p, EPA
f 530/SW
87-015
Part I
•-- -':-;
c-
Background Document on
Proposed Liner and Leak
Detection Rule
NUS Corp., Rockville, MD
Prepared for
Environmental Protection Agency
Washington, DC
PB87-191383
May 87
U, S,
AGENCY
-
eO
i
CT
••ma
ls':$.
i d
-------
• 101
REPORT DOCUMENTATION »•
PAGE
joy
/V /
530- SlO-
icn No .^
4. T.ti« tnd SuMItt*
Background Document on Proposed Liner and Leak Detection Rule j.
7. Auth.ru> Dr_ R_ Bonaparte( Dri j.p.ciroud, Messrs. R.B. Wallace, ;r
.T.Pri_1
t. Ptrtorminf Urgimxiition N»mt and Addrou
Geo. Services Inc.Consulting Engineers
1200 South Federal Highway, Suite 204
Boynton Beach, Florida 33435
10. P'OKCt/Ttm/vvork Unit No
11. Conir«et(C) or Gnnt(G) No
to No.68-01-7310
I (C)
It. Sponiorinf O'(lnitition Nim« «nd Addr«ti
U.S. Environmental Protection Agency
Office of Solid Waste (WH-565E)
401 M Street, S.W.
MashJngi-nn,. P.P.
11. T»p« o' R.oort & (••nod Cdv«r»d
Background Document
14.
IS. Supp«m«ntiry Notn
Prepared under Contract No.68-01-7310, Work Assignment No.1 (Amendment 3)
to NUS Corporation
i«. Aeitrict (Limit: MO w»rd»i ihe purpose of this document is to provide the technical rationale and
support for the 3 main portions of the proposed Liner Leak Detection Rule: (1) leak detec-
tion system requirements; (2) extension of the double-liner system requirements to waste
piles, significant unused portions and certain other units; and (3) construction quality
assurance program requirements.
Itie portion of this document devoted to leak detection (Chapter 2) will: (1) present
the technologies and materials available to construct lining systems and leak detection
systems to meet the proposed regulations; (2) recomiend best demonstrated available tech-
nologies (BOAT) to meet the requirements stenming frcn these regulations; and (3) quantify
leak detection system performance capabilities associated with BOAT.
This document, to the greatest extent possible, presents a state-of-the-art review of
available technologies and achievable performance levels in leak detection systems at haz-
ardous waste management units. This review is necessary to establish the best demonstra-
ted available technology (BDAT) for leak detection systems. It provides the critical
information considered in developing EPA's regulatory options and describes the criteria
used to select among the options. The document also presents a rationale and technical
data to support extension of the double-liner system requirements to waste piles, signifi-
cant unused portions, and certain other units. Lastly, the document presents a comprehen-
sive discussion of the issues, methodologies, and benefits associated with construction
ality Qoouronoo programs at hagardouo wacto manQgomont-facilitiooi
M<-*«.-tJ-*-T T^.ZJ*J\JH-\
Jr. Oocunvfht AntlytJ*
b. Id»ntlf<»ri/Op«n-Cn4«d Tormt
e. COSATI n«li)/an>up
IL Avtilibllity S»t*m*nt
19. Security Cl«99 (This Report)
unc 1nss if a ed
Release Unlimited
; 20. Security Clfltl
'
f icd
21. No ol P»f«
22. Frtc.
$«• tnilrvctrent an Ktrt.'tt
OTWrKAI. FORM J72 (4-77)
(Fomwriy NTIi-JS)
-------
Pbd7-19
EPA/530-SW-87--015
BACKGROUND DOCUMENT
PROPOSED LINER AND LEAK DETECTION RULE
Prepared for
U.S. Environmental Protection Agency
Office of Solid Waste
Washington, O.C. 20460
Under
NUS Corporation
Contract No. 68-01-7310
Work Assignment Ho. 7
(Amendment 3)
by
GeoServlces Inc. Consulting Engineers
1200 South Federal Highway, Suite 204
Boynton Beach, Florida 33435
May 1987
REPRODUCEDBY
U.S. DEPARTMENTOF COMMERCE
NATIONAL TECHNCAL
-------
DISCLAIMER
This Report entitled "Background Document on Proposed Liner and
Leak Detection Rule" was furnished to the U.S. Environmental
Protection Agency by GeoServlces Inc. Consulting Engineers, 1200 S.
Federal Highway, Boynton &each, Florida 33435, under subcontract to
NUS Corporation and 1n fulfillment of Contract No. 68-01-7301, Work
Assignment No. 7, Amendment 3. In addition to Dr. R. Bonaparte,
Project Manager, the primary authors of this document include Dr. J.P.
Giroud, Messrs. R.B. Wallace and C. Ah-Line, and Ms. J. Prillaman.
The opinions, findings, and conclusions expressed are those of the
authors and not necessarily those of the Environmental Protection
Agency or cooperating agencies. Mention of company or product names
Is not to be considered an endorsement by the Environmental Protection
Agency.
ACKNOWLEDGEMENT
This report was prepared under the guidance of the USEPA, Office
of Solid Waste, Land Disposal Branch. USEPA Task Managers for this
assignment were Messrs. Alessi D. Otte and Walter DeRieux, P.E. Their
support and contributions to this work are appreciated.
-------
TABLE OF CONTENTS
Chapter 1
INTRODUCTION
1.1 LEGISLATIVE HISTORY
1.1.1 Solid Waste Disposal Act/Res'-'jrce Conservation -nd Recovery Act
1.1.2 Hazardous and Solid Waste Amendments of 1984
1.1.Z.I Hazardous and Solid Waste Amendment Policies and Findings
1.1.2.2 Major Changes in Hazardous Waste Management
1.2 PURPOSE AND SCOPE OF LINER/LEAK DETECTION RULE
1.2.1 Background
1.2.2 Leak Detection Systems
1.2.3 Extension of Minimum Technology Standards
1.2.3.1 Waste Piles
1.2.3.2 Significant Portions
1.2.4 Construction Quality Assurance Program
1.2.5 Applicable Units
1.3 PURPOSE AND SCOPE OF THE BACKGROUND DOCUMENT
1.3.1 Purpose of the Background Document
1.3.2 Scope of the Background Document
1.3.2.1 Technical Information Areas
1.3.2.1.1 Leak Detection Systems
1.3.2.1.2 Double Liner System
for New Waste Pile Units
1.3.2.1.3 Significant Portions
1.3.2.1.4 Double Liners for New Units, Replace-
ments and Lateral Expansions of Facilities
Permittee rior to November 8, 1984
1.3.2.1.5 Construction Quality Assurance Program
1.3.2.1.6 Land Treatment Units
-------
*:
Chapter 2
LEAKAGE DETECTION
2.1 INTRODUCTION
2.1.1 Scope of Chapter 2
2.1.2 Waste Management Units
2.1.2.1 Introduction
2.1.2.1.1 Definition
2.1.2.1.2 Purpose of this Section
2.1.2.2 Description
2.1.2.2.1 Types of Land Disposal Units
2.1.2.2.2 Geometry of Land Disposal Units
2.1.2.3 Ground Pollution Mechanism
2.1.2.3.1 Surface Impoundments
2.1.2.3.2 Landfills
2.1.2.3.3 Waste Piles
2.1.3 Lining Systems Used in Land Disposal Units
2.1.3.1 Introduction
2.1.3.1.1 Importance of Lining Systems
2.1.3.1.2 Scope of this Section
2.1.3.1.3 Definition of Lining Systems
2.1.3.2 Materials Used in Lining Systems
2.1.3.2.1 Introduction
2.1.3.2.2 Liner Materials
2.1.3.2.3 Drainage Materials
2.1.3.2.4 Transition Materials
2.1.3.2.5 Reinforcement Materials
2.1.3.3 Double Liners
2.1.3.3.1 Introduction
2.1.3.4 Use of Double Liners in Land Disposal Units
2.1.3.4.1 Current Regulations
2.1.3.4.2 Examples of Uses of Double Liners
in Land Disposal Units
2.1.3.4.3 Influence of Top and Bottom Liners on Leak
Detection
-------
2.1.4 Leakage Definition and Detection
2.1.4.1 Definitions
2.1.4.1.1 Leak and Leakage
2.1.4.1.2 Leak Size and Leakage Pate
2.1.4.1.3 Leakage Collected and Leakage
Out of the Unit
2.1.4.2 Leak Detection System
2.1.4.2.1 Definition
2.1.4.2.2 Purpose of Leak Detection
2.1.4.2.3 Performance Requirements of Leak
Detection Systems
2.2 TOP LINER PEurj.lMANCE
2.2.1 Introduction
2.2.1.1 Scope
2.2.1.2 Organization
2.2.2 Top Liners
2.2.2.1 Types of Top Liners
2.2.2.1.1 FML
2.2.2.1.2 Low-Permeability Compacted Soil
2.2.2.1.3 Composite Liner
2.2.2.2 Types of Materials Used for Top Liners
2.2.2.2.1 FMLs
2.2.2.2.2 Low-Permeability Compacted Soils
2.2.2.3 Permeability of Liner Materials
2.2.2.3.1 Introduction
2.2.2.3.2 Permeability of Compacted Soils
2.2.2.3.3 Permeation through FMLs
2.2.2.4 Typical Defects of Liner Materials
2.2.2,4.1 Introduction
2.2.2.4.2 FML Defects
2.2.2.4.3 Low-Permeability Compacted Soils
2.2.2.4.4 Composite Liner Defects
-------
I."' . * ' -• ..'
2.2.3 Leakage throuy,', ,"ML lop Liner
2.2.3.1 Introduction
2.2.3.1.1 Scope of the Section
2.2.3.1.2 Organization of the Section
2.2.3.2 Evaluation of Leakage through FML Top Liners
2.2.3.2.1 Introduction
2.2.3.2.2 Leakage due to Permeation through FML
2.2.3.2.3 Leakage due to Pinholes in the FML
2.2.3.2.4 Leakage due to Holes in the FML
2.2.3.3 Frequency and Size of FML Defects
2.2.3.3.1 Purpose
2.2.3.3.2 Data from Construction Quality Assurance
2.2.3.3.3 Data from Forensic Analyses
2.2.3.3.4 Conclusions on Frequency of Defects
2.2.3.3.5 Estimation of Size of Defects
2.2.3.3.6 Standard Hole Size and Frequency
2.2.3.4 Conclusions on Leakage through FML Top Liners
2.2.3.4.1 Summary
2.2.3.4.2 Leakage Rates
2.2.4 Leakage Through Composite Top Liners
2.2.4.1 Introduction
2.2.4.1.1 Purpose of the Section
2.2.4.1.2 Leakage Mechanisms
2.2.4.1.3 Organization of the Section
2.2.4.2 Analytical Studies
2.2.4.2.1 Introduction
2.2.4.2.2 Analyses Assuming Perfect Contact
2.2.4.2.3 Analyses Assuming Flow between FML and Soil
Z.2.4.3 Laboratory Models
2.2.4.3.1 Introduction
2.2.4.3.2 Review of T->sts by Brown et al.
2.2.4.3.3 Review of Tests by Fukuoka
2.2.4.4 Conclusions on Leakage through Composite Top Liners
2.2.4.4.1 Conclusions from Analytical Studies
-------
2.2.4.'1.2 Conclusion from Model 1331.5
2.2.4.4.3 Conclusions for Leakage Rate Evaluation
2.2.5 Conclusions on Leakage through Top Liners
2.2.5.1 Defects ana Qua ty Assurance
2.2.5.2 Summary of Leakage Rate Values
2.2.5.3 Consents on Leakage Rate Values
2.3 LEAK DETECTION TECHNOLOGIES
2.3.1 Review of Available Technologies
2.3.1.1 Introduction
2.3.1.2 Leachate Collection and Removal Systems
2.3.1.2.1 Principles
2.3.1.2.2 Evaluation
2.3.1.3 Electrical Resistivity
2.3.1.3.1 Principles
2.3.1.3.2 Recent Studies
2.3.1.3.3 Evaluation
2.3.1.4 Time Dorrain Reflectometry
2.3.1.4.1 Principles
2.3.1.4.2 Recent Studies
2.3.I.4."1 Evaluation
2.3.1.5 Acoustic Emissions Monitoring
2.3.1.5.1 Principles
2.3.1.5.2 Recent Studies
2.3.1.5.3 Evaluation
2.3.1.6 Other Leak Detection Technologies
2.3.1.6.1 Lysimeters
2.3.1.6.2 Seismic Measurement
2.3.1.6.3 Electromagnetic Techniques
2.3.1.6.4 Moisture Blocks
2.3.2 Selection of Leachate Collection and Removal
System as Leak Detection System
2.3.2.1 Drainage Layer Technology
2.3.2.2 Innovative Technologies
-------
2.4 LEAK DETECTION SYSTEMS BETWEEN LINERS
2.4.1 Functions of a Leak Detection System
2.4.2 Materials
2.4.2.1 Introduction
2.4.2.2 Drainage Layers
2.4.2.2.1 Granular Drainage Materials
2.4.2.2.2 Synthetic Drainage Materials
2.4.2.3 Filter Layers
2.4.2.4 Cushions Layers
2.4.2.5 Pipes
2.4.2.6 Structures
2.4.2.6.1 Manholes
2.4.2.6.2 Sunps
2.4.2.6.3 Auxiliary Cleanouts
2.4.3 Properties of Materials
2.4.3.1 Introduction
2.4.3.2 Hydraulic Conductivity
2.4.3.2.1 Granular Drainage Materials
2.4.3.2.2 Synthetic Drainage Mater Is
2.4.3.3 Hydraulic Transmissivity
2.4.3.3.1 Granular Drainage Materials
2.4.3.3.2 Synthetic Drainage Materials
2.4.3.4 Filter Characteristics
2.4.3.4.1 Mechanisms of Filtration
2.4.3.' 2 Granular Materials
2.4.3.^.3 Geotextile Filters
2.4.3.5 Durability
2.4.3.5.1 Abrasioi and Fatigue
2.4.3.5.2 Physico-Chemical-Biologica] Degradation
2.4.3.6 Mechanical Effects of Drainage Materials on FML Liners
2.4.3.6.1 Granular Drainage Materials
2.4.3.6.2 Synthetic Drainage Materials
2.4.4 Conclusions
-------
' t ' ' v t - ; _,^ T^' f ' ' -' J • ^ ^ ^_^ -.umainft "*
2.5 LIQUIDS MEASURED IN LDCRS AT OPERATING UNITS
2.5.1 Introduction
2.5.2 Institute of Chemical Waste I'^nagement Data
2.5.3 Case Study - Landfill in South East U.S.
2.5.-I Case Study - Two Landfills in North Contra! U.S
2.5.5 Case Studies - Surface Impoundments in East Central and
South West U.S.
2.5.5.1 Surface Impoundments in East Centra! U.S.
2.5.5.2 Surface Impoundments in South West U.S.
2.6 ANALYSES OF THE FUNCliONING OF LEAK DETECTION SYSTEMS
2.6.1 Introduction
2.6.1.1 Purpose
2.6.1.2 Overview of Leak Detection System Functioning
2.6.1.3 Definitions
2.6.1.3.1 Leakage
2.6.1.3.2 Time of Initial Leakage
2.6.1.3.3 Initial Detection Time
2.6.1.3.4 Leak Detection Time
2.6.1.3.5 Detection Sensitivity
2.6.1.3.6 Action Lea'^age Rate (ALR)
2.6.1.3.7 Rapid and Extremely Large Leaks
2.6.1.4 Organization of this Section
2.6.2 Two-Dimensional Analytical Study
2.6.2.1 Introduction
2.6.2.1.1 Purpose of this Section
2.6.2.1.2 Approach
2.6.2.1.3 Organization of this Section
2.6.2.2 Assumptions
2.6.2.2.1 Assumptions Related to the Leak
Detection System
2.6.2.2.2 Assumptions Related to the Flow
2.6.2.3 Steady-State Flow
2.6.2.3.1 Introduction
2.6.2.3.2 Steady-State Leak Detection Time
2.6.2.3.3 Leak Detection System Capacity
-------
P^w^?.^T!r-^S?|^^f?t»p^
2.6.2.4 Ini Nation of How
2.6.2.4.1 Introduction
2.6.2.4,2 Retention by Capillarity
2.6.2.4.2 Boundaries of the Flow Mechanism
2.6.2.5 Conclusions
2.6.2.5.1 Discussion of the Results
2.6.2.5.2 Extension of the Study
2.6.3 Two-Dimensional Numerical Study
2.6.3.1 Introduction
2.6.3.1.1 Purpose o; this Section
2.6.3.1.2 Approach
2.6.3.1.3 Complementarity of Analytical and
Numerical Studies
2.6.3.1.4 Organization of this Section
2."3.3.2 Method
2.6.3.2.1 Description of the Finite Element Program
2.6.3.2.2 Assumptions
2.6.3.3 Results of the Numerical Study
2.6.3.3.1 Summary of the Results
2.6.3.3.2 Comparison between the Analytical and
Numerical Study
2.7 PERFORMANCE CRITERIA FOR LEAK DETECTION SYSTEMS
2.7.1 Introduction
2.7.1.1 Scope of the Section
2.7.1.1.1 Purpose of the Section
2.7.1.1.2 Organization of the Section
2.7.1.2 Design Requirements
2.7.1.2.1 The Concept of Performance Criteria
2.7.1.2.2 The Concept of Design Specifications
2.7.1.2.3 Performance Criteria and Design
Speci fications
2.7.1.3 Performance Characteristics
2.7.1.3.1 Scope of the Section
0
-------
2.7.1.3.2 Leak Detection System Capaoilities
2.7.1.3.3 Selection of Performance Characteristics
2.7.2 Detection Sensitivity
2.7.2.1 Introduction
2.7.2.1.1 Scope of the Section
2.7.2.1.2 Definition and Importance of
Detection Sensitivity
2.7.2.2 Establishment of the Detection Sensitivity Criterion
2.7.2.2.1 Sunroary of Relevant Information
2.7.2.2.2 Ratior = for the Criterion
2.7.2.3 Presentation of t-he Detection Sensitivity Criterion
2.7.2.3.1 Expression of the Criterion
2.7.2.3.2 Discussion
2.7.3 Detection Time
2.7.3.1 Introduction
2.7.3.1.1 Scope of the Section
2.7.3.1.2 Definition and Importance of Detection Time
2.7.3.2 Establishment of the Detection Time Criterion
2.7.3.2.1 Summary of Relevant Information
2.7.3.2.2 Rationale for the Criterion
2.7.3.3 Presentation of the Detection Time Criterion
2.7.3.3.1 Expression of the Criterion
2.7.3.3.2 Discussion
2.7.4 Leachate Collection Efficiency
2.7.4.1 Introduction
2.7.4.1.1 Scope of the Section
2.7.4.1.? Definition of Leachate Collection Efficiency
2.7.4.2 Discussion of the Concept of Leachate
Collection Efficiency
2.7.4.2.1 Analysis of Leakage Types
2.7.4.2.2 Evaluation of Leachate Collection Efficiency
-------
2.8 DESIGN SPECIFICATIONS FOR LEAK DETECTION SYSTEMS
2.8.1 Introduction
2.8.1.1 Scope of the Section
2.8.1.1.1 Purpose of the Section
2.8.1.1.2 Organization of the Section
2.8.1.2 The Concept of Design Specifications
2.8.1.2.1 Definition
2.8.1.2.2 Demonstration
2.8.1.2.3 Usefulness of Design Specifications
2.8.1.2.4 Conservativeness of Design Specifications
2.8.2 Selection of Design Parameters
2.8.2.1 Introduction
2.8.2.1.1 Purpose
2.8.2.1.2 Method
2.8.2.2 Review of Design Parameters
2.8.2.2.1 Review of Parameters Governing Detection
Sensitivity
2.8.2.2.2 Review of Parameters Governing Detection Time
2.8.2.2.3 Review of Parameters Governing Liquid Head
2.8.2.3 Design Parameters to be Considered in Specifications
2.8.3 Establishment of Design Specifications
2.8.3.1 Summary of Relevant Data
2.8.3.1.1 Performance Criteria to Meet
2.8.3.1.2 Relevant Technical Information for
Design Specifications
2.8.3.2 Rationale for the Sp.cifications
2.8.4 Presentation and Discussion of the Design Specifications
2.8.4.1 Presentation of the Design Specifications
2.8.4.2 Comments
2.9 ACTION LEAKAGE RATE (ALR)
2.9.1 Introduction
2.9.1.1 Purpose ~ the Section
10
-------
2.9.1.2 Organization of the Section
2.9.2 Overview of the Concepts of the Proposed Rule
2.9.2.1 The Concept of a Trigger
2.9.2.2 The Concept of Action Leakage Rate
2.9.3 Technical Support for the ALR
2.9.3.1 Introduction
2.9.3.2 Relevant Technical Information
2.9.3.3 Rationale for the Action Leakage Rate Value
2.9.3.4 Monitoring Requirements
2.10 RESPONSE ACTION PLAN (RAP)
2.10.1 Introduction
2.10.1.1 Scope
2.10.1.1.1 The Response Action Plan (RAP)
2.10.1.1.2 Technical Support
2.10.1.2 Organization of the Section
2.10.2 Technical Elements of the Response Action Plan
2.10.2.1 General Description of Unit
2.10.2.2 Hazardous Constituent Assessment
2.10.2.3 Description of Events Causing Leakage
2.10.".4 Factors Influencing Liquid Quantities in the LDCRS
2.10.U.5 Me hanisms Preventing Migration Out of the Unit
2.10.2.6 Assessment of Response Actions
2.10.2.7 Sources of Information for the RAP
2.10.3 Leakage Bands
2.10.3.1 Introduction
2.10.3.2 Rapid and Extremely Large Leak
2.10.3.2.1 Discussion
2.10.3.2.2 Technical Support
2.10.4 Sources of Liquids other than Leakage
2.10.4.1 Introduction
2.10.4.1.1 Scope
2.10.4.1.2 Organization of the Section
2.10.4.2 Rainwater Entrapped in the Leak Detection System
2.10.4.3 Water Expelled by Consolidation from Top Liner
11
-------
2.10.4.3.1 Introduction
2.10.4.3.2 Analysis
2.10.4.3.3 Review of Results
2.10.4.4 Leakage Into a Land Disposal Unit Due to Giound Water
2.10.5 Conclusions
2.10,5.1 Technical Elements of the RAP
2.10.5.2 Rapid and Extremely Large Leak
2.10.5.3 Source of Liquids other than Leachate
Chapter 3
EXTENSION OF DOUBLE LINER SYSTEM REQUIREMENTS
3.1 INTRODUCTION
3.1.1 Scope of Chapter 3
3.1.1.1 Double Liners and LCRS for Waste Piles
3.1.1.2 Double Liners and LCRS for Significant Unused
"ortions of Existing landfills. Surface
iipoundments, and Waste Piles
3.1.1.3 Double Liners and LCRS for Certain Land Disposal
Units at Facilities Permitted Before November 8, 1984
3.1.2 Organization of Chapter 3
3.2 WASTE PILES
3.2.1 Description of Waste Piles
3.2.2 Background
3.2.3 Rationale for Double Liner System Requirements
3.2.3.1 Operating Characteristics
3.2.3.2 Analytical Calculations
3.2.3.3 Numerical Simulations
3.2.4 Exemption for Totally Enclosed Waste Piles
3.2.5 Variances
3.3 SIGNIFICANT PORTIONS
3.3.1 Definition of Significant Portions
3.3.2 Background
3.3.3 Rationale for Double Liner System Requirements
12
-------
3.3.4 Proposej Exemplion from Leak Detection Rcqu ireiiiei.Ls
3.3.5 Examples of Significant Portions
3.3.6 Variances
3.4 NEW UNITS, REPLACEMENTS AND LATERAL EXPANSIONS AT
FACILITIES PERMITTED PRIOR TO NOVEMBER 8, 1964
3.4.1 Background
3.4.2 Rationale for Double Liner System Requirements
3.4.3 Exemptions for Certain Replacement Units
3.4.4 Variances
Chapter 4
CONSTRUCTION QUALITY ASSURANCE
4.1 INTRODUCTION
4.1.1 Scope of Chapter 4
4.1.2 Rationale for the CQA Program
4.1.3 Definitions Related to CQA
4.1.4 Parties to CQA - Roles and Responsibilities
4.2 CONSTRUCTION QUALITY SSURANCE PLAN
4.2.1 General Description of the Unit
4.2.2 Responsibi1ity and Authority
4.2.2.1 Organizations Involved in CQA
4.2.2.2 Project Meetings
4.2.3 Personnel Qualifications
4.2.3.1 CQA Officer
4.2.3.2 CQA Monitoring Personnel
4.2.3.3 Consultants
4.2.4 CQA Monitoring and Sampling Activities
4.2.5 Documentation of Construction Quality Assurance Activities
4.2.5.1 Daily Record Kjeping
4.2.5.2 Photographic Reporting Data Sheets
4.2.5.3 Block Evaluation Reports
4.2.5.4 Acceptance of Completed Components
4.2.5.5 Final Documentation
4.2.5.6 Document Control
4.2.5.7 Storage of Records
13
-------
4.3 NEED FOR A CQA PROGRAM
4.3.1 Background
4.3.2 Effect of Construction Procedures on Lining System Performance
4.3.2.1 Soils-Related Construction Prcolems
4.3.2.1.1 Moisture Control and Compaction
4.3.2.1.2 Weather and Cl imate
4.3.2.1.3 Availability of Suitable Soils
4.3.2.1.4 Subgrade Soils Problems
4.3.2.1.5 Soils Homogeneity and Layering
4.3.2.1.6 Disturbance Due to Traffic
4.2.3.1.7 Soil Components of Top Composite Liners
4.3.2.1.8 Testing Problems - Field Compaction of Soils
4.3.2.1.9 Testing Problems - Laboratory Moisture-Den-
sity Tests
4.3.2.1.10 Testing Problems - Laboratory Permeability
Determination
4.3.2.1.11 Testing Problems - Field Permeability
Determination
4.3.2.2 Geosynthetic-Related Construction Problems
'.3.2.2.1 Manufacturing Quality Control
4.3.2.2.2 Fabrication Quality Control
4.3.2.2.3 Shipping and Handling
4.3.2.2.4 Sheet Material Defects
4.3.2.2.5 Seaming Procedures
4.3.2.2.6 Seaming Constraints
4.3.2.2.7 Contamination
4.3.2.3 Qualifications of Personnel
4.3.2.3.1 CQA Officer
4.3.2.3.2 CQA Manager
4.3.2.3.3 CQA Monitors
4.3.2.4 Documentation of Problems
4.3.3 Materials Considerations
4.3.3.1 Soils Materials Considerations
4.3.3.2 Geosynthetics Considerations
4.3.3.2.1 Manufacturing Considerations
4.3.3.2.2 Fabrication Considerations
4.3.3.2.3 Transportation and Handling Considerations
4.3.4 Benefits of Construction Quality Assurance
14
-------
4.3.4.1 Nature and Description of the Benefits
4.3.4.1.1 Fewer Leaks
4.3.4.1.2 High Confidence in the Integrity of the Liit
4.3.4.1.3 High Confidence in the Public's Eyes
4.3.4.1.4 Benefits After Construction
4.3.4.1.5 Benefits to the Owner/Operator
4.3.4.1.6 Benefits to the Other Parties
4.4 SCOPE OF CONSTRUCTION QUALITY ASSURANCE
4.4.1 Pre-Construction Stage
4.4.1.1 Design
4.4.1.2 Materials Specifications
4.4.1.3 Materials Procurement
~~-itr
4.4.2 Construction Stage
4.4.2.1 Site Preparation and Foundations
4.4.2.2 Dikes
4.4.2.3 Compacted Soil Liners
4.4.2.4 FMLs and Other Geosynthetics
4.4.2.4.1 Delivery and Conformance Checking
4.4.2.4.2 Deployment and Visual Examination
4.4.2.4.3 Seaming and Joining
4.4.2.4.4 Nondestructive Testing of Sea^s
4.4.2.4.5 Destructive Testing
4.4.2.4.6 Other Considerations
4.4.2.5 Leachate Collection and Removal Systems
4.4.2.5.1 Leachate Collection Pipes
4.4.2.5.2 Obstructions to Leachate Flow
4.4.2.6 Final Cover Systems
4.4.2.6.1 Subsidence
4.4.2.6.2 Installation Procedures
4.4.2.6.3 Vegetative Layers
4.4.3 Post-Construction Stage
4.4.3.1 Reporting
4.4.3.2 Monitoring
4.4.3.3 Coupon Testing
4.5 TESTING PROCEDURES
15
-------
4.5.1 Soils
4.5.1.1 Procedures
4.5.1.1.1 Laboratory Testing - Soil Compaction
4.5.1.1.2 Field Testing - Soils Compaction
4.5.1.1.3 Laboratory Testing - Soil Permeability
4.5.1.1.4 Field Testing - Soil Permeability
4.5.1.2 Effectiveness of the Tests - Acceptance Criteria
4.5.1.2.1 Compaction-Related Tests
4.5.1.2.2 Hydraulic Conductivity Tests
4.5.1.3 Current State of Practice
4.5.1.4 Test Fills
4.5.2 Flexible Membrane Liners
4.5.2.1 Procedures
4.5.2.1.1 Laboratory Testing - Specifications
Conformance
4.5.2.1.2 Laboratory Testing - Destructive FML Seam
Quality Control
4.5.2.1.3 Field Testing - nondestructive FML Seam Tests
4.5.2.2 Effectiveness of the Tests - Acceptance Criteria
4.5.2.2.1 Conformance Tests
4.5.2.2.2 Seam Quality Control Tests
4.5.2.2.3 Nondestructive Seam Tests
4.5.2.3 Current State of Practice
4.5.3 Other Geosynthetic Materials
4.5.3.1 Procedures
4.5.3.1.1 Laboratory Testing - Specifications
Conformance
4.5.3.2 Effectiveness of the Tests - Acceptance Criteria
4.5.3.2.1 Geotextile Conformance Tests
4.5.3.2.2 Geonet Conformance Tests
4.5.3.2.3 Geogrid Conformance Tests
4.5.3.3 Current State of Practice
16
-------
CHAPTER 5
LAND TREATMENT UNITS
5.1 INTRODUCTION
5.2 .EGULATORY APPROACH TO LAND TREATMENT
5.3 PERMITTED LAND TREATMENT UNITS
5.3.1 Current Regulations
5.3.2 Approach to Statutory Requirements
5.3.3 Detection Confidence Level
5.3.4 Monitoring Periods
5.3.5 Inspection
5.3.6 Response Action Plan
5.4 INTERIM STATUS LAND TREATMENT UNITS
5.4.1 Current Regulations
5.4.2 Proposed Interim Status Monitoring Requirements
5.4.3 Interim Status Implementation Differences
5.4.4 Monitoring Plan Amendments
17
-------
CHAPTER 1
INTRODUCTION
-------
LEGISLATIVE HISTORY
Solid Waste Disposal Act/Resource Conservation and Recovery
Act
The Solid Waste Disposal Act was substantively amended by the
Resource Conservation and Recovery Act (RCRA) of 1976 which, with
protection of human health and the environment as Its objective,
Instituted a "cradle to the grave" management system to ensure the
safe treatment, storage and disposal of hazardous waste.
Statutory changes made by the Resource Conservation and Recovery
Act were so comprehensive that references to the statute dealing with
hazardous waste management are made to the Resource Conservation and
Recovery Act, although technically it is codified as the Solid Waste
Disposal Act. This document will follow the tradition of citing the
Resource Conservation and Recovery Act.
1.1.2 Hazardous and Sol Id Waste Amendments of 1984
1.1.2.1 Hazardous and Sol 1d_Waste_A/nendment PoHcles and Findings
The Hazardous and Solid Waste Amendments of 1984 (HSWA) made many
changes in Resource Conservation and Recovery Act sections covering
regulation of hazardous waste.
Congressional findings relating to the threat to human health and
the environment were the rationale for mandating hazardous waste
management regulatory changes. Two of these findings acknowledged
that "placement of inadequate controls on hazardous waste management
will result in substantial risks to human health and the environment"
(RCRA 1002(b)(5) as amended) and that "if hazardous waste management
1s improperly performed in the first instance, corrective action is
likely to be expensive, complex, and time consuming" (RCRA
1002(b)(6)).
1-1
-------
The HSWA also established a national policy that (hazardous) waste
"should be treifed, stored or disposed so as to minimize the present
and future thr r. to human health and the environment" fRCRA I003(b)).
The minimum technology requirements of the H5WA require the
Environmental Protection Agency (EPA) to revise regulations for liners
and leak detection systems at hazardous waste management units. The
EPA is preparing to address through regulation three objectives
mandated by the HSWA:
• To revise existing minimum technology requirements for
landfills, surface impoundments, and waste piles;
• To establish minimum standards for leak detection and response;
and,
• To augment existing rules with construction quality assurance
(CQA) requirements for owners or operators of hazardous waste
management facilities to help assure that land disposal units
perform as designed.
1.1.2.2 Major_Changes 1n Hazardous Waste Management
The HSWA amendments regarding hazardous waste are comprehensive
and are requiring changes in the way that the United States manages
Its hazardous wastes. They are also significant because of the
ambitious schedules Congress established to meet the requirements. For
Instance, minimum technology requirements relating to leak detection
systems at landfills, surface impoundments, and waste piles, and land
treatment units are required to be imple: ted within 30 months of the
HSWA enactment (Nov. 8, 1984).
Some major changes reouired by the HSWA in hazardous waste
management Include:
• Minimum technology requirements for landfills, surface
Impoundments, and waste piles.
1-2
-------
• Requirements for retrofitting certain existing sui face
impoundments with liners.
• Authority for EPA to expedite permits for new and innovative
treatment technologies to foster research and development.
• A new program for identifying the health risks presented by
surface Impoundments and landfill facilities.
• Expanded requirements for ground-water monitoring and cleanup
at permitted facilities, including land treatment units.
• A new program for banning wastes from land disposal.
1.2 PURPOSE AND SCOPE OF LINER/LEAK DETECTION RULE
The purpose of the set of regulations being proposed by EPA
through its "liner/leak detection rule" is to continue implementation
of the Congressional directives regarding hazardous waste management
set out in the Resource Conservation and Recovery Act amendments of
1984 (RCRA 3004(o)). The objectives are being implemented through the
liquids management strategies EPA has .nstituted for each type of land
disposal unit under Code of Federal Regulations, Subparts K through tl.
In general, the liquids management program is a systematic effort to
prevent to the greatest extent possible (using current technology)
migration of hazardous constituents out of the unit.
1.2.1 Background
The new regulations proposed in the liner/leak detection rule
should be considered as companions to the proposed changes in the
minimum technology double liner system requirements presented in the
Federal Register on March 28, 1986 (51 FR 10706). The Agency's
proposed rule codification of March 28, 1986 requires the following
double liner system standards at surface impoundments and landfills:
• A top liner designed, operated and constructed to prevent
migration of liquids into it; and,
1-3
-------
OR
One of two possible bottom liners;
--A bottom liner designed, operated and constructed so that
liquids do not migrate through It. The minimum standard Is a
3-foot-thlck layer of compacted soil with a maximum hydraulic
conductivity of 1 x 10~' centimeters per second (cm/s);
—A composite bottom liner made of two components. The upper
component would be designed, operated and constructed to
prevent migration of hazardous constituents into it and a lower
component designed, operated and constructed to minimize
migration of hazardous constituents through the upper component
if the upper component were breached before the post-closure
care period ends. This lower bottom liner component nust be a
compacted soil with a maximum hydraulic conductivity of 1 x
10"' cm/s.
Subsequent to the proposed rule codification, EPA issued a Notice
of Data Availability on "Bottom Liner Perfornunce in Double-Lined
Landfills and Surface Impoundments" (April 17. 1987) which concluded
that compact°d low-permeability soil bottom liners can impede the
proper functioning of a leak detection system. The Notice showed that
in comparison tc composite bottom liners, a compacted soil bottom
1iner:
• does not maximize leachate removal in the LCRS between the
liners because the compacted soil will absorb some of the
liquid from the LCRS;
• does not allow detection of leakage through the top liner at
the earliest practicable time; and
• does not detect small amounts of leakage thrx .gh the top liner.
For the above reasons, compacted soil bottom liners are believed
to be inferior to composite bottom liners and do not represent best
1-4
-------
demonstrated available technology (BOAT). EPA 1s considering the
possibility of eliminating the single compacted soil bottom liner as
an option.
1.2.2 Leak Detection Systems
The Hazardous and Solid Waste Amendments require an "approved leak
detection system" to be utilized at new landfills, surface
impoundments, waste piles, and land treatment units. The leak
detection system must be capable of detecting leakage at the "earliest
practicable time over all areas likely to be exposed to waste and
leachate during the active life and post-closure care period of the
unit". The basis for the proposed leak detection system is the
leachate collection and removal system (LCRS) between the top and
bottom liners as required by the Proposed Codification Rule of March
28, 1986 (51 FR 10707-12). The Proposed Codification Rule requires a
LCRS between the liners that is designed, constructed, maintained and
operated to detect, collect, and remove liquids that leak through any
area of the top liner during the active life and post-closure care
period. The proposed rule further requires that the LCRS between the
liners be constructed of materials that are chemically resistant to
the waste or leachate in the unit and be designed and operated to
function without clogging dMring the active life and post-closure care
period of the unit.
1.2.3 Extension of Minimum Technology Standards
1.2.3.1 Waste Piles
A double liner system meeting the minimum technology requirements
Is proposed for all newly-constructed waste piles. This is not
specifically mandated by HSWA, but it is being proposed to satisfy the
understood intent of Congress for equivalent levels of protection of
human health and the environment for all units manifesting similar
potentials for migration of hazardous constituents out of the unit.
1-5
-------
1.2.3.2 S1gn1ftcant_Port1ons
A significant portion"is any unlined area of a unit that has r 't
received waste and, 1f double lined before receiving waste, wou d
significantly reduce the potential for migration of hazat
constituents out of the unit, thereby reducing the potential . ^r
ground-water and surface-water contamination.
A double liner system that meets the minimum technology
requirements of RCRA 3004(o) is proposed for all significant portions
of existing landfills, surface impoundments, and waste piles. This
specific provision is not mandated by HSWA jut is proposed for the
purpose of equivalence.
1.2.4 Construction Quality Assurance Program
This program is proposed to be added to the hazardous waste
management regulations to minimize leachate migration Into the
environment by first ensuring the effectiveness of the facility's
design, construction and operating plan. A construction quality
assurance program is proposed for all permitted and interim status
landfills, surface impoundments, waste piles, and land treatment
units. This would apply to all units where construction begins after
promulgation of the liner/leak detection rule.
1.2.5 Applicable Units
These proposed regulations apply to landfills, surface
impoundments, waste piles, and land treatment units. Underground tank
leak detection rules were promulgated under RCRA 3004(o)(4) and
3004(w) on July 14, 1986.
1.3 PURPOSE AND SCOPE OF THE BACKGROUND DOCUMENT
1.3.1 Purpose of the Background Document
The purpose of this document is to provide the technical rationale
and support for the three main portions of the proposed Liner/Leak
1-6
-------
Detection Rule: ) leak detection system requirements; (2) extension
of the double-liner system requirements to waste piles, significant
unused portions and certain other ui.its; and (3) construction quality
assurance program requirements.
The portion of this document devoted to leak detection (Chapter 2)
will: (1) present the technologies and materials available to
construct lining systems and leak detection systems to meet the
proposed regulations; (2) recommend best demonstrated available
technologies (BOAT) to meet the requirements stemming from these
regulations; and (3) quantify leak detection system performance
capabilities associated with BOAT.
This document, to the greatest extent possible, presents a state-
of-the-art review of available technologies and achievable performance
levels In leak detection systems at hazardous waste management units.
This review 1s necessary to establish the best demonstrated available
technology (BOAT) for leak detection systems. It provides the
critical information considered in developing EPA's regulatory options
and describes the criteria used to select among the options. The
document also presents a rationale and technical data to support
extension of the double-liner system requirements to waste piles,
significant unused portions, and certain other units. Lastly, the
document presents a comprehensive discussion of the issues,
methodologies, and benefits associated with construction quality
assurance programs at hazardous waste management facilities.
1.3.2 Scope of the Background Document
1.3.2.1 Technical Information Areas
1.3.2.1.1 Leak Detection Systems
The proposed regulations will require leak detection systems for
new units of landfills, surface impoundments, waste piles, and land
treatment units. The technical areas to be discussed Include: lining
systems material and performance; leak detection technologies;
performance characteristics of leak detection systems between liners;
1-7
-------
analyses of the functioning of leak detect.on systems; action leakage
rate; and response action plans. This information is presented in
Chapter 2.
1.3.2.1.2 Double Liner System for New Waste Pile Units
The proposed regulations will require minimum technology double
liner systems at all newly-constructed waste piles. he technical
areas discussed Include the description and number of waste piles; the
rationale for the proposed changes; the comparable risks to landfills;
and the Issue of equivalent protection. This information is
presented 1n Chapter 3.
1.3.2.1.3 Significant Portions
The proposed regulations will require minimum technology double
liner systems at significant unused portions of existing landfills,
surface impoundments, and waste piles. The technical discussion of
this proposed regulation includes a description of the number of units
it will affect; the rationale for the proposed changes; and the issue
of equivalent protection. This information is presented in Chapter 3.
1.3.2.1.4 Double Liners for New Units, Replacements, and Lateral
Expansions of Facilities Permitted Prior to November 8, 1984
The proposed regulations will require minimum technology double
liner systems at new units, replacements, and lateral expansions of
landfills, surface impoundments, ana waste piles at facilities
permitted prior to November 8, 1984 (except for certain replacement
units permitted prior to November 8, 1984). The technical areas to be
discussed are the comparable risks to landfills and surface
impoundments that these units represent; and the equivalent protection
a double liner system will provide. This information will be presented
In Chapter 3.
1-8
-------
1.3.2.1.5 Construction Quality Assurance Program
The proposed regulations will require a construction quality
assurance program (CQA) for all permitted and interim status
landfills, surface Impoundments, waste piles and land treatment units
regulated under 40 CFR Parts 264 and 265 for which construction
commences later than 12 months after promulgation of the liner/leak
detection rule. Technical issues to be discussed in this section of
the background document include: rationale for a CQA program;
components of a CQA plan; sensitivity of lining system performance to
construction procedures; benefits of a construction quality assurance
program; and testing procedures to evaluate construction quality. This
Information will be presented In Chapter 4.
1.3.2.1.6 Land Treatment Units
The proposed regulations will require leak detection for all land
treatment units. These requirements include a 95 percent confidence
level of detection of hazardous constituents below the treatment zone,
detection of leakage Into the area below the treatment zone at the
earliest practicable time, monitoring above the seasonal high water
table, and a response action plan for widespread leakage. This
Information will be presented 1n Chapter 5.
1-9
-------
Chapter 2
LEAK DETECTION
-------
2.1 INTRODUCTION
2.1.1 Scope of Chapter 2
The purpose of Chapter 2 "Leak Detection" is to provide in a
single document a complete discussion of Issues, technologies, design
and construction, and achievable performance levels for leak detection
systems at surface impoundments, landfills, and waste piles. This
chapter is intended to provide EPA with comprehensive technical
background information on leak detection systems to support its
rulemaking activities related to the proposed liner/leak detection
rule.
In the Introduction (Section 2.1), the following general
introductory material is provided: the definition of hazardous waste
management units and lining systems used in these units; definition of
leakage (what is leakage? what is a leak?); purpose of leak detection
ystems (why Is 1t important to detect leakage? complementarity of
leak detection and leachate collection); and performance
characteristics of leak detection systems (such as ab :ty to evaluate
leakage rate, leak detection time, and determination of leak
location). This section is included as background to ensure that all
readers are familiar with the important basic concepts of hazardous
waste management units, waste containment, lining systems, leachate
collection and removal systems, and leak detection systems.
Section 2.2 is concerned with lining systems materials and
performance. This section first summarizes information regarding
materials used in lining systems. A discussion of the performance of
lining systems is then presented. Failure mechanisms for geomembrane
and soil liners are discussed. Case study information Is presented on
the frequency of seam defects observed in actual geomembrane
installations. Calculations are presented to estimate leakage through
holes in geomembranes and composite liners. The information presented
in this section will be useful in establishing the best demonstrated
2.1-1
-------
available technology (BOAT) for top liner systems, from which action
leak rate recommendations will subsequently be derived (Section 2.9).
Section 2.3 is devoted to leak detection technologies and includes
two parts. The first part is a review of available technologies and
Includes: leachate collection and removal systems used as leak
detection systems, systems based on acoustic emissions, electric
resistivity, and time domain reflectometry, and other technologies.
The second part of Section 2.3 provides a rationale for the selection
of leachate collection and removal systems (LCRS) as leak detection
systems. Thus, the remaining eight sections of Chapter 2 deal
exclusively with leachate collection and removal systems which are
used to satisfy the statutory requirements of RCRA for leak detection
systems capable of detecting leakage in the shortest practicable time.
Section 2.4 provides a detailed description of LCRS type leak
detection systems. Performance requirements, materials used to
construct the LCRS, and the properties of these materials are covered.
In particular, this section compares the properties of natural and
synthetic drainage layer materials. This section is included as
background to ensure that all readers have a source of basic
information on drainage layer materials and properties to draw upon in
the latter sections of Chapter 2 (when leak detection system
pe.'foimance standards and minimum drainage layer component
specifications are discussed).
Sections 2.5 and 2.6 are concerned with the performance of lining
systems and leak detection systems. This Information is Important to
establi h BOAT standards for leak detection systems. The information
will be useful in developing recommendations for leak detection system
performance criteria (Section 2.7) and minimum component
specifications (Section 2.8), and for establishing the action leakage
rate (Section 2.9). The Information presented in Section 2.5 was
obtained from field performance data (leakage collected in sumps).
Section 2.6 presents the results of numerical simulations of lining
system performance carried out by Radian Corporation and GeoServices
Inc.
2.1-2
-------
Section 2.7, entitlea "Performance Criteria for Leak Detection
Systems". provides recommendations on BOAT standards for leak
detection systems. Recommendations are provided for minimum criteria
for leak detection sensitivity, leak detection time, leachate
collection efficiency, and design life of leak detection systems.
These recommendations are based on the review of performance
capabilities presented In Sections 2.5 and 2.6.
Section 2.8 1s devoted to design specifications for the components
of a leak detection system. Minimum specifications for the hydraulic
conductivity and the hydraulic transmisslvHy of the materials used to
construct the leak detection system are discussed, as is the minimum
bottom slope requirement for the waste management unit and minimum
properties for the leak detection system sump. The rationale for
recommend ing certain minimum values is also discussed in this section.
Section 2.9 discusses the concept of an "action leakage rate
(ALR)" which is the leakage rate through the top liner beyond which
EPA intends to require an owner or operator of a unit to initiate
response actions at the unit. ALR recommendations are suggested based
on the BOAT top liner capabilities discussed In Sections 2.2 and 2.6.
Section 2.10 discusses the concept of a "response action plan",
I.e., the actions that should be undertaken when the measured leakage
rate exceeds the "action leakage rate" defined 1n Section 2.9. In
Section 2.10 a discussion is provided on the important variables to
consider in establishing appropriate response actions.
2.1.2 Waste Management and Land Disposal Units
2.1.2.1 Introduction
2.1.2.1.1 Definition
"Waste management unit" is a generic term which encompasses all
containment facilities used to treat, store, or dispose of hazardous
waste. A type of waste management unit is a land disposal unit used
for the temporary storage or permanent burial of hazardous waste.
2.1-3
-------
Land disposal units include surface impoundments, waste piles, and
landfills. In ...Ms document the term "land disposal unit" will be
used as a term to identify surface impoundments, waste piles and
land;iMs. To a lesser extent, the term "waste management unit" will
be used to identify these units (although in its general usage, waste
management unit Includes other types of facilities as well).
2.1.2.1.2 Purpose of this Section
It Is not possible to discuss leakage without a knowledge of:
• the containment facilities from which leakage Is taking place;
and
• the lining systems through wh i leakage is taking place.
The purpose of Section 2.1.2 is to briefly describe hazardous land
disposal units, and to discuss pollution mechanisms that may be
associated with leakage from these units. The next section (2.1.3)
will be devoted to lining systems used in those units.
2.1.2.2 Description
2.1.2.2.1 Types of Land Disposal Units
Three types of land disposal units are considered In this
document: landfills, surface impoundments, and waste piles. These
three types of units are Illustrated 1n Figure 2.1-1 and their usage
is as follows:
• landfills are used for permanent disposal of solid waste
(hazardous waste in "hazardous waste landfills" or municipal
waste 1n "sanitary landfills");
• surface impoundments are used to store liquids (with, possibly,
particles 1n suspension, which settle progressively) or sludges
(which consolidate progressively); and
2.1-4
-------
• waste piles are used for temporary storage of solid waste.
2.1.2.2.2 Geometry of Land Disposal Units
Surface Impoundments - The overall shape of surface impoundments
is roughly that of an inverted truncated pyramid with "side slopes"
and a "bottom". The side slopes can be as steep as permitted by
geotechnical considerations and they typically range between 2H/1V and
4H/1V, while the bottom Is nearly horizontal with just the slope
(e.g., 2%) required for the drainage layer if there is a double liner.
Landfills - The lower part of a landfill has roughly the shape of
an Inverted truncated pyramid, like a surface impoundment. This 1s
the part of a landfill that is lined prior to waste placement. The
side slopes of the bottom part of a landfill can be as steep as
permitted by geotechnical considerations and they typically range
between 2H/1V and 4H/1V, while the bottom is nearly horizontal with
just the slope (e.g., 27.) required for the drainage layers) that is
(are) incorporated into the lining system.
The upper part of a landfill includes a cap that is placed on top
of the waste to close the landfill after completion of waste placement
operations. The cap is a lining system used to prevent (or, at least,
minimize) penetration of rain water into the landfill.
Large landfills may be divided into cells which are operated
sequentially.
Waste Piles - A waste pile can have any shape compatible with
waste stability. The lining system placed under the waste pile is
nearly horizontal, with just the slope (e.g., 2%) required for the
drainage layer(s) that is (are) incorporated into the lining system.
2.1-5
-------
2.1.2.3 Ground Pollution Mechanism
2.1.2.3.1 ''•irface Impoundments
A surface impoundment can cause pollution of soil and ground water
1f the hazardous liquid contained In the Impoundment leaks through the
lining system and Into the ground.
In rare occasions, waves of liquids stored in surface impoundments
have overtopped the crests of the impoundments thereby causing ground-
water pollution.
2.1.2.3.2 Landfills
The mechanism by which a landfill can cause soil and ground-water
pollution Includes two steps:
• first, leachate Is generated In the landfill; and
• then, pollution occurs if some leachate migrates through the
lining system into the ground.
Leachate can be produced by two mechanisms, intrusion of water
into the waste and generation of leachate within the waste:
• Intrusion of Water in the Waste. The main cause of leachate
production is infiltration of rain water into the waste. The
rain water seeping through the waste becomes progressively
polluted and the resulting polluted liquid is called
"leachate". In exceptional cases, leachate can be produced by
Intrusion -f ground water into the waste (if the ground water
table rl3uj), or, even more exceptionally, by Intrusion of
flood water into the waste.
* Generation of Leachate within the Waste. Leachate can
originate in the waste if liquid is entrapped In the waste
during waste placement. Drums containing liquids are not
allowed in hazardous waste landfills, and the only possibility
2.1-6
-------
for entrapping liquids is through moisture in the waste or in
the earth used for the daily covers (i.e., the layers of
compacted earth, placed every day on the waste). Part of the
moisture included in the waste or the daily covers can be
expelled by consolidation (i.e., decrease in volume of the
waste and the dally covers due to compression caused by the own
weight of the waste and the dally covers themselves).
To prevent pollution of soil and ground water by landfills, all
efforts should be made to prevent production of leachate:
• A low-permeability cap must be placed on the landfill
Immediately after completion of waste placement operations to
prevent Intrusion of rain water.
• Selection of landfill location and appropriate design should
minimize Intrusion of ground water and flood water.
• Waste and dally cover material should not contain excess
1iquids.
Since leachate production cannot be totally prevented, especially
during landfill operation (i.e., during waste placement) when rain can
fall freely on the landfill, a lining system is necessary at the
bottom and on the side slopes of the landfill.
2.1.2.3.3 Waste Piles
The two-step mechanism by which waste piles can cause soil and
ground-water pollution 1s similar to the mechanisms related to
landfills which were described in Section 2.1.2.3.2. Waste piles are
temporary storage units and the waste 1s normally removed after some
time.
2.1-7
-------
2.1.3 Lining Systems Used In Land[Disposal Units
2.1.3.1 Introduction
2.1.3.1.1 Importance of Lining Systems
From the above discussion, it Is clear that the lining s>;tem
placed on the bottom and the s*de slopes of a land disposal unit has a
critical role: the ground is polluted as soon as liquid leaks through
the lining system. Therefore It is essential to have a good knowledge
of lining systems prior to discussing leakage.
2.1,3.1.2 Scope of this Section
The purpose of this section is to provide basic Information on the
types of lining systems used 1n hazardous waste land disposal units,
and on the materials used to construct these lining systems. This
section should familiarize the reader with the vocabulary used to
describe lining systems.
This section will addres the following: definition of lining
systems, materials used in lining systems, double liners, and
composite liners. (Experience shows that it is not practical to
discuss double liners and composite liners without a knowledge of
materials used to construct lining systems.)
2.1.3.1.3 Definition of Lining Systems
The terms "Uner" and "lining system" are not synonymous.
A liner Is a low-permeability barrier used to impede liquid or gas
flow. Note that "low oermeabi1ity" is used, and not "impermeable".
If there was such a thing as an impermeable barrier, it would be
possible to prevent leakage, and many of the discussions and
considerations presented in this background document would be
pointless. Although it may be possible that a glass is impermeable to
water, in modern technology there is no material that is impermeable
2.1-8
-------
at the scale of a land d> >posal unit where the area to be lined can be
as large as tens of hectares (dozens of acres).
Since no liner Is Impermeable, pollution control can only result
from a combination of liners and drainage layers, performing
complementary functions:
• Liners (which are low-per-eability barriers) impede the flow of
undesirable (polluted) liquids toward the ground.
• Drainage layers (which have a high permeability) convey the
undesirable flow away fr-m the ground.
Such combination of liners and drainage layers is called a "lining
system".
2.1.3.2 Materials Used in Lining Systems
2.1.3.2.1 Introduction
• low-pcimcabl 11 ty ihJtcM lu I r> to cunsliucL Hie linc'is;
• high-permeability materials to construct the drainage layers;
• transition materials (or interface materials) acting as filters
or protective layers (i.e., providing filtration or protection)
between various layers jf a lining system; and
• reinforcement materials which increase the strength of a lining
system (1f required).
These materials are briefly discussed below.
2.1-9
-------
?.' .2.2 Liner Materials
Low-permeability materials used In civil engineering to construct
liners Include: compacted low-permeability soils, geomembranes,
concrete, and asphaltic concrete. Concrete and asphaltic concrete are
not used 1n hazardous waste units for the following reasons:
• Concrete liners tend to undergo cracking and therefore tend to
leak . gnlflcantly.
• Asphaltic concrete cannot be used because asphalt has a poor
therefore, only low-permeabl 11 ty soils and gcomciiibranes are
discussed in this document.
Compacted Soils - Compacted low-permeability soils used to
construct liners include: clay, sllty clay, clayey sands, and silty
sands. If such soils are not available at the site, 1t is possible to
make a compacted low-permeability soil by mixing bentonite with sand.
Bentonite is composed of extremely small particles of sodium
montmori1lonite. When it is dry, it becomes a powder which can be put
in bags, and 1s purchased and transported like cement.
In the remainder of this document compacted }r -permeability soil
liners will be referred to simply as compacted soil liners.
Geomembranes - Geomembranes are low-permeability membranes used in
civil engineering as fluid barriers. By definition, a membrane is a
material that 1s thin and flexible.
All geomembranes presently used in hazardous waste management
units are synthetic geomembranes. (Asphaltic geomembranes, which are
used for lining water storage facilities, are not us-. in hazardous
waste containment units because they do not have adequate resistance
to chemical attack.) Typical examples of geomembranes used in
2.1-10
-------
hazardous waste containment units include: high density polyethylene
(HOPE) geomembranes; linear low density polyethylene (LLDPE)
geomembranes; polyvinyl chloride (PVC) geomembranes; and
chlcrosul fonated polyethylene (CSPE) geomen-.branes.
The term geomembrane is often used by the engineering community in
place of the term "flexible membrane liner" (FML). EPA is using the
term "flexible membrane liner" or "FML" to be consistent with the
terminology used in the past in documents discussing waste management
units. Therefore, for consistency with previous EPA documentation,
"flexible membrane Uner" or "FML" will be used in the remainder of
this document to describe synthetic membranes used as low-permeability
liners.
2.1.3.2.3 Drainage Materials
High-permeability materials used to construct drainage layers
Include: high-permeability soils, synthetic drainage materials, and
pipes. High-permeabil if-" soils and synthetic drainage materials are
discussed below.
High-Permeability Soils - High-permeability soils include a wide
variety of sands and gravels ranging from fine to coarse in size and
well-graded to uniform in gradation. Selection of a high-permeability
soil for specific conditions must consider the following:
• the drainage layer should be able to collect and rapidly remove
liquids entering the leak detection, collection and removal
system as a result of leakage through the top liner;
• the high-permeability soils should not damage FMLs when the
FMLs are directly 1n contact with the soils; and
2.1-11
-------
• the drainage layer should be physi 'lly compatible with
transition materials to prevent any potential migration of the
transition materials into the drainage layer which could lead
to clogging.
Synthetic Drainage Materials - Synthetic drainage materials are
made of planar structures which are thick enough to convey fluids in
their plane. Synthetic drainage materials are usually made from
polymers. Typical polymers Include polypropylene, polyester, and
polyethylene. These polymers are highly Inert to biological and
chemical degradation.
Four types of synthetic drainage materials are currently
available. These are thick needlepunched nonwoven geotextiles,
geonets, geomats and corrugated or waffled plates. With the exception
of needlepunched nonwovens, these materials can be combined with
geotextile filters to form drainage geocomposites.
2.1.3.2.4 Transition Materials
Transition materials include filters or protective layers.
Fi 1ters - Filters are located between the drainage layer and the
soil to be protected. They usually consist of a granular layer or a
combination of granular layers, or a geotextile. Their function is to
allow free flow into the drainage layer and at the same time prevent
the migration of particles of the protected soil into the drainage
layer.
Protective Layers - Protective (cushion) layers are located
between the drainage layer and the FML. Their function is to protect
the FML from damage by the drainage material. Cushion layers usually
consist of a sand layer or a thick needlepunched nonwoven geotextile.
2.1.3.2i.5 Reinforcement Materials
Reinforcement materials are typically placed in a soil layer.
Typical functions include reinforcing the lining system on steep
2.1-12
-------
slopes to prevent sliding along the slope, reinforcing slopes to
prevent slope failure, or bridging over cavities, depressions or soft
spots. The materials most frequently used in reinforcement
applications at waste management units are geogrids and geotextiles.
2.1.3.3 Double Liners
2.1.3.3.1 Introduction
- Double Liner
A "double iner lining system" simply called a "double liner
system" or a "double liner" 1s a lining system which includes two
liners with a leachate collection and removal system between the two
1iners.
Clearly, two liners in contact (i.e., without a leachate
collection and removal system between the two liners) do not
constitute a double liner (they constitute a single liner, as
discussed below).
- Single Liner
A lining system which includes only one liner is called a "single
liner".
- Composite Liner
A composite liner is a liner comprised of two or more low-
permeability components made of different materials in contact with
each other. For example, a FML and a compacted soil layer placed in
contact with each other constitute a composite liner. Composite
liners do not constitute a double liner because there is no leachate
collection and removal system between the two low-permeability
components.
The purpose of a FML-compacted soil composite liner is to combine
advantages of FMLs and soils. FHLs have a much lower permeability
2.1-13
-------
than compacted soils, but they may have holes through which leakage
can occur if the FML is placed on a pervious medium and then subjected
to a hydraulic head on its top surface. The leakage rate through a
FML hole is reduced if there is compacted low-permeability soil under
the FML.
- Terminology Related to the Liners
In this document, the upper liner of a douole liner is called "top
liner" and the lower liner 1s called "bottom liner". We recognize
that this terminology may be confusing since the term "bottom liner"
may be mistaken for "bottom lining system", I.e., the lining system
located at the bottom of a waste management unit.
"Top liner" is synonymous with "upper liner" or "primary liner".
"Bottom liner" is synonymous with "lower liner" or "secondary
liner".
- Terminology Related to the Leachate Collection and Removal Systems
In all land disposal units lined with a double liner there is a
pervious layer between the two liners. This layer is called the
"leachate collection and removal system (LCRS) between f-a liners".
If this system is also used as a leak detection system (LDS), its name
becomes "leak detection, collection, and removal system" (LDCRS).
While in surface impoundments there is only one pervious layer
(I.e., the LDCRS mentioned above), there are two pervious layers in
landfills: the LDCRS and the layer located above the top liner and
called the "leachate collection and removal system (LCRS) above the
top liner".
2.1-14
-------
2.1.3.4 yse_of Double Llnersjn Land Disposal Units
2.1.3.4.1 Current Regulations
Current EPA regulations (40 CFR Parts 264 and 265) require a
double Uner system 1n '•11 new hazardous waste landfill and surface
impoundment units except units permitted prior to November 8, 1984 or
units where variances are allowed. Furthermore, as discussed in
Chapter 1, the two liners comprising the double liner system should
meet the following requirements:
• "A top liner designed, operated, and constructed of materials
to prevent the migration of any constituent Into such liner
during the period such facility remains In operation (including
any post-closure monitoring period}".
• "A bottom Hner designed, operated and constructed to prevent
the migration of any constituent through such liner during such
period. For the purpose of the preceding sentence, a lower
liner shall be deemed to satisfy such requirement if it is
constructed of at least a 3-foot thick layer of recompacted
clay or other natural material with a permeability of no more
than 1 x 10"' centimeter per second."
According to the Draft Minimum Technology Guidance on Double Liner
Systems of May 24, 1985 (see EPA/530-Srf-35-012):
• The top Hner FML should be at least 0.75 mm (30 mil) thick, if
1t 1s protected in a timely manner after placement; if 1t 1s
not protected in a timely manner the top liner FML should be at
least 1.15 mm (45 mil) thick.
• The upper FML component of a bottom composite liner should be
at least 0.75 mm (30 mil) thick.
2.1-15
-------
2.1.3.4.2 Examples of Uses of Double Liners 1n Land Disposal Units
- Types of Double Liners Used in Land Disposal Units
From the above discussion, it, appears that four types of double
liners are currently allowed by existing EPA regulations (Figure
2.1-2). Such double liners can be used for landfills, surface
Impoundments, and waste piles. The double liner using two composite
liners (Figure 2.1-2(b)), called "double composite liner", is
increasingly used in order to minimize the amount of leakage through
the top liner while maximizing the collection efficiency of the LDCRS.
- Caution on the Use of Top Composite Liners in Surface Impoundments
The use of a top composite liner in a surface impoundment requires
special caution. If the FML (which is the upper component of the top
composite liner) is not covered with a heavy material (such as a layer
of earth, or concrete slabs), and if there is leakage through the FHL,
liquids tend to accumulate between the low permeability soil (which is
the lower component of the top composite liner) and the FML since the
submerged portion of the FML (whose specific gravity is close to 1) is
easily uplifted. Then, if the impoundment is rapidly emptied, the FML
is subjected to severe tensile stresses because the pressure of the
entrapped liquids is no longer balanced by the pressure of the
impounded liquid. Therefore, a top composite liner shuuld always be
loaded, which 1s automatically the case in a landfill or in a waste
pile, and which must be taken into account in the design of a liquid
impoundment.
2.1.3.4.3 Influence of Top and Bottom Liners on Leak Detection
The LCRS between the top and bottom liner is also used as a leak
detection system to form a leak detection, collection, and removal
system (J.DCRS). The leakage through the top liner flows in the LDCRS
over the top surface of the bottom liner. The two liners have the
following influence:
2.1-16
-------
• The top liner governs the amount of leakage entering the LOCP3.
Many hazardous waste management units include a top composite
liner in order to minimize leakage through the top liner.
• The bottom liner has a major influence on tha performance of
the LDCRS. As was shown in EPA's recent Notice of Availability
of Data on "Bottom Liner Performance In Double-Lined Landfills
and Surface Impoundments" (US EPA, 1987), a compacted soil
Hner allows greater leakage Into and through the bottom liner
than does a composite. For this reason, a composite bottom
liner (Figure 2.1-2 (a and b}) 1s preferable to a compacted
soil bottom liner (Figure 2.1-2 (c and d)).
As was shown In the Notice of Availability of Data, owners and
operators of hazardous waste management units rarely use compacted
soil bottom liners (Figure 2.1-2 (c and d)) because of the
performance deficiencies associated with them in comparison to
composite bottom liners (Figure 2.1-2 (a and b)).
2.1.4 Leakage Definition and Detection
2.1.4.1 Definitions
2.1.4.1.1 Leak and Leakage
According to Webster:
• A leak 1s "a crack or opening that permits something to escape
from or enter a container or conduit".
• Leakage 1s "something that escapes by leaking" or "an amount
lost as the result of leaking".
From these definitions, 1t clearly appears that what is monitored
between the top and the bottom liners 1s the leakage, not the Ipaks.
Therefore, the monitoring system should be called "leakage detection
system". While "leakage detection system" Is grammatically correct,
the phrase "leak detection system" has been codified by RCRA. For the
2.1-17
-------
sake of consistency with the law, the phrase "leak detection system"
will be used In this document.
On the other hand, systems used in quality assurance of
geomembrane Installation, si'ch as the vacuum box, are clearly Intended
to find leaks.
2.1.4.1.2 Leak Size and Leakage Rate
According to the above definitions, the term "leak size"
designates the size of a hole, expressed as a surface area or
dimensions such as a diameter (e.g., a 1 cm* leak, a 1 in.* leak, a 2
mm diameter leak, a l/4-1n. diameter leak). The term "leak size" is
sometimes mistakenly used for "leakage rate" which 1s the flow rate
through a leak or a group of leaks, which is expressed as a volume per
unit of time (m'/s, liters/day, gallons/day). The term "leakage rate"
will often be used In this document as an abbreviation for "leakage
rate per unit area", which 1s expressed as a volume per unit of time
per unit of area (m'/s/m* (which Is equivalent to m/s),
liters/hectare/day, 11ters/lOOOm2/day (Ltd), gallons/acre/day (gpad)).
(Note: 1 hectare - 100 m x 100 m = 10 000 m*.)
The fc1' wing conversions apply:
1 gallon/acre/day = 1.08 x 10"" m/s
« 9.35 liters/hectare/day
- 0.935 liters/1000 mVday
1 liter/hectare/day = 1.16 x 10~12 m/s
» 0.11 gallons/acre/day
•= 0.1 liters/1000 mVday
1 llter/lOOOmVday * 1.16 x 10"" m/s
» 1.1 gallon/acre/day
= 10 liters/hectare/day
2.1-18
-------
1 m/s = 8.64 x 1Q10 1 Hers/lOOOm'/day
= 8.64 x 10" liters/hectare/day
= 9.21 x 10'° gallons/acre/day
From a practical standpoint, the approximate conversion can be
used:
1 liter/lOOOm'/day = 1 gallon/acre/day
1 Ltd = 1 gpad
2.1.4.1.3 Leakage Collected end Leakage Out of the Unit
The possible fates of liquids entering a double liner system arc
shown 1n Figure 2.1-3.
The leakage discussed in the previous sections is the leakage that
the LDCR5 system is intended to collect and detect. This is the
leakage through the top liner (C in Figure 2.1-3).
The leakage out of the unit, which is {.he leakage through the
bottom liner (J in Figure 2.1-3), is only a fraction of the leakage
through the top liner. Other fractions Include:
• leakage entrapped in the LDCRS by absorption, capillarity,
ponding, etc. (F in Figure 2.1-3);
• leakage collected at the LDCRS sump (G in Figure 2.1-3); and
• leakage absorbed in the bottom liner (I in Figure 2.1-3).
If the LDCRS is properly designed, the liquid head on the bottom
liner is very small, and leakage through the bottom liner (which is
governed by head on the bottom liner) is very small. This is
consistent with EPA's goal of protecting human health and environment
through system impermeability and not liner impermeability. No liner
is perfectly impermeable but proper design can almost achieve system
impermeabil ity.
2.1-19
-------
2.1.4.2 Leak Detection System
2.1. -1.2.1 Definition
In the context of this background document, leak detection refers
to leakage through the top liner and, therefore, a leak detection
system 1s a system which Is placed between the two liners of a double
Hrer system to monitor the leakage through the top liner.
2.1.4.2.2 Purpose of Leak Detection
As indicated 1n the above definition, the purpose of a leak
detection system 1s to monitor leakage through the top Uner.
Monitoring leakage through the lop liner 1s sn important component to
EPA's systems approach to the containment of hazardous constituents
using double liner systems.
2.1.4.2.3 Performance Requirements of Leak Detection Systems
- Review of Potential Performance Requirements
What performance requirements should be considered when designing
or evaluating a leak detection system? In other words, what do we
expect from a leak detection system.
Based on the discussion presented previously it can be deduced
that the leak detection system should essentially:
• evaluate the leakage rate; and
• provide this information rapidly so action can be taken without
delay, if necessary.
2.1-20
-------
Leak detection systems may alr.o perform other functions such as:
• leachate collection; and
• leak location Identification.
If the drainage layer placed between the top and bottom liners Is
used for leak detection, It collects leakage as well as detects 1t.
On the other hand, electric resistivity or acoustic emission leak
detection systems detect leakage without collecting 1t. It thus
appears that leakage detection and collection are two separate
functions (although, 1n some cases, they can be performed
simultaneously). Leachate collection s already provided in all
hazardous waste management units because there 1s a statutory
requirement for a leachate collection and removal system between the
two liners. Therefore, the ability to collect leakage cannot be
considered as a performance characteristic for a leakage detection
system. However, 1t can be considered as an additional benefit.
The need to locate a leak 1n a land disposal unit will vary based
on factors such as the type of unit, stags of active life, rate of
leakage and available response actions. In many instances, the need
to locate a leak will be limited because:
• In landfills, no significant leakage is likely to occur after
closure of the unit. If the rate of leakage through the top
Hner 1s In the range of a typical remedial action consists of
capping a certain area of the unit (i.e., covering the waste
with a Uner) to prevent Infiltration of rain water, thereby
substantially reducing leachate production. It is not
necessary to know exactly where leakage occurs to design the
capping of an area of the unit.
• In surface Impoundments, the strategy in the case of a large
leak 1s different. The Impoundment can be emptied and the
Uner either repaired or replaced (retrofitted). It is
therefore useful to now the location of the leaks. However,
leaks can be located by means that do not need to be built into
2.1-21
-------
the lining system. These means Include portable electric
resistivity equipment, vacuum testing, or visual inspection of
the lining system after emptying of the impoundment.
- Selection of Performance Requirements for Leak Detection Systems
From the abov discussion, It appears that the two key performance
characteristics thuc should be considered when designing or evaluating
a leak detection system are:
• the ability of the system to correctly evaluate leakage rate;
and
• the ability of the system to provide rapid information on
leakage rate.
It also appears from the discussion that: (i) the ability of the
system to collect leachate Is not a relevant performance
characteristic for leak detection systems (however it is the key
performance characteristic for leachate collection .ind removal
systems, and since the LDCRS serves as both a leak detection system
and a leachate collection and removal system the LDCRS must have
leachate collection and removal capability); and (ii) the ability of
the system to locate leaks 1s not a primary performance
characteristic.
2.1-22
-------
•/-
(
-------
Ol)
eP Lir.«sr- COM rented So, I
Top Lme- . LbCR
Cb)
-*L_ r^
l.ner ^ o, >0.1... (3
(c)
i lie.*- Coni'aclfiJ Soi
ML
o Mom oM,atf So.
(i o ' < -/-,)
. , . -
so, I | > O-1.., (3 n) ^
-------
Leachate Collection and
Removal System (LCRS)
Top Liner
A = le.ichat" collected in the LCR5 - - •-
B = leachate stored* in LCI'3
C = leacliate from the LCRS into top liner
D = leachatc stored* in top liner
E = leakage from the top liner into the
LDCRS
Leak Detection
Collect ion and
Removal System
(LDCRS)
G - leakage collected in the LDCRS sump
F = leakage stored* in LDCRS
H = leakage from the LDCRS into the
bottom 1iner
Bottom Liner
Ground
I = leakage stored* in the bottom liner
J = leakage from the bottom liner
into the ground
* Stored liquids due to capillarity, absorption, etc.
Figure 2.1-3.
Fate of liquids entering a double liner system at a
landfill unit.
2.1-25
-------
2.2 TOP LINER PERFORMANCE
2.2.1 Introduction
2.2.1.1 Scope
It is not possible to discuss leakage detection without first
discussing leakage through top liners since the leakage detected by
the leek detection system is the leakage that has migrated through the
top liner, as indicated in Sections 2.1.4.1.3 and 2.1.4.2.1. Also,
it is necessary to understand the information presented in this
section concerning how well top liners perform in order to establish
recommendations for the action leakage rate (ALR) discussed in Section
2.9.
2.2.1.2 °£93Dl?ati20
Section 2.2 is organized as follows:
• Section 2.2.2 first presents the two types of top liners, i.e.,
the FML top liners and the composite top liners. Then, Section
2.2.2 discusses top liner materials, and presents an overview
of typical defects likely to occur in top liners.
• Sections 2.2.3 reviews and analyzes data pertinent to leakage
through FML top liners, while Section 2.2.4 does the same thing
for composite top liners.
• Section 2.2.5 presents conclusions drawn from data presented in
Sections 2.2.3 and 2.2.4.
2.2-1
-------
2.2.2 Tog_Liners
2.2.2.1 Types of Top Liners
2.2.2.1.1 FML
EPA's Propo^d Double Liner Rule of Marc/i 28, 1986 (FR 10706-
10723) requires a top liner designed, constructed, and operated to
prevent migration of liquids into It. This requirement has been
interpreted to mean that the top liner, at a minimum, must consist of
a FML. According to the Draft Minimum Technology Guidance on Double
Liner Systems of May 24, 1985 (EPA 530-SW-85-012), the FML top liner
should be at least 0.75 mm (30 mil) thick, 1f it is protected in a
timely manner after placement; If it is not protected in a timely
manner, the top FML should be at least 1.12 mm (45 mil) thick.
2.2.2.1.2 Low-Permeability Compacted Soil
Based on the above interpretation of EPA's Proposed Double Liner
Rule, a layer of low-permeability soil alone (i.e., without an FML) is
not accepted as a top liner for hazardous waste land disposal units,
However, a layer of low-permeability compacted soil can be used in
association with an FML to form a composite liner, as discussed below.
2.2.2.1.3 Composite Liner
Composite liners were defined in Section 2.1.3.3.1 as liners
comprised of two or more low-permeability components made of different
materials in contact with each other. The composite liner used as a
bottom liner for hazardous waste land disposal units in the Proposed
Double Liner Rule must have an upper component designed, operated, and
constructed to prevent migration of hazardous constituents into it and
a lower component designed, operated, and constructed to minimize
migration of hazardous constituents through 1t 1f the upper component
were breached before the post closure care period ends. At a minimum,
this lower bottom liner component must be a compacted soil with a
maximum hydraulic conductivity of 10~* m/s (10"' cm/s).
2.2-2
-------
The EPA does not require that the top liner be a composite liner.
However, composite top liners are accepted by the EPA and they are
increasingly used by owners and operators in order to minimize leakage
through the top liner. In addition, since composite top liners are
not required, the low-peimeabi1ity soil layer used in a composite top
liner does not need to meet the hydraulic conductivity and thickness
requirement Indicated above.
2.2.2.2 Types_of Materials Used for_Top Liners
2.2.2.2.1 FMLs
FMLs have been described in [USEPA, 1983] and [Giroud and Frobel,
1983] and the remainder of this Section 1s reproduced from the latter.
Polymers are chemical compounds of high molecular weight. Only
synthetic polymers are used to make FMLs. The most common types of
polymers presently used as base products in the manufacture of FMLs
can be classified as indicated in Table 2.2-1. FMLs most often used
in hazardous waste management units include HOPE, LLDPE, CSPE. and PVC
(the latter mostly for caps on top of landfills).
FMLs may be non-reinforced or reinforced with a fabric. According
to Giroud and Frobel [1983], fabric reinforcement is used for one or
several of the following reasons: (1) to impart stability to the
compound (e.g., asphalt, CSPE) during the manufacturing process; (2)
to provide dimensional stability to compounds that would shrink or
expand excessively as a result of cha.ige in environmental conditions
such as temperature; (3) to increase the mechanical strength
(tensile, tear, burst, puncture) of the FML to prevent it from being
damaged during handling and installation, and to allow it to withstand
design stresses; and (4) to increase the deformation modulus of the
FML 1n order to decrease its elongation when subjected to tensile
stresses.
Fabric reinforcement can be of various types depending on the
manufacturing process of the FML. Fabric reinforcement used with FMLs
2.2-3
-------
are woven or nonwoven (typically needlepunched} fabric. An
increasing quantity of FMLs are reinforced with a ?onwoven fabric
placed on one side of the FHL. However, most reinforced FMLs
presently available are reinforced with a woven fabric placed inside
the FML. The most typical case is a "scrim" (1ight-«ight open weave
fabric) placed between two plies of polymeric compousf. Stability of
the three-ply FML is ensured by adhesion of the twcpolymeric plies
through openings of the scrim. This adhesion medanism is called
"strike-through".
Some FMLs have a roughened or embossed surfacf which increases
their friction coefficient in order to prevent the d»elopment of slip
surfaces on FMLs installed on slopes.
2.2.2.2.2 Low-Permeability Compacted Soils
While the types of low-permeability compacted jails that can be
used in bottom composite liners are limited by the EPf requirement for
a maximum hydraulic conductivity of 1 x 10~* m/s (1x 10"' cm/s), a
variety of low-permeability soils can be used to construct the low-
permeability compacted soil layer .f composite top 1 Tiers. These low-
permeability compacted soils include clays, silty cl%;sf clayey sands,
and si 1ty sands.
2.2.2.3 Permeab!l1ty_of L1ner_Mat3r1als
2.2.2.3.1 Introduction
FMLs and low-permeability compacted soils are tsed co construct
lining systems essentially because of their low permeability to
liquids. However, these liner materials are not totally impermeable
and, to discuss leakage, it is essential to have a gjod understanding
of the permeability of liner materials.
2.2.2.3.2 Permeability of Compacted Soils
Flow rate of liquids through porous media sue* as soils can be
expressed by Darcy's equation:
2.2-4
-------
Q = k i A (Equation 2.2-1)
where: Q = flow rate; k = soil hydraulic conductivity; i = hydraulic
g;adient; and A = area perpendicular to the flow. Recommended SI
units are: Q (m'/s), k (m/s), i (dimensionless), and A (m2).
The apparent flow velocity can ba defined as the velocity the
liquid would have f the liquid were flowing over the entire area.
The apparent flow velocity is therefore obtained by dividing the flow
rate by the area perpendicular to the flow:
v = Q/A = k i (Equation 2.2-2)
The liquid flows only In a fraction of the area perpendicular to
the flow, the portion that 1s not occupied by soil particles.
Therefore the actual velocity of the flow is larger than the apparent
velocity v. It would be extremely difficult to determine the actual
velocity of liquid molecules bec-'ise of the tortuosity of the flow
paths between soil particles. Hi..,ever, it 1s easy to determine the
average velocity of the flow parallel to the direction of the flow,
using the following relationship:
vs = v/n (Equation 2.2-3)
where: vs .- seepage velocity (i.e., average component of the actual
velocity of liquid molecules parallel to the average direction of the
flow); v = apparent velocity; and n = soil porosity.
Typical values of soil porosity are 0.25 - 0.50 (also expressed as
percent: 25% - 50%).
The apparent velocity should be used when calculating the flow
rate, while the seepage velocity should be used to calculate the time
1t takes a given liquid to flow from one point to another.
2.2-5
-------
As indicated by Darcy's equation, the degree of permeability of a
soil is expressed by its hydraulic conductivity, k. Typical values of
soil hydiaulic conductivities are given in Fable 2.2-2.
Leakage rate through low-permeability compacted soils can
significantly increase if the soil layer has defects such as those
discussed in Section 2.2.2.4.3.
2.2.2.3.3 Permeation through FMLs
The mechanism by which liquids migrate through FHLs is very
complex and Is discussed in Section 2.2.3.1.2.
FMLs are not porous media like soils and, therefore, flow of
liquids or gases through FMLs is not governed by Oarcy's equation.
This is why the terminology "permeation through FMLs" is preferred to
the terminology "permeability of FMLs". However, for the sake of
comparison, permeability of FMLs can be expressed by an equivalent
hydraulic conductivity. Typical values of equivalent hydraulic
conductivities for FMLs range between 10~" m/s and 10~" m/s. These
values show that permeation rates through FMLs are many orders of
magnitude smaller than the flow rate through clays which are the least
permeable soils. Leakage rates due to permeation through FMLs are
usually very small compared to leakage rates due to FML defects (which
are discussed In the next section).
2.2.2.4 Typical Defects of_L1ner_Mater1a]s
2.2.2.4.1 Introduction
As indicated in Section 2.2.2.3, leakage rates through low
permeability soils as well as FMLs can significantly increase if these
materials have defects.
There are several possible causes of defects in the FML component
and low-permeability soil component of top liners that could lead to
leakage through the top liner. Typical defects are discussed in the
following two sections.
2.2-6
-------
2.2.2.4.2 FML Defects
Defects tt'at are likely to occur to nils are numerous and may be
caused by .a wide variety of factors including improper design,
defective manufacturing and defective installation. A number of
publications are available which discuss various types of defects
observed in FML-lined units [Bass, 1985,; Giroud, 1984a; Giroud,
1984c; Mitchell, 1984J.
Typical defects observed in FMLs Include:
• defective seams during fabrication, installation, or unit
operation as a result of factors including excessive moisture
or humidity, improper temperature, contamination by dust or
A • dirt, undone seams, or excessive stresses during unit operation
caused by improper design;
• damage to the FML during construction or facility operation as
a result of excessive stresses caused by equipment;
• puncturing of FML by stones in the support or cover material;
• tensile failure of the FML due to excessive stresses generated
by weight of stored material and movements of materials in
contact with the lining system;
• fatigue failures caused by slow crack growth mechanisms
prompted by repeated stresses such as thermal expansion-
contraction; and
• inadequate connections between FMLs and appurtenances.
The last type of defect is illustrated by the following case
history presented in Section 2.5.3. After the construction of a
lining system for a landfill, and before placement of the waste, the
lower portion of the landfill was filled with water to test the lining
system. A leakage of 1000 liters/day (250 gallons/day) was observed
at the connection between the FML and the sump, with a head of only
2.2-7
-------
0.15 m (6 In.) of water above the defective connection.
The potential for the above mentioned defects to occur is iv.iniral
in properly designed and constructed FHL-lined units. However, even
in properly designed and constructed units, there is no guarantee that
these defects will not occur. A discussion of the frequency and size
of FML defects is presented in Section 2.2.3.2.
2.2.2.
-------
Defects 1n a low-p"-meabi1ity soil layer increase its hydraulic
conductivity, thereby increasing leakage rate through the top liner.
2.2.2.4.4 Composite Liner Defects
The preceding two sections address defects specific to each of the
two components of a composite liner: the FML and the low-permeability
soil. In addition to these defects, there are defects that are
inherent to the composite liner itself, such as a poor contact between
the FML and the low-permeability compacted soil. Poor contact may
result from wrinkles in the FML and/or irregularities or clods at the
surface of the low-permeability soil. FML wrinkles may exist even
under very high pressures as shown by Stone [1984J.
In addition, it should be noted that the placement of the low-
permeability compacted soil of a composite top liner may damage the
underlying geosynthetics, e.g., the FML of the composite bottom liner.
This is especially true if the leak detection system is comprised of a
geosynthetic material such as a geonet. Because synthetic drainage
layers are thin and are not able to provide protection to the
underlying FML component of a composite bottom liner, compaction of
the bottom part of the low-permeabi1ity soil layer of composite top
liners must be carried out with ca.e. A frequent solution consists of
placing a thicker first lift (e.g. 0.3 to 0.45 m (12 to 18 in.) in
thickness) and compacting with heavy, rubber-tired equipment.
However, with such thick lifts, the potential exists for not
compacting sufficiently the bottom part of the HTL adjacent to the
leak detection drainage layer. As a result, part of this first lift
probably has a higher hydraulic conductivity than the overlying,
better compacted thin lifts.
2.2-9
-------
2.2.3 Leakage through FML Top Liner
2.2.3.1 Introduction
2.2.3.1.1 Scope of the Section
As Indicated In Section 2.2.2.1, 'op liners can be either FMLs
alone or composite liners. Section 2.^.3 discusses leakage through
FML top liners. Leakage through composite top liners will be
discussed In Section 2.2.4.
In Section 2.2.2.3.3, it was pointed out that leakage through a
FML is very complex and can occur as a result of: (1) permeation
through an Intact FML; and (ii) flow through defects in a FML.
Accordingly, Section 2.2.3 will discuss first leakage due to
permeation through FMLs and, then, leakage due to FML defects. It
should be understood that permeation through FMLs is an area of active
research and that new information will become available in the near
future. EPA will be soliciting new information and comments on FML
permeation during the public comment period for the proposed
Liner/Leak Detection Rule and this document will be updated to reflect
the new Information once it is available.
2.2.3.1.2 Organization of the Section
The first part of Section 2.2.3 (i.e., Section 2.2.3.2) is devoted
to leakage evaluation:
• Evaluation of leakage due to permeation through FMLs is
addressed in Section 2.2.3.2.2.
• Evaluation of leakage due to flow through FML defects is
addressed In Section 2.2.3.2.3 (pinholes) and 2.2.3.2.4
(holes).
This leakage evaluation indicates that leakage due to permeation
Is generally negligible as compared to leakage due to defects.
Therefore, to determine leakage through a FML top liner it is
2.2-10
-------
necessary to estimate the number and size of defects in the ri'L.
Accordingly the second part of Section 2.2.3 (i.e.. Section 2.2. .3)
is devoted to an assessment of the frequency and size of FHL defects
using data from field experience.
Finally, conclusions regarding leakage through FHL top liners are
presented In Section 2.2.3.4.
2.2.3.2 Evaluat1on_of Leakage through FML Top Liners
2.2.3.2.1 Introduction
Three mechanisms of leakage are considered:
• leakage due to permeation through an intact FHL;
• leakage through plnholes in the FML; and
• leakage through holes 1n the FML.
2.2.3.2.2 Leakage due to Permeation through FML
- Permeameter Tests
Tests conducted at the University of Grenoble (France) by Giroud
from 1973 to 1978 and, then, by Gourc and Faure, using a permeameter
similar to those used to measure clay permeability ^Figure 2.2-1),
have shown that water permeates a FHL.
Results of these tests have been published by Giroud [1984a,
1984c]. In these publications, Darcy's equation has been used to
Interpret the test results and calculate equivalent hydraul;:
conductivities which vary significantly with the hydraulic head (and,
consequently, the hydraulic gradient).
2.2-11
-------
- The Concept of Coefficient of Migration
It is preferable to interpret the permeameter tests discussed
above using the following equation proposed by Giroud et a). [1987]:
Q/A = Ug/T (Equation 2.2-4)
where: 0 - flow rate due to permeation through the FML; A = surface
area of the considered FML; Q/A = flow rate per unit area; ug =
coefficient of migration of the FML; and T - FML thickness.
Recommended SI units are: Q (m'/s), A (m2), Q/A (m/s), ug (m'/s), and
T (m).
Values of the coefficient of migration for various FMLs are given
1n Table 2.2-3. Although there are not enough test results to draw a
firm conclusion, 1t appears that the coefficient of migration
Increases as the hydraulic head Increases to some maximum value, umax.
For hydraulic heads larger than approximately 10 meters {30 ft), u =
Vmax- Tne value of umax depends on the polymer used to make the FML.
The value of u 1s obviously zero for a hydraulic head equal to zero.
Ther :>re, the typical thape of the curve of the coefficient of
migration versus hydraulic head 1s given as shown in Figure 2.2-2.
It Is difficult to conduct water permeability tests on FMLs with a
head of water smaller than 5 m (16 ft) because the flow rates are too
small to ba accurately measured. The hydraulic heads that are
relevant to hazardous waste land disposal units are usually smaller
than 5 m (16 ft). Therefore it is useful to complement results from
the permeamcter tests cited above with results from water vapor
transmission tests which are typically conducted with a pressure on
the order of 1 to 10 kPa (0.15 to 1.5 psi), i.e., a hydraulic head on
the order of 0.1 m to 1 m (4 1n. to 40 in.).
- Water Vapor Transmission Tests
Water vapor transmission tests are typically performed on thin
membrane materials because the mechanism for fluid transport through
membranes 1s believed to be one of molecular diffusion through a
2.2-12
-------
nonporous membrane LHaxo et al., 1984]. With tliis mechanis-n,
transport through the membrane Involves three steps: (i) dissolution
of the fluid Into the membrane; (ii) diffusion of the fluid through
the membrane; and (iii) evaporation or dissolution of the fluid on the
downstream side of the membrane. According to Haxo et al. [1984], the
major driving force for the movement of a given fluid through a
membrane 1s its concentration gradient across the membrane. In the
case of water, the Important concentration gradient is suggested to be
the water vapor pressure, and moisture is thought to move through the
membrane by water vapor diffu: 3n. It is important to note that water
vapor diffusion decreases when the thickness of the membrane
increases, but is not dependent on the hydraulic head acting on the
membrane.
Haxo et al. [1984] have described a water vapor transmission test
(ASTM E96, Procedure BW) and have used it to measure water vapor
transmission rates for the range of FML materials given in Table
2.2-4. Values of water vapor transmission rates oLtained from other
sources are given in Table 2.2-5.
Knowing the water vapor transmission rate of a given FML obtained
In a given test, the quantity of vapor permeating through this Fl". can
be calculated using Fick's equation:
M/(At) - (WVT) (T./T) (Ap/ApJ (Equation 2.2-5)
where: M = mass of vapor migrating through the FML; A = FML surface
area; t = time (I.e., duration of the permeation); WVT = water vapor
transmission rate; T0 = FML thickness used in the water vapor
transmission test; T =- considered FML thickness; Ap = vapor pressure
difference between the two sides of the considered FML; and fip0 =
vapor pressure difference between the two sides of the FML used in the
water vapor transmission test. Recommended SI units are: M (kg), A
(m2), t (s), WVT (kg/(ml.s)), T0 and T (m), and Ap and Ap0 (N/m2).
(Note: 1 g/(m'.day) = 1.16 x 10"' kg/(ml.s)).
Vapor pressure is given by:
2.2-13
-------
p = ps "H (Equation 2.2-6)
where: ps = vapor pressure at s=.tjr;ted point; and H = relative
humidity.
Therefore, Equation 2.2-: can t~ written as follows:
M/(At) - (WVT) (Ta/T) {AH/AK,} (Equation 2.2-7)
where: AH - relative humidUj difference between the two sides of the
considered FML; AH. « relai-ve hurridity difference between the two
sides of the FML used In the water i-aprr transmission test; and other
notation as for Equation 2.2-?.
It should be pointed out that f-» use of Equations 2.2-5 and 2.2-7
should be restricted to presiares t~jt are not too different from the
pressures typically used t^ :-nduct the water vapor transmission test
(e.g., pressures on the order of 1.CC3 ta 10,000 Pa (0.15 to 1.5 psi),
I.e., hydraulic heads on tr* order of 0.1 to 1 m (4 to 40 in.) of
water).
According to Pick's e:<:at1on (Eruation 2.2-5), there is no
permeation through an FML if trie re'stiv* humidity is the same on both
sides of the FML. This is fi case, in particular, If there is water
on both sides, even if trier* is a rressure difference. This is in
disagreement with results c::24ned .-si-g a permeameter, which were
presented at the beginning :' Sect-*n 2.2.3.2.2. More research is
therefore needed on this subject.
- Relationships between Vario.i Exprersi—s of Flow Rate
In order to use water vapc- transmission test results to
complement permeameter test -esult;, U 1s necessary to establish
relationships between the va-;ous cref'icients used to express flow
rate.
An equivalent hydraulic :.:nduct" «i I? for FMLs can be obtained by
expressing flow rate throucn 2 FML L3--.g Carey's equation:
2.Z-14
-------
V = Q/A = k(J I (I <|lMllUII ,'.,' I!)
wheie: v = apparent velocity of the flow; 0 = flow rate; A = area
perpendicular to the flow; kg = equivalent hydraulic conductivity of
the FML; and i - hydraulic gradient.
By comparing Equation 2.2-4 with Equation 2.2-8, It appears that:
ug - kg h (Equation 2.2-9)
where: ug - coefficient of migration of the FML; kg = equivalent
hydraulic conductivity of the FML; and h - hydraulic head.
Recotmiended SI units are: Ug (m*/s), kg (m/s), and h (m).
By comparing Equation 2.2-8 (Darcy's equation) with Equation 2.2-5
(Pick's equation), 1t appears that:
WVT = p kg/(g T) = p kg h/T (Equation 2.2-10)
By combining Equations 2.2-9 and 2.2-10, it conies:
WVT = p ug/T (Equation 2.2-11)
where: kg » FML equivalent hydraulic conductivity; g = acceleration
of gravity; T » FML thickness; WVT = FML water vapor transmission
rate; p = pressure; p - liquid density; h » hydraulic head; and Ug =
coefficient of migration. The recommended SI units are: kg (m/s), g
(m/s2), T (m), WVT (kg/(m2.s)), p (Pa), p (kg/rn'), h (m), and
Ug (m2/s). A useful conversion factor for WVT is:
lg/(m*.24h) = 1.16 x 10"' kg/(m'.s)
Using Equation 2.2-11, the measured water vapor transmission (WVT)
values given 1n Tables 2.2-4 and 2.2-5 have been converted into values
of the coefficient of migration. It 1s interesting to see 1n Table
2.2-4 that series of tests on a given product (e.g., series of four
tests on PVC) with various thicknesses generally give consistent
values of the coefficient of migration.
2.2-15
-------
There are not enough values in Tables 2.2-3, 2.2-4 and 2.2-5 to
establish a complete table of values of coefficient of migration, Ug,
for FHLs. It is therefore necessary to draw curves such as those in
Figure 2.2-3 to make interpolations and extrapolations for small
values of the hydraulic head. Also, Tables 2.2-3, 2.2-4 and 2.2-5
contain discrepancies and apparently erratic results due to the
difficulty of the tests and the sometimes great differences between
FMLs of the same type. Therefore some averaging was necessary.
Values cf the coefficient of migration from Tables 2.2-3, 2.2-4 and
2.2-5 are summarized 1n Table 2.2-6. Figure 2.2-3 was established
using values of the coefficient of migration given in Table 2.2-6.
The large discrepancy between water vapor transmission rates
measured on PVC at 0.14 m head (Table 2.2-4) and 0.6 m head (Table
2.2-5) probably results from the fact that the PV tested at 0.14 m
head was a FML made of plasticized PVC and the PVC .-sted at 0.6 m was
pure PVC. Plasticized PVC is swelled by the plasticizers and tends to
oe more permeable than pure PVC (such as the stiff PVC used to make
bottles, which has a very low permeability).
- Leakage Rate Evaluation
From Figure 2.2-3, It 1s possible to establish Table 2.2-7 which
gives our best estimate of coefficient of migration values from the
analyzed data. From Table 2.2-7, 1t is possible, using Equation
2.2-4, to establish Table 2.2-8 which gives leakage rates due to
permeation through FMLs, assuming an FML thickness of 1 mm (40 mils).
- Migration of Chemicals
Many types of FMLs swell when placed in contact with chemicals.
As a result, the distance between polymeric chains increases and
permeability increases. Therefore, an FML can have a low permeability
to water and a high permeability to some chemicals. Data regarding
permeation of FMLs by chemicals can be found in [Haxo et al., 1984,
1986] and [Telles et al,, 1986].
2.2-16
-------
2.2.3.2.3 Leakage due to Pinholes In the FML
- Definition of Pinholes
According to Giroud [1984b]: pinholes should be distinguished from
holes and can be defined as openings having a dimension (such as
diameter) significantly smaller than the FML thickness. The primary
source of pinholes are manufacturing defects. Early manufacturing
techniques for FMLs often resulted in a significant number of FML
defects. However, manufacturing processes and polymer formulations
have advanced to a degree that pinholes are now relatively rare. Good
resin selection and manufacturing quality control should virtually
eliminate pinholes. This discussion of pinholes is included herein
for completeness. Leakage through FML pinholes will not, however, be
included in the final calculations of total leakage through FML top
1 iners.
- Basic Equation
For leakage calculation purposes, pinholes can be considered as
pipes and, therefore, according to Giroud [1984b], Poiseuille's
equation can be used:
Q = it p g h dV(128nT) (Equation 2.2-12)
where: Q = leakage rate; h = hydraulic head on fop of the FML; T =
thickness of the FML; d - pinhole diameter; p and n. = density and
dynamic viscosity of leachate; and g » acceleration of gravity.
Recommended SI units are: Q (m'/s), h (m), T (m), d (m), p (kg/mj), n
(kg/m.s), and g (m/s2).
The above equation is different from the equation used for
evaluating leakage through holes (see Equation 2.2-12, Section
2.2.3.2.4).
2.2-17
-------
- Calculations
E-jmticn 2.2-12 has been used to calculate leakage rates for Un-
typical pinhole diameters, 0.1 mm (O.C04 in.) and 0.3 nro (0.012 in.),
assuming a FML thickness of 1 mm (40 mils). Results are in Table
2.2-9. The hydraulic heads used in these calculations are as follows:
• 0.03 m (0.1 ft) which is an average head that can normally be
expected on the top liner of a landfill with a well designed
and constructed leachate collection and removal system;
• 0.3 m (1 ft) which is the maximum head considered in the design
of the leachate collection and removal system of a landfill;
and
•3m (10 ft) which is a typical head on the top liner of a
surface impoundment.
2.2.3.2.4 Leakage due to Holes In the FML
- Definition of Holes
Accordin to Giroud [1984b] holes should be distinguished from
pinholes and can be defined as openings having a dimension (e.g.,
diameter) about as large as, or larger larger than, FML thickness.
- Assumption Regarding Underlying Material
Leakage rates through FML holes are significantly affected by the
nature of the material underlying the FML. Two extreme cases can be
considered: a high-permeability material such as a granular or
synthetic drainage medium, and a low-permeability compacted soil such
as a clay layer placed under an FML to form a composite liner. The
case of a composite liner is addressed in Section 2.2.4.
In this section, the material underlying the FML is assumed to
have an infinite hydraulic conductivity. Tests by Brown et al. [no
date] have shown that underlying soils with a hydraulic conductivity
2.2-18
-------
higher than 10"' m/s (10"1 cm/s) do not affect free flow through a FML
defect, which justifies the assumption of an infinite hydiaulic
conductivity for drainage materials pporting a FML top liner.
- Assumption Regarding Overlying Material
Leakage rates through FML holes are affected by the material
overlying the FHL. The more permeable this material, the higher the
leakage rate will be. Therefore, in subsequent calculations, the
overlying material will conservatively be assumed to be infinitely
pervious.
Since It is possible that the hydraulic conductivity of the soil
overlying the FML has a marked influence on leakage rates through top
liners, especially in the case of surface Impoundments where hydraulic
heads are high, research should be done on this topic.
- Basic Equation
Assuming that the considered FML is located between two infinitely
pervious media, Bernouilli's equation for free flow through orifices
can be used to evaluate the leakage rate through a hole in the FML:
Q = C a 7 2gh (Equation 2.2-13)
where: Q = leakage rate; h = hydraulic head on top of the FML; a =
hole surface area; and g = acceleration of gravity. C is a
dimensionless coefficient, valid for any Newtonian fluid, and is
related to the shape of the edges of the aperture; for sharp edges,
C = 0.6. Recommended SI units are: Q (m'/s), h (m), a (m2), and g
(m/s2).
- Calculations
Equation 2.2-13 has been used to calculate leakage rates for two
typical holes:
2.2-19
-------
• a 2 mm (0.08 in.) diameter hole which is assumed to be a
"small" hole due to defective seaming (as discussed in Section
2.2.3.3.5} and might escape detection by construction quality
assurance; and
• a 11.3 mm (0.445 in.) diameter hole which is a "standard" 1 cm2
hole conservatively recommended for design, as indicated in
Section 2.2.3.3.6.
Results from these calculations are given in Table 2.2-9.
Hydraulic heads considered in these calculations are as follows:
• 0.03 m (0.1 ft) which is assumed to be the average head acting
on the top Uner of a landfill with a well designed and
constructed learhate collection and removal system.
• 0.3 m (1 ft) which is the maximum head considered in the design
of tiie leachate collection and removal system of a landfill.
• 3 m (10 ft) which 1s assumed to be the maximum head on the top
liner of a surface impoundment.
2.2.3.3 Frequency and_S1ze_of FML_Defects
2.2.3.3.1 Purpose
The purpose of this section is to evaluate the size and frequency
of defects which can occur In a FML. (Causes of defects were
discussed in Section 2.2.2.3.) This information is necessary for
making analytic calculations to evaluate leakage through top liners
(FML alone as well as composite liners). Although this section Is
devoted to all types of defects, it focuses primarily on seam defects
because forensic analyses have shown that leakage through FML liners
is often due to defective seams, and the most complete documentation
of FML defects is for seam defects.
2.2-20
-------
This section is organized as follows: first, data from
construction quality assurance and forensic analyses are reviewed,
then conclusions are drawn from these data.
2.2.3.3.2 Data from Construction Quality Assurance
- Smal1 Liquid Reservoir
This project, constructed in 1981, Is described In detail by
Giroud and Stone [1984], and Stone [1984]. Information regarding seam
defects can "-2 summarized as follows.
The double Hner system Includes two 2.5 mm (100 mil) thick HOPE
FMLs which were welded using an automated extrusion welder.
Ultrasonic testing, carried out as part of the quality control and
quality assurance program, showed that approximately 0.57. of the seam
length was defective. The detected defects were repaired and the
reservoir was filled with water. Leakage occuned and an Inspection
showed that leakage was taking place through approximately 0.015'/. of
the seam length. The ratio 0.5/0.015 shows that, in this project,
intensive quality assurance divided the length of defects by
approximately 30.
This project is particularly interesting because it provides an
evaluation of the benefits from construction quality assurance.
- Large Landfill with Single Liner
Kastman [1984] indicates that in a carefully monitored landfill
liner installation done in 1983, approximately one defect every 15 m
(50 ft) of seam was detected and repaired, as part of the quality
assurance process. The liner was a 1 mm (40 mil) thick HOPE FML and
seaming was achieved with a fillet extrusion weld done using a hand
welder.
2.2-21
-------
- Large Landfill with Double Liner
Giroud and Fluet [1986] report the result of an analysis conducted
on the basis of data collected during the quality assurance process of
liner installation in a large landfill, lined in 1985 with an HOPE
FML. The surface area of the liner is approximately 35 000 m'
(350,000 ft') and seam length is approximately 5000 m (16,000 ft).
During the quality assurance process, an average of approximately one
seam defect every 9 m (30 ft) of seam length was discovered and
repaired.
- Large Landfill with Single Liner
This case history presents the results of an analysis conducted on
the basis of data available in GeoServices files. The data were
collected during the installation of the lining system in a large
landfill, in 1987, as part of the quality control provided by the FML
installer and quality assurance provided by an independent firm. The
surface area of the liner is approximately 53 000 m' (570,000 ft2) and
seam length is approximately 8000 m (26,000 ft). The liner was a 1.5
mm (60 mil) thick HOPE FML. Half of the seam length was welded using
a hand welder which made fillet extrusion welds; the other half was
welded using an automated flat welder. An average of approximately
one seam defect every 11.5 m (38 ft) of seam length was discovered by
the FML installer and the independent quality assurance firm. All
these defects were repaired. Seam inspection was performed first by
the installer, and then by the independent firm after the installer
had completed his inspection. The installer detected approximately
one seam defect every 17 m (56 ft) of seam length. The independent
firm detected approximately one seam defect every 35 m (115 ft) of
seam length.
This project is interesting because it provides an evaluation of
tt/e benefits from construction quality assurance. The independent
firm discovered additional seam defects, after the installer had
completed his quality control inspection. The defects discovered by
the independent firm totaled one third of the total seam defects. The
benefits of quality assurance are probably greater than that: it is
2.2-22
-------
probable that, without the continuous presence at the site of the
independent quality assurance firm, the FML installer would have found
fewer defects than he did as part of his quality control effort.
2.2.3.3.3 Data from Forensic Analyses
- Smal1 Indoor Tank
A power generating station icquired a small acid holding tank,
which was constructed of concrete and lined in 1985 with a high
densi polyethylene (HOPE) FML which required approximately 100 m
(300 ft) of field seaming. The seams were fillet welds done with a
hand welder. The design and installation included no third party
quality assurance, but careful quality control of seaming was provided
by the installer, using visual inspection and vacuum box.
Upon completion of the liner installation, the tank was filled
with water to check for leaks. The liner did leak, so the tank was
emptied, repairs were made and the tank was filled again. This cycle
was repeated several times, with leaks found on every filling. Leaks
were found at 15 deferent locations, i.e. an average of one leak per
7 m (23 ft) of seam. This incidence of seani defects is probably
significantly larger than the incidence which would be experienced in
FML installations in land disposal units, given the complex geometry
and difficult welding conditions in the holding tank compared to a
land disposal unit. However, the complex geometry of this tank is
probably representative of the difficulties encountered in land
disposal units at the connections between FMLs and appurtenances such
as sumps.
- Large Surface Impoundment
The following case history is reported by Giroud and Fluet [1986].
A large reservoir, lined with a single reinforced chlorosulfonated
polyethylene (CSPE-R) FML, had been constructed to contain phosphoric
2.2-23
-------
DHL- >LMI alter the lii'^L Illllm], UK? icscivoli '.u.Mf.'iily ri.plicil.
The analysis of the failure Indicated that phosphoric acid, leaking
through several defective seams, attacked the ground, creating
cavities. The largest cavity was one meter (three feet) in diameter
and half a meter (iO inches) deep. Under the pressure of the
impounded liquid, the FHL spanning this largest cavity burst,
releasing all of the Impounded phosphoric acid Into the ground.
Quality assurance during installation had consisted of only two
one-day visits by an engineer who specialized in roofing membranes.
Therefore, it is not surprising that defective seams were not detected
prior to filling.
During the forensic analysis, visual observation showed that
approximately 0.1% of the seam length was defective. It 1s probable
that a higher percentage would have been obtained if a vacuum box had
been used instead of the visual inspection.
2,2.3.3.4 Conclusions on Frequency of Defects
- Consistency of the Observations
Sections 2.2.2.2 and 2.2.2.3 present data related to frequency of
seam defects. Some of these data are expressed as an average seam
length exhibiting one defect (e.g., one defect per 7 m (23 ft) of
seam), while other data are expressed as percentage of defective seam
length (e.g., 0.57. of the total seam length was defective).
If an average length of seam defect (prior to quality assurance)
of 10 mm (0.4 in.) is considered, a percentage of defective seam
length of 0.1% is equivalent to one defect every 10 m (30 ft).
Therefore, the observations from the previous case studies appear to
be consistent.
2.2-24
-------
- Conclusion Regarding Frequency of Seam Defects
It is not possible to draw general conclusions from only six
cases. However, since the observations made tn these six cases were
consistent H is possible to diaw the following tentative conclusions:
• An average of one defect per 10 m (30 ft) of seam can be
expected without quality assurance.
• An average of one defect per 300 m (1,000 ft) of seam can be
expected with reasonably good Installation, adequate quality
assurance, and repair of noted defects. Even better results
(possibly up to one defect per 1000 m (3000 ft)) may be
possible with the very best Installation procedures and
intensive quality assurance. (Quality assurance followed by
repair drastically decreases the number of seam defects but may
not totally eliminate them.)
The average of one seam defect per 10 m (30 ft) without or before
quality assurance will probably decrease in the future as a result of
the increa-ing use of nev, automated methods of seaming which are now
available. However, the number of seam defects after quality
assurance may not decrease significantly because, In the present state
of practice for construction quality assurance, great emphasis 1s put
on finding seam defects and repairing them. Nonetheless, the better
seaming methods that are now available are highly beneficial for at
least the following reasons: (1) less seam repair is required during
installation; (11) frequency of destructive seam testing may be
decreased; (111) quality assurance efforts may shift toward other
areas where Improvement is sorely needed such as connections of FMLs
with appurtenances and placement of drainage materials (which is
essential for the functioning of leak detection systems); and (iv)
stronger seams that are less likely to fail when subjected to
stresses.
As a result of the above discussion, a frequency of one defect per
300 m (1,000 ft) of seam will be used as a conservative working
assumption. If FML panels 6 to 10 m (20 to 30 ft) wide are used, one
2.2-25
-------
defect per 3r~' m (1,000 ft) of seam is equivalent to 3 to 5 seam
defects per hectare (1 to 2 seam defects per acre) of installed FML.
As soon as possible, these tentative conclusions must be
supplanted and modified as required by conclusions established on a
broader base of well documented case histories. In the meantime (and
in the absence of better data), in arbitrary frequency of one defect
calculations for estimating leakage
on drainage layers. Such frequency
per 4000 m2 (acre) will be used
rates in order to size leak dete
is assumed to Include all types c defects, not only seam defects.
2.2.3.3.5 Estimation of Size of Defects
The seam defect documentation reported above addressed primarily
the frequency of seam defects. Extensive documentation of defect size
does not exist. On the basis of interviews with quality assurance
personnel it appears that the maximum size of defects which may still
exist after Intensive quality assurance is equivalent to hole
diameters en the order of 1 to 3 mm (0.04 to 0.12 in.) for seam
defects and possibly up to 5 mm (0.2 in.) for special areas such as
connections of FML with appurtenances. This is confirmed by the case
history presented in Section 2.5.3.
There are also defects that cannot be observed by the quality
assurance personnel, such as: (i) puncture of the FML during
Installation of the protective earth cover; and (ii) puncture of the
FML as a result of stresses due to the weight of waste or traffic
related to the operation of the hazardous waste management unit.
Therefore, for design purposes it may be approoriate and conservative
for subsequent calculations to consider a hole larger than the
expected size of defects at the end of Ff-'_ installation (wnich were
estimated above as 5 mm (0.2 In.) maximum in diameter). However, for
the establishment of the best demonstrated available technology (BOAT)
for liner Installations, a smaller hole size (in the range of 1 to 2
(Tm (0.04 to 0.08 1n.)) Is probably more appropriate.
2.2-26
-------
2.2.3.3.6 Standard Hole Size and Frequency
For the consistency of calculations and discussions supporting the
proposed Liner/Leak Detection Rule, 1t 1s recommended that a standard
hole size and frequency be selected. The same standard hole size and
frequency will also be useful as guidance for designers of leak
detection systems.
As a result of the above discussions, a "standard" hole area of 1
cm1 (10~4 m2 or 0.16 1n2.) has conservatively been selected, and, on
the basis of the discussion presented in Section 2.2.3.3.4, a
frequency of one standard hole per 4000 m2 (acre) Is considered. The
standard hole area and frequency are used in this background document
for calculations done to evaluate the leakage rates used to establish
the required drainage capabilities of the LDCRS.
It should be kept in mind that the standard hole size and
frequency have been selected with the assumption that good quality
assurance monitoring will be performed. Also, the standard hole size
and frequency does not take into account cases where design flaws or
poor construction practices would lead to many seam defects or a large
tear in the FML.
Lastly, as previously noted, the standard ho'e size is for design
and calculation purposes and not for the purpose of defining BOAT.
Selection of a standard hole size for design should include a margin
of safety. Selection of a hole size for BOAT should be based on the
best actual installation and construction quality assurance practice.
On this basis, the small hole size i probably m-re representative of
BOAT than the standard hole size.
2.2.3.4 Conclusions on_Leakage_through FHL_Top_L1ners
2.2.3.4.1 Summary
In Section 2.2.3, the following was done. The results of
permeameter tests and water vapor transmission tests were used to
2.2-27
IMUflttn
-------
evaluate permeation through FMLs. Equations to evaluate leakage rate
through plnholes and holes were presented. The following
reccrnrendatlon regarding frequency and size of holes to be considered
in calculations was made on the basis of field experience: one 1 cm1
(0.16 in2.) hole per 4000 m* (acre).
2.2.3.4.2 Leakage Rates
By combining Tables 2.2-8 and 2.2-9, it Is possible to establish
Table 2.2-10 which gives orders of magnitude of typical leakage rates
which can be expected when a FML Is used alone as a top Hner.
It appears that the leakage rate due to only one hole per 4000 m1
(acre) Is very large, while the leakage rate due to permeation and
plnholes 1s small. It must be remembered that the leakage rates given
in Table 2.2-10 are related to a top liner comprised of a FML placed
directly on the leak detection, collection, and removal system. As
indicated 1n Section 2.2.3.2.4, the equation used is valid 1f the
hydraulic conductivity of the leak detection, collection, and removal
f " system is larger than 10~' m/s (0.1 cm/s), which should always be the
case.
2.2.4 Leakage Through Composite Liners
-, 2.2.4.1 Introductjon
2.2.4.1.1 Purpose of the Section
This section discusses leakage through composite liners due to a
hole in the FML. The purpose of this discussion is to draw practical
conclusions regarding the evaluation of leakage rate through composite
top liners. These conclusions will be used in the calculations made
in subsequent sections to compare top liners comprised of a FML alone
with composite top liners, and to evaluate the rate of leakage that
Impinges into the leak detection, collection, and removal system.
Composite top liners are not now required by EPA. However, some
owners and operators are opung for composite top liners because, as
/ shown in this section, they reduce leakage through the top liner In
J
2.2-28
-------
comparison to the leakage through a FML alone. However, as will be
shown in Section 2.10, the use of a top composite liner has one
drawback from the standpoint of the proposed Liner/Leak Detection
Rule: consolidation of the compacted soil component of the top
composite liner will result in a significant flew of "consolidation
water" into the LDCRS.
2.2.4. .2 Leakage Mechanisms
A composite liner is comprised of a FML (which is the upper
component of the composite liner) and a low-permeability compacted
soil layer (which is the lower component of the composite liner). If
there is leakage through a composite liner, the leachate first
migrates through the FML. then may travel laterally in the space, if
any, between the FML and the low-permeability compacted soil, and,
finally, migrates through the low permeability soil.
There are three mechanisms by which leakage can migrate through a
FML:
• permeation through the FML (I.e., flow through a FML that has
no defects such as holes or pinholes);
• flow through pinholes in the FML (pinholes are very small
openings that have a diameter that is significantly smaller
than the thickness of the FML); and
• flow through holes in the FML (holes are openings that have a
diameter larger than the thickness of the FML).
Leakage rate due to permeation through the FML should not be
significantly affected by the presence of the low-permeability
compacted soil layer under the FML because even a soil with a very low
permeability 1s still very permeable as compared to a FML without
holes and pinholes. The case of permeation through n FML without
holes or pinholes was discussed in Section 2.2.3.2.2.
2.2-29
-------
Section 2.2.4 s dev"'-3d to leakage through composite liners iue
to a defect such as pinhole or hole in the n-IL. However, pinholes are
only briefly discussed since pinholes are rare and obviously cv^se
much less leakage than a hole.
2.2.4.1.3 Organization of the Section
The remaining three subsections are devoted to:
• analytical studies;
• laboratory models; and
• practical conclusions regarding the evaluation of leakage rate
through composite liners.
2.2.4.2 Analytical Studies
2.2.4.2.1 introduction
As indicated in Section 2.2.4.1.2, the leachate that has passed
through the FHL can flow laterally to a certain extent between the FML
and the low-permeability compacted soil, before it migrates through
the low permeability soil. This is possible if there is a space
between the FML and the low permeability soil.
Two types of analytical studies can be found in the literature:
• anc ytical studies assuming that there is perfect contact
between the FML and the low-permeability compacted soil, and,
consequently, that the leachate does not flow laterally between
the FML and the low-permeability compacted soil; and
• analytical studies assuming that leachate flows laterally
between the FML and the low-permeability compacted soil before
it migrates through the low permeability soil.
2.2-30
-------
2.2.4.2.2 Analyses Assuming Perfect Contact
- Assumptions
Faure [1979] has made an extensive study of the leakage rate
through a composite liner due to a hole In the FML, assuming perfect
contact between the FML and the underlying low permeability soil.
First, Faure considered two simple two-dimensional cases:
• flow net established by considering that the entire soil layer
Is saturated (Figure 2.2-4 a); and
• radial flow (Figure 2.2-4 b) which leads to a convenient close
form solution for the leakage rate (the radial flow was thought
to be a reasonable assumption for thick soil layers, but in
fact Is not, as shown by Faure (see Figure 2.2-7)).
These two types of flow lead to absurd results (such as flow rate
increasing whan soil thickness increases). However, those cases are
useful because Faure showed that they provide upper boundaries for the
actual flow rate through the composite liner when the FML and the
underlying soil are in perfect contact. Also the leakage rate in the
case of the radial flow is expressed by a close form solution for the
three-dimensional case (circular hole), which provides a convenient
upper boundary for the three-dimensional case. This 1s very useful
because the three-dimensional case is very difficult to analyze and
this upper boundary is one of the few theoretical data available for
the three-dimensional case.
A lower boundary of the leakage rate is obtained by assuming that
the flow Is vertical (Figure 2.2-4 c).
The actual flow if the FML and the low-permeability compacted soil
are in perfect contact is shown 1n Figure 2.2-4 d. This has been
demonstrated in the two-dimensional case by:
2.2-31
-------
• Faure [1979] who used numerical methods; and
• Sherard [1985] who traced flow nets by trial and error.
Both Faure and Sherard have shown that, in a two-dimensional flow:
• there is horizontal flow in the soil along a portion of the
interface (although there is no flow between the FML and the
soil because there is no space between the FML and the soil
when perfect contact is assumed); and
• there is a phreatic surface beyond which the soil is not
saturated.
These qualitative characteristics of the flow are certainly also
applicable to the three-dimensional case (circular hole). Typical
flow nets for the two-dimensional case are given in Figure 2.2-5 and a
chart giving the location of the phreatic surface in the two-
dimensional case is presented in Figure 2.2-6.
- Leakage Rates for the Two-dimensional Case
Leakage rates obtained with the various assumptions discussed
?bove are given in Figure 2.2-7 adapted from Faure. This figure
shows that:
• absurd results are obtained with the upper boundaries, cases
(a) and (b), when the low-permeability compacted soil
thickness, H, is large; and
• case (c) is a very low lower boundary when the low-permeability
compacted soil thickness, H, is large.
A chart giving the actual leakage rate (i.e., the leakage rate
obtained in case d) when the FML and the undei lying soil are in
perfect contact has Men prepared by Faura [1979, 1984] for the two-
dimensional case (Figure 2.2-8). The results given by Sherard [1985]
2.2-32
-------
for a limited number of cases are consistent with Faure's. Faure's
chart (figure 2.2-8) Is used with the following equation:
0/B = C kc (II + h) (Equation 2.2-14)
where: Q - leakage rate; B - length of the slot in the direction
perpendicular to the figure; Q/B = leakage rate per unit length; C =
dimensionless coefficient given by the chart; kc « hydraulic
conductivity of the low-permeability compacted soil underlying the
FML; II • thickness of the low-permeability compacted soil; and h -
hydraulic head on top of the FML.
The equation for the two-dimensional radial flow (case (b) in
Figures 2.2-4 and 2.2-7) which gives an upper boundary for the actual
leakage rate Is obtained by integrating Darcy's equation for a
circular domain:
Q/B = H kc (h + H)/Log (2H/b) (Equation 2.2-15)
where: Q - leakage rate; 3 = length of the slot in the direction
perpendicular to the figure; Q/B » leakage rate per unit length; kc =
hydraulic conductivity of the low-permeability compacted soil; h =
head on top of the FML; b -> width of the slot; and H » thickness of
the low permeability soil underlying the FML. Recommended SI units
are : Q (m'/s); Q/B (m'/s/m, I.e., m'/s); kc (m/s); h (m); b (m); and
H (m).
The equation for the vertical flow (case (c) in Figures 2.2-4 and
2.2-7), which gives a lower boundary for the flow rate, is obtained by
writing Darcy's equation for a rectangular domain:
Q/B = kc b (h + H)/H (Equation 2.2-16)
where '.he notation is the same as above.
This lower boundary gives a good approximation of the actual
leakage rate if the rdtio between the Width bf the Fill hole' and the
thickness of the low permeability soil is large, which is rare.
2.2-33
-------
The upper boundary provided by the radial flcrf (Equation 2.2-15)
is excessively high In many cases and increases when H/h is large and
increases, as shown in Figure 2.2-5. Sines the leakage rate cannot
increase if the thickness of th
-------
Q = n kc (n <- H) d/(l - 0.5d/H) (Equation 2.2-18)
where: Q = leakage rate; V;,- • hydraulic conductivity of the Icw-
penneabi1ity compacted soil; h - hydraulic head on top of the FHL; d =
diameter of the circular hole; and H = thickness of the low
permeability soil. ReccmniencJod SI units are: Q (m'/s), kc (m/s), h
(m), d (m), and H (m).
The equation related to the vertical flow (similar to the two-
dimensional case (c) 1n Figure 2.2-4), which gives a lower boundary
for the actual leakage rate, is obtained by writing Darcy's equation
for a 'cylindrical domain:
Q = kc a (h f H)/H (Equation 2.2-19)
where: a = surface Area of the hole in the FML (a = u dJ/4 if the
hole is circular); and other notation as above.
As discussed for . *ie two-dimensional case, Equation 2.2-18 can be
rewritten as fol lows:
Q = IT k n d/(l - 0.5d/H) (Equation 2.2-20)
It is possible that this equation gives a lower boundary of the
actual leakage rate when d/H \'t small (like Equation 2.2-17 for the
two-dimensional case). It is interesting to note that Equation 2.2-20
;ends toward a very simple limit when d/H tends toward zeio:
Q - n kc h d (Equation 2.2-21)
where: Q = leakage rate; kc - hydraulic conductivity of the low-
permeability compacted soil underlying the FML; h = hydraulic head on
top of the FML; and d = diameter of the circular hole in the FML.
Due to the lack of any better solution, Equation 2.2-21 will be
used as an approximation for the actual leakage rate.
2.2-35
-------
Another approach for evaluating leakage rate in the three-
dimensional case is to use the chirt established by Faure for t,';2 two-
dimensional case (Figure 2.2-8) and modify Equation 2.2-14 by
replacing the length B of the slot by the perimeter nd of the circular
hole (and not half the perimeter, nor the diameter of the hole as one
may be tempted to do):
Q = IT C kc (H + h) d (Equation 2.2-22)
where: 0 - leakage rate; C = dimensionless coefficient given by
Faure's chart (Figure 2.2-8); kc = hydraulic conductivity of the low-
permeability compacted soil; H = thickness of the low permeability
soil layer; h - hydraulic head on top of the FML; and d = hole
diameter.
2.2.4.2.3 Analyses Assuming Flow between FML and Soil
- Introduction
Analytical studies have been conducted by Fukuoka [1986] and Brown
et al. [no date]:
• Fukuoka considered the case where there is a geotextile
(without a hole) between the FML (with a hole) and the soil.
The liquid leaking through the FML hole first flows
horizontally in the geotextile, then vertically through the
soil layer.
• Brown et al. considered that there is a space between the FML
and the soil layer. The liquid leaking through the FML hole
first flows horizontally in the space, then vertically through
the soil layer.
- Flow in Ceotextile between FML and Soil
Fukuoka [1986] considered the case of a geotex' le between the FML
and the low-permeability soil, and assumed that the leachate flows
2.2-36
-------
horizontally and radially within the geotextlle before it flows
vertically in the soil underlying the geotextile. Although
ge^U'xtiles are not used in composite liners, the analysis made by
Fukuoka is pertinent to :omposite liners because equations similar to
those derived by Fukuoka can be used for flow in the narrow space (if
it exists) between a FML and soil.
The following differential equation has been established by
Fukuoka [1986]:
(1/r) (dh/dr) + d'h/dr* » h kc/(H8) (Equation 2.2-23)
where: r = radius from center of hole; h = hydraulic head at radius r
in the geotextile; kc - hydraulic conductivity of the low-permeability
compacted soil underlying the geotextile; H - thickness of the soil
layer; and 6 » hydraulic transmissivity of the geotextile.
The only assumption is that the flow in soil Is vertical. No
assumption is made regarding the hydraulic head in the geotextile.
This head decreases from a maximum value at the FML hole, to zero at
the periphery of the wetted portion of the geotextile. Consequently,
flow through soil is faster at the center of the wetted area than at
the periphery. Solving the above equation would give the radius of
the wetted area and would allow the leakage rate to be determined.
Fukuoka did not solve the equation, but the solution proposed by Brown
et al. for Equation 2.2-29, which is similar, can be adapted to
Equation 2.2-23 If the thickness of the geotextile (and, therefore,
its transmissivity) is assumed not to vary with the radius r (while,
In fact it varies since the effective stress on the geotextile varies
with the radius r).
Equation 2.2-23 was established by combining Darcy's vertical flow
in the soil with Darcy's radial flow in the geotextile, Qr, which 1s
governed by the classical differential equation:
Qr - - 2 n r kp s dh/dr (Equation 2.2-24)
2.2-37
-------
where: kp - hydraulic conductivity of the geotextile in the direction
of its plane; s « thickness of the geotextile (I.e., spacing between
FML add soil); and other notation as above.
This equation can also be written:
Qr - - 2 TT r 8 dh/dr (Equation 2.2-25)
where: 6 » hydraulic transmissivity of the geotextile.
- Flow in Space between FML and Soil
This study was made by Brown et al. principally to extrapolate
results obtained with their small diameter model to real situations
where the flow may laterally extend over a large area.
The approach used by Brown et al. is similar to Fukuoka's. They
combine vertical Darcy's flow in the low-permeability compacted soil
with radial flow in the space between the FML and the underlying soil.
Brown et al. integrated Newton's equation for viscous fluids in a
circular domain and demonstrated that the radial flow is governed by:
Qr „ _ [TI r s1 p g/(6 n)] (dh/dr) (Equation 2.2-26)
where: r = radius from center of hole; s = spacing between FML and
low permeability soil; p = density of leachate; o = acceleration of
gravity; n = viscosity of leachate; and h = hydraulic head at radius r
in the space between FML and soil.
By comparing Equations 2.2-25 and 2.2-26, it appears that a space
s between the FML and the underlying soil is -..juivalent to a hydraulic
transmissivity 9 given by:
6 = p g s'/(12 n) (Equation 2.2-27)
For example, using the density (p = 1000 kg/m') and the viscosity
(n = 10"' kg/ms) of water, this equation shows that a spacing s = 1 mm
is equivalent to a hydraulic transmissivity of 8 x 10~2 m2/s, and a
2.2-38
-------
spacing s = 0.1 mn is equivalent to a hydraulic transmissivity of 8 x
1[T5 m2/s. These transmissivity values are consistent with
trdrr.ni ^s i vi t ies of synthetic drainage layers.
The differential equation obtainea oy Grown et al. is:
d(r dh/dr)/d r - [12 n kcr/(p 9 S')J 0 + h/H) (Equation 2.2-28)
which can be written:
(l/i; (dh/dr) + d'h/dr1 =
[12n kc/(p g s1)] (1 + h/H) (Equation 2.2-29)
Combining Equation 2.2-27 and 2.2-29, it appears that Equation
2.2-29 [Drown et al.] is identical to Equation 2.2-23 [Fukuoka, 1986]
except for the last term, h/H for Fukuoka and (1 + h/H) for Brown et
al. Brown et al. solved this differential function using Bessel
functions to interpret results from their laboratory model (see
Section 2.2.4.3.2). However, the charts they proposed for field
conditions were established with a simplifying assumption: the
hydraulic gradient for the vertical flow in soil is one. In other
words, they rssume that the hydraulic head en top of the low-
p
-------
s^vy^^^"^^^^^^^^
£?k<«~>--^-' ••"' ••-?*;,••••- .; >"-» - '---
which gives the following relationship [Brown et at.]:
h + H = [3 n kc dV(4 p g s1)]
[2 (2R/d)' Log (2R/d) - (2R/d)' + 1] (Equation 2.2-31)
where: h = hydraulic head on top of the FHL; H « thickness of the
low-permeability soil layer; n = viscosity of the leachate; kc =
hydraulic conductivity of the low-permeability compacted soil; d •*
diameter of tha hole in the FML; p - density of the leachate; g =
acceleration of gravity; s - spacing between the FML and the low-
permeability compacted soil; and R = radius of the wetted area.
Equation 2.2-31 gives the radius of the wet.ed area if the spacing
s between the FML and the low-permeability compacted soil is known.
Guidance regarding selection of spacing values can be obtained through
backcalculation of Brown et al.'s test results (see Section
2.2.4.3.2).
When the radius R of the wetted area is known, the leakage rate
can be determined by using the following equation which derives from
Darcy's equation with the assumption that the hydraulic gradient is
one in the low-permeability compacted soil:
Q = TT R' kc (Equation 2.2-32)
The above equations were used by Brown et al. to establish charts
giving the leakage rate and the radius of the wetted area (Figures
2.2-9 through 2.2-12). To summarize results presented in these charts
and extrapolate or interpolate them, we propose the following
equations:
Q - 0.7 a"" kc°"" h (Equation 2.2-33)
R = 0.5 a""" kc~°"" h°-B (Equation 2.2-34)
These empirical equations are only valid with the units indicated:
Q = leakage rate (m'/s); a = surface area of FML hole (m2); kc =
hydraulic conductivity of low-permeability compacted soil (m/s); h =
? 2-40
-------
hydraulic head on top of FML (m) ; and R - radius of we...ed area
between FML and soil (m).
2.2.4.3 Laboratory_Models
2.2.4.3,1 Introduction
Tests to evaluate leakage through composite liners due to a hole
in the FML were conducted by Fukuoka [1985, 1986] and Brown et al. [no
date]. It is Important to recognize that neither the Brown et al.
tests or Fukuoka tests were developed to model the field condition of
leakage through composite liners. The Brown et al. tests wer
preliminary and conceptual in nature. The Fukuoka tests do not even
directly relate to field conditions existing at landfills and surface
impoundments. However, both sets of tests (and in particular the
Brown et al. tests) can be used to develop an understanding of the
mechanics of flow through composite liners and to relate design
equations to field conditions.
In both cases, tests were conducted with a FML having a circular
hole, and various hole diameters were used in both testing programs.
Additional tests by Brown et al. included FHL flaws that are not
circular such as slits or seam defects. The tests were intended to be
full-scale models of the reality since hole size, rML thickness, and
(approximately) soil"layer thickness were similar to what they are in
the field. However, the permeameters used had a limited diameter
(e.g., 0.6 m for Brown et al., and 1.5 m for Fukuoka) and the
extension of lateral flow between the FHL and soil was limited-fey the
walls of the permeameter.
In he tests conducted by Brown et al., the FML was always covered
by 0.15 m (6 in.) of gravel to ensure contact between FML and soil,
and, in some tests, an additional load up to 160 kPa (3340 psf)
(equivalent to 10 m of soil) was applied to evaluate the effect of
overburden pressure. In many of the tests conducted by Fukuoka, the
FML was not covered, and the only load applied on the FML was the
water pressure.
2.2-41
-------
Water heads in Brown et al. tests were up to 1 m. while in FiAutxi
tests, they were up to 40 m. Tests by Brown et al. were conducted f:*
landfill applications while Fukucka was working on the design of i
large dam and reservoir.
Fukuoka used only a PVC FML, while Brown et al. considered i
variety of FMLs: HOPE, PVC, CSPE, and EPDH, with various thicknesses.
Tests by Fukuoka as well as tests by Brown et al. showed that
there is flow between the FML and the soil. Some of the test?
conducted by Fukuoka and by Brown et al. included a geotextile betwee^
the FML and the soil. With a geotextile, flow between the liners
would be expected and the liners do not represent a true composite
liner. However, from these tests with geotextiles, some understands:
of the effect of an imperfect FML to soil contact can be obtained.
2.2.4.3.2 Review of Tests by Brown et al.
These tests are presented in a report by Brown et al. [no date].
- Description of the Tests
Tests were conducted in a C.6 m (24 in.) diameter permeatr.eter.
Hole diameters ranged between 0.8 mm (1/32 in.) and 13 mm '1/2 In.).
and non-circular holes such as slits and seam defects were considered.
The FMLs were: HOPE (0.8 am to 2.5 mm) (30 to 100 mils); PVC (O.r
to 0.8 mm) (20 to 30 mils); C^PE (0.9 to 1.15 mm) (36 to 45 mils); ar:
EPDM (0.8 mm) (30 mils).
In some tes^, geotextiles were included between the FML and tr =
soil. The geott/.ciles were needlepunched nonwovens with a mass pe'
unit area of 250 to 350 g/m2 and a thickness (under no load) on tti
order of 2.5 to 4 mm.
The soils used were a silty sand (k = 2 x 10"' m/s), and a clayey
silt (k = 2 x 10"' m/s).
2.2-42
-------
- Approach
The diameter of the permeametcr used by Brown et a), was small
(0.6 m) and lateral flow could not extend beyond a radius of 0.3 m as
it would have 1n most cases without the limitation imposed by the
permeameter walls. Therefore, the calculations presented in Section
2.2.4.2.3 were used to backcalculate the value of the spacing between
the FML and soil from the test results. The value of the spacing thus
obtained can then be used In similar equations to determine the radius
of the wetted area and, therefore, the leakage rate in actual
situations where lateral expansion of the flow is not impeded by
permeameter walls. The backcalculated spacing values are as follows:
0.02 mm for clayey silt regardless of FML
0.08 mm for silty sand and flexible FML (PVC)
0.15 mm for silty sand and stiff FML (HOPE)
Spacing between the FML and the soil, and, therefore, the leakage
rate, appears to increase if the FML stiffness increases (at least in
the case of the more permeable soil). It also appears that spacing
increases 1f the soil is coarse, which is illustrated by:
0.02 mm = d1() of clayey silt
0.08 mm = dls of silty sand
The above spacing values are related to the case of a FML with
15 cm of gravel overburden. This is an unrealistically low overburden
pressure in relation to field conditions. This fact, coupled with the
fact that Brown et al.'s tests were short term, implies that the
spacing values cited above are somewhat conservative.
Following is a review of the influence of various parameters on
test results.
2.2-43
-------
- f.ff«?v.t cf Overburden Pressures
When a compressive stress of 160 kPa (equivalent to 10 m of rMl)
Is applied on a HOPE FML 0.75 mm (30 mil) thick placed on a soil with
a hydraulic conductivity of 2 x 10"' m/s, the flow rate through an
FML hole i. divided by 200 and the backcalculated theoretical spacing
betwepn FML and soil is divided by 10 (thers are no results for the
soil with a hydraulic conductivity of 2 x 10"' m/s).
- Effect of Flaw Shape
Erratic results were obtained with slits and seam defects on the
soi1 with kc = 2 x 10"' m/s:
• Some tests showed that a 50 mm slit or seam defect is often
equivalent to a 0.5 to 1 mm diameter circular hole (however
other tests showed that a 50 mm seam defect can be equivalent
to a 75 imi diameter hole).
• Tests showed that a 150 rnrn slit or seam is often equivalent to
a 75 mm diameter circular hole (which is very different from
the 0.5 to 1 mm diameter circular hole indicated above as
equivalent to a 50 mm seam defect).
It was difficult to compare slits, seams and circular holes with
the 2 x 10""* m/s soil because for that soil there is wore lateral flow
and permcameter walls disturbed the flow.
- Conclusions from Brown et al.'s Tests
In order to ext,ipolate to field conditions, Brown et al. make the
following recomnenoatlons regarding the values of the spacing between
FML and soil to be used in the equations presented in Section
2.2.4.2.3 to evaluate leakage rate and radius of wetted area in actual
field conditions where lateral extension of flow is not impeded by
wall parmeameter:
2.2-44
-------
soil hydraulic
ronductivity, kc
(m/s)
10"'
1(T'
10"'
10''
TML-soil
spacing, s
{mm)
0.15
0.08
0.04
0.02
These values are the upper boundary of (or even larger than) the
backcalculated spacing values previously given in the discussion of
the approach. Also, these spacing values are for the case when there
1s little or no overburden (e.g., 15 cm of gravel), and they are
expected to be larger than 1n the case when there Is a large
overburden. Therefore, for these two reasons, leakage rates
calculated by Brown et al. are likely to be conservative. It is
clear, however, that the results of the Brown et al. study Indicate a
significant benefit of a composite Uner design consisting of a FML
upper component and compacted soil lower component.
2.2.'.3.3 Review of Tests by Fukuoka
These tests are described 1n [Fukuoka, 1985; and Fukuoka, 1986].
They were conducted for the design of the lining system for a dam and
a reservoir with - maximum water head of 40 m (130 ft). Although
these conditions are not representative of hazardous land disposal
units, the study conducted by Fukuoka, when combined with the findings
of Brown et al., provide a good understanding of the mechanisms
governing leakage through composite liners.
- Description of the Jest
All tests discussed below were conducted with the following
equipment, conditions, and materials: permeameter diameter is 1.5 m (5
ft); water pressure 1s 200 or 400 kPa (4,000 or 8,000 psf); soil
permeability 1s on the order of 10"7 to 10~* m/s (10~* to 10~" cm/s);
2.2-45
-------
soil ckness Is 0.45 m (1.5 ft) wh°n no soil cover is placed on the
FHL am 0.225 m when a 0.225 m (0.75 ft) thick soil cover Is placed on
the FML; the FML is a 1 rr:n (-50 mil) thick PVC; the reotextile is a
needl epunched nonwoven geotextile (mass per- unit area 450 g/m2 (13
oz/sq. yd), 4 mm (160 mil) thick, permeability 0.001 m/s (0.1 cm/s)
under no pressure and 0.0005 m/s (0.05 cm/s) under a 400 kPa (8,000
psf) pressure).
- Tests with FML Alone on Soil (no geotextile, no cover)
In this case, '.jsts show that the diameter of the FML hole needs
to be larger than 2 mm (0.08 In.) approximately in order to ensure
that free flow through the hole (assuming there 1s nothing under the
FHL) is larger than flow rate through soil alone. This indicates that
the soil layer has less influence in reducing leakage rate in the case
of very small holes than in the case of large holes.
Tests showed that the leakage rate becomes equal to the leakage
rate with no FML at all when the diameter of the FML hole is larger
than approximately 20 mm (3/4 in.) (Figure 2.2-13). This indicates
that leakage flows laterally between the FML and the soil and reaches
the walls of the permeameter (diameter 1.5 m (5 ft)) when the diameter
ot the ho1* is 20 mm (3/4 in.) or more. This also indicates that the
pressure in the Mquid located between the FML and soil is the same as
the pressure on top of the FML.
Pressure measurements in the soil (Figure 2.2-14a) showed that
the full water pressure is applied on top of the soil, which confirms
that there 1s a space between FML and soil where water flows freely.
In other words "-he FML was slightly uplifted by water. (Note that
pressure on tcr of the FML, plus the weight of the FML (specific
gravity 1.2) exceeds the pressure under the FML by 2 Pa (0.04 psf).
This is an extremely small pressure (i.e., of the order of the
pressure exerted by a couple of sheets of paper in dry conditions) and
1t is easily overcome by the stiffness of the FML, even a FML as
flexible as PVC - a PVC FML wrinkle can easily carry a couple of
sheets of paper.)
2.2-46
-------
- Tests with FML on Geotextile on Soil
The geotextile had no hole (only the FML had a ' le). The
geotextile and the FML were not glued together (I.e., the FML was
simply laid on the geotextile). (This detail is important in the
discussion presented hereafter.)
When FML hole was smaller than 30-50 mm (1-2 in.) approximately,
flow rate was approximately 20 times smaller than flow rate through
soil alone. In other words, when FML ho' 'iameter was smaller than
30-50 mm, using a geotextile under the _ decreased the flow rate
by approximately one order of magnitude or more.
Pressure measurements in the soil in the case of a 20 mm (3/4
1n.) diameter FML hole (Figure 2.2-14 b) showed that the water
pressure on the soil surface (i.e., under the geotextile) was roughly
uniform and 15 times smaller than the uniform pressure in the case
without geotextile between FML and soil. This indicates that the head
and, consequently, flow rate was 15 times smaller with geotextile than
without geotextile, which is consistent with the observations
mentioned above.
Pressure measurement in the soil in the case of a 50 mm (2 in.)
diameter FML hole (Figure 2.2-15) shewed that water pressure on the
soil surface was less uniform than in the case of a 20 mm (3/4 in.)
diameter FML hole. Pressures were larger in the vicinity of the ho'e
which indicated that there was we' ;r flowing in the geotextile within
a radius smaller than the radius of the test permeameter.
It may be concluded that FML, geotextile and soil stay in close
contact when the FML hole is smaller than 50 mm (2 in.). This appears
clearly because:
2.2-47
-------
• if water were accumulating between FML and geotextile, the
water pressure on the soil would be uniformly high, almost
equal to the water pressure on the FML (i.e., 200 or 400 kPa)
(4,000 or 8,000 psf) since geotextile permittivity (i.e.,
permeability/thickness) is much larger than soil permittivity
and, therefore, head loss through geotextile would be small;
and
• if water were accumulating between geotextile and soil, both
geotextile and FML would uplifted and the water pressure on
the soil would be equal to the water pressure on the FML (i.e.,
200 or 400 kPa (4,000 or 8,000 psf}).
FML, geotextile, and soil stay in close contact because the
pressure on top of the FML (200 or 400 kPa) (4,000 or 8,000 psf) is
much higher than the pressure below the geotextile. The same would
happen with the FML alone (i.e., water pressure on top of the FML
would be higher than water pressure under the TIL) if the FML were in
close contact with the soil. But, if the FML were not in close
contact with the soil because of small soil surface irregularities,
and if there were preferential channels for the flow of water between
the FML and soil, water pressure between FML and soil might become
equal to water pressure on top of the FML. If the soil surface were
perfectly smooth, and if the FML had no wrinkle, there would be no
preferential path for the water: the FML and the soil would stay in
close contact (the same way two pieces of polished steel stick to each
other because there is no air or water pressure between them).
- Tests with Earth Cover on the FML, but no Geotextile
In this case, the tests (conducted with FML hole diameter of 10
and 20 mm (3/8 and 3/4 in.)) show a flow rate reduction of the order
of 407. (i.e., a factor of 1.66) as compared to the case where thera is
no earth cover on the FML (Figure 2.2-13). The thickness of the earth
cover was 0.225 m (0.75 ft), and the thickness of the soil under the
2.2-48
-------
FML was 0.225 m (0.75 ft) (I.e., a total
ft) as in the tests discussed above).
soil thickness of 0.45 m (1.5
More tests would be necessary to f'-aw conclusions, such as tests
with a permeable cover material and comparable tests with identical
low-permeability compacted soil layer thickness under the FML.
However, the tests by Fukuoka show that an earth cover, even on a
flexible FML such as PVC, does not have a marked effect on leakage
rate probably because it i, not sufficient to force the FML Into soil
Irregularities.
2.2.4.4 Conclusions on_Leakage_through Compos1te_Top Liners
2.2.4.4.1 Conclusions from Analytical Stud es
It appears that the theoretical analyses involved in the
apparently simple problem of leakage through a hole in a FML placed on
a low permeability soil to form a composite liner are extremely
complex.
If perfect contact between the FML and soil is considered, the
two-dimensional problem has been solved but the three-dimensional
problem still requires research. There is no satisfactory approximate
solution and the analytical lower and upper boundaries are too far
from the actual solution to give valuable information.
Differential equations have been proposed and some approximate
numerical solutions are available for the case of imperfect contact
between the FML and soil. To use these equations, it is necessary to
know the bpacing between the FML and the underlying low-permeability
soil. Spacing values backcalculated from model tests are only
preliminary and are probably smaller than actual spacing values in the
field. The field conditions listed below „i 11 affect actual site-
specific leakage rates. While the quality of the FML-compacted soil
contact was probably better 1n the laboratory tests than 1t would be
In the field, the laboratory tests to date have been carried out only
for short durations and at unrealistically low levels of overburden
pressure. Field conditions affecting spacing between the FML
2.2-49
-------
component and the soil component of a composite liner are the
following:
• subgrade surface preparation;
• FHL wrinkles;
• overburden; and
• time.
As a result, act 1 leakage rates in the field are likely to
differ from those calculated using equations incorporating FML-soil
spacings backcalculated from model tests. Also, it is Hkely that
there will be some spatial variation throughout the liner.
2.2.4.4.2 Conclusion from Model Tests
Tests show that, 1n all cases where a FML 1s placed '.n direct
contact with a low permeability soil, seme liquid that has passed
through a hole 1n the FML flows laterally in the space between the FHL
and the underlying soil. Tests show that, as a result of lateral
flow, leakage rates observed are higher than leakage rates which would
be obtained if there was a perfect contact between the FML and the
underlying soil. The degree of contact between the FML and soil in
the model tests can be considered good (smooth soil surface, no cracks
in clay) but not perfect since flow takes place between the FML and
the soil.
From a construction standpoint, it is recommended to make every
effort to ensure a good contact between FML and low-permeability soil
which includes: (1) having a low-permeability soil with a smooth
surface and no cracks; and (ii) wrinkles in the FML. Ideally, the FML
should be sprayed on the low permeability soil instead of being made
in a plant and transported to the site: in this case, the contact may
be very good.
2.2-50
-------
From a design standpoint, H 1s necessary to take into account the
flow of leachate between the FML and the soil for leakage evaluation
as well as for any other appropriate design consideration such as
damage caused to the soil layers by liquid flowing 1n the space
between the FML and the underlying soil layer.
Although the tests provided a good understanding of the mechanisms
Involved, the diameter of the permeameter and test conditions used by
Brown et al. and Fukuoka limit the usefulness of the test results for
the development of design recommendations. Although extrapolation of
test data to field conditions was done by Brown et al. using a sound
theoretical analysis, test conditions were too far from actual
conditions to ensure that extrapolated values are adequate.
In spite of their limitations, the preliminary tests described in
this report show that composite liners are significantly more
effective than low-permeability compacted soil alone or FML alone.
However, the test results indicate that a FML in perfect contact with
a low-permeability soil (i.e., a FML sprayed on the soil) wou'd
exhibit even better performance than composite liners made uy
unrolling FMLs directly onto the soil.
2.2.4.4.3 Conclusions for Leakage Rate Evaluation
- Review of Methods for Leakage Rate Evaluation
A series of methods have been discussed to evaluate leakage rate
through a composite liner due to a hole in the FML. These methods can
be ranked as follows:
• An absolute minimum of the leakage rate is given by the
vertical flow equation assuming perfect contact between the FML
and the underlying s-oil (Equation 2.2-19).
2.2-51
-------
An approximate value (possibly an underestimate) of tt.o le>.a]o
rate in case of perfect contact between the FHL and the
underlying soil is given by Equation 2.2-21. Since this
equation has not been tested, it is appropriate to have the
absolute minimum mentioned above to make sure that no absurd
result is considered.
on
• Leakage rate obtained using charts prepared by Brown et al.
the basis of their tests (Figures 2.2-9 through 2.2-12) or the
empirical equations we have proposed to summarize these charts
(Equations 2.2-33 and 2.2-34) may be smaller than actual
leakage rate because in the field FHLs have at least sone
wrinkles and subgrade preparation may not be as good as in the
model tests, thereby allowing mere flow between the FHL and the
soil in the field than in the models. However, a counteracting
influence is that the overburden pressure in the model tests
was well below overburden pressures representative of field
condi tions.
• Finally, leakage through a hole in a FHL alone (i.e., with
nothing underneath it) is certainly much larger than leakage
through a composite liner with the same FML hole, even in field
conditions with a far from perfect contact between the FML and
the underlying soil. This case, therefore, provides an
absolute maximum of the leakage rate.
Leakage rate through a hole in a FML alone is given by
Bernouilli's equation (Equation 2.2-13). By combining Equations 2.2-
12 and 2.2-32, it appears that, if the spacing between the FML and the
soil is large enough to ensure free flow, the radius of the wetted
area is given by:
TT RJ kc = 0.6 a / 2 g h (Equation 2.2-35)
2.2-52
-------
R - 0.44 a0'6 (2 g h) °-2S kc~°-6 (Equation 2.2-36)
and, in the case of a circular hole:
R - 0.39 d (2 g h)0'2' kc~0ts
(Equation 2.2-37)
where: R = radius of the wetted area; a = hole area; d = hole
diameter; g - acceleration of gravity; h = hydraulic head on top of
FML; and kc = hydraulic conductivity of the low-permeability compacted
soil underlying the FML. Reconmended SI units are: R (m), a (m2), d
(m), g (m/s2), h (m), and kc (m/s).
A summary of pertinent equations is presented in Table 2.2-11.
- Leakage Rate ard Radius Graphs
Because of the uncertainties in the analyses as well as the wide
variety of contact conditions, it is appropriate in each given case to
plot leakage rates obtained with all the methods described above in
order to make interpolations. It is also appropriate to use a semi-
logarithmic scale for the plot since leakage rates vary within a range
of several orders of magnitude, as is usually the case in hydraulic
problems. The graph in Figure 2.2-16 has been established with a
1 cm* hole, which, as described previously, represents a large hole
(larger than current BOAT capabilities), possibly representative of a
design flaw or operational damage. This standard hole is used herein
for design calculations, as indicated in Section 2.2.3.3.6. The graph
i;i Figure 2.2-16 has been established for a hydraulic head of 30 mm
(0.1 ft) on top of the FML. Numerical values used to establish the
graph in figure 2.2-16 are given in Table 2.2-12.
2.2-53
-------
Similarly, a graph can be established for the radii of wett. .
areas (i.e.. the area covered by leakage flowing between the FML and
the 1ow-peimeabi1ity compacted soil, before it flows into the
compacted soil) obtained with all the methods described above and
summarized in Table 2.2-11. The radius graph related to a hydraulic
head of 30 nrn (0.1 ft) on top of the FML is given In Figure 2.2-17.
Similar graphs were established for a head of 0.3 m (1 ft) and 3 m
(10 ft). These graphs are presented in Figures 2.2-18 through 2.2-21.
Numerical values used to establish Figures 2.2-18 through 2.2-21 are
given in Tables 2.2-13 and 2.2-14.
- Use of Leakage Rate and Radius Graph
The leakage rate graph permits the determinati"- of the leakage
rate for any given field condition by interpolation cween the best
case and the worst case:
• In the best case: (i) the soil is well compacted, flat and
smooth, has not been deformed by rutting due to construction
equipment, and has no clods nor cracks; and (ii) the FML is
flexible and has no wrinkles.
• In the worst case: (i) the soil i:> poorly compacted, has an
irregular surface, and is cracked; and (ii) the FML is stiff
and exhibits a pattern of large, connected wrinkles. (Note:
this worst case condition is improbable with proper
construction quality assurance.)
The conditions in the best case may be as good as the conditions
in the tests by Brown et al. discussed in Section 2.2.4.3. Therefore,
on the graphs, the best case for field conditions is represented by
the vertical line corresponding to test results.
In order to locate the worst field case (which is unrealistic with
current construction quality assurance standards) for a range of
hydraulic heads on the top liner of 0.03 m (0.1 ft) to 3 m (10 ft), we
2.2- 1
-------
fl
have ic-su.-ned that the radius of the wetted area cannot exceed a value
ran-ji-^ frcm 3 to 30 m (10 to 100 ft). A val'je of the compacted soil
hydraulic conductivity of 10~" m/s (10~* cm/'s) was a'',o assumed. This
value of kj has been selected to represent a worst case condition.
The location of the worst case line thus obtained shows that the
conditions In the worst case are still nuch better than the case of
free flow through holes In the FML. Free flow Is an extreme case
which is possible only if the FML is very far from the low-
permeability compacted soil over a very large area (radius of 10 to
100 m), which is practically impossible.
Between the best field case and the worst field case we have
selected a vertical line representing good field conditions and a
vertical line representing poor field conditions. As a result, it
appears in Figure .' 2-16 that, for a head of 30 mm (0.1 ft), a leakage
rate of 0.08 liters/day (0.02 gallon/day) corresponds to good field
conditions and a hydraulic conductivity of kc = 10~* m/s (10~7 cm/s)
for the 'ow-permeabi1ity compacted soil underlying the FML. Poor
field conditions were based on poor FML contact and a hydraulic
conductivity, k^ = 10"* m/s (10~' cm/s). Under these conditions
(which might represent the worst possible realistic field conditions)
the leakage rate value obtained from Figure 2.2-16 is 4 liters/day
(1 gallon/day).
Figure 2.2-21 gives the radius of the wetted area in the case of a
head of 3 m (10 ft), which is a reasonable maximum of the hydraulic
head at hazardous waste surface impoundments. (The wetted area is the
area where leakage flows between the FML and the underlying compacted
soil before it seeps into the compacted soil.) Figure 2.2-21 shows
that the radius of the wetted area is between 3 and 24 m (10 and
80 ft) depending on the hydraulic coMuctivity of the compacted soil
and the quality of the contact between the FML and the compacted soil.
In other words, the large leakage estimated in the case of surface
impoundments corresponds to a large wetted area. This indicates that
leakage rate through top liners in the case of surface impoundments
can be decreased if flow between the FML and the underlying compacted
soil is restricted. Ther-fore, an overburden on the FML will be
beneficial since it will impede lateral flow between FML and
2.2-55
-------
underlying soil, thereby decreasing leakage. It 's preferable that
the o/erburcien be provided by a low-permeabi1ity •soi1 which will
further decrease leakage rate.
- Leakage Rate due to Permeation and Holes
Leakage rates through composite liners due to hole in the FML,
obtained from Figure 2.2-16 (head of 30 mm (0.1 ft)), Figure 2.2-18
(head of 0.3 m (1 ft)), and Figure 2.2-19 (head of 3 m (10 ft)), are
summarized in Table 2.2-15, which also gives leakage rates due to
permeation obtained from Table 2.2-10. (Although Table 2.2-10 was
established for a FML alone, leakage rates due permeation from
Table 2.2-10 can be used for composite liners since leakage rate due
to permeation should not be significantly affected by the underlying
soil because all soil are very pervious as compared to FMLs.)
To the best of our knowledge. Table 2.2-15 summarizes the best
demonstrated available technology on leakage rate through composite
top liners. This table shovs that leakage rates through top composite
liners can be larger in the case of surface impoundments than in the
case of landfills or waste piles.
2.2.5 Conclusions on Leakage through Top L1rer_s
General conclusions are drawn hereafter from all discussions
presented in Section 2.2, which is a thorough review of theoretical
analyses, laboratory tests, and field data. However, it should be
pointed out that these are only tentative conclusions since they
should be subjected to peer review. The ten' twe conclusions are as
follows.
2.2.5.1 Defects and Qual1ty_Assurance
With good quality assurance, it is reasonable to expect 3 to 5 FML
defects per hectare (one or two seam defects per acre). These defects
can be caused by inadequate seaming (which is the most common type of
defect) and FML defects caused by puncture, tear, excessive stresses,
etc. Defects may also be due to inadequate FML connections to sumps
2.2-55
-------
r
and otr.fr appurtenances v,hich are often problem areas. Also, the FML
may undergo excessive stresses in the vicinity of these connections
that v.iy cause defects to develop. Under the very best conditions
with excellent design, installation and construction quality
assurance, a frequency of FML defects below 3 to 5 per hectare (1 to 2
per acre) may be possible.
The leakage rate values summarized in Table 2.2-16 are based en a
FML defect frequency of 3 per hectare (1 per acre) and on a "standard"
hole size of 1 cm2 (0.16 in2.). It must be remembered that the
results presented in luble 2.2-16 are for a properly designed and
constructed land disposal unit in which: (i) good quality assurance is
provided; (ii) extreme care is taken for FML connections to sumps and
other appurtenances; and (iii) a proper design minimizes the risk of
excessive stresses, which could generate holes much larger than 1 cm2
(0.16 in'.).
2.2.5.2 Summary of Leakage_Rate Values
Table 2.2-16 summarizes leakage rates through top liners. This
table has been obtained by combining Table 2.2-10 for FMLs alone wHh
Table 2.2-15 for composite liners. To the best of our knowledge, this
table represents a reasonable summary of the rates of leakage through
properlj designed and constructed top liners.
This table has been established with the following assumptions:
• The FML is 1 mm (40 mil) thick and has one standard defect with
a surface area of 1 cm2 (0.16 in2.) per 4,000 m2 (acre).
• The low-permeability compacted soil underlying the FML has a
thickness of 0.9 m (1 ft) and a hydraulic conductivity of 10"'
m/s (10"' cm/s) in the case of good FML/compacted soil contact
and 10"' (10"' cm/s) in the case of poor FML/compacted soil
contact.
2.2-57
-------
r
The three hydraulic heads used in Table 2.2-i6 represent the
following conditions:
• 0.03 m (0.1 ft) is assumed to be the average head acting on the
top Hner of a landfill or waste pile with a well designed and
constructed leachate collection and removal system (LCRS),
above the top liner;
• 0.3 m (1 ft) is the maximum head considered in the design of
the leachate collection and removal system (LCRS) of a landfill
or waste pile; and
• 3 m V10 ft) is assumed to be the maximum head on the top liner
of a surface impoundment.
2.2.5.3 Comments on Leakage Rate_Values
It appears from Table 2.2-.J that leakage rates through FML top
liners are relatively high if there is one hole per 4,000 m2 (acre) in
the FML: typically 30 to 300 Ltd (gpad) 1n landfills and waste piles
(and up to 1,000 Ltd (gpad) in the extreme) and 300 to 3,000 Ltd
(gpad) in surface impoundments. Since it is impossible to guarantee
that there will be no hole in a FML, owners and operators may
increasingly consider the use of a composite top liner.
Table 2.2-16 shows that leakage rates in the case of a landfill
with a composite top liner can be small (i.e., almost always less than
10, and usually less than 1 Ltd (gpad)). However, it should be kept
in mind that such low leakage rates can be achieved only 1f the lining
system is constructed with very good quality assurance and if the FML
is not subjected to excessive stresses likely to cause a large breach.
2.2-58
-------
r
Table 2.2-16 also shews that, even with a composite top liner the
leakage rate may remain high (e.g., from 10 to 50 Ltd (gpad)) in a
surface Impoundment with a hydraulic head of 3 n (10 ft). As
Indicated In Section 2.2.4.4.3, leakage rates can be decreased if the
FML is covered with a layer of soil, preferably a soil with a low
permeability. It also appears that leakage rates due to permeation
through the FML may not be negligible in the case of surface
Impoundments.
2.2-59
-------
Table ?.2-l. Summary of Typical P'Ls
CAIEGORY POLYMER
Thermoplastics
SYMBOL"
Polyvinyl Chloride
Oil-resistant PVC
Thermoplastic Nitrile-PVC
Ethylene Interpolymer Alloy
PVC
PVC-OP.
TN-PVC
EIA
Crystal!ine
Thermoplastics
Low Density Polyethylene LDPE
Linear Low Density Polyethylene LLDPC
High Density Polyethylene MO, _
High Density Polyethylene- HDPE-A
Alloy
Polypropylene PP
Elasticized Polyolefin ELPO
Thermoplastic
Elastomers
Chlorinated Polyethylene CPE
Chlorinated Polyethylene- CPE-A
Alloy
Chlorosulfonated Polyethylene CSPE
("Hypalon")
Thermoplastic Ethylene-Propylene T-EPDM
Diene Monomer
Elastomers Isoprene-Isobutylene Rubber
("Butyl Rubber")
Ethylene-Propylene Diene
Monomer
Polycholoroprene ("Neoprene")
Epichlorohydrin Rubber
IIR
EPOM
CR
CO
symbols consistent witn those used by the National Sanitation
FOL Cation (NSF) Joint Committee of Flexible Membrane Liners (FML)
Note: Polymers are usually compounded with various additives such
as fillers, fibers, carbon black, plasticizers, stabilizers,
antioxidants, fungicides, and other polymers. These additives perform
various functions without altering the very low permeability of the
base product.
2.2-60
-------
Table 2.2-2. Typical values of soil hydraulic conductivities [Giroud,
1975],
m/s
cm/s
feet/year
Soil with a smal 1
coefficient of
uniformity that
does not contain
particles finer
than indicated
Soil mixtures
that do not
contain particles
finer than
Indicated
1
10'
10'
1(T'
1
10'
Gravel
10-
10'2
•0'
Sand
(coarse,
medium)
Gravel
+
Sand
10"'
10""
10'
10''
10''
1
Fine Sand
Silt
Cracked
Clay
Sand
+
Silt
10"10
10"§
1(T'
10'"
10"10
10""
Non Cracked
Clay
Fine Sand
(or si
t)
•«•
Clay
2.2-61
-------
Table 2.2-3. Values of the ...gration coefficient, u, obtained frcrn permeability
tests conducted at the University of Grenoble (France) with t'-e
apparatus shown in Figure 2.2-1.
FML Type
CSPE
Butyl
Butyl
EF
PVC
PVC
PVC
Asphal tic
Asphal tic
hydraulic head, h, in m
5
3.5x10""
4.2xlO~"
10
3.8x10""
7,7x10""
1.7x10""
1.1x10""
1.7x10""
1.6x10""
8.1x10""
7.4x10""
1.6x10""
25
1.9x10""
6.7x10""
3.2xlO~"
50
5.0x10""
3.9x10""
2.9x10""
2.3x10""
2.'5xlO~"
2.1x10""
2.0x10""
6.5x10""
6.5x10""
75
7.4x10""
4.5x10""
100
5.5x10""
3.1x10""
3.0x10""
2.2x10""
1.1x10""
4.4x10""
1.0x10""
Values of u in m2/s
2.2-62
-------
Table 2.2-4. Water vapor transmission (WVT) rates of FHLs from [Haxo
et. al.t 1984] and values of the coefficient of migration
derived from WVT values using Equation 2.2-9 (See also
Table 2.2-5). All these tests were conducted at 23°C
with a relative humidity difference of 50%, which is
equivalent to a pressure of 1.4 kPa, i.e., a head of 0.14
m of water.
Polymer
Thickness,
(mm)
Water Vapor
Transmission
WVT,
(g/m'.day)
Coefficient
of migration
V
(m'/s)
Butyl rubber
CPE
CSPE
ELPO
CO
EPDM
0.85
0.85
1.85
0.53
0.79
0.79
0.85
0.94
O.S7
0.74
0.76
0.89
0.91
0.94
1.07
0.72
1.160
1.650
0.51
0,94
1 70
0.384
0.020
0.097
0.643
1.400
0.320
0.264
0.305
0.64.,
0.333
0.663
0.438
0.748
0.422
0.252
0.142
20.18
14.30
0.270
0.190
3.8 x 10~"
2.0 x 10"'*
2.1 x 10""
3.9 x 10""
1.2 x 10""1
2.9 x 10~"
2.6 x 10""
2.2 x 10""
7.2 x 10""
2.9 x 10""
5.8 x 10""
4.5 x 10~"
7.9 x 10""
4.6 x 10""
3.1 x 10""
1.2 x 10""
2.7 x 10""
2.7 x 10""
1.6 x 10""
2.1 X 10""
0 1 •• 1ft"11
«>. 4 A i V
2.2-63
-------
£"—--
Table 2.2-4, continued
Neoprene
0.51
0.91
1.27
1.59
Nitrile rubber
PBa
PEEL5
LOPE
HOPE
MDPE-A
PVC
PVC-E
PVC-OR
Saran Film
0.
0.
0.
0.
0.
2.
0.
0.
0.
0.
0.
0.
0.
0.
76
69
20
76
80
44
86
28
51
76
79
91
83
013
0.304
0.473
0.429
0.237
5.51
0.084
10.50
0.0573
0.0172
0.0062
0.0472
4.42
2.97
1.94
1.85
2.78
4.17
0.563
1.
5.
6.
4.
4.
6.
2.
5.
1.
1.
4.
1.
1.
1.
1.
2.
4.
8.
8
0
3
4
8
7
4
0
6
8
7
4
7
7
7
9
0
5
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
10"
10"
10"
10"
10"
10"
10"
10"
10"
10"
10"
10"
10"
10"
10"
10"
10~
10"
1 1
1 1
i >
1 1
1 4
1 I
1 4
I i
1 t
1 «
1 t
1 4
1 4
1 4
1 4
t 4
1 4
I 7
i*PB = polybutylene.
PEEL = polyester elastomer.
(Other symbols are defined in Table 2.2-1}
2.2-64
-------
Table 2.2-5. Water vapor transmission (WVT) rates of Fills [Rogers, ]
and values of the coefficient of migration derived frcii
WVT values using Equation 2.2-9. (See also Table 2.2-4.)
Reference
FMl Pressure
Type
P
(kPa)
Hypalon 6.4
Butyl 6.4
PVC 6.1
HOPE 0.92* 6.4
0.94* 5.8
0.95* 6.1
0.96* 5.8
Water Reference Coefficient
Vapor Thickness of
Transmission Migration
WVT T u
(g/m'.day) (mm) (m'/s)
161 0.025 4.6xlO~"
26 0.025 7.5x10""
32 0.025 9.2x10""
28 0.025 8.1x10""
14 0.025 4.1x10""
6.7 0.025 1.9x10""
4 0.025 1.1x10""
Notes: (1) the test pressure, p, is derived from the test relative
humidity difference using Equation 2.2-6; (ii) a 6 kPa pressure
1s equivalent to a water head of 0.6 m (2 ft); (iii) the values
with (*) refer to the polymer density 1n g/cm1.
2.2-65
-------
Table 2.2-6. Summary of values of the coefficient of migration, \i,
from Tables 2.2-3. 2.7-1 and 2.2-5.
Hydraul ic
head
h
0.14 m
0.6 m
10 m
50 m
100 m
FML Type
CSPE
5x10""
4.6x10"''
3.8x10""
5.0xlO~"
5.5x10""
PVC
1.7x10""
9.2x10""
1.6x10""
2.0xlO"'z
1.0x10""
HOPE
1.7x10""
4.1x10""
-
-
-
Values of coefficient of migration, u (m2/s)
2.2-66
-------
Table 2.2-7. Value- of coefficient of migration resulting from
extrapolations and Interpolations in Figure 2.2-3.
FML
Type
CSPE
HOPE
Hydrau! ic head in m (ft)
0 m
(0 ft)
0
0
0.03 m
(0.1 ft)
3.5xlO~"
1.5xlO~17
0.3 m
(1 ft)
1.5x10"'*
1x10'"
3 m
(10 ft)
6xlO~"
7xlO"14
Values of coefficient of migration, K, in m'/s
> 10 m
(> 30 ft)
6xlO"12
1x10""
Pmax
Table 2.2-8. Values of rate of leakage due to permeation through FML
derived from values of coefficient of migration given in
Table 2.2-7, using Equation 2.2-4 and assuming an FML
thickness of 1 mm (40 mils).
FML
Type
CSPE
HOPE
Hydraul ic head in m (ft)
0 m
(0 ft)
0
0
0.03 m
(0.1 ft)
0.035
0.0015
0.3 m
(1 ft)
1.5
0.1
3 m
(10 ft)
60
7
> 10 m
(> 30 ft)
600
100
Values of leakage rate in IHers/lOOOm'/day (Ltd) or
gallons/acre/day (gpad)
2.2-67
-------
Tabie 2.2-9. Leakage rate due to pinholes and holes in an fML placed
on a very pervious medium such as a drainage }*-j°r.
Note: in the case of pinholes, FML thickness is 1 nin (40
mils) while in the case of holes, leakage rate is
Independent of FML thickness.
Pinholes
Holes (*)
Defect
diameter
0.1 mm
(0.004 In.)
C.3 mm
(0.012 in.)
2 mm
(0.08 In.)
11.3 mm
(0.445 in.)
Hydraul ic head
0.03 m
(0.1 ft)
0.06
(0.015)
5
(1)
125
(30)
1,260
(330)
0.3 m
(1 ft)
0.6
(0.15)
50
(13)
400
(100)
4,000
(1,000)
3 m
(10 ft)
6
(1.5)
500
(130)
1250
(300)
12,600
(3,300)
Values of leakage rate In liters/day
(gal Ions/day)
(*) The il.3 mm diameter circular hole has a surface area of 1 cm1.
2.2-68
-------
r
Table ?.2-10. Leakage rate through a FML top liner assuming one
Dinhole, srr.ill hole, or standard hole per -I.CCO rn1
(acre). This table has been established by combining
Tables 2.2-8 and 2.2-9 and rounding up the numbers. The
considered pinhole has a diameter of 0.1 mm (0.004 in.),
the small hole has a diameter of 2 mm (0.08 in.), and the
standard hole has a surface area of 1 cm2 (0.16 in.2),
I.e.. a diameter of 11.3 mm (0.445 in.).
Permeation
Pinhole
Small hole
Standard
Hole
Hydraul ic head
0.03 m
(0.1 ft)
0.001
0.01
3Q
300
0.3 m
(1 ft)
0.1
0.1
1QQ
1,000
3 m
(10 ft)
10
1
30Q
3,000
Values of leakage rate in 1 iters/1000m2/day (Ltd)
or gallons/acre/day (gpad)
2.2-69
-------
n
T-ible 2.2-11.Summary of equations giving leakage rate, 0, and radius
of wetted area, R, for composite liners when there is a hole in the
ABSOLUTE MINIMUM (MIN) in Figures 2.2-16 through 2.2-21
(Vertical flow)
Q - kc a (h + H)/H (Equation 2.2-19)
R « d/2
PERFECT CONTACT (P.C.) In Figures 2.2-16 through 2.2-21
(Approximate value of Q given by radial flow)
Q = it kc h d (Equation 2.2-21)
R = unknown
EXCELLENT CONTACT (TEST) In Figures 2.2-16 through 2.2-21
(Empirical equations from model tests)
Q = 0.7 a'" kc0'" h (Equation 2.2-33)
R - 0.5 a0'0' kc~">01 h°" (Equation 2.2-34)
LARGE SPACE BETWEEN (MAX) in Figures 2.2-16 through 2.2-21
FML AND SOIL
(Q given by Bernouilli's equation)
Q - C a /~2gh - 0.6 a / 2gh (Equation 2.2-13)
R = 0.39 d (2 g h)"'2' kc"°" (Equation 2.2-37)
where: kc - hydraulic conductivity of low-permeability compacted soil
underlying the FML; a = area of hole in FML; h » hydraulic head on
FML; H = thickness of compacted soil layer; d = diameter of hole in
FML; and g - acceleration of gravity. Recommended SI units: kc
(m/s), a (m1), h, H, and (m); and g (m/s1). These units are mandatory
for the two empirical equations.
2.2-70
-------
r
Table 2.2-12. Numerical values used to establish the graphs presented n Figures
2.2-15 and 2.2-17. This tible has been established for a
hydraulic head of 30 rnn (0.1 ft) on top of the FML, a hole area
of 1 cm1 (0.16 In'.), and a low-permeability compacted soil
thickness of 0.9 m (3 ft).
Hydraulic Conductivity of Compacted Soil
Underlying the FML
Leakage
Rate
g
(m'/s)
Radius
of
Wetted
Area
R
(m)
Case
Absolute minimum
Perfect contact
(approximate
theory)
Good contact
(model tests)
Free flow
(Bernouilli 's
equation)
Absolute minimum
(hole radius)
Perfect contact
(unknown)
Good contact
(model tests)
Free flow
Equation
2.2-19
2.2-21
2.2-33
2.2-13
R - d/2
2.2-34
7.2-37
10"' m/s
1.0x10'"
l.lxlO"10
5.8x10"'
4.6xlO"s
0.0056
-0.032(*)
0.14
12
10"' m/s
1.0x10""
1.1x10""
7.6x10"'°
4.6x10"'
0.0056
•0.032(*)
0.17
38
10"' m/s
1.0x10""
1.1x10""
1.0x10"'°
4.6xlO~5
0.0056
-0.032(*)
0.19
122
(*) Value obtained by interpolation in Figure 2.2-17,
2.2-71
I
-------
Table 2.2-13. V,ner1cal values used lo establish the graphs presented in Figures
?.2-3 and 2.2-19. This tiole has been establish?-] fcr a
hydraulic head of 0.3 m (1 ft) on top of the FML, a hole area
of 1 cm2 (0.15 in2.), and a low-permeability compacted soil
thickness of 0.9 m (3 ft).
Hydraulic Conductivity of Compacted Soil
Underlying the FML
Leakage
Rate
9
(m'/s)
Radius
of
Wetted
Area
R
(m)
Case
Absolute minimum
Perfect contact
(approximate
theory)
Good contact
(model tests)
Free flow
(Bernouilli's
equation)
Absolute minimum
(hole radius)
Perfect contact
(unknown)
Good contact
(model tests)
Free flow
Equation
2.2-19
2.2-21
2.2-33
2.2-13
R = d/2
2.2-34
2.2-37
10'7 m/s
1.3 '0'"
1.1x10"'
5.8x10"'
l.Sxlt/"4
0.0056
-0.045(*)
0.45
22
10 ' m/s
1.3x10""
l.lxlO"10
7.6x10"'
1.5X10"'
?.0056
-0.045(*)
0.52
69
10"' m/s
1.3x10""
1.1x10""
1.0x10"'
1.5x10"'
0.0055
•0.045(*)
0.60
217
(*) Value obtained by interpolation in Figure 2.2-19.
2.2-72
-------
r
Table 2.2-14. Numerical values used to establish the graphs pre^.-nted in Figures
2.2-20 and 2.2-21. This table has been established for a
sydraulic head of 3 m (10 ft) on top of the FHL, a hole area of
i cm1 (0.16 in2.), and a low-permeability compacted soil thickness of
0.9 m (3 ft).
Hydraulic Conductivity of Compacted Soil
Underlying the FML
Leakage
Rate
g
(m'/s)
Radius
of
Wetted
Area
R
(m)
Case
Absolute minimum
Perfect contact
(approximate
theory)
Good contact
(model tests)
Free flow
(Bernoui 1 1 i ' s
equation)
Absolute minimum
(hole radius)
Perfect contact
(unknown)
Good contact
(model tests)
Free flow
Equation
2.2-19
2.2-21
2.2-33
2.2-13
R = d/2
2.2-34
2.2-37
10"' m/s
4.3x10""
1.1x10"'
5.8x10"'
4.6x10""
0.0055
-0.11(*}
1.4
39
10"' m/s
4.3x10""
1.1x10"*
7.6x10"'
4.6x10""
0.0056
-O.IK*)
1.7
122
10"' m/s
4.3x10""
l.lxlO"10
1.0x10"'
4.6x10""
0.0056
-o.ii(*)
1.9
386
(*) Value obtained by interpolation in Figure 2.2-21,
2.2-73
-------
Table 2.2-15. Leakage rates through composite liners. Leakage due to
permeation is obtained from Table 2.2-10 and leakage dj
to holes is obtained froii^ F igures 2.2-16, 2.2-17 ana
2.2-18, as a function of the quality of contact between
the FML component and the compacted soil component of
the top liner. This table has been established with:
hole frequency = 1 per 4000m* (1 per acre); hole area =
1 cm2 (0.16 in*.); compacted soil thickness = 0.9 m (3
ft); compacted soil hydraulic conductivity 10"
(10~* cm/s); and FML thickness = 1 m (40 mils).
m/s
Qual ity
of
contact
Good
Poor
Leakage
mechanism
Permeation
Hole
TOTAL
Permeation
Hole
TOTAL
Hydraul ic head, h
0.03 m
(0.1 ft)
0.001
0.02
0.02
0.001
1
1
0.3 m
(1 ft)
0.1
0.2
0.3
0.1
8
8
3 m
(10 ft)
10
3
13
10
50
60
Values of leakage rate in Ltd
or gpad
2.2-74
-------
Table 2.2-15.
Leakage rates through top liners. This table has been
obtained by combining Tables 2.2-iO and 2.2-15. The
small hole has a dia.r.eter of 2 irm (0.08 in.). The
standard hole has a surface area of 1 cm2 (0.1G in!).
The frequency of holes is 1 per 4000rnz (1 per acre).
The thickness of the compacted soil layer is 0.9 m (3
ft) and its hydraulic conductivity is 10"' m/s (10"'
cm/s). Note: Ltd = I iter/1000m2/day; gpad -
gallons/acre/day; 1 Ltd = 1.1 gpad.
Type
of
Liner
FHL
alone
Composite
1 iner
(good)
contact)
Compos i te
1 iner
(poor)
contact)
Leakage
mechanism
Permeation
Small hole
TOTAL
Permeation
Standard hole
TOTAL
Permeation
Standard hole
TOTAL
Permeation
Standard hole
TOTAL
Hydraulic head, h
0.03 m
(0.1 ft)
0.001
30
30
0.001
300
300
0.001
0.02
0.02
0.001
1
1
0.3 m
(1 ft)
0.1
100
100
0.1
1,000
1,000
0.1
0,2
0,3
0.1
8
8
3 m
(10 ft)
10
300
300
10
3,000
3,000
10
3
13
10
50
60
1
Values of leakage rate in Ltd
or gpad
2.2-75
-------
f
^
A/fc UWbER.
PRESSURE
WATER
POROUS STOfJE
FML
Figure 2.2-1. Permeometer used to evaluate flow through intact FMLs at
the University of Grenoble (France).
2.2-76
-------
\v.
Figure 2.2-2. Typical shape of the curve giving the coefficient of
migration, Ug, as a function of the hydraulic head, h.
2.2-77
-------
lE-ii
1E-12:
.2 1E-13J
-i_>
ra
t_
en
z ! 1-U
o
CJ
LJ
O
a
1E-15J
1E-16J
1E-17-
1E-3 1E-2 1E-1 IE 0 5E i
Hydraulic Head (in)
IE 2
IE 3
Figure 2.2-3. Values of coefficient of migration, ju, for various FMLs
from Table 2.2-6.
2.Z-78
-------
(-1
Figure Z.2-4. Flow nets for the four cases considered in two-
dimensional theoretical studies related to leakage
through composite liners due to hole in FHL, assuming
perfect contact between FI'.L and soil layer: (a) entire
soil layer saturated; (b) radial flow; (c) vertical
flow; (d) actual flow. As demonstrated by Faure [1979],
the actual flow is limited laterally by a phreatic
surface. Note that in cases (a), (b), and (d), there is
flow in thp soil along the interface, although there is
no flow between the FHL and the soil because there is no
space between the FML and the soil in the considered
cases since perfect contact is assumed.
2.2-79
-------
Figure 2.2-5. Typical flow net? for leakage through a composite liner
due to a FML hole (two-dimensional study assuming that
the FML and the underlying soil are in perfect contact)
(see case (d) in Figure 2.2-4). The cases shown above
are: (a) b/ll = 0.005 and h/H = 1; (b) b/tl = 0.005 and
h/ll = 3; (c) b/H = 0.05 and h/ll = 1/3; and (d) b/H =
0.05 and h/ll = 1. Notation: b = width of infinitely
long hole (slot) in the FML; h = hydraulic head on top
of the FML; and II = thickness of the soil layer
underlying the FML [Faure, 1979].
2.2-80
-------
Figure 2.2-6. Lateral extent of the phreatic surface limiting the flow
in the soil layer due to a hole in the FHL. ' is chart
is related to the two-dimensional case (the i.cle is a
slot of width b) and perfect contact is assumed between
the FHL and the soil layer [Faure, 1979J.
2.2-81
-------
r
i it ii i ii in
II
h
Figure 2.2-7. Leakage rates through a composite liner due to a slot of
width b in the FML (two-dimensional case), assuming
perfect contact between the FML and the soil.
Calculations were made with several assumptions
regarding flow: (a) soil entirely saturated by the
flew; (b,) radial flow using Equation 2.2-13; (b,)
radial flow using Equation 2.2-15; (c) vertical flow;
(d) actual flow. Cases (a) through (d) are illustrated
in Figure 2.2-4.
2.2-02
-------
r
- _Q
U= k(hH
1.0
o.o i
Figure 2.2-8. Chart giving dimensionless coefficient C to be used in
Equation 2.2-14 which gives the leakage rate through a
composite liner due to a slot in the FML (two-
dimensional case). Coefficient C can also be used in
Equation 2.2-22 to make an approximate evaluation of the
leakage through a composite liner due to a circular hole
1n the FHL (three-dimensional case). Notation: h =
hydraulic head on top of the FML; b = width of the slot
(to be replaced by the diameter d of a circular hole
when the chart is used for the three-dimensional case);
and H = thickness of soil layer.
2.2-83
L
-------
5.0
3.S
3.0
tf 2.3
2
g-
d Z.O
1.5
1.0
0 .3
10 20
30 40 50 60
70
80
90
5S
5O
•10
35
30
ZS
20
10
100
Figure 2.2-9. Leakage through a composite liner due to a hole in the
FML [Brown et al.]. Chart giving the leakage rate, 0,
and radius, R, of the wetted area as a function of the
hydraulic head on the FML, for a compacted soil
hydraulic conductivity kc = 3.4 x 10"' m/s (3.4 x 10~'
cm/s). Notation: d = diameter of the FML hole; and h -
hylraulic head on the FML. Note: although the chart in
[Brown et al.] is labeled "kc = 10"' cm/s", it seems to
us that it was established for 3.4 x 10"" cm/s.
2.2-84
-------
1.0
.9
.8
.7
.6
•»
§
c! .<
r)
F.
0 I-
O
95
90
85
60 j.
75
70
65
60
•55
50
45
40
30
20
10
10 ZO
10 50 60
IEAO (cm)
70
00
90
IOO
Figure 2.2-10.
Leakage through a composite liner due to a hole in the
FML [Brown et al.]. Chart giving the leakage rate, 0,
and radius, R, of the wetted area as a function of the
hydraulic head on the FML, for a compacted soil
hydraulic conductivity kc = 3.4 x 10"' m/s (3.4 x 10"'
cro/s). Notation: d = diameter of the FML hole; and h
= hydraulic head on the FML. Note: although the chart
In [Brown et al.] is labeled "kc = 10"' cm/s", it seems
to us that it was established for 3.-1 x 10"' cm/s.
2.2-85
-------
.10
.09
.08
.07
.06
.05
a
_i
"• .04
.03
.02
.01
90
85
60
75
70
65
60
55
50
45
40
30
20
10
10
2O
30
4O
50
60
70
OO
90
100
(cm)
Figure 2.2-11.
Leakage through a composite liner due to a hole in the
FML [Brown et al.]. Chart giving the leakage rate, Q,
and radius, R, of the wetted area as a function of the
hydraulic head on the FML for a compacted soil
hydraulic conductivity kc = 3.4 x 10"' m/s (3.4 x 10"'
cm/s). Notation: d = diameter of the FML hole; and h
hydraulic head on the FML. Mote: although the chart
in [Brown et al.] is labeled "k(
10"
cm/s", it seems
to us that it was established for 3.4 x 10 ' cm/s.
2.2-86
-------
.OO9
.008
.oor
.DOS
.003
K
O
G! .C
.003
.002
.001
O' IO 2O 3O 40 SO 6O
00
90
85
80
?S
70
65
60
• 55
50
<5
«0
30
ZO
10
Figure 2.2-12. Leakage through a composite liner due to a hole in the
FML [Crown et al.]. Chart giving the leakage rate, Q,
and radius, R, of the netted area as a function of the
hydraulic head on the FML, for a compacted soil
hydraulic conductivity kc = 3.4 x 10"' m/s (3.4 x 10"'
cm/s). Notation: d = diameter of the FML hole; and h
- hydraulic head on the FML. Note: altho'"ih the chart
in [Brown et al.] is labeled "kc = 10"' ci. .5", it seems
to us that it was established for 3.4 x 10"' cm/s.
2.2-07
»&«*,,
L
-------
luo
U)
y
./
OX.';.
'&
tf
<*
X
-v&-.-
2
fi,., ml:, UfnjV I"-'- Of Kite.
—I—"•;
5 10
Diameter of Defect (nun)
50
Figure 2.2-13. Leakage rates measured in tests conducted with a FML
having a circular hole [Fukuoka, 1985]: (A) no soil
cover on the FML, no geotextile between the FML and the
soil; (B) there is a geotextile between the FML and the
soil, but there Is no soil cover on the FML; and (C)
there is a soil cover on uhe FML and no geotextile
between the FML and the soil, flotation: Q = leakage
rate measured in the tests; and QQ = leakage rate when
there is no FML (i.e., leakage rate governed by Darcy's
flow through the soil).
2.2-80
-------
J
1131 ,.i9' O 5 2-1
IS
V'j
'Oi-1 COM UIH5
it 1oo Kf"^.
Figure Z.2-14. Water pressure in the soil under the FML in the case of
a 20 mm (3/4 in.) diameter hole in a 1 mm (40 mil)
thick PVC FML; (a) the FML 1s placed directly on the
soil; and (b) there is a geotextiie between the FML and
the soil [Fukuoka, 1985].
2.2-09
-------
r0p
IS
Figure 2.2-15. Water pressure in the soil under the F-ML in the case of
a 50 run (2 In.) diameter hole in a 1 mm (-10 mil) thick
PVC FML placed on a needlepunched nonwoven geotextile
(mass per unit area 450 g/mj (13 oz/sq. yd)) resting on
the soil [Fukuoka, 1985].
2.2-90
-------
fnL-50/L CONTACT '.
Figure 2.2-16.
Graph giving the leakage rate in case of leakage
through a FHL hole in a composite liner. The hydraulic
head is 30 nvn (0.1 ft) and the hole area is 1 cm2
(i.e. diameter of 11.3 mm). Because of uncertainties
in the analytical analyses as well as the large
Influence of soil conditions and contact between the
FML and the soil, only a range of values can be given.
Field conditions can be anywhere between the two
extremes: (1) best, i.e., the soil is well compacted,
flat and smooth, has not been deformed by rutting
during construction, and has no clods and cracks, and
the FML is flexible and has no wrinkles; and (2) worst,
i.e., the soil is poorly compacted, has an irregular
surface and is cracked, and the FML is stiff and
exhibits a pattern of large, connected wrinkles.
Abbreviations: GOOD and POOR = good and poor field
conditions; MIfi, P.C., TEST, and MAX are defined in
Table 2.2-11.
2.2-91
L
-------
•0'.
t ^
>,-> (OUL)
t,.
(ft) ('tit) «co rx^ (>-A<)
-SOIL CONTACT CONDITION
Figure 2.2-17. Graph giving the radius of the wetted area in case of
leakage throuah a FML hole in a composite liner. The
hydraulic head s 30 im (0.1 ft) and the hole area is 1
cm' (i.e., diameter of 11.3 mm). Because of
uncertainties in the analytical analyses as well as the
large influence of soil conditions and contact between
the FML and the soil, only a range of values can be
given. Field conditions can be anywhere between the
two extremes: (1) best, i.e., the soil is well
compacted, flat and smooth, has not been deformed by
rutting during construction, and has no clods and
cracks, and the FML is flexible and has no wrinkles;
and (2) worst, i.e., the soil is poorly compacted, has
an irregular surface and is cracked, and the FML is
stiff and exhibits a pattern of large, connected
wrinkles. Abbreviations: GOOD and POOR = good and
poor field conditions; MIN, P.C., TEST, and MAX are
defined in Table 2.2-11.
2.2-92
-------
FML -SOIL CONTACT COND/T/OA/
Figure 2.2-18. Graph giving the leakage rate in case of leakage
through a FML hole in a composite liner. The hydraulic
head Is 0.3 m (1 ft) and the hole area is 1 cm2 (i.e.,
diameter of 11.3 mm). Because of uncertainties 1n the
analytical analyses as well as the large influence of
soil conditions and contact between the FML and the
soil, only a range of values can be given. Field
conditions can be anywhere between the two extremes:
(1) best, i.e., the soil is well compacted, flat and
smooth, has not been deformed by rutting during
construction, and has no clods and cracks, and the FML
is flexible and has no wrinkles; and (2) worst, i.e.,
the soil is poorly compacted, has an irregular surface
and is cracked, and the FML is stiff and exhibits a
pattern of large, connected wrinkles. Abbreviations:
GOOD and POOR = good and poor field conditions; M1N,
P.C., TEST, and MAX are defined in Table 2.2-11.
2.2-93
-------
I '-•
K • 0 J in (if.)
(••IN) (rc)
FML- SOIL
CONTACT
(i'.«i)
CONDITION
Figure 2.2-19. Graph giving the radius of the wetted area in case of
leakage through a FHL hole in a composite liner. The
hydraulic head is 0.3 m (1 ft) and the hole area is 1
cm2 (i.e., diameter of 11.3 mm). Because of
uncertainties in the analytical analyses as well as the
large influence of soil conditions and contact between
the FML and the soil, only a range of values can be
given. Field conditions can be anywhere between the
two extremes: (1) best, i.e., the soil is well
compacted, flat and ^nooth, has not been deformed by
rutting during construction, and has no clods and
cracks, and the FHL is flexible and has no wrinkles;
and (2) worst, i.e., the soil is poorly compacted, has
an irregular surface and is cracked, and the FML is
stiff and exhibits a pattern of Targe, connected
wrinkles. Abbreviations: GOOD and POOR = good and
poor field conditions; M1N, P.C., TEST, and MAX are
defined in Table 2.2-11.
2.2-94
L
-------
r
X
FML-SOIL CONTACT CONDITION
Figure 2.2-20.
Graph giving the leakage rate
through a FML hole in a composite
head i s 3 m (10 ft? and the hole
diameter of 11.3 mm)
analytical analyses
soil conditions and
soil, only a range
conditions can
(1) best, i.e.
in case of leakage
liner. The hydraulic
area is 1 cm2 (i.e.,
. Because of uncertainties in th^
as well as the large influence of
contact between the FML and the
of values can be given. Field
be anywhere between the two extremes:
the soil is well compacted, flat and
smooth, has not been deformed by rutting during
construction, and has no clods and cracks, and the FML
is flexible and has no wrinkles; and (2) worst, i.e.,
the soil is poorly compacted, has an irregular surface
and is cracked, and the FML is stiff and exhibits a
pattern of large, connected wrinkles. Abbreviations:
GOOD and POOR = good and poor field conditions; MIN,
P.C., TEST, and MAX are defined in Table 2.2-11.
2.2-95
-------
CONDITION
Figure 2.2-21.
I'Vf.) (IT ) (rfSl) fflb f
FML - SOIL CONTACT
Graph giving the radius of the wetted area in case of
leakage through a FML hole in a composite liner. The
hydraulic head is 3 m (10 ft) and the hole area is 1
cm2 (i.e., diameter of 11.3 rr.m). Because of
uncertainties in the analytical analyses as we i 1 as the
large influence of soil conditions and contact between
the FML and the soil, only a range of values can be
given. Field conditions can be anywhere between the
two extremes: (1) best, i.e
compacted, flat and smooth, ha
rutting during construction, a
cracks, and the ri-(L is flexible
and (2) worst, i • . , the soi 1 is
an irregular sunace and is cracked, and the FML is
stiff and exhibits a pattern of large, connected
wrinkles. Abbreviations: GOOD and POOR = good and
poor field conditions; MIN, P.C., TEST, and MAX are
defined in Table 2.2-11.
2.2-96
the soil is well
.t been deformed by
ml has no clods and
and has no wrinkles;
poorly compacted, has
-------
2.3 LEAK DETECTION TECHNOLOGIES
2.3.1 Review of Available Technologies
2.3.1.1 Introduction
The technologies available 'or leak detection at land disposal
units fall Into two categories: (1) drainage layer technologies using
the- leachate collection and removal system between the top and bottom
liners as the leak detection system; and (2) innovative technologies
Involving the use of remote sensing techniques, hereafter referred to
as geophysical leak detection systems.
The use of leachate collection and removal systems as leak
detection systems 1s discussed in Section 2.3.1.2. Three geophysical
methods are promising for use as leak detection systems and are
described and evaluated in Section 2.3.1.3 through 2.3.1.5. These are
electrical resistivity, time domain reflectometry, and acoustic
emission monitoring. Other less developed innovative technologies are
briefly discussed 1n Section 2.3.1.6.
2.3.1.2 Leachate Collection and Remova] Systems
The concept of using the leachate collection and removal system
(LCRS) between the top and bottom liners for leak detection is that by
monitoring the liquids that accumulate in the LCRS sump, the presence
of leaks can be detected. This method of leak detection has several
attractive features. In addition to providing leak detection, the
method provides Information on the volume of leakage collected and, if
the leakage Is chemically analyzed, information on its chemical
constituents. The LCRS technology 1s proven through in-field use and
the necessary components are readily available. The components of the
LCRS are also durable. It is a direct method of detection which does
not require sophisticated data interpretation. In addition, since a
LCRS between liners is already a statutory requirement of RCRA, no new
component is added to the lining system.
2.3-1
-------
F1
2.3.1.2.1 Principles
A leac.nate collection and removal system between the top and
bottom liners consists of a layer of drainage media designed and
constructed to transmit flow laterally, between the liners, to
collection points. The drainage layer can consist of either granular
material:) or synthetic materials. Detection is possible with such
systems because the layer between the liners functions as a drain. To
function as a drain, the layer between the liners must possess a much
greater capability of transmitting flow than the bottom liner. This
is possible only if the layer (drainage layer) Is much more permeable
than the bottom 1iner.
The mechanism by which lateral flow may occur in the drainage
layer Is as follows. Liquid entering the drainage layer will flow
vertically (with some lateral wetting) until the less permeable bottom
liner is encountered. Once encountered, the predominantly vertical
flow will be retarded and liquid will begin to accumulate on the
liner. Vertical flow will continue at a rate determined primarily by
the hydraulic conductivity of the bottom liner. Lateral flow will
begin when a threshold level of saturation above the bottom liner is
reached such that the gravitational force on the liquid is greater
than the capillary and adsorptive forces of the drainage layer media.
If, on the other hand, the rate of infiltration into the bottom liner
is faster than the rate of liquid accumulation above the bottom liner,
then lateral flow will never occur. Flow in a porous drainage layer
is governed by Darcy's equation for both saturated and unsaturated
conditions [Todd, 1980]. Flow is a function of hydraulic
[ transmissivity, hydraulic gradient and width of the cross-sectional
area in the direction of flow [Figure 2.3-1]. Hydraulic
transmissivity is the prcHuct of the in-plane hydraulic conductivity
and the saturated thickness of the drainage layer. In-plane hydraulic
conductivity is a property of the drainage material and the liquid.
i This property is typically evaluated under saturated flow conditions.
Under unsaturated flow conditions, capillary and adsorptive forces
restrict the movement of Mquids in porous materials and therefore
decrease hydraulic conductivity. The lower the degree of saturation,
I. the lower the hydraulic conductivity of the material, other factors
1 being constant.
i
j
[ 2.3-2
L
-------
The hydraulic gradient in a satu, ; analysis.
The rapidity with which a LDCRS will permit leak detection is
reflected by the detection time calculated for the leak detection
system. This LDCRS performance characteristic refers to the time
between when leakage passes through the top liner into the LDCRS, and
the time when it is available for detection in the LDCRS sump. For a
given type of leak and rate of leakage, the detection time is
primarily dependent on the hydraulic conductivity and thickness of the
2.3-3
L
-------
I
i
> drainage media in the LDCRS, the capillary stresses in the d~iir£:e
• media, the slope of the LDCRS and the length of the drsinaqe pa'.-.
Through proper design and selection of materials, leak detectic" ives
> in the range of a day can be achieved. (This represents B^AT ':r
I LCRS-type leak detection systems, as discussed in S.-tior. 2.7.)
I Detection of leakage witnin a few days or less of a leak occurrence
I provides a system with minimal potential for leakage frcm the la*d
di-^osal unit prior to leak detection. On this basis, it is conduced
that leak detection times of a few days satisfy the criterion that t-e
leak detection system detect leaks rapidly. Design of the LCCRS to
meet the performance standards described In this paragraph is
discussed in Section 2.8.
The ability to locate a leak within a LDCRS is limited and deperis
on the design of the system. The location of a leak can be narrc>?1
to a zone if the system is divided into separate collection a-d
removal zones. For a given unit, the size of zones is usual'y
dictated by design and economic factors and site conditions. However.
even with zoning, ic is clear that these systems have limited
capability for precisely locating leaks. However, as noted in Sectira
2.1.2.4.3, leak location identification is usually not very important.
LDCRS are reusable because the occurrence of a leak does n:t
physically damage the system. However, .he sensitivity of the systn
for detecting multiple leaks in the same collection sump is less th'-s
the sensitivity for a single leak.
LDCRS are highly reliable systems that can be expected to fmctici
well over time with minimal maintenance. Their primary con-pone'!:
(i.e., the drainage media) are comprised of relatively inert
materials, such as lean silica sands or high density polyethylene
(HOPE) synthetic drainage materials.
LDCRS can function with minimal maintenance and at the same tire
provide continuous monitoring of top liner leakage. Continuous
monitoring of top liner leakage is inherent to the sytem because
leakage entering the system is automatically transmitted by the
drainage layer to the LDCRS sump for monitoring. Leakage ~iy thus t?
monitored by only performing periodic measurements and recorcs of
liquids at the LDCRS sump.
2.3-4
-------
r
! LOCRS are capable of detecting leaks In all areas of the bottCM
and sidewalls of a waste management unit. This capability results
from the consideration that drainage layers of LDCRS are required to
| be installed under all areas that are likely to be exposed to the
•• waste or its leachate.
i
\ 2.3.1.3 Electrical Resistivity
j _ — —
' This section has been excerpted from GCA [1984b]. It has been
; reproduced here with modifications and additions.
i
i
Electrical resistivity is a geophysical technique whereby an
electrical current is Introduced into the ground by a pair of surface
electrodes and the resultant potential field, as measured by a second
pair of electrodes,1s interpreted to detect anomalies (leakage).
Dry soil and rock materials are typically highly resistive (resist
the flow of electric current). As the moisture content of soil or
rock increases, the resistivity generally decreases significantly
(I.e., the material becomes more conductive), and as various salts and
free ions become Incorporated into ground water, the resistivity
decreases even further. Because of this phenomenon, resistivity has
been used successfully to locate and map leachate contaminant plumes
[Benson et al., 1982].
The potential application of the electrical resistivity technique
to leak detection 1n FHL-lined waste management units has been
Investigated by EPA at the Southwest Research Institute (SRI) [Schultz
et al., 1984] using the pole-dipole resistivity technique. More
recently, Foote Mineral Compa-v [1986] has proposed a patented leaV
detection technique based on tne dipole-dipole resistivity technique.
The discussion that follows is based primarily on data from studies
conducted at SRI. However, recent information from Foote indicates
promise for Foote's ER technique as well.
2.3.1.3.1 Principles
Because of their high electrical resistance, FMLs act as
electrical Insulators between the waste material above and the LDCRS
below the FML. A failure in the FML establishes a path of low
2.3-5
L
-------
resistance which can be detected and located with measuring devices at
the surface.
Resistivity measurements are taken using two current electrodes
and two potential electrodes which may be placed in different
configurations. The electrical resistivity (ER) method used by SRI to
detect leaks from waste management units lined with FMLs utilizes the
pole-dipole electrode configuration illustrated 1n Figure 2.3-2. A
current source electrode is placed inside the waste area and a remote
current electrode 1s placed outside of the area at a remote distance.
This distance should be at least 50 times the distance between the two
potential electrodes [Waller and Davis, 1982]. One of the potential
electrodes is used as a reference point and remains stationary while
the other potential electrode is positioned at preselected measuring
points over the waste site. Theoretically, a fixed array of potential
measurement electrodes could be installed where frequent or continuous
monitoring 1s desired. However, a fixed array arrangement would
require such sophisticated wiring and data interpretation systems that
the practicality of the arrangement is questionable.
A direct current (DC) or low frequency alternating c rent (AC)
Injected at the source electrode flows to the remote electrode (Figure
2.3-3) The flow of current will be inhibited by the FML except c r
the upper edges of the liner and at leak points. Since surface
potential measurements are affected by the current distribution, plots
of equipotential lines will show distortions in the at ea of the leak.
The use of computers for recording and processing data has greatly
reduced the time required for data interpretation.
2.3.1.3.2 Recent Studies
Electrical resistivity (ER) techniques have
-------
t\
The ER system used by SRI is illustrated in Figure 2.3-4. The
current source electrode, remote current electrode, and reference
electrode remained stationary during the surveys. The potential
measurement electrode was attached to cable which was positioned
over preselected measurement lines radioing from the current source
electrode to the edge of the impoundment. The selected lines were 5
or 10 degrees apart. During actual measurements, a motorized winch
reeled the electrode from the edge of the Impoundment across the water
surface back to the current source electrode. Figure 2.3-5 shows
equipotentlal plots generated by SRI for (a) no leaks, and (b) a 0.3 m
(1 ft) diameter leak in 1.5 m (5 ft) of water.
Testing of the ER method by SRI Indicates that the technique is
highly sensitive to leakage from surface impoundments. The presence
of a 25 mm (1 1n.) diameter leak was detected in the lined test
impoundment. In addition, the position of the leak was located to
within one foot of the actual location in 1.5 m (5 ft) of water.
Several other experiments were conducted at SRI.
demonstrated the following capabilities of the ER system:
They
• An increase in water depth from 1.5 to 2.0 m (5.0 to 6.5 ft)
reduced the surface potential voltages near the leak but did
not hinder data interpretation.
• The distortion of equipotential lines produced by a 9.1 m (30
ft) long simulated -ear leak was readily apparent and
significantly different from equipotential lines produced by a
point source leak.
• The locations of three simultaneous leaks ranging in size from
2.5 to 30 cm (1 to 12 in.) were identified.
2.3.1.3.3 Evaluation
The field experiments conducted at SRI demonstrated the potential
of the ER system for leak detection at surface impoundments lined with
a single FML liner. The potential for applying this technique to
actual hazardous waste management units is evaluated hereafter.
2.3-7
-------
In Section 2.1.4.2.3 it was noted that the Important performance
requirements for leak detection systems are that they evaluate the
leakage rate and provide the information rapialy. With ER techniques,
it is not possible to evaluate the leakage rate. This is a drawback
of ER-type systems. ER techniques can be used to detect leakage
rapidly If the ER system is used as a continuous monitoring system or
If ER surveys are done frequently. With only portable equipment
however, ER surveys are time consuming and expensive. An advantage of
ER techniques 1s that they can be used to determine leak location.
This Is especially advantageous in surface impoundments (see Section
2.1.4.2.3).
With respect to leak location, SRI researchers demonstrated that
the ER systems can be sensitive to a 2.5 cm (1 in.) diameter leak in a
0.4 hectare (1 acre) impoundment, and that the leak can be located
accurately to within 30 cm (1 ft). A 2.5 cm (1 in.) diameter leak is
the smallest leak simulated in the SRI tests and is not necessarily
the smallest leak the system is capable of detecting. However, it is
clear that liquid transported through the liner as a vapor would not
be detected by the ER technique because no current path would be
produced by the vapor.
While the results cf the research at SRI are impressive, there are
a number of factors which could be expected to interfere with the
performance of the system under routine use. These are addressed in
the following paragraphs.
The electrical resistance properties of the liner materials used
are of critical Importance to the performance of the ER method. The
primary liner must be composed of a highly resistant material such as
an FML. The method cannot be used to detect leaks from an impoundment
with a single or top liner consisting of compacted soil alone due to
Its comparatively low electrical resistance. In double liner systems,
the composition of the drainage layer and the bottom liner are also of
Importance. Dry sand, gravel, and synthetic drainage materials all
exhibit relatively high electrical resistance. When wetted, the
resistance Is lowered significantly. But current will flow to the
remote current electrode along the leak path only if the bottom liner
has a 'ow electrical resistance (e.g., consists of compacted soil
alone) or if the bottom liner is also leaking. A compacted soil
2.3-8
-------
bottom liner would allow the leak current path to be established while
an Intact composite bottom liner would not. Therefore, ER is not
applicable to facilities with a composite bottcm liner.
The application of the ER technique to landfills has not been
demonstrated on a large scale. SRI tested the method on a 3 m by 3m
(10 ft. by 10 ft.) scale model which was lined with a 0.15 rrm (6-mil)
thick black polyethylene sheet and filled with soil. Soil depth was
varied from 7 to 14 cm (2.75 to 5.5 1n.), and simulated leaks 15 cm
(0.5 ft) 1n diameter were accurately detected and located [SchuKz et
al., 1984]. This experiment demonstrated that the ER technique could
potentially be used for detecting leaks in a homogeneous, solid
material.
The performance of resistivity techniques is affected bj
: interferences such as buried metal pipes, metal fences, railroad
;j tracks and stray currents from power lines. Highly conductive waste
i materials have the potential for limiting the effectiveness of the ER
|j technique at waste management units [Waller and Davis, 1982]. Some
limited experiments have been conducted at SRI with highly conductive
metallic objects placed at various locations within the 3 m (10 ft)
scale model filled with water. Their studies suggest that the
metallic objects have no significant influence on the distribution of
, equipotential values. More research is needed in this area to assess
| the effects on equipotential measurements from heterogeneous deposits
I as will probably occur in any landfill.
I-
[
', The ER leak detection system used at SRI is not a continuous
monitoring system but rather a survey technique. The equipment must
be set up and operated, and potential measurements must be interpreted
each time leak information is sought. Detection time, therefore, is a
; function of the technique but, more importantly, is a function of how
'; often the technique is used.
SRI reports that it takes 8 hours to set up equipment, survey, and
: analyze the results from the one-acre test impoundment [ENR, 1984].
'• While the set-up time would remain fairly constant as the unit size
• Increased, survey and data interpretation time would be expected to
Increase proportionally to size.
2.3-9
L.
-------
n
At some hazardous waste management units, frequent or continuous
leakage monitoring may be necessary. An ER leak detection system
could theoretically je installed permanently at such units by
replacing the mobile potential measurement electrode with a set of
stationary potential measurement electrodes. Equipotential plots
would be constructed from the potential readings at each electrode.
The resolution of the system would be a function of the spacing of
measurement electrodes. While the cost of equipment for a continuous
monitoring system would be sharply higher than for a portable system
due to the need for multiple stationary potential electrodes, time for
detection would be reduced and labor associated, with mobilization and
demobilization would be eliminated. Data interpretation, on a unit-
cost basis, would be unchanged.
The ER leak detection system does not require the installation of
below-ground equipment at the hazardous waste management unit. The
electronic equipment needed, whether portable or permanently
Installed, 1s easily accessible for inspection and maintenance
purposes.
The current source, reference potential, and potential measurement
electrodes all come Into direct contact with waste material. The
material used in these electrodes must be resistant to chemical
degradation and at the same time possess the required electrical
properties. In the SRI study, brass was selected for the current
source electrode, and copper for the reference potential electrode.
The reliability of the ER method of leak detection, as
demonstrated at the test site, is high, provided interferences are
minimal, electronic equipment 1s functioning properly, operators
follow prescribed procedures, and data Interpretation is performed by
trained personnel. These conditions can be met with the
Implementation of a strict program of quality assurance/quality
control.
2.3.1.4 Time Domain Reflectometry
This section has been excerpted from GCA [1984b]. It has been
reproduced here with modifications and additions.
2.3-10
-------
Historically, the Time Domain Reflectometry (TDR) technique has
been used in the communications industry to locate fault in long
transmission lines. Recently, TDR has been used to map soil moisture
content along buried electrical transmission lines. The latter
application has shown promise for early detection of liner failures at
hazardous waste landfills and surface impoundments and has been the
subject of a numbe. of recent studies [Davis . al., 1984, 1983a,
1983b; Huck, 1982; Waller and Davis, 1982]. None of these studies has
been conducted with actual waste containment :^tes.
2.3.1.4.1 Principles
The TDR technique measures the electrical property variations in
soil along a pair of parallel transmission line conductors (Figure
2.3-6). A high frequency step-pulse is Introduced by the TDR system
which sets up both an electric field and a magnetic field between the
two conductors. The soil 1n the zone between the two conductors thus
becomes part of the transmission pathway.
The electrical properties of soils are different from water. The
dielectric constant for dry soil ranges from 2 to 5, while that of
water is about 80. Because of this rather large contrast, the TDR
system 1s sensitive to soil moisture content, and is relatively
Insensitive to soil type. These properties make the TDR system
potentially attractive for use under FML liners installed at hazardous
waste rinagement units (Figure 2.3-7).
2.3.1.4.2 Recent Studies
Several EPA-funded studies nave been conducted to test the
efficacy of TDR as a leak detection technique. These studies have
ranged from scale-model laboratory studies to limited field studies on
a model waste Impoundment site [Davis et al., 1984, 1983a, 1983b;
Huck, 1982]. Results have shown that leak detection by TDR Is
technically feasible with presently-available instrumentation.
Transmission lines have been tested which varied from 2.0 to 20 m (6.6
to 65.6 ft) In lenyth although theoretically 1t Is possible to monitor
up to 40 m (131 ft) of cable with present instrumentation.
Transmission line separation in these experiments varied from 0.1 to
2.0 m (0.3 to 6.6 ft).
2.3-11
-------
Some of the EPA findings include:
• Dry sandy soils (as would be present in a drainage layer) ta/e
dielectric properties similar to air and electromagnetic enetgy
losses are not significant. Thus, these materials are suitable
for use as a host material beneath an FHL liner. The host
material thickness should be at least that of the transmission
1 ine spacing;
• The ability of a TOR system to measure changes in dielectric
properties of the host material 1s decreased as the spacing
between the transmission line conductors s increased;
• The minimum detectable leak size that can be detected Increases
with increasing transmission line conductor spacing; and
• Present technology would permit the design of a dedicated TDR
leak detection system along 1000 m (328 ft) transmission lines
capable of detecting a 0.1 m2 (1.0 ft2) leak, although such
resolution would be difficult.
2.3.1.4.3 Evaluation
The TDR technique tested showed promise in allowing accurate
definition of the location and approximate size of a leak in a timely
manner. Present TDR technology enables the design of a leak detection
system that will detect and locate a leak i,i an area of 0.1 m2 (1
ft'). This would require a conductor separation no larger than 0.6 m
(2 ft.), although 0.3 m (1 ft.) would be better. Detection of a leak
of this size 1s subject to certain assumptions concerning the
dielectric properties of the dry host material and those same
properties when saturated by the leachate or impounded liquid, as
discussed later in this section.
If a TDR leak detection system is monitored, it will have a rapid
response time which may be on the order of minutes in the event of top
liner failure, providing the leak is large enough to be detected.
However, because monitoring is expensive, monitoring will typically be
much less frequent.
2.3-12
-------
r
The most critical aspects to be considered in the design of a TDR
system are the physical and electrical properties of the granular-
drainage media and the geometry and spacing of the transmission line
conductors. Each site will be different, depending on the size,
materials available, and specific performance criteria required of the
leak detection system employed.
; The designer must know the minimum s'ze of leak which must be
j detected to design a TDR leak detection system. The transmission line
j conductor separation must be no greater than twice the diameter of the
! "design leak". Sufficient contrast in the dielectric constant must
ex'st between the drainage media hosting the transmission line
conductors and the leachate or liquid to be detected by the TDR
technique. In addition, the granular drainage layer must be at least
' as thick as the separation between transmiss'on line conductors. For
this reason, TDR cannot be used with synthetic drainage layers.
The water content of the drainage media when installed must be
sufficiently low so that an adequate contrast in moisture content
exists once a leak develops. During the model waste impoundment
study conducted by EPA [Davis et al., 1983a] the sand drainage media
1n the test section had to be dried three times over a period of
several months before they were placed.
Water content considerations underline the importance of selecting
the proper materials for the drainage layer. A material with a low
specific retention Is necessary to avoid costly drying operations.
Specific retention 1s defined as the ratio (expressed as a percentage)
of the volume of water a material will retain after it has been
saturated and allowed to drain. A coarse sand with a porosity of 40
percent may have a specific retention of 5 percent. A silty coarse
sand may have a specific retention of 15 percent or greater.
Low specific retention materials are also desirable for
reusability. If a leak develops in the primary liner of a surface
impoundment, a TDR system may become functional again after the leak
has been repaired and most or all of the leakage has drained by
gravity.
i
2.3-13
-------
The drainage layer should be made up of a wel 1-compacted modium-
to fine-grained sand with sufficient fines to horizontally disperse
the wetting front of a leak thus increasing the TDR response [Davis et
al., 1903a]. Horizontal dispersion of the wetting front will produce
a wetted zone much larger in horizontal cross-sectional area than the
actual leak. The drainage layer should not contain too many fines,
however, as silts and clays rapidly attenuate the TOR signal. The
apparent dilemma created by a simultaneous need for low specific
retention and a good TDR signal transmission on the one hand, and
horizontal dispersion on the other points out the need for further
study. In addition, horizontal dispersion reduces leakage collection
at the LCRS sump.
The transmission line conductors should be approximately 2.5 mm
(0.1 In.) diameter stranded copper [Davis et al., 1983a]. Wires of
this diameter are not easily damaged during Installation and have
performed well In this application. However, the potential for wire
corrosion Is uncertain. Durability Is, therefore, a concern with this
method.
Data acquisition would take place at an interval determined by the
nature of the Impounded waste materials and the potential impacts
involved in Impoundment failure. Design of a continuously monitored
TDR :-ystem is feasible, but expensive. Such a system might involve
manual or automatic switching from one transmission line pair to the
next so that a central monitoring device could access information from
any portion of the site. Newly-acquired data from each transmission
line could be compared (manually or by computer) with an earlier data
set from the same transmission line. Any differences between the
earlier and new data sets would be evaluated in the context of
possible top liner failure.
2.3,1.5 Acoustic Em1ss1ons_MonHoMng
This section has been excerpted from GCA [1984b]. It has been
reproduced here with modifications and additions.
Acoustic emission monitoring (AEM) is based on the. concept of
detecting vibrations (acoustic emissions) produced by -l-eakfrig liquids
through the use of highly sensitive piezoelectric sensors
(transducers). The potential of the technique as a means of detecting
2.3-14
-------
and locating leakage beneath FML liners was Investigated following
successful attempts to locate turbulent flow and seepage through earth
dams. [Davis et al., 1983a; Koerner et al., 198-1].
To date, the technology has not be = n proven at a full scale
containment site, although installations jf AEM systems have been
planned at several of such sites.
2.3.1.5.1 Principles
Acoustic emission monitoring relies on the fact that turbulent
" )w with a velocity greater than 1 mm/s (0.04 1n./s) through soil
creates low frequency acoustic vibrations that can be detected and
monitored [Koerner et al., 1984; Davis et al., 1983b].
Monitoring equipment consists of a highly sensitive transducer (a
microphone or an undamped accelerom°.ter) connected to an amplifier
with an adjustment band-pass filter which provide a signal suitable
for subsequent Interpretation. Interpretation may be done manually by
listening to an audible output or viewing a visual display, or
electronically by an electronic spectrum analyzer. Portable single-
channel field equipment is available which permits use of earphones to
monitor acoustic emissions. This equipment also provides a strip
chart record of monitored events. Output connectors are provided for
additional recording, analysis, and display devices. Interpretation
could be done by automated, continuous monitoring of multiple sensors
at large scale Installations.
2.3.1.5.2 Recent Studies
The potential of AEM for leak detection at lined containment sites
has been evaluated by various researchers [Koerner et al., 1981, 1984;
Waller and Davis, 1982; Huck, 1982; Davis et al., 1983a, 198Jb].
Koerner et al. [1981 and 1984] conducted a series of experiments which
investigated the use of different sensor configurations as well as the
effect of flow rate on acoustic emissions and their characteristics.
They observed that increased flow rates through sand and gravel media
increased acoustic emissions and that these emissions showed an
Initial peak and then decreased to less than a threshold value after
several minutes. In addition, they found that use of turbid water (38
2.3-15
-------
r
I g/liter clayey silt) increased the sensitivity to flow detection.
• Sensors located directly beneath a FML liner proved to be effective.
f Koerner et al. reasoned, however, that the cost of the large numbei of
sensors required for a full-scale installation would be prohibitive
[ and Investigated the use of waveguides for transmission of raw signals
\ to the sensor.
I Davis et al., [1983a, 1983b] Investigated many of the variables
believed to have possible Influence on the effectiveness of AEM for
I leak detection at containment sites. Davis confirmed the generation
of acoustic emissions by liquid flow through sand and gravel media and
concluded that the sounds have significant amplitude at frequencies up
to 500 Hz and peak amplitude 1n the range of 100 to 200 Hz. The work
also showed that sound amplitude increased with liquid velocity and
with increased variation in soil grain size. However, sound
amplitudes were observed to decrease to background levels after about
one minute of water flowing through soil. The sound was believed to
have been caused by collisions between soil particles and by air
bubbles moving through the soil. Davis et al. further concluded that
water alone flowing through soil does not produce significant sound
energy. For sounds to be produced the turbulent movement of water
mixed with fine grained soil or air through soil is required. The
range of the sensors used in these experiments was about 1 meter.
Davis concluded, as aid Koerner, that emplacement of individual
sensors at a containment site of four hectares (10 acres) would be
i Impractical.
I
[ 2.3.1.5.3 Evaluation
t
I Laboratory and controlled field experiments have demonstrated the
r ability of AEM to detect leakage through FML liners under specified
j. conditions. Proper spacing of sensors offers the potential to detect
and locate leaks witi a resolution of ±1 m (3.3 ft) 1n the case of
i. individual sensors and with a resolution of ±0.3 m (1 ft) in the case
! of wire waveguides laid out 1n a grid. However, this technology has
I not been proven at a full scale installation.
I AEM sensing equipment appears compatible for installation between
| double liners at newly constructed containment sites. Either
I individual sensors or wire waveguides could be Installed in a granular
I
! 2.3-16
-------
/
drainage layer. Sensor leads would then be routed between liners to
the ground surface at the edge of the containment basin for subsequent
connection to monitoring and interpretation equipment. Figure 2.."1 ^
shows the installation of an AEM sensor in tne granular drainage la
between two FMLs. Because of their size (current design is about 23
mm (0.9 In.) diameter X 200 mm (7.9 in.) length), individual sensors
probably cannot be placed between FHL liners separated only by a
synthetic drainage layer. A wire waveguide grid may be more suitable
for Installation 1n this instance. The waveguides would be equipped
with sensors at one end for connection to monitoring and processing
equipment. Use of wire waveguides would have several advantages over
Individual sensors:
• Greater resolution in locating a leak may be obtainable.
• The number of sensing devices required may be reduced
significantly. Thus capital costs may be reduced accordingly.
• Use of wire waveguides would permit locating sensors should a
malfunction occur, in contrast to the inaccessibility of
embedded sensors.
Wire waveguides may be used
located between two FMLs.
with synthetic drainage layers
A number of drawbacks exist however, with the use of AEM in
hazardous waste management units. These ?.re:
• Individual sensors and leads buried between two liners must
retain functional ability without repair during the units'
active life and the 30 year post-closure care period. Should a
malfunction occur in one sensor, repair would be very
difficult. The use of wire waveguides with serviceable sensors
reduces this concern somewhat; however, selection of waveguide
material would still need to be done with full consideration of
the potential for corrosion.
• AEM appears limited to the detection of leaks which result in
turbulent flow and collisions between soil particles. Small
leaks that increase in size gradually, and low velocity seepage
2.3-17
-------
are likely to be undetected. In addition, the le k must be
detected within a few minutes of occurrence before sound
intensl1".1 diminishes to threshold values. This would
necessiiuie design of an "on-line" continuous monitoring system
to reduce the likelihood of missing a leak. High speed
automatic electronic switching and processing would be a
necessity, especially at large facilities.
AEM equipment Is sensitive to site background noise. Sensors
placed at the end of wire waveguides would be susceptible to
airborne as well as seismic noise. Individual sensors placed
between liners would be protected from airborne noise sources;
but may ba sensitive to noises caused by nearby equipment or
machinery.
2.3.1.6 Other Leak Detection Technologies
This section has been excerpted from GCA [1984b].
reproduced here with modifications and additions.
It has been
Leak detection technologies other than those described in the
previous sections are not sufficiently developed to be considered as a
primary alternative for leak detection in hazardous waste management
units. These technologies may be useful as a confirmatory or backup
techniques. A brief description follows on the applicability of each
of these technologies.
2.3.1.6.1 Lyslmeters
Lysimeters are used to collect liquid samples from the zone of
aeration or vadose zone in soils to monitor for the presence of
leachate plumes. Monitoring wlliiin the vadose zone is desirable
because it allows early detection of contaminants before significant
contamination to soil and groundwater has occurred.
These are two common types of lysimeters: pressure-vacuum or
suction lysimeters, and collection or trench lysimeters. Experience
with both pressure-vacuum [Parizek and Lane, 1970; Apgar and Langmuir,
1971; Gerhardt, 1977; Johnson and Cartwright, 1980; Morrison and Ross,
1978] and collection lysimeters [Kmet and Lindorff, 1983] for leak
detection at containment sites has been reported.
2.3-18
-------
Pressure-vacuum lysirrsters (Figure 2.3-9a) consist, of a p. jus cup
or sleeve attached to a small receiver vessel. The lysiineter is
installed at the location and depth where a sample of soil water is
desired (typically in a vertical borehole). The lysiineter is
connected to the ground surface by two small diameter tubes. Sample
collection 1s achieved by applying a vacuum to the lysimeter through
one of the access cubes. Soil moisture held in tension then moves
along the pressure gradient created by the vacuum, through the porous
cup, and collects Inside the lysimeter. The sample is transferred to
the surface by releasing the vacuum and applying pressure to one of
the access tubes. The water sample 1s forced to the surface through
the other tube.
Pressure-vacuum lyslmeters have a history of problems with
clogging [Waller and Davis, 1982]. In addition, the porous ceramic
cup typical of many lysimeters has been shown to affect sample
Integrity [Hanson and Harris, 1975]. It is now possible to purchase
pressure-vacuum lyslmeters manufactured entirely from PVC or Teflon.
This should reduce or eliminate sample alteration problems associated
with ceramic cups. In addition, greater strength and a lesser
tendency to clog Is reported for the Teflon cups [Castro and Timnons,
1983]. To reduce or prevent clogging, pressure-vacuum lysimeters are
packed with fine silica sand during installation (Figure 2.3-9b).
Collection, or trench, lyslmeters intercept and collect water as
1t percolates through the unsaturated zone and consist of a lined
basin or trough filled with a well-drained material and placed below
the area to be monitored (Figure 2.3-10). They are connected to a
sample collection well or manhole. As in trie case of pressure-vacuum
lyslmeters, leachate would be Identified by analysis of line collected
sample for appropriate parameters.
Use of lyslmeters between double liners is redundant, since use of
a double Uner system with leachate collection between liners is the
equivalent of Installing a single collection lysimeter or Innumerable
pressure-vacuum lysimeters under the entire containment site.
Moreover, pressure-vacuum lysimeters collect a sample from a very
small region 0.03 to 0.06 m' (typically 1 to 2 ft') making it
difficult and expensive to pinpoint the location of a leak. For this
2.3-19
-------
* JM
r
reason, the utility of lyslmeters as a leak detection alternative i?
1 iml ted.
2.3.1.6.2 Seismic Measurements
Seismic measurements may be considered as a possible alternative
for leak detection, although such systems would be costly and of
questionable reliability. Se'smic measurements are generally made to
determine the thickness and depth of geologic layers and the velocity
of seismic waves within the layers. Seismic measurements are commonly
done at the surface, although they can also be made in boreholes
(surface to borehole or borehole to borehole). None of these
variances can be Implemented in a thin drainage layer without extreme
difficulty.
A feasible alternative consists of installing a network of
seismometers within a sand drainage layer below a synthetic liner to
measure the seismic velocity in the sand layer. This velocity is
expected to Increase significantly when that layer, or a portion
thereof, becomes fully saturated with leachate. But factors such as
implementabi 1 i ty (up to thousands of seismometers per acre),
generation of appropriate seismic signals to detect a saturated
condition in a relatively thin medium (no more than several feet
thick) lead to the belief that such a system would be prohibitively
expensive and of questionable reliability.
2.3.1.6.3 Electromagnetic Techniques
Electromagnetic techniques include mutual inductance, ve-y low
frequency and fvigh frequency techniques. Each of these has limited or
no application to the problem of leak detection and 1s discussed
briefly below.
- Mutual Inductance
This technique, also referred to as terrain conductivity
measurement, measures the apparent conductivity of a volume of earth
material between the transmitter and receiver coils. The depth of
penetration 1s a function of the orientation of the coils, the
separation between them, and the frequency of the electromagnetic
2.3-20
L
-------
field generated by the transmitter. Measurements are made at the
surface and have been used traditionally for mineral exploration and,
more, recently to trace contaminate migration at waste disposal sites.
The technique Is considered to be semi-quantitative and is probably
not capable of detecting infi.tration of leachate or impounded '.quids
Into a thin drainage layer.
- Very Low Frequency
The very low frequency (VLF) technique has traditionally been used
for mineral prosoecting and utilizes the 10-20 KHz electromagnetic
waves generated by the U.S. Navy for global submarine communications.
The VLF technique transmitters measure the electrical properties
(inductive conductivity) of the earth materials In the vicinity of the
t Instrument. The depth of penetration can be varied by utilizing
| transmitters with different frequencies. VLF shows promise for plume-
! tracing [Waller and Da\ s, '"'SZ], but its capability for detecting
1 leachate infiltration Into a tnin drainage layer remains questionable.
- High Frequency
; These techniques may be subdivided into pulsed and continuous wave
! systems and operate by transmitting energy between frequencies of 1
| MHz and 1000 MHz into the materials being tested.
I The pulsed wave system is more commonly referred to as ground
probing radar (GPR), and is usually operated from the surface.
Discrete pulses of energy are transmitted into the material to be
tested. The electromagnetic waves emitted penetrate the material and
are reflected from boundaries across which there is a change in the
i I dielectric properties. The reflections are detected by a receiver and
;• i are displayed on a strip chart recording. Data may also be recorded
!' on magnetic tape for later pullback or signal enhancement.
i
! For the continuous wave system, the signal transmitter operates
:- continuously making it more effective in applications where the
receiver can be Isolated from the transmitter by the material to be
: tested (as In adjacent boreholes).
2.3-21
-------
The two high frequency systems tend to be expensive, both in terms
of the electronics required, and the skilled staff necessary for data'
acquisition and interpretation. Both systems suffer from signal
attenuation in the presence of conductive materials, as might be
present at a waste repository. Accordingly, both high frequency
techniques would seeii to have very limited application to the problem
of leak detection.
2.3.1.6.4 Moisture Blocks
The use of moisture blocks to meagre changes in soil moisture
content has been widespread for many years, particularly in
agricultural applications. A typical block consists of two electrodes
embedded In a porous block of gypsum 16 cm' (0.4 in1) in size. The
blocks are responsive, typically within hours, to a wide range of
changes in the electrical resistance between the embedded electrodes.
The soil moisture content can be determined if the block has been
calibrated to the particular soil in which It 1s placed.
A network of moisture blocks embedded in z. drainage layer beneath
a FML at a waste repository is feasible. Such an installation would
probably serve as a reasonably reliable leak detection system. A
recent leak detection experiment using moisture blocks embedded in a
compacted subgrade demonstrated the efficacy of moisture blocks in
this application [Myers et al, 1983],
Moisture blocks have drawbacks. They are fragile and require
careful Installation to avoid damage. Moreover, a moist environment
may eventually lead to dissolution of the gypsum and system failure.
A 2- to 8- year lifetime is often quoted by manufacturers for
agricultural applications. Therefore, the reliability of the system
may become questionable with time.
2.3.2 Selection of Leachate Collection and Removal System as Leak
Detection System
Several technologies for leak detection were reviewed and
evaluated in Section 2.3.1. These are: (1) drainage layer technology
which utilizes leachate collection and removal systems (LCRS) between
the liners as leak detection, collection and removal systems (LDCRS);
2.3-22
-------
and (2) Innovative technologies which involve electrical resisti;ity
(ER), time domain reflectometry (TOR), acoustic emission monitoring
(AEM) and other techniques. This review and evaluation showed that,
for general usage, drainage layer technology currently has significant
advantages over other innovative (and sometimes promising)
technologies. Due to these general advantages, drainage layer
technology Is the method advocated to satisfy the statutory leak
detection requirements of RCRA. A summary of leak detection
technologies is provided below,
2.3.2.1 Drainage Layer Technology
LDCRS have several distinct advantages, for general application at
all land disposal units, over the other methods evaluated.
• Properly designed LDCRS can provide rapid leak detection;
• LDCRS provide information on the volume of leakage. In
addition, if leakage is analyzed, information on chemical
constituents can be obtained. Measuring the quantity of
liquids in the LDCRS requires generally unsophisticated
equipment and can be done on a daily basis with relatively
little effort. The leakage rate can be estimated by simple
liquid level measurements at the LDCRS sump.
• LDCRS is a proven technology that has been demonstrated through
actual field use.
• LDCRS are very durable. Their primary component (i.e., the
drainage media) are comprised of relatively inert materials,
such as clean silica sands or high density polyethylene (HOPE)
synthetic drainage materials.
• LDCRS are highly reliable, low maintenance systems that can be
expected to function well over time.
• LDCRS provide 100% leak detection coverage under all portions
of the top liner potentially exposed to waste. The drainage
system of the LDCRS is capable of detecting leakage in all
areas of the bottom and sidewalls of the unit.
2.3-23
-------
LDCRS combine two important functions, teak detection and
leakage j\lection ana removal, s :e drainage layer technology
is currently the basis for existinj LCRS.
LDCRS satisfy the requirement of a leachate collection and
removal system between the liners as described in the Proposed
Double Liner Rule (51 FR 1070' 12, March 28, 1986).
The use of a leachate collection and removal system between the
top and bottom liners is state of practice in currently
constructed waste management units. An owner or operator can
continue to utilize his current design approach to meet the
requirements of an approved leak detection system rather than
developing new and potentially incompatible design concept for
the various components. Using leachate collection and removal
systems between liners to detect leaks also minimizes
additional operational and cost requirements associated with
leak detection.
2.3.2.2 Innovative Technologies
Under Ideal conditions, each of the Innovative technologies
evaluated is capable of performing better than LDCRS with regard to
leak location and detection time. However, none of these technologies
directly measures leak volume nor enables leakage collection for
chemical analysis. In addition, all of these technologies are still
in the field testing stage and none is routinely used in permanent
applications at waste management units. The long-term reliability and
durability of all of the innovative systems must be further
established before they cat be considered for use in permanent
applications.
Eiectrical resistivity (ER) is a geophysical technique whereby an
electrical current is Introduced into the ground by a pair of surface
electrodes and the resultant potential field, as measured by a second
pair of electrodes, is Interpreted to detect anomalies (leaks). ER is
limited in that It does not provide information on the rate of leakage
and cannot be practically used with either composite top or bottom
liners. Routine monitoring using ER techniques would be time
consuming and expensive and the reliability and durability of an ER
2.3-24
-------
system Is not well documented. ER shows promise for detecting the
leak location at surface 1'rpoundinent. known to be leaking. It also
has the potential to be used for CQA verification of certain portions
of Hie liner such as the sump.
Time domain reflectometry (TDR) measures the electrical property
variations In the material along a pair of parallel transmission line
conductors. TDR is sensitive to soil moisture content making it
attractive for leak detection. However, TDR has several drawbacks:
(1) it must be Installed in a granular drainage media with a
relatively low moisture content to ensure detection sensitivity; (2)
1t is advantageous to use a drainage layer of well-compacted medium to
fine sand because it increases horizontal dispersion of the wetted
front of a leak increasing the TDR response; however, too many fines
rapidly attenuate the TDR signal and also decreases the leachate
collection and removal efficiency of the LCRS; and (3) the durability
of TOR systems are limited due to the potential for corrosion of the
exposed transmission line conductors.
Acoustic emission monitoring (AEM) detects vibrations produced by
liquids leaking from a containment site through the use of
transducers. The technology has not been proven at a full scale site
and has several drawbacks: (1) the potential exists for sensors and
wires to corrode during the active life and post-closure care period;
(?) AEM may not detect small leaks or low velocity leaks where the
flow is not turbulent; and (3) AtM Is sensitive to background noises
(i.e., nearby equipment or machinery). Also, the technology is only
reliable 1f It Identifies the leak within minutes of its occurrence.
For this reason, the method must be used with a costly automated
continuous monitoring system. A brief malfunction of the monitoring
system would be unacceptable since any leaks starting during the
downtime may not be detected by the system.
Other technologies were found to be inappropriate as a primary
leak detection system for waste management units. These include
lysimeters, seismic measurements, electromagnetics and loisture
blocks.
2.3-25
-------
Darcy's Equation for Saturated Flow Conditions:
Q = kiA
Q = rate of flow (m'/s or gpm);
k = in-plane hydraulic conductivity (m/s or cm/s);
i = Ah/1 = hydraulic gradient (dimensionless);
A = TC = cross-sectional area in the direction of flow (mj or ft1)
Darcy's Equation may be rewritten as:
Q = 8iB
8 = kT = hydraulic transmissivity (mz/s or ft'/s);
i = hydraulic gradient;
B = width of the cross-sectional area in the direction of flow (m or
ft).
Figure 2.3-1. Definition of terms for flow in a LDCRS drainage layer.
2.3-26
-------
POLt -{ill-OLE
C - CURRENT ELECTRODE
P - POTENTIAL ELECTRODE
Figure 2.3-2. Electrode configuration used to detect leaks from
waste management units lined with FMLs.
2.3-27
-------
nEMOIECUMMEUr
ELECinOUE
cunntfu SOUMCE
ELECIIIOOE
NOf TO SCALE
Figure 2.3-3. Conceptual electrical resistivity testing technique
applied to detect and locate leaks through a FML liner.
[Schultz et al., 1984]
2.3-28
-------
cunitfiir Pttumi
ntcinooifu AWAY
•™;r^ ,>4:^n>V:
~*^
NOT TO SCALE
Figure 2.3-4. Conceptual drawing of the test impoundment and operation
of the test equipment used to evaluate electrical
resistivity leak detection system. [Schultz et al.,
1984]
2.3-29
-------
Figura 2.3-5. Equlpotentlal contour plots obtained from electrical
resistivity survey. [Schultz et al., 1984]
2.3-30
-------
MAGNETIC FIELD LINES
El ECHUC FIELD LINES
\
OA-H^^ J
©^-|^M3
\
V
^—I-
N
Figure 2.3-6. Parallel transmission line conductors used for Time
Domain Reflectometry (TOR).
2.3-31
-------
HAZARDOUS WASTC LANDFILL
LEVEI or :'••': ; .•:.'.''•. •' •• ;'.;;: -; •'•/
V
LIQUID IMPOUNDMENT
SiNMIETIC
I INCH
10R TRANSMISSION
l.lllt COI.'DUC K>l( PAIflS
e
SAHO DLAHKET
Figure 2.3-7 Conceptlonal TDR Installation at a hazardous waste
management unit.
2.3-32
-------
-------
r
ACM jC"~'Cn i
tAO 10 SunracE
LI'tcB BEOOIHG
V
sunrACE
IEAK COUECIIOH CIPE
AtM 5CNSOR-
MOT TO SCALE
Figure 2.3-8 AEH sensor Installed below the top liner at a new
surface impoundment.
2.3-33
-------
-S '-I flUI* *ITM I OCHAlti r CAP
-«" fo t" »o"i note
LTSIMClift IOOV-
ronoul cur *H
b - Typical Installation
a - Typical Pressure-Vacuum
Ivsimeter
HOT TO SCALE
Figure 1.3-9. Schematic of a pressure-vacuum lysimeter and typical
Installation. [GCA, 1984b]
2.3-34
-------
—-- oo
PLAN VIEW
ACCtSS UAMIIOlt
SAWI'llHG MAHHOLE ,
IIA.'AriDOUS WAS 1C LANDflLL
•/ LIQUID
r^r^^-
GfiANUlAn Ttt.L /
rcnrOHAUO COLLCCTiOU ^ \UOM- I'FnronATtO
SECTION
NOT TO SCALE
Figure 2.3-10. Layout of a collection i>J;,;!Oter. [GCA,
2.3-35
-------
2.4 LEAK DETECTION SYSTEMS BETWEEN LINERS
2.4.1 Functions of a Leak Detection System
The functions of a leak detection system between the top and
bottom liner are to detect leakage through the top liner at the
"earliest practicable time" and to provide information about the rate
of leakage. This information will enable the owner or operator and
EPA to determine what actions (such as accelerated pumping of the
LDCRS sump, etc.) are required to ensure that the hydraulic head on
the bottom liner is minimized (and thereby ensure that leakage into
and through the bottom liner is minimized).
For a given rate of top liner leakage, the hydraulic head on the
bottom liner can be minimized by removing the leachate in the LDCRS as
fast as possible. Thus, the greater the leachate collection and
removal capability of the LDCRS, the smaller the possibility of
generating significant hydraulic heads on the bottoro liner.
The key parameters to be considered in evaluating liquid
collection and removal capabilities include the hydraulic conductivity
and hydraulic transmissivity of the leak detection system drainage
media, the hydraulic gradient (which is a function of the slope of the
drainage layer), and the hydraulic conductivity of the bottom liner.
As was shown in EPA's recent Notice of Availability of Data on "Bottom
uiner Performance in Double-Lined Landfills and Surface Impoundments"
[USEPA, 1987], a compacted soil bottom liner allows significantly
greater leakage Into and through the bottom liner than does a
composite. On this basis, a composite bottom liner is preferable to a
compacted soil bottom liner. In the remainder of this document, a
composite bottom liner is assumed.
The hydraulic head above the bottom liner should be kept at a
minimum at all times. The hydraulic head affects the rate of leakage
through the bottom liner. The larger the head, the larger the leakage
rate. In the case of free surface flow, and neglecting dynamic
effects, the head of liquids on the underlying bottom liner is [Giroud
and Bonaparte, 1984]:
h = (Q/B) / (kd tan B) (Equation 2.4-1)
2.4-1
-"-^
-------
where: Q/B =• rate of flow to be transmitted by the drainage layer per
unit width; k^ « In-plane hydraulic conductivity of the drainage
material; and 0 » slope of the drainage layer. Recommended SI units
are: Q/8 (m2/s) and k,j (m/s). Equation 2.4-1 indicates that for a
given flow rate and drainage layer slope, the head decreases with
increasing drainage material hydraulic conductivity. In the case of
leak detection systems, the rate of flow to be transmitted by the
drainage layer is the rate of leakage through the top liner assuming
there is no other source of liquids entering the leak detection system
(e.g. ground water, construction water, etc.). In-plane hydraulic
conductivity 1s a property of the drainage media and the liquid and Is
typically evaluated under saturated flow conditions. In-plane
hydraulic conductivity of various drainage materials is discussed in
Section 2.4.3.2.
Leak detection time is the time Between when leakage enters the
LDCRS until the time It is collected in the LDCRS sump and available
for removal. The time between the occurrence of a leak and its
detection should be as small as possible. Detection time is
essentially a function of travel distance and velocity. Travel
distance Is established by the geometrical characteristics of the land
disposal unit. For a given unit, detection time decreases with
increasing velocity. Key parameters affecting velocity include in-
plane hydraulic conductivity of the drainage material and hydraulic
gradient. Velocities are higher 1n materials which are highly
permeable in their plane.
A leak detection system is an integral part of EPA's "liquids
management strategy" and "systems approach" to waste con' .inment.
EPA's liquids management strategy has two main objectives: (1)
minimize leachate generation in the waste management unit; and (ii)
maximize leachate removal from the waste management unit at the
earliest practicable time. It is through these two operational
objectives that EPA will achieve the Congressional goal of preventing
migration out of the unit,
The "systems approach" to waste containment applies specifically
to the second part of the "liquids management strategy", namely,
maximizing leachate removal from the land disposal unit. The double
2.4-2
-------
liner system is the mechanism by which leachate collection and removal
can be maximized. The top and bottom liner together with the LCRS
above the top liner (in the case of landfills) and the LDCRS between
the liners function in an integrated, interdependent manner to prevent
leachate migration out of the unit by maximizing its collection and
removal. Each of the system elements reinforces and supports the
other: the liners serve as a barrier to leachate migration and
facilitate Us collection and removal; the leachate collection and
removal system (l.CRS) above the top liner in landfills enables
collection and removal of leachate and minimizes the buildup of the
liquid pressure on the top liner; the leachate collection and removal
system between the liners serves to minimize the buildup of head on
the bottom liner; and the leak detection system provides the owner or
operator and EPA with notification of leakage through the top liner,
which enables the review of existing conditions and may leaa to the
taking of certain re^onse activities.
Clearly, the LDCRS, which combines leak detection and leachate
collection and removal, provides an important element to EPA's liquids
management strategy.
2.4.2 Materials
2.4.2.1 Introduction
Bonaparte et al. [1985] have identified a discussed two
categories of leachate collection and removal systems: (1) those
involving the use of coarse-grained granular materials exclusively or
a combination of granular and synthetic materials (conventional
leachate collection and removal systems); and (2) those involving the
use of synthetic materials exclusively (herein called "synthetic"
leachate collection and removal systems). Their discussion could be
adapted to leak detection systems. Conventional leak detection
systems typically consist of network of collection pipes, drainage
layers made of granular materials, and collection/monitoring
structures. Filters, when needed, are placed adjacent to granular
drainage media 1n order to prevent or reduce clogging of the granular
material due to intrusion of small particles from adjacent soil.
Filters are especially needed when the soil adjacent to the granular
drainage media consists of fine-grained soil such as the compacted
soil component of a composite top liner, or a bottom liner made of
2.4-3
-------
compacted soil alone. These filters typically consist of fine-grained
granular soils or geotextiles.
Synthetic leak detection systems differ from conventional systems
in that they typically consist of a synthetic drainage layer, in lieu
of granular materials, to transmit flow to a collection/monitoring
structure. In addition, almost all synthetic systems must use
geotextiles as filters, where necessary, to prevent adjacent soil from
clogging the synthetic drainage layer.
Figure 2.4-1 Illustrates the use of synthetic drainage materials
to create leak detection systems.
The following sections discuss materials for the components of
leak detection systems. Components include drainage layers,
collection pipes, collection/monitoring structures, filter layers
(where necessary) and cushion layers (where neces ary) to protect
FMLs.
2.4.2.2 Dra]nage_Layers
2.4.2.2.1 Granular Drainage Materials
A wide range of sands and gravels can be uscJ in 'eak detection
systems which may -ry from medium to coarse in size and well-graded
to uniform in gradation. Selection of the drainage materials depends
on the following considerations:
• The drainage layer should be able Lo collect and remove rapidly
liquids entering the system as a result of leakage through the
top liner, as discussed in Section 2.4.1. The drainage layer
should have adequate hydraulic conductivity and hydraulic
transmi ssivi ty.
• The drainage material should not darmqe FMLs when the FMLs are
in direct contact with the drainage material. This scenario
may occur when sharp gravel is placed adjacent to a top liner
consisting of a FML alone and/or a bottom liner consisting of a
>\ ).,.- • composite liner. Protection of the FML against damage may be
accomplished by placing cushion layers between drainage
material and FML. Cushion layers may consist of either sand or
- '. V.
•'-V "_
N
V
2.4-4
-------
geotextlle and arc further described in Section 2.4.2.4.
Criteria for the FML protection have been provided in U5EPA
[1905].
• The drainage layer should be physically compatible with filters
or cushions made of fine-grained granular soils to minimize the
migration of the filter or cushion materials into the drainage
layer.
2.4.2.2.2 Synthetic Drainage Materials
Synthetic drainage layers are made of planar materials thick
enough to convey fluids in their plane. Synthetic drainage materials
are usually made from polymers. Typical polymers include
polypropylene, polyester, and polyethylene. Formulations of these
polymers can be manufactured to be highly inert to biological and
chemical degradation.
Four types of available synthetic drainage materials have been
identified by Glroud and Bonaparte [1984]. They are:
• Nets, which are the most widely used synthetic drainage
materials in leak detection systems, with thicknesses ranging
from approximately 4 mm to 7 mm (160 to 280 mils).
• Needlepunched nonwoven geotextiles, thickness 2 to 5 mm (80 to
200 mils).
• Mats, thickness 10 to 20 nrn (400 to 800 mils).
• Corrugated, waffled or alveolate plates, thickness 10 to 20 mm
(400 to 800 mils).
NeJ.j> (Figure 2.4-2) consist of two sets of parallel extruded
polymer strands Intersecting at a constant angle (gene 'ly between
60° and 90°). Strands of one set lie on top of strands ' the other
set, and the two sets are melt-bonded at the intersection. The two
sets of strands -reate two sets of channels which can convey liquids.
This ability to transmit flow depends on the net geometry and
properties as discussed in Section 2.4.3.3.
2.4-5
-------
A large variety of nets are available. They differ by: size and
shape of cross section of strands; depth of channels; opening size;
and nature of the polymer. Nets used as drainage materials in leak
detection systems are usually made of medium density polyethylene
(HOPE) or high density polyethylene (HOPE) and their geometry makes
them suitable for drainage. Their strands are typically 1 to 3mm
(1/24 to 1/8 In.) in height and width, and their overall thickness is
almost twice the strand height (channel depth 1s approximately equal
to strand height). Opening size is typically from 5 to 10 mm (1/4 to
1/2 in.).
Not all nets are suitable for drainage. Nets with both sets of
strands in the same plane do not have channels deep encugh to allow
significant flow of liquid. Nets with large openings (typically
larger than 10 mm (0.5 in.)) and deep channels could be used as a
drainage medium between two rigid materials, such as concrete; but
flexible materials, such as geotextlles and FMLs, can penetrate the
channels when subjected to soil or liquid pressure and decrease the
rate of liquid flow 1n the net.
Except in a few locations where they are in contact with coarse
granular materials such as gravel, nets used for drainage should never
be put in direct contact with soil. Fine soil particles would rapidly
penetrate net openings and hamper flow of liquids or gases. In most
cases, nets used for drainage are in contact with c^otextiles or FMLs.
Nets, geotextlles and FMLs can be supplied independently and Installed
successively. Composites comprising nets, geotextiles and/or FMLs are
also available or can be custom fabricated.
Needlepunched nonwoven geotextiles (Figure 2.4-2) are formed from
staple fibers or filaments extruded after melting a polymer. These
fibers or filaments are arranged into a planar structure in an
oriented or random pattern and are mechanically bonded by
needlepunching. Needlepunching is accomplished by thousands of small
barbed needles set into a board, which punch through the loose fiber
web and withdraw, leaving fibers entangled.
Mats (Figure 2.4-2) are open structure., which are made of coarse
and rather rigid filaments with a tortuous shape, bonded at their
intersections.
2.4-6
-------
-asas
Corrugated, waffled or alveolate p_la_tes (Figure 2.4-2) are made by
forming a plastic sheet Into the desired profile. Forming techniques
include extrusion, molding, pressing, et~.
2.4.2.3 F1Uer Layers
Two types of filters are typically used in engineering practice.
These are granular filters and geotextile filters.
Granular filters were first introduced in the 1920's [Terzaghi and
Peck, 1967] and have since been extensively used in hydraulic
structures su-ch as dams and reservoirs. They consist of a granular
layer or combination of granular layers having a coarser gradation in
the direction of seepage than the soil to be protected. Granular
filters typically utilize materials which are more permeable than fine
sands.
Geotextiles have been traditionally manufactured in two varieties,
i.e. wovens and nonwovens. However, a variety of other related
products (geotextile-related products) has been introduced recently on
the market. A detailed discussion of available geotextiles and
geotextile-related products is beyond *he scope of this section and
can be found elsewhere [Giroud, 1984a; Koerner, 1986].
The varieties of geotextiles which are most commonly used as
filters are wovens and nonwovens. The following definitions are
reproduced from [Giroud and Carroll, 1983].
Woven geotextiles are composed of two sets of parallel yarns
systematically interlaced to form a planar structure. The manner in
which the two sets of yarns are interlaced determines the weeve
pattern. The two sets of yarns are generally perpendicular, but some
wovens weave at a skew angle.
Nonwoven geotextiles are formed from staple (short) fibers or
filaments that are arranged, using oriented or random patterns, into a
planar structure. The fibers or filaments are bonded together by one
of the following process:
• Chemical bonding: a cementing medium such as glue, rubber,
latex, cellulose derivative, or, more frequently, synthetic
2.4-7
-------
resin, is added to fix the fibers or filaments together. One
thus obtains chemically bonded non-woven geotextiles.
• Theimal bonding: heat causes partial melting of the fibers or
filaments and makes them adhere together at their cross-over
points. One thus obtains heatbonded nonwoven geotextiles that
are relatively thin, typically 0.5 to 1 mm (20 to 40 mils).
• Mechanical bonding by needlepunching: thousands of small
barbed needles, set into a board, punch through the loose fiber
web and withdraw, leaving fibers or filaments entangled. One
thus obtains needlepunched nonwoven geotextiles that are
relatively thick, typically 1 to 5 mm (40 to 200 mils) or more.
2.4.2.4 CusMon Layers
Cushion layers are used in leachate collection and removal systems
to protect FMLs from potential damage caused by drainage materials.
Potential for damage to the FML depends on a number of factors
including nature of drainage material (e.g. particle size and shape
for granular materials, flexibility or tr ttleness for synthetic
materials) and compressive stresses.
Cushion layers typically consist of a few inch thick sand layer or
geotextiles placed between the drainage layer and the "ML. Sand
cushions are exclusively used with granular drainage materials and are
never placed in direct contact with some types of synthetic drainage
material as this could clog the synthetic drainage layer. The
particle size distribution of the cushion sand should be physically
compatible with the granular drainage material to prevent migration of
the sand Into the drainage layer. If not, sand migration may occur
under high seepage forces. Sand migration may result in either one or
both of the following:
• decrease of the effective thickness of the cushion layer, thus
Increasing the potential for puncture of the FML;
• clogging of the leachate collection and removal system.
Geotextiles do not exhibit the drawbacks described above. Typical
geotextile cushions consist of needlepunched nonwovens with a minimum
2.4-8
-------
mass per unit area of 400 g/m* (12 oz/yd2). Meedlepunched nonwoven
geotextlles have been briefly described In Sections 2.4.2.2.2 and
2.4.2.3.
2.4.2.5 Pipes
The primary use of pipes in leak detection systems is to collect
the leachate 1n the drainage layer and convey 1t to
collection/monitoring points. Pipes are ?lso used In the construction
of monitoring ports and system cleanouts.
Theimoplastlc pipes (PVC, HOPE) are widely used In leak detection
systems due to their wide range of chemical resistance and their lower
potential for damaging FMLs. However, if the expected leachate
contains chemicals which may be harmful to these materials, other pipe
materials should be considered.
Pipe structural properties range from flexible to rigid. Flexible
and semi-flexible pipes rely on bedding materials for much of their
structural support. Bedding requirements are usually less crucial for
rigid pipe.
Pipes may or may not be available factory-perforated. For
perforated pipes, leakage enters through slots or circular openings.
Slots or openings should be sufficiently large and spatially arranged
to allow free flow of liquid, but not result 1n significant reduction
1n pipe strength. Alternatively, porous wall concrete pipe may be
applicable for system use.
Pipes are available in various sizes. The diameter of the pipes
is normally calculated based on the design flow quantity. Pipe
diameters as small as 5 and 10 cm (2 and 4 in.) have been utilized in
leak detection systems. However, 0.15-m (5-in.) diameter pipes have
also been used and these pipes may simplify the system maintenance.
EPA has recommended 0.15-m (6-in.) minimum diameter pipes for use in
leachate collection and removal systems at land disposal units [USEPA,
19351.
2.4-9
-------
2.4.2.6 Structures
Structures In leachate collection and removal systems can function
as collection -lonitoring ports, cleaning access points, or
collection/removal structures. Selection of structure type will
depend on system layout and site specific conditions. A description
of various types of structures is presented below.
2.4.2.6.1 Manholes
Manholes act as the ultimate collection/monitoring point where
leachate will flow by gravity. They also serve as a cleaning access
for system Inspection and maintenance.
Manholes used in leachate collection and removal systems are often
Identical to those used in conventional construction projects.
Manholes are normally precast concrete structures. However,
structures manufactured using other types of material, such as high
density polyethylene (HOPE), are also available. In cases where
expected leachate and concrete are incompatible, a different type of
material should be specified.
The diameter of the manhole should be large enough to permit
personnel and equipment entry. Manholes with an inner diameter of 0.6
to 1.2 m (2 to 4 ft) have been used. Manholes may or may not be
associated with sumps and they may be located inside or outside of the
waste management unit boundaries.
2.4.2.6.2 Sumps
Sumps are typically located at the lowest points of a land
disposal unit. They essentially perform the same system functions as
manholes. The benefit of sumps over manholes in leachate collection
and removal systems is that they can be designed to operate between
liners. Penetration of the top liner would therefore be unnecessary.
Sumps are usually made of concrete. Similar to manholes, chemical
compatibility of the sump material with the expected leachate must be
evaluated. An altern-Mve approach consists of lining the sump with
the same type of FML jpecified for the top and/or bottom liners. This
alternative approach might preclude the need for connections between
the FML and sump which is a potential source of leakage.
2.4-10
-------
2.4.2.6.3 Auxiliary Cleanouts
Auxiliary cleanouts permit further denning access to collection
laterals. They also may be designed to extend between linars to the
surface. Auxiliary cleanouts can be made of materials used to ake
pipes and their diameter Is usually equivalent to the diameter of the
lateral col lection pipe.
2.4.3 Properties of Materials
2.4.3.1 Introduction
The LDCRS 1s required to perform a number of functions, as
identified in Section 2.4.1. These functions are to allow leak
detection rapidly (at the "earliest practicable time"), to maximize
collection and removal of liquids present in the leak detection system
and to minimize the hydraulic head on the bottom liner. Two basic
properties of the LDCRS drainage medium have been identified which
govern the drainage performance of leak detection systems under
steady-state, saturated conditions. These properties are hydraulic
transmisslvity and hydraulic conductivity. Additionally, each
component of the LDCRS must be durable to perform its respective
function during the waste management unit's active life and post-
closure care period. Further, the materials used in the LDCRS should
not damage the FML liners.
This section discusses properties required of LDCRS components,
I.e. hydraulic transmissivity and hydraulic conductivity for the
drainage layer, filter characteristics for granular or geotextile
filters, durability, and mechanical effects on FMLs.
2.4.3.2 Hydraulic Conductivity
Hydraulic conductivity describes the velocity of liquid flow
through the drainage layer under a hydraulic gradient equal to one.
Since the velocity of liquid flow is directly proportional to
hydraulic conductivity, hydraulic conductivity Is the single most
Important variable controlling leak detection time. The larger the
hydraulic conductivity of the drainage layer, the shorter the leak
detection time.
2.4-11
-------
Hydraulic conductivity of a drainage layer is the rate of flow per
unit cross-sectional area of drainage layer per unit hydraulic
gradient. The hydraulic conductivity of a drainage layer, kj (m/s),
may be calculated as follows using Darcy's equation:
kd - v/1 » (Q/A)/t (Equation 2.4-2)
where: v - apparent fluid velocity (m/s); Q/A 1s the rate of flow per
unit area of drainage layer In a cross-section normal to the direction
of flow (m/s); and 1 5 the hydraulic gradient (dlmensionless).
Hydraulic conductivity 1s almost always calculated under saturated
flow conditions. When the leak detection system material is not
saturated, the phenomenon of capillarity takes place and the flow of
liquid 1s no longer governed by the above equation. The development
of capillarity 1s linked to soil particle size distribution, which 1s
linked to hydraulic conductivity. This Is discussed In Section
2.6.2.4.2.
The two sections which follow compare granular drainage materials
and synthetic drainage materials in the context of hydraulic
conductivity. Typical hydraullr conductivity values are also
provided.
2.4.3.2.1 Granular Drainage Materials
Granular drainage materials Include sands and gravels, as
discussed 1n Section 2.4.2.2.1. Typical ranges of hydraulic
conductivity of granular media are:
• k,j - 10"* to 10"1 m/s (10~J to 1 cm/s) for fine, medium, coarse
sands; and
• kd - 10~a to 1 m/s (1 to 100 cm/s) for gravels.
Hydraulic conductivity of a granular media depends on a variety of
physical factors. For unconsolidated materials, hydraulic
conductivity varies directly with particle size [Todd, 1980]; the
larger the particle size, the higher the hydraulic conductivity.
Accordingly, as shown by the above values, gravels exhibit the highest
2.4-12
-------
hydraulic conductivity value. Hydraulic cjnductlvlty values for
gravels are variable. Uniform size gravels generally exhibit higher
hydraulic conductivity values than well-graded gravels for given
appioxiniate sizes. Well-graded gravels exhibit a broad range of
conductivity values due to variations in particle size; for example
well-graded, alluvial gravel often has a large content of fines which
blocks flow pathways. Hydraulic conductivity values for fine to
coarse sand vary from higher values for coarse sands to lower values
for fine sands.
2.4.3.2.2 Synthetic Drainage Ma^Mals
Synthetic drainage materials include nets, needlepunched nonwoven
geotextlles, mats and corrugated, waffled or alveolate plates, as
discussed In Section 2.4.2. The hydraulic conductivity of synthetic
drainage materials Is not measured directly. It is typically
calculated by dividing hydraulic transmlssivity of a synthetic
drainage material (which is measured through testing, as discussed in
Section 2.4.3.3) by Us thickness. Both hydraulic transmisslvity and
thickness are affected by overburden compressive stress and boundary
conditions, as discussed subsequently. As a result, hydraulic
conductivity 1s also affected by compressive stress and boundary
condi tions.
- Nominal Hydraulic Conductivity
A nominal hydraulic conductivity can be defined for the flow of
water at 20°C (68°F) a hydraulic gradient of one, and the smallest
compressive stress necessary to keep the sample flat (practically, 10
kPa, i.e., 200 psf). Nominal values are determined with the synthetic
drainage layer placed between two smooth steel plates.
Under these conditions, the following values have been obtained
from a review of the literature and from manufacturers' data:
• nets: kj = 0.1 to 0.5 m/s (10 to 50 cm/s)
• thick needlepunched nonwcven geotextiles: k,j = 10~4 to 10"1 m/s
(0.01 to 0.1 cm/s)
2.4-13
-------
• mats: kd - 0.1 m/s (10 cm/s)
• waffles: kd = 0.1 to 1 m/s (10 to 103 cm/s)
- Influence of Compressive Stress and Boundary Conditions
As previously noted, the hydraulic conductivity of synthetic
drainage materials 1s not determined directly but 1s calculated from
the results of hydraulic transmissivity tests. The hydraulic
transmissivlty of a synthetic drainage layer is strongly affected by
the applied compressive stress (since the synthetic drainage layer is
compressible) and by the material In contact with the drainage layer
(since this material tends to penetrate the openings of the synthetic
drainage layer). Since the hydraulic transmissivity Is strongly
affected, so Is the hydraulic conductivity. The effect of compressive
stress and non-rigid boundary conditions 1s to reduce the hydraulic
conductivity of the synthetic drainage layer below Us nominal value.
The reduced conductivity can be calculated from the reduced hydraulic
transmissivities measured in tests with non-rigid boundary materials
(such as geotextlles and FMLs) and at elevated compressive stresses
(simulating waste overburden stresses). Hydraulic transmissivities
under these conditions are presented in Section 2.4.3.3.
2.4.3.3 Hydraulic Transmissivi'•y
Hydraul': transmissivity of a drainage layer \z the rate of flow
per unit width of drainage layer, under a unit hydraulic gradient.
The hydraulic transmissivity of a drainage l-.yer related to a given
liquid may be calculated as follows using Darcy's equation [Williams
et al,, 1984];
(Equation 2.4-3)
where: Q/B • rate of flow per unit width of drainage layer (m2/s);
and 1 = hydraulic gradient (dlmenslonless). Recotrmended SI units are:
v (m/s) and Q/B (m1).
The hydraulic transmissivity and in-plane hydraulic conductivity
of a drainage layer are related as follows:
8d - kdT
(Equation 2.4-4)
2.4-14
-------
4?'.i'>''" - '•-. • 1>1-" * * •*<-*" * - ' ;• J2-~
where: kd - In-plane hydraulic conductivity of the drainage layer
(m/s); and T = thickness (").
The hydraulic transnii ss i v i ty of a drainage layer depends on
several parameters. Not all the parameters affect the transmissivity
of all types of drainage materials, e.g. parameters which affect
synthetic materials may ..ave a negligible influence on granular
materials. The discussions which follow have been grouped In two
sections. Both the section on granular drainage materials and the
synthetic drainage materials discuss:
• how hydraulic transmissivity is obt-;ned; and
'• the 'nfluence of various parameters such as corrpressive stress
an.. )undary conditions on h_,-,raulic transmissivity.
In addition, both sections provide typical hydraulic
transmissivity values.
2.4.3.J.I Granular Drainage Materials
Granular drainage materials Include sands and gravels, as
discussed in Section Z.4.2.2.1. The hydraulic transmissivit., of
granular drainage materials is typically calculated by multiplying the
hydraulic conductivity of the drainage material by the thickness of
the drainage layer, although transmissivity can be directly measured
through testing. Typical ranges of hydraulic transmissivity for a 0.3
m (1 ft) thick layer of granular materials are:
• 8d = 10"' to 10"' m2/s (0.005 to 5 gpm/ft) for sands;
• Od = 10"' to 10"' m'/s (5 to 500 gpn/ft) for gravels.
- Influence of Compressive Stress
Compresslve stresses generally have little to no effect on the
hydraulic conductivity of granular drainage materials and the
thickness of granular drainage layers. Therefore compressive stresses
generally have little to no effect on the hydraulic transmissivity of
drainage layers made of granular drainage materials.
2.4-15
-------
- Inf'-jence of Doundary Conditions
The hydraulic transmissivity of drainage layers made of granular
materials is not significantly affected by boundary conditions
because:
• granular dralncge layers are relatively thick (compared to
synthetic drainage layers); and
• the boundary materials cannot significantly penetrate the
granular drainage layer (except In case of clogging).
2.4.3.3.2 Synthetic Drainage Materials
Synthetic drainage materials include nets, needlepunched nonwoven
geotextlles, mats and corrugated, waffled or alveolate plates, as
discussed In Section k.4.2. The hydraulic transmisslvity of synthetic
drainage materials 1s typically measured in the laboratory. Hydraulic
transmissivity testing is described in detail in Williams et al.,
[1984], The hydraulic transmissivity test device is shown in Figure
2.4-3. Transmissivlty tests should simulate as accurately as possible
the conditions that will exist in the land disposal unit by accounting
for such variables as overburden con^ressive stress and boundary
conoitions. The Influence of each of these parameters is discussed
below.
- Nominal Hydraulic Transmissivity
A nominal hydraulic transmissivity can be defined for the
following standard conditions: (i) the fluid is water; (ii) the
temperature is 20°C (68°F); (111) the hydraulic gradient is one; and
(iv) the compressive stress is the smallest necessary to keep the
specimen flat (as a practical matter, 10 kPa, i.e., 200 psf). Nominal
values are dete.,nined with the synthetic drainage layer placed between
two smooth steel plates.
Under these conditions, the following values have been obtained
from a review of the literature and from manufacturers' data:
2.4-16
-------
• nets: Od - 10~4 to 10"' m'/s (0.5 to 5 gpm/ft)
• thick needlepunchod nonwcven geotextiles: Gd = 10"' to 10~4
ci'/s (0.005 to 0.5 gpm/ft)
• mats: 6d = 10~' to 10"' m'/s (0.5 to 5 gpm/ft)
• waffles: 6d » 10"' to 10~2 m!/s (5 to 50 gpm/ft)
- Influence of Compressive Stress
Compresslve stresses are caused 1n the field by the weight of the
material overlying the synthetic drainage layer: weight of the solid
waste or earth located above the considered drainage layer in the case
of a landfill or waste pile unit, and pressure of the impounded liquid
exerted on a liner overlying a drainage layer 1n the case of a
surface Impoundment unit.
All synthetic drainage layers are compressible and their thickness
and hydraulic conductivity decrease with increasing compressive
stress. As a result their transmissivity decreases. The Influence of
compressive stress on the transmissivity and thickness of nets,
need!epunched nonwoven geotextiles, mats and waffles is shown 1n
Figure 2.4-4.
Some nets should be used with caution at compressive stresses
larger than 300 kPa (6,000 psf) because their hydraulic transmissivity
decreases rapidly beyond this value of the compressive stress.
Williams et al. [1984] reported that the hydraulic transmissivity of
two corrmercial nets could be 10 to 100 times greater at 10 kPa (200
psf) than at 500 kPa (10,000 psf). The hydraulic transmissivity of
other nets decreases more progressively with Increasing compressive
stresses. These other nets can be used for compressive stresses as
high as 1000 kPa (20,000 psf), as shown by test results presented by
Bonaparte et al. [1985]. It can be concluded that the performance
characteristics of nets (and other synthetic drainage layers) must be
determined on a product specific basis.
Waffles exhibit a relatively low compressibility due to their
stiffness and their transmissivity varies only slightly up to a
2.4-17
-------
certain critical value of compresslve stress. Above this critical
value, the waffle collapses. The critical value of compressive stress
is typically 200 to 300 kFa (4,000 to 6,000 psf) which corresponds to
a depth of earth or solid waste of appr~ • imately 10 to 20 m (30 to 60
ft).
Mats do not collapse, but they are very compressible and lose much
of their hydraulic transmissivHy at a compressive stress of 50 to 150
kPa (1,000 to 3,000 psf) and the manufacturer of the mats recommends
rightfully that their use be limited to depths smaller than 5 to 10 m
(15 to 30 ft).
Need!"Bunched nonwoven geotextiles do not collapse, but they are
very compressible and lose much of their hydraulic transmlssivity at a
compressive stress of the order of 100 kPa (2,000 psf).
- Influence of the Combination of Boundary Conditions and Compressive
Stress
The transmlssivity of a synthetic drainage layer depends on the
type of geotextile or FML adjacent to the drainage layer. As shown by
Williams et at., [1984], the effect of gootextile boundaries
(compared to steel plates) on the hydraulic transrnissivity of a net is
particularly significant. Thick, compressible geotextiles such as
typical needlepunched nonwovens penetrate into net channels, which
decreases the hydraulic transmissivity of the net:
• A thick compressible nonwoven geotextile with a mass per unit
area of 400 g/m2 (12 oz/y'2) on both sides of a net decreases
the hydraulic transmi ss i, i ty by a factor of 100 to 1000
compared to the nominal value.
• A thick compressible nonwoven geotextile with a ass per unit
area of 400 g/m2 (12 oz/yd2) on one side of a net and a rigid
boundary on the other side decreases the transmissivity by a
factor of 5 to 10 (and not only by a factor of two because, at
each rib, the liquids goes from one side of the net to the
other side where geotextile fibers slow down the flow).
2.4-18
-------
Williams et al. [1981] observed that the effect of a FML adjacent
to a not was less significant than the effect of an adjacent
gectoxtlie:
• A thin unreinforced FML (such as a 0.5 mm (20 mil) thick PVC)
on both sides of a net decreases the transmissivity by a factor
of 2.
• A thicker reinforced FML (such as a 0.9 mm (36 mil) thick
Hypalon) or a thick, stiff FML (surh as a 1 mm (40 mil) thick
HOPE) on one side of a net has a negligible effect on the net
transmlsslvity.
The above observation illustrate the influence boundary materials
such as geotextlles or FMLs have on the transmissivi ty of a net.
Figure 2.4-5 Illustrates the effect of boundary conditions and
compressive stress on the hydraulic transmissivity of a net by
plotting the hydraulic transmissivity values of the following systems:
• 1.5 mm (60 mil) thick HOPE FML/ 5 .:;m (200 mil) thick
polyethylene net / 2 mm (80 mil) thick HOPE FML; and
• 25 mm (1 in.) thick clay layer compacted to 907. standard
proctor dry density / special polyester needlepunched nonwoven
geotextile with a very high needling density, mass per unit
area 250 g/m2 (7.4 oz/yd2) / 5 mm (200 mil) thick polyethylene
net / aluminum plate simulating a thick, stiff FML.
The hydraulic transmissivity values of a thick polyester
needlepunched nonwoven geotextile made with continuous filaments, mass
per unit area of 600 g/m2 (18 oz/yd2) under conipressive stress are
also shown for comparative purposes. Figure 2.4-5 shows that:
• the hydraulic transmissivity of a FML/nat/FML system is
typically higher than the hydraulic transmissivity of the
clay/geotextile/net/FML system by approximately one order of
magnitude; and
2.4-19
-------
EZ"
• the hydraulic transmissivlty of the clay/geotextile/net/FML
system 1s typically higher than the hydraulic transmisslvi ty of
the needlepunched nonwoven geotextile by approximately two
orders of magnltude.
2.4.3.4 Filter Characteristics
A filter is required to prevent clogging of a drainage layer when
particles of the soil adjacent to the drainage layer are capable of
migrating into the layer. In the case of leak detection systems, a
filter will be needed when the top liner is a compo: te • the bottom
liner consists of compacted soil alone. This filter ITU. %e either a
single granular layer or a combination of granular layers having a
coarser gradation In the direction of seepage than the soil to be
protected, or a geotextile, as discussed in Section 2.4.2.5.
2.4.3.4.1 Mechanisms of Filtration
A filter must prevent particles of the soil adjacent to the
drainage layer from migrating into trie drainage layer. At the same
time, the filter must allow free flow of liquid from the adjacent soil
into the drainage layer. Soil particles which are smaller than the
pore spaces between the granular filter particles or the geotextile
openings may pass through the granular or geotextile filter when they
are subjected to shear stresses or a high seepage gradient. The
ability of a filter to provide soil particle retention therefore
depends on the size of its pore spaces or openings. The smaller the
size of the pore spaces or openings, the better the retention provided
by the filter. However, the size of the filter pore spaces or
openings should not be too small and impede free flow of liquids. The
functions of a filter may be summarized as follows: the filter must
be permeable enough to allow free f'rw into the drainage layer (which
implies that the size of pore spaces or openings must be larger than a
minimum value) and at the same time retain the soil to be protected
(which implies that the size of pore spaces or openings cannot exceed
a maximum value).
2.4.3.4.2 Granular Materials
Various design criteria are available for granular filters. The
original work was done by Terzaghi [1922] arr' was followed by
2.4-20
-------
considerable work done by many researchers and organizations. As a
result, various expressions of the criteria can be found. The
following expression of the criteria can be found in [Cedergren,
19/7J.
- Retention Criterion
The 15% size (015) of granular filter material (I.e., the particle
size larger than 15 percent :. f the soil particles on a dry weight
basis) must not be more than four or five times the 85% size (095) of
the adjac.nt soil.
D15 (filter)
- < 4 or 5 (Equation 2.4-5)
D85 (soil)
Granular materials are always composed of ranges of particle
sizes. The retention criterion assumes that if the pore spaces 1n
granular filters are small enough to retain approximately the 85% of
the protected soil, then the majority of the finer soil particles will
also be retained (Figure 2. 4- J). Exceptions to this criterion are
gap-graded soils and soil-rock mixtures [Cedergren, 1977].
- Permeability Criterion
The 15% size (Djs) of a granular filter material should be at
least four or five times the 15% (0^5) of a the adjacent soil.
DIB (filter)
- > 4 or 5 (Equation 2.4-6)
DIS (soil)
The permeability criterion requires that granular filters be
sufficiently permeable to prevent the buildup of large seepage forces
and hydrostatic pressures in filters and drainage layers.
The use of granular • Iters implies that the drainage layer
consists of natural drainage materials. In order to prevent the
particles of the granular filter from migrating into the drainage
layer and to ensure that free flow will occur through both the
granular filter and drainage layer, both the retention and
permeability criteria presented above must also be met by the drainage
2.4-21
-------
material-filter system. If lhe drainage material-filter system does
not meet these criteria bec-.se the drainage material particles are
too coarse, an intermediate layer may be necessary to achieve
transition between the granular filter and the drainage layer.
2.4.3.4.3 Geotextlle Filters
Geotextile filters should meet one or two requirements depending
on the nature of the drainage materials:
• Granular Drainage Materials: the geotextlle should meet filter
criteria.
• Synthetic Drainage Materials: (1) the geotextile should meet
filter criteria; and (2) under compresslve stress, penetration
of the geotextlle into the drainage material, should be limited
as penetration could reduce both the hydraulic transmissivity
and conductivity of the synthetic drainage material, as
discussed 1n Section 2.4.3.2.2.
- Filter Criteria
A number of filter criteria nave been proposed by various authors.
Most authors agree that three criteria should be considered for
selecting a filter:
• A retention criterion to ensure that the geotextile openings
are small enough to prevent migration of soil particles.
• A permeability criterion to ensure that the geotextile 1s
pe- 3able enough to allow liquids to pass almost freely through
It (I.e., without significant buildup of liquid pressure
upstream of the filter).
• A clogging criterion to ensure that the geotextlle has a large
number of openings so that blocking of a few of them will not
significantly impair the performance of the filter.
These criteria are summarized in Table 2.4-1. Experience shows
that filter criteria are often fulfilled by the following geotextiles:
2.4-22
-------
• Nonwoven geotextiles ..".eatr rnde: or neealepunched) for fine-
grained soils (I.e., soil:- hiding more than 507. of their
particles smaller than 3.075 ~r, -,e., US sieve Mo. 200).
• Woven monofllament geotextiles ft— sands.
Mul tifllament woven geotsxtlles an slit film woven geotextiles
are often not acceptable filters, usually because they do not meet
clogging criteria.
- Penetration of Net Channels
The discussion presented in Sectiin 2.4.3.2.2 showed that tr
hydraulic transmlsslvlty of s:me synthetic drainage materials can be
significantly reduced by plac;,*g ccnpressible or flexible geotextiles
in contact with the synthetic drainc:e materials. Traditional
needlepunched geotextiles are thic«. a-r compressible; consequently
they penetrate net channels, eve"1 ur-er low compressive stress.
Unless specially required, trailti:~a1 needlepunched nonwoven
geotextiles should not be used in contact with some synthetic drainage
materials. If a needle-punched nonwr.en teotextile is necessary (for
example, to meet strict filter c"it=-ia), the designer has two
alternatives:
• Use two (or more) layers of synfstlc drainage material with
the first layer in contact with the needlepunched geotextile
being neglected in the c=sign.
• Conduct hydraulic tranj-issi • ity tests simulating the actual
site conditions (as inditated in Iictlon 2.4.3.2.2) to evaluate
the transmlssivity of the s_,nt~2t1c drainage material and
needlepunched nonwove- gertextile under the simulated
conditions with a needletanche^ gettextile in contact.
• Specify a special needl:aunc!-ed rrnwoven geotextile which has
been specially treated by e
-------
Special needlepuncfied nonwoven geotextiles do not penetrate net
channels as much as the traditional need! epunched geotextiles
discussed above. However, depending on the stiffness of the soil
located adjacent to the geotextile and on the magnitude of compresslve
stresses, they may penetrate the synthetic drainage layer enough to
significantly affect hydraulic transmissivity. Therefore, hydraulic
transmissivity tests are usually recommended, even with these special
materials.
Although heatbonded nonwoven geotextiles and monofilament woven
geotextiles do not significantly penetrate Into synthetic drainage
layers, they slightly reduce the hydraulic transmissivity of the
synthetic drainage layer as c^-npared to the hydraulic transmissivity
measured between steel plates. Therefore, hydraulic transmissivity
tests are recommended even with heatbonded nonwovens and monofilament
wovens.
2.4.3.5 Durability
The term durability Indicates resistance to progressive
deterioration. Durability of the components of LDCRS depends on the
action exerted by external forces or materials (leachate), on the raw
material, and on the physical structure of the material. The type of
action exerted by external forces can be mechanical (abrasion,
fatigue) or physico-chemical-biological (degradation).
2.4.3.".! Abrasion and Fatigue
Granular materials generally have good resistance to abrasion and
fatigue and this 1s usually not a concern in waste management
applications.
Abrasion and fatigue resistance of a synthetic drainage material
In a given application can be evaluated by tests simulating actual
environmental and loading conditions. Presently there are no widely
accepted standard procedures for such tests. Geotextiles with high
abrasion resistance are available.
2.4-24
-------
2.4.3.5.2 Physlco-Chemlcal-Blologlcal Degradation
Physlco-chemical-bio'ogical degradation of LOCKS components can be
caused by external agen.j, such as: contact with soil, atmospheric
conditions (Including exposure to the sun), and contact with chemicals
or biological agents.
- Contact With Soil
Granular materials are naturally occurring soil materials. As a
result, they are not affected by contact with soil.
ri
. Experience Indicates that geotextiles (usually made of
polypropylene or polyester) have a durability In excess of several
dozens of years when placed in most naturally occurring soil
environments. Nets, mats and waffled structures are made of
polyethylene or polypropylene. Polyethylene and polypropylene both
tend to be very stable polymers when placed in contact with naturally
occurring soils. Since, additionally, the strands of nets, filaments
of mats, and thickness of waffled structure sheets are much coarser
than the fibers or filaments of a geotextile, the durability of nets,
mats and waffled structures buried in the ground is expected to be
equal to or greater than the durability of geotextiles buried In the
ground.
- Outdoor Exposure
Granular materials do not degrade significantly with outdoor
exposure.
Mover id nonwoven geotextiles are usually very sensitive to ultra
violet (UVy light and should not be exposed to sunlight for a period
of time exceeding a few weeks to a few months depending on the type of
polymer and UV resistant additives used to manufacture the geotextile.
For longer duration of exposure the :eotextile should be covered with
a protective material (e.g., soil).
2.4-25
-------
Mats and waffled structures are usually made of polypropylene. The
filaments of mats and thickness of waffled structures are much coarser
than the fibers or filaments of a geotextiles. As a result, their UV
resistance is equal to or greater than that of geotextiles.
Nets have coarse strands and typically contain a certain amount of
carton black which has been added to the polyethylen- resin to provide
UV stabilization. As a result, they can be exposed to UV light for
extended periods of time.
- Contact with Chemicals
Each component of the LDCRS must be chemically resistant to attack
by the expected waste liquids during the active life and post-closure
care period of the unit. LDCRS components include drainage layers,
filters, cushions, and collection pipes and structures. The materials
Involved are granular drainage materials, synthetic drainage
materials, granular filters, geotextlle filters, sand cushions,
geotextile cushions, plastic pipes, structures and pipes made of other
types of material, as discussed in Section 2.4.2.
These materials may be divided in several categories:
• Granular Materials, which include granular drainage materials,
granular filters, and sand cushions.
• Synthetic drainage materials, which Include nets,
needlepunched nonwovens, mats and waffled structures.
• Geotextiles, which Include geotextile filters and geotextile
cushions.
• Plastic pipes, which include PVC, HOPE.
• Structures and pipes made of other types of materials.
This section is intended to provide the owner or operator with
information on chemical compatibility of each component of the LDCRS.
The chemical compatibility of system components is typically evaluated
based on a two-stage process procedure: (1) knowledge of the chemical
2.4-26
-------
resistance of component materials which enables a preliminary
selection of component material; and (2} "accelerated testing"
immersion of the compune.it In the expected waste leachate for
extended periods of time and testing to evaluate property changes
(this second stage enables the owner or regulatory agency to assess
the performance of the component during the active life and post-
closure care period with some degree of confidence). Each of the
following subsections provides first a brief discussion on the
chemical compatibility of the materials Identified above and second
references currently available sting procedures.
Granular Materials. Granular materials are composed of a variety of
minerals such as quartz and limestone. The chemical resistance of
granular materials depends on their mineral content and physical
structure. The chemical compatibility of granu'T materials with the
expected waste liquids has not received much attention from designers
or engineers to date, primarily based on the belief that granular
materials are natural and, as such, will last forever. This belief is
questionable even when one considers natural materials in a natural
environment. The formation of solution cavities in limestone 1s a
typical example that illustrates the attack of natural materials by
natural agents in a natural environment. The belief becomes even more
questionable when natural materials are In a chemical, man-made
environment.
To meet the requirement that granular materials be chemically
resistant to the expected waste liquids, Landreth [1987] suggested a
testing procedure for leachate collection systems which may be adapted
to leak detection systems:
"When rock or gravel is used as drainage material in the leachate
collection system, the owner/operator should verify that the mineral
content of the rock Is compatible with the waste/leachate mixture.
The owner/operator will need to demonstrate that the rock will not be
dissolved or form a precipitate that would clog the leachate
collection system."
Synthetic Drainage Materials. The chemical compatibility of synthetic
drainage layers 1s poorly documented. Needlepunched nonwoven
geotextlles, mats and waffled structures are made with polypropylene
2.4-27
-------
or polyester. Both of these polymers have a high resistance to a wide
range of chemicals. They can both be attacked, however, by chemicals
from certain chemical fj.. Hies.
Nets are made of polyethylene. Polyethylene FHLs such as high
density polyethylene (HOPE) are most often selected to line hazardous
waste management unite because of their resistance to attack by a wide
range of chemicals. For this reason, and because the thickness of
their strands 1s comparable to the thickness of HOPE FMLs, nets are
also expected to exhibit a high resistance to attack by a wide range
of chemicals.
Testing of synthetic drainage layers should be performed to
evaluate their chemical compatibility. Testing should Include
immersion of synthetic drainage layer specimens In the expected waste
liquids for extended periods of time in accordance with USEPA Method
9090. Specimens should be removed at time intervals specified by
Method 9090 for hydraulic transmissivlty testing. Testing should be
conducted on specimens that have been immersed and on a control
specimen (I.e., a specimen that has not been immersed) to evaluate the
effect of Immersion on synthetic drainage layer hydraulic
transmissivity.
Geotextiles. As discussed previously, the polymers used to
manufacture geotextiles, polypropylene and polyester, have high
resistance to a wide range of chemicals. In a given category of
polymer, however, woven geotextiles are usually expected to exhibit a
better chemical resistance than nonwoven geotextiles because of their
usually thicker yarns.
A preliminary selection of candidate geotextiles should be
performed using available chemical compatibility data available from
polymer manufacturers. Once a given type of geotextile has been
selected, specimens of the selected geotextile should be immersed In
the expected waste liquids in accordance with USEPA Method 9090.
Specimens should be removed at specified time intervals for index
property testing. Testing should be conducted on specimens that have
been immersed, and on control specimen (i.e., a specimen that has not
been immersed) to evaluate the effect of immersion on the geotextile
properties. The tests to be performed should include as a minimum the
2.4-28
-------
ASTM D1682 grab strength test, ASTM D751 puncture strength test, and
ASTM D1004 trapezold tear strength test.
Pj.i'tlc Pipes. The most commonly used category of plastic pipes in
wa 2 management units are the thermoplastic pipes. Thermoplastic
pipes include polyvinyl chloride (PVC) and high density polyethylene
(HOPE). PVC and HOPE are known for their high resistance to a wide
variety of chemicals. HOPE pipes are expected to exhibit a chemical
resistance equivalent to HOPE P'Ls. PVC pipes are expected to exhibit
a chemical resistance superior to the chemical resistance of PVC
geomembranes since pipes are made from almost pure PVC, while
geomembranes are made from plasticlzed PVC, I.e., PVC mixed with
plasticizers whose chemical stability is not always as good as PVC's.
The chemical resistance of other plastic pipes should be evaluated
based on available compatibility data.
Testing of plastic pipes (thermoplastic or others) should be
performed to evaluate their ability to withstand loads after Immersion
with expected waste liquids. The testing procedure which follows has
been adapted from Landreth [1987]. Specimens of the plastic pipe
should be prepared for strength testing per ASTM 02412 or equivalent
and at least one prepared specimen should be immersed In the expected
waste liquids 1n accordance with USEPA Method 9090. After the
immersion test, the pipe specimen should be dried (per Method 9090)
and subjected to a strength test (ASTM D2412, paragraphs 6-9).
Testing of an Identical non-immersed (control) specimen should be
performed. A report should be prepared similar to that outlined In
ASTM D2412 paragraph 11 (Including 11.1.7 and 11.1.9) comparinj. the
test results of the Immersed and control samples.
Structures and Other Pipes. Structures and other pipes have been
Identified in Section 2.4.2.5 and 2.4.2.6. They are made with either
one of the following materials: plastics, reinforced concrete,
metals, fiberglass, and acrylonitrlle-butadiene-styrene. The chemical
ccmpatibi11ty of the structures and other pipes should be evaluated
with as much care as exercised for other components of leak detection
systems. Ideally, compatibility testing should be performed on every
one of these structures and other pipes.
2.4-29
-------
However, such testing may be Impractical to perform because of the
large variety of parameters Involved:
• Shape of structure: some structures maybe circular (e.g.
manholes), others may oe quadrangular (e.g. sumps).
• Size of structure: some manholes may have a 1.2 m (4 ft) .ner
diameter.
• Type of material.
Available literature on the chemical behavior of these structures
and other pipes should be used to evaluate their compatibility with
the expected waste liquids.
2.4.3.6 Mechanical Effects of Drainage Materials on FML Liners
2.4.3.6.1 Granular Drainage Materials
Sand and small rounded gravel should not damage FHLs. In fact,
sand cushions are often used to protect FHLs, as discussed in Section
2.4.2.6. Angular gravel (e.g., crushed aggregate) can abrade,
scratch, and puncture FHLs. A sand or geotextiles cushion should
always oe placed between angular gravel and the FML.
/*'! granular drainage materials are placed Msing equipment such as
bulldozers or front-end loaders. Stress generated by ths wheels or
the tracics and misuse of he equipment can damage the FML.
2.4.3.6.2 Synthetic Drainage Materials
Needlepunched nonwoven geotextiles and mats do not damage FMLs
even under relatively high compressive stresses because they are
flexible and smooth. Under low compressive stresses, nets will not
damage FMLs. Under high compressive stresses, nets can damage FMLs if
the net Is stlffer than the FML. Nets should not be used with some
types of FMLs unless a cushion layer 1s placed between the net and the
FML. Waffled structures collapse In a brittle mode at a compressive
stress of 200 to 300 kPa (4,000 to 6,000 psf) and can damage FMLs.
For this reascn waffle structures are usually not used with FMLs In
waste management applications.
2.4-30
-------
Unlike granular drainage materials, all synthetic drainage
materials are placed by hand, which reduces the potential for damage
to the underlying FML by construction equipment. However, care should
be exercised not to entrap stones nor leave other sharp objects in the
synthetic drainage material during Us placement.
The potential for damage common to all available synthetic
drainage materials results from placement of the low-permeability soil
component of a composite top liner. Synthetic drainage layers are
relatively thin and do not protect the FML from construction
equipment-generated stresses as a 0.3 m (1 ft) thick drainage layer
made of natural materials would. A loosely compacted 11ft of low
permeability soil should be Initially placed on top of the geotextlle
covering the synthetic drainage layer to protect the underlying FML,
until subsequent lifts of the top Hner low-permeability soil can be
placed and compacted.
2.4.4 Conclusions
This section of the background document has provided a review of
the materials used to construct leachate collection and removal
systems. This section has also presented a discussion of the
properties of these materials, the factors that affect the materials,
and typical properties. The Important hydraulic properties of
drainage media were listed as hydraulic conductivity and hydraulic
transmlsslvlty and ranges were given for both of these properties for
both granular and synthetic drainage media. These ranges of
properties are Important because the limits of the range define the
limits of performance of LDCRS type leak detection systems. The
values of the parameters used to define BOAT performance must fall
within the ranges of properties defined in this section. The ranges
of hydraulic conductivities (k,j) and hydraulic transmlssivlties (8^)
reported in this section for granular and synthetic drainage materials
were (0^ for sands are based on a 0.3-m (1-ft) thickness):
• sand
kd - 10~" to 10"2 m/s (10"' to 1 cm/s)
6d - 10"' to 10"1 m'/s (0.005 to 5 gpm/ft)
2.4-31
-------
• gravel
kd - 10"2 to 1 m/s (1 to 10! cm/s)
0d - 10"' to 10"' m'/s (5 to 500 gpm/ft)
• synthetic drainage materials (nominal values)
kd - 10"4 to 1 m/s (10~J to 101 cm/s)
8d - 10"* to 10~2 m'/s (0.005 to 50 gpm/ft),
2.4-32
-------
Table 2.4-1. Filter criteria for geotextiles, adapted from
[Christopher and Holtz, 1984; Gircud, 1932], Note:
0», is the apparent opening size of the soil.
1. RETENTION CRITERION
1.1 Soils with less than 50% particles < 0.075 mn (US Sieve 200)
Density Index Coefficient of
of the soil uniformity of the soil
(Relative density)
1 < Cu < 3 Cu > 3
loose q
soil ID < 35% 0,, 10 ksoi1
2.2 Non-critical and Non-severe Applications
kgeotextile > ksoil
3. CLOGGING CRITERION
Nonwoven geotextiles: porosity n > 30%
Woven geotextiles: percent open area A > 4%
2.4-33
-------
LA.£/1
0 lOPLIA/CR CoNSISilMG 01- A FML ALONE
LfACUNTE COLLECTION AHb
REMOVAL SYSTEM »ar SHOW/I
I-I1L
ml' Lif/EP. LOV,'
i.VbOTEX,7-|LE FILTER
-br'/HIEIIC
-Borrow FhL
LOW
b. COMPOSITE TOP LIIILR
Figure 2.4-1. Synthetic leak detection collection, and removal
systems.
2.4-34
-------
(a)
(c)
(.1)
Figure 2.4-2. Available synthetic drainage materials: (a)
needlepunched nonwovcn geotextile; (b) mat; (c) net; and
' (d) waffle. Scale: the diameter of the coin is 24 mm
(1 In.) [Bonaparte et a)., 1985].
2.4-35
-------
Load
L
H
Q
i H/L
Figure 2.4-3. Hydraulic transmissivity test device [Giroud and
Bonaparte, 1984].
2.4-36
-------
(pi)
/'<>X
^roco'irnsi I ^ •• i / (I (tnc»iri~
C.fl.C A>lb C'.CofS' I itg /fLl
ZOO
*» (M,.) ""'
1
crj
Figure 2.4-4. Influence of comoressive stress on synthetic drainage
material hydraulic transrnissivity and thickness [Giroud,
1987].
2.4-37
-------
HYDRAULIC TRANSMISSIVITY (m2s-')
IU
W3
10"'
10's
iO'b
/n-;
, .. . — — _
•^
\^^ ^
N^
__i=_p.03
T^T
"^
^^
-v^^
^-^
— —
^^
^^^
<;^0/'(£//fi/?AA/£
— • t~,GotJr-r
GEoiiBn&nAHE
CEorBxriLE
f,Eout=r
(,EonEne>K\h£
6cor£xrii.E
0 500 1000
COMPRESSIVE STRESS (kPa)
Figure 2.4-5. Influence of boundary conditions and compressive stress
on hydraulic transmissivlty of nets. Mote: Chart
established by GeoServices using values published by
Bonaparte et al. [1985].
2.4-38
-------
LEGEND
= in-place soil
= D^, soil parlide,
entrapped in lilter
= soil which has
migrated into filler
and is held by D^,
size soil particles
Nominal boundary
before stabilization
under seepage
Uer, (soil)
Figure 2.4-6. Illustration of prevention of piping [Cedergren, 1977].
2.4-39
-------
2.5 LIQU'DS MEASURED IN LDCRS AT OPERATING L'lITS
2.5.1 Introduction
To date, only very limited data have been gathered on the amount
of liquids collected in the LDCRS sumps of land disposal units with
double liner systems designed and constructed to current standards.
Undocumented claims from a number of sources indicate that the amount
of liquids collected in facilities constructed to current standards is
small. However, without documentation, and without a clear
understanding of the design, construction and operation of each
individual unit, interpretation of data on the amount of leakage
collected is extremely difficult. While the ir ial undocumented
reports are encouraging, more (and better documer.:ed) data must be
solicited from owners and operators so that the performance of double
liner systems, and top liners and LDCRS in particular, can be
documented and understood.
In the remainder of this section, currently available, documented
1 r]fnrrm Mon (r |'n"|n''F!,
I I I i v > ' - ->' •- ' - ' •
2.5.2 Institute of Chemical Waste Management Data
In September 1986, the Institute of Chemical Waste Management
(ICWM) issued a report providing data on the quantity of liquids
collected in LDCRS sumps at four double-lined landfills and one
double-lined surface impoundment. Data were provided on lining system
design details and quantities of liquids collected in the LCRS above
the top liner and LDCRS between the top and bottom liner. No
information was provided on the geographi location or site
hydrogeology of the five units described in the ICWM report.
The lining system details for the four landfills and the surface
impoundment of the ICWM report are shown in Figure 2.5-1. It can be
seen that in all cases the lining systems utilized composite top and
bottom liners (a double-composite double liner system). In three of
pi ii v i id1 • .n I" I i I I (i II. I I i ii I
-------
ICVIM study. An important point to be observed in Table 2.5-1 is that
at four of the five facilities construction had been completed for a
year or less. Table 2.5-1 provides a summary of the IC'nM LDCRS data.
These data indicate that the average rate of liquid collection in the
LDCRS sump at the units varied from 1.5 to 18.9 1 iters/1000m2/day (1.5
to 18.9 gallons/acre/day), with three of the four reported values
being below 6.7 11 ttrs/lOOOmVday (6.7 gallons/acre/day). The maximum
rate of liquid collection for each unit (based on a monitoring period
of one week) 1s also reported in Table 2.5-1. It can be seen that,
with the exception of Landfill D, the maximum rate of liquid
collection tends to be on the order of several times the average rate
of collection. In Landfill D, the maximum rate of liquid collection
is almost an order of magnitude larger than the average rate.
Overall, the amounts of liquid collected in the units reported by
ICWM are small. The following conclusions can be drawn if the leak
detection systems at the considered units have been properly designed
and constructed:
• The hydraulic heads acting on the bottom liners of al. jf these
units are very small or negligible.
• Since tr hydraulic heads on the bottom liners are very small,
and the bottom liner at each unit is a composite, leakage into
the bottom liner at each unit should be extremely small or
negliglble.
While it can be inferred (provided that design and construction
are adequate, as indicated above) that leakage into the bottom liner
at each unit is extremely small or negligible, the data presented in
Table 2.5-1 allows only limited inferences to be made on the rate of
leakage through the top liner into the LDCRS since:
• Landfills A, B and D contain a sand LDCRS drainage medium
having a hydraulic conductivity on the order of 10"" m/s (10~*
cm/s). As will be shown in Section 2.6, a sand with this
hydraulic conductivity exhibits significant capillarity (See
Table 2.6-4). As a result leakage through the top liner will
be held by capillary tension in the pore space of the sand.
Drain flow will not occur until the sand has absorbed enough
2.5-2
-------
water to fill much of its pore space. As indicated in Section
2.6, for a sand with kj = 1 x 10"" m/s (1 x 10"J cm/s), the
time requ 2d to fill this rT« space can be as laigo is
several years, even for re!atu..y long top liner leakage rates
such as 100 Ltd (100 gpad). Since Landfills A, B and D are
only 1 to 2 years old, leakage that has occurred through their
top liners may not have yet appeared in the LOCRS sump.
• In aJ five ICWM units, the liquids collected in the LDCRS sump
may not be only due to top liner leakage. The liquids in the
su.':ps may be due to other sources of liquids such as rainwater,
ground-water infiltration, and/or water expelled by
consolidation from the compacted soil component of the
composite top liners. In fact, it will be shown in Section
2.10 that liquid quantities in excess of those cited in the
ICWM report can be entirely accounted for by consolidation of
the compacted soil component of the top liner.
• One conclusion can be drawn for Landfill C and Surface
Impoundment A. Since synthetic drainage nets are used as the
LDCRS drainage medium, there is no significant capillary
tension in the pore space of the leak detection layer. Since
there is no capillary tension, there is also no capillary rise
and thus the initial "wetting up" period for the drainage
material in these units is minimal. Therefore, little liquid
is being held in storage in the LDCRS and the quantity of
liquids collected can be considered as an upper-bound of the
actual top liner leakage. In addition, since there is a strong
possibility that the collected liquids are from sources other
than top liner leakage, the actual top liner leakage rate is
probably significantly l»ss than the values cited in Table
2.5-1.
It will be useful to monitor the units cited in the ICWM report
over time. Continued monitoring is strongly encouraged.
2.5.3 :ase Study - Landfill in South East U.S.
An industrial company designed and constructed a double-lined
landfill for the purpose of containing solid hazardous waste. The
2.5-3
-------
site is located in the South East U.S. The ground-water table
elevation is below the bottom of the -it. The lining system of the
unit shown in Figure 2.5-2 is compc J of the following components
(from top to bottom):
• 0.15-m (0.5-ft) thick sand layer;
• geotextile filter;
• 0.3-m (1-ft) thick sand LCRS;
• 2-mm (80-mil) thick HOPE top Flit;
• 0.3-m (1-ft) thick sand LDCRS; and
• composite bottom liner composed of (from top to bottom):
. geotextile filter;
1-m (3-ft) thick compacted clay; and
. 1-mm (40-mil) thick HOPE FML.
(It should be noted that in this lining system cross-section, the
order of the components of the bottom composite liner are reversed
(compacted soil on top of FML) from the order normally used for
composite liners (FMl on top of compacted soil). The reason for this
configuration is unknown.)
The landfill is approximately 39 m by 145 m (128 ft by 474 ft),
and the LCRS and the LDCRS ara continuous across the landfill.
Liquids are removed from the LCRS and LDCRS at ten outlets eat . with
each outlet draining an area of approximately 560 m2 (6000 ft2). In
the following discussion the outlets are referred to as LCRS outlets 1
through 10 and LDCRS outlets 1 through 10.
The quality assurance of installation was performed by the owner
and details regarding the installation procedure of the bottom liner,
geotextile filter or HOPE top liner are not available. However, it is
known that the LDCRS sand was dumped in 150 mm (6 in.) lifts,
saturated and compacted with a vibratory roller. During the
installation of the top liner, the LDCRS sand was allowed to drain,
2.5-4
-------
but, during this period, rainfall repeatedly infiltrated the LDCRS,
and it is possible that th° LDCRS was nearly or fully saturated at trie
tin? cf top liner installation.
After installation of the top FML, but prior to installation of
the LCRS sand, the landfill was flooded with water to an average depth
of 0.6 m (2 ft). (The depth across the landfill varied because of the
27, bottom slope.) The landfill was flooded for a two month period
and the LDCRS was monitored. Initially, leakage was detected 'n all
ten LDCRS outlets, with the rate of leakage decreasing with time. The
initial average rate of leakage observed was 0.9 1 iters/minutp (340
gallons/day) per outlet. After approximately 1 month, leakage from
LDCRS outlets 6,8,9, and 10 had stopped, leakage rates from LDCRS
outlets 1,3, and 5 were decreasing to insignificant levels, and
leakage rates from LDCRS outlets 2 and 4 were fairly constant at rates
of 0.7 liter/minute and 0.3 liter/minute (260 gallons/day and 110
gallons/day) respectively. The rate of leakage from LDCRS outlet 7
was erratic.
The leakage that was initially collected at all ten LDCRS outlets
is attributed to drainage of existing water in the LDCRS sand and
compression of the sand under the surcharge of the water ponded on the
top FML. The leakage from outlets 1, 3 and 5 that was observed latei
on was not attributed to leakage through the top FML, but to continued
d.ainage of the LDCRS. However, the leakage observed at LDCRS outlets
2 and 4 was most likely due to leakage through holes in the top FML.
An equivalent hole size can be backcalrulated from the leakage
rates observed from LDCRS outlets 2 and 4, using Bernoulli's Equation
for flow through an orifice (see Section 2.2). Bernoulli's Equation
is applicable because there is only a small surcharge on the FML (the
weight of water) and the FML will float slightly in the vicinity of
the hole. The rate of leakage observed for LDCRS outlet 2 (0.7
liter/minute or 260 gallons/day) corresponds to a hole 2.7 mm (0.1
in.) in diameter. The rate of leakage observed for LDCRS outlet 4
(0.3 liter/minute or 110 gallons/day) corresponds to a hole 1.8 mm
(0.07 in.) in diameter. These calculated hole sizes are smaller than
the "standard" hole size considered in Section 2.2.3.3.6. On the
basis of these two calculations, the "standard" hole size used for
calculation of leakage through KMI holes is slightly conservative. It
2.5-5
-------
is Interesting to note that the backcalculated hole sizes correspond
almost exactly to the "small" hole size given in Table 2.2-15.
Eventually, waste was placed in the aieas drained by LDCR3 outlets
1 through 5. The flow rate of iqinds draining through the LDCRS was
monitored at the outlets. Initially, a high rate of liquid flow was
observed (0.5 to 1.0 liter/minute (200 to 400 gallons/day)) for all
five outlets. After 30 days, the flow rates had gradually decreased
to a total flow rate of 0.25 liter/minute (100 gallons/day) for all
five outlets. The observed leakage from outlets 2 and 4 after 30 days
were 0.08 and 0.10 liter/minut? (29 and 37 gallons/day), respectively.
This leakage rate is approximately an order of magnitude less than the
rate which was observed durir,- the ponding test. The rate of leakage
was lower than during the ponding test because the hydraulic head
acting on the top liner was lower. The observed rate of leakage was
attributed to consolidation of the bottom compacted clay layer as well
as continued leakage through imperfections in the top FML in the
vicinity of outlets 2 and 4.
After placement of the waste in the area drained by outlets 1
through 5, but prior to the placement of waste in the areas drained by
LDCRS outlets 6 through 10, the landfill was inundated with rain.
The LCRS outlets 6 through 10 were intentionally plugged to perform
an ,er ponding test and approximately 0.3 to 0.6 m (1 to 2 ft) of
water was collected. LDCRS outlets 6 through 10 were monitored and
leakage was observed only from outlet 7. The landfill was drained and
the LCRS sand was removed in the area drained by outlet 7. A
confined ponding test was performed in the vicinity of the point where
the LCRS outlet pipe penetrated the top and bottom liners and a leak
was found in the top FML boot over the pipe. The boot was repaired
and another ponding test was performed; no leakage was collected
during this second ponding test. The leakage through the faulty boot
was approximately 1000 liters/day (250 gallons/day) under an average
hydraulic head of only 0.15 m (6 in.). A leakage rate of this
magnitude could be caused by a hole 4 mm (0.15 in.) in diameter.
Two conclusions can be drawn from this case study:
• First, it is important to distinguish the source of the
collected liquids when evaluating lining system performance.
2.J-6
-------
2.5.4
The rate of l°akage that was initially observed during the
ponding test wus rot alt attributed to lea'r.3ge through the top
FML. At least part of the collected liquid was due to drainage
of rainwater that had been entrapped in trie LDCRS sand. Also,
during placement of the waste, part of the collected liquid was
attributed to consolidation of the compacted soil bottom liner.
Second, leaks can occur at locations where pipes penetrate the
landfill. These penetrations are often the most difficult part
of liner Installation. Nondestructive testing around
penetrations is sometimes impossible. Therefore, such
connections should always be carefully checked.
Case Study - Two Landfills In North Central U.S.
Two double-lined landfill units were constructed in 1985 and 1986
in the North Central U.S. for disposal of solid hazardous waste. The
site of the units is located in a region that experiences about 0.7 m
(2-5 ft) of rainfall annually. The ground-water elevation at the site
is apparently below the elevation of the bottoms of the two units.
Th£ depths of the units vary and range from about 9 to 15 m (30 to 50
ft''. The depth of the units depends primarily on the depth at the
site cf a native clay deposit. The two landfill units will be
referred to as Unit 1 and Unit 2.
Lining system details for Unit 1 and Unit 2 are shown in Figure
2.5-3. Tha most significant difference between the lining systems at
the two units is that the top liner at Unit 1 is a single FML, while
the top liner at Unit 2 is a composite made up of an upper FML
co-conent and a 1.5-m (5-ft) thick lower compacted soil component. In
boii units, the LDCRS drainage media are synthetic drainage nets, so
stc-age of liquid in the LDCRS (due to capillary tension) is
neg'igible. The lining systems for the two units extends all the way
up the 2H:1V side slopes with the top and bottom liner FMLs anchored
int: the same trench at the top of the slopes. In both units, the
LC:.S above the top liner is composed of sand on the bottom of the
units and synthetic drainage nets on the side-slopes. Quality
assurance of lining system installation was carried out by the owner,
wif quality control of FML installation performed by the FML
instil ler.
2.5-7
-------
Information on the quantities of liquid collected in the LDCRS
sumps of the two units are summarized in Table 2.5-2. Referring first
to Unit 1, it can be observed that the average quantity of liquid
collected in the LOCRS sump is on the order of 5 Ltd (gpad) and the
maximum quantity collected is on the order of 40 Ltd (gpad). Since
Unit 1 uses syntnetic drainage nets in the LOCRS (which result in
negligible liquids storage due to LDCRS capillary tension) and an FML
top liner (with no compacted soil component), and since the ground-
water elevation 1s reported to be below the elevation of the LDCRS,
the liquid in the LDCRS sump is assumed to be due to leakage through
the top liner. A rate of top liner leakage of 5 Ltd (gpad) for Unit 1
corresponds to a single FML hole in the Unit 1 tcp liner (over the
entire 14 acres) of about 3 mm (0.04 in.) diameter (assuming a
hydraulic head on f ? top FML of 0.03 m (0.1 ft)).
From Table 2.5-2 it can be seen that the quantities of liquid
collected in Unit 2 are on the order of ten times larger than the
quantities collected in Unit 1. A significant part of this difference
in leakage quantities can be explained by consolidation of the
compacted soil component of the composite top liner. (This may be
partially supported by the fact that GCMS priority pollutant scans of
liquids collected in the LDCRS sump in December 1536 indicated that
the collected liquids "contained no organic constituents" and had
heavy metal concentrations "consistent with locdl background levels".)
On the other hand, it may not be legitimate to attribute all of the
collected liquids to consolidation of the compacted soil. In
addition, it does not seem likely that the liquid is due to leachate
migration through the top liner since the leachate would be required
to migrate through 1.5 m (5 ft) of clay, which would be unlikely in
the short elapsed time since Unit 2 was put into operation. To
understand the source of all of the liquids collected in the LDCRS
sump of Unit 2 will require a very careful revi?^ of the design,
construction and operation of the unit. As pointed out in Section
2.5.1, it is only through a very careful review of all of the factors
affecting each unit that the leakage quantities collected can be fully
understood.
2.5-8
-------
2.5.5 Case Studies -Surface Impoundments 1n__F 3st Central a_nd
South West U._S_._
These case studies provide data on the quantity of liquids in the
LDCRS sumps of surface impoundment units at two sites. The studies
also provide data on lining system design, but contain no data on the
construction quality assurance, .peration or site hydrogeology of the
units. Therefore, the information presented in these studies should
be considered as preliminary until more extensive documentation
becomes available.
2.5.5.1 Surface Impoundments 1n East Central U.S.
Two surface impoundment units were constructed around 1985 in the
East Central U.S. for disposal of liquid waste. These units are each
2000 m1 (0.5 acre) in size and the liquid depth is approximately 6 m
(20 ft). No data is available on the ground-water elevation at the
site.
The lining system is the same for both of these units and details
are shown in Figure 2.5-4a. Both of these units have double liner
systems. The top liner consists of a 1.5 mm (60-mil) thick HOPE FML
and is underlain by a LDCRS composed of a 0.3 m (1-ft) thick layer of
clean sand on the bottom of thu unit and geonets on the side slopes.
The LDCRS is in turn underlain by a composite bottom liner comprised
of a FML placed on top of compacted soil layer. No data is available
on the quality assurance of lining system installation.
During the first 6 to 8 months of unit operation, the quantity of
liquids removed weekly from LDCRS sumps ranged approximately from 60
to 120 liters (15 to 30 gallons). The calculated average quantity of
liquids collected in the LDCRS sump is on the order of 6 Ltd (6 gpad).
During the 6-8 month period, the quantities of liquids removed
decreased. After this period to date, liquids were no longer detected
at LDCRS sumps. The liquids removed from LDCRS sumps were analyzed for
chemical constituents. The liquid total organic constituents (TOC)
concentration ranged from 10 to 40 parts per million (ppml during this
6-8 month period. The TOC of the pond influent during the same period
averaged about 400 ppm. The quality of the liquids removed was
reported to have improved during this 6-8 month period.
2.5-9
-------
The nature of the LDCRS granular drainage medium is unknown and no
data is available on ground-water elevation. Therefore, the
conclusions presented hereafter can only be preliminary and can only
be made by making a number of assumptions. The discussion which
follows assumes that the LOCRS drainage medium consists of sand and
that ground water table is below the LDCRS.
The relatively large quantities of liquids removed in LDCRS sumps
at the beginning of 6-8 month period may be attributed to construction
water (I.e., rainwater or other) draining slowly toward the LDCRS
sumps. Drainage continued in the LDCRS drainage medium up to the
point where the capillary tension in the pore space of the drainage
medium became greater than gravity forces. After the drainage period,
no liquids were observed in LDCRS sumps because any leakage through
the top liner would be held by capillary tension in the pore space of
the sand. As a result, drainage would not occur until the sand has
absorbed enough liquids to fill much of its pore space. As indicated
1n Section 2.6, the time to fill this pore space can be as large as
several years for a sand with kj = 10"4 m/s (10~* cm/s), even for
relatively large top liner leakage rates such as 100 Ltd (100 gpad).
Therefore, if small amounts of leakage are occurring, it is probably
still being held by capillary tension in the LOCRS drainage medium.
The TOC of the liquids removed from LDCRS sumps was significantly
lower than the TOC of the pond influent. This further supports that
the liquids collected in the LDCRS sumps during the initial 6-8
month period were primarily due to drainage of waLcr entrapped in the
LDCRS drainage medium during construction. The relatively minor
amount of TOC in the liquids removed may have been caused by spills
from construction equipment or by a small leakage through the top
liner that diluted with the larger amount of construction water
draining within the LDCRS drainage medium to the LDCRS sumps.
Continued monitoring of the LDCRS is strongly encouraged to verify
the conclusions drawn above for these units.
2.5.5.2 Surface Impoundments 1n South West U.S.
Thirty-two surface impoundment units were constructed in the mid-
1980's in the South West U.S. for disposal of liquid waste. For these
2.5-10
-------
units, My quantities of liquids in LDCRS sumps and lining system
detai 1 s ,.~e aval lable.
The lining system is the same for all of these units and details
are shown In Figure 2.5-4b. All of these units have double liner
systems which meet EPA 1985 minimum technology standards. The top
liner consists of a 2.5-mm (100-mll) thick HOPE FML and the bottom
liner is a composite constructed of a 2.5-mm (100-mil) thick HOPE Fi'L
on top of a compacted soil layer. The LDCRS between the liners
consists of a 4-mm (0.16 1n.) thick synthetic drainage net layer
connected to drainage pipes. The pipes were placed in a collector
trench at the bottom of the unit. The ti ,nch was backfilled with
si 1ty, sandy gravel.
The CQA program at the site called for the FML installer to carry
out ponding tests in each completed unit to search for top liner
leaks. The ponding tests were conducted by filling the ponds to a
level 0.6 m (2 ft) above operating pool level. It is reported that
only one of the 32 ponds produced liquids in the LDCRS sump during the
ponding tests. An evaluation indicated that the water 1n the or°
LDCRS sump was construction water. It is safe to say that the ponding
tests demonstrated that there were no major leaks in the top liner.
However, small leaks (if they were present) in the range of 20 Ltd
(gpad) may have gone undetected during the ponding tests lue to the
storage capacity and capillary tension of the collector trench
backfill.
Since the beginning of operation of the first units at this
facility almost two years ago, no leakage has been reported. These
reports therefore indicate that top liner leakage at the units is
either negligible or very small. Since the bottom liners at these
surface impoundments are composites, the detection sensitivities
(defined subsequently) of the units should be very good. Therefore,
the detection of zero leakage in the LDCRS sumps indicates that the
top liners are not leaking (i.e., zero macroscopic defects were
achieved in these Installations) or, at worst, that only very small
(e.g., not more than 5 to 10 gpad) amounts of top liner leakage is
occurring and it is not being detected because it is evaporating out
of the LDCRS sump or it is being held by ripillary tension in the
collector trench backfill.
2.5-11
-------
Table 2.5-1. Quantities of liquid collected and removed from LDCR3
sump at units cited in the Institute of Chemical i.'-iste
Management Report [1906]
Cod' lycc of Surface Area Date (OCRS
land (acres) Con5tructlon Drainage Medium
Disposal Completed
Unit
quantity of liquids tollcctcd and Removed
from IOCRS during Active life
typical Avei.lg»! fypioil Mi«ipnu
-------
Table 2.5-2. Quantities of liquid collected and removed from LDCRS
sump at two landfill units in North Central U.S.
Code type of Surface Area Da- IDCR5 lop Quantity of liquids C'_ cted and Removed
Land (acres) Con;., .lion Drainage Medium Liner from LDCKS during Active Lire
Disposal Completed
Unit
lyplcal Avtrage Typirji Minimum
gal Ions/a ere/day gaIlons/acie/day
1 landfill 14 Sunnier 1905 Net, 1 layer I ML S 40
2 landfill 9 Simmer 1986 Net. 2 layers Composite 100 500
(FML. clay)
Hole. I - Details of double liner systems sliown on figure 2,5-3,
2.5-13
-------
~T/&.
0.9m (Jfl)
v-tv/-
Landfill A
0.3m(l[t) V^',\.^__-
0.6 n
01- (MO
Landfcll E>
LonJf,ll C
x=w-x--
Landfill D
a 3 r. (ifl)
0.5 1- (3 (t)
5urj««.
LEGEND
___ Geoteictile,
- HDPE
LnJf.!! 0, I.
fll II
sr oil olner Ca
^^el" ^ one. (ntj^r
lif-f I
all el*tr caitt j
Cotnhac'M CWu /
Y j\ Loose claii
Figure 2.5-1.
Details of lining system from units cited in Institute
of Chemical Waste Management Report [1986].
2.5-14
-------
£h>m (XOm.'h
HDPE FM i
tntn
HbPE FML
Figure 2.5-2.
Details of lining system for case study of landfill in
South East U.S.
2.5-15
-------
UN IT 1
(150m,!
Silly fine Sand 0.3 >v-> (.12 in)
< XX X X XX XXXJvl.Smrn
'////////A
MDPE FML
UN IT 2
HDPE
Sill fine sand
(12. i
Cornpac+ed dlq>| Imer/M l.bm C5
ZZ /L/_/l..''-Z Z_/_/JX^2:?0 0/^n1-
\ ( O
XXX XX XX XXXV (Z uwe.O
ft)
PP
Figure 2.5-3. Details of lining systems for case study of two
landfills in North Central U.S.
2.5-16
-------
is. («;o..J;!L.!
•'.' -LbCKS nmnatnr dini-xny.
I
FML
.. STJOC*- irMbo
I./
tnlt, in cos!" Cenira.1 U.S.
X. X X X X X .X.
2.S.rmn ClOO.mil) fliick
HOPE FML
II I' I ' „,(-
^ —, iunulK^C. of^ir^A'\5_ nci
s v—' -* "-i r
HDFE rni_
D- Surtact. imbourxJmenls in OOum WesP U.S.
Figure 2.5-4.
Details of lining systems for case studies of surface
impoundments in East Central and South West U.S.
2.5-17
-------
2.6 ANALYSES OF THE FUNCTIONING OF LEAK DETECTION SYSTEMS
2.6.1 .Introduction
2.6.1.1 Purpose
The purpose of Section 2.6 is to provide insight into the factors
affecting the performance of leak detection, collection, and removal
systems ("leak detection systems"). This is extremely important
because several of the main technical areas addressed in the proposed
Liner/Leak Detection Rule are dependent on the performance of the leak
detection systems. These areas are listed below:
• Performance criteria for leak detection systems, which are
discussed in Section 2.7.
• Design specifications for leak detection systems, which are
discussed in Section 2.8.
• Action Leakage Rate (ALR), which is discussed in Section 2.9.
• Response Action Plan (RAP), which is di-cussed in Section 2.10.
A thorough understanding of the functioning of leak detection
systems is necessary to establish the elements of the proposed
Liner/Leak Detection Rule in the four technical areas listed above.
2.6.1.2 Oyeryjew_of Leak Detectjon_System Functjoning
When leakage occurs through the top liner, it takes some time
before the leakage is detected because of the following three reasons
(see Figure 2.1-3):
• The leakage that impinges into the leak detection system may
be retained by capillarity in the drainage medium during an
Initial per1erl of time, Instead of flowing to the sump, Then,
flow takes place when some portion of the leak detection system
becomes saturated. The size (thickness and width) of this
saturated portion depends on the size of leak, initial
2.6-1
-------
capillary tension in the leak detection system, and pcrosity
and thickness of the leak detection system drainage rciedi--.
• Flow to a drain will be further delayed if the leak detection
system is underlain by a compacted soil bottom 11r°r. As shown
in the Notice of Availability of Data on "Perfor :e of 2ottcm
Liners at Double-Lined Landfills and Surface Impouni-ents"
[USEPA, 1987], compacted soil bottom liners will permit en the
order of 100 Ltd (gpad) of leachate flow into and through the
bottom liner (under steady-state conditions). In contrast,
composite bottom liners having a FML upper component will
permit less than 1 Ltd (gpad). In the remainder of this
document, 1t will be assumed that the bottom liner is a
composite with a FML upper component and a compacted soil
bottom component. Since leakage into composite bottom liners
1s extremely small under almost all circumstances [U5EPA,
1987], the simplifying assumption will be made in subse:-jent
calculations that composite bottom liners are impermeable.
• Finally, once flow begins, it takes time for the leakage to
flow from the location of the leak to the sump.
From the above discussion (and assuming an impermeable composite
bottom liner), 1t is clear that two phases of the functioning of a
leak detection system should be investigated: (i) the peric? of
initiation of the flow; and (11) the steady-state flow.
The parameters which govern the functioning of a leak detection
system are:
• Parameters related to the top liner: leakage rate through the
top 1iner, type of leak.
• Parameters related to the leak detection system:
characteristics governing storage capacity (size of the voids,
porosity, thickness, Initial water content); and
characteristics governing the flow (hydraulic conductivity,
thickness, slope, length of flow path).
2.6-2
-------
• Parameters related to the bottom liner: leakage rate through
the bottom liner, type of leak.
The studies presented in Section 2.6 will provide information
regarding the influence of these parameters on the performance of leak
detection systems. Other pertinent data are provided in Sections 2.2
(on leakage through top liners), Section 2.4 (on characteristics of
leak detection systems), and Section 2.5 (on performance of leak
detection systems). Pertinent Information on the performance of
bottom liners and their influence on the functioning of leak detection
systems may be found in USEPA [1987].
The functioning of a leak detection system may also be
significantly affected by sources of water other than leakage, such
as:
• Rainwater entrapped in the leak detection system during
construction of the unit.
• Water used to compact the low-permeability compacted soil
component, if any, of the top liner.
• Water flowing into the waste management unit as a result of a
rise in the ground water table elevation.
Sources of 1.quid (water) other than leakage through the top liner
affect only the Response Action Plan (RAP) a.nong the four technical
issues listed in Section 2.6.1.1. Therefore the effect of sources of
liquid other than leakage through the top Hner will be discussed only
in Section 2.10.
2.6.1.3 Definitions
Terms and phrases specific to the functioning of leak detection
systems are used in Section 2.6 and need to be defined.
2.6-3
-------
2.6.1.3.1 Leakage
In this context, the term leakage is used for "leakage chrough the
top liner". There is leakag-_- as sr • as leachate has just passed
through the top liner and impinges intt ihe leak detection system.
2.6.1.3.2 T'me of Initial Leakage
The "time of Initial leakage" is the time when leakage first
occurs. A "time of Initial leakage" can be considered for each
specific cause of leakage. For example, the time of initial leakage
related to a given crack through the top liner may be one year after
the beginning of the operation of the hazardous waste land disposal
unit if this Is the time it took the crack to develop, plus the time
for leakage to go through the crack.
2.6.1.3.3 Initial Detection Time
The "initial detection time" is the time required to detect
leakage after it first occurs (I.e., the time difference between the
time of initial leakage and the time when this leakage appears at the
sump)
2.6,1.3.4 Leak Detection Time
The "steady-state leak detection time" (simply called the "leak
detection time") is the time between when a drop of leakage enters the
leak detection system (i.e., the time when this drop has just passed
through the top liner) and the time it appears in the collector pipes
or sump in a steady-state flow. This concept does not apply to the
first drop of leakage; it applies to any drop of leakage after the
leakage and the flow in the leak detection system have reached a
steady state.
2.6.1.3.5 Detection Sensitivity
Detection sensitivity refers to the smallest top liner leakage
rate which can be detected.
2.6-4
-------
2.6.1.3.6 Action Leakage Rate (ALR)
let ion Leakage Rate (ALR) refers to the ra1"1 of leakage frcm the
top liner into the LDCRS that triggers interaction between the owner
or operator and the EPA Regional Administrator to determine the
appropriate response action for leakage.
2.6.1.3.7 Rapid and Extremely Large Leakage
Rapid and Extremely Large Leakage (RLL) Is defined as the maximum
design leakage rate that the leak detection system (LDCRS) can remove
without exceeding a maximum fluid head (pressure) on the bottom
(outside the sump) equivalent to 0.3 m (1 ft) of water.
2.6.1.4 Organ1zat1on_of_th1s Section
Two approaches have been selected to analyze the functioning of
leak detection systems: analytical and numerical.
Section 2.6.2 is devoted to an analytical study conducted by
GeoServices Inc., Boynton Beach, FL and Section 2.6.3 is devoted to a
study using a finite element numerical model conducted by Radian
Corporation, Austin, TX.
2.6.2 Two-Dimensional Analytical Study
2.6.2.1 Introduction
2.6.2.1.1 Purpose of this Section
The purpose of this section is to present an analysis of the
functioning of leak detection systems which provides numerical
information on key parameters involved in the leak detection rule such
as:
• detection sensitivity;
• detection time;
2.6-5
-------
• action leakage rale; and
• rapid and extremely large leak.
This study was performed by GeoServices Inc. Consulting Engineers
of Boynton Beach, Florida, and 1s described 1n [Glroud et a!., 1987b].
2.6.2.1.2 Approach
The approach chosen was to conduct an analytical study. Such
studies can provide results that numerical methods often do not
provide because of Inherent limitations. Also, because of the
complexity of numerical methods, more time and effort are usually
spent in operating the method than in studying the physical problem,
discussing the validity of the results, and drawing practical
conclusions.
2.6.2.1.3 Organization of this Section
Section 2.6.2 is organized as follows:
• Section 2.6.2.2 presents the assumptions made for the
analytical study.
• Section 2.6.2.3 presents the analyses related to steady flow,
including an evaluation of the steady-state detection time
("detection time").
• Section 2.6.2.4 presents the analysis related to flow
initiation, including an evaluation of the initial detection
time.
• Section 2.6.2.5 presents conclusions pertinent to important
aspects of the proposed Liner/Leak Detection Rule such as:
detection sensitivity, detection time, action leakage rate, and
rapid and extremely large leakage.
2.6-6
-------
2.6.2.2 Assumptions
2.6.2.2.1 Assumptions Related to the Leak Detection System
- Leak Detection System Geometry
The leak defection system area considered has the following
geometry (Figure 2.6-1):
• length L along the slope;
• width B across the slope (i.e., horizontal);
• depth D; and
• slope angle p.
- Leak Detection System Material
The pervious material used in the leak detection system has the
following characteristics:
• porosity n; and
• saturated hydraulic conductivity, kj.
2.6.2.2.2 Assumptions Related to the Flow
- Flow Conditions
The flow conditions are as follows:
• the liner underlying the pervious material used in the leak
detection system is perfectly impermeable (i.e., its hydraulic
conductivity is zero and it has no holes, cracks, etc.);
2.6-7
-------
• the only liquids considered in the leak detection system are
those resulting from leakage through the top liner (i.e.,
liquids such as water entrapped during construction and ground
water migrating through the bottom liner are not considered
here - see Section 2.10);
• the only leakage that impinges the considered area of the leak
detection system Is a uniform leak that impinges the leak
detection system along its higher edge; and
• leakage 1s collected uniformly along the lower edge of the leak
detection system.
These assumptions are those of a bidimensional flow. Therefore, B
can be regarded as a unit width.
- Leakage Rate
The leakage rate, 0, that impinges the leak detection system at
its higher edge is expressed in units of volume per unit of time: 0
(m'/s, liters/day, gallons/day). This leakage rate can also be
expressed as a volume per unit of area and unit of time, q (m/s,
m'/m2/s, Uters/lOOOm'/day (Ltd), gallon/acre/day (gpad)), using the
following relationship:
q = Q/(LB) (Equation 2.6-1)
where: q = leakage rate per unit area (often called leakage rate); Q
~ leakage rate; L = length of considered area in the direction of the
flow; and B = width of the considered area. Recommended 51 units are-
q (m/s), Q (m'/s), L (m), and B (m).
- Flow Mechanism
The flow mechanism is assumed to be as follows:
• When leakage first occurs (I.e., when a first drop of liquid
has just passed through the top liner), liquid is first held by
2.6-8
-------
capillarity in the leak detection system. The volume of liquid
which can be held by capillarity depends on the opening size of
the material used in the leak detection system and the moisture
content of the leak detection system material before leakage
occurs.
• Some time after the beginning of leakage, enough liquid has
passed through the top liner to provide ali the volume which
can be held by capillarity. This point is the beginning of
steady flow of liquid In the leak detection system.
These two mechanisms are discussed hereafter. It is more
convenient to discuss steady state first (Section 2.6.2.3), and, then,
to discuss the Initial period when liquid 1s held by capillarity
(Section 2.6.2.4).
2.6.2.3 Steady-State_Flpw
2.6.2.3.1 Introduction
- Scope
In this section, it is assumed that the flow in the leak detection
system is 1n a steady-state condition. This becomes possible when a
portion of the leak detection system which is continuous between the
location of the leak and the sump has become saturated as a result of
progressive accumulation of water held by capillarity.
- Organization of this Section
First, Section 2.6.2.3.2 presents an evaluation of the "steady-
state leak detection time" (also called "leak detection time" or
"detection time"). Then, Section 2.6.2.3.3 discusses the steady-state
flow capacity of the leak detection system. The flow capacity will be
used in Section 2.10 to determine the rapid and extremely large leak
(RLL) that would generate excessive head on the bottom liner.
2.6-9
-------
2.6.2.3.2 Steady-Slate Leak Detection Time
- Dai cy' 5 Equation
Steady-state flow is governed by Darcy's equation. According to
Darcy's equation, the apparent velocity of flow in a saturated
pervious medium, such as the leak detection system material, is given
by:
v = Qd/A = krf i (Equation 2.6-2)
where: Qd - flow rate in the leak detection system; A - cross-section
area of the leak detection system perpendlcu'ar to the flow; k
-------
- Time for Detecting Leakage
The "steady-state leak detection time", supply called "leak
detection time", 1s the time leakage travels from the location •r the
leak to a collector pipe or sump in a steady-state condition. In the
model discussed above, this time Is the time for leakage to travel
along the slope, from the higher edge to the lower edge of the leak
detection system. Using Equation 2.6-5, it appears that this time Is
given by:
td = nL/(kd sin(3) (Equation 2.6-6)
where td - leak detection time; n = porosity of the leak detection
system material; L - length of the flow path (I.e., distance between
the leak in the top liner and the sump); kd = hydraulic conductivity
of the leak detection system material; and 0 = slope of the leak
detection system. Recommended SI units are: t^ (s), L (m), and kd
(m/s); n and p are dimensionless.
Equation 2.6-6 becomes:
td = n L/(86,400 kd sin 0)
- 1.16 x 1CT5 n L/(kd si.i p) (Equation 2.6-7)
with: td (days), L (m), and kd (m/s)
td = n L/(2835 kd sin p)
= 3.5 x 10"" n L/(kd sin P) (Eouation 2.6-8)
with: td (days), L (ft), and kd (cm/s)
For a porosity n = 0.3 of the leak detection system and a 2%
slope, the above equations become:
td = 1.74 x 10"" L/kd (Equation 2.6-9)
with td (days), L (m), and kd (m/s)
td •= 5.29 x 10"' l/kj (Equation 2.6-10;
with td (days), L (ft), and kd (cm/s)
2.6-11
-------
- Chart and Table for Leak Detection Time
Using Equation 2.6-9, a chart giving leak detection time has been
established and fs presented in ";gure 2.6-2. A logarithmic scale was
selected, although it is not iiecessary, to match with the charts
related to flow initiation (presented in tne next section) where a
logarithmic scale is appropriate.
Table 2.6-1 has been established from Equation 2.6-9 or Figure
2.6-2. This table gives leak detection times as a function of: (1)
the distance between the leak and the collector pipe or sump; and (11)
the hydraulic conductivity of the leak detection system material.
Table 2.6-1 and the chart presented in Figure 2.6-2 show that leak
detection time (i.e., steady-state leak detection times) on the order
of one day or less can be obtained using a leak detection system
material hydraulic conductivity of 10~J m/s (? cm/s) or more.
2.6.2.3.3 Leak Detection System Capacity
- Leak Detection System Hydraulic Transmissivity
According to Darcy's equation (Equation 2.6-2), the flow rate in
the leak detection system can be expressed as follows:
Qd/B = kd 1 D = kd D sin (3 (Equation 2.6-11)
where: Qd - flow rate in the leak detection system; B = width of the
considered section of the leak detection system perpendicular to the
flow; Qd/B = flow rate per unit width; k^ = leak detection system
material hydraulic conductivity; i = hydraulic gradient; D = leak
detection system thickness; and B = slope of the leak detection
system. Recommended SI units are: Q (m'/s), B (m), Q/B (m'/s), k,j
(m/s), and D (m); 0 and 1 are dimensionless.
Equation 2.6-11 can be written as follows:
8d i = BCJ sin B (Equation 2.6-12)
2.6-12
-------
where Oj is the hydraulic transmissivity of the leak detection system,
defined by:
6(j = kjj D (Equation 2.6-13)
The recommended SI unit for 64 is m'/s.
According to Equation 2.6-12, the capacity of the leak detection
system is governed by Its hydraulic transmissivlty. Equation 2.6-13
shows that, for a given leak detection system material hydraulic
conductivity, kj, the capacity of the leak detection system is
governed by Its thickness D.
- Relationship Between Leakage Rate and Transmissivity
The flow rate, 0^, in the leak detection system is zero at the
higher edge of the considered model (Figure 2.6-1) and is equal to the
leakage rate, Q, at the lower edge of the considered model. In other
words, the maximum value of the flow rate Q^ in the leak detection
system is:
Qd max = Q (Equation 2.6-14)
18 Etjij^lfill IM\ iljlJiltibn1 i;§-14 caf] ke fwltlifj ki\
max
L B (Equation 2.6-15)
Combining Equation 2.6-11, 2.6-13 and 2.6-15, the following
relationships can be established between the leakage rate through the
top liner and the characteristics of the leak detection system:
q L = k
-------
system material; D = thickness of the leak detection -tern; 0 = slope
of the leak detection system; and 8^ = hydraulic tranj...issivity of the
leak detection system. Recoumended SI units are: q (m/s), L "<), kj
(m/s), D (m), and 6d (m'/s); fl is dimensionless.
- Required Capacity of the Leak Detection System
If the transmissivlty (I.e., thickness x hudraullc conductivity)
of tne leak detection system 1s not sufficient to convey the leakage
through the top liner, hydraulic head builds up 1n the leak detection
system. Leakage Into the ground Increases as a result of Increased
head on the bottom i ner, which should absolutely be avoided.
Therefore, the hydraulic transmissivity and the thickness of the leak
detection system must have the following minimum values derived from
Equation 2.6-16 and 2.6-17:
8d I q L/sin p (Equation 2.6-18)
D I q L/kd sin p (Equation 2.6-19)
(Notation defined after Equation 2.6-17.)
Equations 2.6-18 and 2.6-19 can be used with any compatible
systems of units, including the recommended SI units: 8d (m2/s), q
(m/s), L (m), D (m), and kd (m/s).
Equation ?.6-18 can be written as follows with various units:
6d I 1.16 x 10"" q L/sin p (Equation 2.6-20)
with: 9 (m'/s), q (Ltd) and L (m)
8d I 3.3 x 10~12 q L/sin 0 (Equation 2.6-21)
with: 6d (m'/s), q (gpad), L (ft)
Equation 2.6-19 can be written as follows with various systems of
units:
2.6-14
-------
0 2 1.16 x 10"" q L/(kd sin f) (Equation 2.6-22)
with: D (m), q (Ltd), L (mj, kd (m/s)
D 2 1.16 x 10~* q './(kd sin fl) (Equation 2.6-23)
with: D (ft), q (gpad), L (ft), kd (cm/s)
In the special case of a 2°/. slope (sin 3 = 8.02), Equations 2.6-18
and 19 become:
6d » 5.8 x 10"'° q L (Equation 2.6-24)
with: 9 (m'/s), q (Ltd), L fro)
9d « 1.65 x 10" " q L (Equation 2.6-25)
with: 6d (m'/s), q (gpad), L (ft)
D = 5.8 x 10"10 q L/kd (Equation 2.6-26)
with: D (m), q (Ltd), L (m), kd (m/s)
D = 5.4 x 10~* q L/kd (Equation 2.6-27)
with: D (ft), q (gpad), L (ft), kd (cm/s)
- Tables
Table 2.6-2 was established using Equations 2.6-24 and 2.6-25.
This table gives the required hydraulic transmissivity for various
leakage rates. In order to select the minimum hydraulic
transmissivity a leak detection system should have, it is recofunended
that a leakage rate of 10,000 Ltd (gpad) be considered. This value of
leakage rate provides a factor of safety on the order of 10 vis a vis
the "rapid and extremely large leak" (RLL) which can be on the order
of 1,000 Ltd (gpad). Factors of safety of 10 are typical in hydraulic
designs for critical projects.
Considering a leakage rate of 10,000 Ltd (gpad}, Table 2.6-2 shows
that a minimum hydraulic transmissivity, 8d, on the order of 5 x 10"4
m'/s should be considered for * leak detection system. According to
Equation 2.6-13, such a hydraulic transmissivity can be achieved by
many combinations of hydraulic conductivity, kd, and thickness, 0, of
the leak detection system such as:
2.6-15
-------
• kj = 10~' m/s (10 cm/'s) and . = 5 mm; and
• kd = 10"* m/s (1 cm/s) and D = 50 mm.
Therefore, from a practical standpoint, two types of leak
detection systems siiould be considered:
• synthetic leak detection systems with k^ > .0"' m/s (10 cm/s)
and D > 5 mm (0.2 1n.) ; and
• granular leak detection systems with k,j > 10"* m/s (1 cm/s) and
D > 300 mm (12 in.).
In the latter case, the technical requirement of 50 mm has been
replaced by 300 mm for two reasons: (i) placement of a soil layer
much less than 300 mm (1 ft) is not practical; and, (ii) the
construction equipment used to spread granular materials should not
operate on top of tha FML component of the bottom liner unless the FML
is covered by a protective layer approximately 300 mm (1 ft) thick.
Table 2.6-3 has been established using Equations 2.6-26 and 2.6-
27, for a leakage rate of 10,000 Ltd (gpad). This table gives
required leak detection system thicknesses and illustrates that with a
slope of 27,, a thickness on the order of 5 mm (0.2 in.) is required
with a hydraulic conductivity of 0.1 m/s (10 cm/i), while a thickness
on the order of 50 mm (2 in.) is required with a hydraulic
conductivity of 0.01 m/s (1 cm/s).
2.6.2.4
2.6.2.4.1 Introduction
- Scope
In this section, it is assumed that leakage has just occurred
through the top liner and has just entered the leak detection system.
The purpose of this section is to evaluate the amount of liquid held
2.6-16
-------
by capillarity in the leak detection system before the flow reaches a
steady slate.
- Organization
\ '.{ First, Section 2.6.2.4.2 presents an evaluation of the amount of
liquid held by capillarity 1n the leak detection system, and the rate
at which this amount increases, which govern:: the initial leak
detection time. Then, Section 2.6.2.4.3 discusses the limits of this
mechanism In case of large leakage rates (at whirh point, flow is
governed by Darcy's equation).
2.6.2.4.2 Retention by Capillarity
- Assumption
In this study we assume that the zone where liquid Is held by
capillarity' covers the entire area considered (i.e., L x B) over a
height equal to: (1) the capillary rise, h, for the considered
pervio medium, if h is smaller than the thickness D of the leak
detection system; or (ii) the entire thickness D, if the capillary
rise, h, is larger than D.
- Capi1lary Rise
The capillary rise in a tube is given by the classic Jurin's
equation:
hc = 4T/(p g dt) (Equation 2.6-28)
where: hc = capillary rise; T = capillary tension (which depends on
the liquid and, to a lesser extent, on the ma^ri,ii in rnnt-;»rM • r. -
' 'I ' •
-------
For water:
T = 7.64 x 10~2 N/m
p = 1000 kg/m'
In order to use Junn's equation for granular drainage materials,
we assume that the so-! pore diameter governing capillary Mse is
equal to the d,0 par 'le size of the soil (i.e., the soil particle
size which 1s larger than 10% by weight of the soil particles). In
addition, particle size can be linked to hydraulic conductivity using
Hazen's equation:
kd = C (d,0)2 (Equation 2.6-29)
where: C = Hazen's coefficient, equal to 104 m~'s~' (102 cm"'s~').
A relationship between capillary rise and hydraulic conductivity
can therefore be obtained by combining Equations 2.6-28 and 2.6-29,
and assuming dt = d,0, as indicated above:
hc = 4T/(p g Ad/C) (Equation 2.6-30)
A similar equation related to non-granular drainage materials
could be obtained by combining Jurin's equation with Poiseuille's
equation (instead of Hazen's equation). (Kozeny's approach could be
used to establish a relationship between the two equations.) In the
mean time Equation 2.6-31 will be used for all types of drainage
materials.
Replacing T and p by their values for water (T = 7.64 x 10~2 N/m,
p = 1000 kg/m') and g and C by their values (g » 9.81 m/s2,
C » 10* m~' s~'), the following water capillary rise is obtained:
hc = 3.1 x lO'VArf (Equation 2.6-31)
(Note: this equation is valid only for kd in m/s and gives hc in m)
2.6-18
-------
Typical values of cap ;iary rise as a function of hydraulic
conductivity are given in Table 2.6-4.
- Volume of Liquid Held by Capillarity
While liquid is being held by capillarity, it does not flow
towards the sump, which delays leakage detection. Two cases should be
considered regarding the volume of leakage which can be held by
capillarity in the leak detection system (in addition to the volume of
liquid already held by capillarity before the considered leakage
starts):
• if hc < D:
V - n LBh (Sr - Sro) (Equation 2.6-32)
• if hc > D:
V = n LL, (Sr - Sro) (Equation 2.6-33)
where: V = volume of liquid held by capillarity; hc = capillary rise;
D = thickness of leak detection system; L = length of leak detection
system along the slope; B = width of leak detection system across the
slope; n = porosity of the leak detection system material; Sr = degree
of saturation of the capillary zone at the beginning of leak
detection; and Sro = initial dsgree of saturation of the capillary
zone (i.e., degree of saturation at the time of initial leakage).
Recommended SI units are: V (m3), hc (m), D (m), L (m), B (m); n, Sr,
and Sro are dimensionless and less than one.
The initial degree of saturation, Sro, depends on the water
content of the leak detection system material at the beginning of the
considered leakage. This water content may result from: (i) water
entrapped in the leak detection system during construction; (ii) water
used to compact the low-permeability soil, if any, included in the top
liner; (iii) ground water seeping into the leak detection system; and
(iv) previous leakage.
2.6-19
-------
- Detection Delay Caused by Capillarity
Although the liquid is being held by capillarity, it progresses
toward the sump as the capillary zone grows. Therefore, if we assume
that there is no free flow (i.e., no Darcy's flow), the initial
detection time Is the time necessary for the capillary zone to grow
from the source of leakage (elevated edge of the considered drainage
area as shown in Figure 2.6-1) to the sump (located at the lower edge
of the considered area). This time can be obtained by dividing the
volume of liquid which can be stored in the capillary zone (Equation
2.6-32 or 2.6-33) by the leakage rate (Equation 2.6-1):
t1 = V/(qLB) (Equation 2.6-34)
which leads to:
^ = n (Sr - Sro) hc/q (if h < D) (Equation 2.6-35)
ti = n (Sr - Sro) D/q (if h > D) (Equation 2.6-36)
where: ti = initial detection time; hc = capillary rise; D =
thickness of leak detection system; n = porosity of material used in
leak detection system; Sr = degree of saturation of the capillary zone
at the beginning of leak detection; Sro = initial degree of saturation
of the pu.entia' capillary zone; and q = leakage rate. Recommended SI
units are: t^ (s), hc (m) , D (m) , q (m/s); n, Sr, and Sro are
dimensionless and less than one.
By combining Equations 2.6-30 and 2.6-35, the following equation
vx is obtained:
N
t, = 4n (Sr - Sro) T/(p g q Ad/C) (Equation 2.6-37)
Using the numerical values which were used for Equation 2.6-31,
.^ and using n = 0.3 and Sr = 1, the following equation is obtained:
9.3 x 10"J (1 - Sro)/(q Aj) (Equation 2.6-38)
2.6-20
-------
(Note: this equation is valid only with the following units: tj (s),
q (m/s), and k,j (m/s).)
This equation becomes:
tj = 928 (1 - Sro)/(q J\Td) (Equation 2.6-39)
if the following units are used; tj, in days; q, in 1 ifers/iOOOm2/day
(Ltd); and k^, 1n m/s.
Equation 2.6-38 becomes:
tj - 10,200 (1 - Sro)/(q /i^) (Equation 2.6-40)
if the following units are used: tj, in days; q, in gallons/acre/day
(gpad); and k^ In cm/s.
- Establishment of a Chart
Equation 2.6-38 can be represented by a family of hyperbolas
giving the initial detection time tj as a function of the leakage rate
q, each hyperbola being related to a given value of the hydraulic
conductivity, kj, of the leak detection system. A practical way to
represent the relationship between detection time and leakage rate is
to use a log-log scale (Figure 2.6-3): as a result, the family of
hyperbolas becomes a family of straight lines at 45°. (Note that
Figure 2.6-3 has been established for an initial degree of saturation,
Sro, equal to zero; in other words, Figure 2.6-3 is related to the
case where the leak detection system is dry when leakage begins.)
It is important to notice that the family of curves is bounded by
the curve which 1s related to the hydraulic conductivity that gives a
capillary rise equal to the thickness of the leak detection system.
For example, if the thickness of the leak detection system is D = 0.3
m (1 ft), all curves in Figure 2.6-3 for kj < 10~" m/s (10~2 cm/s)
merge with the curve for kj - 10~4 m/s because the capillary rise for
kg- = lO"' m/s 1s 0.3 m according to Table 2.6-4. In other words, any
calculation done with k,j < 10~4 m/s (10~2 cm/s) gives the same value
2,6-21
-------
of the initial detection time than 3 calculation conducted with kj *
10~" m/s (10"' cm/s), If the thickness of the leak detection system Is
D = 0.3 m (1 ft).
Portions of the curves shown in Figure 2.6-3 are not valid because
they are incompatible with Carey's flow for the reasons explained in
Section 2.6.2.4.2. It is therefore necessary to establish a more
complete chart combining the effects of capillarity and Darcy's flow.
2.6.2.4.2 Boundaries of the Flow Mechanism
- Description of the Phenomenon
The front of the volume which is being saturated by capillarity
cannot move faster than laminar flow 1n a saturated medium, which 1s
governed by Darcy's equation (Equation 2.6-1). Drainage Into pipes or
sumps will not occur until : "ie saturated front reaches the pipe or
sump. Therefore initial detection times, t^, given by Equation 2.6-37
and by the chart presented in Figure 2.6-3 are valid only 1f they are
larger than the leak detection times, t^, related to Darcy's flow and
given by Equation 2.6-6. This provides a boundary to the flow
mechanism, which is discussed below.
- Chart Combining Capillarity and Darby's Flow
A refinement of the chart given 1n Figure 2.6-3 consists of
including thp times tj for Darcy's flow. According to Equation
2.6-6, such times are Independent of the leakage rate q. Therefore,
in the t-q chart, the Darcy's times (I.e., the detection times), t^,
will be represented by horizontal lines. These horizontal lines meet
the 45° lines related to the capillary times (i.e., the initial
detection times), tj, at points defined by:
tj » t(j (Equation 2.6-41)
The locus of these points is obtained by eliminating the hydraulic
conductivity k
-------
td = nL/(k sin|3)
tc = 4n (Sr - Sro) T/(p g q A7c)
This gives:
td « tj - 16 n C sin? L~' [(Sr - Sro) T/(p g q)]' (Equation 2.6-42)
Using the following values:
n = 0.3 (porosfty of material used in leak detection system),
C = 104 nf' s"1 (Hazen's coefficient),
P = arc tan 0.02 (bottom slope),
Sr» 1 (degree of saturation in the capillary zone of the
material used in the leak detection system at the beginning
of leak detection),
Sro= 0 (initial degree of saturation in the potential capillary
zone of the material used in the leak detection system),
T = 7.64 x 10~2 N/m (water capillary tension),
p = 1000 kg/m' (density of water), and
g = 9.81 m/s2 (acceleration of gravity),
the following equations are obtained:
td - t1 - 5.8 x 10~V(Lq') (Equation 2.6-43)
(with the following units: t (s), L (m), q (m/s))
2.6-23
-------
td = tj =. 5 x 10V(Lq2) (Equation 2.6-44)
(with the following units: t (days), L (m), q (1iters/lOOOmVday))
or:
td - t1 » 2 x 10'°/(Lq2) (Equation 2.6-45)
(with the following units: t (days), L (ft), q (gallons/acre/day))
where: t^ - leak detection time; tj =• initial detection time; L -
length of flow path (distance between leak and pipe or sump); and q =
leakage rate per unit area.
The above equation is the equation of dashed line AB in Figure
2.6-4 (for L = 60 m = 200 ft), and dashed lines in Figure 2.6-5. The
dashed lines in these figures are the loci of points where Darcy's
curves meet capillarity curves, i.e., where initial detection time
equals leak detection time. This happens for large leakage rates:
steady-state flow is then established Immediately, i.e., without a
preliminary phase with capillary retention.
- Required Hydraulic Transmissivity
A second boundary of the flow mechanism results from the fact that
Darcy's flow rate 1s limited by the hydraulic transmissivity of the
leak detection system. This is discussed in Section 2.6.2.3.3 where
it is shown that this boundary is reached when the leakage rate
reaches the following value derived from Equation 2.6-16:
q = k D sinp/L (Equation 2.6-46)
Combining Equation 2.6-46 with Equation 2.6-6 (which gives the
Darcy's detection time t^) and eliminating k^ lead to:
2.6-24
-------
530/SW
87-015
Part II
Background Document on
Proposed Liner and Leak
Detection Rule
NUS Corp., Rockvillo, MD
P387-191383
Prepared for
Environmental Protection Agency
Washington, DC
May 87
U, S, ENVIRONMENTAL PROItCTlON
4GENCY
1445 RQvSvS AVENUE
r I
-------
EPA
530/SW
87-015
Part II
td = n D/q (Equation 2.6-47}
where: n = porosity of the leik detection system material; D =
thickness of the leak detection system; and q = leakage rate per unit
area.
Equation 2.6-47 is identical to Equation 2.6-36 which is -elated
to capillarity ove' the entire thickness of the leak detection sy ~i
and to a degree of saturation equal to one (Sr - 1). This is .-t
surprising: In this case, capillarity over the entire thickness of
the leak detection system and Darcy's flow are identical.
This is why there is no possible flow beyond line CO of Figure
2.6-4. In other words, if the leakage rate through the top liner
tends to be larger than the limit given by CD in Figure 2.6-4, the
leak detection system becomes saturated. The pressure buildup in the
leak detection system automatically keeps the leakage rate through the
top liner at a value consistent with curve CD in Figure 2.6-4. As a
result of pressure buildup in the leak detection system, leakage rate
through the bottom liner and into the ground increases.
2.6.2.5 Conclusions
2.6.2.5.1 Discussion of the Results
Although the bidimensional a-alytical study could be further
refined, it has provided useful remits which are summarized in Tables
2.6-1 through 2.6-5 and in Figures 2.6-2 through 2,6-5. These results
are di":ussed below.
- Detection Time
Table 2.6-1 shows that a detection time smaller than one day can
be achieved with a 27. slope and a minimum hydraulic conductivity of
the leak detection system of 10~l m/s (1 cm/s). It is reiterated that
detection times are determined assuming steady-state conditions.
Initial detection times, as discusses below, are much larger.
2.6-25
-------
- Initial Detection Time
The results presented in Table 2.6-5 (established from Figure 2.6-
4) show that Initial detection times depend on leakage rate and are
high. For example, 1f the leak detection system has a hydraulic
conductivity of 10~2 m/s (1 cm/s) this study has shown that for
bidimensional flow it will take 10 days for the initial detection of
large uniform top linei leakage into an initially dry granular leak
detection layer (leakage rate on the order of 1000 Ltd (1000 gpad))
and 1000 days for a leakage rate on the order of the proposed Action
Leakage Rate (ALR) (10 Ltd (10 gpad)).
As shown in Table 2.6-6, initial leak detection times are four
times shorter than the times given In Table 2.6-5 if the potential
capillary zone of the leak detection system has an initial degree of
saturation of 75% (Sro in Equation 2.6-39 and 40). Such an initial
degree of saturation may result from: (i) warer entrapped in the leak
detection system during construction; (11) water used to compact the
low-permeability soil, if any, included in the top liner; (iii) ground
water seeping into the leak detection system; and (iv) previous
leakage.
- Detection Sensitivity
One of the assumptions of the analyses presented in Section 2.6.2
was that the bottom liner is perfectly impermeable. In steady-state
conditions, this assumption leads to a zero detection sensitivity. To
obtain a detection sensitivity different from zero, leakage through
the bottom liner should be considered, which is done in the technical
background document to EPA's April 17, 1987 Notice of Availability of
Data on "Bottom Liner Performance in Double-Lined Landfills and
Surface Impoundments". This document shows that the detection
sensitivity is smaller than 1 Ltd (gpad) if a composite bottom liner
is used, and on the order of 100 Ltd (gpad) if a compacted soil bottom
liner is used.
2.6-26
-------
- Influence of Capillarity on Initial Detection Times
It appears fiom the c arts (Figures 2.6-3 through 2.6-5) that
capillarity plays a very important role and significantly delays
initial detection. It is possible that the fact that the study is
bidimenslonal tends to exaggerate the influence of capillarity (since
in a three-dimensional study, concentrated leakage could be modeled,
which is more realistic than uniform leakage). It is therefore
possible that a three-dimensional study would show that the flow is
governed more by Darcy's equation than by capillarity and therefore is
faster. In other words, it Is expected that initial detection times
would be shorter In a three-dimensional study, especially for small
leakage rates.
The influence of capillarity Is further exaggerated in Table 2.6-5
and In the charts (Figures 2.6-3 through 2.6-5) because the leak
detection system Is assumed to be dry at the beginning of flow.
Shorter Initial detection times are obtained if a leak develops in a
leak detection system which already contains liquid held by
capillarity from various sources such as: previous leaks,
construction water, or water expelled from the compacted soil
component of a composite top liner. For example, Initial leak
detection times are divided by 4 if the initfal degree of saturation
of potential capillary zone In the leak detection system is 75%
instead of 0%, as shown by Equations 2.6-39 and 40, and illustrated in
Tables 2.6-5 and 2.6-6.
2.6.2.5.2 Extension of the Study
The study presented In Section 2.6.2 has provided useful results.
Additional refinements and analyses to the study cculd be conducted to
gain further insight into the performance of leak detection systems.
These refinements and additional analyses are outlined below.
- Leakage through Bottom Liner
In the analyses presented in Section 2.6.2, the bottom liner is
assumed to be absolutely impermeable. Similar analyses could be
2.6-27
-------
conducted with a range of permeabilities for the bottom liner to
Include cases such as: (i) composite bottom liner with a leaking FML;
and (ii) low-permeability soil bottom liner (i.e., without a FML).
- Three-dimensional Study
The study presented in Section 2.6.2 is bidimensional. A similar
study could be conducted considering concentrated leaks and using flow
considerations (as governed by leak detection system material
conductivity) to evaluate the width of the wetted area.
It is expected that such a three-dimensional study will indicate a
lesser influence of capillarity than the bidimensional study presented
in Section 2.6.2. If this is the case, smaller initial detection
times would be obtained.
A major advantage of a three-dimensional study is the ability to
consider a variety of leakage scenarios, including "worst cases" of
concentrated leaks, while the bidimensional study presented in Section
2.6.2 considers a leakage equivalent to an average leakage rate per
unit area, which leads to some semi-paradoxical results, such as
initial detection times independent from distance between leak and
sump.
Finally, experience gained in conducting a three-dimensional study
including various leakage scenarios will make it possible to develop
guidance for designers who have to face the difficult problem of
selecting leakage scenarios.
- Parametric Study
Using the bidimensional and the three-dimensional studies, a
parametric study should be conducted to produce results such as those
given in Tables 2.6-1, 2.6-2, 2.6-5 and 2.6-6 and showing the
inf'tence of the following parameters:
• hydraulic conductivity of the leak detection system material;
2.6-28
-------
• degree of saturation of the leak detection system material
(resulting from previous leakage, construction water, water
expelled from adjacent clay by consolidation);
• thickness and hydraulic transmissivity of the leak detection
system;
• slope of the leak detection system;
• distance between collector pipes in a leak detection system;
• permeability of the bottom liner (which affects In particular
the detection sensitivity); and
• leakage rate.
2.6.3 Two-Dimensional Numerical Study
2.6.3.1 Introduction
2.6.3.1.1 Purpose of this Section
The purpose of this section is similar to the purpose of Section
2.6.2: it 1s to present an analysis of the functioning of leak
detection systems which provides numerical information on key
parameters involved in the proposed Liner/Leak Detection Rule. These
parameters are:
• detection sensitivity;
• detection time; and
• Action Leakage Rate.
This study was performed by The Radian Corporation, Austin, Texa:,
and 1s described 1n [Radian, 1987a, b, c].
2.6-29
-------
2.6.3.1.2 Approach
The approach chosen was to c .''duct a numerical study using a
finite element computer program. Numerical methods have capabilities
that analytical methods do not have. In particular, numerical studies
can accept Input 1n either analytical or numerical form, while, with
analytical studies, input should be made in analytical form.
Therefore, experimental data.(such as those related to capillary
retention) must .be approximated by equations to be used in an
analytical study, while they can be used without any approximation in
numerical studies.
2.6.3.1.3 Complementarity of Analytical and Numerical Studies
One intent of the numerical study was to evaluate the influence of
the bottom liner on the performance of the leak de- -tion system. As
a consequence, compacted soil liners as well as cc. posite liners are
considered for the bottom liner, and leakage through the bottom liner
is taken into account. The analytical stuJy was only intended to
evaluate the performance of the drainage layer which constitutes the
leak detection system and, consequently, leakage through the bottom
liner is neglected. On the other hand, the analytical study was used
to provide detailed information on steady-stata flow as well as
transient flow, while the numerical study concentrated mostly on
transient flow from the beginning of leakage to the establishment of
steady-state flow.
In the case of the analysis of the functioning of leak detection
systems, because of the newness and the difficulty of the subject, it
is appropriate that two studies conducted with two different
approaches and by two different teams are available: the analytical
study by GeoServlces and the numerical study by Radian. Results can
be compared and better conclusions can be drawn.
2.6.3.1.4 Organization of this Section
Section 2.6.3 Is organized as follows:
2.6-30
-------
• Section 2.6.3.2 presents the method used, i.e., the finite
element computer program used for the numerical study, and the
assumptions.
• Section 2.6.3.3 presents the results related to the initial
detection time using the ' -dimensional numerical study, and
compares results of the e-rical study with results of t;,e
analytical study.
2.6.3.2 Method
2.6.3.2.1 Description of the Finite Element Program
The finite element program used in the Radian study is UNSAT2D, a
two-dimensional finite element computer program prepared by S.S.
Papadopulos & Associates, Inc. to simulate soil moisture movement
within waste disposal units including landfills, surface impoundments,
and waste piles. Input parameters to the program include: (i) water
movement across model boundaries and/or hydraulic head on model
boundaries; (11) land disposal unit gecn.itry; (iii) material
properties; and (iv) initial moisture conditions in the land disposal
unit and surrounding soils. The program simulates the transient-state
distribution of hydraulic head and soil moisture within the land
disposal unit for each defined time step.
The program simulates a two-dimensional section through a land
disposal unit. In the formulation of the program it has been assumed
that adjacent parallel sections are identical in their physical and
hydrological characteristics. As a result, the program can only model
linear tears 1n the f"ML (holes cannot be modeled).
The program models FMLs and geotextiles as one-dimensional
(linear) elements which have zero moisture storage. Leakage across
the element 1s proportional to the head difference across the element.
Soil and waste are modeled by two-dimensional, triangular elements
which have moisture storage capacity.
2.6-31
-------
2.6.3.2.2 Assumptions
- Leak Detection System Geometry
The considered leak detection system has the geometry shown in
Figure 2.6-6. The key features are:
• 4H/1V (i.e., 257.) side slopes and 27. bottom slopes;
• 10 m (30 ft) long side slopes in horizontal projection and 18 m
(60 ft) distance between collector pipes on the bottom; and
• 0.3 m (1 ft) thick leak detection layer.
- Leak Detection System Material
The pervious material used in the leak detection system has the
iollowing characteristics:
• porosity n - 0.3;
• a saturated hydraulic conductivity, kd, of 10~6 m/s (10~J cm/s)
in most cases, and 10~4 m/s (10~2 cm/s) in some cases; and
• when unsaturated, the leak detection system material exhibits
capillary suction given by suction-degree of saturation curves
directly input into the computer program.
- Supply of Liquids into the Leak Detection bysum
At the beginning of each run of the computer program, the leak
detection system is assumed to be approximately 75% saturated as a
result of water entrapped during construction or water resulting from
previous leakage. Then, the only supply of liquid to the leak
detection system is leakage through the top liner.
2.6-32
-------
UNSAT2D holds the leakage rate thrcuc the top liner constant and
the hydraulic head on the bottom liner ,s allowed to vary. The
leakage rate through the top liner was controlled by varying th«
hydraulic head on the top liner and the properties of the top lin-r
materia,. Three types of top liner leaks were considered in the
UNSAT2D numerical simulations (Figure 2.6-6b):
• uniform leakage through the entire top liner (uniform leak);
• leakage through a 3 m (10 ft) wide portion of the top liner on
facility side slope (sidewall leak); and
• leakage through a 3 m (10 ft) wide portion of the top liner on
the lower side (i.e., the bottom) of the land disposal unit
(bottom leak).
A limitation of the program is that the smallest top liner
sidewall and bottom leak which could be analyzed is 3 m (10 ft) wide
In reality, FML top liner field defects are more likely to be small
tears or punctures, typically only a few millimeters (fraction of an
inch) in diameter. Thus, the UNSAT2D top liner leak probably better
represents leakage through a composite top liner than through a top
liner consisting of an FML alone.
- Bottom Liner
A thin, very low-permeability layer with zero liquid storage
capacity is placed over a compacted soil layer to simulate the FML
component of a composite bottom liner. In the UNSAT2D simulations, the
migration of liquid across this very thin layer is described by the
FML leakance, which is defined as follows:
(Equation 2.6-48)
where: Q = leakage rate; A = liner area; L = leakance; and h =
hydraulic head. Recommended 51 units are: Q (m'/s); A (m2); L (s'1);
sj
i and h = m.
2.6-33
-------
It appears from Elation 2.6-43 that the leaKance is identical to
the classical permittivity, >!'.
In a double-liner system, the hydraulic head on top of the
bottom liner is almost always very small. However, in the UNSAT2D
simulations the capillary suction acting on the bottom of the FML
component of the bottom liner is significant. In these simulations,
this capillary suction is equivalent to a hydraulic head of 3.4 m
(11.1 ft) acting on the FML from underneath. Such mechanism is
unlikely to be effective 1n reality, and, to counteract the effect of
the large hydraulic gradient set up in the numerical simulations by
the action of capillary suction pulling water through the FML, a very
low leakance value was selected by Radian. In UJJSAT2D simulations, a
leakance of 7 x 10~" s~' was selected to be used with a compacted
soil capillary suction equivalent to 3.4 m (11.1 ft) of negative head.
These values correspond almost exactly to a 1-mm (40-mil) thick FML
with an equivalent hydraulic conductivity kg = 1 x 10~14 m/s (1 x
1CT12 cm/s) subjected to a hydraulic head of 30 tun (0.1 ft). Radian
also carried out simulations with a FML leakance of 3 x 10"" s~'.
This leakance is 4^0 times larger than the one for an "intact" FML and
can be considered .0 approximately represent the FML component of a
bottom liner that has undergone "significant" deterioration. Finally,
Radian carried out several numerical simulations with an intermediate
leakance value, L = 3 x 10"'2 s~', which may represent a FML that has
undergone "some" deterioration.
- Flow Mechanism
The flow mechanism considered in the two-dimensional numerical
study is identical to the flow mechanism considered in the two-
dimensional analytical study:
• When leakage first occurs (i.e., when a first drop of liquid
has just passed through the top liner), liquid is first held t"
capillarity in the leak detection system. The volume of liquid
which can be held by capillarity depends on the opening size of
the material used in the leak detection system and the moisture
2.6-34
-------
• Some time after the beginning of leakage, enough liquid has
passed through the top liner to saturate a sufficient volume in
the vicinity of one of the collector pipes to initiate drain
flow. This point is the beginning of the detection of leakage
in the collector pipes and gives the value of the initial
detection time defined in Section 2.6.1.3.3.
2.6.3.3 Results of the Numerical Study
2.6.3.3.1 Summary of the Results
As indicated fn Section 2.6.3.1.3, low-permeability soil bottom
liners as well as composite bottom liners are r">nsidered in the
numerical study. Only results related to composite oottom liners are
considered here.
The numerical study shows that saturation generally occurs first
in the vicinity of the collector pipe located at the center of the
bottom of the land disposal unit (see Figure 2.6-6). Therefore,
leakage is detected first at this pipe at a time that is the initial
detection time considered here. However, some additional time is
required for steady-state flow to bo est~ •!isht.d. This time was also
determined 1n the numerical study, but is .,ot considered here.
As indicated in Section 2.6.3.2.2, three types of leaks are
considered. Table 2.6-7 shows that, according to the numerical study
there is no significant difference between the various types of leaks
regarding initial detection times.
As indicated in Section 2.6.S.2.2, three cases are considered
regarding the condition of FHL component of the bottom Uner: (i)
intact FML; (11) FML with significant deterioration; and (iii) an
intermediate case. Table 2.6-7 shows that the condition of the bottom
liner does not significantly affect the initial detection times for
the considered cases. However, if smaller leakage rates had been
considered, the effect of the condition of the bottom liner would have
been more marked.
2.6-35
-------
2.6.3.3.2 Comparison between the Analytical and Uimerlcal Study
One of the results of the two-dimensional analytical study is
Equation 2.6-40 which g es the leak detection time as a function of
the leakage rate through the top liner. This equation was used with
the following values of Us parameters:
• an initial degree of saturation of the leak detection system of
75%, which 1s approximately the Initial degree of saturation
used In the two-dimensional numerical study; and
• a leak detection system material hydraulic conductivity of 10"4
m/s (10~* cm/s) although the hydraulic conductivity used in the
numerical study was 10~* m/s (10~3 cm/s), because the
analytical study has shown that, for a given top liner leakage
rate, all hydraulic conductivities smaller than 10~" m/s (10~*
cm/s) lead to values of the leak detection time equal to the
value calculated for 10"4 m/s (10~2 cm/s), if the thickness of
the leak detection system 1s 0.3 m (1 ft) or less.
Leak detection times were calculated using Equation 2.6-40 for the
five values of the top liner leakage rate used in the two-dimensional
numerical study. (These values are listed in the first line of Table
2.6-7.) The leak detection times thus calculated are given in the
second line of Table 2.6-7.
Table 2.6-7 shows that there 1s a very good agreement between leak
detection times calculated using the numerical :r,ethcd on one hand, and
the analytical method on the other hand. This is remarkable since the
leak detection system geometries considered in the two studies were
significantly different, which appears by comparing Figures 2.6-1 and
2.6-6.
The good agreement between the results of the two studies confirms
the validity of the analytical study. This 1s particularly
Interesting since the analytical study has been used for a variety of
values of the hydraulic conductivity, and not only for kc = 10~* m/s
(10"' cm/s) like the numerical study. In fact, the recommended value
2.6-36
/I"
-------
for leak detection system hydraulic conductivity is 10"' in/s (0.1
cm/sx for granular leak detection systems and 10~* m/s (1 cm/s) for
synl ..ic leak detection systems, as discussed In Section 2.8.
It may be concluded that the conclusions of the analytical study,
presented In Section 2.6.2.5.2 and Tables 2.6-5 and 2.6-6, are
confirmed by the numerical study and, therefore, can be used in
subsequent sections. Further, the good agreement between the
analytical and numerical studies helps to validate the extensive
results of the numerical study presented in [Radian, 1987a, b, c].
2.6-37
-------
Table 2.5-1. Leak detection time as a function of the distance, L,
between the leak and the collector pipe or sump, and the
hydraulic conductivity of the leak detection system
material. This table has been established assuming
steady-state flow in the leak detection system and an
Impermeable bottom liner. The slope of the leak
detection system is 2%. The same results can be found in
Figure 2.6-2. [Result of the GeoServices analytical
study]
L
15 m
(50 ft)
30 m
(100 ft)
50 m
(165 ft)
60 m
(200 ft)
100 m
(330 ft)
Hydraulic conductivity of the
leak detection system material, kj
10"" m/s
(10~2 cm/s
26
52
87
104
174
10"1 m/s
(10~' cm/s)
3
5
9
11
18
10~' m/s
(1 cm/s)
0.3
0.5
0.9
1.1
2
10"' m/s
(10 cm/s)
0.03
0.05
0.01
' 0.1
0.2
Leak detection times in days
2.6-38
-------
Table 2.6-2. Required hydraulic transmissivity, 8,3, for the leak
detection system ((In m'/s) as a function of leakage
rate, q, and distance, L, between leak and collector
pipe or sump for a leak detection system slope of 2% and
an Impermsable bottom liner. [Results from the
GeoServices analytical study]
Top liner leakage rate per unit area, q
]Iters/lOOOm'/day (gallons/acre/day)
1000
10 000
100 000
15 m
(50 ft)
1.2 x 10'
1.2 x 10~4
1.2 x 10"'
30 m
(100 ft)
50 m
(165 ft)
1.8 x 10"'
1.8 x 10~4
2.9 x 10"'
2.9 x 10""
1.8 x 10"'
2.9 x 10"'
60 m
(200 ft)
3.6 x 1C"'
3.6 x 10~4
3.6 x 10"'
100 m
(330 ft)
5.8 x 10"'
5 8 x 10"4
5.8 x 10"'
Required hydraulic transmissivity for the
leak detection system, 8^, in m'/s
2.6-39
-------
Table 2.6-3. Required thickness, D, for a leak detection system as a
function of hydraulic conductivity, k^, of the leak
detection system material and the distance, L, between
the leak and the collector pipe or sump. These results
are related to a leakage rate of 10 000
Hters/lOOOm'/day (10,000 gallons/acra/day), a slope of
2% and an Impermeable bottom Hner. [Results from the
GeoServlces analytical study]
Hydraulic conductivity,
Slope
2%
4X
L
15 m
(50 ft)
60 m
(200 ft)
100 m
(330 ft)
15 m
(50 ft)
60 m
(200 ft)
100 m
(330 ft)
0.01 m/s
(1 cm/s)
9 mm
(0.3 in.)
35 mm
(1.3 in.)
58 mm
(2.5 in.)
4 mm
(0.17 in.)
17 mm
(0.64 1n.)
29 mm
(1.3 in.)
0.1 m/s
(10 cm/s)
0.9 mm
(0.03 in.)
3.5 mm
(0.13 in.)
5.8 mm
(0.25 in.)
0.4 mm
(0.017 in.)
1 .7 mm
(0.064 in.)
2.9 mm
(0.13 in.)
1 m/s
(100 cm/s)
0.1 mm
(0.003 in.)
0.3 mm
(0.013 in.)
0.6 mm
(0.025 in.)
0.04 mm
(0.0017 in.)
0.17 mm
(0.006 in.)
0.3 mm
(0.013 in.)
2.6-40
-------
Table 2.6-4. Values of capillary rise as a function of the hydraulic
conductivity of the drainage medium used In the leak
detection system.
Hydraulic Conductivity Capillary Rise
(m/s) (cm/s) (m) (nm) (in.)
1
5
1
5
1
5
1
5
1
5
1
5
x 10"'
x 10"'
x 10"5
x 10"*
x lO"4
x 10""
x 10"'
x 10"1
x 10"2
x 10~2
x 10"'
x 10"'
1
5
1 x 10""
5 x 10""
1 x 10"'
5 x 10"'
1 x 10"2
5 x 1C"2
i x 10" '
5 x 10"'
1
5
10
50
100
500
3.10
1.39
0.98
0.44
0.31
0.14
0.10
0.044
0.031
0.014
0.010
0.004
0.003
0.0014
3100
1386
980.3
438.4
310.0
138.6
98.0
43.8
31.0
13.9
9.8
4.4
3.1
1.4
122
54.6
38.6
17.3
12.2
5.5
3.9
1.7
1.2
0.55
0.39
0.17
0.12
0.055
2.6-41
-------
Table 2.6-5. Initial detection time (In days) as a function of the
top liner leakage rate and the hydraulic conductivity of
the leak detection system material for a 27. slope and an
Impermeable bottom liner. The leak detection system
material is assumed to be Initially dry. [Results from
the two-dimensional andlytical study]
Hydraulic Con-
ductivity of
Leak Detection
Top liner leakage rate per unit area, q
1 Iters/lOOOm'/day (gal lons/acre/day)
10
100
1000
Pea Gravel
10'' m/s
(1 cm/s)
10000
1000
100
10
C -ivel
li m/s
(10 cm/s)
3000
300
30
Coarse Gravel
1 m/s
(100 cm/s)
1000
100
JO
Initial detection times in days
2.6-42
-------
Table 2.6-6. Initial detection time (in days) as a function of the top
Uner leakage rate and the hydraulic conductivity of the
leak detection system material for a 27. slope and an
impermeable bottom liner. The potential capillary zone
Is assumed to have a degree of saturation of 75%.
[Results from the two-dimensional analytical study]
Hydraul1c Con-
ductivity of
Leak Detection
Top liner leakage rate per unit area, q
liters/1000m2/day (gal Ions/acre/day)
10
100
1000
Pea Gravel
10"2 m/s
(1 cm/s)
Gravel
10~' m/s
(10 cm/s)
Coarse Gravel
1 m/s
(100 cm/s)
2500
750
250
250
75
25
25
7.5
2.5
2.5
0.8
0.3
Initial detection times in days
2.6-43
-------
Table 2.6-7. Comparison between the results of the two-dimensional
analytical study (GeoServices study) and the U'o-
dimensional numerical study (Radian study). Initial
detectica times ( in days) are given as a function of the
top liner leakage rate. In both cases, the leak
detection system hydraulic conductivity is smaller than
10"* m/s (10~* cm/3) (the two-dimensional analytical
study has shown_ that hydraulic conductivities, kc,
smaller than 10 * m/s (10~* cm/s) lead to Initial
detection times identical to the initial detection times
related to kc = 10"" m/s (10~a cm/s)). The initial leak
detection times of the analytical study have been
obtained with an initial degree of saturation Sro - 757.,
which Is approximately identical to the degree of
saturation at the beginning of leakage in the numerical
study. The analytical study assumes an ideally
Impermeable bottom liner and should therefore be compared
to the first of the three cases isidered in the
numerical study. (Note: only the nui, ical study cases
with a composite bottom liner are considered in this
table.) Legend: (u) = uniform leak; (s) = sidewall leak
(see Figure 2.6-6b.)
GeoSer"ices (analytical)
Radian
(numerical )
Intact
bottom FML
intermediate
case
deteriorated
bottom FML
Leakage rate per
top liner, q, gal
49
520
600
(s)
—
Initial
60
425
495
(s)
__-
550
(s)
unit area through the
lons/acre/day
92
277
...
334
(u)
351
(u)
leak detection
780 1238
33 21
30
(u)
35
(u)
30
(u)
time, days
2.6-44
-------
Figure 2.6-1. Geometry of the leak detection system considered in the
two-dimensional analytical study.
2.6-45
-------
DISTANCE _ L
Figure 2.6-2. Leak detection time as a function of the distance
between the leak and the collector pipe or sump, and the
hydraulic conductivity of the leak detection system
material. This chart has been established assuming
steady-state flow in the leak detection system and an
ideally impermeable bottom liner. The slope of the leak
detection system is 2%. The ?ame results can also be
found in Tat.e 2.6-1. [Results from the two-dimensional
analytical study]
2.6-46
-------
«? io'
^^o^t /-«./c ft' f*ifa.ff
Figure 2.6-3. Leak detection time 3s a function :
rate and hydraulic conductivity of t
the leak detection system [two-c'--
study] assuming an ideally imperir.ei:
an initially dry leak detection s>:
independent of the thickness of
system prov:ded that the thicki-3
capillary rise (see Table 2.6-4).
that the last curve is valid
conductivities smaller than 10~" m/:
that the capillary rise for a grar
hydraulic conductivity of 10"* m/s
the thickness of the considered le
Caution: as explained in Figure 2.
above curves are not applicable
superseded by curves related to Dar:
f top 1 ir?- leakage
the iTiteriil used in
•ensic-al i*3lytical
:le b::ten ' iner and
:em. The ;_rves are
the l€ak :etection
:s is grener than
Howe.er, :he fact
for all -ydraulic
; is c.e tr the fact
. :ar r^ " eru! wi th a
is 0.3 m, */hich is
;< det=:tir- system.
:-4, c:rti:-3 of the
because :hey are
s flew.
2.6-47
-------
WC 10* 1C* 10' '
001
Figure 2.6-4. Detection time as a function of top liner leakage rate
and hydraulic conductivity of the material used in the
leak detection systerr [two-dimensional analytical study]
assuming an ideally impermeable bottom liner and an
initially dry leak detection system. The capillarity
curves (lines at <15° left of ABC) are independent of the
thickness D, the length L. and the slope 0 of the leak
detection system. The Darcy's curves (horizontal lines,
in zone ABD) are related to L = 60m (200 ft) and are
independent of the thickness D. On a given Darcy's
curve such as XY, the head on the bottom liner is zero
in X and increases linearly between X and Y where it
reaches its maximum value, that is, the thickness D of
the leak detection system. The head on the bottom liner
related to a point such as Y can even be greater than
the thickness of the leak detection system if there is a
large hole in the top liner and if the leakage through
that hole is limited by hydraulic transmi ssi vi ty of the
leak detection system (thereby building up pressure in
the leak detection system).
2.6-48
-------
to' .
r_<
\
\\\^ \
. \ \ \N ~v
01
I 10
<)f, r.lf. Pt
10'
(/itlrs/ /ro, -' /Jay)
( qa ilons /&(_, c/da a)
Figure 2.6-5. Detection time as a function of top liner leakage rate,
hvdraulic conductivity, and length of flow path [two-
c.mensional analytical study] assuming an impermeable
bottom liner. Generic chart which can be used directly
by expert user or can be used to generate series of
charts such as those in Figure 2.6-5. The curves at 45°
are related to capillarity- These curves are valid only
if the thickness of the leak detection layer is larger
than the value indicated on each curve. 1 s curves at
67.5° are related to Carey's flow.
2.6-49
-------
-
COUP*CIEO CLA*
Avvifvc not 10 SCAIE
Figure 2.6-6. Lining system modeled using the UNSAT2D program, to
study leak detection system performance: (a) geometry
of the system; (b) definition of the three types of
leaks considered [Radian, 1987a, b, c].
2.6-50
-------
2.7 PERFORMANCE CRITERIA FOR LEAK DETECTION SYSTEMS
2.7.1 Introduction
2.7.1.1 Scope of the Section
2.7.1.1.1 Purpose of the Section
The purpose of this section is to review the technical information
pertinent to performance criteria for leak detection systems. This
Information supports:
• the selection of the leak detection system performance
characteristics that are most appropriate to establish
performance criteria; and
• the establishment of the criteria themselves, which are
expressed as values (maximum or minimum, depending 'on the
considered characteristics) that must be met by the selected
performance characteristics.
2.7.1.1.2 Organization of the Section
The introduction:
• defines the concept of "performance criteria" and related
concepts; and
• reviews leak detection system performance characteristics that
can be considered to establish performance criteria, and
selects those that are deemed appropriate.
Then, the technical Information relevant to the various
performance criteria and the rationale for the establishment of these
criteria are presented 1n Sections 2.7.2 through 2.7.4.
2.7-1
-------
2. 1.2
Design Requirements
The term "design requirements" and "leak detection capability"
encompasses "performance criteria" and "design specifications". Both
concepts are discussed below.
2.7.1.2.1 The Concept of Performance Criteria
- Definition
Performance criteria quantify the minimum performance for which a
leak detection system should be designed. In other words, the
performance criteria considered here are design performance criteria,
rather than field performance criteria.
- Demonstration
As a result of the above definition, the owner or operator of the
hazardous waste management unit need not conduct a field test or to
make field measurements to demonstrate that the leak detection system,
when built, will meet the performance criteria. The owner or operator
should only be required to demonstrate that the design is such that
the leak detection system is expected to meet the performance
criteria, provided that it is properly constructed.
2.7.1.2.2 The Concept of Design Specifications
- Definition
Design specifications are requirements that selected physical
characteristics of the leak detection system should satisfy, such as
size of the system and properties of the materials used to construct
the system.
- Demonstration
Demonstration by the owner that a leak detection system meets
design specifications can be done in two stages:
2.7-2
-------
• at the design stage, the owner or operator should select leak
detection system geometry and iterials that meet the design
3cecifications; and
• at the construction stage, conformance testing of the materials
and measurements should be made to verify that the materials
used and the geometry of the leak detection system meet the
design specifications and are In agreement with the design.
2.7.1.2.3 Performance Criteria and Design Specifications
- Differences Between Performance Criteria and Design Specifications
Performance criteria and design specifications are different:
• Performance criteria are related to the way a leak detection
system must work as predicted by design.
• Design specifications are related to the way a leak detection
system must be buiIt.
The combination of performance criteria and design specifications
1s EPA's way of ensuring adequate leak detection capability in the
LDCRS of double liner systems. It is believed to be wise to require a
leak detection system to meet both performance criteria and design
specifications because it appears that in many cases performance
criteria will not be met unless design specifications are also imposed
(which implies that it is important for the design specifications to
be carefully chosen).
Design specifications appear to be essential. However, it should
not be concluded that design specifications alone are sufficient and
can replace a good design. Design specifications alone are not
sufficient because, even if a l,eak detection system meets all design
specifications, there is no guarantee that its field performance, or
even its design performance, will meet the performance criteria.
2.7-3
-------
- Complementarity of Performance Criteria and Design Specifications
One may argue that a lea' detection system should meet its
objectives regardless of the means Involved. In other words, one may
argue that only performance criteria are necessary and useful to
ensure adequate leak detection capability while design specifications
are unnecessary and useless. One may even argue that design
specifications are detrimental because, by specifying the means to be
used, they restrict Innovation and prevent designers from developing
and using new technologies. There Is some merit to this second
argument, and 1t should be understood that design specif'cations can
be modified and adapted when knowledge Increases and new technologies
become available. The RCRA amendments of 1984, In fact, call for
minimum technological requirements to be revised "from time to time to
take into account improvements in the technology of control and
measurement" RCRA 3004(o)(l). The recent history of the development
of the use of synthetic drainage layers .eflects EPA's willingness to
adapt design specifications when a pew technology becomes available
which appears to be able to meet design objectives and performance
criteria.
Design specifications are very useful, however, as shown by the
following five-point discussion:
• Special expertise is required to design a leak detection system
and to demonstrate that 1t will meet the performance criteria
and provide adequate leak detection capability.
• Special expertise is also required to evaluate such designs and
demonstrations.
• Because of the increasing use of synthetic drainage layers and
the possible growing use of other innovative technologies, some
designers, reviewers, and permit writers will lack the
necessary expertise to competently evaluate and apply the new
technologies, at least initially.
2.7-4
-------
• Guidelines on leak detection system design are therefore needed
by designers, reviewers, and permit writers.
• The simplest guidelines are d"-,ign specifications. They are
simple to follow and their implementation 1s simple to verify.
It may be concluded that design specifications are necessary
safeguards against errors 1n design that otherwise would be
undetected.
2.7.1.3 Performance Characteristics
2.7.1.3.1 Scope of the Section
- Purpose of this Section
The purpose of this section 1s: (1) to Identify the performance
characteristics, I.e., the parameters which characterize the
performance of a leak detection system; and (11) to select the
characteristics that will be used to establish performance criteria.
- Organization of the Section
This section includes two subsections. The first subsection is
devoted to a review of the leak detection capabilities that leak
detection systems should have in general. In the second subsection,
the capabilities of the leak detection system are expressed in terms
of specific performance characteristics, relevant to the types of leak
detection systems considered in this background document.
2.7.1.3.2 Leak Detection System Capabilities
- Purpose of Leakage Monitoring
There are two reasons for monitoring the presence of leakage
between the two liners: (1) to prevent its causes; and (ii) to
minimize Us consequences. Causes and consequences are discussed
hereafter.
2.7-5
-------
Leakage Is caused by leaks. To p: -ent the causes of leakage,
leaks must be detected, located, and repaired. Leak location
deter:;iination, therefore, Is a required capability of a leakage
monitoring system Intended to provide information in order to prevent
the causes of leakage.
The presence of leakage between the top and bottom liners of a
double liner system has a detrimental consequence: a liquid head Is
created on the bottom Uner. This liquid head will cause leakaga Into
the bottom liner. If the bottom liner Is defective and if the liquid
head becomes large, leakage may pass through the bottom liner and into
the ground. With EPA's "systems approach", the prevention of
migration of hazardous constituents Into the bottom liner and out of
the unit can be minimized by proper design, construction and operation
of two lining system components: the bottom liner and the leak
detection system. With respect to the bottom liner, EPA has recently
shown [USEPA, 1987] that composite bottom liners will minimize the
potential for migration of hazardous constituents into the bottom
liner. With respect to the leak detection system, the potential for
migration into the bottom liner can be minimized by minimizing the
hydraulic head acting on top of the bottom liner. Ideally, to ensure
a minimum hydraulic head, .he liquid head on the bottom liner should
be measured at many locations. This is extremely difficult because of
the amount of monitoring equipment that would have to be installed,
and the complexity and cost of equipment maintenance and data
collection. It 1s much more practical to make an indirect evaluation
of liquid head by measuring leakage rates, since liquid head on the
bottom liner 1s linked to the leakage rate through the top liner.
Therefore, a leakage monitoring system, whose purpose is to provide
useful information so that the consequences of the presence of leakage
can be minimized, must have the capability of evaluating leakage rate.
Whether the purpose of monitoring leakage is to provide
Information for preventing leakage causes or minimizing leakage
consequences, the rapidity with which the information is obtained is
important.
2.7-6
-------
In addition, a leak detection system must perform throughout the
active life and post-closure care period of the land disposal unit.
Therefore, the materials used to construct a leak detection system
must have adequate durability when placed in contact with the ' _-:hate
that results from the treatment, storage or disposal of hazardous
wastes at a specific unit. This Important performance requirement Is
discussed in Section 2.4.3.5. '
- Identification of Leak Detection Systems Capabilities
The conclusion of the above discussion is that, for leak detection
systems 1n general, the following capabilities can be considered:
• leak location determination, if the purpose is to prevent
leakage causes, i.e., to repair leaks;
• leakage rate evaluation, if the purpose 1s to minimize leakage
consequences; and
• rapidity with which the Information is obtained, in both cases.
These general capabilities are discussed below to establish a list
of performance characteristics relevant to the leak detection systems
considered in this background document.
2.7.1.3.3 Selection of Performance Characteristics
- Purpose of the Section
The purpose of this section is to select, after considering
available technologies, the performance characteristics of a leak
detection system which should be used to establish performance
cri teria.
According to Section 2.7.1.3.2, the following leak detection
system capabilities should be addressed in order to select relevant
performance characteristics:
2.7-7
-------
• leak location determination;
• evaluation of leakage rate; and
• rapidity of providing information.
These should be translated into simple performance characteristics
which can be used to express performance criteria 1n a practical
manner. This 1s the purpose of the following discussion.
- Discussion of Leak Location Determination
As discussed in Section 2.3.1, it is very difficult and expensive
to identify the location of leaks. In addition, there 1s not much
Incentive for finding the location of leaks:
• In landfills, the occurrence of significant rates of leakage
are unlikely after closure jf the unit. If - jor leakage
through the top liner is detected during landfill operation, a
typical remedial action consists of capping a certain area of
the unit (i.e., covering the waste with a liner) to prevent
infiltration of rainwater, thereby drastically reducing
leachate production. It is not necessary to know exactly where
1 lakaqe occurs to design >e capping of an area of a landfill.
It is, however, useful to subdivide a large landfill into cells
with independent leak detection systems to be able to identify
in which area leakage occurs.
• In surface Impoundments, the strategy in case of major leakage
1s different. The Impoundment can be emptied and the liner
repaired or replaced ("retrofitted"). It is therefore useful
to know the location of leaks. However, leaks can be located
by means that do not need to be built in the lining system.
These means include portable electric resistivity equipment,
vacuum testing, and visual inspection of the lining system
after emptying the Impoundment.
2.7-8
-------
• In waste plies, leak location determination is usually not
needed because of the small size and the temporary nature of
these units.
It therefore appears that, from a technical standpoint, it is not
crucial to build Into the leak detection system a permanent means to
identify leak location (although owners or operators may e'ect to do
this voluntarily). This conclusion 1s corroborated by the discussion
in Section 2.3.2 that led to the recommendation to use, as a leak
detection system, the leachate collection and removal system located
between the two liners, which cannot determine leak location.
Although the ability to determine the location of leaks is a
desirable characteristic of a leak detection system, it appears frcm
the above discussion that 1t Is neither necessary nor practical (with
presently available technology) to consider the determination of leak
location as a performance characteristic for which a general
performance criterion should be established. However, if a leak
detection system, which meets all general performance criteria and
design specifications, also has the capability of locating leaks, it
clearly has an additional benefit.
- Discussion of Leakage Rate Evaluation
The ability of a leak detection system to evaluate leakage rates
can be expressed by two performance characteristics:
• the leachate collection efficiency; and
• the leak detection sensitivity.
These two performance characteristics can be defined as follows:
• The leachate collection efficiency 1s the ratio between leakage
rate conveyed by the leak detection system and collected at the
sump, and the leakage rate that impinges the leak detection
system (I.e., the leakage rate that has passed through the top
1iner).
2.7-9
-------
• The leak det tion sensitivity is the smallest leakage rate
that can be detected.
As will be discussed 1n Section 2.7.4, it is impractical (although
1t 1s desirable) to establish a leak detection system performance
criterion on the basis of the leachate collection efficiency.
Therefore the only performance characteristic related to leakage rate
evaluation that will be considered in the establishment of performance
criteria 1s the leak detection sensitivit, (which will be discussed in
detail 1n Section 2.7.2).
- Discussion of Speed at ich Information is Provided
The speed at which information is provided by the leak detection
system can be expressed by the "detection time" which is essentially
the time necessary to detect leakage which originates in a leak as far
away as possible from the sump. The concept of detection time 1s
discussed in detail in Section 2.7.3.
- Recapitulation of Performance Characteristics
As a result of the above discussion, the following performance
characteristics are considered:
• detection sensitivity; and
• detection time.
These two performance characteristics are discussed in detail in
Sections 2.7.2 and 2.7.3.
2.7-10
-------
2.7.2 Detection Sensitivity
2.7.2.1 Introduction
2.7.2.1.1 Scope of the Section
- Purpose of the Section
The purpose of this section is: (1) to present the technical
Information available to support the establishment of a performance
criterion regarding detection sensitivity; and (1i) to establish the
detection sensitivity performance criterion.
- Organization of the Section
The Introduction includes a definition of detection sensitivity
and a discussion of the Importance of this concept. A review of the
available information and a rationale for establishing the performance
criterion is presented in Section 2.7.2.2. The criterion is presented
in Section 2.7.2.3, with a discussion.
2.7.2.1.2 Definition arid Importance of Detection Sensitivity
- Definition
The detection sensitivity of a leak detection system can be
defined as:
Detection sensitivity 1s the smallest quantity of
liquid that can pass through the top liner and be
detected by the leak detection system.
- Importance of Detection Sensitivity
Detection sensitivity is an essential performance characteristic
of any leakage monitoring system, as indicated in Section 2.7.1.3.3.
In addition, a high leak detection sensitivity is consistent with the
statutory requirements of the RCRA amendments which require the leak
2.7-11
-------
detection system be able to detect leakage "at the earliest
practicable time".
A leak detection system with a high sensitivity (i.e. able to
detect a small leakage rate) permits a detailed monitoring of the
behavior of a land disposal unit. This 1s very useful to the owner or
operator of the considered unit and it is useful in general, in that
It Increases knowledge about the ' ictioning of lining systems used In
land disposal units. The expe .nee thus gained will benefit the
designers, the owners, the operators, and the environment.
2.7.2.2 Establishment of the Detection SenslUyHy Criterion
2.7.2.2.1 Summary of Relevant Information
Available technical Information wl.ich can be used to establish the
detection sensitivity performance criterion can be summarized as
follows:
• Table 2.7-1 gives calculated leakage rates through FML and
composite top liners. This table (which is a reproduction of
Table 2.2-16) shows that leakage rate under a hydraulic head of
0.3 m (1 ft), which is the specified maximum permissible head
on the top Uner of a landfill, is on the order of 0.3 Ltd
(gpad). This table also shows that leakage rate under a
hydraulic head of 0.03 m (0.1 ft), which may represent a
typical average head on the top liner of a landfill, is on the
order of 0.02 Ltd (gpad).
• Table 2.7-1 also shows that leakage rate under a hydraulic head
of 0.03 m (0.1 ft), which is conservative for a bottom liner
under almost all operating conditions, is very small: 0.001
Ltd (gpad) due to permeation through the FML, and 0.02 Ltd
(gpad) due to one standard hole or defect In the FML component
of a composite liner (i.e., a 1 cm1 (0.16 in'.) hole) per 4000
m' (acre) with good contact between the FML and compacted soil.
2.7-12
-------
• Using Darcy's equation, it is possible to rrake a calculation
showing that a bottom liner made exclusively of compacted soil
(i.e., without a FHL) would exhibit the following leakage rates
(almost regardless of Its thickness): approximately 900 Ltd
(gpad) if the hydraulic conductivity of the compacted soil is
10~* m/s (10~* cm/s), and approximately 90 Ltd (gpad) if the
hydraulic conductivity of the compacted soil is 10"' m/s (10~7
cm/s).
The above technical information will be used in the next section.
2.7.2.2.2 Rationale for the Criterion
- Lower Boundary for the Criterion
Clearly, a lower boundary for the detection sensitivity is the
leakage rate through the bottom liner. If a compacted soil bottom
liner is considered, the leakage rate through the bottom liner ., :n the
order of 90-900 Ltd (gpad) as given in the previous Section) will in
most cases exceed the leakage rate through the top liner given in
Table 2.2-16. As a result, no leakage will be detected. Therefore, a
detection sensitivity of 90-900 Ltd (gpad) is too large to meet the
requirement to detect leaks at the earliest practicable time.
If a composite bottom liner is used, the leakage rate will be
0.001 Ltd (gpad) as a result of permeation through the FML, and in the
range of 0.02 Ltd (gpad) as a result of holes, as shown in Table 2.7-
1. However, 1n evaluating detection sensitivity it is important to
realize that holes should not be considered since there will always be
some leakage that will bypass the holes and be detected. Therefore,
when a composite bottom liner is used, a lower boundary for the
detection sensitivity 1s 0.001 Ltd (gpad), which 1s the leakage rate
due to permeation through the FML.
- Upper Boundary for the Criterion
A leak detection system should be capable of detecting leakage due
to a standard FML hole. Then, if a hole develops 1n a FML during the
2.7-13
-------
active period or post-closure care p~-iod of a hazardous land disposal
unit, 'eakage through tne !-:ie C5n :; detected. Since both FML and
ccmpos e top liners are ar:.;ed -y ETA, both snould be considered in
setting the upper boundary of t~e Detection sensitivity criterion.
Since the composite liner will allow less top liner leakage than the
FML, it will be more critic*! in set:ing the criterion. Therefore,
according to Table 2.7-1, tJ-.s smallest leakage rates due to holes in
composite 1Iners are:
• 0.02 Ltd (gpad) with 2 hydrauKc head of 0.03 m (0.1 ft) which
may be assumed to correspond t: average conditions on the top
liner of a landfll1.
• 0.3 Ltd (gpad; with z. hydraulit head of 0.3 m (1 ft) that is
typically considered is the des-;n of a landfill top liner.
- Conclusion
The lower boundary of O.G01 Ltd 'gpad) 1s very small and would
require a leakage rate measurement precision that is not typically
feasible in the field. Therefore, a \='ue close to the upper boundary
should be considered, such as 0.1 Ltd gpad). This rate of top liner
leakage represents a very hie- level c: detection sensitivity and can
be considered to represent 3CAT for lea< detection systems.
2.7.2.3 Presentation of_the Detect1c-_Sensitiv1ty_Cr1ter1on
2.7.2.3.1 Expression of the Criterion
The proposed detection se-sitivity rerformance criterion resulting
from the above analysis wou'd :e as fol'rws:
The smallest leakage rate that can be
detected by (the "d=:ect1cn sensitivity" of)
a leak detection syste-n r'lould be 0.1
lUer/lOOOm'/day (0.1 gallon/'icre/lay).
2.7-14
-------
This proposed value represents the best demonstrated available
technology (BCAT) to the best of our knowledge. However, since the
collecMon of liquid quantities as small such as 0.1 IHer/lOOOnT Jay
(0.1 g^l lon/acre/day) may not be possible in the field (due to the
finite capabilities of collector systems, sumps, and pumps), a higher
value of detection sensitivity, such as 1 1iter/lOOOm'/day
(1 gallon/acre/day) should be considered.
2.7.2.3.2 Discussion
The above detection sensitivity performance criterion can be met
only 1f the leak detection system is properly designed and
constructed.
In order to design a leak detection system in a way that increases
the detection sensitivity (I.e., that decreases the size of the
smallest leakage rate that can be detected), the designer should
recognize that it 1s necessary to decrease capillarity in order to
minimize the amount of liquid held in (.he system. The following can
be done to decrease capillarity:
• The leak detection system material should have a high hydraulic
conductivity such as 10~z m/s (1 cm/s) to have a small
capillary effect as indicated 1n Section 2.8.
• A judicious arrangement of saw-tooth slopes at the bottom of
the landfill may concentrate the flow in channels, which favors
: Darcy's flow and reduces the relative importance of
j capillarity.
I
i To properly construct a leak detection system, at least the
, following should be done:
• Drainage materials such as coarse sand, gravel or geonets
should be clean. Therefore, they should be washed at the site,
prior to installation (even If they look clean, because
experience shows that they are never really clean), and they
2.7-15
-------
must be protected from dust and dirt prior to being covered by
other materials of the lining system.
• FMLs used as the upper component of a composite bottom liner
must be placed without significant wrinkles, and the quality
assurance monitors should verify that the pattern of small
wrinkles, which cannot be eliminated, do not form closed areas
where liquids will pond. To that end, observations after a
rainfall are very useful, especially if photographs are taken.
2,7.3 Detection Time
2.7,3.1 Introduction
2.7.3.1.1 Scope of the Section
- Purpose of the Section
The purpose of this sectlcT is: (1) to present the technical
information available to support the establishment of a performance
criterion regarding detection time; and (ti) to establish the
detection time performance criterion.
- Organization of the Section
The introduction of this section includes a definition of
detection time and a discussion of the Importance of this concept. A
review of the available information and a rationale for establishing
the performance criterion is presented in Section 2.7.3.2. Finally,
the criterion is presented in Section 2.7.3.3, with a discussion.
2.7.3.1.2 Definition and Importance of Detection Time
- Definition
The detection time of a leak detection system can be defined as:
2.7-16
-------
Detection time is the time from when liquid
enters the leachate collection and removal
system between the liners (i.e., the leak
detection system) to when it is present (and
available, for removal) 1n the leachate
collection and removal system sump.
- Comments
The proposed Liner/Leak Detection Rule will require a leak
detection system that is capable of detecting leakage that migrates
through the top liner into the space between the liners at the
"earliest practicable time". This implies that the detection time
should be short but not so short that it would be impractical to meet
with currently available technology. The proposed performance
criterion should take this consideration into account.
If a FML/compacted soil composite liner is used, is there leakage
when leachate passes through the FML (which is the upper component of
the composite liner) or when leachate passes through the soil (which
is the lower component of the composite liner)? This issue has been
debated in the past. But, clearly, there is one logical answer:
since a composite liner is one liner, not two, there is leakage only
when liquid has passed through both components. In fact, the
definition given above is clear: the initial time for measuring the
detection time is the time when leakage enters the ieak detection
system.
- Importance of Detection Time
Detection time is an essential performance characteristic of any
leakage monitoring system as indicated in Section 2.7.1.3.3. In
addition, strict detection time performance criteria are in agreement
with the proposed Liner/Leak Detection Rule requirement to ensure
detection "at the earliest practicable time".
2.7-17
-------
A short detector time is important, especially for large leaks,
because It Is a prerequisite for prompt triggering of appropriate
Response Action Plan (see Section 2.10).
2.7.3.2 Establishment of the Detect1on_T1me Crjterion
2.7.3.2.1 Summary of Relevant Information
Available technical Information which can be used to establish the
detection time criterion 1s summarized in Table 2.7-2 which gives
detection time as a function of the hydraulic conductivity of the leak
detection system material and the distance between the leak and the
collector pipe or the sump. This table is a reproduction of Table
2.6-1.
2.7.3.2.2 Rationale for the Criterion
- Presentation of the Rationale
The leak detection time should be shorter than the monitoring
period so that any leakage event happening in a given monitoring
period can be detected during that period. Therefore, if a daily
monitoring of the leak detection sumps 1s required, a leak detection
time of one day maximum is recommended.
As shown In Table 2.7-2, such a leak detection time would be
achieved with a hydraulic conductivity of the leak detection system
material of 10~J m/s (1 cm/s) and a distance between the leak and the
collector pipe or sump less than 50 m (165 ft).
2.7.3.3 Presentation of the Detection_T1me Criterion
2.7.3.3.1 Expression of the Criterion
As a result of the above analysis, the proposed detection time
performance criterion of a leak detection system would be as follows:
2.7-18
-------
The time from when liquid enters the leachote
collection and removal system between the liners
(i.e., the leak detection system) to when it is
available for removal in the leakage detection,
removal, and collection sump should be smaller than
one day assuming steady-state flow conditions.
The proposed value represents the best demonstrated ava1l?ble
technology (BOAT) to the best of our knowledge. However, 1t can be
replaced by other values for practical reasons.
2.7.3.3.2 Discussion
The above detection time performance criterion can be met only If
the leak detection system Is properly designed and constructed.
In order to design a leak detection system in a way that decreases
the detection time, the designer should recognize that it 1s necessary
to decrease capillarity in order to minimize the amount of liquid held
1n the system. The following can be done to decrease capillarity:
• The leak detection system material should have a high hydraulic
conductivity such as 10~* m/s (1 cm/s) to have a small
capillary effect as indicated in Section 2.8.
• A judicious arrangement of saw-tooth slopes at the bottom of
the landfill may concentrate the flow in channels, which favors
Darcy's flow and reduces the relative Importance of
capillarity.
To properly construct a leak detection system, at least the
following should be done:
• Drainage materials such as coarse sand, gravel or geonets
should be clean. Therefore, they must be washed at the site,
prior to Installation (even 1f they look clean, because
experience shows that they are never really clean), and they
2.7-19
-------
must be protected from dust and dirt prior o being covered by
other materials of the lining system.
• FMLs used as the upper component of a composite bottom liner
must be placed without significant wrinkles, and the quality
assurance monitors should verify that the pattern of small
wrinkles, which cannot be eliminated, do not form closed areas
where liquids will pond. To that end, observations after a
rainfall are very useful, especially if photographs are taken.
2.7.4 Leachate Collection Efficiency
2.7.4.1 Introduction
2.7.4.1.1 Scope of the Section
- Purpose of the Section
The purpose of this section is to discuss the concept of leachate
collection efficiency and evaluate if it can be used as a performance
criterion for a leak detection system.
- Organization r the Section
The Introduction of this section includes a definition of leachate
collection efficiency. A discusr-ion of the concept is presented in
Section 2.7.4.2.
2.7.4.1.2 Definition of Leachate Collection Efficiency
The leachate collection efficiency of a leak detection system can
be defined as:
The leachate collection efficiency is the ratio of the
leakage collected at the leak detection system sump
divided by the leakage entering the leak detection
system.
2.7-20
-------
2.7.4.2 Discussion of_the C :ept of Leachate_Co]lect1on Efficiency
2.7.4.2.1 Analysis of Leakage Types
Figure 2.7-1 (which is a reproduction of Figure 2.1-3),
illustrates that four leakages can be defined for a lining system:
• C = leakage from the LCRS Into the top liner;
• E = leakage through the top liner (i.e., leakage which impinges
Into the leak detection system);
• H - leakage from the leak detection system into the bottom
liner; and
• J « leakage through the bottom liner (i.e., leakage into the
ground).
Clearly, leakage of hazardous leachate into the ground (leakage J
in Figure 2.7-1) must be absolutely minimized - ideally, prevented.
The following relationships exist between the four leakages:
• Leakage Into the ground (J in Figure 2.7-1) is smaller than
leakage from the leak detection system into the bottom liner (H
in Figure 2.7-1) because a fraction of leakage H is absoiaed
by, or otherwise entraoped in, the bottom liner.
• Leakage H, from the leak detection system, is much smaller than
leakage E, through the top liner, because a portion F of
leakage E remains entrapped in the leak detection system (by
capillarity or otherwise) anJ a portion G of leakage E is
conveyed by the leak detection system to a sump where it is
collected and from where it is removed.
• Leakage E, through tlie top liner, is smaller than leakage C
Into the top Hner because a portion D of leakage E Is absorbed
by, or otherwise entrapped in, the top Hner.
2.7-21
-------
n
The following relationship exists between the four leakages
discussed above:
J < H « E < C (Equation 2.7-1)
where < means smaller than and « much smaller than.
The amount of leakage, G, collected in the sump is given by the
following equation:
G - E-H-F (Equation 2.7-2)
It Is Important that the leakage from the leak detection system,
Into the bottom liner (leakage H 1n Figure 2.7-1) be as small as
possible for two reasons:
• to make sure that leakage into the ground (leakage J in Figure
2.7-1) Is very small; and
• to ensure proper monitoring of the performance of the top liner
by making sure that leakage rate measured at the sump (i.e.,
E-H-F) 1s as close as possible to leakage E through the top
1iner.
A bottom liner that leaks very much (i.e., large H) would give the
Illusion that the lining system functions well (i.e., s^all value of
observed leakage E-H-F), while allowing significant ground pollution
(I.e., large value of J).
Leakage from the leak detection system into the bottom liner
(i.e., H In Figure 2.7-1) 1s small if the following conditions are
met:
• the depth of liquid in the leak detection system is small
(i.e., the hydraulic head acting on the bottom liner is small);
2.7-22
-------
n
• the bottom liner is a FML-ccmpacted soil composite liner, which
absorbs much less liquid than a compacleJ sol! alone (see the
EPA Background Document on "Bottom Liner Performance in Double-
Lined Landfills and Surface Impoundments", USEPA, 1987); and
• extensive quality assurance is performed during the
construction of the two components of the bottom composite
Mner, the soil and the FML, to ensure no major defects in
either component.
'he first of the three above conditions, a small liquid depth in
the leak detection system, requires the following:
• leakage E through the top liner should be small (which is
better achieved if the top liner is a composite liner than a
FML alone);
• the leak detection system _.,ould be adequately designed with
high hydraulic conductivity materials to ensure small liquid
depth; and
• the leak detection system should be properly constructed, with
clean draining materials to prevent clogging, and appropriate
slopes to present ponding of the collected leakage In the leak
detection system.
The last requirement implies that construction quality assurance
of lining systems should not be restricted to liner materials, but
should also be concerned with the construction of the drainage layers.
The first requirement (small leakage E) requires that leakage C be
small, which 1n turns requires that, in the case of landfill, the
leachate depth 1n the leachate collection and removal system above the
top liner be small. This further emphasizes the importance of proper
design and quality assurance of drainage layers.
2.7-23
-------
2.7.4.2.2 Evaluation of Leachate Collection Efficiency
- Expression of Leachate Collection Efficiency
Using the definition presented in Section 2.7.4.1.2 and the
analysis presented 1n Section 2.7.4.2.1, the cumulative leachate
collection efficiency (measured from the time of unit startup to any
other point 1n time) can be expressed as follows:
CLCE - G/E (Equation 2.7-3)
Combining Equations 2.7-2 and 2.7-3 gives:
CLCE « 1 - (H + F)/E (Equation 2.7-4)
where: CLCE = cumulative leachate collection efficiency; G »
cumulative amount of leakage collected at the leak detection system
sump; E = cumulative leakage through the top liner (I.e., leakage
which enters the leak detection system); H =• cumulative leakage from
the leak detection system into the bottom liner; and F = cumulative
leakage that remains entrapped in the leak detection system. The
cumulative leachate collection efficiency, CLCE, 1s dimensionless,
while G, E, H and F should be expressed in 11 ters/lOOOm2 or
gal Ions/acre.
Another way of defining collection efficiency is the steady-state
leakage collection efficiency (SSLCE). The determination of the SSLCE
assumes a constan1. rate of leakage from the top liner into the LDCRS
and from the LDCRb into the bottom liner. Further, under steady-state
conditions the quantity of liquid held in storage in the LDCRS is
assumed to be constant. For this set of assumptions, the SSLCE 1s
given by:
SSLCE - 1 - H/E (Equation 2.7-5)
2.7-24
-------
where: H and E can be expressed in units of 11ters/10CCm2/doy (Ltd)
or gallons/acre/ (gpad), while SSLCE is dimensionless.
- Discussion
Table 2.7-3 provides SSLCE values for several different lining
systems. From Table 2.7-3, 1t can be observed that the SSLCE of land
disposal units with composite bottom liners are better than units with
compacted soil bottom liners. For this reason, as well as for the
better detection sensitivities obtained with composite bottom liners
composites are strongly recommended for the bottom liners at land
disposal units.
- Conclusion
Two leachate collection efficiencies have been identified: the
steady-state value (SSLCE) and the cumulative value (CLCE). These two
values will be similar 1f the leachate stored in the LDCRS is small
(if F is small in Figure 2.7-1) and if the bottom liner minimizes
leakage Into 1t (1f H is small in Figure 2.7-1). Both of these
conditions are met by the leak detection system proposed in this
background document. First, leachate stored in the LDCRS is minimized
by using a drainage medium in the LDCRS with minimal capillarity. As
will be seen in Section 2.8 of this report, by using a drainage medium
with a minimum hydraulic conductivity of 10"2 m/s (1 cm/s),
capillarity will be minimal. Second, leakage into the bottom liner is
minimized through the use of a composite bottom liner. Since these
two conditions will be met in practice, it is acceptable to consider
the steady-state leachate collection efficiency (which is easy to
calculate) rather than the cumulative leachate collection efficiency
(which is more difficult to calculate).
The steady-state leachate collection efficiency (SSLCE) is very
sensitive to the type of bottom Uner in the land disposal unit. This
fact can be seen 1n Table 2.7-3, where it is shown that a certain
landfill with a compacted soil bottom liner has a SSLCE of 457., while
in contrast, the same landfill with a composite bottom liner has a
SSLCE of 99.9'/.. Clearly, the SSLCE highlights differences in the
2.7-25
-------
performance of compacted soil and composite bottom liners. Howeve-,
if only composite bottom liners are considered, which is the case for
tne leak detection systems proposed in this background document, t'-e
SSLCE will be very high for all cases except those where the rate of
top liner leakage is insignificantly small. Since the SSLCE will fee
very high for all of the leak detection systems proposed in this
document, there Is no need to explicitly require SSLCE as a
performance criterion.
2.7-26
-------
Table 2.7-1. Leakage rates through FML and composite top liners. The
small hole has a diameter of 2 ma (0.08 in.). The
standard hole has a surface a- of 1 cm2 (0.15 in;).
The frequency of holes is 1 per ^OiB2 (1 per acre). The
thickness of the compacted soil layer is 0.9 m (3 ft) and
Us hydraulic conductivity is 10~" m/s (10""' cm/s).
Note: Ltd - Uter/lOOCmVday; gpad -gallons/acre/day; 1
Ltd » 1.1 gpad. This table is a reproduction of Table
2.2-16 from Section 2.2.
Type
of
Liner
FML
alone
Composite
1 iner
(good)
contact)
Composite
1 iner
(poor)
contact)
Leakage
mechanism
Permeation
Small hole
TOTAL
Permeation
Standard hole
TOTAL
Permeation
Standard hole
TOTAL
Permeation
Standard hole
TOTAL
Hydraul ic fiead, h
0.03 m
(0.1 ft)
0.001
30
30
O.C01
300
300
0.001
0.02
0.02
0.001
1
1
0.3 m
(1 ft)
0.1
100
100
0.1
1,000
1,000
0.1
0.2
0.3
0.1
8
8
3 m
(10 ft)
10
300
300
10
3,000
3,000
10
3
13
10
50
60
Values of leakage rate in Ltd or gpad
2.7-27
-------
Table 2.7-2. Leak detection time as a function of the distance, L,
between the leak and the collector pipe, and the
hydraulic conductivity of the leak detection system
material. This table nas been established assuming
steady-state flow in the leak detection system. The
slope of the leak detectio,. system is 2'/.. This table is
a reproduction of Table 2.6-1. [Result of the two-
dimensional analytical study presented 1n Section
2.6.2.]
L
15 m
(50 ft)
30 m
(100 ft)
50 m
(165 ft)
60 m
(200 ft)
100 m
(330 ft)
Hydraulic conductivity of the
leak detection system material , k^
10"" m/s
(10~' cm/s
26
52
87
104
174
10"' m/s
(10"1 cm/s)
3
5
9
11
18
10"2 m/s
(1 cm/s)
0.3
0.5
0.9
1.1
2
10"' m/s
(10 cm/s)
0.03
0.05
0.01
0.1
0.2
Leak detection times in days
2.7-28
-------
Table 2.7-3. Examples of steady-.tate leachate collection efficiencies
(SSLCE). Values for leakage through the top liner (E) and into
the bottom liner (H) are obtained from Table 2.7-1. G and SSLCE
were calculated using Equations 2.7-2 and 2.7-4. The various
types of leakage, E, H, and G, are illustrat2d in Figure 2.7-1.
Case
(1)
(2)
(3)
(4)
(5)
Description
Landfill with FML
top liner and
compacted soil
bottom liner.
Landfill with FML
top 1 iner and
composite bottom
liner.
Landfill with
composite top liner
and compacted soil
bottom liner.
Landfill with
composite top
1 iner and
bottom Uner.
Surface Impoundment
with composite
top 1 1ner and
composite bottom
1 Iner.
Various leakages as defined
In Figure 2.7-1 In Ltd (gpad)
E
200
200
1
1
20
H
90
0.2
1
0.2
0.2
G
110
199.8
0
.8
19.8
Steady
State
Leachate
Collection
Efficiency
SSLCE
0.45
0.999
0
0.80
0.99
2.7-29
-------
Leachate Col lection and
Removal System (LCRS)
A = Isachate collected in the LCRS
B = leachate stored* in LCRS
C = leachate from ti.e LCRS into top liner
Top Liner
D = ieachate stored* in top liner
E = leakage from the top liner into the
LOCRS
Leak Detection
Col lection and
Removal System
(LDCRS)
G = leakage collected in the LDCRS sump
F « leakage stored* in LDCRS
H = leakage from the LDCRS into the
bottom liner
Bottom Liner
I = leakage stored* in the bottom liner
Ground
leakage from the bottom liner
into the ground
* Stored liquids due to capillarity, absorption, etc.
Figure 2.7-1.
Fate of liquids entering a double liner system at a
landfill unit. Thie Figure is a reproduction of
Figure 2.1-3.
2.7-30
-------
2.8 DESIGN SPECIFICATIONS FOR LEAK DETECTION SYSTEMS
2.3.1 Introduction
2.8.1.1 Scope of the Section
2.8.1.1.1 Purpose o* the Section
The purpose of this section 1s to review the technical Information
pertinent to design specifications for leak detection systems. This
Information supports:
• the selection of the leak detection system design parameters
that are most appropriate to establish design specifications;
and
• the establishment of the .^eclfications themselves, which are
expressed as values (maximum or minimum, depending on the
considered parameter) that must be met by the selected design
parameters.
2.8.1.1.2 Organization of the Section
This section Is organized as follows:
• The remainder of this introduction disc sses the concept of
"design specifications".
• Section 2.8.2 reviews the leak detection system design
parameters that can be considered to establish design
specifications, and selects those that are deemed appropriate.
• Section 2.8.3 reviews the technical information relevant to the
various design specifications and the rationale for the
establishment of these specifications.
• Section 2.8.4 presents and discusses the design specifications.
2.8-1
-------
2.8.1.2 The Concept of Design Spec'f 1c-it1c .
2.8.1.2.1 Definition
Design specifications are requirements that selected physical
characteristics of the leak detection system should meet. Examples of
such characteristics are the dimensions of the leak detection system
and properties of the materials used to construct the system. These
physical characteristics which are selected to establish design
specifications are those which govern the design and hereafter they
are called design parameters.
2.8.1.2.2 Demonstration
Demonstration by the owner that a leak detection system meets
design specifications is done in two stages:
• at the design stage, the owner should select, and write project
specifications for, leak detection system geometry and
materials that meet the design specifications; and
• at the construction stage, conformance testing and measurements
should be made to verify that :he materials used in the leak
detection system and the geometry of the leak detection system
meet the design specifications and agree ith the design.
2.8.1.2.3 Usefulness of Design Specifications
A comparison between performance criteria and design
specifications 1s presented in Section 2.7.1.2.3. From this
cor^arison, the following conclusions can be drawn regarding the
usefulness of design specifications:
• Spe al expertise is required to design a leak detection system
and to demonstrate that it will meet performance criteria and
provide adequate leak detection capability.
2.8-2
-------
• Special expertise is also required to evaluate such designs and
demonstrations.
• Because of the Increasing use of synthetic drainage layers and
the possible growing use cf other innovative technologies, It
Is expected that some designers, reviewers, and permit writers
will, at least initially, lack expertise 1n these technologies.
• Guidelines on leak detection system design are therefore needed
by designers, reviewers, and permit writers.
• The simplest guidelines are design specifications. They are
simple to follow and their Implementation 1s simple to verify.
it may be concluded that design specifications are necessary
safeguards. Without them, mistakes could be made at the design stage,
? id many would remain undetected.
2.3.1.2.4 Conservatlveness of Design Specifications
When a design must meet performance criteria and satisfy design
specifications, the design specifications do not need to be
conservative because their role in this case, is only to act as
safeguards. For such cases, it may e/en be preferable to have design
specifications that are net conservative In order to allow designers
latitude for Innovation, provided they can demonstrate those
innovative designs meet the performance criteria.
Design specifications must be conservative in those simple
applications where no real design is performed and where the only
performance criterion is that "it should work".
2.8-3
-------
2.8.2 Selection of Design Par-• Deters
2.8.2.1 Introduction
2.8.2.1.1 Purpose
To be adequate, design specifications must Include the design
parameters which have an Important Influence on the performance of the
system. It 1s therefore essential that design parameters which will
be used to establish design specifications be carefully selected.
2.8.2.1.2 Method
Equations, charts, and numerical results pertinent to the
performance of leak detection systems should be reviewed to determine
which parameters are Important. As discussed in Section 2.7, the
essential performance characteristics are detection sensitivity and
detection time. Leak detection systems are also used as leachate
collection and removal systems between the liners. As such, their
relevant performance characteristic is the liquid head on the bottom
liner, which must be small. Consequently, equations, charts, and
numerical results related to detection sensitivity, detection time,
and liquid head should be reviewed.
2.C.2.2 Review of Design Parameters
2.8.2.2.1 Review of Parameters Governing Detection Sensitivity
As discussed in Section 2.7.2, the required -'•'tectlon sensitivity
can be achieved only 1f leakage Into or through the bottom liner is
very small. Such a requirement cannot be achieved by a compacted soil
bottom liner as discussed In the EPA "Background Document on Bottom
Liner Performance In Double-Lined Landfills and Surface Impoundments"
(USEPA, 1987). Therefore, a composite liner should be used, and the
design specifications regarding composite bottom liners discussed In
the above referenced document are useful to ensure proper functioning
of the leak detection system. Bottom Hner specifications will not be
discussed hr -eafter. The design parameters discussed hereafter are
specific to the leak detection system.
2.8-4
-------
2.8.2.2.2 Review of Parameters Governing Detection Time
Accerdlng to the two-dimensional analytical study presented In
2.6.2, :t1on time depends on the following parameters which govern
steady-scate flow: hydraulic conductivity of the leak detection
system material, slope of the leak detection system, and distance
between the leak and the collector pipe or sump. Short detection
times result from high hydraulic conductivities, steep slopes, and
short distances between leaks and collector pipes.
Leak detection system materials with a high hydraulic conductivity
exhibit little capillary rise and, therefore, short Initial detection
time, which 1s an additional benefit.
2.8.2.2.3 Review of Parameters Governing Liquid Head
According to the study presented 1n Section 2.6.2.3.3, the liquid
head on the bottom liner is governed by the following parameters: the
hydraulic conductivity of the leak detection system material, the
slope of the detection system, and the distance between ths leak and
the collector pipe. Small heads are obtained if:
• hydraulic conduct!1, ty 1s large;
• slope 1s steep; and
• distance between leak and collector pipes Is small.
However small the liquid head 1s, it Is never zero and the leak
detection system thickness should be greater than the head,
2.8.2.3 Design Parameters tp_be Coo5!01?!6^ 1n_Spec1f1cat1ons
From the above discussion 1t appears that the relevant design
parameters are:
• the hydraulic conductivity of the leak detection system
material ;
2.8-5
-------
• the thickness and the slope of the leak detection system; and
• the distance between the leak and the collector pipe or su/rp.
From the above discussion it also appears that the leak detection
system will perform better (i.e., the detection time will be shorter
and the head smallcr) If:
• the hydraulic conductivity Is large;
• the slope 1s steep; and
• the distance between the leak and the collector pipe or sump 1s
short.
In addition, the leak detection system thickness must be large
enough to contain the head required by the flow.
The values that these parameters should have are discussed in the
following sections.
Thickness, D, of the leak detection system, and hydraulic
conductivity, kj, of the leak detection system material can be
combined to give the hydraulic transmisslvity, Sj (with 8^ « k,jD), of
the leak detection system. The hydraulic transmlss1v1ty can be used
as a design parameter for granular as well as synthetic drainage
layers. However, 1t 1s particularly convenient for synthetic drainage
layers for which hydraulic transmlssivlty 1s directly obtained from
tests.
2.8.3 Establishment of Design Specifications
2.8.3.1 Summary of Rejevant Data
2.8.3.1.1 Performance Criteria to Meet
From Section 2.7, the performance criteria to meet are as follows:
2.8-6
L
-------
] • retectlon sensitivity: 1 Mter/lOOCmVday (1 gpad): and
i
j • :2tectfon time: 1 day.
1
] 2.8.3.1.2 Relevant Technical Information for Design Specifications
- Available Technical Information
Available technical data which can be used to establish leak
detectict system design specifications can be summarized In four
tables:
• Table 2.8-1 gives detection times as a function of the distance
between the leak and the collector pipe or sump for various
hydraulic conductivities of leak detection system material.
This table (which Is a reproduction of Table 2.6-1 established
considering steady-state flow governed by Darcy's equation)
. Illustrates the influence of leak detection system material
| hydraulic conductivity on detection time.
1 i
• Table 2.8-2 gives the values of capillary nse in the leak
detection system as a function of the hydraulic conductivity of
the drainage medium used to construct the leak detection
; system. This 'able 1s a reproduction of Table 2.6-4.
i • Table 2.8-3 gives required hydrau':c transnissivities as a
I function of the leakage rate and the distance between the leak
and the collector pipes or sump. This table (v/hich Is a
i rep'oductlon of Table 2.6-2) has been established considering a
j steidy-state flow governed by Darcy's equation for several
j very large leakage rates. A leakage rate of 10 000
; llters/lOOOm'/day (10,000 gallons/acre/day) represents a rapid
;j and extremely large leakage (RLL) with a factor of safety.
• Table 2.8-4 gives the required thickness, D, for a leak
detection system as a function of the hydraulic conductivity,
ktf, cf the leak detection system material, the slope of the
leak detection system, and the distance, L, between the leak
2.8-7
L
-------
and the collector pips. This table (which is a reproduction of
Table 2.6-3) Iv.s been established considering steady-state florf
governed by Da y's equation.
-Additional Technical Information
As Indicated 1n Section 2.3.2.1.2, the hydraulic head on the
bottom liner should be as small as possible and, therefore, It Is
important that the characteristics of a leak detection system be such
that the hydraulic head on the bottom liner be as small as possible.
The hydraulic head on the bottom liner can be approximated using
Moore's equation [Moore, 1S83] which assumes that the leak is
uniformly distributed:
h - L (V(q/kd) + tan2 p - tan p] (Equation 2.8-1)
where: h « hydraulic head on top of the bottom liner; L * length of
the considered area impinged by a uniform leak (see Figure 2.6-1); q »
uniform leakage rate; k^ = leak detection system material hydraulic
conductivity; and P = slope of the leak detection system. Recommended
SI units are: h (m), L (m), q (m/s), and k^ (m/sl.
It should be pointed out that Moore's equation is related to
steady-state flow, which is consistent with the other technical data
related to leak detection system performance criteria.
Table 2.8-5 has been established using Moore's equation for a
leakage rate of 10,000 1 iters/1000m2/day (10,000 gallons/acre/day),
which represents a rapid and extremely large leak wi^h a factor of
safety. (A rapid and extremely large leakage may be c.i the order of
1,000-5,000 Ltd (gnad).)
2.8-8
-------
2.8.3.2 Rationale for the Specifications
- Hydraulic Conductivity and Length of Leak Detection System
Table 2.8-1 shows that a detection time equal to or less than one
ujy can be achieved only 1f the leak detection system hydraulic
conductivity 1s equal to or g--iter than 1C"1 m/s (1 cm/s).
In addition, Table 2.8-1 shows that the distance between the leak
and the collector pipe or sump should be less than 50 m (165 ft) In
order to meet the performance criterion of detection time less than
one day. To ensure that the distance between the leak and the
collector pipe or sump be less than 50 m (165 ft), the length of the
leak detection system should be smaller than 50 m (165 ft).
Table 2.8-2 shows that a drainage medium with a hydraulic
conductivity larger than 10"a m/s (1 cm/s) exhibits a small capillary
rise (I.e.. 30 mm (1.2 1n.)). Therefc-e, a leak detection system
constructed with such a material will not retain excessively large
amounts of leakage by capillarity. This confirms that 10"2 m/s (1
cm/s) should be considered as a minimum for the hydraulic conductivity
of the leak detection system.
It is Important to remember that Table 2.8-1 is for a bottom slope
of 2%. For a detection time of one day, a longer drainage path is
possible if a steeper bottom slope is incorporated into the design.
Alternatively, for a given length of drainage path, the detection time
can be decreased b.y steepening the bottom slope.
- Hydraulic Transmlsslvity
The required hydraulic transmissivlty of a leak detection system
depends on the considered leakage rate. As discussed before, a
leakage rate on the order of 10,000 Ltd (gpad) could La considered In
order to handle a rapid and extremely large leakage (RLL) with a
factor of safety. Table 2.8-3 shows that for such a leakage rate a
hydraulic transmissivlty of 1 x 10"4 mVs would be too small. A
hydraulic transmissivlty of 3 x 10"' m'/s is required for a 50 m
2.8-9
-------
(165 ft) distance between the leak and the collector r ie. For
increased safety, a value of 5 x 10'* m:/s is recc-rrended.
- Thickness
The hydraulic transmisslvi ty, 6
-------
of the leak detection system, which is not conservative, while
calculations performed In Section 2.6-2 and leading to Table 2.8-4
consider leakage concentrated at the top of the leak detection
system.)
- Slope
All design specifications discussed above were related to a leak
detection system with a 2% slope. Tables 2.8-4 and 2.8-5 show that a
4% slope would clearly result ,, smaller hydraulic heads on the bottom
Uner. These tables also show that hydraulic heads obtained with a 2%
slope are not excessive and, therefore, a 2% slope is theoretically
appropriate.
However, field experience shows that bottom liners which are
designed with a 21/. slope are often constructed with a slope that 1s
locally less than 2%. Also, settlement can locally reduce the slope
after waste has been placed. In well designed hazardous waste
management units, allowance is made for settlement and the 2'/. slope is
the "after-settlement slope". However, since design specifications
are mostly Intended as guidance to prevent mistakes in cases where
design is Insufficient and/or construction quality Is below standards,
a 4% slope design specification could be 'onsidered.
2.8.4 Presentation and Discussion of the Design Specifications
2.8.4.1 Presentation of the Design Specifications
The design specifications can be summarized as follows:
• Hydraulic conductivity of leak detection system material: k^ I
1Q~* m/s (1 cm/s),
• Hydraulic transmlssivlty of leak detection system (regardless
of type): Gj 1 5 x ^J"" m'/s.
2.8-11
-------
• Thickness of geosynthetic leak detection -ystem: 0 2 5 run (0.2
In.), which Implies a minimum hydraulic conductivity of 10"'
m/s (10 cm/s) for the geosynthetic material in order to meet
the above hydraulic transmissivity requirement.
• Thickness of granular leak detection system: D }.3 m (1 ft).
• Slope of leak detection system (including pipes): tan 3 2 0.02
(2%), with a recommendation that a 4'/. slope be considered.
• Length of leak detection sys TI: L i 50 m (165 ft).
2.d.4.2 Comments
- Comments on Hydraulic Transmlsslvlty
It 1s essential that the hydraulic transmlsslvity (and the two
related properties, hydraulic conductivity and thickness) of the leak
detection system be evaluated by tests conducted with boundary
conditions (such as compressive stress and materials In contact)
representative of conditions at the considered site. This
recommendation 1s so important that it should not be considered as a
simple comment but should be Included in *"he specifications.
- Comments on Slope
As Indicated In Section 2.8.4.1, a 27. slope is theoretically
appropriate, while a 4'/. slope would be advisable 1n some hazardous
waste management units to compensate for Insufficient design regarding
settlement and/or poor grade control during construction. To select
between 2% and 4'/. the following strategy may be considered: the
specified slope would be 4V. unless: (i) the owner or operator can
denonstrate that the design slope 1s such that after settlement the
actual slope will be at least 27. everywhere; and (11) tne quality
assurance plan Includes a detailed survey to ensure that the bottom
Uner will be constructed with the design slope.
2.8-12
-------
Table 7.8-1. Leak detection time as a function of the istance, L,
between the leak and the collector pip» or sump, and the
hydraulic conductivity of the leak detection system
rr erial. This table has been established assuming
steady-state flow in ..the leak detection system. The
slope of the leak detection system is 27,. This table is
reproduced from Table 2.6-1 in Section 2.6.
L
15 m
(50 ft)
30 m
(100 ft)
50 m
(165 ft)
60 m
(200 ft)
100 m
(330 ft)
Hydraulic conductivity of the
leak detection system material , k^
10~« m/s
{10"1 cm/s
26
52
87
104
174
10~' m/s
(10~' cm/s)
3
5
9
11
18
10"J m/s
(1 cm/s)
0.3
0.5
0.9
1.1
2
10"' m/s
(10 cm/s)
0.03
0.05
0.01
0.1
0.2
Leak detection times in days
2.8-13
-------
Table 2.8-2. Values of capillary rises as a function of me hydraulic
conductivity of the drainage medium used in the leik
detection system. This table is reproduced frcni Table
2.6-4 in Section 2.6.
Hydraulic Conductivity
kd
(m/s) (cm/s)
(m)
Capillarv Rise
(mm)
(in.)
1
5
1
5
1
5
1
5
1
5
1
5
x 10"*
x 10"'
x 10"'
X 10"s
x 10~4
x 10~-
x 10"'
x 10"'
x 10~2
x 10~2
x 10"'
x 10"'
1
5
1 x 10""
5 x 10""
1 x 10"'
5 x 10"'
1 x 10"'
5 x 10~2
1 x 10"'
5 x 10"'
1
5
10
50
100
500
3.10
J.39
0.93
0.44
0.31
0.14
0.10
0.044
0.031
0.014
0.010
0.004
0.003
0.0014
3100
1386
980.3
438.4
310.0
138.6
98.0
43.8
31.0
13.9
9.8
4.4
3.1
1.4
122
54.6
38.6
17.3
12.2
5.5
3.9
1.7
1.2
0.55
0.39
0.17
0.12
0.055
2.8-14
-------
Table 2.8-3. Required hydraulic transrnissivity, 8d. for the leak
detection system (1n m'/s) as a function of leakage
rate, q, and distance, L, between leak and collector
pipe for a leak detection system slope of 27.. This
table is reproduced from Table 2.6-2 in Section 2.6.
Leakage rate per unit area, q
liters/lOOOm'/day (gallons/acre/day)
1000
10 000
100 000
15 m
(50 ft)
1.2 x 10"*
1.2 x 10""
1.2 x 10~J
30 m
(100 ft)
50 m
(165 ft)
1.8 x 10"'
1.8 x 10 «
2.9 x 10"'
2.9 x 10""
1.8 x 10"1
2.9 x 10"'
60 m
(200 ft)
100 m
(330 ft)
3.6 x 10"'
3.6 x 10""
5.8 x 10"'
5.8 x 10""
3.6 x 10"'
5.8 '. 10"
Required hydraulic transmissivity for the
leak detection system, 8,j, 1n mj/s
2.8-15
-------
Table 2.8-4. Required thickness, D, for a leak detection system as a
function of hydraulic conductivity, kj, of the leak
detection system material, the slope of the leak-
detection system, and the distance, L, between the leak
-id the collector pipe. These results are related to a
leakage rate of 10,000 1 iters/lOOOm'/day (10,000
gallons/acre/day). (Note: this table is a reproduction
of Table 2.6-3 from Section 2.6.)
Slope
27.
47.
_
15 m
(50 ft)
60 m
(200 ft)
100 m
(330 ft)
15 m
(50 ft)
60 m
(200 ft)
100 m
(330 ft)
Hydraulic conductivity, kj
0.01 m/s
(1 cm/s)
9 mm
(0.3 1n.)
35 mm
(1.3 in.)
58 Km
(2.5 in.)
4 mm
(0.17 in.)
17 mm
(0.64 1n.)
29 run
(1.3 in.)
0.1 m/s
(10 cm/s)
0.9 mm
(0.03 in.)
3.5 mm
(0.13 in.)
5.8 mm
(0.25 ;n.)
0.4 nm
(0.017 tn.)
1.7 mm
(0.064 in.)
2.9 mm
(0.13 in.)
1 m/s
(100 crr/s)
0.1 mm
(0.003 in.)
0.3 mm
(0.013 in.)
0.6 mm
(0.025 in.)
0.04 PTT1
(0.0017 in.)
0.17 mm
(0.006 in.)
0.3 mm
(0.013 in.)
2.8-16
-------
Table 2.8-5 Liquid head on the bottom liner as a function of the leak
detection system hydraulic conductivity, kj, the slcce.
and the distance, L, between the leak and the collector
pipe or sunp. This table was established using Moore's
equation [Moore, 1983] for a leakage rate of 10,000 Ltd
(gpad).
Slope
2%
4%
L
15 m
(50 ft)
30 m
(100 ft)
60 m
(200 ft)
100 m
(330 ft)
15 m
(33 ft)
30 m
(100 ft)
60 m
(200 ft)
100 m .
(400 ft)
Hydraulic conductivity, k^
0.01 m/s
(1 cm/3)
4.3 mm
(0.17 in.)
8.6 mm
(0.3 in.)
17 nrn
(0.7 in.)
29 mm
i in.)
2.1 mm
(0.08 in.)
4.3 nrn
(0.2 in.)
8.7 mm
(0.3 in.)
14 mm
(0.6 In.)
0.1 m/s
(10 cm/s)
0.43 mm
(0.017 in.)
0.86 mm
(0.03 in.)
1.7 mm
(0.07 in.)
2.9 mm
(0.1 in.)
0.21 mm
(0.008 in.)
0.43 mm
(0.02 in.)
0.87 mm
(0.03 in.)
1.4 mm
(0.06 in.)
1 m/s
(100 cm/s)
.043 mm
(0.0017 in.)
.086 mm
(0.003 in.)
.017 mm
(0.007 in.)
0.29 mm
(0.01 in.)
0.021 mm
(0.0008 in.)
0.043 mm
(0.002 in.)
0.087 rrm
(0.003 in.)
0.14 mm
(0.006 in.)
Liquid head on the bottom liner
2.8-17
-------
2.9 ACTION LEAKAGE RATE (ALR)
2.9.1 Introduction
2.9.1.1 Purpose of the Section
The Action Leakage Rate (ALR) constitutes a trigger for initiating
Interactions between fie owner or operator of a land disposal unit and
EPA. Unlike the detection capability criteria described In Sections
2.7 and 2.8 of this document (which require the owner or operator to
carry out a demonstration), the ALR is a standard that is compared to
the leakage rates that the owner or operator measures (as part of a
leak detection monitoring program) at the leak detection system su-np.
If the measured leakage rate exceeds the ALR (as described
subsequently under monitoring requirements) the owner or operator will
enter a response action mode and will be required to initiate
interactions with the EPA Regional Administrator. In the proposed
Liner/Leak Detection Rule owners and operators may elect to use a
standard value of ALR defined by EPA or they may submit to the EPA for
approval a site-specific ALR.
The purpose of this section is to review the technical information
pertinent to the Action Leakage Rate. This information is used as a
basis to support the rationale given in the proposed Liner/Leak
Detection Rule for the establishment of the standard Action Leakage
Rate.
2.9.'..2 Organjzat1on_of the_5ection
This section Includes three parts:
• the first part presents an overview of the concepts of the
proposed rule regarding the Action Leakage Rate (ALR);
• the second part presents the available technical data that
support the rationale for the establishment of the standard
Action Leakage Rate; and
2.9-1
-------
• the third part presents suggested leachate monitoring
requirements.
2.9.2 Overview of the Concepts, of the Proposed Rule
2.9.2.1 The Concept_of_a_Tr1gger
The Action Leakage Rate triggers interactions between the owner or
operator and the Regional Administrator (RA) and requires the owner or
operator to submit a Response Action Plan (RAP) for leakage rates less
than rapid and extremely large (RLL) or to Immediately Implement the
RAP for leakage rates exceeding the RLL. This approach takes into
account the characteristics of the double-liner system, 1s consistent
with the capabilities of current technology, allows the owner or
operator flexibility, and uses leachate volume rather than hazardous
constituent concentrations as a trigger (except for land treatment
units).
In developing the concept of a trigger for interactions between
the owner or operator and the EPA, two different types of triggers
could be considered: a hydraulic trigger based on the rate of leakage
through the top Uner, or a leakage quality trigger based on the
concentrations of hazardous constituents 1n the leakage. It is
believed that for a trigger mechanism, a hydraulic criterion has
several distinct advantages over a leakage quality criterion.
However, once the ALR is exceeded, hazardous constituent monitoring
should be required as part of the response action activities. The
advantages of a hydraulic trigger over a leakage quality trigger
Include:
• a measurement of leakage rate 1s more Indicative of the
magnitude and severity of a top liner breach than is a
measurement of leachate quality;
• changes 1n leakage rate over time are indicative of progressive
changes in the condition of the top liner; a knowledge that
progressive changes are taking place 1s critical to selection
of the appropriate response actions;
2.9-2
-------
• the day-to-day monitoring program for a leakage rate trigger is
fast, relatively Inexpensive and can be conducted by
maintenance personnel using relatively unsophisticated
equipment; the day-to-day monitoring progra for leakage
quality would require complex, expensive chemical analyses, is
more time consuming, and requires more highly trained
personnel.
2.9.2.2 The Concept of Action Leakage_Rate
The action leakage rate (ALR) logically extends EPA's systems
approach to Its liquids management strategy. An ALR not only
Initiates an Interaction between an owner or operator, but also Is a
mechanism for the EPA to evaluate the leak detection program on a
site-specific basis. The ALR and the Response Action Plan (discussed
1n Section 2.10) are key elements of an overall containment system.
They function In an Integrated Interdependent manner with: (1) the
top and bottom liners; (2) the leachate collection and removal system
above the top Uner (for landfills); (3) the leak detection,
collection and removal system; and, where applicable, (4) the final
unit cover.
The total system ?ch1eves the objective of preventing hazardous
constituent migration from the unit by maximizing leachate collection
a.id removal. The total system falls only if there is a fatal
combination of failures among system components and 1f response
actions triggered by the ALR are Inappropriate or inadequate. The
probability of a fatal combination of component failures 1s believed
to be extremely low.
2.9-3
-------
2.9.3 Technical Support for the AIR
2.9.3.1 Introduction
- Purpose of the Section
The purpose of this section 1s to provide technical support for
the Issues of the proposed Liner/Leak Detection Rule related to the
Action Leakage Rate (ALR).
- Organization of the Section
Section 2.9.3.2 reviews the available Information. Then, the
rationale for the Action Leakage Rate 1s provided In Section 2.9.3.3.
2.9.3.2 RelevantTechn1cal
The ALR Is based on the technical capability of the top liner to
prevent leakage through the top liners at properly designed and very
well constructed units. The ALR therefore represents the best
demonstrated available technology (BOAT) for top 11 -s. (Note that
the ALR is established Independent of other possible sources of liquid
1n the leak detection system such as rainwater entrapped during
construction or water expelled by consolidation from the soil
component of the top composite Uner. The way to account for these
other sources of liquid 1s through an owner/operator demonstration as
part of the response action plan.) The relevant technical
information to use 1n establishing the ALR is the analyses presented
1n Section 2.2 on leakage through top liners and the information in
Section 2.5 on existing land disposal units.
The technical Information rrom Section 2.2 on leakage rates
through top liners Is summarized 1n Table 2.9-1. For the case of
composite liners only the condition for good FML-compacted soil
contact 1s shown since this represents the best demonstrated available
technology (BOAT) for composite top liners. Table 2.9-1 gives leakage
rates through top liners as a function of the hydraulic head on the
2.9-4
L
-------
liner. The three heads used in Table 2.9-1 represent the following
condltlons:
• 0.03 m (0.1 ft) Is assumed to be the average head acting on the
top liner of a landfill or waste pile with a well designed and
constructed leachate collection and removal system (LCRS).
• 0.3 m (1 ft) 1s assumed to be maximum head acting on the top
liner of a landfill or waste pile with a well designed and
constructed leachate collection and removal system (LCRS).
• 3 m (10 ft) 1s assumed to be the maximum head acting on the top
liner of a surface Impoundment.
2.9.3.3 Batl°Dai?_for tte Action Leakage_Rate_Value
The owner or operator has the option in the proposed Liner/Leak
Detection Rule of using an EPA-spedf1ed ALR which 1s generic and
applicable to all land disposal units, or aU°rnatively to use a site-
specific ALR obtained after a site-specific demonstration. The site-
specific ALR is Intended to provide the owner or operator with a
mechanism to account for conditions that reduce the potential for
migration of hazardous constituents through the top liner such as the
thickness of the top Uner and Us capacity to attenuate the specific
hazardous constituents in the leachate at that site. The generic ALR,
on the other hand, 1s based strictly on the best demonstrated
available technology (BOAT) for hydraulic containment by the top
liner. This section of the report presents the rationale for EPA's
selection of a generic ALR in the range of 5-20 Ltd (gpad).
2.9-5
-------
- Leakage Due to Permeation Versus Holes
The ALR is based on best demonstrated available technology (BOAT)
for leakage through top liners. This leakage can be due to two
sources: permeation through the FML and leakage through defects
(holes) in ' FML. Since permeation will occur through all FMLs and
1s not indicative of a breach or defect in the top liner, it is not
necessary to account for this source of liquid in the ALR
determination. The permeation values presented in this report must be
considered preliminary due to the limited data upon which they are
based. Continuing research is providing a better understanding of
permeation of liquids through FMLs. Once additional definitive
permeation data is available, EPA may wish to consider allowing the
owner or operator to subtract liquid associated with permeation
through the FML when determining whether the liquid collected in the
unit sump has exceeded the ALR. In conclusion, the ALR should be
based solely on i.akage through defects (holes) in the top liner.
- Selection of a Value for ALR
Inspection of Table 2.9-1 shows that the leakage rate through a
FML top liner will be dependent on the presence of FML defects and on
the hydraulic head acting on the top liner. Best demonstrated
available technology (BOAT) for FML top liners implies very high
quality construction and intensive quality assurance. Therefore, BOAT
should be based on an FML with, at most, only a small defect (rather
than a standard defect which might be considered for design
calculations). In fact, based on the very best installation
techniques and very intensive construction quality assurance, holes
smaller than 2 mm (0.08 in.) diameter occurring at frequencies less
than 1 hole per 4,000 ma (acre) may be considered. If a range of one-
1 mm (0.04 in.) diameter hole per 8,000 m* (2 acres) to one-2 mm (0.8
in.) diameter hole per 4,000 m2 (1 acre) is considered to represent
the range of BOAT for FML top liners, then the range for the ALR
(assuming a 0.03 m (0.1 ft) hydraulic head on the top liner) would be
from about 5 Ltd (gpad) to 30 Ltd (gpad). Choosing a value from the
2.9-6
-------
lower portion of this range, an ALR of 5 to 20 Ltd (gpad) may be
reasonable.
The range of ALR calculated above is based on a hydraulic head cf
0.03 m (0.1 ft). The reasonableness of this value for hydraulic head
must be considered. In landfills and waste piles, the average
hydraulic head on the top liner is expected to be about 0.03 m (0.1
ft) or less. The depth of liquid in the LCRS above "°e top liner will
periodically be higher, due primarily to precipitation, and this will
increase top liner leakage rates temporarily. However, the best way
to account for these temporary increases is not to increase the ALR.
The best way to account for these temporary increases is by using a
time-weighted averaging procedure to process results from the
monitoring program. The time-averaging process could be used to
smooth out temporary increases in leakage rate due to precipitation so
that a unit would not be found to be "leaking" at a rate exceeding the
ALR each time a major storm occurred.
As can be deduced from Table 2..-1, the range of ALR mentioned
above is not appropriate for an FML top liner with holes in a surface
impoundment with a 3 m (10 ft) depth. If the ALR were to be set at a
level corresponding to an FML with holes, an ALR value in the range of
100 to 500 Ltd (gpad) would need to be considered in a surface
impoundment. Clearly, alleging this much liquid to enter the leak
detection system before initiating interactions between the
owner/operator and EPA (to at least assess the consequences of this
level of leakage) is undesirable because it would permit the buildup
of high heads on the bottom liner.
However, the frequency of holes in operating surface impoundments
should be smaller than the frequency of holes in landfills and waste
piles. The reasons for this include: (i) hazardous waste surface
impoundment units tend to be smaller than landfills and aste piles;
(1i) there 'Is less overburden pressure or construction equipment
operating on top of liners in surface impoundments than in landfills
and waste piles; (iii) because they are small and the consequences of
holes are large, the CQA program for surface impoundments will often
involve steps to identify and repair any holes (such as ponding tests
2.9-7
far -•
-------
and electrical resistivity surveys); and (iv) if a hole should occur
during operation, repair or retrofitting of a surface impoundment top
liner is often feasible. For the above reasons, it can be tentatively
concluded that the frequency of holes in .op liners of surface
Impoundments designed and constructed to BOAT standards can be less
than the frequency at landfills and waste piles and may even approach
zero defects (the "no holes" case 1n Table 2.9-1). This observation
is consistent with the data on surface impoundments presented in
Section 2.5. Based on the preliminary information presented in
Section 2.5 for the ICWM surface impoundment (Section 2.5.2), the
surface impoundment in East Central U.S. (Section 2.5.5.1) and the
surface impoundments in South West U.S. (Section 2.5.5.2) it appears
that top liner leakage rates of 5 to 20 Ltd (gpad) or less can be
achieved and that the suggested ALR range represents BOAT for top
liners at surface impoundments.
2.9.3.4 Monitoring Requirements
- Recommended Monitoring Interval
In order to determine whether the ALR has been exceeded (as well
as to maintain a minimum hydraulic head in the leak detection system
sump), the owner or operator should be required to institute a regular
monitoring program at the leak detection system sump. The monitoring
program will simply consist of taking regular measurements and keeping
detailed records of the amount of liquid removed from the leak
detection system sump.
Owners and operators should be encouraged to monitor the leak
detection system as frequently as possible and as regularly as
possible. Based on the leak detection system capabilities presented
in Sections 2.7 and 2.8 a one-day monitoring interval is suggested
during the active life and closure period of a land disposal unit. A
longer monitoring interval during the post-closure care period is
clearly a'-ceptab1e. The recommendation for a one-day monitoring
interval during the active life and closure period is based on three
factors:
2.9-8
L
-------
• The recommended maximal leak detection time for land disposal
units is one day; the monitoring interval should be similar to
the detection time in order to derive the full benefit (i.e.,
rapid leak detection) of this detection capability.
• Frequent monitoring allows the liquid level in the LDCRS sump
to be kept to a minimum, which is essential to minimize leakage
through that portion of the bottom liner underlying the sump.
• In the event of a top liner leakage rate in excess of the rapid
and extremely large leakage rate (defined in Section 2.10), a
short monitoring interval will ;,,inimize the length of time that
the bottom Uner of the land disposal unit could be subjected
to significant hydraulic heads. In a one-day interval the
leakage through a composite bottom liner subjected to 3.0 m
(10 ft) of head (representing the worst case of a catastrophic
failure of the top liner in a surface impoundment) would be
small and within environmentally acceptable limits.
- ALR Determination
Having obtained daily leakage rate data from the monitoring of the
LDCRS sump, the owner or operator must determine if the .R has been
exceeded. As noted previously, the daily leakage into the LDCRS will
vary somewhat from day to day. Recent observations from active units
Indicate that even in the absence of precipitation leakage rates can
easily vary by 10 to 20% or more. Much larger variations can be
associated with major precipitation events. Due to these variations,
1t is appropriate to use time-weighted averages to determine if the
ALR has been exceeded. Further, the period for conducting the time-
weighted average should consider the maximum leakage rate measured.
If a very high leakage rate is measured, the duration for the time-
weighted average should be short so that the owner or operator moves
quickly to begin Interactions with EPA. EPA is recommending the
following procedures to determine 1f the ALR has been exceeded (based
on a ALR of 5 to 20 Ltd (gpad)):
2.9-9
L
-------
• if no leakage rate measurement exceeds 50 Ltd (gpad) during the
active life, average the measurements over 30 days to determine
if the ALR has been exceeded;
• if no leakage rate measurement exceeds 3bO 1iters/1000m2/week
(gallons/acre/week) during the post-closure care period,
average the measurements over 7 days to determine if the ALR
has been exceeded; and
• if any leakage rate measurement exceeds 50 Ltd (gpad) during
the active life, consider that the ALR has been exceeded.
These time-averaging procedures should be conducted on a daily,
forward-roll ing basis.
2.9-10
-------
fable 2.9-1. Leakage rates through top liners. The small hole has a
diameter of- 2 mm (0.08 1n.). The standard hole has a
surface area of 1 cm2 (0.16 in.*). The frequency of
holes Is 1 per 4000m' (1 per acre). The thickness of the
compacted sc laye_r 1s 0.9 m (3 ft) and its hydraulic
conductivity ,s 10 ' m/s (10"' cm/s). Good contact
refers to the quality of the contact between the FML and
the underlying low-permeability soil, which governs the
flow between the FML and the low-permeability soil and,
therefore, has a large Influence on the leakage rate.
Note: Ltd - Uter/lOOGmVday; gpad = gallons/acre/day; 1
Ltd - 1.1 gpad. This table 1s adapted from results
presented In Section 2.2.
Type of
Liner
FML
alone
Composite
1 1ner
(good)
contact)
Composite
1 iner
(good)
contact)
Leakage
mechanism
Permeation
No hole
TOTAL
Permeation
Small hole
TOTAL
Permeation
Standard hole
TOTAL
Permeation
Small hole
TOTAL
Permeation
Standard hole
TOTAL
Hydraulic head, h
0.03 m
(0.1 ft)
0.001
C
0.001
0.001
30
T "•
0.001
3UO
30o
0.001
0.01
0.02
0.001
0.02
0.02
0.3 m
(1 ft)
0.1
0
0.1
0.1
100
100
0.1
1,000
1,000
0.1
0.1
0.2
0.1
0.2
0,3
3 m
(10 ft)
10
0
10
10
300
300
10
3,000
3,000
10
2
12
10
3
13
Values of leakage rate 1n Ltd or gpad
2.9-11
-------
2.10 RESPONSE ,flOH PLAN (RAP)
2.10.1 Introduction
2.10.1.1 Scope
2.10.1.1.1 The Response Action Plan (RAP)
As part of the implementation of EPA's leak detection standards, a
written Response Action Plan (RAP) Is proposed to be required of
owners or operators of land disposal units. The RAP 1s a vehicle for
site-specific response actions to be taken when leakage above the
Action Leakage Rate (ALR) is detected within the leak detection
system. The goal of the RAP is to prevent the migration of hazardous
constituents out of the unit by providing a mechanism to initiate
appropriate actions to mitigate the potential for such migration
should the LDCRS reveal the presence of leakage above the ALR.
EPA is proposing RAPs for all newly constructed landfills, surface
impoundments, and waste piles; for replacement landfill, surface
impoundment and waste pile units; and for landfill and surface
impoundment units required to have double liners after November 8,
1984, at both permitted and interim status facilities. RAPs are
proposed for two leakage rates: (1) Rapid and extremely Large Leakage
(RLL); and, (2) leakage rates less than the RLL but larger than the
ALR. The Action Leakage Rate (ALR) was defined in Section 2.9. The
RLL will be defined subsequently.
2.10.1.1.2 Technical Support
The purpose of this section is to present technical information
supporting the concept of Response Action Plan. Three areas have been
identified where technical support is required: elements of the RAP,
leakage bands, and sources of liquids other than leakage. These three
areas are defined below.
2.10-1
-------
- Technical Elements of the Response Action Plan
The first technical issue related to the RAP concerns the question
of what information should be Included in the RAP. The general answer
to this question is that the RAP should include all information which
would help the EPA, as well as the owner/operator, understand the
design, construction, and operation of the land disposal unit, as well
as the current performance of the unit and the anticipated future
performance. The quantities and quality of leachate collected in the
LDCRS should be documented. [Note: the quality of the leachate is
not required 1n making an ALR determination; however, once the ALR is
exceeded leachate quality must be evaluated.] The owner/operator
should use this Information In the RAP to develop a full range of
response activities to actual or anticipated leakage events. The
objective of any response activity 1s to ensure, to the extent
feasible with current technology, that hazardous constituent migration
out of the unit 1s prevented.
A more thorough discussion of the technical elements of the
Response Action Plan are presented in Section 2.10.2.
- Leakage Bands
Since leakage through the top Hner will fluctuate during a unit's
active life and post-closure care period, 1t Is logical to develop a
RAP that has a range of leakage bands and a response or set of
responses for each band. With a specific response tied to a leakage
band Instead of an Individual leakage rate, the leakage rate can
fluctuate over time without the need to implement a different rescpnse
for each small fluctuation. A leakage band refers specifically to a
range of top liner leakage rates. For example, the owner or operator
may recommend In a RAP the following leakage bands and response
actions for a unit with an ALR of 20 Ltd (gpad) and a RLL of 2500 Ltd
(gpad):
2.10-2
-------
Example of
Leakage Band;
Typical Scope
Responses
< 20 Ltd (gpad)
20 Ltd 'ipad)
20 - 250 Ltd (gpad)
250 - 2,500 Ltd (gpad)
> 2,500 Ltd (gpad)
pump collected leachate
notify Regional Administrator (ALR)
increased pumping and monitoring
of the leak detection system
several changes in operating
practices to reduce leakage to
lower leakage band
repair leak or close unit (or
part of unit)
Leakage bands are site-specific and the values given above should
only be considered simple examples. In this exa.-ple, 20 Ltd (gpad) is
the Action Leakage Rate (ALR) defined in Section 2.9, and 2,500 Ltd
(gpad) is the Rapid and extremely Large Leak (RLL) defined in Section
2.10.3.
Technical support for the Action Leakage Rate was provided in
Section 2.9. Section 2.10 provides technical support for the other
elements of the leakage bands.
- Sources of Liquids Other than Leakage
Liquids collected in the leak detection system sump may come from
sources other than leakage through the top liner, such as:
• rainwater, entrapped in the leak detection system during
construction, which drains as soon as pumping of the LDCRS
starts;
2.10-3
-------
• water contained in the low-permeability soil component, if any,
of the top liner, expelled as a result of expression of this
layer under the pressure of the waste (a phenomenon known as
consolidation); and
• ground water intruding into the leak detection system through
the bottom 1Iner.
If one or more of the above sources of w..ter are present at the
unit", the owner or operator has the opportunity to demonstrate that
the leakage rate measurements have been disturbed by sources of
liquids other than leakage through the top liner. If the Regional
Administrator approves the demonstration, a variance to the Response
Action Plan may be considered. Also, as noted in Section 2.9, the
owner or operator may demonstrate that some small component of the top
liner leakage 1s due to permeation through the FML. Usually, however,
leakage associated with permeation is insignificant compared to other
sources of 1iquid.
Technical support is presented in Section 2.10 regarding sources
of liquids other than leakage through the top liner.
2.10.1.2 Organ1zat1on_of_tne Section
Two issues requiring technical support have been identified above:
leakage bands and sources of liquids other than leakage. Accordingly:
• Section 2.10.2 is devoted to a discussion of the necessary
technical elements of the RAP;
• Section 2.10.3 is devoted to a technical discussion pertinent
to leakage bands; and
• Section 2.10.4 is devoted to an evaluation of sources of
liquids other than leakage.
2.10-4
-------
2.10.2 Technical Elemer.cs of the Response Action Plan
As previously noted, the RAP should include all information tri-it
would help EPA and owners/operators understand the design construction
and operation of the land disposal unit, as well as the current
performance of the unit and its anticipated future performance. The
quantities and quality of leachate collected in the LDCRS should be
documented. The RAP should then go on to use this information to
develop a full range of response activities to actual or anticipated
leakage events. To achieve this end, 1t would appear that the RAP
would need to contain, at a ~ nimum:
(1) a general description of the unit;
(2) a description of the hazardous constituents contained in the
unit;
(3) a description of the range of events that may potentially
cause leakage (and the anticipated leakage rate associated
with each event);
(4) a discussion of the factors that can affect the amount of
liquid entering the LDCRS;
(5) a description of the design or operational mechanisms that
can be used to prevent the migration of hazardous
constituents out of the unit; and
(6) an assessment of the effectiveness of a range of possible
response actions.
Each of these six technical elements of the RAP are briefly
discussed below.
2.10.2.1 Genera] Descr]pt]on_of Unit
The response action plan should include a general description of
the unit including whether at closure the wastes will be
2.10-5
-------
decontaminated In place, removed from the unit, or left in place. The
: site-specific Information should include, as a minimum, the type,
si;e, a'id location of the unit; the d e s 13 n of the unit including
• details of the lining system; the geographic and climatic setting; and
the operating history and practices at the unit including the age of
i the unit, planned unit active life, ongoing activities at the unit,
volume of wastes being stored or disposed, methods of waste placement,
• equipment used, Intermediate cover practices, and the closure plan.
i 2.10.2.2 Hazardpus_Const1tuent_Assessment
i •
j j The response action plan should include a general discussion of
ti the hazardous constituents contained in the unit. This discussion
i should include, at a minimum, a summary of the results of analyses
\ carried out as part of the site-specific wa:.te analysis plan (Sections
264.13(b) and 265.13(b)) as well as description of the physical
characteristics of the waste. Of particular importance is the
; chemical quality of the leachate collected in the LDCRS and the
i compatibility of the waste in the unit and the leachate In the LDCRS
• with the lining system components.
2.10.2.3 DescrlpUon of Events Causing Leakage
; The response action plan should include a discussion of all events
that may potentially cause leakage exceeding both the ALR (if
appropriate) and the RLL. These potential causes will be site-,
( design-, and operation-specific. In general, they may include
I operational accidents, design deficiencies identified subsequent to
j the start of unit operation (such as inadequate connections between
liners and liner penetrations such as pipes and manholes), unforeseen
Incompatible wastes, equipment damage or unforeseen site subgrade
: settlements.
2.10.2.4 Factors Influencing Liquid Quantities |n_the LDCRS
The response action plan should include a discussion of the
important factors that can affect the amount of liquid entering the
leachate col lee1 on and removal system between the liners. These
factors should include, but not be limited to, the size and type of
2.10-6
-------
r>
top liner breach, the potential for additional breaches in the future,
the amount of liquid head !n the leachate collection and removal
system above the top liner, the potential for leachate generation in
the unit due to the moisture content of the waste, the anticipated
amount and frequency of precipitation, and the potential for surface
water run-on. The potential for sources of liquid other than top
liner leakage should also be considered, Including liquids from
construction water, consolidation of any compacted soil component of
the top liner, or water due to ground-water infiltration.
2.10.2.5 Mechanisms Preventing_H1grat]cn Out_of_the_UnH
The response action plan should include a description of the
design and operational mechanisms that will prevent migration of
hazardous constituents out of the unit. These mechanisms should
consider the capabilities of the entire land disposal unit as well as
the capability of each individual unit component. Particular
attention should be given to: the condition of the composite bottom
Uner; the condition and operational capability of the leak detection
system between the top and bottom liners; the condition and
operational capabilities of the top liner and the leachate collection
and removal system above the top liner; the potential to repair or
retrofit the top liner 1f the RLL is exceeded; and the potential for
the use of intermediate covers and run-on controls to limit leachate
production potential in the unit.
2.10.2.6 Assessment of_Response Actions
Last, the response action plan should include a detailed
assessment describing the feasibility of each of a range of responses
for preventing hazardous constituent migration out of the unit. For
top liner leakage rates exceeding the RLL, the top liner leakage rate
f 'St be dramatically reduced 1n order to ensure, to the extent
..chnically feasible, that hazardous constituent migration out of the
unit will be prevented. It therefore appears that only a limited
number of response action options are available. These options are:
(1) the owner or operator terminates receipt of waste and closes
the unit (or part of the unit);
2.10-7
L
-------
(ii) the owner or -perator provides expeditious repair of the
leak(s) (or rcirofitting of the top liner); or
(ill) the owner or operator Institutes operational changes at the
unit that will reduce leakage into the space between the
liners so that leakage will be less than rapid and extremely
large.
In the case of leakage between the liners 1n excess of the ALR,
but less than the RLL, the owner or operator has additional response
action options. Therefore, the assessment in the RAP should include
the three options listed above, plus:
(iv) the owner or operator continues to remove and treat the
leakage with Increased ground-water monitoring activities;
(v) the owner or operator maintains current operating
procedures.
2.10.2.7 Sources_of_Informat1on_for the_RAP
In developing the site-specific information for the response
action plan, the owner or operator should evaluate the condition of
the liners by reviewing activities that have occurred at the unit from
the tiire of construction to the present. An analysis of the results
of a rigorous construction quality assurance (CQA) plan should provide
a good data base to assess the condition of the liners after
construction of the unit. Results of CQA testing will be particularly
valuable if key areas of the liner were tested hydraulically for
leaks.
Other Information that the owner or operator may us3 in
development of a RAP includes: (1) the Part B permit application for
the unit (for permitted units), (2) a review of operational records
practices during the active life, (3) leachate analysis to indicate
whether unanticipated waste constituents are present, (4) coupon
testing 1n the sump above the top liner of a landfill or waste pile or
1n the waste at a surface Impoundment to determine any chemical
2.10-8
-------
compatibility problems, and (5) an assessment of operating activities
that may have damaged the liner. A review of the double liner system
design can also reveal whether the design concept had any weaknesses
that could increase the probability of a liner breach. The evaluation
of the design will also indicate areas that include redundancy or
design concepts that will minimize leakage if a breach occurs. This
type of review of site-specific information may help isolate the
location and extent of damage to a liner and can provide information
showing that the breach is the result of a design, construction, or
operational activity.
2.10.3 Leakage Bands
2.10.3.1 Introduction
An example of leakage bands was given in Section 2.10.1.1.2. The
two main elements of a set of leakage bands are:
• the Action Leakage Rate (ALR), which is the lower boundary of
the set of leakage bands; and
• the Rapid and extremely Large Leakage (RLL), which is the upper
boundary.
Technical support for the determination of the Action Leakage Rate
has been provided 1n Section 2.9. Therefore, Section 2.10.2 will be
entirely devoted to providing technical support for the Rapid and
extremely Large Leakage (RLL).
2.10.3.2 Rapid and_Extreme]y Large Leakage
2.10.3.2.1 Discussion
As intended for use in the proposed Liner/Leak Detection Rule,
Rapid and extremely Large Leakage (RLL) is leakage that is equal to or
larger than the maximum design leakage rate that the LDCRS can remove
under gravity flow conditions (i.e., so that the fluid head on the
bottom liner does not exceed the thickness of the LDCRS drainage
medium, which is about 0.3 m (1 ft) for granular drainage materials
2.10-9
-------
and 5 mm (0.2 in.) for synthetic drainage materials). Leakage rates
in excess of the RLL significantly increase the potential for
migration of hazardous constituents into the bottom liner and out of
the unit and therefore should be avoided. Remediation for RLL may
include repair of the leak or closure of the unH.
2.10.3.2.2 Technical Support
- Scope
As a result of the above discussion, calculations should be
carried out to determine typical leakage rates which can generate a
0.3 m (1 ft) hydraulic head on the bottom liner.
- Evaluation of Hydraulic Head
According to Giroud and Bonaparte [1984], the hydraulic head in a
drainage layer, such as a leak detection system, with uniform flow is
given by:
h = (Qd/B)/(kd tan p) (Equation 2.10-1)
where: h = hydraulic head; Q^ = flow rate in the considered drainage
layer; B » width of the drainage layer perpendicular to the flow; Qj/B
- flow rate per unit width; k—
-------
(m 2), and Q - m'/s. With the units used in this document: q is in
Ltd or cjpad; N 1s in number of holes per acre; and Q is In liters per
day cr gillens per day,
Combining Equation 2.10-1 and Equation 2.10-2 gives:
h - [q/(N b)]/(kd tan p)
- q/(N b kd tan p)
(Eouation 2.10-3)
where: b - width of the wetted area (b is used instead of B to
prevent confusion with the width of the leak detection system).
If we consider the typical design case of one hole per 4,000 m!
(1 hole per acre), N = 1/4000 mz and Equation 2.10-3 becomes:
h » 4000 q/(b kd tan P) (Equation 2.10-4)
with: h (m/s), q (m/s), b (m), and kd (m/s), or:
h - 4.6 x 10"' q/(b kd tan P) (Equation 2.10-5)
with: h (m), q (Ltd), b (m), and kd (m/s).
For a 2% slope (tan p = 0.02) and a hydraulic conductivity of the
leak detection system, kd = 10~2 m/s (1 cm/s), Equation 2.10-5
becomes:
h = 2.3 x 10"" q/b
with: h (m), q (Ltd), and b (m).
(Equation 2.10-6)
Table 2.10-1, established using Equation 2.10-6, shows that a head
of 0.3 m (1 ft) is obtained for a width of the flow of 1.5 m (5 ft)
(which may be considered as a reasonable value and a leakage rate of
2000 Ltd (gpad)). The results in Table 2.10-1 illustrate that, In a
typical case, the Rapid and extremely Large Leakage (RLL) is on the
order of 2,000 Ltd (gpad).
2.10-11
L
-------
This example also shows that the 'culated head is dependent on
the width of flow, b. Unfortunately, the correct value for b is
unknown and more information will be required to develop guidelines
for the selection of this parameter.
2.10.4 Sources of Liquids other than Leakage
2.10.4.1 Introduction
2.10.4.1.1 Scope
Sources of liquids other than leakage in the LDCRS include:
• rainwater entrapped in the leak detection drainage layer during
construction, which will drain progressively by gravity (except
for that portion of the water held by capillarity);
• water present in the low-permeability soil component of a top
composite liner, which will be expelled when the soil component
compresses under pressures exerted by the waste; and
• ground water intruding into the leak detection system through
the bottom 1iner.
The purpose of this section is to evaluate the flow rate that can
be generated by these sources of water and to determine if this can
significantly disturb leakage rate monitoring.
2.10.4.1.2 Organization of the Section
The three sources of water are discussed successively:
• Entrapped rainwater in Section 2.10.4.2;
• Consolidation water 1n Section 2.10.4.3; and
• Ground water in Section 2.10.4.4.
2.10-12
-------
2.10.4.2 Rainwater tntrapped 1n_the Leak Detection System
In this section the rate of flow of water entrapped in the leak
detection drainage layer is estimated. It is assumed that all water
initially entrapped in the drainage layer is due to rainfall that
occurred during construction.
It is conservatively assumed that during construction the
rainwater collected 1s not removed from the sump, and that this
precludes free draining of the rainwater during construction. As a
result, all the rainfall during construction is entrapped in the leak
detection system. The volume of water i.ius entrapped is:
V = e A At (Equation 2.10-7)
where: V = volume of water entrapped in the leak detection drainage
layer; e = rainfall impingement rate; A = considered surface area of
leak detection system; and At = duration of rainfall. Recommended SI
units are V (m'), e (m/s), A (m*), and At (s).
The maximum time it takes the water to reach a collection pipe is
given by Darcy's equation:
n L
td = (Equation 2.10-8)
kd sin 0
where: t
-------
Q - V/t,< = e A At '<<-( sr £/. n) (Equation P.10-9)
who-fi Q = rite cf flow of viter c:;!i::>ng at the leak detection sirp
(m'/s), ard V, td, 5, A, At, '
-------
related to flow under steady-state conditions. Therefore, the above
conclusion means that rainwater entrapped during construction does not
affect the leak defection tine (i.e., the detection time assuming flow
in steady-state condition). It does, however, affect the initial
detection time. Rainwater entrapped during construction will Increase
the degree of saturation (Sr) of the LDCRS drainage medium, thereby
decreasing capillary stresses and thus the initial detection time.
2.10.4.3 Water Expelled_by Consolidation from Top Liner
2.10.4.3.1 Introduction
- Presentation of the Mechanism
The low-permeability compacted soil layer of a composite top liner
consolidates under the pressure exerted by the solid waste (for
landfill) or liquid waste (for surface impoundments). The pressure
exerted on the low-permeability compacted soil layer creates a buildup
of excess water pressure in the soil pores which will tend to
dissipate by drainage into the leak detection system.
Consolidation of the low-permeability compacted soil layer results
in settlement of the soil. Primary consolidation refers to the
compression of the soil which occurs as a result of pore water
pressure dissipation. Secondary consolidation refers to the
compression that occurs at constant effective stress as a result of
the rearrangement of soil structure, following the completion of
primary consolidation. This section addresses only the water expelled
by the low-permeability compacted soil layer during primary
consolidation.
- Assumptions Regarding Low-Permeability Compacted Soil Layer
The calculations performed below assume that:
• The low-permeability soil has been preconsolidated by
compaction equipment during construction; its consolidation
behavior is governed by the recompression curve (rather than
2.10-15
... ..,,^^s
-------
the virgin curve) when it is subjected to pressure exerted by
solid waste (for landfills) or liquid waste (for surface
impoundments).
• Consolidation of the low-permeability soil layer is one-
dimensional (which Is a legitimate assumption In the case of a
relatively tMn layer).
• Excess pore water pressure drains only downward and the maximum
distance of drainage through the low-permeability soil layer is
therefore the thickness of the soil layer.
• The low-permeability compacted soil layer is fully saturated
and the quantity of water expelled is equal to the volume
decrease experienced by the low-permeability compacted soil
layer as a result of settlement; this assumption is reasonable
because low-permeability soils are typically compacted at or
slightly above their optimum moisture content (which is close
to the saturation moisture content).
- Assumptions regarding Rate of Water Expulsion during Consolidation
Consolidation 1s assumed to commence after the unit is completely
filled. In reality the consolidation process commences well before
the unit is full.
The rate of water expelled from consolidation is assumed to be
equal to the total quantity of water expelled divided by the active
life of the unit (i.e. time it takes to fill the unit) plus time at
which most of the consolidation has taken place.
2.10.4.3.2 Analysis
- Total Settlement
The settlement of the low-permeability compacted soil layer may be
estimated based on parameters obtained in a consolidation test:
2.10-lfa
-------
s - H log (1 + o/o,) (Equation 2.10-10)
1 t e0
where: s = total settlement of the low-permeability compacted soil
layer; Cr - recompression index; e0 - void ratio; H - thickness of low
permeability soil layer; o = vertical pressure exerted on the surface
of the low permeability soil layer by the solid waste or liquid waste;
oc = existing pressure in the middle height of the low-permeability
compacted soil layer prior to placement of any waste or overlying
leachate collection system materials. Recommended 51 units are: s
(m), H (m), o (N/m'), o0 (N/mJ). Cr and ec are dimensionless.
In the absence of data, the reccinpression index may be estimated
from the compression index of a consolidation test virgin curve
[Navfac OM-7.1, 1982]:
1 1
— Cc < Cr < — Cc (Equation 2.10-11)
10 5
where Cr - recompression index (dimensionless); and Cc = compression
index (dimensionless).
The compression index, in turn, may be estimated from the low-
permeability compacted soil liquid limit [Navfac DM-7-1, 1982; Bowles,
1977]:
Cc = 0.009 (LL - 107.) (Equation 2.10-12)
where: Cc = compression index (dimensionless); and LL = liquid limit
The vertical pressure exerted on the low-permeability compacted
soil layer by the waste is:
o = yn (Equation 2.10-13)
2.10-17
-------
where: o = vertical pressure; Y =• total unit weight of waste; and h
height of waste. Recommended 51 units are: o (fi/m'), y C'/m1}. ancl
The pressure existing in the middle height of the low-permeability
compacted soil layer prior to any placement of waste and overlying
leachate collection system materials is :
o, - Y' i
-------
The tirre t, may be calculated as follows [Mavfac DM-7,1, 1982]:
Tv II2
(Equation 2.10-17)
where: Tv - consolidation time factor; H « distance of drainage
through the low-permeability compacted soil layer (i.e., thickness of
the low-permeability compacted soil layer); and cv = coefficient of
consolidation. Recommended SI units are: t, (s), H (m), and cv
(m'/s). Tv is dlmensicnless and is given in Figure 2.10-3. In order
to ensure that most of the consolidation has taken place, a value of
Tv of approximately one should be used, according to Figure 2.10-3.
- Rate of Water Expulsion
The rate of expulsion of water due to consolidation is the volume
of water expelled divided by the drainage time:
Q = V / t (Equation 2.10-18)
where: Q = rate of water expulsion; V = volume of water expelled; and
t = drainage time. Recommended SI units are: Q (m'/s), V (m' , and t
(s).
2.10.4.3.3 Review of Results
- Calculations
The following characteristics were considered for the low
permeability soil layer: thickness H = 0.60 m (2 ft); unit weight Y'
= 18.8 kN/m' (120 pcf); void ratio e0 = 1; and liquid limit LL = 507..
For the waste, a unit weight of 15.7 kN/m3 (100 pcf) was assumed. For
a liquid limit of 507., Equation 2.10-12 yields a compression index of
0.36. This value of the compression index is substituted into
Equation 2.10-11 which then yields a recompression index between 0.036
and 0.072.
2.10-19
rifrrViiiim-'llfiiinli
-------
Several cases pertinent to landfills were considered using the
following parameters in addition to those described previously:
• waste height between 3 m (10ft) and 20 m (100 ft);
• recompression index between 0.036 and 0.072; and
• area of 1CCC m2 (10,750 ft2).
Table 2.10-3 summarizes the total quantities of water expelled
from consolidation of the low permeability soil layer. Table 2.10-3
indicates that the total quantities of water varies from 10,000 to
40,000 liters/1000 m2 (10,000 to 40,000 gallons/acre).
The time to attain a given percentage of consolidation was
calculated by Equation 2.10-17 considering: H = 0.60 m (2 ft); cv =
4.6 x 10"' m2/s (0.4 ft'/day) (Figure 2.10-2); and Tv = consolidation
time factor, as a function of the percentage of consolidation (Figure
2.10-3). Figure 2.10-4 presents a curve showing the percentage of
consolidation versus the calculated time to attain it. Figure 2.10-4
indicates that for a 0.60 m (2 ft) thick low penr.eabil ity soil layer,
most of this consolidation (90 to 95 percent consolidation) may be
attained after 100 days following the completion of waste placement in
the unit.
Considering an active life of 3 years (1095 days) for the unit and
a time or 100 days for most of the consolidation to take place, the
drainage time 1s then 1195 days, according to Equation 2.10-16.
Table 2.10-4 presents the rates of water expulsion calculated with
Equation 2.10-18. Table 2.10-4 indicates that the rates of water
draining through the leak detection drainage layer as a result of low
permeability soil layer consolidation range from 8.5 to 34 liters/1000
m2/day (8.5 to 34 gallons/acre/day).
2.10-20
-------
- Conclusions
The rates of watc. expelled from ccusol ic-'at ion of the ' ;-
permeability compacted soil layer were estimated for typical landi.il
scenarios incorporating a composite top liner. The calculations were
carried out only for the primary consolidation of the low-permeability
compacted soil layer, assuming that the soil layer is fully saturated.
The calculations further considered a variety of other parameters
judged to be representative of typical low-permeability soils used in
composite top liners. The calculated rates of water expelled from
consolidation of the low-permeability compacted soil layer ranged
between:
• 8.5 liter/1000 m'/day (8.5 gallon/acre/day) for a waste height
of 3 m (10 ft) and a recompression index of 0.036; and
\
• 34 liter/1000 m'/day (34 gallon/acre/day) for a waste height of
30 m (100 ft) and a recompression index of 0.072.
These flow rates are on the order of leakage rates considered for
the Action Leakage Rate or more. Therefore, water expelled from a
composite top liner as a result of consolidation of the low-
permeability compacted soil component of the top liner is expected to
significantly disturb leakage rate measurements. Furthermore, this
disturbance is expected to affect leakage rate measurements for a long
period of time since consolidation is a slow mechanism.
2.10,4.4 Leakage Into a Land Disposal Unit Due to Ground Water
Leakage Into a land disposal unit will occur if the ground-water
table rises above the bottom part of the bottom FML thereby causing an
artesian condition with flow directed toward the inside of the land
disposal unit. The inward leakage rate can be simply estimated using
the bottom half of Table 2.2-16 in Section 2.2, which is related to
composite liners.
2.10-21
-------
n
M
It appears that inward leakage rates can be:
• on the order of n.3-10 Ltd (gpad) if the ground-water table is
approximately O.j m (1 ft) above the top of the bottom liner;
and
• on -e order of 10-50 Ltd (gpad) If the ground-water table is
app.jxlmately 3 m (10 ft) above the top of the bottom liner.
As with the rate of water expelled by soil consolidation, the rate
of ground water entering the LDCRS can be significant and In the range
of the Action Leakage Rate. Therefore, water from this source, 1f It
1s present, Is expected to significantly disturb leakage rate
measurements.
2.10.5 Conclusions
This sect'on has presented information on the selection of
technical elements of the RAP (2.10.2), the Rapid and extremely Large
Leakage (Section 2.10.3), and sources of liquid other than top liner
leakage (2.10.4).
2.10,5.1 Technkal Elements of the RAP
Section 2.10.2 presents a discussion of the minimum technical
elements that should be included in the Response Action Plan (RAP).
Six major elements were discussed, these being: (1) a general
description of the unit; (2) a description of the hazardous
constituents contained in the unit; (3) a description of all events
that may eventually cause leakage; (4) a discussion of the factors
that can effect the amount of leakage entering the LDCRS; (5) a
description of the design or operational mechanisms that can be used
to prevent the migration of hazardous constituents out of the unit;
and (6) an assessment of the effectiveness of a range of possible
response actions.
2.10-22
-------
MO.5.2 Rapid and e.'^reme^y Large_Leakage
Table 2.10-1 provides a preliminary assessment cf the expected
magnitude of the RLL. These results show that leakage rates on the
order of 2000 Ltd (gpad) appear to be a reasonable value. However, it
should be kept in mind that the evaluation of the RLL is a site-
specific determination and highly dependent upon the Assumed leak
size. As a result, the designer will require guidance for the
j selection of hole size and the width over which flow occurs.
, i
'j 2.10.5.3 Source of Ljquids_other than Leachate
• i ~~~ ~
I Section 2.10.3 showed that a significant amount of liquid can
;• enter the LDCRS from sources other than leakage through the top liner.
These sources include:
• Rainwater entrapped in the leak detection drainage layer during
construction, which will drain progressively by gravity. The
; amount of water which can be collected will be largely a
function of the drainage layer properties, however, the leakage
rate of a material with a hydraulic conductivity of 10"' m/s
(10~' cm/s) will be on the order of 10,000 Ltd (gpad). While
this Is a large quantity, it will occur over a short period of
time (a few days) and should not interfere with the long term
: monitoring of the LDCRS.
• Water present in the low-permeability soil component of a top
composite liner and expelled ^hen this soil layer compresses
under pressures exerted by the waste. Typical rates at which
the water can be expelled are on the order of 10 to 30 Ltd
(gpad) and can occur over several years.
• Ground water intruding into the leak detection system through
the bottom liner when the ground-water table rises above the
elevation of the bottom FML. The rate of leakage can be on the
order of 1 to 50 Ltd (gpad), depending upon the elevation of
the ground-water table. This leakage can occur indefinitely 1f
the water table remains high.
2.10-23
L
-------
Table 2.10-1. Hydraulic heads on the bottom liner calculated using
Equation 2.10-^ for large leaks flowing over a width b
of the leak detection system. Characteristics of the
leak detection system used to establish this table are:
hydraulic conductivity, k^ = 1CT1 m/s (1 cm/s); and
slope 0 = 2%.
b
1.0 m
(3.3 ft)
1.5 m
(5 ft)
2 m
(6.6 ft)
Leakage rate, In Ltd (gpad)
100
0.023 m
(0.08 ft)
0.015 m
(0.05 ft)
0.010 m
(0.03 ft)
1,000
0.23 m
(0.75 ft)
0.15 m
(0.5 ft)
0.10 m
(0.3 ft)
2,000
0.46 m
(1.5 ft)
0.31 m
(1 ft)
0.21 m
(0.7 ft)
10,000
2.3 m
(7.5 ft)
1.5 m
(5 ft)
1.1 m
(3.6 ft)
Hydraulic head on the bottom liner
2.10-24
-------
Table 2.10-2. Flow rate in 11ters/lCCCm'/day (Ltd) or gallons/acre/day
(cjpad) [1 Ltd - 1.1 gpid] of water entrapped in a leak
detection system as 01 function of hydraulic
conductivity, kjj, of the leak detection system material
and the drainage distance, L, to a collection pipe.
Hydraulic Conductivity, kj
10"4 m/s
(10~2 cm/s)
10"* m/s
(10"' cm/s)
10"' m/s
(1 cm/s)
10~' m/s
(10 cm/s)
1 m/s
(100 cm/s)
15 m
(50 ft)
60 m
(200 ft)
960
(26.1)
240
(105)
9,600
(2,61)
2,400
(10.5)
96,000
(0.26)
24,000
(1.05)
960,000
(0.026)
240,000
(0.105)
9,600.000
(0.003)
2,400,000
(0.020)
Values of flow rate In Ltd or gpad
(The values In parentheses are those
of drainage time in days)
2.10-25
-------
Table 2.'0-3. Total quantities of water expelled from consolidation of
low permeability soil layer, q, in 1 iters/1000mz or
gallons/acre, as a function of the waste height, h, and
the recompresslon Index, Cr.
0.036
0.072
Waste height, h
3 m
(10 ft)
30 m
(100 ft)
10,000
20,000
20,000
40,000
2.10-26
-------
Table 2.10-4. Rates of water expelled from consolidation of low
permeability soil layer, q, 1n 1 iters/1000m2/day or
gallons/acre/day, as a function of waste height, h, and
the recompression Index, Cr.
Waste height, h
0.036
0.072
3 m
(10 ft)
8.5
17
30 m
(100 ft)
17
34
2.10-27
-------
f \Ovl dirrcl (OH
\a\rJ?.t~ Qti'^e
dov'r,- ^IV t* 1 '
Figure 2.10-1.
Plan view of a leak detection system with a large
leak flowing over a width b.
2.10-28
-------
10*
e
u
ui
c^ «
2
o \
>
ii 2
8
.0-1
6
6
5
4
3
CCtFFICIENT OF CONSOUDAT K*l
VS LIQUID LIMIT
COMPLETELY FttMOLDED SAWPuES'
CY LIES BELOW THIS Uf'FtR |
UNDISTURBED SAMPLES'
Cy IN RANGE OF VIRGIN COWPflESStON
Cv IN RAJJCE Of RECOMPRESSION LIES ABOVE '
THIS LOWER LIMIT
1
' 0.7
. 0.5
O.S
0.2
O.I
0.07-
O.O5 •
O.OJ
OD07
.006;
20 4O 6O 80 OO ^2
LIQUID LIMIT (LL) (7°)
160
Figure 2.10-2.
Coefficient of consolidation, cv, as a function of
soil liquid limit, LL. [NAVFAC, 1982]
2.10-29
-------
\
8
4
o
J
o
8-.
o
UJ '
o
UJ
o
u
o
•<
«E
UJ
>
<
S
8
8
5-W4Y OaAINAGE
(ALL VALUES O)
O.I
CCNSCUPAT>ON WITH VtHTlCAL 0«AINA.,£
INSTANTANEOUS UJAOING
0.01 0.10
TIME FACTOR, Ty
W~'
WAY DRAIfiACC
1.00
K)
I tltt.
ONE WAY DRAINAGE
TWO WAY DRAINAGE
U|
DISTRIBUTION OF INITIAL
POKE PRESSURE
Figure 2.10-3. Consolidation time factor, Tv. [NAVFAC, 1982]
2.10-30
-------
"Time, tj
50
Percan'arjt- o/ cooSo/idauon ( / -)
Figure 2.10-4.
Time, t,, required to achieve various percentages of
consolidation.
2.10-31
-------
I
CHAPTER 3
EXTENSION OF
DOUBLE LINER SYSTEM REQUIREMENTS
\ V ,\
>v-
, \
-------
3.1 INTRODUCTION
j.1.1 Scope of Choptgr 3
The purpose of Chapter 3 is to provide a discussion of the
proposed extensions of t minimum technology double Uner system
requirements to waste pil-. , significant unused portions of existing
facilities, and certain units permitted prior to November 8, 1984.
This chapter provides EPA with a summary of the proposed changes to
Part 264 and Part 265 regulations, technical rationale for the
proposed changes, and regulatory issues and options.
3.1.1.1 Double Liners and LCRS ."or Waste Piles
EPA is proposing that six months after promulgation (the
"effective date") of the proposed Liner/Leak Detection Rule, owners
and operators must install double liners and leachate collection
systems at new waste pile units, lateral expansions, and replacement
units at both permitted and interim status facilities. As a result of
this proposed rule, the lining systems at designated waste piles will
have technological requirements equivalent to those of landfills and
surface impoundments. EPA believes that it is critical that waste
piles have equivalent lining systems because, as will be shown in this
chapter, the potential for leachate migration from a waste pile can be
similar to or greater than the potential for migration from a landfill
for an equivalent time perioa.
Under the proposed rule, owners or operators of permitted and
interim status waste pi s will be allowed to seek the same variances
as those allowed to owners and operators of landfills and surface
impoundments from the minimum technology requirements described under
Sections 3094(o) (2) and 3004(o) (3) of RCRA. Owners or operators of
totally enclosed waste piles that meet the requirements of Section
264.250(c) will remain exempt from these requirements.
3-1
-------
3.1.1.2 D?yb1e_1-iners_and LCR5 for Significant Unused Portions of
Cxisting_Landf il 1 s_,_Surf<3C9 Impoundments, and Waste Piles
In the proposed Liner/Leak Detection Rule, existing l.Mdfill,
surfaca impoundment, and waste pile units will be required tc install
double liners and a LCR5 between the liners on those portions of the
unit that are not defined as existing portions in Section 260.10, do
not have a liner system that meets the Part 264 single liner standard,
and meet the definition of a significant portion (which is defined
subset ently). EPA takes the oosition that double liners should be
installed at significant portions of existing units where the
opportunity to do so is the same as for new units. The installation
of double liner systems at significant portions of existing units
reduces the potential for adverse human health and environmental
impacts by preventing, to the extent feasible with current technology,
the migration of hazardous constituents out of the unit.
3.1.1.3 9°y^e_L1ners_and_LCRS_for Certain Land_Disposal_Units_at
Before November 8, 1984
New units, and lateral expansions and replacements rf existing
landfill, surface impoundment, and waste pile units at facilities
per cted before November 8, 1984, will be required to have double
liners and leachate collection and removal systems meeting EPA minimum
technology standards. This requirement of the proposed Liner/Leak
Detection Rule will apply to those units that begin construction 24
months after the date the final rule is published in le Federal
Register. The proposed rule, thcjgh not required by RCm, is
presented because the potential for migration of hazardous
constituents from these units is similar to the potential at units
permitted after November 8, 1984. Because units permitted after
November 8, 1984 are required to have double liners and leachate
collection systems, the Agency believes it is appropriate to require
units that are not yet constructed at facilities permitted before
November 1984 to also meet these requirements.
There is, however, an exception to the applicability of the
requirements discussed above. Under 40 CFR 264.221(f) and 264.Z54(f),
the Agency is proposing to exempt c-'tain replacement surface
impoundments and waste piles from the jouble liner and leachate
AGENCV p .
1445 ROSS AVENUE /
DALLAS, TEXAS 75202
-------
collection system requirements. In essence, owners or operators -..-o
demonstrate that th-?y havs a single liner at a surface impoundment, or
w-isto pile that cut ently nieets the Part 265 single liner re'ji.nr.3
and who have no reason to suspect that the liner is leaking .vili be
exempt from the double liner and Icachate col 1"ction system
requirements.
EPA takes the position that if the owner or operator made a good
fi'ith effort to satisfy single liner requirements in effect at the
time of permitting, it is unreasonable to require the owner or
operator to assume the expenses of a new double Hner system.
3.1.2 Organization of Chapter 3
Chapter 3 is comprised of four sections which are briefly
summarized below.
Section 3.1 is primarily devoted to a discussion of the scope of
the proposed extensions to the minimum technology double liner system
requi rements.
Section 3.2 is concerned with the extension of the minimum
technology double liner system requirements to waste piles. This
section addresses: description of waste p^les; background and issues;
comparative performance of waste piles, landfills and surface
impoundments; and rationale for proposing double liner system
standards.
Section 3.3 is concerned with extension of the minimum technology
double liner system requirements to significant portions of existing
facilities. This section addresses: definition of significant
portions; background and issues; performance of lining systems under
significant portions; and rationale for proposing double liner system
standards.
Section 3.4 is concerned with extension of fhe minimum technology
double liner system standards to new units, re. acements and lateral
expansions at facilities permitted prior to November 8, 1984. This
section addresses: scope of proposed rule changes background and
issues; and rationale for proposing double liner system standards.
3-3
-------
I", -.*••-*-
3.2 WASTE PILES
3.2.1 Descrlptlon of Waste Piles
Waste piles are defined as facilities that store or treat v.aste in
piles (40 CFR 264.250). A waste pile is typically, but not
necessarily, an above-ground facility. A typical waste pile is
illustrat . In Figure 3-1. A prerequisite for waste piles is that at
the end of the active ' fe of the waste pile, all waste residues are
either removed or decontaminated, as are all contaminated containment
system components, contaminated subsoils, and structures and equipment
contaminated with waste or leachate. If all wastes, waste residues,
and contaminated materials cannot be removed or decontaminated at the
end of the active life, the facility must be closed in accordance with
closure and post-closure care requirements that apply to landfills.
EPA has previously estimated the number of existing waste piles
containing hazardous waste at about 80 units. Of the 80, it is
believed that 8 were permitted and 72 are operating under interim
status. The exact number of hazardous waste piles currently permitted
or operating under interim status is not documented. The current
number is believed to be somewhat larger then 60 units. These waste
pile units are used in a variety of industrial applications and for a
variety of purposes. These include temporary storage of hazardous
waste, ore storage at mining facilities, and heap leach pads.
While the operating characteristics of waste piles varies from
waste pi. to waste pile, some generalized characteristics may be
defined:
• waste piles have long active lives (the active life of a waste
pile may be as long as the active life plus post-closure care
period of a landfi11);
• waste piles are usually not covered;
• waste piles are frequently active facilities; and
• lining systems underlying waste piles are more prone to damage
from heavy equipment use than are lining systems at landfills.
3-4
-------
3.2.2 Background
•10 CFR 264.251(a) current^ require; permitted vvaste piles to ;..r.c
a single liner that is designed, constructed and installed to prevent
any migration of leachate out of the waste pile and into the
surrounding environment during the active life (and the closure
period, If applicable) of the «aste pile. The liner may be
constructed of materials (such as low- ermeability soils) that allow
migraMon of leachate into the liner itself as long as the leachate
does not migrate into adjacent soil, ground water, or surface water.
A leachate collection and removal system that is designed,
constructed, maintained, and operated to collect and remove leachate
from the waste pile is required to be placed immediately above the
liner (40 CFR 264.251(a)). Owners or operators whose waste pile is
Inside or under a structure that provides protection from
precipitation, so that neithe un-off or leachate is generated, are
exempted from liner and ' ;hate collection and removal system
requirements provided that: xi) liquids or materials containing free
liquids are not place in the waste pile; (2) the wiste pile is
protected from surface water run-on by the structure or in some other
manner; (3) the waste pile is designed and operated to control the
dispersal of waste by wind, where necessary, by means other than
wetting; and, (4) the waste pile will not. generate leachate through
decomposition or other actions. For interim status units, with
respect to waste received after May 8, 1985, the owner or operator of
a waste pile is subject to the requirements for liners and leachate
collection systems under 40 CFR 264.251 with respect to each new unit,
replacement of existing unit, or lateral expansion of an existing unit
that is within the area identified in the Part A permit application.
The variance provisions under 264.251 are applicable to these units.
EPA is proposing to modify the current regulations to require
double liners and leachate collection and removal systems at new waste
pile units, lateral expansions, and replacement units. This action is
being proposed based on evidence (discussed subsequently) which
indicates that waste piles pose a potential threat to human health and
the environment similar to that posed by landfills.
3-5
-------
3.2.3
Rationale for Double Liner System Requirements
EPA's rationale for a double liner system requirement at v,-;ste
piles is based on the premise that landf'ils and waste piles pose
similar potential risk"; to human health and the environment an^ should
be regulated in such a way as to provide the public and env jnment
with equal levels of protection. The premise has been derived from:
(1) comparison of typical operating characteristics of landfills and
waste piles; (2) analytical calculations comparing the lining system
currently allowed in 40 CFR Part 264 with the proposed minimum
technology double liner system; and (3) comparison of numerical
simulatic of leachate generation and migration in landfills and
waste piles.
3.2.3.1 Operating Characteristics
Certain operating characteristics of waste piles may result in
conditions that increase the potential for leachate migration from the
waste pile unit beyond the levels of migration observed in landfills.
For instance, at some waste piles, the waste is periodically
removed and replaced with new waste. This moving process involves the
use of heavy equipment. The equipment has the potential to damage the
lining system through careless operation, construction accidents, or
insufficient protective cover above the lining system. Fhick
protective covers can, to some degree, mitigate the potential for
equipment related damage. However, in a typical facility, the
thickness of the protective cover will be limited to 0.3 m (1 ft.) to
0.6 m (2 ft.) and the potential for equipment related damage will be
present. In contrast, at landfills, waste is not removed above '•he
liner, and the liner is not exposed to equipment operation to the su.,ie
extent as in waste piles. Thus, the potential for equipment related
liner damage appears to be greater for waste piles than for landfills.
This potential cause of liner damage increases the probability that
hazardous constituents could migrate out of the waste pile.
Moreover, there are other factors which tend to indicate that
waste piles possess a potential for leachate generation and migration
equaling or exceeding the potential for migration from landfills.
Waste piles generally have a higher percentage of their waste area
exposed to precipitation than do landfills. In addition, waste is
3-6
-------
?p
generally exposed to precipitation for a longer period of time at
waste piles than at landfills. This is because landfill units are
periodically closed by placing a temporary or final cover c-'er the ir.-
pi.ice waste. The normal operating practice for landfills is to
minimize the time that waste is exposed to precipitation in order to
to minimize the generation of leachate. Also, as wastes are removed
from an unprotected waste pile and replaced with new waste, more
hazardous constituents become available for exposure to precipitation,
and therefore, more constituents are available for leaching from
precipitation in a waste pile than in a landfill, in addition, the
active life for a new landfill unit is typically 6 months to 5 years,
while waste piles may be used for storage for a much longer time
period, in some cases 20 years or more.
Because the potential for liquids to migrate" through a liner is
similar for landfills and unprotected waste piles, the same level of
protection of human health and environment should be provided at
landfills and unprotected waste piles. This equates to installing a
minimum technology double liner system at waste piles.
In general, waste piles are also equivalent to landfills with
respect to the types of hazardous constituents that will be accepted
by the owners and operators of these units. This fact can be seen in
the results of the 1986 "National Survey of Hazardous Wasts Treatment,
Storage, Disposal and Recycling Facilities", conducted by EPA in 1986
[USEPA, 1986d], Data from this survey are shown in Table 3.1.
Presented in the table are the number of landfill and waste p'le units
found by the survey to "treat, store, dispose, or recycle" each of
eight different classification of hazardous waste. While we were not
able to determine the total number of landfill and waste pile units
included in the survey, we have made estimates of these numbers in
order to compare -the percentages of landfill and waste pile units
accepting hazardous waste from each of the eight considered waste
categories. The estimated percentages are also shown in Table 3.1.
From inspection, it can be seen that similar percentages of landfills
and waste piles accept each of th> eight considered waste types. On
this basis, the conclusion is drawn that the owners and operators of
landfills and waste piles will, in general (there are, of course
exceptions), accept similar types of hazardous waste at their landfill'
and waste pile units. Therefore, with respect to the waste type
3-7
-------
contained at the unit, landfills and waste piles can be considered to
be roughly equivalent.
3.2.3.2 Analytical Calculations
Current 40 CFR Part 264 regulations for waste piles require a
single liner "that is designed, constructed and installed to prevent
any migration of wastes out of the le into the adjacent subsurface
soil or ground water or surface water at any time during the active
life (including the closure period of the waste pile)". The
regulations also require a "leachate collection and removal system
that is designed, constructed, maintained, and operated to collect and
remove leachate from the waste pile. The Regional Administrator will
specify design and operating conditions in the permit to ensure that
the leaci.ate depth over the liner doos not exceed 30 cm (one foot)".
The single liner requirement described above can be satisfied
using a compacted soil liner. The performance of this compacted soil
liner can be evaluated in terms of the potential for migration of
leachate into the bottom-most liner in the lining system as described
in the minimum technology double liner system ;n the proposed Double
Liner Rule of March 28, 1986 (51 FR 10706-10723). For leachate to
migrate out of a lining system and into the environment, it must pass
through the bottom-most liner in the lining system. Thus, an
evaluation of migration into the bottom-most liner provides an
indication of the potential for migration out of the unit. The
performance of compacted soil liners is evaluated below using a one-
dimensional, steady-state saturated flow analyses based on Darcy's
Equation (Equation 2.4-2). Theses results are then compared to the
results on leakage into and through FML and composite liners.
Using Darcy's Equation, the steady-state flow into a uniform layer
of saturated compacted soil can be calculated. If it is assumed that
the compacted soil has a hydraulic conductivity, kc, of 1 x 10 "' m/s
(1 x 10"' cm/s) the following results will be obtained for leakage
into (and through) the liner (given in units of Ltd (gpad)):
3-8
-------
Ccn-pacted soil
thickness, in (ft)
0.9 (3 ft)
1.5 (5 ft)
3.0 (10 ft)
llydraul ic he.
li=0.03 in (0.1 ft)
89
88
87
id on liner
h=0.3 m (1 ft)
112
103
95
From the auove table it
Into and through a compacted
kc = 1 x 10"' m/s (1 x 10~f
0.9 m (3 ft). The amount of
1iner wil1 be larger if the
permeable than kc = 1 x 10"
result, it can be concluded
in excess of 89 Ltd (gpad),
the compacted soil liner.
can be seen that the steady-st^te leakage
soil liner is at least 89 Ltd (gpad) for
cm/s) and a liner thickness of at least
leakage into and through a compacted soil
liner is thinner than 0.9 m (3 ft) or more
' m/s (1 x 10"' cm/s). From this simple
that for waste piles generating leachate
at least 89 Ltd (gpad) will migrate into
The value of leakage into a compacted soi' ner given above can
be compared to a inimum technology double linei .ystem comprised of a
FHL top liner and a composite bottom liner. (Note: A composite
bottom liner has been selected based on EPA's April 1987 Background
Document on "Bottom Liner Performance in Double-Lined Landfills and
Surface Impoundments" [USEPA, 1987] which showed that composite bottom
liners provide higher levels of leachate containment and improved
LDCRS leak detection sensitivities and leachate collection
efficiencies than compacted soil bottom liners.) Using the results
from Section 2.2 of this document, a properly designed top FML
installed with good construction quality assurance monitoring may have
on the order of one FML hole per acre. Depending on the hole size,
the leakage through the hole may be in the range of 30 to 300 Ltd
(gpad) for a hydraulic head on the top liner of 0.03 m (0.1 ft), as
shown in Table 2.2-16. Leakage through the top liner will be greatly
reduced if the top liner is a composite rather than a FML alone.
Furthermore, most of this liquid will be collected by the LDCRS. (A
small amount may migrate into the composite bottom liner.) Using
3-9
-------
Table 2.2-16 and a hydraulic head of O.C3 in (0.1 ft) on the bottom
liner (which Is very conservative) and assuming one "standaid" hole
per acre (with good contact), the est'!inted leakage into the hottc.n
liner hould not exceed 0.02 gpad.
By comparing the above res'ilts of a single compacted soil liner
and a minimum technology double liner system, it can be observed that
leakage Into the bottom liner (and potentially out of the unit) is on-
the order of 5000 times more for the single compacted soil liner than
for the minimum technology double liner system. (Note: The previous
calculation treated the double liner system very conservatively. A
less conservative calculation might show that the double liner system
reduces the leakage into a composite bottom liner by a factor of
several tens of thousands or more compared to the leakage into a
single compacted soil liner.)
An important question to consider in evaluating the equivalence of
landfills and waste piles is the leachate production potentials of the
two units. In general, it is believed that waste piles have longer
active lives (10 to 20 years) than landfills (1 to 5 years).
Therefore, waste piles will be open to proportionally more
precipitation than landfills (which are covered at the end of their
active lives). To illustrate this point the following scenario is
considered: a 4,000 m2 (i acre) landfill and a similar sized waste
pile are located in an i>rea receiving 0.4 m (15 in.) of rainfall
annually. The active life of the landfill is 2.5 years, while thb
active life of the waste pile is 15 years. It is assumed that in both
units only 25% of the precipitation impingi g the unit reaches the
LCRS above the top liner, with the rest going into surface-water
collectors, evaporation and field storage (257. may be realistic for
landfills and is probably conservative for waste piles). Based on the
above assumptions, at the end of its active life the LCRS in the
landfill win have intercepted 1,000 m3 (250,000 gallons) of leachate.
In contrast, at the end of its active life, the LCRS in the waste pile
will have intercepted 6,000 m' (1,500,000 gallons) of leachate. From
this very simple comparison, it is clear that a waste pile at a given
site has a leachate production potential at least equal to, and
probably greater than, the leachate production potential of a landfill
at the same site.
3-10
-------
3.2.3.3 Numerical Simulations
In 1903, EPA Offi'-.e cf Solid Wast« conducted a study entitled
"Evaluation of Land Di ..osal facility Technologies and Integration of
Waste/Environment/Technology Characteristics to Produce Facility
Profiles" [EPA, 1983b]. In thi? study, EPA evaluated the hydrologic
performance of landfills, surface impoundments, and waste piles using
the Hydrologic Evaluation of Landfill Perfc mance (HELP) computer
program [Schroeder, P.R., et al., 1984a, 1984b]. The HELP program is
a water-budget model that can be used to estimate the magnitudes of
various components of a water budget at a land disposal unit. For
instance, the HELP model can be used to estimate the volume of
leachate generated at any point in time at the base of a landfill or
waste pile. The model can also be used to estimate the migration of
leachate out of a land disposal unit. However, to carry out this last
calculation, simplifying assumptions are required regarding migration
through the lining system. Results from the HELP model can be used to
compare the leachate production potential of a range of simulated
landfill and waste pile facilities. Input to the HELP model includes
cl imatologic, unit (liners and leachate collection and removal
layers), soil, and waste data. The output fron the model can include
daily estimates of water and leachate movement into, through and out
of the land disposal unit.
EPA used the HELP model to evaluate leachate migration out o
landfills, surface Impoundments, and waste piles at three different
geographic locations (Hartrord, CT, New eans, LA, and Denve,r,. CO).
The rate of migration out of the units was investigated for a 100 year
unit life, which included each unit's active life, post-closure care
period, and post-care period. The rate of migration out of each unit
was investigated for all three types of units as a function of various
types of liner and cover systems (using both cl^.ys and FMLs).
Results from the study showed that the migration of leachate out
of a unit was controlled largely by the assumptions regarding the
performance of the FML components of the lining and cover systems. If
the FML was assumed to be intact and functioning properly, migration
out of the unit was zero. If, however, the FML was assumed to fail at
some point in time (e.g., 50 years), the migration of leachate out of
the unit would suddenly increase. It is clear that the results
3-11
-------
I, I,,,,,., i immurr. i J.i .11 II ...II . L _l[ 'I"'.""""."
presented are dependent 0:1 the assumptions about the perfornunce of
the r,'!L components of the lining and ccver s> stems. Onre the r,"i.
components failed and migration out of the unit was initiated, the
rate of migration was largely controlled by cl'matic conditions (i.e.,
the rate was proportional to the average annual rainfall), and by the
permeability of clay layers in the cover and lining systems.
While the absolute results from the EPA study should be used with
caution (since the assumptions about long-term FML liner performance
are, by necessity, somewhat arbitrary), they are useful for direct
comparisons of landfill and waste pile units having imilar lining
systems and operating characteristics, and exposed to similar
climatoiogic conditions. The results from these types of comparisons
showed thdt for a similar set of climatoiogic and lining system
assumptions, migration rates of liquid into the lining system are
similar for landfills and waste piles. Therefore, for a given set of
climatoiogic and unit operating conditions, equivalent lining systems
would be required at landfills and waste piles in ordei to provide
protection to human health and the environment.
3.2.4 Exemption for Totally Enclosed Waste Piles
Some waste piles are completely enclosed within structures a-id are
thereby protected from wind, rain and snow. If the waste placed in an
enclosed facility 1s dry the waste pile will not generate leachate
during its active life (assuming that the enclosure structure
continues to function as planned). Since the potential for migration
of leachate from the waste pile 1s non-existent, EPA is proposing to
exempt these units from minimum technology double liner system
requirements. This exemption is consistent with the existing EPA
exemption for enclosed waste pile (40 CFR 254.25C(c)).
For a waste pile unit to be eligible for an exemption undei 40 CFR
264.250(c), the unit must be totally enclosed and it must be shown
that there 1s no potential for the migration of leachate or hazardous
constituents into the surrounding environment. If an owner or
operator of a waste pile desires to be exempt from the double linei
and leachate collection and removal system requirements, the following
conditions under Section 264.250(c) must be met: (1) the waste pile
must be inside or under a structure that provides protection from
3-12
-------
precipitation so that neither run-off " leaclute 15 generated; (2)
liquids or materials containing free li,jids may not be placed in the
pile; (3) the pile is protected from surface water run-en !:y the
structuie or in some other manner; (4) the pile is desigp.e-J and
operated to control dispersal of the waste by wind, where necessary,
by means other than wetting; and (5) the pile will not generate
leachate through decomposition or other reactions. It is important
to recognize that the foregoing limitations require that the waste in
the waste pile have such a low water content that no free liquids will
be present, and further that no leachate will drain out of tha waste
pile at any time after placement. Therefore, totally enclosed waste
piles which contain moist waste or where liquids are added to the
waste do not qualify for the exemption. It is recognized that since
they are enclosed, waste piles with moist waste have a greatly
diminished capacity for leachate generation compared to waste piles
exposed to the environment with equally moist wastes. However, since
the active life and operating practices (frequency of waste
"turnover") of the waste pile are unrestricted, significant amounts of
leachat--1 can be generated within enclosed units or liquids can be
added to the waste. In addition, in an enclosed waste pile, no
restrictions exist on the height of leachate above the top liner other
than the existing 40 CFR 264.250(c) requirement to limit the hydraulic
head to no more than 30 cm (one foot). This level of liquids above
the top liner represents a mechanism for migration potential similar
to that for landfills and unenclosed waste piles. EPA believes it is
appropriate to require minimum technology doub,. liner systems for
enclosed waste piles containing moist wastes that will generate
leachate.
EPA's new proposed regulations add an enclosure inspection
program to the enclosed waste pile requirements for permitted and
interim status units under Section 26
-------
and waste pile after every precipitation event (i.e., rain, snow, or
ice) and check for leaks.
3. ?. 5 Variances
Current regulations provide owners or operators of permitted (40
CFR 264) and 1nter1ir. status (40 CFR 265) surface Impoundments and
landfills with certain exemptions from the minimum technology double
liner standards. One type of exemption (e.g., 40 CFR 264.221(d))
applies if the owner or operator can demonstrate that alternative
design and operating procedures, together with location
characteristics, will prevent the migration of any hazardous
constituents Into ground water or surface water at least as
effectively as e '.inimum technology double liner system. The second
type of variance (e.g. 40 CFR 264.221(e)) applies to certain types of
monofills. EPA is proposing to extend to waste piles these two types
of exemptions for landfills and surface impoundments. It is EPA's
position that extension of these exemptions to waste piles is
appropriate because: (1) waste piles falling under the exemptions
will handle similar wastes as landfills and surface impoundments; and
(2) waste pile lining systems have similar designs and design lives as
landfills and surface Impoundments.
Owners or operators of permitted and inter 11 status units may
receive a variance from the minimum technology double liner
requirements if they are able to demonstrate that the alternative
design and operating procedures, together with location
characteristics, will prevent migration of any hazardous constituents
into ground water or surface water at least as effectively as the
minimum technology double liner system.
3.3 SIGNIFICANT PORTIONS
3.3.1 Definition of Significant Portions
"Significant portions" has not previously been defined in RCRA or
its amendments, but has evolved in concept from regulations pursuant
to RCRA. EPA is currently proposing to define significant portions
as: "any unlined area of a unit that has not received waste and, if
double lined before receiving waste, would significantly reduce the
3-14
-------
polentiil for ground-water and surface-, 'er contamination frcm the
uni t".
3.3.2 Background
EPA's current regulations require the lining of partial units at
permit issuance. Section 264.221(a) (for surface impoundments),
Section 264.251(a) (for waste piles), and Section 264.301(a) (for
landfills) require the portions of units not covered with waste at
permit issuance to Install a single liner (with a leachate collection
and removal system above the liner in the case of a landfill or waste
pile). This means that even if a indfill, surface impoundment or
wasie pile unit 1s exempt from the doub liner standards, any portion
of the unit not covered with waste at permit issuance is still subject
to EPA's current single liner standards in Sections 264.221 (a),
264,251(a), and 264,301(a). Hence, in this instance, EPA's current
standards have not been superseded by HSWA.
EPA believes the number of units that can be characterized as
having "significant portions" is small, probably less than 10.
However, the potential adverse impact to human health and the
environment of not double-lining the "significant portion" is large.
The potential benefit associated with the use of a minimum tech-
nology double-liner system compared to a single compacted soil liner
(with a LCRS above the liner in the case of a landfill) can be
assessed from the results of calculations presented in Section
3.2.3.2. In that section, it was pointed out that the last step
before leachate passes through a lining system and into the environ-
ment (where it can potentially adversely affect human health and the
environment) is its passage into the bottom-most liner in the lining
system. Therefore, migration of leachate into the bottom-most liner
should be minimized. In Section 3.3.3.2, it was shown that a single
compacted soil liner with kc = 1 x 10"' m/s (1 x 10~7 cm/s) would al-
low on the order of 89 Ltd (gpad) of leachate to enter into the liner.
In contrast, the migration of liquid into the bottom composite
liner of a double liner system is on the order of 0.2 Ltd (gpad) or
less. The difference between 89 Ltd (gpad) and 0.2 Ltd (gpad)
represents reduction in the potential for migration out of the unit by
3-15
-------
a fic'.or of 500. Since the above results are conservative, the actml
factcr pay be several thousand rather than 500. If a 5 acre
s i -.-,]' • -rant portion is considered and it has a 10 year active life,
and if the rate of leachate generation in the unit is in excess of 35
Ltd (gpad), the total quantity of liquid migrating into the bottom-
most liner will be on the order of 6 million liters (1.5 million
gallons) for a single compacted soil liner, and on the order of 15,000
liters (3,600 gallons), or less for the composite bottom liner of a
minimum technology double liner system.
3.3.3 Rationale for Double Lin"r System Requirements
The proposed "significant portions" regulations changes the
current single liner requirements for unused portions of existing
units to a requirement for a double liner system (for a significant
portion) or to a complete waiver of liner requirements (for a
nonsignificant portion). The objective of this change is to correct
inequities in the existing regulations, which would have owners or
operators of some units install liners which would not serve to reduce
threats to human health or the environment. However, EPA recognizes
the need to retain or enhance this protection in areas where liners
would serve to reduce the potential for the migration of hazardous
constituents out of the unit.
The primary purpose behind a requirement for minimum technology
double liner systems at significant portions is to provide these
portions with the same level of protection (by controlling migration
of hazardous constituents out of the unit to prevent ground water
contamination) of human health and the environment that is provided by
other newly constructed land disposal units. By requiring a minimum
technology double liner system for significant portions, EPA would be
minimizing the total number of land disposal units that can receive
hazard- -s waste and not be as protective of human health and the
environment as other units with minimum technology double liner
systems.
3.3.4 Proposed Exemption from Leak Detection Requirements
The proposed Liner/Leak Detection Rule does not require a leak
detection system to be installed at significant portions of existing
3-16
-------
units. Thus, the leachate collertion and removal system between the
liners would be exempt from the leak detection system performance
requirements outlined In diopter 2. Frcm a technical viewpoint, this
is a reasonable position because: (1) the possibility of leakage from
other areas of the unit that could cause a false -indication of leakage
through the top liner of the significant portion; and (2) potential
implementation problems from response actions being required for
leakage through the top liner when other portions of the unit may not
have any liner. These implementation problems are caused by having
different operational requirements for the "existing portions" part
and the "sign1f1cant portions" part of a unit. Without having a
consistent requirement for migration out of the whole unit into the
subsurface, 1t would be difficult or impossible to determine if the
portion of the unit with more stringent operational controls (the
"significant portion") is meeting its requirements. This is because
current monitoring techniques would not be able to determine which
areas of the unit were leaking. Therefore, EPA would not know whether
or not the "significant portion" was in compliance with the double
liner standards and any ootential response actions. However, owners
and operators should st II be encouraged to voluntarily comply with
the minimum technocal gu.delines for double liner systems that are set
forth in this document.
v 3..3.5 Examples of Significant Portions
Precise criteria have not been developed for categorizing
significant portions and nonsignificant portions. However, the
following examples provide guidance on EPA's thinking of what are
significant and nonsignificant portions:
• An example of a "significant portion" of an existing landfill
unit would be an exposed unlincd bottom area of several acres
that was not covered by waste. If double liners and a leachate
collection and removal system were installed in this area prior
to its receiving waste, a significant benefit to human health
and the environment would 1'kely result because large amounts
of leachate wcu'1 be collected and removed over a 5 year
period.
3-17
-------
• An example of a portion of an existing unit that in::y not be a
"significant port>on" is the unlined area of a surface
impoundment located above the liquid surface level that would
be covered with waste if the liquid level were raised.
• In most cases, "significant portions" will be those areas in a
unit where the addition of a double liner system will provide
hydraulic control of leachate or liqui waste and assure
collection and removal.
• "Significant portions" may include both the bottom and
sidewalls of existing units.
Examples of significant portions -re given in Figure 3-2.
3.3.6 Variances
In EPA's proposed rule changes for lining systems for significant
portions, owners or operators would be eligible for an alternative
technology variance. Owners or operators of significant portions of
permitted and interim status units wishing to use designs different
from those specified under the minimum technology requirements will be
allowed to do so if they are abl2 to demonstrate that the alternative
design and operating procedures, t~gether with location
characteristics, will prevent the migration of hazardous constituents
into ground water or surface water at least as effectively as a
minimum technology double liner system.
The proposed rule changes also provide a provision for owners or
operators of significant portions of permitted or interim status
facilities to seek a waiver of the double liner system requirements
for monofills containing only hazardous wastes from foundry furnace
emission controls or metal casting molding sands if such wastes do not
contain constituents which would render the waste hazardous for
reasons other than the EPA toxicity characteristics 1n Section 261.24,
40 CFR Ch.l. Further requirements to obtain such a waiver were given
previously in Section 3.2.5 of this report.
3-18
-------
3.4 NEW UNITS, REPLACEMENTS AND LATERAL EXPANSIONS AT FACILITIES
PERMITTED PRIOR TO NOVEMBER 8, 1934
3.4.1 Background
The statutory requirements of RCRA (Section 3004(o)(1)(i)) and
current regulations specify minimum technology double liner systems
only for new units, lateral expansions and replacements of existing
landfills and surface impoundments that receive permits after November
8, 1984. As previously discussed in Section 3.3.2, units permitted
prior to November 8, 1984, may currently be required to have single
liners (and leachate collection and removal systems above the liner
for landfills and waste piles) on unused portions of a unit.
3.4.2 Rationale for Double Liner System Requirements
EPA is proposing to extend the minimum technology double liner
system standards to new units, replacement units, and lateral
expansions of surface impoundments, waste piles and landfills at
facilities permitted prior to November 8, 1984. The rationale for
this proposal 1s to assure that these units provide the same level of
protections of human health and the environment as is provided at
other newly constructed units. The EPA's prcpo.,al will minimize the
number of units in which waste can be placed that are not as
protective of human health and the environment as units having minimum
technology double liner systems. EPA is of the belief that the
opportunities for construction, and impacts on owners or operators,
are similar to those for units permitted after November 8, 1984. In
other words, the technical and resource requirements for implementing
the minimum technolog., double liner system standards are identical for
new units replacements and lateral expansions permitted prior to, and
after, November 8, 1984. No additional requirements are placed on
owners and operators of units permitted prior to November 8, 1984 that
aren't placed on units permitted after that date.
3.4.3 Exemptions for Certain Replacement Units
EPA Is proposing to exempt certain replacement units permitted
prior to November 8, 1984 from the minimum technology double liner and
leachate collection and removal requirements, as well as the leak
3-19
-------
detection system requirements proposed today. (EPA can exempt these
units from the leak detection requirements because they are not
req.'i'-ed by the statute to have loak detection.) Die number of units
affected by this proposed extension of the double liner standard is
small. The, total number of affected facilities is believed to be
approximately eight. It is believed that all of these cases will
involve lateral expansions or replacements (i.e, no new units).
The types of land disposal units that the Agency believes are most
likely to be defined as replacement units are surface impoundments and
waste piles. As discussed in the Federal Register (50 FR 28742, July
15, 1985), a unit qualifies as a replacement unit when: (a) the unit
is taken out of service (i.e., the receipt of waste is stopped or the
normal input of waste is significantly reduced); (b) all or
substantially all of the waste is removed; and, (c) the unit is
reused. However, a unit is not considered a replacement unit if the
waste is removed from the unit for treatment, treated, and then put
back in' the same unit as part of the unit closure plan.
EPA is considering exemptions from the double liner system and
leak detection system requirements for those replacements of
landfills, surface impoundments, and waste piles that meet the
following condi tions:
• The existing unit eceived a final permit prior to November 8,
1984;
• The existing unit was constructed in compliance with the single
liner requirements (and leachate collection and removal system
requirements for landfills and waste piles^ or requirements for
equivalent protection (i.e., the variance) contained in Part
264 and the liner or leachate collection and removal system is
not replaced; and
• There is no reason to believe that the liner or leachate
collection system is not functioning as designed.
EPA is considering exemption of units that meet the above criteria
from the double liner system and leak detection system requirements,
because the owner or operator of these units made a good faith effort
3-20
-------
to satisfy the liner system requirements that were in effect at the
tin"? the facility was permitted (aivJ the liner cr leachate collection
syslc,"i are still functioning as designed); EPA also considered that in
many cases the owner or operator would be required to totally replace
the whole unit. Retrofitting an additional liner on top of the
existing liner would not be feasible. TiVs is for three reasons: (1)
existing single liners would not meet double liner system bottom liner
requirements; (2) reduced capacity would not meet unit owner or
operator needs; and, (3) retrofitting a design concept does not allow
the owner or operator to meet new BOAT technology for liners.
3.i.4 Variances
In EPA's proposed rule changes owners or operators of new units,
replacement units, and lateral expansions of units at facilities
permitted prior to November fa, 1984, would be eligible for the same
variances as previously described in Section 3.2.5 of this report.
3-21
-------
Table 3.1 Comparison of waste types accepted at lanr(fills arrl v
pile i-nits. [This data is from the fcPA 1986 ".';a t J
Screening Survey of Hazardous '.-,'iste Treatment, jf.3M.-je,
Disposal Facilities" (USEPA, 1936d)J
Landfills'
Waste Piles'
Acidic Corrosives
(PH<2)
Metals
Cyanides
Solvents
PCGs
Dioxins
Other Halogenated
Organics
Other Hazardous
Waste
Total Number of
Units (approximate)
No. of Units
35
94
37
37
3
2
36
90
150
(70 '
23
63
24
24
2
1
24
60
No. Of Units
13
75
4
12
11
1
12
54
80
<%)•
16
94
5
15
14
1
15
67
Notes: (1) The number of units refers to the nuirber of units willing
to treat, store, dispose or recycle the particular waste
type. The (%) refers to the percentage of waste pile
units willing to treat, store, dispose, or recycle that
waste type.
(2) Percentages are estimates based on the estimated total
number of units.
3-22
-------
Figure 3-1.
Typical waste pile incorporating minimum technology
double 1iner systems.
3-23
-------
(O La-clfL
i srk"
Figure 3-2. Examples of significant portions.
3-24
-------
71
CHAPTER 4
CONSTRUCTION QUALITY ASSURANCE
-------
n
4.1 INTRODUCTION ]
i
4.1.1 5ccpe of Chapter 4 j
The purpose of Chapter 4 is to provide a discussion of •.
Construction Quality Assurance Program requirements in the proposed j
Liner/Leak Detection Rule. This chapter therr -e includes: a summary j
of the proposed additions to 40 CFR Part 264 and Part 265
Regulations; technical rationale for the proposed changes; and,
regulatory Issues and options.
Chapter 4 1s comprised of 6 sections which are briefly summarized j
below.
Section 4.1 describes the background and rationale for the
Construction Quality Assurance Program, as well as the roles and
responsibilities associated with implementation of a CQA Program.
Section 4.2 Is concerned with the elements of the written CQA
Plan: the organizations which will be involved in the program; the CQA
Officer and the CQA personnel; sampling strategies; and reports and i
documentation. !
i
i
Section 4.3 addresses the role that the Construction Quality
Assur ;e plays In double liner systems. This section discusses: the
sensitivity of system performance to construction procedures;
materials issues; and the benefits of Constructior Quality Assurance.
Section 4.4 1s concerned with the scope of a Construction Quality
Assurance Program. This section addresses: the role of CQA during the
pre-construction and design stage; CQA tasks during construction; and
post-construction CQA, including reporting and monitoring
requi rements.
Section 4.5 1s concerned with the testing procedures used in a
Construction Quality Assurance Program. This section addresses:
laboratory and field soils tests; soils tests acceptance criteria;
laboratory and field flexible membrane liner (FHL) tests; FML
4-1
-------
accept'ince criteria; and testing proceduies and acceptance criteria
for c'.f-jr qecsynthetic materials.
Section 4.6 discusses Const.-., .ion Quality Assurance issues. This
section addresses: the qualifications of the CQA Officer; levels of
control within the CQA Program; administrative timing of review and
approval of CQA Plans; interim status vs. permitted CQA Plan
requirements; and length of comment periods.
4.1.2 Rationale for the CQA Program
Historically, a conscientious and well-managed Construction
Quality Assurance Program has served to ensure that a completed
project meets or exceeds the specified design. The provision of
quality assurance is a standard component of virtually any project
involving the placement and compaction of soils. Similarly,
geosynthetics require close monitoring during placement in order to
ensure a quality irxstallation. As a result, EPA believes that the
Construction Quality Assurance (CQA) Program represents an essential
element of its overall liquids management strategy. The CQA Program
must ensure that all foundations, low-permeability compacted soils,
flexible membrane liners (FML's), dikes, leachate collection and
removal systems (LCRS's), and final cover, meet or exceed all design
criteria, plans, and specifications.
The first element of the COA Program is the preparation of a site-
specific Construction Quality Assurance Plan. The CQA Plan addresses
activities such as monitoring, documenting, and sampling for the
Individual components. By providing the CQA Plan during the design
stages of the unit, permit granting agencies and regulators are able
to review the specific procedures that the Owner/Operator will use to
comply with CQA requirements. For permitted units, the CQA Plan must
be submitted to, and approved by, the Regional Administrator (RA)
before construction will be allowed to begin.
The second element of the CQA Program is the implementation of the
CQA P'^n by the CQA Officer (i.e., a registered professional
engine.,-.) Thorough and complete documentation of all of the
4-2
-------
» rconitoring activities and signature(s) nctiny the compliance with the
Construction C'-^lity Assurance Plan are required in the repor
suL:;u t (.•:•] by the CQA Officer to the ov-ner/cper ator and by ti
cwner/ocerator to the Regulatory Agency. The submission and approva1
of this report are prerequisites to the granting of permission to
receive waste at the unit.
\ The Construction Quality Assurance Program is directly related to
both parts of EPA's liquids management strategy: minimizing the
generation of leachate and maximizing leachate removal. To ensure that
the waste management system will meet EPA's objectives, all components
of the total system must function as designed: top and bottom liners,
leachate collection and removal systems above and between the liners,
the leak detection system, and the final cover. The CQA Program
establishes specific activities that the Owner/Operator must implement
to ensurn the quality of each component of the system.
Use of systematic CQA programs at waste piles, landfills, and
surface impoundments will heip ensure that each unit is designed and
constructed to the same general standards. This applies to the lining
systems and covers of waste piles, landfills, and surface
impoundments, and U IE covers for land treatment units. The latter
involves covers only, because the treatment of wastes in these units
involves the application of the waste on the soil surface or into the
upper soil layers in order to degrade, transform, or immobilize
hazardous constituents present (See Chapter 5).
There are significant benefits to tr- provision of a CQA Program.
particular, by the provision of a comprehensive, thorough program
of Construction Quality Assurance, '"here will be greater scrutiny and
diligence, as well as the establishment of formal protocols for
conducting examinations and testing for flaws in materials.
Consequently, when flaws are found and corrected there will be a
„ reduced potential for nrgration of hazardous constituents out of the
unit and into the environment will be reduced.
Oy the provision of a high quality and intensive CQA program, the
responsible parties (Owner/Operator, Designer, and Regulators) will
4-3
-------
have greater confidence that every effort has been made to ensure
confor!"3,ire with the design and specifications, and that the unit will
therefoie perform as intended. As a natural extension, there will be
greater confidence in the integrity of the unit in the ?yes of the
public.
4.1.3 Definitions Related to CQA
Within the context of the Construction Quality Assurance Program,
many of the terms and definitions used may differ from their common
usage. The following definitions are those which are used in Chapter
4. A discussion or c' -ification is also provided in order to avo; '
any misinterpretation uf these terms.
• Quality Assurance - "means all those planned and systematic
actions needed to provide adequate confidence that products or
services will satisfy specific requirements." [Canadian
Standards Association (CSA), 1986].
In this document, Construction Qmllty Assurance includes
the provision of quality assurance services for the
man'.'facture, fabric3tion, and installation of the geosynthetic
components of lining systems, including flexible membrane
liners (FMl's) (also referred to as geomembranes), geotextiles,
geonets, and geogrids; as well as the testing, placement, and
compaction of the soils components of land disposal units,
Including foundations and dikes, compacted low-permeability
soil layers and high-permeability grant ar drainage layers.
Construction Quality Assurance actually has pre-construction,
construction, and post-construction components.
In addition, Giroud and Fluet [1986] state that: "In the
context of geomembrane-1ined facilities: ... Quality assurance
refers to means and actions employed by the owner through the
quality assurance team to assure conformity of the design,
production (i.e., manufacture and fabrication) and Installation
with the quality assurance plan, as well as with drawings and
specifications. 'Third party quality assurance' refers to a
4-4
-------
quality assurance team which is independent of the designer,
mrvjfacturer, fabricator, installer or ov.ner".
Qua 1i ty Contr 1 - "those actions which provide a means to
measure and regulate the characteristics of an item or service
to contractual or regulatory requirements." [Giroud and Fluet,
1986].
In addition, Giroud and Fluet [1986] state: "In the context
of geomembrane-1ined facilities: Quality control refers to
those action? taken by the designer, manufacturer, fabricator,
and/or installer to ensure that their methods, materials and
workmanship are accurate and correct and meet the requirements
of regulations, plans and specifications. ... Quality control
is provided by each party for its own work (e.g., quality
control of the installation is provided by the installer) while
quality assurance is provided by a party independent from
design, production and Installation. Quality control is
therefore associated with the offering (selling) of a product
or service, wnereas quality assurance relates to the acceptance
of a product or service".
In the case of the quality assurance of soils placement and
compaction, the CQA Officer normally undertakes the
nondestructive quality assurance testing, and, since the
Earthworks Contractor does no other testing, this also serves
as quality control testing. In this c.ase, this testing is
required to determine the acceptability of the work, and
traditionally, Earthworks Contractors have accepted the results
of the testing by the CQA Contractor, but this in no way
relieves the Earthworks Contractor of his responsibility for
conformance.
Construction Quality Assurance Plan - A Construction Quality
Assurance Plan is a site-specific document which describes the
Construction Quality Assurance Program. This CQA Plan is
normally prepared by the CQA Contractor or Designer, and
includes, at the least:
4-5
-------
Site-Specific Application - the n.nne, location, nature,
general description, and type of waste management unit;
Roles and Responsibilities - identifies all parties to the
program, qualifications of the parties and their personnel,
and their responsibilities and levels of authority;
Procedures - outlines or includes all procedures to be
followed 1n the performance of the work, Including:
test procedures;
acceptance/rejection criteria;
repair procedures;
samp!ing procedures;
documentation procedures; and
reporting procedures.
Documentation Requirements - lists all elements of the
required documentation, including the final report,
signatures required, and the record drawings.
Specific requirements of the Construction Quality Assurance
Plan are discussed in Sectior 1.2.
Nondestructive Testing - Nondestructive testing consists of
procedures for testing the materials in relation to the
specifications, without damaging or otherwise requiring
reconstruction or replacement of the tested material. For
example, in the context of soils testing, nuclear test
procedures determine the in situ density of the soil without
disturbance of those materials. In this case, a quantitative
result 1s obtained (e.g., soil density). In the context of
nondestructive testing of FML's, vacuum testing (for instance)
is carried out without damaging the seams. In this case, a
qualitative result is obtained (e.g., confirmation of seam
continuity).
Destructive Testing - Destructive testing consists of
procedures for testing the materials in relation to the
4-5
-------
specifications using procedures U.at result in destruction of a
p:r:ien of the inslaPed c-ite-ijl. Destructive testing thus
requires some form of" replacement or repair of the
installation. In the context of soils testing, destructive
testing would normally consist of the field sampling of the
compacted soil material" ror laboratory testing, which
requires replacement of thL Uerial removed. In the context of
flexible membrane liner seam testing, samples are cut 'om the
field-formed seams for laboratory testing to failure, to
determine the ultimate strength properties of the seam. The
sampled location must then be repa' ""d.
• Con forma nee Testing - Conformance testing is carried out on
samples collected from the materials supplied to the site, but
before installation. This testing is carried out in the
laboratory to verify the properties and record the conformance
of these properties with the specifications. Sufficient
additional materials must be provided to the site such that the
samples removed for conformance testing (at a specified
frequency) do not result in a shortage of material to complete
the work.
4.1.4 Partiesto CQA - Roles and Responsibilities
There are many parties involved in the Construction Quality
Assurance Program and these interact in different ways. It is
important to identify these parties, as well as their roles and
responsibilities, in the CQA Program.
As outlined 1n Giroud and Fluet [1986]: "The following is a
listing of the various parties along with a brief description of
their roles and responsibilities:
- Designer - responsible for the design, drawings, plans and
specification of the lining system and the supporting soil.
- Civil Engineering Contractor - responsible for the preparation
of the supporting soil on which the lining system is to be
4-7
-------
instal'-d and for the construction of the concrete structures
and t.. pipe systems to which t"e lining system ts to be
connected; may also be the party responsible for placing earth
or concrete cover and granular drainage or filter materials, if
any.
Polymer Supplier - produces and delivers raw polymer (typically
in the form of flake or pellets) to the manufacturer.
Manufacturer - responsible for production of geosynthetics from
raw polymer. !n the case of geomenbranes, produces rolls of a
constant width.
Fabricator - responsible for the fabrication of geosynthetic
panels from geosynthetic rolls. (Geomembranes are the most
1ikely materials to require fabrication.)
Transporter - transports geosynthetic rolls from manufacturer
to fabricator or the site and/or geosynthetic panels from
fabricator to the site. (Transportation is usually not critical
for geosynthetics other than geomembranes.)
Installer - responsible for field handlin storing, placing,
seaming and other site aspects of the geosynthetics; may also
be responsiole for anchor trenches and all temporary anchoring
or loading required to support the lining system during
installation.
Quality Assurance Contractor - party (independent from the
designer, manufacturer, fabricator, installer and owner)
responsible for observing and documenting activities related to
the quality assurance of the geosynthetic lining system. ...
The quality assurance contractor is the employer of the quality
assurance team.
Quality Assurance Team - the quality assurance team includes:
(i), one quality assurance managing engineer, who is located at
the offices of the quality assurance contractor and is
4-8
-------
responsible for managing the quality assurance team; (11), a
quality assurance manager, who is physicr y present at the
fabrication factory and/or site throughout the lining system
fabrication and/or installation, and assigns tasks to the
quality assurance monitors and otherwise manages their
activities; and Mil), quality assurance monitors who are
responsible for documentation of all Installation actl/lties
which they observe. ...
- Quality Assurance Laboratory - party (independent from the
designer, manufacturer, fabricator, Installer and owner)
responsible for conducting tests on samples of geosynthetics
taken from the fabrication factory and/or the site. For large
projects, the laboratory may be located on site.
- Owner/Operator - owns and/or is responsible for the lined
facility. For quality assurance purposes, the term 'owner'
usually applies equally to 'operator', i.e., the party
responsible for operating the lined facility.
- Project '^an^ger - the official representative of the owner;
i.e., cne Individual in charge of coordinating field
activities.
- Regulator(s) - responsible for enforcing compliance with
regulatory statutes and/or codes."
In addition:
- Earthworks Contractor - the party responsible for the placement
and compaction of soils for earthworks and the compacted soil
1iner at the site;
- Soil Supplier - the party responsible for supplying soils to
the site;
- Soils CQA Contractor - the party (Independent from the
designer, manufacturer, fabricator, installer and owner)
4-9
-------
responsible for observing and documenting activities related to
the quality assurance of the earthworks and compacted soil
components of the lining system... The Soils Quality Assurance
Contractor Is the employer of the Soils Quality Assurance Team
and may also be the Geosynthetics Quality Assurance Contractor.
Within the context of this document, the roles of the CQA Manager
and CQA Managing Engineer are fulfilled under the title "CQA Officer",
whose function may or may not coincide with the two separate roles
described above. Similarly, the Quality assurance laboratory is
referred to herein as the "Independent Test Laboratory". In addition,
the roles and responsibilities of the CQA Contractor, and hence the
CQA Team, will also Include the Construction Quality Assurance of the
soils and other non-geosynthetic components of the work. In any event,
the Construction Quality Assurance Plan should clearly delineate the
responsibilities of each party.
4.2 CONSTRUCTION QUALITY ASSURANCE PLAN
The unit Owner/Operator must submit a written CQA Plan as part of
the permit application. Although the overall content of the CQA Plan
will depend on the site-specific conditions for the proposed hazardous
waste management unit, at a minimum, several elements should be
included in the Plan. These elements of the Construction Quality
Assurance Plan are summarized in this section.
• General Description of the Unit—Plans for the design,
construe ion, operation, and closure of the unit should be
discussed. The description should identify the construction
stages for the components at the unit.
• Responsibility and Authority—The responsibility and author.ty
of organizations and key personnel (by title^ involved In
permitting, designing, constructing, and quality assuring the
hazardous waste management unit should be described In the CQA
Plan. The description must assure that the objective of the CQA
Program identified in 40 CFR 264.19 will be met.
4-10
-------
* CQA Personnel Qualif1cat1ons--The qualifications of the CQA
Officer and reporting CQA personnel should be presented in the
CQA Plan in terms of the training and experience necessary to
fulfill their identified responsibilities.
• CQA Monitoring and Sampling Actlv11ies--The observations and
tests that will be used to ensure that the construction or
Installation meets or exceeds all design criteria, plans, and
specifications for each hazardous waste management unit
component should be described in the CQA Plan. The sampling and
monitoring activities for all constructed components, sample
size and sample locations, frequency of testing, data
evaluation procedures, acceptance and rejection criteria, and
plans for Implementing corrective measures as addressed in the
project specifications should all be presented In the CQA Plan.
• Documentation of Construction Quality Assurance Actlvities--
Reportlng requirements for CQA activities should be described
in detail in the CQA Plan at the time of its submittal for
approval. This should Include such Items as daily summary
reports, observation data sheets, problem identification and
corrective measures reports, block evaluation reports,
acceptance reports, and final documentation. Provisions for
the final storage of all records should also be presented in
the CQA Plan.
Each of these elements is described in the following subsections. A
detailed outline of the requirements of the CQA Plan is presented in
USEPA [1986b], from which these requirements have been adapted.
4.2.1 General Description of the Unit
The land disposal unit should be clearly identified In the CQA
Plan, Including direct references to the project plans and
specifications, and a description of the type of waste management
unit, its dimensions and capacity. The proposed operating duration,
and plans for closure must also be presented.
4-11
-------
H
4.2.2 Respon5lbHlti_and Authority
4.2.2.1 Organizations |nvo]ved_1n .QA
The principal organizations involved in permitting, designing, and
constructing a hazardous waste management unit Include the Permitting
Agency or Regulator, the unit Owner/Operator, the Designer, the CQA
Contractor, and the Construction Contractor(s). Except for the
Regulator, the principal organizations will not necessarily be
completely Independent of each other; e.g., the unit Owner/Operator
may also be the Construction Contractor, and the CQA Contractor
personnel may be employees of the unit Owner/Operator, the Designer,
or an Independent Firm. Regardless of the relationships among the
organizations, It Is essential that the areas of responsibility and
lines of authority for each organization be clearly delineated early
In the CQA Pian. This will help establish the necessary lines of
communication that will facilitate an effective decision-making
process during Implementation of the site-specific CQA Plan. It Is
also essential that the organization performing CQA operates
Independently of and Is not responsible to the organization(s)
Involved in constructing the unit.
4.2.2.2 Project Meetings
Periodic meetings held during the life of the project strengthen
responsibility and authority by enhancing communication between
personnel responsible for designing, constructing, and documenting
construction of a hazardous waste management unit. Since conducting
periodic project meetings Is not a mandatory feature of the proposed
Liner/Leak Detection Rule, the decision to hold project meetings is at
the option of the unit Owner/Operator; however he may delegate that
responsibility to one of his supporting organizations (e.g., Designer
or CQA Contractor). Regardless of which party conducts them, periodic
project meetings benefit all those involved with the unit by ensuring
familiarity with unit design, construction procedures, and any design
changes. Examples of the types of meetings that may be held are:
4-12
-------
* Preconstruct ion CQA Meeting
Immediately prior to construction, a meeting may be held to
resolve any uncertainties following the completion of the unit
design, completion of the site-specific CQA Plan, and award of
the construction contract. The unit Owner/Operator, Designer,
CQA personnel, and Construction Contractor would normally all
be present. This meeting would serve to familiarize all
parties with the CQA Plan, their various responsibilities,
lines of authority, etc., prior to the commencement of any site
construction activity. The meeting would also resolve
differences between the parties in sufficient time to allow
modifications to the CQA Plan prior to commencement of
construction.
• Daily Progress Meetings
A progress meeting may also be held daily at the work area just
prior to commencement, or following completion of work. This
meeting would normally be attended by the Construction
Contractor and the CQA personnel. The purpose of the meeting
would be to review the activities of the previous day or shift,
review the activities for the upcoming day or shift, and
discuss any potential construction problems.
* Problem or Work Deficiency Meetings
A special meeting may be held when and if a problem or
deficiency Is present or likely to occur. At a minimum, the
meeting would be attended by the Construction Contractor and
the CQA personnel. The purpose of the meeting would be to
define and resolve a problem or recurring work deficiency.
4-13
-------
4.2.3 Personnel Qualifications
Ihe CQA Plan should identify the required qualifications of the
CQA Officer and the CQA Team, and describe their expected duties.
4.2.3.1 CQA_Offleer
The CQA Officer Is assigned responsibility for all aspects of CQA
Plan implementation. The CQA Officer is responsible to the unit
Owner/Operator, and should function Independent of the Construction
Contractor. The location of the CQA Officer within the overall
organizational structure of the project, Including the unit
Owner/Operator, Designer, Construction Contractor, and Regulator,
should be c'early described within the CQA Plan as noted in the
previous discussion on respnnsibi1ity and authority.
The CQA Officer should possess adequate formal academic training
ir. engineering, engineering geology, or closely associated
disciplines; and sufficient practical, technical, and managerial
experience to successfully oversee and implement Construction Quality
Assurance activities for hazardous waste management units. In almost
all states, the responsibilities of a CQA Officer are of a nature that
lead to the requirement t:iat he/she be a registered Professional
Engineer. Because the CQA officer may have to Interrelate with all
levels of personnel Involved In the project, good communication skills
are essential. The CQA Officer should ensure that communication of
all CQA-related matteis is conveyed to and acted upon by the affected
organizations. The CQA Officer should have specific training,
experience, and knowledge of the materials (soils and geosynthetlcs)
for which he will be providing Construction Quality Assurance.
On some projects (particularly large ones), the duties of the CQA
Officer may be divided between two or more arsons. For example, the
engineering aspects may be performed by a Managing Engineer, while the
construction and managerial aspects may be performed by an on-site
Quality Assurance Manager.
4-14
-------
4.2.3.2 CQA Monitoring
:t;e CQA Monitors should possess aJcquato forrril training and
sufficient practical, technical, and admin i strati vt experience to
execute and record CQA activities successfully. This should include
demonstrated knowledge of specific field practices relating to
construction techniques used for hazardous waste management units, all
codes and regulations concerning material and equipment installation,
observation and testing procedures, equipment, documentation
procedures, and site safety. They should also have spscific training
and knowledge regarding the materials (soils and geosyflthetics) which
they will be monitoring.
4.2.3.3 Consultants
Authorities in engineering geology, geotechnical engineering,
civil erglneerlng, and other technical disciplines may be called In
from external organizations in the ?nt of unusual site conditions or
observations. The final report should present detai les! documentation
of Consultant qualifications when expert tecMical judgments are
obtained and used as a basis for decision in seme aspect of
Construction Quality Assurance. Expert opinions should not be used as
a substitute for objective data collection and interpretation when
suitable observations and test procedures are available,
4.2.4 CQA Monitoring and Samp IJng Activities
The CQA Plan should describe the monitoring and sampling j
activities (observations and tests) that will M perforaed by the CQA j
Team during hazardous waste management unit cc ,:truct}ia. The scope j
of this discussion should address only the construction and j
Installation of all unit components and the r'inufacture/'fabr1cat1on of j
various components and subcomponents when pertinent. It is assumed |
that the site has been cnaracterized adequately, including evaluation I
of the hydrogeologlc environment. It is also assumed that a site-
specific i It design has been prepared that meets regulatory
requirements and is acceptable to the unit Owner/Operator, and that
this design has been evaluated to ensure its technical correctness and
feasibility.
4-15
-------
I fie components of the unit which require the provision of CQA
sorv ,:es include (but are not limited to):
• foundations;
• dikes;
• compacted low-permeability soil liners;
• other soils; j
• flexible membrane liners; ;
• leachate detection, collecLion, and removal systems; <
• other geosynthetics; and
• final cover systems. ;
i
For many materials and construction processes, it is necessary to j
estimate the quality of the overall material or process from the •.
observed or measured quality of a representative samp'e that is a ;
small fraction of the total material or process. Examples of these ;
situations Include assessment of characteristics of a compacted soil ';
Hner (e.g., permeability, moisture content, density, particle size j
distribution) and destructive testing of FKL seams. >.
i ^
Some of the key characteristics of corrronly used sampling •;
strategies Include: •
[
• data type; ;
• acceptance/rejection criteria; '
• sampling uni ts; i
• number of sampling units and number of measurements per unit;
• location of sampling units and/or measurements within units; I
• treatment of outliers; and i
j
• corrective measures. 1
j
The current state of knowledge on sampling strategies for \
hazardous waste management unit CQA is not well enough developed to '
enable EPA to recommend a specific approach for designing a sampling \
strategy. For Instance, the measurement error inherent in test j
methods is an Important piece of information when devising a j
statistical sampling strategy. However, the measurement error ]
associated with certain important test methods (e.g., laboratory and j
i
4-16 1
-------
field permeability) is not known. Until nore information is
available, the selection of appropriate sa.-pling strategies should te
conducted with the guidance of knowledgeable engineers and
statisticians.
4.2.5 Documentation of Construction Qua]jty^Assurance Activities
The ultimate value of a CQA Program depends to a large extent on
recognition of all of the construction activities that should be
monitored, the selective assignment of responsibilities to the various
members of the CQA Team for the monitoring of each activity, and the
careful documentation of all observations. This documenting of CQA
activities must be addressed in the CQA Plan. The CQA Team should be
reminded of the Hems to be monitored, and should note, through
required descriptive remarks, data sheets, and checklists signed by
them, that the monitoring activities have been accomplished.
4.2.5.1 Dally Record Keeping
Standard dally reporting procedures should include the preparation
of a summary report with supporting logs and data sheets and, when
appropriate, problem Identification and corrective measures reports.
In particular, the following documentation should be prepared on a
dally basis:
• logs;
• data sheets;
• problem identification/corrective measures reports;
• photographic reporting data sheets;
• documentation of correction (cross-referenced to data sheets);
• final results;
• suggested methods to prevent similar problems; and
• signature of the appropriate CQA monitoring personnel and
concurrence by the CQA Officer.
In some cases, not all of the above information will be available
or obtainable. However, when available, such efforts to document
problems could help to avoid similar problems 1n the future.
4-17
-------
Upon receiving the CQA Officer's written concurrence, copies of
the report should be sent to the Designer and the unit Gorier/Operator
for their comments and acceptance. These reports should not be
submitted to the Permitting Agency at that time unless they have been
specifically equested. However, a summary of all data sheets and
reports may be required by the Permitting Agency upon completion of
construction.
4.2.5.2 Photographic Report1ng_Data Sheets
Photographic reporting data sheets may also prove useful. Such
data sheets could be cross-referenced or appended to data sheets
and/or problem identification and corrective measures reports. These
photographs will serve as a pictorial record of work progress,
problems, and corrective measures.
4.2.5.3 Block Evaluation Reports
Within each construction block (a block Is a group of related
activities that are carried out at a certain point in time), there may
be 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. There block evaluation reports may
then be used to summarize all of the site construction activities.
4.2.5.4 Acceptance of_Comp^eted Conponents
All dally summary reports, data sheets, problem identification
and corrective measures reports, and block evaluation reports should
be reviewed by the CQA Officer. The documentation should be evaluated
and analyzed for Internal consistency and for consistency with similar
work. Timely review of these documents will permit errors,
Inconsistencies, and other problems to be detected and corrected as
they occur, when corrective measures are easiest.
The above Information may be assembled and summarized Into
periodic Acceptance Repc .s, or otherwise summarized In the final CQA
4-18
-------
Report. The reports should Indicate that the materials and
construction procedures comply with the plans and specifications.
These reports should be Included in project records, submitted to the
unit Owner/Operator, and submitted to the Permitting Agency.
4.2.5.5 Final Documentation
At the completion of the project, the unit Owner/Operator must
submit a final report to the Permitting Agency. This report should
Include all of the dally summary ,sports, data sheets, problem
Identification and corrective measures reports, block evaluation
reports, photographic reporting data sheets, acceptance reports,
deviations from design and material specifications (with justifying
documentation), and record drawings. This document should be signed
by the CQA Officer and Included as part of the CQA Plan documentation.
4.2.5.6 Document Control
The CQA Plan and all CQA documentation should be maintained under
a document control procedure. This indexing procedure should provide
for convenient replacement of pages 1n the CQA Plan, thereby not
requiring a revision to the entire document; should Identify the
revision status of the CQA documents; and should enable the CQA
documents to be organized 1n terms of their relationship to each
other, the CQA Plan, and the time and location of the materials and/or
workmanship that they represent.
4.2.5.7 Storage of Records
During the construction of a hazardous waste management unit, the
CQA Officer should be responsible for all unit CQA documents. This
Includes the CQA Officer's copy of the design criteria, plans, and
specifications, the CQA Plan, and the originals of all data sheets and
reports. It 1s suggested that duplicate records be kept at another
location to avoid loss of this Information if the originals are
destroyed.
4-19
-------
Once unit construction is cc.rplete, the document originals should
be stored by the Owner/Operator 1n s. manner that will allow for easy
access while still protecting them f- -n any damage. An additional
copy should also be kept at the unit if this is in a different
location from the Owner/Operator's files. A final copy will be kept
by the Permitting Ag :y in a repository accessible by the public.
All documentation should be maintained through the operating and post-
closure monitoring periods of the unit.
4.3 NEED FOR A CQA PROGRAM
4.3.1 Background
The performance of a lining system is sensitive to both the
methods and procedures followed during construction as well as to the
quality of materials used to construct 1t. Once constructed, each
component must be able to perform Its intended function within the
lining system. This need has been recognized by the EPA, and resulted
In the preparation of a preliminary technical guidance [USEPA, 1987],
and finally a Technical Gu1d""ce Document, "Construction Quality
Assurance for Hazardous Waste Land Disposal Facilities" [USEPA,
1986b]. In this document, the background information provided
Identifies those parameters which are critical to the successful
completion of the unit. The result 1s that CQA has been Incorporated
as an integral component of the proposed Liner/Leak Detection Rule. As
component design criteria and construction methods Improve, the CQA
Program should provide a means of monitoring and documenting them. By
doing so, the CQA Program will become a part of the cost of the
facility. The following sections examine the considerations and
constraints that cumulatively Illustrate the need for the CQA Program.
4.3.2 Effect of Construction Procedures on Lining System
Performance
The methods and procedures used to construct all components of
landfills, surface impoundments, land treatment units, and waste piles
directly affect the performance of the respective components of the
facility. In addition, those methods and orocedures used to construct
4-20
-------
a particular cmponent of the system may, either directly or
indirectly, affect the performance of other components of the
facility.
Certain types of problems may potentially arise which must be
specifically addressed, In order to ensure conformance to the plans,
specifications, Construction Quality Assurance PJ?n, and good
construction practice in general. The objective of the CQA program is
to help minimize the occurrence of these potential problems. The
problems, which can be divided into soils-related problems and
geosynthetlcs-related problems, are reviewed below to demonstrate
potential benefits arising from the CQA program.
4.3.2.1 Soils-Related Construction Problems
4.3.2.1.1 Moisture Control and Compaction
Compaction criteria which are developed for the placement of soil
materials, primarily for site earthworks, but also tor the compacted
soil components of the lining system, are generally based on ^ne of
two standard laboratory tests: ASTH D698 "Test Methods for Moisture-
Density Relations of Soils and Soil-Aggregate Mixtures, Using 5.5-lb
(2.49-Kg) Rammer and 12 in (305 mm) Drop"; and ASTM 01557 "Test
Methods for Moisture-Density Relations of Soils and Soil-aggregate
Mixtures Using 10 Ib (4.54 Kg) Rammer and 18 in (457 rrm) Drop" (see
also Section 4.5.1.1.1). Commonly referred to as the Standard Proctor
and Modified Proctor tests, respectively, these tests produce, for a
given soil, the relationship between achievable dry density a -\
moisture content and a particular level of compactlve effort. The
moisture content for which the highest dry density (referred to as the
maximum dry density) is achieved, 1s called the optimum moisture
content. The field compaction specification 1s then based on a given
percentage (usually 90 percent to 100 percent) of this maximum dry
density and on a given range of water contents referenced to the
optima, water content. Density is specified 1n cases wherein strength
or soil support 1s a primary requirement for the compacted soil. In
other situations, such as for the compacted soil components of the
liner systems, for which low permeability 1s the requirement,
4-21
-------
specifications may call for a particular hydraulic conductivity,
regardless of the compaction of the soil. In fact, optimum conditions
for maximizing soil density are not necessarily the same as optimum
conditions for minimizing soil permeability.
Compaction testing criteria, when applied to site earthworks such
as berms or other site soils which are not components of the lining
.Astern, are an effective Indicator of conformance to the
specifications. For tne compacted soil liners, however, It Is
hydraulic conductivity which Is the specifics property. The density
and water content of the soil can, however, still be used as an
approximate indicator of the permeability, provided that the
relationships between the degree of compaction, density, moisture
content, and hydraulic conductivity are known. These relationships
are discussed further In other sections of this chapter.
Very few soils are indifferent to their moisture condition, In
rngard to the maximum dry density. In fact, the characteristic curves
% igure 4-1) taken from the results of either Proctor test show much
lower values of attainable density (for that given, fixed level of
compactive effort) at moisture contents both lower than, and higher
than optimum (curve 1). As can be seen, however, for Increasing levels
of compactive effort (curves 2 to 4), the attainable density of the
soil Increases, while the optimum moisture content decreases. A line
can be drawn through the 'family' of maxima, which 1s indicative of
the universe of maximum densities for different levels of compactive
effort. For saturated samples, this line would coincide with the zero
air voids line, shown on the figure.
It Is known, however, that the hydraulic conductivity of clayey
soils is not directly related to dry density. There has been
considerable research into this phenomenon, the reasoning Is well
documented, and a number of theories exist to explain this fact.
[Bjerrum et al., 1957; Lambe, 1954; Lambe, 1958; Mitchell et al.,
1965; and Seed et al., 1959; Hllf, 1975]. Hermann and Elsbury [1987]
present a recent review of theories related to the structure and
properties of compacted soils.
4-22
-------
One theory commonly used to explain the relationship between water
content, dry density and hydraulic conductivity is the flocculated
versus disperse structure theory of Lambe [1953]. In short, clay
particles are relatively flat, and in effect are essentially two-
dimensional. At low moisture contents, and in a relatively loose
state, they generally have a flocculated structure, with nearly random
particle orientations -./Hhin the floes. At higher moisture contents,
clay particles tend to orient themselves preferentially in one
direction along their two-dimensional plane. This Is called
dispersion. It 1s known that for a given compactlve effort, the
tendency of a clay to disperse during compaction Is greater at higher
moisture contents, and that, due to the ability of the particles to
fit much more closely together, lower hydraulic conductivities are
possible at higher moisture contents. At the top of Figure 4.2 Is a
curve representing the degree of dispersion of the clay. This curve
tends to show that for a given degree of compaction, at higher levels
of dispersion, and higher moisture contents, the attainable hydraulic
conductivity 1s much lower than at lower dispersion levels, and at
lower moisture contents.
Figure 4.2 Is an example which, within reasonable limitations,
Illustrates the 'typical1 case for a clay soil. As can be seen, the
moisture content for a minimum hydraulic conductivity Is higher than
the optimum compaction moisture content. Additionally, considerable
variation in hydraulic conductivity results from different moisture
conditions of the soil, given the same level of compactlve effort.
The variation In hydraulic conductivity spans two or more orders of
magnitude, whereas the difference in dry density for different
moisture contents is relatively small. In fact, the ratio of maximum
to minimum hydraulic conductivity 1s about 100, whereas the ratio of
maximum density to minimum density is about l.l, over the same range
of moisture contents, and with the same level of compactlve effort.
[Mitchell et al., 1965].
From- the above, 1t Is apparent that the required moisture content
to meet a compaction criterion is generally dry of the moisture
content required to meet a permeability criterion. Because of the
relationship Illustrated in Figure 4-2, it may be possible, however,
4-23
-------
to use the compacted dry density and moisture content in the fill as
an indicator of hydraulic conductivity. This correlation will
ncnrolly be determined during the Test Fill program. This is discussed
further in Sections 4.3.2.1.11 and 4.5.1.4.
For earthworks for berms and other structural soil components of
the unit, the dry density and moisture content compaction criteria are
most appropriate for specification and conformance testing. For
example, figure 4-3 shows an example moisture-density relationship for
the compaction of a soil. If, for this soil, a specification of 95
percent of the maximum dry density is desired, then it can be seen
that this should be attainable if the moisture content of the soil
falls within the range wmjn(g5-/a) to Umx(95'/,)> for a de9ree °f
compaction that Is consistent with the laboratory test. Therefore, for
the modelled level of compactlve effort, an area of the fill with a
moisture content above wmax(g5y.j or below wmin(g5«/,) will require
moisture conditioning (wetting or drying, as required).
The composition of the soils in the field dictate the required
ranges In moisture contents and compactive effort to achieve the
specified dry density. Although soils vary in constituents,
mineralogy, and moisture condition from one location to another,
several general guidelines ran be applied to field compaction of
soils:
• clays and other fine grained soils, which wou'd typically be
used In the soil component of a liner, often have natural
moisture contents, in cold or very wet climates, between 5 and
20 percent 1n excess of their optimum moisture content; 1n hot
or drier climates, natural moisture contents on the dry side of
optimum are more usual;
• the range of moisture contents over wnich a given degree of
compaction can be attained is generally smaller for fine
grained soils (silt, clay) than for granular soils (sand,
gravel); and
4-24
-------
• gra"j'" soils above the ground-water table commonly have
natjr -oisture contents dry of the optimum roisture content,
•j.;p;--j:-g On local drainage, fines content, and other
considerations.
In any eve't, when a high degree of compaction (say, greater than
95 percent) is required, moisture conditioning, in the form of wetting
or drying, is often required for both fine grained and coarse grained
soils. This poses a considerable constraint on the scheduling and cost
of earthworks, 'ncluding installation of the compacted soil component
of the lining system. The quality control testing to confirm that the
specifications l-jve been attained is therefore an important component
of the Construction Quality Assurance Program. Various well-
established test procedures exist (See Section 4.5.1), ;ich are
reliable and easy to perform. The measurements of dry density and
moisture content do not, in themselves, represent a problem.
Considerable difficulty does exist, however, with the moisture
conditioning ap;roach, and close Construction Quality Assurance is
necessary to con'frm the uniformity and preservation of the drying or
wetting process.
4.3.2.1.2 Weathe- and Climate
Closely related to the quality assurance of compaction are the
cliratic constraints imposed by virtue of the geographic location of
the site. High-quality earthworks (i.e., soils placement and
compaction to stringent specifications) cannot reliably be attained In
below-freezing (O'C or 32°F) temperatures. Seasonal constraints will
therefore llnit field operations in major geographic portions of the
United States.
Preclpitatic^ also Impacts soils placement and compaction. The
tolerable range of moisture contents noted 1n 4.3.2.1.1 is often
difficult to neet in very wet or very dry climates. Excessive wetting
through precipitation and subsequent ponding can result in softening
of both in-place and stockpiled materials. This may necessitate
removal of porticis of the soils that have already been placed and
compacted. Because hydraulic conductivity is more affected by changes
-,-25
-------
i compaction moisture content than is dry density, the moisture
requirements for soil liners are more stringent than for general
earthworks, and hence this problem is considerably more prevalent for
compacted soil components of liners. The roperties of clay soils
which make them particularly effective for low-permeability liners
also make them sensitive to compaction moisture content. Placement
of 'hese clays at moisture contents wet of optimum is typically
required to allow attainment of the specified hydraulic conductivity
criterion. The Test Fill program will be able to identify the
relationship between the compaction criteria and the specified
hydraulic conductivity (see Section 4.4.2.2).
Similarly, excessive drying of the soils due to the hot summer
sun, high winds, and/or climates with low relative humidity can cause
problems not onlv during, but subsequent to placement. Fine-grained
soils such as clays are frequently very difficult to moisten in a
uniform manner. Soils which will meet the low-permeability criterion
for compacted soil liners typically exhibit high plasticity, and hence
a propensity for volume change in response to a change in moisture
content. Consequently, drying-out of these soils after placement and
compaction may result in the formation of desiccation cracks, which
can be as mu as 25 to 50 mm (1 to 2 1n.) wide and up to 300 mm (12
In.) deep. If this condition is encountered, the soil will require
scarification and recompaction over the entire depth of desiccated
soil.
The problems of hot, cold, wet, or dry periods will, to a greater
or iesser deg—'e, prove to be a construction constraint on almost
every project. The Construction Quality Assurance of the earthworks
1s therefore essential, because contractors are often anxious to
compensate for weather-related delays and may be Inclined to relax
their standards in the absence of a stringent Construction Quality
Assurance Program.
4.3.2.1.3 Availability of Suitable Soils
Earthworks soils can frequently be specified to acconmodate those
materials excavated from the site. In this manner, soil materials
4-26
-------
handling can be optimized, and costs minimized. The requirements for
compacted soil components of lining systems, however, are much more
inflexible. As discussed in other chapters, the principal requirement
of the compacted soil liner Is specified as a maximum permeability
criterion (10~f m/s (10"' cm/sec)). Soils which will satisfy this
requirement may be unavailable 1n many locations. Similarly, granular
drainage media within the lining system have a minimum planar
permeability requirement of 10~J m/s (1 cm/sec). This represents a
clean, coarse-grained sand, and availability car. also be a problem.
The Construction Quality Assurance Prograr is a useful mechanism to
ensure that proposed Imported fine-grained soils and granular drainage
materials have properties which will enable them to meet the
specif lotions.
4.3.2.1.4 Subgrade Soils
The CQA Program can ensure that site selection process precludes
the location of landfills, surface impoundments, waste piles, and land
treatment units on sites with organic soils, badly fractured rock or
other undesirable foundation materials. Lining systems In which the
secondary liner rests directly on the recompacted subgrade soil will
be construction-sensitive to the condition of the subgrade immediately
prior to placement of the FML. Regardless of the soil types existing
at the site, the compaction of these soils must be properly completed
1n order to attain the required s oort and surface. In some cases,
soils may even have to be Imported from off-site, 1n order to obtain
specified compaction, and a proper surface and liner support.
4.3.2.1.5 Soils Homogeneity and Layering
In nature, the sedimentary process deposits soils gradually and
preferentially (coarse-grained materials first, fine-gnined last).
Therefore, any given source of clay may exhibit varying properties.
No matter what level of care is taken in placing soils In lifts, there
ma.y be zones of non-homogeneity within a soil lift and at the
Interfaces of adjacent lifts. Lift thicknesses are selected based on
the soils conditions and the nature of the compaction equipment being
used. The desire for * -ck lifts must be weighed against the ability
4-27
-------
to compact the material to a relatively uniform density and
permeability throughout the lift. Consequently, soil lift thicknesses
will tenJ to be in the 150 to 250 irm (6 to 10 in.) range. From
compaction and permeability viewpoints, this provides a "relatively"
homogeneous soil mass, 1n which the degrees of compaction and
permeability are acceptably uniform throughout, and the sizes of soil
clod, ninimized. It should be recognized that, from a permeability
viewpoint, a 1ift-to-11ft interface or interlayer zone is created that
will differ In some manner from the other portions of the fill. For
example, the homogeneity of the soil will be reasonably consistent,
but the horizontal and vertical hydraulic conductivities of the soil j
within this zone may vary. j
j
The potential problem of a preferential flow path, created by a ]
potentially higher lateral permeability along ths layer interface, can
be minimized by the scarification of the top 50 to 75 mm (2 to 3 in.)
of the previous lift, Immediately before the placement of a new lift.
This allows better intermixing of the two layers and breaks up any ; j
desiccated crust (which can form after only a few hours exposure to ;1
sun and wind) on the previous lift. i]
' i
4.3.2.1.6 Disturbance Due to Traffic ;!
- i
Disturbance due to traffic is wholly preventable, but still '• '
sometimes occurs. The concentrated lo?ds imposed by vehicles can
severely rut the surface of compacted soils. Remedial measures I
required will vary depending on the soils, nature of the disturbed i
area, and severity of the damage, and can range from minor dressing, j
to excavation and removal. Control of all construction traffic, and |
the prevention of access to finished lifts or areas, will preclude |
this particular problem from arising.
4.3.2.1.7 Soil Components of Top Composite Liners \
1
It should be noted that soil components of top composite liners
will be typically placed over one or more geosynthetic layers, and
care must be taken to ensure that no damage occurs to the
geosynthetics during the placement and compaction of these overlying
4-28
-------
1
soil layers. As a consequence, it is usually preferable to not specify ;
completion or permeability requirements for the bottc.i-'iost lift of j
the soil component of the top composite liner. It is better to simply j
place this lowest layer with low ground pressure tracked equipment and )
no compaction equipment. Additionally, the specification should j
require the first 11ft over und lying geosynthetics to be a minimum j
of 300 mm (12 In.) thic'~. The specification for soil components of top j
composite liners should therefore require the first lift to be a 1
minimum of 300 mm (12 in.) thick of relatively uncompacted soil. (This ]
lift could even be granular material if available clays are too ]
difficult to place without compaction equipment.) Overlying lifts can j
receive Increasing degrees of compaction and the top lift should be ' ]
fully compacted and smooth rolled to act as the surface upon which the
FML component of the top composite liner will be placed.
4.3.2.1.8 Testing Problems - Field Compaction of Soils
The current state of practice regarding the field ccmpaction
testing of soils Is relatively sophisticated. Consequently, few
problems are encountered with the testing of soils for density
determination. Nuclear density gauges are typically used, and these
devices provide both density and moisture content to depths up to 200
mm (8 in.). These apparatus are simple to use, accurate, and fast.
As a result, the productivity of CQA personnel using this equipment
can be high, producing many tests 1n a shift. But more Importantly,
nuclear density gauges provide CQA personnel with an immediate answer.
Other more traditional procedures, such as the sand cone test,
although reliable, take longer to perform, and require overnight
laboratory determination of moisture content, and are sensitive to the
care taken by, and the experience of, the CQA technician. It Is
appropriate, however, to cross reference and correlate the nuclear
test results with periodic sand cone testing.
4.3.2.1.9 Testing Problems - Laboratory Moisture-Density Tests
Laboratory moisture-density relationships for all soils used at
the site are carried out in conformance with either the Standard or
Modified Proctor tests (see Section 4.5.2.1.1). The moisture-density
4-29
-------
tests provide the 'maximum' (for the given I eve of ccn-pactive effort)
dry Jensity attainable, and the corresponding moisture content. These
tests are relatively easy to perform, anJ pro/ide the basis for the ;
evaluation of the re1 tive degree of compaction of the soil, i.e.,
they are the basis for the in situ tests discussed 1n the previous ;
section. J
Few problems exist with these tests, although frequently some very >
highly plastic clays will have optimum moisture contents very much j
higher than the natural, in-place moisture content, and are difficult \
to work with In this wet condition. In that instance, the performance !
of the test may be very difficult, and may be a function of the I
technician's experience and ability, and laboratory technique. 1
i
In addition, some uniform sandy soils are very moisture content j
Insensitive, and virtually the same density may be obtained over a j
range of moisture contents. In this case, the test is still j
relatively easy to perform, however the interpretation of the results |
is sometimes difficult. in general, however, an experienced 'j
technician performing the test can alleviate or overcome this problem. !
4.3.2.1.10 Testing Problems - Laboratory Permeability Determination
At present, laboratory permeability testi.g for Construction
Quality Assurance purposes is carried out on samples of the compacted
soil liner and sand drainage layers in the absence of generally
accepted In-place procedures which can be performed relatively
quickly. Several procedures exist for the performance of permeability
testing In the laboratory (see also Section 4.5.1.1.3). Constant head
\ tests (In which flow 1s maintained through the soil under a fixed
level of head) and falling head tests (i vhich an initial head of
water Is not replenished as the water flows through the sample) can be
performed in sach of the following apparatus:
• Soil Permeameter;
• Oedometer Cel1;
• Triaxlal Cell; and
• Rowe Cel1.
4-30
-------
Each of these apparatus are fairly well developed as a laboratory tool
for the determination of soil hydraulic conductivity. The nature of
the actual test performed (constant head or falling head) is a
function of the nature of the soil.
A recurrent problem related to laborator testing of this nature
1s the sensitivity of the test result to sample disturbance. Cohesive
soils (e.g., clay) can be collected 1n a 'relatively1 undisturbed
state, however problems related to handling, transport, extrusion, and
sample preparation Introduce an unquantiflable error to the test.
Fairly stringent permeability requirements are applied to clay
components of lining systems (10"' m/s (10~7 cm/sec)) and granular
components of drainage systems (10"1 m/s (1 cm/sec)). As a result,
the test c--"i be properly carried out and a result obtained, but the
representat veness of the test result to the field conditions (the 1n
situ permeability) 1s uncertain. In fact, sample disturbance can
cause tests to err in either direction from the "correct" result.
Cohesionless soils are especially difficult to represent in the
laboratory and, 1n fact, laboratory permeability tests on granular
soils will almost always have to be performed on reconstructed
samples. In this case, clearly, the suitability of the test result as
an Indication of field permeability is in question.
In spite of these problems, this mode of testing is still
acceptable In that the quality control of the test itself 1s very
reliable. In each case, a test result can be obtained which can be
considered to be representative, within an often unspecified margin,
of the permeability of the sample in Us as-tested condition. The
Interpretation of the relationship between the field condition and the
sampled and tested condition remains the problem. At present,
however, pending resolution ;>f the Issues 1n dispute over field
permeability tests, laboratory testing will continue to be accepted,
with some Interpretation required.
4-31
-------
4,3.2.1.11 Testing Problems - Field Permeability Determination
Field testing of soil permeability is currently in the
developmental stage. Several procedures and apparatus have been
developed [Daniel, 1984; Boynton and Daniel, 1985; Daniel et al,
1985], but are not yet In common usage and, certainly, no standards
yet exist. The procedures which have been proposed for the
determination of field permeability (hydraulic conductivity), are
thoroughly documented, but sufficient studies of the accuracy and
relevance of the test results have not yet been carried out.
The results of field hydraulic conductivity tests often yield
higher values than the laboratory tests (by un to as much as two
orders of magnitude, or 100 times) [Daniel, 1984; Daniel et al, 1985].
Considerable discussion over the meaning of this variation has been
carried out. In the case of careful prevention of sample disturbance
(by the collection of block samples, for example) the resultant
laboratory hydraulic conductivity will be representative of the field
condition on a micro-scale. In contrast, field tests carried out on
larger 'sample areas' incorporate the effects of desiccation,
secondary structure, rootholes, etc. In the case of a prepared fill
for a compacted soil Hnr TT which CQA has been provided, many of
these anomalies will be m 'zed, and the assumption of a homogeneous
soil mass, as discussed previously, may be more closely met. In
general, however, the measured laboratory hydraulic conductivity will
tend to be lower than the field measurements of hydraulic
conductivity.
The problem with field hydraulic conductivity testing at present,
and the reason that it cannot be used for quality control purposes in
its present form, is that it can take up to 3 to 4 months to obtain
results which are considered to be representative (due to the time
factors associated with saturation and soil swelling). Among the
methods now available, some apparatus which provide results in shorter
times are being evaluated. The nature of this equipment and these
test methods is discussed in Section 4.5.1.1.4 of this documer1.
-------
Ai a means of reconciling the time dela.,. associated with the
field tests and the potential inaccuracy of the laboratory tests, the
current state of practice calls for the use of field tests on Test
Fills in order to determine the suitability of available materials and
placement techniques, and correlate these data to moisture/density
relationships, followed by laboratory testing of the actual liners
(see Section 4.5.1.1.4).
4.3.2.2 Geosynthetlc-Related Construction Problems
4.3.2.2.1 Manufacturing Quality Control
Manufacturers of geosynthetlcs (including flexible membrane
liners, geotextHes, geonets, geogrids, etc.) provide In-house quality
control of their manufacturing process, In part due to the need to
confirm '.hat the process is functioning properly, and in part to
provide documentation and certification that the properties specified
for the material for the specific project have been satisfied.
Nevertheless, in some cases, substandard materials, which do not
eet the project specifications and must therefore be rejected, have
been shipped to a project site. If a thorough CQA program 1s
Instituted on site, the performance of conformance tests (see Section
4.5.2.1) will Intercept these materials, but the cost of rejection can
be high if construction delays result from re-ordeiing and
remanufacturing of materials. For this reason, CQA of the
manufacturing process can play an important delay and cost-saving role
In the project, in addition to confirming the adequacy of the
materials.
4.3.2.2.2 Fabrication Quality Control
Geosynthetlc products are sometimes prefabricated prior to
shipment to the site, and this may either take place at the
manufacturing facility or at some other location. This fabrication
process typically consists (in the specific case of FML's) of the
seaming together of rolls and portions of rolls into larger panels, to
facilitate placement at the site. Although the conditions under which
4-33
-------
this fabrication seaming takes place are normally much better than 1n
the field (e.g., indoors, flat surface, etc.), this is still a most
critical operation. Fabrication seams are not always tested once the
panels are deployed in the field; therefore the CQA Program should
include the material fabrication process in order to thoroughly review
all of the methods and procedures, and to monitor and document the
testing on these materials.
•..3.2.2.3 Shipping and Handling
Geosynthetics, particularly geotextiles and flexible membrane
liners, are very susceptible tc mage if improperly shipped, stored,
or handled. In addition, their performance 1s, at least in part,
dependent on the continuity of the sheet material, and damage
occasioned by mishandling or other poor practices which will require
repair. Geonets are somewhat more forgiving by the nature of their
function, and damage such as a hole or a tear can more easily be
repaired, or the damaged section removed, without disruption of the
Integrity of the remainder of the roll. At the other extreme,
geogrids which are required for reinforcement must be continuous in
the direction of the applied load. Consequently, damage to rolls of
geogrid will normally result in rejection of the entire roll ^unless
an alternate location requiring a shorter length is available).
Geotextiles and FHL's must be protected from dust, dirt, moisture,
and ultraviolet radiation during shipping, storage, and handling.
Protective wrac ings should be provided, in order to alleviate such
problems. Geonets must be protected from dust and dirt and protective
wrappings should therefore be provided for them also.
Handling problems usually relate to damage caused by handling
equipment, such as forkllfts or loaders, which can puncture or tear
the rolls while picking them up and transporting them. In addition, a
common source of geosynthetics damage 1s abrasion caused by dragging
rolls or panels of material when positioning them for deployment.
This can cause severe damage, and frequently can wear completely
through one or more layers on the roll. Damage of this nature may not
be repairable, in which case the iamaged material should be rejected.
4-34
-------
Storage of the geosynthetlcs on site, prior to deployment and
Installation, requires a location In which damage or contamination due
to wind, dust, dirt, rain, and ultraviolet exposure is not possible.
Particularly in the case of geotextiles which are to pe,.orm filter
and drainage functions, contamination and clogging by dust or mud may
well be Irreparable. Most other geosynthetics (e.g., FML's, geonets,
and geogrlds) can usua'ly be cleaned by washing.
The provision of Construction Quality Assurance for these
components of construction can effectively minimize the incidence of
these problems.
4.3.2.2.4 Sheet Material Defects
In general, sheet material defects will normally be Intercepted at
the manufacturing stage, 1f good quality control Is present, and
especially If Construction Quality Assurance is provided for that
operation.
There are several types of flaws or defects which are found in
homogeneous sheeting material for flexible membrane liners, and which
are all a function of some Irregularity or eccentricity of the
nanufactuMng process. These Include:
• plnholes, which are very small holes and are relatively rare,
likely caused by moisture or Inclusions 1n the system;
« holes, which can be up to 25 mm (1 1n.) 1n size which could be
caused by moisture, inclusions, or other manufacturing
Irregularity;
• blisters, which can be caused by the presence of vapors during
the manufacturing process;
• craters, which are "holes" which have not completely penetrated
the sheet, and which can be created by foreign matter;
4-35
-------
• small bumps, which could be caused by excess concentrations of
carbcn black;
• Insufficient thickness, caused by the feed process; and
• scratches or gouges, caused by Impact or contact with external
objects, which can result In weak points 1n the sheet.
Similar defects can occur In geotextiles. In addition, instances
have been recorded In which needle-punched geotextiles have damaged
the FML, or been damaged themselves, by the presence of broken needles
within the fabric. These needles may be detected visually, and could
possibly be found by the use of a metal detector used to sweep the
geotextlle surface.
In general, because of thd structural nature of the materials,
flaws In geogrld or geonet "sheet" materials would normally be
relatively easy to detect visually.
It should be noted that flaws Inherent 1n the material which do
not manifest themselves by a visual Indicator will be difficult to
detect In the field. In fact, flaws such as insufficient thickness
can only be detected at the edges of a roll or panel, using a
micrometer or similar form of caliper. Thinness in the middle of a
sheet cannot be measurably detected, axcept by sampling. Interception
of these problems during the manufacturing process by observation, and
close control of the extrusion process, would be the superior means of
detection. In any event, the best assurance of detect,on of these
types of flaws 1s to provide CQA at both the manufacturing and
deployment stages of construction of the lining system. Even so,
nondestructive test procedures are not presently available to al.low
really effective testing of the entire sheet surfaces. Of course, the
CQA personnel can carry out sampling and destructive testing of the
sheet material 1n areas which appear to be of concern, but good
experience and judgment are required of the CQA pers.mel
-------
4.3.2.2.5 Seaming .-Yocedures
Frequently, the problems with seaming the geosynthetic materials
dictate the amount of destructive testing that ; carried out, and
also highlight the value of CQA. Problems related to the seaming
operations Include:
• Flexible Membrane Liner Seaming
The field seaming of the FML components of lining systems 1s
perhaps the single most impprtant activity for which the CQA
Team provides monitoring and documentation services. There are
many methods used to seam FML's, but, for purposes of this
document, those methods and procedures utilized for the field
seaming of polyethylene (PE) flexible membrane liners will be
emphasized.
There are two processes for seaming PE flexible membrane liners
currently 1n common use. These procedures are:
fusion welding, with either a single track, or double track,
In which the bond of the two FML sheets is achieved by the
heating of the two sheets, and upon melting, the application
of pressure to produce a homogeneous joint upon cooling and
solidification; and
extrusion welding, consisting of an extruded hot
polyethylene bead, which 1s either sandwiched between the
sheets, or placed as a fillet weld at the edge of the upper
sheet (extrusion welding is also used for repairs and spot
patching).
The seaming system used Is normally either a proprietary one,
or a Manufacturer/installer-specific system. Regardless of the
system, the seam that >s prepared must exhibit, at the least,
the following characteristics:
4-37
-------
the bond, or continuity of the seam, must be continuous over
the entire length of the seam, and any discontinuity must be
considered to be a flaw;
the seam strength, or integrity, must be consistent and be
within a tolerable statistical variation from the strength
of the sheet Itself; and
t*"> seam must be able to be formed in the field, under less-
than-1deal conditions.
Under adverse weather conditions It 1s difficult to obtain a
field seam which satisfies all three of the above-noted
criteria. In order to determine the seam continuity and
Integrity, they are subjected to continuous nondestructive
testing and selective destructive testing (see Sections
4.5.2.1.2 and 4.5.2.1.3). In general, the difficulties
experienced with the formation of these seams, which result in
Inadequate seams Include:
1n general:
moisture on the liner sheets can result in bubbles
within the seam, which might result in low strength
values and/or a bond break;
dirt or other foreign objects on the liner sheets
can result 1n their Inclusion within the fused seam;
a dlsc-ntlnuity may be formed that results 1n low
seam strength; and
very high sheet temperatures will result In many
folds in the sheet due to thermal expansion, and
cause the formation of 'Hshmouths' along the edge,
resulting in the need to cut and patch before
seaming, as well as special seaming techniques.
4-38
-------
for fusion welding, In particular:
Improper heat settings can result in lack of fusion
due to the material not melting (setting not hot
enough for the conditions), or can result in low
strength values, or burning through, or Inconsistent
fusion (setting too hot for the conditions);
Improper sheet overlap can result 1n trimming and
consequent waste of material (If too much overlap,
such that the hot wedge cannot hook over the lower
FML), or loss of the double weld (if too little
overlap, such that the track of the double weld is
wider than the overlap); and
Improper pressure settings can result in the
squeezing out of the softened material (usually in
hot weather, 1f too much pressure 1s applied),
mechanical damage to the sheeting material (usually
1n cold weather, 1f too much pressure Is applied),
or lack of fusion (if not enough pressure is
applled).
for extrusion weeding,_1n particular:
Improper heat settings can result In a lack of
fusion (setting not hot enough), or can result in
uneven bond or burning throuah (setting too hot);
working in moist (e.g., morning fog or dew)
conditions can result in moisture contaminating the
extrudate supply, particularly 1f partlculate resin
1s utllIzed; and
very low sheet temperatures can Inhibit the proper
fusion by cooling the extruded material too quickly.
4-39
-------
Naturally, the combination of all of the site-specific
conditions will tend to govern the incidence and degree of
these potential problems.
The provision of high quality CQA for the seaming process 1s
absolutely necessary. Many of these potential problems may not
be detected by nondestructive test procedures, and It 1s
necessary for the CQA Monitors to document the presence of
contaminants or foreign objects, lack of sufficient overlap,
moisture, or Inadequate settings (via trial seams).
^eotextile Seaming
Geotextlle materials, whether used as transmission media,
filters, separators, or cushions, should be seamed in the
field. This seaming Is achieved by sewing, fusion, or gluing.
Various shapes of overlap and types of stitch or solvent are
utilized, but there is usually no mechanical property criteria
which must be satisfied to accept or rejact the seam. Many
Designers believe that sewing 1s the only acceptable means of
seaming geotextlles. In that case, the CCA Monitor need only
confirm the existence of the seam (in addition to the normal
monitoring requirements - no dust, mud, damage, etc.). In the
case of geotextile filters, fusion seaming 1s considered
Inappropriate because of the propensity for burning through the
geotextile. Solvent seaming 1s usually not acceptable In
hazardous waste management units,
Other Geosynthetlc Seaming
Geonets usually do not require seaming as there is usually no
design load Imposed within the plane of the material. Tying of
geonets together to hold them 1n place and ensure continuity 1s
usually the sole extent to which jolting 1s attempted.
4-40
-------
Geogrlds require Integrity In the-direct ion of applied load
(e.g., downslcpe). For this reason, gecgrids should be placed
in a single piece In this load direction, and it is preferable
that they not be joined, although methods are available.
CQA efforts will, 1n these cases, be devoted to conformance
testing, placement methods, and the documentation of continuity
In the required orientation.
4.3.2.2.6 Seaming Constraints
Aside from those factors Identified in Section 4.3.2.2.5 with
respect to seaming procedures, other constraints apply with regard to
the seaming of flexible membrane liners In particular. Weather, as
with soils Installation, poses the single greatest constraint on the
construction operations.
Hot weather (I.e., ambient temperature 1n excess of 40°C (104°F)
measured 150 mm (6 In.) above the liner) can be detrimental to
flexible membrane liner Installation operations, In that seaming
equipment can damage the sheet materials. In addition, due to the
black color of the FML, surface temperatures under hot weather
conditions can reach 80°C (176°F), which makes work by Installation
personnel virtually impossible.
Cold weather (I.e., ambient temperature below 5°C (40°F)) makes
seaming very difficult due to the seaming problems cited :n Section
4.3.2.2.5; and special techniques, as well as considerable operator
skill, are necessary 1n order to achieve adequate seams. In this
particular case, with cold weather seaming, very thorough CQA Is
required for the seaming operation.
Seaming cannot be undertaken during any type of precipitation
unless special measures are takpn to protect the seams from moisture,
humidity, and accelerated coo!1;d.
4-41
-------
4.3.2.2.7 Contamination
One of the greatest problems with ceosynthetics installation is
the occurrence of prolonged periods of rainfa while the system is
only partially completed. This Invariably leads to some form of
contamination of the lln'-:g system with respect to the Intrusion of
soil into the LOCRS at ut. ealed edges (not to mention damage to the
underlying clay). In some cases, geotextile filters can be Irreparably
damaged. In any event, removal of some portions of the system, In
order to obtain access to the contaminated materials is necessary.
Washing or replacement of geonets and geotextlles 1s required, so that
these components can operate in the manner Intended. In extreme cases,
clogging of drainage pipes and sumps could occur.
This Is certainly one of the very important roles of the CQA
Program, In that the degree of contamination can be examineJ and
evaluated, and the repair documented for reference.
4.3.2.3 Qualifications of Personnel
The Construction Quality Assurance personnel must be capable of
recognizing, Interpreting, or otherwise identifying all of the
potential problems which have been discussed herein. In order to
fulfill this requirement, it 1s imperative that 'he CQA personnel be
familiar with every aspect of the installation of geosynthetics and/or
soils. In addition, these personnel must be familiar with the
properties of each of the materials, as well as the tests which are
required to confirm these materials properties. Finally, the CQA Team
must have a strong backup engineering capability available in order to
properly fulfill this role. This engineering capability, whether on or
off site, must be thoroughly familiar with all aspects of the design
of geosynthetlc lining systems, and the properties and behavior of
polymeric and soil ma rials.
4.3.2.3.1 CQA Officer
The Construction Quality Assurance Officer should hold a B.S.,
M.S., or Ph.D. engineering degree and be registered as a Professional
-42
-------
Engineer (P.E.) In accordance with the laws of the state in question.
He should be experienced with geosynthetIcs, Including flexible
membrane liners of the '.ype specified on the project, geonets,
geogrlds, and gcotextiles. He should also ba experienced in quality
assurance, particularly for the Installation of geosynthetlcs and
soils materials. The CQA Officer should be experienced in the
preparation of quality assurance documentation, Including quality
assurance forms, reports, certifications, and manuals.
4.3.2.3.2 CQA Manager
If the CQA Officer is not resident full-time on the site, he
should designate a CQA Manager, who would be on site continuously, and
should be specifically experienced In the Installation of
geosynthetlcs, Including flexible membrane liners of the type
specified for the project, geonets, geogrlds, and geotextiles. In
addition, he should be experienced in the placement and compaction of
soils, particularly compacted soil components of lining systems. He
should be specifically tialned and certified by the CQA Contractor in
the duties of a CQA Manager, and should be experienced In the quality
assurance documentation required for Installation of geosynthetics and
soils. This CQA Manager performs those duties of the CQA Officer which
Involve on-site responsibilities.
4.3.2.3.3 CQA Monitors
The CQA Monitors should be fully trained and/cr experienced in the
proper documentation of all CQA activities, In addition, CQA Monitors
Involved with the monitoring of soils placement and compaction should
be fully trained and experienced 1n that aspect, and should be
qualified and/or licensed to carry out the nondestructive testing
required for the quaVty control of the placement of those materials.
Similarly, CQA Monitors Involved with the monitoring of the placement,
seaming, and testing of the geosynthetlcs should be trained In those
aspects of CQA.
1
I
I
4-43
-------
4.3.2.4 Documentation of Problems
Ihe literature documenting the problems wit'i geosyntliet Ics, in tr.e
foim of case histories of failures, is somewhat sparse, because there
has been a disinclination for the industry to document Its problems.
As a consequence, there have been reservations by some engineers about
the effectiveness of these polymeric materials (also caused by human
Inertia with regard to new methods). Naturally, there Is also a legal
question with regard to the documentation of litigious Issues. Several
papers have been produced, however, which address the types of
problems which have arisen with geosynthetics, and discuss the nature
of the problems, wUh measures that can be taken to preclude those
problems through design modifications or different construction
practices [Bass et ")., 1985; Ghassemi et a!., 1986; Giroud and
Goldstein, 1982; Giroud, 1984f; Giroud and Fluet, 1986; Morrison et
a)., 1981; and US EPA, 1986b]. The documentation of problems is
discussed further in the context of the benefits of CQA in Section
4.3.4.3 of this document.
Soils problems are generally well documented, and in fact are
cover 'd in the geotechnical engineering programs at most universities.
There are, however, a few aspects of the design of these land disposal j-
unlts that distinguishes t n from other engineering structures. In \
particular, the emphasis on permeability, as opposed to strength and i
density, as an installation criterion Is not co'vmon. The solution of !
problems that are encountered is therefore dependent on the \
qualifications of the Construction Quality Assurance Team. These j
problems are documented in many sources, including [Turnbull et al., j
1956; Hilf, 1957; Seed et al., 1957; Lambe, 1958a; Lambe, 1958b; Seed j
et al., 1959; Lowe et al., I960; and Gibbs et al., I960].
Considerable work has been carried out since the era of these
references, however, they remain the cornerstone of research into the
behavior of compacted clays. The reader is referred to US EPA
[1986b], US EPA [1986c], and US EPA [1987] for contemporary
discussions of the construction of, and CQA requirements for,
compacted soil 1 iners.
4-44
-------
4.3.3 M-itorlaJs..Con si derations
4.3.3.1 Soils Materials Cons eratlons
The soils components of land disposal units differ from the
geosynthetlc and otht, manufactured components In that, for the most
part, they originate on site or near site. In fact, wherever
possible, all of the soils used at the site should be from the site,
from the viewpoint of handling, transport, and cost. Sometimes,
however, some or all of the soils components may originate off-site.
From the viewpoint of the Construction Quality Assurance Program,
It will be necessary for the Contractor to work with the soils, making
the necessary changes (e.g., wetting or drying) In order to conform to
the specifications. This places considerable demands on the CQA Team.
Even though soils placement and compaction are better understood
than geosynthetlcs placement and seaming, the operations Involving
Construction Quality Assurance of the soils components of a unit can
be more drastically affected by external factors, such as the weather,
than for the geosynthetlcs.
4.3.3.2 Ge synthetics Considerations
4.3.3.2.1 Manufacturing Considerations
There are obvious issues affecting the Cons.,uct1on Quality
Assurance operations, which relate to the manufacture of the
geosynthetlcs (including flexible membrane liners, geotextlles,
geonets, and geogrlds). ,hese Issues are discussed as follows:
• Feedstock, Process, and Additives
The resins used to manufacture flexible membrane liners
originate from a relatively small number of suppliers. As a
result, there is good control of the quality of the feedstock
for the FML's. There are, however, considerable differences
between the different resins. These differences will, of
4-45
-------
necessity, tend to make specification of base resins a
significant Issue.
* ln-Plant Quality Control
The processes by which the flexible membrane liners are formed
can be very closely controlled, and it can normally be expected
that the quality of the sheet would be very uniform.
Nevertheless, substandard materials are delivered to sites on a
regular, although not frequent basis, which indicates that the
system 1s not without its shortcomings. Although it is often
not possible to conduct CQA during the manufacture of tne site-
specific materials, Construction Quality Assurance can be
carried out during the manufacture of similar materials (and
other geosynthetlcs, to a lesser degree of Importance), thereby
comprising a spot check of the process In general, as well as
1n-plarit quality control procedures.
4.3.3.2.2 Fabrication Considerations
The Issues which are of particular concern with regard to the
fabrication of the rolls of geosynthetlcs into panels are generally
the same as those for the manufacturing process. The two of primary
concern are the quality control of the 1n-plant seaming operations,
and the conditions under which the work 1s carried out.
• In-Plant Quality Control
The fabrication of the rolls of materials Into panels 1s
primarily applicable to the synthetic components of the lining
systems. The apparatus and procedures used In the fabrication
process are typically the same as are used In the field for
field seams. The monitoring of the operations associated with
this seaming should be as thorough as the field requirements,
and ensure that all personnel and apparatus are 1n conformance
with the Construction Quality Assurance Plan and the project
specifications for those activities.
4-46
-------
" Operating Environment
As an extension of the above point, even though the operating
conditions for fabrication may be much more controlled than at
the site (e.g., indoors), the degree of quality assurance
required for the fabrication should justifiably be similar in
scope. The external constraints such as the weather are more
controlled for a fabricating plant, but the variables
representing the problems which can arise are not different.
4.3.3.2.3 Transportation and Handling Considerations
The transportation and handling of the geosynthetics,
particularly the FML's and the geotextlles, are Important to the
preservation of the quality of the materials, since they are extremely
fragile, and susceptible to damage, even in cases where proper care is
being exercised. In general, specifications are written In such a
manner that 1t Is up to the Manufacturer to see that care is taken
during handling, but often the Transporter Is not conscious of the
Importance of protection of the materials. The specifier and the CQA
Contractor should ensure that the handling criteria for the materials
are specified and that these requirements are met.
4.3.4 Benefits *f Construction Quality Assurance
4.3.4.1 Nature and Description of_the Benefits
The benefits of a detailed CQA Program are often Intangible, and
1n fact, as with any QA-related service, If the CQA Program 1s
properly carried out, the benefits that can be directly attributed to
that program may not be apparent at all. There are, however, some
obvious benefits of the program that are real, regardless of the
specifics of the work. These are discussed 1n the following sections.
4.3.4.1.1 Fewer Leaks
The fundamental goal of EPA's hazardous waste management
regulations 1s the protection of human health and the environment. The
4-47
-------
strategy for the achievement of this goal has been to set a 'no
migration' lining system goal for land disposal units. The EPA
recognizes, however, that this 'no migration' goal Is not always
achievable. Through the EPA's 'liquids management strategy' and
through the use of BOAT for double Uner systems, It Is believed that
waste management units incorporating double Uner systems can come
very close to the 'no migration' goal. As discussed In other Chapters,
EPA's liquids management strategy has two goals: minimize leachate
generation, and maximize leachate removal. Construction Quality
Assurance of the earthworks and lining system Installation applies
directly to the second of these goals. The higher the quality of the
eart^^orks and lining system, the closer the attainment of this goal.
It Is possible, with proper design, construction, and CQA
procedures, to control the leakage through a lining system and retain
It within the closed system. One purpose of the CQA Program Is to
minimize the numbers of leaks which can occur through a given liner,
and thereby maximize the security of the overall system. The detection
of problems, by the careful monitoring of all Installation operations,
which would otherwise have gone undetected, will result, in most
cases, 1n a lining system that meets the EPA goal of minimized
leakage, thereby preventing the migration of waste materials through
the system, and benefiting the health and safety of the people In the
are*.. Simply stated, EPA believes that 1t is not possible to construct
a liner which does not leak; however, with good design, construction,
and CQA, 1t 1s possible to construct a lining system which does not
measurably leak and thereby meets the EPA goal of mlnimlzfrg ''akage
Into the environment.
The consequences of a hole 1n a given Uner have been discussed 1n
previous sections of this document. The benefits of detecting and
repairing a flaw which would become a leak are therefore apparent.
There exists IHtle published documentation, at present, which
Indicates the direct benefit of Construction Quality Assurance, 1n the
context of Identification of the proportion of flaws that are detected
by the CQA personnel, relative to the flaws which were detected and
would have been repaired by the Installer anyway. However, those data
which are available clearly support the benefits of CQA. Based on
4-48
-------
experience, 1t is reasonable to assume that the Incidence of leaks
will be between one and two orders of magnitude lower in a project for
which intensive quality assurance has been provided. In other words,
If a flexible membrane liner, for whicn quality assurance c.f the
Installation was not provided, exhibits leaks at a frequency of 1
every 30 m (1 per 100 ft.) of seam, the same Installation, If
Intensive quality assurance had been provided, would have exhibited
leaks at a frequency of 1 per 300 m (1 per 1000 ft.J of seam to 1 per
3000 m (1 per 10,000 ft.) of seam.
This quantification of the benefit 1s Important for several
reasons. First, the cost of Construction Quality Assurance 1s high,
representing an appreciable proportion of the facilities cost [Giroud
and Fluet, 1S86]. As with any CQA-related activity, the product which
Is received Is not visible. The Owner/Operator must therefore
understand the benefits of this cost, In that this additional capital
cost can be shown to reduce future maintenance costs, and 1n the case
of leakage resulting 1n decommissioning and repair, can reduce or
eliminate future capital costs and costs associated with loss of use,
assuming repair to be a viable alternative. Additionally, good CQA
greatly diminishes the liability of all concerned parties, including
the Designer, Manufacturer, Fabricator, Installer, and Owner/Operator.
Other benefits of Construction Quality Assurance are not
necessarily quantifiable, but may be equally as beneficial as the
monetary Issue of future savings These benefits are extremely
Important 1n the decision-making process, and so are briefly discussed
In the following sections.
4.3.4.1.2 High Confidence In the Integrity of the Unit
One benefit of CQA is a high confidence level by all of the
parti.j that the system will perform as it is Intended. In the early
years of FML's, the Industry had several failures, which resulted 1n a
lack of confidence by some In flexible membrane liners. It 1s
extremely likely that almost all of those failures could have been
prevented by better design and/or the Institution of programs of
Construction Quality Assurance (including CQA of the design). Within
4-49
-------
the context of this document, CQA new includes:
• quality assurance of manufacturing, which would identify
manufacturing defects, and would encourage the manufacturers to
initiate very good in-plant quality control procedures;
• quality assurance of fabrication, which would detect flaws In
factory seams;
• quality assurance of construction, which would detect most of
the leaks which lead to failures; and
• quality assurance in the post-construction phase, which can
prevent catastrophic failures by the anticipation of failure to
the extent that protocols exist for the response to a potential
problem, Identified by the monitoring system and program.
Some failures of structures which were lined with flexible
membrane liners 1n the early days of the industry have been
documented. Construction Quality Assurance Programs, properly
performed, would almost certainly have prevented these failures. There
are, of course, other circumstances and mltlga' ">g factors Influencing
Individual cases, but as a general statement, the benefit of
Construction Quality Assurance Programs 1s that they would have
prevented the overwhelming majority of those failures. Furthermore,
most of the remainder of the problems would have been prevented by CQA
of the design.
The high confidence level that we are therefore given by the
addition of the Construction Quality Assurance Programs to the
construction of these systems 1s a definite benefit. Confidence in the
system, by all of the parties Involved, will clearly assist 1n the
provision of a better product. If, 1n future, CQA is extended to
Include the design of the lining system, the level of confidence will
Increase even more.
4-50
-------
4.3.4.1.3 High Confidence In the Public's Eyes
Another benefit of Construction Quality Assurance is an increase
In the confidence of the public 1n relation to the Integrity of the
unit. This Is always a problem, particularly In the Immediate vicinity
of the facility. The demonstraclon that the other benefits of
Construction Quality Assurance are real (i.e., fewer leaks, and
containment and control of leaks), will increase public confidence
that the potential adverse consequences, which would affect the people
directly (in particular, their health), will not arise. This public
confidence In turn will be of considerable intangible benefit to both
the Owner/Operators and the responsible Permitting Agencies.
4.3.4.1.4 Benefits After Construction
One of the greatest tangible benefits of the provision of a
Construction Quality Assurance Program is the detailed documentation
of the Installation, wtr'h Is of considerable benefit in the diagnosis
of any post-construction problems which may arise. Complete
documentation should be provided to include, at the least, the
following:
• documentation of all seaming activities, including ambient and
apparatus temperatures, personnel identification, times,
location and length of seams, etc.;
• documentation of all nondestructive testing activities;
• documentation of all sample locations for destructive testing,
test results, and actions taken;
• documentation of all repairs carried out, including locations,
type of repair, dates, and confirmation of nondestructive
testing of the repair;
4-51
-------
• photographic documentation of all activities; and
• preparation of record ' vlngs for the lining system
Installation, Illustrating che location and nature of all
repairs, construction details of the system, and any components
of the system which differ from the project plans and
specifications (1n this regard, these are not as-built
drawings,.but rather a record of the sequence and activities of
the construction).
4.3.4.1.5 Benefits to the Owner/Operator
The benefits of a thorough, well-documented CQA Program are
considerable. Notwithstanding the regulatory requirements and
benefits, in the context of the Owner/Operator, these benefits can be
described as follows:
• greater assurance that the lining system will perform as
designed;
• reduced costs of maintenance or remedial work during
operations;
• reduced cost of management of the unit, due to continued
satisfactory performance;
• reduced amortized maintenance cost over the lifetime of the
unit;
• reduced time required for issuance of operational permits since
regulators have an Increased level of confidence; and
« reduced liability.
4-52
-------
\
4.3.4.1.6 Benefits to th«. Other Parties
In the context of the other parties responsi'-'e for the supply of
the materials and services, these benefits can be described as
follows:
• greater assurance that the products meet the specified
requirements, which reduces warranty cost for the
Manufacturers, but also enhances the Image and reputation of
the Manufacturers, ensuring long term growth and prosperity;
• lower production and Installation costs due to less rework,
repair, replacement, etc.;
• better productivity for all parties;
• reduced risk of litigation for all pa.ties, resulting 1n
reduced legal costs;
• better communications between all parties, which helps
Intercept potential problems on a timely basis; and
• better marketability of the products of the Manufacturers, and
of the services of the Designer, due to the better quality of
the product.
4.4 S:. PE OF CONSTRUCTION QUALITY ASSURANCE PROGRAM
The proposed Liner/Leak Detection Rule currently addresses only
Construction Quality Assurance. However, we believe that the CQA
process should extend back to the pre- nstruction design stage of a
project.
4.4.1 Pre-Constructlon Stage
The pre-constructlon phase of the CQA Program Includes all of the
activities which will affect, directly or Indirectly, the construction
of the unit. Consequently, although not usually perceived to be
4-53
-------
components of a project, which would be subject to review from the CQA
personnel, this 1s a very important part of the program, directed
toward anticipation and prevention of situations which could present
problems during construction and operation.
4.4.1.1 Design
The design of tv.c un'L should undergo considerable QA review. The
design should be p r reviewed, and the optimum source for this review
1s the CQA Officer, since he/she is responsible for the quality of
implementation. For this reason, the C;A Officer must be qualified and
experienced 1n the design of composite soils/geosynthetic lining
systems, as well as the soils components of the structures. This peer
review must be carried out in detail, and include the review of all
design calculations for soils and geosynthetics, such as sizing of the
drainage layers to accommodate the anticipated volume of leachate,
soils factors such as stability considerations, and the final plans
and specifications.
In addition, at this stage of the pre-construction activities, the
detailed, site-specific Construction Quality Assurance Plan (Section
4.2) should be prepared by the CQA Contractor or the Designer for both !
soils and geosynthetic components of the system. Whoever (CQA j
Contractor or Designer) prepares the CQ.% Plan, the other should j
provide QA peer review. :
4.4.1.2 Materials Specifications j
i
The CQA Contractor should review the desig- specificat ns for the ;
lining system materials, Including bot). the soils and the j
geosynthetics. This process should include the testing and evaluation 1
of the candidate materials on the basis of chemical ccnvjatibility with \
the specific contained wastes, as well as the mechanical and/or 1
hydraulic characteristics required to satisfy the design, and the
constructability of the materials.
Soils materials specifications should be based on the
geo- ;hn1cal evaluations for the site, and the testing which has been
peri.rmed on the on-site and borrow soils for the system. Peer review
4-54
-------
of the testing ar.d evaluation of soil properties can be carried out 3t
this stage.
The final specifications for the soils and geosynthetic materials :
can be conducted as part of this task, if not previously performed in
conjunction with the peer review of the final design. ' :
!
4.4.1.3 Materials Procurement j
The Manufacturer's quality control documentation which must be j
submitted with the bids should be reviewed by the CQA Contractor to I
ensure conformance with the materials specifications. j
j
The Independent Testing Laboratory should be selected by the CQA . j
Contractor, on the basis of quality of work, experience, capabilities, j
and data turnaround. i
i
4.4.2 Construction Stage I
4.4.2.1 Site Preparation and Foundations j
The site preparation includes all of the earthworks required to j
develop the unit. No geosynthetlcs Installation is conducted in this j
particular phase, unless a geosynthetic pressure relief system 1s '
Installed. Construction Quality Assurance of the site grading and ]
development 1s necessary, including the subgrade preparation,
earthwork operations, and temporary roads or struct, es required for !
construction purposes. Important steps in subgrade preparation to •>,
ensure a structurally stable foundation should Include excavation, j
placement, and compaction of soil lifts, nmbankment and slope j
construction, surface smoothing and soil sterilization. 1
i
Quality control testing, consisting in part of the selection of |
samples for laboratory testing for the development of compaction j
stanc'^rds (see Section 4.5.1.1.1) is required for all soils placement i
at tht '.ite. In addition, nondestructive in place density and moisture
content tests should be carried out by the CQA Contractor to monitor
the acceptability of the soils as placed, and to document the tests so
4-55
-------
that there is an accurate record of the opcrati:ns, recorded as they
are being undertaken.
In conjunction with the monitoring and documentation of the soils
placemo~t and compaction operations, the COA Contractor should also
collect undisturbed samples frcm compacted fills for laboratory
testing to ensure conformance with the plans and specifications.
4.4.2.2 Dikes
A dike in a waste management unit functions as a hydraulic barrier
as well as a retaining structure, resisting the lateral forces of the
wastes, liners, and leachate collection and removal systems. A dike
1s also the above-ground extension of the foundation, providing
support to the unit components above. In addition, dikes can be used
to create separate cells for different wastes within a large landfill,
surface impoundment, or waste pile unit. Dikes, therefore, must be
designed, constructed and maintained with sufficient structural
stability to prevent failure.
The CQA Program for dikes can be used to ensure that completed
dikes meet or exceed design criteria, plans, and specifications.
These activit.es may include examining the prepared dike foundation,
monitoring fill materials, placement and compaction, construction of
drainage systems and implementing erosion control measures.
Materials to be used for dike construction should be monitored to
confirm that they are the same as those specified by the design and
that they are uniform, so that no unsuitable materials are included in
the dike. A test fill may be considered to verify that the specified
nil dry density, moisture content, and strength can be obtained with
reasonable compactive effort.
4.4.2.3 Compacted Sojl Uners
The compacted soil component(s) of the lining system is one of the
most important parts of the soils Construction Quality Assurance
Program. As part of the CQA Program a Test Fill should be required and
4-56
-------
the in-place hydraulic conductiv r.y of the Test Fill should be
measured. The Test Fill should be constructed using the same borrow
soil, compaction equipment, and construction procedures as proposed in
the unit. A test fill should be required because laboratory hydraulic
conductivity tests frequently overestimate the actual field hydraulic
conductivity [US EPA, 1987]. A field hydraulic conductivity test on
the Test Fill is necessary to confirm that the materials and
procedures used 1n the field will result in a compacted soil liner
with a hydraulic conductivity of 1 x 10"' m/s (1 x 10"' cm/s) or
lower. The Test Fill program will establish the optimum methods and
procedures (e.g., number of passes, moisture conditioning required) to
follow to meet the minimum hydraulic conductivity criterion.
The CQA Contractor will provide the Construction Quality Assurance
of the placement and compaction of compacted soil liners, including
sampling for laboratory testing, and nondestructive field density and
moisture content tests required to monitor the acceptability of the
soils as-placed, and to document the tests so that there 1s an
accurate record of the operations, recorded as they are being
undertaken.
The acceptance criterion for the compacted soil component of the
lining system is a saturated hydraulic conductivity criterion (i.e., <
10"' m/s (10~7 cm/sec)). Until th; laboratory and/or field testing
program provides results which satisfy this criterion, the compacted
soil liner component cannot be accepted. The CQA Contractor is
responsible for the monitoring of all testing carried out on or off
the site, and must document the results of this testing, as well as
all measures taken to rectify non-conforming test results. As
previously discussed (Section 4.3.1.1.9), laboratory testing of low-
permeability soil samples for permeability characteristics is
sensitive to sample disturbance and simulation of field stress
conditions. The CQA Contractor must ensure that the collection,
handling, and transportation of these samples is carried out in such a
manner that the disturbance of the samples 1s minimized, and that the
results which are obtained are as representative as possible of the
field permeability of the compacted soil layer.
4-57
-------
•' 1.2.4 FMLs_and Other Geo-yithetics
The Ccnstr'jction Quality Assuranc2 of the r. .synthetics components
of the lining system includes the monitoring of all activities by the
Installer's personnel and CQA Monitors. This is absolutely essential
because there are potential sources of problems at every stage of
handling, deployment, seaming, nondestructive testing, and destructive
testing.
Direct monitoring of these to involves the observation of the
activity as 1t 1s carriad out, recc -jing the relevant data, and proper
documentation that the activity was consistent with the plans and
specifications, as well as good construction practice in general. The
following specific tasks must be monitored, at the least.
4.4.2.4.1 Delivery and Conformance Checking
As the materials arrive at the site and are unloaded, the CQA
personnel are responsible for noting that the handling of the
materials did not appear to damage them in any way, and document any
damage which may ~ccur due to accident (e.g., dropping a roll) or poor
practice (e.g., dragging a roll to move it). It is the responsibility
of the CQA personnel to check the documentation that arrives with the
geosynthetics materials (i.e., packing notices, roll labels and
identification) to ensure that the correct materials have been j
received (e.g., correct gauge thickness shown on the flexible membrane •
liner roll label). Lastly, it is the responsibility of the CQA
Contractor to collect samples for conformance testing at a frequency <
specified in the specifications and/or Quality Assurance Plan for the j
site. The CQA personnel are also responsible for the packing and !
shipping of these samples to the Independent Test Laboratory, with |
appropriate instructions. Conformance testing by the Independent Test j
Laboratory should include physical properties and testing to confirm
that the polymer and additives used to manufacture the sheet are as
specified.
4-58
-------
4.4.2.4.2 Deployment and Visual E*3-inat1on
The CQA Monitor should f — st ::serve and test the compacted soil
liner or other material uvon -mch ".a FML or other geosynthttic will
be placed to confirm that tre s.'i or material has been properly
compacted and 1s smooth and ready for placement of the FML or other
geosynthetic. When the rolls cr p=-els of FML or other geosynthetics
are deployed (i.e., unrolled), tn? CQA personnel should undertake a
visual examination of the entire sjrface of the roll or panel, to
confirm the orientation and lccat::n of that roll or panel, and to
detect any flaws or damage. Any fU^s or damage which are encountered
should be appropriately marked for repairs, and it 1s the
responsibility of the CQA Contractrr to ensure that they are noted,
documented, and that the appropriate parties are informed.
Documentation of subsequent reoair, testing, and acceptance based on
the test results 1s carried out by :^e CQA personnel. Where excessive
damage 1s noted, a meeting of the Owner/Operator, CQA Officer, and
Installer 1s held to determine whether or not the roll or panel is
repairable. If the decision to reje:: the panel is made, a CQA Monitor
should record it and confirm that -'••& roll or panel 1s taken up and
removed from the site. If portions :f a panel can be salvaged for use
elsewhere, their accsptabi11ty nust re confirmed by the Designer.
4.4.2.4.3 Seaming and Joining
Each seaming crew at the site should be accompanied by a CQA
Monitor. This is of primary i-pcr-.ance with the flexible membrane
liner seaming, as the integrity of fese seams is the most critical to
the successful performance of tie l;iing system. In general, geonets
and geogrids do not require strirrent joining criteria (since the
design should preclude tension at seams), but aa. quate overlaps or
ties should be monitored and crnfi^-ed prior to •> overplacement of
the next layer of materials. Geotextile filters should be sewn, and
destructive tests performed in accordance with the requirements of the
plans and specifications. In any s/ent, the CQA Contractor should
confirm that all geotextile searrs that require sewing have been
completed.
4-55
-------
In the specific case of lexible membrane liners, the CCA
Contractor should monitor a trial sea^n for each piece of seamng
apparatus in use at the site, as well as for each equipment operator.
This trial seam (carried out on a sacrificial or waste piece of liner)
should be field tested in shear and peel (see Section 4.5.2.2), and
should pass that testing in accordance with the acceptance criteria
established In the lining system specifications. If this trial seam
falls 1n either shear or peel, the equipment should be adjusted
appropriately, and another trial seam prepared. No piece of seaming
apparatus, and no seaming operators should be allowed to prepare any
flexible membrane liner production seams before passing a trial seam
test. These trial seam tests should be conducted at the beginning of
every production shift, and other times as deemed necessary by the CQA
personnel (e.g., a change in temperature or weather during a shift).
The CQA Monitor should record all of the relevant apparatus settings,
weather conditions, name of the seamer, and the pass or fail result
for every trial seam carried out at the site.
4.4.2.4.4 Nondestructive Testing of Seams
All flexible membrane liner seams should be nondestructlvely
tested over their entire length, using appropriate Industry-accepted
procedures, as outlined In the Construction Quality Assurance Plan f—
the site. These procedures may include, but are not necessari ,,
limited to visual observation, prote testing, vacuum testing, spark
testing, air lance testing, and ultrasonic testing. Other test
procedures may become appropriate as new technology develops and is
proven to be effective. The CQA Contractor should monitor the
nondestructive testing of every crew undertaking this activity at the
site. All seams failing the test should be marked by the CCA
Contractor, including an estimate of the extent along the seam of the
Inadequate section.
The CQA Contractor should document that the failed area is
repaired by the Installer, and retested with the nondestructive test
apparatus, and that this retesting results in a passed test. The
location, date, type of repair, and retest should be properly
documented by the CQA Contractor, along with other pertinent weather
4-60
-------
data, apparatus identification, and seamer data which are necessary
for the proper identification of ' 3 location ar.-j confirmation of the
repa;r.
4.4.2.4.5 Destructive Testing
Test strips should be removed from the flexible membrane liner
seams at locations determined t>y the CQA Contract..-, on the basis of
unusual conditions, suspicious seam quality, or at predetermined
Intervals at a frequency which meets or exceeds the specified minimum
frequency. These seams should be destructively field tested on site,
and based on whether the seams pass or fail, appropriate action taken.
The Construction Quality Assurance Plan for the site should outline
the protocol for the follow-up of failed on-5ite destructive tests,
and failed areas should be investigated and repaired in accordance
with these procedures. The CQA Contractor should ensure that the
sampled locations are repaired and nondestructively retested. This
follow-up process should be thoroughly documented, including location,
test result, sample number, etc.
In addition, the CQA Contractor should oe responsible for the
selection of • 3am sample location, seam sample collection and sample
shipment to the Independent Test Laboratory for testing. These
samples should be of sufficient size to yield the required number of
test specimens, and to allow division of the sample into three: one
for the Independent Test Laboratory, one for the Installer's test
laboratory (if he has one), and an archive sanple to be retained by
the Owner/Operator. The sampling frequency for this testing should be
specified 1n the Construction Quality Assurance Plan, but should
normally be at a specified minimum frequency (such as one test for
every 150 m (500 ft.) of seam), or more frequently, as dictated by
site conditions, anticipated problems, etc.
4.4.2.4.6 Other Considerations
TI.e CQA Contractor should b- responsible for the examination of
each prepared surface prior to the placement of the next layer of the
geosynthetlc or soil. This should include an examination of the
4-61
-------
compacted soil subgrade surface prior to the pla .nient of the first
layer of geosynthetics, and similarly for each successive layer of
soil or geosynthetics comprising the lining sy""?m. Continuous CQA
monitoring is suggested during placement of ;il cover or other
materials over FML components of the lining system. Any observed
damage to the FML during cover placement should be expeditiously
repaired.
In the case of the compacted soil subgrade, the Installer should
submit documentation accepting the condition of the subgrade, at which
time he accepts responsibility for ensuring that the subgrade
condition is preserved up to the point at which it 1s covered. The CQA
Contractor should ensure that this documentation is presented 1n the
final report of the Construction Quality Assurance Program.
The proper functioning of the geosynthetlc components of the
system 1s predicated upon their Installation in the manner prescribed
in the specifications. As a result, the CQA Contractor should ensure
that the geosynthetics are clean prior to being covered. Any dust,
debris, or accumulated dirt due to operations or other contamination
occasioned by rainfall, erosion of the soil components, or uncovered
portions must be removed prior to covering the contaminated layers or
the material must be replaced. The geonet drains and geotextile
filters are particularly sensitive to this type of problem, and are
the most likely to be adversely affected, 1f measures to clean them
are not taken. In some cases, it may be necessary to remove one or
more in-place layers of geosynthetics, in order to gain access to the
contaminated geosynthetlc, particularly when the edge of several
layers 1s exposed and damaged by washing soil particles into the
e^osed edge of the system. If the contaminated layers cannot be
cleaned, they must be replaced.
All of these activities should be documented by the CQA Contractor
in his dally reports.
4-62
-------
4.4.2.5 Leachate CoHectyn and Re-noval_Sys terns
The CQA Program for leachate collection and removal systems must
provide reliance that the installed system meets or exceeds the design
specifications. The functions of a leachate collection and removal
system above the top liner (LCRS) in a double-lined landfill or waste
pile unit are to minimize leachate head on the top liner and to
collect and remove liquids from the unit, during the active life and
post-closure care period. The purpose of a leachate detection,
collection and removal system between the two liners (LDCRS) of a
double-lined waste unit is to rapidly collect and remove liquids
entering the system, also through the post-closure care period. By
providing for rapid leachate removal, the LDCRS will greatly minimize
the hydraulic head on vue bottom liner and, thereby, minimize or
eliminate leachate migrai..on out of the unit. As the LDCRS will also
be used to detect leaks in the top liner, the CQA Program must ensure
that the system is installed as designed for that purpose by meeting
the detection sensitivity and detection time performance standards
discussed in Chapter 2 of this report.
Observing and testing the subcomponent materials of the leachate
collection and removal systems as they are delivered to the site and
Installed are necessary to confirm and document nat these materials
conform to the design criteria, plans, and specifications. This
observation and testing applies to the granular materials,
geosynthetic materials, piping and sumps, and any other materials that
make up a leachate collection and removal system.
Below are summaries of key factors that need to be addressed while
constructing the LCRS and LDCRS. Two potential problems related to
installation are (1) damage to the collection system during
installation resulting from excessive stress and (2) leachate flow
obstruction through the system. A third potential problem is
contamination of geosynthetic components of the leachate collection
and removal system by dust, debris or other materials. This was
discussed in Section 4.4.2.4.6 and is no1 repeated here.
4-63
-------
4.4.2.5.1 Leachate Collection Pipes
leachate collection pipes installed in trenches at the base of a
landfill or waste pile and between the iiners in a landfill, surface
Impoundment, or waste piles are subjected to loads from construction
equipment during installation, operation activities during the active
life, and the waste itself. In a well-designed trench, only a small
fraction of the load of a wheel or tracked vehicle applied at the top
of the trench should be transmitted through the trench backfill to the
pipe. However, the percentage of the load transmitted increases
rapidly as the vertical distance between the loaded surface and the
top of the pipe decreases. In addition, moving loads cause impact
loading, which 1s generally considered to have a one and one half to
two times the effect of stationary loading. Thus, backfill procedures
and equipment traffic over pipe trenches must be monitored carefully
to prevent damage to pipes.
4.4.2.5.2 Obstructions to Leachate Flow
The second consideration when installing a leachate collection and
removal system '• to provide confidence that the flow of leachate
through the syscem is not impaired by construction activities or
occurrences. Collection systems generally are designed so that
leachate generated within the unit drains first through a soil or
geosynthetic filter before entering the drainage layer. The purpose
of this filter is to remove any fine particles that otherwise would
clog the drainage layer and prevent its functioning. The filter,
therefore, must be designed and constructed carefully to perform under
the expected conditions. The leachate then flows through tne drainage
layer, which comprises permeable soils or geosynthetic drainage
materials placed over the liner. If this layer does not have
sufficient transmissivity (thickness times hydraulic conductivity) to
accommodate the maximum leachate flow, the flow will be held up, and
hydraulic head will build up on the Hner. Achieving the designed
thickness can be made more difficult by Improper Installation
procedures, such as placii.g a granular drainage layer during high wind
or intense rain, which may displace the soil so that it is no longer
of uniform thickness. Another weather-related problem is drainage
4-64
-------
material contamination with fine soil particles, which decreases the
drainage layer's hydraulic conductivity. This c?n occur as a result
of soil particle erosion into granular or geosyntrit^c drainage layers
from runoff from facility side slopes, mud, or windblown dust. These
types of problems can be minimized by monitoring and testing
activities that cherk the critical factors in the leachate collection
system.
Installation procedures must be monitored to confirm that granular
soils used fir the drainage layer meet design specifications for size
distribution of particles. In particular, excessively fine soils must
not be allowed, because they will decrease the hydraulic conductivity
of the layer and will clog collection pipes. On-site washing of the
granular soils to remove fines may be' necessary to achieve the
required properties. Similarly, geosynthetlc materials must be
conformance tested to ensure that they meet design specifications, and
they also must be covered to keep them clean.
4.4.2.6 Final Cover Systems
The successful construction of the final cover, like the other
unit components, relies on following recommended practices for
construction, employing experienced personnel, and conducting a CQA
Program. The CQA Program for final covers at all land disposal units
must provide assurance that (1) all layers of the final cover are
monitored for uniformity, imperfections, and damage; (2) the materials
for each layer are as specified in the design specifications; and (3)
each layer 1s Installed or constructed to meet the design
requirements.
The following is a summary of the key factors that should be
addressed for final cover construction cover at landfills, surface
impoundments, and land treatment units.
4.4.2.6.1 Subsidence
Subsidence under a final cover may ca'"5e problems similar to those
experienced when the subgrade under a ;,ner subsides. A flexible
4-65
-------
membrane liner may fail in tension if the waste that comprises its
subgrade subsides differentially. If the final cover uses a layer of
compacted soil, the soil layer may develop cracks as a result of
differential subsidence that allows rainwater to infiltrate. In
addition, differential subsidence may result in rainwater ponding
above the final cover. The ponded rainwater may have an increased
chance of penetrating the cover even if the soil is intact because of
the increased pressure head on the liner. If a cover of any type has
failed, ponding prevents runoff from leaving the area and provides
additional opportunities for leachate production.
For covers, the problem of subgrade subsidence begins with waste
placement. The waste may not have sufficient bearing strength to
support the weight of additional waste and soil cover material placed
above It. In addition, if the waste 1s not well compacted and placed
so that void spaces are filled, proper compaction of the Uner bedding
material will not be sufficient to prevent subsidence. Therefore, to
minimize subsidence, waste placement must be considered a part of
final cover subgrade preparation. Cover subsidence resulting from
Improper waste compaction may be less of a problem today than it has
been In the past. Wastes were not compacted well or at all in older
landfills or disposal surface impoundments when problems associated
with final cover subsidence were not well known. Now, however,
virtually all landfills compact their waste. Nonetheless,
differential settlement because of waste subsidence continues to be a
serious problem that must be anticipated in the cover system design.
Some key considerations in the design of cover systems Include:
(1) The stress-strain propei ,ies of the cover system FML,
geosynthetics and soils;
(11) the ability to maintain minimum slopes for gravity drain
systems;
(lii) the slope stability of layers above FML's and
geosynthetics;
4-66
-------
(iv) the use of subgrade reinforcement of stabilization
methods, such as geosynthetic reinforcement or dynamic
compaction to improve stability.
4.4.2.6.2 Installation Procedures
The construction process for final covers at landfills and
disposal surface impoundments involves subcomponents similar to many
of the components previously discussed, such as foundations, compacted
low-permeability soil liners, flexible membrane liners, and drainage
layers (leachate collection and removal systems). There is too little
documented information to substantiate the quality of final covers
that are constructed to comply with the landfill and surface
Inmpoundment requirements in Parts 264 or 265. However, it is
believed that most of the installation problems for final covers for
these units sr .uld be similar to those experienced installing liners,
dikes, and leachate collection and removal systems.
For example, the compacted low-permeability soil layer and FML in
a final cover is constructed much like the low-permeability soil and
FML liner. However, tM foundation for the final cover may have a
lower bearing strength than the soil liner foundation; this may
require using different construction techniques to achieve the
required compacted soil hydraulic conductivity in the field.
Additionally, the design may specify foundation (waste) soil
reinforcement and such soil reinforcement must be Carefully monitored
during installation by Construction Quality Assurance personnel. As
with the compacted low-permeability soil and FML liner, it is
necessary to monitor the construction of the compacted low-
permeability soil and FML cover layer.
Installation procedures for FMLs in a • -il cover include proper
on-site storage, handling and placing of the ^anels to ensure proper
positioning, allowing enough slack in the material for it to fit
around angles and penetrations, proper seaming and anchoring
procedures, and installation only during proper weather conditions.
4-67
-------
4.4.2.6.3 Vegetative Layers
The k?y factors that need to be addressed for constructing the
vegetaMve layer of the final cover at land disposal units include:
vegetative layer soil quality and thickness, seeding uniformity .id
timing, and vegetation establishment. The vegetative layer is the
only layer of the final cover required for properly operated land
treatment units under a permit.
Vegetation establishment and maintenance ;an be accomplished only
by carefully addressing the soil type and the nut ient and pH levels
to provide the proper soil conditions for successful seed germination
and vigorous growth. The thickness of the vegetative soil layer also
must be as specified in the design to provide proper root development
and a sufficient moisture reserve to sustain the vegetation during dry
periods.
The timing of the seeding is probably the most important factor in
successfully establishing a vegetative cover. The timing will depend
on whether the plant species selected is a cool- or warm-season
species and on local climate conditions. The recommendations of the
locan country agricultural extension agent or seed company should be
used. The CQA Plan must address seeding procedures so that the
recommendations are followed.
4.4.3 Post-Construction Stage
4.4.3.1 Reporting
The CQA Contractor should be responsible for the preparation of a
final report on the construction of the land disposal unit. This
report should contain, at the least, all of the activities identified
herein, properly documented to allow the retrieval and interpretation
of the information. This includes, but is not necessarily limited to
the following:
4-68
-------
detailed documentation of the laboratory testing car <3d out en
the soils used for earthworks, the granular cc^onents of
drainage and leachate collection and removal systems, and the
compacted soi! components of the lining system;
detailed documentation of the Test Fill program, including all
Iterations and modifications utilized to develop the final
criteria for attainment of the specified compaction and/or
hydraulic condu-- -. i vity requirements in the earthworks and
compacted soil components of the lining system;
detailed documentation of the field testing to determine as-
placed densities and moisture contents for earthworks and
compacted soil liners (this should include every test result
obtained in this program, identifying the equipment used to
conduct th« test, the operator, the locati^ and depth of the
test, and a cross-reference in cases where retesting was
required);
detailed documentation of all of the rolls and panels of
geosynthetics supplied to the site, with appropriate
Identification;
detailed documentation of all of the conformance testing
carried out on the geosynthetic materials supplied to the site,
and a cross-reference to indicate the action taken for any
materials which were rejected;
detailed documentation of every trial seam carried out for each
piece of apparatus, and each operator, for every shift worked
in which seaming was carried out, including failed trial seams
and the retests;
detailed documentation of the seaming operations carried out at
the site, including the location of the seam, length of seam,
panel numbers, operator, apparatus number, weather conditions,
and time seamed;
4-69
-------
detailed documentation of the nondestructive testing of the
seams, including the location of the seam, length of s?3n,
panel numbers, operator, apparatus number, weather conditi:ns,
time tested, test result, and a cross reference, for failed j
seams, to the retest of t t seam which obtained a passing •
result;
detailed documentation of the field destructive test results
carried out on specimens cut from the 1n-place seams, Including
pass/fail results, operator, apparatus number, location, and a
cross reference to the repair and nondestructive testing
results;
detailed documentation of the location and nature of the
destructive test samples collected for testing by the
Independent Testing Laboratory, Including seam and panel
Identifications;
a final report from the Independent Testing Laboratory showing
the results of all of the destructive test samples collected
from the FML seams, and the rt^Jlts of the conformance testing
conducted for all geosynthetics;
detailed documentation of the fabricated seam quality control
testing carried out in the fabrication facility whether or not
this was observed by the CQA Consultant (if not, this
Information should constitute a required submittal of the
Fabricator);
detailed documentation from the Installer indicating his
acceptance of the soil subgrade prior to the placement of any
components of the lining system (if the same Installer Is
responsible for both the earthworks and the compacted soil
Uner, as well as the geosynthetics Installation, then this may
not be required);
4-70
-------
• detailed photographic documentation of the activities at the
site, highlighting problems encountered, actions taken, and
final resolutions; and
• detailed record (as-built) drawings of the lining system.
Indicating the locations of all repairs, destructive sample
tests, the layout of panels and/or rolls and their numbers, and
all relevant details and cross sections required to obtain an
accurate picture of the system as it was constructed.
The final report for the CQA Program should include a statement,
signed and sealed by the responsible CQA Officer, that all of the
components of the system were installed in conformance with the plans,
specifications, and good construction practice. Any exceptions to this
statement should be noted, and the action taken fully described. The
CQA Officer should be a registered professional engineer (P.E.) in
accordance with the laws of the state in which the facility is
located.
4.4.3.2 Monitoring
Any monitoring program involving the lining system, or the
leachate collection and removal system, in t^e form of leak detection
or investigat-on should at least indirectly include the CQA
Contractor, by virtue of his primary knowledge of the system as
installed. In any event, any problems with the system that are
encountered, should be investigated by a team which should include, at
the least, the Designer, the Installer, the Owner/Operator, and the
CQA Contractor. Any corrective construction work carried out to repair
damage or other problems should be provided with full Construction
Quality Assurance by the CQA Contractor, and ideally the same
personnel as were involved in the original program.
4.4.3.3 Coupon Testing
Coupon testing is sometimes incorporated into post-construction
CQA activities, although it is an optional component of the
Construction Quality Assurance Program. Samples of all of the
4-71.
-------
materials are collected, and after start of operations, immersed in
the leachate collection and removal system surp. Subsequently,
san-ples frcm the coupon may be removed and tested in the laboratory t2
evaluate the durability and aging characterises of the coupon
material. This may be incorporated into any jnit system, for
assistance in the evaluation of the performance and condition of the
1iner.
4.5 TESTING PROCEDURES
4.5.1 Soils
4.5.1.1 Procedures
The soil components of the construction activities should be
tested, wherever possible, in accordance with accepted American
Society for Testing and Materials (ASTM) standards. The same standards
are applied to different soils, with slight variations, regardless of
the purpose of those soils (e.g., the same field compaction density
test 1s carried out on soils for the compacted soil liner as for the
soil materials used to construct dikes and embankments, although the
required test result may differ). The following test procedures should
be utilized for the testing of soil '•—oonents of ti.e waste management
unit.
4.5.1.1.1 Laboratory Testing - Soil Compaction
Laboratory tests are carried out on the particular soils to be
placed at the site, in order to develop the required compaction
criteria. The following tests are those which are utilized for this
purpose:
• Standard Proctor Test - ASTM D698 - "Test Methods for Moisture-
Density Relations of Soils and Soil-Aggregate Mixtures, Using
5.5 Ib (2.49 kg) Rammer and 12 in (305 mm) Drop" [ASTM 4.08].
This test is referred to as the Standard Proctor test, and
determines the maximum density attainable for a given level of
. ,compactive effort in a laboratory mold. The field degree of
4-72
-------
^
compaction 1s then compared to this density, expressed as a
percentage. The typical field ccr-pactlon specification would
therefore call for some percentage, e.g., 97 % of the Standard
Proctor maximum dry density.
• Modified Proctor Test - ASTM D1557 - "Test Methods for
Moisture-Density Relations of Soils and Soil-Aggregate Mixtures
Using 10 Ib (4.54 kg) Rammer and 18 1n (457 mm) Drop" [ASTM
4.08]. This test Is referred to as the Modified Proctor test,
and 1s carried out 1n essentially the same manner as 0698,
except that the soil 1s compacted Into the mold 1n thinner
lifts (I.e., 5 lifts Instead of 3), and greater compaction
energy Is used. As a result, the compaction criterion 1s
virtually always more stringent (i.e., 97 '/. of Standard Proctor
maximum dry density will almost always be a lower density than
97 7. of Modified Proctor maximum dry density).
4.5.1.1.2 Fltid Testing - Soils Compaction
Field density testing 1s today predominantly performed by one
procedure (the nuclear density test), which provides values for both
the field density, and the moisture content. Sometimes, however,
other methods are employed in order to provide a check of the
repeatability and consistency of the results.
• Nuclear Density Test - ASTM D2922 - ''Density of Soil and Soil-
Aggregate In Place by Nuclear Methods (Shallow Depth)" [ASTM
4.08]. This procedure measures the density and moisture content
of the soil at the surface and at depths up to 250 mm (10 In.)
by the transmission of gamma rays, the intensity of which are
affected by the density of the medium. The soil density can
therefore be computed by the calibration of the equipment to
known standards. This measured density in the field is related
to the specified density (a percentage of either the Standard
or Modified Proctor maximum dry :nsity). This then determines
the degree of compaction, and whether or not further compactive
effort 1s required.
4-73
-------
• Sand Cone Density,Test - ASTH D1556 - "Density of Soil In Place
by tf:» Sind Cone Method" [ASTM 4.OBJ. This procedure measures
tne hi-place density of the soil. Soil is carefully excavated
from a hole In the fill, and e volume of the hole detetniined
by filling 1t with a pre-cal,jratsd sand. The weight of sand
used to fill the hole 1s measured, so the dry density can be
calculated from the weight of sand, volume of the hole, and
moisture content jf the soil removed from the hole. This
procedure Is well established, having been 1n common use for 30
years. It Is, however, somewhat time-consuming since the
excavated sample must be returned to the laboratory for
moisture content determination. It has value primarily as a
cross reference and a check on the calibration of the nuclear
densometers being used for production field testing.
4.5.1.1.3 Laboratory Testing - Son Permeability
The laboratory determination of soil hydraulic conductivity 1s
well established within the geotechnical engineering profession.
However due to the degree of sophistication of sorn° of the equipment
that is required, not all of the procedures are available to
laboratories outside of the university or research environment. ere
are, however, two well established tests for the laboratory
determination of the hydraulic conductivity of soils that can be
carried out in any of four pieces of apparatus; three of these four
are relatively 'standard' geotechnical laboratory apparatus. It should
be noted that not all of these have established ASTM test procedures.
In addition, as discussed previously, the relevance of the test result
to the field condition or the soils
disturbance caused during sampling,
laboratory, and sample preparation,
Is, In part, a function of the
handling, transportation to the
as well as by the presence of
nonuniformlties, secondary structures, etc., 1n the field.
• Falling Head Permeability Test - This test 1s commonly used for
the determination of the hydraulic conductivity of fine-grained
soils, notably silts and clays, which exhibit intermediate to
low values of hydraulic conductivity (I.e., less than 10~' m/s
(10"* cm/sec)). The test can be performed on trimmed samples 1n
4-74
-------
any of the t^st apparatus discussed below.
Is outlined _y [Head. 1982].
Another procedure
• Constant Head Permeability Test - This type of test ,s
Intended for use solely for granular soils (sands and gravels)
containing little or no fine-grained silt or clay particles.
The procedure Is very similar to that of the falling head tes^,
except that the head of water Is maintained at the same level
throughout the test. This is relatively easy to understand
since the falling head tests, If conducted on these types of
soils, would either require a very large reservoir, or the test
would be of such short duration that the test result would be
questionable. Other procedures also exist using different
apparatus [Head, 1982].
The falling-head and constant-head permeability tests described
above can be performed on representative soils 1n any of four well
established laboratory apparatus, although there are only three
apparatus that can be considered to be common laboratory equipment,
and only two apparatus which are represented in ASTH procedures. These
four apparatus are briefly described as follows:
• Soil Permeameter- The soil permeameter cell is described 1n
ASTM 02434 - "Permeability of Granular Soils (Constant Head)"
[ASTM 4.08]. This test can be performed fairly easily on coarse
grained soils with moderate to high permeability (hydraulic
conductivity). It consists of a plastic cylinder which is
filled with the soil, and through which water is allowed to
flow. Maintaining the constant head of water on the sample, the
flow through the sample 1s directly measured, from which the
hydraulic conductivity of the soil can be calculated using
Carey's equation. Falling head tests are carried out in the
same apparatus on fine-grained soils, with similar procedure
except that the head is allowed to draw down as the water flows
through the sample. Both of these tests are also described in
[Head, 1982].
4-75
-------
* Pedometer Cel 1 - ASTM D2435 - "Cne-Dimensional Consolidation
Properties of Soils" [ASTM 4.C3J. The hydraulic conductivity of
a ; ine-gra mod soil on be aeterr-irvjd for jny lead in:i—-?r,t
during the performance of a standard oedcmeter consolic ion
test. If ''he soil consolidation properties are not of specific
interest, the permeability test can still be performed in this
apparatus by following the procedure cited for tne
consolidation test. This is a standard piece of geotechnical
laboratory apparatus, and n fact, if the test is monitored by
a data acquisition system, then the values for the hydraulic
conductivity of the sample under each increment of load can be
obtained directly. In the context of waste management units,
this can provide an indication of the response of the soil
hydraulic conductivity under load. This is a feature that is
not readily available with the Permeameter. The drawback of the
Oedometer, however, is that the sample size is frequently far
from ideal, and the lack of representativeness to site
conditi is is 1ikely maximized.
* Triaxial Cell - There is no standardized test procedure for
this apparatus for determining soil hydraulic conductivity,
although several different procedures are contained in the
literature [Head 1986]. In this reference three procedures are
outlined which accommodate the variab; ity in the type of
equipment which may be available in a given laboratory. In the
test, the flow of water through the sample is upwards, which
distinguishes it from the other tests, which are gravity-
driven. This does, however, allow the control of the gradient
across the sample, and provides a high liability factor to
the test result. One of the features of tnis apparatus is that
both constant head and falling head tests can be performed,
which provides considerable flexibility in the sense that the
same apparatus can be utilized for all types of soils. In
addition, the flexible containment of the sample more closely
models the field overburden conditions, which can be taken to
very high levels easily with this system.
4-75
-------
'Cell - This is another test which has not been
standardized, but whicn is familiar to research and university
laboratories. The apparatus is a relatively large-scale
consolidation cell (up to 250 mm (10 in.) in diameter) on which
the load is imposed hydraulically through a diaphragm and
pressure system (such as the standard laboratory mercury-pot
system). Due to the configuration of the cell, the test can be
set up to determine the hydraulic conductivity of the soil
sample with drainage either vertically or horizontally. In
addition to allowing the test to be carried out under known
conditions of effective stress, the loads can al-o be cycled.
In addition, both constant head and falling head tests can be
performed in this apparatus. A thorough treatment of this test
and its variations is contained in [Head, 1986].
4.5.1.1.4 Field Testing Soil Permeability
The least developed component of soils testing related to the
design and construction of land disposal units is the field
determination of soil hydraulic conductivity. The test procedures
currently in use have a response time occasioned by the low hydraulic
conductivity of the soil, which results in the test taking a very long
time to perform. Hence, the existing tests are primarily of value as
design tests. Due to the duration, they are presently of little value
as a quality control test.
For example, the sealed double ring infiltrcmeter (SDRI) developed
by Daniel [1985] is based in part on ASTM D3385 "Infiltration Rate of
Soils in Field Using Double-Ring Inf1Itrometers" [ASTM, 4.08]. The
Daniel apparatus has been developed to overcome the constraint imposed
that the test as written is not suitable for use in clay soils. This
modified procedure achieves results which are reported to be more
representative of field conditions over laboratory test procedures.
The SDRI test tends to give a higher (up to two orders of magnitude)
value for hydraulic conductivity than do those obtained from
laboratory hydraulic conductivity tests. One shortcoming of the
procedure is that the overburden conditions experienced in a landfill
waste pile or surface impoundment are not modeled. In landfills, r~r
4-77
-------
instance, overburden stresses on the order of 5CO kPa (10,500 psf)
could be exerted, which would tend to alter the hydraulic conductivity
from that measured in the absence of this condition.
The SDR! procedure reported by Daniel is,/ at present, the most
thoroughly documented, and is current beifng used for in-place
hydraulic conductivity testing of Test Fills ;:(see Section 4.5.1.4'i.
Other procedures are being developed as well-[Reynolds and Elrick,
1985a, 1985b, 1986]. In the context of CQA, however, 1t is necessary
to rely on the :are and skill of the persons sampling and handling the
laboratory hydraulic 'conductivity tests, to provide accurate results.
The extrapolation of these results to the field situation will remain
an area requiring considerable experience and interpretation on the
part of the Designer and CQA Contractor.
4.5.1.2 Effectiveness of the Tests - Acceptance Criteria
-.5.1.2.1 Compaction-delated Tests
The compaction-related tests for soils are very well established,
and even the most recent of these tests (D2922 "Density of Soil and
Soil-Aggregate In Place By Nuclear Methods (Shallow Depth)") has been
an ASTM standard test method since 1971. It i: generally accepted that
the reliability of the results has been proven, and this equipment is
now almost universally used for this purpose.
The Modified Proctor Test (D1557) is used for a compaction
standard, primarily when a very high degree of compaction is necessary
for a high strength fill (e.g., dam construction). For the purposes of
waste management projects, for the general earthworks other than the
soils for the lining system, it is believed that the less restrictive
Standard Proctor Test (D698) is more appropriate. The acceptance
criteria for compaction are a function of the nature of the particular
component of the structure. The general earthworks and fills would
normally be compacted to a degree between 95 percent and 98 percent of
the Standard Proctor maximum dry density fc,- that soil, and densities
falling below the 95 percent level are not desirable.
4-78
-------
The low-permeability soils comprising compacted soil components cf
the lining system are specified on the basis of a hydraulic
conductivity criterion, and cc-paction to a specified degree rray r.ot
be necessary, except in cases where the level of c -paction, soil
density, moisture content, and hydraulic conductivity have been
correlated. One of the purposes ~r the Test Fill program [USE^A,
1986] Is to allow this cor-elatio compactive effort and density
achieved at a given moisture concent, to the 1n-place hydraulic
conductivity. Consequently, for the compacted soil lining components,
the field density will likely only be used for this cross reference.
4.5.1.2.2 Hydraulic Conductivity Tests
The laboratory tests to determine hydraulic conductivity are well
established. The apparatus such as the Triaxial Cell and the Rowe Cell
are somewhat more desirable in that they can allow the use of larger
samples. Regardless of the particular test selected, care must be
taken at all stages from sample collection through performance of the
test, to avoid the potential problem of sample disturbance. In
addition, there 1s concern with regard to the representativeness of
the sample to the 'macro-scale' field condition, whereby the secondary
structure of the soil, and other discontinuities may not be modeled
iiito the test because of the relatively small sample tested. The
Interpretation of the engineer carrying out the test therefore becomes
an Important factor in assessing the results.
4.5.1.3 Current_State_of PracUce
The test procedures outlined herein and recommended for use in the
Construction Quality Assurance Program represent the current state of
practice. A useful tool for providing field hydraulic conductivity
measurements with fast turn-around of results Is still required,
however, in order to be of value for Construction Quality Assurance
Programs. This state of practice is underdeveloped at present. As new
methods and procedures are developed, they will require considerable
field testing before they can be relied on for CQA purposes.
4-79
-------
&*' *-
4.5.1.4 Test Fills
Although Test Fills may be constructed for the unit in order to
evaluate the compatibility of the soils over a range of moisture
contents, the Test Fill program is primarily a means of evaluating the
hydraulic conductivity of the compacted soil components of the lining
system.
A field hydraulic conductivity test of the compacted soil in the
Test Fill should be performed to confirm that the materials and
procedures used in the field will result in a compacted soil liner
with a hydraulic conductivity of 10~' m/s (10~7 cm/sec) or lower.
Field testing Is not intended to preclude the use of laboratory
testing in the design or construction stage or as a means of
evaluating 11ner/leachate compatibility. The design ?tage and the
Construction Quality Assurance Program will be expected to Include a
mixture of Doth field and laboratory hydraulic conductivity tests.
As appropriate methods are developed and verified, it will be
desirable to conduct field hydraulic conductivity tests on the entire
unit. Until that time, field hydraulic conductivity tests can be
performed in the Test Fill without causing delays during construction
of the ent- a unit. The field test used in the Test Fill should be
performed for 2.n adequate length of time to achieve "stable" results
and to verify that the hydraulic conductivity of the compacted soil
liner 1s 10"' m/s (10~7 cm/sec) or lower.
In addition to being used for a site for the field hydraulic
conductivity test, the Test Fill should also verify other elements of
the design and construction of the soil liner. The Test Fill
construction will allow the Construction Quality Assurance Monitors to
verify that equipment and construction procedures for breaking up
clods, moisture conditioning (wetting and/or drying), and compacting
the soil are adequate to meet the specified density, moisture content,
and permeability criteria. In addition, construction monitoring
activities, including measurement of lift thickness and compaction
equipment coverages, can be correlated with in-place density and
moisture content tests and with the field hydraulic conductivity.
4-80
-------
^
EPA has published a valuable Technical Resource Document: "Design,
Construction, and Evaluation of Clay Liners for Hazardous Waste
Facilities" [USEPA, 1986c] that provides detai'?d information on
construction of a compacted soi. liner.
The Construction Quality Assurance Program for compacted, low-
permeability soil line must confirm that the liners meet or exceed
the design specificat .jn. The purpose of a compacted, low-
permeability soil Uner depends on the overall liner system design. In
the case of soil liner1; used as the lower component of a composite
Hner, the soil component serves as a protective bedding material for
the upper component of the FML and minimizes the leakage rate through
any flaws in the upper component. An objective shared by all low-
permeability soil liners is, therefore, to serve as long-term,
structurally stable bases for all overlying material.
Prior to construction, adequate studies should be conducted to
confirm that the low-permeability soil liner design meets or exceeds
regulatory requirements. These studies should include an evaluation of
the proposed borrow source to confirm the existence of an adequate
quantity of suitable material, with testing to determine particle size
distribution, Atterberg limits, moisture/density relationships,
hydraulic conductivity, 1 iner-leachate compatibility tests, and
appropriate consolidation and strength tests of fabricated samples of
the proposed soil liner. This work can be replicated on the Test Fill.
4.5.2 Flexible Membrane Liners
4.5.2.1 Procedures
Many of the procedures with which FML's are being tested were
developed for polymeric materials in general, and in some cases may
not be totally appropriate for flexible membrane liners. ASTM
Committee 035 on Geotextiles and Related Products is working to
develop FML-specific tests to supplant existing tests.
4-81
-------
r~ \yff^f--tfyTy T*"^'«->--JiSSl * * -
4.5.2.1.' Laboratory iest1ng - Specifications Conformance
The laboratory tests for conformance are those which are carried
out on s'.mples of f'.dxible membrane liners taken frcm the mils and/or
panels delivered to the site. These tests are used to verify that the
FML's supplied comply with the specified properties. In some cases,
for particular applications, different or additional properties may be
specified for this conformance testing, but in general, the following
tests will allow the evaluation of flexible membrane liner conformance
with the specifications.
• Density - ASTM D7l)2 "Test Methods for Specific Gravity and
Density of Plastic by Displacement", Method A, or ASTM D1505
"Test Method for Density of Plastics by the Density-Gradient
Technique" [ASTM, 8.01]. These procedures test to verify that
the density of the material is within the range specified.
Tensile Strength and Elongation - ASTM D638 "Test Method for
Tensile Properties of Plastics" [ASTM, 8.01]. This test J-s a
simple shear test carriad out to record the tensile strength
and elongation on bredk of the flexible membrane liner, to
ensure that its stress-strain characteristics comp j with the
specifications. Requisite stress is a constant, regardless of
the gauge of the material, and this can therefore be easily
verified by this test, given the thickness of the material.
Carbon Black Content - ASTM D1603 "Test Method for Carbon Black
1n Olefin Plastics" [ASTM, 8.02]. Carbon black content in FML's
1s a necessary component required to resist degradation due to
ultraviolet exposure. In polyethylene flexible membrane liners,
the otherwise white material is susceptible to degradation
after exposure to sunlight after even relatively short periods.
A minimum proportion of carbon black is required to be mixed
with the resin to avoid this problem.
4-82
-------
syjafsr?*!**!*^^ :'^5^t^
.-_•_-, - 4
• Thickness - ASTM 01593 "Specification for Nonrigid Vinyl
Chloride Plastic Sheeting" or ASTM D374 "Test Methods for
Thickness of Solid Electrical Insulation'1 Method C [ASTM,
8.01]. As the mechanical properties are, at least in part, a
function of the thickness of the sheet, the thickness can give
an Indication of whether or not there will be problems with the
physical characteristics of the material.
4.5.2.1.2 Laboratory Testing - Destructive FML Seam Quality Control
As an Important component of the overall Construction Quality
Assurance Program, the quality of the flexible membrane liner field
seams formed by either extrusion or fusion procedures is measured by
the performance of testing to determine the Integrity of the seams.
These samples are collected from actual production seams at the site,
and the testing carried out by an Independent Testing Laboratory, with
results of the testing reported quickly, to allow action before any
cover materials are placed on the liner.
• Bonded Seam Strength - ASTM D3083 "Specification for Flexible
Poly(Vinyl Chloride) Plastic Sheeting for Pond, Canal, and
Reservoir Lining" Modified by NSF 54 [ASTM, 4.04; NSF, 1985].
In this test, a tensile test 1s carried out across the FML
field seam, and the conformance of the test result to the
specification checked.
' Peel Adhesion - ASTM 0413 "Test Methods for Rubber Property -
Adhesion to Flexible Substrate" Modified by NSF 54 [ASTM, 9.01;
NSF, 1985]. This 1s a test across the flexible membrane liner
seam, 1n which the two pieces of the seamed rolls or sheets are
pulled apart, so that the s'-'-ess is exerted directly onto the
seam. It 1s generally rr,..t appropriate for polyethylene
flexible membrane liners.
4.5.2.1.3 Field Testing - Nondestructive FML Seam Tests
All FML seams formed in the field must be nondestructively tested
over their entire length. A major component of the overall
4-83
-------
Construction Quality Assurance Program, these technicues can be an
effective -.sans of confirming seam continuity, but are not intended to
be .^jant'.titiva 'n nature.
• Vacuum Test - No ASTM establisheu test method presently exists
for this test, although the test 1s widely used by Installers
for the nondestructive testing of ^xible membrane liner
seams. In the test, a section of the seam is covered with a
soapy mixture and subjected to a vacuum. Leakage is detected
through the presence of bubbles, as air 1s drawn from below,
through the seam. The test is very fast and effective at
locating flaws which may not have been visually apparent.
• Air Lance Test - Another test .hich is not standardized, the
Air Lance Test, is also used to nondestructively test flexible
membrane liner seams. In this test, a high pressure air jet is
trailed along the edge of the flexible membrane liner seam, and
leaks detected by the ballooning r the upper liner.
• Air Pressure Test - This test is not standardized, but is
frequently utilized in the case of double fusion seams of
polyethylene FML's. In this test, a pressure is exerted into
the gap between the two tracks, and the pressure is monitored
over a brief period. A drop in the sustaine^ pressure is
indicative of a leak in the seam.
• Spark Test - In this (also non-standardized) test, a piece of
copper wire 1s placed, so as to be embedded within the seam,
after seaming. A DC current is then passed through the wire
through exposed ends, and a negatively charged wand passed
along the edge of the seam. Leaks will be detected by a spark
jumping from the copper wire to the rod. Some problems exist
with the method, such as the necessity of securing the copper
wire free ends after the test. It is gaining increased usage,
however.
4-84
-------
" UJtrasonic Testing - Nondestructive testing using ultrasonic
equipment is receiving considerable develccmental attention at
this time, although it is not presently widely accepted, and
sufficient field truthing has not been carried out to dat2
[Peggs et al, 1985; and Koerner et al, 1987]. The procedures
being evaluated include both direct transmission, as well as an
echo/pulse procedure. Discontinuities within the seam are
indicated by variations in the oscilloscope signal received. At
present, the need for considerable operator skill and
Interpretation is a major constraint on its use,
4.5.2.2 Effect1veness_of the Tests - Acceptance CMteMa
4.5.2.2.1 Conformance Tests
These tests, as a measure of conformance of the flexible membrane
liner to the specifications for the project, are generally very
effective. Most of the tests were developed specifically for non-
flexible membrane liner plastics, having been adopted for general
purposes. It is anticipated that ASTM Committee D35 will derive and
issue FML-specific tests, and these should be adopted as they become
available. In the interim, the tests procedures which are being used
are not defic;-nt for testing the particular characteristics of the
flexible membrane liners. As confonnance tests, the attractiveness is
that the" are generally easy to perform. In fact, seme of these tests,
such as .ne thickness test can readily be carried out on the spot
using an appropriate micrometer or ether caliper.
The required specifications which must be satisfied for the
conformance testing for flexible membrane liners will be project-
specific, and the properties will therefore be a specific function of
the design.
4.5.2.2.2 Seam Quality Control Tests
The effectiveness of the current seam quality control testing is
the subject of considerable discussion at present, not necessarily
with specific regard to the performance or the results of the test,
4-85
-------
but rather with the Interpretation of he results. This has been -e
topic of technical papers [Peggs et ai, 1985; and Peggs, 1937], and
clearly there are two issues. First, the current NSF Standard 54
states that in peel, for instance, the seam must not peel (pull apart
at the fusion joint), and specifically that Film Tearing Bond must be
exhibited. There are, however, a wide variety of failure patterns
which can occur In this test, and no guidance exists with respect to
how to determine whether or not a problem is indicated by a particular
failure mode.
Another problem is more critical to the operations at the site,
and the Construction Quality Assurance Program in general. When the
samples are collected from the seams formed at the site, they are
split in such a manner that one sample goes to the Independent Testing
Laboratory, and one to the Manufacturer's (or Installer's, if a
different party) laboratory. The laboratories split each sample into
ten specimens, ?->d conduct five shear tests and five peel tests per
sample.
The problem is that there is no guidance with regard to the
interpretation of the results. For a case in which all five peel
specimens fail, for example, the c ;lusion is clear with regard to
the test. In the more frequent circumstance, however, one specimen may
fail totally, but the others pass. The question will arise if the
average of the five specimens meets the specification, then is the
sample a pass or a fail? At present, this must be decided at the
discretion of the CQA Contractor, unless a protocol for the
acceptability of flexible membrane liner destructive seam tests is
developed, in which case it should be incorporated into the
Construction Quality Assurance Plan. Similarly, the procedure for the
evaluation of differing results from the Independent and the
Installer's test laboratories should be iterated in the CQA Plan. Work
Is proceeding on the development of a acceptance protocol for
destructive seam tests [Peggs 1987], and this problem will be resolved
1n due course.
4-86
-------
4.5.2.2.3 Nondestructive Seam Tests
The nondestructive seam tests which are presently camel out are
reasonably effective in the non-quantitative evaluation of seam
continuity. They are, however, not standardized, and for that reason
there is very little control over the procedure, i.e., since the test
method 13 often not detailed, a vacuum test, for instance, can be
performed that goes through an approximate procedure which satisfies
the spirit of the test, but may provide a result of dubious merit. The
Construction Quality Assurance Plan for the project should, therefore,
outline the procedure to be followed, in the absence of a standard to
reference. The CQA Contractor should then include in his
responsibilities, the confirmation that the method and procedure are
as contained in the CQA Plan.
Acceptance cr".eria for a qualitative test are of necessity vague.
The purpose of the test is to provide an evaluation of the continuity
of the flexible membrane liner seams. This is indicated by the lack of
detection of a leak by the test. This 'evidence1 of continuity must
not be ccnfused with indication of the quality of the seam, which
at present can only be determined by quantitative methods (i.e.,
destructive testing). Finding no leaks can lead to the conclusion of
seam continuity, but such is not a result in itself.
4.5.2.3 tate of
The testing requirements identified herein for flexible membrane
liners represent the state of practice. The A5TM Committee 035 on
Geotextlles and Related Products is actively pursuing new test methods
prepared specifically to address the testing of flexible membrane
liners. This 1j a slow process, however, and the current procedures
will be used for some time yet. As the new tests become available,
however, they should be substituted for the current practice of
modifying an otherwise inappropriate test method. The lack of test
procedures is not so much an indication of the negligence of the
standards-writing systems, but is rather an indication of the growth
of the industry and the proliferation of applications of flexible
membrane liners.
4-87
-------
It should be noted that there are many independent test methods,
some of which are very applications-specific. These tests are r.st
standardized, and in many cases, never will be. In particular
circumstances, however, they may represent the state of the art, and
1t may be desirable to use some of these procedures for specific
projects. Consequently, the designers of these facilities should
remain current with the testing literature, 1n which these ^st
procedures are normally well documented.
4.5.3 Other Geosynthetlc Materials
4.5.3.1 Procedures
Although the testing requirements for the various geosynthetics
(other than FML's) vary from material to material, they are discussed
together primarily because the same tests are normally used, and the
differences lie more -vith the menu of tests required for each
material. Discussion he.ein shall be limited to geotextiles, geonei.3,
and geogrlds, those being the geosynthetic materials most commonly
used in these types of lining systems. As other products, such as
composite products, gain acceptance for use in hazardous waste
management units, they will require additional attention.
The following tests are used for the determination of the physical
and mechanical characteristics of these materials in the performance
of a Construction Quality Assurance Program. It should be noted that
all of these tests are performed in the laboratory. At present,
neither destructive nor nondestructive tests arc carried out on these
materials in the field.
4.5.3.1.1 Laboratory Testing - Specifications Conformance
Conformance testing is carried out for geotextiles, geonets, and
geogrlds for compliance with the specifications, on samples taken from
the rolls of material supplied to the site. The requirement for this
testing will vary from one site to another, but these samples should
be collected at a minimum frequency of one per 10 000 m2 (100,000
4-88
-------
ft'). The testing requirements for conforrance for each of these
materials are as follows, at the least.
• For geotextiles:
. Mass Per Unit Area - ASTH D3776 "Mass Per Unit Area
(Weight) of Woven Fabric" [ASTM, 7.01]. The mass per unit
area Is always one of the primary criteria specified for
geotextiles. It is particularly important for geotextiles
which are intended to function as a cushion, but also in
general it is an indicator of the adequacy of the mechanical
properties.
Grab Strength - ASTH D1682 "Breaking Load and Elongation of
Textile Fabrics" Section 16, using a 4 in. x 8 in. (100 mm x
200 mm) specimen, 3 in. (75 mm) gauge length, 1 in. wide x 2
in. long grips, 12 in. (300 mm) per minute strain rate, and
a Constant Rate of Extension (CRE) Machir°" [ASTM, 7.01].
This test is well suited for use as a quality
control/conformance test as it eas.,', and can be performed
very quickly. Most of the strength tests for geotextiles
(including this one) are borrowed from the textile test
methods, and are contained in this standard.
Tear Strength - ASTH D4533 "Trapezoid Tearing Strength of
Geotextiles" [ASTM, 4.08]. One of the few geotextile-
specific test methods available at present, this test
determines the tearing strength of geotextiles, propagating
from an initial cut.
Burst Strength - ASTM D3786 "Hydraulic Bursting Strength of
Knitted Goods and Nonwoven Fabrics - Diaphragm Bursting
Strength Tester Method" [ASTM, 7.01]. This test is borrowed
from the textile test procedures, and is commonly referred
to as the Mullen Burst Test. It consists of the rupture of a
specimen of gectextlle by the application of a fluid load,
on a membrane below the fabric. This parameter is frequently
Included in the specification of geotextiles.
4-89
-------
Puncture Strength - ASTM D3787 "Burst-'-} Strength of Knitted
Goods - Constant-Rate-of-Traverse (~-.T) Ball Burst Test"
Modified for Geote:;ti1es [ASTM, 7.01]. This is a test in
which a plunger with a round head is pushed through the
specimen ana the load recorded. This test will very shortly
be replaced by a new test for puncture specific to
geotextiles, currently being finalized by ASTM 035.
Wide Strip Tensile Test - ASTM D459S "Wide Width Strip
Tensile Strength of Geotextiles" [ASTM, 4.08]. This test is
normally Intended more as a design test than as a quality
control or confornance test. If, however, the geotextile
will undergo very :igh tensile loadings, it should be
Included as a conformance test.
Thickness - ASTM D1777 "Measuring Thickness of Textile
Materials" [ASTM, 7.01]. This procedure nas been slightly
modified to identify the "nominal thickness" as that
thickness under a compressive stress of 2 kPa (42 psf). Its
use as a conformance test 1s frequently confined to
applications in which the :eotextile is to be used as either
a cushion, or as a fluid transmission medium, for which the
thickness Is an important component of the specification.
Normal Permeability or Permittivity - ASTM D4491 "Water
Permeability of Geotextiles by Permittivity" [ASTM, 4.08].
Permittivity of geotextiles refers to the flow rate of a
fluid through the plane of the fabric. Both permittivity and
normal permeability are determined in this test, one being
derived from the other, and having different units. This
particular property is of interest primarily if the
geotextile 1s designed to act as a filter.
Apparent Opening Size - United States Corps of Engineers CW
02215 "Plastic Filter Fabric". This test, scheduled to be
replaced by an ASTM procedure, is of value in the evaluation
of geotextiles as filters for different types of soils. It
4-90
-------
determines an equivalent soil grain size for which 90
percent of the openings in the geotextile are smaller.
For geonets:
Polymer Specific Gravity - ASTM D7S2 "Specific Gravity and
Density of Plastics by Displacement" [ASTM, 8.01]. The
specific gravity of the geonet is usually specified to
conform to that required for the flexible membrane liner.
. Mass Per Unit Area - ASTM D3776 "Mass Per Unit Area (Weight)
of Woven Fabric" [ASTM, 7.01].
. Thickness - ASTM D1777 "Measuring Thickness of Textile
Materials" [ASTM, 7.01]. This procedure has been slightly
modified to identify the "nominal thickness" as that
thickness under a compressive stress of 2 kPa (42 psf).
For geoqrids:
. Mass Per Unit Area - ASTM D3776 "Mass Per Unit Area (Weight)
of Woven Fabric" [ASTM, 7.01].
Measurement of Spacing Between Strands - There is no test
method for this requirement, which should comprise the
measurement of the grid openings at several locations in the
sample.
. Wide Strip Tensile Test - ASTM D4595 "Wide Width Strip
Tensile Strength of Geotextiles" [ASTM, 4.08]. This test is
normally intended more as a design test than as a quality
control or conformance test. For geogrids, however, the
narrow strip tests are not sufficiently representative of
the strength.
Node Strength - The only test of this nature presently in
existence is a manufacturer-specific test in which a tensile
force 1s applied to the geogrid node, rather than the rib of
4-91
-------
the material. A specified proportion of the geogrid rib
strer,gth rnust be achieved through the node.
4.5.3.2 Effectiveness of the_Tests -^Acceptance Criteria
4.5.3.2.1 Geotextile Conformance Tests
The conformance 'ests used for geotextiles are at present a
mixture of textile a.1 geotextile test methods. The applicability of
some of the former is suspect because of the nature of geotextiles
versus most textiles, e.g., the strengths of geotextiles are typically
orders of magnitude higher than are required for textiles. As a
component of the overall Construction Quality Assurance Program, these
tests are effective in the contex1- that they provide a relatively fast
and easy way to confirm that the properties of the material provided
to the project site conform to the specifications.
Not all of the tests cited will necessarily he required for
conformance testing on a given project, and for particular situations,
the design engineer may wish to add other tests. In general, some of
the textile test methods are inappropriate for geotextiles. The
development of standard test methods is very slow, but progress is
being made. At present, ASTM Committee D35 has over 20 test methods
for geotextiles in draft, and as these test procedures become
available, they should supersede the k ts described or referenced
herein.
The acceptance criteria for geotextile conformance tests are a
function of the use of the geotextile (e.g., cushion, filter) as well
as the configuration of the unit in which it is to be incorporated.
The properties related to the tests indicated in Section 4.5.3.1,
above, are those which relate mo:t appropriately to the normal
application of geotextiles. and for which the conformance of the
material to the specification is critical to performance. It is the
responsibility of the Designer to write specifications for testing
that clearly indicate that the requisite properties of the geotextiles
have been satisfied.
4-92 \
\
-------
4.5.3.2.2 Geonet Conformance Tests
rcr~an:e testing for goonets is more difficult in that there is
only one real design criterion which is absolutely critical, that
being the transmissivity of the geonet. This is a very specialized
test, and Is design-.'elated more than Construction Quality Assurance
related. Consequently, the most that can be done in tne form of
conformance testing is to confirm the materials properties, and the
dimensions of the geonet, which will confirm its compatibility and
stress-strain response (in particular, long term creep under constant
load), as well as its flow capacity.
4.5.3.2.3 Geogrid Conformance Tests
Just as only one property of a geonet is of concern
(transmissivity), only one property of geogrids is of concern with
regard to conformance of geogrids, that being tensile modulus
(tension/strain characteristics). Confirmation of the tensile
properties of geogrids requires an understanding of the manner in
which tensile properties, as reported by the Manufacturers, are
derived. This is more of a design concern, and the values required for
the conformance tests must reflect the numbers reported or guaranteed
by the Manufacturers as well as the numbers used for design.
4.5.3.3 Current State of Practice
The current state of practice with regard to the testing of
geotextiles, geonets, and geogrids for conformance, and even in
general, is that the industry is coping in the absence of a well
defined set of appropriate geotextile, geonet, and geogrid standard
test methods. It is more of a problem with these materials than with
FMLs because the established t°sts for plastic sheeting, for example,
are generally applicable to FHLs. Due to the differences in
properties of geotextiles versus textiles, on the other hand, many
textile test methods have been modified to make them usable, but often
leave gaps in their usefulness. This problem is, however, decreasing,
and in fact, as the ASTM D35 Committee prepares and issues a more
complete set of geosynthetics-specific tests, the problem will
4-93
-------
indually, tut eventually, go away. The designers and Construction
'it./ Ajj'jranc: Consultants must remain current en the status of the
activities of Committee D35, and ensure that the properties and tests
specified are current. This will be parties.arly important, because
the Manufacturers will likely all change over their reporting and in-
plant testing as soon as the new tests become available.
4-94
-------
96
92
12
14
16
18 20 22
CCMItMI (1)
2G
Flgtr- 4-1. Characteristic moi sture-densHy curves for a low-
permeabllHy soil for different levels of compactlve
effort (the compactlve effort Increases from Curve 1 to
Curve 4).
4-95
-------
12 14 16 18 20 22 24 25 23
HJ:S!IS£ CCSltllT (11
Figure 4-2. Characteristic curves for a low-permeability soil,
illustrating the moisture/density, moisture/permeability,
and moisture/degree of dispersion relationships at a
given level of ccmpactive effort. Adapted from [Mitchell
et al, 1965; and Seed and Chan, 1959].
4-96
-------
U 16
18 20
iwsn.se ccHffir (:)
ICO lb/ff wcpt - 22 •!.
Required compaction - 55 % (0.95 « ICQ - 95 lb/ff)
Range of moistur0 contents - 19 to 24 %
Figure 4-3. Moisture-density relationship: parameters pertinent to
compaction specifications.
4-97
-------
CHAPTER 5
LAND TREATMENT UNITS
-------
5.1 INTROD'JCT'''I
'!;e purpose of Chapter 5 is to provide a discussion of the new
standards proposed by EPA for new permitted and interim status land
treatment units.
Chapter 5 is comprised of 4 sections which are summarized below.
Section 5.1 outlines the scope of this chapter.
Section 5.2 is a discussion of regulatory approaches to land
treatment and EPA's proposal to expand the soil-core and soil-pore
liquid monitoring plans required of land treatment units. This
section addresses: the viability of land treatment; the unique
challenges leak detection at land treatment units presents.
Section 5.3 is a discussion of current regulations and proposed
changes to those regulations at permitted land treatment units. This
section addresses: the current regulations; the approach to statutory
requirements; detection confidence levels; inspection of monitoring
equipment; and response action plans.
Section 5.4 is a discussion of the additional requirements
proposed for Interim status land treatment facilities. This section
addresses: current interim status regulations; interim status
implementation differences; and amendments to the interim status
monitoring plan.
5.2 REGULATORY APPROACH TO LAND TREATMENT
Land treatment is viewed by EPA as a viable method of hazardous
waste land treatment and disposal for some types of hazardous waste.
Land treatment involves applying hazardous waste on the soil
surface or incorporating it into the upper layers of the soil in
order to degrade, transform, or immobilize hazardous constituents.
Unlike landfills, waste piles or surface impoundments, land treatment
does not use double liners and leachate collection systems to contain
the wast. Rather land treatment relies on the physical, chemical,
and biological processes occurring in the upper layers of the soil for
5-1
-------
the degradation, transfoi ma t ion , and i;i,.<:cbi 1 i Z3t ion of Hazardous
co"S-' t'.::.>f ts. In this sense, land treatment can be viewed as an c^en
system.
Land treatment units depend upon a number of soil and waste
interactions for success, therefore it is especially important that
the unit be carefully operated and monitored. The current design and
operating requirements under Parts 264 and 265 require owners or
operators of land treatment units to include monitoring in the
unsaturated zone to provide information that the owner or operator
will use in modifying his operating practices to maximize the success
of treatment processes. The orinciple objective of the current
unsaturated zone monitoring requirements is to provide effective
management of liquids at the unit to minimize the risk of groundwater
contamination. At surface impoundments, waste piles, and landfills
this objective is met by the double liner and leachate collection
system, and the final cover that prevents liquids entering the unit
and migrating into the subsoils.
The regulatory approach to land treatment, however, does seek to
minimize uncontrolled migration of hazardous constituents into the
environment. This is accomplished by using a defined layer of surface
and subsurface soils (referred to as the "treatment zone") to degrade,
transform or immobilize the hazardous constituents contained in the
leachate passing through the system. These treatment processes
achieve the same general objectives as the liquids management strategy
used at other types of land disposal units in that they act to prevent
hazardous constituents from migrating into the environment.
Wh the objective remains the same, the general approach must be
modified somewhat for land treatment units. Land treatment units are
dissimilar to other land disposal units in that they are not designed
and operated to minimize all releases to ground water. On the
contrary, they are open systems that freely allow liquid to move out
of the unit. The goal of land treatment, therefore, is to reduce the
hazardousness of waste applied in or on the soil through degradation,
transformation and immobilization processes.
5-2
-------
Two monitoring procedures, soil-cora dtrJ soil-pore liquid
monitor'.ng, are required in EPA existing rules. They are Intended to
complement one another. Soil-core monitoring primarily will provide
information on the movement of "slower-moving" hazardous constituents
(such as heavy met?ls). Soil-pore liquid monitoring will provide
essential additional data on the movement of fast-moving, highly
soluble hazardous constituents that soil-core monitoring may miss.
For example, 1f a significant increase of a hazardous constituent
Is detected in unsaturated zone monitoring, the owner or operator is
required under the existing Part 264 requirements to examine more
closely the unit characteris s that significantly affect the
mobility and persistence of that constituent. These significant unit
characteristics may Include treatment zone characteristics (e.g., pH,
cation exchange capacity, organic matter content), or operational
practices (e.g., waste application metho- and rate). Modifications to
one or more of these characteristics may be necessary to maximize
treatment of the hazardous constituent within the treatment zone and
to minimize additional migration of that constituent to below the
treatment zone".
5.3 PERMITTED LAND TREATMENT UNITS
5.3.1 Current Regulations
EPA's current regulations (-10 CFR 264, Subpart M) for land
treatment units require that the owner or operator of the unit must
(among other requirements): (1) establish a land treatment program
that is designed to ensure that azardous constituents placed in or on
the treatment zone are degraded, transformed, or immobilized within
the treatment zone (264.271(a)); (2) demonstrate, prior to application
of waste, that hazardous constituents in the waste can be completely
degraded, transformed or immobilized in the treatment zone
(264.272(a' (3) design, construct, operate and maintain the unit to
maximize the degradation, transformation, and immobilization of
hazardous constituents in the treatment zone, to minimize run-off and
run-on, and to control wind dispersal and to provide a run-off, run-
on, and wind dispersal inspection program (264.273(a)-(M); (4)
establish an unsaturated zone monitoring program (264.278), and (5)
5-3
-------
imolc" t certain operational, ronit,,Mng and inspection requirements
Juri"9 the closure and post-closure care periods (25-1.280''.
As stated above, 40 CFR 264.278 requires that all land treat-rent
units have an unsaturated zone monitoring program. This program must
provide the capability of determining whether hazardous constituents
have migrated below the treatment zone. The monitoring program must
Include the use of both soil cores and devices to measure soil-pore
liquid (such as lysimeters). The unsaturated zone monitoring program
system must consist of a sufficient number of monitoring points at
appropri e locations and depths to yield samples that:
(1) Represent the quality of background soil-pore liquid and the
chemical make-up of soil that has not been affected by
leakage from the treatment zone; and
(2) Indicate the quality of soil-pore liquid and the chemical
make-up of the soil below the treatment zone.
40 CFR 264.278(d) requires the owner or operator to conduct soil
monitoring and soil-pore liquid monitoring immediately below the
treatment zone at frequency u.id timing specified by the EPA Regional
Administrator. The owner or operator must determine whether there is
a statistically significant change over background values for all
hazardous constituents requireJ under Section 254.278 (a). This
determination must be made belov, the treatment zone each time the
owner/operator conducts soil monitoring and soil-pore liquid
monitoring (Section 264.278(f)).
RCRA Section 3004(o)(4) requires a leak detection system capable
of detecting leakage at the earliest practicable time for all new and
existing land treatment units. This requirement will become effective
24 months after promulgation of the final liner/leak detection rule.
T^ satisfy the statuary leak detection requirements of RCRA, EPA
will propose the following additions to the current do CFR 264.278
unsaturated zone monitoring requirements: (1) a 95-percent confidence
level for detecting hazardous constituents below the treatment zone;
(2) detection of leakage at the earliest practicable time; (3)
monitoring to be conducted above the seasonal high water table; (4) a
5-4
-------
rescon3e action plan for major and widespread leakage; and (5)
irsnecticn of unsatuiated zone monitoring equipment.
5.3.2 Approach to Statutory Requirements
The land treatment process is fundamentally different than land
disposal. The land treatment process involves biodegradation of
wastes in the upper layers of the soil thereby reducing the levels of
hazardous constituents during the degradation process (USEPA, 1983).
Land disposal essentially involves the use of "liquids management and
containment" technologies. Since the land treatment and land disposal
processes require fundamentally different types of waste management
structures, fundamentally different approaches are required to satisfy
the statutory leak detection requirement of RCRA.
EPA has elected to employ the existing unsaturated zone monitoring
requirements under 40 CFR 264 to satisfy statutory leak detection
requirements at new and existing permitted units. The EPA is planning
to expand on the monitoring requirement by adding a monitoring
confidence level of detection of 95 percent and requiring detection in
the earliest practicable time. The unsaturated zone monitoring is the
vehicle that EPA is proposing to be ! for defection of leakage.
If the owner or operator detects concentrations of constituents
statistically exceeding background levels, then appropriate
operational controls will be required such as reducing the application
rate.
The owner or operator of new and existing units at permitted or
interim status facilities will be required to include in the permit
application or operating plan a response action plan for widespread
leakage. This will be prepared and submitted before waste can be
received at a new unit. Leakage less than widespread will not be
required to have a RAP because the treatment requirements, as noted
above, have a process ^ address smaller leakage rates through
operational changes at the unit.
5-5
-------
5.3.3 Detection Confldence Level,
t?A is proposing to a a 95 percent confidence level for
detection of hazardous constituent migration out of the land treatment
unit to the existing 40 CfR 264.278 requirements. Land treatment
relies on the deg.adation, transformation or immobilization of
hazardous waste within the treatment zone. However, land treatment
units have no barrier to downward migration, and ground water can be
located as close as one meter to the bottom of the treatment zone.
For these reasons, EPA takes the position that the owner or operator
must detect leakage out of the unit at the earliest practicable time
and at a 95-percent confidence level
By requiring a 95-percent confidence level, EPA will be assuring
that the unsaturated zone monitoring system will consist of a
sufficient number of sampling points at appropriate locations and
depths to determine the spatial and temporal variations in constituent
concentrations beneath the entire treatment zone. A properly designed
and well managed site with uniform waste application will require
fewer sample locations than a poorly managed s'te. The owner or
operator must consider site-specific variations and the relative
uncertainty associated with soil-pore liquid sampling procedures in
developing the unsaturated zone monitoring program.
EPA 1s proposing the confidence level value of 95 percent as a
result of recently developed guidance [USEPA, 1986a; U3EPA, 1986b] on
unsaturated zone monitoring. This guidance explains what is required
of owners or operators to ccmpl> with the confidence level
requi rements.
A confidence level is the range within which the true value of a
parameter 1s to be found with a given probability. The reliability
expressed by the confidence level states the level of precision of the
sampling study. Three levels of confidence are commonly used: (i) 68
percent; (ii) 95 percent; and (Hi) 99 percent. These can be
expressed as ± 1 standard deviation, ± 1.96 standard deviations, and ±
2.58 standard deviations of the mean, which covers 68 percent, 95
percent and 99 percent respectively. Another way to describe a
confidence level Is to say that the probability 1s 0.32 (or 1 in 3}
that the value 1s outside of 1 standard deviation on either side of
5-6
-------
the rean, 0.05 (or 1 in 20) that the value is outside 1.96 standard
deviations on either side of the mean, or O.C1 (or 1 in 100) that the
.ilue s outside 2.58 standard aeviati^-s en cither side of the -rean.
When results have to be absolute, a 99 percent confidence level is
used. When funding or other resources are limited, or whsn
reliability Is comparatively unimportant, the 68 percent confidence
level nay be acceptable. Environmental sampling, however,
traditionally attempts to attain a level of 95 percent confidence.
Establishing the level of confidence is often a judgment exercised
by the researcher based on the degree of reliability desired.
Selection of the confidence level affects many decisions relating to
the research studies, including the number of samples that must be
taken. Consequently, statistical sampling studies require a known
confidence level to be established before sampling begins.
A 95 percent confidence level was selected by EPA because it is
the generally accepted level of confidence for environmental studies,
it provides a high level of reliability and it sets a reasonable
standard for reliability and precision.
5.3.4 Monitoring Periods
The mandate of RCRA 3004(0)(4)(B) requires an approved leak
detection system to be utilized which is capable of detect,.ig leaks at
the earnest practicable time. To meet this requirement, EPA is
proposing to require quarterly monitoring to detect hazardous
.onstituents at the earliest practicable time.
Two kinds of monitoring are used at land treatment units: soil
core and soil-pore liquid monitoring. Soil-core monitoring is used to
evaluate the transport of relatively slow-moving waste constituents
while soil-pore liquid monitoring is designed to detect rapid pulses
of mobile waste constituents that rr- be rapidly transported through
the unsaturated zone to ground wa and that are unlikely to be
observed through regularly schedule? analysis of soil cores.
Appropriate timing of soil-pore liquid monitoring is essential,
and ideally sampling is performed after precipitation or snow melt is
5-7
-------
S3T.pl ing depends on specific
'y of water movement
sufficient, to g e n e ~ 31 e ' e :: • 11 e !
soil and s:te ccnd?cicns t",ii o'c
in f'.e so)!}.
In the EPA's te.nnica! ucu~e-t c- land treatment [USEPA, 1983c],
quarterly sampling of soi i-rore '. 'Q-j'i is recommended although more
frequent sampling r,ay ts rjcomre-dec! in areas of high rainfall or
highly perreable soils or jc la~d treatment sites where waste is
frequently applied.
The requirement for cui-terN restoring of soil-pore liquids is
considered to be a reascnat'j min;TUTI frequency because even if fast-
moving hazardous ccnsti tue-rs are to ~ove out of a treatment zone,
they usually migrate within PI days foT:wjng waste application.
Soil-cere monitoring is used to evaluate the transport of waste
constituents which ray rove thrc^'in t~» soil profile fairly slowly.
Waste constituents transport wiy t= sl:v because of insufficient soil
moisture to leach through t-= sy5-«Ti, or a natural or artificially
occu'-ing layer or horizon -f lev hjrraulic conductivity or waste
consi.cuents that cr.ly exnif: a lev t. moderate mobility relative to
water in soil.
EPA reccirnisnds [USEF.-., :i:3c] t
at least se~i annually. ^.,i~ter
provide the earliest pract'-nble
sampling and soil-pore sa~ol"c.
5.3.5 Inspection
at ^ail-core sampling be conducted
sapling therefore is a way to
sa^ detection for both soil-core
EPA is also pressing to =
to require c«ners or operate:;
the unsaturated zon? moritr*
post-closure care period cf a
owner or operator to est
deterioration, malfunction, :
monitoring equipment. T-
effectiveness of controls
constituent "igratic" bejir:
exceed background le.-els. T1
-------
keep a detailed log of all inspection information to demonstrate
ccirn 1 iance with unsaturated zone i:;oni tcring oeniiit require;: 2.'its. The
Reyk.'ial Adinir "tratcr (RA) may requiie additional inspection aid
monitoring requi -iients which will be specified in the permit.
5.3.6 Response Action Plan
Existing regulations (40 CFR 264.278) require the owner or
operator to notify the Regional Administrator within 7 days when there
is a statistically significantly increase in hazardous constituents
below the monitoring zone. The owner or operator also must submit to
the RA within 90 days an application for a permit modification to
change the operating practices at the facility to maximize the success
of degradation, transformation, or immobilization processes in the
treatment zone.
EPA is proposing a requirement that after the effective date of
the liner/leak detection rule, owners or operators of land treatment
units develop response action plans (RAPs) for "widespread leakage"
before waste is received. The RAP will set forth actions to be taken
upon finding widespread leakage. Widespread leakage is defined as a
statistically significant increase of hazardous constituents at 50-90
percent or more of the unsaturated one sampling points. This
increase represents migration from the total areal A tent of the unit
posing a threat to groundwater quality. Widespread leakage requires
an immediate response action plan.
The EPA believes that isolated leakage (a statistically
significant increase of hazardous constituents at fewer than 50
percent, of the unsaturated zone sampling points) presents a minimal
threat to ground water design and operation may need to be adjusted
within the unit to minimize local hazardous constituent migration out
of the land treatment unit. However, if statistically significant
increases in hazardous constituents occur at 50 to 90 percent of the
unsaturated ~one sampling points (i.e., widespread leakage), then it
is believed the threat to human health and the environment at this
level of migration out of the unit increases to the point that
substantive changes in the operation of the unit or closing the unit
may be necessary.
5-9
-------
The possible coui3es of action to take ucicn finding widespread
leakage include charging the operating practices or closing t';e
facility. Changing operating practices iray include changing the type
of waste treated, the timing of application, a reduction of the amount
applied, or a reduction in the application frequency. Closing the
facility may be necebsary if changing operating practices cannot be
shown to protect ground and sut face water or if the owner or operator
finds the changes to be cost prohibitive.
EPA considered other possible response actions for widespread
leakage, but chose not to include them. These actions include
increasing the frequency of groundwater monitoring, excavating the
unit, and Installing a cover over the unit. The EPA lias taken the
position that using increased groundwater monitoring would be too slow
to detect contamination and does not achieve the goal of preventing
groundwater contamination. Excavating a land treatment unit and
disposing of the contaminated soil would be excessively expensive and
would achieve protection similar to closing the unit. The last option
considered is installing a temporary landfill cover over the unit or
part of the unit. This is counter to the principles of land
treatment, which are to allow natural aerobic processes to degrade
waste.
5.4 INTERIM STATUS LAND TREATMENT UNITS
5.4.1 Current Regulations
EPA's current regulations (40 CFR 265, Subpart M) require owners
or operators of interim status land treatment units to install and
operate an jnsaturated zone monitoring system with the capability of
detecting the migration of hazardous constituents vertically from the
land treatment facility's treatment zone.
Data on the background concentrations of hazardous waste and
hazardous waste constituents also must be provided by the monitoring
plan, which must include the use of soil cores for soil monitoring and
lysimeters (or a similar device) for soil-pore liquid monitoring.
5-10
-------
The current regulations for interim status land treatment unit
unsaturatod z-jr.e monitoring arc less stringent than those proposed by
EPA. Cu;rent regulations are:
(1) The owner and operator must demonstrate that the depth at
which soil and soil-pore water samples are taken is below
the treatment zone;
(2) The owner and operator must base the number of soil and
soil-pore water samples on the variability of the hazardous
waste constituents and soil types. Furthermore, sampling
frequency and sampling time must be based on the frequency,
time, and rate of waste application, proximity to ground
water, and soil permeability.
A copy of the unsaturated zone monitoring plan and the rationale
used to develop the plan must be filed at the land treatment facility
by the owner or operator according to 40 CFR 265.278 (d).
The owner and operator also must analyze the soil and soil-pore
water samples for hazardous waste constituents that exhibit an
extraction procedure (EP) toxicity characteristic (40 CFR 265.273(a)),
and he must determine the presence and conc'itration 1n the sample of
any substances on the federal hazardous waste list specified in CFR
2G1, Subpart D (40 CFR 265.273(b)).
The Hazardous and Solid Waste Amendments of 1984 (RCRA 300-1(0^(4))
require interim status land treatment facilities to utilize an
approved leak detection system at the earliest practicable time. This
requirement becomes effective 24 months after the final liner/leak
detection rule Is promulgated.
To satisfy the statutory requirements, EPA is proposing to require
interim status units to achieve the same monitoring standards as
permitted units. The more stringent standards are based on the
proposition that Interim status land treatment poses the same threat
of adverse impacts to human health and the environment as permitted
land treatment. Therefore, EPA believes existing units can comply
with the standard in the same manner as new units and thus provide the
same level of protection to human healfh and the environment.
5-11
-------
5.4.2 Proposed Interim Status Monitoring Requirements
EPA proposes to replace current regulations governing interim
status land treatment monitoring (40 CFR 265.270) with present and
proposed additional monitoring requirements (40 CFR 264.278) for
permitted land treatment units. These proposed regulations will
require essentially the same monitoring plan at all land treatment
units. The upgraded monitoring plan that is proposed has the following
additional elements: (1) a 95 percent confidence level for detecting
hazardous constituent migration from the treatment zone, (2) detection
of hazardous constituent leakage at the earliest practicable time, (3)
conduction of soil and soil-pore liquid monitoring above the seasonal
high water table (SHWT), (4) a response action plan for widespread
leakage and (5) unsaturated zone monitoring equipment inspections.
5.4.3 Interim Status Implementation Differences
Droposed requirements for permitted and interim status land
tr-j.i.ent units are similar, but implementation procedures differ in
two ways: (1) The interim status land treatment unit owner or operator
must have a written leak detection plan at the facility and (2) a copy
of the plan must be sent to the EPA Regional Administrator.
The minimum requirements for a written interim status leak
detection plan are:
(1) A description of how soil and soil-pore liquids
will be monitored to determine at 'he earliest
practicable time whether hazardous constituents have
migrated out of the treatment zone. This monitoring
program must cover all areas likely to be exposed to
hazardous waste and leachate through the facility's
active life and post-closure care period. Hazardous
constituents or principal hazardous constituents to be
monitored must be identified.
(2) A description of the number, location and depth of
soil-pore liquid monitoring devices necessary to
represent to a 95 percent confidence level the
following characteristics;
5-12
-------
a) soil and -soil-pore liquid quality below
the treatrent zone,
b) the quality of L\ic; jrcur'.d soil i;id soil-
pore liquid,
(3) A description of methodology for establishing
background values for each hazardous constituent to be
monitored.
(4) A description of the frequency, time, and depth of
soil and soil-pore liquid monitoring based on the
frequency, time and rate of waste application and on
the soil permeabi1ity.
(5) A description of sampling and analysis procedures
that are designed to ensure that sampling results
provide a reliable indir3tion of soil-pore liquid
quality and the chemical composition of soil below the
treatment zone. Procedures for sample Collection,
preservation and shipment, along with analytical
procedures and chain of custody control should be
included.
(6) A description of the statistical procedure to
determine if there is a significant increase over
background values In the monitoring data. This
description must include the amount of time allowed
between drawing a sample and determining the
statistical significance of that sample. The plan must
specify a statistical procedure that is appropriate
for the distribution of data used to establish
background values. This procedure must provide a
reasonable balance between the probability of a false
determination and failure to identify migration of
hazardous waste
(7) A response action plan that describes action to
take if 1t Is determined that a widespread leakage has
occurred.
5-13
-------
5.4.4 Monitoring P1?n Afpcn'Jir.ent_5
Once a statistically sij.iificant -u^r^ase of hazardous wa->te
constituents is determined by the o-,ner or operator to have occurred
at an interim status facility, the EPA Regional Administrator must be
contacted in writing within 7 days. This notice must identify the
constituents detected and include the preliminary concentration
levels. An amendment to the operating plan must be submitted in 90
days of the determination. This plan must show that operating
practices have been modif'-d sufficiently to maximize the success of
degra . ,t1on, transformatiun, or immobilization processes in the
treatment area.
5-14
-------
REFERENCES
I. CITED REFERENCES
"•met : can Society foe lasting and "aerials, A;>rujj _Qock of"._^.lTM
r-'j'lJv-'js, Vol. 7.01 (r tiles - 'tarns, FatM .s, Ge::erii Test
•;?in JJ3/1, ii'54.
American Society for Testing and Materials, Annual Book of ASTM
Standards. Vol. 4.08 (Soil and Rock, Building Stones), 1985a.
American Society for Testing and Materials, Annual Book of ASTM
Standards, Vol. 8.01 (Plastics (I): C 177 - D 1600), 1985b.
American Society for Testing and Materials. Annual Book of ASTM
Standards, Vol. 8.02 (Plastics (II): 0 1601 - 0 3099), 1985c.
American Society for Testing and Materials, Annual Book of ASTM
Standards. Vol. 9.01 (Rubber, Natural and Synthetic - General Test
Methods; Carbon Black), 1985d.
American Society for Testing and Materials, Annual Book of ASTM
Standards, Vol. 4.04 (Roofing, Waterproofing, and Bituminous
Materials), 1986.
Apgra, M. and Langmuir, D., "Groundwater Pollution Potential of a
Landfill Above the Water Table", Groundwa'ier. Vol. 9, No, 1971,
pp. 76-96.
Bass, J.M., Lyman, W.J., and Tr^tnyek, J.P., "Assessment of Synthetic
Membrane Successes and Failires at..Waste Storage and Disposal
Srte_s", ' -oort to EPA, NTIS No. PB85-245 637,'AS, Arthur D. Little
Inc., ISuj.
Benson, R.C., Glaccum, R.A., and Noel, M.R., "Geophysical Techniques
for Sensing Buried Wastes and V/aste Migration". Report to EPA, Las
Vegas, IJV, Contract No. 68-03-3050, Dec 1982.
Bjerrum, L. and Huder, J., "Measurement of the Permeability of
Compacted Clays", Proceedings, Fourth Inteir Uiorial Conference on
Soil Mechanics and Foundation Engineering, London, Vol. 1, 1957,
pp. 6-10.
Bonaparte, R., Williams, N., and Giroud, J.P., "Innovative Leachate
Collection Systems for Hazardous Waste Contains ~t Facilities",
Proceedings, G technical Fabrics Conference '85, Cincinnati. CH,
Jun 1985, pp. 9-34.
Boynton, S.S. and Daniel, D.E., "Hydraulic Conductivity Tests on
Compacted Clay", Journal of Geotechmcal Engineering, ASCE,
Vol. Ill, No. 4, Apr 1985, pp. 465 - 478.
-------
lock, 1977.
Brown, K.U.. Thomas, J.C., Lytton, R.L., Ja> awi ckra-u P., and fi.ihrt,
S.C., "Quant HicaU?|i__oJ_L_pjik_Rj:tes Jj"iouJ'i !lPl?3_jn 1. 3ndfj.il
Lu-".-3"," U3EPA Report CR 6109 10," Cincinnati , U'i', ~[D/UE~ M!55!f.G CM
Canadian Standards Association, "Guide for Selecting and Ir'plcmentjng
the CAN3-Z299-85 Quality Assurance, Program Standard", CAN3-Z299-
05, Canadian Standard Association Publishers, Ontario, Canada,
Nov 1986, 63 p.
Castro, A. and Timmons, R., "Porous Teflon: Its Application _ ui
Groundwater Sampling", Timco Manufacturing Inc., Prairie Ou Sac,
Wl, 1983.
Cedeigren, II. R., "Seepage, Drainage, and Flow Nets", John Wiley and
Sons Inc., New York, 1977.
Christopher, B.R. and Hultz, R.D., "Geotextile Engineering Manual",
FHWA-DTFH61-80-C-00094, 1984.
Daniel, D.E., "Predicting Hydraulic Conductivity of Clay Liners",
Journal of Geotechnical Engineering, American Society of Civil
Engineers, Vol. 110, No. 2, Feb 1984, pp. 285-300.
Daniel, D.E., Trautwein, S.J., and McMurtry, D.C., "A Case History of
Leakage from a Surface Impoundment", Proceedings of a Symposium on
Seepage and Le aka go _from Dams and Impoun-j 'ien t_s, edited by R.L.
Volke and W.E. elly, American Society of Civil Engineers, Denver,
CO, 1985, po. 220-235,
Davis, J.L., Singh, R., Waller, M.J., and Gower, P., "Time Domain
P-ef 1 ec tometry and Acoustic E mission Monitoring Techniques for
Locating Liner Failures", Report to EPA, Cincinnati, OH, Contract
No. 68-03-3030, Jun 1983.
Davis. J.L., Wa''?r, M.J., Stegman, B.C., and Singh, R., "Evdluat ion
of Time Dc::,oin Ref lectcmetry and Acoustic Emission Techniques to
Detect and Locate Leaks in Waste Pond Liners". Report to EPA,
Cincinnati, OH, Contract NO. 606-19-83-018, Sep 1983b.
Davis, J.L., Singh, R., Stegman, B.C., and Waller, M.J., "Innovative
Concepts f or Detecting and Locating Leaks in Waste Impoundment
Liner Systems: Acoustic Emissions Mcnito ring and Time Domain
Reflectometry", Report to EPA, Cincinnati, OH, Contract NO. 68-03-
3030, Apr 1984.
Engineering News Record (ENR), "Test Tracks Toxics Leaks". Jul 19,
1984, p. 12.
-------
ia.,- , i..-;., 'jc-jign u f Or a in iict-3.K!i u,--. CIML :•:•_•:. - "'.'.a :e
Estimation and Flow Patterns in Case of Leak", flL°:L§LLllllL5_.2_f_l!!S
]_ntcrn-311 _onaj Conference _o_n _Gpom?f"Lra"es, Vol. 2, Denver,
Colorado, Jun 1934, pp. <163--)b3.
fi'.i'O, i.H., "Manxes Et-inches: D"bi_t_e_t Forme '!e ! - -:•- ij 1 p-'-
J 2_J "ii '_L5." > "Thesis, University of Grenccle, r ice,
^'63 p.
Fukuoka, M., "Outl ine -f La_.-qg_ Scale Model Test on Waterproof
Membrane,", Unpubl ishea "Report, May 1985, 24 p.
Fukuoka, H., "Large Scale Permeability Tests for GecTiembrane-Subgrade
System", Proceedings of the Third International Conference on
Geotextijes. Vol. 3, Vienna, Apr 1986, pp. 917-922.
GCA, "Trip Report - Hughes Aircraft (Tucson)", Technology Division,
Bedfort, MA, 1984a.
GCA Corporation, "Evaluation of Leak Detection and Collection Systems
for Hazardous Waste Land Disposal Facilities", Draft Report, EPA
Contract No. 63-01-6871, U.S. Environmental Protection Agency,
Cincinnati, OH, Dec 1964b.
GCA Corporation, "Technical Suppott for Development and Analysis of
Hazardous Waste Disposal Regulations", EPA Contract No. 68-01-
6871, U.S. Environmental Protection Agencv, Cincinnati, OH, Apr
1986.
Gerhardt, R.A., "leachate Attenuation in the L':~saturated Zone Beneath
Three Sanitary Landfills in Wisconsin", Wisconsin Geological and
Natural History Survey, Information Circular 34, 1977.
Ghassemi, M., Crawford, K., and Haro, M., "Leach-ite Collection and Gas
Migration and Emission Proolems_ at Lindfins_ and Surface
Impoundments". Project Summary EPA/6CO/S2-86/017, U.S.
Environmental Protection Agency, Cincinnati, OH, Jut 1986, 3 p.
Gibbs, H.J., Hilf, J.W., Holtz, W.G., and Walker. F.C., "Shear
Strength of Cohesive Soils", Proceedings, cgsearch Conference_on
the Shear Strength of Cohesive Soiis. ASCE, Boulder, CO, 1960,
pp. 33-162.
Giroud, J.P., "Filter Criteria for Geotextiles , Proceedings, Second
International Conference on Geotextiles. Vol. 1, Las Vegas, NV,
Aug 1982, pp. 103 - 108.
Giroud, J.P. and Goldstein, J.S., "Geomembrane Liner Design", Waste
Age, Sep 1982, pp. 27-30.
Giroud, J.P. and Carroll, R.G., "Geotextile Products:i, Geotechnical
Fabrics Report, Summer 1903, pp. 12-15.
-------
\
\
-------
Giroud, J.P. and Frobel, R.K., "Geomembrane Products", Geo'ieclm ic
Fabrics Report. Fall 1983, pp. 38-42.
Gircud, J.P., "Geote/t i le a/id_J^qmej:ibr_an_es _- D_e_fin_ ''_s. Proper t •
ii:"! 2?3iiP"- industrial Fabrics Association irr, 31 'Ml K, mi , 3
Giroud, J.P., "Impermeability: The Myth and a Rational Approach1',
Pro ceedings of the International conference on Gecmenbranes ,
Vol". 1, Denver, CC, Jun 19840, pp. 157-162.
Giroud, J.P., " Assessment of Synthetic Membrane Performance at Waste
Disposal Facilities". Contract 68-03-1772, Woodward-Clyde
Consultants, Chicago, IL, Nov 1984c, 59 p.
Gircud, J.P. and Bonaparte, R., "Waterproofing and Drainage:
Geomembranes and Synthetic Drainage Layers", R.I.L.E.M. Symposium
Mo. II , Liege, Belgium, Jun 1984.
Giroud, J.P. and Stone, J.L., "Design of Geomembrane Liner for the
Proton Decay Experiment", Proceedings of the International
Conference on Geomembranes. Vol. 2, Denver, CO, Jun 1984,
pp. 469-474.
Giroud, J.P. and Fluet. J.E., Jr., "Quality Assurance of Geosynthetic
Lining Systems", Journal of Geotextiles and Geomembranes, Vol. 3,
No. 4, 1986, pp. 249-287.
Giroud, J.P., "Geotextiles and Related Products", Chapter 8 in, Vesic
Memorial Book. S. Sayed (Ed.), Gulf Publishing Company, 1987.
G'roud, J.P., Bonaparte, R., and Beech, J.F., "Leakage Through
Gecmembrane Liners" , to be published, 1987a.
Giroud, J.P., Bonaparte, R., and Beech, J.F., "Design of Leakage
Detection Systems", to be published, 1987b.
Hanson, E.A. and Harris, A.R., "Soil Water Samples Collected with
Porous Ceramic Cups", Soil Science Society of America Proceeding,
Vol. 39, 1975, pp. 529-536.
Haxo, H.E., "Lining of Waste Impoundment and Disposal Facilities".
Report to EPA, Cincinnati, OH, Report No. SW-870, 1983.
(Hazardous and Solid Waste Amendments of 1984).
Haxo, H.E., Miedema, J.A., and Nelson, N.A., "Permeability of
Polymeric Membrane Lining Materials, Proceedings, International
Conference on Geomembrane, Vol. 2, Denver, CO, Jun 1984,
pp. 151 - 156.
-------
-------
n'jxo, M.E., Nelson, N.A.,.and 'liede:-:!, J.A., "Solubility rj-a;:ieters
for Predicting Membrane - Waste Liquid Compatibility",
Proceedings, Eleventh Annua 1 _Ke_5e} rc_h__GvrrjD3j_u2_2!L-L/iricJ_ p i sp_o_3aj
of Hazardous Waste', fc>A/60079-85/013, Api" 1985," ;p. 198"- 212.
Head, K.H., "Vol. 2: Permeability. r
-------
F.-.-JI MCI , <.M., McCabe, '.'.'!., and Caldi.eso, L.F., "A.custic i^n ss i>:n
Monitoring of Seepage", Journal of the Geotechnicjl Engineering
Division. ASCE, Vol. 107, No. GT 4, Apr jo'si. pp. 521-525.
Koerner, R.M., Lord, A.E., Jr., and Luciano, V.A., "A Detection arj
Monitoring Techr.icue for Location of Geemcr'brane L •:-.3'•. s",
•''rocogjir-js of Int^-r.'.jt.io.'ial Conference on G;on-; j^Kmjjs, Vol. li,
Denver, CO, Jun 198T, pp. 379-184.
K erner, R.H., Lord, A.E., Crawford, R.A., and Cadwallader, M.,
"Ultrasonic NOT Testing of Geomembrane Seams", Proceedings,,.
GsosyntheLics '87. 1FAI, New Orleans. LA, Feb 1987, pp. 493-504.
Lambe, T.W., "The Permeability of Compacted Fine-Grained Soils", STP
163, ASTM, 1954, pp. 55-67.
Lambe, T.W., "Compacted Clay: Structure", Journal of the Soil
Mechanic: ind Foundation Engineering Division, ASCE, Vol. 125,
Part I, May 1958, p. 682.
Lambe, T.W., "Compacted Clay: Engineering Behavior" Journal of the
Soil Mechanics and Foundation Engineering Division, ASCE, Vol.
125, Part 1, May 1958, p. 718.
Landreth, R.E., "The EPA Testing Program .for Components of Treatment,
Storage, and Disposal Facilities", Proceedings, Geosynthetics '87,
IFAI, New Orleans, LA, Feb 1987, pp. 609-615.
Lowe, J. and Karafiath, L., "Effect of Anisotiopic Consolidation on
the Undralned Shear Strength of Compacted Clays", Proceedings,
Research Conference on the Shear Strength of Cohesive Soils, ASCE,
Boulder, CO, I960, pp. 837-858.
Mitchell, J.K., Hooper, D.R., and Campanella, R.G., "Permeability of
Compacted Clay", Journal of the Soil Mechanics and Foundation
Engineering Division, ASCE, Vol. 91, SM4, Jul 1965.
Mitchell, D.H., "Technology for Uranium Mill Ponds Using
Geomembranes". NUREG/U-3890, PNL-5164, Prepared for U.S. Nuclear
Regulatory Commission, Washington, D.C., Dec 1984.
Moore, C.A., "Landfill and Surface Impoundment Performance
Evaluation". EPA Report SW-869, Apr 1983.
Morrison, R. and Ross, D., "Monitoring for Groundwater Contamination
at Hazardous Waste Disposal Sites" ^rocegdings of the 1978
National Conference on Control of I-, jrdous Materials Spills,
Miami Beach, FL, Apr 11-13, 1978, pp. 281-286.
-------
Morrison, W.R., Gray, E.W., Jr., Paul, D.B., -ir.d Frobel, R.*..,
"lHJ>tallatipn of Flexible Membrane Joining in J^ U EJ_be_r_t__Fo j_e bji v
l^"Jiy_9JJL''• U.S. Department of the Interior, liurc-iu of
Reclunation, Denver, CO, 1981, 46 pp.
M/O'-', D.A.. Tyler, S.U., Gulnecht, 0.,)., and Mitch--?! 1 . D.M.. "Lc]k
"e t?_c_tj.9A-jLyjJ?™l_C?Jl__y_rJL'! Lb-iH_L!iU l9iilG31_ i'1p_p_yn:!;;!injments and Waste
Piles, Phase II". USEPA, Appendix A, Draft Final Report, 1987b,
270 p.
Radian Corporation, "Performance Analysis of Alternative and Minimum
Technology Design for Landfi11s, Surface Impoundments and Waste
Piles, Phase II", USEPA, Section 10, Craft Final Report,
Apr 1987c, 64 p.
Reynolds, W.D. and Elrick, D.E., "In-Situ Measurement of Field-
Saturated Hydraulic Conductivity, Sorptivity and the Alpha-
Parameter Using the Guelph Permeameter", Soil Sciences, Vol. 140,
No. 4, 1985a, pp. 292-302.
Reynolds, W.D. and Elrick, D.E., "Measurement of Field-Saturated
Hydraulic Conductivity, Sorptivity and the Conductivity - Pressure
Head Relationship Using the Guelph Permeameter", Proceedings,
National Water Well Association Conference on Characterization and
Monitoring of the Yadose (Unsaturated) Zone. Denver, CO,
Nov I985b.
-------
P'jjnj.as, 'rt.O. jnd Eirick, D.t., "A Method for S ih.j! taneous In-Situ
Measurement in the Vadose Zone of Field Saturated Hydraulic
Conductivity, Sorptivity and the Conductivity - Pressure Head
Relationship", Groundwater Monitoring Review. Winter 1986,
pp. 84-95.
^;-':-i'>, C.E., "Engineeriri_3__Dgs ign for Plaslici", E. Baer, Reinbold
P'jol. Corp., New York, 196<1, pp. 609-688.
Schrceder, P.R., Morgan, J.M., Walski, T.M., and Gibson, A.C., "The
Hydro logic Evaluat - on of Landfill Performance (HELP) Mode] ^
volume I User's Guide for Version 1", USEPA Technical Resource
Document, EPA/530-SW-84-009, Ji.- 1934a, 120 p.
Schroeder, P.R., Morgan, J.M., Walski, T.M., and Gibson, A.C., "The
Hydrologlc Evaluation of Landfill Performance (HELP) Model -
Volume II Documentation for Version 1", USEPA Technical Resource
Document EPA/530-SW-84-010, Jun 1984b, 256 p.
Schultz, D.W., Duff, B.M., and Peters, W.R., "Electrical Resistivity
Technique to Assess the Integrity of Geomembrane Liners", Report
to EPA, Contract No. 68-03-3033, Cincinnati, OH, Jul 1984.
Seed, H.B. and Chan, C.K., "Thlxotropic Characteristics of Compacted
Clays", Journal of the Soil Mechanics and Foundation Engineering
Division, ASCE, Vol. 83, SM4, Proceedings Paper 1427, Nov 1957.
Seed, H.B. and Chan, C.K., "Structure and Strength Characteristics of
Compacted Clays", Journal of the Soil Mechanics and Foundation
Engineering Division. ASCE, Vol. "85, SMS, Cct 1959.
Sherard, J.L., "The Upstream Zone in Concrete-Face Rockfill Dams",
Proceedings of a Symposium on Concrete Faces Rockfill Dams -
Design, Construction, and Performance, ASCE Geotechnical
Engineering Division, J. Garry Cooke and James L. Sherard, Eds.,
Detroit, MI, Oct 1985, pp. 618-641.
Stone, J.L., "Leakage Monitoring of the Geomembrane Liner for Proton
Decay Experiment", Proceedings, International Conference on
Geomembranes. Vol. 2, Jun 1984, pp. 475-480.
Telles, R.W., Unger, S.L., and Lubwoitz, H.R., "Technical
Considerations for De Minimis Pollutant Transport by Polymeric
Liners". Contract No. 68-03-3218, U.S. Environmental Protection
Agency, Cincinnati, OH, Sep 1986, 80p.
Terzaghi, K. and Peck, R.S., "Soil Mechanics in Engineering Practice",
John Wiley & Sons, New York, 1967, 729 p.
Todd, D.K., "Groundwater Hydrology", John Wiley & Sons, Inc., New
York, 1980.
-------
lurnuull, W.J. and rosier, C.R., "Stabilization of Materials by
Compaction", Journa 1 of the So H M_e_cha n i cs and Foundation
Engineering Division. ASCE, Vol. 82, SM4, r 1956.
U.S. Department •„, the f.'avy, "5qi_1__Me_c Panics Design Minual _7.J.",
,'JAVMC DM-7.1, May 1982, 354 p7
USEPA, "Lining of Waste Impoundment and__DJ_lP_o_sa_1—Eig.llLLLgl" • 5W-870,
U.S. Environmental Agency, Cincinnati, OH, Mar 1933a, pp. 45-113.
USEPA, "Evaluation of Land Disposal Facility Technologies and
Integration of Waste/Environment/Technology Characteristies to
Produce Facilities Profile", 1983b.
USEPA, "Hazardous Waste Landfill Treatment". Revised Edition,- SWr_JA^_
Apr 1983c.
USEPA, "Minimum .echnology Guidance oni_ Double Liner Systems for
Landfills and Surface Impoundments — Design, Construction, and
Operation", Draft Secona Version, U.S. Environmental Protection
Agency, Cincinnati, OH, May 24, 1985, 71 p.
USEPA, "Permit Guidance Manual on Unsaturated Zone Monitoring for
Hazardous Haste LandTreatment. Uni's". EPA/530-SW-86-040, U.S.
Environmental Protection Agency, Washington, D.C., Oct 1986a,
111 p.
USEPA, "Technical Guidance Document: Construe'ion Qua!ity Assurance
for Hazardous, Waste Land Disposal Facilities'". EPA/530-SW-86-031,
Oct 19865, 88 p.
USEPA, "Design, Construction, and Evaluation of Clay Liner for
Hazardous Waste Facilities". Draft EPA/530-SW-86-007, Mar 1986c.
USEPA, "1906 National Screening Survey of Hazardous Waste Treatment^
Storage, Disposal and Recycling Facilities - Summary of Results
fo.r/'DR Facilities Active in 1985", 1986d.
USEPA, "Background Document: _Bottom Liner Performance in Double-Lined
Landfills and Surface Impoundments", EPA/53U-SW-87-013, Prepared
by GeoServices Inc.", Apr 1987, 301 p.
Wallace, R.B. and Eigenbrod, K.D., "An Unprotected HOPE Liner in a
Subartic Environment", Proceedings of the International Conference
on Geomembranes, Vol. 1, Denver, CO, Jun 1984, pp. 73-78.
Waller, M.J. and Davis, J.L., "Assessment of Innovative Techniques to
Detect Waste Impoundment Liner Failure", Final Report, EPA
Contract No. 68-03-3029, U.S. Environmental Protection Agency,
Cincinnati, OH, Jun 1982, 139 p.
-------
-------
Day, j.R., and Daniel, D.-E., "Hydraulic Conductivity of Tv»o Prototype
Clay Liners". Journal of Geotechnical Engineering, ASCE, Vol. Ill,
Mo. 0, Aug 1985, pp. 957-970.
DJ,;, 5.R.. Daniel, O.E., and Ooynton, S.5., "Field PermeoDiIity lest
f or Clay Liners". A5_TM_ SIP 374 Hydr3ul ic ^i!lL'_LQ_JJL-5°-! ' - l'ld-
(llxi. 19'35> PP- 276-287.
Dunn. R.J. and Mitchell, J.K., "Fluid Conductivity Testing of Fine-
Grained Soils", Journal of Geotechnical Engineering, ASCE,
Vol. 110, No. 11, No» 1984.
Ertec Atlantic, "Land Disposal Liner/Locational Analysis Project",
Revised Draft Final Report to U.S. EPA Office of Solid Waste,
Washington, DC, Earth Technology Corporation, Somerset, NJ, Jan
1984.
Ghassemi, M., Haro, M., Metzger, J., and Powers, M., "Assessment of
Technology for Constructing and Installing Cover and Bottom Liner
Systems for Hazardous Waste Facilities". Vol. II, Report to EPA,
TRW, Torrance, CA, 1983.
Giroud, J.P., "Aging of PVC Geomembranes in Ut lium Mine Tailing
Ponds", Proceedings, International Conference of Geomembranes,
Denver, CO, Jun 1984, pp. 311-316.
Gosse, M.M. and Mclnnes, R.G., "Construction Techniques for Double
Lined .Systems", Draft Final Report to U.S. EPA. Office of Solid
Waste, Washington, DC, GCA Corporation, Bedford, MA, Dec 1984.
Griffin, R.A., Hughes. R.E., Follmer, L.R., Stohr, C.R., Morse. W.J.,
Johnson, R.M., Bartz, J.K., Sttele, J.D., Cartwright, K., Killey,
M.M., and OuMontelle, P.B., "Migration of Industria, Chemical and
Soil-Waste Interactions at Wilsonville, Illinois", Proceedings of
the Tenth Annual Research Symposium on Land Disposal of Hazardou_s
Waste, EPA 600-/9-84-007, U.S. EPA Municipal, Cincinnati. OH,
Environmental Research Laboratory, 1984.
Gunkel , R.C., "Membrane Liner Systems for Hazardous Waste Landfills",
hi Land Disposal of Hazardous Waste, Proceedings of the Seventh
Annual Symposium. EPA-600/9-81-002b, 1981, pp. 131-139.
Harrop-Wi11iams, K., "Clay Liner Permeability: Evaluation and
Variation", Journal of Geotechnical, Engineering, ASCE, Vol. Ill,
NO. 10, Oct 1985, pp. 1211-1225.
Haxo, H.E., Jr. and Nelson, N.A., "Factors in the Durability of
Polymeric Membrane Liners", Proceedings, International Conference
on Geomembranes. Denver, CO, Jun 1984, pp. 287-292.
11
-------
Herzcg, B.L. and Morse, W.J., "A Comparison of Laboratory and Field
Determined Values of Hydraulic Conductivity at a Waste Disposal
Site", Proceed ings, Se_ve_n lh Anruja_i_ Madison Wiste_ _Cor' ^^I'^ce ,
University of Wisconsin-Extension, Madison, WI, 1'JJl, op. -J-S2.
>tt. P.r., i.'recic, R.. Ojeshina, A., "An A$ses-::.eri!: of HLVC Liter
Duianility: A Report on Selected Installations", Pro coed i r.-js,
Internationa1 Conference on Geomefbranes, Denver, CO, Jun 1984,
pp. 317 320.
Jordan, E. C., "Performance Standard for Evaluating Leak Detec Mem,"
Draft Final Report to the EPA ~GCA Corporation, Bedford, MA and
E.G. Jordan Co., Portland, ME, ^ec 1984.
Kleppe, J.H., and Olson, R.E., "Desiccation Cracking of Soil
Barriers", ASTM 5TP 874 Hydraulic Barriers for Soil and Rock,
1985, pp. 263-275.
Kmet, P., Quinn, K.J., and Siavik, C., "Analysis of Design Parameters
Affecting the Collection Efficiency of Clay Lined Landfills",
Presented at the Fourth Annual' Madison Conference of
Applied Research and Practice on Municipal and Industrial Waste,
Sep 28-30, 1981.
Knipschid, F.W., "Selected Aspects of Dimensioning Geomembranes for
Ground-Water Protection Applications", Proceedings, International
Conference on Geomembranes, Vol. 11, Denver, CO, Jun 1984,
pp. 439 - 444.
Knipschid, F.W., Taprogge, I.R., and Schneider, I.H., "Quali ty
Assurance in Production and Installation of Large Area Sealing
Sections of High Density Polyethylene", Schlegal Engineering,
Bredowstrasse 33d-2000 -amburg 74, Germany.
Mitchell, J.K., "The Fabric of Natural Clays and its Relation to
Engineering Properties", Proceedings, Highway Research Board, Vol.
35, 1956, pp. 693-713.
Mitchell, J.K., "Fundamentals of So i 1 Behavior", John Wiley and Sons,
1976, New York, NY, 422 pp.
Pertusa, M., "Materials to Line or to Cap Disposal Pits for Low-Level
Radioactive Wastes", Geotechnical Engineering Report GR80-7, Dept.
of Civil Engineering, University of Texas, Austin, TX, 1980,
62 pp.
Reades, D.W. and Thompson, C.D., "Quality Control Testing and
Monitoring of Performance of Clay Till Liner, Stage 1, Keele
Valley Landfi 11", Maple, Ontario. Reprint of paper submitted to
CEO-CGS Seminar on Design and Construction of Municipal and
Industrial Waste Disposal Facilities, Ontario, Canada, 1984.
12
-------
Kogo«ski, A.S., "Effectiveness of a Compacted Clay Liner in Preventing
Ground Water Contamination",
9IL Aqui fer Res ^or^
~
ul, Cli, Mjy~ 1W5, pp. 417-429.
Po.r;.. -ki. 'VS., Wairirich, B.f.. and Si'i-;cn-, O.L., "Pc-i .- e il.> i ! : t j
Assoss.r-ant in a Contacted Clay Liner", Proceed ing: of the___Ej_ght_h
Annual Madison Waste Conference, Department cf Engineering and
Appfied Sciences, University of Wisconsin-Extension, Madison, WI ,
Sep 1905, pp. 31C-336.
Rogowski, A.S. and Richie, E.B., "Relationship of Laboratory and field
Determined Hydraulic Conductivity in Compacted Clay Soils",
Proceedings of the Mid-Atlantic Industrial Waste Conference ,
University Park, PA, 1984, pp. 520-533.
Schevon, G.R. and Damas, G. "Using Double Liner In Landfill Design
and Operation". GRCDA Convention, Orlando, FL, Aug. 1983 .
S c htn i d t , R . K . , "Specification and Construction Methods for Flexible
Membrane Liners in Hazardous Waste Containment", Technical Report
No. 102, Gundle Lining Systems, Houston, TX.
Schultz, D.W., .Duff, B.H., and Peters, W.R., "Performance of an
Electrical Resistivity Technique of Detecting and Locating
Geomembrane Failures", Prqceedjngs, Inter-nt ional Conference on
Gepinembranes, Vol. 11, Denver, CO, Jun 198'i, pp. 445 - 450.
Schultz, D.W. and Hiklas, M.P., Jr., "Procedure for Installing Liner
Systems." Proceedings . Eighth Annual Symposium Land Disposal of
Hazardous Waste, EPA-600/9-82-002, 1982, pp." 224-238.
U.S. Army, Corps of Engineers, "Construction Control for Earth and
Rock-Fill Dams". EH 1110-2-1922, Washington, DC, Jan 1977.
U.S. Army, Corps of Engineers, "Earth-fill and Rock-fill
Construction", Construction Control for Earth and Rock-Fill Dams ,
U.S. Army Engineer Manual EH1110-2-1911 , 1977.
U.S. Department of the Interior, Bureau of Reclamation, "Earth
Manual " . 2nd. Ed. S/N 2403-00079, Superintendent of Documents,
U.S. Government Printing Office, Washington, DC. 1974.
USEPA. "Cost Model", Appendix E of Liner Location Risk and Cost
Analysis Model , Draft Report. United States Environmental
Protection Agency, Washington, DC, 1985, pp. E-l to E-54.
USEPA, " Design, Construction, and _Eva luation of Clay Liners for
Hazardous Waste Facilities". Public Connie nt Draft, OSW,
Washington, DC, EPA/530-SW-86-007, 1986.
13
-------
USEPA, "Proceedings of the Sixth through Eleventh Annu.il Research
S.ymposla on Land Disposal, of...Hazardqii.s__Was_te". 1900 through 1985
available from NUS as PB-60175086, >B-81173074, PB-31173PS2, PB-
82173022, PB-8-1113/77, PB-841777999, and FB-?5n5376.
\. "_?C.v\_G,jidafire_ Occui^n^- _Land
Final Cover ', ~19d27~Draft", lT9~pp~
9es: -,n , L : "?r S . 3'.. .1:. acj
14
-------
------- |