USF Libraries
USF Digital Collections

Hydraulic, diffusion, and retention characteristics of inorganic chemicals in bentonite

MISSING IMAGE

Material Information

Title:
Hydraulic, diffusion, and retention characteristics of inorganic chemicals in bentonite
Physical Description:
Book
Language:
English
Creator:
Muhammad, Naim
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla.
Publication Date:

Subjects

Subjects / Keywords:
landfill
clay liner
coefficient of permeability
electrical conductivity
adsorption capacity
partition coefficient
retardation factor
Dissertations, Academic -- Civil Engineering -- Doctoral -- USF   ( lcsh )
Genre:
government publication (state, provincial, terriorial, dependent)   ( marcgt )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Summary:
ABSTRACT: Inorganic contaminants, while transported through the bentonite layer, are chemically adsorbed onto the particle surfaces and exhibit a delay in solute breakthrough in hydraulic barriers. Transport of inorganic leachate contaminants through bentonite occurs by advection, diffusion or a combination of these two mechanisms. During the process of chemical solute transport through low permeability bentonite, the amount of cation exchange on the clay particle surface is directly related to the cation exchange capacity (CEC) of montmorillonite and other mineral constituents. The process of diffusion and advection of various inorganic leachate contaminants through bentonite is thoroughly investigated in this study. Diffusion characteristics are of specific interest as they have a prominent effect on the long term properties of bentonite compared to advection. This is mostly true if the hydraulic conductivity of the material is less than 10-8 cm/s and if the thickness of the barrier is small. Chemical reactions in the form of cationic exchange on the clay particle surfaces has been incorporated in the analysis of the diffusion process. Adsorption-desorption (sorption) reactions of chemical compounds that influence the concentrations of inorganic leachates during transport in bentonite clay have been modeled using the Fick's fundamental diffusion theory. Partition coefficients of the solutes in pore space, which affect the retardation factor of various individual ions of chemical solutions, have been investigated during transient diffusion and advection processes. Several objectives have been accomplished during this research study. An evaluation has been carried out of the hydraulic conductivity of bentonite with respect to single species salts and various combinations of electrolyte solutions. Diffusion properties of inorganic leachates through bentonite have been characterized in terms of apparent and effective diffusion coefficients. Time-dependent behavior of the diffusive ions has been analyzed in order to determine the total retention capacity of bentonite before electrical conductivity breakthrough and steady-state chemical stability are reached. An analytical solution of the attenuation of various inorganic ions concentrations through bentonite has been developed. Finally, recommendations were made for landfill liners exposed to highly concentrated inorganic leachates.
Thesis:
Thesis (Ph.D.)--University of South Florida, 2004.
Bibliography:
Includes bibliographical references.
System Details:
System requirements: World Wide Web browser and PDF reader.
System Details:
Mode of access: World Wide Web.
Statement of Responsibility:
by Naim Muhammad.
General Note:
Includes vita.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 251 pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001478742
oclc - 56389463
notis - AJS2431
usfldc doi - E14-SFE0000383
usfldc handle - e14.383
System ID:
SFS0025076:00001


This item is only available as the following downloads:


Full Text

PAGE 1

Hydraulic, Diffusion, and Retention Characteris tics of Inorganic Chemicals in Bentonite By Naim Muhammad A dissertation submitted in partial fulfillment of the requirement for the degree of Doctor of Philosophy in Civil Engineering Department of Civil and Environmental Engineering College of Engineering University of South Florida Major Professor: Alaa K. Ashmawy, Ph.D. Manjriker Gunaratne, Ph.D. A. Gray Mullins, Ph.D. Luis Garcia-Rubio, Ph.D. V.R. Bhethanabotla, Ph.D. Date of Approval: June 18, 2004 Keywords: Landfill, Clay Liner, Coeffi cient of Permeability, Electrical Conductivity, Retardation Factor, Partition Co efficient, Adsorption Capacity. Copyright 2004, Naim Muhammad

PAGE 2

DEDICATION To my beloved family

PAGE 3

ACKNOWLEDGEMENTS The author expresses his deep gratitude and sincere appreciation to his major supervisor Associate Professor Dr. Alaa K. Ashmawy for his valued advice, guidance, encouragement and constructive criticism thr oughout this study. The author wishes to express his sincere appreciation and thanks to all his committee members: Dr. Manjriker Gunaratne, Dr. A. Gray Mullins, Dr. Luis Ga rcia-Rubio and Dr. Thomas Pichler for their interest, useful suggestions and constant support in this study. The aut hor is also grateful to Dr. Audrey D. Levine and Ms. Barbara Dodge of Environmenta l Lab, Mr. Jay Bieber of Nanomaterials and Nanomanufacturing Rese arch Center (NNRC), Mr. Robert R Smith and Mr. Tom Gage Machine Shop for their technical s upport and using their facilities. The valuable discussions with and cons tructive suggestions from Mr. Darwish ElHajji, Ms. Maysson Sallam, and Ms. Jessica A Schenning are gratefully achnowledged. The author is also grateful to all his co lleagues for their constant help and support throughout this study. The author deeply appreciates the funds pr ovided in part by the Florida Center for Solid and Hazardous Waste Management and th e financial assistance in the form of Graduate Research Assistantship and Teachi ng Assistantship and facilities given by the Department of Civil and Envi ronmental Engineering of the University of South Florida.

PAGE 4

i TABLE OF CONTENTS LIST OF TABLES iv LIST OF FIGURES vi LIST OF SYMBOLS xiii ABSTRACT xv CHAPTER ONE: INTRODUCTION 1 1.1 Scope and Significance 1 1.2 Research Objectives 5 1.3 Dissertation Outline 6 CHAPTER TWO: MATERIALS AND METHODOLOGY 8 2.1 Bentonite in Landfills 8 2.2 Bentonite Clay 11 2.2.1 Basic Clay Mineralogy 11 2.2.1.1 Classification and Chemical Composition 14 2.2.2 Cation Exchange Capacity 20 2.2.3 Cation Replaceability 23 2.3 Permeant Characteristics 28 2.3.1 MSW Leachate 28 2.3.2 Ash Leachate 31 2.3.3 Other Sources of Inorganic Leachates 34 2.4 Water-Bentonite Interaction 36 2.4.1 Mechanism of Interaction 37 2.4.2 Diffuse Double Layer 39 2.4.2.1 Theory and Mathematical Models of DDL 39 2.4.2.2 Factors Affecting DDL 45 CHAPTER THREE: BENTON ITE CHARACTERIZATION 48 3.1 Source of Bentonite 48 3.1.1 Mineralogy Through XRD 48 3.1.2 Mineral Compositions 52 3.1.3 Chemical Composition 53 3.1.3.1 EDS Analysis 53 3.1.4 Electrical Conductivity and pH 57

PAGE 5

ii 3.1.5 Loss of Ignition 60 3.2 Grain Size Distribution 60 3.2.1 Hydrometer Test 61 3.2.2 Test Results and Discussion 62 3.3 Physical Properties 64 3.3.1 Specific Gravity 64 3.3.2 Atterberg Limits 66 3.4 Swell Index 72 3.4.1 Test Procedure 72 3.4.2 Effect of Chemical Solution Species 73 3.5 Cation Exchange Capacity of Bentonite 75 3.5.1 Methylene Blue Test Procedure 75 3.5.2 Test Results and Discussion 78 CHAPTER FOUR: EQUIPMENT DESIGN & FABRICATION 80 4.1 Permeability Equipment 80 4.1.1 Design Concept 81 4.1.2 Materials and Fabrication 83 4.2 Diffusion Equipment 87 4.2.1 Design Concept 89 4.2.2 Materials and Fabrication 91 CHAPTER FIVE: HYDRAULIC CHAR ACTERIZATION OF BENTONITE 93 5.1 Hydraulic Conductivity of Bentonite 93 5.1.1 Inorganic Chemical Permeants 94 5.1.2 Flexible Wall Permeability 95 5.1.2.1 Test Procedure 97 5.1.2.2 Sample Preparation 97 5.1.2.3 Sample Saturation 102 5.1.2.4 Permeation Phase 103 5.1.2.5 Termination Criteria 104 5.1.3 Rigid Wall Permeability 105 5.1.3.1 Sample Preparation 105 5.1.3.2 Permeation Phase 106 5.1.4 Factors Affecting Hydraulic Conductivity 107 5.1.4.1 Permeant Chemical Composition 107 5.1.4.2 Void Ratio 112 5.1.4.3 Hydraulic Gradient 115 5.1.4.4 Pre-hydration 120 5.1.4.5 Type of Permeameter 123 5.2 Chemical Analysis of Effluent 127 5.2.1 General 127 5.2.2 pH Measurement 127 5.2.3 Electrical Conductivity 128 5.2.4 Ionic Analysis 132

PAGE 6

iii CHAPTER SIX: DIFFU SION IN BENTONITE 134 6.1 Experimental Methods 134 6.1.1 Test Set-up 135 6.1.2 Sample Preparation and Procedure 138 6.1.2.1 Sample Preparation 138 6.1.3 Synthetic Inorganic Chemicals 141 6.1.4 Sample Collection for Chemical Analysis 142 6.2 Chemical Analysis 142 6.2.1 pH Measurement 143 6.2.2 Electrical Conductivity 147 6.2.3 Ionic Analysis 150 CHAPTER SEVEN: TRANSPO RT THEORY AND ANALYSIS OF DIFFUSION OF BENTONITE CLAY 153 7.1 Fluid Transport Mechanisms 153 7.1.1 Advection Flow 153 7.1.2 Diffusion Flow 155 7.1.2.1 Mathematical Solution to Diffusion Equation 159 7.1.3 Chemico-Osmotic Flow 162 7.1.4 Determination of Diffusion Parameters 164 7.2 Analysis of Diffusion Test Results 167 7.2.1 Lag Time and Time to Steady-State 167 7.2.2 Diffusion Coefficient 170 7.2.3 Retardation Factor 175 7.2.4 Partition Coefficient 176 7.2.5 Diffusion Coefficient Through Numerical Solution 177 CHAPTER EIGHT: SUMMARY AND RECOMMENDATIONS 181 8.1 Summary 181 8.2 Design Recommendation 183 REFERENCES 185 APPENDICES 204 Appendix A: Test Results of pH and EC of Permeability Tests 205 Appendix B: Test Results of pH and EC of Diffusion Tests 219 Appendix C: Ionic Analysis Test Results 230 ABOUT THE AUTHOR End Page

PAGE 7

iv LIST OF TABLES Table 2.1 Some Clay Minerals Charac teristics (after Mitchell, 1993) 21 Table 2.2 Radii of Ions 24 Table 2.3 Hydrated Radius of Cations 24 Table 2.4 Hydration Energy of Me tal Cations (after McBride, 1994) 28 Table 2.5 Chemicals in Leachate as Found by Different Researchers (after Reinhart & Grosh (1998) 32 Table 2.6 Chemical Composition of Two MSW Landfill Leachates 33 Table 3.1 Chemical Com position of Bentonite 57 Table 3.2 Dimensions of Me thylene Blue Single Molecule (After Taylor, 1985) 77 Table 5.1 Chemical Solutions Used in Hydraulic Conductivity Using Flexible Wall Permeameter 108 Table 5.2 Rigid Wall Permeability Te sts with Void Ratio Variation 113 Table 5.3 Flexible Wall Permeability Test s with Hydraulic Gradient Variation 116 Table 5.4 Flexible Wall Permeability Tests with Various Pre-Hydration Solutions 121 Table 5.5 Permeability Tests Using Flexible Wall and Rigid Wall Permeameters 124 Table 5.6 Lists of Flexible Wall Permeability Tests with pH Results 128 Table 5.7 Theoretical and Actual Chem ical Retention During Permeability 131 Table 6.1 Lists of Diffusion Samples with Source Solutions 143 Table 6.2 Lists of Diffusion Test s with Out-Fluxed pH Results 145 Table 6.3 Comparison of Diffusi on Tests with ‘Lag Time’ and Steady-State Equation 148 Table 6.4 Ionic Analysis of Diffu sant of Two Molar Solutions Through Bentonite 151 Table 7.1 Summary of Statistical Method for Steady-St ate Diffusion 169 Table 7.2 Lag Time and Time to Steady-State of Various Diffusants 170

PAGE 8

v Table 7.3 Worksheet for the Calculati on of Effective Diffusion Coefficient, D* of Various Cations 173 Table 7.4 Apparent Diffusion Coefficien t for Various Cations in Bentonite 174 Table 7.5 Retardation Factor of Various Cations in Bentonite 175 Table 7.6 Partition Coefficient of Various Cations in Bentonite 177

PAGE 9

vi LIST OF FIGUES Figure 1.1 Schematic Represen tation of Clay Particles 3 Figure 2.1 Cross-Section Sketches of Various GCLs 10 Figure 2.2 Diagrammatic Sketch Showing Clay Tetrahedral 12 Figure 2.3 Diagrammatic Sketch Showing Octahedral 12 Figure 2.4 Sheet Representation 13 Figure 2.5 Repeated Sheet Representation for 1:1 (Tetrahedral : Octahedral) Layer 15 Figure 2.6 Diagrammatic Sketch of th e Structure of the Kaolinite Layer (After Grim, 1968) 15 Figure 2.7 Charge Distribution on Ka olinite (after Mitchell, 1993) 16 Figure 2.8 Repeated Sheet Representation for 2:1 (Tetrahedral : Octahedral : Tetrahedral) Layer 17 Figure 2.9 Diagrammatic Sketch of the Montmorillonite 18 Figure 2.10 Charge Distribution in Mont morillonite (after Mitchell, 1993) 19 Figure 2.11 Schematic Diagram of the Structures of (a) Illite and (b) Vermiculite 20 Figure 2.12 The Three Mechanisms of Cation Adsorption on a Silicate Surface; e.g. Montmorillonite (after Sposito, 1989) 22 Figure 2.13 Schematic Diagram of the Clay Surface-Exchange Cation Interaction 27 Figure 2.14 Application of Bentonite in Embankment or Earthen Dam 35 Figure 2.15 Application of Benton ite in Manhole-Pipe Connection 36 Figure 2.16 Possible Mechanisms of Wa ter Adsorption by Clay Surfaces 38 Figure 2.17 Helmholtz Model 40 Figure 2.18 Gouy-Chapman Model 42 Figure 2.19 Stern Gouy-Chapman Model 42

PAGE 10

vii Figure 2.20 Effect of Concentration on Ion Distributions with Distance (after Mitchell, 1993) 46 Figure 2.21 Effect of Cati on Valence on Double Layer (after Mitchell, 1993) 47 Figure 3.1 Basics of X-ray Diffraction Technique 50 Figure 3.2 XRD Spectrometer Fundamentals 51 Figure 3.3 XRD Test Results for Bentonite 52 Figure 3.4 Schematic Diagram of EDS System 55 Figure 3.5 Spectrometer Fitted with Scanning Electron Microscope (HITACHI S-800) 55 Figure 3.6 Dry Bentonite Powder Under SEM 56 Figure 3.7 Energy Peaks for Bentonite Chemical Elements Using EDS 56 Figure 3.8 Accumet (Model AB 30) 4-cell Conductivity Meter 58 Figure 3.9 Accumet Portable (Model AP63) pH Meter 58 Figure 3.10 Electrical Conductivity a nd pH of Bentonite Suspension 59 Figure 3.11 Bentonite Particle/Aggregate Distribution with Various Inorganic Chemical Solutions of 0.1 Molar of Concentration 63 Figure 3.12 Bentonite Particle/Aggregat e Distribution with NaCl Solutions of Various Concentrations 64 Figure 3.13 Experimental Vari ation of Specific Gravity 65 Figure 3.14 Plasticity Chart (after Holtz and Kovacs, 1981) 66 Figure 3.15 Wyo-Ben Bentonite on the Plasticity Chart 68 Figure 3.16 Penetration vs Water/Solution Content (Water and 1 Molar Solution) 68 Figure 3.17 Penetration vs Water/Solution Content (Water and 0.5 Molar Solution) 69 Figure 3.18 Penetration vs Water/Solution Content (Water and 0.1 Molar Solution) 70 Figure 3.19 Penetration vs Water/Solution Content (Water and 0.01 Molar Solution) 70 Figure 3.20 Variation of Liquid Limits with Electrolyte Concentration 71 Figure 3.21 Variation of Liquid Limits with Types of Electrolyte Solutions 71 Figure 3.22 Swell Index of Bentonite in Inorganic Chemical Solutions 73

PAGE 11

viii Figure 3.23 Variation of Swell Index w ith Concentration of Salt Solutions 74 Figure 3.24 Methylene Blue Chemical Structure 76 Figure 3.25 Schematic Diagram of Methylene Blue Molecules (after Taylor, 1985) 76 Figure 3.26 Cation Exchange Capacity of Bentonite 79 Figure 4.1 Schematic Diagram of Permeability Test Setup 81 Figure 4.2 Schematic Diagram of Permeameter Cell 84 Figure 4.3 Schematic Diagram of Flexible Permeameter (a) Permeameter Cell (b) Bottom Connection 86 Figure 4.4 Specified Volume Diffusion Cell (After Lake and Rowe, 2000) 88 Figure 4.5 Diffusion Set-up with Clay Slurry (a) Initial Before Consolidation (b) Final After Consolidation 90 Figure 4.6 Modification of Porous Stone 92 Figure 5.1 Schematic Diagram of Flexible Wall Permeameter Set-up 95 Figure 5.2 Flexible Wall Permeameter 96 Figure 5.3 Components of Fl exible Wall Permeameter 97 Figure 5.4 Sample Preparation for Flexible Wall Permeability Test 99 Figure 5.5 Schematic Diagram of Rigid Wall Permeameter Set-up 106 Figure 5.6 Permeability vs. Duration for 1M Salt Solutions Using Flexible Wall Permeameter 109 Figure 5.7 Permeability vs. Pore Volume for 1M Salt Solutions Using Flexible Wall Permeameter 109 Figure 5.8 Permeability vs. Duration for All Salt Solutions (K-5, K-6, & K-7) 110 Figure 5.9 Permeability vs. Pore Volume for All Salt Solutions (K-5, K-6, & K-7) 111 Figure 5.10 Variation of Permeability with Molarity of Combined Salt Solutions 112 Figure 5.11 Variation of Permeability with Duration of 1M CaCl2 Permeant Used in Bentonite of Various Void Ratios 113 Figure 5.12 Variation of Permeability with Pore Volume of 1M CaCl2 Permeant Used in Bentonite of Various Void Ratios 113 Figure 5.13 Variation of Permeability with Duration of 1M NaCl Permeant Used in Bentonite of Various Void Ratios 114

PAGE 12

ix Figure 5.14 Variation of Permeability w ith Pore Volume of 1M NaCl Permeant Used in Bentonite of Various Void Ratios 114 Figure 5.15 Variation of Permeability with Void ratio for 1M CaCl2 and 1M NaCl Permeants 115 Figure 5.16 Variation of Permeability with Duration of Flow for K-5 117 Figure 5.17 Variation of Permeability w ith Pore Volume of Flow for K-5 117 Figure 5.18 Variation of Permeability with Duration of Flow for K-6 118 Figure 5.19 Variation of Permeability w ith Pore Volume of Flow for K-6 118 Figure 5.20 Variation of Permeability with Duration of Flow for K-7 119 Figure 5.21 Variation of Permeability w ith Pore Volume of Flow for K-7 119 Figure 5.22 Variation of k with A pplied Hydraulic Gradient in Combined Salt Solutions 120 Figure 5.23 Variation of Permeab ility with Duration of Flow for K-1 & K-9 121 Figure 5.24 Variation of Permeab ility with Duration of Flow for K-4 & K-10 122 Figure 5.25 Variation of Permeab ility with Duration of Flow for K-3 & K-13 122 Figure 5.26 Variation of Permeability with Duration of Flow for K-2 & K-12 123 Figure 5.27 Comparison of Permeame ters for DI Water Permeant (K-11 & KD-1) 125 Figure 5.28 Comparison of Permeameters for 1M CaCl2 Permeant (K-1 & KD-6) 125 Figure 5.29 Comparison of Permeameters for 1M MgCl2 Permeant (K-2 & KD-4) 126 Figure 5.30 Comparison of Permea meters for 1M NaCl Permeant (K-4 & KD-5) 126 Figure 5.31 Chemical Retention Measurement for KD-6 130 Figure 6.1 Schematic Diagra m of Diffusion Cell Set-up 136 Figure 6.2 Diffusion Set-up 137 Figure 6.3 Components of Diffusion Cell 138 Figure 6.4 Sample Preparation for Diffusion Test 141

PAGE 13

x Figure 6.5 Variation of pH for Group #1 Diffusion (D-6, D-11, D-12, and D-13) 145 Figure 6.6 Variation of pH for Group #2 Diffusion Tests (D-5, D-10, D-14, and D-16) 146 Figure 6.7 Variation of pH for Group #3 Diffusion Tests (D-8, D-9, D-10, and D-11) 146 Figure 6.8 Cumulative EC fo r Group #1 Diffusion Tests (D-6, D-11, D-12, and D-13) 148 Figure 6.9 Cumulative EC fo r Group #2 Diffusion Tests (D-5, D-10, D-14, and D-16) 149 Figure 6.10 Cumulative EC fo r Group #3 Diffusion Tests (D-8, D-9, D-10, and D-11) 149 Figure 7.1 Advection of Solute Transport 154 Figure 7.2 Mathematical Representation of Advection 154 Figure 7.3 Molecular Diffusion of Solute 156 Figure 7.4 Diffusion as a Func tion of Distance and Time 159 Figure 7.5 Chemico-Osmosis of Solute Transport 162 Figure 7.6 Induced Chemico-Osmotic Pressure Observed for Clay Membranes (Shackelford and Lee, 2003) 165 Figure 7.7 Cumulative Solute Ma ss Through Clay Specimen due to Diffusion (Shackelford and Lee, 2003; Malusis, et al 2001) 165 Figure 7.8 Diffusion Profile of Mg2+ Ions Using Numerical Method 178 Figure 7.9 Diffusion Profile of K+ Ions Using Numerical Method 179 Figure 7.10 Diffusion Profile of Na+ Ions Using Nume rical Method 180 Figure 7.11 Diffusion Profile of Ca2+ Ions Using Nume rical Method 180 Figure A.1 Ionic Analysis of Permea bility Test K-1 (a) Electrical Conductivity vs. Pore Volume and (b) pH vs. Pore Volume 205 Figure A.2 Ionic Analysis of Perm eability Test K-2 (a) Electrical Conductivity vs. Pore Volume and (b) pH vs. Pore Volume 206 Figure A.3 Ionic Analysis of Perm eability Test K-3 (a) Electrical Conductivity vs. Pore Volume and (b) pH vs. Pore Volume 207 Figure A.4 Ionic Analysis of Perm eability Test K-4 (a) Electrical Conductivity vs. Pore Volume and (b) pH vs. Pore Volume 208

PAGE 14

xi Figure A.5 Ionic Analysis of Perm eability Test K-5 (a) Electrical Conductivity vs. Pore Volume and (b) pH vs. Pore Volume 209 Figure A.6 Ionic Analysis of Perm eability Test K-6 (a) Electrical Conductivity vs. Pore Volume and (b) pH vs. Pore Volume 210 Figure A.7 Ionic Analysis of Perm eability Test K-7 (a) Electrical Conductivity vs. Pore Volume and (b) pH vs. Pore Volume 211 Figure A.8 Ionic Analysis of Perm eability Test K-8 (a) Electrical Conductivity vs. Pore Volume and (b) pH vs. Pore Volume 212 Figure A.9 Ionic Analysis of Perm eability Test K-9 (a) Electrical Conductivity vs. Pore Volume and (b) pH vs. Pore Volume 213 Figure A.10 Ionic Analysis of Perm eability Test K-10 (a) Electrical Conductivity vs. Pore Volume and (b) pH vs. Pore Volume 214 Figure A.11 Ionic Analysis of Perm eability Test K-11 (a) Electrical Conductivity vs. Pore Volume and (b) pH vs. Pore Volume 215 Figure A.12 Ionic Analysis of Perm eability Test K-12 (a) Electrical Conductivity vs. Pore Volume and (b) pH vs. Pore Volume 216 Figure A.13 Ionic Analysis of Perm eability Test K-13 (a) Electrical Conductivity vs. Pore Volume and (b) pH vs. Pore Volume 217 Figure A.14 Ionic Analysis of Perm eability Test K-14 (a) Electrical Conductivity vs. Pore Volume and (b) pH vs. Pore Volume 218 Figure B.1 Diffusion Test Results for D-5 (a) pH and Electrical Conductivity and (b) Cumulative EC versus Cumulative Diffusion Time 219 Figure B.2 Diffusion Test Results for D-6 (a) pH and Electrical Conductivity and (b) Cumulative EC versus Cumulative Diffusion Time 220 Figure B.3 Diffusion Test Results for D-8 (a) pH and Electrical Conductivity and (b) Cumulative EC versus Cumulative Diffusion Time 221 Figure B.4 Diffusion Test Results for D-9 (a) pH and Electrical Conductivity and (b) Cumulative EC versus Cumulative Diffusion Time 222 Figure B.5 Diffusion Test Results for D10 (a) pH and Electr ical Conductivity and (b) Cumulative EC versus Cumulative Diffusion Time 223 Figure B.6 Diffusion Test Results for D-11 (a) pH and Electr ical Conductivity and (b) Cumulative EC versus Cumulative Diffusion Time 224 Figure B.7 Diffusion Test Results for D-12 (a) pH and Electr ical Conductivity and (b) Cumulative EC versus Cumulative Diffusion Time 225 Figure B.8 Diffusion Test Results for D-13 (a) pH and Electr ical Conductivity and (b) Cumulative EC versus Cumulative Diffusion Time 226

PAGE 15

xii Figure B.9 Diffusion Test Results for D-14 (a) pH and Electr ical Conductivity and (b) Cumulative EC versus Cumulative Diffusion Time 227 Figure B.10 Diffusion Test Results for D-16 (a) pH and Electr ical Conductivity and (b) Cumulative EC versus Cumulative Diffusion Time 228 Figure B.11 Diffusion Test Results for D-17 (a) pH and Electr ical Conductivity and (b) Cumulative EC versus Cumulative Diffusion Time 229 Figure C.1 Concentration of Va rious Cations in Effluent During Permeability (K-1) 230 Figure C.2 Concentration of Vari ous Cations in Effluent During Permeability (K-2) 230 Figure C.3 Concentration of Va rious Cations in Effluent During Permeability (K-3) 231 Figure C.4 Concentration of Va rious Cations in Effluent During Permeability (K-4) 231

PAGE 16

xiii LIST OF SYMBOLS = Angstrom unit = 1 x 10-10 m Go = standard free energy change of the reaction o = Gibbs free energy at standa rd pressure and temperature Keq = thermodynamic equilibrium constant KE = revised equilibrium constant Eatt = electrostatic attraction energy Etot = total energy change Etot = overall change of energy rs = effective radius of the charge surface rA = radius of displaced ion A rB = radius of displaced ion B EA = hydration energy of displaced ion A EB = hydration energy of displaced ion B = electrical potential ch = charge volumetric density (C m3), D = relative permittivity of the medium D = Dielectric Constant D* = effective diffusion coefficient D*A = apparent diffusion coefficient o = dielectric consta nt of the void (C V-1 m-1) k = Boltzmann’s constant (1.38045 x 10-23 J/oK) k = coefficient of permeability T = absolute temperature i = ionic concentration of the specie i e = unit electronic charge (16 x 10-20 Coulomb)

PAGE 17

xiv e = void ratio o = electrical poten tial at concentration i o eq = equivalent charge fw = formula weight K = Debye-Huckel parameter = ionic valence m = mass of a particle v = velocity k = the Boltzman constant Ws = weight of the dry sample Wfs = weight of the flask filled with soil and water Wfw = the weight of the flask Gs = specific gravity of soil solids

PAGE 18

xv HYDRAULIC, DIFFUSION, AND RETENTION CHARACTERISTICS OF INORGANIC CHEMICALS IN BENTONITE Naim Muhammad ABSTRACT Inorganic contaminants, while transported through the bent onite layer, are chemically adsorbed onto the particle surfaces and e xhibit a delay in solute breakthrough in hydraulic barriers. Transport of inorgani c leachate contaminants through bentonite occurs by advection, diffusion or a combina tion of these two mechanisms. During the process of chemical solute transport through low permeability bentonite, the amount of cation exchange on the clay pa rticle surface is di rectly related to the cation exchange capacity (CEC) of montmorillonite a nd other mineral constituents. The process of diffusion and advection of various inorganic leachate contaminants through bentonite is thoroughly investigated in this study. Diffusion characteristics are of specific interest as they have a prominent e ffect on the long term properties of bentonite compared to advection. This is mostly true if the hydraulic conductivity of the material is less than 10-8 cm/s and if the thickness of the barrier is small. Chemical reactions in the form of cationic exchange on the clay partic le surfaces has been incorporated in the analysis of the diffusion process. Adsorptiondesorption (sorption) reactions of chemical compounds that influence the c oncentrations of in organic leachates during transport in bentonite clay have been modeled using the Fick’s fundamental diffusion theory. Partition coefficients of the so lutes in pore space, which aff ect the retardation factor of various individual ions of ch emical solutions, have been investigated during transient diffusion and advection processes.

PAGE 19

xvi Several objectives have been accomplished during this research study. An evaluation has been carried out of the hydrauli c conductivity of benton ite with respect to single species salts and various combinati ons of electrolyte solutions. Diffusion properties of inorganic leachates through benton ite have been characterized in terms of apparent and effective diffusion coefficients. Time-dependent behavior of the diffusive ions has been analyzed in order to determ ine the total retention capacity of bentonite before electrical conductiv ity breakthrough and steady-stat e chemical stability are reached. An analytical solution of the attenuation of various inorganic ions concentrations through bentonite has been de veloped. Finally, r ecommendations were made for landfill liners exposed to hi ghly concentrated inorganic leachates.

PAGE 20

1 CHAPTER ONE INTRODUCTION 1.1 Scope and Significance One of the main problems in the geoenvir onmental field is the intrusion of toxic contaminants from waste disposal and ot her sources into the underlying ground water supply. Clays are commonly used as barrier s in landfills, slurry walls, and similar structures to slowdown the movement of c ontaminants because of their higher water absorption capacity. Bentonite clays are also being used as buffers in nuclear fuel waste disposal sites to control the spread of ra dioactive materials in to the ground (Hancox, 1986; Cheung, 1994). Bentonite clay, when used in the field as a hydraulic barrier, comes in contact with various inorganic chemicals which even tually cause the perf ormance of bentonite clay to diminish in terms of permeability and chemical outflux (Anderson et al ., 1985; Cadena et al ., 1990; Chapuis, 1990; Cheung et al ., 1980). Earlier research carried out at USF on ash monofill leachate revealed a signifi cant amount of inorganic chemicals such as sodium, calcium, magnesium, and potassium with initial concen trations well above the accepted drinking water standard (Muhammad and Ashmawy, 2003). Attempts were made to use an alternate lin er system with sand-ash-bent onite mixture to arrest the chemical outflux while permeation without much success because of the porous structured formed within the mixture. Bentonite is a very highly plastic swel ling clay of the smectite mineral group, and is mineralogically known as “montmorillonite”. Because of the low permeability of bentonite clay, and the low hydraulic gradie nts to which it is typically subjected,

PAGE 21

2 molecular diffusion and advection are both equally important transport mechanisms. Molecular diffusion coefficients are therefor e important parameters in predicting rates and fluxes of various species of contaminants flowing into the natural soils. Inorganic contaminants, while transported through the be ntonite layer, are chemically adsorbed onto the particle surfaces a nd experience a delay in solute breakthrough in hydraulic barriers. Transport of inor ganic leachate contaminants through bentonite could occur either by advection or diffusion or a combina tion of these two types. During the process of chemical solute transpor t through a low permeability bent onite layer, cation exchange takes place on the clay particle surfaces due to the high cati on exchange capacity (CEC) of montmorillonite minerals. The process of diffusion and advectio n using various inorganic leachate contaminants through bentonite is thoroughly in vestigated in this dissertation. Diffusion study is particularly interest ing in bentonite barriers as it is found to be prominent compared to advection, when the hydraulic cond uctivity of the material is less than 2.0 x 10-8 cm/s (Shackelford, 1988). In addition, the diffusion characteristics of bentonite have not been thoroughly studied and have gained little attention in the geoenvironmental literature until recently. Chemical reactions in the form of cation exchange on the clay particle surfaces must be incorporated during the diffusion process study. Adsorptiondesorption (sorption) reactions of chemical compounds that influence the concentrations of inorganic leachates during transport in be ntonite clay may be modeled using Fick’s diffusion theory. “Partition coefficients” of solutes in pore space, which affect the retardation factors of various individual i ons of chemical solutions, are investigated during transient diffusion and advection processes. The time dependent degradation of hydrauli c conductivity of th e bentonite portion of conventional geosynth etic clay liners (GCL’s) is an ur gent concern particularly for ash monofills. The increase in hydraulic conductivit y of bentonite is caused by aggressive leachates containing high amounts of divalent or higher valence cat ions, especially in landfills subjected to high percolation. The leve ls of some soluble metals and chlorides in landfill leachates exceed USEPA drinking water standards, indicating the importance of liners with high retention capacity of chem ical elements that can sustain their

PAGE 22

3 characteristics for a long duration. Since bent onite is used to cont ain and to reduce the flow of liquids in inorganic contaminant e nvironments, further investigation has become necessary to validate its usage in retaining ce rtain ions from the leachate solutions before reaching chemical equilibrium between influent and effluent. In addition, the increase in hydraulic conductivity of bentonite, caused by leachates containing high amounts of divalent or higher valence cations, is investigated in this research study. It has been reported that Ca2+ and Mg2+ ions, often present in municipal solid waste (MSW) and incinerator ash, can be detr imental to the bentonite if permeated over extended periods of time (Petrov and Rowe, 1997 ). Due to high cati on (+ion) exchange capacity (CEC) and isomorphic replaceable characteristics of montmorillonite microstructure layers, the in crease in hydraulic c onductivity of bentonite can even be observed within a very short period (48 hours) with highly concentrat ed ionic solutions. The low hydraulic conductivity characteristics of bentonite are cause d by the hydration of interlayer spacings through a process calle d “inner-crystalline swelling”. Further adsorption of monovalent cations on the nega tively charged interlayer and external surfaces (osmotic swelling) causes the forma tion of the electrical “double layer” in between the mutually repellant surfaces a nd thus causes separation. As the osmotic swelling is only caused by the hydration of monovalent (namely, Na+) ions, presence of highly concentrated polyvalent cations will inevitably negate the formation of a dispersed clay microstructure and will cause the staggered formation of aggregated clay due to the reduction in the thickness of diffuse double layer (Van Olphen, 1977) as shown in figure 1.1 (Ashmawy, et al. 2002). (a)(c) (b) Figure 1.1 Schematic Representation of Clay Particles Under (a)Initial Saturation with Multivalent Cations; (b) Initial Saturation with Water or Monovalent Caions; and (c) Pre-Hydration Followed by Multivalent Cations (a)(c) (b) Figure 1.1 Schematic Representation of Clay Particles Under (a)Initial Saturation with Multivalent Cations; (b) Initial Saturation with Water or Monovalent Caions; and (c) Pre-Hydration Followed by Multivalent Cations

PAGE 23

4 Aggregated formation of bentonite clay layers from a dispersed structure will increase the free pore space, thus resulting in higher hydraulic conductivity and higher free flow of highly concentr ated soluble metallic ions into the ground. When the permeant contains monovalent cations, i.e., no ionic exchange o ccurs, the amount of interlayer bound water and interlayer spacing wi ll vary according to the variation in the concentration of the permeated liquid (Jo et al ., 2001; Van Olphen 1977). Since the volume of bound water is affected by the si ze of the hydrated cation, solution pH, and anion concentration, these factors also aff ect the hydraulic conductivity of the bentonite (Mitchell, 1993; McBride, 1994; Egloffstein, 1995). The rate of cation exchange in a sodi um bentonite is dependent on, among many other factors, hydraulic gradient, solution c oncentration, temperature, and time (Mitchell 1993; Egloffstein 1995). As the bentonite lin ing system would be laid underneath the leachate collection system in a landfill, the effects of hydrau lic gradient and temperature would be minimal on the degradation of the hydr aulic conductivity of bentonite layer. Another potential degradation mechanis m involves changes in the mineral microstructure. This is most likely to occur at low pH values due to dissolution of clay particles. Alumina in the octahedral layers of the montmorillonite can be dissolved by hydrolysis, thus causing ionic exchange of Al3+ for Na+ in the interlayer spacing and a reduction in the amount of bound water (Norrish and Quir k, 1954; Mathers et al ., 1956; Egloffstein, 1995). In this study, inorganic contaminant leach ates, such as those typically found in ash monofill landfills, were synthesized in the laboratory by combin ing various chemical compounds in deionized (DI) water. Diffu sion and hydraulic conductivity tests were conducted on bentonite materials unde r various boundary conditions, and the concentration of various ions, namely, s odium, calcium, potassium and magnesium, of influent and effluent solutions were determined at various stages of flow. The chemical composition of the bentonite was determined by Energy dispersive spectroscopy (EDS), while mineral compositions were carried out by the XRD method. Commercially available WyoBen bentonite was in this study in conjunction with various inorganic ions comm only found leachate in contamin ants such as NaCl, MgCl2,

PAGE 24

5 KCl, and CaCl2. As the ionic retention capacity of bentonite clay materials can be beneficially exploited in vari ous flow barriers, the ion ab sorption capacity of bentonite was determined under various saturation and loading conditions. 1.2 Research Objectives The specific objectives of this research are itemized as follows: (a) Evaluation of the hydraulic conductivity of the bentonite clay with respect to single salts and various combinati ons of electrolyte solutions under a range of hydraulic gradients. (b) Evaluation of the change of hydraulic conductivity of the bentonite clay for various pre-hydrated conditions, sequencing of inorganic electrolyte solutions, testing method (i.e. flexible wall and rigid wall permeameter), and porosity values of bentonite clay. (c) Determination of “lag time”, breakthrough time, and rate of diffusion of various inorganic dissolved salt so lutions through be ntonite clay under various chemical gradients. (d) Characterization of diffu sion properties of inorga nic leachates through the bentonite layer in terms of apparent and effective diffusion coefficients, and adsorption capacity of the partic les under various loading conditions. (e) Analysis of the time-dependent behavior of the diffusive ions in order to determine the total retention capacity of the bentonite layer before electrical conductivity br eakthrough and steady-state chemical stability are reached. In order to achieve the above objectives, it was also very important to characterize the bentonite clay material in terms of its chemical co mpositions and physical and hydraulic properties.

PAGE 25

6 1.3 Dissertation Outline Chapter Two of this dissertation pres ents the general usage of bentonite, information related clay mineralogy with detailed bentonite clay mineralogy, permeant characteristics, and genera l background material on wate r-bentonite interaction. Literature review on diffuse-doubl e layer (DDL) of clay particle s is also presented in this chapter, which includes mathematical models of DDL and the factors that affect the size of DDL. Characterization of the bentonite used in this research is presented in Chapter Three. Mineral and chemical compositions of bentonite as determined by X-Ray Diffraction (XRD) and Energy Dispersive Spec troscopy (EDS) methods, respectively, are presented in this chapter. P hysical and geotechnical propertie s of bentonite clay, such as grain size distribution, Atterberg limits, sp ecific gravity, swell index, and cation exchange capacity with or without s ynthetic dissolved salts are included. Chapter Four presents the experimental apparatus, along with the design concept and materials and fabrication of permeability and diffusion equipment. In order to prevent any chemical reaction due to aggres sive chemical leachates during permeability, modification to conventional flexible wall permeameters were introduced. Hydraulic characterization of bentonite clay is discussed in chapter Five of this dissertation. Comparison of hydraulic conductiv ity test results carried out on flexible wall and rigid wall permeameters is discussed in this chapter. Vari ous factors affecting hydraulic conductivity are also disc ussed. Results of the chemical analysis of effluent at various stages of permeation are presented. Chapter Six presents experimental me thods of diffusion tests and chemical analysis of the diffusant. IN addition, pH measurements, electri cal conductivity (EC), and ionic analysis test results ar e included in this chapter. The fundamentals of transport theory a nd an analysis of diffusion of chemical solutions through bentonite clay are discu ssed in chapter Seven. Determination and analysis of various diffusion parameters are al so discussed. In this chapter, the main

PAGE 26

7 contributions in terms of characterizing the partition coefficient, retention factor, and retention capacity of bentonite are presented. Chapter Eight summarizes the resear ch findings and provi des recommendations for future work.

PAGE 27

8 CHAPTER TWO MATERIALS AND METHODOLOGY 2.1 Bentonite in Landfills Bentonite, named after an American geologist who discovered this type of clay in about 1890 in Fort Benton, Wyoming, is a cl ay mineral with expa nsive characteristics and low permeability, where montmorillonite is the main mineral. Montmorillonite, named after a deposit located in southern France, swells when contacted with water approximately 900% by volume or 700% by wei ght. When hydrated under confinement, the bentonite swells to form a low permeability clay layer with the equivalent hydraulic protection of several feet of compacted clay when used in traditional landfill applications (Bruno, 2002). Because of its low permeability characteris tics, bentonite clay, with or without treated materials, is being used in combination with geosynthetics to form a composite commonly known as a geosynthetic clay liner (G CL), which has been in use in the USA in the landfill construction since 1988 (Koe rner, 1999). GCLs are rolls of factoryfabricated thin layers of be ntonite clay sandwiched between two geotextile layers or glued to a geomembrane which are used in the lining system as well as cover construction. GCLs are used as a hydraulic barrier and/or contaminant layer for leachate, either in place of a composite layer or in addition to other layers in bottom landfill lining system. Due to surrounding environmental conditions and applied superimposed loads, conventional compacted clay liners (CCLs) develop internal cracks and shrinkage that lead to significant increase in seepage and leakage of contaminant liquid into the ground

PAGE 28

9 soil and water. Bentonite used in GCLs is commonly a sodium bentonite, where sodium ions are located in the interstitial water, betw een clay platelets, in an adsorptive layer that results in the bentonite swelling characteristics. This swelling allows the bentonite to seal around penetrations, giving the GCL self seal ing characteristics. During hydration, a confined layer of dry bentonite changes into a dense monolithic mass with no observable individual particles. A fully hydrated s odium bentonite layer can have a hydraulic conductivity of approximately one hundred time s lower than a typical compacted clay liner (CCL). A single GCL of less than 25 mm provides superior hyd raulic performance than of a meter of typical compacted clay. Bentonite, within geosynthetic clay liners, has been used extensively over the past two decades, and is being investigated furt her to improve quality and performance in many other applications, including lining systems. It is also being used as part of landfill cover systems in landfill construction (Daniel, 1995). Besides GCLs, bentonite clay is also being used in mixed-in-plant (in-situ ) systems, where a mixture of one or two different types of soils as a base material is enriched with be ntonite to obtain low permeability clay base liners (Koch, 2002). As the mixing of in-situ materials with bentonite is becoming popular, the mixed-in-pla nt option represents a very flexible, fast and economical way of landfill construction, es pecially in Europ ean countries (Koch, 2002). Bentonite with cement is also us ed in various construction processes and temporary and permanent sealing barriers, su ch as slurry walls during construction of diaphragm walls or cut-off walls. The tec hnical properties of these materials are well documented, and their integrity as a sealing barrier has b een demonstrated in field applications. Since the bentonite clay is now processed and produced in bulk in factory, its properties and qualities ar e well documented, which gives the design engineers more confidence in predicting its behavior, charac teristics and cost analysis in landfill and other geotechnical applicatio ns (Lin and Benson, 2000). Most of the GCL products manufactured in North America use sodium bentonite clay of mass per unit area of 3.2 to 6.0 kg/m2 with an average clay thickness of 4.0 to 6.0 mm and of hydraulic conductivity ty pically in the range of 1 x 10-9 to 5 x 10-9 cm/s

PAGE 29

10 (Koerner, 1997). Cross-sections of some of the presently available GCLs are shown in figure 2.1. Figure 2.1 Cross-Section Sketches of Various GCLs Bentonite + Adhesive ~ 5 mm Woven Geotextile Woven/non-woven Geotextile a) Adhesive glued bentonite with geotextiles ~ 5 mm Geotextile Geotextile b) Stitch bonded bentonite with geotextiles Bentonite Stitch 4 ~ 6 mm Non-woven Geotextile Non-woven Geotextile c) Needle punched bentonite with geotextiles Bentonite Needle punched fibers 4 ~ 5 mm Lower / Upper PVC/HDPE sheet d) Adhesive bond bentonite to a geomembrane Bentonite + Adhesive Bentonite + Adhesive ~ 5 mm Woven Geotextile Woven/non-woven Geotextile a) Adhesive glued bentonite with geotextiles ~ 5 mm Geotextile Geotextile b) Stitch bonded bentonite with geotextiles Bentonite Stitch 4 ~ 6 mm Non-woven Geotextile Non-woven Geotextile c) Needle punched bentonite with geotextiles Bentonite Needle punched fibers 4 ~ 5 mm Lower / Upper PVC/HDPE sheet d) Adhesive bond bentonite to a geomembrane Bentonite + Adhesive

PAGE 30

11 2.2 Bentonite Clay Fundamentals of bentonite in terms of its mineralogy, cation exchange capacity, and interaction with water are di scussed in this sub-section. 2.2.1 Basic Clay Mineralogy Clay minerals are generally classified according to their crystal structure and geometry. Basic elements of clay minerals are two-dimensional arrays of silicon-oxygen (Si-O) tetrahedron called “tetrahedral sheet “ and aluminumor magnesium-oxygenhydroxyl (Al-, Mg-O-OH) octahedron called “oct ahedral sheet”. The tetrahedron unit in a tetrahedral sheet is composed of four equidistant oxygen atoms arranged in the form of a tetrahedron with a silicon at om at the center as shown in figure 2.2(a) and (b) (after Grim, 1968; Holtz and Kovacs, 1981). All the bases of tetrahedrons are connected to form a single plane in a single sheet, and th e tips of oxygen are pointed in the same direction. A top view of the silica sheet, shown in figure 2.2(c) reveals the linkage of the silicon atoms with the oxygen that forms a hexagonal network with “holes” in the middle (after Warshaw and Roy, 1961). The octahedral sheet in clay minerals is a group of octahedron units, which are composed of six oxygen atoms or hydroxyl groups positioned at equal distance from each other, with an aluminum, magnesium, iron, or other atom at the center as shown in figure 2.3. An octahedron unit is shown in figure 2.3( a), and the linkage of octahedron units to form an octahedral sheet is shown in figure 2.3(b) (after Gr im, 1968). Octahedral sheets are represented as a recta ngular diagram, while the schematic diagram of a silica tetrahedral sheet or silica is represented by a trapezoid in the clay mineralogy as shown in figure 2.4.

PAGE 31

12 and Oxygens Silicons and (a) (b) Figure 2.2 Diagrammatic Sketch Showing Clay Tetrahedral (a) a Single Silica Tetrahedron, (b) Isometric View of Silica Sheet, and (c) Top View of Silica Sheet (after Holtz and Kovacs, 1981) Oxygenslinked to form network Outline of bases of silica tetrahedra Outline of hexagonal silica network (2-D), indicates bonds from silicons to oxygens. (c) and Oxygens Silicons and (a) (b) Figure 2.2 Diagrammatic Sketch Showing Clay Tetrahedral (a) a Single Silica Tetrahedron, (b) Isometric View of Silica Sheet, and (c) Top View of Silica Sheet (after Holtz and Kovacs, 1981) Oxygenslinked to form network Outline of bases of silica tetrahedra Outline of hexagonal silica network (2-D), indicates bonds from silicons to oxygens. (c) Figure 2.3 Diagrammatic Sketch Showing Octahedral (a) a Single Octahedral Unit and (b) the Sheet Structure of the Octahedral Units (after Grim, 1968). and Hydroxyls Aluminums, magnesiums, irons, etc. (a) (b) Figure 2.3 Diagrammatic Sketch Showing Octahedral (a) a Single Octahedral Unit and (b) the Sheet Structure of the Octahedral Units (after Grim, 1968). and Hydroxyls Aluminums, magnesiums, irons, etc. (a) (b)

PAGE 32

13 Figure 2.4 Sheet Representation It can be highlighted that two of every three central spaces of an octahedron are filled with aluminum atoms, keeping the third one vacant. The octahedral sheet where the anions are hydroxyls and two thirds of its available spaces are filled with cations (aluminum) is known as gibbsite as represente d by ‘G’ in the alumina lattice shown in figure 2.4(b). The cations in the octahedral sheet can be substituted with other cations through a geological process ca lled isomorphous substitution. When all the available spaces of cations are filled with magnesium atoms, the mineral is then called brucite shown in figure 2.4(c). Depending on the comb inations of various sheets and cations, which in turn form different crystal basic stru ctures, clay minerals have been divided into various groups. When Al3+ cations are located in two of ev ery three available sites in an octahedral sheet, such minerals are known as dioctahedral. In cont rast, when divalent cations such as Fe2+, Mg2+, Zn2+, etc., are found to be filled in all the available sites, then such clay minerals are called trioctehedral. The tetrahedral (T) and octahedral (O) sheet s are joined in such a way so as to form two-layer clays (T-O), three-layer clay s (T-O-T), or mixed-layer clays that are mixtures of two and three laye rs clays. The linkage between tetrahedral and octahedral sheets causes the sharing of oxygen atoms and hy droxyls at their interface. Clay minerals show various types of chemical compositions due to the fact that Al3+ in octahedral sheets can be replaced by other tr ivalent cations, such as Fe3+, Cr3+, or divalent cations, such as (a) Silica lattice (Tetrahedral) or(b) Alumina lattice (Di-Octahedral) Gibbsite sheet S S Gor B (c) Magnesium lattice (Tri-Octahedral) Brusitesheet (a) Silica lattice (Tetrahedral) or(b) Alumina lattice (Di-Octahedral) Gibbsite sheet S S Gor B (c) Magnesium lattice (Tri-Octahedral) Brusitesheet

PAGE 33

14 Fe2+, Mg2+, Zn2+, or other cations (Faure, 1998). Furthermore, silicon ions (Si4+) in tetrahedral sheets can also be replaced by Al3+ ions due to isomorphous substitution, which takes place during the geological forma tion of various clay minerals. All these substitutions of ions produce excess imbalan ced negative charges on the clay particles that, in turn, adsorb positively charged cations to the outer surfaces of tetrahedral sheets of adjacent clay units in order to satisfy electrical neutrality. 2.2.1.1 Classification and Chemical Composition Clay minerals are classified into grou ps according to the number of layers and their crystal structure. Each group is divide d into subgroups accordi ng to their chemical composition in octahedral sheets, and furthe r divided into individual species of clay minerals. Clay minerals are mainly divided into two-layer, threelayer, and mixed-layer clays as follows: (a) Two-Layer Clays (1 : 1 layer = One Tetrahedral : One Octahedral) Two-layer clay minerals consist of re peated combinations of one layer of tetrahedral sheet and one layer of octahedral sheet as shown by a representative sheet in figure 2.5. The repeated sheets are bonded by sharing O2ions between octahedral cations (Al3+) and tetrahedral cations (Si4+) as shown in the struct ure of a kaolinite layer in figure 2.6 (Grim, 1968). The mineral gr oup of these clays is know n as kaolinite with each layer thickness of 0.72 nm as shown in a schematic diagram in figure 2.6. Depending on the isomorphic substitution of cati ons of octahedral sh eets, kaolinite group minerals are further divided into two subgr oups, namely, kaolinite (dioctahedral) and serpentine (trioctahedral).

PAGE 34

15 The common minerals of the kaolinit e subgroup are kaolinite and halloysite which are represented by the same chemical formula Al2Si2O5(OH)4 –nH2O, where n is the number of water molecules that occupy the in terlayer spaces of the clay aggregates. The value of n is zero for kaolinite clay and 4 for halloysite clay. The ideal structure of the kaolinite subgroup minerals produces no io nic charge imbalance, as shown in figure 2.7, and therefore no cations are affected in th eir interlayer spaces. The individual layers are bonded by strong hydroge n bonds between the OHgroups of the octahedral sheet and O2ions of the adjacent tetrahedral sheet. As hydration is no t possible within the Figure 2.5 Repeated Sheet Representation for 1:1 (Tetrahedral : Octahedral) Layer Octahedral Tetrahedral Basal Spacing Figure 2.5 Repeated Sheet Representation for 1:1 (Tetrahedral : Octahedral) Layer Octahedral Tetrahedral Basal Spacing Oxygens Silicons Hydroxyls Aluminums Figure 2.6 Diagrammatic Sketch of the Structure of the KaoliniteLayer (After Grim, 1968) 0.72 nm Oxygens Silicons Hydroxyls Aluminums Figure 2.6 Diagrammatic Sketch of the Structure of the KaoliniteLayer (After Grim, 1968) 0.72 nm

PAGE 35

16 interlayer spaces, kaolinite clays do not commonly swell when submerged in water, whereas the halloysite mineral contains a layer of water in its interlayer space which causes an increase in layer thickness of 10.1 (McBride, 19 94). The interlayer water molecules of halloysite mineral can easily be irreversibly removed by slightly increasing the temperature, after which it behaves like kaolinite clay. In the serpentine subgroup of kaolinite, th e gibbsite dioctahedral sheet is replaced by a brucite trioctahedral sheet, where three magnesium ions replace two aluminum ions and produce ionic balance on its surface. The chemical formula of serpentine is Mg3Si2O5(OH)4 or Fe3 2+Si2O5(OH)4 which is known as greenalite, where three Fe2+ ions replace two Al3+ ions in the octahedral sheet. (b) Three-Layer Clays (2: 1 layer = two tetrahedral : one octahedral) These clay minerals consist of an octa hedral sheet sandwiched in between two sheets of tetrahedrals with the oxygen tips of the tetrahedrons combining with the hydroxyls of the octahedron to form a single la yer as shown in the figure 2.8 (Holtz and Kovacs, 1983; Faure, 1998). Depending on their chemical composition, crystal Figure 2.7 Charge Distribution on Kaolinite (after Mitchell, 1993) Net charge 28 –28 = 0 7.2 Silica tetrahedron 6 (OH) (OH) O Si Al O -6 +12 4 2 4 4 6 -10 +16 -12 Aluminum octahedron Figure 2.7 Charge Distribution on Kaolinite (after Mitchell, 1993) Net charge 28 –28 = 0 7.2 Silica tetrahedron 6 (OH) (OH) O Si Al O -6 +12 4 2 4 4 6 -10 +16 -12 Aluminum octahedron

PAGE 36

17 structures and physical propert ies, these minerals have b een divided into six groups, namely, pyrophyllite, smectite, vermiculite, mica, brittle mica and chlorite (Faure, 1998). [Smectite] is the largest group in the thre e-layer clays, where the minerals are produced due to full or partial replacement of Al3+ in the octahedral sheet and partial replacement of Si4+ in the tetrahedral sheet (Grim, 1968, Faure, 1998). The smectite group is divided into two subgroups, na mely, dioctahedral when isomorphous substitution occurs in alumina (gibbsite) octa hedral sheets and silica tetrahedral sheets, and trioctahedral when substitution occurs in magnesium (brucite) octahedral sheets and silica tetrahedral sheets. Substitution of Si4+ in the tetrahedral layer is commonly limited to only 15% by mainly Al3+ ions, while Al3+ in the octahedral sheets are generally replaced by various types of cations such as Mg2+, Fe2+, Zn2+, Ni2+, Li+, etc. (Grim, 1968). Montmorillonite is the most commonly f ound mineral in the dioctahedral smectite subgroup, where substitution of one Mg2+ occurs in every sixth Al3+ in octahedral sheets, as shown in figure 2.10, and no substitution takes place in tetrahedral sh eets. This results in a net charge deficiency of about 0.66 – per unit cell as calculated in figure 2.10. This net charge deficiency is balanced by exch angeable cations adsorbed between the unit layers and around their edges as shown in th e crystalline structure in figure 2.9. The Octahedral Tetrahedral Tetrahedral Basal Spacing Figure 2.8 Repeated Sheet Representation for 2:1 (Tetrahedral : Octahedral : Tetrahedral) Layer Octahedral Tetrahedral Tetrahedral Basal Spacing Figure 2.8 Repeated Sheet Representation for 2:1 (Tetrahedral : Octahedral : Tetrahedral) Layer

PAGE 37

18 stoichiometric formula for a unit cell of Na-m ontmorillonite where the interlayer cation is sodium is written as [Si8(Al3.34Mg0.66)O20(OH)4].Na0.66. Other commonly found exchangeable cations adsorbed within the interlayer spaces are Ca2+, K+, and Mg2+. The trioctahedral smectites include the mineral species saponite, hectorite, and sauconite (Faure, 1998). In saponite, the oc tahedral sheet is fully occupied by Mg2+ instead of Al3+, and the charge deficiency is due to the isomorphous substitution of Si4+ by Al3+ in its tetrahedral sheet. The chemical formula of unit cell of saponite is given by Grim (1968) as [Mg6(Si7.34Al0.66)O20(OH)4].Na0.66. Figure 2.9 Diagrammatic Sketch of the Montmorillonite Oxygens Hydroxyls Aluminum, Iron, Magnesium and Silicon, Occasionally Aluminum Exchangeable Cations nH2O Figure 2.9 Diagrammatic Sketch of the Montmorillonite Oxygens Hydroxyls Aluminum, Iron, Magnesium and Silicon, Occasionally Aluminum Exchangeable Cations nH2O

PAGE 38

19 Figure 2.10 Charge Distribution in Mont morillonite (After Mitchell, 1993) [Vermiculite], like smectite, has 2:1 layer sheet structures with both the dioctahedral and trioctahedral forms of clay mineral. The chemical formula of a typical vermiculite is given by McBride (1994) as [(Mg, Al, Fe3+)6(Si8-xAlx)O20(OH)4] (Mg.Ca)x where, x = 1 to 1.4. The stru cture is unbalanced mainly due to the substitutions of Al3+ for Si4+ in tetrahedral sheet and ca uses a residual net charge deficiency of 1 to 1.4 per unit cell. The higher charge deficiency in the te trahedral sheet causes exchangeable cations in the interlayer (mainly Mg2+ with small amount of Ca2+) to electrostatically pull the layer together and thus reduce the layer thic kness. As reported by Grim (1968), many researchers have concluded that vermiculite has only two molecules sheets of water present in the interlayer, cr eating the characteristic spaci ng of 14, as shown in figure 2.11(b). In trioctahedral vermiculite, the char ge deficiency in the tetrahedral sheet is partly compensated by an additional positive ch arge in the Al or Fe octahedral sheet. [Illite] is a nonexpandable dioctahedral clay under the mineral group called mica. Its basic unit is a layer com posed of two inward-pointing si lica tetragonal sheets with a central octahedral sheet, as shown in figure 2.11 (a). In the illites, one-sixth of Si4+ ions 6 O 4 O 2 (OH) 4 Al 6 O 4 Si 4 Si 4 O 2 (OH) -12 +16 -10 +12 -10 -16 -12 Net Charge + 44 –44 = 0 6 O 4 O 2 (OH) 3.34 Al 6 O 4 Si 4 Si 4 O 2 (OH) -12 +16 -10 +11.34 -10 -16 -12 0.66 Mg After substitution Net Charge +43.33 –44 = -0.666 O 4 O 2 (OH) 4 Al 6 O 4 Si 4 Si 4 O 2 (OH) -12 +16 -10 +12 -10 -16 -12 Net Charge + 44 –44 = 0 6 O 4 O 2 (OH) 4 Al 6 O 4 Si 4 Si 4 O 2 (OH) -12 +16 -10 +12 -10 -16 -12 Net Charge + 44 –44 = 0 6 O 4 O 2 (OH) 3.34 Al 6 O 4 Si 4 Si 4 O 2 (OH) -12 +16 -10 +11.34 -10 -16 -12 0.66 Mg After substitution Net Charge +43.33 –44 = -0.66

PAGE 39

20 are replaced by Al3+ in octahedral sheets, which ge nerates the net-unbalanced-charge deficiency of 1.3 per unit cell (Grim, 1968). The resultant charge deficiency is compensated by the potassium ions in the inte rlayer spaces, which are fitted into the hexagonal holes formed by the silica sheets. Therefore, illite has a low cation exchange capacity with very little or no water ad sorption, which prevents it from swelling. [Chlorites] are the 2:1 layered clay mi nerals which can be trioctahedral or dioctahedral in nature. In chlorites, the ne gative charge produced due to replacement of Si4+ by Al3+ is neutralized by the positive charge of brucite sheets generated due to the replacement of Mg2+ by Al3+ sandwiched in the interlayer position which bonds two tetrahedral sheets of two adjacent layers. 2.2.2 Cation Exchange Capacity Cations are attracted and held in between the sheets, on the surfaces, and on the edges of particles in order to maintain the electro-neutrality of particle charges. The cations, which are exchangeable and readily available to be replaced by similar or other Figure 2.11 Schematic Diagram of the Structures of (a) Illite and (b) Vermiculite 10 ~ 14 G B fixed Exchangeable B B G G (b) (a) K K K K K K K K K Ca Mg 10 Figure 2.11 Schematic Diagram of the Structures of (a) Illite and (b) Vermiculite 10 ~ 14 G B fixed Exchangeable B B G G (b) (a) K K K K K K K K K K K K K K K K K Ca Mg 10

PAGE 40

21 types of cations under different environmen tal and phase conditions, are quantified in terms of the cation exch ange capacity of clay. Cation Exchange Capacity (CEC) is define d as the quantity of cations reversibly adsorbed by clay particles, expressed as m illiequivalents (meq) per 100 grams of dry clay mineral. As shown in the table 2.1, the cation exchange capacities of montmorillonite and vermiculite minerals are the highest (in the range of 80~150 meq/100g) among all clay minerals because of their high isomor phous substitution within the octahedral and tetrahedral layers, respectively, which results in a large ionic deficiency. Table 2.1 Some Clay Minerals Ch aracteristics (after Mitchell, 1993) Mineral Interlayer bond Basal sapcing Specific surface (m2/gm) Cation exchange capacity (mEq/100 g) Kaolinite Hydrogen strong 7.2 10-20 3-15 Montmorillonite Oxygen-Oxygen Very weak 9.6 700-840 80-150 Illite K ions: strong 10 65-100 10-40 Vermiculite Weak 10.5-14 870 100-150 Chlorite Strong 14 10-40 When water comes in contact with clay particles, adsorption of positively charged ions with hydrated water molecules occurs at the interface between the solid phase and the aqueous phase. According to Sposito (1989, 1981), adsorption of cations on clay particle surfaces and interlayers can take pl ace by any of the three mechanisms as shown in figure 2.12. The siloxane surface, the plane of oxyge n atoms on the surface of a 2:1 layer silicate, is characterized by a series of hexagonal cavi ties among its c onstituent oxygen atoms, which are formed by six corner-sha ring tetrahedra. The diameters of these cavities are found to be around 0.26 nm and are surrounded by six sets of electron orbits originating from the nearby oxygen atoms (Sposito, 1989, 1981).

PAGE 41

22 The activity of a siloxane surface cavity de pends on the charge distribution of the surrounding layer silicate structur e. A siloxane cavity can act as a mild electron donor if the near layer charge deficiency is low or zero, and can produce a complex with neutral dipolar molecules such as water. The co mplexes formed in the cavity on a neutral interlayer silicate structure are very unstable and easily separable from their constituents. On the other hand, if negative charges are pr esent in the octahedral layer, complexes formed in the cavity with interlayer catio ns and water molecules become strong enough to be immobile and can even get much st ronger when formed near the surface of a negatively charged tetrahedral sheet where the la yer charges are much closer to the cavity surface oxygen atoms. Two types of surface complexes are s hown in figure 2.12, namely, the innersphere complex, which is the result of the en trapment of ions or molecules within the surface cavity without the interf erence of water molecules, and the outer-sphere complex, which is produced by ions or molecules with at least one molecule of water attached to the surface functional group. Outer-sphere complexes, which are formed due to electrostatic bonding, are generally weaker th an inner-sphere complexes involving either ionic or covalent bonding mechanisms. Diffuse ion Inner-sphere complex Outer-sphere complex Figure 2.12 The Three Mechanisms of CationAdsorption on a Silicate Surface; e.g. Montmorillonite(after Sposito, 1989) Diffuse ion Inner-sphere complex Outer-sphere complex Figure 2.12 The Three Mechanisms of CationAdsorption on a Silicate Surface; e.g. Montmorillonite(after Sposito, 1989)

PAGE 42

23 Besides forming inner-sphere and outer-sphe re complexes, the interlayer cations can also be adsorbed and neutralized by the negatively charged clay particles to form a diffuse-ion swarm, as shown in figure 2.12. Su ch diffuse ions are dissociated from the surface functional groups and are free to move in the interparticle solution. Readily exchangeable ions in soil are t hose that can be easily replaced by other ions in an electrolyte soluti on passing through the soil. I ons located within the diffuseion swarm and the outer-sphere complex are th e main readily exchangeable ions in the soil. 2.2.3 Cation Replaceability Exchangeable cations are hydrated when mi xed with water or liquid solutions and are readily displaced into solutions by catio ns of other types of higher replaceability (McBride 1994). The capacity of cationic re placeability depends mainly on the valence, the relative abundance of differe nt ion types in the solution and the silicate exchangeable layer, and the hydrated ion size. As ge nerally reported in the geochemistry and fundamental soil mineralogy literature (Mitch ell, J. K, 1993; Schulze, D.G. 1989; Kelly, W.P. 1948; McBride, M.B. 1994), higher valence cations replace lower valence cations and smaller hydrated cations or larger ionic ra dius cations replace larger hydrated cations or smaller ionic radius cations of the same valence that are present in the exchangeable sites. Besides the above criteria for cation repla ceability, the concentration of cations in the solution plays an important role in the replacement process. In general, the replaceability series, also known as th e “lytropic series,” is as follows: Li+ < Na+ < K+ < Rb+ < Cs+ < Mg2+ < Ca2+ < Ba2+ < Cu2+ < Al3+ < Fe3+ An exception to the above replaceability is possible when the cations of lower replacing power exist in very high concentrations in solution relative to high replacing power cations (Mitchell, 1993). Table 2.2 and 2.3 show the radii of ions in dry and hydrated condition respectively.

PAGE 43

24 Table 2.2 Radii of Ions Table 2.3 Hydrated Radius of Cations Ions Ionic radius () Ions Hydrated Ionic radius () Li+ 0.68 – 0.82 Li+ 7.3 – 10.0 Na+ 1.07 – 1.40 Na+ 5.6 – 7.9 K+ 1.46 – 1.68 K+ 3.8 – 5.3 Mg2+ 0.66 – 0.97 Mg2+ 10.8 Ca2+ 0.83 – 0.95 Ca2+ 9.6 Al3+ 0.47 – 0.61 Fe3+ 0.57 – 0.63 After Faure, G. 1998 After Mitchell, 1993 Ion exchange can also be viewed a chemi cal reaction, but exchange of ions occurs only due to broken bonds and long range elec trostatic bonds of low energy (McBride, 1994). As such, ion exchange “reactions” are similar to inorganic chemical reactions and are typically written in the same form as given in equation (2.1), where Na+ ions from a layer of silicate clay surface are exchanged by Ca2+ in a CaCl2 solution. CaCl2 (aq) + 2NaX(s) = 2NaCl (aq) + CaX2(s) (2.1) where (aq) and (s) refer to the aqueous elec trolyte solution and solid (exchanger) phases, respectively, and X represents the relatively insoluble aluminosilicat e portion of the clay mineral. The aluminosilicate can be assumed to act as a single anion with an equivalent charge of one. Thermodynamic theories that are applicable to inorganic chemical reactions are also applicable in the same way to those of cation exchange reactions (Sposito, 1981). The thermodynamic potential of a reaction is commonly described by the Gibbs-Duhem equation as expressed in equati on (2.2). The standard free energy change of the reaction ( Go) defines the direction of the reaction as follows: ts reac o products o oGtan (2.2)

PAGE 44

25 where the superscript o refers to the conventional standa rd state which is at standard temperature (25oC) and standard atmospheric pre ssure (101.3 kPa). The symbol refers to Gibbs free energy of each chemical species. When Go is negative, the forward reaction has excess energy when it occurs in the standard state. An example of soil thermodynamics theory has been cited by Sposito (1981) in terms of cation exchange reaction that occurs between an aqueous electrolytic solution of Ca2+ cations and Na+ saturated Camp Berteau montmorillonite. The cation exchange reaction can be expressed as: 2NaX (s) + Ca2+ (aq) CaX2 (s) + 2Na+ (aq) (2.3) where 97 5 85 29 955 0 2 045 0 3 612 0 358 4 94 11OH O Mg Fe Fe Al Si X represents the aluminosilicate part of the montmorillonite normalized to the fractional charge deficiency [obtained by dividing each stoichiometric coefficient in the chemical formula of Camp Berteau montmorillonite by 0.335 eq/fw, the cati on exchange capacity due to isomorphous substitutions]. The standard free energy for the above cation exchange reaction, as given in equation (2.3), can be calculated from the i ndividual reactants’ and products’ free energy (o) (Sposito, 1981; Faure 1998). o (Na-mont) = -5,346.1 kJ mol-1 o (Ca-mont) = -5,352.3 kJ mol-1 o (Na+ (aq)) = -261.9 kJ mol-1 o (Ca2+ (aq)) = -553.5 kJ mol-1 The standard free energy change for the reaction in equation (2.3) can be calculated for the Naand Ca-montmorillo nite by dividing the above corresponding values by 0.335 and multiplying by the valence of the exchangeable cation to place them on an equivalent basis as follows: o (NaX(s)) = (1/0.335) o (Na-mont) = -15,958.5 kJ mol-1 o (CaX2(s)) = (2/0.335) o (Ca-mont) = -31,954.0 kJ mol-1

PAGE 45

26 Therefore, the net change in free energy, Go, is: Go = {-31,954.0 + 2 (-261.9)} – {2 (-15,958.5) + (-553.5)} = 7.3 kJ mol-1 Since the free energy due to the cation exchange reaction according to equation (2.2) is negative, the forward reaction has an excess energy when it occu rs in the standard state, which favors the direction as writte n. Thus, the reaction and formation of Camontmorillonite is thermodynamically favorable. The hydration energy of cations, defined as the amount of energy released when dry cationic substances are mixed or hydrated in water, has also been used in the Eisenman energy model of cation exchange, wh ere the behavior of ions of different radius has been incorporated. As desc ribed by McBride (1994), the electrostatic attraction energy, Eatt, between an adsorbed cation and th e surface charge site is inversely proportional to the finite distance between the charge centers, as shown in figure 2.13, and is given by equati on (2.4) as follows: A s attr r e E 2 (2.4) where e is the electronic charge unit. This is the energy that is required to displace the water molecules present between the catio ns and the charged clay surface. The presence of water molecules on the cl ay surface is the resu lt of the hydration of the clay surface and the exchangeab le cations. The total energy change, Etot, in excess of the attraction energy due to th e movement of a monovalent ion, A+, from the solution to the surface is give n by McBride (1994) as: A s A s totE E r r e E2(2.5) where rA and EA are the radius and hydration ener gy of cation A, respectively. The parameter rs is the effective radius of the charge surface, as shown in figure 2.13, and Es is the hydration energy of the surface. For the cation exchange of ion B+ by ion A+ on the same clay surface, the overal l change of energy would be:

PAGE 46

27 A B A s B s totE E r r e r r e E2 2 (2.6) where rB and EB are the radius and hydrati on energy of displaced ion B. When the distance between the cation charge center and the location of negative structural charge in the clay ( rs + rA or rs + rB) is large, as is the case for montmorillonite minerals where isomorphous substitution occurs in the octahedral laye r, the electrostatic term of equation (2.6) is negligible. Therefore, in the weak field condition, as depicted in figure 2.13, the total change of energy due to cation exchange would be equivalent to the Figure 2.13 Schematic Diagram of the Clay Surface-Exchange CationInteraction in (a) Dry Condition, (b) Water on a “Weak Field”, (c) Water on a “Strong Field” Exchanger (after McBride, 1994) (a) (b) (c) Figure 2.13 Schematic Diagram of the Clay Surface-Exchange CationInteraction in (a) Dry Condition, (b) Water on a “Weak Field”, (c) Water on a “Strong Field” Exchanger (after McBride, 1994) (a) (b) (c)

PAGE 47

28 difference in ionic hydration energies. From Table 2.4, it is clear that the cations of higher hydration energies can be easily replaced by cations of lower hydration energies to come into contact with the surface and release energy during the process. It can be noted that the Eisenman model is not considered to be a complete solution as it does not cover the changes of entropy (disorder) of various cations during exchanges. Table 2.4 Hydration Energy of Me tal Cations (after McBride, 1994) Ion Hydration energy (kcal/mol) Ion Hydration energy (kcal/mol) Li+ 124 Mg2+ 460 Na+ 97 Ca2+ 381 K+ 77 Ba2+ 312 Rb+ 71 Al3+ 1114 Cs+ 63 Fe3+ 1046 2.3 Permeant Characteristics Bentonite clay is being used in various applications of solu tion containment as well as a water barrier, in which a number of chemicals are dissolved. These chemicals may be generated from many different industr ial, commercial, and household application processes. This section is mainly focused on sources of various chemical solutions that are blended in water which are required to be contained by clay liners and similar barrier materials. 2.3.1 MSW Leachate Bentonite clay, as an active component of Geosynthetic Clay Liners (GCL) is being widely used in Municipal Solid Wa ste (MSW) landfill construction where the

PAGE 48

29 proper functioning of the lining system is critic al in terms of contai nment effectiveness of generated leachates. Leachate is formed when water infiltrates the waste in the landfill cell. The water within the landfill could be generated either from a combination of precipitation from rain and melted snow, or fr om the waste itself. As the liquid moves through the landfill, many organic and inorga nic compounds, such as heavy metals, are transported through the leachate. The amount of leachate produced is direc tly linked to the amount of precipitation around the landfill. The amount of liquid waste in the landfill also a ffects the quantity of leachate produced. Leachates are potentially haza rdous wastes in landfill sites. It is of the utmost importance that leachates are tr eated and contained within the landfill to prevent any contamination and mi xing with fresh ground water. Leachate generated from municipal so lid waste (MSW) and hazardous waste (HW) landfills is a mixture of organic and inorganic compounds, as well as dissolved and colloidal solids. In order to design a collec tion and treatment system for leachate, it is important to have an understanding of the wa stes placed in the landfill, as well as the physical, chemical, and biological processes that are occurring within the landfill. The quality and chemical composition of leachates vary tremendously depending of a number of factors which include mainly: (a) Waste Composition The waste composition of MSW, especi ally household refuse (eg. food, garden wastes, animal residues, etc), contributes and determines the range and extent of biological activity with in the landfill (Chen and Bowerman, 1974). Inorganic constituents in l eachates are mainly derived from construction and demolition debris, i ndustrial wastes, household furniture and electrical appliances, vehicle parts and tires, etc. (b) Depth of Waste Higher depth of waste is found to cont ribute to higher concentrations of leachate at the base of the waste layer before entering into the lining

PAGE 49

30 systems. Deeper waste also requir es a longer time to decompose as the water takes longer to reach larger dept hs. As the water percolates through the deeper waste, it travels a long dist ance and reacts with larger quantities of waste material, which eventually yi elds a highly concentrated chemical solution at the base lining sy stem (Qasim and Chiang, 1994). (c) Moisture Availability The quantity of water or the degree of saturation of waste materials within the landfill is the most important cont rolling factor of leachate quality. High quantities of moisture within loose or less compacted waste landfills increase the rate of flushing, which removes the majority of the contaminants during the early stages of filling, whereas in more compacted or low permeability landfills, high moisture causes an increase in the rate of anaerobic microbial ac tivity which generate s high strength of organic leachates (McBean et al ., 1995; Chen and Bowerman, 1974). Low amounts of moisture take longer to fully react with all the available inorganic and organic agents of wast e materials and therefore develop a slow stabilization rate of the landfill chemistry (McBean et al ., 1995; Miller et al ., 1994) (d) Oxygen Availability The amount of available oxygen controls the type of decomposition (i.e. anaerobic or aerobic) of organic co mponents in landfill wastes. Aerobic decomposition happens when the oxygen is available within the landfill, i.e., during the operation stage, at the top layer of the waste, and within loosely compacted waste fills where air voids are available. Carbon dioxide, water, lightly concentrat ed organic compounds, and heat are generated during aerobic decomposition while highly concentrated organic acids, ammonia, hydrogen, carbon dioxide, methane, and water are produced during anaerobic degradation (McBean et al ., 1995).

PAGE 50

31 (e) Temperature Temperature within the landfill is responsible for bacterial growth, which controls organic and chemical reactions of the waste materials. The solubility of many inorganic salts [e.g. NaCl, KCl, MgCl2, Ca3(PO4)2] increases with temperature. However, the solubility of a number of other chemical compounds that are present in leachates, such as CaCO3 and CaSO4, deceases with increasing temper ature as investigated by Lu et al (1985). (f) Age of Landfill The age of a landfill directly controls the quality of leachate. Leachates with maximum contaminants are found within 2-3 years of the final placement of wastes in the landfill, after which the amount of contaminants decline steadily over the next 10-15 years (McBean et al ., 1995; Lu et al ., 1985). Depletion of inorgani c compounds is much faster than that of organic compounds whic h continue for a long period of time due to bacterial and other microorganism reactions (Lu et al ., 1985). Table 2.5 shows the wide variation in l eachate quality as investigated by various researchers (after Reinhart and Grosh, 1998). A more deta iled breakdown of organic and inorganic compounds of two MS W landfill leachates is given in Table 2.6, which was published in a report by the Ontario Minist ry of Environment, Canada, in 1996. 2.3.2 Ash Leachate Ash from Waste-To-Energy (WTE) faciliti es is being generated in abundance in the United States of America as the volume of solid waste increases with the increasing growth of population. The incinerated residues composed of bottom ash and fly ash, are commonly disposed in landfills under Sub title D ash monofills, provided that the

PAGE 51

32 materials are non-hazardous according to USEPA's recommended Toxicity Characteristics Leaching Procedure (TCLP) test. The main factors, among many others, whic h affect the variation in chemical composition of ash are believed to be the s ource of burning materials (type of solid waste), methods of incineration, and additiv es used in the process of neutralizing hazardous materials (Muhammad and Ashmawy, 2003). Table 2.5 Chemicals in Leachates as Found by Different Research ers (after Reinhart and Grosh, 1998) Parameter Ehrig, 1989 Qasim and Chiang, 1994 South Florida* Landfills, 1987 Pohland and Harper, 1985 BOD (ppm) 20 – 40,000 80 – 28,000 4 – 57,700 COD (ppm) 500 – 60,000 400 – 40,000 530 – 3,000 31 – 71,700 Iron (ppm) 3 – 2,100 0.6 – 325 1.8 – 22 4 – 2,200 Ammonia (ppm) 30 – 3,000 56 – 482 9.4 – 1340 2 – 1,030 Chloride (ppm) 100 – 5,000 70 – 1330 112 – 2360 30 – 5,000 Zinc (ppm) 0.03 – 120 0.1 – 30 0.06 – 220 P (ppm) 0.1 – 30 8 – 35 1.5 – 130 0.2 – 120 pH 4.5 – 9 5.2 – 6.4 6.1 – 7.5 4.7 – 8.8 Lead (ppm) 0.008 – 1.020 0.5 – 1.0 BDL – 0.105 0.001 – 1.44 Cadmium (ppm) <0.05 – 0.140 <0.05 BDL – 0.005 70 – 3,900 BDL – below detection limit South Florida Water Management District, 1987. Chemical analysis of various types of fly ash conducted by many researchers revealed that the major four minerals present in the fly ash are silica (SiO2), alumina (Al2O3), calcium oxide (CaO), and iron oxide (Fe2O3). Other minor minerals, which are normally less than 5% in total weight, are magnesium oxide (MgO), sodium oxide (Na2O), titanium oxide (TiO2), potassium oxide (K2O), phosphorus oxide (P2O3), sulfur trioxide (SO3), and trace metals oxide (Edil et al ., 1992; Joshi et al ., 1994;

PAGE 52

33 Wentz et al .,1988; Porbaha et al. 2000; Hettiaratchi et al ., 1999). The four major minerals found in combined MSW ash are the sa me as those in fly ash but the amount of calcium oxide (CaO) is predominant compared to other minerals because of the presence of free-lime used in the process of incineration (Keith and Goodwin, 1990). Table 2.6 Chemical Composition of Two MSW Landfill Leachates Parameter Muskoka Guelph Benzene (ppb) 18 19 Toluene (ppm) 263 201 Ethylbenzene (ppm) 35 80 m + p-xylene (ppm) 66 148 O-xylene (ppm) 37 85 NH4+ (ppm) 103,000 865,000 K (ppm) 114 1301 Ca (ppm) 203 883 Mg (ppm) 29 525 Fe (ppm) 38 1 B (ppm) 1 8 Cl(ppm) 98 2464 EC (mS/cm) 1.4 9.9 pH 5.4 7.0 The electrical conductivity (EC) of the effl uent solution is found to be reduced to around 1000 microsiemens/cm from their init ial high values of 100,000 microsiemens/cm within less than 5 pore volumes of flow t hough the specimens of compacted ash materials Therefore it is concluded that the majority of the salts (chlorides and sulfides) are “flushed” out of the sample within a maximum of 5 pore volumes (Muhammad and Ashmawy, 2003). The research conducted by Muhammad and Ashmawy (2003) on ash leachates also reveals the pattern of a ttenuation of sodium, calcium a nd potassium in the effluent

PAGE 53

34 permeant with pore volumes of permeation. It was observed that the initial high calcium concentration of 15,000 to 35,000 ppm was re duced to below 3,000 ppm within 5 pore volumes of permeation, with further reducti on to less than 500 ppm after around 12 pore volumes. The same trend was also ob served for sodium and potassium ion concentrations. The concentration of sodium ions was reduced from an initial high concentration of around 10,000 -12,000 ppm to less than 500 ppm within 8 pore volumes. Similarly, potassium ions decreased in concentration from around 6,000-9,000 ppm to less than 500 ppm within 5 por e volumes of permeation. The trend of attenuation of all the main elements replicates the atte nuation of EC values of effluent. 2.3.3 Other Sources of Inorganic Leachates Bentonite waterproofing has proven reliab le for a wide range of applications, including underslab, back-filled walls, plaza deck, and property line construction such as soldier piles and lagging. U nderslabs typically are inst alled directly on a properly compacted substrate, eliminating the requi rement for a mud slab. The swelling properties of bentonite are effective in sealing small conc rete cracks caused by settlement, seismic action or other simila r conditions. For installations where groundwater is contaminated or has a high level of salt concentr ation, contaminantresistant bentonite characte ristics are required. Bentonite waterproofing systems are empl oyed on fresh concrete as soon as the concrete forms are removed in order to pres erve concrete water / cement ratio and to prevent any external ingress into the concrete Limitations of bentonite waterproofing include proper confinement for maximum perf ormance. Bentonite waterproofing should not be installed when properly compacted back-fill or concrete cover is absent, as proper confinement is required. Bentonite can be used to form a cut-o ff wall by injection or pressure grouting and/or slurry trenching. It is also being used in repair ing cracks of earth dams or

PAGE 54

35 embankments used in storing industrial bypr oducts containing or ganic or inorganic contaminants, as shown in figure 2.14. Bentonite is also used in pi pe connections such as at the joints between concrete or synthetic pipes and manholes in sewe r construction, where a large amount of contaminated slug flows constantly, as shown in figure 2.15. Other uses include earthen ponds and lagoons, where bentonite is exposed to the contained water, which affects its performance if highly concentrated dissolved salts are present. The swelling property when hydrated allows bentonite to fill voids or unexpected opening in sandy soils, where it acts as a “self-healing” material. Drilling fluids have been used for year s to stabilize bore holes during drilling operations. In the 1950’s, civil and geotechni cal engineers discovere d that deep, narrow trenches excavated in granular soils could also be stabilized using the same technology to prevent collapse of the sidewalls. The excav ated materials could then be mixed with bentonite slurry and backfilled, providing an economical barrier to lateral flow of water and many fluid pollutants since fluid loss of th e pure bentonite plays can affect long term performance. The amount of fluid loss is also affected by the quality of the water that is expected to be in contact with the bentonite. Figure 2.14 Application of Bentonite in Embankment or Earthen Dam In all of the above applications, bentonite is expected to encounter water-borne contaminants or highly concentr ated organic or inorganic salt solutions where Ca, Mg, K, Pervious embankment Storage Level Pervious layer Impervious layer Grout holes Pervious embankment Storage Level Pervious layer Impervious layer Grout holes

PAGE 55

36 and Na dissolved cations are present. The ex istence of these cations in salt solutions is responsible for the deteriorati ng performance of the bentonite component of the structure. Figure 2.15 Application of Bent onite in Manhole-Pipe Connection 2.4 Water-Bentonite Interaction Adsorbed cations needed to neutralize th e negatively charged pa rticles are tightly held on the clay surface in th e dry phase of the clay. Dried clays adsorb water from the atmosphere at low relative humidities. Cl ays in the smectite group swell when they adsorb water, and need temperatures above 100oC to remove most of the water within the pore spaces. Much higher temper atures in the range of 500 ~ 1000oC are needed to remove all the water within clay interlayer spaces, which is held tightly on the clay particles due to the negative charge on the surface. In the clay chemistry literature, clays are considered to be lyophobic (liquid hating) or hydrophobic (water hating) colloids rather th an lyophilic or hydrophilic colloids, even though water is adsorbed by th e clay particles. Hydrophilic colloids are

PAGE 56

37 those that adsorb water so as to form a co lloidal solution instantaneously (van Olphen, 1977). Clays are considered hydrophobic because: (a) it has a two-phase system with a large interfacial surface area, (b) clay-water behavior is dominate d by clay surface forces, and (c) it can flocculate in the pres ence of small amount of salts. 2.4.1 Mechanisms of Interaction The following mechanisms for clay-w ater interaction are possible: (a) Hydrogen Bonding Because the clay mineral’s exposed surf aces are either composed of oxygens or hydroxyls ions, hydrogen bonding develops with oxygen attracting the positive corner (H+) of water molecules and hydroxyl attracting the negative portion (O-), as shown in figure 2.16(a). This bond will redistribute and reorient the charges in normal water, and the bonded water molecules will progressively al ter the direction of adjacent molecules. The bonding will become less rigid with distan ce from the surface of the clay due to the surface force fields as well as the increase in the force fields of the water structure (Mitchell 1993). (b) Exchangeable Cations Exchangeable cations that are attracte d on the negatively charged surfaces get hydrated when mixed with water and are attr acted to the clay surface in the form of hydrated molecules, as shown in figure 2.16(b). Positively charged cations are surrounded by the negative corner of the water molecules. (c) Attraction by Osmosis The concentration of hydrated cations near to the charged surface is higher due to the electrostatic attraction. Du e to this electrostatic attrac tion, cations are prevented to diffuse away from the surface so the concentra tion of water molecules is lower at near the surface of clay particles. This variation in water concentration causes water molecules to

PAGE 57

38 diffuse toward the vicinity of the charged surface due to osmotic pressure, as shown in figure 2.16(c) (Mitchell, 1993). Water dipoles Inward diffusion of H2O Increasing ion concentrationSurface oxygens Water molecules Clay surface Surface (OH) Clay surface Clay surface Clay surface Water molecules Cations(a) (b) (c) (d) Figure 2.16 Possible Meachanismsof Water Adsorption by Clay Surfaces (a) Hydrogen bonding, (b) Ion hydration, (c) Attraction by Osmosis, and (d) Dipole Attraction. (after Mitchell, 1993) Water dipoles Inward diffusion of H2O Increasing ion concentrationSurface oxygens Water molecules Clay surface Surface (OH) Clay surface Clay surface Clay surface Water molecules Cations(a) (b) (c) (d) Figure 2.16 Possible Meachanismsof Water Adsorption by Clay Surfaces (a) Hydrogen bonding, (b) Ion hydration, (c) Attraction by Osmosis, and (d) Dipole Attraction. (after Mitchell, 1993)

PAGE 58

39 (d) Charged Surface – Dipole Attraction As the water is a dipole molecule, even though it is electricall y neutral, it is electrostatically attracted to the charged clay surface due to Van der Waals attraction force (Mitchell 1993, Holtz 1981). Water mol ecules dipoles are directed to the negative charged surfaces, with the degree of orient ation decreasing gradually with increasing distance away from the surface. 2.4.2. Diffuse Double Layer Tightly held interlayer cations within the clay particles, due to electrostatic attraction of the negatively charged surfaces pull water molecules because of their hydration energy upon wetting. Highly concen trated cations along the charged surfaces try to diffuse away from the surfaces in or der to equalize the con centration throughout the clay-water solution. The escaping tendency of cations from the surface and the opposing electrostatic attraction lead to a specific i on distribution along the clay particles in the clay-water suspension. The negative charge of the clay surface and the distribution of cations in the soil solution are known as “D iffuse Double Layer” or DDL (Mitchell, 1993; Shackelford, 1994). 2.4.2.1 Theory and Mathematical Models of DDL The concept of diffuse double layer has b een developed from the basics of the electrical double layer, which describes the va riation of electric pot ential near a charged surface, and plays an important role in the behavior of colloids and other surfaces which are in contact with electrolyte solutions. The earliest concepts of the double layer were proposed and developed by Helmholtz (1853-187 9) where the double layer refers to the counterions (cations) and co-ions (anions) in a rigid layer adjacent to the clay charged interfaces (Endo et al. 2001). Figure 2.17 illustrates the Helmholtz model which is

PAGE 59

40 analogous to the parallel plate-capacitor in which the negatively charged surface would form one plate and rigidly linked opposite charged cations to the surface would form another plate (Endo et al. 2001). In this model no interactions occur further away from the first layer of adsorbed ions, and the electric potential drops sharply from its maximum value at the charged surface to an almost neg ligible value at the center of the first fixed layer of cations adjacent to the surface, as shown in figure 2.18. Two principal shortcomings were discovered in this m odel during subsequent research by Gouy and Chapman in 1913 (Mitchell, 1993; Endo et al. 2001; Wikopedia, 2004; Van Olphen, 1977) as follows: (a) It neglects interactions of cations an d anions occurring further away from the charged surfaces and (b) The extent and thickness of diffuse double layer takes into account no dependence on electrolyte concentration. Gouy and Chapman (1910-1913) made a significant improvement by introducing a diffuse double layer model, in which the potential decreases expone ntially away from the surface due to adsorbed counter-ions ( cations) from the solution away from the charged surface. Thus, the double layer would not be compact as in Helmholtz’s model, but of variable thickness as th e ions are free to move away in the bulk electrolyte solution as shown in figure 2.19 (after Mitchell 1993). Potential Solution Potential Distance from surface Figure 2.17 HelmholtzModel Potential Solution Potential Distance from surface Figure 2.17 HelmholtzModel

PAGE 60

41 The electrical potentia l of the electrolyte solution decreases exponentially from the face of the charged surface, which gradua lly extends into the bulk solution, and the concentration of ions have been calculated to be very high at the surface due to the assumption of point ionic charges (van Ol phen 1977, Mitchell 1993). The hydrated ionic size is not considered in this model. Stern, in 1924, developed a model in corporating Helmholtz model and GouyChapman model, commonly known as SternGouy-Chapman model, which is widely acceptable at present (figure 2.19). This model consists of a compact layer of cations of finite radius at the close vicinity of the negatively charged surface known as the “Stern layer”, similar to the Helmholtz model, and a diffuse layer of cations and anions extending into the bulk solution similar to the Gouy-Chapman model (van Olphen, 1977; Mitchell 1993). The effect of the stern layer on the surface electrical potential and cationic concentration is shown in figure 2.19 (Mitchell 1993). The thickness of the Stern layer increases with cationic size, a nd its presence would lim it the predicted cation concentration at the surface, as shown in figure 2.19. Mathematical representations of the diffuse double laye r phenomenon were provided using the following assumptions (Mitchell 1993): (a) Ions are point charges with no intera ction among opposite ch arges within the interlayer and bulk pore spaces, (b) The charge on the particle surface due to isomorphous substitution is uniformly distributed, (c) The dimensions of the surface on which the charge deficiency is uniformly distributed are much larger th an the diffuse double layer, and (d) The permittivity of the medium present in between the surfaces is constant regardless of the position. From electrostatics, Poisson’s equation gi ves the charge balance in an electric field, and the general expression for a homogeneous dielectrical medium in a onedimensional situation is:

PAGE 61

42 D dx do ch 2 2 (2.7) where, is the electrical potential in front of the charged clay surface, ch is the charge volumetric density (C m-3), D is the relative permittivity of the medium, and o is the dielectric constant of the void (C V-1 m-1). Potential Distance from surface Figure 2.18 Gouy-Chapman Model Potential Distance from surface Figure 2.18 Gouy-Chapman ModelPotential Distance from surface Figure 2.19 Stern Gouy-Chapman Model Stern layer Diffuse layer Potential Distance from surface Figure 2.19 Stern Gouy-Chapman Model Stern layer Diffuse layer

PAGE 62

43 On the other hand, the Boltzmann equation represents the distribution of ions within an electrical field: kT e e io ii o i exp (2.8) where, k is Boltzmann’s constant, T the absolute temperature, i = ionic concentration of the species i, e = unit electronic charge (16 x 10-20 Coulomb), and o = electrical potential at concentration i o. As the potential at great distance from the interface is equal to zero, the term ieo can be set to zero. The volume charge can be expressed as: i i che (2.9) Using equation (2.9), Boltzmann e quation (2.8) can be written as: kT e io i chie exp (2.10) Substituting into Poisson’s equation lead s to the general expression for the Poisson-Boltzmann equation in a one-dimensional field: D e dx do kT e io ii exp2 2 (2.11) For a solution of single cation and an ion species of equal valence, i.e. i = 2, + = = o + = o = and sinh p = (ep – e-p)/2 equation 2.11 can be rewritten as: kT e D e dx do o sinh 22 2 (2.12) It is convenient to rewrite the above equation (2.12) in terms of the following dimensionless quantities: x kT e kT e yo z (2.13) where, DkT eo o 2 2 22 (2.14)

PAGE 63

44 Putting the above relationships of equati on (2.13) in equation (2.12), we get: y d y d sinh2 2 (2.15) K, dimensionally a length, is called th e Debye-Huckel parameter. Using the boundary conditions for the first integration, = y = 0, and dy/d = 0 the following can be obtained: 2 sinh 2 2 cosh 22 / 1y y d dy (2.16) This condition holds for a large pore, as it assumes that the do uble layers of two platelets, one in front of the other, do not overlap. The boundary condition for the second integration, = 0, y = z (i.e. = o), yields: exp 1 exp 1 exp exp 1 exp 1 exp exp2 / 2 / 2 / 2 / 2 /z z z z y (2.17) Equation (2.17) describes the decay of th e potential as a function of the distance from the surface at a given surface potential (i.e., z ) and at a given electrolyte concentration (i.e., K2). If the surface potential is small ( << 25 mV), then e/kT << 1 (i.e., z << 1) and the relation e-x 1-x is often adopted in order to e xpand the exponential equation (2.11) as follows: D kT e e dx do io i io i /2 2 2 (2.18) Because of the electrical ne utrality of the bulk solution, the first term in the parentheses (iio) has to be equal to zero from the charge equation (2.9), and equation (2.18) then becomes: 2 2 2 dx d (2.19) The solution of the above equation (2.19) can be written as: x o exp (2.20)

PAGE 64

45 In this case, the center of gravity of the counter ions (cations) atmosphere coincides with the plane Kx = 1 or x = 1/K. Hence 1/K is often called the double layer thickness; it is also equal to the “character istics length” in the Debye-Huckel theory of strong electrolytes. 2.4.2.2 Factors Affecting DDL The thickness of the diffuse double layer (DDL), 1/K, can be rearranged from equation 2.14 as follows: 2 / 1 2 22 1 e DkT Ko o (2.21) The variable factors in e quation (2.21), which affect the DDL thickness, can be summarized as follows: (a) Electrolyte Concentration ( o) By keeping all other factors constant, an increase in electrolyte concentration will decrease DDL exponentially. A one hundred fold increase in concentration will cause a 10 fold d ecrease in DDL distance as calculated from equation (2.21), an example of wh ich is shown in figure 2.20 (Mitchell, 1993). (b) Electrolyte Cation Valance () It is found from equation (2.21) that the thickness of DDL is inversely proportional to the valence of the electrolyte solution. An increase in valence will suppress the midplane concentrations and potential between interacting plates, which leads to a decr ease in interplate repulsion as given in figure 2.20 (after Mitchell, 1993)

PAGE 65

46 (c) Effects of Dielectric Constant (D) The DDL thickness is directly proportional to the square root of the dielectric constant of the concentrated electrolyte solution. The value of D also affects the electrical potential (o) as per the hyperbolic expression given in the following equation: 2 / 18 2 sinh DkT kT eo o o (2.22) It is found from equation (2.22) that for a constant value of surface charge density, the electrical potential increases as the dielectric constant decreases. Figure 2.20 Effect of Con centration on Ion Distributi ons with Distance (after Mitchell, 1993) Distance from surface ()Concentrations (ions/cm3) Anions Distance from surface ()Concentrations (ions/cm3) Anions

PAGE 66

47 Figure 2.21 Effect of Cati on Valence on Double Layer (after Mitchell, 1993) Distance from surface ()Concentrations (ions/cm3) Distance from surface ()Concentrations (ions/cm3)

PAGE 67

48 CHAPTER THREE BENTONITE CHARACTERIZATION In this chapter the index and physicoche mical properties of bentonite that has been used in this study are highlighted. Geotechnical tests such as Atterberg limits, particle/grain size distribution, specific gravity, and swell index have been conducted using ASTM standards. Modifications have been made to the conventional standards to suit the type of bentonite clay used in this study after thor ough investigations of various studies published in the literature. 3.1 Source of Bentonite Extra High Yield Bentonite powder manufactured by W yo-Ben, Inc., has been used in this study. Widely known as “W yoming Bentonite” (sodium montmorillonite), this bentonite is being co mmercially used in the cons truction industry for mining exploration, water wells, and directional drilling opera tions. When one 50-lb bag bentonite powder is mixed with 300 gallons of water, it provides a funnel viscosity of 3035 seconds. 3.1.1 Mineralogy Through XRD X-Ray diffraction (XRD) has been used for many years to determine the mineralogy based on basal spacing of the clay minerals (Suzuki et al., 2001; Hwang and

PAGE 68

49 Dixon, 2000; Chmielova et al., 2000; Kozaki et al., 2001; Song and Sandi, 2001; Mayayo et al., 2000; Cases et al., 1997). X-rays are electromagnetic radiations of wavelength of about 1 , which is approximately the same size as an atom They occur in that portion of the electromagnetic spectrum between gamma-rays and ultraviolet. X-ray diffraction has been in use in two main areas: characte rization of crystalline materials and the determination of their structure. Each crys talline solid has its unique characteristic X-ray powder pattern, which may be used as a "fi ngerprint" for its identification. Once the material has been identified, X-ray crysta llography may be used to determine its structure, i.e., atomic packing in the crys talline state and interatomic distances and angles. X-ray diffraction is a routine method in mi neralogy, particularly for fine-grained material study. It is one of the primary techniques used by mineralogists and solid state chemists to examine the physicochemical composition of unknown solids. XRD can provide additional information beyond basic identif ication. If the sample is a mixture, XRD data can be analyzed to determine the pr oportion of the different minerals present. Other information obtained can include the de gree of crystallinity of the mineral(s) present, possible deviations of the minerals from their ideal compositions (presence of element substitutions and solid solutions), structural state of the minerals, and degree of hydration for minerals that contain water in th eir structure. Some mineralogical samples analyzed by XRD are too fine-grained to be identified by optical li ght microscopy. XRD does not, however, provide the quantitative compositional data obtained by electron microprobes or textural and qualitative com positional data obtained by scanning electron microscope. The XRD technique requires placing a pow dered sample of the material in a holder, then illuminating it with X-rays of a fixed wave-length. The intensity of the reflected radiation is then r ecorded using a goniometer. This data is analyzed for the diffraction angle to calculate the inter-at omic spacing (d valu e in Angstroms 10-8 cm).

PAGE 69

50 The three-dimensional structure of non-amor phous materials, such as minerals, is defined by regular, repeating plan es of atoms that form a crys tal lattice. When a focused X-ray beam interacts with these planes of atoms, part of the beam is transmitted, part is absorbed by the sample, part is refracted and scattered, and part is diffracted. Diffraction of an X-ray beam by a crystalline solid is an alogous to diffraction of light by droplets of water, producing the familiar rainbow. X-ra ys are diffracted by each mineral differently, depending on atom make up and arrange ment in the crystal lattice. In X-ray powder diffractometry, X-rays ar e generated within a sealed tube under vacuum. A current is applied that heats a fila ment within the tube; the higher the current the greater the number of electrons emitted from the filament. This generation of electrons is analogous to the production of electrons in a te levision picture tube. A high voltage, typically 15-60 kilovolts, is app lied within the tube This high voltage accelerates the electrons, which then hit a ta rget, commonly made of copper. When these electrons hit the target, X -rays are produced. The wave length of these X-rays is characteristic of that target. These X-rays are collimated and directed onto the sample, which is a fine powder of particle size of less than 10 microns. A detector detects the Xray signal; the signal is then processed e ither by a microprocessor or electronically, converting the signal to a count rate. Cha nging the angle between the X-ray source, the sample, and the detector at a controlled rate between preset limits, an X-ray scan is obtained. Figure 3.1 shows how X-ray waves re veal the atomic structure of crystals. Figure 3.1 Basics of X -ray Diffraction Technique d = 4.8 Atomic planes Scattered beam = 30.0o = 2.6 Incident beam o d = 4.8 Atomic planes Scattered beam = 30.0o = 2.6 Incident beam o

PAGE 70

51 When an X-ray beam hits a sample and is diffracted, it measures the distances between the planes of the atoms that constitute the sample by applying Bragg's Law as given in the equation (3.1). sin 2 d n (3.1) where the integer n is the order of the diffracted beam, is the wavelength of the incident X-ray beam, d is the distance between adjacent planes of atoms (the d-spacing), and is the angle of incidence of the X-ray beam. By knowing and measuring the d-spacing can be calculated. The characterist ic set of d-spacings generated in a typical X-ray scan provides a unique “fingerprint” of the mineral or minerals present in the sample. When properly interpreted, by comp arison with standard reference patterns and measurements, this “fingerprint” allows for identification of the material. A typical spectrometer with XRD fundamentals is s hown in figure 3.2 where the value of or 2 determines the composition of minerals in the specimen. Figure 3.2 XRD Spectrometer Fundamentals Extra high yield bentonite from Wy o-Ben, Inc. was dried at 105oC to remove all the mobile pore fluid from the sample for XRD analysis. The sample’s dry particles were passed through sieve # 200 (particles less than 75 microns) and ke pt dry until placed Matched filters Crystal Cillimators Detector X-ray tube 2 Matched filters Crystal Cillimators Detector X-ray tube 2

PAGE 71

52 within the holder of XRD device. Copper radiation (K = 1.5405 ) in the 5 < 2 < 65 range was applied to generate the XRD signature of the material. 3.1.2 Mineral Compositions Composition of minerals of “Wyo-ben” bentonite from the XRD diffractograms is shown in figure 3.3. Most of the peaks match with those for montmorillonite and quartz minerals. 0 200 400 600 800 1000 1200 1400 1600 1800 2000 51525354555Angle (2 theta)Counts Figure 3.3 XRD Test Results for Bentonite

PAGE 72

53 3.1.3 Chemical Composition Chemical properties of the bentonite pow der were investigated in both dry and colloidal states. Analysis of the chemical composition of bentonite has been carried out on dry specimens using Energy Dispersive Spectroscopy (EDS ). Electrical conductivity and pH measurements were conducted on a be ntonite-water suspensi on as described in the following sections. 3.1.3.1 EDS Analysis Earlier research on chemical composition of various types of bentonite revealed that the major four minerals present in the sodium montmorill onite are silica (SiO2), alumina (Al2O3), sodium oxide (Na2O), calcium oxide (CaO), and iron oxide (Fe2O3). Other minor minerals, which normally consti tute less than 1% in total weight, are magnesium oxide (MgO), titanium oxide (TiO2), potassium oxide (K2O), manganese oxide (MnO), and trace metals oxide (Kaufhold et al., 2002; Nakashima, 2003; Singh et al. 2002; Guillaume et al., 2003; Christidis, 2001). The four major chemical compounds found in the bentonite are similar to those found in other clay minerals, except for the amount of calcium oxide (CaO) and the relati ve amounts of other constitutes (Ramirez, 2002; Guillaume et al., 2003; Bradbury and Baeyens, 2003; Nakashima, 2003; Benito et al., 1998). In this research, chemical analysis of the as-received bentonite was conducted by Energy Dispersive Spectroscopy (EDS). The as-received samples were oven dried and a 20-gram portion was used for EDS testing. The EDS technique uses X-rays resulting from interactions between applied fast beam electrons and the specimen atoms. X-ra ys, which are electromagnetic radiations of extremely short wavelength, are emitted when a specimen is bombarded with fast electrons. The X-ray energy and wavelength are related to the specimen’s elemental composition. When the specimen is bombar ded by the electron beam of a Scanning

PAGE 73

54 Electron Microscope (SEM), el ectrons are ejected from the innermost shell of the atoms comprising the specimen. An electron from an outer atomic shell drops into the vacancy in the inner shell in order to return the atom to its normal (balanced) state. This drop results in the loss of energy due to the diffe rence in energy between the vacant shell and the shell contributing the electron. The energy is given up in the form of electromagnetic radiation or X-rays. Since ener gy levels are different for different elements, characteristic rays are generated accordingly. Energy Dispersive X-ray micr oanalysis uses detection equipment to measure the energy values of the characteristic X-rays ge nerated within the electron microscope. An X-ray micro-analyzer system converts X-ra y energy into an electronic count by using semiconductor materials that can detect th e X-rays. The accumulation of these energy counts creates a spectrum, which is then plotte d against relative count s of the detected Xrays and evaluated for qualitative and quantit ative determination of the elements present in the specimen. The energy peaks are essentia lly fingerprints of the specific elements in a specimen. Figure 3.4 illustrates the basi c layout of an EDS system. Details of EDS have been described by Russ (1984) and Goldstein et al. (1981). EDS characterization of bentonite was conducted using a Hitachi S-800 spectrometer located at the Metrology Laboratory of the Nanomaterials and Nanomanufacturing Research Cent er (NNRC) at the University of South Florida. This spectrometer is also fitted with a Scanni ng Electron Microscope (SEM) as shown in figure 3.5. Bentonite powder was scanned usin g SEM to find its aggregated particle size, which is also shown in figure 3.6. Energy peaks for various elements for a bentonite specimen are shown in figure 3.7, which shows the major elements found usi ng EDS. The main chemical elements in the composition of bentonite are found to be oxygen, chlorine, silicon, aluminum, calcium, sodium, iron, sulfur, magnesium and some other trace metals. The chemical compounds that constitute the bentonite powder used in this research are SiO2, Al2O3, Fe2O3, Na2O, MgO, CaO, TiO2, K2O, MnO, and some other trace metal oxides. Table 3.1 shows the quantitative chemical compos ition of all the elements and trace metals derived from EDS.

PAGE 74

55 Figure 3.4 Schematic Diagram of EDS System Figure 3.5 Spectrometer Fitted with Sca nning Electron Microscope (HITACHI S-800) Display X-ray signal Amplifier Specimen Electron Beam Detector Basic Data Input Display X-ray signal Amplifier Specimen Electron Beam Detector Basic Data Input

PAGE 75

56 Figure 3.6 Dry Bentoni te Powder Under SEM Figure 3.7 Energy Peaks for Benton ite Chemical Elements Using EDS

PAGE 76

57 Table 3.1 Chemical Composition of Bentonite Major Chemical compounds Trace metals SiO2 66.5% Arsenic 0.1 ppm Al2O3 16.9% Barium < 1.0 ppm Fe2O3 7.4% Cadmium <0.01 ppm Na2O 2.3% Chromium < 0.05 ppm MgO 2.3% Lead < 0.1 ppm CaO 2.1% Mercury < 0.02 ppm TiO2 0.2% Selenium < 0.02 ppm K2O 0.4% Silver < 0.05 ppm It can be seen from the table that the main exchangeable ca tions in the double layer space would be sodium, calcium, magne sium and a small amount of potassium. 3.1.4 Electrical Conductivity and pH Electrical conductivity and pH of a bentonite-water suspension were investigated at various colloidal concentra tions. The electrical conductiv ity was measured by using an Accumet (model AB30) 4-cell conductivity meter (shown in figure 3.8) and two epoxy body electrodes of cell constant 1.0 cm-1 and 10.0 cm-1. These electrodes are capable of measuring a wide range of el ectrical conductivity from 10 to 200,000 microsiemens/cm. Whenever a change of electrodes was re quired to obtain a measurement within a particular range, it was necessary to recalib rate it using its own standard solution. Bentonite samples of various amounts were soaked into deionized water for at least 48 hours in order to adsorb as much water as possible with all the pores and interlayer spacing. Quantities of 5g, 10g, 15g, 20g, 30g, and 50g air-dry bentonite powder were mixed with 1 liter of deionized water to obtain 0.5%, 1%, 1.5%, 2%, 3%, and 5% of suspension respectively.

PAGE 77

58 Figure 3.8 Accumet (Model AB 30) 4-cell Conductivity Meter After soaking for at least 48 hours, the be ntonite suspension in deionized water was stirred for at least 15 minutes using a mechani cal stirrer/mixer before being poured into a 1 liter capacity glass beaker for self flocculation. After flocculation and subsequent settlement, supernatant water samples were co llected for electrical conductivity and pH measurement. Immediately following the sample collection, the pH of the non-acidified original sample was measured using an Accume t portable (model AP63) pH meter and polymerbody combination pH/ATC Ag/AgCl electrode as shown in figure 3.9. The pH meter was calibrated at three levels, using th ree standard color-code d buffer solutions of pH 4.00, 7.00 and 10.00. The variations of el ectrical conductivity and pH with respect to the percentage of suspension of bentonite in deionized water are shown in figure 3.10. The electrical conductivity of bentonite increases with in creasing amount of bentonite suspension in a s econd-order polynomial manner, wh ile the pH decreases with increasing bentonite suspension in a power e quation as shown in figure 3.10. At higher suspension concentrations, bentonite aggregate particles are unable to deflocculate and disperse in water, thus cont ributing less towards the total el ectrical conductivity of the water solution.

PAGE 78

59 Figure 3.9 Accumet Portable (Model AP63) pH Meter Figure 3.10 Electrical Conductivity and pH of Bentonite Suspension y = 10.009x-0.0341y = -29.987x2 + 399.62x 0 250 500 750 1000 1250 1500 1750 2000 01234567Bentonite suspension (%)Electrical Conductivity, S/cm 5 6 7 8 9 10 11pH EC pH

PAGE 79

60 3.1.5 Loss of Ignition The amount of organic content in the bent onite powder is indicated by the value of loss of Ignition (LOI), which is traditionall y expressed as total per cent of the material. LOI of fine grained powdered be ntonite is determined by burni ng at high temperatures in a controlled temperature oven. Loss of igniti on has been reported for various types of bentonite within a range of 0.2% to 5% (Keijer and Loch, 2001; Keijer et al., 1999; Lehikoinen et al., 1996). Bentonite clay specimens in this study we re burned at two different temperatures (550oC and 1000oC) in two separate specimens, 4.0 g and 2.0 g, respectively. After burning at 550oC temperature, the LOI was f ound to be 3% while at 1000oC the value rose to 5.6%. According to the technical information provided by the manufacturer of this bentonite (Wyo-Ben, Inc), LOI has been found to be 4.4%, wh ich falls within the range obtained in this study. LOI cannot be measured at very high temperatures because of the evaporation of the volat ile components of the bentonite material. Bentonite with higher LOI may or may not inte rfere with the chemical solu tions used during long term diffusion as well as hydraulic conductivity, but would produce organic compounds under long and sustained chemical and hydraulic flow as obser ved later in some of our experiments. 3.2 Grain Size Distribution Particle size distribution of bentonite cannot be obtain ed using either conventional dry sieve or hydrometer testi ng because of the aggregated nature of the particles. However, both sieve analysis a nd hydrometer test were carrie d out in order to investigate the amount of coarse fraction and fine-grain ed characteristics w ith various types of inorganic chemical solutions, and to gain a ge neral idea about the re lative distribution of clay aggregated particles.

PAGE 80

61 3.2.1 Hydrometer Test Bentonite particles, because of their hi gh surface charges, repel each other and exist as individual particles when mixed with bul k water. The size of the particle is in the range of 0.01 to 1 m and can be considered as colloidal Because of their high colloidal nature, bentonite particle sizes cannot be m easured by hydrometer analysis. Nonetheless, a review of the literature show that bentonite particle size distributions were carried out by Kozaki et al., (2001), Eriksen et al., (1999), Zhang et al., (1995), and others, where hydrometer and dry sieve mesh were used. A clay-water solution is a result of ho mogeneous dispersion of very small clay particles. The colloidal st ate lies somewhere between a solution and a suspension. Colloidal clay minerals are among the smallest crystalline particles known to exist, and are neither a suspension nor a solution. Clay colloids ar e hydrophobic in nature, meaning they have an inherent resistance to interacti on with water. Aggregated microscopic clay particles in colloidal solutions are usually less than 2 microns in diameter. Colloidal solutions do not settle under gravity within a reasonable time. When the dispersed particles accumulate into a larger lump or aggregate, which settles relatively rapidly under gravity, then the dispersion is called as “suspension”. The distinction in particle size between colloidal solutions and suspensions are arbitraril y taken in geochemistry as a Stokes radius (equivalent spherical radi us) of 1 micron (van Olphen, 1977). The equivalent particle size of any shape is co mputed in hydrometer tests velocity using Stokes Law. Particles smaller than 1 micr on are known as colloidal and larger than 1 micron are clay suspensions. Colloidal clay solutions produce Brownian motion, where the small clay particles display an erratic and random motion in al l directions. The water molecules are in constant thermal agitation, and their velo city distribution is determined by the temperature of the system. The motion of the water molecules, due to the fact that the fluid contains heat, causes the molecules to st rike the suspended clay particles at random. The impact makes the particles move, and the net effect is an erratic, random motion of the particle through the fluid. Brownian moti on is the result of thermal activity of water

PAGE 81

62 molecules around the clay particles. Water mo lecules in solution constantly collide with clay particles and push the particles in random direction due to the ne t resultant force. The mean kinetic energy of a molecule in the liquid, which is equal to the average translational kinetic energy of the particles, is given by: kT mv E2 3 2 12 (3.2) where, m is the mass of a particle, v is the velocity, k is the Boltzman constant, and T is the temperature. From this formula (equa tion 3.2) it can be seen that the mean kinetic energy of Brownian motion is proportional to th e temperature. Equati on (3.2) can also be used to find the velocity of a particle (Van Olphen, 1977). It can be seen from the above energy conservation theory that the average pa rticle velocity decreases with increasing mass, and Brownian motion does not exist for higher clay particles sizes. Collision with fluid molecules can also make a suspended particle rotate. This phenomenon is called rotational Brownian motion. It has been found that bentonite clay particles can flocculate in the presence of an electrolyte leading to an increase in particle sizes and a reduction in reactivity (Van Olphen, 1977; Sridharan et al., 1999; Zhang et al., 1995; Quirk and Schofield, 1955; Keren and Singer, 1988). To observe the relative particle sizes of clays, th e hydrometer technique was used with a dispersing agent in deionized water, and in various synthetic inorganic solutions without a dispersing agent. This method gives the effective particle size in different pore flui ds. Though Stoke’s law is not strictly valid for non-spherical particles settling at high velocity, it has been used to find out the relative particle sizes in different pore fluids. Six different solutions of various concentrations were used in addition to deionized water with 0.1M NaCl, 0.1M KCl, 0.1M MgCl2, and 0.1M CaCl2 as the lowest electrolyte concentrations. 3.2.2 Test Results and Discussion Figures 3.11 and 3.12 give the relative particle size distribution of bentonite in 0.1 molar concentration of four different solutions and in NaCl solutions with three different

PAGE 82

63 concentrations respectively. It can easily be observed that, relative ly speaking, bentonite is least aggregated and exists as finer particles with 0.1M NaCl and 0.1M KCl solutions in comparison with 0.1M MgCl2 and 0.1M CaCl2 solutions (figure 3.11). It can be concluded that an increase in the solution’s ionic valence increases particle aggregation and flocculation. With an increase in elec trolyte concentration, the bentonite particles have also become coarser. Compared to NaCl solutions, KCl solutions cause more aggregation of particles. A si milar trend was observed where CaCl2 caused more aggregation than MgCl2. Thus 0.1M CaCl2 causes maximum aggreg ation of particles among all the 0.1 molar solutions. Figure 3.11 Bentonite Particle/Aggregat e Distribution with Various Inorganic Chemical Solutions of 0.1 Molar of Concentration The flocculation increases with an increase in electrolyt e concentration, as shown in figure 3.12. Higher concentrated electr olyte solutions reduce the diffuse double layer thickness by attracting the nei ghboring particles, t hus creating aggregated particles which can easily flocculate and settle with time. 0 20 40 60 80 100 0.0010.010.11Particle Size (mm)% Fine r 0.1M NaCl 0.1M CaCl2 0.1M KCl 0.1M MgCl2 Deionized water

PAGE 83

64 Figure 3.12 Bentonite Particle/Aggregate Distribution with NaCl Solutions of Various Concentrations 3.3 Physical Properties Physical properties of be ntonite in terms of its specific gravity of solids and Atterberg limits, namely liquid limit and plastic lim it, are described in this section. Both ASTM standard and British standards have been used in these investigations. 3.3.1 Specific Gravity Specific Gravity, also known as SG, is a measure of the density of minerals compared to water. Minerals with a specific gravity under 2 are considered light, between 2 and 4.5 average, and greater th an 4.5 heavy (Faure, 1998). The specific gravity may slightly vary for a given mineral because of impurities present in the mineral 0 20 40 60 80 100 0.0010.010.11Particle Size (mm)% Fine r 0.1M NaCl 0.5M NaCl 0.7 NaCl Deionized water

PAGE 84

65 structure. Many researchers involved with bentonite materials have reported specific gravity values within a range of 2.4 to 2.65, depending on the percentage of montmorillonite mineral content (Malusis and Shackelford, 2002; Keijer and Loch, 2001; Keijer et al., 1999). The specific gravity of the bentonite pa rticles was measured according to ASTM D854-02 (2002) using a 500 ml pycnometer volumetric flask. Air-dry samples were soaked in tap water for at least 24 hours under vacuum so as to facilitate the removal of fine pore air bubbles from the water-clay so lution. The specific gravity of the solid particles was calculated usi ng the following equation. fs fw s s sW W W W G (3.3) where, Ws is the weight of the dry bentonite (taken after 24 hrs of oven dry at 105oC), Wfs the weight of the flask filled with bentonite and water, and Wfw the weight of the flask filled with deaired water only. The average specific gravity of the bentonite used in this study was measured to be 2.55, as shown in the figure 3.13. Figure 3.13 Experimental Va riation of Specific Gravity 0 1.5 3 4.5 123456Experiment numberSpecific gravity .

PAGE 85

66 3.3.2 Atterberg Limits The plasticity behavior of bentonite clay and the effect of pore fluid on the liquid limit help predict the long-term performance of the liner. Atterberg limits of soils can also be used to identify the mi neral contents of the soil materials using the plasticity chart shown in figure 3.14 (after Holtz and Kovacs, 1981). The liquid limit of bentonite is very high compared to other clay minerals because of its ability to disperse into extremely small particles with a tremendous amount of potentially absorbing su rface. The liquid limit of bentonite is primaril y controlled by its diffuse double layer thickness. The numerous factors affecting the thickn ess of diffuse double layer depend upon the characteristics of the pore fluid which are ex plained in chapter two, namely dielectric constant, electrolyte concentration, valence of the electrolyte, and temperature. An increase in the diffuse double layer thickness causes an increase in the liquid limit. Figure 3.14 Plasticity Chart (after Holtz and Kovacs, 1981) The liquid limit (LL) of bentonite was determined using water and electrolyte solutions. The Casagrande method for liquid limit testing, as de scribed in ASTM D4318-00 (2000), was tried unsuccessfully for be ntonite. Because of the “stickiness” of Halloysites Chlorites Illites A-line U-line Kaolinites Montmorillonites 020406080100 0 10 20 30 40 50 60 Liquid Limit LL (%) Plasticity Index PI (%) PI = 0.73(LL –20) PI = 0.9 (LL –8) CL-ML ML or OL MH or OH C L o r O L C H o r O H Halloysites Chlorites Illites A-line U-line Kaolinites Montmorillonites 020406080100 0 10 20 30 40 50 60 Liquid Limit LL (%) Plasticity Index PI (%) PI = 0.73(LL –20) PI = 0.9 (LL –8) CL-ML ML or OL MH or OH C L o r O L C H o r O H

PAGE 86

67 the bentonite particles, no groove can be cut through the materi al placed in the Casagrande apparatus. Instead the liquid lim it of bentonite clay was determined by the cone penetration method (BS1377-1975), wher e it is defined as the water content corresponding to 20-mm penetration. The plastic limit of the soil was obt ained by ASTM Test Method (D 4318-00) using a rolling apparatus. Th e test specimens were prepar ed by mixing with water and storing for at least 24 hrs for uniform absorpti on. The average values of liquid limit and plastic limit using deionized wa ter as pore fluid for the bentonite used in this research were found to be 546% and 56% respectively. Montmorillonite minerals plot at extreme locations on the plasticity chart, close to the U-line, because of their high absorption capacity (figure 3.15). The bentonite clay used in this study lies slightly above the Uline, which falls out of the montmorillonite mineral zone depicted in the A-chart shown in figure 3.15. This slight de viation of plasticity index (PI) from the theoretical U-line might be due to the arbitrary st raight-line definition of the Uline, especially at such high liquid limits. A similar deviat ion was also reported by Ma lusis and Shackelford (2002) for the bentonite used in thei r research. Other researchers have found liquid limit values of smectite minerals as high as 1000% (Mesri and Olson, 1971; Alther et al., 1985; Reschke and Haug, 1991). Deionized water, tap water, and four different inorganic salt solutions with four different concentrations were used in liquid limit investigations of bentonite material in this research study. Figure 3.16 shows the variation of cone pe netration with water content for deionized water, tap water, and one molar solutions, while figures 3.17, 3.18, and 3.19 shows the results for 0.5 molar, 0.1 molar, and 0.01 molar salt solutions respectively.

PAGE 87

68 Figure 3.14 Wyo-ben bent onite in Plasticity chart Figure 3.15 Wyo-Ben Bentonite on the Plasticity Chart Figure 3.16 Penetration vs. Water/Soluti on Content (Water a nd 1 Molar Solution) yDeionized water = 27.511x 3.7004 yTap water = 22.124x + 86.144 y1M NaCl = 4.3834x + 80.37 y1M CaCl2 = 2.889x + 71.054 y1M KCl = 3.8519x + 83.912 y1M MgCl2 = 3.6687x + 70.275 0 100 200 300 400 500 600 700 800 900 05101520253035Penetration (mm)Water or solution Content (%) deionized Tap water 1M NaCl 1M KCl 1M MgCl2 1M CaCl2

PAGE 88

69 Figure 3.17 Penetration vs. Water/Solution Content (Water and 0.5 Molar Solution) It can be seen from figures 3.16 and 3. 17 that there is a distinct difference between the liquid limit in water and that in hi gher concentrated salt solutions. This is due to fact that the bentonite particles aggregate due to the reduction in double layer thickness. However, at lower concentrations of salt, the bentonite behavior resembles that of water as seen in figure 3.19. As tap water contains some inorganic ionic compounds, the liquid limit using tap water is sli ghtly lower than that of deionized water, which produces the maximum liquid limit for bentonite clay. Comparing the liquid limits at various con centrations of salt solutions, it can be concluded that their value decrease with increasing electrolyte concentrations. For monovalent cationic solutions, it can be shown from figures 3.16 – 3.19 that the liquid limit for KCl solutions are lower than those of NaCl, possibly due to stronger ionic linkage of potassium ions with the negative interlayer and inter-par ticle surface charges, which causes a reduction in diffuse double layer. The variation of liquid limit for electrolyte solutions of monova lent and divalent cationic solutions can be seen in figures 3.20 and 3.21. Divalent cations are able to replace the exchangeable monovalent cations in the interlayer surfaces, causing a yDeionized water = 27.511x 3.7004 yTap water = 22.124x + 86.144 y0.5M NaCl = 5.0024x + 113.85 y0.5M CaCl2 = 2.6x + 79.241 y0.5M KCl = 4.7778x + 88.999 y0.5M MgCl2 = 3.9926x + 78.304 0 100 200 300 400 500 600 700 800 900 05101520253035Penetration (mm)Water or solution Content (%) deionized Tap water 0.5M NaCl 0.5M KCl 0.5M MgCl2 0.5M CaCl2

PAGE 89

70 reduction in double layer thickness, which even tually reduces the absorption capacity of bentonite. As a result, the liquid limit of th e material decreases. Liquid limits for CaCl2 solutions were found to be the least among all the solutions because of its higher replacement capacity and its lowe r water absorption affinity. Figure 3.18 Penetration vs. Water/Solution Content (Water and 0.1 Molar Solution) Figure 3.19 Penetration vs. Water/Solution Content (Water and 0.01 Molar Solution) yDeionized water = 27.511x 3.7004 yTap water = 22.124x + 86.144 y0.1M NaCl = 12.463x + 158.94 y0.1M CaCl2 = 4.4898x + 127.51 y0.1M KCl = 13.141x + 112.87 y0.1M MgCl2 = 5.385x + 110.98 0 100 200 300 400 500 600 700 800 900 05101520253035Penetration (mm)Water or solution Content (%) deionized Tap water 0.1M NaCl 0.1M KCl 0.1M MgCl2 0.1M CaCl2 yDeionized water = 27.511x 3.7004 yTap water = 22.124x + 86.144 y0.01M NaCl = 17.869x + 161.04 y0.01M CaCl2 = 15.095x + 147.81 y0.01M KCl = 18.487x + 154.75 y0.01M MgCl2 = 12.652x + 209.84 0 100 200 300 400 500 600 700 800 900 05101520253035Penetration (mm)Water or solution Content (%) deionized Tap water 0.01M NaCl 0.01M KCl 0.01M MgCl2 0.01M CaCl2

PAGE 90

71 Figure 3.20 Variation of Liquid Limits with Electrolyte Concentration Figure 3.21 Variation of Liquid Limits with Types of Electrolyte Solutions 0 100 200 300 400 500 600 00.10.20.30.40.50.60.70.80.911.1Solution Molarity (M)Liquid Limit (LL) % NaCl KCl MgCl2 CaCl2 0 100 200 300 400 500 600 NaClKClMgCl2CaCl2SolutionLiquid Limit (LL) % 0.01 M 0.1 M 0.5 M 1.0 M

PAGE 91

72 3.4 Swell Index Swell index tests were carried out according to the ASTM Standard Method (ASTM D 5890-02) using the various concentr ation electrolyte solutions which were presented in the previous section. A m odified method was used by Reschke and Haug (1991), where 3g of samples were allowed to soak for 24 hours, and the swell index was calculated by dividing th e swell volume by the specific grav ity of sample. Swell index of Wyoming bentonite has also b een investigated by Alther et al., 1985, Xeidakis (1996), Zhang et al., (1995), Jo et al., (2001), Stern and Shackel ford (1998), Shackelford et al., (2000), and others. It has been found to ra nge between 25 and 65 ml/2g. The main causes of swelling of smectite clay (i.e. bentonite) are (1) the magnitude of cation exchange capacity of the clay mineral inte rlamellar surface, (2) the type of cations present within the clay surfaces, and (3) the interaction between cations and water molecules (Odom, 1984; Alther et al., 1985; Kster, 1996; Kjellander et al., 1988; Shackelford et al., 2000). 3.4.1 Test Procedure The newly published ASTM D 5890-02 sta ndard was adopted in determining the swell index of the bentonite. To perform these tests, a 2g sample of dried and finely powdered bentonite clay is dispersed into a 100 ml graduated cylinder in 0.1g increments. A minimum of 10 minutes must pass between additions to allow for full hydration and settlement of the clay to the bottom of the cy linder. These steps are repeated until the entire 2g sample has been added. The sample is then covered and protected for a period of 16 24 hours, at which time the level of the settled and swollen clay is recorded to the nearest 0.5 ml. The swell index is expressed in ml/2g of bentonite.

PAGE 92

73 Deionized water and four types of salt so lutions at various concentrations were used in swell index experiments. The result s and effects of various chemical solutions are discussed in the following sub-section. 3.4.2 Effect of Chemical Solution Species The swell index of Wyo-Ben bentonite was found to be 60 ml/2g for the suspension in deionized water. However, th e maximum swell index of 67 ml/2g has been measured in 0.01 molar NaCl solution as show n in figure 3.22, which could be due to the possibility of more quasi-crystalline wate r layer formations in and around bentonite particles. As suggested by Odom (1984), th e maximum adsorption occurs when Ca and Mg cations together constitute to 5 1 of the total exchangeable cations, which allows several layers of water to be developed by Na ions. The 0.01 molar NaCl solution may supply the appropriate amount of free sodium ions that c ould form several layers of quasi-crystalline water layer around the surface of the bentonite, which are responsible for higher swelling and hydration. Figure 3.22 Swell Index of Bentonite in Inorganic Chemical Solutions 0 10 20 30 40 50 60 70 80 NaClKClMgCl2CaCl2Liquid solutionSwell Index mL/2g 1 M 0.5 M 0.25 M 0.1 M 0.05 M 0.01 M

PAGE 93

74 It can be seen from figure 3.22 that th e swell index is higher for monovalent solutions compared to divalent solutions of si milar concentrations. Swell index values in KCl solutions are lower than those of NaCl so lutions of equal concentrations because of the ability to form rigid linkage between potassium ions and ne gatively charged clay surfaces. This strong linkage could reduce th e diffuse double layer thickness, which is directly responsible for the redu ced swelling volume of the bentonite clay materials. In the case of divalent solutions, the divalent cations replace the m onovalent cations on the surface exchangeable space and reduce the thic kness of the diffuse double layer. As shown in figure 3.23, minimum swell index values are obtained in the case of higher concentrations of CaCl2 solutions since at higher co ncentration, calcium cations can substantially replace monovalent cations and water molecules on the surface of the bentonite. Figure 3.23 Variation of Swell Index with Concentration of Salt Solutions 0 10 20 30 40 50 60 70 80 00.20.40.60.811.2Liquid Solution Concentration (M) Swell Index mL/2g NaCl KCl MgCl2 CaCl2

PAGE 94

75 3.5 Cation Exchange Capacity of Bentonite Researchers have been measuring the cation exchange capac ity (CEC) of clay minerals in many different ways. CEC can be determined by the sodium saturation method as described by Chapman (1965), where the soil sample is first saturated with sodium, and the sodium is subsequently repl aced by ammonium ions. The concentration of the recovered sodium is determined by fl ame photometry which is then expressed in terms of meq/100 g of oven dry soil (Wentink and Etzel, 1972). Othe r methods used in measuring CEC are X-ray diffraction (Ben et al., 2000; Kaufhold et al., 2002), infra-red chromatography/spectroscopy (Petit et al., 1998; Hwang and Dixon, 2000), cesium chloride adsorption (Itami and Tamamura, 1999 ), adsorption of a c opper ethylenediamine complex (Bergaya and Vayer, 1997), and strontium chloride adsorption (SrCl2) (Schaefer and Steiger, 2002). Research on methylene blue adsorption on clay minerals has been conducted extensively by analyzing clay samp les collected from various parts of the world (Hang and Brindley, 1970; Grim 1968; Higgs, 1988; Taylor 1985; Santamarina et al., 2002). 3.5.1 Methylene Blue Test Procedure Methylene blue (MB) adsorption was found to be one of the most reliable and simple methods to obtain information on the pr operties of clay minerals, including cation exchange capacity (CEC) of soils and other fine grained mine rals. It is also used as an indirect quality indica tor for swelling activity of clay materials. If a significant amount of methylene blue is adsorbed by the clay mi nerals, this may lead to the conclusion that the clay’s swelling activity is higher, ev en though some other minerals which do not swell might also adsorb methylene blue. The cations in the diffuse double layer ar e exchangeable with those in the free water. Therefore, methylene blue will excha nge cations from both of these sources. In order to eliminate the excessive value of free water cations, clay samples are needed to be

PAGE 95

76 mixed with sufficient deionized water in or der to dissolve the precipitated salts and cations existing in the free pore fluid. The methylene blue molecule consists of an organic base in combination with an acid as shown in figure 3.24. The size of a single molecule of methylene blue as drawn in figure 3.25 has been reporte d by a number of authors and is tabulated in Table 3.2. Figure 3.24 Methylene Blue Chemical Structure When methylene blue (dye/powder) is dissolved in water, it will disperse to form a monomer (single molecules) at lo wer concentrations (less than 10-3 mol/m3), or in monomer-dimer (2 molecules) equilibr ium at higher concen trations (about 10-2 to 1 mol/m3) (Taylor, 1985). The chemical formula of C16H18N3SCl corresponds to a molecular weight of 319.87 g/mol for methyl ene blue dry die. The methylene blue C16H18N3SCl CH3CH3 Cl-H3C H3C N N S N+C16H18N3SCl CH3CH3 Cl-H3C H3C N N S N+ H B L H B L Figure 3.25 Schematic Diag ram of Methylene Blue Molecules (after Taylor, 1985)

PAGE 96

77 molecule contains a negatively charged (Cl-) ion and a large positively charged ion which drives away the positively charged cations loosely bonded with the clay surfaces at the internal and external faces. The process cont inues until all the cations have been expelled and replaced by methylene blue molecule s with higher fixation attraction. Table 3.2 Dimensions of Methylene Blue Single Mol ecule (After Taylor, 1985) L B LxB L H LxH B H BxH Author (nm) (nm) (nm2)(nm) (nm) (nm2) (nm) (nm) (nm2) Hofman et al. 1.95 White and Cowen 1.95 0.25 Kipling and Wilson 1.6 0.84 1.34 1.60 0.47 0.75 0.84 0.47 0.39 1.25 0.57 0.71 1.25 0.51 0.64 0.57 0.51 0.29 Hofmann et al. 1.50 0.65 0.98 1.50 0.77 0.65 0.33 Hang and Brindley 1.29 0.55 0.25 Methylene blue replaces the clay cations irre versibly as indicated by following reactions. Ca-Na-Mg clay + Methylene blue (MB) hydrochloride MB Clay + Ca-Na-Mg chloride Na-bentonite indicates higher CEC because of its higher interlayer spacing as compared to Ca-bentonite where the entry of methylene blue molecules is expected to be restricted because of its limited in terlayer (lattice) expansion. Methylene blue chloride powder (Fisher Scientific, Pittsburgh, PA) was used in this research and the spot method (European standard) has been adopted for measuring CEC for bentonite material. The test pr ocedure can be briefly described as follows: (a) The methylene blue solution is prepar ed by mixing methyelene blue powder and deionized water at the ratio of 1g to 200 cc water. (b) A sufficient amount of deionized water is added to the bentonite clay at about 500 mg to 2 g, so as to produce a suspen sion or slurry consis tency of the clay particles.

PAGE 97

78 (c) A magnetic stirrer with a speed of 400 to 700 rpm is used to stir the bentonite in a glass container continuously until the end of titration by methylene blue. (d) A methylene blue solution is added to the clay suspension in 0.5 ml increments and stirred for at least 15 minutes. (e) After each addition of methylene blue, a small amount of clay suspension after stirring is removed by a glass rod and then placed on Fisher brand filter paper P5. (f) The “end point” is expected to be reac hed when a permanent light blue halo around the wet soil spot is formed by th e unabsorbed methylene blue in excess of the amount required to replace the exchangeable cations of the clay particles. (g) In order to confirm the end point, step 5 is repeated after a longer stirring for about an hour in order to totally ad sorb the methylene blue on the clay surface. If the halo disappears on the f ilter paper, 0.5 ml of MB is added and steps 5-7 are repeated until a permanent halo appears around the wet clay spot on the filter paper. The total volume of methylene blue solution added in this process is recorded and used to calculate the ca tion exchange capacity by the following equation. (mEq/100 g) (3.4) 3.5.2 Test Results and Discussion CEC is normally expressed in meq/100 g of clay sample. The CEC of relatively pure smectite clays ranges betw een 70 and 130 meq/100 g (Keijer et al., 1999; Triantafyllou et al., 1999; Shackelford and L ee, 2003; Odom, 1984; Sanchez et al., 1999; (g) dry wt 100 (cc) solution MB of 1000 87 319 (g) dry wt MB x (cc) added . Clay g x Vol x MB C E C

PAGE 98

79 Gleason et al ., 1997; Kahr and Madsen, 1995). However, Malusis and Shackelford (2002) investigated a bentonite havi ng a CEC of 47.7 meq/100 g. Figure 3.26 Cation Exchange Capacity of Bentonite CEC values were calculated for various amounts of clay content and MB solution concentrations using the above equation (3.4). Figure 3.26 shows the values of CEC with respect to the amount of water added to the bentonite samples used in this study. Three different amounts of air-dry bentonite samp les (0.5 g, 1 g, and 2 g) were mixed and stirred with various amounts of water before adding MB. It can be concluded that no distinct variation of CEC is noticeable due to the amount of bentonite and water added in these experiments. The CEC of the Wyo-Ben bentonite is, therefore, measured to be between 80 and 90 meq/100 g, with an average value of 83 meq/100 g. 50 60 70 80 90 100 050100150200250Water added (mL)CEC (mEq/100g) 0.5g Bentonite 1g Bentonite 2g Bentonite

PAGE 99

80 CHAPTER FOUR EQUIPMENT DESIGN & FABRICATION 4.1 Permeability Equipment Standard pressure panels for permeabili ty tests are commonly used to perform hydraulic conductivity tests, but cannot be used directly with highly corrosive leachates collected from landfill sites and or synthetically mixed chemicals. Aggressive inorganic or organic chemicals can cause the panel t ubes and fittings to corrode. To protect the pressure panel, special buffer cells were designed and fabricated as shown in the schematic diagram in figure 4.1. These buffe r cells are connected w ith the various inlets and outlets of the main pressure panels. Th e main criteria considered during design and fabrication of these cells are (a) to be used as a substitute for the burette attached to the pressure panels, (b) to prevent corrosive liquids or leachates from contacting metal fittings and pressure regulators, and thus prev enting corrosion, clogging and damaged panel parts, (c) to accommodate a large quantity of influent to be permeated through the specimen in an uninterrupted fashion, (d) to fill the cells in an uninte rrupted fashion whenever necessary, (e) to collect the influent and e ffluent samples at any pore volume of permeation while continuing the permeability test, and (f) to apply and maintain any specifi c hydraulic gradient during the test.

PAGE 100

81 Figure 4.1 Schematic Diagram of Permeability Test Setup 4.1.1 Design Concept The permeability cells designed and fabricat ed in this study are suitable for use in both constant head as well as variable head conditions in rigid wa ll and flexible wall permeameters. The design of the permeability buffer cells has been incorporated to suit the existing pressure panels available in the Geoenvironmental Laboratory at the University of South Florida (USF). Compre ssed air pressure regulated from the panel can be utilized through the buffer cells. Connect ions of the cells with the pressure panels were made in such a way that the influent and effluent pressure during the permeability tests can be monitored through the electroni c indicator mounted on the main pressure InfluentTank Permeability Cell Ash waterPressure Lines Pressure supply Pressure panel Effluent Tank InfluentTank Permeability Cell Ash waterPressure Lines Pressure supply Pressure panel Effluent Tank

PAGE 101

82 panel board. Buffer cells were designed to accommodate permeability tests on many different types of soils with a wide range of permeability. Soils with low permeability such as bentonite and other clays, caus e small amounts of flow through the specimen even at higher hydraulic gradie nts while soils with high perm eability such as coarse sand generate a high rate of flow under low gradient s. Buffer cells have been designed with three chambers of different diameters to i ndicate with high precisi on any amounts of flow in the influent and effluent tubes. Since the duration or pore volume of flow is an important factor in investigating permeability of soils to chemicals, the cells were designed considering a high volume capacity so that the tests c ould be carried out uninterrupt ed overnight or for longer intervals. Uniformity of synthetic chemical solutions can be maintained within the effluent cell for a long duration of time during the pe rmeability since the cell chambers are made of chemically inert materials, and because of their high storage capacity. To verify the chemical composition of synthetic solutions, to monitor leachates quality in the influent cell, and to conduct chemical analysis of the effluent solution, control valves have been provided at the bottom of both influent and effluent cells which provide easy sample collections at any interval of time while running the permeability tests. Replenishment of influent can be carr ied out in any quantity through the bottom control valves by hydrostatic force or under pr essure, and through the top control valves using a syringe or fill pump connected with th e pressure panel. Replenishment from one chamber to another can also be achieved by using a pressure differential across the connecting bridge tubing during the test. Highly concentrated chemical solutions and contaminant leachates used as permeants might cause chemical precipitation and deposition within the cells as well as in the connecting tubings. Light precipitation and deposition wi thin the cell chambers can be dissolved or removed in one chamber at a time by water jetting or using cleaning solutions without discontinuing the tests. High precipitation can be cleaned after completion of the tests when the individua l parts and tubings are dismantled and are typically cleaned using conventi onal or special acid cleaners.

PAGE 102

83 4.1.2 Materials and Fabrication The cells, as shown in figure 4.2, consist of three different diameters clear acrylic (Plexiglass) cylinders placed in between two metallic plates. To prevent the chemical corrosion of the metallic plates due to highly concentrated synthetic chemical solutions, the bottom metallic plate was made of highl y corrosion-resistance Type 316L stainless steel. Plexiglass cylinders are fitted into grooves cut on the top and bottom plates. The cells are fabricated in such a way that, after tightening, no leakage of liquid is possible even at high applied pressures (100 to 120 ps i). All cylinders are 11 inches long and graduated in length and volume, which allows the measurement of th e permeants with an accuracy of 1 mm in elevation and 0.1 cm3 in volume. The cylinders are connected at the top and bottom by flexible transparent rubbe r tubings to two central blocks, which are made of stainless steel. All the connecting lines are made of tran sparent nylon tubes of -inch outside diameter. Stainless steel central blocks were chosen for their high strength and chemical /corrosion resistance ch aracteristics. The top central block is connected to the burette of the pressure panel, through which th e pressure is regulated. The other central block, which is attached at the base, is connected to the permeation chamber where the specimen is placed. The top central block, connecting each of the three buffer cylinders controls the applied pressure from panel regulator with an accuracy of 0.1 psi. However, since no separate regulator is attached to individual buffer cells, the pressure applied on the each of the cylinders is constant. Each cylinder is equipped with a vent valve which is used to release pressure and to facilitate backfilling of the permeant liquid. The bottom central block connecting each of the three cylinders at their bases controls the flow of permeant in and out of th e cylinders. Nickel plated ball valves are connected to each of the outlet nylon tubes at the bottom central block, and are used to control flow of the permeant from individual cylinders and to collect liquid samples at any time during the permeability tests without interrupting flow through other tubes. The required amounts of permeant samples (leac hate) before and after passing through

PAGE 103

84 the test specimen are collected from the corresponding (active) cylinder for further chemical analysis. Figure 4.2 Schematic Diagram of Permeameter Cell Large cylinder Medium cylinder Small cylinder Bottom block Top block Graduated scale in length and volume Large cylinder Medium cylinder Small cylinder Bottom block Top block Graduated scale in length and volume Top view of bottom block To large cylinderInfluent EffluentTo small cylinder To medium cylinderVent Vent Control valve Top view of bottom block To large cylinderInfluent EffluentTo small cylinder To medium cylinderVent Vent Control valve Air Pressure from PanelTop view of top blockTo large cylinder To small cylinder To medium cylinder Air Pressure from PanelTop view of top blockTo large cylinder To small cylinder To medium cylinder

PAGE 104

85 Since the inside diameter of the largest cylinder is 5-inch, over one gallon (4 liters) of permeant can be stored. Each test can therefore, run for quite a substantial period of time depending on the applied hydr aulic gradient, soil permeability, and thickness of the test specimen. However, for very low permeability specimens, where the amount of permeant is very small, the small diameter ( -inch ID) cylinders are used for flow measurement as well as for collecti ng the permeant. The small cylinders can also be used for highly permeable specimens to determine the coefficient of permeability (hydraulic conductivity) and to collect sma ll amount of liquid permeants over a short period of time. The flow of permeant from any cylinder can be cut-off and switched to another cylinder within the same cell using the botto m central block. Throughout the process, the flow remains uninterrupted. By cutting o ff the flow, the permeant can be collected or replenished up to any desired level while the permeability test is continuing using the other cylinder. Backfilling into the cylinde r can be expedited by releasing the attached vent valve placed at the top of the cylinder. Any cylinder can be separately cleaned of any chemical deposition or sedimentation by flus hing it with cleaning agents or tap water. The entire cell can also be dismantled afte r completion of any test, and cleaned and reassembled for subsequent experiments. The permeameter, as shown in figure 4.3, consist of 5--in OD and 5-in ID clear acrylic (Plexiglass) cylinder placed in betw een two metallic plates and two 4-in diameter stainless steel platens. Plexiglass cylinders are fitted into the grooves cut on the top and bottom plates. The height of the permeameter cylinder is 12 inches which can easily accommodate specimens of up to 8-in long. A control valve at the top plate of the permeameter is connected with the pressure panel through which compressed air is applied to the cell. Pressure applied at the cell liquid surface acts as the cell pressure for the specimen, which is submerged in. Three stainless steel control valves are connected at the base plate of the permeameter as shown in figure 4.3 (b). The middle control valve is connected at the bottom of the platen through which the infl uent enters into the test specimen. After permeating through the sp ecimen, the effluent flows through the top

PAGE 105

86 platen into the effluent buffer cell. All the connecting lines are of transparent nylon tubes -inch in outside diameter. A split mold 4-in ID and 1-in long has been designed and fabricated to be used in preparing the clay specimens onto the bottom pl aten within the permeameter. The split mold fitted with a rubber membra ne is placed flush with th e bottom platen so that the loose dry bentonite powder does not slip through the sides of the platen while preparing the specimen. Both top and bottom plates of the permeameter are made anodized iron for corrosion resistance and l onger service life. (a) (b) Figure 4.3 Schematic Diagram of Flexible Permeameter (a) Permeameter Cell (b) Bottom Connection Water filled Air Pressure from Panel Rubber membrane Bentonite specimen Bottom platen O-ring Porous stone Water filled Air Pressure from Panel Rubber membrane Bentonite specimen Bottom platen O-ring Porous stoneTop view of bottom connection To fill/drain cell To effluent tank From influent tank Control valve To / from cell To cell From cellTop view of bottom connection To fill/drain cell To effluent tank From influent tank Control valve Control valve To / from cell To cell From cell

PAGE 106

87 4.2 Diffusion Equipment Cheung, (1994) described the use of an used apparatus for in-diffusion (electrolyte solution passes through the speci men from high to low concentration) and through-diffusion (electrolyte solution passe s over one side of the specimen and is collected from the other side), Eriksen, et al ., (1999) used a diffusion cell where bentonite was statically compacted in the diffusion cy linder (internal diamet er of 10 mm and length of 5 mm) to a dry density of 1800 kg/m3. Inlet and outlet channels were fitted with a metallic filter (0.82 mm thick), and the clay was equilibrated with the aqueous solution for at least three weeks by pumpi ng a groundwater solution. Higashi et al ., (1990) investigated the diffusivity of nuclide transpor t in water (titrated wa ter) through bentonite by using a diffusion cell 4 cm long and 2 cm in diameter. Pre-saturation of samples was carried out by submerging the specimen in water after placing it into the cell. Diffusivity of ions, especially radio-nuclides, through comp acted sodium bentonite were investigated by Kim et al ., (1993) by a method called “back-to-back ”, where the source solution is allowed to diffuse in plane from the center toward both ends of the specimen. In this method the bentonite clay was saturated with the solution to form a slurry before being dried and cut into slices 2.5 cm in diameter and 2 cm in length. The sliced specimens were then placed into the diffusion circular metallic cells where the specimens were allowed to swell upon saturation with the solution to a predet ermined size and volume. An equipment called “DKS permeameter” (diffusion, convection, sorption) was used to study soil-contaminant transport m echanisms by Mahler and Velloso (2001). In this technique, the soil sample is molded in the middle of the pe rmeameter, and both source solution and distilled water are allowed to percolate into the top and bottom channels, which are made of highly permeable porous materials, thus creating a constant concentration gradient through the specimens. The diffusion characteristics of compacted sodium bentonite in terms of ionic charges and orientation of clay particles were investigated by Sato (2000) and Sato and Suzuki (2003) using through-diffusion tec hniques, where bentonite specimens were placed in a diffusion cell and then compacted and saturated with various electrolyte

PAGE 107

88 solutions. Diffusion tests for compacted clay using compaction mold type cells were also used by Shackelford (1988, 1990, 1991, 1994), Shackelford and Daniel (1991), Shackelford et al ., (1999), where the clay samples were compacted in the mold at optimum water content before being saturated wi th water for a period of 17 to 160 days. Lake and Rowe (2000, 1997) devised an apparatus similar to the one used in this study to measure the diffusion coefficient of GCL materials under specified volume diffusion (SVD) condition, where various type s of inorganic chemical solutions were used as a source (figure 4.4). Here the clay specimens were allowed to swell by hydration up to any degree, resulting in a wi de range of void ratios. SVD allows the comparison of diffusion results in various co ntrolling solutions by controlling the final saturated void ratios of the be ntonite clay specimens. Most of the above diffusion cells were adopt ed to satisfy the i nvestigators interest in particular factors, fiel d requirements and environmenta l conditions such as highly traceable chemical elements, long periods of diffusivity, simulating in-situ compaction and saturation, or automation of the set-up among others. Figure 4.4 Specified Volume Diffusi on Cell (After Lake and Rowe, 2000)

PAGE 108

89 4.2.1 Design Concept The following factors have been considered in designing the diffusion cell in this study. (a) It is necessary to obtain a uniformity swelling bentonite clay specimen in terms of density and water content. (b) Full saturation of the clay specimens must be achieved within a short period of time. (c) The void ratio of the specimens must be varied by controlling the thickness and changing the dry benton ite weight, or by controlling the weight while thickness at full saturation. (d) Collection of the source and receptor solutions must be easily done while continuing diffusivity testing thr ough the specimen when necessary. (e) Disturbance to the prepared samples within the mold must be avoided. (f) Provisions must be made to appl y a hydraulic gradient through the specimen for further advection analysis if required. (g) High storage capacity of the source solution must be secured in order to continue the diffusion experiment for a long period of time without affecting the quality and con centration of the solution. The diffusion apparatus designed in this st udy is of the considered to specified volume diffusion (SVD) type as shown in figure 4.5, where the volume of the specimen remains the same throughout the whole diffusi on process. Specimens are made by slowing consolidating the sl urry samples prepared by mixing bentonite with high amounts of water (above their liquid limit). The idea of making specimens from slurry has come up from the fact that dry bentonite powder starts to swell as soon as it comes into contact with water. An outer sealed la yer is created, so uniformity of the specimen cannot be achieved for small amounts of bentonite mixed with water. Furthermore, the amount of swelling bentonite is not uniform across portions of the specimen, which could develop channels for the fluid to pass through during diffusion. It is also not possible to make uniform bentonite samples by mixing with low amount of water (below liquid

PAGE 109

90 limit) because of air bubbles trapped within the specimen during preparation. Preparing the specimens outside the diffusion cell may cause disturbance to specimens, making it difficult to place the specimen uniformly in the cell. Both source and receptor containers need to be transparent so that the level of fluid can be monitored during diffusion and refi lled if necessary to maintain a constant level. It is noted that fl uid levels may change due to osmotic flow, thus creating an unwanted hydraulic gradient. Internal piston O-ring Soil slurrry Porous stone Graduated pipette Figure 4.5 Diffusion Set-up with Clay Slurry (a) Initial Before Consolidation (b) Final After Consolidation (a)(b) Internal piston O-ring Soil slurrry Porous stone Graduated pipette Internal piston O-ring Soil slurrry Porous stone Graduated pipette Figure 4.5 Diffusion Set-up with Clay Slurry (a) Initial Before Consolidation (b) Final After Consolidation (a)(b)

PAGE 110

91 4.2.2 Materials and Fabrication Constant head rigid wall permeameters manufactured according to ASTM D-2434 have been modified to create the diffusi on cells as shown in figure 4.5. Rigid permeameter walls which contain the source fluid and the clay specimen have been fabricated in the machine shop at the University of South Fl orida. The cylinder is made of transparent acrylic material (plexiglass) 6-in long, 3-in ID and -in thick. A thick Oring is placed within the groove on both the bottom base plate and the top metallic platen in order to prevent any leakage during diffusi on. In order to furt her prevent leakage, sufficient vacuum grease was pasted in and around the O-rings. The acrylic chamber permits viewing of the sample during testing. The end plates are constructed of anodized aluminum for rust resistance. An internal piston, which is placed inside the diffusion cylinder, is also made of acrylic material a nd 2-in OD and -in thick. The receptor is a graduated pipette -in internal diameter a nd 10-in long which can c ontain up to 30 mL of solution. A cap is fitted at the open end of the pipette to prevent any ingress of impurities and evaporation from the solution during testin g. A hand-held rubber suction pump with a smaller diameter pipette is used to collect receptor fluid samples for further chemical analysis during diffusion tests. Both the source chamber and receptor are gr aduated so as to m onitor the level of the fluid. Porous stones 3-in in diameter and -in thick are placed at top and bottom of the clay specimen inside the source ch amber to provide filtering during sample preparation and to maintain uniformity of th e specimen thickness. Because a tight seal was required, fitting of the porous stones inside the source chamber was one of the most difficult tasks in the whole assembling pro cess. The bottom porous stone had to be placed before the slurry sample was poured into the chamber, while the top one was placed after pouring the slurry. Grooves were cut along the perimeter porous stone so as to fit O-rings as shown in figure 4.6. A sufficient amount of vacuum greas e was pasted along the O-ring and circumference of the top porous stone before placing at the top of the source chamber and subsequently pushing through the plexiglass of the chamber. Tight dimensional

PAGE 111

92 tolerance O-ring is necessary so that the por ous stone assembly does not fit too tightly into the chamber which might cause cracki ng and eventually break ing upon pushing with the internal piston. Figure 4.6 Modification of Porous Stone Original porous stone Porous stone after grooving Porous stone with o-ring Figure 4.6 Modification of Porous Stone Original porous stone Porous stone after grooving Porous stone with o-ring

PAGE 112

93 CHAPTER FIVE HYDRAULIC CHARACTERIZATION OF BENTONITE In this chapter, characterization of bent onite in terms of its hydraulic conductivity is presented. Two types of permeameters, namely flexible wall and rigid wall, with various inorganic chemical permeants under various hydraulic gradients and prehydration conditions have been used in this study. Chemical analysis of effluent following permeation through bentonite is also reported in this chapter. 5.1 Hydraulic Conductivity of Bentonite Deionized water, tap water, and synthe tic inorganic salt so lutions of various concentrations and combinations have been us ed as permeants for bentonite clay in this investigation. Various chemical permeants, pe rmeameter types, and the effects of various factors controlling conductivity ar e discussed in this section. Fifty grams of air-dried bentonite samples, 7.5 mm thick and 10.16 cm (4 -in) in diameter were used in most of the hydraulic conductivity experiments conducted in the flexible wall permeameter. The corresponding dry density is 0.83 g/cm3. A different setup was used for rigid wall permeameters, as discussed later in this chapter.

PAGE 113

94 5.1.1 Inorganic Chemical Permeants Deionized water with less than 5 ppm of impurities, tap water with 200 to 300 ppm of ionic concentration, and four di fferent salt solutions (NaCl, KCl, CaCl2 and MgCl2) of various concentrations and combinations as shown in Table 5.1, were used in hydraulic conductivity tests as pe rmeants through bentonite clay specimens. All the salts are Fisher Scientific Lab certified brands a nd have been used according to their formula weights for preparing synt hetic inorganic solutions. NaCl, KCl, and CaCl2 are in the form of anhydrous granular salt while MgCl2 is a hexahydrate crystalline salt having the chemical formula of MgCl2.6H2O. Deionized water, commercially available in plastic one-gallon bottles was used as a solvent for those salt solutions. The salt solutions have been chosen to investigate the effects of various concentrations, cation size, valence, and ionic strength. Deionized (D I) water is used as the reference and controlling solution, in a ddition to initial pore fluid saturation. Concentrations of the electrolyte solutions were varied from 0.1M to 5M and were prepared by dissolving crystallin e/granular salts with DI wa ter. NaCl and KCl were chosen to investigate the effects of monovalent cations and hydrated ion size (Na+ and K+ have different hydrated radius) while CaCl2 and MgCl2 were chosen to investigate the effect of divalent cations (Ca2+ and Mg2+) that are commonly found in natural aqueous systems. Sufficient quantities of solutions were prepared in order to last for the whole period of conductivity experiments so that the uniformity of the solutions can be maintained. The synthetic solutions were tran sferred to the largest chamber of the cells after being prepared in the lab at normal room temperature (21o ~ 22oC). 5.1.2 Flexible Wall Permeability Permeability tests have been performed according to ASTM standard (D-5084) for flexible wall permeameters. Since the effl uent (tailwater) level increases with time

PAGE 114

95 during the tests, falling head assumptions w ith increasing tailwater pressure calculations have been adopted for the calculation of the coefficient of permeability. The modified formula used in the calculati on is given in equation (5.1). final w out in final out in initial w out in initial out in out in sample sample out inp p h h p p h h In a a t A L a a k (5.1) where a is the area of the apparatus pipette, Lsample is the length of the sample, Asample is the cross sectional area of the sample, t is the time between initial and final readings, h is the water elevation in the pipette, p is the pressure in the pipette, and subscripts in and out denote inflow and outflow, respectively. A schematic diagram has been shown in figure 5.1 to describe all the terms of equation (5.1). Figure 5.1 Schematic Diagram of Flexible Wall Permeameter Set-up Influent Chamber Permeability Cell Cell water Pressure LinesEffluent Chamber Influent Effluent PinPout Lsample Asamplehinhoutaout ain Datum Influent Chamber Permeability Cell Cell water Pressure LinesEffluent Chamber Influent Effluent PinPout Lsample Asamplehinhoutaout ain Datum

PAGE 115

96 The detailed description of the test apparatus has been given in Chapter 4. Three stainless steel ball valves are connected w ith the bottom permeameter plate, which are used to (i) fill and drain the chamber thr ough the base, (ii) supply influent from the influent cell to the specimen through the base platen, and (iii) discharge effluent from the specimen to the effluent cell through the top platen. An opening at the top of the permeameter plate is connected through the -inch (OD) nylon tubing to the pressure panel through which the regulated air pressure is applied to the cell water. Figure 5.2 shows the connections of all the tubings with various components of influent chamber, permeability cell, and effluent chamber with the pressure panel. Figure 5.2 Flexible Wall Permeameter Influent Chamber Effluent Chamber Clay specimen Base connections Permeameter Cell Panel Pressure Burette Pressure Panel Pressure Regulator Pressure Indicator Influent Chamber Effluent Chamber Clay specimen Base connections Permeameter Cell Panel Pressure Burette Pressure Panel Pressure Regulator Pressure Indicator

PAGE 116

97 5.1.2.1 Test Procedure The test procedure described in this section includes sample preparation, sample saturation, permeation phase, and termination cr iteria. Sample preparation is one of the most critical steps in the flexible wall pe rmeability experiments since proper preparation is crucial in minimizing experimental errors from leakage, sample loss, and disturbance. 5.1.2.2 Sample Preparation All the components of the permeameter and supporting devices are shown in figure 5.3. A vacuum pump was used duri ng sample preparation for fitting the rubber membrane into the split mold so that the mo ld can be easily placed on the bottom platen of the permeameter, and to secure the bottom porous stone in place while placing the powdered sample. Figure 5.3 Components of Flexible Wall Permeameter Plexiglass Cylinder O-ring Expander O-ring for Platen Porous Stone Permeameter Cap Compaction Device Top Platen Bottom Platen Permeameter Stop Watch Rubber Membrane O-ring on Cap Groove Split Mold Plexiglass Cylinder O-ring Expander O-ring for Platen Porous Stone Permeameter Cap Compaction Device Top Platen Bottom Platen Permeameter Stop Watch Rubber Membrane O-ring on Cap Groove Split Mold

PAGE 117

98 The following steps are carried out in se quence during the preparation of clay specimens in the flexible wall permeameter. Step 1: The groove of the base plate of permeameter is cleaned thoroughly with a brush and cloth/tissue paper. The O-ring, lubric ated with silicon grease, is then fitted to prevent base leakage. A porous stone and filter paper (Fisher Scientific brand) are placed on the bottom platen of the permeameter. Step 2: A leak free latex membrane is wrapped over the split mold and manually stretched to fit the inner side of th e mold in an unwrinkled fashion. Step 3: A vacuum pump is connected w ith the mold, and suction is applied to remove air from the space between membrane and mold so that the membrane can follow the inner shape of the mold. The split mold wrapped with the membrane is then placed over the bottom platen while th e vacuum pump is kept on. Step 4: An accurately measured amount ( 50 g) of air-dried bentonite powder is spread over the filter paper in the mold and is lightly compacted uniformly using a specially designed compaction rod. A spoon can also be used to lightly fill-up any voids left along the perimeter. Step 5: The top filter paper and porou s stone are placed on the bentonite powder carefully to prevent disturban ce or loss of clay powder. Step 6: The top platen is placed on the porous stone and the vacuum pump is then disconnected. The rubber membrane is unrolled from both ends of the split mold and is extended over both top and bottom platens. The mold is then split open and is removed carefully. Step 7: An O-ring expander whose diameter is at least -in larger than that of the specimen is used to place one or two O -rings on the grooves at both top and bottom platens in order to prevent any leakage from the cell into the specimen and to keep the membrane in place. Step 8: The top platen is connected tigh tly with the outlet t ubings to prevent cell water from seeping through. The plexiglass cylinder is then placed on the O-ring fitted on the chamber base. The top cover of the ch amber is tightened to seal the cell while pressure is applied to the chamber during permeability testing.

PAGE 118

99 Step 9: Tap water is filled into the permeameter cell through the bottom opening slowly to minimize disturbance to the newly prepared sample. Water is filled up to a level of 1-2 inches be low the top cover. After completion of sample preparation and water filling, the permeameter is shifted to the pressure panel where the in fluent and effluent chambers are already assembled. The permeameter is then connected to both chambers through the base tubings as shown in figure 5.2. All th e above steps are shown in figure 5.4. (a) Sample Preparation (Step 1) (b) Sample Preparation – (Step 2) Figure 5.4 Sample Preparation fo r Flexible Wall Permeability Test

PAGE 119

100 (c) Sample Preparation – (Step 3) (d) Sample Preparation – (Step 4) (e) Sample Preparation – (Step 5) Figure 5.4 Continued

PAGE 120

101 (f) Sample Preparation – (Step 6) (g) Sample Preparation – (Step 7) (i) Sample Preparation – (Step 9) (h) Sample Preparation – (Step 8) Figure 5.4 Continued

PAGE 121

102 5.1.2.3 Sample Saturation Back pressure saturation is used as outlined in ASTM D5084 to ensure full saturation. Both influent and effluent chambers are filled up to the same level so that the same pressure head can be maintained during saturation of the specimen. The permeameter cell is also fille d with tap water through which the confining pressure is applied to the specimen. All the pressure regulators connected with the chambers and permeameter cell, are turned to zero pressure before opening any controlling valves of the chambers and cell. All the vent valves are closed and the valves that control flow through the specimen are opened so that the wate r can flow into the specimen and the air bubbles can be flushed out of the specimen. Air pressure is first applied to the permeameter cell so that the cell water can deve lop an initial confini ng pressure of about 20 to 35 kPa (3 to 5 psi). Air bubbles will s queeze out of the specimen to the connecting chambers due to this initial confining pre ssure. Influent and effluent chambers are connected through a bridge on the pressure panel so that the same pressure is applied to both chambers through a single regu lator. Air pressure is app lied to the chambers as well as to the permeameter cell gradually so as to maintain a pressure difference of 20 to 35 kPa (3 to 5 psi) between the confining pr essure and chamber pressures at all times. It is imperative that the pressure head in both influent and effluent chambers are the same during saturation, so that no flow o ccurs from one chamber to the other. To achieve this, the same air pressure regulator is used to apply pressure to both chambers through a bridge connection in the pressure pa nel as mentioned earlier. The pressure in the cells is raised gradually up to 415 kPa (60 psi) and is kept 20 to 35 kPa (3 to 5 psi) below that of the permeameter chamber during the whole period of saturation. A similar backpressure was used by Boynton and Daniel (1985) in flexible wall permeability tests. Backpressure applied to the specimen through the top and bottom platens, is not allowed to surpass the confining cham ber pressure to prevent bulgi ng of the specimen. This process of saturation by backpressure was con tinued for at least 5-7 days to completely dissolve air or gaseous substances from the te st specimen and to complete any chemical

PAGE 122

103 primary reaction that might occur with wa ter and chemical compounds present on the surface of the bentonite clay materials. It is worth noting that removal of air bubbl es and gaseous substances from the test specimens by flowing water from one end to the other is not advisable because various dissolved substances will leach out of the speci men with the water flow during saturation. Maximum saturation of the specimen is considered to be accomplished when no significant drop of influent and effluent water levels was obs erved over a period of 2 to 3 hours. 5.1.2.4 Permeation Phase Selection of the appropri ate hydraulic gradient is of great importance in determining the suitability of flow rate so that no cracks or channels develop through the specimen during the experiment. Variation of the measured coefficient of hydraulic conductivity with hydraulic gradients in excess of 100 was found to be insignificant for clay by Shackelford et al. (2000). Hydraulic gradients in the range of 25 to 100 had also been used by many researcher s (Boynton and Daniel, 1985; Jo et al ., 2001; Stern and Shackelford, 1998; Shackelford and Redmond, 1995; Lin and Benson, 2000;) for clay and sand mix samples. Higher hydraulic gradie nts in the range of 100 to 600 have also been used for measuring hydraulic conductivity of the clay compone nt of geosynthetic liners (Day and Daniel, 1985; Fernandez a nd Quigley, 1985; Shackelford, 1994; Petrov et a l., 1997; Petrov and Rowe, 1997; Ruhl a nd Daniel, 1997; Daniel, 1993). Rad et al ., (1994) used hydraulic gradients as high as 2800 and found that the hydraulic conductivity of GCL was not affected when water is used as permeant. In order to investigate the effects of hydraulic conductivity on bentonite samples, various hydraulic gradients from 250 to 3500 were used in this research st udy with various inor ganic solutions as permeants. The results are presented later. The effluent cell pressure was reduced fr om its applied backpressure of 415 kPa (60 psi) during saturation to the required pressure level to develop the pre-calculated

PAGE 123

104 hydraulic gradients. Since the hydraulic gradient also depends on the difference in levels between influent and effluent in addition to the applied regulated pr essure, the fluctuation of the actual hydraulic gradient has been account ed for in the calculations. The constant pressure difference of 20 kPa (3.0 psi) between influent and permeameter cell was maintained until the completion of the experiment in order to avoid piping and specimen collapse. 5.1.2.5 Termination Criteria Since the test set-up represents falling head conditions with increasing tailwater elevation, Eq. 5.5 is again used here to calculate the coefficient of permeability, k Permeation through the bentonite specimen was continued until at leas t four consecutive values of hydraulic conductivity were obtaine d over an interval of time (24-hour) in which: (1) the ratio of outflow to inflow ra te is between 0.75 and 1.25, and (2) the values of hydraulic conductivity (coefficient of perm eability) are steady within 50% of each other. Besides the above standard criteria for termination of permeation in permeability tests, electrical conduc tivity of the effluent was measur ed until its value exceeded 90% of the influent before termination. This is one way of ensuring that a chemical steady state has been reached before the test is stopped. The coefficient of permeability, and the pH and electrical conductivity of the effluent were plotted with respect to the pore volume of the specimen. A pore volume is defined as the volume of void space in the permeated medium. In these experiments, it represents the total volume of voids available for the leachate to flow through the specimen. The pore volume, Vp, is calculated from the following Equation: w s c pW w V (5.2) where wc is the sample’s water content at the end of the test, Ws is the weight of the dry sample, and w is the unit weight of water. In order to allow interactions between

PAGE 124

105 permeant and porous medium to take place, at least one pore volume should be allowed to pass through (Joshi et al ., 1994). Termination of the permeability test was carried out by gradually and simultaneously lowering the applied air pr essure in the influent, effluent and permeameter cell to zero. Cell pressure was always kept higher than chamber pressure so that the confining rubber membrane does not separate from the specimen surface. Water was then forced out of the cell by slightly increasing the confining pressure through the base opening valve. After achieving the termination criteria, the specimens were carefully taken out of the permeameter in such a way so as to pres erve their structural integrity for further determination of void ratio, physical shape, and size. 5.1.3 Rigid Wall Permeability Rigid wall permeability experiments were carried out using the diffusion cells described in Chapter 4. The sample prepar ation was the same as for the diffusion tests, which are described in Chapter 6. Like flexible wall permeability, the effluent (tailwater) level increases with time during the tests and the modified ASTM equation is used in the calculation (equation 5.1). A sc hematic diagram is shown in figure 5.5 to describe all the terms of equation (5.1). 5.1.3.1 Sample Preparation Sample preparation steps are the same as those of diffusion tests as described in Chapter 6. After accomplishing a pre-determined thickness of the clay sample due to consolidation, the water from the effluent t ube and influent permeameter cell is replaced with deionized water or required permeants. The piston inside the permeameter used during consolidation was kept dur ing the permeability test in order to maintain a constant

PAGE 125

106 Figure 5.5 Schematic Diagram of Rigid Wall Permeameter Set-up thickness of the sample. To prevent leakage of influent and compressed air from the bottom and top of the permeameter cell respectiv ely, a sufficient amount of silicon grease was applied on both ends of the plexiglass cylin der of the permeameter cell as well as on the protruded portions of the O-rings placed on both end plates. 5.1.3.2 Permeation Phase After completion of sample preparation, the cell influent and pipette tube are filled, and the permeameter is shifted to the pressure panel where the influent is subjected to compressed air pressure supplied from the pre ssure panel. The effluent pipette tube is left open under atmospheric pressure (gauge pressure, Pout = 0 ). Lower hydraulic gradients (compared to flexible wall), in th e range of 200 to 500, are applied so that the Rigid Wall Permeability Cell Influent Pressure Line Effluent PinPout=0 LsampleAsamplehoutaout Datum hinain Rigid Wall Permeability Cell Influent Pressure Line Effluent PinPout=0 LsampleAsamplehoutaout Datum hinain

PAGE 126

107 effluent can be collected w ithout spilling from the pipette within a reasonable time interval. Termination criteria for the rigid wall permeability are the same as those for flexible wall, except that the test needs to be interru pted should it be necessary to replenish the influent cell before chemical and steady-state flow are achieve d. The effluent sample is collected from the pipette at any desired time interval for further analysis for chemical equilibrium in terms of electrical conductivity and pH. The collected sample needs to be diluted for chemical analysis if the amount of effluent collected is in sufficient due to slow permeation through bentonite samples. 5.1.4 Factors Affecting Hydraulic Conductivity Hydraulic conductivity of s odium montmorillonite ha s been found to be the lowest among most of the clays followed by calcium montmorillonite or other divalent cation montmorillonites as investigated by many researchers (Benson et al ., 1994; Mitchell, 1976; Lambe, 1953). The various factor s that are investigated in this research work are permeant chemical composition, voi d ratio, hydraulic grad ient, first wetting liquid, and boundary conditions. 5.1.4.1 Permeant Chemical Composition The permeant chemical solution is one of the most important factors affecting permeability of bentonite due to its interaction with the negatively charged clay mineral surfaces, which is responsible for the diffuse doubl e layer variation of be ntonite particles. Table 5.1 shows the synthetic ch emical solutions with their mo larity used in the hydraulic conductivity experiments using flexible wall permeameters. Tests K-1 to K-8 were carried out using single salt solutions while K-9 to K-14 were conducted using multiple solutions in a sequential permeation fashion. Permeability test results for single salt

PAGE 127

108 solutions are shown in figure 5.6 and 5.7 as a function of duration and pore volume, respectively. Table 5.1 Chemical Solutions Used in Hydraulic Conductivity Using Flexible Wall Permeameter Test number Pre-hydration 1st solution 2nd solution 3rd solution K-1 DI water 1M CaCl2 K-2 DI water 1M MgCl2 K-3 DI water 1M KCl K-4 DI water 1M NaCl K-5 DI water All salts (1M each) K-6 DI water All salts (0.1M each) K-7 DI water All salts (0.01M each) K-8 DI water 5M CaCl2 K-9 1M CaCl2 1M CaCl2 1M NaCl K-10 1M NaCl 1M NaCl 1M CaCl2 K-11 DI water DI water All salts (0.01M each) All salts (0.1M each) K-12 1M MgCl2 1M MgCl2 1M KCl K-13 1M KCl 1M KCl 1M MgCl2 K-14 DI water 1M CaCl2 1M MgCl2 Note: All salts means NaCl, KCl, CaCl2 and MgCl2 K-11 was also tested for all salts of 1M each solutions following 0.1M solution.

PAGE 128

109 Figure 5.6 Permeability vs. Duration for 1M Salt Solutions Using Flexible Wall Permeameter Figure 5.7 Permeability vs. Pore Volume for 1M Salt Solutions Using Flexible Wall Permeameter 1.00E-10 1.00E-09 1.00E-08 051015202530354045Duration (day)Coefficient of permeability, k (cm/s) DI water pre-saturation 1M CaCl2 permeant (K-1) 1M MgCl2 permeant (K-2) 1M KCl permeant (K-3) 1M NaCl permeant (K-4) Hydraulic Gradient = 500 Bentonite Specimen Thickness = 3/10 in, dia = 4 in. Air-Dry weight = 50 g 1.00E-10 1.00E-09 1.00E-08 0.001.002.003.004.005.006.007.00Pore volumeCoefficient of permeability, k (cm/s) DI water pre-saturation 1M CaCl2 permeant (K-1) 1M MgCl2 permeant (K-2) 1M KCl permeant (K-3) 1M NaCl permeant (K-4) Hydraulic Gradient = 500 Bentonite Specimen Thickness = 3/10 in, dia = 4 in. Air-Dry weight = 50 g

PAGE 129

110 It can be seen from figures 5.6 and 5. 7, that the variation of steady-state coefficients of permeability (from 2.4x10-9 cm/s to 3.5x10-9 cm/s) is not that significant among those four different salt solutions. Ho wever, an increase in permeability can be observed initially for CaCl2, MgCl2 and KCl permeants, which could be the results of initial exchange of bentonite surface exchangeable cations. It can also be found that the final stable values of coefficient of permeability are achieved after at most two pore volumes of permeants of all four types of salt solutions. Permeability test results for experiments K-5, K-6, and K-7 are shown in figures 5.8 and 5.9 with respect to duration and pore volume, respectively. Test K-7, where 0.01 molar of each salt was used as permeant, s hows the minimum coefficient of permeability (k = 1.0x10-9 cm/s). The variation of permeability with respect to molar concentration of all salts (K-5, K-6, and K-7) permeants is s hown in figure 5.8 where a trend of increasing permeability is observed with increasing molarity of the permeants. Figure 5.8 Permeability vs. Duration for All Salt Solutions (K-5, K-6, & K-7) Regardless of applied hydraulic gradient, the final permeability of bentonite clay using various combinations of salt solutions can be achieved after at most two pore 1.00E-10 1.00E-09 1.00E-08 0510152025303540Duration (day)Coefficient of permeability, k (cm/s) DI water pre-saturation all salts (1M each) (K-5) all salts (0.1M each) (K-6) all salts (0.01M each) (K-7) Thickness = 0.31 in, dia = 4 in. Bentonite air-dry weight = 50 g

PAGE 130

111 volumes of permeant through the specimens, as shown in figure 5.9. No further variation of permeability is observed until te rmination at around 9 pore volumes, which lasted for 35 days (figure 5.8) Figure 5.9 Permeability vs. Pore Volume for All Salt Solutions (K-5, K-6, & K-7) Figure 5.10 Variation of Permeability with Molarity of Combined Salt Solutions 1.00E-10 1.00E-09 1.00E-08 0.01.02.03.04.05.06.07.08.09.010.0Pore volumeCoefficient of permeability, k (cm/s) DI water pre-saturation all salts (1M each) (K-5) all salts (0.1M each) (K-6) all salts (0.01M each) (K-7) Thickness = 0.31 in, dia = 4 in. Bentonite air-dry weight = 50 g 1.0E-10 1.0E-09 1.0E-08 0.0010.010.1110Concentration of solution (M)Coefficient of permeability, k (cm/s) all salt permeants (K-5, K-6, K-7)

PAGE 131

112 5.1.4.2 Void Ratio In order to investigate the effect of void ratio of bentonite specimens on permeability using various synthetic chemical solutions, rigid wall permeability tests were carried out with various amounts of air-dry bentonite clay. Tabl e 5.2 lists the tests that were conducted in the laboratory in orde r to identify the e ffects of void ratio on CaCl2 and NaCl salt solution permeants. Table 5.2 Rigid Wall Permeability Tests with Void Ratio Variation Permeant chemical solution Test No. Sample Size (diameter x thickness) Air-dry sample weight (g) Void ratio (e) Coefficient of permeability (k, cm/s) KD-6 76.2 mm x 7 mm 30 1.98 1.26x10-9 KD-7 76.2 mm x 7 mm 15 4.97 5.84x10-9 1M CaCl2 KD-8 76.2 mm x 3 mm 2.5 14.35 6.44x10-8 KD-5 76.2 mm x 7 mm 30 1.98 6.0x10-10 KD-9 76.2 mm x 7.84 mm 15 5.69 1.82x10-9 1M NaCl KD-10 76.2 mm x 7 mm 7.5 10.94 1.55x10-8 Experimental test results of KD-6, KD-7, and KD-8 for 1M CaCl2 permeant are presented in figures 5.11 and 5.12 in terms of duration and pore volume, respectively, while those of KD-5, KD-9, and KD-10 for 1M NaCl permeant are presented in figures 5.13 and 5.14. It can be seen from figure 5. 11 that the permeability increases slightly as the test proceeds for 1M CaCl2 permeant, which could be due to the cationic exchange process that results in aggregation of clay pa rticles in higher void ratio specimens. No increase in permeability can be observed fo r more compacted (i.e. lower void ratio) specimen shown in figure 5.12.

PAGE 132

113 Figure 5.11 Variation of Permeab ility with Duration of 1M CaCl2 Permeant Used in Bentonite of Various Void Ratios Figure 5.12 Variation of Permeability with Pore Volume of 1M CaCl2 Permeant Used in Bentonite of Various Void Ratios 1.00E-10 1.00E-09 1.00E-08 1.00E-07 051015202530Duration (day)Coefficient of permeability, k (cm/s) DI water pre-saturation 1M CaCl2, e = 1.98 (KD-6) 1M CaCl2, e = 4.96 (KD-7) 1M CaCl2, e = 14.35 (KD-8) Rigid Wall Permeability 1M CaCl2 Permeant 1.00E-10 1.00E-09 1.00E-08 1.00E-07 0.02.04.06.08.010.012.014.016.018.020.0Pore volumeCoefficient of permeability, k (cm/s) DI water pre-saturation 1M CaCl2, e = 1.98 (KD-6) 1M CaCl2, e = 4.96 (KD-7) 1M CaCl2, e = 14.35 (KD-8) Rigid Wall Permeability 1M CaCl2 Permeant

PAGE 133

114 Figure 5.13 Variation of Permeability with Duration of 1M NaCl Permeant Used in Bentonite of Various Void Ratios Figure 5.14 Variation of Permeability w ith Pore Volume of 1M NaCl Permeant Used in Bentonite of Various Void Ratios 1.00E-10 1.00E-09 1.00E-08 1.00E-07 051015202530Duration (day)Coefficient of permeability, k (cm/s) DI water pre-saturation 1M NaCl, e = 1.98 (KD-5) 1M NaCl, e = 5.68 (KD-9) 1M NaCl, e = 10.94 (KD-10) Rigid Wall Permeability 1M NaCl Permeant 1.00E-10 1.00E-09 1.00E-08 1.00E-07 0.02.04.06.08.010.012.014.016.018.020.0Pore volumeCoefficient of permeability, k (cm/s) DI water pre-saturation 1M NaCl, e = 1.98 (KD-5) 1M NaCl, e = 5.68 (KD-9) 1M NaCl, e = 10.94 Rigid Wall Permeability 1M NaCl Permeant

PAGE 134

115 Figure 5.15 Variation of Permeability with Void Ratio for 1M CaCl2 and 1M NaCl Permeants It is clearly revealed from figure 5. 15 that the permeability of bentonite clay increases with increasing void ratios with both permeants of one molar CaCl2 and NaCl salt solutions. The permeability is higher at hi gher void ratio simply due to the fact that the amount of higher pore volume and space of flow exist in higher void ratio specimens. As divalent cations like calcium Ca2+ replace monovalent negatively charged ions like sodium Na+, potassium K+, and others during flow, th e diffuse double layer thickness between clay platelets is reduced. Shrinkage of the clay specimens occurs, which causes higher flow of solution during perm eability as shown in figure 5.15. 5.1.4.3 Hydraulic Gradient A combination of salt solutions was used in flexible wall permeability tests (K-5, K-6, and K-7) in order to find the effect s of hydraulic gradient on permeability of chemical solution permeants through bentonite clay materials. Hydraulic gradients from 1.0E-10 1.0E-09 1.0E-08 1.0E-07 0246810121416Void ratio (e)Coefficient of permeability, k (cm/s) 1M CaCl2 (KD-6, KD-7, KD-8) 1M NaCl (KD-5, KD-9, KD-10)

PAGE 135

116 450 to 3115 were applied for this investigati on. Table 5.3 shows the coefficient of permeability with respect to the applied hydrau lic gradient on various test specimens. Table 5.3 Flexible Wall Permeability Tests with Hydraulic Gradient Variation Sample No. & permeant Hydraulic gradient (i) Coefficient of permeability (k, cm/s) 450 2.52x10-9 1320 2.27x10-9 K-5 (all salts, 1M each) 2190 2.22x10-9 445 2.27x10-9 1310 1.98x10-9 K-6 (all salts, 0.1M each) 2165 1.76x10-9 455 1.21x10-9 1335 1.09x10-9 2230 9.93x10-10 K-7 (all salts, 0.01M each) 3115 1.01x10-9 It can be seen from the permeability result s listed in Table 5.3 that the coefficient of permeability decreases slightly within th e range of 10% to 25% due to 5 to 6 fold increase in hydraulic gradient. The reducti on in permeability could be attributed to the effect of clay consolidation at higher seepage forces during permeability experiments. However, the higher applied hydraulic gradie nt would cause the permeant to flow at a much higher rate, which expedites chemical equilibrium. Chemical equilibrium is required in order to obtain a stable and constant value of k for low permeability clays. Although ASTM D 5084 recommends a maximu m gradient of 30 for fine grained soils of low hydraulic co nductivity (k less than 10-7 cm/s), higher hydraulic gradient in the range of 50 to 600 are commonly used fo r measuring hydraulic conductivity of the clay component of geosynthetic liners (Dan iel, 1994; Pertrov and Rowe, 1997; Petrov and Rowe, 1997; Petrov et al ., 1997; Ruhl and Daniel, 1997; Lin, 1998). A high

PAGE 136

117 hydraulic gradient of 2800 was used by Rad et al ., (1994), for GCL clays using tap water permeant where the k value was found to be inse nsitive to variations in applied gradients. Figure 5.16 Variation of Permeability with Duration of Flow for K-5 Figure 5.17 Variation of Permeability with Pore Volume of Flow for K-5 1.0E-10 1.0E-09 1.0E-08 05101520253035Duration (day)Coefficient of permeability, k (cm/s) DI water pre-saturation i = 450 i = 1300 i = 2200 Test No. K-5 (Flexible wall permeameter) (1M NaCl + 1M CaCl2 + 1M MgCl2 + 1M KCl) Permeant Thickness = 0.31 in, dia = 4 in. i = 450 i = 1300 i = 2200 1.0E-10 1.0E-09 1.0E-08 0.01.02.03.04.05.06.07.08.09.010.0Pore volumeCoefficient of permeability, k (cm/s) DI water pre-saturation i = 450 i = 1300 i = 2200 i = 450 i = 1300 i = 2200 Sample No. K-5 (1M NaCl + 1M CaCl2 + 1M MgCl2 + 1M KCl) Permeant

PAGE 137

118 Results of the permeability experiments using flexible wall permeameter for K-5, K-6, and K-7 are given in figures 5.16 to 5.21. DI water was used during pre-hydration of the test specimens by applying back pressure. Figure 5.18 Variation of Permeability with Duration of Flow for K-6 Figure 5.19 Variation of Permeability with Pore Volume of Flow for K-6 1.0E-10 1.0E-09 1.0E-08 0.01.02.03.04.05.06.07.08.0Pore volumeCoefficient of permeability, k (cm/s) i = 450 i = 1300 i = 2200 DI water pre-saturation i = 45 0 i = 130 0 i = 2200Sample No. K-6 (0.1M NaCl + 0.1M CaCl2 +0.1M MgCl2 + 0.1M KCl) Influent Solution 1.0E-10 1.0E-09 1.0E-08 05101520253035Duration (day)Coefficient of permeability, k (cm/s) i = 450 i = 1300 i = 2200 DI water pre-saturation Sample No. K-6 (Flexible wall permeameter) (0.1M NaCl + 0.1M CaCl2 +0.1M MgCl2 + 0.1M KCl) Influent Solution Thickness = 0.32 in, dia = 4 in. Aiih0 i = 450 i = 1300 i = 2200

PAGE 138

119 Figure 5.20 Variation of Permeability with Duration of Flow for K-7 Figure 5.21 Variation of Permeability with Pore Volume of Flow for K-7 1.0E-10 1.0E-09 1.0E-08 0510152025303540Duration (day)Coefficient of permeability, k (cm/s) i = 450 i = 1300 i = 2200 i = 3100 DI water pre-saturation Sample No. K-7 (Flexible wall permeameter) (0.01M NaCl + 0.01M KCl + 0.01M CaCl2 + 0.01M MgCl2) Influent Solution Thickness = 0. 31 in, dia = 4.1 in. Air-Dry weight = 50 g i = 450 i = 1300 i = 2200 i = 3100 1.0E-10 1.0E-09 1.0E-08 0.01.02.03.04.05.06.07.08.09.010.0Pore volumeCoefficient of permeability, k (cm/s) i = 450 i = 1300 i = 2200 i = 3100 DI water pre-saturationSample No. K-7 (0.01M NaCl + 0.01M KCl + 0.01M CaCl2 + 0.01M MgCl2) Influent Solutioni = 450 i = 1300 i = 2200 i = 3100

PAGE 139

120 Figure 5.22 Variation of k with Applied Hydr aulic Gradient in Combined Salt Solutions The reduction in permeability is found to be minimal even at an applied hydraulic gradient of around 2000 for the combined salt permeants, as shown in figure 5.22. 5.1.4.4 Pre-hydration Four pairs of tests are compared to find the effects of pre-hydration on the permeability of bentonite clay as listed in Table 5.4. The same salt solutions applied as permeants are also used as hydration liquid a nd the “k” test results are compared with those of deionized (DI) wate r pre-hydrated values. CaCl2 hydrated clay (K-9) is found to develop a higher k value compared to DI water pre-hydrated clay (K-1), as shown in Table 5.4 and figure 5.23. Structured aggregate particles are assumed to be fo rmed during the process of hydration by CaCl2 solution, which contributes to the higher pe rmeability due to shrinkage of the diffuse double layer and higher pore openings of flow. Hydration by other chemical solutions does not seem to be effective in reorganizing and reorienting the clay platelets, rather 1.0E-10 1.0E-09 1.0E-08 0500100015002000250030003500Hydraulic gradient (i)Coefficient of permeability, k (cm/s) all salts, 1M each (K-5) all salts, 0.1M each (K-6) all salts, 0.01M each (K-7)

PAGE 140

121 forming more dispersed structur e of clay particles and thus reducing the flow of chemical solutions. Table 5.4 Flexible Wall Permeability Te sts with Various Hydration Solutions Permeant Sample No. Pre-hydration Coefficient of permeability, k (cm/s) K-1 DI water 3.23x10-9 1M CaCl2 K-9 1M CaCl2 5.93x10-9 K-4 DI water 3.22x10-9 1M NaCl K-10 1M NaCl 1.43x10-9 K-3 DI water 3.55x10-9 1M KCl K-13 1M KCl 1.55x10-9 K-2 DI water 2.30x10-9 1M MgCl2 K-12 1M MgCl2 1.95x10-9 Figure 5.23 Variation of Permeability with Duration of Flow for K-1 & K-9 1.0E-10 1.0E-09 1.0E-08 1.0E-07 05101520253035Duration (day)Coefficient of permeability, k (cm/s) 1M CaCl2 permeant DI water pre-hydration (K-1) 1M CaCl2 pre-hydration (K-9) Flexible wall permeameter Thickness = 3/10 in, dia = 4 in. Air-Dry weight = 50 g

PAGE 141

122 Figure 5.24 Variation of Permeability with Duration of Flow for K-4 & K-10 Permeability test results for 1M NaCl, 1M KCl, and 1M MgCl2 permeants under two different hydration conditions ar e given in figures 5.23 to 5.26 Figure 5.25 Variation of Permeability with Duration of Flow for K-3 & K-13 1.0E-10 1.0E-09 1.0E-08 051015202530354045Duration (day)Coefficient of permeability, k (cm/s) 1M KCl permeant DI water pre-hydration (K-3) 1M KCl pre-hydration (K-13) Flexible wall permeameter Thickness = 3/10 in, dia = 4 in. Air-Dry weight = 50 g 1.0E-10 1.0E-09 1.0E-08 05101520253035Duration (day)Coefficient of permeability, k (cm/s) 1M NaCl permeant DI water pre-hydration (K-4) 1M NaCl pre-hydration (K-10) Flexible wall permeameter Thickness = 3/10 in, dia = 4 in. Air-Dry weight = 50 g

PAGE 142

123 Figure 5.26 Variation of Permeability with Duration of Flow for K-2 & K-12 5.1.4.5 Type of Permeameter Two types of permeameters, namely, flexible wall and rigid wall, are compared in terms of their performance in achieving th e least permeability coefficient values for chemical solution permeation through bentonite clay specimens. Four pairs of tests are compared with the same type of compacted bentonite clay samples (i.e. having same void ratios) under the same pre-hydration conditi ons (using DI water) and back pressure saturation, as listed in Tabl e 5.5. Two divalent (Ca2+ and Mg2+), one monovalent (Na+) cationic salt solutions and DI water were used as permeants in this investigation. It is clearly found from the test results give n in Table 5.5 that the ‘k’ values due to all types of permeants are lower in rigid wall permeameters than in flexible wall permeameters. During the process of pre-hydr ation in rigid wall permeameters, bentonite clay particles tend to expand as a result of osmosis pressure due to adsorption but are restrained due to the boundary surfaces of rigid wall and fixe d porous plates on both ends 1.00E-10 1.00E-09 1.00E-08 051015202530354045Duration (day)Coefficient of permeability, k (cm/s) 1M MgCl2 permeant DI water per-hydration (K-2) 1M MgCl2 pre-hydration (K-12) Flexible wall permeameter Thickness = 3/10 in, dia = 4 in. Air-Dry weight = 50 g

PAGE 143

124 of the specimen. The sides of clay specimen form a seal against the rigid walls and thus prevent development of any side wall leakag e, which is quite prev alent in non-expansive clay soils. Table 5.5 Permeability Tests Using Flexible Wall and Rigid Wall Permeameters Permeant (DI presaturated) Sample No. Permeameter type Coefficient of permeability, k (cm/s) K-11 Flexible wall 7.98x10-10 DI water KD-1 Rigid wall 4.96x10-10 K-1 Flexible wall 3.25x10-9 1M CaCl2 KD-6 Rigid wall 1.26x10-9 K-2 Flexible wall 2.33x10-9 1M MgCl2 KD-4 Rigid wall 7.96x10-10 K-4 Flexible wall 3.22x10-9 1M NaCl KD-5 Rigid wall 5.94x10-10 In rigid wall permeameters, the void ratio and the physical dimensions of the specimens can be maintained constant as the porous plates are restrained at predetermined levels, thereby making the permeability calculation less erroneous. The only two disadvantages associated with rigid wall permeameters are that (1) the influent cannot be replenished with the same permean t or replaced with other permeants during progress of permeability tests without disrupti ng the flow and (2) the influent cannot be collected intermittently for further chemical analysis while the test is in progress. Four pairs of permeability test results as listed in Table 5.5 and are given in figures 5.27 to 5.30.

PAGE 144

125 Figure 5.27 Comparison of Permeameters for DI Water Permeant (K-11 & KD-1) Figure 5.28 Comparison of Permeameters for 1M CaCl2 Permeant (K-1 & KD-6) 1.0E-10 1.0E-09 1.0E-08 0.01.02.03.04.05.06.07.08.09.010.0Pore volumeCoefficient of permeability, k (cm/s) 1M CaCl2 permeant Flexible wall permeameter (K-1) Rigid wall permeameter (KD-6) DI water pre-hydration Compacted void ratio, e = 1.98 1.0E-10 1.0E-09 1.0E-08 0.01.02.03.04.05.06.0Pore volumeCoefficient of permeability, k (cm/s) DI water permeant Flexible wall permeameter (K-11) Rigid wall permeameter (KD-1) DI water pre-hydration Compacted void ratio, e = 1.98

PAGE 145

126 Figure 5.29 Comparison of Permeameters for 1M MgCl2 Permeant (K-2 & KD-4) Figure 5.30 Comparison of Permeameters for 1M NaCl Permeant (K-4 & KD-5) 1.0E-10 1.0E-09 1.0E-08 0.01.02.03.04.05.06.07.08.09.010.0Pore volumeCoefficient of permeability, k (cm/s) 1M MgCl2 permeant Flexible wall permeameter (K-2) Rigid wall permeameter (KD-4) DI water pre-hydration Compacted void ratio, e = 1.98 1.0E-10 1.0E-09 1.0E-08 0.01.02.03.04.05.06.07.08.09.010.0Pore volumeCoefficient of permeability, k (cm/s) 1M NaCl permeant Flexible wall permeameter (K-4) Rigid wall permeameter (KD-5) DI water pre-hydration Compacted void ratio, e = 1.98

PAGE 146

127 5.2 Chemical Analysis of Effluent 5.2.1 General Chemical analysis of effluent afte r permeation in this section includes measurements of pH and electrical conductivit y at regular intervals of permeant flow, their ionic analysis, and further analysis in te rms of solute storage or retention within the bentonite specimens. Chemical equilibrium in terms of electrical conductivity between influent and effluent is also used as one of the criteria for termination of hydraulic conductivity tests after steady-state permeability is achieved. Measurements of pH and electrical conduc tivity (EC) of influent and effluent were carried out in conjunction with flexib le-wall as well as rigid wall permeability tests at 24-hour intervals, or in shorter interval s when the permeability was found to be high. The main goal was to obtain a representative profile of the electrical conductivity of chemicals in the leachate as a function of pore volumes of flow until it reaches its chemical equilibrium. During the collection of permeant, the connect ing ball valves with the buffer cylinders were kept closed wh ile the permeant was diverted through other tubes to minimize disturbance to the flexible wall permeability test. In order to conduct full cationic chemical an alysis of dissolved solids, 2 ml samples of permeant were collected and transported to the envi ronmental engineering lab upon completion of the collection. To prevent a ny further chemical reaction in the permeant liquids, care was taken to ensure that hand ling and transportation time was kept to a minimum. The solution was preserved by adding 1% of nitric acid (HNO3) and then kept in the refrigerator until the chemical analysis was carried out. 5.2.2 pH Measurement Immediately following sample collection, the pH of the non-ac idified original sample was measured using an Accume t portable (model AP63) pH meter and

PAGE 147

128 polymerbody combination pH/ATC Ag/AgCl el ectrode. The pH meter was calibrated at three levels, using three sta ndard color-coded buffer solution of pH 4.00, 7.00 and 10.00. As mentioned earlier, DI water was used for pre-hydration and back pressure saturation in the flexible wall permeability tests that are listed in Table 5.6. The figures in Appendix A show the variation of pH of th e effluent solution with pore volumes passed through the bentonite specimen with referen ce to influent pH values. It maybe highlighted that the pH values are found to be slightly higher at the beginning of the experiment, before they gradually reduce to e quilibrium values at steady-state conditions. These steady state values are higher than th e corresponding influent values, except in the case of CaCl2 permeants. In the case of CaCl2, hydroxyl [OH-] ions are retained on the clay surfaces during permeability, and therefore the pH of the effluent is reduced. Table 5.6 Lists of Flexible Wall Permeability Tests with pH Results Test number Source solution Influent pH Effluent pH range Effluent Mean pH K-1 1M CaCl2 7.4 6.81 – 7.74 7.2 K-2 1M MgCl2 6.63 6.27 – 7.59 7.09 K-3 1M KCl 7.1 6.98 – 7.7 7.38 K-4 1M NaCl 7.35 7.28 – 7.92 7.44 K-8 5M CaCl2 8.2 7.25 – 8.25 7.74 5.2.3 Electrical Conductivity The electrical conductivity of leachates was measured for the same specimens using an Accumet (model AB30) 4-ce ll conductivity meter and two epoxy body electrodes of cell constant 1.0 cm-1 and 10.0 cm-1. These electrodes are capable of measuring a wide range of electrical conductivities from 10 to 200,000 microsiemens. Whenever a change of electrodes was re quired to obtain a measurement within a particular range, it was necessary to recalibra te it using its own standard solution. The

PAGE 148

129 figures in Appendix A show the variation of EC of the effluent solution with pore volumes passed through the bentonite specimen w ith reference to influent EC values. From figures A.1 to A.27 (odd numbers), it can be clearly concluded that the EC values reach the influent values at chemical equilibri um after about 3 to 4 pore volumes of flow. Chemical retention in the bentonite specimen can therefore be happe ning within the first 3 to 4 pore volumes of flow until chemical equilibrium is attained. In order to calculate the total chemical retention within the bentonite clay specimen during permeability, it can be assumed that the existing chemical elements of the bentonite clay mineral have been “flushed ” out within the first pore volume of flow and the influent chemical elements start to accumulate then, until chemical equilibrium is reached. Integrating the area in between the in fluent EC line and the best fitted effluent EC line from zero pore volume to the pore volum e at chemical equilibrium (3 to 4 pore volumes), the total retention capacity of the dissolved salt permeant can be calculated. The area under the electrical conductivity (EC) curve ( S/cm. pore volume) represents the total amount of solute permeated through the clay specimen. The area under the effluent EC curve within any inte rval of pore volumes provides the amount of dissolved chemical salts permeated through th e specimen, while that under influent EC represents the amount of chemi cal salts present in the influe nt that flows into the clay specimen during the same interval of pore volumes. The difference in areas is the amount of chemicals retained within the be ntonite clay during pe rmeation of inorganic dissolved chemicals. Since the testing specimens are saturated with deionized water before the chemical solution permeation is carried out, no chemicals are added to test specimens. After saturation of the clay specimens, EC is measured for the deionized water in the effluent cylinder. Any value measured is due to the diffusion of chemicals present within the specimen during the saturation phase. The total amount of outfluxed chemicals during saturation is to be incorporated in the calculation of actual amounts of chemical retained within the specimen during permeability. In order to obtain a distinct variation of effluent EC, permeation through bentonite cl ay is required to be carried out following deionized water pre-hydration and back pressu re saturation. The calculated amount of

PAGE 149

130 retained chemicals can be checked against th e actual amount retained which is obtained by measuring the dry weight of the sp ecimen after completion of the test. An example of the calculation of any part icular salt permeant is shown below. The influent of the test is one molar CaCl2 solution (test # KD-6) with an electrical conductivity of 128,000 S/cm. The best fit curve for the effluent EC is obtained using any statistical analysis software (‘excel’ wo rksheet in this study). A fourth-order polynomial equation is generated for the effl uent EC curve as shown in figure 5.16, which merges with the influent EC line at around 5 pore volumes of flow. The total amount of chemicals present in the effluent, until chemical equilibrium is achieved, is calculated by the area under this curve (area ABCD as in figure 5.31) from a pore volume of zero to a pore volume of five, as gi ven in the following equation (5.3). Area ABDE = 5 0 3 3 4 5 6 5 02500 25288 3 7076 5 2238 68 185 5822 3 0951 0 dx x x x x x x ydx (5.3) = 379,340 S/cm pore volume Figure 5.31 Chemical Retention Measurement for KD-6 y = -0.0951x6 3.5822x5 + 185.68x4 2238.5x3 + 7076.3x2 + 25288x + 2500 R2 = 0.9806 1.0E+03 1.0E+04 1.0E+05 1.0E+06 0.01.02.03.04.05.06.07.08.09.010.0Pore volumeElectrical conductivity ( S/cm) Effluent Influent EC = 130,000 microS/cm A B C D EKD-6 1M CaCl2 permeant

PAGE 150

131 The total amount of chemicals present in the influent during the same flow volume is the area under the influent EC lin e from zero to five pore volumes (area ACDE). Therefore, the amount of chemi cals influxed into the clay specimen is: Area ACDE = 130,000 x 5 = 650,000 S/cm. pore volume The total of chemicals retained within the specimen = Area ACDE – Area ABDE = 650,000 379,340 S/cm. pore volume = 270,660 S/cm. pore volume = 270,660 x 0.66 mg/L x pore volume [since 1 S/cm = 0.66 ppm] = 178,635 mg/L x 21.2 ml [since 1 pore volume for KD-6 = 21.2 ml] = 3,787 mg = 3.78 g The actual increase in mass recorded for th e test specimen after drying in the oven at 105C for 24 hours was found to be 3.2 g wh ich is 15% smaller th an the theoretical value, as calculated above from the EC m easurements of the effluent and influent solutions. Other values in terms of actual and theoretical chemical retention are given in Table 5.7. Table 5.7 Theoretical and Actu al Chemical Retention During Permeability Test # Influent Size Thickness x mass Void ratio Calculated chemical retention (g) Actual chemical retained (g) KD-4 1M MgCl2 7mm x 30g 1.98 2.0 1.48 KD-5 1M NaCl 7mm x 30g 1.98 0.8 0.48 KD-6 1M CaCl2 7mm x 30g 1.98 3.78 3.2 KD-11 1M KCl 7mm x 30g 1.98 1.51 1.18 KD-7 1M CaCl2 7mm x 15g 4.96 3.08 2.1 KD-8 1M CaCl2 3mm x 2.5g 14.35 1.86 1.03 KD-10 1M NaCl 7mm x 7.5g 10.94 1.12 0.93

PAGE 151

132 It can be seen from Table 5.7 that the actual amounts of chemical retention are lower than those calculated th eoretically from the electri cal conductivity plots. These lower values may be attributed to the fact that some loss of clay specimen mass occurs due to the dissolution of chemicals present in the clay during saturation. Also the presence of chemical solution, and the precipita tion left within the test apparatus after the completion of permeability test may contribute to the difference. By comparing the first four tests as list ed in Table 5.7 having the same size, mass, and void ratio, it is found that the amount of chemical retain ed in divalent permeants is higher than that of monovale nt permeants since the highe r valence cations replace the lower valence cations on the su rface of the clay platelets. 5.2.4 Ionic Analysis All influent and effluent samples were co llected in 60 ml polyethylene chemically resistant bottles and mixed with 1% nitr ic acid (0.6 ml) fo r preservation at 4oC in the refrigerator until the actual ch emical analyses were done. The acidification is a required step in the preservation and chemical analysis of the samples, and does not interfere with the accuracy of the measurement in any way. The acidified samples were analyzed for all major metal ions, namely sodium (Na+), calcium (Ca++), magnesium (Mg++), and potassium (K+). This was done using the “AAnalyst – Atomic Absorption Spectrometer” at the environmental engineering lab of the University of South Florida. Liquid samples, which were collected and preserved previously during the hydraulic conductivity tests at different EC values, were analyzed, and the amounts (concentration) of their four major chemical elements were determined. Test results obtained from the permeability tests (Test # K-1, K-2, K-3, a nd K-4) are given in Appendix C. It is found from the plots, in fi gures C.1 to C.4 of Appendix C, that most of the cation exchange happens until around 2 to 3 pore volumes except in sodium solution where no cation exchange is visible, as s hown in figure Appendix C.4 (test # K-4). Sodium and calcium chemical elements ar e found to be present within the supply

PAGE 152

133 bentonite either in the form of precipitation or exchangeable cations on the clay platelets as evidenced from the ionic analysis plots.

PAGE 153

134 CHAPTER SIX DIFFUSION IN BENTONITE In this chapter, test set-up and expe rimental procedures of diffusion through bentonite is described. A ne w test procedure and apparatus for diffusion using parts of a rigid wall permeameter is proposed in order to obtain relevant diffusion properties of bentonite clay material. In this study, a number of inorga nic chemical permeants were used in diffusivity of bentonite at various so lid-water conditions (i.e various void ratios). Solutions are collected from outflux tubes conne cted to the diffusion apparatus at regular intervals of time during the pr ogress of diffusion. Test resu lts of chemical analysis of diffusion solutions are also reported in this chapter. 6.1 Experimental Methods A specially fabricated diffusion cell is us ed for the diffusion experiments, which was also used in rigid wall permeability experiments (described in Chapter 5). Commercially available deioni zed water and synthetic inorga nic salt solutions of various concentrations and combinations were used as permeants for bentonite clay in this investigation. Apparatus set-up, test sample preparati on and procedure, and sample collection for chemical analysis are described in this section. Test results and chemical analysis of diffused collected samples are discussed in sectio n 6.2. In order to find the diffusivity of various chemical elements in bentonite clay, a number of dissolved salts solutions, used

PAGE 154

135 as permeants, were placed in the highly concentrated source chamber. Bentonite specimens with various thicknesses were prep ared to provide different void ratios. Void ratios of the specimens were vari ed by taking different amounts of air dry samples for the same physical dimension of the specimens. The thickness of the specimens was kept constant at 3 mm while the weights of air-dried bentonite powder were varied from 2.5 g to 7.5 g. Highly concentrated dissolved salt solutions of one molar and above were prepared and applied as a single salt or a combination of various salts in the source chamber. 6.1.1 Test Set-up The full description of the test apparatus was given in chapter 4. Diffusion of the highly concentrated solutions through clay was carried out by keeping the liquid levels of both source chamber and receiving tube the same. A schematic diagram in figure 6.1 shows the relevant terms necessary to i nvestigate the diffusion characteristics of bentonite clay materials. The relevant terms are: Lsample = length of the sample Asample = cross-sectional area of the sample as = area of the source chamber which is equal to the cross-sectional area of the sample ar = area of the receptor tube A stainless steel ball valve connected at the bottom of the receptor tube is used to separate the solution in the tube from th at in the source chamber and bottom porous stone. The valve is to be closed while coll ecting the out-fluxed solution from the receptor tube so that no disturbance or hydraulic gradient is created within the diffusion cell. The connection between the receptor an d ball valve is required to be leak proof so that no outfluxed solution is lost. The grooves on the bottom plate of the diffusion cell need to be cleaned periodically from any deposited solu tes by using pressurized tap water and a cleanser. The plexiglass diffusion cylinder, which is placed in between top and bottom

PAGE 155

136 plates, is to be tightened fi rmly with the bottom plate so that diffusion is prevented through the perimeter of the bottom porous stone Sufficient vacuum grease is applied on both ends of the plexiglass cylinder in order to prevent any leakage. Figure 6.1 Schematic Diag ram of Diffusion Cell Set-up Porous stones are assumed to be non-r eactive to the source solutions; however, a small amount of precipitation of dissolve d solutes may occur during the process of diffusion. A the end of each diffusion test, each porous stone is thoroughly cleaned of any deposited chemical solutes using commerc ially available cleanser or diluted acids and flushed with pressurized tap water and th en submersed in DI water for at least 48 hours to remove all residue. The neutrality of the DI water with submersed porous stones is checked by a pH meter before being used in any new set-up of diffusion cells. The porous stones on both sides of the clay specimen are placed in such a way that soft clay slurry does not squeeze out through the joints between the plexiglass and the porous stones while preparing the clay specimen inside the diffusion cell. A sufficient amount of silicon sealant is to be added along the O-ring and circumference of the top and bottom porous stones before being placed inside the chamber. A small Diffusion Cell Source Chamber Pressure Line (closed) Receptor LsampleAsamplear as Porous stones Solution height Diffusion Cell Source Chamber Pressure Line (closed) Receptor LsampleAsamplear as Porous stones Solution height

PAGE 156

137 amount of silicon vacuum grease is also app lied along the perimeter edge of the porous stones to create frictionless joints so that the stones can be pushed into the top of the specimens with ease. No filter paper is placed in between the specimen and porous stones to reduce any reaction which might occu r after a certain period of time between the constituents of the paper, chemical solutions, and clay minerals during the process of diffusion. Choosing the right size O-ring is essential so th at the porous stone assembly does not fit too tightly into the chamber, which might cause it to crack and eventually break while pushing the porous stone the internal piston. A full diffusion set-up picture is given in fi gure 6.2. A highly concentrated solute flows from the source chamber towards th e receptor tube with time due to the concentration gradient. Figure 6.2 Diffusion Set-up Cylindrical piston Receptor Source chamber Control ball valve Porous stone Clay specimen Diffusion Cell Cylindrical piston Receptor Source chamber Control ball valve Porous stone Clay specimen Diffusion Cell

PAGE 157

138 6.1.2 Sample Preparation and Procedure The sample preparation procedure and test sequence for diffusion experiments were followed strictly in order to obtain reproducible and relia ble test results, and minimize experimental errors due to leakag e, sample non-uniformity, sample loss, and disturbance. 6.1.2.1 Sample Preparation A pre-determined amount of air-dried bentoni te powder (2.5 g to 7.5 g) is taken in a plastic bowl of sufficient capacity (0.5 liter to 1.0 liter. DI water is slowly added to the bentonite powder and then mixed thoroughly using a high speed mechanical mixer until a slurry consistency bentonite-wat er suspension is obtained. Th e bentonite-water slurry is then kept in the bowl with a cover for at leas t 24 hours so that the water molecules adsorb uniformly on the clay platelets. Figure 6.3 Components of Diffusion Cell Plastic bowl Piston cylinder Top support Top plate Pipette Mixture machine Receptor Base plate Plexiglasscylinder Porous stone Preservation bottle Rubber pump Collection bottle Plastic bowl Piston cylinder Top support Top plate Pipette Mixture machine Receptor Base plate Plexiglasscylinder Porous stone Preservation bottle Rubber pump Collection bottle

PAGE 158

139 After the 24 hour soaking period, the mixe r is again used to disintegrate any lumped or aggregated clay particles so th at a uniform slurry suspension is achieved before being poured inside the diffusion cell. Constant care is to be taken during the mixing so that no material is lost or left adhering to the parts of the mixer or mixing bowl. All the components of the diffusi on cell and its supporting accessories are shown in figure 6.3. The following steps are carried out in sequence during preparation of clay specimens in the diffusion cell. Step 1: The grooves of the base plate, connecting fittings of the diffusion cell, and receptor tube are cleaned thoroughly w ith a brush, pressurized tap water and cloth/tissue paper so that no deposited salt or other impurities are left behind. Step 2: Sufficient silicon vacuum greas e is applied on the perimeter edges of a porous stone and then positioned inside one of the ends of the plexiglass cylinder, flush with the edge of the cylinder. Step 3: Additional vacuum grease is a pplied on the both edges of the cylinder. The end with the bottom porous stone from step 2 is then placed on the O-ring seated on the based plate of the diffusion cell. Step 4: The top support (ring frame) is placed on the top end of the cylinder and tightened with screws so th at no leakage is allowed through the bottom connection of the cylinder and the plate. Step 5: The prepared bentonite slurry is then poured into the plexiglass cylinder (already fitted with the bottom porous stone) in such a way that no bentonite clay is left on the bowl surface or the spoon. Step 6: After applying vacuum grease al ong the sides of the O-ring placed on its perimeter edge, the top porous stone is caref ully placed on the top side of the cylinder and then pushed into the cylinde r with the help of a smaller diameter cylinder until the porous stone touches the top of the bentonite slurry. The porous stone is pressed down inside the cylinder in such a manner that it re mains horizontal all the way to the top of the slurry surface. Any inclination in placi ng the porous stone would allow the bentonite slurry to squeeze out during the subsequent consolidation process. Erroneous results

PAGE 159

140 would be encountered should there be any gap along the perimeter joint of the porous stone and cylinder. Step 7: After leaving the piston cyli nder inside the diffusion cylinder, the top plate of the diffusion cell is placed on the pist on cylinder. Three wing nuts are then used to push the top plate down with the piston, which eventually presses the top porous stone down and squeezes the slurry bentonite sample. The three nuts are to be turned slowly and uniformly in order to apply a uniform pressure on the porous stone. During this process of consolidation, the receptor ball valve is kept open so as to create a double drainage flow path. Step 8: Water accumulated within the r eceptor tube due to consolidation of the slurry is flushed out. The th ickness of the specimen is calcula ted from the height of the piston when the top plate t ouches the top support ring afte r pressing the piston down by turning the screws. The pistons are fabricated in such a length that produces the required thickness of the bentonite specimen at wh ich the diffusion test is performed. Step 9: The water as well as some susp ension clay particles that are squeezed out through the porous stones from top and botto m of the specimen and are accumulated within cylinder and receptor respectively, ar e collected and dried in an oven overnight. The dried weight of the clay is deducted from the initial amount of the bentonite in order to calculate the final amount of benton ite used in the diffusion experiments. Step 10: A synthetic inorganic salt solu tion is prepared by dissolving the required amount of salt in DI water. The concentrat ed salt solution (about 300 ml to 400 ml) is poured into the source chamber. The top pl ate is then placed on top of the source chamber and is tightened with the wing nut s so that no opening in the connection is available for air-flow. Step 11: DI water is poured into the r eceptor tube up to the same level as the source solution in the chamber. A cap is then placed on top of the receptor tube to prevent any air circulation or evaporation of the receptor solution during the process of diffusion.

PAGE 160

141 Various important steps of the above pr ocedure are shown in figure 6.4. The prepared test assembly with specimen and synthetic solution is then kept in an undisturbed place free of air flow/circula tion or temperature variation. (a) Step 2 (b) Step 3 (c) Step 4 (d) Step 5-8 (e) Step 9-11 Figure 6.4 Sample Preparation for Diffusion Test 6.1.3 Synthetic Inorganic Chemicals Deionized water with less than 5 ppm dissolved ions and four different salt solutions (NaCl, KCl, CaCl2 and MgCl2) of various concentrati ons and combinations as shown in Table 6.1 were used in diffusion test s as the source solution. All the salts are Fisher Scientific Lab certified brands and were used according to their formula weights for preparing synthetic inorganic solutions. NaCl, KCl, and CaCl2 are anhydrous granular salts while MgCl2 is a hexahydrate crystalline sa lt having the chemical formula

PAGE 161

142 MgCl2.6H2O. Deionized water commercially available in one-gallon plastic bottles was used as a solvent for those salt solutions. The salt solutions have been chosen to investigate the effects of various concentrations, cation size, and valence on the diffusion characteristics of bentonite clay. Concentrations of the electrolyte solutions were varied from 1M to 5M and were prepared by dissolving crystalline/ granular salts with DI water. In order to determine the adsorption capacity and replaceabilty of cations on negatively charged clay mineral surfaces, NaCl and KCl were chosen to st udy the effects of monovalent cations and hydrated ion size (Na+ and K+ have different hydrat ed radius) while CaCl2 and MgCl2 were chosen to investigate the be havior of divalent cations (Ca2+ and Mg2+) that are commonly found in natural aqueous systems a nd at higher concentr ations in polluted groundwater and landfill leachate. 6.1.4 Sample Collection for Chemical Analysis Measurements of pH and electrical c onductivity (EC) of outfl uxed diffusant were taken at 48-hour intervals, or shorter inte rvals when the diffusion rate was found to be high. The main goal was to obtain a represen tative profile of the flow of chemicals through the bentonite clay as a function of time. In order to conduct EC and pH measurements, as well as a full cationic chem ical analysis of di ffusant, DI water was added to the receptor solution up to a level of 25 ml. By using a long slender pipette and a handheld rubber suction pump, the diffusant was collected from the receptor tube for chemical analysis which includes pH and EC measurements and ionic analysis. 6.2 Chemical Analysis Chemical analyses in terms of pH, electri cal conductivity (EC) and ionic analysis were carried out on the diffusant solution coll ected from receptor tube. In addition, pH

PAGE 162

143 and EC measurement of source solutions were car ried out intermittently in order to verify the uniformity of influx concentration during th e whole process of diffusion. Table 6.1 lists the diffusion tests carried out with s ynthetic inorganic salt so lutions of different molarities. Void ratios of the specimens we re varied according to their size and the amount of air-dry bentonite in the specimen. Table 6.1 Lists of Diffusion Samples with Source Solutions Test number Source solution Specimen size Diameter x thickness Void ratio D-5 1M NaCl 76.2 mm x 7.84 mm 5.69 D-6 2M CaCl2 76.2 mm x 8 mm 9.23 D-8 2M MgCl2 76.2 mm x 3 mm 14.35 D-9 2M KCl 76.2 mm x 3 mm 14.35 D-10 2M NaCl 76.2 mm x 3 mm 14.35 D-11 2M CaCl2 76.2 mm x 3 mm 14.35 D-12 5M CaCl2 76.2 mm x 3 mm 14.35 D-13 5M CaCl2 76.2 mm x 3 mm 4.11 D-14 5M NaCl 76.2 mm x 3 mm 14.35 D-16 5M NaCl 76.2 mm x 3 mm 6.67 D-17 All salts (1M each) 76.2 mm x 3 mm 14.35 6.2.1 pH Measurement Immediately following sample collection, the pH of the non-ac idified original sample was measured using an Accume t portable (model AP63) pH meter and polymerbody combination pH/ATC Ag/AgCl el ectrode. The pH meter was calibrated at three levels, using three standa rd color-coded buffer solutions of pH 4.00, 7.00 and 10.00. During the measurement, the solution must be st irred constantly with the pH probe for at least a minute in order to obtain a stable readi ng. In each subsequent use of the pH probe,

PAGE 163

144 it is important to wash the probe thoroughl y using DI water in order to prevent contamination with previous ly measured solutions. The test results for the diffusants collected from the receptor tube are given in a series of figures in appendix B. The results are also summarized in Table 6.2, where the range of pH and their mean pH along with th e specimens’ void ratios are highlighted. In order to compare the variation of pH values with respect to source solutions, tests results are grouped into three categories as follows: Group # 1 – Source solution CaCl2 of various molarities (D-6, D-11, D-12 and D-13) Group # 2 – Source solution NaCl of various molarities (D-5, D-10, D-14 and D-16) Group # 3 – Source solution of two molars of various salt solutions for same void ratio (e = 14.25) specimens (D-8, D-9, D-10, and D-11). Combined test results for group 1, 2, and 3 are given in figures 6.5, 6.6, and 6.7 respectively. The values of pH were found to be widely scattered within a range of 5.3 to 10.95, as given in Table 6.2. It can be seen from figure 6. 5 of group # 1 tests, where CaCl2 solutions of various concentrations were used as the source, th at pH values of out -fluxed diffusants were found to be slightly higher than those of group # 2 (figure 6.6), where NaCl solutions were used as a source. It can be hi ghlighted that the pH values of CaCl2 source solutions were found to be between 10.0 and 10.5, while those of NaCl solutions were in the range of 7.1 to 7.5. Therefore, it may be conclude d that the pH value decreases in the case of CaCl2 source solutions due to retention of hydroxyl ions [OH-] within the bentonite clay during diffusion. In the case of NaCl source solutions, an increase in pH values can be observed, which could be due to the supply of hydroxyl ions [OH-] from the bentonite clay during the process of diffusion. It can be observed from figure 6.7 and Ta ble 6.2 that the pH values for KCl and MgCl2 source solutions are lower than those of NaCl and CaCl2 source solutions, which could be due to the fact that or iginal source pH for KCl and MgCl2 are 6.55 and 6.65 respectively, which are lower than those of NaCl and CaCl2.

PAGE 164

145 Table 6.2 Lists of Diffusi on Tests with Out-Fluxed pH Results Test number Source solution Void ratio pH range Mean pH D-5 1M NaCl 5.69 7.33 – 8.86 7.95 D-6 2M CaCl2 9.23 5.85 – 10.95 9.18 D-8 2M MgCl2 14.35 6.32 – 8.87 7.35 D-9 2M KCl 14.35 6.49 – 9.2 7.64 D-10 2M NaCl 14.35 6.6 – 10.67 8.4 D-11 2M CaCl2 14.35 5.3 – 9.74 8.36 D-12 5M CaCl2 14.35 5.74 – 9.8 8.54 D-13 5M CaCl2 4.11 6.62 – 9.35 8.08 D-14 5M NaCl 14.35 7.24 – 8.64 7.98 D-16 5M NaCl 6.67 6.75 – 9.2 8.21 D-17 All salts (1M each) 14.35 7.12 – 8.9 7.68 Figure 6.5 Variation of pH for Group #1 Diffusion Tests (D-6, D-11, D-12, and D-13) 0 2 4 6 8 10 12 0102030405060708090100Cumulative diffusion time (day)pH pH for D-6 (2M CaCl2) pH for D-11 (2M CaCl2) pH for D-12 (5M CaCl2) pH for D-13 (5M CaCl2)Group # 1

PAGE 165

146 Figure 6.6 Variation of pH for Group #2 Diffusion Tests (D-5, D-10, D-14, and D-16) Figure 6.7 Variation of pH for Group #3 Diffusion Tests (D-8, D-9, D-10, and D-11) 0 2 4 6 8 10 12 0102030405060708090100Cumulative diffusion time (day)pH pH for D-5 (1M NaCl) pH for D-10 (2M NaCl) pH for D-14 (5M NaCl) pH for D-16 (5M NaCl)Group # 2 0 2 4 6 8 10 12 0102030405060708090100Cumulative diffusion time (day)pH pH for D-8 (2M MgCl2) pH for D-9 (2M KCl) pH for D-10 (2M NaCl) pH for D-11 (2M CaCl2)Group # 3

PAGE 166

147 6.2.2 Electrical Conductivity The electrical conductivity of leachates was measured for the same specimens using an Accumet (model AB30) 4-ce ll conductivity meter and two epoxy body electrodes of cell constant 1.0 cm-1 and 10.0 cm-1. These electrodes are capable of measuring a wide range of electrical conductivity from 10 to 200,000 microsiemens. Whenever a change of electrodes was re quired to obtain a measurement within a particular range, it was necessary to recalibra te it using its own st andard known solution before using. Test results of electrical conductivity measurements for all the diffusion experiments are listed in Table 6.1 and are presented in appendix B. Cumulative diffusion time in days, shown on the horizon tal axes of the figures in appendix B, represents the elapsed time from the beginning of the diffusion test. As the receptor tube is replenished with DI water after each collection of diffusant solution, electrical conductivity values presented in the “a” series of figures in appendix B measure the EC for the duration between two c onsecutive sample collections. In the “b” series of the figures in appendix B, the cumulative electri cal conductivity values, which are calculated from the raw data of the “a” series, are plotted on the vertical axis. A diffusion test is considered to have reached at steady-stat e condition when the curve of cumulative EC versus cumulative diffusion time starts to ta ke the shape of a st raight line. After achieving a constant variation of cumulative EC with respect to elapsed diffusion time, as shown by the dotted straight lines in figures “b” in appendix B, diffusion tests were terminated and the bentonite clay specimens were collected and dried in the oven for further analysis. Three groups of tests, as outlined in section 6.2.1, were also considered for comparison of out-fluxed cumulative electri cal conductivity with respect to diffusion duration. The results are tabulated in Ta ble 6.3, along with th eir duration intercept known as the “Lag Time”, and their steady-stat e equation. Combined test results of EC for groups 1, 2, and 3 are also given in fi gures 6.8, 6.9, and 6.10 respectively.

PAGE 167

148 Table 6.3 Comparison of Diffusion Te sts with ‘Lag Time’ and Steady-State Equation Group # Test number Source solution Void ratio Lag Time Steady-state equation D-6 2M CaCl2 9.23 39 Y= 382.82X 14913 D-11 2M CaCl2 14.35 25 Y = 847.67X 20905 D-12 5M CaCl2 14.35 8 Y = 1664.1X 13584 1 D-13 5M CaCl2 4.11 54 Y = 756.82X – 40684 D-5 1M NaCl 5.69 16.5 Y = 59X – 984 D-10 2M NaCl 14.35 40 Y = 453.78X – 17907 D-14 5M NaCl 14.35 14 Y = 1116.7X – 15682 2 D-16 5M NaCl 6.67 19 Y = 677.77X – 13165 D-8 2M MgCl2 14.35 45 Y = 566.66X – 25387 D-9 2M KCl 14.35 40 Y = 359.66X – 14468 D-10 2M NaCl 14.35 40 Y = 453.78X – 17907 3 D-11 2M CaCl2 14.35 25 Y = 847.67X 20905 Figure 6.8 Cumulative EC fo r Group #1 Diffusion Tests (D-6, D-11, D-12, and D-13) 0 20,000 40,000 60,000 80,000 100,000 120,000 0102030405060708090100Cumulative diffusion time (day)Cumulative EC ( S/cm) 5M (D-12) e = 14.35 2M (D-11) e = 14.35 5M (D-13) e = 4.11 2M (D-6) e = 9.23Group # 1

PAGE 168

149 Figure 6.9 Cumulative EC fo r Group #2 Diffusion Tests (D-5, D-10, D-14, and D-16) Figure 6.10 Cumulative EC fo r Group #3 Diffusion Tests (D -8, D-9, D-10, and D-11) 0 10,000 20,000 30,000 40,000 50,000 60,000 70,000 0102030405060708090100Cumlative diffusion time (day)Cumulative EC ( S/cm) 1M (D-5) e = 5.69 2M (D-10) e = 14.35 5M (D-14) e = 14.35 5M (D-16) e = 6.67Group # 2 0 10,000 20,000 30,000 40,000 50,000 60,000 0102030405060708090100Cumlative diffusion time (day)Cumulative EC ( S/cm) 2M MgCl2 (D-8) e = 14.35 2M KCl (D-9) e = 14.35 2M NaCl (D-10) e = 14.35 2M CaCl2 (D-11) e = 14.35Group # 3

PAGE 169

150 It can be seen from figure 6.8 that th e fastest diffusion rate and shorter lag time were achieved for test D-12 where the bent onite specimen had higher void ratio (e = 14.35) with a higher concentrated source so lution of 5 molars. For consolidated bentonite clays of lower void ratios, lower rates of diffusion and longer lag times were found, even at higher concentrated sour ce solutions, as shown in figure 6.8. The same trend can also be observed for NaCl source solutions used in diffusion through bentonite clay specimens, as shown in figure 6.9. The rate of diffusion of one molar NaCl solution through bentonite clay of void ratio 5.69 was so slow that it was terminated after 45 days of diffusion. By comparing figures 6.8 and 6.9, it can be concluded that the rate of di ffusion is much faster in CaCl2 solutions than in NaCl solutions. In group # 3 diffusion tests, various salt so lutions of the same molarity (2M) were used as source solutions for bentonite clay sp ecimens of the same void ratio (e = 14.35). The results are given in figure 6. 10. It can be clearly seen from figure 6.10 that the rate of diffusion is much higher and la g time is much shorter for CaCl2 solution in comparison with other source solutions. 6.2.3 Ionic Analysis All diffusant samples were collected in 2 ml polyethylene chemically resistant bottles and mixed with 1% nitric acid for preservation at 4oC in the refrigerator until the actual chemical analyses were done. The acidification is a required step in the preservation and chemical analysis of the samples, and does not interfere with the accuracy of the measurement in any way. The acidified samples were analyzed for all the relevant cations, namely sodium (Na+), calcium (Ca2+), magnesium (Mg2+), and potassium (K+). This was done in the USF environmental lab using the “Optical Emission Spectrometer” which is known to be a highly accurate method for that purpose.

PAGE 170

151 Table 6.4 Ionic Analysis of Diffusant of Two Molar Solutions Through Bentonite Test # Time lapsed (days) Na+ (mg/l) K+ (mg/l) Mg++ (mg/l) Ca++ (mg/l) Total (mg/l) 1.05 9 5.5 3 3.3 20.8 9.09 0.6 0.7 0 0.5 1.8 13.16 3.3 1.1 11.1 1.2 16.7 16.1 5.8 1.1 20.4 1.6 28.9 22.92 3.5 0.5 12.5 0.8 17.3 29.99 2 6.7 49 1.6 59.3 42.15 2.2 0.4 300 2 304.6 53.31 2.3 14.8 351 3 371.1 D-8 2M MgCl2 71.35 4 2.9 608.8 3.8 619.5 1.04 1.1 8.5 4.2 7.1 20.9 2.3 0.3 8.7 3.4 2.7 15.1 16.1 7.2 10 0.3 0.5 18 18.96 2.2 36.5 0.4 1.1 40.2 22.92 1.8 58.4 0.5 0.7 61.4 29.99 1.3 163.4 0.3 0.9 165.9 39.91 1.8 212.3 0.5 1.3 215.9 D-9 2M KCl 53.31 1.8 244.9 1.5 3.3 251.5 1.03 7.2 4.5 2.2 2.2 16.1 2.3 4.6 5 1.6 1.5 12.7 5.14 11 3.2 0.4 0.3 14.9 13.16 45 2.2 2.1 1.6 50.9 16.1 99.3 1.03 0.2 0.3 100.8 22.92 90 1.4 0.2 1 92.6 33.53 108 1 0.3 1.4 110.7 D-10 2M NaCl 78.08 2666 2 0.4 5.8 2674.2 5.31 4.1 3.6 0.5 16.5 24.7 9.38 11 19.4 0.9 0.9 32.2 18.44 7.5 10.4 5.1 108 131 D-11 2M CaCl2 29.27 22.9 45.3 1 643.6 712.8 Solutions, which were collected and pr eserved previously during the diffusion tests at different EC values, were analy zed, and the amounts (concentrations) of their major four chemical elements were determined. Table 6.4 shows the test results of ionic analysis conducted on various diffusants coll ected from receptor tubes. By comparing the diffusants of 2M MgCl2 and 2M CaCl2 in tests # D-8 and D-11, respectively, it is found that calcium divalent cations replace mo re monovalent cations from the negatively charged surface of clay platelets than ma gnesium divalent cations. This may be

PAGE 171

152 attributed to the larger hydr ated radius of magnesium cati ons compared to calcium. Small amounts of divalent cations were detected, even when using monovalent diffusants, as shown in test # D-9 and D-10 in Table 6.4. This may be due to the presence of loose precipitated divalent cations mixed in the bentonite powder. However, no significant traces of cati ons are encountered at the st eady-state condition other than those of the diffusant solutions.

PAGE 172

153 CHAPTER SEVEN TRANSPORT THEORY AND ANALYSIS OF DIFFUSION IN BENTONITE CLAY 7.1 Fluid Transport Mechanisms There are four different types of flow which occur through soils, namely, fluids, electricity, chemicals, and heat flow. Thes e flows occur due to the variation in their respective potentials at various locations. In addition, coupled flow is defined as the flow of one type due to the flow potential of a nother type. Water flow, chemical flow and coupled hydraulic-chemical flow are investigated in this research. Transport of dissolved chemicals or so lutes in the subsurface is generally considered to be the result of three importa nt processes: advection, dispersion, and diffusion. The following sections are designa ted to describe the advection and diffusion flow theories and their related characteristics. 7.1.1 Advection Flow Advection is defined as a movement (flo w) of fluid (or leachate) through a porous medium due to a potential (hydraulic gradient) as shown in figure 7.1. Advection occurs in the pore fluid where the flowing fluid is responsible for carrying chemicals in the form of dissolved or suspended particles.

PAGE 173

154 Figure 7.1 Advection of Solute Transport The mathematical representation of 3-D a dvection is shown in figure 7.2, where dh is the change of hydraulic head across an infinitesimal distance dx, dy or dz The terms q and k with subscripts in their respectiv e directions are known as Darcy flux [LT-1] and hydraulic conductivity or co efficient of permeability [LT-1], respectively. The chemical flux, Ja [MT-1L-2] through a unit area due to a hydraulic influx of a solution of concentration C [ML-3] can be written as (Malusis, 2001; Mitchell, 1993): Cq Ja (7.1) The mass of chemical solute accumulated by advection during any time interval t1 to t2 can be calculated by integrating equation (7.1) as follows: 2 1 2 1) (t t t t a adt q t C dt J M (7.2) In equation (7.2), C ( t ) is the concentration of the chemical during the time interval, and q is the Darcy flux defined as k(dh/dx) Hydraulic head Higher concentrated solution Lower concentrated solution Soil porous medium Hydraulic head Higher concentrated solution Lower concentrated solution Soil porous medium x z y dx dz dy qx= -kx dh dx qy= -ky dh dy qz= -kz dh dz x z y x z y dx dz dy qx= -kx dh dx qx= -kx dh dx qy= -ky dh dy qy= -ky dh dy qz= -kz dh dz qz= -kz dh dzFigure 7.2 Mathematical Representation of Advection

PAGE 174

155 7.1.2 Diffusion Flow Transport of chemicals through a porous medium by dispersion consists of two processes, namely, molecular diffusion (com monly known as diffusion) and mechanical (or hydrodynamic) mixing. Diffusion is de fined as the process whereby ionic or molecular constituents are tran sported under the influence of their kinetic activity in the direction of their con centration gradient as shown by the schematic diagram in figure 7.3. The solute (chemicals) still flows through the porous medium even when the hydraulic gradient is zero, as shown in figure 7.3. Dissolved chemicals flow from the hi gh concentration location to the low concentration location. The amount of mass flux, Jd [MT-1L-2], depends on its chemical concentration gradient. Figure 7.3 shows the variation of concentration gradient (also known as chemical potential gradient) where it changes with time and eventually reaches a constant at steady state condition. No concentration gradient, and accordingly no net solute flow, exists when the concentration on both sides of the medium is the same. Diffusion flux, Jd, as given by the Fick’s first law for steady state condition, is written in equation 7.3 (Mitchell, 1993; Malusis, 2001, 2004; Shackelford, 1993, 1996, 2001). dx dC D Jo d (7.3) Do [L2T-1] is known as the coefficient of diffu sion in “free solution” (normally when the chemical is in infinite dilution). Several investigators have studied the influential factors cont rolling the value of Do as expressed in equation (7.4) (Shackelford and Daniel, 1991; Robinson and Stokes, 1959; Beek, et al. 1999) r z T f Do (7.4) where |z| is the absolute va lue of the ionic valence, is the absolute viscosity of the solution, r is the molecular or hydrated ionic radius, and T is the absolute temperature of the solution.

PAGE 175

156 Figure 7.3 Molecular Diffusion of Solute In order to incorporate Fick’s first law in a soil medium, modifications to equation (7.3) have been introduced by many research ers, (e.g., Cheung and Gray, 1989; Eriksen, et al. 1999; Foose, 2002; Shackelford and Lee, 2003; Malusis and Shackelford, 2004). Chemical diffusion in soils is much slower th an in the free solution because of the effect of porosity, especially in fine grained soils where the permeability is lower and where a tortuous pore channels exist. Further reducti on of diffusivity happens in clays since the particles are adsorptive due to the negatively charged particle surface. Factors affecting the diffusivity of chemical solutes thr ough a soil mass can be summarized as follows: (a) Cross-sectional area of flow within the soil mass: The availability of the flow path depends on the porosity of th e soil and the degree of saturation. Diffusivity is directly proportional to the values of porosity (n) and degree of saturation (Sr). The maximum flux for liquid phase diffusion occurs when the soil is fully saturated (degree of saturation, Sr = 1.0). (b) Flow path tortuosity: Tortuosity ( ) of a soil mass, which depends on the shape and arrangement of clay/soil pa rticles, reduces the flow rate of Concentration gradient produces mass fluxHydraulic head (constant) No fluid flow High concentration Low concentration Solute Particle Concentration gradient produces mass fluxHydraulic head (constant) No fluid flow High concentration Low concentration Solute Particle

PAGE 176

157 chemical solutes through diffusion. Si nce it is not possible to measure the tortuous flow path directly, the e ffect of tortuosity is typically incorporated into the value of diffusi vity coefficient of solute flow as suggested by many researchers (Shack elford and Daniel, 1991; Quigley and Rowe, 1986; Quigley, et al. 1987). (c) Fluidity or mobility of the fluid adja cent to clay particles: The viscosity of the fluid adjacent to the clay mineral surfaces is higher than that of bulk fluid because of the immobility of the clay surface water and the higher adsorption capacity of the negatively charged clay particle surfaces. A fluidity factor ( ) has been introduced by Kemper et al. (1964), Olsen et al. (1965) which reduces the diffusivity of chemical solutes through finegrained adsorptive clay particles. (d) Anion exclusion: Electrical im balance might occur on clay mineral surfaces due to the exclusion of anionic charges which are expelled from the pores between diffuse double layers when subjected to high stresses (Porter et al. ,1960; Freeze and Cherry, 1979; Drever, 1982; Shackelford and Daniel, 1991). However, it is not quite possible to separate the anion exclusion factor ( ) from other factors in determining the diffusivity. The only factor from the above list that can be readily measured for any clay material is porosity. Th erefore, the chemical mass fl ux due to diffusion through finegrained (non-reactive) clay can be written by adopting an effective diffusion coefficient, D *, which incorporates all other control ling factors as given in equation (7.5). x C n D Jd (7.5)

PAGE 177

158 The effective diffusion coefficient, D *, can be expressed wi th all the relevant factors including those expre ssed in equation (7.4) as gi ven in the following equation (7.6). r z T f D (7.6) However, the transport of solutes that are subjected to chemical reactions or chemical exchanges (cation exchange for bent onite clay minerals), which are analogous to “reactive solutes,” differ from the transport of nonreactive solutes as calculated using equation (7.5). In order to accommodate th e effects of cation exchange on the clay mineral surfaces, an additional factor known as “retardation factor”, Rd, has been added in the diffusion formulation which inversely affects the flow of solutes as given in equation (7.7). x C R n D Jd d (7.7) The retardation factor can be define d in terms of partition coefficient Kp, as given by equation (7.8). p d dK n R 1 (7.8) The partition coefficient, Kp, is defined as the amount of a given constituent that is adsorbed or desorbed by a soil for a unit increase or decrease in the equilibrium concentration in solution (Mitchell, 1993, Shackelford and Daniel, 1991). Fick’s first law is only applicable for di ffusive flux of solutes under steady-state condition when the concentration gradient with in the medium does not change with time. The rate of change in concentration with ti me and distance within the transport medium, as shown in figure 7.4, is described by Fick’s second law which can be expressed mathematically for non-reactive solute diffusion as follows:

PAGE 178

159 2 2* x C D t C (7.9) Fick’s second law for reactive solutes, where adsorption on clay mineral surfaces occurs during diffusive transport in clay soil, can be expressed by equation (7.10) incorporating the retardation factor (Freeze and Cherry, 1979, Shackelford and Daniel, 1991, Mitchell, 1993; and many others). Figure 7.4 Diffusion as a F unction of Distance and Time 2 2 2 2 * x C D x C n R D t CA d (7.10) The value of ( D* n/Rd), replaced byAD *, is defined as the “apparent diffusion coefficient” by many researchers (Quigley et al. 1987; Li and Gregory, 1974). 7.1.2.1 Mathematical Solution to Diffusion Equation The partial (second order) differential e quation (7.10) which has been solved mathematically by various researchers in th e form of equation (7.11) as suggested by Ogata (1970) and Freeze and Cherry ( 1979) is most popular among engineers. Distance, xC [ML-3] t2t1 t2 > t1 Final constant gradient (i.e. steady state condition) Distance, xC [ML-3] t2t1 t2 > t1 Final constant gradient (i.e. steady state condition)

PAGE 179

160 t D x erf t D x erfc C CA A O* 2 1 2 (7.11) where C is the concentration at any time, Co is the constant supply concentration, and x is the distance of travel at time t. The initi al concentration of the medium through which diffusion occurs is considered to be zero at t = 0. Equation (7.11) can also be fitted to a forward difference numerical solution which can be easily implemented in a spreadsheet as outlined below. The purpose of solving the differential equation would be to calculate the concentration of solute ( Cx,t) at any depth of the clay medium as time progresses. The subscripts (x,t) of concentra tion C, have been changed so as to provide more arithmetic representation as follows: (time) (depth) where,, ,t j x i C Cj i t x Since the solution is required for 1-D ve rtical diffusion flow, the clay layer has been divided into a number of thin layers with a distance or depth of x. It is first required to calculate the concentration at 1 j iC based on neighboring locations on previous time as ,, 1 j i j iC Cand j iC, 1 as follows: To find 2 2/ dx C d in a finite difference scheme, the concentration function C = f(x) about point i can be expanded using Ta ylor’s series expansion. Time Depth 1 1 , 1 j i j i j i j iC C C CTime Depth 1 1 , 1 j i j i j i j iC C C C

PAGE 180

161 ... 4 3 22 2 4 2 2 3 2 2 2 1 i i i i i ix C x x C x x C x x C x C C (7.12) ... 4 3 22 2 4 2 2 3 2 2 2 1 i i i i i ix C x x C x x C x x C x C C (7.13) Adding equations (7.12) and (7.13) would result in equation (7.14). ... 4 2 22 2 4 2 2 2 1 1 i i i i ix C x x C x C C C (7.14) Since the value of 2x is small, the value of 4x is even smaller and can therefore be neglected. By rearranging e quation (7.14), the expression of 2 2/x C can be written as follows: 2 , 1 1 2 22 x C C C x Ct i t i t i (7.15) The partial differential term on the left side of Fick’s second law in equation (7.10) can be refined as follows: t C C t Ct i t t i , (7.16) By substituting equations (7.15) and (7.16) into equation (7.10), the concentration of solute at any location i after an infinitesimal time interval t can be calculated using equation (7.17). t i t i t i A t i t t iC C C x D t C C, 1 1 2 ,2 (7.17) It can be seen from equation (7.17) that the value of concentration at a node at the next time step (t + t) is determined from the values at the current time at the three adjacent nodes (i-1, i, and i+1). In this formulation, t is the incremental time step in the numerical solution, and x is the increment in sp ace in the direction, x. Boundary conditions on the sides of the clay medium during diffusion play an important role in calculating and representi ng graphically the diffus ion profile with the variation of time during the transient period be fore achieving steadystate condition. In this research, a source of constant concentration, Co, has been applied and the

PAGE 181

162 concentration at a depth is pres ented in proportion to the initia l concentration, in the form of C/Co, as shown later in the chapter. Gra phs are plotted to show the change in concentration as isochrones of C/Co with respect to elapsed time. 7.1.3 Chemico-Osmotic Flow Osmosis flow is considered when the cl ay material acts like a semipermeable membrane. Osmosis is a process when a memb rane restricts the passage of solutes while allowing the flow of solvent due to the differ ence in concentration of the solvent between the both sides. The transport of solvent (eg. water) stops when the concentrations of the solutions on both sides are the same, or when the hydraulic pressure across the membranes equals the osmotic pressure differ ence between the two solutions, as sketched in figure 7.5. Figure 7.5 Chemico-Osmosis of Solute Transport It can be seen from figure 7.5 that the chemico-osmosis phenomenon counteracts the flow of solute and therefore reduces the contaminant outward flux (Malusis et al., 2001; Shackelford et al., 2001). Higher concentrated solution Lower concentrated solution membrane Water flux (osmosis) Chemical flux (diffusion) Hydraulic or Osmosis pressure at equilibrium Higher concentrated solution Lower concentrated solution membrane Water flux (osmosis) Chemical flux (diffusion) Hydraulic or Osmosis pressure at equilibrium

PAGE 182

163 The chemico-osmosis efficiency coefficient, also known as the reflection coefficient, is defined as the ratio of the pr essure difference induced across the membrane as a result of prohibiting chemico-osmotic flux of solution (P) to the theoretical chemico-osmotic pressure di fference across an ‘ideal’ semipermeable membrane () subjected to an applied difference in solute concentration as shown in equation (7.18) (Malusis et al. 2003; Keijer, 2000). P (7.18) It can also be defined as the ratio of the developed hydraulic pressure over the applied osmotic pressure after equilibrium i. e. at zero solution fl ux (Keijer, 2000). The chemico-osmotic efficiency coefficient, ranges from zero ( =0) for non-membranes to unity ( =1) for ‘ideal’ membranes that complete ly restrict the passage of solutes. Clay minerals can be considered to be ‘non-ideal’ membranes with < 1. The theoretical value of osmotic pressure () is calculated with respect to concentration variation at the membrane boundaries by the van’t Hoff equation (7.19) (Malusis and Shackelford, 2002; Mitchell, 1993). N i L i H iC C RT1 , (7.19) where, R = the universal gas constant [8.314 J mol-1K-1 or 0.0821 atm mol-1K-1], T = the absolute temperature [K], Ci,H = the initial high concentration of solute i species [mol L3], and Ci,L = the initial low concentration of solute i species on the other side of the membrane. The induced hydraulic pressure can be calculated or measured from the levels of the standpipes connected to the so lutions on both sides of the membrane, which varies with time until it reaches equilibrium w ith constant elevations of solution on both sides. A steady-state solute flux through the semi permeable clay specimen is established and maintained when (a) the osmotic pre ssure is counterbalanced by the hydraulic pressure and (b) constant flow of solute diffusion occurs due to the difference in concentration gradient.

PAGE 183

164 7.1.4 Determination of Diffusion Parameters The developed hydraulic pressure due to osmosis can be measured by observing the solution levels of the higher concentrated standpipe or by using a differential pressure transducer placed on both sides of the membrane (Shackelford and Lee, 2003). The pressure gradually increases with time until it reaches its peak value and then decreases due to diffusion of solute until it reaches its steady-state condition. A typical graph of these processes with respect to induced chemico-osmotic pressure (P) and elapsed time is drawn in figure 7.6 (Shackelford and Lee, 2003). During the process of solute transport due to diffusion and chemico-osmosis, the concentration of the solution is measured from the concentration of individual species of solute. In the steady-state condition, the meas ured concentrations for a given solute are converted to cumulative mass per unit cross-sectional area, Qt, as given in equation (7.20) (Malusis et al. 2001; Shackelford and Lee, 2003). N i i i N i i tV C A m A Q1 11 1 (7.20) where, A = cross-sectional area of the specimen, mi = mass increment of the solute species i collected over a time increment ( t), Vi = increment volume of the solution from which the outflow flux is collected, Ci = the concentration of the solute species in the incremental volume, and N = number of incremental samples (solution) collected during the total elapsed time, t. The values of Qt calculated from equation (7.20) with respect to elapsed time can be plotted as s hown in figure 7.6 (Shackelford and Lee, 2003; Malusis et al., 2001). It is seen from figure 7.7 that the consta nt slope line, which represents the steadystate diffusion, intersects with the time axis at tL commonly known as lag time (Shackelford, 1991; Malusis et al., 2001). The time, tss, in figure 7.7 denotes the time required for steady-state diffusion or the ti me until which transient diffusion occurs within the specimen due to chemico-osmosis of the semipermeable clay membrane.

PAGE 184

165 Figure 7.6 Induced Chemico-Osmotic Pr essure Observed for Clay Membranes (Shackelford and Lee, 2003) Figure 7.7 Cumulative Solute Mass Th rough Clay Specimen due to Diffusion (Shackelford and Lee, 2003; Malusis et al., 2001) The analytical solution for cumulative mass flux (Qt) due to diffusion in 1-D direction under steady-state condition has been investigated by many researchers. The expression given in equation (7.21) by Crank (1 975) and Shackelford (1 991) is applicable for a constant source concentration, Co, and a perfectly flushing boundary condition (concentration of solute is kept zero). Time, tInduced chemico-osmotic pressure, P>0 Psteady-state= Ppeak Psteady-state< Ppeak Increasing C C > 0 0 Time, tInduced chemico-osmotic pressure, P>0 Psteady-state= Ppeak Psteady-state< Ppeak Increasing C C > 0 0 Time, tCumulative mass per unit area, Qt Qt/ t = constant Time to steady state Time lag C > 0 0 tLtss Time, tCumulative mass per unit area, Qt Qt/ t = constant Time to steady state Time lag C > 0 0 tLtss

PAGE 185

166 6 *o d o tLC nR t L C nD Q (7.21) where, n = the specimen porosity, L = length or thickness of the specimen, D* = effective diffusion coefficient, Rd = the retardation factor, and t = total elapsed time of diffusion flow. The slope t Qt / of figure (7.6) which represen ts the steady-state diffusion condition is obtained by best-fit regression of the straight port ion of the graph. The value of effective diffusion coefficient, D*, of any solute species can be computed using equation (7.21) by considering the term 6 /o dLC nR as zero at steady-state condition, which gives the following equation (7.22) o tnC L t Q D (7.22) The value of retardation factor, Rd, of any solute species can be computed by using equation (7.21) and lag time, tL, at time intersection when Qt = 0 as follows: L dt L D R2* 6 (7.23) The value of D* calculated from equation (7 .22) is used to evaluate the value of Rd from equation (7.23). The total solute mass flux of any dissolved chemical species (i), Ji, through low permeability clay due to advection, chemicoosmosis, and diffusion can be written as follows (Mitchell, 1993; Malu sis and Shackelford, 2002, 2004): x R C nD C q C q J J J Jd i i A i i h i d i i a i , ,* 1 (7.24) where, Ja,j = advection solute flux due to hydraulic gradient (ih), = chemico-osmotic efficiency coefficient (0 1), qh = Darcy’s flux (=khih, where kh = hydraulic coefficient), J = chemico-osmotic solute flux, q [= khi, where i = the gradient in chemico-osmotic pressure head] is the chem ico-osmotic solute flux for a unit difference in concentration from lower solute concentration to higher solute concentration (i.e.,

PAGE 186

167 opposite to the direction of so lute diffusion), n = porosity, i AD,*= effective salt-diffusion coefficient, and Ci = initial influent molar solute concentration. The chemico-osmotic coefficient controls th e types of flow to be considered for evaluating the solute mass through the semipermeab le membrane barriers. If the value of is close to zero (i.e. non-membrane, permeable layer), then, q 0, and equation (7.24) would become the conve ntional advection-diffusion so lute mass flux expression as given in equation (7.25). x R C nD C q J J Jd i i A i h i d i a i , 0* (7.25) However, when the value of is close to unity (i.e. ideal membrane, impermeable layer), then, the advection flow would be zero and the chemico-osmotic flux and diffusion will become the same which would cancel each other, eventually causing the resultant solute flux of equati on (7.24) to become zero (J 0) (Malusis and Shackelford, 2003). 7.2 Analysis of Diffusion Test Results Analysis of diffusion test results was car ried out in order to determine the various diffusion parameters, namely, effective diffu sion coefficient of inorganic chemical elements (D*), retardation factor (Rd), partition coefficient (Kp), and apparent diffusion coefficient (AD *). The following sub-sections desc ribe these diffusion parameters in detail. 7.2.1 Lag Time and Time to Steady-State In order to obtain the lag time and time to steady-state as expl ained in figure 7.6, electrical conductivity values taken at regula r interval of time were accumulated with

PAGE 187

168 time of diffusion and plotted as cumulative electrical conductivity vs diffusion time, as shown in figures B.1(b) to B.11(b) in appendix B. A st raight line was drawn for each graph using linear regression, and the intercept of the straight line on the diffusion time axis was taken as the lag time. The point where the curve generated during the initial stage of diffusion joins the stra ight line is known as time to steady-state. The lag time and time to steady-state of diffu sion tests carried out in this study are tabulated in Table 7.1. Since the values of lag time and time to steady-state vary with the void ratio, specimen thickness, and concentration of diffusa nt, diffusion tests were divided into three groups, as shown in Table 6.3 of Chapter six. Lag time and time to steady-state are required to be measured as accurately as po ssible using consistent statistical methods. Because of the slight scattering in data fr om a theoretical straight line, even after attaining the steady-state condi tion, a sequential linear regression method was used to obtain the best possible straight line for the steady-state condition. The sequential linear regression was carried progressively from the last 3 data points of the EC versus time data with an increment of one additional data point in successive regression cycles. In each regression cycle, the coefficient of determination, R2, was calculated and compared with the next regression analysis coeffi cient until a significa nt deviation in R2 was found. The data point corresponding to the location where R2 starts to drop significantly from the previous regression cycle represents the time when the transient diffusion ends. This data point also represents the transition from the initial non-linear curve to the linear slope line. Therefore, the elapsed time associated with the earliest maximum R2 value of the regression analysis represents the time required to establish steady-state diffusion of the solute, tss, as explained in figure 7.6. The straight line representing the st eady-state diffusion, obtained from the sequential linear regression, is then exte nded to the horizontal (time) axis, and the intercept value on the time scale is established as the lag time, tL, of the diffusion solute. A summary of the statistical methods for all the diffusion tests with R2 values are tabulated in Table 7.1.

PAGE 188

169 Table 7.1 Summary of Statisti cal Method for Stea dy-State Diffusion Test No. Number of data points used R2 Steady-state equation D-6 18 0.9982 Y = 382.82X 14913 D-8 9 0.9978 Y = 566.66X – 25387 D-9 16 0.9896 Y = 359.66X – 14468 D-10 15 0.9959 Y = 453.78X 17907 D-11 17 0.9940 Y = 847.67X 20905 D-12 24 0.9979 Y = 1664.1X – 13584 D-13 7 0.9904 Y = 756.82X – 40684 D-14 23 0.9973 Y = 1116.7X – 15682 D-16 17 0.9979 Y = 677.77X 13165 The time to steady-state and lag time obtai ned using the above statistical methods for all the diffusion tests conducted in th is study are listed in Table 7.2 with corresponding void ratios of the test specimens. The times to steady state are compared to theoretical values generated from the numerical analysis as explained later in this chapter. The amount of influx coming out of the specimen after achieving the steadystate condition is found to be constant for any in terval of time as defined by the slope of the straight line drawn on the graphs shown in figures B.1(b) to B.11(b). From the test results of group #3 of equal void ratio of test specimens, it was found that the lag time of cations follows the sequence Ca2+ < Na+ < K+ < Mg2+, which means that the calcium cations lag time is the shortest, follo wed by sodium, potassium, and magnesium.

PAGE 189

170 Table 7.2 Lag Time and Time to Steady-State of Various Diffusants Group # Test number Source solution Void ratio Lag Time (days) Time to Steady-state (days) D-6 2M CaCl2 9.23 39 53 D-11 2M CaCl2 14.35 25 36 D-12 5M CaCl2 14.35 8 9 1 D-13 5M CaCl2 4.11 54 63 D-10 2M NaCl 14.35 40 58 D-14 5M NaCl 14.35 14 18 2 D-16 5M NaCl 6.67 19 26 D-8 2M MgCl2 14.35 45 58 D-9 2M KCl 14.35 40 58 D-10 2M NaCl 14.35 40 58 3 D-11 2M CaCl2 14.35 25 36 7.2.2 Diffusion Coefficient Two types of diffusion coefficient, na mely, effective diffusion coefficient (D*) and apparent diffusion coefficient (AD *) are calculated in this study. The effective diffusion coefficient (D*) of any cation is calculated using equation 7.22 as follows: o tnC L t Q D (7.22) The slope of the steady-state line as s hown in figures B.1(b) to B.11(b) is converted from a change in electrical c onductivity per unit time [microSiemen/day] to change of mass flux per unit area per unit time [mg/(cm2 x s)]. This conversion is carried out according to the following steps:

PAGE 190

171 (a) The electrical conductivity in micros iemens is multiplied by 0.66 to obtain the mass flux concentration in ppm (mg/liter). (b) The mass flux in ppm (mg/liter) is multiplied by the volume of receptor solution (after dilution) to calcu late the total mass flux in mg. (c) The total mass flux is distributed among its cations and anions components. At steady-state c ondition, in order to fulfill the electroneutrality requirement, the charge flux of the anions (in this case is the chloride anion, Cl-) is of the same magnitude as the charge flux of the cations (that constituents the salt solution, namely, Na+,K+,Mg2+, and Ca2+) according to the following equation 7.26. cation cation anion anionz J z J (7.26) where Janion and Jcation are the steady-state diffusive molar fluxes of anions and cations, and zanion and zcation are the charges of anions and cations respectively. The steady-state diffu sive molar flux of chloride (Cl-) anion will therefore be the same magnitude of the steady-state diffusive molar flux of monovalent cations and twice the magnitude of divalent cations. However, in order to obtain the mass fluxes of Cl-, the above ratios are required to be multiplied by the ratio of atomic weight of cation to the atomic weight of Cl-. For example, for the NaCl solution, the magnitude of the steady-state diffusive mass flux of Clwill be [= 1 x (23/35.453)] 0.648 times the magnitude of the sodium (Na+) cation mass flux. Similarly, for CaCl2 solution, the rati o of mass flux for Ca2+ and Clat steady-state will be 1:2.26. (d) The value obtained in step 2 is divided by the cross-sectional area [cm2] of the clay specimen. (e) The units are them converted to a consis tent set of unties to finally attain the unit of mg/(cm2 x s). After calculating the slope of the steady-state line in mg/(cm2 x s) for a particular cation mass flux per unit area per duration of diffusion, the source concentration of the same cation, Co, is then calculated in mg/cm3. For example, for a 2M CaCl2 solution, the

PAGE 191

172 theoretical value of Co would be [= 80,000 ppm = 80,000 mg/liter = 80,000 mg / 1000 cm3 ] = 80 mg/cm3. The value of [L/(nCo)] is then calculated in cm4/mg, since the porosity, n, is dimensionless, and the length or thickness of the specimen is in cm. The units of the effective diffusion coefficient is therefore [mg/(cm2 x s) x cm4/mg ] or cm2/s which is then converted to the more commonly used unit of cm2/day as shown in the Table 7.3. The source concentration, Co, was calculated based on individual cation concentration of the synthetic salt so lution. For example, for the 2M CaCl2 source solution, the concentration Co is 40,000 x 2 = 80,000 ppm or mg/l.

PAGE 192

173 D* (m2/s) (Cations) 5.45E-12 6.61E-12 2.94E-12 8.04E-12 4.37E-12 3.43E-12 1.81E-12 7.91E-12 5.16E-12 D* (cm2/s) (Cations) 5.45E-08 6.61E-08 2.94E-08 8.04E-08 4.37E-08 3.43E-08 1.81E-08 7.91E-08 5.16E-08 (L/nCo) cm4/mg 0.011086 0.006602 0.004103 0.006975 0.004011 0.001604 0.001863 0.002790 0.002999 Co (mg/l) 80000 48600 78200 46000 80000 200000 200000 115000 115000 ( Q/ T) mg/cm2/s 4.9E-06 1.0E-05 7.2E-06 1.2E-05 1.1E-05 2.1E-05 9.7E-06 2.8E-05 1.7E-05 Cations (mg) 19.3698 39.4325 28.2212 45.4125 42.8901 84.1995 38.2933 111.755 67.8285 Outflux (mg) 63.1653 93.4989 59.3439 74.8737 139.866 274.577 124.875 184.256 111.832 Receptor Vol. (ml) 250 250 250 250 250 250 250 250 250 Slope (EC/day) 382.82 566.66 359.66 453.78 847.67 1664.1 756.82 1116.7 677.77 L (cm) 0.8 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 Porosity n 0.902 0.935 0.935 0.935 0.935 0.935 0.935 0.935 0.935 Source Solution 2M CaCl2 2M MgCl2 2M KCl 2M NaCl 2M CaCl2 5M CaCl2 5M CaCl2 5M NaCl 5M NaCl Test No. D-6 D-8 D-9 D-10 D-11 D-12 D-13 D-14 D-16 Table 7.3 Worksheet for the Calculation of Effec tive Diffusion Coefficient, D* of Various Cations

PAGE 193

174 In order to calculate the appa rent diffusion coefficient, (AD *), the relationships of retardation factor, Rd, with diffusion coefficients as given in equations 7.10 and 7.23 were rearranged and the following expression was deduced in terms of the lag time and the physical dimensions of the test specimen. L At nL D6 *2 (7.27) The values of apparent diffusion coefficients, (AD *), of various cations used as source solution during the diffusion tests thr ough bentonite are calcu lated using equation 7.27 and are tabulated in Table 7.4. By comparing the tests D-8, D-9, D-10, and D-11 of the same porosity and thickness specimens usi ng the same concentrated diffusants, it can be found that the apparent diffusion coefficient of Ca2+ (i.e. 6.49x10-13 m2/s) is higher than those of other cations due to its highe r replaceability capacity as compared with others cations used in bentonite. Table 7.4 Apparent Diffusion Coeffici ent for Various Cations in Bentonite Test No. Source Solution Porosity n L (cm) Lag Time (days) AD* (m2/s) (Cations) D-6 2M CaCl2 0.902 0.8 39 2.86E-12 D-8 2M MgCl2 0.935 0.3 45 3.61E-13 D-9 2M KCl 0.935 0.3 40 4.06E-13 D-10 2M NaCl 0.935 0.3 40 4.06E-13 D-11 2M CaCl2 0.935 0.3 25 6.49E-13 D-12 5M CaCl2 0.935 0.3 8 2.03E-12 D-13 5M CaCl2 0.805 0.3 54 2.59E-13 D-14 5M NaCl 0.935 0.3 14 1.16E-12 D-16 5M NaCl 0.87 0.3 19 7.95E-13

PAGE 194

175 7.2.3 Retardation Factor Lag time is again used to cal culate the retardation factor, Rd, of individual cations using the expression given in equation 7.23. Re tardation factor is di rectly proportional to both effective diffusion coefficient and lag time. The values of retardation factor of all the cations used in this st udy are given in Table 7.5. Table 7.5 Retardation Factor of Various Cations in Bentonite Test No. Source Solution L (cm) Porosity n Lag Time (days) D* m2/s (Cations) Rd (Cations) D-6 2M CaCl2 0.8 0.902 39 5.45E-12 1.722 D-8 2M MgCl2 0.3 0.935 45 6.61E-12 17.126 D-9 2M KCl 0.3 0.935 40 2.94E-12 6.771 D-10 2M NaCl 0.3 0.935 40 8.04E-12 18.522 D-11 2M CaCl2 0.3 0.935 25 4.37E-12 6.287 D-12 5M CaCl2 0.3 0.935 8 3.43E-12 1.580 D-13 5M CaCl2 0.3 0.805 54 1.81E-12 5.633 D-14 5M NaCl 0.3 0.935 14 7.91E-12 6.381 D-16 5M NaCl 0.3 0.87 19 5.16E-12 5.649 The retardation factor of all the individua l cations is found to be more than unity, as listed in Table 7.4, which indicates th at adsorption happens on the surface of clay platelets due to diffusion flow of the cations (Ca2+, Mg2+, K+, and Na+). The smallest retardation factor was obtained for calcium cations with a highly porous bentonite specimen (D-12), probably because of its mi nimum resistance to diffusion on bentonite clay platelets at steady-stat e condition. The 2M NaCl source solution produces the maximum retardation factor (test. D-10) i ndicating a maximum resistance to diffusion on bentonite clay platelets at st eady-state condition. The thickness of diffuse double layer is higher in sodium concentrated solution than in calcium concentrated solution, which

PAGE 195

176 results in a zone of immobility within the pore spaces. By comparing test D-10 and D-14 for 2M NaCl and 5M NaCl source solutions, re spectively, it can be concluded that the higher concentrated solutions generate lowe r retardation factors because of the lower diffuse double layer thickness which eventually creates more available pore spaces for solute mobility. The same trend can be observed in calcium solutions where higher concentrated source solution (D-12) develops lower retardation factor compared to lower concentrated source solution (D-11) fo r bentonite of the same porosity. 7.2.4 Partition Coefficient The partition coefficient, Kp, is defined as the ratio of the adsorbed concentration on the clay surfaces to the concentration of so lution in equilibrium. It be calculated using equation 7.8 after calculating th e value of retardation factor of each individual cations. The values of partition coefficient of all the cations used in this study are given in Table 7.6. It can be highlighted from Table 7.6 that the minimum partition coefficient was found in calcium source solutions (e.g. Kp = 0.492 from D-12 of 5M CaCl2) which represents the minimum adsorption on the clay platelet surfaces. The maximum partition coefficient (Kp = 14. 88) was found in test D-10 with the 2M NaCl solution, indicating the maximum adsorption on the clay pl atelet surfaces. It can be concluded, from tests D-11 and D-12 on 2M CaCl2 and 5M CaCl2 diffusants, respec tively, that for clay specimens of the same porosity, the highe r concentrated diffusant results in a lower partition coefficient due to the collapse of cl ay platelets as a result of shrinkage in the diffuse double layer and the formation of mo re aggregated partic les where the total adsorption capacity per unit surface area decreases comp ared with the increasing concentration of pore fluid in equilibrium. The 2M MgCl2 diffusant also resulted in a higher partition coefficient, which could be due to its higher hydrated ionic radius that gets obstructed in the diffuse double layer. Mo re diffusion tests are necessary in order to

PAGE 196

177 conclude the mechanism of diffusivity of ma gnesium and potassium cations in bentonite clay. Table 7.6 Partition Coefficient of Various Cations in Bentonite Test No. Source solution porosity n Bulk density b (g/ml) Rd (Cations) Kp (Cations) D-6 2M CaCl2 0.902 1.152 1.722 0.565 D-8 2M MgCl2 0.935 1.101 17.126 13.694 D-9 2M KCl 0.935 1.101 6.771 4.901 D-10 2M NaCl 0.935 1.101 18.522 14.880 D-11 2M CaCl2 0.935 1.101 6.287 4.490 D-12 5M CaCl2 0.935 1.101 1.580 0.492 D-13 5M CaCl2 0.805 1.303 5.633 2.862 D-14 5M NaCl 0.935 1.101 6.381 4.570 D-16 5M NaCl 0.87 1.202 5.649 3.365 7.2.5 Diffusion Coefficient Through Numerical Solution The apparent diffusion coefficient (AD *) obtained from the lag time method can be used in calculating the time to steady-st ate using numerical method, as explained in section 7.1.2.1. By knowing the boundary conditions and dividing the specimen into a number of thin layers, the time to steady-stat e of any particular diffusant can be obtained by the forward numerical difference method, as expressed in equati on (7.17). A simple spreadsheet was formulated to calculate th e diffusion mass flux at any particular time interval and location within the test specime n. The process of calculating the diffusion mass flux through the bentonite clay specimen continues until a stead y-state condition is reached which satisfies the constant ma ss flux of diffusant obtained from the

PAGE 197

178 experiments. In theory, complete steady state will never be reached, but a condition where the variation in concentr ation is almost linear with distance could be viewed as steady state. The diffusion profile of Mg2+ cations using the 2M MgCl2 diffusant through bentonite is shown in figure 7.8. Using the value of AD *obtained from the time-lag method, it can be found that the time required to achieve a constant diffusion mass flux at the receptor end is 56 days as compared to 58 days as calculated from lag time method given in Table 7.2 (D-8). Figure 7.8 Diffusion Profile of Mg2+ Ions Using Numerical Method The time required to satisfy the conditi ons of steady-state, which were derived from the lag time method for K+ cations using a 2M KCl di ffusant through bentonite specimen layer is 47 days as shown in figure 7.9. However, it took about 58 days to reach the steady-state condition using the la g time analysis as shown in Table 7.2. 0 0.5 1 1.5 2 2.5 3 010,00020,00030,00040,00050,00060,000Mg2+ (ppm)Depth (mm) t = 5 t = 10 t = 15 t = 20 t = 30 t = 40 t = 50 t = 55 t = 56 2M MgCl2 (n = 0.935), D A = 3.61x10-13 m 2/s

PAGE 198

179 Figure 7.9 Diffusion Profile of K+ Ions Using Numerical Method Figures 7.10 and 7.11 show the diffusion profiles of Na+ and Ca2+ cations respectively using numerical methods at vari ous depths and durations until steady-state conditions as obtained by the lag time method were satisfied. The time to steady-state with Na+ cation using the numerical method was found to be 41 days while that for Ca2+ cation was about 35 days, compared to 58 days and 36 days, respectively, from the lag time method (Table 7.2). It is therefor e concluded that the numerical method underpredicts the time to st eady state, compared to the lag ti me method. However, it can be seen from the slope of the diffusion profiles (f igures 7.8 to 7.11) that further diffusion would result in a condition that better approxi mates the theoretical steady-state condition (straight line). 0 0.5 1 1.5 2 2.5 3 010,00020,00030,00040,00050,00060,00070,00080,00090,000K+ (ppm)Depth (mm) t = 5 t = 10 t = 15 t = 20 t = 30 t = 40 t = 46 t = 47 2M KCl (n = 0.935), DA* = 4.06x10-13 m2/s

PAGE 199

180 Figure 7.10 Diffusion Profile of Na+ Ions Using Numerical Method Figure 7.11 Diffusion Profile of Ca2+ Ions Using Numerical Method 0 0.5 1 1.5 2 2.5 3 05,00010,00015,00020,00025,00030,00035,00040,00045,00050,000Na+ (ppm)Depth (mm) t = 5 t = 10 t = 15 t = 20 t = 30 t = 40 t = 41 2M NaCl (n = 0.935), DA* = 4.06x10-13 m2/s 0 0.5 1 1.5 2 2.5 3 010,00020,00030,00040,00050,00060,00070,00080,00090,000Ca2+ (ppm)Depth (mm) t = 5 t = 10 t = 15 t = 20 t = 30 t = 34 t = 35 2M CaCl2 (n = 0.935), D A = 6.49x10-13 m 2/s

PAGE 200

181 CHAPTER EIGHT SUMMARY AND RECOMMENDATIONS 8.1 Summary While bentonite has been used as a flow barrier in many applications, its performance has been found to deteriorate wh en in contact with inorganic chemicals present in the leachate. Attempts have b een made to find a relation between bentonite performance, in terms of coefficient of pe rmeability, and its geotechnical properties. Aggregated platelets size di stribution obtained using hydromet er tests were found to be inconsistent with the liquid limits tests obtained using the cone penetrometer, with finer aggregated platelets produci ng higher liquid limit in the sequence of Na>K>Mg>Ca. However, this finding was limited to 0.1 molar concentrated electrolyte solutions of the above inorganic salts. No distinctive relationship was found to exist between the li quid limits and the coefficient of permeability obtained using both flexible wall and rigid wall permeameters. However, a strong correlation was found to exist between swell index and hydraulic conductivity and was attributed to the fact that similar mechanisms control both the swelling behavior and the hydr aulic conductivity. Swell inde x values obtained using 1 molar concentration of various salt solutions (figure 3.22) were f ound to be in sequence with the values of coefficients of perm eability obtained using rigid wall permeameter (Na
PAGE 201

182 electrical conductivity measurements. The coefficient of permeability measured from the flexible wall permeameter was found to be erratic because of the unpredictable and immeasurable shape of the swe lling clay specimen during testi ng. Significant swelling of specimen occurred, especially while performing permeability tests using water and lower concentrated electrolyte solutions. Sidewa ll leakage channels, normally developing in rigid wall type experiments, can easily be eliminated when the bentonite clay specimens are hydrated and saturated for at least 48 hour s before the actual permeation is carried out. Swelling of the bentonite upon hydration acts as a self sealing mechanism for all internal and sidewall channels. Other than the type of permeameter, th e most important factors affecting the permeability of bentonite are permeant chemical composition, void ratio, and initial hydration condition. A distinct variation in coefficient of permeability is observed between permeants containing sodium and calcium cations. Higher k values for permeants containing calcium is a ttributed to the fact that Ca2+ replaces monovalent cations, such as Na+, K+ and others attached on to the negatively charged clay surface, and thereby reduces the thickness of the di ffuse double layer. Relationships for the variation in coefficient of permeability betw een sodium and calcium solutions, as well as void ratio relationships, were es tablished. No significant vari ation of k of bentonite clay was observed upon applying hydraulic gradients at high as 3100, which discredits the concept of consolidation of be ntonite at higher gradients. Pre-hydration of bentonite clay plays an important role in its permeability. Different structures of the platelets are fo rmed when hydrating with water and various inorganic electrolyte solutions before the actual permeation of the solution through the specimen. Electrical conductivity of the effluent can be used as an indicator to monitor the amount of chemical retention during permeati on by various inorganic electrolytes. No significant correlation could be found between pH and electr ical conductivity of the effluent during permeation with various electrolyte solutions. Various diffusion parameters, namely effective and apparent diffusion coefficients, retardation factor, and partiti on coefficient, of four different inorganic

PAGE 202

183 chemical diffusants were investigated in this study. The time to steady-state determined from the lag time method was found to be sl ightly shorter than that from the forward numerical difference method. Ca lcium cations were found to di ffuse faster than the other three cations (Na+, K+, Mg2+), as identified by their retardation factor values. Higher sodium retardation factors indicate its str ong affinity for the negatively charged clay surface for a longer period of time. Maximum sodium cation adsorption on the clay surface was also confirmed by its highest parti tion coefficient compared to other cations used in this study. Partition coefficients of the cations were found to be in the order of Ca2+ < K+ < Mg2+ < Na+. 8.2 Design Recommendation Design recommendation can be made to be ntonite clay of similar physical and engineering characteristics to be used as hydr aulic barriers in the field. The relevant physical, chemical and engineering propertie s necessary for design are Atterberg limits, and swell index. Particle or aggregated cl ay platelets size distri bution from hydrometer tests could be misleading because of the cha nges in specific gravity of the aggregated particles formed upon hydrated. Other limitati ons of the hydrometer test stem from the use of Stokes’ law, which assumes the aggreg ated particle to be solid single spherical particle of pre-determined specific gravity. Liquid and plastic limits of the bentonite used in this study were found to be 546% and 56%, respectively, while the free swell index determined using deionized water was 60 ml/2g of dry bentonite. The fr ee swell index can also be measured using various inorganic salt solutions to provide a rapid indica tor of the coefficient of permeability of the bentonite when permeated with the corresponding solutions. However, the coefficients of permeability of bentonite with various inorganic chemical permeants should be applied with caution si nce statistical confid ence levels are not available. More data are required to validate the reproducibility of th e test results before they can be used in design.

PAGE 203

184 The amounts of chemicals retained in the saturated bentonite clay during permeation can be calculated using electrical conductivity measurements of the effluent, as described in Chapter 5, and correlate we ll with the weight increase of the actual bentonite clay after at the end of the test An equivalent flow of five pore volume permeation is required in order to predict the maximum chemical retention within the bentonite clay during advection flow. The amou nts of chemical retained in the case of divalent permeants are higher than in monova lent permeants. If the total amount of chemicals in the total influent of any cont ainment can be calculated by the designer, the chemical outflux can be predicted by subtrac ting the total amounts of chemical retained in the bentonite clay during the first five pore volumes or so. The designer can also choose the thickness of the bent onite layer such that the re tention capacity meets certain performance limits or criteria. Diffusion profiles of various inorganic ch emicals can be simulated up to steadystate conditions and allow the calculation of mass flux and di ffusant concentration at any elapsed time and at any location within the bentonite clay barrier. By plotting the normalized concentration in terms of initial concentration of inorganic source chemical [C/Co] versus depth factor normalized with respect to total depth [d/do], the diffusion mass flux can be predicted for any concentratio n of source diffusants at any depth within the bentonite clay medium. The solution can be easily adapted to varying source concentrations since it can be eas ily programmed in a spreadsheet.

PAGE 204

185 REFERENCES 1. ASTM D 5890-02, “Standard Test Method for Swell Index of Clay Mineral Component of Geosynthetic Clay Liners,” Annual Book of ASTM Standards 2002, Vol. 04.13, American Society for Testi ng and Materials, Philadelphia, PA. 2. ASTM D 422-63 (2002), “Standard Test Me thod for Particle-Size Analysis of Soils,” Annual Book of ASTM Standards 2002, Vol. 04.08, American Society for Testing and Materials, Philadelphia, PA. 3. ASTM D854-02 (2002) “Standard Test Methods for Specific Gravity of Soil Solids by Water Pycnometer” Annual Book of ASTM Standards Vol. 04.08, American Society for Testing a nd Materials, Philadelphia, PA. 4. ASTM D 4318-00 (2000) “Standard Test Met hods for Liquid Limit, Plastic Limit, and Plasticity Index of Soils,” Annual Book of ASTM Standards Vol. 04.08, American Society for Testing a nd Materials, Philadelphia, PA. 5. D2434-68(2000) “Standard Test Method for Permeability of Granular Soils (Constant Head)”, Annual Book of ASTM Standards Vol. 04.08, American Society for Testing and Mate rials, Philadelphia, PA. 6. D5084-00e1 “Standard Test Methods for Measurement of Hydraulic Conductivity of Saturated Porous Materials Us ing a Flexible Wall Permeameter”, Annual Book of ASTM Standards Vol. 04.08, American Society for Testing and Materials, Philadelphia, PA. 7. BS 1377-1975 Test 2(A), “Determination of Liquid Limit, Preferred Method using Cone Penetrometers,” British Standard Institution 8. ASTM C837-99 (2003) Standard Test Met hod for Methylene Blue Index of Clay. 9. Alther, G., Evans, J., Fang, H., and Witmer, K. (1985), “Influence of Inorganic Permeants upon the Permeability of Bentonite,” Hydraulic Barriers in Soil and Rock, ASTM STP 874, A.I. Johnson et al., eds., ASTM, West Conshohocken, Pa., pp. 64-73.

PAGE 205

186 10. Anderson, D.C., Crawley, W., and Zabci k, J.D. (1985), “Effects of Various Liquids on Clay Soil : Bentonite Slurry Mixtures,” Hydraulic Barriers in Soil and Rock, ASTM STP 874, A.I. Johnson et al ., eds., ASTM, West Conshohocken, Pa., pp. 93-103. 11. Anderson, D., and Brown, K.W. (1981) “Organic Leachate Effects on the Permeability of Clay Liners,” in La ndfill disposal: H azardous Waste, Proceedings of the 7th Annual Research Symposium, Washington, D.C., U.S. Environmental Protection Agency, pp. 119-130. 12. Anderson, D., Brown, K.W., and Green, J. (1982), “Effects of Organic Fluids on the Permeability of Clay Soil Liners,” in Landfill disposal: Hazardous Waste, Proceedings of the 8th Annual Research Symposium Fort Mitchell, Kentucky, U.S. Environmenatl Protection Agency, pp. 179-190. 13. Ashmawy, A.K., El-hajji, D., Sotelo, N., and Muhammad, N. (2002), “Hydraulic Performance of Untreated and Polymer-Treated Benton ite in Inorganic Landfill Leachates”, Clays and Clay Minerals, Vol. 50, No. 5, pp. 546-552. 14. Barbour, S.L. and Fredlund, D.G. (1989) “Mechanisms of Osmotic Flow and Volume Change in Clay Soils,” Canadian Geotechnical Journal vol. 26, pp. 551562. 15. Beek, W.J., Muttzall, K.M.K., and Van Heuven, J.W. (1999), Transport Phenomena, 2nd edition, John Wiley and Sons Ltd., West Sussex, England. 16. Ben, R.H., Tessier, D., and Ben H. A.A., (2000), “Mineralogy of the <2 micron Fraction of Three Mixed-Layer Clays fr om Southern and Central Tunisia,” Clay Minerals, vol.35, no.2, pp.375-381. 17. Bennett, R.H., and Hulbert, M.H., (1986) Clay Microstructure, International Human Resources Development Corpor ation, Publishers, Boston, MA, USA. chapters 1, 2 & 3. 18. Benson, C.H. and Daniel, D.E. (1994a), “Minimum Thickness of Compacted Clay Liners: I. Stochastic Models,” Journal of Geotechnical Engineering ASCE, vol. 120, no. 1, pp. 129-152. 19. Benson, C.H. and Daniel, D.E., (1994b) “Minimum Thickness of Compacted Clay Liners: II. Analysis and Case Histories,” Journal of Geotechnical Engineering, ASCE, vol. 120, no. 1, pp. 153-172. 20. Benson, C.H., Zhai, H., and Wang, X. (1994), “Estimating Hydraulic Conductivity of Compacted Clay Liners,” Journal of Geotechnical Engineering, ASCE, vol. 120, no. 2, pp. 366-387.

PAGE 206

187 21. Benson, C.H., Gunter, J.A., Boutwell, G.P ., Trautwein, S.J., and Berzanskis, P.H., (1997), “Comparison of Four Methods to Assess Hydraulic Conductivity,” Journal of Geotechnical and Geoenvironmental Engineering ASCE, vol. 123, no. 10, pp. 929-937. 22. Bergaya, F, and Vayer, M. (1997), “CEC of clays; measurement by adsorption of a copper ethylenediamine complex,” Applied Clay Science vol.12, no.3, pp.275280. 23. Bogardi, I., Kelly, W.E., and Bardossy, A. (1989), “A Reliability Model for Soil Liners: Initial Design,” Journal of Geoetchnical Engineering ASCE, vol. 115, no. 5, pp. 658-669. 24. Bogardi, I., Kelly, W.E., and Bardossy, A. (1990), “Reliability Model for Soil Liner: Postconstruction,” Journal of Geoetchnical Engineering ASCE, vol. 116, no. 10, pp. 1502-1520. 25. Bourg, I.C, Bourg, A.C.M, Sposit o, G. (2003), “Modeling diffusion and adsorption in compacted bentonite; a critical review”, Journal of Contaminant Hydrology vol. 61, no. 1-4, pp. 293-302. 26. Boynton, S.S. and Daniel, D.E. ( 1985), “Hydraulic Conductivity Tests on Compacted Clay,” Journal of Geotechnical Engineering ASCE, vol. 111, no. 4, pp. 465-478. 27. Bradbury, M. H. and Baeyens, B. (2003) “Porewater Chemistry in Compacted Re-saturated MX-80 bentonite,” Journal of Contaminant Hydrology vol.61, no.1-4, pp.329-338. 28. Broderick, G.P. and Daniel, D.E. (1990) “Stabilizing Compacted Clay Against Chemical Attack,” Journal of Geotechnical Engineering, ASCE, vol. 116, no. 10, pp. 1549-1567. 29. Brownlow, A.H. (1970), Geochemistry, Pr entice-Hall, Inc., Englewood Cliffs, New Jersey. 30. Bruno, H. (2002), “Geosynthetic Clay Liners Offer Many Advantages for Containment Applications ”, http://www.esemag.com/0102/clay.html, Accessed 01/17/2004. 31. Bujdak, J., Janek, M., Madejova, J., Komadel, P. (2001), “Methylene Blue Interactions with Reduced-Charge Smectites,” Clays and Clay Minerals vol.49, no.3, pp.244-254, Jun 2001.

PAGE 207

188 32. Cadena, F., Rizvi, R., and Peters, R. W. (1990), “Feasibility Studies for the Removal of Heavy Metals from So lution Using Tailored Bentonite,” Hazardous and Industrial Wastes Hazardous and Industr ial Wastes Proceedings of the MidAtlantic Industrial Waste Conference. Publ by Technomic Publ Co Inc, Lancaster, PA, USA. p 77-94. 33. Canty, G.A., Atalay, A., Laguros, J.G., Robertson, J., and Pandey, K. K. (1995), “A Preliminary Assessment of Utilizing Fluidized Bed Ash in Landfill Liner Applications,” Journal of Environmental Science and health, Part A: Environmental Science & Engineer ing and Toxic & Hazardous Substance Control, vol. 30, no. 2, Feb, pp. 439-459. 34. Cases, J.M., Berend, I., Francois, M., Urio t, J.P., Michot, L.J., Thomas, F. (1997), “Mechanism of Adsorption and Desorption of Water Vapor by Homoionic Montmorillonite; 3, The Mg2+, Ca2+, and Ba3+ exchanged forms,” Clays and Clay Minerals, vol.45, no.1, pp.8-22. 35. Chapman, H.D. (1965), “Cation Exchange Capacity,” In Methods of Soil Analysis – Part 2, American Society of Agronomy Inc., Madison, Wisconsin. 36. Chapuis, R.P. (1990), “Sand-Bentonite Liners: Permeability from Laboratory Tests,” Canadian Geotechnical Journal vol. 27, pp. 47-57. 37. Chapuis, R.P. (2002), “The 2000 R.M. Hardy Lecture: Full-Scale Hydraulic Performance of Soil-Bentonite a nd Compacted Clay Liners,” Canadian Geotechnical Journal, vol. 39, pp. 417-439. 38. Chen, K.Y. and Bowerman, F.R. (1974) “Mechanisms of Leachate Formation in Sanitary Landfills, in Recycling and Di sposal of Solid Wastes: Industrial Agricultural, Domestic,” Yen, T.F. E d.;56 Ann Arbor Science Pub., Ann Arbor. 39. Cheung, S.C.H. (1994), “Modeling of Inorga nic Contaminant Transport in Dense Bentonite,” Journal of Soil Contamination vol. 3, no. 2, pp. 137-157. 40. Cheung, S.C.H. and Gray, M.N. (1989), “Mechanisms of Ionic Diffusion in Dense Bentonite,” Proceedings of Scientific Basis for Nuclear Waste Management XII symposium, Materials Research Society, Pittsburgh, Pennsylvania, vol. 127, pp. 677-681. 41. Cheung, S.C.H., Oscarson, D.W., and Lopez, R.S. (1984), “Factors Influencing Mass Diffusion in Bentonite and Mixt ures of Bentonite and Sand,” Materials Research Society Symposium, Boston, MA, vol. 26, pp. 711-718.

PAGE 208

189 42. Chmielova, M., Malac, Z., Merinska, D., Capkova, P., and Weiss, Z., (2000), “XRD analysis and modelling of Na -montmorillonite intercalated with octadecylamine,” Sixteenth conference on Clay mineralogy and petrology in Karlovy Vary (Czech Republic) Acta Univ ersitatis Carolinae. Geologica, vol.44, no.2-4, pp.91-93. 43. Christidis, G.E. (2001), “Formation and Growth of Smectites in Bentonites; a Case Study from Kimolos Island, Aegean, Greece,” Clays and Clay Minerals vol.49, no.3, pp.204-215. 44. Christidis, G.E. (1998), “Physical and Ch emical Properties of Some Bentonite Deposits of Kimolos Island, Greece,” Applied Clay Science vol.13, no.2, pp.7998. 45. Crank, J. (1975), The Mathematics of Diffusion, 2nd edition, Clarendon Press Oxford, England. 46. du Plessis, J.P. and Ross, L.I. (1993), “Permeability Predicti on for Water Seepage Through Low Porosity Granular Porous Media,” Water South Africa, vol. 19, no. 2, Apr. pp. 147-152. 47. Daniel, D.E. (1984), “Predicting Hydrau lic Conductivity of Clay Liners,” Journal of Geotechnical Engineering ASCE, vol. 110, no. 2, pp. 285-300. 48. Daniel, D.E. (1985), “Predicting Hydrauli c Conductivity of Clay Liners,” Reply. Journal of Geotechnical Engineering ASCE, vol. 111, no. 12, pp. 1466-1467. 49. Daniel, D.E. (1989), “In Situ Hydrau lic Conductivity Tests for Compacted Clays,” Journal of Geotechnical Engineering ASCE, vol. 115, no. 9, pp. 12051227. 50. Daniel, D.E. (1990), “A Rati onal Basis for Determining Safety of Containment,” Proceedings of the ASCE 17th Annual National Conference on Optimizing the Resources for Water Management Forth Worth, TX, pp. 489-493. 51. Daniel, D.E., et al. (1984), “Perm eability Testing with Flexible-Wall Permeameters,” Geotechnical Testing Journal vol. 7, no. 3, pp. 113-122. 52. Daniel, D.E. (1993), “Clay Liners,” Geotechnical Prac tice for Waste Disposal, D.E. Daniel, ed., Chapman & Hall London, England, pp. 137-163. 53. Daniel, D.E. (1994), “State-of-the-Art: Laboratory Hydraulic Conductivity Tests for Saturated Soils,” In Hydraulic Conductivity and Waste Contaminant Transport ASTM, STP 1142, D.E. Daniel and S. J. Trautwein, eds., ASTM, West Conshohocken, Pa., pp. 30-78.

PAGE 209

190 54. Daniel, David E., (1995), “Soil Barrier La yers Versus Geosynthetic Barriers in Landfill Cover Systems,” Proceedings of the 1995 Conference of the Geotechnical Engineering Division of ASCE in conjunction with the ASCE Convention, San Diego, CA, Geotechnical Sp ecial Publication, no. 53, pp. 1-18. 55. Daniel, D.E. and Benson, C.H. (1990): “W ater Content-Density Criteria for Compacted Soil Liners”, Journal of Geotechnical Engineering, ASCE vol. 116, no. 12, pp. 1811-1830. 56. Day, S.R. and Daniel, D.E. (1985), “H ydraulic Conductivity of Two Prototype Clay Liners,” Journal of Geotechnical Engineering ASCE, vol. 111, no. 8, pp. 957-970. 57. Dorsey, J.D., Ward, A.D., Fausey, N.R., and Blair, E.S. (1990), “A Comparison of Four Methods for Measuring Saturated Hydraulic Conductivity,” Transactions of the American Society of Agricultural Engineers (ASE), vol. 33, no. 6, pp. 19251931. 58. Drever, J.I. (1982), The Geochemistry of Natural Waters, Prentice-Hall, Inc., Englewood Cliffs, N.J. 59. Edil, T.B. and Berthouex, P.M. (1990), “Earthen Barriers Technology For Waste Containment”, Journal of Waste Management vol. 10, pp. 147-153. 60. Edil, T. B., Sandstrom, L. K., and Berthouex, P.M. (1992), “Interaction of Inorganic Leachate with Compacted Pozzolanic Fly Ash”, Journal of Geotechnical Engineering vol. 118, no. 9, September 1992, pp. 1410-1430. 61. Egloffstein, T. (1995), “Properties and Test s Methods to Assess Bentonite Used in Geosynthetic Clay Liners,” Geosynthetic Clay Liners Balkema, Rotterdam, The Netherlands, pp. 51-72. 62. Endo, T., Yamamoto, S., Honna, T., En eji, A.E. (2002), “Sodium-Calcium Exchange Selectivity as Influenced by Clay Minerals and Composition”, Soil Science vol. 167, no. 2, pp.117-125. 63. Eriksen, T.E., Jansson, M., and Molera, M. (1999), “Sorption Effects on Cation Diffusion in Compacted Bentonite,” Journal of Engineering Geology, vol. 54, pp. 231-236. 64. Faure, G. (1998), Principles and Applications of Geochemistry, 2nd edition, Prentice-Hall Inc., New Jersey.

PAGE 210

191 65. Fernandez, F. and Quigley, R.M. (1986) “Organic Liquids and the Hydraulic Conductivity of Barrier Clays,” Proceedings of the International Conference on Soil Mechanics and Foundation Engineering June 22-26, Cambridge, MA, pp. 1867-1870. 66. Fernandez, F. and Quigley, R.M. (1985), “Hydraulic Conductivity of Natural Clays Permeated with Simple Liquid Hydrocarbons,” Canadian Geotechnical Journal vol. 22, pp. 205-214. 67. Fletcher, P. and Sposito, G. (1989), “The Chemical Modeling of Clay/Electrolyte Interactions for Montmorillonite,” Journal of Clay Minerals vol. 24, pp. 375-391. 68. Fletcher, P., Sposito, G., and LeVe sque, C.S. (1984), “Sodium-CalciumMagnesium Exchange Reactions on a Montmo rillonitic Soil: I. Binary Exchange Reactions,” Soil Sci Soc Am J vol. 48 no. 5 Sep-Oct 1984. p 1016-1021. 69. Foose, G. J. (2002), “Transit-Time Design for Diffusion Through Composite Liners,” Journal of Geotechnical and Geoe nvironmental Engineering, ASCE, vol. 128, no. 7, pp. 590-601. 70. Freeze, R.A. and Cherry, J.A. (1979), Groundwater, Prentice-Hall, Inc ., Englewood Cliffs, N.J. USA. 71. Gleason, M.H., Daniel, D.E., and Eykholt, G.R., (1997), “Calcium and Sodium Bentonite for Hydraulic Cont ainment Applications,” Journal of Geotechnical and Geoenvironmental Engineering, ASCE, vol. 123, no. 5, pp. 438-445. 72. Goldstein, J.I., Newbury, D.E., Echlin, P., Joy, D.C., Fiori, C., and Lifshin, E. (1981) Scanning Electron Microscopy and X-Ray Microanalysis, Plenum Press. 73. Gordon, B.B. and Forrest, M. (1981), “Per meability of Soils Using Contaminated Permeant,” In Permeability and Groundwater Contaminant Transport ASTM, Special Technical Publication, STP 746, pp. 101-120. 74. Grim, R.E. (1968), Clay Mineralogy, 2nd edition, McGraw-Hill Book Company 75. Grim, R.E. and Guven, N. (1978), Bent onites: Geology, Mineralogy, Properties, and Uses, Elsevier Science Publis hing Co., Inc., New York, N.Y. 76. Grube, W.E. J. (1990), “Measuring Perfor mance of Clay Containment Barriers,” Proceedings of the ASCE 17th Annula National Conference on Optimizing the Resources for Water Management Forth Worth, TX, pp. 482-488.

PAGE 211

192 77. Guillaume, D., Neaman, A., Cathelineau, M., Mosser, R.R., Peiffert, C., Abdelmoula, M., Dubessy, J., Villieras, F ., Baronnet, A., and Michau, N., (2003), “Experimental Synthesis of Ch lorite from Smectite at 300oC in the Presence of Metallic Fe,” Clay Minerals vol.38, no.3, pp.281-302. 78. Gngr, N. and Ece, .I. (1999), “Effect of the Adsorption of Non-Ionic Polymer Poly(vinyl) Pyrolidone on the Rheological Properties of Na-Activated Bentonite,” Materials Letters, vol 39, pp. 1-5. 79. Gymer, R.G. (1973), Chemistr y: an Ecological Approach, Harper & Row, Publishers New York. 80. Hajjaji, M., Kacim, S., Alami, A., El B. A., and El M.M. (2001), “Chemical and Mineralogical Characterization of a Clay taken from the Moroccan Meseta and a Study of the Interaction Between its Fine Fraction and Methylene Blue”, Applied Clay Science vol.20, no.1-2, pp.1-12. 81. Hajra, M.G., Reddi, L.N., Glasgrow, L. A., Xiao, M., and Lee, I.M. (2002), “Effects of Ionic Strength on the Fine Particle Clogging of Soil Filters,” Journal of Geotechnical and Geoenvironmental Engineering, ASCE vol. 128, no. 8, Aug., pp. 631-639. 82. Hang, P.T. and Brindley, G.W. (1970), “Methylene Blue Absorption by Clay Minerals Determination of Surface Areas and Cation Exchange Capacities (ClayOrganic Studies),” Clays and Clay Minerals vol. 18, pp. 203-212. 83. Hermann, J.G. and Elsbury, B.R. (1987), “I nfluential Factors in Soil Practice for Waste Disposal ’87,” edited by R. D. Woods, ASCE, NY, pp. 522-536. 84. Hettiaratchi, J.P.A., Achari, G., Joshi, R. C., and Okoli, R.E. (1999), “Feasibility of Using Ash Admixtures in Landfill Bo ttom Liners or Vertical Barriers at Contaminated Sites,” Journal of Environmental Science and Health, Part A: Toxic/Hazardous Substances and Environmental Engineering, vol 34, n 10, pp. 1897-1917. 85. Higashi, K., Yamaguchi, T., Takabatake, M. N., and Fujita, H. (1990), “Diffusion of Water in Bentonite,” Memoirs of the Faculty of Engineering, Kyoto University vol. 52, no. 2, pp. 106-113. 86. Higgs, N.B. (1988), “Methylene Blue Ad sorption as a Rapid and Economical Method of Detecting Smectite,” Geotechnical Testing Journal vol. 11, no. 1, pp. 68-71. 87. Holtz, R.D. and Kovacs, W.D. (1981), An Introduction to Geotechnical Engineering, Prentice-Hall, Inc ., New Jersey.

PAGE 212

193 88. Holtz, W.G. (1985), “Predicting Hydrau lic Conductivity of Clay Liners,” Discussion: Journal of the Ge otechnical Engineering Division ASCE, vol. 111, no. 12, pp. 1457-1459. 89. Hwang, J.Y. and Dixon, J.B., (2000), “Flo cculation Behavior and Properties of Na-Montmorillonite Treated with Four Organic Polymers,” Clay Science vol. 11, no. 2, pp.137-146. 90. Itami, K. and Tamamura, T. (1999), “Evalu ation of Cation Exchange Capacity of Montmorillonite using Caesium Chloride,” Clay Science vol. 10, no. 6, pp. 469476. 91. James, A.N., Fullerton, D., and Drake, R. (1997), “Field Performance of GCL under Ion Exchange Conditions,” Journal of Geotechnical and Geoenvironmental Engineering ASCE, vol. 123, no. 10, pp. 897-901. 92. Jo, H.Y., Katsumi, T., Benson, C.H., and Edil, T.B. (2001), “Hydraulic Conductivity and Swelling of Nonprehydrated GCLs Permeated with SingleSpecies Salt Solutions,” Journal of Geotechnical and Geoenvironmental Engineering ASCE, vol. 127, no. 7, pp. 557-567. 93. Johnson, G.W., Crumley, W.S., and Boutwell, G.P. (1994), “Field Verification of Clay Liner Hydraulic Conductivity,” In Hydraulic Conductivity and Waste Contaminant Transport in Soils edited by D.E. Daniel and S. Trautewein, ASTM, Special Technical Publication STP 1142, pp. 226-245. 94. Joshi, R.C., Hettiaratchi, J.P.A., and Achari, G. (1994 ), “Properties of Modified Alberta Fly Ash in Relation to Utilizati on in Waste Management Applications”, Canadian Journal of Civil Engineering, vol. 21, no. 3, Jun 1994, pp. 419-426. 95. Jurcek, P., Krizova, V. J., Ivanova, P ., and Aguete, E.C. (1999), “Study of Sorption and Diffusion Processe s in Natural Bentonites,” Acta Universitatis Caroliae Geologica, vol. 43, no. 3, pp. 581-585. 96. Kacimov, A.R. and Obnosov, Y.V. (2000), “Two-Dimensional Seepage in Porous Media with Heterogeneities,” Journal of Geochemical Exploration, vol. 69, pp. 251-255. 97. Kahr, G. and Madsen, F. T. (1995), “Det ermination of Cation Exchange Capacity and the Surface Area of Bentonite. Illit e, and Kaolinite by Methylene Blue Adsorption,” Applied Clay Science vol. 9, pp. 327-336. 98. Kemper, W.D., Maasland, D.E.L., and Po rter, L.K. (1964), “Mobility of Water Adjacent to Mineral Surfaces,” Proceedings of Soil Science Society of America vol. 28, no. 2, pp. 164-167.

PAGE 213

194 99. Kaufhold, S., Dohrmann, R., Ufer, K., and Meyer, F. M. (2002), “Comparison of methods for the quantification of montmorillonite in bentonites,” Applied Clay Science vol. 22, no.3, pp.145-151. 100. Kayabali, K. (1997), “Engineering Aspect s of a Novel Landfill Liner Material: Bentonite-Amended Natural Zeolite,” Journal of Engineering Geology, vol. 46, pp. 105-114. 101. Keijer, T.J.S. (2000), “Chemical Osmo sis in Natural Clayey Materials,” Geologica Ultraiectina, 1 996, Ph.D. Thesis, Universiteit Utrecht, The Netherlands, pp. 166. 102. Keijzer, T.J.S and Loch, J.P.G. ( 2001), “Chemical Osmosis in Compacted Dredging Sludge,” Soil Science Society of America Journal vol. 65, no. 4, pp.1045-1055. 103. Keijzer, T.J.S., Kleingeld, P.J., and Loc h, J.P.G. (1999), “Chemical Osmosis in Compacted Clayey Material and the Prediction of Water Transport,” Journal of Engineering Geology, vol. 53, pp. 151-159. 104. Keren, R. and Singer, M.J. (1988), “Effect of Low Electrolyte Concentration on Hydraulic Conductivity of Sodium/Calci um-Montmorillonite-Sand System,” Soil Science Society of America Journal vol. 52, no. 2, pp. 368-373. 105. Kim, H.T., Suk, T.W., and Parks, S.H. (1993), “Diffusivities for Ions Through Compacted Na-Bentonite with Varying Dry Bulk Density,” Waste Management vol. 13, pp. 303-308. 106. King, K.S., Quigley, R.M., Fernandez, F ., Readers, D.W., and Bacopoulos, A., (1993), “Hydraulic Conductivity and Diffu sion Monitoring of the Keele Valley Landfill Liner,” Maple, Ontario, Canadian Geotechnical Journal vol. 30, pp. 124-134. 107. Kitsopoulos, K.P. (1997), “Comparison of the Methylene Blue Absorption and the Ammonium Acetate Saturati on Methods for Determination of CEC Values of Zeolite-Rich Tuffs”, Clay Minerals vol. 32, no. 2, pp.319-322. 108. Kjellander, R., Marcelja, S., and Qu irk, J. (1988), “Attractive Double-Layer Interactions Between Calcium Clay Particles,” Journal of Colloid and Interface Science vol. 126, no. 1, pp. 194-211. 109. Koch, D. (2002), “Bentonites As A Basic Material For Technical Base Liners And Site Encapsulation Cut-Off Walls”, Applied Clay Science vol. 21, no.1-2, pp.1-11.

PAGE 214

195 110. Koerner, R.M. (1999), Designing with Geosynthetics, 4th edition, Prentice-Hall, Inc. Upper Saddle River, N.J. 111. Koerner, R.M. and Soong, T.Y. (2000), “Leachate in Landfills: The Stability Issues,” Journal of Geotextiles and Geomembranes, vol. 18, pp. 293-309. 112. Kster, H. M. (1996), “Mineralogical a nd Chemical Heterogeneity of Three Standard Clay Mineral Samples, ” Clay Minerals, vol. 31, pp. 417-422. 113. Kozaki, T., Inada, K., Sato, S., and Ohashi, H. (2001), “Diffusion Mechanism of Chloride Ions in Sodium Montmorillonite,” Journal of Contaminant Hydrology vol.47, no.2-4, pp.159-170. 114. Lake, C.B., and Rowe, R.K. (2000), “Di ffusion of Sodium and Chloride Through Geosynthetic Clay Liners,” Journal of Geotextiles and Geomembranes, vol. 18, pp. 103-131. 115. Lehikoinen, J., Carlsson, T., Muurinen, A., Olin, M., and Salonen, P. (1996), “Evaluation of Factors Affecting Diffusi on in Compacted Bentonite,” Scientific Basis for Nuclear Waste Management XIX Materials Research Society Symposium Proceedings vol 412, Materials Research Society,USA., pp. 675-682. 116. Lambe, T.W. (1953), “The Stru cture Of Inorganic Soil,” Proceedings of the American Society of Civil Engineers vol. 79, no. 315, 49 pp. 117. Li, Y.H. and Gregory, S. (1974), “Diffusion of Ions in Sea Water and in Deep-sea Sediments,” Geochimica et Cosmochimica Acta ., vol 38, no. 5, pp. 703-714. 118. Lide, D.L. (2002), CRC Handbook of Chemistry and Physics, 82nd ed, CRC Press LLC., FL. 119. Lin, L.C. and Benson, C.H. (2000), “Effect of Wet-Dry Cycling on Swelling and Hydraulic Conductivity of GCLs,” Journal of Geotechnical and Geoenvironmental Engineering, ASCE, vol. 126, no. 1, pp. 40-49. 120. Lu, J.C.S. et al., Ed. (1985), Leachate from Municipal Landfills, Production and Management, Noyes Pub., Park Ridge. 121. Mahler, C.F. and Velloso, R.Q. (2001) “Diffusion and Sorption Experiments Using a DKS Permeameter,” Journal of Engineering Geology, vol. 60, pp. 173179. 122. Malusis, M.A. and Shackelford, C.D. (2004), “Predicting Solute Flux through a Clay Membrane Barrier,” Journal of Geotechnical and Geoenvironmental Engineering vol. 130, no. 5, pp. 477-487.

PAGE 215

196 123. Malusis, M.A. and Shackelford, C.D. (2002a), “Chemico-Osmotic Efficiency of a Geosynthetic Clay Liner,” Journal of Geotechnical and Geoenvironmental Engineering, vol. 128, no. 2, pp. 97-106. 124. Malusis, M.A. and Shackelford, C.D. (2002b), “Coupling Effects During SteadyState Solute Diffusion Through A Se mipermeable Clay Membrane,” Environmental Science and Technology, vol. 36, pp. 1312-1319. 125. Malusis, M.A., Shackelford, C.D., and Olsen, H.W. (2001), “A Laboratory Apparatus to Measure Chemico-Osmotic Effi ciency Coefficients for Clay Soils,” Geotechnical Testing Journal vol. 24, no. 3, pp. 229-242. 126. Mayayo, M. J., Bauluz, B., and Gonzalez, L.J.M. (2000), “Variations in the chemistry of smectites from the Calatayud Basin (NE Spain),” Clay Minerals vol. 35, no. 2, pp.365-374. 127. McBean, E.A., Rovers, F.A., and Far quhar, G.J. (1995), Solid Waste Landfill Engineering and Design.Prentice, Hall PTR, Englewood Cliffs. 128. McBride, M.B. (1994), Environmental Chemistry of Soils, Oxford University Press, New York, Oxford. 129. McNeal, B.L. and Coleman, N.T. (1966), “Effect of Solution Composition on Soil Hydraulic Conductivity,” Proceedings of Soil Science Society of America vol. 30, pp. 308-312. 130. McNeal, B.L., Norvell, W.A., and Coleman, N.T. (1966), “Effect of Solution Composition on the Swelling of Extracted Soil Clays,” Journal of Soil Science Society of America vol. 30, pp. 313-317. 131. Mesri, G. and Olson, R..E. (1971), “Mechan isms Controlling the Permeability of Clays,” Clays and Clay Minerals vol. 19, no. 3, pp. 151-158. 132. Miller, W.L., Townsend, T., Earle, J., Lee, H., and Reinhart, D.R. (1994) “Leachate Recycle and the Augmentati on of Biological Decomposition at Municipal Solid Waste Landfills,” Presented at the Second Annual Research Symposium, Florida Center for Solid and Hazardous Waste Management Tampa, FL. 133. Mitchell, J.K. (1993), Fundamentals of Soil Behavior, John Wiley and Sons, Inc ., New York, N.Y.

PAGE 216

197 134. Mitchell, J.K. and Madsen, F.T. (1987), “Chemical Effects on Clay Hydraulic Conductivity,” Geotechnical Practice foe Waste Disposal ’87, Proceedings of Specialty Conference Ann Arbor, Michigan, June 15-17, Geotechnical Special Publication, no. 13, pp. 87-116. 135. Muhammad, N. and Ashmaw y, A.K, (2003), “Compatibil ity of Incinerator AshSoil Mix as an Alternative Material for Landfill Liners and Covers,” report submitted to Florida Center For Solid And Hazardous Waste Management Gainesville, FL, pp. 72. 136. Mundell, J.A. (1985), “Predicting Hydr aulic Conductivity of Clay Liners,” Discussion: Journal of Geotechnical Engineering ASCE, vol. 111, no. 12, pp. 1459-1464. 137. Mundell, J.A. and Bailey, B. (1985), “The Design and Testing of a Compacted Clay Barrier Layer to Limit Percolation Through Landfill Covers,” In Hydraulic Barriers in Soil and Rock edited by A.I. Johnson, R.K. Frobel, N.J. Cagalli, and C.B. Petterson, ASTM, Special Tec hnical Publication, STP 874, pp. 246-262. 138. Nakashima, Y. (2003), “Diffusivity Measurement of Heavy ions in Wyoming Montmorillonite Gels by X-ray Computed Tomography,” Journal of Contaminant Hydrology vol.61, no.1-4, pp.147-156. 139. Nordquist, J.E. (1990), “Comparison of La boratory and Field Measured Hydraulic Conductivities of Soil Liners,” Proceedings of the 1990 Annual Symposium on Engineering Geology and Geotechnical Engineering no. 26, April 4-6, Pocatello, Idaho, pp. 11/1-11/5. 140. Odom, I.E. (1984), “Smectite Clay Minerals; Properties and Uses,” Philosophical Transactions of the Royal Society of London, Series A: Mathematical and Physical Sciences vol.311, no.1517, pp.391-409. 141. Ogata, A. (1970), “Theory of Dispersion in a Granular Medium, U.S. Geological Survey Professional Paper, no. 411-1. 142. Olsen, S.R., Kemper, W.D., and van Schaik, J.C. (1965), “Self-Difusion Coefficients of Phosphorous in Soil M easured by Transient and Steady-State Methods,” Proceedings of Soil Science Society of America vol. 29, no. 2, pp. 154158. 143. Peterson, S.R., and Gee, G.W. (1985), “Int eractions between Acidic Solutions and Clay Liners: Permeability and Neutralization,” Hydraulic Barriers in Soil and Rock ASTM STP 874, A.I. Johnson et al., eds., ASTM, West Conshohocken, Pa., pp. 229-245.

PAGE 217

198 144. Petit, S., Righi, D., Madejova, J., a nd Decarreau, A. (1998), “Layer Charge Estimation of Smectites using Infrared Spectroscopy,” Clay Minerals vol. 33, no. 4, pp. 579-591. 145. Petrov, R.J. and Rowe, R.K. (1997), “Geos ynthetic Clay Liner (GCL) – Chemical Compatibility by Hydraulic Conductivity Testing and Factors Impacting Its Performance,” Canadian Geotechnical Journal vol. 34, pp. 863-885. 146. Petrov, R.J., Rowe, R.K., and Quigle y, R.M. (1997), “Selected Factors Influencing GCL Hydraulic Conductivity,” Journal of Geotechnical and Geoenvironmental Engineering ASCE, vol. 123, no. 8, pp. 683-695. 147. Picornell, M. (1985), “Predicting Hydr aulic Conductivity of Clay Liners,” Discussion: Journal of Geotechnical Engineering ASCE, vol. 111, no. 12, pp. 1464-1465. 148. Porbaha, A., Pradhan, T.B.S., and Yamane, N. (2000), “Time Effect on Shear Strength and Permeability of Fly Ash”, Journal of Energy Engineering, vol. 126, no. 1, April 2000, pp. 15-31. 149. Porter, L.K., Kemper, W.D., Jackson, R.D ., and Stewart, B.A. (1960), “Chloride Diffusion in Soils as Influenced by Moisture Content,” Proceedings of Soil Science of America vol. 24, no. 6, pp. 400-403. 150. Qasim, S.R. and Chiang, W. (1994) Sa nitary Landfill Leachate, Technomic Publishing Co., Inc., Lancaster. 151. Quigley, R.M. and Rowe, R.K. (1986), “Leachate Migration Through Clay Below a Domestic Waste Lanfill, Sarnia, Ontario, Canada: Chemical Interpretation and Modeling Philosophies,” Hazardous and Industrial Sol id Waste Testing and Disposal STP 933, Lorenzen, et al., (eds.) ASTM, Philadelphia, PA., pp. 93-103. 152. Quigley, R.M., Yanful, E.K., and Fern andez, F. (1987), “Ion Transfer by Diffusion Through Clay Barriers,” Geotechnical Practice for Waste Disposal A.S.C.E. Geotechnical Special Publication, vol. 13, pp. 137-158. 153. Quirk, J.P. and Schofield, R.K. (1955), “The Effect of Electro lyte Concentration on Soil Permeability,” Journal of Soil Science vol. 6, no. 2, pp. 163-178. 154. Rad, N.S., Jacobson, B.D., and Bachus R.C. (1994), “Compatibility of Geosynthetic Clay Liners with Or ganic and Inorganic Permeants,” Proceedings of the 5th International Conference on Geot extiles, Geomembranes and Related Products September 5-9, Singapore, pp. 1165-1168.

PAGE 218

199 155. Ramirez, S., Cuevas, J., Petit, S., Righ i, D., and Meunier, A. (2002), “Smectite Reactivity in Alkaline Solutions Geologi ca Carpathica (Bratislava),” vol. 53, no. 2, pp. 87-92. 156. Reinhart, D.R. and Grosh, C.J. (1998) “Analysis of Florida MSW Landfill Leachate Quality,” State University System of Fl orida, Florida Center for Solid and Hazardous Waste Management Gainesville, FL, Report #97-3, pp. 108. 157. Reschke, A.E. and Haug, M.D. (1991), “The Physico-Chemical Properties of Bentonites and the Performance of Sand-Bentonite Mixtures,” Proceedings of the 44th Annual Canadian Geotechnical Conference Calgary, Alberta, September 29-October 02, vol. 2, paper no. 62, pp. 62/1-62/10. 158. Rogowski, A.S. (1986), “Hydraulic Conduc tivity of Compacted Clay Soils,” Proceedings of the 12th Annual Research Symposium on Land Disposal, Remedial Action, Incineration, and Trea tment of Hazardous Waste U.S.EPA, Cincinnati, Ohio, pp. 29-39. 159. Rogowski, A.S. (1990), “Comparison Between the Field and Laboratory Measured Properties of a Clay Liner,” Proceedings of the 17th Annual National Conference on Optimizing the Re sources for Waste Management Forth Worth, TX, pp. 476-481. 160. Rowe, R.K. and Booker, J.R. (1985), “1-D Pollutant Migration in Soils of Finite Depth,” Journal of Geotechnical Engineering ASCE, vol. 111, no. 4, pp. 479499. 161. Rowe, R.K., Lake, C.B., and Petrov, R.J. (2000), “Apparatus and Procedures for Assessing Inorganic Diffusion Coefficients Through Geosynthetic Clay Liners,” ASTM Geotechnical Testing Journal vol. 23, pp. 206-214. 162. Rowe, R.K., Petrov, R.J., and Lake, C. (1997), “Compatibility Testing and Diffusion Through Geosyntheti c Clay Liners (GCL),” Proceedings of Sardinia 97, Sixth International Landfill Symposium Cagliari, Italy, October 13-17, pp. 301-310. 163. Rowe, R.K., Quigley, R.M., and Booker, J.R. (1995), “Clayey Barrier Systems for Waste Disposal Facilities,” E & FN Spon (Chapman and Hall) London. 164. Ruhl, J.L. and Daniel, D.L. (1997), “Geo synthetic Clay Liners Permeated with Chemical Solutions and Leacheates,” Journal of Geotechnical and Geoenvironmental Engineering, ASCE, vol. 123, no. 4, pp. 369-381.

PAGE 219

200 165. Russ, J.C. (1984), Fundamentals of En ergy Dispersive X-Ray Analysis, Butterworths & Co (Publishers) Ltd 166. Sai, J.O. and Anderson, D.C. (1991), “Sta te-of-the-Art: Field Hydraulic Tests for Compacted Soil Liners,” Geotechnical Testing Journal vol. 13, no. 3, pp. 215225. 167. Sllfors, G., and berg, H., and Anna, L. (2002), “Determination of Hydraulic Conductivity of Sand-Bentonite Mixtures for Engineering Purposes,” Journal of Geotechnical and Geological Engineering, The Netherlands, vol. 20, pp. 65-80. 168. Sanchez, A.G., Alastuey, A., and Querol X. (1999), “Heavy Metal Adsorption by Different Minerals: Application to th e Remediation of Polluted Soils,” The Science of the Total Environmental Journal, vol. 242, pp. 179-188. 169. Santamarina, J.C., Klein, K.A., Wang, Y. H., and Prencke, E. (2002), “Specific Surface: Determination and Relevance,” Canadian Geotechnical Journal vol. 39, no. 1, pp. 233-241. 170. Sato, H. (2000), “Effect of Ionic Charge on Effective Diffusion Coefficient in Compacted Sodium Bentonite,” Proceedings of Materials Research Society Symposium vol. 608, pp. 267-274. 171. Sato, H. and Suzuki, S. (2003), “Funda mentals Study on the Effect of an Orientation of Clay Particles on Diffusi on Pathway in Compacted Bentonite,” Applied Clay Science vol. 23, pp. 51-60. 172. Schaefer, M. and Steiger, M., (2002), “A Rapid Method for the Determination of Cation Exchange Capacities of Sa ndstones; Preliminary Data,” Geological Society Special Publications vol.205, pp.431-439. 173. Schulze, D.G. (1989), “An Introduction to Soil Mineralogy”, Dixon, J.B. and Weed, S.B. (eds.), In Minerals in Soil Environments Madison, WI, Soil Science Society of America, pp. 1-34. 174. Shackelford, C.D. (1988), “Diffusion of Inorganic Chemical Wastes in Compacted Clay,” thesis presented to the University of Texas, at Austin, TX, in partial fulfillment of the requirements for the degree of Doctor of Philosophy. 175. Shackelford, C.D. (1990), “Transit-t ime Design of Earthen Barriers,” Journal of Engineering Geology vol. 29, no. 1, pp. 79-94. 176. Shackelford, C.D. (1991), “Laboratory Di ffusion Testing for Waste Disposal – A Review,” Journal of Contaminant Hydrology vol. 7, pp. 177-217.

PAGE 220

201 177. Shackelford, C.D. (1994), “Waste Soil Interactions that alter Hydraulic Conductivity,” In Hydraulic Conductivity and Waste Contaminant Transport, ASTM STP 1142, D.E. Daniel and S.J. Trautwein, eds., ASTM, West Conshohocken, Pa., pp. 111-168. 178. Shackelford, C.D. and Daniel, D.E. (1991a ), “Diffusion in Saturated Soil, I: Background,” Journal of Geotechnical Engineering ASCE, vol. 117, no. 3, pp. 467-484. 179. Shackelford, C.D. and Daniel, D.E. ( 1991b), “Diffusion in Saturated Soil, II: Results for Compacted Clay,” Journal of Geotechnical Engineering ASCE, vol. 117, no. 3, pp. 485-506. 180. Shackelford, C.D. and Lee, J.M. (2003), “The Destruction Role of Diffusion on Clay Membrane Behavior,” Journal of Clays and Clay Minerals vol. 51, no. 2, pp. 186-196. 181. Shackelford, C.D. and Redmond, P.L. ( 1995), “Solute Breakthrough Curves for Processed Kaolin at Low Flow Rates,” Journal of Geotechnical Engineering ASCE, vol. 121, no. 1, pp. 17-32. 182. Shackelford, C.D., Malusis, M.A., and Stern, R.T. (1999), “Electrical Conductivity Breakthrough Curves,” Journal of Geotechnical and Geoenvironmental Engineering, ASCE, vol. 125, no. 4, pp. 260-270. 183. Shackelford, C.D., Benson, C.H., Katsumi, T., Edil, T.B., and Lin, L. (2000), “Evaluating the Hydraulic Conductivity of GCLs Permeated with Non-Standard Liquids,” Journal of Geotextile s and Geomembranes, vol. 18, pp. 133-161. 184. Singh, S. K., Baser, B. L., and Shyampur a, R. L. (2002), “Chemical composition and charge behaviour of smectites in Vertisols of Rajasthan,” Journal of the Indian Society of Soil Science vol.50, no.1, pp.106-110. 185. Snyder, K.A. (2001), “The Relationship Between the Formation Factor and the Diffusion Coefficient of Porous Mate rials Saturated with Concentrated Electrolytes: Theoretical and E xperimental Considerations,” Concrete Science and Engineering, vol. 3, pp. 216-224. 186. Song, K. and Sandi, G. (2001), “Character ization of montmor illonite surfaces after modification by organosilane,” Clays and Clay Minerals vol.49, no.2, pp.119-125. 187. Sposito, G. (1981), “The Thermodynamics of Soil Solutions”, Oxford University Press Oxford, London.

PAGE 221

202 188. Sposito, G. (1989), The Chemistry of Soils, Oxford University Press Oxford, London. 189. Sposito, G. and Fletcher, P. (1985), “S odium-Calcium-Magnesium Exchange Reactions on a Montmorillonite Soil : III. Calcium-Magnesium Exchnage Selectivity,” Soil Science Society of America Journal vol. 49, pp. 1160-1163. 190. Sivapullaiah, P.V. and Savitha, S., (1999) “Index Properties of Illite-Bentonite Mixtures in Electrolyte Solutions,” Geotechnical Testing Journa l, GTJODJ, vol. 22, no. 3, September pp. 257–265. 191. Stern, R.T. and Shackelford, C.D. (1998) “Permeation of Sand-Processed Clay Mixtures with Calcium Chloride Solutions,” Journal of Geotechnical and Geoenvironmental Engineering, ASCE, vol. 124, no. 3, pp. 231-241. 192. Stewart, J.P. and Nolan, T.W. (1987), “Infiltration Testing for Hydraulic Conductivity of Soil Liners,” Geotechnical Testing Journal vol. 10, no. 2, pp. 41-50. 193. Suzuki, S., Fujishima, A., Ueno, K., Ichi kawa, Y., Kawamura, K., Shibata, M., Sato, H., and Kitayama, K. (2001) “Mic rostructural Modeling of Compacted Sodium-Bentonite and Application of Unified Molecular Dynamics/Homogenization Analysis for Diffusion Process,” Journal of the Clay Science Society of Japan vol. 41, no. 2, pp.43-57. 194. Taylor, R.K. (1985), “Cation Exchange in Clays and Mudrocks by Methylene Blue,” Journal of Chemical Technology and Biotechnology Vol. 35A, pp. 195207. 195. Tissa H.I. (2004), Environmental Scie nce and Engineering and Center for Experimental Study of Subsurface Envir onmental Processes (CESEP), Colorado School of Mines, www.hsrc.org/hsrc/h tml/ssw/rio/tissa2.pdf, Accessed April 03, 2004. 196. Trautwein, S. and Boutwell, G. (1994), “In Soil Hydraulic Conductivity Tests for Compacted Liners and Caps,” In Hydraulic Conductivity and Waste Contaminant Transport in Soils edited by D.E. Daniel and S. Trautwein, ASTM, Special Technical Publication, STP 1142, pp. 184-226. 197. Triantafyllou, S., Christodoulou, E., a nd Neou-Syngouna, P. (1999), “Removal of Nickel and Cobalt from Aqueous Soluti ons by Na-Activated Bentonite,” Journal of Clays and Clay Minerals vol. 47, no. 5, pp. 567-572. 198. Van Olphen, H. (1977), Clay Colloid Chemistry, 2nd edition, John Wiley & Sons Inc., New York.

PAGE 222

203 199. Volzone, C., Rinaldi, J.O., and Or tiga, J., (2000), “Swelling of TMA (tetramethylammonium)and HDP (hexad ecylpyridinium) – Montmorillonites in Water and Toluene Media: Influence of the Type Montmorillonites,” Journal of Materails Research, vol. 3, no. 4, pp. 115-118. 200. Walter, C. E. (1976), “Practical Refuse Recycling,” Journal of Environmental Engineering Division, ASCE vol. 102, no. 1, pp. 139-148. 201. Warshaw, C.M. and Roy, R. (1961), “Classification And A Scheme For The Identification Of Layer Silicates”, Geological Society of America Bulletin vol. 72, no. 10, pp.1455-1492. 202. Weaver, C.E. and Pollard, L.D. (1975), The Chemistry of Clay Minerals, 1st edition, Elsevier Scientific Publishing Company, Amsterdam, The Netherlands. 203. Wen S., Yang D., and Chen J. (2001), “X-ray diffraction characterization of bentonite and acidation of bentonite, al kalization of bentonite, and surface characteristics under scanni ng electron microscope,” Acta Mineralogica Sinica vol. 21, no. 3, pp. 453-456. 204. Wentink, G.R. and Etzel, J.E. (1972), “Removal of Metal Ions by Soil,” Journal of Water Pollution Control Federation vol. 44, no. 8, pp. 1561-1574. 205. Xeidakis, G.S. (1996), “Stabilization of Swelling Clays by Mg(OH)2. Changes in Clay Properties after Add ition of Mg-hydroxide,” Journal of Engineering Geology, vol. 44, pp. 107-120. 206. Xiao, S., Moresoli, C., Bovenkamp, J., and Kee, D.D. (1997), “Sorption and Permeation of Organic Environmental Contaminants Through PVC Geomembranes,” Journal of Applied Polymer Science, vol. 63, no. 9, Feb 1997, pp. 1189-1197. 207. Zhang, F., Low, P., and Roth, C. (1995), “Effects of Monovalent Exchangeable Cations and Electrolytes on the Rela tion Between Swelling Pressure and Interlayer Distance in Montmorillonite,” Journal of Colloid and Interface Science., Academic Press New York, vol. 173, pp. 34-41. 208. Zimmie, T.F. (1981), “Geotech nical Test Considerations in the Determination of Laboratory Permeability for Hazardous Waste Disposal Siting”, American Society for Testing and Materials, Special Technical Publications 760. 1981. ASTM pp. 293-304.

PAGE 223

204 APPENDICES

PAGE 224

Appendix A: Test Results of pH and EC of Permeability Tests 205 (a) (b) Figure A.1 Ionic Analysis of Permeabil ity Test K-1 (a) Elec trical Conductivity vs. Pore Volume and (b) pH vs. Pore Volume 1.0E+02 1.0E+03 1.0E+04 1.0E+05 1.0E+06 0.01.02.03.04.05.06.0Pore volumeElectrical conductivity ( S/cm) .DI water pre-hydration (K-1) Influent (1M CaCl2) EC = 128,000 microS/cm Effluent 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 0.01.02.03.04.05.06.0Pore volumepH DI water pre-hydration (K-1) Influent (1M CaCl2) pH = 7.4 Effluent S/cm)

PAGE 225

Appendix A: (Continued) 206 (a) (b) Figure A.2 Ionic Analysis of Permeabil ity Test K-2 (a) Elec trical Conductivity vs. Pore Volume and (b) pH vs. Pore Volume 5.00 5.50 6.00 6.50 7.00 7.50 8.00 8.50 9.00 9.50 10.00 0.001.002.003.004.005.006.007.00Pore volumepH DI water pre-hydration (K-2) Influent (1M MgCl2) pH = 6.63 Effluent 1.0E+02 1.0E+03 1.0E+04 1.0E+05 1.0E+06 0.01.02.03.04.05.06.07.0Pore volumeElectrical conductivity ( S/cm) DI water pre-hydration (K-2) Influent (1M MgCl2) = 115,000 microS/cm 1M MgCl2 permeant S/cm)

PAGE 226

Appendix A: (Continued) 207 (a) (b) Figure A.3 Ionic Analysis of Permeabil ity Test K-3 (a) Elec trical Conductivity vs. Pore Volume and (b) pH vs. Pore Volume 5.00 5.50 6.00 6.50 7.00 7.50 8.00 8.50 9.00 9.50 10.00 0.001.002.003.004.005.006.007.00Pore volumepH DI water pre-hydration (K-3) Influent (1M KCl) pH = 7.1 Effluent 1.0E+02 1.0E+03 1.0E+04 1.0E+05 1.0E+06 0.01.02.03.04.05.06.07.0Pore volumeElectrical conductivity ( S/cm) DI water pre-hydration (K-3) Influent (1M KCl) =105,000 microS/cm 1M KCl permeant S/cm)

PAGE 227

Appendix A: (Continued) 208 (a) (b) Figure A.4 Ionic Analysis of Permeabil ity Test K-4 (a) Elec trical Conductivity vs. Pore Volume and (b) pH vs. Pore Volume 5.00 5.50 6.00 6.50 7.00 7.50 8.00 8.50 9.00 9.50 10.00 0.001.002.003.004.005.006.00Pore volumepH DI water pre-hydration (K-4) Influent (1M NaCl) pH = 7.35 Effluent 1.0E+03 1.0E+04 1.0E+05 1.0E+06 0.01.02.03.04.05.06.0Pore volumeElectrical conductivity ( S/cm) DI water pre-hydration (K-4) Influent (1M NaCl) = 86,000 microS/cm 1M NaCl permeation S/cm)

PAGE 228

Appendix A: (Continued) 209 (a) (b) Figure A.5 Ionic Analysis of Permeabil ity Test K-5 (a) Elec trical Conductivity vs. Pore Volume and (b) pH vs. Pore Volume 5.00 5.50 6.00 6.50 7.00 7.50 8.00 8.50 9.00 9.50 10.00 0.01.02.03.04.05.06.07.08.09.010.0Pore volumepH DI water pre-hydration (K-5) Influent (1M NaCl + 1M KCl + 1M CaCl2 + 1M MgCl2) pH = 7.2 Effluent 1.0E+03 1.0E+04 1.0E+05 1.0E+06 0.01.02.03.04.05.06.07.08.09.010.0Pore volumeElectrical conductivity ( S/cm) DI water pre-hydration (K-5) Influent (1M NaCl + 1M KCl + 1M CaCl2 + 1M MgCl2) = 270,000 microS/cm Effluent S/cm)

PAGE 229

Appendix A: (Continued) 210 (a) (b) Figure A.6 Ionic Analysis of Permeabil ity Test K-6 (a) Elec trical Conductivity vs. Pore Volume and (b) pH vs. Pore Volume 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 0.01.02.03.04.05.06.07.08.0Pore volumepH DI water pre-hydration (K-6) Influent (0.1M NaCl + 0.1M KCl + 0.1M CaCl2 + 0.1M MgCl2) pH = 7.15 Effluent 1.0E+03 1.0E+04 1.0E+05 0.01.02.03.04.05.06.07.08.0Pore volumeElectrical conductivity ( S/cm) DI water pre-hydration (K-6) Influent (0.1M NaCl + 0.1M KCl + 0.1M CaCl2 + 0.1M MgCl2) = 50,000 microS/cm Effluent S/cm)

PAGE 230

Appendix A: (Continued) 211 (a) (b) Figure A.7 Ionic Analysis of Permeabil ity Test K-7 (a) Elec trical Conductivity vs. Pore Volume and (b) pH vs. Pore Volume 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 0.01.02.03.04.05.06.07.08.0Pore volumepH DI water pre-hydration (K-7) Influent (0.01M NaCl + 0.01M KCl + 0.01M CaCl2 + 0.01M MgCl2) pH = 6.95 Effluent 1.0E+03 1.0E+04 1.0E+05 0.01.02.03.04.05.06.07.08.0Pore volumeElectrical conductivity ( S/cm) DI water pre-hydration (K-7) Influent (0.01M NaCl + 0.01M KCl + 0.01M CaCl2 + 0.01M MgCl2) = 7,500 microS/cm Effluent S/cm)

PAGE 231

Appendix A: (Continued) 212 (a) (b) Figure A.8 Ionic Analysis of Permeabil ity Test K-8 (a) Elec trical Conductivity vs. Pore Volume and (b) pH vs. Pore Volume 1.0E+03 1.0E+04 1.0E+05 1.0E+06 0.02.04.06.08.010.012.0Pore volumeElectrical conductivity ( S/cm) DI water pre-hydration (K-8) Influent (5M CaCl2) = 270,000 microS/cm Effluent for 5M CaCl2 permeant 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 0.02.04.06.08.010.012.0Pore volumepH DI water pre-hydration (K-8) Influent (5M CaCl2) pH = 8.2 Effluent S/cm)

PAGE 232

Appendix A: (Continued) 213 (a) (b) Figure A.9 Ionic Analysis of Permeabil ity Test K-9 (a) Elec trical Conductivity vs. Pore Volume and (b) pH vs. Pore Volume 5.00 5.50 6.00 6.50 7.00 7.50 8.00 8.50 9.00 9.50 10.00 020406080100120140160Pore volumepH 1M CaCl2 pre-hydration (K-9) Effluent for 1M CaCl2 permeant Effluent for 1M NaCl permeant Influent (1M CaCl2, pH = 6.95) & (1M NaCl, pH = 7.35) 1M CaCl21M NaCl 1.0E+04 1.0E+05 1.0E+06 020406080100120140160Pore volumeElectrical conductivity ( S/cm) 1M CaCl2 pre-hydration (K-9) Effluent for 1M CaCl2 permeant Effluent for 1M NaCl permeant Influent (1M CaCl2, EC = 132 mS/cm) & (1M NaCl, EC = 86 mS/cm) 1M CaCl21M NaCl S/cm)

PAGE 233

Appendix A: (Continued) 214 (a) (b) Figure A.10 Ionic Analysis of Permeabil ity Test K-10 (a) Elec trical Conductivity vs. Pore Volume and (b) pH vs. Pore Volume 5.00 5.50 6.00 6.50 7.00 7.50 8.00 8.50 9.00 9.50 10.00 0.005.0010.0015.0020.0025.00Pore volumepH 1M NaCl pre-hydration (K-10) Effluent for 1M NaCl permeant Effluent for 1M CaCl2 permeant Influent (1M NaCl, pH = 7.35) & (1M CaCl2, pH = 7.1) 1M NaCl 1M CaCl2 1.0E+04 1.0E+05 1.0E+06 0.05.010.015.020.025.0Pore volumeElectrical conductivity ( S/cm) 1M NaCl pre-hydration (K-10) Effluent for 1M NaCl permeant Effluent for 1M CaCl2 permeant Influent (1M NaCl, EC = 86 mS/cm) & (1M CaCl2, EC = 133 mS/cm) 1M NaC l 1M CaCl2 S/cm)

PAGE 234

Appendix A: (Continued) 215 (a) (b) Figure A.11 Ionic Analysis of Permeabil ity Test K-11 (a) Elec trical Conductivity vs. Pore Volume and (b) pH vs. Pore Volume 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 051015202530Pore volumepH DI water permeant all Salts (0.01M each) permeant all Salts (o.1M each) permeant all Salts (1M each) permeant Influent (DI, pH = 6.95; 0.01M, pH = 7.05; 0.1M, pH = 7.2; 1M, pH = 7.35) DI wate r all salts (0.01M) all salts (0.1M) all salts (1M) K-11 1.0E+02 1.0E+03 1.0E+04 1.0E+05 1.0E+06 051015202530Pore volumeElectrical conductivity ( S/cm) DI water permeant all Salts (0.01M each) permeant all Salts (0.1M each) permeant all Salta (1M each) permeant Influent (DI = 10 microS/cm, 0.01M = 7.2 mS/cm, 0.1M = 56 mS/cm, 1M = 270 mS/cm) DI water all salts (0.01M) all salts (0.1M) all salts (1M) K-11 S/cm)

PAGE 235

Appendix A: (Continued) 216 (a) (b) Figure A.12 Ionic Analysis of Permeabil ity Test K-12 (a) Elec trical Conductivity vs. Pore Volume and (b) pH vs. Pore Volume 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 0.05.010.015.020.025.030.035.0Pore volumepH 1M MgCl2 pre-hydration (K-12) Effluent for 1M MgCl2 permeant Effluent for 1M KCl permeant Influent (1M MgCl2, pH = 6.65; 1M KCl, pH = 7.05) 1M MgCl21M KCl 1.0E+04 1.0E+05 1.0E+06 0.05.010.015.020.025.030.035.0Pore volumeElectrical conductivity ( S/cm) 1M MgCl2 pre-hydration (K-12) Effluent for 1M MgCl2 permeant Effluent for 1M KCl permeant Influent (1M MgCl2 = 110 mS/cm, 1M KCl = 106 mS/cm) 1M MgCl21M KCl S/cm)

PAGE 236

Appendix A: (Continued) 217 (a) (b) Figure A.13 Ionic Analysis of Permeabil ity Test K-13 (a) Elec trical Conductivity vs. Pore Volume and (b) pH vs. Pore Volume 1.0E+04 1.0E+05 1.0E+06 0.05.010.015.020.025.0Pore volumeElectrical conductivity ( S/cm) 1M KCl pre-hydration (K-13) Effluent for 1M KCl permeant Effluent for 1M MgCl2 permeant Influent (1M KCl = 105 mS/cm; 1M MgCl2 = 109 mS/cm) 1M KCl 1M MgCl2 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 0.05.010.015.020.025.0Pore volumepH 1M KCl pre-hydration (K-13) Effluent for 1M KCl permeant Effluent for 1M MgCl2 permeant Influent (1M KCl, pH = 7.1; 1M MgCl2, pH = 6.65) 1M KCl 1M MgCl2 S/cm)

PAGE 237

Appendix A: (Continued) 218 (a) (b) Figure A.14 Ionic Analysis of Permeabil ity Test K-14 (a) Elec trical Conductivity vs. Pore Volume and (b) pH vs. Pore Volume 1.0E+03 1.0E+04 1.0E+05 1.0E+06 0.05.010.015.020.025.0Pore volumeElectrical conductivity ( S/cm) DI water pre-hydration (K-14) Effluent for 1M CaCl2 permeant Effluent for 1M MgCl2 permeant Influent (1M CaCl2 = 130 mS/cm; 1M MgCl2 = 110 mS/cm) 1M CaCl21M MgCl2 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 0.05.010.015.020.025.0Pore volumepH DI water pre-hydration (K-14) Effluent for 1M CaCl2 permeant Effluent for 1M MgCl2 permeant Influent (1M CaCl2, pH = 7.4; 1M MgCl2, pH = 6.65) 1M CaCl21M MgCl2 S/cm)

PAGE 238

Appendix B: Test Results of pH and EC of Diffusion Tests 219 (a) (b) Figure B.1 Diffusion Test Results for D-5 (a) pH and Electrical Conductivity and (b) Cumulative EC versus Cumulative Diffusion Time 0 50 100 150 200 250 300 0.05.010.015.020.025.030.035.040.045.0Cumulative diffusion time (day)Electrical Conductivity ( S/cm)0 1 2 3 4 5 6 7 8 9 10pH Electrical Conductivity pH D-5 1M NaCl diffusant 0 400 800 1,200 1,600 2,000 05101520253035404550Cumulative diffusion time (day)Cumulative EC ( S/cm) 1M NaCl diffusant e = 5.687 D-5 Thickness = 7.84 mm Air-dry weight = 15 g

PAGE 239

Appendix B: (Continued) 220 (a) (b) Figure B.2 Diffusion Test Results for D-6 (a) pH and Electrical Conductivity and (b) Cumulative EC versus Cumulative Diffusion Time 0 5,000 10,000 15,000 20,000 25,000 0102030405060708090100Cumulative diffusion time (day)Cumulative EC ( S/cm) 2M CaCl2 diffusant e = 9.23 D-6 Thickness = 8 mm Air-dry weight = 10 g 0 500 1,000 1,500 2,000 2,500 3,000 020406080100Cumulative diffusion time (day)Electrical Conductivity ( S/cm) .0 2 4 6 8 10 12pH Electrical Conductivity pH D-6 2M CaC l 2 diffusant

PAGE 240

Appendix B: (Continued) 221 (a) Figure B.3 Diffusion Test Results for D-8 (a) pH and Electrical Conductivity and (b) Cumulative EC versus Cumulative Diffusion Time 0 500 1,000 1,500 2,000 2,500 3,000 01020304050607080Cumulative diffusion time (day)Electrical Conductivity ( S/cm) .0 1 2 3 4 5 6 7 8 9 10pH Electrical Conductivity pH D-8 2M MgCl2 diffusant 0 5,000 10,000 15,000 20,000 25,000 0102030405060708090Cumulative diffusion time (day)Cumulative EC ( S/cm) 2M MgCl2 diffusant e = 14.35 D-8 Thickness = 3 mm Air-dry weight = 2.5 g(b)

PAGE 241

Appendix B: (Continued) 222 (a) (b) Figure B.4 Diffusion Test Results for D-9 (a) pH and Electrical Conductivity and (b) Cumulative EC versus Cumulative Diffusion Time 0 400 800 1,200 1,600 2,000 2,400 2,800 3,200 3,600 4,000 020406080100Cumulative diffusion time (day)Electrical Conductivity ( S/cm) .0 2 4 6 8 10 12pH Electrical Conductivity pH D-9 2M KCl diffusant 0 5,000 10,000 15,000 20,000 25,000 0102030405060708090100Cumulative diffusion time (day)Cumulative EC ( S/cm) 2M KCl diffusant e = 14.35 D-9 Thickness = 3 mm Air-dry weight = 2.5 g

PAGE 242

Appendix B: (Continued) 223 Figure B.5(a) pH and Electr ical Conductivity for D-10 (a) (b) Figure B.5 Diffusion Test Results for D10 (a) pH and Elec trical Conductivity and (b) Cumulative EC versus Cumulative Diffusion Time 0 500 1,000 1,500 2,000 2,500 3,000 3,500 4,000 020406080100Cumulative diffusion time (day)Electrical Conductivity ( S/cm) .0 2 4 6 8 10 12pH Electrical Conductivity pH D-10 2M NaCl diffusant 0 5,000 10,000 15,000 20,000 25,000 30,000 0102030405060708090100Cumulative diffusion time (day)Cumulative EC ( S/cm) 2M NaCl diffusant e = 14.35 D-10 Thickness = 3 mm Air-dry weight = 2.5 g

PAGE 243

Appendix B: (Continued) 224 (a) (b) Figure B.6 Diffusion Test Results for D11 (a) pH and Elec trical Conductivity and (b) Cumulative EC versus Cumulative Diffusion Time 0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000 0102030405060708090Cumulative diffusion time (day)Electrical Conductivity ( S/cm) .0 2 4 6 8 10 12pH Electrical Conductivity pH D-11 2M CaCl2 diffusant 0 10,000 20,000 30,000 40,000 50,000 60,000 0102030405060708090100Cumulative diffusion time (day)Cumulative EC ( S/cm) 2M CaCl2 diffusant e = 14.35 D-11 Thickness = 3 mm Air-dry weight = 2.5 g

PAGE 244

Appendix B: (Continued) 225 (a) (b) Figure B.7 Diffusion Test Results for D12 (a) pH and Elec trical Conductivity and (b) Cumulative EC versus Cumulative Diffusion Time 0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000 9,000 10,000 01020304050607080Cumulative diffusion time (day)Electrical Conductivity ( S/cm) .0 2 4 6 8 10 12pH Electrical Conductivity pH D-12 5M CaCl2 diffusant 0 20,000 40,000 60,000 80,000 100,000 120,000 01020304050607080Cumulative diffusion time (day)Cumulative EC ( S/cm) 5M CaCl2 diffusant e = 14.35 D-12 Thickness = 3 mm Air-dry weight = 2.5 g

PAGE 245

Appendix B: (Continued) 226 (a) (b) Figure B.8 Diffusion Test Results for D13 (a) pH and Elec trical Conductivity and (b) Cumulative EC versus Cumulative Diffusion Time 0 1,000 2,000 3,000 4,000 5,000 020406080100Cumulative diffusion time (day)Electrical Conductivity ( S/cm) .0 1 2 3 4 5 6 7 8 9 10pH Electrical Conductivity pH D-13 5M CaCl2 diffusant 0 5,000 10,000 15,000 20,000 25,000 30,000 35,000 0102030405060708090100Cumulative diffusion time (day)Cumulative EC ( S/cm) 5M CaCl2 diffusantt e = 4.11 D-13 Thickness = 3 mm Air-dry weight = 7.5 g

PAGE 246

Appendix B: (Continued) 227 (a) (b) Figure B.9 Diffusion Test Results for D14 (a) pH and Elec trical Conductivity and (b) Cumulative EC versus Cumulative Diffusion Time 0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000 01020304050607080Cumulative diffusion time (day)Electrical Conductivity ( S/cm) .0 2 4 6 8 10 12pH Electrical Conductivity pH D-14 5M NaCl diffusant 0 10,000 20,000 30,000 40,000 50,000 60,000 70,000 80,000 01020304050607080Cumulative diffusion time (day)Cumulative EC ( S/cm) 5M NaCl diffusant e = 14.35 D-14 Thickness = 3 mm Air-Dry weight = 2.5 g

PAGE 247

Appendix B: (Continued) 228 (a) (b) Figure B.10 Diffusion Test Results for D16 (a) pH and Elec trical Conductivity and (b) Cumulative EC versus Cumulative Diffusion Time 0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000 01020304050607080Cumulative diffusion time (day)Electrical Conductivity ( S/cm) .0 2 4 6 8 10 12pH Electrical Conductivity pH D-16 5M NaCl diffusant 0 5,000 10,000 15,000 20,000 25,000 30,000 35,000 40,000 01020304050607080Cumulative diffusion time (day)Cumulative EC ( S/cm) 5M NaCl diffusant e = 6.67 D-16 Thickness = 3 mm Air-Dry weight = 5 g(a)

PAGE 248

Appendix B: (Continued) 229 (a) (b) Figure B.11 Diffusion Test Results for D17 (a) pH and Elec trical Conductivity and (b) Cumulative EC versus Cumulative Diffusion Time 0 5,000 10,000 15,000 20,000 25,000 30,000 35,000 40,000 0102030405060Cumulative diffusion time (day)Electrical Conductivity ( S/cm) .0 2 4 6 8 10 12pH Electrical Conductivity pH D-17 All salts (1M each) diffusant 0 50,000 100,000 150,000 200,000 250,000 0102030405060Cumulative diffusion time (day)Cumulative EC ( S/cm) 1M NaCl + 1M KCl + 1M CaCl2 + 1M MgCl2 permeant e = 14.35 D-17 Thickness = 3 mm Air-dry weight = 2.5 g(a)

PAGE 249

Appendix C: Ionic Analysis Test Results 230 Figure C.1 Concentration of Various Cati ons in Effluent During Permeability (K-1) Figure C.2 Concentration of Various Cati ons in Effluent During Permeability (K-2) 100 1,000 10,000 100,000 0.01.02.03.04.05.06.0Pore volumeConcentration of Cations (mg/L) .DI water pre-hydration (K-1) Influent (1M Ca++) = 40,000 pp m Effluent (Na+) Effluent (Ca++) 100 1,000 10,000 100,000 0.01.02.03.04.05.06.07.0Pore volumeConcentration of Cations (mg/L) DI water pre-hydration (K-2) Influent (1M Mg++) = 24,000 ppm Effluent (Mg++) Effluent (Ca++) Effluent (Na+)

PAGE 250

Appendix C: (Continued) 231 Figure C.3 Concentration of Various Cati ons in Effluent Duri ng Permeability (K-3) Figure C.4 Concentration of Various Cati ons in Effluent During Permeability (K-4) 100 1,000 10,000 100,000 0.01.02.03.04.05.06.07.0Pore volumeConcentration of Cations (mg/L) ... DI water pre-hydration (K-3) Influent (1M K+) =39,000 ppm Effluent (K+) Effluent (Ca++) Effluent (Na+) 1 10 100 1,000 10,000 100,000 0.01.02.03.04.05.06.07.0Pore volumeConcentration of Cations (mg/L) DI water pre-hydration (K-4) Influent (1M Na+) = 23,000 ppm Effluent (Na+) Effluent (Ca++) Effluent (Mg++)

PAGE 251

ABOUT THE AUTHOR Naim Muhammad received his Bachelor of Science in Civil E ngineering (B.Sc. Engg.) in 1988 from the Bangladesh University of Engineering and Technology (BUET). He received his Master of Engineering (M.E ng.) in Civil Engineering specializing in Geotechnical Engineering in 1993 from the Na tional University of Singapore (NUS). Following his Masters degree, Naim worked as a Geotechnical Engineer for four years in various consulting and construction companies in Singapore. Naim also worked as a Lecturer in Singapore Polytechnic for three ye ars before heading to United States in 2000 for his Ph.D. program. During his PhD program at University of South Florida (USF), Naim has carried out research works on landfill liner materials an d bentonite clay. He has authored jointly with NUS faculty members a number of jour nal papers and international conference proceedings on land reclamation and marine clay slurry and is currently in the process of publishing a number of technical papers with USF faculty members on hydraulic, chemical, and diffusion characteris tics of bentonite clay.


xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001478742
003 fts
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 040811s2004 flua sbm s000|0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0000383
035
(OCoLC)56389463
9
AJS2431
b SE
SFE0000383
040
FHM
c FHM
090
TA145 (ONLINE)
1 100
Muhammad, Naim.
0 245
Hydraulic, diffusion, and retention characteristics of inorganic chemicals in bentonite
h [electronic resource] /
by Naim Muhammad.
260
[Tampa, Fla.] :
University of South Florida,
2004.
502
Thesis (Ph.D.)--University of South Florida, 2004.
504
Includes bibliographical references.
500
Includes vita.
516
Text (Electronic thesis) in PDF format.
538
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
Title from PDF of title page.
Document formatted into pages; contains 251 pages.
520
ABSTRACT: Inorganic contaminants, while transported through the bentonite layer, are chemically adsorbed onto the particle surfaces and exhibit a delay in solute breakthrough in hydraulic barriers. Transport of inorganic leachate contaminants through bentonite occurs by advection, diffusion or a combination of these two mechanisms. During the process of chemical solute transport through low permeability bentonite, the amount of cation exchange on the clay particle surface is directly related to the cation exchange capacity (CEC) of montmorillonite and other mineral constituents. The process of diffusion and advection of various inorganic leachate contaminants through bentonite is thoroughly investigated in this study. Diffusion characteristics are of specific interest as they have a prominent effect on the long term properties of bentonite compared to advection. This is mostly true if the hydraulic conductivity of the material is less than 10-8 cm/s and if the thickness of the barrier is small. Chemical reactions in the form of cationic exchange on the clay particle surfaces has been incorporated in the analysis of the diffusion process. Adsorption-desorption (sorption) reactions of chemical compounds that influence the concentrations of inorganic leachates during transport in bentonite clay have been modeled using the Fick's fundamental diffusion theory. Partition coefficients of the solutes in pore space, which affect the retardation factor of various individual ions of chemical solutions, have been investigated during transient diffusion and advection processes. Several objectives have been accomplished during this research study. An evaluation has been carried out of the hydraulic conductivity of bentonite with respect to single species salts and various combinations of electrolyte solutions. Diffusion properties of inorganic leachates through bentonite have been characterized in terms of apparent and effective diffusion coefficients. Time-dependent behavior of the diffusive ions has been analyzed in order to determine the total retention capacity of bentonite before electrical conductivity breakthrough and steady-state chemical stability are reached. An analytical solution of the attenuation of various inorganic ions concentrations through bentonite has been developed. Finally, recommendations were made for landfill liners exposed to highly concentrated inorganic leachates.
590
Adviser: Ashmawy, Alaa K.
653
landfill.
clay liner.
coefficient of permeability.
electrical conductivity.
adsorption capacity.
partition coefficient.
retardation factor.
690
Dissertations, Academic
z USF
x Civil Engineering
Doctoral.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.383