Sal 1 sit 2 27m 38m El 4 — 50m & |e Sls Giay: G,=27, w= 95% : Clay oa m 8 ha a Pore pressure _Etfective ©) Figure 2.11, Example 25 Example 25 For the soil profile shown in Figure 2.11¢a plot the distribution of foal and effective tial stresses and pore pressure taking account ofthe capillary rise inthe silt layer. Solution: Calculate the dry and saturated densities of silt: py = 1.88(1+0.105) = 1.701 Ma/m’. Assuming V= 1m: ; V,= 1701 /2.65 = 0.642 mV, = 100.682 =0.358 a, Pay =1701 40.358%1.0= 2.059 Men Calculate equivalent tube diameter: e = 0.358/0.642=0.558, X Dyp = 0.558x0.02 = 0.01116 mm, d= 1.116 x 10° m, fig =AT I.gd = 40.000074/9.81X1,116x10°5 = 2.7m, Calculate the saturated density of clay eqssturted clay)= WG, =0.35%2.7 = 0945, V, =1MU+e) =14L+0.945) = 0.514 m’, Vy = eMl+e) = Pqe = 0.514%2.7 + 0.486%1.0 = 1.874 Mgfin? At epth 38-27 = Lm 4G, #1.88%9.81x1.1= 20.3 KPa, 0 6%, =203+265= 468 kPa. At depth 3.8 m 6, = 203+2059%9.81x2.7 = 74.8 kPa, « At depth 5.0m 0, = 748+ 2059%9.81x1 6, =99.1-118=87.3kPa At depth §.0:m 6, =99.1~ 1874x9.813 1945 /(1+0.945) = 1.0x9.81x2.7 6.5 kPa, OkPa, 0, =74.8 kPa. 99.1kPa, w= 1.09.81) 1.8kPa, 54.2 kPa, w= 1.0x9.81x4.2 = 41.2 kPa,64 Soil Mechanics: Basic Concepts and Engineering Applications N N 4 Pressure measurement device eS FES Valve @ o) © Figure 2.12, Conceptual model for consolidation, 0%, =154.2-41.2 = 113. 0KPa. The results are plotied in Figure 2.11(b) 2.3. STATE OF STRESS AT A POINT DUE TO EXTERNAL FORCES 23.1 Conceptual model for consolidation Extemal loading inereases the total stress at every point ina saturated soil above its initial value, The magnitude of this increase depends mostly on the location of the point and is estimated from appropriate stress distribution theories based on the mechanical properties of the saturated soit (Chapter 5). In clay soils immediately aftr the application of the load, the pore pressure increases to almost equal the applied load and the pressurised water tends to move towards the free drained boundaries. Thus, overtime, the excess pore pres- sure dissipates and the pore pressure approaches its inital value. The volume of the clay layer decreases to balance the volume of the transported water and remains saturated. Thus the applied load is gradually transmitted to the Solid particles and the associated vol- lume change causes the settlement ofthe layer. This phenomenon, known as consolidation, can be explained by the mathematical-physical model showa in Figure 2.12 Consider a cylinder with a spring supporting a piston and filled with water tothe top of| the spring, as shown in Figure 2.12(a). The piston seals the water but has smooth contact with the walls of the cylinder. initially a valve, which isthe only escape for the water, is closed and the system is in equilibrium, There is no force in the spring and no pressure in the water. Ia normal force Vis applied to the piston the water will resist the entire force since, because itis incompressible, there will be no deformation in the water and conse quently no deformation and therefore 10 force in the spring. IF the valve were to be opened for a specified period of time, the pressurised water drains through the valve (Figure 2.12(b)). As a result, the spring undergoes a shortening equal to the vertical displacement of the piston and the load N is resisted by both the water and the spring. IF the valve re- ‘mains open until the water pressure decreases to zero (Figure 2.12(¢)) the spring fakes the entire force applied to the piston. The time over which the force Nis transmitted from the water to the spring depends on the diameter of the valve, the volume of water inside theEffective Stress and Pore Pressure in Saturated Soils 65 cylinder and the elastic characteristis of the spring. In real soil the valve represents the voids between the solid particles and the spring represents the solid particles themselves, and the load Vis normally applied through « footing or a similar type of strcture. The pore pressure at every point within the Volume increases initially 10 2 maximum value ‘equal othe increment of stress at that point caused by the applied loading (excess pore pressure). A. gradual process of redistribution of load to the particles starts immediately after application of the external foad (consolidation). The process is associated with vol- ume change (equal tothe drained water) and surface setlement. At the end of this process the applied load is resisted totally by the soil particles. In unloading the volume change due to consolidation is not recoverable or the rate of expansion is very small in comperi- som 10 the rate of compression. Consequently, the conceptual model shown in Figure 2.12 has 1 be modified to represent this behaviour ‘The rate at which the total load is applied also affects the soil deformation behaviour. If the external load is applied in small increments over a long period of time and the soil has free draining boundaries, then there will be no excess pore pressure and the applied incre- ments of load will be resisted by the solid particles, Volume change will occur in incre- ments similar © the loading. In general this type of te loading, whether in the field or in the laboratory, is termed drained loading. In undrained loading the water in the voids cannot drain 10 the fiee boundaries and volume remain unchanged. However, in the field, the lateral boundaries are not as rigid as the walls of the cylinder model, and an element of soil will undergo vertical and horizontal deformations most probably in the everse direc- tions to keep the volume unchanged. The term undrained loading can also be used to de- seribe the state of stress inthe field when the load is applied very quickly andthe seepage of water from the voids tothe free boundaries takes place slowly. In laboratory techniques the slow application of load is assumed to be equivalent to drained loading whilst rapid load application represents undrained loading. However, the full undrained loading condi- tion can be created regardless ofthe rate of load application. ‘The compression of a clay layer may occur ifthe water table is drawn downwards as a result of pumping. In this case the effective stresses increase due to the reduction in the level of the water table. However, ifthe pumping is stopped, the water table will recover ‘0 its intial position after time but most of the settlement that occurred in the clay layer will not be recovered. This forms the basis of a method of soil stabilization to minimize the volume change due to external loading. Example 2.6 A layer of clay of 4 m thick is overlain by a sand layer of 5 m, the top of which is the ground surface. The clay overlays an impermeable stratum. Initially the water table is at the ground surface but it is lowered 4 metres by pumping. Calculate o’y at the top and base of the clay layer before and after pumping. For sand e = 0.45, G,=2.6, 5; (sand, after pumping) = 50%. For clay ¢= 1.0, Gy 7, Solution: Calculate the density of sand at S; = 50% and 100%: For V=1m', 3 V, = W/-+e) =11+0.45) = 0.690 m,66 Soil Mechanics: Basic Concepts and Engineering Applications 104 Me/m? 1, Pyar (clay) = 1.850 Mei’ At the top of the clay layer before pumping: Pay = 0.690%2.6+0.310x1.0 Using similar caleutation for ¢ 6, =2.104%9.81%5.0 = 103.2 kPa, w= 1.0x9.81x5.0= 49.0kPa of, =1032-49.0=54.2 kPa, At the base of the clay layer before pumping 103.2 + 1.850%9.81x4,0=175.8kPa. = LOX981x9, 9, =175.8-88.3 AA the top of the clay layer after pumping: 949%9.81x4.0+2.104%9.81x1,0 =97.1 kPa, 1.0x9.81x1. ‘Ao’, (at the top) = 87.3 ~ 54.2 = 33.1 kPa, Ao’y (atthe base) = 120.7 = 87.5 = 33.2 kPa. ‘The increase in the effective vertical stress throughout the clay layer is uniform, 232 The case of partially saturated soil Bishop (1959) extended Equation 2.1 10 represent the state of stress in an unsaturated soil: O20 Hug Klug ~ty) 2.8) where ua is the pore air pressure, uy is the pore water pressure and is a constitutive pa- rameter depending on the degree of saturation. Nore that (ua ~ ny) represents the soil suc- tion. To explain the physical meaning of x we rearrange the terms in Equation 2.8 OOF Ay HUAI g PU= Hy Hy es Ie can be seen that xis the fraction of the unit area occupied by water, and 1 ~ 7 represenes the area of the air. The magnitude of x can be determined experimentally, however a lin- car relationship between degree of saturation and x may be adopted between the following limits: 5, 0,5, = 1, = 1. As the pote water and pore air are assumed to have an interface due to surface tension, then the pore water pressure must be always less than the pore air pressure. Therefore the reliability of Equations 2.9 reduces with a decreasing de- gree of saturation; however, for higher degrees of saturation which are on the wet side of the optimum moisture content, Equations 2.9 work well Fredlund (1973, 1979) and Fredlund & Morgenstern (1977) investigated the stress state in an unsaturated soil assuming four independent phases viz., solids, pore air, pore waterEffective Stress and Pore Pressure in Saturated Soils 67 and the air-water interface. Three stress variables of G— ig, 6 ~ thes and Wg — Hews Were introduced to express the state of stress in a three dimensional system of which only (wo are independent, as adding the frst variable to the third variable will yield the second vari- able. The strength parameters of the unsaturated soil were then defined differently to those of saturated soils in classical soil mechanics. The conditions of transition to the saturated. case were successfully applied in the proposed models. 24 PROBLEMS 2.4 For the soil profile shown in Figure 2.13(a) plot the variation of total vertical stress, Pore pressure and effective Vertical swess and indicate their values on the boundaries of each layer Answers: fy = 31.5 KPs, 98.5 kPa, 92.3 kPa 2.2. For the given soil profile of Figure 2.13(b) calculate the effective vertical stress at a depth of 7.5 m. Answer: $7.2 kPa A clay layer of 4 m thick with paar 2 Me/m” is overiin by @ 4 m sand with Pe 1.9 Mg/m’ and pary = 1.65 Mp/m’, the top of this layer being the ground surface. ‘The water table is located 2 m below the ground surface. The clay layer is undertain ‘bya sand stratum that i in artesian conditions withthe water level ina standpipe be- ing 4 m above the ground surface Calculate the effective vertical stresses atthe top and the base of the clay layer. IF the dry sand is excavated, in what depth the effective stress atthe bottom of the clay layer will become zero? ‘Answers: 50.0 KPa, 30.4 KPa, 1.9 m 24 A clay layer 10 m thick has a density of 1.75 Mg/m* and is underlain by sand. The top of the clay isthe ground surface. An excavation in the clay layer failed when the depth ofthe excavation reached to 6.5 m from the ground surface Calculate the depth of water in a standpipe sunk to the sand layer. Answer: 3.875 m 00m SEIS ————_— 0.0m . Using Equation 3.14 foreach testThe Movement of Water through Soil 77 Testh: q =281.5/10.0 = 28.15 ml/min =28,15%10~% /60,0 = 0.46910" m/s, 469%10°S 0.110.139 -0.100)x4.4179%10"9] = 2.7210 mis, 0.46910" x0.1/[(0.100-0.060)x4.4179x10"> | = 2.6510" nv. 22.5/10.0 = 42.25 ml/min = 42.25x10~ (60.0 = 0.70410 m'Vs, 704 10-* x0,1/{(0.2085 =0.1505) 4.417910] = 2.75x10~ ms, 0.704310 x0.1/{(0.1505 ~ 0.0915) x4.4179x 10] =2.70x10 mys. Test 3: q = $91,0/10.0 = $9.10 m/min =$9.10x10"® /60.0=0.985x10" m'is, (a) k=0.985x10 x0.1/[(0.292-0.210)x4.4179%10-] =2.72%10 mis, (b) k = 0.98510 ®0.1/[(0.210 0.127) 4.417921" ] = 2.6810" mvs, ‘Average k= 2.70 107 mvs Example 33 ‘A falling head test was conducted in cell of 100 mm diameter and 127 mm high. The re coded data were: f= 560 min p= 465 mm, 1 1081s, and the ameter of the vertical standpipe was 7 mun. Determine (a) kin mvs, (6) the time require for similar head and drop if astandpipe 4 mm in diameter was used Include the effect of capillary rise Solution (a) A=0.12 xn/: From Eqetion 3.15 k =2.3x3.848%10 > x0.127/(7.854% 10" 1081.0) lag(560.0/ 465.0) = 1.069x10~7 mis. Correction fr capillary tse: hy = AT gd = 4%0.000074/981%0.007%1000= 4.3 mm. fy = $600 ~4.3 =585.7 mm, hz = 465.0 ~ 4:3 = 460.7 nm . kK =2.3x3.848x10~> x0.127/(7.854%107> x 1081.0) log(555.7 / 460.7) = 1.078x1077 mis. (0) he = ATF yd = £%0,000074/9.81%0.004x1000=7-5 mm, 552.5 mm, fp =465.0~7.5 = 487.5 mm 257 x 10” m?. Using Fquation 3.15: 2.3%1.257 107 x0,127 x log(552.5/ 457.5) (7.854% 107 1.06910 From Equation 3.15 itcan be shown that in the absence of capa rise yf = DPI D3. 1081 ir, =7.02 1402 1p =353, 854x107 m?, a =0,007? xR/4=3.848%10°S a 3585, Example 3.4 A permeability test is arranged according to Figure 3.6. The cell has an inner diameter of 100 ‘myn ané a total length of 300 mm. Two soil specimens of equal length are fitted into the cell as shown. Soil 1 has a coefficient of permeability of 6.5 x 10"? mis. The total heads at points A and B (ia and fg) have been recorded 650 mm and 320 mm respectively. The amount of discharged water is 210 mf in 5 minutes.Caleulate the cocfficient of permeability of soil 2 Solution: A=7.854 107 m, L=03/2=0.15 Ak = OLIKA For soil 2: k=QLI MA using Equation 3.14 for sot 1: 10,010 x0,15.(6.510"* x7.854x10"° x5.0%60) = 0.2057 m. 65 - 0.2057 - 0.32 = 0.1243 m, 10.0x10§ x0,150.1243%7 854X109 x5.0% 60) = 1.075%10~4 ms78 Soil Mechanies: Basic Concepts and Engineering Applications Figure 3.6, Example 3.4 3.3.3. Inssitu permeability ests ‘The main disadvantage of a laboratory permeahility testis the small size of the sample. Iti, inappropriate to use only laboratory results to study low conditions in the Field. These results must he supported by in-situ tests before selecting an average & for the conditions involved in the project. In-situ permeability tests require che construction of test and observation wells both of which are expensive. Moreover, the complete hydraulic behaviour of any aquifer may need to be known. A comprehensive pumping test provides the necessary information about permeability characteristics and is a reliable basis for the prediction of ground water behav- jour, In-situ tests are normally based either on the constant or variable head methods. The flow rate and the position of the piezometrc level in the aquifer must also be determined which could he unconfined, confined or a combination of both. Care has to be taken to avoid distur bance, which may happen if aquifers are connected through the test wells and boreholes. A typical section of a modern test well is shown in Figure 3.7 (AS 2368). The well is sometimes protected against instability of its sides by casing but in dense and consolidated soils there ‘may not be a need for casing. Water is extracted from the screened jnterval(s) that may also be perforated or unlined, Pumping rest in an unconfined aquifer. An idealized section of an unconfined aquifer with a pumping well and two observation wels is shown in Figure 3.8. A pump is selected in accor- dance with the relevant code with regard co the maximum discharge and is mounted at the top. of the pumping well to discharge a specified flow rate of q. Asa result of pumping, the water Fevel in the well is lowered (drawdown) cteating a localized hydraulic gradient as shown in Figure 3.8, This causes the water to flow into the well in an axisymmetric condition making a cone type potentiometric surface around the well. The shape of the inverted cone (cone of de- pression) depends on the quantity ofthe discharge and hydeaulie properties of the aquifer. Af- ter equilibrium is reached, che water levels in the two observation wells are measured using ‘manual, automatic, electronic recorders or pressure gauges and the total heads of) and x2 are established. The flow passing through any cylindrical soil volume of radius rand height of zis equal tothe discharged flow:The Movement of Water through Soit 79. Piezometic tube ee plate ‘Weir Sion Pressure gauge, alr valve fang sample tap \ Control Non-retumn waive Drawdown Pump bowls measurement tube | AR pump intake Well sereen In(40.0/20.0) (7112.45? ~12.202)} = 6.56%10~ nis. Using Equation 3.19: 15, = 20.0exp{nx6.56%10 413.0? ~12.2°)/0.018333]=193 m. Example 3.6 ‘A pump test was carried out in an unconfined aquifer of k= 3 x 10-° ms with a flow rate of 20 1m fous, The radius of the well is 0.4 m and the aquifer has a depth of 80 m above an imper- Imeable stratum. The drawdown in an observation well at a distance of 150m from the well is 2.5 m. Caleulate the radius of influence and the depth of water in the well. Solution: 80.0 -25=77.5 mn, From Equations 3.19 and 3.20: ry =150.0xexplax3%10~ (80.0 ~17.52)/5.5 frq.52 5555x107 m0 a/i500)Aax3n10 Dy, =80.0 ~ 77.5? +5.555% 10 In10.4/150.0) (3% 10 50 m, Fe = 04 m, q = 20 m*Vhour= 5.555 x 10" m°Vs. Example 3.7 ‘A well is constructed to fully penetrate a confined aquifer of thickness of 25 m. The water fevel at two observation wells 40 m and 150 m from the well are 1.1 m and 0.4 m below the ‘original piezometric evel respectively. Determine the value of &. q=2.4 m/min. Solution: 9724/60 =0.04 mis, ell k= 0.041n(150.0/ 40.0) (28% 25.0%0.7) = 4.8104 ms. 2 = ty 04 0.7 m. From Equation 3.21 Example 3.8 An unconfined aquifer has a thickness of 40 m. A well of diameter 0.3 m is constructed for 3 pump test. For flaw rates of 43 m'Yhour and 135 m'/hour the actual drawdowns in the well are measured 3.7 m and 16.1 m respectively. Estimate the dravedown in the well for a flow rate of 60 m'thour assuming the radius of influence is proportional ta the flow rate (a) 37=400- fo00?+5.0/3600 91NO TST DH 161=400- fior GEO AGODA oe nr (0161 400-00 + €350/96000)100 1573 HH Solve two equations of (a) and (b) for ro and &84. Soil Mechanies: Basic Concepts and Engineering Applications 184.5 m, roa = 184.5 x 3.139 = $79.2m,k =9.58 x 10 mvs. For g (60.0 / 135.0) x $79.2 or (60.0/ 43.0) x 184.5 = 257.4 m. 40.0 90.0? + (60.0/3600.0)1n(0.15/257.4) xx9.58x1 33.4 Permeability of stratified sails Consider the stratified soil section shown in Figure 3.11. The average values of & in the x and z directions are estimated assuming that the water moves either parallel or perpendicular to the layers. When the seepage is parallel to «: qzav (lh 21) = Alka (Wy [2)) 5009 = Akg (Mig 24) ‘where Alt, Ait, and Aly are the losses in the total head as the water passes through the layers Of thickness of £1, 22, ... and 2. The total loss is equal to the sum of head losses, hence: I= Ah + Aly +--+ Ay OF: Aki = eee Aly lee Aba where 2=2, +2) tert gq and ke isthe average for the stratified soil in z direction, thus: (3.25) Blk tglkeg + Fe, hy, Note that the xz coordinate is selected to match the geometry of the layered structure and the z- axis may not be in the direetion of gravity. For the seepage inthe x direction the total flow rate is the sum of flow rates passing through each layer: 90,442 tt dy OF Akagi = Akg Agkgal t+ Agha where A= axb= (2) +2) tc#Q) x1. AL S21 % Toe ated Ay = z9 1, Substituting A values imo the equation above, we obtain: Keg =Car taken tet Skea)! 2 626) Kar Kets 21 : as Koos Fe 2 aoe Kone 2 a Figure 3.11. Analysis of average horizontal and vertical coefficient of permeability in strife SilThe Movement of Water through Soil 85 Example 3.9 ‘The profile of a stratified soil contains three horizontal layers each haying a thickness of 0.3 im. The top ofthe first layer is the ground surface. The & values are 10" mis for the top, 2.5 10 mvs for the middle and 4,0 x 10™ m/s for the bottom layer. There is 0.2 m of water above the surface of the soil. The section is drained at the base level. Calculate the flow sate per m” of the stratified soil in lirehour. If the overflow mechanism of the constant head is deleted, ‘what will be the time required for the water levet to drop to the ground surface. Solution: Using Equation 3.26: &, =0.940.3/10"? +0.3/25x10~ +0.3/4.0x10° 4g = Aki = 1,0x1.0310 + «(02 + 0.9)/0.9% 1000x3600 = 453.2 Vhour Using Equation 3.15 and assuming that, fy = 1 m, f= 0.9 m, L= 0.9 mand A k =L.03x10 = 2.3(aL/ Arjlog(h hg) = 2.3(0.9/:)log(11/09) ~> 1 =1751.s=29 min, Ls. 03x10 mis im 34 FLOW NETS 34.1 Flow nets in an isotropic material A flow net is a graphical representation of two-dimensional seepage and consists of two groups of curves of flow lines and equipotential lines (Figure 3.12). A flow line in a steady flow condition represents curve on which a particle of water commences its movement fron a known head (point $, Figure 3.12) and terminates ata specified lower head (point F, Figure 3.12). The tangent ata point on a flow line represents the direction of the velocity defined by Darcys law. Thus the velocity at each point is proportional to the hydraulic gradient which in tum depends on the coordinate of the point. Asa result the average velocity on a flow line be- ccames dependent on the length of the flow line. A shorter flow line means a higher hydraulic aradient for a specified loss of head, resulting in a higher velocity. Refersing to Figure 3.12 it can be seen thatthe velocity in the vicinity of the base of the dam is much higher than the ve- locity immediately above the impermeable stratum. An equipotential line is a curve on which Figure 3.12. Flow and equipotent lines under an impeemeable dar.86 Soil Mechanics: Basic Concepts and Engineering Applications the total head is constant (water rises toa constant level) and whose magaitude ranges between the highest and lowest total heads related to the seepage problem. The flow lines are every- ‘where normal to the equipotential lines, and thus the velocity at every point is perpendicular to the equipotential line passing through that point. From the definition of flow and equiporential lines we can see that their numbers are infinite, however a finite number of flow and equipo- ‘ential lines (low net) can give valuable information on seepage characteristics such as veloc- ity and flow rate. In steady flow there is no flow occurring actoss the flow lines and the rate of| flow hetween two flow lines is constant, Figure 3.13 is an example of a soil element bounded ‘with (0 flow Fines and two equipotential lines. The flow net is constructed in a way that the average distance between two flow fines within an element is equal tothe average distance be- ‘owen two equipotential lines () in Figure 3.13). Furthermore, the head foss between succes- sive equipoteutial lines is equal. The flow lines selected to represent the flow net are those in which the amount of flow between two pair of flow lines is equal. Two flow lines represent 2 lane (Wu, 1966) with a variable section snd the total flow rate is the flow rate corresponding to one lane multiplied by the number of lanes. The flow rate passing through an element abcd (Figure 3.12) is the constaut flow between two flow lines of gj and qi 1 and is given by h v= Ak (xy = kay =e 4 Ne NG where isthe numberof equal drops in total head anh isthe total loss du tothe seepage fhe ott number of ls lanes i Nhe kta low ates ag Ny g=NySq= hha! om ia Flow nets are constructed either analytically or by mumerical methods. Both techniques are based on the solution of the continuity equation (Section 3.5), In seepage problems with sim: ple boundary conditions the flow net is coustructed by sketching, as demonstrated in Figure 3.14. This may be done by hand or by computer graphics in accordance with the following. 1. The boundary between soil and water is an equipotential line, Be Figure 3.13. Flow and equipotential linesThe Movement of Water through Soil 87 Figure 3.14. Step by step construction of 2 Flow net. 2. Any boundary etween the soil and an impermeable material is a flow line. The imperme- able material is either an impermeable stratum(s) ofthe aquifer or the base ofthe Foundations ‘and sheet piles that ouch the aquifer. 3. Sketching may start by constructing the first low fane as shown in Figure 3.1$(a). The flow lane is divided into a numberof squares and equipotental lines are projected outwards into the second flows lane. 4. The second flow line is draw to form the squares of the second flow lane. 5. The procedure is continued until the fast flow lane is formed. This lane must consist of squares and satisfy the boundary conditions; otherwise the first flow line has to be reposi= tioned and the whole procedure repeated. Example 3.10 A shee piling system with its corresponding Flow net is shown in Figure 3.15. (a) Estimate the flow rate in m’/day per 1 m eun of piling, (b) forthe element A with {= 1.5 meafculate the av- rage velocity and the effective vertical stress, (c) determine the magnitude of the effective Vertical stress atthe base and atthe right-hand side ofthe sheet pile, (d) calculate the factor of safety against the quick condition, defined as the ratio of the existing hydraulic gradient along, the downstream face of sheet pile to the ertcal hydraulic gradient defined by Equation 3.11. k =0.02 mms, Your = 20 KNAn Solution: ()Np=S and Ny= 11, h=3.0 m, 5 = KAN 7 Ng = (0.021000) >24%<3600%3.0%5/11 = 2.356m° day. (6) v= Ai = 0.02x((3.0111)/1.5]%3600 = 13.1 mmvhour88 Soil Mechanics: Basic Concepts and Engineering Applications impermeable siraturn igure 3.15. Example 3.10. Head loss atthe centre of the clement = 2.5 drops x 3.0/ 11 = 0.682 m, diy. = dept of A from water surface ~ loss of head = (4.5 + 1.5) 0.682 318x981 = 52.2 kPa, 6, = 4.5%200+1.5x9.81=1047 kPa, 6, =104.7 ~52.2 = 52.5 kPa (6) Head loss atthe base ofthe sheet pile = 6 (number of equal drops) x (3.0/ Lt) = 1.636, Height of water at this point = depth from the water surface ~ loss of head (4.5 +1.5)~1,636 = 4.366 mu = 4.364x9.81 = 42.8 kPa, =3x200= GOOKPa, 0}, = 60.0-42.8=17.2 kPa, (4.364 ~3.0)/3.0= 0.455, {, = 7p, = 200-9.981)/9.8 (039/0.455= 2.28. ‘The factor of safety against the quick condition is also estimated by evaluation of the stability ‘of a prism of soil D (embedment depth) by D/2 (width) by unit thickness attached to the hack of the sheet pile. The factor of safety is defined as the effective weight ofthe prism divided by the upward force resulting from the Seepage pressure at the base of the prism. 318 m, 342 Flow nets in an anisotropic material Consider an anisotropic soil with horizontal and vertical coefficients of permeability of ky and A, respectively. The material can be treated as an isotropic soil by assuming an isowopic coefficient of permeabilty of: bade a2 and using a ansfrmed scale of anf lh 629 Considering an element of soi in x direction, having the length of and cross-sectional area of ‘Act flow tate nthe dietion inthe utansformed stats: = Ak, = Ak! In the transformed state:The Movement of Water through Soil 89 80m 50m 58X10 ms 00x 105m SAT 80m Impermeabie stratum impermeable stratum @ © Figure 3.16, Example 3. Example 3.11 A concrete dam has 8 base length of 8 m and retains $ m of water as shown in Figure 3.16(a). ‘The water level on the downstream side is atthe ground surface. Under the dam there exists an 8 m thick layer of anisotropic permeable soil with the & values shown in the Figure 3.164a). Caleufate the flow rate under the concrete dam in Vda. Solution: — The scale factor in the x direction is: fk, /E, = ¥L.0%10™$ /2.56x10 The length of the transformed base is: 8.0 x 0.625 = $.0 m. ‘The equiv coefficient of permeability is calculated using Equation 3.28: ms Figure 3.16(b) shows the transformed section of the soil-dam, with the flow net drawn by sketching in which: Na =, Ny ‘Using Equation 3.27: q= kiN, IN, 625, 6X10" «5.0% 1000x3600 24/8 34.3 Control af the flow rate under impermeable dams The flow rate in the soil under an impermeable dam (Figure 3.17(a)) and the factor of safet against the quick condition may be controlled by introducing horizontal oF vertical imperme= able blankets and a toe filter. A horizontal impermeable blanket may contain a thin layer of compacted clay constructed on the upstream side (Figure 3.17(b)). This increases the length of the flow line immediately under the darn, projecting all the flow fines tothe left of the blanket ‘Asa result, she number of equiporential lines increases, thereby reducing the flow rate (Equa- tion 3.27) to a required valve. It may be shown tha ifthe number of equipotential lines is kept the same as in the uomodified section, then the number of flow lines and the flow rate will both decrease. In addition, the pore pressure atthe toe decreases, improving the overall stabil ity of the dam and the factor of safety against the quick condition. An example of a vertical90 Soil Mechanics: Basie Concepts and Engineering Applications ZL (\eoagmoatie dem | eas x impermeable blanket @ o pr IOS ” ‘Sheet ple Toe drain © @ Figure 3.17. Methods to control the flaw rate and quick coalition impermeable blanket is a sheet pile system driven into the soil, either atthe toe or at another position within the base of the dam (Figure 3.17(c)). Sheet piles are more effective in increas- ing the factor of safety but less effective in decreasing the flow rate. A toe drain shown in Fig ture 3.17(d) may also be used to significantly increase the factor of safety but it will result ina higher flow rate that may or may not be desirable for the project. A combination of the above ‘methods may be used 0 optimise the length of the base which és designed mainly to provide stability against hydrostatic and gravity forces. ample 3.12 Resolve Example 3.11 with a horizontal impermeable blanket of 8 m length added co the sys tem atthe upstream side. Solution: Impermeable stratum Figure 3.18, Example 3.12.The Movement of Water through Soil 91 The transformed length of the horizontal blanket is: 8:0 x 0.625 = 5.0 m, Figure 3.18 shows the low net drawn schematically for the transformed section. The number of equipotential lines were kept the same: N= 8, and Np= 4, Using Equation 3.27: g = 1.6%10 ° x5.0%1000%3600%24%4/ The flow rate is reduced by: (4320 ~ 3456) / 4320 = 0.2 = 20%, Example 3.13 For the concrete dam shown in Figure 3.19 construct the flow net under the dam, calculate the flow rate (in m°/day) and the distribution of uplift pressure on the base of the dam. The base is | m below the original ground surface and a sheet pile,7 m long is driven into the soit at the vpstream side. =9 m, k= 4510" mys, Yoo = 21 RNA Solution: The flow net is shown in Figure 3.19, from whigh: Nu= 10 and N= 4. Using Equation 3.27: 4g = 40x10" x9.03600%24%4/ 10 = 12.44 Aday (per mette run). The total head across the length of the base is formulated as follows: f= 10.0 + L.O(depth of foundation) ~ 0.9n,, where ny is the number of equal drops of 9.0/ 10 =0.9 m up to the point of interest on the base, The distance of the intersection points of equi potential lines from the left corner of the dam and the relevant heads and water pressures are presented in the table below. For point (I) we may use interpolation or, couservatively. simply ‘accept the higher magnitude corresponding to ny = 3. Poi . 0 zim) 000 1500 19.35 hm 83) 36 47 wikPa) 814 54946. i 250m 200m Figure 3.19, Example 3.13,92. Soil Mechanics: Basic Concepts and Engineering Applications ‘The diagram of uplift pressure is also shown in Figure 3.19. The average hydraulic gradient bherween the base and ground surface on the downstream side is: i=Ah/1=0,911.0=09, ic=7 ly = @L.0-9.81)/9.81= 114, Factor of safety against the quick condition = 1.14/09 = 1.27, 35 MATHEMATICS OF THE FLOW IN SOIL 35.1 Continuity equation fora steady ow Figure 320 shows thre dimensional element of soil in which th tou hea of wate ish ‘Assuming steady sate Now, the quantity of wate emerng the element pall wane ofthe anes (eB 1aN) ts gee Av bere vb the component of velocity in the = drestion and Ay ip x des the sectional area of the element perpendicular the destion of lo. Hence Ea Ea The negative signi iar and causes the vale off to decrease in the destin ofthe ve Toc Substatng vs into the lw rate equation we obua: k 30) an Paya ean or : ‘he amount of fl that exits the element inthe x direction is qx + ds im which dy the satiation of the low rate inthe xdtestion and is expressed by: Ph gc aya a3 aF Assuming zero volume change the low rate etering the element is equal tothe low ite xe iting the element. Therefore, the continuity condition can be writen a 4s day = mks Ga tay tas ix +49, +9, +49, +9; +0q2 =A, dq, +g, +g 33) one +09, ard % oz[ te o,09, al, y i Figure 3.20. Steady flow through a three dimensional elementThe Movement of Water through Soil 93 Substituting Equation 3.32 into the continuity condition we have: ° Gay) In two-dimensional low, where flow in, say, the y direction is zero, the corresponding term in the continuity equation vanishes. Equation 3.34 therefore becomes: ah, Ph eh be SE nga hea (for ke = &) 3.35) The solution to this differential equation yields the total head within the seepage zone. It is seen that the distrihution of the total head is independent of the coefficient of permeability (ovhen ky = kp) and depends entirely on the geometry of the problem. The continuity condition expressed by Equation 3.35 may be arranged in terms of the components of the velocity by substituting Equations 3.30, resulting in 636) Equation 3.35 (or 3.36) isthe classical differential equation of flow in two dimensions and is, referred 10 as the Laplace equation. Traditional methods of solving tis equation include direct integration and numerical analysis. Direct integration is carried our only for seepage problems with simple boundary conditions by using complex functions. Whilst there are an infinite number of solutions tothe Laplace equation, the houndary conditions contol the selection of te function. For one dimensional flow, inthe x direction: 2 ae Equation 3.37 may be integrated (wice to give the following linear relationship for the cota head: f= Cre + C2, where C1 and Cz are integration constants found by substituting two known boundary conditions into the linear equation. These known boundary conditions may be selected as the total heads fy andl ha corresponding to two sections with vy =O and x2 = L where £ is the distance between the two sections. Substituting these values into the linear equation of total head we ge: CEH gy L. Ce ° 37) hy The total head between two sections of x1 and 12 is represented by: =m ig) Lh, 338) By differentiating Equation 3.38 and applying Darcy's law, the velocity is found to be: an 39) Or ‘This shows that the velocity is constant between sections | and 2 and is consistent with the ba- sie assunption stated previously (Figure 3.1).94 Soil Mechanics: Basic Concepts and Engineering Applications 35.2 Potential and stream functions Equations 330 represent the projections ofthe velocity in a thee dimensional system and in Diy tha he voc is pepedicalr wo the constant h passing through te pin. Te anal {al sofuion of two dimensonal flow trough an scope mater expressed by Equation 335 Soe Ne sehen i ee ee ME eS oo eae eee ex where ® and ae sled potential fnction and stream f Equations 340 ito 3.38 We obtain ao ao eco ee tased onthe definion of n.2)= 2) #C ea) here Cis constant depending onthe boundary condiions. For diffrent bat onstant ates OF, such as yy and Os, Eeuation 842 represen seis of cures for Which the pote tal faeton and to head are constant e-equipotenial lines, From th definition of the function iis seen tar it tangential tthe veloy and represents the dein ofthe Now fn ow lines By comparing Eqntions 40 and 82, we chain 20 a¥ 30 a ax 8: ox “Teve ae called the Cauchy Riemann equation whose soliton yes series of stream ne sons or Row Tins for constant vals of Wi, Was asigned for Wi). Taking the derivative = 3.40) nction respectively. Substituting ° BaD \which can be expressed as follows (3.43) ofthe first term of Equation 3.43 with respect to < and the second tefm with respect to x and ali the results, then zy at oy ot 3.48) Stream or Flow ines Figure 3.21. Caleulanon ofthe quantity of flow between two flow TinesThe Movement of Water through Soil 95 Which means that the stream functions are in the form of the Laplace equation. Figure 3.21 shows two flow lines represented with ‘1 and 'P2. Atan arbitrary point within these lov fines the velocity v is shown with its projections in the xz plane. With the arrangement shown for the positive variation of x and z along the equipotential line, one of the projections of the ve+ Tocity is in the negative direction of the corresponding plane axis, The flow dg corresponding to the length of ds on the equipotential line is: dg =vxds =v, xde—v. xe ‘Substituting velocities from Equations 3.40 we obtain: avy ae ar ag Yor og fe aH =, a ‘This shows that the flow rate between two successive flow lines is equal tothe difference be- tween the stream functions. Ifthe stream function is known in terms of x and then the veloc- ity components can be calculated using Equation 3.40. Conversely, with known functions for the velocity components, the stream function can be obtained by: ve {Mare [Baec 646) A useful result is abiained by substicuting Equations 3.43 into Equation 3.41 B47) ‘This indicates that a stream function obtained from Equation 3.46 automatically satisfies the ccomtinuity equation, Equations 3.35 (or 3.41) and 3.44 are a set of differential equations for ‘which closed form solutions are available only in very simple cases. Analytical solutions are obtained by using the theory of complex variables. For example, for an impermeable damn with a base length of b, and a total head loss of fh, resting on a permeable layer, similar to Figure 3.12 but with infinite depth, Wu (1966) reported the following closed form solutions for po- tential and stream functions: pallet yy ee =. coed > Eur eos™2)? sin tees)? sinh ™ In general, the solution of two-dimensional flow is achieved by numerical techniques such as finite element and finite difference methods, I the Finite element method the seepage zone is divided into discrete elements with common nodes. The parameter of interest (kor y) is as- sumed 10 vary within each element according to a linear ar a non-linear fonction. A solution is obtained for specified boundary conditions that covers the entre seepage 2ove. Inthe finite difference method the seepage zone is divided into a network of nodes usually Known as grid points. The partial derivatives at each point are replaced witha finite difference approximation G48)96 Soil Mechanics: Basie Concepts and Engineering Applications Figure 3.22. Deflection of flow lines atthe boundary between wo Soils with different permeabilities. that results in a linear algebraic equation. With known boundary conditions, and application of the continuity condition at the boundaries with unknown numerical values for the specified function (f, © or y), a unique solution can be obtained for grid point values. The variation of 1k, @ or v is assumed to be linear between grid points, which is a major disadvantages com> pared to the finite element method, However, the finite difference method is easy to formulate and a useful alternative 1 the analytical solution, 35.3. The ransfer conditions The line ab in Figure 3.22 represents the boundary between two soils with different & values. The flow lines at the ky side are deflected as they enter into the kz side, This i represented by different values of B, the angle between a flow line and the direction normal to the boundary. The flow net at the hy side is comprised of squares of length f}. At the fp side the squares change into rectangles fg by fs which represent the distance between the equipotential ines and flow lines respectively. Consider the case where one flow line and one equipotemtiat lin inter- sect at a point B located on the boundary. According to the continuity condition the flow rate inside a flow ane atthe ky side is equal to the flow rate in the deflected lane atthe kz side. gy = Abi = (1, XIDR SR/ = kh, Agy = ARI = (by xI)ky Ata From the geometry of Figure 3.22, and can be expressed in terms of fy, 82 and fh: fz = BCsinf =/, sing /sinB, by = ABcosB = ,cosBy/cosB, Substituting /2 and (5 in the flow equations and equating Agy by Aq2 we obtain: ky MtanB = kp /tanBy > ky tanBy = ky tanB 6.49) 354 Numerical analysis of two-dimensional flow using the finive difference method In this method two-dimensional flow, represented by the continuity Equations 3.35 or 3.41 or the differential equation of stream function in Equation 3.44, is approximated by finite differ-‘The Movement of Water through Soil 97 ence equations. In this section We develop numerical values for the solution of the Laplace equation for total head given in Equation 3.35. Let Ax and Az be the change in the coordinate of a point represented by i and j as shown in Figure 3.23(a). Restricting the Taylor's series to a finite number of terms, the total head at points (/+ 1, and (I~ 1, representing +x and —Ae respectively, are a Fag = Wj FAG) 3.50) panany, oA asp Adding Eautions 3.50 an 3.1, we obtain ? ety t as) Sin, for Now nthe = vection, we have then thot 2h ah, Rigen TMi J G3) Oe (dey Sabsiutng Equation 352 and 3.53 no ie coninity Equation 3.35: haus thewy-2hy haya sp tietat ics Bhhy Must . (an ©) Figure 3.23, The basic finite difference grid.98 Soil Mechanics: Basic Concepts and Engineering Applications Pasty theo * Miso hi po fy, oe tc hier sot 3.55) hig 7 55) ‘This equation shows that, in square grid, the total head at every gid point isthe average of the total heads atthe four adjacent grid points, Ic also implies that the total head at every point inthe grid represents the average total bead within a flow lane of thickness Av or Az, and that the point is located on the centre line ofthe flow lane (Figure 3.23(b)). This means that if we apply Darcy's law in the form g = Aki, and the continuity condition in the form Aq = 0 (at every point) we can obtain Equation 3.55. Note that in this case, the bydrauli gradient isthe first derivative of the total head expressed in finite diterence form. Equation 3.55 can be modified (using the concept of flow lane and continuity condition) for the case when the point (i,j) is located on an impermeable boundary (Figure 3.23()): int thing #2 i ee 336) 4 If the point (i,j) is located at the base of a vertical impermeable boundary with negligible thickness (e.g. sheet ple), as iflstrated in Figure 3.23(d), the total head at the grid point (i, + 1) can be taken asthe average of the total heads a the left and righ ofthe boundary: O51, ota +h, jor PR *h + If the point (i,j) is located on the houndary of two soils with different permeabilities, and the ‘boundary is parallel to the horizontal axis, then Equation 3.55 becomes: 2 os Eke where kr and kg are coefficients of permeability at the top zone and the bottom zoe of the boundary respectively (Figure 3.23(e)). For each unknown bead at an interior or a boundary ‘grid point, there is a condition expressed by Equations 3.55 to 3.58. Thus, the total number of linear equations and unknowns are equal, and a unique solution can be obtained. Traditionally ‘he Set of equations was solved using a relaxation method, which included the readjustment of assumed values of the unknowns to satisfy the appropriate condition at each grid point. How: ever, withthe improvement of computational tos, a direct solution is now preferred. +h G51 1 ag yin ions ed 659) Example 3.14 Figure 3.24(a) shows a section of a concrete dam, with an $ m base, resting on a permeable layer of thickness 8 m. Taking the impermeable stratum as the datum for toral heads, calculate the total heads at the grid poimts using the finite difference method and construct the equipo- stial line for a head loss of 1.25 mm, Repeat the solution using the relaxation method by a re fined grid (occupying the same area) with Ax Solution: Amesh of Ax ‘mis constructed to evaluate the total heads atthe grid points. Since the flow net in the permeable layer is symmetric, only one half of the section is considered.The Movement of Water through Soil 99 impermeable stratun a @ Figure 3.26, Example 3.14 Boundary conditions: fy = hag =f 3 = 5.0+8.0= 13.0m, Fora symmetrical low net: hay hay =hg3 =25+80=10.5m, ‘Assume at the left boundary (grid points 1,1-1,2 and 1,3) the flow rate is 2er0 and the pore pressures at these points are hydrostatic: fy y =f, = h,3 =13.0m. Applying Equation 3.55 at the interior grid points 2,2 and 3,2 and Equation 3.56 at the grid points 2,] and 3,1 on the impermeable boundary, and substituting the knowa boundary heads: why +My 2 hig =260, “hy 2 hy) +452 = 28.5 ig ~2hy hy, =13.0, gg + Alyy ~2ly2 =105 Solving for the unknowns: fa,1 = 12.49 m, ig. 2 = 12.61 m, fis = 11.73 m,hs2= 11.96 m. The total had for the equipotential line with a 1.25 m loss is 13.0 ~ 1.25 = 11.75 m. The ddashed line in Figure 3.24(b) shows the equipotential line for = 11.75 m, where the locations of poims A, B, C and D have been established by linear interpolation between the two grid poimts at oth sides of the point of interest. For the finer grid the first estimations of the total head at the grid points are shown in Figure 3.25 (first row). The first iteration is applied using Equation 3.55 for the interior points and Equation 3.56 for the points on the impermeable boundary. These results are presented in the second row whilst the third row shows the results ‘of the fourth and final iteration, The cquipotemtial line corresponding to h = L175 m is repre sented by the solid line, while the dashed line represents the previous analysis withthe coarse rid, Sample calculations forthe first iteration for grid points 3,2 (interior) and 3,1(00 the im permeable boundary) are is hay + hag +hg thy )4 hag tga thy thy Ay But as the selected values are not the exact values, a residual R is defined according to:100 Soil Mechanics: Basic Concepts and Engineering Applications 2.0m. 15: 25 38, 4s ‘< isco 13.09 13.00 |13.00 10.50 20m na 24 34. 4a 74 1300/1290 [1280 fi2ea 10.50 W290 1273/1260, 1200 [1278/2598 1.3 23 38 49 7.8 i300 -fiz80 [ize [12.90 10.50 p278 i261 12.95 howe |i2er—fr2se 2 22 32 42 72 1300 fia7s izes fizze 1050 ro74 [ize frzas har fi2s1 frase Ol 24 3a 4a 51 6A 7.4 7300 1270-2491.) ——«1080 1270 | 12A8 1216 7s is Name | ear said | 1190. | sa Figure 3.25, Example 3.14 Ry =hhga tig tis 3 +h) ~4hyy = 1219 412.75 + 12.61412.49 ~4%12.55 = -0.16 m. The corrected value of fy2 to satisty the continuity condition is dig =hyg +Rya A= 12.59 + (-0.16)/4 = 12.51, For grid point 3.1: Igy = (gg +h +259), dating +2lg ~Aigy = 12124127 2x12.55-4x12.49 =-0.04m, figy + Ry 42 12.49 + (-004)/4=12.48.m More precise values may be obtained ifthe left boundary is moved further wo the left in order to justify the assumptions made on the total heads forthe points om this boundary 3.6. SEEPAGE THROUGH E: RTH DAMS 3.6.1 Entry and exit conditions ‘An idealized section of a typical earth dam is shown in Figure 3.26, Iis constructed on an im- permeable base and the seepage zone is limited to the embankment soil only. The height of the ‘water atthe upstream side is whilst at the downstream side, the water table is assumed to beThe Movement of Water through Soil 101 Impermesble base cemeny Figure 326, Am idealized section of permeable cath dam. ‘on the ground surface. A toe drain may be constructed at the downstream side to control the "uppermost flow line or phreatic surface. Iss length is designed to stop the phreatic surface from exiting at che downstream slope, The main objective is wo establish the phreatic surface and the flow net within the dam in order to calculate the flow rate as well a the parameters necessary for a stability calculation, such as pore pressure and seepage pressure. Since the co- ordinate system shown in Figure 3.26 does not agree with the conventions of the previous sec tion, the relevant equations are modified for this arrangement. The problem involves three cases of transfer conditions: 1. Enary of water fram the & =< zone (water) to the soil with coefficient of permeability of & 2, Possible exit af water from the soil tothe air (k= ©) at the downstream slope. 3. Entry of water from the soil to the free draining at toe with k ‘The entry and exit conditions for a boundary between a soil of permeability of k and a zone of infinite permeabitity may be established by constructing @ flow net around the boundary sini lar to Figure 3.22. However, in this section, a complete proof of the transfer conditions is not presented. Figure 3.27(a) shows a case in which water enters from dhe upstream face to the soil, AS the upstream face is an equipotential line, the flow line must start perpendicular to this face. The flow line rises up a litde above the normal fine tothe face as shown in the figure. If there is a gravel wall between the water and the earth dam, the phreatic surface will be tangent to the water level in the gravel and will drop downwards, as shown in Figure 3,27(b). Note that the downstream side of the gravel wall does not represent an equipotential line. Figure 3.28(a) shows the entry condition from the soil to the air at the downstream face of the earth Flow line “rowtine | Sol a> 90° | (3) © Figure 327. Entry conditions fom the water 10 the soil102 Soil Mechanics: Basic Concepts and Engineering Applications ‘7 Free raining 90" Flow ine @ 5 ow tne son ~--}- draining «00 © ) Figure 328, Entry conditions from the silt the ar and drain damn where the phireatic surface is tangent to the slope. Figures 3.28(b) 10 3.26(d) show the exit conditions from the soil to a drain with a horizontal, a vertical and an inclined surface respec tively. In all cases the flow line is tangent toa vertical line invespective of the slope of the sur face of the drain, 3.6.2 The equation of the phreatic surface: the basic parabola A waditional (Casagrande) method defines the phreatic surface as a parabola with its focus lo- cated at the origin O of the xz coordinate system (Figure 3.29). This basic parabola can be de- fined mathematically if the coordinate of one point on the boundary (or within the seepage zone) is known, and is shown as curve FMC in Figure 3.29. Experimental investigations have om ! Directix impermeable stratum Free draning rae Figure 329. Casagrande method 1 esiablish the phreaie surface in an earth dam,‘The Movement of Water through Soil 103, Directrix Iimper rmeable stratum @) Figure 330. Correction of the phreati suave to saisy the eit conditions. shown that the intersection point of the basic parabola and the water surface, point F, is to cated such that FA = 0.3 EA, which means that the coordinates of point F are known. The pa rabola has fo be corrected at point A to ineet the requirements of the entry conditions. The dis- tance af any point Af on the Basic parabola from the focus is equal to the distance of this point ‘from the directrix, which is located at an unknovn distance of p from the z-axis. + px=(2?=p?)/2p G9) By substituting the coordinates of point Fin Equation 359, the valve ofp so obtained is 3.60) ‘The flow rate is estimated by constructing the flow net schematically as explained earlier. An alternative solution is to assume a constant hydraufic gradient in the vertical sections: iki = (ex)Rdz/ dx, from Equation 3.59 dz/dx = p/2 and: Bsr) The horizontal length of the toe drain L must be sufficient to allow the basic parabola to be lo: ccated inside the earth dam. It is calculated by intersecting the equation of the hasic parabola with the equation of the downstream face z = tana. (x +L} and seeking the condition in which both functions become tangent to each other: L= p(ltcot® a)/2 (3.42) In the absence of a toe drain the basic parabola intersects the downstream face at point 8 as shown in Figure 3.30(a). In reality. the phreatic surface must be tangent to the downstream, face at point T with a distance a from the origin O. In the Casagrande method, the correction. length of Sa (Figure 3.30(b)) is found from Figure 3.31 which is based on experimental re- sults. Note that the distance OB = Aa + a can be easily established, as the equations of the ba- sic parahola and the downstream face are both known,104 Soil Mechanics: Basic Concepts and Engineering Applications aa/ (ata) 8 8 60 90 120 150 180 (degrees) Figuee 331. Casagrande method to comect the preatc surface Table 3.2 da / (da + @) in ers of angle (depress) 3080S). 18D Aavide+a) 036 032 026 018 010 0 Table 3.2 shows the approximate values of Aa / (Aa + a) for given angles of a based on Casagrade's experimental work. An estimation of Aa / (Aa + a) can be made by assuming the following linear relationship ofthe form reported by Linsley etal (1992). dada +a) =(180-a)/ 400 3.63) ‘which is shown as the dashed line in Figure 3.31. The top flow tine can now be corrected by sketching a smooth curve tangent to the basic parabola a 8” and downstream face at T (Figure 3,30(b)). An alternative solution, called the Dupuit method, defines the correction curve as fol- lows: (a) at any vertical section the hydraulic gradient defined by ds dx has a constant value, (©) the curve passes through point F which is already defined in the construction of the basic parabola, (c) The curve is tangent tothe downstream face It can be shown thatthe fist condi tion yields the correction curve as a parabola ia the form: x= Cyz? +Cy ‘The integration constants C1 and C are found by substituting the second and third conditions in the above equation. Hence, the equation ofthe flow line adjacent to the downstream face is r=cosa(s? ~2})/2asin? at xp 3.64) TThe coordinates of point Tare x= a cost and 2 =a sinc. Substituting these values into Equa tion 3.64 two answers are obtained. the positive value of which is the distance a. 8.65)‘The Movement of Water through Soil 10S | 200m woemeane wen a 200m Direct Figure 332. Example 3.15 ‘The flow rate can be obtained by calculating dz / dx from Equation 3.64 and substicuting this result and 27a sinc. into the flow equation: g= Aki= (ep xDK(de/dx) > q = kasin tance 6.66) Example 3.15 AA vertical section ofa large earth dam is shown in Figure 3.32(a), Estimate the flow rate by sketching the flow net and use the Dupuit assumption. ky =k. =3 x 10"° ms, Solution: tan c= 60.0/{(200.0 -20.0)/2] = 0.666 ~ cx = 33.69°. Referring to Figure 3.29: b 100.0 = 82.5 + 0.3x82.5 - 60. A= 550/130 33.69° = 82.50 m, From Equation 3.60: p = ¥8225? + 55.07 -§2.25 =16.69m, ‘The equation of the basic parabola is: x= (2? ~ p?)/2p = Figure 3.32(b) shows the flow net within the earth dam in which Nj GH HAN, [Ng =3.0%10~* x24x3600%55.0%3/ 10 = 4.28m'day. When the Dupuit condition applies: 4 = kp =3%10° x16.69%24%3600= 4.33 m'/day. Example 3.16 For the carth dam sectiop shown in Figure 3.33 calculate the values ofa and 4a and the flow rate across the dam in m’/day. ky 10° mis Solution:106 Soil Mechanics: Basic Concepts and Engineering Applications 140m 160.0m__Impermeable stratum Figure 3.33. Example 3.16 32.0/1(160.0 14.0)/2 43836 —> «= 23.67°, Referring to Figure 3.28: FA=300/1a023.67° = 68.44, Xp = 160.0~68.44 +0.3x68.44 = 112.09 m, zp = 30.0 m, From Equation 3.60: p = vi12.09 + 30.07 ~112.09 =3.94m The equation of the basic parabola is: x = (2? - p*)/2p = (2? ~3.942)/7.89. The length OB = Aa + ais calculated by substituting: xp = OB cox 23.67° and

Permeable base IIDIIS IIIT ELLE TOS Impecmeable stratum © (Cito trench Impermeable stratum ) Figure 3.36. Different types of earth dams. 37 PROBLEMS 3.1 Ina constant head permeability est, a cylindrical sample 100 mm in diameter and 150 ‘mm high is subjected to an upward flow of $40 ml/min. The head loss over the length of the sample is 360 mm. Calculate the coefficient of permeability in mn Answer: k=4.8x107 mis In a laboratory falling head test, the recorded data are: diameter of the tube = 20 mm. di- ameter of the cell = 100 mun, length of the sample = 1000 mm. The head measured from veto" Figure 3.35. Problem 3.3,108 35 36 Soil Mechanics: Basic Concepts and Engineering Applications the top level of the sample dropped from 800 mm to 600 mm within I hour and the temperature ofthe watey Was 30°C, Calelate the egeeticint of permeability at 20°C. 1 = 1.005 x 10" Nisin? (at 20° C), n = 0.801 x10" N/m” (at 30°C), Answer: k = 4.010 mis For the test arrangement shown in Figure 3.35, calculate the volume of water discharged in 20 minutes. The cross sectional area of the soil is 4000 mn* and k = 4.0 mms Answer: Q= 23081 ‘Along wench is excavated parallel toa river, as shown in Figure 3.36. The soil profile consists of @ permeable soil of thickness D confined bewween two impermeable layers Initially the water level in the trench isthe same as that inthe river. Water is pumped out Of the trench ata flow rate of ¢ (4) Formulate g in terms of the geometecal parameters shosun in Figure 3:36, (6) for L = 100m, D = 5 m,and k= 4% 10" ns, calculate the flow rate corresponding toa drawdown of Dy =2 m, (c) calculate q when the water table in the trench is 2 m below the surface of the permesble layer. Answers: q= kD(2,,~D)/L (for full flow), q= 2D(2y ~ D)+ D? = (zy ~ Diy)? 2L, (fgr the water table in the trench under the surface of permeable layer), 0.346 and 0.622 1 Ydayimette run of trench, ‘A wel of diameter 0.3 m is constructed 0 the full depth of an unconfined aquifer of Uhickgess of 150 m, The water table is 101m below the ground surface. A pumping test of 12 mMour has resulted a drasdown of 10-m, Assuming ro = 400 my calouiate the coef ficient of permeability of the aquifer. I the flow este increases to 18 m'/day, and in the absence of any other data what willbe the best estimate forthe drawdown inthe well? Answers: 3.1 x 10 mvs, 15.3 m ‘A pumping test carried out in @ 50 m thick confined aquifer (well dia. = 0.6 m) results in flow rate of 600 Vinin. The thickness ofthe impermeable layer above the aquifer is 20 ‘mand the original water level in the well is 2 m below the ground surface (which is also the top of the impermeable layer). Dravdowns in two observation wells focated 50 m and 100 m from the well are 3 and 1m respectively. Calculate: (a) the coefficient of permeability of the aquifer, (b) the drawdown in the well, (c) the radius of influence. (4 Lf. ° Impermeable Permeable Figure 3.36, Problem 3.4