USF Libraries
USF Digital Collections

Evaluation of the long term effect of inorganic leachate on geosynthetic clay liners

MISSING IMAGE

Material Information

Title:
Evaluation of the long term effect of inorganic leachate on geosynthetic clay liners
Physical Description:
Book
Language:
English
Creator:
El-Hajji, Darwish
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
Montmorillonite
Permeability
Plasticity
Swell index
X-ray diffraction
Dissertations, Academic -- Civil Engineering -- Doctoral -- USF
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Abstract:
ABSTRACT: Because of its low permeability and high swelling characteristics, bentonite is used in various hydraulic barrier systems and in the manufacturing of Geosynthetic Clay Liners (GCLs). Exposure to inorganic solutions containing elevated concentrations of electrolyte can significantly increase their permeability. To enhance the bentonite's chemical resistance to inorganic solutions, the manufacturers of GCL materials introduced propriety soluble polymeric compounds as an additive to bentonite. The resulting materials are referred to as polymer-treated, chemically-enhanced, or contaminant-resistant clays, and are arguably resistant to a host of inorganic chemicals. In this study, the response of both regular and polymer treated bentonite clays to ordinary tap water and inorganic landfill leachate is evaluated using permeability tests, index tests and x-ray diffraction. The results indicate the high dependence of performance on sample preparation techniques, pre-hydration conditions, and first wetting liquid and, to a lesser extent, polymer treatment. The x-ray diffraction results indicate that the samples reached chemical equilibrium during the permeation process, as demonstrated by a full shift in d-spacing from Na-bentonite to Ca-bentonite. Further, the results show that the cation exchange capacity, the clay plasticity ratio, and the swell index appear to be reliable indicators of the hydraulic compatibility of bentonite permeated with inorganic chemicals
Thesis:
Dissertation (Ph.D.)--University of South Florida, 2006.
Bibliography:
Includes bibliographical references.
System Details:
System requirements: World Wide Web browser and PDF reader.
System Details:
Mode of access: World Wide Web.
Statement of Responsibility:
by Darwish El-Hajji.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 98 pages.
General Note:
Includes vita.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001910885
oclc - 173610793
usfldc doi - E14-SFE0001718
usfldc handle - e14.1718
System ID:
SFS0026036:00001


This item is only available as the following downloads:


Full Text
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 001910885
003 fts
005 20071002113228.0
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 071002s2006 flu sbm 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001718
040
FHM
c FHM
035
(OCoLC)173610793
049
FHMM
090
TA145 (ONLINE)
1 100
El-Hajji, Darwish.
0 245
Evaluation of the long term effect of inorganic leachate on geosynthetic clay liners
h [electronic resource] /
by Darwish El-Hajji.
260
[Tampa, Fla] :
b University of South Florida,
2006.
3 520
ABSTRACT: Because of its low permeability and high swelling characteristics, bentonite is used in various hydraulic barrier systems and in the manufacturing of Geosynthetic Clay Liners (GCLs). Exposure to inorganic solutions containing elevated concentrations of electrolyte can significantly increase their permeability. To enhance the bentonite's chemical resistance to inorganic solutions, the manufacturers of GCL materials introduced propriety soluble polymeric compounds as an additive to bentonite. The resulting materials are referred to as polymer-treated, chemically-enhanced, or contaminant-resistant clays, and are arguably resistant to a host of inorganic chemicals. In this study, the response of both regular and polymer treated bentonite clays to ordinary tap water and inorganic landfill leachate is evaluated using permeability tests, index tests and x-ray diffraction. The results indicate the high dependence of performance on sample preparation techniques, pre-hydration conditions, and first wetting liquid and, to a lesser extent, polymer treatment. The x-ray diffraction results indicate that the samples reached chemical equilibrium during the permeation process, as demonstrated by a full shift in d-spacing from Na-bentonite to Ca-bentonite. Further, the results show that the cation exchange capacity, the clay plasticity ratio, and the swell index appear to be reliable indicators of the hydraulic compatibility of bentonite permeated with inorganic chemicals
502
Dissertation (Ph.D.)--University of South Florida, 2006.
504
Includes bibliographical references.
516
Text (Electronic dissertation) in PDF format.
538
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
500
Title from PDF of title page.
Document formatted into pages; contains 98 pages.
Includes vita.
590
Adviser: Alaa Ashmawy, Ph.D.
653
Montmorillonite.
Permeability.
Plasticity.
Swell index.
X-ray diffraction.
690
Dissertations, Academic
z USF
x Civil Engineering
Doctoral.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.1718



PAGE 1

Evaluation of the Long Term Eff ect of Inorganic Leachate on Geosynthetic Clay Liners by Darwish El-Hajji A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Civil and Environmental Engineering College of Engineering University of South Florida Major Professor: Alaa Ashmawy, Ph.D. Abla Zayed, Ph.D. Audrey Levine, Ph.D. Muhammad Rahman, Ph.D. James Chastain, Ph.D. Date of Approval: July 7, 2006 Keywords: Montmorillonite, Permeability, Plasticity, Swell Index, X-Ray Diffraction Copyright 2006, Darwish El-Hajji

PAGE 2

i Table of Contents List of Tables................................................................................................................. ....iv List of Figures................................................................................................................ .....v Abstract....................................................................................................................... .....viii Chapter 1 Introduction......................................................................................................1 1.1 Background......................................................................................................1 1.2 Research Scope................................................................................................3 1.3 Research Objective..........................................................................................3 1.4 Materials..........................................................................................................4 1.5 Testing Procedures...........................................................................................6 1.5.1 Permeability M easurements....................................................................7 1.5.2 Swell Index Test......................................................................................7 1.5.3 Liquid Limit............................................................................................8 1.5.4 Plastic Limit............................................................................................9 1.5.5 Methylene Blue Adsorption....................................................................9 1.5.6 X-Ray Diffraction.................................................................................11 1.6 Testing Facilities............................................................................................11 1.7 Dissertation Organization..............................................................................12 Chapter 2 Literature Review...........................................................................................15 2.1 Overview........................................................................................................15 2.2 Bentonite Origin and Characteristics.............................................................16 2.3 Water Adsorption and Swelling Capacity.....................................................17 2.4 Diffuse Double Layer Theory........................................................................19 2.5 Cationic Exchange Capacity..........................................................................20

PAGE 3

ii 2.6 Factors Affecting Permea bility Measurement...............................................23 2.6.1 Permeant Type and First Wetting Liquid..............................................23 2.6.2 Confining Pressure................................................................................24 2.6.3 Hydraulic Gradient................................................................................25 2.6.4 Electrolyte Concentration......................................................................26 2.7 X-Ray Diffraction..........................................................................................26 Chapter 3 Index and Permeability Testing.....................................................................30 3.1 Overview........................................................................................................30 3.2 Introduction....................................................................................................30 3.3 Bentonite in GCL...........................................................................................31 3.4 Experimental Program...................................................................................33 3.4.1 Description of Clays and Leachates......................................................33 3.4.2 Specimen Preparation and Test Parameters..........................................35 3.5 Experimental Results and Discussion............................................................36 3.5.1 Hydraulic Conductivity.........................................................................36 3.5.2 Effect of Prehydration...........................................................................39 3.5.3 Relationship to Swell Index and Mineralogy........................................41 3.6 Summary........................................................................................................47 Chapter 4 Evaluation and Modeling of Permeability-Plasticity Relationship................48 4.1 Overview........................................................................................................48 4.2 Background....................................................................................................48 4.3 Materials........................................................................................................49 4.4 Equipment and Specimen Preparation...........................................................50 4.5 Sorption Characteristics.................................................................................52 4.6 Hydraulic Conductivity vs Swell Index Relationship...................................56 4.7 Hydraulic Conductivity vs Plas ticity Ratio Relationship..............................58 4.8 Diffusion and Sorptive Retention Characteristics.........................................61 4.9 Summary........................................................................................................62

PAGE 4

iii Chapter 5 X-Ray Diffraction..........................................................................................63 5.1 Overview........................................................................................................63 5.2 Background....................................................................................................63 5.3 Bentonite Soil Mineralogy.............................................................................64 5.4 XRD Principles..............................................................................................65 5.5 XRD Methods................................................................................................66 5.6 Qualitative Analysis Method.........................................................................66 5.7 Quantitative Analysis Method.......................................................................67 5.8 XRD Equipment............................................................................................68 5.9 Sample Preparation........................................................................................70 5.10 Particle Size and Sample Grinding................................................................72 5.11 Specimen Preparation....................................................................................72 5.12 XRD Results and Discussion.........................................................................73 5.12.1 Intensity and Peak Widths.....................................................................80 5.12.2 Data Analysis........................................................................................80 5.13 Summary........................................................................................................83 Chapter 6 Conclusion and Recommendations................................................................84 6.1 Summary........................................................................................................84 6.2 Conclusions....................................................................................................85 6.3 Recommendations for Future Work..............................................................86 6.4 Engineering Implications...............................................................................89 References..................................................................................................................... ....94 About the Author...................................................................................................End Page

PAGE 5

iv List of Tables Table 1-1 Pasco County leachate chemical composition....................................................5 Table 2-1 Cationic exchange capacity (after Mitchell 1993)............................................20 Table 3-1 Bentonite properties, as supplied by the manufacturers (ASTM D5890a, ASTM D5887b, based on typical GCL thickness)..........................................34 Table 3-2 Relevant leachate properties, measured at the USF environmental laboratory.........................................................................................................34 Table 3-3 Measured hydrau lic conductivities in cm/s......................................................37 Table 3-4 Mineralogical compositi on of clays from XRD analyses.................................45 Table 4-1 Regression statistics – hydr aulic conductivity vs swell Index.........................58 Table 4-2 Regression statistics – hydraulic conductivity ratio vs plasticity ratio.............59 Table 4-3 Diffusion parameters for pure Wyoming bentonite..........................................61 Table 5-1 Leachate chemical composition.......................................................................64

PAGE 6

v List of Figures Figure 1-1 Cone penetrometer............................................................................................8 Figure 2-1 Two-to-one (2:1) clay mineral........................................................................17 Figure 2-2 Water adsorption of Ca-bent onite and Na-bentonite (adopted from Egloffstein 1995).............................................................................................18 Figure 2-3 Diffuse double la yer (adopted from Das 1990)...............................................19 Figure 2-4 Diffuse double layers surrounding clay part icles (adopted from Ruhl 1994)................................................................................................................20 Figure 2-5 Change from s odium to calcium bentonite.....................................................21 Figure 2-6 Soil particle association effect on permeability (adopted from Egloffstein 2002).............................................................................................22 Figure 2-7 Hydraulic conduc tivity vs hydraulic gradient for needle-punched GCL permeated with distilled water and Na Cl solution (adopted from Jo et al. 2001)................................................................................................................26 Figure 2-8 XRD pattern (adopted from Wikipedia).........................................................27 Figure 2-9 Constructive and destruct ive waves (adopted from Wikipedia).....................28 Figure 3-1 Variation in hydraulic conductivity with por e volumes of flow for prehydrated specimens exposed to leachate L-3 .............................................38 Figure 3-2 Effect of pr ehydration on the hydraulic conductivity of specimens exposed to leachate L-3 ...................................................................................40 Figure 3-3 Clay fabric unde r different saturation conditi ons: (a) initia l saturation with multivalent cations; (b) initia l saturation with water; and (c) prehydration followed by permeation with multivalent cations......................40 Figure 3-4 Relationship between swe ll index and hydrauli c conductivity for nonprehydrated specimens....................................................................................42

PAGE 7

vi Figure 3-5 Relationship between swell i ndex ratio and hydraulic conductivity for non-prehydrated specimens.............................................................................44 Figure 3-6 Relationship between mo ntmorillonite content and hydraulic conductivity of GCL clays..............................................................................46 Figure 4-1 Schematic of flexible wall permeability setup................................................50 Figure 4-2 Influence of sample pr eparation on the hydraulic conductivity......................52 Figure 4-3 Liquid limit vs plasticity index of GCL bent onite materials...........................54 Figure 4-4 XRD pattern of T1 bentonite: (a ) as received, (b) permeated with water, and (c) permeated with high con centration inorga nic leachate.......................56 Figure 4-5 Relationship between hydr aulic conductivity and swell index.......................57 Figure 4-6 Relationship be tween hydraulic conductivity and plasticity ratio..................59 Figure 4-7 Relationship between hydraulic conductivity and CEC..................................60 Figure 5-1 XRD pattern (reproduced from Whittig 1986)................................................65 Figure 5-2 Conceptual represen tation of x-ray diffractometer.........................................69 Figure 5-3 Photograph of XRD instrument......................................................................70 Figure 5-4 Sample preparati on technique for XRD testing..............................................74 Figure 5-5 XRD pattern for BT (GSE new treated)..........................................................75 Figure 5-6 XRD pattern for AT (GSE old treated)...........................................................75 Figure 5-7 XRD pattern for BU (GSE new untreated).....................................................76 Figure 5-8 XRD pattern fo r AU (GSE old untreated).......................................................76 Figure 5-9 XRD pattern for CT-1 (CETCO old treated #1).............................................77 Figure 5-10 XRD pattern for CT-2 (CETCO old treated #2)...........................................77 Figure 5-11 XRD pattern for CT-3 (CETCO old treated #3)...........................................78 Figure 5-12 XRD pattern for CT-4 (CETCO old treated #4)...........................................78 Figure 5-13 XRD pattern for CT-5 (CETCO new treated #1)..........................................79 Figure 5-14 XRD pattern fo r CU (CETCO untreated).....................................................79

PAGE 8

vii Figure 5-15 GCL permeabilities.......................................................................................82 Figure 6-1 Example of plas ticity ratio calculation............................................................91

PAGE 9

viii Evaluation of the Long Term Effect of Inorganic Leachate on Geosynthetic Clay Liners Darwish El-Hajji ABSTRACT Because of its low permeability and high sw elling characteristics, bentonite is used in various hydraulic barrier systems a nd in the manufacturing of Geosynthetic Clay Liners (GCLs). Exposure to inorganic soluti ons containing elevated concentrations of electrolyte can significantly increase their permeability. To enhance the bentonite’s chemical resistance to inorganic soluti ons, the manufacturers of GCL materials introduced propriety soluble polymeric com pounds as an additive to bentonite. The resulting materials are referred to as polymer-treated, chemically-enhanced, or contaminant-resistant clays, a nd are arguably resistant to a ho st of inorganic chemicals. In this study, the response of both regular a nd polymer-treated bentonite clays to ordinary tap water and inorganic landfill leachate is evaluated using permeability tests, index tests and x-ray diffraction (XRD). The results i ndicate the high dependence of performance on sample preparation techniques, prehydration conditions, and fi rst wetting liquid and, to a lesser extent, polymer treatment. The XRD results indicate that the samples reached chemical equilibrium during the permeation process, as demonstrated by a full shift in d spacing from Na-bentonite to Ca-bentonite. Further, the results s how that the cation exchange capacity, the clay plasticity ratio, and the swell index appear to be reliable

PAGE 10

ix indicators of the hydraulic compatibility of bentonite permeated with inorganic chemicals.

PAGE 11

1 Chapter 1 Introduction 1.1 Background The issue of environmental protection has been in the forefront for decades. Explosive population growth in states lik e Florida is placing extreme demands on the groundwater supply. Furthermore, population grow th leads to a substantial increase in the solid waste generation and the need to di spose of this waste in an environmentally responsible way. Past waste disposal activit ies have evolved from simply placing solid waste in pits dug in remote ar eas to the current state-of-the -art landfills. To protect groundwater supplies from potential cont amination by landfill leachate, the environmental permitting agencies mandate landfills to have a bottom containment layer that includes a synthetic liner overlaying a 2-ft thick low pe rmeability layer of compacted clay, with a hydraulic co nductivity smaller than 10-7 cm/s. Low permeability clays are not readily available in many parts of this c ountry and, in many cases, this clay has to be imported at a substantial cost. To meet the environmental agencies’ regul ations, and to provid e a cost effective alternate to the Compacted Clay Liner (CCL), the manufacturers introduced a Geosynthetic Clay Liner (GCL) that can be used as an alternate to the CCL. The GCL is manufactured by placing a thin layer of bentonite, mined from the State of Wyoming, between two geotextile filter fa brics. Because of their availability, low permeability, ease

PAGE 12

2 of use and relatively inexpe nsive construction costs compar ed to conventional CCLs, the use of GCLs has gained wide acceptan ce and is becoming the norm in landfill construction (Ashmawy et al. 2002). At the same time, a number of metropolitan areas incinerate their solid waste a nd dispose of the generated ash in the landfills. During the incineration process, slurry lime is introdu ced to reduce pollutant emission to the atmosphere. The use of lime causes the ash generated from the incineration process to have elevated concentration of electrolytes t ypically in the form of calcium salts which end up in the leachate generated by the landfill. It has been demonstrated that the bentonite’s permeability, upon exposure to calcium salts, tends to increase by orders of magnitude (e.g., Ruhl 1994, Lin and Benson 2000, Shan and Daniel 1991). The manufactur ers introduced polymer modified GCL that can maintain its low permeability (k<10-7 cm/s) upon exposure to inorganic solutions containing elevated electrolyte concentrati ons. The performance of GCL hydrated with non-standard liquids (other than water) has b een studied by others including Jo et al. 2001, and Ruhl (1994). However, these studies us ed simulated liquids rather that actual landfill leachate and, according to the authors, chemical equilibrium was not reached during the permeation process. Without achieving chemical equilibrium, the long-term performance of GCLs when permeated with highly concentrated inorganic solutions remains undocumented. According to Ashmaw y et al. 2002 the long-term performance of such materials in aggressive enviro nments has not been fully evaluated.

PAGE 13

3 1.2 Research Scope The scope of the research program is to ev aluate the suitability of the use of GCLs in lining applications containing aggressive inorgani c chemical compounds (high salt concentration), with the followi ng specific objectives in mind: 1. Simulate long-term permeability pe rformance of polymer-treated and untreated soil component of GCLs upon exposure to actual inorganic landfill leachate containing elevated electrolyte concentrations. 2. Assess the effect of the inorganic leac hate on the soil’s mineralogical makeup and validate the results. 3. Analyze data and evaluate whether a correlation can be established between index tests as early indicators to th e soil’s compatibility with the actual permeating solution. 1.3 Research Objective The research’s main objective is to establ ish a procedure that can be adopted to predict the expected perm eability performance of the materials upon hydration with inorganic leachate and prior to performing extended duration compatibility tests. Specifically, the hypothesis that there exists a relationship between index test parameters and hydraulic conductivity will be explored. While earlier studies have attempted to establish such a relationship (e.g., Jo et al. 2001 ; Lee et al. 2005), thes e studies failed to identify any significant mathematical re lationship between index parameters and hydraulic conductivity. In contra st, the principal hypothesis of the current research is the existence of a mathematical relationship between the relative values (ratios) of hydraulic

PAGE 14

4 conductivity and index parameters upon perm eation of the bentonite with water and leachate, respectively. As such, the re sults will be an indicator of the chemical compatibility of the clay with th e leachate at hand. 1.4 Materials The leachate used in this study was obtained from three facilities in Florida: Pasco County, Bay County, and Palm Beach County ash disposal landfills. The soil component of both treated and untreated GCLs was obtained from Colloid Environmental Technologies Company (CETCO) headquartered in Arlington Hei ghts, Illinois and Gundle/SLT Environmental, Inc. (GSE) hea dquartered in Houston, Texas. These two companies are the primary manufacturers and suppliers of a multitude of geosynthetic lining products in the United States. The chemical composition of the vari ous leachates was determined through chemical analysis at the University of South Florida (USF) College of Engineering Environmental Laboratory. The composition of the Pasco County leachate (Table 1-1) indicates the presence of calcium cations in high concentration, main ly in the form of calcium chloride. The chemical compositions of the other two leachates (Bay and Palm Beach Counties) exhibited smaller concentra tions of these chemicals and are presented later in the dissertation.

PAGE 15

5 Table 1-1 Pasco County leachate chemical composition This elevated electrolyte concentrations is not typical in ash landfills and is specific to Pasco County. For example, le achate generated from Bay County and West Palm Beach ash landfills has electrolyte concentrations in the range of 4,500 to 6,000 mg/L in the form of calcium chloride. Th e elevated electrolyte concentration in the Pasco County leachate is due to the end pr oduct generated by their leachate treatment facility. This facility implements the use of evaporation technology to remove the dissolved salt from the leachate. The end bypr oduct of the treatment is water and pure salt crystals. The water byproduc t is further treated at the adjacent wastewater treatment plant and the salt crystals are taken in i ndustrial-sized canvas bags and deposited in the active landfill ash cell. Subse quent rain causes the dissolutio n of the salt crystals back into the leachate collection system. This leac hing process, coupled with more salt being leached out of the deposited ash, causes the salt concentrations to be abnormally high. Therefore, the use of this leachate as the pe rmeant for this research clearly represents a conservative condition that will simulate a worst condition scenario. The GCL manufacturers use propriety polymer treatment processes to enhance the physical properties (mainly permeability) of sodium-bentonite (Na-bentonite) soils mined from the State of Wyoming that are intende d for use in a variet y of environmental applications. In the geosynthetics industry, the term Contaminant Resistant Clay (CRC) Chemical Composition Concentration Alkalinity 72 mg/L as CaCO3 Conductivity 37.6 ms/Sec pH 6.52 Total Hardness 20,000 mg/L as CaCO3 Calcium 12,800 mg/L as CaCO3

PAGE 16

6 has become synonymous with these polymer-treated products. A total of ten soil samples were obtained from these manufacturers in 1999. CETCO provided a total of six samples that consisted of five polyme r-treated samples and one untreated sample. GSE delivered a total of four samples of which two were pol ymer-treated and the ot her two were of the regular (untreated) type. These samples were sent to the USF’s Geoenvironmental Laboratory in multiple shipments. All of the samples that arrived as part of the first shipment were labeled as “Old” and the subsequent shipment were assigned the term “New”, preceded by the manuf acturer’s name in order to differentiate between the samples. Concurrently, the polymer-treated materials were assigned the symbol (T) whereas the untreated materials were labeled (U ). For example, the first polymer-treated shipment received from GSE was labeled as “GSE Old T” and the subsequent shipment was labeled as “GSE New T”. CETCO provi ded four polymer-treated samples in the first shipment followed by one treated and one untreated sample. The first shipment was labeled as “CETCO Old T #1, CETCO Old T # 2 and so on, while the subsequent polymer-treatment sample was labeled as “CETCO New T”. The untreated sample was labeled as “CETCO New U”. Visual inspection of the soils samples revealed varying textures, colors and appearances that ranged from fine to coarse a ggregate with colors va rying from light gray to charcoal. 1.5 Testing Procedures The soils samples were tested to measure their permeabilities (hydraulic conductivities), index properties and mineral ogical makeup. The results were further

PAGE 17

7 analyzed to determine whether meaningful re lationships can be established between the index properties and the materials’ permeabilit y. The index testing included the swell index tests, Atterberg limits (plasticity) te sts, and methylene blue adsorption tests. 1.5.1 Permeability Measurements The permeability measurements were perf ormed in general accordance with the ASTM D5887 method using a flexible wall pe rmeameter. In this dissertation, as in geotechnical practice, the terms permeabilit y, coefficient of permeability, and hydraulic conductivity will be used interchangeably. One of the termination criteria in the ASTM Standards stipulates the termination of the test when two pore volumes of liquid have passed through the sample. This terminati on criterion was followed for water permeation condition. When leachate was used, the pe rmeation process continued for extended periods that lasted, in some instances, three months with over 100 pore volumes of leachate passing through the samples. The read er is referred to Chapters Two and Four for detailed discussions a nd results interpretation. 1.5.2 Swell Index Test The swell index tests measure the swelli ng properties of a clay mineral in a reagent (water or leachate). The testing pro cedure is simple and entails the placing of 100 ml of permeant in a graduated cylinder. To that, a total of 2 g of the soil is added gradually and allowed to swell freely for a period of 48 hours. The final soil volume is then recorded and graphically plotted. Bentonite is believed to achieve its low permeability by virtue of its swelling behavior. This swelling constricts the pore spaces available for the permeating liquid to trav el through, which in turn reduces the

PAGE 18

8 permeability of the material. The swell index test can provide an insight to the expected permeability performance of the material. However, there have not been any mathematical models that can be used to predict the permeability based on the material’s swelling index. 1.5.3 Liquid Limit In the United States, the ATSM 423 met hod is typically used to calculate the index properties, collectively called Atterberg limits, of fine -grained (clay) soils. Because of the high plasticity of the bentoni te materials it was not possible to form a standard grove as called for in the testing procedure, and subsequently, this testing method was abandoned. Instead, the British St andard Method BS 1377:1975 was used to calculate the liquid limit. The British Standard specifies the use of a cone penetrometer apparatus, shown in Figure 1-1 below, a nd requires the mixing of 200 to 250 g of soil with distilled water and then allowing the mi xture to cure in a covered container for a period of 24 hours. Figure 1-1 Cone penetrometer

PAGE 19

9 Following the curing period, the prepared sa mple is re-mixed for a minimum of ten minutes and placed in a special meta l cup. The cup is placed under the cone penetrometer and the initial disp lacement gauge reading is reco rded. The cone is released and allowed to penetrate into the soil sample and held into that position for 20 seconds. The final gauge reading is reco rded. This step is repeated by adding more water to the soil to obtain a range of c one penetration from about 15 to 25 mm. The liquid limit versus the displacement values is plotted a nd the liquid limit value at 2 cm displacement is recorded as the liquid limit (LL). 1.5.4 Plastic Limit The plastic limit was measured in accordance with the ASTM D4318 method which requires the mixing of a small amount of water with the soil material and the rolling of the mixed material into a thread-like shape having a diameter of about 3 mm. The procedure is repeated by adding additional water until th e mixed soil material begins to crumble at the 3 mm diameter. The rolled material is weighted and placed in a drying oven to determine the water cont ent of the soil, which represents the plastic limit (PL). The plasticity index (PI) is calculated as the difference between the liquid limit and plastic limit. 1.5.5 Methylene Blue Adsorption The methylene blue adsorption test (MBA) is a semi-quantitative analysis used to determine the presence of clay minerals in a soil sample and to calculate the soil’s cationic exchange capacity (CEC). The underl ying principle is that if the amount of methylene blue adsorbed by the clay is an in dication of its CEC and, consequently, of its

PAGE 20

10 swelling potential. According to Taylor 1984, the cations in the MB solution irreversibly replace those on the exterior clay surface as follows: Ca-Na-Clay + MB hydrochloride ----MB-Clay + Ca-Na-Chloride The MBA is a simple and relatively quick lab procedure and can be performed using either the “turbidimetric” method or the "spot” method. The turbidimetric method is more accurate, but more complex, than the spot method, and is performed by the mixing of finely ground soil with methylene bl ue solution and then leaving the mixture for a few days. A spectrometer is then used to determine the amount of methylene blue adsorbed by the clay. The spot method, used for this study, is a ti tration procedure in which a methylene blue solution is added in measured quantity to a mass of finely ground soil particles, thoroughly mixed, and a single dr op is extracted from the suspension using a pipette and dropped on a filter paper. The c oncentration of the methylene blue in the solution is increased until a blue ring begins to appear ar ound the “spot” on the filter paper. The appearance of such a ring or ha lo signifies that the clay has reached its maximum adsorption capacity due to cation excha nge at the surface. The total amount of methylene blue solution adsorbed is used to calculate the CEC using the following mathematical equation: g) (mEq/100 (g) dry wt Clay g 100 x (cc) solution MB of Vol 1000 x 87 319 (g) dry wt MB x (cc) added MB C E C The methylene blue used in the current st udy is available commercially and was acquired from Fisher Scientific.

PAGE 21

11 1.5.6 X-Ray Diffraction The XRD technique was used to evaluate the soil’s mineralogical makeup in the as-received, water-permeated and leachate-per meated samples in order to determine whether chemical equilibrium was achieved during the leachate permeation process. The XRD analysis can be performed using quali tative, quantitative or semi-quantitative analysis or a combination thereof. The qua litative analysis reveals the mineralogical make up of the minerals present in a sample while the quantitative analysis is used to measure the abundance of those minerals. For this research, the qualitative analysis was used because the intent of the research is to determine whether or not any alteration has occurred in the mineralogical makeup upon leach ate permeation. The reader is referred to Chapters Two and Five of this disser tation for an expanded discussion of the XRD technique and conclusions. 1.6 Testing Facilities The various testing procedures were c onducted at the appropriate laboratories situated in the USF. The permeability and index tests were conducte d at the College of Engineering Geoenvironmental Laboratory. The permeability test apparatus consists of a state-of-the-art permeability panel with the capability of performing multiple permeability tests concurrently while mainta ining the regulated pressure for each permeability cell. At the onset of the testing program, the panel’s burettes were used as influent and effluent chambers. This pract ice was quickly terminated because the metal components of the panel began to corrode from the highly corrosive leachate.

PAGE 22

12 Subsequently, external clear plastic cells were fabricated at the College of Engineering machine shop and the metal co mponents were coated with epoxy-based paint. These tanks provided large storage cap acity for the leachate and enabled the user to suspend the permeability test to obtain samples or add additional leachate without disturbing the sample. The XRD analysis was conducted at th e USF Material Testing Laboratory equipped with a state-of-the-art Philips PW 3040/60 XRD instrument that was directly connected to a PC for data acquisition a nd control. The lab was equipped with the necessary sample preparation instruments which includes a pe stle and mortar, and sieves with varying mesh sizes. Furthermore, XR D results were analyzed using a software system, X’Pert Pro, developed by Philips. The leachate chemical analysis was perf ormed by a qualified technician at the USF College of Engineering Environmental Laboratory. The lab is fully equipped and has the capabilities of performing various ch emical analyses on a host of chemical compounds. 1.7 Dissertation Organization The organization of this dissertation is mainly based on professional journal papers co-authored by the dissertation author that were either published, submitted for publication consideration, or being submitted fo r publication. Each of these papers was authored and formatted with the intent of being a standalone document suitable for publication. These papers are in corporated in this dissertation in Chapters Three, Four and Five and as such some of the informa tion included in these pa pers may be partly

PAGE 23

13 duplicated in other parts of the dissertati on. However, the paper sequence provides an integrated view of the topic at hand a nd serves to support the hypothesis of this dissertation. Chapter Two provides a review of past lite rature and discusses related topics that include bentonite origin and characterist ics, diffuse double layer theory, cationic exchange capacity, XRD, factors affecting pe rmeability measurement, and related data published by other researchers. This chapte r provided a road map to the development and evolution of this research project. Info rmation collected from the various literature sources illustrate the gaps that exist in the published data and lead to the specific goal and objectives of this research program. Chapter Three provides a discussion of mate rials index testing that include swell index, liquid limit and plasticity index test, plastic limit and USCS classification. This chapter was published as a jour nal paper (Ashmawy et al. 20 02) and lays out the main shortcomings in the current state-of-the-art by validating some of the empirical testing relations that were proposed or established by others. The findings illustrate the inability of the current models to capture the depe ndence of bentonite permeability on the various index properties. Chapter Four includes some of the more recent data that was recently published in ASCE Geotechnical Special Pu blication No. 142 (Ashmawy et al. 2005). It discusses advection, diffusion, and sorption characteri stics of inorganic chemicals in GCL bentonite. This chapter util izes the data published by Ashm awy et al. 2005 to further examine the bentonite material characteristics and its interaction with inorganic leachate. This chapter forms the core of the dissertat ion because the data wa s analyzed with the

PAGE 24

14 purpose of developing mathematical models that can be used to pr edict the bentonite’s permeability performance and its resistance to inorganic chemicals vis--vis what is called “performance compatibility” in engine ering practice. The results and the new mathematical models are being prepared for publication in a journal paper. Chapter Five provides an in-depth disc ussion of the XRD technique and how it was used to evaluate the cha nges in lattice parameters or d -spacing of th e crystalline component of the bentonite. A detailed discus sion of the XRD data collected during the research process is presented. The material contained in this chapter is essential in verifying that chemical equilibrium, that is, full replacement of sodium cations by calcium cations, has been achieved during th e permeation process, thus validating the findings in the earlier chapters. Chapter Six is intended to provide conc lusion and present a concise summary of the testing program, and to present the key findings. Further this chapter includes a section on engineering implications intended to provide the engineering community with a useful procedure based on statistical m odels obtained through nonlinear regression analysis. This statistical re gression model can be used by design professionals to make informed decisions, early in the design proce ss, regarding GCL materi al selection and its suitability and chemical compatibility for the intended use.

PAGE 25

15 Chapter 2 Literature Review 2.1 Overview This chapter presents a comprehensive re view of the litera ture published on the permeability performance of geosynthetic cl ay liners (GCL), XRD technique and material index testing. Throughout this revi ew, the words bentonite, montmorillonite and clay may be used interchangeably despite th e subtle technical differences between these terms. Bentonite is characterized as a low permeability soil having high swelling and “self-healing” potentials. Clays are typi cally used in environmental containment applications that include la ndfill bottom liners, lagoon lining and tank farms. Bentonite has a fine powder texture and is usually mi xed with local soils us ing pug mills to reduce their permeabilities. GCLs are manufactured by placing approximately a half centimeterthick layer of pure sodium or calcium bentonite between two geotextile fabrics. The use of GCLs has evolved from simple water ir rigation pond containments to environmental containment applications to separate certa in organic and inorganic compounds from the surrounding environment. It has been illustra ted that certain orga nics and inorganic contaminants can substantially increase the permeability values of bentonite material (e.g., Ruhl 1994, Lin and Benson 2000, Shan and Daniel 1991). To counter the negative effects of these contaminants on the pe rmeability of the bentonite material, the

PAGE 26

16 manufacturers introduced a propriety treatm ent process to enable the bentonite to maintain its low permeability upon exposure to inorganic compounds. Conceptually, the treatment process involves the use of polymer materials to encapsulate the bentonite with a sacrificial layer, thus promoting the ionic exchange within that layer instead of the cations present in the bentonite (McKelvey 1996). 2.2 Bentonite Origin and Characteristics The term “clay" is used, in the technical sense, to describe fine-grained soils having particle diameter size less than 2 microns (0.002 mm) and characterized by their high plasticity index and swell potential. Benton ite is classified as clayey soil that was formed by ash spewed by volcanic eruptions an d deposited into the seas that covered much of Wyoming during the cr etaceous age. Wyoming bentonite is composed mainly of the mineral sodium (Na+) montmorillonite, which has an intrinsically high swelling capacity and low permeability values. Alternatively, calcium (Ca2+) montmorillonite, readily available in Europe and parts of th e United States, exhibits less swelling and has higher permeability va lues than the Na-montmorillonite type. Clay minerals are classified into th e following four groups 1) kaolinite; 2) smectite; 3) illite and 4) chlorite. Montmor illonite falls within the smectite group and is characterized as a 2:1 clay mineral compos ed of one gibbsite sheet held between two tetrahedral sheets as illustrated in Figure 2-1 below.

PAGE 27

17 Figure 2-1 Two-to-one (2:1) clay mineral The soil interlayers for the smectite group ar e weakly held together and can have a basal spacing ( d -spacing) that varies between 9.6 Angstrom in the dry condition to a complete separation when the soil is treate d with Ethylene Glycol (Moore and Reynolds 1989). Because the interlayer is expansible, smectites exhi bit high swelling potential. The separation between individual smectite sheets varies and depends on: 1. The interlayer cations ty pe; for example monovalent cations like sodium (Na+) cause more expansion than do di valent cations like calcium (Ca2+), 2. The concentration of ions in the surrounding solution, and 3. The amount of water present in the soil. 2.3 Water Adsorption and Swelling Capacity The liquid limit of bentonite is appr eciably high because of its high water adsorption capacity which is attributed to its small particles that have a high amount of absorbing surface. According to Egloffstein 1995, it could take up to ninety minutes for sodium bentonite to reach its maximum ad sorption capacity while calcium bentonite could reach it in about twenty minutes as illustrated in Figure 2-2 below.

PAGE 28

18 0 50 100 150 200 250 300 350 400 450 0.1110100100010000 Time (min)Water Adsorption ( % Ca-Bentonite Na-Bentonite Water Adsorption (%) Figure 2-2 Water adsorption of Ca-bentonite and Na-bentonite (adopted from Egloffstein 1995) Sodium bentonite’s high water absorp tion capacity contributes to its high plasticity, fracture resistance, high swelling and low permeability. The low permeability of bentonite is attributed to its minera logical make up and swelling capacity. For example, sodium montmorillonite can free sw ell fifteen to twenty times its original volume while calcium montmorillonite swells to about five times of its original volume (Egloffstein 1995). Upon swelling, the permeability of the soil tends to decrease because the pore spacing becomes constricted, thus reducing the available conduits for the water to travel through the soil’s fabric (Ruhl 1994).

PAGE 29

19 2.4 Diffuse Double Layer Theory According to Das 1990, clay particles carry a net negative charge on their surface mainly due to isomorphous substitution and break in the structural cont inuity at its edge. In dry clay, the electrical ne utrality is preserved by elec trostatic attraction between the exchangeable cations such as Mg2+, Ca2+, Na+ and K+ and the negatively charged particle surface. The diffuse double layer is formed when water is added to the dry clay whereby the cations and some anions encapsulate the cl ay particle. Cation c oncentration decreases with increased distance from the clay particle while anions concentration increases with increased distance. At a known distance from the clay particle, the cations and anions reach equilibrium as illustrated in Figure 2-3 below. Figure 2-3 Diffuse double layer (adopted from Das 1990) The double layer thickness is directly proportional to th e soil’s water adsorption capacity and inversely proportional to its pe rmeability. An increas e in the double layer thickness decreases the soil’s permeability by co nstricting the flow path available to the permeating liquid to travel between the soil’s particles as illustrate d in Figure 2-4 below.

PAGE 30

20 Figure 2-4 Diffuse double layers surrounding clay particles (adopted from Ruhl 1994) The high water adsorption capacity of Na-montmorillonite causes an increase in its double layer thickness and a decrease in its permeability. Alternatively, Camontmorillonite has a low water adsorption capacity which causes the diffuse double layer thickness to remain relatively unchange d and the free water flow path unobstructed thus resulting in higher permeability than Na-montmorillonite. 2.5 Cationic Exchange Capacity CEC is used to measure the soil’s activ ity and interactions with the various chemical compounds and is expressed in milliequivalents per 100 g of dry soil (mEq/100g). It can be measured using th e MBA test in accordance with the ASTM C0837 and D2330 methods. The CECs for the va rious clay minerals vary widely as illustrated in Table 2-1 below. Table 2-1 Cationic exchange capacity (after Mitchell 1993) Clay Mineral CEC (meq/100 g) Montmorillonite 80-150 Illite 10-40 Kaolinite 3-15

PAGE 31

21 Based on the above CEC values, soils com posed of the mineral montmorillonite tend to have lower permeability than those having illite or kaolinite as the main mineral. Within the same mineral group, the CEC can vary widely. For example sodium montmorillonite has a higher CEC value than that of calcium montmorillonite, as noted by Lin and Benson 2000 and Egloffstein 2002 The soil mineralogical makeup and its CEC di rectly affect the soil’s permeability. Cationic exchange is one of the mechan isms responsible for altering the soil’s mineralogical makeup. For instance, when s odium montmorillonite is permeated with inorganic liquids having elevated calcium ch loride concentration, a cationic exchange process occurs whereby the higher divale nt calcium cations replace the weaker monovalent sodium cations. This cationic exchange process continues until chemical equilibrium is reached whereby the sodium cations are completely exhausted and replaced by those of calcium (Figure 2-5) and resulting in soil having low expansion potential and higher permeability value. Figure 2-5 Change from sodium to calcium bentonite According to Brown and Anderson 1983, th e cationic exchange mechanism also affects the soil particle orientation. This al teration increases the clay permeability when its particle arrangement is altered from a welldispersed state into a flocculated one. The low permeability of Na-montmorillonite is attribut ed, in part, to its very fine soil particles Cationic Exchange N a+ N a+ N a+ N a+ N a+ N a+ Ca++ Ca++ Ca++

PAGE 32

22 that are dispersed and randomly oriented. Becau se of this particle arrangement, it would take a considerable amount of time for the permeating liquid to trav el through the soil’s fabric paths. Upon transfor mation to Ca-montmorillonite, the soil’s particles become flocculated and oriented, thus providing a direct travel path for the permeating liquid to travel through the soil fabric, resulting in higher permeability as illustrated in Figure 2-6 below. Figure 2-6 Soil particle association effect on permeability (adopted from Egloffstein 2002) Flocculation of the soil particles causes an increase in the soil’s permeability, and diminishes with increased thickness of the double layer. In turn, the characteristics of the double layer are influenced by: 1) perm eating liquid electrolyte concentration ( no ), 2) temperature ( T ), 3) ion valence ( v ) and 4) the permeating liq uid dielectric constant ( ) which is a measure of the ease with which molecules can be polarized and oriented in an electric field (Goldman et al .1998). The double layer thickness ( H ) can be calculated using the following mathemati cal relation (Mitchell 1993): 2 2 08 v e n kT H (2.1) where, k is Boltzman’s constant (1.38 x 10-16 erg/K) and e is the unit electronic charge (1.6 x 10-19 coloumb). Randomly dispersed Aggregated &Oriented Clay Particle Clay Particle Permeant Path Permeant Path

PAGE 33

23 2.6 Factors Affecting Permeability Measurement The soil’s permeability is influenced by inhe rent properties that include the soil’s mineralogical makeup, water adsorption cap acity, diffuse double layer thickness and CEC. These properties are interrelated, inte rdependent and are specific to the soil’s makeup. They can be altered by external fact ors that include permeating liquid chemical make up, test confining pressure and hydrauli c gradient. In an effort to establish uniformities while performing permeability measurement, the ASTM D5887 and D5084 methods established certain standards that must be followed during laboratory sample preparation and testing. These standards ar e specific to sample preparation methods, magnitudes of the confining pressure a nd hydraulic gradient, length of hydration, permeation, test termination crite ria and calculation procedures. 2.6.1 Permeant Type and First Wetting Liquid To perform laboratory permeability measurements, the ASTM D5084 method recommends a 48-hour hydration period using ta p water (first wetting liquid) followed by permeation with the intended chemical com pound. Water hydrati on can substantially increase the thickness of the diffuse double la yer, albeit by varyi ng degree depending on the soil’s mineralogy. The increase in the double layer thickness could result in artificially low permeability value. Ruhl 1994 reported a permeability of 2x10-9 cm/s for a GCL material that was firs t hydrated with water and then permeated with simulated landfill leachate. When leachate was used to hydrate and permeate the GCL, the measured permeability increased to 2x10-5 cm/s.

PAGE 34

24 Water, because of its chemical neutrality, is not detrimental to the soil’s low permeability while liquids containing certain organic and/or inorganic compounds can adversely affect the soil’s permeability. Fu rthermore, organic liquids like aromatic hydrocarbons can increase the montmorillonite’s permeability by order of magnitudes. When Brindley and Brown 1980 permeated be ntonite with water, they reported a permeability value of 3.6x10-8 cm/s and when the same bentonite was permeated with xylene, the permeability increased to 1.76x10-4 cm/s, almost a four-orders-of-magnitude increase. Inorganic chemical compounds like calcium chloride, magnesium chlorides and sodium chloride can significantly increase the bentonite’s permeability. Lin and Benson 2000 reported a permeability increase from 5x10-9 cm/s to 8x10-6 cm/s for bentonite material permeated with water and then with a solution cont aining 0.0125 mol/L of calcium. 2.6.2 Confining Pressure Using high confining pressure is not recommended during permeability testing because it could lead to soil consolidation thus artificially lowering its permeability. When Daniel et al. 1997 conducted permeab ility testing on GCL material using a confining pressure between 5 and 10 kPa, they achieved a permeability of 10-9 cm/s and upon increasing the confining pressure to 300 kPa, the permeability decreased by one order of magnitude. According to Ruhl 1994, when Shan and Daniel 1991 permeated a GCL material with tap wate r under a confining pressure of 4 kPa they reported a permeability of 2x10-9 cm/s and when the confining pressure increased to 140 kPa, the

PAGE 35

25 permeability decreased to 3x10-10 cm/s. The ASTM D5887 method recommends a maximum confining pressure of 35 kPa for permeability testing with no minimum value specified. The confining pressure should be specific to the inte nded use and should be consistent with the pressure pr esent under normal field conditions. 2.6.3 Hydraulic Gradient It is the norm to use high hydraulic gradie nt to perform permeability tests for fine grained soils. Due to their low permeabilities it would take consid erable amount of time for the permeant to fully saturate and penetrat e the soil layer. Conve rsely, Jo et al. 2001 recommend the hydraulic gradient be as close as what can be expected in the field under actual conditions. Testing with gradients higher than that could result in an artificially low permeability due to sample conso lidation. Although the ASTM D5887 method recommends a maximum hydraulic gradient of 30, researchers and pr actitioners continue to test low permeability soils using elevated gradients. Hydraulic gradients that ranged from 50 to 550 were reportedly used by others for the same material with no noticeable change of the reported permeability values. Jo et al. 2001 reports that Rad et al. 1994 demonstrated that using a hydr aulic gradient of 2800 to permeate a GCL sample with water had no measurable effect on the permeability value, and Petrov et al. 1997 illustrated the same findings as shown in Figure 2-7.

PAGE 36

26 Figure 2-7 Hydraulic conductivity vs hydraulic gradient for needle-punched GCL permeated with distilled water and NaCl solution (adopted from Jo et al. 2001) 2.6.4 Electrolyte Concentration According to Sivapullaiah and Savitha 1999 the inter-particle separation of the bentonite particles in water can lead to a higher swell index while electrolyte solutions tend to limit or inhibit the inter-particle sepa ration, thus leading to a lower swell index and a potential decrease in the diffuse double layer thickness. Sykes et al. (1982) reported a permeability of 4.14x10-8 cm/s when the salt concentration in the permeating liquid was at zero percent, but when the salt concentration increased to five percent, the bentonite permeability increased to 1.31x10-7 cm/s. 2.7 X-Ray Diffraction The theory and practical applications of XRD techniques are well documented in the literature (e.g., Brindley and Brown 1980; Whittig 1986; Bueno 2002 (). XRD is used to determine the mineralogical ma ke up of soil samples in qualitative and quantitative manners, and will be used for that purpose in this dissertation. According to

PAGE 37

27 Whittig 1986 x-rays are electromagnetic radi ation generated from the oscillation of electrostatic and electromagnetic fields of a wavelength equal to the interatomic distance in a crystal. XRD is based on Bragg’s law, and depends on the crystal’s physical properties. Crystalline structures are formed by three dime nsional arrangements of atoms situated at a fixed interplanar distance, commonly referred to as d -spacing or basal spacing, within a crystal. Diffraction occurs when the electromagnetic waves with a wavelength bombard an atom causing the incident wave to refract. Successive diffraction enforces the diffracted beam resulting in signal strength sufficient for recording as illustrated in Figure 2-8 below. Figure 2-8 XRD pattern (adopted from Wikipedia) The XRD pattern is expressed mathematically using Bragg’s law (E quation 2.2) that must be satisfied for the diffraction process to be successful. n = 2 d sin (2.2)

PAGE 38

28 where n is an integer, is the wavelength of the x-rays, d is the spacing (basal spacing) between the planes in the atomic lattice, and is the angle between the incident ray and the refracted beam. According to Moore and Reynolds 1989, the diffraction process and the resulting diffraction data are highly dependable on th e sample preparation techniques. Poor sample preparation could result in a dest ructive diffraction patte rn while a good sample preparation technique can result in a c onstructive diffraction pa ttern. Constructive diffraction occurs when two electromagnetic waves are in phase whereby their troughs and peaks line up thus magnifying the amplitude of the refracted wave. Alternatively, destructive diffraction occurs when the waves are 180 out of phase and the resulting troughs of one wave line up with the peaks of the next wave resulting in waves having little or no amplitude as illustrated in Figure 2-9 below. Figure 2-9 Constructive and destructive waves (adopted from Wikipedia) The types of refracted waves have an impact on the XRD results whereas narrow peaks with lower amplitude as an indication that destructive interference was introduced during the diffraction process while wider peak s with higher amplitude is an indication that constructive waves were r ecorded (Moore and Reynolds 1989).

PAGE 39

29 Depending on the scope of the study, the XRD process can be conducted with a qualitative or quantitative anal ysis. The qualitative analys is is the simpler of two methods and little experience is required to pe rform the analysis, whereas the quantitative analysis requires a great deal of experi ence and, if not used properly, can lead the researchers to wrong conclusions (Moore and Reynolds 1989). The quantitative analysis is conducted using either an internal standa rd method or an external standard method. According to Brindley and Brown 1980 the c hoice of method depends largely on the type of material being analyzed. The internal st andard method is typically used in analyzing powdered soil specimens and is applied to mineral or materials for which the composition is unknown. Alternatively, the external standa rd method is used on solid materials like alloys and allows the quantific ation of one or more components in the material while the direct comparison method is only applicable to fully crystalline mixtures and does not require any standards. There are various potential sources of erro rs that could occur before, during and after the diffraction process has been co mpleted which could cause the user to misinterpret the data. The mo st common type of error that occurs before the diffraction process involves sample preparation techniques. The sample must be prepared with great care to ensure that near random orientation of the soil particles is achieved so that each set of crystals can be diffracted. Errors c ould also result during the diffraction process by the improper mounting of the sample and instrument misalignment. Following the diffraction process, data misint erpretation is the major source of error and usually occurs due to the lack of user’s experience.

PAGE 40

30 Chapter 3 Index and Permeability Testing 3.1 Overview In this chapter, experimental results are presented to evaluate the immediate change in hydraulic conductivity of seve n types of GCL clays upon permeation with leachate generated from three ash landfills. The composition of the ash, which is a byproduct of the incineration of municipal soli d waste (MSW), in turn influences the composition of the resulting leachate. Falling head permeability tests were performed on flexible-wall permeameter specimens, with b ack-pressure saturation. Chemical analysis shows that the three leachate products cont ain high, medium, and low concentration calcium and magnesium cations. The result s are interpreted using existing models developed by other researchers, and the limita tions of such models are discussed. Most notably, the inability of such models to provide a systematic methodology or a general framework for evaluating the compatibility of bentonite with specific inorganic chemicals is demonstrated. 3.2 Introduction The clay component of GCL materials tested in this study consists of regular and polymer-treated bentonite. As mentioned in Chapters One and Two, polymer treatment arguably renders the clay non-reactive to ma ny organic and inorganic chemicals. In municipal solid waste landfills where the inorganic or organic contaminant concentration

PAGE 41

31 in the leachate is low, lining systems can rely on natural or untreated bentonites (Egloffstein 1995; Ruhl and Daniel 1997). Such GCL be ntonites typically contain natural sodium montmorillonite although calcium montmorillonite is sometimes used. In regions where leachates with hi gh concentrations of contaminant are present, the use of polymer-treated or polymer-coa ted bentonite is beneficial as it possibly renders the montmorillonite non-reactive towards most organic and inorganic chemical compounds. Laboratory test results (El-Hajj i et al. 2001) suggest that pol ymer-treated GCLs maintain their low hydraulic conductivity when hydrat ed with liquid containing single-species solutions at low concentration. In this chap ter, laboratory data are presented for seven untreated and treated GCL bentonites permeated with natural landfill leachate solutions with different concentrations. The leachat es were obtained from three MSW landfills where large quantities of inci nerator ash rich in calcium and magnesium are disposed. 3.3 Bentonite in GCL Bentonites, a key component in commerc ially-manufactured GCLs, are natural clays formed as a result of mechanical and chemical weathering of volcanic ash that has been deposited in salt or fresh water. Sodium and Calcium montmorillonite, which constitute the main mineral in bentonite clay s, are present in salt and fresh water deposits, respectively. Bentonite exhibits low hydraulic conductivity (up to 10-10 cm/s), high fluid adsorption capacity, and can swell to severa l times its original volume. Its swelling capacity depends on factors such as mineral composition, grain size, aggregate size, CEC, chemical concentration of the permeating li quid, and chemical composition of the first wetting liquid. Extrinsic factor s that affect the hydrauli c conductivity of bentonite

PAGE 42

32 include, among other things, confining pr essure, laboratory pr essure-saturation conditions, and hydraulic gradient (Shackelford et al. 2000). The hydraulic conductivity of bentonite can increase gradually, albeit by several orders of magnitude, due to alterations in it s micro-fabric, stemming from a decrease in the diffuse double layer thickness. Such a decrease in th ickness can occur upon increasing the solution concentration, or as a result of the replacemen t of monovalent ions (e.g., Na+) with higher valence cati ons present in an intruding inorganic solution. Changes in the bentonite hydraulic conductivity by several orders of magnitude have been reported in the literature upon permeati on with high concentra tion solutions (e.g., Petrov et al. 1997; Shackelford et al. 2000; El-Hajji et al. 2001). Over the past several years, GCL ma nufacturers introduced products such as CRCs or polymer-treated bentonites (PTB ) to minimize bentonite degradation upon exposure to contaminants. Polymer treatment processes and formulas used in GCL manufacturing are proprietar y and are not documented (Ruhl and Daniel 1997; Kajita 1997). According to Theng 1979, clay-polymer interactions can be classified according to the polymer’s surface charge; uncharged polymers are electrically neutral whereas anionic and cationic polymers carry net negative and positiv e surface charges, respectively. The chemical modification mech anism in bentonite clays, which involves the use of cationic polymers, has been descri bed extensively in the literature since the 1970’s (e.g. Pezerat and Vallet 1973; Bart et al. 1979). Two mechanisms exist by which polymer treatment enhances the resistance of bentonite to cation exchange, the main cause of flocculation which, in turn, results in an increase in hydraulic conductivity. The first mechanism involves the replacement, during

PAGE 43

33 manufacturing, of the adsorbed sodium or calci um ions in the clay by a cationic polymer. This process is irreversible because a single polymer chain contains thousands of cations which would need to be displaced simultaneously if cation exchange is to take place later. The second mechanism relies on a weaker bond (dipole attraction) between the cationic polymer and the sodium ions. In this case, sodium ions are not re placed, but rather the sodium montmorillonite sheets are “coated” with the polymer. 3.4 Experimental Program 3.4.1 Description of Clays and Leachates In the present study, seven bentonite clay s, produced by two GCL manufacturers were used. Three were desc ribed by the manufacturers as “polymer treated”, although the exact treatment process was not disclo sed. Some of the samples were obtained directly in the form of a dr y clay powder from the manufacturer while others were extracted for the testing program by cutting a sample of the parent GCL material and extracting the clay component. Table 3-1 cont ains a description of the seven bentonites with the nominal properties provided by the ma nufacturers. The principal mineral in all seven clays is sodium montmorillonite. The l eachates were obtained from incinerator ash disposal facilities in three different counties in the Stat e of Florida: L-1 from Bay County, L-2 from Palm Beach County, and L-3 from Pasco County. The Pasco County facility ( L-3 ) is an ash monofill while the other tw o facilities use landfill co-disposal where both incinerator ash and MSW are dispos ed. Table 3-2 summarizes the relevant chemical characteristics of the leachates.

PAGE 44

34 Table 3-1 Bentonite properties, as supplied by the manufacturers (ASTM D5890a, ASTM D5887b, based on typical GCL thickness) Manufacturer Treatment Label Swell Index a ml/2g Hydraulic conductivity b cm/s GSE Untreated AU 24 510-9 GSE Treated AT 24 N/A GSE (Bentofix) Untreated BU 24 510-9 GSE (Bentofix) Treated BT 24 110-9 CETCO Untreated CU 24 510-9 CETCO Treated CT 1 24 510-9 CETCO Treated CT 2 24 510-9 Table 3-2 Relevant leachate properties, meas ured at the USF environmental laboratory Leachate L-1 L-2 L-3 Landfill type Co-disposal Co-disposal Ash monofill pH 6.30 6.55 7.24 Cations (mg/L) Ammonia, NH4 + 60 260 15 Sodium, Na+ 2,200 300 1,640 Calcium, Ca2+ 1,150 2,625 5,120 Magnesium, Mg2+ 750 1,525 2,075 Anions (mg/L) Chloride, Cl – 5,300 3,800 6,836 Bicarbonate, HCO3 – 227 870 100 Sulfur Species 920 7 1,100 Other (mg/L) 33 13 14 Total Dissolved Solids (mg/L) 10,640 9,400 16,900

PAGE 45

35 All three leachates were found to possess high levels of calcium and magnesium compared to other multivalent cations (e.g., al uminum, copper, and zinc). The abundance of calcium and magnesium, in the form of ch loride, bicarbonate and sulfide, has been traced back to the chemicals added during the waste incineration process. Because of the aggressive nature of the ash chemistry, the c oncentrations reported here are significantly higher than those reported by earlier research ers for real and simulated MSW leachates and salt solutions (e.g. Ruhl and Daniel 1997 ; Shackelford et al. 2000; Jo et al. 2001). 3.4.2 Specimen Preparation and Test Parameters Falling head permeability tests with back-pressure saturation were performed following the ASTM D5887 method. The only exception entailed the preparation of specimens from the clay component only rather than the intact GCL with the geosynthetic backing. For consistency purposes 60 g of dry bentonite were tamped to a constant thickness of 6 mm in a specially fabricated mold to produce the 100 mm diameter specimens, which were then placed in the flexible wall permeameter. Test parameters including confining pressure, back pressure saturation, and hydraulic gradient were followed per the ASTM D5887 met hod. The experimental design randomized block model (Neter et al 1990) was implemented to minimize the number of experiments. The model permits the selection of specific combinations of parameters in multifactor investigations while avoiding parame ter bias due to systematic or subjective selection. Consequently, not all combinations of clays and leachates were tested in the present study.

PAGE 46

36 In order to simulate the worst possible field conditions, the specimens were backpressure saturated directly with the leachat e and allowed to soak for 48 hours before the initial hydraulic conduct ivity reading was recorded. Howe ver, five additional specimens subjected to leachate L-3 were prehydrated to investigate the influence of the first wetting liquid on the results. Swell Index test s per the ASTM D5890 method were also performed on all clays with all leachates. 3.5 Experimental Results and Discussion 3.5.1 Hydraulic Conductivity To measure the variation of hydraulic conductivity with quantity of flow, a constant gradient of approximately 150 was maintained, and measurements were recorded at regular intervals. Both prehydrated and non-pr ehydrated specimens exhibited an increase in hydraulic conductivity with time. Each test was continued until no trend was observed in the readings, and a st eady value was obtained for the hydraulic conductivity, following the criteria proposed by Ruhl and Daniel ( 1997), Shackelford et al. 2000 and Jo et al. 2001. Due to the use of back-pressure satu ration technique and equipment limitations, it has not been possible to sample the influent and effluent for chemical analysis. Therefore, chemical equilibrium was not established during testing, and the tests are not necessarily re presentative of long-term conditions. The hydraulic conductivity values given in Table 3-3 indicate that leachate L-3 was very detrimental, even to polymer-treated clays. Some of these measurements were confirmed through independent testing by a co mmercial lab in Orlando, Florida, and by

PAGE 47

37 one of the GCL manufacturers. Initial satu ration with the leachate resulted in high hydraulic conductivities, with the values bei ng stable from the beginning of the test. Table 3-3 Measured hydraulic conductivities in cm/s Leachate Clay Water L-1 L-2 L-3 L-3 Prehydrated AU < 910-9 1.310-5 4.510-8 AT < 310-9 1.210-6 1.010-8 BU < 110-8 1.510-8 BT < 110-8 1.210-8 CU < 310-9 4.010-9 8.510-6 3.110-8 CT1 < 110-7 1.410-6 2.010-5 CT2 < 510-8 1.610-8 7.510-7 This implies that the cation exchange pr ocess occurs almost instantaneously upon initial saturation. On the other hand, prehydrat ed specimens exhibited a gradual increase in hydraulic conductivity, but a steady-stat e condition was always reached beyond a maximum quantity of flow of 3 pore volumes (Figure 3-1). Shackelford et al. 2000 reported a gradual increase in hydraulic condu ctivity up to at least 40 pore volumes of flow when a diluted CaCl2 solution was used. Ruhl and Daniel 1997 found that diluted leachates from MSW facilities did not cause significant degradation in the hydraulic properties of untreated GCLs, even up to nine pore volumes of permeation. In contrast, the results presented here, as well as t hose published by Jo et al. 2001 suggest that significant degradation and steady values can be obtained at smalle r pore volumes. This

PAGE 48

38 discrepancy can be attributed to the low con centration of the solutions used by the former researchers. Pore Volumes of Flow 02468 AU AT CU CT-1 CT-2 10-910-810-710-610-510-4Hydraulic Conductivity (cm/s) Figure 3-1 Variation in hydraulic conductivity with pore volumes of flow for prehydrated specimens exposed to leachate L-3 According to the current state of practic e worldwide, hydraulic conductivities of approximately 10-8 cm/s are typically specified as the minimum acceptable for GCLs in landfill liner applications. A more specific value for the acceptable hydraulic conductivity can be calculated by demonstrating hydraulic comp atibility of the GCL with a 300-mm thick clay liner. The results in Table 3-3 indicate th at all materials exhibited hydraulic conductivities higher than those published by the manufacturers, even though the difference is small in most cases. Both untr eated and treated bentonites were marginally acceptable under L-1 and L-2 In contrast, L-3 caused significant degradation in the

PAGE 49

39 majority of the cases. Specifically, all specimens directly saturated with leachate L-3 were well above the acceptab le hydraulic conductivity. Even though this is clearly attributed to the unusually aggressive nature of the leachate, it is interesting to note that even polymer-treated clays did not perform as expected. Chemical tests on leachate samples obtained several months apart from the L-3 landfill indicate that the calcium and magnesium concentrations were consistently high as a result of the ash monofill practice at the waste disposal facility and therefore pose a possible threat to the existing landfill liner. 3.5.2 Effect of Prehydration The results presented in Table 3-3 indicat e a significant difference in hydraulic conductivity, depending on the initial wetting conditions. Prehydrated specimens performed, in general, far better than those e xposed directly to the le achate, even at large numbers of pore volumes of flow (Figure 3-2) Petrov et al. 1997 and Shackelford et al. 2000 reported similar findings. The interpretatio n of such behavior can be drawn from information available in various literature sources on the particle arrangement of clay particles (van Olphen 1977; Mitchell 1993; Then g 1979; Pusch 1998). In general, initial saturation of sodium montmorillonite with a multivalent-rich solution, such as calcium chloride, causes the divalent cations (Ca2+) to immediately occupy most of the cationic sites originally taken by the sodium. This abrupt transformation limits water migration into the interlayer space because the electros tatic forces between the cation and the clay particle surface are larger than hydration forces of the divalent cation. Consequently, an aggregated structure with macro-voi ds is achieved (Figure 3-3(a)).

PAGE 50

40 Hydraulic Conductivity to Water (cm/s) NonPrehydrated 10-910-810-710-610-510-4Hydraulic Conductivity (cm/s) 10-910-810-710-6Prehydrated Water Figure 3-2 Effect of prehydration on the hydraulic conductivity of specimens exposed to leachate L-3 (a)(b)(c) Figure 3-3 Clay fabric under different saturati on conditions: (a) initial saturation with multivalent cations; (b) initial saturation wi th water; and (c) prehydration followed by permeation with multivalent cations In contrast, initial hydration of the monovalent cations (Na+) present in sodium bentonite attracts large quanti ties of water into the interl ayer space, thereby creating a dispersed structure, as shown in Figure 3-3(b). The interlayer spacing in this case can be as large as tens of nanometers, which cau ses extensive swelling of the clay. Upon

PAGE 51

41 permeation with a leachate containing mu ltivalent cations, the spacing between the platelets should gradually decrease due to changes in the double layer thickness, but the dispersed and uniform arrangement of the clay fabric is retained (F igure 3-3(c)). The homogeneous arrangement of the clay platelet s in this case results in a lower hydraulic conductivity than when the particles are aggr egated. Shackelford et al. 2000, however, appropriately question the effectiveness of prehydration on the long-term hydraulic conductivity especially that the tests available in th e literature do not go beyond a limited number of pore volumes. Such longterm tests as well as procedures to investigate the mechanism of improvement due to prehydration are being currently researched by the authors. A large variability in results was obser ved between samples obtained from the same source. Samples CT-1 and CT-2 were both intended to be marketed as the same brand polymer-treated GCL product. Howeve r, upon visual and manual inspection of the samples it was evident that they were different namely in terms of color and aggregation. Large bentonite agglomerations, in general, flocculate upon satura tion, especially when cations are present in the so lution (Shackelford et al. 2000). In turn, much higher hydraulic conductivities are obtaine d. Sample CT-1 indeed contained a high percentage of bentonite agglomerations and exhibited unusually high conductivity with both water and prehydrated L-3 leachate. 3.5.3 Relationship to Swell Index and Mineralogy While performing the permeability experiments, higher swelling was observed in conjunction with specimens with lower hydrau lic conductivity, and th e opposite was true

PAGE 52

42 for highly permeable specimens. The relatio nship between swell behavior and hydraulic conductivity has been studied by Jo et al. 2001. A strong correlation was found to exist in their study and was attributed to the fact that similar mechanisms control both the swelling behavior and the hydraulic conductivity A similar trend was observed in the present study, as shown in Figure 3-4. A math ematical model is proposed, together with the model statistics, in Chapter Four of th is dissertation to relate swell index and hydraulic conductivity. Swell Index (ml/2g) 010203040 AU AT BU BT CT-1 10-910-810-710-610-510-4Hydraulic Conductivity (cm/s)CT-2 Figure 3-4 Relationship between swell index and hydraulic conductivity for nonprehydrated specimens Although normalization of the swell inde x and the hydraulic conductivity per Jo et al. 2001 did not produce a significant correl ation, the results pres ented in Figure 3-5 support their conclusion that the hydraulic c onductivity ratio is constant for free swell ratios larger than 20. Here, the hydraulic conductivity ratio is defined as the ratio

PAGE 53

43 between kleachate and kwater. The free swell ratio is define d as the free swell index divided by the volume of solids contained in 2 g of clay. The level of scatter in the data was very high for low swell combinations, mainly because the hydraulic conductivity is sensitive to changes in soil fabric. Polymer-treated clays consistently produced s lightly lower swell indexes co mpared to their untreated counterparts. Overall, a swell index of 25 or higher always resulted in hydraulic conductivities within the acceptable range. Howeve r, the free swell of the CU clay (not shown in Figure 3-4 and Figure 3-5), was uncha racteristically high compared to the other materials; values as high as 50 and 90 ml/2g-clay were obtained with leachate L-3 and water, respectively. Based on the data presen ted, it is recommended that the swell index ratio be used with caution when correlati ng with the hydraulic c onductivity ratio. In other words, the data at hand does not support the implementation of swell index ratio as an early indicator of bentonite comp atibility with inorganic chemicals.

PAGE 54

44 Free Swell Ratio 010203040 AU AT BU BT CT-1 10-1100101102103104Hydraulic Conductivity RatioCT-2 50 Figure 3-5 Relationship between swell inde x ratio and hydraulic conductivity for nonprehydrated specimens To evaluate the influence of the mineral composition on the hydraulic and swelling characteristics, a seri es of XRD tests was conducted. Aggregated clays were grinded to a size of approximately 5 microns and dried at a low temperature of 50C to avoid shifting in the diffraction peaks. Unoriented powder samples were prepared following the Poppe et al. (2001) back-loading procedure, and tested in the Philips PW3040 Theta-2 diffractometer. Incident a nd diffracted beam optics settings were selected based on values recommended by M oore and Reynolds 1989. A more detailed description of the XRD testi ng program, results, and interp retation is given in Chapter Five. For quantitative interpretation of the diffraction patterns, the reference intensity ratio method was employed, with mineral intensity factors reported by Burnett 1995,

PAGE 55

45 Moore and Reynolds (1989), and Hillier 2000. The results of the analyses are summarized in Table 3-4. Table 3-4 Mineralogical composition of clays from XRD analyses Weight Percentage Mineral AU AT BU BT CU CT-1 CT-2 Montmorillonite 92 91 73 78 56 49 74 Illite 15 10 22 39 12 Quartz & Cristobalite 3 4 4 7 12 8 6 Feldspar 4 5 6 6 6 3 8 Calcite <2 <2 <2 <2 <2 Gypsum <2 Siderite, Mica, and others <2 <2 Because of the abundance of montmorillonite in bentonites, as well as its large specific surface and surface charge, it is often identified as the main mineral affecting the hydraulic and swelling properties of clays. In addition, because of severe changes in the montmorillonite double layer chemistry upon exposure to different leachates, significant changes can occur in the hydraulic conductiv ity. Figure 3-6 shows the relationship between montmorillonite content and hydraulic conductivity for all samples tested in this study, with the exception of CU whic h, as mentioned earlier, exhibited uncharacteristically high free sw ell and low hydraulic conductivity. The results plotted in Figure 3-6 indicat e that the hydraulic conductivity decreases as montmorillonite content increases for: 1) samples permeated with water, and 2) prehydrated samples permeated with L-3 leachate. The data available for L-1 and L-2

PAGE 56

46 leachates are insufficient for establishing a patt ern, but the values are comparable to those of water, possibly because of the lower concentration of calcium in solution. On the other hand, the hydraulic conductivity appears to be insensitive to montmorillonite content when the samples are initially saturated with leachate L-3. This is in agreement with the mechanism described in Figure 3-3, as well as with data published by Petrov et al. 1997 and Shackelford et al. 2000. Percent Montmorillonite 406080 10-910-810-710-610-510-4Hydraulic Conductivity (cm/s)100 L-3(prehydrated) Water L-1 L-2 L-3 Untreated clay Treated clay Figure 3-6 Relationship between montmorillo nite content and hydraulic conductivity of GCL clays The results also indicate that polymer tr eatment is not significantly beneficial when the clay is permeated with water or low concentration solutions (e.g. L-2). When high concentration solu tions are used (e.g. L-3), and at a given montmorillonite content, polymer-treated GCL clays exhibit, at best, hydraulic conductiv ities one order of magnitude smaller than the corresponding untreated clay. Although such level of

PAGE 57

47 improvement may be adequate in certain si tuations, the resulting hydraulic conductivities are generally higher than typically specified for landfill GCLs. Therefore, acceptance of polymer-treated clays as hydraulic barriers in aggressive inorganic environments should not be recommended without proper verificat ion of their chemi cal compatibility. 3.6 Summary Hydraulic conductivity tests we re conducted to evaluate the response of untreated and polymer-treated bentonites to water and as h fill leachates, and to compare the results to those published in ea rlier studies. The results were interpreted using methods and models available in the literature. The prelim inary findings of this study indicate that: 1) polymer treatment is generally more beneficial if the clay is first saturated with water and not directly with the leacha te, 2) high swell potential of the bentonite is more advantageous than polymer treatment, espe cially when low hydraulic conductivity is required in the short term and if the clay is prehydrated, 3) prehydrat ion is generally more beneficial than the use of treated bentonites fo r up to at least eight pore volumes of flow, and 4) a relationship exists between the sw ell index and the hydraulic conductivity, as suggested in earlier studies. Even th ough all leachates contained relatively high concentrations of divalent cations, both untreated and treated bentonites produced acceptable results, except when extremely hi gh concentration solutions were used. Compared to untreated GCL clays with si milar mineralogical composition, polymertreated clays exhibit only limited improveme nt in terms of reduction in hydraulic conductivity.

PAGE 58

48 Chapter 4 Evaluation and Modeling of Permeability-Plasticity Relationship 4.1 Overview In this chapter, the geotechnical inde x properties of the bentonite clay are measured, and their use as early indicators of chemical compatibility of bentonite is explored. The specific geotechnical index pr operties addressed in this study are liquid and plastic limits (Atterberg limits) and swell in dex. In addition, the advection, diffusion, and sorption characteristics of untreated and polymer-treated bentonite clays are examined in this chapter. Long-term a dvection properties are determined from the flexible and rigid wall permeability tests described in Chapter Three, where both chemical and hydraulic equilibriu m are established. In this chapter, it is demonstrated that the cation exchange capacity, the swell index, and the clay plasticity ratio appear to be more reliable than the swell index ratio and the plasticity index as indicators of the hydraulic performance and chemical compatibil ity of bentonite permeated with inorganic chemicals. 4.2 Background Little work has been conducted to rela te the permeability of bentonite to its plasticity behavior. In a most recent study by Lee et al. 2005, no significant relationship was found between the plasticity index of bentonite clay and its hydraulic conductivity when permeated with inorgani c solutions. In addition, on ly a few studies have been

PAGE 59

49 conducted to evaluate the sorptive, diffusion, and retention characte ristics of bentonite materials within the context of their e xposure, as GCL components, to chemicals abundant in landfill leachates. While Eglo ffstein 1995 provided a general overview of candidate test methods for comprehensive evaluation of the chemical and hydraulic compatibility of GCLs, little research ha s since been conducted in this regard. In this chapter, the experimental re sults presented in earlier chapters are complemented with additional tests to de velop a relationship between permeability and index tests using a new framework. The result s provide a broad perspe ctive of the factors relevant to long term design of GCL-base d leachate containment systems, while the model presents a specific methodology to be followed as a means of evaluating the chemical compatibility of a GCL material. The same untreated and polymer-treated bentonite materials described in earlier chapters were te sted for hydraulic conductivity, ionic sorption characteristics, and index pr operties, while diffusion and retention data on only untreated bentonite are pres ented. In addition, rigid wa ll tests were also performed to verify the influence of sample preparati on procedures and test methods on the results. 4.3 Materials Eleven types of commercially availa ble Wyoming sodium bentonite, with different index properties, were tested and are reported in this chapter, including the seven materials presented in Chapter Three. One additional untreated bentonite material (CU-2), and one polymer-treated benton ite (CT-3) were provided by the same manufacturers. While bentonite, in a ge neral sense, is composed mostly of montmorillonite, the type and relative abundance of other minerals such as quartz, illite,

PAGE 60

50 calcite, and feldspar, can have a significan t influence on its engi neering properties. Moreover, the properties of bentonite depe nd not only on the source and composition, but also on preparation processes such as wetting and drying, heating, chemical modification, and mechanical grinding. Engineering prope rties of identical GCL materials obtained from the same source that may be altere d during manufacturing include consistency limits, swell properties, hydrau lic conductivity, and long-term chemical compatibility. 4.4 Equipment and Specimen Preparation Permeability tests were conducted using both flexible and rigid wall permeameters. The specimens were back-pre ssure saturated with water (prehydrated) or permeant (non-hydrated) before being subjected to a hydraulic gradient of 250. Special interconnected buffer tanks with different capaci ties were used to store the influent and effluent solutions, and were equipped with three-way valves to sample the solution for chemical analysis, as shown in Figure 4-1. Figure 4-1 Schematic of flexible wall permeability setup

PAGE 61

51 By sampling the leachate while the test is runni ng, this setting allowe d for verification of steady-state electrical conductiv ity and pH conditions (chemical stability) as an additional condition for terminating the tests. The void ratio in the rigid wall tests was controlled by maintaining a constant specimen height. Specimen preparation is a cr ucial component of GCL tes ting, and the preparation procedure recommended in the ASTM D 5887 and D6766 methods resulted in a nonuniform specimen thickness. Post-testi ng inspection of the permeated specimens indicated that the thickness was reduced near the edges due to swe lling and material loss while cutting the sample. Channels and cracks around the specimen edges were also observed. To ensure homogeneity of the specimens, the bentonite component was first removed from the GCL, and a given weight of dry bentonite was tamped to a constant height of 7.5 mm into a 101-mm diameter rigid mold within the permeameter, in a procedure similar to that used to prepare sand samples. The rigid mold was removed before testing, and the specimen shape was preserved by applying vacuum before removing the mold. The specimens prepared us ing this method gave repeatable results, and post-testing visual inspection showed no si gn of cracking. The only drawback of the method is that only the clay component of the GCL is tested as the geotextile backing is discarded. However, the method allows fo r unbiased comparisons between the hydraulic conductivities of different clay s since the initial mass and th ickness are constant. Figure 4-2 illustrates the importance of sample preparation on the measurement of hydraulic conductivity of GCL bentonite.

PAGE 62

52 Figure 4-2 Influence of sample preparation on the hydraulic conductivity Samples of five different clay types pr epared using the ASTM method exhibited hydraulic conductivity values up to two orders of magnitude greater than the USF method at an effective confining pressure of 30 kPa. The effect was particularly prominent when the samples were permeated with non-hydrated high concentration leachate from an ash landfill. The leachate contained calcium a nd magnesium chlorides at concentrations close to the saturation li mit of the solution (Ca2+ at 5000 mg/L and Mg2+ at 2000 mg/L). 4.5 Sorption Characteristics Exposure to inorganic chemicals causes gradual degradation of the hydraulic performance of bentonite, until the total sorptive capacity of the clay is reached. Sorption of inorganic cations occurs mainly through ionic bonding by replacement of the existing cations – usually Na+ – in the diffuse double layer of the clay with higher valence cations such as Mg2+ and Ca2+. The long-term resistance of be ntonite to inorganic chemicals can

PAGE 63

53 be enhanced by sorbing polymers or other organic compounds during preparation (Soule and Burns 2001; Ashmawy et al. 2002). Depending on their chemical composition, sorption of organics occurs through a combination of ion exchange and adsorption by van der Waals forces (Soule and Burns 2001). The sorption of organic compounds in th is case is viewed as beneficial, as the organic sorbents shield the clay surface from interactions with other detrimental organic or inorganic chemicals. However, the additi on of large quantities of organics can cause the total sorptive capacity of the clay to be reached, and the hydraulic conductivity of the clay to consequently increase. This can be prevented by mixing small quantities of treated or amended clay into a larger quantity of pure benton ite. The treated fraction of the material adds to the chemical resistance, while the pure bentonite helps to maintain the low hydraulic conductivity. The sorptive capacity of bentonite is rela ted to its CEC, which can be measured directly or indirectly by a variety of methods including sodium saturation (Wentink and Etzel 1972), XRD analysis (Kaufhold et al 2002), infrared chromatography (Hwang and Dixon 2000), or chemical adsorption of a wide array of chemicals. MB adsorption is widely accepted as one of the most relia ble methods to obtain information on the properties of clay minerals, including CEC (Grim 1968; Santamarina et al. 2002). It is also used as an indirect quality indicator for swelling activity of clay materials, even though minerals which do not swell might also adsorb MB. The CEC of relatively pure montmorillonite, measured by MB adsorp tion, typically ranges from 70 to 130 mEq/100g. However, the presence of other mi nerals in bentonite can cause its CEC to fall within a lower range. The eleven GCL cl ays presented in this chapter exhibited CEC

PAGE 64

54 values between 45 and 90 mEq/100g, as measured by MB adsorption. The relationship between CEC and hydraulic conduct ivity is discussed later. Other indicators of sorptive capacity incl ude swell index, XRD descriptors, and consistency limits (LL and PL). Because of its extremely high plasticity, it is often difficult to cut a groove into a bentonite pa ste, which makes it impossible to accurately determine the LL using the ASTM Standard (Casgrande) Method. The European (fall cone) method was used instead. The eleven ma terials presented in th is chapter exhibited a wide range of plasticity beha vior (Figure 4-3), but the majo rity fell within the typical range for montmorillonite. Figure 4-3 Liquid limit vs plasticity index of GCL bentonite materials The swell index was measured in both water and leachate. The relationship between hydraulic condu ctivity and both swell index and plasticity limits is discussed later. XRD tests were also conducted to observe the sh ift in diffraction angle or d-

PAGE 65

55 spacing upon permeation with water and a hi gh concentration calcium and magnesium chloride leachate from an ash landfill. All tests were conducted on air-dried bentonite powder specimens prepared using the back-l oading method. While dry-prepared XRD specimens may reflect preferred particle or ientations, the results provide an unbiased comparison between different specimens. Furt her description and an alysis of the XRD test data is provided in Chapter Five. The results shown in Figure 4-4 indi cate no significant difference in the diffraction pattern between the as-received a nd water-permeated bentonites. In contrast, the bentonite underwent a distin ct shift and broadening of the 7 peak upon permeation with the leachate. The peak at or around 7 is the characteristic peak for montmorillonite, and the shift signifies a change in the thickness of the interparticle spacing due to the associated cation exchange process. This sh ift from 7 to 6 corre sponds to a change in dspacing of approximately 17%, and the broadening of the peak indicates a larger range of interparticle spacing, possibly due to a partial exchange. Similar behavior was observed with all other bentonite samples, with the ex ception of T4 which exhibited a shift in the 7 peak and intensity up on permeation with water.

PAGE 66

56 Figure 4-4 XRD pattern of T1 bentonite: (a) as received, (b) permeated with water, and (c) permeated with high concentration inorganic leachate 4.6 Hydraulic Conductivity vs Swell Index Relationship Swell index and LL have been proposed as indicators of the hydraulic conductivity of bentonite (e.g., Shackelford et al. 2000; Jo et al. 2001). While the present data supports the argument that the hydrauli c conductivity to wate r and leachate depends on the swell index, there is a clear distinction in behavior between treated and non-treated bentonites. This is due to the differe nce between clay and clay-polymer surface chemistries. The nature of clay-polymer-w ater interaction depends on the composition, ion size, and surface charge of the polymer. For instance, very high swell index values have been associated with high permeability values and vice vers a for polymer-treated bentonite (Schenning 2004).

PAGE 67

57 1.E-09 1.E-08 1.E-07 1.E-06 1.E-05 1.E-04 020406080100 Swell Index (mL/2g)k (cm/sec) AU AT BU BT CU CT1 CT2 Model k ( cm/s ) Swell Index (ml/2g) Figure 4-5 Relationship between hydraulic conductivity and swell index The data presented in Figure 4-5 indicat e that there is a strong dependency between the hydraulic conductivity and the swell index. To this end, non-linear regression analysis was performed on the data set using a numberof candidate statistical functions, including logarithmi c, exponential, and polynom ial equations. Polynomial equations were quickly discarded as they do not provide reasonabl e boundary conditions such as asymptotic values at very low and very high swell i ndex. Correlation coefficients were determined for the various candidat e functions, and the mathematical model presented in Equation 4.1 result ed in the highest correlation. 3 12424 2 605 iS k (4.1) where k is the hydraulic conduc tivity in cm/s, and Si is the swell index in ml/2g. The corresponding regression statisti cs are listed in Table 4-1.

PAGE 68

58 Table 4-1 Regression statistics – hydraulic conductivity vs swell Index Correlation Coefficient (R) 0.891 Adjusted R2 0.761 (adjusted for nu mber of observations) Standard Error 85.146 Observations 15 Standard Error in Intercept 36.2 Standard Error in Slope 359 4.7 Hydraulic Conductivity vs Plas ticity Ratio Relationship The plasticity ratio is a newly introduced pa rameter as part of this research, and is essentially an indicator of the relative change in soil plasticity when tested with leachate as opposed to water. To calcu late the plasticity ratio, the consistency limits (LL and PL) of the sample are first measured using water, and the soil is mapped on the plasticity chart, as shown in Figure 4-3. The relative plasticity, a measure of how far the soil is from the U-line, is then calculated by divi ding the plasticity in dex of the soil by the corresponding ordinate on the U-line. The LL and PL are then measured by mixing the soil with the leachate, instead of water, and the relative plasticity is calculated accordingly. The ratio between both relative pl asticity values (leachat e and water) is then defined as the plasticity ratio. This parameter can be view ed as the relative shift away from the U-line due to sample exposure to leachate. Figure 4-6 presents the rela tionship between the plastici ty ratio and the hydraulic conductivity ratio. A strong corre lation is evident from the da ta presented on the graph. The conductivity ratio for each bentonite type is defined here as the ratio between the hydraulic conductivity to leachate and the hydraulic conductivity to water. A non-linear regression analysis was performed on the pl asticity and hydraulic conductivity ratios

PAGE 69

59 resulting in the development of a mathematical model (Equation 4.2) that can be used as an early indicator to the permeability perf ormance of non-prehydrated bentonite upon permeation with inorganic leachate solution. This model was, again, developed by testing nine different statisti cal functions (logarithmic, e xponential, and polynomial) and selecting the one that provided th e highest correlation coefficient. 0 20 40 60 80 100 120 0.40.50.60.70.80.91.0Plasticity Ratiokleachate / kwater CU BU CT-2 CT-3 CT-1 A T CU-2 A U BT Model Figure 4-6 Relationship between hydraulic conductivity and plasticity ratio 235 0 4831 0 2232 0 r rP k (4.2) where kr is the hydraulic conductivity ratio (kleachate / kwater) and Pr is the plasticity ratio. Table 4-2 lists the regression statistics associ ated with the non-linear regression analysis. Table 4-2 Regression statistics – hydraulic conductivity ratio vs plasticity ratio Correlation Coefficient (R) 0.997 Adjusted R2 0.993 (adjusted for nu mber of observations) Standard Error 3.71 Observations 9 Standard Error in Intercept 1.5 Standard Error in Slope 0.0147

PAGE 70

60 Figure 4-7 shows a general decrease in hydraulic conduc tivity to landfill leachate as the CEC increases, regardless of polymer treatment, although some level of scatter exists at high CEC values. Figure 4-7 Relationship between hydraulic conductivity and CEC While extreme care was taken during sample preparation and testing, the reproducibility of the hydrauli c conductivity values has only been verified for a limited number of tests. The initia l results indicate that the range of variability on hydraulic conductivity is within 10 0% for flexible wall tests, and 50% for rigid wall tests. A consistent trend was, however, observed wher e flexible wall tests resulted in hydraulic conductivity values larger than rigid wall tests for specimens pr epared at the same initial void ratio. The increase was as low as 60% for specimens permeated with water and as high as 450% for high concentration single species solutions. This difference is attributed to the inability of rigid wall specimens to swell vertically and laterally during

PAGE 71

61 testing, which causes the voi d ratio to remain constant throughout. The hydraulic conductivity values reported in Figure 4-2 are for flexible wall tests while the hydraulic conductivity data reported in Figure 4-5 is for a flexible wall permeameter tests. 4.8 Diffusion and Sorptive Re tention Characteristics The diffusivity and retenti on of cations in bentonite depend on the void ratio, CEC, solution concentration, and cation type. Because of the small thickness of GCLs, diffusion plays an important role in solute tr ansport, and can cause the bentonite to reach its sorption capacity within relatively short pe riods of time. Chemical diffusion can also cause substantial degradation to the hydraulic conductivity of the liner. Diffusion tests were conducted on pure Wyoming bentonite in conjunction with 1M, 2M, and 5M solutions of NaCl, CaCl2, and MgCl2. The range of values of effective diffusion coefficient, D*, and retardation factor, Rd, is given in Table 4-3. Table 4-3 Diffusion parameters for pure Wyoming bentonite Solution CaCl2 MgCl2 NaCl Effective diffusion coefficient, D (m2/s) 1.8-5.5x10-126.6x10-12 5.2-8.0x10-12 Retardation factor, Rd 1.6-6.3 17.1 5.6-18.5 The low retardation factor values in th e case of calcium indicate a rapid cation exchange process, resulting in collapse of the double layer and increase in the space available for ions to diffuse. This finding is important in understanding the rate of cation of exchange upon permeation with calcium and magnesium leachates. The low retardation factors support the notion that chem ical equilibrium is reached within a short permeation duration, possibly within a few pore volumes. This conclusion was also verified through XRD testing as de scribed later in Chapter Five.

PAGE 72

62 4.9 Summary Results were presented from studies on the hydraulic conductivity, diffusion, and sorption characteristics of polymer-treated and untreated GCL bentonite exposed to inorganic solutions. Hydraulic and chemical equilibrium were verified throughout each of the tests to ensure the si mulation of long-term conditions Polymer-treated bentonites retained their hydraulic conduc tivity after exposure to hi gh concentration inorganic leachate, with the exception of one sample that exhibited high k values when permeated with both water and landfill leachate. Swell in dex and plasticity ratio proved to be better indicators of the performance of polymer-treated clays than CEC. This may be attributed to the fact that the surfac e chemistry of polymer-treated clays does not follow the classical Gouy-Chapman double-layer theory. Changes to the double layer chemistry in both polymer-treated and untreated bentonite can be characterized by the shift in bentonite characteristic peak in the XRD trace. A sta tistical regression model was proposed to relate hydraulic conductivity with swell i ndex. A second model was developed to correlate the conductivity ratio with a newly introduced parameter, the plasticity ratio. The second model is of immense value in evaluating the compatibility of GCL bentonite with inorganic leachates.

PAGE 73

63 Chapter 5 X-Ray Diffraction 5.1 Overview In this chapter, all ten samples of both treated and untreated bentonite materials, permeated with ordinary tap water and inorganic ashfill leachate during hydraulic conductivity tests, were exhumed at the end of the experiment and subjected to XRD testing. The results indicate that the samples permeated with leachate underwent full ionic exchange whereby the higher valance calcium cations exchanged with the lower valance sodium cations thus re sulting in a calcium montmor illonite. This chapter will present testing procedures, discuss results and provide recommendations. 5.2 Background As part of the current research, a sepa rate task was undertaken to determine whether the leachate permeation altered th e soil samples’ mineralogical makeup. Specifically, the use of XRD analysis techni que was undertaken to determine whether the samples have achieved chemical equilibrium, and whether significant alteration to the mineral structure has occurred upon permeation. The XRD process is highly reliable, albeit a destructive test in the sense that the permeability test must be terminated, the sample extracted and processed for the diffraction measurement. The inorganic leachate, obtained from the Pasco County facility, was analyzed for chemical composition at the USF Environmental Laboratory, and the resu lts are summarized in Table 5-1 below.

PAGE 74

64 Table 5-1 Leachate chemical composition Alkalinity Conductivity pH Total Hardness Ca++ 72 mg/L as CaCO3 37.6 mS/cm 6.52 20,000 mg/L as CaCO3 12,800 mg/L as CaCO3 5.3 Bentonite Soil Mineralogy The contraction and expansion of the interlayer spacing in bentonite clay directly affects the permeability of the soil material. Highly swelling bentonites tend to have lower permeability than those with lower sw elling capacity. In general, the smectites’ high swelling capacity is attributed to the weak interlayer attraction force which is insufficient to keep the interlayer spacing intact (Moore and Reynolds 1989). Ca-montmorillonite and Na-montmorilloni te are the most common minerals found in bentonite soils. Na-montmorillonite typically has a permeability that is an order of magnitude lower than that of Ca-montmor illonite. Because of its weak surface charge, Na-montmorillonite is vulnerable to atom ic exchange when exposed to solutions containing exchangeable cations having highe r valence than the sodium cations. For example, when sodium bentonite is permeated with a solution containing electrolytes in the form of calcium chloride, a cationic ex change process occurs whereby the higher valance calcium cations exchange with the lo wer valance sodium cations thus altering the mineralogical makeup of the soil and its inhe rent low permeability. The purpose of the XRD testing is to identify such changes in clay mineralogy and to establish whether or not full exchange between Na+ and Ca2+ has occurred during permeation.

PAGE 75

65 5.4 XRD Principles A basic overview of XRD theory was pres ented with the liter ature review in Chapter Two. In this Chapter, the theory and implementation are reviewed in more detail. X-rays are electromagnetic radiation of a wavelength equal to the size of an atom and are generated from the oscillation of electrostatic and electromagnetic fields perpendicular to each other and to the pl ane of propagation through space. XRD occurs when Bragg’s Equation 5.1 is satisfied. n = 2 d sin (5.1) where n is an integer, is the wavelength, d is the spacing between the lattice plane, sometimes referred to as basal spacing, and is the diffraction angle. Figure 5-1 shows a schematic of the diffraction process, where the x-rays are refracted upon colliding with the atoms in the lattice structure. d Lattice Planes Figure 5-1 XRD pattern (reproduced from Whittig 1986) In three dimensional spacing, the interato mic spacing of each mineral is distinct producing a unique diffraction a ngle for each mineral. These diffraction angles can be used to identify the minerals present in the material being diffracted. For the XRD

PAGE 76

66 process to be perfectly successful, the incide nt rays should be perfectly parallel, atoms must be perfectly ordered and the crystal should be perfectly oriented. When the incident beam waves hit the atoms and refract in phase, their amplitudes are amplified and form a refracted beam of higher amplitude and wide r peak in what is known as constructive interference. Alternatively, when the incide nt beam waves are out of phase, they form smaller amplitudes which in turn results in a narrow peak due to destructive interference (Moore and Reynolds 1989). 5.5 XRD Methods XRD can be performed using either a quali tative method, quantitative method or a combination of both. The qualitative method is used to identify the minerals that make up the sample with no regard to their quanti ties. The quantitative method is used to determine the relative concentrations of thos e minerals. For this research program, the qualitative method was used because the intent of the research is to: 1) identify the type of mineral present in the sample, 2) determine whether the sample has reached chemical equilibrium during permeation, and 3) dete rmine whether the mineralogical makeup was altered following the permeation process. 5.6 Qualitative Analysis Method The qualitative analysis is based on identif ying the minerals associated with the diffraction maxima obtained from a diffr acted sample. Identification can be accomplished wither by: 1) using the standa rd comparison method, or 2) comparing the diffraction spacing with known spacings of standa rd minerals. In the direct comparison method, the unknown mineral can be identified by comparing the resulting strongest

PAGE 77

67 diffraction peak(s) with those of known mineral standards. Successive association of the peaks with minerals serves to identify the mineralogical make up of the soil sample. For this research program, the diffraction spacing was used to identify the minerals. Each mineral has a unique d-spacing value that serves to id entify that mineral. Sodium montmorillonite, for example, has a d-spacing value of 12.4 (Moore and Reynolds 1989) while the d-spacing for calcium montmorill onite is 15 (Shang et al. 2002). Following the diffraction process, commercial software was used to identify the minerals associated with the strongest peak s, followed by validation of the identified minerals by the operator. The software is a very effective means for identifying the minerals; on the other hand, it also tends to identify all possible minerals that could be associated with the peaks, resulting in an exhaustive list of common and uncommon minerals. The operator post-identification valid ation relies heavily on the experience and judgment of the user and serves to narrow dow n the list of identified minerals to a more plausible roster based on the ge ology and source of the material. 5.7 Quantitative Analysis Method The quantitative analysis method requires deliberate and careful sample preparation, good quality data and a detail ed understanding of the material being analyzed. This method is difficult to perform on clayey soils due to the variations in their chemical composition, random particle orientation and particle sizes. The quantitative analysis can be conducted using either an internal standard method, or an external standard method. The choice of method depends largely on the type of material being analyzed. The inte rnal standard method is typically used in

PAGE 78

68 analyzing powdered soil specimens and is ty pically applied to mineral or materials for which the composition is unknown. The exte rnal standard method is used on solid materials like alloys and allows the quantification of one or more components in the material while the direct comparison method is only applicable to fully crystalline mixtures and does not require any standards. The internal standard method requires a known quantity and weight of a standard substance to be added to the unknown mixture to be analyzed. Th e substance selected must have the same adsorption characteristic s as the sample being analyzed. The mass absorption of a mixture need not be known in advance and any number of constituents in a mixture may be independently quantified. Unlike the external standard method, the mass absorption coefficient must be known in advance which may require either full elemental chemistry or prior knowledge of the chem istry, as in the case of alloys. All of the components in the mixture mu st be quantified for solution. 5.8 XRD Equipment An x-ray diffractometer, in its simplest fo rm, is composed of an x-ray source and a detector that picks up and r ecords the diffracted rays as schematically illustrated in Figure 5-2 below.

PAGE 79

69 Figure 5-2 Conceptual representation of x-ray diffractometer The Philips PW 3040/60 X’Pert Pro XRD system, located in the College of Engineering Material Testing Laboratory at the USF, was used to perform the XRD analyses. The diffraction system is compos ed of an upper console that houses the instrumentation and a lower part containi ng the power supply and the measuring and control electronics. The soil specimen is held in a stationary hori zontal position and the angle of incidence of the x-rays is varied by rotating the goniometer as illustrated in Figure 5-3 below.

PAGE 80

70 Upper Console Lower Console Rotating Arms Stationary Sample Holder Control Panel Figure 5-3 Photograph of XRD instrument To ensure complete identification of the soil minerals, the incident beam was set to bombard the specimen at a scanning rate of 2/min over an angular range between 4o and 70o degrees. The 2 values were used to calculate the d-spacing by applying Bragg’s law. 5.9 Sample Preparation There is a strong dependency between th e quality of the XRD results and the sample preparation techniques. The speci men powder mass must contain large numbers of crystals so that enough cr ystals will be oriented to ensure that every set of lattice planes will be diffracted, and the crystals must be correctly oriented with respect to the

PAGE 81

71 incident beam. Both of these conditions can be achieved with the proper specimen preparation. For quantitative analysis, the specimen part icles need to be pulverized to produce extremely fine particles in the range 1 to 5 m (Brindley and Brown 1980) which will ensure that enough particles ar e “participating” in the diffr action process. For routine qualitative evaluation the recommended partic le size is finer than a 325 mesh sieve (44 m). Achieving the recommende d particle size requires samp le grinding using either a manual or mechanical method. Excessive grindi ng should be avoided so as not to distort the lattice planes or alter th e mineralogy by excessive frictio n-induced heat. Mechanical grinding of the soil material should be used as a last resort because it could cause the disorder of the crystalline materials wh ich will reduce the diffracted intensities. Quantitative analysis relies heavily on th e intensity ratios in identifying mineral concentrations, thus requiring the particles to be randomly or iented in the sample being diffracted. Preferred particle orientation causes distortion of the intensities and should be avoided. In qualitative analys is it is also desirable to have randomly oriented particles to produce the maximum number of overlapping reflection. Randomly oriented specimen preparation can be achieved by using either vibrating table, side packi ng or back packing. Other advanced preparation methods include the coatings of minerals with various solutions like aerosol plastic spray, acetone and thermoplastic organic cement. Achievi ng a totally random orientation is extremely difficult and requires a great deal of experi ence whereas near-random orientation is more manageable.

PAGE 82

72 5.10 Particle Size and Sample Grinding Particle size distribution plays an import ant factor in achieving a meaningful XRD data. Coarse particles will produce “b lemished” diffraction data and therefore a very fine particle distribution is desirable. Fine powders will include a sufficient number of particles to provide the maximum number of diffraction when bombarded with the incident beam. Achieving 44 m particle size is very tedious and may involve a combination of wet sieving and gravity or cen trifugal sedimentation. Thus, preparing the sample by crushing of the sample using pestle and mortar is the preferred method, at least at the start of the grinding process. Th e finer particles are separated by sieving the crushed material through No. 200 mesh sieve. The sieved material can be crushed further and sieved through finer mesh sieves until the de sired particle size is attained. Achieving particles finer than No. 325 sieve would requ ire special techniques such as grinding under water. Because of equipment limitations particles passing the 200 mesh sieve (75 m) were used to perform for the XRD for this research program. 5.11 Specimen Preparation The bentonite materials were classified into the following three groups: 1) asreceived bentonite; 2) bentonite permeated wi th water; and 3) bentonite permeated with leachate. Following the permeation process, the permeated samples were allowed to air dry at room temperature and then crushed us ing pestle and mortar and sieved through the No. 200 mesh. The sieved material was plac ed inside glass sample holders, securely capped and labeled accordingly.

PAGE 83

73 There are various XRD sample preparati on techniques that include back-loading, front-loading, side-loading and spray-drying. Achieving rand omly oriented crystals is crucial to the quality of the XRD results and as such, smoothing of the specimen surface during preparation should be avoided to minimi ze preferential crystal orientation. Under this research program, the back-loading technique was used to prepare the samples. The technique entailed the placing of the samp le holder over a strip of glass followed by randomly placing the crushed and sieved soil sample in the sample holder and the excess material was removed using a straight edge The metal backing plate was placed on top of the soil and securely snapped into place on the sample holder. The sample holder was then carefully inverted and the glass strip wa s carefully removed so as not to disturb the soil material. The prepared sample was imme diately placed in the XRD machine. The sample preparation procedure is detailed in Figure 5-4. 5.12 XRD Results and Discussion The d–spacing of the diffracted samples was calculated using Bragg’s law and the 2 diffraction angle values; for the Phillips XRD instrument, the wavelength = 1.5406. The d-spacing values were graphically plotted for all samples as illustrated in Figure 5-5 through Figure 5-14.

PAGE 84

74 Specimen Holder Figure 5-4 Sample preparation technique for XRD testing Specimen Holder Mortar and Pestle Sample Bottle #200 Mesh Sieve Cationic Exchange Cationic Exchange Pulverized soil Back plate Packed bentonite

PAGE 85

75 Na+ Montmorillonite Na+ Montmorillonite Ca++Montmorillonite0 500 1000 1500 2000 2500 0 2 4 6 8 10 12 14 16 18 20 22 24 d-Spacing ()Intensity Water As-Is Leachate Figure 5-5 XRD pattern for BT (GSE new treated) Na+ Montmorillonite Na+ Montmorillonite Ca++ Montmorillonite0 500 1000 1500 2000 2500 0 2 4 6 8 10 12 14 16 18 20 22 24 d-Spacing ()Intensity Water As-Is Leachate Figure 5-6 XRD pattern for AT (GSE old treated)

PAGE 86

76 Na+ Montmorillonite Na+ Montmorillonite Ca++Montmorillonite0 500 1000 1500 2000 2500 3000 0 2 4 6 8 10 12 14 16 18 20 22 24 d-Spacing ()Intensity Water As-Is Leachate Figure 5-7 XRD pattern for BU (GSE new untreated) Na+ Montmorillonite Na+ Montmorillonite Ca++Montmorillonite0 500 1000 1500 2000 2500 0 2 4 6 8 10 12 14 16 18 20 22 24 d-Spacing ()Intensity Water As-Is Leachate Figure 5-8 XRD pattern for AU (GSE old untreated)

PAGE 87

77 Na+ Montmorillonite Na+ Montmorillonite Ca++Montmorillonite0 500 1000 1500 2000 2500 3000 3500 4000 4500 0 2 4 6 8 10 12 14 16 18 20 22 24 d-Spacing ()Intensity Water As-Is Leachate Figure 5-9 XRD pattern for CT-1 (CETCO old treated #1) Na+ Montmorillonite Na+ Montmorillonite Ca++Montmorillonite0 500 1000 1500 2000 2500 3000 3500 4000 0 2 4 6 8 10 12 14 16 18 20 22 24 d-Spacing ()Intensity Water As-Is Leachate Figure 5-10 XRD pattern for CT-2 (CETCO old treated #2)

PAGE 88

78 Na+ Montmorillonite Na+ Montmorillonite Ca++Montmorillonite0 500 1000 1500 2000 2500 0 2 4 6 8 10 12 14 16 18 20 22 24 d-Spacing ()Intensity Water As-Is Leachate Figure 5-11 XRD pattern for CT-3 (CETCO old treated #3) Na+ Montmorillonite Na+ Montmorillonite Ca++Montmorillonite0 500 1000 1500 2000 2500 3000 0 2 4 6 8 10 12 14 16 18 20 22 24 d-Spacing ()Intensity Water As-Is Leachate Figure 5-12 XRD pattern for CT-4 (CETCO old treated #4)

PAGE 89

79 Na+ Montmorillonite Na+ Montmorillonite Ca++Montmorillonite0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 0 2 4 6 8 10 12 14 16 18 20 22 24 d-Spacing ()Intensity Water As-Is Leachate Figure 5-13 XRD pattern for CT-5 (CETCO new treated #1) Na+ Montmorillonite Na+ Montmorillonite Ca++Montmorillonite0 500 1000 1500 2000 2500 3000 3500 4000 4500 0 2 4 6 8 10 12 14 16 18 20 22 24 d-Spacing ()Intensity Water As-Is Leachate Figure 5-14 XRD pattern for CU (CETCO untreated)

PAGE 90

80 5.12.1 Intensity and Peak Widths The above figures were plotted with th e diffraction intensity versus the basal spacing (d-spacing) to show the location of the peak position which is used to identify the mineral of concern. The term “intensity” as used in this qualitative XRD analysis refers to the diffraction intensity which is directly pr oportional to the instrument’s receiving slit, (detector) width. Increasing the detector’s width will result in sharper peaks and will have a direct impact on the length of the s canning time. The diffraction intensity data merely provides an insight to the kind of at oms and their position within a unit cell and is not an indication to the quantity of that mineral (Moore and Reynolds 1989). It is important to understand that na rrow peaks are cau sed by destructive interference during the XRD process while wi der peaks are indicat ive of constructive interference. The peak’s width is a concep tual indication as to the quality of the diffraction data and has no relation to the quantit y of the mineral in the diffracted sample. The relative width of the peak is also a m easure of the uniformity and purity of the mineral. 5.12.2 Data Analysis The data reveals that the tested samp les, with the exception of CETCO Old Treated #3 and #4, contained Na-montmorillonite as the main mineral when diffracted in the as-received and water-permeated condi tions. Upon leachate permeation followed by diffraction analysis, the data illustrate a distinct shift in the basal spacing (d-spacing) of the strongest peak from 12.4 (Na-montmor illonite) to 15 (Ca-montmorillonite). The CETCO Old Treated #3 sample appeared to contain Ca-montmorillonite under all

PAGE 91

81 diffracted conditions which are an indicati on that the bentonite source used in the fabrication process contained an abundance of calcium cations. The CETCO Old Treated #4 appeared to contain Na-Montmorilloni te in the as-receive d condition and CaMontmorillonite when permeated with water and leachate. Since water is chemically neutral and contains no exchangeable cati ons, it is not possible for the sample’s mineralogical makeup to have significantly changed upon water perm eation. The reason for such a disparity remains inconclusive, alt hough it is possible that the clay contained a calcium salt or hydroxide that was dissolved upon permeation with water, causing the calcium ions to replace the sodium in the clay double layer. The samples that achieved a basal spaci ng shift from the 12.4 to the 15 imply that: 1) these samples have achieved chemical equilibrium during the permeation process, and 2) the final calculated perm eability value is accurate and will remain constant for the life of the liner system because the exchangeable cations were completely exhausted. Figure 5-15 provides a comparison of the samples’ permeability values following permeation with water and leachate and prior to the XRD process.

PAGE 92

82 1.00E-10 1.00E-09 1.00E-08 1.00E-07 1.00E-06 1.00E-05 1.00E-04 1.00E-03 1.00E-02 1.00E-01 1.00E+00CETCO New Treated #1 CETCO Old Treated #2 CETCO Old Treated #1 GSE New Treated GSE Old Treated CETCO Old Treated #3 GSE New Untreated CETCO Untreated GSE Old Untreated CETCO Old Treated #4ManufacturerPermeability (cm/s) Water Leachate Figure 5-15 GCL permeabilities The permeability data reveals that sa mples achieved low permeability values upon water permeation. Alternatively, the permeabilities increased upon permeation with leachate. The magnitude of permeability increase varies from one sample to another. For example, the permeability of the CETCO Untreat ed sample increased by as much as two orders of magnitude. Furthe r, it appears that the permeabilities of the treated samples remained relatively low after extended leachate permeation, except for CETCO Old Treated #4. This conclusion supports the manuf acturers claim that the polymer treatment will enable the bentonite to maintain its low permeability upon permeation with high concentration of inorganic liquids. Th e CETCO Old Treated #4 sample was the exception and achieved the highest permeability values upon leachate permeation. This could be attributed to its mineralogical make up which indicated that the sample contained

PAGE 93

83 a calcium compound to begin with. Similar observations hold true for the CETCO Old Treated #3 sample, which also contained Ca -Montmorillonite and achieved the second highest permeability values of all treated samples. 5.13 Summary The XRD has proven to be an accurate and re liable test that can be performed to determine the mineralogical changes to cl ay specimens upon permeation with inorganic leachate. On the other hand, XRD is a time-consuming analysis that requires a good deal of experience and familiarity with sample preparation techniques, instrument operating procedure and results interpretation. Sample preparation is the most crucial step in the analysis and is the main source of errors. Fo r example if the sample preparation results in a preferred particle orientation, the results will be erroneous and will not provide for a true determination of the sample’s mineralogical makeup. The permeabilities results indicate that the amount of pore volume that was introduced into the samples during the perm eation process was adequate to ensure chemical equilibrium. Simple calculations can demonstrate that such large pore volumes will take a number of years to be achieved, if ever, under actual field conditions. As such, it is the conclusion of this author that the treated samples will achieve low permeabilities upon exposure to highly concentrat ed inorganic liquids. It also appears that the treatment process achieved its desirable e ffect. However, the untreated samples also performed relatively well under the same cond itions. Therefore, th e decision whether or not to use a treated GCL for a nominal adva ntage in performance over an untreated GCL material rests solely on the design professional.

PAGE 94

84 Chapter 6 Conclusion and Recommendations 6.1 Summary Alterations to the clay’s mineralogi cal makeup upon permeation with inorganic solution containing elevated concentration of electrolyte increase its permeability. The overall intent of this research was to valid ate whether the polymer-t reated bentonite can maintain its low permeability upon permeation with aggressive inorganic solution, to compare its performance with that of the unt reated bentonite, and to develop a rational procedure for evaluating the chemical comp atibility of bentonite with inorganic chemicals based on index testing. To that end, permeability measurements were conducted using tap water and leachate genera ted from ashfill disposal facilities. Permeation with tap water was performed to establish a permeability base value that was used as a baseline to further evaluate th e permeability performa nce of the leachate permeated soils. The ultimate objective of this research was to explore whether an empirical model can be developed and used as an early predictor of the bentonite’s permeability performance prior to performing extended permeab ility duration tests. To that end, the index properties (swell inde x, plasticity index and MBA) of the various bentonite materials were measured and the resulting da ta was analyzed for meaningful correlation with the permeability. Analysis of the testing results concluded in the development of

PAGE 95

85 two empirical models that can be used to predict the permeabili ty of the bentonite material upon permeation with inorganic l eachate prior to performing extended duration tests. These models use the swell index and the soil properties to predict either the permeability or the increase in permeability due to the presence of inorganic substances in the leachate. The developments of thes e models are addressed in Chapter Four and their practical applica tions are discussed in Section 6.4 below. 6.2 Conclusions The results clearly indicate the existe nce of a strong correlation between the plasticity ratio and the permeability ratio, whic h can be used to establish the extent to which a particular bentonite is able to maintain its low hydraulic conductivity upon exposure to inorganic leachate. A mathem atical relationship wa s also established between the swell index and the permeability va lue. Furthermore, the results indicate that the polymer treatment of the bentonite en ables it, to a limited extent, to maintain its low permeability upon exposure to highly concentr ated electrolyte solution. This finding should be qualified by further sta ting that the results are specif ic to the leachate chemical composition and the GCL liners used in this study. Actual results may vary when testing with leachate having di fferent concentrations. The results indicate that the water-perm eated samples, both regular and polymertreated, achieved comparable permeability values that remained constant for the permeation duration. This is a clear indicati on that the polymer treatment, in the case of water permeation, did not affect the permeability of the bentonite either adversely or favorably. However, when leachate was used as the permeant, the permeability of both

PAGE 96

86 types of bentonite increased, albeit, by vary ing degrees. Overall, with the exception of one material (CETCO Old Treated #4), the polymer-treated bentoni te samples achieved lower final permeability values than those of the untreated type. It was also concluded that Ca-montmorillonite is a more stable product than Na-montmorillonite in that its permeability is not adversely affected when permeated with leachate containing elevated concentrations of sodium, ma gnesium, and calcium cations. 6.3 Recommendations for Future Work The findings of this research provided valu able information that can be used to further our understanding of the long-term permeability performance of the GCL materials. However, there remains a need to perform additional research on these materials in addition to further validate and generalize the findings established in this study. The scientific and the engineering communities can greatly benefit from the following research topics in the future: 1. There is a need to evaluate the longterm effect of concentrated organic compounds on regular and polymer-treated bentonite. This can be achieved by collecting leachate generated from landfills known to have high concentrations of organics that are ty pically detrimental to the permeability performance of the soil component (ben tonite) of the GCL. The testing methods and analyses can pa rallel those followed in this study which include extended permeation with no prehydrati on of the soils with the organic leachate followed by XRD of the permeat ed sample, index tests (swell index, plasticity index, and MBA). The goa ls of the study should be to verify

PAGE 97

87 whether or not the statistical models pr esented in this study are applicable in the case of permeation with organic l eachate – as opposed to inorganic – and to evaluate whether there is a benefit in using the polymer-treated bentonite in those applications. 2. It is imperative to evaluate the use of Ca-montmorillonite as an alternate to Na-montmorillonite in applications c ontaining elevated concentrations of electrolytes. The testing program can, ag ain, parallel the steps implemented in this research, with the ultimate goal of defining the resistance of Camontmorillonite to ashfill leachate. 3. Additional swell index and permeability te sts need to be carried out using inorganic leachate with varying inorgani c concentrations to further evaluate the statistics of the mathematical mo dels established in this study for the relationship between the swell index a nd permeability and for the relationship between the plasticity ratio and permeab ility ratio. This can be accomplished by testing bentonite against inorganic leachate from various ash monofills to cover a wide spectrum of electrolyt e concentrations and establish the appropriate levels of confidence and e rror of the current empirical models. 4. The geotextile fabric used in the GCL permeated with inorganic and organic leachates have not been evaluated to determine whether these fabrics will experience any degree of degradation upon extended exposure to these contaminants. While the vast majority of existing data suggests that geotextiles can withstand inorganic ch emicals, the resistance to specific organic compounds has not been adequa tely established. This can be

PAGE 98

88 accomplished by directly submerging thes e fabrics in leachate for extended duration while observing changes, if any, to their physical composition. 5. It would be of great prac tical value to simulate the performance of the GCL under actual field conditions using a s caled landfill cell. This can be accomplished by using plexiglas panels to form a rectangular box equipped with leachate detection system overlai n by GCL topped with a layer of bottom ash generated from the incineration of MSW. The box should remain uncovered and placed in an open environm ent to allow for the percolation of rainwater into the material. To further simulate actual field conditions, additional layers of ash could be added at pre-determined intervals. The rate of leachate flow from the detection system can be observed over time and the variation of the flow can be used to evaluate the performance of the GCL. The chemical composition of the leachate at the start of the testing should be determined and used as a basis to determine when chemical equilibrium is achieved. A representative sample of the GCL can then be obtained to measure the permeability value in the lab and to examine the changes in mineralogy using XRD. 6. Throughout this research, the need becam e evident to devise a practical and cost-effective method to identify when chemical equilibrium is achieved during the permeation process. One po ssibility is to add and mix a known quantity of inert dye material to the be ntonite prior to the permeation process and observe the change in the effluent color. A calibration process could be undertaken to determine the rate of leach ability of the dye under actual testing

PAGE 99

89 conditions and its relationship with chem ical influent-effluent equilibrium. This determination can be correlated to the number of pore volumes introduced into the permeating sample. 6.4 Engineering Implications The short-term duration index tests used in this study along with the mathematical models established are valuable tools that can be used by the desi gn professional at an early stage in the design process to determin e whether further testing and evaluation of the material under design consideration is warr anted. These tools will greatly benefit the engineering community by saving valuable time and by reducing the costs associated with lab testing. It should be noted that these short-durati on tests are not intended to replace the permeability test and the design professiona l should perform the extended duration permeability tests after the initia l screening of the material through the index testing. For example, in extremely environmentally se nsitive engineering projects, the design professional may want to use XR D analysis to have a high le vel of confidence as to the composition and mineralogical makeup of the material. Extended duration permeability tests under leachate-specific conditions may n eed to be carried out if the long-term compatibility of the GCL bentonite is in question. One of the key contributions of the cu rrent research is the development of methods for early determinati on and/or quality control of GCL bentonite compatibility. To this end, the design professional should obt ain a sample of the GCL material from the manufacturer and associated ce rtification. Upon receipt of the sample, the bentonite components of the GCL should be separated from the geotextile fabric and securely

PAGE 100

90 capped in laboratory-approved ai r tight containers. A certified geotechnical lab can run both the swell index and plasti city index testing on the be ntonite using both water and actual landfill leachate. If the design is for a new f acility, a leachate sample can be obtained from existing landfills with similar waste characteristics. When the results are available, the design professional can use th e mathematical models presented in this dissertation to decide whether further testing of the material is warranted. For the swell index results, the following relationship can be used to predict the permeability performance of the GCL upon permeation with the intended leachate: 3 12424 2 605 iS k (6.1) where k is the hydraulic conduc tivity in cm/s, and Si is the swell index in ml/2g. Based on the mathematical model and the results obtained in this study, bentonite material having a swell index of 25 ml/2g or higher shou ld have a final permeability in the range of 10-8 cm/s and lower, which is acceptable by USEPA and all state-EPA standards. The design professional can also plot the results of the plasticity index tests for both water and leachate on a standard Unified Soils Classi fication chart with the U-line and A-line clearly depicted. The material’s plasticity ratio (leachate/water) can then be calculated by measuring the relative distance of the plotted data points to the U-line for both water and leachate conditions and taking the ratio between those values, as described in Chapter Four.

PAGE 101

91 0 50 100 150 200 050100150200Liquid LimitPlasticity Inde x U-line wate r leachate Figure 6-1 Example of plasticity ratio calculation For example, consider the data presented in Figure 6-1 for a bentonite material tested with water and leachate. The liquid limit and plasticit y index of the material, as determined using water and leachate, are ( 130, 90) and (110, 50), respectively. To calculate the plasticity ratio of the material shown in Figure 6-1, the following steps are implemented: Calculate the ordinate to the U-line co rresponding to the “wat er” liquid limit of 130. The equation of the U-Line is given by Equation 6.1: PIu = 0.9 (LL-8) (6.1) Where PIu is the plasticity index (y-coordinate ) of the U-line corresponding to a liquid limit LL. Thus for a liquid limit of 130, the Plasticity Index coordi nate for the U-line equals to 109.8. The relative distance to ULine (relative plasticity) is then calculated as the ratio (90/109.8) which is equal to 0.820. The same procedure is repeated using the “leachate” liquid limit and plasticity index va lues. The coordinates corresponding to the

PAGE 102

92 material tested with leachate from Figure 6-1 above are (110, 50). From Equation 6.1, for a liquid limit of 110, the Pl asticity Index coordinate is equal to 91.8. The relative distance of the material to the U-Line (rela tive plasticity) is calculated as (50/91.8), which is equal to 0.545. The plasticity ra tio is then calculated by dividing the relative plasticity for the leachate by the relative plastic ity for the water. For this material, the plasticity ratio is calculated at 0.545/0.820=0.66 It has been established through the mode ls described by Equation 6.2, which was presented earlier in Chapter Four, that the higher the material’s permeability ratio, the smaller the permeability ratio: 235 0 4831 0 2232 0 r rP k (6.2) where kr is the hydraulic conductivity ratio (kleachate / kwater) and Pr is the plasticity ratio. A material with a plasticity ratio of 1.0 is expected to undergo minimal changes upon permeation with the leachate, whereas a material with a plasticity ratio close to 0.4 is expected to exhibit a drastic increase in permeability when permeated with leachate, compared to water. Based on the mathema tical model presented by Equation 6.2, the plasticity ratio for the example above (0.66) results in a permeability ratio of 5.25. In general, a permeability ratio of 10 should be deemed acceptable since it signifies a potential increase of one order of magnitude upon permeation with leachate compared to water, provided that the re quirement for maximum acceptable permeability as specified by the regulatory agency is satisfied. In most cases, the permeability of GCL bentonite as provided by the GCL manufacturers is smaller than 10-9 cm/s. As such, a

PAGE 103

93 permeability ratio of 10, corresponding to a plas ticity ratio of approximately 0.6, will result in a hydraulic conductivity of 10-8 cm/s, which is acceptable by all federal and state EPA standards. Therefore, as a first indicato r of quality of GCL, it is recommended that a plasticity ratio of 0.6 or higher be achieved.

PAGE 104

94 References Ashmawy, A.K., El-Hajji, D., Sotelo, N., and Muhammad, N. (2002), Hydraulic performance of untreated and polymer-tre ated bentonite in inorganic landfill leachates, Clays and Clay Minerals, 50 (5), pp. 546-552. Ashmawy, A.K., Muhammad, N., and El-Hajji, D. (2005), Advection, Diffusion, and Sorption Characteristics of Inor ganic Chemicals in GCL Bentonite, ASCE Geotechnical Special Publication No. 142, Waste Containment and Remediation, E.M. Rathje, Editor, pp. 3249-3258. ASTM C0423-02A (2005), Test Method for Sound Absorption and Sound Absorption Coefficients by the Reverberation Room Method, ASTM Annual Book of Standards 2005, Vol. 04.06, American So ciety for Testing and Materials, West Conshohocken, PA. ASTM C0837-99R03 (2005), Test Method for Methylene Blue Index of Clay, ASTM Annual Book of Standards 2005, Vol. 15.02, American Society for Testing and Materials, West Conshohocken, PA. ASTM D2330-02 (2005), Test Method for Methylen e Blue Active Substances, ASTM Annual Book of Standards 2005, Vol. 11.02, American Society for Testing and Materials, West Conshohocken, PA. ASTM D4318-00 (2005), Test Methods for Liquid Limit, Plastic Limit, and Plasticity Index of Soils, ASTM Annual Book of Standards 2005, Vol. 04.08, American Society for Testing and Materi als, West Conshohocken, PA. ASTM D5084-03 (2005), Test Methods for Measurement of Hydraulic Conductivity of Saturated Porous Materials Us ing a Flexible Wall Permeameter, ASTM Annual Book of Standards 2005, Vol. 15.02, Ameri can Society for Testing and Materials, West Conshohocken, PA. ASTM D5887-04 (2005), Test Method for Measurement of Index Flux Through Saturated Geosynthetic Clay Liner Specimens Using a Flexible Wall Permeameter, ASTM Annual Book of Standards 2005, Vol. 04.13, American Society for Testing and Materials, West Conshohocken, PA.

PAGE 105

95 ASTM D6766-02 (2005), Test Method for Evaluation of Hydraulic Properties of Geosynthetic Clay Liners Permeated with Potentially Incompatible Liquids, ASTM Annual Book of Standards 2005, Vol. 04.13, American Society for Testing and Materials, West Conshohocken, PA. Bart, J.C., Cariati, F., Erre, L., Gessa C., Micera, G., and Piu, P. (1979), Formation of Polymeric Species in the Interlayer of Bentonite, Clays and Clay Minerals, 27 (6), pp. 429-432. Brindley, G. W. and Brown G. (1980) Quantitative X-ray Mineral Analysis of Clays, Mineralogical Society, pp. 411-438 London, UK. British Standards Institution (1975), Methods of the Test for Soils for Civil Engineering Purposes, BS 1377, London. Brown, K. W. and D.C. Anderson (1983), Effect of Organics on the Permeability of Clay Soils, (EPA 600/2-83-06) United States Envi ronmental Protection Agency, p. 153. Bueno, Benedito de Souza (2002), Laboratory studies for the development of a GCL, Zanzinger, Koerner & Ga rtung (eds) 2002, pp. 365-370. Burnett, A.D. (1995), A Quantitative X-Ray Diffraction Technique for Analyzing Sedimentary Rocks and Soils, J. of Testing and Evaluation, JTEVA, 23 (2), pp. 111-118. Daniel, D. E., Bowders, J. and Gilbert, R. B. (1997), Laboratory Hydraulic Conductivity Testing of GCLs in Fl exible-Wall Permeameters, Symposium on Testing and Acceptance Criteria for GCLs, ASTM ST P-1308, dir. L. Well, pp. 208-226. Das B.M. (1990), Principles of Geotechnical Engineering, 2nd Edition, PWS-Kent., pp.10-73. Egloffstein, T.A. (1995), Properties and Test Methods to Assess Bentonites Used in Geosynthetic Clay Liners, Proc. Geosynthetic Clay Liners, A.A. Balkema, Rotterdam/Brookfield, Nurnberg, pp. 51-72. Egloffstein, T.A. (2002), Bentonite as sealing material in geosynthetic clay liners – Influence of the electrolytic concen tration, the ion exchange and ion exchange with simultaneous partial desiccation on permeability, Clay Geosynthetic Barriers, Zanzinger, Koerner & Gartung (eds). El-Hajji, D., Ashmawy, A., Darlington, J., and Sotelo, N. (2001), Effects of Inorganic Leachate on Polymer Treated GCL Material, Proceedings, Geosynthetics 2001, Portland, Oregon, pp. 663-670. Goldman, L. J., Greenfield, L. I., Damle, A.S. Kingsbum, G.L., Northein, CM.M. and Truesdale, R.S. (1998).

PAGE 106

96 Grim, R. (1968), Clay Mineralogy, 2nd Edition, McGraw-Hill. Hillier, S. (2000), Accurate Quantitative Analysis of Clay and Other Minerals in Sandstones by XRD: Comparison of a Rietveld and a Re ference Intensity Ratio (RIR) Method and the Importance of Sample Preparation, Clays and Clay Minerals, 35, pp. 291-302. Hwang, J.Y. and Dixon, J.B. (2000), Flocculation behavior and properties of Namontmorillonite treated with four organic polymers, Clay Sci, 11 (2), pp. 137146. Jo, H.Y., Katsumi, T., Benson, C.H., and Edil, T.B. (2001), Hydraulic Conductivity and Swelling of Non-Prehydrated GCLs Perm eated with Single Species Salt Solutions, J. of Geotechnical and Geo-environmen tal Eng., Vol. 127, No. 7, pp. 557-567. Kajita, L.S. (1997), An Improved Contaminant Resistan t Clay for Environmental Clay Liner Applications, Clays and Clay Minerals 45 (5), pp. 609-617. Kaufhold, S., Dohrmann, R., Ufer, K., and Meyer, F.M. (2002), Comparison of methods for the quantification of mont morillonite in bentonites, Applied Clay Sci, 22 (3), pp. 145-151. Lee, J.M., Shackelford, C.D., Benson, C.H., Jo, H.Y., and Edil, T.B. (2005), Correlating Index Properties and Hydraulic Conductiv ity of Geosynthe tic Clay Liners, J. of Geotechnical and Geo-environmental Eng., Vol. 131, No. 11, pp. 1319-1329. Lin, L.C., Benson, D.H. (2000), Effect of Wet-dry Cycling on Swelling and Hydraulic Conductivity of GCLs, J. of Geotechnical and Geoenvironmental Eng., Vol. 26, No. 1., pp. 40-49. McKelvey, J.A. (1996), Geosynthetic Clay Liners in Alkaline Environments, Symposium on Testing and Acceptance Criteria for Geosynthetic Clay Liners, January 1996. pp. 139-147. Mitchell, J.K. (1993), Fundamentals of Soil Behavior, 2nd Edition, John Wiley and Sons, New York, pp. 402. Moore, D.M., and Reynolds, R.C., Jr. (1989), X-Ray Diffraction and the Identification and Analysis of Clay Minerals, 2nd Edition, Oxford University Press, p. 378. Neter, J., Wasserman, W., and Kutner, M.H. (1990), Applied Linear Statistical Models, 3rd Edition, Irwin, p. 1181. Petrov, R. J., Rowe, R. K. and Quigley, R. M. (1997), Selected Factors Influencing GCL Hydraulic Conductivity, J. of Geotechnical and Ge o-environmental Eng., Vol. 123, No. 8, pp. 638-695.

PAGE 107

97 Pezerat, H. and Vallet, M. (1973), Formation de Polymre Inser dans les Couches Interlamellaires de Phyllites Gonflantes, Proceedings of the Fourth International Clay Conference, Madrid, Spain, pp. 683-691. Poppe, L.J., Paskevich, V.F., Hathaway, J.C., and Blackwood, D.S. (2001), A Laboratory Manual for X-Ray Powder Diffraction, USGS Open-file Report 01-041, U.S. Geological Survey. Pusch, R. (1998), Transport of Radionuclides in Smectite Clay, Environmental Interactions of Clays – Clays and the Environment (A. Parker and J.E. Rae, editors), Springer, pp. 7-35. Rad, J. L., Jacobson, B. D. and Bachus, R. C. (1994), Compatibility of Geosynthetic Clay Liners with Organic and Inorganic Permeants, 5th Intl. Conf. Geotextiles, Geom. & Related Products, IGS, Singapore, pp. 1165-1168. Ruhl J.L. (1994), Effects of Leachates on the Hydraulic Conductivity of GCLs, M.S. Thesis, Texas Univ., Austin, p. 226. Ruhl, J. L. and Daniel, D.E. (1997), Geosynthetic Clay Liner Permeated with Chemical Solutions and Leachates, J. of Geotechnical and Ge o-environmental Eng., Vol. 123, No. 4, pp. 369-381. Santamarina, J.C., Klein, K.A., Wa ng, Y.H., and Prencke, E. (2002), Specific surface: determination and relevance, Canadian Geotech J, 39 (1), pp. 233-241. Schenning, J.A. (2004), Hydraulic Performance of Polymer Modified Bentonite, MSCE thesis, University of South Florida. 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, Elsevier, Vol.18, Nos.2-3, pp. 133-161. Shan H.Y., Daniel D.E. (1991), Results of Laboratory Test s on a Geotextile/Bentonite Liner Material, Proc. Geosynthetics’91, NAGS, pp. 517-535. Shang, C., Rice, J.A., and Lin, J.S. (2002), Small-Angle X-Ray Scat tering Study of the Quasi-Crystal Structure of Mont morillonite-CTAB in Suspension, Soil Science Society of America Journal 66(4), pp. 1225-1230. Sivapullaiah, P.V., Savitha, S. (1999), Index properties of illit e-bentonite mixtures in electrolyte solutions, ASTM Geotechnical Testing Journal, Vol. 22, No. 3, pp. 257-266. Soule, N.M. and Burns, S.E. (2001), Effects of organic cation structure on behavior of organobentonites, J. of Geotechnical and Geo-en vironmental Eng, Vol. 127, No. 4, pp. 363-370.

PAGE 108

98 Sykes J.F., Soyupak S., Farquhar, G.J. ( 1982), Modelling of leachate organic migration and attenuation in groundwater below sanitary landfills, Water Resources Research, 18(1), pp. 135-145. Taylor R.K. (1984), Cation Exchange in Clays and Mudrocks by Methylene Blue, Paper presented at the Meeting of the Board and Building Materials Group Society of Chemical Industry held in London on February 1984, pp. 195-207. Theng, B.K.G. (1979), Formation and Properties of Clay-Polymer Complexes, Elsevier, p. 362. van Olphen, H. (1977), An introduction to clay colloid chem istry: For clay technologists, geologists, and soil scientists, 2nd Edition, John Wiley and Sons, New York. Wentink, G.R. and Etzel, J.E. (1972), Removal of metal ions by soil, J. of Water Pollution Control Fed, 44 (8), pp. 1561-1574. Whittig, L.D. and Allardice, W.R. (1986), X-ray diffraction techniques. In A. Klute (ed.) Methods of soil analysis. Part 1. Ph ysical and mineralogical methods. 2nd Edition. Agronomy 9, pp. 331-362.

PAGE 109

About the Author Darwish El-Hajji received his Bachelor’s Degree in Civil Engineering from the University of South Alabama in 1983 and a M. S. in Environmental Engineering from the University of South Florida in 1997. He has been working as a consulting engineer in the private sector with concentration on envi ronmental projects dealing with municipal landfills. While in the Ph.D. program at the University of South Florida, Mr. El-Hajji continued to work as a full-time consulti ng engineer and has coauthored several publications addressing landfill designs, geosynt hetic clay liners uses and testing that were published in journals and conference proc eedings. He has also presented papers at various conferences and professional societies.