|USFDC Home | USF Electronic Theses and Dissertations||| RSS|
This item is only available as the following downloads:
The Relationship Between Plasticity Ratio and Hydraulic Conductivity for Bentonite Clay During Exposur e to Synthetic Landfill Leachate by Whitney M. Allen A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in Civil Engineering Department of Civil and Environmental Engineering College of Engineering University of South Florida Co-Major Professor: Alaa Ashmawy, Ph.D. Co-Major Professor: Je ffrey Cunningham, Ph.D. Manjriker Gunaratne, Ph.D. Date of Approval: November 14, 2005 Keywords: Atterberg limits, permeability, landfill liners, index properties, chemical compatibility Copyright 2005, Whitney M. Allen
DEDICATION I would like to dedicate this thesis to my parents, William R. Allen and Patricia S. Allen, along with my entire family. Without their love, support, and guidance, I would not have been able to continue my education to this degree. I would also like to express my gratit ude to Jeremy Runkle for his persistent encouragement and support.
ACKNOWLEDGEMENTS I would like to express my sincere ap preciation to my major professors, Dr. Alaa Ashmawy and Dr. Jeffrey Cunningham, for their inspiration, guidance, and support throughout this study. I also would like to thank Dr. Manjriker Gunaratne for his aid as a committee member. Appreciation is also extended to Delfin Carreon for his assistance in the lab and Jessica McRory for her guidance.
i TABLE OF CONTENTS LIST OF TABLES iii LIST OF FIGURES iv ABSTRACT ix CHAPTER 1 INTRODUCTION 1 1.1 General Concept 1 1.2 Landfills and Liners 1 1.3 Objective and Scope 3 1.4 Prediction of Chemical Compatibility 4 CHAPTER 2 BACKGROUND INFORMATION 6 2.1 Introduction to Bentonite 6 2.2 Structural Composition of Bentonite 6 2.3 Influence of Water and Chemicals 10 2.4 The Adsorbed Layer and its Influences 12 2.5 Transport through Bentonite Clay 17 2.6 Physical Structure of Bentonite Clay 19 CHAPTER 3 THEORY AND APPLICAT ION OF ATTERBERG LIMITS 22 3.1 Introduction to Plasticity 22 3.2 Testing Materials 23 3.3 Moisture Content Determination 24 3.4 Determination of Liquid Limit 26 3.5 Determination of Plastic Limit 27 CHAPTER 4 THEORY AND MEASUREMENT OF HYDRAULIC CONDUCTIVITY 31 4.1 Theoretical Background 31 4.2 Influences on Hydraulic Conductivity 33 4.3 Testing Materials and Set-Up 35 4.4 Initiation of Hydraulic C onductivity Testing and Duration 37 4.5 Termination Criteria 40
ii CHAPTER 5 EXPERIMENTAL RESULTS 43 5.1 Liquid Limit Testing 43 5.2 Plastic Limit Results 46 5.3 Plasticity Index 48 5.4 Casagrande Classification 49 5.5 Plasticity Ratio Results 51 5.6 Hydraulic Conductivity Testing and Ratios 53 CHAPTER 6 CONCLUSIONS AND RECOMMENDATIONS 59 6.1 Conclusions 59 6.2 Summary of Study 61 6.3 Recommendations 61 REFERENCES 63 APPENDICES 66 Appendix A: Liquid Limit Data 67 Appendix B: Hydraulic Conductivity Data 72
iii LIST OF TABLES Table 2.1 Typical Ion Dist ribution Found by Egloffstein in Sodium Bentonite 9 Table 2.2 Typical Hydrated Ra dii for Selected Cations 14 Table 2.3 Diffusion Coefficients of Se lected Ions in Water at 25C 18 Table 4.1 Chemical Compositions of DI Water and Salt Solutions 36 Table 5.1 Plasticity Ratio s for Pure Bentonite 52 Table 5.2 Plasticity Ratios for Bentofix 53 Table 5.3 Hydraulic Conductivity Values, k and Hydraulic Conductivity Ratios for Pure Bentonite Clay 55 Table 5.4 Hydraulic Conductivity Values, k and Hydraulic Conductivity Ratios for Bentofix Clay 55
iv LIST OF FIGURES Figure 1.1 Example of Landfill Design Possibility 1 Figure 1.2 Diagram of Bentofix GCL 4 Figure 2.1 Synthesis Pa ttern for Smectite 2:1 Unit 7 Figure 2.2 Montmorillonite Mineral Layers 8 Figure 2.3 Water and Cation Adsorption on Clay Surfaces 12 Figure 2.4 Three Mechanisms of Cation Adsorption 13 Figure 2.5 Distribution of Ions Along a Clay Surface Explained by the Diffuse Double Layer Concept 15 Figure 2.6 Relationship Between Cation Va lence and DDL Cation Concentration 16 Figure 2.7 Clay Particle Association and Modes 20 Figure 3.1 Atterberg Limits and Stages 22 Figure 3.2 Diagram of Fall Cone Equipment 27 Figure 3.3 Plastic Limit Rolling Device and Sample Preparation 28 Figure 4.1 Hydraulic Conductivity Diagram 32 Figure 4.2 EC Meter 35 Figure 4.3 pH Meter 35 Figure 4.4 Diagram of Permeameter for Hydraulic Conductivity Testing 38 Figure 4.5 Consolidation Phase of Hydraulic Conductivity Testing 40 Figure 5.1 Liquid Limit Results for Pure Bentonite Hydrated with DI Water 43
v Figure 5.2 Liquid Limit Values for Pure Bentonite Samples Exposed to Salt Solutions & DI Water 44 Figure 5.3 Liquid Limit Results of Pure Bentonite Represen ted by Molarity 45 Figure 5.4 Liquid Limit Values for Bentof ix Samples Exposed to Salt Solutions & DI Water 46 Figure 5.5 Plastic Limit Values for Pure Bentonite Samples Exposed to Salt Solutions & DI Water 47 Figure 5.6 Plastic Limit Values for Bentof ix Samples Exposed to Salt Solutions & DI Water 47 Figure 5.7 Plasticity Index Values for Pu re Bentonite Samples Exposed to Salt Solutions & DI Water 48 Figure 5.8 Plasticity Index Values fo r Bentofix Samples Exposed to Salt Solutions & DI Water 49 Figure 5.9 Casagrande Chart for Pure Bentonite Samples Exposed to Salt Solutions and DI Water 50 Figure 5.10 Casagrande Chart for Be ntofix Samples Exposed to Salt Solutions and DI Water 50 Figure 5.11 Hydraulic Conductivity Versus PVF Data for Pure Bentonite Exposed to DI Water 54 Figure 5.12 Hydraulic Conductivity Versus PVF Data for Bentofix Clay Exposed to 0.1M CaCl2 56 Figure 5.13 ECeffluent/ECinfluent Versus PVF Data for Bentofix Clay Exposed to 0.1M CaCl2 56 Figure 5.14 Hydraulic Conductivity Vers us PVF for Bentofix Clay Exposed to DI Water, 0.1M CaCl2, and 0.1M NaCl 57 Figure 6.1 Hydraulic Conductivity Ratio Ve rsus Plasticity Ratio for Pure Bentonite and Bentofix Clay 59 Figure 6.2 Plasticity Ratio Vers us Hydraulic Conductivity Ratio 60 Figure A.1 Liquid Limit Experimentation Results for KCl Concentrations on Pure Bentonite 67
vi Figure A.2 Liquid Limit Experimentation Results for CaCl2 Concentrations on Pure Bentonite 67 Figure A.3 Liquid Limit Experimentation Re sults for NaCl Concentrations on Pure Bentonite 68 Figure A.4 Liquid Limit Experimentation Results for MgCl2 Concentrations on Pure Bentonite 68 Figure A.5 Liquid Limit Experimentation Results for De-ionized Water on Bentofix Bentonite 69 Figure A.6 Liquid Limit Experimentation Results for KCl Concentrations on Bentofix Bentonite 69 Figure A.7 Liquid Limit Experimentation Results for CaCl2 Concentrations on Bentofix Bentonite 70 Figure A.8 Liquid Limit Experimentation Re sults for NaCl Concentrations on Bentofix Bentonite 70 Figure A.9 Liquid Limit Experimentation Results for MgCl2 Concentrations on Bentofix Bentonite 71 Figure A.10 Liquid Limit Results of Bentofix Bentonite Represented by Molarity 71 Figure B.1 Hydraulic Conductivity Versus PV F Data for Pure Bentonite Clay Exposed to 0.1M CaCl2 72 Figure B.2 ECeffluent/ECinfluent Versus PVF Data for Pure Bentonite Clay Exposed to 0.1M CaCl2 72 Figure B.3 Hydraulic Conductivity Versus PV F Data for Pure Bentonite Clay Exposed to 0.5M NaCl 73 Figure B.4 ECeffluent/ECinfluent Versus PVF Data for Pure Bentonite Clay Exposed to 0.5M NaCl 73 Figure B.5 Hydraulic Conductivity Versus PV F Data for Pure Bentonite Clay Exposed to 0.1M NaCl 74 Figure B.6 ECeffluent/ECinfluent Versus PVF Data for Pure Bentonite Clay Exposed to 0.1M NaCl 74
vii Figure B.7 Hydraulic Conductivity Versus PV F Data for Pure Bentonite Clay Exposed to 0.01M NaCl 75 Figure B.8 ECeffluent/ECinfluent Versus PVF Data for Pure Bentonite Clay Exposed to 0.01M NaCl 75 Figure B.9 Hydraulic Conductivity Versus PVF Data for Pure Bentonite Clay Exposed to 0.5M MgCl2 76 Figure B.10 ECeffluent/ECinfluent Versus PVF Data for Pure Bentonite Clay Exposed to 0.5M MgCl2 76 Figure B.11 Hydraulic Conductivity Versus PV F Data for Pure Bentonite Clay Exposed to 0.1M MgCl2 77 Figure B.12 ECeffluent/ECinfluent Versus PVF Data for Pure Bentonite Clay Exposed to 0.1M MgCl2 77 Figure B.13 Hydraulic Conductivity Versus PV F Data for Pure Bentonite Clay Exposed to 0.01M MgCl2 78 Figure B.14 ECeffluent/ECinfluent Versus PVF Data for Pure Bentonite Clay Exposed to 0.01M MgCl2 78 Figure B.15 Hydraulic Conductivity Versus PV F Data for Pure Bentonite Clay Exposed to 0.5M KCl 79 Figure B.16 ECeffluent/ECinfluent Versus PVF Data for Pure Bentonite Clay Exposed to 0.5M KCl 79 Figure B.17 Hydraulic Conductivity Versus PV F Data for Pure Bentonite Clay Exposed to 0.1M KCl 80 Figure B.18 ECeffluent/ECinfluent Versus PVF Data for Pure Bentonite Clay Exposed to 0.1M KCl 80 Figure B.19 Hydraulic Conductivity Versus PV F Data for Pure Bentonite Clay Exposed to 0.01M KCl 81 Figure B.20 ECeffluent/ECinfluent Versus PVF Data for Pure Bentonite Clay Exposed to 0.01M KCl 81 Figure B.21 Hydraulic Conductivity Versus PVF Data for Bentofix Clay Exposed to 0.1M CaCl2 82
viii Figure B.22 ECeffluent/ECinfluent Versus PVF Data for Bentofix Clay Exposed to 0.1M CaCl2 82 Figure B.23 Hydraulic Conductivity Versus PV F Data for Bentofix Clay Exposed to 0.1M NaCl 83 Figure B.24 ECeffluent/ECinfluent Versus PVF Data for Bentofix Clay Exposed to 0.1M NaCl 83
ix THE RELATIONSHIP BETWEEN PLASTICITY RATIO AND HYDRAULIC CONDUCTIVITY FOR BENTON ITE CLAY DURING EXPOSURE TO SYNTHETIC LANDFILL LEACHATE Whitney M. Allen ABSTRACT In landfill design, the contai nment of solid and liquid c ontaminant is essential. Leachate is produced from the biodegradation of the waste with the migration of liquid including rain-water through the heap. This liquid can become a health hazard if it leaches into the groundwater. Liners are pl aced beneath leachate collection systems to prevent leachate from seeping into the soil underneath the landfill. Compacted clay liners, usually contai ning bentonite clay, are widely used. Bentonite can be characterize d by its low hydraulic conductivity and high swell potential. With a low hydraulic conductivity, the liner can serve as a barrier. The high swell potential aids in the integrity of a liner when suffering from cracking or puncturing. The chemicals that can be found in leach ate are capable of increasing the clays hydraulic conductivity due to chemical interact ions. Chemical compatibility testing laboratory hydraulic conductivity tests using sp ecific chemical solutions as a permeant are performed to determine the effects. Laboratory hydraulic conductivity tests, regardless of the permeant, can be time-consuming and expensive. In this study, pure Wyoming bentonite clay and Bentofix clay were used. Deionized water and 0.01M, 0.1M, 0.5M concentr ations of four i norganic salt (NaCl, KCl, MgCl2, and CaCl2) solutions were the liquids to which both clays were exposed during testing. Plastic limit and liquid limit tests were run on bot h clays with all 13 liquids. Laboratory hydraulic conductivity te sting with pure Wyoming benonite clay was done with 12 different permeantsall solutions except 0.01M CaCl2 and 0.5M CaCl2.
x The hydraulic conductivity testing on Bentof ix clay was run with 3 permeantsdeionized water, 0.1M CaCl2, and 0.1M NaCl. The purpose of this study was to determ ine if a correlation exists between the experimentally determined liquid limit and plastic limit of a specific clay and its hydraulic conductivity when exposed to a synthetic leachate. It was determined that a trend exists that will allow for less expensive and timeconsuming determination for hydraulic conductiv ity of a clay liner when exposed to a specific chemical solution. However, more expe rimental data need to be collected before a definite trend is verified. The proposed pr ocedure requires that a hydraulic conductivity test of the clay be run using deionized water as the permeant, and plasticity index tests be performed using the leachate.
1 CHAPTER 1 INTRODUCTION Not in my backyard is a common thought of any re sident when discussing the placement of landfills. 1.1 General Concept Landfills can pose a serious threat to their local environment and residents if not properly designed. Beyond their main function of storing municipal waste, it is essential that landfills properly handle the produced le achate. Leachate is the term used to describe the liquid that deve lops from the municipal solid waste. This potentially harmful liquid is generated in the landfills by the movement of water (usually from rainfall) through the buried waste. As this water passes through the pile of decomposing solid waste, hazardous chemicals dissolve, or le ach, into the vertically moving liquid. Regardless of the possible harm of these waste disposal systems, the function of landfills is a necessity that must be rec ognized. Recycling, com posting, and combustion are acceptable means of processing solid wast e and currently account for 44% of the US discharge. (EPA, 2005a) The remaining 56% of non-hazardous waste can be found in approximately 1,700 municipal landfills located throughout the 50 states. 1.2 Landfills and Liners As municipal waste is broken down in these containment regions, it is essential that the produced leachate does not contaminate nearby soil or groundwater. Therefore, a leachate collection system is in stalled as part of the landf ill design (see Figure 1.1). One important piece of this containment system is the liner. The liner serves as a barrier for the potentially pollutant li quid. To prevent contamination of the underlying ground water, a liner with a low hydraulic conductiv ity is required. No material is fully
2 impermeable when in contact with various forms of chemicals, but if the liner material is capable of reducing the amount of flow to n egligible levels, it is deemed impermeable. The liners used in the leachate collection syst em often are made with clay. The clay liners can be made from compacted clays (CCL ), or geosynthetic cl ay systems (GCL). Not only is there the benefit of having a low hydraulic conductivity characteristic, but these liners can also stand the test of time. Th e presence of leachate in landfills is not a temporary phenomenon but also a by-product of th e aging landfill. The liners need to last longer than the lifetime of th e landfill. Top cover liners ar e placed above the waste in landfills to limit the amount of rainfall that can enter the landfill and to limit the amount of methane gas that can escape the decompos ing heap. The existing moisture travels through the waste and becomes contaminated dur ing this movement. The leachate that is produced regardless of the capability of the top liner must not be introduced into the groundwater. Any leachate that escapes from a poorly designed municipal solid waste (MSW) landfill can travel in the underlying soil. This harmful liquid is a detrimental addition to groundwater and nearby surface waters. Therefore, a clay liner is placed on the bottom of the landfill to wo rk in coordination with the l eachate collection system. The collection system not only prevents the migration of le achate into the underlying groundwater but creates a means for treating a nd properly disposing of the liquid waste. Some designs may demand multiple clay line rs. This study focuses only on the bottom clay liners and not on the top cover liner. Figure 1.1 Example of Landfill Design Possibility (Vesilind et al., 2002)
3 The chemicals that are found in leachate ar e deleterious to compacted clay liners; therefore, it is imperative to predetermine how a leachate will affect a specific clay layer (Shackelford et al, 2000; Shan and Lai, 2002; Simpson, 2000; Pe trov and Rowe, 1997; Ruhl and Daniel, 1997; Rad et al, 1994). H ydraulic conductivity testing is the most acceptable means of determining the hydraulic c onductivity of a clay soil. This testing can be expensive and time-consuming. A soil liner has few major functions and each is of extreme importance. First, the liner must exhibit a structural integrity for stability against slope failure and other earth movements. The other principal aspect s include a low hydraulic conductivity and chemical compatibility with leachate. Testi ng of the liner before and during installation is imperative to ensure a reliable leachate collection system. 1.3 Objective and Scope Other common characteristics of clay include the consistency limits known as Atterberg limits. The plastic limit and liquid li mit are two types of Atterberg limits. The plastic limit is the moisture content at which a clays behavior transforms to act more like a plastic than a solid. At an increased moisture content, the liquid limit can be determined. This is when the clay enters its liquid state. The Atterberg limits are used to determine the commonly utilized plasticity index. After determining all three of these properties (plastic limit, liquid limit, and plasticity index), the type of soil can be ascertained by the aid of a Casagrande Chart. With the use of a Casagrande Chart and the plasticity index, the pl asticity ratio of a clay sample e xposed to a chemical solution can be determined (Ashmawy et al., 2005). All of these terms will be el aborated on later in the thesis. The intent of this study is: (1) to establish whether the dete rmination of the Atterberg li mits can be a surrogate for the traditional hydraulic conductivity testing method, (2) to compare Atterberg limits and hydrauli c conductivity values of pure Wyoming bentonite clay against Bentofix bentonite clay used in GCLs and (3) to evaluate the effect of different chemical s of characteristics of these clay soils
4 Thirteen main solutions were involved during the experimental process. These included de-ionized water and three different solution concentrations of NaCl, KCl, MgCl2, and CaCl2. The concentrations cons isted of 0.01M, 0.1M, and 0.5M. Concentrations were added of 0.03M and 0.3M in the liquid limit testing of pure bentonite to further evaluate the trend of the concentration of the salt solution effect. In this series of testing, two clay so ils will be used for comparison: powdered, Wyo-Ben Wyoming sodium bentonite (referr ed to as pure bentonite) and granular Bentofix bentonite (see Figure 1.2) Bentofix bentonite is a sodium-activated bentonite that is used in the constructed GCLs (Geos ynthetic Clay Liners). A GCL consists of a thin clay layer placed between two geotext iles or a geomembrane. The bentonite clay used by Bentofix is traditionally needle-punched be tween two geotextiles prior to installation. All data that is us ed in this report was collected from the use of clay only. The geotextiles were not attached prior to or during any experiments to obtain a controlled comparison. Figure 1.2 Diagram of Bentofix GCL The main objective of this research is to determine if the plasticity ratio can be used as a surrogate for determination of th e permeability of a clay soil by laboratory hydraulic conductivity testing. 1.4 Prediction of Chemical Compatibility As mentioned earlier, hydraulic conductiv ity testing can be potentially expensive and a major time investment. A quicker method would reduce cost and save time. To experimentally determine the Atterberg limits no more than two days would be needed. In this study, some hydraulic conductivity tests ran for up to 90 days. The simplicity of Atterberg limits testing also provides lo wer lab fees comparatively to hydraulic conductivity testing.
5 A relationship between Atterberg limit te sts and hydraulic conductivity tests with salt solutions and de-ionized water as hydrating liquids is hypothesized. Therefore; by establishing the Atterber g limits of a clay exposed to a le achate, a prediction can be made on the effects that a leachate will have on the hy draulic conductivity of a clay liner. This will reduce testing expenses, save laborato ry time, and enhance the design of landfill liners.
6 CHAPTER 2 BACKGROUND INFORMATION 2.1 Introduction to Bentonite Bentonite is a common component of clay liners used in landfill construction. It is known for its high swell pot ential and low hydraulic conduc tivity. Both are extremely desirable characteristics for an impermeable material. The formation of bentonite clay rock or deposits is from the weathering of volcanic ash. Calcium bentonite is a regul ar occurrence while sodium bentonite is preferred for CCLs (compacted clay liners ) and GCLs (Geosynthetic clay liners). Sodium bentonite has a greater swell capabi lity than the calcium bentonite. The swell improves the performance of a clay liner becau se it is self-healing if a fracture or crack occurs during installatio n or desiccation. Semi-arid climat e mixed with alkaline soil, as found in Wyoming, are necessary for the form ation of sodiumbentonite. (Goldman et al., 1990 ) 2.2 Structural Composition of Bentonite Montmorillonite is the main clay minera l found in bentonite deposits. It is also responsible for the ideal qua lities of swelling and low hydraulic conductivity. Wyoming bentonite, a type of bentonite used in this study, is composed of montmorillonite and beidellite minerals (Hunter, 1993). The amount of montmorillonite in bentonite is the main factor in determining how well a certain sample of a clay will do as a clay liner. This mineral, though, is vulnerable to leachat e due to possibility of alteration of the chemical structure because of the chemicals f ound in leachate. In high quality bentonite, the montmorillonite mineral is approx. 75-90% by weight (Egloffstein, 2001). The clay in a clay liner is not always 100% bentonite. By adding percentages of bentonite to other more accessible soils, the performance of the soil is improved. Typically, 47-67% of the clay in GCLs is made up of montmo rillonite (Shackelford et al.,1999).
7 Montmorillonite belongs in the smectite group of clay minerals. The smectite group is notable by the 2:1 chemical structure. All clay minerals, which belong to the phyllosilicates mineral family, are not chemically st ructured the same way, but they have similar components. Smectites are distinguishe d by an alumina octahedral sheet structure that is composed of magnesium or aluminum coordinated with the placement of oxygens or hydroxyls. This sheet is located betw een two layers of interconnected silica tetrahedra, termed silica sheets. Hence, th e 2:1 unit is two tetrahedral sheets to one octahedral sheet. In montmorillonites, the octahedral sheet is mainly composed of aluminum ions. The typical chemical fo rmula for montmorillonite is stated as (OH)4Si8Al4O20nH20 (Lee and Shackelford, 2005). Figure 2.1 Synthesis Pattern for Smectite 2:1 Unit Extensive isomorphic substitution for silicon and aluminum by other cations is a unique characteristic of the smectite mineral (M itchell, 1993). Isomorphic substitution is the substitution, during formation, of a cation by another cation that is not part of the main structure of the crystal. Usually th e replacing cation has a lower valence, but maintains the original crystal structure. For instance, in montmorillonite, the aluminum ion (Al3+) is replaced by a divalent ion. A magnesium ion (Mg2+) is a common divalent cation in this interaction. This excha nge will result in a net negative charge (1). A portion on the trivalent ions will undergo isomorphic substitution. For every ion replaced, a negative charge will incur. Clay minerals, including montmorillonite, have a net negative surface charge due to the isomorphic substitution (Mitchell 1993, Shackelford 2005). Absorbed water and ions Alumina Sheet Silica Sheet
8 In montmorillonite, a surface charge defi ciency is commonly found because of the magnesium replacement of alum inum. Two-thirds of availa ble sites on the octahedral sheet consist of aluminum in a smectite. Ever y sixth aluminum is typically substituted in a montmorillonite structure. The isomorphic substitution is the main source of exchange capacity in smectites. In other clay mine rals, with low or no swell, the isomorphic substitution occurs in the tetrahedron layer. In montmorillonite, the substitution occurs in the octahedron layer (see Figure 2.2), otherwis e known as the alumin a sheet (Madsen et al., 1994). Because of the isomorphic substitu tion, cations such as magnesium, iron (II), and manganese are commonly found in the m ontmorillonite structur e. It has been determined by Lee and Shackelford (2005) that for pure montmorillonite with magnesium as the only replacing cation, a surf ace charge deficiency can be expected in the range of 0.5esu/unit cell1.2esu/unit cell. The beidelli te, a mineral also found in bentonite, is related to the montmorillonite mineral but the aluminum ion is replaced for silicon (Hunter, 1993). (a) (b) Figure 2.2 Montmorillonite Mineral Layers (a) Silicon Tetrahedron and Tetrahedra Arrangement (b) Octahedral Unit and Sheet Arrangement (Mitchell, 1993)
9 Because of the generated negative charge and a moderate surface charge density, clay minerals are capable of adsorbing a significant amount of water and cations to neutralize the negative charge. Montmor illonite has a large specific surface of 800m2/g (Shackelford et. al., 1999). Smectites, in ge neral, have a very large cation exchange capacity. Cation exchange capacity (CEC) is used to describe the maximum number of exchangeable cations for adsorption potenti al. Typical values range from 80 to 150 meq/100gm of dried weight of clay (Mitchell, 1993). It is because of these mechanics that montmorillonite has interlayer cations as shown in Figure 2.1. The cation exchange capacity of bentonite can be expected to be lower due to bentonite containing a small amount of minerals other than montmorill onite (Shackelford et al., 2000). The sodium bentonite that is commonly found in clay li ners can be distinguished by the Na+ ions existing on the surface of the cl ay layers to neutralize the negative charge. Other ions are typically found in s odium bentonite as seen in Table 2.1. During exposure to a solution, the cations ava ilable in the liquid will replace the Na+ ions (Jo et al., 2005). This type of cati on exchange can increase the hyd raulic conductiv ity that is usually associated with the mineral mont morillonite. Hydrated cations (like Na+) restrict the flow through the pore space. The change in cations will alter th e magnitude of this restriction. Montmorillonite is vulnerable to the chemical composition of solutions (Shackelford et al., 2000). This vulnerability extends into the clay liners of landfills. Table 2.1 Typical Ion Distribution Found by Egloffstein (2001) in Sodium Bentonite Besides the attraction of cations to the negatively charged clay surface, water molecules are also attracted. Water molecule s are dipolar with a positive charge at one end of the molecule and a negative charge at the other. The positive side of the H2O atom is attracted to the negative charge and th e negative side is attracted to the positively charge cations. The water molecule is also drawn to the surface of the clay surface Na+ Ca2+ Mg2+K+Fe2+ Al3+ 50-90% 5-25% 3-15% 0.1-0.8% <0.5% <0.5%
10 because of the hydrogen bonding. As mentioned in Section 2.2, the outer silica layer is composed mostly of oxygen atoms. During hydrogen bonding, the hydrogen is shared with the water molecules and the oxygen on the clay surface (Das, 2002). The water held to the surface because of attraction mechanisms is called the adsorbed water. The van der Waal forces that originally attract the laye rs together are weaker than the hydration energy and hydrogen bonds; therefore, the spa ce between the layers increases. This causes swelling (Mitchell, 1993). The swelling characteristic in bentonite ai ds in the reliability of compacted clay liners as well as geosynthetic clay liners. With an increase in swelling ca pability, there exists an increase in selfhealing capabil ity (Madsen et al., 1994). During installation and desiccation, punctures and cracks can develop in the cl ay liner. This occurrence would eliminate the resistance th at exists for leachate to es cape into the ground and allow for contamination if swelling di d not materialize. 2.3 Influence of Water and Chemicals The leachate that comes in contact with the clay liner in a municipal waste landfill can have an impact by several mechanisms Not only is the leachate introducing new chemicals to the clays chemical structure, but the leachate is also moving particles through the liner, adding stress to the liner through seepage forces, and removing some of the chemicals that were previously in the clay (Mitchell, 1993). The introduction of chemicals to clay has been the recent focu s of research towards improving landfill design (Shackelford et al., 2000; Shan and La i, 2002; Simpson, 2000; Petrov and Rowe, 1997a,b; Jo et al., 2001; Ruhl and Daniel, 1997). An impor tant piece of information about a solution is the type of chemical or chemicals present. The leachate that is produced from the decomposition of solid waste contains high concentrations of sodium, calcium, magnesium, chloride, and su lfate ions (Quasim and Chiang, 1994). The pH of a leachate can have a large a nd unpredictable effect on the clay liner. The pH of a liquid indicates if it is an acid or base. A str ong acid solution tends to break down the carbonates, iron oxides, and alumin a octahedral layers of clays (Mitchell, 1993). Basic solutions affect mostly the si lica sheet. As pH increases, the net proton
11 charge decreases (Sposito, 1989). Therefore; as the pH increases, there is a higher demand for metal cations on the clay surface. This movement of particles from the acids or bases creates the chance of (1) increased hydraulic conductivity due to an increase in pore space or (2) a decrease in hydraulic conductivity because of pore clogging. The bonding of the layers in a smectit es chemical structure occurs by weak, easily broken van der Waals forces and charge deficiencies. The adsorption of water and chemicals can easily penetrate and alter bentonit e at the microscopic level. Cations have an extreme effect on the inter-p article forces of clay dependi ng on the size and valence of the cation. When clay mixes with water, a stabilizing process occurs. Any cations in a clay mineral, that are unnecessary will mix with associated anions and precipitate out as a salt. When water passes through the clay, these salt precipitates will be flushed from the mineral. The clay structure, though, needs to stay in equilibrium with the pore fluid. Therefore, the ions that are present in the solution will affect the equilibrium of concentrations throughout the clay particle. Ions are interchangeable but the ease of replacing one i on with another ion depends on many factors. The valence of the ions is a major contributor to the restraints. Other factors include the concentration of th e ion in the solution and the radius of the ions. Because of the increased positive char ge and stronger attracti on to the negatively charged clay surface, the higher valence cati ons are more difficult to replace than the lower valence cations. The typical trend of cation exchangeability is as follows: Na+ < K+ < Mg2+ < Ca2+ This trend is under the situation that all concentrations are eq ual. If a concentration of the more easily replaced cations (e.g. Na+) is higher than a more difficult cation (e.g.Ca2+), the exchange will occur differently. Sodium bentonite is characterized by an abundance of sodium between the clay layers. A clay layer is the 2:1 unit cell. The sodium (Na+) is needed to balance the typical negative net surface charge of clay (F igure 2.3). Sodium attracts water molecules and aids in decreasing the hydr aulic conductivity of a clay lin er. Because of the cation exchange trend, a divalent calcium ion will easily replace the sodium ion. This will
12 potentially increase the hydrauli c conductivity of the clay line r. The reasons for this occurrence will be discussed further. Figure 2.3 Water and Cation Adsorption on Clay Surfaces 2.4 The Adsorbed Layer and its Influences Adsorption takes place at the interface of a solid phase a nd a liquid phase. Sposito (1989) describes three mechanisms th at can cause cation adsorption to a clay surface. The mechanisms are presented in Figure 2.4. The inner-sphere complex requires ionic or/and covalent bonding and no water molecule between the cation and clay surface. If a water mol ecule is present then an outer-s phere complex is formed with the aid of electrostatic bonding. The fina l mechanism, diffuse ion, does not form a complex with the surface but with the nearby surface water. Electrostatic bonding is also present in this mechanism. The ions pres ent in the diffuse ion swarm are considered readily exchangeable ions because of the w eak attraction. The inner-sphere complex has the strongest attraction and is not co nsidered a readily exchangeable ion. + + + + ++ + + + + + + + + + + + + + + + + ++ + + + + + Water dipoles cations Clay Surface
13 Figure 2.4 Three Mechanisms of Cati on Adsorption (Sposito, 1989) The molecular adsorption that takes place on the clay surface can be described by the theoretical diffuse double-layer. The diffuse doublelayer (DDL) represents the negatively charged clay surface and the charge d phase adjacent to the surface (Mitchell, 1993). The Stern-Gouy model is a commonly accepted method of explaining the diffuse double-layer and calculating the ion distribution but is lim ited in its accuracy. The basis of the Stern-Gouy model was set by the Gouy-Chapman theory. To follow this theory, assumptions must be made The cations and anions are considered point charges while the negative charge on the clay surface is homogeneous.1 Other assumptions include the thickness of the double layer becoming negligible with respect to the clay surface to achieve a one-dimensi onal condition and the electrical permittivity (ease that molecules can be polarized and orient ed in an electric fiel d) is independent of placement relative to the clays surface (Mitchell, 1993). The importance of these theories is to understand what influences a diffuse double layer. The relationship presented by th e Gouy-Chapman theory is presented as: 2 / 1 2 2 0 02 1 v e n DkT K where 1/K is the term to describe the thickness of the double layer, 0 is the permittivity in a vacuum, D is the dielectri c constant, k is the Boltzmann constant (1.38x10-23 JoK-1), T is the temperature in Kelvins, no is concentration of ions, e is the electronic charge in coulombs, and v for the ionic valence (Mitchell, 1993). 1 It is known that ions are of a finite size and this fact will be taken in to consideration later.
14 The assumptions create non-realistic ion concentrations because ions have sizes that take up space and vary from ion to ion. Values from the hydrated radii of the cations that were used in this study are listed below in Table 2.2. Ions with a larger radius need a thicker diffuse double layer and create greater interparticle repulsion. Table 2.2 Typical Hydrated Radii for Selected Cations Ion Hydrated Radius K+ 3.8-5.3 Na+ 5.6-7.9 Ca2+ 9.6 Mg2+ 10.8 The Stern-Gouy model is commonly used to describe the diffuse double layer with clay chemistry. The Stern layer is the location of the water molecules and the hydrated cations that lay along the negatively charged clay surface. In this model, the diffuse double layer is limited to the space beyond the Stern layer where more hydrated cations are attracted. Followi ng this model, the concentratio n of cations is higher along the negatively charged surface. As the distance from the surface increases, the concentration on the cations decreases (see Fi gure 2.5). This is a function of the electrical potential of the clay surface (Shack elford et al., 2000). The electrical potential, described by Mitchell (1993), is negative beca use of the negative su rface charge. This potential also varies with distance and is defi ned as the work to bring a positive charge to a specified point (Mitchell, 1993). As th e distance for the surface increases, the approaches zero.
15 Figure 2.5 Distribution of Ions Along a Clay Surfa ce Explained by the Diffuse Double Layer Concept (Mitchell, 1993) This model renders the following expression for the Debye length, (Mitchell, 1993): 2 22 F v RTo The Debye length is the centroid of the diffuse layer. In this equation, is the dielectric constant of the pore water, T is the absolute temperature, is the electrolyte concentration, R is the Universal gas constant, F is Faradays constant, and 0 and v symbolize the same items as in the previous equation. Another name for the Debye length is the thickness of the double diffuse layer. In all reality, this is an arbitrary thickness because the DDL has smudge d borders (Shackelford, 1999). The inorganic chemicals that are present in leachate will have an effect on the performance of the clay liner. This effect can also be explained with the Stern-Guoy Model. An alteration of the cation valence that is attracted to the clays surface will alter the electrical potential. From the equation for the Debye length, the change in valence of
16 Figure 2.6 Relationship Between Cation Valence and DDL Cation Concentration (Mitchell, 1993) the cation will also change the thickness of the diffuse double layer. The cation valence will also affect the possible concentration of ca tions as seen in Figure 2.6. A cation will a higher valence (Ca2+) will decrease the concentration more rapidly compared to a monovalent cation. With less ions between th e layers, the repulsion forces will decrease allowing for a shrinkage in the DDL. This will then increase hydr aulic conductivity and decrease swell. The thickness of the diffuse double layer is related to the hydraulic conductivity and swell potential of clay. If the diffuse double layer (DDL) shrinks, the available space for flow will increase. This directly cau ses an increase in the hydraulic conductivity (Shackelford et al., 2000). The replaceability of cations is relate d to the change in thickness of the diffuse double layer. Sodium ions are monovalent with a hydrated radius smaller than typical divalent ions as seen in Table 2.1. When a Ca2+ cation replaces the Na+ cation the DDL will shrink. The rate of the exchange will be dependent on the concentration of the Ca2+ in the solution. The concept was proved by experiments
17 conducted by Shackelford et al. (2000). Th e shrinkage of the double diffuse layer not only increases the hydraulic c onductivity but also decrease s the swell potential. The components of the chemical soluti on will be important in determining the effect on a clay liner. The changes to the hydraulic conductivity will typically be consistent with the internal alterations. Therefore; it is impera tive that not only should clay liners have a low hydraulic conductivity, but they also must be compatible with the chemicals present in a specific leachate. 2.5 Transport Through Bentonite Clay The interaction between the clay particles and ions is not exclusive to adsorption. Two other possible mechanisms of ion transp ort through a clay liner include advection and dispersion. Dispersion includes both molecular and mechanical processes. Advection is the transport of the dissol ved solids in flowi ng fluid (Fetter, 1999), such as the chemicals that come in contact with the landfill wate r during its movement through the biodegrading municipal waste. Th e movement of leachat e is typically onedimensional downward. With this assumption, the advective transport equation is given as y C v t Cy which relates the rate of change of the concentration of the chemical in the water to the vertical concentration gradient and the advection velocity. For low hydraulic conductivity soils, diffu sion is primarily responsible for the transport of chemicals (Fetter, 1999). Di ffusion is the dispersion that results from chemicals in solutions moving from areas of higher concentration to areas of lower concentration independent of fluid velocity. Following the same assumptions as needed for the advective transport of the contamin ant, the equation representing diffusive transport is 2 2 x C D t Cd
18 In this situation, the concen tration is no longer a function of velocity but of the diffusion coefficient. In transport of dissolved solutes through porous media the diffusion coefficient must reflect the slowing process that will occur due to the tortuosity induced by the clay particles. This effective diffusion coefficient, D*, is related to the diffusion coefficient by tortuosity which measures the effect of the shape of the flowpath followed by water molecules in a porous medium (Fetter, 1999). Diffusion is also affected by the negative charge of the clays particle su rface because electrical neutrality must be maintained during the diffusion process. Table 2.3 Diffusion Coef ficients of Selected Ions in Water at 25oC K+ 1.96*10-9 m2/sec Na+ 1.33*10-9 m2/sec Ca2+ 7.93*10-10 m2/sec Mg2+7.05*10-10 m2/sec Cl-2.03*10-9 m2/sec Research was conducted by Rowe (1998) to evaluate the effect of diffusion on GCL liners and found that the diffusion coefficients were dependent on several characteristics of the liner. These properties include void ratio a nd confining stress. It was also found that the concentration of i norganic ions in a permeant will also affect the diffusion coefficients. Because the chemicals within the flow ing liquid do not move at the same pace because of interactions and path interrupti ons from the clay particles, mechanical dispersion takes place. The mechanical disper sion is a function of the velocity of the fluid proportional to the dynamic dispersivity (i ) in the respective direction. The summation of mechanical dispersion and molecular diffusion are generally labeled hydrodynamic dispersion. All means of dispersi on obey Ficks law. Ficks first law simply states that areas of hi gher concentration of contaminan t typically move to areas of lower respective concentrations. This is illustrated by the negative sign in the equation:
19 dx dC D F In this equation, F stands for the mass flux (mass/areatime), D is the diffusion coefficient, and dC/dx is the change in c oncentration with respect to position. In landfills, there is a change in the concentr ation of contaminant with the increase of leachate as time passes Ficks second law is needed : 2 2x C D t Cd To determine the transport of a chemical in ground water, both processes must be considered. A form of the advective-disper sion equation must be used to determine the transport of the contaminant a nd its concentration, e.g.: x C v x C D t C 2 2 Many adaptations are possible to this equation cons idering such possibilities as sorption and de-sorption properties as mentioned be fore and reactions (such as from the biodegradation of organic c ontaminants) (Cunningham, 2005). 2.6 Physical Structure of Bentonite Clay The physical characteristics of a clay also impact its behavior. Clay minerals, including montmorillonite, have a platy particle shape. In Figure 2.6, common particle arrangements are laid out. A dispersed struct ure describes clay part icles with no face to face association while an aggregated struct ure implies face to f ace association between several particles. The term deflocculated is used to express no association between aggregates while flocculated specifies edge to edge or edge to face associations. The arrangement of the particles is rela ted to the charges of the surfaces. The face or surface of a clay particle has a negati ve charge. The edge of the platey particle has a positive charge. The edge to face flocculation occurs because opposite charges attract from the electrostatic attraction (Das, 2002). As described by Das (2002), when clay particles are exposed to salt solutions van der Waals forces are more prevalent allowing for the platelets to aggregate in a more parallel manner.
20 The fabric of the clay liner (i.e. the arrangement of particles) will affect the hydraulic conductivity performance. It is e xpected that dispersed arrangements allow a lower hydraulic conductivity than aggregated arrangements because of smaller pore volume. Through consolidation, compaction, a nd shearing, the fabric of a clay can be modified. The moisture content of clay dur ing the compaction process will determine the ease that the particle groups will be rearranged. If compaction occurs when the moisture content is above the optimum, the platelets will easily line-up along the failure plane (Mitchell, 1993). If the clay is compacted dry of the optim um moisture content, a much higher hydraulic conductivity value will be enco untered from an identical clay at the same void ratio and density exposed to the same permeant. By decrea sing the large pores possibly found in bentonite cl ay and creating a homogenous fa bric in a clay slurry, the potential for a lower hydraulic conductiv ity is increased. Also, by creating a homogenous fabric, the testing of cl ay liners is more controlled. The arrangements of clay platelets are also related to the diffuse double layer. The DDL is inversely proportional to the ability of the particles to aggregate. With an a ) e) c) d) f) b) g) Figure 2.7 Clay Particle Association and Modes (Mitchell, 1993) a) Dispersed and Deflocculated; b) Aggregated and Deflocculated; c) Edge-Face Flocculated but Dispersed; d) Edge-Edge Flocculated but Dispersed; e) Edge-Face Flocculated and Aggregated; f) Edge-Edge Flocculated and Aggregated; g)Edge-Face and Edge-Edge Flocculated and Aggregated
21 increase in the diffuse double layer thickness, the dispersion of cl ay platelets also increases (Goldman et al., 1990). Both of th ese characteristics resu lt in lower hydraulic conductivities compared to the other possible situations. From the earlier sections in this chapter, the importance of the microstructure is detailed. From Section 2.6, it should be unde rstood that the macrostructure (i.e. physical structure) is as essential as the chemistr y of clay to determining the capability of bentonite serving its purpose in a clay liner.
22 CHAPTER 3 THEORY AND APPLICATION OF ATTERBERG LIMITS 3.1 Introduction to Plasticity Atterberg limits is the general term gi ven to include the plastic limit, liquid limit, shrinkage limit, and plasticity index characteristics of a soil. Atterberg limit tests are relatively quick and reproduc ible under constant conditions The plasticity index is not determined from a standard test but is calculated from the liquid limit and plastic limit of a soil exposed to the same liquid. The different states of consistency of a soil are separated by these different limits. This is demonstrated in Figure 3.1. The solid state and semi-solid state of so il are at a low moisture content (i.e. little water will exist in the clay sample) where the soil acts as a solid. Some of the characteristics the soil would show are non-re shapeability and hardness. The moisture content that no longer controls the volume of a soil sample is termed the shrinkage limit (SL). Clays will swell when exposed to a li quid; the degree of the swelling is a function of the chemistry of the clay. Montmorilloni tes, such as bentonites, tend to have the greatest swell potential. The shrinkage limit is not a measure of this quantity. A swell Plastic state Liquid State Solid State Semi-solid State Shrinkage Limit Plastic Limit Liquid Limit Moisture Content, wc (%) 0% Figure 3.1 Atterberg Limits and States
23 test would be needed. The shrinkage limit is simply the moisture content, wc, below which the volume will cease to change as the soil is further dried. The term plastic state is used to descri be a soil that can be remolded and still hold a shape. The plastic limit separates the plastic state from the semi-solid state; therefore, the plastic state will posses higher mo isture contents than the semi-solid state and the solid state of the same soil. The liquid state is considered when the soil acts more like a li quid, for lack of a better word. It will no longer hold a shape. This will occur at higher moisture contents than the plastic limit. The liquid limit was set when the soil possesses a specific shear strength (2.5 kPa). The mois ture content under this situat ion is considered the liquid limit. To get the clay to act as desired one can simply increase or decrease the moisture content until the correct consis tency is achieved. All limits were arbitrarily set but, with standardized testing, are cons istent and reliable means of characterizing a soil. Further details about the plastic limit and liquid limit ar e located in subsequent sections in this chapter. A description of the plas ticity index can be found in Chapter 4. 3.2 Testing Materials For all experimentation in this study, two different clay soils were used. In the tests labeled pure bentonite clay, WYO-BEN (Billings, Montana) Extra High Yield, High Performance Bentonite was used. The powdered clay was stored in sealable plastic bins at air moisture content and room temperatur e. The Bentofix clay sample was provided by Bentofix Technologies, Inc. (Ontario, Canada). For all testing in this study, the same batch of the respective clay sample was used. The chemicals used in creating the synt hetic leachate solutions were purchased from Fisher Scientific. The Fisher Chemicals are MgCl2 (F.W. 203.31), CaCl2 (F.W. 110.99), NaCl (F.W. 58.44), and KCl (F.W.74.56). All were purchased in solid form and mixed with a specific amount of deionized wate r for the desired concentrations of the salt solutions. For testing with deionized water and for the creation of the chemical solutions, Publix Purified Water was utilized.
24 3.3 Moisture Content Determination To determine the Atterberg limits of a soil, the moisture content of the used samples of the soil for both liquid limit testing and plastic limit testing were needed (these will be discussed in section 3.4 and 3.5). At least thirteen sets of both types of test were run with a minimum of thirteen different liquids. A controlled test was run using deionized water. The other twelve liquids consisted of three diffe rent concentrations, 0.01M, 0.1M, and 0.5M, of the four inorganic salt solutions, MgCl2, NaCl, KCl, and CaCl2. For liquid limit testing for the pure Wy oming Bentonite powder, two additional molarities for all four salts were added, 0.03M and 0.3M. For the samples consisting of deionized water, ASTM Standard D2216 was followed for determination of samples moisture contents. A few exceptions to the standard were needed for the samples involving the clay exposed to electrolyte solutions. The procedure set forth by the ASTM standard was followed until calculating the moisture content. The calculation of the moisture content of samples with salt solutions had to be slightly modified. The traditional ca lculation for moisture content, as used for the samples with deionized water, is w= [(Mcws-Mcs)/ (Mcs-Mc)]*100, where w is the moisture content (%), Mcws is the mass of the wet specimen and the container, Mcs is the mass of the oven-dried specimen and the container, Mc is the mass of the container. This equation results in the ratio of the mass of th e water during the test over the mass of the solid particles that consists of pure bent onite clay represented as a percentage. For the samples involving the salt solutions, this calculation would have resulted in inaccurate moisture content. Respective with the concentration of the soluble salt in the solution, an additional mass will result in the dried sample due to the chemical in the salt solutions. The procedure of determining the water content was consistent in all 33 sets of tests (21 liquid limits and 13 plastic limits). An oven compatible pan was labeled and weighed. An approximate mass was collected of the clay from the test and weighed in the pan. All masses were recorded in grams with values to the 2nd decimal place. Clay samples of about 10-15 grams for liquid limit tests and 3-4 grams for plastic limit tests were used. The difference in mass collect ed between the tests was dependent on the
25 amount of the clay required to run the corres ponding test. More clay was needed for a liquid limit test than a plastic limit test. Thes e collected samples were then placed in an oven to be dried for at least 14 hours under 1005C. After a consistent mass had been obtained due to all moisture having evaporat ed from the sample, the dry specimen mass was recorded. As mentioned previously, the mass of the chemical in the dried specimen was accounted for when presenting the moisture content of the indivi dual samples. The calculation of the water content was modified for the samples exposed to salt solutions as follows: ch c cs cs cwsM M M M M w where Mcws is the mass of the wet sample and container, Mcs is the mass of the oven-dried specimen and container, Mc is the mass of the container, and Mch is the mass of the chemical due to the salt solution used dur ing tests. This mass was calculated by M-ch=M*(Mcws-Mcs)*(1L/ 1000mL)*(FW)*(1 mL H2O/ 1 g H2O) where M is the molarity of the respective salt solution and FW is the molecular weight of that salt. The density of the deionized water us ed in the equation was the standard assumption of water equaling 1g/cm3 at room temperature. It was experimentally determined that the actual density of the deionized water was 0.997 g/cm3. This difference is considered negligible for these calculations. It was also recognized that the pr eviously reported calculation for Mch does not take into account that some of the moisture in the oven dried specimen is due to the moisture in the air-dried bentoni te and not just the applied salt solution. To include this, the mass of the chemical in the dried specimen is given by Mch= M*(Mcws-Mcs-wa(Mcs-Mc-Mch))*(1L/1000mL)*(FW) where wa is the moisture content(%) of the air dried bentonite which was experimentally obtai ned to be 10%. The determination of Mch becomes an iterative process. Six trials were run of this process, two on each concentration level on random sa lts. The former method and the latter (iterative) method never resulted in a difference of the final water content value greater than 1%. The difference in the mass of the chemical when taking the moisture from the air-dried
26 bentonite was determined to be negligib le; therefore the initial calculation for Mch not considering the moisture content alre ady in the air dried clays was used. 3.4 Determination of Liquid Limit Casagrande, in 1932, standardized the liquid limit of a soil at a water content corresponding to a shear strength of 2.5 kP a (Das, 2002). This now defines the common liquid limit perception and is us ed for ASTM standards and the British Standards. The liquid used to determine the liquid limit of a so il has a great effect on the recorded value. It has been found that the cat ion valence found in a solution will decrease the liquid limit as the valence is increased (Mitchell, 1993). The liquid limit of the air-dried pure bentonite clay and Bentofix sample was determined using the fall cone method following the British StandardBS1377. The equipment used during this testing consists of a standard 0.78 N cone connected with a dial gauge that measures displa cement with an accuracy of 1/10th of a millimeter. At a penetration distance of 20mm, the soil has reach ed its liquid limit. This is measured by placing the tip of the cone on the top edge of the mold containi ng the hydrated clay. Then, placing the tip of the cone on the surface of the mid-point of the clay sample, the cone is released from its initial set hei ght for 5 seconds. The displacement is then recorded. A sample of the clay from the mold is then collected to determine its water content. This procedure was done an average of five times for each liquid limit test with at least two points belo w and two points above the liquid limit. The liquid limit of the clay was obt ained using deionized water and five molarities of four inorganic salt solutions for the pure Wyoming bentonite powder as described above. For the Bentofix clay, deionized water and three molarities of the four salt solutions were used to gather liquid limit values. Th e concentrations used for the solutions of NaCl, CaCl2, KCl, and MgCl2 were 0.01M, 0.1M, and 0.5M. A total of 13 sets of tests were run on the Bentofix clay.
27 Figure 3.2 Diagram of Fall Cone Equipment 3.5 Determination of Plastic Limit The standard means of obtaining a value fo r the plastic limit of a soil sample is the rolling method. A small mass of a clay wi th a low moisture content, when compared to the moisture content of liquid limits, is rolled to 1/8th of an inch. The rolling is done on the palms of the testers hands or on a ro lling device. Rolling by hand is less accurate than using the rolling device, but both me thods have complications. The hand method allows for more time exposure between the so il and the skin which slowly dries a sample. It is also difficult to measure exactly 1/8th inch and to ensure cons tant and even pressure while rolling between ones hands. The rolling device method also is technician based. The determination of the extent of fracturing of the rolled clay (failure) is dependent on the perception of the person running the plastic limit test. The same amount of pressure needs to be applied by the t op plate during each test and the rolling speed should be consistent. A common problem is the actual determination of the plastic limit. There are guidelines as to when a sample fails but it is ultimately the testers opinion on whether that state has been reached. To prevent di fferences in these mentioned variables, all plastic limit tests were run by the same tester. 30o 55m m 40m m Cone Release b utton DialGauge Cone Reset b utton
28 Figure 3.3 Plastic Limit Rolling Device and Sample Preparation To determine the plastic limit for the bentonite samples exposed to deionized water and salt solutions, ASTM Standard D 4318-00 was followed with a slight modification. The concept of the standard is to begin with a sample of a soil containing a low moisture content. At a quick pace, a pproximately 2 grams should be shaped by hand into an ellipsoid. The ellipsoid is then placed on the center of the bottom plate of the rolling device. This step is repeated on a piece from the same sample with no change in initial moisture content. This is done to ensure reproducibility and to place a larger sample in the oven during the moisture determination phase. These samples are then rolled to a diameter of 1/8th inch. Measurement of 1/8th inch is reliable for the rolling device method because of a resting place for the top plate on the interior of the sidewalls of the bottom plate. If the sample su ccessfully reaches the criterion of a 1/8th inch diameter and can be remolded into an ellips oidal shape, then the sample has a moisture content above the plastic limit. The procedure is then repeated. The testing continues until the sample has failed. Failure is when the sample crumbles, fractures, or barrels under this tr eatment. For bentonite clay, the sample crumbles. This is the moment when th e soil is entering the solid state. When a sample is able to be molded, it is acting plastically. To reach the plastic limit, the moisture content needs to decrease The decrease occurs slowly with the process of shaping and rolling. When shap ed and rolled, over a nd over, enough moisture is eventually removed to determine the plastic limit.
29 Only the plastic limit of the sample e xposed to deionized water was determined using this procedure. The process causes a complication with salt solutions because the clay sample is drying during this interaction. The procedure of th e standard had to be adjusted to accommodate this happening. If the samples moisture content is decreased due to the procedure, the concentration of the sa lt is increased. Molarity is a function of the mass of the salt over the volume of existi ng water. If the mass of the salt is held constant but the volume of water decreases, mo larity (M) increases. This would tamper with the recordings and the modified calculati ons of moisture content that were presented in the previous section. To decrease the likeliness of drying the samples that are exposed to salt solutions, an iterative process was established. An imag e of this procedure is presented in Figure 3.3. Prior to testing, 100mL of the chosen salt solution was created. This batch was the only solution used for determining the plastic limit of the pure be ntonite powder and the Bentofix granular bentonite with that corres ponding permeant. A six gram portion of the clay was measured out into individual ceramic bowls. Initially, 4mL of the salt solution was quickly mixed2 by hand with the six gram portion. The procedure of rolling the sample to a 1/8th diameter was attempted following the ASTM standard. If the sample failed before achieving the desired di ameter, a new air dried six gram portion was mixed with a higher volume of solution and rolled. If the sample achieved the 1/8th diameter and could be remolded, a new six gr am portion of clay was mixed with a lower volume of solution. If the sample did not achieve the 1/8th diameter as before, again a new six gram portion was shaped into an ellip soid and rolled on the rolling device. This process is continued until a sample failed just at the 1/8th inch diameter. As with the clay exposed to deionized water, failure is said to occur when the sample crumbles. To ensure that this moisture content is the correct in dication of plastic limit, a sample with an additional solution volume of 1mL had to reach the 1/8th in. diameter without failing, and a sample with a less volume of solution by 1mL had to fail before the 1/8th in. diameter. 2 A folding method was used to mix the solution with the clay; this is similar to a method used in cooking.
30 The methods were consistent betwee n the pure bentonite powder and the Bentofix granular bentonite.
31 CHAPTER 4 THEORY AND MEASUREMENT OF HYDRAULIC CONDUCTIVITY 4.1 Theoretical Background Hydraulic conductivity is a re presentative measure of the ease of a fluid to travel a medium while experiencing a certain hydrauli c gradient. Darcys Law quantifies the principle of hydraulic conductiv ity. The simple equation,nki v illustrates the principle of the concept. This rela tes the hydraulic conductivity, k, 3 to the hydraulic gradient, i, and the discharge velocity, v, including the effect of the porosity, n, of the clay specimen. The hydraulic gradient is defined by: L h i The h and L variables represent the head loss and the distance over which the head loss occurs, respectively. The disc harge velocity is equal to nA Q v where Q is the volumetric discharge and A is the cross-secti onal area of the clay specimen. Both of these properties are measurable in a labor atory hydraulic con ductivity experiment allowing calculation of the discharge velocity. The flow is expressed and calculated as a one directional scalar. Although hydraulic cond uctivity is delivered in cm/s, it is not a velocity. The units th at are achieved through th e equation result in cm3/cm2s. By mathematical cancellation, the cm/s is derived. The hydraulic conductivity, k, of a soil is dependen t on the liquid used to determine the hydraulic conductivity value. The prior equations are independent of the fluids properties. To include certain properties of the fluid in to the analysis, the absolute permeability can be calculated. The absolute permeability, K is used to relate the 3 The hydraulic conductivity, k is also commonly known as coef ficient of permeab ility.(Das,2002)
32 hydraulic conductivity to the uni t weight of the fluid, and its viscosity, with the equation: K k The units of the equation result in the absolute permeability, K being delivered in cm2. Several types of hydraulic conductivity te sting are possible. In-situ testing and testing in a laboratory are the two main branch es. For a laboratory test, a permeameter is utilized such as in this study. There are a variety of ways to set-up the permeameter. The test can be run with a flexible wall or ri gid wall permeameter. There are equations for constant head, falling head tests with constant tailwater levels only, falling head tests with constant headwater only, and falling head tests with decreasing headwater and increasing tailwater. For the presented data in this document, the last option was used. Figure 4.1 Hydraulic Conductivity Diagram The equation needed to determine the hydraulic conductivity fr om this set-up is as follows: 2 1lnh h a a At L a a kout in out in. aout h1 A h h2 L a in Datum
33 The a values are the cross-sectional areas of the respective reservoirs, and L and A are the length and cross-sectional ar ea of the porous medium. The t and h values are the independent values represen ting the time interval (t) in which the head (h1 and h2) values were taken. To quantify the effect that the chemi cals in the permeant could have on a clay liner, a compatibility test is used. Simply a laboratory hydraulic conductivity test is run using the solution of interest. The hydrau lic conductivity is calc ulated as presented above. In this study, chemicals that are common in landfill leachate are of interest. Therefore, compatibility tests were run with various molarities of CaCl2, NaCl, MgCl2, and KCl. 4.2 Influences on Hydraulic Conductivity Montmorillonite is the mineral in bent onite that allows for the high swell potential. The mineral also is related to the hydraulic conductivity by the adsorption of the water molecules and free ions. The adsorp tion restricts the flow path by limiting pore space. The greater the percentage of mont morillonite found in a bentonite deposit, the lower the hydraulic conductivity when the clay is permeated with water. Montmorillonite, though, is extremely vulnerabl e to a chemical attack; therefore, the previous concept does not necessarily apply to chemical permeants. The permeant will affect the clay liner by the adsorption of the cations found in the solution. The effect will be directly influenced by the type and amount of cations present in the non-standard liquids. Any liquid other th an waters is labeled a nonstandard liquid. Water is incl usive of de-ionized water, distill ed water, tap water, etc. The ion exchange will affect the thickness of the double diffuse layer of the bentonite; thereby, affecting the hydrau lic conductivity. As noticed by Shackelford et al. (2000), while running a flexible wall compatibility test on a needle punched GCL exposed to a weak 0.0125 CaCl2 permeant, the Ca2+ slowly substitutes for the Na+ in the DDL. With this exchange, the thickness of the adsorb ed layer decreases while the experimentally determined hydraulic conductivity increases. Along with the valence of the ion in the electrolyte, the concentration of the exchangeable ions in the hydrating liquid has a great
34 influence. The cation replaceab ility trend is altered when there is an overwhelmingly high amount of an ion that is less likely to replace an ion wi th a higher valence or smaller hydrated radius. Along with aff ecting the cation exchange, an in crease of electrolytes in the DDL will decrease the thickness a nd increase the hydraulic conductivity. In clay specimens that are not pre-hy drated prior to the hydraulic conductivity testing, aggregate size distribu tion can be a concern. When the sample is permeated with water, the aggregate size has a negligible effect. However, when exposed to a nonstandard liquid, the various sizes of aggr egates will not hydrate at the same rate (Shackelford, 1999). This will decrease the swell and potentially increase the hydraulic conductivity. Pre-hydration of the clay sa mple is known to have other effects (Shackelford, 1999). Clay liners that are pr e-hydrated with water prior to permeation have a much lower hydraulic conductivity th an non-pre-hydrated specimens directly exposed to the chemical solutions. In th is study, a common method of pre-hydration was used: imbibition. In this situ ation, the clay specimen swells before consolidation. It has been noted that slightly higher k values can be expected comp ared to if the clay had not been pre-hydrated (Petrov et. al, 1997b). The reason to ensure that the clay specimen is completely saturated prior to testing is to increase the tendency for even dispersion of the clay platelets. This will decrease the value of the experimentally determined hydraulic conductivity because of the reduced pore space. A relationship also exists between the void ratio of the clay sample and the determined hydraulic conductivity of the clay sample. (Shackelford et. al., 1999; Petrov and Rowe,1997; Mesri and Olson,1971) The exact empirical rela tionship is still debatable. The trend is cons istent, though; an increase in void ratio correlates to an increase in hydraulic conductivity. The relationship with the void ratio is a result of the void space in the specimen and not with the DDL. During th is study, the void ratio is held constant for all hydraulic conductivity tests with pure Wyoming bentonite and the Bentofix bentonite. The hydraulic gradient that is applied during the hydr aulic conductivity tests is another variable. The effect of the gradient on estimated values of k has been determined to be relatively low as long as the specime n is not experiencing additional consolidation
35 due to the seepage forces. According to Ra d et al. (1994), it has been deemed acceptable to use hydraulic gradient values as high as 2800 for GCL testing, but the typical values for low hydraulic conductivity soils (k<10-7 cm/s) are 50-550. (Shackelford et al., 1997) In this study, gradients were set at approximately 2000-2700. 4.3 Testing Materials and Set-Up As with the liquid limit and plastic limit testing, the same bentonite soils were used. The same chemicals were also used in creating the chemical solutions. The procedure for making the salt solu tions is the same as presente d in Section 3.2. For the hydraulic conductivity testing, ce rtain properties of the chemi cal solutions were needed for analyzing data. The chemical compositi on, namely pH and EC, of the primary 13 permeants were determined and recorded pr ior to conductivity test ing. The Accumet AP63 pH meter (Figure 4.4) was used in the determination of the pH of the solution and the YSI Model 3100 (Figure 4.3) with a conductiv ity cell with a cell c onstant (K) equal to 1.0/cm. This cell constant was needed to be sure the conductivity cell could accurately measure the electrical co nductivity values typical of these salt solutions. Figure 4.2 EC Meter Figure 4.3 pH Meter
36Table 4.1 Chemical Compositions of DI Water and Salt Solutions Type of Solution Molarity pH EC (mS/cm) Deionized Water 5.02 0.0029 CaCl2 0.01M 8.76 2.20 0.1M 9.90 18.70 0.5M 10.31 75.2 MgCl2 0.01M 5.47 2.255 0.1M 5.50 18.27 0.5M 5.53 68.80 NaCl 0.01M 5.23 1.170 0.1M 6.09 10.98 0.5M 6.00 47.3 KCl 0.01M 5.46 2.20 0.1M 5.62 18.70 0.5M 5.85 75.2 The ASTM Standard Test for Measurem ent of Hydraulic Conductivity of Porous Material using a Rigid-Wall, Compactio n-Mold Permeameter (ASTM D 5856-95) was consulted for test apparatus construction a nd performance. A compaction mold was not used in these experiments; therefore, some modification from the st andard exists. The primary unit was purchased fr om CETEC. The modifications to the apparatus were required because available cells were origin ally intended for constant head testing (Schenning, 2004). The permeameter cell consis ted of an acrylic ri gid wall cylinder with an inner diameter of 76.2mm and a thickness of 6.3 mm. On the outside of the cell, markings were applied lengthwise to indicate the height measurements in mm. The markings began at the designated base of the cylinder and increased until the rim. Capability of measuring the outflow was also needed. A graduated burette that was able to hold at least 35mL of volume was attached to the base of the permeameter. To ensure a secure fit with no leakage of liquid or air, a combination of Swagelok fittings, a silicone-curing agent, and Teflon was used. Sw agelok fittings were also used for the inlet valve for the applied air.
37 Per the ASTM standard, porous stones with a 76.2mm diameter were used within the cell during testing. An O -ring was placed around the stone to ensure a secure fitting into the cylinder and to prev ent sidewall leakage of the be ntonite slurry. A groove was cut along the center of the wall of the porous stones using a Dremel handheld drill with a diamond tip (Dremel 7144) to prevent disp lacement of the O-ring around the porous stone during the consolidation phase of testing set-up. Al ong with the porous stones and O-rings, Fisher brand P4 filter paper is used as a medium to maintain the integrity of the bentonite slurry and prevent bentonite loss. 4.4 Initiation of Hydraulic C onductivity Testing and Duration Prior to the permeameter set-up of an individual hydraulic conductivity experiment, a slurry of the tested bentoni te clay was made. For the pure Wyoming bentonite clay powder, the slurry was crea ted by mixing 200mL of de-ionized water to 22 grams of the bentonite sample. It was hand mixed for 45 minutes. The sample was then covered with a sealable plastic wrap. The sa mple was allowed to sit for at least 24 hours to guarantee complete hydration of the samp le. For the Bentofix bentonite clay, the slurry consisted of 22 grams of the clay with 150mL of de-i onized water. This slurry needed at least 36 hours for comp lete hydration. The slurries were securely covered to prevent escape of water due to evaporation. After this phas e, the procedure is identical for the pure Wyoming bentonite clay and the Bentofix bentonite clay.
38 Figure 4.4 Diagram of Permeameter for Hydraulic Conductivity Testing For use in the permeameter (Figure 4.2) two porous stones with attached O-rings were placed in an abundant amount of de-ionized water. Prio r to being placed inside the acrylic cylinder, the porous stones were soaked for 24 hours. One of these stones was then placed in the base of the cylinder with a thin layer of non-curing silicone compound. This grease was used as another precau tion to prevent side-wall leakage, as recommended by the ASTM Standard D5856. A piece of filter paper that had been wet with de-ionized water was then placed above th e porous stone. Twenty (20) grams of the bentonite from the bentonite slurry were plac ed in the cylinder above the filter paper. The weight of the water was taken into consid eration and the slurry is assumed to be a homogenous mixture. The ratio of solid bent onite to de-ionized wa ter was calculated and used so that 20 grams of solid bentonite w in the cell along with the proportional amount of water. A porous stone and a piece of filter paper were placed above the slurry with the filter paper below the stone. Any exce ss non-curing silicone compound was removed Millimeter Markings Inlet Pressure Valve Bentonite Sample Internal Piston Tightening Knobs Outflow Graduated Burette RigidWall Cylinder Porous Stones with O-rings
39 from the interior wall of the cylinder prior to slurry application and again after top porous stone. The complete cylinder was then placed in to the base of the cell. Along the upper face of the base, in the center of the side wa lls, is a channeling system allowing the liquid to escape the cylinder and to flow into the burette. A thin layer of the non-curing grease is applied between the edge of this circular channeling syst em and the rim of the large acrylic cylinder. The purpose, as before, is to prevent any leakage of liquid or air. The grease is not touching the face of the porous stone or interfering with the outflow of liquid into the channeling system. For the purpose of consolidation and te sting, a small acrylic piston with a diameter of 50.8mm was placed onto the top porous stone. A sufficient amount of deionized water was poured into the cylinder to aid in consol idation and keep the slurry saturated. Three knobs are screwed onto the surface of the aligned top plate resting on the internal piston. The effl uent flow valve is opened to allow discharge. The knobs were turned slightly to appl y a small amount of pressure ont o the top porous stone to aid in the consolidation process. There was at least a 15 minute interv als between turns to allow for the stress to dissipate between the t op porous stone and the slurry. This allowed for an evenly consolidated specimen. The consolidation took 7-10 days on average for each cell. The final height of the sample was approximately 7 mm with a diameter of 76.2 mm. When a slurry was completely cons olidated, the top plate laid securely on top of the unit. The piston stayed in place a nd made certain that the 7mm was a constant height. The piston prevented sw elling during the testing phase.
40 Figure 4.5 Consolidation Phase of Hydraulic Conductivity Testing The hydraulic conductivity testing followe d Test Method D of ASTM D 5856, which consists of decreasing headwater level and increasing tailwater level. Headwater and tailwater are also called inflow and out flow, respectively. After the consolidation phase, the de-ionized water was removed from the cell in both the cylinder and the graduated burette. Four hundred (400) mL of the selected solution was made and poured into the cylinder above the top porous stone di sc. The level of the solution in the large acrylic cylinder was determined using the millimeter markings and recorded. The permeameter was then connected to a laboratory pressure pane l by the air vent port. The pressure panel allowed for connections to the hydraulic conductiv ity cells with the applied pressure to be held constant. E ach connection had a monitoring device connected to a digital reading device. Two milliliters of de-ionized water was placed in the graduated burette to allow for immediate measur ement of change in head of the effluent flow. Readings were recorded on average once a day to allow a recognizable difference in the headwater and tailwater measurements. 4.5 Termination Criteria According to the ASTM Standard D5856, permeation is not to be terminated until steady hydraulic conductivity values are obtai ned. To calculate hydraulic conductivity using this standard, readings at recorded times are necessary. The standard does not
41 specify time intervals between readings. Duri ng this experiment, readings were taken every 24-48 hours. To terminate a test, four or more consecutive k value calculations must fall within 25% of the mean value. The calculations of th e hydraulic conductivity were done with the equation in Section 4.1 fo r this testing set-up. The standard also states that for termination, the inflow/outfl ow ratio of the permeameter should fall in the range of 0.75 to 1.25. The precision on the inflow was too low for this experiment to completely depend on inflow/outflow ratio but the hydraulic conductivity values were confidently calculated. It has also been suggested by the Standa rd that at least two pore volumes is sufficient for termination. Through prior re search on hydraulic conductivity testing, it has been evaluated that two pore volumes does not ensure that chemical equilibrium has been reached (Shackelford et al.,1999; R uhl and Daniel, 1997). It is suggested by Shackelford et al. (1999) that the chemical composition of the effluent in comparison to the influent should also be considered. Measurements of the pH and electrical conductivity of a sample of the effluent di scharge are quick dete rminations of the chemical composition. The pH and electrical conductivity of the initial pe rmeant was determined prior to placement in the permeameter. During the experiment, the effluent discharge was removed from the graduated burette when 2025mL was available for testing. When the effluent discharge was pulled from the rese rvoir, 2mL of de-ionized water was again placed in the bottom of the burette to a llow for immediate readings. The pH was determined for the fluid using an Accumet porta ble pH meter. The electrical conductivity was measured with an YSI 3100 conductivity instrument in temperature compensation mode. Both pieces of equipment were calibrate d on the days of pH and EC testing. The 2mL initial dilution of the effluent disc harge was taken into consideration. The possibility of the chemical compos ition of the influent fluid being altered during the experiment was also examined. It is not possible to determine any characteristics of the influent fluid after the application of the air pressure without interrupting the hydraulic conductivi ty testing. This would comp romise the results. With the chemistry of the influent permeant questionable, mainly a steady hydraulic
42 conductivity and at least 2 pore volumes of flow were used to determine termination. After an individual hydraulic conductivity experiment was considered ready for termination, the permeameter was disconnected from the pressure panel and the pH and electrical conductivity of the permeant was recorded. It was found that the influent permeant properties had been altered during te sting. The change is believed to be attributed to the fluid being affected by the gases in the air dissolving into the liquid and altering the permeants chemistry. Both the EC and pH of the perm eant had increased in comparison to the initial readi ngs made prior to exposure to the air pressure. Lee and Shackelford (2005) also observed the change in pH during hydraulic conductivity tests involving several con centrations of CaCl2. It was then conclude d that the pH of the influent versus the effluent would not be used during evaluation. The electrical conductivity ratio (ECeffluent/ ECinfluent) was still calculated to verify the trend that as the test approached completion th is ratio approaches 1. Once a hydraulic conductivity test had b een terminated, the permeameter was disconnected from the pressure panel. The consolidated bentonite clay and porous stones were removed from the large cylinder using a soil extractor. The stones were carefully removed from the clay. The moisture cont ent was then determined per the procedure detailed in the ASTM Standard D2216.
43 CHAPTER 5 EXPERIMENTAL RESULTS 5.1 Liquid Limit Testing Following the procedure outlined in S ection 3.4, the liquid limit of the two bentonite clays was determined. As previous ly mentioned, during each liquid limit test 5-6 samples were taken from the hydrated batch of clay. The moisture content was determined for each sample. The displacement during the falling cone test was also recorded for each correspondi ng moisture level. When moisture content (w, %) values ar e plotted against the penetration depth, the liquid limit can be calculated from the equati on of the line for that particular series. The liquid limit is defined to be the moisture content that corresponds to a displacement, or penetration, of 20 mm. The graph for th e liquid limit test of the pure bentonite powder hydrated with de-ionized water is presented in Figure 5.1. The graphs for the remaining pure bentonite and Bentofix tests are presented in the Appendix B. 300 350 400 450 500 550 600 650 700 750 010203040Penetration, d(mm)Moisture Content, w(%) Figure 5.1 Liquid Limit Results for Pu re Bentonite Hydrated with DI Water
44 When the determination of the liquid limit wa s completed for the pure bentonite powder, the following values (Figure 5.2) were obtained. 0 100 200 300 400 500 600Moisture Content, w(%) 0.5M 200140220170 0.1M 380210380220 0.01M 460440480430 De-ionized Water 540 KClCaCl2NaClMgCl2DI Figure 5.2 Liquid Limit Values for Pure Bentonite Samples Exposed to Salt Solutions & DI Water Generally, as the concentration of the salt increases in the solution the moisture content (%) of the clay sample at the liquid limit decreases. The valence of the cation also appears to have an effect. For the pure bentonite powder, this concept was further examined. In addition to the previous mentioned concentrations, additional tests were conducte d on concentrations of 0.3M and 0.03M for all four soluble sa lts (Figure 5.3).
45 0 100 200 300 400 500 600 0.010.11 Molarity (M)Moisture Content, w(%) NaCl CaCl2 KCl MgCl2 Figure 5.3 Liquid Limit Results of Pu re Bentonite Represented by Molarity In these experiments, the two monovalent salts seem to follow one trend while the two divalent salts follow another. The low concentrations of the four solutions have a liquid limit that is extremely high due to th e solutions acting similar to the pure deionized water. As the solu tions become more concentrat ed, the influence on the liquid limit of the clay strengthens. As the molarity of the concentrations increase, the paths seem to approach a merger. This pattern is consistent with the liquid limit tests performed on the Bentofix sample.
46 0 100 200 300 400 500 600Moisture Content, w(%) 0.5M 120130140130 0.1M 250160290170 0.01M 450370480360 De-ionized Water 520 KClCaCl2NaClMgCl2DI Figure 5.4 Liquid Limit Values for Bentofix Samples Exposed to Salt Solutions & DI Water 5.2 Plastic Limit Results Each sample determined to be at the pl astic limit (Section 3.5) was placed in an evaporating dish of known mass. The dish including the sample was then weighed, labeled, and placed in the drying oven. Moist ure content was determined using the same method and formula described earlier (Section 3.3). The results of this testing are presented in Figure 5.5 and Figure 5.6.
47 0 20 40 60 80 100 120Moisture Content, w(%) 0.5M 60506060 0.1M 60707070 0.01M 90908080 De-ionized Water 110 KClCaCl2NaClMgCl2DI Figure 5.5 Plastic Limit Values for Pure Bentonite Samples Exposed to Salt Solutions & DI Water 0 20 40 60 80 100 120Moisture Content, w(%) 0.5M 50807060 0.1M 80808080 0.01M 1001009090 De-ionized Water 110 KClCaCl2NaClMgCl2DI Figure 5.6 Plastic Limit Values for Bentofix Samples Exposed to Salt Solutions & DI Water
48 The plastic limit follows the same trend as the liquid limit. As the concentration of the salt increased in the solutions, the plas tic limit moisture contents decreased. This is demonstrated in both clay samples. Un like the liquid limit, the trends observed for plastic limit do not appear to depend strongly on the valence of the cation. 5.3 Plasticity Index The plasticity index can be calculated after the liquid limit and plastic limit are determined for a clay sample using the fo rmula PI=LL-PL where PI is the plasticity index, LL is the liquid limit, and PL is the plastic limit. All of these quantities are presented by the moisture cont ent (%) of the sample at that specific characteristic. 0 50 100 150 200 250 300 350 400 450 0.5M 14090150120 0.1M 320140310150 0.01M 370350390340 De-Ionized Water 420 KClCaCl2NaClMgCl2DI Figure 5.7 Plasticity Index Values for Pure Bentonite Samples Expose d to Salt Solutions & DI Water
49 0 50 100 150 200 250 300 350 400 450Moisture Content, w(%) 0.5M 70507070 0.1M 1708021080 0.01M 350340390380 De-ionized Water 420 KClCaCl2NaClMgCl2DI Figure 5.8 Plasticity Index Values for Bentofix Samples Exposed to Salt Solutions & DI Water 5.4 Casagrande Classification Soils are typically characteri zed with the assistance of a Casagrande Chart. This chart demonstrates the relationship between th e plasticity index and the respective liquid limit of a soil. Bentonite clay traditionally falls above the A-line and below the U-line on this chart. On a Casagrande Chart, the A-line is the lower of the two diagonal lines while the U-line is the higher dia gonal line (Figure 5.9). The Aline separates the inorganic clays from the silty and organic soils. A vertical line is also present to distinguish soils with a liquid limit greater than 50%. Such soils are described as havi ng a high plasticity. Soils consisting of montmorillonite are expected to plot 1) above the A-line and 2) to the right of the 50% marker on the horizontal axis (liquid limit). The following graphs exhibit the results of th e 13 sets of testing for the four salt solutions and deionized water.
50 CL MLML or OL MH or OH CL or OL CH or OH0 50 100 150 200 250 300 350 400 450 0100200300400500Liquid Limit (LL) Plasticity Index (PI) .0.5M MgCl2 DI 0.01M NaCl 0.5M KCl 0.1M MgCl2 0.5M NaCl 0.5M CaCl2 0.01M MgCl2 0.01M CaCl2 0.1M NaCl 0.1M CaCl2 0.1M KCl 0 01M K C l Figure 5.9 Casagrande Chart for Pure Bentonite Sa mples Exposed to Salt Solutions and DI Water CL MLML or OL MH or OH CL or OL CH or OH0 50 100 150 200 250 300 350 400 450 0100200300400500Liquid Limit (LL) Plasticity Index (PI) .0.5M MgCl2 DI 0.01M NaCl 0.5M KCl 0.1M MgCl2 0.5M NaCl 0.5M CaCl2 0.01M M g Cl2 0.01M CaCl2 0.1M NaCl 0.1M CaCl2 0.1M KCl 0 01M K C l Figure 5.10 Casagrande Chart for Bentofix Samples Exposed to Salt Solutions and DI Water
51 In both graphs, a pattern of the plots is noticed. The clay samples hydrated by deionized water are in the uppe rright hand corner. For the samples hydrated with a salt solution, the plots move lower and to the left as the concentrations increase. This follows the previous observations from the liquid limit and plastic limit tests independently. In the Casagrande Chart for the pure bentoni te sample, there is a cluster of data representing all four 0.01M so lutions. Then the monovalent 0.1M solutions plot together. Close to the 0.5M monovalent solutions, the 0. 1M divalent solutions point are present. The furthest left and down are the 0.5M CaCl2 and MgCl2 solutions. The last six mentioned solutions all four 0.5M and 0.1M of divalent solutions lie very close to the A-line. Along with the down and left movement associated with the increase in molarity of the solution, there also is a movement towards the A-line. For the chart presenting the Bentofix data, similar characteristics of the pure bentonite samples can be noticed. As mentione d, it was expected for the samples to plot above the A-line and to the ri ght of the 50% marker on th e liquid limit axis. In all thirteen data points, the second condition occurred. W ith the exception of 0.5M MgCl2 and 0.5M CaCl2, all points are in the expected range for clay soils above the A-line. The points corresponding to 0.5M MgCl2 and 0.5M CaCl2 lie below the A-line, in the region usually indicative of hi gh-plasticity silts or organics. O bviously, the two samples did not become silts or organics4. The behaviors that are typical of clays do appear to be altered by the presence of a high amount of divalent cati ons in the Bentofix sample but it is still a Bentofix clay sample. 5.5 Plasticity Ratio Results According to previous work of Ashmawy et al. (2005), the plas ticity ratio can be used to indicate the effect that a leachate ha s on a soils plasticity as compared to testing performed with de-ionized water. The pl acements of the samples on the Casagrande chart are needed in calculating the plasticity ratio. The relative plasticity is first determined. The relative plasticity can be de fined as a measure of how far the soil is 4 The area that the two samples fall in usually labels them as a high plasticity silt or organic soil but once a clay, always a clay.
52 from the U-line and is ascertained by dividi ng the plasticity index of the sample by the optimum plasticity index. The optimum plas ticity index is the U-line placement at the corresponding liquid limit on the Casagrande char t. The plasticity ratio is equal to the ratio of the relative plasticity of a clay exposed to a leachate to th e relative pl asticity of the clay exposed to de-ionized water. Table 5.1 Plasticity Ra tios for Pure Bentonite KCl CaCl2 NaCl MgCl2 0.5M 0.9 0.8 0.9 0.9 0.1M 1.1 0.9 1.0 0.9 0.01M 1.0 1.0 1.1 1.0 As the molarity of the salt solution incr eases, the plasticity ratio for the pure bentonite decreases. There is an exception in this trend. The pure bentonite sample permeated with the 0.1M KCl solution had the highest plasticity ratio when compared to the 0.01M and 0.5M KCl solutions. If the plasticity ratio is e qual than 1, this hydrating sa lt solution exposed to the clay acts similarly to the soil saturated by de-ionized water. This is seen in three of 0.01M salt solutions combined with pure bentonite, the exce ption being the bentonite hydrated with 0.01M NaCl. It is expected that the lower conc entrations of salt solutions would behave similar to the de-ionized water because of the high dilution. The two 0.1M monovalent salt solutions also share the characteristic of th e 0.01 M salt solutions with the 0.1M KCl and 0.1M NaCl having plasticity ratios of 1.1 and 1. 0 respectively. The divalent 0.1M salt solution and all four 0.01M so lutions have plasticity ratios below 1. This signifies that the samples with lower pl asticity ratios are receding from the U-line in comparison to the de-ionized water hydrated pure be ntonite sample.
53Table 5.2 Plasticity Ratios for Bentofix KCl CaCl2 NaCl MgCl2 0.5M 0.8 0.7 0.8 0.7 0.1M 0.9 0.8 0.9 0.8 0.01M 1.0 1.0 1.0 1.0 The plastic limit and liquid limit tests for the Bentofix have demonstrated similar trends as the testing on pure bentonite. Th e plasticity ratio trends for the Bentofix clay are not identical to the pure bentonite pl asticity ratios trends, but a resemblance does exist. In both types of clays, with the incr ease of molarity in the salt solutions, there is a decrease in plasticity ratio of the respective salt. With the Bentofix, a trend dependent of the valence of the salt solution exists (Table 5.2). All four 0.01M salt solutions exposed to th e Bentofix clay delivered plasticity ratios equal to 1.0. The salt solutions containing m onovalent cations have identical plasticity ratios. For the 0.5M and 0.1M solutions of KCl and NaCl, the plasticity ratios equaled 0.8 a nd 0.9, respectively. The salt so lutions with divalent cations also had identical results. These solutions resu lted in lower plasticity ratios than the KCl and NaCl solutions. The 0.5M and 0.1M solutions of CaCl2 and MgCl2 had plasticity ratios of 0.7 and 0.8, respectively. 5.6 Hydraulic Conductivity Te sting Results and Ratios As mentioned in Section 4.5, a specific criterion was set for termination of an individual rigid wall hydraulic conductivity test. To eval uate the data, the calculated k values are plotted against the pore flow volum es. Using the data from the pure Wyoming bentonite hydraulic conductivity test with deionized water as a permeant, Figure 5.11 is presented. From the data, the hydraulic conductivity value, k, was determined to be 2.0x10-10 cm/s. This hydraulic c onductivity was also observed for the Bentofix clay exposed to de-ionized water.
54 1.0E-11 1.0E-10 1.0E-09 1.0E-08 1.0E-07 00.511.522.533.544.55 Pore Volumes of Flow, PVFHydraulic conductivity, k (cm/s ) Figure 5.11 Hydraulic Conductivity Versus PVF Data from Pure Bentonite Exposed to DI Water This method was used in determination of all hydraulic conduc tivity values. The calculated hydraulic conductivity values are pr esented with an accura cy expected to be within one order of magnitude. To evaluate the chemical compatibility tests, the hydraulic conductivity ratio, kl / kw, was calculated. The term kl represents the experiment ally determined hydraulic conductivity value of a clay exposed to a specif ic leachate. The leachates used in this study are the synthetic leachates (i.e. the salt solutions). The kw represents the hydraulic conductivity of the same type of clay with deionized water as the permeant. All hydraulic conductivity values and hydraulic cond uctivity ratios are presented in Table 5.3 for the pure bentonite clay and in Tabl e 5.4 for the Bentofix clay.
55Table 5.3 Hydraulic Conductivity Values, k and Hydraulic Conductivity Ratios for Pure Bentonite Clay Permeant k (cm/s) kl/ kw De-ionized Water 2.0E-10 0.5M MgCl2 4.0E-10 2.0 0.5M NaCl 2.3E-10 1.1 0.5M KCl 4.5E-10 2.3 0.1M CaCl2 7.3E-10 3.6 0.1M MgCl2 5.0E-10 2.5 0.1M NaCl 3.0E-10 1.5 0.1M KCl 6.0E-10 3.0 0.01M MgCl2 7.5E-10 3.8 0.01M NaCl 1.3E-10 0.6 0.01M KCl 4.0E-10 2.0 Table 5.4 Hydraulic Conductivity Values, k and Hydraulic Conductivity Ratios for Bentofix Clay Permeant k (cm/s) kl/ kw De-ionized Water 2.0E-10 0.1M CaCl2 1.5E-9 7.5 0.1M KCl 3.8E-10 1.9 For the chemical compatibility tests, ch emical equilibrium was also important in determination of a rigid-wall hydr aulic conductivity test being applicable for termination. In Figure 5.12, the results of the Bentofix clay exposed to 0.1M CaCl2 are presented. In Figure 5.13, the results of the ECeffluent/ ECinfluent ratio are presented versus pore volumes of flow. For chemical equilibrium to be achieved, the electrical conductivity of the effluent flow should be equal to the electrical co nductivity of the influent flow (i.e. the permeant). As the hydraulic conductivity test approached completion, the ECeffluent-/ECinfluent ratio should approach 1. This is demonstr ated in Figure 5.13. It is can be noted
56 that the point at which the hydraulic conductivity values be came consist corresponds to the PVFs of the ECeffluent/ ECinfluent equals about 1. 1.E-11 1.E-10 1.E-09 1.E-08 1.E-07 05101520 Pore Volumes of Flow, PVF Hydraulic Conductivity, k (cm/s) Figure 5.12 Hydraulic Conductivity Versus PVF Data for Bentofix Clay Exposed to 0.1M CaCl2 0 1 2 024681012141618 Pore Volumes of Flow, PVFECeffluent/ECinfluent Figure 5.13 ECeffluent/ ECinfluent Versus PVF Data for Bentofix Clay Exposed to 0.1M CaCl2
57 As mentioned in Section 4.5, the electrical conductivity of the influent permeant may have changed slightly during ap plication of air pressure du ring the hydraulic conductivity testing. For the determination of termination and the ECeffluent/ECinfluent ratio, the initial EC reading was used. Using the data collected from the Bent ofix clay hydraulic c onductivity tests, the effect of the permeant on the experimentally determined hydraulic conductivity can be observed. The clay exposed to the de -ionized water had the lowest hydraulic conductivity equal to 2.0x10-10 cm/s. The salt solution with the monovalent cation, 0.1M NaCl, has a slightly increased hydraulic conductivity of 3.8x10-10 cm/s. This difference in experimentally determined hydraulic conductiv ity is basically negligible because of the difference being less than an orde r of magnitude. Estimates of k are probably only accurate to within a fact or of 10. For the purpose of evaluation, all k values are given with two significant figures. For the salt solution with the diva lent cation, 0.1M CaCl2, hydraulic conductivity is about one order of magn itude greater than the clay with the deionized water permeant. From the literature (C hapter 2), it is expected that the divalent cation would increase the hydraulic conductiv ity of a clay more than a monovalent cation. 1.E-10 1.E-09 1.E-08 012345678910 Pore Flow Volumes, PVF (mL)Hydraulic Conductivity, k (cm/s) DI Water, k_w=2.0E-10 0.1M CaCl2, k_l=1.5E-9 0.1M NaCl, k_l=3.75E-10 Figure 5.14 Hydraulic Conductivity Versus PVF fo r Bentofix Clay Exposed to DI Water, 0.1M CaCl2, and 0.1M NaCl The hydraulic conductivity values for th e pure bentonite did not follow such a distinguishable trend. As seen in Table 5.3, all hydraulic conductivity ratios for the pure
58 bentonite are in the range of 0.63.8. From Table 5.4, the Bentofix clay exposed to the 0.1M CaCl2 had a hydraulic conductivity an orde r of magnitude gr eater than the Bentofix clay exposed to de-ionized water. This resulted in a hydraulic conductivity ratio of 7.5. In all hydraulic conduct ivity tests using a NaCl solution, the kl/kw ratio was close to one. The NaCl solutions appear to have the weakest effect compared to the remaining salts. This is expected because a lot of cation exchange would not be expected between a sodium bentonite and a sodium solution. By compar ison, it is noticed that all hydraulic conductivity ratios for th e pure bentonite are close to 1. Therefore, the clays exposed to the various salt solutions were not dramatically affected. The graphs of the hydraulic c onductivity versus PVF and the ECeffluent/ ECinfluent versus PVF for all hydraulic conductivity tests are presented in Appendix B.
59 CHAPTER 6 CONCLUSIONS AND RECOMMENDATIONS 6.1 Conclusions The main objective of this study is to dete rmine if the plasticity ratio can be used as a surrogate for determination of chemical co mpatibility. To evaluate the data for both types of clays, pure bentonite and Bentof ix clay, the hydraulic conductivity ratio was plotted versus the plasticity ratio. Th is graph is presented in Figure 6.1. 0 1 2 3 4 5 6 7 80.511.5Plasticity Ratio kl/ kw MgCl2 NaCl CaCl2 KCl Bentofix CaCl2 Bentofix NaCl Figure 6.1 Hydraulic Conductivity Ratio Versus Plasticity Ratio for Pure Bentonite and Bentofix Clay The plots in Figure 6.1 are for all hydraulic conductivity tests that went to completion. There are twelve chemical compatibility tests conducted that were able to be successfully terminated not including the de-ionized wate r permeant. Twelve hydraulic conductivity ratios were calculated from the chemical compatibility tests with the aid of the two
60 hydraulic conductivity tests run with deionized water as the permeant (one with pure bentonite and one with Bentofix clay). As seen in the data presented in the tabl es in Chapter 5, the plasticity ratios and hydraulic conductivity ratios for a ll samples are close to 1. From Figure 6.1, all data points fell into a narrow range around the 1 to 1 intersect. If a salt solution does not drastically alter the physical properties of clay (plasticity ratio), the hydraulic conductivity should not be exp ected to drastically increa se during exposure to the solution. This is demonstrated in the figure. Superpositioning the data collected from this study on to a graph presented by Ashmawy et al. (2005), the narrow range of ra tios calculated during this study are further shown. The plots in the lowe r right hand corner of this ch art were collected during the experiments of this study. One sample had a plasticity ratio greater than one and therefore is not included in Fi gure 6.2. Figure 6.2 includes th e ratios collected in this study along with ratios collected from testing on an untreated bentonite along and on a polymer-treated bentonite clay. The untreat ed bentonite is notated by a U and a number while the polymer-treated bentonite is notated by a T and a number. From the previous study, plasticity ratios as low as ~0.4 were determined. For the cl ay samples with lower plasticity ratios, higher hydr aulic conductivity ratios were observed. The results presented in Chapter 5 did not show this amount of variation in either se t of ratios. 0 5 10 15 18.104.22.168.81Plasticity Ratio kl/ kw MgCl2 NaCl CaCl2 KCl Bentofix CaCl2 Bentofix NaCl Figure 6.2 Plasticity Ratio Vers us Hydraulic Conductivity Ratio
61 6.2 Summary of Study During the determination of the Atterberg limits, a general trend was noticed. For both the liquid limit and plastic limit testing, as the concentration of the salt solution increased, the respective moisture content at th e limits decreased. When the plastic limits and the liquid limits were plotted on a Casagrande chart, a trend conti nued to persist. As the concentration of the salt solutions increase d, the plots moved left and downward. The placement of the clay sample exposed to deio nized water was in the top right hand corner for the Casagrande Chart, for both the pur e bentonite and for the Bentofix clay. The hydraulic conductivity ratio was not able to be described by such a trend. For the pure bentonite clay, all chemical compatibility tests delivered k values with an order of magnitude 10-10 cm/s. All ratios are basically equal to one. The chemical compatibility tests for Bentofix clay are lackin g sufficient data to be certain of the same trend. From the two results presented in this paper, a similar trend can be expected. Ideally, as the plasticity ratio decreases, the hydraulic conductivity ratio should increase. From the experiments, no drastic chan ges were seen in either ratio for any sets of testing. When the plasticity ratio is close to one, the hydr aulic conductivity ratio should also be close to one. With a hydraulic conductivity ra tio of one or approximately equal to one, the experimentally determined hydraulic conductivity of a clay exposed to a specific salt solution will be similar to the hydr aulic conductivity of th at clay exposed to deionized water. 6.3 Recommendations The continuation of this study would be extremely profitable. The reduction in testing fees with an increased reliability in landfill design w ould be advantageous to both civil engineers and to citizens w ho could be affected by a faulty clay liner. From the data collected during this study, a definitive relationship could not be presented for determining the extent of the effect that a leachate would have on a clay liner through Atterberg limit testing.
62 With additional experimentation, further analysis could be done. The tested properties of pure bentonite clay were not drastically altered by the exposure to salt solutions of various concentrations. The Be ntofix clay was not thoroughly tested to conclusively determine the effect that sa lt solutions had on th e hydraulic conductivity. The two chemical compatibility tests presente d in this thesis for the Bentofix clay delivered hydraulic conductivity ratios close to one. Starti ng with a soil that is more susceptible to chemical attack would be a dvantageous in determining if an empirical relationship can be expressed. For the hydraulic conductivity testing, dur ing the consolidatio n phase of set-up, the clay was pre-hydrated with de-ionized wa ter prior and during th e consolidation phase of set-up. It is encouraged that hydraulic conductivity test s also be conducted with nonpre-hydrated clay specimens. During the Atterberg limit determination, the clay was exposed only to the solution of evaluation. Th is concept should also be applied to the hydraulic conductivity testing. Simply, with more determinations of hydraulic conductivity rati os and plasticity ratios for various bentonite clay s, a clear relationship is expected to present itself.
63 REFERENCES Ashmawy, A.K., Muhammad, N., and Elhajji, D. (2005), Advecti on, Diffusion, and Sorption Characteristics of Inorga nic Chemicals in GCL Bentonite, ASCE Geotechnical Special Publication, No. 142, Waste Containment and Remediation, E.M. Rathje, Ed. 10 pp. Benjamin, M. (2002), Water Chemistry, McGraw-Hill, Singapore. Cunningham, J.A. (2005), Lecture Notes, Tra nsport in Porous Medi a, University of South Florida, Spring Semester 2005. Das, B.M. (2002), Principles of Geotechnical Engineering, Fifth Edition, Thomas Learning, Inc., Pacific Grove, CA. Egloffstein, T.A. (2001), Natur al BentonitesInfluence of the Ion Exchange and Partial Desiccation on Permeability and Self-Healing Capacity of Bentonites Used in GCLs, Geotextiles and Geomembranes, Vol.19, pp.427-444. Fetter,C.W. (1999), Contaminant Hydrogeology, Second Edition, Prentice Hall, Upper Saddle River, NJ. Goldman, L.J., Greenfield, L.I., Damle, A.S., Kingsbury, G.L., Northeim, C.M., and Truesdale, R.S. (1990), Clay Liners for Waste Management Facilities: Design, Construction, and Evaluation, Noyes Data Corporation, Park Ridge, New Jersey. Guyonnet, D., Gaucher, E., Gaboriau, H., Pons, C.H., Clinard, C., Norotte, V. and Didier, G. (2005), Geosynthetic Clay Liner Inter action with Leachate: Correlation between Permeability, Microstructure, and Surface Chemistry, Journal of Geotechnical and Geoenvironmental Engineering, Vol.131, No. 6, pp.740-748. Hunter, R. J. (1993), Introduction to Modern Colloid Science, Oxford University Press, Oxford. Jo, H.Y., Benson, C.H., Shackelford, C.D., Lee, J., and Edil, T.B. (2005), Long-Term Hydraulic Conductivity of a Geosynthetic Cla y Liner Permeated with Inorganic Salt Solutions, Journal of Geotechnical and Geoenvironmental Engineering, Vol.131, No.4, pp.405-417.
64 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, Vol. 127, No.7, pp.557567. Lee, J.M. and Shackelford, C.D. (2005), Impact of Bentonite Quality on Hydraulic Conductivity of Geosynthetic Clay Liners, Journal of Geotechnical and Geoenvironmental Engineering, Vol.131, No.1, pp.64-77. Madsen, F. and Nesch, R. ( 1994), Characteristics and Sea ling Effect of Bentonites, Proceedings of an International Symposium, Nrnberg, Germany, pp.31-49. Mesri,G. and Olsen, R.E. (1971), Mechanisms Controlling the Perm eability of Clays, Clays and Clay Minerals, Vol. 19, pp.151-158. Mitchell, J.K. (1993), Fundamentals of Soil Behavior, Second Edition, John Wiley and Son, Inc., New York, New York. Mitchell, J.K. and Madsen, F.T. (1987), Chemical effects on Clay Hydraulic Conductivity, Proceedings of the Geotechnical Prac tice for Waste Disposal Conference, Ann Arbor, Michigan, pp. 87-116. Petrov, R.J. and Rowe, R.K. (1997), Geo synthetic Clay Liner (GCL) Chemical Compatibility by Hydraulic Conductivity Testing and Factors Impacting its Performance, Canadian Geotechnical Journal, Vol.34, pp.863-885. Petrov, R.J., Rowe, R.K., and Quigley, R.M. (1997), Selected Fact ors Influencing GCL Hydraulic Conductivity, Journal of Geotechnical and Ge oenvironmental Engineering, Vol.123, No.8, pp.683-695. Qasim, S.R. and Chiang, W. (1994), Sanitary Landfill Leachate: Generation, Control, and Treatment, Technomic Publishing Company, Inc., Lancaster, Pennsylvania. Rad, N.S., Jacobson, B.D., and Bachus, R. C. (1994), Compatibility of Geosynthetic Clay Liners with Organic and Inorganic Permeants, Proceedings of the Fifth International Conference on Geotextile s, Geomembranes, and Related Products, Singapore, pp.1165-1169. Rowe, R.K. (1998), Geosynthe tics and the Minimization of Contaminant Migration through Barrier Systems Beneath Solid Waste, Proceedings of the Sixth International Conference on Geosynthetics, Atlanta, Ga, pp.27-102. Ruhl, J.L. and Daniel, D.E. (1997), Geosynthetic Clay Liners Permeated with Chemical Solutions and Leachate, Journal of Geotechnical and Geoenvironmental Engineering, Vol. 123, No. 4, pp.369-381.
65 Schenning, J.A. (2004), Hydraulic Performance of Polymer Modified Bentonite, MSCE thesis, University of South Florida. Shackelford, C.D., Malusis, M.A., Majeski, M.J., Stern, R.T. (1999), Electrical Conductivity Breakthrough Curves, Journal of Geotechnical and Geoenvironmental Engineering, Vol. 125, No. 4, pp.260-270. 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, Geotextiles and Geomembranes, Vol.18, pp.133-161. Shan, H. and Lai, Y. (2002), Effect of Hydr ating Liquid on the Hydraulic Properties of Geosynthetic Clay Liners, Geotextiles and Geomembranes, Vol. 20, pp.19-38. Simpson, B.E. (2000), Evaluation of Leach ate Compatibility to Soil Clay for Three Geosynthetic Clay Liner Products in Advances in Systems Using Geosynthetics, American Society of Civil Engineers, pp. 117-128. Sposito, G. (1989), The Chemistry of Soils, Oxford University Press, New York, New York. Sridharan, A. and Nagaraj, H.B. (2005), Hydraulic Conductivity of Remolded Finegrained Soils Versus Index Properties, Geotechnical and Geol ogical Engineering, Vol. 23, pp.43-60. U.S. Environmental Protection Agency (2005a), Basic Facts: Municipal Solid Waste (MSW). Accessed on July 5, 2005. http://www.epa.gov/epaoswer/nonhw/muncpl/facts.htm Vesilind, P.A., Worrell, W.A ., and Reinhart, D.R. (2002), Solid Waste Engineering. Brooks/Cole, Crawfordsville.
67 Appendix A Liquid Limit Data 0 100 200 300 400 500 600 010203040 Penetration, d (mm)Moisture content, w(%) 0.5M KCL 0.3M KCl 0.1M KCL 0.03M KCl 0.01M KCl Figure A.1 Liquid Limit Experimen tation Results for KCl Concentrations on Pure Bentonite 0 100 200 300 400 500 600 010203040 Penetration, d(mm)Moisture Content, w(%) 0.5M CaCl2 0.3M CaCl2 0.1M CaCl2 0.03M CaCl2 0.01M CaCl2 Figure A.2 Liquid Limit Experi mentation Results for CaCl2 Concentrations on Pure Bentonite
68 Appendix A: (Continued) 0 100 200 300 400 500 600 010203040 Penetration, d(mm)Moisture Content, w(%) 0.5 NaCl 0.3M NaCl 0.1 M NaCl 0.03M NaCl 0.01M NaCl Figure A.3 Liquid Limit Experimenta tion Results for NaCl Concentrations on Pure Bentonite 0 100 200 300 400 500 600 010203040 Penetration, d(mm)Moisture Content, w(%) 0.5 MgCl2 0.3M MgCl2 0.1M MgCl2 0.03M MgCl2 0.01M MgCl2 Figure A.4 Liquid Limit Experi mentation Results for MgCl2 Concentrations on Pure Bentonite
69 Appendix A: (Continued) 300 350 400 450 500 550 600 010203040Penetration, d(mm)Moisture Content, w(%) Figure A.5 Liquid Limit Experimenta tion Results for De-ionized Water on Bentofix Bentonite 0 100 200 300 400 500 600 010203040 Penetration, d (mm)Moisture content, w(%) 0.5M KCL 0.1M KCL 0.01M KCl Figure A.6 Liquid Limit Experimentation Results for KCl Concentrations on Bentofix Bentonite
70 Appendix A: (Continued) 0 100 200 300 400 500 600 010203040 Penetration, d(mm)Moisture Content, w(%) 0.5M CaCl2 0.1M CaCl2 0.01M CaCl2 Figure A.7 Liquid Limit Experi mentation Results for CaCl2 Concentrations on Bentofix Bentonite 0 100 200 300 400 500 600 010203040 Penetration, d(mm)Moisture Content, w(%) 0.5 NaCl 0.1 M NaCl 0.01M NaCl Figure A.8 Liquid Limit Experimenta tion Results for NaCl Concentrations on Bentofix Bentonite
71 Appendix A: (Continued) 0 100 200 300 400 500 600 010203040 Penetration, d(mm)Moisture Content, w(%) 0.5 MgCl2 0.1M MgCl2 0.01M MgCl2 Figure A.9 Liquid Limit Experi mentation Results for MgCl2 Concentrations on Bentofix Bentonite 0 100 200 300 400 500 600 0.010.11 Molarity (M)Moisture Content, w(%) NaCl CaCl2 KCl MgCl2 Figure A.10 Liquid Limit Results of Bentof ix Bentonite Represented by Molarity
72 Appendix B Hydraulic Conductivity Data 1.E-11 1.E-10 1.E-09 1.E-08 1.E-07 024681012 Pore Volumes of Flow, PVFHydraulic Conductivity, k (cm/s) Figure B.1 Hydraulic Conductivity Versus PVF Data for Pure Bentonite Clay Exposed to 0.1M CaCl2 0 1 2 024681012 Pore Volumes of Flow, PVFECeffluent/ECinfluent Figure B.2 ECeffluent/ECinfluent Versus PVF Data for Pure Bentonite Cl ay Exposed to 0.1M CaCl2
73 Appendix B: (Continued) 1.E-11 1.E-10 1.E-09 1.E-08 1.E-07 0246 Pore Volumes of Flow, PVFHydraulic Conductivity,k (cm/s) Figure B.3 Hydraulic Conductivity Versus PVF Data for Pure Bentonite Cl ay Exposed to 0.5M NaCl 0 1 2 012345 Pore Volumes of Flow, PVFECeffluent/ECinfluent Figure B.4 ECeffluent/ECinfluent Versus PVF Data for Pure Bentonite Cl ay Exposed to 0.5M NaCl Clogging started
74 Appendix B: (Continued) 1.E-11 1.E-10 1.E-09 1.E-08 1.E-07 02468 Pore Volumes of Flow, PVFHydraulic Conductivity, k (cm/s) Figure B.5 Hydraulic Conductivity Versus PVF Data for Pure Bentonite Cl ay Exposed to 0.1M NaCl 0 1 2 01234567 Pore Volumes of Flow, PVFECeffluent/ECinfluent Figure B.6 ECeffluent/ECinfluent Versus PVF Data for Pure Bentonite Clay Exposed to 0.1M NaCl
75 Appendix B: (Continued) 1.E-11 1.E-10 1.E-09 1.E-08 1.E-07 01234 Pore Volumes of Flow, PVFHydraulic Conductivity, k (cm/s) Figure B.7 Hydraulic Conductivity Versus PVF Data for Pure Bentonite Clay Exposed to 0.01M NaCl 0 1 2 01234 Pore Volumes of Flow, PVFECeffluent/ ECinfluent Figure B.8 ECeffluent/ECinfluent Versus PVF Data for Pure Bentonite Clay Exposed to 0.01M NaCl
76 Appendix B: (Continued) 1.0E-11 1.0E-10 1.0E-09 1.0E-08 1.0E-07 02468 Pore Volumes of Flow, PVFHydraulic Conductivity, k (cm/s) Figure B.9 Hydraulic Conductivity Versus PVF Data for Pure Bentonite Clay Exposed to 0.5M MgCl2 0 1 2 0123456 Pore Volumes of Flow, PVFECeffluent/ECinfluent Figure B.10 ECeffluent/ECinfluent Versus PVF Data for Pure Bentonite Clay Exposed to 0.5M MgCl2
77 Appendix B: (Continued) 1.E-11 1.E-10 1.E-09 1.E-08 1.E-07 024681012 Pore Volumes of Flow, PVFHydraulic Conductivity, k (cm/s) Figure B.11 Hydraulic Co nductivity Versus PVF Data for Pure Bentonite Cl ay Exposed to 0.1M MgCl2 0 1 2 024681012 Pore Volumes of Flow, PVFECeffluent/ECinfluent Figure B.12 ECeffluent/ECinfluent Versus PVF Data for Pure Bentonite Clay Exposed to 0.1M MgCl2
78 Appendix B: (Continued) 1.E-11 1.E-10 1.E-09 1.E-08 1.E-07 024681012141618 Pore Volumes of Flow, PVFHydraulic Conductivity, k (cm/s) Figure B.13 Hydraulic Co nductivity Versus PVF Data for Pure Bentonite Clay Exposed to 0.01M MgCl2 0 1 2 024681012141618 Pore Volumes of Flow, PVFECeffluent/ECinfluent Figure B.14 ECeffluent/ECinfluent Versus PVF Data for Pure Bentonite Clay Exposed to 0.01M MgCl2
79 Appendix B: (Continued) 1.E-11 1.E-10 1.E-09 1.E-08 1.E-07 0246810 Pore Volumes of Flow, PVFHydraulic Conductivity,k (cm/s) Figure B.15 Hydraulic Co nductivity Versus PVF Data for Pure Bentonite Cl ay Exposed to 0.5M KCl 0 1 2 0246810 Pore Volumes of Flow, PVFECeffluent/ECinfluent Figure B.16 ECeffluent/ECinfluent Versus PVF Data for Pure Bentonite Clay Exposed to 0.5M KCl
80 Appendix: (Continued) 1.E-11 1.E-10 1.E-09 1.E-08 1.E-07 02468101214 Pore Volumes of Flow, PVFHydraulic Conductivity, k (cm/s) Figure B.17 Hydraulic Co nductivity Versus PVF Data for Pure Bentonite Cl ay Exposed to 0.1M KCl 0 1 2 02468101214 Pore Volumes of Flow, PVFECeffluent/ECinfluent Figure B.18 ECeffluent/ECinfluent Versus PVF Data for Pure Bentonite Clay Exposed to 0.1M KCl
81 Appendix B: (Continued) 1.E-11 1.E-10 1.E-09 1.E-08 1.E-07 012345 Pore Volumes of Flow, PVFHydraulic Conductivity, k (cm/s) Figure B.19 Hydraulic Co nductivity Versus PVF Data for Pure Bentonite Clay Exposed to 0.01M KCl 0 1 2 0246 Pore Volumes of Flow, PVFECeffluent/ECinfluent Figure B.20 ECeffluent/ECinfluent Versus PVF Data for Pure Bentonite Clay Exposed to 0.01M KCl
82 Appendix B: (Continued) 1.E-11 1.E-10 1.E-09 1.E-08 1.E-07 05101520 Pore Volumes of Flow, PVF Hydraulic Conductivity, k (cm/s) Figure B.21 Hydraulic Co nductivity Versus PVF Data for Bentofix Clay Exposed to 0.1M CaCl2 0 1 2 024681012141618 Pore Volumes of Flow, PVFECeffluent/ECinfluent Figure B.22 ECeffluent/ECinfluent Versus PVF Data for Bentofix Clay Exposed to 0.1M CaCl2
83 Appendix B: (Continued) 1.E-11 1.E-10 1.E-09 1.E-08 1.E-07 01234567 Pore Volumes of Flow, PVF Hydraulic Conductivity, k (cm/s) Figure B.23 Hydraulic Co nductivity Versus PVF Data for Bentofix Clay Exposed to 0.1M NaCl 0 1 2 02468 Pore Volumes of Flow, PVFECeffluent/ECinfluent Figure B.24 ECeffluent/ECinfluent Versus PVF Data for Bentofix Clay Exposed to 0.1M NaCl
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 001709490
007 cr mnu|||uuuuu
008 060508s2005 flu sbm s000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001393
Allen, Whitney M.
The relationship between plasticity ratio and hydraulic conductivity for bentonite clay during exposure to synthetic landfill leachate
h [electronic resource] /
by Whitney M. Allen.
[Tampa, Fla.] :
b University of South Florida,
Thesis (M.S.C.E.)--University of South Florida, 2005.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
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 96 pages.
ABSTRACT: In landfill design, the containment of solid and liquid contaminant is essential. Leachate is produced from the biodegradation of the waste with the migration of liquid including rain-water through the heap. This liquid can become a health hazard if it leaches into the groundwater. Liners are placed beneath leachate collection systems to prevent leachate from seeping into the soil underneath the landfill. Compacted clay liners, usually containing bentonite clay, are widely used. Bentonite can be characterized by its low hydraulic conductivity and high swell potential. With a low hydraulic conductivity, the liner can serve as a barrier. The high swell potential aids in the integrity of a liner when suffering from cracking or puncturing. The chemicals that can be found in leachate are capable of increasing the clays hydraulic conductivity due to chemical interactions.Chemical compatibility testing laboratory hydraulic conductivity tests using specific chemical solutions as a permeant are performed to determine the effects. Laboratory hydraulic conductivity tests, regardless of the permeant, can be time-consuming and expensive. In this study, pure Wyoming bentonite clay and Bentofix clay were used. Deionized water and 0.01M, 0.1M, 0.5M concentrations of four inorganic salt (NaCl, KCl, MgCl2, and CaCl2) solutions were the liquids to which both clays were exposed during testing. Plastic limit and liquid limit tests were run on both clays with all 13 liquids. Laboratory hydraulic conductivity testing with pure Wyoming benonite clay was done with 12 different permeants- all solutions except 0.01M CaCl2 and 0.5M CaCl2. The hydraulic conductivity testing on Bentofix clay was run with 3 permeants- de-ionized water, 0.1M CaCl2, and 0.1M NaCl.The purpose of this study was to determine if a correlation exists between the experimentally determined liquid limit and plastic limit of a specific clay and its hydraulic conductivity when exposed to a synthetic leachate. It was determined that a trend exists that will allow for less expensive and time-consuming determination for hydraulic conductivity of a clay liner when exposed to a specific chemical solution. However, more experimental data need to be collected before a definite trend is verified. The proposed procedure requires that a hydraulic conductivity test of the clay be run using deionized water as the permeant, and plasticity index tests be performed using the leachate.
Adviser: Alaa Ashmawy, Ph.D.
Co-adviser: Jeffrey Cunningham, Ph.D.
x Civil Engineering
t USF Electronic Theses and Dissertations.