Luận án tiến sĩ: Development of a protocol for the assessment of unsaturated soil properties

ABSTRACT In this study, shear strength and volume change behavior of unsaturated soils 1s evaluated using axis-translation test methods and advanced testing equipment designed to provid

DEVELOPMENT OF A PROTOCOL FOR THE ASSESSMENT OF UNSATURATED

SOIL PROPERTIES

by Natalia Perez

A Dissertation Presented in Partial Fulfillment

of the Requirements for the Degree Doctor of Philosophy

ARIZONA STATE UNIVERSITY

December 2006

UMI Number: 3241336

INFORMATION TO USERS

The quality of this reproduction is dependent upon the quality of the copy submitted Broken or indistinct print, colored or poor quality illustrations and photographs, print bleed-through, substandard margins, and improper alignment can adversely affect reproduction

In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted Also, if unauthorized copyright material had to be removed, a note will indicate the deletion

®

UMI

UMI Microform 3241336 Copyright 2007 by ProQuest Information and Learning Company

All rights reserved This microform edition is protected against unauthorized copying under Title 17, United States Code

ProQuest Information and Learning Company

300 North Zeeb Road P.O Box 1346 Ann Arbor, MI 48106-1346

DEVELOPMENT OF A PROTOCOL FOR THE ASSESSMENT OF UNSATURATED

ABSTRACT

In this study, shear strength and volume change behavior of unsaturated soils 1s

evaluated using axis-translation test methods and advanced testing equipment designed to

provide accurate volume change determination during specimen shearing Tests were performed on a wide range of soil types with net normal stress from 20 to 250 kPa and matric suction from 20 to 700 kPa This data set adds significantly to the existing

database of unsaturated soil properties

The triaxial test data was analyzed using the extended Mohr-Coulomb equation proposed by Fredlund et al (1978) A hyperbolic curve was fit to the o° versus suction relationship, and a correlation between the hyperbolic curve parameters and soil index

properties was developed Volume change behavior for the soils was related to soil

suction, and it was found that for a given suction, the test soils exhibited greater tendency

to dilate as soil suction is increased For a given suction, soils exhibited higher

compressive strains as the net normal stress was increased, as expected

A new oedometer device for soil water characteristic curve (SWCC)

determination was evaluated by testing a wide range of soil types, and recommendations for best practices for use of this device have been made The new device allows

application of overburden stress to the specimen, measurement of specimen volume

change due to suction and overburden pressure changes, and the use of a single specimen for determination of the SWCC The evaluation includes issues of temperature control, overburden stress, air diffusion through high air entry disks, and a study of potential sources of error in water content determination for SWCC’s It was concluded that the new oedometer device can be used to obtain accurate SWCC determinations on a single

11

specimen up to 1500 kPa General recommendations for appropriate corrections to SWCC data obtained using axis translation pressure plate-type devices are made

1V

DEDICATION

To my family and husband

ACKNOWLEDGEMENTS

This work was supported by the National Science Foundation under the grant No.CMS-0099800 The author is truly indebted to Dr Sandra Houston and Dr W.N Houston who contributed so much to the completion of this dissertation, most the ideas that were followed during the testing came from them The innumerably meetings and

phone conversations in which they spent so much time explaining the advantages and

disadvantages of using different testing procedures and data analysis will never be

forgotten

Acknowledgements are given to Dr Delwyn Fredlund who has given so many

ideas during the development of this research project The author also appreciates the time that he took to answer some of her questions even when he was traveling overseas

Acknowledgments are also for Dr Claudia Zapata who is always willing to share

her knowledge of unsaturated soil mechanics and data analysis

This research work was also greatly supported by Dr Manuel Padilla, Dr

Christopher A Lawrence, and Peter Goguen, no words exist to show them the

appreciation for their help

Thanks are also given to the Mexican Transport Institute that provided support to the author to continue studying postgraduate studies

vi

Vil

3.3.2 Shear strength prediction based on empirical models -+ 33 3.4 Historical review on the volume change measurements and negative pore water pressure on unsaturated SOIÏS c-creererererrierrrrrtrrtrrrrtrrrtrrrrree 38

CHAPTER 4 EXPERIMENTAL PROGRAM, EQUIPMENT AND PROCEDURES 81

A.2 Index properties .c.ccccesecseseeseetecressesssereseeseseneersnsensncssencsceeneereanernsgaenensgetenens 81 4.3 Soil water characteristic CUIVe tEStING .scccseseneteeneesesseeeteeseaeteeneresseeeetens 81 4.3.1 Pressure plate apparafUS -cerereirherretrtrtrrerdrrnntrrttrire 81 4.3.2 Specimen preparation .erererrerrrrrrrrdtrrrrrtrtrrrtrrreie 84 4.3.3 Specimen testing - eerrererirrrerrrrrrreddtrrrntltrrnrrrrnrire 86 4.4 Tempe celÌ testing - -cccccchhtnhrhdtrdtrrrrdrdirrdrrdminrtrntrrrndrrrr 91 4.4.1 Equipmen( -. cccnehhhhhthrrdtrrdtrdrdrrrrtridtrtrndrnrrrret 91 4.4.2 Specimen preparation eeceeeerrrreeerrrerrrtrerrrrtttrrrrrrtrrdrrrie 92 4.4.3 Testing ofthe specimen - -cccerreeerirrterrrrrdirtdrrrrrtnrdrrrtrnrire 92 4.5 Triaxial testing Traditional syst€mS -.«ceceeerrrrrrrrrtrrrtrtrnrrttrid 93 4.5.1 Descriptlon OfapparafUS -ecerererrrrrrrtrrrderrrrrdndrtrrdtrrtrnitrdrre 93 4.5.2 Traditional systes -cc-cccenenhhhhhrhhHrdrrrrrrdrddrrnittrdrrrtrrid 95 4.5.2.1 Pore wafter controÌ mneasuremenI - -‹ - «+ nhe nh nhe nh nh 95 4.5.2.2 Pore air controÏ meaSurem€nt - -‹ «che hhhhhhhhrhennntreem 96 4.5.2.3 De-airing the high air entry stone « ceềehhhnhhenrnnernnertrnenrin 96

VIH

4.5.2.4 Compaction of the specimen ‹ «cành nhehh nhe nhhnhhenenree 97 4.5.2.5 Setting up the sample in the system -.- + hhnhhhhhritmre 98 4.5.2.6 Shearing of specimen -.‹ cà nền nh nhhnhehehh he hen Hong 99 4.5.3 Advanced triaxial sySf€ITS - ch HH thư 100 4.5.3.1 Triaxial system and modIfications - «+ ch hhhhhhehetmrreree 100 4.5.3.2 Modifđed triax1al celÏ ‹ căn nh nh nh nh kh hy nh 101 4.5.3.3 Total volume change cOrr€CfIOPS ‹ cà cà nen he nhe HH 104 4.5.3.4 Pore-water controÏ sySf€m - «cành nh nh nhe nh nh nhớ 106 4.5.3.5 Pore air contro] sySfem ‹-‹ sec sàn nh nh nh nh khe Hinh nhe 108 4.5.3.6 Volume change measuremenf sysfem - «+ hhhhe nhe 108

4.5.3.8 Electronic software and hardWar€ cà nen he nhe 112 4.5.3.9 Data refrICVAÌ - ng ng ng Ki Ki kh th nĩn Bà th ng 114 4.5.4 Testing procedure for advanced triaxial sySt€IS -‹ -ccccceeneiee 114 4.5.4.1 Preparation of specimen for saturated and unsaturated

4.5.43.1 — Suction control testing sDeCImeH - hé nhnheh 117

45.4.3.2 Pre-equilibrated sample or suction adjustmeni : 117

1X

Page 4.5.4.3.3 Specimen compacted at a desired water content which corresponds

to a desired matric suction VAÌU© no SH HT n nh n Kế BH nh nh cá nh ch 120 4.5.4.4 Assembling and filling the pressure ceÌÏ - - {nề ewes 121 4.5.4.5 Pulsing f€SÍS HH» nh nh BE ng km hà Khen kề kn n th kh bu 122 4.5.4.6 Isotropic consolidation SfaØ€ nh nh nh nh kh nhe 123 4.5.4.6.1 Specimens tested under sucfion control ‹-‹-«- 123

4.5.4.6.2 Pre-equilibrated specImens - sec che 124

4.5.4.6.3 Specimen compacted at a desired water content which corresponds

to a đesired matric suctlon vaÏUe « «cà cà se 124

4.5.4.7 Shearing specÏmen - ch nh nh nh nh HH kh te 125 4.5.4.8 Reliability of the resuÏfS nen kề nh kh nhe hinh 126

CHAPTER 5 DESCRIPTION OF TEST SOILS -cc cà ieeHhee 128

6.2.4 Recommendation for correction to SWCC àc che 159

Page 6.3 Other important aspects for determination of soil water characteristic curves 162 6.3.1 Correction of the soil water characteristic curve for 100 % saturation 162 6.3.2 Measurement ofmatric suction at very low vaÌues c.ceceiee 167

6.3.3 Tmportance ofmatching ceramic stone to suction range/soil type 169

6.3.4 Diffused air through the high air entry ceramic stones during SWCC

(505207 173

6.3.5 Effect of overburden pressure on the SWCC ccecehheeere 176

6.3.6 Shrinkage during suction appliCatiOn ‹-ccsssseheinhieerrererriree 188 6.3.7 Effect of contact between the specimen and ceramic stone 191 6.4 Final soil water charaCf€TISẨICS uc ng HH Hi it 193 6.5 Correlation of air entry value with basic soil properties cesses 202 CHAPTER 7 TRIAXIAL TESTING SATURATED AND UNSATURATED SHEAR

S126 209

7.1.1 Sensitivity ofthe $° at low matric suction -ccccreeererrrrrirrriree 220 7.2 ASU east soil (SM| SH Hà HH Hàn HH HH tà Hà th 221

7.2.1 Saturated testing Traditional system -:sccssthhheehhrrerree 221

7.2.2 Unsaturated shear strength (Traditional triaxial system)-ASU east soil 223

7.2.3 Unsaturated shear strength (Advanced triaxial system) —ASU east soil 225

7.3 Price Club soil (CL-ML) ca SH HH nhau the 237

7.4 Sheely soil r€SUÏ(S cà nh HH HH re 245 7.4.1 Saturated f€sting cuc nhìn nh HH nàn kh bề kh nh nh tk ng 245

XI

7.4.2 Unsaturated t€StInE co HH HH nh nh nh kh HH k HH 245 7.5 Volume change m€aSUT€Tm€TS .- cv nhi HH Hi ha 250 7.5.1 Calibration ofthe double walled cell «cà se nh hưe 250 7.5.2 Volume change measurements for ASU east soil (SM) 253 7.5.3 Price Club soil -Volume change measurernens - - 258 7.5.4 Sheely soil -Volume change measuremenfS ‹‹ +: 262

7.6 Comparison of suction control procedure, “pre-equilibrated” procedure and

specimen compacted at a đesired wafer content/suction vaÌue 265

CHAPTER 8 MODEL FOR THE PREDICTION OF ð? VALUE . - 271

CHAPTER 9 CONCLUSIONS AND RECOMMENDATIONS 289

9.1 Conclusions with respect to soil water characteristic curve testing 289 9.2 Unsaturated triaxial testing COMCIUSIONS cssseseresreeteeteeeeeteeeeseneretereenees 293

PS NN:(cuou oi ae 297

;$232:4:0/0 21 .ỐỐ 299 APPENDIX

A OPERATING PROCEDURES FOR THE ADVANCED TRIAXIAL SYSTEMS

xH

Page CHARACTERISTICS OF SAMPLES FOR SOIL WATER CHARACTERISTIC CURVE TESTING HH HH HH KH TH HH HH HH nhiệt 342

UNSATURATED SHEAR STRENGTH RESULTS FOR YUMA SAND

TRADITIONAL TRIAXLAL SYSTEM che 344

UNSATURATED SHEAR STRENGTH RESULTS FOR ASU EAST SOIL TRADITIONAL TRIAXLAL SYSTEM St nnehhhhiHrhreree 351

UNSATURATED SHEAR STRENGTH AND VOLUME CHANGE FOR ASU EAST SOIL PRE EQUILIBRATED SAMPLES ADVANCED TRIAXIAL SQYSTEMS ung Ho HH Hư TH KH HH g0 1011101510111 101801001110 357

UNSATURATED SHEAR STRENGTH RESULTS FOR PRICE CLUB SOIL

UNSATURATED SHEAR STRENGTH RESULTS FOR PRICE CLUB SOIL TRADITIONAL TRIAXIAL SYSTEM ceccccescesesseessesecseeeteenessesneeneeeenaeees 381

UNSATURATED SHEAR STRENGTH RESULTS FOR SHEELY SOIL

ADVANCED TRIAXIAL SYSTEM uu .cccccescesseeeseessecseeseeeesssesssessessessenees 392

xII

SOUS †€XẦUTE HH0 11101 HH HH TH TH ưu 130 Gradation Of SOI SapÏ€S 2 tt th nh ng nryếu 131 Selection OÝ C€TaTnIC SỈOTIS th TH TT HT TT ng no 172 Soils to be used in the air entry value correlation cccececscecsecsrsceea 205 Summary of shear strength at failure for Yuma sand and corresponding o° 214 Computation of o° based upon the total cohesion intercepfs 216 Characteristics of samples tested under saturated conditions 221

o° values for ASU east SOiL s.s.sccsssssssssssccsssesssssscssssesecsssssssecessssssscsessssssssevessen 231

Computations of 6° based upon the total cohesion InferCepfsS 232

Water content of samples after shearing (samples tested on modified triaxial SYSCOM) oo.cecscccscssessscsetsteescsesesessecssscscsesesssevsvevssscssssssvacsessavsccavaceavavasaseuvavavseavas 235 Summary of soils for which a and b parameters were obtained 286 Summary of a and Ð paTaIn€f€FS Sàn TT ng ng tren 286

XIV

Predicted SWCC based on the Dạo and wPI (Zapata, 1999) cccccscersey 13 Comparison of a soil water characteristic curve obtained in laboratory versus soil water characteristic curve predicted with Zapata”s mođel - se cs se srsea 14

Yuma sand soil water characteristic curves versus predicted curve 15

Flow chart for hierarchical approach (From Fredlund, Houston, and Houston,

2002) HH 11111111 TH HH HH Họ ty 18

Typical relationship between the shear strength and matric suction (From Gan et

al 1988, cited by Khalili and Kabbaz 1988) - net re 28 Definition for a and b parameters .ccccccccsccccsessscsscssscssescsessvscsesesvscseasesvscuanacees 35

Small volume of soil considered to derive the equation (From Brahtz, Zanger and

Bruggeman, 1939) oo csscssssesssesssssssesscscscesscsvscssssecsesscsssceseevaceeavevsvarscacseseas 39 Layout of the apparatus for triaxial compression tests with pressure measurement

XV

Figure Page 3.9 A series of tests made on specimens of different mixtures of clay and silt (After

3.10 Clay specimens (After Gibbs and Coffey, 1963; Gibbs, 1965) - - 51

3.11 Exposed end plate test for measuring negative pore water pressure (Gibbs and Coffey, 1963) cccsssesesecsssensesscnessseesececsscesecseetsesseseeseecsesessssssseseesssseeasensens 52

3.12 Pore pressure apparatus as operated with exposure at the center of the end platen (€105-100i2 0650 .a.a 53 3.13 Schematic of pressure cell and appurtenances (Langfelder, 1964) 55 3.14 Initial negative pore water pressure versus water content for Goose Lake clay

3.15 Initial negative pore water pressure versus specific surface for samples compacted with 42.4 psi (Langfelder, 1964) c1 12111121 1111611011101 1 Hy tin Hết 58 3.16 Triaxial cell and volume change device (Garlanger, 1970) ccccssssseseseeessesees 59 3.17 Pore pressure changes and volumetric strains during shear (specimen 8-3, 03 =

188 psi) (Garlanger, 1970) .Ả 61 3.18 Isometric view of failure surface (afier Garlanger, 1970) .-cccccsreee 62 3.19 Projections of failure surface onto strength-applied stress and strength-capillary E5 (€rlät›:1x106 0017 63 3.20 Schematic of the system (Cambell, 19773) ác tt SH HH ren 64 3.21 Volumetric strain for Peorian Loess (Cambell, 19773) - + +cccsvxsvscssses 65 3.22 Volumetric strain for Champaign TiII (Cambell, 1973) c.ccccceskexes 66 3.23 B-coefficients for Peorian Loess (a) and Champaign Till (b) (Cambell, 1973) 67 3.24 Double-walled triaxial cell (Wheeler, 1986) uc nh Ha rêu 68

XVI

A new total volume measuring system for unsaturated soils (Ng, et al 2002) 73 Cylindrical specimen of sand with mounted collars (Kolymbas and Wu, 1989) 74 General arrangement and instrumentation (Hird and Yung, 1989) 75 Progressive development of the lateral profiles of the specimen (Romero et al,

2 76 a) Schematic of the controlled suction system; b) Location of LVDTs (Blatz and i1 0820900 00007 .ố.ố 78 Modified high pressure cell with controlled suction (Blatz and Graham, 2000) 79 Set of new pressure plate apparatuses at Arizona State niversity - 82

(a) Grooved water compartment in which stone is placed, (b) stone set on bottom Plate ốố cna 83 Specimen safuratiOI se S2 12 nh HH HH1 H11 86 Tempe Cell ccccesecscssescssessrererensescsserstsnsncnensnessecessseseensrssssssseeerenenessensnennenenetseas 91 Traditional Triaxial systems to test small samples (d=7.1em, h=171cm) 94 Ceramic stone placed on top of the bottom pedestal (Picture taken from Lawrence, 2004Ì) ch, 1H H0 Hà hà Hà Ha 1 101011 1g 95

Mold and tamping rods to prepare specimens of 7.1 cm diameter and 17 cm

height nh HH2 00 HH HH HH HH0 111g 98 Advanced Triaxial System Arizona State University Single pressure cell 100

Cell expansion test (Lawrence, 2004) chia 102

XVil

Representation of factors which influence cell volume change measurements: (a) components due to specimen, (b) effects during consolidation, and (c) effect of piston penetration during shearing stage (Head, 1998) ciee 105 Bottom pedestal in which the high air entry cerarnic đisk 1s glued 107 Pressure/volume cylinders (From Lawrence, 2004) che 109 Main screen of the GCTS controller for unsaturated testing .-‹ - 113

Photograph of the tampering rods, compaction mold for 4 in X 9 in samples and

New heater placed on the wall of the soil water characteristic curve cell 155

Soil water characteristic curve obtained after a new was placed 156 ASU east soil water characteristic curve run with modified cell and jacket heater

Corrected points for a Yuma sand {€Sf - ác chinh 161 Soil water characteristic curve for Rocky Mountain soil in which corrected points

Comparison of two ASU east soil water characteristic curves that started at

different degrees of saturation .cccccceesesseseseseeeeeetseesteeeseessetecseeesenensenensees 163 Soil water characteristic curves when samples are compacted statically and

dynamically (From Olivera, 2004) chau 165 Soil water characteristic curves for Ocotillo soil (as-measured and corrected for 0002-19 0 ố 167 Pool of water on high air entry ceramic stone

xix

Figure Page 6.25 Effect of the high air entry ceramic stone on the equilibration time .4 134 6.26 Soil water characteristic curves for Yuma sand when using 1 and 5 bar stone high 8lT ©€n†rY C€TạIC S{OTI€S nhàn Hi Hit Hi Hà Hi Hà 171 6.27 Diffused air through 15 bar high air entry ceramic stones (thickness = 0.71 cm)

sevevseescsdevsasesnsevseseesuensensssecsessessseaeenseasesessaseesenssssaussesansusessseneonsessuensnseesesaseneagen nas 134

6.28 Air đifusion through 5 bar high air entry ceramic stOn€S :-‹-:-c sẻ 176 6.29 Summary of soil water characteristics for Price Club soil for three different COnfining PF€SSUT€S tt 12121112 tà th th HH Hi 0 10 1 0 134 6.30 Summary of soil water characteristics for Aurora soil for two different confining DTSSUT€S nh HH HH Hà kg HH 011 22101110 178 6.31 ASU east soil water characteristic curves for three overburden pressures 179

6.32 ASU east soil water characteristic curves for three overburden pressures without

Ẩi{fÏE CUTVS cá Lành HH 1211111 1.11 101kg g1 0101 180 6.33 Increase in air entry value with pseudo net normal stress (Vanapalli, 1994) 181 6.34 Soil water characteristic curves for wet of optimum condition and three net

6.35 e vs log s plot for e = 0.52 specimen ccc chu khe 183

6.36 Increase in air entry value with net normal stress from unsaturated shear strength (Vanapalli, 1994) .cccnnH212112H HH HH 184 6.37 Soil-water characteristic curves: volumetric water content (a) and degree of saturation (b) versus matric suction with variation of the net confining stress (From Lee et al 2005) cọ nh HH Hi HH HH TH 186 6.38 Effect of ko stress condition on the SWŒCs (From Bao and Ng, 2000) 187 6.39 Shrinkage of the soil specimen at the end of the tesf -cccoreheehree 189 6.40 Soil water characteristic curve obtained with and without correction for vertical and horizontal deformation at the last pO1nIf «ch nhhhhhehere 190 6.41 Yuma sand soil water characterIsfIC CUTV€S share 192

XX

Measured and predicted air entry values for soils with PI>0 - 208 Saturated triaxial tests for Yuma sand oo cccscsccseessseesecreeseeeessereeessseseeeceneenes 210

Extended Mohr-Coulomb plot for Yuma sand for tests run with 20 kPa suction

ÔÒÔ 212

XXI

Extended Mohr-Coulomb plot for Yuma sand tests run with 250 kPa suction 213 Extended Mohr-Coulomb plot for Yuma sand tests run with 400 kPa suction 213

Shear strength at failure as a function of suction for three net normal stress values

Unsaturated shear strength for an average suction of 345 kPa 229

Unsaturated shear strength for an average suction of 625 kPa -.-: 230

o° values calculated using cohesion intercepts and points at failure 232

Suction versus shear strength at failure for different confining pressures 233

Distribution of water content along the specimen for a suction of 600 kPa 234

Water content at the end of shearing specimen superimposed on the SWCC curve 100.18 1 236 Price Club soil effective stress paraI€f€rS cnnenhhhHhhnehhhehihihe 237

XXii

20, and 75 KPav.cccscsccsesseensesssssssssessesessessseseesesneaceetensensesecsesssessesenssenenenesenenenens 246 Unsaturated soil testing for 500 kPa suction and net normal stress of 20 kPa 247 Unsaturated soil testing for 725, and 685 kPa suction and net normal stresses of

20, and 75 kPa càng tt Hà HH HH HH HH Hit H4 213

Suction versus increase in cohesion intercept due fO sucf1OT -cee 249 Suction versus 6° for Sheely soil ccccerieerirrttrrrrieirdrirriiriien 249

Calibration of đouble walled cell for system no Š ‹ c che 251 Calibration of double walled cell for system no Ố -sàc cành 525 Volume change measurements for 20 kPa net normal stT©ss -: -: 253 Volume change measurements for 75 kPa net normal Sfr€SS - 254 Volume change measurements for 250 kPa net normal str€ss - 255 Volume change measurements for 250 kPa net normal stress . -: 256

XXII

Axial deformation versus deviator stress for net normal stress 20 kPa and different matrIc suc†ion VẠU€S L c cọc n ng ng vế 260 Volume change measurements on Price Club soil at 250 kPa suction 261

Axial deformation versus deviator stress (Price Club soil at 250 kPa confining

9x 1 262 Volume change behavior Sheely clay samples for 20 kPa net normal stress 263 Volume change measurements for Sheely clay (Gz-uạ = 75 kPa) 213 Comparison of volume change behavior for CL-ML and CL samples 264 ASU east soil Shear strength plot to compare the effect on the three methods 267

ASU east soil Volumetric strain following the three procedures 268

Three samples of ASU east soil run under three procedur€s - 269 Volume changes of three samples run under three different procedures to attain TEQUIFE SUCIOTI So hình g2 11 T11 1H11 HH te 270 ASU east test results of suction Versus 0° .cccssssssecsecsecsseecsessteessesneesneeesneessneesns 273

Hyperbolic shape for the suction versus o° values for a clay of low plasticity 274

Physical interpretation of the b paraIm€€F cà Sàn nhhhhe 275 Transformed hyperbolic model L -‹- các cà ct ng tt He Hà 277 Transformed plot for Red clay data presented by Escario and Juca (1989) 280 Transformed hyperbolic model for Madrid clayey sand ccscseeseee eres 281

XXIV

Figure Page 8.7 ASU east soil (analysis of data based on points at failure) 282 8.8 Transformed hyperbolic model for Sheely clay ssc teers 283 8.9 Transformed hyperbolic plot for Price Club soil -.-:ersrerereeerererrre 284 8.10 Transformed hyperbolic model for Vanapalli°s đafa -ccceneeeeereee 285

8.12 Correlation for b paraImef€T - + cà cnhhhhhhhhrertHrrrrerirrrrdrrrrnrinti 288

XXV

NOMENCLATURE

American Soil Testing Materlals

Arizona State University

Area total

Effective water content area

Air entry value

Increase in shear strength due to suction for Khallili et al approach Initial specific volume of free air at pressure po

Specific volume of free air at pressure po

Cohesion in Coulomb’s equation

Total cohesion of Fredlund’s shear strength equation

Effective cohesion intercept

Clay of low plasticity

Soil classified with double symbol

Maximum cohesion value

Percent clay

Diameter

Grain diameter corresponding to 10 % passing by weight or mass

Grain diameter corresponding to 60% passing by weight or mass

Correction term for unsaturated conditions

Specific gravity of solids

Slope of the linear dropping of the soil water characteristic curve

Intercept of the linear dropping of the soil water characteristic curve

Residual degree of saturation when using soil water characteristic curve

Soil water characteristic curve

Pore-air pressure

XXV1

Pore-water pressure Matric suction

Is the matric suction in excess of the air entry value Bubbling pressure or air entry value

United States Gravimetric water content Optimum water content obtained from Standard Proctor test Saturated volumetric water content

Weighted PI = Passing 200 X PI

Fitting parameter that corresponds to c’

Fitting parameter Fitting parameter

Bulk density Dry density Maximum dry density obtained from Proctor Standard test

Is the change in volume of adsorbed air at pressure po

Bishop’s fitting parameter

Fitting parameter Effective angle of internal friction Angle indicating the increase of shear strength due to suction Fitting parameter for Vanapalli’s equation

Matric suction Air entry value Normalized volumetric water content Volumetric water content

Volumetric water content

Volumetric water content under fully saturated condition

Residual volumetric water content Total stress

Is the effective stress Hilf's equation Net normal stress on the surface of failure Net normal stress

Net normal stress Cell pressure or minor principal stress

Major principal stress Shear strength at failure

Test parameter Shear strength at failure Shear strength contribution due to suction

Is the drained shear strength of saturated soil

XXVH

CHAPTER 1: INTRODUCTION

1.1 Overview

Lately unsaturated soils has become a subject of great interest because of

the increasing awareness that the principles developed for saturated soil

mechanics can be applied partly to unsaturated soil problems An example is the use of the effective stress law in the shear strength equation which appears to be applicable up to the air entry value of the soil Even though many researchers have tried to modify the effective stress equation for the case of unsaturated soils

by introducing different parameters related to soil suction, none have been

successful so far

The engineer is basically interested in the mechanical response of soils because it is the material in which structures are built or it is the material from which some structures are constructed, as is the case of embankments,

highways, liners, etc The response of unsaturated soils to application of loads

(which simulates the load of an engineering structure) is studied based on

volume change and shear strength response For many years, these responses have been the subject of many investigations, however, researchers have

realized that shear strength of unsaturated soil is not extremely hard to

determine, though it may be time-consuming, but it is the volume change

measurement that is troublesome due to the many factors which affect it

Needless to say, to run unsaturated soil testing in the laboratory to

determine the mechanical properties in which the engineer is interested requires

2 that the laboratory be equipped with advanced and very complex equipment that

is very costly and is also complex to work with on a routine testing basis Thus, if

empirical models can be determined to predict these unsaturated soil properties based on easily determined index properties and soil water characteristic curves

of the soils, this would be of great benefit to the engineering community who can use this information as a preliminary tool to determine an approximate soil

behavior without the necessity of conducting lengthy and complex testing

The objectives of this research are threefold:

1 Improve the procedures for performing unsaturated soil testing:

Soil Water Characteristic Curves (SWCC) The improvement in the

procedures will take into account the temperature control and related

compensating corrections, the importance of the improvement on specimen- ceramic stone contact, the proper use of the high air entry ceramic stones

depending on the suction range or soil type, relevance of and corrections for diffused air, and use of apparatus that allows for volume change determination during SWCC measurement

Triaxial Testing Volume change has been one of the most challenging properties to determine during unsaturated soil triaxial testing This part of the objective will be the study of volume change determination during unsaturated soil testing, and will include evaluation of proper corrections to volume change

measurements during consolidation and shearing phases of triaxial testing The use of a double walled cell for volume change determination is studied and

recommendations for its appropriate use are made

Time required for unsaturated soils testing is of concern for

implementation into practice and with regard to development of a large database for evaluation of unsaturated soil behavior and properties Thus, decreasing the

time of unsaturated soil triaxial testing is one objective of this study Pre-test

specimen equilibration methods were used to reach a desired soil suction, and the results compared with results of specimens tested when they are compacted

at desired water content (suction) values, as well as the comparison to samples tested under the longer-test-time method of the suction control (a given suction is imposed and the sample is allowed to equilibrate by drawing or losing water)

2 Determine the shear strength and volume change behavior of saturated and unsaturated soils to obtain 4’, c’, and 6°

This objective includes the development of a database compiled by

extracting data from published literature, dissertations, and other documents During the selection of the information, care was taken to include only results in which the authors presented a complete explanation of the procedures followed

to obtain the information When there was ambiguous information given in a particular literature source, this resulted in the study not being used in this

research to reduce the variability of the database This database is

supplemented with an extensive database of results of unsaturated soil testing

obtained by the author from samples which were sampled at various U.S

locations and represent a wide range in soil type For shear strength

determination the triaxial tests were performed using traditional triaxial cells,

modified with high air entry disks for suction control, and also with advanced computer controlled triaxial unsaturated soils testing systems The advanced unsaturated soils testing equipment included capability of shear strength

determination and volume change determination because of the use of the

double walled cell for triaxial chamber

3 The third and last objective is the evaluation of all the information gathered to find relationships between volume change behavior and shear strength with soil water characteristic curve parameters, saturated soil properties, index properties, and gradation

The organization of the research is outlined briefly below:

Chapter 1 It is used to describe the main objectives of this research work

Chapter 2 This chapter describes the hierarchical approach that can be

followed to determine unsaturated soil property functions based on direct

measurement with advanced unsaturated soils testing equipment to

determination of properties using saturated soil properties and other index properties

Chapter 3 Background on shear strength and volume change

measurements that have been or are current in use is presented, emphasizing that volume change measurement during triaxial testing is a problem that can be solved by taking into account the many factors which influence its measurement

Chapter 4 Procedures that were followed during the development of the testing methodologies used in this research are presented The procedure to determine soil water characteristic curve was previously presented by Perera (2003), but some slight modifications to improve the procedure are presented

The procedure to determine unsaturated triaxial properties on the

advanced triaxial systems was originally presented by Lawrence (2004), with

emphasis on hollow cylinder testing His procedure was modified to include the

use of the double walled cell and a flushing device at the base of the high air

entry disk

Chapter 5 Presents the index properties and general soil characterization

for several soils that were used in this research The rationale for selection of

soils for triaxial shear strength and volume change determination is discussed

Chapter 6 The results and findings on the soil water characteristic curve testing are presented Soil water characteristics curves are shown for some of the 11 test soils that were sampled at U.S locations These curves were

determined for overburden pressures of 20 kPa, 75 kPa and 196 kPa The soil

water characteristic curves were obtained by using a set of modified pressure plate apparatuses For sandy soils, the Tempe cell was also used in the study at low suction values Many testing conditions affecting SWCC testing results are evaluated with respect to their impact on accuracy of results These conditions

include temperature effects, air entry value of the stone, volume change

determinations during SWCC testing, and equilibration times

Chapter 7 Provides results of a comprehensive unsaturated soil triaxial

testing program carried out under net normal stresses of 20, 75, and 250 kPa

The suction was varied from 20 kPa to 700 kPa for the advanced triaxial systems and from 50 to 400 kPa for the modified traditional systems The procedures that were followed are: 1) Suction control testing, 2) pre-testing equilibration method

to reach a desired soil suction, and 3) specimen compacted at a desired water content which corresponds to the desired matric suction value Shear strength and volume change data are presented for 3 test soils ranging in unified soil

classification from SM to CL.

7

Chapter 8 The development of correlations for unsaturated shear strength

behavior with index properties, saturated soil properties, and SWCC data are presented in this chapter Recommendations for estimating unsaturated soil

properties are presented and an extensive discussion of findings is given

Finally, in chapter 9, some conclusions are drawn from the research

carried out and recommendations are proposed for future research activities.

CHAPTER 2: HIERARCHICAL APPROACH TO UNSATURATED SOIL

PROPERTY ASSESSMENT

2.1 Introduction

Unsaturated soil behavior has been a matter of research since the 1930’s

in the Civil Engineering field, but before in the field of Soil Science Although a great amount of knowledge has been put forward by many researchers it remains

a science that can not be put into practice and probably the primary drawback for its implementation is the excessive cost required to experimentally measure

unsaturated soil properties Practicing engineers generally would not invest in

sophisticated and costly testing equipment Furthermore the time of performing a single test may prohibit almost all geotechnical engineering firms from testing unsaturated soils (Lawrence, C., 2004)

Lately, the use of the soil-water characteristic curve has been shown to be one of the keys to the implementation of unsaturated soil mechanics as a means

of quantifying unsaturated soil property functions The concept of use of the SWCC for estimating unsaturated soil properties is presented in this chapter

2.2 Unsaturated soil property functions

The quantification of unsaturated soil property functions, more than any other single factor provides the key to the implementation of unsaturated soil mechanics in geotechnical engineering practice The main challenge involves the

determination of economically viable procedures for the assessment of

unsaturated soil property functions (Fredlund, Houston, and Houston, 2002)

Among the most common unsaturated soil property functions are the hydraulic conductivity function, water storage functions, shear strength function, and void ratio function Each of these functions can be used when solving

engineering problems related to slope stability, seepage analysis, lateral earth

pressure behavior, heave analysis, and soil collapse, among others This

research is only concerned with the shear strength and volume change functions

2.3 Assessment of unsaturated soil property functions

Difficulties have been encountered when testing unsaturated soils

because of the lengthy times to obtain unsaturated soil properties in addition to the problems associated to the equipment and materials used during testing (e.g cavitation of the measuring system, leakage of air through high air entry ceramic stones, and diffusion of air through membranes in triaxial testing) Also, the

testing procedures have proven to be costly from an experimental point of view because it involves the use of high precision pressure/volume controllers, high air entry ceramic stones, means for control of pore air and pore water pressures at

the same time, and transducers with high precision In addition, the unsaturated

soil property functions are generally non-linear, which further complicates the

problem.

10 Lately, many models and ways to predict the unsaturated soil behavior have been published in the unsaturated soil mechanics field, such is the case of

the hierarchical approach to the determination of unsaturated soil property

functions presented by Fredlund, Houston and Houston (2002), and Houston, S.L (2002) This hierarchical approach is depicted in Figure 2.1

Figure 2.1 Approaches to determine unsaturated soil property functions (From

Fredlund, Houston and Houston, 2002)

The categories or levels of the hierarchical approach to the assessment of unsaturated soil properties can be described as follows (Fredlund, Houston and Houston, 2002):

11 2.3.1 Level 1

In the words of Fredlund, Houston and Houston (2002), this level was

described as follows:

“, 1s intended for high profile infrastructure design involving significant cost or projects with profound implication in case of failure, research applications, large, costly projects of high risk, or where there could be potential cost savings

At this level, advanced unsaturated testing equipment and procedures are used, such as triaxial equipment, modified direct shear devices, hollow cylinder

devices, etc., capable of following a variety of soil suction and net normal stress paths along with the measurement of total and water volume changes”

At this level the soil properties needed are shown in Table 2.1

Table 2.1 Properties needed at level 1 of the hierarchical level

Index properties Gradation, specific gravity, compaction characteristics,

hydrometer analysis, Atterberg Limits, etc

Saturated soil condition properties | Triaxial testing on saturated soils to obtain c’ and o’, volume

change, hydraulic conductivity, consolidation testing, etc

Unsaturated soil properties Triaxial testing on unsaturated soil to obtain the relationship

suction versus shear strength, suction versus ÉP, volume change characteristics, unsaturated hydraulic permeability,

soil water characteristic curve

12 2.3.2 Level 2

In this level the soil water characteristic curve has to be obtained in order

to be used in the determination of unsaturated soil property functions This level also implies that conventional testing (saturated soil properties and index

properties) will be carried out and coupled with soil water characteristic curve data for estimation of unsaturated soil properties

The soil water characteristic curve can be obtained using tensiometers,

pressure plate apparatus, pressure membrane, Tempe cells, and relative

humidity-controlled dessicators, depending on which device is available and also the range of suction for which the soil water characteristic curve needs to be

obtained

2.3.3 Level 3

For Level 3 analyses, the intent is to use soil index properties and

correlations with index properties to estimate the soil water characteristic curve The required information may also be derived from a database of available

curves, for example, using the “Soilvision” database (Fredlund, 2006) It is

assumed that the saturated soil properties are known, or measured This level may represent the most common scenario used on routine geotechnical

engineering projects (Houston, S.L 2002; Fredilund, Houston and Houston,

2002).