EPA/530-SW-86-007-F
DESIGN, CONSTRUCTION, AND EVALUATION OF CLAY LINERS
FOR WASTE MANAGEMENT FACILITIES
by
L.J. Goldman and L.I. Greenfield
NUS Corporation
Gaithersburg, Maryland 20878
A.S. Damle, G.L. Kingsbury
C.M. Northeim, and R.S. Truesdale
Research Triangle Institute
Research Triangle Park, North Carolina 27709
EPA Contract No. 68-01-7310
Project Officer
M.H. Roulier
Waste Minimization, Destruction
and Disposal Research Division
Risk Reduction Engineering Laboratory
Cincinnati, Ohio 45268
OFFICE OF SOLID WASTE AND
EMERGENCY RESPONSE
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, DC 20460
RISK REDUCTION ENGINEERING LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
DISCLAIMER
The Information in this document has been funded wholly by the United
States Environmental Protection Agency under Contract 68-03-3149 to Research
Triangle Institute, Research Triangle Park, North Carolina and Contract
68-01-7310 to NUS Corporation, Gaithersburg, Maryland. It has been subject
to the Agency's peer and administrative review, and it has been approved for
publication as an EPA document. Mention of trade names or commercial
products does not constitute endorsement or recommendation for use.
it
-------
FOREWORD
Today's rapidly developing and changing technologies and industrial
products and practices frequently carry with them the increased generation
of materials that, if improperly dealt with, can threaten both public
health and the environment. The U.S. Environmental Protection Agency is
charged by Congress with protecting the Nation's land, air, and water
systems. Under a mandate of national environmental laws, the agency strives
to formulate and implement actions leading to a compatible balance between
human activities and the ability of natural systems to support and nurture
life. These laws direct the EPA to perform research to define our environ-
mental problems, measure the impacts, and search for solutions.
The Risk Reduction Engineering Laboratory is responsible for planning,
implementation, and management of research, development, and demonstration
programs to provide an authoritative, defensible engineering basis in
support of the policies, programs, and regulations of the EPA with respect
to drinking water, wastewater, pesticides, toxic substances, solid and
hazardous wastes, and Superfund-related activities. This publication is
one of the products of that research and provides a vital communication
link between the researchers and the user community.
The Office of Solid Waste is responsible for issuing regulations and
guidelines on the proper treatment, storage, and disposal of hazardous
wastes to protect human health and the environment from the potential harm
associated with improper management of these wastes. These regulations are
supplemented by guidance manuals, technical guidelines, and technical
resource documents, made available to assist the regulated community and
facility designers in understanding the scope of the regulatory program.
Publications like this one provide facility designers with state-of-the-art
information on design and performance evaluation techniques.
This technical resource document is a compilation of all of the available
information relevant to the design, construction, and performance of clay-
lined waste management facilities. The broad topics covered are: clays;
geotechnical testing of soils; the compatibility of clays and chemical
wastes; the design, construction, and construction* quality assurance of
.clay liners; potential failure mechanisms; the performance of existing
facilities; and methods for predicting the useful life (transit time) based
on the modeling of leachate flow through soils.
E. Timothy Oppelt, Acting Director
Risk Reduction Engineering Laboratory
-------
PREFACE .
Subtitle C of the Resource Conservation and Recovery Act (RCRA) requires
the U.S. Environmental Protection Agency (EPA) to establish a Federal hazard-
ous waste management program. This program must ensure that hazardous wastes
are handled safely from generation until final disposition. EPA issued a
series of hazardous waste regulations under Subtitle C of RCRA that are
published in 40 Code of Federal Regulations (CFR) 260 through 265 and 122
through 124.
Parts 264 and 265 of 40 CFR contain standards applicable to owners and
operators of all facilities that treat, store, or dispose of hazardous
wastes. Wastes are identified or listed as hazardous under 40 CFR Part 261.
Part 264 standards are implemented through permits issued by authorized
States or EPA according to 40 CFR Part 122 and Part 124 regulations. Land
treatment, storage, and disposal (LTSD) regulations in 40 CFR Part 264 issued
on July 26, 1982, establish performance standards for hazardous waste land-
fills, surface impoundments, land treatment units, and waste piles.
EPA is developing three types of documents for preparers and reviewers
of permit applications for hazardous waste LTSD facilities. These types
include RCRA Technical Guidance Documents, Permit Guidance Manuals, and
Technical Resource Documents (TRD's). \
The RCRA Technical Guidance Documents present design and operating
specifications or design evaluation techniques that generally comply with or
demonstrate compliance with the Design and Operating Requirements and the
Closure and Post-Closure Requirements of Part 264.
The Permit Guidance Manuals are being developed.to describe the permit
application information the Agency seeks and to provide guidance to appli-
cants and permit writers in addressing information requirements. These
manuals will include a discussion of each step in the permitting process and
a description of each set of specifications that must be considered for
inclusion in the permit.
This document is a Technical Resource Document. It was prepared by the
Hazardous Waste Engineering Research Laboratory of the Office of Research and
Development at the request of and in cooperation with the Office of Solid
Waste and Emergency Response. The TRD was first issued as a draft for public
comment under the title, "Design, Construction, and Evaluation of Clay Liners
for Waste Management Facilities" (EPA/530-SW-86-007) dated March 1986. The
draft TRD was also made available through the National Technical Information
Service (Order No. PB86-184496/AS). All comments received on the draft TRD
have been carefully considered and, if appropriate, changes were made in this
final document to address the public's concerns. With issuance of this docu-
ment, all previous drafts of the TRD are obsolete and should be discarded.
IV
-------
ABSTRACT
.Z- ^ '•' ' - ' -.&• -•"&
This Technical Resource Document (TRD) is a compilation of an of the
available information on the design, construction, and evaluation of clay
liners for waste landfills, surface impoundments, and wastepiles. The
information was obtained from the literature and from in-depth interviews
with design and construction engineers and other knowledgeable individuals in
both the private and government sectors. As a consequence, some information
is presented for the first time in this document. The broad topics covered
are: clays, with emphasis on their composition, fabric, and hydraulic con-
ductivity; geotechnical test methods and soil properties including index
properties, soil classification, and hydraulic conductivity testing; clay
chemical compatibility, including a discussion of the mechanisms of interac-
tion and a comprehensive compilation of existing test data from the litera-
ture and private sources; construction and quality assurance; clay liner
failure mechanisms; the performance of existing clay liners based on case
studies of 17 sites; and clay liner transit time prediction methods featuring
an in-depth discussion of many available techniques and models.
This TRD was submitted in September 1987 by NUS Corporation in fulfill-
ment of Contract No. 68-01-7310 with the U.S. Environmental Protection
Agency. The TRD has been revised to address issues that were raised during
the public comment period on the draft TRD (EPA/530-SW-86-007); the revised
TRD also includes technical information that became available after the draft
TRD was completed by the Research Triangle Institute 1n August 1985.
-------
TABLE OF CONTENTS
Chapter Page
Foreword iii
Preface 1 v
Abstract v
Figures „.. Xiv
Tables ..„.. Xix
Acknowledgments „. J xxi i
1 Introduction „.. i-i
1.1 Scope „.. i_3
1.2 Summary of Current Practices 1-3
1.2.1 Investigation of Site Conditions
(Field) 1-3
1.2.2 Material Selection and Characterization
(Laboratory) 1-4
1.2.3 Develop Liner Design/Construction
Plans 1-5
1.2.4 Pilot Construction Test (Test Fill) 1-5
1.2.5 Construction 1-5
1.3 Analysis of Current Practices 1-7
1.3.1 Liner Material 1-7
1.3.2 Clod Size 1-7
1.3.3 Water Content i-a
1.3.4 Pilot Construction (Test Fill) 1-8
1.4 Summary i_g
1.5 Reference i_g
2 Clay Soil ....! 2-1
2.1 Clay Minerals ....' 2-2
2.1.1 Clay Mineral Structure 2-2
2.1.2 Clay Mineral Groups 2-4
2.1.2.1 Kaolinite Minerals 2-4
. 2.1.2.2 Illite Minerals *..... 2-9
2.1.2.3 Chlorite Minerals 2-10
2.1.2.4 Smectite Minerals ..-. 2-13
2.2 Clay Formation and Occurrence 2-14
2.2.1 Clay Mineral Paragenesis 2-14
2.2.2 Clay Soil Formation and Occurrence 2-15
2.2.2.1 Fluvial Soils 2-16
2.2.2.2 Glacial Soils 2-17
2.2.2.3 Residual Soils ......... 2-17
2.3 Clay Chemistry 2-18
2.3.1 Electrical Double-Layer Theory 2-18
2.3.2 Cation-Exchange Capacity and Cation
Affinity 2-20
2.3.3 Significance of the Electrical Double
Layer to Clay Liners 2-22
VI
-------
Chapter
TABLE OF CONTENTS (continued)
Page
2.4 Clay Soil Fabric and Hydraulic Conductivity 2-24
2.4.1 Soil Porosity and Hydraulic
Conductivity 2-24
2.4.1.1 Soil Microstructure and Primary
Porosity 2-24
2.4.1.2 Soil Macrostructure and Secondary
Porosity 2-27
2.4.2 Sail Structure and Hydraulic
Conductivity in Compacted Soils 2-32
2.5 References 2-38
Test Methods and Soil Properties 3-1
3.1 Introduction 3_i
3.2 Fundamental Relationships '/.'.'. 3-2
3.2.1 Water Content 3_2
3.2.2 Density 3_2
3.2.3 Specific Gravity 3_2
3.2.4 Unit Weight 3.5
3.2.5 Void Ratio 3.5
3.2.6 Porosity 3.5
3.2.7 Degree of Saturation ! 3-6
3.3 Atterberg Limits 3_6
3.4 Soil Classification 3-10
3.4.1 Grain Size Analysis 3-10
3.4.2 The Unified Soil Classification System 3-15
3.4.2.1 Field Classification 3-15
3.4.2.2 Laboratory Classification 3-17
3.4.2.3 Field Identification Procedures for
Fine-Grained Soils or Fractions .... 3-19
3.5 Compaction 3_2Q
3.5.1 Fundamentals of Compaction !!!!! 3-20
3.5.2 Compaction and Permeability 3-25
3.6 Field Measurement of Density and Moisture
Content 3_2*;
3.6.1 Traditional Methods 3-25
3.6.2 Nuclear Methods 3-28
3.6.2.1 Nuclear Density Gauge 3-28
3.6.2.2 Nuclear Moisture Gauge 3-31
3.7 Testing for Shear Strength 3-33
3.8 Hydraulic Conductivity Testing 3-35
3.8.1 Darcy's Law 3-36
3.8.2 Hydraulic Gradient 3-38
3.8.3 Permeability Measurement and Factors
That Influence Test Results 3-39
3.8.3.1 Sample Selection, Size, and
Preparation 3-40
3.8.3.2 Hydraulic Gradient 3-45
3.8.3.3 Sample Saturation 3-47
vii
-------
TABLE OF CONTENTS (continued)
Chapter
3.8.3.4 Permeant Characteristics 3-49
3.8.3.5 Test Duration «... 3-49
3.8.4 Laboratory Permeability Tests 3-50
3.8.4.1 Pressure Cell 3-50
3.8.4.2 Compaction Permeameter ....... 3-50
3.8.4.3 Triaxial Cells 3-55
3.8.4.4 Consolidation Cells 3-57
3.8.5 Field Permeability Tests .......' 3-57
3.8.5.1 Bore Hole Tests .... 3-57
3.8.5.2 Porous Probes 3-61
3.8.5.3 Air Entry Permeameter ............... 3-63
3.8.5.4 The Guelph Permeameter ............... 3-66
3.8.5.5 Ring Infiltrometers 3-69
3.9 References 3-73
4 Clay-Chemical Interactions and Soil Permeability ............ 4-1
4.1 Parameters Determined in Permeability
Testing for Compatibility 4-2
4.2 Clay-Chemical Interactions that Influence
Permeability 4-3
4.2.1 Soil Fabric and Permeability 4-4
4.2.2 Dissolution by Strong Acids or Bases .......... 4-11
4.2.3 Precipitation of Solids 4-11
4.2.4 Effect of Microorganisms 4-11
4.3 Measuring Clay-Chemical Compatibility
Through Permeability Testing 4-12
4.3.1 Measurement Devices 4-12
4.3.2 Test Setup 4-13
4.3.3 Compatibility of Materials With Test
Fluids 4-13
4.3.4 Effect of Backpressure 4-14
4.3.5 Effect of Hydraulic Gradient 4-14
4.3.6 Criteria for Concluding a Test 4-15
4.4 Summary of Available Research Data ...«. 4-16
4.5 Permeability Studies To Investigate Clay-Chemical
Interactions (test methods, data, and discussion of
results for 23 individual studies) 4-16
4.5.1 Observations by Macey (1942) on
Effects of Organics on Fireclay 4-16
4.5.2 Tests With Kaolinite and Organic
Solvents by Michaels and Lin (1954) 4-26
4.5.3 Study by Buchanan (1964) of the Effect
of Naphtha on Montmorillonite 4-29
4.5.4 Study by Reeve and Tamaddoni (1965)
of the Effect of Electrolyte
Concentration on Permeability of a
Sodic Soil 4-29
viii
-------
TABLE OF CONTENTS (continued)
Chapter " ,:, ;;|,v ' ; 4*, 14 Page
4.5.5 Tests by van Schaik and Laliberte (1968)
of Permeability of Soils to a Liquid
Hydrocarbon „ 4_31
4.5.6 Study by Everett (1977) of Permeability
of Lacustrine Clay to Four Liquid Wastes 4-32
4.5.7 Tests by Sanks and Gloyna (1977) of
Permeability of Lacustrine Clay to
Liquid Waste 4.33
4.5.8 Investigation of the Effect of
Organic Solvents on Clays by
Green, Lee, and Jones (1979) 4-35
4.5.9 Anderson's Study (1981) of the Effects
of Organics on Permeability 4-39
4.5.10 Schramm's Study (1981) of the
Permeability of Soil to Organic Solvents 4-54
4.5.11 Evaluation by Monserrate (1982)
of the Permeability of Two Clays to
Selected Electroplating Wastes 4-59
4.5.12 Research by Brown, Green, and Thomas
(1983) on the Effect of Two Organic
Hazardous Wastes on Simulated Clay Liners 4-60
4.5.13 Study by Brown, Thomas, and Green
(1984) of the Effect of Dilutions
of Acetone and Mixtures of Xylene
and Acetone on Permeability of a
Micaceous Soil 4.54
4.5.14 Tests by Brown, Thomas, and Green
(1984) to Determine the Permeability
of Micaceous Soil to Petroleum Products 4-64
4.5.15 Study by Brown and Thomas (1984) of
the Permeability of Commercially
Available Clays to Organics 4-70
4.5.16 Studies Conducted for EPA by Daniel
(1983) and Foreman and Daniel (1984)
at the University of Texas, Austin 4-72
4.5.17 Tests Conducted for Chemical
Manufacturers Association by Daniel
and Liljestrand (1984) 4-73
4.5.18 Study by Dunn (1983) of the Effects
of Synthetic Lead-Zinc Tailings
Leachate on Clay Soils 4-81
4.5.19 Studies by Acar and Others (1984) on the
Effect of Organics on Kaollnite 4-82
4.5.20 Finding by Olivieri (1984) of
Impermeability of Montmorillonite to
Benzene 4_8-:}
4.5.21 Study of Permeability of Clays to
Simulated Inorganic Textile Wastes
by Tulis (1983) 4_84
IX
-------
TABLE OF CONTENTS (continued)
Chapter Page
4.5.22 Tests Conducted by Engineering
Consulting Firms for Specific Applica-
tion (unpublished data) ....,, 4-84
4.5.23 Tests Reported by Bentonite Companies 4-94
4.6 References . ... 4-93
5 Current Practices: Clay Liner Design and Installation 5-1
5.1 Design 5-1
5.1.1 Site Investigation 5-2
5.1.2 Liner Material Selection and
Characterization 5-6
5.1.2.1 Native Soils 5-13
5.1.2.2 Admixed Soils 5-14
5.1.3 Facility Design 5-16
5.1.3.1 Configuration 5-16
5.1.3.2 Foundation Design 5-16
5.1.3.2.1 Settlement 5-16
5.1.3.2.2 Seepage 5-17
5.1.3.2.3 Dike Design 5-18
5.1.3.2.4 Sidewall Design .. 5-21
5.1.3.2.5 Bottom Design 5-27
5.1.3.3 Liner Design 5-27
5.1.3.4 Special Design Considerations 5-32
5.1.3.4.1 Control of Erosion 5-32
5.1.3.4.2 Control of Scouring 5-33
5.1.3.4.3 Cold Climate Design 5-34
5.1.3.4.4 Control of Piping 5-34
5.1.3.4.5 Control of Desiccation .. 5-35
5.1.3.4.6 Seismic Design ...» 5-35
5.1.3.4.7 Intergradient Facility
Design ..... 5-39
5.1.4 Construction Specifications and CQA Plan ...... 5-40
5.1.5 Design Case Studies . 5-42
5.2 Clay Liner Construction: Methodology -
and Equipment „ 5-46
5.2.1 Preinstallation Activities 5-46
5.2.1.1 Foundation Preparation 5-46
5.2.1.2 Groundwater Control . 5-47
5.2.1.3 Leak Detection System Installation .. 5-48
5.2.2 Clay Liner Installation 5-48
5.2.2.1 Natural Soil Liners 5-48
5.2.2.1.1 Liner Material
Emplacement ............. 5-49
5.2.2.1.2 Clod Size Reduction 5-49
5.2.2.1.3 Moisture Control .„ 5-52
5.2.2.1.4 Compaction 5-55
5.2.2.2 Admixed Bentonite Liners 5-70
5.2.2.3 Climatic Effects 5-75
5.2.3 Postinstallatlon Activities 5-80
-------
TABLE OF CONTENTS (continued)
Chapter " i4i-**.-\-"J ^..r. Page
5.3 Quality Assurance and Quality Control
5.3.1 Key Terms
5.3.2 Personnel
5.3.3 Observations and Tests
5.3.4 Documentation
5.4 Clay Liner Design and Construction: Problems
and Preventive Measures
5.5 References
Failure Mechanisms
6.1 Desiccation Cracks
6.1.1 Description
6.1.2 Studies of Cracking
6.2 Slope Instability
6.2.1 Description
6.2.2 Discussion of Slope Instability
6.3 Settlement
6.3.1 Description
6.3.2 Studies of Settlement
6.4 Piping
6.4.1 Description „
6.4.2 Studies of Piping
6.5 Penetration „
6.5.1 Description „
6.5.2 Studies of Penetration
6.6.1 Description
6.7 Cold Climate Operations
6.8 Earthquakes
6.9 Scouring
6.10 Failures from Design or Construction Errors
6.11 References
Clay Liner Performance
7.1 Introduction i
5-82
5-83
, . . . . 5-86
, . . . . 5-88
..... 5-98
..... 5-109
5-112
6-1
6-1
6-1
6-2
6-3
.... 6-3
.... 6-4
.... 6-5
6-5
6-5
.... 6-6
6-6
6-6
6-8
.... 6-8
.... 6-9
6-9
.... 6-9
.... 6-9
.... 6-11
.... 6-15
.... 6-16
6-17
.... 7-1
7-1
7.2 Case Studies (physical description; startup date;
geology and hydrology; waste type; liner description,
installation, and performance for 17 clay-lined
" 7-1
7-2
7-5
7-7
7-9
7-14
7-18
7-21
7-25
7-28
7-33
7-38
xi
7.2.1
7.2.2
7.2.3
7.2.4
7.2.5
7.2.6
7.2.7
7.2.8
7.2.9
7.2.10
7.2.11
Criteria
Site A .,
Site B .,
Site C .,
Site D .,
Site E .,
Site F ..
Site 6 ..
Site H .,
Site I ..
Site J ..
for Site Selection
-------
TABLE OF CONTENTS (continued)
Chapter
Page
7.2.12 Site K .... 7-42
7.2.13 Site L 7-46
7.2.14 Site M 7-49
7.2.15 Site N 7-55
7.2.16 Site 0 7-59
7.2.17 Site P 7-62
7.2.18 Site Q 7-67
7.3 Liner Types 7-70
7.3.1 Unlined Facilities 7-71
7.3.2 Recompacted Soil Liners 7-72
7.3.3 Admixed Liners 7-72
7.4 Site Characterization . ....' 7-73
7.4.1 Case Studies 7-74
7.5 Installation of Clay Liners 7-74
7.5.1 Installation Methods ......... 7-75
7.5.2 Quality Assurance/Quality Control
for Clay Liners ....... 7-76
7.6 Waste Types ....... 7-77
7.6.1 Free Liquids 7-77
7.6.2 Stabilized or Solidified Liquids .............. 7-78
7.6.3 Sludges and Solid Wastes .. 7-78
7.6.4 Waste Compatibility ....... 7-79
7.7 Performance Monitoring 7-81
7.7.1 Unsaturated Zone Monitoring 7-82
7.7.2 Groundwater Monitoring .......: 7-84
7.7.3 Leachate Level and Quality
Monitoring 7-85
7.8 Conclusions . 7-85
7.9 References 7-86
8 Prediction of Clay Liner Performance 8-1
8.1 Introduction 8-1
8.2 Background Considerations ....... 8-1
8.2.1 Performance Criteria — ,..«,... 8-1
8.2.2 Clay Liner System ....... 8-2
8.2.3 General Equations 8-4
8.3 Transit Time Prediction Methods .„...; 8-6
8.3.1 Simple Transit Time Equation ....... 8-6
8.3.2 Modified Transit Time Equation 8-8
8.3.3 Green-Ampt Wetting Front Model „.. 8-9
8.3.4 Transient Linearized Infiltration
Equation 8-10
8.3.5 Numerical Solutions „... 8-12
8.4 Comparison of Different Approaches .; 8-14
8.5 Batch-Type Absorption Procedures for Estimating
Clay Liner Performance ,..„... 8-14
8.6 References 8-16
Appendix
A Test Method Descriptions A-l
xii
-------
FIGURES
Number v;'f'i ' ,'»•$-•..*? Page
1-1 Cross section of an idealized clay liner system 1-2
2-1 Clay mineral tetrahedral sheet structure 2-3
2-2 Clay mineral octahedral sheet structure 2-3
2-3 Kaolinite group minerals 2-7
2-4 mite clay minerals 2-11
2-5 Chlorite and smectite clay minerals 2-1?
2-6 Electrical double layer 2-19
2-7 Effect of solution pH on clay mineral surface
charge (EPM) 2-21
2-8 Comparisons of clay mineral sizes and surface areas 2-23
2-9 Clay soil fabrics 2-25
2-10 Fractures in glacial till 2-28
2-11 Root cast in glacial till 2-29
2-12 Permeable strata in glacial till deposit 2-30
2-13 Compaction curve from a standard compaction test 2-33
2-14 Compaction curves for different compactive efforts
applied to a silty clay 2-34
2-15 Permeability as a function of molding water content
for samples of silty clay prepared to constant
density by kneading compaction 2-36
2-16 The effect of dispersion on hydraulic conductivity 2-37
2-17 Effect of method of compaction on the permeability
(hydraulic conductivity) of a silty clay 2-39
3-1 Schematic representation of soil illustrating the
fundamental relationships among the solid, liquid,
and air constituents 3-4
3-2 Consistency limits of cohesive soils 3-6
3-3 Device for determining the liquid limits of a
cohesive soil. The dish contains a grooved sample 3-7
3-4 Clay sample being grooved for liquid limit test 3-7
3-5 Rolling a clay sample for plastic limit test 3-9
3-6 Results of rolling clay with moisture content below
the plastic limit 3_9
3-7 Typical relationships between the liquid limit
and the plasticity index for various soils 3-11
3-8 Idealized particle size distribution curves for
well-graded, poorly-graded, and gap-graded soils 3-14
3-9 Unified soil classification chart 3-16
3-10 Typical soil compaction curve illustrating maximum
dry density and optimum water content 3-21
3-11 Compaction curves for different compactive efforts
applied to a silty clay 3-23
3-12 Four types of compaction curves found from
laboratory investigation 3-24
3-13 Permeability as a function of molding water content
for samples of silty clay prepared to constant density
by kneading compaction 3-26
xiii
-------
FIGURES (continued) ;
Number Page
3-14 Influence of the method of compaction on the
permeability of silty clay 3-27
3-15 Schematic diagram of triaxial compression apparatus
for Q test 3-34
3-16 Effect of backpressure on permeability to water,
Sasumua clay 3-48
3-17 Apparatus for pressure cell method 3-51
3-18 Modified compaction permeameter 3-53
3-19 Detail of the base plate for a double-ring permeameter 3-54
3-20 Schematic of a constant head triaxial cell permeameter ...... 3-56
3-21 Consolidation permeameter 3-58
3-22 Two-stage, borehole permeability text (Boutwell and
Derick, 1986) 3-59
3-23 Installed porous probe (Daniel, 1987) 3-62
3-24 Modified air-entry permeameter 3-64
3-25 Schematic diagram of Guelph permeameter 3-67
3-26 Double-ring infiltrometer 3-70
3-27 Sealed double-ring infiltrometer 3-72
4-1 Change in a pore diameter (400%) corresponding to a
permeability increase of 25,600% 4-5
4-2 Distribution of ions adjacent to a clay surface
according to the concept of the diffuse double layer ........ 4-7
4-3 Intrinsic permeabilities as a function of void space
(e) measured for different permeants 4-28
4-4 Coefficient of permeability of Ranger shale to
various chemicals „..; 4-40
4-5 , Permeability of the four clay soils to water
(0.01N CaS04) 4-43
4-6 Permeability of the four clay soils to acetic acid .......... 4-44
4-7 Permeability and breakthrough curves of the four
clay soils treated with aniline 4-45
4-8 Permeability of the four clay soils to ethylene glycol ...... 4-46
4-9 Permeability of the four clay soils to acetojie 4-47
4-10 Permeability of the four clay soils to methanol
and the breakthrough curve for the methanol-treated
mixed cation illitic clay soil 4-48
4-11 Permeability of the methanol-treated mixed cation
illitic clay soil, at two hydraulic gradients 4-49
4-12 Permeability and breakthrough curves of the four
clay soils treated with xylene 4-50
4-13 Permeability and breakthrough curves of the four
clay soils treated with heptane 4-51
4-14 Variation of intrinsic permeability with solvent
for each soil 4-58
4-15 Permeability of White Store clay to 0.01 N calcium
sulfate, chromic acid (1 molar), and zinc chloride
(1 molar) as a function of moisture at compaction 4-61
xiv
-------
FIGURES (continued)
Number 'I'^F ;':|. » page
4-16 Hydraulic conductivity versus pore volume for
laboratory-compacted micaceous soil exposed to
kerosene at a hydraulic gradient of 91 4-65
4-17 Hydraulic conductivity versus pore volume for
laboratory-compacted micaceous soil exposed to
diesel fuel at a hydraulic gradient of 91 4-66
4-18 Hydraulic conductivity versus pore volume for
laboratory-compacted micaceous soil exposed to
paraffin oil at. a hydraulic gradient of 91 4-67
4-19 Hydraulic conductivity versus pore volume for
laboratory-compacted micaceous soil exposed to
gasoline at a hydraulic gradient of 91 4-68
4-20 Hydraulic conductivity versus pore volume for
laboratory-compacted micaceous soil exposed to
motor oil at a hydraulic gradient of 91 4-69
4-21 Permeability versus number of pore volumes of flow
for kaolinite permeated with methanol at a hydraulic
gradient of 250 or 300 4-74
4-22 Permeability versus hydraulic gradient for kaolinite
permeated in flexible-wall permeameters 4-75
4-23 Permeability versus hydraulic gradient for kaolinite
permeated in consolidation cell permeameters 4-76
4-24 Permeability versus hydraulic gradient for kaolinite
permeated in compaction mold cell 4-77
5-1 Compacted clay cutoff seal 5-19
5-2 Dike components and typical configurations 5-20
5-3 Methods of liner sidewall compaction 5-22
5-4 Liner design for collection system pipes and sump 5-28
5-5 Methods of keying-in liner segments 5-30
5-6 Liner material emplacement 5-50
5-7 Emplacement of liner material over foundation
excavation underneath a collection pipe 5-51
5-8 Use of pulvi-mixer for clod size reduction 5-54
5-9 Moisture addition to liner material prior to
compaction 5_56
5-10 Joints and seepage along lift boundaries 5-61
5-11 Sketches of different types of roller feet 5-65
5-12 Various compacting rollers .....^... 5-67
5-13 Compaction on a 2(H) to 1(V) slope with a towed
sheepsfoot roller 5-69
5-14 Central plant mixing of bentonite and soil 5-72
5-15 Truck-loaded bentonite spreader 5-73
5-16 Pneumatically fed bentonite spreader 5-76
5-17 Blending bentonite with soil using a disk harrow 5-78
5-18 Soil stabilizer mixing bentonite in place 5-79
5-19 Inflatable dome over a hazardous waste landfill 5-81
5-20 CQC test location and data summary 5-103
5-21 Statistical analysis of CQC test data 5-104
xv
-------
FIGURES (continued)
Number Page
6-1 Location of past destructive earthquakes in the
Uni ted States „.. 6-12
6-2 Differences in propagation of damage for eastern
and western earthquakes „.. 6-12
7-1 Plan view of site A 7-6
7-2 Plan view of site B „.. 7-8
7-3 Plan view of site C ..... 7-10
7-4 Cross-sectional view of site C (vertical
dimensions are to scale) 7-12
7-5 Plan view of site D 7-15
7-6 Cross-sectional view of site D liner 7-17
7-7 Plan view of site E .........: 7-19
7-8 Cross-sectional view of site F 7-22
7-9 Plan view of site G 7-26
7-10 Plan view of site H 7-29
7-11 Cross-sectional view of site H liner showing
details of leachate collection system and lysimeter
constructi on 7-30
7-12 Plan view of site I 7-36
7-13 Cross section of liner at site I 7-37
7-14 Plan view of site J ... 7-39
7-15 Cross-sectional view of site J liner ... 7-41
7-16 Cross-sectional view of site K liner „.. 7-43
7-17 Cross section of containment system at site L „.. 7-47
7-18 Cross section of site M „.. 7-50
7-19 Plan view of site M leachate collection and leak
detection systems 7-51
7-20 Plan view of site N 7-57
7-21 Cross-sectional view of site N liner and leachate
management systems 7-53
7-22 Cross section of site P showing relationship of
liner and dikes 7-63
7-23 Detailed cross section of site P liner 7-64
7-24 Cross section of site Q liner 7-69
8-1 Flow domain for leachate flow 8-3
xvn
-------
TABLES
• *"- '' ' Jg L •:• .
No. Page
2-1 Clay Mineral Characteristics 2-5
3-1 Soil Tests Summarized in Appendix A 3-3
3-2 U.S. Standard Sieve Sizes and Their Corresponding
Open Dimension 3-13
3-3 Z/A of Various Soil Components 3-30
3-4 Summary of Potential Errors in Laboratory
Permeabi 1 ity Tests -on Saturated Soi 1 3-41
3-5 Summary of Sources of Error in Estimating Field
Permeability of Compacted Clay Liners
from Laboratory Tests 3-42
3-6 Test Results Showing Effect of Sample
Diameter on Permeability Measurements 3-45
4-1 Results of Permeability Tests With Organic Chemicals 4-17
4-2 Results of Permeability Tests With Wastes 4-24
4-3 Void Ratio and Coefficient of Permeability
Relationships for Calcium- and Sodium-
Montmorillonite Permeated by Water and Naphtha 4-30
4-4 Summary of Soil Permeability With Soltrol C and Water 4-30
4-5 Permeabilities Measured With Lacustrine Clay
Exposed to Water and Waste Liquids 4-34
4-6 Properties of Soils Tested 4-36
4-7 Classification of Clay-Organic Solvent Systems
According to Swell Properties 4-37
4-8 Percent Swell for Clay Soils in Contact With
Organic Liquids and Water 4-38
4-9 Grain Size Distribution, Mineralogy, and Properties
of the Four Clay Soils 4-41.
4-10 Characteristics of Soils Used in Permeability Tests 4-56
4-11 Permeability Coefficients (cm/s) Determined
in Soils Tested With Organic Solvents 4-57
4-12 Mean Conductivity of Each Soil to Each Fluid „
Tested (Brown and Thomas, 1984) 4-71
4-13 Properties of Clay Soils Tested by Daniels and
Liljestrand (1984) 4-79
4-14 Leachate Characteristics 4-80
4-15 Permeability Test Results (Pennsylvania Case A) 4-87
4-16 Permeability Test Results (Pennsylvania Case B) 4-88
4-17 Chemical Characteristics of Waste Permeants, Project E 4-91
4-18 Results of Permeability Tests, Project E 4-92
4-19 Results of Permeability Tests, Project L 4-95
4-20 Initial and Final Permeabilities Determined
in Triaxial Cell Tests With Leachates, Project N 4-95
4-21 Effect of Concentrated Organics on a Treated
Bentonite Seal 4-97
4-22 Permeability (cm/s) of a Treated Bentonite
Seal to Kerosene 4-97
xvn
-------
TABLES (continued)
No.
Page
5-1 Accessible Methods of Subsurface Exploration 5-7
5-2 Nonaccessible Methods of Subsurface Exploration 5-9
5-3 Properties of Soils Used To Construct Soil Liners 5-11
5-4 Properties of Soils Used To Construct Soil Liners 5-112
5-5 Factors Controlling Stability of Sloped Cut
in Some Problem Soils „ 5-25
5-6 Relative Volume Change of a Soil as Indicated
by Plasticity Index and Other Parameters 5-36
5-7 Soil Volume Change as Indicated by Liquid Limit and Grain
Si ze . s_36
5-8 Effect of Clod Size on Permeability of Laboratory
Compacted Clay „ 5.153
5-9 Compaction Equipment and Related Specifications for
Constructing Soil Liners 5-58
5-10 Compaction Equipment and Methods 5-63
5-11 Current QA Practices for Clay Liner Construction 5-99
5-12 Recommendations for Construction Documentation
of Clay-Lined Landfills by the Wisconsin Department
of Natural Resources 5-101
5-13 Elements of a Construction Documentation Report 5-102
5-14 Potential Clay Liner Design and Installation
Problems and Preventive Measures 5-110
7-1 Clay-Lined Facility Information 7-3
7-2 General Occurrence of Chemical Parameters in the
Groundwater at Site C 7-13
7-3 Lysimeter (L) and Leachate Collection System (LCS)
Liquid Volumes (gal) at Site H 7_'(2
7-4 Monitoring Data for Site H 7-34
7-5 Heavy Metal Content and Percent Solids of Lime
Sludge Disposed at Site M 7-Bi2
7-6 Groundwater Monitoring Well Sample Analysis at
Site M o.. 7.54
7-7 Leachate Analysis at Site M .....!!!! 7-56
7-8 Leachate Volumes at Site M 7-57
7-9 Water Sample Analysis: BOD, COD, Total Coliform,
and Fecal Coliform 7.50
8-1 Comparison of Transit Time Predictions 8-15
xviii
-------
ACKNOWLEDGMENTS
The support of the U;S. Environmental Protection Agency, Cincinnati,
Ohio, and Dr. M. H. Roulier, EPA Project Monitor, is greatly appreciated.
The draft version of this document was prepared at the Research Triangle
Institute by Dr. L. J. Goldman, Project Leader (currently with the NUS
Corporation), along with Mr. R. S. Truesdale, Ms. G. L. Kingsbury, Ms. C. M,,
Northeim, and Mr. A. S. Damle. This final version was prepared by the NUS
Corporation by Dr. L. J. Goldman, Project Leader, and Mr. L. I. Greenfeld.
The authors wish to acknowledge the firms, agencies, and individuals
that provided much of the information contained in this document. We also
wish to acknowledge the contributions of those individuals who provided
reviews, comments, and suggestions on the draft version of this TRD.
xix
-------
-------
CHAPTER 1
INTRODUCTION
A clay liner consists of one or more layers of cohesive soil that have
been compacted to achieve a low permeability. The purpose of a clay liner'in
a waste management facility (landfill, waste pile, or surface impoundment) is
to serve as a barrier between waste materials and the hydrogeologic environ-
ment by limiting seepage from the .facility and to provide support for over-
lying components of the facility.
A clay liner is usually constructed of native soil that contains
appreciable amounts of clay-sized particles; in some cases other materials
such as bentonite, are mixed with the soil when it does not contain suffi-
cient clay. A clay liner may be overlain by one or more flexible membrane
liners and primary and secondary leachate collection (drain) systems.
Figure 1-1 illustrates one possible liner configuration for a surface
impoundment.
A compacted clay soil liner, because of its low permeability (hydraulic
conductivity), limits the steady-state seepage from a facility. In waste
management facilities the clay liner also is designed to delay the release of
leachate for the longest possible time (transit time) and to have sufficient
structural stability to support itself and other components of the facility
that may lie above it. Clay liners are used not only in waste management
facilities but also in many other applications such as water storage and
conveyance structures.
Although clay liners are a widely used technology, there is very little
information on their performance in limiting seepage from operating waste
facilities. [7.2.1; 7.3.2]* Performance predictions have been based
primarily on results from permeability tests conducted on laboratory-
compacted soils. Laboratory testing has provided a great deal of data that
demonstrate the permeability (hydraulic conductivity) that can be achieved
with various soil materials and compaction methods and how various types of
wastes can interact with soil materials to change this permeability. Even
though it is common practice to use such laboratory results to predict the
field behavior of a liner, there are few field data that confirm the validity
of laboratory-derived soil permeability values as a measure of field permea-
bility. There are some field data showing that permeability and seepage
rates for field-compacted clay soil liners are much greater than for
laboratory-compacted soils. [2.3.2; 3.8.3.1] Until more field data are
available on the performance of clay liners constructed with the best current
*Numbers in brackets refer to relevant sections of this document.
1-1
-------
Cover Soil and/or
Riprap
Leachate Collection
System
Foundation
Not to scale.
Bottom Liner-
Compacted Low
Permeability Soil
Component
Top Liner-
Synthetic Membrane
Bottom Liner-
Synthetic Membrane
Component
Figure 1-1. Cross section of an idealized clay liner system.
-------
technology, permeabilities of laboratory-compacted soils should be regarded
as a goal rather than an accurate estimate of the performance of field-
constructed clay liners. « >&. >ti. ,*
1.1 SCOPE
The objective of this Technical Resource Document (TRD) 1s to provide,
in a single source, all of the available information on design, construction,
and performance of clay-lined waste management facilities. The broad topics
covered are: clay properties and characteristics, [2.0]; geotechnical testing
of soils [3.0]; the compatibility of clays and chemical wastes [4.0]; the
design, construction, and construction quality assurance of clay liners
[5.0];, potential failure mechanisms [6.0]; the performance of existing facil-
ities [7.0]; and methods of predicting the useful life (transit time) based
on the modeling of leachate flow through soils [8.0]. It should be noted
that the design and construction section [5.0] is limited to practices
currently in use by engineers. Information on recommended practices is
available in guidance documents issued by the USEPA Office of Solid Waste,
Washington, D.C.
One of the major sources of information for this document was a series
of in-depth interviews with over 30 design/construction engineers who have
hands-on experience with clay-lined facilities in all areas of the country.
Some of the material obtained in these interviews appears here, in print, for
the first time. Opinions and preferences stated in this document, except
when referenced to published material, were obtained from these interviews
and do not necessarily represent the preferences of the authors or USEPA.
This document contains some soils information taken from a Technical
Resource Document on flexible membrane liners, Lining of Waste Impoundment
and Disposal Facilities (Haxo, 1983). Other important sources of information
were the published literature, discussions with researchers, interviews with
personnel from waste management companies and industries, and contacts with
State and Federal regulatory personnel. Waste management companies and
regulatory agencies were the primary sources for the information contained in
Chapter 7, Clay Liner Performance.
1.2 SUMMARY OF CURRENT PRACTICES
*
The following is a summary of major current practices in clay liner
construction that are believed to affect the permeability of the liner; other
aspects of earthwork (cut and fill volumes, etc.) are not covered. The
practices summarized are average or typical of a large number of the cases
observed by the authors at operating sites or in construction documentation.
Instances of better or poorer practices were also frequently encountered by
the authors.
1.2.1 Investigation of Site Conditions (Field)
Site investigations are conducted before construction to identify and
investigate borrow areas for liner material. The borrow area and the soils
that will underly the liner are investigated through borings, pits, and
trenches cut across the area. In situ soil properties and groundwater condi-
tions are identified and taken into account in the design to avoid structural
1-3
-------
problems such as hydraulic uplift (heaving) or settlement of the soil under-
lying the liner. The foundation (soil) underlying the proposed facility site
is evaluated for structural properties through laboratory tests on removed
samples and in situ testing. [5.1.1; 5.3.3.2.1]
1-2.2 Material Selection and Characterization (Laboratory)
1.2.2.1 Foundation Soil--
Soil underlying the proposed facility must possess bearing and shear
strength adequate to support expected loading. If tests on samples examined
In the laboratory and onsite bearing tests show inadequate properties, the
site design and construction plans will include specifications that provide
for excavation and recompaction of an adequate foundation. [5.1.3-2.1; 6.2;
6.3]
1.2.2.2 Index Properties of Liner Materials--
Samples of soils from potential borrow areas are analyzed to determine
their index properties (grain size distribution, Atterberg limits, and
Unified Soil Classification). [3.3; 3.5.2; A-26; A-28; A-29; A-34; A-37]
This information is used to identify and reject undesirable materials (i.e.
low plasticity index or low fines content) and to select desirable
materials.
1.2.2.3 Engineering Properties of Liner Material —
A very important criterion for a suitable liner soil is whether it can
be compacted so that the permeability (hydraulic conductivity), as
measured in the laboratory, is 10-' cm/s or less. A number of engineer-
Ing firms and regulatory agencies have guidelines for the plasticity index
and amount of fines (amount passing a No. 200 sieve) and some have acceptance
criteria that consider only plasticity index or fines, but, at present,
permeability appears to be the most heavily weighted criterion. i
Selection of a liner material that can be compacted to the required
permeability involves a series of laboratory tests of the engineering proper-
ties of the candidate materials. A moisture content/density relationship is
established for the material by compacting samples of the material at various
moisture contents with a set compactive effort. [2.3.2; 3.4.1; 3.8.4.5;
uf5^ 1V47; A~5°J The °Pt1mum molding water content is the water content at
which the maximum dry density is achieved. The permeability (hydraulic
conductivity) of a sample of the soil compacted with the same compactive
effort at a water content slightly greater than optimum is measured. [2.3.2;
l2 1 -1 he relat1onsfiip between density, moisture content, compactive
effort, and permeability derived from these data will be used to develop
construction specifications and, for quality control of the construction
process, to ensure that the water content is slightly wetter than optimum and
that compaction has been sufficient to achieve the desired permeability.
* fu ;,?*6; 5<1'4; 5.3.3.2.4] It is also necessary to construct a test fill
of the liner material prior to construction to confirm that the laboratory-
derived relationship can be achieved in the field. [1.2.4; 5.2.2.1.4;
5.3.3.1]
In addition to meeting permeability requirements, liner material must be
compatible with the waste it is meant to contain. A laboratory-compacted
sample of the son will be tested to determine whether wastes or leachates
1-4
-------
from the proposed facility will react with it to increase its permeability
and decrease Its ability to control seepage from the facility. [4.3; 4.4;
7.6.4] . ••?- f ' " =>
1.2.3 Develop Liner Design/Construction Plans
1.2.3.1 Lift and Liner Thickness—
The depth of loose soil laid down for compaction (lift thickness) is
selected so that the equipment and operating conditions will be capable of
imparting the required compactive effort throughout the lift. [5.2.2.1.1]
Minimum liner thickness is usually set by State or Federal regulations.
Beyond these minimums, liner thickness may be established on the basis of the
time required for liquid or waste to pass through the liner (transit or
containment time). [5.1.3.3; 7.7; 8.0] Greater than average liner thickness
may be specified at critical points such as the junction of the bottom and
side and below leachate collection lines or sumps. [5.2.2.1.1]
1.2.3.2 Compactive Effort--
Equipment and operating conditions (weight, number of passes, etc.) are
selected in conjunction with lift thickness to ensure that the same compac-
tive effort will be applied in the field as was used in the laboratory to
achieve the specified permeability. [3.4; 5.2.2.1.4] This selection will be
confirmed in the test fill. [1.2.4]
1.2.3.3 Adjust Design to Avoid Failure Mechanisms—
The design of the liner and the foundation (underlying soil) will be
adjusted to avoid common failure mechanisms, as appropriate:
« Erosion [5.1.3.4; 6.6.2]
« Freezing [5.1.3.4.3; 5.2.2.3.2; 6.7]
• Piping [5.1.3.4.4; 6.4.3]
o Slope instability [6.2]
« Settlement [6.3]
« Earthquake [5.1.3.4.6; 6.8]
» Scouring (surface impoundments) [5.1.3.4.2; 6.9]
(i Hydraulic uplift (heaving) [5.1.3.4.7; 7.4].
1.2.4 Pilot Construction Test (Test Fill)
— Him _-_.-!• i.-r-.i-j..-.jii-_-___ll___u.L ^ *i —
A small-scale construction test is conducted to determine whether the
selected combination of lift thickness, equipment, operational procedures,
moisture content, and density will result in the design permeability identi-
fied during the laboratory testing. [5.2.2.1.4; 5.3.3.1]
1.2.5 Construction
1.2.5.1 Foundation Preparation—
The native soil is the foundation for the clay liner. The soil surface
will be, at a minimum, cleared, grubbed, stripped, and cleaned of organic or
otherwise deleterious material prior to placement of the liner material.
Resulting holes and depressions in the soil or underlying rock will be
filled. Where necessary, the soil may be excavated and recompacted using
standard earthmoving equipment. Water content and compactive effort will be
1-5
-------
adjusted to give a firm surface for construction of the Uner. The surface
may be proof-rolled to locate soft spots prior to placing liner material.
[5.2.1.1; 5.3.3.2.1]
1.2.5.2 Liner Material Preparation-
Soil material for the liner will be either excavated on site or brought
from a nearby borrow area. Areas containing material that is unsuitable
because the fines content is too low or because it contains rocks, organic
matter, etc., will be identified; material from these areas will be segre-
gated from the liner material and not used in the liner. [5.3.3.1; 7,5.1.2]
If the material is too dry or heterogeneous, construction equipment will
be used to mix 1t and the water content may be adjusted by addition of water
prior to placement. Adequate curing time must be allowed to enable added
water to uniformly penetrate all the liner material. Stockpiles of liner
materials are often covered or seal-rolled to retard drying or erosion.
[5.2.2.1]
1.2.5.3 Material Placement--
Liner material will be hauled to the site from a stockpile or borrow
area and spread to the specified loose lift thickness as estimated with a
shovel blade or staff. [5.2.2.1.1] Rocks and other foreign material may be
removed 1f apparent. Clods that are larger than the 11ft thickness may be
reduced 1n size with a disc or harrow. [5.2.2.1.2; 5.3.3.2.2]
Water content of the soil will have been measured on samples from the
emplaced material. [3.6.2.2; A-6; A-7] If water content 1s less than speci-
fied in the design, water will be added by spraying from a truck or large
hose before the soil is compacted. Adequate curing time must be allowed. If
the son is too wet, 1t will be allowed to dry somewhat before compaction.
[5.2.2.1.3; 5.3.3.2.3]
1.2.5.4 Compaction—
The soil Is compacted using equipment such as sheepsfoot or rubber-tired
rollers until the density of the soil, at the specified moisture content, has
reached the value specified in the design. [5.2.2.1.4; 5.3.3.2.4] During
construction, density may be measured by direct methods such as the sand cone
or balloon but is more commonly measured with a nuclear density gauge.
[3.6.2.1; 5.3.3.2.4; 5.3.3.2.5; A-19; A-20; A-22] Samples of soil may be
taken during compaction to determine 1f the water content 1s close enough to
the design value. Alternatively, a nuclear moisture gauge may be used.
The top of a 11ft (layer of soil) may be scarified with a harrow or
other equipment so that there will be an adequate bond with the lift placed
above 1t. The edges of lifts are often beveled or overlapped to ensure
complete coverage. [5.1.3.3; 5.2.2.1.1]
1.2.5.5 Completion—
Upon completion of the liner, samples may be taken for laboratory perme-
ability measurements; alternatively, field permeability measurements may be
made on the completed liner.
1-6
-------
The surface of the liner may be proof-rolled to seal it and prevent
erosion and the liner may be covered with a protective layer of soil to
reduce desiccation cracking if there will be considerable time before the
liner is placed in service or before a leachate collection system or flexible
membrane liner is placed over the clay liner. [5.1.3.4.5; 5.2.2.3.1; 6.1.3]
1.3 ANALYSIS OF CURRENT PRACTICES
The current practices discussed in this section are those that would bp
most likely to .improve clay liner performance (lower permeability) if they
were changed. Most of this analysis is based on laboratory studies of the
behavior of compacted soils; few .field studies of the relationship between
construction practice and permeability have been conducted.
1.3.1 Liner Material
Changes in the amount of fines (material passing a No. 200 sieve) in a
soil strongly affect the relationship between compactive effort and water
content and the maximum density and permeability that can be achieved.
Consequently, efforts in the following areas are likely to result in a liner
with a more uniform and lower permeability:
Identification of the range of soil properties in the borrow
area
« Determination of the water content, compactive effort, density, and
permeability relationships for all the major and significantly dif-
ferent bodies of soil 1n the borrow area, and verifying these rela-
tionships in a field-compacted test fill
• Sorting soils so that those delivered to the construction site do not
have properties significantly different from the soils that were used
in developing the construction specifications.
1.3.2 Clod Size
One laboratory study has shown that, for the same water content and
compactive effort, permeability increased as the size of the clods
Increased. The mechanism responsible has not been identified but it is
ass.umed to be nonuniform compaction, nonuniform moisture distribution, or
inadequate bonding between adjacent aggregates, leaving planes of weakness in
the soil sample and areas of higher permeability.
fn to r??uce clod s1ze dur1"9 excavation and placement of the soil
for the liner would improve chances for achieving a lower permeability in
several ways: J
• A small clod is more likely to have a uniform water content; the non-
uniform water content of larger clods will lead to differences in
density and permeability after application of the same compactive
effort.
• At any given water content, the material could be compacted more
uniformly because the range in clod sizes would be smaller.
1-7
-------
• If water addition were necessary, the added water could be distrib-
uted-.more evenly; shorter times would be required for water to move
from the outside to the inside of smaller clods than of larger
clods.
1.3.3 Water Content
Laboratory studies using homogenized soil samples that have been ground
to pass a 2-mm sieve have demonstrated that as water content is increased
several percentage points above optimum, the permeability of the soil when
compacted decreases sharply. Permeability may decrease by a factor of 100
over a water content range of only 2.0 to 4.0 percent above optimum.
During construction, a uniform water content slightly higher than the
optimum for maximum density is likely to result in lower permeabilities than
would water contents that are optimum or less. Considerable effort may be
required to control water content'under field conditions but all available
information indicates that a considerable return may be realized in terms of
lower permeability.
1.3.4 Pilot Construction (Test Fill) '.
There are a number of differences between laboratory and field condi-
tions (uniformity of material, control of water content, compactive effort,
compaction equipment, etc.) that make it unlikely that permeabilities
measured on laboratory-compacted samples can be achieved during construc-
tion. In addition, there are several processes and conditions"that cannot be
examined or anticipated through laboratory work (e.g., control of desiccation
cracking, bonding of lifts, degree of compaction on sidewalls),, Pilot
construction (test fill) provides an opportunity to verify that the materi-
als, equipment, and personnel that will be used to construct the liner can
produce a liner that performs according to the design. Specific factors that
can be examined/tested during construction of a test fill to increase the
probability of achieving minimum permeability in the actual clay liner are:
• Preparation and compaction of foundation material to the
required bearing strength
• Methods of controlling uniformity of material, clod size, and water
content
• Types of equipment and estimates of operational conditions (e.g.,
number of passes).required to achieve design density, compactive
effort, and permeability
t Lift thickness and placement procedures necessary to achieve uniform-
ity of density throughout a lift and the absence of apparent boundary
effects between lifts or between placements in the same lift
t Procedures for protecting against desiccation cracking or other site-
and season-specific failure mechanisms for the finished liner or
intermediate lifts
1-8
-------
• Measuring the permeability on the test fill in the field and collect-
ing samples of field-compacted soil for laboratory testing
• Test procedures for controlling the quality of construction
°f S011 t0 meet Permeabi11ty requirements
• Skill and competence of the construction team.
1.4 SUMMARY
nnnhaSlS 11ners.are a widely used technology for management of hazardous and
nonhazardous wastes. Field data on the performance of clay liners or on the
?-I?JV yariouf construction procedures on the permeability of liners are
inn ted. Laboratory data show that low permeabilities (less than
lu cm/s) can be achieved by compacting soils. The few case studies
™ *!! been1conducted Su99est tnat permeabilities measured on laboratory-
compacted samples or on small-diameter intact samples are poor predictors of
performance of the actual liner. A large body of information exists that
like?v ?0W?InrnufP??nf °f C5nstruct1on Practice, if changed, would be most
nMiTOZi^
with values approaching
1.5 REFERENCE
Haxo, H. E. et al. 1983. Lining of Waste Impoundment and Disposal
Facilities. SW-870, U.S. Environmental Protection Agency, Cincinnati,
1-9
-------
-------
CHAPTER 2
CLAY SOIL
Clay liners are composed of layers of cohesive soil, engineered and com-
pacted to form a barrier to liquid migration. From an engineering stand-
point, soil has been defined as unconsolidated accumulations of solid
particles produced by the physical and chemical disintegration of rocks
(ASTM, 1985) or all materials in the surface layer of the Earth's crust that
are loose enough to be moved by a spade or shovel (Winterkorn and Fang,
1975).
Soil may be viewed as a three-phase system composed of solids, liquids,
and gases. The solid phase is composed of inorganic and organic particles of
varying shapes and sizes. The liquid phase is usually an aqueous electrolyte
solution. The gaseous phase is basically air with variations in composition
resulting from biological activity and chemical processes in the soil. Soils
are normally characterized by.the size and composition of their solid (par-
ticulate) components, with air- and water-filled voids considered together as
porosity. However, the relative amounts of air and water (usually expressed
as the degree of saturation) also influence soil behavior.
The term clay can be defined in several ways. Clay can refer to all soil
particles less than a given size, usually 2 urn (Mitchell, 1976). In soils,
this particle size range is composed of clay minerals and other components.
Clay minerals give a clay soil its plastic and cohesive properties. Other
components of the clay size fraction include nonclay minerals (e.g., marl and
chalk), amorphous material, and organic material.
Geotechnical engineers use the term clay to refer to soils that contain
enough clay size particles to affect their behavior CHoltz and Kovacs,
1981). Because of its emphasis on the engineering properties of clay as a
liner material, this document uses the term clay to refer to clay soil, with
clay mineral being used when referring to mineralogy and clay size being used
when referring to particle size.
This chapter discusses the characteristics of clay soils and the clay
minerals that are important soil constituents. It is intended to give the
reader a brief overview of these materials, to present some basic defini-
tions, and to discuss the properties of clay soils and clay minerals that
influence the performance of clay liners. The formation and occurrence of
low-permeability, clay-rich soils are also discussed. For a more thorough
discussion of clay mineralogy or of the geotechnical behavior of soil, the
reader is referred to Grim (1962, 1968), Mitchell (1976), and Perloff and
Baron (1976).
2-1
-------
2.1 CLAY MINERALS
Clay minerals are hydrous silicates, largely of aluminum, magnesium, and
iron, that, on heating, lose adsorbed and constitutional water and yield
refractory material at high temperatures. Plasticity is characteristic of
clay minerals and is largely due to an affinity of the clay surface for
water, resulting from a net negative charge on the surface of a clay particle
that causes it to adsorb water and other polar fluids. This net negative
surface charge results from defects in the clay mineral crystal structure and
from surface chemical reactions, as described below. Because of their
electrochemical surface activity and high surface area (resulting from their
small size and lamellar shape), clay minerals can profoundly affect aisoil's
engineering behavior, even when present in small quantities. As the clay
content of a soil increases, the influence of the clay fraction on its
behavior also increases. The strong influence of clay minerals on soil
behavior can be illustrated by the addition of bentonite to a granular soil.
Bentonite 1s a clay material composed largely of the clay mineral sodium
montmorillonite. An addition of only 2 to 3 weight percent bentonite can
reduce a soil's permeability when it is compacted by 2 to 3 orders of
magnitude (Kozicki and Heenan, 1983). The influence of clay minerals on a
soil's behavior increases with increasing clay content to a range of 33 to
50 percent, at which point the nonclay size material is essentially floating
1n a clay matrix and has little effect on the engineering behavior of the
soil (Mitchell, 1976; Holtz and Kovacs, 1981).
2.1.1 Clay Mineral Structure
Most clay minerals have a sheet-like layered crystalline structure and
thus fall Into the phyllosilicate mineral family. Exceptions are the clay
minerals attapulgite, palygorsite, and sepiolite, which have structures com-
posed of double chains of silica tetrahedra. These minerals are not common
1n soils and are not discussed further in this document.
Clay mineral sheet structures consist of two different layer types, one
composed of tetrahedral units and the other of octahedral units, that are
arranged 1n different sequences to form the different clay minerals. The
tetrahedral unit 1s composed of silica tetrahedra in which four oxygens sur-
round a silicon atom in tetrahedral coordination (Figure 2-1). The octahe-
dral sheet, which is made up of cations octahedrally'coordinated with oxygen
(Figure 2-2), occurs 1n two forms. If the cation is trivalent, only two-
thirds of the possible spaces in a layer are filled and the structure Is
dloctahedral. The most commonly occurring dioctahedral sheet in clay min-
erals 1s the gibbsite sheet, in which the cations are aluminum. If the
cation 1n the octahedral sheet is divalent, all of the available cationic
spaces are filled and the structure is termed trioctahedral. The most
commonly occurring trioctahedral sheet in clay minerals is the brucite sheet,
in which the cations are magnesium.
Isomorphous substitution, or the substitution of different, similar-size
cations for those present in the ideal crystal structure without a change in
structure, 1s common in clay minerals and 1s an important factor in their
behavior. Common cation replacement 1n clay minerals Includes aluminum
(A1+3) for silicon (S1+4), magnesium (Mg+2) for aluminum (Al+3)} and
ferrous iron (Fe+z) for magnesium (Mg+2) in the ideal tetrahedral and
2-2
-------
and
Oxygens
Q and • " Silicons
Si
Si
(c)
(a) Silica tetrahedron.
(b) Silica tetrahedral sheet.
(c) Schematic of silica sheet. See Table 2-1.
Source: Lambe, 1958; Grim, 1968
Figure 2-1. Clay mineral tetrahedral sheet structure.
O and O * Hydroxyls or £ Aluminums, magnesiums, etc.
oxygens
Al
Al
(c)
(a) Octahedron.
(b) Octahedral sheet.
(c) Schematic of gibbsite octahedral sheet. See Table 2-1.
Source: Lambe, 1958; Grim, 1963
Figure 2-2. Clay mineral octahedral sheet structure.
2-3
-------
octahedral sheets described above. Isomorphous substitution in clay minerals
results in a charge deficiency in the crystal structure and a net negative
charge on the mineral's surface.
The variety of cation substitutions, both in and between the crystalline
sheets, and the intergrowth of layers of different character results in the
diversity of actual clay minerals. However, despite the large number of clay
minerals, only a few are important soil-forming constituents. Table 2-1,
(from Mitchell, 1976) summarizes important chemical and physical
characteristics of the clay minerals that commonly occur in soils.
2.1.2 Clay Mineral Groups
This section describes the basic structural makeup of common clay min-
eral groups. Although each clay mineral has a definite "ideal" structure,
many naturally occurring clays are.complex and do not fit the ideal formulas
described herein. Mixed-layer clays can occur, with crystals containing
structural units of more than one clay mineral group. In addition, soils
composed of a single clay mineral or clay mineral group are relatively rare;
multimlneral soils are more commonly encountered in most areas. In a study
of 137 soils across the United States, more than one clay mineral occurred in
about 70 percent of them (Lambe and Martin, 1953-1957). Therefore, it is not
possible to predict soil behavior accurately by assuming that only one clay
mineral predominates through soil material. For more information on clay
minerals and their complex natures, the reader is referred to Grim (1968) and
Van Olphen (1963).
All clay minerals, except those with chain structure, may be roughly
categorized into four groups based on the height of the unit cell, the com-
position of the sheets, and the kind of intersheet bonding that forms the
layers of unit cells. These groups are kaolinite, smectite, illite (or
mica-like), and chlorite. This grouping is convenient since members of the
same group have comparable engineering behavior (Mitchell, 1976). The
following subsections describe these groups and the mineral characteristics
that are Important determinants of the engineering behavior of clay.
2.1.2.1 ICaolinite Minerals--
The basic structural unit (unit cell) of the kaoJInite group is a 1:1
arrangement of a silica tetrahedral sheet and an alumina (gibbsite) octahe-
dral sheet (Figure 2-3). In the tetrahedral sheet, the tips of the silica
tetrahedra all point toward the center of the unit. The oxygen atoms at the
tips of the tetrahedra are common with one of the planes of oxygens in the
octahedral sheet and compose two-thirds of the octahedral oxygens. The
remaining positions In this plane are occupied by hydroxyls that are located
directly below each hexagonal hole in the network formed by the bases of
the silica tetrahedral (Figure 2-3). The kaolinite basal spacing
is 7.2 A*.
Minerals 1n the kaolinite group are composed of stacks of the 1:1
structural units (unit cells) described above. These unit cells are held
together by van der Waals forces and hydrogen bonding between the tetrahedral
sheets and the octahedral sheets of adjacent unit cells. These bonds are
strong enough to preclude the introduction of water between the unit cells
and thus any Interlayer swelling.
2-4
-------
TABLE 2-1. CLAY MINERAL CHARACTERISTICS3
Type
Subgroup and
schematic''
Ml neral
Ideal formula/unit cellc
Cations
octahedral/tetrahedral
1:1 Kaolinite Kaolinite
Halloysite
(dehydrated)
(hydrated)
(OH)8
{OH)8 Si4Al4Oio
(OH)8
Al4/Si4
Al4/Si4
M
2:1 Illite
Smecti te
mite
(KiH20)2(Si)8(A1,Mg,Fe)4>6 020(OH)4 (Al ,Mg,Fe)4.6/(Al,Si)8
Vermiculite (OH)4(Mg.Ca)x(Si8_x.Alx)(Mg.Fe)6 02o.YH20 {Mg,Fe)6/(Si,Al)8
x = 1-1.4, y = 8
Montmorillbnite (OH)4Si8(Al3.34.Mg.66)02o.nH2od
Na.66
Al3.34Mg.66/Si8
2:1:1 Chlorite
Chlorite
(OH)4(SiAl)8(Mg.Fe)6 02o(2:l layer)
(MgAl)e(OH)i2 (interlayer)
(2:1 layer)/(Si,A1)8
(Mg,Al)6 interlayer
See notes at end of table.
(continued)
-------
TABLE 2-1 (continued)
INJ
0>
Mineral
Kaolinlte
Halloysite
(dehydrated)
(hydra ted)
Illite
Vermiculite
Montmorillonite
Chlorite
Isomorphous
substitution
Little
Little
Si always replaced
by some Al. Balanced
by K between layers
Al for Si; net
charge of 1 to
1.4/unit cell
Mg for Al ; net
charge always =
0.66-/unit cell
Al for Si (2: I/layer)
Al for Mg (interlayer)
Inter! ayer bond
0-OH
Hydrogen-strong
0-OH
Hydrogen-s.trong
K ions-strong
Weak
0-0;
Very weak
expanding lattice
0-OH
Hydrogen-strong
Basal
spacing
7.2A"
7.2J
10. 1A
M
10.5-14&
9. eft-complete
separation
14^
CECe
(meq/100g)
3-15
5-10
5-40
10-40
100-150
80-150
10-40
Specific
surface (m2/g)
10-20
35-70
65-100
40-80 primary
870 secondary
50-120 primary
700-840 secondary
Liquid
limit
(%)
30-110
35-55
50-70
60-120
100-900
44-47
Plastic
limit
<«)
25-40
30-45
47-60
35-60
50-100
36-40
Shrinkage Activity
Umi t plasticity Index
(%) % < 2pm
25-29 0.5 !
0.1-0.5
15-17 0.5-1
8.5-15 1-7
aAfter Mitchell (1976).
bS indicates silica tetrahedral sheet.
G indicates gibbsite octahedral sheet. *
B indicates brucite octahedral sheet.
K indicates potassium Ions.
0 indicates water layer.
cTwo formula units required per unit cell.
dArrow indicates source of charge deficiency. Equivalent fia listed as balancing cation.
eCation exchange capacity.
-------
(a)
Oxygens
(OH) Hydroxyls
Aluminums
• O Silicons
(b)
10.1 A
\
{C)
Water Molecules
7.2 A
(a) Diagram of kaolinite structure.
(b) Hydrated hailoysite.
(c) Kaolinite or dehydrated hailoysite.
G = Gibbsite sheet.
Source: Grim, 1963; Mitchell, 1976
Figure 2-3. Kaolinite group minerals.
2-7
-------
2.1.2.1.1 Kaoli ni te—Kaoli ni te is the most common mineral in this group
and consists -of stacks of 1:1 unit cells comprised of silica tetrahedral and
gibbsite (Al) octahedral sheets. The stacks generally range from 0.05 to
2 /jm 1n thickness and can attain thicknesses up to 4,000 jum; the stacks can
range from 0.1 to 4 urn laterally. The specific surface area of kaolinite is
on the order of 10 to 20 m2/g of dry clay.
A small net negative charge on kaolinite particles results in a cation
exchange capacity of 3 to 15 meq/100 g. This charge has been attributed to
a small amount of isomorphous substitution in the silica or gibbsite sheets,
replacement of exposed hydroxyl hydrogens by exchangeable cations, broken
bonds around particle edges, or diffuse charges resulting from the large
size and surface accessibility of 0~2 and OH" molecules (Mitchell, 1976;
Winterkorn and Fang, 1975). The stacked crystal structure of kaolinite
results in a blocky form for this clay mineral and a larger size and lower
surface-to-volume ratio than other clay minerals. This low surface area,
combined with the relatively small'negative surface charge, results in
kaolinite being the least electrochemically active and least plastic clay
mineral.
Because of the blocky structure of kaolinite particles, crystal edges
of this mineral group comprise 10 to 20 percent of the total crystal area
(Theng, 1974) and therefore exert a stronger influence on the electro-:
chemical behavior of these minerals than do the crystal edges of smectites
or illites (crystal edges comprise 2 to 3 percent of total crystal area for
montmorillonite; Theng, 1974). Broken bonds on these edges result in
unsatisfied valences that can be satisfied by cation or anion adsorption.
However, unlike the structurally generated negative charges on the platy
crystal surfaces, these charges are affected by the pH of the environment.
Evidence suggests that the edges are positively charged at low pH and
negatively charged at high pH. This results in kaolinite having a low
cation exchange capacity at low pH and higher cation exchange capacity at
high pH (see Section 2.3.1). Kaolinite has a higher anion exchange capacity
than most clay minerals. This may result from the presence of replaceable "
hydroxyl ions on the outside of structural sheets. Kaolinite thus has the
ability to fix certain negative ions (Deer et al., 1966).
Compared with other clay minerals, kaolinite has. a lower affinity for
water, has a lower dispersivity (see Section 2.3), and does not achieve as
low a permeability upon compaction. On the other hand, because it is not as
electrochemically active, its behavior may be less affected by chemicals
than other clay minerals. Thus, a kaolinitic clay liner may have a higher
permeability than liners composed of other clays, but the permeability of a
kaolinitic clay liner may not be as sensitive to changes in moisture content
or to chemical attack.
2.1.2.1.2 Ha Hoy site—Ha Hoy site is another kaolinite group mineral
that 1s a common soil constituent in some areas. This mineral occurs in two
forms: a nonhydrated type with a structural composition similar to Icaolin-
1te and a hydrated form with a single layer of water interposed between unit
2-8
-------
kaolinite layers (Figure 2-3). This layer increases the basal spacing to
10.1 A, compared with 7.2 % for nonhydrated halloysite and kaolinite.
Partially hvdrated halloysite (metahalloysite)-with basal spacing from
7.4 to 7.9 A can also occur. The interlayer water molecules in hydrated
halloysite are believed to be in a rather flat hexagonal network linked
to each other and to adjacent halloysite layers by hydrogen bonding.
The hydrated form of halloysite occurs in cylindrical tubes of over-
lapping kaolinite sheets. The outside diameters of the tubes range from 0.05
to 0.20 ym, with a median value of 0.07 urn, and range in length from a fraction
to several micrometers. The specific surface area of halloysite ranges from
35 to 70 m2/g (Mitchell, 1976).
Because of the interlayer water sheet in hydrated halloysite, inter-
calation (introduction between the unit cells) of chemicals can occur. This
also results in a slightly higher cation exchange capacity for hydrated
halloysite (5 to 40 meq/100 g) than for kaolinite (3 to 15 meq/100 g).
Halloysite also may be more affected by chemicals than kaolinite.
The interlayer water in halloysite is easily removed during drying, and
this dehydration is irreversible. Because of this phenomenon, soil engi-
neering tests on air-dried samples can give different results than those
performed on samples at the original field moisture content. For this
reason, it is especially important that laboratory tests on soils with
appreciable halloysite content be carried out on samples at the original
field moisture content (Holtz and Kovacs, 1981; Hilf, 1975).
2.1.2.2 Illite Minerals—
The illitic clay minerals are also known as the mica-like clay minerals
because of their structural similarity to hydrous micas, mites are com-
posed of mica-like three-layer sandwiches with an octahedral sheet between
two silica tetrahedral sheets (2:1 layers). These "sandwich" layers are in
turn bound together by fixed or exchangeable cations. The specific mineral
species in this group are determined by differences in octahedral sheet com-
position and the type of interlayer cations. Two minerals in this group
common in soils are illite and vermiculite.
2.1.2.2.1 Illite—Illite is an important constituent of clay soils and
has been described by Mitchell (1976) as "perhaps the*most commonly occur-
ring clay mineral found in soils encountered in engineering practice."
Illite has almost the same crystalline structure as muscovite mica. This
structure is comprised of a silica-gibbsite-silica sandwich, with the tips
of the silica tetrahedra p.ointing toward the octahedral gibbsite sheet and
the oxygens at the tips being common with the octahedral sheet (Figure 2-4),,
Isomorphous substitution of aluminum for silicon in the tetrahedral sheet
results in a negative charge at the surface of these layers. This charge is
balanced by potassium, cesium, and ammonium ions between the 2:1 layers;
these ions fit tightly in the 1.32-X-radius holes in the bases of the
silica sheet and as a result are fixed in position and are not exchangeable,,
Illite differs from muscovite in having less isomorphous substitution in the
2-9
-------
tetrahedral sheet, a lower negative surface charge, and a lower amount of
potassium between the layers. The stacking of illite layers is also more
random, and mite occurs with a much smaller particle size than muscovite.
In terms of properties important to clay liner performance, illite lies
between kaolinite and the smectite clay minerals. Although extensive iso-
morphous substitution results in a net negative charge on the c'lay mineral
surface, the fixed potassium cations balance the charges and strongly bond
adjacent 2:1 sheets together. As a result, illite has intermediate
values for surface area (65 to 100 m2/g), cation exhange capacity (10 to
40 meq/100 g), swelling index, and activity. It is also intermediate in
its reaction to chemicals. Because of the strength of the interlayer
potassium bonding, the basal spacing of illite remains at 10 A* when it is
exposed to polar liquids (Mitchell, 1976). The potassium ions effectively
prevent the intercalation of water, organic liquids, and other cations (Deer
et al., 1966).
2.1.2.2.2 Vermiculite—Vermiculite is a fairly common mineral in clay
soils and usually occurs with other clay minerals. Vermiculite has a 2:1
structure with a poorly organized octahedral sheet sandwiched between two
silica tetrahedral sheets (Figure 2-4). The octahedral sheet contains iron
and magnesium ions.
As with illite, isomorphous substitution of aluminum for silicon is
extensive in the tetrahedral sheet, resulting in a net negative charge on the
crystal surface. This positive charge deficiency is larger than that of the
smectite minerals (see Section 2.1.2.4) and is usually balanced by interlayer
layers of divalent cations and water. This larger charge deficiency results
1n Vermiculite having the highest cation exchange capacity of all clay min-
erals (Deer et al., 1966). The most common interlayer cations in vermiculite
are magnesium and, to a lesser extent, calcium.
The amount of water that is intercalated in vermiculite is less vari-
able than that in smectite and usually is limited to two layers of water
molecules. The interlayer spacing is therefore fairly constant for
vermiculite but varies to some extent depending on the cations present
between the layers. Vermiculites can absorb organic liquids between their
layers but take up less than the smectite minerals (Deer et al.} 1966).
The primary specific surface area for vermiculite ranges from 65 to
100 mz/g. This is within the range reported for montmorillonite and, as
with montmorillonlte, the secondary (interlayer) surface area can reach
very high values (870 m2/g) (Mitchell, 1976).
2.1.2.3 Chlorite Minerals—
Chlorite minerals in clay soils are almost always found in association
with other clay minerals. Chlorites are composed of 2:1 layers of silica
tetrahedral sheets surrounding a gibbsite or brucite octahedral sheet, with
another octahedral sheet between the 2:1 mica layers (Figure 2-5). Chlorites
thus may be termed 2:1:1 clay minerals.
Chlorites can have Isomorphous substitution and may be missing a few of
the octahedral sheets between the 2:1 layers. This can result in some
swelling from water uptake between the layers. Chlorites are less active
2-10
-------
Oxygens. (OH) Hydroxyls, Aluminum, ( ) Potassium
O and • Silicons (One-Fourth Reolaced by Aluminums)
(a)
\
/
n
\ X
s
/ \
\ X
£
X \
- Water Molecule
10 to u A
(14 A as
shown)
(b)
(a) Diagram of illite structure
(b) Illite schematic.
(c) Vermiculite schematic
Source: Mitchell, 1976
(C)
G = Gibbsite sheet.
B = Brucite sheet.
Figure 2-4. Illite clay minerals.
2-11
-------
T
(a)
Gibbsite
or brucite
X 3.
S.
(b)
Several
water
Layers
1.4 nm
Several
Water
Layers
1
J.
nH20 layers and exchangeable cations
(~\ Oxygens (OH) Hydroxyls A Aluminum, iron, magnesium
O and 9 Silicon, occasionally aluminum
(a) Schematic diagram of chlorite.
(b) Schematic diagram of montmoriilonite.
(c) Diagram of smectite structure.
Figure 2-5. Chlorite and smectite clay minerals.
2-12
-------
than the smectites, have a cation exchange capacity similar to illite (10 to
40 meq/10 g),- and may be similar to illite in engineering behavior.
2.1.2.4 Smectite Minerals—
The smectite group of clay minerals includes 2:1 minerals whose unit
cell is composed of an octahedral sheet sandwiched between two silica tetra-
hedral sheets. The bonding between these 2:1 layers is by van der Waals
forces and cations that may be present to balance out structural charge
deficiencies in the 2:1 layer. This bonding is weak, and, as a result,
the layers are easily separated by adsorption of water or other polar
liquids. Thus, the interlayer spacing of smectites can vary from 9.6 A
to complete separation, and this results in the high swelling behavior and
high activity of these clay minerals.
The smectite minerals may be divided into two groups, based on the com-
position of the octahedral sheet. .The montmorillonites have a dioctahedral,
aluminum-based (gibbsite) octahedral sheet; the saponites have a triocta-
hedral magnesium-based (brucite) sheet. Only montmorillonite is commonly
found in soils. The saponites are relatively unimportant as soil constit-
uents and are not discussed further in this document.
2.1.2.4.1 Montmori11oni te—Montmori11oni te is a 2:1 clay mineral with a
dioctahedral gibbsite sheet sandwiched between two silica tetrahedral sheets
(Figure 2-5). Extensive substitution of magnesium and other cations for
aluminum and aluminum for silicon results in a charge deficiency of 0.5 to
1.2 (usually 0.66) on the unit cell (Mitchell, 1976). Most of the substitu-
tion in montmorillonite occurs in the octahedral gibbsite sheet, usually one
magnesium for every sixth aluminum. This results in a charge on the mineral
surface that is more diffuse or evenly spread than that of vermiculite, which
has mostly substitution of aluminum for silicon in the outer, tetrahedral
layers (Deer et al., 1966; Winterkorn and Fang, 1975). The charge defi-
ciencies on the montmorillonite unit cells are balanced by exchangeable
cations between the unit cells, and, as a result, montmorillonite exhibits a
high cation exchange capacity (generally 80 to 150 meq/100 g).
The bonding forces between unit cells of montmorillonite are weak, and
water and polar fluids can easily penetrate between the unit cell layers. As
a result, montmorillonite particles are very small and can be dispersed
to sheets of unit cell thickness (10 8) in water (Mitchell, 1976). The
specific surface area of montmorillonite is very high, with a primary
surface area of 50 to 120 m2/g and a secondary surface area (including
interlayer surfaces) of 700 to 840 m2/g. Because of its high specific
surface and tendency to adsorb interlayer water, montmorillonite 1s very
susceptible to swelling and is the most active of the clay minerals.
Montmorillonite is especially sensitive to alteration by chemical attack.
The type of cations occupying the interlayer spaces strongly influ-
ences the behavior of montmorillonite. The most commonly occurring Inter-
layer cation 1s calcium, a divalent cation. Like vermiculite, caldum-
montmorlllonites usually take up two layers of water between the unit cell
layers (Deer et al., 1966). This results in limited swelling to a maxi-
mum interlayer spacing of 19 A* (Theng, 1974). However, when sodium 1s
2-13
-------
the interlayer cation, as occurs in the Wyoming bentonites (see 2.1.2.4.2),
the amount of interlayer water is not so limited and the interlayer spac-
ing can range" from 10 A (oven-dry) to over 50 X (Theng, 1974). This
results in high swelling, which is characteristic of sodium-montmori11onite;
1t can expand to 13.8 times its dry volume when fully hydrated.
2.1.2.4.2 Bentbnite—Bentonite is not a clay mineral. It is a rock
(or clay deposit) composed largely of the clay mineral montmori'llonite. The
swelling and dispersive properties of this mineral give bentonite the ability
to lower the permeability of a soil, even when added in small quantities
(e.g., 1 to 3 percent by weight). The swelling capacity of bentonite depends
on its sodium-montmorillonite content. Low-swelling bentonite has signifi-
cant quantities of calcium-montmori11onite, which, because of limited inter-
layer water uptake, does not swell to the extent of sodium-montrnorillonite.
High-swelling sodium bentonite has a liquid limit of 500 percent or more and
can swell 15 to 20 times in volume. Calcium bentonite will increase in
volume 0 to 5 times when wetted with water; this swelling capacity has been
reported to increase 700 to 1,000 percent by treating calcium bentonites with
a 0.25-percent solution of ^003 (Fisher, 1965).
Bentonite is formed by the weathering of volcanic ash. Environmental
conditions favorable to sodium bentonite formation are semiarid climate and
alkaline soil and groundwater. The type locality for bentonite is Wyoming,
and most sodium bentonite comes from the western United States and Canada
(Hosterman, 1985). Calcium or low-swelling bentonite is mainly obtained from
deposits in the Gulf Coastal Plain formed from weathering of volcanic ash
deposits (Hosterman, 1984). i
2.2 CLAY FORMATION AND OCCURRENCE
This section presents some of the factors that influence the formation
and occurrence of clay soils in the United States. It is intended to help
the reader comprehend the complexity of soil-forming processes and the degree
of heterogeneity and variability that may be expected in naturally occurring
soils that may be used for clay liners. It is not intended to be a complete
treatise on the subject.
2.2.1 Clay Mineral Paragenesis
«
Clay minerals are products of weathering or hydrothermal alteration,
with different minerals resulting from differences in physical-chemical
conditions and differences in parent material (Deer et al., 1966). Clay
minerals may be found in their place of origin or may be transported and
deposited in sediments. In general, acid conditions favor the removal of
cations from the soil and kaolinite formation, and alkaline conditions favor
the formation of other clay minerals, with the predominant type(s) of cation
Influencing the clay mineral species. A brief discussion of clay mineral
formation follows. A more complete discussion may be found in Grim (1968),
Keller (1964), and Weaver and Pollard (1973).
Kaolinite mineral deposits are formed primarily by the weathering of
alkali feldspars and other silicate minerals common to sialic rock types such
as granites and quartz diorite. Kaolinite occurs in residual, hydrothermal,
2-14
-------
and sedimentary deposits (Patterson and Murray, 1984). In the soil environ-
ment, acidic "conditions favor kaolinite formation through the dissolution and
removal of bases from the parent material. Kaolinite formation is favored
where alumina is abundant and silica is scarce. Kaolinites may be formed in
situ but are usually a product of weathering and transport (Deer et al.,
1966). Kaolinite minerals may be found alone or in association with other
clay minerals in soils. Kaolinite soils are prevalent in the southeastern
United States, where they are formed from weathering, erosion, and redeposi-
tion of granitic material (Mason and Berry, 1968) and in other humid regions
with intense chemical weathering and good drainage.
Illites are formed from the degradation of muscovite and from the
alteration of other clay minerals. The formation of illite is favored by
alkaline conditions with high concentrations of alumina and potassium (Deer
et al., 1966); illite can be formed from smectites under these conditions.
Illite is very common in soils and. is almost always found in association with
other clay minerals; its,stability is responsible for its abundance.
Vermiculite in soils is formed from alteration of biotite, muscovite, or
chlorite by weathering; it is usually found in association with other clay
minerals in soils. Chlorite is also commonly found in soils in association
with other clay minerals; it may be formed from the alteration of other clay
minerals (e.g., montmorillonite) in the presence of magnesia or by the
degradation of ferro-magnesian minerals. Alkaline environments rich in iron
and magnesium favor its formation.
In general, the formation of smectites is favored by alkaline conditions
and the presence of calcium and magnesium. However, smectites occur in
association with illite in the prairie soils of the Midwest (pH 4.5 to 5).
Calcium is usually the dominant exchange cation in smectites, except for
deposits in the western United States, where high-sodium-montmorillonite-
content (sodium bentonite) clay deposits are formed from the alteration of
volcanic rocks. Montmorillonite is also found in soils formed by the
weathering of basic igneous rocks; poor drainage conditions promote the
formation of montmorillonite in soils because magnesium is not removed during
weathering (Deer et al., 1966). High calcium and magnesium and low potassium
and sodium are also favorable to montmorillonite formation, as are semiarid
and arid climates (Mitchell, 1976). -
2.2.2 Clay Soil Formation and Occurrence
Clay soils occur throughout the United States in a variety of deposi-
tional environments. The characteristics of a clay soil are influenced by
the parent material of the soil and the soil-forming processes (e.g.,
weathering, deposition). Clay soil may occur as transported soil or residual
soil. Transported clay soil includes soil laid down by water action (fluvial
soil) and soil deposited by ice action (glacial deposits). Residual soil is
soil that is chemically weathered in place from the parent bedrock (USDI,
1974). Each of these soil-forming processes results in soils with different:
characteristics. The processes involved in soil formation are diverse and
multivariate, resulting in a wide range of chemical and physical variations.
2-15
-------
2.2.2.1 Fluvial Soils—
Fine-grained, high-clay-content soils are found in the outer edges of
torrential outwash deposits or alluvial fans (deposits laid down where an
abrupt flattening of stream gradient occurs), especially in humid climates.
Other alluvial (steam laid) deposits can also have high clay contents; sedi-
ments from older, slow-moving, meandering streams have a larger fraction of
fine-grained components (USDI, 1974). Overbank, floodplain deposits laid
down in still waters also tend to be fine-grained.
Individual strata of alluvial deposits can vary considerably, both
vertically and horizontally and in clay content and permeability (Freeze
and Cherry, 1979). A predominantly fine-grained floodplain or valley-fill
deposit should be expected to have many lenses or strata of coarse-grained,
permeable material. Complex vertical and lateral facies changes, from fine-
grained overbank deposits to coarse-grained river channel deposits, are
common in valley-fill and valley terrace deposits. These lateral facies
changes result from changes in a river course and changes in a flow regime
and reflect past and present geomorphic processes and environments. Meander-
Ing stream deposits are especially variable in this regard because of the
continuous changes in river course common to this geomorphic environment.
Thus, there is much heterogeneity in alluvial sediments resulting from
textural variability. Hydraulic conductivities can vary more than 2 or 3
orders of magnitude in these deposits (Freeze and Cherry, 1979). Because of
this heterogeneity, knowledge of fluvial processes can be very helpful when
site investigations are conducted in areas of alluvial soils. Alluvial
deposits are found in river valleys across the United States, in glaciated
regions, and in the large intermountain basins of the Southwest (Heath,
1984). \
Lacustrine deposits are deposited by the still water of lakes. They
are fine-grained, finely stratified deposits with low permeability, high
compressibility, and low shear strength. Lacustrine deposits have flat
surfaces and are surrounded by high ground (USDI, 1974). Near the edges of
lake deposits, alluvial influences can result in lenses of coarse-grained,
permeable material being interstratified with the fine-grained lake sedi-
ments. As with the valley-fill deposits, these heterogeneities can result
in rapid groundwater migration through the permeable layers. Lacustrine
deposits are common in glaciated regions, along the Gulf Coastal Plain, and
1n the Intermountain basins of the West (Freeze and Cherry, 1979).
Because they are laid down in still or slow-moving water, fine-grained,
low-permeability fluvial deposits tend to have a dispersed soil fabric (see
Section 2.4.1 for a discussion of fabric). However, flocculated fabrics also
can result from fluvial deposition. Fluvial soils tend to be anisotropic
with respect to permeability from stratification, with permeability greater
1n the horizontal than in the vertical direction. In one study of alluvial
and lacustrine sediments (Johnson and Morris, 1962), 46 samples had a greater
horizontal than vertical hydraulic conductivity, 11 samples were isotropic,
and 4 samples had greater vertical conductivities. Horizontal conductivities
were 2 to 10 times larger than vertical conductivities in this study.
2-16
-------
2.2.2.2 Glacial Soils--
Glacial deposits are those,that are laid down by the advances and
retreats of the North American continental ice sheets and by smaller mountain
glaciers in the Rocky Mountains, the Sierra Nevadas, and Alaska. Glacial
soils tend to be very heterogeneous because the source material of a single
soil can extend over hundreds of miles. Deposits of the glacier proper are
referred to as glacial-till or morainal deposits and are heterogeneous, well-
graded deposits with particle sizes ranging from clay to large boulders.
They are deposited directly from glacial ice with little or no sorting by
moving water. The rock fragments are generally contained in a matrix of fine
silt and clay, resulting in a low-permeability soil with considerable shear
strength. These are excellent qualities for landfill sites and clay liner
material. For these deposits, the fine clay material can have a flocculated,
random fabric because till is deposited in mass from the wasting ice. Fine-
grained glaciofluvial and glaciolacustrine deposits tend to have a dispersed
fabric because these deposits are fluvial in origin (see preceding discussion
on fluvial soils). However, flocculated fabrics can result from fluvial
deposition.
The heterogeneities common in glacial soils make a detailed site survey
necessary during investigations of a landfill site or borrow pit in glacially
deposited soils. Field permeability tests should be part of any site sur-
veys in glacial soils because of the potential for fractures and stratified
heterogeneities. Knowledge of glacial processes is very helpful in delineat-
ing heterogeneities during a site investigation in glacial soils.
2.2.2.3 Residual Soils-
Residual soils result from the in situ weathering of underlying bedrock
and are always present in the place where they have formed. The character of
a residual soil is determined by the parent bedrock and the type and
degree of weathering to which it has been subjected (USDI, 1974). Residual
soils are rare in glaciated regions because the action of ice removed such
soils during the last glaciation, and not enough time has passed since then
(10,000-12,000 years) for weathering processes to form new soils from rocks
in place. Residual soils generally occur where transported soils or surfi-
cial bedrock are absent (USDI, 1974).
The mineralogy and physical grain morphology for- residual soils are
different from those of other soils. Residual soils are more leached than
glacial or fluvial soils, with lower lime, magnesium, and potash contents and
higher silica, alumina, and iron-oxide contents. Individual mineral grains
in residual soils tend to be more rounded and weathered (decomposed) than in
other soils. The fabric of residual soils can be flocculated, or randomly
oriented, as the mineral grains are formed in place and not deposited with
preferential orientation as are fluvial soils. However, weathering processes
can also result in dispersed fabric, depending on the mineralogy of the soil
and the geochemical environment (see Section 2.4.1.1).
Although most soils have unique compaction curves, for a residual soil.,
compaction curves at a given compactive effort can change, depending on the
soil moisture content at the start of the test. This has been attributed to
the presence of halloysite (which exhibits irreversible drying) and to
particle breakdown during air drying and subsequent compaction (HiIf, 1975),
2-17
-------
Irreversible changes during drying can also occur for other soil types (see
Section 3.8.3.1; Sangrey et al. 1976).
2.3 CLAY CHEMISTRY
2.3.1 Electrical Double-Layer Theory ;
Most clay mineral particles fall into the size range that defines col-
loids. Colloids are particles that are sufficiently small to allow inter-
facial forces to be significant in their behavior. For clay minerals in
soil environments, these interfacial forces are present as a charged solid-
solution interface that gives rise to an electrical or diffuse double layer.
An electrical double layer is composed of a layer of charges fixed on or
within the surface of a solid and a layer of mobile ions of opposite charge,
or counter ions, distributed in the liquid adjacent to the surface. These
counter ions are weakly held close to the surface by electrostatic forces and
can be displaced with respect to the charged, solid surface with movement of
the liquid.
The theory of the diffuse double layer was originally developed by Gouy
(1910) and Chapman (1913), with later modifications by Stern (1924) and
others, to explain colloid behavior. Parks (1975) and Stumm and Morgan
(1970) describe several models of the double layer. A simple model is
illustrated in Figure 2-6. In this case, electrostatic attraction between
the adsorbent (negatively charged clay surface) and adsorbate (positive ions
in solution) results in a layer of cations at the surface (the Stern layer)
and a diffuse layer of cations in solution (the Gouy layer). The cation
concentration in the Gouy layer decreases with distance from the negatively
charged surface. The thickness of the Stern layer is fixed by the hydrated
radius of the adsorbed counter ions; the plane through the center of the
Stern layer counter ions is known as the Stern plane (Figure 2-6,
diagram A). The thickness of the Gouy layer is defined as the distance from
the charged surface at which the cation and anion concentrations reach that
of the bulk solution. From this point, for a negatively charged surface,
cation concentrations increase and anion concentrations decrease as one moves
toward the surface (Figure 2-6, diagram B).
«
The negative surface charge that generates the electric double layer
around clay minerals may be attributed to two causes (Stumm and Morgan,
1970): (1) imperfections or substitutions in the crystal lattice of the clay
minerals (I.e., isomorphous substitution) and (2) charge deficiencies arising
from chemical reactions at the mineral surface. These surface chemical
reactions include those involving broken bonds at particle edges and non-
cleavage surfaces and the interactions of potential-determining ions (H+
or OH~), with hydroxyl groups at the exposed surface of the clay parti-
cle. Isomorphous substitution is the major cause of negative surface charge
for all clay minerals except perhaps the kaolinite group. Surface chemical
reactions contribute about 20 percent of the surface charge for smectites and
may be the major source of surface charge for kaolinites (Mitchell, 1976).
Negative surface charge from structural causes is not affected by the pH
of the solution surrounding the surface in question. However, charges aris-
ing from surface chemical reactions are directly affected by solution pH.
2-18
-------
A.
Solid Surface
Stern Plane
Solution
Gouy layer, charge = 0G incl. charge
in Stern plane
• Surface Charge = Oa
Concentration
of
B. Counter- and
Co-Ions in the
Double Layer
— Concentration
in Solution
Distance from Surfaffe
After Parks, 1975.
Figure 2-6. Electrical double layer.
2-19
-------
Decreasing pH lessens the negative surface charge. For a given solution
and clay mineral, there is a characteristic pH, the pHpzc, at which the
surface charge is zero. At pH values lower than pHpzc, the surface
becomes positive and thus can act as an anion exchanger (Stumm and
Morgan, 1970).
Figure 2-7 compares the effect of solution pH on surface charge, in
terms of electrophoretic mobility (EPM), for kaolinite and fuller's earth, a
clay material usually composed principally of calcium montmorillonite. One
may see that kaolinite is strongly affected by solution pH, with fuller's
earth being much less affected. This supports the contention that most of
the surface charge for kaolinite is due to surface chemical effects. Most of
the total surface charge of the 2:1 minerals (e.g., montmorillonite) results
from defects in the crystal structure inside the minerals and therefore is
less affected by changes in pH.
The thickness of the double layer of clay minerals results from a
balance between the electrostatic attraction force holding the diffuse cation
layer to the surface of the mineral and osmotic forces tending to diffuse
cations away from the mineral surface. The electrostatic attraction force
depends on the mineral properties (i.e., surface charge), while the osmotic
forces depend upon the properties of the solution (including dissolved ions)
surrounding the clay mineral.
A more detailed discussion of these effects may be found in Mitchell
(1976) and in Chapter 4 of this document, which discusses the different
effects of chemicals on clay permeability. Quantitative treatment of the
electrical double layer also is discussed in Chapter 4. However, it is dif-
ficult to apply the electrical double layer equations to predict the behavior
of clay soils because of the complexity of the system. Clay soils are almost
always composed of a variety of clay minerals, each with different double
layer characteristics.
2.3.2 Cation-Exchange Capacity and Cation Affinity
The cation-exchange capacity (CEC) of a clay mineral may be defined
as the excess of cations in the electrical double layer that can be ex-
changed for other cations in the bulk solution (i.e., the shaded area in
Figure 2-6b). The three-layer (2:1) clay minerals exhibit the largest
exchange capacities. Typical CEC ranges for clay minerals are presented in
Table 2-1. Ranges in CEC result from variations in composition and environ-
mental factors; a given clay mineral does not have a single CEC value
(Mitchell, 1976). It should also be noted that because clay soils are rarely
composed of a single clay mineral and because other nonclay substances (e.g.,
sand and rock flour) usually constitute a significant portion of the soil
mass, the cation exchange capacities of bulk soils are lower than those of
pure clay minerals.
Cations differ in their probability of being adsorbed by a colloid.
There is a well-defined affinity series of cations for clay mineral sur-
faces. For cation-exchange equilibria dominated by electrical clipole
2-20
-------
1O
I 0.5
f I o
--J,
1 -CL5)-
10 11 12— pH
for filler's earth
Source: Stumm and Morgan, 1970
Figure 2-7. Effect of solution pH on clay mineral su/face charge (EPM).
2-21
-------
Interactions between the cations and water molecules, the following affinity
series exist: : :
and
Cs+ > K+ > Na+
Ba++ > Sr++ > Ca++ > Mg++ > Be
Thus, affinity increases with decreasing ionic radius for a given cation
valence. This series does not apply to ion exchange sites that have affini-
ties for specific cations (e.g., interlayer illite potassium sites).
As predicted by double layer theory, clay minerals have a greater affin-
ity for bivalent cations than for monovalent cations and this selectivity
decreases with increasing solution.ionic strength (Stumm and Morgan, 1970).
In addition, the selectivity of clay surfaces for different ions in a mixed
ion system is temperature-dependent. Mass action effects (i.e., high con-
centrations of specific cations in solution) can override the affinity series
(Mitchell, 1976).
Cation exchange affects the double layer thickness and this, in turn,
can affect the permeability of clay soil through its effects on particle
arrangements. Replacement of a monovalent cation by a divalent cation
results in a reduction in double layer thickness, lower interparticle repul-
sion and dispersion, increased flocculation, and increased permeability (see
Section 2.4.2). This phenomenon is discussed in Chapter 4. The properties
of clay minerals with higher cation exchange capacities (e.g., montmoril-
lonite) are more affected by cation exchange than those of clay minerals with
low cation exchange capacities (e.g., kaolinite). Cation exchange capacity
increases with increasing mineral surface area and surface charge.
2.3.3 Significance of the Electrical Double Layer to Clay Liners
The thickness of the electrical double layer is an extremely important
determinant of the engineering properties of a clay soil. For clay soils,
increasing clay mineral electrical double layer thickness results in a more
dispersed fabric and lower hydraulic conductivity (see Section 2.3). Reduc-
tion in electrical double layer thickness for clay minerals can occur through
desiccation, cation exchange, and interactions with certain chemicals (see
Chapter 4 for further discussion of clay/chemical interactions)„ Reduction
in double-layer thickness results in a more flocculated soil structure and
increased permeability. Ttie effect of double-layer thickness in a clay soil
on its behavior also depends on the size and specific surface area of the
clay particles in the soil. The behavior of smaller clay minerals, with
higher specific surface area and surface charge (e.g., smectites), is more
affected by changes in double-layer thickness than clay minerals with lower
specific surface areas (e.g., kaolinite). Relative sizes of clay minerals
are illustrated in Figure 2-8.
2-22
-------
(a)
Adsorbed water
Kaolinite crystal
(1000 X 100 nm)
Montmoriilonite
crystal
(100 X 1 nm)
Edge View
Typical
Thickness
(nm)
Typical
Diameter
(nm)
Specific
Surface
(knv>/kg)
Montmoriilonite
100-1000
0.8
(b)
Illite
30
10000
0.08
Chlorite
Kaolinite
30
10000
50-2000 300-4000
0.08
0.015
(a) Relative sizes of hydrated kaolinite and montmorillonite crystals.
(b) Relative sizes of clay minerals.
Source: Lambe, 1958
Figure 2-8. Comparisons of clay mineral sizes
and surface areas.
2-23
-------
2.4 CLAY SOIL FABRIC AND HYDRAULIC CONDUCTIVITY
This section briefly discusses the fabric of fine-grained soils and the
engineering property of soils most relevant to clay liners; i.e., the rela-
tionship of compaction to hydraulic conductivity (permeability). For more
information on these subjects, see Grim (1962, 1968), Mitchell (1976), and
Hilf (1975). A discussion of clay/chemical compatibility, including the
effects of chemicals on soil hydraulic conductivity, may be found in
Chapter 4 of this document.
2.4.1 Soil Porosity and Hydraulic Conductivity
Porosity and hydraulic (or fluid) conductivity are two fundamental soil
properties from the standpoint of fluid transport. The pore size of a soil
is the most important factor influencing the soil's hydraulic conductivity.
Although a statistical relationship exists between porosity and hydraulic
conductivity for relatively uniform porous media (increasing porosity
increases hydraulic conductivity), it is difficult to relate, quantitatively,
porosity and hydraulic conductivity for clay soils (Mitchell, 1976). How-
ever, this relationship can be discussed qualitatively.
The physical structure* of a soil determines its porosity and its
hydraulic conductivity. Two types of porosity may be delineated in soils:
primary and secondary. These types of porosity are controlled by, respec-
tively, the microstructure (fabric) and macrostructure of the soil, as
described below. In addition to primary and secondary porosity, macro-
structural heterogeneities (e.g., sand lenses and sand seams) also influence
the hydraulic conductivity of a soil mass.
2.4.1.1 Soil Microstructure and Primary Porosity--
Primary porosity is the porosity of the soil mass or soil matrix and is
controlled by the microstructure or fabric of the soil, which is influenced
by the particle size distribution and the arrangement of mineral grains.
Soils with a significant percentage of fine materials usually have low
hydraulic conductivities. Most often this fine-grained material is composed
of clay minerals. Some soils have low permeabilities- with only a small
percentage of fines as in the case of admixtures containing bentonite.
*
Clay minerals generally have flat, platy particle shapes. The orienta-
tion of the clay platelets in the soil is one of the most important param-
eters determining effective porosity and hence hydraulic conductivity in
fine-grained soils (Mitchell, 1976). Moreover, the primary cause of micro-
scale anisotropy in fine-grained soils is the orientation of clay particles
1n a dispersed manner (Freeze and Cherry, 1979). Figure 2-9 illustrates some
simple soil fabrics that can occur in clay soils. These are simplified
pictures. Much more complicated particle arrangements usually occur, and a
single soil may have many different zones of fabric.
*Herein, structure refers to the physical arrangement of the soil
constituents. :
2-24
-------
b.
(a) Flocculated clay platelets.
(b) Dispersed clay platelets.
(c) Flocculated clay platelet groups.
(d) Dispersed clay platelet groups.
Source: From Holtz and Kovacs, 1981
d.
Figure 2-9. Clay soil fabrics.
2-25
-------
In fine-grained, clayey soils, fabric can refer to the arrangement of
Individual clay platelets (Figure 2-9, diagrams a and b) or to the !
arrangement of oriented groups of platelets (Figure 2-9, diagrams c and d)
and the relationship of these platelets to larger mineral grains (silt and
sand). A soil's fabric helps determine its pore size. In a flocculated
soil structure (Figure 2-9, diagrams a and c), the clay platelets or
platelet groups tend to be edge-to-face oriented. Soils with this fabric
can be expected to have a larger pore matrix with hydraulic conductivity
that is fairly equal in all directions. In a dispersed soil structure
(Figure 2-9, diagrams b and d), clay platelets or platelet groups tend to be
oriented face to face, resulting in a deposit that is stratified on the
microscale. A dispersed soil structure will be anisotropic with respect to
hydraulic conductivity with higher matrix hydraulic conductivity in the
horizontal direction (parallel to the oriented clay particles) than in the
vertical direction (perpendicular to the oriented clay particles) (Mitchell,
1976). The pore size of a dispersed soil structure is smaller than that of
a flocculated structure. ;
Hydraulic conductivities in soils with dispersed fabrics tend to be
lower than the hydraulic conductivities of similar soils with flocculated
structures. This phenomenon may be responsible for the difference between
the measured hydraulic conductivity of a clay compacted wet of optimum
and the hydraulic conductivity of the same clay compacted dry of optimum.
Although equal density can be achieved in both cases, the hydraulic conduc-
tivity of the soil compacted wet of optimum will be less (vertical direction)
than that of the soil compacted dry of optimum. This difference in hydraulic
conductivity can approach 2 orders of magnitude (Lambe and Whitman, 1979)
and can be attributed to the increased dispersion of clay platelets and the
resulting smaller pores when a soil is compacted at the higher moisture con-
tent (Daniel, 1984). Mercury porosimetry data, reported by Acar and Seals
(1984) for kaolinite compacted to the same dry density, wet and dry of opti-
mum, suggest that higher hydraulic conductivities in dry of optimum samples
result from a bimodal pore size distribution; although the cumulative poros-
ity of the two samples is the same, the dry of optimum sample has both large
(~ 25 urn) and small (~ 5 x 10~2 /am) pore sizes, whereas the wet of optimum
sample has only the small pore size. .
The fabric of an in situ clay soil is largely determined by soll-
formlng processes responsible for its formation, its depositional environ-
ment, and the action of postdepositional processes. A very important factor
influencing a clay soil fabric is the electrochemical environment at the
time of its deposition, including the mineralogy of the clay and the electro-
chemical properties of the depositional medium. These factors Influence the
thickness of the electrical double layer surrounding the clay particles (see
Section 2.3 for a discussion of double-layer theory). Increasing the double-
layer thickness increases the dispersivity of clay particles, thus decreasing
their tendency to flocculate during deposition. Dispersed soil fabrics tend
to occur when a clay with exchangeable sodium 1s deposited in water with a
relatively low electrolyte level (Rogowski and Richie, 1984). In addition,
dispersed, anisotropic fabric may be developed in isotropic soils from post-
depositional shear or compressive forces (Mitchell, 1976). Flocculated
fabric can be created in virtually all depositional environments (Holtz and
Kovacs, 1981) and can result from postdepbsitional changes in the soil
environment (e.g., cation exchange).
2-26
-------
2.4.1.2 Soil Macrostructure and Secondary Porosity-
Secondary porosity is controlled by the macrostructure of the soil and
encompasses the heterogeneities.ofssoil deposits including joints, fissures,
root and animal holes, and other defects in the soil mass (Holtz and Kovacs,
1981). Joints, fissures, and/or holes can occur in any soil as a result of
stresses on the soil (i.e., earthquakes, slumping, and compression), desicca-
tion, dissolution, or the action of vegetation or animals. In compacted
engineered soils, secondary porosity can result from poor quality control of
the compaction process or from desiccation of the liner after compaction.
Secondary porosity affects the hydraulic conductivity of the entire soil
mass and generally overrides the effect of primary porosity. Thus, a son
that is composed of material of low matrix hydraulic conductivity can have a
very high hydraulic conductivity because of secondary porosity.
Many natural clays have joints. Joints can result from postdeposi-
tional expansion and contraction from wetting and drying. Joints and fis-
sures in preconsolidated clays result from shrinkage cracking and unloading.
Cracks can also occur in clays that have a water content consistently above
the shrinkage limit. These cracks have been postulated to result from
syneresis—the formation of closely knit aggregates from the mutual attrac-
tion of clay particles or from the removal of potassium by weathering.
Figure 2-10 shows a fracture in a glacial till with a low matrix permeabil-
ity. Root systems can cause cracks by sucking up water, causing desiccation
cracks to develop in the soil mass, and can leave holes in the soil mass
after rotting away. Figure 2-11 shows a root cast in a till, including a
fracture and a zone of higher hydraulic conductivity material left after
roots have decayed. Griffin et al. (1985) found in a study of a site in
Illinois that hydraulic conductivities measured in the laboratory were much
lower than conductivities measured by three methods in the field; they con-
cluded that small laboratory specimens, whether 'undisturbed' or recompactecl,
were unable to simulate the flow through relatively large joints or partings
present in natural materials.
The ability of fissures or holes to heal in a soil depends largely upon
soil moisture content, soil plasticity, the size of the fissure or hole, and
ambient stress. Wetter, more plastic soils have a greater healing capability
(USDI, 1974). Certain organic chemicals can affect a soil's self-healing
capability by reducing a soil's moisture content and .plasticity.
Zones of permeable material in a soil with low hydraulic conductivity
are generally referred to as lenses and seams and may be composed of silt and
sand. In certain soil deposits, lenses are isolated and discontinuous and
have little influence on the overall hydraulic conductivity of the soil mass
and on potential contaminant migration. However, in other soils, these
lenses may be continuous in one or more directions, in which case they
greatly increase a soil's overall hydraulic conductivity and provide poten-
tial pathways for the rapid migration of contaminants in the groundwater.
Continuous zones of permeable material are usually deposited by moving water
and are common in river floodplain deposits, around the perimeter of lake
deposits, and in glacial outwash deposits. Permeable zones also may be pres-
ent in marine and glacial till deposits. Figure 2-12 depicts a sand seam
outcropping in an otherwise low hydraulic conductivity (K^10~7 cm/s) glacial
till. Leachate from a nearby landfill has migrated through this sand seam and,
by evaporating, has left white salt crystals that outline the sand seam on the
surface of the cut.
2-27
-------
Source: Photo courtesy of Wisconsin Department of Natural Resources
Figure 2-10. Fractures in glacial till. Note pen for scale (arrow).
2-28
-------
Source: Photo courtesy of Wisconsin Department of Natural Resources
Figure 2-11. Root cast in glacial till. (Root cast is the lighter colored material in the
right center. Note fracture that has developed along root cast (arrow).)
2-29
-------
Source: Photo courtesy of Wisconsin Department of Natural Resources
Figure 2-12. Permeable strata in glacial till deposit. (Permeable zone is the outlined band
across the center of photo. Note white salt crystals on surface of permeable
zone left by evaporating leachate flowing from a nearby landfill.)
2-30
-------
For residual soils, products of chemical weathering of rocks in place,
the primary and secondary porosity-is influenced by the degree of weathering
and the composition and structure of the parent rock. Fractures and fis-
sures in the parent rock may persist in the weathered mass (although not in
highly weathered deposits), affecting the secondary porosity. Hydrothermally
deposited veins of quartz in the original rock can weather into veins of sand
in the resulting residual soil. In addition, for soils derived from strati-
fied rock, the different rock strata can result in a soil with strata of
different hydraulic conductivities.
Anisotropy and heterogeneity are common in all soils. Anisotropy and
heterogeneity can be attributed to the existence and variation of primary
and secondary porosity. On a small scale, theoretical considerations have
led to the attribution of anisotropy to the orientation of clay minerals and
the effect the orientation has on primary porosity. On a larger scale, a
relationship exists between heterogeneity and anisotropy (Freeze and Cherry,
1979). Heterogeneities can be attributed largely to secondary porosity and
stratification in low-permeability soils. Trending heterogeneities, where
transport parameters (e.g., hydraulic conductivity) change gradually in a
given direction, also exist in fine-grained, low-permeability soils.
Trending heterogeneities are most common in soils deposited in deltas,
alluvial fans, and glacial outwash plains (Freeze and Cherry, 1979).
It should be stressed that secondary porosity, if present and inter-
connected, can completely override the primary porosity of a soil mass,
making a deposit of low-permeability matrix material highly permeable. The
prevalence of heterogeneities in natural soil deposits is the primary reason
that hazardous waste containment facilities should be lined regardless of the
hydraulic conductivity of the in situ soil.
Heterogeneities and anisotropies in soils can lead to significant dif-
ferences between field and laboratory permeability measurements (Olson and
Daniel, 1981; Griffin et al., 1985). There are at least three factors
contributing to this difference:
• Most laboratory tests are set up to measure only vertical per-
meability, while field permeability test results are usually a
function of both vertical and horizontal permeabilities. In
anisotropic soils, horizontal permeability can be higher than
vertical permeability, resulting in higher field test results.
• Bias in sample selection is a factor for those soils cohesive
enough to withstand handling so that cracking or breaking is
avoided. This leads to a predominance of measurements of matrix
permeability rather than bulk permeability.
• Samples taken for laboratory permeability tests are smaller than
the area covered by field tests. Field permeability tests are
more likely to include heterogeneities such as fissures and zones
of high-permeability material that can significantly increase soil
permeability.
2-31
-------
Because of these factors, properly performed field permeability measure-
ments can give a more accurate estimate of soil permeability than can lab-
oratory permeability tests that are run on "undisturbed" samples (Griffin
et al., 1985; Day and Daniel, 1985). For soils with heterogeneities that
affect permeability, the larger the field permeameters or the more that are
used for a site investigation, the more accurate will be the measure of the
soil's permeability. These factors should be considered when the protec-
tlveness of potential hazardous waste facility sites is evaluated and during
the measurement of hydraulic conductivity of compacted clay liners.
2.4.2 Soil Structure and Hydraulic Conductivity in Compacted Soils
Compaction is the application of force to a soil to reduce the percent-
age of air-filled voids and thus to increase its density. For cohesive
soils, of which clay liners are composed, compaction with a given type and
amount of compactive effort at various water contents will result in a,com-
paction curve such as the one shown in Figure 2-13. Examination of this
curve shows that, with a given compactive effort, a maximum density is
achieved at a certain water content. This is the optimum water content for
this type and amount of compactive effort. The "S" lines in this figure
represent the computed relationship between water content and dry density at
a constant degree of saturation.
Figure 2-14 illustrates that the moisture/density relationship for a
given soil also depends upon the type and amount of compactive effort. This
figure shows that the line of optimum moisture contents and maximum dry den-
sity for these compactive efforts approximately parallels the line of con-
stant saturation. This figure also shows that the type of compactive effort
influences the shape of the moisture content/density curve (see Chapter 3).
i
Much earthwork compaction is directed toward achieving structural sta-
bility, as with roadbed or foundation compaction. For these applications,
strength, stability, and resistance to changes in volume are the important
parameters. Density is used to control these parameters in the field.
Figures 2-13 and 2-14 illustrate that the same density can be achieved either
wet or dry of optimum. In many cases, foundations and roadbeds are compacted
dry of optimum for greater shear strength because soil shear strength de-
creases with Increasing moisture content (Mitchell, 1976). However, when
cohesive soils are compacted to achieve lower hydraulic conductivity, both
the dry density (reduction of air voids) and the fabric of the soil are
important. Molding water content becomes important when attempts are made to
rearrange the fabric of the clay.
The reduction of hydraulic conductivity during the compacting of a
cohesive soil 1s a function of the following two processes:
• The reduction of the voids in the soil
• The rearrangement of the fabric of a soil to a more dispersed
structure.
2-32
-------
18.5
(-'
5
^-
Z 18.0
c
Q
17.5
i
0.41
0.43
0.46
0.49
0.52
e
a
5
« 8 10 12 14 16 18
WATER CONTENT w(%)
Source: Lambe, 19S5
Figure 2-13. Compaction curve from a standard compaction test.
2-33
-------
19
g 18
2
JC
17
S 16
e
Q
15
I
10 15 | 20
WATER CONTENT (%)
25
No.
1
2
3
4
Layers
5
5
5
3
Blows per
Layers
55
26
12
25
Hammer
Mass
4.54 kg
5.54
4.54
2.50
Hammer
Drop
457 mm (mod. AASHO)
457-
457 (std. AASHO)
305
Source: Lambs and Whitman, 1979
Figure 2-14. Compaction curves for different compactive efforts applied to a silty clay.
2-34
-------
In a dispersed soil structure, the clay mineral platelets are oriented in a
face-to-face fashion. A dispersed structure has smaller pores than a more
flocculated (edge-to-face) soil structure and also has more tortuous paths
for liquid movement through the soil, resulting in a lower hydraulic
conductivity.
Factors that influence the extent to which compaction acts on a cohesive
soil fabric to produce a more dispersed structure include:
• Molding water content
t Dispersivity of the clay minerals in a soil
• Amount of compactive effort
• Type of compactive effort.
Molding water content strongly influences the ease of rearranging clay
particles and particle groups under the compactive effort used (Mitchell
et al., 1965; Carpenter, 1982). The profound effect of molding water content
on soil hydraulic conductivity is shown in Figure 2-15. This figure illus-
trates that, for clay samples compacted to the same dry density at different
water contents, permeability decreases with increasing moisture content, and
this decrease can be more than 3 orders of magnitude (Mitchell, 1976).
Permeability has also been shown to be strongly affected by the amount and
type of compactive effort used to reach the desired density (Mitchell et al.,
1965). Because of these phenomena, clay liner designs should specify that
clay liners be compacted wet of optimum and moisture content, density, and
compactive effort measured and controlled for proper clay liner compaction.
If only density is specified or measured and moisture content and compactive
effort are not carefully controlled, the specified density can be achieved at
too low a moisture content, resulting in a liner with a higher hydraulic
conductivity than required.
Soil dispersivity is affected by the double-layer thickness of the clay
minerals in the soil. Soils composed of minerals with thicker double layers
have higher dispersivities. The dispersivity of clay minerals in a soil
affects its hydraulic conductivity. Figure 2-16 illustrates this phenomenon,
showing that adding a chemical dispersant to a soil reduces hydraulic conduc-
tivity upon compaction. (The chemical dispersant increases the double-layer
thickness of the clay minerals.)
A liner soil must be carefully screened as it arrives onsite to ensure
that the arriving soil type remains the same. If a change in soil type
occurs, it could have a lower dispersivity. If this soil is compacted
according to soil compaction parameters (density and moisture content)
developed for the original soil, a more permeable liner could result. Thus,
if the soil characteristics change during the construction of a clay liner
it is necessary to reestablish density/moisture content/compactive
effort/hydraulic conductivity relationships for the new soil before it is
used as a liner material.
2-35
-------
o
£
1x10
-9
Saturation = 95%
12
>
l€
S i
> ^
a:
a
114
106
14
Kneading
t i
16
13
compaction curve *••...
r
Density of Samples
I
i
20
Line of optimums
13 15 17 19
MOLDING WATER CONTENT (%)
Source: Mitchell, 1976
Figure 2-15. Permeability as a function of molding water content for samples
of silty clay prepared to constant density by kneading compaction.
2-36
-------
1x10'~S I-
o
e
u
I 1*10'
o
> 1x10~7
<
CO
10 11 12 13 1* 15
MOLDING WATER CONTENT
(% dry soil mass)
20.0
* 19.0
Z
> 18.0
0
0=0.1
10 11 12 13 14 15
MOLDING. WATER CONTENT
(% dry soil mass)
Source: Lambe, 1955.
Figure 2-16. The effect of dispersion on hydraulic conductivity. D = 0 is compaction
curve for a soil compacted in its natural state; D = 0.1 is a compaction
curve for a soil compacted with 0.1% (of dry soil weight) of sodium-
polyphosphate.
2-37
-------
The method of compaction also affects the hydraulic conductivity of a
compacted soil liner. Figure 2-17 illustrates that, for a cohesive soil com-
pacted to the" same density, kneading compaction results in a lower hydraulic
conductivity than static compaction. This occurs because kneading action
introduces shear strains to the soil, which tends to realign particles and
results in a more dispersed soil fabric (Mitchell, 1976). This phenomenon
occurs only for cohesive soils compacted wet of optimum because the clay
particles can be rearranged easily only at a high moisture content. This
phenomenon is the reason for a general preference for sheepsfoot rollers for
compacting clay liners; although equivalent densities at a given moisture
content may be obtained with a smooth-wheeled roller, the kneading action of
the sheepsfoot should result in lower hydraulic conductivities for most
soils.
The effect of method of compaction on soil fabric is partly responsible
for the difference observed between field and laboratory compaction methods
(Lambe and Whitman, 1979). Laboratory compaction methods generally result in
permeabilities lower than those attained using field equipment (Dunn and
Mitchell, 1984; Day and Daniel, 1985). This is because compactive efforts
used in laboratory compaction tests are not the same as those produced by
specific field compaction equipment. Thus it is necessary to compact a
test fill in the field and conduct permeability tests on this material to
verify laboratory results (see Section 5.3.3.1.1).
2.5 REFERENCES
Acar, Y.B. and R. K. Seals. 1984. Clay Barrier Technology for Shallow Land
Waste Disposal Facilities. Hazardous Waste. 1(2):167-181.
ASTM. 1985. American Society of Testing Materials Annual Book of ASTM
Standards. Vol. 04.08.
Carpenter, G. W. 1982. Assessment of the Triaxial Falling Head Perme-
ability Testing Technique. Ph.D. dissertation, University of Missouri,
Rolla, Missouri.
Chapman, D. L. 1913. A Contribution to the Theory of Electrocapillarity.
Philosophica Magazine. 25(6):475-481.
«
Daniel, D. E. 1984. Predicting Hydraulic Conductivity of Clay Liners.
Journal of Geotechnical Engineering. 110(2):285-300.
Day, S. R. and D. E. Daniel. 1985. Hydraulic Conductivity of Two Prototype
Clay Liners. ASCE Journal of Geotechnical Engineering. August 1985.
111(8):957-970.
Deer, W. A., R. A. Howie, and J. Zussman. 1966. An Introduction to the
Rock-Forming Minerals. Longman Group Limited. London.
Dunn, R. J. and J. K. Mitchell, 1984. Fluid Conductivity Testing of Fine
Grained Soils. J. of Geotechnical Engineering. 110(11):1648-1665.
Fisher, W. L. 1965. Rock mineral resources of east Texas: Texas University
Bureau of Economic Geology. Report of Investigations 54. 439 pp.
2-38
-------
c
CD
r\>
CO
c o
~ •*
0 3
ll
a o
§ o
21
O 13
-#» QJ
•fl) ^
52. 5'
O 3
•a
a>
CD
iu
er
3.
CO A
I 5
o a
x
3
a
o
o
o
3 « o
» i i
Sol
r- » e
§ a s
o o
-* 3 3
M
in
§
DRY DENSITY
(Ib/ll3)
HYDRAULIC CONDUCTIVITY (cm/aoc)
O 3
3 -
O u.
o
5
m
3)
o
o
z
ni
z
o
O
Z
O
-I
m
3J
O
O
z
m
z
H
-------
Freeze, R. A., and J. A. Cherry. 1979. Groundwater. Prentice-Hall, Inc.
Englewood-Cliffs, New Jersey.
Gouy, G. 1910. Sur la Constitution de la Charge Electrique a la Surface
d'un Electrolyte. Anniue Physique (Serie 4) 9:457-468. Paris.
Griffin, R. A., B. L. Herzog, T. M. Johnson, W. J. Morse, R. E. Hughes,
S. F. J. Chou, and L. R. Follmer. 1985. Mechanisms of Contaminant
Migration Through a Clay Barrier-Case Study, Wilsonville, Illinois.
Eleventh Annual Research Symposium, U.S. Environmental Protection Agency,
Cincinnati, Ohio. EPA/600/9-85/013. , p. 27-38.
Grim, R. E. 1962. Applied Clay Mineralogy. McGraw-Hill. New York.
Grim, R. E. 1968. Clay Mineralogy (second edition). McGraw-Hill.
New York.
Heath, R. C. 1984. Groundwater Regions of the United States. Water
Supply Paper 2242. United States Geological Survey, Reston, Virginia.
Hilf, J. W. 1975. Compacted Fill. Chapter 7 in Foundation Engineering
Handbook. Eds., Winterkorn, H. F., and H-Y. Fang. Van Nostrand Reinhold
Co. New York.
Hosterman, J. W. 1984. Ball Clay and Bentonite Deposits of the Central and
Western Gulf of Mexico Coastal Plain, United States. U.S. Geological
Survey Bulletin 1558-C. USGS; Alexandria, Virginia.
Hosterman, J. W. 1985. Bentonite and fuller's earth resources of the United
States. Mineral Investigation Resources Map: MR-0092. U.S Geological
Survey; Alexandria, Virginia.
Holtz, R. D., and W. D. Kovacs. 1981. An Introduction to Geotechnical
Engineering. Prentice Hall, Inc. Englewood Cliffs, New Jersey.
Johnson, A. T., and D. A. Morris. 1962. Physical and Hydrological
Properties of Water-Bearing Deposits From Core Holes in the Las Barros—
Kettleman City Area, California. USGS, Open File, Department, Denver,
Colorado.
Keller, W. D. 1964. Processes of Origin and Alteration of Clay Mineral.
Eds., Rich, C. I., and G. W. Kunze Soil Clay Mineralogy. University of
North Carolina Press, Chapel Hill.
Kozicki, P., and D. M. Heenan. 1983. Use of Bentonite as a Soil Sealant for
Construction of Underseal Sewage Lagoon Extension, Glenboro, Manitoba.
Shortcourse on Waste Stabilization Ponds, Winnipeg, Manitoba.
Lambe, T. W. and R. T. Martin. 1953-1957. Composition and Engineering
Properties of Soil. Highway Research Board Proceedings. 1-1953,
11-1954, III-1955, IV-1956, V-1957.
Lambe, T. W., and R. V. Whitman. 1979. Soil Mechanics (SI version). John
Wiley and Sons, Inc. New York. , .
2-40
-------
Mason, B., and L. G. Berry. 1968. Elements of Mineralogy. W. H. Freeman
and Co. San Francisco.
Mitchell, J. K. 1976. Fundamentals of Soil Behavior. John Wiley and Sons,
Inc. New York.
Mitchell, J. K., D. R. Hooper, and R. G. Campanella. 1965. Permeability of
Compacted Clay. Journal of the Soil Mechanics and Foundations Divisions,
A.S.C.E. 91(SM4):41-65.
Olson, R. E., and D. E. Daniel. 1981. Measurement of the Hydraulic
Conductivity of Fine-Grained Soils. ASTM STP 746,:18-64. ASTM.
Philadelphia, Pennsylvania.
Parks, G. A. 1975. Adsorption in the Marine Environment. Eds., Riley,
J. P., and G. Skirrow. Chemical Oceanography 1. (second edition)
241-308. Academic Press. New.York.
Patterson, S. H. and H. H. Murray. 1984. Kaolin, Refractory Clay, Ball
Clay, and Hallosite in North America, Hawaii, and the Caribbean Region.
Prof. Paper 1300. U.S. Geological Survey. Alexandria, Virginia.
Perloff, W. H., and W. Baron. 1976. Soil Mechanics. Ronald Press Co.
New York.
Rogowski, A. S., and E. B. Richie. 1984. Relationship of Laboratory and
Field Determined Hydraulic in Compacted Clay Soils. Proceedings of the
16th Mid-Atlantic Industrial Waste Conference. The Pennyslvania State
University, University Park, Pennsylvania.
Sangrey, D. A., D. K. Noonan, and G. S. Webb. 1976. Variation in Atterberg
Limits of Soil Due to Hydration History and Specimen Preparation. Soil
Specimen Preparation for Laboratory Testing, ASTM STP 599, American
Society for Testing and Materials, pp. 158-168.
Stern, 0. 1924. Zur Theorie der Elektrolytischen Doppelschrift.
Zietschrift Electrochem. 30:508-516.
Stumm, W., and J. J. Morgan. 1970. Aquatic Chemistry. John Wiley and Sons,
Inc. New York.
Theng, B. K. G. 1974. The Chemistry of Clay-Organic Reactions. John Wiley
and Sons, Inc. New York.
U.S. Dept. of Interior (USDI). 1974. Earth Manual (second edition). U.S.
Government Printing Office, Washington, D.C.
Van Olphen, H. 1963. An Introduction to Clay Colloid Chemistry. Wiley
Interscience. New York.
Weaver, C. E., and L. D. Pollard. 1973. The Chemistry of Clay Minerals.
Developments in Sedimentology 15. Elesevier, Amsterdam.
2-41
-------
Wlnterkorn, H. F., and H-Y. Fang. 1975. Son Technology and Engineering
Properties of Soils. Chapter 1 In: Foundation Engineering Handbook.
Eds., Wlnterkorn, H. F., and H-Y. Fang. Van Nostrand Reinhold.
New York.
2-42
-------
CHAPTER 3
TEST METHODS AND SOIL PROPERTIES
3.1 INTRODUCTION
The engineering properties of soils are as varied as their chemical
compositions, their modes of formation, and their past and present environ-
mental settings. Consequently, when designing and constructing earthwork
projects, the design or construction engineer is faced with determining
whether or not a particular soil is suitable for its intended application.
To aid this decisionmaking process, a wide assortment of test procedures
have been developed that are used either to predict the probable engineering
performance of a soil (index tests) or to measure the soil performance
directly with respect to parameters of concern (e.g., permeability and shear
strength).
This chapter presents and discusses the soil testing procedures that
are of particular interest to designers of clay liners for hazardous waste
containment facilities. Emphasis is on determining Atterberg limits and
density, compaction, and permeability testing, with the bulk of the chapter
devoted to the latter.
Many test procedures have been standardized by organizations like
the American Society of Testing and Materials (ASTM) and are applied
routinely and uniformly by all competent soil testing groups (ASTM, 1984).
Permeability testing, on the other hand, has not.been standardized.
Different types of permeability testing equipment are in current use, and
test protocols also can vary from laboratory to laboratory and even from
test to test within a laboratory. The situation is further complicated by
the need to measure not only the permeability to water but also, and perhaps
more importantly, the permeability to leachates or other complex chemical
mixtures. In addition, there is much discussion and controversy among soil
experts concerning the interpretation of permeability test results, the
relationship between field and laboratory permeability testing, and the
relative merits of different types of testing equipment and test protocols.
Many soil tests, because of their routine nature and the general
availability of the test protocols, are not discussed in detail in this
chapter. However, capsule summaries of many important tests are provided
in Appendix A. These summaries have been taken directly from the EPA docu-
ment Geotechnical Quality Assurance of Construction of Disposal Facilities
(Spigolon and Kelly, 1984).The summaries are intended only to give the
reader general information about the specific test methods. More detailed
information can be obtained from the references noted for each test. A list
3-1
-------
of the summarized tests contained in Appendix A is provided in Table 3-1.
Other important soil test procedures may be found in the ASTM Standards,
Part 19 - Natural Building Stones; Soil and Rock (ASTM, 1984).
One of the most commonly used pieces of field test equipment is the
nuclear moisture and. density gauge. This device provides a rapid and reli-
able means of making in-place field measurements of the moisture content
and density of soil. However, for accurate measurements in all test circum-
stances, it is necessary to understand the theoretical basis of the tests and
to be mindful of certain caveats Regarding the application of nuclear soil
testing techniques. For these reasons, nuclear methods are described in
detail in Section 3.6.2.
3.2 FUNDAMENTAL RELATIONSHIPS
The presence of liquid and gas-filled voids in soils is an inescapable
consequence of the particulate nature of the material. The multiphase nature
of soils is diagramatically represented in Figure 3-1, which depicts the
fundamental relationships of the weights and volumes of the basic components
of a soil mass. For soil engineering purposes, the important parameters are
unit weight, void ratio, porosity, and degree of saturation. The measure-
ments that are made to compute these parameters are the weight and volume of
the wet specimen, the weight of the same specimen after drying, and the
volume of the solids. Discussion of the appropriate standard procedures for
these measurements can be found in Laboratory Soil Testing, U.S. Department
of the Army (1970); in ASTM (1984); and in Appendix A of this document.
3.2.1 Water Content
A fundamental property of any soil is its water content. This is de-
fined as the ratio, expressed as a percentage, of the weight of water in a
given soil mass to the weight of the solid particles.
3.2.2 Density
The density of a material is its mass per unit volume. As a consequence
of voids being an integral part of any soil, soil may be considered to have
two "densities." One of these densities, called the dry or solid density, is
that of the solid particles. The other density, calfed the total or wet
density, is the density of the total soil mass including water- and
air-filled voids.
3.2.3 Specific Gravity
Specific gravity (of solids), Gs, is defined as the ratio of the density
of solid particles, Ps, at a stated temperature to the density of dis-
tilled water, Pw, at 4° C:
Ps W.
Gs-p-0rGs=HT (3<1)
w s 7w
3-2
-------
TABLE 3-1. SOIL TESTS SUMMARIZED IN APPENDIX A
Method
Method number3
Parameter measured: water content
Standard oven-dry j
Standard nuclear moisture/density gauge 2
Gas burner i 3
Alcohol burning - 4
Calcium carbide (speedy) 5
Microwave oven g
Infrared oven 7
Parameter measured: unit weight
Standard laboratory volumetric 8
Standard laboratory displacement g
Standard field sand-cone in
Standard field rubber balloon 11
Standard field drive-cylinder 12
Standard nuclear moisture/density gauge 13
Parameter measured: ispecific gravity
Standard laboratory 14
Parameter measured: grain-size distribution
Standard sieve analysis (+200 fraction) 15
Amount of soil finer than No. 200 screen (wash) standard 16
Standard laboratory hydrometer (-200 fraction) 17
Pipette method for silt and clay fraction 18
Decantation method for silt and clay fraction 19
Parameter measured: liquid limit
Standard multipoint 20
Standard one poiht 21
Parameter measured: plastic limit
Standard laboratory 22
Parameter measured: cohesive soil consistency
Standard unconfined compression 23
Field expedient unconfined compression 24
Hand penetrometer * 25
Hand-held torvane 26
Parameter measured: water content/density/compactive effort
25 blow standard;Proctor compaction 27
25 blow modified;Proctor compaction 28
Nonstandardized Proctor compaction 29
Rapid, one-point Proctor compaction 30
Rapid, two-point Proctor compaction TI
Hilf's rapid 32
Ohio Highway Department nest of curves 33
Harvard miniature compaction ™
i JH
^
aMethod numbers have been assigned for the sole purpose of providinq
an easy way of finding the test method in the appendix. The number
is meaningless outside of the context of this document.
3-3
-------
Weight
W w
w
We
Air
Water
Volume
i, i
'w
9
Jj
MOIST SOIL
Weight
Volume
W,
W
w
W.
Water
VWVV
V
SATURATED SOIL
Figure 3-1. Schematic representation of soil illustrating the fundamental relationships
among the solid, liquid, and air constituents.
3-4
-------
where
7 = the-unit weight of water (1.00 g/cm3 or 62.43 lb/ft3) at 4°C
w
3.2.4 Unit Weight
Dry unit weight 73 is defined as the weight of the oven-dried
soil solids per total volume of the wet soil mass:
Ws
7d = -f (3.2)
Wet unit weight ym is the total weight (solids and water) per unit
volume of soil:
rm = v (3-3)
Both wet and dry unit weights are usually expressed in pounds per cubic
foot (lb/ft3).
3.2.5 Void Ratio
Void ratio, e, is the ratio of the volume of the voids to the volume of
the solids in a given soil mass:
Vy V - V
e = -v = —r~ (3-4)
s s
where
ws
V_ = volume of solids = 7*-=—
s Gs ^w
3.2.6 Porosity
Total porosity, n, is the ratio, usually expressed as a percentage, of
the total volume of the voids of a given soil mass to the total volume of the
soil mass:
V V - V
n = x 100 = s x 100 (3.5)
3-5
-------
Effective porosity is defined as that fraction of the total volume
through whlch-flow can occur. If all of the pores and void space contrib-
uted to flow, the effective porosity would simply equal the volume of voids
divided by the total volume. However, some pores in soil are isolated or
dead-end pores that do not contribute to flow. Water that is tightly
adsorbed to the wall.of a fluid-conducting pore also decreases the flow area
and effective porosity. The total porosity typically ranges from 0.2 to 0.5
for clays; however, the effective porosity in clay is thought to be on the
order of 0.01 to 0.1 (Zimmie et al., 1981; Daniel, 1981; Horton et al., 1987;
Peyton et al., 1986).
3.2.7 Degree of Saturation
Degree of saturation, S, is the ratio (expressed as a percentage) of the
volume of the water in a soil mass to the total volume of the voids.
S =
w
(3.6)
3.3 ATTERBERG LIMITS
Depending upon the water content, the consistency of a cohesive soil
may range from that of a viscous liquid to that of a very hard solid. This
range of consistencies may be arbitrarily divided into the four stages illu-
strated in Figure 3-2, with the divisions between these states referred to as
the limits. These limits are called the Atterberg limits after the Swedish
soil scientist, A. Atterberg, who in 1911 developed a series of tests for
determining the plasticity of cohesive soils.
To understand the significance of the Atterberg limits, assume that a
very wet fine-grained soil is slowly dried. In the very wet condition, the
soil will act like a viscous liquid. As it dries, a reduction in volume
occurs that is very nearly proportional to the water loss. When the water
content is reduced to the value of the liquid limit, the soil consistency
becomes plastic (i.e., the soil mass can be shaped or deformed without
cracking). The liquid limit is determined experimentally using the device
illustrated in Figure 3-3. The soil sample is placed,in the cup of the
apparatus and a specially designed tool is used to cut through the sample
(Figure 3-4). To perform the test, the crank is turned causing the sample
cup to be tipped up by a cam and then dropped onto the hard surface below it,
Stages of consistency
_A
rSolid
stata
Semisolid
stata
Plastic state
Range indicated by
the plasticity index (PI)
PI - LL-PL
Liquid state
SL PL LL
Sourca: U.S. D«pt of Interior, 1974
Figure 3-2. Consistency limits of cohesive soils.
3-6
-------
Figure 3-3. Device for determining the liquid limits of a cohesive soil. The dish contains a
grooved sample.
Figure 3-4. Clay sample being grooved for liquid limit test.
3-7
-------
thereby supplying a "blow" to the soil sample. The liquid limit is defined
as the water content at which one-half inch of the groove is closed after 25
blows (U.S. Department of Interior, 1974).
In practice, while it is hard to adjust the soil moisture so that the
groove closes at exactly 25 blows, the liquid limit can be determined by
plotting the water contents at which closure occurs with other blow counts
against the log of the number of blows needed to achieve groove closure.
This procedure produces a straight line known as the flow line from which
the water content corresponding to groove closure at 25 blows can be found
easily. This method is described in ASTM method D4318 (ASTM, 1984).
A one-point liquid limit test also can be performed provided that the
practitioner is already familiar with the soils from which the sample is
obtained. The basis for the one-point procedure is the fact that the slopes
of the flow lines for soils within a given soil environment are generally
constant. Therefore, if the flow line slope, the water content, and the
number of blows to obtain closure at that water content are known, the liquid
limit can be computed from the following formula:
LL = WN
('tan 0 \ (3.7)
where
WN » water content of sample
N = number of blows required to !close the groove
tan 0 = slope of the flow line.
The one-point test is usually performed on soil samples prepared to a con-
sistency that requires 20 to 30 blows to close the groove (U.S. Department
of the Army, 1970). A one-point liquid limit test method is described in
ASTM method D4318 (ASTM, 1984).
As the water content 1s reduced below the liquid limit, the soil mass
becomes stiffer and will no longer flow. The soil w1Jl continue to be
plastic (deformable) until the plastic limit is reached. The plastic limit
is defined as the water content, expressed as a percentage of the weight of
the oven-dried soil, where the soil begins to crumble when rolled Into a
thread 1/8 inch in diameter (U.S. Department of the Army, 1970)., The rolling
and crumbling are Illustrated 1n Figures 3-5 and 3-6. The test method for
determining the plastic limit 1s ASTM D 4318 (ASTM, 1984)
As stated earlier, a soil mass will decrease 1n volume 1n proportion to
the amount of water removed. However, 1f enough water is removed, a point
will be reached below which very little shrinkage will occur. The moisture
content at this point, which is referred to as the shrinkage limit, divides
the solid from the semlsolid phases in the soil consistency continuum
(Figure 3-2). Below the shrinkage limit, the soil is considered a solid.
In most fine-grained plastic soils, the plastic limit will be appreciably
greater than the shrinkage limit. However, for coarser grained soils, with
predominantly coarse silt and fine sand sizes, the shrinkage limit will be
3-8
-------
Figure 3-5. Rolling a clay sample for plastic limit test.
* *
"*
Figure 3-6. Results of rolling clay with moisture content below the plastic limit.
3-9
-------
near the plastic limit (U.S. Department of Interior, 1974). The test
procedure for determining the shrinkage limit is ASTM D 427 (ASTM, 1984).
Very often the term plasticity index is encountered. It is the
numerical difference between the liquid limit and the plastic limit and
represents the range of moisture within which the soil remains plastic.
The plasticity index can be used to compute the activity, A, of a
cohesive soil. A is the ratio of the plasticity index to the percentage of
clay-size particles (smaller than 0.002 mm) and is an expression of the
plasticity of the clay fraction of the soil. The mineralogy of the clay
fraction is suggested by the activity, with low activity (less than 1)
characteristic of kaolinite, high activity (greater than 4) characteristic of
montmorillonite, and intermediate activity (1-2) characteristic of illite
(Sowers and Sowers, 1970).
Soils may be grouped and located on a plasticity chart according to
their liquid limits and their plasticity indices (Figure 3-7). The location
of a soil on the chart provides a basis for estimating its engineering
properties based on knowledge of other tested soils that plot in the same
region of the chart. The plasticity chart is roughly divided into quadrants
by the "A" line and the 50-percent liquid limit line. Soils that plot above
the "A" line are classified clays; those below are not clays. The 50-percent
liquid limit line is an arbitrary division between high and low liquid limit
soils. In general, soils plotting above the "A" line have low
permeabilities; for soils with the same liquid limit, the greater the
plasticity index the greater the strength at the plastic limit;
compressibility increases with increasing liquid limit (U.S. Department of
Interior, 1974).
3.4 SOIL CLASSIFICATION
Based upon his early field work, Professor A. Casagrande of Harvard
University developed a soil classification system for grouping soils by a
plot of their plasticity index versus their liquid limit. In 1942, the U.S.
Army Corps of Engineers adopted this system for airfield work. The system
was modified slightly in 1947 to prevent dual classification of soils; in
1952 the U.S. Bureau of Reclamation and the U.S. Army. Corps of Engineers, in
consultation with Dr. Casagrande, adopted the system under the title of the
Unified Soil Classification System (USCS). The USCS is descriptive, easy to
use, adaptable to field or laboratory use, and takes into account the soils'
engineering properties. One of its greatest advantages is that a soil can be
classified readily by visual and manual examination without extensive
laboratory testing. The system is based on the size distribution of the soil
particles and the activity of the very fine grains (U.S. Department of the
Army, 1953; U.S. Department of Interior, 1974).
3.4.1 Grain Size Analysis
Within the USCS, soils are divided into coarse-grained and fine-grained
types. Coarse grains are larger than the openings in the U.S. Standard
Series No. 200 sieve; fine grains can pass through the No. 200 sieve. The
3-10
-------
ISO
160
140
120
x
Ul
a
z
100
80
60
40
20
SANOY CLAY, NORTHEASTERN NEW MEXICO.
SILTY LOESS, KANSAS-NEBRASKA
CLAYEY LOESS,KANSAS.
KAOLIN CLAY, DECOMPOSED GRANITE,SINGAPORE.'
KAOLIN CLAY, RESIDUAL SOIL,CALIFORNIA.
SILTY CLAY, SALT LAKE BASIN SEDI MENTS,UTAH.
PORTERVILLE CLAY "EXPANSIVE" (CALCIUM
BEIOELLITE ) , CALIFORNIA.
HALLOYSITE CLAY, SANOY , TRACE OF
MONTMORILLON.ITE, GUAM .MARIANAS ISLANDS.
TULE LAKE SEDIMENTS (DIATOMACEOUS AND
PUMICE), NORTHERN CALIFORNIA.
MONTMORILLONITE CLAY, S-LI.GHTLY ORGANIC,
GUAM, MARIANAS ISLANDS.
> CLAY.GLACIAL LAKE DEPOSIT, NORTH DAKOTA.
CLAY,"EXPANSIVE" (SODIUM MONTMORILLONITE)
GILA RIVER VALLEY, ARIZONA.
i BENTONITE,WYOMING AND SOUTH
DAKOTA.
200 400 600
LIQUID LIMIT
80 100 120
LIQUID LIMIT
140
160
180
200
Source: U.S. Department of Interior, 1974
Figure 3-7. Typical relationships between the liquid limit
and the plasticity index for various soils.
3-11
-------
particle size distribution of coarse-grained soils can be determined readily
by a sieve analysis, wherein the sample is passed through a set of sieves and
the material retained on each sieve is weighed (U.S. Department of the Army,
1970). Sieve sizes commonly used are listed in Table 3-2.
The size distribution of particles smaller than the No. 200 sieve can
be determined by the hydrometer or by the pipette method based on Stokes1
Law, which relates the terminal velocity of a sphere flowing freely through
a fluid to the sphere's diameter. The relationship is expressed as;
v - 7s = 7F D2 (3.8)
V * 1,800 rj U
where :
V ~ terminal velocity of sphere, cm/s
7 s density of sphere, g/cm3
7 = density of fluid, g/cm3
TJ » viscosity of fluid, g-s/cm3
D » diameter of sphere, mm.
The test assumes that Stokes1 Law can be applied to a mass of dispersed soil
particles of various shapes and sizes. The test is conducted by treating
the soil sample with a chemical dispersant and then mechanically dispersing
the sample in water to form a suspension. Hydrometer readings or pipette
samples at a specific depth in the column are taken every several minutes,
and Stokes1 Law is used to compute the maximum grain size in suspension.
Methods for determining grain size distribution are summarized in
Appendix A, Methods 15-19. Directions for conducting the tests can be
found in U.S. Department of the Army (1970) and in ASTM Method D422-63
(ASTM, 1984).
Grain size distributions are usually displayed graphically as illus-
trated in Figure 3-8, which plots graded and poorly graded distributions.
Well-graded soils have good representation of all particle sizes over a
fairly broad size range. A poorly graded soil may have an excess or
deficiency of particles within a certain size range (gap-graded or skip-
graded), or all the particles occur within a fairly narrow range.
The following coefficients have been defined to determine whether a soil
is well graded or poorly graded:
Dfin (3.9)
Coefficient of uniformity, C = ^
u D
3-12
-------
TABLE 3-2. U.S. STANDARD SIEVE SIZES AND THEIR
CORRESPONDING OPEN DIMENSION
U.S. standard Sieve opening
sieve number (mm)
4 4.75
10 2.00
20 0.85
40 0.425
60 0.25
100 0.15
140 0.106
200 0.075
3-13
-------
CO
I
IOC
90
10
70
$*
>•
•
| 30
I
0 40
£
30
10
0
U S STANDARD SIEVE Of ENING tH INCHES U S STANDARD SIEVE NUMIERS HYDROMETER
A 4 3 1 \'A 1 'A 14 K 3 4 6 • 10 14 16 JO 30 40 30 70 MO 1 40 300
-
,_
_.
-
--
...
—
i rp
.... .
- • — •
:•„..—. .
if
. ,|~
t -
4
It
1 i
i
- + -
_ _i.j .
' i
- t h-
J i
"T"
|
i
1
1
....
i
--
....
!
r
*n
I
r*»»4^ i
\
S
V
Gc
.
M
* *i
S
N
- r S
p-
•*.
S,
Veil _.
aded
- - -
- -
T"
•"•„
...,
00
—
-<
=*>.
—
^
•v*
1 1
\
^
\
X
>
500 100 -50 10 5
GRAIN SUE
COMUS
SAMM NO
envoi oenv
C«AV(l
COAttl
IHl
COAMf |
ClAMtflCAnCN
GRADATION CURVES
\
"!>
'
S
S
S
S
S
s
S
---•
1
S
-
S
---
—
\
X
N
-,-.-
i
L
N
x
S
I
-
.....
y
R
1
.... jl
1
1
r- Pool
Grad
i;
i
"A
s
' ^
y
i
i
0.5
MIUIMETtRS
SAND
MIOHJM | fWI
MAI W
& . u
n
-
Iv
•y
ed
L
•"* *j
v v
ri"
p
i
i
-
....
""«.
:
O.t 005
pi
~-
._
*>s.
^
1
....
.
^f c
^v
^^
001
Gap
traded
^
S.
0.00.
—
**^
-..,
„.
—
=
1
HK
0.(
SHI OR CIA*
0
20
30 £
Q
i
40 *
«
•c
IU
tn I*
o
u
An 5
IU
u
•tf
70 *
100
M\
ta
tOtMGNO
OAH
Figure 3-8. Idealized particle size distribution curves for well-graded, poorly graded, and gap-graded soils.
-------
Coefficient of curvature, C
_
c U10 x U60
where "D" with a numerical subscript represents the size that is larger
than a given percentage of the soil mass (as Indicated by the subscript).
As an example, DIQ = 0.5 mm means that 10 percent of the sample by weight
is smaller than 0.5 mm in diameter. To be well graded, the coefficient
of curvature (Cc) must be between 1 and 3 and, in addition, the coefficient
of uniformity (Cu) must be greater than 4 for gravels and greater than
6 for sands.
3.4.2 The Unified Soil Classification System (USCS)
Except for small editorial changes, the following sections (Sections
3.4.2 through 3.4.2.3.3) are taken. directly from the Earth Manual (U.S.
Department of Interior, 1974).
Soils in nature seldom exist separately as gravel, sand, silt, clay, or
organic matter, but are usually found as mixtures with varying proportions of
these components. The Unified Soil Classification System (USCS) is based on
recognition of the type and predominance of the constituents, considering
grain size, gradation, plasticity, and compressibility. It divides soils
into three major divisions: coarse-grained soils, fine-grained soils, and
highly organic (peaty) soils. In the field, identification is accomplished
by visual examination for the coarse grains and by a few simple hand tests
for the fine-grained soils or fraction, as described in the classification
chart (Figure 3-9 and fully described in Section 3.4.2.3). In the labora-
tory, the grain-size curve and the Atterberg limits can be used in addition
to visual examination. The peaty soils (Pt) are readily identified by color,
odor, spongy feel, and fibrous texture and are not further subdivided in the
classification system.
3.4.2.1 Field Classification—
A representative sample of son (excluding particles larger than 3
inches) is first classified as coarse grained or fine, grained by estimating
whether 50 percent, by weight, of the particles can be seen individually by
the naked eye. Soils containing more than 50 percent. of Individual particles
that can be seen are coarse-grained soils; soils containing more than 50
percent of particles smaller than the eye can individually distinguish are
fine-grained soils. (The No. 200 sieve size [74 /nn] particles are about the
smallest that can be seen individually by the unaided eye.) If the soil is
predominantly coarse grained, it is then identified as being a gravel or
a sand by estimating whether 50 percent or more, by weight, of the coarse
grains are larger or smaller than the No. 4 sieve size (about 1/4 inch).
If the soil is a gravel, it is next identified as being "clean" (con-
taining few or no fines) or "dirty" (containing an appreciable amount of
fines). For clean gravels, final classification is made by estimating the
gradation: the well -graded gravels belong to the GW group, and uniform and
3-15
-------
I
I-*
0>
UNIFIED SOIL CLASSIFICATION
INCLUDING IDENTIFICATIOH AND DESCRIPTION
FIELD IDENTIFICATION PROCEDURES
(deluding parliclis lorgtrjheA J uKMS V4 Dating frtctrtnt ottttunatid vtighlsl
Uort than holt of mettnoU jmoJ]jE than Na 200 sitvt tut. J Mart then tali of mattrwl is larftr than No, zoo swvt sut u
(Tht No. 200 titvt sttt is about tht snalltst oertKte visihtt to tht noktd tyt)
•KAVCLS
Uort than halt of coarst traction
sizt may bt ustd as tquivalmt
SAMDI
Mort than halt of coarst fraction
is smaller than No.4 sitvt silt
(For visual classifications, tht i"
totht NO.4 sitvt Slit,)
CLEAN (RAVELS
(Lift It er no
tinit)
MAVELS WITH
FINES
(Apprtcioblt
cmount of tints)
CLEAN SANDS
(Litllt or no
tints)
SANDS WITH
PINES
(Apprtcioblt
amount of tints)
IDENTIFICATION PROCEO
S SILTS AND CLAYS
Liquid limit
) Itss than 50
SI!
JH
HIGHLY CRSAH1C SOILS
Widl rongt in grotn tut ond iwbitontMj amounts
otMuitirmtdrttt parhcU lilts
fVidommanlly ont Hit or a rangt of Sim
with somt intirmidialt um mining.
Non-plastic finis (for idtntification proctdurts
sn UL btiow)
Plastic finis ((or idinhti cation prociduns
sit CL bllow)
Widt rangt in gram tills ond substantial
amounts of all inttrmtd^alt particlt silts
Prtdomuiantly ont silt or a rongt of silts with
sent inttrmiAatt suts missing.
Non-plastic lints (for idintification procidurts
sit ML bilow)
Plastic twits (for idintification procidurts
IM CL ttlow).
U*ES ON FRACTION SMALLER THAN No. 40 SIEVE SIZE
OUT STMNtTH
KRUSKING
CHAHACTEftlSTICS)
Nont to slight
MiaVum to high
Slight to mtdium
Slight to roidium
High to vtry high
Mi*um to high
OILATANCT
(REACTION
TO SHAKING I
Quick to Stow.
Nont to viry stow
Slow
Slow to noni
Nont
Nona to vary slow
TOiifHNESS
(CONSISTENCY
NCAR PLASTIC LNJITI
Nont
Mtdium
Slight
Slight to radium
High
Slight to mtditun
Rtsdily idtntilitd by cotor. odor, spang? fwf end
trtqutntly by fibrous ttiturt.
GROUP
EYU60LS
11
6W
CP
GU
CC
sw
SP
SM
SC
ML
CL
OL
UH
CH
OH
Pt
TYPICAL NAUES
Will o/odid grtvils, gravit-tond mutu/is,
hltli or no tints
Poorly grodid gronlls, grevil-|0fld miituris,
liltlt or no (mis
Silly gravils, poorly gradid gravil-sond*
silt muturts.
Cloyiy grovils, poorly graatd gravtl-sond-
cloy mixturts.
Will grodtd toads, gravtlly sondi. littlt or
no fints.
no (IMS.
Stlty sands, poorly gradtd sond-silt nuituris.
Clayiy sands.poorly grodtd sand-clay muluris.
Inorganic silts and wiry tint sandf, rock Hour, silty
or clayty fint sands with slight plasticity.
Inorganic clays of low to mi&um plasticity, grovilly
clays, sandy clays, silty clays, lion ctays
Organic silts and organic silt-ctays at tow
plasticity.
Inorganic Silts, micaetous or dtattHnactous tint
Inorgaiuc clays of high plasticity, fat clays.
Organic clays of medium TO high plasticity.
Plot and othtr highly organic soils '
INFORMATION R£QOtft£0 FOA
DESCRIBING SOILS
Gut typical nomt, intfcdt approiunatt
pireintogts of sand andgnh>il,md(
nil, angularity, surfact condition,
ond hardmss of tht coarsi grams t
local or gtotog4C namt and othir
ptrtinint dtscriptivt information.
For undislurtwd I a It odd information
on strotificalift^dia/it ol compact-
niis^iirwntatun, moisturt conditions
SMtysflftd, gravtlly ; about 20% hard,
anguioTgravtl particlts [-n maximum
SHI[ roundtd and subangular sand
grains coarst to tintj about ISX non-
plastic tints with low dry slrtnglh ;
will compacfid and moist tn ploct;
alluvial sand; ISM)
Givi typical nnmt; indicatt dtgrttand
charactir of plasticity, amount and
maximum silt at coarst grains-.color
m wit condition , odor it any, local or
giologk norm, and othtr pirlmint
discriptivt intormottoni and lymbol
in partntntsts.
For undisturbed soils add information
on ttructurt, stratification, consisttncy
in undtsturbtd and rtmoWtd statts,
moisturt onddrainagt conditions.
Clayty silt, brown-, slightly plastic ;
Small ptrctntogt of tint sand;
numtraus vtrtical root holts ; firm
and dry m plact, lots t; (ML)
LABORATORY CLASSIFICATION
CRITERIA
c
0
.1
S
1
1
s
Is
5 ** o>
*z £
!si 1°
IIL-J!
IfiSi
Hi
lii Sp
ii° j
CB - -§^- Crtotw than 4 *
Cc - g^gy Bltwttn one and 3
Not milling all gradalion rtquirt
Atltrsirg limili bllow V lint,
or PI liss than 4
Alttrbtrg limits abovtVlmt
with PI grt«ltr than 7
Cu - -{}£- Grtalir than C
Ce " b^p^; fl'*«Bft ont ond 3
Net muting all gradation rtquirtrm
Atttrfairg limits bllow V lint
or PI Itss than 4
Atttrbirg limits obovt V lint
with PI grtattr Ihm T
s» =
sSS5S^s==
\==== = ^\
LIQUID LIMIT
PLASTICITY CHART
FO* LtKMTONV CLftltlPICkTIM « fINI M»
minis for Cw
Afeovt V (wit with
PI bttwtin4andr
art bgrdtrlint costs
riqujipng uil of dual
Symbols.
nil for SW
Abovt 'A* lint with
PI bttwttn4antf7
art bordirlint casts
rt quiring ust of dual
symbols.
Nil f OIL*
t agyndorjf clotsifications--Soils posstssino choroctirislici of two groups art disignattd by combinations ot group symbols. For txamplt GW-6C, wtll grodtd gravil-sand muture with clay bindtr
• All siivt silts on this chart art us. standard.
ADOPTED Br -CORP!
Source: U.S. Department of Interior, 1974
Figure 3-9. Unified soil classification chart.
-------
• •
skip-graded gravels, the poorly graded gravels, belong to the GP group.
Dirty gravels are of two types: those with nonplastic (silty) fines (GM)
and those with plastic (clayey) fine (GC). The determination of whether
the fines are silty or clayey is made by three manual tests for fine-
grained soils (described in Section 3.5.2.3).
If a soil is a sand, the same steps and criteria are used as for the
gravels to determine whether the soil is a well-graded clean sand (SW),
poorly graded clean sand (SP), sand with silty fines (SM), or sand with
clayey fines (SC).
If a material is predominantly (more than 50 percent by weight) fine-
grained, it is classified into one of six groups (ML, CL, OL, MH, CH, OH)
by estimating its dilatancy (reaction to shaking), dry strength (crushing
characteristics), and toughness (consistency near the plastic limit) and by
identifying it as being organic or inorganic. (The test procedures and
behavior of the various groups of fine-grained soils for each of the hand
tests are shown on the classification chart (Figure 3-8) and are described
in Section 3.4.2.3).
Soils typical of the various groups are readily classified by the
foregoing procedures. Many natural soils, however, will have property
characteristics of two groups because they are close to the borderline
between the groups either in percentages of the various sizes or in
plasticity characteristics. For this substantial number of soils, border-
line classifications are used; i.e., the two group symbols most nearly
describing the soil are connected by a hyphen, such as GW-GC.
If the percentages of gravel and sand sizes in a coarse-grained soil
are nearly equal, the classification procedure is to assume the soil is a
gravel and then continue the classification procedure using the chart until
the final soil group, say GC, is reached. Since it could have been assumed
that the soil is a sand, the correct field classification is GC-SC because
the criteria for the gravel and sand subgroups are identical. Similarly,
within the gravel or sand groupings, borderline classifications such as
GW-GP, GM-GC, GW-GM, SW-SP, SM-SC, and SW-SM can occur.
Proper classification of a soil near the borderline between coarse-
grained and fine-grained soils is accomplished by classifying it first as a
coarse-grained soil and then as a fine-grained soil. Such classifications
as SM-ML and SC-CL are common.
Within the fine-grained division, borderline classifications can occur
between low-liquid-limit soils and high-liquid-limit soils as well as between
silty and clayey materials in the same range of liquid limits. For example,
one may find ML-MH, CL-CH, and OL-OH soils; ML-CL, ML-OL, and CL-OL soils;
and MH-CH, MH-OH, and CH-OH soils.
3.4.2.2 Laboratory Classification--
Although most classifications of soil will be done visually and by
simple hand tests, the USCS has provided for precise delineation of the
3-17
-------
soil groups by gradation analyses and Atterberg limits tests in the labora-
tory. Laboratory classifications are often performed on representative
samples of so'ils which are to be subjected to extensive testing for shear
strength, compressibility, and permeability. They can also be used to
advantage in training the field classifier of soils to improve his ability
to estimate percentages and degrees of plasticity.
The grain-size curve is used to classify the soil as being coarse-
grained or fine-grained and, if coarse-grained, into gravel or sand using
the 50-percent criterion. Within the gravel or sand groupings, soils
containing less than 5 percent finer than the No. 200 sieve size are con-
sidered "clean" and are then classified as well-graded or poorly graded by
their coefficients of uniformity and of curvature. In order for a clean
gravel to be well-graded (GW), it must have both a coefficient of uniformity,
Cu, greater than 4 and a coefficient of curvature, Cc, between 1 and 3;
otherwise, it is classified as a poorly graded gravel (GP). A clean sand
having both Cu greater than 6 and Cc between 1 and 3 is in the SW group;
otherwise, it is a poorly graded sand (SP).
"Dirty" gravels or sands are those containing more than 12 percent
fines, and they are classified as either siHy or clayey by results of
their Atterberg limits tests as plotted on the plasticity chart shown in
Figure 3-8. Silty fines are those that have a PI less than 4 or that plot
below the "A" line. Clayey fines are those that have a PI greater than 7
and that plot above the "A" line. ;
Coarse-grained soils containing between 5 and 12 percent fines are
borderline cases between the clean and dirty gravels or sands (i.e., GS,
GP, SW, SP, GM, GC, SM, and SC). Similarly, borderline cases may occur
in dirty gravels and dirty sands where the PI is between 4 and 7 (GM-GC,
SM-SC). It is theoretically possible, therefore, to have a borderline
case of a borderline case; but this refinement is not used, and the rule
for correct classification is to favor the nonplastic one. For example,
a gravel with 10 percent fines, a Cu of 20, a Cc of 2.0, and a plasticity
index of 6, would be classified GW-GM rather than GW-GC.
If a soil is determined to be fine-grained by using the grain-size
curve, it is further classified into one of the six groups by plotting the
results of Atterberg limits tests on the plasticity otiart, with attention
being given to the organic content. Inorganic, fine-grained soils with
a plasticity index greater than 7 and above the "A" line are CL or CH,
depending on whether their liquid limits are less than 50 percent or more
than 50 percent, respectively. Similarly, inorganic, fine-grained soils with
a plasticity index less than 4 or below the "A" line are ML or MH, depending
on whether their liquid limits are less than or more than 50 percent, respec-
tively. Fine-grained soils that fall above the "A" line but that have a
plasticity index between 4 and 7 are classified as ML-CL.
Soils below the "A" line that are definitely organic are classified as
OL if they have liquid limits less than 50 percent and as OH if the liquid
limits are above 50 percent. Organic silts and clays are usually distin-
guished from inorganic silts, which have the same position on the plasticity
3-18
-------
chart, by odor and by color. Howey_er, when the organic content is doubtful.,
the material can be oven dried, remixed with water, and retested for liquid
limits. The "plasticity of fine-grained organic soils is greatly reduced
on oven drying due to irreversible changes in the organic colloids. Oven
drying also affects the liquid limit of inorganic soils but to a much smaller
degree. A reduction in liquid limit after oven drying to a value less than
three-fourths of the liquid limit before oven drying is considered positive
identification of organic soils.
3.4.2.3 Field Identification Procedures for Fine-Grained Soils or
Fractions—
These procedures are to be performed on the minus No. 40 sieve size
particles (approximately 1/64 inch). For field classification purposes,
screening is not intended; simply remove by hand the coarse particles that
interfere with the tests.
3.4.2.3.1 Dilatancy (Reaction to Shaking)—After removing particles
larger than No. 40 sieve size, prepare a pat of moist soil with a volume of
about one-half cubic inch. Add enough water if necessary to make the soil
soft but not sticky.
Place the pat in the open palm of one hand and shake horizontally,
striking vigorously against the other hand several times. A positive
reaction consists of the appearance of water on the surface of the pat,
which changes to a livery consistency and becomes glossy. When the sample
is squeezed between the fingers, the water and gloss disappear from the
surface, the pat stiffens, and finally it cracks or crumbles. The rapidity
of appearance of water during shaking and of its disappearance during
squeezing assist in identifying the character of the fines in a soil.
Very fine clean sands give the quickest and most distinct reaction,
whereas a plastic clay has no reaction. Inorganic silts, such as a typical
rock flour, show a moderately quick reaction.
3.4.2.3.2 Dry Strength (Crushing Characteristics)—After removing
particles larger than No. 40 sieve size, mold a pat of soil to the
consistency of putty, adding water if necessary. Allow the pat to dry
completely by oven, sun, or air drying, and then test its strength by
breaking and crumbling between the fingers. This strength is a measure
of the character and quantity of the colloidal fraction contained in the
soil. The dry strength increases with increasing plasticity.
High dry strength is characteristic for clays of the CH group. A
typical inorganic silt possesses only very slight dry strength. Silty
fine sands and silts have about the same slight dry strength but can be
distinguished by the feel when powdering the dried specimen. Fine sand
feels gritty, whereas a typical silt has the smooth feel of flour.
3.4.2.3.3 Toughness (Consistency Near Plastic Limit)—After removal
of particles larger than the No. 40 sieve size, a specimen of soil about
3-19
-------
one-half cubic inch in size is molded to :the consistency of putty. If too
dry, water must be added and, if sticky, the specimen should be spread out
1n a thin layer and allowed to lose some moisture by evaporation. Then the
specimen is rolled out by hand on a smooth surface or between the palms into
a thread about one-eighth inch in diameter. The thread is then folded and
repelled repeatedly. During this manipulation the moisture content is
gradually reduced and the speciment stiffens, finally loses its plasticity,
and crumbles when the plastic limit is reached.
After the thread crumbles, the pieces should be lumped together and a
slight kneading action continued until the lump crumbles.
The tougher the thread near the plastic limit and the stiffer the lump
when it finally crumbles, the more potent is the colloidal clay fraction in
the soil. Weakness of the thread at the plastic limit and quick loss of
coherence of the lump below the plastic limit indicate either inorganic clay
of low plasticity or materials such as kaolin-type clays and organic clays
that occur below the A-line.
Highly organic clays have a very weak and spongy feel at the plastic
limit.
3.5 COMPACTION
Soil compaction is the rapid application of mechanical force to a soil
to increase its density. In the construction of clay liners, compaction is
done to decrease the permeability of the liner material. Compactive effort
is the term used to describe the amount of mechanical energy applied to the
soil. The units of compactive effort are foot pounds of energy per cubic
foot of soil (ft-lb f/ft3) or joules per cubic meter (J/m3).
3.5.1 Fundamentals of Compaction
In a 1933 series of publications on soil compaction, R. R. Proctor de-
scribed a laboratory procedure that, in its modern form, has become one of
the standard methods for determining the moisture, dry density, and compac-
tive effort relationship of compacted soils (Proctor, 1933). In the Proctor
compaction test, soil densification is achieved through the application of
a standard dynamic impact (compactive effort). A standardized procedure
has been adopted and described in ASTM test method D698-78 (ASTM, 1984).
The test is performed by placing a layer (lift) of test soil, approximately
2 Inches thick, in a cylindrical compaction mold and dropping a 5.5-pound
weight onto its surface 25 times from a height of 1 foot. This procedure is
repeated for two subsequent lifts, resulting in a compacted sample with three
lifts. The dry density of the compacted sample is then determined. This
procedure is repeated several times on samples of the same soil at different
moisture contents. When the testing is completed, the water contents of the
samples are plotted against the dry densities achieved after compaction.
This exercise produces a compaction curve (also called a moisture density
curve) as illustrated in Figure 3-10.
3-20
-------
19.0
18.0
c
Q
17.5
ine of Optimums
Optimum Water
Content
J L_J
10 12 14
Water Content, w (%)
Source: Lambe, 1955
16 18
0.41
0.43
to
CC
0.46 2
o
0.49
0.52
Figure 3-10. typical soil compaction curve illustrating maximum dry density
and optimum water content.
3-21
-------
The Illustration presents only one curve achieved with one compactive
effort. Changing the compactive effort will produce similar curves with
different maximum dry densities and optimum water contents (optimum water
content is the water content corresponding to the maximum dry density
achieved in the test). Increasing the compactive effort increases the
maximum dry density and decreases the optimum moisture content. A typical
series of compaction- curves for a single soil at different compactive efforts
1s presented in Figure 3-11. Note that changing the compactive effort also
can change the shape of the compaction curve.
The most important feature on the compaction curve is its peak, which
represents the maximum dry density that can be achieved with a given compac-
tive effort. As stated above, the moisture content corresponding to this
peak is called the optimum water content. The line of optimums connects the
maximum densities and optimum water contents of compaction curves produced by
different compactive efforts applied to samples of the same soil. The line
of optimums runs nearly parallel to the zero air voids line (also called the
100-percent saturation line), which represents the saturated water contents
for different dry densities of a given soil.
The compaction curves in Figure 3-11 clearly illustrate that, for a
given soil, compacted density is a function of soil moisture content and
compactive effort. Care must be taken in compaction tests to ensure that
moisture is evenly distributed through the soil mass. Winterkorn and Fang
(1975) point out that:
... to avoid irregular and meaningless moisture density curves
in laboratory and field tests on highly cohesive soils, it is
imperative that the moisture be evenly distributed throughout the
secondary soil aggregates. This may take from 1 to 7 days in the
case of highly cohesive soils to which water is added in the dry
condition.
The Interpretation of compaction test results is not always straight-
forward. Winterkorn and Fang (1975) point out:
. . . even properly performed tests may not yield the generally
expected paraboloid curves. Andrews, et al. (1967) found two
optimum moisture contents for the compaction of ila-bentonlte,
one at 100 percent H20 yielding 73 lb/ft3 and one at 50 per-
cent FLO yielding 65 lb/ft3. Lee and Suedkamp (1972) reported the
existence of four types of moisture-density curves [as shown
in Figure 3-12]. They are (A) single-peak, (B) 1'peak, (C) double
peak, and (D) oddly stiaped with no distinct optimum moisture con-
tent. They related the different types to the liquid limit ranges
as follows: LL<30, double and 1-1/2 peaks; LL 30 to 70, typical
single peak; LL>70, peaked and oddly shaped curves.
These unusual curves are not likely to be encountered with clays suitable
for liners (Mitchell, J. K., 1985, Dept. of Civil Engineering, University
of California, Berkeley, personal communication).
3-22
-------
£
•«••.
Z
JC
•^
•
I-
e
o
19
18
17
Z 16
2
15
14
10 15 20
WATER CONTENT (%)
25
lo.
1
2
3
4
Layers
5
5
5
3
Blows per
Layers
55
26
12
25
Hammer
Mass
4.54kg
5.54
4.54
2.50
Hammer
Drop
457mm (mod.AASHO)
457
457 (std. AASHO)
305
Source: Lambe and Whitman, 1979
Rgure 3-11. Compaction curves for different compactive efforts
applied to a siity ciay.
3-23
-------
Type A
\
\
TypeC
Type B
TypeD
A
After Winterkorn and Fang, 1975
Figure 3-12. Four types of compaction curves found from laboratory investigation.
3-24
-------
3.5.2 Compaction and Permeability
During fhstallation of a clay liner, compaction is controlled by
measuring density and moisture content in each lift. However, these
measurements by themselves are meaningless unless they ultimately can be
related back to the permeability. (See Figures 3-13 and 3-14.)
Mitchell, Hooper, and Campanella (1965) investigated the relationship
between these variables and found that, at a given density, permeability
was very sensitive to variations in the molding water (moisture) content.
They performed a series of compaction tests on a silty clay, samples of
which were kneaded to constant density at different water contents. When
permeability tests were performed, the permeability increased slightly
between 12 and 18 percent molding water content but then decreased by
around 3 orders of magnitude as the water content increased from 18 to
19.5 percent. The decreased permeability corresponded to compaction per-
formed at water contents on the wet side of optimum; minimum permeability
occurred on the "wet-of-optimum" side of the compaction curve and not at the
maximum density. Mitchell et al. (1965) also found that, in the laboratory,,
lower permeabilities are achieved with kneading compaction than with static
compaction. These phenomena are discussed in Chapter 2 of this report.
This discussion gives rise to the following conclusion: to achieve a
specified permeability, soil moisture, compactive effort, and dry density
must be carefully measured and controlled.
3.6 FIELD MEASUREMENT OF DENSITY AND MOISTURE CONTENT
Controlling clay compaction in the field so that a specified per-
meability is achieved is done through control of the moisture content
and density. Methods for measuring moisture content are summarized in
Appendix A, Methods 1 through 7. Field density or unit weight measurements
are summarized in Methods 8 through 13. The following text discusses the
more commonly used techniques in more detail.
3.6.1 Traditional Methods
3.6.1.1 Density--
Two traditional methods are used for measuring density in the field. In
one type of test, a small hole is dug in the compacted fill and the excavated
material is saved and weighed. The volume of the hole is measured by filling
it with sand or liquid with a device that measures the amount of material
required to fill the hole. The sand cone and rubber balloon methods are
examples of this category of test (Appendix A, Methods 10 and 11).
Another technique is to drive a hollow cylinder into the fill, remove a
core, trim it to a known volume, and then determine its weight. This drive-
cylinder method (Appendix A, Method 12) and the sand cone and rubber balloon
methods take time because the sample must be oven dried before the dry den-
sity can be determined.
3-25
-------
0 1x10
§ 1X10-7
Q
Z
O
O
O
^ 1x10*8
ce
a
1x10
-9
r i
12
14
i r r
10 13
i r
co
>
ac
a
114
110
108
Kneading
compaction curva ''••.
\
20
Line of optimums
13 15 17 19
MOLDING, WATER CONTENT (%)
Source: Mitchell, 1975
Figure 3-13. Permeability as a function of molding water content for samples
of silty clay prepared to constant density by kneading compaction.
3-26
-------
1x10
-5
* 5x10
S
o
-6
-6
> 1x10
§ 5X10-7
Q
O 7
3 1x10 7
ff 5x10"8
Q
_ O
Optimum water content
_L
Static compaction
Kneading compaction
I | I ! |
15 17 19 21 23 25 27
MOLDING WATER CONTENT (%)
108 -
>
H-
35 -.
Z ^"9
at c
Q ^
> S
cr
Q
106
104
102
—
° ^-^•°*»^
X^^ V Npv
/ ° °N
V
X
Y*
100 -
I
I
15 17 19 21 23 25 27
MOLDING WATER CONTENT <%)
o Kneading compaction 1' x 2.8" 0 mold
• Kneading compaction 3.5* x 1.4" 0 mold
v Static compaction 1" x 2.8" 0 mold
Source: Mitchell, 1976
Rgure 3-14. Influence of the method of compaction
on the permeability of silty clay.
3-27
-------
3.6.2 Nuclear Methods
Nuclear "gauges offer a faster and more convenient method for measuring
field density and moisture content than those methods described in Section
3.6.1 and are presently widely used for earthwork compaction quality con-
trol. Nuclear gauges are designed to give very rapid measurements of density
and moisture content.
3.6.2.1 Nuclear Density Gauge— :
Nuclear density gauges consist of a source of gamma rays, typically
Cs-137 (0.662 Mev), and a detector (or nest of detectors) separated hori-
zontally from the source. The gauges are used in two modes: a "back-
scattering" mode in which the gamma rays are directed, by collimation,
downward into the soil where they are scattered with some fraction reaching
the detector and a transmission mode in which the source is lowered into a
hole in the soil (2 to 12 inches deep) and the number of gamma rays pene-
trating the soil to reach the detector on the surface is recorded. In
the range of densities of interest (70 to 160 lb/ft3), the gamma rays
that reach the detector decrease with increasing soil density.
Gamma rays interact with the electrons that are a part of the atoms
making up the soil. Two types of interactions are exploited for measuring
density. The primary interaction is Compton scattering wherein the gamma
ray is deflected (scattered) by an electron and continues in a different
direction at a reduced energy. For the elements found in soil, Compton
scattering is the predominant reaction in the energy range above 100 keV.
The second interaction is photoelectric absorption wherein the gamma ray
gives up all its energy to an electron and is removed from the beam. The
frequency of this interaction becomes greater as the gamma ray energy
decreases from 100 keV. The probability of this reaction also increases
with the atomic number of the element. :
The relative frequency of these two interactions is a function of the
soil's chemical composition. The effects can best be illustrated by first
considering the predominant reaction, Compton scattering, and then the
modification introduced by photoelectric absorption.
The Compton scattering cross section, i.e., the probability of a
scattering event, is proportional to the electron den-sity of the medium.
This 1s given by:
De = Dm N = W1 VAi <3-n)
where ;
Dm - mass density of the medium
N - Avogadro's number
V/i - weight fraction of element
3-28
-------
Zj = atomic number of element
A-J = atomic weight of element.
Thus, the electron density is related to the mass density by the factor
Z/A of the medium. The Z/A of the various soil components is given in
Table 3-3. Except for Fe2<33 (Z/A - 0:476), the major soil components
have Z/A values that fall within the range of 0.489 to 0.500. Thus, if
Compton scattering alone were responsible for the attenuation of the gamma
rays, the mass densities would be within a few tenths of a percent of the
instrument gauge indications except for soils containing appreciable amounts
of F6203. Water also has a measurable effect on nuclear density measure-
ments. Because of its high Z/A, its electron density (per gram) is
high. As a result, a dry soil of a given density will produce a slightly
lower nuclear gauge density reading than a wet soil of the same mass den-
sity. The different readings are a consequence of the difference in the Z/A
of the two soils (Table 3-3; clay and clay + 10 percent H20).
Thus, with respect to the Compton scattering component, the gauge will
read iron-rich soils lower than their actual density and damp soils somewhat
higher than their actual density.
The cross section for the photoelectric effect interaction of gamma
rays with electrons is experimentally determined to be about proportional
to Z^/A. Thus, this effect becomes increasingly important for higher Z
elements such as calcium, which has a Z of 20. The other major earth
elements—oxygen, silicon, and aluminum—have relatively low Z values
(8, 14, and 13, respectively), so the photoelectric effect for these is
less than for calcium. The consequence is that more gamma rays are absorbed
per unit of electron density in calcareous soils than in noncalcareous soils
and the nuclear gauge density reading is therefore higher than the actual
density.
To minimize this effect, ASTM Standard 2922 (ASTM, 1984) requires
that nuclear density gauge calibration blocks include two that bracket
the attenuation coefficient of soils and suggests limestone and granite.
The calibration curve is adjusted to fall halfway between the limestone
and granite data. As a result of this compromise, highly calcareous
soils will read 2 to 3 lb/ft3 high and highly siliceous soils will read
2 to 3 lb/ft3 low when measurements are made in the transmission mode.
This effect is exaggerated further in the backscattering mode, where the
photoelectric effect is greater. In the backscattering mode, highly
calcareous soils will read 3 to 4 lb/ft3 high and highly siliceous
soils will read 3 to 4 lb/ft3 low.
Although iron is not a major component of the earth's crust, some soils
contain sufficient iron to affect the reading of a nuclear density gauge.
Biotite mica, other iron silicates, and iron oxides (e.g., goethite and
hematite) are some of the more common iron-containing minerals in soils;
nuclear density measurements in soils containing iron minerals should be made
with care. Because the Z of iron is 26, the photoelectric effect becomes
quite important and the gauge will indicate a density higher than the actual
value when calibrated according to ASTM Standard 2922.
3-29
-------
TABLE 3-3. Z/A OF VARIOUS SOIL COMPONENTS
Components
ZI/AI
Silica (Si02)
Feldspar (KAlSi308)
Lime (CaC03)
Alumina (A1203)
Soda (Na203)
Magnesia (MgO)
Clay (Al6Si2Oi3)
Clay + 10% H20
Fe203
H20
0.500
0.495
0.499
0.490
0.489
0.495
0.496
0.502
0.476
0.555
3-30
-------
A method for overcoming the chemical composition error is to calibrate
the gauge against a series of sand cone tests performed on the soil of
interest. TKe precision of a series of sand cone tests is poorer than the
precision of nuclear gauge tests, but the accuracy is generally better if
the sand cone tests are done carefully. The ratio of the sand cone density
to the nuclear density may then be applied to the nuclear gauge reading to
achieve the corrected density.
When used in the backscattering mode, "surface roughness" may cause
inaccurate density measurements. If the soil surface is not a perfect plane,
air voids are formed between the gauge and the soil, reducing the density of
the volume beneath the gauge. A 50-mil air gap will result in a density
reading that is about 4 lb/ft3 low.
The uniformity of the soil density is also important in the back-
scattering mode. About 60 percent of the gauge response is from the top
inch of soil, 25 percent from the second inch, and 10 percent from the third
inch. Thus, while a gauge may be quoted as having 95 percent of its response
from a 3-inch depth, the response versus depth in this region is in no way
uniform. In summary:
t Nuclear density gauges are factory calibrated to give compromised
measurements in two soil extremes. In highly calcareous dry
soils, the gauge will read high by 2 to 3 lb/ft3 at 120 lb/ft3
in the transmission mode and in highly siliceous soils low by 2
to 3 lb/ft3. In the backscattering mode, the gauges are 3 to
4 lb/ft3 high and low, respectively, at 120 lb/ft3. The gauge
would be expected to read properly in clays with up to 10 to
12 percent water content if no high Z elements such as calcium
and iron are present in appreciable quantities. Nuclear gauges
will read iron-rich soils lower than their actual density and
damp soils somewhat higher than their actual density.
• In the backscattering mode, the nuclear gauge is sensitive to any
density variations in the upper 3 inches of soil and to the flatness
of the soil surface upon which the gauge rests.
• It is customary to calibrate the nuclear gauge against a series of
sand cone tests to overcome chemical composition problems in the
field. A gauge reading (preferably in the transmission mode) is
taken, and then a sand cone test is made on the same volume measured
by the gauge. The average of the ratio of the sand cone test to the
gauge reading for a series of tests then becomes the correction
factor to be applied to the gauge reading. This correction factor
will apply as long as the chemical composition of the soil does not
change appreciably.
3.6.2.2 Nuclear Moisture Gauge--
Nuclear moisture gauges consist of a source of fast neutrons (usually
an americium-beryllium source) and a slow neutron detector mounted as close
to one another as possible. The principle of operation is that of neutron
3-31
-------
moderation, i.e., the slowing down of neutrons caused by elastic scattering
from the nucl-ei in the scattering medium. Hydrogen is an excellent moderator
because its nuclear mass is the same as the mass of a neutron and each inter-
action with hydrogen results in a major energy loss to the neutron. On the
average, about 19 collisions with hydrogen nuclei will thermalize the neu-
tron, i.e., reduce its energy to that of! the surrounding nuclei. The other-
elements in the earth's crust are not nearly as effective—oxygen requires
about 158 collisions on the average and silicon 273. The readings obtained
with a nuclear moisture gauge are essentially independent of the density of
the medium because, except for hydrogen, the medium is nearly transparent to
the neutrons.
By using a slow neutron detector, i.e., one responsive only to thermal
neutrons, one can measure the number of neutrons thermalized, which is a
function of the number of hydrogen atoms in the medium. In inorganic soils,
water provides the bulk of the hydrogen and thus the water content can be
measured. The greater the water content, the greater the number of neutrons
thermalized.
Because of the system's geometry, the gauge is most sensitive to the
hydrogen content in the immediate vicinity of the detector. At low moisture
contents, the neutron moves further from the source-detector position before
it has undergone enough collisions to thermalize it and is far enough from
the detector that the probability of returning to the detector is low. On
the other hand, a high moisture content will permit thermalization much
closer to the detector. This leads to the conclusion that the active volume
of measurement is a function of the moisture content.
Experimentally determined depths of measurement for a typical neutron
gauge give the following expression for the depth where 95 percent of the
detected interactions will occur:
D (inches) = 11 - 0.17 MC (3.12)
where
MC * moisture content in pounds of water per cubic foot of soil.
When Equation (3.12) is applied to a hypothetical case, if a soil
sample has a density of 120 lb/ft3 and moisture content of 12 percent, it
contains 14.4 pounds of water per cubic foot and the sensitive depth is
8.5 inches. Note, however, that if the moisture is nonuniform, the gauge
response will be biased toward the moisture content near the detector.
Another reaction can modify the gauge reading when certain soil elements
are present. Aside from the moderation interactions, certain elements have
a high probability of absorbing a neutron. The only ones normally found in
soil are iron and chlorine, the latter where saline soils exist* Thus,
coastal soils may contain sufficient chlorine to be significant and iron
3-32
-------
oxide or silicate contents of 35 to 40 percent will cause errors. In
summary:
fc. , £-'' ' '• ft* -*
• Neutron gauges are useful for measuring the moisture content of
inorganic soils of uniform water content. Although the sensitive
depth of measurement may normally be 6 to 8 inches, the response
is biased toward the moisture content near the detector.
9 If hydrogen-containing materials (including clay minerals) other than
water are present, the gauge will respond to this hydrogen. Coastal
soils may have sufficient chlorine content to interfere, and high
iron oxide or silicate content will influence the moisture
measurement.
» In order to compensate for the soil compositions that may affect
the neutron response, it is customary to calibrate the gauge against
tests on the oven-dry soil. This is generally done with the soil
samples extracted for the density calibration.
3.6.2.3 Modern Nuclear Density and Moisture Gauges--
Current nuclear gauges contain both density and moisture-measuring
components. A key pad, display unit, and microprocessor enable various
correction factors to be entered into the gauge for particular soils.
Operating manuals provide complete instructions for the use of the gauge
under various conditions. In addition, gauge manufacturers provide training
seminars on the proper use of their gauges, including safety considerations",
3.7 TESTING FOR SHEAR STRENGTH
The shear strength of a son must be known before an earthen structure
can be designed and built with assurance that the slopes will not fail (see
Chapter 5).
A soil mass may be considered to be a compressible skeleton of solid
particles. In saturated soils the void spaces are completely filled with
water; in partially saturated soils the void spaces are filled with both
water and air. Shear stresses are carried only by the skeleton of solid
particles, whereas the normal stress on any plane is carried by both the
solid particles and the pore water. -
The development of shear in a saturated cohesive (clay) soil is com-
plex because an applied load is initially supported by the stress in the
pore water (neutral stress) and is not immediately transmitted to the soil
structure. Because of the very low permeability of clays, neutral stresses
are transmitted to the soil structure very slowly, sometimes requiring months
or years before the soil structure feels the full stress increase. Another
complicating factor is the attractive and repulsive interaction between the
clay particles (Sowers and Sowers, 1971).
The triaxial compression test is commonly used to measure the shear
strength of a soil under controlled drainage conditions. A triaxial com-
pression chamber consists primarily of a head plate and a base plate
separated by a transparent plastic cylinder (Figure 3-15). In the basic
3-33
-------
-Dial Indicator
Pressure Gage
Pressure
Regulator
*"""?
Air Pressure Line
Chamber
Pressure
Reservoir
Source: U.S. Department of the Army, 1970
Rgure 3-15. Schematic diagram of triaxiai compression apparatus for Q test.
3-34
-------
triaxial test, a cylindrical specimen of soil, encased in a rubber membrane,
is placed in the chamber. Connections at the ends of the soil specimen
permit contra!led drainage of pore water. The chamber is filled with water
and pressurized, thereby establishing a confining pressure on the specimen.
The specimen is then loaded axially to the point of failure. The axially
applied stress is called the "deviator stress." In general, a minimum of
three specimens, each under a different confining pressure, are tested to
establish the relation between shear strength and normal stress (U.S.
Department of the Army, 1970).
Three types of triaxial compression tests have been developed to measure
the shear strength of clays. The tests differ in the drainage conditions
established for the sample and shear stress application rate. These tests
are:
« Consolidated-Drained (CD) Test or Slow (S) Test—The desired con-
fining stress is applied to the specimen and drainage is allowed
until 100-percent consolidation is achieved. The rate of applied
shear stress is slow enough that essentially no change in the initial
pore pressure occurs.
« Consolidated-Undrained (CU) Test or Consolidated Quick (R) Test—As
in the CD test above, the derived confining stress is applied to the
specimen and drainage is allowed until 100-percent consolidation is
achieved. No drainage is allowed during the application of shear
stress and the specimen remains at a constant volume and water
content.
9 Unconsolidated-Undrained (UU) Test or Quick (Q) Test—No drainage
is allowed during application of the confining stress or during
application of shear stress. The unconfined compression test is a
special case of the UU test with the confining stress equal to zero.
The principal stress difference at failure is called the unconfined
compressive strength.
Detailed discussion of the protocols for shear strength testing and interpre-
tation of the test results are beyond the scope of this document. The reader
is referred to the U.S. Department of Interior (1974) and U.S. Department of
the Army (1970) for test protocols. Explanations of<*he theory and interpre-
tation of the test can be found in geotechnical engineering texts such as
Lambe and Whitman (1979) and Holtz and Kovacs (1981).
3.8 HYDRAULIC CONDUCTIVITY TESTING
Laboratory and field tests to quantify the hydraulic conductivity
(permeability) of low-porostty soils to water have been used for years.
Historically, information on permeability has been required to evaluate
seepage characteristics of soils in a variety of applications ranging from
earthen dams to septic tank drainage fields. These same test procedures
have been adapted, with some modifications, to evaluate the performance of
compacted clay liners.
3-35
-------
Hydraulic conductivity testing of clay liner material is used for
facility design, for construction quality control (CQC), and for clay/
chemical compatibility evaluation. CQC tests are conducted to ensure
that the hydraulic conductivity of an installed clay liner meets the
performance specification. A compatibility test is conducted to determine
if a particular soil's permeability is altered by a liquid waste with which
it might come in contact in a specific application. Chemical compatibility
is discussed, at length, in Chapter 4.
Theoretical knowledge has not yet reached the stage of providing reli-
able estimates of permeability. Thus,, field and laboratory measurements are
needed. Standard procedures, however, have not been adopted for permeability
testing related to hazardous waste containment.
Effective porosity, pore-size distribution, fluid viscosity, and fluid
density determine the permeability of clay soils to fluids. Any action that
affects the soil's fabric may affect the effective porosity and, in turn,
the permeability. Such modification of the soil fabric might be physical,
as with soil compaction, or chemical, through the various processes that can
occur when a chemical waste interacts with the clay (see Chapter 4).
In general, the procedure for measuring the permeability of a compacted
soil is to enclose the sample tightly in a cylinder (permeameter) and then to
pass the liquid (permeant), usually under pressure, through the sample. The
pressure differential across the sample is expressed in terms of hydraulic
gradient (a dimensionless quantity) that is the head loss across the sample
divided by the sample's height. The gradient can be controlled by super-
imposing air pressure above the permeant supplied to the sample and by
regulating the pressure applied at the effluent end of the column. High
hydraulic gradients reduce testing time by forcing permeant through the
sample at a greater rate than could be achieved with low gradients; however,
the amount of gradient may influence the test results.
3.8.1 Darcy's Law
Hydraulic conductivity (K) or permeability is a measure of how rapidly
a permeant fluid can move through porous soils under a hydraulic gradient.
Quantitatively, K is defined as the constant of proportionality in Darcy's
law (Equation 3.14). To calculate K for a given soil* sample, one must be
able to measure the volumetric flow rate through the sample, the cross-
sectional area of the sample perpendicular to the direction of fluid flow,
and the hydraulic gradient across the sample.
The flow of liquids through clay has been found to obey Darcy's Law,
which states that the volumetric flow rate of water through a porous medium
is proportional to the cross-sectional area of the medium and to the hydrau-
lic gradient impressed across it. The constant of proportionality (K) is
defined as the soil's hydraulic conductivity. This is expressed in
Equation (3-13):
Q = KAh/L (3.13)
3-36 ;
-------
where
Q = volumetric flow rate, em3/s
K = hydraulic conductivity (permeability), cm/s
A = cross-sectional area of specimen, cm2
h = change in hydraulic head (head loss) across the specimen, cm
L - length of sample, cm.
Total or hydraulic head (h) is the combination of the elevation dif-
ference between the in-fTow and the out-flow fluid levels and any applied
pressure or vacuum expressed as an equivalent height of water column.
Note that h/L is the hydraulic gradient. Also note that although K appears
to have the dimensions of a velocity, this is an artifact due to the can-
cellation of units. The true dimensions are cm3/cm2 s (i.e., volume
per unit area per unit time). Darcy's Law assumes a direct proportionality
between the hydraulic gradient and the flow rate.
When fluids other than water are used, the permeability as defined
above will be different if the viscosity and/or density of the fluid differs
from that of water. It is convenient to define an intrinsic permeability
coefficient that considers these two parameters. The equation for intrinsic
permeability is:
(3.14)
where
k = intrinsic permeability, cm2
K = permeability, cm/s
M = dynamic viscosity, g/cm s
p = density, g/cm3
*
g = gravitational constant, cm/s2.
A fluid's viscosity is a measure of its resistance to flow while its
density measures the degree to which gravity influences its flow behavior.
For liquids other than water, these two factors influence the fluid conduc-
tivity in a porous media.
Knowing the hydraulic conductivity of a clay and the hydraulic con-
ditions at the top and bottom of the liner enables one to calculate the
volumetric flow rate (Q) through a saturated, homogeneous liner composed
3-37
-------
of that clay. Given a liner area (A) of 20,000 ft2 (about half an
acre), a liner thickness of 3 ft (L), 1 ft of water over the liner making
h equal 4 (1 ft water + 3 ft of saturated liner), and a permeability
(K) of 10-' cm/s (2.88 x 10"4 ft/d), the volumetric flow rate from
Equation (3.13) will be:
Q = KAh/L i
Q = (2.88 x 10-4 ft/d) (20,000 ft2) (4 ft/3 ft)
= 7.68 ft3/d
= 57 gal/d.
Decreasing the permeability by a factor of 2 will halve the flow rate.
Note that this is the flow rate after equilibrium has been reached and
the liner is fully saturated. A time lapse will occur between the intrusion
of the leachate into the liner and its appearance at the bottom of the
liner.
3.8.2 Hydraulic Gradient
It is customary to conduct the tests with hydraulic gradients that
are substantially greater than those encountered in the field in order to
measure the permeability of compacted clay within a reasonably short time
period. The two implied conditions in the Darcy equation are that the flow
rate is directly proportional to the hydraulic gradient and that a plot of
the relationship between flow rate and hydraulic gradient passes through the
origin. There is no single accepted hydraulic gradient for use in permea-
bility testing. Thus, gradients of 5 to 20 are recommended by some (Zimmie,
1981) while gradients as high as 362 have been used by others (Anderson and
Brown, 1981).
Over the past several decades, several studies have been aimed at evalu-
ating the validity of Darcy's law by measuring the dependence of permeability
on hydraulic gradient. Oakes (1960), Hansbro (1960), Mitchell and Younger
(1967), and others have published data that indicate -a departure from lin-
earity at low hydraulic gradients. Bowles (1979) observes that in clays a
threshold gradient of 2 to 4 may be necessary to produce any flow. The
departure from linearity at low hydraulic gradients may not be unexpected
according to Yong and Warkentin (1975). The binding forces between water
molecules and clay surfaces, in effect, create immobilized hydrodynamic
layers of water surrounding each clay particle. The thickness of these
immobilized hydrodynamic layers depends upon the extent of interaction
between the water and clay surfaces and upon the driving force for flow.
At sufficiently low hydraulic gradients, the "effective" pore diameter
available for flow could be decreased by the immobilized hydrodynamic
layers. Several workers, however, have attributed apparent threshold
gradients to experimental artifact.
The current regulations require that a landfill liner have a leachate
collection system that will ensure that the leachate depth over the liner
3-38
-------
does not exceed 30 cm (1 ft). Thu;s, for liners exceeding 1 foot in thick-
ness, the hydraulic gradient will be less than 2. In view of the larger
hydraulic gradients used in laboratory tests, the question of the linearity
of permeability with hydraulic gradient should be resolved, if meaningful
translation of laboratory results to the field situation is to be expected.
3.8.3 Permeability Measurement Factors That Influence Test Results
Permeability may be determined by either a constant head test or a
falling head test. In a constant head test, as implied by the name, the
test sample ,is subjected to the permeant fluid under a constant head (i.e.,
the hydraulic gradient is constant). In falling head tests, the head of
permeant fluid is allowed to decrease (fluid passes through the sample)
during the timed interval of the test.
Equations (3.15) and (3.16) are used to calculate permeability through
the methods of constant head and falling head, respectively.
For constant head, K = rjr (3.15)
for falling head, K = AT ln F (3.16)
where
K = hydraulic conductivity, cm/s
L - length of soil path across which head is impressed, cm
Q = volumetric flow rate, cm3/s
A = cross-sectional area of sample, cm2
h = hydraulic head, total head
t = time interval over which the sample is collected (or readings
are taken)
a = cross-sectional area of in-flow column
h ,h = height of fluid in in-flow column at beginning of test and at
1 z the end of test.
The total head is actually comprised of three components: velocity,
elevation, and pressure head. Because the seepage velocity in geotechnical
problems is relatively low, the velocity head is insignificant compared to
elevation and pressure heads.
Most permeability test devices may be operated in the laboratory in
either the falling head or constant head mode, depending on the hydraulic
systems connected to the permeameter. Similarly, in the field either a
constant head or a falling head test may be used to determine flow through
the soil. Theoretically, test results (i.e., the value of K that is
3-39
-------
calculated) should be the same whether the measurements are made in a
constant head-or falling head test.
Laboratory permeability tests are conducted to predict performance of
a natural or compacted clay liner in the field. Frequently, however, to
complete a test within a reasonable time frame, laboratory conditions are
established that differ substantially from conditions in the field (e.g.,
elevated hydraulic gradient). An understanding of the many factors that
affect permeability tests is necessary to extrapolate, in a meaningful way,
from test results to performance prediction.
Important factors that influence permeability measurements include
sample characteristics and preparation, permeant properties, design of the
test apparatus, and selection and controlof variables during test perform-
ance. Another important factor that has not been addressed adequately is
interlaboratory variability in permeability tests. Regarding variability in
permeability, Zimmie (1981) has concluded that, "One is always dealing with
orders of magnitude in permeability problems, and it is unrealistic to expect
results to agree within less than several hundred percent." Bryant and
Bodocsi (1986) have also examined precisibn and reliability that can be
achieved in laboratory measurements.
A summary compiled by Olson and Daniel (1979) of potential errors in
laboratory permeability tests is given in Table 3-4. The information is
based on reported permeability test results from several investigations.
Another summary of sources of error in estimating field permeability from
laboratory tests is given in Table 3-5. This table is reproduced from
Daniel (1981). i
3.8.3.1 Sample Selection, Size, and Preparation-
Measuring the permeability of compacted fill is often accomplished by
bringing undisturbed samples into the laboratory. Samples obtained with
Shelby tubes or similar sampling devices are considered "undisturbed,"
although in reality this is not the case. Griffin et al. (1985) and Herzog
and Morse (1984) found that hydraulic conductivities measured in the labora-
tory on Shelby tube samples of a glacial till were lower than conductivities
measured in the field. They concluded that collection of the Shelby tube
samples caused some compaction and closing of naturaljy-occurring cracks and
fissures in the till. Boutwell and Donald (1982) measured hydraulic con-
ductivities on large diameter core samples from a compacted clay liner and
found them similar to conductivities measured in the field. The specifi-
cation of sample selection, sample size, sample preparation, and the number
of samples subjected to testing should be oriented toward adequately
representing field conditions. The number of samples and tests should be
determined by the level of statistical confidence desired combined with the
precision and accuracy of the test method employed.
Large samples tested in situ (see Section 3.8.5) may be of sufficient
size to include some macropores, holes, natural cracks, or sand lenses.
Such samples are more likely than much smaller samples to provide good esti-
mates of actual field performance. However, the time required for such tests
may be prohibitive, particularly if the purpose of the test is construction
quality control. Laboratory tests are performed on smaller samples but have
the advantage of a shorter test time and limited interruption of construction
activities. :
3-40
-------
TABLE 3-4. SUMMARY V>OTENTIAL% ERRORS IN LABORATORY
PERMEABILITY TESTS ON SATURATED SOIL3
Source of error
Direction
of deviation of
measured permeability
from correct value
Ratio of measured
permeability to
correct value
Voids formed in sample
preparation
Smear zone formed
during trimming
Use of distilled water
as a permeant
Air in sample
Growth of microorganisms
Use of excessive hydraulic
gradient
Use of wrong temperature
Ignoring volume change
due to stress change
Flowing water in a direction
other than the one of
highest permeability
Performing laboratory
rather than in situ
tests
High
Low
Low
Low
Low
Low or high
Varies
High
Low
Usually low
0.005 to 0.1
0.1 to 0.5
0.001 to 0.1
<1 to 5
0.5 to 1.5
1 to 20
1 to 40
<0.0001 to 3
aData from Olson and Daniel (1979)
3-41
-------
TABLE 3-5. SUMMARY OF SOURCES OF ERROR IN ESTIMATING FIELD
PERMEABILITY OF COMPACTED CLAY LINERS FROM
LABORATORY TESTS3
Potential
sources of error
Laboratory K too
high or low?
Possible number of
orders of magnitude
of error
Compaction at a higher
water content in
laboratory than in field
Maximum size of clay
chunks smaller in
laboratory than in field
Deleterious substance (e.g.,
plant roots) present in
the field but not in
laboratory samples
Use of more compactive effort
in the laboratory than in
the field, resulting in
optimum water content being
higher in field than in
laboratory
Air in laboratory samples
Use of excessive hydraulic
gradient causing particle
migration
Lack of steady-state seepage
due to stress change
Sample size too small in
laboratory test
Desiccation cracks in field
Low
Low
Low
Low
Low
Low
I
High
i
Low
Low
1 to 3
1 to 2
1 to 3
1 to 3
0 to 1
0 to 1
0 to 1
0 to 3
No data
Reproduced from Daniel (1981).
bA rough estimate based on available data.
3-42
-------
During the design phase of a clay liner construction project, samples of
clay soils are obtained from the site and subjected to a series of laboratory
tests, which include the measurement of their permeability in the compacted
state. The manner in which these samples are handled and the procedures
followed during compaction can greatly influence the test outcome. At pres-
ent, there are no accepted standard protocols for test sample preparation.
A recent study of test methods used by commercial soil laboratories
(Truesdale et al., 1985) revealed that some used protocols requiring air
drying the soil while others required that it be maintained at or near the
field moisture content before clod size reduction. Reasons given for drying
were that it made it easy to break up clods, sieve the soil, and obtain a
very homogeneous soil mass for testing. The problem with drying, however,
is that there is no way of knowing if the dried soil can be rehydrated to
its former condition. Sangrey et al. (1976) found that drying and rewetting
significantly altered the liquid limits of several clays. Typically, re-
hydration took several weeks rather than the 24 hours commonly allowed for
most laboratory tests. More important was their finding that some clays were
irreversibly altered by drying.
Truesdale et al. also found a variety of methods for reducing clod size
and obtaining a uniform representative sample for compaction. When dry soil
was used, the clod size reduction methods included grinding the clods in an
electric mill, breaking the clods up with a hammer, and manually crushing
them between a hand-held steel plate and a counter top. Typically, the
ground sample is passed through a No. 4 sieve before being rewet and
compacted.
Wet samples were also handled in a variety of ways. These included
manually breaking the clods, forcing the moist clay through a ,No. 4 sieve,
and mixing and grinding the soil in a Hobart®vegetable shredder.
Several techniques have been developed for compacting clay soil
samples. In general, the soil is compacted in a cylindrical compaction
mold. If a fixed-wall permeation test is to be run, the clay can be
compacted directly in the permeameter, which in this instance serves as
the compaction mold. Alternatively, clay can be compacted in a separate
compaction mold, extruded, and then trimmed to fit th,e permeameter. At
some facilities, a split mold is used.
A variety of methods are available for compacting test samples in the
laboratory. Among these are:
t Static Compaction—A hydraulic or mechanical press is used to
compress a predetermined weight of soil into a mold of known
volume (to the required density).
• Impact Compaction—A drop hammer is used to compact the specimen.
Often the compaction is done according to the ASTM procedure for
determining the moisture density curve. The procedure is often
modified to accommodate various-sized compaction molds.
3-43
-------
• Kneading Compaction—This may be accomplished with a device known as
a Harvard Miniature Compactor. Some investigators think this device
mimics the kneading compaction obtained in the field with sheepsfoot
rollers.
0 Manual Compaction—In some laboratories, soil samples are weighed and
pushed into a mold of known volume with a handheld ram or rod.
Recently, Dunn and Mitchell (1984) reported on a series of permeability
tests in which the only difference was the method of compacting the sample;;.
When samples were compacted by different methods to both 90 and 95 percent of
their maximum dry density, differences in their hydraulic conductivities were
notable. At both dry densities, the static compaction produced samples with
higher hydraulic conductivities than kneading, which was second highest, or
impact, which was lowest. Manual compaction methods were not tested.
The results of Dunn and Mitchell's (1984) study indicated that static
compaction produced replicate samples with the least effort. The higher
hydraulic conductivity obtained with this method correlates better with
field-compacted samples than the other commonly used compaction methods.
Regardless of the compaction method used in the laboratory, the final
sample should not be considered equivalent to the same material compacted
in the field, for several reasons. Field compaction is done with machines
whose size and compactive effort are not duplicated in the laboratory.
Clod sizes, which are very small and fairly well controlled in the labora-
tory, are much more variable in the field, where clod size may range up to
6 or 8 inches or even greater in some cases. Finally, moisture content and
distribution in laboratory samples are controlled much more carefully than
in the field, where the combination of large clods and inadequate curing
(soaking) time after water addition may result in nonuniform moisture dis-
tribution within a liner. The profound effect of moisture content on
permeability is discussed in Chapter 2 (Section 2.4.2).
The'significance of sample diameter has been investigated by Daniel
(1981). His data show that permeability determined on the specimens tested
in the laboratory drastically underestimated the actual permeability of a
clay liner. Test results on different diameter samples of a compacted clay
liner are shown in Table 3-6. The average permeability of the in-place
liner was back-calculated from measured leakage rates and found to be
1 x 10~5 cm/s. Boynton and Daniel (1985) compacted test samples with
various diameters ranging between 1.5 and 6 inches. The measured hydraulic
conductivities showed an increase with sample diameter, with the smallest
diameter having the lowest conductivities. However, the highest and lowest
conductivities differed only by a factor of 2, which the authors did not
consider to be of practical significance.
Anderson and Bouma (1973) experimented with a series of cores of
different lengths to determine the effect of sample size on permeability.
They found that permeabilities for cores ,17 cm in length were lower by
one-half an order of magnitude than for the 5-cm-length samples.
3-44
-------
TABLE 3-6. TEST RESULTS SHOWING EFFECT OF SAMPLE
DIAMETER ON PERMEABILITY MEASUREMENTSa
Sample diameter (cm) Permeability (cm/s)
3.8 1 x ID"7
6.4 8 x ID-9
243.8 3 x ID'5
aData from Daniel (1981).
3-45
-------
Whether a sample is compacted dry of optimum or wet of optimum can
profoundly influence the resulting permeability. Daniel (1981) reported
that the permeabilities of soils compacted dry of optimum might typically
be 10 to 1,000 times greater than permeabilities of the same soils compacted
wet of optimum (this phenomenon is discussed in Section 2.3). Because mois-
ture conditions at compaction can strongly influence permeability measure-
ments, gross errors in predicting field permeability from laboratory tests
may occur if compaction is performed at different water contents. Permea-
bility tests can be conducted with samples compacted at several points in a
range of moisture contents to ensure that actual field moisture conditions
are reproduced.
Another consideration is that equipment used to compact laboratory
samples does not resemble field compaction equipment. Lambe and Whitman
(1979), in comparing field and laboratory compaction efforts, have concluded
that generally the laboratory curves yield a somewhat lower optimum water
content than the actual field optimum.
In the field, a compacted low-permeability soil may display anisotropic
flow characteristics. The horizontal permeability (parallel to the plane
of compaction) of anisotropic soil may be much higher than the vertical
permeability (normal to the plane of compaction). Such a condition will
lead to predominantly horizontal flow. For in-place stratified clays, the
ratio of horizontal to vertical permeabilities may exceed 10 (Olson and
Daniel, 1979).
In contrast, however, Boynton and Daniel (1985) report that no anise-
tropic flow was found in laboratory-prepared samples of fireclay compacted
wet of optimum. Anisotropic flow found in compacted soils may be the result
of poor lift bonding or variations in density within the test sample.
3.8.3.2 Hydraulic Gradient--
Si nee Darcy's law indicates a linear relationship between flow rate and
hydraulic gradient, many workers have used elevated hydraulic gradients to
reduce testing time. However, if hydraulic gradients are excessive, piping
(opening flow channels and increasing hydraulic conductivity) or particle
migration (blocking flow channels and reducing hydraulic conductivity) may
occur and can significantly influence permeability measurement. Although
such effects can occur and have been reported (Daniel, 1981; Mitchell and
Younger, 1967), studies have been conducted at elevated gradients with no
evidence of piping or particle migration (Anderson, 1981).
Excessive hydraulic gradients can result in deviations from Darcy's law,
which is only valid for laminar flow conditions. Even under gradients as
high as 361, however, velocities low enough to assure a Reynolds number
within the laminar flow regime are assured due to the very small particle
diameters 1n fine-grained soils.
Zimmie et al. (1981) have recommended use of hydraulic gradients
between 5 and 20 for laboratory studies. Research performed at Louisiana
State University has lead to the conclusion that tests conducted under
hydraulic gradients as high as 100 are "acceptable for testing, reducing ,
testing times to realistic and practical duration" (Acar and Field, 1982).
3-46
-------
Dunn and Mitchell (1984) reported that increasing the gradient in steps from
20 to 200 caused an irreversible decrease in hydraulic conductivity. They
attributed mast of the observed,,changes to particle migration due to seepage
forces. They recommend that test gradients be kept as low as possible while
still allowing the test to be completed in a reasonable time.
3.8.3.3 Sample Saturation--
A soil sample, even when compacted, has some porosity. The pores are
filled with either gas (generally air) or liquid. Because liquid water
cannot flow through a gas bubble, entrapped air within the interconnected
pores that form flow channels for the permeant fluid causes a reduction in
flow and a corresponding apparent decrease in permeability. Soaking the
sample from the bottom with the top open to the atmosphere may not result in
complete saturation. Smith and Browning (1942) found that in 200 specimens
soaked from the bottom, the degree of saturation averaged 91 percent, with
the lowest value at 78 percent.
The extent of the error in permeability measurement attributable to
entrapped gas bubbles is not fully known, although decreases in permeability
by factors ranging from 2 to 5 have been reported (Johnson, 1954).
The use of backpressure in permeability testing is considered by some
researchers to be necessary for saturating a soil sample. Matyas (1967),
investigating the effect of saturation on permeability, applied a back-
pressure of 5 psi to "saturate a clay specimen." According to his calcula-
tions, this pressure was not sufficient to guarantee complete saturation.
After the saturation attempt at the low pressure, the specimen was subjected
to different hydraulic gradients, and discharge velocities were measured.
Similar samples were saturated at 24 psi and at 70 psi. The plot of
hydraulic gradient versus discharge velocity is shown in Figure 3-16.
Matyas1 data indicate that at 5 psi backpressure the velocity gradient
plot does not pass through the origin and is nonlinear. However, when the
backpressure was raised to either 24 psi or 70 psi, the velocity gradient
relationship was found to be linear and to pass through the origin. The
apparent nonlinear velocity gradient relationship was attributed to the
presence of air in the voids. There was no evidence of a threshold gradient
(i.e., a gradient below which flow would not occur).
*
Matyas (1967) did not address the difference in the velocity gradient
relationships at backpressures of 24 and 70 psi. One might conclude that
the application of higher backpressures caused some particle rearrangement
resulting in lower observed permeability values. The lowest practical
backpressure to obtain saturation appears desirable in order to minimize
these effects. The effects can also be minimized if the backpressures are
increased incrementally and sufficient time is allowed between successive
increases so that high seepage forces do not develop in unsaturated portions
of the specimen. Daniel et al. (1984) reports using backpressures of 40 to
60 psi applied in 10-psi increments. Each increment is held for a period
ranging from several minutes to several hours. Full backpressure is main-
tained for 1 to 5 days before the test is begun.
3-47
-------
3 X 10
2 X 10~5
u
o
o
> 1 x 1(T5
246
Hydraulic Gradient
After Matyas, 1967
Figure 3-16. Effect of backpressure on permeability to water, Sasumua clay.
3-48
-------
3.8.3.4 Permeant Characteristics--
Permeant fluids that are commonly used to determine baseline permea-
bility values-include deionized water, tap water, groundwater (representative
of a specific site) and standard permeant solution, typically 0.01 N calcium
sulfate or calcium chloride. The standard solution is intended to simulate
the hard water frequently used in the field during installation and subse-
quent hydration of a.clay barrier.
While some researchers claim that use of different fluids can influence
baseline permeabilities, others do not believe the effects are significant
(Olson and Daniel, 1981). Recently, the trend has been toward using the
standard calcium solutions for initial saturation and baseline permeability
determinations.
In quality control and in compatibility testing, factors that influence
the flow characteristics of a permeant fluid will alter the apparent soil
permeability to that fluid. Because viscosity and density are temperature-
dependent properties of a fluid, seasonal temperature changes in the field
could influence the range of permeabilities to be anticipated. Large
variations in temperature over a test sample are unlikely, however. Soil
permeabilities are sometimes expressed as intrinsic permeabilities (see
Equation 3.15 in Section 3.8.1) to account for differences in flow properties
between different permeant fluids, or at different temperatures. However,
the use of the permeability coefficient, K, is much more common and is used
throughout the remainder of this section.
The presence of certain chemicals in permeant fluids has been shown to
alter the permeability of clay soils (see Chapter 4). Shrinkage of clay
particles may result from changes in interlayer spacing due to interaction
of chemicals with the clay particle's double layer. Changes in adsorbed
cations or the presence of salts or acids in the permeant fluid can also
effect significant permeability increases. Such effects are the reason for
conducting compatibility tests. Effects of chemical permeants may be evident
immediately after the permeant penetrates the sample, or the effect may not
be seen until one or more pore volumes have passed through the sample. These
effects are discussed fully in Chapter 4.
3.8.3.5 Test Duration—
A number of factors can cause changes in permeabjlity with time. It
is essential in permeability testing that flow through the sample be con-
tinued until stable permeability measurements are obtained. In compatibility
testing, several pore volumes of fluid should be passed through the sample
to ensure that any tendency toward an increased or decreased permeability is
observed. The ultimate permeability cannot be established if the permea-
bility changes appreciably with time.
Under conditions of constant applied stress, changes in pore pressure
can cause changes in sample volume. In a constant head test, some of the
initial measured in-flow into the sample compensates for the volume change
rather than being the result of steady-state seepage (Olson and Daniel,
1979). Permeability tests should be run long enough to ensure that steady-
state values are obtained. Peirce and Witter (1986) discuss several methods
for developing termination criteria in tests with water and other chemical
solutions. -.'"-"
3-49
-------
Some clays exhibit thixotropic behavior; their internal structure
and flow characteristics change with time. These changes can increase
permeability with time because of larger effective pore sizes in the clay.
Mitchell et al. (1965) found measured permeabilities to be as much as six
times greater for samples tested at 21 days compared to samples tested
immediately after compaction. Dunn and Mitchell (1984) reported similar
results. Boynton and Daniel (1985) did not observe thixotropic changes
in stored samples; but they attributed this to their having compacted them
dry of optimum, whereas Mitchell et al. (1965) and Dunn and Mitchell (1984)
compacted wet of optimum.
Although not usually a significant factor, microorganisms present in
test samples have been shown to influence permeability by clogging the flow
channels with organic matter or with gases produced by the microorganisms.
Allison (1947) reported long-term permeabilities of sterile soils to be 8 to
50 times lower than the values of nonsterile soils.
3.8.4 Laboratory Permeability Tests
Examples of permeability test devices and the methods used for testing
are described briefly below. The emphasis of the discussion is on features
that distinguish the various devices and their advantages and disadvantages.
The relative merits of fixed- and flexible-wall permeameters are discussed by
Daniel et al., 1985.
3.8.4.1 Pressure Cell--
The accepted standard test for laboratory determination of saturated
permeability in the agricultural sciences (but not the geotechnical sciences)
1s called the pressure cell test. In this procedure a soil sample, which may
be an undisturbed core, a sample compacted in a mold, or a volume of soil, is
placed in a metal pressure cell. After the soil is initially saturated, it
1s connected to a standpipe. The permeant fluid is introduced through the
standpipe and forced through the pressure cell under a falling head. The
standpipe may or may not be connected to a source of air pressure to super-
Impose a pressure head over the fluid column. The pressure cell apparatus
for a falling head test is illustrated in Figure 3-17. A constant head test
could also be conducted in the pressure cell by measuring the volume of the
effluent forced through the sample within a timed interval.
Saturation of the sample core is usually accomplished by submerging one
end of the core 1n a pan of water for 16 hours with the other end open to the
air. This procedure may not be adequate to saturate the sample completely
prior to the permeability test. Vacuum wetting and fluctuating external gas
pressure are alternate saturation techniques that may be used with the pres-
sure cell.
3.8.4.2 Compaction Permeameter— ;
The compaction permeameter (also called a fixed-wall permeameter) has
been developed for testing permeant fluids with a compacted soil layer. One
advantage over the pressure cell is that the sample is compacted directly in
the permeability test device. As a result, a better seal is obtained between
the sample and the walls of the test vessel. To ensure a good sidewall seal,
some workers apply a bentonite slurry to the Inside of the chamber before the
sample is introduced. Silicon grease has*also been tried. As with the pres-
sure cell, the compaction mold may be used in either a falling head test or
a constant head test. The compaction permeameter may be modified so that
3-50
-------
To drain
Porous plate
Standpipe
"O" ring seals
Porous plate
Overflow (Fixed level)
To water source •
for filling standplpe
2-way stopcock
After Klute, 1965
Figure 3-17. Apparatus for pressure cell method.
3-51
-------
elevated pressures can be superimposed to reduce the testing time. A
modified compaction permeameter is illustrated in Figure 3-18. Not shown
on the diagram is the source of compressed air with a water trap, regulator,
and pressure meter. Also not shown is a fraction collector with automatic
timer that handles the collection and isolation of effluent samples.
The sample is leveled and the fluid^chamber is slowly filled so as not
to disturb the sample surface. The filled permeameter is allowed to stand
for a period of time (typically overnight or longer) to allow hydration and
swelling of the clay. Pressure is applied to the sample only after it has
completely hydrated.
In compatibility testing, baseline permeability determinations are
often obtained with the standard permeant fluid before the test permeant is
introduced. Alternatively, baseline values are obtained in separate tests
with other samples of the same soil.
A recent modification of the compaction permeameter device is the
double-ring permeameter. This apparatus has been developed for use in
compatibility tests involving waste leachates that may cause sample volume
changes during testing (Anderson, 1983). In the double-ring permeameter
test, a soil sample is compacted in a cylindrical mold that is mounted on
a base plate with a short cylindrical section mounted on its upper surface
(Figure 3-19). Separate outlets in the inner and outer rings serve to
separate sidewall leakage from flow through the central soil matrix. This
type of device may be especially useful for tests with soils that are liable
to undergo volume changes during permeability testing. Also, the double-ring
permeameter is appropriate for use in compatibility tests that involve
liquids that may cause shrinkage of the sample due to development of soil
structure. ;
Testing in the double-ring permeameter begins by passing sufficient
standard permeant fluid through the soil to obtain stable baseline perme-
abilities for both the inner and outer compartments. If the permeability
of the outer compartment is significantly higher than that of the inner
compartment, the soil core is discarded and a new core is compacted in the
mold. After stable baseline permeabilities are obtained, the standard
permeant is replaced by the test fluid, which is then passed through the
sample. The volume passed through each compartment to measured and reported
based on the pore volume of that compartment.
Tests in the double-ring permeameter are usually conducted with a con-
stant elevated hydraulic gradient. They also can be performed as a constant
head test without superimposed air pressure. As in other tests, 100-percent
saturation of the sample may not be achieved unless backpressure is used.
Permeability value, leachate volume, and time increment are recorded
for discrete volumes of permeant fluid passed through both chambers during
a compatibility test. Permeability measurements are then plotted versus
the total number of pore volumes passed through the sample. Data plotted
in this way indicate changes in permeability that might be expected for a
clay liner in service over a period of time that corresponds to the number
of pore volumes exchanged in the test.
3-52
-------
Pressure intake
Sealing gaskets
Pressure release
Top plate
Clamping stud
Outlet
Base plate
Porous stone insert
Source: Anderson and Brown, 1981
Figure 3-18. Modified compaction permeameter.
3-53
-------
01
Permeameter
Base Plate
Outlet for
Inner Ring
Outlet for
Outer Ring
Outer Ring
Source: Anderson, 1983
Figure 3-19. Detail of the base plate for a double-ring permeameter.
-------
Because the double-ring device has been introduced only recently,
a criterion has not been established to determine what constitutes a
significant difference between the permeability in the central compart-
ment and that in the annular space. This criterion needs to be developed
if the device is to be widely used.
3.8.4.3 Triaxial Cells--
Samples prepared for triaxial testing consist of cylindrical columns
of compacted soil encased laterally with a flexible membrane (often latex
rubber) and enclosed at the ends with porous stones. The enclosed soil
sample is placed in a water-filled cell that can be pressurized to provide
a confining pressure on the sides of the sample (Figure 3-20). The sample
to be tested may be prepared in a compaction mold and extruded for testing
in the triaxial cell. Samples are typically 2 to 4 inches in diameter.
Undisturbed Shelby tube samples may also be tested.
Permeability tests are usually performed by passing permeant liquids
upwards into the sample under pressure while maintaining a lower pressure
at the exit port on the top of the sample. A confining pressure somewhat
greater than the pressure under which the liquids enter the soil is imposed
to press the flexible membrane firmly against the soil sample, preventing
flow along the sidewall. The confining pressure on the soil sample should
be selected to simulate the lateral pressures a material will experience in
the clay liner.
Sample saturation is accomplished by forcing a standard permeant liquid
upward through the sample. Backpressure may be applied to hasten the dis-
solution of trapped air bubbles or gases generated by reactions between
permeant liquid and the sample. The use of backpressure ensures virtually
complete saturation of the sample, a factor that probably contributes to the
good reproducibility of triaxial tests. Pressure regulators and electronic
pressure transducers are used to control and monitor sample stress conditions
and to assess the saturation state within the sample during testing.
In a typical test, sufficient standard liquid is passed through the
sample to establish a stable baseline permeability before the test permeant
fluid is introduced. For some tests the test permeant is introduced directly
with baseline being determined with another soil sample. The test may be
conducted as a constant head or falling head test. Permeability value,
leachate volume, and time increment are reported for each volume of test
fluid passed through the sample. These data are used to plot permeability
versus pore volumes of test fluid passed.
Excellent precision (+20 percent) based on four samples has been
reported for the apparatus~s.hown 1n Figure 3-20 (Haji-Djafari and Wright,
1982)•
Questions have been raised about whether triaxial cells are appropriate
for compatibility testing of permeant fluids that may cause the soil sample
to shrink (Chapter 4). Because of the confining cell pressure, the flexible
membrane will contract with the soil sample so that shrinkage cracks that
could occur 1n a field situation and that might appear 1n fixed-wall devices
as sidewall leaks would not be simulated in the triaxial cell. Not all
researchers agree on this issue since the effect has not been demonstrated
3-55
-------
CELL
PRESSURE
SOURCE
(COMPRESSED AIR)
-------
clearly in comparative tests. Boynton and Daniel (1984) have shown that
cracks in des-iccated and rewet samples would only partially close unless
substantial effective stress were applied.
Although conventional triaxial cells are not designed to withstand
concentrated chemical permeants, several modifications are possible to
minimize the problem of chemical attack on the test device. Polyethylene,
nylon, or Teflon® fittings may be used in place of the customary brass or
steel fittings. The use of colostomy bags (of the type used in surgical
applications) in place of the conventional latex membrane has also been
reported for use with wastes such as gasoline (J. Withiam, D'Appolonia
Consulting Engineers, Inc., Pittsburgh, PA, personal communication, 1983).
Samples may also be wrapped in thin Teflon® tape to prevent contact between
the permeant and the latex membrane.
3.8.4.4 Consolidation Cells--
Consolidation cells (consolidometers) are commonly used in the field of
geotechnical engineering to determine the compressibility and rate of settle-
ment of soils. Consolidation occurs when water is squeezed out of the soil
and is therefore a function of permeability. A fixed-ring consolidation cell
can be used to measure permeability (Figure 3-21).
The consolidation cell method is routinely used in testing permeability
for applications such as earth dams, retaining walls, and slurry trenches.
The method has not been widely used in the evaluation of chemical compat-
ibility with clay liner material.
3.8.5 Field Permeability Tests
The permeability of clay liners can be determined in the field by per-
forming bore hole permeability tests or by using porous probes, air entry
permeameters, the Guelph permeameter, or ring infiltrometers. The reader
is advised to consult EPA method 9100 "Saturated Hydraulic Conductivity,
Saturated Leachate Conductivity, and Intrinsic Permeability" (U.S. EPA,
1986), which contains the EPA recommended procedures for the Subtitle C
(Hazardous Waste) RCRA program.
3.8.5.1 Bore Hole Tests—
The two stage bore hole permeability test was developed by G. Boutwell
and described in Boutwell and Derick, 1986. The discussion provided here is
based on Daniel, 1987.
To perform this test, a hole is drilled into the clay liner and a casing
is installed and grouted around the outside. The depth of the hole must be
at least 5 times greater than its diameter and the distance between the
bottom of the hole and the liner bottom must be greater than 5 diameters.
After placement, the casing is capped and both casing and stand pipe are
filled with water (Figure 3-22). The Stage I test consists of a series of
falling head tests in which the head (H) driving the flow is measured from
3-57
-------
Vertical Stress
T Pressure
Permeant Ruid
,n,e«
Effluent
Outlet-
LT
Outlet
irn
Soil
u
Porous Stone
Figure 3-21. Consolidation permeameter.
3-58
-------
H
STAGE I
.2.
Clay Liner
'Casing
• Grout
>5D
>5D
H
STAGE II
.2.
Clay Liner D
>5D
>5D
Figure 3-22. Two-stage, borehole permeability test (Boutwell and Oerick, 1986).
3-59
-------
the bottom of the bore hole. The Stage I hydraulic conductivities (ki)
are computed as follows:
(t -
in OyH-) (3.17)
The steady state value of ki is found by plotting the computed ki values
as functions of time. For the second stage of the test the top of the per-
meameter is removed and the hole is made deeper. The new uncased segment of
the hole must have a length to diameter ratio (L/D) of between 1 and 1.5. A
second series of falling head tests Is performed in which the head (H) is is
measured from the midpoint of the new uncased section of the bore hole. The
Stage II hydraulic conductivities (k2) are computed as follows;
k2 = (A/B) in (Hj/Hg) (3.18)
where .
A - d2 (in [(L/D) + (1 + (L/D)2)172]) (3.19)
B = 8D (L/D) (1- 0.562 exp [-1.57 (L/D)]) (3.20).
The steady state value of k2 is found by plotting the k2 values as
functions of time.
A parameter m is defined as:
where
kfo = horizontal hydraulic conductivity
kv = vertical hydraulic conductivity "
Arbitrary values of m typically ranging from 1 to 10, are chosen and
used to compute values of k£/ki by substituting into the following
expression:
211/2.I
In J(mL/D) + [1 + (mL/D)2]1/2£
3-60
-------
Next, the chosen values of m and the corresponding computed values of
k2/ki, are plotted. The real value of k£/ki is then determined using the
k2 and kj valties obtained from equations 3.18 and 3.17. The k2/kj vs.
m graph is then used to find the m value corresponding to the real
value. Finally, the horizontal (k^) and vertical (kv) hydraulic
conductivities are calculated from the following equations:
kh = mki (3.21)
kv = (l/m)ki (3.22)
According to Daniel (1987), "The advantages of borehole tests are that
the devices are relatively easy to install, they can be installed at great
depth, the cost is relatively low, the hydraulic conductivity in both the
vertical and horizontal directions can be measured, and relatively low
hydraulic conductivities (as low as about 1 x 10~9 cm/s) can be measured.
The disadvantages are that the effects of incomplete and variable saturation
are unknown, the influence of soil suction upon the results is ill defined,
the test cannot be used near the top or bottom of a liner, and the volume of
soil that is permeated is relatively small. Boutwell and Derick indicate
that the test has worked well and present several case histories."
3.8.5.2 Porous Probes—
The porous probes are small cylindrical devices, with a porous section
near the tip, that can be pushed or driven into the soil. Following emplace-
ment, the tip is filled with water and measurements of the rate at which the
water passes out into the surrounding soil are used to compute the horizontal
permeability. A porous probe is illustrated in Figure 3-23. In one com-
mercially available unit, air pressure is used to supply the driving head
(Torstensson, 1984). When using a porous probe, several important
assumptions have to be made (Daniel, 1987). These are:
t The pore water pressure in the soil surrounding the probe is known
t The degree of saturation does not vary in the soil volume through
which the permeant flows
« The soil has not been unduly disturbed by the insertion of the probe.
«
Porous probes are easy to install and, in the case of one commercially
available unit, produce results within several hours. Among their dis-
advantages is their small size which only allows the permeation of small
volumes of soil .
In field tests, porous probes yielded permeability values that were in
good agreement with both laboratory and sealed double ring infiltrometer
tests. (Chen and Yomamoto, 1987; Petsonk et al., 1987).
3-61
-------
Refill
Seal
Flush
Porous
s^ Material
V
Figure 3-23. Installed porous probe (DanieJ, 1987).
3-62
-------
3.8.5.3 Air-Entry Permeameter--
An air-efitry permeameter is used to measure soil air-entry values from
which permeabilities can be calculated. In a saturated soil in which the
pressure head is decreasing, the negative pressure head at which air first
enters the soil and becomes essentially continuous in the soil pores is
called the air-entry, value. Conversely, when an unsaturated soil is wetted,,
the pressure head of water in the soil increases and that point at which
water has displaced most of the air and has become essentially continuous in
the pores is called the water-entry value (or air-exit value). In a number
of granular materials the water-entry value is about half the air-entry value
(Bouwer, 1978).
An air-entry permeameter consists of a metal cylinder, approximately
25 cm in diameter, fitted with an air tight top equipped with a water supply
reservoir, a vacuum gauge, an air escape valve, and as depicted in Figure
3-24, a tensiometer. The cylinder is embedded in the soil test patch to a
depth of approximately 10 cm and the unit is filled with water, all air being
allowed to escape through the air escape valve. This valve is closed and
water is periodically added to the reservoir as infiltration of the soil
takes place. When the wetting front has reached a depth of 10 cm, no more
water is added to the reservoir and the supply valve is closed. The wetting
front depth is estimated on the basis of past experience with the unit or, if
a modified unit is being used, on the basis of the tensiometer readings.
After closing the supply valve, the wetting front no longer advances and the
pressure gauge will start to display negative pressures that reach a minimum
value when the air-entry value of the wetted zone is reached. (The negative
pressure develops as a result of water being sucked into the unsaturated soil
below the wetting front). At this point, air will start to bubble up through
the wetted zone increasing the pressure of the aboveground water in the
device. Once the minimum pressure is achieved, the test is terminated, the
device is removed, and the actual depth of the wetted front is determined by
digging a hole and observing the extent of infiltration (This may be
difficult in clay that has been compacted wet of optimum.) This last step is
not required if a tensiometer has been used. The air-entry value (Pa) is
calculated by the equation:
where
Pa = air-entry value of soil expressed as pressure head at point of
air entry, cm water
pmin s minimum pressure head as measured with the pressure gauge,
cm water
G = height of gauge above soil surface, cm
L = depth of wetting front (depth of tensiometer), cm
3-63
-------
H
Soil surface
r
Reservoir
Vacuum guage
Supply valve
Tensfometer
Air escape valve
C-cIamp
Gasket
Cylinder wall
Wetting front
After US. EPA, 1984
Figure 3-24. Modified airrentry permeameter.
3-64
-------
The saturated hydraulic conductivity is calculated by the following
equation: - V ":•••
K 2 (dH/dt) L (Rr/Rc)2
" Ht + L + l/2(Pa)
where
dH/dt = rate of fall of water level in reservoir just before the supply
valve is closed, cm/s
Ht =-height above soil surface of reservoir water level at time the
supply valve is closed, cm
Rr = radius of reservoir, cm
Re = radius of cylinder, cm.
Daniel (1987) states that several important assumptions underlying the
use of air entry permeameters are:
• The suction at base of the wetting front is the water-entry value
• The water-entry value is one-half the air entry value
• The water pressure read on the pressure gauge during the second stage
of testing is (when corrected for elevation) the negative of the
air-entry suction
• The air-entry permeameter is completely rigid
• The soil is incompressible
• The air in the soil beneath the wetting front is at atmospheric
pressure.
Daniel asserts that these assumptions are unprotfen and in some cases
unlikely to be correct.
The advantages of the air-entry permeameter are that relatively rapid
measurements can be made, a relatively large surface area is tested and the
permeability is measured vertically within the confines of the embedded
ring. The disadvantages are that the depth of soil tested is shallow, and
very low infiltration rates are difficult to measure because of thermal
effects and compliance of the permeameter itself (Daniel, 1987).
Topp and Bins (1976) found that the air-entry permeameter gave repro-
ducible values of permeability at various depths in a number of different
soils. They also found that results were consistent with values obtained
from determinations made on laboratory cores. Although the laboratory core
3-65
-------
permeability was generally lower than the permeability determined by the
air-entry permeameter, this was probably due to lack of worm holes or cracks
in the laboratory samples. Aldabagh and Beer (1971) found the method to be
consistent and reasonable with variability substantially less than the
natural variation of soil.
Knight and Haile (1984) used an air-entry permeameter on an earthen liner
and measured permeabilities between 5 x 10~9 an 3 x 10~7 cm/s. Undis-
turbed samples subjected to laboratory tests had permeabilities about one-
half order of magnitude less than measured with the air-entry device.
3.8.5.4 The Guelph Permeameter--
The Guelph permeameter (manufactured by Soilmoisture Equipment Corp.,
Santa Barbara, California) derives its name from the University of Guelph, in
Canada, where it was developed. The device consists of a simple Mariotte
bottle device that, when installed in a bore hole, can be used to measure the
"field saturated" hydraulic conductivity of the surrounding soil. A sche-
matic diagram of the device is shown in Figure 3-25.
The use of this device is based on the theory that, when a bore-hole 1n
soil is filled with water to a constant height, a bulb shaped zone of satura-
tion is established around the hole. (See Figure 3-25). Once the saturated
zone is established, the outflow of water becomes constant. The rate of out-
flow associated with each of two different water depths along with the radius
of the bore-hole can be used to calculate the field saturated conductivity of •
the soil .
The following generalized equation solves for field saturated con-
ductivity (permeability) using data generated by the Guelph permeameter
(product literature):
where
=
2 TT [ 2 Hj H2 (H2 - Hj);* a2
[H
G s G —
c]
Q2 - (X) (R2)
Q - (x) (R)
3-66
-------
Air Tube
Water Supply
Seal
Partial Vacuum, PI
* Water Column, P£
Atmospheric Pressure, PQ
Soil
Zone of Saturation
Figure 3-25. Schematic diagram of Guelph permeameter (P<\ + P£ = PQ)-
3-67
-------
and ;
a ="wen radius, cm
C.., C2 = proportionality factors dependent primarily on ^1 and ^2
respectively , a a
HI, \\2 = well height for first and second measurements respectively, cm
K = permeability, cm/sec.
R!» RE = steady-state rate of fall of water, in the reservoir of the
permeameter, corresponding to HI and H2, respectively, cm/sec.
X = reservoir constant corresponding to the cross-sectional area,
cm^
The advantages of the Guelph permeameter in measuring the permeabili-
ties of soils in situ include the following (product literature):
t Instrument is portable, lightweight, easily assembled in the field,
and easy to use.
• Assembly and measurements require only one person, and only 2.5
liters of water is needed to operate instrument.
• Design is simple so that serviceability is easy.
• Measurements are easily defendable because the method is based on the
fundamental, firmly established constant head well principle.
In spite of these advantages, there are some significant disadvantages
in using the Guelph permeameter to measure clay liner permeabilities. One
disadvantage is that it takes a long time to obtain results in fine grained
materials. Measurements in coarse grained soils can be completed within a
couple of hours at most, but compacted clays can require 2 to 3 days
(Bradshaw, 1986). This makes it impossible to take more than a few tests
over a short period of time unless many permeameters jtre used.
Another drawback in measuring the permeabilities of clay liners is that
the permeameter itself can be affected by temperature changes brought on by
the sun. The air in the reservoir expands and causes water to flow quickly
from the permeameter when -the sun is out. At the end of the day, the tem-
perature decreases and causes the air in the reservoir to contract, making
the water flow back in from the well. While heating and cooling effects are
insignificant when coarse soils are involved, measurements in compacted clays
must be averaged between both warm and cobl hours of the day. Tests are run
twice because temperature related water level fluctuations are large compared
to water level change due to percolation. (Bradshaw, 1986)
One problem that occurs when testing compacted clays is that it is dif-
ficult to ensure that the permeameter 1s working properly. The air bubbles
that result from displacement by percolation are infrequent and large as they
3-68
-------
are observed in the calibrated reservoir. A technique has been developed
which allows smaller bubbles to appear and at a more regular rate. Crushed
quartz can be-placed in the hole around the,permeameter which reduces the
volume of water necessary for each increment of head drop. Since the perme-
ability of the quartz is greater than that of the soil, the quantity of water
migrating from the well is the same. (Bradshaw, 1986)
An additional problem arises from the influence of atmospheric
moisture. Tests in compacted clays take long enough and involve small enough
changes in water level that condensation and evaporation affect results.
Plastic hole covers can be used to decrease these effects. (Bradshaw, 1986)
The Guelph permeameter is limited in its usefulness, as designed,
since the instrument is only capable of measuring permeabilities as low
as 10-D cm/sec. (Guelph permeameter product literature). Permeabilities
of 10-' cm/sec have been measured experimentally using larger volume well
holes (Bradshaw, 1986). Since regulatory requirements for clay liners
state that 10-' cm/sec is the maximum permeability allowable, it follows
that instrumentation should be capable of measuring permeabilities that are
lower than this for data to be reliable.
Another limitation arises from the assumption that the soil being tested
is homogeneous (Guelph permeameter product literature). This cannot be
easily verified under field conditions. The assumption does not allow for
variations in the clay liner due to large clod size, for example. If the
test was taken in a clod, the permeability might be lower than what is
representative of the liner as a whole.
A final drawback to testing with the Guelph permeameter is the diffi-
culty in quantifying results. The calculations required to obtain accurate
permeabilities from measurements in compacted clays are cumbersome. There
are standardized equations that use simplifying assumptions (provided in the
product literature), but these are only useful when greater permeabilities
are involved.
3.8.5.5 Ring Infiltrometers—
In its simplest form, a ring infiltrometer is a short length of tubing
that can be partially embedded in soil and filled with water. The rate at
which the water infiltrates into the soil can be measured by recording the
water depth or by keeping track of the amount of water needed to maintain a
constant depth within the infiltrometer. Ring infiltrometers can be of the
single or double ring type (Figure 3-26). The double ring infiltrometer has
a large diameter and small diameter ring placed concentrically. Both rings
are filled with water and, in theory, the water infiltrating from the outer
ring restricts the lateral spread of water infiltrating from the inner ring.
Data from the inner ring is used to calculate the infiltration rate.
The ASTM method for double ring infiltrometers (D3385-76) states that
the method is "difficult to use and the resultant data may be unreliable in
very coarse or heavy clay soils, or in frozen or highly fractured ground"
(ASTM, 1984). In the ASTM procedure the volume of water added to maintain a
3-69
-------
Outer ring water level
Soil surface
Scale
Water leval
Inner ring
Outer ring
1 Burlap (to prevent puddling)
Source: EPA/Army Corps of Engineers/USDA, 1977
Figure 3-26. Double-ring infiltrometer.
-------
constant level 1n the inner ring is taken as the measure of the volume that
infiltrates the soil. The volume that infiltrates over a time interval can
be converted to an infiltration velocity expressed as inches or centimeters
per hour. The ASTM method does not directly yield a permeability value for
the test soil.
The infiltration rate (I) can be calculated from the equation:
I = q/A
where
q = volume of flow per unit time
A = inside area of the ring
The hydraulic conductivity (K) can then be calculated from the equation.
K = I/i
where
i = hydraulic gradient.
If the test is run until the wetting front reaches the bottom of the
liner then the hydraulic gradient is equal to the total head divided by the
liner thickness. Alternatively, the depth of the wetting front can be used
in place of the liner thickness. This can be determined with tensiometers
as depicted in Figure 3-27 which is an illustration of a sealed double-ring
infiltrometer with two tensiometers in place. If the tensiometers are
attached to a differential pressure gauge or manometer, the passage, and,
therefore, depth of the wetting front can be observed. The hydraulic
gradient is found from the equation:
i = (H + L)/L
where
H = depth of water above liner surface
L = depth of wetting front or liner thickness if break through has
occurred.
Sealed double ring infiltrometers, as depicted in Figure 3-27 are the
newest development in ring inflltrometers. In these units the center ring is
fitted with an air-tight top. A tube leads from the top to a flexible
plastic bag that serves as the reservoir supplying make-up water to the
sealed inner ring. The entire inner ring assembly is kept completely
sub-merged within the outer ring. Periodically, the flexible bag is removed
and weighed to determine the water volume that has infiltrated the soil under
the inner ring. Submerging the sealed inner ring and flexible bag assures
that no pressure differences develop between the inner and outer rings
(providing the temperature is the same).
3-71
-------
Sealed Inner Ring
Tensiometers
Flexible Bag
Outer Ring
Wetting Front
^S. /> Clay Liner
Grout
Figure 3-27. Sealed double-ring infiltrometer.
3-72
-------
Ring infiltrometers, compared to other devices, have the advantage that
they cover a relatively large area thus subjecting a large volume of soil to
testing. The conversion of the infiltration rate into a permeability value
is based on the assumptions of one-dimensional vertical flow through the soil
mass and a uniform advance of the wetting front. Nolan (1983) performed a
series of tests with a single ring infiltrometer and found that under his
experimental conditions neither of these assumptions were strictly correct.
The sealed double ring infiltrometer has the disadvantage of requiring
more setup time than other types of infiltrometers. Ring infiltrometers,
in general require long test periods, 30 to 90 days being typical for clay
liners and even longer periods required in some situations, for example if
the liner has a low degree of initial saturation. Another disadvantage
is that very low permeabilities (less than 10~8 cm/s) cannot be accu-
rately measured with these devices. (Daniel, 1987)
3.9 REFERENCES
Acar, Y. B., and S. D. Field. 1982. Organic Leachate Effects to Hydraulic
Conductivity in Fine-Grained Soil, Volume 1. Report No. GE-82/01,
Louisiana State University.
Aldabagh, A. S. Y., and C. E. Beer. 1971. Field Measurement of Hydraulic
Conductivity Above a Water Table with an Air-Entry Permeameter.
Transactions of the American Society of Agricultural Engineers.
14:29-31.
Allison, L. E. 1947. Effect of Micro-organisms on Permeability of Soil
Under Prolonged Submergence. Soil Science. 63:439-450.
Anderson, D. 1983. Effects of Organic Solvents on Clay Liners - Contaminant
Resistant Bentonite Slurry Mixtures. Report Prepared for U.S.
Environmental Protection Agency, Cincinnati, Ohio, 46 pp.
Anderson, D. C. 1981. Organic Leachate Effects on the Permeability of Clay
Soils. M.S. Thesis, Soil and Crop Sciences Department, Texas A & M
University, College Station, Texas.
«
Anderson, D. C. and K. W. Brown. 1981. Organic Leachate Effects on the
Permeability of Clay Liners, pp. 119-130. In D. W. Shultz (ed.) Land
Disposal: Hazardous Waste. EPA-600/9-81-002b. National Technical
Information Service, Springfield, Virginia.
Anderson, J. L., and J. Bouma. 1973. Relationships Between Saturated
Hydraulic Conductivity and Morphometric Data of an Argillic Horizon.
Soil Science Society of America Proceedings. 37:408-413.
Andrews, R. E., J. J. Gawarkiewicz, and H. F. Winterkorn. 1967. Comparison
of the interaction of 3 clay minerals with water, dimethyl sulfoxide,
and dimethyl formamide. Highway Research Record. No. 209. pp. 66-78.
ASTM. 1984. Annual Book of ASTM Standards, Part 19, Soil and Rock; Building
Stones. American Society for Testing and Materials, Philadelphia,
Pennsylvania.
3-73
-------
Boutwell, G. P., and R. K. Derick. 1986. Groundwater Protection for
Sanitary-Landfills in the Saturated Zone. Paper presented at Waste
Tech '86. National Solid Waste Management Association, Chicago,
Illinois.
Boutwell, G. P., and. V. R. Donald. 1982, Compacted Clay Liners for
Industrial Waste Disposal. ASCE National Meeting, Las Vegas. Nevada,
April 1982. 23 p.
Bouwer, H. 1978. Groundwater Hydrology. McGraw-Hill Book Company, New
York, New York. 480 pp.
Bowles, J. E. 1979. Physical and Geotechnical Properties of Soils.
McGraw-Hill Book Company, New York. 478 pp.
Boynton, S. S., and D. E. Daniel. 1985. Hydraulic Conductivity Tests on
Compacted Clay. J. of Geotechnical Engineering. 111(4):465-478.
Bradshaw, K. L., 1986. An Assessment of the Standard Percolation Test.
Masters Thesis. University of Waterloo, Waterloo, Ontario.
Brown, K. W., J. Green, and J. Thomas. 1982. The Influence of Selected
Organic Liquids on the Permeability of Clay Liners (draft). Texas
Agricultural Experiment Station, Soil and Crop Science Department,
College Station, Texas.
Bryant, J., and A . Bodocsl. 1986. Precision and Reliability of Laboratory
Permeability Measurements, EPA-600/2-86-097, U.S. Environmental
Protection Agency, Cincinnati, Ohio. 177 pp.
Chen, H. W., and L. 0. Yamamoto. 1987. Permeability Tests for Hazardous
Waste Management Unit Clay Liners. In: Geotechnical and Geohydrologi-
cal Aspects of Waste Management. D. J. A. van Zyl et al. eds. Lewis
Publishers, Inc., Chelsea, Michigan, pp. 229-243.
Daniel, D. E. 1987. Hydraulic Conductivity Tests for Clay Liners. In:
Ninth Annual Symposium on Geotechnical and Geohydrological Aspects of
Waste Management. Colorado State University, Fqrt Collins, Colorado.
Daniel, D. E. 1981. Problems in Predicting the Permeability of Compacted
Clay Liners. In: Symposium on Uranium Mill Tailings Management, Fort
Collins, Colorado, pp. 665-675.
Daniel, D. E., D. C. Anderson, and S. S. Boynton. 1985. Fixed-Wall vs.
Flexible-Wall Permeameters. In: Hydraulic Barriers for Soil and Rock,
ASTM STP 874, American Society for Testing and Materials, Philadelphia,
Pennsylvania, pp. 107-123.
Daniel, D. E., S. J. Trautwein, S. S. Boynton, and D. E. Foreman. 1984.
Permeability Testing with Flexible Wall Permeameters. Geotechnical
Testing Journal. 7(3):113-122.
Dunn, R. J., and J. K. Mitchell. 1984. Fluid Conductivity Testing of Fine
Grained Soils. J. of Geotechnical Engineering. 110(11):1648-1665.
3-74
-------
Griffin, R. A., G. L. Herzog, T. M. Johnson, W. J. Morse, R. E. Hughes,
S. F. J. Chou, and L. R. Folmer. 1985. Mechanisms of Contaminant
Migration Through a Clay Barrier -- Case Study, Wilsonville, Illinois.
In: Land Disposal of Hazardous Waste: Proceeding of the Eleventh
Annual Research Symposium. (EPA/600/9-85-013) U. S. Environmental
Protection Agency, Cincinnati, Ohio. pp. 27-38.
Haji-Djafari, S., and J. C. Wright. 1982. Determining the Long-Term Effects
of Interactions Between Waste Permeants and Porous Media. Presented at
American Society for Testing and Materials Second Symposium on Testing
of Hazardous and Industrial Solid Wastes. 26 p.
Hansbro, S. 1960. Consolidation of Clay with Special Reference to the
Influence of Verticle Sands Drains. In: Proceedings 18, Swedish
Geotechnical Institute, Stockholm, Sweden.
Herzog, B. L. and W. J. Morse. 1984. A Comparison of Laboratory and Field
Determined Values of Hydraulic Conductivity at a Waste Disposal Site.
In: Seventh Annual Madison Waste Conference Proceedings. University of
Wisconsin Extension, Madison, Wisconsin, pp. 30-52.
Holtz, R. D., and W. D. Kovacs. 1981. An Introduction to Geotechnical
Engineering. Prentice-Hall, Englewood Cliffs, New Jersey. 733 pp.
Horton, R., M. L. Thompson, and J. F. McBride. 1987. Method of Estimating
the Travel Time of Noninteracting Solutes Through Compacted Soil
Material. Soil Sci. Soc. Am. J. 51(l):48-53.
Johnson, A. I. 1954. Symposium on Soil Permeability. ASTM STP 163,
American Society for Testing and Materials, Philadelphia, Pennsylvania.
pp. 98-114.
Klute, A. 1965. Laboratory Measurement of Hydraulic Conductivity of
Saturated Soil. In: Methods of Soil Analysis, C. A. Black, Ed.
Amer. Soc. of Agronomy, Madison, Wisconsin, pp. 210-220.
Knight, R. B., and J. P. Haile, 1984. Construction of the Key Lake Tailings
Facility. In Proceedings of the International Conference on Case
Histories in Geotechnical Engineering. St. Louis, Missouri.
Lambe, T. W. 1955. The Permeability of Fine-Grained Soils. ASTM 163,
American Society for Testing and Materials, Philadelphia, Pennsylvania.
pp. 55-67.
Lambe, T. W., and R. V. Whitman. 1979. Soil Mechanics, SI Version.
John Wiley and Sons, Inc., New York. 553 pp.
Lee, D. Y. and R. J. Suedkamp. 1972. Characteristics of irregularly shaped
compaction curves of soils. Highway Research Record, No. 381, p. 1-9.
Luxmoore, R. J., B. P. Spalding, and I. M. Munro. 1981. Area! Variation and
Chemical Modification of Weathered Shale Infiltration Characteristics.
Soil Science Society of America Journal. 45:687-691.
3-75
-------
Mason, D. D., J. F. Lutz, and R. G. Peterson. 1957. Hydraulic Conductivity
as Related to Certain Soil Properties in a Number of Great Soil Groups-
Sampling Errors Involved. Soil Science Society of America Proceedings.
21:554-561.
Matyas, E. L. 1967.. Air and Water Permeability of Compacted Soils:
Permeability and Capillarity of Soils. Standard Technical Publication
417, ASTM, Philadelphia, Pennsylvania, pp. 160-175.
Mitchell, J. K. 1976. Fundamentals of Soil Behavior. John Wiley and Sons,
New York. 422 pp.
Mitchell, J. K., and J. S. Younger. 1967. Abnormalities in Hydraulic Flow
through Fine-Grained Soil. ASTM STP 417. pp. 106-139.
Mitchell, J. K., D. R. Hooper, and R. G. Campanella. 1965. Permeability of
Compacted Clay. Journal of the Soil Mechanics and Foundation Division,
ASCE. 91:41-66.
Nolan, T. W. 1983. Evaluation of the Single-Ring Infiltrometer for
Measuring Hydraulic Conductivity of Soil Lines. Masters Thesis.
Syracuse University, Syracuse, New York. 93 pp.
Oakes, D. J. 1960. Solids Concentration on Effects in Bentonite Drilling
Fluids. Clay and Clay Minerals. 8:252-273.
Olson, R. E., and D. E. Daniel. 1979. Field and Laboratory Measurements
of the Permeability of Saturated and Partially Saturated Fine-Grained
Soils. In: ASTM Symposium on Permeability and Groundwater Contaminant
Transport, Philadelphia, Pennsylvania. 67 pp.
Peirce, J. J. and K. A. Witter. 1986. Termination Criteria for Clay
Permeability Testing. ASCE Journal of Geotechnical Engineering,
September 1986. 112(9):841-854.
Petsonk, A. M., W. Romanowski, and V. Richards. 1987. Field Permeability
Testing for a RCRA Landfill Final Cover. In: Proceedings of Hazmacon,
87. Santa Clara, California. April 22.
Peyto?Ao~* R" J' P< Gibb' M' H* LeFalvrei, J. D. Burch, and M. J. Barcelona.
1986. Effective Porosity of Geologic Materials. In: Land Disposal,
Remedial Action, Incineration and Treatment of Hazardous Waste -
Proceedings of the Twelfth Annual Research Symposium. (EPA/600/
9-86/022) U.S. Environmental Protection Agency, Cincinnati, Ohio.
pp. 21-28.
Proctor, R. R. 1933. Fundamental Principles of Soil Compaction. Engi-
neering News-Record. 111(9):245-258.
Sangrey, D. A., D. K. Noonan, and G. S. Webb. 1976. Variation in Atterberg
Limits of Soil Due to Hydration History and Specimen Preparation. Soil
Specimen Preparation for Laboratory Testing, ASTM STP 599, American
Society for Testing and Materials, pp. 158-168.
3-76
-------
Smith, R. M., and D. R. Browning. 1942. Persistent Water-Unsaturation of
Natural -Soil in Relation tq Various SoiJ and Plant Factors. Soil
Science Society of America Proceedings. 7:114-119.
Sowers, G. B., and G. F. Sowers. 1970. Introductory Soil Mechanics and
Foundations. The Macmillan Company, New York. 556 pp.
Spigolon and Kelley. 1984. Geotechnical Assurance of Construction of
Disposal Facilities. EPA 600/2-84-040, U.S. Environmental Protection
Agency, Cincinnati, Ohio.
Topp, G. C., and M. R. Binns. 1976. Field Measurement of Hydraulic
Conductivity with a Modified Air-Entry Permeameter. Canadian Journal
Soil Science. 56:139-147.
Torstensson, B. 1984. A New System for Ground Water Monitoring. Ground
Water Monitoring Review. Fall, pp. 131-138.
Truesdale, R. L., L. Goldman, J. Peirce, B. Cox, K. Witter, and T. Peel.
1985. Laboratory Methods for Testing the Permeability and Chemical
Compatibility of Inorganic Liner Material (in preparation).
U.S. Department of the Army. 1953. The Unified Soil Classification System,.
Corps of Engineers, Technical Memorandum No. 3-357, Volumes 1 and 3.
U.S. Department of the Army. 1970. Laboratory Soils Testing. EM
1110-2-1906, Office of the Chief of Engineers, Washington, D.C.
U.S. Department of the Army. 1977. Construction Control for Earth &
Rock-Fill Dams. EM 1110-2-1911, Washington, D.C.
U.S. Department of Interior. 1974. Earth Manual. U.S. Government Printing
Office, Washington, D.C. 810 pp.
U.S. Environmental Protection Agency. 1984. Soil Properties, Classifica-
tion, and Hydraulic Conductivity Testing (Draft).. SW-925, Office of
Solid Waste and Emergency Response, Washington, D.C. 167 pp.
U.S. Environmental Protection Agency. 1986. Test Methods for Evaluating
Solid Waste (Third Edition). SW-846, Office of Solid Waste and
Emergency Response, Washington, D.C.
Winterkorn, H. F., and H. Y. Fang. 1975. Foundation Engineering Handbook.
Van Nostrand Reinhold, New York.
Withiam, J. 1983. D'Appolonia Consulting Engineers, Inc. Personal
communication.
Yong, R. N., and B. P. Warkentin. 1975. Soil Properties and Behavior.
El sever Scientific Publishing Company, New York. 449 pp.
Zimmie, T. F., J. S. Doynow, and J. T. Warden. 1981. Permeability
Testing of Soils for Hazardous Waste Disposal Sites. In: Proceedings
of the Tenth International Conference on Soil Mechanics and Foundation
Engineering, Vol. 2, Stockholm, Sweden, pp. 403-406.
3-77
-------
-------
CHAPTER 4
CLAY-CHEMICAL INTERACTIONS AND SOIL PERMEABILITY
It has been known for many years that the permeability of clay soils may
be drastically altered by chemicals present in the permeating liquid. Apart
from the many recent studies stemming from concern over the effects of haz-
ardous waste and waste leachates on clay liners, much research has been
carried out over the last several decades to determine the effects of various
chemicals on agricultural soils or on geological formations important for oil
production. The potential effects of certain organic fluids on clay perme-
ability were recognized as early as 1942, when Macey's experiments with
fireclay showed that the rates of flow for certain organic liquids through
clay were "of an enormously higher order than for water" (Macey, 1942).
Since Macey's experiments, many researchers have investigated the effects of
organic and inorganic fluids on clays in an effort to elucidate the causes of
the observed changes in permeability.
The current state of the knowledge in this area is complicated by the
dilemma of how to measure the permeability changes that appear to be caused
by clay-chemical interactions. The question of what types of permeameters
give valid measures of clay-chemical compatibility remains an important
issue. Because permeability studies with different fluids have been carried
out with different test protocols, different test devices, and different
clays, quantitative data comparisons cannot be made except in a few of the
more recent studies. Much can be learned from a review of the research that
has been carried out, however, and qualitative statements can be made regard-
ing the behavior of certain clays in the presence of many types of fluids.
Unifying theories of soil physics that explain the reported findings have
been advanced by several researchers. Such theories are useful for predict-
ing clay-chemical incompatibilities that could lead to performance failures
in clay-lined hazardous waste disposal facilities.
This chapter presents the experimental findings that pertain to the
effects of chemicals on clay barrier permeability as well as the theories and.
mechanisms proposed to explain the observed effects. Section 4.1 defines the
terms most important in clay-chemical compatibility testing— permeability
(or, if the permeating liquid is water, hydraulic conductivity) and intrinsic
permeability. Section 4.2 is a discussion of the clay-chemical interactions
that influence soil permeability. A summary of the relevant permeability
studies is presented in Section 4.3. Approaches that have been used in
studies to determine clay-chemical compatibility and problems associated with
different test methods are addressed in Section 4.4. Section 4.5 reviews in
more detail permeability testing efforts that have been carried out to
measure clay-chemical interactions.
4-1
-------
4.1 PARAMETERS DETERMINED IN PERMEABILITY TESTING FOR COMPATIBILITY
When results of clay-chemical compatibility tests are compared, it is
important to understand the parameters that are being measured and calcu-
lated. The parameter usually determined in compatibility tests is permea-
bility, K, which is defined by Darcy's Law as expressed in Equation (4.1):
K = Q/Ai (4.1)
where
Q = volumetric flow rate (L3/t)
A s cross-sectional area of flow (L2)
1 = hydraulic gradient (dimensionless).
K has units of length per unit time (e.g., cm/s). The density and viscosity
of the permeating liquid as well as the pore size distribution within the
soil matrix will influence the value of K. Hydraulic conductivity refers to
the value of K when the permeating fluid is water. Darcy's Law is limited to
saturated soil conditions and to laminar flow conditions.
In order to separate the effects of the liquid properties (viscosity and
density) from those of the medium (pore size distribution), a different
parameter, the intrinsic permeability, should be used. Intrinsic permeabil-
ity, usually referred to as k, is a property of the medium that is dependent
on the shape, size, and continuity of the pore spaces. Intrinsic permeabil-
ity is "a measure of the relative ease with which a porous medium can trans-
mit a liquid under a potential gradient. ! It is a property of the medium
alone and 1s independent of the nature of the liquid and of the force field
causing movement" (Lohman et al., 1972).
Intrinsic permeability, k, has units of length squared (e.g., cm2) and
Is related to permeability, K, by Equation (4.2):
K * k £ g or k = Kpg (4'2)
where »
p = density of the fluid (M/L3)
u a dynamic viscosity of the fluid (M/Lt)
g = acceleration due to gravity (L/t2).
In clay-chemical compatibility testing, the value of K for a clay soil
permeated by a certain chemical is determined from measurements of the fluid
Inflow or outflow. A change or lack of change in the value of K (when com-
pared to K for water or other baseline fluid) may be due to a combination of
two factors—
4-2
-------
9 Difference in the permeant fluid viscosity and density (compared to
baseline permeant fluid)
o Change in porous medium characteristics as a result of clay-chemical
interactions.
In order to separate these effects, it is usually necessary to report
the results of the tests in terms of intrinsic permeability (k) both for
the tests with the baseline permeant fluid and for the tests with the chemi-
cal permeant fluid in question. In practice, most researchers report and
discuss their test results in terms of permeability (K) rather than intrinsic
permeability (k). Provided the density and viscosity of the test fluid (at
the test temperature) are known, one could calculate k to correspond to each
K value reported. In general, only substantial permeability changes are
meaningful in clay-chemical compatibility testing. When large changes in
permeability are measured during the course of a test, the clay-chemical
interaction is apparent regardless of whether the k values are computed and
plotted.
4.2 CLAY-CHEMICAL INTERACTIONS THAT INFLUENCE PERMEABILITY
Mechanisms whereby the chemical nature of a permeant fluid may alter
clay soil permeability and theories to predict clay-chemical interactions
have been described by several researchers, among them Mitchell (1976),
Brown and Anderson (1980), Acar and Field (1982), Evans, Chaney, and Fang
(1981), Anderson and Jones (1983), Daniel (1982, 1983), Daniel and
Liljestrand (1984), Dunn (1983), Monserrate (1982), Peirce (1984), and
Griffin and Roy (1985). In addition to laboratory and field permeability
tests, methods such as X-ray diffraction, shrink-swell measurements, settling
tests, and other techniques have been used to investigate the clay-chemical
interactions that influence permeability.
Changes in the permeability of clay soils due to chemical interactions
may result from--
• Alterations in soil fabric stemming from chemical influences
on the diffuse double layer surrounding clay particles
• Dissolution of soil constituents by strong a&ids or bases
• Precipitation of solids in soil pores
t Soil pore blockage due to the growth of microorganisms.
The permeability of a soil may also be affected by the pore fluid veloc-
ity; high velocities can displace small particles in the soil matrix. The
fluid flow velocity can also influence chemical interactions that depend on
the time of contact between the soil and some chemical component of the
permeating fluid. Thus, permeability and fluid flow velocity are inter-
related characteristics of a soil-permeant fluid system.
4-3
-------
4.2.1 Son Fabric and Permeability
The permeability of any soil depends upon the geometric characteristics
of the area available for fluid flow. Influencing factors are the size,
shape, tortuosity, and degree of interconnection between the pore spaces.
The geometric arrangement of the soil particles will determine these pore
space characteristics.
The term "fabric" refers to the arrangement of particles, particle
groups, and pore spaces in a soil. Modes of particle association orienta-
tion in clay suspensions were described by van Olphen (1963) as "dispersed,"
"aggregated," "flocculated," or "deflocculated" (see discussion in
Section 2). Particle associations corresponding to these descriptors are
illustrated in Figure 2-10. Dispersion and flocculation represent the ex-
tremes in soil fabric classification, and a chemical present in the permeat-
ing liquid may influence the permeability of a clay soil by altering the soil
fabric toward either of these extremes. Figure 4-1 illustrates how a change
in pore diameter can drastically alter permeability.
A dispersed deflocculated soil fabric tends to have a large number of
very small pore spaces; with flocculation, relatively large-sized inter-
particle and interaggregate pores are formed. These large diameter pores
can cause drastic changes in the permeability of the clay soil since the
flow rate is proportional to the square of the diameter of the flow channel.
The pulling together of groups of clay particles into aggregates results
when cohesive forces between individual clay particles outweigh the repulsive
forces. The forces of attraction result from London-van der Waals forces and
do not vary significantly with the chemistry of the pore water. Attractive
forces are strongest close to the clay surface and diminish rapidly with
increasing distance from the surface.
The forces of repulsion between adjacent clay surfaces, however, are
primarily electrostatic and are influenced by the clay surface charge and the
chemistry of liquid adjacent to the clay surfaces. Interparticle spacing is
a function of the thickness of the diffuse double-layer cationic clouds that
form the Gouy layer of the diffuse double layer (Anderson and Jones, 1983).
(See also Section 2.3 and discussion below). In theory, the direction of
change in permeability associated with varying pore f-luid chemistry could be
predicted if the variables that affect the thickness of the diffuse double
layer are known (Acar and Seals, 1984). However, the heterogeneous mineral
composition and wide particle-size distribution common in many soils along
with the complicated nature of chemical-soil Interactions make this difficult
to accomplish in practice.
4.2.1.1 Diffuse Double-Layer Theory—
The theory of the diffuse double layer (also called the electrical
double layer) has evolved from the studies of colloid chemistry directed
at the description of surface interactions of small particles in a water-
electrolyte system. The description that follows 1s excerpted from Mitchell
(1976, pp. 112-113).
-------
Source: Anderson, 1981
Figure 4-1. Change in a pore diameter (400%) corresponding
to a permeability increase of 25,600%.
4-5
-------
In a dry clay, adsorbed cations are tightly held by the negatively
charged clay surfaces. Cations in excess of those needed to neutralize
the electronegativity of the clay particles and their associated anions
are present as salt precipitates. When the clay is placed in water the
precipitated salts go into solution. Because the adsorbed cations are
responsible for a much higher concentration near the surfaces of par-
ticles, there is a tendency for them to diffuse away in order to equal-
ize concentrations throughout. Their freedom to do so, however, is
restricted by the negative electric field originating in the particle
surfaces . . . The negative surface and the distributed charge in the
adjacent phase are together termed the DIFFUSE DOUBLE LAYER . . .
The distribution of cations adjacent to a negatively charged clay
particle in suspension (the diffuse double layer) is depicted in Fig-
ure 4-2 (see also Figure 2-6). The distribution for a particular soil-
water-electrolyte system results from a balance between the tendency of the
cations to escape due to diffusion, and the opposing electrostatic attraction
of the clay surface for the cations. The Gouy-Chapman theory of the diffuse
double layer (Gouy, 1910; Chapman, 1913) is widely recognized, and mathe-
matical descriptions of the diffuse double layer have been formulated for
both planar and spherical surfaces.
Despite the fact that the Gouy-Chapman theory does not account for all
the factors that can influence the behavior of the soil-water-electrolyte
system, it has been useful as a generalized model for explaining clay-
chemical interactions that affect permeability. Since the thickness of the
double layer influences the level of interlayer and interparticle repulsion,
system variables that influence the double-layer thickness consequently
affect the physical interactions among clay particles.
The nature and thickness of the double layers, and thus the repulsive
forces, depend upon characteristics of clay particles and the pore fluid.
In general, the tendency for particles in suspension to flocculate decreases
with increased thickness of the double layer. An approximate quantitative
indication of the relative influences of several factors on the thickness of
the double layer is given by Equation (4.3) below (see Mitchell, 1976,
p. 118):
H, -DkT
8;
where
H » thickness of the double layer
D » dielectric constant of the medium
k * Boltzman constant (1.38 x 10~16 ;erg/K)
T » temperature in degrees Kelvin
4-6
-------
©
(S)
© ©
- © ©
Disianc*
Figure 4-2. Distribution of ions adjacent to a clay surface according
to the concept of the diffuse double layer.
4-7
-------
no = electrolyte concentration
e » unit-electric charge, 16 pvp 10~6 coulomb
v = valence of cations in the pore fluid.
The thickness varies inversely with the valence of the cations present:
and inversely with the square root of the concentration; the thickness
increases with the square root of the dielectric constant and the tempera-
ture, other factors remaining constant. Bas.ed on the Gouy-Chapman model,
Lambe (1958) noted that the following variables in the soil-water system
affect double-layer thickness and colloidal stability: dielectric constant,
electrolyte concentration, temperature, ionic valence, size of hydrated ions
present, pH, and anion adsorption.
It has been found that attractive forces exceed repulsive forces when
interlayer spacing is about 0.5 nm. (Yong and Warkentin, 1975). Thus, a
sufficient reduction in repulsive forces (i.e., reduced double-layer
thickness) could "transform a massive, structureless, and slowly permeable
clay barrier into an aggregated, structured, and more permeable barrier"
(Anderson and Jones, 1983).
4.2.1.1.1 Dielectric Constant—The dielectric constant is a measure of
the ease with which molecules can be polarized and oriented in an electric
field (Mitchell, 1976, p. 113). It represents the ability of a fluid to
transmit a charge. Quantitatively, the static dielectric constant is defined
by D 1n Coulomb's equation (Equation 4.4), where F is the force of electro-
static attraction between two charges, Q and Q1, separated by a distance d.
As the dielectric constant decreases, the fluid film surrounding the
clay that contains positive cations must be thinner for the negative surface
charge on the clay to be neutralized. For a constant surface charge, the
surface potential function will increase as the dielectric constant de-
creases. Since most organic liquids have dielectric constants substantially
lower than water, it is to be expected that the double-layer thickness would
be reduced (with an associated tendency toward flocculation) when an organic
liquid rather than water surrounds the clay particle. Due to the effects of
dielectric constant on the electrical double layer, there 1s a relationship
between the dielectric constant of an adsorbed fluid and interlayer spacing
exhibited by clay particles. In general, interlayer spacing decreases with a
decrease 1n the dielectric constant, although this apparent relationship can
be complicated by the other .factors that affect interlayer spacing.
4.2.1.1.2 Electrolyte Concentration—As the electrolyte concentration
in the pore fluid increases, the thickness of the double layer tends to
decrease, promoting flocculation. An analysis of the effect of electrolyte
concentration on the double layer indicates that an increase 1n concentration
4-8
-------
reduces the surface potential for the condition of constant surface charge.
Also, the dec_ay of potential with distance is more rapid with increased
electrolyte concentration. In essence, the double layer is suppressed by an
increase in electrolyte concentration. Interparticle interactions extend to
much greater particle spacings for a low electrolyte concentration (e.g.,
0.83 x 10~4 M NaCl) than a higher concentration (e.g., 0.83 x 10~2 M
NaCl) (Mitchell, 1976, p. 122). The effect of salt concentration on the
behavior of clays has been discussed by Anderson (1981):
As salt concentration in interparticle spaces increases, the
cationic cloud is compressed closer to the clay surface, resulting in a
decrease in electrostatic repulsion and interlayer spacing. Weiss
(1958) noted the direct relationship between salt concentration and
interparticle spacing in smectitic clay minerals in a study using
distilled water and several concentrations of sodium chloride in water.
Both distilled water and 0.01 N NaCl gave infinite interlayer spacing
values (the clay was completely dispersed), while 1.0, 3.0, and 5.0 N
NaCl gave interlayer spacings of 0.93, 0.61, and 0.58 nm, respectively.
4.2.1.1.3 Temperature—An increase in temperature causes an increase
in double-layer thickness with a corresponding tendency toward dispersion.
However, the value of the dielectric constant for various fluids is also
affected by temperature, generally decreasing with increased temperature.
For water, the value of the product DT (Equation 4.3) is reasonably constant,
and temperature effects tend to cancel out.
4.2.1.1.4 Ionic Valence—The cation valence affects both the surface
potential and the thickness of the double layer. For solutions of the same
molarity and a constant surface charge, increasing the cation valence will
cause a decrease in the thickness of the double layer and a tendency toward
flocculation. It is also shown that an increase in valence will suppress the
midplane concentrations between parallel plates, leading to a decrease in
interplate repulsion (Mitchell, 1976, p. 122).
4.2.1.1.5 Size of Hydrated Ions—The smaller the size of the hydrated
ion, the closer it can approach the surface of the clay particle (Lambe,
1958). Thus, for a given cation valence, the thickness of the double layer
will tend to decrease with decreasing hydrated radii "of double-layer cations.
4.2.1.1.6 p_H~Changes in pH can affect the thickness of the double
layer in several ways. The electrolyte concentration as well as the net
negative charge on the clay particle are influenced by the solution pH. The
formation of stable suspensions or dispersions of clay particles often
require high pH conditions. There are two ways in which pH can change the
surface charge that results from chemical reaction at the surface of the clay
particles. First, a high pH can cause dissociation of hydroxyl groups at
the edge of clay particles, increasing the net charge and expanding the
double layer. The dissociation reaction is given below:
H2° - +
SiOH ——> SiO + H+ (4.5)
4-9
-------
The higher ths pH, the greater the tendency for the H+ to dissociate and
the greater the effective negative charge. Low pH discourages this
dissociation, lowering the surface charge, reducing the thickness of the
double layer, and promoting flocculation (Evans et al., 1981; Mitchell. 1976,
p. 127).
Second, pH affects alumina exposed at the edges of clay particles.
Alumina ionizes positively at low pH and negatively at high pH. Thus, in an
acid environment, positive double layers may develop at the edges of clay
particles with H+ serving as the potential determining ion (Mitchell,
1976, p. 126).
Changes in pH do not significantly affect surface charge resulting from
isomorphous substitution of the crystal lattice of the clay mineral. Thus,
clays with most of their surface charge attributed to Isomorphous substitu-
tion (e.g., smectites and mites)'are less affected by changes in pH than
the kaolinite minerals, which have most of their surface charge resulting
from surface chemical reactions (see Section 2.3).
4.2.1.1.7 Anion Adsorption—Adsorption of anions (e.g., Cl~s PO/p3
and certain surfactants) by the clay particle can increase the net negative
charge and increase the double-layer thickness. Dispersion can occur as a
result.
4.2.1.2 Displacement of Water—
If adsorbed water within clay particles is displaced by a fluid with a
different dielectric constant or electrolyte concentration, the result may be
a change 1n the interlayer spaces of the clay. Such changes can cause the
clays to shrink and crack. This can result 1n the formation of large
conducting channels through the soil along with drastic increases in the
permeability. Desiccation of clays by certain organic fluids has been
reported by Anderson (1981) and others.
4.2.1.3 Cation Exchange-
Cations adsorbed in the diffuse double layer are exchangeable with other
cations in solution. In general, the affinity of a cation for a clay
Increases with cation valence arid decreases with increasing ionic radius
within an element group. (See Section 2.2.)
Cation exchange will usually result in a change in double layer
thickness. The thickness of the double layer decreases with increasing
cation valence for montmorillonite; replacement of Na+ with Ca++ results
1n a reduction of interlayer basal spacing from over 4 nm to about 1.9 nm.
For montmorillonite, only two layers of water are incorporated between layers
when calcium is the adsorbed cation; with sodium, the number of water layers
between layers is practically unlimited. ; Thus, cation exchange can cause
double-layer collapse, desiccation and shrinkage, flocculation, and increased
permeability for certain clays.
Theng (1974) has summarized the Importance of cation exchange in
clay-organic complexes:
4-10
-------
. . . Since uncharged polar organic molecules are adsorbed essen-
tially by replacement of the interlayer water, the behavior of such
molecule's is likewise strongly influenced by the exchangeable cation.
Evidence is accumulating to show that, at least at low water contents,
cation-dipole interactions are of paramount importance in their effect
on the adsorption of polar organic species by clay materials.
4.2.2 Dissolution by Strong Acids or Bases
Both organic and inorganic acids react with and dissolve aluminum,
iron, and silica in the crystal lattice of clay minerals. Strong bases can
have a similar effect. This dissolution can result in a release of mineral
fragments that may migrate from their original position and leave enlarged
pore spaces for conducting the permeant fluid through the clay. Anderson
and Jones (1983) have pointed out that "whether the permeability of the
clay barrier increases or decreases will depend on the fate of the migrating
particles." The increase in conducting pore size may give rise to an
increased permeability if the particles migrate through the clay mass. If
the particles lodge in pore constrictions, clogging the conducting pore
spaces, a decrease in permeability may result (Anderson and Jones, 1983).
Data originally reported by Pask and Davies (1945) show that sulfuric
acid dissolves 3, 11, and 89 percent of the aluminum present in kaolinite,
illite, and smectite. Other studies also suggest that kaolinite is less
soluble than smectite in strong acids (Grim, 1953).
4.2.3 Precipitation of Solids
The precipitation of solids in the soil-water system is controlled by
ionic concentration and equilibrium solubility. If the concentration of
certain ions exceeds the solubility limits, minerals such as gypsum
(CaS04 2H20) or jarosite (KFe3(S04)2(OH)6) may precipitate from solution
(Shepard, 1981; Dunn, 1983). Formation of such precipitates can clog
pore spaces in the soil matrix.
4.2.4 Effect of Microorganisms
The presence of microorganims can affect the mobility of fluids
through the soil. Fuller (1974) classified activity-of microorganisms in
terms of three classes of chemical reactions: (1) oxidation and reduction,
(2) mineralization and immobilization, and (3) reactions with organic consti-
tuents. In certain circumstances, the attenuation of contaminants in soils
is drastically affected by microorganisms. Processes that can contribute to
attenuation of contaminants by microorganisms include the following (Fuller,,
1974; Dunn, 1983):
o Degradation of carbonaceous wastes
o Transformation of cyanide to mineral nitrogen and denitrifi-
cation to nitrogen gas
4-11
-------
• Oxidation-reduction reactions With metal ions
• Reduction of sulfate to sulfide
• Production of carbon dioxide from organic molecules
• Production of simple organic acids
• Production of humic and fumaric acids that can react with
trace contaminants
• Production of organic species on which trace contaminants
can be adsorbed.
Apart from the benefits of attenuation, the physical presence of the
microorganisms as well as the gases produced as a result of the reactions
listed above can cause blockage in the conducting pore spaces within soils,
reducing the area for flow and decreasing permeability. This effect has been
noted 1n permeability tests, particularly when the permeant fluid is
conducive to the growth of microorganisms.
4.3 MEASURING CLAY-CHEMICAL COMPATIBILITY THROUGH PERMEABILITY TESTING
Permeability is a highly variable engineering property of soils, and
slight changes in the measurement technique or test equipment can cause
order of magnitude changes in the values determined (Mitchell et al.,
1965). Because one is always dealing with orders of magnitude in perme-
ability testing, it is unrealistic to expect test results to agree within
less than several hundred percent (Zirnnie et al., 1981). Bryant and Bodocsi
(1986) have also examined precision and reliability that can be achieved in
laboratory permeability measurements. No widely accepted method exists for
measuring clay-chemical compatibility through permeability testing. Thus,
a variety of equipment and techniques have been employed in this area of
research. Some of the variables in test methods are described 1n this
section, and advantages and disadvantages are highlighted. For additional
information and recommendations see Bowders et al. (1986).
4.3.1 Measurement Devices
The permeability of clay soils to various liquids is usually deter-
mined in either constant head or falling head tests in fixed-wall, flexible-
wall, or consolidation permeameters. Each category of test device can have
many variations. Fixed-wall permeameters that have been used in clay-
chemical compatibility testing include stainless steel permeameters, plexi-
glass permeameters, thick-walled pyrex glass permeameters, glass cylinders,
PVC (polyvinyl chloride) tubes, and shrink tubing. Shrink tubing may, in
fact, lend to fixed-wall tests a major desirable feature of flexible-wall
tests—reduction in the possibility of sidewall leakage.
Tests with water and other solutions that do not interact with clays
have shown that comparable permeability measurements may be obtained
through different types of test devices. Everett (1977) measured the perme-
ability of Lacustrine clays in falling head tests over a 2-month test period
1n three types of fixed-wall permeameters;. Test results showed almost
4-12
-------
equivalent permeability readings in all test devices. Values measured
were 9.0 x IQj9 cm/s in a commercial metal permeameter, 6.9 x 10~9 cm/s
in PVC pipe, and 8.3 x 10~9 cm/s in shrink tubing.
Permeability tests performed in both fixed-wall and flexible-wall tests
at Duke University (Peirce, 1984; Peirce and Peel, 1985) also indicate that
comparable data can be obtained from the different devices. Test fluids used
in the Duke investigation did not significantly alter the soil permeabilities
as compared to baseline permeabilities obtained with 0.01 N calcium sulfate.,
More than 100 tests were carried out. The reproducibility of the permeabil-
ity results suggests that side-wall leakage in fixed-wall devices can be
virtually eliminated through careful quality control during sample prepara-
tion, compaction, and testing if the permeant fluid does not alter the clay
that is tested.
Test results may differ somewhat with different types of permeameters
when the test fluid is other than water (or comparable baseline fluid).
Daniel (1983) found that methanol in tests with kaolinite in three types of
permeameters produced similar curves (permeability versus pore volumes passed
through sample) but with compaction mold permeabilities somewhat higher than
values measured in flexible-wall or consolidation cells.
4.3.2 Test Setup
Whether permeability tests are carried out as constant head or falling
head should not affect the validity of the test results. Effectively, one is
doing the same thing in both tests. In the constant head test, readings are
converted to a flow rate; in the falling head test, readings are converted to
changes in head. Most of the permeability tests described in this section
are essentially constant head tests because of the very low permeabilities of
compacted clay soils. Test results reported by Monserrate (1982), however,
were computed as falling head permeabilities.
Buettner and Haug (1983) noted that leakage is the controlling factor on
how low a permeability can be measured. They measured total leakage from
their flexible-wall permeameter setup by placing a solid metal block in the
cell and monitoring the changes in inflow and outflow. Volume change in-
dicators with an accuracy of +0.005 cm3 were used to determine the leak-
age rate and also enabled the researchers to measure"the change in volume of
clay soils due to consolidation or swelling. Buettner and Haug found that
if leakage is not corrected for, the potential error at low permeabili-
ties (less than 10~12 cm/s) can be as high as 2 orders of magnitude.
4.3.3 Compatibility of Materials with Test Fluids
A problem that has become apparent in testing organic fluids in triaxial
cells is the incompatibility between the commonly used latex membranes and
the fluid to be tested (Acar et al., 1984a; Foreman and Daniel, 1984).
Excessive deformations (i.e., wrinkling and expansion) in the membrane may
occur with possible effects on the test results. One technique used by
4-13
-------
Acar et al. (1984a) to alleviate problems resulting from incompatibility was
to wrap the samples with two rounds of 0.03-mm sheet Teflon®. After place
ment around the wrapped samples, the membranes were coated with a
contaminant-resistant (silicon base) grease to decrease chemical diffusion.
Even with these provisions, however, acetone used as the test fluid was found
to diffuse through the membrane into the: cell water. A procedure and
apparatus that can be used to test the membrane-permeant fluid compatibility
have been described by Rad and Acar (1984).
Foreman and Daniel (1984) also found a workable technique to be wrapping
the soil sample along with the top cap and base pedestal with two or three
revolutions of 15-cm (6-inch) -wide Teflon® tape before placement of the
latex membrane. Long-term flexible-wall! tests with heptane have been carried
out successfully with this method.
Materials compatibility has also been a problem in fixed-wall tests with
corrosive test fluids. Stainless steel permeameters and pressure fittings
were damaged in tests with chronic acid (Monserrate, 1982). Several
researchers have used plexiglass devices with Teflon® fittings to avoid such
problems.
4.3.4 Effect of Backpressure
Slight deviations from full saturation have been shown to significantly
affect measured permeability values (Mitchell et al., 1965). Recognizing
this, some researchers include in the test method a saturation step involving
backpressurlng. Zimmie et al. (1981) noted that many permeability determina-
tions are made using backpressure to promote complete saturation and then
releasing or lowering the backpressure during the permeability test. When
this 1s done, dissolved gases immediately begin to come out of solution,
causing the measured permeability values to decrease. Zimmie et al» (1981)
concluded, "The maximum, fully saturated permeability value should be deter-
mined. It is necessary to utilize backpressure to properly saturate the
specimen whether or not the actual permeability test utilizes backpressure."
4.3.5 Effect of Hydraulic Gradient
Mitchell, Hooper, and Campanela (1965), in tests*with a silty clay,
found evidence that rapid infiltration of water could cause migration of fine
particles that tended to plug conducting pore spaces and reduce flow rates.
Mitchell and Younger (1967) measured permeability to water of a compacted
silty clay as a function of hydraulic gradient. A gradual increase in the
gradient from 0 to 17 over 26 days resulted in a corresponding increase
1n permeability from less than 5 pvp 10~7 to 5.4 x 10~6 cm/s. The in-
crease was approximately linear in the region of gradients of about 4 to 10,
the curve of permeability versus hydraulic gradient tending to flatten at
the lower and higher gradients. The nonlinear behavior at the higher gradi-
ents may have been the result of movement of fine soil particles. Other
researchers—Olsen (1965) and Hamilton (1979)—have reported data
4-14
-------
that show no significant effect of hydraulic gradient on either discharge
velocity or permeability to water., ,
Under a cooperative agreement with the U.S. Environmental Protection
Agency (EPA), Daniel (1982) investigated the effect of hydraulic gradient on
permeability. Water, (actually 0.01 N calcium sulfate) as well as other
permeant fluids were tested at gradients of 10, 50, 100, and 300 in flexible-
wall cells, fixed-wall permeameters, and consolidation cells. The perme-
ability of kaolinite to water did not vary with hydraulic gradient in any
test device. Changes in hydraulic gradient caused a slight increase in
permeability for kaolinite permeated with methanol in flexible-wall cells.
In fixed-wall permeameters and consolidation cells, however, the trend was
for decreasing permeability with decreasing gradient in tests with methanol.
Due to large variations in the test data obtained for fixed-wall permeam-
eters, the results of the tests were somewhat inconclusive.
In other research funded by the EPA, Brown, Thomas, and Green (1984)
studied the effects of hydraulic gradient on the permeability of three clay
soils to two organic wastes--a xylene waste and an acetone waste. It was
concluded that hydraulic gradients of 31, 91, and 181 did not greatly affect
the permeabilities in either presaturated or unsaturated samples.
4.3.6 Criteria for Concluding a Test
Clay-chemical compatibility tests should be continued until all changes
in permeability resulting from the interaction of the chemical with the clay
have been observed. To satisfy this condition, compatibility tests should be
concluded only if the slope of the permeability versus time curve does not
vary significantly from zero (steady-state permeability has been reached) and
at least one pore volume of fluid has passed through the clay. If the first
condition is not met, there is no assurance that reactions of the chemical
with the clay are complete. If the second condition is not met, there is no
assurance that the permeant has contacted all of the clay in the column. In
order to satisfy the above conditions, it is necessary to determine the
chemistry of both the influent and effluent (to detect breakthrough) and
report pore volumes through samples as well as real time. (Conclusion from a
workshop on Permeability Testing, Atlanta, GA, January 1984, unpublished.)
*
A statistical procedure was developed at Duke University (Monserrate,
1982; Peirce and Witter, 1986) to determine when a test should be concluded,,
Readings of column level and time are taken at certain intervals throughout
the test. The hydraulic conductivity is computed for each time interval.
From this set of data the first 10 points are taken and a linear regression
analysis is performed to determine the slope of the hydraulic conductivity
vs. time curve. The first point is then dropped and another value is added
on the other end. The slope is calculated again, and so on. In the
beginning of the test this slope will be fairly large, but as the test
progresses it will decrease and approach zero when steady-state is obtained,,
The criteria for when steady-state is obtained will be taken when two
criteria are met--
4-15
-------
t When the slope at the curve does not vary significantly from zero
at the 95-percent confidence level, and
t At least one pore-volume of the liquid is passed through the sample
(Peirce, 1984).
4.4 SUMMARY OF AVAILABLE RESEARCH DATA
Numerous studies pertaining to the effects of chemicals on clay soil
permeabilities have been carried out by researchers at universities and by
private firms across the country during the last few decades. Research has
been sponsored by the EPA, the Army Corps of Engineers, the Chemical
Manufacturers Association, and others. Among the classes of compounds that:
have been tested are aliphatic hydrocarbons, chlorinated aliphatics, aromatic
hydrocarbons, alcohols, glycols, ketones, carboxylic acids, amines, and
aromatic nitro compounds. In addition to tests with single-compound test
solutions, many studies have been carried out using complex chemical mix-
tures. In tests involving actual wastes or leachates (obtained by passing
water through wastes), the exact composition of the fluid is usually not
known although major components and important parameters are usually identi-
fied and quantified.
The results of some of the most significant permeability tests involving
specific organic solvents are presented in Table 4-1. The results of
permeability tests involving wastes are presented in Table 4-2.
Some of the studies summarized here and in more detail in Section 4.5
have shown that certain pure, concentrated organics can drastically alter the
permeability of clay soils under the conditions of the laboratory test.
These test results have led to widespread concern over the possible seepage
of wastes into the environment from clay-lined disposal facilities. It
should be emphasized that clay liners in disposal facilities are most often
in contact with leachate having much lower organic solvent concentrations
than the test solutions associated with the drastic permeability increases
seen in some laboratory studies. Laboratory studies performed with less
concentrated test solutions (either actual wastes or leachates or dilutions
prepared in the laboratory) do not appear to produce such effects.
Because of the wide variations in the soils used* in clay liners and the
leachates to which they will be exposed during service, testing of the actual
Uner/leachate system is necessary to confirm compability.
4.5 PERMEABILITY STUDIES TO INVESTIGATE CLAY-CHEMICAL INTERACTIONS
The major findings of various research efforts to investigate effects on
permeability of clay-chemical interactions are presented in this section.
4.5.1 Observations by Macey (1942) on Effects of Organics on Fireclay
As a result of permeability experiments with fireclay, Macey (1942)
concluded that the rates of flow of benzene, nitrobenzene, and pyridine
4-16
-------
TABLE 4-1. RESULTS OF PERMEABILITY TESTS WITH ORGANIC CHEMICALS
* ^r '
ALIPHATIC AND AROMATIC HYDROCARBONS
Heptane
Fixed-wall permeameter tests at high gradient with four clays showed rapid
permeability increases and breakthrough. The increase in permeability was
more than 3 orders of magnitude in an illite clay. Baseline permeabilities
were established in the same sample with 0.01 N calcium sulfate prior to
introducing the test solution. (Anderson, 1981)
Benzene
The flow rate of benzene through fireclay was of an enormously higher order
of magnitude than the flow rate for water. (Macey, 1942)
In a column test under a head of 701 cm (23 feet), signs of full penetration
throughout the clay material were observed after 36 days; in a similar column
with water as the permeant fluid the liquid level dropped less than 2.54 cm
(1 inch) over a 100-day period. Samples were 91 cm in height and 2.54 cm in
diameter. (White, 1976, unpublished data)
Following an initial decrease in permeability (compared to permeability to
deionized water established in a similar sample), total breakthrough occurred
on the eighth day of testing when Ranger shale was exposed to benzene in a
fixed-wall permeameter under low hydraulic gradient. (Green et al., 1979)
Benzene did not penetrate compacted Ca-montmorillonite that was first
saturated with 0.01 N calcium sulfate even at hydraulic gradients as high at
150. (Olivieri, 1984)
In flexible-wall tests with a Georgia kaolinite, permeability decreased until
the tests were terminated. The final value was approximately 2 orders of
magnitude lower than the initial permeability. (Acar et al., 1984a)
Xylene
*
Following an initial decrease in permeability, total breakthrough occurred on
the 25th day of testing when fireclay was tested in a fixed-wall permeameter
under low hydraulic gradient. In Ranger shale, a slight decrease in perme-
ability was observed and remained steady until the test was terminated at 40
days. In Kosse kaoline, a slight decrease in permeability was followed by an
increase to about the initial level, which persisted until the test was
terminated at day 36; changes did not exceed half an order of magnitude.
(Green et al., 1979)
In column tests under very low gradient, intrinsic permeabilities were higher
than permeability to water by at least 1 order of magnitude in samples of
Lake Bottom, Nicholson, Fanno, Chalmers, and Canelo clays. Permeabilities
~~(continued!
4-17
-------
TABLE 4.1 (continued)
Xylene (con.)
were slightly higher than for water in three other soils. The samples (5.8
cm in diameter and 5 cm in height) were presaturated with xylene. (Schramm,
1981)
Fixed-wall tests at high gradient with four clays showed permeability in-
creases and breakthrough followed by nearly constant permeabilities roughly
2 orders of magnitude higher than baseline permeabilities. Baseline perme-
abilities were established in the same sample with 0.01 N calcium sulfate
prior to introducing the test solution. (Anderson, 1981)
In fixed-wall permeameter tests at-high gradient, the permeability of an
unsaturated micaceous soil was 4 orders of magnitude higher when exposed to
xylene than when tested with 0.01 N calcium sulfate. (Brown et al., 1984)
Xylene/Acetone ;
In fixed-wall permeameter tests at high gradient, the permeability of an
unsaturated micaceous soil to either pure acetone or pure xylene was greater
than the permeability determined for mixtures of the two solvents. The
permeability of a mixture of 87.5 percent xylene and 12.5 percent acetone
was lower by 3 orders of magnitude than the permeability measured with pure
xylene (though still higher than the permeability to 0.01 N calcium sul-
fate). When the acetone component was increased to 75 percent, the perme-
ability was approximately the same as that determined with pure acetone
(i.e., about 1.5 orders of magnitude greater than the permeability to 0.01 N
calcium sulfate). (Brown et al., 1984)
Kerosene (a mixture of aliphatics and aromatics)
In column tests under very low gradient, intrinsic permeabilities were higher
than permeability to water by approximately 1 order of magnitude in samples
of Lake Bottom, Nicholson, Fanno, Chalmers, and Canelo clays. Permeabilities
were slightly higher than for water in three other soils. The samples (5.8
cm in diameter and 5 cm in height) were presaturated with kerosene.
(Schramm, 1981)
In fixed-wall permeameter tests at high gradient, the permeability of an
unsaturated micaceous soil increased by 3 to 4 orders of magnitude compared
to the permeability determined with 0.01 N calcium sulfate. (Brown et al.,
1984)
Naphtha
The permeabilities of two clays (Na-saturated and Ca-saturated montmorillon-
1te) to naphtha were greater by several orders of magnitude than their
permeabilities to water. (Buchanan, 1964)
(continued)
4-18
-------
TABLE 4-1 (continued)
Soltrol C (a light hydrocarbon liquid)
Intrinsic permeabilities for samples tested with the light hydrocarbon were
significantly higher than the permeabilities measured in similar samples
exposed to water, (van Schaik, 1970)
Diesel Fuel
Fixed-wall permeameter tests at high gradient with an unsaturated micaceous
illite produced highly variable data inconsistent with the pattern of perme-
ability changes seen with other liquid hydrocarbons. Permeability was
greater by 1 to 2 orders of magnitude than the permeability measured with
0.01 N calcium sulfate in a similar sample. (Brown et al., 1984)
Paraffin Oil
In fixed-wall permeameter tests at high gradient with an .unsaturated mica-
ceous illite, permeability was greater by about 1 order of magnitude than
the permeability measured with 0.01 N calcium sulfate in a similar sample.
Maximum values were obtained after the passage of one pore volume. (Brown et
al., 1984)
Gasoline
In fixed-wall permeameter tests at high gradient with an unsaturated mica-
ceous illite, permeability was greater by 1 to 2 orders of magnitude than
the permeability measured with 0.01 N calcium sulfate in a similar sample.
(Brown et al., 1984)
Motor Oil
In fixed-wall permeameter tests at high gradient with an unsaturated mica-
ceous illite, permeability increased by about 1 to 2 orders of magnitude as
2.5 pore volumes of fluid were passed through the sample. (Brown et al.,
1984) „
ETHERS
Dioxane
Kaolinite initially packed and permeated with water was permeated with
anhydrous dioxane until complete displacement of the water was achieved.
Replacement of water by dioxane was accompanied by about a 20- to 30-percent
increase in intrinsic permeability. This permeability was much lower,
however, than the values determined for kaolinite beds initially prepared
with dioxane. (Michaels and Lin, 1954)
(continued)'
4-19
-------
TABLE 4-1 (continued)
KETONES
i
Acetone ;
In fixed-wall permeameter tests under low hydraulic gradient, three clays
showed slight decreases 1n permeability (compared to permeability to
delonlzed water established in a similar sample). All tests were concluded
before 40 days. Less than 0.5 pore volumes were passed through the sample,,
(Green et al., 1979) ;
Fixed-wall permeability tests at high gradient with four clays showed initial
permeability decreases followed by increases compared to baseline. Baseline
permeabilities were established in-the same sample with 0.01 N calcium sul-
fate prior to introducing the test solution. Extensive shrinking and crack-
Ing in the soils were observed after permeation. (Anderson, 1981)
In flexible-wall tests with a Georgia kaolinite, an immediate decrease in
permeability was followed by an increase, the final value stabilizing at
approximately double the initial permeability (Acar et al., 1984)
Acetone (high and low concentration)
An Increase over baseline permeability (established with 0.01 N calcium sul-
fate in similar samples) was seen in an unsaturated micaceous soil for solu-
tions where the acetone concentration was 75 or 100 percent. Samples tested
with lower concentrations of acetone did not show appreciable changes in
permeability compared to the 0.01 N calcium sulfate. Tests were carried out
1n fixed-wall permeameters at high gradient. (Brown et al., 1984)
Acetone (low concentration)
Permeability decreased slightly 1n a Georgia kaolinite clay tested 1n a
flexible-wall permeameter with a solution containing a low concentration of
acetone (I.e., below 0.1 percent) prepared in 0.01 N calcium sulfate. (Acsir
et al., 1984a)
ALCOHOLS, GLYCOLS, PHENOL
Methanol
Permeability decreased slightly (compared to permeability to deionlzed water
established in a similar sample) when Ranger shale was exposed to methanol
under low hydraulic gradient in a fixed-wall permeameter. The test was
terminated after 30 days. (Green et al., 1979)
Fixed-wall permeameter tests at high gradient with four clays showed steady
permeability increases compared to baseline. Baseline permeabilities were
established in the same sample with 0.01 N calcium sulfate prior to introduc-
ing the test solution. Examination of the methanol-treated samples revealed
development of large pores and cracks. (Anderson, 1981)
(continued)
4-20
-------
TABLE 4-1 (continued)
Methanol (con.)
Test results with Lufkin clay in flexible-wall cells showed essentially no
change in permeability with time when samples were permeated with methanol.
The permeability to methanol was virtually the same as with 0.01 N calcium
sulfate. (Daniel, 1983)
At high hydraulic gradients, kaolinite was found to have a higher conduc-
tivity to methanol than to water regardless of the permeameter types; fixed-
wall, flexible-wall, and consolidation permeameters were used. In the
flexible-wall and consolidation permeameters, kaolinite is about twice as
permeable to methanol as to water. (Foreman and Daniel, 1984)
Isopropyl Alcohol
In column tests under very low gradient, intrinsic permeabilities were higher
than permeability to water by almost 1 order of magnitude in samples of
Nicholson, Fanno, and Canelo clays. Permeability values were the same or
slightly higher than baseline in five other soils. The samples (5.8 cm in
diameter and 5 cm in height) were presaturated with the alcohol. (Schramm,
1981)
Glycerol
Permeability decreased slightly (compared to permeability to deionized water
established in a similar sample) when Ranger shale was exposed under low
hydraulic gradient to glycerol in a fixed-wall permeameter. The test was
terminated after 36 days. (Green et al., 1979)
Ethylene Glycol
Fixed-wall permeameter tests at high gradient showed permeability decreases
compared to baseline followed by increases in three clays; a smectitic clay
showed an initial rapid increase followed by a slower but continuous increase
in permeability. Baseline permeabilities were established in the same sample
with 0.01 N calcium sulfate prior to introducing the test solution (Anderson,
1981)
In column tests under very low gradient, intrinsic permeabilities were an
order of magnitude lower than permeabilities to water in Chalmers clay,
Mohave clay, and River Bottom sand. Values were slightly lower in Lake
Bottom, Nicholson, Canelo, and Anthony clays and slightly higher in Fanno
clay. The samples (5.8 cm in diameter and 5 cm in height) were presaturated
with the ethylene glycol. (Schramm, 1981)
(continuedj
4-21
-------
TABLE 4-1 (continued)
Phenol (949 mg/L)
In column tests with a lacustrine clay (packed to discharge 2 ml/day), no
significant effect on permeability was noted when deionized water was
replaced by the phenol solution as the permeant fluid. (Sanks and Gloyna,
•Ly / / j
Phenol (low concentration)
Permeability decreased slightly in a Georgia kaolinite clay tested in a
flexible-wall permeameter with a solution containing a low concentration of
phenol (I.e., below 0.1 percent) prepared in 0.01 N calcium sulfate. (Acar
et al., 1984a) .
Phenol (high concentration)
In flexible-wall tests with a Georgia kaolinite and a high-strength phenol
solution, an immediate decrease in permeability was followed by an increase,
the final value stabilizing at approximately double the initial perme-
ability. (Acar et al., 1984a)
AMINES
Aniline
Fixed-wall permeability tests at high gradient with four clays showed perme-
ability increases and breakthrough. Baseline permeabilities were established
in the same sample with 0.01 N calcium sulfate prior to introducing the test
solution. Extensive structural changes in the upper half of the soil columns
were observed following permeation with aniline. The aggregated structure
was characterized by visible pores and cracks on the surface of the soils.
(Anderson, 1981)
Pyridine
«
The flow rate of pyridine through fireclay was of an enormously higher order
of magnitude than the rate of flow for water. (Macey, 1942)
CHLORINATED ALIPHATICS
Carbon Tetrachloride
Permeability decreases slightly (compared to permeability to deionized water
established in a similar sample) when Ranger shale was tested under low
hydraulic gradient in a fixed-wall permeameter. The test was terminated
after 14 days. (Green et al., 1979)
(continued)
4-22
-------
TABLE 4-1 (continued)
Trlchloroethylene
Permeability decreased slightly (compared to permeability to deionized water
established in a similar sample) when fireclay was tested under low hydraulic
gradient in a fixed-wall permeameter. The test was terminated after 36 days.
(Green et al.f 1979)
OTHER
Acetic Acid
Tests at high gradient in fixed-wall permeameters showed continuous perme-
ability decreases to baseline in two clays. Tests with smectitic and ill He
clays showed permeability increases after initial decreases. Baseline
permeabilities were established in the same sample with 0.01 N calcium
sulfate prior to introducing the test solution. The permeability decreases
were attributed to partial soil dissolution and migration of particles, which
temporarily clogged the fluid conducting pores. Progressive soil piping
eventually caused the increase in permeability. (Anderson, 1981)
Nitrobenzene
The flow rate of nitrobenzene through fireclay was of an enormously higher
order of magnitude than the flow rate for water. (Macey, 1942)
In flexible-wall tests with a Georgia kaolinite, permeability decreased until
the tests were terminated. The final value was approximately 2 orders of
magnitude lower than the initial permeability (Acar et al., 1984a)
4-23
-------
TABLE 4-2. RESULTS OF PERMEABILITY TESTS WITH WASTES
"Xylene Waste" (paint solvent containing xylene with 25 percent paint pig-
ments and trace amounts of water)
In fixed-wall permeameter tests, the permeability of three clay soils, pre--
saturated with 0.01 N calcium sulfate, increased rapidly upon exposure to
xylene waste after the cumulative flow exceeded 0.2 to 0.4 pore volume.
Permeabilities were 2 to 4 orders of magnitude greater than permeabilities
measured with 0.01 N calcium sulfate. Highest permeabilities measured on
Initially unsaturated samples were greater by 1 to 2 orders of magnitude than
for samples that were initially saturated with the calcium sulfate. (Brown
et al., 1983)
"Acetone Waste" (a chemical manufacturing waste containing 91.7 percent
acetone, 4 percent benzene, and 0.6 percent phenol)
In fixed-wall permeameter tests, the permeability of three clay soils, pre-
saturated with 0.01 N calcium sulfate, initially decreased (minimum perme-
ability at approximately 0.5 pore volume) and then steadily increased.
Permeabilities were 2 to 4 orders of magnitude greater than permeabilities
measured with 0.01 N calcium sulfate. Highest permeabilities measured on
initially unsaturated samples were greater by 1 to 2 orders of magnitude than
for samples that were initially saturated with the calcium sulfate. (Brown
et al., 1983)
Perchloroethylene Waste
There is evidence that a perchloroethylene waste, which formed a separate,
denser than water phase, contributed to the failure of a clay liner at a
surface impoundment. (Personal communication, 1984)
"Acid Prowl" (pesticide wash of very low pH; higher viscosity than water)
After several days of exposure to a lacustrine clay packed in a fixed-wall
column, the water reacted with the soil to produce chlorine gas. Over a
5-week period, the flow of liquid from the column was, irregular due to
clogging of pores by the gas. (Everett, 1977)
"Acid Wash" (42 percent sulfuric acid with about 5 percent organics: higher
viscosity than water)
Permeability of a lacustrine clay packed in shrink tubing increased by about
1 order of magnitude over a period of 19 ;days. The permeability was lower,
however, than values obtained when the soil sample was exposed to water.
(Everett, 1977)
~(continued!
4-24
-------
TABLE 4-2 (continued)
"Mother Liquor" fan acid wash with pH of 0.37; higher viscosity than water)
Permeability of a lacustrine clay packed in shrink tubing was lower than the
value obtained when the soil sample was exposed to water. The waste may have
reacted with the soil, liberating gases and increasing pore pressures and
clogging flow. (Everett, 1977)
"Hydrazo Benzene" (33 percent methanol, 12.8 percent sodium hydroxide, 15.5
percent sodium formate, and 1.5 percent hydroazobenzene and azobenzene; more
viscous than water)
Permeability of a lacustrine clay packed in shrink tubing was slightly lower
than the value obtained when the soil sample was exposed to water. Tests
were carried out for 34 days under low gradient (less than 100 cm).
(Everett, 1977)
"Acid Waste" (100 mM HC1/L)
Due to reaction with carbonates, much higher permeabilities were observed in
tests with lacustrine clay compared to permeabilities determined with
deionized water. (Sanks and Gloyna, 1977)
"Basic Waste" (100 mM NaOH/L)
Permeabilities in a lacustrine clay decreased compared to the values measured
with deionized water. (Sanks and Gloyna, 1977)
4-25
-------
through clay are "of an enormously higher order than for water." Analytical
reagent-grade_organics were used. All the organic liquids passed through the
clay at flow rates 100,000 to 1,000,000 times greater than the flow rate of
water through the same clay. The high rates of flow produced experimental
difficulties because of the high resistance to flow of the testing apparatus,
and the experiments were finally abandoned. Differences between the viscosi-
ties of the organic fluids and of water are not sufficient to account for the
very large permeability changes that were observed.
4.5.2 Tests With Kaolinite and Organic Solvents by Michaels and Lin (1954)
Michaels and Lin (1954) measured the permeability of high-purity
kaolinite to nitrogen gas, cyclohexane, acetone, dioxane, methanol, and
distilled water. Tests were also conducted to determine the effect of
replacement of one permeant liquid by another. The investigation was one
part of a broad program of fundamental research sponsored by the U.S. Army
Corps of Engineers and by industrial contributions and was carried out at the
Massachusetts Institute of Technology (MIT) Soil Stabilization Laboratory.
4.5.2.1 Test Method--
Permeability tests were conducted in a 5.1-cm (2-inch) -diameter fixed-
wall permeameter cell. Bed thickness was 1.1 to 1.8 cm. The base-exchange
capacity of the kaolinite was determined to be 5.0 + 0.5 meq/100 g.
Samples were dried for 24 hours at 160°C prior to saturation by the
organic liquids for the permeability test.
Prior to permeability testing, each sample was submerged in the fluid to
be tested (either organic liquid or water). The wetted samples containing
known amounts of clay were poured into the permeability cell, entrapped air
was removed, and the sample was consolidated by means of a confining piston.
Samples were consolidated to several void ratios (volume of voids/volume of
solid) for each test fluid. Void space was determined through column bulk
density and the true density of kaolinite.
For the permeability tests, a hydrostatic head not exceeding 10 percent
of the compaction pressure was maintained. Permeabilities were determined
for each void ratio.. Permeability coefficients determined for separate clay
samples prepared with a given fluid and compacted to essentially the same
void ratio were found to be reproducible to within 10-percent.
The effect of change of fluid medium on the permeability of a confined
clay was determined through a series of desolvation experiments. Kaolinite
initially packed and permeated with water was subsequently permeated with
dioxane until the water was completely displaced (as evidenced by constant
permeability to dioxane). The dioxane was then displaced with acetone under
similar conditions. Finally, warm, dry nitrogen was passed through the
sample until the bed was solvent-free. Beds initially tested with the
organic liquids (methanol, dioxane, acetone, and cyclohexane) were similarly
treated with dry nitrogen until all solvent was removed.
4-26
-------
The permeability to nitrogen of the desolvated samples was then
determined and compared to the permeability with the original permeant
fluid. J -
4.5.2.2 Test Results—
The intrinsic permeabilities (cm2) of samples packed at different
void ratios, measured for organic solvents and for nitrogen, are shown in
Figure 4-3. Predictably, the permeabilities increased with increasing void
ratio. The data also indicate that permeability at any given void ratio
decreases with increasing polarity of the permeant.
As a result of the desolvation experiments, the authors concluded that:
Whatever specific effects these permeant liquids exert on the kaolin-
ite persist when the clay is thoroughly dried in the confined state.
The reduced permeabilities of kaolinite to these liquids relative to the
values observed for this clay to gas when packed in the dry state
apparently cannot be ascribed to adsorbed liquid films, abnormally high
liquid viscosities, or electro-osmotic effects. ... It seems most
probable, therefore, that the lower permeabilities of kaolinite to these
liquids than to nitrogen are caused by improved dispersion, and possibly
more orderly packing of the solids in more polar media.
Results of the desolvation experiments show the effect of replacement of
permeant fluid on the permeability of the kaolinite. For samples sedimentecl
from and initially permeated with methanol, dioxane, acetone, or cyclohexane
and then exhaustively permeated with warm, dry, nitrogen gas, the final gas
permeability is found to be equal, within the limits of experimental accu-
racy, to the permeability to the initial permeant liquid at the corresponding
void ratio. This is particularly true at low void ratios. An increase in
permeability upon drying is observed if the initial void ratio is high.
Specific observations by Michaels and Lin include the following:
o Replacement of water by dioxane is, in all cases, accompanied
by about a 20- to 30-percent increase in permeability; the dioxane
permeabilities through these beds are, however, much lower than the
values determined for kaolinite beds prepared^ initially in dioxane.
o Replacement of the dioxane with acetone leads to a small additional
permeability increase approximately when water is the initial
permeant fluid.
o Displacement and evaporation of the acetone with dry nitrogen is
accompanied by a great increase in permeability; the magnitude of
this increase is greatest for kaolinite samples confined initially at
high void ratios.
4-27
-------
Permeants
x - Water
0 - N2
a - Cyclohexane
A - Acetone
A - Oioxane
+ - Methanol
2.4
Source: Michaels and Lin, 1954
Figure 4-3. Intrinsic permeabilities as a function of void space (e)
measured for different permeants.
4-28
-------
4.5.2.3 Discussion—
Intrinsic permeabilities reported by Michaels and Lin cannot be directly
compared with permeability data from other tests because the measurement pro-
cedure is unique.. No information is provided about the amount of fluid passed
through the sample in determining each of the reported permeabilities.
While samples used in permeability tests are usually air-dried, the
clay samples used by Michaels and Lin were dried for 24 hours at 160°C, a
condition that could cause irreversible changes in clays and organic matter.
It is possible that the adsorptlve properties of the kaolinite were
permanently altered by this drying procedure. The kaolinite mineral,
halloysite, will not rehydrate if the interlayer water is removed.
4.5.3 Study by Buchanan (1964) of the Effect of Naphtha on Montmorilignite
Buchanan (1964) at Texas A&M University studied permeabilities of sodium-
and calcium-saturated montmorillonite when these clays were permeated by water
and naphtha. Permeabilities of the two clays to naphtha were greater by
several orders of magnitude than their permeabilities to water, even with
reductions in void ratios for the samples treated with naphtha. The data also
suggest that the permeability of the sodium-saturated clay is more
dramatically affected by the naphtha. Permeability differences were thought
to be at least partially due to the inability of naphtha to form an
•immobilized liquid film on the clay mineral surfaces. Buchanan's data are
summarized in Table 4-3.
4.5.4 Study by Reeve and Tamaddoni (1965) of the Effect of Electrolyte
Concentration on Permeability of a Sodic Soil
Reeve and Tamaddoni (1965) of the U.S. Department of Agriculture studied
the effects of high-salt, high-sodium solutions on the permeability of a
highly sodic soil. California Waukena clay loam containing about 15 percent
of expanding lattice-type clay was tested in the laboratory and in the field.
4.5.4.1 Test Method-
Solutions of varying concentrations but with constant sodium adsorption
ratio (SAR) were used in the tests. Test solutions were prepared from sodium
chloride (NaCl) and calcium chloride (CaCl2) at SAR levels of 0 (100
percent CaCl2), 80 (a partially reclaiming solution),-180 (the equilibrium
solution of the natural soil), and <» (100 percent NaCl). For each SAR level,
seven ionic concentrations (63, 125, 250, 500, 1,000, 2,000, and 4,000
meq/L) were tested in a selected sequence. Laboratory tests were carried out
in fixed-wall 5.4-cm (2.125-inch) ID cylinders.
4.5.4.2 Test Results—
The test results indicate that permeability is a function of both the
absolute concentration and the SAR of the initial solution. Permeability
increased with increasing SAR value and ionic concentration • The
permeability-concentration relationship was found to depend markedly on the
initial solution concentration (i.e., if the initial solution concentration
4-29
-------
TABLE 4-3. VOID RATIO AND COEFFICIENT OF PERMEABILITY RELATIONSHIPS
-FOR CALCIUM- AND SODIUM-MONTMORILLONITE PERMEATED BY
WATER AND NAPHTHA3
Clay
Calcium-saturated
smectite
Sodium-saturated
smectite
Void
ratio
1.72
3.75
Water
Permeability
(cm/s)
1.6 x 10-9
5.2 x 10-11
Void
ratio
1.52
1.31
Naphtha
Permeability
(cm/s)
6.4 x 10-5
3.8 x 10-5
aData from Buchanan (1964).
TABLE 4-4. SUMMARY OF SOIL PERMEABILITY WITH SOLTROL C AND WATER3
Soil
(bulk density in g/cm
pi
i3) W,
Intrinsic
ermeabllity (urn2) Permeability. K (cm/s)
'ater
Oil
Water
Oil
Cavendish loamy sand 4.94
(bulk density:
1.44 g/cm3;
6 percent clay)
Chin sandy clay loam 0.73
(bulk density:
1.25 g/cm3;
18 percent clay)
Chin sandy clay loam 0.22
(bulk density:
1.38 g/cm3;
18 percent clay)
Lethbridge clay loam 0.51
(bulk density:
1.22 g/cm3;
37 percent clay)
9.38 4.8 x lO'3 3.0 x 10~3
5.45 7.1 x lO-4 5.1 x 10~3
2.53 2.1 x 10~4 1.4 x 10'3
6.10 5.0 x 10-4 3.3 x lO"3
aData from van Schaik (1970).
4-30
-------
was low, the range of permeability values with varying solution concentra-
tions was lower than if the initial solution concentration was high, and
vice versa.) Apparently the degree of flocculation or dispersion that
occurred with the application of the initial solution influenced the magni-
tude of the permeability.
In some of the tests a large change in solution concentration (e.g.,
from 125 to 1,000 meq/L) was accompanied by a large increase in outflow due
to sidewall leakage as a consequence of the contraction of the previously
swollen mass.
In the test series in which initially low concentrations (63 meq/L)
were increased stepwise to 4,000 meq/L and then reduced to the initial con-
centration, reproducibility of permeability values as a function of concen-
tration was good. These results suggested that a reversible swelling
process was occurring. In the test series in which initial concentrations
of 1,000 meq/L were increased to 4,000 meq/L and then reduced to 63 meq/L,
results showed a gradual decrease in permeability with a given solution
concentration. This was attributed to slaking or particle rearrangement
that led to permeability decreases that progressed with time.
The permeability of the Waukena soil to a saturated calcium sulfate
(CaS04) solution was an order of magnitude lower than the lowest values
measured with the test solutions.
Test results showed that intake rates measured in the field at high
electrolyte concentrations were approximately three times as great as
corresponding permeabilities measured in the laboratory. However, the
fractional change was essentially the same at equal electrolyte concentra-
tions.
4.5.5 Tests by van Schaik and Laliberte (1968) of Permeability of Soils
to a Liquid Hydrocarbon
Permeability data are reported by van Schaik (1970) for three soils
tested with tap water and with a light hydrocarbon liquid (trade name
Soltrol C). Permeability measurements were made on samples that had been
vacuum-saturated. The intrinsic permeabilities with .the oil were first
reported by van Schaik and Laliberte (1968).
4.5.5.1 Test Results—
The saturated intrinsic permeability of the three samples varied between
0.22 and 4.94 isnf- for water. For oil, the values ranged between 2.53 and
9.38 urn*. The data are summarized in Table 4-4.
4.5.5.2 Discussion—
The test apparatus used in the study was designed to measure properties
of unsaturated porous media. The number of pore volumes displaced by the
test fluid during the permeability determination was not reported. Details
of the test procedure used also were not given in the references cited. In
spite of these limitations, the intrinsic permeabilities are useful for
4-31
-------
comparing the relative permeabilities of the various soils to water and oil.
The Increased-permeabilities of the soils when exposed to the oil are con-
sistent with the later findings by Anderson (1981).
4.5.6 Study by Everett (1977) of Permeability of Lacustrine Clay to Four
Liquid Wastes-
Everett (1977) investigated the permeability of lacustrine clays from
Bay County, Michigan, to four liquid v/astes. The purpose of the study was to
establish the feasibility of a hazardous waste landfill in Bay County. Three
types of permeability test devices were used. The results of the comparative
tests with water show no significant differences in the permeabilities
measured in the different devices.
The predominant mineral in the soil ;was quartz, estimated to be 40 to 50
percent. Clay minerals present were illite and chlorite, each at about 10 to
15 percent. Other important constituents were dolomite (20 to 25 percent)
and calcite (10 to 15 percent). The clay fraction was reported to average
18.1 percent. The mean cation exchange capacity of the soil (determined with
ammonium acetate solution) was low.
Wastes used in the compatibility tests were "Acid Prowl," "Acid Wash,"
"Mother Liquor," and "Hydrazo Benzene." "Acid Prowl" is a pesticide wash of
very low pH. The "Acid Wash" was about 42 percent sulfuric acid 1n water and
also contained about 5 percent organics including dichlorobenzidine,
orthochloroaniline, and tars. The waste labeled "Hydrazo Benzene" actually
contained about 1.5 percent hydroazobenzene and azobenzene. Other components
were methanol (33 percent), sodium hydroxide (12.8 percent), sodium formate
(15.5 percent), and water (36.7 percent). The waste has a pH of 12.10. The
"Mother Liquor" was an acid wash (pH of 0.37) of unknown use. All waste
samples were more viscous than water.
4.5.6.1 Test Method-
Permeability to water was measured i:n three different test devices--
commercial Soil Test® units (fixed-wall, 10.2-cm [4-inch] diameter, 10.2-cm
[4-inch] sample height), PVC pipe (10.3-cm [4-inch] diameter, 15.2-cm
[6-1nch] sample height), and shrink tubing (10-2-cm [4-inch] diameter,
15.2-cm [6-1nch] length). Samples tested in the SoiLTest units and the
shrink tubing were prepared in a standard Proctor mold according to ASTM
specifications for modified Proctor compaction (D-1557-70). The objective of
preparing the samples was to obtain uniform packing. The PVC pipe was packed
1n a different manner, not specified.
Samples were saturated by backwashing with water for more than 2 weeks
before permeability data were recorded. Readings were taken over a 2-month
period to determine the permeability of the samples to water. After testing
with water, three shrink tube columns and the PVC pipe columns were tested
with wastes. Samples were allowed to saturate by backwashing with the waste
until the effluent showed the presence (by color or pH) of the waste.
4-32
-------
The tests were all conducted 1n falling head setups with a head of less
than 100 cm (39.37 inches). The metal test units were not used for testing
the wastes because of the corrosive nature of the wastes.
4.5.6.2 Test Results—
"Acid Prowl" was tested in the PVC column. After several days of con-
tact with the soil, .chlorine gas became noticeable at the outlet pipe. After
2 weeks, the gas was no longer apparent, and liquid flow from the permeameter
ceased. The column was allowed to stand with the waste fluid head for 3 more
weeks. During this time the fluid reacted irregularly. This behavior was
attributed to clogging of pores by chlorine gas generated with the sample.
Valid permeability readings could not be obtained.
"Acid Wash," "Mother Liquor," and "Hydrazobenzene" were tested in shrink
tubes. The permeabilities measured with the wastes were lower than the
corresponding permeabilities to water. This may be due in part to the higher
viscosity of the wastes compared to water. Values obtained after 19 days of
testing with "Acid Wash" showed a slight increase in permeability (about 1
order of magnitude) compared to the initial value. For "Mother Liquor," test
data with the waste showed permeability to be lower by more than 2 orders of
magnitude than values measured with water. These acid wastes may have
reacted with the soil-liberating gases, which would increase pore pressures
and consequently decrease the permeability. In tests with hydrazobenzene,
permeability values were slightly lower than values obtained with water.
Permeability data are compared in Table 4-5.
4.5.6.3 Discussion—
Although the compatibility tests were carried out for approximately 1
month, the volumes of liquid forced through the columns under the low gradi-
ents had to be very small. The number of pore volumes displaced was not
specified. Sufficient data were not presented to determine quantitatively
the effect attributable to the viscosity of the wastes.
4.5.7 Tests by Sanks and Gloyna (1977) of Permeability of Lacustrine
Clay to Liquid Waste
Sanks and Gloya (1977) tested five simulated liquid wastes with three
clays in column tests to determine effects on permeability. The aqueous
waste solutions tested were chosen to be representative of materials that
might leak from containers of solid wastes. The various substances contained
acid, base, and heavy metals. Clays tested contained large percentages of
montmorillonite. The columns were packed at densities low enough to dis-
charge 2 mL/day of distilled water at 152-cm (5-ft) head with an allowable
error of 10 percent. After the tests with water, the columns were drained
and refilled with the waste fluid to be tested.
Permeabilities with deionized water ranged from 4 x 1Q-8 to 1.2 x
10~7 cm/s. Much higher permeabilities were observed when the acid waste
(100 mM HC1/L) was passed through the columns due to reaction with carbo-
nates. Permeabilities decreased in tests with the basic waste (100 mM
NaOH/L). Phenol at 10 mM/L (940 mg/L) appeared to have little effect on
permeability.
4-33
-------
TABLE 4-5. PERMEABILITIES MEASURED WITH LACUSTRINE CLAY
EXPOSED TO WATER AND WASTE LIQUIDS3
Day of
test data
1
3
5
15
19
29
32
34
3.4 x 10-10
1.3 x 10-9
1.5 x 10-9
With
"acid wash"
7.1 x 10-11
No flow
2.9 x 10-10
5.8 x 10-10
6.9 x 10-10
Permeability (cm/s)
With water
6.7 x 10-7
6.5 x 10-7
5.7 x 10-7
With
"mother liquor"
9.0 x 10-10
1.0 x 10-9
8.3 x 10-10
2.1 x 10-9
3.1 x 10-10
4.0 x 10-7
6.0 x 10-7
5.1 x 10-7
With
"hydrazobenzene"
9.0 x lO-8
7.8 x 10-8
No reading
1.3 x 10-7
1.3 x 10-7
aData from Everett (1977).
4-34
-------
Heavy metals tested, HgCl2 at 30 mM/L (8,100 mg/L) and ZnS04 x 7H20
at 30 mM/L (8-,610 mg/L), showed no effect except for one anomaly.
4.5.8 Investigation of the Effect of Organic Solvents on Clays by Green.
Lee, and Jones (1979T
Under a grant from EPA's Kerr Environmental Laboratory, Ada, Oklahoma,
to the University of Texas at Dallas, Green, Lee, and Jones (1979) studied
the impacts of organic solvents on the shrink/swell behavior and permeability
of four clay soil materials—Ranger shale, fire clay, Kosse kaoline, and
Parker soil. The test results were compared to results with water. The
studies included extensive soil characterization. Soil properties are
summarized in Table 4-6. Organic solvents tested were benzene, xylene,
carbon tetrachloride, trichlorethylene, methanol, and glycol (Green et al.,
1981, 1983).
4.5.8.1 Test Methods--
Swell properties of the clays in contact with water, the organic
solvents, and various solvent mixtures v/ere investigated. The swell
properties were measured with a 6.4-cm-diameter consolidometer. Care was
taken to prevent evaporation. Clay core samples used in the tests were
compacted at optimum moisture content. In the consolidometer, samples were
flooded with the test fluid and measurements of the swell or shrink behavior
were made as a function of time.
Permeability data were collected on 15 clay/solvent systems. Test
durations ranged from 8 to 40 days. Fixed-wall permeameters us&d in the
study were thick-walled Pyrex glass, and all joints were Tefloi^-lined. The
test procedure was adapted from an ASTM method. Samples were compacted at
optimum moisture conforming to standard compaction procedures before transfer
to the permeameters. The test fluid was then introduced, the liquid level
being adjusted in an 8-mm graduated standpipe. Equilibrium permeabilities
were estimated from curves of K(cm/s) versus time. The number of pore
volumes of permeant fluid passed through the samples was not reported.
4.5.8.2 Test Results--
Each of the clay soils swelled to a greater extent when exposed to
deionized water than with any of the pure solvents tested. The shrink/swell
behavior of the various clays in water and in organic solvents is described
in Table 4-7. The final percent swells for the samples are,listed in Table
H~O «
Based on the swelling data for both pure solvents and for mixtures, the
authors concluded that "in mixtures of solvents, a clay will tend to swell as
though it were immersed in the component of higher dielectric constant only"
(Green et al., 1979).
When Ranger shale was presented with benzene, breakthrough occurred on
the eighth day of testing. Breakthrough occurred at about day 25 in the fire
clay sample presented with xylene.
4-35
-------
TABLE 4-6. PROPERTIES OF SOILS TESTED*
Clays
Properties
Particle size distribution
(wt. %)
Clay
Silt
Sand
Total carbon (%)
Carbonate (%)
Cation exchange capacity
Atterberg limits
Plastic limit (moisture %)
Liquid limit (moisture %)
Optimum moisture content (%)
Corresponding dry density
(g/cm3)
Mineralogy (% of clay
fraction)
Kaolinite
Quartz
11 lite/mica
Chi ori te-montmori 11 oni te
Ranger
shale
40
59
1
0.60
0.32
54.4
36
46
17.5
1.73
24
28
24
10
Fire
clay
44
55
1.5
0.03
0
11.2
31
32
16
1.81
78
16
10
Negligible
Kosse
kaoline
53
47
0
0.12
0
13.4
38
50
31
1.36
85
10
5
Negligible
Parker
soil
10.5
70.5
19
3.18
0.42
14.1
22
26
18
1.86
14
Negligible
37
49
aData from Green et al., 1979.
4-36
-------
TABLE 4-7. CLASSIFICATION OF CLAY-ORGANIC SOLVENT SYSTEMS
ACCORDING TO SWELL PROPERTIES3
Swelling Swelling then
only shrinking
RS/H20 RS/acetone
RS/glycerol KK/xylene
RS/methanol FC/acetone
RS/CC14 FC/TCE
RS/TCE FC/xylene (NS)
KK/water
KK/acetone
FC/water
RS = Ranger shale
FC = Fire clay
NS = Net shrinkage (net swell
observed unless indicated
otherwise)
Shrinking then
swelling
RS/benzene
KK/TCE
FC/benzene
Shrinking
only
RS/xylene (NS)
KK/CC14 (NS)
FC/CC14 (NS)
-
KK = Kosse kaoline
TCE = Trichloroethylene
Reproduced from Green, Lee, and Jones, 1979.
4-37
-------
TABLE 4-8. PERCENT SWELL FOR CLAY SOILS IN CONTACT
WITH ORGANIC LIQUIDS AND WATERa
Clay-soil
Ranger shale
Kosse kaoline
Fire clay
Solvent
Benzene
Benzene/acetone (3:l)c
Xylene
Carbon tetrachloride
Trichloroethylene
Acetone
Acetone/benzene (3:l)c
Acetone/water (l:l)d
Methanol
Glycerol
Water
Xyl ene
Acetone
Water
Xylene
Carbon tetrachloride
Acetone
Water
Percent swell"
0.05
5.75
-0.11
1.1
1.0
4.0
4.6
11.0
11.4
5.3
11.7
0.16
8.7
11.7
-0.25
-0.6
3.6
8.2
aAdapted from Green et al., 1979.
bNegative value indicates net shrinkage.
cMole percent.
dVolume percent.
4-38
-------
For organic solvents other than benzene, a decrease in permeability of
Ranger shale .was observed for 5 to 10 days. This decrease was followed by
stable readings at a minimum value. The permeability coefficients reported
are shown in Figure 4-4. Although there is some scatter in the measurements
characteristic of normal experimental error, the decreasing permeability
trends over the duration of the test period are evident. The largest
decrease in permeability compared to water is seen with glycerol. With
carbon tetrachloride, the decrease in permeability was followed by an
increase that continued to the end of the test period.
4.5.8.3 Discussion--
The authors based their final conclusion, an empirical relationship for
estimating the coefficient of permeability from dielectric constant of the
permeant and packed bulk clay density, on what they judged to be final per-
meabilities. Their data do not indicate that stable permeabilities were
reached, however, in all of the solvent clay systems tested. Certainly the
permeabilities measured just prior to breakthrough could not be classified as
stable, yet these were used in obtaining the empirical equation to predict
permeability.
4.5.9 Anderson's Study (1981) of the Effects of Organics on Permeability
Anderson (1981) at Texas A&M University studied seven organic fluids in
comparative permeability tests. Four native clay soils were tested—two with
predominantly montmorillonite (smectitic) clay minerals but different chemi-
cal properties, one with predominantly kaolinite minerals, and one with
predominantly illite. Each soil contained a minimum of 35 percent by
weight clay minerals and exhibited a baseline permeability of less than
1 x 10~7 cm/s when compacted at optimum water content. Organic solvents
tested were reagent grade ethylene glycol, acetone, heptane, xylene (mixed
isomers), aniline, glacial acetic acid and methanol. The control fluid used
to establish the baseline permeability for each soil was a standard aqueous
solution of calcium sulfate (0.01 N CaS04). (Although it was not noted
in the initial publication of this research, the methanol used in the
experiment contained 20 percent water [Anderson 1982].)
Characteristics of the four soils used in the tests are presented in
Table 4-9. The percentage of soil voids filled withjvater at compaction at
optimum moisture content was approximately 75 percent for the two smectitic
clays and approximately 90 percent for the other two samples. The minimum
permeability for each sample occurred at or just above the optimum moisture
content (Anderson et al., 1981; Brown and Anderson, 1983).
4.5.9.1 Test Method--
Permeability was measured in constant head tests on soil cores compacted
at or above optimum water content. Pressurized compaction permeameters with
an air-induced elevated hydraulic gradient were used. For two montmoril-
lonite clays, a gradient of 361.6 (equivalent to a hydraulic head of 42.2 m
of water) was used. For the illite and kaolinite clay soils, a hydraulic
gradient of 61.6 (equivalent to a hydraulic head of 7 m of water) was
imposed. No signs of particle migration or turbulent flow resulted from the
elevated gradients.
4-39
-------
+ Carbon Tetrachlorfde
o Xylene
A Trichloroethylene
• Deiom'zed Water
0 Glycerol
x Acetone
A Methano1
50 i
£52; Carbon Tetrachlorid
10
30
15 20 25
Time (day)
Source: Green, Lee, and Jones, 1979
Figure 4-4. Coefficient of permeability of Ranger shale to various chemicals.
25
4-40
-------
TABLE 4-9. GRAIN SIZE DISTRIBUTION, MINERALOGY, AND PROPERTIES
OF THE EOUR CLAY SOILS*
Clay soil
description
% Sand (>50 nm)
% Silt (50-2.0 nm)
% Clay (<2.0 nm)
Coarse clay (2.0-0.2 nm)
% of total
Mineralogy'3
Fine clay (<0.2 nm)
% of total
Mineralogy0
Cation exchange capacity
(meq/100 g)
Total alkalinity (meq/100
Fe203 (%)
Organic matter (%)
CaC03 Equiv. (%)
Shrink-swell potential
Liquid limit
Plasticity index
Optimum water content0
Maximum density (kN rrr3)
Non-
calcareous
smectite
35-37
26-28
36-38
16
QZ-1
KK-2
MI-2
84
MT-1
24.2
g) 3.3
0.42
0.9
-
Very high
51-67
30-45
20.0
15.0
Calcareous
smectite
7-8
42-44
48-50
25
MT-1
KK-2
QZ-3
75
MT-1
KK-3
36.8
129.2
0.2
3.0
33
Very high
*
58-98
34-72
21.5
14.4
Mixed
cation
kaolinite
39-41
17-18
42
33
KK-1
QZ-2
67
KH-1
MT-3
8.6
0.8
13.2
0.6
Trace
Moderate
41-60
18-30
20.0
16.3
Mixed
cation
illite
14-15
38-39
47
61
1-1
QZ-2
39
1-1
MT-2
18.3
4.2
-
-
-
Moderate
-46
-27
19.0
16.6
aAdapted from Anderson, 1981.
DKey to mineralogy: MT =
KK =
I =
Smectite 1
Kaolinite 2
Illite 3
= >40% QZ
= 10-40% MI
= <10% KH
= Quartz
= Mica or il
= Halloysite
lite
cPercent by dry weight.
4-41
-------
Special precautions were taken to minimize several sources of error that
occur frequently 1n permeability testing. At least one pore volume of the
CaS04 solution was passed through the soil cores to minimize trapped
air. Before the permeameter was pressurized, the top of the soil sample
was exposed to 10 cm of standard CaS04 solution for 48 hours. This
procedure helps to prevent channel formation and bulk flow and promotes
sealing of the permeameter sidewalls. Provision was made to view the efflu-
ent passing through the soil sample so that trapped air or evidence of piping
could be monitored. Care v/as taken to minimize evaporative losses from the
collected effluent. Teflon fittings were used wherever possible to avoid
deterioration from exposure to the organic solvents.
After seating the soils at low pressure, the selected gradient was
applied to the permeameter fluid chamber .until stable permeability values
were obtained with the standard CaS04 solution. At this point, pressure
was released and permeameters were disassembled. Soil that had expanded
out of the mold was removed and additional standard leachate was passed
through the remaining sample in the permeameter. Percent swelling was deter-
mined for each soil. At this point, the soil sample was assumed to be com-
pletely saturated.
After the saturation step to ensure stable baseline permeability, the
remaining standard CaS04 solution was removed and replaced with one of
the organic solvents to be tested. The selected gradient was imposed and
between 0.5 and 2.0 pore volumes of fluid were passed through the sample 1n
the permeameter. The permeameters were then depressurized, disassembled, and
the cores removed and examined for evidence of structural changes. Following
the permeability tests of the noncalcareous smectite clay soil with aniline,
methanol, and ethylene glycol (all highly water-soluble), water was reintro-
duced to the test columns and permeabilities were again determined.
During the permeability tests, effluent was carried to an automatic
fraction collector, which separated the samples obtained from 10 permeameters
during specific time increments. The percentage of organic fluid in the
effluent samples was determined to develop breakthrough curves. A permeabil-
ity value, leachate volume, and time increment were obtained on each volume
of fluid passed through a sample.
4.5.9.2 Test Results--
Baseline permeability (two pore volumes) for the four soils is depicted
1n Figure 4-5. Permeability data obtained for the various organlcs tested
are presented 1n Figures 4-6 through 4-13 (Anderson, 1981; Anderson et al.,
1981; Brown and Anderson, 1983). Findings as described by the authors are
excerpted below:
Acetic Acid (glacial)—All four clay soils permeated with acetic acid
showed initial decreases in permeability . . . thought to be due to
partial dissolution and subsequent migration of soil particles. These
migrating particle fragments could lodge in the fluid-conducting pores,
thus decreasing cross-sectional area available for fluid flow.
4-42
-------
o
-J
-------
NCNCALCAREOUS SMECTITE
CALCAREOUS SMECTITE
MIXED CATION KAOLJN1TE
MIXED CATION I LUTE
ACETIC AGO
as to 1.5
PORE VOLUMES
Source: Anderson, 1981
Figure 4-6. Permeability of the four clay soils to acetic acid.
4-44
-------
100-
LU ^ 4-
s B t
-»-!-» ••
I
LU
NCNCALCAREOUS SMECTITE
CALCAREOUS SMECTITE
MIXED CATION KAOLJNITE
MIXED CATION ILL1TE
A
o
0.5
ao
as LO 1.5
PORE VOLUMES
2.0
2.5
Source: Anderson, 1981
Figure 4-7. Permeability and breakthrough curves of the four clay
soils treated with aniline.
3.0
4-45
-------
NONCALCAREQUS SMECTITE
CALCAREOUS • SMECTITE.
MIXED CATION KAOUNITE
MIXED CATION I LUTE
ETHYLENE GLYCOL
0.5 L'0. 1.5
PORE VOLUMES
Source: Anderson, 1981
Figure 4-8. Permeability of the four ciay soils to ethylene glycoi.
4-46
-------
NCNCALCAREOUS SMECTITE
CALCAREOUS SMEC
MIXED CATION KAOUNITE
MIXED CATION ILL1TS
ACETONE
IO
as to 1.5
PORE VOLUMES
Source: Anderson, 1981.
Figure 4-9. Permeability of the four clay soils to acetone.
4-47
-------
<
'LU
100-
B ::
LU
!z -
f
i
LU
NCNCALCAREOUS SMECTITE
CALCAREOUS SMECTITE
MIXED CATION* KAOUNITE
MIXED CATION 1LLJTE
METHANOL
A
O
10
CL5
ao
as to' 1:5
PORE VOLUMES
2.0
2.5
3.0
Source: Anderson, 1981
Figure 4-10. Permeability of the four clay soils to methanol and the breakthrough
curve for the methanoi-treated mixed cation illitic clay soil.
4-48
-------
as ao as to i.s
PORE VOLUMES
2,0 2.5
3.0
Source: Anderson, 1981
Figure 4-11. Permeability of the methanol-treated mixed cation illitic clay soil
at two hydraulic gradients.
4-49
-------
NCNCALCAREOUS SMECTITE
: CALCAREOUS SMECTITE
MIXED CATION KAQLiNITE
MIXED CATION I LUTE
A
O
XYLENE
0.5 :|.0 1.5
PORE VOLUMES
2.0
2.5
3.0
Source: Anderson, 1981
Figure 4-12. Permeability and breakthrough curves of the four clay
soils treated with xylene.
4-50
-------
100-
NCNCALCAREOUS SMECTITE
CALCAREOUS SMECTITE
MIXED CATION KAOUNITE
MIXED CATION I LUTE
HEPTANE
0.5
ao
0.5 10 1.5
PORE VOLUMES
2.0
Z.5
3.0
Source: Anderson, 1981
Figure 4-13. Permeability and breakthrough curves of the four ciay
soils treated with heptane.
4-51.
-------
Two of the soils treated with acetic acid (calcareous smectite and mixed
cation kaolinite) showed continuous permeability decreases throughout
the test period. After passage of approximately 20 percent of a pore
volume, the acid treated kaolinitic clay generated a dark red-colored
effluent . . . probably due to dissolution of iron oxides. The acid
treated calcareous smectite began passing cream-colored foamy effluent
after passage of about 28 percent of a pore volume. . . . The creamy
material was probably dissolved calcium, while the foam was the result
of C02 liberation from the dissolved carbonates.
Both noncalcareous smectite and the mixed cation illite eventually
showed permeability increases after . . . passage of 39 percent and 62
percent of a pore volume, respectively. . . . Permeability increases
with both of these soils were probably due to progressive soil piping
that eventually cleared initially clogged pores.
Aniline—Both noncalcareous smectite and mixed cation ill He had
breakthrough of aniline with concurrent permeability increases at pore
volume values (below 0.5). . -. . Aniline broke through after passage of
one pore volume for the kaolinitic soil. The calcareous smectite . . .
permeability increased rapidly at first, but showed substantial de-
crease concurrent with aniline breakthrough. After the permeability
decrease, this soil exhibited a slow but steady permeability increase.
There were no signs of migrating soil particles in any effluent
samples. . . . Apparently, aniline is too weak a base to cause signifi-
cant dissolution of clay soil components. However, . . . the organic
base caused extensive structural changes in the upper half of the soil
cores. The massive structure of the four soils . . . was altered by
aniline into an aggregated structure characterized by visible pores and
cracks in the surface of the soils.
Ethylene Glycol— . . . Permeability values indicated that it was the
ability of ethylene glycol to alter ;the soil fabric that was the domi-
nating influence on permeability. Three of the clay soils treated with
ethylene glycol showed initial permeability decreases. The kaolinitic
clay soil continued to undergo permeability decreases as long as it was
being tested. The illitic clay soil began showing a permeability in-
crease after passage of 0.5 pore volumes. In contrast, the calcareous
smectite followed its initial permeability decrease with a substantial
increase, a second decrease, and finally reached a nearly constant value
that continued until the end of the test period.
The noncalcareous smectitic clay soil treated wfth ethylene glycol
showed an initial rapid increase in permeability and a slower but con-
tinuous increase after passage of 0.5 pore volume.
Acetone—All soils treated with acetone had initial permeability de-
creases. These decreases continued until passage of approximately 0.5
pore volume. During passage of the next 0.5 pore volume, however, the
soils underwent large permeability increases. One possible explanation
for this sequence of permeability changes is as follows:
1. The higher dipole moment of acetone caused Initial Increase in
Interlayer spacing between adjacent clay particles as compared to
water only.
2. As more acetone passed through the soil cores, more water layers
were removed from clay surfaces. Due to its larger molecular
weight, however, fewer acetone layers were adsorbed than had
adsorbed when water was the only fluid present. This resulted 1n a
larger effective cross-sectional area available for fluid flow.
4-52
-------
While acetone can displace water from clay surfaces due to its higher
dipole moment, it cannot fgrnKas many adsorbed fluid layers as water due
to its higher molecular weight. '
Examination of the soil after acetone treatment showed extensive shrink-
age and cracking. Such soil shrinkage is usually associated with de-
hydration, indicating that acetone had extracted water from soil
particle surfaces.
Methanol—Unlike soils treated with acetone, methanol-treated soils
underwent no initial permeability decrease.
Percent methanol in the effluent from the illitic clay soil paralleled
an increase in permeability of the soil. After passage of 1.5 pore
volumes, the hydraulic gradient was reduced from 61.1 to 1.85 and
another pore volume of methanol passed. After an initial decrease,
permeability of the soil steadily increased at the lower hydraulic
gradient. ... No particle migration was detected in effluent from
methanol-treated cores, and therefore soil piping was discounted as
a mechanism for observed permeability increases. Examination of
methanol-treated soil cores revealed development of large pores and
cracks visible on the soil surface. The lower dielectric constant of
methanol may have caused a decrease in interlayer spacing of the clay
minerals present in the soils and thereby promoted the structural
changes.
Xylene—Xylene-treated soils showed rapid permeability increases fol-
lowed by nearly constant permeabilities roughly two orders of magnitude
greater than their permeabilities to water. Permeability increases
may be caused by ... structural changes in the xylene-treated soils,
exemplified by massive structure before treatment and blocky structure
after the soils were treated with xylene.
Heptane—Permeabi 11 t.v patterns for the heptane cores closely approxi-
mated those shown by the xylene-treated cores (i.e., large initial per-
meability increases). Following these initial large increases, rate of
permeability increase slowed until nearly constant permeability values
were observed.
Only the calcareous smectitlc clay showed . . . permeability to heptane
well below its permeability to xylene. The constant permeability values
eventually reached by the neutral nonpolar treated cores were probably
related to the limited ability of these fluids to penetrate interlayer
spaces of the clay minerals. ,
4.5.9o3 Reintroduction of Water--
When the standard calcium sulfate solution was reintroduced to the non-
calcareous smectite test columns permeated with aniline, methanol, and
ethylene glycol, a subsequent decrease in the permeability (roughly 1 order
of magnitude) was observed. The final permeability (after one pore volume of
standard solution) remained well above the baseline permeability measured
prior to introduction of the organic solvents (Anderson, 1981).
4.5.9.4 Discussion-
Viscosity and density differences between the organic fluid and water
do not account for the large changes in permeability observed 1n the tests.
The data from these permeability tests illustrate the importance of passing
multiple pore volumes of fluid through the soil samples in order to determine
the effects of solvents on permeability.
4-53
-------
The work by Anderson has been criticized as being "unrepresentative of
real conditions" because pure organic solvents (rather than waste leachate)
were used in the permeability tests (Gray and Stoll, 1983). The test results
stand, however, as a demonstration of potential effects on clay liners should
they come in contact with certain types of undiluted organic liquids.
The high gradients used in the tests have also been questioned (Zoeller,
1982; Gray and Stoll, 1983) since it is unlikely that gradients as high as
361 would be encountered in the field. The use of elevated gradients in
laboratory testing in order to reduce testing time is not uncommon. Although
the gradients used by Brown and Anderson were higher than those used by most
researchers, the permeability changes observed cannot be attributed to the
elevated gradients. Conditions were maintained well within the laminar flow
regime (I.e., Reynolds number was below 10) so that Darcy's Law for computing
the permeabilities remained valid. In general, the trends observed in these
tests are consistent with data from other researchers, and explanations can
be linked to the findings from clay-chemical complex studies.
Complete saturation in the compaction permeameter without backpressure
cannot be ensured. It is possible that air introduced with the permeant
fluids could have influenced the test results to some extent, but probably
by no more than 1 order of magnitude.
Finally, the use of fixed-wall permeameters for these studies has been
criticized (because of the potential for sidewall leakage) by advocates of
trlaxlal test methods. As there are advantages and disadvantages with either
method, the argument as to which test is most meaningful is finally a matter
of opinion.
Acetic acid, a weak organic acid, has a dissociation constant of
1.75 x 10-b in aqueous solution (pKa = 4.75) at 25°C. Since glacial
acetic acid was used, the extent of ionizatlon was a fraction of this. The
permeability decreases observed probably would have been followed by sharp
increases In permeability due to dissolution of the basic soil components if
a dilution had been used rather than glacial acetic acid. Progressive
soil piping, as pointed out by Brown and Anderson, follows the soil dissolu-
tion and eventually results in large pores with an associated increase in
permeability.
«
This effect (i.e., increased permeability) is to be expected with
carboxylic adds or sulfonic acids in general. This is of particular
significance since carboxylic acids are among the biological degradation
products of a wide range of organic compounds. Soils that are high in
carbonates are most affected by acids (even weak acids such as acetic acid),
but other soil constituents such as Fe£03 are also susceptible.
4.5.10 Schramm's Study (1981) of the Permeability of Soil to Organic
Solvents
Schramm (1981) evaluated permeabilities for eight soils tested with
kerosene, Isopropyl alcohol, ethylene glycol, and mixed xylenes.
Permeability with water was also measured. A total of 211 column tests were
run to determine saturated permeabilities to the various test fluids. The
research was performed at the University of Arizona.
4-54
-------
Clay content in the soils ranged from 1 percent in River Bottom sand to
70 percent in-Lake Bottom clay. Characteristics of the soils are presented
in Table 4-10. ' •** -* '
4.5.10.1 Test Method--
Permeabilities were obtained in glass or plexiglass cylinders through
the falling head method. Cylinders 5.8 cm in diameter were used in the tests
with soil packed to a depth of 5 cm. Soil was retained in the cylinders by a
galvanized screen glued to the cylinder. Prior to testing, samples were
immersed overnight in the test solvent. During the test, the height of the
solvent column above the soil surface varied between 6 and 2 cm. Six
replicates were tested for every solvent-soil combination. Experiments were
repeated until a constant permeability value was obtained. Length of runs
varied from 1-1/2 minutes to several weeks. An analysis of variance was
performed to evaluate differences between intrinsic permeability for the
various soils and solvents.
4.5.10.2 Test Results—
A summary of the saturated permeability values (expressed in cm/h)
obtained from these tests is given in Table 4-11. It is notable that with
each soil tested the highest permeabilities were obtained with the non-
polar organics, kerosene, and xylene. (Kerosene is a mixture of Ci2-Ci«
compounds and includes both aliphatics and aromatics.)
Intrinsic permeabilities were also calculated to take into account
differences in viscosity and density for the various permeant fluids.
Analysis of the intrinsic permeability values Included a one-way variance
test to determine if the intrinsic permeability varies significantly with
solvents or with soils. The intrinsic permeabilities for a given soil are
expected to be constant unless there is interaction between the soil and the
permeating liquid. The differences in intrinsic permeability with the
different solvents are shown in Figure 4-14 for five soils. The trend for
increased intrinsic permeability is readily apparent.
The analysis also shows rather uniform behavior of intrinsic perme-
ability for solvents in the different soil types, the Fanno, Mohave, and
River Bottom Sand having highest relative intrinsic permeabilities with all
organic solvents.
From the data summary and analysis of variance, Schramm (1981) concluded
that:
Calculated values of intrinsic permeability vary for the same soil
depending on solvent. However, the variation is relatively minor com-
pared to variation due to differences in soil properties among several
soils. ...
Since intrinsic permeability values vary for the same soil depending on
solvent, the soil and solvents are interacting.
4-55
-------
TABLE 4-10. CHARACTERISTICS OF SOILS USED IN PERMEABILITY TESTS3
Soil name
Lake Bottom clay
Nicholson
Fanno
Chalmers
"f1 Canelo
en
O) .
Anthony
Mohave
River Bottom sand
Soil
order
Entisol
Alfisol
Alfisol
Mollisol
Alfisol
Entisol
Aridisol
Entisol
Electrical
Cation conductivity
Soil exchange of saturated
Clay Silt Sand paste capacity extract
% % % pH (meq/100 g) (emho/cm)
70.6 24.0 5.4 7.7 34.7
-
49.0 47.0 3.0 6.7 37.0
46.0 19.0 35.0 7.0 33.0
31.0 52.0 14.0 6.6 22.0
28.0 28.6 43.4 5.6 5.76
15.0 14.0 71.0 7.8 10.0
11.1 37.0 52.0 7.3 10.0
1.0 2.0 97.0 7.2 2.0
t
1,111
176
392
288
240
328
615
210
Column
bulk
density
(g/cm3)
1.52
1.53
1.48
1.53
1.72
.1.87
1.78
1.80
Soil
surface Predominant
area cl ay
(m^/g) minerals
142.0 11 lite
Kaolinite
120.5 Vermiculite
122.1 Montmorillonite
Mica
95.6 Montmorillonite
Vermiculite
35.0
49.8 Montmorillonite
Mica
38.3 Mica
Kaolinite
3.6 Kaolinite
Mica
aAdapted from Schramm (1981).
-------
TABLE 4-11. PERMEABILITY COEFFICIENTS (cm/s) DETERMINED IN SOILS TESTED WITH
ORGANIC SOLVENTS9
CJl
•vl
Soil
Lake Bottom clay
Nicholson
Fanno
Chalmers
Canelo
Anthony
Mohave
River Bottom sand
Water
5.0 x 10'5
4.2 x 10~6 "
.1.5 x 10-4
1.8 x 10-5
5.0 x 10-6
9.9 x ID"5
1.9 x ID'3
1.7 x 10-2
Kerosene
3.4 x ID"4
9.5 x 10-5
5.2 x 10~3
3.8 x ID'5
5.6 x 10-5
2.0 x ID'4
2.6 x ID'3
3.4 x ID'2
Solvent
Isopropyl
alcohol
2.1 x 10-4
2.3 x 10-5
3.1 x 10-3
2.3 x lO-5
3.5 x 1C-5
9.5 x ID'5
1.6 x ID'3
5.8 x 10-3
Ethyl ene
glycol
1.1 x 1C-5
6.8 x 10~7
3.0 x ID'4
1.8 x ID'6
1.4 x 10~6
5.8 x 10'6
2.0 x ID'4
6.4 x ID'4
Xyl ene
1.7 x 10-3
3.5 x ID'4
1.8 x lO-2
1.7 x ID'4
1.5 x 10-4
4.4 x ID'4
8.5 x 10-3
1.8 x ID'2
aData from Schramm (1981).
-------
14
12-
10-
8.
6
4-
2
LAKE BOTTOM dAX
^_^
••MOM
•«•••
•H^B"
1
Kay:
f
]
WATR - Water
ETGL • Ethelyne Glycol
ISO? • Isopropyl
KERO « Xerosine
XYL « Xylene
u
a,
o 4.
UJ
I 3
i »
NICHOLSON
n
ANTHONY
nn
2 ,
CHALMERS
I^H^H^
CANELO
••^•H
n
WATR ETGL ISOP KERO XS. WATS ETGL ISO? KERO XXL
SOLVENTS
Source: Schramm, 1981
Rgure 4-14. Variation of intrinsic!permeability with solvent for each soil.
4-58
-------
4.5.10.3 Discussion—
The test method used by Schramm differed significantly from that
employed by Anderson (1981) to measure permeability. Notable differences
include the following:
• Gradients imposed on the sample were very low.
• No estimate was made of the pore volumes passed through the test
samples.
• Samples were presaturated with the test fluid rather than a standard
calcium-sulfate solution.
The impact of the low hydraulic gradient on the test results is not
known. The concept of "threshold gradient" has been postulated, although
there is no evidence of this effect in the results reported by Schramm.
Although Schramm reports that tests were continued until constant per-
meability values were obtained, the criterion to establish an acceptable
difference in consecutive measurements was not discussed. There is no indi-
cation of the number of pore volumes exchanged.
It may be argued that saturating the samples using the test permeant
fluid rather than water or a standard permeant solution is representative of
conditions typical for a liner in service in a land disposal application.
4.5.11 Evaluation by Monserrate (1982) of the Permeability of Two Clays
to Selected Electroplating Wastes
Monserrate (1982) at Duke University investigated the permeability of
two clays exposed to simulated electroplating wastes. The purpose of the
research was to examine the effect of compaction on the permeability response
of the two clays. The clays tested were a Wyoming bentonite and a White
Store clay from North Carolina. The White Store clay was reported to be an
active clay.
The simulated electroplating wastes used as permeant fluids were a
1-molar solution of zinc chloride (ZnCl2) (136.3 g/L, pH = 5.5) and a
1-molar solution of chromic acid (t^Crtty) (100 g/L, pH = 1.5).
4.5.11.1 Test Method--
Standard Proctor tests were conducted on each clay to establish the
relationship between water content and compacted density. Procedures
outlined in ASTM D-698 Method A (ASTM 1985) were followed. Permeability
studies were subsequently conducted at several different moisture contents.
For the permeability tests, the samples were prepared by mixing 0.01 N
calcium sulfate with the clay to achieve the desired moisture content. The
clays were compacted into either zinc-plated or lucite compaction permeam-
eters using compactive efforts comparable to those used in the standard
Procter tests. The compaction permeameters measure 4 inches (10.16 cm) in
diameter and 4.6 inches (11.68 cm) in height. Clays were compacted to a
depth of 2 inches (5.08 cm). The compaction permeameters were adapted to
4-59
-------
allow measurement of flow as the permeant fluid enters the unit. The data
were evaluated as a falling head test. A standard solution of deaired 0.01 N
calcium sulfate was used to saturate the sample prior to introducing the test
permeant fluids. The samples were exposed to the permeant fluids under
approximately 75 pounds of air pressure. In the permeability tests,a
statistical procedure was used to determine when enough data had been
accumulated to make a determination of the k value for a certain clay-
permeant fluid combination. (See details in Section 4.3).
4.5.11.2 Test Results-
Tests with bentonite showed that at imoisture contents in the proximity
of the optimal compacted moisture content, permeabilities to calcium sul-
fate, chromic acid, and zinc chloride were all on the order of 10~10
cm/s. At lower moisture contents, the permeability to calcium sulfate was
higher by a factor of 2 than the permeability to the chromic acid test
solution. At wet-of-optimum water-contents, however, the permeabilities
determined for the two fluids were nearly identical. The same low water
content divergence in permeability was observed for the zinc chloride test
fluid, although the magnitude of the difference was less.
The permeability of White Store clay to standard calcium sulfate solu-
tion and to the test fluids is shown in Figure 4-15 as a function of the
moisture content at compaction. The White Store clay appeared to be more
sensitive than the bentonite clay to changes in compacted moisture content.
The permeability to the chromic acid test fluid aligns closely with that of
calcium sulfate at moisture contents higher than about 20 percent (optimum
moisture content = 21.5 percent). At lower than about 20 percent moisture,
the permeability to the chromic acid was higher than to calcium sulfate.
Zinc chloride test fluid showed permeabilities lower than for calcium sulfate
at low moisture.
Monserrate (1982) concluded that for both bentonite and White Store
clays, the application of zinc chloride and chromic acid did not alter the
clay's permeability response, indicating that the structure of the clay, as
compacted, is the predominant influence in determining permeability. She
noted that the test results differed markedly from those of Cola (1981) who
carried out similar studies at Duke University using the chromic acid.
Monserrate noted that there was extensive corrosion of the steel per-
meameters and some of the pressure fittings that were exposed to the chromic
acid test fluid.
4.5.12 Research by Brown. Green, and Thomas (1983) on the Effect of Two
Organic Hazardous Wastes on Simulated Clay Liners
Brown, Green, and Thomas (1983) at Texas A & M University reported
results of permeability tests in which clay soil samples representative of
three types of simulated clay liners were subjected to a xylene waste and an
acetone waste in both laboratory and field studies. The three clay materials
tested were predominantly kaolinite, mica, or bentonite. The soils were
blended with sandy loam soil to attain permeabilities in the range of
1 x 10~7 to 1 x 10-°. The laboratory studies in fixed-wall permeameters
4-60
-------
o>
3.6 X 10
,-7
,-7
3.0 X 10
~ 2.4 X 10"7
fj
1 1.8 X 10~7
Q)
Q.
-7
1.2 X 10
6.0 X 10~8
Moisture Content at Compaction (%)
Source: Monseratte, 1982
Figure 4-15. Permeability of White Store clay to 0.01 N calcium sulfate, chromic acid
(1 molar), and zinc chloride (1 molar) as a function of moisture at compaction.
-------
were directed at determining the influence of different initial moisture
contents andjelevated gradients on the clay permeabilities. The procedures
used in the laboratory tests were similar to those described by Anderson
(1981) and Brown and Anderson (1983) in earlier permeability tests with
organics. The research was sponsored by the U.S. EPA.
The xylene waste, a paint solvent that was used to clean factory sprayer
lines, contained 25 percent paint pigments and trace amounts of water. The
acetone waste was a chemical manufacturing waste containing 91.7 percent
acetone, 4 percent benzene, 0.6 percent phenol, and 3.7 percent unknown.
4.5.12.1 Test Methods—
For the laboratory tests, soils were compacted wet of optimum to at
least 90 percent Proctor density. Prior to exposure to the waste solvents,
samples were saturated with 0.01 N calcium sulfate. Unsaturated cores were
also subjected to the permeability tests. Pressures of 5, 15, and 30 psi
(equivalent to hydraulic heads of 31, 91, and 181) were tested.
Twenty-eight field cells (1.5 m x 1.5 m x 1.8 m inside dimensions) were
constructed of reinforced concrete and lined with 100-mil high-density
polyethylene to facilitate leachate collection. The clay soils were com-
pacted with a vibratory compactor to 95 percent Proctor in two 7.5-cm-thick
lifts. The liquid waste to be tested was introduced into each cell through a
standpipe Into barrels that were placed above the liners. A head of 1 m of
liquid (gradient of 7) was maintained throughout the test. Permeabilities
were calculated based on the volume of cell drainage collected during dis-
crete time intervals. Following the permeability tests, each cell was dis-
assembled and subjected to chemical and morphological analysis. The wastes
to be tested in the field cells were dyed with Automate Red B and Fluorescent
Yellow to facilitate detection within the sample cores.
4.5.12.2 Laboratory Test Results—
To compare the results from the various tests, calculated permeabilities
were plotted as a function of the number of pore volumes of effluent that
passed through the sample. Permeabilities to standard calcium sulfate solu-
tion on similar soils in the laboratory were generally found to be repro-
ducible to within 0.25 order of magnitude. Permeabilities increased, how-
ever, when the organic-solvent wastes were substituted as the permeant
fluids. Permeability to the organic-solvent wastes was typically 2 to 4
orders of magnitude greater than the permeabilities measured with the calcium
sulfate solution. Presaturation with 0.01 N calcium sulfate appeared to
retard the effect of the permeant fluids on the samples (I.e., more pore
volumes were displaced before breakthrough occurred 1n the presaturated
samples). Highest permeabilities measured on initially unsaturated samples
were typically 1 to 2 orders of magnitude greater than for samples that were
Initially saturated.
Permeability of the clay soils increased rapidly upon exposure to xylerie
waste after the cumulative flow exceeded 0.2 to 0.4 pore volume. The behav-
ior of acetone was characterized by an Initial small decrease 1n permeability
(minimum at approximately 0.5 pore volume), which was followed by a steady
Increase In permeability. This behavior was explained as follows: acetone
first caused swelling of the clay soil with low concentrations Initially
4-62
-------
diffusing into the pores; as more of the water was displaced by the solvent,
shrinkage occurred.
— • t .3.-- ' • • m. s
-:*v •: *
Similar increases in permeability were observed for all three soil
types. The increase in permeability, however, for the bentonite samples
occurred after a larger pore volume was displaced.
If the permeability values calculated from the test data are plotted as
a function of pore volumes, the different hydraulic gradients used in the
experiments (31, 91, 181) have little influence on either the increase in
permeabilities observed or on the final values achieved. (This conclusion
holds for both presaturated and unsaturated samples [Brown et al., 1984]).
4.5.12.3 Field Test Results—
In the field studies, xylene had penetrated 10 of the 12 clay liners
within 12 months after installation; acetone had penetrated 2 of the liners
but at a slower rate. Increases in permeability of the bentonite clay soil
to xylene were more dramatic (as much as 2 orders of magnitude increase) than
of the other soils and occurred after passage of 0.5 pore volume of permeant
fluid. Kaolinitic and micaceous soils showed immediate increases in perme-
ability to xylene. Permeability values continued to increase as more pore
volumes of fluid passed through the sample. Increases of more than 2 orders
of magnitude were observed in some samples after passage of two pore volumes
of xylene.
Field tests with the acetone waste exposed to micaceous clay soil showed
a permeability decrease with the first 0.5 pore volume. This was followed,
however, by an increase in permeability. After two pore volumes had passed
through the liner, the permeability was slightly higher than the initial
value measured.
Visual inspection of the clay liners following permeation by the liquid
wastes was facilitated by the dyes and by the paint pigment in the xylene
waste. Dyed surfaces observed throughout the liner Indicated that the liquid
had moved through the soil through preferential channels rather than in a
clearly defined wetting front. Preferential channels were occasionally found
to extend from the top to the bottom of the liner. Evidence that xylene
moved through cracks in the liner or around blocky subangular structural com-
ponents was also obtained through chemical analysis of sections of the per-
meated liners. The channels observed in the clays are attributed to the
displacement of water by the chemicals resulting in desiccation that caused
the clay to shrink and crack.
4.5.12.4 Discussion—
The results of this study, in close agreement with those previously
found by Brown and Anderson (1983), indicate that severe permeability in-
creases can occur when certain clay materials are in contact with concen-
trated organic solvents. Although the hydraulic heads used in the laboratory
test procedures are higher than are likely to be encountered in the field,
the data suggest that the permeabilities measured are independent of the
hydraulic gradient used in the test. These results further support the
conclusions by Anderson based on tests involving a gradient of 361.
4-63
-------
The effect of presaturatlon with calcium sulfate on the highest perme-
abilities 1s .significant. Laboratory tests by other researchers (Green
et al., 1979; Foreman and Daniel, 1984) have also involved introducing the
test permeant fluid onto the unsaturated sample.
4.5.13 Study by Brown. Thomas, and Green (1984) of the Effect of Dilutions
of Acetone and Mixtures of Xylene and Acetone on Permeability of a
Micaceous Soil
To determine the effect of diluting a polar organic solvent with water,
Brown, Thomas, and Green (1984) measured the permeability of an unsaturated
micaceous compacted soil to several acetone-water dilutions. Samples were
tested at a gradient of 91 in laboratory tests with fixed-wall permeameters.
Solutions tested were 100, 75, 50, 25, 12.5, and 2 percent acetone. Methods
were similar to those used previously by Anderson (1981).
Compared to permeability to water (actually 0.01 N calcium sulfate), an
Increase in permeability was seen for solutions where the concentration of
acetone was 75 percent or 100 percent. Samples permeated with lower concen-
trations of acetone did not show appreciable changes compared to the perme-
ability to water.
Similar tests were conducted to determine the effect on permeability of
mixtures of xylene and acetone (Brown et al., 1984). For 100 percent xylene,
the permeability was reported to be 4 orders of magnitude greater than the
permeability to water (0.01 N calcium sulfate). When a mixture of 87.5 per-
cent xylene and 12.5 percent acetone was tested, the permeability was dramat-
ically reduced (i.e., more than 3 orders of magnitude below the permeability
to pure xylene). Values were comparable for a mixture containing 25 percent
acetone and 75 percent xylene and for a 1:1 mixture. When the acetone compo-
nent was increased to 75 percent, however, the permeability increased to
approximately the value obtained in similar tests with pure acetone. This
value was about 1.5 orders of magnitude greater than the permeability of the
sample to 0.01 N calcium sulfate.
4.5.14 Tests by Brown. Thomas, and Green (1984) to Determine the
Permeability of Micaceous Soil to: Petroleum Products
Brown, Thomas, and Green (1984) also; reported permeability tests with
five commercial petroleum products. A compacted micaceous soil, unsaturated,
was permeated with the test fluids at a gradient of 91. The results of the
tests with kerosene, diesel fuel, paraffin oil, gasoline, and motor oil are
shown in Figures 4-16 through 4-20, respectively. Conclusions of the tests
show that the micaceous soil is more permeable to these petroleum products
than to water (0.01 N calcium sulfate).
In tests with kerosene, the permeabilities were 3 to 4 orders of magni-
tude higher than values measured with 0.01 N calcium sulfate in similar
samples. One-half the difference was achieved within the first 0.1 pore
volume. The permeability to paraffin oil was about 1 order of magnitude
greater than the comparable value to the calcium sulfate solution despite
Its high viscosity. Maximum values were achieved by the passage of one pore
volume.
4-64
-------
i64-
1,5
CD
<
UJ
"S.
ce
UJ
Q_
10^
io7-
,OgO.
LAB MICA
KEROSENE
NONSATURATED
GRADIENT 91
• REP I
-X REP 2
O REP3
PORE VOLUME
Source: Brawn, Thomas, and Green, 1984
Figure 4-16. Hydraulic conductivity versus pore volume for laboratory compacted
micaceous soil exposed to kerosene at a hydraulic gradient of 91.
4-65
-------
i
0>
CD
u
»
in
U
-J
CD
<
UJ
2
K.
10"
fo8
O
1.0
20
LAB MICA
DIESEL FUEL
NONSATURATED
GRADIENT 91
! • REP I
! X REP 2
O REP 3
• REP 4
PORE
4.0
5.0
6.0
Source: Brown, Thomas, and Green, 1984
Figure 4-17. Hydraulic conductivity versus pore voiume for iaboratory compacted
micaceous soil exposed to diesel fuel at a hydraulic gradient of 91.
-------
i66-
>io7-
03
<
UJ
2
tr
LU
Q_
.68-
I09-
LAB MICA
PARAFFIN OIL
NONSATURATED
GRADIENT 91
• RER I
X R£f>2
O REP 3
1 PORE VOLUME 2
Source: Brown, Thomas, and Green, 1984
Figure 4-18. Hydraulic conductivity versus pore volume for laboratory compacted mica-
ceous soil exposed to paraffin oil at a hydraulic gradient of 91.
4-67
-------
:66
o
03
m
>I07
CD
<
LU
5
cr
LU
CL
:68-
LAB MICA
GASOLINE
NONSATURATED
GRADIENT 91
RER 4
1 PORE VOLUME 2
Source: Brown, Thomas, and Green, 1984
Figure 4-19. Hydraulic conductivity versus pore volume for laboratory compacted
micaceous soil exposed to gasoline at a hydraulic gradient of 91.
4-68
-------
i6s
o
Q>
m
>io7-
I-
03
<
LJ
2
0=
LU
0.
io9-
LAB MICA
MOTOR OIL
NONSATURATED
GRADIENT 91
X RER 2
PORE VOLUME2
Source: Brown, Thomas, and Green, 1984
Figure 4-20. Hydraulic conductivity versus pore volume for laboratory compacted
micaceous soil exposed to motor oil at a hydraulic gradient of 91.
4-69
-------
Test data obtained with diesel fuel were variable and did not appear to
Increase along the patterns of most organics tested. No explanation is
offered for tfie unpredicted behavior. Values were, however, greater by 1 to
2 orders of magnitude than permeability to water. This was also the case for
gasoline. The test with motor oil showed a steady increase in perme-
ability from 2 x 10~7 cm/s at 0.5 pore volume to 1.5 x 10~6 cm/s at 3
pore volumes.
4.5.14.1 Discussion—
The data obtained in tests with the commercial petroleum products
illustrate the importance of extended permeability tests to allow passage of
several pore volumes of fluid. Earlier data indicate that the effects of
organic permeant fluids are retarded in tests where the samples are presatu-
rated with water (or 0.01 N calcium sulfaite). The tests on unsaturated
samples may serve as better estimators of maximum permeabilities that may
result when clay liners are exposed to petroleum products.
4.5.15 Study by Brown and Thomas (1984) of the Permeability of
Commercially Available Clays to Organics
To determine the effect of organic fluids on currently available clays
that are sold for sealing and lining impoundments, Brown and Thomas (1984)
tested permeabilities of three cpmmercial;ly available clays admixed with
sand.
4.5.15.1 Test Method--
Each clay was mixed with sand to obtain a permeability to water of
about 1 x ID-8 cm/s. Smectite is the dominant mineral in soil CC1 (blue
bentonite) and soil CC2 (a synthetically treated bentonite); a micaceous
mineral is dominant in soil CC3 (Ranger Yellow). The soils were compacted at
optimum moisture in 10-cm fixed-wall molds using a mechanical compactor as
described in ASTM Procedure 698-70.
The compacted soils were exposed to the test fluid without presatura-
tion. Following a 24-hour equilibrium period, a pressure of 15 psi (equi-
valent to a hydraulic gradient of 91) was applied to the liquid surface.
Leachate passed through the samples was collected at intervals to allow
calculation of the soil permeability. Fluids tested in replicate were water
(0.01 N calcium sulfate), acetone, xylene, kerosene,-tiiesel fuel, gasoline,
and used motor oil. A statistical analysis was performed to establish the
variance among the data set for each permeant fluid. Means were separated
using a Duncan's Multiple Range test at a significance level of P = 0.05.
4.5.15.2 Test Results—
The test results for all permeant fluids are summarized in Table 4-12.
All of the organic fluids tested produced dramatic increases over the cor-
responding permeabilities to water, the increase ranging from 1 to 5 orders
of magnitude. Some of the increases, though large, were not statistically
significant compared to the permeabilities measured with water due to the
large variability seen 1n the tests. The, increases in permeability were
significant for all clays permeated with xylene and for both the smectitic
clays when permeated by gasoline and kerosene.
4-70
-------
TABLE 4-12. MEAN CONDUCTIVITY OF EACH SOIL TO EACH FLUID TESTED
(Brown and Thomas,, 1984)
Fluid
Water
Acetone
Xylene
Gasoline
Kerosene
Diesel fuel
Motor oil
CC1
3.61 x 10-8ba
5.05 x 10~5b
1.76 x 10-4a
1.96 x 10-4a
1.49 x 10-4a
5.17 x 10-5b
6.13 x 10-6b
CC2
2.58 x 10-8b
1.41 x 10-6b
7.28 x 10-4a
9.07 x~10-5a
9.10 x 10-5a
4.53 x 10-5ab
2.13 x 10-6b
CC3
1.57 x 10-8b
2.51 x 10-7b
1.00 x 10-4a
6.19 x 10~5b
5.68 x 10~5b
6.29 x 10-7b
9.48 x 10-7b
aValues in a given column followed by the same letter do not
differ significantly (P = 0.05).
4-71
-------
Generally the increase 1n the micaceous clay permeability was less by 1
order of magnitude than the increases seen in the smectitic clays. The
untreated bentonite (CC1) showed the greatest permeability increase in
response to each organic test fluid. Since the untreated bentonite is the
soil that is most subject to shrinkage and the micaceous soil is least sub-
ject to shrinkage, these findings are corisistent with the theory that shrink-
age resulting in greater spacing between peds is responsible for the changes
in soil permeability.
4.5.16 Studies Conducted for EPA by Daniel (1983) and Foreman and Daniel
(1984) at the University of Texas, Austin
Under a cooperative agreement with EPA, Daniel and others have carried
out research to address how the permeability of clay soils is affected by
hydraulic gradient and by the test device when the permeant fluid is other
than water. Permeability tests were performed using three different types
of test devices—flexible-wall cells, fixed-wall compaction mold permeam-
eters, and consolidation cell permeameters—at hydraulic gradients of 10,
50, 100, and 300. The three soils studied were a noncalcareous smectite
(Lufkin clay), a mixed-cation ill He (from Hoytville, Ohio), and a commer-
cially processed kaolinite (Hydrite R). Fluids tested were water (actually
0.01 N calcium sulfate), methanol, and heptane (Foreman and Daniel, 1984).
4.5.16.1 Test Method-
Most of the tests for the project were flexible-wall tests. The fixed-
wall and consolidometer tests were performed to allow comparison of the data
from the different devices. Soil samples were compacted at or slightly dry
of optimum moisture content according to ASTM D-698. Soils were scarified
between each lift. Samples to be tested in flexible-wall cells were extruded
from the 10.2-cm (4-inch) -diameter compaction mold and trimmed to a height
of 9.4 cm (3.7 inches). Similar diameter samples with a height of 11.9 cm
(4.7 inches) were tested in the fixed-wall permeameter setup. Samples tested
in the consolidation cell were taken from the central portion of a compacted
10.2-cm (4-1nch) sample. The sample v/as trimmed to a height of 1.90 cm (0.75
Inch) and a diameter of 6.4 cm (2.5 inches).
Prior to the permeability measurements, the soil samples were saturated
with the permeant fluid to be tested. For the flexibje-wall tests great care
was taken to completely saturate the samples. This was accomplished through
soaking under vacuum and backpressuring. During the permeability tests,
the backpressure was kept at about 2.8 kg/cm2 (40 psi) and the cell pres-
sure at 3.87 kg/cm2 (55 psi). The average effective stress was 1.05 kg/cm2
(15 psi) for tests at the .lower hydraulic gradients; an average effective
stress of 1.76 kg/cm2 (25 psi) was used when the gradient was 300.
4.5.16.2 Test Results--
Test results with Lufkin clay in the flexible-wall cells showed
essentially no change in permeability with time when methanol was used as the
permeant fluid. Data were obtained for passage of about 0.9 pore volume.
The permeability to methanol was virtually the same as with 0.01 N calcium
sulfate (Daniel, 1983).
4-72
-------
Methanol test results with the kaolinite show that when permeability is
plotted as a function of pore volumes of flow, the curves have similar shapes
for the three types of permeameters"V The compaction mold yields somewhat
higher permeability results than the other two test devices. Test data are
shown in Figure 4-21.
Permeability values for the kaolinite at different hydraulic gradients
were obtained with water and with methanol in the different test devices.
The following conclusions were drawn:
• At high hydraulic gradients (values of 150 or larger), kaolinite has
a higher conductivity to methanol than to water regardless of
permeameter type.
• At high hydraulic gradients, the flexible-wall and consolidation-
cell permeameters yield similar hydraulic conductivities for both
methanol and water; kaolinite is roughly twice as permeable to
methanol as to water for these two types of permeameters.
t At high gradients, use of compaction-mold permeameters leads to large
sdatter in measured hydraulic conductivity. On the average,
compaction-mold devices showed kaolinite to be approximately 10 times
more permeable to methanol than to water. The lowest values of K
measured with the compaction-mold devices are nearly identical to
values measured with the other two devices. It is possible that
sidewall leakage contributed to the causes ....
• With flexible-wall permeameters and two liquids and one soil, hydrau-
lic gradient appears to have little effect on hydraulic conduc-
tivity . . . .
• With consolidation-cell permeameters, hydraulic gradient has a very
substantial effect on hydraulic conductivity. At a hydraulic gradi-
ent of 50, K to methanol is only half of K to water, but at gradients
of 200 to 300, K to methanol is twice as large as K to water ....
Figures 4-22, 4-23, and 4-24 show the effect of hydraulic gradient on
tests with kaolinite in flexible-wall cells, consolidation cells, and fixed-
wall permeameters, respectively. The scatter in the Tixed-wall permeameter
data does not allow conclusions to be drawn although there appears to be a
trend to increased permeability with increased gradient.
4.5.17 Tests Conducted for Chemical Manufacturers Association by Daniel
and Liljestrand (1984T
In work sponsored by the Chemical Manufacturers Association (CMA),
Daniel and Liljestrand at the University of Texas at Austin carried out a
laboratory study of the permeability of five clay soils to six aqueous fluids
representative of waste leachates or actual liquid wastes. The purpose of
the testing was to determine if dilute organic/water mixtures produced the
increases in permeability demonstrated by other researchers using concen-
trated organic liquids as permeant fluids.
4-73
-------
o
O)
U)
E
o
>.
>
o
3
C
o
O
o
T3
>s
0
Pore Volumes of Flow
Source: Foreman and Daniel, 1984
Figure 4-21. Permeability versus number of pore volumes of flow for kaolinite
permeated with methanol at a hydraulic gradient of 250 or 300.
4-74
-------
-6
0 h
u
-------
,20-0 300
Hydraulic; Gradient
Source: Foreman and Daniel, 1984
Figure 4-23. Permeability versus hydraulic gradient for kaoiinite
permeated in consolidation cell permeameters.
4-76
-------
-6
o
o>
en
£
o
o
3
-O
C
o
p
u
T3
>^
X
-7
iO
0
-8
0
M e t h a n o
100 200
Hydraulic Gradient
Source: Foreman and Daniel, 1984
Figure 4-24. Permeability versus hydraulic gradient for kaolinite
permeated in compaction moid cell.
300
4-77
-------
The soils studied included Texas Lufkin clay (also used in tests at
Texas ASM) and four soils supplied by CMA (identified as SI, S2, S3,, and S4)
from actual landfill sites. Properties of the clays are listed in Table
4-13. Permeability tests were not performed on the SI clay since its index
properties and mineralogy were similar to those of the S2 clay.
Permeant liquids used in the tests were: a leachate from a solid-waste
landfill (L2); two liquids (LI and L3) from industrial waste impoundments; an
aqueous solution containing methanol at 5 percent (50,000 ppm); an aqueous
solution saturated with xylene (near 196 ppm); and the leachate, 11, spiked
with chloroform (200 ppm) and trichloroethylene (200 ppm) to simulate a land-
fill leachate contaminated with chlorinated hydrocarbons. Water used to pre-
pare the dilute methanol and xylene solutions was actually 0.01 N calcium
sulfate. Characteristics of the leachates LI, L2, and L3 are listed in
Table 4-14.
4.5.17.1 Test Method--
Permeability tests were carried out as described previously in Section
4.5.16. All testing was performed usingiflexible-wall permeameters and
hydraulic gradients of 150 or 200. Rates of flow were determined from the
rate of inflow for each test chamber.
4.5.17.2 Test Results—
The permeability tests were performed for several months, allowing the
passage of more than one pore volume of fluid for most soil/leachate combi-
nations. The permeabilities of the soils S2, S3, S4, and Lufkin clay appear
to be related to the plasticity index, the permeability decreasing with
Increasing plasticity. For any one soil, the permeabilities to the various
liquids were all about the same. Permeabilities to the leachates did not
differ significantly from the permeability to water (actually 0.01 N
calcium sulfate). All permeabilities measured were 1 x 10~8 cm/s or
lower. In three of the four soils tested, the permeability to 196 ppm xylene
was slightly lower than the permeability to the 5 percent methanol. All
permeabilities were plotted as a function of pore volumes of fluid passed
through the sample.
In addition to the permeability tests, Atterberg limits of the soils
were determined using the various test fluids and pure xylene and methanol in
place of water. Test fluids LI, L2, L3, and spiked L-l did not significantly
alter the plasticity of any of the clay soils during the short-term exposure
of the test. Mixing any of the five soils with pure xylene or pure methanol,
however, caused a drop in the liquid limit and destruction of the soil's
tendency to be plastic. The dilute solutions of methanol and xylene (as used
in the permeability tests) had much less significant effects on the plas-
ticity. Only the soil S3 showed a large drop in plasticity when mixed with
the 5-percent methanol or the aqueous solution containing 196 ppm xylene.
The significance of Daniel and Liljestrand's findings is that they
appear to show that dilute organic/water mixtures are not capable of causing
significant changes in the permeability of natural clay liners.
4-78
-------
TABLE 4-13. PROPERTIES OF CLAY SOILS TESTED BY DANIELS AND LILJESTRAND (1984)
VI
to
Property
Natural water content (%)
Optimum water content (%)
Specific gravity
Percent finer than #200
sieve
Percent sand
Dominant minerals
Secondary minerals
Organic carbon content
(% of dry weight)
Cation exchange capacity
SI
23
17
2.73
93
7
Illite
Chlorite
1.46
10
S2
22
24
2.81
93
7
Chlorite
Smectite,
kaolinite
0.83
20
S3
47
31
2.71
94
6
Smectite
Kaolinite
1.39
40
S4
32
18
2.71
87
13
Quartz
kaolinite
1.79
5
Lufkin clay
i
23
21
2.66
81
19
Smectite
Kaolinite,
illite
0.28
25
(meq/100 g)
Plasticity index
21
32
59
16
42
-------
TABLE 4-14. LEACHATE CHARACTERISTICS*
Parameter measured
or reported
Suspended solids (mg/L)
Dissolved solids (mg/L)
Volatile solids (mg/L)
Total organic carbon (TOC)
(mg/L)
Chemical oxygen demand (COD)
(mg/L)
Potential (ORP) (mV)
PH
Conductivity (jdnho/cm)
Metals (mg/L)
Cr
Cu
Pb
Ni
Zn
Organics (mg/L)
Ethyl benzene
Toluene
Nltrotoluene
Dinitrotoluene
Formaldehyde
Dichlorobenzene
Aniline
Toluenediamine
LI
465
20,202
5,768
1,440
2,160
398
6.9
23,700
1.0
5.3
0.36
0.18
6.0
b
Leachates
L2
162
1,043
121
13
12
268
7.1
865
b
b
*
L3
107
4,593
1,052
82
120
130
496
1.5
4,620
b
0.120
0.035
8
117
14
8
4
2
aData from Daniel and Liljestrand, 1984.
bNot reported.
4-80
-------
4.5.18 Study by Dunn (1983) of the Effects of Synthetic Lead-Zinc Tailings
Leachate on Clay Soils
Dunn (1983) evaluated the effects of a synthetic lead-zinc tailings
leachate, of low pH and containing large concentrations of heavy metals, on
the permeability of two clay soils. The permeability of the clays was shown
to be affected by the waste permeant, with cation exchange and precipitation
apparently being the most important processes.
The soils used in the testing, Altamont soil and Rockville soil, had
been identified as potential liner construction materials. Altamont soil, a
"valley alluvium" with moderate plasticity, is a silty clay soil with 1 or 2
percent shale fragments. Montmorillonite is the predominant clay mineral in
Altamont soil. Rockville soil, a yellow-brown silty clay with high plas-
ticity, is composed of the fine fraction from a sand and gravel plant. The
predominant clay mineral is kaolin. This soil probably contained micro-
organisms since some evidence of bacterial growth was noted in soil that was
stored for several months.
The permeant fluids tested were tap water, distilled water, and the
synthetic lead-zinc tailings leachate. Characteristics of this-fluid were:
. t Conductivity 1,550 ^mho/cm
• pH = 2.6
• Lead = 5.8-15.0 mg/L
• Zinc = 200 mg/L.
The leachate was prepared with zinc sulfate, lead sulfate, and lead nitrate.
Sulfuric acid was used to attain the desired pH.
Two tests with each clay were conducted with the synthetic leachate.
Permeabilities obtained with the test fluid were compared to the results of
samples tested with tap water and with distilled water. All the tests were
run at a compacted dry density of about 90 percent of maximum dry density and
at a water content approximately 1 to 2 percent above optimum. Tap water was
used for the molding water. Soil samples were prepared with static compac-
tion since this method was found to be most appropriate for producing repli-
cate samples at approximately the same dry density and water content. The
static compaction involves compressing the soil to a known density with
applied hydraulic pressure.
Tests were performed in a triaxial cell at a hydraulic gradient of 50.
Samples tested were 3.81 cm (1.5 inches) in diameter and 2.54 cm (1 inch)
thick. The samples were presaturated (using a backpressure technique) with
the fluid to be tested. Prior to permeability testing, the samples were
consolidated at 1.5 kg/cm2 (21.3 psi) effective stress. Permeability
measurements were carried out at an effective stress of 1.25 kg/cm2
(17.8 psi).
4-81
-------
The permeability values measured in both soils exposed to the synthetic
leachate were_higher than those measured with tap water. Values measured
with distilled water were lower than permeabilities measured with tap water.
The ultimate effect of the synthetic leachate on the permeability to either
soil could not be determined from the tests, however, because the perme-
ability values did not reach steady state during the duration of the test
(5,000 minutes or 3.5 days). The tests with tap water did appear to reach
steady state during this test time.
For one Altamont sample, the measured permeability to the synthetic
leachate was approximately 2 orders of magnitude higher than that of tap
water. This appeared to correlate with a long curing period (the period of
time from compaction to the beginning of the permeability test). Thixotropic
alterations of the soil fabric may have occurred during this interval,
resulting in the higher K value.
4.5.19 Studies by Acar and Others (1984) on the Effect of Organics on
Kaolinite
In research funded by EPA, Acar, Olivieri, and Field (Acar et al., 1984b
and c) and Acar, Hamidon, Field, and Scott (Acar et al., 1984a) studied the
effect of four organic fluids on the saturated permeability of Georgia kaoli-
nite. The fluids tested—benzene, acetone, phenol, and nitrobenzene—were
chosen because they represent a wide range of dielectric constants. Compara-
tive tests were performed with 0.01 N calcium sulfate. In addition to the
pure solvents, 0.1 percent (1,000 ppm) solutions of acetone and phenol pre-
pared in 0.01 N calcium sulfate solutions were tested. This research was
carried out at Louisiana State University, Hazardous Waste Research Center.
4.5.19.1 Test Method—
The kaolinite was cured at 32 percent moisture for 1 week before compac-
tion in a Harvard miniature mold at a compactive effort corresponding to
standard Proctor compaction. Sample dimensions were restricted to 3.55 cm in
diameter and 3.8 to 5.1 cm in height.
Tests were conducted in triaxial cells with continuous backpressure.
Hydraulic gradients of less than 100 were used. Backpressures of 414 to 449
kPa (60 to 65 ps1) were used to fully saturate the samples prior to the per-
meability testing. Approximately one pore volume of D.01 N calcium sulfate
was passed through the samples to establish the reference permeability
value. The Influent liquid was then switched to the organic fluid to be
tested. Tests were continued until the permeability readings and the efflu-
ent concentrations were stable.
A mercury Intrusion method was used to characterize the pore size dis-
tribution in the samples before and after permeation with the organic test
fluids and with the calcium sulfate (Acarlet al., 1984b).
A fixed-wall test with acetone as well as flexible-wall tests at vari-
able effective stresses were also carried out in order to evaluate the test
scheme.
4-82
-------
Free-swell and liquid limit tests were conducted with the organic solu-
tions and with the standard calcium sulfate to determine if these properties
were affected by the chemicals.
4.5.19.2 Test Results--
All tests with chemicals at low concentrations resulted in slight de-
creases in permeability. It was found, however, that the chemicals diffused
through the flexible latex membrane so that the concentration of the organics
that actually permeated the soil sample could not be ascertained.
Reference permeabilities in all kaolinite samples tested with pure
organics were between 5.0 x ID"8 cm/s and 6.0 x 10-° cm/s. When pure
organic fluids were introduced into the test cells, an immediate decrease
in permeability occurred. For acetone and phenol, this decrease was followed
by a permeability increase, the final value stabilizing at approximately
double the initial or reference value. In the tests with benzene and nitro-
benzene, the permeability decreased until the tests were terminated. Final
permeability values measured in these tests were 2 orders of magnitude lower
than the reference values.
When acetone was tested in a fixed-wall permeameter, the perme-
ability stabilized at 2 x 10-° while tests under comparable conditions in
flexible-wall cells yielded values between 6 x 1Q-8 and 9 x lO"8. The
difference in these test results is attributed to sidewall leakage in the
fixed-wall test resulting from sample shrinkage during permeation with
acetone. The free-swell tests also confirm that shrinkage would occur with
acetone permeation.
Both swelling behavior and liquid limit determined with the specific
pel-meant fluid were found to relate to the changes in permeability measured.
The liquid limit and free swell were increased significantly with benzene and
nitrobenzene, slightly increased with phenol, and decreased with acetone.
Although absolute permeabilities were altered considerably by permeation
with the pure organic fluids, the mercury intrusion investigations showed
that the size and distribution of pores greater than 80 A were not
significantly affected. Since pores with diameters of less than 80 A*
are not expected to contribute appreciably to the total flow, these results
suggest that physicochemical properties of the pore fluid close to the clay
surfaces lead to variations in flow characteristics.
4.5.20 Finding by Olivieri (1984) of Impermeability of Montmorillonite to
Benzene'
Olivieri (1984) found that benzene did not penetrate compacted Ca-
montmorillonite that was first fully saturated with 0.01 N calcium sulfate
solution even at a hydraulic gradient as high as 150. This finding was
attributed to hydraulic pressures being less than the required flow
initiation pressure to two-phase flow (Acar and Seals, 1984).
4-83
-------
4.5.21 Study of Permeability of Clays to Simulated Inorganic Textile
Wastes by lulls (1983T!
lulls (1983) tested Wyoming bentonite, Grolley kaoline, vermiculite, and
White Store clay in compaction permeameters with alkaline metal hydroxides.
Test solutions were ferrous hydroxide, cupric hydroxide, and manganese
hydroxide prepared from 1 N solutions of the respective sulfates by adding
sodium hydroxide until a precipitate formed. The final solutions had pH
readings as follows: ferrous = 12.9, manganese = 13.0, and cupric = 12.5.
Test results show an increase in permeability of the bentonite when it was
exposed to the alkaline permeant fluids. The permeability of the kaoline
decreased with the application of the bases. Neither the field clay nor the
vermiculite showed any significant variation in permeabilities with the test
fluid.
Cracks observed in the bentonite indicate that shrinkage was the mech-
anism responsible for the observed increase in permeability. The decreased
permeability of the kaoline was attributed to clogging of the pore space by
dispersed clay. All of the leachates from the tests contained silica that
had been dissolved by the alkaline solutions.
4.5.22 Tests Conducted by Engineering Consulting Firms for Specific
Application (unpublished data)
4.5.22.1 Tests Conducted by the Trinity Engineering Testing Corporation—
The tests discussed below were performed by the Trinity Engineering
Testing Corporation at the request of the Corpus Christi, Texas, City Water
Department (White, 1976).
4.5.22.1.1 Soil Characteristics and Test Method—A soil sample from a
proposed toxic waste landfill site was subjected to permeability testing with
isopropyl alcohol, benzene, and charcoal starter fluid. The material tested
was a subsurface clay. Composite samples: were compacted at optimum moisture
content to a height of approximately 91.4 cm (36 inches) at 95.8 percent
standard Proctor density in 2.54-cm (1-inch) inside diameter cylinders. Each
of the three organic fluids was added to a level of approximately 61 cm (24
Inches) above the compacted soil samples. Water was tested in a fourth
sample. A constant pressure head of 7.01 m (23 feet^was Imposed with com-
pressed air. Liquid level in each test cylinder was recorded every 4 hours
until all the liquid had penetrated the full 91.4-cm (36-inch) column of
compacted material.
4.5.22.1.2 Test Results—In the experiments with Isopropyl alcohol and
water as the test fluid, the total drop 1n liquid level was less than 1 inch
over a 100-day period. In the column tested with the charcoal starter fluid,
liquid began dripping from the bottom of the test cylinder after 122 days.
In the benzene test column, signs of full penetration throughout the clay
material were observed after 36 days.
In a second experiment series, two 91.4-cm (36-inch) compacted clay soil
samples were tested with benzene. The samples differed only 1n their mois-
ture content (sample A at 10 percent moisture and B at 20 percent moisture),.
Sample A showed signs of full penetration of the benzene after 20 hours.
4-84
-------
Full penetration of benzene occurred after 32 days in sample B, and all
liquid passed_ through the sample in 71 days.
i* • "'•?,• ;
The conclusions drawn from the permeability tests are stated below:
• Under optimum compaction, the clays are highly impermeable to
domestic water and, conversely, are very permeable to lighter
hydrocarbons.
• When properly compacted to a finished thickness of 91.4 cm (3 feet),
the clays will serve as a suitable tank liner for domestic water but
will not contain the lighter hydrocarbons.
t Under optimum compaction, a 91.4-cm (3-foot) liner of the clay in a
tank 7.01 m (23 feet) deep containing benzene can begin to leak
within 36 days.
4.5.22.1.3 Discussion—Although the soil sample used in the Trinity
Engineering experiments is not adequately characterized, the results of the
test clearly indicate potential for large permeability increases resulting
from exposure to concentrated nonpolar hydrocarbons. The apparatus and test
procedures used differ substantially from those used in permeability tests
conducted by other investigators.
The data also illustrate the importance of moisture content during com-
paction. The performance of clay-soil liner in contact with chemicals such
as benzene could be drastically influenced by the uniformity of the moisture
content when the liner material was installed and compacted.
4.5.22.2 Test Data Submitted to Pennsylvania Department of Environmental
Resources—
The Pennsylvania Department of Environmental Resources has received data
pertinent to clay liner/chemical compatibility. The data pertain to specific
sites and specific wastes and were submitted to the State by consulting engi-
neers. Testing procedures vary and information needed to evaluate the data
is not always provided in the reports. All test results show the clay liner
permeabilities to be "within the range required" after exposure to the wastes
tested.
*
Report A deals with tests to evaluate effects on a liner material in
contact with a waste comprised of one part oil contaminated soil and four
parts water. For the permeability tests, samples were air dried and then
recompacted, at +2 percent of optimum moisture content, to 95 percent of
maximum dry density. Stainless steel molds 10.2 cm (4 inches) in diameter
and 11.7 cm (4.6 inches) in height were used. After trimming, the molded
sample was transferred to a constant head permeability device. A back-
pressure was applied to two of the four samples during saturation with 0.01 N
calcium sulfate solution. Permeability measurements with the calcium sulfate
showed no significant differences between the results for samples with or
without backpressure. Thus, it was concluded that the effect of entrapped
air was minimal.
After permeability tests with the standard calcium sulfate, the liner
samples were placed in contact with the test fluid, sealed, and placed on a
4-85
-------
rocking table for 30 days. Three of the samples were tested with the waste
fluid and one-with calcium sulfate solution. After 30 days, the samples were
drained and the permeability tests repeated. Reported results are shown in
Table 4-15.
No discernible .color change was observed after a 30-day exposure to the
waste fluid. No shrinkage along the sides of the mold was observed, although
a change in height of approximately 0.64 cm (0.25 inch) was reported.
The waste tested—one part oil contaminated soil and four parts water-
was not characterized further. Neither the extent of the oil contamination
nor the characteristics of the contaminated oil were reported. It may be
assumed that either tap water or deionized water was used to prepare the
waste fluid. The extent to which the oil or its contaminants would be
extracted by the water is unknown but probably very small. Although it is
not discussed in the report, the waste material in contact with the clay
liner sample may have involved three phases, with oil being the lightest
phase.
No indication of the length of the permeability testing procedure or
pore volumes displaced is given. The hydraulic heads used in the tests also
are not reported. These details may be stated in a letter that is referenced
in the report. Soil characterization data were not included.
Report B describes the results of a similar investigation in which liner
samples from a disposal site were tested with four different wastes. The
permeabilities are reported in Table 4-16. All samples showed slight (two-
fold) increases in permeability after a 30-day exposure to the wastes. It
is notable that "control sets" exposed for 30 days to 0.01 N CaS04
showed similar increases.
Soil properties are listed below:
Cation exchange capacity: 11.1 meq/100 g soil
Predominant exchangeable cation: Calcium
Major mineral fraction: Alpha quartz
Other minerals (trace to minor): Microcline, Adularia, Muscovite,
Kaolinite
Percent clay size ( 0.002 mm): 13
Percent silt size (0.002 mm -0.05 mm): 30
Percent sand size (0.05 mm -2.0 mm): 22
Percent larger than 2.0 mm: 35
Liquid limit (percent water): 31
Plastic limit (percent water): 22
Wastes used in the tests were not characterized beyond the description
given in the table. Presumably, they were diluted with water as was done in
Case A, but this is not stated in the report. The duration of the tests,
pore volumes replaced, and hydraulic head are not provided in the report.
Report C provides permeability data on a soil sample tested with chrome
ore leachate. The leachate (pH = 13.0) contained 1,400 mg/L total chromium
and 1,200 mg/L hexavalent chromium. A constant head test, conducted at a
pressure of 20 psi, gave a permeability of 1.2 x 10~8 cm/s. With the
falling head method, a permeability of 2.2 x 1Q-8 cm/s was measured.
4-86
-------
TABLE 4-15. PERMEABILITY TEST RESULTS*
(Pennsylvania Case A)
Sample
AC
BC
C
D
With
1
5
9
1
0
.1
.9
.4
.7
.01
X
X
X
X
Coefficient
N CaS04
10-7
10-8
10-8
10-7
of permeabili
ty
After exposure
1.2
3.9
1.1
Not
X
X
X
(cm/s)
to test
10-7d
10-8
10-7
fluidb
tested
aTest data reported to Pennsylvania Department of Environmental
Resources.
bTest fluid was one part oil-contaminated soil to four parts water,
cSamples under backpressure during initial permeability tests.
dControl sample—tested after 30-day exposure to 0.01 N CaS04«
4-87
-------
TABLE 4-16. PERMEABILITY TEST RESULTSa
(Pennsylvania Case B)
Coefficient of permeabil
Sample
A
B
C
Before exposure
to waste
1.4 x 10-8
1.4 x lO-8
1.8 x ID'8
After exposure
to waste
2.0 x 10-8
2.1 x 10-8
3.0 x 10-8
ity (cm/s)
Waste
Electric furnace
baghouse dust
Tar decanter sludge
(high in organics)
Neutralized pickle
liquor rinse water
sludge
D 1.8 x 10-8 3.0 x 10-8 Hot strip mill recycle
system sludge (high in
oil)
aTest data reported to Pennsylvania Department of Environmental
Resources.
4-88
-------
Soil characterization and details of the test method were not provided
in the brief_report.
4o5.22.3 Studies Sponsored by Waste Management, Inc.—
Waste Management, Inc., has sponsored a series of studies to determine
compatibility between landfill leachate and in situ clay soil, which func-
tions as the landfill liner. (Weaver and Brissette, Canonie Environmental
Services Corporation, 1982). Permeability data were obtained in triaxial
devices. Each report includes details of the sampling effort, soil char-
acterization data, and the permeability test results. One important con-
clusion by Canonie is as follows:
... It has been our experience that in concentrations of less than 1
percent organics have no significant impact on soil permeability.
The laboratory permeability results presented in each of the reports
indicate that no appreciable change in permeability will occur on soils at
the specific sites due to contact with the waste leachate obtained from the
facility.
4.5.22.4 Test Data from D'Appolonia Consulting Engineers, Inc.—
D'Appolonia Consulting Engineers, Inc. (D'Appolonia), has performed
numerous permeability studies involving site-specific soil samples and
leachates. Projects have involved slurry trench walls as well as liners for
landfills. In 1983 D'Appolonia compiled their permeability test data from
14 projects (D'Appolonia Consulting Engineers, Inc., 1983). Leachates
tested in the D'Appolonia projects were frequently high in salts, and some
were highly acidic. All data were obtained from triaxial tests. Perme-
ability data from the various projects are summarized below.
Project A—No significant changes were observed in long-term perme-
ability tests on soil samples containing 2 percent or 3 percent commercial
bentonite when they were subjected to an aqueous waste fluid of pH 7. The
fluid, collected during dredging operations and shipped to D'Appolonia by
the client, contained minor concentrations of a number of inorganic salts
(sulfate, chloride, fluoride). Specific conductance was reported as 1,710
/^mho/cm.
The permeability tests involved the exchange of"up to 4.7 pore volumes
with test times of 24 to 76 days. The hydraulic gradients used varied
from 36 to 168; cell pressures were 1.0 to 1.5 kg/cm2.
Project B—Extensive permeability tests were performed on reddish-brown
clay samples (undisturbed Shelby tube samples) using a stabilized pH 4
fluid. The fluid was prepared in the laboratory using pond water and waste
from the site. The most significant characteristic of the stabilized
pH 4 fluid was the specific conductance at 94,000 //mho/cm at 25° C; pH was
3.88. It should be noted that Shelby tube samples may not always be com-
pacted to exactly the same density so that slight deviations in measured
permeability are not considered to be significant.
Although all permeabilities measured were below 1 x 10~7 cm/s, a
trend toward increased permeability was apparent as more pore volumes were
exchanged. Changes in soil chracterization were also in evidence as more
4-89
-------
pore volumes of test permeant fluid were passed through the samples.
Elevated hydpaulic gradients 110 and 450;were used for the tests, and
cell pressures were 2.5 and 4.5 kg/cm2. Test times ranged from 100 to
350 days. In three of the samples, more1than 12 pore volumes were
exchanged. !
Project C—- A waste fluid characterized by high salt concentration and
high total organic carbon was tested in four soil samples. The samples
contained varying proportions of two soils—a fine-to-coarse sand, and a
sandy clay with 1 to 3 percent commercial bentonite.
Permeability tests were run at a hydraulic gradient of 20. After ini-
tial permeability increases in some samples, the permeabilities decreased.
Samples were tested for 692 days. For each sample, between 2 and 13 pore
volumes were exchanged. The permeability decrease noted was approximately 1
order of magnitude for a cement bentonite sample. For the other three soil
mixtures, permeability decreased by a factor of 2 to 4.
The constituents present in the permeant are not known precisely.
Available chemical analysis data indicate that the chemical concentrations
1n the waste material vary considerably from year to year.
Project D—Slight decreases in permeability were reported for clay-soil
samples (with bentonite) exposed to contaminated groundwater samples taken
from piezometers. Soil samples were 25 percent flyash, 25 percent uncon-
taminated clay, 25 percent contaminated clay, and 25 percent silt.
Bentonite (1 percent) was added.
Two permeant fluids were tested; pH values were 10.57 (Sample A) and 8.8
(Sample B). Specific conductance was reported at 35,300 and 13,200 ymho/cm
at 25°C. :
Samples were tested at a gradient of 190 with cell pressure at 1.5
kg/cm2. Test duration was 90 days with more than three pore volumes
exchanged. Final permeabilities were determined to be below 1 x 10~8
cm/s.
Project E—Permeability tests were run on a silky clay soil from a pro-
posed facility with four waste leachates as the permeant fluid. Soil
samples were compacted in the laboratory (95 percent of standard Proctor
density). Chemical characteristics of the waste fluids tested are shown in
Table 4-17.
Permeability tests were performed at a hydraulic gradient of 47 and
cell pressure of 0.75 kg/cm2. Total test time was 40 to 50 days. The
results of the permeability tests are given in Table 4-18. All samples were
saturated with 0.01 N calcium sulfate solution prior to introduction of the
waste fluid.
Project F—Permeability data on undisturbed Shelby tube samples
tested with tap water showed permeability values ranging from 6.1 x 10~8
to 2.0 x 10~9 cm/s with an average of 1.0 x 10~8 cm/s. Fifty-seven
samples were tested with total test time varying from 6 to 10 days.
Hydraulic gradient was 25 to 50, and cell pressure was 3.5 to 5.5 kg/cm2.
4-90
-------
TABLE 4-17. CHEMICAL CHARACTERISTICS OF WASTE PERMEANTS, PROJECT E*
Waste leachate permeant fluid0
Parameter
Column test designation
Ph
Specific conductance
Filterable residue at
180 °C
Acidity
Alkalinity
Phenolpthalein
alkalinity
Chloride
Sulfate
Dissolved metals:
Cadmi urn
Calcium
Chromium (hexavalent)
Chromium (total)
Iron
Lead
Magnesium
Manganese
Nickel
Selenium
Sodium
Zinc
Units
pH units
^mho/cm
'at 25°C
mg/L
mg/L CaCOs
mg/L CaC03
mg/L CaCOs
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
No. 3
P-l
12.63
20,500
12,100
0
4,140
2,660
1,600
50
0.02
610
450
880
<0.1
0.49
<0.1
0.07
0.23
1.86
945
3.82
No. 4
P-2
11.23
11,900
8,620
0
130
90
2,300
2,550
<0.01
170
14
16
<0.1
0.32
0.4
0.04
0.10
0.196
1,860
0.12
Sludge
P-3
8.37
10,100
8,720
0
20
2
570
500
-
<0.01
1,800
0.06
0.08
<0.6l
135
0.14
.0.46
<0.001
12
0.02
Composite
leachate
P-4
11.23
11,000
8,380
0
224
143
1,800
2,250
<0.01
505
21
52
0.01
2.8
0.06
0.10
0.342
1,350
0.03
aData from D'Appolonia Consulting Engineers, 1983.
DLeachates were generated from various wastes by 1:4 shake extraction of
solid waste with water.
Composite leachate obtained by mixing four waste leachates in the following
proportion:
No. 3 leachate - 5 percent
No. 4 leachate - 75 percent
No. 5 leachate - 5 percent (composition not specified)
Sludge leachate - 15 percent.
4-91
-------
TABLE 4-18. RESULTS OF PERMEABILITY TESTS, PROJECT Ea
Sample
fluid
P-l
P-2
P-3
P-4
Permeant
0.01 N CaS04
No. 3 leachate
0.01 N CaS04
No. 4 leachate
0.01 N CaS04
Sludge leachate .
0.01 N CaS04
Composite leachate
Pore volumes
exchanged
0
3.2
0
0
4.7
13.5
0
0.8
- 3.2
- 12.7
- 8.8
- 4.7
- 13.5
- 17.2
- 0.8
- 5.6
Permeability
(cm/s)
1.6 x ID'7
1.7 x ID-7
1.4 x ID'7
1.1 x 10-7
1.3 x 10-7
1.4 x 10-7
1.5 x 10-7
2.4 x 10-7
4.2 x 10-7
1.1 x 10-7
1.2 x 10-7
8.1 x 10-7
aData from D'Appolonia Consulting Engineers, Inc., 1983.
4-92
-------
Project G—Slurry wall test data were provided for two different
leachates identified as low pH or high pH. A commercial bentonite product
was added to a elay-sand mix. Sixty-day tests at hydraulic gradients of
170 to 205 and cell pressure at 1.5 kg/cm2 gave permeability values below
7 x 10~8 cm/s. Between 19 and 47 pore volumes were exchanged in the
tests.
Project H--Potential clay liner materials were tested with a permeant
fluid generated in the laboratory by leaching a young slag from a pilot
plant. The pH of the leachate was 4.90; specific conductance was 2.70
jumho/cm at 25°C.
Hydraulic gradients of 25 and 100 and cell pressures of 0.75 and
1.5 kg/cm2 were used in the permeability tests. Distilled water was
used to saturate the samples. Small permeability decreases were observed in
tests with the slag leachate. Total testing time was approximately 2 to 3
months.
Project I—Groundwater spiked with several chlorinated hydrocarbons was
used as the permeant fluid in tests on undisturbed Shelby tubes samples.
The total concentration of chlorinated ethanes was 500 ppm. The soil sam-
ples tested were comprised of smectite (50 to 75 percent), kaolinite (10 to
25 percent), vermiculite (10 to 25 percent), mica (10 to 25 percent), and
quartz (10 to 25 percent).
Permeability tests were conducted at a hydraulic gradient of 150 and
a cell pressure of 1.5 kg/cm2. Four samples were tested for up to 140
days with a maximum of 24 pore volumes exchanged. Distilled water was used
as the initial permeant fluid. Slight decreases in permeability were
observed with the spiked groundwater as permeant fluid.
Project J—Two soil samples were tested with a highly acidic waste
fluid (pH = 1.5) that was collected from waste ponds. Other significant
characteristics of the fluid were specific conductance = 22,200 ^mho/cm at
25°C and sulfate = 15,000 ppm.
Slight decreases in permeability were observed after exchange of more
than 10 pore volumes. Hydraulic gradients of 20 to U)0 were used with
cell pressures of 0.7 to 2.0 kg/cm2.
Project K—Three waste fluids were tested with a sandy soil mixed with
a commerical bentonite product. Significant characteristics of the waste
fluids are shown below.
Parameter Waste fluids
pH 8 9 10
°K 6.65 7.65 5.00
Specific conductance 920 1,000 430
(jumho/cm at 25°C)
Sulfate (ppm) 470 460 180
4-93
-------
No significant change in permeability occurred when the pH 9 fluid was
tested with the sandy clay soil. The test involved exchange of 7.4 pore
volumes. A slight increase in permeability was noted with the pH 8 fluid
tested with the sandy clay soil after 4.7 pore volumes were exchanged. Waste
fluid at pH 10 tested with the sandy clay soil mixed with bentonite decreased
permeability after exchange of 4.2 pore volumes.
A hydraulic gradient of 22 was used for the tests involving the waste
fluids and soil. A hydraulic gradient of 100 was imposed on the soil
bentonite tested with pH 10 fluid. Cell 'pressure was 1.1 kg/cm2. The
pH 8 and 9 samples were tested at a hydraulic gradient of 22 and a cell
pressure of 1.0 kg/cm2.
Project L—- Two waste fluids with neutral pH were used in tests with soil
mixed with approximately 1 percent treated bentonite from three vendors. The
only notable characteristic of the waste fluids was specific conductance.
The permeability data are summarized in Table 4-19. The hydraulic gradi-
ent used in the tests was 80 to 90; cell pressure was 1.0 kg/cm2.
Project M—Permeability tests were performed on 20 soil samples to
determine the effect of a permeant fluid of pH 1.5. The fluid, collected
from waste ponds at a disposal site, had a high salt concentration
(specific conductance = 22,200 jumho cm at 25°C; sulfate = 15,000 ppm).
The only permeability Increases noted were for a soil characterized as
silt stone and one characterized as sand !stone. For these samples, perme-
ability increases were just less than 1 order of magnitude after passage of
approximately 12 pore volumes. Hydraulic gradients used in the tests
were 15 to 285; cell pressures were Oo75 to 4.00 kg/cm2.
Project N—Permeability studies on three composite soil samples were
conducted. Three waste leachates used as the permeant fluid were prepared in
the laboratory by extracting tailings from a pilot plant. Fluids tested were
characterized by pH (3, 6, or 9).
Initial permeabilities were determined with groundwater from the pro-
posed site. The results of the permeability tests are shown in Table 4-20.
Permeability increases of approximately 1 order of magnitude were observed in
the glacial till samples tested with permeant fluids of pH 6 and 9.
4.5.23 Tests Reported by Bentonite Companies
4.5.23.1 American Colloid Company— ; .
The American Colloid Company produces Volclay®soil sealants. These pro-
ducts are a special type of high swelling sodium montmorillonite that has
been treated by a proprietary process to render the material unreactive
toward most chemical materials. American Colloid Saline Seal 100®is a
patented product Intended for use in containing wastes with high levels of
dissolved salts, acids, or alkali.
4-94
-------
TABLE 4-19. RESULTS fOF PERMEABILITY TESTS, PROJECT La
-
Waste fluid
(1,900 wmho/cm
A
at 25°C)
Pore volumes
K-max/Mnitial replaced
Cement
bentonite
Aqua gel
Saline seal
1.0
2.2
•1.6
6
8
7
Waste fluid B
(3,800 /amho/cm at 25°C)
Pore volumes
fynax' ^initial replaced
1.0 13
6.9 7
3.9 13
aData from D'Appolonia Consulting Engineers, Inc., 1983.
TABLE 4-20. INITIAL AND FINAL PERMEABILITIES DETERMINED IN
TRIAXIAL CELL TESTS WITH LEACHATES, PROJECT Na
Sediment sample
Glacial till
(Composite No. 1)
Stratified drife
(Composite No. 2)
47. Bentonite/till
Admixture
Leachate
permeant
PH
3
6
9
9
3
6
9
9
3
9
Laboratory permeability @ 20ob
(cm/s)
initial with
site groundwater
5.5 x 10-8
3.8 x 10-8
5.7 x 10-8
3.6 x 10-7
1.8 X 10-5
1.5 x 10-5 ^
1.3 x 10-5
2.1 x 10-5
1.0 x 10"10
1.0 x 10'10
Final with
waste leachates
1.4 x 10-7
3.4 x 10-7
5.6 x 10-7
5.7 x 10-7
1.3 x 10-5
1.2 x lO-5
1.3 x 10-5
2.1 x 10-5
1.5 x 1010C
1.5 x 10-10d
aData from D'Appolonia Consulting Engineers, Inc., 1983.
bPermeability calculations based on final column sample dimensions.
cDetermined as 83.1 percent saturation based upon final moisture content
measurements.
dDetermined as 83.4 percent saturation based upon final moisture content
measurements.
4-95
-------
American Colloid has conducted permeability tests of Saline Seal 100
with gasoline-, kerosene, and 1,1,2-trichToroethane. The trichloroethane
tested was waste solvent that had been contaminated with other unchar-
acterlzed materials. It was said to be predominantly trichloroethane
(Jepsen, 1983).
The American Colloid tests were conducted with a fixed-wall compaction
permeameter. Inside walls of the cylinder were coated with a thin film of
slurry to provide a barrier against capillary effects along the wall.
Samples were compacted to at least 90 percent Proctor. Soil samples were
either 5.1 or 10.2 cm (2- or 4-inch) -thick cores and consisted of a uniform
standard silica sand mixed with between 6 and 15 percent bentonite (dry
weight). Test samples were prehydrated with deionized water for at least 48
hours, and provision was made for deairing. After the test fluid was added
to the permeameter, head loss was recorded periodically until stable read-
ings were established. The results of these permeability tests are shown in
Table 4-21.
4.5.23.2 Federal Bentonite--
Another major bentonite company, Federal Bentonite, produces, among
other bentonite products, petroleum tank farm sealants. The products are
made by treating sodium-bentonite with specific polymers in order to obtain
the desired sealing characteristics. PPS-21 is a free-flowing granular
bentonite product designed to promote an impermeable barrier in the event of
failure or leak In petroleum tank farms.;
Permeability tests with water and with kerosene were performed on 5.1 cm
(2-1nch) samples of test soil consisting of PPS-21 mixed with washed beach
sand. Samples were prehydrated with deionized water under a 136-cm head
prior to introduction of the kerosene. Tests results are summarized in
Table 4-22.
4.5.23.3 Discussion—
The behavior of polymer-treated bentonites over a long time period is
not demonstrated in the test results presented here. Although the duration
of the American Colloid tests exceeded 40 days, only a fraction of a pore
volume of fluid was displaced during the tests. The number of pore volumes
was not expressed in the data presented by Federal Bejitonite.
Suggestions that the polymers in the treated bentonites will degrade
after 3 to 4 years have been made, and at least two laboratories claim to
have data 1n support of this time-degradation behavior (Seattle, 1983;
Zlamal, 1983). Many applications for the treated bentonites involve short-
term performance requirements. In long-term applications, such as barriers
for landfills', the time-degradation issue could have serious implications.
The long-term viability of treated bentonite seals needs to be verified.
4-96
-------
TABLE 4-21. EFFECT OF CONCENTRATED ORGANICS ON A
TREATED BENTONITE SEAL3
Test duration
Organic permeant (days)b
Gasoline 68
Kerosene 40.2
1,1,2-Trichloroethane 71
(waste)
Pore volumes Permeability
displaced (cm/s)
0.6 4.7 x 10-7
0.16 4.7 x 10-8
0.58 4.2 x 10~7C
Unpublished data on Saline Seal 100® from American Colloid Company,
personal communication, January 23, 1983.
bAll tests conducted using a hydraulic head of 76.2 cm (2.5 feet).
Permeability of prehydrated soil was 1.5 x 10~7 cm/s prior to
addition of organic permeant.
TABLE 4-22. PERMEABILITY (cm/s) OF A TREATED BENTONITE SEAL
TO KEROSENE3»b
Sample
Sample 1:
(Prehydrated for
Sample 2:
(Prehydrated for
Sample 3:
(Prehydrated for
Sample 4:
(Prehydrated for
24 h)
48 h)
72 h)
96 h)
Prior to
addition
of kerosene
5.1 x 10-8
3.2 x 10-8
2.0 x ID'8
1,3 x 10-8
After exposure
to kerosene
for 7 days
3.4 x 10-8
2.2 x 20-8
1.3 x 10-8
1.1 x 10-8
After exposure
to kerosene
for 42 days
2.5 x 10-8
1.5 x ID"8
1.6 x 10-8
9.6 x 10-9
3Data from Federal Bentonite (1983) on tank farm sealant PPS-21.
bTests conducted under a standard 136-cm head using a falling head
permeameter.
4-97
-------
4.6 REFERENCES
Acar, Y. B. and-R. K. Seals. 1984. Clay Barrier Technology for Shallow Land
Waste Disposal Facilities. Hazardous Waste. 1(2):167-181.
Acar, Y. B., A. Hamidon, S. Field, and L.;Scott. 1984a. Organic Leachate
Effects on Hydraulic Conductivity of Compacted Kaolinite. ASTM Symposium
on Impermeable Barriers for Soil and Rock, Denver, Colorado.
Acar, Y. B., and S. D. Field. 1982. Organic Leachate Effects to Hydraulic
Conductivity in Fine-Grained Soil, Volume 1. Report NO. 6E-82/01,
Louisiana State University.
Acar, Y. B., I. Olivieri, and S. D. Field. 1984b. The Effects of Organic
Fluids on the Pore Size Distribution of Compacted Kaolinite. ASTM
Symposium on Impermeable Barriers for;Soil and Rock, Denver, Colorado.
Acar, Y. B., I. Olivieri, and S. D. Field. 1984c. Organic Fluid Effects on
the Structural Stability of Compacted Kaolinite. In: Land Disposal of
Hazardous Waste: Proceedings of the Tenth Annual Research Symposium at
Ft. Mitchell, Kentucky, April 3-5, 1984. EPA-600/9-84-007.
Anderson, D. C. 1981. Organic Leachate Effects on the Permeability of Clay
Soils. M.S. Thesis, Soil and Crop Sciences Department, Texas A & M
University, College Station, Texas.
Anderson, D. C., 1982. Does Landfill Leachate Make Clay Liners More
Permeable? ASCE. pp. 66-69.
Anderson, D. C., K. W. Brown, and J. Green. 1981. Organic Leachate Effects
on the Permeability of Clay Liners. National Conference on Management of
Uncontrolled Hazardous Waste Sites, Washington, D.C., pp. 223-229.
Anderson, D. C., and S. G. Jones. 1983. Clay Barrier-Leachate Interaction.
National Conference on Management of Uncontrolled Hazardous Waste Sites,
Washington, D.C., October 31-November 2, 1983.
Anderson, D. C., and S. G. Jones. 1984. Fate of Organic Liquids Spilled on
Soil. National Conference on Management of Hazardous Wastes and
Environmental Emergencies, Houston, Texas, March 1984. pp. 1-22.
ASTM. 1985. American Society of Testing Materials. Annual Book of ASTM
Standards. Vol. 04.08..
Seattle, B. 1983. Federal Bentonite, Montgomery, Illinois, personal
communication with Research Triangle Institute.
Bowders, J. J., D. E. Daniel, G. P. Broderick, and H. M. Llljestrand. 1986.
Methods for Testing the Compatibility of Clay Liners with Landfill
Leachate. In: Hazardous and Industrial Solid Waste Testing: Fourth
Symposium, ASTM STP 886, American Society for Testing and Materials,
Philadelphia, Pennsylvania, 1986, pp. 233-250.
4-98
-------
Brown, K. W., and D. C. Anderson. 1980. Effect of Organic Chemicals on Clay
Liner Permeability: A Review of the Literature. In: Disposal of
Hazardous Waste: Proceedings of the Sixth Annual Research Symposium at
Chicago, Illinois, March 17-20, 1980, D. Shultz, ed. EPA-600/9-80-010.
Brown, K. W., and D. C. Anderson. 1983. Effects of Organic Solvents on the
Permeability of Clay Soils. EPA 600/2-83-016. U.S. Environmental
Protection Agency, Cincinnati, Ohio.
Brown, K. W., and J. C. Thomas. 1984. Conductivity of Three Commercially
Available Clays to Petroleum Products and Organic Solvents. Hazardous
Waste. 1(4):545-553.
Brown, K. W., J. Green, and J. Thomas. 1983. The Influence of Selected
Organic Liquids on the Permeability of Clay Liners. Land Disposal of
Hazardous Waste: Proceedings of the Ninth Annual Research Symposium at
Ft. Mitchell, Kentucky, May 2-4, 1983. EPA-600/9-83-018.
Brown, K. W., J. C. Thomas, and J. W. Green. 1984. Permeability of
Compacted Soils to Solvents Mixtures and Petroleum Products. In: Land
Disposal of Hazardous Waste: Proceedings of the Tenth Annual Research
Symposium at Ft. Mitchell, Kentucky, April 3-5, 1984. EPA-600/9-84-007.
Bryant, J., and A. Bodocsi. 1986. Precision and Reliability of Laboratory
Permeability Measurements, EPA-600/2-86-097, U.S. Environmental
Protection Agency, Cincinnati, Ohio, 177 pp.
Buchanan, P. N. 1964. Effect of Temperature and Adsorbed Water on
Permeability and Consolidation Characteristics of Sodium and Calcium
Montmorillonite. Ph.D. Dissertation, Texas A&M University, Colleqe
Station, Texas.
Buelt, 0. L., and S. Barnes. 1981. A Comparative Evaluation of Liner
Materials for Inactive Uranium-Mi 11-Tailings Piles. Prepared for the
U.S. Department of Energy under Contract No. DE-AC06-76RLO 1830.
Buettner, W., and M. D. Haug. 1983. Permeability Testing of Fine Grained
Soils. In: Potash Technology Mining, Processing* Maintenance,
Transportation, Occupational Health and Safety, Environment: Proceedings
of the First International Potash Technology Conference, October 3-5,
1983. Saskatoon, Saskatchewan, R. M. McKercher, ed. Pergamon Press,
Toronto, Canada.
Chapman, D. L. 1913. A Contribution to the Theory of Electrocapillaritv.
Philosophical Magazine. 25(6):475-481.
Coia, M. F. 1981. The Effect of Electroplating Wastes Upon Clay as an
Impermeable Boundary to Leaching. M.S. Thesis, Department of Civil and
Environmental Engineering, Duke University, D.urham, North Carolina.
Daniel, D. E. 1982. Effects of Hydraulic Gradient and Field Testing on
Hydraulic Conductivity of Soil. U.S. EPA Cooperative Agreement CR 810165
to the University of Texas, Austin.
4-99
-------
Daniel, D. E. 1983. Third Quarterly Progress Report to the U.S. EPA for the
Project:_ Effects of Hydraulic Gradient and Field Testing on Hydraulic
Conductivity of Soil. U.S. EPA Cooperative Agreement CR 810165 to the
University "of Texas, Austin. For the period February 1983 to May 1983.
Daniel, D. E., and H. M. Liljestrand. 1984. Effects of Landfill Leachates
on Natural Liner Systems. A Report to Chemical Manufacturers Association
by Department of Civil Engineering, University of Texas, Austin.
D'Appolonia Consulting Engineers. 1983.! Data Compilation and Presentation
for Column Testing Associated with Hazardous Waste Isolation. Prepared
for Research Triangle Institute by DIAppolonia Consulting Engineers,
Inc., Pittsburgh, Pennsylvania.
Dunn, R. J. 1983. Hydraulic Conductivity of Soils in Relation to Subsurface
Movement of Hazardous Wastes. Ph.D. Dissertation, Department of Civil
Engineering, University of Cal-ifornia, Berkeley.
Evans, J. C., R. C. Chaney, and H. Fang. 1981. Influence of Pore Fluid on
Clay Behavior. Department of Civil Engineering, Lehigh University,
Bethlehem, Pennsylvania. Sponsored by Woodward-Clyde Consultants,
Plymouth Meeting, Pennsylvania. Fritz Engineering Laboratory Report No.
383.14, December 4, 1981.
Everett, E. E. 1977. Suitability of the Lacustrine Clays in Bay County,
Michigan for Land Disposal of Hazardous Waste. M. S. Thesis, The
University of Toledo, Toledo, Ohio.
Foreman, D. E., and D. E. Daniel. 1984. Effects of Hydraulic Gradient and
Method of Testing on the Hydraulic Conductivity of Compacted Clay to
Water, Methanol, and Heptane. In: Land Disposal of Hazardous Waste:
Proceedings of the Tenth Annual Research Symposium at Ft. Mitchell,
Kentucky, April 3-5, 1984. EPA-600/9-84-007.
Fuller, W. H. 1974. The "State of the Art" of Migration and Attenuation of
Some Potentially Hazardous Polluting Trace and Heavy Metals, Asbestos and
Cyanide in Soil. University of Arizona, Tucson, Arizona.
Gouy, G. 1910. Sur La Constitution de la Charge Eleetrique a la Surface d'un
Electrolyte, Anniue Physique (Paris). 4(9):457-468.
Gray, D. H., and U. U. Stoll. 1983. Readers Write—Leachate Land Liners.
ASCE. 52(1):20. ;
Green, J., G. Lee, R. Jones, and Ted Palit. 1983. Interaction of Clay Soils
with Water and Organic Solvents: Implications for the Disposal of
Hazardous Wastes. Environmental Science and Technology. 17:278-282.
Green, W. J., G. F. Lee, and R. A. Jones. 1979. Impact of Organic Solvents
on the Integrity of Clay Liners for Industrial Waste Disposal
Pits—Implications for Groundwater Contamination. Final Report to U.S.
EPA, Ada, Oklahoma.
4-100
-------
Green, W. J., G. F. Lee, and R. A. Jones. 1981. Clay-Soils Permeability and
Hazardous. Waste Storage. Journal Water Pollution Control Federation.
54(8):1347-.1354. "
Griffin, R. A., and W. R. Roy. 1985. Interaction of Organic Solvents with
Saturated Soil-Water Systems. Report prepared for the Environmental
Institute for Waste Management Studies, University of Alabama by the
Illinois State Geological Survey. December 1984.
Grim, R. E. 1968. Clay Mineralogy (second edition). McGraw-Hill, New York.
596 pp.
Hamilton, J. M. 1979. Measurement of Permeability of Partially Saturated
Soils. M.S. Thesis, University of Texas, Austin, Texas, 1979.
Jepson, C. 1983. American Colloid Company, Skokie, Illinois, personal
communication with Research Triangle Institute.
Lambe, T. W. 1958. The Structure of Compacted Clay. Journal of the Soil
Mechanics and Foundations Division, ASCE. 84(SM2):34.
Lohman, S. W., et al. 1972. Definitions of Selected Groundwater Terms-
Revisions and Conceptual Refinements. U.S. Geological Survey Water-
Supply Paper No. 1988.
Macey, H. H. 1942. Clay-Water Relationships and the Internal Mechanism of
Drying. Transactions of the British Ceramic Society. 41(4):73-95.
Michaels, A. S., and C. S. Lin. 1954. Permeability of Kaolinite.
Industrial and Engineering Chemistry. 46(6):1239-1246.
Mitchell, J. K. 1976. Fundamentals of Soil Behavior. John Wiley and Sons,
Inc., New York, New York.
Mitchell, J. K., and J. S. Younger. 1967. Abnormalities in Hydraulic Flow
Through Fine-Grained Soil. ASTM STP 417. pp. 106-139. -
Mitchell, J. K., D. R. Hooper, and R. G. Campanella. 1965. Permeability of
Compacted Clay. Journal of the Soil Mechanics and Foundation Division,
ASCE, 91:41-66
Monserrate, M. L. 1982. Evaluation of the Hydraulic Conductivity of Two
Clays Exposed to Selected Electroplating Wastes. M.S. Thesis, Duke
University, Durham, North Carolina.
Olivieri, I. 1984. The Effect of Organic Fluids on the Fabric of Compacted
Clay. M.S. Thesis, Department of Civil Engineering, Louisiana State
University, Baton Rouge.
Olsen, H. W. 1965. Deviations From Darcy's Law in Saturated Clay. In:
Soil Science Society of America Proceedings, Volume 29. pp. 135-140.
4-101
-------
Pask, J. A., and B. Davies. 1945. Thermal Analysis of Clays and Acid
Extraction of Alumina from Clays. U.,S. Bureau of Mines, Technical Paper
664. pp. 56-78.
Peirce, J. J. 1984. Effects of Inorganic Leachates on Clay Liner
Permeability. Final Report. U.S. EPA Contract No. 68-03-3149, 24-1.
Department of Civil and Environmental Engineering, Duke University.
September 28, 1984.
Peirce, J. J., and T. A. Peel. 1985. Effects of Inorganic Leachates on Clay
Soil Permeability. In: Land Disposal of Hazardous Waste: Proceedings
of the Eleventh Annual Research Symposium. EPA/600/9-85/013. April.
Peirce, J. J. and K. A. Witter. 1986. Termination Criteria for Clay Per-
meability Testing. ASCE Journal of Geotechnical Engineering, September
1986. 112(9) .-841-854.
Rad, N. S., and Y. B. Acar. 1984. A Study on Membrane-Permeanto
Compatibility. Geotechnical Testing Journal. 7(2).
Reeve, R. C., and G. H. Tamaddoni. 1965. Effect of Electrolyte Concentra-
tion on Laboratory Permeability and Field Intake Rate of a Sodic Soil.
Soil Science. 99(4).
Sanks, R. L., and E. F. Gloyna. 1977. Clay Beds for Storing Solid
Industrial Wastes~A Survey. In: Proceedings of the 32nd Industrial
Waste Conference, Purdue University, Lafayette, Indiana. Ann Arbor
Science, Ann Arbor, Michigan.
Schramm, M. 1981. Permeability of Soils to Four Organic Solvents and
Water. M.S. Thesis, Department of Hydrology and Water Resources,
University of Arizona.
Shepard, T. A. 1981. Geochemical Retardation Processes, Subsurface
Contaminant Migration from Mine and Mill Waste Impoundments. Colorado
State University.
Socash, S. M. 1982. Liner Compatibllty Information. Commonwealth of
Pennsylvania Department of Environmental Resources, Meadville,
Pennsylvania. .
Theng, B. K. G. 1984. The Chemistry of Clay-Organic Reaction. John Wiley
and Sons, New York.
Tulls, D. S. 1983. Response of Clay Permeabilities to Simulated Inorganic
Textile Wastes. M. S. Thesis, Department of Civil and Environmental
Engineering, Duke University, Durham, North Carolina.
van Olphen, H. 1963. An Introduction to Clay Colloid Chemistry, Wiley
Intersdence, New York.
van Schaik, J. C. 1970. Soil Hydraulic Properties Determined with Water and
with a Hydrocarbon Liquid. Canadian Journal of Son Science. 50:79-84.
4-102
-------
van Schaik, J. C., and G. E. Laliberte. 1968. Soil Hydraulic Properties
Affected by Saturation Technique. Canadian Journal of Soil Science.
49:95-102-.
Weaver, J. W., and R. F. Brissette. 1982. Personal communication with Waste
Management, Inc., three reports. A, B, and C.
Weiss, A. 158. Interlamellar Swelling as a General Model of Swelling
Behavior. Chem. Ber. 91:487-502.
White, R. 1976. The Permeability of a Clay Material Taken from a Proposed
Toxic Waste Landfill Site as Affected by Different Liquids. Personal
communication from Trinity Engineering Testing Corp. to the City of
Corpus Christi, Texas.
Yong, R. N., and B. P. Warkentin. 1975. Soil Properties and Behavior.
Elsevier Scientific Publishing. Company, New York.
Zimmie, T. F., J. S. Doynow, and J. T. Warden. 1981. Permeability Testing
of Soils for Hazardous Waste Disposal Sites. In: Proceedings of the
Tenth International Conference on Soil Mechanics and Foundation Testing,
Vol. 2. Stockholm, Sweden, pp. 403-406.
Zlamal, F. 1983. Slurry Systems Contractors and Consultants, Gary, Indiana,
Personal communication with Research Triangle Institute.
Zoeller, A. 1982. Review of Texas A&M University Studies on Clay Soils and
Organic Reagents. Commonwealth Edison, Chicago. Appendix C to Comments
of Utility Solid Waste Activities Group, The Edison Electric Institute
and The National Rural Electric Cooperative Association on Sections 3004
and 3005 of the Resource Conservation and Recovery Act of 1976:
Permitting Standards for Land Disposal Facilities.
4-103
-------
-------
CHAPTER 5
CURRENT PRACTICES: CLAY LINER DESIGN AND INSTALLATION
This chapter describes the state-of-the-art knowledge and techniques
currently used for clay liner design and installation. Information in this
chapter was compiled from a variety of sources, including a review of exist-
ing literature on design and installation for various applications and inter-
views with key personnel from selected engineering firms, waste management
companies, government agencies, professional associations, and contractors,
all of whom are experienced in various aspects of clay liner installation and
construction. The number of clay-lined facilities designed specifically for
hazardous waste containment and information on these sites are somewhat
limited. In addition, the design and installation practices for clay liners
for other applications are similar to those for hazardous waste facilities
(though possibly not as rigorous in design and installation procedures). For
these reasons, information from these applications is included as well.
Clay liner construction utilizes equipment and techniques similar to
those used in other earthwork construction projects. However, the soil
compaction practices used to achieve low permeability employed in clay liner
construction differ from compaction practices used for strength and stabil-
ity, as used in foundations and roadbeds. For cohesive soils (i.e., clays),
maximum strength is achieved when soil is compacted at dry of optimum
moisture content (see Chapter 3 for a discussion of compaction parameters),
but the lowest permeability is achieved 2 to, 3 percent wet of optimum
moisture content. Therefore, when clay liners are compacted, it is necessary
to specify and control carefully compactive effort, density, and moisture
content to ensure that the desired permeability is achieved.
The first two sections of this chapter describe current practices for
designing and constructing clay liners. Literature reviews and interviews
with design engineers and contractors indicate that construction quality
assurance (CQA) and construction quality control (CQC) are considered among
the most important elements of successful facility construction, with clay
liner failures often attributed to deficiencies in these areas. Because of
the importance of CQA and CQC activities during the construction of clay
liners, this subject is covered separately in the third section of this
chapter, including a description of the element of a CQA plan. Common
problems encountered during clay liner construction and solutions to these
problems are tabulated in the final section of this chapter.
5.1 DESIGN
The fundamental aspects that must be considered during the design of a
clay liner are:
t Stability of the liner against major earth movements such as
slope failure, settlement, and bottom heave
5-1
-------
• Resistance of the liner to fluid flow (I.e., permeability)
• Compatibility of the liner material with the wastes 1t 1s
meant-to contain
• Long-term durability of the liner.
The design effort applies standard geotechnical engineering practices to
address these considerations and thus to meet the special performance
requirements of clay liners.
Clay liner design is very site and material specific. Waste volume
requirements and waste characteristics must be considered during the design
effort, and facility designs must be tailored to the site-specific condi-
tions. The design effort may be divided into the following activities:
• Site investigation
• Liner material selection and characterization
t Facility design
• Preparation of construction specifications and the quality
assurance (QA) plan.
Site investigation, liner material selection, and Uner material characteri-
zation are done prior to facility design. Preparation of construction
specifications and the QA plan follow facility design. Most of the design
engineers interviewed during the course of this project emphasized that
design activities usually continue through the construction of the clay liner
because unexpected situations often arisfe that necessitate modifications in
the original plans.
5.1.1 Site Investigation
As with any earthwork project, a clay-lined hazardous waste containment
facility must be designed to be compatible with the geological conditions at
the specific site. For this reason, the first step in the design effort is a
comprehensive evaluation of the site. Ideally, sites should be selected
according to the suitability of the in situ earth materials for containing
hazardous wastes. However, in actual practice this is not always possible-
Constraints such as land use, zoning laws, land ownership, and distance to
waste generators often result in sites being selected for reasons other than
their technical suitability for containing wastes. This fact, combined with
the subsurface heterogeneity and spatial variability that is the rule 1n most
geologic environments, makes adequate site investigation a critical part of
facility design.
5.1.1.1 Purpose—
Site investigations are conducted to delineate a site's topography, sub-
surface geology, and hydrogeology. Topography influences facility configura-
tion and drainage system design (runon/runoff control). Subsurface site
Investigations are necessary to determine whether soils suitable for liner
material are available at the facility site or whether it is necessary to
5-2
-------
identify and investigate borrow sources. In addition, knowledge of in situ
soil properties is important for foundation design. Soil characteristics
influence seTection of the method of slope stability analysis appropriate for
facility destpr are a necessary input to stability analyses, and determine
the necessity for special design measures to control settlement or to ensure
maximum protection against contaminant migration.
Hydrogeologic information about the site is important for the proper
monitoring of well placement. From a design standpoint, it is important to
determine the depth to water table for the site, including seasonal variabil-
ity. Some States require a specific thickness of unsaturated soil between
the facility base and the water table. For sites with high water tables,
this can necessitate aboveground facility design. In States where facilities
can extend below the water table, high groundwater levels can necessitate
special intergradient (below water table) designs. Groundwater elevations
are necessary for assessing liquefaction potential for in situ soils where
significant seismic ground motion .can occur. Hydrogeologic investigations
also are necessary to locate, identify, and delineate hydrologic pathways
(e.g., fractures and sand seams) at the site so that provisions for sealing
them can be incorporated into the facility design. These pathways can con-
tribute to rapid migration of wastes from the facility if a liner leak
occurs. In addition, when liners constructed below the groundwater table
intersect these pathways, hydraulic pressures can build against the outside
of the liner. In unfilled facilities this can result in heaving, slope
failure, and liner rupture.
5.1.1.2 Approach and Methodology—
The investigation of a facility site should address the following:
• Regional and site-specific investigations to relate the site
geology to the regional geological picture.
• Topography, including drainage patterns.
• Analyses of representative soil samples. Important tests can
include Atterberg limits, particle size distribution, shrink/
swell potential, cation exchange capacity, total organic
carbon, mineralogy, shear strength, dispersivity, compres-
sibility, consolidation properties, density and moisture
content, Proctor density, laboratory (compacted) permeability,
and chemical compatibility.
t In-place soil characteristics including depth to bedrock,
in-place permeability, and the presence of features that can
act as failure planes or hydrologic pathways (e.g., slicken-
sides, fractures, faults, silt and sand lenses and seams, and
root holes).
• Bedrock characteristics including type, form, fractures, solu-
tion cavities, and joints.
• Hydrogeologic site characteristics including depth of the
water table, horizontal and vertical flow components, hydro-
geologic pathways, seasonal variability, and location and use
of aquifers.
5-3
-------
• Land use and ownership.
t Climate.
This information 1s necessary for facility design and, to some extent, for
planning efficient borrow site development.
5.1.1.2.1 Indirect Methods—Indirect investigative methods include
collection of existing site information and remote sensing techniques. These
methods do not require drilling or excavation and are appropriate for the
initial stages of site investigation.
Site investigations usually begin with compilation and review of exist-
ing information pertinent to the site. Sources of information include Soil
Conservation Service County Soil Surveys, U.S. Geological Survey topographic
and surficial geology maps, published literature, State geological survey
information, and county records of geotechnical tests associated with
previous construction projects. This information can be very useful for
planning the scope and approach of further site investigation activities.
Geophysical remote sensing techniques that can be applied during site
investigation include electrical survey methods, ground-penetrating radar,
and seismic refraction. All of these techniques are conducted on the surface
but provide information about the subsurface. The selection of geophysical
techniques depends to a large degree on the geologic setting (White and
Brandwein, 1982).
Electrical resistivity surveying can be used to delineate the depth of
the water table as well as the presence of subsurface layers or lenses of
different permeability that have .contrasting resistivities (e.g., clay and
sand layers). However, electrical resistivity methods cannot be applied in
certain geologic settings where general subsurface resistivity 1s relatively
high and are best used in areas (e.g., the Atlantic Coastal Plain) where
electrical resistivities of subsurface materials contrast strongly (White and
Brandwein, 1982). Further information on electrical surveying may be found
in the U.S. Environmental Protection Agency (1978) and Freeze and Cherry
(1979).
<»
Seismic refraction surveys can give valuable information about the depth
to bedrock, the subsurface bedrock topography, and the condition (fracturing)
of the bedrock (Cichowicz et al., 1981). In addition, the seismic velocity
of a geologic material is altered by the degree of weathering and water
saturation and therefore can provide information about the variability of
these parameters in the subsurface. However, because of the multitude of
variables that can affect a material's characteristic seismic velocity,
seismic results can be difficult to interpret, especially in areas with
complex subsurface geology or In areas where there is little contrast in
seismic propagation velocities in the subsurface. For this reason, limited
exploratory drilling will usually be necessary in conjunction with seismic
surveys to confirm interpretations based on this technique (Cichowicz et a!.,
1981). More detailed information on seismic refraction surveying may be
found in Dobrin (1960).
5-4
-------
Ground-penetrating radar also has some utility to site investigations
for locating buried structures and pipes and for indicating depth to shallow
bedrock. However, it is limited by a shallow depth of penetration when
compared to ottrer techniques (White and Brandwein, 1982).
The advantage of indirect techniques during the early stages of site
investigation is that their use can reduce drilling costs and costs asso-
ciated with laboratory tests and analysis. Much information about a site,
including an indication of its technical suitability as a containment facil-
ity site, can be gained at a relatively low cost. In addition, the informa-
tion gathered indirectly can be used to plan direct site investigations,
ensuring that these are carried out as efficiently and economically as
possible.
5.1.1.2.2 Direct Methods—Direct methods of site investigation include
drilling boreholes and wells and excavating pits and trenches. The purpose
of these methods is to expose subsurface material so that the physical condi-
tions can be observed and measured (e.g., faults, slickensides, sand seams,
depth to bedrock and to the water table, penetration tests, and in situ
permeability) and to obtain samples of subsurface material for laboratory
testing of engineering properties.
Direct investigations are conducted during the final stages of site
investigation and must provide sufficient information for input to the facil-
ity design. The scope of investigation necessary to accomplish this goal
will vary from site to site according to the complexity of the subsurface
geology, the potential for seasonal variability in site conditions, and the
amount of information about the site that is already available.
Regulatory personnel from the Wisconsin Department of Natural Resources
have recently published the following recommendations for site investigations
for clay-lined landfills (Gordon et al., 1984):
• Drill an adequate number of soil borings across the site to
characterize the soil deposits within and beneath the site.
The borings should extend a minimum of 25 feet below the
anticipated site base grade or to the water table, whichever
is deeper.
«
• Install a sufficient number of water table observation wells
and piezometers to define both the horizontal and vertical
groundwater flow directions.
• Excavate backhoe pits on a grid pattern across potential clay
borrow sources to characterize their depth, areal extent, and
uniformity and to obtain samples of the clay material for
testing.
• Perform appropriate laboratory tests on samples from the
potential clay borrow sources to determine if they will meet
the design specifications.
These recommendations are presented as examples; detailed exploration needs
are site and facility specific.
5-5
-------
Subsurface heterogeneities can lead to increased permeability (seepage)
or loss in strength in the foundation. Where these are suspected, it may be
appropriate to "dig test pits and trenches to identify and determine the
prevalence of these features. Downhole television monitors also can be used
in boreholes to identify important subsurface features such as faults,
fractures, slickensides, and zones of permeable material.
Accessible (pits, trenches, and large boreholes) and inaccessible (bore-
holes and wells) methods of site investigation are summarized in Tables 5-1
and 5-2. These methods are discussed in more detail in U.S. Department of
Interior (1974). Methods of obtaining disturbed and undisturbed samples
during both accessible and inaccessible site exploration are discussed in
U.S. Department of Interior (1974) and ASTM (vol. 04.08, 1985). Methods of
conducting laboratory tests of engineering properties on these samples are
discussed in Chapter 3 and Appendix A of this document. Detailed discussions
of general geotechnical site investigation techniques may be found in
Winterkorn and Fang (1975).
Hydrogeologic site investigations are necessary for planning the ground-
water monitoring system and for estimating hydraulic stresses that may act on
the facility so that they may be properly considered during facility design.
Further information on conducting hydrogeologic investigations and on
installing monitoring wells and piezometers may be found in U.S. Environ-
mental Protection Agency (1983, 1986a), Fenn et al. (1977), Johnson Division
(1975), and Lutton et al. (1983).
5.1.2 Liner Material Selection and Characterization
Soil liner materials are selected based on their ability to meet
specific performance standards and the costs to bring the material onsite.
Requirements that must be met for a soil to perform properly as a liner
material include:
• Low permeability (usually < 1 x 10~7 cm/s) when compacted
• Sufficient strength to support itself and the overlying facility
components without failure when compacted to the required
permeability and thickness *
• Compatibility with waste or waste leachate to be contained (i.e.,
no significant loss in permeability or strength when exposed to
waste or waste leachate).
The U.S. Environmental Protection Agency has compiled data on the
characteristics of soils used for constructing liners in a variety of
locations nationwide (Elsbury et al., 1985; Ely et al., 1983). These data
are presented in Tables 5-3 and 5-4. In addition to soil characteristics,
cost considerations also enter into material selection when liner material
must be brought from offsite.
The In-place, native soil at the facility site is the ideal Uner
material from the standpoint of cost and convenience; this material will be
excavated during foundation preparation and therefore does not need to be
transported to the site. If the native soil is not suitable as a liner
5-6
-------
TABLE 5-1. ACCESSIBLE METHODS OF SUBSURFACE EXPLORATION
Methods
Procedure
Type of soil and
in-place condition
Limitations
Use
Trenching
Cuts
Test pits
in
i
Accessible boring
Excavate 3 ft min. width
by hand, dragline, power
shovel, bulldozer prefer-
able; explosives if neces-
sary; min. bracing or
slope unstable soils.
Same as trenching,
performed on gentle to
fairly steep slopes;
steps up slope may be
necessary.
Excavate rectangular hole,
3 ft by 5 ft min. at
working level, by hand or
hand-operated power tools;
explosives if necessary.
Cribbing required over
S ft depth. Log and
sample as excavation
progresses when sheeting,
Inclined poling, or
notched-box cribbing Is
required in unstable
soils and for ground-
water control.
Drill 28 in. min. dia.
hole, using heavy power-
operated disc, bucket,
helical augers, single
tube or core barrels in
stable soils; log and
sample as excavation
progresses; casing required
for protection during sam-
pling and inspection.
Coarse-grained soils,
Including those containing
large quantities of gravel
and cobbles, and soft
weathered rock; and all
fine-grained soils, dense
consolidated, wet or satu-
rated or dry and hard;
loose unconsolldated, wet
or saturated and soft or
dry and granular.
Depth about 20 ft or
to groundwater or
unstable material.
Depth to 50 ft,
Infrequently 80 ft,
or groundwater if
pervious strata and
high flow.
Depth of 100 ft in
soil, 150 ft in rock.
Requires heavy drill
rig.
Access for logging and disturbed sampling for laboratory
test, for reconnaissance and feasibility design stage;
and for hand-cut undisturbed sampling for final design
or for field tests such as field density, permeability,
full-sized bearing capacity tests. Unsatisfactory in
unstable cohesionless soils. Economical and besjt
method for shallow explorations of borrow, foundation,
and aggregate deposits. '
Use same as above, except is more expensive and used in
areas of limited access by heavy equipment and for
greater depths. Best method for "hand cut" undisturbed
sampling except in unstable soils or below groundwater.
Nonaccesslble methods, Table 5-2, recommended for undis-
turbed sampling of fine-grained unstable soils bel-pw
water table.
Use same as above for stable soils in place of test
pits; very economical if equipment is available and
area is accessible.
(continued)
-------
TABLE 5-1. (continued)
Methods
Procedure
Type of soil and
in-place condition
Limitations
Use
Accessible caissons
Tunnels and drifts
Blasting
Same as accessible borings;
casing and air pressure
required in unstable soils.
Excavate accessible holes,
5 by 7 ft min., using hand
or hand-operated power
tools, lagging required.
Expose strata using ex-
plosives and hand or
power tools.
Same as above but primarily
for consolidated dry soils
and bedrock.
Sane as above, used
only when caisson 1s
part of construction.
Expensive, used only
under special conditions.
Limited to exposed
faces or outcrops.
Limited use, used primarily in establishing footing
grade during construction for Individual caissons,
under very poor foundation conditions and/or under
water.
I
Limited use, for final exploration of dan-site Abutments
when other methods have disclosed questionable condi-
tions that cannot be resolved otherwise.
Use to expose rock faces and outcrops for rip-rap and
crushed aggregate sources and to indicate size and shape
of particles that may be expected during quarrying.
Source: U.S. Department of Interior, 1974.
ui
00
-------
TABLE 5-2. NONACCESSIBLE METHODS OF SUBSURFACE EXPLORATION
Methods
Procedure
Type of soil and
in-place condition
Limitations
Use
Auger boring (hand)
Auger boring (power)
Drive-tube boring
O1
l
Percussion Jchurn)
drilling
Wash boring
Rotate and force auger
bit into soil, withdraw
and empty when full.
Auger bits, 2 to 8 in.,
helical or post hole.
Same as above, using
powered drill rigs.
auger bits, 4 to 24 in.,
helical, disc, or bucket.
Over 28 in. considered
to be accessible.
Force open pipe or tube,
with sharpened edges,
without rotation, into
soil; withdraw and
remove soil. Thin- or
thick-wall tubing or
pipe, 2 to 8 in. dia.
Chopping and cutting
action by Impact of heavy
chisel-edged bit. Water
added and cuttings form
slurry that is removed
intermittently by pump or
bailer. For holes larger
than 4 in.
Chopping and cutting by
Impact and twisting action
of lightweight bit, and »
jetting action of circulat-
ing water to remove cut-
tings. For holes from 2
to over 8 in. dia.
Fine-grained cohesive, fairly
hard to soft or fine-grained,
noncoheslve, dense to loose,
weakly cemented, or dry or
moist; with particles 1/4 in.
to 1-1/2 in. depending upon
size of auger.
Fine-grained as above, and
coarse-grained soils with
particles as large as 3 In.
depending upon auger.
Fine-grained cohesive and
slightly cohesive soils such
as loess, firm to soft clays,
and silts.
Coarse-grained soil
containing cobbles and
boulders, and hard, dense,
fine-grained soils and
rock.
Fine- or coarse-grained
soils, with small
amounts of gravel
and few cobbles;
fairly hard to soft;
weakly cemented to
loose; above or below
water table.
About 20 ft, 80 ft with
tripod; unsatisfactory
1n unstable cohesionless
soils below groundwater;
slow in hard soils.
Economical depth about
40 ft, over 100 ft
with special equipment;
unsatisfactory in
unstable cohesionless
soils below groundwater;
slow In hard, dense soil.
About 80 ft depending
upon equipment. Not
satisfactory In coarser
fine-grained soils, clean
sands, or cohesionless
soils below water table.
Unsatisfactory in
unstable soil or
fractured rock; no
information for log-
ging or samples for
classification.
No Information for
logging or samples for
classification; slow
in hard or cemented
layers.
(1) Advance hole. (2) Data for logging. (3) Represen-
tative disturbed samples for classification, index
tests, and standard properties tests. (4) Access for
field penetration and permeability tests. (5) Access
for undisturbed sampling. ' I
Same as above.
Same as above.
Used with other methods to advance hole through hard,
cemented strata, coarse gravel, boulders, or other
obstructions.
(1) Used with other methods to advance hole particularly
through unstable soils requiring casing. (2) Penetrate
fine-grained soils to establish depth to bedrock.
(3) Drill holes for groundwater observation.
(4) Provide access for sampling and penetration testing
of Impervious soils above groundwater or pervious
or impervious soils below.
(continued)
-------
TABLE 5-2 (continued)
Methods
Procedure
Type of soil and
tn-place condition
Limitations
Use
Jetting
Rotary drilling
Rotary drilling
Ul
t-»
o
Continuous sampling
High-velocity water jet
directed downward from
pipe raised and lowered
In short strokes; erodes
soil, which is carried
upward by water. For
holes 2 In. to over 10 in.
dia.
Power rotation of bit;
cuttings removed by cir-
culation of drilling mud
or water; holes 1-1/2 In.
to over 10 In. dia.
Power rotation of bit;
cuttings removed by cir-
culation of air. Holes
2 In. to over 10 in. dia.
Drive-tube boring or
rotary drilling (core bor--
1ng) that provides samples
as a result of advancing
the hole.
Fine- or coarse-grained
soils; weakly cemented
noncoheslve or cohesive;
above or below water
table.
Fine- or coarse-grained,
compact or cemented soils,
and rock.
Fine- or coarse-grained,
compact or cemented
soils and rock.
Ho information for
logging or samples for
classification; slow
in hard cohesive soils.
Ho Information for
logging or samples for
classification;
difficult in loose,
coarse-grained soil with
cobbles and boulders.
Information for logging
and samples for
classification;
unsatisfactory In loose
coarse-grained soil
with cobbles and
boulders.
Depends upon the
method selected.
Same as (1), (2), and (3) for wash boring.
(1) Advance hole. (2) Access for field penetration test
(not suitable for well permeameter test or groundwater
observation If drilling mud used). (3) Access for dis-
turbed or undisturbed sampling.
(1) Advance hole. (2) Access for field penetration
test. (3) Well permeameter test. (4) Groundwater
observation. (5) Access for disturbed and undisturbed
sampling. (6) Advance hole to install casing for
nuclear moisture-density probes.
Source: U.S. Department of Interior, 1974.
-------
TABLE 5-3. PROPERTIES OF SOILS USED TO CONSTRUCT SOIL LINERS*
01
Geographic
Location
Alabama
Al abama
Al abama
California
California
California
California
Colorado
Georgia
1 1 1 i noi s
I ndi ana
Indiana
Michigan
Michigan
New York
Ohio
Oklahoma
South Carolina
Texas
Texas
Utah
Geologic
Origin
residual
residual
sedimentary
sedimentary
residual
sedimentary
sedimentary
residual
residual
glacial
glacial
glacial
glacial
glacial
glacial
glacial
residual
sedimentary
sedimentary
sedimentary
sedimentary
Percent
Passing
No. 200
Sieve
96
75
75
78
95
95
>30
40-80
85
--
57
--
80
80
—
90-98
--
—
--
75-90
Atterberg
Limits
Liquid
L i mi t
62-79
32-48
40-90
35
__
35-60
50-65
--
44-70
19-29
—
23
36
30
30
--
30-45
56
37
—
20-40
Plasticity
Index
29-53
14-28
15
_ _
above A line
above A line
__
20-34
6-15
__
12
20
15
15
—
18-25
35
19
__
5-20
Unified
Classification
CH, MH
CL, CH
CL
CL, CH
CH
. CL, CH
CL, ML
CL
CL
CL
CH
CL,CH
CH
CL
CL, ML
Hydraulic
Conductivity
cm sec-3 '
— 1 —
<1.0 x 10-7
<1.0 x 10-7
~6.9 x 10~^
3.5 x 10-6
<1.0 x 10"7
^ J. • W /\ J.V/
<1.0 x 10-7
<1.0 x 10-7
<1.0 x 10-7
6.0 x 10-7
<1.0 x 10-7
<1.0 x 10~7
~8.5 x 10-8
<1.0 x 10-7
<1.0 x 10-8
<1.0 x 10-8
2.4 x 10-8
<1.0 x ID'7
<1.0 x 10-7
2.9 x 10-8
6.0 x ID"9
<1.0 x 10-7
aData from Part B Permit Applications.
-------
TABLE 5-4. PROPERTIES OF SOILS USED TO CONSTRUCT SOIL LINERS*
01
I—«
ro
Geographic
Location
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Geologic
Origin
glacial
__ •
• __
_ _
_ _
__
glacial
—
__
__
__
—
Percent
Passing
No. 200
Si eve
31
—
84
78
86
70
80
--
—
__
__
—
Atterberg
Limits
Liquid
Limit
28
33-51
31
21-54
40-43
38
29
16-27
25-40
.. . 45
28-39
11-13
42
Plasticity
Index
7
16-32
14
8-30
22-24
20
11
3-10
5-9
18
15-17
3
18
Unified
Classification
CL,ML
—
CL
CL.CH
CL
CL
CL,CH
__
CL
__
__
CL
Hydraulic
Conductivity (
cm sec-1 ,
<1.0 x lO-7
2.9 x 10-8
to 3.5 x 10-8
2.7 x 10-8
<1.0 x 10-8
1.3 x 10-8
<1.0 x 10-7
~2.6 x 10-9
2.0 x 10-8
2.0 x 10-;
1.0 x.10-7..
3.8 x 10-7
to 4.9 x 10-8
1.0 x 10-7
4.9 x 10-8
aFrom Ely et al., 1983.
-------
material, a suitable soil from a nearby borrow source can be utilized. Most
engineers interviewed reported an economic haul distance of 8 to 10 miles for
borrow clay, -a-1-though haul distances of up to 25 miles were reported in some
instances. When suitable soils are not available at economic distances from
the facility, it may be necessary to blend an additive, such as bentonite,
with the native soil.to improve its performance as a liner material or to
blend together local soils to achieve the proper material properties. These
solutions are used in areas where suitable clay is scarce.
5.1.2.1 Native Soils-
Native soils include those obtained onsite and those obtained from
nearby borrow areas. Design engineers' opinions differed over suitable soil
types for a clay liner. These differences are, to some extent, due to
regional variability in soil types. However, the differences are also due to
the fact that soil selection based on the desirable characteristics for liner
soils (low permeability, high strength, self-healing capacity, chemical
compatibility, and low settlement and shrink/swell potential) involves some
compromise between these characteristics. The soils with the lowest
permeability and the highest flexibility and self-healing capacity (CH or fat
clays—see Section 3.4 of this document) have the lowest strength and highest
shrink/swell potential and, are more affected by chemicals than other soils.
Gravelly or sandy clays have high strength and a relatively low potential for
settlement. However, they also tend to be brittle and crack when stressed,
and they may be more likely to contain sandy or gravelly inclusions that
could locally raise permeability.
Opinions differ on the suitability of fat(CH) clays for liners. Some
consider that their tendency to wick moisture; to dry, shrink, and crack; and
to expand upon moistening, combined with a higher likelihood of permeability
changes when exposed to certain chemicals, makes them inherently unsuitable
as liner materials. Others consider that their low permeability and high
self-healing capacity makes them the preferred liner material as long as
provisions are made to prevent moisture change in the liner during construc-
tion and operation and after closure. Interestingly, the location of the
persons interviewed seemed to affect their opinion on the use of fat clays in
liners. Engineers in regions where these clays were common were more
positive about their use than engineers in regions where these clays are not
common and other clay types are more readily availably. However, all
engineers agreed that the major factors influencing soil selection were its
cost, and its ability to be compacted to the required permeability.
If sufficient quantities of soil suitable for use as a liner material
are not available at the facility site, a borrow source must be identified.
When a borrow source is selected, routine testing procedures are used to
screen various potential sites. The time and expense involved in permeabil-
ity testing discourage its use for this routine screening. Design firms
often estimate clay suitability (i.e., permeability and strength) based on
quicker, more easily performed tests, including Atterberg limits, gradation
(particle size distribution), and compaction tests (standard Proctor).
Once a potential borrow source has been identified, the site should be
investigated, with the methods described in Section 5.1.1, to determine the
amount of suitable soil present at the site and the degree of spatial
variability of soil properties in the soil deposits and to confirm that the
soil is sufficiently impermeable to serve as a liner material. Borrow source
5-13
-------
Investigation results then can be used to plan an efficient extraction
procedure for-the liner material.
Prior to facility design, representative samples of the liner material
are subjected to laboratory compaction and permeability tests to establish
the relationship among moisture content, density, compactive effort, and
permeability. This information is needed for preparation of the facility
design and the construction specifications. Although both standard and
modified Proctor compaction tests are used, standard Proctor is preferred by
most engineers because it results in a wetter optimum moisture content for a
given soil. Fixed-wall, flexible-wall, and consolidation cell permeameters
are used to measure laboratory permeabilities and to establish the moisture
content and compactive effort at which minimum permeabilities may be achieved
(for further discussion of these techniques, see Chapter 3).
Strength of the liner and foundation soils also must be measured for
input to the design. The triaxial compression strength test (ASTM D2850-82)
is generally preferred for measuring soil strength. However, the direct
shear test (ASTM D3080-72, D2573-72) or vane shear test may also be used.
The shear test chosen should mimic the type of failure most likely to occur
1n the soil liner.
The results of compaction, permeability, and strength tests are used to
establish an envelope of acceptable values for parameters that will be used
during design to establish material and compaction specifications. Suffi-
cient testing should be conducted to determine the range of variability in
these soil properties and to determine the suitability of all soil types that
may be encountered at the soil source. Correlating these test results with
soil Index properties, appearance, and feel enables ranges of acceptable
values of these properties to be established and specified for routine
screening of liner material as it leaves the borrow source (see
Section 5.3.4).
Most design engineers interviewed also recommended that compatibility
studies of liner soils with wastes or waste leachates be conducted as a part
of the material selection, especially when the soils are comprised of fat
clays. However, no cases were uncovered (by the authors) where compatibility
problems with a natural soil resulted in rejection ot that soil as a liner
material. One of the biggest problems with compatibility testing is the
selection of a representative waste or leachate composition. Another is
deciding how much of a change in soil properties (e.g., permeability) on
exposure to the waste or waste leachate constitutes incompatibility. More
discussion on the techniques and the problems associated with compatibility
testing may be found in Chapter 4 of this document.
5.1.2.2 Admixed Soils—
When the in situ native soil is not suitable as a clay liner and nearby
borrow sources of suitable soils do not exist, it may be economical to use
bentonite or other clay materials as a soil additive to decrease the
permeability of the native soil. Bentonite is a clay material composed of
mostly sodium-montmorillonite (with minor amounts of calcium-montmorlllonite
and other clay minerals) and 1s, as a result, highly expansive with the addi-
tion of water. Bentonite1s expansive nature enables relatively small amounts
(5 to 10 percent) to be added to a noncohesive soil and makes it cohesive and
behave similarly to a soil containing 50 percent nonbentonite clay (Kozlcki
5-14
-------
and Heenan, 1983). Blending two native soils can also be used to produce a
soil with the-desired material characteristics.
Important parameters to consider in the design of bentonite liners
include aggregate particle size of the dry bentonite, mineralogy (as
reflected by swelling potential), and the required bentonite application
rate. Both granular and milled (powdered) bentonite are currently used as
admixed liner material. However, a Canadian firm with extensive experience
in designing bentonite liners recommends that only powdered bentonite be used
in liners (Kozicki, Ground Engineering, Ltd., Regina, Saskatchewan, personal
communication, 1984) because powdered bentonite mixes uniformly throughout
the soil mass, making intimate contact with the soil grains. Granular
bentonite cannot be as intimately mixed with the soil. A failure of one
bentonite/sand admixed liner has been attributed to the use of granular
bentonite (Diamond, 1979).
The type of exchangeable cations present in a bentonite influences its
ability to lower the permeability of a native soil and is an important
property to control. Sodium is the predominant cation in high-swelling
bentonite; its high-swelling capacity minimizes the amount that must be used
to lower the permeability of a soil. Other bentonites can contain
significant amounts of exchangeable calcium, resulting in a clay that has a
lower swelling capacity. Thus, it is important to select and specify a
bentonite that has the degree of swelling desired. This can be done easily
with a simple test (Rollins, 1969).
The amount of bentonite to admix with the native soil to achieve a
specified compacted permeability varies according to soil conditions. In
general, 3 to 8 percent bentonite will lower the permeability of most
granular material to between 1 x ICT7 cm/s to 1 x ID'9 cm/s (Kozicki and
Heenan, 1983). However, it is necessary to determine the optimum application
rate and moisture content for the specific soil/bentonite admixture. Usually
this is accomplished by conducting a series of compacted permeability tests
on admixtures with different percentages of bentonite. Alternatively, one
design engineer determines the proper percentage by the amount necessary to
bring the liquid limit of the soil to 45 or to produce a CL classified soil
(Pacey, Emcon Associates, Inc., San Jose, California, personal communication,
1984). Once the proper percentage of bentonite is determined, density,
moisture content, compactlve effort, and permeability relationships need to
be established for facility design purposes. Proper soil percentages for
blended soil liners can also be determined through the above procedures.
All personnel experienced with bentonite additives agree that
compatibility tests are especially critical for admixed liners. The high-
swelling clay minerals are generally more affected by chemicals than other
clay minerals. As an example, sodium-montmorillonlte is easily changed to
calclum-montmorillonite when it undergoes ion exchange with solutions high 1n
calcium salts. This change will seriously reduce the swelling potential of
bentonite and thereby increase the permeability of the admixture.
One approach to dealing with the compatibility problem is to pretreat
the admixed bentonite with the waste liquid it 1s to contact. This method is
currently being used for bentonite admixed linings for brine ponds. When
brine is used to wet the bentonite admixtures, all chemical effects take
place during installation, precluding any change once the brine pond 1s
5-15
-------
filled (Kozickl, Ground Engineering, Ltd;, Regina, Saskatchewan, personal
communication", 1984). Although it may not be reasonable to pretreat benton-
1te with hazardous wastes or waste leachate, it may be possible to pretreat
admixtures with a nonhazardous material that will have a similar effect on
the bentonite.
5.1.3 Facility Design
Once the site and liner material have been selected and characterized,
the design of the hazardous waste containment facility can begin. Facility
design is accomplished through standard geotechnical practices but must be
tailored to the individual site geology and facility operational require-
ments. The following text summarizes some important points about the design
of clay liners gathered through our interviews and review of the literature.
Other facility components (e.g., leachate collection systems and caps) neces-
sary for proper facility performance are not covered by this document. More
detailed discussions of earthwork design engineering may be found in refer-
ences such as Winterkorn and Fang (1975), U.S. Department of the Navy (1982),
and U.S. Department of the Interior (1974).
5.1.3.1 Configuration—
The configuration of the clay liner is determined by the configuration
of the containment facility, which 1s determined by topography, geology,
hydrogeology, land ownership, existing structures, and waste volume require-
ments. Generally, facilities are rectangular, but topographic constraints or
land availability can result in Irregular shapes. Facility size is usually
determined by projected waste volume requirements and planned modes of facil-
ity operation. However, facility size also can be limited by technical
considerations, such as seismic design criteria.
Facilities may be excavated below ground, built above ground and
contained by dikes, or built partially above and below ground. Generally the
designs of clay liners for these types of facilities are similar, except that
dike design is required for aboveground facilities and groundwater control
measures are usually required for facilities sited below the water table
(intergradient design).
5.1.3.2 Foundation Design— «
Foundations for clay liners are designed to control settlement and
seepage and to provide structural support for the liner. The natural founda-
tion should provide satisfactory contact with the overlying liner, minimize
differential settlements and thereby prevent cracking of the liner, and
provide an additional barrier to leachate migration from the facility.
5.1.3.2.1 Settlement—Sett!ement is usually not a problem for clay
liner foundations. Most clay liners are sufficiently thick to withstand some
differential settlement of the foundation soils. As long as the topography
is fairly uniform and significant soil heterogeneities are not present, dif-
ferential settlement should be minimal. However, several design engineers
recommend excavating and recompacting the upper 1 to 2 feet of foundation
soil to control local settlement and seepage prior to liner installation.
Several engineers also recommend that foundation settlement analysis based on
the site's soil properties (determined during site investigation) be
conducted during the design of the facility. These analyses should take into
account the weight of all facility components on the foundations, especially
5-16
-------
footings for pile-type structures such as leachate collection risers, which,
if Improperly-designed, can be forced into or through the Uner. Compensated
foundation, wfttCTi implies that the weight of soil extracted from the site
balances the weight of fill material, also can be used to minimize subgrade
settlement (Vesilind et al., 1983). Techniques for conducting settlement
analyses are given in any standard soil mechanics text.
Haxo (1980) notes that differential settlement is a localized structural
stress phenomenon; therefore, the greater the liner's thickness and elastic-
ity, the greater the tolerance range for differential settlement. A
sufficiently thick liner can engage in self-healing if the subgrade settles
nonuniformly.
The ability to predict the extent of settlement depends upon the type of
process anticipated to cause settlement. Primary consolidation, which is a
reduction in void ratio due to removal of pore fluids by mechanical loading,
generally occurs according to the consolidation theory developed for soil.
Basically, the theory states that the rate and amount of compression is equal
to the rate and amount of pore fluids squeezed out of the soil (Anderson,
1982).
Secondary consolidation (densification) depends upon the applied load
and the chemical and physical nature of the solid particles and the waste.
Therefore, secondary consolidation is more irregular and less predictable
than primary consolidation and may be significant in settlement of plastic
clay soils, heterogeneous fill materials, organic materials, and other
compressible materials.
Tertiary consolidation (densification) occurs when the volume of solids
is reduced. The effect of tertiary consolidation on mineral soils is
minimal; however, it may be a major concern with organic soils, organic
waste, soluble materials, and materials subject to chemical attack. Tertiary
consolidation is highly irregular and is influenced by a number of environ-
mental factors that make 1t difficult, if not impossible, to predict
(Anderson, 1982).
5.1.3.2.2 Seepage—Seepage both into and out of the facility must be
controlled during construction and site operation. Although the clay and
flexible membrane liners are designed to accomplish this goal, for their
optimum performance and because it may function as a backup liner, the
foundation also should be designed to control seepage. For 1ntergrad1ent
facilities, seepage can reactivate slickensides in the foundation soil. If
these features are present near the toe of sidewall slopes, slope failures
can result (Boutwell, Soil Testing Engineers, Inc., Baton Rouge, Louisiana,
personal communication, 1984). Opening of slickensides or joints can occur
from stress removal by excavation and when soils heave in response to
unbalanced water pressures in underlying permeable strata (Boutwell and
Donald, 1982).
Heterogeneities such as large cracks, sand lenses, or sand seams In the
foundation offer pathways for leachate migration and could cause piping
failures. Soft spots 1n the foundation can cause differential settlement,
possibly causing cracks in the liner and damage to the leachate collection
and leak detection system. Cracks and sand lenses or seams also can cause
5-17
-------
problems during liner construction if the facility penetrates the water
table. An illustration of this problem was provided by a design engineer
during one of"the interviews. In the case he cited, a 2-foot clay liner was
installed without"adequate site investigation or foundation preparation. A
small fracture in the foundation base was connected to an artesian (geopres-
sured) aquifer 30 feet below the landfill bottom. The overburden that was
removed during construction originally provided confining pressure on the
fracture. After the liner was installed, hydrostatic pressure in the crack
exceeded the confining pressure provided by the liner, causing a rupture of
the liner. The expense of excavating the ruptured liner and crack, grouting
the crack, and pumping the aquifer to reduce the hydraulic head could have
been avoided or minimized if adequate site Investigation and foundation
preparation had been practiced prior to liner installation.
Solutions to these problems include the various dewatering systems
(e.g., pumping wells, slurry walls, trenching, and pumping) to lower the
hydraulic gradient on the facility.(Boutwell et al., 1980). Further discus-
sion on dewatering methods may be found in Cashman and Haws (1970)„ and
Section 5.1.3.4.7 describes intergradlent facility design in more detail.
Other methods to control foundation seepage Include properly keying dikes
Into the foundation subsoil, selecting impermeable materials for dike cores,
grouting cracks and fissures in the foundation soil with bentonite or other
grouting material, and designing compacted clay cutoff seals to be emplaced
1n areas of the foundation where lenses or seams of permeable soil occur.
Figure 5-1 Illustrates typical clay cutoff seal designs.
5.1.3.2.3 Dike Design—Containment facilities built above ground or
partially above ground require dikes to support the aboveground portion.
Hazardous waste containment facilities may be designed to be above ground for
several reasons. Regulations may require a certain distance between the
facility bottom and the nearest underlying aquifer or, in some cases, the
water table. Aboveground design may be necessary to achieve this separa-
tion. In regions with high water tables, limits placed on excavation depth
in order to prevent bottom heaving or rupture may necessitate a partial
aboveground design (see Section 5.3.4.1.7). In addition, low-cost, suitable
dike construction material may be readily available at or near the facility
site, making aboveground construction more economical than excavated
facilities.
*
A containment facility dike serves as a retaining wall to resist the
lateral forces Imposed by the stored wastes. Design for retaining character-
istics requires slope stability analysis, which is normally accomplished by
using the Bishop method of slices (Boutwell et al., 1980). As the waste
produces much of the outward force on the dike, the geotechnical properties
of the wastes must be defined. In addition, time-related changes in the
properties of both the waste and the dike material resulting from consolida-
tion, settlement, changes in saturation, or chemical interactions must also
be evaluated (Boutwell et al., 1980). Figure 5-2 illustrates typical dike
configurations.
Erosion resistance and control of desiccation must also be considered
during the design of dikes. Berms and vegetation may be used to control
erosion. In arid regions special designs incorporating gravel-filled troughs
1n the dike crest have been used to provide a method to prevent desiccation
5-18
-------
• Interior Side Slope
Cut required to overexcavate
permeable seam or zone and
replace with compacted seal.
Permeable
Seam or Zone
Clay compacted in lifts not exceeding
9" in loose thickness to a minimum of 95%
of the standard proctor density (ASTMI D-698)
- Interior Side Slope
Cut required to overexcavate
permeable seam or zone and
replace with compacted seal.
Permeable
Seam or Zone
Clay compacted in lifts not exceeding
9" in loose thickness to a minimum of 95%
of the standard proctor density (ASTM D-698)
Base Grade
Corner
Clay compacted in lifts not exceeding
. 9" in loose thickness to a minimum of 95%
of the standard proctor density (ASTM D-698)
Base Grade
Cut required to overexcavate
permeable seam or zone and
replace with compacted seal.
Bottom
Permeable
Seam or Zone
After Waste Management, Inc.
Figure 5-1. Compacted clay cutoff seal.
5-19
-------
Cover Soil
Berm
Toe Drain
Homogeneous Dike
and Soil Liner
Synthetic Membrane
Liners
Leachate Collection
System
HOMOGENEOUS DIKE
Toe Drain
Synthetic Membrane
Liners
Leachate Collection
System
Low-Permeability
Soil Liner
ZONED DIKE
Figure 5-2. Dike components and typical configurations.
5-20
-------
cracking. If the trough is kept filled with water, the exposed upper portion
of the dike can be kept moist. For further information on dike design, the
reader is referred to Sherard et al. (1963), U.S. Bureau of Reclamation
(1973), Winterkorn and Fang (1975), U.S. Department of the Army (1977), and
U.S. Department of the Navy (1982).
5.1.3.2.4 Sidewall Design—Sidewall slopes of the hazardous waste
containment facilities identified during this project ranged from 4 to 1
(horizontal to vertical) to vertical. Factors that influenced the selection
of sidewall slopes included waste volume/landfill area considerations,
foundation and liner soil stability, equipment operation constraints, and
stability of other facility components.
One factor that influences the choice of sidewall slope is whether the
sidewall liner is to be compacted in horizontal lifts or in continuous lifts
parallel to the liner surface (Figure 5-3). Generally, the continuous lift
method cannot be used for side slopes steeper than around 2.5 to 1 because of
operational limitations of compaction equipment. However, this limitation
varies according to soil characteristics and equipment type. In one case,
described during an interview, continous lifts were compacted on slopes as
steep as 2 to 1 with a sheepsfoot roller. However, the facility was small
the slopes were short, and the roller had to be both pushed and towed to
negotiate the slope. In soils common to Louisiana, a bulldozer can operate
on slopes as steep as 2 to 1 but tends to tear the liner material. A slope
of 2.5 to 1 is recommended for bulldozer operation, and a slope of 2.8 to 1
is recommended when sheepsfoot rollers are used (Boutwell and Donald, 1982).
Some design engineers interviewed preferred horizontally compacted side
slopes to those with continuous lifts because the former are more stable and
they allow steeper slopes and hence greater facility waste capacity. How-
ever, most engineers interviewed believe that, for horizontal sidewall lifts,
the orientation of lift boundaries and the compacted clay fabric perpendic-
ular to the liner surface increase the likelihood of seepage through the
liner, limiting the desirability of horizontally compacted sidewall slopes.
Thus, most engineers consider 1t especially critical to ensure that lifts are
adequately tied together in horizontally compacted side slopes. However, no
case studies were found during this investigation, which demonstrated
increased seepage due to horizontal sidewall lift compaction.
Other sidewall slope considerations collected during our Interviews and
literature review include:
• Flexible membrane liner (FML) manufacturers generally recommend
that, when an FML is part of the liner system, sldewalls should not
be steeper than 3 to 1. Tracked vehicles placing earth materials
on FMLs tend to stall, spin their tracks through the lodse earth,
and damage the FML on steeper slopes (Morrison et al., 1982).
However, FML-lined facilities with side slopes of 1 to 1 were
encountered during our survey. Opinions differed among design
engineers on this subject.
• For admixed bentonite liners, slopes of 3 (H) to 1 (V) are
generally preferred. Mechanical spreading methods can be used on
2.5-to-l slopes, but this is the marginal case. Hand placement can
5-21
-------
Overbuild and
Cut to Slope
• ^
Horizontal Lifts
Continuous Lifts
Figure 5-3. Methods of liner sidewall compaction.
5-22
-------
be used on slopes of 2 to 1, with mixing and compaction performed
by lowing equipment up and down the slope (Kozicki and Heenan,
—
• The angle of repose of sand corresponds to approximately a 2-to-l
slope. Thus, if a granular leachate collection system is designed
to extend up the slopes, sidewall slopes cannot exceed 2 to 1.
• Although steep sidewalls with horizontal lifts maximize volume and
provide greater stability, construction costs can be greater than
with continuous lifts because of logistic and scheduling require-
ments and the extra liner material that must be used and then
trimmed away.
• With thinner linings (2 to 3 feet), extra width to accorrcnodate
equipment may result in the sidewall lining being thicker than the
bottom lining for horizontally compacted sidewalls.
• One engineering firm preferred horizontally compacted sidewalls for
thick linings (>5 feet) and continous sidewall liner lifts for thin
linings (<2 feet).
• One engineer recommended that 4-inch lifts (versus 6-1 nch bottom
lifts) be used in horizontally compacted liner sidewalls.
t Slopes of 3 to 1 or less tend to collect water if not properly
smoothed.
• Maximum sidewall slopes of 2 to 1 are advisable for liners composed
of highly plastic soils because of the loss of stability of these
soils when they are saturated (Day, 1970). Highly plastic soils
contain high amounts of smectite clay minerals. The high disper-
sivity of these minerals results in a loss of strength upon
wetting.
Slope stability analysis must be conducted for both foundation and liner
soils to ensure that shearing stresses developing within sidewall slopes
following excavation or dike construction will not exceed the available shear
strength of the soil and cause a failure of the slope. The shear strength of
the soil and its variability in the soil mass, degree of saturation of the
soil, pore water pressure (if effective stress analyses are to be performed),
slope height and inclination, heterogeneities in the soil mass, and expected
stresses on the slope are all important inputs to slope stability analyses.
Applicable strength tests for the waste and soil include a measure of
relative shear strength by a cone penetrometer, vane shear tests, remolding
index, drained direct shear, triaxial compression, and consolidation tests.
Strength tests should be selected to mimic the expected mode of failure in
the soil (Haxo, 1983).
Several methods of slope stability analysis are currently used. Classic
limit equilibrium methods are generally used for earthwork design. These
methods take a free body from the slope and, using estimates of the forces
acting on the body, calculate the equilibrium shear resistance of the soil
and compare it to the shear strength of the soil to indicate the factor
5-23
-------
of safety (Fang, 1975). Regardless of the specific procedure used for carry-
ing out the computations, the following principles are common to all limit
equilibrium methods: a potential failure surface is postulated; the shearing
resistance required to equilibrate the assumed failure mechanism (i.e., the
shear stresses required along the failure surface to balance the driving
forces of the unsupported slope) is calculated by means of statics; the
calculated shearing resistance required to maintain equilibrium is compared
to the available shear strength of the soil along the assumed failure plane
(the ratio of the existing strength to induced shear stress is one definition
of a factor of safety); and the mechanism with the lowest factor of safety is
found by iteration. ;
Choosing the method depends to some extent on the properties and
hydrologic conditions of the soils to be evaluated. For homogeneous soils
with equal pore pressure distributions, methods that consider the whole free
body are appropriate, such as the friction circle method (Taylor, 1948). If
soil properties vary through the soil mass, the method of slices, which
divides the free body into several vertical slices, is appropriate (Bishop,
1955; Fellenius, 1927). If straight line failure planes can be identified in
the soil mass, the wedge method can be used (Lambe and Whitman, 1969). The
wedge method has been recommended for evaluating the stability of clay liners
placed on side slopes (Boutwell and Donald, 1982). For facilities below the
water table it is necessary to consider pore water pressure, and the effec-
tive stress (total stress minus pore water pressure) should be used instead
of total stress (Fang, 1975). Methods of stability analysis that consider
these factors Include the Bishop and Morgenstern method (Bishop and
Morgenstern, 1960) and the Spencer method (Spencer, 1967). Factors that must
be considered when slope stability is calculated in special problem soils are
presented in Table 5-5.
Essentially two types of slope stability problems occur in clay:
short-term stability (end-of-construction case) and long-term stability
(steady-seepage case). The short-term case applies just after the excavation
is completed and assumes that time has been insufficient for any water to
move in or out of any representative soil element within the soil profile.
When a soil profile is believed to be relatively homogeneous and without
discontinuities, this is a reasonable assumption and has been used with much
success. Should the clay have substantial joints and fissures, drainage may
occur so quickly along these discontinuities that the* problem may not be
adequately represented by the assumption (Esu, 1966). In general, only
short-term stability analysis is necessary for interior containment facility
slopes 1f the facility is to be filled shortly after construction; long-term
slope stability should be considered for the outer slopes of aboveground
facilities (Boutwell and Donald, 1982).
A factor of safety is customarily calculated during slope stability
analysis. These factors are based on the values of various parameters that
can affect the stability of a slope. These include available shear strength
versus required shear strength, required soil cohesion versus available soil
cohesion, actual friction angle versus stable friction angle, actual height
versus stable height, and resisting moments versus moments tending to cause
failure (Fang, 1975).
5-24
-------
TABLE 5-5. FACTORS CONTROLLING STABILITY OF
SLOPED CUT,INCOME PROBLEM SOILS
Stiff-fissured Clays
and Shales
Field shear resistance may be less than suggested by
laboratory tests. Slope failures may occur progres-
sively and shear strengths reduced to residual values
compatible with relatively large deformations. Some
case histories suggest that the long-term performance
is controlled by the residual friction angle which
for some shales may be as low as 12°. The most
reliable design procedure would involve the use of
local experience and recorded observations.
Loess and Other
Collapsible Soils
Strong potential for collapse and erosion of rela-
tively dry material upon wetting. Slopes in loess
are frequently more stable when cut vertical to
prevent infiltration. Benches at intervals can be
used to reduce effective slope angles. Evaluate
potential for collapse as described in DM 7.1,
Chapter 1. (See DM-7.3, Chapter 3 for further
guidance.)
Residual Soils
Sensitive Clays
Talus
Loose Sands
Significant local variations in properties can be
expected depending on the weathering profile from
parent rock. Guidance based on recorded observation
provides prudent basis for design.
Considerable loss of strength upon remolding generatfjd
by natural or man-made disturbance. Use analyses
based on unconsolidated undrained tests or field vane
tests.
Talus is characterized by loose aggregation of rock
that accumulates at the foot of rock cliffs. Stable
slopes are commonly between 1-1/4 to 1-3/4 horizontal!
to 1 vertical. Instability 15. associated with
abundance of water, mostly when snow is melting.
May settle under blasting vibration, or liquify,
settle, and lose strength if saturated. Also prone to
er&sion and piping.
Source: U.S. Department of the Navy, 1982).
5-25
-------
A factor of safety of 1.0 means that the slope is constructed at the
limit of equilibrium; the mobilized stresses equal the available strength.
In practice, "larger safety factors are used for designing slopes for critical
facilities. The acceptable value of factor of safety depends very much on
whether the slope will be temporary or permanent, whether the analysis is for
short- or long-term conditions, whether conservative assumptions have been
made about soil properties, and other factors. Temporary slopes are often
designed for factors of safety of 1.2 to 1.5. Permanent slopes are often
designed with different factors of safety for undrained conditions compared
with drained conditions. For public earth works projects (e.g., dams) in
California, a safety factor of 2.0 is used. One design engineer considered
a safety factor of 1.7 to be the minimum acceptable for use on outer slopes
of hazardous waste facilities (Reynolds, California Department of Health
Services, Sacramento, California, personal communication, 1984).
Static slope stability analysis is appropriate in areas with little or
no seismic risk. Dynamic slope stability must be considered 1n areas where
significant ground motion can occur (Fang, 1975; also see Section 5.1.3.4.6).
In summary, the following guidance should be applied to the design and
construction of slopes (Lutton et al., 1979):
• Examine, sample, and test to ensure that the foundation is not weak
and likely to participate 1n displacement.
• Conduct detailed engineering stability analyses for any site where
the consequences of slope failure are serious. Estimate changes in
hydrology and seismic stability, identify average and worst case
patterns, and calculate factors of safety.
• When selecting soils, consider soil shear strength, allowing for
compaction and the corresponding strengthening effect.
t Specify slope inclination; decreasing the design inclination
effectively increases the stability of soil slopes.
• Use underdralns, toe drains, cutoffs, and leachate collection and
disposal systems for seepage control. *
• Allow for freeze/thaw and dry/soak conditions in the selection of a
sufficiently thick side slope soil.
• Compact soil as specified, using field tests for quality control or
quality assurance. Prescribe lift thickness.
• Consider other miscellaneous factors including toe protection
(e.g., from flooding conditions) the use of berms or systems of
berms rather than having a single unbroken inclination, and the use
of reinforcement to strengthen embankments.
5-26
-------
5.1.3.2.5 Bottom Design—The bottom of the containment facility must
be shaped to facilitate leachate collection,in landfills and drainage in
surface impoundments and to prevent puddling and ponding on the liner during
construction. Most design engineers use a 1- to 2-percent slope for the
facility base, with 2 percent being preferred.
Kmet et al. (1981) used an analytical model of landfill leakage
developed by Wong (1977) to evaluate several landfill design parameters,
including clay liner bottom slope. In general, increasing the bottom slope
decreased the liner leakage by providing more rapid movement of leachate to
the collection sump. Leakage rapidly increased when bottom slopes decreased
to below 2 percent. Bottom slope increase to 5 percent provided additional
reduction in leakage, with little additional benefit for steeper slopes.
The configuration of the bottom slope depends on the facility configura-
tion and the leachate collection system design. Most designs incorporate a
sand/gravel drainage layer with a system of pipes, with the liner bottom
sloping to the pipes and the pipes and underlying liner sloping to a collec-
tion sump. Based on the analytical model described previously (Wong, 1977)
Gordon et al. (1984) recommend a maximum leachate flow distance to the pipe
network of 150 feet, and 50 feet has been stated as the reasonable minimum
due to construction practicalities (Kmet et al., 1981). Small facilities
may have a pipeless collection system, and one design engineer interviewed
preferred a properly designed granular drainage layer without pipes to a
piped system because it precludes the necessity of cleaning out pipes.
5.1.3.3 Liner Design--
Clay liners are constructed of compacted clay soils installed in a
series of lifts of specified thickness. The liner must be sufficiently thick
and impermeable to retard leachate flow and to provide structural support to
overlying facility components. For clay liners, a permeability of
1 x 10-' cm/s is required by Federal regulations*. One design engineer
recommended that a permeability of 1 x lO"8 cm/s be specified to provide
a factor of safety.
Liner thicknesses of 1 to 12 feet of compacted clay were encountered
during the course of this study, although most design engineers recommended
•£? i feet of Clay< State re9ulations usually detejmine liner thickness,
with 2 feet (as recommended by EPA guidance) as the minimum. Transit time
prediction methods also have been used to specify liner thickness.
In general, the liner is designed to be uniformly thick over the entire
facility except for thicker areas that will be excavated to accommodate the
leachate collection sump and any leachate collection pipes recessed into the
landfill bottom. This is necessary so that, following excavation for the
sumps and pipes, the liner will have the uniform, specified thickness over
the entire landfill bottom as illustrated in Figure 5-4. In addition to
this, engineers working for the Department of Natural Resources, State of
Wisconsin, recommend an extra foot of liner under leachate collection lines
(Gordon et al., 1984). For further information on leachate collection
systems, see Bass, 1986, and U.S. EPA, 1985, 1986 (pages 39-46), and 1987.
*A11 permeabilities in this Section are laboratory measurements unless
otherwise specified.
5-27
-------
01
I
00
Drainage
Layer
Recess for Pipe
Leachate Collection Pipe
Figure 5-4. Liner design for collection system pipes and sump.
-------
Extra liner thickness and compactive effort have been recommended for the
toes of sidewall slopes to combat seepage and to ensure that the bottom and
sidewall linersjare adequately tied together.
Lift thickness is selected by the design engineers and the contractors
depending upon soil characteristics, the compaction equipment to be used, and
the required compactive effort. Lift thickness is usually specified in the
construction specifications. The liner lifts must be thin enough so that
adequate compactive effort reaches the lower portion of the lift. Emplace-
ment of thin lifts ensures that the entire lift 1s adequately compacted and
because more additional compactive effort is transferred to lifts below the
lift being installed. However, these advantages must be balanced with the
added construction expenses associated with thinner lifts since more lifts
must be compacted to achieve the specified liner thickness. Interviews with
design engineers indicate a general preference for loose lift thickness of 6
to 9 Inches. However, loose lift thicknesses of up to 15 inches have been
encountered during our survey, and. a 2-foot lift was used over a sand
drainage layer at one facility (see Section 5.1.3.3). Some engineers think
that adequate compaction can be achieved with thick lifts if heavy enough
compaction equipment is used. Four-inch lifts were recommended by one
engineer for horizontally compacted sidewalls.
Clay liners may be designed to be installed over the entire facility
(small facilities) or in segments (large facilities and continuous-operation
facilities). If the liner is installed In segments, a beveled or step-cut
joint between segments (Figure 5-5) should be specified to ensure that they
are properly tied together. A bevel with an 8 to 1 slope was specified for a
large facility where the liner was installed in wide transverse strips
(Boutwell, Soil Testing Engineers, Baton Rouge, Louisiana, personal
communication, 1984).
Most admixed bentonite liners are only 4 to 6 inches thick. Bentonite
company literature Indicates that liner thickness can be as small as 4 to
6 inches because of the low permeability (l(r9) that can be achieved with
bentonite (IMC, 1982). However, Federal guidance recommends a 2-foot clay
liner, and many States require liners up to 10 feet thick. Regulatory
personnel in some of these States did not think that thin bentonite linings
satisfy the regulations. Besides failing to meet the thickness requirements,
thin liners may not provide adequate structural stability for the overlying
facility components such as leachate collection risers and may be hard to
construct to uniform thickness. In general, clay liner thickness and
permeability requirements must be independently satisfied; the use of a clay
with a lower permeability than required does not justify a thinner liner
(Boutwell and Donald, 1982).
Two facilities with bentonite liners are included in the case studies
section of Chapter 7. At Site 0, local soils were augmented with 3 percent
bentomte to achieve the required permeability. The permeability
achieved was 8.3 x 10-°; however, a 4-foot liner was still used at the
site. In addition, 3 percent lime was added to stabilize the bentonite by
reducing its swelling capacity (by replacing sodium with calcium in the
bentomte). The same effect and permeability could have been achieved
without lime addition by using lower cost bentonite with a h1gh-calc1um-
montmorillonite content.
5-29
-------
Bevel Cut
Step Cut
Figure 5-5. Methods of keying-in liner segments.
5-30
-------
Site P presents an approach to achieving both low permeability and high
strength white minimizing the addition of bentonite. In this
double-clay-Uned. site, the upper liner consists of a 1-foot compacted
layer of admixed polymer bentonite and native soil (5 x 10~8 cm/s)
sandwiched between two layers (3 ft above and 1 ft below) of recompacted
native soil of a higher permeability (10-° cm/s). The lower liner is
composed of 6 inches of compacted admixed liner underlain by 5 feet of
recompacted or in situ native soil. The sides are single lined with 18
inches of compacted native soil overlying 6 inches of compacted admixed
soil. The designers of the site used the recompacted native soil to provide
structural strength and protection to the admixed liner with the admixed
liner providing lower permeability. It is not certain whether this liner
design would satisfy current regulatory requirements, especially in States
requiring thick liners.
Liner system configuration differs according to the type of containment
facility. In landfills, the clay liner usually lies below a synthetic liner
that lies below the leachate collection system. In some cases, the clay
liner may be sandwiched between two synthetic liners or vice versa. Leak
detection systems may be installed between the liners or under the liner
system. A leak detection system may consist of a sand, gravel, and/or
geotextile layer underlying the entire facility (continuous coverage) or
individual lysimeters or other instruments at specific points under or
between liners (discrete coverage). Collection lines from the leak detection
system either pass through the clay liner or pass under the liner to exit at
the landfill periphery. The latter design is preferred by many engineers
because objects penetrating a liner offer potential pathways for leakage
through the liner.
The relationship of the clay liner to other system components for a
waste pile is the same as for a hazardous waste landfill, except that a
primary synthetic liner is not required and thus may not be present in a
waste pile. The leachate collection system will rest directly on a clay
liner when it is the primary liner.
The major difference between the liner system in an Impoundment and that
of a landfill or waste pile is that because impoundments by definition are
designed for holding bulk liquids, leachate collection systems are not
installed on top of the liner. Surface impoundment liner system designs
incorporating clay liners can consist of a single clay liner or may consist
of a system of redundant liners (all clay or clay and synthetic) with leak
detection or collection between or under the liners. If the wastes are to
remain in the impoundment after closure, synthetic linings are required.
Outlet pipes from leak detection/collection systems (below the primary liner)
may pass through the primary liner or may run under it to the periphery of
the facility.
A rip-rap layer on the upper side slopes of surface impoundments may be
necessary to protect against wave erosion (Section 5.1.3.4.2). A discharge
structure for loading wastes Into the pond will prevent scouring of the liner
surface (Section 5.1.3.4.2). Adequate free board (sidewalls above the liquid
5-31
-------
surface) to contain runoff from large rain storms should be Included 1n
surface Impoundment designs (U.S. Environmental Protection Agency, 1982b).
When clay liners are installed over granular leak detection layers, the.
granular layer often does not provide enough stability for compacting the
first lift of liner material. Techniques that allow equipment to 'bridge'
over sand layers without causing rearrangement and damage include:
• The use of a geotextile over the drainage layer (this also prevents
piping)
• The use of lighter compaction equipment that can 'bridge' the sand
• The specification of thicker lower lifts.
At one large facility, this problem was solved by using a loose lift thick-
ness of 2 feet for the first lift over the sand layer. The normal lift
thickness at this site was 15 inches; the use of large compaction equipment
enabled thick lifts to be specified. ;
5.1.3.4. Special Design Considerations—
This section describes some special design practices that are needed to
prevent specific causes of clay liner failure. A full discussion of each of
these failure mechanisms, along with Illustrative case studies, may be found
1n Chapter 6 of this document.
5.1.3.4.1 Control of Erosion—Erosion can be a problem during liner
construction and on dikes and freeboard areas in completed facilities. The
most accurate soil loss prediction tool that is now field-operational is the
U.S. Department of Agriculture's Universal Soil Loss Equation (USLE)
(W1schme1er and Smith, 1978). This equation has been used for agricultural
erosion control planning for more than a decade. The maximum erosion rate
should not exceed 2 tons per acre using the USLE (U.S. Environmental Protec-
tion Agency, 1982b). Important design considerations for erosion control are
listed below (Lutton et al., 1979): '
• Erosion-resistant soils (erosion resistance is quantified by the
soil erodibility factor, K, in the USLE) should be selected.
• Dispersive clays and soils in a dispersed condition should be
protected from erosion or avoided in facility areas where erosion
can occur.
• The overall landfill configuration is important. The two factors,
L (slope-length factor) and S (slope-steepness factor), in the USLE
dominate the erosion aspects of runoff.
• It is generally essential to divert natural drainage from outside
the immediate site (through external runoff diversion) to prevent
erosion 1n the facility excavation.
• Final dike slopes more than 5 feet high should be protected from
erosion by building berms and gutters along the top and sides of
the slope.
5-32
-------
• Vegetation should be started as soon as possible on exterior dike
waljsjind be well maintained throughout the life of the site.
• The use of soil covers is often specified to minimize exposure and
to protect otherwise bare soil at the construction site.
Applications should always be encouraged for intervals when
construction is interrupted.
t A program of long-term maintenance can be used to avoid erosion
problems after closure of the landfill site. This is especially
important for flood-retaining structures for facilities sited in
the 100-year flood plain.
5.1.3.4.2 Control of Scouring—Scouring is the erosion of the liner or
sidewalls of a containment facility by the force of moving water. Surface
impoundments often need to be protected from wave erosion on their upper side
slopes. Rip-rap (loose rock) is the most widely used material for this
purpose. Concrete aprons can be used, but studies by the U.S. Department of
Interior have shown that rip-rap is generally more effective and easily
placed (Small, 1981). Rip-rap must be properly sized and placed to result In
a stable protective blanket (U.S. Department of Interior, 1974). One design
engineer recommends that a wind rose for the site and the pond fetch should
be used to determine maximum wave height. The rip-rap layer can then be
designed, based on wave height and expected fluctuations of liquid levels in
the pond. Rip-rap generally is uniform in size, and dumped rip-rap performs
better than hand-placed rip-rap. More information on the design of rip-rap
layers may be found in U.S. Department of the Interior (1974) and U.S.
Environmental Protection Agency (1982b).
Clay liners in surface impoundments must also be protected from erosion
by discharge of waste into the impoundment. At least one clay liner failure
identified during this study was attributed to erosion from discharging
wastes into the pond. Discharge structures used to reduce this erosion
include concrete or rip-rap aprons, discharge tubes with upward facing
outlets, and various weirs, such as a reverse duckbill weir (Day, 1970).
Johnson and Cole (1976) reported that a bentonite liner in a papermill
lagoon was protected from scouring by covering with coarse-grained material.
First the liner was covered with 3 to 4 inches of till, and then a 6-inch
gravel cover was placed over the till in the most susceptible areas (e.g., at
the base of the aerators). Sidewalls with 3-to-l slopes were protected from
scour by 6 inches of crushed stone that was covered with rip-rap. No
problems have resulted in this facility. Johnson and Geisel (1979) reported
that a clay liner used in a municipal wastewater treatment lagoon was
protected by covering with gravel.
Voigts and Savage (1974) described a wastewater treatment lagoon lined
with natural clay. Rip-rap was added to the slopes to protect against wave
action and erosion, and a concrete scour pad was placed under the aerators to
prevent scouring the liner. Stone was added around the scour pad out to 35
feet as an additional safety factor to prevent erosion. The aerator
supports, which penetrated the clay liner, were sealed with neoprone water
stops in the middle of the compacted clay.
5-33
-------
5.1.3.4.3 Cold Climate Design—Failures from operations in cold
climates are-minimized by following design considerations outlined by Lutton
(1979). Although- these considerations were developed for covers, the follow-
ing are relevant to liners as well:
• Evaluate soil susceptibility to undesirable frost actions and
locate the earth barrier below the frost zone.
t Maintain an unfrozen soil supply.
• Consider seasonal scheduling and alternating the choice of liner
soils depending on the availability of unfrozen materials.
In addition, temporary liner covers of soil or organic mulch have been used
for protection from freezing temperatures.
5.1.3.4.4 Control of Piping—Piping is a form of internal soil erosion
that occurs below the ground surface (see1 Section 6.4). It occurs when fine
particles migrate away from a cohesive soil layer and may result from
physical causes (discordant grain size) or chemical dissolution.
Compatibility testing of the liquid waste or leachate and the liner
material (see Chapter 4) is a prerequisite to minimizing piping from chemical
dissolution. The compatibility testing can document the chemicals' effect on
permeability. During the permeability test, observations for signs of
migrating soil particles should be made and recorded. Dissolution effects
(e.g., observation of an unusual color of effluent) and obvious structural
changes 1n the soil material should also be noted. The waste and liner would
be considered compatible and piping from dissolution would be considered
unlikely 1f the test showed no permeability increases, no migrating soil
particles, and no other dissolution effects.
Four laboratory tests to determine soil susceptibility to dispersive
erosion have been developed by the U.S. Soil Conservation Service (SCS).
Dispersion tests are a specialized group of tests for characterizing fine-
grained soils suspected of having a tendency to erode rapidly. A major
conclusion of a recent symposium on soil piping was that these four tests
should be performed on soils where piping would cause, unacceptable damage
(Sherard and Decker, 1977). The four tests are the pinhole test, a test of
dissolved salts in the pore water, the SCS dispersion test, and the, crumb
test. The test methods and extensive test data are available in ASTM Special
Technical Publication No. 623.
When a clay Uner overlies a granular drainage layer, control of piping
can be accomplished by the incorporation of a filter layer below the clay
layer. A filter layer will have a grain size slightly larger than the clay
and will capture the migrating clay particles. For example, clay over gravel
represents the joining of discordant grain sizes such that the clay particles
can penetrate to the voids of the gravel. When a filter is placed under the
clay, Internal erosion is largely eliminated. In addition, the filter would
help to minimize drainage problems in the sand or gravel layer (or the
leachate collection system) that occur from clogging by the fine clay
particles.
5-34
-------
.Filter layers may be composed of geotextiles or a layer of graded
material. Resource Conservation and Recovery Act (RCRA) guidance documents
have been dev«4-0ped by EPA, and these documents contain criteria for select-
ing filter layers to stop migrating particles and to avoid plugging of
drainage layers (U.S. Environmental Protection Agency, 1982b). The criteria
were developed by the U.S. Army Corps of Engineers and are based on qrain-
size ratios for the adjacent layers.
5.1.3.4.5 Control of Desiccation—Desiccation cracks are prevented by
several testing, design, and construction procedures. Waste compatibility
with the liner should be confirmed to avoid cracking from chemical attack.
If the waste leachate creates cracks or channels in the liner material,
alternative liner materials should be evaluated. The liner material should
be tested to determine the liquid, plastic, and shrinkage limits. These
criteria may be used to evaluate a soil's cracking potential (Lutton et al.,
19/9). Tables 5-6 and 5-7 show the expected volume changes associated with
these indices. Soils with high volume changes have a greater tendency to
crack with decreased moisture.
Clay liners may be subject to developing desiccation cracks during and
immediately after installation. The clay may be protected from desiccation
after construction by installing a synthetic membrane; by installing 1 to
2 feet of soil; or for surface impoundments, by putting liquids into the
impoundment immediately after construction.
A1JUU 5.1.3.4.6 Seismic Design—Earthquakes occur across the United States.
Although they are less frequent in eastern States, geological conditions are
such that damage from ground motion is more widespread in the East than in
California when earthquakes do occur (see Section 6.8). If a facility
is located in an area likely to experience ground motion from seismic events,
it should be designed to withstand this ground motion. The most likely types
of damage from seismic events include (LARG, 1982):
• Failure of structures from ground shaking
§ Failure of facility components due to soil liquefaction,
liquefaction-induced settlement and landslidlng, and soil slope
failure in foundations and embankments
• Failure of facility components due to fault rupture
• Landsliding and collapse of surrounding structures.
Of these failures, current Federal locational standards protect only against
fault rupture by requiring a 200-foot setback from active faults. The other
three types of failure are caused by earthquake-induced ground motion.
LARG (1982) found that ground motion is much more important as a failure
mechanism than fault rupture. More sites are impacted by ground motion than
by surface faulting for a given seismic event because the only sites impacted
by faulting are those actually located on a surface fault trace. LARG's
study of six hazardous.waste facilities in California also found that
ground-motion-induced failure of tanks and poorly designed surface
5-35
-------
TABLE 5-6. RELATIVE VOLUME CHANGE OF A SOIL AS INDICATED BY
PLASTICITY INDEX AND OTHER PARAMETERS
Likelihood of volume
change with changes Plasticity index
in moisture Arid regions
Little 0 to 15
Little to moderate 15 to 30
Moderate to severe 30 or more
Humid regions Shrinkage limit
0 to 30 12 or more
30 to 50 10 to 12
50 or more 10 and less
TABLE 5-7. SOIL VOLUME CHANGE AS INDICATED BY
LIQUID LIMIT AND GRAIN SIZE
Passing No. 200
sieve (%)
>95
60 to 95
<30
Liquid limit
(%)
>60
40 to 60
<30
Probable expansion
(%)
>10
3 to 10
<1
Source: Lutton et al., 1979.
5-36
-------
impoundments were potentially the most significant contributors to hazardous
waste release_during earthquakes.
The procesTbf designing earthquake-resistant structures may be divided
into four steps. They are:
• Determining the maximum credible or maximum probable earthquake for
the site
• Determining the expected peak ground acceleration at the site from
the maximum earthquake, based on regional and site-specific
geologic factors
• Determining site-specific seismic hazards, such as potential for
soil liquefaction, slope failure, and landslides
• Designing the facility to withstand peak ground acceleration.
The current California Water Resources Control Board hazardous waste
regulations encompass these steps in addition to requiring a 200-foot setback
IT ^ly6 Ho1ocene faults. Specifically, the California Administrative
code (CAC) requires the following for new hazardous waste facilities or
expansion of existing facilities:
• A determination of the expected peak ground acceleration at the
waste management unit associated with maximum credible earthquake
(CAC Article 9, Section 2595)
t Consideration of regional and local seismic conditions and faulting
and site-specific surface and subsurface conditions in the above
determination (CAC Article 9, Section 2595)
• Use of the peak ground acceleration to determine the stability of
and safety factors for all embankments, cut slopes, and associated
fills during the design life of the unit (CAC Article 9, Section
• Design of the waste management unit to withstand the maximum
credible earthquake without damage to the structures that control
leachate, surface drainage, erosion, and gas (CAC Article 4,
Section 2547).
Determining the maximum expected seismic event and the resultant seismic
loading at the site are the key elements in seismic risk analysis. The two
approaches to this determination are deterministic and probabilistic
(Bernreuter and Chung, 1984). The first step in both methods is to delineate
regional (tectonic province) and local (fault) sources. For the deter-
ministic method, the next step is to select the governing earthquake or
maximum credible earthquake, usually the most damaging historical earthquake
associated with the site. Attenuation of the seismic energy with distance
from the source is then determined based on regional and local subsurface
conditions. The maximum (peak) ground acceleration at the site is then
determined based on surface and subsurface site characteristics.
5-37
-------
The probabilistic approach involves developing an earthquake recurrence
model for each source that could impact the site. The ground motion at the
site from different earthquakes at different distances is calculated (based
on local and site-specific subsurface geology), and this information is used
to calculate the probability that a given level of ground motion will not be
exceeded within a specified time period. One weakness of the probabilistic
approach is that it results in the same probability of occurrence of a large
earthquake for any time period. In the real world, the longer the time
period since the last earthquake, the mor.e likely it is that another earth-
quake will occur (Bernreuter and Chung, 1984). In California, the deter-
ministic approach is considered to be the conservative or worst case approach
and is required for determining peak ground acceleration at hazardous waste
facilities and for large public work projects, e.g., dams (Reynolds,
California Department of Health Services, Sacramento, California, personal
communication, 1984).
In order to adequately estimate peak ground acceleration associated with
the maximum credible or maximum probable earthquake at a site, it is very
important to conduct comprehensive assessments of the regional geology and
adequate site-specific subsurface investigations. Site-specific subsurface
geology determines the magnitude and direction of propagation of seismic
energy to such an extent that it is impossible to generalize attenuation of
seismic energy with distance for seismic events. In addition, determining a
site's geology is necessary for identifying features that are vulnerable to
seismic ground motion such as unstable soil or rock slopes prone to
landslides and unconsolidated, saturated deposits prone to liquefaction-
induced settlement and failure. Liquefaction is primarily controlled by the
character of ground motion, soil type, soil moisture content, and in situ
soil stress conditions. Slopes most vulnerable to earthquake shocks are:
t Very steep slopes of weak, fractured, and brittle rocks or
unsaturated loess that are vulnerable to transient shocks due to
the opening of tension cracks
t Loose, saturated sand that may:be liquefied by shocks with sudden
collapse of structure and flow slides
• Sensitive cohesive soils with natural moisture exceeding the liquid
limit
t Dry, cohesion!ess material on a slope at the angle of repose that
will respond to seismic shock by shallow sloughing and slight
flattening of the slope (U.S. Department of the Navy, 1983).
Estimates of ground motion are derived from peak bedrock acceleration at
a site. The methodology used to determine peak bedrock acceleration should
take into account regional and local seismicity, surface and subsurface
geology, seismotectonic features, faulting mechanisms (e.g., thrust and
strike-slip) and regional attenuation. This peak bedrock acceleration is
then modified for site-specific surface and subsurface conditions that focus
or attenuate seismic energy or that are sensitive to ground motion. These
conditions include type, thickness, and density of materials overlying the
bedrock, surface topography, and depth to groundwater (for liquefaction
assessment). The goal of this approach is to define peak ground acceleration
5-38
-------
and ground motion response spectra at a site (Reynolds, California Department
of Health Services, Sacramento, California, personal communication, 1984).
Peak ground acceleration is then used for tfie seismic design of the
facility. Methods for estimating earthquake ground motions may be found in
Hays (1980).
Dynamic side-slope stability may be calculated through standard methods
such as those described by Makdisi and Seed (1978), Sherard (1967), and Seed
(1975). Computer programs are also available for designing earthworks to
withstand seismic events (EERC, 1984). The risk of densification and
settlement in response to ground shaking can be reduced by compacting soils
to densities high enough to prevent further settlement. The potential for
liquefaction can be reduced by removing and replacing sensitive (granular)
soils or by lowering the water table (Eagling, 1983).
In general, seismic design of earthworks involves building structures
with more strength, density, mass, and thickness. Deeper foundations,
greater cross-sectional area, and better materials are specified for
seismic-resistant designs. Size and configuration are also important;
generally, smaller facilities are more resistant to damage from ground
shaking. A safety factor (Section 5.1.2.3.4) also must be incorporated into
the design (Reynolds, California Department of Health Services, Sacramento,
California, personal communication, 1984).
5.1.3.4.7 Intergradient Facility Design—The construction of sites
below the water table (intergradient facilities) presents problems due to
seepage and hydraulic forces on the compacted clay liner. Excavations below
the water table can experience hydraulically induced side-slope slippage and
heave or rupture of the foundation base. In addition, a buildup of water
pressure behind a recompacted clay liner can threaten the structural integ-
rity of the liner. Although not allowed in some States (e.g., New York)
intergradient facilities are constructed in other States (e.g., Louisiana)
where sites with low-permeability soils above the water table are rare.
When the excavation extends below the groundwater table, it is necessary
to consider the long-term or steady-seepage case for slope stability
analysis. For this case, pore pressures are assumed to be in equilibrium arid
are determined from considerations of steady seepage (generally from the
construction of a flow net or finite element analysis'); the excess pore
pressures generated due to the total stress changes in the slope during the
excavation are assumed to have dissipated. This case is analogous to the
drained-shear test, and effective stress parameters should be used. A number
of analytical techniques are available that are appropriate for intergradient
design. Summaries of the available total stress and effective stress methods
that describe the particular assumptions on which the methods are based are
presented in a number of foundation engineering texts and technical
publications (e.g., Winterkorn and Fang, 1975; Schuster and Krizek, 1978).
If the clay soil in which the containment facility is excavated is
located above an underlying permeable stratum (aquifer), there is potential
for heave or rupture of the floor of the excavation. To prevent this from
occurring, the downward pressure at the interface of the clay soil layer and
the aquifer must exceed the upward hydrostatic pressure. Where the width of
the excavation floor is narrow compared to the thickness of the clay soil
5-39
-------
layer over the aquifer, the shear strength of the soil can add substantially
to the calculated stability of the soil layer. However, for large floor
dimensions with—respect to clay layer thickness or for long-term considera-
tions, the overburden weight should exceed the upward hydrostatic force.
A related but distinctly different problem involves the stability of the
compacted soil liner. Assuming it is properly designed and installed, the
liner may be expected to have a nearly uniform low permeability. Water pres-
sure can build up behind the clay liner sidewalls or under the clay liner on
the excavation floor, cause it to separate from the parent soil, and cause it
to heave or rupture. Whether a liner heaves or ruptures depends on the
plasticity of the soil, with more plastic soils tending to heave and stiffer
soils tending to rupture. The analysis of this problem involves, again, a
comparison between the "uplift" water pressure and total overburden stress
due to the weight of the liner (soil density and liner thickness) and any
waste material in place.
Control of these problems may be achieved by at least three methods:
(1) decrease the depth of excavation into the clay layer, (2) decrease the
hydrostatic pressure acting on the bottom of the clay layer, or (3) increase
the overburden pressure. Assuming the design depth is an economically based
decision, the best permanent solution is the added weight of the waste
material. However, in the short term the use of some type of dewatering
system or soil grouting may be necessary if the desired depth is too deep and
results in excessive uplift pressures. If the imbalance of forces is small
and it is estimated that the time period between excavation of the overburden
and replacement by the waste material is short, a method for evaluating the
rate of heave proposed by Boutwell and Donald (1982) can be applied. They
note that the uplift of the bottom of the excavation must be small with
respect to the effective diameter of the excavation and that the worst case
occurs when the underlying permeable stratum is very thick. The authors
indicate that in some cases the analysis may show additional dewatering
measures to be unnecessary.
Dewatering methods that can be used to reduce hydraulic head on the
liner include dewatering wells, slurry wall cutoffs, and sump pumping.
Further Information on the design and applications of these dewatering
processes may be found in Cashman and Haws (1970).
5.1.4 Construction Specifications and CQA Plan
These documents establish the lines of communication among the contrac-
tor, design engineer, owner-operator, and regulatory personnel and are
critical to the proper construction of a clay liner. Both specifications and
the CQA plan must be proposed prior to the construction. These documents
usually must be approved by the permitting agency prior to construction and
must be complete and detailed enough to assure them that the liner will be
properly constructed.
5-40
-------
5.1.4.1 Construction Specifications— ,
The two iypes of construction specifications are:
?ji.-' '.~£ ',*
• Performance, which specifies a required level of performance for
the completed facility components (e.g., the liner will have a
permeability of 1 x 10~8 cm/s).
• Method, which specifies methodology and equipment to be used in
constructing the liner.
In general, engineers designing clay liners for hazardous waste facilities
prefer a combination of method and performance specifications. The
performance must be specified to ensure that the liner performs as required.
Most designers consider some method specifications necessary, especially if
the contractor is not experienced in constructing clay liners for hazardous
waste facilities and may not be aware of the importance of operational
methodology in constructing clay liners so that they will not fail. In
addition, moisture content, density, and compactive effort must be controlled
in the field if the specified permeability is to be achieved; specification
of these parameters helps ensure that it is achieved. Combination method and
performance specifications must be very carefully drawn to solve the problem
of the specified method that does not yield the specified performance.
Equipment (e.g., sheepsfoot rollers for compaction) and lift thickness
are usually specified, and maximum clod size and scarification between lifts
is sometimes specified. Prior to liner construction most design engineers
specify the construction of a test fill with the same materials, equipment,
and methodology that is to be used for the liner construction. Test fills
can give the constructor valuable experience with the equipment, methodology,
and soil to be used during construction and can help convince the constructor
of the necessity of using certain equipment or methodology. Test fills are
also necessary to ensure that the permeability measured in the laboratory can
be achieved in the field with the equipment to be used in constructing the
liner. Test fills are further discussed in Section 5.3.4.1.
Design tolerances are usually present in the specifications and/or the
QA plan. Although statistical methods are available for determining design
tolerances, most engineers use a 'rule of thumb' method based on their
experience. A safety factor approach (e.g., specifying an order of magnitude
lower permeability than required) is one approach. Another is to allow a
certain percentage of test failures based on experience during quality
control (QC) activities. Specifications usually include design drawings and
text and for clay liners often include the following:
• Facility configuration and size
t Foundation preparation
• Liner material characteristics (e.g., index properties)
• Liner thickness and permeability
§ Sidewall slope
5-41
-------
• Bottom slope and configuration
• Lif-t-orientation on sidewalls
• Lift thickness
• Maximum clod size
a Percent Proctor density
• Percent wet of optimum moisture content
• Scarification between lifts
• Compaction equipment and number of passes
• Test-fill compaction.
5.1.4.2 CQA Plan--
The design effort does not end with the start of construction but con-
tinues until the facility is completed. The CQA plan is prepared prior to
construction and establishes the lines of communication and the testing
program necessary to inform the designers, owner-operator, and regulatory
agencies about whether the construction process is producing a liner that
performs as required by regulation. The design engineer uses the CQA and CQC
results to identify unexpected problems encountered during liner construction
that can necessitate changes in the original design. The CQA program also
informs the designers and/or owner-operator whether the construction
specifications are being followed by the contractor. Further discussion of
CQA and of the CQA plan may be found in Section 5.3 of this document.
5.1.5 Design Case Studies
This section presents, as examples of current practices, important
design features of some of the clay-lined hazardous waste facilities
identified during the course of this study. These examples are presented to
Illustrate the variability in clay liner design and construction practices.
Further Information on these facilities, including diagrams of the facil-
ities, may be found in Chapter 7 of this report.
5.1.5.1 Site D—
This site, a secure landfill in the Southeastern United States, has a
clay liner constructed of compacted clay from a nearby borrow area (about
3 miles away) with a maximum compacted permeability of 1 x 10~7.* The
Uner 1s 10 feet thick and lies on a low-permeability deposit of opaline
claystone or Fuller's earth. The facility is constructed below the water
table, and a French drain system was installed around the site prior to
excavation to drain the overlying red sand material. Side slopes are 3 to
1. A 17-foot-wide bench with a 3-foot-wide drainage trench was constructed
at the top of the opal claystone. This bench enabled a thicker liner to be
Installed above the claystone unit. Below the bench, a minimum of 10 feet of
*A11 permeabilities in this section are laboratory measurements unless
otherwise specified.
5-42
-------
the claystone was left as a foundation, except for the leachate collection
sump, which 1-s underlain by 4.5 feet of claystone. A test fill was specified
for this facrH-ty- prior to construction to check equipment performance. The
clay liner was compacted 1n 6- to 8-inch lifts, with a sheepsfoot roller or a
smooth drum vibratory roller. The clay liner is overlain by a flexible mem-
brane liner (FML). Each landfill cell is completed prior to Its operation.
5.1.5.2 Site F—
This site is a series of five clay-lined landfill cells each lined with
10 feet of recompacted clay, an FML, and a 1- to 2-foot overlying compacted
clay layer to protect the FML. The clay liner was constructed of material
obtained from a borrow area about 15 miles from the site. This facility was
constructed above the ground with dikes to contain the wastes. The high
water table at the site, regulatory considerations, and ready availability
of suitable dike material (industrial slag) at the site contributed to the
selection of an aboveground design. Foundation preparation included removal
of industrial slag and organic silt material previously disposed on the
site.
A test fill was specified prior to construction. The liner was con-
structed in 6-inch lifts with a sheepsfoot roller. The sidewall slope is
2 to 1, and the sidewalls were compacted in horizontal lifts. The bottom
slope of the liner ranged from 1 to 2 percent. The entire liner was
installed prior to waste placement.
5.1.5.3 Site H~
This facility is a rectangular 9-acre cell used as a sanitary landfill.
it is lined with a 4-foot clay liner. Prior to clay liner Installation, the
fill area was rough-graded. Clay was brought to the facility from a nearby
borrow area. The variable nature of the borrow area required a soil's tech-
nician to be present during all removal operations to ensure that the mate-
rial met the project specifications. A technician present at the land-fill
rechecked the material by performing the required soil tests as the clay was
emplaced. The liner was compacted in 12-inch lifts with either a rubber-
tired or sheepsfoot roller.
This clay liner is underlain by two 12-foot x 100-foot lysimeters for
leak detection.
5.1.5.4 Site I —
This facility consists of three clay-lined surface impoundments that
cover approximately 8 acres 1n a semiarid region of the United States. The
impoundments are lined with two 5-foot compacted clay liners separated by a
15-inch granular leak detection layer. The Impoundments are partly above
ground and partly below ground and are contained by dikes. The dikes are
12 feet wide on top.
The interior dike sidewall slopes are 3 to 1. The soil excavated for
the facility was used as the liner material. The liners were compacted in
8-inch lifts with a sheepsfoot roller. A 2-foot layer of sandy soil was
placed on top of the liner for protection. Troughs for loading wastes into
the impoundment were lined with rip-rap. This facility experienced failure
from desiccation cracking of the liner because it was left exposed for 7
months prior to waste emplacement.
5-43
-------
5.1.5.5 Site J—
This facility consists of six ponds and a landfill. The ponds range in
size from l/2-ttr4 acres. The bottom liners at this facility consist of two
recompacted clay layers separated by a leak detection system. The lower clay
Uner is 1 foot thick and the upper clay liner is a minimum of 5 feet thick.
The sidewalls are constructed partially underground and partially above
ground. The aboveground sections are built upon dikes. The leak-detection
system is a 6-inch drainage blanket that slopes to a trench containing a
2-inch slotted polyvinyl chloride (PVC) collection pipe (Figure 7-15). The
dikes around the ponds are topped with a gravel-filled trench. This trench
1s filled with water to help prevent desiccation of the dikes during dry
months.
5.1.5.6 Site K—
This site in the Western United States consists of six double-lined
ponds ranging from 1.2 to 2 acres.. The pond liners were constructed of the
local excavated claystone material. The liner system consists of a 1-foot
recompacted basal liner. Sump and collection trenches are excavated into
this bottom liner. A 1-foot sand layer covers the lower clay liner. The
sand layer is overlain with 3 feet of compacted clay. The liner was
compacted with a segmented smooth-wheeled roller. The facility is
constructed partially in ground and partially above ground with dikes to
contain the wastes.
5.1.5.7 Site L—
The landfill consists of a flat double clay liner on top of which a dike
was placed to contain the wastes. The liner extends beyond the dike to form
the Uner for a 12-foot-wide drainage ditch that encircles the site.
The liner system at this facility consists of six layers listed below
from the top down:
• An 8- to 12-inch drainage layer (sand or gravel)
• An 18- to 25-inch compacted clay layer
• A 12-inch sand leak-detection layer
d
• A 12-inch compacted clay layer
• A 6- to 18-inch compacted soil layer
t A bidim type C34 (synthetic soil stabilization geotextile) layer.
5.1.5.8 Site M--
This 12-acre site is designed to consist of three cells of approximately
equal size. The facility is lined with 1 foot of recompacted clay. A
leachate collection system is above the clay liner, and a leak detection
system lies 2 feet below the clay liner. The first cell was excavated to a
maximum depth of 6 feet. This provides a minimum of 5 feet of separation
between the lowest layer of waste and the highest seasonal groundwater eleva-
tion. The sidewalls were excavated to a maximum slope of 3 to 1. The bottom
slopes 1 percent to the center of the landfill.
5-44
-------
The clay for the liner was obtained from a nearby borrow area. Prior to
the placement-of the clay liner, all large rocks, roots, and other foreign
matter were r-enwwed from the foundation. The foundation was then graded to a
1-percent slope and scarified to permit better bonding between the foundation
and the first clay lift. Uncompacted liner material was placed in 12-inch
lifts and moistened as necessary. The clay was then compacted with
"approved" equipment. Documentation of the specific types of equipment used
was not available.
5.1.5.9 Site P«
The facility consists of one double-lined hazardous waste cell. The
bottom liner is composed of leachate collection and detection systems as well
as a series of natural soil and bentonite/soil liners. The side liner and
dike containment system includes a soil and a bentonite/soil liner.
The bottom liner extends over the entire bottom and 6 feet up the
sides. It is a layered system containing two drainage or collection layers
and two soil liners. Proceeding from the top layer downward, the bottom
liner components are as follows:
• Leachate collection system—1 foot of No. 78 gravel and sand with
4-inch perforated PVC pipes
• Upper soil barrier—5 feet of compacted soil further subdivided
into three layers:
3 feet of compacted native soil. Permeability =
1 x lO'4 cm/s.
1 foot of enchanced soil, i.e., native soil blended with 9- to
12-percent polymer-treated bentonite. Permeability on the
order of 5 x 10~8 cm/s.
1 foot of compacted native soil. Permeability =
ID'4 cm/s.
• Leak detection layer—A 1-foot sand/gravel layer with perforated
pipe to detect leaks and/or to control seepage through the upper
barrier. The pipes are connected to several independent monitoring
stations to determine the approximate location of any leaking that
might occur.
• Lower soil barrier—A 6-inch layer of enhanced soil, i.e., native
soil blended with 9 to 12 percent polymer-treated bentonite.
Permeability on the order of 5 x 10~8 cm/s.
• Buffer zone—5 feet of either in situ or recompacted native
soil. Permeability = 1 x 10~4 cm/s.
The side liner system extends from a point 6 feet above the cell bottom
to the top of the cell. This section will not have liquid impounded against:
it; therefore, the liner system is not as extensive as the bottom liner.
5-45
-------
Proceeding from the top layer downward, the side liner components are as
follows:
t 1 foot of No. 78 gravel
• 18 inches of compacted native soil; permeability =
1 x lO-4 cm/s
• 6 inches of enhanced soil, i.e., native soil blended with 9-to
12-percent polymer-treated bentonite; permeability on the order
of 5 x 10~8 cm/s.
5.1.5.10 Site Q—
The landfill consists of a single containment cell covering an area of
approximately 3 acres. It is located in a former sand and gravel pit. The
cell has a double liner consisting of two 4-inch layers of a bentonite/soil
mixture on the bottom and side slopes up to a vertical elevation of 20 feet
above the cell bottom. The bentonite/soil layers are separated by a 12-inch
layer of sand on the bottom of the cell and a 6-inch layer of sand on the
side slopes. The side slopes above the 20-foot vertical level are covered
with a single 6-inch layer of the bentonite/soil mixture. All bottom and
side slope bentonite/soil surfaces are covered with a 12-inch protective
layer of gravel. The slope of the cell sidewalls varies from 2.5 to 1 to 3
to 1.
The central-plant (pugmill) mixing method was used "to blend the benton-
ite. It was spread with a dump truck, grader, screened boards, and hand
labor. A backhoe-mounted hydraulic tamper was used to compact the liner.
5.2 CLAY LINER CONSTRUCTION: METHODOLOGY AND EQUIPMENT
This section describes the methodology and the equipment that are
presently used for constructing clay liners. For clarity and convenience,,
the discussion is broken down into preinstallation, installation, and
postinstallation construction phases.
5.2.1 Preinstallation Activities
Before the liner is installed, the foundation is" prepared, groundwater
control measures are initiated for sites below the water table, leak detec-
tion systems may be installed, and the groundwater monitoring program is
implemented.
5.2.1.1 Foundation Preparation—
The foundation of a clay-lined hazardous waste containment facility is
the native soil substrate either unaltered or recompacted. For aboveground
facilities, dikes constitute part of the foundation. Operations during the
construction of foundations should include the following to accomplish these
goals (U.S. Department of the Army, 1977):
• Stripping and excavating to remove all soft, organic, permeable,
and otherwise undesirable materials. Proof-rolling with heavy
5-46
-------
equipment such as rubber-tired rollers or dozers should be done to
detect soft areas likely to cause settlement.
• Filling of rock joints, clay fractures, depressions, and any areas
where undesirable material has been removed. Fill material should
be engineered backfill compacted to the required specifications or
grouting material.
These items are important to ensure secondary containment and to ensure that
the liner retains its integrity during and after construction.
Foundation construction will essentially determine the configuration of
the clay liner. Whether the foundation is excavated or built above ground,
the sidewall slope and bottom slope must be properly controlled and shaped.
Excavation and shaping are accomplished with standard earth-moving equipment
such as dozers, scraper-pans, and road graders. Slope control is achieved
through traditional instrument surveying or through electronic (laser survey-
ing) devices. Trenches are cut for the collection sump and for leachate-
collection pipes if they are to be recessed into the liner.
Removal of soft spots and permeable areas in the foundation is
accomplished with standard excavating equipment. Once these are removed, the
resulting holes and irregularities are backfilled and compacted in a manner
similar to clay liner construction. At some facilities, the entire founda-
tion surface is disked or tilled to a depth of 1 to 2 feet and recompacted in
one or two lifts. At other facilites, the foundation is left unaltered. At
the end of foundation construction, the entire landfill base can be seal-
rolled (rolled smooth) to seal the soil and to ensure that precipitation that
may fall on the site prior to liner placement will run off properly and will
not puddle or pond on the foundation surface. Final proof rolling also can
be used to ensure the integrity of the finished foundation.
If the facility is constructed entirely or partially aboveground, dikes
are constructed around the periphery of the excavation to serve as retaining
walls for the liner (see Chapter 7 for examples of diked facilities). Dikes
are generally earth or rockfill embankments and are constructed with the same
techniques and equipment used to construct earth or rockfill dams. Founda-
tions for dikes are prepared to control underseepage, to provide satisfactory
contact with the overlying compacted fill, and to minimize differential
settlement (U.S. Department of the Army, 1977). Dike foundation preparation
operations are generally the same as described for the clay liner founda-
tion. Dikes may be constructed in horizontal compacted lifts in a manner
similar to clay liners. For further information on dike construction, the
T^er,,!s referred to Sherard et al., 1963; U.S. Department of the Interior,
1974; Winterkorn and Fang, 1975; U.S. Department of the Army, 1977; and U.S.
Department of the Navy, 1982.
5.2.1.2 Groundwater Control--
When a foundation excavation extends below the water table, unbalanced
hydrostatic pressures from groundwater can develop in the foundation bottom
and sidewalls. These pressures can cause seepage, sidewall failure and
collapse, and bottom heave or rupture during subsequent facility construction
and operation. Groundwater control measures can reduce this hydraulic
5-47
-------
pressure and thus reduce the likelihood that these failures win occur. The
most common methods of reducing hydrostatic head around a facility are:
• Construction of a slurry cutoff wall
• Trenching and pumping
• Installation of dewatering wells for lowering the water table by
pumping.
Slurry cutoff walls can be constructed by excavating a trench with a
back hoe (or similar equipment) and backfilling with bentonite, asphalt,
cement grout, or other suitable material. Alternately, slurry walls can be
installed with the vibrating-beam technique, in which bentonite is injected
along vibrating beams driven into the ground. Trench and pump methods
involve digging trenches and installing sump pumps to remove infiltration.
Dewatering wells are constructed by standard well-drilling techniques
(Johnson Division, 1975) and are pumped to lower the water table, thereby
reducing hydrostatic pressure on the facility sides and bottom.
A complete discussion of these techniques 1s beyond the scope of this
document. For more information, the reader 1s referred to D'Appolonia, 1980;
U.S. Environmental Protection Agency, 1984; Schmednecht and Harmston, 1980;
U.S. Environmental Protection Agency, 1982a; and Cashman and Haws, 1970.
Groundwater control to preserve liner integrity 1s necessary only during
construction and operation of the facility. Once the facility is filled, the
weight of the waste will balance the inward hydrostatic forces on the liner,
eliminating the need for groundwater control measures.
5.2.1.3 Leak Detection System Installation—
If a leak detection system is part of the facility design, it is neces-
sary to install this prior to liner emplacement. This involves laying
granular drainage layers and pipes over part or all of the foundation.
Following Its installation, the system can be covered with a properly graded
soil blanket or a geotextile to prevent damage during subsequent construction
activities, to provide a stable base for placement of the basal-liner lifts,
and to prevent piping of clay liner material into the porous drainage layer.
For further information on leak detection system desi-gn and installation, the
reader Is referred to related material on leachate collection systems in
Bass, 1986, and U.S. EPA, 1985, 1986 (pages 39-46), and 1987.
5.2.2 Clay Liner Installation
Clay liners are constructed by compacting clay soil. Clay liner
materials covered in this document include natural, untreated soil and soil
mixed with bentonite additives to achieve a lower permeability. Installation
procedures differ for these materials, and they will be addressed separately
in this document.
5.2.2.1 Natural Soil Liners-
Natural soil liner material may be excavated from the site during
foundation preparation or may be transported to the site from a nearby
borrow source 1f soil material at the facility is insufficient or if the
1n situ soil is not a suitable liner material. QC measures are necessary at
5-48
-------
the borrow site and/or at the facility to ensure that the liner material
meets design specifications (see Section 5.3.4). Excavation at the borrow
site is usually-accomplished with 'backhoes," front-end loaders, or other
standard excavating equipment. One engineer recommended a side-cut excavator
for use at the borrow site. This equipment is apparently very productive and
produces material with a uniform, small clod size.
Prior to emplacement, a borrow pile or storage pile of the liner
material usually is established at the site. Depending on climatic condi-
tions and the condition of the clay, the borrow pile may have to be protected
from moisture loss and erosion. If the clay is wet when originally placed in
the pile, protection from moisture loss will reduce the amount of water that
must be added to the clay prior to compaction. This is especially important
in arid climates. In areas of heavy rainfall, erosion prevention may be
necessary to prevent loss of liner material. Plastic or soil covers may be
installed to control borrow pile moisture content and erosion. Alterna-
tively, the pile may be graded and seal -rolled with motor graders, bull-
dozers, and smooth-wheeled rollers.
5.2.2.1.1 Liner Material Emplacement— Thickness requirements for clav
liners for hazardous waste facilities (usually 2 to 12 feet) necessitate
installing the clay in a series of lifts (layers) to ensure uniform compac-
tion throughout the liner. For each liner lift, material from the borrow
pile is emplaced into the facility with scraper-pans or trucks and uniformly
distributed over the site with dozers or graders (Figure 5-6). Lift
thickness is controlled during emplacement by using measuring staffs, shovel
blades, or instrument surveys. Figure 5-7 shows liner material being
emplaced in the area of a collection pipe. The foundation was excavated for
the pipe, and, as a result, the lowest liner lift was thicker in this area.
(This is also illustrated in Figures 5-4 and 7-11.)
4.^ J:1ner emplacement methods vary with the size of the facility and the
method of facility operation. For small facilities, individual liner lifts
are often installed over the entire excavation, and the liner is constructed
as a single unit. For larger facilities and for continuous operation facil-
ities, where the wastes are emplaced in the facility as parts of the liner
are built, the liner is installed in segments. After each liner segment is
.pnmnil6*' 1J '? .b5v?1!d °f steP-cut with grading equipment so that the next
segment may be tied into the previously installed segment, eliminating a
potential pathway for seepage through the liner along the boundary (see
Section 5.1.3.3, Figure 5-5).
. a5'?'?;J;2. c]od Size Reduction— Following placement, the liner material
for each 11ft is broken up for homogenization and clod size reduction (clods
are unbroken aggregates of liner material). The clod size of the liner
material affects moisture control and compaction operations. Reduction in
clod size increases the surface-volume ratio and decreases the time it takes
for moisture to become evenly distributed within the clod (curing time),
thereby facilitating moisture control operations. In addition, clod size
reduction allows more effective and homogeneous distribution of compactive
energy through the lift than would be achieved in lifts with clods of greater
?QQ! (Nithiam, D'Appolonia Consulting Engineers, personal communication,
}ffii; Sey' EMCON Assoc1ates> San Jose, California, personal communication,
ia«4;. Homogeneous compaction helps ensure homogeneous permeability.
5-49
-------
* ''
Source: Photo courtesy of Wisconsin Department of Natural Resources
Figure 5-6. Liner material emplacement.
5-50
-------
Source: Photo courtesy of Wisconsin Department of Natural Resources
Figure 5-7. Emplacement of liner material over foundation excavation underneath a
collection pipe.
5-51
-------
Opinions differ among design engineers on optimum clod size for clay
Uner construction.' Clod size recommendations gathered through interviews
Include 1 1nch^-2-to 3 inches, no larger than one-half the lift thickness,
and no larger than the lift thickness (see Table 5-11, Section 5.3.3.4).
Clod size reduction is especially important in restricted areas where hand
compactors must be used (e.g., around penetrating objects and in the corners
of some facilities). Specifications obtained from the U.S. Department of the
Navy (1982) and from the nuclear industry require that clod size be reduced
to 1 to 3 inches in areas that are to be compacted by hand. Construction
specifications sometimes specify maximum clod size, but not in most cases.
In actual practice, clod size may not be carefully controlled. Clods with
horizontal dimensions exceeding the lift thickness have been observed by the
authors at some facilities under construction.
There is no documented information on the effect of clod size on the
achievement of specified liner permeability in the field. However, in a
laboratory study of compacted clay, Daniel (1981) demonstrated that clod size
can significantly affect permeability. Table 5-8 presents the results of
this experiment. Daniel (1984) has also described a field case study where
large clods and Inadequate curing have resulted in nonuniform moisture
distribution in compacted clay liners.
The importance of clod size control depends, to some extent, on how much
moisture must be removed or added to meet the specified compaction moisture
content. If moisture must be added, large clods may necessitate long curing
times and repeated moisture applications to ensure uniform moisture content
across the clods, resulting in delays in construction schedules. Thus,
efficient clod size reduction can save construction time and money.
Clod size reduction is usually accomplished using disk harrows or rotary
tillers with various shaped tilling blades. Availability usually determines
equipment selection. High-speed pulvi-mixers (e.g., BOMAG® MPH 100),
designed for breaking up soil and old asphalt pavement, have proven to be
superior to tillers or disks for blending bentonite with soil (Kozickl and
Heenan, 1983) and may be very good for breaking up soil clumps for untreated
soil liners (Figure 5-8). The manufacturer claims that these machines are
capable of reducing clod size of cohesive soils to 2 inches in a single pass
and 80 percent of the clods to less than 1.5 inches after two passes.
5.2.2.1.3 Moisture Control—Moisture control includes the addition or
removal of water from the liner soil to achieve the specified compaction
moisture content (molding water content). Correct and uniform moisture
5-52
-------
TABLE 5-8. EFFECT OF CLOD SIZE ON
PERMEABILITY OF LABORATORY
COMPACTED CLAY
Maximum size Permeability
of clods (in.) (cm/s)
3/8 2.5 x 107
3/16 1.7 x 1Q8
1/16 8.5 x 109
Source: Daniel, 1981.
5-53
-------
Source: Photo courtesy of Bomag, Inc.
Figure 5-8. Use of pulvi-mixer for clod size reduction
5-54
-------
content is essential for compacting to the specified permeability. Minimum
permeability-can be obtained onJy ,,if the molding water content is within
several percentage points on the "wet side of optimum." (This is fully
explained in Chapter 3.) For this reason, it is important to control the
moisture content of the liner material carefully prior to and during liner
construction to ensure that the moisture is uniformly applied and distributed
throughout the soil of each lift. Nonuniform moisture distribution in clay
liners has been attributed to inadequate breakup of large clods prior to
compaction; uneven water distribution by sprinkling devices, especially on
slopes; and inadequate curing time allowed for the water to penetrate the
soil (Ghassemi et al., 1983).
Moisture is added to liner material prior to placement. Often it is
most convenient to do this at the borrow area, although it may be necessary
to add moisture during liner material emplacement if the soil has dried
during handling and transport (Figure 5-9). Moisture addition is accom-
plished with sprinkler trucks, sprinkling systems, or other sprinkling
devices during spreading and mixing operations. Added moisture must be
thoroughly mixed into the soil with mixing devices such as disk harrows,
rotary cultivators, or pulvi-mixers, to ensure that moisture is uniformly
distributed throughout the soil mass.
Adequate equilibration time after moisture addition is critical to
ensure that moisture is uniformly distributed throughout the soil. Clod size
reduction reduces penetration time and helps achieve uniform moisture content
across all clods within reasonable time. Equilibration times may reach days
or weeks if soil material is very dry or if soils with a high montmorillonite
content are used in the liner. Moisture addition in the borrow area may be
necessary in these cases to avoid construction delays that would result if
the material were moistened in place prior to compaction and then allowed to
equilibrate. In dry areas or during dry periods of the year, it may be
necessary to cover the soil during equilibration to prevent additional
moisture loss from evaporation.
If the liner material becomes too wet to work during construction
moisture reduction may be accomplished through a combination of mechanical
agitation, aeration, and solar drying.
Moisture content may be maintained during inactive periods by sloping
and seal-rolling the liner to ensure proper runoff or by covering the liner
with plastic or a layer of moist soil to prevent drying or overwetting of the
liner material. Prevention of drying is important because desiccation can
cause cracking of the liner, which can greatly increase its permeability.
Moisture content measurement is a QC activity; techniques for measurinq
soil moisture content are discussed in Chapter 3, and techniques for estimat-
ing soil moisture during construction are discussed in Section 5.4.3.2.3.
5.2.2.1.4 Compaction—The important theoretical aspects of compaction
of fine-grained soils to achieve low permeability are discussed in Chapters 2
and 3. Compaction quality control and quality assurance are discussed in
Section 5.3.4. The discussion below is limited to the practical aspects of
compaction in the field.
5-55
-------
Figure 5-9. Moisture addition to liner material prior to compaction.
5-56
-------
Compaction of a clay liner is accomplished through standard compaction
practices as used in other earthwork construction. The following variables
exist for compaction operations* -v ^ <«
• Lift thickness and number
• Equipment type and size
• Number of equipment passes
t Soil quality "
• Soil moisture content.
Because of their interrelated nature, it is necessary to control all of these
variables to achieve adequate compaction in the field. Compaction of clay
soils to obtain low permeability differs from compaction of earth to obtain
structural stability in that compaction is performed at a higher moisture
content (usually 2 to 3 percent wet of optimum). Thus, moisture content is
one of the most critical factors to control when clay liners are compacted.
In a recent study EPA compiled information on the compaction practices
followed during the construction of a number of soil liners (Elsbury, 1985),.
These data are presented in Table 5-9. Additional information on construc-
tion practices followed at one German and 22 U.S. waste disposal facilities
was gathered by Peirce et al., 1986.
A critical aspect of the compaction of clay liners is to tie (join)
together adjacent lifts properly. Improperly tied lifts can result in
greatly increased horizontal permeability along the lift interface, an
especially serious problem when sidewall lifts are compacted in a horizontal
manner (see Section 5.1.3.2.4). Figure 5-10 illustrates joints and seepage
along lift boundaries in sections of two experimental liners. These liners
were compacted with a small self-propelled sheepsfoot roller (first figure)
and a vibratory plate hand compactor (second figure). Two measures recom-
mended by the interviewed design engineers for tying together liner lifts are
(1) scarification of the surface of the last installed lift, through a disk
harrow or other device, before the next lift is installed and (2) control of
the moisture content of the adjacent lifts so that they are equivalent and as
specified. Haxo (1983) recommended the use of compaction equipment with feet
that are at least 50 percent longer than the height of the compacted lift.
However, most sheepsfoot rollers have feet that are 7 to 10 inches long
(Hilf, 1975), which could hamper implementing Haxo's recommendation.
Johnson and Sallberg (1960) reported on several studies of the
development of "compaction planes" or laminations between lifts. The results
of these studies indicate the following:
t For all types of rollers, laminations or lift partings are much
more likely to occur in soils compacted wet of optimum than in
those compacted dry of optimum.
• Laminations in a compacted soil are produced primarily from
"springing" of the lift under compactive equipment.
5-57
-------
TABLE 5-9. COMPACTION EQUIPMENT AND RELATED SPECIFICATIONS FOR CONSTRUCTING SOIL LINERS
Ul
01
00
Geographic
Location
Alabama
Al abama
Alabama
California
Cal i forma
California
California
Colorado
Compaction
Equipment
— —
Self-propelled
23,000# sheeps-
foot roller
Sheepsfoot roller
Sheepsfoot roller
w/ water ballast
Sheepsfoot roller
20-30 ton sheeps-
foot roller or
Compaction
Moisture
Content
2 to 3% above
optimum
0 to 3% above
optimum
-1 to +3% of
opti mum
+5% of optimum
M
0 to +4% of
opti mum
> +1% of
optimum
+1 to +2% of
opti mum
Maximum*
Density
88-80% standard
Proctor
95% standard
Proctor
95% standard
Proctor
90% modified
Proctor
'1 hi
90% relative
compaction
90% relative
compaction
98% modified
Proctor
Lift Maximum
Thickness Clod Size
! 1
i
15 cm loose
23 cm loose 10 cm nominal
15 cm cmptd effective
di ameter
15 cm loose
15 cm loose —
15 cm cmptd
20 cm loose 2.5 cm
15 cm cmptd
20 cm loose 2.5 cm
15 cm cmptd
segmented wheel
roller
-------
TABLE 5-9. (continued)
01
I
Geographic
Location
Georgia
111 i noi s
Indiana
Indiana
Michigan
Michigan
New York
Ohio
Okl ahoma
South
Carolina
Compaction
Equipment
—
Rubber-tired
roller
—
Vibratory
padfoot
Scraper
traffic
Heavy rubber-
tired roller t
—
Padfoot
Towed vibratory
sheepsfoot
Compaction
Moisture
Content
—
+11 to +13 of
opti mum
__
optimum
—
—
—
2% above
optimum
Wet of optimum
Maximum*
Density
95% max. dry
density
—
90% modified
Proctor
90% modified
Proctor
90% modified
Proctor
93% modified
Proctor
90% modified
Proctor
—
95% standard
Proctor
—
Lift Maximum
Thickness Clod Size
20 cm loose
23 cm loose --
15 cm loose
23 cm loose
--•
20-30 cm
loose
15-23 cm
loose
-- __
15 cm loose 2.5 cm
10 cm cmptd 2.5 cm
15 cm cmptd 2.5 cm
-------
o>
o
TABLE 5-9. (continued)
Geographic
Location
Compaction
Equi pment
Compaction
Moisture
Content
Maximum*
Density
Lift
Thickness
Maximum
Clod Size
Texas
Texas
Utah
Padfoot
optimum
95% standard
Proctor
20-23 cm
loose
; t
2.5 cm
^Specification.
-------
Source: Photo courtesy of Richard Warner, University of Kentucky
Joints Between Linerlifts
Source: Photo courtesy of Kirk Brown and Assoc., Austin, Texas
Joints Between Lifts Coated With Seepage
Figure 5-10. Joints and seepage along lift boundaries.
5-61
-------
• Tamping feet tend to mesh the boundary between successive layers.
Because clayilnars must be compacted wet of optimum to achieve minimum
permeability, this information suggests that a great deal of attention must
be paid to ensure proper bonding of clay liner lifts.
Two methods of sidewall compaction are used for clay-lined hazardous
waste facilities. Depending on soil conditions, if sidewall slopes are less
than around 2.5 (H) to 1 (V), it may be feasible to compact the sidewalls in
lifts that are continuous with the bottom liner lifts. This ensures
continuity between the liner bottom and sidewalls and also orients the lift
boundaries and compacted clay fabric parallel to the liner surface. However,
if the sidewall slope is ,to be steeper than 2.5 to 1, the sidewalls may have
to be compacted in horizontal lifts because most compaction equipment cannot
operate on such steep slopes. When sidewalls are compacted in a horizontal
manner, they are overbuilt and trimmed back to the final slope with a motor
grader or excavator.
Compaction equipment is usually selected based on techniical performance
and availability. Some of the technical factors to be considered are whether
side slopes or bottom slopes are to be compacted, lift thickness, and liner
material. In some cases technical considerations are not weighed as heavily
as they should be, and the choice is made based on what equipment is
available that can "do the job." The following is a list of the kinds of
equipment currently used for compacting clay liners:
• Sheepsfoot or clubfoot roller—self-propelled and towed
• Padfoot (pegfoot or wedgefoot) roller—self-propelled and towed
• Vibratory sheepsfoot roller—self-propelled and towed
• Rubber-tired roller
• Wobble-wheel, rubber-tired roller
• Smooth-wheeled roller
• Vibratory smooth-wheeled roller
• Vibratory plate compactor (for compaction around penetrating
objects—hand operated)
• Bulldozers
• Tractors.
Table 5-10 describes the types and typical uses of compaction equipment
(U.S. Department of the Navy, 1982). This table covers all types of earth-
work compaction operations and is not limited to clay liner construction;
only that equipment suitable for cohesive soils at wet-of-optimum moisture
levels should be used for clay liners. Geotechnical textbooks and most
experts Interviewed recommend the use of sheepsfoot or tamping foot rollers
for compacting cohesive (clay) soils to achieve low permeability. The
different types of roller feet are illustrated in Figure 5-11. All of these
5-62
-------
TABLE 5-10. COMPACTION EQUIPMENT AND METHODS
Requirements for Compaction of 95 to 100 Percent Standard
Maximum Density
Equipment
Type
Applicability
Compacted
Lift
Thickness, Passes or
in. Coverages
Dimensions and Weight of Equipment
Possible Variations in
Equipment | '
Sheepsfoot
Rollers
For fine-grained soils or
dirty coarse-grained soils
with more than 20 percent
passing No. 200 sieve. Not
suitable for clean coarse-
grained soils. Particularly
appropriate for .compaction of
impervious.zone for earth dam
or linings where bonding of
lifts is important.
en
i
o>
CO
Rubber Tire
Roller
Do.
For clean, coarse-grained
soils with 4 to 8 percent
passing the No. 200 sieve.
For fine-grained soils or well
graded, dirty coarse-grained
soils with more than 8 *
percent passing the No. 200
sieve.
Smooth Wheel Appropriate for subgrade or
Rollers base course compaction of
well-graded sand-gravel
mixtures.
Uo May be used for fine-grained
soils other than in earth
dams. Not suitable for
clean well-graded sands or
silty uniform sands.
10
6 to 8
4 to 6 passes
for fine-
grai ned soi1.
6 to 8 passes
for coarse-
grained, soil.
3 to 5
coverages
4 to 6
coverages
8 to 12 4 coverages
6 to 9 6 coverages
Soil Type
Foot Foot
Contact Contact
Area Pressures
sq. ft. psi
Fine-grained 5 to 12 250 to 500
soil PI>30
Fine-grained 7 to 14 200 to 400
soil PK30
Coarse-grained 10 to 14 150 to 250
soil :
Efficient compaction of soils wet of
optimum requires less contact pres-
sure than the same soils at lower
moisture contents.
Tire inflation pressure of 35 to 130
psi for clean granular material or
base course and subgrade compac-
tion. Wheel load 18,000 to 25,000
Ibs.
Tire inflation pressures in excess of
65 psi, for fine-grained soils of
high plasticity. For uniform clean
sands or silty fine sands, use
large size tires with pressures of
40 to 50 psi.
Tandem type rollers for base course
or subgrade compaction 10 to 15 ton
weight, 300 to 500 Ibs per lineal
in. of width of rear roller.
3-wheel roller for compaction of
fine-grained soil; weights from 5
to 6 tons for materials of low
plasticity to 10 tons for materials
of high plasticity.
For earth dam, highway and
airfield work, articulated
self propelled rollers are
commonly used. For smaller
projects, towed 40 to 60
inch drums are used. Foot
contact pressure should be
regulated so as to avoid
shearing the soil on the
third or fourth pass.
Wide variety of rubber tire,
compaction equipment is
available. For cohesive
soils, light-wheel loads,
such as provided by wobble-
wheel equipment, may be
substituted for heavy-wheel
load if lift thickness is
decreased. For granular
soils, large-size tires are
desirable to avoid shear
and rutting.
3-wheel rollers obtainable
in wide range of sizes.
2-wheel tandem rollers are
available in the range of 1
to 20 ton weight. 3-Axle
tandem rollers are gener-
ally used in the range of
10 to 20 tons weight. Very
heavy rollers are used for
proof rolling of subgrade
or base course.
-------
TABLE S-10. (continued)
en
i
o>
Requirements for Compaction of 95 to 100 Percent Standard
Maximum Density
Equipment
Type
Vibrating
Sheetsfoot
Rollers
Vibrating
Smooth Drum
Rollers
Compacted
Lift
Thickness, Passes or Possible Variations in
Applicability in. Coverages Dimensions and Weight of Equipment Equipment
For coarse-grained soils 8 to 12 3 to 5 1 to 20 tons ballasted weight. May have either fixed or
sand-gravel mixtures Dynamic force up to 20 tons. variable cyclic frequency.
For coarse-grained soils 6 to 12 3 to 5
sand-gravel mixtures - rock (soil)
fills ' to - do - - do -
Vibrating For coarse-grained soils with
Baseplate less than about 12 percent
Compactors passing No. 200 sieve. Best
suited for materials with 4 to
8 percent passing No. 200 sieve,
placed thoroughly wet.
4 to 6
8 to 10 3 coverages
Single pads or plates should weigh
no less than 200 Ibs. May be used in
tandem where working space is avail-
able. For clean coarse-grained soil,
vibration frequency should be no less
than 1,600 cycles per minute.
Vibrating pads or plates
are available, hand-
propelled, single or in
gangs, with width of cover-
age from 1-1/2 to 15 ft.
Various types of vibrating-
drum equipment should be
considered for compaction
in large areas.
Crawler Best suited for coarse-grained
Tractor soils with less than 4 to 8
percent passing No. 200 sieve,
placed thoroughly wet.
60 to 10
3 to 4
coverages
Vehicle with "Standard" tracks having Tractor weight up to 85 tons.
contact pressure not less than 10
psi.
Power Tamper
or Rammer
For difficult access, trench
backfill. Suitable for all
inorganic soils.
4 to 6 in. 2 coverages
for silt
or clay,
6 in. for
coarse-
grained
soi 1 s .
30-1 b minimum weight. Considerable
range is tolerable, depending on
materials and conditions.
Weights up to 250 Ibs.,
foot diameter 4 to 10 in.
Source: U.S. Department of the Navy, 1982.
-------
Foot
(a)
Drum
Tapered or
Wedge Foot
Cross Section
of Tapered Foot
(d)
Drum
V3
Clubfoot
Pegfoot
Sheepsfoot
Not drawn to scale.
After Johnson and Sallberg, 1960
Figure 5-11. Sketches of different types of roller feet.
5-65
-------
tamping rollers are often referred to gerierically as sheepsfoot rollers. The
general consensus is that the kneading action of these devices affects the
soil fabric •hr-a-beneficial way (see Chapter 3 for a discussion of soil
fabric and permeability). However, this opinion is derived from laboratory
studies of impact versus kneading compaction, which show that lower
permeabilities can be achieved with the latter. We know of no field studies
that demonstrate the superiority of tamping or sheepsfoot rollers over other
types of rollers in reducing the permeability of cohesive soils. This is an
important point because laboratory kneading compactors use a tamping foot of
about 0.5 inch in diameter, and the test procedure ensures that its impacts
cover the entire surface area of the test soil. In contrast, the feet on a
sheepsfoot roller are several inches across and are separated on the roller
by a space of several inches (Figure 5-12). Thus, they do not induce shear
strains as intensely or as uniformly through the soil mass as the laboratory
technique does.
One engineer interviewed related a case in which the required compacted
density for a clay liner could not be achieved with a sheepsfoot roller but
was achieved with a vibratory smooth-wheeled roller. The differences between
permeabilities for the two types of rollers were not determined in this
case. An interview with another design engineer revealed that at another
site (South Central United States) adequate compaction in near-saturated
clays was being achieved with several passes of a bulldozer. One major engi-
neering firm interviewed requires by specification that for cohesive soil
compaction at nuclear power plants a variety of equipment (including a towed
sheepsfoot roller, a self-propelled static wedgefoot roller, and a towed
static wedgefoot roller) be evaluated with several different lift thicknesses
in a test fill prior to construction. This facilitates selection of the most
suitable equipment and lift thickness for each soil type to be compacted (see
section 5.3.3.1).
Hilf (1975) mentions that sheepsfoot or clubfoot rollers are preferable
to other roller types because their mixing action produces a more homogeneous
liner with respect to moisture content and physical characteristics. Hilf
(1975) also states that pad-type tamping foot rollers, because of larger foot
end areas, do not blend and mix embankment materials as effectively as
conventional sheepsfoot rollers. Figures 5-11 and 5-12 illustrate the
difference between sheepsfoot and padfoot (or pegfoot.) rollers.
The size and configuration of a facility may place some limits on
equipment selection. Large compaction equipment (necessary for thick 11ft
compaction) may not be usable at small sites because of large turning
rad11. If sidewalls are steep, some equipment may not be able to negotiate
the slope. Sidewall slopes of 2.8 to 1 or less have been recommended for
sheepsfoot rollers (Boutwell and Donald, 1982). Figure 5-13 shows the
compaction of a 2-to-l slope with a sheepsfoot roller. In this case it was
necessary to both push and tow the roller to negotiate the slope. If
sidewalls are compacted in horizontal lifts, lift width may restrict the size
of the compaction equipment.
The number of passes necessary to achieve the specified compactive
effort depends upon the size, weight, and configuration of the equipment. In
general, small equipment requires more passes than large equipment.
Compactive effort can be estimated from the towing force required to tow the
equipment per unit distance multiplied by the number of equipment passes. It
5-66
-------
Source: Photo courtesy of Wisconsin Department of Natural Resources
Figure 5-13. Compaction on a 2(H) to 1(V) slope with a towed sheepsfoot roller.
5-69
-------
1s usually expressed on a per unit volume of fill basis (e.g., foot pounds
per cubic foot) (Selig, 1982; Johnson and Sallberg, 1960) as follows:
( J ( ) (L) „ compactive effort per unit volume of fill (ft lb/ft3),
(W) (T) (L) }'
where:
F * draw bar pull (Ib)
N * number of passes
L 3 length of each pass (ft)
W = roller width (ft)
T = lift thickness (ft).
For each soil/moisture content/equipment combination, a different num-
bers of passes are required to achieve a specified permeability and density.
Thus, it 1s extremely important to determine the compactiye effort necessary
to achieve the design permeability with each type of compaction equipment to
be used 1n a test fill prior to construction (Section 5.3.3.1.1). Mitchell
et al. (1965) demonstrated that, with some clays, Increasing compactive
effort can decrease permeability hundredfold without changing density or
moisture content by additional shearing that breaks down soil structure.
Compaction should be controlled in the field by measuring density, moisture,
and compactive effort. The specified values are based on the relationship,
previously derived 1n the laboratory and confirmed 1n the field test fill,
among these variables and the permeability for the specific soil and the
specific compaction equipment to be used.
Moisture content, density, and compactive effort measurements are neces-
sary for controlling compaction to ensure that the specified permeability is
achieved 1n the field. Visual observations of construction operations are
also critical to compaction quality control. A full discussion of compaction
quality control 1s found 1n Section 5.3.
5.2.2.2 Admixed Bentonlte Liners— *
In areas where suitable soils for clay liners are not available or
cannot be economically delivered to the construction site, bentonite addi-
tives may be blended Into the unsuitable native soils to enable them to be
compacted to the required .permeability. Selection of the type and the proper
percentage of bentonite additive 1s described 1n Section 5.1.2. The differ-
ence between Installation of natural soil liners and bentonite admixtures
mainly lies in the liner emplacement methods. When bentonite 1s stored
onslte, 1t is critical to keep 1t covered and protected from precipitation as
1t cannot be worked 1n the wet state.
5.2.2.2.1 Bentonite Mixing and Spreading—When bentonite soil additives
are used, they must be thoroughly and uniformly mixed with the native soil.
5-70
-------
This 1s most easily done when the native son is relatively dry. Mixing can
be accomplished in place with the additive applied evenly over the site and
then mixed iirtp the native soil, or mixing can be done in a central plant
where the add!Lives and soil are blended lira mixing device and the final
mixture is spread over the site.
Central plant mixing is the preferred method of several design engineers
interviewed and in a study by Lundgren (1981) has been shown to be more
effective than tilling in place. In this method, the bentonite and the soil
are mixed in a pugmill, cement mixer, or other device where moisture is added
during the mixing process. Moisture content, particle size, and bentonite
content must be monitored and controlled during this mixing process. Central
plant mixing is illustrated in Figure 5-14. Laboratory tests of central
P^nt mixing of bentonite and soil yielded the following results (Lundgren,
• More than 10 minutes mixing time 1s preferable.
• The soil may be dry, naturally conditioned (drained), or saturated
when the bentonite is added.
• Water should be Introduced into the mixer after the bentonite and
after a couple of minutes of homogenizing.
t All investigated bentonites (five types) were homogenized to the
same degree.
Following central plant mixing, spreading may be accomplished as with natural
soil liners (using trucks or pans and graders or dozers) or with a continuous
asphalt paving machine (Geo-Con, 1984).
Currently, central plant mixers are commercially available that are
specifically designed for bentonlte/soil admixtures. These mixers are
capable of producing 1,000 yd3/day of admixed material. Computer
controls for these devices are capable of achieving an accuracy of +0.5
percent moisture 1n the admixture throughout the project (Geo-Con, T984).
In-place spreading and mixing Is a commonly practiced construction
method for bentonite/soll liners. Locally available^lner material 1s first-
spread uniformly over the prepared foundation. The bentonite 1s then spread
uniformly over the native Uner material. If the side slopes are steep
( Lto Lor 9reater) or the Uner is small, bags of bentonite can be placed
on the site in a predetermined pattern and then the bentonite 1s manually
raked over the Uner material. Bag placement must be determined carefully so
that the specified quantity of bentonite per cubic foot of Uner 1s
maintained uniformly throughout the fill. This is frequently accomplished by
placing the bags of bentonite 1n a grid pattern over the facility site.
Close visual scrutiny is necessary during manual mixing to ensure that the
spreading 1s adequate. Alternatively, for larger sites with sldewall slopes
of 2.5 to 1 or less, mechanical or pneumatic spreaders can be used. A belt-
feed cement spreader has been found to be particularly suitable for benton-
ite. This spreader, pictured 1n Figure 5-15, requires only two men and
provides uniform spreading rates at up to 25 ton/hr (Koz1ck1 and Heenan,
1983). The spreader can be operated with feed from dump trucks or, when
5-71
-------
en
i
XJ
ro
Source: Photo courtesy of Geo-Con, Inc., Pittsburgh, Pennsylvania
Figure 5-14. Central plant mixing of bentonite and soil.
-------
Source: Photo courtesy of Ground Engineering, Ltd.. Regina, Saskatchewan
Figure 5-15. Truck-loaded bentonite spreader.
-------
ut
i
Source: Photo courtesy of Ground Engineering, Ltd.. Regina, Saskatchewan
Figure 5-15. Truck-loaded bentonite spreader (continued).
-------
modified with a cyclone, fed pneumatically in bulk from tank trucks
(Figure 5-16)-
________ ..$* . >••"• :>'** r>t
Following in-place spreading, the bentonite must be mixed into the soil
along with enough water to bring the mixture to the proper moisture content.
This step activates (swell) and disperses the bentonite. Mixing can be
accomplished with disk harrows, rototillers, tined rotovators, or a high-
speed pulvi-mixer (soil stabilizer). Disks (Figure 5-17) and tillers should
make several passes in a crisscross pattern to help break up clods and ensure
more complete mixing (Ghassemi et al., 1983). Water may be added from
sprinklers attached to the mixer or by sprinkling between passes.
Rototillers have been demonstrated to be more efficient at mixing than disk
harrows (Ghassemi et al., 1983), and larger wheeled rototillers are more
effective than the wheel! ess types (Lundgren, 1981). Six or eight passes are
generally required for adequate mixing. Ha««
4. u-^ Severa1 Canadian installations, a high-speed pulvi-mixer (soil
stabilizer) achieved very good mixing to a depth of 200 mm in the first pass
ana to a depth of 300 to 350 mm on subsequent passes (Figure 5-18). Water
can be added after the first dry-mix pass (ICozicki and Heenan, 1983) or, as
illustrated (Figure 5-18), it can be added during mixing. Two passes are
generally sufficient to mix the bentonite thoroughly and reduce all clods to
less than 1 inch. Although this equipment is considerably more expensive
than conventionally used mixing devices, the reduced number of passes and
ensurance of better mixing make it cost competitive (Kozicki, 1983, Ground
Engineering Ltd., Regina, Saskatchewan, Canada, personal communication).
4 ,In7?lac? sPread1"9 and mixing are generally recommended only for
single-lift (4- to 6-inch) liners; for thicker liners, the central plant
method is preferred. One contractor with admixed liner Installation
experience has stated that because of the difficulty of conducting stringent;
quality assurance/quality control for in-place spreading mixing, Eentral
e"111/? 9 1f««_? Preferred metnod f°r hazardous waste containment facility
(Kyan, 1984).
5.2.2.2.2 Compaction— Foil owing spreading arid mixing, the bentonite
admixture is compacted. Vibratory smooth-wheeled rollers or vibratory-plate
compactors are preferred for this operation, for two reasons. First, because
admixed liners are often thin (4 to 6 Inches), tamplng-foot or sheepsfoot
rollers can penetrate the Uner. However, this may not be an Issue for
hazardous waste facility liners because most regulations require liners
L ?e:- h1ck or sreater. The second reason stated for this preference 1s
that the native soils used, at these sites often have a high sand content and
are most effectively compacted with smooth vibratory rollers (Kozlckf, Ground
Engineering, Ltd., Regina, Saskatchewan, Canada, personal communication,
i y GV j •
5.2»2.3 Climatic Effects—
The following section 1s a discussion of climatic Influences on clay
liner construction activities and the measures to avoid problems resulting
from climatic stresses.
... 5'2.2.3.1 Precipitation and Desiccation— Precipitation can Interfere
with construction operations by eroding or flooding the site or by
5-75
-------
Ol
I
VI
o>
Source: Photo courtesy of Ground Engineering, Ltd.. Regina, Saskatchewan
Figure 5-16. Pneumatically fed bentonite spreader.
-------
Ul
I
Source: Photo courtesy of Ground Engineering, Ltd., Regina, Saskatchewan
Figure 5-16. Pneumatically fed bentonite spreader(continued).
-------
Figure 5-17. Blending bentonite with soil using a disk harrow.
5-78
-------
U1
I ,
VJ
(O
Source: Photo courtesy of Ground Engineering. Ltd., Regina, Saskatchewan
Figure 5-18. Soil stabilizer mixing bentonite in place.
-------
over-moistening the Uner material. Provisions for protecting a borrow pile
from erosion or overwetting have already been discussed. For Uner Installa-
tion, when construction 1s interrupted at night or by rain, the compacted
11ft 1s usua44y-s~eal -rol 1 ed (rolled smooth with a smooth drum or wheeled
roller). If the site is properly graded, this ensures that water will run to
the lowest point of the site and not puddle or pond on the Uner surface.
Two hazardous waste management companies have used or suggested
Inflatable domes over secure landfills for protection from the elements dur-
ing construction and operation (Figure 5-19). These domes enable construc-
tion activities to proceed despite inclement weather. In addition, for
facilities with continuous operation where wastes are emplaced in the
landfill while other parts of the liner are still being constructed, the dome
prevents rainwater from falling on the site, thus eliminating the need to
collect and treat leachate generated during operation. Upon closure of the
facility, the dome is deflated and moved to another location.
Plastic covers may be used during inactive periods to prevent drying or
wetting of the Uner material. Soil covers are sometimes used to prevent
desiccation and erosion. It 1s important to protect the liner against desic-
cation, especially if high-swelling soils have been used, because desiccation
cracks can seriously Increase the liner's permeability. In certain areas, 6-
to 8-inch dessication cracks can develop in 1 day (Ghassemi et al., 1983).
If desiccation cracks occur, it is necessary to disk and recompact the
portion of the Uner that has been affected. A liner failure documented 1n
Chapter 7 (Site I) illustrates the importance of controlling desiccation.
5.2.2.3.2 Freezing—Liners should not be constructed of frozen soils.
From the standpoint of compaction, frozen soils are difficult to work and the
compactlve effort needed to achieve a specified density and permeability
Increases with decreasing temperature, often to the point that the required
permeability and density cannot be achieved. Thus, 1n colder climates,
liners cannot be properly constructed during the winter months.
Freezing of a liner can cause surface cracking and degradation of the
Uner soil fabric, resulting in increased permeability (Mercuric, Ebasco
Services, Inc., New York, New York, personal communications, 1984). At the
end of a construction season, as well as at night in the fall and spring when
temperatures drop below freezing, the liner should be;protected from
freezing. Liners can be protected from freezing with a blanket of soil or
organic mulch.
5.2.3 Postlnstallatlon Activities
Upon completion, the Uner 1s rolled smooth to seal the surface so that
precipitation and/or leachate can run freely to the leachate collection
sump. The completed Uner 1s surveyed to ensure that thickness, slope, and
surface topography are as required by the design specifications. Seals
around objects penetrating the Uner (e.g., antiseep collars around leak
detection system pipes) should be checked for Integrity. The Uner can be
covered with plastic or a soil cover to prevent desiccation 1f any time will
pass before the next Uner system component (e.g., FML or leachate collection
system) 1s Installed or before the liner 1s covered with waste. This 1s
5-80
-------
Source: Photo courtesy of Waste Management, Inc., Oakbrook, Illinois *
Figure 5-19. Inflatable dome over a hazardous waste landfill.
5-81
-------
especially important for bentonite liners or for liners composed of highly
swelling native soils where desiccation cracking can occur very quickly.
For surface impoundments, rip-rap can be placed on the upper sidewall
slopes to protect them from wave erosion. Experience in dam construction has
shown that in most cases rip-rap is best placed by dumping. A U.S. Army
Corps of Engineers' survey cited by Small (1981) found that dumped rip-rap
failed only 5 percent of the time, whereas the failure rate for hand-placed
rip-rap was 30 percent. The failure rate for concrete pavement used as a
substitute for rip-rap was 36 percent. Proper sizing of dumped rip-rap is
critical to its performance. Rip-rap slopes also must be maintained to
provide reliable liner protection. Erosion-resistant structures for
channeling or loading liquid wastes into surface impoundments also may be
constructed after liner installation.
5.3 QUALITY ASSURANCE AND QUALITY CONTROL
A recent survey of hazardous waste surface impoundment technology
revealed that rigorous construction quality assurance (CQA) and construc-
tion quality control (CQC) are necessary to achieve good site performance
(Ghassemi et al., 1983). Peirce et al . (1986) found a wide variety of
different construction and testing methods in a survey of 1 German and 22
;r4ua!t! J1sP°sa1 facilities. Liner failures at several impoundments were
attributed to various factors including "failure to execute proper quality
assurance and control." The success of surveyed facilities that have per-
formed very well is attributed to many factors including "the use of com-
petent design, construction and inspection contractors, close scrutiny of all
phases of design, construction, and CQA inspection by the owner/operator,
excellent CQA and CQC and recordkeeplng during all phases of the project, and
good communications between all parties involved in establishing the sites"
(Ghassemi et al., 1983). EPA has recognized the Importance of CQA and has
proposed an extensive program for both new and Interim status hazardous waste
facilities (U.S. EPA, 1987). A recently developed Technical Guidance Docu-
ment provides the framework for the CQA program required at hazardous waste
facilities (U.S. EPA, 1986b). These documents should be consulted for
specific applications of the material discussed in the remainder of this
section.
mA ^nHJ1!1 ?auses of Clay I1ner fa11ure that can 4>e avoided with careful
CQA and CQC include:
t Use of materials that do not meet the design specifications
• Inadequate foundation preparation
• Inclusions of roots and other organic matter, large rocks, pockets
of permeable materials, and other foreign objects 1n the Hner
material
• Inadequate moisture control both prior to and after compaction
• Inadequate clod size reduction, mixing, and spreading of Uner
materials
5-82
-------
Emplacement of Inadequate amounts of liner materials (especially
Important with bentonite/soil liners)
• Failure to follow installation procedures specified in the design
• Use of improper or inadequate construction equipment
• Specification or application of inadequate compactive effort
t Failure to tie lifts together properly
• Inadequate control of soil moisture content and density during
compaction and poor maintenance after construction.
This section specifically addresses quality assurance and quality
control for the construction of clay liners for hazardous waste landfills,
waste piles, and surface Impoundments. It addresses liners constructed of
both recompacted soil and bentonlte/soil admixtures and includes QA
activities that are necessary to ensure that the liner material 1s as
specified and that Installation procedures will result 1n a liner that will
perform as specified in the facility design. This text is a compilation of
information on current CQA and CQC practices obtained 1n the course of this
study through Interviews with design engineers and through literature
reviews. The section on current practices (Section 5.3.4.4) tabulates much
of the Information obtained during this effort.
Proper Installation of all of the components of a hazardous waste
storage or disposal unit (I.e., clay liners, synthetic liners, leachate
collecting system, dikes, and cover systems) 1s necessary to ensure the
specified performance of the clay liner and the total containment system. As
with clay liners, an adequate CQA and CQC program 1s necessary during
installation of these components to ensure that they will perform as
specified; however, a discussion of CQA and CQC for these components 1s
beyond the scope of this document.
5.3.1 Key Terms
The following concepts and terms are used throughout this section.
8
• Construction Quality Management—The process whereby scientific and
engineering principles and practices are used to ensure that a
land-based hazardous waste facility 1s constructed 1n conformance
with Its design. The emphasis on construction quality management
must begin as early as possible in the design of the facility, must
be stressed throughout the actual construction of the facility, and
ceases upon closure of the facility. Managing construction quality
is the responsibility of the permit applicant, the construction
contractor, and the design engineering firm. It consists of two
components, CQC and CQA.
• CQA--A planned series of overview activities, the purpose of which
1s to provide assurance that CQC 1s being Implemented effectively.
The system Involves a continuing evaluation of the adequacy and
effectiveness of CQC, the Inspections performed, the data
collected, and the Interpretation of the data made 1n response to
5-83
-------
the CQC activities. For hazardous waste land disposal facilities,
this series of overview activities may involve verifications,
audits, and evaluations of the factors that affect the Installa-
tion, Inspection, and performance of a hazardous waste land
disposal facility to ensure that the facility meets the design
specifications.
CQC—Planned and unplanned inspection activities, the purpose
of which is to help provide the level of construction quality
that will result in a facility that meets the design specifica-
tions. The overall system involves integrating several related
factors including: (1) proper selection of specified materials,
(2) Installation to meet the full intent of the specification,
and (3) inspection to determine whether the resulting product is
according to the specification.
CQA Plan—The CQA plan that is discussed in this chapter is a
written approach that may be followed to attain and maintain
consistent high quality in the construction of hazardous waste
storage and disposal facilities. The purpose of a CQA plan 1s
to ensure that a completed facility meets or exceeds all design
criteria. The CQA plan is tailored to the specific facility to
be constructed and documents the permit applicant's commitment
to CQA.
The CQA plan 1s prepared as part of the facility design
activities. It is usually prepared by the design engineers but in
some cases may be prepared by an independent third party respon-
sible for CQA for the facility. The CQA plan Indicates what tests
and observations will be made during construction to ensure that
design criteria are met, as well as test frequency and test spacing
requirements for CQC. Acceptance/rejection criteria for specific
tests are specified in the plan and reflect the precision of the
specific test methods used and the specified values, design
tolerances, and expected field variability of the tested param-
eters. Also specified in the CQA plan are corrective actions to be
implemented if some part of the work is substandard and conse-
quently rejected.
While the overall content of the CQA plan will depend on the site-
specific nature of the proposed hazardous waste land disposal facility, as a
minimum, several specific elements should be included in the plan. These
elements are summarized briefly below.
• Responsibility and Authority—The responsibility and authority of
all organizations involved in permitting, designing, and construct-
ing the hazardous waste land disposal facility should be discussed
fully in the CQA plan.
• CQA and CQC Personnel Qualifications—The CQA officer, CQC
inspector(s), and all other CQA and CQC personnel should possess
the training and experience necessary to fulfill their identified
responsibilities, and their qualifications should be presented in
the CQA plan.
5-84
-------
• CQC Activities (Observations and Tests)—The specific observations
and tests that will be used to control or monitor the Installation
of--the hazardous waste land disposal facility components should be
sufflrasFFfzed In the CQA plan.
• Construction Quality Evaluation—The sampling activities, sample
size, sample location, frequency of testing, acceptance and rejec-
tion criteria, and plans for implementing corrective measures, all
•thVcQA Ian 1" the project 5Pec1f1cat1ons, should be presented in
• Documentation—The CQA plan should Include a discussion of the
reporting requirements for the project. This should include such
items as daily.summary reports, observation and testing data
sheets, problem Identification and corrective measures reports
block evaluation reports, design acceptance reports, and final
documentation. Provisions for the final storage of all records
should also be discussed 1n the CQA plan.
m , rn? I?ll0?1n9JS a m2re deta1led 11st of important Items to be Included
IIrS.?2\5iX I°/ddrKs the above e1ements Properly. These Items have bee
Plan" of SPn8o !"°2 m%prSvan J"d !ffect1ve "Contractor Quality Control
Plan of ER 1180-1-6 (U.S. Department of the Army, 1978). They are presented
here only as an example and should not be considered an obligatory or
complete or exclusive outline. These items are:
t A planned QA organization
• An education plan to ensure that the workmen, construction manage-
S?nfh/nrn ^EeCt2rLare aware of the Var1ous ^^^ requirements
of the project and the reasoning behind the requirements
• Proposed methods for performing CQC Inspections, both for construc-
tion process control and for acceptance sampling and testing (qual-
ity evaluation); this Includes Inspections of subcontractors' work
* Name and qualifications of each Individual assigned a CQA or CQC
function; method of establishing and verifying personnel qualifi
tions to perform specific tests and/or tasks «""
• Discussion of how CQC Inspections will be performed:
by Inspectors employed by the permit holder designated by
name, their qualifications, and the specific tests and
observations to be made
by subcontract Inspectors designated by name, their qualifica
tions, and the tests and observations to be made subject to
approval of the regulatory agency
5-85
-------
t The test and/or observation method to be used for each specifica-
tion quality requirement; whether 1t 1s a standard or alternative
mefebSiL the reference standard or literature reference for each;
the method for establishing size of the unit of construction for
acceptance testing; the method for establishing sample sizes and
sampling Locations; acceptance/rejection criteria for each test;
and the method for dealing with outlying observations
• Location, availability, applicability, and calibration of test
facilities and equipment
• Procedures for advance notice and coordination of special
inspections when and where required
t Procedures for reviewing all drawings, samples, certificates, or
other documents for compliance with permit requirements and for
certification of their acceptability to the regulatory agency;
qualifications required of the!Individual performing the reviews
0 Procedures for reviewing inspection test results and observation
records; qualifications required of individuals performing the
reviews
t Procedures for observing and testing fresh exposures of the site
media (soil profile) and for comparing the results with evaluations
made during site characterization studies; qualifications required
of individuals performing the observations and tests
• Action to be taken by inspectors and reviewers (re: 10, 11, and 12
above) when deficiencies are identified and/or reported, including
who is to be notified and in what manner
• Reporting procedures, providing for submittal and/or storage of all
test and observation reports, at specified Intervals, and reporting
of all actions taken under Item 13 above; report formats to be used
• Definition
-------
Successful CQA and CQC requires clarity 1n written and oral communica-
tlons among all the parties, especially 1n defining key terms and 1n
delineating ajsas. and lines of authority and responsibility 1n scope of work
and other contractual arrangements. For example, "inspector," "Inspection,"
"certification," and "verification" are terms that are used ubiquitously and
frequently without regard to precise definition. The definitions of these
terms should be understood and accepted by all parties Involved in the clay
liner project and clearly documented 1n a CQA plan.
5.3.2.1 Responsibility and Authority—
The overall responsibility of the CQA and CQC personnel 1s to execute
activities specified under the CQA plan. As a minimum, CQA personnel
includes a CQA officer and a CQC inspector. Specific responsibilities and
authority of each of these persons are defined in the CQA plan and associated
contractual arrangements with the owner. For the CQA officer, specific
responsibilities and areas of authority may include:
• Reviewing and fully understanding all aspects of the specified
landfill design and proposed construction techniques
« Serving as the owner's or design engineer's liaison with the
contractor in Interpreting and clarifying contract documents
• Providing CQA reports to the owner on the results of Inspections
and testing
• Advising the owner or design engineer of work that the CQA officer-
believes should be corrected, rejected, or uncovered for Inspection
and of work that may require special testing, inspection, or
approval
• Reviewing inspection and test results and rejecting defective work
when authorized to do so, by the owner or design engineer
• Directing the CQC inspector 1n performing site Inspections and
testing
• Stopping construction site activities 1n cases where deviations
from design plans and specifications are defected and Implementing
corrective actions.
For the CQC Inspector, specific responsibilities may Include:
t Conducting onslte observations and tests of the work 1n progress to
assess compliance by the contractor with the plans, specifications,
and construction-related contractual provisions for the project
t Reporting to the CQA officer results of all Inspections Including
work that does not meet the specifications or falls to meet
contract requirements
• Monitoring reviews and tests conducted by the contractor as
required by the specifications and contract
5-87
-------
• Verifying that tests, equipment» and system startups are conducted
by .qualified personnel and proceed according to standardized
procedures defined by contract documents.
5.3.2.2 Qualifications of CQA Personnel-
Inspectors and CQA officers should be trained 1n the proper use of all
test methods and equipment. They should!have the ability to calibrate
equipment, administer the required tests, record and interpret data, and make
pertinent observations. The training that 1s required to obtain these skills
may come from the classroom or through field experience. However, emphasis
is generally placed on first-hand (field) testing experience.
In addition to the above requirements, CQA officers should be
registered professionals who thoroughly understand the theory and application
of all physical and observational tests, the overall site design, the proper
use of the various types of construction equipment, and project management
and who have sufficient practical experience 1n landfill construction.
5.3.3 Observations and Tests
This section describes the observations and tests that should be
specified 1n a CQA plan for clay liner construction. The following section
1s divided according to the CQA and CQC activities that will take place
during preconstructlon, construction, and postconstructlon periods of clay
liner Installation activities. All ASTM test methods referenced 1n this
chapter may be found in ASTM (1985).
5.3.3.1 Reconstruction—
The first activity under preconstructlon CQA 1s to review the design
drawings and construction specifications for the clay liner that 1s to be
Installed with emphasis on CQA and CQC. The design drawings and construction
specifications need to be clear and understandable from the standpoint of
both the onslte CQC Inspectors and the contractor. If the design is deemed
Inadequate or unclear by the CQA officer, 1t should be returned to the design
engineer for clarification and/or modification.
Prior to construction, the CQA officer must also assess the capabilities
of the construction contractor's personnel so that he can determine the type
and amount of training, Instruction, and supervision that will be needed
during construction operations. The contractor's prior performance 1n
general earthwork activities, experience In construction of hazardous waste
facilities, and experience 1n working the specific type of soil and equipment
to be used 1n constructing the facility in question need to be addressed 1n
this assessment.
A preconstructlon training plan should be Included 1n the CQA plan, as
stated by the U.S. Department of the Army's Construction Control Manual
(1977): —
Preconstructlon Instructions and training should be given
to field Inspection personnel to acquaint them with design
concepts and to provide them with a clear understanding of
expected conditions, methods of construction, and the scope
of plans and specifications. This may be done by training
sessions, preferably with design personnel present, using a
5-88
-------
manual of written Instructions prepared especially for field
personnel, to discuss engineering considerations Involved and
to explain control procedures and required results.
If significant quantities of the soil liner material are to be stored at
the construction site, the CQA officer should make provisions to ensure that
storage facilities are adequate to prevent alteration of the liner material.
For natural clay soils, this could include seal-rolling, grading, or covering
the storage pile to encourage runoff and to reduce erosion. If bentonite/
soil admixtures are to be used as liner material, 1t 1s critical that benton-
ite material stored onsite 1s sheltered from precipitation because this
material cannot be worked except when in a dry state.
The CQA plan must provide assurance that any liner material brought onto
the site is as specified. Soil material screening begins as a preconstruc-
tion activity and continues during construction as long as material is being
brought onsite or excavated from the site. This activity can be accomplished
in several ways, depending on the source of the liner material and site con-
ditions. If the Uner material 1s obtained onsite, the inspection can be
accomplished as 1t is placed in the borrow pile for storage. If the
excavated soils are heterogeneous, it may be necessary to segregate the soil
material as 1t 1s excavated, with suitable soil placed 1n a borrow pile for
future use and son that does not meet specifications discarded. The CQC
inspector observes the segregation operations carefully to ensure that only
suitable material 1s retained for Uner construction.
Similarly, 1f the Uner son 1s obtained from a nearby borrow area,
the son material may be Inspected at the borrow site or as the material
arrives at the construction site. Borrow site Inspection 1s more desirable,
especially 1f the soil is heterogeneous, because this will ensure that only
suitable material 1s transported to the site, saving transportation costs.
Subsurface characterization of the borrow site may expedite this effort for
heterogeneous borrow sources. The U.S. Department of the Navy (1982) recom-
mends that during Initial exploration of the borrow pit area, borings or test
pits be made on a 200-foot grid. If variable conditions are found during
this Initial exploration, Intermediate borings or test pits should be made.
Borrow pit exploration should produce the following Information:
• A reasonably accurate subsurface profile do>n to the anticipated
excavation depth
• Engineering properties of each material considered for use
• Approximate volume of each material considered for use
• Water level
• Presence of sands, gypsum, or other undesirable materials
• Extent of organic or contaminated soils, 1f encountered.
For extremely heterogeneous borrow areas, 1t may be necessary for the
Inspector to guide the excavating equipment to avoid substandard soil
material.
5-89
-------
Inspection of the soil can be largely visual; however, CQC personnel
conducting thjs Inspection must be experienced with visual-manual soil
classification techniques (ASTM D 2488). Changes 1n color or texture may
Indicate a charrge 1n soil type or soil moisture content. The soil also may
be inspected for roots, stumps, and large rocks. In addition, as a check of
visual observations, samples of the soil usually are taken and tested to
ensure that the soil's index properties are within the range stated in the
specifications; tests of moisture/density characteristics (ASTM D 698) also
should be conducted to ensure that these relationships do not change. The
number of tests to be conducted depends on site-specific conditions (I.e.,
soil type and heterogeneity) and the experience of the CQC personnel.
Usually a minimum number of tests per cubic yard of material is specified,
with additional tests required by the inspector 1f visual observations
suggest a change 1n soil type.
Soil index properties are simplified tests that provide indirect
Information about the engineering properties of soils beyond what can be
gained from visual observations. Although the correlation between index
properties and engineering properties is not perfect, it is generally
adequate for CQC purposes. Index property tests commonly used to screen
soils are described below.
Atterberg limits include the liquid limit and the plastic limit (ASTM
D 4318). These tests are commonly used along with grain size distribution
for monitoring changes in soil type. A significant change in Atterberg
limits usually reflects a change 1n important engineering properties, such as
the relationship among moisture content, density, compactlve effort, and
permeability.
Grain size analysis 1s another Important screening test for changes In
soil composition. The percentage of clay-size particles and the overall
particle size distribution of a soil affects its engineering properties,
especially permeability and strength. Rough estimates of grain size may be
obtained through manual estimates (ASTM D 2488) and may be sufficient for
screening. A 200-mesh sieve may be used to separate coarse (sand and gravel)
and fine (silt and clay) particles. More detailed grain size distributions
may be obtained by sieving the coarse fraction and by using several settling
methods (hydrometer, decantatlon, or pipette) for the fine fraction
(ASTM D 422). Cases where samples of Incoming or Installed liner materials
contain Inadequate quantities of the necessary soil particle size may be
referred to the CQA officer for evaluation. Moisture content and consistency
tests are also needed for screening soils. Again, 1t 1s Important to monitor
carefully for soil type changes as long as liner material is being placed. A
change in soil type requires the compaction of another test fill as described
1n the following text.
When bentonite/soll liners are specified, Incoming bentonite should be
Inspected to ensure that Its quality 1s as specified. For all bentonite
shipments, certification of compliance with material specifications should be
obtained from the manufacturer or supplier. In addition, the quality of the
arriving bentonite should be tested frequently for dry fineness, pH, and
viscosity and fluid loss of a slurry made from the bentonite. Dry fineness
1s the percentage passing a 200-mesh sieve. It 1s necessary to control dry
fineness to ensure proper mixing of the bentonite. Slurry viscosity, slurry
fluid loss, and pH are standard tests specified by the American Petroleum
5-90
-------
m , *r -• , '"I - *>
Institute (API, 1982). These tests are necessary to ensure the quality of
the bentonitejn terms of its swelling potential. Electric pH meters should
be used for pJLmeasurements; pH papers are usually not reliable (Xanthakos,
1979). If bentonite additives are specified, the manufacturer's certificate
contained with each shipment should state compliance with the specified
characteristics.. More information on testing bentonite quality may be found
1n Xanthakos (1979).
5.3.3.1.1 Test Fill—An important preconstruction CQA activity is to
determine the suitability of the equipment and methodology to be used to
compact the liner. To accomplish this, most design engineers recommend that
a representative test fill of liner material be compacted with the designated
equipment to see 1f the specified density/moisture content/permeability
relationships determined in the laboratory can be achieved 1n the field with
the compaction equipment to be used and at the specified lift thickness.
Test fill dimensions must be sufficient to accommodate the compaction
equipment. Several lifts are usually compacted 1n the test fill to check the
methodology to be used to tie lifts together. The test fill also is used to
determine the number of equipment passes (or amount of compactlve effort)
needed to achieve the specified permeability and to determine the ability of
mixing equipment to break up large clods of uncompacted liner soil.
Field permeability tests can be conducted on the compacted test fill
material. Field compactlve effort is different from the compactlve effort
applied in the laboratory. Although densities may be the same for different
types of compactlve efforts, the fabric and the permeability of son
compacted by different methods can differ significantly (Mitchell, 1976).
These permeability tests, therefore, are necessary to ensure that the
compactlve effort that 1s applied 1n the field will result in the same or
lower permeability than was demonstrated 1n the laboratory tests of the liner
material. Additionally, the test fill is useful in establishing a relation-
ship between field permeability and laboratory permeability measurements on
undisturbed samples of compacted liner material, which can vary by as much as
3 orders of magnitude (Day and Daniel, 1985). The long time necessary for
field permeability measurements often limits their use during construction
operations because of scheduling problems. However, field permeability
measurements can be scheduled conveniently during test fill compaction.
In addition to the number of passes necessary to*achieve the specified
permeability, equipment type, size, and compatibility with the soil type are
evaluated and recorded during test fill compaction. Equipment items to be
checked and recorded include:
• For sheepsfoot rollers—drum diameter and length, empty and
ballasted weight, length and face area of feet, and the yoking
arrangement
• For rubber-tired rollers—the tire inflation pressure, spacing of
tires, and empty and ballasted wheel loads
t For vibratory rollers—the static weight, imparted dynamic force,
operating frequency of vibration, and the drum diameter and length
(U.S. Department of the Army, 1977).
This information is necessary to estimate the compactlve effort and
compactlve force applied in the field.
5-91
-------
Compacting the test fm 1s also valuable 1n that 1t gives the CQA
officer, CQC Inspector, and construction personnel some experience with the
behavior of the specific soils, equipment, and methodology to be used at the
site during compaction. This 1s a very useful base of knowledge to apply
when observing compaction of the actual containment facility liner. The test
fin compaction thus serves to calibrate ;the Inspectors' and contractors'
observations to the conditions that will be encountered during the compaction
of the liner.
The successful use of density and moisture content measurements for CQC
of clay liner compaction depends upon the relationship established among
density, moisture content, compactive effort, and permeability for the
specific soil and for the specific compaction equipment and methodology used
during test fill compaction. If a change in son characteristics or a change
of compaction equipment or methodology occurs during liner construction,
another test fill should be compacted with the new soil, equipment, and/or
methodology because the original relationships may no longer apply. If this
is not done, there is no assurance that CQC using density/moisture
content/compactlve effort measurements will result 1n a Uner with the
specified permeability.
5.3.3.2 Construction—
An Important CQC activity during clay liner installation 1s observation
of the construction process, Including personnel performance, by the
Inspectors. Observations by an experienced inspector, coupled with a soundly
developed and implemented CQC testing plan, will ensure that the liner is
Installed as specified, that any potential problems are identified in a
timely manner, and that proper corrective actions are implemented.
5.3.3.2.1 Foundation Base Preparation—COG for excavation and construc-
tion of foundations 1s not fully addressed 1n this document. Standard
methods for controlling the quality of foundation preparations and earthen
embankments may be found elsewhere (e.g., Splgolon and Kelley, 1984; U.S.
Department of Interior, 1974; U.S. Department of the Army, 1977). The
foundation base must be adequately prepared before the clay liner 1s
constructed. The natural foundation should provide satisfactory contact with
the overlying compacted Uner, minimize differential settlements and thereby
prevent cracking of the liner, and provide an additional barrier to leachate
migration from the facility. To ensure that these goals are met, observa-
tions during the construction of foundations should include the following
(U.S. Department of the Army, 1977):
• Observations of stripping and excavation to ensure that all soft,
organic, and otherwise undesirable materials are removed. Proof-
rolling with heavy equipment can be used to detect soft areas
likely to cause settlement. Consistency of the foundation soil may
be checked with a hand penetrometer, field vane shear test, or
similar device.
• Inspection of soil and rock surfaces for adequate filling of rock
joints, clay fractures, or depressions and removal and filling of
sand seams.
5-92
-------
• Inspection of the dep|h and slope of the excavation to ensure thai:
1t meets design requirements. r ?
• Observations to ensure proper placement of any recessed areas for
collection or detection pipes and sumps.
• Tests and observations to ensure the quality of compacted fill.
Visual observation of the construction process is the major means of
ensuring that the foundation is constructed as designed. Some Instrument
surveying may be necessary to ensure that facility dimensions, side slopes,
and bottom slopes are as specified. Visual-manual soil identification
techniques (ASTM D 2488) and index property tests (ASTM D 422; D 4318) may
be used to monitor foundation soil composition.
Soil consistency may be checked with a cone penetrometer (ASTM D 3411),,
This method 1s widely used to determine the consistency of cohesive soils for
classification and 1s more accurate than the visual-manual method. It is
less accurate than the laboratory method (ASTM D 2166), but can be used to
give broad consistency classifications for cohesive soils. Shear strength
of foundation soils can be checked 1n the field with a field vane shear
device (ASTM D 2573) or a torvane device (Lanz, 1968). The ASTM D 2573
method is standardized and, although it Is less accurate, can be correlated
with the standard laboratory method for unconfirmed compresslve strength
(ASTM D 2166). Although all of these field expedient methods give only
approximate values, they are usually sufficient for construction. Compaction
is controlled as described 1n Section 3.3.3.2.3.
Further Information on quality control of foundations may be found 1n
Splgolon and Kelley (1984), U.S. Department of the Interior (1974), and U.S.
Department of the Army (1977).
5.3.3.2.2 Liner Lift Placement—Liner 11ft placement Includes the
operation of spreading the Uner material(s) over the floor of the facility;
breaking and homogenizing clods of soil; and, for blended clay liners and
admixed liners, uniformly blending the mixed materials.
During placement of soil materials, the son 1s spread uniformly as
specified. The loose 11ft thickness of the son should be measured system-
atically over the entire site, with a marked staff or shovel blade, and
survey levels should be made every few lifts for verification and documenta-
tion of Uner thickness. Following spreading, the Uner material 1s disked
or tilled to break up large soil aggregates and to homogenize the material.
All large clods of liner material should be reduced 1n size as much as pos-
sible to facilitate moisture penetration and to ensure uniform compaction
through the 11ft. Opinions differ on the acceptable maximum clod size; 1n
a series of Interviews with design engineers, recommendations ranged from
1 Inch to no greater than the 11ft thickness (see Table 5-11 1n Section
5.3.3.4). Close observation by the onslte CQC Inspector 1s critical to
ensure that this 1s properly accomplished.
If bentonlte additives are to be admixed with the natural soil, the
proper percentage of the additive must be controlled. For spreading and
1n-place mixing, this 1s accomplished by visual observations of the additive
5-93
-------
as it Is spread over the site by hand or by a suitable spreading device. In
addition, a spreader can be roughly calibrated by using 1t to spread the
additive over^a=£!ast1c sheet or pan of given dimensions and then weighing
the collected material. Visual observations also should be made to ensure
that additives are uniformly distributed in each liner lift. To ensure the
proper bentonite percentage 1n admixtures, the methylene blue test (see
Appendix A) can be carried out on representative samples of the admixed
material. The soil mixture's liquid limit or plasticity Index has also been
used to ensure proper bentonite percentage.
If bentonite additives are to be blended with the native soil prior to
emplacement, proper bentonite percentage can be ensured by using the
methylene blue test on samples of the mixed material. CQC personnel also can
inspect the mixing operations to ensure that the proper amounts of bentonite
and soil are placed Into the mixing device and that mixing time and force are
sufficient.
5.3.3.2.3 Moisture Control—Moisture content of the liner soil should
be measured and controlled both before and after 11ft placement to ensure
that the soil moisture content 1s as specified and for activating bentonite
additives. Nuclear probes (ASTM D317-78) and/or manual moisture content
measurements are generally used to control moisture 1n the field (see
Chapter 3). The following description of manual moisture estimation is
adapted from Johnson et al. (1983).
In order for quality control personnel to satisfactorily estimate fill
water contents, they must become thoroughly familiar with the fill
material prior to the start of fill operations. Preferably, they should
spend some time in the field laboratory, performing several compaction
tests to become familiar with the differences 1n appearance and behavior
of the various fill materials, to recognize when they are too dry or too
wet, as well as when they are at the specified water content.
Inspection personnel should also run several Atterberg limits on
fine-grained soils so they can compare the appearance and feel of the
soils when they are at the plastic limit with that at proper water
content for compaction.
Trained personnel should then be able to pick up a handful of soil and
make a reasonable estimate of the relation of Its water content to Its
optimum water content by feel and appearance (experienced technicians
often can estimate deviation of water content from optimum within +1
percent). Occasional moisture tests should be made to confirm these
manual estimates.
A few reference bags of liner material at known moisture contents stored at
the site can be very helpful for calibration of manual moisture content
estimates.
Many design engineers recommend that oven moisture content measurements
by the appropriate test method (ASTM D2216-80) be taken at Intervals to
ensure that nuclear measurements and manual estimates are accurate. Proper-
moisture content is achieved and maintained with sprinkling devices (1f too
dry) or by a combination of mechanical agitation, aeration, and solar drying
(1f too wet).
5-94
-------
If the liner 1s to be left exposed between the Installation of succes-
sive lifts or^after completion, the QC personnel make sure that 1t 1s
protected from moisture content^.change by ensuring that seal-rolling 1s
uniformly performed over the site and that cover material, 1f used, 1s
properly emplaced. If desiccation does occur, the QC inspector reports this
to the QA officer, who notes this and specifies remedial actions, such as
disking and recompactlon of the affected portion of the liner.
5.3.3.2.4 Compaction—Son selection based on laboratory compaction and
permeability tests is made on samples of the liner material during design of
the facility. A relationship among moisture contents, densities, and
permeabilities is established based on test results. Design specifications
usually require achievement of a minimum percentage of the maximum density
(Proctor or modified Proctor) at a specified range of water contents, based
on these results. The specified density/water content corresponds to the
density/water content at which the specified permeability can be achieved as
established by the laboratory and test fill compaction tests. This density/
water content 1s then tested during quality control of clay Uner
installation.
Additionally, during compaction of each lift, compactlve effort and
uniformity of compaction are observed and recorded. Compactlve effort 1s
estimated by the number of passes of equipment of a known size and weight.
It is important to make coverage uniform, especially at fill edges and 1n
equipment turnaround areas and at the top and bottom of slopes (Spigolon and
Kelley, 1984), In addition, permeability tests often are conducted
periodically on the compacted liner material as a check on the moisture/
density/permeability relationship. It 1s generally agreed that all of the
above measurements are necessary to ensure that the specified permeability
1s being achieved in the field. Density measurement should never be used
alone for quality control of clay Uner Installation.
The relationship among moisture content, density, compactlve effort, and
permeability is unique for a specific soil or soil mixture and specific type
of equipment. Laboratory compaction tests to establish density/moisture
curves are determined regularly on field samples of the clay Uner to
determine changes 1n optimum water contents. If these tests or field
Inspection of the Incoming Uner materials Indicate a significant change,
laboratory permeability tests and a test fill compact-Ion must be conducted to
establish the density, moisture, and permeability relationship for the new
soil. Otherwise, attainment of the moisture/density relationship that was
specified for the original soil may result 1n an unacceptable permeability
with the new soil. Similarly, 1f different compaction equipment or
methodology 1s used, another test fill should be compacted with the new
equipment because the type of compactlve effort applied affects the final
permeability.
The design engineers Interviewed during this Investigation stressed the
Importance of visual and manual observations by a qualified Inspector for
Uner compaction quality control. Some consider specific tests only a backup
documentation to observations by the qualified CQC Inspector. All
professionals interviewed agreed that observation of the construction process
1s the primary and most effective approach to CQC. Testing 1s secondary;
beyond the minimum test frequency and spacing, visual observations are used
5-95
-------
to Identify problem areas and to call for more Intensive testing to document
and delineate_any substandard liner areas. Remedial actions (e.g., removal
and reconstruction) are then ordered for the substandard areas so
delineated. All engineers stressed the Importance of having a qualified
Inspector on the site at all times during construction.
An experienced observer can determine how compaction 1s proceeding
(e.g., moisture content and density) by observing how the equipment moves
across the area to be compacted, how the soil contacts or sticks to the
compaction equipment, how the soil heaves during compaction, how deep the
compaction equipment sinks into the soil, and how the roller walks out of the
soil (sheepsfoot rollers only) (Johnson et al., 1983). In addition, it is
Important to observe uniformity of coverage by compaction equipment,
especially at fill edges, in equipment turnaround areas, and at the top and
bottom of slopes (Splgolon and Kelley, 1984).
5.3.3.2.5 Specific Tests—Testing and sampling of the liner are
necessary to ensure compliance with the design requirements and to document
the as-built conditions of the clay liner. Specific tests to ensure that
compaction results 1n the specified liner permeability Include field density
tests (nuclear, sand-cone, and others), field moisture content measurements,
laboratory compaction tests, and both field and laboratory permeability
tests. The methods and QC measures for conducting these tests may be found
1n several documents (MSHTO, 1978; ASTM, 1985; U.S. Department of the Army,
1970; and U.S. Department of the Interior, 1974) and are briefly discussed in
Chapter 3 and Appendix A. The main tools used for controlling the quality of
compaction are field density and moisture content measurements, with supple-
mentary laboratory compaction tests providing a means of monitoring changes
1n soil material. A laboratory compaction test should be conducted for every
10 to 20 field density/moisture determinations, depending on soil variability
(U.S. Department of the Navy, 1982). Nuclear probes may be used to measure
field density and moisture content, but these must be calibrated for each
soil that 1s to be tested. In addition, if nuclear devices are used, other
field density and moisture content measurements, such as sand cones and oven
drying, should be made periodically to confirm nuclear results. Again, 1t
1s necessary to measure density, moisture, and compactlve effort 1n the
field to ensure that .the required permeability 1s achieved during clay liner
compaction.
0
In addition to density and moisture measurements and estimates of
compactlve effort, permeability tests should be made regularly to confirm
that the measured moisture/density levels correspond to those required for
the specified permeabilities. Shelby tube or block samples may be taken for
laboratory permeability tests (ASTM D 1587-74; ASTM, 1985), or field
permeability tests may be performed. Laboratory permeability tests are
easy to conduct, do not consume valuable construction time, and are quicker
than some field permeability tests. However, they can underestimate field
permeability by as much as 3 orders of magnitude (Day and Daniel, 1985).
Field permeability tests more accurately represent actual permeabilities, but
some can take many days to complete, seriously Interrupting construction
activities. Field permeability tests may be conducted on test fills prior to
construction so that construction activities will not be Interrupted. (For
further discussion of field versus laboratory permeability measurements and
for a discussion of test methods, see Daniel, 1981; Olson and Daniel, 1981;
5-96
-------
Rogowski and Richie, 1984; and Daniel, 1987; also see Chapter 3 of this
document.)
Several -d^Kjn engineers reeooimended that moisture/density measurements
and Shelby tube samples for laboratory permeability tests be obtained from
the 11ft underlying the lift that has just been compacted. These engineers
believe that during compaction of a lift, significant compactlve effort is
being applied to the underlying lift. A more accurate representation of the
degree of compaction across the liner is achieved by testing the underlying
1 1 T t •
Following Shelby tube sampling, nuclear density measurements, or field
permeability testing, the resulting hole is filled with liner material and
hand tamped or is grouted, with bentonlte. Over excavation of the hole with a
shovel to slope the sides of the hole prior to backfilling may further
facilitate sealing. Test locations should be staggered from lift to 11ft so
that the testing or sampling holes. do not line up. This also gives better
test coverage of the liner.
Index property tests (e.g., grain size, clay content, and Atterberg
limits) are used to evaluate liner soils prior to emplacement in order to
monitor changes in soil type. These test methods are well established and
are described 1n Chapter 3 and in Appendix A.
Minimum test frequency and test spacing should be specified for all
tests in the test plan. Test spacing practices and frequency are discussed
in Section 5.3.3.4 of this report.
5.3.3.3 Completion Tests--
Upon completion of the liner, CQA personnel should check that it is
rolled smooth to seal the surface so that precipitation and/or leachate can
run freely to the leachate collection sump. The completed liner should be
surveyed to ensure that thickness, slope, and surface topography are as
required by the design specifications. Seals around objects penetrating the
Uner (e.g., leak detection system stand pipes) also should be checked for
integrity.
Field permeability tests should be conducted on the completed Uner as a
final QA check. It appears that field measurements o£ permeability (e.g.,
with sealed double ring inflltrometers) are preferable to laboratory measure-
ments because they subject a larger portion of the Uner to permeability
testing. Recent work has shown that field permeability measurements yield an
average hydraulic conductivity close to a liner's actual hydraulic conductiv-
ity. Laboratory tests, even on undisturbed samples, can give a hydraulic
conductivity 19000 times less than the actual value, measured by collecting
seepage through the Uner (Day and Daniel, 1985). Several field permeameters
can be set up over the site, or 1f the site 1s not too large or 1s a surface
impoundment the facility can be filled with water and seepage from the site
can be measured after accounting for evaporation. The latter method, when
feasible, ensures that the entire site will function according to specifica-
tions once 1t is filled, assuming that no waste/ liner compatibility problems
occur.
5-97
-------
If the completed Uner 1s to be left exposed prior to Installation of
the overlying, facility components, CQC Inspectors should ensure that the
Uner 1s covered adequately with soil or plastic sheeting to prevent
desiccation aiRTwind erosion or that 1t 1s maintained at suitable moisture
content by managed water application.
5.3.3.4 Current Sampling Program Design Practices--
Current CQC practice relies on the Inspector's observations and judgment
for accepting most of the work, with actual sampling and testing of the
material serving only to document compliance and to help guide the Inspec-
tor s judgment. Table 5-11 is a summary of information on current CQC
practices obtained during Interviews with design engineers active 1n desiqn-
ing and controlling the construction quality of clay liners. Also Included
in this table is information from earthwork construction manuals of the U.S.
Department of the Interior and the U.S. Department of the Navy. Tables 5-12
and 5-13 11st recommendations for construction documentation recently
published by the Wisconsin Department of Natural Resources personnel.
Statistical sampling methods for earthwork quality control (geostatls-
tics) have been developed by the U.S. Department of the Interior and the "
U.S. Army Corps of Engineers (U.S. Department of the Interior, 1974; U.S.
Department of the Army, 1977; Davis, 1966; Turnbull et al., 1966;
Wlllenbrpck, 1976). These documents Include statistical methodology to
eft??1lfh de9re? of confidence for a testing program. Other discussions of
statistical methods may be found 1n Wlnterkorn and Fang (1975), Sellg (19821
Wahls et al. (1968), Jorgenson (1971), and Kotzlas and Stamatapoulous
(1975) «
n™ £l9ures ?720 ^JL5"?1 11lustrate a simple, concise method of documenta-
tion of clay Uner CQC statistics developed by Soil Testing Engineers of
Baton Rouge, Louisiana. Figure 5-20 Illustrates the sampling locations for
f?nne^%Sf a ^""/U^ ]?nf 111 cfil ' • "«"« 5-21 is a graphic presenta-
1 ^°!,the mo1sture/dens1ty/permeab1l1ty data and clearly Illustrates the
statistical analyses performed on these data. These figures are presented as
examples of good CQC documentation and statistical analysis.
5.3.4 Documentation
,.J dePends heavily on recogn1«1ng all construction
that should be Inspected and assigning responsibilities to CQA and
CQC personnel for Inspecting each activity. This 1s most effectively
accomplished by documenting CQA and CQC activities and should be addressed as
the fifth element of the CQA plan. CQC personnel will be reminded of the
factors to be Inspected and will note by signing required descriptive
remarks, data sheets, and checklists that , the Inspection activity has been
accomplished.
5.3.4.1 Daily Recordkeeplng—
Standard dally reporting procedures should Include preparation of a
summary report with data sheets for supporting observation and testing and,
when appropriate, for problem identification and corrective measures.
.. 5.3.4.1.1 Daily Summary Report— A summary report, or project Inspection
diary, should be prepared dally by the CQA officer. This report provides the
5-98
-------
TABLE 5-11. CURRENT QA PRACTICES FOR CLAY LINEK CONSTRUCTION
CO
to
Source of
information3
A
B
C
D
E
F
G
H
Minimum frequency of
in-place moisture or
density tests
1 per 1,000 yd3 of fill
Moisture tests only (used
with permeability tests
and compact! ve effort to
control compaction)
Frequency per unit volume
of fill based on experience
1 per 2,500 ft2 of lift
Frequency, per unit volume
of fill, based on experience
and liner material
homogenei ty
Statistical approach
9 per acre of lift
1 per 2,000 ft2 or at least
1 per lift
Permeability testing
Occasional lab test
Trlaxial and some field tests
Lab tests
Triaxlal--! per 3 acre feet
with minimum of 1 per acre
Modified trlaxial, fixed-wall,
and consolidation lab tests—
per unit volume of fill.
based on experience
Lab tests— 5 to 6 per acre
Falling head field test at
density test location— 1 per
acre of lift
Lab test-1 per 10,000 to
20.000 ft2 or every third
lift
Sample or
test hole Test fill
filling method recommended
Mot usually
Bentonite Yes
—
Liner material Yes
hand-tamped
-
Yes
Yes
Liner material
or bentonite
hand- tamped
Maximum clod size
allowed
1/2 lift' thickness; none
greater than lift
thickness
2 inches
1/2 lift thickness
1/2 lift thickness, if
possible; none greater
than lift thickness
-
1 inch in upper lifts;
may be larger in lower
lifts
Maximum rock or
root size allowed
1
2 inches
-
4 to 6 inches
3 to 6 inches or
1/2 lift thickness
1 inch .
' <
(continued)
-------
TABLE 5-11. (continued)
ui
o
o
Source of
information3
I
J
K
HlniBun frequency of
In-place Moisture or
density tests
1 per 2,500 ft? aininua
of 4 per lift or 1 per
day— More as needed
1 per 10,000 ft? or 1 per
1 1 It
Site specific, based on
experience
Permeability testing
Laboratory test— 1 per 2,000
yd3 or 2 per lift— occasional
field test
Lab and field
Lab trl axial
Sample or
test hole
filling method
Bentonite
—
Bentonite
Test fill
recontended
Yes
Yes
Sometimes,
especially
for admix-
tures
Maximum clod size
allowed
1/2 lift thickness
1/2 lift thickness if
possible; none greater
than lift thickness;
1 to 3 Inches for hand-
compacted areas
3 to 6 inches, depending
on compaction equipment;
1 inch In upper 1 foot
of liner If overlain by
Max i num rock or
root size allowed
-
II
1 inch
FML; none greater than
lift thickness
L 1 per 1,000 yd3 or 1 per
day
M 1 per 1,000 yd3 or 1 per
day
Nd 1 per 1,000 yd3—! per 200
yd3 if hand tanped
Lab constant head—1 per 25.000
yd3 or 1 per week
Lab falling head—1 per
25,000 yd3
Yes
No
Yes
3 inches; 3/4 inch for
blended materials
'Sources A through H were design and geotechnical engineers with clay liner construction experience.
DU.S. Bureau of Reclamation, 1974. .
<-U.S. Department of the Navy, 1982. *
"Ghasseni et al., 1983.
3 Inches
Oc 1 per 500 to 1,000 yd3
Pd 1 per 2,000 yd3
3 inches; 1 to 3 Inches
for hand-compacted areas
Lab— 1 per 16,000 yd3, field— — — 3 to 6 Inches
1 per 40,000 yd3
-------
TABLE 5-12. RECOMMENDATIONS FOR CONSTRUCTION DOCUMENTATION OF CLAY-LINED
LANDFILLS BY THE WISCONSIN DEPARTMENT OF NATURAL RESOURCES
Item
Testing
Frequency
1. Clay borrow source
testing
2. Clay Uner testing
during construction
3. Granular drainage
blanket testing
Grain size
Moisture content
Atterberg limits
(liquid limit and
plasticity index)
Moisture-density curve
Lab permeability
(remolded samples)
Density
(nuclear or sand cone)
Moisture content
Undisturbed permeability
Dry density
(undisturbed sample)
Moisture content
(undisturbed sample)
Atterberg limits
(liquid limit and
plasticity index)
Grain size (to the
2-m1cron particle size)
Moisture-density curve
(as per clay borrow
requirements)
Grain size
(to the No. 200 sieve)
Permeability
1,000 yd3
1,000 yd3
5,000 yd3
5,000 yd3 and all
changes in material
10,000 yd3
5 tests/acre/lift
(250 yd3)
5 tests/acre/lift
(250 yd3)
1 test/acre/lift
(1,500 yd3)
1 test/acre/lift
(1,500 yd3)
1 test/acre/lift
(1,500 yd3)
1 test/acre/lift
(1,500 yd3)
1 test/acre/lift
(1,500 yd3)
5,000 yd3 and all
changes 1n material
1,500 yd3
3,000 yd3
Source: Gordon et al., 1984.
5-101
-------
TABLE 5-13. ELEMENTS OF A CONSTRUCTION DOCUMENTATION REPORT
Major—etements
Components
A. Engineering plans
Completed sub-base elevations.
Final clay Uner grades.
Top of drainage blanket grades.
Leachate collection lines, cleanouts,
and manholes with spot elevation every
100 feet along the lines and at all
manhole entrances and exits.
Drainage features.
All monitoring devices.
Spot elevations at all breaks In slope
and on approximate 100-foot centers.
Document testing locations.
Other site information as appropriate.
B. Engineering cross-sections
C. Comprehensive narrative
D. Series of 35-mm color prints
E. Construction certification
Minimum of one east-west and one
north-south through the completed
area.
Explaining how construction of the
project was accomplished along with an
analysis of the soil-testing data
obtained in 1 through 3 above. This
report should also include an appendix
containing all the raw data from the
field and laboratory testing.
Documenting all major aspects of site
construction.
Should be certified by a registered
professional engineer to have been
completed according to the approved
plans. Any deviations from the plans
should be noted and explained.
Source: Gordon et al., 1984.
5-102
-------
01
I
o
CJ
Loc.tlo. IUIU1 u«. rio.1 jin. tlB,r tu.-kE...
*la°° "MOO W1300 WI200 -Jft. if*.*' i.'t.tl (f..n ATTTISiM LIMIT IES7 ?E
"I~ "*" """ ~1| »-SO 3I.S7 1..7
« 2I-18 <*.T3 1. 15 Life Li^Jld Lt.U/M».tlsl
« 1 20.00 ;j!e) |(7 ^^ "
1 1 =0.<0 H.59 IS, 12,» 5.^37 5$;
1 „ a 1 "•"> 50.30 .«7 S!/15 S!/>| ij,,,
« I " 0 I 8 'n Jo'Sl ''f «''F -i'-° '"'?
« « 1 ? * ° I 1 -J= ' i« »:
o o 1 ° I 1 t s 1J'91 '°* 76/s< 6D/JT T2/5°
1 » ^1 o" " " 1 ° ° 1 ' :" ":" •»
1 0 0 1 ° II ! •'• ».™ .11
11 1 « 1 " •" 10-M »-
1 1 a \ M !f \ ' " '" "•" »
1 1 11 a S \ 0 ° 1 0 » .91 1I.5« tl
1 jo 1 „ 0 ° 1 1 HI 2» .« u.ll .M
1 o 0 I ° 1 1 * " " "••' »
1 ° \ 1.1 22 « 31.3« 17
1 I t \ " 0 * \ *\ " "'I* "
1 \ „ \ *> i' 0 » " "•" « Llfl =„..„ ..„•„ _„„
1 ff \ " 0 1 lH* M 5T 19 31 IT '°- 2&Liia "i»t r.jc. j»«t sinf
1 ° \Ji-—T, — ^"/X0 I I i "^ " "-" ?
\'°/j °° ° °^4 -i ?S si i ^ ^ '
1 y — 8 31 Tl ti'.at 31 'ow
I/ ___ H! W M ,2.01 $
\k. — '" « 30 11.2t 51
ELEVATION POINTS « »'jj ij']i »j
SUITS
»T/30
46YZB
52 3<
65 42
T» »9 .
<5 M 1|
£
x 1C T)
jt gp)jtTi SlpFft
« «.2I 13.M C5
NOTE, It '•!! »•» » LABORATORY TEST DATA
Licur IhickMtioi givin for poinli oo llopn on \\ ,l'" ]!'!? !1
pirptndiculor to tlopi lou. " "
El«votion doto wot corrtctid for slop* onglt
in tho cott of such points.
LINER THICKNESS DATA
SUMMARY of RESULTS
THICKNESS DATA- ATTERBERG LIMIT DETERMINATIONS'-
Number of data points - 49 »» Number of t«"i M 30
Average thickness value, 5.5 Av*"|J «""•' . 5g
Standard deviation » 0.7 ft. *H- ^ ^
plasticity mdex * 40
Standard deviation!
COMPACTION DATA: liquid limit » 13
Number of tests .148 plasticity mdex - 17
Average compaction, 9I% PERMEAB1L|TY TESTS:
Number of tests • 14
PERMITTING STUDY
•
I SOL INVE3TOA710H SI
A^OL [ESINGENGMEERS, INC.
II IV 1 CM 1 JPMMIM
I DOW 1 SJC 1 [ Q-2A
1 1 1 1
Source: Courtesy of Soil Testing Engineers, Inc., Baton Rouge, Louisiana
Figure 5-20. CQC test location and data summary.
-------
U1
i
to
»0
40
30
20
10
0
I
.o4
4
1
id*
4
t
4
t
10*
4
2
Id'
,0-'
. t
i«94.3%
• >I2»M
•
— -F^FTf
O SO TO tO t
SATURATION
~~"-*~^
~~**"-^^
•-^
• '• ._ —
L^-sos_ |
^»-^
~^
1
1
~^s
^>..
7
o
^
-
N
to
30
20
o
\
II
1
r
K«U>K
tf^Q4!1
• •15
J*M '
13
** 5
o
0
0
0
0
et93.0%
• •129 iMk
100 lo"144^'44*)-*14'!?7*44!^14**)-'144*)-4 "70 at
Cofflclot •' Ptrmiklllty (K) -ci/«ic
LABORATORY TESTS ON UNDISTURBED SAMPLES D>
PERMEABILITY
CHARACTERISTICS AS COMPACTED
NOT
1 1
Q1E.IRI grin Ian fron
Ctll 10 Toll.
^ ..,..]
%^
— —
W tO TO tO (C
REGRESSION CORRELATION
ill
-
^N^
*5
&
=
-«
^—
^
\
^===1
^s
S
N^
i
M
130
120
HO
100
90
to
TO
0
\\\l-
» \ v ^
1 \ \
\ \ >X
* \
\
\
\
v \
\\
\\
N V]
\ \
\
\ \
\ v.
\
\ ',
\
\
t«r»'
^ Av«rt|
s.
\
\ ^
\ j
\
\
\
•E,
\\.
\ xx
\ %
*, \
. \
\
S
S.
N.
r Ci»rv« (0 68.8)
t FUN Otntlty Ttit
e UD 3o«f>U f«r
• ••blllty T.et
x^
1^
N
x N^-
N.
"v
'N
>
(O 8O 34
COMPACTION DATA
K.
x N
\% X
*>*•
^
x.
>
X*"
x-
k
'Pvrniiklllty
N.r»,| Olitrlkul ••
Laatt-S^utr«t Uvltlpla
SOTL TESTING ENGINEERS, INC
/ j \Battl RCXJM. Lo Uto CtolM. U,
BY Glut Fa NO
DOW GPB 2
Source: Courtesy of Soil Testing Engineers, Inc., Baton Rouge, Louisiana
Figure 5-21. Statistical analysis of CQC test data.
-------
chronologic framework for identifying and recording all other reports. At a
minimum, the jummary reports should include the following information
(Spigolon andJ£e_Lley, 1984): ; #
• Unique identifying sheet number for cross-referencing and document
control
• Date, project name, location, and other identification
• Data on weather conditions
• Reports on any meetings held and their results
• Unit processes, and locations, of construction underway during the
time frame of the report
t Equipment and personnel in each unit process, including
subcontractors
• Descriptions of areas or units of work (blocks) tested and/or
observed and documented
• Description of offsite materials received, including any quality
verification (vendor certification) documentation
• Calibrations, or recallbratlons, of test equipment, Including
actions taken as a result of recallbration
• Decisions made regarding approval of units of material or of work
(blocks) and/or corrective actions to be taken 1n Instances of
substandard quality
t Unique identifying sheet numbers of observation and testing data
sheets and/or problem reporting and corrective measures data sheets
used to substantiate decisions described In the preceding item
• Signature of the CQC Inspector and concurrence by the CQA officer.
Items above may be formulated Into site-specific^checklists and data
sheets so that details are not overlooked.
5.3.4.1.2 Observation and Testing Data Sheets—All observations and
field and/or laboratory tests should be recorded on appropriate data sheets.
Required data to be Included 1n a test report (data sheet) for most of the
standardized test methods are included 1n the pertinent ASSHTO (1983) and
ASTM (1985) standards. Examples of field and/or laboratory test data sheets
are given 1n U.S. Department of the Army (1970, 1978) manuals and 1n Spigolon
and Kelley (1984).
Due to their nonspecific nature, no standard format can be given for
data sheets to record observations. Recorded observations may take the form
of notes, charts, sketches, photographs, or any combination of these. Where
possible, a checklist may be useful to ensure that no pertinent factors of a
task-specific observation are overlooked.
5-105
-------
Observation and testing data sheets should Include at least the
following Information (Splgolon and Kelly, 1984):
• Unique Identifying sheet number for cross-referencing and document
control
• Description or title of the observation/test
t Location of the observation/test or location from which the sample
Increment was obtained
• Type of observation/test; procedure used (reference to standard
method when appropriate)
• Recorded observations or test data, with all necessary calculations
• Results of the observation/test; comparison with specification
requirements
• Personnel Involved 1n the observation/test
• Signature of the CQC Inspector and concurrence by the CQA officer.
Items above may be formulated into site-specific checklists and data sheets
so that details are not overlooked.
cu ^5'3:4^*3 Problem Identification and Corrective Measures Data
Sheets—Problem reporting ana corrective measures data sheets should be
nrnh?Ime!!ce?S!dfJ«S58C1J!C Obfervat1on or testing data sheets where the
problem was Identified. They should, at a minimum, Include the following
information: a
• Unique Identifying sheet number for cross-referencing and document
control
t Detailed description of the problem
• Location of the problem
«
• Probable cause
t Method and time frame of locating the problem (reference to data
sheets)
• Estimated duration of the problem
• Suggested corrective action
• Documentation of correction (reference to data sheet)
• Final results
• Suggested methods to prevent similar problems
Signature of the CQC inspector and concurrence by the CQA officer.
5-106
a
-------
In some cases, not all of the above information wm be available or
obtainable. However, when available, such efforts to document problems could
help to avoid-similar future problems.
The CQA officer should be made aware of any significant recurring
nonconformances. The CQA officer will then determine the cause of any
problems and recommend appropriate changes to prevent future recurrence.
When this type of evaluation 1s made, the results should be documented.
Upon receiving the CQA officer's written concurrence, copies of the
report(s) should be sent to the design engineer and the facility owner/
operator for their comments and/or acceptance. These reports should not be
submitted to the permitting agency unless they have been specifically
requested. However, a summary of the data sheets along with final testing
results and inspector certification of the facility may be required by the
permitting agency upon completion of construction.
5.3.4.2 Photographic Reporting Data Sheets--
Photographic reporting data sheets may also prove useful. Such data
sheets should be cross-referenced with observations or testing data sheets
and/or problem identification and corrective measures data sheets.
Photographic reporting data sheets should include the following minimum
information:
t A unique identifying number for cross-referencing and document
control
t The date, location, and weather conditions for the photograph
t Location and description of the work or work product
• Purpose of the photograph
t Signature of the photographer and CQC Inspector.
These photographs will serve as a pictorial record of work progress,
problems, and mitigation activities. They should be kept in plastic file
sheets in the chronological order in which they were taken. The basic file
should contain color prints; negatives should be stored 1n a separate file 1n
chronological order.
A video recording of problem areas and/or conditions and of the
completed Installation of each soil component may also prove useful.
5.3.4.3 Block Evaluation Reports-
Each Inspection block may have several quality characteristics, or
parameters, that are specified to be observed or tested, each by a different
observation or test, with the observations and/or tests recorded on different
data sheets. At the completion of each block, these data sheets should be
organized Into a block evaluation report. These block evaluation reports may
then be used to summarize all of the site construction activities.
5-107
-------
Block evaluation reports should be prepared by the CQA officer and, at a
minimum, shouJd Include the following Information (Splgolon and Kelley,
1984)t -.
• Unique Identifying sheet number for cross-referencing and document
control
0 Description of block (use project coordinate system to Identify
areas and appropriate identifiers for other units of materials or
work)
• Quality characteristic being evaluated; references to sections of
specifications
• Sampling method; how 1t was established
• Sample increment locations (describe by project coordinates or by a
location sketch on the reverse of the sheet)
• Tests and/or observations made (define procedure by name or other
identifier; give unique identifying sheet number for observation
test data sheets)
t Summary of test data (give block average and, 1f available, the
standard deviation for each quality characteristic)
• Define acceptance criteria (compare block observation/test data
with design specification requirements; indicate compliance or
noncompllance; 1n the event of noncompliance, Identify documenta-
tion that gives reasons for acceptance without specification
compliance)
t Signature of the CQA officer.
5.3.4.4 Design Engineer's Acceptance of Completed Components--
All daily inspection summary reports, observation and testing data
sheets, problem identification and corrective measures sheets, and block
evaluation reports should be reviewed by the CQA officer and then submitted
to the design engineer. The reports should be evaluated and analyzed for
Internal consistency and for consistency with similar work. Timely
submission of these documents will permit errors, inconsistencies, and other
problems to be detected and corrected as they occur, when corrective measures
are easiest.
The design engineer should assemble and summarize the above information
Into a periodic design acceptance report. The reports should indicate that
the materials and workmanship comply with design specifications and permit
requirements. These reports should be Included 1n project records and, 1f
requested, submitted to the permitting agency.
5.3.4.5 Final Documentation—
At the completion of the project, the facility owner/operator should
submit a final documentation report to the permitting agency. This report
should Include all of the design engineer's acceptance reports (I.e.,
5-108
-------
periodic summaries of all dally Inspection summary reports, observation and
test data shejts, problem Identification and corrective measures data sheets,
and block evaluation reports), deviations from design and material specifica-
tions (with justfylng documentation), and as-built drawings. This document
should be prepared by the CQA officer and Included as part of the CQA plan
documentation.
5.3.4.5.1 Responsibility and Authority—The final documentation should
reemphaslze that areas of responsibility and lines of authority were clearly
defined, understood, and accepted by all parties involved 1n the project.
Signatures of the facility owner/operator, design engineer, CQA officer, CQC
inspector, and construction contractor should be included as confirmation
that each party understood and accepted the areas of responsibility and lines
of authority and performed their function(s) according to the CQA plan.
5.3.4.5.2 Relationship to Permitting Agencies—Final documentation
submitted to the permitting agency as part of the CQA plan documentation does
not sanction the CQA plan as a guarantee of facility construction and
performance. Rather, the primary purpose of the final documentation is to
improve confidence 1n the constructed facility through written evidence that;
the CQA plan was Implemented as proposed and that the construction proceeded
according to design plans and specifications.
5.3.4.6 Storage of Records—
During the construction of a hazardous waste land disposal facility, the
CQA officer should be responsible for the facility records, including the
originals of all the data sheets, summary reports, and block evaluation
reports; the design engineer's acceptance of completed components reports;
and facility drawings. With the CQA officer 1n charge of the facility
construction records, any documentation problems should be noted and
therefore remediated quickly. Once the facility construction 1s complete,
the document originals should be stored by the owner/operator 1n a manner
that will allow for easy access. An additional copy should also be kept at
the facility if this 1s 1n a different location from the owner/operator's
files. A final copy should be kept by the permitting agency 1n a publicly
acknowledged repository.
5.4 CLAY LINER DESIGN AND CONSTRUCTION: PROBLEMS AND PREVENTIVE MEASURES
*
Table 5-14 lists common problems encountered during clay liner design
and construction, their probable causes, and suggested solutions. This table
was compiled from Information assembled from a wide variety of sources,
including literature surveys, case studies, and Interviews with experts 1n
the field. This table summarizes Information presented 1n this chapter.
5-109
-------
TABLE 5-14. POTENTIAL CLAY LINER DESIGN AMD INSTALLATION PROBLEMS AMD PREVENTIVE MEASURES
Problems
Cause
Preventive measures
Sidewall slump and collapse
Improper characterization
of soil strength profile
that results in inproper
sidewall design.
High inward hydraulic
pressure on sidewalls
(sites below water table).
bottom heave or rupture
o High inward hydraulic
pressure on bottom
(sites below water table).
01
i
Accumulation of water in • Rainfall
landfill during construction
Seepage into site
(sites below water table)
Drying and cracking of clay
liner, greatly increasing
permeability
• Desiccation
• Properly characterize subsurface condi-
tions.
t Design a more gentle sidewall slope
(depending on shear strength
of foundation soil.)
• Reduce hydraulic head by:
— Installing slurry wall around site
perimeter to cut off groundwater.
— Trenching and pumping around site
to cut off groundwater.
— Pumping from wells to lower local
groundwater table.
• Control depth of excavation to
lower head potential.
• Reduce hydraulic.head by slurry
wall, pumping, or other technique.
Fill landfill before heaving occurs.
• Cover site with inflatable dome (re-
duces leachate treatment requirements
for continuous operation facilities).
• Seal roll liner at end of construc-
tion day to ensure proper runoff of
precipitation into .sump.
• Reduce hydraulic head in surrounding soil.
• For all infiltration:
— Operate leachate collection
system to remove water.
-- Design leachate collection system
or detection system (between liners)
to handle extra water input.
t Do not construct during extremely
hot, dry periods.
• Wet down liner during dry periods.
• Cover liner with plastic sheet or
soil layer if liner construction is
interrupted.
9 Do not leave liner exposed prior to
waste emplacement or leachate collection
system installation.
(continued)
-------
TAULE b-14 (continued)
ui
Problems
Cause
Preventive measures
Loss in liner density
(increased permeability)
• Freeze/thaw
Reduction in clay workability • Low temperatures
Erosion of upper liner after
construction
(liquid impoundments)
Visible partings between
liner lifts and increased
permeability parallel to
lifts
Pockets of high-permeability
material (e.g., sand and
gravel) in liner material
Leakage around designed
liner penetrations
Leachate collection
system clogging
Leachate collection syatein
damage during waste
emplacement
• Wave erosion
Liner lifts not properly
tied together
Heterogeneous clay liner
material
• Improper sealing around
penetrating objects
• Sediment entering into
systems
• Inadequate management
of personnel
Areas of high permeability • Substandard compaction
• Do not construct during winter in
cold climates.
• Cover liner with soil blanket or other
Insulation material when cold weather is
anticipated.
* Increase compact!'ve effort.
• Cease construction till spring.
• Place rip-rap on lagoon sidewalls
extending from below liquid level into
the freeboard area.
• Scarify or disk lower lift prior to
installing next lift.
• Ensure moisture content of last lift
and lift being installed are the same.
• Closely inspect liner material at
borrow site or as it is being
installed and reject coarse-grained
material.
0 Avoid liner penetrations in design.
• Seal properly around penetrating objects.
• Cover system with geotextile or
graded soil layer.
Initiate personnel training program.
Cover system with geotextile or soil
cover prior to waste management.
Design system to minimize protrusions
(manholes, etc.) in waste emplacement
areas.
• Conduct proper CQA and CQC, including:
-- Monitoring for'material variability
-- Observation of compaction operations
— Moisture/density/permeability tests.
-------
5.5 REFERENCES
AASHTO: The Anarjcan Association of State Highway and Transportation
Officials. 1978. Standard Specifications for Transportation Materials
and Methods of Sampling and Testing, Part 2, Methods of Sampling and
Testing, 12th edition. Washington, D.C.
Anderson, D. 1982. Inplace Closure of Hazardous Waste Surface
Impoundments (draft), Chapter 5—Evaluating Stabilized Waste Residuals.
EPA-68-83-2943.
API. 1982. American Petroleum Institute. API Recommended Practice:
Standard Procedure for Testing Drilling Fluids. API RP13B. American
Petroleum Institute, Dallas, Texas.
ASTM. 1985. The American Society for Testing and Materials. 1985 Annual
Book of ASTM Standards, Volume 4.08, Soil and Rock; Building Stones.
Philadelphia, Pennsylvania.
Bass, J. 1986. Avoiding Failure of Leachate Collection and Cap Drainage
Systems, EPA 600/2-86-058, U.S. Environmental Protection Agency,
Cincinnati, Ohio 55268.
Bernreuter, D. L., and D. H. Chung. 1984. Earthquake Hazard Analysis for
Nuclear Power Plants, Energy and Technology Review. Lawrence Livermore
National Laboratory, Livermore, California.
Bishop, A. W. 1955. The Use of the Slip Circle in the Stability Analysis
of Slopes. Geotechnique. 5(1):1-17.
Bishop, A. W., and N. R. Morgenstern. 1960. Stability, Coefficients for
Earth Slopes. Geotechnique. 10(4):129-150.
Boutwell, G. P., Jr., R. B. Adams, and D. A. Brown. 1980. Hazardous Waste
Disposal in Louisiana. Geotechnical Aspects of Waste Disposal.
American Society of Chemical Engineers, A Two-Day Seminar.
Boutwell, G. P., Jr., and V. R. Donald. 1982. Compacted Clay Liners for
Industrial Waste Disposal. American Society of Slvll Engineers,
National Meeting, Las Vegas, Nevada.
Cashman, P. M., and E. T. Haws. 1970. Control of Groundwater by Water
Lowering. Chapter 3 in: Ground Engineering, Institution of Civil
Engineers, London.
C1chow1cz, N. L., R. W. Pease, Jr., P. J. Stoller, and H. J. Yaffe.
1981. Use of Remote Sensing Techniques 1n a Systematic Investigation of
an Uncontrolled Hazardous Waste Site. EPA 600/2-81-187, U.S.
Environmental Protection Agency, Cincinnati, Ohio.
Daniel, D. E. 1981. Problems 1n Predicting the Permeability of Compacted
Clay Liners. Symposium on Uranium Mill Tailings Management, Fort
Collins, Colorado, pp. 665-675.
Daniel, D. E. 1984. Predicting Hydraulic Conductivity of Clay Liners.
Journal of Geotechnical Engineering. 110(2):285-300.
5-112
-------
Daniel, D. E. 1987. Hydraulic. Conductivity; Tests for Clay Liners. Ninth
Annual Symposium on Geotechnical and Gebhydrological Aspects of Waste
Management-— Colorado State University, Fort Collins, Colorado.
Daniel, D. E., and R. E. Olson. 1980. Geotechnical Aspects in Design
of Disposal Sites for Low-Level Radioactive Wastes. American Society of
Civil Engineers, Texas Section Meeting, San Antonio, Texas.
D'Appolonia, D. J. 1980. Soil-Bentonite Slurry Trench Cutoffs.
Journal of the Geotechnical Engineering Division, ASCE. GT-4:389-417.
Davis, F. J. 1966. Summary of Bureau of Reclamation Experience in
Statistical Control of Earth Dam Embankment Construction. National
Conference on Statistical Quality Control Methodology in Highway and
Airfield Construction, University of Virginia, Charlottesville.
Day, M. E. 1970. Brine Pond Disposal Manual. Office of Solid Waste
Contract No. 14-001-1306, U.S. Bureau of Reclamation, Denver, Colorado.
Day, S. R., and D. E. Daniel. 1985. Hydraulic Conductivity of Two
Prototype Clay Liners. In: Land Disposal of Hazardous Waste: Proceed-
ings of the Eleventh Annual Research Symposium. EPA/600/9-85/013.
U.S. Environmental Protection Agency, Cincinnati, Ohio.
Diamond, S. 1979. Preliminary Report on Analysis and Testing Concerning
Polymer Bentonite Lagoon Seal at Wurtsmith Air Force Base, Oscoda,
Michigan. Prepared for Slurry Systems, Inc., Gary, Indiana.
Dobrin, M. B. 1960. Introduction to Geophysical Prospectinq.
McGraw-Hill, New York.
Eagling, D. G. 1983. Seismic Safety Guide. LBL-9143, Engineering and
Technical Services Division, Lawrence Berkeley Laboratory, University of
California, Berkeley, California.
EERC: Earthquake Engineering Research Center. 1984. Computer Software
for Earthquake Engineering. National Information Service for Earthquake
Engineering, University of California, Berkeley, California.
Elsbury, B. R., J. M. Norstrom, D. C. Anderson, J. A. Rehage, J. 0. Sai,
R. L. Shiver and D. E. Daniel. 1985. Optimizing Construction Criteria
for a Hazardous Waste Soil Liner Phase I Interim Report. EPA-68-08-
3250, U.S. Environmental Protection Agency, Cincinnati, Ohio.
Ely, R. L., Jr., G. L. Kingsbury, M. R. Branscome, L. J. Goldman, C. M.
Northelm, J. H. Turner, and F. 0. M1xon, Jr. 1983. Performance of Clay
Caps and Liners for Disposal Facilities. Final Report. EPA Contract
No. 68-03-3149. U.S. Environmental Protection Agency, Cincinnati, Ohio.
Esu, F. 1966. Short-Term Stability of Slopes 1n Unweathered Jointed
Clays. Geotechnique. 16(4):321-328.
Fang, H.-Y. 1975. Stability of Earth Slopes. Chapter 10 1n: Wlnterkorn,
H. F., and H.-Y. Fang, Foundation Engineering Handbook, Van Nostrand
Reinhold, New York.
5-113
-------
Fellenius, W. 1927. Erdstatische Berechnungen (Calculation of the
Stability of Slopes). W. Ernst und $ohn, Berlin.
Fenn, D. G., et al. 1977. Procedures Manual for Ground-water Monitoring
at Solid Waste Disposal Facilities. EPA-530/SW-G11, U.S. Environmental
Protection Agency, Office of Solid Waste, Washington, D.C.
Freeze, R. A., and J. A. Cherry. 1979. Groundwater. Prentice-Hall,
Inc., Englewood Cliffs, New Jersey.
Geo-Con, Inc. 1984. Soil Bentonite Liners. Pittsburgh, Pennsylvania.
Ghasseml, M., M. Haro, and L. Fargo. 1984. Assessment of Hazardous Waste
Surface Impoundment Technology (Case Studies and Perspectives of
Experts) (draft). EPA-68-02-3174, U.S. Environmental Protection Agency,
Multidisciplinary Energy and Environmental Systems and Applications,
Torrance, California.
Ghasseml, M., M. Haro, J. Metzgar, M. Powers, S. Qulnlivan, L. Sdnto,
and H. White. 1983. Assessment of Technology for Constructing Cover
and Bottom Liner Systems for Hazardous Waste Facilities (Final Report)
EPA 68-02-3174, U.S. Environmental Protection Agency, Cincinnati, Ohio.
Gordon, M. E., P. M. Huebner, and P. Kmet. 1984. An Evaluation of the
Performance of Four Clay-Lined Landfills in Wisconsin. Seventh Annual
Madison Waste Conference, University of Wisconsin, Madison, Wisconsin.
Hays, W. W. 1980. Procedures for Estimating Earthquake Ground Motions.
USGS Prof. Paper 1114. U.S. Government Printing Office, Washington,
u.C.
Haxo, H. E. 1980. Interaction of Selected Lining Materials with Various
Hazardous Wastes—II. In: Disposal of Hazardous Wastes—Proceedings of
the Sixth Annual Research Symposium, pp. 160-180.
Haxo, H. E., et al. 1983. Lining of Waste Impoundment and Disposal
Facilities. SW-870, U.S. Environmental Protection Agency, Cincinnati,
Ohio.
«
H1lf, J. W. 1975. Compacted Fill. Chapter 7 1n: Wlnterkorn, H. F., and
H.-Y. Fang, Foundation Engineering Handbook, Van Nostrand Relnhold. New
York.
IMC. 1982. The Role of Bentonite 1n Soil Sealing Applications (Promotional
Literature). Mundeleln., Illinois.
Johnson Division. 1975. Groundwater and Wells. UOP, Inc., St. Paul,
Minnesota.
Johnson, A. W., and J. R. Sallberg. 1960. Factors that Influence Field
Compaction of Soils. Bulletin 272. Highway Research Board, Washington,
D.C.
5-114
-------
Johnson, H. V., S. J. Spigolon, and R. J. Lutton. 1983. Geotechnlcal
Quality .Control: Low-Level Radioactive Waste and Uranium Mill Tailings
D1sposaUajC-1l1t1es. NUREG/CR-3356, U.S. Nuclear Regulatory Commission,
Washington, D.C.
Johnson, J. J., and F. J. Gelsel. 1979. Clay Caps Lagoon System. Water
and Sewage Works. 126(11):30-31.
Johnson, P. E., and S. W. Cole. 1976. Benton1te/Glac1al Till: The
Right Mixture for Lining Brown Co. Cascade Mill Lagoons. Paper Trade
Journal. 160(16):36-38.
Jorgenson, J. L. 1971. Development and Trial Use of Acceptance Sampling
Plans for Compacted Embankments. Highway Research Record. 357:24-34.
Kmet, P., K. J. Qulnn, and C. Slavlk. 1981. Analysis of Design Parameters
Affecting the Collection Efficiency of Clay Lined Landfills. Fourth
Annual Madison Conference of Applied Research and Practice on Municipal
and Industrial Waste, University of Wisconsin, Madison, Wisconsin.
Kotzlas, P. C., and A. C. Stamatopoulos. 1975. Statistical Quality
Control at Kastrakl Earth Dam. Journal of the Geotechnlcal Engineering
Division, ASCE. 101(GT9):837-853.
Kozickl, P., and D. M. Heenan. 1983. Use of Bentonlte as a Soil Sealant
for Construction of Undersea! Sewage Lagoon Extension, Glenboro
Manitoba. Short Course on Waste Stabilization Ponds, Winnipeg,
Manitoba.
Lambe, T. W., and R. V. Whitman. 1969. Soil Mechanics. John WHey and
Sons, New York. pp. 353-373.
LARG. 1982. Seismic Location Standard. LARG-830131. Llvermore
Associated Research Group, Inc., Llvermore, California.
Lundgren, T. A. 1981. Some Bentonlte Sealants 1n Soil Mixed Blankets.
In: Proceedings, of the Tenth International Conference on Soil Mechanics
and Foundation Engineering, Stockholm.
Lutton, R. J., G. L. Regan, and L. W. Jones. 1979. Design and
Construction of Covers for Solid Waste Landfills. EPA-600/2-79-165,
U.S. Environmental Protection Agency, Cincinnati, Ohio.
Lutton, R. J., w. E. Strohm, Jr., and A. B. Strong. 1983. Subsurface
Monitoring Programs at Sites for Disposal of Low-Level Radioactive
Waste. NUREG/CR-3164, NTIS, Springfield, Virginia.
Makd1s1, F. I., and H. B. Seed. 1978. Simplified Procedures for Estimating
Dam and Embankment. Earthquake Induced Deformations. Journal of the
Geotechnlcal Division, ASCE. 109 (GT7).
Miller-Warden Associates. 1965. Development of Guidelines for Practical
and Realistic Construction Specifications. NCRHP Report 17. Highway
Research Board, Washington, D.C.
5-115
-------
Mitchell, J. K. 1976. Fundamentals of Soil Behavior. John Wiley and
Sons, Inc., New York.
Mitchell, J. K., D. R. Hooper, and R. G. Campanella. 1965. Permeability
of Compacted Clay. Journal of the Soil Mechanics and Foundations
Division, ASCE., 91(SM4):41-65.
Morrison, W. R., E. W. Gray, Jr., D. B. Paul, and R. K. Frabel. 1982.
Installation of Flexible Membrane Lining 1n Mt. Elbert Forebay
Reservoir. REC-ERC-82-2. U.S. Bureau of Reclamation, Denver, Colorado,
Olson, R. E., and D. E. Daniel. 1981. Measurement of the Hydraulic
Conductivity of Finegrained Soils. ASTM STP 746. pp. 18-64.
Pelrce, J. J., G. Sallfors, and E. Peterson. 1986. Clay Liner Construction
and Quality Control. ASCE Journal of Environmental Engineering,
February 1986. 112(1):13-24."
Perloff, W. H. 1975. Pressure Distribution and Settlement. Chapter 4
1n: Winterkorn, H. F. and H.-Y. Fang, Foundation Engineering Handbook,
Van Nostrand Reinhold, New York.
Rogowskl, A. S., and E. B. Richie. 1984. Relationship of Laboratory
and Field Determined Hydraulic Conductivity in Compacted Clay Soils.
In: Proceedings of Sixteenth Mid-Atlantic Industrial Waste Conference,
Pennsylvania State University, University Park.
Rollins, M. B. 1969. Sealing Properties of Bentonite Suspensions. Clays
and Clay Minerals. 16:415-423.
Ryan, C. R. 1984. Barriers-Technology and Construction. Seminar
Presented at the Hazardous Materials Control Research Institute
Conference, Washington, D.C., November 7.
Schmednecht, F. C., and J. Harmston. 1980. Vibrated Beam Techniques for
Cutoff Wall Construction. Presented at ASCE and AEG Meeting, North
Carolina. State-University, Raleigh, N.C., American Foundations Inc.,
Morr1sv1lle, N.C.
Schuster, R. L., and R. J. KMzek. 1978. Landslides: Analysis and
Control. TRB Special Report 176, National Academy of Sciences,
Washington, D.C.
Seed, H. B. 1975. Earthquake Effects on Soil-Foundation Systems.
Chapter 25 1n: Winterkorn, H. F., and H.-Y. Fang, Foundation
Engineering Handbook, Van Nostrand Reinhold, New York.
Sellg, E. T. 1982. Compaction Procedures, Specifications, and Control
Considerations. Transportation Research Record 897:1-8, National
Academy of Sciences, Washington, D.C.
Sherard, J. L. 1967. Some Considerations 1n Earth Dam Design. Journal
of the Soil Mechanics and Foundations Division, ASCE. 93(SM4):377-401.
5-116
-------
Sherard, J. L., and R. S. Decker. 1977. Summary Evaluation of Symposium
in Dispersive Clays. In: Dispersive Clays, Related Piping, and Erosion
In Geotetbflical Projects, ASTM STP 623, ASTM. Philadelphia,
Pennsylvania, pp. 467-479.
Sherard, J. L., R. j. Woodward, S. F. C1z1enski, and W. A. Clevenger.
1963. Earth and Earth-Rock Dams. John Wiley and Sons, New York.
Small, D. M. 1981. Establishing Installation Parameters for Installing
Liner Systems. Universal Linings, Inc., Haverford, Pennsylvania.
Spencer, E. 1967. A Method of Analysis for Stability of Embankments Using
Parallel Intersllce Forces. Geotechnlque 17:11-26.
Spigolon, S. J. 1984. Construction Quality Assurance for Dikes
(Preliminary Draft). U.S. Environmental Protection Agency, Cincinnati,
Ohio.
Spigolon, S. J., and M. F. Kelley. 1984. Geotechnlcal Quality Assurance of
Construction of Disposal Facilities. EPA 600/2-84-040, U.S. Environ-
mental Protection Agency, Cincinnati, Ohio.
Taylor, D. W. 1948. Fundamentals of Soil Mechanics. John Wiley and Sons.
New York. pp. 406-479.
Turnbull, W. J., J. R. Compton, and R. G. Ahlvln. 1966. Quality Control
of Compacted Earthwork. Journal of the Son Mechanisms and Foundation
Division, ASCE. 92(SM1).
U.S. Bureau of Reclamation. 1973. Design of Small Dams. 2nd edition.
U.S. Department of the Interior, Washington, D.C.
U.S. Department of the Interior. 1974. Earth Manual, 2nd edition.
Washington, D.C.
U.S. Department of the Army. 1970. Laboratory Soils Testing. EM
1110-2-1906, Washington, D.C.
U.S. Department of the Army. 1977. Construction ConYrol for Earth and
. Rock-Fill Dams. EM 1110-2-1911, Washington, D.C.
U.S. Department of the Army. 1978. Contractor Quality Control Plan.
ER 1180-1-6, Washington, D.C.
U*S' SnEr^r* of the Navy' 1982' Foundations and Earth Structures.
NAVFAC DM-7. 2, Naval Facilities Engineering Command, Alexandria,
V Tl ly 1 fl 1 £1 •
U.S« Department of the Navy. 1983. Son Dynamics, Deep Stabilization,
and Special Geotechnlcal Construction. NAVFAC DM-7.3, Naval Facilities
Engineering Command, Alexandria, Virginia.
U.S. Environmental Protection Agency. 1978. Electrical Resistivity
5-117
-------
U.S. Environmental Protection Agency. 1982a. Handbook: Remedial
Action at Waste Disposal Sites. EPA-625/6-82-006, Washington, D.C.
U.S. Environmental Protection Agency. 1982b. Surface Impoundments—Liner
Systems, Final Cover and Freeboard Control. Draft RCRA Guidance
Document, Office of Solid Waste, Washington, D.C.
U.S. Environmental Protection Agency. 1983. Groundwater Monitoring
Guidance for Owners and Operators of Interim Status Facilities.
PB83-Z09445, NTIS, Springfield, Virginia.
U.S. Environmental Protection Agency. 1984. Slurry Trench Construction for
Pollution Migration Control. EPA-59d/2-84-001, Washington, D.C.
U.S. Environmental Protection Agency. 1985. Draft Technology Guidance
Document on Double Liner Systems for Landfills and Surface
Impoundments—Design, Construction, and Operation. EPA/530-SW-85-014.
Office of Solid Waste and Emergency Response, Washington, D.C. 71
pages.
U.S. Environmental Protection Agency. 1986a. RCRA Ground-Water Monitoring
Technical Enforcement Guidance Document. OSWER-9950.1, September 1986.
Office of Solid Waste and Emergency Response, Washington, D.C.
U.S. Environmental Protection Agency. 1986b. Technical Guidance Document:
Construction Quality Assurance for Hazardous Waste Land Disposal
Facilities. EPA/530-SW-86-031, OSWER Policy Directive No. 9472.003,
October 1986. Washington, D.C.
U.S Environmental Protection Agency. 1987. Notice of Proposed Rulemaklng:
Liners and Leak Detection for Hazardous Waste Land Disposal Units.
Federal Register, Friday, May 29, 1987, Vol. 52, No. 103.
pp.20218-20311.
Vesillnd, P. A., J. J. Pelrce, and 6. Salfors. 1983. Physical Failures of
Compacted Clay Liners and Caps (draft). Center for Environmental
Engineering, Duke University, Durham, North Carolina.
Voigts, D. L., and E. S. Savage. 1974. Engineering-Approach to a Secondary
Treatment System. Tappl Press Publications. 57(6):96-100.
Wahls, H. E., C. P. Fisher, and L. J. Langfelder. 1968. The Compaction of
Son and Rock Materials for Highway Purposes. Bureau of Public Roads,
Raleigh, North Carolina.
White, R. M., and S. S. Brandweln. 1982. The Application of Geophysics to
Hazardous Waste Investigations. American Defense Preparedness
Association Symposium, Washington, D.C.
WHlenbrock, J. H. 1976. A Manual for Statistical Quality Control of
Highway Construction, Vols. 1 and 2. Purchase Order No. 5-1-3356,
Federal Highway Administration, Washington, D.C.
Winterkorn, H. F., and H.-Y. Fang. 1975. Foundation Engineering Handbook,
Van Nostrand Reinhold, New York.
5-118
-------
Wischmeier, W. H., and D. D. Smith. 1978. Predicting Rainfall Erosion
Losses—A Guide to Conservation Planning. USDA Agriculture Handbook
No. 537.-U.S. Department of Agriculture, Washington, D.C.
Wong. J. 1977. The Design of a System for Collecting Leachate From a
Lined Landfill Site. Water Resources Research. 13(2).
Xanthakos, P. P. 1979. Slurry Walls. McGraw-Hill Book Company, New
York, New York. 621 pp.
5-119
-------
-------
CHAPTER 7
CLAY LINER PERFORMANCE
7.1 INTRODUCTION
Knowledge of the performance of existing liner designs under field con-
ditions is an important factor for evaluating the adequacy and practicality
of various designs for compacted clay liners. As part of this project, an
attempt was made to gather clay liner field performance information. At the
onset of this effort, only clay-lined hazardous waste sites with lysimeters
or leak detection systems were deemed suitable for our purposes. As work
progressed, however, it became apparent that additional sites without
lysimeters or leak detection systems that contained predominantly nonhazard-
ous wastes might also be useful. In addition, several facilities that are
unlined but located in deposits of low-permeability soil were included in the
case study section. The inclusion of such sites is generally for the
illustration of a specific performance problem.
The success or failure of a disposal site depends upon many factors,
most of which are unrelated to the type of waste contained within. Hazardous
and nonhazardous waste sites are probably equally susceptible to the physical
failure mechanisms discussed in previous chapters. This is not the case,
however, where chemical and clay interactions are important; in these cases,
industrial and chemical waste disposal sites have a greater probability of
displaying these effects than do municipal or nonchemical waste sites.
Therefore, by including as many different types of sites (municipal, indus-
trial, and chemical) as possible, problems resulting from most types of
failure mechanisms can be "tested for" through the analysis of the site data.
This chapter discusses the various factors affecting clay liner perform-
ance as well as the techniques and problems of clay Jiner performance moni-
toring. Section 7.2 of this chapter is a detailed discussion of 17 facility
case studies. This section includes such information as the age of the
facility, liner specifications, construction techniques, waste types, local
hydrology and geology, and facility performance. Section 7.3 contains a sum-
mary of the information presented in the case study discussions. The
relationship between the number of successes and failures of the three facil-
ity types (lined with compacted soil, lined with admixed materials, or
unlined) is also presented. Sections 7.4 to 7.7 include discussions on
various factors that influence a facility's performance and methods for moni-
toring that performance. The final section, Section 7.8, presents the
conclusions that can be drawn from the chapter.
7.2 CASE STUDIES
This section contains descriptions of 17 clay-lined facilities listed in
Table 7-1. The data for these facilities were obtained from State and Fed-
eral agencies, commercial waste disposers, clay liner design and construction
7-1
-------
firms, and the industrial sector. Information on approximately 50 other dis-
posal facilit4es was also obtained during this effort and previous data-
gathering efforts. In most cases, either necessary information was not
available or the facility did not have any unique features worthy of dis-
cussion. Several medium-scale field studies also provide an insight into
the performance of field-compacted clay liners (Day and Daniel, 1985; Bagchi,
1987; Rogowski et al., 1985; Rogowski, 1986).
7.2.1 Criteria for Site Selection
Several important sets of data are needed in order to evaluate the
effectiveness of a disposal site. These include:
• Geologic and hydrogeologic site characteristics
• Types of waste in the site
• Geotechnical characterization of the clay
e Leachate collection, leak detection, or groundwater monitoring
data
• Physical description of the site
a Liner description
• Construction procedures.
Many times, while the files on specific sites were reviewed, it became
obvious that critical sets of data were either fragmentary or not available.
One reason for this is that many sites had been "grandfathered in" when the
States started permitting disposal facilities. In this circumstance, impor-
tant information, which is now required before a site can receive a permit,
was never developed or submitted to the States when these older sites were
started. For this reason, even though we know about the existence of many
old clay-lined sites, their Incomplete documentation makes them unsuitable
for our purposes.
In some cases, when sites have been suspected of* causing groundwater
pollution, the State or landfill operator has initiated a hydrogeological
study designed to evaluate the performance of the clay or soil leachate
barrier. Sites where these studies have been performed are among the best
documented and are included in our data base. However, the motivation for
performing these studies was the suspicion of poor performance, and their
Inclusion here may distort our perception of the capabilities of clay liners
in general. We have no way of including the hundreds of undocumented clay-
lined sites that may be performing well.
As one might suspect, the best documented sites are those relatively few
that have been built after the institution of strict permitting requirements
at the State and Federal level. In general, permit files contain hydrogeo-
logical reports and engineering reports that document the installation of the
facility and the characteristics of the clays used. Typically, however,
these sites are only a few years old and therefore do not supply much data
relevant to the prediction of the long-term performance of clay liners.
7-2
-------
TABLE 7-1. CLAY-LINED FACILITY INFORMATION
Site Startup date
Waste type
Liner description
(liner components listed from
the top down)
Facility performance and comments.
CO
1976 NHa and HUb
1971 NH and HW
(including un-
solidified liquids)
1977 . HW (including
organic solvents,
heavy metal sludges
and acids) high-
density polyethylene
1979 HW
Late 1970's HW
1976
HW
1955 HW
Leachate collec-
tion system
added in 1982.
1980
1980
1979
75% MSWC
25% paper mill
sludge
Liquid HW
HW (liquids and
solids)
-Unrecompacted in situ glacial till.
-Zone-of-saturation landfill.
-In situ glacial till. Sand or
gravel seams were excavated and
then backfilled with clay.
-Leachate collection system.
-In situ clay-shale soil contain-
ing calcium carbonate nodules
and seams. Sand or gravel
seams were excavated and then
backfilled with soil.
-Leachate collection system.
-80-mil (HOPE) liner.
-5-foot recompacted clay liner.
-French drain above liner in
each cell.
-Recompacted Demopolis Chalk.
-Leachate collection system
-Flexible membrane liner (FML)
(Hypalone® or HOPE).
-10-foot recompacted clay
liner from borrow site.
-Leachate collection system.
-Unrecompacted glacial till.
-Leachate collection system.
-4-foot recompacted clay liner.
-Three lysimeters.
-5-foot recompacted clay liner.
-Leak detection system.
-5-foot recompacted clay liner.
-5-foot recompacted clay liner.
-Leak detection system.
-1-foot recompacted clay liner.
Original monitoring well installation
problem has been corrected. No
recent performance problems have been
reported.
Bathtub effect has caused inward
hydraulic gradient to reverse.
Severe groundwater contamination
has resulted.
Groundwater contamination has
occurred, possibly due to a
reaction between the low pH waste
and the calcium carbonate inclusions
in the local soil.
No performance problems.
Minor problems, including buildup
of leachate head, have been
corrected.
Leachate removed from collection
system is less contaminated than
the groundwater. Contamination
was caused by an adjacent facility
requiring remedial action.
Contamination in well is thought
to have come from an adjacent
abandoned drum storage facility by
way of a sand and gravel seam.
Initial collection of liquid in
lysimeters thought to be soil mois-
ture release following construction.
Desiccation cracks formed in
unprotected liner prior to filling
with waste. When pond was filled,
waste migrated into the detection
system.
Chlorinated organic liquid placed
in pond caused liner failure.
(continued)
-------
TABLE 7-1 (continued)
Site Startup date
Waste type
1981
Liquid
1978
HW
1977
1974
1980
1980
HU
MSW
HU
HW
1976
Liner description
(liner components listed fro*
the top down)
-3-foot recompacted clay liner.
-Leak detection system.
-1-foot recompacted clay liner.
-Leachate collection system.
-Approximately 2-foot
recompacted clay liner.
-Leak detection system.
-1-foot recompacted clay liner.
-Approximately 1-foot compacted
soil liner.
-Geotextlle.
-Leachate collection system.
-1-foot recompacted borrow clay
soil liner.
-Leak detection system.
-Leachate collection system.
-4-inch bentonlte (averaging
11 percent by weight) and
sand liner.
-Two lysimeters.
-Leachate collection system.
-4-foot-thlck recompacted local
clay Hner with 3 percent benton-
ite and 3 percent lime added.
-Leachate collection system.
-5-foot recompacted soil liner
Including a 1-foot layer of
bentonlte (9 to 12 percent) and
soil.
-Leak detection system.
-6-inch bentonite and soil layer.
-5-foot in situ soil layer.
-Leachate collection system.
-4-inch bentonite/soil liner.
-Leak detection system.
-4-inch bentonite/soil liner.
Facility performance and cownents
Liquid volu«es collected in detection
system were used to calculate liner
permeability. Values ranged from
4 x 10-8 to 3 x 10? cnt/s. Recent
major earthquake 100 miles north of
the facility caused no damage.
Failure in upper liner has occurred.
Lower Hner 1s still functioning.
No performance problems.
Groundwater contamination has been
detected at several monitoring points.
Cap failure caused increased leachate
volumes. Problem was corrected by
replacing section of cap.
Small amounts of liquid have been
collected in the leak detection
system. Monitoring wells have
shown no significant changes.
Liquid volumes collected in detection
system were used to calculate the
liner permeability. Values ranged
from 3 x 10~8 to 6.5 x 10'8 cm/s.
aNH = Nonhazardous.
bHW = Hazardous waste.
CMSH = Municipal solid waste.
-------
The site data presented in the following sections include all of the
relevant information that was made available to the authors for each site.
As discussed in the previous paragraphs, the quality, quantity, and type of
these data were quite varied.
7.2.2 Site A
7.2.2.1 Physical Description—
This sanitary and hazardous waste landfill is located in an area of
glacial till soils. The liner at this facility consists of the unrecompacted
glacial till, which has a relatively low permeability. A plan view of the
facility is presented in Figure 7-1.
7.2.2.2 Startup Date—
This facility began operations in 1976.
7.2.2.3 Local Geology and Hydrology—
This facility is located in the midwestern United States. Average
annual precipitation at this site is 35 inches.
The soil in the vicinity of the landfill is described as a clay-loam
till. Soil borings indicate that it extends to depths of 150 feet. Underly-
ing the till is a thick formation consisting of limestone, dolomite, sand-
stone, and slate. Occasional gravel, silt, or sand lenses are present in the
upper till layers. It is not known if these areas were excavated and back-
filled with clay prior to waste placement.
There is very little information about the groundwater levels in this
area. However, it is known that the groundwater flows in a northeasterly
direction and that the facility is located in an area where the possibility
of groundwater contamination is low.
Monitoring wells are located both upgradient and downgradient of the
site. Samples are analyzed quarterly for abnormalities in groundwater
conditions.
7.2.2A Waste Type--
Both sanitary and hazardous wastes (including drummed combustibles and
Teachable metals) are accepted at this facility.
7.2.2.5 Liner Description—
The liner at this facility consists of unrecompacted in-situ glacial
till. This material has the following characteristics.
» Permeability 3.0 x 10-6 to 5 x 10-8 cm/s
(triaxial tests with
leachate and water)
o Liquid limit 22 - 26%
o Plasticity index 2 - 10%
» Moisture content 12 - 22%
« Amount passing No. 200 sieve 75 - 95%.
7-5
-------
Hydraulic Gradient
Monitoring /
Wells \
Hazardous
Waste
. 1,000 feet
' (approximate)
Figure 7-1. Plan view of site A.
7-6
-------
Compatibility testing performed by a private firm has indicated that the
leachate win not react with the surrounding liner soils in a way that would
significantly increase its hydraulic conductivity. The major organic
components of the test leachate were phenol, methylene chloride, 1,1,1-
trichloroethane, toluene, 1,1-dichloroethane, and diethylphthalate.
7.2.2.6 Liner System Installation—
This is an unlined facility. Methods used for the excavation were not,
available.
7.2.2.7 Performance--
Analysis of groundwater samples taken from the present monitoring wells
has not revealed any leachate migration problems. However, the original
monitoring wells were improperly installed, having contained steel mill slag
where clean gravel should have been used. Samples taken from these wells in
1976 had pH levels between 10.7 and 11.3. The analysis also revealed a
chemical oxygen demand (COD) of 3,193 ppm and a chloride level of 1,028 ppm.
These elevated levels were attributed to the steel-mill slag in the well
casings. Samples taken from newly constructed wells in 1981 had pH levels of
around 7, COD levels ranging from less than 10 ppm to approximately 60 ppm,
and chloride levels ranging between 4 ppm and 12 ppm.
7.2.3 Site B
7.2.3.1 Physical Description—
This 166-acre facility contains an 80-acre zone-of-saturation, or inter-
gradient, landfill. A small section of the landfill has a recompacted clay
liner; the remainder of the landfill is unlined. A leachate collection
system (LCS) lies between the waste and the landfill bottom. A plan view of
the facility is presented in Figure 7-2.
7.2.3.2 Startup Date—
This landfill began accepting waste in 1971.
7.2.3.3 Local Geology and Hydrology—
This facility is located in the northern central portion of the United
States. Average annual precipitation at this facility is approximately
30 inches.
«
The site is located in a glacial till, which consists primarily of clay
and silt. Site investigations have shown that sand and gravel lenses are '
present throughout the till. Underlying the glacial till is a layer of
moderately fractured dolomite up to 200 feet thick. A permeable aquifer
exists in this bedrock layer. The glacial till is saturated to within
10 feet of the ground surface. Groundwater flow was in a southeasterly
direction prior to the development of the disposal facility.
7.2.3.4 Waste Type—•
This facility contains municipal solid waste, nonhazardous industrial
waste, and hazardous waste. Included in this are large amounts of plating
sludge and pickle liquor as well as smaller amounts of various solvents.
Liquid wastes were not solidified prior to their disposal.
7-7
-------
Monitoring Wells
VJ
00
Leachate
Seeps
Old Section
No Data
Figure 7-2. Plan view of site B.
-------
7.2.3.5 Liner Description—
«.... The.u 11nfr at this facility is a minimum of 30 feet of in situ glacial
till. The glacial till has the following characteristics: 9'acnai
Soil Characteristics Average Value
• Amount passing No. 200 sieve 68%
• Clay 4Q%
• Liquid limit ' 29%
• Plastic index . 14%
• Field permeability 7.5 x 10-6 cm/s
t Laboratory permeability 9 x lO"8 cm/s.
Lenses of sand or gravel that were encountered during site construction were
excavated and backfilled with a minimum of 5 feet of recompacted clay. A
leachate collection system is located on top of the landfill base. Specifi-
cations for this system were not available.
7.2.3.6 Liner System Installation--
No information on the excavation and construction of this facility was
aval labl e.
7.2.3.7 Performance—
Recently, three major and numerous minor leachate leaks have been
observed in the western side of the final cover (see Figure 7-2) The statP
?,eni??!?tS Sttr1bute these 1eaks to the 26-foot hydraulic head w thin the
landfill One geologist estimated that, at times, the head is as great Is 40
feet. This high hydraulic head has caused the inward gradient of this zone-
of-saturation landfill to reverse.
in*-« JhfeC°nd Prob.1em.enco"ntered at this site is the migration of leachate
into the groundwater in a very small area on the northern edge of the land-
fill (see Figure 7-2). Excavations in the problem area revealed that in one
5h2 ilP?T*Ie dep2Slt £hat had been 1mPr°Perly sealed with clay was belSw
2! JXeV! ?he Waste4mater1al • Leachate entered this permeable layer and
was detected n a monitoring well. Remedial action consisting of the instal-
lation of a clay cutoff wall and leachate removal has resulted in some
groundWater 9ual1ty; however, significant contamination is
7.2.4 Site C
7.2.4.1 Physical Description—
«« H The JaS1l1ty consists of. 31 small drum disposal trenches, 4 treatment
ponds, and 2 evaporation ponds (see Figure 7-3). The total storage and
disposal area is approximately 16.5 acres.
The ponds and trenches were excavated in the local natural soil If
sand seams or other potentially troublesome areas were located in the pond
7-9
-------
a
a
a
-s]
l-»
o
Evaporation
Ponds
D
D
Treatment
Ponds
Disposal
Trenches
N
a
•$• Monitoring wells in the upper
water-bearing zone
O Monitoring wells in the lower
water-bearing zone
Approximate Scale
I I 1
0 200 ft 400 ft
Figure 7-3. Plan view of site C.
-------
bottoms, these areas were recompacted. No recompacting was done 1n the
disposal trenches. Several monitoring wells surround the facility.
7.2.4.2 Startup Date—
The facility was opened in 1977, but because of groundwater contamina-
tion it was forced to close in January 1982. Since that time, changes have
been made and parts of the facility are again in operation.
7.2.4.3 Local Geology and Hydrology—
This facility is located in the midwestern plains of the United States.
Average annual precipitation in this area 1s approximately 25 inches.
Investigations have indicated the presence of two water-bearing zones
beneath the facility at depths of approximately 40 and 50 feet below the
ground surface. Several springs and a creek are to the north of the facil-
ity. The relationship between the groundwater zones and springs is unknown.
Soil borings indicate that approximately 3 feet of topsoil overlie 30 to
40 feet of weathered clay-shale. This weathered clay-shale is a plastic
silty to highly plastic clay (CL to CH) with calcium carbonate nodules and
seams throughout. Bedrock lies under the clay-shale layer. An example of a
typical soil boring is illustrated in Figure 7-4.
7.2.4.4 Waste Type-
Various hazardous wastes including organic solvents, heavy metal
sludges, and acids were held in the treatment and evaporation ponds and
disposed of in the drum storage area.
7.2.4.5 Liner Description-
While this facility is almost completely unlined, potential problem
areas uncovered by the pond excavations, such as sand seams, were
recompacted.
The local clay-shale soil was the only barrier to waste migration. A cross
section of the facility is presented in Figure 7-4.
7.2.4.6 Liner System Installation-
Information about the procedures used to excavate and recompact the
disposal areas was not available. Methods used to install the monitoring
wells were also not available.
7.2.4.7 Performance—
Analysis of groundwater samples indicates that contamination has
occurred. Elevated concentrations of chromium, arsenic, barium, cadmium,
lead, and mercury were detected near the disposal ponds and trenches.
However, except for chromium, none of the heavy metals were detected in
offslte monitoring wells. In addition to heavy metal contamination, several
volatile organics such as chloroform, trichloroethane, dlchloroethane,
benzene, and methylene chloride were detected in both onsite and offsite
monitoring wells. The highest levels of volatile organics were found onsite
in the uppermost water-bearing zone in wells closest to the disposal areas.
Base-neutral-extractable organics and acid-extractable organics were also
detected in onsite and some offsite wells (see Table 7-2).
7-11
-------
Evaporation
Pond
Treatment
Pond
Disposal
Trench
VI
i-»
to
Upper Water-
Bearing Zone
Lower Water-
Bearing Zone
40ft
Topsoil
Clay-shale
soil with
calcium carbonate
nodules and
seams
Bedrock
Figure 7-4. Cross-sectional view of site C (vertical dimensions are to scale).
-------
TABLE 7-2. GENERAL OCCURRENCE OF CHEMICAL PARAMETERS IN THE GROUNDWATER AT SITE C
No.
Parameter parameters
tested for tested for
Volatile organics-
No. found 31
Highest
concentration
Trace metals-
No, found 8
Highest
concentration
Acid-extractable
organics
No. found n
Highest
concentration
Base-neutral- t
extractable organics
No. found 46
Highest
concentration
No. parameters
Onsite
detected
Upper water- Lower water-
bearing zone bearing zone
20
370,000 3
8
20,000
6
24,000
11
410
19
,100
8
500
ND
ND
3
83
and highest concentration
Off site
Upper water-
bean" ng zone
18
±\j
79,000
8
\j
300
3
3,300
6
250
(ppb)
Lower wat,er-
bearing zone
£
D
510
7
/
500
MD
IMU
ND
i
J.
160
ND = No parameters of this type were detected.
-------
Investigations of the groundwater contamination at this facility suggest
that the treatment ponds are the major source of contamination 1n the upper-
most water-bearing zone. This conclusion is based on the fact that the
materials 1n the ponds were usually of low pH and would be reactive with the
carbonate seams and inclusions in the surrounding soil. Several of the
trenches were used for the disposal of highly acidic waste oil reprocessing
sludges. These sludges may have also attacked the surrounding materials and
allowed contamination to occur.
7.2.5 Site D
7.2.5.1 Physical Description--
This recompacted clay- and synthetic-lined hazardous waste disposal
facility consists of two large disposal sections, each of which is divided
into several smaller disposal cells. These in turn are divided into three
subcells each for the segregation of different waste types. Four of the five
disposal cells in Section I are closed, and the remaining cell is currently
active. The first of 10 proposed disposal cells in Section II 1s under
construction. This construction is scheduled to be completed in the fall of
1984, at which time landfill ing operations will be initiated. A plan view of
the facility is presented in Figure 7-5.
7.2.5.2 Startup Date--
Construction of this facility was started in 1979.
7.2.5.3 Local Geology and Hydro!ogy~
This facility is located in the southeastern United States. Average
annual precipitation at this site is approximately 47 inches.
Information on the local geology and hydrology was obtained through the
use of 32 exploratory borings. These borings have confirmed that there are
several distinct soil layers. The lowest of these layers 1s the Tuscaloosa
Formation, which is overlain by the Black Mingo Formation, which in turn is
overlain by Quaternary deposits. The upper surface of the Tuscaloosa
Formation 1s located approximately 55 to 120 feet below the ground surface-
This material consists of clayey sands and clays. Lenses or beds of coarse
to fine sand are located in this formation. In situ hydraulic conductivity
testing of this formation revealed that it has a permeability 1n the
range of 1.09 x 10~6 to 4.01 x 10~6 cm/s. *
The Black Mingo Formation encountered at this site consists of four
subunlts: the Basal Clay Unit, the Opal Claystone Unit, the Red Sand Unit,
and the Buhrstone Unit.
The Basal Clay Unit consists of very hard black clay with frequent sand
and silt lenses. The upper surface of this material is located approximately
60 to 115 feet below ground level and varies in thickness from 1.5 to 15.5
feet. In a few of the borings, this material was not present.
The Opal Claystone Unit, on the other hand, is either above the
Tuscaloosa or Basal Clay Formations over the entire site. This material con-
sists of low-permeability (exact value unknown) silt and clay with very few
sand Inclusions. The thickness of this formation varies from 20 to 60 feet.
7-14
-------
VI
l-»
U!
Section II
Direction of
Ground and
Surface Water
Flow
Clay stone
Processing
Facility
Drum Storage and
Waste Handling Facility
Section I
N
Scale
I
400'
I
800'
Figure 7-B. Plan view of site D.
-------
The Red Sand Unit, which is located above the Opal Claystone Unit,
consists of sand to clayey silts. The silts and clays are generally present
as 1/4-inch"lo 2-inch layers interbedded with the sand. The thickness of
this formation varies from 0 to 15 feet. In situ hydraulic conductivity
testing of this material revealed that its permeability ranges from
6.24 x 1C-5 to 6.43 x 10'4 cm/s.
The final unit of the Black Mingo Formation, the Buhrstone Unit, is
found scattered over the entire area. When present, it ranges up to 14 feet
thick. This material is described as clayey silt to silty clay.
The Quaternary deposits contain layers of variegated clay that were
originally thought to be a source of low-permeability liner material.
Further investigations revealed that the silt and sand content was much too
high for it to be used as a base liner.
An aquifer is present in both the Tuscaloosa and the Red Sand Forma-
tions. The Tuscaloosa aquifer is approximately 60 to 70 feet below the
ground surface. The Red Sand aquifer is roughly 40 feet above this aquifer.
The direction of flow in both cases, as well as the direction of surface-
water flow, is toward a lake located southwest of the facility.
7.2.5.4 Waste Type—
This facility accepts most types of hazardous wastes. These wastes are
classified as alkaline, acid, or organic and are disposed of with similar
wastes only. This is accomplished through the use of three subcells that are
incorporated Into all of the cells in Section II of the facility. The size
of each of the subcells was determined so that maximum separation of
potentially incompatible wastes was provided and based on past operating
experience.
7.2.5.5 Liner Description—
The liner at this facility consists of 5 feet of recompacted clay
(maximum permeability of 1 x 10-' cm/s) brought to the facility from a
borrow pile located several miles away. Ten-foot sidewall liners are present
above the top of the Opal Claystone Unit (see Figure 7-6). An 80-m1l high
density linear polyethylene (HOPE) membrane liner lies on top of the
recompacted clay. This liner is protected with a soil layer that is 9 inches
deep on the bottom and 24 inches deep on the sidewafls. A leachate collec-
tion system 1s located on top of the protective soil layer (see Figure 7-6).
The subcell separation berms are constructed of recompacted clay and are
built up in 5.5-foot-high sections as the landfill operations progress. The
side slopes of these berms are 1:1.
7.2.5.6 Liner System Installation-
Prior to excavation of the landfill cells, a channel to drain the
perched water table within the Red Sand Unit was constructed. After this
channel was completed and in operation, excavation of the landfill cells was
Initiated. Side slopes were cut on a 3:1 (horizontal:verticle) slope down to
the top of the Opal Claystone Unit. At this location, a 17-foot-wide "bench"
with a 3-foot-wide trench was excavated into the top of the opal Claystone
(see Figure 7-6). This area was excavated to provide room for a thicker
Uner. Below the "bench," side slopes were also 3:1. A minimum of 10 feet
of the opal Claystone was left as foundation material. At the leachate
7-16
-------
10 ft Recompacted
Clay Liner
Side Wall
1 Slope
I
}-»
VI
In Situ
Opal Claystone
24 in Protective
Soil Cover
80 mil HOPE
Liner
1-ft Drainage
Layer
Typical 5.5 ft-High
Section of a Cell
or Subcell Separation
Berm
.,*-
9-in Protective
Soil Cover
4-in Perforated
Collection Pipe
rr*
5 ft Recompacted
Clay Liner
\\ // \\ // \\ // \v\//
\
Filter Fabric
\\
Figure 7-6. Cross-sectional view of site D liner.
-------
collection system sump locations, approximately 4.5 feet of opal claystone
remained as the foundation.
After the sides and base of a cell were graded, clay was brought to the
site from the borrow area. Before the clay was placed in the excavation, a
"test fill" was constructed with borrow soil that was selected based on the
basis of laboratory screening procedures. The "test fill" was constructed in
the field with equipment and procedures that were to be employed during liner
construction. Undisturbed Shelby tube samples were taken from the completed
"test fill" and were tested in the laboratory for density, moisture content,
and permeability. The clay was compacted in 6- to 8-inch lifts with either a
sheepsfoot or smooth-drum vibratory roller. The liner specifications
required a permeability of 1 x 10~7 cm/s or less. Testing results from
the construction quality assurance (CQA) tests on the installed liner were
not available; however, results from tests performed on Shelby tube samples
taken from the "test fill" were as follows:
a Density 114.9 - 120.0 lb/ft3
• Water content 22.6 - 29.8%
• Permeability 1.8 x 10~8 to 8.0 x ID'8 cm/s.
7.2.5.7 Performance--
Groundwater samples taken from upgradient and downgradient monitoring
wells have not been statistically different since the start of waste
placement.
7.2.6 Site E
7.2.6.1 Physical Description—
The facility presently occupies approximately 100 acres, with room for
expansion to 340 acres. There are over 20 disposal trenches at various
stages of operation as well as 4 active evaporation ponds (see Figure 7-7).
7.2.6.2 Startup Date—
This facility became active in the late 1970's.
7.2.6.3 Local Geology and Hydrology—
This facility is located in the southeastern United States. Average
annual precipitation at this facility is^50 inches.
It is located in a natural clay formation called the Demopolis Chalk
Formation. Testing of the Demopolis Chalk Formation has shown that it has
the following properties:
t Average permeability 4.0 x 10"8 cm/s
a Permeability to polychlorinated biphenyl (PCB) liquids:
- Light oils and PCB's, 9 x lO'9 cm/s
7-18
-------
-4-
Key
N
Active Impoundment
or Trench
Completed Impoundment
or Trench
Monitoring Well
200 ft (approx.)
-4-
o
o
O
i I
Active
Active
o
Q.
O
Proposed
Figure 7-7. Plan view of site E.
7-19
-------
- Heavy oils and PCB's, 1.5 x 10-10 cm/s
• 88% of the material passes the No. 200 sieve
t 35% of the material is less than 0.001 mm in diameter
• Liquid limit of 31%
t Plastic limit of 20%.
Soil borings in the area have shown that the clay is a minimum of 500 feet
thick and in some places extends to depths of over 700 feet.
The closest usable groundwater is located below this clay formation.
However, a few locations of perched water are in the vicinity of the site.
In addition, the Demopolis Chalk Formation is saturated to within 10 to
30 feet of the ground surface.
The 100-year projected flood elevation for a creek, located 3,000 feet
north of the facility, is 125 feet above mean sea level. This should present
no problem to the facility because the disposal trenches are located at
points that range from 180 to 220 feet above mean sea level. The lowest
elevation of the site is 148 feet above mean sea level.
7.2.6.4 Waste Type--
The site is designed to be a full-service facility that accepts all
types of hazardous wastes. The majority of the material received for dis-
posal consists of acidic and caustic Industrial wastes, heavy metal sludges,
contaminated containers, sludge, and oil sludges. All wastes received by the
facility are handled separately, and only compatible materials are placed
within proximity of each other.
PCB wastes are handled separately and only PCB's and PCB-compatible
wastes will be located in the same disposal trench.
7.2.6.5 Liner Description--
Each disposal trench is lined with recompacted native soil. Construc-
tion information concerning the number and thlckness.of lifts was not
available. A calculated amount of absorptive material is placed on top of
the recompacted bottom of each trench» The wastes are then placed on top of
this material and covered with 6 to 8 Inches of the natural clay. The final
trench covers are also constructed with natural clay.
The current monitoring system consists of the following:
• A subsurface collection system and a secondary retention dam.
This system would collect any materials that might migrate.
• Several monitoring wells placed within proximity to active
trenches.
• Observation wells surrounding the entire facility.
7-20
-------
A French drain system is located at the lowest point in each trench. In
the event that liquid is collected in this drain, it would be pumped out and
treated as incoming waste.
7.2.6.6 Liner Installation--
No information was available.
7.2.6.7 Performance--
In general, this facility is operating as planned. There have, however,
been a few minor problems. Groundwater from a more permeable weathered sec-
tion of the Demopolis Chalk Formation has infiltrated the sidewalls of
trenches 15 and 19. At trench 15, this problem was remedied by installing a
cutoff wall on the south-southeast side of the trench. Recompaction of the
entire sidewall at trench 19 eliminated the groundwater flow into this
trench. More recently, large volumes of leachate (up to 50 feet deep) have
been accumulating in some of the trenches with interim final covers. These
large volumes of leachate are attributed to the infiltration of groundwater
and precipitation. Frequent pumping of the leachate has helped to reduce
these volumes. Final covers consisting of 3 to 5 feet of recompacted
Demopolis Chalk and a HOPE liner are scheduled to be installed by the end of
1986. In addition, the State monitoring well analysis has shown trace levels
(92 ppb maximum) of 12 organic compounds in groundwater samples. An investi-
gation of their source indicates that they may have come from the well casino
materials.
7.2.7 Site F
7.2.7.1 Physical Description—
The 380-acre facility consists of five secure land disposal cells with a
total area of 25 acres. Cells 1, 2, and 3 are closed; cell 4 is currently
being filled; and cell 5 is under construction. The remainder of the facil-
ity is devoted to sanitary landfill ing, waste treatment operations; and
recovery of lime from the old slag disposal operation.
The five cells are lined with 10 feet of recompacted clay soils. On top
of the clay liner is a flexible membrane liner (FML), which in turn is
covered with a 1-foot layer of recompacted clay. A layer of geotextile lies
above the 1-foot clay liner. Finally, a 12-inch drainage blanket containing
slotted 4-inch corrugated HOPE pipe at a minimum of tOO-foot intervals is
installed on top of the geotextile (see Figure 7-8).
7.2.7.2 Startup Date—
The facility is located at the site of a slag disposal operation that
has not been active for 90 years. The clay-lined facility as it currently
exists has been in operation since 1976.
7.2.7.3 Local Geology and Hydrology--
This facility 1s located in the northeastern United States. Average
annual precipitation in this area 1s approximately 36 Inches.
Numerous onsite borings and test pits have revealed that the site 1s an
old disposal area covered with waste industrial slag ranging in depth from
approximately 7 feet to 45 or 50 feet. The disposed slag fill ranged in size
7-21
-------
VI
Flexible
Membrane
Liner
(Not to scale)
Interior
Clay Dike
1-ft Drainage Layer
Slotted 4-inch
Geotextile Wrapped
Leachate Collection Pipes
\\
1 to 2 ft Compacted Cover Soil
10-ft Compacted Clay Liner
\\ // \\ // \\ // \\ // \\ // \\ // \\
Figure 7-8. Cross-sectional view of site F.
-------
from silt-sized particles to boulders. Beneath the slag was a layer of marsh
silt that ranged in thickness from a few inches to as much as 5 feet.
Below the slag and marsh silt materials covering the entire site was a
6-foot layer of lacustrine clays. These clays contained lenses of gray
nonplastic silts and clay with minor amounts of fine sand. The presence of
these lenses throughout the lacustrine clay eliminated it from consideration
as an in situ liner material. Laboratory analysis of the lacustrine clav
indicated that it had the following properties:
• Natural water content 26.5 - 42.8%
• Liquid limit . 50%
• Plastic limit 22%
• Dry unit weight 99 ib/ft3
• Permeability (laboratory 1 x lO"8 to 6 x 10~8 cm/s
compacted)
• Amount passing No. 200 99%
sieve
• Amount passing 2 microns 54 - 63%.
Underlying the lacustrine clay was a 5-foot layer of glacial till
consisting of brown to red-brown silts, clays, and sands with varying amounts
of gravel. Laboratory testing demonstrated that the permeability of the
till was 3 x 10-' cm/s, making it unsuitable for use as a liner material.
Groundwater under the facility occurs in three zones:
• Unconfined water table above the lacustrine clays
t Confined aquifer in the upper 10 feet of bedrock
• Immobilized groundwater held within the impermeable confining
beds.
Piezometers installed in the soil-boring holes indicated that the
groundwater elevations ranged from 0.5 to 6 feet below the ground surface.
7.2.7A Waste Type-
Each of the five disposal cells are, or will be, divided into four or
five subcells. The purpose of these subcells is to isolate the various waste
groups accepted at the facility, thereby preventing the interactions of
incompatible wastes. When five subcells are used, as in cells 1, 2, 3, and
4, the waste categories are as follows:
« General wastes. These wastes represent approximately 44 percent
of the total waste volume. General wastes are defined as
7-23
-------
materials of both an organic and inorganic nature that do not
contain a significant quantity of any of the other waste
categories.
The hazardous acidic or acid-generating materials are
covered with lime to ensure that any acid that is generated will
be neutralized.
• Pseudo metals. This type of material represents approximately 6
percent of the total waste volume. Pseudo metals are arsenic,
antimony, bismuth, and phosphorous. Chalcogens, beryllium, and
any of their compounds as well as alkaline-sensitive materials
are also disposed in this subcell.
This subcell has a pH buffer system that maintains pH
levels between 6 and 8.
• Heavy metals. These wastes represent approximately 15 percent
of the total waste volume. This group is comprised of all heavy
metals and asbestos. This subcell contains the smallest amount
of organic materials, which helps to reduce fire hazards caused
by the reaction of strong oxidizing agents with organics.
• Highly flammable wastes. This type of material represents
approximately 12 percent of the total waste volume. These
materials generally exhibit a flash point between 80 and
100°F. These materials are kept apart from powerful oxidiz-
ing agents, materials that are prone to spontaneous heating, or
materials that react with air or moisture to evolve heat.
• Toxic materials. These wastes represent approximately 23 per-
cent of the total waste volume. Included in this category are
all highly toxic organic compounds, carcinogens, PCB's, and
other halogenated wastes. No solvent-type wastes were permitted
in this subcell.
If only four subcells are used, the psuedo-metals subcell is eliminated.
7.2.7.5 Liner Description—
The liners at the five secure land disposal cells are very similar.
Minor design changes have been made as each cell was constructed. The
components of the liners include:
e A minimum of 10 feet of local and borrow soil placed above the
in situ glacial till. The soil is compacted so that the
permeability is no greater than 1 x 107 cm/s.
t An FML. Both Hypalon® and HOPE ranging from 30 to 80 mils have
been used.
• Cover soil of 12 to 24 inches compacted to a permeability of
1 x 10~7 cm/s or less.
7-24
-------
t In some of the cells, an 8-Inch layer of slag material placed
over_the cover soil. This layer 1s to provide a firm base for
equipment movement.
• A leachate collection system located on top of the liner system.
This system consists of a 12-inch drainage blanket, slotted
4-inch geotextile wrapped pipes located 100 feet apart (maximum),
and standpipes or manholes to enable collection and removal of
the leachate.
7.2.7.6 Liner System Installation--
Prior to liner installation, all of the industrial slag material and
the organic marsh silt were removed. The lacustrine clays were not excavated
because of the high water table. Dewatering was conducted as necessary.
Borrow liner material was transported to the facility. Prior to liner
compaction a test patch was required. This patch enabled the construction
contractor to determine the compactive effort necessary to achieve the
required permeability. Density, moisture content, and permeability measure-
ments were conducted on the test patch to establish the ability of the equip-
ment to compact the liner to the design specifications.
It was determined that the liner should be compacted 1n 6-inch lifts
with a sheepsfoot roller. The finished clay liner thickness was 10 feet.
The sidewalls were Installed in horizontal lifts and cut back to a final
slope of 2:1. The slope of the bottom liner ranged from 1 to 2 percent. The
clay liner was protected by the FML. The entire liner was Installed before
waste placement was initiated.
A system of dikes was used to separate the large cell into the various
subcells. The interior dikes were constructed in 4-l/2-foot-h1gh sections.
New sections of the interior dikes were constructed by keying into the too of
the previous dike (see Figure 7-8).
CQA consisted of soil density and moisture content measurements (nuclear
gauge) and laboratory permeability tests on undisturbed Shelby tube samples
of the liner.
7.2.7.7 Performance--
The performance of this facility is very difficult to determine at this
time. A potential Superfund site located next to this facility has caused
extensive groundwater contamination in the area. For this reason, baseline
groundwater data have not been obtained. However, it 1s known that the
leachate removed from the collection system of the newly constructed facility
Is less contaminated than the local groundwater.
7.2.8 Site G
7.2.8.1 Physical Description—
The facility consists of several disposal areas covering 80 acres. A
100-acre section is planned for future development. Some sections of the
landfill are active, while others are capped (see Figure 7-9). Each disposal
area is lined and/or capped with unrecompacted local soil. A leachate
7-25
-------
i
f\3
O)
£~~~- Intermittent Creek
r
-it-
Leachate Collection System
Old Drum
Storage Area
Leachate Collection Tank
0-
Monitoring Well
N
Figure 7-9. Plan view of site G.
-------
collection system was installed in 1982. Several monitoring wells are
located within the area.
7.2.8.2 Startup Date--
This site was first used as a dump in 1955. Sanitary landfilling opera-
tions began in 1967. A leachate collection system was added in 1982.
7.2.8.3 Local Geology and Hydrology—
This facility is located in the midwestern United States. Average
annual precipitation at this facility is approximately 35 inches.
The soil in the vicinity of the landfill is described as silty clay to
clayey silt, which extends to a depth of 200 feet. A layer of shale under-
lies the till material. Discontinuous sand and gravel seams are located
throughout the area. When encountered, they have been removed prior to
landfilling.
Groundwater is known to occur in sand and gravel seams within 6 feet of
the surface. The static water level in the clay soil is not known.
7»2.8.4 Waste Type—
The facility has accepted various types of hazardous wastes including
organics, heavy metals, and pesticides.
7.2.8.5 Liner Description—
The local soil (unrecompacted) used to line the site has the following
physical properties:
• Permeability 1 x 10-7 to 1 x 10-9 cm/s
• Natural moisture content 9 to 12%
• Plastic limit 20 to 21%
• Liquid limit 11 to 13%
• Plasticity index 8 to 10%
«
• Cation exchange capacity 1.2 to 2.0 (milleq./lOO g sample).
7.2.8.6 Liner System Installation—
This information was not available.
7.2.8.7 Performance--
Analysis of groundwater samples from well No. 1 (see Figure 7-9) has
indicated high levels of lead, chloride, COD, total dissolved solids,
chloroform, and various chlorinated organics.
This contamination is thought to come from an old drum disposal site
previously owned by another company located adjacent to the landfill. The
drum disposal site is located on a sand and gravel seam that extends upward
near the ground surface. Contaminants from this site can reach this
permeable area and travel southward to Site G's monitoring well No. 1.
7-27
-------
Therefore, Indications are that the contamination 1n well No. 1 has not come
from Site G._
7.2.9 Site H
7.2.9.1 Physical Description—
This site has a rectangular 9-acre cell lined with 4 feet of locally
obtained clay that was put down in four lifts. A leachate collection system
composed of 8-inch perforated polyvinyl chloride (PVC) pipe is on top of the
clay liner. Directly below the clay liner are two lysimeters. A third
lyslmeter is located under the leachate storage basin (see Figure 7-10).
The present cell is filled to near capacity, and a new cell is being con-
structed 1n such a manner that the bottom clay liners will be tied together
to form one large cell with no vertical wall dividing the old and new
disposal areas. The presence of lysimeters almost directly below the
leachate collection drain pipes presents a unique opportunity to evaluate the
performance of the intervening layer of clay.
7.2.9.2 Startup Date--
Operation began in December 1980.
7.2.9.3 Local Geology and Hydrology—
The facility 1s located in the northern central United States. Averaqe
annual precipitation in this area is 30 inches.
The site 1s underlain by approximately 50 to 100 feet of dense silty
sand. This material contains boulders, cobbles, and pockets of sandy
material. The permeability of this glacial till ranges from 1 x 10-* to
9 x 10-' cm/s. Bedrock is encountered beneath this layer. The water
table 1s at a depth of 50 to 80 feet In the silty sand or bedrock. Ground-
water flow is primarily in the southeast direction.
7.2.9.4 Waste Type--
Wastes accepted at this facility consist of approximately 75 percent
municipal and 25 percent paper-mill sludge. Small quantities of other
Industrial wastes are also disposed of at this facility.
7.2.9.5 Liner Descrlption—
The liner system consists of 4 feet of recompacted clay underlain by two
collection lysimeters. The clay for the liner was obtained from a' nearby
borrow area. It consists of a reddish-brown to reddish-gray sllty clay. The
specifications for the material used were as follows:
• Liquid limit >30%
§ Plasticity Index >15%
• P-200 >50%
• Permeability
-------
1
*•*
Under
Construction
o
1
1L1
\
Active
i
i
\
\
U— Drain
i Pipe
i
I
|t2 ^
^s J
—":.•'..'. ...•'.•- - - . . ^*^
a i
^
'
-,
^
ft
-
'
-
1
Road
/
P
1 _»
rn no" r-
j I
Leachate
Storage
Basins
•f
M
Single Monitoring Well
-<>- Monitoring Well Nest
a Lysimeter Sump
o Piezometer
^=» Lysimeter Area
Figure 7-10. Plan view of site H.
7-29
-------
Silty Sand
Protective
Layer
Sand
1% Slooe \
' >
T
t
5'
ElS
x 8" Perforated Pipe Le
X \*% ;^****?T * , * ";"
a-^^ t
4'
Clay Liner i
2 . 1
lachatc
Rock
\\
Lysimeter
20 mil PVC Sheeting
\\
Sand
4" Perforated Pipe
Figure 7-11. Cross-sectional view of site H liner showing details
of leachate collection system and lysimeter construction.
7-30
-------
The leachate collection system consists of a 1-foot trench excavated
into 5-foot-thick sections of the clay liner. Eight-inch perforated PVC pipe
is positioned in the trench, which is backfilled with 3/4-inch crushed rock.
The entire trench is overlain by a 2-foot-high mound of sand. A 1-foot-thick
layer of silty sand covering the entire bottom of the landfill serves as a
protective layer for the clay liner (see Figure 7-11).
The lysimeters located under the clay liner are approximately 12 feet by
100 feet. They are lined with 20-mil PVC sheeting and filled with medium to
coarse sand. Four-inch perforated PVC pipe positioned in the bottom of each
lysimeter (see Figure 7-11) connects them to a manhole, thus enabling
leachate detection and collection.
7.2.9.6 Liner System Installation--
Prior to clay liner installation, the fill area was rough graded. Clay
was brought to the facility from a nearby borrow area. The variable nature
of the borrow area required a soils technician to be present during all
removal operations to ensure that the material met the project specifica-
tions. The technician rechecked the material by performing the required soil
tests as the clay was emplaced. The liner was compacted in 12-inch lifts
with either a rubber-tired or sheepsfoot roller.
The construction quality control (CQC) program called for 29 Shelby tube
samples taken at various locations in each of the four 1-foot-thick clay
lifts. These samples were tested for various parameters. The results of
these tests were:
• Permeability 7 x 1Q-8 cm/s, 5 x 10-10 Cm/s
0 Liquid limit 39 - 82%
0 Plastic limit 20 - 37%
0 Plasticity index 16 - 54%
0 P-200 >50%
0 Density 92 - 102.8% of maximum dry density
(nuclear gauge)
95 - 110.8% of maximum dry density
(sand cone).
7.2.9.7 Performance--
Quarterly sampling of all monitoring points began in December 1980.
This sampling included determining leachate volumes and composition,
lysimeter liquid volumes and composition, and groundwater composition. The
liquid volumes collected from both the lysimeters and the leachate collection
system are shown in Table 7-3.
7-31
-------
TABLE 7-3. LYSIMETER (L) AND LEACHATE COLLECTION SYSTEM (LCS)
LIQUID VOLUMES (GAL) AT SITE H
Date
Lia
12
L3
LCSb
12/80
38
66
Dry
3/81
24
9.4
1
NA
6/81
1.9
0.6
Dry
NA
9/81
0.03
0.03
1
89,603
12/81
Dry
Dry
0.75
NA
3/82
Dry
Dry
Dry
NA
6/82
Dry
Dry
Dry
32,995
9/82
Dry
Dry
Dry
63,321
12/82
Dry
Dry '
Dry
89,678
12/83
1 —
Dry
Dry
Dry
NA
<*Lysimeter 1 is located below the leachate holding pond.
bThe numbers indicate the total monthly volume of leachate that was pumped from the collection system at
irregular intervals.
CO
N>
NA = Data not available.
-------
As Table 7-3 Indicates, the volume of liquid collected 1n Ivsimeters ha
declined steadily over time. The liquid that'was " iXlTJ cSl ecS ?n the
lysimeters may be soil moisture released after site construction.
T ^ Analysis of leachate and groundwater samples is presented in Table 7-4
To date, there is no apparent difference in water quality between the base-
line groundwater samples and the more recently taken samples from upgradient
and downgradient monitoring wells. uHaiauient
7.2.10 Site I
7.2.10.1 Physical Description—
a* * Taln/??111^ conslst! °f three clay-lined surface impoundments as well
rJJLl! 2 I Jhe P0^ that c°ver approximately 8 acres are lined with two
compacted clay liners that are separated by a leak detection system.
7.2.10.2 Startup Date—
fan «2n?™!!Ct1Sn !f the t?ree ponds at th1s fac11ity was completed in the
S1U • ??' Jf- S WaS> Placed 1n two of the ponds that fa11 but was not
placed in the third pond until the following spring.
7.2.10.3 Local Geology and Hydrology —
c* 4. Tn1snfac11ity Is located in a semiarid section of the western United
States. Average annual precipitation at the facility is approximately
lo inches. J
Approximately 75 percent of the site area was used for the land applica-
tion of sewage sludge. This material was disked into the top layer of so}}
7°f2ro? ?^?f J3yr °f -I9"1* °rgan1c topso11' Below tn1* layfir lies 5 to
7 feet of residual clay soil underlain by 5 to 20 feet of sandstone, which
of clay stone bedrock- The
o Liquid limit 36 to 57 (average 45) %
« Plasticity index is to 27 (average - 21) %
i> Natural water content 14.6 - 26.1 (average - 23) %
• Dry density 80.9 (1 sample) Ib/ft3
' PH 7.5 - 7.7 (average - 7.6)
• Unified Soil Classification CH, CL, and CL-CH.
System designation
The groundwater at the site ranges from 100 to 200 feet below the
surface. Drainage across the site is to the east.
7.2.10.4 Waste Type—
n»Ci--I?S thrK6 E°ndf nave received most types of hazardous waste, except
pesticides, herbicides, PCB's, dioxin, reactive materials, and any material
7-33
-------
TABLE 7-4. MONITORING DATA FOR SITE H
Parameter
Sample date
Chloride (mg/L)
COD (mg/L)
pH
Alkalinity, total
(mg/L)
•jj Conductivity
u (micro/cm)
Total hardness
CaCO (mg/L)
Iron, dissolved
Baseline
groundwater
analysis
6/80
1.5 -
4.0 -
7.2 -
120 -
300 -
122 -
0.49 -
5.4
21.0
7.6
240
385
228
2.65
Upgradient
well
9/81
2.5
11
7.4
206
405
222
0.06
Range for
all other
wells
9/81
1 -
4 -
7.2 -
150 -
320 -
162 -
0.01 -
3
50
7.4
168
350
182
0.09
Leachate
6/81
230
5,315
6.1
960
5,000
3,650
161
Range for all
lysimeters
i —
6/81
16 -
40 -
7.1 -
291 -
625 -
344 -
0.06 -
19
57
7.2
445
840
480
0.20
-------
containing more than 5 percent solvents. Pond No. 1 was the only pond to
receive oily-wastes.
Wastes were placed in the ponds through a trough leading to the base of
each pond.
7.2.10.5 Liner Description—
The liner at this facility consists of two 5-foot layers of recompacted
local clay soil that are separated by a 15-inch leak detection layer. The
three ponds are separated by dikes that are 12 feet wide at the top (see
Figure 7-12). The interior slopes of the dikes (sidewalls) are 3:1. A
cross-sectional diagram of the liner system is illustrated in Figure 7-13.
Properties of the local clay soil that was used as the liner material
are listed in Section 7.2.10.3. Additional laboratory testing of this
material indicated that it had a permeability ranging from 2 x 10~9 to
4.8 x 10-° cm/s. ^e value for the permeability of the claystone
material ranged from 3 x 10~9 to 1 x lO'8 cm/s.
7.2.10.6 Liner System Installation—
The local soil was excavated with dozers and scrapers. Excavated clayey
materials were placed on a stockpile. Here the material was spread in 8-inch
lifts and then watered to achieve a moisture content between optimum and
3 percent above optimum. After the excavation was complete, the pond bottoms
were graded according to design specifications. This included the additional
excavation of a 10-foot-wide trench area beneath the leak detection system
collection pipes.
The stockpiled liner material was disked and then placed in the pond
excavations in horizontal lifts (which, before compaction, were 8 inches
thick). The lifts were compacted to 95 percent standard Proctor density with
a sheepsfoot roller.
A backhoe was used to excavate a shallow trench into the bottom liner
for the leak detection drain and sump. Perforated pipe was placed in the
trench and then covered with gravel. The entire bottom liner was then
covered with a sand drainage layer.
The final 5-foot (minimum) clay liner was constructed through the same
procedure as previously discussed. A 2-foot protective layer of sandy soil
was placed on top of the finished clay liner. Waste unloading troughs were
protected at the bottom with rip-rap.
7.2.10.7 Performance—
The construction of the liners of these ponds was completed in the fall
of 1980. Ponds No. 1 and 3 were filled with waste shortly thereafter. Pond
NOo 2, however, was left unfilled and uncovered until the following spring.
Approximately 3 months later, liquid started accumulating in the leak
detection system. This liquid had elevated levels of chloride, ammonium, and
total dissolved solids as well as increased conductivity. Liquid was not
detected in the sumps of ponds No. 1 and 3.
7-35
-------
CO
0>
Figure 7-12. Pian view of site !.
-------
CO
VI
Truck Maneuvering Area ^^-18-in Gravel Layer
Freeboard
Waste
2 ft Liner Protection Layer
10 ft (max.)
Liquid
Depth
18-in Leak Detection Layer
5 ft (min.)
Compacted Clay Liner
5 ft (min.)
Compacted Clay Liner
Leak Detection Pipe
\\
\\ // \\
\\
\\
10ft
Figure 7-13. Cross section of liner at site I.
-------
The most probable cause of the failure at pond No. 2 was waste migration
through desiccation cracks in the upper liner. Cracks ranging from 2 to
10 mm wide were observed on the upper portion of the sidewalls. In most
cases, these cracks could not be traced for any more than a few feet along
the surface. An exception to this was in the corners of the liner, where the
cracks appeared to be up to 20 feet long.
7.2.11 Site J
7.2.11.1 Physical Description--
This facility consists of a drum disposal pit, a solvent recovery facil-
ity, and seven ponds, one of which is used for landfill ing. The ponds range
from approximately 1/2 to 4 acres. All of these areas are lined with two
layers of compacted clay separated by a leak detection system. A 10-foot-
high retention dike located northeast of the disposal areas prevents any
overflowing liquid, such as may occur during a 100-year storm, from leaving
the facility. Several groundwater monitoring wells are Installed around the
facility (see Figure 7-14).
7.2.11.2 Startup Date--
Ponds No. 1 through 4 and the drum pit went into service in June of
1979. The remaining ponds (No. 5 and 6) started operation in May of 1981.
Pond No. 7 was used as a landfill starting in May of 1981.
7.2.11.3 Local Geology and Hydrology—
This facility is located in the southern central United States. The
average annual precipitation at the facility is approximately 26 inches.
The top layer of soil at this site consists of red silty clay. Thin
sandstone and gypsum lenses are present at depths ranging from 8 to 18 feet
below the ground surface. A layer of gray siltstone encountered at depths
ranging from 14 to 27 feet underlies the silty clay. The physical properties
of the silty clay are as follows:
• Soil type Very fine silts or clays
• Liquid limit 31.0 - 46.0%
• Plastic limit 22.3 - 29*5%
• Plasticity Index 7.4 - 20.1%
• Optimum moisture content 18.0 - 23.7%
• Maximum dry density 99.3 - 110.4 lb/ft3
• Laboratory permeability 2.0 x 10-9 _ 2.9 x lO'8 cm/s.
Six borings were dug in an attempt to determine the location of the
water table. At the time of exploration, water was encountered in two of the
borings at depths of 7 and approximately 27 feet below the ground surface.
Ten days later, water was standing in all six borings at depths ranging from
5 feet 8 Inches to 23 feet below the ground surface. Surface drainage from
the site is in two directions—north and east.
7-38
-------
VI
I
CO
tO
Groundwater
Flow
Solvent
Recovery
Facility
Figure 7-14. Plan view of site J.
-------
7.2.11.4 Waste Type--
Th1s facility receives most types of hazardous liquids and sludges.
7.2.11.5 Liner Description—
The bottom liners at this facility consist of two recompacted clay
layers separated by.a leak detection system. The lower clay liner is 1 foot
thick and the upper clay liner is a minimum of 5 feet thick. The sidewa'lls
are constructed partially underground and partially above ground. The above-
ground sections are built upon dikes. The leak detection system is a 6-inch
drainage blanket that slopes to a trench containing a 2-inch slotted PVC
collection pipe (See Figure 7-15). Quality assurance test results of the
completed liners are as follows:
• Dry density 94.2 - 115.8 Ib/ft3
• Moisture content 15.8 - 31.0%
e Permeability 5.3 x 1Q-9 - 2.8 x 10-Q cm/s
• Liquid limit 35.6 - 63.0%
• Plastic limit 16.1 - 31.6%
• Plasticity index 6.7 - 36.6%.
7.2.11.6 Liner System Installation--
No specific information on the liner construction or the equipment used
was available. However, it is known that construction of the second phase
(ponds No. 5, 6, and 7) took place during the winter. Problems due to frozen
and unworkable liner materials caused many construction slowdowns.
Quality control (QC) and quality assurance (QA) inspectors were present
at the facility during construction operations. They conducted tests, made
observations, and prepared a project diary and final documentation report.
7.2.11.7 Performance--
Liquid, presumed to be construction water, was being collected and
removed from all of the detection systems by early 1982. This liquid was
periodically analyzed and found to be "clean" until February 1982. At this
time, samples removed from the pond No. 5 detection system contained a
yellow-brown oily liquid with an organic-solvent odor. A small water phase
was present in the sample. Analysis of this sample showed that 1t contained
over 11.2 percent perch!oroethylene (perc) plus small amounts of other
chlorinated and nonchlorinated organics. The facility was allegedly not
accepting wastes with more than a trace of "perc."
An Investigation of the problem revealed that aqueous wastes coming from
one of the disposal facility's customers contained approximately 5 percent
degreasing fluid waste, which contained perchloroethylene. This waste was to
have been stored in a separate tank at the originator's plant. Evidently, it
was not made clear to the workers where to put the "perc" waste, and it was
placed 1n an aqueous waste storage tank. Being insoluble and heavier than
7-40
-------
Dike
2 Sidewall Slope
I/
2-in Slotted PVC
Leak Detection Pipe
5 ft (min.)
Compacted Clay
Liner
6 in Drainage Blanket
(Leak Detection System) •
1-ft Compacted Clay Liner
6-in (min.)
Figure 7-15. Cross-sectional view of site J liner.
-------
water, 1t sank to the bottom. When a tank truck from the disposal facility
would collect a load of aqueous waste, it would also receive a few hundred
gallons of tfie "perc" waste. The waste sampling method used by the disposal
facility on all loads of incoming waste was not able to sample the bottom few
inches of each load. In addition, the parking area where all truckloads of
waste were sampled was slightly sloped; this sloping caused the relatively
small amount of "perc" waste in the bottom of the tanker to flow to the back
of the truck, where it could not be reached by the sampling device.
Over approximately 9 months, the aqueous waste containing the "perc" was
placed in one of the clay-lined ponds and went unnoticed until the liner
failure occurred. The apparent incompatibility is consistent with current
research on the effects of chlorinated solvents on clays.
Presently, the leak in the pond No. 5 liner has slowed down consider-
ably. Perch!oroethylene concentrations in the detection system liquid are
less than 100 ppm. It is estimated that several thousand gallons of the
"perc" waste are adsorbed into the clay liner and the sludges in the pond's
bottom and will present minor problems for many years.
7.2.12 Site K
7.2.12.1 Physical Description—
This facility consists of six double-lined ponds ranging in size from
1.2 to 2 acres. The liners are composed of a 3-foot recompacted clay layer,
below which lies a leak detection system and a 1-foot layer of recompacted
clay. A typical cross-section of the pond and liner system is illustrated in
Figure 7-16. The design depth of all six ponds is 10 feet.
7.2.12.2 Startup Date—
The construction of the first pond at this facility was completed in
December 1981. Five additional ponds have been constructed since then.
7.2.12.3 Local Geology and Hydrology™
This facility is located in the western United States. Average annual
precipitation at this facility is approximately 6 inches, while the average
annual evaporation is 63 inches.
A total of 10 trenches and 12 borings ranging from 8 to 100 feet in
depth were used to investigate the geology at this facility. This investiga-
tion, combined with prior knowledge, revealed three distinct layers of silty
claystone, siltstone to sandstone, and silty claystone. The uppermost layer
caps a flat ridge at the facility. Here the silty claystone is up to 80 feet
thick. The next layer ranges up to 120 feet thick. The lower layer is in
excess of 80 feet thick across the entire site.
The uppermost claystone layer was characterized most extensively. It
has the following properties:
• Liquid limit 56 - 84%
0 Plasticity index 36 - 55%
e P200 83 - 99%
7-42
-------
VI
I
CA>
Waste
10ft
\\
\\
• _ • •• ' _L^M,
/A K^N
//\ 3ft St
^^^ J
\\
\\
1% Slope
1% Slope
Dike
3 ft (min.)
Recompacted
Clay Liner
\\ // \\ .
1 ft (min.)
Sand Basket
(Leak Detection
System)
1-ft (min.)
Recompacted Clay Liner
4 in Slotted PVC Pipe
Figure 7-16. Cross-sectional view of site K liner.
-------
t Laboratory permeability 5.0 x 10~7 - 2.8 x lO"8 cm/s
(remolded)
• Laboratory permeability 3.7xl07-6.1x lO"8 cm/s
(undisturbed)
Investigations also revealed the presence of occasional gypsum veins through-
out the claystone material. These veins ranged in thickness from less than
1/10 inch to about 3 inches and had an average permeability of
2.4 x 10-3 on/s. a y
A review of published data indicated that several seismic faults exist
in the region of the disposal facility. However, within 4 miles of the site
no seismic faults show evidence of recent creep. The two seismic faults
within 1/2 mile of the facility are both inactive.
State investigations indicate that groundwater is scarce in the vicinity
of the site. This fact was confirmed by drilling 12 borings on the site
property. Of the 12, only 1 encountered water at a depth of 32 feet perched
on claystone bedrock.
The scarcity of groundwater is due to low rainfall, high evaporation,
and the fine-grained nature of the sediments.
7.2.12.4 Waste Type--
The facility accepts liquid-scrubber wastes that are generated while
stack gases are cleared from oil-refining facilities. The wastes are highly
saline with pH values in the wide range of 3.5 to 9.0. The waste is
temporarily stored in the evaporation ponds. When it has been reduced to a
semlsolid, it is removed and disposed of onsite.
7.2.12.5 Liner Description—-
The pond liners were constructed of the local excavated claystone
material. The liner system consists of a 1-foot recompacted basal liner.
Sump and collection trenches are excavated into this bottom liner. A 1-foot
sand layer covers the lower clay liner. The sand layer is overlain with
3 feet of clay.
Four undisturbed samples were taken from one of "the constructed liners
and tested for their Atterberg limits, percent passing the No. 200 sieve,
moisture content, and dry density. The results of these tests are presented
below.
_L 2 _3 _ 4
• Atterberg limits
-Liquid limit (%) 70 87 84 93
-Plasticity index (%) 49 67 64 69
t Amount passing No. 200 86.9 88.1 88.3 87.2
sieve (%)
7-44
-------
• Moisture content (%) 24.9 21.5 20.8 23.0
• Dry density (Ib/ft3) 94.6 103.0 104.4 99.8
Permeability tests were conducted in which remolded liner samples were
subjected first to water and then to typical wastes. The permeability of the
liner material to both water and the scrubber waste ranged from
2 x ID-7 to 1 x ID-8 cm/s. The design specifications called for the
liner permeability to be no greater than 1 x lO'6 cm/s.
7.2.12.6 Liner System Installation—•
The leak detection system was constructed by first excavating and sub-
sequently recompacting the clay soil to provide a 3-foot foundation beneath
the collection drains and sump. A 1-foot clay liner was then placed over the
base and side slopes of the excavated area. The sump and drain-pipe trenches
were then excavated over the 3-foot foundations. After the installation of
the drainage system, a 1-foot layer of sand was placed over the entire clay
liner. Finally, a 3-foot recompacted clay liner was placed on top of the
sand layer. A segmented steel-wheel compactor was used for liner
compaction.
Construction activities were inspected by the design firm. Included in
these inspections were numerous density, moisture content, and permeability
tests. At the completion of construction, the design firm certified that the
facility was constructed according to design specifications and would there-
fore perform as designed.
7.2.12.7 Performance--
Liquid volumes collected in the leak detection system were used to
determine the installed liner permeability. The permeabilities for five of
the six ponds were calculated based on Darcy's Law, the collected leachate
volumes, and the impounded liquid depth and pond area. The value for pond
No., 4 was not available. These values are as follows:
Pond Average Permeability (cm/s)
1 2.95 x ID'7
*
2 1.8 x 10-7
3 4.1 x ID'8
4 Data not available
5 1.4 x 10-7
6 1.4 x 10-7
The permeability of the installed liner system is less than the
specified value of 1 x 10~6 cm/s.
7-45
-------
This facility 1s located approximately 100 miles south of the epicenter
of a recent major earthquake. After the quake, the excavation, slopes, and
berms were examined carefully, but no indications of seismic damage were
noted. Careful monitoring of the ponds since the quake has not revealed any
unusual changes in fluid levels or leachate collection volumes.
7.2.13 Site L
7.2.13.1 Physical Description--
The landfill consists of a flat double-clay liner on top of which a dike
has been placed to contain the wastes. The liner extends beyond the dike to
form the liner for a 12-foot-wide drainage ditch that encircles the site (see
Figure 7-17). v
The landfill is divided into two separate sections, one for each of two
types of waste. The largest area is designed to contain dewatered sludge
pumped from settling tanks. It has a 5-foot-high perimeter dike and a
drainage blanket on top of the liner composed of 1 foot of graded clean sand
with drain pipes installed to collect and remove leachate. Sludge is
transferred to this portion of the landfill through a pipeline from offsite
settling tanks.
The smaller section of the landfill is designed to hold chemical wastes
from the plant manufacturing units and QC laboratory. It has a gravel bottom
over its liner. The working area was developed in sections across the width
of the landfill with the waste layer maintained at approximately 3 feet in
depth. Each 11ft of waste was covered with river sediment material
previously deposited on the site.
7.2.13.2 Startup Date—
The site became active in 1978.
7.2.13.3 Local Geology and Hydrology--
The facility is located in the northeastern United States. Average
annual precipitation at this facility 1s approximately 40 inches.
The facility is located in an area where there is approximately 3 feet
of topsoil over a subsoil varying from silt to sand to fine gravel and sandy
clay. Groundwater is located in the subsoil layer a* depths ranging from 4.5
to 14.5 feet below the ground surface.
7.2.13.4 Waste Type--
The landfilled waste material can be defined as either sewage treatment
plant sludge or chemicals; The sewage treatment plant sludge consists of
dewatered (12 to 15 percent solids) sludge removed from settling tanks and
transferred to the landfill. The chemical waste consists of Inert material
from Hme slaker operations (inert rocks and insoluble calcium and magnesium
salts), hard-pitch residue that is only slightly water soluble and that
crystallizes at 180°C, filter aid (dicalite) wetted with phosphate
esters, filter paper wetted with phthalate esters, and residue from the
manufacture of tetrachlorophthalic anhydride.
7-46
-------
VI
I
Waste
Dike
Drainage Layer
Leachate Collection Pipe
Leak
Detection
Layer
Figure 7-17. Cross-section of containment system at site L.
-------
7.2.13.5 Liner Description—
The liner system at this facility consists of the six layers listed
below from the top down:
t 8- to 12-in. drainage layer (sand or gravel)
• 18- to 25-ini compacted clay
• 12-in. sand leak detection layer
• 12-in. compacted clay
• 6- to 18-in. compacted soil
• Bidim type C34 (synthetic soil stabilization geotextile).
The physical properties of the clay used for the liner are as follows:
• Soil classification - grey silty clay
0 Particle size distribution: Gravel - 1%
Sand - 29%
Clay and colloids - 707.
• Liquid limit 45%
0 Plastic limit 17%
0 Plasticity index 28
0 Liquidity index 0.1
0 Specific gravity 2.67
0 Moisture content 28.5%
0 Dry unit weight 95.0 lb/ft3
0 Maximum dry density 101.5 *
0 Optimum moisture content 19.5%
0 Permeability . 6.6 x lO"8 to 3.8 x 10'9 cm/s.
7.2.13.6 Liner System Installation--
No information was available on installation procedures.
7.2.13.7 Performance™
Within the first year of operation, problems developed in the upper clay
liner. Evidence of this was the appearance of contaminated water in the
leachate monitoring layer manhole. Slight contamination in local groundwater
monitoring wells was also discovered. A study conducted to determine the
source of the groundwater contamination revealed that the source of contami-
nation was not the double-lined pond.
7-48
-------
7.2.14 Site M
7.2.14.1 Physical Description—
This 12-acre site is designed to consist of three cells of approximately
equal size. At the time of this writing, the first cell was still in use and
construction had not yet started on either of the remaining two cells.
The facility is lined with 1 foot of recompacted clay. A leachate
collection system is above the clay liner and a leak detection system lies
2 feet below the clay liner. In addition to the leachate collection and
subsurface monitoring (leak detection) systems, three groundwater monitoring
wells are situated around the facility as well as drainage ditches to prevent
surface runoff from entering the landfill (see Figures 7-18 and 7-19).
7.2.14.2 Startup Date—
10-70 Construction was initiated 1n 1977. The first wastes were accepted in
1978 ii
7.2.14.3 Local Geology and Hydrology—
This facility is located in the northern central portion of the United
States. Average annual precipitation in the vicinity of this facility is
approximately 26 inches. *
A total of five borings were used to investigate the subsurface
conditions at the proposed landfill site. The borings indicated a uniform
geologic profile over the entire facility. The top 1 to 2 feet of soil
consists of a sllty sand underlain by approximately 25 feet of fine- to
medium-grained sandy alluvium.
Groundwater was encountered in all five borings at depths ranging from
approximately 10.5 to 15 feet below the original ground surface. Groundwater
flow was toward a marshy area located east of the proposed facility. Other
information indicates that in a second deep aquifer groundwater flow is in a
southeasterly direction.
7.2.14.4 Waste Type and Placement--
The landfill is used for the disposal of inorganic lime sludge. Samples
of the sludge have been analyzed for heavy metals content and percent solids.
The results of these analyses are presented in Table*7-5.
Due to the high moisture content of the waste material, a special proce-
dure for waste placement was developed. This procedure involved mixing a
thin layer of the sandy soil obtained from the site excavation with a thin
layer of the waste 1n order to reduce the overall moisture content and to
give the material structural stability. The range of soil to waste ratios
varied from 0.5:1 to 1.5:1. Exact mixing proportions were determined in th«
field and judged sufficient when the landfill equipment was able to drive
over the mixture. Due to the great amount of soil mixed with the waste and
the waste's inorganic nature, no dally cover was used at this facility. The
proposed final cover will consist of a 6-inch compacted clay cap, 2 feet of'
topsoll, and a vegetative cover. No gas venting or collection will be
necessary.
7-49
-------
en
o
Original Ground Contour
1ft
Coarse
Sand or
Fine Gravel
Subsurface Monitoring
System Pipe
(Leak Detection)
Leachate
Collection
System Pipes
1-ft Compacted
Clay Liner
Leachate
Collection
System Sump
and Manhole
Subsurface
Monitoring
(Leak Detection)
System Sump
and Manhole
Figure 7-18. Cross-section of site M.
-------
O Location of borings
Location of monitoring wells
MH Manhole
Area for
Future Development
Direction of
Groundwater
Flow
Leachate Collection System
Perforated Pipes
Subsurface
Monitoring
System (Leak
Detection)
Figure 7-19. Plan view of site M leachate collection and leak detection
systems.
7-51
-------
TABLE 7-5. HEAVY METAL CONTENT AND PERCENT SOLIDS OF LIME SLUDGE
DISPOSED AT SITE M
Sample date
7/79
7/80
10/81
7/82
Cadmium
(mg/kg)
6.6
1.6
3.3
0.89
Chromium
(mg/kg)
200
110
1,300
170
Copper
(mg/kg)
4,500
5,300
3,800
520
Cyani de
(mg/kg)
750
140
2
52
Iron
(mg/kg)
5,400
10,000
11,000
880
Lead
(mg/kg)
5,200
5,100
4,800
520
Nickel
(mg/kg)
1,300
1,100
340
64
Zinc
(mg/kg)
5,700
4,900
5,200
640
PH
7.8
8.6
7.9
8.5
Total
sol i ds
(%),
—
11.1
28.7
12.1
01
ro
-------
7.2.14.5 Clay Liner Description—
The clay for the liner was obtained from a nearby borrow area. The
specifications for the clay are as follows: ?
t Liquid limit 50 - 70%
t Plasticity tndex >28%
• P200 >75% (by weight)
• Permeability
-------
TABLE 7-6. GROUNDWATER MONITORING WELL SAMPLE ANALYSIS AT SITE M
VJ
I
U1
Well
number
1
1
2
2
3
3
Sample
date
7-79
7-82
7-79
7-82
7-79
7-82
Cadmium
(mg/kg)
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
Chromi urn
(mg/kg)
<0.05
<0.05
<0.05
<0;05
<0.05
<0.05
Copper
(mg/kg)
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
Cyani de
(mg/kg)
<0.01
<0.02
<0.01
<0.02
<0.01
<0.02
Iron
(mg/kg)
0.30
0.10
0.25
0.4
0.1
0.05
Lead
(mg/kg)
-------
The leachate from the landfill has also been sampled and analyzed since
1978. Examples of these data appear in Table 7-7.
The subsurface monitoring system tank also has been checked on a
quarterly basis. There has been no indication of leachate entering this
system.
The above facts confirm that the landfill has performed as designed
since its construction and initial waste placement in 1977 and 1978.
7.2.15 Site N
7.2.15.1 Physical Description—
The site is located in an old sand and gravel pit and covers
approximately 160 acres. The eight cells range in size from 9 acres to 25
acres with 20 acres being the average cell size. At the time of writing, the
liner covers 108 acres.
The facility bottom is lined with a 4-inch bentonite and sand
layer. The sides of the landfill are unlined. Two 1,000-ft2 PVC-lined
lysimeters are below the liner as well as a system of 24-inch perforated
pipe, which lowers the groundwater table. A leachate collection system is
above the liner. In addition to these systems, several monitoring wells and
surface-water monitoring points are around the facility (see Figures 7-20 and
7—21)*
7.2.155.2 Startup Date-
Construction of cell 1 was initiated in 1974. Since then, seven
additional cells have been installed.
7.2.15.3 Local Geology and Hydrology—
This facility is located in southeastern Canada. The average annual
precipitation in the vicinity of the landfill is approximately 35 inches.
Very little information was available concerning the local geology and
hydrology. As previously mentioned, however, the facility is located in an
old sand and gravel pit. The local soil is very sandy and, for this reason,
a bentonite liner was chosen.
«
Because natural groundwater elevations at the facility are very near the
surface, a drainage system was installed to lower the groundwater table. The
final groundwater elevation is maintained at 5 feet below the liner bottom.
7.2.15.4 Waste Type—
The facility accepts municipal solid waste only. At the time of
this writing, the facility contained approximately 25 million yd3 of
waste material.
7.2.15.5 Liner Description-
Laboratory studies have shown that 6 percent by weight of bentonite
(sodium montmorillonite with 55 to 75 percent by weight passing the No. 70
mesh sieve) in a 6-inch sand layer would produce a liner with the specific
7-55
-------
TABLE 7-7. LEACHATE ANALYSIS AT SITE M
Sample date
7/79
7/80
10/81
7/82
Cadmium
(rag/L)
<0.01
<0.01
<0.01
0.02
Chronri urn
(rag/L)
<0.05
<0.05
<0.05
0.05
Copper
(mg/L)
0.15
0.20
0.30
0.35
Cyarri de
(mg/L)
0.02
0.03
0.09
<0.02
Iron
(mg/L)
3.5
4.0
3.0
3.5
Lead
(mg/L)
<0.1
<0.1
0.1
<0.2
Nickel
(mg/L)
0.10
0.05
0.35
0.15
Zinc
(mg/L)
0.13
0.10
0.10
<0.10
PH
'7.4
7.5
6.9
7.2
VI
I
en
o>
-------
I
01
Groundwater
Pumping Station
Figure 7-20. Plan view of site N.
-------
en
oo
Groundwater
Pumping Lysimeter
Station Manhole
Leachate
Pumping
Station
Vii'".£'i::iu'.'lii'i-'°il>•'—'&:'£'''^'''£-&Zy/\'-'.*.\'':'?;
6 in Sand Layer
(Collection System)
} Bentonite-Sand Liner
Lysimeter
20 mil PVC Membrane
Lysimeter Drain Pipe
Perforated Groundwater
Drain
Compacted
Native Sand
18 in (min.)
Figure 7-21. Cross-sectional view of site N liner and ieachate management systems.
-------
permeability of 1 x 10-8 cm/s. Duping liner installation, a 4-inch layer
produced a moje uniform mix. Therefore, a 4-inch liner was installed with an
additional 3 percent by weight of bentonite as a safety margin. The liner
was compacted to a minimum 90 percent standard Proctor density.
7.2.15.6 Liner System Installation—
7.2.15.6.1 Excavation—Little additional excavation was necessary due
to the location of the facility in an old sand and gravel pit. Areas that
were to be lined were excavated to bottom contours and graded to a 1-percent
slope with large-capacity self-loading scrapers. A 7-ton, self-propelled
vibratory, smooth-drum roller was used to compact the basal sand and to
complete the grading of the cell bottom into a relatively smooth surface.
7.2.15.6.2 Liner—The following methods were used to construct the
liner.
Bentonite was spread to a 5/8-inch thickness with three to five passes
of a large scraper. Next, a tractor-mounted rotatiller was used to mix the
bentonite into the sand to a depth of approximately 4 inches. The liner
material was allowed to hydrate naturally. Additional water was added as
necessary to achieve the required moisture content. The liner was compacted
with a vibratory smooth-drum roller a minimum of four passes.
The leachate collection system was then installed on the liner and
covered with clean pea gravel. A 6- to 8-inch layer of loose sand was placed
over the finished liner. No traffic was allowed on the completed area before
refuse was spread over the liner from the top.
QA testing of the constructed liner indicated that the bentonite content
ranged from 8.5 to 15 percent by weight and averaged about 11.2 percent by
weight. Greater than 90 percent compaction was achieved in all areas with
variations from 93 to 100+ percent standard Proctor density. The moisture
content varied from 3 to 12 percent.
7.2.15.6.3 Monitoring Systems—Methods used to install the various
monitoring systems were not available.
7.2.15.7 Performance— -
Analysis of water samples collected from lysimeter 1, surface-water
monitoring points, and groundwater monitoring wells indicates that leachate
has passed through the liner and entered the groundwater. Liquids have also
been detected but not analyzed in lysimeter 2. Table 7-9 contains the
results of the analysis for biochemical oxygen demand (BOD), COD, total
coliform, and fecal coliform at several of the monitoring points.
7.2.16 Site 0
7.2.16.1 Physical Description—
This landfill is approximately 2 acres in size. It is lined with a
mixture of local soil combined with a 1:1 mix of bentonite and lime. The
ime was added to reduce the swelling potential of the bentonite. A
leachate collection system is on top of the clay liner. Several groundwater
7-59
-------
TABLE 7-9. WATER SAMPLE ANALYSES: BOD, COD, TOTAL COLIFORM,
AND FECAL COLIFORM
Sampling location
(see Figure 7-20)
L ia
SW lb
SW 2b
SW 3b
SW 4b
MW lb
MW 2b
MW 3b
MW 4b
BOD
(mg/L)
336
<1
<1
1
260
>800
16
13
2
COD
(mg/L)
1,560
3.9
5.8
19
400
1,870
45
100
10
Total col i form
per 100 mL
14,000
500
1,900
500
0
0
190
0
0
Fecal coli form
per 100 mL
14,000
140
320
90
0
0
0
0
0
aSample date - 12/5/83.
bSample date - 10/11/83.
7-60
-------
monitoring wells are located adjacent to the landfill. The landfill has been
capped and covered with asphalt pavement.
The leachate collection system consists of a stone-packed sump (drywell)
with a casing pipe extending to the surface that is used for observation and
leachate withdrawal.
7.2.16.2 Startup Date--
Construction of this facility began in June of 1980. It was filled and
capped by September 1980.
7.2.16.3 Local Geology and Hydrology—
This facility is located in the southeastern United States. Average
annual precipitation at the facility is approximately 50 inches.
The soil in the vicinity of the landfill is composed of a mixture of
clay, silt, and sand. Soil borings of the site indicate that this soil
mixture extends to a minimum of 10 feet. Analysis of the native soil gave
the following results:
« Average permeability 8.7 x 10-5 cm/s
» Plasticity index 7.1%
« Liquid limit 27.1%
» Plastic limit 20.0%
• Amount passing No. 200 sieve 31.3%.
Monitoring wells have indicated that groundwater is well below the
bottom of the site and poses no potential problem. Water samples from the
monitoring wells are taken monthly and analyzed for PCB's, pH, specific
conductance, and chlorinated organics.
7.2.16.4 Waste Type—
The major material disposed at this site was PCB's. The facility also
contains solvents, waxes, and oils, all of which were solidified with sawdust
prior to their disposal.
7.2.16.5 Liner Description—
The 4-foot-thick liner consists of a homogeneous mixture of 3 percent
bentonite, 3 percent lime, and 94 percent native soil.
The physical properties of the liner material are:
• Liquid limit 27.8%
• Plastic limit 20.7%
• Plasticity index 7.1%
• Permeability 8.3 x 1Q-8 cm/s
• Amount passing No. 200 sieve 35.4.
7-61
-------
The leachate collection system consists of a stone-packed sump (drywell)
with a casing pipe extending to the surface that is used for observation and
leachate withdrawal.
7.2.16.6 Liner Installation—
The liner at this facility was installed in 8-inch lifts and compacted
to at least 95 percent of maximum Proctor density. The bottom of the facil-
ity slopes at a rate of 2 percent to the leachate collection system. The
sidewall slopes are 3:1 maximum. The 3-foot-thick cap constructed of the
same mixture was compacted to 90 percent of maximum density. No information
on the Installation of the asphalt cover was available.
7.2.16.7 Performance--
Due to a poor seal around the leachate collection system casing pipe
(standpipe), surface water eroded through this area and was collected in the
sump. This problem was solved by removing a 12-foot-diameter section of the
cap surrounding the pipe. The area was then filled with tightly compacted
pure bentonite. A concrete dome was placed on top of the bentonite to direct
the flow of water away from the casing pipe. Because additional liquids were
still being collected in the collection system, the entire cap was paved with
asphalt. No information was available on the performance of the facility
after the asphalt was installed.
7.2.17 Site P
7.2.17.1 Physical Description—
The facility consists of one double-lined hazardous waste cell. The
bottom liner is composed of leachate collection and detection systems as well
as a series of natural soil and bentonite/soil liners. The side liner and
dike-containment system include a soil and a bentonite/soil liner (see
Figures 7-22 and 7-23).
7.2.17.2 Startup Date--
The facility was constructed and began accepting waste in 1980.
7.2.17.3 Local Geology and Hydrology—
This facility is located 1n the southeastern United States. The average
annual precipitation in the vicinity of this facility is approximately
52 Inches. No information on the local geology was available.
Groundwater at this site occurs in soil and fractured rock. The maximum
groundwater table 1s located approximately 14 to 15 feet below the liner
bottom. Groundwater levels at the site fluctuate with precipitation, tending
to be high 1n winter and spring and low in the remainder of the year.
7.2.17.4 Waste Type-
Approximately 80 percent of the disposed waste 1s brine purification
mud. This material has the consistency and appearance of fine, wet sand.
The remaining 20 percent of the wastes are all mercury-contaminated
substances including sulfide treatment filter cake, retort ash,
Reductone® filters, decomposer packing, and contaminated earth.
7-62
-------
0>
CO
Side Liner System
(see Figure 7-23
for details)
Waste
6ft
Bottom Liner System
(see Figure 7-23 for details)
Dike
\\ // \\ // \\ // \\ // \\ // \\
\\ // \\
Native Soil
Figure 7-22. Cross section of site P showing relationship of liner and dikes.
-------
VI
CO
6-in Bentonite/Soil Liner
18 in Recompacted Clay Liner
1-ft Drainage Layer
6ft
'•:*V"V-&^'->V^ Leachate
?&:i<&::-£+x£&t&?}:^i%X&&}^ Collection La
Collection Layer
-3-ft Recompacted
Clay Liner
/XHf»-1-ft Bentonite/Soil Liner
•1-ft Recompacted Clay Liner
t Leak Detection Layer
5 in Bentonite/Soil Liner
• 5-ft Recompacted
Clay Liner
\\
\\
\\
\\
Figure 7-23. Detailed cross section of site P liner.
-------
7.2.17.5 Liner Description—
The liner at this facility is composed of two sections: a bottom liner
and a side (dike) liner. Each of these sections has its own specifications.
The bottom liner extends over the entire bottom and 6 feet up the
sides. It is a layered system containing two drainage or collection layers
and two soil liners. From the top layer downward, the bottom liner
components are as follows:
• Leachate collection system— 1 foot of No. 78 gravel and sand with
4-inch perforated PVC pipes.
• Upper soil barrier— 5 feet of compacted soil further subdivided into
three layers:
- 3 feet of compacted native soil; permeability =
1 x 10~4 cm/s.
- 1 foot of enhanced soil, i.e., native soil blended with 9 to
12 percent polymer-treated bentonite; permeability on the
order of 5 x 10~8 cm/s.
- 1 foot of compacted native soil; permeability =
1 x 10-4
• Leak detection layer. A 1-foot sand/gravel layer with perforated
pipe, to detect leaks and/or to control seepage through the upper
barrier. The pipes are connected to several independent monitoring
stations to allow the determination of the approximate location of
any leaking that might occur.
» Lower soil barrier. A 6-inch layer of enhanced soil, i.e., native
soil blended with 9 to 12 percent polymer-treated bentonite;
permeability on the order of 5 x ICT8 cm/s.
o Buffer zone. 5 feet of either in situ or recompacted native
soil. Permeability = 1 x KT4 cm/s.
The side liner system extends from a point 6 feet above the cell bottom
to the top of the cell. This section will not have liquid impounded against;
it; therefore, the liner system is not as extensive as the bottom liner.
From the top layer downward, the side liner components are as follows:
« 1 foot of No. 78 gravel
• 18 inches of compacted native soil; Permeability = 1 x lO"4 cm/s
t 6 inches of enhanced soil, i.e., native soil blended with 9 to
12 percent polymer-treated bentonite.
7-65
-------
Liner-waste compatability tests were conducted from June through
December 1980 with a percolation column. In it was placed a compacted blend
of 1,600 g native soil and 360 g bentonite (i.e., 18.4 percent bentonite).
The column was filled with actual facility liquids, with analysis as follows:
• 1.87 ppm mercury
• 154,036 ppm chlorides
t pH 6.1.
Short-term permeability results are as follows:
Date Permeability (cm/s)*
6-2-80 - 2.23 x 10~7
6-10-80 2.92 x ID'7
6-17-80 3.71 x 10-7
7-2-80 4.05 x 10-7
7-18-80 4.81 x 10-7
7-24-80 5.88 x 1Q-7
Ca. 12-80 Approximately 0
7.2.17.6 Liner System Installation—
The construction methods used are as follows:
• Soil spreading. All soils and gravel were placed with scrapers
and spread with bulldozers.
• Soil compaction. All soils were compacted to 95 percent standard
Proctor density. This was done by repeated front-end loader or
scraper passes (rubber-tired). An independent soils laboratory
took samples as the work progressed, and any fill not meeting the
minimum 95 percent compaction was upgraded accordingly.
• Mixing of soils and bentonite. The bottom 6-inch layer of
"enhanced" soil was constructed by first spreading two lifts of
approximately 4.5 to 5 inches of loose native soil. Likewise, the
top 12-inch layer of "enhanced" soil was constructed by first
spreading four such lifts. Scrapers were used to spread the
soil. After placement of loose soil for each layer, the bentonite
product was hand-placed at an application rate of 50
lb/40 ft2.
^Calculated values based on leachate absorbed into test specimen. All
data supplied by American Colloid Company.
7-66
-------
Mixing was accomplished with a rotiyator until the color of the
bentojiite became unnoticeable.
t "Enhanced" soil compaction was done by repeated passes of a
front-end loader or rubber-tired scraper until a minimum of
95 percent Proctor was achieved.
• Quality assurance. Soil layers were sampled as the work progres-
sed to ensure 95 percent standard Proctor density. During con-
struction of the "enhanced" soil layers, a minimum of one
field density test was performed per 2,500 ft2 of lift. Tests
for moisture, permeability, grain-size distribution, and liquid
and plastic limits were also performed by qualified soils
personnel on an "as-needed" basis.
Results from the quality assurance of the constructed liner indicate
that the wet unit weight ranged from 101 lb/ft3 to 115 lb/ft3, the
moisture content ranged from 23 to 28 percent, and the laboratory
permeability ranged from 2 x 108 cm/s to 7 x 108 cm/s.
7.2.17.7 Performance—
Leachate collected in the leachate collection layer was analyzed and
found to contain high levels of chlorides and measurable levels of mercury.
Small amounts of liquid have also been collected from the leak detection
system. This liquid does not contain high levels of either chlorides or
mercury. The monitoring wells have not shown any significant changes.
The performance of this facility is very difficult to determine at this
time. A complete analysis of the liquid collected 1n the leak detection
system is being performed. Attempts are being made to correlate these
results with changes in precipitation, waste type, and water table level.
The complete analysis was not available at the time of this writing.
7.2.18 Site Q
7.2.18.1 Physical Description—
The landfill consists of a single containment cell covering an area of
approximately 3 acres. It is located in a former sand and gravel pit. The
facility has a double liner consisting of two 4-inch tentonite/soil layers.
These layers are separated by a leak detection system.
7.2.18.2 Startup Date—
This facility was placed in operation in October 1976.
7.2.18.3 Local Geology and Hydrology—
This facility is located in the northeastern United States. Average
annual precipitation at this facility is approximately 35 inches.
The immediate area around the facility is mostly sand and gravel. The
facility 1s located above the 100-year flood plain.
7.2.18.4 Waste Type—
The sludge deposited in the landfill is a black, semisolid, relatively
odorless material, which will support the weight of small grading equipment.
7-67
-------
The sludge 1s nonvolatile and noncombustlble. It consists primarily of metal
hydroxides and waste pigments. Typical sludge composition 1s as follows:
Metal Percentage
Iron 10 - 24
Lead 5-20
Chromium 6-12
Barium 2-12
Calcium 4-48
Aluminum 1-3
Copper 1-2
Cadmium 1-2
Zinc 0-1
Carbon 2-10
7.2.18.5 Liner Description—
The cell has a double liner consisting of two 4-inch layers of a
bentonlte/soil mixture on the bottom and side slopes up to a vertical eleva-
tion of 20 feet above the cell bottom. The bentonite/soil layers are
separated by a 12-inch layer of sand on the bottom of the cell and a 6-inch
layer of sand on the side slopes. The side slopes above the 20-foot vertical
level are covered with a single 6-inch layer of the bentonite/soil mixture.
All bottom and side slope bentonite/soil surfaces are covered with a 12-inch
protective layer of gravel. The slope of the cell sidewalls varies from
2.5:1 to 3:1. A typical cross section of this facility is illustrated in
Figure 7-24.
7.2.18.6 Liner System Installation--
Most of the bentonite/soil mixture used for the liner was premlxed in a
pug mill to predetermined proportions by closely controlling the feed rate of
each material. Soil was fed from a hopper onto a conveyer belt at a constant
rate. Dry bentonite was fed from a hopper onto the conveyor belt on top of
the son through a variable-rate vibratory feeder. The feed rate of the
bentonite was periodically checked and adjusted as necessary by weighing
timed samples. The two materials were discharged into the pug mill where
water was added at a controlled rate to produce a thoroughly mixed product at
desired moisture content (specific value unknown). The mixture was dis-
charged from the pug mill into dump trucks and was placed 1n position at a
predetermined thickness with a grader, screened boards, and hand labor. The
mixture was then compacted to the specifications with a backhoe-mounted
hydraulic tamper.
During all construction phases, CQA and CQC required full-time supervi-
sion by qualified personnel and a soils consultant as required.
7-68
-------
o>
ID
12-in Gravel Layer (Leachate Collection)
6-in Bentonite/
Soil Layer
12-in Sand Layer (Leak Detection)
O O O
6-in Sand Layer
4-in Perforated
Pipes
4-in Bentonite/
Soil Layers
Figure 7-24. Cross section of site Q liner.
-------
7.2.18.7 Performance—
This facility is presently active and uncapped. As a result, a great
deal of leachate is being collected. Liquid has also been collected in the
leak detection system. Analysis of this liquid shows that it has slightly
elevated levels of cadmium, lead, zinc, iron, copper, cyanide, and COD. The
permeability of the upper bentonite/soil liner has been determined from the
volume of liquid pumped from the sump, the waste head, and the facility
area. This value, as reported by the site owner, varies from
3 x 10"8 cm/s to 6.5 x 10~8 cm/s.
Analysis of the groundwater from the facility monitoring wells has shown
no significant changes since construction of the facility and is well within
drinking water standards.
7.3 LINER TYPES
The feasibility of using a clay liner at a waste disposal facility
depends on several factors, the most important being availability of suitable
liner material. The location of a facility in a deposit of low-permeability
soil lowers the facility cost by greatly reducing or eliminating liner
material transportation costs. The cost incurred by transporting suitable
borrow soils a short distance (e.g., 5 miles or less) may allow this to be a
viable liner option. Another type of soil liner consists of a relatively
high-permeability soil that has been augmented with certain natural or
treated bentonlte additives. The addition of bentonite to a highly permeable
material greatly reduces the material's permeability. These three types of
facilities (I.e., unlined, recompacted soil lined, and bentonite/soil lined)
were included in this study.
Two of the four unlined facilities discussed have had groundwater con-
tamination problems. One facility had a problem due to a faulty well instal-
lation. No other problems have been reported at this facility. The fourth
unlined facility has had contamination detected in one of its monitoring
wells. A study of this contamination indicated that its source was an
adjacent drum disposal facility.
Nine facilities with recompacted soil liners are discussed in this
chapter. Three of these nine facilities have had some type of performance
problem. These three facilities all have leak detection systems that were
responsible for quickly detecting and providing the Information necessary for
determining the cause of the failures.
Four facilities with admixed (bentonite/soil) liners are discussed in
this chapter. Two of these facilities have had small amounts of liquid
detected 1n between their double liners. However, overall they appear to be
functioning according to their design specifications. One facility had a cap
problem that was rapidly detected and corrected. The liner at this facility
has had no detected problems. The final facility with a bentonite/soil liner
that 1s discussed in this section has had performance problems. Leachate has
migrated through the 4-inch liner and has resulted in extensive groundwater
contamination.
7-70
-------
7.3.1 Unlined Facilities ( .,* • ^ ,
Although~the purpose of this document is to discuss clay liners, in some
cases information relating to waste disposal and clay liners may be obtained
from the case histories of unlined disposal sites. Generally speaking,
unlined sites are located in soil formations that are relatively low in
permeability. However, such soil formations are rarely homogeneous. They
generally contain sand or gravel seams, varying amounts of organic matter,
and cracks or fissures. In some cases, they may also contain lenses of
gypsum, limestone, or other soluble material. Discontinuities such as these
are often excavated and then backfilled with recompacted clay prior to waste
disposal. When this is not specified or when highly permeable areas are
unnoticed during facility construction, the siting of an unlined hazardous
waste disposal facility in a heterogeneous soil may cause severe performance
problems.
(Examples of unlined waste disposal facilities are sites A, B, C, and G,,
These four facilities are 8 to 30 years old. Extensive groundwater contami-
nation has occurred at two of these facilities: a slight problem due to poor
well installation procedures occurred at one facility, and contamination in a
well at the fourth facility is thought to have come from an adjacent
abandoned drum disposal operation.
Facility B, a zone-of-saturation or intergradient landfill, has had
severe groundwater contamination problems. This facility is located in a
low-permeability glacial till deposit. Lenses of sand or gravel that were
encountered during the site construction were excavated and backfilled with a
minimum of 5 feet of recompacted clay. A buildup of leachate in the landfill
due to the disposal of unsolidified liquids, precipitation, and seepage
caused waste to enter a permeable deposit that had been improperly sealed
with clay. This resulted in leachate migration and groundwater
contamination.
Facility C is also unlined. Geological investigations revealed that
calcium carbonate seams and nodules were present throughout the entire area.
The design specifications called for the excavations of such seams and then
required the excavated areas to be backfilled with the local clay-shale
soil. Groundwater contamination at this facility can be attributed to
inadequate excavation of the calcium carbonate deposits, inadequate recompac-
tion of the excavated areas, or lack of sufficient waste/liner material
compatibility testing.
Initial groundwater samples taken from a monitoring well at site A had
pH values between 10.7 and 11.3. An investigation into the problem revealed
that waste migration from the unlined facility was not the cause. Instead,
the contamination was traced to some steel-mill slag that was used as well
packing instead of the specified clean gravel. This problem was remedied,
and since that time no other problems have been reported.
Facility G is an old, unlined hazardous waste disposal facility.
Samples from a groundwater monitoring well at this facility have contained
high levels of various pollutants. A study of this contamination indicated
that its source was an old drum disposal site (previously owned by another
company) located adjacent to the landfill. This drum disposal site is
7-71
-------
located on a sand and gravel seam. This permeable seam was determined to be
the mechanism by which the contaminants were transported to the monitoring
well at SiteTS.
7.3.2 Recompacted Soil Liners
Recompacted soil liners are the most common type of soil liner in
current use. This category consists of both recompacted in situ soil liners
and liners constructed of material that has been transported to the facility
from a nearby borrow area. Sites D, E, F, H, I, J, K, L, and M are all in
this category. Some of these facilities have a double liner system (in the
case of sites D and F, one of the liners is a synthetic). When this is the
case, the drainage layer separating the clay liners is called a leak detec-
tion layer or system. All of these facilities have either a leachate collec-
tion or a leak detection system (in some cases, both). These facilities are
4 to 9 years old. The performance at these facilities is quite varied. Six
of the facilities (sites D, E, F, H, K, and M) appear to be functioning as
planned with only a few minor problems. In the case of site K, the liquid
volumes removed from the detection system were used to calculate the liner
permeability. The calculated values were all lower than the design
specifications required.
The remaining three facilities (sites I, J, and L) have all had
performance problems. For example, the clay liner for a pond at site I was
left empty and unprotected for several months. During this time desiccation
cracks formed in the liner. Failure to repair the liner prior to waste
placement resulted 1n waste migration into the leak detection layer. Waste
migration into the leak detection system also occurred at site L shortly
after waste placement. An investigation was unsuccessful in determining the
location of the leak. However, it was determined that the lower compacted
clay liner was not leaking. The other recompacted clay-lined facility with
performance problems, site J, also has a leak detection system. In this
case, perchloroethylene was mistakenly placed into a pond. Approximately 6
to 9 months later, and literally overnight, the "perc" penetrated the liner
and was detected in the leak detection system. The apparent incompatibility
of the "perc" and the liner material is consistent with current research on
the effect of chlorinated solvents on clays.
7.3.3 Admixed Liners -
Admixed liners are those composed of permeable soil and an additive
designed to decrease its permeability. Commonly used additives include,
asphalt, fly ash, soil cement, and polymer-treated or natural bentonite.
Bentonlte, a natural clay mineral, is the only additive discussed in this
document.
The addition of bentonite (generally 3 to 15 percent is used) to a
highly permeable soil can decrease its permeability so that 1t can be used
as a Uner material. A discussion of the physical and chemical properties of
bentonite as well as its advantages and disadvantages as a liner material may
be found 1n Chapters 2 and 5 of this document.
Facilities N, 0, P, and Q are all lined with bentonite/soil admixtures.
These facilities are 5 to 11 years old. The liner configurations at these
7-72
-------
facilities are quite varied. Site'O has a 4-foot-thick native soil liner
containing 3 percent bentonite and 3 percent lime. The lime was added to
reduce the swelling potential of the bentonite. Sites P and Q are lined with
a layered system containing bentonite/soil liners ranging from 4 to 12 inches
thick and leak detection systems. Site P has the additional protection of a
series of compacted native soil liners. Facility N is lined with a 4-inch
bentonite/sand liner. Two lysimeters and a groundwater pumping system are
located below the liner, and a leachate collection system is on top of the
liner.
Just as the liner systems at these four facilities are varied, so are
their performances. Both sites P and Q have had small amounts of liquid
detected in their leak detection layers; however, overall they seem to be
functioning according to specifications. In the case of site Q, the liquid
volumes collected were used to calculate the as-built liner permeabil-
ity. This value ranged from 3 x 10"8 cm/s to 6.5 x l(r8 cm/s, which is
less than the specifications required. The source and quality of the liquid
being collected from the detection system at site P are currently being
investigated by the site owner.
Site 0 had a problem with liquid eroding the cap and infiltrating
through the eroded section. This problem was corrected by repairing the
eroded portion and then paving the entire cap with asphalt. The liner at
this facility has not had any detectable problems.
Finally, facility N, a very large municipal waste disposal facility
located in an abandoned sand and gravel pit, has produced extensive contami-
nation that has been detected in the subsurface monitoring layer, the
lysimeters, and several groundwater monitoring wells. Analysis of the con-
tamination indicates that the leachate is passing through the 4-inch liner
and entering the groundwater.
7.4 SITE CHARACTERIZATION
The location of a clay-lined hazardous waste disposal facility may have
a significant effect on its performance. The site-specific geology, hydrol-
ogy, and climate must.be evaluated thoroughly prior to the facility design
and construction. Some of the geological factors that must be investigated
include the local soil type, permeability, and nature*of the soil deposit
(e.g., whether it contains continuous or discontinuous sand or gravel seams,,
gypsum, or other reactive material seams and the depth of the deposits).
Hydrological factors that must be investigated include location of the
water table, location of any perched water tables above the main water
table„ and the groundwater and surface-water flow patterns. Problems
associated with high water tables include excessive pressure on the side or
bottom liners. This hydraulic pressure may cause sidewall slumping or
collapse or bottom heaving. High water tables may also result in the
unwanted infiltration of groundwater. For a facility under construction,
this may make compaction and other liner installation procedures difficult or
impossible without continuous groundwater pumping. For a closed facility,
groundwater infiltration may result in overloading the leachate collection
system, causing high leachate levels in the facility.
7-73
-------
Site-specific climate data such as average temperature and amount of
prec1pitation_also must be known. Locations that have below-freezing temper-
atures for long periods of time have the potential for freeze/thaw cycling of
unprotected liners. This may result in increased liner permeability and poor
liner performance. Hot climates, on the other hand, may affect clay liners
by causing excessive evaporation and possibly desiccation of the liner. Clay
liners located in hot, arid climates are even more susceptible to desiccation
and cracking. On the other hand, excessive precipitation at an open clay-
lined facility may cause a buildup of hydraulic head on the liner, thus
increasing the flow of liquid through it. Excessive precipitation may also
cause a rise in the groundwater table and the previously discussed problems
associated with this phenomenon.
A complete discussion of these site-specific factors, their associated
problems, and the potential failure mechanisms involved may be found in the
preceding chapters of this document.
7.4.1 Case Studies
Several performance problems or liner failures that may have been
partially caused by inadequate site characterization are illustrated by case
studies B, G, and I.
Site B, a zone-of-saturation or intergradient facility, is located in a
low-permeability soil formation. Sand or gravel seams that were encountered
during excavation were to be removed and replaced with recompacted clay. No
additional recompaction was specified. A buildup of leachate within the
Hner due to precipitation, the disposal of free liquids, and seepage caused
the gradient to reverse. Within the site, apparently a sand seam was either
not discovered or not properly sealed during construction. The leachate
flowed through this sand seam and caused severe groundwater contamination in
a limited area of the property.
Another problem caused by a sand seam was discovered at site G. Contam-
ination in one of the facility monitoring wells was initially attributed to
waste migration from within the site. However, an investigation Into the
problem revealed that the source of contamination was an adjacent abandoned
drum disposal facility. Leachate from this facility entered a water-bearing
sand seam that encountered one of the wells at site G*.
Site I is located in a semiarid climate. Three ponds with double clay
liners were installed in the fall of 1980. Two of the ponds were filled
shortly after construction, while the third pond was not filled until the
following spring. During this time, the unprotected clay Hner developed
severe desiccation cracks. These were not repaired prior to filling; con-
sequently, contaminated liquid passed through the liner and began accumulat-
ing in the leak detection system. This problem, which was caused by the arid
climate and the loss of water from the liner, could have been avoided with
adequate liner maintenance and a rigorous preservice inspection.
7.5 INSTALLATION OF CLAY LINERS
This section presents a very brief discussion of clay Hner installation
methodology and procedures. Only one case study 1s Included for discussion
7-74
-------
because 1t is very difficult to attribute a liner failure to poor construc-
tion techniques. For a complete discussion of this topic, see Chapter 5.
7.5.1 Installation Methods
It is widely believed that the performance of clay liners is affected by
the equipment and procedures used to install them. Conclusive field data to
support this belief, however, are not available. Specific items that need to
be controlled during clay liner construction are discussed in the following
sections.
7.5.1.1 Excavating, Grading, and Foundation Preparation--
The excavation and grading of the bottom and side slopes of the liner
foundation should be conducted according to design specifications. This
generally requires the use of construction equipment such as dozers,
backhoes, scrapers, and graders. For unlined facilities constructed in low--
permeability soil, special care must be taken to locate and fully excavate
all permeable zones that may allow for leachate migration. For facilities
that are to be lined with recompacted local soil or admixed materials,
suitable liner materials are generally excavated, stockpiled, and used as
required. In all cases, the specified slopes and elevations of sidewalls and
bottoms as well as those of collection system pipes and manholes must be
carefully controlled. This will help ensure that the liner is properly
located regarding the water table and local geological features. Specified
side and bottom slopes are also necessary so that leachate will flow to the
collection system pipes and sump, thus preventing the "bathtub" effect.
7.5.1.2 Liner Materials--
Native soil liner material may be excavated from the site during
foundation preparation or may be brought to the site from a nearby borrow
area. Admixture materials may be shipped to the facility in bags or bulk
form. In all cases, the use of materials that are specified in the facility
design is extremely important. As previously discussed, materials such as
sand, rocks, roots, or other organic matter, if included in the liner, will
greatly affect its permeability and ultimate performance. When materials
such as bentonite are to be added to in situ materials to form a liner, the
bentonite content must be carefully controlled. The use of inadequate
quantities of bentonite will produce a liner with higher than required
permeability and lower than specified performance.
7.5.1.3 Moisture Content/Density/Compactive Effort--
The relationship among moisture content, density, and compactive effort
is discussed in Chapter 2. Careful control of these parameters is necessary
throughout all phases of liner construction. The final liner permeability
and performance depend on proper moisture control and distribution, mixing,
and compaction techniques. Too little or too much water added to the liner
material will make it difficult to compact to the specified density, thus
affecting permeability. Inadequate mixing of materials may result in a
heterogeneous liner. Insufficient compactive effort may result in a liner
that is more permeable than required.
7-75
-------
7.5.2 Quality Assurance/Quality Control For Clay Liners
A survey of hazardous waste surface impoundment technology has found
that rigorous quality assurance is necessary to achieve good site performance
(Ghassemi et al., 1984). Liner failures at several impoundments included in
the survey were attributed to various factors including "failure to execute
proper quality assurance and control." The success of surveyed facilities
that have performed very well is attributed to many factors including "the
use of competent design, construction and inspection contractors, close
scrutiny of all phases of design, construction and QA inspection by the
owner/operator, excellent CQA/CQC and recordkeeping during all phases of the
project, and good communications between all parties involved in establishing
the sites."
Specific problems that may result in clay liner failure and that can be
avoided with careful CQA include:
• Use of materials with specifications other than the ones in the
approved design
• Lack of careful screening and testing of incoming materials to remove
roots and other organic matter, rocks, pockets of permeable
materials, and other foreign objects prior to placement
• Lack of adequate moisture control both prior to and after compaction
• Improper size reduction, mixing, and spreading of Uner materials
t Use of inadequate liner materials (especially important with
bentonlte/soil liners)
t Failure to follow installation procedures specified in the design
t Use of improper construction equipment
• Application or specification of inadequate compactive effort.
The following case study is of a surface impoundment that has had
performance problems. Due to insufficient information, this case study is
not Included In Section 7.2. The problems at this facility may have resulted
from a combination of factors, one being the lack of adequate CQA.
A sewage-treatment lagoon was to be lined with 3.7 pounds of
polymer-treated bentonite per square foot of liner mixed with the native soil
to a depth of 4, inches. The particle size of the bentonite that was supplied
did not meet the original specifications, making it very difficult to spread
and mix the material adequately. Problems were also encountered in achieving
proper moisture content. After the bentonite/soil layer was compacted, the
final permeability of the liner was to be determined by filling the lagoon
with water to a depth of 15 feet. The decrease in .liquid head and the
evaporation rate were then to be used to determine the as-built liner
permeability. Unfortunately, the as-built liner permeability was much
greater than expected. Actual as-built permeabilities were never determined,
however, because the water flowed through the Uner faster than it could be
7-76
-------
pumped Into the lagoon. Examinations of the liner profile revealed large
discontinuities in the liner materials, which explains the high
permeability. - v- ^
A second attempt to construct the liner to the same specifications was
successful. All phases of liner installation (by a different contractor)
were carefully monitored by the site owner. Final permeability testing con-
firmed that the liner exceeded the required specifications. A cross section
of this liner revealed uniform thickness and material content.
A complete discussion of CQA/CQC may be found in Chapter 5 of this
document.
7.6 WASTE TYPES
Prior to recent Federal regulations (Federal Register, 1982), unsolidi-
fied liquid wastes either containerized or in bulk form were often disposed
of in landfills. Since these recent regulations, free-standing liquid wastes
are required to be removed from drums and solidified prior to final dis-
posal. While the option still exists to store liquids in surface impound-
ments or evaporation ponds, free liquids must be removed prior to final
facility closure.
7.6.1 Free Liquids
Many performance problems have resulted from the practice of placing
containerized or bulk free liquids in waste disposal facilities. With bulk
liquids, a liquid head is imposed on the liner. If the head becomes too
great or if the waste material is incompatible with the liner material, the
waste has a much greater chance of infiltrating into or through the liner.
Fifty-five-gallon drums, which are often used to contain free liquids,
have the potential to degrade with time. This can result in leaky drums,
which, in turn, may cause leachate levels to build and the performance
problems associated with this phenomenon to occur. Several problems are also
associated with the waste placement and closure operations of a facility that
contains drums. Drums are usually stacked upright 1n several layers or
lifts. Absorptive material such as soil is then placed between the drums and
compacted to fill all voids. It is difficult to determine when all of the
lower voids are filled and when the material has been compacted properly. If
the spaces are not completely filled, piping Into the spaces between the
drums may occur. This can result in differential settlement and cap failure,
which may lead to Increased liquid infiltration and hydraulic gradient on the
liner, thus affecting its performance.
An example of a facility where a performance problem has resulted, at
least in part, from free liquid disposal is site B. This facility is
approximately 13 years old. It is a zone-of-saturation (intergradient)
facility where liquids were not solidified prior to disposal. These
unsolidlfied liquids along with the accumulated precipitation and seepage
resulted in a "bathtub" with leachate seeps penetrating the cap/liner
interface.
7-77
-------
Present landfill regulations require that leachate collection and
removal systems must be installed immediately above the liner. These systems
must be designed to ensure that the leachate depth does not exceed 30 cm.
Properly functioning leachate collection systems along with the regulation
preventing free liquids from being disposed of in landfills should prevent
similar future performance problems from occurring.
7.6.2 Stabilized or Solidified Liquids
As previously mentioned, Federal regulations require that free liquids
not be placed in landfills. In response, free liquids are now removed,
solidified, or stabilized prior to their final disposal. Removed liquids may
be recycled, impounded, used as fuel, or in some way treated and then placed
in a landfill either in drums or bulk. The solidification process may be
accomplished by mixing the liquid with absorptive material such as lime,
cement kiln dust, fly ash, or some other substance. Stabilization of wastes
may include processes such as neutralizing acidic or basic wastes or
precipitating heavy metals. These processes generally decrease the toxic or
hazardous nature of the waste and are currently practiced at many of the case
study sites in Section 7.2.
A procedure that provides protection from leaking drums is practiced at
site E. Here, a "calculated amount" of "absorptive material" is placed
beneath each lift of waste. This material will theoretically absorb any
liquids that might leak from the drums. The success of this method is not
proven at this time as site E has had large volumes of leachate accumulating
in some of its waste trenches.
7.6.3 Sludges and Solid Wastes
7.6.3.1 Sludges--
Depending on the water content, sludges may be either landfilled or
placed in evaporation ponds. However, in both cases, it is important that
most of the moisture be removed prior to final facility closure. Case
studies M, P, and Q are all landfills that contain various types of sludges.
All of these facilities contain both leachate collection systems and some
form of leak detection system.
Site M uses a special procedure for waste placement. This procedure
Involves mixing a thin layer (approximately 2 feet) of locally available soil
with a thin layer of the inorganic lime sludge, thereby reducing the overall
moisture content of the material and giving it structural stability. The
range of the soll-to-waste ratio varies from 0.5:1 to 1.5:1. Exact mixing
proportions are determined in the field and judged sufficient when the
landfill equipment is able to drive over the mixture.
Facilities P and Q contain brine purification mud and metal hydroxide
and waste pigment sludge, respectively. These facilities are both lined with
bentonlte/soil admixtures. The bottom and lower side liners at both of these
facilities are thicker than the upper side liners to provide extra protection
1n areas where a liquid head may be present.
The following case study 1s not included in Section 7.2 due to insuf-
ficient information. Enough information is available, however, to allow a
7-78
-------
a failure that occurred at this facility to be attributed to improper waste
placement procedures. This double-lined facility was originally designed for
the disposal of solid wastes. A change in the parent company's waste
disposal needs resulted in this facility accepting high-liquid-content
sludges. The sludges were transported to the site in tank trucks that would
back up to the facility and dump their load through a pipe whose outlet was
well above the base of the liner. The impact of the sludge on the liner
caused severe liner erosion and groundwater contamination probably as a
result of an erosional breach in the liner.
7.6.3.2 Solid Waste--
Solid wastes may be landfilled in containers such as 55-gallon drums or
in bulk form. When wastes are landfilled in drums, all free liquids must be
removed or solidified prior to final disposal. Care must be exercised to
make certain that the voids between adjacent drums are filled with well-
compacted soil or other material. Problems that may result from the disposal
of drums are discussed in Section 7.6.1.
Solid wastes such as contaminated soil or other fine uniform material
may be emplaced and compacted in lifts. When this method is used, the void
volume in the landfill is reduced; thus, future problems due to settlement or
piping are less likely to occur. Solid wastes that are not uniform, such as
unshredded tires, scrap metal, and municipal solid wastes, on the other hand,
are more difficult to compact thoroughly and therefore have the potential to
settle differentially. This can be minimized by crushing or shredding the
waste, thereby reducing its void spaces and overall volume. The use of a
landfill compactor will also help to prevent future problems due to piping
and settling.
7.6.4 Waste Compatibility
7.6.4.1 Waste/Liner Compatibility--
Laboratory tests suggest that dilute aqueous leachates will not affect
the permeability of clay liner materials if the moisture at compaction is
uniform and close to optimum and if compaction is uniform. However, data
from several studies suggest that drastic increases in permeability can
result from certain clay-chemical interactions. Examples of organic
chemicals that have been demonstrated to increase clay soil permeability are:
aliphatic and aromatic hydrocarbons (e.g., cyclohexarre, heptane, kerosene,
naphtha, benzene, and xylene), alcohols (e.g., methanol and ethylene glycol),
ketones (e.g., acetone and dioxane), amines (e.g., aniline and pyridine),
carbon tetrachloride, and nitrobenzene. Changes in permeability have also
been noted with strongly acidic permeant fluids.
Bentonite is sometimes given a polymer treatment to improve its
resistance to the effects of normally incompatible fluids. The long-term
viability of polymer-treated products needs to be verified.
The enormous variability in clay soil from different locations
complicates the task of predicting clay-chemical compatibility. Data from
several laboratory studies suggest the possibility of drastic increases in
permeability as a result of certain clay-chemical interactions. It should be
emphasized, however, that many aqueous leachates have been tested and
7-79
-------
produced no significant increases in permeability. For a detailed discussion
of clay-chemical interactions, see Chapter 4 of this document.
Facilities C and 0 have both had liner failures that have been attrib-
uted to liner/waste incompatibility. In the case of site C, an investigation
of groundwater contamination identified the facility's treatment ponds as the
probable source of the pollution. This conclusion was based on the fact that
the materials in the treatment ponds were usually of low pH and would have
reacted with carbonate seams and inclusions in the surrounding soil. Several
of the disposal trenches at this facility were also used for the disposal of
highly acidic waste, in this case oil reprocessing sludges. These acidic
sludges may have also reacted with the carbonate inclusions, creating paths
for leachate seepage and groundwater contamination.
The Uner failure at site J occurred almost overnight. Samples of
liquid removed from the leak detection system at one of the ponds on one day
were reported to be "clean." Samples taken the following day contained over
11.2 percent perchloroethylene plus small amounts of other chlorinated and
nonchlorinated organics. The facility was allegedly not accepting wastes
with more than a trace of perchloroethylene.
An investigation into the problem revealed that aqueous wastes coming
from one of the disposal facility customers contained approximately 5 percent
degreaslng fluid waste that contained perchloroethylene. This waste was to
have been stored in a separate tank at the originator's plant. Evidently, it
was not made clear to the workers where to put the "perc" waste, and it was
placed in an aqueous waste storage tank. Being insoluble and heavier than
water, it sank to the bottom. When a tank truck from the disposal facility
would collect a load of aqueous waste, it would also receive a few hundred
gallons of the "perc" waste. The waste sampling method used by the disposal
facility on all loads of incoming waste was not able to sample the bottom few
inches of each load. In addition, the parking area where all incoming
truckloads of waste were sampled was slightly sloped, causing the relatively
small amount of "perc" waste in the bottom of the tanker to flow to the back
of the truck where it could not be reached by the sampling device.
Over a period of approximately 9 months, the aqueous waste containing
"perc" was placed into one of the clay-lined ponds and went unnoticed until
the Uner failure occurred. The apparent incompatibility is consistent with
current research on the effects of chlorinated solvents on clays.
7.6.4.2 Waste/Waste Compatibility--
Large disposal facilities that accept most types of wastes, such as
sites E and F, generally have a waste placement plan. Such a plan is used to
separate incompatible or reactive materials and in some cases to place
together materials that produce "favorable reactions."
Most cells at site F are divided into five subcells. The purpose of
these subcells is to isolate the various waste groups accepted'at the
facility, thereby preventing the interactions of incompatible wastes. The
7-80
-------
waste categories, percentage of total wastes, and separation rationale at
site F are as_follows:
• - * .«?•' • • •
• General wastes. These wastes represent approximately 44 percent of
the total waste volume accepted at site F. General wastes are
defined as materials of both an organic and inorganic nature that do
not contain a significant quantity of any of the other waste
categories.
The hazardous acidic or acid-generating materials are covered
with lime to ensure that any acid that is generated will be
neutralized.
• Pseudo metals. This type of material represents approximately 6
percent of the total waste volume accepted at site F. Pseudo metals
are arsenic, antimony, bismuth, and phosphorous. Chalcogens,
beryllium, and any of their compounds as well as alkaline-sensitive
materials are also disposed of in this subcell.
This subcell has a pH buffer system that maintains pH levels
between 6 and 8.
• Heavy metals. These wastes represent approximately 15 percent of the
total waste volume accepted at site F. This group is comprised of
all heavy metals and asbestos. This subcell contains the smallest
amount of organic materials, which helps to reduce fire hazards
caused by the reaction of strong oxidizing agents with organics.
« Highly flammable wastes. This type of material represents
approximately 12 percent of the total waste volume accepted at site
F. These materials generally exhibit a flash point between 80 and
100 F. These materials are kept apart from powerful oxidizing
agents, materials that are prone to spontaneous heating, or materials
that react with air or moisture to evolve heat.
(i Toxic materials. These wastes represent approximately 23 percent of
the total waste volume accepted at site F. Included in this category
are all waste materials containing more than 15 percent by weight of
highly toxic organic compounds, carcinogens, PCS's, and other
halogenated wastes. No solvent-type wastes are permitted in this
subcell.
7.7 PERFORMANCE MONITORING
Several methods can be used to monitor the performance of clay-lined
landfills. These methods generally involve groundwater monitoring or
monitoring of the unsaturated zone directly beneath the clay liner.
Monitoring the quantity and quality of the leachate above the liner may also
give some indication of the facility performance. This type of monitoring
program for a facility should be able to detect performance problems, such as
leachate migration, soon after they occur. This will enable the rapid
remediation of any problems and prevent serious groundwater contamination.
7-81
-------
7.7.1 Unsaturated Zone Monitoring
Two systems that are commonly used to monitor the unsaturated zone
beneath a clay liner are lysimeters and continuous-coverage leak detection
systems. A continuous-coverage leak detection system generally consists of a
flexible membrane overlain by a porous layer that contains perforated
collection pipes. Just as with a leachate collection system, the porous
layer is usually sloped so that any detected fluid will drain to a sump.
Here the liquid may be sampled and analyzed for hazardous constituents.
Total liquid volumes removed from a detection layer may also be used to
estimate the liner permeability.
The advantages of a continuous-coverage leak detection system include
its ability to provide early indications of the liner performance and its
ability to detect and collect all liquids that appear in the detection
system. The disadvantages of a continuous-coverage detection system include
its relatively high cost and the possibility that the leachate, if a chemical
compatibility failure should occur, might spread out upon and damage a large
portion of the lower clay liner or foundation soil.
Another approach to unsaturated zone monitoring involves the use of
either pressure-vacuum lysimeters or collection lysimeters. Pressure-vacuum
lysimeters consist of a porous cup, placed in a bore hole, that a vacuum may
be applied to. This vacuum causes pore liquids to collect in the porous
cup. The liquid is then removed and analyzed for hazardous constituents. A
collection lyslmeter, on the other hand, is a lined gravel or sand-filled
basin or trench beneath the primary liner. A collection pipe located at the
low point of the lysimeter leads to a sump from which any detected liquids
can be sampled and analyzed.
The following discussion on the advantages and disadvantages of
collection and pressure-vacuum lysimeters is taken from Kmet and Lindorff
(1983).
Both pressure-vacuum lysimeters and collection lysimeters can be used to
monitor the unsaturated zone beneath a landfill and provide an early
Indication of landfill performance. Each has its own advantages and
disadvantages.
«>
The major advantage of a collection lysimeter is that it provides a
method of monitoring the quantity as well as the quality of leakage from
the base of a landfill. If one assumes that the leakage reaching the
lysimeter is representative of leakage over the entire base of the site,
it is possible to calculate the volume of leakage from the entire
landfill and thus assess the effectiveness of the liner in attenuating
the leachate and in limiting the rate of leachate movement.
Also, collection lysimeters do not require the special equipment or the
care necessary to sample pressure-vacuum lysimeters properly. Samples
can be taken from a manhole or riser by a bailer rather than a vacuum
pump. In addition, collection lysimeters do not require that a vacuum
be placed on the system at least several days prior to sampling, as is
the case with pressure-vacuum lysimeters. When the monitoring results
from a landfill indicate that one or more pressure-vacuum lysimeters are
7-82
-------
dry, one is never sure whether there was insufficient soil moisture or
whether jmproper sampling techniques were used. With collection
lysimeters, one is reasonably certain that if liquid is present in the
subsurface, it will be collected. Collection lysimeters also eliminate
the problem of frozen lines in cold weather that can occur with
pressure-vacuum lysimeters.
Research, including work done by Apgar and Langmuir (1971); Johnson and
Cartwright (1980), has suggested that some chemical constituents may be
filtered out by the porous membrane of the pressure-vacuum lysimeter,
thereby producing inaccurate water quality data. This is not a concern
with collection lysimeters since no membrane is present.
The major disadvantage of a collection lysimeter in comparison with a
pressure-vacuum lysimeter is the lower cost and shorter time required
for pressure-vacuum lysimeter installation. Whereas a pressure-vacuum
lysimeter is installed in a bored hole, construction of a collection
lysimeter necessitates excavation of sufficient area to appropriate
grade to install the basin, collection pipe, and manhole. A substantial
number of pressure-vacuum lysimeters could be installed in the time and
for the cost of one collection lysimeter. A pressure-vacuum lysimeter
can be installed for less than $100 plus drilling costs. Although data
is [sic] limited, the cost of collection lysimeter installation,
including excavation, materials, and construction costs, is expected to
be several thousand dollars and can approach $10,000 for the larger
lysimeters.
Another disadvantage is that only one collection lysimeter can be
installed at a given location. Therefore, data is [sic] generated for
only one depth in the unsaturated zone, typically just below the clay
liner. In contrast, pressure-vacuum lysimeters are frequently installed
in a nest at different depths to monitor changes in water quality with
depth.
Several of the facilities discussed in Section 7.2 of this document have
some type of unsaturated zone monitoring system. Six of these
facilities—sites H, L, M, N, P, and Q—are landfills and three—sites I, J,
and K—are ponds. Sites H and N both have lysimeters beneath their clay
liners. Facility H has two I,200-ft2 lysimeters directly beneath the
4-foot-thick recompacted clay liner. An additional lysimeter is located
under the leachate storage basin. For approximately 9 months after the start
of landfill ing operations, small quantities of liquid were removed from the
lysimeters under the landfill liner. These liquid volumes declined steadily
over time and were reduced to zero 1 year after the start of landfill
operations. The liquid that was initially collected in the lysimeters is
most likely soil moisture that was released after liner construction.
Facility N also has two lysimeters beneath the liner. This facility
also has a groundwater drain that is used to lower the groundwater table. An
interesting fact about this facility is that although highly contaminated
liquids have recently been removed from the lysimeters, 'contaminated liquids
were first discovered in the groundwater drain manhole. This has continued
to be the location where the majority of contamination has been detected.
7-83
-------
The three ponds with unsaturated zone monitoring all have continuous-
coverage leak_detection systems between two recompacted clay liners. At two
of the ponds, facilities I and J, major performance problems have developed
in the upper clay liners. In both cases, the problems were quickly detected
(overnight at site J) and corrected, thus preventing any contamination from
penetrating the lower liner. The third double-lined pond, site K, has had
small quantities of liquid accumulating in its leak detection system. These
volumes were used to calculate the overall liner permeability, which
ranged from 4 x 10~8 cm/s to 3 x 10~7 cm/s, demonstrating that the
original specifications of 1 x 10~6 cm/s had been met.
Four of the six landfills, facilities L, M, P, and Q, have continuous-
coverage leak detection systems. Liquids have been detected in the leak
detection layers of sites L, P, and Q. However, only one, site L, is
considered a failure. Here, slightly contaminated liquids were removed from
the detection system manhole less than 1 year after the start of operation.
Facility Q also had slightly contaminated liquids removed from its detection
system. The volume of liquid removed was used by the facility owner to
calculate the admixed liner's permeability, which ranged from
3 x ICT8 cm/s to 6.5 x 10~8 cm/s, indicating that the liner met the
original permeability specification of 1 x 10~7 cm/s. The as-built liner
permeability at facility P has not been determined. The source of small
amounts of uncontaminated liquids in the detection system at this facility
was being investigated by the site owner. The results of this investigation
were not available at this writing.
7.7.2 Groundwater Monitoring
Information on Federal groundwater monitoring requirements may be found
in "Groundwater Monitoring Guidance for Owners and Operators of Interim
Status Facilities" (U.S. EPA, 1983) and in the RCRA Ground-Water Monitoring
Technical Enforcement Guidance Document (U.S. EPA, 1986).
Groundwater monitoring wells are the most commonly used type of
performance monitoring systems. They are relatively inexpensive to install
and maintain, and samples are easily obtained from them. They may be placed
in areas where contamination is estimated to be most likely to occur. They
may also be placed in groups or clusters with each well being at a different
depth and therefore monitoring the vertical as well as lateral movement of
any contamination. Generally, wells are placed both upgradient and
downgradient from a facility prior to the start of operations. This enables
baseline data for both upgradient and downgradient wells to be obtained.
Future groundwater samples may then be compared with the baseline data and
facility performance can be estimated.
Two major problems with relying solely on monitoring wells to determine
facility performance are that leachate plumes may be missed and that the
monitoring data are only qualitative. If an adequate number of wells at
varying depths are used, the possibility of missing a contaminant plume is
greatly decreased.
An additional problem that may occur with monitoring wells is poor
construction procedures such as the use of inappropriate materials. This can
7-84
-------
often lead to well-water contamination that is Incorrectly attributed to the
waste disposal facility. Examples of such problems occurred at sites A and
E. Groundwater samples taken from the wells at site A in 1976 had pH levels
between 10.7 and 11.3, COD's of more than 3,000 ppm, and chloride levels
greater than 1,000 ppm. Originally, it was thought that this contamination
originated in the disposal facility. A complete analysis of the problem,
however, revealed that the monitoring wells were Improperly installed and
contained steel-mill slag where clean gravel should have been used. Analysis
of samples taken from new wells, constructed according to the specifications,
has not indicated elevated levels of any hazardous constituents.
Monitoring well analysis at site E has shown trace levels (92 ppb
maximum) of 12 organic compounds in the groundwater. An investigation of
their source indicated that the organics most probably came from the plastic
well casing materials. This case study points out the importance of testing
the compatibility of all materials with the local groundwater as well as the
expected leachate.
7.7.3 Leachate Level and Quality Monitoring
Leachate collection systems are designed to remove leachate that would
otherwise accumulate in the liner. By doing this, they prevent the buildup
of leachate head, thus keeping liquid infiltration into the liner to a
minimum and therefore contributing to the overall performance of the liner.
Some facilities incorporate leachate level and leachate quality monitor-
ing into their overall monitoring program. These parameters serve as indices
of the efficiency of treatment methods that they may have been used on waste
materials and the leachate collection system. Analysis of the leachate may
be used to predict the quality of the liquid, if any, that may be passing
through the liner. Leachate and liner samples can be subjected to laboratory
compatibility tests to indicate the facility performance.
7.8 CONCLUSIONS
The sites included in our study and a brief description of their clay
liners and performance are presented in Table 7-1 and are fully described in
Section 7.2. The 17 sites included in this section were chosen because they
illustrate a specific performance problem or facility characteristic. While
conducting our information search, we encountered many facilities both with
and without failures that were not included here because they would not have
provided any additional Information.
Comparing the data from successful sites with the data from leaking
sites does not reveal any significant differences in clay or liner character-
istics that might explain the differences in performance. Similarly,
performance does not seem to be related to the use of recompacted borrow clay
versus recompacted native clay or admixed materials. The performance of
facilities without recompacted liners is, however, generally poorer than the
performance of facilities with a recompacted clay or admixed liner. Also
note that many of the sites discussed in this section are relatively new
(after 1975). It is therefore difficult to state whether the facilities
without detection systems are performing well or poorly as sufficient time
for leaching chemicals to reach monitoring wells may not have elapsed.
7-85
-------
Performance of the facilities that have detection systems, on the other hand,
is much more quickly determined than is the case with facilities I, J, L, N,
and Q. Sites I, J, L, and N have reported performance problems. In all
cases, the leak detection system was instrumental in the rapid detection of
the performance problems. Site Q has had small amounts of liquid detected in
its leak detection system. The volume of liquid removed was used to deter-
mine the permeability of site Q's upper liner. This value was lower than the
design specification, indicating good performance. Of the 17 facilities dis-
cussed in Section 7.2, 6 have had major liner failures. Of these six, the
cause of four of the failures is known and the cause of the failure of the
remaining two is unknown. Generally, the cause of a failure in a clay-lined
landfill Is very difficult to determine because this would require removing
the waste and other overburden, which may in itself cause liner damage and
Impose further hazards. Determining the cause of a liner failure in a lagoon
or evaporation pond is somewhat less difficult because the waste is more
easily removed.
The four facilities with explained failures include three lagoons or
ponds and a landfill that accepted unsolidified liquids. Two of these
facilities had leak detection systems; the others did not. The failures at
the facilities with detection systems were noticed soon after they occurred,
thus preventing groundwater contamination. The other two failures were not
detected until groundwater contamination was found in a monitoring well. Two
of the failures were due to chemical compatibility problems, one to leachate
migration through an improperly sealed sand seam, and the other to desicca-
tion cracks that formed in an unprotected liner.
Sites with a single layer of clay or those relying on in situ clay for-
mations are less secure since a leak cannot be detected until pollutants are
found in groundwater samples. The security of a landfill would be greatly
Increased if it were constructed with a continuous-coverage leak detection
system Installed between two layers of clay. If the Inner layer failed, the
leak detection layer would serve as a buffer zone that might preserve the
integrity of the outer layer. The use of lysimeters as a method for deter-
mining liner performance also may provide an early indication of leachate
migration. However, these devices are not foolproof because they are only
able to detect leaks 1n the section of the liner directly above them.
7.9 REFERENCES
Apgar, M., and D. Langmulr. 1971. Groundwater pollution potential of a
landfill above the water table. Groundwater 9(6):76-96.
Bagchl, A. 1987. Discussion on "Hydraulic Conductivity of Two Prototype
Liners. ASCE Journal of Geotechnical Engineering, July 1987,
113(7):796-820.
Day, S. R., and D. E. Daniel. 1985. Hydraulic Conductivity of Two Prototype
Clay Liners. ASCE Journal of Geotechnical Engineering, Agust 1985,
111(8):957-970.
Federal Register, Monday, July 26, 1982, Vol. 47, No. 143.
7-86
-------
Ghassemi, M., M. Haro, and L. Fargo. 1984. Assessment of Hazardous Waste
Surface Jmpoundment Technology (Case Studies and Perspectives of
Experts). Draft. Prepared for the U.S* Environmental Protection
Agency, Contract No. 68-02-3174, Work Assignments 97 and 123.
Multidisciplinary Energy and Environmental Systems and Applications,
Torrance, California.
Johnson, T. M., and K. Cartwright. 1980. Monitoring of Leachate Migration
in the Unsaturated Zone in the Vicinity of Sanitary Landfills.
Circular 514, Illinois State Geological Survey, Champaign, Illinois,
82 pp.
Kmet, P., and D. W. Lindorff. 1983. Use of Collection Lysimeters in
Monitoring Sanitary Landfill Performance. Paper presented at the
National Water Well Association Conference on the Characterization and
Monitoring of the Vadose (Unsaturated) Zone, Las Vegas, Nevada,
December 8-10.
Rogowski, A. S. 1986. Hydraulic Conductivity of Compacted Clay Soils.
In: Land Disposal, Remedial Action, Incineration, and Treatment of
Hazardous Waste - Proceedings of the Twelfth Annual Research Symposium.
(EPA/600/9-86/022) U.S. Environmental Protection Agency, Cincinnati,
Ohio. pp. 29-39.
Rogowski, A. S., B. E. Weinrich, and D. E. Simmons. 1985. Permeability
assessment in a compacted clay liner. In: Proceedings of the 8th
Annual Madison Waste Conference on Municipal and Industrial Waste,
University of Wisconsin, Madison, pp. 315-336.
U.S. Environmental Protection Agency. 1983. Ground-water Monitoring
Guidance for Owners and Operators of Interim Status Facilities. Office
of Solid Waste and Emergency Response, Washington, D.C.
U.S. Environmental Protection Agency. 1986. RCRA Ground-Water Monitoring
Technical Enforcement Guidance Document, OSWER-9950.1, September 1986.
Office of Solid Waste and Emergency Response, Washington, D.C.
7-87
-------
-------
CHAPTER 8
PREDICTION OF CLAY LINER PERFORMANCE
8.1 INTRODUCTION
Although recompacted clays may have low hydraulic conductivity, they
are, nonetheless, permeable porous media. If a finite depth of liquid waste
or leachate is maintained indefinitely over a clay liner, the liquid and
leachate chemicals will eventually seep through. The goal of transit time
prediction is to determine both the rate of seepage with time and the time it
will take for liquids to seep through a liner.
Generally, transit time prediction methods may be used in two ways.
First, they may be used to facilitate the design of new clay liner systems
and second, they may be used to predict the performance of existing clay
liner systems and to determine the potential for groundwater contamination
from leachate or liquid wastes.
Seven transit time prediction methods are discussed in this chapter.
Section 8.2 is. a review of the assumptions and basic equations that underlie
the use of all these methods. Section 8.3 is a discussion of the derivation
and use of each method. Section 8.4 is a comparison of the consequences—
i.e., the predicted transit time, in years—of using each method.
The prediction of transit times generated by various models cannot be com-
pared with any actual liner data because no information regarding liner
infiltration and breakthrough under certain conditions is currently avail-
able. Section 8.5 offers a summary and some conclusions regarding the use of
the different methods. Section 8.6 is a discussion of the use of batch type
adsorption data for predicting liner performance. References are contained
in Section 8.7. *
8.2 BACKGROUND CONSIDERATIONS
8.2.1 Performance Criteria
In order to design or evaluate a liner system it is necessary to
establish specific performance criteria that will provide a definition for
transit time." These performance criteria should be designed so that, if
they are satisfied, the liner system will ensure safe operation over the
intended life of the facility or up to a certain time after closure.
Performance criteria may be expressed 1n terms of seepage flux (volumetric
flow rate of seepage in a porous medium per unit cross-sectional area)
contaminant flux (indicated by the concentration of a chemical that is
present in the leachate), or the time taken to reach a specified flow rate or
8-1
-------
chemical concentration at the bottom of the liner. The following are five
examples of possible specific liner performance criteria:
t Specified seepage flux at the liner bottom
• Specified contaminant flux for a particular leachate component
at the liner bottom
• Time taken to reach a specified seepage flux at the liner bottom
• Time taken to reach a specified leachate chemical flux at the
liner bottom
• Time taken for the concentration of a specific leachate com-
ponent to reach a specified value at the liner bottom.
Selection of specific performance criteria is very important in predict-
ing corresponding "transit times," and it can also influence the applicabil-
ity of a given method.
Recommendations regarding the use of specific performance criteria are
beyond the scope of this work. However, examples of the consequences of dif-
ferent performance criteria will be considered in the evaluation of transit
time prediction methods.
8.2.2 Clay Liner System
The equations and boundary conditions used in transit time prediction
methods are based on a specific site geometry or flow domain. This flow
domain describes the path of waste leachate through a clay liner system. For
the sake of this discussion it consists of, in a vertically downward direc-
tion, saturated solid waste; a sand/gravel bed on top of the clay liner
(leachate collection system); a clay liner; a sand/gravel leak detection
system; underlying local soil; and an underlying saturated aquifer zone.
Current regulations (40 CFR 264.301(a)(2)) require a leachate collection sys-
tem on top of a clay liner to maintain a leachate head of less than 1 foot.
This makes 1t unnecessary to Include the saturated solid waste zone and the
leachate collection system zone in the flow domain. The sand/gravel leak
detection layer may not be present in some designs. -A schematic of the flow
domain that is used to illustrate the transit time equations discussed below
1s shown in Figure 8-1.
Because the hydraulic conductivity of the clay liner material 1s much
lower than the underlying "site soil, the rate of leachate flow is predomi-
nantly controlled by the clay liner with the underlying soil influencing only
the boundary conditions at the bottom of the clay layer. Some of the transit
time methods are thus based on leachate flow through the clay Uner alone.
Due to the large ratio of liner width to liner depth, it is adequate to
consider vertical flow alone to describe the leachate flow.
Any attenuation capacity of the clay or temporary immobilization of some
contaminants tends to slow the leachate migration. Failure to account for
such factors will result in underestimates of transit time.
8-2
-------
Impounded Liquid Head
in Leachate Collection Zone
Clay Liner
Leak Detection System
Local Site
Soil
Saturated
Aquifer
Figure 8-1. Flow domain for leachate flow.
8-3
-------
Note that throughout this section only aqueous leachate systems are con-
sidered; i.e._, we are considering only the flow of water and dissolved
species through the flow domain. Although some organic-solvent-based leach-
ates may be immiscible with water, analysis of such a three-phase system (two
immiscible liquid phases + air) is extremely complex and none of the transit
time prediction methods discussed below (except numerical methods) apply to
this situation.
8.2.3 General Equations
A general one-dimensional equation may be used to describe the vertical
flow of fluid through the saturated/unsaturated flow domain. The equation
may be written as follows (Bear, 1979; Huyakorn and Pinder, 1983):
O O, . ^ ti>Tf
,worv,I./o*1x-i fs •n
05TT- = 5^ I.KC Kr (^r ~ 1/J (°'L)
*dt az s r oz
z = vertical coordinate, expressed as positive downward distance
Ks = saturated hydraulic conductivity
kr * relative hydraulic conductivity with respect to Ks
* - pressure head
- porosity
Sw s fractional saturation (equal to 1 for saturated media)
t = time.
For saturated environments, the equation transforms into a steady state
formulation and one can determine the steady state flux. For unsaturated
conditions, this equation describes the advancement of the saturated wetting
front and also the unsteady state fluid flux as a function of position.
The general flow equation does not consider the transport of soluble
chemicals in the leachate. A general equation to destribe the vertical
transport of a nonradioactive solute species may be written as follows
(Bear, 1979; Huyakorn and Pinder, 1983):
n m 3C^
WC) 9 (D ) 3
_ a _
9z 9z 3z
where
D = dispersion coefficient that includes both molecular diffusion
and mechanical dispersion
C » solute concentration in liquid phase
z * vertical coordinate, expressed as positive downward distance
8-4
-------
v = Darcian velocity in z direction
"s '& *f .-.-
R = retardation factor to account for attenuation capacity of the
clay or soil.
The source and sink terms are not included because they are not
expected to be significant. For a conservative species, the retardation
factor equals 1. The Darcian velocity, v (cm/s), is obtained from the flow
equation and is given by:
v ' -Ks kr
-------
methods. A review of currently available techniques is presented in "Soil
Properties, Classifications, and Hydraulic Conductivity Testing." (U.S. EPA,
1985)
The solute transport equation involves two additional parameters: a
retardation factor and the axial dispersion coefficient. The retardation
factor is dependent upon the attenuation capacity (e.g., adsorption) of the
soil medium and needs to be determined experimentally. Note also that the
attenuation capacity of soil is time dependent and it is reduced as the soil
approaches saturation with respect to adsorbed species. For conservative
liner thickness predictions, it is probably best to assume a retardation
factor of unity (no attenuation).
The axial dispersion coefficient consists of two parts: hydrodynamic
dispersion and molecular diffusion. The first process is caused by mixing
due to variations in fluid velocity associated with distance from pore
walls. Diffusion occurs in response to concentration gradients and by
random thermal motion. The molecular diffusion coefficient should be
readily available from the literature, whereas measurement of the hydro-
dynamic dispersion coefficient is tedious and time consuming, and no
standardized methods are available for this purpose.
8.3 TRANSIT TIME PREDICTION METHODS
The general equations describe leachate flow and solute transport in
the flow domain shown in Figure 8-1. Solving these equations will give the
seepage flux, leachate flux, moisture profiles, and concentration profiles
as a function of time. Together with the appropriate performance criteria,
these equations may be used to assess the performance of an existing
facility or to design an effective new facility. These equations are com-
plex, and they can only be solved numerically; but with certain assumptions
1t 1s possible to simplify the equations and allow analytic solutions. Each
of the different transit time prediction methods described below uses some
form of simplification to achieve this end. In a report attached as
Appendix A to U.S. EPA 1984, Cogley et al. reviewed several transit time
equations.
8.3.1 Simple Transit Time Equation
a
A simple transit time equation can be used to estimate the necessary
bottom liner thickness as a function of the design life of an impoundment
(U.S. EPA, 1984, Appendix A). This equation assumes a flow domain that
includes only a saturated clay liner.
Due to saturated conditions, Equation (8.1) is transformed into a steady
state equation, and the steady state Darcian velocity (cm/s) or flux
(cm3/cmzs) is given by Equation (8.3). At the top of the liner (z = 0),
the liquid pressure head is equal to the impoundment liquid head (* » h); at
the liner bottom, z is equal to liner thickness (z = d). Because the liner
1s saturated, the pressure head is taken as zero (* = 0) at the liner bottom,
assuming free drainage. Thus, the steady state Darcian velocity and flux,
assuming linear potential gradient, are given by:
v = Ks (§ + 1) (8.4)
8-6
-------
Because the liner is assumed to be saturated, the seepage flux is
established a_s soon as the impoundment liquid head is established. The time
taken by the leachate chemicals;at!"the top df the liner to arrive at the
liner bottom under steady state saturated conditions, due to advection only,
may be obtained by solving Equation (8.2). Assuming D = 0 and R = 1,
Equation (8.2) gives the required transit time, t, as:
The liner thickness that is required to achieve a given transit time
may then be obtained based on Equations (8.4) and (8.5):
d = 0.5
(8.6)
Because only the advective transport of solute is considered when the
solute transport equation is solved, the leachate concentration at the
bottom of the liner after the transit time would be the same as that of
waste leachate on top of the liner. Before this transit time, the leachate
chemical concentration in the seepage at the liner bottom would be zero.
The leachate chemical flux at the liner bottom after the transit time can
be obtained simply by multiplying the seepage flux (cm3/cm2s) by the leachate
concentration (ppm).
The key assumptions in this approach are:
• Steady state saturated Darcian flow
t Pore fluid pressure at the bottom of the liner is equal to atmos-
pheric pressure; i.e., the bottom layer of the liner remains
saturated indefinitely
• Solute transport is by advection only; i.e., there is no diffusion
or dispersion
• Attenuation capacity of the clay is ignored. *
A limiting assumption in this approach is that molecular diffusion or
dispersion is not significant compared to convective flux. In view of the
low permeability of clays and the corresponding low fluid velocity, the
molecular diffusion process may be a significant factor. Diffusion may
allow the dissolved chemicals to migrate faster in a liner compared to
advective flow and will thus reduce the transit time of the chemical
species.
The simple transit time equation is very easy to use, however. One
needs to know only the saturated hydraulic conductivity of the liner
material, its total (or effective) porosity, and the leachate head at the
top of the liner. Because these parameters should be readily available,
8-7
-------
this equation provides a method to obtain quick estimates of liner perfor-
mance or required liner thickness.
8.3.2 Modified Transit Time Equation
Cogley et al. (U.S. EPA, 1984, Appendix A) modified the simple transit
time equation to account for effective porosity (0e) and the suction potential
at the liner bottom. However, other assumptions regarding steady state condi-
tions, saturated Darcian flow in a homogeneous liner, and advective solute
transport were made as before.
Due to the steady state saturated conditions, Darcian velocity and flux
are again given by Equation (8.3). At the top of the liner (z = 0), the pres-
sure head is given by impoundment height (* = h), and at the bottom of the
liner (z » d) the pressure head is negative due to suction (* = - hj). When
these boundary conditions are incorporated and a linear potential gradient
assumed, the Darcian velocity and flux are given by:
h+h .
v = Ks (-g-S + D (8.7)
Equation (8.2), the solute transport equation, may then be solved as
before—assuming saturated steady state conditions and ignoring dispersion
and attenuation processes (D = 0 and R = 1). The resulting transit time for
a solute species for a given liner thickness is given by:
(h+hd+d)
(8.8)
The required Uner thickness for a specified leachate transit time is
then obtained as:
0.5
(8.9)
Since only advective transport is considered for-solv1ng the solute
transport equation, the leachate concentration at the bottom of the liner
(after transit time) would be the same as the concentration of waste leach-
ate on top of the liner. Before this transit time, the leachate concentra-
tion 1n the seepage at the Uner bottom would be zero. The leachate chemi-
cal flux at the liner bottom after the transit time is given simply by
multiplying the seepage flux by the leachate concentration.
Like the simple transit time equation, the modified equation assumes
saturated steady state conditions and negligible diffusion processes 1n
solute transport. The incorporation of suction potential at the liner
bottom 1s actually somewhat inconsistent with the assumption of saturated
flow since the negative pressure would induce unsaturated conditions in
lower sections of the liner. Although the suction potential at the bottom
is included, the higher hydraulic conductivity value corresponding to
8-8
-------
saturated conditions is used in this equation. As a result, for a speci-
fied transit time, this modified equation yields greater values for liner
thickness compared with the simple transit time equation.
8.3.3 Green-Ampt Wetting Front Model
The transit time equations discussed above ignore both the initial,
unsaturated nature of the liner and infiltration dynamics. Green and Ampt
(1911) derived a simple model to describe the infiltration process that has
been proposed as a method to assess liner reliability (U.S. EPA, 1984,
Appendix A). A similar approach was also analyzed by McWhorter and Nelson
(1979) and discussed by EPA (U.S. EPA, 1983).
The Green-Ampt model" (Green and Ampt, 1911) describes moisture movement
in unsaturated soil during ponded infiltration by assuming a sharp,
saturated wetting front moving down the soil column as a square wave.
Above the wetting front, the soil is fully saturated, with a moisture
content, 0S, while below the wetting front the moisture content is equal to
its initial value, 8,. At a given time, t, following establishment of the
ponded leachate head on top of the liner, the wetting front will have moved
down a certain distance, L. Saturated flow analysis may be applied to the
saturated zone above the wetting front to determine the Darcian velocity or
flux as given by Equation (8.3). Th |