Requirement for HW Landfill De

8/8/2019 Requirement for HW Landfill De

http://slidepdf.com/reader/full/requirement-for-hw-landfill-de 1/132



Technology Transfer EPA162514-89lO22

Seminar Publication

Requirements for Hazardous

Waste Landfill Design,Construction, and Closure

August 1989

Center for Environmental Research InformationOffice of Research and DevelopmentU.S. Environmental Protectin Agency

Cincinnati, OH 45268



NOTICE

The information in th is document has been funded wholly or in part by the United States

Environmental Protection Agency under Contract 68-C8-0011 to Eastern Research

Group, Inc. It has been subject to the Agency's peer and admini strative review, and it has

been approved for publication as an EPA document. Mention of trade names or

commercial products does not constitute endorsement o r recommendation for use.



CONTENTS

Page

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi

Regulations for Hazardous Waste Landfills . . . . . . . . . . . . . . . . . . . . . . 1

Leak Detection Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

Construction Quality Assurance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9Sum mar y of Minimum Technology Requi rements . . . . . . . . . . . . . . . . . . . . . . 10

1. Overview of Minimum Technology Guidance and

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Double Liners and Leachate Collection anRemoval Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Closure and Fina l Cover . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

. . . . . . . . . . . . 10 eferences . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2. Liner Design: Clay Liners . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Attack by Waste Leachate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

3. Flexible Membrane Liners . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

. . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

Gas Collector and Removal Systems . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . 70

. . . . . . . . . . . . . . . . . . 75. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .e Caps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75

. . . . . . . . . 75

. . . . . . . . . . . . . . . . . . . . . . 76

. . . . . . . . . . . . . . . . . . . . . . . 78

. . . . . . . . . . . . . . . . . . . . . . . .

.. .111



6 . Construction. Qual ity Assurance. and Control:Construction of Clay Liners . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89

CompactionVariables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89The Construction Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94

Construction Quality Assurance (CQA) Testi ng . . . . . . . . . . . . . . . . . . . . . . . . 95Test Fills . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . .. . 96

7. Construction of Flexible Membrane Liners . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99

Responsibility and Aut rit y . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99CQA Personnel Qualifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100Inspection Activities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100Sampling Strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101

Documentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103

8. Liner Compatibility with Wastes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109

Exposure Chamber . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109Representative Leachate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110

Compatibility Testing of Components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110Blanket Approvals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 Interpreting Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111

9 Long-Term Con ns: Problem Areas and Unknowns . . . . . . . . . . . . . . 113Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113Flexible Membrane Liners . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113

Clay Liners . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118Leachate Collection and Removal S ystem s . . . . . . . . . . . . . . . . . . . . . . . . . . . 118Cap/Closure Systems . . . . . . . . . . .. . . . . . . . . . . . .. . . . . . . . . . . .. . . . . . . . . . . 119

Action Leakage Rate (ALR) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121

Rapid a nd Large Leakage (RLL)Response Action Plans (RAPS) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124

Preparing and Submitting the RAP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125

List of Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127

. . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .0 Leak Response Action Plans 121Background . . . . . . . . . . .. . . . . . . . . . . . .. . . . . . . . . . . .. . . . . . . . . . . . .. . . . . . 121

122. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

iv



ACKNOWLEDGEMENTS

This sem in ar publication is based wholly on presentations made a t the U.S.Environmental Protect ion Agency (EPA) Technology Tra nsf er sem in ar s on

Requiremen ts for Hazardous Was te Land f i l l Des ign , Cons truc tion , and Closure . These

seminars were held from June 20 to September 16, 1988 in S an Francisco, California;

Seattle, Washington ; Dallas, Texas; Chicago, Illinois; Denver, Colorado; Kans as City,

Missouri; Philadelphia, Pennsylvania; Atlanta, Georgia; New York, New York; and

Boston, Massachusetts. The present ers were:

Sarah A. Hokanson, the Earth Technology Corporation, Alexandria, Virginia

(Chapters 1 and 10)

Dr. David Daniel, Universi ty of Texas, Austin, Texas (Chapters 2 and 6)

Dr. Gregory N . Richardson, Soil & Materials Engineers, Inc., Raleigh, NorthCarolina (Chapters 3,5, nd 7).

D r. Robert M.Koerner, Drexel University, Geosynthetic Research Institute,

Philadelphia, Pennsylvania (Chapters 4 and 9).

Robert Landreth, U . S . Environmental Protection Agency, Risk Reduction

Engine ering Laboratory, Cincinnati, Ohio (Chapter 8)

Susan Edwards, Linda Saunders, and Heidi Schultz of Eastern Research Group, Inc.,

Arlington, Massachusetts, prepared the text of this document based on the speakers'transcr ipts and slides. Orville Macomber (EPA Cen ter for Env iro nme nta l Res earc hInformation, Cincinnati, Ohio) provided sub stan tive guidance and review.

V



PREFACE

The U.S. Environmental Protection Agency's (EPA's) minimum technological require-ments for hazardous waste landfill design were set forth by Congress in the 1984 Hazard-

ous and Solid Waste Amendments (HSWA). HSWA covered requiremen ts for landfill lin-er s and leachate collection and removal systems, as well as leak detection systems for

landfills, surface impoundments, and waste piles. In response to HSWA and other Con-

gressional mandates, EPA has issued proposed regulations and guidance on the design of

these systems, and on construction quality assurance, final cover, and response actionplans fo r responding to landfill leaks.

This seminar publication outlines in detail the provisions of the minimum technology

guidance and proposed regulations, and offers practical and detailed information on the

construction of hazardous waste facilities that comply with these requir ements . ChapterOne presents a broad overview of the minimum technology guidance and regulations.

Chapter Two describes the use of clay liners in hazardous waste landfills, including theselection and testing of mate rial s for the clay component of double liner sys tems. Chap-te r Three discusses materia l and design considerations for flexible membrane liners, andthe impact of the proposed regulations on these considerations. Chapte r Four prese nts a n

overview of the three parts of a liquid management system, including the leachate collec-

tion and removal system; the secondary leak detection, collection, and removal system;

and th e surface water collection system. Chapte r Five describes the element s of a closuresystem for a completed landfill, including flexible membrane caps, surface water collec-

tion and removal systems, gas control layers, biotic barriers, and vegetative top covers.

Chapte rs Six and Seven discuss the construction, quality assur ance, and control criteriafor clay liners and flexible membrane liners, respectively. Chapter Eight discusses thechemical compatibility of geosynthetic and nat ura l liner materi als with waste leachates.

Chapter Nine presents an overview of long-term considerations regarding hazardouswaste landfills, surface impoundments, and waste piles, including flexible membrane

and clay liner durability, potential problems in liquid management systems, and aesthe t-

ic concerns. Chapter Ten reviews proposed requirements for response action plans forleaks in hazardous waste landfills.

This publication is not a design manual nor does it include all of the latest knowledge

concerning hazardous waste landfill design and construction; additional sources should

,be consulted for more detailed information. Some of these useful sources can be located in

the reference sections at the end of several chapters. In addition, State and local author i-ties should be contacted for regulations and good management practices applicable to lo-

cal areas.

v i



1. OVERVIEW OF MINIMUM TECHNOLOGY GUIDANCE ANDREGULATIONS FOR HAZARDOUS WASTE LANDFILLS

This chapter presents a summary of existing andproposed regulations and guidance on the design of

double liners and leachate collection and removalsystems, leak detection systems, final cover, andconstruction quality assurance. An overview of

proposed regulations concerning leak response

action plans is given in Chapter Ten. More technicaldiscussion of these a nd other components of landfill

design and construction are given in Chapters Two

through Y'me.

Background

EPA's minimum technological requirements forhazardous waste landfill design and construction

were introduced by Congress in the 1984 Hazardous

and Solid Waste Amendments (HSWA). In HSWASection 3004(0)(1 (A), Congress required all new

landfills and surface impoundments to ha ve doubleliners and leachate collection and removal systems

(LCRS). n Section 3004(0)(4),Congress also requiredleak detection systems at all new land disposal units,

including landfills, surface impoundments, andwaste piles. In response to other Congressional

mandates, EPA ha s issued proposed regulat ions orguidance on the design of these systems. In addition,EPA has issued guidance on construction qualityassurance programs and final cover. While notspecified in HSWA, the guidance and regulations in

the addi t ional areas were issued bv the U . S .

to maximize leachate collection and removal throughuse of the lining system a nd LCRS.

To date, EPA has issued regulations and guidanceprimarily focusing on double liners and leachate

collection and removal systems. Four F e d e r a lRegister notices and guidance documents have beenpublished by EPA in the last 4 ears in this ar ea (seeTable 1-1). EPA h as issued proposed regulations

and/or guidance in the additional areas listed inTable 1-2. The draft guidance on the final coverissued in July 1982, which was never widely

distribu ted, i s being revised for reissuance by the end

of 1989. EPA also plans to issue f inal regulations fordouble l iners and f o r leak detect ion sys tems,

including cons truc t ion qua l i ty as sura nce andresponse action plans.

Table 1-1. Guidance and Regulations Issued to Date

(Double Liners and LCRS)

0 Codification Rule (July 15, 1985)

0 Draf t Minimum Technology Guidance (May 24, 1985)

0 Proposed Rule (March 28, 1986)

0 Notice of Availabil ity of Information and Request for Comments

(April 17, 1987)

Environmental Protection Agency (E PA) to ensur eprotection of human heal th and the environment. Table 1-2. Guidance and Regulations Issued to Date

(Additional Areas)

For these new hazardous waste landfills and surface

impoundments, EPA and Congress have set forthperformance objectives of preventing hazardous

constituent migration out of a unit through t he endof post-closure care (or approximately 30 to 50 years).The approach EPA has developed to meet those

performance objectives is cal led the LiquidsManagement Strategy. The goal of the st rategy is tom i n im i z e l e a c ha te ge ne r a t i on t h r oug h bo t h

operational practices and the final cover design, and

Leak Detection Systems

0 Proposed Rule (May 29, 1987)

Construction Quality AssuranceProposed Rule (May 29. 1987)

0 Technical Guidance Document (October 1986)

Response Action Plan

0 Proposed Rule (May 29. 1987)

Cover Design

0 Draft Guidance (July 1982)

I



Double Liners and Leachate Collectionand Removal Systems

Figure 1-1 is a simplified schematic diagram of ahazardous waste landfill, showing the geometry an dplacement of double liners and LCRSs in a landfill.

In a double-lined landfill, there are two liners andtwo LCRSs. The prima ry LCRS is located above the

top liner, and the secondary LCRS is located betweenthe two liners. In this diagram, the top liner is a

flexible membrane liner (FML) and the bottom lineris a composite liner system consisting of a FML

overly ing compacted low perm eabili ty soil (o r

compacted clay).

Existing (D raft ) Guid ance for DoubleLiners

The EPA draf t gu idance i s sued in Ju ly 1985

discusses three types of liners: flexible membrane

liners (FMLs); compacted clay liners; and compositeliner systems (a F M L overlying a compacted low

permeability soil layer). Material specifications inthe guidance for FMLs and compacted clay line rs ar e

briefly reviewed below, along with existing and

proposed regulat ions regarding a l l three l iner

systems.

The m inimum thickness specification for a FML topliner covered with a layer of soil is 30 mils; for a FML

without a soil cover layer, the specification is 45mils. A FML in a composite bottom liner system

must be at least 30 mils thick. Even though theseFML thicknesses meet E PA specifications, 30 mils is

not a suitable thickness for all F M L materials. Infact, most FML materials installed at landfills are in

the range of 60 to 100 mils in thickness. Other key

factors affecting selection of FML materi als includechemical compatibility with waste leachate, agingand durability characteristics, stress and strain

characteristics, ease of installat ion, and watervaporlchemical permeat ion. These factors are

discussed in grea ter detail in Cha pter Three.

For compacted, low permeability soil liners, the EPA

draft guidance recommends natural soil materials,such as clays and silts. However, soils amended or

blended with different additives (e.g., lime, cement,

bentonite clays, borrow clays) also may meet thec u r r e n t s e l e c t i on c r i t e r i a o f l ow hyd r a u l i c

conduct ivi ty, o r permeabi l i ty , and suf f i c i en t

t h i c kne s s t o p r e ve n t ha z a r dous c ons t i t ue n tmigration out of the landfill unit. Therefore, EPA

does not currently exclude compacted soil liners t ha tcontain these amendments , Addit ional factorsaffecting the design and construction of compacted

clay liners include plasticity index, Atterburg limits,grain sizes, clay mineralogy, and attenuationproperties. These factors are discussed further in

Chapte r Two.

Existing and Proposed FederalReg ulations for Double Liners

Figure 1-2 shows cross sections of thre e double linerdesigns that have been used to meet existing or

proposed regulations. The double liner design on theleft side of the figure meets the existing minimum

technological requirements (MTR)as codified in Ju ly

1985. The center and right-hand designs meet theMTR as proposed by EPA in March 1986. The

existing regulations for MTR call for a double linersys tem cons is t ing of a FML top l iner and acompacted clay bottom liner that is 3 feet thick an d

has a maximum satu rated hydraulic conductivity of

no more than 1 x 10-7 cent imet ers per second

(c ds ec ). The 1986 proposed rule on double l inersgives two design options fo r MTR landfills: onesimi lar to the exi sting MTR design (differing only in

that the compacted clay liner must be sufficientlythick to prevent hazardous constituent migration);and one calling for a FML top liner and composite

bottom liner.

EPA i s cur ren t ly l ean ing toward requi r ing acomposite bottom liner in the final rule to bepublished in the su mmer of 1989.The Agency also isconsidering allowing use of a composite liner as a n

optional top liner system, instead of a FML. The finalrule, however, probably will not have minimum

thickness or maximum hydraul ic conduct ivi ty

s tand ards associated wi th the compacted c lay

component of such a composite top liner.

EPA’s rationale for favoring the composite bottomliner option in the final double liner rule would bebased on the relative permeability of the two liner

systems. Figures 1-3 through 1-5 show the results ofnumerical simulations performed by EPA (April1987) th at compare the performance of a composite

bottom liner to tha t of a compacted soil bottom linerunder various top liner leakage scenarios. In thesescenarios, liquids pass through defects in the top

FML and enter the secondary LCRS above the

bottom liners. As illustrated in these numerical

results, the hydraulic conductivities of these bottomliner systems greatly affect the amounts of liquids

detected, collected, and removed by the secondary

LCRS.

Figure 1-3 compares th e compacted soi l and

composi t e bo t tom l i ne r sys tems in t e rm s of

theoretical leak detection sensitivity, o r the minimalleak rate that can be detected, collected, andeffectively removed in the secondary LCRS. Thetheoretical leak detection sensitivity is less than 1

gallon per ac re per day (gallacrelday) for a composite

liner having an intact FML component. This leakdetection sensitivity value reflects water vaportransmission rates for FMLs with no defects. In

contrast, with well-constructed clay bottom liners

2



Double Liners and Leachate Collection System

Components

Protective Soil or

Cover (Optional) r Top Liner (FML)

Drainage

Material 0 0

Drainaqe BottomComposite Liner

Upper Component

Leachate Detection, ( F W

Collection, and Lower Component

Removal System (LDCRS (compacted soil)

Leachate Col lectiori

and Removal System

Native Soil Foundation

Source: EMCON, 1988

Figure 1-1. Schematic of a double liner and leachate collection system for a landfill.

(10-7 cmlsec permeability ), liquid s e nte ri ng the

secondary LCRS may go undetected and migra te into

the bottom liner until the leak rates approach 100gallacrelday. With a s l ight ly more permeablecompacted clay bottom l iner with 10-6 cmlsec

permeability, the secondary LCRS may not detect,collect, or remove the liquid flowing from a leak inthe top liner until leak rat es a re very serious (on the

order of 1,000 gallacrelday).

Figure 1-4 compares theoretical leachate collection

efficiencies f o r landfil ls having compacted soilbottom liners with those having composite bottom

liners. Leachate collection efficiency is th e amount ofliquid collected and removed in the secondary

leachate collection system divided b y the total

amount enter ing into the secondary LCRS through abreach in the top liner. For low leakage rates, the

leachate collection efficiency of a landfill with a

composite bottom liner system, even a compositesystem with tears o r small defects in the FML, isvery high (above 95 percent for leak rates in the

range of 1 to 10 gallacrelday). In comparison,

landfills with compacted clay bottom liners have 0

percent leachate collection efficiency for low leak

rates, and only 50 percent efficiency for leak rate s ofapproximately 100 gallacrelday. These re sults

demonstrate that leachate collection eficiency of the

secondary LCRS improves significantly simply by

\

Leachate CollectionSystem Sump

(Monitoring Compliance

Point)

installing a FML over the compacted clay bottom

liner.

Figure 1-5 shows the total quanti ty of l iquids

entering the two bottom liner systems over a 10-year

time span with a constant top liner leak rate of 50gallacrelday. A composite bottom line r with a n intac t

F M L accumulates around 70 gallacre, primarily

through water vapor transmission. Even with a 10-foot tear, which would constitute a worst-case

leakage scenario, a composite line r system will allow47,000 to 50,000 gallacre to enter that bottom liner

over a 10-year time span. Compacted soil li ners

meeting the 10-7 c d s e c permeability standard wil lallow significant quantities of liquids into the bottomliner, and potentially ou t of the un it over time, on the

order of hundreds of thousands of gallons per acre.

The numerical results indicate superior performance

of composite liner systems over compacted clay line rsin preventing hazardous constituent migration out ofthe unit and maximizing leachate collection and

removal. Consequently, many owners of new unitssubject to the double liner re quir emen t of HSWA are

proposing and installing composite bottom liners or

double composite liner systems, even though theyare not required currently. A survey conducted inFebruary of 1987 and revised in November of that

year ha s indicated that over 97 percent of these MTR

3



landfills and surface impoundments have one ofthese two double liner designs.

Existing (D raf t) Guidance for LeachateCollection and Removal Systems

Double-lined landfills have both primary andsecondary LCRS. The design of the secondary LCRS

in the landfill receives particular attention in EPA'sproposed leak detection requirements. Described

below are the existing guidance and proposed

regulations applicable to both LCRSs in double-linedlandfills.

The components of a LCRS include the drainage

layer, f il ters, cushions, sumps, and pipes andappurtenances. Of these components, the drainagelayer receives the most atten tion in the guidance andregulations. The drain age layer can consist of either

granular or synthetic material. If granular, it mustbe either clean sand or gravel with very little fines

content in order to facilitate t he rapid collection and

removal of the liquids tha t accumulate above the topliner and between the two liners. This minimizeshydraulic head on both liner systems.

Interim

Statutory Design

Waste

Leachate Collection

System Between

Liners

3 0

-

-

--

3 Feet-

- t-

-

-- -1 -

Native Soil

Design 1

Waste

Leachate Collection

System Between

Liners

0

Native Soil

According to the draft guidance, the main selectioncriteria for granular drainage materials are high

hydraulic conductivity and low capillary tension, orsuction forces. Figure 1-6 shows a range of hydraulicconductivities for natural granular materials. Fortypical drainage layer materials, permeabilitiesrange between 10-3 c d s e c and 1 cmfsec. A silty sanddrainage layer with significant fines content will

have a lower permeability (Le., 10-3 cm/sec) and

significant capillary tension. At the upper end of thescale, drainage layers consisting of clean gravel canachieve a permeabi lity on the order of 1 c d s e c to 100

cd se c . In this upper range of permeability, capillarytension is negligible. Therefore clean sands andgravels ar e preferred over silty sands.

T ab le 1 - 3 s h ow s t h e c o r r e l a t i o n b e t w e e npermeability and capillary rise (the elevation heightof liquids retained by granular particles within the

drainage layer by surface tension under unsaturated

conditions). At 10-3 cmlsec, there is significantcapillary rise (approximately 1 meter) while at the

upper end of the permeability scale (1 cdsec ) , thecapillary rise is only on the order of an inch.Reduction in fines content, therefore, significantlyreduces capillary rise while increasing hydraulic

Proposed Designs

Design 2

Waste

- Leachate Collection

System Between

Liners

Top LinerDesigned, Constructed, and

Operated to Prevent

Migration-AL 8, PCCP

/ ottom Liner \ I

Composite -Upper Component Prevent (FML) -Lower Component Minimize (clay)

Compacted SoilSufficient Thickness

' o Prevent Migration

During AL & PCCP

I -

I -

Native Soil

Figure 1-2. Interim statutory and proposed double liner designs.

4



I r l860

0.001

Composite

Type of Bottom Liner

Source: 52 FR 12570, pril 17, 987

Figure 1-3. Comparison of leak detection sensitivities for 3-foot compacted soil and composite liners (one-

dimensional flow calculations).

conductivity. -Increasing hydraulic conductivity , inturn, results in rapid collection and removal of

liquids.

Synthetic drainage materials have only recentlybeen introduced to the waste management industry.

Unl ike granular mater ia ls , synthet ic dra inagematerial s come in various forms and thicknesses:

0 Nets (1 60-280 mils)

0 Needle-punched nonwoven geotextiles (80-200mils)

Mats (400-800 mils)

0 Corrugated, waffled, or alveolate plates (400-800

mils)

sec

Top Liner Leakage Rate (gal/acre/day)

Source: 52 FR 12572, pril 17, 1987

Figure 1-4. Comparison of leachate collection efficiencies

for compacted soil and composite bottom liners.

Construction mater ials also vary. The most common

synthetic materials are polypropylene, polyester, orpolyethylene. More detailed discussion of these

drainage materials is presented in Chapter Four.Because synthetic drainage layers ar e much thinner(less tha n 1 inch) than g ranul ar drainage layers ( 1

foot) and have sim ilar design liquids capacity, theiruse in a landfill results in increased space for waste

storage and disposal. This advantage translates intoincreased revenues for the ownerloperator of alandfill.

The main selection criteria for synthetic drainagematerials are high hydraulic transmissivities, or

inplane flow rates, and chemical compatibility withthe was te leachate . Discuss ion o f c he m i c a lcompatibility of synthetic liners an d drainage layersis given in Chapter Eight.

Hydraulic transmissivity refers to the value of thethickness times the hydraulic conductivity for thatdrainage layer. Over the lifetime of d facility, theactual hydraul ic t ransmiss ivi t ies of synthet ic

drai nage layers a re affected by two key factors: ( 1 )

overburden stress and (2) boundary conditions. The

first factor pertains to the increasing loads (i.e.,

5



result in clogging, or reduced transmissivity, of the

LCRS.

200.000

:-

150,000

&

m9)-E-I

EI=0

a,

0

Ea,

m

r 100,000

c

BYma,-I

P)

m.-”-

50,000

50

0

160.000

Ik =- 1 x 10-7 cm/secl

Type of Bottom Liner

Source: 52 FR 12574 , April 17, 1987

Figure 1-5. Cumulative 10-year leakage into the bottomliner for a leak of 50 gal/acre/day through the

side wall of the top liner.

Table 1-3. Capillary Rise as a Functionof the Hydraulic

Conductivityof Granular Materials

Hydraulic Conductivity ofDrainaqe Medium (k) Capillary Rise (h)

cmlsec in_ _ _ _ ~ ~ ~ ~

1 x 10-3 38.6

1 x 10-2 12.2

1 1.2

Source: EPA, May 1987

wastes, operating equipment, and final cover)

applied to the liner that an LCRS experiences overthe lifetime of the facility. The second factor pert ain s

to the stress-strain characteri stics of adjacent laye rs(i.e., FMLs, filters, cushions, compacted clay). Overtime and with increasing stress, adjacent lay ers willintrude, or extrude, into the drainage layer and

Proposed Regulations for LeachateCollection and Rem o va Systems

Proposed regul ations applicable to LCRSs in doublelined landfills (March 1986) differ in two principalways from existing standards for LCRS in single

lined landfills and waste piles. First, LCRSs must be

designed to operate through the end of the post-closure care period (30 to 50 years ), and not simplythrough the active life of the unit. Secondly, in adouble-lined landfill with primary and secondary

LCRSs, the primary LCRS need only cover thebottom of the unit ( i .e . , sidewall coverage is

optional). The secondary LCRS, however, must coverboth the bottom and the side walls.

As in the existin g standa rds for single-lined landfills

and waste piles, the proposed regulations alsorequire t hat LCRSs be chemically resistant to waste

and leachate, have sufficient str ength and thicknessto meet design criteria, and be able to functionwithout clogging.

Applicability of Double Liner and LCRSRequirements

According to HSWA Section 3004(0)(1),all new units(landfills and surface impoundments) and lateralexpansions or replacements of existing units forwhich permit applications were submitted after

November 8, 1984 (the dat e HSWA was enact ed) wilbe required to comply with these double-liner and

LCRS requi rements once they ar e finalized. If permitapplications for these units were submitted beforethis dat e, new units need not have double li ners and

LCRSs unless the applications were modified afterthe d ate HSWA was enacted. However, EPA can usethe omnibus provision of HSWA to require doubleline rs and LCRSs that meet th e liner guidance on a

case-by-case basis at new facilities, regardless ofwhen the permit applications were s ubmitted. Table1-4 identifies facilities that will be requ ired to

comply with the new regulations.

Leak Detection Systems

Described in thi s section are proposed leak detection

system requirements that apply to the secondaryLCRS between the two liners in a landfill. These

requirements focus on the drainage layer componenof the LCRS. Figure 1-7 illu strat es the location of a

leak detection system in a double-lined landfill thatmeets these requirements.

Proposed Design Criteria

The proposed minimu m design s t an da rds fog r an u la r d r a in ag e l ay e r ma te r i a l s r eq u i r e a

minimum thickness of 1 foot and a minimum

6



Table 1-4. Landfills and Surface Impoundments Subject to

Proposed Regulations

APPLICABLE UNITSPermit Applications Submitted

I I I

Before 1 1/8/84 After 11/8/84

New Facilities No New Facilities YesIf permit

modified after

1 1 .8.84: Yes

Interim Status Interim Status

Facilities No Facilities

If permit modified Existing units: No

after 1 1/8/84: Yes New units

(operational)

after 5/8/85): Yes

New units at

Permitted Facilities Permitted Facilities

New units at

previously Interim previously Interim

Status facilities No" Status facilities: Yes

'Proposing to require MTR for these units on site-by-site basis.

I I I

hydraulic conductivity of 1 cd se c. In order to meet

this proposed m ini mum hy drauli c conductivity

s tandard for granular dra inage mater ia ls , thesecondary LCRS, or leak detection system, must be

constructed of clean gravels.

For synthetic drai nage mater ials, EPA has proposeda minimum hydraulic transmissivity of 5 x 10-4

square meters per second (mVsec). Th e hy dra uli c

transmissivity of a drainage material refers to thethickness of the drainage layer multiplied by the

hydraul ic conduct ivi ty. The t ransmiss ivi ty of

Drainage

Potential

soil

Types

granular drainage layers (1 oot x 1 cm/sec) is withinan order of magnitude of the 5 x 10-4 m%ec standardfor sy nthe t i c dra inage l ayer s . The proposed

hydraul i c t ransmis s iv i ty va lue for synthe t i c

drainage materials was developed to ensur e that thedesign performance for a geonet, geocomposite, o rother syn thetic drainage layer is comparable to thatfor a 1-foot thick gr anu lar drainage layer.

The proposed standards for leak detection systemsalso specify a mi nimum bottom slope of 2 percent, asis recommended in the existing draft guidance for

LCRS, and require the installat ion of a leak

detection sump of appro pri ate size to allow dailymonitoring of liquid levels or inflow rates in the leak

detection system. Specifically, the sump should bedesigned to detect a top liner leakage rate in the

range of the action leakage rate (ALR) specified in

the proposed leak detection rule. Chapter Tendiscusses the proposed ALR in more detail.

Proposed Design Performanc e

RequirementsThe proposed leak detection rule also establishesdesign performance standards for the leak detection

system. Design performance standards mean thatthe facil i ty design must include materials andsystems that can meet the above-mentioned design

criteria. If the liners and LCRS materials meet thedes ign cr i ter ia , then the des ign performance

standards will be met. Compliance with design

performance standards can be demonstrated through

Coefficient of Permeability (Hydraulic Conductivity) n CM/SEC

(Log Scale)

2 10' 1 o lo- ' 10-2 10-3 10.4 10-5 1 0 6 11

Good Almost

ImperviousI Poor

Clean gravel Clean sands, andclean sand and

gravel mixtures

Very fine sands, organic

and inorganic silts, mixtures

of sand, silt and clay, glacial

till, stratified clay deposits,

etc.

-7

Adapted from Terzaghi and Peck (1967)

Figure 1-6. Hydraulic conductivitiesof granular materials.

7



1 ft Granular Drainage Layer

Compacted Soil

Protective Soil Cover

Leachate Collection and

Removal System (above top liner)

Top Liner (Composite)

Leachate Detection,

1 ft granular drainage layer(k 5 1 cm/sec)

L 3-fl min compacted soil

(k c 1 x cm/sec)

Collection, and Removal System

Bottom Liner (Composite)

LeQend

(synthetic fibers-woven, nonwoven or knit)

(plastics formed into an open, netlikeconfiguration (used here in a redundant manner))

*e Geotextile

---- eonet

- lexible Membrane Liner (FML)

Figure 1-7. Location of a leak detection system in a double-lined landfill that meets proposed requirements.

numerical calculations rather than through field

demonstrations.

The proposed leak detection rule outl ines two design

per formance s t andards : ( 1 ) a l eak de tec t ion

sensitivity of 1 gallacrelday and (2) a leak detection

time of 24 hours. The leak detection sens itivity refersto the minimum top l iner leak ra te that can

theoretically be detected, collected, and removed bythe leak detection system. The leak detection time is

the minimum time needed for l iquids passing

through the top liner to be detected, collected, andremoved in the neares t downgradient collection pipe.In the case of a composite top liner, the leak detectiontime refers to the period starting at the point when

liquids have passed through the compacted soilcomponent and ending when they a re collected in thecollection pipe.

E P A bases i t s 1 gal lacrelday leak detect ion

sensitivity on the results of calculations that showthat , theoretically, a leak detection system overlyinga composite bottom liner with an intact F M Lcomponent can detect, collect, and remove liquids

from a top liner leak rate less than 1 gallacrelday

(see Figure 1 - 3 ) . This pe r formance s t an dard ,therefore, can be met with designs that include acomposite bottom liner. Based on numerical studies,

one cannot meet the leak detection sensitivity with acompacted soil bottom liner, even one with ahydraulic conductivity of 10-7 c d se c. Therefore, the

emphas i s of th i s s t anda rd i s on se lec t i ng anappropriate bottom liner system.

Meeting the 24-hour leak detection time , however, isdependent on the design of the leak detection system.

A drainage layer meet ing the des ign cr i ter ia ,

t o g e t h e r w i t h a d e q u a t e d r a i n s p a c i n g , c a ntheoretically meet the 24-hour detection t ime

standard. The emphasis of the proposed standards,therefore, is on designing and selecting appropriatematerials for the secondary LCRS.

A s s ta ted previous ly, compliance wi th E P A ' sproposed design performance standards can bedemonstrated through one-dimensional, steady-stat e

flow calculations, instead of field te sts. For detectionsensitivity, the calculation of flow rates should

a



as drain spacing, drainage media,

tom slope, and top and bottom liners should a ll be

equirements

s a nd ope r a t o r s o f l a nd f i l l s , s u r f a c e

w a s t e p i l e s o n w h i c h6 months after the date the rule

leak detection systems.

egulations concerning the design of the final cover

he guidance for final covers. The recommended

1982 draft version, with the exception t ha t some of

he design values for components of the final cover

ave been upgraded. EPA plans to issue th e reviseduidance for final covers in 1989.

Draft Guidance and Existing RegulatoryRequirements

E P A i s s ue d r e gu l a t i ons a nd d r a f t gu i da nc e

was te fac i l i t i e s in Ju ly 1982. Bas ica l ly , the

regulations require that the final cover be no more

permeable than the liner system. In addition, thecover must be designed to function with minimummaintenance, and to accommodate settlement andsubsidence of the underlying waste. The regulati ons

o not specify any design criter ia for liner mate rial sto meet the performance sta ndar d for permeability.

he draft guidance issued i n July 1982 recommends

a three-layer cap design consisting of a vegetative

top cover, a middle drainage layer, and a composite

liner system composed of a FML over compacted low

ermeability soil. The final cover is to be placed over

each cell as it is completed.

Since the regulations do not specify designs ofmaterial s for the final cover, o r cap, design enginee rscan usually use thei r own judgmen t in designing the

final cover and selecting materi als. For example, if

t h e l i n i n g s y s t e m c o n t a i n s a h i g h d e n s i t ypolyethylene (HDPE) membrane, the final cover does

not necessarily need to have a HDPE membrane. Theamount of flexibility in selecting FML mate rial s for

the final cover varies from region to region, based on

how s t r i c t ly the s t a tu tory phrase “no more

permeable than“ is interpre ted. Nevertheless, from adesign perspective, the selection of FML materi als in

the final cover should emphasize the physical rather

than the chemical properties of the liner material,

since the main objective is to minimize precipitation

infiltration. Precipitation infiltration is affectedmainly by the number of holes o r tears in the liner,

not by the water vapor transmission r ates.

F o r t h e v e g e t a t i v e c o v e r , E P A ’ s g u i d a n c e

recommends a minimum thickne ss of 2 feet and final

upper slopes of between 3 and 5 percent, after tak ing

into account total settlement and subsidence of the

waste. The middle drainage layer should have am i n i m um t h i c kne s s o f 1 f oo t a nd m i n i m um

hydraulic conductivity of 10-3 c d s e c . EPA’s revised

draft guidance upgrades th at standard by a n order of

magnitude to 10-2c d s e c to reduce capillary rise andhydraulic head above the composite liner system. For

the composite liner syste m a t the bottom of the cap, itis critical th at both the FML and the compacted soil

components be below the average depth of frost

penetration. The FML should also have a minimumthickness of 2 0 mils, but 20 mils will not be asufficient thickness for all FML materials. The soil

component under the FML must have a minimum

thickness of 2 feet and a maximum sa tura tedhydraulic conductivity of 10-7cd se c. The final upper

slope of the composite liner system must be no lesst h a n 2 p e r c e n t a f t e r s e t t l e m e n t . T a b l e 1 - 5

summariz es specifications for each pa rt of the final

cover.

Construction Quality Assurance

The final component of th e re gulatorylguid ance

summary discusses construction of a hazardouswaste landfill. The following section summarizes

EPA’s cons t ruc t ion qua l i ty as surance (CQA)program, as it is presented in existing guidance

(October 1986) and proposed regulations (May 1987).Chapter Seven contains a more detai led discussion of

CQA implementation.

Guidance and Proposed Regulations

The proposed regula t ion s an d ex i s t ing CQAguidance require the ownerloperator to develop aCQA plan that will be implemented by contracted,third-party engineers. The ownerloperator also must

submit a CQA report contai ning the following:

Summ ary of a l l observat ions , dai ly inspec-

tiodphotolvideo logs.

0 Problem ident i f ica t ionlcorrect ive m e a s u r e s

rep0rt

0 Design engineer’s acceptance r eports ( for errors,

inconsistencies).

0 D ev i at i on s f r o m d e s i g n a n d m n t e r i a l

specifications (with justifying dociimentation).



Table 1-5. Cover Design

Vegetative Cover

0 Thickness 2 2 ft

0 Minimal erosion and maintenance (e g , fertilization, irrigation)

Vegetative root growth not to extend below 2 ft

0 Final top slope belween 3 and 5 % after settlement or

subsidence. Slopes greater than 5% not to exceed 2 0tondacre erosion (USDA Universal Soil Loss Equation)

cap without rills and gullies

Surface drainage system capable of conducting run-off across

Drainage Layer Design

Thickness s 1 ft

Saturated hydraulic conductivity s 10-3 cm/sec

Bottom slope s 2% (after settlementisubsidence)

Overlain by graded granular or synthetic filter to prevent

Allow lateral flow and discharge of liquids

clogging

Low Permeability Liner Design

FM L Component

0 Thickness s 20 mil

Final upper slope 2 2% (after settlement)

0 Located wholly below the average depth of frost

Soil Component

Thickness 2 2 ft

0 Saturated hydraulic conductivity 5 1 x

0 Installed in 6-in lifts

penetration in the area

cm/sec

0 Summary of CQA activities for each landfill

component.

Thi s repor t mus t be s igned by a r e g i s t e r e d

professional engineer or the equivalent, the CQA

officer, the design engi neer, and the ownerloperator

to ensure tha t all partie s are satisfied with th edesign and construction of the landfill. EPA w i l lreview selected CQA reports .

The CQA plan covers all components of landfill

construction, including foundations, liners, dikes,leachate collection and removal systems, and finalcover. According to the proposed rule (May 19871,

EPA also may require field permeability testing of

soils on a test fill constructed prior to construc tion ofthe landfill to verify tha t the final soil liner will meetthe permeability stan dar ds of 10-7 cmlsec. Th is

requirement, however, will not preclude the use of

laboratory permeabi l i ty tes ts and other tes ts

(correlated to the field permeability tests) to verify

that the soi l l iner wi l l , as ins ta l l ed , have apermeability of 10-7 c ds e c .

Summary of Minimum TechnologyRequirements

E P A ’ s m i n i m u m t e c h n o l o g y g u i d a n c e a n d

regulations for new hazardous waste land disposalfacilities emphasize the importance of proper designand construction in the performance of the facility.The current trend in the regulatory programs is to

develop standards and recommend designs based onthe current state-of-the-art technology. Innovations

in technology are , therefore, welcomed by EPA andare taken into account when developing theseregulations and guidance.

References

1.

2.

3.

4.

5.

6.

7.

8.

9.

EMCON Associates. 1988. Draft backgrounddocument on the final double liner and leachatecollection system rule. Prepared for Office of

Solid Waste , L.S. EPA. NCS Contract N o. 68-01-

7310, Work Assignment N o. 66.

U.S. EPA. 1987a. Liners and leak detection for

hazardous waste land disposal units: notice of

proposed ru lema king. Fed. Reg. Vol 52, No. 10320218-20311. May 29.

U.S. EPA. 1987b. Hazardous waste management

systems: minimum technology r equir ement s:

notice of availability of information and request

for comments. Fed. Reg. Vol. 52, No. 74, 12566-12575. April 17.

U.S. EPA. 1987c. Background document on

proposed liner and leak detection rule. E PN5 30-

SW-87-0 15.

U.S. EPA. 1986a. Technical guidance document:

construction quality assurance for hazardouswaste land disposal facilities. EPAf530-SW-86-

031.

U.S. EPA. 198613. Hazardous waste managementsystems: proposed codification rule. Fed. Reg.

Vol. 51, No. 60, 10706-10723. March 28.

U.S. EPA. 1985a. Hazardous waste managementsystems: proposed codification rule. Fed. Reg.

Vol. 50, N O . 135,28702-28755. Ju ly 15.

U.S. EPA. 1985b. Draft minimum technology

guidance on double liner systems for landfillsa n d su r f ac e i m p o u n d m e n t s - d e s i g n ,

construction, and operation. EPAl530-SW-84-014. May 24.

U.S. EPA. 1982. Handbook for remedial action a t

was te d i sposa l s i t e s . EPA-62516-82-006 .Cincinnati, O H : U.S. EPA.

10



2. LINER DESIGN: CLAY LINERS

IntroductionThis chapter discusses soil liners and their use inhazardous waste landfills. The chapter focusesprimarily on hydraulic conductivity testing, both inthe laboratory and in the field. It also coversmaterials used to construct soil liners, mechanismsof contaminant transport through soil liners, and the

e f fec t s o f chemica l s and was te l eacha tes oncompacted soil liners.

Materials

C la y

Clay is the most important component of soil linersbecause the clay fraction of the soil ensures low

h y d r a u l ic c o n d u c t i v i t y . I n t h e U n i t e d

States,however, ther e is some ambiguity in definingthe term "clay" because two soil classification

systems are widely used. One system, published bythe American Society of Testi ng and M ate ria ls

(ASTM), is used predominantly by civil engineers.

The other, the U.S. Department of Agriculture's(USDA's) soil classification system, is used prima rily

by soil scientists, agronomists, a nd soil physicists.

The distinction between various particle sizes differs

between ASTM and USDA soil classification systems

(see Table 2-11, In the ASTM system, for example

sand-sized particles ar e defined as those able to pass

a No. 4 ieve but not able to pass a No. 200 sieve,fixing a gra in size of between 0.075 millimeters (mm)an d 4.74mm. The USDA soil classification system

specifies a grai n size for sand between 0.050 mm and

2 mm.

Th e USDA classification system is based entirelyupon grain size and uses a thre e-p art diagram toclassify all soils (see Figure 2-1). The ASTM sys tem,

however, does not have a grain size criterion forclassifications of clay; clay is distinguished from silt

en t i re ly upon p las t i c i ty c r i t e r i a . The ASTM

classification system uses a plasticity diagra m and asloping line, called the "A" line (see Figure 2-21 to

distinguish between silt and clay. Soils whose data

Table 2-1. ASTM and USDA Soil Classification by Grain Size

A S T M USDA

Gravel4.74 2

(No. 4 Sieve)

Sand0 075 0.050

(No. 200 Sieve)Silt

None 0.002

(Plasticity Criterion)

points plot above the A line on this classificationchart are, by definition, clay soils with prefixes C inUnified Soil Classification System symbol. Soilswhose da ta point s plot below the A line ar e classified

as silts.

EPA requires that soil liners be built so that the

hydraulic conductivity is equal to or less than 1 x10-7 cm/sec. To meet this requirement , cer ta in

character istics of soil materials should be met. Fir st,

the soil should have at least 20 percent fines (fine silt

and clay sized particles). Some soils with less tha n 20percent fines will have hydraulic conductivitiesbelow 10-7 cd se c , but at such low fines content, therequired hydraulic conductivity value is much

harder to meet.

Second, plasticity index (PI) should be greater than10 percent. Soils with very high PI, greater than 30to 40 percent, are sticky and, therefore, difficult to

work with in the field. When high PI soils are dry,

they form hard clumps that are difficult to breakdown durin g compaction. On the Gulf Coast of Texas,

fo r example, clay soils are predominantly highly

plastic clays and require additional processingduring cons t ruct ion. F igure 2-3 represent s acollection of data from the University of Texaslaboratory in Austin showing hydraulic conductivity

as a function of plasticity index. Each data point

represent s a separa te so i l compac ted in the

11



conductivity fr om 10-4 to 10-7 cm/sec, a r a therdramatic reduction.

Percent S a n d

Figure 2-1. USDA soil classification.

laboratory with s tandard Proctor compact ionprocedures and at a water content about 0 to 2

percent wet of optimum. Hydraulic conductivitiesar e consistently below 10-7 c d s e c for soils with PISgreater tha n 10 percent.

Third, coarse fragments should be screened to nomore than about 10 percent gravel-size particles.

Soils with a greater percentage of coarse fragments

can contain zones of gravel th at have high hydraul icconductivities.

Finally, the mater ial should contain no soil particleso r chunks of rock larger than 1 to 2 inches indiameter. if rock diameter becomes a significantpercentage of the thickness of a layer of soil, rocksmay form a permeable "window" through a layer. Aslong as rock size is small compared to th e thickness

of the soil layer, the rock will be surrounded by theother materia ls in the soil.

Blended Soils

Due to a lack of naturally occurring soils a t a site, itis sometimes necessary to blend imported clay

minerals with onsite soils to achieve a sui tableblended material. The most common blend is acombination of onsite sandy materials and importedsodium bentonite.

Figure 2-4 shows the influence of sodium bentoniteon the hydraulic conductivity of the silt/sand soil.

The addition of only 4 or 5 percent sodium bentoniteto th i s par t ic u lar so i l d rops the hy drau l i c

Calcium bentonite, though more permeable thansodium bentonite, has also been used for soil blends

Approximately twice as much calcium benton ite

typically is needed, however, to achieve a hydraulicconductivity comparable to th at of sodium bentoniteOne problem with using sodium bentonite, howeveris its vulnerability to attac k by chemicals and waste

leachates, a problem that will be discussed later in

the chapter.

Onsite sandy soils also can be blended with otherc l a y soils available in the ar ea , but n atu ral clay soiis likely to form chunks that are difficult to breakdown into small pieces. Bentonites, obtained in dry

powdered forms, ar e much easier to blend with onsitesandy soils than are wet, sticky clods of c l a yMaterials other t han bentonite can be used, such asatapul gite, a clay mineral tha t is insensitive toattack by waste. Soils also can be amended withlime, cement, or other additives.

Clay Liners versus Composite Liners

Composite liner systems should outperform eitherflexible membrane liners (FMLs)or clay liners alone

Leachate lying on top of a clay liner wil l percolate

down through the liner at a rate controlled by thehydraulic conductivity of the liner, the head of theleachate on top of the liner, and t he liner's total a rea

With the addition of a FML placed directly on top othe clay and sealed up against its upper surfaceleachate moving down through a hole or defect in the

FML does not spread out between the FML and theclay liner (see Figure 2-5) . The composite linersystem allows much less leakage than a clay liner

acting alone, because the area of flow through theclay liner is much smaller.

The FML must be placed on top of the clay such t ha

liquid does not spread along the interface betweenthe FML and the clay and move downward through

the ent ire ar ea of the clay liner. A FML placed on abed of sand, geotextiles, or other highly permeablematerials, would allow liquid to move through thedefect in the FML, spread over the whole are a of the

clay liner, and percolate down as if the FML was nothere (see Figure 2-6). With clay liner soils tha

contain some rock, it is sometimes proposed that awoven geotextile be placed on top of the soil lineunder the FML to prevent the puncture of rocks

through the FML. A woven geotextile between theFML and the clay, however, creates a highlytransmissive zone between the FML and the clay

The surface of the soil liner instead should becompacted and the stones removed so that the FMLcan be placed directly on top of the clay.

12



60

50

40

-a,%

2 30

U

-

0

ma

c

-

20

1c

7

4

(10 16 20 30 40 50 60 70 80 90 100 110

Llquid Limit (LL)

Figure 2-2. ASTM plasticity determination for fine-grained soils.

10.7

10.8

Lower Bound

10-10 1I I 1

10 20 30 40 50 60

Plasticity Index ('70)

Figure 2-3. Hydraulic conductivity as a function of plasticity

index for soils in Austin Laboratory Tests.

Darcy's Law, Dispersion, and Diffusion

Figure 2-7 illustrates Darcy's law, the basic equation

used to describe the flow of fluids through porousmaterials. In Darcy's l a w , coefficient k, hydraulic

conductivity, is often called the coefficient of

permeability by civil engineers.

Darcyk law applied to a soil liner shows the rate of

flow of liquid q directly proportional to the hydraulic

conductivity of the soil and t he hydraulic g radien t, a

measure of the driving power of the fluid forcing

itself through the soil and the cross-sectional area

"A "of the liner (see Figure 2-7) .

If hydraulic conductivity is 10-7 c d s e c , the am ount

of leakage for a year, per acre, is 50,000 gallons. If

the conductivity is 10 t imes t hat value (1 x 1 0 - 6cd se c) , the leakage is 10 times greater, o r 500,000gallons. Table 2-2 summarizes quant ities of leakageper annu m for a l-acre liner with a n amount of liquid

ponded on top of it, assuming a hydraulic gradient of1.5.Cutting the hydraulic conductivity to 10-8c d s e c

reduces the quantity of leakage 10-fold to 5,000gallons per acre per year. These data demonstrate

how essential low hydraulic conductivity is tominimizing the quantity of liquid passing through

the soil liner.

Contaminants

The transport of contam inants through the soil lineroccurs by either of two mechanisms: advective

13



10.3

1 0 -4

10-5

10-6

10-7

10-8

10-9

lo-"

1 I I I I

0 4 8 12 16 20 24

Percent Bentonite

Figure 2-4. Influence of sodium bentonite on hydraulic

conductivity.

Clay Liner Composite Liner

Leachate Leachate

A = Area of Entire Area c Area of Entire

Liner Liner

Figure 2-5. Leachate infiltration in clay and composite liner

systems.

Leachateeachate

-o Don't

Figure 2-6. Proper placement of FMLs on clay liners.

I l Hnfluent

Liquid

SOl l LI:

u iquid

H

Lq = k - A

q = r a t e o f f l o w H = h e a d l o s sk = h y d rau l i c L = l e n g t h o f f l o w

conduct iv i ty A = t o t a l a r e a

9Leachate

Subsoil

transport, in which dissolved contaminants are q = kiA

carried by flowing water, and molecular diffusion ofthe waste through the soil liner. Darcy's law can beused to estimate r ate s of flow via advective transpor t

water. Seepage velocity is the hydraulic conductivity

times the hydraulic gradient, divided by the effectiveporosity of the soil. The effective porosity is defined

as the volume of the pore space that is effectively

i = H y d r a u l ic G r a d i e n t

H + D

D

-by calculating the seepage velocity of the flowing - -

(Assumes No Suction Below Soil Liner)

Figure 2-7. Application of Darcy's Law.

14



2-2. Effects of Leakage Quantity on HydraulicConductivity for a 1-Acre Liner

divided by the hydraulic conductivity times thehydraulic gradient (Figu re 2-91.

Hydraulic Conductivitv (cmisec) Annual Lcakaoe (oallons)

1 x 10-6 500,000

i 10-7 50,000

1 x 1 0 - 8 5,000 L

Vs kiTOT = - -

1.5

sample (Figure 2-81.

Figure 2-9. Time of Travel (TOT).

i = Hydraulic

H + T

T

Gradient

It is possible to confirm these calculations andmeasure some of the parameters needed t o makethem by performing labora tory permeab il i ty

exper iments . In these experim ents, clean soil is

placed into columns in the laboratory, and theleachate or some other waste liquid is loaded on topof each soil column, forcing the liquid through the

column over a period of time, while keeping theconcentration of influent liquid constant. The

concentration of one or more chemicals in theeffluent liquid is measured over time.

- -

(No Suction)

Flux

v, = seepage velocity

ki

n= -

Subsoil ne = effective porosity

igure 2-8. Advective transport.

f the liquid uniformly passes through all the pores

in the soil, then the effective and total porosities ar equal. However, if the flow takes place in only a

small percentage of the total pore space, for example,hrough fractures o r macropores, the effective

orosity will be much lower than the total porosity.the effective porosity i s one of the problems

f estimating seepage velocities.

If effective porosity and other parameters ar e known,

the time of travel (TOT) for a molecule of waste

transported by flowing water through the soil linercan be calculated. TOT equals the length of theparticular flow path times the effective porosity,

A plot called a breakthrough curve shows theeffluent liquid concentration c divided by the

influent liquid concentration co as a function of porevolumes of flow (see Figure 2-10). One pore volume offlow is equal to the volume of the void space in the

soil. The effective porosity of the soil is determinedby measuring a breakthrough curve.

It can be expected that as the leachate invades thesoil, none of the waste chemical will appear in the

effluent liquid at first, only remn ant soil and water.Then a t some point, the invading leachate will makeits way downstream through the soil column, and

come out in more or l e ss f u l l s t r e n g t h . A n

instantaneous breakthrough of the waste liquid

never occurs, however. The breakthrough is a l w a y sgradual because the invading leachate mixes withthe remnant soil water through a process called

mechanical dispersion.

Many of the waste constituents in the leachate are

attenuated or retarded by the soil. For example, lead

migra tes very slowly through soil, while chloride andbromide ions migrate very quickly. With noretardation or attenuation, breakthroughs wouldoccur at c/coo f 0 . 5 to 1 pore volume of flow and below

(see Figure 2-1 11. With effective and tota l porosities

equal, a much delayed breakthrough of chemicals

15



C

C

-

LeachateConcentration

of Solute = C

1

0.5

0

Concentration

of Solute = cEffluent

(forn = fie)

0 1 2

Pore Volumes of Flow

Figure 2-10. Effective porosities.

that have been absorbed or attenuated by the soil

could be expected.

The best way to determine effective porosity is toperform a test using a ''tracer" ion tha t will not be

absorbed significantly by the soil, such as chloride o rbromide. If the breakthrough occurs in one porevolume of flow, the effective and total porosity is

equal. If, instead, the breakthr ough occurs at half apore volume of flow, then the effective porosity i s halfthe total porosity.

Molecular Diffusion

Chemicals can pass through soil liners by moleculardiffusion, as well a s by advective transpor t. One can

study the molecular diffusion of chemicals in the soil

by compacting soil in the bottom of an impermeablebeaker a nd ponding waste liquid or leachate on top of

No Retardation Retardation

0 1 2 3


Figure 2-11. Effective porosity of soils with retardation andwithout retardation of waste ions.

the so i l. At the s t a r t of t he e xpe r i m e n t , t he

concentration c is equal to c, in the waste liquid. The

soil is clean. Even though no water flows into the soi

by advection, chemicals move into the soil by the

process of molecular diffusion. Eventually, the

concentration of the waste liquid and the soil will beone and the same (see Figure 2 - 1 2 ) .

0 CO

Concentration (c)

Figure 2-12. Molecular diffusion.

Calcula t ions show tha t a f t e r 10 to 3 0 ye a r s

molecular diffusion begins to transport the firsmolecules of waste 3 feet downwards through acompacted soil liner. Accordingly, even with aperfectly im permeab le l iner wit h 0 hydr a u l i cconductivity, in 1 to 3 decades contaminants wil

begin to pass through the soil liner due to molecula

diffusion.

The rate of diffusion is sensitive to a number o

parameters. For conservative ions tha t a re noattenuated, the transfer time is 1 to 3 decades. For

ions that are attenuated, transfer t ime is muchlonger. The mass rate of transport by molecular

diffusion, however, is s o slow that even thoughchemicals begin to show up in 1 to 3 decades, thetotal amount released per unit of area is small.

Flexible membrane liners permit the release oorganics and vapors via molecular diffusion by

16



ctly the s ame process. Transport times for

anic chemicals t hroug h F M L s typically range

a few hours to a few days.

conductivity of a soil liner is the key

ic conductivity testing in the laboratory are:

Representativeness of the sample.

Degree of water sa tura tion .

Stress conditions.

Confinement conditions.

Testing details.

of Sample: Case

tativeness of the soil sample being tested is

P A

a pad of75 feet long, and 1 foot thick.

ay liner was built in t hree lifts, or layers, each4 nches thick. The liner was built up on a

so that researchers could crawl underlect an d measur e the liquid coming out of the

1 foot of water ponded over the surface.

a number ofeach subunit measur ing 3 feet by 3 feet. A

of 250 different collection units underneath the

60 1-foot diameter ringsf the liner, so that he could measure

different infiltration rates on theof the 3-square-foot (ft2)

c ks w a s a s s i gne d a n a v e r a ge hyd r a u l i c

2-13 shows the results. T he zone

uctivity of 10-5 c d s e c probably resulted from

the way the liner was built. The sheepsfoot roller

used to compact the liner probably bounced on theramp causing lower compaction, which resulted in are la t ive ly h igh conduc t iv i ty a t the end . Theconductivity for the rest of the liner varies between10-6 and 10-8 c d s e c , a 100-fold variation of hydraulic

conductivity

10-5 cm/s

10-6 cm/s

[IIII] 10-7 cmis

10-8 cm/s

7s n

30 t

Figure 2-13. Hydraulic conductivity zones from Klingerstown,

PA Tests.

For a laboratory test on this soil, the test specimen

would need to measure about 3 inches in diameter

and 3 inches in height. Finding a 3-inch diamet ersample representative of this large mass of soilpresents a challenge, since small s amp les fromlarger quantitie s of material inevitably va ry in

hydraulic conductivity.

D r . Rogowski 's experiments resul ted in two

interesting sidelights. First, the average of all thehydraul ic conductivities was 2 to 3 x 10-7cd sec . Dye

was poured into the water inside some of the 1-footdiameter rings installed in the surface of the liner todetermine if the dye came out directly beneath thering or off to the side. In some cases it came outdirectly beneath t he ring and in some it wandered offto the side. It took only a few hours, however, for thedye to pass through the soil liner, even with anaverage conductivity only slightly greater than 1 x

1 0 - 7 c d s e c . A few preferential flow paths connected

17



to some of the rings allowed very rapid t ransit of the

dye through the soil liner.

The second inter esting sidelight was Dr . Rogowski’sconclusion that no relationship existed between insitu hydraulic conductivity and either molding watercontent of the soil or the dry density of the compacted

soil.

The soil used in the experiment w a s a low plasticitysandy material with a PI of about 11 percent. The

varia t ions in hydraul ic conduct ivi ty probably

reflected zones of materi al t ha t contained more sandin some places and more clay in o thers. Tests have

been performed on a couple of liners in the fieldwhere liquid flowing into the soil liners has beendyed and traced by cutting a cross section or trenchthrough the liner. Typically, a pattern such as thatshown in Figure 2-14 emerges, with the horizontal

dots indicating lift interfaces. The results seem to

indicate th at dyed liquid finds a defect in the top lift,moves down and spreads along a more permeablezone between lifts; finds another defect, moves

downward, spreads; finds another defect and so forth.

V

1 1 1Figure 2-14. Liquid flow between lift interfaces in a soil liner.

The problem arises in determining from where arepresentative sample should be taken. Even if 25samples were picked randomly in a grid pattern fromthat zone for 25 independent measures of hydraulic

conductivity, it would be unclear how to arrive at a

single representative measure. The flow through a 3-inch diameter specimen is much too small to mimicthe pat terns of fluid flow that occur in the field undersimilar conditions.

Houston, TX

A second case h i s tory tha t demons t ra tes thedifficulty of obtaining representative samples

involves a trial pad constructed in Houston in 1986.

A 1-foot thick clay line r was compacted over a gravelunderdrain with an ar ea roughly 50 feet by 50 feet.

The entire a re a of the liner was drained and the flowfrom an are a roughly 16 feet by 16 feet was carefullycollected and measured .

The liner was first built on top of the underdr ain, the

soil compacted with a padfoot roller, and water

ponded on top of the liner. Infiltrometers measured

the rate of inflow, and a lysimeter measured the rat eof outflow. The soil used in the experiment w a shighly plastic with a PI of 41 percent.

The l iner was compacted with two lifts, each 6 inches

thick. A 1-ft3 block of soil was carved from the liner ,

and cylindrical test specimens were trimmed fromupper and lower lifts and measured for hydraulicconductivity. A 3-inch diameter specimen also was

cut, and hydraulic conductivity parallel to the lift

interface was measured.

Table 2-3 summarizes the results of these varioustests. The actual in situ hydraulic conductivity, a

high 1 x 10-4 cmlsec, was verified both by t heinf i l t ra t ion measurements and the underdrainmeasurements.

Table 2-3. Hydraulic Conductivities from Houston Liner Tests

Actual k: 1 x cmis

Lab K’s:

Location Sampler K (cm/s)

Lower Lift 3-in Tube 4 x 10-9

1 x 10-9

1 x 1 0 - 7

Lower Lift Blocka

x 10-5

Upper Lift Block 1 x 1 0 - 8

Upper Lift 3-in Tube

Lift Interface 3-in Tube

The tests were replicated under controlled conditionsusing soil collected from the liner in thin-walled 3-inch d iamet e r s ample tubes . The l abora torymeasures of hydraulic conductivity were consistently

1 x 10-9 c d s e c , five orders of magnitude lower tha nthe field value of 1 x 10-4cd se c. The laboratory tests

yielded a hydraulic conductivity 100 ,000 tim es

different than that from the field test. Apparentlythe flow through the 3-inch specimens did not mimicflow on a larger scale through the ent ire soil liner.

The sample tr immed horizontally at the lift interfacewas actually obtained by taking a %inch diameter

sample from a sample collected with a 5-inchdiameter tube. The hydraulic conductivity with flowparallel to the lift interface w a s two orders of

magnitude higher.

Of all the values recorded from the lab tests, only the

one obtained from the upper lift of the block samplewas close to the field value of 1 x 10-4 cm/sec.

18



f the more permeable zones and, more or less bya lab measurement that agreed

measurement.

a laboratory

lso can be affected by the amount of gas present

l primarily is filled with ai r. Becauseer does not flow through air-filled voids,

only through water-filled voids, the dryness of ar permeabi lity.

conduct ivi ty tes ts by backp ressu reof the soil. This technique pressurizes the

of air, therebyng hydraulic conductivity.

, or confining pressure, acting on the soil.1 foot of soil overburden is roughly

1 pound per squa re inch (psi). If two

at depth, the soil near the

a more compactby the overburden pressure. Thus, soil

as a lower porosity with increasing dep th.

a series of experiments performed a few years ago,

of-clay were compacted in t he lab and then

f the clay was compacted and then trimmedtest specimen immediately, while the other was

a period of time before beinge tha t desiccated had tiny cracks as

result of the desiccation process, and was much

acted into a less porous condition.

w a s cracked f romwas obviously more permeable, at a very

ning pressure acting on the soil, the cracks t hat

had existed earlier closed up completely so that the

soil w a s no longer highly permeable .

1 0 - 7

10-8

(k Pa)

0 50 100

-Sample Containing --

--

----

No Desiccation

Cracks -

0 4 8 12 16

Effective Confining Pressure (psi)

Figure 2-15. Hydraulic conductivity as a function of confining

pressure.

One implication of these experiments for laboratory

hydraulic conductivity testing is that conductivityvalues can vary remarkably depending on theconfining stress. I t is essential that the confiningstress used in a laboratory test be of the same

magnitude as th e stress in the field.

Another impor tant implicat ion is t ha t highl ypermeable soil liners generally have defects, such ascracks, macropores, voids, and zones that have notbeen compacted properly. One opportunity toeliminate those defects is at the time of construction.Another opportunity arises after the landfill is in

operation and the weight of overlying solid waste o rof a cover over the whole system furth er compresses

the soil. This compression, however, occurs only onthe bottom liners, as there is not much overburdenstress on a final cover placed over a solid waste

disposal unit. Thi s is one reason it is more difficult todesign and implement a f inal cover with low

hydraulic conductivity than it is a bottom liner. Notonly is there lower stres s acting on a cover than on al iner, but the cover is also subjected to man yenvironmenta l forces which th e liner is not.

19



Double-ring and Flexible Wall Permeameters

A double-ring permeameter separates flow that

occurs through the central part of the soil samplefrom flow that occurs near the side wall . Thepermeameter i s designed such that a ring sticks intothe bottom of the soil sample, thereby detecting

sidewall leakage tha t might invalidate the results of

laboratory conductivity tests. Almost all of the rigidwall permeameters now being installed in the

University of Texas laboratories have double rings.Another kind of permeameter cell is a flexible-wall

permeameter in which the soil specimen is confinedby a thin, flexible membrane, usually made of latex.

The latex membrane conforms to any irregularitiesin the sample, an advant age when col lect ing

irregular ly shaped specimens from the field.

Termination Criteria

When conducting laboratory hydraulic conductivitytests, two criteria should be met before testing is

termina ted. F irst , the ra te of inflow should be within

10 percent of the ra te of outflow. Measuring both the

rat e of inflow and the rat e of outflow is necessa ry todetect problems such as a leak in the system o revaporation from one of the reservoirs. Second,a plotof hydraulic conductivity versus t ime or pore volume

of flow should essentially level off, indicating thathydraulic conductivity is steady.

ASTM has no standards at the present time fortesting low-hydraulicconductivity soil, but is in the

final stages of developing a standard for tests withflexible wall per meameters that should be available

within the next 2 years.

Field Hydraulic Conductivity TestingIn si tu, or field, hydraulic conductivity testing

operates on the assumption that by testing largermasses of soil in the field one can obtain more

realistic results. There ar e actually four kinds of in

situ hydraulic conductivity tests: borehole tests,porous probes, infiltrometer tests, and underdrain

tests. To conduct a borehole test one simply drills ahole in the soil, fills the hole with water, and

measures the ra te a t which water percolates into the

borehole.

The second type of tes t involves driving or pushing aporous probe into the soil and pouring water through

the probe into the soil. With this method, however,the advantage of testing directly in t he field is

somewhat offset by the limitations of testing such asmal l volume of soil.

A third method of testing involves a device called a n

infiltrometer. This device is embedded into thesurface of the soil liner such that the ra te of flow of a

liquid into the liner can be measured. lnfiltrometers

have the advantage of being able to permeate large

volumes of soil, which the f irst two devices cannot .

A fourth type of test utilizes a n underdrain, such asthe one at the Houston test site discussed ear lier .

U n d e r d r ai n s a r e t h e m o s t a c c u r a t e i n s i t upermeability testing device because they measure

exactly wha t comes out from the bottom of the liner.They are, however, slow to generate good data forlow permeability liners because they take a while to

accumulate measurable flow. Also, underdrainsmust be put in during construction, so there are

fewer in operation than there are other kinds oftesting devices. They are highly recommended for

new sites, however.

The two forms of infiltrometers popularly used are

open and sealed. Four variations are illustrated in

Figure 2-16. Open rings are less desirable becausewith a conductivity of 10-7 cmlsec, it is difficult toseparate a 0.002 inches per day drop in water level ofthe pond from evaporation a nd other losses.

Open, Single Ring Open, Double Ring

Sealed, Single Ring Sealed, Double Ring

Figure 2-16. Open and sealed single- and double-ringinf iltrometers.

With sealed rings, however, very low rates of flowcan be measured. Single-ring infiltrometers allow

lateral flow beneath the ring, complicating the

interpretation of test results. Single rings are also

susceptible to the effects of temperatu re variat ion; asthe water heats up, the whole system expands and asit cools down, the whole system contracts. Thissituati on could lead to erroneous measu rement s

when the rat e of flow is small .

The sealed double-ring infiltrometer has proven themost successful and is the one used currently. Theouter ring forces the infiltration from the inner ring

to be more or less one dimensional. Covering the

20



es it substantially from

a 12-foot by 12-foot outer

a 5-foot diameter inner ring. Tensiometers

at various depths to establish the

so tha t

ydraulic conductivity can be calculated.

\Sealed Inner Ring

\ ,Tensionmeters

Clay Liner Grout

Figure 2- 17. Details of a sealed double-ring infiltrometer.

ate of infiltration is measured by a small flexible

ag. As water infiltrates from the i nner ring into the

oil, the flexible bag is gradually compressed asater leaves it to enter the ring. To determine howuch flow has taken place, the flexible bag is

be refilled or replaced with a fresh bag.

ween the inner and outer rings. If the water leveln the outer ring changes, the hydrostatic pressure

mount. Thus, even though the water level in the

verall; this simple device compensates for water

evel changes and allows a range of measurements.

nstallation of th e Sealed Doub le-ringnfiltrometer

ed double ring infi ltrometer is best used on aest pad. The width of the te st pad is usually about 40

eet by 50 feet; the th ickness of the test pad usually 23 feet. The test pad is a l w a y s covered to prevent

on after construction has been completed.

foot by 12-foot ou te r ri ng is made of four

luminum panels that ar e assembled a t the site. Arefabricated design allows the panels to be boltedogether to prevent breaching. The outer box canhen be lifted u p and put into position embedded in

the liner. If the site is sloping, the elevation of the

four corners is measured at the site with a handheld

level or a transit, so th at the top of the infiltrometer

is more or less horizontal and the water level is evenwith the top of the infiltrometer.

A rented ditching machine is used to excavate atrench about 18 inches deep and 4 nches wide for the

outer ring. The ring is embedded into the trench a nd

the elevations ar e checked again.

The sealed inner ring typically is made of fiberglass

and measures 5 feet by 5 feet. I t slopes from left to

right and from side to side in a dome-shaped slope

such that it has a high point. As the ring fills with

water from the bottom up, gas is displaced out thetop. When the inner ring is completely full of water,

the ga s is purged out of the system.

The trench for the inner ring is not dug with the

ditching device because the vibration and churning

action might open up fractures in the soil and changethe measurements. Instead, the trench is dug in one

of two ways: by a handheld mason's hammer or by achain saw. A chain saw with a specially equipped

blade is the state-of-the-art in excavation of trenchesfor the inner ring.

While the excavation is being done, the working are a

is covered with plastic. Before the system is ready to

be filled with water, a pick or rake is used to scrape

the surface thoroughly to ensure that smeared soil

has not sealed off a high hydraulic conductivity zone.

After the trench has been excavated, it is filled with

a grout containing bentonite that has been mixedwith water to the consistency of paste. A grout mixer

rather than a concrete mixer is used to provide morecomplete mixing. The inner ring is then forced intothe grouted trench. The grout is packed up against

the ring to obtain the best possible seal. To pretestthe seal, the inner ring is filled with about 3 inches of

water. If ther e is a gross defect a t the seal, water will

spurt out of it. Next, the outer ring is placed in its

grout-filled trench .

The next step is to tie steel cords in the middle of the

four sections to prevent th e outer rin g from bowing

out from the pressure of the wat er. Tensiometers ar e

installed in nests of three a t three different depths tomonitor the depth of the wet ting process. To cushion

the tensiometers, soil is excavated and mixed withwater to form a paste. The tensiometer is then

inserted into the hole which is then sealed with

bentonite. The depths of the porous tips a re typically6 inches, 12 inches, and 18 inches in each nest of

thre e. Finally , the system is ready to fill with wate r.

The small flexible bag used can be an intravenous

bag from the medical profession, available in a range

21



of sizes from a few hundred millilite rs to larger sizes.

A ruler taped to the inside of the outer ring can be

used to monitor the water level, which should be keptto within +/- 1 inch of its or iginal level.

When the construction process is complete, the en tire

uni t is covered with a tarp. The t ar p minimizes waterevaporat ion and keeps the temperature f rom

fluctuating.

A f te r t he i n f i l t r ome t e r h a s be e n i n s t a l l e d ,measurements are taken over a period lasting a tleast 2 weeks and often as much as 1 to 2 months.

Readings involve removing the bag, weighing thebag, and refilling it with water. Readings might betaken as infrequently as once a week or as frequently

as once a day, depending on the situation.

An experienced group of people can put in a sealeddouble ring infiltrometer in 1 day. Two days is more

typical for less experienced people.

The cost of the equipment to build a sealed doublering infiltrometer is about $3,000.The tensiometers,

grout, and equipment rental typically add another$1,500.The total cost for equipment and installation,plus the periodic monitoring of the flow rate andanalysis of test data is approximately $10,000, not

including the cost of a trial pad. The sealed doublering infiltrometer itself is reusable, therefore the

$3,000 cost of the rings is recoverable. In comparison

to the cost of in f i l t romete r ins ta l l a t ion and

operation, a single laboratory hydraulic conductivitytest costs only $200 to $400.

Issues Associated with F ield Hy draulicCon ductivity Testing

A number of issues are associated with all fieldhydraulic conductivity tests. Most importantly, the

tes t s do not d i rec t ly measure the hyd raul i c

conductivity (k) f the soil. Instead they measure the

infiltration rate ( I ) for the soil. Since hydraulicconductivity is the infiltration rate divided by the

hydraulic gradient (i)(see equations in Figure 2-18),it is necessary to determine the hydraulic gradient

before k can be calculated. The following equation(with terms defined in Figure 2-18) can be used to

estimate the hydraulic gradient:

i = ( D + Lf)/Lf

This equation assumes the pressure head at the

wetting front equal to zero. The value of the pressure

head is, however, a source of disagreement and one ofthe sources of uncert a inty in this tes t . Theassumption that the pressure head is zero is aconservative assumption, tending to give a high

hydraulic conductivity.

I

Soil Liner

I = infiltrat ion Rate

= (Quantity of Flow/Area)TTime

= (QiA)/t

k = Hydraulic Conductivity

= Q/(iAt) = iii

Figure 2-18. Hydraulic gradient.

A second issue is that of effect ive s t ress o roverburden stress. The overburden stress on the soilis essentially zero at the time the test is performed,whi le under operat ing condi t ions , i t may be

substantial. The influence of overburden stress on

hydraulic conductivity cannot be estimated easily in

the field, but can be measured in the lab ora tor y.Using conservative estimates, the shape of the field

curve should be the same as that obtained in thelaboratory (see Figure 2-19). If there is significant

overburden stress under actual field performance,the infiltrometer test measurements would need tobe adjusted according to the laboratory result s.

A third i ssue that must be considered is the effect ofsoil swelling on hydraulic conductivity (see Figure 2-20). Tests on highly expansive soils almost always

take longer than tests with other soils, typicallylasting 2 to 4 months. This is a particular problem

with soils along the Texas Gulf Coast.

A hydraulic conductivity test is termina ted when thehydraulic conductivity drops below 10-7 cd se c ( see

Figure 2-21). It usually takes 2 to 8 weeks to reach

tha t point, and is usually clear after about 2 monthswhether or not that objective will be achieved.

The ASTM standard for double-ring infiltrometers iscurrently being revised. The existing double-ringtest (D3385) was never intended for low hydraulic

conductivity soil and should not be used on clayliners. A new standard for double-ring infiltrometers

22



I w0

Effective Stress

2-19. Hydraulic conductivity asa function of effective

stress.

Swelling Zone

Soil Liner

2-20. Soil swelling.

bly be availab le in 1990.

Tests versus Laboratory Tests

of test ing soil l iner

rial s will involve both laboratory and field tests.

a

tes ts . A f i e l d t e s t is a l s o m o r e

ive and more reliable.

primary advantage of laboratory tests is tha t they

e less expensive so more of them can be performed.

Termination of Test

stopTest Time

Figure 2-21.Termination of testing.

satur ate th e soil fully in the laborato ry, getting rid ofall the gas. One can also vary the overburden st ress

in the lab, which cannot be done conveniently in the

field. Finally, in the lab, actual waste liquids can bepassed through a column of material for testing, acondition that could not be duplicated in th e field.

There is a radical variation in the reliability of field

tests versus laboratory tests. In the Houston t est paddiscussed earl ier the real value for hydraulicconductivity in the field was 1 x 10-4 c d s e c while the

lab values were 1 x 10-9 cm/sec, a 100,000 -folddifference i n th e values.

At the Keele Valley landfill, just outside Toronto,

however, some excellent field data have been

obtained. At this particular site, a %foot clay liner

spanning 10 acres is monitored by a series ofunderdr ains. Each underdr ain measure s 1 5 m2 and

i s made of h igh dens i ty po lye thylene . Theunderdrains track the liquid as it moves down

through the soil liner. The underdrains have been

moni tored for more than 2 y e a r s a n d h a v econsistently measured hydraulic conductivities ofabout 1 x 10-8 c d s e c . Those field values essentially

are identical to the laboratory values.

The clay liner at Keele Valley w a s built verycarefully with strict adherence to constructionquality assurance. The laboratory and field values

ar e the same because the line r is essential ly free ofdefects. Lab and field values differ when the soilliner in the field contains defects that cannot be

simulated accurately on small specimens. If the soil

23



is homogeneous, lab and field tests should compare

very well.

Attack by Waste Leachate

Acids and Bases

Acids can attack soil by dissolving the soil mineralsinto other constituents. Typically, when acids are

passed through soil, hydraulic conductivity dropsbecause the acids dissolve the mater ials tha t help to

neutralize them. After large amounts of acid washinto the soil, hydraulic conductivity decreases.

There is real concern over waste impoundments usedto store acidic liquid. Small amounts of acid such asthat contained in a barrel in a solid waste landfill

underla in by a 3-foot thick liner will not inflict majordamage on the soil liner. A large volume of liquid in

the impoundment, however, can damage the soilseriously.

Neutral Inorganic Compounds

Nonacidic liquids can change hydraulic conductivity

in other ways. Soil is made up of colloidal particlesthat have negative charges along the surface. Wateris a polar molecule, with atoms arranged or alignedasymetrically. This alignment allows the wate r

molecule to be attracted electrochemically to thesurfaces of the negatively charged soil particles (seeFigure 2-22).

Figure 2-22. Water and clay particle molecules.

I t is also possible for ions in the water, especiallypositively charged ions, o r cations, to be attracte d tothe negatively charged surfaces. This leads to a zone

of water and ions surrou nding the clay particles,

known as the diffuse double layer.

The water and ions in the double layer a re attra ctedso strongly electrochemically to the clay particlestha t they do not conduct fluids. Fluids going throughthe soil go around th e soil particles and, also, aroundthe double layer. The hydraulic conductivity of thesoil , then, is controlled very strongly by the

thickness of these double layers. When the double

layers shrink, they open up flow paths resulting inhigh hydraulic conductivity. When the layers swell,they constrict flow paths, resulting in low hydraulic

conductivity.

The Gouy-Chapman Theory relates electrolyte

concentration, cation valence, and dielectric constantto the thickness of this double layer (see Figure 2-23). This theory was originally developed for dilute

suspens ions of sol ids in a l iquid. However ,experience confirms that the principles can be

applied qualita tively to soil, even compacted soil th atis not in suspension.

Gouv-Chapman:

gzhickness Q

D = Dielectric Constant

no = Electroylte Concentration

V = Cation Valence

Figure 2-23. Gouy-Chapman Theory.

The following application of the Gouy-ChapmanTheory uses sodium bentonite. The ion in the soil issodium, which has a charge of + 1 . The electrolytevalence in the Gouy-Chapman Theory is v = 1 . Th epermeating liquid is rich in calcium, and calcium has

a charge of +2. A s calcium replaces sodium, thevalence (v) in the Gouy-Chapman equation goes from

1 to 2. A rise in v increases the denominator, thusdecreasing the thickness (TI. A s T decreases and thedouble layer shrinks, flow paths open up making t hesoil more permeable, as shown in Figure 2-24. !

Since calcium bentonite, typically, is 100 to 1,000times more permeable than sodium bentonite, the

24



Clay

Particle

\ ouble

Layer

Figure 2-24. The diffuse double layer.

introduction of this permeating liquid could change

hydraulic conductivity substantial ly.

able 2-4 shows that soils containing polyvalentcations having high valence and high electrolyteconcentration have a high conductivity, while the

soils containing monovalent cations, like sodium,have a low k. Distilled water at the extreme end of

the spectrum is free of electrolytes. In the Gouy-Cha pm a n e qua t i on , t he n , no the e lec t ro ly teconcentration, would be 0 . T he de nom i na t o r ,

therefore, would go to 0 and the T value to infinity.

Table 2-4. Electroylte Concentration

High k Water with Polyvalent Cations

Tap Water (Note Variation)

Water with Monovalent Cations

Distilled Water

1Low k

Consequently, if the free ions in the soil water are

leached out, the double layers swell tremendously,

pinching off flow paths and resulting in very low

hydraulic Conductivity. Data have shown hydraulicconductivity to be as much a s two to th ree orders of

magnitude lower when measured with distilledwater than with other kinds of water. F o r th i sreason, distilled water should not be used in testing

the hydraulic conductivity of a clay liner.

An ASTM standard under development recommends

using 0.005 normal calcium sulfate as the standard

permeating water, because of it s medium range

electrolyte concentration. Calcium sulfate, withdivalent calcium, will usually not reduce hydraulic

conductivity.

Neutral Organic CompoundsOrganic chemicals can cause major changes in

hydraulic conductivity. The dielectric constant (D ) of

many of the organic solvents used in industry is very

low. For example, the dielectric constant of water isa b o u t 80, w h i l e t h e d i e l e c t r i c c o n s t a n t o f

trichloroethylene is about 3. Using the Gouy-Chapman equation, if D decreases, which means the

numerator decreases, the value for T will also

decrease, causing the double layer to shrink. Theeffect of replacing wate r with a n organic solvent then

is to shrink the double layer and open up flow paths.

In addition to opening up flow paths, as the doublelayers shr ink, the solvent f loccula tes the soi lparticles, pulling them together and leading to

cracking in the soil. Permeation of the soil with an

organic chemical, such as gasoline, may produce

cracking similar to th at associated with desiccation.The organic solvent, however, produces a chemical

desiccation rather than a desiccation of the soil by

drying out.

Laboratory test data indicate that if the organic

chemical is present in a dilute aque ous solution, the

dielectric constant will not be dangerously low. A

dielectric constant above 30 generally will not lowerthe conductivity substanti ally enough to damage thesoil. Two criteria need to be met for a liquid not to

attack clay liners: (1) the solution must contain atleast 50 percent water, and (2) no separate phase o rorganic chemicals can be present .

Termination Criteria

Chemical compatibil i ty s tudies with hydraulic

conductivity tests must be performed over a long

enough period of time to de term ine the fu ll effects ofthe waste liquid. Termination criteria include equalinflow and outflow of liquid, steady hydraulic

conductivity, and influentleffluent equilibrium. At

least two pore volumes of liquid must be passedthrough the soil to flush out the soil water and bring

the waste leachate into the soil in s ignificant

q u a n t i t i e s ( s e e F i g u r e 2 - 2 5 ) . R e a s o n a b l eequilibrations of the influent and effluent liquids

occur when the pH of the waste influent and effluentliquids ar e similar and the key organic and inorganic

ions are at full concentrations in the effluent liquid.

Resistance to Leachate A ttack

It is possible to make soils more resi stan t to chemicalattack. Many of the same methods used to lower

hy drau ic conduct vi ty can s tab 1 ~e ma e r ia 1s

a g a i n s t 1e a c h a t e a t t a c k , i nc 1 lid i n g g r e a t c I'

25



Water + --) Organic Chemical

I I ,0 1 2


Figure 2-25. Hydraulic conductivity as a function of pore

volumes of flow.

compaction, an increase in overburden stress, andthe mixing of additives such as lime cement orsodium silicate with the na tura l soil materials.

Figure 2-26 shows the results of an experiment

conducted using a soil called S1, an i l l i t ic c lay

containing chlorite from Michigan. Two sets of datashow the results of permeation of the regular soil,first with water and then with pure reagent gradehe p t a ne. T he he p t a ne c a use d t he hyd r a u l i c

conductivity of the regular compacted soil toskyrocket. About 8 percent cement was then added tothe soil.

After treatment of the soil with Portland cement,however, the heptane did not affect the soil even

after a pore volume of flow. The Portland cementglued the soil particles together so that the soilbecame invulnerable to attack, rat her tha n causing

it to undergo chemical desiccation.

References1 . Daniel , D.E. 1987. Ear the n l i ners for land

disposal facilities. Proceedings , Geotechnical

Practice for Waste Disposal 1 987, Uni v. ofMichigan, Ann Arbor, Michigan, June 1987: 21-

39. New York, N Y : American Society of CivilEngineers.

Trautwein Soil Testing Equipment Company.1989. Installation and operat ing instructions forthe sealed double-ring infiltrometer. Houston,

Texas : Trautwein Soi l Tes t ing EquipmentCompany.

2.

t;

Clay Stabilization Research

soil. s1

Stabilization: CementOrganic Chemical: Heptane

10 4

HO 4 eptaneI I

1 0 - 6

l o - \

t

r'

: /

Unamended

Cement Stabilized

-2.00 0.00 2.00 4.00 6.00 8.0


Figure 2-26. Illitic-chlorated clay treated with heptane and

with Portland cement.

3. U.S. EPA. 1986a. U .S . Environment al ProtectionAgency. Design, construction, and evaluation of

clay liners for waste management facilities.Technical Resource Document. Report N o .EPA/530/SW-86/007, NTIS Or de r N o . PB86-

184496/AS. Cincinnati, Ohio: EPA.

4. U.S . EPA. 298613. U.S. Environment al Protection

Agency. Saturated hydraul ic conduct ivi ty,saturated leachate conductivity, and intrinsicpermeabi lity. EPA Method 9100. EPA.

26



3. FLEXIBLE MEMBRANE LINERS

highlights some of the problems encountered in

ons on materia l and design considerations.

a basin has

or more

of water-retaining material (liners) that form

a network of pipes and collector

mer shee t s o r of c l ay . EPA's mi n i mu m

a composite liner that utilizes advantages

ing both liner systems.

liners is very important in

a composite liner.

(Q = kiA)

is discussed in more detail in Chapter

In clay liners, the factors th at most influencermance ar e hydraul ic head and soil p e r m e -

Clay l ine rs have a higher hydraul i c

ctivity and thickness tha n do synthetic liners.

ionally, leachate leaking through a clay liner

tion of contaminants in the leachate.

a synthetic liner is controlled bys fi rst law, which applies to the process of liquid

fusion through the liner membrane. The diffusion

is similar to flow governed by Darcy's lawis driven by concentration gradients and not

s. In synthetic liners, therefore, the factor th at

p e n e t r a t i o n s .

Synthetic l in ers may have imperfect se ams o rpinholes, which can greatly increase the amount of

leachate tha t leaks out of the landfill.

Clay liners, synthetic liners, or combinations of both

ar e required in landfills. Figure 3 -1 depicts the

syntheticlcomposite double liner system that appear s

in EPA's minimum technology guidance. The systemhas two synthetic flexible membrane liners (FMLs):the p r i m a r y F M L , which lies between two leachate

collection an d removal sys tem s (LCRS), and the

secondary F M L , which overlies a compacted c l a yliner to form a composite secondary liner. The ad-

vantage of the composite liner design is that byputting a fine grain mat erial beneath the membrane,

the impact of given penetrations can be reduced by

many orders of magnitude (Figure 3-2). In the figure,

Qg s the inflow rate with gravel a nd Qc is the inflow

rat e with clay.

Figure 3-3 is a profile of a liner that appeared in anEPA design manual less than a year ago. This

system is already dated. Since this system was

designed, EPA ha s changed the m inimum hydraulic

conductivity in the secondary leachate collectionsystem from 1 x 10-2 c d s e c to 1 c d s e c to improve

detection time. To meet this requirement, either

gravel or a net made of synthetic material must be

used to build the secondary leachate collection

system; in the past, sand was used for th is purpose.

Material Considerations

Synthetics are made up of polymers-natural or

synthetic compounds of high molecular weight.Different polymeric materials may be used in the

construction of FMLs:

0 Thermoplastics-polyvinyl chloride (PVC)

Crystalline thermoplastics-high den sit y poly-

ethylene (HDPE), linear low densi ty polyethylene

(LLDPE)

27



Native Soil Foundation

(Not to Scale)

Figure 3-1. Synthetidcomposite double liner system.

-rf c

Q, 1 0 5 Q,

Figure 3-2. Advantage of composite liner.

0 T he rmop1as ic e1as omer s-c hlo r ina ed PO 1y-

ethylene (CPE), chlorylsulfonated polyethylene

(CSPE)

Elastomers-neoprene, ethylene propylene dienemonomer ( E P D M )

Typical compositions of polymeric geomembranes

are depicted in Table 3-1 . A s the table shows, themembrane s contain various admixtures such as oils

and fillers that a re added to aid manufacturing of the

F M L but may affect future performance. In addition,many polymer F M L s will cure once installed , and the

strength and elongation characteristics of certain

F M L s will change with t ime. I t is i m por t a n t

15 cm Filter Media

I I Hvdraulic Conductivity I

3~~

cmlx 1 0 - 7 cmiseQ I

Hydraulic Conductivib

Primary LCR

- Primary FML

Secondary LCR

'- Secondary FML

Hydraulic Conductivity

5 1 x 10.7 cmisec Clay Liner

.-Unsaturated Zone

Native Soils

'Minimum hydraulic conductivity is no w 1 cmisec

Figure 3-3. Profile of MTG double liner system.

therefore to select polymers for F M L construction

with care. Chemical compatibility, m anufa cturi ngconsiderations, stress- strain characteristics, sur -

28



must be considered.

ical C ompatibility

a FML with waste

is a n important material consideration.

d to construct FMLs. (EPA Method 9090 tests a re

U n f o r -is a lag period between the

of a facility. It is very rare that at the

ction lot as the synthetics installed in the field.

ular st ruc ture of different polymers can be

r thermogravimetr ic tes t ing. This tes t ing o r

fo r the 9090 test was used in th e field. Figure 3-

was provided by a HDPE manufacturer, and the

a

ne sheets are produced in various w a y s :

Extrusion-HDPE

Calendaring-PVC

Spraying-Urethane

e t s . H ow e ve r , t he

3-1. Basic Composition of Polymeric Geomembrane

180"C, 800 psig-

28

Time (min)

Figure 3-4. Comparison of "fingerprints" of exothermic peak

shapes.

compatibility of extrusion welding rods and high

density polyethylene sheets can be a problem. Somemanufacturing processes can cause high density

polyethylene to crease. When this material creases,

stress fractures wil l result. If the material is takeninto the field to be placed, abrasion damage will

occur on the creases. Manufacturers have been

working to resolve this problem and, for the most

part, sheets of acceptable quality are now being

produced.

Stress-Strain Characteristics

Table 3-2 depicts typical mechanical properties ofHDPE, CPE, and PVC. Tens i le s t rength is a

Composition of Compound Type

(parts by weight)

Component Crosslinked Thermoplastic Semicrystalline

Polymer or alloy 100 100 100

Oil or plasticizer 5-40 5-55 0-10

Fillers:

Carbon Black

lnorganics

5-40

5-40

5-40

5-40

2-5

Antidegradants 1-2 1-2 1

Crosslinking system:.._norganic system 5-9

Sulfur system 5-9 _.-

E. 1986. Quality Assurance of Geomembranes Used as Linings for Hazardous Waste Containment. In: Geotextiles and

Geomembranes, Vol. 3, No. 4. London, England.

29



fundamental design consideration. Figure 3-5 shows

the uniaxial str ess- str ain performance of HDPE,

CPE, and PVC. As 600,800,1,100,and 1,300 percent

strain is developed, the samples fail. When biaxialtension is applied to HDPE, the material fails atstrains less than 20 percent. In fact, HDPE can fail a t

strains much less than other flexible membraneswhen subjected to biaxial tensions common in thefield.

Another stress- strain consideration is th at highdensity polyethyl,ene, a material used frequently athazardous waste facilities, has a high degree of

thermal coefficient of expansion - three to four times

that of other flexible membranes. This means thatduring the course of a day (par t icular ly in the

summer) , 100-degrees-Fahrenheit ( O F ) variations inthe temperature of the sheet ing are rout inely

measured. A 600-foot long panel, for example, maygrow 6 feet during a day.

Survivability

Various tes ts may be used to determine thesurvivability of unexposed polymeric geomembranes(Table 3-3). Puncture tests frequently are used toestimate th e survivab ility of FMLs in the field.

During a puncture test, a 5/16 steel rod with rounded

edges is pushed down through the membrane. A very

flexible membrane that has a high strain capacityunder biaxia l tens ion may al low that rod to

penetrate almost to the bottom of the chamber

r up t u r e . S uc h a m e m b r a n e h a s a very low

penetration force but a very high penetration elonga-

tion, and may have great survivability in the field.High density polyethylenes will give a very high

penetration force, but have very high brittle failure.

Thus, puncture data may not properly predict fieldsurvivability.

Permeability

Permeability of a FML is evaluated u sing the WaterVapor Transmission test (ASTM E96). A sample of

the membrane is placed on top of a small aluminum

cup containing a small amount of water. The cup ist he n p l ac e d i n a c o n tr o l le d h u m i d i t y a n dtemperature chamber. The humidity in the chamber

is typically 20 percent relative humidity, while the

humidi ty in the cup is 100 percent . Thus , a

c onc en t r a ti on g r a d ie n t i s s e t up a c r os s t hemembrane. Moisture diffuses through the membrane

and with time the liquid level in the cup is reduced.The rate at which moisture is moving through the

membrane i s measured . F rom tha t ra t e , the

permeability of the membrane is calculated with the

simple diffusion equation (Fick’s first l a w ) . I t isimportant to remember that even if a l iner i sinstalled correctly with no holes, penetrations,

punctures, or defects, liquid will stil l diffuse through

the membrane.

Design Elements

A number of design elements must be considered in

the construction of flexible membrane liners: (1m i n i m u m t e c hn o l og y g u i d a n c e , ( 2 ) s t r e s s

considerations, (3 ) structural details, and (4) anel

fabrication.

Minimum Technology Guidance

EPA has set minimum technology guidance for thedesign of landfill and surface impoundment liners to

achieve de minimis leakage. De minimis leakage is 1

gallon per acre per day. Flexible membrane liners

must be a minimum of 30 mils thick, or 45 mils thickif exposed for more than 30 days . There may,

however, be local variations in the requirement of

minimum thickness, and these variations can have

an impact on costs. For example, membranes costapproximately $.01 per mil per square foot, so that

increasing the required thickness of the FML from

30 mils to 60 mils, will increase the price $.30 cents

per square foot or $12,000per acre.

Stress

Stress considerations must be considered for sideslopes and the bottom of a landfill. For side slopes,

self-weight (the weight of the membrane itselfl andwaste settlement must be considered; for the bottom

of the facility, localized settlement and normalcompression must be considered.

The primary FML must be able to support its own

weight on the side slopes. In order to calculate self-

weight, the FML specific gravity, friction angle,

FML thickness, a nd FML yield st res s must be known(Figure 3-6).

Waste settlement is another consideration. As waste

settles in the landfill, a downward force will act on

the primary FML. A low friction component betweenthe FML and underlying material prevents that

force from being transferred to the underlying

materia l, putting tension on the primary FML. A 12-inch direct shear te st is used to measure the friction

angle between the FML and underlying materi al.

An example of the effects of waste se ttlement can be

illustrated by a recent incident at a hazardous waste

landfill facility in California. At this facility, wastesettlement led to sliding of the waste, causing the

standpipes (used to monitor second ary lea cha te

collection sumps) to move 60 to 90 feet downslope in

1 day. Because there was a very low coefficient of

friction between the primary liner and the geonet,

the waste (which was deposited in a canyon) slid

down the canyon. There w a s also a failure zone

between the secondary liner and the clay. A two-

30



3-2. Typical Mechanical Properties

HDPE CP E PVC

Density, gm/cms >,935 1.3 - 1.37 1 24 - 1.3

Thermal coefficient of expansion 12.5 x 10-5 4 x 10-5 3 x 10-5

Tensile strength, psi 4800 1800 2200

Puncture, Ib/mil 2.8 1.2 2.2

4000 I 1 I I I

./ * /'

-" /'2000 ' ' /

/;t / CP E (Uniaxial) x

CPE (Biaxial)

I0 100 200 300

Strain, %

3-5. FML stress-strain performance

(uniaxial-Koerner, Richardson; biaxial-Steffen).

ed a factor of safety grea ter than one. A three-

FM L

Figure 3-7 is a simple

on the F M L. Settlements along trenchesl cause stra in in the membrane, even if the trench

is applied to high density polyethylene, theial fails at a 16 to 17 percent strain , it is

400 500

~~

PSI at 1180%

possible that the membrane will fail at a moderate

settlement ratio.

Another consideration is the normal load placed onthe membranes as waste is piled higher. Many of thenew materials on the market, particularly some ofthe l inear low density polyethylene (LLDPE) liners,will take a tremendous amount of normal loadwithout failure. The high density polyethylenes, onthe other hand, have a tendency to high brittlefailure.

Structural Details

Double liner systems a re more prone to defects in thest ructural detai l s (anchorage, access ramps,collection standpipes, and penetrations) than singleliner systems.

31



Table 3-3. Test Methods for Unexposed Polymeric Geomembranes

Membrane Liner Without Fabric Reinforcement

Property Thermoplastic Crosslinked Semicrystalline Fabric Reinforced

Analytical Properties

Volatiles MTM-la MTM-la MTM-la MTM-la

(on selvage and

reinforced sheeting)

(on selvage andreinforced sheeting)

ASTM D297, Section 34

(on selvage)

ASTM D792, Method A

(on selvage)

MTM-2aTM-2a MTM-2a MTM-2axtractables

ASTM D297, Section 34 ASTM D297, Section 34 ASTM D297, Section 34sh

Specific gravity ASTM D792. Method A ASTM 0297, Section 15 ASTM 0792, Method A

Thermal analysis:

Differential scanning

calorimetry (DSC)

Thermogravimetry

(TGA)

NA

Yes

NA

Yes

Yes

Yes

NA

Yes

Physical Properties

Thickness - total

Coating Aver fabric

Tensile properties

ASTM 0638

NA

ASTM D638

(modified)

ASTM 0638

NA

ASTM D882,

ASTM D638

ASTM 0412

NA

ASTM D412

ASTM 0751, Section 6

Optical method

ASTM D751, Method A

and B (ASTM D638 on

selvage)

ASTM D751, Tongue

method (modified)

NA

ASTM D2240 Duro A

or 0 (selvage only)

FTMS 101B,

Methods 2031 and 2065

ASTM D751, Method A

Tear resistance ASTM 01004

(modif ed)

NA

ASTM D2240

Duro A or D

FTMS 101B,Method 2065

NA

ASTM 0624 ASTM D1004

Die C

ASTM D882, Method A

ASTM 02240

Duro A or D

FTMS 1018,

Method 2065

ASTM D751, Method A

Modulus of elasticity

Hardness

NA

ASTM D2240

Duro A or D

FTMS 1018,

Method 2065

NA

Puncture resistance

Hydrostatic resistance

Seam strength:

In shear ASTM D882, Method A

(modified)

ASTM D413, Mach

Method Type 1

(modified)

NA

ASTM 0882, Method A

(modified)

ASTM D413, Mach

Method Type 1

(modified)

NA

ASTM D882, Method A

(modified)

ASTM 0413, Mach

Method Type 1

(modified)

NA

ASTM 0751, Method A

(modified)

ASTM D413, Mach

Method Type 1

(modified)

ASTM D413, Mach

Method Type 1

ASTM D751, Sections

39-42

In peel

Ply adhesion

Environmental and Aqing

Effects

Ozone cracking

Environmental stress

cracking

Lo w temperature testing

ASTM D1149

NA

ASTM D1790

ASTM D638 (modified)

ASTM D1204

ASTM D1149

NA

ASTM D746

ASTM 0412 (modified)

ASTM D1204

NA

ASTM Dl6 93

ASTM D1790

ASTM 0746

ASTM D638 (modified)

ASTM D 1204

ASTM D1149

NA

ASTM D2136

Tensiie properties at

Dimensional stability

elevated temperature

ASTM 075 1 Method B

(modified)

ASTM D1204

32



3-3. Test Methods for Unexposed Polymeric Geomembranes (continued)

Design Ratio:'10 ou Y I E L D

Membrane Liner Without Fabric Reinforcement

References:

Property Thermoplastic Crosslinked Semicrystalline Fabric Reinforced

Air-oven aging ASTM D573 (modified) ASTM D573 (modified) ASTM D573 (modified) ASTM D573 (modified)

Water vapor transmission ASTM E96, Method 0W ASTM E96, Method B W ASTM E96, Method BW ASTM E96, Method B W

Water absorption

Immersionin standard ASTM D471, D543 ASTM D471 ASTM D543

Immersion in waste

Soil burial

Outdoor exposure

ASTM D570 ASTM D471 ASTM D570 ASTM 0570

ASTM D471. D543

liquids

liquids EPA 9090 EPA 9090 EPA 9090 EPA 9090

ASTM D3083 ASTM D3083 ASTM D3083 ASTM D3083

ASTM D4364 ASTM D4364 ASTM D4364 ASTM 04364

e reference (8) .e e reference (12).

= not applicable.

Haxo, 1987

Cell Component: FLExlbLE M e M 0 R A U E L IUE P

Figure 3-6. Calculation of self-weight.

33



20

15

5

Circular Trough Model

0 0.1 0.2

Settlement Ratio, Si2L

Figure 3-7. Settlement trough models (Knipschield,1985).

Anchorage

Anchor trenches can cause FMLs to fail in one of twoways: by ripping or by pulling out. The pullout modeis eas ier to correct. It is possible to calculate pulloutcapacity for FMLs placed in various anchorageconfigurations (Figures 3-8 an d 3-9). In the "V "anchor configuration, resistance can be increased byincreasing the "V" angle. A drawback to using the"V" design for getting an accurate estimate of

pullout capacity is that it uses more space. Theconcrete trench is not presently used.

Ramps

Most facilities have access ramps (Figure 3-10),which are used by trucks dur ing construction and bytrucks bringing waste into the facility. Figure 3-11depicts a cross section of a typical access ramp. Thedouble FML integrity must be maintained over theentire surface of the ramp. Because ramps can faild u e t o tr a f f ic - i nd u ced s l i d i n g , r o ad wa yconsiderations, and drainage, these three factorsmu s t -b e co ns i de red d u r i n g t h e d e s i gn an d

construction of access ramps.

The weight of the roadway, the weight of a vehicle onthe roadway, and the vehicle braking force all mustbe considered when evaluating the potential for

0.3

2L

slippage due to traffic (Figure 3-12). The vehiclebraking force should be much larger than the deadweight of the vehicles that will use it. Wheelloadsalso have an impact on the double FML system and

the two leachate collection systems below theroadway. Trucks with maximum axle loads (somemuch higher than the legal highway loads) and 90psi tires should be able to use the ramps. F igure 3-13illustrates how to verify that, wheel contact loadingwill not damage the FML. Swells or small drains

may be constructed along the inboard side of aroadway to ensure that the ramp w i l l adequatelydrain water f rom the roadway. F i g u re 3-14 illustrates how to verify tha t a ramp will drain water

adequately. The l iner system, which must beprotected from tires, should be armored in the are a ofthe dra inage swells. A sand subgrade contained by ageotextile beneath the roadway can prevent local

sloughing and local slope failures along the side ofthe roadway where the dra ins are located. The sandsubgrade tied together with geotextile layers forms,basically, 800-foot long sandbags stacked on top ofone another.

Vertical Standpipes

Landfills have two leachate collection and removalsystems (LCRSs): a primary LCRS and a secondaryLCRS. Any leachate that penetrates the primary

34



Material Properties Range

fP*,=Ts.y U $ L € 12-20'

5rnL f Q t c T 8 e U A U q L C 2 5 - 38 "

'. $ 0. H

PC F,+,ac,3 q e 3 IO

n Ratio: References:

N O T At'PLlC4bLL

3-8. Calculationof anchor capacity.

oved. Vertical standpipes (Figure 3-15) a r eary leachate collection sumps.

s waste settl es over time, downdrag forces can haven impact on standpipes. Those downdrag forces can

of the primary FM L beneath theFigure 3-16 illustrates how to verify that

reduce the amount of downdrag force on the waste

of H D P E . This material has a

lo w coefficient of friction that helps reduce theoun t of downdrag force on the waste piles. Figure

illustrates how to evaluate the potential

e forces.

IExampie No. 3.17

Downdrag forces also affect the foundation orsubgrade beneath the standpipe. If the foundation i srigid, poured concrete, there is a potential for sig-nificant s tra in gradients. A flexible foundation willprovide a more gradual transition and spread thedistribution of contact pressures over a largerportion of the F M L than will a rigid foundation(Figure 3-18). To so f t en r ig id foundat ions ,

encapsulated steel plates may be installed beneaththe foundation, as shown in Figure 3-15.

Standpipe Penetrations

The secondary leachate collection system is accessedby collection standpipes that must penetrate theprimary liner. There are two methods of makingthese penetrations: rigid or flexible (Figure 3-19). Inthe rigid penetrations, concrete anchor blocks are setbehind the pipe with the membranes anchored to theconcrete. Flexible penetrations are preferred sincethese allow the pipe to move without damaging the

35



Horizontal Anchor

tL

" V " Anchor

A T

Concrete Anchor

Figuri3-9. Forces and variables- anchor analysis.

liner. In either case, standpipes should not be weldedto the liners. If a vehicle hits a pipe, there is a high

potential for creating major tea rs in t he li ner a tdepth.

36



structure

3-10. -Geometry of typical ramp.

18' Typicalt

Figure 3-1 1. Cross section of typical access ramp.

Wind Damage

During the installat ion of FMLs, care must be takento avoid damage from wind. Figure 3-20 showsmaximum wind speeds in the United States.Designers should determine if wind wi l l affect an

37



Reauired Material Properties I Range I Test IStandardI 1

Figure 3-12. Calculation of ramp stability.

installation and, if so, how many sandbags will be

needed to anchor the FML panels as they are beingplaced in the field. Figure 3-21 shows how tocalculate the required sandbag spacing for FMLpanels during placements. Wind-uplift pressuremust be known to make this calculation. Using thedata in Table 3-4, the uplift pressures acting on themembranes may be calculated.

Surface Impoundments versus LandfillsThere are significant differences in structuralconsiderations between landfil ls and surfaceimpoundments. Fi rst , l iners used in surfaceimpoundments have a long-term exposure to thewaste and to sunl ight . In addi t ion, surface

impoundments have a potential for gas in theleachate collection and removal system becausethere will always be the potential for organicmateri al beneath the system.

[Example No. 4.1

Long-term exposure can be stopped using either soilor a nonwoven fabric to cover the membrane in asurface impoundment. Figure 3-22 illu strates how tocalculate the stability of a soil cover over themembrane. Another option is to drape a heavy,nonwoven fabric with base anchors in it over themembrane. This nonwoven material is cheaper,safer, and more readily repaired th an a soil cover.

Gas or liquid generated "whales" can be a serious

problem in surface impoundments. Water-induced"whaling" can be a problem in facilities that are

located where there is a high water table. Stormwater can also enter a collection system through gasvents. Figure 3-23 illustra tes two gas vent designs inwhich the vent is placed higher than the maximumoverflow level. If excess water in the leachate

collectors is causing whaling, the perimeter shouldbe checked to determine where water is entering. Torepair a water-generated whale, the excess watershould be pumped out of the sump and its source



I 1 C Im n

:

3-13. Calculationof wheel loading capacity.

is gas in the whale (liner is inflatedsible above th e w ater surface ), the facility

t be rebuilt from scratch.

pect to consider is the FML panel

a FML panel layout: (1)

(2 ) the field seam length should be

(3) there shoulda FML below the top of the

he facility. Figure 3-24 depicts the panel-

g scheme also assures a high qua lity

ry al l samples cut from the FML panel dur ing

I

IExample No. 4.3

installation. The samples cut from the panels aretested to ensure the installation is of high quality.Quality assurance and the panel-seam identificationscheme are discussed in more detail in Chapter

Seven.

References1. Haxo, H.E. 1983. Analysis and Finge rprinting of

Unexposed and Exposed Polymeric Membrane

Liners. Proceedings of the Ninth Annual

Research Symposium, Land Disposal ofHazardous Waste, U S . EPA 600/8-83/108.

Knipschield, F.W. 1985. Materia l, Selection, an dDimensioning of Geomembranes for Ground-water Protection. W a s te a n d Re f u s e . SchmidtPublisher, Vol. 22.

2.

39



D R -

Design Ratio:

DR.,,> 1.5

QAGT

4

References: IIExample No. 4.2

Figure 3-14. Calculation of ramp drainage capability.

40



Figure 3-15. Details of standpipddrain.

42



I2 [Example No. 4 . 5uR a - 5 - 5 . 2 -r

3-16. Verification that downdrag induced settlement will not causeLCR failure.

43



Design Ratio:

IJ~ AP PUCAOLE

[Example No. 4 .4

Figure 3-17. Evaluation of potential downdrag forces on standpipes with and without coating.

44



P edge

0.1 0 rla

1 I I 1 I I 1 I

2a

-1.0 -- Rigid Foundation------

A) Elastic Settlement Constant

Clay Subgrade Sand Subgrade

‘I (after Terzaghi and Peck)

Elastic Subgrade

I

B)Distribution of Contact Pressures

3-18. Standpipe induced strain inFML.

45



Geotextile J

w

Rigid Penetrations

Geqtextile

Flexible Penetrations

Primary FML

Secondary FML

Figure 3-19. Details of rigid and flexible penetrations.

46



3-20. Design maximum wind speeds.

47



Cell Comwnent: F L L ~ , C + ELMBUALJE

Required Material Properties Range

* d u . r clc-iwr 3 Z b , Y

F L ~ I ~ L L' M B ~ L U BIF

Example:-4 I*Eh(:

Figure 3-21. Calculationof required sandbag spacing fo r FMUFMC panels.

Table 3-4. Wind-Uplift Forces, PSF (Factory Mutual System)

Height Wind Isotach, mph

Above

GroundFlat, Open Country, or Open Coastal Belt > 1500 ft from Coastity, Suburban Areas, Towns, and Wooded Areas

(ft) 70 80 90 100 110 70 80 90 100 110 120

10 13 17 21 25 16 21 27 33 40 48

12 15 19 24 29 18 24 30 37 44 35

14 18 22 27 33 20 26 33 40 49 85

1oa 11 14 17 20 14 18 23 29 35 140-15

30

50

75

aUplift pressures in PSF

48



inalysis Procedure:

REFEREUCEIS3.1L

IExample No. 3.18

3-22. Calculationof soil cover stability.

49



Creation of “whales.”

50



Place Vent Higher than Maximum Liquid Level

at Over-Flow ConditionsIn";- AlrIGas Vent

Two-Inch Minimum

Drainage Composite

Openings in Vent to be Higher than

Top of Berm or Overflow Liquid Level

AirIGas Vent Assemblyr

Liner1 &Drainage Cor$&

Approx Six-Inches

Bond Skirt of

Gas Flow

3-23. Gas vent details.

51

Wind Cowl DetaileGeomem brane

'e

-oncrete

Vent to Liner



l- ---- -I-----1 I;1---- r--- --l--"

I I II II

1 II

A A A A A h i0 Panel Number

A Seam Number

Figure 3-24. Panel-seam identificaiton s c he m e .

52



4. ELEMENTS OF LIQUID MANAGE MENT AT WASTE CONTAINMENTSITES

a liner has no leaks, the phenomenon of molecular

lection and removal of that leachate ist the heart of this chapter.

presents an overview of collector designmater ials , followed by a discussion of the three

a liquid management system: the leachate

e surface water collection system above

a discussion of gas collector and

Overview

Drainage MaterialsFiltration MaterialsGeosyntheticsDesign-by-function Concepts

Primary Leachate Collect ion and Removal(PLCR) Systems

Granular Soil (Gravel) Drainage DesignPerforated Collector Pipe DesignGeonet Drainage Design

Granular Soil (Sand) Filt er DesignGeotextile Filte r DesignLeachate Removal Systems

Leak Detection, Collection, and Removal (LDCR)Systems

Granular Soil (Gravel) Drainage DesignGeonet Dra inage Design

Response TimeLeak Detection Removal Systems

Surface Water Collection and Removal (SWCR)Systems

Gas Collector and Removal Systems

OverviewLeachate refers to rainfall and snowmelt thatc o m b i n e s w i t h l i q u i d i n t h e w a s t e a n dgravitationally moves to the bottom of a landfillfacility. During the course of its migration, t he liquidtakes on the pollutant characteristics of the wasteitself. A s such, leachate is both site specific andwaste specific with regard to both its quantity andquality. The first part of the collector system tointercept the leachate is the primary leachatecollection and removal (PLCR) syst em locateddirectly below the waste and above the primaryliner. This system must be designed and constructedon a site-specific basis to remove the leachate for

proper tr eat ment and disposal.

The second part of a leachate collection system isbetween the primary and secondary liners. Varyingwith State or region, it is called by a number ofnames including the secondary leachate collectionand removal (SLCR) system, the leak detecti onnetwork, o r the leak questioning system. It wil l bereferred to here as the leak detection, collection, andremoval (LDCR) system. The main purpose of thissystem is to determine t he degree of leakage, if any ,of leachate through the primary liner. Ideally, thissystem would collect only negligible quantities ofleachate; however, it must be designed on the basis ofa worst-case scenario.

The third part, called the surface water collectionand removal (SWCR) system, lies above the wastesystem in a cap or closure above the closed facility.Its purpose is to redirect surface water comingthrough the cover soil from off of the flexiblemembrane in the cap to the outside perimeter of the

5 3



system. The location of all thrre parts of t h e liquidmanapcment system is illustra ted in Figure 4 1 .

Drainage Mate ials

The drainage materials for the liquid managementsystem must allow for unimpeded flow of liquids forthe intended lifetime of the facility. In a leachatecollection system, the drains may consist of pipes,soil (gravel), geonets, o r geocomposites. These

mate rial s will be described in the following sections.

Perforated drainage pipes have the advantage ofcommon usage an d design, and they tr ansmi t fluidsrapidly. They do, however, require considerablevertical space, and are susceptible to particulateclogging, biological clogging, and creep (deflection).Creep is of concern for both polyvinyl chloride (PVC)and h igh dens i ty po lye thy lene (HDPE) p ipematerials.

According to proposed EPA regulat ions , thehydraulic conductivity value for soil used as thedrainage component of leachate collection systems

will increase over previous regulations by two orders

of magnitude, from 0.01 cmlsec to 1 cd se c, in thevery near fu ture. This regulat ion essent ial lyeliminates the use of sand, and necessitates the useof gravel. Gravel that meets this regulation hasparticle sizes of 114 to 112 inches and must be quiteclean with no fines content. While gravels of thist y pe a r e d u r a b l e a n d h a v e h i g h h y d r a u l i cconductivities, they require a filter soil to protectthem. They also tend to move when waste is loadedonto the landfill o r personnel walk on them. For the

latt er reason, they ar e practically impossible to placeon side slopes.

The synthetic materials that best meet inplane flowrate regulations are called geonets. Geonets requireless space than perforated pipe or gravel, promoterapid transmission of liquids, and, because of theirrelatively open apertures, are less likely to clogThey do, however, require geotextile filters abovethem and can experience problems with creep andintrusion. Geonets have the disadvantage of beingrelatively new and , therefore, less familiar to ownersand designers than are sand and gravel drainage

materials.

---..-Operational Cover

Figure 4-1. The three elements of a liquid management drainage system in a double-lined solid waste facility.

54



eonets. They generally ar e notor secondary leachate collection

is not sufficient to carry thea large landfill . Geocomposites ar e useful,

er, for surface water collector systems, wherermal stresses a re quit e low.

or apertures in geonets,

as a filter. Commonly in aor gravel drain, a medium-coarse to fine sandy

as a fil ter. Sand, however, has the

ng under various loading conditions.

as filters avoid these problems. The

A s

extiles a re a new technology much about them is

as cushioningials above and/or below FMLs.

a key role in liquid

are :

Geotextiles

Geogrids

Geonets

Geomembranes

Geocomposites

scussion of each type follows.

xtiles ar e either woven or nonwoven fabrics

of webbed f ibers that run in

ween t he fibers are the most important element.or voids must be l arge enough to allow

out invading particulates. The geotextiles also mustbe sufficiently strong to cover and reinforce theapertures, or openings, of the drainage materialsthey are meant to protect.

In nonwoven geotextiles the fibers a re much thinnerbut far more numerous. The various types areneedle-punched, resin-bond, and melt-bond. Allcontain a labyrinth of randomly oriented fibers thatcross one another so that there is no direct line offlow. The fabric must have enough open space toallow liquid to pass through, while simultaneouslyretaining any upstream movement of particles. Theneedle-punched nonwoven type is very commonlyused as a filter material.

Geogr ids a re very s t rong in t ransverse andlongitudinal directions, making them useful asreinforcing materials for either soil o r solid waste.Generally, they ar e used to steepen the side slopes ofinterior cells or exterior containment slopes of afacility. Recently they also have been used in theconstruction of ''piggyback'' landfills, i.e., landfills

built on top of existing landfills, to reinforce theupper landfill against differential settlements withinthe lower landfill.

Geonets ar e formed with inters ecting ribs made froma counter-rotating extruder. A typical geonet isabout 1M-inch thick from the top of the upper rib tothe bottom of the lower rib, yet the flow capability isapproximately equivalent to tha t of 12 inches of sandhaving a 0.01 cdsec permeability. (The proposedregulation wil l increase this value to 1 cm/sec, asmentioned earlier.) The rapid transmission rate isdue to clear flow paths in the geonets, as opposed toparticle obstructions in a granular soil material.

There a re two main concerns with geonets. First, thecrush strength at the rib's intersection must becapable of maintaining its structural stabili tywithout excessive deformation or creep. Second,adjacent mat eri als must be prevented from intrudinginto the rib apertures, cutting off or reducing flowrates.

Foamed geonets are relatively new products madewith a foaming agent that produces a thick geonetstructure (up to 1/2-inch) with very high flow rates.These improved flow rates result from the thickerproduct, but eventually the nitrogen gas in the ribvoids diffuses through the polymer structure, leavingbehind a structure with reduced thickness. Theresult over the long term is a solid rib geonetthickness equiva lent to other nonfoamed geonets.

The fourth type of geosynthetic is a geomembrane, orFML. It is the primary defense against escapingleachate and of crucial importance. FMLs are thefocus of Chapt er Three.

55



The final category of geosynthetics is dra inagegeocomposites. These are polymeric materials withbu i l t -up co lumns , nubs , cuspa t ions , or o t h e rdeformations that allow planar flow within theirstructure. A drainage geocomposite having 1-inchhigh columns can carry the flow .of a 4- o 5-inchdiameter pipe. Many products, however, have lowcru s h s t r en g t h s t h a t a r e i n ad eq u a t e fo r d eeplandfills or surface impoundments. They are useful,however, for surface water collector systems above

the closed facility where they only need to supportapproximately 4 feet of soil and constructionplacement equipment .

Design- by- function Con cepts

Whatever parameter of a specific material one isevaluating, a required value for the mat eria l must befound using a design model an d a n allowable valuefor the material must be deter.mined by a t es tmethod. The allowable value divided by the requiredvalue yields the design ratio (DR), or the resultingfactor of safety (FS). This design-by-function conceptis necessary to design and evaluate new materials

tha t are both feasible and safe for a variety ofsituations.

In evaluating drainage and filtration materials, a nallowable flow rate is divided by a required flow rateto obtain the design ratio or factor of safety accordingto th e equ ations below:

(a) For Drainage:

where DR = design ratio

q

Y = transmissivity

= flow rate per unit width

(b) for Fcltration:

DR = qallow/qreqd (3)

DR = Yallow/Yreqd (4)

or

where DR = design ratio

q

Y = permittivity

= flow rate per unit area

Transmiss iv i ty i s s imply the coeff ic ien t o fpermeability, or the hydraulic conductivity (k),within the plane of the material multiplied by thet h i ck n es s ( t ) of t h e m a t e r i a l . B e c a u s e t h ecompressibility of some polymeric materials is very

high, the thickness of the ma teri al needs to be takeninto account. Darcy's law, expressed by the equation

q = kiA, is used to calculate rate of flow, withtransmiss ivi ty equal to k t and i equal to thehydraulic gradient (see Figure 4-2):

Figure 4-2. Variables for calculating inplane flow rates(transmissivity).

q = kiA

= k(Ah/L) (w x t)

qlw = (kt)(Ah/L)

if 8 = kt

q/w = e(;)

where qlw = flow rate per unit width8 = transmissivity

Note that when i = 1.0, (q/w) = 8; otherwise it doesnot.

Wi th a l iquid flowing across the plane of thematerial, as in a geotextile filter, the permeabilityperpendicular to the plane can be divided by thethickness, t, to obtain a new value, permittivity (seeFigure 4-3 ) . In c rossp lane f low, t is i n t h edenominator; for planar flow it is in t he numerat or.Crossplane flow is expressed as:

q = kiA (6)

= k(Ah/t)A

q =(Wt)AhA

Y = (Ut) = (q/AhA)

where Y = permittivity

q/A = flow rate per uni t a rea (''flux'')

56



Aht

Figure 4-3. Variables for calculating crossplane flow rates

(permittivity).

4-1 shows some of the ASTM test methods and

lection systems. Test methods a re determinedD18, the Soil and Rock Committee of ASTM, andD35, the Committee on Geosynthetics.

llectors, geonet drains , sand filte rs,Figure 4-4 hows a cross

a primary leachate collection system withgeonet drain on the side slope leading into a gravel

a perforated pipe collector. A geotextile acts as aecting the geonet and s and acts as a filter

asFigure 4-4.

(G rav el) Dra inag e Design

s must:

Be 30 centimeters (12 inches) thick.

Have 0.01 c d s e c ( = 0.02 ft/min) permeability(hydraul ic conductivity).

Have a slope greater t han 2 percent.

Include perforated pipe.

Include a layer of filter soil.

0 Cover the bottom and side w a l l s of the landfill.

There are two ways to calculate the required flowrate, q, in granular soil drainage designs. One isbased on the above MTG values; the o ther is based onthe Mound Model (see Figure 4-5). Based on MTGvalues:

q = kiA

= (0.02) 0.02)(1 x 1)= 4 1 0 -4 ftslmin

(7)

Note that if MTG increases the required hydraulicconductivity of the drainage soil to 1 cmlsec, theabove flow ra te will be increased to 0.04 ftslmin.

In the Mound Model, the maximum height betweentwo perforated pipe unde rdra in systems is equal to:

tana+ 1 -- tan’a + c

2 C

wherec = qlk

k = permeability

q = inflow rate

The two unknowns in the equat ion are L , thedistance between pipes, and c, the amount ofleachate coming through the system. Using amaximum allowable head, hmax, of 1 foot, theequations ar e usually solved for L.

One method of determining t he value of c is using theWate r Balance Method:

PERC = P - W O - S T - A E T (9)

where PERC = percolation, i.e. the‘ liquidt h a t p e rm ea t e s t h e s o l i dwaste (gallacrelday).

P = precipitation €o r which themean monthly values aretypically used.

WO = surfacerunoff.

ST = soil moisture storage, i.e.,

moisture retained in the soila f t e r a g i v en am o u n t o faccumulated potential waterloss or gain ha s occurred.

AET = actual evapotranspirat ion,i.e., actual amount of waterloss during a given month.

57



Table 4-1. Test Methods and Standards

ASTM Test Desigiiatioii(or other) Used to Determine Material Value Used for

Perineability soil PLCR, LDCR2434

D24 16 Strength Underdrain pipe PLCR, LDCR

General specification HDPE pipe PLCR, LDCR405, F667

D4716 Transmissivity Geonet, geocomposite PLCR, LDCR

D4491 Permittivity Geotextile PLCR filter

D4751 Apparent opening size Geotextile PLCR filter

cw -0 22 15 a Gradient ratio Geotextile PLCR filter

GRI-GTlb Long-term flow Geotextile PLCR filter

a U S . Army Corps of Engineers Test Method.

bGeosynthetic Research Institute Test Method.

& aste

Perforated

Pipe

Figure 4-4. Cross section of primary leachate collection systems.

The range of percolation r ates in the United State s i15 to 36 incheslyear (1,100 to 2,700 gallacrelday(U.S. EPA, 1988).

The computer program Hydrologic Evalu atio nLandfill Performance Model (HELP) can also be usedto calculate c. HELP w a s developed to assist in

1 1 1 1 1 1 1 1 1 1 es t i m a t i n g t h e m ag n i t u d e o f wat e r -b a l an cecomponents and the height of water-saturated soiabove the barrier layers. HELP can be used withthree types of layers: vertical percolation, lateradrainage, and barrier soil l iner. By providingclimatological data for 184 cities throughout the

Inflow

Drainage Layer

///\W\V/A ///,W/A \//A,\ United Sta tes, HELP allows the user to incorporate

e L extended evaluation periods without having toassemble large quantities of data (Schroeder et al

Figure 4-5. Flow rate calculations: Mound Model. 1984).

58



A s landfills became higher, the

PV C pipes are commonly used, as are

as part of

s are :

.

.

.

Obtain the required flow value using knownpercolation and pipe spacing.

Obtain the required pipe size using the requiredflow and the maximum slope.

Check the pipe strength and obtain its ringdef lec t ion to determine tolerance againstcrushing.

ed using the curve in Figure 4-6. The amountleachate percolation at the particular site is

is th e

of flow rate and the bottom slope ofcan find the required diameter for the

Figure 4-7). Finally, the graphs in Figures and 4-9 show two ways to determine whether o r

Figure 4-8, the vert ical soilis located on the y-axis. The density of th e

is used to

so it will deform underis often

as the limiting deflection value for plastic pipe.Figure 4-9 the applied pressure on the pipe is

ion value is checked for its adequacy.

rainage Design

4 - 2 presents a compilation of currently

hemically resis tant to the leachate,

D 4716 as to allowable

y-function equation presented ear lier.

D 4716 flow test, the proposed collector crossas closely as possible. The

a,

3c

c_

E

-

% 100

c 1 2 0 i

Percolation, in inches per month

'Where b = width of area contribut ing to leachate collection pipe

Figure 4-6. Required capacity of leachate collection pipe

between a FML beneath and a geotextile above. Soil,perhaps simulatidg the waste, is placed above thegeotextile and th e load platen from the t est device isplaced above the soil. Applied normal stress istransmitted through the entire system. Then planarflow, at a constant hydraulic head, is initiated andthe flow ra te through the geonet is measured.

(after US. EPA, 983).

Figure 4-10 shows the flow rate "signatures" of a

geonet between two F M L s (upper curves) and thesame geonet with the cross section describedabove(1ower curves). The differences between the twos e t s o f c u r v e s r e p r e s e n t i n t r u s i o n of t h egeotextile/clay into the apertures of the geonet.Irrespective of the comparison in behavior, thecurves ar e necessary in obtaining a n allowable flowrate for the parti cula r geonet being designed.

The required flow rate can be calculated by threedifferent methods: ( 1 ) directly from minimumtechnology guidance, (2) u s i n g a n e q u a t i o ndeveloped in the design manua l, or (3)on the basis ofsurface water inflow rate. To be conservative, allthree calculations should be performed and theworst-case situation (e.g., tha t with t he highest flowrate) used €or the required flow rate. The variousequations to determine the required flow rate o rtransmissivity appear below:

1 . Geonet must be equivalent to MTG regulationsfor nat ura l materials:

59



G.P.M. C.F.S

P5 1000V,

0)

ai

"!30

= O 0.

Slope of Pipe in Feet per Thousand Feet

Figure 4-7. Sizing of leachate collection pipe ( U S E P A, 1983).

0 2 0.02 ftalmin-ft (10)

2. Based on estimated leachate inflow (Richardsonand Wyant , 1987):

Q L2(11)

4h + 2Lsinamax

3. Based on surface water inflow (U.S .EPA, 986):

Q = CIA 112)

where Q= surface water inflow

C = runoff coefficient

I = average runoff intensi ty

A = surface ar ea

Generally geonets result in high factors of safety or

design ratios, unless creep becomes a problem or ifadjacent materials intrude into the apertures.

Granular Soil (Sand) filter Design

There are th ree part s to an analysis of a sand filter tob e p l aced ab o v e d ra i n ag e g rav e l . T h e f i r s tdetermines whether or not the filter allows adequate

Pipe Flowing Full

on Manning's Equation n = 0.0

flow of liquids through it. The second evaluateswhether the void spaces ar e small enough to prevensolids being lost from the upstream materials. Thethird pa rt estimates the long-term clogging behavioof the filter.

Required in the design of granular soil (sand) filtermaterials is the particle-size distribution of thedrainage system and the particle-size distribution othe invading (or upstream) soils. The filter m ater iashould have its large and small size particlesintermedi ate between th e two extremes (see Figure4-11). Adequate flow and adequate retention are thetwo focused design factors, but perhaps the mosimportant is clogging. The equations for adequateflow and adequ ate retenti on are:

0 Adequate Flow: d ~ 5 ~(3 to 5) d15~,,. (13)

0 Adequate Retention: d1 5~ (3 to 5)d85,,f, (14)

There is no quantita tive method to assess soil filteclogging, although empirical guidelines are found ingeotechnical engineer ing references.

Geotextile filter Design

Geotextile filter design parallels sand filter designwith some modifications. The three elements o

60



12,000

10,000

8,000

6,000

4,000

2,000

0

Ring Deflection, AY/D (% ) = E Except as Noted

4-8. Vertical ring deflection versus vertical soil pressure for 18-inch corrugated polyethylene pipe in high pressureSoil

cell.

tion remain the same.

is assessed by compar ing thew a bl e p e r m i t t i v i t y w i t h t h e r e q u i r e d

design.

DR = ' Y a l l ow / yreqd (15)

Y al l ow = permittivity from ASTM TestD4491

q- = inflow rate pe r unit ar eaA

h,,, = 12 inches

The second part of the geotextile filter design isdetermining the opening size necessary for retainingthe upstream soil or particulates in the leachate. It iswell established that the 95% opening size is relatedto the parti cles to be retained in the following type ofrelationship

095 < fct. (d50,cu,DR) (16)

095 = 95% opening size of geotextile (U.S.Army Corps of Engineers CW 02215tes t method)

where

d50 = 50% size of upstream particles

CU= uniformity of the upstream particlesizes

61



4

0 3

-

:K>

'I 2a

1

0

E=- -Vertical Soil Strain E

for Native Soil @75% Densitv

i 1 B/D = 0-

5 10 15 20 25 30

Ring Deflection, A Y/D (YO)

Figure 4-9. The effect of trench geometry and pipe sizingon ring deflection (after Advanced Draining

Systems, Inc., 1988)

DR = relat ive densi ty of the upstreamparticles

Geotextile literature documents the relationshipfurther.

The 0 9 5 size of a geotextile in the equation is theopening size at which 5 percent of a given size glassbead passes through the fabric. This value must beless than the particle-size characteristics of theinvading materials. In the te st for the 0 9 5 size of thegeotextile, a sieve with a very coarse mesh in thebottom is used as a support. The geotextile is placed

on top of the mesh and is bonded to the insideso

thatthe glass beads used in the t est cannot escape aroundthe edges of the geotextile. This particular testdetermines the 0 9 5 value. To verify the factor ofsafety for particle retention in the geotextile filter,the particle-size distribution of retained soil iscompared to the allowable value using any of anumber of existing formulae .

The third consideration in geotextile design is long-ter m clogging. The test method t ha t probably will beadopted by ASTM is called the Gradi ent Ratio Test.It was originally formula ted by the U.S. Army Corpsof Engineers and is listed in CW 02215. In the test,

the hydraulic gradient of 1 inch of soil plus theunderlying geotextile is compared with th e hydraulicgradient d inches of soil. If the gradient ratio isless than 3, the geotextile probably will not clog. Ifthe gradient ratio is greater than 3, the geotextileprobably will clog. An al te rnate to this procedure is along-term column flow test that also is performed ina laboratory. The test models a given soil-to-fabricsystem at the anticipated hydraulic gradient. The

flow rate through the system is monitored. A long-term flow rate at a constant value indicates anequilibrium between the soil and the geotextilesystem. I f clogging occurs, the f l o w r a t e w i l lgradually decrease until it stops altogether.

Leachate Removal Systems

Figure 4-12 shows a low volume sump in which thedistance from the upper portion of the concretefooting to the lower portion is approximately 1 footOne foot is an important design number becauseEPA regulations specify a maximum leachate headof 1 foot. Low volume submersible sumps presentoperational problems, however. Since they run drymost of the time, there is a likelihood o f the i rburni ng out. For this reason, landfill operators preferto have sumps with depths between 3 and 5 feetinstead of 1 (Figure 4-13), even though the leachatelevel in a high volume sump will be greater than the1-foot max imum.

The leachate removal standpipe must be extended

through the entire landfill from liner to cover andthen through the cover i tself. It also must bemaintained for the entire post-closure care period of30 years or longer.

Leak Detection, Collection, and Remova(LDCR) Systems

The leak detection, collection, and removal system(LDCR) is located between t he pri mar y an dsecondary liners i n landfills, surface impoundmentsand waste piles. It can consist of eithe r granular soils(i.e., gravels) or geonets.

Granular Soil (G rave l) Drainag e DesignAs with the primary leachate collection systemabove the liner, leak detection systems betweenliners are designed by comparing allowable flowrates with required flow rates, The allowable flow isevaluated a s discussed in the section on granular soil(gravel) draina ge design for PLCR systems. Therequired flow is more difficult to estimate. This valuemight be as low as 1 gallacrelday o r many times thatamount. It is site specific and usually is a roughestimate. P ast designs have used 100 gallacreiday forthe required flow rat e. Data from field monitoring ofresponse action p lans (RAPS) will eventually furn ish

more realistic values. A pipe network for leachateremoval is required when using granular soils.

Geonet D rainage Design

For a geonet LDCR system, the flow rate for thegeonet is determined in the laboratory from ASTMD4716 tes t method, and the value is modified to meetsite-specific situations. The geonet flow rate design

62



4-2. Types and Physical Properties of Geonets (all are polyethylene)

0 Tex-Net (TN)

Poly-Net (PN)

Product Name

FX-2000 Geo-NetFX-2500 Geo-Ne1

FX-3000 Geo-Ne1

XB8110

XB8210

XB8310XB84 10

XB8315CN

TN-1001

TN-3001

TN-4001

TN-3001CN

PN- 1000

PN-2000

PN-3000

PN-4000

GSI Net 100

GSI Net 200GSI Net 300

Gundnet XL-1

Gundnet XL-3

Lotrak 8Lotrak 30

Lotrak 70

CE 1

CE 2

CE 3

CE 600

DN1-NS1100

DN3-NS1300

-NS1400

Roll Size, widthllenqth Thickness Approx. Apperture Size

Structure

extruded ribs

extruded ribs

extruded ribs

extruded ribsextruded ribs


extruded ribs

extruded ribs

extruded ribs

extruded ribs

extruded ribs

foamed, and

extruded ribs

extruded ribs

foamed, and

foamed, and


extruded ribs

extruded ribs

extruded mesh

extruded mesh

extruded mesh

extruded ribs

extruded ribs

extruded ribs

extruded ribs

extruded ribs

extruded ribs

extruded ribs

extruded ribs

extruded ribs

extruded ribs

f t m

7.51300 2.3191

7 51300 2.3191

7.51220 2.3167

6.91300 2.1191

6,91300 2 1/91

6.91300 2.11916.91220 2.1167

6.91300 2.1191

7.51300 2.3191

7.51300 2.3191

7.51300 2.3191

7.51300 2.3191

6.75i300 2.0191

6.751300 2.0191

6.751300 2.0191

6.751300 2.0191

6.21100 1.9130

6.21100 1.9130

6.61164 2.0150

6.61164 2.0150

6.611 64 2.0150

4.8166 1.5120

7.4182 2.2125

5.51100 1.67130.5

5.2198 1.6130

6.2198 1.9130

6.2/98 1.9130

7.4182 3.8125

mils

200

250

300

250

160

200300

200

250

200300

200

250

160

200

300

250

160200

250

200

120

200

290

250

200

160

160

220

150

200

mm

5 1

6 3

7 6

6 34 1

5 17 6

5 1

6 3

5 1

7 6

5 1

6 3

4 1

5 1

7 6

6 3

4 15 1

6 3

5 1

3 0

5 2

7 3

6 3

5 14 1

4 1

5 6

5 1

3 8

in. mm

0.3 x 0.3 8 x 8

0.3 x 0.4 8x10

8 x 8

0.35 x 0.35

0.25 x 0.25

0.3 x 0.3

9 x 9

6 x 6

0.3 x 0.3 8 x 8

0.3 x 0.4 9 x 9

6 x 6

0.35 x 0.35 a x i o0.25 x 0.25

0.3 x 0.3

0.3 x 0.3

0.3 x 0.3

1.2 x 1.2

2.8 x 2.8

0.3 x 0.25

0.3 x 0.35

0.3 x 0.25

0.3 x 0.25

0.3 x 0.3

0.3 x 0.3

0.3 x 0.3

8 x 8

8 x 8

8 x 930 x 27

70 x 70

8 x 6

9 x 9

8 x 6

8 x 6

8 x 8

8 x 8

8 x 8

is then determined in the same way as for theN o pipe network is needed.

concern when using geonets with a compositeary. liner .design is the effect of geotextile

. In composite primary l iner systems, theis placed immedia tely below a clay liner with

geotextile as an intermediate barrier. The design ofis important because clay particles

of a needle-punched

8 to 10 ounces per

ion in every detai l.

PA specifies that the minimum detection time forof a

LDCR system is less than 24 hours. Response timecalculations are based on velocity in the geonetandlor granular soil drainage layer. Darcy's law isused to calculate flow velocity in the geonet, and a"true" velocity must be used for granul ar soil.

Figure 4-15 shows the response time calculation for aleachate leak through a primary liner traveling 40feet through the geonet on the side wall and 20 feetthrough the sand a t the bottom. The res ultingresponse times are 1.5 hours in the geonet and 6.2hours in the soil; giving a total response time of 7.7

hours.

The travel time in a geonet is very short; so a 24-hourresponse time can easily be achieved. With granularsoils, the t rave l time will be much longer.

Leak Detec tion Rem oval Systems

Leak detection removal systems require monitoring,sampling, and leachate removal. Any leachate that

63



frE1

mmv

040 5,000 10,000 15,000 20,000

Normal Stress (Ibsisq. ft)

(a) FML - Geonet - FM L Composite

10 -

a -

6 -

4 -

0 5,000 10,000 15,000 20,000

Normal Stress (Ibsisq. ft)

(b) FML - Geonet - Geotextile - Clay Soil Composite

. i = o 5

1=0.125

= I = O 0625

I = 0.25

e i=0.125

i=O.O625

0 I = 0.03125

Figure 4-10. Flow rate curves for geonets in different composite situations.

64



Gravel Sand Sllt Clay

Particle Size (log)

4-1 1. Design basedon particle-sized curves.

liner system and enters the

n removal system consists of aely l arge diameter pipe run ning down the sideetween the primary and secondary liners to the

point (s ump) in t he LDCR. The pipe m us tt the top. A submersible

ning" of the quan tit y of fluid coming into the(see Figure 4-16). The choice of monitoring

and retrieval pump depends on the quantity ofleachate being removed.

An alternate system, one based on gravity, requirespenetration of both the FML and clay components ofthe secondary composite liner system as shown inFigure 4-17. I t also requires a monitoring andcollection manhole on the opposite side of the landfillcell (see Figure 4-18). The manhole and connectingpipe, however, become an underground storage tanktha t needs its own Secondary containment and leakdetection systems.

Surface Water Collection and Removal

(SWCR) Systems

The third pa rt of liquids management is the surfacewate r collection and removal system (SWCR). It isplaced on top of the completed facility and above thecover F M L . The rainwater and snowmelt thatpercolate through the top soil and vegetative covermust be removed to a proper upper drainage system.Figure 4-19 illustrates the major components of a

surface water collector system. The design quant ityfor the amount of fluid draining into the surfacewater collector system can be determined by either awater balance method o r the computer programH E L P discussed previously (see Figure 4-20).

Concrete Base

I . ( Gravel 'J

4-12. Leachate removal system with a low volumesump.

65



Figure 4-13. Leachate removal system with a high volume

sump.

Clay Beneath Primary FM L -Possible IntrusionGeotextile I ~ / iocations

Geonet4 x p~ G S-FML

Clay Beneath Secondary FML-

Figure 4-1 4. Geotextile used as barrier material to prevent

extrusionof upper clay into geonet drain.

Surface water drainage systems can be composed ofgranular soils, geonets, o r geocomposites, but themajority of drainage systems use gra nul ar soil. This

is particularly tr ue in frost, regions wher e it isnecessary to have 3 to 6 feet of soil above the FML tosatisfy the requirements for frost penetration. Insuch cases, 1 foot of gra nul ar soil thickness can serveas the surface water collector. If good drainagematerials ar e not available, if the si t e is tooextensive , or if na t ura l mat er ia l s would addundesired thickness, a geonet or geocomposite can beused. The advantage of drainage geocomposites istheir higher flow rate capabilities over geonets orgranula r so i l s . Tab le 4-3 l i s t s a n u m b e r o fgeocomposites that can be used for drainage sys tems.All of these systems have polymer cores protected bya geotextile filter. Although many of the polymers

cannot withstand aggressive leachates, this is not anissue in a surface drainage collector where the onlycontact is with water. The crush strengths of thegeocomposites ar e generally lower t han for geonets,but that too is not a problem in a surface watercollector. The heaviest load the geocomposite wouldbe requi red to suppor t p robably would beconstruction equipment used to place the cover soiland vegetation on the closed facility.

The design for the surface water collector system isdetermined by an allowable flow rate divided by a

required flow rate. Allowable rat es for geocompositesar e determined experimentally by exactly the s ame

method as for geonets. Figure 4-21 shows the flowrate behavior for selected drainage geocomposites.

The specific cross section used in the test procedureshould replicate the intended design as closely as

possible. For the required flow rate, Darcy's l a w or

€ f E L P can be used. Then the design-by-functionconcept is used to determine th e design ratio ( D R ) , or

factor of safety (FS).

allowable flow rate

required flow rateDR = FS =

Gas Collector and Removal SystemsDegradation of solid waste materials in a landfillproceeds from aerobic to anaerobic decompositionvery quickly, thereby generating gases that collect

beneath the closure FML. Almost 98 percent of t hegas produced is either carbon dioxide ((202) o rmethane (CH4). Because COz is heavier th an air , itwill move downward and be removed with theleachate. However, CH4, representing about 50percent of the generated gas, is lighter than airand,therefore, wil l move upward and collect at thebottom of the facility's "impermeable" FML. If thegas is not removed, it w i l l produce a buildup ofpressure on the FML from beneath.

In gas collector systems, either a granu lar soil layeror a needle-punched nonwoven geotextile is placeddirectly beneath th e FML or clay of a composite cap

system. Gas compatibility and air transmissiv ity ar ethe design factors tha t must be considered. Methane,the most predominant gas, should be compatiblewith most types of geotextiles including polyester,polypropylene, and polyethylene.

The thickness design should be based on gastransmissivity tests. Since water has a viscosity of1,000 to 10,000 times t ha t of gas, gallowfor gas flowshould compare very favorably with the results of awater transmissivity test. A s an example, Figure 4-22 shows air transmissivity versus normal stress fora 1 2 oz/yd2 needle-punched nonwoven geotextile.A l t e rn a t i v el y , o n e c o u l d l oo k d i r e c t l y a t

permeability coefficients where geotext ile ai r flow isseveral orders of magnititude greater th an the MTG-required values as shown in Figure 4-23. In the testmethod, the geotextile specimen fits underneath aload bonnet. Then the load, equivalent to the coversoil, is added and gas is brought to the inside of thegeotextile. The gas flows through the geotextile andinto a shroud th at goes on the outside of the flangesand registers on an air meter. The resulting applied

66



4-15. Example problem for calculationof primary liner leak response time.

if necessary, converted into gas

d the design ratio, or factor of safety.

at least that long to avoid gas pressure on the

underside of the cover.

Gas generation might also cause problems in"piggyback" landfills, landfills that have been buil ton top of one another. It is still unknown whathappens to gas generated in an old landfill after anew liner is placed on top of it. To minimizeproblems, the old landfill should have a uniform

ation occurs over a period of 70 to 90 years,gas collector and removal systems must work for

67



Leakage

Removal

I I ,

Figure 4-16.

v

Submersible Pump

Secondary leak detection removal system via pumping between liners and penetrationof pimary liner.

\ equires

Penetration

Wi-See p Collars Of S-FMLand Clay

Figure 4- 17. Secondary leak detection removal system via gravity monitoring via penetrationof secondary liner.

68



Note: Manhole will be equipped with discharge line to

leachate removal system or with discharge pump

Figure 4-18. Monitoring and collection manhole(E . C. Jordan, 1984).

[Layer 1

Vegetative

Layer

. Drainage

Layer

Low Permeability

Layer

Manhole Frame and Cover

with Vented Lid

Granular Backfill

4 ' LD Manhole

Point of Continuous or' Intermittent Monitoring

t2 12"

FML.

Tt 0 mils in

thickness)

2 24"

Iompacted

Soil Layer

Vegetation or Other

Erosion Control Materialat and Above Surface

Top Soil for RootGrowth

Remove Infiltrating

Water

Increases Efficiency

of Drainage Layer and

Minimizes Infiltration

into UnitI

Figure 4-19. Surface Water Collection and Removal (SWCR) system.

69



100

3zEE.

101

0

6ma

c

0

85 l o :

z

-LL

c

0

0 1

700

600

-8 500-e

Q

0

Lo

5

-

a, 4oa

2.

U

a,

8 30C>a,0

??

g 2oc2a

1oc

OO (

1 1 1

Normal Pressure Iblin*/lb/ft*

, 10

1O i l 440

1512160

2012880 {

,, ,,,:

,,,&?-r, i:,

-,,a%': ; 5 '

-

-

1 I l l I !

0.01 0 10 1 0 10 0

/

(/

//

Thickness of Mat, d(in)

L' " I 1 1 I

0 25 50 75 10

(b)

Residual Thickness, n(Y0)

Figure 4-21. Flow rate behavior of selected geocomposite drainage systems.(a ) Mirardrain6000 at hydraulic gradients of 0.01 to 1.0

(b) Enkadrain at hydraulic gradient of 1.0

72



I

g 3 0cE

E

2'.2 2 0 -

$%

C

25 0 t00 -

M - u, = 1 0 PSI- 1

u, = 1 0 PSI

u, = 3 .5 psi

L-a

1 0 -

1

1

u, = 0 1 PSI

u, = 0 1 PSI

C I -- m - - - -

I I I I

- -

7 3

0 500 1000 1500 2000 2500 b 1 1 I 1 1



Boot Seal at FMC

Cover SoilRiser4

Operational CoverFlange Seal at FMC

7 iner Place Vent Higher than Maximum

at Over-Flow Conditions

Two-inch Minimum

Air/Gas Vent

Drainage Composite

Liquid 1eve1

Openings in Vent to be Higher than

Top of Berm or Overflow Liquid Levelr T A i r l G a s Vent Assembly

Approx. Six inches

Drainage Composit

6Cjeotextile+T'---Hr1pprox. Six inches

Wind Cowl Detaile

-oncrete

Vent to Liner

Figure 4-24. Miscellaneous details of a gas collectorsystem.

74



5. SECURING A COMPLETED LANDFILL

IntroductionThis chapter describes the elements in a closure or

cap system of a completed landfill, including flexiblemembrane caps, surface water collection andremoval systems, gas control layers, biotic ba rrie rs,and vegeta t ive top covers . I t a lso discussesinfiltration, erosion control, and long-term aestheticconcerns associated with securing a completed

landfill.

Figure 5-1 shows a typical landfill profile designed tomeet EPA's proposed minimum technology guidance(MTG) requirements . The upper subprofile comprisesthe cap, or cover, and includes the required 2-footvegetative top cover, 1-foot lateral drainage layer,and low permeability cap of barr ier soil (clay), whichmust be more than 2 feet thick. This three-tiersystem also includes a n optional flexible membranecap and an optional gas control layer. The guidanceoriginally required a 20-mil thick flexible membranecap, but EPA currently is proposing a 40 milminimum.

Flexible Membrane Caps

Flexible membrane caps (FMCs) are placed over thelow permeable clay cap and beneath the surfacewater collection and removal (SWCR) system. FMCsfunction primarily in keeping surface water off thelandfill and increasing the efficiency of the drainagelayer. EPA leaves operators with the option ofchoosing the synthetic material for the FMC thatwill be most effective for site-specific conditions. Inselecting materials, operators should keep in mindseveral distinctions between flexible membraneliners (FMLs) and FMCs. Unlike a FML, a FMCusually is not exposed to leachate, so chemicalcompatibility is not a n issue. Membrane caps al sohave low normal s t r esse s act in g on them incomparison with FMLs, which generally carry theweight of the landfill. An advantage FMCs have over

liners is tha t they ar e much easier to repair , becausetheir proximity to the surface of the facility makesthem more accessible. FMCs w i l l , however, be

subject to greater s t ra ins than FMLs due tosettlement of the waste.

Surface Water Collection and Removal(SWCR) SystemsThe SWCR system is built on top of the flexiblemembrane cap. The purpose of the SWCR system isto prevent infiltration of surface water into the

landfill by containing and systematically removingany liquid that collects within it. Actual designlevels of surface water infiltration into the drai nagelayer can be calculated using the water balanceequation or the Hydrologic Evaluation of LandfillPerformance ( HELP) model. (A more detail eddiscussion of HELP is contained in Chapter Four.)Figure 5-2 shows the results of two verificationstudies of the HE LP model published by EPA.

Errors in grading the perimeter of the cap oftenintegrates (or cross-connects) the SWCR system withthe secondary leak detection and removal system,r e s u l t i ng i n a s i gn i f i c a n t a m oun t o f w a t e r

infiltrating the secondary detection system. Thissituation should be remedied as soon as possible if itoccurs. Infiltration of surface water is a particularconcern in nuclear and hazardous waste facilities,where gas vent stacks are found. A containmentsystem should be designed to prevent water fromentering the system through these vents.

In designing a SWCR system above a FMC, threeissues must be considered: (1) cover stability, (2)puncture resistance , and (3) the abi lity of the closuresystem to withstand considerable stresses due to theimpact of settlement. Figure 5-3 il lustrates the

effects of these phenomena.

Cover Stability

The stability of the FMC supporting the SWCR

system can be affected by the materials used toconstruct the drainage layer and by the slope of thesite. In some new facilities, the drainage layer is ageonet placed on top of the flexible membrane cap,

75



Figure 5-1. Typical geosynthetic cell profile.

with the coefficient of friction between those two

elements being as low as 8 to 10 degrees. Such lowfriction could allow the cover to slide. One facility a tthe Meadowlands in New Jersey is constructed on ahigh mound having side slopes steeper than 2:l. Inorder to ensure adhesion of the membrane to the side

slopes of the facility, a nonwoven geotextile w a sbonded to both sides of the FMC. Figures 5-4 and 5-5give example problems that evaluate the slidingstability of a SWCR system in ter ms of sh ear

capacities and tensile stress.

Puncture Resista nc e

Flexible membrane caps must resist penetration byconstruction equipment, rocks, roots, and otherna t u r a l phe nom e na . T r a f f i c by ope r a t i ona l

equipment can cause serious tearing. A geotextileplaced on top of or beneath a membrane increases its

punc ture resistance by three or four times. F igure 5-6 shows the results of puncture tests on severalcommon geotex t i l e /membrane combina t ions .

Remember, however, that a geotexti le placedberieath the FM C and the clay layer will destroy thecomposite action between the two. This will lead toincreased infiltration through penetrations in theFMC.

Impact of Settlement

The impact of settlement is a major concern in thedesign of the SWCR system. A number of facilitieshave settled 6 feet in a single year, and 40 feet ormore over a period of yea rs. The Meadowlands sit e in

New Jersey , for example, w a s built a t a height of 95feet, settled to 40 feet, and then was rebuilt to 135feet. Uniform settlement can actually be beneficial

by compressing the length of the FMC and reducingtensile strains. However, if waste does not settleuniformly it can be caused by interior berms that

separa te waste cells.

In one current closure site in California, a waste

transfer facility with an 18-foot wall is being builtwithin a 30-foot trench on top of a 130-foot highlandfill. Th e waste transfe r facility will sett le fasterthan the adjacent ar ea, causing tension a t the edge ofthe trench. Electronic extensometers ar e proposed a tthe tension points to check cracking strains in theclay cap and FMC.

Settlem ents can be estimated, although the marginfor error is large. Secure commercial hazardouswaste landfills have t he sm allest displacement, lessthan 1.5 percent. Displacements a t new larger solidwaste landfills can be estimated a t 15 percent, while

older, unregulated facilities with mixed wastes havesettlements of up to 50 percent. Figure 5-7 gives anexample problem showing how to verify thedurability of a FM C under long-term settlementcompression.

Gas Control LayerGas collector systems ar e installed directly beneath

the low permeability clay cap in a hazardous wastelandfil l . Landfil ls dedicated to receiving onlyhazardous wastes are relatively new and gas hasnever been detected in these systems. It may take 40years o r more for gas to develop in a closed securehazardous waste landfill facility. Because the long-

term effects of gas generation are not known, and

76



Placing FMC at edge of cap.

costs are minimal, EPA strongly recommends the

use of gas collector systems.

Figure 5-8 shows detail s from a gas vent pipe system.The two details at the left of the illustration showcloseups of the boot seal and flange seals locateddirectly at the interface of the SWCR system withthe flexible membrane cap. To keep the ventoperating properly, the slope of the closure systemshould never be less tha n 2 percent; 5 to 7 percent ispreferable. A potential problem with gas collector

systems is that a gas venting pipe, if not properlymaintained, can allow surface water to drain directly

into the landfill waste.

Figure 5-9 illu strat es two moisture control options ingas collector systems. Gas collector systems will

tolerate a large amount of moisture before airtransmissivity is affected. Figure 5-10 shows air andwater transmissivity in a needle-punched nonwoven

geotextile. Condensates from the gas collector layerthat form beneath the clay and flexible membrane

77



Leqendc- 6? - elp Simulation

c5 5 1 - ield Measurement

2 41

2 j

Cover Stability

Puncture Resistance

= 40

6

c!2n

7 10

2-)

0 0~~~55an Jan Jan Jan Jan

1973 1974 1975 1976 1977

Date

Figure 5-2. Cumulative comparison of HELP simulation and

field measurements, University of Wisconsin,

Madison, uncovered cell.

4 Impact of Settlement

Figure 5-3. SWCR systems considerations.

cap also can be tak en back into the waste, since mosthazardous wastes are deposited very dry.

Biotic BarriersA biotic barrie r is a gravel and rock layer designed toprevent the intrusion of burrowing animals into thelandfill area . This protection is primarily necessaryaround the cap but, in some cases, may also beneeded at the bottom of the liner. Animals cannot

generally penetrate a FMC, but they can widen anexisting hole or tear the material where it haswrinkled.

Figure 5-11 shows the gravel filter and cobblestonecomponents of the biotic barrie r and the ir placementin the landfill system . The proposed 1- met erthickness for a biotic barrier should effectivelyprevent penetration by all but the smallest insects.Note that the biotic barrier also serves as the surfacewater collectioddrainage layer. Biotic barr iers used

in nuclear caps may be up to 1Cfeet thick with rockseveral feet in diameter. These barriers ar e designeto prevent disruption of the landfill by humans bothnow and in the future.

Vegetative LayerThe top layer in the landfill profile is the vegetativ

layer. In the short term, this layer prevents wind anwater erosion, minimizes the percolation of surfacw a te r i n to t h e w as t e l ay e r , an d max imizeevapotranspiration, the loss of water from soil byevaporation and transpiration. The vegetative laye

also functions in the long term to enhance aestheticand to promote a self-sustaining ecosystem on top othe landfill. The latter is of primary importanc

because facilities may not be maintained for a

78



Meadowlands test of slip-resistant FMC.

indefinite period of time by ei ther governmen t orindustry.

Erosion can seriously affect a landfill closure bydisrupting the functioning of drainage layers andsurface water and leachate collection and removalsystems. Heavy erosion could lead to the exposure ofthe waste itself. For this reason, it is important topredict the amount of erosion tha t will occur a t a si teand reinforce the facility accordingly. The UniversalSoil Loss Equation shown below can be used to

determine soil loss from water erosion:

X = RKSLCP

where X = soil loss

R = rainfall erosion index

K = soil erodibility index

S = slope gradient factor

L = slope length factor

C = crop management factor

P = erosion control practice

Figure 5-12 can be used to find th e soil-loss ratio dueto the slope of the site as used in the Universal SoilLoss Equation. Loss from wind erosion can bedetermined by the following equation:

where X' = ann ual wind erosion

I' = field roughness factor

K ' = soil erodibility index

79



C' = climate factor

L' = field length factor

V' = vegetative cover factor

There a re many problems in main ta in ing anagricu ltural layer on top of a landfill site, especiallyin arid o r semiarid regions. An agricultural layer

built on a surface water collection and removalsystem composed of well-drained stone and synthetic

material may have trouble supporting crops of anykind because the soil moisture is removed. In aridregions, a continuous sprinkler system may beneeded to mainta in growth on top of the cap, even ifthe soil is sufficiently deep and fertile. A final

problem involves landfills built on slopes greaterthan 3 : l . Equipment neces sa ry to p lan t andmaintain crops cannot operate on steeper slopes.

(Example No. 5. 2

Figure 5-4. Shear failure for surface water collection and removal system.

Operators should contact their local agriculturale x t e n s i o n a g e n t or S t a t e D e p a r t m e n t ofTransportation to find out what kinds of vegetation

will grow under the conditions at the site. Theimpact of the SWCR system on the soil layer alsoshould be studied before vegetation is chosen. Nativegrasses usually are the best choice because theya l re a d y a r e a d a p t e d to t h e s u r r o u n d i n genvironment. Sometimes vegetation can overcomeadverse conditions, however. At one site in the NewJersey Meadowlands, plants responded to excesssurface water by anchoring to the underlying waste

through holes in a FMC, cre ati ng a sturdy bondbetween surface plants an d underlying material .

For sites on very arid land or on steep slopes, anarmoring system, or hardened cap, may be moreeffective than a vegetative layer for secu rin g alandfi l l . Operators should not depend on anagric ultur al layer €or protection in ar eas w here

vegetation cannot survive. Many States a l low

80



lesign Ratio:

D R ) 5 .o

[Example No. 5.3

References:

Figure 5-5. Tensile stress for a surface water collection and removal System.

asphalt caps as an alternative to vegetative covers.Some closures at industrial sites have involvedconstructing hardened cap "parking lots" on top ofthe cap. membrane and clay layers. A chip seal layerover the asphalt prevents ultraviolet degrada tion ofthe pavement. These caps, however, need to bemaintained and resealed every 5 years. At somesites, a fabric incorporated into the top of th e asphaltminimizes cracking and water intrusion.

Other Considera tions

Filter layers, frost penetration, and cap-liner

connections are other factors to consider in designingthe closure system for a hazardous waste landfill.Before using geotextiles for filter layers in closures,one should conduct pres sure test s and clogging testson the material . Freeze- thaw cycles probably havelittle effect on membranes, but their impact on clayis stil l not known. Because of th is lack of knowledge,membrane and clay layers should be placed below

the frost penetration layer. Figure 5-13 hows frostpenetration depths in inches for the continental

United St ates. Finally, a cap membrane should notbe welded to the primary flexible membrane liner(see Figure 5-14).Differential settlem ent in the capcan put tension on the cap membrane. In such asituation, the seam could separ ate and increase thepotential for integration of the surface wate rcollection system into the leak detection sys tem.

81



. -

65,

60,

55

50,

>F 35

6 osy Geotextile 30

25

20

15

10

5I eomembrane AloneEDPM.75 mm PUC.75 mm 0 Geotextile Both Sides EDPM.75 mm PUC.75 mm

1500'

1350-90

m n

CPE.75 mm HDPE.75 mm 6 Geotextile Front CPE.75 mm HDPE.75 mm

IGeomembrane Alone0Geotextile Both Sides

PGeotextile Front

B Geotextlle Back

1200 55

12 osy Geotextile

EDPM.75 mm PuC .75 mmHDPE.75 mmPE.75 mm HDPE.75 mrr CPE.75 mm

60

50

18 osy Geotextile

EDPM.75 mm PUC.75 mmCPE.75 mm HDPE.75 mm

(Koerner, 1986)1 N = .225 Ib.

" EDPM.75 mm PUC.75 mmCPE.75 mm HDPE.75 mm

1 J = ,738 ft-lb

a. Puncture Resistance b. Imoact Resistance

Figure 5-6. Puncture and impact resistance of common FMLs.

82



I t l T L t M f N l R A TIO. S / Z LD I 0.2

DQ: '71 6 9 OK SET T L EM f N T RATIO, 5/21

7

[Example No.5.4

Figure 5-7. The effects of settlement on a flexible membrane Cap.

83



Boot Seal at FMC

V 111 A\ VI I1, I” - 4 {L J \ ” L

Cover SoilRiser4

Flange Seal at FMC

Figure 5-8. Details of a gas vent pipe system.

I I I \ -

/- a

.

1 Compacted sol1

. . . . . . . ..___.

//A Y/A‘y//. . . .

- -- . . . . . .-

_ - - - _ -_ - - _. . .

. - .. .. ... . . .1 . .FPerforated.Plpe _.: .:. 1 G L ~ent . . ... . . . . . . . . . . . . . . - . . . .

Ooerational Cover

84



UGas Well

Condensate Drain to Collector

U

\Gas Flow

Uas Well

Condensate Drain to Well

- as Flow.- - Water Flow

Moisture Control

(after Rovers, et ai)

Figure 5-9. Water traps in a gas collector system.

100%r\ I I I I I1 00%

(Koerner, et al, 1984)

I I I I0

0 0.2 0.4 0.6 0.8 1 o

Normalized Pressure Ratio

Figure 5-10. Air and water transmissivity in a needle-punched

nonwoven geotextile.

i- Vegetation

Topsoil(60 cm)

Gravel Filter(30 cm)

Cobblestone(70 cm)

1I

BioticBarrier

1Protctive Laver

FMC

Compacted Soil(90 cm)

Gas Vent

(30 m)

Waste

Figure 5-1 1. Optional biotic barrier layer.

85



b

>- > >r - -

-11

Pc

aainn

0

0 100 200 300 400 500 600 700 800

Slope Length (Feet)

Figure 5-12. Soil erosion due to slope.

86



kFigure 5-13. Regional depth of frost penetration in inches.

Figure 5-14. Geosynthetic cell profile with extrusion welds at FML and FMC junctures.

87



6. Construction, Quality Assurance, and Control:Construction of Clay Liners

IntroductionThis chapter focuses on construction cri teri a for clayl ine rs , inc luding impor tan t va r i ab les in so i lcompaction, excavation and placement of liner

materials, and protection of liners after construction.The chapter concludes wi th a discu ss ion o fconstruction quality assurance, and of test tills and

their incorporation into the design and constructionquality assurance plan for liners.

Compaction VariablesThe most important variables in the construction of

soil liners are the compaction variables: soil watercontent, type of compaction, compactive effort, size ofsoil clods, and bonding between lifts. Of these

variables, soil water content is the most critical

parameter.

Soil Water ContentFigure 6-1 shows the influence of molding water

content (moisture content of the soil at the time ofmolding o r compaction) on hydraulic conductivity ofthe soil . The lower half of th e dia gr am is acompaction curve and shows the relationshipbetween dry un it weight, or dry density of the soil,and water content of the soil. A water conten t calledthe optimum moisture content is related to a peakvalue of dry density, called a maximum dry density.Maximum dry density is achieved at the optimum

moisture content.

T he s m a l le s t h yd r a u l i c c onduc t i v it y o f t h ecompacted clay soil usually occurs when the soil ismolded at a moisture content slightly higher than

the optimum moisture content. That minimumhydraulic conductivity value can occur anywhere inthe range of 1 to 7 percent wet of optimum watercontent. Ideally, the liner should be constructedwhen the water content of the soil is wet of optimum.

Uncompacted clay soils th at are dry of their optimumwater content contain dry hard clods that are noteasily broken down during compaction. After

HydraulicConductivity

Dry Un i t

Weight

-Molding Water Content

Figure 6-1. Hydraulic conductivity and dry unit weight as afunction of molding water content.

compaction, large, highly permeable pores are leftbetween the clods. In contrast, the clods in wetuncompacted soi l a r e s o f t a nd w e a k . U ponc om pa c ti on, t he c l ods a r e r e m o l de d i n t o ahomogeneous relatively impermeable mass of soil.

Low hydraulic conductivity is the single mostimportant factor in constructing soil liners. In orde r

to achieve that low value i n compacted soil, the large

voids or pores between th e clods must be destroyed.Soils are compacted while wet because the clods canbest be broken down in tha t condition.

Type of Compaction

The method used to compact the soil is anotherimpor tan t fac tor i n ach iev ing low h ydraul i cconductivity. Stat ic compaction is a method by whichsoil packed in a mold is squeezed with a piston to

89



Kneading CompactionI !

compress the soil. In kneading compaction, a probe o rpie-shaped metal piece is pushed repeatedly into thesoil. The kneading action remolds the soil much l ikekneading bread dough. The kneading method isgenerally more successful in breaking down clodsthan is the st atic compacting method (see Figure 6-21.

r - 74i

Molding w (% )

Figure 6-2. Efficiencies of kneading compaction and staticcompaction.

The best type of field compaction equipment is adru m roller (called a sheepsfoot roller) with rods, orfeet, sticking out from the d rum that penetra te thesoil, remolding i t and destroying the clods.

Compactive Effort

A th i rd compac t ion va r i ab le to cons ide r i scompactive effort. The lower half of the diagram inFigure 6-3 hows that increased compactive effortresu lts in increased maximum density of the soil. In

general, increased compactive effort also reduces

hydraulic conductivity. For this reason, i t isadvantageous to use heavy compaction equipmentwhen building the soil liner.

Two samples of soils with the same water contentand similar densities can have vastly differenthydraulic conductivities when compacted withdifferent energies. The ex tra compaction energy may

12 16 20 24

Molding Water Content ( Y G )

Figure 6-3. Effects of compactive effort on maximum density

and hydraulic conductivity.

not make the soil more dense, but it breaks up theclods and molds the m together more thoroughly.

The compaction equipment also must pass over the

soil liner a sufficient number of times to maximizecompaction. Generally, 5 to 20 passes of theequipment over a given lift of soil ensures that theliner has been compacted properly.

A set of data for clay with a plasticity index of 41percent used in a tria l pad in Houston illustrate s twocommonly used compaction methods. Figure 6-shows the significantly different compaction curvesproduced by standard Proctor and modified Proctorcompaction procedures. The modified Proctorcompaction technique uses about five times morecompaction energy tha n the standard Proctor.

Figure 6-4 hows the hydraulic conductivities of thesoil molded a t 12 percent water content to be lessthan 10-10 c d s e c , using the modified Proctor, and10-3 d s e c , using the standard Proctor. The differentlevels of compaction energy produced a seven order ofmagnitude difference in hydraulic conductivity.Apparently, modified Proctor compaction provided

enough energy to destroy the clods and produce low

90



hydraulic conductivity, whereas standard Proctorcompaction did not .

Figure 6-4 also shows that at 20 percent moldingwater content, the soil's water content and densityare virtually the same when packed with eithermodified or standard Proctor equipment. ModifiedProctor compaction, however, still gave one order of

'magnitude lower hydraulic conductivity. In thiscase, addit ional compaction energy produced

significantly lower hydraulic conductivity withoutproducing g reater density.

130'oisture content between the soil that w a s crushedand passed through a No. 4 sieve and the soil thatmerely w a s passed through a 0.75-inch sieve. The

impl icat ion for laboratory tes t ing is th at theconditions of compaction in the field must be

simulated as closely as possible in the laborato ry toensure reliable results.

125

90 I : I

2.

0

U.....

s .."0-8

Mod. Proctor

I I I 15 10 15 20 25

Molding Water Content ("/o)

Figure 6-4. Effects of modified Proctor versus standard

. Proctor compaction procedures.

Size of ClodsAnother factor th at affects hydraulic conductivity isthe size of soil clods. Figure 6-5 shows the result s ofprocessing the same highly plastic soil from aHouston site in two ways. In one, the soil w a s passedthrough the openings of a 0.75-inch sieve, and in theother, the soil was ground and crushed to passthrough a 0.2-inch (No. 4) sieve. Most geotechnical

l a b o ra t o ri e s p e rf o r m i n g s t a n d a r d P r o c t o rcompaction tests first air dry, cru sh, and pulverizethe soil to pass through a No. 4 sieve, then moistenthe soil to various w ater content s before compaction.

Figure 6-5 shows a 3 percent difference in optimum

-115-

EP

PcC

2 105

PD

95

a5u0 15 20 25

Molding Water Content ("/o)

Figure 6-5. Effects of clod size on hydraulic conductivity.

Table 6- 1 summarizes data concerning the influence

of clod size on hydraulic conductivity for the same

soil. At a molding water content of 1 2 per cen t,hydraulic conductivity of the sample with the

smaller (0.2-inch) clods is abou t four orde rs ofmagnitude lower than the hydraulic conductivity forthe sample with la rge r (0.75-inch) clods. Apparen tly,

at 12 percent water conten t, the clods are strong, butwhen reduced in size, become weak enough to be

broken down by t he compaction process. For the wetsoils with a molding water content of 20 percent, clodsize appears to have little influence on hydraulicconductivity, because the soft, wet clods are easilyremolded and packed more t ightly together . Clod sizethen is a n important factor in dry, hard soil, but lessimportant in wet, soft soil, where t he clods ar e easilyremolded.

91



Table 6-1. Influence of Clod Size on Hydraulic Conductivity

Hydraulic Conductivity (cmis)Molding

w c. (96) 0.2-ln. Clods 0.75-in. Clods

12 2 x 10-8 4 x 1 0 - 4

16 2 x 10-9 1 x 10-3

18 1 x 10-9 a x 10-10

20 2 x 10-9 7 x 10-10

Bonding betw een Lifts

A final important compaction variable is the extentof bonding between lifts of soil. Eliminating highlypermeable zones between lifts cuts off liquid flowfrom one lift to another. If defects in one lift can bemade discontinuous with defects in another lift, thenhydraulic continuity between lifts can be destroyed(see Figure 6-6).

1L Lift 4

r

Lift 3

Lift 1f

1Figure 6-6. Conductivity between lifts.

Compaction equipment with roilers that knead orremold the soil can help destroy hydraul ic connectionbetween defects. Footed rollers are the best kind to

use for this purpose. The two kinds of footed rollersgenerally used for soil compaction are those withlong, thin feet that fully penetrate the soil and thosewith feet that only partially penetrate the soil (seeFigure 6-7). The roller with fully penetrating feet,typically called a sheepsfoot roller, has shafts about 9inches long. Because the lift thickness of a clay lineris typically 8 to 9 inches before compaction and 6inches after compaction, the shaft of the sheepsfoot

roller can push through a n enti re lift of soil.

The fully penetrating feet go all the way through theloose lift of soil to compact the bottom of the liftdirectly into the top of the previously compacted lift,blending the new lift in with the old. The partlypenetrating foot, or padfoot, cannot blend the new lift

Fully-Penetrating Feet:

Loose Liftof soil

7 I / / / / /

Compacted Lift/ I / / / /

Partly-Penetrating Feet:

of sot1

7- / / / / /

Compacted Lift/ , , / ,

Figure 6-7. Two kinds of footed rollers on compaction

equipment.

to the old, s ince i ts shorter shafts do not gocompletely through the lift.

Another way to blend the new lift with the old lift isto make the surface of the previously compacted liftvery rough. Commonly in the construction of soil

liners, the finished surface of a completed lift iscompacted with a smooth, steel drum roller to sealthe surface of the completed lift. The smooth soil

surface of the completed lift minimizes desiccation,helps prevent erosion caused by runoff from heavyrains, and helps in quality control testing. The soilhowever, must be broken up with a disc before a newlift of soil can be placed over it.

In below-ground disposal pits, it is sometimesnecessary to construct a soil liner on the side slopesThis sloping clay liner component can be constructedeither with lifts parallel to t he face of the soil or with

horizontal lifts (see Figure 6-8).Horizontal lifts must

be at least the width of one construction vehicle, orabout 12 feet.

Horizontal lifts can be constructed on almost any

slope, even one tha t is almost vertical. Para llel liftshowever, cannot be constructed on slopes at anglessteeper than about 2.5 horizontal to 1 vertical (a

92



Parallel Lifts

5 Min.

Horizontal Lifts

Figure 6-8. Liner construction on side slopes with horizontal

and parallel lifts.

slope angle of about 2 2 degrees), because th ecompaction equipment cannot operate on them.

On surfaces without steep slopes, soil liners withparall el lifts ar e less sensitive to some of the defectsthat might occur during construction than thosebuilt with horizontal lifts. Figure 6-9 shows a liner

containing a qu anti ty of sandy materia l . Withparallel lifts, the sandy zone is surrounded by zonesof good soil, and so has little influence. But withhorizontal lifts, a window through the soil liner could

allow greater permeability to waste leachate if itwere to.occur on the bottom in a sump area.

Compaction Equipment

In addition to increasing bonding between lifts (asdiscussed in the previous section), the equipme ntused to compact soil l iners should maximizecompactive energy and the remolding capability ofthe soil. The type of rolle r, weight of the roller, and

number of passes the equipment makes over thesurface of the soil are all important factors. Theheaviest rollers available weigh between 50,000 and70,000 pounds. The Cate rpilla r 825 is an example ofone of the heaviest, weighing more than 50,000pounds and having long, penetrating feet. A mediumweight roller weighs between 30,000 an d 50,000pounds and a rela tively light roller weighs 15,000 to30,000 pounds.

Parallel Lifts

Horizontal Lifts

Figure 6-9. Effect of sandy soil zone on liners with parallel

and horizontal lifts.

The best way to compare one roller to another is toexamine weight per linear foot along the drumsurface. A very lightweight roller will typicallyweigh about 500 pounds per linear foot along thedrum surfaces, while a very heavy roller weighs3,000 to 5,000 pounds per linear foot.

Vibratory rollers, weighing typically 20,000 to30,000 pounds static weight may not be effective forclay compaction. A piece of vibration equipmentinside the drum gives the vibratory roller it s name.The drums of static rollers are filled with liquid,

making them very heavy. The vibratory equipmentinside the drum of the vibratory roller, however,prevents it from being filled with wate r, so the totalweight is in the d rum itself. This kind of roller workswell for compacting granular base material s beneathpavements, so contractors frequently have themavailable. However, there is no evidence that thehigh frequency of vibration is effective in compacting

clay.

Vibratory rollers ar e not good rollers for compactingclay liner materials for several reasons. First, thepadfoot (only about 3 inches long) does not fullypenetrate the soil. Second, the area of the foot isfairly large. Because the weight is spread over alarge area , the stresses ar e smaller and the soil is notcompacted as effectively. The sm alle r the a re a of thefoot, the more effective in remolding the soil clods.Third, the roller is relatively lightweight, weighing

93



only 20,000 to 30,000 p o u n d s . I n a d d i t i o n ,approximately half the rollers' weight goes to therear axle and the rubber tires, leaving only about

15,000 pounds or less to be delivered to the dr um.

The feet of a classic sheepsfoot roller, in contrast tothose of the vibra tory roller, a re about 9 inches long.The area of the foot is relatively small so that thecompact stress on the tip typically ranges from 200 to700 pounds per square inch. The drum normally is

filled with liquid so that great weights are achieveddirectly on the d rum. Manufacturers make very fewsheepsfoot rollers now, despite the fact tha t they a rethe most effective roller for clay compaction. TheCaterpillar 815 and 825 ar e two of the few sheepsfootrollers curren tly being produced.

The Construction ProcessTable 6-2 outlines the major steps in the construction

process for clay liners. First, a source of soil to beused in constructing the liner must be found. Thenthe soil is excavated a t this location from a pit calleda "borrow pit." (Excavated soil is referred to as

"borrow soil.") Digging te sts pit s in th e borrow a re ahelps determine the stratification of the soil beforebeginning excavation of the borrow pit itself.

Table 6-2. Steps in the Construction Process

1 Location of Borrow Source- Boreholes, Test Pits- Laboratory Tests

2. Excavation of Borrow Soil

3. Preliminary Moisture Adjustment, Amendments, Pulverization

4. Stockpile; Hydration; Other

5 . Transport to Construction Area; Surface Preparation

6. Spreading in Lifts; Breakdown of Clods

7. Final Moisture Adjustment, Mixing; Hydration

8. Compaction; Smoothing of Surface

9. Construction Quality Assurance Testing

10. Further Compaction, If Necessary

The borrow soil is mixed and blended as i t i sexcavated to produce as homogeneous a soil aspossible. Scrapers are useful for excavating soilsfrom borrow areas, because the soil is mixed up inthe scraper pan by the action of the scraper . The soilalso can be sieved and processed thro ugh a rockcrusher to grind down hard clods. Cutting across

zones of horizontal stratification also will help mixup the soil as it is excavated. Using some of thesemethods, the excavation process can be designed tomaximize so i l mix ing wi thout s ign i f i can t ly

increasing excavation costs.

The next ste p is to moisten or dry t he soil as needed.If the required change in water content is only 1 to 2percent, the adjustme nt in moisture content can be

made after the soil is put in place and before it iscompacted. However, if a substantial change in soimoisture conten t is necessary, i t should be performedslowly so moistening occurs uniformly throughouthe soil. To change the soil moisture content, the soiis spread evenly in a layer, moistened, and thencovered for several days, if possible, while themoisture softens the soil clods. A disc o r a rototillerpassed through the soil periodically speeds up the

process.

If soil moistu re content is too high, the soil should bespread in lifts and allowed to dry. Mixing the soidur ing the drying process will prevent a dry cru st osoi l f rom forming on the top wi th wet soiunderneath.

When the moisture adj ustments have been made, thesoil is transported to the construction are a. Then thesoil is spread in lifts by bulldozer or scraper, and a

disc or rototiller is used to break down soil clodsfurther. A pulvermixer, a piece of equipment widelyused for reclaiming asphaltic concrete pavementalso works well. These machines can pulverize and

mix a lift of soil as much as 24 inches deep.

Once the soil is in place and prior to compactionminor a djustmen ts in moisture content aga in can bemade. N o large changes in water content should bemade a t th is time, however.

In the next step, the soil is compacted. Afterwardsthe surface of the soil may be smoothed by a smoothsteel drum roller before the construction qualitycontrol inspector performs the moisture density testIf the test indicates tha t the soil ha s been compactedadequately, the next lift is placed on top of it. If th ecompaction has not been performed properly, the soi

is either compacted further or that section of theliner is dug up a nd replaced.

Soil-Bentonite Liners

When there is not enough clay available at a site toconstruct a soil liner, the clay can be mixed with

bentonite. The amount of benton ite needed should bedetermined in the laboratory and then adjusted toaccount for any irregularities occurring duringconstruction. Dry bentonite is mixed with the soifirst, and water is added only after the mixingprocess is complete.

The bentonite can be mixed using a pugmill o r byspreading the soil in lifts and placing the bentoniteover the surface. Passing a heavy-duty pulvermixerrepeatedly through the soil in both directions mixesthe soil with the bentonite. After th e bentonite andclay are mixed, water is added in small amountswith the soil mixed well after each addition. Whenthe appropriate moisture content is reached, the

clay-bentonite soil is compacted.

94



After the construction process is finished, the newlycompacted soil line r, along with the last lift of soil,

must be covered to protect against desiccation orfrost action, which can crack the soil liner.

Construction Quality Assurance (CQA)TestingConstruction quality assurance (CQA) control tests

must be performed on the finished liner. There aretwo categories of CQA tests: tests on the quality ofthe material used in construction, and tests on thecompleted lift soil to ensure tha t proper constructionhas taken place. These tests include:

Materials Tests:

0 Atterberg Limits

0 Grain Size Distribution

0 Compaction Curve

0 Hydraulic Conductivity of Lab-compacted Soil

Tests on Prepa red and Compacted Soil:

0 Moisture Conten t

0 Dry Density

Hydraulic Conductivity of "Undisturbed" Sample

Table 6-3 presents recommendations for testingfrequency at municipal solid waste landfills. Thistable was developed by the Wisconsin Department ofNatur al Resources and is also contained in the EPA'stechnical resource document on clay liners.

Tests on prepared and compacted soils often focus on

water content and soi l densi ty . Construct ionspecifications usually require minimums of 95

percent of maximum density with standard Proctorcompaction or 90 percent of maximum density withmo di f ied P r o c to r co mp ac t i o n . A ccept a b epercentages of,water content commonly range from 1

to 5 percent higher than optimum moisture content.

Figure 6-10 shows a typical window of acceptableranges for percentages of water content andminimum density. Typical data for water contentand density gathered in the laboratory ar e plotted inFigure 6-11. The solid data points represent samples

for which the hydraulic conduct ivities were less than

10-7 c d s e c . The open symbols correspond to samplesthat have hydraulic conductivities greater than 10-7c d s e c . The window of acceptable moisture contentsand densities includes all of the solid da ta points andexcludes all of the open data points. Such a windowdefine5 accep tab le mois tu re con te n t lde ns i t ycombinations with respect to methods of soilcompaction (modified Proctor or standard Proctor).

The window in Figure 6-10 is actually a subset of thewindow in Figure 6-11.

Defining a window of acceptable water contents andd e n s i t i e s f o r h y d r a u l i c c o n d u c t i v i t y i s arecommended first step in developing constructionspecifications. Someone designing a clay liner mightchoose jus t a small pa rt of an acceptable range suchas that defined in Figure 6-11. For example,

extremely high water content may be excluded to

establish an upper water content limitation becauseof concerns over the strength of the soil. In an aridregion, where wet soil can dry up and desiccate, the

decision might be made to use only materials at thedry end of the range.

Factors Affecting Construction QualityAssurance (CQA) Control TestingKey fac to r s tha t a f f ec t cons t ruc t ion qual i ty

assurance control testing include sampling patt erns,testing bias, and outliers, or data that occurs outside

of the normally accepted range.

The first problem in construction quality controltesting is deciding where to sample. Some bias islikely to be introduced into a sampling patternunless a completely random sampling pattern isused. For random sampling, it is useful to design agrid pattern with about 10 times as many grids assamples to be taken. A random number generator,

such as those on many pocket calculators, can beused to pick the sampling points.

Bias in tes t results can originate from many areas, soit is important to include in a CQA plan a procedurefor verifying test results. Nuclear density andmoisture content tests , for example, can err slightly.

The CQA plan should specify tha t these test s will bechecked with other tests on a prescribed frequency tocut bias to a minimum. Certain "quick" moisturecontent tests, such as tes ts using a microwave ovento dry soil, can also be biased. The plan must specifythat these kinds of tests be cross-referencedperiodically to more standard tests.

Clay content and hydraulic conductivity te sts on so-called "undisturbed" samples of soil often give li ttleuseful information. To obtain accurate results, the

conditions of the tested sample must match t he fieldconditions as closely as possible. In the Houston testpad, for example, laboratory test s on 3-inch diameter

tube samples gave results t hat differed by five ordersof magnitude from the field value.

Inevitably, because soil is a variable materi al, somedata points will be outside the acceptable range. Thepercentage of such points, or outliers, that will beallowed should be determined i n advance of testing.Figure 6-12 shows that ifenough data points define a

95



Table 6-3. Recommendations for Construction Documentation of Clay-Lined Landfills by the Wisconsin Department of Natural

Resources

Item Testing Frequency

1 Clay borrow source testing Grain size 1,000 yd3

Moisture 1,000 yd3

Atterberg limits (liquid limit and plasticity 5.000 yd3

5,000 yd3 and all changes in materialMoisture-density curve

Lab permeability (remolded samples) 10,000 yd3

index)

2 Clay liner testing during construction Density (nuclear or sand cone) 5 tests/acre/lift (250 yd3)

5 tests/acre/lift (250 yd3)

1 tesVacre/lift (?SO0 yd3)

1 tesVacre/lift (1,500 yd3)

1 testiacre/lif t (1,500 yd3)

1 testiacreilift (1,500 yd3)

1 testiacrellift (1,500 yd3)

5,000 yd3 and all changes in material

Moisture content

Undisturbed permeability

Dry density (undisturbed sample)

Moisture content (undisturbed sample)

Atterberg limits (liquid limit and plasticity

Grain size (to the 2-micron particle size)

Moisture-density curve (as per clay borrow

index)

requirements)

3 Granular drainage blanket testing Grain size (t o the No. 200 sieve) 1,500 yd3

Permeability 3 000 yd3

Source Gordon, M E , P M Huebner and P Kmet 1984 An evaluation of the performance of four clay-lined landfills in WisconsinProceedings, Seventh Annual Madison Waste Conference, pp 399-460.

w o p t W

\ Acceptable\olding Water Content

Figure 6-11. Molding water contents and densities fromHouston test pad.Figure 6-10. Acceptable range fo r water conte nt and

minimum densities.

normal distribution, allowing for a percentage ofoutl ie rs , the acceptable minimum v alue for aparameter can be determined.

Test Fills

Test fills simulate the actual conditions of soil linerconstruction before the full-sized liner is built.

Generally, they ar e approximately 3 feet thick, 40 to80 feet long, and 20 to 40 feet wide. The materialsand construction practices for a test fi l l shouldimitate those proposed for the full-sized liners asclosely as possible. In si tu hydraulic conductivity ofthe soil at the test pad is required to confirm that t hefinished liner will conform to regulations. A sealed

double-ring infiltrometer usually is used for this

96



Mean

Value of Parameter

Minimum

LMean

Dry Density

Figure 6-12. Normal distribution curves.

purpose. (See Chapter Two for a detailed discussion

of testing with infiltrometers.)

When test fills first were used, the results were oftenunsuccessful. Of the first 10 test pads built aroundthe country, about half failed to meet the 10-7 c d s e chydraulic conductivity criterion. In all cases, asecond trial pad was built tha t then passed the t est.

In almost all of these second trials, the moisture

content of the soil was raised and, in some cases, aheavier piece of equipment was used to compact thesoil. The biggest problems with test pads have beenoverly dr y soils an d lightweight compactors. Of th eapproximately 50 test pads built recently, about 90

percent passed the t est t he first time.

The test pad is useful not only in teaching people howto build a soil liner, but also in functioning as a

construction quality assurance tool. If the variablesused to build a test pad that achieves 10-7 mlsechydraulic conductivity are followed exactly, then thecomple ted fu l l - s ize l iner shou ld meet E P Aregulations.

Figure 6-13 i l lus trates an ideal tes t pad. Thecompacted clay liner has been built on top of anunderdra in , the bes t in s i tu te s t ing methodavailable. The liner surface also contains a sealed

infiltrometer to measure percolation into the liner. Aliner with a n extremely low hydraulic conductivitywill lose little liquid out of the bottom over a

reasonable period of time, but inflow from the top ofthe liner can be measured. Water ponded on thesurface of the li ner, and a pile of gravel th at can beadded to the liner to compact the clay, can be used toevaluate the inf luence of a s mal l amount of

overburden stress. Most test s are performed with nooverburden pressure on the soil, yet there are oftenhigh compressive stres ses act ing on a liner in thefield.

The conditions shown in the test pad in Figure 6-13

are not always possible to replicate. For the mostreliable test results , however, the owner or operatorof a landfill should incorporate as many of thesefeatures into the test pad as are practical for the site.

97



Gravel to Load Clayto Evaluate Effect ofOverburden Stress

Collection

Collection Pan Lysimeter

Geomembrane

Figure 6 - 1 3 . A n ideal test pad

98



7. CONSTRUCTION OF FLEXIBLE MEMBRANE LINERS

IntroductionThis chapter describes the construction of flexiblemembrane liners ( F M L s ) , quality control measuresthat should be taken during construction, and EPA' sconstruction quality assurance (CQA) program. TheCQA program for FMLs is a planned series ofactivities performed by the owner of a hazardouswaste facility to ensure that the flexible membraneliner is constructed as specified in the design. Therear e five el ements to a successful CQA program: (1)

responsibility and authority, (2) CQA personnelqualifications, (3) inspection activit ies, (4) sampling

strategies, and (5) documentation. This chapterdiscusses each of these elements.

Responsibility and AuthorityA FML may be manufactured by one company,fabricated by a second company, and installed by athir d company. The FML also may be manufactured,fabricated, and installed by the same company.Depending on how the F M L is constructed, variousindividuals will have responsibilities within theconstruction process. These individuals may includeengineers, manufacturer s, contractors, and owners.

In general, engineers design the components andprepare specifications, manufacturers fabricate th eF M L , and contractors perform the installation .

Any company that installs a FML should have hadpast experience with a t least 10 million square feet ofa similar FML material. Supervisors should havebeen responsible for installing at least 2 millionsquare feet of the FML material being installed atthe facility. Caution should be exercised in selectingfirms to install FMLs since many companies have

experienced dramatic growth in the last severalyears and do not have a sufficient number ofexperienced senior supervisors.

A qualified auditor should be employed to review twokey documents: (1 ) a checklist of requirements forfacilities, which will help ensure that all facilityrequirements are met; and (2) a CQA plan, which

w i ll be u s ed d u r i ng c ons t r uc t i on t o gu i deobservation, inspection, and testing.

Designers are responsible for drawing up generaldesign specifications. These specifications indicate

the type of raw polymer and manufactured sheet tobe used, as well as the limitations on delivery,storage, installation, and sampling. Some specifichigh density polyethylene (HDPE) raw polymer and

manufactured sheet specifications are :

Raw Polymer Specifications

0 Density (ASTM D1505)

Melt index (ASTM D1238)

0 Carbon black (ASTM D1603)

0 Thermogravimetric analysis (TGA) or differ-ential scanning calorimetry (DSC)

Manufactured Sheet Specifications

0 Thickness (ASTM D1593)

e Tensile propert ies (ASTM D638)

0 Tea r res istance (ASTM D1004)

0 Carbon black content (ASTM D1603)

0 Carbon black disp. (ASTM D3015)

0 Dimensional s tabi lity (ASTM D1204)

Str ess crack resistance (ASTM D1693)

Both the design specifications and the CQA plan ar ereviewed during a preconstruction CQA meeting.

This meeting is the most important part of a CQAprogram.

The preconstruction meeting also is the time to

define crit eria for "seam acceptance." Seams are th emost difficult aspect of field construction. Whatconstitutes an acceptable seam should be definedbefore the ins ta l la t ion gets under way. Onetechnique is to define seam acceptance and verify t he

99



qualifications of the personnel installing the seamsat the same time. The installer's seamers producesamples of welds during the preconstruction CQAmeeting that are then tested to determine seamacceptability. Samples of "acceptable" seams areretained by both the owner and the installer in caseof disputes later on. Agreement on the mostappropriate repair method also should be madeduring the preconstruction CQA meeting. Variousrepair methods may be used, including capstripping

o r grind ing and rewelding.

CQA Personnel QualificationsEPA requires tha t th e CQA officer be a professionalengineer (PE), or the equivalent, with sufficientpractical, technical, and managerial experience.Beyond these basic criteria, the CQA officer must

understand the assumptions made in the design ofthe facility and the installation requi rement s of the

geosynthetics. Finding personnel with the requisitequalifications and a ctual field experience can besomewhat difficult. To develop field expertise inlandfill CQA, some consulting firm s routine ly assign

a n inexperienced engineer to work with trained CQApeople on a job site and not bill for the inexperiencedengineer receiving training. This enables companiesto build up a reservoir of experience in a short periodof time.

Inspection ActivitiesBecause handling and work in the field can damagethe manufactured she ets, care must be taken whenshipping, storing, and placing FMLs. At every step,the mater ial should be carefully checked for signs ofdamage a nd defects.

Shipping and Storage Considerations

FML panels frequently ar e fabricated in the factory,rath er t han on site. The panels must be shipped andstored carefully. High crys talline FML, for example,

should not be folded for shipment. White lines, whichindicate str ess failure, will develop if this mate ria l isfolded. Flexible membrane liners that can be folded

should be placed on pallet s when being shipped to thefield. All FMLs should be covered during shipm ent.Each shipping roll should be identified properly withname of manufacturerlfabricator, product type andth ickness , manufac ture r ba tch code, da te ofmanufacture, physical dimensions, panel num ber,

and direc tions for unfolding.

Proper onsite storage also must be provided for thesematerials. All FMLs should be stored in a secure

area , away from dirt, dust, water, and extreme heat.In addi tion, they should be placed where people andanimals cannot d i s turb them. P roper s torageprev ents heat - induced bonding of the rol ledmembran e (blocking), and loss of plas tici zer o rc u r i n g of t he po l ym e r , w h i c h c ou ld c a us e

embrittlement of the membrane and subsequentseaming problems.

Bedding Considerations

Before placing the membrane, bedding preparationsmust be completed. Adequate compaction (90 percentby modified proctor equipment; 95 percent bystanda rd proctor equipment) is a must. The landfillsurface must be free of rocks, roots, and wate r. The

subgrades should be rolled smooth and should be freefrom desiccation cracks. The use of herbicides canalso affect bedding. Only chemically compatibleherbicides should be used, particularly in surfaceimpoundments. Many herbicides have hydrocarbon

carriers that will react with the membranes anddestroy them.

FML Panel Placement

Prior to unfolding o r unrolling, each panel should be

inspected carefully fo r defects. If no defects arefound, the panels may be unrolled. The deliveryticket should describe how to unroll each panel.Star ting with the unrolling process, care should betaken to minimize sliding of the panel. A properoverlap €o r welding should be allowed as each panelis placed. The amount of panel placed should belimited to that which can be seamed in 1day.

Seaming and Seam Repair

After the panels have been inspected for defects, theymust be seamed by a qua l i f ied se amer . Themembrane must be clean for the sea ming process andthere must be a firm foundation beneath the seam.

Figure 7-1 shows the configuration of several types ofseams.

The most important seam repa ir criterion is that anydefective seam must be bounded by areas that pass

fitness structure tests. Everything between suchare as must be repaired. The repair method should bedetermined and agreed upon in advance, andfollowing a repair, a careful visual inspection should

be performed to ensure the repa ir is successful.

Weather and A nchorage Criteria

Weather is an addi t ional cons iderat ion wheninstalling a FML. From the seaming standpoint, it isimport ant not to expose the liner mate rials to rai n ordust. Any time the temperature drops below 50"F,

t h e i n s t a l l e r s h o ul d t a k e p r e c a u t i o n s f o rtemperature. For example, preheaters with the

chambe rs around them may be used in cold weatherto keep the FML warm. There also should be noexcessive wind, because it is very difficult to weldunder windy conditions.

In addit ion, FML panel s should be anchored as soonas possible. The anchor trench may remain open for

100



La p Seam

Lap Seam with Gum Tape

Tongue and Groove Spllce

Extrusion Weld L ap Seam

Fillet Weld Lar, Seam

Double Hot Air or Wedge Seam

Figure 7-1. Configurations of field geomembrane seams.

several days after installation of a panel. However,the anchor trench must be filled when the panel is a t

its coolest tempera ture and is, therefore, shortest inlength . Thi s will occur early in the morning.

Additiona Polymer ComponentsPolymer components, such as geotextiles, geonets,and geogrids, must be carefully inspected, as ther e isno CQA program for these components. ChapterFour discusses polymer components in more detail.To date; CQA activities have focused on FMLs, andthere is no way to "fingerprint" other materials todetermine their characteristic properties over the

long term. Fingerpr inting refers to the evaluation ofthe molecular structure of the polymer. For example,some geonets sold on the market use air-entrainedpolymers to create "foamed" geonets with greaterthicknesses. Over time, however, the air .in theentrained bubbles diffuses through the polymer and

the drainage net goes flat. When loads are left onthese geonets for testing purposes, it is possible toobserve orders-of-magnitude reductions in thecapacity of these materials by the 30th day of testing.

Geotextiles, geogrids, and geonets all should bepurchased from companies that have institutedquali ty control procedures at their plants and

unders tand the l iabi l i t ies , the r isks , and the

problems associated with landfill liner failure.

Sampling StrategiesIn a CQA program, there are three sampl ingfrequency criteria: ( 1 ) continuous (100 percent), (2)judgmental, and ( 3 ) statistical. Every FML seamshould be tested over 100 percent of its length. Anytime a seaming operation begins, a sample should be

cut for testing. A sample also should be taken anytime a seaming operation is significantly modified(by using a new seamer or a new factory extrusionrod, o r by maki ng a major adjust ment to th e

equipment).

Continuous (1 00 Perc ent) Testing

There are three types of continuous tests: visual,des t ruct (DT), and nondes truct (NDT). Visualinspection must be done on all seams, and DT testsmust be done on all st artup seams.

There a re several types of nondestruct (NDT) seamtests (see Table 7-1). The actua l NDT test depends on

the seam type and membrane polymer. An air lance(a low pressure blast of ai r focused on the edge of theseam) can be used on polyvinyl chloride ( P V C ) ,chlorinated polyethylene (CPE), and other flexibleliner materials. If there is a loose bond, the air lance

will pop the sea m open.

In a mechanical point stress test, a screwdriver or apick is pressed into the edge of the s eam to detect aweak bond location. In a vacuum chamber test, theworker applies soapy water to the seam. The vacuumchamber is then moved over the seam. If there is ahole, the vacuum draws air from beneath themembrane, causing a bubble to occur. The chamber

should not be moved too quickly across the seam. Tobe effective, the vacuum box should remain on each

portion of the seam at least 15 seconds before it ismoved. Otherwise, it may not detect any leaks.

The pressurized dual seam test checks air retentionunder pressure. This tes t is used with double hot airor wedge seams that have two parallel welds with a n

air space between them, so that a ir pressure can beapplied between the welds. Approximately 30 psi isapplied for 5 minutes with a successful seam losingno more than 1 psi in tha t time. This seam cannot be

used in sumps or areas in which there is limited

space for the equipment to opera te.

Ultrasonic equipment also may be used in a variety ofseam tests. This equipment measures the energytransfer across a seam using two rollers: one thattransmits a high frequency signal, and one thatreceives it. An oscilloscope shows the signal beingreceived. An anomaly in the signal indicates somechange in proper ties, typically a void (caused by the

101



Overgrind of an extruded seam.

presence of water). Ultrasonic equipment, however,will not detect a tacked, low-strength seam or dirt

contamination, and the tests are very operator-

dirt, debris, grinding, or moisture that may affec

seam quality.

dependent Statistical Testing

Judgmen tal TestingTrue statistical testing is not used in evaluatingseams ; however, a minimum of one DT everv 500 fee

J u d g m e n t a l t e s t i n g i nv o lv e s a r e a s o n a b l eassessment of seam st rengt h by a trained operator or

CQ A inspector. Judgm ental testing is required whena visual inspection detects factors such as apparen t

of seam, with a minimum of one test perlseam, isrequired. Sumps or ramps, however, may have seam s

that are very short, and samples should not be cufrom these seams unless they appear defective. In

102



Table 7-1. Overview of Nondestructive Geomembrane Seam Tests

Noiidestrw tive Test Primary User General Comm ents

Coritrac- Design Third Cost of Speed of Cost of Type of Recording Operatorethod

tor Party Equipment Tests Tests Result Method Depend encyEngr. Inspector ($)lnsp

1, Air lance Yes .. .. 200 Fast Ni l Yes-No Manual V high

2 Mechanical Yes .- ._ Nil Fast Nil Yes-No Manua l V highpoint (pick)stress

chamber(negativepressure)

(positivepressure)

echo

impedance

shadow

3 Vacuum Yes Yes _ _ 1000 Slow V high Yes-No Manua l High

4 Dual seam Yes Yes _ _ 200 Fast Mod. Yes-N o Manua l Low

5 Ultrasonic pulse - - Ye s Ye s 5000 Mod. High Yes-No Automatic Moderate

6 Ultrasonic _ _ Ye s Ye s 7000 Mod. High Oualitative Automatic Unknown

7 Ultrasonic -. Yes Yes 5000 Mod. High Qualitative Automatic Lo w

Source: Koerner, R . M and G. N Richardson. 1987. Design of geosynthetic systems for waste disposal AS CE -GT Spe cialty Conference,

Geotechnical Practices for Wa ste Disposal, Ann Arbor, M ichigan.

addition, a minimum of one DT test should be done

per shift.

There ar e no outlier criteria for statistical testing ofseams. In other words, no failure is acceptable.Typically two test s, a shear test and a peel test, ar eperformed on a DT sample (Figure 7-2). The shear

test measures the continuity of tensile strength in amembrane. It is not, however, a good indicator of

seam qual i ty . The peel tes t provides a goodindication of the quality of a weld because it workson one face of a weld. A poor quality weld will fail

very quickly in a peel test.

In a shear test, pulling occurs in the plane of theweld. This is comparable to grabbing onto theformica on a desk top and trying to pull the formicaoff horizontally. The bond is being sheared. The peeltest, on the other hand, is a true test of bond quality.This test is comparable to getting beneath theformica at one corner of a desk top and peeling up.

Documentation

Documentation is a very important part of the CQAprocess. Documents must be maintained throughoutFML placement, inspection, and testing. A FML

panel placement log (Figure 7-31, which details thepa ne l i d e n t i t y , s ubg r a de c ond i t i ons , pa ne lconditions, and sea m details, should be kept for everypanel that is placed. This form is filled out on sitea nd t yp i c a l l y c a r r i e s t h r e e s i gna t u r e s : t he

engineer’s , the installer’s , and the regu lato ryagency’s onsite coordinator’s (i f appropriate).

In addi t ion , a l l i nspec t ion documents (e .g . ,information on repairs, test sites, etc.) must becarefully maintained. Every repair must be logged(Figure 7-4). Permits should never be issued to afacility whose records do not clearly document al lrepairs.

During testing, samples must be identified by seamn u m be r a n d l o ca t io n a l o n g t h e s e a m . A

geomembrane seam tes t log is depicted in Figure 7-5.This log indicates the sea m number and length, the

test methods performed, the location and date of thetest, and the person who performed the test.

At the completion of a FML construction, an as-builtrecord of the landfill const ruction should be producedthat provides reviewers with an idea of the quality ofwork performed in the construction, as well as whereproblems occurred. This record should contain truepanel dimensions, location of repairs, a nd location ofpenetrations.

103



Dirt within an extruded seam

104



Shear Test

Figure 7-2. Seam strength tests.

Peel Test

Panel Placement Log

______--_.__._--__anel Number

Owner: Weather:

Project: Temperature:

Datef l ime: Wind:

. . ___ _ _ - - - - - _ _ _ .Subgrade C ondlt lons _ _ _ _ .__-.-_-. ~.

Line 8 Grade:

Surface Compaction:

Protrusions:

Ponded Water: Dessica tion

.____._...__- .....- anel Conditions ._._..____.____

Transport Equipment:

Visual Panel Inspection:

Temporary Loading:

Temp. Welds/Bonds:Temperature:

Damages:

____________....._____eam Details_________

Seam Nos.:

Seaming Crews:

Seam Crew Testing:

Notes:

Figure 7-3. Panel placement log.

10 5



Geomembrane Repair Log

Type Repair Type

Figure 7-4. Geomembrane repair log.

Date

106

Seam Panels Location Material Description of Damage Type Of Repair Test Tested B

I!I

!

I

II



Destructive Testontinuous Testing

Seam Seam Visual Air Test Pressure P eel Shear Location Date Tested

No Length Inspect Temp Method IniVFinal Test Test BY

I i !

I

I

i

I I I I

1

i

Geomembrane Seam Test Lo g

Figure 7-5 . Geomembrane seam test log.

107



8. LINER COMPATIBILITY WITH WASTES

IntroductionThis chapter discusses chemical compatibi l i ty

(resistance) of geosynthet ic a nd na tu ra l l ine rmaterials with wastes and leachates. Even in arelat ively iner t environment, certai n mate rial s

deteriorate over time when exposed to chemicals

contained in both hazardous and nonhazardousleachate. It is important to anticipate the kind and

quality of leachate a site will generate and selectliner m aterials accordingly. The chemical resistance

of any flexible membrane liner (FML) materials,

geonets, geotextiles, and pipe should be evaluated

before insta llation .

Chemical compatibility tests using EPA Method

9090 should always be performed for hazardouswaste sites, but some municipal waste sites also

contain hazardous, nondegradable materials. EPA

conducted a 5-year study of the impac t of munic ipalrefuse on commercially available liner materia ls an d

found no evidence of deterioration within that

period. However, in a curr ent study of lea chate

quality in municipal land fills, the Agency ha s

discovered some organic chemical constituents

normally found in hazardous waste landfill facilities.Apparently small qu anti ties of household hazardous

waste .enter municipal sit es or a re disposed of a ssmall quantit y generator wastes. As a result of these

findings, EPA is developing a position on the need €or

chemical compatibi l i ty test s for t hou sa nds ofmunicipal waste disposal sites.

In general, cover materials, including membranes

and geosynthetics, do not need to be checked forchemical compatibility since these materials do not

encounter leachates. Research data indicate th at t hepredominant gases coming from municipal sites are

methane, hydrogen, and carbon dioxide, although afew others may be emitted from household hazardous

waste. These gases pass through cover materials by

diffusion and evidence to date indicates that theyhave caused no deterioration of membranes. Also,chemical compatibility of cover mater ials with gases

has not been a major problem at hazardous waste

facilities.

A primary objective of chemical compatibility testing

is to ensure that liner materials will remain intactnot just during a landfill’s operation but also through

the post-closure period, and preferably longer. I t isdifficul t , however, to predict future chemical

impacts. There is no guarantee that liner materialsselected for a site today will be the s ame as ma terialsmanufactured 20 years from now. For example, the

quality of basic resins has improved considerably

over the la st few years.

The wastes themselves also change over time. Tests

should be performed to ensure that landfill leachatewill not permeate the liner layer. EPA recommends a

variety of physical property degrada t ion tests,

i nc lud ing a f i nge rp r in t p rog ram o f t h e r m o -

g r av i m e t r ic an a l y s i s , d i f f e r en t i a l s can n i n gc a l or i m et r ic t e s t s , a n d i n f r a r e d a n a l y s i s .Fingerprinting involves analyzing the molecular

structure of the leachate components. Sometimes a

particularly aggressive leachate component can beidentified by eval uat ing the finge rprint analy sis

tests af ter exposure of the membrane to the leachate.

Exposure ChamberThe first area of concern in chemical compatibility

testing is the exposure chamber used to hold theleachate and membranes being tested. T he exposure

chamber tank can be made of stainless steel ,

polyethylene, glass, or a variety of other materials.Any geosynthetic liner material being considered

must be tested for chemical compatibility with theleachate. Some leachates have caused rusting and

deterioration of stainless steel tank s in the past, andif polyethylene is being evaluated, the tank shouldbe of another type of materia l to prevent competition

between the ta nk ma terial a nd the te st specimen foraggressive agents in the leachate.

The conditions under which the material is testedar e crucial. The top of the exposure chamber must be

109



sealed and the tan k should contain no free air space.A stirring mechanism in the tank keeps the leachatemixture homogeneous and a heater block keeps it a t

an elevated temperature as required for the test.Stress conditions of the material in the field also

should be simulated as closely as possible. Theoriginal EPA Method 9090 test included a rack to

hold specimens under stress conditions but w a srevised when some materials sh rank in t he leachate.Due to the hazardous natur e of the materi al, testing

should be performed in a contained environment andsafety procedures should be rigorously followed.

In some cases a sump at the waste managementfacility can be used as an exposure chamber if it islarge enough. The designer of a new landfill site candesign a slightly larger sump especially for thispurpose. However, since the temperature of a sumpis colder than room temperature (55°F instead of

72"F), the geosynthetics need to be exposed for alonger period of time. Instead of 120 days, the testmight take 6 months to a year or longer.

Representative Leachate

It is important that the sample being tested isrepresentative of the leachate in the landfil l .Leachate sampled directly from a sump is usuallyrepresentative, but care mus t be taken not to mix itduring removal. This will disturb the sample'shomogene ity and may resu l t i n component ssepara ting out. Another problem is that municipalsolid waste landfill leachate will start to oxidize assoon as it leaves the sump and probably should besampled under a n iner t atmosphere.

A sampler should be fam iliar with the source of allthe leachate at a site before removing a sample. If

radioactive m ateria ls ar e present, extra car e must betaken.

At some exist ing waste management facil i t ies ,operators have placed coupons of geosynthetic

materials into sump areas to monitor leachate

effects. Information gathered from this monitoringprocedure provides an excellent da ta base. Regularrecording of data allows the operator to discovercompatibility problems as they develop, rather thanwaiting until a landfill liner fails. If the coupon

shows early signs of deterioration, the operator canrespond immediately to potential problems in the

facility.

When planning construction of a new s i te , an

operator first assesses the market to determine thequanti ty and quali ty of waste the landfil l willreceive. Representative leachate is then formulated

based on the operator's assessment.

The Permit Appl icant ' s Guidance Manual forTreatment, Storage, and Disposal (TSD) facilities

conta in s addi t io na l in format ion on l eacha t erepresenta t iveness (see Chapt er 5 , pp. 15-17;

Chapter 6, pp. 18-21; and Chapt er 8, pp. 13-16).

Compatibility Testing of Components

Geosynthetics

EPA's Method 9090 can be used to evaluate all

geosynthetic materials used in liner and leachatecollection and removal systems currently being

designed. Method 9090 is used to predict the effectsof leachate under field conditions and has beenverified with limited field da ta . The test is performedby immers ing a geosynthet ic in a c he m i c a le nv i r onm e n t €o r 120 da ys a t t w o d i f f e r e n ttemperatures, room and elevated. Every 30 days,samples are removed and evaluated for changes inphysical properties. Tests performed on F M L s a r e

listed in Table 8-1. The result s of any test should becross-referenced to a second, corollary test to avoiderrors due to the te st itself or to the laboratory

personnel.

Table 8-1. Chemical Compatibility Tests for FMLs

Hardness

Melt Index

Extractibles

Volatile Loss

Peel Adhesion

Tear Resistance

Specific Gravity

Low Temperature

Water Absorption

Puncture Resistance

Dimen sional Stability

Modulus of Elasticity

Bonded Seam Strength

Hydrostatic Resistance

Carbon Black Dispersion

Thickness, Length, Width

Tensile at Yield and Break

Environmental Stress Crack

0 Elongation at Yield and Break

Physical property tests on geotextiles and geonetsmust be designed to assess different uses, weights,and thicknesses of these materials, as well asconstruction methods used in the field. EPA has alimited data base on chemical compatibility withgeotextiles. Some tests for geonets and geotextilesrecommended by EPA are listed in Table 8-2. The

ASTM D35 Committee should be consulted forinformation on the latest testing procedures.

110



Table 8-2. Chemical Compatibility Tests forGeonetdGeotextiles

Puncture

Thickness

Permittivity

Transmissivity

MassNnit Area

Burst Strength

Abrasive Resistant

Percent Open Area

Ultraviolet Resistivity

Grab TensileiElongation

Equivalent Ope ning Size

Hydrostatic Bursting Strength

Tearing Strength (Trapezoidal)

Compression BehavioriCrush Strength

Until recently, EPA recommended using 1 1/2 times

the expected overburden pressure for inplanetransmissivity tests. Laboratory research, however,

has revealed tha t creep and intrusion cause a loss oftransmissivi ty, so the Agency has amended i ts

recommendation to 2 to 3 t imes the overburden

pressure. EPA also recommends tha t th e geotextile

or geonet be aged in leachate, but th at the actua l test

be performed with water. Performing the test withthe leachate creates too great a risk of contamination

to test equipment and personnel. T he transmissivitytest should be run for a minimum of 100 hours. The

test apparatus should be designed to simulate thefield conditions of the actual cross section as closely

as possible.

Pipes

The crushing streng th of pipes also should be tested.

There have been examples where pipes in landfills

have actually collapsed, and thus forced the site tostop operating. Th e ASTM D2412 is used to measure

the str ength of pipe mat erials.

Natural Drainage Materials

Natural drainage materials should be tested toensure that they will not dissolve in the leachate or

form a precipitant tha t might clog the system. ASTM

D2434 will evaluate the ability of the materials to

retain permeability characteristics, while ASTM

D1883 tests for bearing ra tio, or the ability of t hematerial to support the w aste unit.

Blanket ApprovalsEPA does not grant "blanket approvals" for any

landfill construction materials. The quality of liner

mater i a l s var i es cons iderab ly , depend ing on

quantit ies produced a nd on the m anufac turer. Even

high dens ity polyethylene (HDPE) does not receiveblanket approval. The Agency, together with a group

of chemists, biologists, and researchers from thel iner manufacturing industry, determined th at

HDPE varies slightly in its composition among

manufacturers. Because different calendaring aids,

processing aids, or stabilizer packages can changethe overal l characterist ics of a product , each

mat eria l should be individually tes ted.

Landfill designers should select materials on the

bas is of site-specific conditions, as t he composition of

specific leachates w i l l vary from site to site. Adesigner working with the operator determines in

advance of construction wha t ma teri als will be mosteffective. In recent year s, E P A has restricted certain

wastes, including liquids, from land disposal. Theseregulations have expanded the number of potential

candidate materials, thus allowing more flexibility

to landfill designers .

Interpreting DataWhen liner material test data show the rate of

change of the mate rial to be nil over a period of time,then the membrane is probably not undergoing any

chemical change. There have been ins tances ,however, in which a material w a s tested for a year

without change and then suddenly collapsed. For

this reason, the longer the testing process can

continue, the more reliable the data will be. When

test data reveal a continuous rat e of change, then th emater ial is reacting with the leachate in some way. If

the data show an initial continuous rate of changethat then tapers off, new leachate may need to be

added more often. In any case, the situatio n should

be studied in more detail.

A designer should consult with ex perts to inte rpretdata from chemical compatibility tests. To meet this

need, EPA developed a software system cal led

Flexible Membrane Liner Advisory Expert System

(FLEX) to assist in evaluating test data. FLEX is a n

expert system that is based on data from manyc h e m i c a l c o m p a t i b i li t y te s t s a n d c o n t a i n s

interpr etation s from experts in the field.

1 1 1



9. LONG-TERM CONSIDERATIONS: PROBLEM AREAS AND UNKNOWNS

IntroductionThis chapter presents an overview of long-termconsiderations regarding the liner and collectionsystems of hazardous waste landfil ls , surfaceimpoundments, and waste piles. Included in the

discussion are flexible membrane liner and clay linerdurability, potential problems in leachate collectionand removal systems, and disturbance and aesthetic

concerns in caps and closures.

will stay intact indefinitely. This is because stablepolymers in isolation have a n equilibrium molecular

structure that prevents aging. The usual indicatorsof s tre ss ar e changes in density, p , and glasstransition temperature, T,. The glass transitiontemperature is the temperature below which theamorphous region of a crystalline polymer is rigidand brittle (glossy) and above which it is rubbery o rfluidlike,

In judging the impact of any facility, site-specificconditions such as geology and stratigraphy;seismicity; ground-water location and quali t y;population density; facility size; leachate qu an ti tyand quality; and nontechnical political, social, andeconomic factors must be taken into account. Table9-1 summarizes ar eas of concern in various landfillmateri als and components.

One of the most important considerations inplanning a waste facility is estimating t he length oftime the facility is expected to operate. Some

recommended time frames for different kinds offacilities are:

Polymers in the field, however, are subject to manyexternal influences and stresses. Experiments doneon FML materials attempt to simulate the long-termin situ aging processes. One approach is to place a

sample of material in a tes t chamber a t roomtemperature (approximately 70"F), and ano ther

sample in a chamber heated to 120" to 160°F. Theactivation energy for the two FMLs is evaluated.Then a model based on the Arrhenius equation isused to determine the length of time it would takethe sample kept at 70°F to degrade to the sameexten t a s the h igh t empera tur e sample . Thi sprocedure, however, assumes that temperature

increase is physically equivalent to the time ofexposure in the field, an assumption with which

1-5 years many chemists disagree.Heap leach pads

Waste piles 5-10 years

Surface impoundments 5-25 years

Solid waste landfills 30-100 years

0 Radioactive waste landf ills 100-1000+ years

None of these t ime f rames a re s e t for th inregulations, however. The only time frame regulatedby EPA is the 30-year post-closure care period for all

hazardous waste landfill facilities.

Flexible Membrane LinersT he m a j o r l ong - t e r m c ons i de r a t i on i n a nysynthetically lined facility is the durabil i ty offlexible membrane line rs (FMLs). In the absence ofsunlight, oxygen, or stresses of any kind, a properlyformulated, compounded, and manufactured FML

Therefore, in the absence of a direct measurementmethod for FML lifetime, it becomes necessary toevaluate all of the possible degradation mechanismsthat may contribute to aging.

Degradation Concerns

A number of mechanisms contribute to degradation

in the field, many of which can be controlled withproper design and construction. A FML can be

w e a ke ne d by va r i ous i nd i v i dua l p hys i c a l ,mechanical, and chemical phenomena or by thesynergi s t ic effects of two o r more of thesephenomena. Polymeric materials have a n extremely

elongated molecular structure. Degradation cutsacross the length of this structur e in a process knownas chain scission. The more chain breakages thatoccur, the more the polymer is degraded by loss of

113



Table 9-1. Long-Term Concerns in Landfill Mechanisms (V = yes; N = no)

LCR Sec. Liner LDCRap Cap - SWCR

Mechanism FML Clay Nat . SY N P-FML Nat. SYN FML Clay Nat . SYN

Subsoil

Waste

1 . Movement of N N N N Y N N Y Y N N

2. Subsidence of Y Y Y Y N N N N N N N

3. Aging Y Y Y Y Y Y Y Y N N Y

4. Degradation Y N N Y Y N Y Y N N Y

5. Clogging N N Y Y N Y Y N N Y Y

6. Disturbance Y Y Y Y N N N N N N N

where:FM L = geomembrane linerLC RSWCR = surface water co llection and r emov al systemNat. = made from natural soil materialsSyn.

= leachate collection and rem oval system

= made from synthetic polymeric materials

streng th and loss of elongation. Each of the processesinvolved in chain scission will be discussed in thefollowing sections.

Oxidation DegradationOxidation is a major source of polymer degradationleading to a loss of mechanical properties andembrittlement. The steps in this process are as

follows:

0 Heat liberates free radicals.

0 Oxygen uptake occurs.

0 Hydroperoxides accelerate uptake .

0 Hydrogen ions attach to tertiary carbons which

0 Subsequent bond scission occurs.

At high temperatures (over 200°F) oxidation occursvery rapidly. Consequently, oxygen will createserious problems for FMLs built near furnaces,

incinerators, o r in extremely hot cl imates. Theimpact of oxidation is greatly reduced at ambient

temperatures.

ar e most vulnerable.

One can minimize oxidative degradation of FMLs bydesigning facilities with FMLs buried to avoid

contact with oxygen and to dissipate he at generated

by direct rays of the sun. Oxygen degradation is avery serious problem with mater ial s used for surface

impoundments, however, where the FML cannot be

buried or covered.

Ultraviolet Degradation

All polymers degrade when exposed to ultraviolet

light v ia the process of photooxidation. The part ofthe ultraviolet spectrum responsible for the bulk ofpolymer degradation is the wavelength UV-B (3 1 5 -

380 nm). ASTM D4355 uses Xenon Arc appara tus forassessing the effects of this wavelength on polymerictes t specimens . The Xenon Arc app ar atu s i s

essentially a weatherometer capable of replicatingthe effects

ofsunlig ht under laborato ry conditions.

Blocking or screening agents, such as carbon black,ar e commonly used to retard ultraviolet deg radation.

For this reason, FMLs are manufactured wi thapproximately 2 to 3 percent carbon black. Even th atsmall amount effectively retards degradation byultraviolet rays. Although the addition of carbon

black retards degradation, it does not s top i tcompletely. Ultraviolet degr adatio n, however, can beprevented by burying the material beneath 6 to 12inches of soil. FMLs should be buried within 6 to 8weeks of the t ime of construction, geonets w ithin 3 to

6 weeks, and geotextiles within 1 o 3 weeks, i.e., the

higher the surface area of the material , the morerapidly the geosynthetic must be covered.

In surface impoundments where FMLs cannot be

buried, ultraviolet degradation also contributes tothe oxidation degradation . An at tem pt still should bemade to cover the FML, even though sloughing of thecover soils will occur unless the site has very gentleslopes. Various other covering strategies are being

evaluated, such as bonding geotexti les to FMLsurfaces.

High Energy Radiation

Radiation is a serious problem, as evidenced by the

N u c l e a r R e g u l a t o r y C o m m i s s i o n a n d U . S .Department of Energy’s concerns with transuranicand high level nuclear wastes. High energy radiationbreaks the molecular chains of polymers and givesoff various products of disintegrat ion.

A s of 1992, low-level radioactive wastes, such asthose from hospitals and testing laboratories, must

14



be contained in landfills. The effects of low-levelradiation associated with these waste materials onpolymers still needs to be evaluated .

Chemical Degradation: pH Effects

A ll polymers swell to a certain ex tent when placed incontact with water (pH = 7) because they acceptwater andlor water vapor into their molecularstructure. Degrees of swelling for some commonpolymers are listed below:

Polyvinyl chloride (PVC) 10 percent

Polyamide (PA) 4 to 4.5 percent

Polypropylene (PPI 3 percent

Polyethylene (P E) 0.5 to 2.0 percent

Polyester (PETI 0.4 o 0.8 percent

Polymer swelling, however, does not nece ssarily

prevent a material from functioning properly. TheU.S. Bureau of Reclamation has observed polyvinylchloride liners functioning adequately in watercanals for 20 to 25 years despite relatively large

increases in the thickness of the mat eria l.

In very acidic environments (pH < 3 ) , somepolymers, such as polyamides, (i.e., Kevlar andnylon) begin to degrade. On the other end of thespectrum, certain polyesters degrade in e xtre melya l k a l in e e n v i r o n m e n t s ( p H > 1 2 ) . H i g h

temperatures generally accelerate the chemicaldegradation process. Most landfill leachates ar e notacidic or basic enough to cause concern. However, incert ain kinds of landfills, such as those used for ashdisposal, the pH of the leachate might be quitealkaline and needs to be taken into account whenchoosing liner materia ls.

Chemical Degradation: Leachate

EPA's Method 9090 is a test procedure used toevaluate leachate degradation of FML materials. A sdescribed in- Chapte r Eight, t he FML mu st beimmersed in the site-specific chemical environment

for at least 120 days at two different temperatu res.Physical and mechanical properties of the testedmaterial are then compared to those of the originalmaterial every 30 days. Assessing subtle propertychanges can be difficult. Flexible Liner EvaluationExpert (FLEX),a software system designed to assistin the Resource Conservation Recovery Act (RCRA)permitting process, can help in evaluating E P A

Method 9090 test data.

Biological Degradation

Neither laboratory nor field tests have demonstratedsignificant evidence of biological degradation.Degradation by fungi or bacteria cannot take place

unless the microorganisms attach themselves to the

polymer and find the end of a molecular chain, an

extremely unlikely e vent . Chemical companies havebeen unable to manufacture biological additivescapable of destroying high-molecular weightpolymers, like those used in FMLs and relatedgeosynthetic materials. Microbiologists have triedunsuccessfully to make usable biodegradable plastictrash bags for many years. The polymers in FMLshave 10,000 times the molecular weight of thesematerial s, thus a re very unlikely to biodegrade frommicroorganisms.

There also is little evidence t hat insects o r burrowinganimals destroy polymer liners or cover mate rials. Intests done with rat s placed in lined boxes, none of theanimals were able to chew their way through theFMLs. Thus, degradation from a wide spectrum ofbiological sources seems highly unlike ly.

Other Degradation Processes

Other possible sources of degradation includethermal processes, ozone, extraction via diffusion,and delamination. Freeze-thaw cycling, or theprocess by which a material undergoes alternatingrapid extremes of temperat ure, has proven to havean insignificant effect on polymer stre ngth or FMLseam str engt h. Polymeric materials experience somestress due to warming, thereby slightly decreasingtheir strength, but within the range of 0" to 160"F,there is no loss of integri ty.

Ozone is related to ultraviolet degradation in the

photooxidation process, and, therefore, creates amore serious problem for polymeric materials.However, ozone effects can be essentially elim inatedby covering the geosynthetic materials within thetime frames previously mentioned.

I n t he e x t r a c t i on v i a d i f f u s i on m e c ha n i s m ,plasticizers leach out of polymers leaving a tackysubstance on t he surface of the ma teri al. The FML

becomes less flexible, but not, apparently, weakernor less durable.

Delamination of scrim-reinforced material was aproblem until 15 years ago when manufacturers

began using large polymer calendar presses. Thep r e s s e s t h o ro u g h ly i n c o r p o r a t e t h e s c r i mreinforcement, so that delamination rarely occurs

today.

Stress-induced MechanismsFreeze-thaw, abrasion, creep, and stre ss cracking areal l stress mechanisms that can affect polymers. Thefirst two, freeze-thaw and abr asion, are not likely tobe problems if the materia l is burie d su fficientl ydeep. Soil burying will el iminate temperatureextremes. Abrasion is a consideration only in surface

115



impoundments in which water wavcs come into

direct contact with the F M L .

CreepCreep refers to the deformation of the FML over aprolonged period of time under constan t stres s. It can

occur at side slopes, at anchor trenches, sumps,protrusions, set tlemen t locations, folds, and creases.

A typical creep test for a FML, o r a n y o t h e r

geosynthetic material, involves suspending a loadfrom an 8-inch wide tensile test specimen. Initiallyan elastic elongation of the material occurs. Themater ial should quickly reach a n equilibrium state ,

however, and experience no further elongation overtime. This is shown in the stabilized lower curve inFigure 9-1. The second and third curves in Figure 9-1

show test specimens undergoing states of constantcreep elongation and creep failu re, respectively

I A

Creep Failure

Constant Creep

No Creep

c0

w

Time

Figure 9-1. Typical results of a sustained load (creep) test.

One can use the design-by-function concept to

minimize creep by selecting materials in which the

allowable stress compared with the actual stressgives a high factor of safety. For semicrystalline

FMLs, such as polyethylenes, the actual stre ss mustbe significantly less than the yield s tress. For scrim-

re inforced FMLs such a s c h l o r o s u l f o n a t e dpolyethylene (CSPE), o r re inforced e thy lene

propylene diene monomer (EPDM), the act ual stressmust be significantly less tha n the break ing stren gthof the scrim. Finally , for nonreinforced plastics such

as PVC or nonreinforced chlorinated polyethylene

(CPE), the actual s tress must be less th an theallowable stress at 20 percent elongation. In allcases, one should maintain a factor of safety of 3 orhigher on values for these materials.

Stress CrackingS t r e s s c r a c k i ng is a potent ia l problem with

semicrystalline FMLs. The hig her t he crysta llineportion of the molecular structure, the lower theportion of the amorphous phase and the great er t he

possibil i ty of s t r e s s cr a c ki ng . H i gh d e ns i t y

polyethylene (HDPE) is the primary material oconcern.

ASTM defines stress cracking as a brittle failure thaoccurs at a stress value less th an a material's short

term strength. The test usually applied to FMLs isthe "Bent Str ip Test," D1693 . The bent strip test is aconstant strain test that depends on the type and

thickness of the mater ial being teste d. In performingthe test, a specimen of the FML 0.5 inch wide by 1.inches long is prepared by notching i ts surface

approximately 10 mils deep. Then the specimen isbent 180 degrees and placed within the flanges of asmall channel holder as shown in Figure 9-2

Approximately 10 repli cate specim ens can be placedin the holder simultane ously. The assembly is thenplaced in a glass tube containing a surface activewetting agent and kept in a constant temperature

bath a t 122°F. The notch t ips ar e observed focracking and/or crazing. Most specifications call fo

stress-crack free results for 500 or 1,000 hoursCommercial HDPE shee t usua lly performs very welin this particular test.

There are two things, however, that the D1693 tes

does not allow: a constant s tress test ing of thematerial and an evaluation of th e seams. The D1693process bends the te st specimen initia lly, but thenallows it to relax into its original state. Furth ermor e

the notch cannot be made to span over separate

sheet s at seam locations.

ASTM D2552, "Environ mental St re ss Rup tureUnder Tensile Load," tests HDPE materials under

constant stress conditions. In t his partic ular testdogbone-shaped specimens under constant load are

immersed in a surface active agent at 122°F (seFigure 9-3). The test specimens eventually enter

elastic, plastic, or cracked states. Commerciallyavailable HDPE sheet material performs very wel

in this test, resulting in negligible ( = 1 percentstress cracking. The test specimens are generallyelastic for stresses less than 50 percent yield and

plastic for stress levels grea ter th an 50 percent yield.

This appa rat us can be readily modified to test HDPE

seams (see Figure 9-41, Long dogbone-shapedspecimens with a constant-width cross section aretaken from the seamed region. The same tes

procedure is followed and t he tes t res ult s should bethe same. If cracking does occur, the cracks gothrough the fillet portion of the extrusion fillet

seamed samples or through the FML sheet materiaon either side of the fillet. For other types of seams

the cracking goes through the paren t sheet on one othe other side of the seamed region. Results on a wide

range of field-collected HDPE seams have shown are la t iv e ly h igh inc idence o f c r a c k i n g . T h iphenomenon is current ly being evaluated wi th

repl icate tes ts ; careful ly prepared seams; and

116



Test Sample

(A )

F

--4 Ir-

’7

.-Y-

II

J‘ /

IIIE

lI

IIIIIII

,&

Specimen Holder

(6)

Figure 9-2. Details of ASTM D1693 on “Environmental Stress-Cracking of Ethylene Plastics.”

Micro Switch to 76.20 m

Front ViewTray Moved Up andDown on Rack andPinion Arrangement

TestAssembly

( C )

Figure 9-3. Environmental stress rupture test for HDPE sheet ASTM D2552.

117



variations of temperature, pressure, and other

seaming controls.

A A

Front View

tEdge View

tEdge View

Figure 9-4. Details of test specimens from ASTM D2552

modified test for HDPE seams.

T h e t e s t is s o m e t i m e s cr i t i ci z e d a s b e i n gnonrepresentative of field conditions since thewetting agent exposes all sides of the seamed regionto liquid, which does not happen in the field. The

specimen also is too narrow to simulate real lifewide-width action. Also ther e is no confinement on

the surfaces of the material. While these criticismsdo challenge the test as being nonrepresentative,there have been three field failures that mimickedthe laboratory cracking exactly. The first was in asurface impoundment lined with 80-mil HDPE t hatwas an extrusion fillet seam. T he problem w a s not

picked up by construction quality assurance work,but instead occurred after t he facility was closed. Thesecond case involved extensive FML seam failu res a ta site in the Southwest. The se am used there was aflat extrudate. Cracking originated in several areas

due to high localized stresses and exposure to widetemperature fluctuations. The third f ailure occurreda t a site in the Northeast that used 60 mil HDPE and

a hot wedge seam. In th is case a s tress crack 18 to 20inches long formed a t the outer tr ack of the wedge.

Of all the degradation processes reviewed, stresscracking in field seams is of the greatest current

concern. It appear s as though th e field problems onlyoccur in the exposed FML seam areas of surface

impoundments. Work is ongoing to investigate thephenome non f u r t he r a nd t o unde r s t a n d t hemechanisms involved. An emphasis on carefullyconstructed and monitored field seams will cert ainly

be par t of the final recommendations.

Clay LinersClay has a long history of use as a liner material.Data is available on the effects of leachate on clayliners over periods of 10 ye ars o r more, and the

resul ts generally have been satisfactory . While claysdo not experience degradation o r stress cracking,they can have problems with moisture content andclods. High concentrations of organic solvents, and

severe volume changes and desiccation also causeconcern at specific sites. The rapid freezing and

thawing of clay liner materials also affects theirintegrity, but freeze-thaw can usually be alleviatedwith proper design and construction considerations.

For a more complete discussion of clay l inerdurability, see Chapter Two.

Leachate Collection and RemovalSystemsThe leachate collection and removal system includesall materials involved in the primary leachate

collection system and the leak detection collectionsystem. For the proper functioning of these systems,a l l ma te r i a l s being used mus t be chemica l ly

resistant to the leachates th at they a re handling. Ofthe nat ura l soil materia1s;gravel an d san d ar egenerally quite resistant to leachates, with thepossible exception of freshly ground limestone. Withth i s mate r i a l , a so lu t ion conta in ing ca lc ium,magnesium, or other ion deposits can develop as the

flow rat e decreases in low grad ient areas. A form ofweak bonding called "limerock" has been known to

occur. Its occurrence would be disastrous in terms ofl e a c ha t e co l le c ti on a nd r e m ova l . Re ga r d i nggeotextiles and geonets, the re a re no established testprotocols for chemical resistivity evaluation like theEPA Method 9090 test for FMLs. Table 9-2 suggestssome tests tha t should be considered.

Table 9-2. Suggested Test Methods for Assessing Chemical

Resistance of Geosynthetics Used in Leachate

Collection Systems (Y = yes; N = no)

Test Type Geotexti le Geonet Geocomposite

Thickness Y Y Y

Mass/unit area Y Y Y

Grab tensile Y N N

Wide width tensile Y Y N

Puncture (pin) N N N

Puncture (CBR) Y Y Y

Trapezoidal tear Y N N

Burst Y N N

The system for collecting and removing the leachatemust function continuously over the anticipated

lifetime of the facility and its post-closure careperiod. During operat ion , leachate is removedregularly depending on the amount of liquid in theincoming waste and natural precipitation entering

118



the si te. Leachate, however, continues to begenerated long after landfill closure. During the fir stfew post-closure year s th e ra te of leacha te removal i s

almost 100 percent of that during construction.Approximately 2 to 5 years after closure, leachate

generally levels off to a low-level constant cap leakrate or , in a very tight, nonleaking closure, falls to

zero (see Figure 9-51,

Clogging is the primary cause of concern for the long-

term performance of leachate collection and removalsystems. Parti culate clogging can occur in a numberof locations. First, the sand filter itself can clog thedrainage gravel. Second, the solid material withinthe leachate can clog the drainage gravel or geonet.

Third , and most likely, the solid suspended mat eria lwithin the leachate can clog the sand filter or

geotexti le fi l ter The following breakdown o fparticulate concentration in leachate at 18 landfillsshows the potential for par ticula te clogging:

Total solids

0 Total dissolved solids 584 - 44,900 mg/L

Total suspended solids

0 - 59,200 mg/L

10 - 70 0 mg/L

Salts precipitating from high pH leach ate, iro nocher, sulfides, and carbona tes can all contribute toparticulate clogging.

The potential for clogging of a filter or drainagesystem can be evaluated by modeling the system inthe laboratory. This modeling requires an exactreplica system of the proposed components, i.e. , coversoil, geotextile, geonet, etc. Flow rate plotted as afunction of time will decrease in the beginning, buteventually should level off to a horizontal line a t aconstant value. It may take more than 1,000 hoursfor this leveling to occur. Zero slope, at a constantvalue, indicates an equlibrium (o r nonclogging)

situat ion. A continuing negative slope is evidence ofclogging. A s yet, there is no formula or criteria t hatcan be substituted for this type of a long-term

laboratory flow test .

Biological clogging can arise from many sourcesincluding slime and sheath formation, biomass

format ion, ochering, sul f ide depos i t ion, andcarbonate deposition. Ocher is the precipitate leftwhen biological activity moves from one zone toanother. It is an iron or sulfide deposit, and is mostlikely to occur in the smallest apertures of filter

materials. Sand filters and geotextile filters ar e mostl i k e l y t o c l o g , w i t h g r a v e l , g e o n e t s , a n dgeocomposites next in order from most to least likely.

The biological oxygen demand (BOD) value of theleachate is a good indicator of biological cloggingpotential; the higher the number, the more viablebacteria are present in the leachate. Bacterialclogging is more likely to be a problem in municipal

landfills than at hazardous waste facilities becausehazardous leacha tes probably would be fata l to mostbacteria. Curren tly, a n EPA contractor is monitoringsix municipal landfills for evidence of aerobic andanaerobic clogging. The resu lts should be availablein 1989.

The most effective method for relieving particulateand/or biological clogging is creating a high-pressurewater flush to clean out the filter and/or the dra in. In

cases of high biological growth, a biocide may alsoneed to be introduced. Alte rna tively, a biocide mightbe int roduced into the mater ia ls d urin g thei rmanufacture. Geotextiles and geonets can include atime-release biocide on their surface, or within theirstructure. Work by a number of geotexti le andgeonet manufacturers is currently ongoing in thisa re a . Measures to remedy c logging mus t beconsidered in the design stag e.

The final factor to be considered in leachatecollection and removal systems is extrusion andintrus ion of materi als into the leak detection system.In a composite primary liner syst em, clay can readily

extrude through a geotextile into a geonet if thegeotextile has continuous open spaces, i.e., percentopen a rea (POA) > 1 percent. Therefore, relativelyheavy nonwoven geotextiles are recommended.Elasticity and creep can cause geotextiles to intrudeinto geonets from composite prima ry line rs as well. AFML above or below a geonet can also intrude intothe geonet due to elasticity and creep. Design-by-function and laboratory evaluations that simulatefield conditions should alert designers to thesepotential problems. For al l geosynthetics, a highdesign ratio value or factor of safety for strengthshould be chosen.

Cap/Closure SystemsWate r and wind erosion, lack of vege tation , excessivesunlight, and disturbance by soil-dwelling anim als

(or by people) all are potential problems for landfillclosure systems. The effects of ra in, hail, snow,freeze-thaw, and wind a re discussed in Chapter Five.Healthy vegetation growing over the cap minimizes

the erosion of soil on the slopes by these naturalelements.

The effects of animals and of sunlight (ultravioletrays and ozone) can be minimized by adequatelyburying the caplclosure facility. Soil depths overFMLs in covers range from 3 to 6 feet in thickness.

Large rocks above the FML cover can also thwar t theintrusion of anima ls into the are a. Huma n intrusion,either accidental o r intentional, can usually beprevented by posting signs and erec ting fences.

The final long-term consideration related to cap and

closure systems is aesthetic. The New JerseyMeadowlands Commission is planning for the final

119



"'F,a, Weekly I Monthly-I 50

a,i

0"

sCap Leakage

No Leakage1

I - - -e = = - -

Years after Closure

Figure 9-5. Approximate amount of leachate withdrawal after Closure.

closure of 54 landfills in the northeastern par t of the

stat e, all but 3 of which ar e already closed and readyto be capped. A graphic artis t was hired to design anattractive landscape out of one facility along aheavily traveled automobile and rail route. Thedesign included looping pipes for methane recovery,

a solar lake, and an 8-foot concrete sphere allcontributing to a visually pleasing l una r theme.

The performance of a capped and closed wastefacility is critically important. If a breach should

occur many years after closure, there is a high

l ikel ihood t ha t ma inte nan ce forces would beunavailable. In tha t event, surface water could enterthe facility with largely unknown consequences.

Thus the design stage must be carefully thought outwith long-term considerations in mind.

120



IO. LEAK RESPONSE ACTION PLANS

This final chapter reviews proposed requirements forResponse Action Plans, o r RAPs, that ar e containedin the proposed leak detection rule issued in May,1987. It focuses on the concepts behind the RAPs and

the preliminary, technical calculati ons used indeveloping them. The main topics of discussion willbe the technical basis for the two response action

triggers, action leakage rate (ALR) and rapid and

large leakage (RLL) rate ; the RAPs themselves; andthe RAP submit tal process.

BackgroundIn the Hazardous and Solid Waste Amendments(HSWA) of 1984, Congress required tha t leaks fromnew land disposal facilities be detected at t he ea rliest

practical time. However, HSWA did not require or

specify actions to be taken once a leak is detected inthe leak detection system. Therefore, EPA proposedrequirements for response action plans to deal withleaks detected in the leak detection system betweenthe two liners. EPA realizes that even with a goodcons t ruct ion qual i ty as sura nce plan, f lexible

membrane liners (FMLs) will allow some liquidtransmission either through water vapor permeationof an intact FML, or through small pinholes or tearsin a slightly flawed FML. Leakage rates resultingfrom these mechanisms can range from less than 1 to

300 gallons per acre per day (gal/acre/day). Ifunchecked, these leak rates may result in increasedhydraulic heads acting on the bottom liner andpotential subsequent damage to the liner system.

The idea behind the RAP is to be prepared for anyleaks o r clogging of the drainage layer in the leakdetection system that may occur during the activelife or post-closure care period of a waste facility. The

first step is to identify the top liner leak rates thatwould require response actions. Therefore, in theproposed leak detection rule of May 29, 1987, EPAestablished two triggers for response actions: theAction Leakage Rate (ALR) and the Rapid and LargeLeakage (RLL) rate . The ALR is a low-level leak r atethat would indicate the presence of a small hole o rdefect in the top liner. The RLL is indicative of a

severe breach or large tear in the top liner. Adifferent level of responsiveness would be requiredfor leakage rates above these two triggers. RAPsdeveloped by owners or operators may have morethan two triggers as appropriate to cover the range ofleak rates expected for a landfill unit. In addition totriggers, the proposed rule also defines the elemeqtsof a RAP, gives an example of one, and discusses the

procedures for submit ting and reviewing a RAP.

Action Leakage Rate (ALR)EPA has historically used the term de minimusleakage when referring to leaks resulting frompermeation of an intact FML. Action leakage rate(ALR) was developed to distinguish leak r ates due toholes from mere permeation of an intact FML, and toinitiate early interaction between the ownerloper-ato r of the un it and the Agency. The ALR essentiallydefines top liner leakage in a landf ill, and th e

proposed value is based on calculated leak ratesthrough a 1 to 2 mm hole in a FML subject to low

hydraulic heads on the order of 1 inch. The proposed

ALR, therefore, is representative of well-designedand operated landfills, although, as proposed, itwould also apply to surface impoundments and waste

piles.

Because EPA is considering setting a single ALRvalue applicable to landfills, surface impoundments,and waste piles, the Agency calculated top liner leakrates for different sizes of holes and for differenthydraulic heads. In addition, EPA compared leakrate s for a F M L top liner with that for a compositetop liner, since many new facilities have double

composite liner systems. Table 10-1 shows theresults of these calculations for FML and composite

top liners. Even for FMLs with very small holes (Le.,1 to 2 mm in diameter), leak r ates can be significantdepending on the hydraulic head acting on the topl i ne r . T he a dd i t i on of t he c o m pa c t e d l owpermeability soil layer to the FML significantlyreduces these leak rates to less than 10 gal/acre/day,even for large hydraulic heads t hat ar e common insurface impoundments. These results indicate th at,

121



at least for deep surface impoundments with large

hydraulic heads, double composite liner systems maybe the key to reducing the leak r ates to de minim uslevels tha t ar e below the proposed ALR.

Table 10-1. Calculated Leakage Rates through FML and

Composite Liners (gavacrdday)

FML Alone

Hydraulic Head, R

Leakage Mechanism 0.1 1 10

Small Hole ( 1 -2 mm) 30 100 300

Standard Hole (1 cm2) 300 1,000 3,000

Composite Liner (good contact)

Hydraulic Head, R

Leakage Mechariism 0.1 1 10

Small Hole (1 2 mm) 0.01 0.1 2

Standard Hole (1 cm2) 0.01 0.2 3

Source: U S . EPA. 1987. B ackground document on proposed l inerand leak detection rule. EPA /530-S W-8 7-015 .

EPA's proposed rule sets the ALR a t 5 to 20

gallacrelday, a difficult range to achieve with aprimary FML alone (especially for surface impound-m e n t s ) . T he p r opos ed r u l e a l s o e na b l e s t he

ownerloperator to use a site-specific ALR value th atwould t ake in to account meteoro logica l andhydrogeological factors , as well as design factors tha tmight result in leak rates that would frequently

exceed the ALR value. Using these factors, a surfacei m po u nd m en t t h a t m e e t s t h e m i n i m u mtechnological requi rements of a FML top line r could

conceivably apply for a site-specific ALR value.

Daily leakage rates through top liner s can vary by 10

to 20 percent or more, even in the absence of major

precipitation events. Because of these variations,EPA may allow the landfil l ownerloperator toaverage daily read ings over a 30-day period, as long

as the leakage rate does not exceed 50 gallacrelday

on any 1 day. If the average daily leak ra te does notexceed the ALR, then the ownerloperator does nothave to implement a RAP.

Rapid and Large Leakage (RLL)

The Rapid and Large Leakage (RLL) rat e is the high-level trigger th at indicates a serious malfunction of

system components in th e double-lined unit a nd t ha twarrants immediate act ion. In developing theproposed ru le , EPA def ined the RLL as t h e

m a x im um de s i gn l e a ka ge r a t e t h a t t h e l e a kdetection system can accept. In other words, the RLL

is exceeded when the fluid head is g rea ter tha n thethickness of the secondary leachate collection and

removal system (LCRS) drain age layer. The visible

expression of RLL leakage in surface impoundmentsis the creation of bubbles, or "whales," as the FML islifted up under th e fluid pressure. See Chapter Thre efor furt her discussion of "whales".

Because the RLL is highly dependent on the designof the leak detection syst em, EPA's proposed rul erequires that ownersloperators calculate their ownsite-specific KLL values. EPA also proposes to

require tha t ownersloperators submit a RAP forl e a ka ge r a t e s e xc e e d i ng t ha t va l ue p r i o r t obeginning operation of a unit. T he EPA Regional

Administrator must approve the RAP before afacility can receive wastes.

T h e f o ll ow in g e q u a t i o n s r e p r e s e n t E P A 'spreli minar y atte mpt to define a range of potentialRLL values for a hypothetical leak detection system,which consists of a 1-foot granular drainage layerwith 1 cmlsec hydraul i c conduc t iv i ty . Thesecalculations ar e for two-dimensional rath er t ha n

three-dimensional flow. In addition, the equations

apply to flow from a single defect in th e FML, rath erthan multiple defects. Therefore, results from this

analysis are only pr eliminary ones, and the EPA willdevelop guidance on calculati ng RLL values in t he

near future.

RLL values can be calculated using the following

equation:

h = (Qd/B)/(kdtanS) (1

where: h =

Qd =

B =

kd =

P =

When the value

hydraulic head

flow r a t e e n t e r i n g i n t o t h e

drainage layerwidth of the dr ainage layer

hydraul i c conduc t iv i ty of th e

drainage layer

s lope o f t h e d r a i n a g e l a y e rperpendicular to, and in the planeof, flow toward t he collection pipe

for h exceeds the thickness of the

drainage layer (1 foot in th is example), the l eakag erate is greater tha n the RLL value for the unit.

In reality, a leak from a n isolated source, Le., a tear

or a hole in the FML, results in a discreet zone ofsaturation as the liquids flow toward the collectionpipe (see Figure 10-1). The appropriate variablerepre senting the width of flow, then, is not really B,the ent ire width of the drai nage layer perpendicularto flow, but b, the width of saturated flow perpen-dicular to the flow direction. If b were known, theequation could be solved. But to date, the data has

122



not been available to quantify b for all drainagelayers and leakage scenarios.

A ’

Leak

iea k

High Edge(Upgradient)

Collector Pipe -

Cross Section A . A1

Figure 10-1. Plan view of a leak detection system with a large

leak flowing over a width b.

From Equation 1, one can make substitutions for

variables B and Qd and give values for the othervariables kd and tanp . If N repres ents the frequencyof leak s in a well-designed and installed un it, then Q,the flow rate in the dra inag e layer (m31s) is directlyrelated to q, the leakage rate per unit ar ea (d se c) :

Q = NQor Q = q/N (2)

Combining Equations 1 and 2 and substituting b forB, and q for Q:

h = ql(Nbkdtanp) (3 )

Equation 3 now can be used to define the leakagerate (4) that exceeds the leak detection systemcapacity. All that is needed are the values €or theother variables ( N , kd, tanp). For a well-designed andinstalled unit, th e frequency of leaks ( N ) is 1 hole per

acre, or in units of m2; N = 114,000m2. Substituting

this value into Equation 3:

h = 4000q/(bkdtanp) (4)

Where q is in units of litersl1,OOO mzlday (Ltd),

Equation 4 can be wr itte n as follows:

The proposed rule requires leak detection systems tohave a minimum bottom slope of 2 percent (tanp) andminimum hydraulic conductivity of 10-2 d s e c (kd).Substitu ting these values into Equation 5:

h = 2.3 x 10 -4qh (6 )

where h is in un its of m, q is in unit s of Ltd, and b isin un its of m . For the purposes of these calcula tions,it is assumed that Ltd is equivalent to about 1gallacrelday. The final re sult s were derived by usingthre e different values for b (the unknown variable)and determining what values of q between 100 and10,000 gal/acre/day (Ltd) result in hydraulic headsexceeding the 1-foot thickness of the drainage layer

( h ) .

Table 10-2 shows the results of these preliminarycalculations. For values of q between 100 and 10,000

gallacrelday and values of b between 3 and 6 foot, thehydraulic head exceeds 1 foot when leak ra tes are inthe range of 2,000 to 10,000 gallacrelday. Therefore,RLL values for lea k detection system s consisting ofgranular drainage layer are expected to be in therange of 2,000 to 10,000 gallacrelday. Clogging of thedra inage layer would decrease the design capacity of

the leak detection system, and hence the RLL value,over time. With respect to the variables described

above, clogging of the drainage layer could berepresented using smaller values for b, the width ofsaturated flow, since clogging would result in areduced width of saturated flow. As shown in Table 10-2, smaller values of b reduce t he min imu m

leakage rate, q, needed to generate heads exceedingthe 1-foot thickness. E PA plans to issue guidance on

esti mat ing the effect of clogging on RLL values.

Table 10-2. Results of Preliminary Studies Defining Ranges

of RLL values

Width (b) Flow (q )f7 gallacrelday

3.3 1,000 - 2,000

5.0 2,000 - 5,000

6.6 5,000 10,000

123



Response Action Plans (RAPS)

According to the proposed leak detection rule, thekey elements of a RAP are :

0 General description of unit .

0 Description of waste constituents.

Description of all event s that may cause leakage.

0 Discussion of factors affecting amounts of leakage

enter ing LCRS.0 Design and operational mechanisms to prevent

leakage of hazardous const ituen ts.

Assessment of effectiveness of possible response

actions.

In developing a RAP, ownersfoperators of landfillsshould gather information from Part B of the permitapplication, available operational records, leachate

analysis results for exist ing facil i t ies , and the

c ons tr uct i on qua l i t y a s s u r a nc e r e po r t . T he

cons t ruct ion qual i ty assurance report i s veryimportant because it helps define where potential

leaks are likely to occur in the unit.

Sources of Liquids Other than Leachate

Depending on the unit design and location, other

liquids besides leachate could accum ulate in th e leak

detection system and result in apparent leak ratest h a t e xc ee d t h e A L R v a l u e . F o r e x a m p l e ,precipitation may pass through a tear in the FMLthat is located above the waste elevat ion (e.g. a tearin the FML a t a pipe pene tration point). The liquids

entering the leak detection system under thisscenario may not have contacted any wastes andhence would not be considered to be hazardous

leachate. In addition, rainw ater can become trappe d

in the drainage layer during construction andinstallation of the le ak detection system, but these

construction waters are typically flushed through

the system ear ly on in t he active life of the facility. Inthe case of a composite top liner, moisture from the

compacted soil component may be squeezed out overtime and also contribute to liquids collected in theleak detection sump. These sources of nonhazardous

liquids can add significant quantities of liquids to aleak detection system and might result in an ALRbeing exceeded. Therefore, these other sources ofliquids should also be considered when developing aRAP, and steps to verify th at cer tain liquids ar e not

hazardous should be outlined in the plan.

Ground-water permeation is one other possiblesource of nonhazardous liquids in the leak detection

system that can occur when the water table elevationis above the bottom of the uni t. The abi lity of groundwater to enter the leak detection sump, however,raises serious questions about the integr ity of the

bottom liner , which is the backup system in a double-

lined unit. If ground water is being collected in the

leak detection system, then hazardous constituentscould conceivably migr ate out of the landfill and into

the environment when the water table elevationdrops below the bottom of the uni t, e.g ., n th e case ofdry weather conditions. A s a result, while groundwater permeation is another source of liquids, it is

not a source that would ordinarily be used by theownerfoperator to justify ALR exceedances.

Preparing and Submitting the R A PResponse action plans must be developed for twobasic ranges: (1 ) leakage rates that exceed the R L Land (2) leakage rat es that equal or exceed th e ALRbut are less than the RLL. In submitting a RAP, afacility ownerloperator has two choices. First, the

ownerfoperator can submi t a plan to EPA before thefacility opens that describes all mea sures to be taken

for every possible leakage scen ario . The majordrawback to this option is th at the RAP may have tobe modified as specific l eak incidents occur, becausethe re ar e several variables th at affect the selection of

suitab le response actions. One variable is the time awhich the leak occurs. For example, if a leak is

discovered at the beginning of operation, the bestresponse might be to locate and repair the leak, sincethere would be little waste in the unit and the tea r o rhole may be easy to fix. If, however, a leak isdiscovered 6 months before a facility is scheduled to

close, it would probably make sense to close the unitimmedia te ly to min imize in f i l t ra t ion . I f the

ownerfoperator chooses to develop and submit oneRAP before the un it begins operation, he or she musdevelop suitable response actions for different leakrates and for different stages during the active life

and post-closure care period of the un it.

The second choice a n ownerloperator has is to sub mithe RAP in two phases: one RAP for the first rangeserious RLL leakage, tha t would be submitted beforethe sta rt of operation; and an other for th e second

range of leakage rate s (exceeding the ALR bu t les

tha n the RLL) tha t would be submit ted after a leakhas been detected.

EPA developed three generic types of response

actions tha t the ownerloperator must consider when

developing a RAP for leakage rates greater th an or

equal to the RLL. The three responses for veryserious leakage are straightforward:

0 Stop receiving waste and close the unit, or close

0 Repair t he leak or retrofit the top liner.

0 Instit ute operational cha nges

par t of the unit.

These three response actions also would apply toleakage rates less than RLL, although, as moderateto serious responses, they would apply to leakage

12 4



rates in the moderate to serious range, i.e., 500 to

2,000 ga l l ac re lday . For mos t l andf i l l s , 500gallacrelday leak rates would be considered fairlyserious, even though they may not exceed the RLL.In addition, clogging of the leak detection systemcould also result in serious leakage scenarios at rate sless tha n 2,000 gal/acre/day. For lower leak rate s ustabove the ALR, the best response would be promptlyto increase the l iquids removal f requency tominimize head on the bottom liner, analyze the

liquids, and follow up with progress reports.

Another key step in developing RAPs is to set upleakage bands, with each band representing aspecific range of leakage rates that requires aspecific response or set of responses. Table 10-3 shows an example of a RAP developed for threespecific leakage bands. The number and range ofleakage bands should be site-specific and take intoaccount the type of unit (i. e., surface impoundment,

landfill, waste pile), unit design, and operationalfactors.

Table 10-3.

ALR = 20 gal/acre/day and RLL = 2,500 gal/acre/day

Sample R AP for Leakage < RLL

Leakage Band(gallacreiday) Generic Response Action

20

20-250

Notify RA and identify sources of liquids

Increase pumping and analyze liquids insump.

250-2,500 Implement operational changes.

The RAP submitta l requir ements proposed by EPAdiffer for permitted facilities and interim statusfacilities. For newly permitted facilities, the RAP for

RLL must be submitted along with Part B of thepermit application. For existing facilities, the RAP

for RLL must be submitted as a request for permitmodification. Facil i t ies in interim status mustsubmit RAPs for RLL 120 days prior to th e receipt ofwaste.

If the RAP for low to moderate leakage (grea ter than

ALR but less than RLL) has not been submittedbefore operation, EPA has proposed that it must besubmitted within 90 days of detect ing a leak. In anycase, the EPA Regional Administrator’s approvalwould be requi red be fore tha t RAP can beimplemented.

Requ irements for Reporting a Leak

Once a leak has been detected, the proposedprocedure is similar for both ALR and RLL leakagescenarios. The ownerloperator would need to notifythe EPA Regional Administrator in writing within 7

days of th e date the ALR or RLL w a s determined to

be exceeded. The RAP should be implemented if it

ha s been approved (as in the case for RLL leaks), or

submit ted within 90 days for approval if not alreadysubmitted. Regardless of whether the RAP for theleak incident is approved, the ownerloperator wouldbe required to collect and remove liquids from theleak detection sump. Examples of the liquids should

be analyzed for leachate quality parameters, asspecified by the Regional Administrator in an

approved RAP. Both the need for analysis and the

parameters would be determined by the RegionalAdministrator.

In addition to the leachate sampling, the EPARegional Administrator would also specify aschedule for follow-up reporting, once the ALR o rRLL is exceeded. According to the proposed rule, thi sfollow-up reporting will include a discussion of the

response actions taken and t he change in leak ratesover time. The first progress report would besubmitted within 60 days of RAP implementation,and then periodically or annually, thereafter, asspecified in an approved RAP. Additional reporting

would also be required with in 45 days of detecting a

significant increase in the leak rate (an amountspecified in the RAP). This significant increase inleak rate indicates a failure in the response actionstaken and, therefore, may require modifications ofthe RAP and the implementation of other responseactions. These additional reporting and monitoringr e q u i r e m e n t s w o u l d b e p a r t o f t h e R A Pimplementation to be completed only when the

resu lting leak rat e drops below the ALR.

Summary

Although the overall containment system consistingof two liners and two LCRSs may achieve theperformance objective of preventing hazardous

constituent migration out of the unit for a period ofabout 30 to 50 years, the individual components maya t some point malfunction. Liners may l eak orLCRSIleak detection systems may clog during theactive life or post-closure care period. Therefore,EPA has developed and proposed requirements forearly response actions to be taken upon detecting amalfunction of the top liner or leak detection system.These requirements, once finalized, will ensuremaximum protection of human health and the

environment.

125



Abbreviations

= Action Leakage Rate= American Society for Testing and Materials

= biological oxygen demand

= degrees Centigrade= centim eters per second= chlorinated polyethylene

= Construction Quality Assurance= chlorylsulfonated polyethylene

= dielectric constant

= design ratio= differential scanning calorimetry

= destruct tests

= U.S. Envi ronm ental Protection Agency= ethyle ne propylene diene monomer= degrees Fahrenheit= Flexible Liner Evaluation Expert

= flexible membrane caps= flexible membrane liner

= factor of safety= feet

= squar e foot= cubicfoot= gallon per a cre per day

= high density polyethylene= Hydrologic Evaluat ion Landfill Performance Model= Hazar dous and Solid Waste Amendment

= leachate collection and removal system= leak detection, collection, and removal

= linear low density polyethylene

= litersl1,OOO mzlday

= square meters

= squar e meter s per second= minute= millimeters

= minimum technology guidance= minim um technological requ ireme nts= nondestruct tests

ALRASTM

BOD"Ccmlsec

CPE

CQACSPE

D

DRDSCDTEPA

EPDM"FFLEX

FMCFML

FSf tft2ft3gallacreldayHDPEHELPHSWA

LCRSLDCRLLDPE

Ltdm2

m2Isecminmm

MTGMTR

Requirement for HW Landfill De

Documents

Transcript of Requirement for HW Landfill De