EPA
UT.fed States
environmental Protection
Ayency
Mumc'pal Environmental Research ~~t\-"~' -
Laboratory S''pte', ':ei 1 ,
Cincinnati OH 45268
Research and Development
/._. '2JL..
Land Disposal of
Hazardous Waste
Proceedings of the
Ninth Annual
Research Symposium
Do not remove. This document
should be retained in the EPA
Region 5 Library Collection.
-------
EPA-600/9-83-018
September 1983
LAND DISPOSAL OF HAZARDOUS WASTE
Proceedings of the Ninth Annual Research Symposium
at Ft. Mitchell, Kentucky, May 2-4, 1983
Sponsored by the U.S. EPA, Office of Research & Development
Municipal Environmental Research Laboratory
Solid and Hazardous Waste Research Division
and
Industrial Environmental Research Laboratory
Industrial Pollution Control Division
Edited by: David W. Shultz
Coordinated by: David Black
Southwest Research Institute
San Antonio, Texas 78284
Cooperative Agreement CR-810437-01-0
Project Officer
Robert E. Landreth
Solid and Hazardous Waste Research Division
Municipal Environmental Research Laboratory
Cincinnati, Ohio 45268
MUNICIPAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
U.S. Env':-,-^'••"••"• ! -^ '•--,':on Agency
-------
DISCLAIMER
These Proceedings have been reviewed in accordance with the U.S.
Environmental Protection Agency's peer and administrative review policies and
approved for presentation and publication. Mention of trade names or
commercial products does not constitute endorsement or recommendation for use.
-------
FOREWORD
The Environmental Protection Agency was created because of increasing
public and governmental concern about the dangers of pollution to the health
and welfare of the American people. Noxious air, foul water, and spoiled
land are tragic testimony to the deterioration of our natural environment.
The complexity of the environment and the interplay between its components
require a concentrated and integrated attack on the problem.
Research and development is the first necessary step in problem solution;
it involves defining the problem, measuring its impact, and searching for
solutions. The Municipal Environmental Research Laboratory develops new and
improved technology and systems to prevent, treat, and manage wastewater and
the solid and hazardous waste pollutant discharges from municipal and com-
munity sources; to preserve and treat public drinking water supplies, and to
minimize the adverse economic, social, health and aesthetic effects of pollution,
This publication is one of the products of that research—a vital communications
link between the researcher and the user community.
These Proceedings present the results of completed and ongoing research
projects concerning the land disposal of hazardous waste.
Francis T. Mayo
Director
Municipal Environmental
Research Laboratory
iii
-------
PREFACE
These Proceedings are intended to disseminate up-to-date information on
extramural research projects concerning land disposal, incineration, and
treatment of hazardous waste. These projects are funded by the U.S. Environ-
mental Protection Agency's Office of Research and Development and have been
reviewed in accordance with the requirements of EPA's Peer and Administrative
Review Control System.
xv
-------
ABSTRACT
The Ninth Annual Research Symposium on land disposal, incineration, and
treatment of hazardous wastes was held in Ft. Mitchell, Kentucky, May 2,
3 and 4, 1983. The purposes of the symposium were (1) to provide a forum for
a state-of-the-art review and discussion of ongoing and recently completed re-
search projects dealing with land disposal, incineration, and treatment of
hazardous wastes; (2) to bring together people concerned with hazardous waste
management who can benefit from an exchange of ideas and information; and
(3) to provide an arena for the peer review of the Solid and Hazardous Waste
Research Division's and the Industrial Pollution Control Division's research
programs in hazardous waste management. These Proceedings are a compilation
of papers presented by the symposium speakers.
The symposium proceedings are being published as two separate documents.
In this document, Land Disposal of Hazardous Waste, seven technical areas are
covered. They are as follows:
(1) Land Disposal Research Overviews
(2) Landfill Design and Operation
(3) Pollutant Movement
(4) Pollutant Control-Liners
(5) Waste Modification
(6) Cost/Economics
(7) Remedial Action
-------
CONTENTS
SESSION A: HAZARDOUS WASTE LAND DISPOSAL
LAND DISPOSAL RESEARCH OVERVIEW
Technical Resource Documents for Hazardous Wastes
Norbert B. Schomaker and Ihor Melnyk
U.S. Environmental Protection Agency
Current Research on Land Disposal of Hazardous Waste
Norbert B. Schomaker and Ihor Melnyk
U.S. Environmental Protection Agency
Summary of the First Year of Operation
Elvin J. Dantin, Hazardous Waste Research Center
Louisiana State University 22
LANDFILL DESIGN AND OPERATION
A Model for Hydrologic Evaluation of Landfill Performance
Paul R. Schroeder, USAE Waterways Experiment Station
Michael D. Smolen, Virginia Polytechnic Institute
and State University 29
Geotechnical Quality Assurance of Construction of Disposal Facilities
S. Joseph Spigolon and Michael F. Kelley
USAE Waterways Experiment Station 45
POLLUTANT MOVEMENT
Secondary Emissions from Hazardous Waste Disposal Lagoons: Field
Measurements
Charles Springer, Louis J. Thibodeaux, Phillip D. Lunney,
Rebecca S. Parker, University of Arkansas
Stephen C. James, U.S. Environmental Protection Agency 58
Investigation of Clay Soil Behavior and Migration of Industrial
Chemicals at Wilsonville, Illinois
R. A. Griffin, Keros Cartwright, P. B. DuMontelle, L. R. Follmer,
C. J. Stohr, M. T. Johnson, M. M. Killey, R. E. Hughes, B. L. Herzog
W. J. Morse, Illinois State Geological Survey 70
vii
-------
Groundwater and Leachate Plumes: A Model Study
Wayne A. Pettyjohn, Douglas C. Kent, Jan Wagner,
and Arthur W. Hounslow, Oklahoma State University 80
Procedures and Techniques for Controlling the Migration of Leachate
Plumes
Charles Kufs, Kathi Wagner, Paul Rogoshewski, Marjorie Kaplan,
and Edward Repa, JRB Associates 87
POLLUTANT CONTROL - LINERS
The Influence of Selected Organic Liquids on the Permeability of Clay
Liners
K. W. Brown, J. W. Green and S. C. Thomas
Texas Agricultural Experiment Station 114
Laboratory Studies for Determining Flexible-Membrane Soil Bedding
Requirements
Donald M. Ladd and Gordon L. Carr
USAE Waterways Experiment Station 126
Applications of Geotextiles in Land Waste Disposal Sites
R. C. Horz and V. H. Torrey, USAE Waterways Experiment Station .... 139
Potential Mechanisms and Remedies for Clogging of Leachate Drain Systems
Jeffrey M. Bass, Raymond M. Cornish, John R. Ehrenfeld,
Stephen P. Spellenberg and James R. Valentine
Arthur D. Little, Inc 148
Analysis and Fingerprinting of Unexposed and Exposed Polymeric Membrane
Liners
Henry E. Haxo, Jr., Matrecon, Inc 157
Evaluation of a Waste Impoundment Liner System After Long-Term
Exposure
Susan A. Roberts, Malcolm Pirnie, Inc.
Nancy A. Nelson and Henry E. Haxo, Jr., Matrecon, Inc 172
Evaluations of Time-Domain Reflectometry and Acoustic Emission Techniques
to Detect and Locate Leaks in Waste Pond Liners
J. L. Davis, M. J. Waller, B. G. Stegman and R. Singh
EarthTech Research Corporation 188
Pilot Scale Verification of a Liner Leak Detection System
Wendell R. Peters and David W. Shultz
Southwest Research Institute 203
Development of Liner Retrofit Concepts for Surface Impoundments
John W. Cooper and David W. Schultz
Southwest Research Institute 211
vi n
-------
WASTE MODIFICATION
Long-Term Impacts of Hazardous Waste Codisposal in Landfill Simulators
James J. Walsh and W. Gregory Vogt, SCS Engineers
Riley N. Kinman and Janet I. Rickabaugh, University of Cincinnati. . . 219
Application of Solidification/Stabilization Technology to Electroplating
Wastes
Philip G. Malone, Larry W. Jones and J. Pete Burkes
USAE Waterways Experiment Station 247
COST/ECONOMICS
Analyzing the Costs of Remedial Actions
Edward J. Yang and James D. Werner, Environmental Law Institute. . . . 262
Costs of Remedial Actions at Uncontrolled Hazardous Waste Sites
James J. Walsh, John M. Lippitt, Michael P. Scott and
Anthony J. DiPuccio, SCS Engineers 271
Development of a Framework for Evaluating Cost Effectiveness Actions
at Uncontrolled Hazardous Waste Sites
Ann E. St. Clair, Michael H. McCloskey, and James S. Sherman
Radian Corporation 286
REMEDIAL ACTION
Survey and Case Study Investigation of Remedial Actions at Uncontrolled
Hazardous Waste Sites
S. Robert Cochran, Marjorie Kaplan, Paul Rogoshewski, Claudia Furman
JRB Associates 293
Field Scheme for Determination of Waste Reactivity Groups
Ursula Spannagel, Richard Whitney, Dean Wolbach, Acurex Corporation. . 304
Demonstration of the Block Displacement Method at Whitehouse, Florida
Thomas P. Brunsing, Foster-Miller, Inc 327
The Impact of Top-Sealing of the Windham Connecticut Landfill
Rudolph M. Schuller, Alison L. Dunn, SMC Martin, Inc.
William W. Beck, Jr., TRC Environmental Consultants, Inc 334
Evaluation of Remedial Action Alternatives--Demonstration/Application of
Groundwater Modeling Technology
F. W. Bond and C. R. Cole, Battelle, Pacific Northwest Laboratories
D. Sanning, U.S. Environmental Protection Agency 343
Slurry Trench Construction of Pollution Migration Cut-Off Walls
Philip A. Spooner and Roger S. Wetzel, JRB Associates
Walter E. Grube, Jr., U.S. Environmental Protection Agency . 356
-------
Handbook for Evaluating Remedial Action Technology Plans
John R. Ehrenfeld and Jeffrey M. Bass, Arthur D. Little, Inc 373
Collection of Information on the Compatibility of Grouts with Hazardous
Wastes
Gary E. Hunt, Virginia E. Hodge, Pauline M. Wagner, and
Philip A. Spooner, JRB Associates
Herbert R. Pahren, U.S. Environmental Protection Agency 384
Remedial Action Research: Wastewater Lagoons
Elly K. Triegel and Joseph R. Kolmer
Woodward-Clyde and Associates 397
A Literature Survey of Three Selected Hazardous Waste Destruction
Techniques
Danny D. Reible and David M. Wetzel, Louisiana State University. . . . 408
Drum Handling Practices at Abandoned Sites
Roger Wetzel and Kathleen Wagner, JRB Associates
Anthony N. Tafuri, U.S. Environmental Protection Agency 414
List of Attendees 427
-------
TECHNICAL RESOURCE DOCUMENTS FOR HAZARDOUS WASTES
Norbert B. Schomaker
Ihor Melnyk
U.S. Environmental Protection Agency
26 West St. Clair Street
Cincinnati, Ohio 45268
ABSTRACT
The Environmental Protection Agency is preparing documents for permit officials
responsible for hazardous waste landfills, surface impoundments, land treatment
facilities and piles. The Technical Resource Documents (TRDs) describe current
technologies and methods for evaluating the performance of the applicant's design in the
areas of liners, leachate management, closure, covers, and water balance. These
documents are basically a compilation of all research efforts of the Solid and Hazardous
Waste Research Division (SHWRD) to date and are being published or developed to assist
in the implementation of the regulations concerning hazardous waste disposal
facilities. The RCRA Guidance Documents being prepared by the Office of Solid Wastes
(OSW) will present design and operating specifications which the Agency believes comply
with Design and Operating Requirements and the Closure and Post-Closure Requirements
contained in Part 264 of the regulations. The information and guidance presented in
these documents constitute a suggested approach for review and evaluation based on good
engineering practices. There may be alternative and equivalent methods for conducting
the review and evaluation. However, if the results of these methods differ from those
of the Environmental Protection Agency's method, they may have to be validated by the
applicant.
INTRODUCTION
Land disposal of hazardous waste is
subject to the requirements of subtitle C
of the Resource Conservation and Recovery
Act of 1976. This Act requires that the
treatment, storage, or disposal of hazard-
ous waste be carried out in accordance
with a permit. Owners or operators of new
facilities must apply for and receive a
permit before beginning operation of such
a facility.
In reviewing and evaluating the permit
application, the permit official must make
all decisions in a well defined and well
documented manner. Once an initial deci-
sion is made to issue or deny a permit,
the subtitle C regulations require prepa-
ration of either a statement of basis or a
fact sheet that discusses the reasons
behind the decision, and becomes part of
the permit review process specified in 40
CFR 124.6-124.20.
Eight TRDs have been completed to
date. Six of the TRDs are related to
landfills, and one document each is
related to surface impoundments and land
treatment. A listing of these documents
is shown below, along with a brief
description, publication number, and the
project officer's name in parentheses.
TRD 1: EVALUATING COVER SYSTEMS FOR SOLID
AND HAZARDOUS WASTE (SW-867) - This
document presents a procedure for evalu-
ating cover systems for solid and
hazardous waste. It contains a checklist,
with supporting documentation, that a
permit writer can use to evaluate proposed
cover systems. *055-000-00228-2 (R. E.
Landreth)
TRD 2: HYDROLOGIC SIMULATION ON SOLID
WASTE DISPOSAL SITES (SW-868) - This
document provides a computer package to
aid planners and designers by simulating
hydrologic characteristics of landfill
-------
covers to predict percolation as a
function of cover design and climate.
*055-000-00225-8 (D. C. Ammon)
TRD 3: LANDFILL AND SURFACE IMPOUNDMENT
PERFORMANCE EVALUATION (SW-869) - This
document describes how to evaluate
leachate collection systems using
compacted clay or synthetic liners to
determine how much leachate will be
collected and how much will seep through
the liner into underlying soils. The
adequacy of sand and gravel drain layers,
slope, and pipe spacing are also covered.
*055-000-00233-9 (M. H. Roulier)
TRD 4: LINING OF WASTE IMPOUNDMENT AND
DISPOSAL FACILITIES (SW-870) - This
document provides information and guidance
on liner systems. It discusses waste,
liner types, compatibility, liner
selection, specifications, design of
leachate collection systems, and case
study analysis methodology. It also
includes a glossary of liner system
related terms.
*055-000-00231-2 (R. E. Landreth)
TRD 5: MANAGEMENT OF HAZARDOUS WASTE
LEACHATE (SW-871) - This document
discusses leachate composition, leachate
generation, selected management options,
available treatment technologies,
unit-treatment options, and system
economics. It presents management options
that a permit writer or hazardous waste
landfill operator may consider in
controlling and treating leachate.
*055-000-00224-0 (S.C. James)
TRD 6: GUIDE TO THE DISPOSAL OF
CHEMICALLY STABILIZED & SOLIDIFIED WASTES
(SW-872) - This document provides basic
information on stabilization/
solidification of industrial wastes to
*Drafts of the above documents have
been released for public comment and all
have been revised so as to incorporate the
public's comments and reviews. They are
currently being published and the reports
will be available from GPO by requesting
the stock number. Copies can be obtained
for a price from the:
Superintendent of Documents
U.S. Government Printing Office
Washington, DC 20402
(202) 783-3238
insure safe burial of wastes containing
harmful materials. A summary of major
physical and chemical properties of
treated wastes is presented along with a
listing of major suppliers of
stabilization/solidification technology
including a summary of each process.
*055-000-00226-6 (R. E. Landreth)
TRD 7: CLOSURE OF HAZARDOUS WASTE SURFACE
IMPOUNDMENTS (SW-873) - This document
discusses and references the methods,
tests, and procedures involved in closing
a surface impoundment. Problems of
abandoned impoundments causing
environmental degradation and discussed.
Closure methods such as waste removal and
consolidating the waste on site and
securing the site as a landfill are also
discussed.
*055-000-00227-4 (M. H. Roulier)
TRD 8: HAZARDOUS WASTE LAND TREATMENT
(SW-874) - This document discusses land
treatment medium, characteristics of
hazardous waste, allowable waste loadings,
facility planning for land treatment, land
treatment permit processes, closure, and
items to be considered during permit
evaluation.
*055-000-00232-1 (C.C. Wiles)
Each of the completed TRDs is
described in greater detail in the
following section of this paper.
TECHNICAL RESOURCE DOCUMENTS CURRENTLY
AVAILABLE
Evaluating Cover Systems for Solid and
Hazardous Waste (SW-867)
A critical part of the sequence of
designing, constructing, and maintaining
an effective cover over solid and
hazardous waste is the evaluation of
engineering plans. This TRD presents a
procedure for evaluating closure covers on
solid and hazardous wastes. All aspects
of covers are addressed in detail to allow
for a complete evaluation of the entire
cover system.
Covers may be evaluated for confor-
mance to requirements for closure covers
by the following eleven (11) general
procedures in sequence:
1) examine soil test data,
2) examine topography,
3) examine climate data,
-------
4)
5)
6)
7)
8)
9)
10)
11)
evaluate
evaluate
evaluate
evaluate
evaluate
evaluate
evaluate
evaluate
composition,
thickness,
placement,
configuration
drainage,
vegetation,
post-closure plan, and
contingencies plan.
This document describes current
technology for landfill covers. For a
landfill, these state that: 1) runoff
that has touched hazardous waste must be
collected and treated, 2) waste received
must be analyzed, 3) each waste cell's
contents, depth, location, and dimensions
must be documented, and 4) a closure and
post-closure plan is required to address
control of pollutant migration, surface
water infiltration, and erosion.
Guidance is provided on:
characteristics and type of waste,
location of fill with respect to potential
environmental impact of pollutant migra-
tion, climate and precipitation, cover
material and thickness, cover slope and
vegetation, maintaining the final cover,
maintaining groundwater, monitoring the
gas collection system, and restricting
access to the landfill. There are 36
specific steps, regarding the preceding
factors, which should be followed in
evaluating a permit for a cover for
hazardous waste.
Hydrologic Simulation on Solid waste
Disposal Sites (Sw-868)
Percolation and runoff of precipita-
tion at solid waste disposal sites create
a potential for ground and surface water
contamination with leachate. The proper
landfill site design and operational
approach is to minimize or eliminate
percolation through the solid waste and
thus control the formation of leachate.
The purpose of this TRD is to provide
an interactive computer program for simu-
lating the hydrologic characteristics of a
solid and hazardous waste disposal site
operation. A large number of cities
within the U.S. for which five years of
climatic records exist have been put on
tape for easy access and can be used in
lieu of on-site measurements. In addi-
tion, the model stores many default values
of parameter estimates which can be used
when measured and existing data files are
not available. The user must supply the
geographic location, site area and hydro-
logic length, the characteristics of the
final soil and vegetative cover, and
default overrides where deemed necessary.
From minimal input data, the model can
simulate daily, monthly, and annual
runoff, deep percolation, temperature,
soil water, and evapotranspiration.
The model, which is a modification of
the Soil Conservation Service (SCS) number
method for predicting runoff and the
hydrologic portion of the U.S. Department
of Agriculture, Science and Education
Administration (USDA-SEA) hydrologic model
entitled "Chemicals, Runoff, and Erosion
From Agricultural Management Systems
(CREAMS)," has been modified to conform to
the design characteristics of solid and
hazardous waste disposal sites. The model
takes engineering, hydrologic, and
climatologic input data in the form of
rainfall, average temperature, solar
radiation, and leaf area indices and
characteristics of cover material and
performs a sequential analysis to derive a
water budget including runoff,
percolation, and evapotranspiration.
The user can specify up to three soil
layers and may also specify a membrane
liner at the base of the cover. The
decreasing effectiveness of the liner is
simulated. The model is designed for use
in a conversational manner, that is, the
user interacts directly with the program
and receives output immediately. No prior
experience with computer programming is
required. All commands necessary to use
the model are presented in the user's
manual. The model can also be run in
batch mode, which requires some computer
programming experience.
Landfill and Surface Impoundment
Performance Evaluation (SW-869)
This TRD describes how to evaluate the
capability of various liner and drain
designs to control leachate release. The
author has allowed for the widely varied
technical backgrounds of his intended
audience by presenting, in full, the
rigorous mathematics involved in reaching
his final equations. Thus, any evaluator
can take full advantage of the manual up
to the level of his own mathematical
proficiency.
This evaluation system considers the
transport of liquids through simple
-------
modular configurations. The important
parts of these modules are identified as
the Liquid Transport Control (LTC) system,
and the Diversion Interface (DI). The
mathematics consider the efficiency with
which the LTC and DI components convert
leachate flow from vertical to horizon-
tal. The extent of ponding on a bottom
liner is considered in relation to the
position and effectiveness of the drainage
system.
The procedures described in the
document should allow an evaluator to
determine the adequacy of design for:
1) compacted clay liners or synthetic
liners intended to impede the vertical
flow of fluids,
2) sand or gravel drainage layers
intended to convey liquids laterally
into collection systems,
3) slopes on such liner systems, and
4) spacings of collector drains.
The evaluation procedure is indepen-
dent of the structure of the regulation it
supports. If regulations specify
acceptable rates of release for leachate,
or efficiencies of leachate collection
systems, calculation results may be used
directly to determine if the landfill is
acceptable. However, if the regulations
describe acceptable environmental impact,
results from the evaluation procedure can
be used as input for further evaluation in
other manuals to determine if the impact
of the landfill is acceptable,
Lining of Waste Impoundment and Disposal
Facilities (SW-870)
This document provides information on
performance, selection, and installation
of specific liners and cover materials for
specific disposal situations, based upon
the current state-of-the-art of liner
technology and other pertinent technolo-
gies. It contains descriptions of wastes
and their effects on linings, a full
description of various natural and
artificial liners; service life and
failure mechanisms; installation problems
and requirements of liner types; costs of
liners and installation; and tests that
are necessary for pre-installation and
monitoring surveys.
Described in the report are poten-
tially useful lining materials and their
associated technologies, including
re-molded and compacted soils and clays,
admixes, polymeric membrane liners,
sprayed-on liners, soil sealants, and
chemisorptive liners. The specific
characteristics of these liners in service
environments with various types of wastes
are discussed, particularly with respect
to compatibility with wastes, permeability
to water and waste, failure mechanisms,
and service life.
Design and construction of waste
facilities with various types of linings
are described, emphasizing installation of
membrane liners and their problems.
Special problems associated with the
design and construction of solid waste
disposal sites are considered. The
subjects of leachate collection above the
liner, gas control, and top-seals are
addressed.
Several processes for selecting a
specific liner or various satisfactory
liners for a given disposal site are
proposed. These include the use of
compatibility tests, moderate duration
exposure tests, soil conditions tests,
costs, etc. Methods for the preparation
of specifications covering the selected
liners or groups of liners are suggested.
In the appendices, data are presented
on liners and wastes, compatibility
studies, manufacturers and suppliers,
specifications, test methods, along with
indices to lined disposal sites. As new
technology is developed, and as experience
in the use of liners becomes available,
the manual will be upgraded on a
chapter-by-chapter basis.
Management of Hazardous Waste Leachate
(SW-871)
This document has been prepared to
provide guidance for permit officials and
disposal site operators on available
management options for controlling,
treating, and disposing of hazardous waste
leachates. It discusses considerations
necessary to develop sound management
plans for leachate generated at surface
impoundments and landfills. Management
may take the form of leachate collection
and treatment, or pre-treatment of the
waste.
-------
The manual addresses areas that must
be considered in evaluating a leachate
management option: leachate generation and
composition, and available technologies
for treatment of leachate and hazardous
waste.
Leachate composition is examined for
several hazardous waste landfills, syn-
thetic leachates from lab studies, and
sanitary landfill leachates. The expected
composition of the leachate and the con-
centrations of the pollutants in the
leachate are presented along with possible
guidelines for permit writers to use in
determining the relative hazards of a
particular leachate.
A key section enumerates factors which
influence treatment process selections and
provides a suggested approach for system-
atically addressing each. Selected
hypothetical and actual leachate
situations are used as examples for
applying the approach to the selection of
appropriate treatment processes.
Other sections address monitoring,
safety, contingency plans/emergency
provisions, equipment redundancy/backup,
permits, and surface runoff.
Guide to the Disposal of Chemically
Stabilized and Solidified Haste"T5tr872)
Stabilization/sol id ification of
industrial waste is a pre-treatment
process designed to insure safe disposal
of wastes containing harmful materials.
The purpose of this TRD is to provide
guidance in the use of chemical
stabilization/solidification techniques
for limiting hazards posed by toxic wastes
in the environment, and to assist in the
evaluation of permit applications related
to this disposal technology. The document
addresses the treatment of hazardous waste
for disposal or long term storage and
surveys the current state and effective-
ness of waste treatment technology. A
summary of the major physical and chemical
properties of treated wastes is pre-
sented. A listing of major suppliers of
stabilization/solidification technology
and a summary of each process is included.
This guide provides the background
information needed for waste generators
and regulatory officials to determine the
testing program and/or product information
necessary for them to make the best
engineering judgements concerning the long
term effectiveness in site specific
conditions. Some familiarity with general
soil characteristics, water balance,
climatic conditions, and fundamentals of
leachate generation would be helpful to
the user.
Closure of Hazardous Waste Surface
Impoundments (S
"Closure of Hazardous Waste Surface
Impoundments" presents a listing of
closure plan and post-closure care
considerations and details for surface
impoundments containing hazardous wastes.
It is written primarily for staff members
in EPA Regional Offices or state regula-
tory offices, who are charged with
evaluating and approving closure plans for
surface impoundments under regulations of
the Resource Conservation and Recovery Act
of 1976. Methods of assessing site
closure considerations are documented.
The guide describes and references the
methods, tests, and procedures involved in
closing a site in such a manner that (a)
minimizes the need for further mainte-
nance, and (b) controls, minimizes, or
eliminates, to the extent necessary to
protect human health and the environment,
post-closure escape of hazardous waste,
waste constituents, leachate, contaminated
rainfall or waste decomposition products
to groundwater, surface water, or the
atmosphere. Problems that have been over-
looked in abandoned impoundments and have
caused environmental damage are dis-
cussed. The techniques involved are
pertinent to closing an impoundment either
by removing the hazardous waste or by
consolidating the waste on-site and
securing the site as a landfill. Techni-
cal criteria for implementing the closure,
specifically those regarding aspects
substantially different from a landfill,
are given. Relevant literature or proce-
dures are documented for more in depth
review.
Hazardous Waste Land Treatment (SW-874)
Land treatment, which involves using
the surface soil as the treatment medium,
is already widely practiced by some
industries for handling their hazardous
wastes. However, the lack of systematic
study or monitoring of many past or
present facilities has limited the knowl-
edge and data base with regard to the
-------
important parameters and interactions.
Additionally, many potentially land
treatable wastes have not been tested or
have been examined only under a limited
range of conditions. Therefore, informa-
tion is needed on site selection, waste
characteristics, optimum soil and climatic
conditions for degradation of organic
constituents, application rates and
scheduling decomposition products, and the
contingency for environment
contamination. Potential problems include
worker health and safety, organic
components, water pollution hazards, and
local atmospheric emissions.
Land treatment of hazardous waste has
not been under close scrutiny for a long
enough period of time to acquire numerous
documented failures resulting in environ-
mental pollution. The potential for such
failures due to the lack of proper design
and/or management is evident from research
conducted on the land treatment of other
non-hazardous waste materials. For land
treatment to be an effective system, the
process must operate within a narrow range
of parameters. Exceeding them could
result in the uncontrolled release of
pollutants to the environment. The
objectives of "Hazardous Waste Land
Treatment" are to describe current
technology and to provide methods for
evaluating the performance of an
applicant's hazardous waste land treatment
facility design. Land treatment is
approached comprehensively from initial
site selection through final closure, and
additional informational sources are
referenced liberally.
FUTURE TECHNICAL RESOURCE DOCUMENTS
Additional TRDs now being developed or
in planning stages by Office of Research
and development:
A listing of these documents is shown
below, along with a brief description and
the project officers name in parentheses.
1. SOIL PROPERTIES, CLASSIFICATION AND
HYDRAULIC CONDUCTIVITY TESTING
This report is a compilation of
available laboratory and field testing
methods for the measurement of hydraulic
conductivity (permeability) of soils.
Background information on soil classifica-
tion, soil water, and soil compaction are
included along with descriptions of
sixteen methods for determination of
saturated or unsaturated hydraulic
conductivity.
(M. H. Roulier)
2. SOLID WASTE LEACHING PROCEDURE MANUAL
This project is developing laboratory
batch procedures for extracting or leach-
ing a sample of solid waste so that the
composition of the lab leachate is similar
to the composition of leachate from waste
under field conditions.
(M. H. Roulier)
3. BATCH SOIL PROCEDURE TO DESIGN CLAY
LINERS FOR POLLUTANT REMOVAL
This project is studying the feasibi-
lity of using soil/leachate batch
adsorption procedures for designing
compacted clay liners to remove/retain
known amounts of pollutants, and will
contribute to guidance for designing and
evaluating clay liners to limit pollutant
release from landfills.
(M. H. Roulier)
4. METHODS FOR THE PREDICTION OF LEACHATE
PLUME MIGRATION AND MIXING
This project is developing a variety
of computer programs, nearly all of which
are designed for hand-held calculators and
microcomputers. The programs permit
prediction of leachate plume migration and
groundwater recharge. The document also
contains discussions of sorption, case
historical and a field study.
(M. H. Roulier)
5. HYDROLOGIC EVALUATION OF LANDFILL
PERFORMANCE (HELP) MODEL
The original waste disposal site
hydrologic model entitled, "Hydrologic
Simulation on Solid Waste Disposal Sites"
is being modified. This update will
incorporate the two-dimensional aspects of
landfill cover systems, as well as the
addition of the leachate collection
system. This second generation model will
be entitled, "Hydrologic Evaluation of
Landfill Performance (HELP) Model."
(D. C. Ammon)
CONCLUSION
The TRDs that are being prepared by
SHWRD are a series of documents which
represent the Best Engineering Judgement
-------
(BEJ) for the design, operation, and
closure of hazardous waste facilities.
Eight documents have been completed with
five more either in process or planned by
the Office of Research and Development.
These documents have been directed
primarily for use by permit writers for
evaluating facility designs and potential
performance of new waste disposal
facilities. The TRDs present the sum
total of the body of information and
experience gained by the Agency over the
years on a given topic. As such, they
contain factual summaries concerning the
experience and effectiveness of design
alternatives, covering what has been found
not to work as well as what has been found
to be effective. The TRDs can be
considered as technical background or
development documents supporting the RCRA
guidance documents and regulations.
The project officers can be contacted
by writing or telephoning the: U. S. EPA,
Municipal Environmental Research
Laboratory, Solid and Hazardous Waste
Research Division, 26 West St. Clair
Street, Cincinnati, Ohio 45268, Telephone
(513) 684-7871.
-------
CURRENT RESEARCH ON LAND DISPOSAL OF HAZARDOUS WASTES
Norbert B. Schomaker
Ihor Melnyk
U.S. Environmental Protection Agency
26 West St. Clair Street
Cincinnati, Ohio 45268
ABSTRACT
The Solid and Hazardous Waste Research Division (SHWRD), Municipal Environmental
Research Laboratory, U.S. Environmental Protection Agency, in Cincinnati, Ohio, has
responsibility for research in solid and hazardous waste management with respect to land
disposal of wastes. To fulfill this responsibility, the SHWRD is developing concepts and
technology for new and improved systems of solid and hazardous waste land disposal; is
documenting the environmental effects of various waste disposal practices; and is
collecting data necessary to support implementation of disposal guidelines mandated by
the "Resource Conservation and Recovery Act of 1976 (RCRA)" PL 94-580. Also, SHWRD is
collecting data on existing construction technologies and applying these techniques to
the containment of pollutants emanating from uncontrolled dump sites. This is being done
to assist in the implementation of the "Comprehensive Environmental Response,
Compensation and Liability Act of 1980" (CERCLA) PL 96-510. This paper will present an
overview of the land disposal aspects of the SHWRD Hazardous Waste Program Plan and will
report the current status of work in the following categorical areas:
A. LANDFILLS
1. Pollutant Generation
2. Pollutant Movement
3. Pollutant Control
4. Pollutant Treatment
B. LANDFILL ALTERNATIVES
1. Surface Impoundments
2. Underground Mine Disposal
C. UNCONTROLLED SITES/REMEDIAL ACTION
D. COST/ECONOMIC ASSESSMENT
INTRODUCTION
The waste residual disposal research
strategy, encompassing state-of-the-art
documents, laboratory analysis, bench and
pilot studies, and full-scale field
verification studies is at various stages
of implementation. Over the next 2 years
the research will be reported as guidance
documents for user communities. The waste
disposal research program is currently
developing and compiling a data base for
use in the development of guidelines and
standards for waste residual disposal to
the land as mandated by the "Resource
Conservation and Recovery Act of 1976"
(RCRA). Permit Writer's Guidance Manuals,
which will present design and operating
specifications which the Agency believes
comply with design and operating
requirements and the closure and post
closure requirements contained in Part 264
of the regulations are currently being
prepared by the Office of Solid Waste in
Washington, D.C. Technical Resource
Documents (TRD) in support of the Guidance
Manuals are also being prepared by the
Office of Research and Development in
-------
specific areas to provide current
technologies and methods for evaluating
the performance of the applicants'
design. A separate paper entitled,
"Technical Resource Documents for
Hazardous Wastes" has been accepted for
presentation at the Ninth Annual Research
Symposium on Land Disposal, Incineration
and Treatment of Hazardous Wastes.
Therefore, no discussion of the TRDs will
be presented in this paper.
Current hazardous waste land disposal
research has been divided into four
general areas: (a) Design Considerations
for Current Landfill Disposal Techniques;
(b) Alternatives to Current Landfill
Disposal Techniques; (c) Remedial Action
for Minimizing Pollutants from
Uncontrolled Sites; (d) Economic
Assessment of Hazardous Waste Disposal
Practices and Alternatives.
The waste residual research program
has been discussed in the previous eight
symposia. These symposia describe the
land disposal of both municipal and
hazardous waste. The report of the
previous most recent symposium is entitled
"Research on Land Disposal of Hazardous
Waste: Proceedings of the Eighth Annual
Research Symposium," March 8-10, 1982,
Fort Mitchell, Kentucky, EPA-600/9-82-002.
LANDFILLS
Pollutant Generation
The overall objective of this research
activity is to address techniques for
leaching a waste sample to obtain leachate
similar to that which will occur when the
waste is landfilled. A leaching procedure
is being developed for use in determining
which pollutants will be of concern in
movement of leachates out of the landfill
and for evaluating leachate treatment
schemes in order to better develop
pollutant control technology. A Technical
Resource Document (TRD) entitled "Landfill
and Surface Impoundment Performance
Evaluation" (SW-869) reflects current
state-of-the-art in this area and is
available from the Government Printing
Office (055-000-00233-9).
One completed effort (1)* in the waste
leaching area was directed toward
applicable analytical methods for leachate
analysis. Physical and chemical methods
were evaluated for their suitability in
analyzing of contaminated industrial
leachate streams. The results of this
completed effort were published in a
report entitled "Analytical Methods
Evaluation for Applicability in Leachate
Analysis," EPA-600/2-81-046, March 1981.
A second ongoing effort (2) is
developing a detail laboratory procedure
for extracting or leaching a sample of
solid waste so that the composition of
laboratory leachate is similar to the
composition of leachate from the waste
under field conditions. It is anticipated
that this effort will be identified as a
Technical Resource Document (TRD) entitled
"Solid Waste Leaching Procedure Manual"
with anticipated publication late in FY
1983. A subsequent study will be
implemented (2) in FY 1983 to compare
batch and column procedures for waste
leaching.
A third ongoing effort (7) was the
development of a two-dimensional surface
drainage system evaluation model for
hazardous waste disposal facilities. A
Technical Resource Document entitled
"Hydrologic Simulation on Solid Waste
Disposal Sites" (SW-868) reflects the
current state-of-the-art knowledge in this
area and is available from the Government
Printing Office (055-000-00225-8). In
addition to the development of this model,
a detailed evaluation of various designs
was conducted to assist in the design of
numerical performance information for
cover systems after closure and for
leachate collection systems. The
completed first phase of this project
includes four tasks: 1) evaluation of
subsurface drainage system designs for
establishment of numerical performance
standards; 2) development of user oriented
two-dimensional subsurface drainage system
evaluation capabilities; 3) user support
for the current version of the Hydrologic
Simulation on Solid Waste Disposal Sites;
and 4) documentation, report preparation,
and technical coordination. A draft TRD
report entitled "Hydrologic Evaluation of
Landfill Performance (HELP) Model" has
Parenthesis numbers refer to the
project officers who is monitoring
this effort and whose name is listed
immediately following this paper. The
project officer can be contacted for
additional information.
-------
been prepared with anticipated publication
in late FY 1983. The second phase effort
for additional technical support and model
verification is now underway.
A fourth completed effort (4) is a
study of the effectiveness of cover
materials to prevent/control infiltration
of water into waste disposal sites.
Covers made of synthetic materials,
natural materials, and combinations of
both are being studied for their
effectiveness. Disturbances to covers
such as vegetation growth, burrowing
animals, and varying climatic conditions
are being considered. The final report
entitled "Design and Construction of
Covers for Solid Waste Landfills" is
available (EPA 600/2-79-165) August 1979.
A TRD entitled "Evaluating Cover Systems
for Solid and Hazardous Waste" (SW-867)
reflects the current state-of-the-art in
this area and is available from the
Government Printing Office
(055-000-00228-2).
Co-Disposal: The overall objective of the
co-disposal activity is to assess the
impact of the disposal of industrial waste
materials with municipal solid waste.
Since the environmental effects from
landfilling result from not only the
soluble and slowly soluble materials
placed in the landfill but also the
products of chemical and microbiological
transformation, these transformations
should be a consideration in management of
a landfill to the extent that they can be
predicted or influenced by co-disposal
operations.
One ongoing effort (3) involves a
study of the factors influencing (a) the
rate of decomposition of solid waste in a
sanitary landfill, (b) the quantity and
quality of gas and leachate produced
during decomposition, and (c) the effect
of admixing industrial sludges and sewage
sludge with municipal refuse. Some of the
factors under investigation are: varying
moisture infiltration rates, adding pH
buffering compounds, prewetting the wastes
and varying ambient temperatures. A
combination of municipal solid waste and
various solid and semi-solid industrial
wastes were added to several field
lysimeters. The industrial wastes being
investigated are: petroleum sludge,
battery production waste, electroplating
waste, inorganic pigment sludge, chlorine
production brine sludge, and a
solvent-based paint sludge. Also,
municipal digested primary sludge
dewatered to approximately 20 percent
solids was utilized at three different
ratios. The results of approximately nine
years of data generated under this effort
have been reported in several proceedings
of the annual research symposia from the
Solid and Hazardous Waste Research
Division. An annual report entitled
"Evaluation of Landfilled Municipal and
Industrial Hazardous Waste" has been
received and is currently being reviewed.
A second completed effort (4) related
to the co-disposal of chemically treated
and untreated industrial wastes in a
municipal refuse environment. Large
lysimeters were being utilized to
determine the difference in leachate
quality. Results have indicated that
co-disposal of treated or untreated
electroplating sludge with MSW does not
affect leachate quality. Co-disposal of
calcium fluoride/sewage sludge with MSW
apparently improves leachate quality. The
final report is scheduled for publication
in September 1983.
Pollutant Movement
The objective of pollutant movement
work is to develop predictive
relationships which can be used in the
design and regulation of disposal
facilities for municipal and hazardous
solid waste to insure that any releases to
the environment will not be detrimental.
Current research is intended to
develop tools for designing soil liners
that limit pollutant release to a
predictable minimum. The application is
primarily for landfills, but the
information is also useful in design of
soil liners for surface impoundments and
in predicting the amounts of pollutants
that will pass below the zone of
incorporation at land disposal sites. The
work is proceeding in two parallel
phases. In the first, methods for
predicting movement of liquid (leachate)
are being developed and improved and in
the second, methods are being developed
for predicting how well soil liners will
remove pollutants from any leachate that
seeps through the liner. Work is in
progress on empirical predictive
techniques using samples of wastes and
10
-------
soils from locations of specific
interest. This approach appears to be the
most promising in order to integrate the
effect of waste and soil characteristics
on pollutant retention processes. The
precision and accuracy that will be
achieved by this method remains to be
determined.
One completed effort (2) examined the
extent to which hazardous substances from
specific industrial and flue gas cleaning
(FGC) wastes would migrate into
groundwater at disposal sites. Procedures
for conducting such examinations have been
developed. A sequential batch
leaching/soil adsorption procedure was
developed which provides information
comparable to that obtained from soil
column studies, but in a much shorter
time. The results of this effort have
been reported in several preveious
symposia proceedings; a complete report of
the project will be published in a report
entitled "Migration of Hazardous
Substances in Soil" in FY 1983.
A second completed effort (2) was the
development and testing of numerical
simulation models for predicting movement
of solutes in saturated and unsaturated
soils. The accuracy and computational
efficiency of several existing and newly
developed numerical schemes for solving
flow and transport equations were
compared. Additionally, a closed-form
analytical model was developed for
predicting the unsaturated hydraulic
conductivity from available soil moisture
retention data. Considerations in
selection of numerical models, including
the effects of soil types, flow regime and
magnitude of soil solute interaction
parameters are discussed in a final report
entitled "Mass Transport in
Saturated-Unsaturated Media," which is
being prepared for publication in
September 1983.
A third completed effort (2) was the
development of user oriented models for
predicting movement in soil as a basis for
improving selection of sites for disposing
of solid and hazardous wastes. Long term
movement rates were correlated with soil
and leachate properties to develop
regression equations forming a basis for a
simplified predictive tool. The work has
been completed and a report entitled
"Behavior of Cd, Ni, and Zn in Single and
Mixed Combinations in Landfill Leachates"
describing use of this tool in conjunction
with soil column tests to predict solute
movement in soil is being prepared for
publication in FY 1983.
A fourth completed effort (2) studied
vertical and horizontal contaminant
migration of Zn, Cd, Cu and Pb at three
secondary zinc smelting plants and one
organic chemical manufacturing plant.
Migration patterns were defined using soil
coring and monitoring well techniques.
Soil coring was determined to be an
investigative tool, but not suitable by
itself for routine monitoring of waste
disposal activities. The results of this
study are currently being prepared for
publication in a report entitled "Field
Verification of Toxic Wastes Retention by
Soils at Disposal Sites."
A fifth completed effort (2) was the
determination of the attenuation
mechanisms and capacity of selected clay
minerals and soils for
hexachlorocyclopropentadiene (HCCPD) and
"hex" wastes. Also, effects of
caustic-soda brine on the attenuation of
solubility of HCCPD were researched. The
development of a chemical model to predict
HCCPD migration through soil was pursued.
Mobility of HCCPD in soils was
independently assessed by use of soil
thin-layer chromatography and column
leaching techniques. This study has been
completed and the draft final report
entitled "Assessment of Soil, Clay, and
Caustic-Soda Effects on Land Disposal of
Chlorinated-Hydrocarbon Wastes" is
currently undergoing peer review for
publication in FY 1983.
A sixth completed effort (2) studied
organic contaminant attenuation by
selected clay minerals and coal chars.
Initial work was with Polychlorinated
biphenyls (PCBs); subsequent efforts were
with Hexachorobenzene (HCB) and
Polybrominated biphenyls (PBB). The
results of the first phase of this effort
were published in a report entitled
"Attenuation of Water-Soluble
Polychlorinated Biphenyls by Earth
Materials," EPA-600/2-80-027, March 1980.
The results of the second phase were
published under the title "Attenuation of
Polybrominated Biphenyls and
Hexachlorobenzene by Earth Materials,"
EPA-600/2-81-191, March 1981.
11
-------
A seventh completed effort (2) is
looking at the categories of water
movement that occur in connection with
various types of landbased disposal
systems. Methods of soil hydraulic
conductivity determination and
interpretation are discussed. The results
of the effort are currently being prepared
in the form of a technical resource
document (TRD) entitled "Soil Properties,
Classification, and Hydraulic Conductivity
Testing." This TRD will be released by
the Office of Solid Waste in FY 83 for
public comment.
An eighth ongoing effort (4) has
evaluated the effects of organic leachates
on clay liner permeability and procedures
for rapidly testing these effects have
been outlined. The initial results of
this effort have been published in a
report entitled "Effects of Organic
Solvents on the Permeability of Clay
Soils," EPA 600/2-83-016, April 1983..
A ninth ongoing effort (2) is
currently underway to determine how
accurately the EPA Gas Movement Model
predicts the maximum distance that methane
gas will move through soils adjacent to
landfills and how accurately this model
will predict the relative effectiveness of
control systems (e.g., trenches, wells,
barriers) for minimizing methane gas
movement. Methane movement was measured
at three selected landfills. With data
from each landfill, the model will be used
to predict the maximum extent of methane
movement. The measured and predicted
methane movements will be compared,
reasons for any differences analyzed, and
the accuracy and effort of use for the
model determined. At one of the
landfills, active and passive control
systems are being installed. The observed
degree of control and that predicted by
the model are being compared. The project
is scheduled for completion in March
1983. The report will be entitled "Field
Verification of Gas Migration Predictive
Methods and Design Criteria for Gas
Controls."
A tenth completed effort (2) has
resulted in establishment of guidelines
for design of gas migration control
devices for sanitary landfills. The
migration control devices studied included
trenches venting under natural convection,
trenches with exhaust pumping, trenches
with recharge pumping, barriers, hybrid
systems consisting of a barrier with a
pumped (exhaust or recharge) or unpumped
trench on the landfill site, and pumped
(exhaust or recharge) pipe vents.
Computer codes developed under a previous
study (i.e., Moore and Alzaydi, 1977; and
Moore and Rai, 1977) were modified to
incorporate pressure flow as well as
diffusional flow. The results of this
completed effort currently in draft stage
will be published in a report entitled
"Gas Control Design Criteria" in FY 1983.
An eleventh ongoing effort (2) is
concerned with developing a contaminant/
soil interaction protocol. Work on this
project will include the collection,
evaluation, and testing of procedures for
conducting batch adsorption studies with
leachates and soils to determine the
amounts of pollutants that will be
retained by clay liners in disposal
facilities. The results of this project
will be put into a TRD entitled "Batch
Soil Procedure to Design Clay Liners for
Pollutant Removal" for use in designing
and evaluating clay liners to limit
pollutant release from landfills.
A twelfth ongoing effort (2) is
concerned with developing a TRD entitled
"Containment/Soil Interaction Protocol."
The project will address: 1) to what
degree leachate concentrations will be
reduced by mixing with ground water in the
vicinity of waste disposal sites; 2) the
direction of movement and shape of
leachate plumes; and 3) the types of
models that are appropriate for predicting
movement of solutes in ground water. An
interim report entitled "Methods for
Prediction of Leachate Plume Migration and
Mixing" is now being prepared for release
by the Office of Solid Waste for public
comment.
A thirteenth ongoing effort (8) is
concerned with the development of a
handbook designating state-of-the-art
procedures and techniques of control of
leachate plume migration including methods
for priority use rankings based on site
conditions, risk assessments, cost and
long term applications. Leachate
migration controls that are being studied
include impermeable barriers, slurry
walls, grout curtains, groundwater
pumping, water table adjustment, plume
containment, subsurface drains,
12
-------
bioreclamation, drainage ditches, liners
and leachate treatment.
Pollutant Control
The overall objective of this research
activity is to reduce the impact of
pollution from wastewater disposal sites
by technology that minimizes, contains, or
eliminates pollutant release and leaching
from waste residuals disposed to the
land. The pollutant control studies are
determining the ability of in-situ soils
and natural soil processes to attenuate
leachate contaminants as the leachate
migrates through the soil from landfill
sites. The studies are also determining
how various synthetic and admixed
materials may be utilized as liners to
contain and prevent leachates from
migrating from landfill sites.
Natural Soil Processes: The attenuation
by natural soil processes of pollutants
from hazardous waste and municipal refuse
disposal sites is being performed in the
controlled lab studies previously
discussed in the section on pollutant
movement.
Li ners/Membrane/Admixtures: The
liner/membrane/admixture technology (4)
has been studied to evaluate suitability
for eliminating or reducing leachate from
landfill sites of municipal or industrial
hazardous wastes. The test program
evaluated the chemical resistance and
durability of the liner materials over 12
and 36-month exposure periods to leachates
derived from industrial waste, sulfur
oxides (SOX) wastes, and municipal solid
wastes. Acidic, basic, and neutral
solutions were utilized to generate
industrial waste leachates.
The results of these tests along with
a discussion of the overall hazardous
waste liner material program are presented
in a report entitled "Liner Materials
Exposed to Hazardous and Toxic Sludges
First Interim Report," EPA-600/2-77-091,
June 1977.
A second effort (4) relates to the
types of materials tested for use as
liners for sites receiving sludges
generated by the removal of SOX from
flue gases of coal burning power plants.
The admixed materials consisted of the
following: lime; cement with lime;
polymeric bentonite blend (Ml79); gray
powder-quartec (UF); asphaltic concrete;
TACSS 020; TACSS 025; TACSS C400; and
TACSS ST. The prefabricated membranes
liner materials consisted of the
following: elasticized polyolefin; black
neoprene - coated nylon; and black
neoprene - reinforced fabric. The
spray-on liner materials consisted of the
following: polyvinyl acetate; natural
rubber latex; natural latex; polyvinyl
acetate; asphalt cement; and molten
sulphur. A final report has been
published entitled "Effect of Flue Gas
Cleaning Sludges on Selected Liner
Materials," EPA-600/2-81-098, July 1981
A third effort (4) is an ongoing study
to assess actual field procedures utilized
in (a) preparing the support sub-base soil
structure for liners and (b) placing the
various liner materials common to projects
requiring positive control of fluid loss.
Information obtained to date includes:
methods and equipment used to prepare the
subgrade and to place the liner material;
specific design and construction
considerations; and special problems
encountered during installation. The
final report entitled "Case Studies for
Lined Impoundments" is being prepared for
publication in FY 1983.
A fourth completed effort (4) was
concerned with determining the soil
requirements to act as bedding and
protective covers for flexible membranes.
Experiments were conducted using a variety
of flexible membranes, thicknesses and
densities of base and sub-base sands and
soils, and then trafficked employing three
different vehicles representative of those
used for landfill construction. A design
manual, applicable to the majority of the
country, will be developed which will
provide a design engineer with the data
necessary to determine the most economic
soil requirements above and below a
flexible membrane. The final project
report entitled "Laboratory Model Studies
Conducted to Determine Soil Bedding
Requirements for Flexible Membranes" is
being prepared for publication in FY 1983.
A fifth ongoing effort (4) responds to
the need for support of the guidelines and
hazardous waste criteria mandated under
RCRA. This project has been designed to
provide the EPA with data to enable annual
revisions to the landfill liner TRDs now
13
-------
being finalized. Furthermore, it will
provide assistance in permit application
evaluation, assessment of performance and
expert testimony.
A sixth effort (4), now completed, was
primarily concerned with the evaluation of
specific technologies (chemical liners)
for the disposal and recovery of sludges
from the metal finishing industry.
Specifically, this effort evaluated the
use of three chemical liners, i.e.,
agricultural limestone, fly ash, and
hydrous oxides of iron in minimizing the
pollution potential from an electroplating
sludge. Data have been developed for mass
flux values. The final report will be
available in September 1983.
A seventh effort (2), now completed,
relates to a laboratory study of
agricultural limestone and hydrous oxides
of Fe to evaluate their use as landfill
liner materials to minimize the migration
of metal contaminants. The results of
this effort have been published in a
report entitled "Liners of Natural Porous
Materials to Minimize Pollutant
Migration," EPA-600/2-81-122, July 1981.
An eighth completed effort (4)
provided for the long-term testing of
liner materials. Bag and tub testing of
various membrane materials exposed to
hazardous wastes was used. One of the
objectives of this work was to assess the
movement of organic contaminants through
membrane liners. Permeability testing of
various membrane liners was also
attempted. Also, the study of membranes
on rough surfaces was undertaken to
determine their puncture resistance and
creep over long periods of time. The
results of work on this project will be
used to update the technical resource
document entitled "Lining of Waste
Impoundment and Disposal Facilities,"
(SW-870) annually. This document is
available from the Government Printing
Office (055-000-00231-2).
A ninth ongoing effort (5) is
concerned with evaluating innovative
techniques for detecting and locating
leaks in landfill liner systems. The
three methods to be studies are Time
Domain Reflectometry (TDR), the Acoustical
Emission (AE) technique, and the
Electrical Resistivity (ER) technique.
These methods will be evaluated for their
technical and economic feasibility at a
surface impoundment that is presently
being built and later at an actual liquid
waste impoundment. A report entitled
"Assessment of Innovative Techniques to
Detect Waste Impoundment Liner Failings"
has been reviewed and is expected to be
published in FY 1983.
A tenth effort (5) is concerned with
electrical resistivity techniques for
locating liner leaks. Flexible membrane
materials have been used as liners at
landfills and surface impoundments to
inhibit fluids from contaminating
surrounding water resources. The primary
objective of this task is to develop a
method to detect and locate leak paths in
membrane liners. The work consists of
two-dimensional electrical modeling using
a physical scale model. A system is
underway to develop and demonstrate a
search technique based on electrical
measurement and leak-dependent signatures
as observed in the model studies. Results
have been published in several previous
symposia proceedings. A final report
entitled "Methods for Locating Liner
Failures" will be published in 1983.
Pollutant Treatment
Chemical stabilization is achieved by
incorporating the solid and liquid phases
of a waste into a relatively inert matrix
which exhibits increased physical strength
and protects the components of the waste
from dissolution by rainfall or by soil
water. If this slows the rate of release
of pollutants from the waste sufficiently
and no serious stresses are exerted on the
environment around the disposal site, then
the wastes have been rendered essentially
harmless and restrictions on siting will
be minimal. A technical resource document
entitled "Guide to the Disposal of
Chemically Stabilized and Solidified
Waste" (SW-872) reflects current
state-of-the-art knowledge in the area of
chemical fixation and is available from
the Government Printing Office
(055-000-00226-6).
The initial chemical fixation effort
(4) which is completed relates to the
transformation of wastes into an insoluble
form to minimize leaching. The test
program consists of investigating ten
industrial waste streams in both the raw
and fixed states. The waste streams are
14
-------
treated with at least one of seven
separate fixation processes and subject to
leaching and physical testing. The lab
and field studies have been completed.
The final results of these studies are
reported in three reports entitled: 1)
"Physical Properties and Leach Testing of
Solidified/Stabilized are Gas Cleaning
Waste," EPA 600/2-81-166; 2) "Physical
Properties and Leach Testing of
Solidified/Stabilized Industrial Wastes,"
(EPA 600/2-82-099) in press; and 3) "Field
Investigation of Contaminant Loss From
Chemically Stabilized Industrial Sludges,"
EPA 600/2-81-163.
A second completed effort (6)
evaluated and verified several selected
concentration techniques.
Concentration/pretreatment schemes were
selected for further development and
subsequent bench scale verification with
actual waste stream containing hazardous
materials. Studies were carried out using
various groundwaters which had been
severely contaminated with organic
compounds. Bench scale laboratory
treatability studies, included activated
carbon adsorption, resin adsorption,
aerobic and anaerobic biological, and
stripping. The results of this effort to
date have been published in a report
entitled "Concentration Technologies for
Hazardous Aqueous Waste Treatment,"
EPA-600/2-81-019, January 1981. A
technical resource document entitled
"Management of Hazardous Waste Leachate"
(SW-871) reflects the current
state-of-the-art knowledge in leachate
treatment and is available from the
Government Printing Office
(055-000-00224-0).
A third completed effort (5) relates
to encapsulating processes for managing
hazardous wastes. Techniques for
encapsulating unconfined dry wastes are
discussed in a report entitled
"Development of a Polymeric Cementing and
Encapsulating Process for Managing
Hazardous Wastes," EPA-600/2-77-045,
August 1977. Additional evaluations have
been performed whereby containers of
hazardous waste (i.e., 55 gallon drums)
are placed in a fiberglass thermo-setting
resin casing and covered with a high
density polyethylene. The results of
these evaluations have been published in
three reports entitled "Securing
Containerized Hazardous Wastes with
Polyethylene Resin and Fiberglass
Encapsulates," EPA-600/2-81-138, August
1981; "Securing Containerized Hazardous
Wastes with Welded Polyethylene
Encapsulates," EPA-600/2-81-139, August
1981; "Securing Containerized Hazardous
Wastes by Encapsulation with Spray-On/
Brush-On Resins," EPA-600/2-81-140, August
1981.
A fourth effort (4) was to determine
how potentially useful sulfur-asphalt
binder composites are for stabilization/
solidification and disposal of toxic metal
bearing wastes. Indications from work
carried out on paving material suggest
that both solid and aqueous inorganic as
well as organic materials can be readily
stabilized and solidified with binders
comprised of sulfur-asphalt blends. This
program was terminated and the report
status is pending.
LANDFILL ALTERNATIVES
Due to the concern for environmental
impact and economics, alternatives to
waste disposal in landfills and by
incineration have been proposed. In this
regard, SHWRD is actively pursuing
research related to surface impoundments
and underground mines and their
applicability for the disposal of wastes.
Surface Impoundments
The temporary, or permanent retention
of liquid hazardous wastes in surface
impoundments (i.e., pits, ponds, or
lagoons) is an alternative to disposal in
a landfill, which SHWRD is researching.
Surface impoundments require special
consideration to control vaporization and
volatilization in the air.
An ongoing effort (6) is investigating
mechanisms of chemical movement of
pollutants from surface and near-surface
impoundments into the air. Methods for
estimating air emissions of hazardous
chemicals from waste disposal sites are
being developed. These methods include
transport coefficients and conceptual
models for those coefficients not
available. Methods for controlling or
reducing desorption from surface
impoundments will be suggested.
15
-------
A second completed effort (4) studied
the potential for landfill collection
drains to become clogged by chemical,
physical and/or biological reactions.
Individual landfill sites were studied to
determine existence and nature of current
or anticipated problems. Methods to
remedy drain clogging were developed and
refined.
A third completed effort (4) is
concerned with reviewing current designs
and construction practices for disposal
facilities. Geotechnical parameters which
should be tested, observed, and documented
during construction, operation and closure
of disposal facilities will be
identified. Disposal facility
construction sites will be selected for
observation of the procedures used in
verifying the geotechnical data. This
project will lead to the development of a
manual on state-of-the-art testing
procedures. The final report entitled,
"Geotechnical Quality Assurance in
Construction of Disposal Facilities"
should be available in September 1983.
UNDERGROUND MINES
The use of underground salt mines for
the long-term controlled storage of non
radioactive hazardous wastes has been
shown to be promising as a result of
studies by EPA and others and by the
successful commercial operation of such a
facility in West Germany since 1973.
A research program on characterization
and containment of hazardous wastes for
mine disposal was completed in 1977. This
effort included determination of suitable
mine locations, determining waste handling
requirements, specifying operational
techniques, and the preliminary design of
a mine disposal system. This material was
used to develop a feasibility report
entitled, "Evaluation of Hazardous Waste
Emplacement in Mined Openings," EPA
600/2-75-040, December 1975. Based upon
the feasibility report, one or more
candidate mines was selected for further
evaluation and detailed economic design
was completed. A report entitled, "Cost
Assessment for the Emplacement of
Hazardous Materials in a Salt Mine," EPA
600/2-77-215 was published in November
1977. An engineering plan for an initial
pilot- scale operation was developed and a
full- scale demonstration was planned in
cooperation with a private firm, but never
implemented. In FY 1983 an update of
these studies of hazardous waste disposal
into underground mines will be initiated.
UNCONTROLLED SITES/REMEDIAL ACTION
A completed study by OSW has
identified incidence of water well
contamination due to waste disposal
sites. Seventy-five (75) to eighty-five
(85) percent of all sites investigated are
contaminating ground or surface waters.
In order to determine the best practical
technology and economical corrective
measures to remedy these landfill leachate
and gas pollution problems, a research
effort has been initiated and completed
(1) to provide local municipalities and
users with the data necessary to make
sound judgements of the selection of
viable, in-situ, remedial action
procedures and to give them an indication
of the cost associated with such an
action. This research effort is being
pursued at a municipal refuse landfill
site at Windham, Connecticut. The
engineering feasibility study has been
completed and a report entitled "Guidance
Manual for Minimizing Pollution from Waste
Disposal Sites," EPA-600/2-78-142, August
1978 has been published. This guidance
document emphasizes remedial schemes or
techniques for pollutant containment.
The field verification scheme
installed and monitored at the Connecticut
MSW site was a surface capping technique.
This study is now completed and the
results will be published in a final
report entitled "Remediation of an
Inoperative Municipal Waste Landfill -
Windham Landfill, Windham, Connecticut."
The report is expected to be available in
late 1983.
A second completed effort (1 and 6)
provides for a survey of ongoing and
completed remedial action projects. Ten
case study sites were selected and these
sites visited and inspected for remedial
action information. The results of this
study are published in a report containing
case study summary statistics of each of
the sites investigated. The results of
this completed effort were published in a
final report entitled "Survey of On-Going
and Completed Remedial Action Projects,"
EPA-600/2-81-246, September 1981. This
survey will be periodically updated.
16
-------
A third completed effort (6) is
concerned with the assessment of barrel
and drum reconditioning to determine
environmental impact and to make
recommendations for possible improvements
in the reconditioning processes. Based on
industry characterization and specific
facility identification, two facilities
were chosen for sampling and analysis
efforts, which were monitored for air,
land and water pollutant discharges. The
results of this effort have been published
in two reports entitled "Drum
Reconditioning Process Optimization,"
EPA-600/2-81-233, September 1981 and
"Barrel and Drum Reconditioning Industry
Status Profile," EPA-600/2-81- 232,
September 1981.
A fourth ongoing effort (2) is an
investigation of the failure mechanisms
and migration of industrial chemicals at a
hazardous waste disposal site. The site
has been closed and is under court order
to be excavated for reburial elsewhere.
During excavation of the landfill, the
soil, facility, and waste conditions will
be observed and samples will be
collected. Work will also be conducted to
determine to what extent external
examination can identify problems with
soil covers, to estimate service life of
steel drums in landfills cells, and to
correlate quick indicator tests (viscosity
and filtration rates of leachate/clay
mixtures) with soil permeability changes
measured in the field. The final report
on this project will include recommended
improvements in design and operation of
hazardous waste landfills.
A fifth effort (4) was concerned with
evaluating the uses of geotextiles in
stabilizing cover soils on landfills, act
as drainage layers for gases and liquids,
and act as a bedding for membrane
barriers. This program studied available
geotextiles and geotextile products, their
engineering properties, sizes available
(length, width, thickness), chemical
resistance, seaming techniques, relative
cost, life expectancy, and specific
potential usage. This project included an
initial assessment of problems at the
municipal refuse landfill site in Windham,
Connecticut, and recommendations for those
remedial measures in which geotextiles
could play a significant role. The final
report entitled "Evaluaton of the Use of
Geotextiles for Landfill Cover Systems"
will be published late in 1983.
A sixth ongoing effort (4) is
concerned with determining the feasibility
of retrofitting existing surface
impoundments with an effective membrane
liner system. This study evaluates one or
more technically feasible retrofit liner
installation techniques including the
applicable liner candidate materials. The
work assesses under which physical and
chemical in situ conditions the various
techniques will have the greatest chance
for success. External equipment
requirements, safety, and economics as
well as overall limitation are being
determined. Pilot-scale and field
verification will follow the technical and
economical assessment. The data and
procedures developed in this project will
be used to update the existing TRD reports
entitled "Lining of Waste Impoundment and
Disposal Facilities" (SW-870) and "Closure
of Hazardous Waste Surface Impoundments"
(SW-873) available from the Government
Printing Office (055-000-00227-4).
The seventh ongoing effort (7) is
concerned with the assessment of remedial
actions for uncontrolled hazardous waste
disposal sites. Work on this project will
lead to two distinct products. The first
is to develop a detailed modeling program
for site engineering assessment of
remedial action alternatives at
uncontrolled hazardous waste disposal
sites. The second is to develop
simplified desktop procedures for remedial
action alternative assessment based on the
detailed modeling program. Both products
will describe the effectiveness and cost
of remedial actions considering site
factors and characterization of control
technologies. The draft reports entitled
"Remedial Action Technology Evaluation
Models for Uncontrolled Hazardous Waste
Site, Volume I: Detailed Models" and
"Volume II: Simplified Desktop Models,"
respectively will be completed in FY 1983.
An eighth effort (7) is concerned with
determining the effects of human safety
and degree-of-hazard considerations on
remedial action costs at uncontrolled
hazardous waste sites. Cost components
which are of a worker safety/degree-
of-hazard nature will be identified.
After "identifying the component costs,
specific cost figures will be determined.
This information will be used to adjust
total remedial action unit operation
costs. The draft report entitled
17
-------
"Cost Estimation Procedures for Remedial
Action Technologies at Uncontrolled
Hazardous Waste Sites," should be
completed by July 1983.
A ninth completed effort (10) is the
demonstration of a block displacement
technique to seal off large blocks of
contaminated earth. This was proposed to
include a combination of drilling and
fracturing techniques, displacement
through injection of a specially
formulated bentonite slurry, and ultimate
placement of a low-permeability material
across the bottom and around the sides of
the block. This project provides data
which demonstrates the feasibility of
placing a horizontal stratum of pumpable
material at a reasonable subsurface depth
in a region of favorable geology.
Problems associated with unknown geologic
materials and the interrelationship with
slurry injection technology were
identified, serving as a significant data
base upon which further practice can be
more accurately predetermined. The field
construction within this project has been
completed and the report entitled "The
Block Displacement Method Field
Demonstration and Specifications" is
undergoing final review for publication in
1983.
A tenth completed effort (8) has
developed methods for classification of
wastes by chemical and/or reactive
properties and lays out a logical sequence
of activities to determine the
compatibility of two wastes. The key
elements of the document are a
compatibility chart and a flow diagram for
its use. Therefore, simple and rapid
field tests are needed to determine
reactive properties, chemical class,
generic name and results of mixing. Flame
and solubility-reactivity testing
procedures for the flow scheme were
selected in September 1982. In addition,
a workable redox test was established. A
flow scheme and decision tree for liquids
has also been established. The final
report has been completed with anticipated
publication by the end of FY 1983.
An eleventh ongoing effort (9) is
concerned with the compatibility of
various grouts with hazardous wastes.
Data has been collected and is now being
analyzed. Portland cement grouts,
different percentage silicate base grouts,
lignin base grouts, urea formaldehyde
resin grouts, acrylamide grouts, and other
types of grouts were examined. Also to be
included will be a summary of procedures
and tests recommended prior to selecting a
grout for a particular hazardous waste
site.
A twelfth completed effort (10) is
concerned with the development of a
handbook for slurry trench cutoff wall
design, construction, and performance
evaluation. Among the factors being
tested is the use of different bentonites
and their reactivity to hazardous waste
and leachates. A summary of major cost
elements will also be presented in the
handbook. This Handbook will be printed
as a Technical Transfer document in the
Design Manual series published by EPA's
Center for Environmental Research
Information (CERI). Anticipated
publication will be late 1983.
The purpose of a thirteenth effort
(11) is to close a lagoon at the Louisiana
Army Ammunitiions Plant near Shreveport,
Louisiana. The lagoon contains wastewater
from TNT manufacturing operations.
State-of-the-art fixation techniques are
being tested for the lagoon to insure that
contaminants in the sludge will be
immobilized. Also, processes that
generated heat for the purpose of breaking
the TNT ring are being tested. The
results of this project are included in
the Proceedings of the Ninth Annual
Research Symposium.
The need for a practical compilation
of evaluations and potential applications
of various existing remedial action
techniques has led to the development of a
technical handbook. The handbook explains
the nature of contamination at waste
disposal sites and describes some of the
remedial actions that can be applied for
the clean-up of each contaminated medium.
Remedial actions are designed to control,
contain, treat or remove contaminants from
uncontrolled hazardous waste sites.
Remedial actions are divided into surface
controls, groundwater controls, leachate
controls, direct treatment methods, gas
migration, controls, techniques for
contaminated water and sewer lines, and
methods for contaminated sediment
removal. The handbook is for industrial
and governmental technical personnel
involved with the clean-up of uncontrolled
18
-------
entitled "Cost Estimation Procedures for
Remedial Action Technologies at
Uncontrolled Hazardous Waste Sites,"
should be completed by July 1983.
A ninth completed effort (10) is the
demonstration of a block displacement
technique to seal off large blocks of
contaminated earth. This was proposed to
include a combination of drilling and
fracturing techniques, displacement
through injection of a specially
formulated bentonite slurry, and ultimate
placement of a low-permeability material
across the bottom and around the sides of
the block. This project provides data
which demonstrates the feasibility of
placing a horizontal stratum of pumpable
material at a reasonable subsurface depth
in a region of favorable geology.
Problems associated with unknown geologic
materials and the interrelationship with
slurry injection technology were
identified, serving as a significant data
base upon which further practice can be
more accurately predetermined. The field
construction within this project has been
completed and the report entitled "The
Block Displacement Method Field
Demonstration and Specifications" is
undergoing final review for publication in
1983.
A tenth completed effort (8) has
developed methods for classification of
wastes by chemical and/or reactive
properties and lays out a logical sequence
of activities to determine the
compatibility of two wastes. The key
elements of the document are a
compatibility chart and a flow diagram for
its use. Therefore, simple and rapid
field tests are needed to determine
reactive properties, chemical class,
generic name and results of mixing. Flame
and solubility-reactivity testing
procedures for the flow scheme wera
selected in September 1982. In addition,
a workable redox test was established. A
flow scheme and decision tree for liquids
has also been established. The final
report has been completed with anticipated
publication by the end of FY 1983.
An eleventh ongoing effort (9) is
concerned with the compatibility of
various grouts with hazardous wastes.
Data has been collected and is now being
analyzed. Portland cement grouts,
different percentage silicate base grouts,
lignin base grouts, urea formaldehyde
resin grouts, acrylamide grouts, and other
types of grouts were examined. Also to be
included will be a summary of procedures
and tests recommended prior to selecting a
grout for a particular hazardous waste
site.
A twelfth completed effort (10) is
concerned with the development of a
handbook for slurry trench cutoff wall
design, construction, and performance
evaluation. Among the factors being
tested is the use of different bentonites
and their reactivity to hazardous waste
and leachates. A summary of major cost
elements will also be presented in the
handbook. This Handbook will be printed
as a Technical Transfer document in the
Design Manual series published by EPA's
Center for Environmental Research
Information (CERI). Anticipated
publication will be late 1983.
The purpose of a thirteenth effort
(11) is to close a lagoon at the Louisiana
Army Ammunitiions Plant near Shreveport,
Louisiana. The lagoon contains wastewater
from TNT manufacturing operations.
State-of-the-art fixation techniques are
being tested for the lagoon to insure that
contaminants in the sludge will be
immobilized. Also, processes that
generated heat for the purpose of breaking
the TNT ring are being tested. The
results of this project are included in
the Proceedings of the Ninth Annual
Research Symposium.
The need for a practical compilation
of evaluations and potential applications
of various existing remedial action
techniques has led to the development of a
technical handbook. The handbook explains
the nature of contamination at waste
disposal sites and describes some of the
remedial actions that can be applied for
the clean-up of each contaminated medium.
Remedial actions are designed to control,
contain, treat or remove contaminants from
uncontrolled hazardous waste sites.
Remedial actions are divided into surface
controls, groundwater controls, leachate
controls, direct treatment methods, gas
migration controls, techniques for
contaminated water and sewer lines, and
methods for contaminated sediment
removal. The handbook is for industrial
and governmental technical personnel
involved with the clean-up of uncontrolled
19
-------
hazardous waste sites. In tandem with the
proposed National Contingency Plan, it
will assist in development of technically
sound, environmentally protective,
consistent, cost-effective remedies. The
final report entitled "Handbook for
Remedial Action at Waste Disposal Sites"
(EPA 625/6-82-006) is available.
COST AND ECONOMIC ASSESSMENT
The use of market-oriented incentive
(disincentive) mechanisms has received
very little consideration for pollution
control policy in the United States,
particularly in the area of hazardous
waste management. Economic theory
suggests that incremental pricing of
hazardous waste collection and disposal
would reduce the waste generation rate and
enhance resource recovery of recyclable
materials.
An initial completed effort (7)
developed a method to analyze the economic
and social effects of alternative
approaches to hazardous waste management.
This method has been published in a report
entitled "Socioeconomic Analysis of
Hazardous Waste Management Alternatives:
Methodology and Demonstration," EPA
600/5-81-001, July 1981.
A second completed effort (7) has
resulted in a report entitled "Cost
Comparison of Treatment and Disposal
Alternatives for Hazardous Wastes," Volume
I, EPA 600/2-80-188, February 1981 and
Volume II, EPA 600/2-80-208, February
1981. Unit costs are estimated for
sixteen treatment and five disposal
techniques applicable to various hazardous
wastes. Life cycle average unit costs are
presented in both tabular and graphic form.
An ongoing third effort (7) analyzed
unit operations cost data for long term
remedial action programs at uncontrolled
hazardous waste sites. The analysis
included an outline of the process for
costing remedial action unit operations
and combining the unit operations to yield
relative cost data for an entire remedial
action scenario. The final report
entitled "Costs of Remedial Response
Actions at Uncontrolled Hazardous Waste
Sites" should be available after FY 1983.
A continuation effort is evaluating the
effects of human safety and
degree-of-hazard considerations on
remedial action costs. This information
will be incorporated into the report
entitled, "Survey and Case Studies of
Remedial Actions at Hazardous Waste Sites"
which will be available in early 1984.
An ongoing fourth effort (7) is a
cooperative agreement involving economic,
legal and policy assessment related to
CERCLA. The development of a legislative
index of major implementation issues
related to CERCLA has been completed and
has been published by the Office of
Emergency and Remedial Response. This
report is entitled "Superfund; A
Legislative History". Analysis of actual
remedial action costs based on case
histories is the major ongoing activity.
A fifth ongoing effort (7) provides
direct technical assistance to State and
Regional Office programs dealing with
hazardous waste problems covered by
CERCLA. It provides the assistance
required to apply the best technological,
yet cost-effective solutions to the
problem sites and responsible agencies.
The principal objective of a sixth
completed effort (7) is to develop an
analytical framework for conducting cost-
effectiveness and reliability evaluations
of remedial options at uncontrolled
hazardous waste sites. CERCLA requires
that remedial actions to be implemented
under Superfund must be cost-effective.
This is defined in the National
Contingency Plan as "the lowest cost
alternative that is technologically
feasible and reliable and which
effectively mitigates and minimizes damage
to and provides adequate protection of
public health, welfare, or the
environment". A methodology for making
determinations of cost-effectiveness will
be developed and a guidance document
describing the application of the
framework will be prepared, which can be
used by EPA and state regulatory personnel
to aid in consistent and comprehensive
evaluation of the factors affecting
reliability and cost-effectiveness of
remedial actions. A draft report entitled
"Guidelines for Conducting
Cost-Effectiveness Evaluations at
Uncontrolled Hazardous Waste Sites" has
been prepared.
20
-------
CONCLUSIONS
The laboratory and field research
project efforts discussed here reflect the
overall SHWRD effort in hazardous waste
disposal research. The projects will be
discussed in detail in the following
papers. More information about a specific
project or study can be obtained by
contacting the project officer referenced
in the text. Inquiries can also be
directed to the Director, Solid and
Hazardous Waste Research Division,
Municipal Environmental Research
Laboratory, U. S. Environmental Protection
Agency, 26 West St. Clair Street,
Cincinnati, Ohio 45268. Information will
be provided with the understanding that it
is from research in progress and the
conclusions may change as techniques are
improved and more complete data become
available.
PROJECT OFFICERS
All the Project Officers can be
contacted through the Solid and Hazardous
Waste Research Division (SHWRD), whose
address is shown above. The phone number
is 513/684-7871.
1. Mr. Donald E. Sanning
2. Dr. Mike H. Roulier
3. Ms. Norma M. Lewis
4. Mr. Robert E. Landreth
5. Mr. Carlton C. Wiles
6. Mr. Stephen C. James
7. Mr. Douglas C. Ammon
8. Ms. Naomi P. Barkley
9. Mr. Herbert R. Pahren
10. Dr. Walter E. Grube, Jr.
11. Mr. Jonathan G. Herrmann
21
-------
SUMMARY OF THE FIRST YEAR OF OPERATION
Hazardous Waste Research Center
Elvin J. Dantin
Hazardous Waste Research Center
Louisiana State University
Baton Rouge, Louisiana 70803
ABSTRACT
The Hazardous Waste Research Center, located at Louisiana State University, Baton Rouge,
Louisiana, is one of eight research centers in the U.S. EPA Centers of Excellence program.
This program, funded through the Agency's Research and Development Act of 1978, estab-
lished institutional centers to focus on long-term, fundamental environmental research.
The organizational structure and the program areas of the Hazardous Waste Research Center
are discussed, and the first year's accomplishments are highlighted.
The basic research focused on incineration, alternate methods of treatment/destruction,
and chemical/materials interaction and stabilization. A summary of each research project
is given. The current status of each project is also discussed.
In addition to the basic research efforts of the Hazardous Waste Research Center, two
adjunct programs currently being pursued are discussed. A technology transfer program is
being developed to communicate to industry, governmental agencies, academia and the public
the advances being made through research and also to make available information on all
aspects of hazardous waste. An applied research program is also established to work with
industry to define and serve the immediate problem needs to industry.
BACKGROUND
In it's efforts to expand basic envi-
ronmental research, the U.S. Environmental
Protection Agency (EPA) established eight
research centers, each assigned a topic of
specialization. The Hazardous Waste Re-
search Center (HWRC) is one of these cen-
ters. Louisiana State University was
selected as the HWRC after a very competi-
tive selection procedure, and was awarded a
cooperative agreement on September 30,
1981.
CENTER'S STRUCTURE
The organizational structure of the
Center, shown in Figures 1 and 2, exhibits
the unique characteristics which contrib-
uted to LSU's selection for the location of
the HWRC. The Center not only conducts
basic research but has also established two
adjunct programs to complement the basic
research efforts. See Figure 3 for more
detail on the adjunct programs. The funds
to support the first year's activities of
these adjunct programs have been received
from industry.
A technology transfer program is being
developed to communicate hazardous waste
research results to industry, governmental
agencies, academia, and the public, and to
make available information on all aspects
of hazardous waste. This program has a
full-time person, supported by the State
and industry, to assist in developing the
program. An advisory group has been ap-
pointed to guide the program development.
22
-------
FIRST YEAR'S FUNCTIONING
n
TECHNOLOGICAL
RESEARCH CENTERS
POLICY BOARD
(U S EPA)
FIGURE 1. ADMINISTRATIVE STRUCTURE
Also established is an applied re-
search program to serve the immediate
problem needs of industry. An Industry
Advisory Committee was organized to assist
in this needs identification process.
Researchers from Louisiana State
University (LSU) visited with U.S. EPA
researchers from IERL, Cincinnati; RTF,
North Carolina; MERL, Cincinnati; and
Edison, New Jersey. From these meetings,
suggested areas of research were proposed
that would best fill the need in fundamen-
tal research in the hazardous waste area.
An inventory of the faculty from LSU and
scientists from Gulf South Research Insti-
tute (GSRI) was made to determine the
expertise available to the Center as well
as the interest of these individuals in the
program. The interested participants met
with the Project Officer of EPA, who pre-
sented an overall view of the Center's
functions.
Coordinators were assigned to three
research areas—incineration, alternate
methods of treatment, and waste/materials
interaction. The interested researchers
met separately on several occasions to
develop research studies within their
research areas and preproposals were pre-
pared in the three areas. Nineteen prepro-
posals were submitted to the Executive
Committee of the Center for evaluation.
ADJUNCT PROGRAM
(OUTSIDE GRANT FUNDS)
APPLIED RESEARCH
PROGRAM
INDIVIDUAL
PROJECTS
HAZARDOUS WASTE
RESEARCH CENTER
BASIC RESEARCH
PROGRAM
(EPA SPONSORED)
ADJUNCT PROGRAM
(OUTSIDE GRANT FUNDS)
TECHNOLOGY
TRANSFER
FIGURE 2. FUNCTIONAL STRUCTURE
23
-------
APPLIED RESEARCH
FUNCTION:
TO ADDRESS COMMON INDUSTRY PROBLEMS NEEDING
IMMEDIATE ATTENTION
TECHNOLOGY TRANSFER
FUNCTION:
I. TO COMMUNICATE TECHNICAL , STATE-OF-THE-ART
INFORMATION TO INDUSTRY AND OTHERS
PROFESSIONALLY INVOLVED IN HAZARDOUS WASTE
2. TO DEVELOP AN INFORMATION CENTER THROUGH WHICH
GENERAL INFORMATIOM IS COLLECTED, ORGANIZED,
MADE ACCESSABLE AND/OR COMMUNICATED TO ANY
PUBLIC OR PROFESSIONAL USER
FIGURE 3. FUNCTIONS OF THE TWO ADJUNCT
NON-EPA FUNDED PROGRAMS
As a result of the Executive Commit-
tee's evaluation, eleven preproposals were
submitted to the Policy Board of EPA for
their review and approval at the first
meeting of the Board on December 8-9, 1981,
at LSU. At this meeting the Policy Board
also approved the three focus areas of
research for the Center. Specifically, the
Policy Board of HWRC approved three re-
search projects in incineration, two re-
search projects and one planning study in
alternate methods of treatment, and two
research projects and one planning study in
waste/material interaction. A total of
nine projects was approved.
Upon acceptance of the Year 01 pro-
grams by the Policy Board, initial steps
were taken toward establishing the Scien-
tific Advisory Committee (SAC). The rela-
tionship of the SAC to the overall func-
tioning of the Center can be seen by refer-
ring back to Figure 1. At the Policy Board
meeting held in Chicago, June 10, 1982, a
partial list of candidates was presented to
the Policy Board for approval. There are
now eleven members appointed to the Commit-
tee. Membership includes representatives
from U.S. EPA, industry, and academic
communities.
Basic Research
During the first year of activity,
seven research projects and two planning
studies were identified and funded. Of the
seven research projects approved, all were
designed as multi-year projects. The
results of the first year's research ef-
forts have been reviewed by SAC and recom-
mendations for continuation, based on the
project's progress for the first year, have
been made to the Policy Board. The two
planning studies were terminated at the end
of the first year.
The title, principal investiator(s),
and period of support, together with a
brief statement of objectives for each
funded research project during the first
year are given below. Although the first
year funding period for the cooperative
agreement was November 15, 1981 through
November 14, 1982, delays in program evalu-
ation and funding caused a delay in the
starting date for most projects.
1. Title: Destructability of Pure
Hazardous Waste Com-
pounds in a Laboratory
Flame Environment
Principal Investigator: V. A.
Cundy, Assistant Pro-
fessor of Mechanical
Engineering
Period of Support: 15 January
1982 - 14 November 1982
The objective of this study is to
obtain the fundamental combustion kinetic
data associated with the destruction of
chlorinated hydrocarbon wastes. A flame
burner system will be designed and con-
structed to perform this study. Samples of
gas will be obtained through the flame
produced by the waste fuel providing con-
24
-------
centration profiles. This data, along with
temperature profiles and flame velocity
information, will then be utilized to pro-
ject the fundamental chemical kinetic data
occurring. Such information will provide
more exact details concerning the transfor-
mation of the waste material to less toxic
substances as it progresses through the
flame.
This project has been approved for
continuation for the second year.
2. Title: The Effects of Physical
and Thermodynamic
Properties on the
Combustion of Liquid
Wastes
Principal Investigator: A. M.
Sterling, Professor of
Chemical Engineering
Period of Support: 15 January
1982 - 14 November 1982
The purpose of this project is to
investigate the atomization of liquid
wastes. In contrast to furnaces that burn
fuels of known properties, liquid-waste
incinerators will be required to burn
materials with a wide variation in proper-
ties. In order to tailor the operating
parameters of the furnace to achieve opti-
mum combustion efficiency, the effect of
the atomization process on operating param-
eters must be known. The first step in the
project will be to develop an instrument
that can be used to measure spray charac-
teristics. The instrument will consist of
a laser beam and associated lenses and
photo detectors, and will be able to meas-
ure both the size and velocity of droplets
at various locations throughout the spray.
The effect of liquid properties and the
operating parameters of the nozzle on the
spray characteristics can therefore be
measured and correlated.
The proposal has been completely
rewritten in close collaboration with the
Scientific Advisory Committee and is being
evaluate by the entire Incineration Combus-
tion subgroup of the SAC. Recommendation
from SAC will be forthcoming.
3. Title: Incinerability Charac-
teristics of Selected
Hazardous Waste Mate-
rials
Principal Investigator: R. A.
Matula, Professor of
Mechanical Engineering
Period of Support: 15 January
1982 - 14 November 1982
The objective of this study is to
develop a better understanding of the
combustibility or incinerability character-
istics, as reflected by the combustibility
or incinerability characteristics, as
reflected by chemical and/or kinetic prop-
erties, of a selected group of pure hazard-
ous waste compounds. Presently, experimen-
tal work is being carried out to determine
if either infrared emission from carbon
monoxide (CO) or carbon dioxide (CO-) can
be utilized as an effective measure of
ignition delay times for chlorinated hydro-
carbon/oxygen systems. In addition, ana-
lytical models for the combustion of CC1,
are being developed in order to obtain a
better understanding of the reactions
important in the ignition and oxidation of
chlorinated hydrocarbons.
This project has been approved for
continuation for the second year.
4. Title: Destruction of Hazard-
ous Organic leachates
by Photolytic Ozonation
Principal Investigators: D. Roy,
Assistant Professor of
Civil Engineering;
M. E. Tittlebaum,
Assistant Professor of
Civil Engineering
Period of Support: 15 January
1982 - 14 November 1982
The major objective of this research
is to obtain information on the kinetics
and mechanism of destruction of hazardous
organic wastes by photolytic ozonation.
Application of the photolytic ozonation
process for complete destruction as well as
for enhancement of biodegradability of
hazardous wastes and leachates will be
investigated. Based on these findings an
engineering design approach will be devel-
oped.
The proposal is being revised with
major input from SAC and the Policy Board,
and will be re-evaluated by SAC. Recommen-
dations will be forthcoming.
5. Title: Rate Mechanisms in the
Supercritical Extrac-
tion of Organics from
Solid Hazardous Wastes
Principal Investigators: F. C.
Knopf, Assistant Pro-
25
-------
fessor of Chemical
Engineering; x E.
McLaughlin, Professor
of Chemical Engineer-
ing; R. G. Rice, Pro-
fessor of Chemical
Engineering
Period of Support: 15 January
1982 - 14 November 1982
The removal of organic hazardous
wastes from contaminated soil due to chemi-
cal spill is the concern of this project.
In supercritical extraction technology, a
simple gas such as carbon dioxide at high
pressure and moderate temperatures, is
contacted with the contaminated soil and
the organic waste is leached out into this
high pressure gas in which it freely dis-
solves. The gas can then be vented off
into a tank where, if the pressure is
reduced, the hazardous waste falls to the
bottom of the tank and the pure carbon
dioxide gas comes off the top. The gas is
returned to a compressor which compresses
it up to high pressure. It is again re-
turned to the contact tank to remove more
waste from the soil.
The proposal has been revised with
major input from SAC, Policy Board, and
external reviewers. SAC members and exter-
nal reviewers are evaluating this proposal.
Recommendations will be forthcoming.
6. Title: Leachate Effects on
Structural Stability
and Hydraulic Conduc-
tivity
Principal Investigator: Y. Acar,
Assistant Professor of
Civil Engineering;
R. E. Ferrell, Profes-
sor and Chairman of
Geology; S. D. Field,
Assistant Professor of
Civil Engineering
Period of Support: 15 January
1982 - 14 November 1982
This research addresses the question
of what is the long-term fate of hazardous
materials being currently disposed of in
landfills. What happens to wastes buried
in such sites? Over long periods of time
(decades to centuries) it is known that the
wastes will tend to migrate through the
clay liners at very slow rates. At the
same time the wastes are subject to absorb-
tion on soil particles and degradation.
This study investigates the rates of decay
of specific wastes (p-chloro-m-cresol,
arocolor 1254, 2-4-dinitrotolune, and
1,2-dichlorobenzene). This information
will assist in the analysis of whether
these specific wastes can be safely buried
in the future.
A new proposal is being prepared in
close collaboration with SAC members.
The Hazardous Waste Research Center
also supported two research planning stud-
ies in the first year that will not be
continued in the second year. The intent
of the planning studies was to review
existing information in specific subject
areas and identify research needs. The
information from these planning studies
would be used to prepare research projects
to fill the voids which exist in the funda-
mental subject areas.
1. Title: An Assessment of the
Technical Feasibility
of Stabilization of
Hazardous Organic
Liquid Wastes and
Sludges
Principal Investigators: R. K.
Seals, Professor and
Chairman of Civil
Engineering; M. E.
Tittlebaum, Assistant
Professor of Civil
Engineering
Period of Support: 15 January
1982 - 14 November 1982
The objectives of this project are to
determine the present state-of-the-art of
stabilization of hazardous organic liquid
wastes and sludges, and to identify the
principal types and estimated quantities of
these wastes and sludges. Potential bond-
ing mechanisms between organic wastes and
candidate stabilization materials and
mechanisms that interfere with stabiliza-
tion processes will be identified. The
potential use of waste residuals in stabil-
ization processes needs to be determined.
Pretreatment techniques to enhance the
stabilization of organic waste need to be
identified, and a program to evaluate the
performance of these stabilization tech-
niques needs to be developed.
The final report for the first year
has been completed and is being reviewed by
SAC and the Policy Board, A research
proposal based on these findings is being
reviewed by SAC, and recommendations will
26
-------
be forthcoming.
PUBLICATIONS
2. Title: Novel Treatment Tech-
niques for Soils Con-
taminated with Hazard-
ous Substances
Principal Investigators: D. P.
Harrison, Professor of
Chemical Engineering;
D. M. Wetzel, Assistant
Professor of Chemical
Engineering
Period of Support: 15 January
1982 - 14 November 1982
This study will survey possible alter-
nate treatment/destruction methods to
identify those which appear to be most
promising in terms of ultimate destruction
capability, cost effectiveness, and poten-
tial applicability. Key research needs for
each of the most promising methods will be
identified and used to guide the long-term
research effort in this area.
The final report for the first year
has been completed and is being reviewed by
SAC and the Policy Board. A research
proposal based on these findings is being
reviewed by SAC, and recommendations will
be forthcoming.
Applied Research
The faculty coordinator for the ap-
plied research program met periodically
with the Industry Advisory Committee (IAC)
to identify areas of concern as well as the
alternative approaches to solving the
problems. The active participation and the
serious concern of the members of the IAC
was evident in the first year.
Technology Transfer
In technology transfer's first year,
many tasks, as program planning, Center
support functions, numerous publications
and other activities, were performed. The
publications are described in the following
section. Program planning focused on
activities to benefit industry, government
and academic researchers. An assessment of
the support capabilities for research in
hazardous waste by the LSU Library was
begun. Efforts to strengthen this support
capability are planned. Members for the
Technology Transfer Advisory Committee were
selected.
Through the Technology Transfer pro-
gram of the HWRC, publications have been
prepared that will be used to increase
industry and community awareness of the
purpose and functioning of the Center.
Magazine articles to appear in the LSU
College of Engineering Engineering News and
the newsletter, Water Currents, were writ-
ten to publicize the existence and function
of the Center. Currently in printing are a
brochure and a newsletter.
Another publication was the Proceed-
ings of the Symposium and Workshop on
Hazardous Waste Management: Protection of
Water Resources. The Symposium, held in
November 1982 at LSU, was organized to
present the important issues in hazardous
waste management. The Symposium was fol-
lowed by a workshop conducted to define
research needs in the hazardous waste area.
The following technical publications
have resulted from the HWRC research stud-
ies .
"Destruction of Hazardous Wastes
by Photolytic Ozonation," D. Roy,
M. E. Tittlebaum and R. Knox.
Presented at the Louisiana Con-
ference on Water Supply, Sewage,
& Industrial Wastes sponsored by
the Louisiana Water Pollution
Control Association in March,
1982.
"Comparison of Exact and Approxi-
mate Solutions of Transient
Response for Leaching by SCF in
CSTR." F. Carl Knopf, Edward
McLaughlin and Richard G. Rice.
Chemical Engineering Communica-
tions .
"A Quantitative Approach to the
Long-Term Effectiveness of Land-
filling of Organic Hazardous
Wastes." Ron Malone and Michelle
Frey. To be submitted to Journal
of Hazardous Materials.
"Organic Leachate Effects on
Hydraulic Conductivity of Com-
pacted Kaolinite." Hamidon,
Scott, Y. Acar and S. Field. To
be presented at the International
Symposium on Impermeable Barriers
27
-------
for Soil and Rock, and later
published in an ASTM Special
Technical Publication.
5. "Pure Organic Solvent Effects on
the Fabric of Compacted Clay."
Olivieri, Y. Acar and S. Field.
To be presented at the Interna-
tional Symposium on Impermeable
Barriers for Soil and Rock.
Later to be published in an ASTM
Special Technical Publication.
In addition, final reports have been pre-
pared for the two research planning stud-
ies.
FUTURE PLANS
Two new research proposals have re-
sulted from the planning studies. In
addition, eight new research proposals have
been submitted to the Center for evalua-
tion. The two proposals resulting from the
planning studies and two of the new propos-
als are being evaluated by the SAC and
Policy Board, and, if approved and funds
become available, will be included in the
second year of the Center's program.
With support from industry and the
State, the Center will sponsor a symposium
and several short courses in 1983 and 1984.
The technology transfer program will main-
tain a depository of current informtion and
develop materials for transferring informa-
tion to professionals as well as lay peo-
ple. The quarterly publication of a news-
letter as well as numerous other publica-
tions are planned.
A more active participation and rap-
port with industry is anticipated as the
Center develops its applied research pro-
gram in hazardous waste.
The future of the Center will depend
on the support of U.S. EPA to continue the
fundamental research program. The technol-
ogy transfer and applied research program
are the necessary added involvement to
develop the Center into a Comprehensive
Center of Excellence in Hazardous Waste
Research.
28
-------
A MODEL FOR HYDROLOGIC EVALUATION OF LANDFILL PERFORMANCE
Paul R. Schroeder
USAE Waterways Experiment Station
Vicksburg, MS 39180
Michael D. Smolen
Department of Agricultural Engineering
Virginia Polytechnic Institute and State University
Blacksburg, VA 24061
ABSTRACT
The HELP model is an interactive computer-based model designed for use by nonpro-
grammers to evaluate the probable hydrologic performance of existing or proposed landfill
designs. The model simulates runoff, infiltration, evapotranspiration, lateral subsurface
flow, and percolation using accepted theoretical relationships and water routing tech-
niques. Outputs from the model include all aspects of the hydrologic cycle, with particu-
lar emphasis on leachate production and percolation losses. The hydrologic model is
interfaced directly with precipitation and temperature files from representative areas
around the United States and has default parameter values for simulations of most landfill
designs. The theoretical development of the HELP model and its application to evaluation
of probable landfill performance is discussed.
INTRODUCTION
The Hydrologic Evaluation of Landfill
Performance (HELP) model was developed to
assist in the design of waste disposal
sites and in the evaluation of the environ-
mental impact of existing waste disposal
sites. The HELP model is an improved and
extended version of the Hydrologic Simula-
tion model for estimating percolation at
Solid Waste Disposal Sites (HSSWDS) (1,2),
that was developed for the U.S. Environmen-
tal Protection Agency. The resulting model
is a quasi-two-dimensional, deterministic
model that computes the daily water budget
for a landfill design, considering percola-
tion through successive horizontal layers
and lateral drainage from above sloping
barrier layers. The HELP model is based on
accepted, but simplified, soil physical
theory using parameters for which values
are available from the landfill design
specifications or from reports in the
hydrology literature.
The HELP model is implemented, as was
HSSWDS, with a full set of default soil
characteristics and default climatologic
data for representative locations in all
parts of the country. Therefore, the pro-
spective model user can use HELP by speci-
fying only the sequence of soil, waste, and
barrier layers, with layer dimensions and
layer textures.
The model is implemented for use in a
conversational mode so that the user re-
ceives his output at a computer terminal.
A question and answer format is used so
that no programming experience is necessary
to run the HELP program.
MODEL STRUCTURE
The overall configuration of the HELP
model is shown in Figure 1. The user spec-
ifies the sequence and characteristics of
the layers in the model. Any of the layers
29
-------
Rain/Snowmelt
' Evapotranspiration
COVER
WASTE, DRAIN,
LINER SYSTEM
Percolation
Figure 1. Configuration of HELP model.
shown in Figure 1 may be omitted, or addi-
tional waste cells nay be added below the
one shown. An open landfill, prior to
sealing, may be simulated by omitting the
layers of the cover, such that the upper-
most layer is a waste layer.
The amount of moisture that enters the
topsoil layer is determined by a daily in-
filtration submodel that considers the
amount of antecedent moisture and the den-
sity of vegetation on the surface. Infil-
trated water is routed through conceptual
soil segments to the barrier layer where it
is subject to evapotranspiration demand and
gravity drainage. Drainage may occur by
vertical percolation or lateral subsurface
flow. The input to layers below the cover
is the amount computed as percolation
through the barrier layer. Multiple bar-
rier layers and lateral drainage layers
are possible.
The HELP model computes runoff by the
Soil Conservation Service (SCS) curve num-
ber method (3). Percolation is computed by
Darcy's Law, modified for unsaturated flow
(4). Lateral flow is computed by a lin-
earization of the Boussinesq equation and
evapotranspiration is determined by a
30
-------
method developed by Ritchie (5). The ver-
tical percolation and evapotranspiration
components of the HELP model originate-!
with the CREAMS (6) program. The develop-
ment presented here, however, reflects a
significant advance beyond the CREAMS model
in both of these areas. Each of these
model components will be discussed in
detail below.
Runoff and Infiltration
The SCS curve number technique (3) as
first implemented in the CREAMS model, was
selected as a model to partition incoming
rainfall or snowmelt between runoff and in-
filtration in HELP. This technique was
selected because:
1. It is a well-established procedure
and appropriate values for the curve number
can be assigned without great difficulty,
2. It provides a computationally
efficient procedure for estimating daily
runoff values from daily precipitation, and
3. A wide variety of surface vegeta-
tive cover densities can be simulated.
The SCS equation relates daily runoff
volume, Q, to daily precipitation, P, and a
watershed retention parameter, S, as fol-
lows : „
= (P-0 ..2S )
^ (P+0.8S)
where, Q, P, and S are in inches.
(1)
The retention parameter, S, is a nonlinear
function of soil wetness and vegetative
cover density. It has been related to a
series of empirical curves developed by
SCS. An appropriate curve can be selected
from the SCS manual (3) to match the de-
sired surface cover for a landfill. The
curve number, CN, is related to the reten-
tion parameter as:
(2)
The actual value of S for any time
during a simulation period is computed by
the procedure described in CREAMS (6):
max
r SM - WP"I
I UL - WPJ
(3)
where
S = retention parameter value for
max , v
a dry antecedent condition
SM = a depth weighted estimate of
soil moisture (dimensionless)
UL = upper limit of soil water
storage or total porosity of
the soil (dimensionless)
WP = the lower limit of soil water
storage or wilting point.
(dimensionless)
To account for the nonuniform dis-
tribution of moisture in the soil profile,
a weighting function which gives greater
weight to moisture near the surface is
used. The soil profile in the evaporative
zone is divided into seven segments. These
soil segments are used in the moisture rout-
ing scheme, and their soil moistures are
averaged to compute the effective soil mois-
ture.
Soil segments vary in thickness as
follows: the uppermost segment has a
thickness equal to 1/36 times the lesser
of the depth from the surface to the bar-
rier layer and the depth of the evapora-
tive zone; the second layer has thickness
equal to 5/36 of the profile, and the re-
maining five segments each have thickness
equal to one-sixth of the profile. The
depth-weighted retention parameter is
computed as:
(4)
where W. = weighting factor for segment j
SM. = soil moisture content of
segment j
UL. = upper limit of soil moisture
content of segment j
WP = wilting point of moisture con-
tent of segment j
The weighting factors are identical to
those used in CREAMS (3): for segments 1
through 7, these weighting factors are
0.111, 0.397, 0.254, 0.127, 0.063, 0.032,
and 0.016, respectively.
The value of S in Equation 3 cor-
responds to the retention parameter for dry
antecedent moisture conditions (AMC-I).
The curve number for AMC-I, CN , is com-
puted from the average curve number CN as
follows: LI
31
-------
CN = -16.91 + 1.348 CN
- 0.01379 CNj + 0.0001177 CN^
(5)
Infiltration is computed as the dif-
ference between the daily precipitation and
runoff computed by Equation 1, minus the
daily surface evaporation. If the mean
daily temperature is below 32°F, the pre-
cipitation is stored on the surface as snow
and does not contribute to infiltration or
runoff until the mean daily temperature
rises above 32°F. The amount of melted
snow is added to precipitation for use in
Equation 1.
The amount of snowmelt on day i is
computed as follows:
Hi = 0 for TI < 32°F or SNO = 0
MI = 0.06 (T.j-32) for T± > 32°F (6)
and SNO > 0.06 T.
VI± = SN01_1 for T± > 32°F and
SNO < 0.06 T ,
where M = amount of snow melted on day
i, inches
T. = mean temperature on day i, °F
SNO. . = amount of snow water on
surface at the end of day i-1,
inches.
Evapotranspiration
The evapotranspiration (ET) from a
landfill cover is a function of the energy
available, the vegetation, the soil
transtnissivity, and the soil moisture con-
tent. The potential ET Is computed in
HELP by a modified Penman method as
described by Ritchie (5) and used in
CREAMS (6):
E =
1.28 A H
(7)
o A + 0.68
where E = potential ET, mm
A = slope of saturation vapor
pressure
H = net solar radiation, langleys
Daily mean temperatures and solar
radiation values are obtained by fitting a
simple harmonic curve to the monthly mean
values by Fourier analysis. The resulting
equation is evaluated for each day of the
simulation.
The daily potential ET is applied
first to any free water available on the
surface, thereby reducing the computed in-
filtration or the amount of snow on the
surface. Any ET demand in excess of free
surface water is exerted on the soil col-
umn for direct soil evaporation and evapo-
transpiration through the surface vegeta-
tion.
Soil evaporation proceeds in two
stages: stage one, where evaporation is
controlled by atmospheric demand and stage
two where evaporation is limited by soil
transmisslvity. The potential soil
evaporation is computed as:
ES = E exp (-0.4 LAI)
(3)
where ES = potential soil evaporation, mm
E = potential ET, mm
LAI = leaf area index (on scale of
0 to 3)
During the nongrowing season, LAI in Equa-
tion 8 is replaced by a winter cover fac-
tor (WCF) which varies between 0 and 1.8
depending on the density of dead or dormant
vegetation.
In stage one evaporation, soil evapo-
ration is equal to the potential soil
evaporation ES . An upper limit to the
amount of stage one evaporation is computed
as:
U = 9(ag - 3)
0.42
(9)
where
U = the upper limit for stage one
evaporation in mm
a = soil transmissivity parameter
for evaporation in (mm/day)
When the accumulated stage one evaporation
less the infiltration exceeds the upper
limit, U, any further soil evaporation pro-
ceeds by the following equation:
32
-------
where ES_ = stage two soil evaporation, mm
t = Jays since stage one ended
Whenever the infiltration exceeds the cumu-
lative evapotranspiration less the upper
limit for stage one evaporation, evapora-
tion reverts to the stage one process.
Potential plant transpiration is com-
puted as
EP
E LAI
o
(10)
depth-weighted using the same weighting co-
efficients used for computation of the re-
tention factor used in Equation 1.
The rate of change in leaf area index
(DLAI) during different parts of the year
is computed frora 13 LAI values that repre-
sent values for different dates from a year
in which no severe droughts occurred. The
appropriate values of DLAI are used to com-
pute LAI on a daily basis throughout the
year. When drought occurs and the weighted
soil moisture approaches the wilting point,
no increase in LAI occurs.
The potential ET may be satisfied entirely
by evaporation from free water, from stage
one soil evaporation, or from stage two
soil evaporation. If there is remaining
atmospheric demand after these sources of
ET have been exhausted, the demand will be
satisfied by plant ET to a maximum of EP .
The actual ET varies as a function of soil
moisture and atmospheric demand as shown by
Shanholtz and Lillard (7), Saxton et al.
(8), and Sudar et al. (9). The following
relationship was developed from ET-rate
curves for no-till corn presented by
Shanholtz and Lillard (7):
EP =
where EP = actual plant ET, mm
= actual plant ET demand, mm
= E - ES - ESS
o
if EP > E - ES - ESS
o o
= EP
The total ET is the sum of free sur-
face evaporation, stage one and stage two
soil evaporation, and plant ET. The plant
ET is distributed to the conceptual seg-
ments of the soil profile using the same
weighting factors used previously to esti-
mate the average soil moisture. As each
segment is depleted to wilting point, the
next lower segment is subject to an addi-
tional ET demand that Is equal to the re-
maining plant ET of the above segment. The
result of this procedure is that water is
drawn rapidly from the upper segments of
the profile, then moves to deeper segments
as the upper segments are depleted. As the
ET demand extends to deeper segments, the
efficiency of moisture withdrawal is re-
duced by the segments weighting factor.
The HELP model simulates ET moisture with-
drawal to a maximal depth of 120 cm, but
the efficiency of moisture withdrawal below
45 cm is very low. This distribution of ET
demand through the soil profile is consis-
tent with data presented by Saxton et al.
(8).
Vertical Moisture Flow
if EP
-------
INFILTRATION
DR1=DRAINAGE FROM SEGMENT 1
DR2
i
DR3
TOPSOIL LAYER
DR5
DR6
LATERAL DRAINAGE
LAYER
LATERAL DRAINAGE RATE
PERCOLATION RATE
DR?= LATERAL DRAINAGE RATE + PERCOLATION RATE
I BARRIER SOIL
LAYER
PERCOLATION RATE
Figure 2. Vertical flow submodel for Landfill cover.
The profile of the landfill cover is
generally assumed to consist of eight seg-
ments, where the bottom of segment seven is
the upper boundary of the barrier layer.
If the profile consists of several layers,
as shown in Figure 2, the layer properties
are assigned to corresponding model seg-
ments. Where segment and layer boundaries
do not coincide, the segment properties
(conductivity, porosity, etc.) are computed
as thickness-weighted averages of the two
layers represented. In this way a layered
profile consisting of loam over sand may be
modeled as having a gradual change in soil
properties.
Below the landfill cover, the waste
and drainage layers may each be modeled as
a single soil segment. The input to layers
below a barrier is the amount of water that
34
-------
percolates through the barrier above. The
barrier layer itself is always modeled as
being separate from the rest of the pro-
file.
Storage routing through segments of
the model consists of simultaneous solution
of Darcy's Law and an equation of continu-
ity. Darcy's Law is represented as:
1 = KTT
(12)
where q = rate of vertical flow, inches/
day
K = hydraulic conductivity, inches/
day
h = gravitational head, inches
1 = length in the direction of
flow, inches
Free outfall is assumed from each segment
such that dh/dl may be set equal to unity,
and q is equal to the hydraulic conductiv-
ity. This assumption is acceptable only if
the conductivity of the profile is constant
or increases with depth. Because this
assumption is not met at the interface be-
tween the seventh segment and the barrier
layer, a different procedure is employed
there.
The rate of vertical moisture flow in
soil varies with soil moisture content.
When the soil is fully saturated, the flow
rate is equal to the saturated hydraulic
conductivity, and when the moisture content
is reduced to field capacity, the flow rate
is zero. In the HELP model the hydraulic
conductivity is taken to be a linear func-
tion of soil moisture:
k = k
u s
(SM-FC\
UL-FCy
(13)
where
k = unsaturated hydraulic con-
ductivity, inches/day
k = saturated hydraulic con-
ductivity, inches/day
SM = soil moisture content, vol/vol
FC = soil moisture content at field
capacity, vol/vol
UL = total porosity or the soil
moisture content at satura-
tion, vol/vol
The continuity equation is written to
represent the midpoint of the routing time
interval as:
AS =
l
(14)
where
AS is the change in storage during
the time interval,
I. and !„ are inflows to storage at
the beginning and end of the time
interval, and
0 and 0 are outflows from storage
at the beginning and end of the
time interval.
The first term on the right hand side of
Equation 14 represents the average inflow,
and the second term represents the average
outflow. The inflow to the uppermost seg-
ment is infiltration; inflow to other seg-
ments is percolation from the above
segment. Outflows are by gravity drainage
or ET.
Rearranging Equation 14, the change
in storage (of a segment) may be written as
the sum of one-half the net drainage from
the previous time interval and one-half the
net drainage at the end of the current time
interval. Therefore:
aii = SM
where
i-l
SM
NDR-
NDR
-
soil moisture at the end of
the current time interval,
SM. . = soil moisture at the end of
the previous time interval,
NDR = net drainage,
and
NDR.
.
- ET
(16)
1(j)"JAt
where j = segment number,
i = time,
DR = drainage rate,
At = the time interval, and
35
-------
ET = evapotranspiration rate.
The drainage rate from a segment is
written as:
DR. = k
i s
- FC
UL - FC
and Equations 17 and 15 may be solved
simultaneously, knowing SM. . from
vious time interval. The resulting
age from segment j is:
(17)
:he pre-
drain-
2ks (J)
1/2
-t- 1/2 DR1(j-l)
1/2 ETi(j) At - FC(j)
- FC(J)) + k (j) At)
S
(18)
2(UL(J)
The routing proceeds from the top sebment,
where incoming drainage is infiltration
(rainfall less surface evaporation), to the
seventh segment where the drainage oat is
divided between percolation through :he
barrier layer and lateral drainage a'jove
the barrier layer.
Percolation
An accurate estimate of percolation
rate cannot be obtained using the rott-
ing procedure described in the vertical
flow submodel because of the discontinu-
ity at the interface with the barrieij
layer. Percolation is modeled as Dancian
flow where the drainage rate is computed
as:
TH
\
v
(19)
where q
the rate of percolation through
the barrier
TH = the total head in the profile
above the barrier layer,
T = the thickness of the barrier,
and
k^ = the saturated hydraulic con-
ductivity of the barrier.
To reduce complexity of the percolation
model, the drainable porosity of the bar-
rier is considered negligible. This
assumption has the effect that drainage
stops as soon as there is no head above the
barrier. In fact, if the barrier is a
typical clay, it would continue to drain at
a very low rate until the drainable water
is depleted. If the drainable porosity of
the clay were one percent and the thickness
of the barrier were 60 cm, 0.6 cm of water
could drain from the barrier after all
drainable water was removed from the pro-
file above the barrier. Because the
hydraulic conductivity of clay is very low
(on the order of 10 cm/sec), it would re-
quire several months to deplete the drain-
able water from the clay. Further, it must
be considered that the percolation of water
from the soil profile into the drainable
volume of the clay will also proceed at a
slow rate due to the low hydraulic conduc-
tivity of the clay. Therefore, under humid
conditions the barrier assumption will have
a negligible effect, but under arid condi-
tions with occasional wetting the amount of
percolation may be underpredicted. The
underprediction, however, would be largely
compensated over extended simulation
periods.
Lateral Flow Submodel
The lateral flow submodel of the HELP
program is based on a linearization of the
Boussinesq equation:
Figure 3. Configuration of lateral flow submodel.
36
-------
+ R
(20)
where f = drainable porosity
'-i = gravitational heat!
k - hydraulic conductivity for
lateral flow
t = time
x = lateral distance
Q! = fractional slope
K = recharge flux
The model configuration is illustrated in
Figure 3 for a landfill that is 21 wide and
has side slope equal to a. For application
in the HELP model, a steady-state assump-
tion is used:
y = average thickness of flow,
L = maximum length to drain
h = elevation of water surface at
° x = L
The vertical flow submodel simulates
vertical flow for the entire cross-section
shown in Figure 3. Therefore, the thick-
ness of the saturated zone computed in_the
vertical flow submodel corresponds to y.
Using a finite difference model, developed
by Skaggs (10) to solve the Boussinesq
equation, lateral drainage rates and water
profile were generated for nine combina-
tions of a and L with y ranging from 0.2 m
to 1.0 m to obtain correction factors for
the linearized Boussinesq equation.
Equation 23 may be rewritten with the cor-
rection factor as follows:
(21)
2 C ky
(24)
such that no change in head occurs during a
computational time step. The steady-state
assumption places certain restrictions on
the magnitude of computational time steps.
This will be discussed further in the sec-
tion on model testing.
With the steady-state assumption,
Equation 20 may be rewritten as:
dx vy dx
where y = h - ax.
(22)
Equation 22 was linearized by Skaggs (10)
to the following form by substituting y for
y and integrating subject to the boundary
conditions:
where
and
(25)
C = correction factor
= 0.510 + 0.00205 aL
(26)
The variables, y and h , are unknown dur-
ing the simulation; therefore, a relation-
ship between y and y was developed of the
following form:
where
yo - C2
(27)
h = 0 at x = 0
-r— = 0 at x = L
to obtain:
(22a)
(22u)
(28)
Therefore, the final form of Equation 23
is:
2kyh
(23)
2ky(0.510 + 0.00205QL)
T2
US]'16 }
\y\aij + oL/
(29)
where: Q = lateral drainage rate,
LJ
37
-------
where the units of y and L are in inches
and a is dimensionless.
Linkage Between Vertical
and Lateral Flow Submodels
The linkage between vertical and lat-
eral flow occurs through a moisture balance
computation in the vertical model which is
used to estimate the gravitational head.
The gravitational head determines the rate
of lateral flow and the rate of percola-
tion. Because the moisture balance is a
function of drainage from the profile, and
drainage from the profile is a function of
the resulting moisture balance, an itera-
tive scheme is required to solve both
simultaneously.
The computational scheme begins by
taking a previous value as the initial
estimate of drainage out of the vertical
profile. This outflow rate is used in
Equation 15 to compute the moisture content
of the bottom segment. The earlier assump-
tion of free outflow from each segment (see
Equation 12) could result in an excessive
buildup of moisture in the bottom segment.
Therefore, each segment, beginning at the
bottom of the profile, is checked for
apparent moisture content in excess of the
available pore space of the segment. Any
excess moisture in a segment is distributed
upward. If excess moisture results at the
top of the profile, it is added to the run-
off during the time step.
After the segment moisture contents
are balanced by distributing excess mois-
ture upward, the effective gravitational
head for lateral flow and percolation is
computed with the following equation, be-
ginning at the bottom of the profile:
" FC(3)
TH = /. 10 1.1; !„ — -.-< „„, , % i C28)
*J
where TH = total head,
TS(j) = thickness of segment j,
m = the bottom segment, and
n •= the lowest segment that is not
saturated.
The hydraulic conductivity, k, for
lateral flow in Equation 23 is also a
function of the thickness of flow (y), if
the profile is layered. The conductivity
is computed as a head—weighted average of
the saturated hydraulic conductivity of
each segment that contributes to the head.
This can be expressed as:
\ - 2- k (j)D(J)/TH (29)
j=m
where k_ = the effective lateral
hydraulic conductivity,
k (j) = the saturated hydraulic con-
ductivity of segment j
and
SM(j) - FC(j)
UL(j) -
The gravitational head (TH) is substi-
tuted for y and hydraulic conductivity (1O
is substituted for k in Equation 23 to com-
pute the rate of lateral flow (Q,), and the
rate of percolation through the barrier by
Equation 19. The estimates, Q and Q are
summed and compared with the initial esti-
mate of profile drainage. If this estimate
is not within a predetermined tolerance of
the previous estimate, a new drainage
estimate is computed by the method of bi-
section. The moisture balance and head
computations and the computation of Q are
repeated until successive estimates of Q.
are within 5 percent of each other.
The steady-state assumption used to
link the vertical and lateral flow sub-
models requires that the computational time
step be sufficiently short that there is no
appreciable change in head. The stringency
of this requirement is reduced somewhat
because of the midpoint routing procedure.
With the midpoint routing procedure, the
computational time interval must only be
short enough to keep the rate of lateral
flow and percolation constant through the
time interval. The length of computational
time step must be short when y (TH) is
large and the drainable porosity is low.
Figure 4 shows_the rate of change of Q. as
a function of y and k for a soil with
c
drainable porosity equal to 15 percent.
When the soil consists of a loam (k =
10 cm/sec), Q varies_from zero to
9.4 x 10 cm day as y changes.from zero
to 150 cm. For a sand (k = 10 cm/sec),
however, Q varied from zero to 9.34 cm
day " over the same range of >a
-------
10
8 -
X
E
U
o-
•73
6 -
2 .
k = 1.0 x 10 cm/sec
150
Figure 4. Rate of change In Lateral flow rate (Q ) with head for different hydraulic conductivities (k)
-------
sand and the loam (k = 3.4 x 10 cm/sec)
varies from zero to 1.07 cm day . A daily
time step would provide sufficient accuracy
if the soil has an effective lateral con-
ductivity like that of loam, but a shorter
time step is required to simulate lateral
flow accurately in a sand when y is greater
than about 30 cm. A layered soil column
consisting of loam over sand, however,
would be simulated accurately when y is
below about 50 cm. If the drainable poros-
ity were reduced, the rate of change of Q
would be proportionately higher.
To assure acceptable accuracy, simu-
lations were performed varying the compu-
tational time step from one day to
96 day . The accuracy of simulation was
found to converge rapidly as the time step
was reduced from 1 day to 4 day and to
converge very slowly from 12 day to
96 day . In an 18-day simulation of
drainage from a layered, loam over sand,
profile, four time steps per day produced
an error that was less than 0.3 cm
although y was initially at 1 m. It was
concluded that four time steps per day
( t = 6 hrs.) would give sufficient accu-
racy for most applications. The error in-
troduced from this decision is expected to
be small because the high values of y,
which causes this problem, do not occur
with high frequency or last very long.
Use of HELP in Landfill Design
The HELP model, although it is
strictly deterministic in structure, is in-
tended for use in evaluating landfill de-
sign stochastically. The idea here is that
the sequence of precipitation and ET demand
is a stochastic process, and so the model
output is a stochastic process. Therefore,
it Is reasonable to test a landfill design
with the cover off by simulating twenty
years of record then analyzing the fre-
quency of occurrence of certain water table
depths, percolation rates, or volumes of
leachate to be collected and treated. For
this reason it is necessary that the model
perform in a realistic manner and that the
prec lpitation-evapotr
-------
TABLE 1. INPUT DATA FOR EXAMPLE
Parameter
Value
Location
Simulation period
Vegetative cover
Evaporative zone depth
Are,i of site
Number of layers
Layer 1
Layer type
Soil type
Layer thickness
Layer 2
Layer type
Soil type
Layer thickness
Drainage length
Slope at bottom
Layer 3
Layer type
Soil type
Layer thickness
New Orleans, LA
5 years (default data)
fair grass (default data)
20 inches
240,000 sq ft
3
1 (vertical percolation)
12 (silt/loam, default)
24 inches
2 (lateral drainage)
1 (coarse sand, default)
12 inches
175 feet
3 percent
3 (barrier soil without synthetic membrane)
20 (low-permeability clay barrier soil, default)
24 inches
TABLE 2. ANNUAL TOTALS
Water Budget Component
1974
Precipitation
Runoff
Eva po transpiration
Percolation from base of cover
Drainage from base of cover
Soil water at start of year
Soil water at end of year
Snow water at start of year
Snow water at end of year
1975
Precipitation
Runoff
Evapo transpiration
Percolation from base of cover
Drainage from base of cover
Soil water at start of year
Soil water at end of year
Snow water at start of year
Snow water at end of year
(Inches)
72.79
10.29
50.26
1.68
8.98
21.00
22.57
0.0
0.0
80.50
13.79
52.08
1.79
13.60
23.57
21.81
0.0
0.0
(Cu. Ft.)
1,456,000.
205,800.
1,005,000.
33,640.
179,700.
420,000.
451,400.
0.
0.
1,610,000.
275,800.
1,042,000.
35,760.
272,000.
451,400.
436,200.
0.
0.
Percent*
100.00
14.14
69.05
2.31
12.34
100.00
17.13
64.70
2.22
16.89
(continued)
41
-------
TABLE 2. (continued)
Water Budget Component
1976
Precipitation
Runoff
Evapo transpiration
Percolation from base of cover
Drainage from base of cover
Soil water at start of year
Soil water at end of year
Snow water at start of year
Snow water at end of year
1977
Precipitation
Runoff
Evapo transpiration
Percolation from base of cover
Drainage from base of cover
Soil water at start of year
Soil water at end of year
Snow water at start of year
Snow water at end of year
1978
Precipitation
Runoff
Evapo transpiration
Percolation from base of cover
Drainage from base of cover
Soil water at start of year
Soil water at end of year
Snow water at start of year
Snow water at end of year
(Inches)
47.36
4.21
34.80
1.37
5.16
21.81
23.63
0.0
0.0
71.81
12.64
42.94
1.90
16.44
23.63
22.52
0.0
0.0
76.85
18.24
45.34
1.89
14.30
22.52
19.60
0.0
0.0
(Cu. Ft.)
947,200.
84,260.
696,000.
27,360.
103,100.
436,200.
472,600.
0.
0.
1,456,000.
252,700.
858,800.
37,920.
328,900.
472,600.
450,500.
0.
0.
1,537,000.
364,700.
906,900.
37,800.
286,000.
450,500.
391,988.
0.
0.
Percent*
100.00
8.90
73.48
2.89
10.89
100.00
17.35
58.98
2.60
22.58
100.00
23.73
59.01
2.46
18.61
*Percent of annual precipitation.
TABLE 3. AVERAGE MONTHLY TOTALS FOR 1974 THROUGH 1978
Water Budget Component
Precipitation (inches)
Runoff (inches)
Evapo t ranspiration
(inches)
Percolation from base
of cover (inches)
Drainage from base of
cover (inches)
Jan/Jul
6.54
6.58
1.62
0.55
2.53
5.48
0.15
0.14
1.42
0.86
Feb/Aug
3.66
9.84
0.68
1.93
2.58
5.56
0.16
0.14
1.59
0.92
Mar/Sep
4.37
5.91
0.51
1.36
4.13
4.32
0.15
0.15
1.16
1.25
Apr/Oct
4.55
3.16
0.84
0.30
3.50
2.51
0.14
0.13
0.69
0.76
May/Nov
7.01
7.12
1.13
1.45
4.93
2.78
0.13
0.13
0.60
0.58
Jun/Dec
5.95
5.37
0.44
1.03
4.26
2.53
0.15
0.15
0.66
1.20
42
-------
TABLE 4. AVERAGE ANNUAL TOTALS FOR 1974 THROUGH 1978
Water Budget Component
(Inches)
*Percent of average annual precipitation.
(Cu. Ft.)
TABLE 5. PEAK DAILY VALUES FOR 1974 THROUGH 1978
Percent*
Precipitation
Runoff
Evapo transpiration
Percolation from base of cover
Drainage from base of cover
70.06
11.83
45.09
1.72
11.70
1,401,000.
236,700.
901,700.
34,500.
233,900.
100.00
16.89
64.35
2.46
16.69
Water Budget Component
Maximum veg. soil water (vol/vol)
Minimum veg. soil water (vol/vol)
(Inches)
(Cu. Ft.)
Precipitation
Runoff
Percolation from base of cover
Drainage from base of cover
Head on base of cover
Snow water
8.52
4.80
0.02
0.12
35.6
0.00
170400.
97680.
493.
2474.
0.
0.54
0.24
without knowledge of computer programming.
The interactive system includes default
values for all model parameters and repre-
sentative precipitation and temperature
files so that the model can be used in most
areas of the United States.
HELP is an extension of the HSSWDS
model, which was based on the CREAMS model
developed by USDA. Significant refinements
have been introduced to the model compo-
nents for computation of evapotranspiration
and water routing. The evapotranspiration
component includes a representation of bio-
logical control mechanics which limit the
rate of evapotranspiration when atmospheric
demand is excessive, and a mechanism by
which the occurrence of drought (wilting
point moisture levels) prevents leaf area
index (LAI) from increasing. Whereas pre-
viously (in HSSWDS) any moisture which
passed through a shallow root zone was
assumed to percolate into the waste cell,
the HELP model allows a small but signifi-
cant rate of moisture withdrawal by evapo-
transpiration to a depth of 120 cm.
The capability of simulating lateral
flow above a sloping barrier layer and a
rate of percolation through the barrier
adds further options to the analysis of
landfill designs. The vertical flow model
is based on Darcy's Law of flow in porous
media, with modification for unsaturated
flow. The lateral flow component is based
on the Boussinesq equation with a steady-
state approximation. The vertical and
lateral flow submodels are linked by an
iterative procedure that has been optimized
to achieve high numerical accuracy while
maintaining an efficient longterm simula-
tion.
43
-------
REFERENCES
1. Perrier, E. R., and A. C. Gibson.
Hydrologic Simulation on Solid Waste
Disposal Sites. EPA-SW-868,
U. S. Environmental Protection Agency,
Cincinnati, OH, 1980. Ill pp.
2. Schroeder, P. R., and A. C. Gibson.
Supporting Documentation for the Hy-
drologic Simulation Model for Esti-
mating Percolation at Solid Waste Dis-
posal Sites (HSSWDS). Draft Report,
U. S. Environmental Protection Agency,
Cincinnati, OH, 1982. 153 pp.
3. USDA, Soil Conservation Service. Nat-
ional Engineering Handbook, Section 4,
Hydrology. U. S. Government Printing
Office, Washington, D.C., 1972.
4. Freeze, R. A., and J. A. Cherry.
Groundwater. Prentis-Hall, Englewood
Cliffs, N.J., 1979. 604 pp.
5. Ritchie, J. T. A Model for Predicting
Evaporation from a Row Crop with In-
complete Cover. Water Resources Re-
search, Vol. 8, No. 5, 1972.
pp. 1204-1213.
6. Knisel, W. J., Jr., Editor. CREAMS, A
Field Scale Model for Chemical Runoff
and Erosion from Agricultural Manage-
ment Systems. Vols. I, II, and III,
Draft Copy, USDA-SEA, AR, Cons. Res.
Report 24, 1980. 643 pp.
7. Shanholtz, V. 0., and J. B. Lillard.
A Soil Water Model for Two Contrasting
Tillage Practices. Bulletin 38, Vir-
ginia Water Resources Research Center,
VPISU, Blacksburg, VA, 1970. 217 pp.
8. Saxton, K. E., H. P. Johnson, and
R. H. Shaw. Modeling Evapotranspira-
tion and Soil Moisture. In: Proceed-
ings of American Society of Agricul-
tural Engineers 1971 Winter Meeting,
St. Joseph, MI, 1971. No. 71-7636.
9. Sudar, R. A., K. E. Saxton, and R. G.
Spomer. A Predictive Model of Water
Stress in Corn and Soybeans. Trans-
action of American Society of Agri-
cultural Engineers, p. 97-102, 1981.
10. Skaggs, R. W. Modifications to
DRAINMOD to Consider Drainage from and
Seepage through a Landfill. Draft
Report, U.S. Environmental Protection
Agency, Cincinnati, OH, August, 1982.
44
-------
GEOTECHNICAL QUALITY ASSURANCE OF CONSTRUCTION OF DISPOSAL FACILITIES
S. Joseph Spigolon and Michael F. Kelley
Soil Mechanics Division
Geotechnical Laboratory
Waterways Experiment Station
Vicksburg, Mississippi 39180
ABSTRACT
This paper summarizes the technical resource document. The need for a quality
assurance program as a management practice which assures the protection of the human
health and environment is discussed within the framework of the EPA regulations promul-
gated on 26 July 1982. The permittee of a hazardous waste disposal facility is shown
to be somewhat unique in the construction industry given the three responsibilities:
design; construction and operations; and quality assurance; all consolidated with him.
The geotechnical parameters which should be tested, observed and documented during the
construction, operation and closure of a disposal facility are identified. The limita-
tions of design properties tests and the applicability of index properties tests are
discussed. A summarized statistical analysis of sampling size, frequency and distri-
bution to support a hazardous waste disposal facility quality assurance program is
presented. Accepted and commonly used methods for testing water content; unit weight;
specific gravity; grain-size distribution; liquid limit; plastic limit; cohesive soil
consistency; water content/density/compactive effort; cohesionless soil relative density;
and geomembranes/geotextiles are identified. The criteria for the selection of a partic-
cular test method are discussed. A quality assurance program, suitable for the somewhat
unique permittee of a hazardous waste disposal facility, is presented. Considerations
for such a quality assurance program are also addressed.
INTRODUCTION
As stated in the Federal Register,
26 July 1982, "EPA wants to make sure that
the issuance of a RCRA (Resource Conser-
vation and Recovery Act) permit for a
facility means that a certain level of
protection is provided and that the pub-
lic can be assured that the prescribed
level of protection will be achieved."
"Section 3004 of RCRA provides that
EPA has authority to issue regulations
covering owners or operators of (hazard-
ous waste) treatment, storage and dis-
posal facilities as may be necessary to
protect human health and the environ-
ment." Given this authority the EPA
has charted a two "line of defense"
protection strategy for hazardous waste
disposal facilities. This strategy
calls for facility design and operating
standards to prevent contamination and a
monitoring and response program to detect
and correct a condition of contamination
should it occur. The EPA believes that
adequate protection will be afforded if
the technical and environmental perfor-
mance standards promulgated to execute
their two "line of defense" strategy are
met. A quality assurance program for the
construction, operation and closure of a
hazardous waste disposal facility, though
not mandated by the current regulations,
is a "management practice" which can
assure that these standards are met, that
45
-------
the strategy is executed and that human
health and the evironment are protected.
Objective of Study
The study is intended to provide
technical background for use by designers,
specifiers, quality assurance engineers,
and permit writers. It addresses the
subject of quality assurance of site geo-
technical parameters during the construc-
tion, operation, and closure of landfills,
surface impoundments, waste piles, and
land treatment units. It does not address
injection wells or seepage facilities.
Scope of the Study
Four major topics are discussed as
they relate to the responsibilities of the
permittee during construction, operation,
and closure of a hazardous waste disposal
facility:
(1) Those geotechnical parameters that
should be tested and/or observed and
documented are identified and de-
scribed. The reasoning for their
selection is given.
(2) Sampling and frequency of testing are
discussed. Random sampling plans,
sample size selection, and calibration
of test equipment are presented using
rational, unbiased statistical
methods.
(3) The commonly used laboratory and field
testing and/or observation methods,
needed for investigation of the geo-
technical parameters, are identified
and described. A discussion of merits
and limitations of the various methods
is given. Criteria are proposed for
selection among alternatives.
(4) A quality assurance program, suited to
the unique construction responsibil-
ities of the permittee, is presented
and discussed. The documentation
necessary to provide a valid record
of the quality of the work is
recommended.
The Unique Permittee
In most areas of construction there
is a contractual relationship between the
owner, whether public or private, and the
contractor. The owner's designers prepare
plans and specifications. The contractor
is expected to use those "control of
quality" procedures necessary to control
his materials and workmanship to comply
with the plans and specifications. The
owner's inspectors perform those accep-
tance sampling and testing functions
necessary to give confidence to the owner
that the materials and workmanship pro-
vided are acceptable. This then results
in a near adversary relationship between
the owner and the contractor.
The permittee of a hazardous waste
disposal facility is somewhat unique in
the construction industry. Because of
EPA policy, the permittee is responsible
to the EPA for all three construction
related functions:
(1) The Design Function: Including site
selection, site characterization,
analysis and design, plans and speci-
fications, preparation of the permit
application, and verification of
design for exposed site media.
(2) The Construction and Operations
Function: Provision of all mate-
rials and workmanship.
(3) The Inspection Function: All quality
assurance activities, including
sampling, testing and/or observation,
and documentation.
That he is responsible for both con-
struction and inspection poses a poten-
tial conflict of interest. The quality
assurance program presented later in this
paper accounts for the permittee's unique
situation and this potential conflict of
interest.
GEOTECHNICAL PARAMETERS
In the design of a hazardous waste
disposal facility the permittee's design
group makes estimates of the engineering
properties of the natural site soils and
assumptions about the engineering prop-
erties of the constructor-placed mate-
rials. This requires quantitative and
specific information on the physical and
strength properties of soils that have
direct and valid application to the
design problem (ASTM, 1970). These
properties include permeability,
46
-------
compressibility, and shear strength. The
function of the quality assurance program
during construction is to assure the
standard of quality specified by the de-
sign is achieved. Ideally, quality
assurance observations and tests should
measure the design properties directly.
Design Properties Tests
The geotechnical tests used to define
the engineering properties of soils are:
(1) Permeability
a.. Laboratory permeability tests
b_. Laboratory capillarity tests
Ł. Field permeability tests
(2) Compressibility
a.. Laboratory consolidation tests
b_. Swell pressure tests
(3) Shear strength
a_. Laboratory direct shear
b_. Laboratory triaxial compression
c^. Laboratory unconfined compression
Limitations of Design Properties Tests
Although these are the tests that
should be made as part of a quality assur-
ance program, they are not normally used
because they require (ASTM, 1970; DA,
1970; and Terzaghi, 1967):
a_. Extensive time: The performance of
any of these tests can take from one
full day to as long as two weeks.
Efficient control of construction
requires much faster test times.
b_. Complex equipment.
c_. Complex procedures.
cU Highly trained operators.
e_. Controlled laboratory environment:
This implies temperature control and
vibration-free surroundings not nor-
mally found at a construction site.
f_. High quality undisturbed samples.
Instead, various simpler, faster, index
tests, whose results relate to permeabil-
ity, compressibility and shear strength,
are performed.
Criteria for Inspection Tests and
Observations
Construction efficiency and control
of quality require that nearly continuous
tests and observations be made during the
progress of the work. Therefore, an
inspection test or observation should:
Ł. Be a good indicator of a design
property. It should correlate well
with the appropriate engineering
design property by itself or with
other, similar tests.
b_. Be accurate and precise. Accuracy
can be improved by calibration and
should be stable. Precision should
be within acceptable, known limits.
c_. Have results quickly available. The
test or observation should be capable
of being performed in time for an
acceptance decision to be made so
that minimum interference with
contractor performance occurs.
d_. Be easily run using simple, rugged
equipment. The test or observation
should not require extensive train-
ing of the operator or require judge-
ment decisions on his part.
e_. Not require an undisturbed sample,
except for density.
f_. Provide test/observation results
that are documentable. The test or
observation should be a number or a
well defined term or phrase.
Ł. Preferably be non-destructive. Field
tests often require sample removal
and replacement or patching.
Index Properties Tests
In the practice of geotechnical
engineering, there has been developed a
small group of laboratory tests called
indicator, or index properties, tests.
These are physical tests that have proven
to be suitable for quality assurance
47
-------
programs, and some of which have been used
in such and similar applications for over
50 years.
The index properties tests useful
in a quality assurance program may be
placed into three categories:
(1) Weight-volume relationships.
a_. Water content test
b_. Density test
c_. Specific gravity test
cl. Calculated values — void ratio,
porosity, and degree of saturation
(2) Soil classification tests.
a. Grain size distribution tests
b_. Atterberg limits tests
c^. Consistency tests — using simple
field methods
cl. Calculated values — plasticity
index, liquidity index, activity,
and USCS Classification
(3) Laboratory compaction.
ja. Density/water content/compactive
effort tests; for cohesive soils
and "dirty" granular soils
b_. Maximum and minimum density tests;
for clean granular soils
All of the index properties result from
standardized soils tests.
Construction Processes
The main tests/observations that
should be made for the various construc-
tion processes are described below. The
emphasis is on a quality assurance pro-
gram that focuses on catching mistakes
before or as they occur, correcting them,
testing for specification compliance, and
documenting the entire process to provide
the necessary confidence in the inspec-
tion procedure. This requires continuous,
cooperative interaction between the design
group, the construction group, and the
inspection group. The tests and/or
observations that should be made and docu-
mented are:
(1) Site preparation.
a_. Observation of the newly exposed
ground surface to confirm site
characterization
b_. Observation of surface drainage
contouring
c^ Observation of behavior of the
drainage system
(2) Excavation.
a.. Observation of exposed walls and
floor for site media verification
b_. Observe construction surveys for
exact trench locations and slopes
c^. Observation of excavation related
soil movements, using surveying
methods
jd. Observation of placement of
erosion control measures
(3) Foundation preparation.
a_. Observation of stripping
b_. Observation of soil and rock
surfaces
Ł. Observation of construction of
underseepage cutoffs
(4) Compacted earth fill.
a_. Quality of the borrow soils
b^. Atmospheric conditions recorded
jc. Equipment suitability
cl. Lift thickness
e_. Compactive effort (number of
passes) and uniformity of rolling
f_. Compacted density and water
content
(5) Liner materials and geotextiles.
a. Identification of materials used
48
-------
b_. For admixed materials, the appli-
cation rate, blending, spreading,
rolling, layer thickness, and
density (see Compacted earth fill
above)
c_. For geomembranes and geotextiles,
the identification of material,
smoothness of placement surface,
method of placement, and
imperfections
ci. For geomembranes and geotextiles,
the seam overlap, seam sealing,
and seam integrity tests
(6) Hydraulic collector.
&. Observation of the surface of the
impermeable base for imperfections
b_. Observation of the amount and
direction of slope
c^ Quality of permeable materials
ei. Tests of placement of granular
materials
Variability in measurements occurs as a
result of (a) material composition vari-
ability, (b) placement (formation) process
variability, and (c) measurement process
variability. All measurements contain
variability from all three causes combined.
Each individual measurement of a
quality characteristic may be considered
to be one of a "countably infinite"
number of similar measurements that could
be made in a universe of the material.
The difference or error between the
measured value and the "true" or reference
value is the sum of components, an ac-
curacy (systematic or calibration) error
and a precision (random) error. This
can be written as (ASTM, 1982):
X = yR + 6 + e
and
where
(1)
(2)
X = numerical value of a measurement
SAMPLING TECHNIQUES
The method and frequency of sampling
and testing in construction had tradi-
tionally been based on intuitive decisions
of the sampler or his superiors. The
purpose of sampling and testing is to pro-
vide data for evaluation of a portion of
the work. Unless sampled by probability,
or random, sampling methods, the result
is often a biased and nonrepresentative
estimate of a quality characteristic.
Frequency of sampling and testing can only
be rationally discussed in statistical
analysis terms. It is a function of
desired level of confidence, maximum
error or percent error, and the actual
variability of the tested quality
characteristic in the sampled portion of
work. Comparison of standard and alterna-
tive test methods can be made on the basis
of sample sizes needed for equivalence of
confidence levels.
Frequency of Sampling and Testing
There is an inherent variability in
measurement data for any specified quality
characteristic of any soil mass, whether
natural or construction-processed.
y_ = true or accepted reference
level of the property of the
material to be measured
Up = expected value or the average
of many observations recorded
for the process
5 = correction for consistent or
systematic error of the measure-
ment process at the reference
level u_
e = correction for random deviation
about the average of the
observations y
Accuracy refers to the degree of
agreement of individual measurements or
the average of a large number of measure-
ments with a "true" or accepted refer-
ence value. This implies the idea of a
consistent deviation, or systematic or
bias error, which is fixed or common
contribution to error in each of a sample
of measurements. Precision is the degree
of mutual agreement among individual
measurements of a consistent material
made under prescribed, like conditions.
The imprecision of measurement may be
49
-------
characterized by the standard deviation
of the errors of measurement.
In Equation 1, the 6-term for cor-
rection for systematic, or accuracy,
error is the sum of corrections from
each of the three causes of nonrandom
variation. This is then written:
6 =
(3)
where
6 = correction for systematic error
due to material composition
6 = correction for systematic error
due to placement process
S - correction for systematic error
due to testing process
Because the effects of material com-
position, placement process, and testing
process are interrelated, it is not
usually possible to differentiate these in
a group of measurements. If the material
composition and placement process effects
are combined into a material quality
effect, then:
and defining
then
where
2 2
O = CTQ
(7)
(8)
a = overall variance
a = variance due to material
M . .
composition
a = variance due to placement
process
o = variance due to testing process
2
CT = variance due to material quality
If the variance due to a particular
2
testing process,
is known, or can
be estimated, then its effect on the over-
2
all variance, a , can be evaluated
from Equation 8 as:
and
- 6M + 6P
(4)
"o • •„
(9)
6 =
(5)
where
6_ = correction for systematic error
of the measurement process at
the reference level UR due to
the material quality, the com-
bined effects of material com-
position and placement process
The precision term of Equation 1,
e , is a random effect, and is the combi-
nation of random effects due to the three
causes discussed above. The statistical
variace, en , of a group of measure-
ments of a characteristic of a consistent
material using a consistent test method
is the sum of the variance due to the
three causes of random variability. This
may be given as:
2
CT0 =
(6)
where
CT_ = total standard deviation
a = quality standard deviation
CT = testing standard deviation
When comparing alternative test
methods, the effect of different standard
deviations of the test methods on the
overall standard deviation should be
considered. It can be seen from Equa-
tion 9 that when material quality is
uniform (low aq) the precision of the test
method (CT^) is of great importance.
However, when material quality is not uni-
form (high CTQ) the relative precision of
one test method (o^) versus another is of
little overall importance.
Types of Sampling
Probably the most satisfying method
50
-------
to the engineer concerned about sampling
all parts of the block is a combination
of (a) stratified random sampling and
(b) systematic sampling with a random
start.
If a universe, or block, consists of,
or can be divided into, a group of strata,
then a random sample can be drawn from
each stratum instead of drawing a single
random sample from the entire population.
This is called stratified random
sampling. The division of a soil mass
or a construction process into blocks is
a form of stratified random sampling.
A popular sampling method is
systematic sampling with a_ random start.
This method involves the selection of
successive sample units at uniform inter-
vals of time, distance, area, or volume.
It is argued that if the first sample
unit is selected at a random location,
then all successive units are random also.
By subdividing each block of work
into a convenient large number of sub-
blocks, using a random start, and follow-
ing a systematic pattern, a random sample
with uniform coverage of the block is
obtained.
Selection of Sample Size
A statistically rational and valid
method of selecting sample size is given
in ASTM (1982) Designation E 122-72,
Standard Recommended Practice for Choice
of Sample Size to Estimate the Average
Quality of a Lot or Process. The equation
for the number of units (sample size, n)
to include in a sample in order to esti-
mate, with a prescribed precision, the
average of some characteristic of a lot
is:
n = (ta'/E)'
(10)
or, in terms of coefficient of variation
,2
n = (tV'/e)'
(11)
where
n = number of units in the sample
t = a probability factor, from the
Tables of Students'-t
a' = the known or estimated true
value of the universe, or lot,
standard deviation
E = the maximum allowable error
between the estimate to be
made from the sample and the
result of measuring (by the
same methods) all the units in
the lot
V1 = coefficient of variation
= CT'/X , the known or estimated
true value of the universe or
lot
e = E/X , the allowable sampling
error expressed as_a percent
(or fraction) of X'
X = the expected (mean) value of
the characteristic being
measured
TESTING
Previously in this paper the geo-
technical parameters, that should be
observed or tested during the construction
of a disposal facility, were identified.
(For the purpose of this paper, an
observation usually does not involve a
formal or specific method, whereas a test
almost always involves a specific tech-
nique and often a specific piece of equip-
ment.) Standard laboratory or field test
methods exist for each of the soil index
properties that were identified and
defined.
This section of the paper identifies
the standard test methods for the soil
index properties. For most of these,
there exist commonly used alternative
test methods; these are also identified.
The following discussion includes
only those tests that were identified
earlier in the paper. They are the ones
considered most appropriate for both
process control and acceptance inspection
during construction. Although not dis-
cussed here, other tests may be used for
special investigations during construc-
tion, but usually not on a routine basis.
These include such tests as the engineer-
ing properties tests (mentioned earlier),
geochemical tests, hydrology-related
tests, off-site evaluations of off-site
51
-------
materials, and material-specific tests
of nonsoil materials such as portland
cement, asphalt, bentonite, and others.
Criteria for Test Method Selection
With the exception of the nuclear
gage moisture/density test, the density
part of volumetric field density tests,
the air lance test, the vacuum seam test
and the conductivity test, all of the
standard test methods require (a) procure-
ment of a sample from the field and
(b) performance of a test in a laboratory
environment. This usually involves the
availability of electric power, a water
supply, freedom from dust and vibration,
and controlled temperature humidity.
These are conditions found in a central
testing laboratory, and are available
only in a few well-equipped construction
site laboratories, usually found only
on large construction projects. In
response to this, and for the other
reasons given below, alternatives to
the standard test methods have come into
common use for construction quality
control. The criteria for selection
among alternatives include:
a_. Time; speed of testing and/or
speed of obtaining test results.
The efficient use of men and
machinery on a construction
site demands that the results
of quality control testing be
available quickly. Construction
costs can be one or more hundred
dollars per hour. The cost of
rapid testing must be compared
to the total cost to the
permittee due to a slower test
method.
b_. Correlation with the standard
test method. A reasonable
reproducibility of test results
must be obtainable if test
operator or specific test equip-
ment is changed. Accuracy must
be stable so that a calibration
correction can be made; preci-
sion must be such that a
reasonable number of tests can
be averaged to yield the
required confidence.
c_. Adaptability to field laboratory
conditions. The absence of
electric power, the absence of
water, and/or the absence of
environmental controls can
effectively eliminate certain
standard test techniques.
d_. Adaptability to on-site use.
For example, soil density is
more readily measured in place
rather than by use of an undis-
turbed sampling method with
laboratory measurement of
density.
e_. Equipment cost; comparison can
be made in terms of capital cost
of test equipment, relative
operator cost, cost of field
laboratory services, and need
for rapid test method. Expen-
sive, automated equipment
cannot be justified if less
costly apparatus will provide
equivalent results in sufficient
time for effective action,
especially during small, slow-
moving construction processes.
_f_. Simplicity and ruggedness of
equipment; acceptable precision
and sensitivity of equipment
can often be achieved with
apparatus that is simpler to
use and more durable under field
conditions than more sophisti-
cated devices. The level of
necessary operator training,
and operator fatigue and/or
boredom, can seriously affect
repeatability of a test method.
Ł. Nondestructive testing; this is
of particular importance where
the act of sampling causes a
discontinuity that must be re-
paired. For example, securing
a test coupon from a geomembrane
seam requires placement of a
secure patch to maintain
impermeability.
Soil Index Properties Test Methods
The listing given below includes the
standard laboratory test method for each
of the index properties. In most
instances, alternative standard or non-
standard methods are in common use. There
are undoubtedly other alternative methods
for some of the tests that are used by
various testing groups for construction
52
-------
(1) Water content test
a. Standard oven-dry method.
b_. Standard nuclear moisture/density
gage method.
Ł. Gas burner (frying pan) method.
d_. Alcohol burning method.
Ł. Calcium carbide (speedy) method.
_f. Infrared oven or hot air method.
(2) Unit weight test
a.. Standard laboratory volumetric
method.
b_. Standard laboratory displacement
method.
Ł. Standard field sand cone method.
d_. Standard field water balloon
method.
e^. Standard field drive cup method.
f^. Standard nuclear moisture/density
gage method.
(3) Specific gravity test
Standard laboratory method.
(4) Grain-size distribution test
a.. Standard sieve analysis
~~ (+200 fraction) method.
b_. Amount of soil finer than No. 200
screen (wash) standard method.
c^. Standard laboratory hydrometer
(-200 fraction) method.
d_. Pipette method for silt and clay
fraction.
e^. Decantation method for silt and
clay fraction.
b. Standard one-point me
(6) Plastic limit test
Standard laboratory method.
(7) Cohesive soil consistency test
a.. Standard unconfined compressioi
test method.
b_. Field expedient unconfined com-
pression test method.
c^. Hand (pocket) penetrometer method
cL Hand-held Torvane method.
(8) Water content/density/compactive
effort tests
a_. 25-blow standard Proctor compac-
tion method.
1}. 25-blow modified Proctor compac-
tion method.
Ł. Nonstandardized Proctor compac-
tion methods.
cL Rapid, one-point Proctor compac-
tion method.
Ł. Rapid, two-point Proctor compac-
tion method.
Ł_. Hilf's rapid method.
Ł. Ohio Highway Department nest of
curves method.
h_. Harvard miniature method.
(9) Cohesionless soil relative density
test
a_. Standard laboratory maximum
density method.
b_. Standard laboratory minimum
density method.
c. Modified Providence method.
53
-------
(10) Geomembrane/Geotextile test
a_. Bonded seam strength method.
b_. Breaking strength method.
Ł. Peel adhesion method.
d_. Air lance method.
e_. Vacuum seam method.
f_. Conductivity method.
Observation Techniques
Observations are not dependent on the
specifics of a particular technique. Most
involve visual evaluations, or simple
counting and tabulation, or are simple
length or thickness measurements. The
usual procedures for these observations
are either straightforward, requiring
no special instruction or equipment, or
they involve visual comparisons that are
either site-specific or material-specific.
Specification writers and permit writers
should simply state the desired end
result and leave the determination of
technique to the experience and ingenuity
of the inspector or his superior.
As an example of the statements made
above, the smoothness and freedom from
sharp objects of the soil surface to
receive a geomembrane can only be accom-
plished by 100 percent visual inspection.
This is also true in the observation for
freedom of the placed membrane from tears.
The thickness of a soil lift prior to
compaction can be quickly and easily
checked with a marked rod, or even a
suitably marked spade. The observation
of a freshly cut trench wall or floor
starts with visual inspection for texture
and color patterns, supplemented by field
expedient soils tests using the visual-
manual procedure. Sampling and testing
plans for site media verification studies
can then be effectively made. The re-
sults of the observation depend more on
the technical expertise and willingness
of the inspector than on specific details
of technique.
THE QUALITY ASSURANCE PROGRAM
As stated earlier in this report the
permittee of a hazardous waste disposal
facility is somewhat unique in the con-
struction industry. Because of his unique
situation the permittee is confronted with
a potential conflict of interest between
his construction and operations responsi-
bility and his quality assurance
responsibility.
Those factors that make his situation
unique, and that affect the quality assur-
ance program, are:
(1) The permittee has total responsibility
for design, construction, and inspec-
tion within his own organization.
(2) The permittee must give confidence,
or assurance, to the EPA in the
quality of his work.
(3) The permittee passes judgement on his
own work.
(4) A disposal facility can be a rela-
tively small project with only a few
workmen and pieces of equipment
operating at any time.
(5) The permittee1s organization is rela-
tively informal, with sometimes
blurred lines of authority and
responsibility.
Considerations for a Quality Assurance
Program
The permittee should consider all of
the following before establishing a qual-
ity assurance plan: quality personnel
management; attitude of the workmen,
construction management, and inspectors
toward quality; the nature and cost of a
quality assurance program; th« human
factor; and the problem of scale.
Quality Personnel Management within
the Permittee's Organization. The
permittee can be assured that his desired
standard of quality is achieved by the
implementation of an effective quality
assurance program within his organiza-
tion. Such a program would undoubtedly
include management practices with the
objective of effecting worker performance.
For the purpose of this paper, the
implementation of management practices
to effect worker performance, such as
those recommended by Harris and Chaney
54
-------
(1969), shall be called quality personnel
management. Quality personnel management
includes but is not limited to practices
such as proper supervision, personnel
selection, inspection techniques, and
quality assurance jobs definitions. It
also includes measures to enhance worker
education, participation and satisfaction.
It includes other positive motivators
such as measures to enhance worker identi-
fication with the work, including having
workers sign their work. It also includes
negative motivators which "push" towards
the desired standard, e.g. the threat of
a reprimand, as opposed to positive
motivators which "pull." In summary
quality personnel management includes
all the management practices which effect
worker performance to give a measure of
assurance that the desired standard of
quality is achieved.
Quality personnel management within
the permittee's organization is a respon-
sibility of the permittee. Also quality
personnel management within the per-
mittee 's organization is, in general, of
no more peculiar significance in the con-
struction of a hazardous waste disposal
facility than it would be in any other
construction endeavor undertaken by the
permittee's organization.
Workman/Management/Inspector
Attitude Towards Quality. The individual
workman is the ultimate determinant of
quality. There is a continual conflict
between his desire to achieve high
productivity and to do the work "right"
as he knows it. He, and his supervisors,
may feel that the project is overdesigned
at times. They may then develop a "rebel"
attitude, in which they will show the
inspector that they have as much knowledge
as the designers.
The inspector can be part of a team
or can be the adversary of construction.
He may develop a bias towards or against
the construction group. The inspector
must be qualified for his work and must
not be given the job as a reward for long
service.
The Nature of a Quality Assurance
Program. The nature of a quality assur-
ance program should match the nature of
the design. That is, the design for a
relatively high standard of quality
(e.g. for a site where the complex
geology and the protection of the human
health and environment require that exten-
sive measures be undertaken) calls for a
program which gives a great deal of assur-
ance that the high standard of quality is
achieved. Conversely, the design for a
lesser standard of quality (e.g. for a
site with an ideal geology and climatic
conditions, and where little threat is
posed to the protection of the human
health and environment) does not call for
that same program which gives that great
deal of assurance that this lesser standard
of quality is achieved. Simply stated, a
design for a relatively high standard of
quality calls for a program which gives
a high assurance that the standard is
achieved. And a design for a relatively
low standard of quality calls for a pro-
gram which gives a lesser, but sufficient,
assurance that the standard is achieved.
The Human Factor. There can never be
absolute certainty that the desired qual-
ity is everywhere achieved in the con-
struction of a hazardous waste disposal
facility. There exist too many variables,
particularly given the inevitable gaps in
our knowledge of the soil conditions.
However, a quality assurance program does
give a measure of assurance, of confi-
dence, that the standard of quality de-
sired and designed for is achieved. The
better the quality assurance program the
greater this measure of confidence can
be. And given the EPA's two-pronged,
prevention and cure, protection strategy,
the inevitable uncertainty of the quality
has been accounted for by the regulatory
provision of the monitoring and remedial
action procedures.
One significant variable that can be
addressed and limited by a quality assur-
ance program is the human factor. Harris
and Chaney treat this entire subject in
their text, Human Factor in Quality
Assurance. Their text is recommended to
the reader interested in all aspects of
the human factor in quality assurance.
The remainder of this section will focus
on the human factor variable as it is of
peculiar significance in the considera-
tion of a quality assurance program for
the construction of a hazardous waste
disposal facility.
Through its permit issuing agency,
EPA can be assured that the desired
standard of quality is designed for by
55
-------
the permittee. But what assurance is there
that the standard of quality contained in
the design is achieved in the construction
and operation of the facility?
The permittee suffers from an inher-
ent conflict of interest in the fulfill-
ment of his construction and operations
responsibilities vis-a-vis the fulfill-
ment of his quality assurance responsi-
bilities. Possible manifestations of
this conflict of interest are obvious.
What does the permittee do when told by
his quality inspector that an entire block
of work is rejected and must be done again
because it is just barely below the
standard of quality in the design? What
does the permittee do when motivated by
the opportunity to maximize profits and
believing to himself that government
approved designs always include an
excessive margin of safety?
This conflict of interest, this human
factor, must be addressed in an effective
quality assurance program.
The Problem of Scale. As stated
earlier, the permittee of a hazardous
waste disposal facility is somewhat uni-
que. Although there are other organiza-
tions in the construction industry which
also have all these responsibilities
(design, construction and operations,
and quality assurance) and thus are simi-
lar in concept, there exist significant
differences between these organizations
because of the problem of scale. An
example of such a seemingly similar
organization is that of an organization
constructing a nuclear power facility.
In the construction of a nuclear
power facility a large organization, what
might be considered an institution, has
the three responsibilities. In order to
protect the reputation of the organiza-
tion, the present and future jobs of its
employees, the huge financial stake in
the project, and the professional status
of its engineers, this institution has
many positive motivators to achieve the
standard of quality contained in the
design. In order to avoid the adverse
consequences of all possible future
litigation, the institution also has a
strong negative motivator to achieve the
standard of quality contained in the
design.
In the construction of a hazardous
waste disposal facility, it is entirely
possible for a very small organization to
have all three responsibilities. Hypo-
thetically, that small organization could
be dominated by one individual. Thus,
this hypothetical organization would act
less like an institution and more in a
manner reflective of the dominant indivi-
dual's personality. Thus this organiza-
tion might be willing to cut corners and
take risks. The standard of quality
achieved by the organization would then
be that set by the dominant individual
and not that contained in the design.
So although the two organizations
described above are similar in concept,
in that both have all three responsibil-
ities, there exists the real likelihood
that they might differ in action because of
their dissimilarity in size: the problem
of scale.
The Quality Assurance Plan
It is recommended that the permittee
develop a formal, written plan of action
for implementing a QA program. This
should include such items as: an organi-
zation plan; details of QA methods; names
and qualifications of individuals for QA;
the procedures for review of data; proce-
dures for reporting; and procedures for
assigning responsibility.
Quality Assurance Documentation
Considering all of the factors dis-
cussed above, it is recommended that the
following documentation of tests and
observations be made:
(1) Daily inspection reports
(2) Data sheets for all tests and
observations
(3) Block evaluation reports, showing all
data and statistical analyses
(4) Design acceptance reports, presenting
a review of the block reports and
their comparison with design
requirements
(5) A summary and checklist for each major
section of the project made after its
completion
56
-------
The documentation program given above
is intended to provide (a) motivation for
an effective QA program, (b) data in the
event remedial action is required in the
future, and (c) factual information in
the event of litigation.
REFERENCES
1. The American Society for Testing and
Materials. 1982 Annual Book of ASTM
Standards, Part 15, "Road and Paving
Materials...", Philadelphia, 1982.
2. American Society for Testing and
Materials. 1970. Special Procedures
for Testing Soil and Rock for
Engineering Purposes. Special Techni-
cal Publication 479. Philadelphia.
3. Department of the Army. Laboratory
Soils Testing, EM 1110-2-1906, w/Chl,
Washington, 1970.
4. Harris, Douglas H., and Chaney,
Frederick B. 1969. Human Factors in
Quality Assurance. John Wiley &
Sons, Inc. New York.
5. Hald, A. 1952. Statistical Theory
with Engineering Applications.
John Wiley & Sons, Inc. New York.
6. Proctor, R. R. "Four Articles on the
Design and Construction of Rolled-
Earth Dams," Engineering News-Record,
Vol 111, 1933.
7. Terzaghi, K. and Peck, R. B. Soil
Mechanics in Engineering Practice,
John Wiley & Sons, Inc., New York,
1967. 729 pp.
57
-------
SECONDARY EMISSIONS FROM HAZARDOUS WASTE DISPOSAL LAGOONS: FIELD MEASUREMENTS
Charles Springer
Louis J. Thibodeaux
Phillip D. Lunney
Rebecca S. Parker
University of Arkansas
Department of Chemical Engineering
Fayetteville, Arkansas 72701
and
Stephen C. James
U. S. Environmental Protection Agency
Cincinnati, Ohio 45268
ABSTRACT
Field measurements of the emission of methylene chloride and toluene from surface
impoundments at a site intended for hazardous waste disposal were made by the concen-
tration profile technique. The technique employs profiles of chemical concentrations,
temperature and wind speed to quantify chemical emission rates. One significant feature
of the method is that it does not disturb the natural transport processes in effect on the
water surface. The method is limited with respect to sample time, wind speed, fetch-to-
height ratio of upwind disturbances, etc. Observations of chemical concentration in air,
wind speed, and temperature are made at six heights within the turbulent boundary layer
above the surface. The chemical concentration and velocity vs In-height profiles should
be linear over the full range of six observation heights only if the air boundary layer is
neutral. The profiles will likely be non-linear under stable and unstable micrometeoro-
logical conditions and thus display some curvature. Various equations may be used to
determine flux values of the chemicals depending on the conditions. Flux values deduced
from these measurements were compared to those calculated by the Cohen, Cocchio and Mackay
model and were found to be in general agreement, although a number of inconsistencies were
observed.
INTRODUCTION
It is probable that volatile chemicals
can be and are emitted from various types
of waste impoundments, and in particular,
from surface impoundments such as oxidation
lagoons, holding basins, evaporation ponds,
etc. Such emissions constitute possible
sources of undesirable air pollution
and have been the subject of a number of
laboratory studies and reviews (1-4) which
have proposed models and simulated waste
sites in wind tunnels in an effort to
quantify the processes.
Thus far, however, it has been diffi-
cult to establish the validity of the
models and/or predictive equations by
comparison with field measurements of
emissions from actual impoundments.
Not the least of the difficulties has
been the frequent reluctance of the
operators to allow the required access to
the waste impoundments. However, there
are also technical problems in the field
measurements as well, and in particular,
some methods will interfere with the
process of volatilization under study as
for example, floating a container or
vessel on the water surface, so as to
confine and/or capture the escaping
58
-------
vapors will seriously interfere with the
process being studies. Other methods,
such as sampling upwind and downwind,
then estimating the area emission based
on dispersion models requires the estimation
of atmospheric parameters as well as evalu-
ation of the surrounding terrain and its
effect on the dispersion. There are inade-
quate methods of accounting for other
than ideal flat terrain, and methods based
on dispersion also require relatively
small changes in wind velocity and direction
over an extended time.
One method of field measurement devel-
ed by Thibodeaux (5), called the "concen-
tration-profile" (CP) method eliminates
some of the difficulties in that it doesn't
disturb the process and relies on simul-
taneously measureable atmospheric parameters.
This method was used in the present work
and will be briefly described.
The CP method involves measurement
of the concentration gradient of a chemical
of interest above the surface. Then,
combined with simultaneously measured
air stability parameters (velocity and
temperature gradients), the average flux
(mass rate of flow per unit area) can
be determined for the immediate upwind
surface. If the upwind flux were variable
with the fetch, then this is an averaging
method.
The method was employed in the present
instance by drawing samples of air through
six different adsorbent tubes simultaneously
from six different elevations off the
water surface. Hwang (8) has recently
proposed a modification of the CP technique
which reduces the number of air samples
from six to one. This alternative has
not yet been adopted.
The CP method must be used in the
presence of a wind, prefereably between
two (2) and six (6) m/s at a height of
one to two meters above the surface.
Theoretically the method would be immune
to the presence of a uniform background
concentration of chemical since it depends
only on the concentration gradient with
height off the surface. In practice,
however, a background presence would be
undesireable, not only from a measurement
standpoint, but also would preclude testing
a model for emissions which assumes no
such background concentration. Additional
details on the use of the method are pre-
sented later. It has also been described
in detail elsewhere (6,7).
The CP method was used to determine
the emission flux for two chemicals (methy-
lene chloride and toluene) from three eva-
poration ponds located at a waste treatment
facility in Western United States.
Nature of the Site
The three equally sized ponds were
located adjacent to each other as shown
in Figure 1. The berm heights were vari-
able, but ranged from approximately three
feet to approximately twelve feet.
These ponds constitute what the facil-
ity operators call "evaporation ponds."
They regularly receive aqueous waste con-
taining organic materials and water loss
then occurs by evaporation. Some biologi-
cal decomposition is anticipated as evi-
denced by the presence of the surface
aerators on ponds No. 1 and No. 3.
The circled numbers in Figure 1 show
the approximate location of the sampling
apparatus for the samples (profiles) indi-
cated. In each case, the location was
selected so as to be downwind of the maxi-
mum expanse of pond surface with the exist-
ing wind.
The volatile priority pollutants in
each of the ponds are listed in Table 1
(10). It should be noted that other vola-
tile species were also present, but are
not indicated in Table 1.
One of the ponds, No. 1 as indicated
in Figure 1, was partially covered with
an oil slick, approximately as outlined
in Figure 1. The oil slick was moved about
considerably by the wind and tended to
collect on the downwind side of the pond.
During sampling on pond No. 1, the slick
covered only about ten (10) percent of
the lagoon beyond the oil booms. Since
the booms held a much heavier layer of
oil behind them, the wind did not expose
much water behind the oil booms.
During sampling, neither of the aera-
tors were in operation. The weather was
mostly sunny and warm, with winds ranging
from near calm to around (5.4 m/s) at a
few feet off the surface. Sampling was
done only during the daylight of two days.
59
-------
0
POND 2
©
225'
O
POND 3
FIGURE 1 - PLAT AND SAMPLE LOCATIONS
Circled numbers are sample locations
60
-------
Table 1 - Water Analysis Results
Concentrations in ug/1
Component
Methylene Chloride
Toluene
1,1,1-Trichloroethane
Benzene
Ethylbenzene
Chloroform
Tetrachloroethylene
Pond 3
(Profiles 1 & 2
10
28
57
Pond 2 Pond 1
(Profiles 3 & 4 (Profiles 5 & 6
440
350
45
14
13
330
51
670
21
18
SELECTION OF SAMPLING
Description of the Sampling Apparatus
Samples of volatile chemicals in the
air were collected and concentrated on an
adsorbent (Tenax). The apparatus consisted
of various devices installed onto a boat
for the simultaneous measurement of dry-
bulb temperature and wind speed and obtain-
ing air constituent samples. Placement of
the apparatus on a boat allowed optimal
location on the impoundment surface.
The sample collection devices (sample
concentration tubes) were short sections
of tubing packed with an adsorbent polymer
(Tenax, GC 60/80 mesh). These tubes were
15cm (5.98 inches) in length, 0.125 inch
(3.2mm) O.D. and 0.085 inch (2.2mm) I.D.
Each tube was packed with approximately
llcm of the Tenax and approximately 2.5cm
of silica gel. The tubes were precondi-
tioned by heating to 270°-300°C while
purging with helium for not less than one
hour. The tubes were fitted with Swagelok
caps on each end, and were sealed before
and after sampling. Air was drawn through
the Tenax end using a vaccuum supplied by
a hand-pumped tank.
Water sample containers were small-
mouth, amber glass, 1 liter bottles with
Teflon lined caps.
Wind speed was measured with a wind
profile register system, Model 106 as
manufactured by the C. W. Thornthwaite
Associates. This device has six cup ane-
mometers spaced one above the other at
heights of 20, 40, 80, 120, 240 and 320cm.
The sampling and temperature measure-
ment mast was designed and constructed at
the University of Arkansas. The tubing
leading from the sampling location to the
sample concentration tubes (previously
described) was 0.25 inch (6.4mm) O.D. by
0.21 inch (5.3mm) I.D. These tubes termi-
nated at a Swagelok reducing union to which
a sample concentration tube could be attached.
The outer ends of the lead in tubes were
fitted with Swagelok caps so that they
could be sealed and maintained in a clean
condition during transport and storage.
At the outer end, or air entry loca-
tion, a temperature sensor was attached to
each of the sampling tubes. The sensors
were Chromel-Alumel thermocouples in each
case which terminated at a multiposition
selector switch. The thermocouple junctions
were centered inside 2-1/2 inch long pieces
of white polyvinly chloride pipe, 3/4"
nominal size, which provided radiation
shields for the junctions. From the se-
lector switch, a single pair of Chromel-
Alumel extension wires was connected to an
Omega Engineering Co., Model 410A digital
readout meter.
61
-------
The complete sample collection mast
could be moved up or down without changing
the relative position of the sampling and
temperature measurement points. The
sampling elevations were such that if the
bottom inlet were 5cm above the water sur-
face, the sampling elevations would be
approximately 5, 11, 25, 45, 90, and 180cm
above the surface. The elevations were
chosen to fit the logarithmic scale used in
calculations.
Downstream of each sample concentration
tube, the air passed through a rotometer
(Matheson Model 7400) and from the roto-
meter to a common vacuum manifold. A vacuum
was used to draw the ambient air through
the sampling tubes in order to control flow
rates. The samples must be taken over a
long enough time period to accommodate
varying wind speeds. Thus, the flow rate
over this time period was regulated to
obtain a predetermined volume of air
passing through the sampling tubes. The
vacuum was provided by a hand-pumped tank
of about 70 liter volume. The flow through
all the samplers could be started and
stopped simultaneously and each could be
individually adjusted by a needle valve on
the rotometer, to maintain a steady, equal
flow rate in each tube.
The equipment described in the foregoing
was mounted on the deck of a boat, which in
turn, was moored to the shore to provide
stability.
Description of the Sampling Procedure
The apparatus, as described was assembled
on a small boat moored to the shore so as
to provide the maximum fetch with the avail-
able wind. Air samples were drawn through
the sample concentration tubes, with the
air entering the end containing the Tenax.
The air flow rate was adjusted so as to
pass one liter of air through each sampler
in thirty minutes, to obtain the sample
over a reasonable period of time.
Simultaneously with the start of sam-
pling, the anemometer counters were started.
Similarly, they were stopped simultaneously
with the stopping of the sampling.
At one or more times during the thirty
minute sampling times, the temperatures
were read at each of the six elevations.
Also during the thirty minute period,
two water samples were collected at a loca-
tion close to the sampling location.
The water sample was immediately acidified
with concentrated hydrochloric acid prior
to capping in order to minimize biological
activity. The water samples were refrig-
erated upon return from the field until
they were shipped to the contractor's labo
ratory for analysis.
At the finish of a sampling, the
sample concentration tubes were tightly
capped and held for later analysis.
Analytical Equipment and Procedures
The analysis of the water samples as
reported in Table 1 were performed by a
contractor (Mead Compuchem; Durham, N.C.)
using a gas chromatograph/mass spectro-
photometer. Quality assurance measures
were adopted for the analytical techniques.
Based on the information supplied by the
facility operators, some knowledge was
available on the identity of the volatile
chemicals present. In the laboratory,
external standards were prepared using the
sample tubes with Tenax and silica gel
spiked with selected volatile chemicals and
optimal desorption operating conditions and
recovery efficiencies were obtained. These
standards were prepared by purging water
solutions and adsorbing in the tubes. This
was done prior to analyzing any of the
field sample tubes.
The method employed to analyze the air
samples was essentially as follows: The
Tenax traps were thermally desorbed at
180°C for 12 minutes in an aluminum block
heater with the helium carrier gas flow (40
ml/min). The effluent from the traps was
collected in a cryogenic, capillary trap
cooled with liquid nitrogen.
The samples were then heated and in-
jected into 8 ft x 1/8 in O.D. stainless
steel column packed with 60/80 mesh Carbo-
pack B/1% SP-1000. Temperature programmed
gas chromatography was performed with a
Perkin Elmer Sigma 3 gas chromatograph (3
min. initial hold at 45°C, 8°C/min to 220°C
and 15 min final hold at 220°C, detector
temp. 250°C, flow rate 40 ml/min of helium.
Detection was by flame ionization.
Peak areas determined by the analysis
were compared to those obtained from exter-
nal standards as described above. The
62
-------
external standards were injected in the
same manner as described above for the
field samples.
The field samples could not be sub-
jected to duplicate analysis since purging
used the entire sample. Duplication of
results on the external standards was con-
sidered satisfactory. The peak areas were
within less than 5% disagreement from the
mean for six replications, and retention
times agreed to within one percent or less.
COMPUTATIONAL PROCEDURE
Data Reduction
As stated previously, the underlying
principle and methodology of the concen-
tration profile method has previously been
described (5-7). However, a brief descrip-
tion of the computational methods used
herein is nevertheless appropriate, espe-
cially it will be helpful to describe the
source and/or significance of the variables
and parameters employed.
The flux of chemical arising from an
area source immediately upwind of the
apparatus is given by the following data
reduction equation developed by Thibodeaux:
N (DA-Air?2/3 SV S
fo 2 Sc
where:
N/\ = flux of chemical A (ng/cm^.s)
diffusivity of chemical A in Air
(cm2/s)
diffusivity of chemical B (water
vapor) in Air (cm^/s)
slope of the velocity profile plot
of velocity (of the wind) vs In
y where y = height off the water
in cm. (this slope has units of
cm/s)
slope of the concentration profile
plot of chemical concentration
vs. In y (this slope will have
2 units of concentration, ng/cm^)
$m = diabatic velocity profile cor-
rection factor (dimensionless)
Sc(t) = turbulent Schmidt no. for water
vapor in air (dimensionless)
von Karman constant (k = 0.4,
dimensionless)
-Air =
Sy =
S=
k =
Of the terms in the data reduction,
all except the diffusivities and the von
Karman constant were determined from the
field measurements (if concentration is a
field measurement).
2
The term(|>m scU) as it can be applied
for water vapor, is approximately related
to the Richardson No., Ri, as follows:
(I'm Sc(t)H= (1
50 Ri)±V2
When the Richardson No. is negative (un-
stable) then -50 and +1/2 apply. When the
Richardson No. is positive (stable) then
+50 and -1/2 apply.
The Richardson No., Ri, is a measure of
the atmospheric stability and was determined
from velocity and temperature profile data
as follows:
Ri = ^
-------
Table 2 - Concentration Profile Data
(Profile No. 4; Example Data)
Sample Concentrations
Sampling Height (y)
(cm)
5.08
11.43
24.8
44.5
90.2
181.6
Methylene Chloride
(ng/cm3)
0.280
0.763
0.371
0.021
0.288
0.203
Toluene
(ng/cm3)
0.584
0.345
0.151
0.024
0.001
0.040
Temperature
°C
27.9
31.4
31.5
30.8
31.2
31.1
Velocity
cm/s at cm
198 at 44cm
220 at 64
247 at 104
277 at 184
292 at 264
309 at 344
Sv [slope of vel vs In y] = 52.82 cm/s (correlation coefficient = 0.999)
Sp [slope of cone, vs In (y)]: (ng/cm3)
For CH2C12, using 4 of 6 points: -0.179 (Corr. Coef. = 0.901)
For C/Hg, using 6 of 6 points: -0.159 (Corr. Coef. = 0.909)
Richardson No., Ri, = +0.0187, ( $ 2 Sc^)"1 = (1 + 50 (0.0187))~1/2 = 0.719
n ^
rH n
Diffusivity rations: Ln2 2-Air = 0.120 = Q 465.
DH20-Air 0.258
Flux: NcH2Cl2 = - (0.465)2/3-52.82 (-0.179)(0.2)2(0.719) = 0.653 ng/cm2-s
DTol-Air
DH20-Air
Ntoi = -(0.333)2/3 (52.82) (-0.159) (0.4)2(0.719) = 0.464 ng/cm2-s
-------
Table 3 - Summary of Profile Data
Profile
1
2
3
4
5
6
CH2q2
Velocity
Ri = + 0
CH9C19
C 2H 2
Velocity
Ri = + 0
CH2C12
C7 H8
Velocity
Ri = + 0
CH2C12
C7 H8
Velocity
Ri = + 0
CH9C19
C72H82
Velocity
Ri = - 0
CH9C19
r <- ii <-
C7 H8
Velocity
Ri = - 0
Slopes*
- 0
+ 0
+37
.0066
- 0
- 0
+20
.0040
- 0
- 0
+42
.0608
- 0
- 0
+52
.0187
- 0
- 0
+51
.049
+ 0
+ 0
+57
.089
Correlation
Coefficient
.566
.101
.14
.093
.008
.94
.214
.006
.49
.179
.159
.82
.097
.047
.17
.092
.024
.79
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.972
.782
.998
.795
.767
.996
.936
.878
.998
.901
.952
.999
.753
.861
.966
.664
.787
.997
No.
5
6
5
4
6
5
4
4
4
5
4
4
of Points
Used
of
of
of
of
of
of
of
of
of
of
of
of
6
6
6
6
6
5
6
6
6
6
4
4
* Concentration slopes, Sp, are in units of ng/crn^ (P^ vs In y
Velocity slopes, Sv, are in units of cm/s (v vs In y)
65
-------
available from many sources, handbooks,
etc. such as Treybal (9). In general, and
specifically in the present case, the
diffusivity of the chemicals (methylene
chloride and toluene) in air were estimated
by the method of Wilke and Lee as given by
Treybal (9).
Table 2 presents the data for one
profile and the calculated values resulting.
Table 3 presents a summary of all the pro-
files and the slopes.
Model Prediction Calculations
Flux of evaporating chemicals can be
readily predicted from mass transfer princi-
ples if the rate constants can be predicted.
Wind tunnel tests, as for example, Cohen
et al . , (2) will normally attempt to pre-
dict a mass transfer coefficient which is
related to wind speed, fetch, topography
and so forth. At present, the mass transfer
predictions suffer from many of the limi-
tations of such predictions as dispersions
and so forth, that is, they apply only for
certain ideal terrain configurations.
Nevertheless, it may be that many waste
disposal lagoons conform in many ways to
the ideals.
Cohen, et al . , (2) is perhaps the
latest study to appear in the open litera-
ture concerning mass transfer coefficients
for wind blowing across lagoons. Cohen's
paper suggests the mass transfer coeffi-
cient for low-solubility, volatile chemi-
cals can be predicted by the following:
7.07 x TO"3 Z10U10
1.25
KL = 11.4
where:
- 5.0
K|_ = overall liquid phase mass transfer
coefficient (cm/hr)
Re* = roughness Reynolds No. (dimension-
less)
The roughness Reynolds No. evaluation
requires the estimation of some parameters
which are difficult to determine, but the
authors suggest an approximation for it as
follows:
Re* =
'a exp(56.6/U,Q
0.25,
where:
Z1n = elevation at 10m in cm (thus,
IU = 1000)
= wind velocity at 10m elevation
(cm/s)
v a = kinematic viscosity of air,
(centistoke)
Finally, the rate of mass transfer (evapo-
ration of the chemical) from the water sur-
face is estimated as:
NA = KL ( PA -
PA
where PA = concentration of the chemical
in water (ng/cm^)
liquid concentration in equili-
brium with the actual gas phase
concentration of the chemical
(ng/cm3)
It is normally assumed t(iatPA = 0
when a wind is blowing across a lagoon.
While the simplified flux equation
above is adequate for low solubility, vola-
tile chemicals (said to be liquid phase
controlled), a more general treatment
requires consideration of the gas phase
coefficient, kg, as well, according to the
two-resistance theory which is presented in
a number of references , including Treybal
(9).
One drawback of the Cohen, et al., (2)
equation is that it does not have a fetch
dependence. The work of Hwang (8) and
Sutton (12) suggest that the flux from an
area source is a function of the fetch.
The gas phase coefficient, kg, has been
found to be a weak function of fetch, such
that kg is proportional to (fetch)'^-1 (13).
Ongoing work in our laboratories also indi-
cates that the liquid phase coefficient is
a weak function of fetch (14). At this
point it appears that the fetch correction
will be of the order of a few percent,
perhaps an increase of less than five (5)
percent.
66
-------
Cohen, et al., (2) determined the mass
transfer coefficient for toluene. In order
to apply their relation for other chemicals
it must be multiplied by a diffusivity
correction factor which is a ratio of
diffusivities in water to the two-thirds
power.
RESULTS AND DISCUSSION
As previously mentioned, the water
analyses were as shown in Table 1. The
concentration of methylene chloride and of
toluene, which were of interest in this
study, were both rather low in Pond 3, and
were so close to the limit of detection
(10 ng/cm^) that the values became suspect.
In Pond 2, the chemicals of interest are
high enough to provide for reliable analysis
but not nearly high enough to invalidate
any assumptions based upon dilute solutions.
In Pond 1, the concentration of methylene
chloride is similar to that in Pond 2,
but the toluene concentration is only about
five times the limit of detection and could
be suspect.
The flux values determined from the
measurements are compared with those calcu-
lated by the Cohen (2) model shown in
Table 4. Of the twelve determinations (six
profiles of two chemicals) there are six
(both chemicals in profiles 3 and 4 and
methylene chloride in profiles 5 and 6)
where the water phase concentration of
chemical is suitably high for this study.
Of these six,four show reasonably
good agreement, but two do not, they being
the unreasonably low flux value for toluene,
resulting from profile 3 and the negative
methylene chloride flux from profile 6.
The other four determinations show
agreement within about a factor of two,
which is an accurate as could be expected
when comparing models to the turbulent
environment.
The results of the other six deter-
minations, that is profiles 1 and 2 for
both chemicals and profiles 5 and 6 for
toluene are somewhat disconcerting, espec-
ially those which show the calculated
(Cohen model) flux to be substantially
lower than that which was seemingly observed
in the field.
In profiles 1 and 6 there are negative
toluene fluxes which presumably means
toluene was in the air and was being absorbed
by the water in the pond. While this might
be a satisfactory explanation for profile
1, it doesn't seem reasonable for profile
6, since only a short while earlier, profile
5 at the same location showed an unreason-
ably high toluene flux away from the pond.
The authors cannot point to any known
causes of the several discrepancies. How-
ever, possible reasons abound. Perhaps one
of the most likely reasons for unreasonable
discrepancies would be a non-uniform
quality of the sample concentration tubes.
These tubes were made up of Tenax porous
polymer and silica gel packing inside 1/8"
tubes. It is quite impossible to examine
a tube and determine that adequate packing
is present, or, that if present that it is
properly desorbed, and if not, it would have
a dramatic effect on the concentration vs.
height relationship from which the flux is
deduced.
There is also a possibility of leaks
between the rotometer inlests and the con-
centration tube outlets. Even though
efforts were made to detect and correct
such leaks in the field, there remains
the possibility which would again have
a pronounced effect on the concentration
vs. height relationship.
Finally, the unpredictability of
the micro-environment close to the pond
surface can certainly introduce seeming
inconsistencies in the results. Although
a uniform background concentration of
chemical would not, presumably, interfere
with the concentration profile method
of measurement, a background concentration
at a waste disposal site would not neces-
sarily be uniform, and background would
cause errors in the calculated (predicted)
values of flux, since these are based
on an assumed zero concentration in the
air.
A quantitative determination of the
confidence level of the several measure-
ments of the concentration profile method
is currently underway, but not yet avail-
able (15).
67
-------
CONCLUSIONS
It has been shown that some of these
measurements confirm that secondary emis-
sions from surface impoundments do occur
and can be predicted for low solubility,
volatile chemicals which evaporate under
liquid phase control. Thibodeaux, et al.,
(5-7) have previously confirmed similar
predictability for cases where the gas
phase controls.
It is also apparent from this work
that the field measurement of volatile
chemical emissions from area sources is
encumbered by a serious problem of non-
reproducibility and that spurious results
may be anticipated. The authors feel that
the CP method, however, will generally be
the method of choice when suitable con-
ditions (such as constant wind speed, a
reasonable fetch-to-height ratio of upwind
disturbances, etc.) exist for its employ-
ment.
Maybe it can be said that this work
has accomplished as much to confirm the CP
method as it has to confirm the volatili-
zation modesl, even though one would have
to address such an issue carefully since
it may be confirming two unknowns by
mutual comparison.
It is also apparent that some improve-
ments in the measurement methodology may
be in order. Indeed, that has been the
history of the method during its develop-
ment. Consistency has improved and opera-
tion has become simpler, faster and easier
through the evolution of the apparatus and
method, from that first used by Thibodeaux,
et al., (5-7) through that used in 1981 by
Thibodeaux, Springer, Lunney, James and
Shen (11), to the present state. The
authors are convinced that the method
represents state-of-the-art emission moni-
toring for waste sites where it is appli-
cable.
The authors likewise feel that addi-
tional field work will be required before
the emission models such as that of Cohen,
et al., (2) can be finally confirmed.
REFERENCES
1. Mackay, D. and P. J. Leionen, Journal
Environ. Sci. & Tech., 9, 1178 (1975T
2. Cohen, Y., W. Cocchio and D. Mackay,
Env. Sci. & Tech., J[2, 5, 553 (1978.
3. Mackay, D., Y. Cohen, "Prediction of
Volatilization Rate of Pollutants in
Aqueous Systems," Symposium Nonbiologi-
cal Transport & Transformation of
Pollutants on Land and Water," NBS,
Gaithersburg, MD (May 1976).
4. Thibodeaux, L. J., "Air Shipping of
Organics from Wastewater" A Compen-
dium," Proceedings 2nd Nat'l Conf. on
Complete Water Use, Chicago, IL
(May 1978).
5. Thibodeaux, L. J., D. G. Parker, H. H.
Heck, "Measurement of Volatile Chemical
Emissions from Wastewater Basins,"
Final Report prepared for USEPA under
Grant No. R-805534 (December 1981).
6. Heck, H. H., D. G. Parker, L. J.
Thibodeaux, and R. L. Dickerson,
"Measurement of Volatile Chemical
Emissions from Aerated Stabilizaiton
Basins," National Conference on
Environmental Engineering, ASCE,
Atlanta, GA (July 1981).
7. Thibodeaux, L. J., D. G. Parker, H. H.
Heck, and R. L. Dickerson, Quantifying
Organic Emission Rates from Surface
Impoundments with Micrometeoroligical
and Concentration Profile Measurements,'
Paper No. 127a, AIChE National Meeting,
New Orleans, LA (November 1981).
8. Hwang, S. T., "Measuring Volatile
Chemical Emission Rate from Large
Waste Disposal Facilities," Paper 61c,
AIChE National Meeting, Cleveland, OH,
(August-September 1982).
9. Treybal, Robert E., "Mass Transfer
Operations," 2nd Edition, McGraw-Hill,
New York (1968).
10. Mead Compuchem Corp., Private Communi-
cation, Reports for Sample Nos. 18590,
18591, 18592 (August 1982).
11. Thibodeaux, L. J., C. Springer, P.
Lunney, S. C. James and T. T. Shen,
"Air Emission Monitoring of Hazardous
Waste Sites," Washington, D.C.,
(November 29 - December 1, 1982).
12. Sutton, 0. G., "Micrometeorology,"
Krieger Publishing Co., Huntington,
New York (1953).
68
-------
13. Harbeck, G. E., Jr., "A Practical
Field Technique for Measuring Reser-
voir Evaporation," Geological Survey
Prof. Paper, 272-E, U.S. Government
Printing Office, Washington, D.C.
(1962).
14. Lunney, P. D., "Master's Thesis,"
University of Arkansas, Fayetteville,
AR (December 1982).
15. Cox, R. D., J. L. Steinmetz and D. L.
Lewis, "Evaluation of VOC Emissions
from Wastewater Systems (Secondary
Emissions)", Draft Final Report, EPA
Contract No. 68-03-3038, Radian Corp.,
Austin, TX (1982).
69
-------
INVESTIGATION OF CLAY SOIL BEHAVIOR AND MIGRATION
OF INDUSTRIAL CHEMICALS AT WILSONVILLE, ILLINOIS
R. A. Griffin, Keros Cartwright, P. B. DuMontelle,
L. R. Follmer, C. J. Stohr, T. M. Johnson, M. M. Killey,
R. E. Hughes, B. L. Herzog, and W. J. Morse
Illinois State Geological Survey
Champaign, Illinois 61820
ABSTRACT
Clay soil behavior and migration of industrial chemicals are being investigated at
a hazardous waste disposal facility in Wilsonville, Illinois (Macoupin County). The
study was initiated after the Illinois Supreme Court affirmed a trial court order
requiring the wastes at this site to be exhumed and removed because of their proximity
to Wilsonville and the potential for harm to the town from continued operation of the
site. The May 1981 order provided a unique opportunity to examine in detail the effects
of the wastes on soils below and adjacent to the site and to measure the migration of
contaminants from the trenches. Work is currently in progress, but results are not yet
sufficient to permit substantial conclusions to be reported. The approach is presented
here because it may be helpful to others evaluating hazardous waste disposal sites.
The project provides detailed geologic descriptions of the site, including macro,
meso, and micro features and the effects of the waste leachate on these features. A
detailed hydrogeologic and geochemical investigation is being carried out, including
soil sampling, monitoring wells, piezometers, and hydraulic conductivity measurements.
The latter include in situ vertical and secondary fracture flow, laboratory measurements
of recompacted samples, and horizontal and vertical tests of undisturbed cores. Mea-
surements of effective porosity will be attempted where feasible.
Detailed investigations of materials present within the landfill trenches include
sampling for chemical and physical analyses. Trench materials are being examined with
respect to the nature of the backfill and voids, and the condition of the drums and
cover. Stereo-pair photography is being used to record trench conditions and correlate
them with measurements and notes.
Laboratory studies and chemical analyses are being carried out to support the field
investigations.
INTRODUCTION
This study was initiated after the
Illinois Supreme Court affirmed a trial
court order requiring the exhumation and
removal of wastes at a hazardous waste
disposal facility near Wilsonville,
Illinois (Macoupin County). The reason
for the order was the proximity of the
site to Wilsonville and potential for harm
to the town from continued operation of
the site. The order provided a unique
opportunity to examine in detail the
effects of the wastes on soils below and
adjacent to the site and to measure the
migration of contaminants from the
trenches. Work is currently in progress,
but results are not yet sufficient to
70
-------
permit substantial conclusions to be
reported. The approach is presented here
because it may be useful to others
evaluating hazardous waste disposal sites.
The study of the site is a coopera-
tive effort between several agencies and
the site owner. The US EPA has supplied a
major part of the funding through a coop-
erative agreement with the Illinois State
Geological Survey whose personnel are per-
forming a majority of the work. The
Illinois EPA has provided a drill rig and
crew for the field studies and is perform-
ing organic analyses on soil and water
samples from the site. The site owner,
SCA Services, Inc., has provided access to
the site, safety training, and a substan-
tial amount of the materials for construc-
tion of the monitoring wells.
BACKGROUND EVENTS
Earthline Corporation, a subsidiary
of SCA Services, Inc., began operating a
130-acre landfill near Wilsonville,
Illinois on November 15, 1976. The opera-
tion was a trench-and-fill procedure that
relied on a clay-till deposit native to
the site for natural attentuation of any
leachate. A compacted clay liner was used
to supplement native till in at least one
of the trenches. The company had applied
for and received a permit from the
Illinois Environmental Protection Agency
(IEPA) to dispose of industrial and haz-
ardous wastes at the site. Several months
after the landfill was in operation, the
citizens of Wilsonville became alarmed at
the disposal of hazardous wastes in the
proximity of their community. They filed
suit to stop the disposal of wastes and to
have them removed from the site. A
lengthy court battle ensued, and Earthline
continued to bury wastes. In March 1982,
the Illinois Supreme Court affirmed the
May 1981 trial court's ruling that the
hazardous wastes buried in the approxi-
mately 26 trenches (each 10 to 20 ft deep,
50 ft wide, and 250 to 350 ft long) at
Wilsonville must be exhumed and removed
from the site. SCA Services, Inc.,
announced in March 1982 that they were
dropping further appeals and would comply
with the court order. The preparations
began in the summer of 1982, and the
actual exhumation and removal process,
begun on September 7, 1982, is expected to
continue over a 4-year period.
Earlier that year, in February 1982,
the IEPA discovered that routine monitor-
ing of wells at the site had shown migra-
tion of organic pollutants as far as 50 ft
from the trenches in a 3-year period.
This discovery was a separate issue from
the court proceedings and exhumation
order. The migration rates were 100 to
1000 times faster than predicted. Two
obvious questions were posed: (1) Why
were these organic compounds migrating
faster than predicted, and (2) what are
the implications to land disposal of
similar wastes at other sites? This
research project has been designed to
provide answers to these and many other
questions regarding the impact of land
disposal of hazardous wastes, particularly
organic liquids.
OBJECTIVES
The project aims to determine why
several organic contaminants detected in
monitoring wells around the trenches are
migrating faster than predicted and what
the implications are to land disposal of
similar wastes at other sites. An outline
of the planned scope of work follows.
Site Characterization
1. A detailed geologic description is
to be developed for the site, in-
cluding macro, meso, and micro geo-
morphology, soil materials, and
hydrogeology.
2. Differences will be noted for
laboratory- and field-measured
values of hydraulic conductivity.
3. The occurrence and magnitude of
secondary hydraulic conductivity or
fracture flow will be determined.
^. The effective porosity will be
measured for both field-scale and
laboratory samples of clay till
from the site.
Organic Chemical Effects
1. Measurements will be made of actual
migration rates of organic chemi-
cals through soil materials at the
site.
2. Studies will be made to determine
the effects of certain concentrated
organic liquids on permeability.
3. Studies will be made of the effects
of organic chemicals on pore struc-
71
-------
ture, volume, and size distribution
in clay from the site.
4. A screening test will be developed
to determine the effects of leach-
ate and organic chemicals on clay
structure and hydraulic conductiv-
ity.
Clay Liner Construction
Sections of Trench No. 24 are lined
with a 5-ft layer of recompacted native
clay. A study in in progress to determine
how the waste leachate affects this
liner's ability to retain liquids.
Liquid Level in Waste Cells
The hydrogeology and water balance of
the cells are being examined to determine
effects of hydraulic head on liner and
side wall permeability.
Acid Drainage
The strong acidity and high inorganic
salt content of leachate from the adjacent
pile of coal cleaning refuse (gob) are
being studied with regard to trench con-
ditions and the hydraulic conductivity of
trench bottoms and walls.
Condition of Trench Covers
The condition of the covers is being
examined as they are removed to determine
the role of surface and internal erosion
or the effects of settlement resulting
from waste subsidence on the water balance
of the trenches.
Subsidence
The site is being surveyed at regular
intervals to determine the extent of sub-
sidence in the trenches caused by the
abandoned coal mine 300 ft below the site.
Condition of Drums and Wastes
The condition of the drums and other
wastes is being documented periodically as
they are removed from the trenches to help
determine the effects of the leachate on
the drums and earth materials, the
relative leachate strength, and drum life
expectancy.
RESEARCH PLAN
MULTIDISCIPLINARY APPROACH
A multidisciplinary/multiagency
approach was adopted early in the planning
of this project. The coordinated efforts
and resources of all the interested par-
ties are expected to produce superior
results for several reasons:
1. Access to the site and legal prob-
lems are reduced through coopera-
tion with the site operators
2. The number of borings and samples
are reduced when the needs of all
participants are incorporated into
the experimental design,
3. Field personnel and travel needs
are reduced by consolidation of
efforts.
4. Duplicate testing and analyses are
eliminated.
5. Materials required for construc-
tion, sampling, and analyses are
reduced.
6. Interferences and delays caused by
competing priorities are avoided.
7. Stronger conclusions may be drawn
because of the contributions of
several disciplines and parties.
The multidisciplinary nature of the
project and the uncertainty over the po-
tential hazards that would be encountered
at the site demanded detailed preparations
for the field work. These involved organ-
izational meetings with representatives of
the three government agencies and the
operators of the site to delineate areas
of responsibility, liability, and
allocation of resources. Special safety
measures prescribed for all onsite staff
were: (1) successful completion of an
8-hr first-aid training course, (2) in-
struction in use of protective breathing
apparatus, and (3) a thorough medical
examination before beginning field work.
In addition, a special first-aid kit has
been prepared, and the route from the site
to the nearest hospital emergency room has
been traveled by the field team.. Protec-
tive clothing routinely worn during field
work includes hard hat, disposable cover-
alls, rubber boots, and protective gloves
appropriate for the particular task.
72
-------
Organic vapors are monitored during dril-
ling and other activities, and when con-
centrations exceed 100 ppm, protective
breathing apparatus is worn. Positive
pressure breathing apparatus is worn at
all times when working in the trenches.
Explosive gas mixtures are monitored with
an explosive gas meter. A fire truck is
onsite at all times. Other routine and
special safety precautions are also em-
ployed as deemed appropriate.
GEOLOGICAL CHARACTERIZATION
An extensive geologic investigation
and description of the site is being
carried out to place this site in the
proper regional geologic framework and to
collect sufficient baseline data for ex-
trapolating the results from our investi-
gation to other sites.
The geologic characterization is be-
ing carried out by four principal means:
(1) examination of all previously gathered
data and information available for the
site and for the regional geology, (2)
investigation of outcrops and exposures at
and in the vicinity of the site, (3) study
of the trench walls themselves and backhoe
pits at selected locations on and around
the site, and (4) study of drill samples
collected on and around the site by split
spoon, Shelby tube, and Giddings soil tube
techniques.
A map of the site study area and the
overall drilling plan appear in Figure
1. The drilling includes 11 nests of pie-
zometers and monitoring wells labeled A
through K and two profiles of monitoring
wells labeled V and W, a total of more
than 70 holes. The deepest hole at each
nest and profile location was continuously
sampled using push tubes near the surface
and split-spoon methods for the deeper
samples. The samples were logged in the
field, preserved in core boxes, and
transported to the laboratory, where
Profile V
Groundwater interceptor dram
(approx. loc.)
I II I I
L-ILJI-U—IL_U-I
~~ Property line
O
Nest F
Figure 1. Site map showing locations of trench areas and
SCA monitoring wells, borings, and observation
wells in relation to project well nests and
profiles.
73
-------
detailed sampling and descriptions are
being carried out. Sixteen locations at
the site were continously sampled for
stratigraphic, geologic, and chemical
characterization. These 16 holes were
commonly 30 to M5 ft deep, but some holes
were deeper, including a deep strati-
graphic control hole at nest I that was
continuously sampled to 92 ft, where
bedrock was encountered. Split-spoon
samples were also collected at inter-
mittent depths from all the other holes
drilled at the site.
Geologic units in the Wilsonville
area have been found to extend regionally
and undergo areal facies changes. Soil
units (pedologic types) in the area also
vary areally from one type to another.
Changes in geologic materials are directly
expressed in the soil types of the area,
but these changes have not necessarily
been reflected in the agronomic soil
classifications.
Special emphasis has been placed on
assessing the normal pedologic and
geologic features (facies change phenom-
ena, soil horizonation, etc.) in and
around the trenches to develop a classifi-
cation for materials that distinguishes
classes of normal and disturbed materi-
als. Assuming that all materials within
the trenched area are disturbed to some
degree, sites away from the trenches have
been sought to establish the normal re-
lationships. For increased confidence in
the stratigraphic interpretations, cores
from the surrounding region are being
collected and evaluated. In addition,
more than seven backhoe pits in un-
disturbed locations adjacent to the
trenched area have been examined to com-
plement the study of the exposures
produced by the trench excavations.
The geologic characterization will
focus particular attention on morpho-
logical changes in the soil materials
because it is suspected that organic
solvents may affect soil structure. Meso-
and micro-morphological features will be
studied at selected distances from the
trench walls and bottoms. These features
will be compared with those of normal
samples to determine whether the wastes
have imparted patterns of change into the
morphologies of exposed soils.
Detailed physical and chemical
laboratory characterizations are also
being conducted on selected samples from
the drilling and trench sampling pro-
grams. These include particle-size
analysis, clay and mineralogical char-
acterization, petrographic (micro-
morphologic) evaluation, hydraulic con-
ductivity, pore size distribution, surface
area, pore volume, moisture/density, and
other geologic and engineering tests as
appropriate. Complete chemical charac-
terization includes analyses of volatile
and semivolatile organics and 30 to 45
inorganic parameters.
HYDROGEOLOGIC/GEOCHEMICAL STUDIES
An extensive hydrogeologic and geo-
chemical investigation of the site is
under way and will include borings of 11
piezometer nests (nests A to K, in Figure
1) with three to six piezometers per
nest. These piezometers are being used
initially for insitu hydraulic conduc-
tivity tesha at the various depths of the
piezometers. Later these piezometers will
be used to establish the long-term piezo-
metric surface and in turn the hydraulic
gradient and flow across the site. Core
samples from these borings will be used
for chemical analysis. But water chem-
istry will not be analyzed because of
interference caused by the addition of tap
water during the hydraulic conductivity
slug tests. In addition, withdrawal of
water for analysis is also likely to dis-
turb the piezometric surface measurements.
A separate set of monitoring wells is
being constructed for water chemistry
samples. These include at least two mon-
itoring wells at each nest and up to 22
wells positioned in a geometric progres-
sion from the trenches along two lines.
These lines have been designated as V and
W in Figure 1. Cores and disturbed
samples from the drilling will be sub-
jected to hydraulic conductivity and
chemical measurements.
Vertical fracture permeability will
be measured by angle drilling at separate
locations. Three nests containing three
to four holes per nest will be drilled on
an angle to intersect possible vertical
fractures. Insitu hydraulic conductivity
74
-------
measurements will be carried out, and the
results will be compared with previous
measurements of vertical and horizontal
hydraulic conductivity. Measurements of
effective porosity will be attempted where
feasible.
The construction of the monitoring
wells, screened piezometers, and open-hole
piezometers is shown in Figure 2. Mon-
itoring wells were constructed by boring a
hole to a selected depth with a rotary
hollow-stem auger drill rig. A 2-in.-ID
well casing with a slotted well screen was
lowered to the bottom of the hole through
the hollow-stem auger. All well screens
were 2 ft long, and screen and casing
materials were stainless steel for mon-
itoring wells. Flush screw joint PVC was
used for screened piezometers.
Following placement of the casing and
screen, the hollow-stem auger was with-
Monitoring well or
screened piezometer
Open-hole piezometer
.: expanding
• cement
metal or
PVC casing
2Vi-\n ID
PVC casing
sand-bentonite t f
'* slurry
expanding ^ '
cement :
«— sand pack
-i — well screen
shelby tube hole — ,
1-
»•
1
Figure 2. Diagram illustrating con-
struction of monitoring
wells, screened piezometers,
and open-hole piezometers
used at the site.
drawn from the hole, and clean medium
silica sand was placed to approximately
1.0 ft above the well screen. A plug of 2
to 5 ft of expanding cement was then
placed above the sand pack. Expanding
cement was used for sealing rather than
bentonite to preclude the possibility of
cracking because of the presence of or-
ganic solvents. A mixture containing 70 %
(by volume) clean silica sand and 30%
granular bentonite was used to backfill
the hole to within approximately U ft of
the surface. If water was standing in the
hole above the lower cement plug at the
time of construction, a 5-gal pail of
bentonite pellets was occasionally used.
To avoid vertical cross contamination,
drill cuttings were not used for back-
fill. The annulus was then plugged to the
surface with expanding cement and mounded
slightly around the casing to promote
drainage away from the well.
The open-hole piezometer was con-
structed in a similar manner. The hollow-
stem auger was used to bore to the desired
depth. A 2.5-in.-ID PVC glue joint casing
was lowered through the hollow-stem auger
to the bottom of the hole. The augers
were then withdrawn from the hole, the
bottom sealed with 2 to 5 ft of expanding
cement as a plug, and the hole backfilled
as described above. A 2-in. Shelby tube
was then lowered through the casing, and a
2-ft long sample was pushed and retrieved
with the drill rig, leaving a 2-in. x 2-ft
open hole at the bottom. The hole and
casing were then immediately filled to the
top with water, and the insitu hydraulic
conductivity slug test was initiated. The
immediate introduction of water applied a
positive pressure to the soil, which kept
the hole from caving in. Changes in water
levels were recorded as a function of
time, and the hydraulic conductivity was
computed. The Shelby tube sample was
sealed and taken to the laboratory for
hydraulic conductivity measurements, which
can later be compared with field tests at
each depth in each nest.
ENGINEERING GEOLOGY INVESTIGATIONS
The engineering geology investiga-
tions include: (1) measurement of surface
response to potential coal mine subsidence
at the site, (2) measurement of settlement
and condition of trench covers, (3) deter-
mination of physical properties and char-
75
-------
acteristics of geologic materials compos-
ing strata and fill at the site, (4)
recording of trench conditions using
close-range photogrammetry, stereograms,
and mapping, and (5) obtaining and using
aerial photography to prepare a model for
the final base map of the site and to
interpret geologic features.
Measurement of surface and near-
surface response to subsidence of the
underground coal mine workings involves
careful construction of monuments and
precise surveying measurement. The con-
struction of the survey monuments and set-
tlement probes is illustrated in Figure 3.
Because of the anticipated small
movements likely to be measured at the
site, surveying monuments must be con-
structed in such a manner that movements
of the monuments are those that reflect
movements of the target strata and are not
complicated by other forces such as
shrink-swell soil, freeze-thaw phenonema,
groundwater fluctuations, vehicular
traffic, and other possible earth move-
ments. Third order or better surveys are
necessary to assure the accuracy needed
for recording subtle movements of the
monuments.
Settlement probe installation
5-ft probe 14-ft probe
Monuments for land and trunk cover survey net
3-ft probes
standard
monument
monument
with tilt plate
. protective .^
H-y "1 caP ^*f—
-*— plastic casing — »»
galvanized
"* pipe (1 inch)
loess
o f
o till °
I
^^. protective -.^
•— « »••»" i-^^ cap ^^>
o
c
c
' ^ removable *•; )
^ \ insulation l—
) ~+ — plastic casing — *• f^\
1 L ^ |"j
fl"* ~ -steel pin — ».-j:j
i 'J -^ cement ^ )>
note
casing not attached
to cement U
j
3
5
0 x
pipe driven 1 to 2 ft , — '
nto compact till (soil) * /
o /
\ o NX
Figure 3. Diagram illustrating the construction of the settlement
probe and monument installations used at the site.
76
-------
Measurement of trench cover
settlement will likely be less accurate
than that proposed for surface response to
coal mine subsidence. The principal
problem is that the covers are approx-
imately as thick as the frost penetration
expected at the site; consequently, found-
ing a survey monument below the frost
penetration depth would have meant boring
into the wastes—an undesirable prospect
for various reasons. Movements of monu-
ments on the trench will therefore be
subject to nonsystematic errors. Further
study is under way to mitigate this
problem.
Trench cover settlement data from
this phase of the study will provide
unique information in this regard. These
data will be of particular value in de-
signing limited-infiltration, multiple-
layered earth trench covers that appear
likely to be sensitive to settlement
stresses.
The condition of the cover is impor-
tant as a potential failure mechanism
because any reduction in the effective
thickness of the soil cover by external or
internal erosion or by cracking because of
waste subsidence would increase the amount
of water percolating into the waste
cells. The amount of leachate released
from the cells may be due to increased
hydraulic head on the bottom and side
walls of the trenches. The condition of
the cover will be studied in place and as
it is removed to determine the extent to
which external examination can identify
such problems.
The number and condition of the drums
and their contents relative to other
wastes will be investigated during the
waste removal process. This information
will help interpret the observed effects
on soils in the context of predicting
whether maximum leachate strength has
occurred. The information will also help
estimate the effective service life of
steel drums in landfill cells in this
environment.
In conjunction with the trench inves-
tigations and sampling, the condition of
drums and backfill in the trenches will be
recorded by still photography, employing
techniques for stereographic recording.
Stereograms have been made of the features
in the surface of one trench cover and are
being applied to studies of features in
the trenches.
To extrapolate better the measure-
ments taken in the trench and to record
the location of samples taken, a 2-ft
aluminum cube with holes drilled on
centers will be incorporated into stereo-
graphic photographic pairs and tripli-
cates. Plumbing and orientation of the
cube will enable researchers to orient
directions in space. Negatives can be
placed into a stereoplotter to make exact
measurements of thickness and area. In
addition, where joints occur, measurements
of slope magnitude and direction can be
made where the use of a clinometer would
be clumsy or impractical.
Aerial photographs will be obtained
over the study site for use as a base map
and for interpretation of geologic fea-
tures. An effort will be made to differ-
entiate bedrock from glacial features.
Lineaments of unknown origin will be ex-
amined by field studies. We are attempt-
ing by this study to incorporate into
waste disposal site selection the current
photointerpretation techniques used for
mineral, oil, and gas exploration, and
structural geologic mapping.
LABORATORY INVESTIGATIONS
Hydraulic Conductivity Studies
Laboratory studies are designed to
provide data to help interpret field
measurements and observations. These
studies involve laboratory measurements of
hydraulic conductivity using selected
aqueous and organic liquids. These
results will be compared with those ob-
tained in the field. In addition, we will
measure the effects of varying concentra-
tions of an organic solvent (such as
methylene chloride) on hydraulic conduc-
tivity. These data will help interpret
the insitu hydraulic conductivity measure-
ments and correlate laboratory measure-
ments with field data for predictive
purposes.
Additional laboratory studies will
include determination of moisture release
curves, measurements of effective poros-
ity, and delineation of unsaturated flow
characteristics of soil materials from the
site. These measurements will be carried
77
-------
out in aqueous media and as a function of
organic solvent concentration.
Screening Test For Leachate Effects On
Clay
A related question that this project
will attempt to answer is how can the po-
tential for clay liner failure from
leachate attack be identified at operating
and closed facilities. Because it is dif-
ficult to predict leachate composition
(particularly for mixed-waste facilities),
and because leachate composition changes
gradually with time, it is possible that
incompatible waste/clay liner combinations
may fail to be identified during the
design of the facility. If reference
samples of the clay are taken during
construction and preserved, then testing
can be conducted during facility operation
or after closure using samples of leachate
taken at those times. If appropriate
tests have been previously identified and
correlated with the behavior of clays in
the field, then the potential for clay
liner failure can be evaluated with
greater confidence than is possible during
the design phase of a facility. In this
project, samples of currently produced
leachate, affected clays from the facil-
ity, and unaffected clays from outside the
facility are available. This situation
presents a unique opportunity to evaluate
the feasibility of developing a screening
test for clay liner failure.
Measurements of the viscosity and
filtration rates of leachate-clay suspen-
sions will be evaluated for use as a
screening test to identify leachate
effects on the hydraulic conductivity of
clays and as aids in interpreting the
field observations and measurements. If
the results of these studies correlate
with the effects observed in the trenches,
these tests could be used for post-closure
tests at other facilities.
We anticipate that organic solvents
will play a major role in the migration of
pollutants through earth materials at the
site. The reason is that large increases
in permeability to organic liquids in
landfill liners could possibly develop by
at least two conceivable mechanisms: (1)
interlayer collapse of clay minerals be-
cause of dehydration, or (2) flocculation
resulting from changes in surface proper-
ties of clay minerals. A test of the
cause of these failures is essential. If
the sudden increase in permeability is due
to flocculation of clay particles, a vis-
cosity test could provide an excellent
means of quantifying the effect. Changes
in viscosity of clay suspensions can be
related directly to degree of flocculation
of the suspension. A viscosity test will
allow a rapid screening of different
chemicals, give a quick test of syner-
gistic interaction between chemicals, and
provide an evaluation of any corrective
actions. Viscosity tests can be run on
all earth materials, and experimental
consistency can be maintained by adding
water or clay to achieve a comparable
starting viscosity for each test.
A measure of the filtration rate of
liquids from dilute suspensions of clay
materials could also be used to study the
failure of clay liners. A filtration rate
experiment is probably a better simulation
of the response of a clay liner material,
but the test cannot be used to distinguish
between permeability changes resulting
from flocculation and those owing to
shrinkage of the clays caused by organic-
induced dehydration and interlayer col-
lapse of clays. Once the mechanism of
failure is clarified, a filtration rate
measurement could represent a rapid method
of screening numerous chemicals and clays.
If part or all of the increased
permeability (syneresis effect) that is
due to organic chemicals results from
shrinkage of clays (because of dehydration
and replacement of associated interlayer
water), this phenomenon should be measur-
able by X-ray diffraction. This experi-
ment would be carried out using smectite
clay and indigenous materials. Appropri-
ate quantities of organic chemicals will
be added to the test samples, and X-ray
diffraction will be used to check replace-
ment of intercrystalline water. If the
rate of replacement is determined to be a
significant factor, this variable will
also be measured with diffraction
techniques.
All of the above field and laboratory
data will serve as input for computer
models that will be used to attempt quan-
titative predictions of water balance,
ground-water movement, and pollutant
migration across the site. Predictions
made with the computer models will be
compared with actual field measurements,
78
-------
and the implications of predicting future
migration of contaminants from the site
will be discussed. Discrepancies between
predicted and actual observations will be
interpreted with respect to the design of
earthen liners and covers for landfills,
and with regard to the implications to the
permit process for new waste disposal
sites and post-closure processes for
existing sites.
CURRENT STATUS
The major efforts of this project to
date have been the installation of wells
and the collection of samples outside of
the disposal pits. Soil borings and in-
stallation of monitoring wells have been
completed. Analyses of water, leachate,
and soil samples are under way. Prelim-
inary examination of samples from the soil
borings and observations made in backhoe
pits on and around the site indicate that
sandy layers within the clay till are not
continuous and thus cannot account
entirely for the observed contaminant
migration. The significance of discon-
tinuities and joints exposed in the till
by backhoe pits remains to be deter-
mined. With the decrease in field work
because of winter weather, laboratory
studies of hydraulic conductivity, con-
taminant adsorption, and screening tests
for leachate attack on clays are being
initiated.
ACKNOWLEDGMENTS
The authors wish to acknowledge
partial support of this project by SCA
Chemical Services, Inc., Wilsonville,
Illinois, the Illinois Environmental Pro-
tection Agency, and the U.S. Environmental
Protection Agency, Cincinnati, OH, under
Cooperative Agreement No. R810HM2-01.
Dr. Michael Roulier was the US EPA project
officer.
The authors also gratefully acknow-
ledge J. Reid, S. Otto, J. Hurley,
D. Tolan, and K. Bosie of the Illinois
Environmental Protection Agency, and
G. Rush, J. DiNapoli, V. Poland, and
R. Cloud of SCA Services, Inc., for their
assistance during this project.
79
-------
GROUNDWATER AND LEACHATE PLUMES:
A MODEL STUDY
Wayne A. Pettyjohn, Douglas C. Kent
Jan Wagner, and Arthur w. Hounslow
Oklahoma State University
Stillwater, Oklahoma 74078
ABSTRACT
A TRD has been prepared to aid in the evaluation of permit applications as they deal with
leachate plume migration and mixing. The document describes the development of a variety
of computer programs, nearly all of which are designed for hand-held calculators and
microcomputers. The programs permit prediction of leachate plume migration and ground-
water recharge. The document also contains discussions of sorption, case histories, and
a field study.
INTRODUCTION
This paper describes a technical re-
source document (TRD) being developed to
assist U.S. Environmental Protection Agency
(EPA) permit writers in evaluating permit
applications for hazardous waste storage,
treatment, and disposal facilities. The
document describes a method for predicting
leachate mixing and movement in ground
water; it has been submitted to EPA and is
expected to be released for public comment
in the near future.
In addition to the TRD, several pro-
jects are under way that will eventually
support additions and revisions. These
supporting projects are conducted through
the National Groundwater Research Center
and are sponsored by the Municipal Environ-
mental Research Laboratory, Cincinnati,
Ohio, and by the R.S. Kerr Environmental
Research Laboratory, Ada, Oklahoma. The
projects include:
1. A vertical cross-section model of a
leachate plume,
2. A groundwater recharge model,
3. Field tests of dispersion coefficient
estimates,
4, Measurement of adsorption phenomena and
incorporation into mass transport
equations,
5. Use of a case history to test the TRD,
and
6. Conversion of the model from VS BASIC
to FORTRAN.
TECHNICAL RESOURCE DOCUMENT
The TRD addresses two major questions:
— How much will concentrations of
leachate contaminants be reduced
after mixing with ground water?
— In what direction will a leachate
plume travel and how will its con-
figuration change along the travel
path?
Permit applicants are assumed to use a
variety of methods for predicting the
effects of a disposal site on ground-water
quality and to need a method for screening
these predictions to identify potential
errors and the need for more detailed exa-
mination. To accompish this task the TRD
will provide a variety of predictive tech-
niques ranging from simple to complex.
The document is a primer on leachate move-
ment and mixing. A variety of sophisti-
cated models are available elsewhere for
use by those with a strong background in
hydrogeology.
This report summarizes the TRD
(Pettyjohn et al., in press), and includes
work completed, in progress, and soon to be
undertaken.
Because our charge from EPA implied a
series of less-complicated techniques, the
literature was searched for an adequate
80
-------
chemical transport equation that could be
modified to suit these specified conditions.
The fundamental equation on which our ana-
lytical techniques are based was described
by Wilson and Miller (1978) as follows:
C = f 'm exp
4nn
V (u.r/B) (1)
where: C = the predicted concentration of
a contaminant
m = the aquifer thickness
f 'm = the mass injection rate of
contaminant
D & D = longitudinal and transverse
dispersion parameters
n = porosity
W(u,r/B) = well function for leaky
aquifers described by
Hantush (1956 and 1964).
Some obvious oversimplifications exist
with this model — the lack of a natural
recharge term for example. But it was
chosen because it is simpler and has fewer
parameters than the other equations exa-
mined. The simpler solution lends itself
better to developing a nomograph and user-
oriented calculator and computer programs.
The most significant consequence of adopt-
ing these simplifying assumptions is that
the user cannot predict changes in contam-
inant concentration with depth because the
model is two-dimensional and averages con-
centration over the entire thickness of the
aquifer. This deficiency is being remedied
by development of the plume cross-section
model discussed later in this paper.
The techniques used to predict leach-
ate plume migration and mixing were divided
into three stages. The first is repre-
sented by a nomograph that can be used to
calculate the centerline concentration in a
plume at some designated time and distance
along a flow path that begins at the point
source. The system is not designed to map
a plume (Figure l). The second relies on a
programmable calculator. The calculator
code allows modeling of dispersion, retar-
dation, and radioactive decay. Repeated
calculations will lead to concentration
mapping. The program was devised for a
Texas Instruments TI-59 hand-held calcu-
lator. The third and most complete tech-
nique is a computer program developed to
calculate plume concentration by means of an
IBM 370 computer. The program is written in
VS BASIC for use with a TSO interactive
facility and can calculate and display con-
centration values for a specified time at
selected points on a grid map. An output
example is shown in Figure 2.
These techniques, with examples, have
been described and discussed in detail by
Pettyjohn et al. (1981, 1982) and Kent et
al. (1982).
VERTICAL CROSS-SECTIONAL MODEL OF A PLUME
The model in the TRD predicts the
average concentration at each point in a
plan view of the contaminant plume. This
concentration is averaged over the entire
depth of the plume. In practice, concen-
trations will vary with depth at each point.
An effort is under way to develop a model
for vertical distribution of contaminant
concentrations to supplement the horizontal
distribution model in the TRD. Instead of
being based on a two-dimensional solution
(Equation 1 ) of the conservation of mass
problem, the vertical cross section model
will be based on the full three-dimensional
solution.
The differential equation that des-
cribes the conservation of mass in a homo-
geneous aquifer with uniform, steady flow in
the x-direction can be written as
_ 3C .. 3C
"d at * ' R •
32C
-2
+ uz ° ^ _ R XC (2)
where' 3z
C = the component mass per unit volume
of fluid phase,
D , D , D = the dispersion coeffic-
ients in x, y, and z coordinate
directions,
V = the seepage velocity (Darcy velo-
city is a flux: seepage velocity
is Darcy velocity divided by
effective porosity.
R, = the retardation coefficient, and
d '
X = the first order decay constant.
The retardation coefficient accounts
for the adsorption/absorption of the compo-
81
-------
nent on the solid matrix assuming a linear
adsorption isotherm. The retardation coef-
ficient used in Equation 2 is defined as
PI
'B
(3)
where:
K, = a distribution constant that des-
cribes the linear relationship
between concentrations on the
solid matrix and in the liquid
phase;
PR & 6 = bulk density and porosity of
the porous medium, respectively.
Equation 1 is a vertically averaged solu-
tion — that is, the equation is integrated
over the saturated thickness, and the re-
sulting two-dimensional solution describes
concentrations in a horizontal plane as
shown in Figure 2. This is the model des-
cribed by Pettyjohn et al. (in press).
A horizontally-averaged solution to
Equation 2 would represent a line source of
infinite length located on the water table
perpendicular to the direction of ground-
water flow. The coordinate system is shown
in Figure 3- An interactive or command
language program has been written for both
the steady-state and transient solutions.
A nonuniform source loading has been incor-
porated using multiple source rates and
superposition in time. The program is
menu-driven, and the user can modify any or
all of the input variables using two-key-
stroke commands.
As presently written, the solution is
for aquifers of infinite thickness or for
regions of confined aquifers that are near
the source where the plume has not pene-
trated the entire saturated thickness. An
option has been added to include solutions
for aquifers of finite depth by using
superposition in space (i.e., image wells).
Computer codes are in FORTRAN and BASIC. A
code for the Tl-59 programmable calculator
is also available.
NATURAL GROUND-WATER RECHARGE MODEL
One of the major unknowns in any
ground-water study is the rate at which
water infiltrates the land surface and per-
colates to the water table. In many
ground-water pollution studies, recharge is
important because the infiltrating fluid
leaches water-soluble compounds from a dis-
posal site to form a leachate plume. Con-
versely, recharge may be a mechanism that
tends to dilute a plume, change its geo-
metry, or influence the vertical distribu-
tion of contaminants. Several ground-water
quality transport models require recharge
as an input parameter.
Recharge can be determined or estima-
ted by a number of methods, including
stream hydrograph separation, flow-net ana-
lysis, water-table fluctuations, infiltra-
tion rings, etc. Some tests (such as in-
filtration rings) are highly site-specific,
whereas others (such as stream hydrograph
separations), are more regional. Both
local and regional techniques have short-
comings.
The method we developed is regional,
but it may provide an estimate of monthly
and annual recharge rates. The method is
based on stream hydrograph separation
techniques and is described in detail by
Pettyjohn and Henning (1979).
Ground-water recharge eventually dis-
charges into streams or is removed by eva-
potranspiration. A long-term balance
exists between recharge and discharge, and
in an unstressed aquifer, the amount of
water continually in storage is nearly con-
stant. The recharge program calculates
effective ground-water recharge rates. The
effective recharge rate is the total quan-
tity of water that originates from downward
infiltration to the water table and upward
leakage (if any) from deeper zones to a
shallow or surficial aquifer and then even-
tually finds its way to a nearby stream.
Effective recharge is synonymous with
ground-water runoff. The effective re-
charge is less than total recharge because
the latter is depleted — largely by eva-
potranspiration, and to a smaller extent by
pumping.
The recharge methods (fixed interval,
sliding interval and local minima) general-
ly provide two calculations that are simi-
lar, whereas the third may be substantially
different. This result reflects the
streamside geology. Most commonly, values
for the two closest calculations are aver-
aged.
The data set needed for the analysis
can be obtained from the U.S. Geological
Survey, which annually publishes stream
flow and chemical quality data for each
82
-------
state. The data are listed by water year,
which begins October 1. Water year 19§0,
for example, started on October 1, 1979,
and ended on September 30, 1980.
Presently the program is written in
FORTRAN for use on an IBM 370. The code
is being written in BASIC for a microcom-
puter. An example of one of the output
options, using a VERSATEC printer/plotter,
appears in Figure 4-
DIMENSIONAL ANALYSIS OF SOLUTE PLUMES IN
GROUNDWATER: FIELD-SCALE EXPERIMENTS
The dispersion coefficients (Dy, Dx,
and Dz in Equations 1 and 2) significantly
affect the predicted plume shape and con-
centration distribution within the plume.
Most of the available values for disper-
sion coefficients are based on laboratory
studies, and considerable controversy
exists over values of the coefficients
under field conditions. Because of the
significant effect of the value of the
dispersion coefficient on model predic-
tions, we have arranged to compare litera-
ture values with values from a carefully
controlled field test in a similar mater-
ial. This test will use dye tracers and a
series of injection and observation wells
to evaluate the flow system. Although the
dispersion coefficients will be applicable
only to the system studied, the magnitude
of the difference between literature and
field values will also be a basis for de-
termining the need for similar work in
other hydrogeologic settings.
The field investigation is being con-
ducted by the Illinois State Water Survey
under the leadership of T.G. Naymic. The
site for the experiments, located in Sand
Ridge State Forest in west-central
Illinois, has been in use since February
1982. Full cooperation was obtained from
the owner (Illinois Department of Conser-
vation), and three well drilling contrac-
tors installed wells to sample the aqui-
fer. Sixteen wells are in place at the
experiemental plot.
In March 1982, the first three wells
at the site were installed 300 feet apart
in a triangular pattern. Water levels in
these wells have been measured every 2
weeks since April 1982 to monitor the
ground-water gradient and flow direction.
These data were used to establish the
proper location of three additional wells
that lie along a flow line.
A preliminary tracer experiment was
conducted in August using only four wells
spaced 20 feet apart. The purpose of this
experiment was to gain information on sol-
ute travel times, but the tracer was not
detected in any of the down-gradient wells.
A flow direction shift caused by irrigation
wells was determined during August, but the
main problem was a lack of control on the
system brought about by an insufficient
number of observation wells.
Subsequently, 10 additional wells
were installed near the source well.
During the drilling of one well using a
hollow-stem auger, split-spoon aquifer
samples were collected. The interval of
the aquifer that will receive the tracer
consists of 11 feet of windblown sand that
is underlain by 9 feet of fluvial sand and
gravel. The hydraulic conductivity in the
lower sand and gravel (1600 gpd/ft^) was
obtained from sieve analyses and was four
times greater than that of the upper sand
(400 gpd/ft2). The results of porosity
experiments showed higher actual (0.29) and
effective (0.28) porosity in the upper unit
than in the lower. Actual and effective
porosities in the lower unit are 0.25 and
0.23, respectively.
Other tests included two adsorption
experiments designed to measure the amount
of adsorption of the tracer (Rhodamine WT)
on the aquifer materials and the drilling
fluid (Revert). No adsorption was detected
on either experiment.
A continuous water-level recorder was
installed in a 6-inch-diameter observation
well approximately 3 feet from the source
well. These two wells and the well from
which the aquifer samples were taken will
serve as the three source wells for another
experiment, which will be conducted in
1983-
After the water-level recorder was in-
stalled, a water injection test was con-
ducted using the current source well as the
injection well. Native ground water (400
liters) was used to create a local ground-
water mound around the source well. The
purpose of this test was to check the hy-
draulic connection between the source well
and the nearby observation wells. The
water-level rise in the wells screened only
in the lower sand and gravel unit was twice
that of the wells screened only in the
upper sand unit. As a result of this
biased response, future tracer injection
83
-------
volumes will be kept to a minimum in order
to keep the injection coordinates the same
for both units.
Currently, water samples are being
analyzed monthly at the site with the hope
of directly correlating water-level changes
(recharge events) to changes in natural
quality. Regional recharge rates, based on
the model described earlier, are also being
examined.
Throughout the project, model develop-
ment has continued in preparation for ana-
lyzing the results of the tracer experi-
ments. Computer codes have been written to
solve the one-dimensional transport equa-
tion for several conditions. For example,
with the measured water levels and tracer
concentrations, the physical and chemical
characteristics of the site are estimated
and used to predict the response of the
system at down-gradient wells. A two-
dimensional numerical model has been de-
signed for the Sand Ridge site.
FATE AND TRANSPORT STUDY
This segment of the investigation re-
views the methods that have been used to
quantify the movement of pollutants in the
subsurface in relation to the adsorption
characteristics of the subsurface solids.
This study has two primary goals: One is
to evaluate methods for determining ad-
sorption isotherms, and the second is to
examine methods for incorporating adsorp-
tion phenomena into the mass-transport
equation.
Primary emphasis in the literature is
on the adsorption of nonpolar organic com-
pounds on soil organic matter. With the
exception of certain agricultural chemi-
cals, little attention has been given to
polar-organic compounds or to the adsorp-
tion on clay or amorphous hydroxides of
iron, manganese, and aluminum.
The two main approaches for indirectly
determining adsorption isotherm coeffi-
cients for hydrophobic, nonpolar compounds
are the inverse correlation between adsorp-
tion and solubility and the relation of
octanol-water partition coefficients to
adsorption. See Kenaga and Goring (1980)
for details. Considerable caution must be
exercised in the use of these formulas be-
cause different assumptions are made for
each equation. Quantative relationships
are generally derived by lease squares fit
of experimental data. Also, not all or-
ganic materials behave in the same way or
even in a predictable manner. For some
organic compounds, the solubility equations
may give completely different adsorption
coefficients from those derived from the
octanol-water coefficient technique.
Present work attempts to establish the
assumptions and applications of the various
methods for indirectly determining the
adsorption coefficient. Concurrently, work
is proceeding on the evolution of mass
tansport equations using adsorption coef-
ficients.
CASE HISTORY STUDY
An examination of published materials
and an evaluation of data found in files of
consulting firms regarding ground-water
contamination cases have revealed some in-
teresting but disturbing points. The exa-
mination of case histories included several
hundred examples. Unfortunately, many
regulatory agencies have assumed that the
literature contains an abundance of cases
that describe leachate plume mixing and
migration in considerable detail. Because
of this assumption, many administrators
believe it is no longer necessary to con-
duct field studies. But examination has
shown that despite the hundreds of publish-
ed examples, probably fewer than a half
dozen contain sufficient information to
make a prediction. In nearly every case,
only enough information has been collected
to show that a problem actually exists, and
mere guesses have been made on the proper-
ties required to make a prediction. The
significance of this finding is clear:
More field studies must be undertaken where
such parameters as dispersion, adsorption,
and the areal distribution (both vertically
and horizontally) of water-bearing strata
are examined in some detail.
MODEL OF A CASE HISTORY
The model applications described in
the TRD were tested using the chronium
study on Long Island. The Babylon landfill
on Long Island also was chosen to extend
the model approach to other studies. This
study provided an opportunity to calibrate
the model using data representing different
depths. Calibration techniques included
adjustments in velocity, mass rate, and
thickness (vertical mixing).
84
-------
A complete set of water quality
measurements from the vicinity of the Baby-
Isn landfill was obtained and entered into
,: central computer data file. The data re-
present three depths from which samples
were collected and subsequently analyzed.
Thus a three-dimensional analysis of the
plume geometry can be represented by the
three depth intervals.
The plume was described at three sep-
arate depth zones (a, b, and c). Each zone
was set at a thickness (m) of 25 feet with
each representing the sampling interval of
the wells in the monitoring well nest at
the site (see Figure 5). Selected chemical
constituents such as chloride were contour-
ed for each depth horizon. These data were
then used to help calibrate the plume
model.
The simulation included nine starting
times, with a different mass flow rate for
each time period. The initial starting
times and mass flow rate were estimated by
calibrating a model with measured chloride
concentrations along the centerline of the
plume. The most successful calibration re-
sulted during a simulation of chloride
movement in zones b and c.
The use of a mixing thickness (m) for
depth zones b and c provided another dimen-
sion in model calibration, which can be
used with the modeling procedures described
in the original manual.
FUTURE OBJECTIVES
In addition to tasks described above,
several others are in progress. When com-
pleted, all of the tasks will be described
in separate chapters of the original leach-
ate prediction manual. These include the
following:
1. Conversion of the transport program
from VS BASIC to FORTRAN language.
2. Modification of the program to include
a plume cross-sectional output, super-
position of several sources for dif-
ferent time periods, and superposition
of a subdivided source.
3. Development of a users manual for the
FORTRAN version.
4. Conversion of the vertical model from
FORTRAN to BASIC.
5. Development of a technique to examine
the effect of natural recharge on a
plume.
6. Completion of a computer code for re-
charge mound geometry to be used on
handheld and microcomputers.
7. Development of stochastic models to
gain some insight into the effects that
errors or uncertainties in site char-
acterization have on predicted leachate
plume concentrations. Deterministic
models such as those discussed pre-
viously provide a specific result or
set of results for the specified input
parameters that are assumed to be
exact. In practice, these input para-
meters provide the physical and chemi-
cal characteristics of the site and the
contaminant; but they are seldom, if
ever, known exactly. Parameters such
as porosity or hydraulic conductivity
may represent average values for a very
heterogeneous system. Dispersion
coefficients incorporate a variety of
mechanisms that can result in the
spread of a contaminant: but values of
these parameters might only be guessed
within an order of magnitude. These
uncertainties are not reflected in the
results of a deterministic model.
Fairly simple statistical methods
are being evaluated for possible use in
conjunction with the deterministic
models included in the manual. The ob-
jective of this effort is to quantify
the range of probable concentrations
that might be expected at a particular
time and position in a leachate plume.
Rather than predicting a single value,
the attempt is to produce an expected
value and an associated confidence in-
terval that reflects the uncertainties
or errors in the input data.
REFERENCES
1. Hantush, M. S. 1956. Analysis of data
from pumping tests in leaky aquifers.
In: Transactions, American Geophysical
Union, Vol. 37, No. 6, 1956, pp. 702-
714.
2. Hantush, M. S. 1964. Hydraulics of
wells, - In: Advances in Hydroscience,
Vol. 1, 19?>4, pp. 282-432.
85
-------
Kenaga, E. F. and C. A. T. Goring.
1980. Relationship between water solu-
bility, soil sorption, octanol-water
partitioning, and concentration of
chemicals in biota. In: Proceedings
of the Third Annual Symposium on
Aquatic Toxicology (STP 707). Edited
by J. G. Eaton, P. R. Parrish, and A.
C. Hendricks. American Society for
Testing and Materials, Philadelphia,
Pennsylvania, pp. 78-115-
Kent, D. C., W. A. Pettyjohn, T. A.
Prickett, and F. E. Witz. 1982.
Methods for the prediction of leachate
plume migration. In: Proc. Symp. on
Aquifer Restoration and Ground Water
Rehabilitation, Nat. Water Well Assn.,
pp. 246-261.
Pettyjohn, W. A. and R. J. Henning.
1979- Preliminary estimate of ground-
water recharge rates, related stream-
flow and water quality in Ohio. Ohio
State Univ. Water Resources Center,
Project Completion Rept. No. 552,
323 P.
Pettyjohn, W. A., T. A. Prickett, D. C.
Kent, and H. E. LeGrand. 1981. Pre-
diction of leachate plume migration.
In; Proc. 7th Ann. Research Symp. on
Land Disposal of Hazardous Wastes,
EPA-600/9-81-002B, U.S. Env. Prot.
Agency, pp. 71-84-
Pettyjohn, W. A., T. A. Prickett, D. C.
Kent, H. E. LeGrand, and F. E. Witz.
1982. Predicting mixing of leachate
plumes in ground water. In; Proc. 8th
Ann. Research Symp. on Land Disposal of
Hazardous Wastes, EPA-600/9-82-002,
U.S. Env. Prot. Agency, pp. 122-136.
Pettyjohn, W. A., D. C. Kent, T. A.
Prickett, H. E. LeGrand, and F. E.
Witz. In press, Methods for predicting
leachate plume migration and mixing.
A Technical Resource Document. U.S.
Env. Prot. Agency, 202 p.
Wilson, J. T. and P. J. Miller. 1978.
Two-dimensional plume in uniform
ground-water flow. Jour, of Hydraulics
Div., Am. Soc. of Civil Eng., Paper No.
13665, HY4, pp. 503-514-
86
-------
PROCEDURES AND TECHNIQUES
FOR
CONTROLLING THE MIGRATION
OF LEACHATE PLUMES
Charles Kufs
Kathi Wagner
Paul Rogoshewski
Marjorie Kaplan
Edward Repa
JRB Associates
McLean, Virginia
ABSTRACT
The problem of leachate plume management has been aggravated, to some extent,by a lack of
understanding of plume dynamics and the various remedial options available. This paper
summarizes information in the areas of leachate plume dynamics and plume management
alternatives. The paper describes factors that affect leachate plume movement and key
considerations in delineating the current and future extent of the leachate plume. Four
technologies for controlling the migration of the plume are also discussed: (1)
groundwater pumping involves extracting water from or injecting water into wells to
capture a plume or alter the direction of groundwater movement; (2) subsurface drains
consist of permeable barriers designed to intercept groundwater systems; (3) barriers
consist of a vertical wall of low permeability materials constructed underground to
divert groundwater flow or minimize leachate generation and plume movement; and (4)
in-situ treatment methods which biologically or chemically remove or attenuate
contaminants in the subsurface.
INTRODUCTION
Cleaning up the thousands of hazard-
ous waste release sites identified across
the U.S. is a serious environmental
challenge to the Nation. Of particular
importance is protecting and cleaning up
contaminated groundwaters that now serve
or could serve as drinking water supplies.
The problem of leachate plume management
has been intensified to some extent by a
lack of understanding of plume dynamics
and the various site remediation options
available. Generally, methods for con-
trolling the migration of a leachate plume
fall into one of four categories—ground-
water pumping, subsurface drains, low
permeability barriers, or in-situ
treatment. The purpose of this project,
then, was to prepare a reference manual on
factors that affect the movement of
hazardous leachate plumes, and con-
siderations necessary to develop sound
plans for managing leachate plumes. The
purpose of this paper is to present an
overview of the manual's contents.
PLUME DYNAMICS
The simplest type of leachate plume,
in terms of plume movement, is one which
mixes with groundwater and moves at the
same rates and in the same directions as
87
-------
groundwater. In this case, plume velocity
(ignoring attenuation and other geochemi-
cal factors) can be expressed by Darcy's
Law:
V = K i/n=Ti/bn
where
V is the velocity of the plume
K is the effective permeability or
hydraulic conductivity of the aqui-
fer
i is the hydraulic gradient or the
slope of the water table or the
potentiometric (pressure) surface
n is the porosity of the aquifer
T is the transmissivity of the aqui-
fer or the aquifer's ability to
transmit water to wells
b is the saturated thickness of the
aquifer through which groundwater
flows.
Thus, the velocity of a plume moving with
groundwater will be directly related to
the transmissivity and the gradient, and
inversely related to the saturated thick-
ness and the aquifer's porosity.
The direction of movement of simple
plumes is controlled by the gradient of
the potentiometric surface. In unconfined
aquifers, this gradient commonly follows
the surface topography. The gradient,
however, can be affected by a variety of
hydrologic and geologic factors. Some of
those factors include:
• Aquifer Geometry—The shape of the
water bearing zone through which
contaminants flow. For instance,
a "shoestring" aquifer formed by
buried stream deposits can channel
contaminants in a sinuous path
from a site independent of surface
topography.
• Aquifer Uniformity—The consist-
ency of the hydrologic properties
of the aquifer. For example,
lenses of higher permeable materi-
als can cause contaminants to
break off from the main body of
the plume and migrate
independently.
• Geologic Discontinuities—Faults,
fractures, bedding planes, joints,
and other planes of greatly
different permeability. For ex-
ample, sealed (low permeability)
fault planes can restrict plume
movement in a given direction
while open (high permeability)
fault planes can channel the plume
in a new direction.
• Hydrologic Barriers—Areas of re-
charge or discharge which control
potentiometric gradients.
In addition to hydrogeologic factors,
man-made alterations to surface or sub-
surface conditions can also affect the
direction and rate of leachate plume
migration. For example, plume movement
can be redirected by the presence of
pumping wells, sewer lines, mines, canals
and impoundments, tunnels, construction
activities, and buried foundations. The
cumulative effect of these and other
factors can make predicting plume migra-
tion patterns an extremely complex task
even for water soluble plumes that move
with groundwater flow. Consequently, it is
imperative to delineate plumes on a site
specific basis.
PLUME DELINEATION
Before the options for leachate plume
management at a given site can be de-
veloped, it is essential to know the
spatial dimensions of the plume. There are
a variety of direct and indirect methods
for obtaining data on a plume's dimensions
including:
• Aerial image interpretation
• Hydrogeologic investigations
• Geophysical investigations
• Groundwater sampling
• Hydrologic (computer) modeling.
88
-------
Aerial imagery refers to pictorial
representations produced by electromag-
netic radiation that is emitted or reflec-
ted from the earth and recorded by
aircraft mounted sensors. The simplest,
most common kind of imagery is the
photograph, which uses only the visible
part of the electromagnetic spectrum.
The most common type of image outside
of the visible spectrum is the infrared
image. Infrared images indicate areas
that are hotter or cooler than the general
surroundings such as areas of dead or
stressed vegetation. If historical aerial
imagery is available, comparisons between
new and old imagery can indicate changes
in an area that may be due to leachate
migration. For example, the gradual
development of a barren area in a forest
near a disposal site may indicate the
presence of hazardous leachates.
A hydrogeo logic site investigation
typically consists of installing a number
of wells and conducting tests which
characterize the aquifer's ability to
store water ( storat ivity ) and transmit
water ( transmissivity ) . Three types of
tests can be used to obtain this in—
format ion:
is monitored. Pressure tests are
generally used only in bedrock
aquifers .
In addition to conducting aquifer
tests, a hydrogeologic investigation
should also provide detailed information
on site geology and pedology from well
logs as well as document the presence of
leachate seeps, geologic discontinuities,
and discharge and recharge barriers.
The well logs obtained in the hydro-
geologic investigation will provide ex-
cellent data on the subsurface geology at
discrete points. However, this infor-
mation may not be sufficient (unless a
large number of borings are taken) if the
geology of the site is highly variable.
In this case, a geophysical survey of the
site can provide a cost effective solu-
tion. The geophysical methods most
commonly used in site investigations are
resistivity, seismic, metal detection, and
radar. With proper application of geo-
physical survey methods, the lateral and
vertical extent of the plume can be
determined, changes in contamination with
depth can be approximated, depth of
bedrock can be established, and metal
drums or debris can be located.
• Pumping Tests — During a pumping
test, one well is pumped and the
decline in the water level in the
aquifer (i.e., the drawdown) is
monitored in other wells over
time, generally from a few hours
to over a week depending on the
aquifer. Pumping tests are appli-
cable to both confined and uncon—
fined aquifers and are the most
common type of aquifer test used.
• Slug Test — In a slug test, water
is typically injected into a well
and the decline in the water level
is monitored over time. Slug
tests are used most commonly in
shallow, unconfined aquifers.
• Pressure Tests — In a pressure test
(also called a packer test) a zone
of the aquifer is isolated and
water is injected into the forma-
tion while the injection pressure
The most direct means of delineating
a leachate plume is by groundwater
sampling. Typically, the wells installed
during the hydrogeologic investigation are
also used for obtaining samples of the
contaminated groundwater. However, ob-
taining valid groundwater chemistry data
is extremely difficult because of the
large number of variables involved. Some
of the factors that can affect the
validity of groundwater sampling results
are outlined in Table 1. The results of
the chemical analyses of these samples can
be plotted on a site map and contoured in
much the same way as water table and land
surface elevations. These concentration
contour maps for individual chemical
contaminants can then be used to ascertain
plume movement directions, which can be
critical in cases where the plume does not
follow groundwater flow paths.
Groundwater models have been used to
evaluate the extent and expected movement
89
-------
TABLE 1: FACTORS AFFECTING THE VALIDITY
OF GROUNDWATER SAMPLING RESULTS
Influencing Factor
Inherent Problems
Well Installation
Well Purging
Sample Acquisition
Sample Preservation and Holding
Sample Analysis
Sanples can be contaminated by well
material or equipment used to install
the well.
Analytical results can be skewed if
the sample obtained is stagnant water
(insufficient purging) or is diluted
(excessive purging), or if the time
between purging and sampling is too
short or too long.
The methods used to extract the water
from the well (e.g., pumping, bailing)
and place the water in the sample con-
tainer can change the chemical properties
of the sample (e.g., pK, volatile
organic content).
Samples that are not preserved properly
(e.g., through refrigeration or pH
adjustment) or that are held too long
prior to analysis can undergo chemical
changes.
Rigorous laboratory controls are re-
quired to preserve sample integrity,
especially when working in the parts
per billion range. Even so, common
laboratory contaminants (e.g., phthalates)
may adversely affect sample analysis.
90
-------
of plumes. Numerous models are available
for the purpose (e.g., AT123D, SWIFF,
Random Walk Solute Transport) which can be
categorized into two groups—analytical
and numerical. Analytical models estimate
waste constituent concentrations and dis-
tributions using simplified, explicit
expressions generated from partial dif-
ferential equations. Numerical models
reduce the partial differential equations
to a set of algebraic equations that are
solved through linear algebra using matrix
techniques. Both types of models are
capable of evaluating plume movement in
three dimensions, but analytical models
typically simplify complex physical and
chemical characteristics (i.e., isotropy,
homogeneity).
A comparison of the advantages,
disadvantages, and examples of data ob-
tained using the previously discussed
plume delineation methods is given in
Table 2. The methods used for delineating
a plume at a site will probably be based
upon available resources (e.g., funds,
personnel, equipment), and the amount of
data already available.
GROUNDWATER PUMPING
Previous utilization of pumping tech-
nologies to manage plumes has shown that
these methods are most effective at sites
where underlying aquifers have high in-
trinsic permeabilities (e.g., coarse
grained sands) and where the contaminants
move readily with groundwater (e.g.,
benzene). Pumping methods have also been
utilized with some effectiveness at sites
where pollutant movement is occurring
along fractured or jointed bedrock.
However, the fracture patterns must be
traced in detail to ensure proper well
placement.
The shape and size of a well's cone
of depression is dependent upon the
pumping rate and cycle, slope of the
original water table, location of hydra-
ulic barriers, aquifer characteristics,
and location of recharge zones. Two
aquifer characteristics that are important
in determining the cone's configuration
are the coefficients of transmissivity (T)
and storage (s). Both of these parameters
are inversely related to drawdown and the
deepening of the cone.
The various equations used in analy-
zing the flow to wells under different
aquifer conditions are given in Table 3.
Equilibrium formulas are simpler than
non-equilibrium formulas because once equi-
librium conditions are reached the shape
of the cone of depression does not change
with time. Using these equations the
effects of barrier conditions, cumulative
drawdown, and discharge-recharge well
interactions on the potentiometric surface
can be obtained.
The equations presented in Table 3
allow the radius of influence of a well to
be determined. The radius of influence is
used in determining well spacing, pumping
rates, pumping cycles, and screen lengths.
Calculations of the radius of influence of
a well can be performed exactly or
approximated depending on the available
data for the aquifer. Table 4 gives the
various methods that can be used to
calculate the radius of influence in
confined and unconfined aquifers under
equilibrium or non-equilibrium pumping
conditions.
Two useful relationships for under-
standing the effects of drawdown on
pumping volume and radius of influence,
and for making checks on calculations are:
Q/H-h = 2 K/(ln Ro/r) = constant (for
confined aquifers)
2 2
Q/H -h = K/(ln Ro/r) = constant
(for unconfined aquifers)
Therefore, for a given aquifer being
pumped at equilibrium, the radius of
influence will not change with pumping
rate. Under non-equilibrium conditions,
the size of the radius of influence is
determined mainly by the pumping cycle.
91
-------
TABLE 2. SUMMARY OF PLUME DELINEATION METHODS
Methods
Advantages
Disadvantages
Examples of Data Obtained
Aerial image
interpretation
Hydrogeologic
Invest igations
• Can be a fairly inexpensive
and rapid means of estimating
plume limits
• Can also be used to verify
site geology, hydrology,
and pedology
• Many different types of
imagery exist for special-
ized applications
• Nearly every part of the
Country is covered by
some type of imagery
• The primary method of
obtaining site data needed
for remedial action design
and implementation
Evidence of leachaLe is
indirect and may be mis-
leading
May provide ambiguous data
• Available imagery may not
show site at an appropri-
ate scale
Can be very expensive
• Can be fairly hazardous to
invest igators
Areas of stressed
vegetat ions
• Surface water flow pat-
terns and other charac-
teristics
• Locations of springs,
faults, fracture traces,
and other natural features
• Aquifer characteristics
(e.g., gradient, trans-
missivity, storage
coefficient)
• First hand evidence of
wastes and leachate
effect s
Does not provide much in-
formation on plume
dimensions
Geophysica 1
Investigat ions
Fairly inexpensive and safe
method of obtaining site
data
Data is sometimes ambiguous
Areas of high/low
groundwater conductivity/
res 1st ivity
-------
TABLE 2. (Continued)
Methods
Advantages
Disadvantages
Examples of Data Obtained
Geophysical
Investigations
(Cont inued)
Groundwater
Sampling
Can provide data on both site
geology and plume dimensions
• Can provide continuous as
well as discrete data
• The primary method of
obtaining data on plume
dimensions
• Techniques can be depth
limited at some sites
Can be fairly expensive
Locations of buried
drums (metal detection,
magnetometer)
Site geology (i.e.,
lithology, stratigraphy,
structure)
Concentrations of in-
dividual chemicals or
chemical groups at
specific locations and
depths at a site
• Can be hazardous to
investigators
Hydrologic
Mode 1 ing
A fairly rapid method of
assessing site data in
great detail
Can be fairly inexpensive
depending upon the model
chosen
A variety of computer
models exist for data
analysis
• Discrete data points are
sometimes difficult to
interpret
• Requires a fairly extensive
amount of high quality data
Data output must be evalu-
ated in terms of input
data accuracy and precision
and assumptions
Requires field and analyti-
cal verification
• Answers are only estimates
of actual conditions.
• Present and future plume
movement patterns
Effects of site remedi-
ation alternatives on
plume dynamics
-------
TABLE 3. BASIC WELL ANALYSIS EQUATIONS
Aquifer Type
Pumping Type
Equilibrium
Non-Equilibrium
Unconfined
Confined
H2-hw2 = (Q/TTk) In R /
rw
H-hw = (Q/2TT) In R0/rw
H-h = (Q/4TTT) W (yA, yB,r)
UA - r2 S/4Tt (early drawdown)
yg = r2Sy/4Tt (late drawdown)
T. = rK2/H2KH
H-h = (Q/47TT) W (y)
y=r2S/4Tt
H =saturated thickness of aquifer
hw=height of water in well
Q =pumping rate
K =coefficient of permeability (hydraulic conductivity)
Ro=radius of influence
rw=radius of well
T =coefficient of transmissivity
W(y) = well function
r =distance from well to drawdown measurement
S = coefficient of storage
t =time elapsed since pumping started
Sy=specific yield
K2 ,K]-j=vertical and horizontal permeability
94
-------
TABLE 4. METHODS FOR CALCULATING RADIUS OF INFLUENCE (R )
o
Pumping Condition Method
Equilibrium
. „ K(H -hw2) , „ T(H-hw) + In r
-Exact ln V 458 Q + ln --
-Approximate Ro= 3(H-hw) (0
Non-Equilibrium
-Exact Drawdown vs. log distance plots
Tt \
-Approximate Ro = rw + ( ' )
R0=radius of influence (ft)
K =permeability (gpd/ft^)
H =total head (ft)
hw=head in well (ft)
Q =pumping rate (gpm)
\fo=well radius (ft)
T =transmissivity (gpd/ft)
t =time (min)
S =storage coefficient (dimensionless)
95
-------
Applications
Well Design and Installation
Well systems can be designed to
perform several functions with or without
the assistance of other technologies
(e.g., barrier walls). The main applica-
tions in plume management are groundwater
level adjustment, plume containment, and
plume removal.
Well systems to adjust groundwater
levels can be designed using extraction
wells to lower water levels or using
injection wells to create groundwater
mounds or barriers. By adjusting ground-
water levels, plume development can be
stopped at the source or the speed and
direction of the plume can be altered. In
either case contaminated water is not
extracted from the groundwater system as
is the case with containment and removal
techniques.
Well systems used to contain a plume
may incorporate extraction wells or ex-
traction and injection wells in combina-
tion. Containment differs from removal in
that the source of contamination generally
is not stopped, so that contamination is
an ongoing process. Because containment
requires removing contaminated ground-
water, a treatment or disposal method must
be developed to handle the system's
discharge.
Plume removal implies a complete
purging of the groundwater system to
remove contaminants. Removal techniques
are suitable when contaminant sources have
been stopped (e.g., by waste removal or
site capping) or contained (e.g., by
barrier walls) and aquifer restoration is
desired. Extraction or extraction and
injection well systems can be used in
plume removal. Numerous arrays and pat-
terns are available for injection/extrac-
tion wells and the choice is usually made
based on suitability for a specific job.
Extraction/injection techniques can also
be used with flushing compounds to accel-
erate contaminant removal. As with con-
tainment designs, treatment of pumped
water is necessary.
Selecting and designing the appro-
priate well system requires that adequate
information be available on the site's
hydrogeologic conditions and the plume's
characteristics. Four basic types of wells
are used in plume management: well points;
suction wells; ejection wells; and deep
wells. Using the site data, selection of
an appropriate well type can be made.
Table 5 lists selection criteria for these
well types.
The installation of wells varies with
geologic materials, well type, and diame-
ter, but most installations consist of
opening the well hole, installing casing,
completing the well, and developing the
well. The first of these steps consists
of dislodging and removing the earth
materials located in the path of the well.
Screens and casings may be installed
simultaneously with the opening of the
hole or after the hole has been completed.
The sequence of events depends upon the
technique used to open the hole and the
geologic characteristics in the immediate
vicinity of the hole. Installation of
screens, filters, pumps, and grout com-
pletes the construction of the well. Well
development is the last step and consists
of removing material that has built up on
the well screen and walls of the borehole
during the previous steps. Removal of
these particles maximizes well yield and
prevents pump damage. Table 6 gives
various installation methods used for
wells, and methodrestrictions.
SUBSURFACE DRAINS
Subsurface drains include any type of
buried conduit used to collect liquid
discharges (i.e., contaminated ground-
water) by gravity flow. The major
components of a subsurface drainage system
include: drain pipe(s), envelope and/or
filter, backfill, manholes or wetwells,
and pumping station(s).
96
-------
TABLE 5. CRITERIA FOR WELL SELECTION (adapted from Reference 12)
Parameters
Wellpoints
Suction Wells
Ejector Wells
Deep Wells
Hydrology
o Low permeability
(e.g., silty or
clayey sands)
o High permeability
(e.g., clean sands
and gravel)
o Heterogeneous materials
(e.g., stratified soils)
o Proximate recharge
o Remote recharge
Depth of Well
Normal spacing
Normal range of
capacity (per unit)
Efficiency
Good
Good
Good
Good
Good
Shallow < 20 ft
5-10 ft
0.1-25.0 gpm
Good
Poor
Good
Poor
Poor
Good
Shallow < 20 ft
20-40 ft
50-400 gpm
Good
Good
Poor
Good
Good to Fair
Good
Deep > 20 ft
10-20 ft
0.1-40.0 gpm
Poor
Fair to Poor
Good
Fair to Poor
Poor
Good
Deep > 20 ft
> 50 ft
25-3000 gpm
Fair
-------
TABLE 6. METHODS OF WELL INSTALLATION
Method
Basic Variations
Hand Augers
Driving
points
Boring Rotary Auger
bucket
Spiral Auger
Jetting Self Jetting
Wellpoint/
Riser Unit
Separate
temporary
jetting pipe
Applications
Geologic Material Maximum Well
Dia. (in)
Soft, without excess 6
sand and water, no
boulders
Soft soil free of 3
boulders
Soft soils without 48
excess boulders
Soft soil without 6
excess sand and
water, no boulders
Soft soils free 8
of boulders
Soft soils free 8
of boulders
Soft soils free 8
of boulders
Casing/Screen
Maximum Well
Depth (ft)
20 Driven after
opened
30 Driven as hole
proceeds
90 Driven after
hole opened
90 Driven after
hole opened
50 Driven as hole
opened
50 Driven as hole
opened
100 Placed after
hole opened
Comments
Wider and
deeper can
be machine
driven
Typically
l.imited to
24-in. diameter
wells or less
Cannot be used
below water
table, must be
used in
combination
with other
techniques
Jetted wells
require less
development
Method uses
large quantities
of water
-------
TABLE 6. (Continued)
Method
Basic Variations
Jetting Separate
(Continued) Permanent
jetting pipe
Jet
Percussion
Rotary hole
puncher
Jg Rotary Conventional
Drilling hydraulic
Reverse flow
Air Rotary
Air rotary
with pneumatic
Applications
Geologic Material Maximum Well
Dia. (in)
Soft soils free 8
of boulders
Soft soils free 24
of boulders
Soft soil, sand 24
stone, schist
Any type -
Any type, boulders 60
may be a problem
Any type 12
Any type 8
Casing/Screen
• — - • • ' ' - - - — — — Treatment
Maximum Well
Depth (ft)
100 Driven as hole
opened
200 Placed after
hole opened
200 Placed after
hole opened
- Placed after
hole opened
- Placed after
hole opened
- Placed after
hole opened
- Placed after
hole opened
Comments
Typically used
for deep wells,
quickest drill-
ing method
Require large
quantities of
water
Very fast
method
hammer
-------
Theory and Applications
Subsurface drains function similarly
to an infinite line of extraction wells.
That is, they effect a continuous zone of
depression which runs the length of the
.drainage trench.
Functionally, there are two basic
types of drains—relief drains and inter-
ceptor drains. Relief drains are instal-
led in areas where the hydraulic gradient
is relatively flat. They are generally
used to lower the water table beneath a
site or to prevent contamination from
reaching a deeper, underlying aquifer.
Relief drains are installed in parallel on
either side of the site such that their
areas of influence overlap and contami-
nated groundwater does not flow between
the drain lines. They can also be
installed completely around the perimeter
of the site.
Interceptor drains, on the other
hand, are used to collect groundwater from
an upgradient source in order to prevent
leachate from reaching wells and/or
surface water located hydraulically down-
gradient from the site. They are instal-
led perpendicular to groundwater flow. A
single interceptor located at the toe of
the landfill, or two or more parallel
interceptors may be needed, depending upon
the circumstances.
Whether a drain functions like an
interceptor or a relief drain is de-
termined by the hydraulic gradient. The
design of these two drain types differs
but construction and installation are the
same.
Although subsurface drains perform
many of the same functions as pumping
systems, drains may be more cost effective
under some circumstances. For example,
they may be particularly well suited to
relatively low permeability sites where
the cost of pumping may be prohibitively
high because of the need to space wells
very close together.
There are, however, a number of
limitations to the use of subsurface
drains as a remedial technique. They are
not well suited to areas of high perme-
ability and high flow rate. Also,
contamination at great depth may cause
construction costs to be prohibitive,
particularly if a substantial amount of
hard rock must be excavated. Subsurface
drains are also not suitable when the
plume is viscous or reactive, since this
type of leachate may clog the drain
system.
When a drainage system is being
designed the following elements must be
specified:
• Drain type
• Drain depth
• Spacing, if parallel drains are
required
• Drain diameter and gradient (hy-
draulic design).
Drainage systems are classified by
function (i.e., as either -relief or
interceptor drains), by the type of drain
pipe, and by their configuration. Per-
forated pipe, available in chemically re-
sistant concrete, vitrified clay, and
various plastics, is the most widely used
drain pipe for remedial action work,
particularly if the drains are to be
placed deep in the substratum. Other
types of drainage conduits include jointed
clay and concrete tiles and flexible
corrugated plastic pipe.
The configuration of the drainage
system depends on the site configuration
and size, and the flow rate. The system
may be singular, consisting only of
lateral pipes discharging to a collection
sump, or composite where laterals dis-
charge to larger collector pipes which may
in turn discharge into a main before
reaching a collection sump.
In a subsurface drainage system
containing parallel relief drains, depth
and spacing are interdependent design
variables. In theory, the deeper the
drains are the greater the spacing that
can be used to obtain the same zone of
100
-------
depression. This relationship between
depth and spacing is critical to the
design of effective parallel drainage
systems. In designing a system for
hazardous waste sites, it must be de-
termined whether the drawdown curves of
parallel drains located on either side of
the site will intersect and thereby
capture the entire plume. It may be
necessary to manipulate depth and spacing
in order to capture the entire plume.
However, in designing parallel drains for
hazardous waste sites, a minimum spacing
is often imposed by the boundaries of the
waste since excavation through the waste
material can be extremely hazardous. A
maxiumum depth may be imposed by the
prohibitive cost of trench excavation.
Numerous equations, models, and
analogues have been developed for de-
termining appropriate spacing of parallel
relief drains under various idealized
conditions. Donnan(5) developed a formula
for determining drain spacing for drainage
systems which extend to the depth of an
impermeable barrier (Table 7-A). It is
evident from this formula that spacing is
a function not only of drain depth, but
also of the hydraulic conductivity and
flow rate. In many instances, relief
drains will not be installed to the full
depth of an impermeable barrier due to
high costs. Under these circumstances,
design equations must account for radial
flow to the drain from the underlying
permeable layers as well (Table 7-B).
Hooghoudt's formula (9) was developed to
account for this radial flow. A number of
nomographs are available to aid in solving
Hooghoudt's equation.
Design of interceptor drains is
different from that of relief drains. In
order to decide where to position the
drain to intercept the desired flow, the
relationship between depth and flow, and
the upgradient and downgradient influence
of the drain must be known.
for a distance which is inversely pro-
portional to the water table gradient.
The distance to which the water table is
lowered downgradient of the interceptor
drain is directly proportional to the
depth of the drain. Theoretically, a true
interceptor drain lowers the water table
downgradient to a depth equal to the depth
of the drain. The distance downgradient
to which it is effective in lowering the
water table is infinite provided there is
no recharge. This however is never the
case since infiltration from precipitation
recharges the groundwater. The quantity
of flow is also proportional to the depth
of the drain. If an interceptor drain is
placed at the midpoint between the water
table and an impervious layer, a little
less than 50 percent of the flow will be
intercepted. The upgradient and down-
gradient influence of an interceptor drain
and the quantity of flow intercepted can
be determined theoretically or in the
field.
The upgradient influence can be
determined from the equation
D = 4/3 H tan a
u
where
D = effective distance of drawdown
u .
upgradient
H = saturated thickness of the
water bearing strata not
affected by drainage
tan a = angle between the initial
water table or ground surface
and the horizontal plane.
The theoretical expression for the
downgradient influence is
= K tan q
where
The shape of the drawdown curve
upgradient of the site is independent of
hydraulic conductivity but is a function
of head or hydraulic gradient. The
upgradient influence or drawdown extends
K = hydraulic conductivity
q = drainage coefficient
h = effective depth of the drain
h_ = desired depth to the water
table after drainage
10]
-------
TABLE 7. STEADY STATE DRAIN SPACING FORMULA FOR PARALLEL RELIEF DRAINS
Situation
Formula
Description
Flow to drains resting on an
impermeable barrier (5).
Q =
4 K(H2 - D2)
when D « H,
Q =
Q = discharge rate per unit surface
area (m/day)
K = hydraulic conductivity (m/day)
H = height of the water table above
the impervious floor (m)
D = Thickness of the aquifer below
drain level (m)
L = drain spacing (m)
h = height of water table above drain
level, midway between 2 drains (m)
Flow to drains at the interface
of two permeable layers (9).
Presuming D « H
= 8K?dh + 4Kih2
X .^•-
d = thickness of the "equivalent
layer" or area'of radial flow
(m)
-------
D = distance from the ground sur-
face to the water table at the
drain
D_ = distance from the ground
surface to the water table
before drainage at the distance
D downgradient from the drain
D = distance downgradient from the
drain where the water table is
lowered to the desired depth.
In the equation given above, D and D_ are
interdependent variables. In obtaining the
solution to the equation, it is necessary
to estimate the value of D- and then make
a trial computation. If trie actual value
of D_ at distance D is appreciably
different, a second calculation is neces-
sary. Where tan a is uniform throughout
the area, D. can be considered equal to D .
D is the distance downgradient from the
drain to which the water table is lowered
to the depth required to capture the
plume. If a second interceptor is needed,
it would be located the distance D
downgradient from the first (16).
In addition to designing the sub-
surface drainage system for appropriate
depth and spacing, the hydraulic design
must also be developed. The drain's pipe
size and gradient must be adequate to
carry away the water after it enters the
pipe. The gradient of the drain pipes
should be great enough to result in a flow
velocity that prevents siltation yet will
not cause turbulence. SCS has published
data on minimum recommended grades for
various pipe diameters and maximum
velocities recommended for various soil
types (16). The diameter of the drain for a
given capacity is dependent on flow,
hydraulic gradient, and the roughness
coefficient, n, which is a function of the
hydraulic resistance of the drain materi-
al. The formula for hydraulic design is
based on the Manning formula for rough
pipes which is written as follows:
Q = av = (l/n)(R2/3)(i1/2)
in which
Q = discharge
R = hydraulic radius (equal to the
wet cross-sectional area, a,
divided by the wet perimeter)
i = hydraulic gradient
1/n = roughness coefficient
v = velocity
a = area
SCS(16) and Cavelaars(3) have developed
nomographs for estimating pipe diameter
based on the Manning formula when hydrau-
lic gradient, flow rate, and drainage area
are known. Roughness coefficients have
been published for all drain types.
The other criterion for successful
hydraulic design is to ensure that the
pipe will accept the drainage water when
it arrives at the drainline. To meet this
criterion, the relationship between the
hydraulic conductivity of the gravel
envelope and/or filter material, the
perforations in the drain pipe, and the
hydraulic conductivity of the base soil
material must be evaluated. The primary
function of a filter is to prevent soil
particles from entering and clogging the
drain. The function of an envelope is to
improve water flow into the drains by
providing a material that is more per-
meable than the surrounding soil. Although
filters and envelopes have distinctly
different functions, it is possible to
meet the requirements of both a filter and
an envelope. SCS(16) has developed dis-
tinct design criteria for gravel filters
and envelopes whereas the Bureau of
Reclamation (2) has developed one set of
standards for a well-graded envelope which
meets the requirements of both a filter
and an envelope. For tile drains the
criteria are based upon a comparison of
the grain sizes of the envelope and base
soil material. Where perforated pipe
drains are being used, the minimum size of
the envelope material is based on the size
of the perforations. Geotextile fabrics
offer an alternative to the more con-
ventional sand and gravel filters provided
the fabrics are compatible with the waste
components in the leachate so that the
drains do not clog.
Design of the drainage sump and
pumping plant are also considered part of
the total hydraulic design. The main steps
in their design include:
103
-------
• Determining the maximum inflow to
the sump
• Determining the amount of storage
required
• Calculating the pumping rate
• Determining the start, stop, and
discharge levels
• Determining the type and size of
the storage tank
• Selecting the pump and the motor.
Construction and Installation
Construction and installation of
subsurface drains can be divided into two
major phases—trench excavation and drain
installation. Trench excavation is often
the most complex and costly aspect of
construction and installation. The ease
or difficulty of excavation can have a
dramatic effect on the cost of the total
installation. A difficult excavation may
result in exclusion of subsurface drainage
as a viable technique because of prohi-
bitive costs. Table 8 summarizes the
various activities involved in trench
excavation, equipment or methods suitable
for carrying out these activities, and the
major application(s) and limitation(s) of
these methods. Many of these methods are
also used in construction of barrier
walls, discussed in the next section. Once
trench excavation is completed the com-
ponents of the subsurface drain can be
installed. This process consists of
installing drain pipe bedding, drainage
pipe, a gravel or soil envelope, filter
fabric, backfill, and auxiliary components
including manholes and pumping stations.
BARRIER WALLS
Low permeability barriers can be used
to divert groundwater flow away from a
waste disposal site or to contain contami-
nated groundwater leaking from a waste
site. There are three major types of low
permeability barriers that are applicable
to leachate plume management—slurry
walls, diaphragm walls, and grout curtains
(10, 17).
A slurry wall is a subsurface barrier
that is formed through the excavation of a
trench using a bentonite and water slurry
to support the sides. The trench is then
backfilled with materials having far lower
permeability than the surrounding soils.
The slurry backfill trench or slurry wall
reduces or redirects the flow of ground-
water .
A diaphragm wall is a subsurface
barrier designed for structural strength
and integrity in addition to low per-
meability. Diaphragm walls can be made of
cast-in-place concrete or precast panels
with cast-in-place joints.
A grout curtain is a subsurface
barrier formed through the pressure injec-
tion of one of a variety of special
grouts, into a rock or soil body to seal
and strengthen it. Once in place, these
grouts set or gel into the rock or soil
voids, greatly reducing the permeability
of and imparting increased mechanical
strength to the grouted mass. As this
process is carried out, a grout wall or
curtain results. Because a grout curtain
can be three times as costly as a slurry
wall, it is rarely used when groundwater
has to be controlled in soil or loose
overburden. Grout is used primarily to
seal voids in porous or fractured rock
when other methods of controlling ground-
water are impractical.
Barrier walls are classified by the
materials of which they are composed and
the position in which they are placed with
respect to the pollution source. There
are three major types of barrier walls
that are categorized according to the
material that is used to backfill the
trench:
• Soil-bentonite
• Cement-bentonite
• Diaphragms.
Soil-bentonite walls are composed of
soil materials (often the trench spoils)
mixed with small amounts of the bentonite
slurry from the trench. Cement-bentonite
walls are composed of a slurry of portland
cement and bentonite. Diaphragm walls are
104
-------
TABLE «. METHODS AND EQUIPMENT APPLICABLE FOR TRENCH EXCAVATION (4, 12, 14)
Activity
Me thod/Equipment
Applications/Limitations
Subsurface Exploration
Direct excavation (using backhoe
and ripper)
Drilling
Seismic testing
sledge hammer-disk
blasting
Applicable under most circumstances.
Usually carried out to determine extent of
groundwater and soil contamination. This data
may be applicable if appropriate drilling logs
are kept.
Suitable for limited explorations - distances
to 300 feet and depths up to 100 feet.
Not recommended for hazardous waste sites; may
result in rock fracturing, which can cause
movement of the plume to an underlying aquifer.
May result in secondary detonation of explo-
sives or air pollution.
Rock Fragmentation
Mechanical ripping
Impact methods
Blasting
Non-explosive demolition
Use is limited to trench depth of about 6 feet
or less, unless the ripper can be lowered into
the trench to reach greater depths.
Low production rate generally limits usage to
a small scale.
(see comment above)
Alleviates environmental problems associated
with blasting, but the process is expensive
and time consuming.
-------
TABLE 8. (Continued)
Activity
Method/Equipment
Applications/Limitations
Excavation
Trenchers
Backhoe
Clamshell and draglines
Can provide continuous excavation, but is only
suitable in soil or well fragmented rock. Maximum
depth for a large wheel trencher is 8.5 feet; 27
feet for bucket ladder type trencher.
Excavate to depths of up to 70 feet; can excavate
rock up to one-half the bucket diameter; diameters
up to 4 yd^ are available. Large size may preclude
use in urban areas
Used when the trench depth is below the limit of
other equipment. Limited to loose rock and earth.
Wall Stabilization Methods
Shoring
Ground freezing
Open cuts
Most widely used stabilization method. Method
varies with the depth of the trench.
Expensive; may require very low temperature to
freeze clay.
Limited to very shallow trenches
Dewatering
Open pumping
Predrainage using well points
or wells
Groundwater cutoff
Applicable to shallow trenches with stable soils of
low permeability. May require treatment of contami-
nated groundwater; Inexpensive
Applicable under most circumstances but depth of
trench and soil permeability may make this method
cost prohibitive
Often,used together with open pumping to prevent low
rates of seepage
-------
TABLE 8. (Continued)
Activity
Method/Equipment
Applications/Limitations
Grade Control
Visual method (target grade control)
Automatic method (laser beam)
Precision and accuracy depend on machine operators
skills and alertness
Trenches and backhoes can be equipped with this
system; high accuracy and precision
-------
composed of pre-cast or cast-in-place
reinforced concrete panels (diaphragms)
installed by excavating a short, slurry-
supported trench section, using a clam-
shell bucket or other suitable piece of
equipment. Upon completion of excavation,
the trench section is filled with a
pre-cast, reinforced concrete panel or
tremied concrete around a reinforcement
cage. Alternate or primary panels are
installed first followed by the secondary
panels. Joints between panels are formed
using stop end tubes that are concreted
after adjacent panels are completed.
Another method involves using a cement-
bentonite slurry as the excavation fluid
which forms the joint between panels when
set.
In general, soil-bentonite walls can
be expected to have the lowest perme-
ability, the widest range of waste com-
patibilities, and the lowest cost. They
also offer the least structural strength
(highest elasticity), usually require the
largest work area, and are restricted to a
relatively flat topography unless the site
can be terraced.
Cement-bentonite walls can be instal-
led at sites where there is insufficient
work area to mix and place soil-bentonite
backfill. By allowing wall sections to
harden and then continuing the wall at a
higher or lower elevation, they can be
installed in a more extreme topography.
Although cement-bentonite walls are
stronger than soil-bentonite walls, they
are at least an order of magnitude more
permeable, resistant to fewer chemicals,
and more costly.
Diaphragm walls are the strongest of
the three types as well as the most
costly. Provided the joints between
panels are installed correctly, diaphragm
walls have approximately the same per-
meability as cement-bentonite walls and,
because of a similarity of materials,
about the same chemical compatibilities.
Diaphragm walls are most typically used in
situations requiring structural stength
and relatively low permeability.
Combinations of these three major
backfill types may be included within the
same wall. For example, a soil-bentonite
backfill may be used for the majority of a
wall with cement-bentonite backfill being
used in areas that require greater
strength, such as a road or rail crossing.
Being able to combine the various types of
walls makes this technique adaptable to a
wider range of site characteristics.
The vertical and horizontal posi-
tioning of a barrier wall with respect to
the location of the pollution source and
groundwater flow characteristics is known
as configuration. There are two types of
configurations—vertical and horizontal.
The wall configuration, combined with any
other remedial measures (e.g., pumping,
capping), determine how effective a
barrier wall will be in controlling
leachate plume migration.
Vertical configuration refers to
the depth of the wall with respect to both
geologic formations and the water table.
Based on vertical positioning, walls are
either "keyed" into a low permeability
formation below the aquifer, or placed to
intercept only the upper portion of the
aquifer. The latter type, known as a
"hanging" barrier wall, can be used to
control contaminants, such as petroleum
products, which do not mix with the
groundwater but float on top of it. In
these situations, the barrier need only
extend to a depth in the water table
sufficient to intercept the contaminants.
Keyed-in barrier walls are excavated
through the water table to a confining
layer below it to contain contaminants
that mix with or sink to the bottom of the
aquifer. The connection between the wall
and the low permeability zone is very
important to the overall effectiveness of
the barrier.
Horizontal configuration refers to
the positioning of the barrier relative to
the location of the pollution to be
controlled and the direction of ground-
water flow (i.e., the gradient). Based on
horizontal configuration, barrier walls
may completely surround the pollution
source or be placed upgradient or down-
gradient from it.
108
-------
Circumferential placement refers to
placing a barrier wall completely sur-
rounding the site. Although this requires
a greater wall length and higher cost than
either upgradient or downgradient
placement alone, it does offer some
advantages and is a common practice. A
circumferential barrier wall, when used
with a surface infiltration barrier such
as a cap, can greatly reduce the amount of
leachate a site will release to the
environment. If a leachate pumping or
drainage system is used, as often is the
case, a waste site can be virtually
dewatered. In addition to vastly reducing
the amount of leachate to be treated,
dewatering can help increase the longevity
of the wall.
Upgradient wall placement refers to
the positioning of a wall on the ground-
water source side of a waste site. This
type of placement can be used where there
is a relatively steep gradient across the
site in order to divert uncontaminated
groundwater around the wastes. In such
cases, clean groundwater is prevented from
becoming contaminated thereby reducing
leachate treatment requirements.
Placement of a barrier wall on the
side opposite the groundwater source is
referred to as downgradient placement.
This placement configuration does nothing
to limit the amount of leachate being
generated and so is practical only in
situations where there is a limited amount
of groundwater or contaminant flow such as
near drainage divides. This type of
barrier can contain leachate so it can be
recovered for treatment. Although this
wall configuration may be keyed-in for
miscible or sinking contaminants, it is
most often hung to contain and recover
floating contaminants.
The effectiveness of barrier walls
can be dramatically increased by incor-
porating additional remedial measures into
the overall plan addressing the problem.
A barrier wall combined with a groundwater
pumping system, for example, can be used
to trap contaminated groundwater and then
pump it to a treatment system. Addition-
ally, surface drains can prevent water
from infiltrating into a waste disposal
area while the barrier wall reduces
groundwater flow into the site. Other
remedial measures which may be used in
conjunction with barrier walls include
surface sealing, sheet piling, synthetic
membrane installation, and grouting.
Grouting materials fall into three
basic groups—cement, bituminous, and
chemical. Some specific grout mixtures
include portland cement, sand-cement,
clay-cement, clay-bentonite, bituminous
emulsions, sodium silicate, and
acrylamide. The applicability of each
material is based on the size of the
openings in the soil or rock formation,
and the anticipated area of grout penetra-
tion. The major grouts in use are cement
and clay which make up approximately 95
percent of all grouts used.
Cement grouts utilize materials that
set, harden, and do not disintegrate in
water. Because of their large particle
size, cement grouts are more suitable for
rock than for soil applications.
Materials may be added to cement grouts to
improve their applicability. Sand may be
added to portland cement to create a grout
suitable for coarse materials, while
bentonite may be added to improve the
penetration of cement in alluvial soils.
Clays have been used widely as
grouts either alone or in formulations. In
general, coarse sands and gravels are
initially grouted using clay or clay-
cement because they are inexpensive.
Bentonite is an excellent clay grouting
material because of its swelling and gel
formation properties (1).
Grout curtains are the most practical
and efficient method for sealing fissures,
solution channels, and other voids in
rock. There are four basic techniques for
installing a grout curtain. These are:
• Stage—up method
• Stage-down method
• Grout-port method
• Vibrating beam method (8).
109
-------
The first three methods are injection
methods in which the grout is injected
from either the bottom of a borehole to
the top (stage-up), or the top of a
borehole to the bottom (stage-down), or
through a slotted injection pipe that has
been sealed into the borehole with a
brittle portland cement/clay mortar jacket
(grout-port). The grout port method
utilizes rubber sleeves which cover the
outside of each slit (or port) permitting
grout to flow only out of the pipe (8). The
vibrating beam method is not an injection
technique, rather, it is a way of placing
the grout in which an I-beam is vibrated
into the soil to the desired depth and
then raised at a controlled rate. As the
beam is raised, grout is pumped through a
set of nozzles mounted on the base of the
beam, thus filling the newly formed
cavity. When the cavity is completely
filled, the beam is moved in the direction
of the wall, leaving a suitable overlap to
ensure continuity. The process is con-
tinued until a wall of the desired length
is constructed (?)•
As with slurry walls, placing a grout
curtain upgradient from a waste site can
redirect flow so that groundwater no
longer contacts the wastes that are
creating the leachate plume. However,
placement of a grout curtain downgradient
from a hazardous waste site may not be
successful because of grout/leachate
interactions. For example, in many
instances it would be difficult to control
the grout setting time, and thus, be
difficult to emplace a curtain of reliable
integrity. Additional problems can occur
in attempting to grout a horizontal
curtain or layer beneath a waste site in
that injection holes must be drilled
either directionally from the site
perimeter, or directly through the wastes
(10). The first case may not always be
feasible and may be quite costly and the
second might be quite hazardous to con-
struction crews.
IN-SITU TECHNIQUES
Leachate plumes may be treated in-
place by certain biological and chemical
treatment methods that are specifically
designed to remove plume contaminants.
Contaminant plumes consisting chiefly of
biodegradable organics may be treated by
in-place biodegradation, or bio-
reclamation. Chemicals may also be injec-
ted into the leachate plume to neutralize,
stabilize, or mobilize contaminant ma-
terials. For example, soil flushing is
used as an in-situ technique to flush
residual contaminants from soil particles
in order to mobilize them for collection
and treatment. Usually water is used as
the flushing media, however, the use of
dilute solutions of surfactants and or-
ganic solvents has also been proposed.
Bioreclamation
Bioreclamation is an in—situ ground-
water treatment technique based on the
concept of utilizing microorganisms, com-
bined with aeration and the addition of
nutrients, to accelerate the biodegrada-
tion rate of groundwater contaminants.
Many species of bacteria, actino-
mycetes, and fungi have been found to
degrade hydrocarbons associated with pe-
troleum. Bacteria are the prime micro-
organisms involved with biodegradation of
organics in groundwater. Naturally
occurring species of the genera
Pseudomonas, Arthrobacter, Nocardia
Achromobacterium, and Flavobacterium have
been found to attack petroleum hydro-
carbons and other organic chemicals (13).
These bacteria can be stimulated by adding
nutrients and oxygen to develop a po-
pulation that is adapted to readily
degrade organic chemicals present in
groundwater.
An alternative to developing adapted
populations from naturally occurring
bacteria is to inoculate the subsurface
with mutant microorganisms developed in
the laboratory to degrade specific organic
chemicals or chemical groups. The
particular advantages of this alternative
is that overall biodegradation rates of
specific organics may be increased and the
time required for adaptation of a natural-
ly occurring population is eliminated. In
this method, the parent microorganisms are
collected from a naturally occurring
110
-------
source containing the specific chemical
groupings, such as oil refinery treatment
plant sludge for phenol and cyanide
treatment. Mutant strains are developed
from irradiation and these mutant bacteria
are cultured in large amounts and packaged
for use. The mutant bacteria can be
cultured in even larger amounts in holding
ponds at the site and introduced into the
subsurface by spray irrigation, surface
flooding, or subsurface injection.
As mentioned earlier, nutrients and
oxygen are required for proper cell growth
and respiration. Cell nutrients include
nitrogen, phosphorus, and trace elements
(i.e., potassium, sulfur, sodium, calcium,
magnesium, iron, and copper). Groundwater
usually contains sufficient amounts of
nitrogen and phosphorous for degradation
of only 10 to 20 milligrams per liter
(mg/1) of organic material. More nutrients
must be added for biodegradation of higher
contaminant levels. Oxygen is also re-
quired for the aerobic decomposition of
organics by bacteria. Roughly 3 or 4 mg/1
of oxygen are required for every milligram
per liter of organic constituent
degraded (15). A constant supply of dissolved
oxygen must be supplied to the contami-
nated groundwater. Subsurface aeration
has been the method most commonly used
during previous bioreclamation of ground-
waters contaminated with gasoline.
However, since this method can only
provide a maximum of about 10 mg/1 of
dissolved oxygen, biodegradation of high
levels of organics in the subsurface can
be limited. Alternative oxygenation tech-
niques include the use of pure oxygen
systems, ozone, and low levels of hydrogen
peroxide.
The selection of bioreclamation as a
plume management technique depends on the
biodegradabi1ity of the components in the
contaminant plume. Biodegradabi1ities of
various organic substances are based on
the ratio of Biochemical Oxygen Demand
(BOD) to Chemical Oxygen Demand (COD).
Compounds with a BOD/COD ratio of less
than 0.01 are considered relatively
undegradable, compounds with a ratio of
0.01 to 0.1 are considered moderately
degradable, and compounds with a ratio of
0.1 or greater are considered
degradable (11).
Implementation of the bioreclamation
process involves the placement of extrac-
tion wells to control migration of the
contaminant plume by pumping. Groundwater
pumped to the surface is mixed with
nutrients and reinjected upgradient of the
extraction wells. Specialized mutant
bacteria may also be added along with the
nutrients. The groundwater may be oxy-
genated via air, oxygen, or hydrogen
peroxide.
The bioreclamation technique has been
used successfully in a number of cases to
treat contaminated groundwater plumes from
underground gasoline and hydrocarbon
leaks. It has not yet been demonstrated
for groundwater treatment at uncontrolled
hazardous waste disposal sites. However,
its potential to treat hydrocarbon conta-
minated groundwater establishes it as a
viable technique.
In-Situ Chemical Treatment
In-situ chemical treatment techniques
involve the injection of a chemical into a
leachate plume to neutralize, detoxify,
precipitate, or othewise affect the con-
taminant materials. These techniques are
highly dependent on the contaminant and
have in the past been used only for spills
of specific chemicals. Dilute solutions of
acids or bases, such as nitric acid or
sodium hydroxide, could theoretically be
used to neutralize acidic or basic ground-
water contaminants. A system of extrac-
tion/injection wells could be used to
disperse the neutralizing agent, and
contain and cycle groundwater until the
appropriate pH was attained. Similarly,
chemical agents could be used in this
manner to detoxify plume contaminants.
Sodium hypochlorite, for instance, has
been used to oxidize cyanide—contaminated
groundwaters. Other oxidizing chemicals
such as hydrogen peroxide or ozone may
find potential application in this type of
remedial approach. Solutions of sodium
sulfide have also been proposed to pre-
cipitate toxic metals from groundwater,
thereby resulting in their immobilization.
Recently, an underground spill of acrylate
monomer was attenuated by the injection of
a catalyst which caused the plume of
acrylate monomer to polymerize and
solidify (18).
Ill
-------
THE INFLUENCE OF SELECTED ORGANIC LIQUIDS ON
THE PERMEABILITY OF CLAY LINERS
K. W. Brown, J. W. Green and J. C. Thomas
Texas Agricultural Experiment Station
Soil and Crop Sciences Department
College Station, Texas 77843
EPA Grant No. CR-808824020
ABSTRACT
Research on the influence of organic chemicals on permeability of
clay soils used to line waste impoundments has continued. Three types
of simulated clay liners have been prepared by mixing selected clay
minerals with sandy loam soil to achieve permeabilities to water
normally considered acceptable for waste impoundments. These liners
were subjected to two hazardous wastes, a xylene paint solvent waste
and a contaminated acetone waste, in both laboratory permeameters and
field cells. Wastes were tagged with tracer dyes to facilitate visual
and microscopic observation of their movement through soil. Laboratory
studies were directed at determining the influence of different initial
moisture contents and elevated pressures on the permeabilities of clay
liners subjected to these wastes. Both the initial moisture content and
elevated pressures influenced the time required for permeability
changes to occur. The pressures at which the soils were tested did not
influence the final permeabilities attained. Samples tested at
compaction moisture typically attained final permeabilities one to two
orders of magnitude greater than those attained on samples presaturated
wi th water.
The field study is still in progress, but the preliminary data
corresponds closely to the laboratory predictions of increased
permeability. Visual observations of dye paths and chemical analyses of
small soil samples from both laboratory and field studies indicated
that the wastes moved through preferential channels in the center of
the soil mass and not along the edges of the permeameters or test
cells.
INTRODUCTION
Compacted clay soils have the past been developed using
long been used to line waste only water as the permeant
storage and disposal facilities liquid. Such installations have
including waste piles, surface generally been successful when
impoundments, and landfills. the primary liquid to be retained
Design specifications have in was relatively pure water.
114
-------
It has long been known that
smectite clay subjected to some
organic chemicals exhibited
smaller spacing between adjacent
crystalline layers than when
exposed to water (Barshad, 1952).
Only recently, however, have
measurements been made of the
impact of organic liquids on
permeability of recompacted clay
soils. Previous reports by
Anderson e^t_ aj^. (1982) and
Brown and Anderson (1983)
evaluating the impact of
concentrated organic liquids on
the permeability of four native
clay soils indicated that the
permeabilities may be 2 to 3
orders of magnitude greater than
those measured with water.
Observations of the permeated
soil cores indicated physical
changes in the soil structure.
The present studies were
conducted with three clay soils
compacted in fixed wall
permeameters . Liquids were passed
through the soils under elevated
hydraulic gradients (61.1 and
361.6) to shorten the testing
time. Although laboratory
determined permeabilities have
been traditionally used to
estimate field permeability of
such soils, it was believed
necessary to collect data using
wastes under conditions as
similar as possible to those
anticipated in a field
installation of clay soils
compacted to serve as a landfill
liner. Thus, a field study using
three soils of different clay
mineralogies permeated by two
organic solvent wastes was
undertaken to further evaluate
the validity of the laboratory
testing procedure and to provide
field data on the impact of
organic liquid wastes on the
permeability of clay liners.
MATERIALS AND METHODS
Soj.1
Each of the three clay soils
were blended with a predetermined
amount of sandy loam soil to
attain perme ab i l_i± ie s in tnf)
range of 1 x 10 to 1 x 10~
cm/sec. These three clay soils
were selected to represent the
range of materials most widely
available and used for
constructing waste impoundment
liners. In all following
discussion, the clay soil blends
will be referred to by their
dominant mineralogies of
kaolinite, mica, and bentonite.
The textures and mineralogies of
the clay soil blends are given in
Table 1. The chemical properties
are given in Table 2.
Exchangeable cations in the
kaolinitic and micaceous clay
soils were predominantly calcium,
while the bentonitic clay soil
was sodium saturated.
Was tes
The wastes were selected to
be representative of polar and
nonpolar organic chemical wastes
usually generated in large
quantities. The xylene waste was
a paint solvent used to clean
factory paint sprayer lines and
contained 20% by volume solids.
The solids were predominantly
paint pigments composed of Fe ,
Cr, Ti, Cu, K, S, P, Si, and Al .
The waste was allowed to settle
in drums for 4 months prior to
use. The liquid following
settlement, still containing 25%
by weight plasticizers and
suspended pigments, was decanted
for use in the test cells. A
representative sample of the
waste was distilled, and an
infrared analysis confirmed that
the liquid was xylene with only
trace amounts of water present.
The acetone waste was a chemical
manufacturing waste consisting of
of 91.7% acetone, 4% benzene,
0.6% phenol, and 3.7% unknown
(percent by weight). Two dyes
were added to each waste to
facilitate visual observations of
their movement. In both wastes,
the dyes used were Automate Red B
and Fluorescent Yellow at 154 and
50 mg/1, respectively.
115
-------
CROSS SECTION OF TEST CgLLS
Figure 1. Schematic diagram of a
crossection of a field
cell.
Typically, 4 vertical profiles
were sampled in 2.5 cm depth
increments for chemical analysis,
and 10 sample profiles were taken
for morphological study. Addi-
tional samples were taken from
readily observable cleavage
planes, which were often colored
by dye, and from cut surfaces
representing the inside of single
natural soil aggregates or peds.
Chemical Analy_s_is
The determination of xylene
in liner soil samples was
accomplished by mixing the bulk
soil-sample, taking a 25 g
subsample, and blending the
subsample with 50 ml of 20%
in an
Waring Blender.
The mixture was filtered and
brought to a 50 ml volume with
CH0C1,,. An aliquot was then
by a high performance
chromatography (HPLC
Model 3390) with a 4 mm
CH OH:80%CH Cl
explosion proof
analyzed
liquid
Beckman
column and 250 nm UV detector.
The flow rate was 1 ml/min with
80:20 methano1:water as solvent
system.
Analysis of leachate collec-
ted from the field cells for
xylene was accomplished by
measuring the volumes of the
aqueous and organic phases after
they separated.
For the determination of
acetone in leachate samples, a 5
ml aliquot was transferred to a
septum-capped vial and
equilibrated at 25°C for 30
min. A 2 ml sample of the
headspace was analyzed by flame
ionization gas chromatography
(GC, Tracer). Acetone in soils
was determined by first mixing
the bulk sample and selecting a 5
g subsample. The subsample was
then transferred to a septum-
capped vial and analyzed by GC
as above. The GC was equipped
with a 1.8 m x 4 mm id glass
column packed with 30 cm of
Porapak Q followed by 150 cm of
5% Carbowax 20 M on 80/120 Gas
Chrom-Q. The conditions were as
follows: N^ carrier gas at 20
i sotherma1 oven
ml/min and the
temper a ture at
40°C.
"ultrasphere ODS" reverse phase
RESULTS AND DISCUSSION
Comparison of tests per-
formed at different pressures and
times was accomplished by plot-
ting the calculated permeability
as a function of the pore volume
of effluent which passed through
the 1ine r .
Laboratory Results
The laboratory studies using
the three soils at a series of
pressures resulted in conclu-
sions quite similar to those
reported previously (Anderson
e_t_ aAj^ 1982; Brown and
Anderson, 1983) for other soils
and solvents. In most cases, the
results from replicates were
virtually identical with + 0.5
orders of magnitude. Only in
118
-------
cases where flow was too rapid to
collect adequate data were major
differences between replicates
ev iden t.
Representative data are shown
in Figures 2 through 5.
Permeabilities to water on
similar soils were within + 0.25
orders of magnitude before
solvents were introduced. After
the solvents began to permeate
the soil, the permeabi1ites
increased, typically to values
which were two to four orders of
magnitude greater than those
obtained with water. As can be
seen by comparing Figures 2 and
3, the solvents broke through the
clay soils at lower pore volumes
when the soils were initially at
the as-compacted moisture
contents ( nonsaturated ) , than
when the soils were initially
saturated with water. For the
water-saturated soils, the pores
were initially filled with water
that had to first be displaced
before the solvent could move
through. Highest permeabilities
measured on soils which were not
presaturated were typically one
or two orders of magnitude
greater than those which were
initially saturated.
Differences in the behaviors
exhibited by polar and nonpolar
solvents were the same as those
seen previously (Anderson e_t^
a 1., 1982; Brown and Anderson,
1983). A comparison of break-
through of the nonpolar xylene
and polar acetone are shown in
Figures 4 and 5b. The tests
showed that the xylene rapidly
increased the permeability of the
clay soils after the cumulative
flow exceeded 0.2 to 0.4 of a
pore volume. Permeability
changes caused by acetone
characteristically resulted in a
small decrease in permeability
reaching the lowest value at 0.5
pore volume; this was followed by
a subsequent increase in
permeability. The different
response to the polar acetone and
nonpolar xylene is believed to be
caused by acetone first causing
swelling of the clay soil with
low concentrations initially
diffusing into the pores,
followed by shrinkage when more
of the water was displaced by the
solvent.
Although data are still being
collected for all replicates, it
appears that hydraulic gradients
used in conducting tests of
permeability have little
influence on either the increase
in permeabilities observed or on
the final permeability values
achieved after the solvent
penetrates the compacted soil
mass. This conclusion is reached
by comparing the data presented
in Figures 5a-5c for kaolinitic
clay exposed to acetone at three
different hydraulic gradients.
These data show a trend similar
to others that have been
collected (Anderson e_t^ a 1 . ,
1982).
Increases in permeability as
described in detail above were
observed for all three soil
types, but the data collected
from bentonitic clay soils thus
far did not exhibit an increase
in permeability until a greater
fraction of the pore volume was
displaced than was necessary for
the other soils.
Fie Id Results
Permeab i1i t y
The field studies are still
in progress. However, within 12
months after installation, xylene
had penetrated 10 of the 12 clay
soil liners, while acetone had
penetrated 2 soil liners.
Permeability data for xylene are
presented in Figures 6 to 8. The
kaolinitic clay soil liner
(Figure 6) shows a very clear,
almost immediate increase in
permeability. After passage of 1
pore volume of leachate through
the compacted clay, the
permeability had increjj.sed to an
average of ^7 x 10 cm/sec.
The largest permeability,
replicate 1, was an order of
119
-------
PORf VOMIME
Figure 2. Permeability of presaturated
soil (micaceous clay dominant)
to xylene-rich waste. The tests
were performed using a hydraulic
gradient of 91.
LAS MICA
XYLENE
NONSATUKATEO
GRADIENT 91
LABORATORY K/ALUE WITH WATER
PORE VOLUME"
Figure A.
I/O
o REP I
• REP 2
Figure 3. Permeability of nonsaturated
soil (micaceous clay dominant)
to xylene-rich waste and 0.01N
CaSO^ as a function of pore
volume. The tests were performed
using a hydraulic gradient of 91.
LAB. KAOUNITE
XYLENE
PRESATURATED
GRADIENT 91
PORE ' VOLUME
Permeability of presaturated
soil (kaolinite clay dominant)
to xylene as a function of
pore volume. The tests were
performed using a hydraulic
gradient of 91.
PORE VOLUME
Figure 5a. Permeability of presaturated
soil (kaolinite clay dominant)
to acetone as a function of
pore volume. The tests were
performed using a hydraulic
gradient of 31.
120
-------
o REP I
• REP 2
KAOUNITE
ACETONE
PflESATTJRATED
GRADIENT 91
Figure 5b. Permeability of presaturated
soil (kaolinite clay dominant)
to acetone as a function of
pore volume. The tests were
performed using a hydraulic
gradient of 91.
KAOUNITE
ACE row
PflE SATURATED
GRADIENT 181
Figure 5c. Permeability of presaturated
soil (kaolinite clay dominant)
to acetone as a function of
pore volume. The tests were
performed using a hydraulic
gradient of 181.
1,0s
m
<
10
O REP I
• REP 2
O REP 3
• REP 4
15 cm KAOLINITE
XYLENE
GRADIENT 7
5.5xl69 LABORATORY VALUE WITH WATER
PORE VOLUME'
T
10'5
id"61
10
10
REP 1
REP 2
REP 3
REP 4
15 CM MICA LINER
XYLENE
GRADIENT 7
r m
^LABORATORY VALUE
WITH WATER
1 2
PORE VOLUME
Figure 6. Permeability of kaolinite soil
to xylene in a field test
using a hydraulic gradient of
7.
Figure 7.
Permeability of a mica soil to
xylene in field test using a
hydraulic gradient of 7.
121
-------
15cm BENTONITE
XYLENE
GRADIENT 7
LAB VALUE W/ WATER
'PORE VOLUME2
Figure 8. Permeability of
bentonite soil to
xylene in a field test
using a hydraulic
gradient of 7 .
15 cm MICA LINER
ACETONE
GRADIENT 7
• ; LAB VALUE W/ WATER
'PORE VOLUME2
Figure 9. Permeability of mica
soil to acetone in a
field test us ing
hydraulic gradient of
7 .
magnitude above the average. Data
from the micaceous clay soil
liner (Figure 7) show a similar
family of curves. Beginning
permeabilities were well below 1
x 10 cm/sec. However, after
the passage of 1 pore volume, the
perme ab i_l.i t ies had increased to
6 x 10 cm/sec. Most data is
available for replicate 1 which
continued to increase in
permeability for the entire
experimental period reaching 1 x
1 0
cm/ sec .
Increases in permeability of
the bentonitic clay soil liner
(Figure 8) were more dramatic
than that of the other two clays.
Permeability of replicate 1 of
the bentonitic clay soil liner
was initially close to 1 x 10
cm/sec_.and very quickly rose to 2
x 10 cm/sec. Replicate 2
also exhibited a similar trend
although not as dramatic an
increase. The increases in
permeability of as much as 2
orders of magnitude occurred in
0.5 of the pore volume as
c ompa red
vo1ume s
similar
clay soil
to 1
required
increase
1iner s .
or
to
in
more pore
cause a
the other
The effect of acetone on the
permeability of micaceous clay
soil liners is shown in Figure 9.
There was an initial decrease in
permeability during the first 0.5
of a pore volume, after which
there was an increase in
permeability with final values
greater than the initial
permeability. This behavior
be characteristic of
of acetone on the
of compacted clay
behavior has been
appears to
the impact
perme ability
soils: this
consistently observed in both
122
-------
laboratory and field studies of
all soils tested.
Morphology
The compacted clay soil
liners in the field cells were
carefully removed and dissected
to allow for both visual
inspection and photographic
documentation of the exposed
surfaces. The path of the xylene
waste was much more evident than
that of the acetone waste because
of the presence of the paint
pigments. When the HOPE was cut
away at the level of the clay
liner, it was evident in most
cases that the waste had flowed
along the sidewall next to the
liner down to the level of the
flap (See Figure 1). However, the
clay liner adjacent to the
sidewall was free of dye below
the flap which indicated that the
flap had functioned as planned in
preventing the waste from
bypassing the clay liner. Using a
spade, the clay liner was cut in
sections 10 cm thick. Wherever
possible, large sections of the
clay were broken free so that
fracture planes that were likely
to be paths of substantial flow
would be exposed. In most
instances, dye and pigments were
evident on the surfaces of these
cracks. The waste saturated the
upper 1 or 2 cm of clay soil
liners in nearly all the cells
examined. In addition, it
appeared, based on the dyed soil
surfaces, that the waste moved to
lower depths through preferential
channels. When these channels
reached the interface between the
first and second lift, the waste
generally spread out and
permeated the lower 1 to 2 cm of
the upper lift. Again, it
appeared, based on dyed soil
surfaces, that the waste then
moved downward through
preferential channels through the
lower layer of compacted soil.
Occasionally, preferential
channels, the faces of which were
coated with dye, were found to
extend from the top to the bottom
of the liner. When the soil mass
was broken along spontaneously
developed cleavage planes,
vertical surfaces 10 to 20 cm
wide were often coated with dye.
It is not likely that the cracks
resulted from dessication after
compaction since care was taken
to immediately cover all clay
line'rs after compaction to
prevent evaporation. The cracks
observed were either present at
the completion of compaction or
were formed as a result of
physical alteration of the clay
by the wastes.
In all cases, dyed surfaces
were found at apparently randomly
developed spacings throughout the
entire clay liner. This further
indicated that the liquid moved
through the bulk of the soil and
did not bypass the liner. Morpho-
logy (macro scale) of all the
soil mass comprising the liners
could be described as massive,
with few observed strong
horizontal and vertical planes.
A system to impregnate soils,
which had been penetrated with
dyed wastes, was developed to
allow polishing of thin sections
(30 urn) for petrographic micro-
scopic analysis. Preliminary
photographs of the cross sections
of the soils indicate that the
waste moved through selected
pores in the clay liners. Efforts
are underway to develop automatic
techniques to quantify these
observations and permit larger
numbers of sections to be more
readily examined.
Concentrations
Typical xylene concentra-
tions found in profile samples of
each clay soil liner are presen-
ted in Table 3. The variability
of concentrations occurring with
depth in any one profile and
across the clay soil liner at any
given depth indicates that the
movement of the liquid was
through preferential pathways and
did not move in a clearly defined
wetting front through the entire
soil mass. Some samples which
123
-------
Table 3. Concentration of Xylene In MR/B in Soil Saaplea from B Typical Cell of the Three
Different Clay Soil Lliiera.
Depth
0-2.5
2.5-5.0
5.0-7.5
7.5-10.0
10.0-12.5
12.5-15.0
13.0-17.5
17.5-20.0
20.0-22.5
+ randow
4 + not de
Kaollnlte
location No.+
1
235
1214
5060
352
13
9391
3757
3401
termtned.
' 2
18
45
333
419
12
2
722
860
3
44
238
806
564
71
0
2015
nd
4
170
303
122
36
160
1669
nd"
nd
1
237
251
584
714
1289
204
452
2005
nd
Mica
Location No.+
2
112
1027
115»
4035
4310
1903
55
nd
nd
3
51
105
611
2031
77
232
1887
3952
4609
Bentonlte
Location Ho.+
123
533 16 4
5689 8 38
15969 31 685
23099 139 1520
29168 2585 1085
4144 4750 7272
12706 2409 7500
20900 nd 11352
71
4
1
822
17785
1336
2542
6960
had a volume of approximately 25
cm at a particular location
were nearly free of xylene, while
others had contents as high as
20,000 mg/kg. Thus, large
segments of the clay soil liner
had little penetration by the
solvent, while other adjacent
segments were nearly saturated.
Similar evidence that xylene
moved through the soil in cracks
or around blocky sub-angular
structural components of the
liner is presented in Table 4 in
Table 4. Xylene Content in mg/kg
of Dyed and Undyed Sur-
faces (1 mm thick soil
fragment surface) of
Clay Soil Liners.
Lysimeter Dyed
No. Surface
11 19,404
9,149
12 ,456
8 2,814
2,691
Undyed
1 , 147
0
nd +
155
704
which the xylene content of dyed
and undyed surfaces of clay soil
liner fragments are compared. The
xylene content of the dyed
surfaces exposed by breaking the
liner along spontaneous cleavage
planes in a given cell were
least 10 t ime s
found in samples
cut surfaces on
not apparent. The
cent ent of the
add s credence to
that xylene moved
preferential pathways,
along cracks and ped
typ ica1 ly at
greater than
s c raped from
which dye was
lower xylene
undyed surfaces
the cone 1 us ion
through
perhaps
faces, and not uniformly through
the compacted clay soil mass.
Ong^o ing Stud ig s^
The present paper is a report
of progress on an ongoing
project, and several elements of
the research remain to
completed. Approximately half
the field cells remain to
excavated, photographed,
sampled. Laboratory studies
the influence of dilutions
water soluble organics
presently being conducted. It
anticipated that
laboratory evaluations
be
of
be
and
on
of
are
is
fur ther
will
+ not determined
include mixtures
wastes and testing
treated clays.
o f organ ic
of polyme r-
Morpho logical
124
-------
studies to determine the effect
of solvents on soil particle
orientation and arrangement are
also underway; efforts are being
made to quantify the changes in
pore size distribution which may
occur as a result of organic
liquid interactions with the
soil. Other mechanisms of
organic liquid-soil interactions
which may explain the observa-
vations reported here are also
being hypothesized and evaluated.
REFERENCES
Anderson, D. C . ,
J. W. Green.
o f organ ic
pe rmeab i 1 i ty
liners . ^n
(ed.). Land
K. W. Brown and
1982. Effect
fluids on the
of clay soil
D. W. Shultz
Disposal of
Hazardous Waste. EPA-600/9-
82-002, Office of Research
and Development, U.S.
Environmental Protection
Agency, Cincinnati, Ohio.
Barshad, I. 1952. Factors
affecting the interlayer
expansion of vermiculite and
montmori1lonite with organic
substance. SSSAP 16:176-182.
Brewer , R
Miner a 1
Robert
Company ,
York. p.
1976. Fabric and
Analysis of Soils.
E. Krieger Publishing
Huntington, New
50.
Brown, K. W.
1983.
ef f ec t s
of clay
for EPA
806825010.
Hazardous
Division ,
Protect ion
and D. C. Anderson.
Organic leachate
on the permeability
soils. Final Report
Cincinnati, Ohio.
Grant No. R.
Solid and
Was te Res earch
Environmental
Agency,
125
-------
LABORATORY STUDIES FOR DETERMINING FLEXIBLE-MEMBRANE SOIL BEDDING REQUIREMENTS
Donald M. Ladd and Gordon L. Carr
Geotechnical Laboratory
U. S. Army Engineer Waterways Experiment Station
Vicksburg, Miss. 39180
ABSTRACT
Three laboratory tests were developed to simulate field loading conditions on flex-
ible membrane liners during construction of hazardous waste landfills. One test method
utilized a moving pneumatic tire loading, another used a rotating gyratory load, and the
third used a cyclic vertical plate load. Loading conditions and thickness of cover
material over the membrane were scaled using Boussinesq equations to produce vertical
stresses on the membrane similar to those encountered under field conditions.
Test results showed that the moving pneumatic tire load test would be most useful
for determining cover and bedding criteria using available site soils and candidate mem-
branes. Also, a layer of clay soil was effective in preventing puncture of the membrane
by the subgrade.
INTRODUCTION
Background
In recent years the Environmental Pro-
tection Agency (EPA) has sought means for
limiting the hazards involved in disposal
of industrial waste in landfills. The use
of impervious flexible membrane placed
over the subgrade in hazardous waste land-
fill could be a method of reducing the
danger of toxic material from contaminat-
ing the soil and ground water. One prob-
lem with the use of membranes in waste
landfills is that such membranes are sub-
ject to damage during construction and
operation of the landfill. In November
1979, the EPA requested the U. S. Army En-
gineer Waterways Experiment Station (WES)
to conduct a study aimed at determining an
economical method for preventing damage to
flexible membranes during construction and
testing of prototype simulation of waste
landfills utilizing various flexible mem-
branes as liners. Twelve variations of
membranes, subgrades, and protective cov-
ers were tested in this test series. The
testing indicated that with a granular
subgrade all of the membranes would sus-
tain some damage during construction, but
if a clay or a concrete sand were used as
a subgrade, the damage would be reduced.
The test data was reported in "Land Dis-
posal: Hazardous Waste," Proceedings of
the Seventh Annual Research Symposium,
No. EPS-600/9-81-002, March 1981, edited
by D. W. Shultz, Southwest Research Insti-
tute, San Antonio, TX 78284. Only a lim-
ited number of test items were evaluated
in this study and results were variable.
However, the test indicated a sand clay or
silt (small grain size soil material)
would be required under the membrane as
well as on top of the membrane to protect
the membrane from puncture caused by the
action of the traffic to the surface and
subgrade. At the conclusion of the ini-
tial phase of testing and after the re-
sults had been reported, the EPA requested
that additional tests be conducted. In
order to increase the number of tests that
could be conducted and produce more con-
sistent results, laboratory tests were de-
vised that would simulate field testing.
Test Objective
The objective of this investigation
was to develop laboratory tests that could
be used to determine soil cover and
126
-------
bedding requirements for protecting flex-
ible membrane liners from puncture during
construction of hazardous waste landfills.
Scope of Work
Three laboratory tests were developed
to simulate field loading conditions on
flexible membrane liners during construc-
tion of hazardous waste landfills. One
test used a moving pneumatic tire loading,
another used a rotating gyratory load, and
the third used a cyclic vertical plate
load. Tests were conducted using a grav-
elly sand or limestone subgrade under the
membrane liner and a gravelly sand cover.
In some tests, a lean clay or a fabric was
placed between the liner and subgrade to
serve as protection for the membrane
liner. One type and thickness of cover
and three membrane liners were used in
the tests. Other special tests were con-
ducted to develop a test that possibly
could be used as a screening test for mem-
brane liners.
Definitions
Cover material: Material placed on
top of a membrane liner to protect the
liner from puncture during construction of
a landfill.
Bedding material: Material placed be-
tween the membrane liner and the subgrade
to protect the liner from puncture.
Subgrade: Soil used in these tests to
represent natural material upon which a
landfill may be constructed.
DEVELOPMENT OF MODEL PARAMETER
Vertical stresses on the membrane were
selected as the modeling parameter in the
laboratory test to simulate the field
test. It was felt that this was an impor-
tant parameter affecting the behavior of
the membrane and one that could be trans-
lated from the field to the laboratory.
The stress on the membrane is basically
dicated by the type and magnitude of the
load, surface contact pressure, and thick-
ness of the cover over the membrane. The
estimated stress on the membrane for cover
thickness of 15.2, 30.5, and 45.7 cm (6,
12, and 18 in., respectively) used in the
field tests were 206.7, 158.5, and
110.2 kilopascals, respectively (30, 23,
and 16 psi, respectively). The laboratory
tests were designed to duplicate these
stresses on the membrane at a 15.2-cm
(6-in.) depth by calculating that loading
which would produce the desired stress.
These calculations were made using
Boussinesq equations for stress which is
an accepted geotechnical procedure. A
schematic drawing showing the general ar-
rangement of the layers during testing is
presented in Figure 1.
DESCRIPTION OF EQUIPMENT AND TESTS
General
Three types of test equipment were
selected for testing the membranes. One
consideration was that the equipment be
readily available to most commercial-type
laboratories, or be easily obtained, and
that it be adaptable for testing membrane
materials. The initial tests were con-
ducted using the plate loading equipment
since this method has been used to test
fabrics used as reinforcement in pave-
ments. The gyratory tests were conducted
next since they required a small sample
and were easy to conduct. The pneumatic
tire tests were conducted last and re-
quired the development of test equipment
to simulate the effects of a moving tire
load. The various types of equipment are
described below.
Pneumatic Tire Load
The model load cart and soil test box
are shown in Figure 2. The load cart was
moved for tracking purposes by the force
generated by an air cylinder on a ram that
moved through a maximum travel distance of
61 cm (24 in.). The load wheel was cap-
able of being maneuvered into three dif-
ferent positions in the soil box for traf-
fic test purposes. A total load of
3.5 kilonewtons (800 Ib) was positioned on
BEDDING MATERIAL
V \7 SUBGRADE ~^J
7 ^ ^ ^ V ^
- LINE Ft
Figure 1. Typical test sample.
127
-------
Figure 2. Testing equipment for pneumatic
wheel load tests.
the load cart and the 5.00-5, 4-ply tire
inflated to 220.5 kilopascals (32 psi).
The dry weight and water content were de-
termined for the soil used for each of
four lifts. Each lift was compacted by
hand tamping to produce the density de-
sired. Following this, the membrane was
placed taut over the subgrade and the
3.2-cm- (1.25-in.-) deep collar was C-
clamped in position to confine the sand
cover. The tracking lanes were 10.2 cm
(4 in.) wide by 70 cm (24 in.) long; how-
ever, only the center 35.6 cm (14 in.) of
the lanes were used for comparison pur-
poses (Figure 3). The ends of the track-
ing lanes were not used for evaluating
punctures as the load was shifted in this
area when the cart direction was reversed.
Sand was placed, compacted, and leveled in
the box collar and on top of the membrane
as a cushion (Figure 2). Three tracking
lanes (three tests) were available for
traffic tests each time the soil box was
filled with soil. After three tests were
conducted, the top 15.2 cm (6 in.) of sub-
grade material was reworked and/or re-
placed, and recompacted for additional
traffic tests. In some tests, 2.5 cm
(1 in.) of the subgrade was replaced with
a lean clay to protect the membrane from
puncture.
Gyratory Shear
The gyratory shear test was conducted
in the gyratory testing machine shown in
Figure 5. In this test a vertical load is
applied by a piston to a material sample
contained in a tilted mold. By applying a
rotating load to the mold, shear strain is
induced throughout the sample. The com-
bined action of the vertical stress and
the shear strain is similar to the knead-
ing action of a rolling tire load. During
the test, the applied load (pressure) tilt
or angle of gyration (usually 1 degree)
and number of revolutions are controlled.
Two types of subgrade were used in
these tests. One was the same crushed
limestone used with the model load cart
tests and the other subgrade was a 5.1-cm-
(2-in.-) thick rubber block with a Cali-
fornia bearing ratio (CBR) value of 16.
The rubber subgrade contained steel barbs
at the surface to simulate rocks. Three
of the barbs were 0.6, 1.3, and 0.9 cm
Figure 3
H3 membrane placed over com-
pacted soil collar for sand
clamped in place and tracking
lanes marked.
Figure 4. Soil box for plate loading
tests.
128
-------
Figure 6.
Overall view of Instron,
recorder and soil membrane
test setup.
Figure 5. Gyratory compaction machine.
(0.25, 0.5, and 0.37 in., respectively)
high with a conical angle of 60 degrees,
radius of the apex angle of 0.04 cm
(0.015 in.). Another 0.6-cm- (0.25-in.-)
high barb had a 0.08-cm (0.03-in.) radius
at the apex angle; one had a 0.15-cm
(0.06-in.) radius, and the last one was
a half sphere with a 0.6-cm (0.25-in.)
radius referred to as smooth. The test
membranes with a sand cover were placed
above the subgrades of both type samples.
In tests where a lean clay was used,
2.5 cm (1 in.) of the subgrade was re-
placed with the clay as a bedding to pro-
tect the membrane from puncture.
Plate Loading
The plate load tests were performed
using the Instron equipment (Figure 6).
The load was applied to a 29.2-cm-
(11.5-in.-) diam steel plate, 2.5 cm
(1 in.) thick. A cyclic load of 1.48 kil-
onewtons (3324 Ib) was applied to the
668.8-sq-cm (103.7-sq-in.) plate to
achieve 220.5-kilopascal (32-psi) pressure
to the sand-covered membrane. The soil
box 0.6 by 0.6 by 0.45 m (2 by 2 by
1.5 ft) deep was filled with gravelly
sand and compacted in four layers. Gra-
dation of the gravelly sand is shown in
Figure 7. Each layer of soil was weighed
and compacted in a known volume to produce
the desired density. The water content
was controlled prior to compaction and the
density and water content measured with
the Troxler Nuclear Densitometer after
compaction. The membrane and collar
(6.4 cm (2.5 in.) deep to confine the con-
crete sand) was clamped in place followed
by the addition of the concrete sand which
was tamped and leveled (Figure 4).
The cyclic load was applied at a rate
of 10-12 cycles per minute to a maximum of
129
-------
GRADATION CURVES
Figure 7. Soil gradation curves.
1000 cycles. After a varied number of
cycles had been applied (300, 400, or
500 cycles) the membrane was removed, in-
spected , and punctures marked. The mem-
brane was then replaced as nearly as pos-
sible in its original test position with
the collar and sand cover replaced and the
test continued. The subgrade after these
cyclic loads were applied was not dis-
turbed or replaced. As additional tests
were conducted, only the top 15.2 cm
(6 in.) of the subgrade was scarified and
recompacted.
DESCRIPTION OF MATERIALS
Soils
The soils selected for use in testing
the membranes were as follows:
_a. A brown gravelly sand (SP) (clas-
sified by Unified Soil Classifica-
tion System) having a gradation as
shown on Figure 7 was used as a
cover to protect the membrane dur-
ing tests. Only the material that
passed the 0.9-cm (0.37-in.) sieve
was used for test purposes.
b_. A red gravelly sand (SP) having a
gradation as shown on Figure 7 was
used as a subgrade material.
c. A lean clay (CL) was used as a
bedding material.
ji. A crushed limestone with approxi-
mately 85 percent of the aggregate
between the 1.9-cm (0.75-in.) and
0.9-cm (0.37-in.) screens. The
crushed limestone gradation curve
is shown on Figure 7.
Soil data for the wheel load tests are
shown in Table 1.
Membranes
Three membranes tested during this in-
vestigation as well as a geotextile used
as bedding material were used in the lab-
oratory tests and are described as
follows:
Desig-
nation
M2
M3
Polymer
Polyvinyl
chloride
(PVC)
Nominal
Thickness Thread
mm (mils) Count
0.51 (20) *
Chlorinated 0.76 (30) *
polyethylene
(CPE)
Chorosulfated 0.91 (36) 10x10
polyethylene-
reinforced
(CSPE-R)
Bl Spunbonded 2.06 (81)
continuous
polyester
f ilament,
261.1 g/sq m
(7.7 oz/sq yd)
* Not applicable,
FAILURE CRITERIA AND DATA COLLECTED
Failure Criteria
The criteria used to determine mem-
brane liner failure consisted of the vis-
ual inspection of pinholes in the membrane
liner caused by tests and then the exami-
nation of the pinholes in accordance with
ASTM Test Method D 3083. When light used
in accordance with the ASTM Test Method
was observed to pass through any one of
the pinholes, this was identified as
failure of the membrane liner.
130
-------
TABLE 1. SOIL DATA FOR WHEEL LOAD TESTS
Test
No.
Wl
W2
W3
W4
W5
W6
W7
W8
W9
W10
Wll
W12
W13
W14
W15
Soil Type
Concrete sand
Gravelly sand
Concrete sand
Gravelly sand
Concrete sand
Gravelly sand
Concrete sand
Gravelly sand
Concrete sand
Gravelly sand
Concrete sand
Clay (loess)
Gravelly sand
Concrete sand
Gravelly sand
Concrete sand
Gravelly sand
Concrete sand
Gravelly sand
Concrete sand
Gravelly sand
Concrete sand
Gravelly sand
Concrete sand
Gravelly sand
Concrete sand
Gravelly sand
Concrete sand
Clay (loess)
Gravelly sand
Concrete sand
Clay (loess)
Gravelly sand
Soil Use
Cover
Subgrade
Cover
Subgrade
Cover
Subgrade
Cover
Subgrade
Cover
Subgrade
Cover
Bedding
Subgrade
Cover
Subgrade
Cover
Subgrade
Cover
Subgrade
Cover
Subgrade
Cover
Subgrade
Cover
Subgrade
Cover
Subgrade
Cover
Bedding
Subgrade
Cover
Bedding
Subgrade
Oven Water Content
percent
3.5
5.9
3.2
—
3.7
—
3.5
—
4.9
—
3.7
19.4
—
4.6
—
4.4
—
4.4
—
4.8
—
4.2
—
4.4
—
4.5
—
3.9
18.3
6.0
3.7
19.2
—
(Continued)
Troxler*
Water Content
percent
7.0
—
6.8
6.0
....
6.4
—
6.4
—
6.5
—
6.0
—
6.0
5.9
—
6.2
__.
6.2
—
6.1
—
6.1
—
5.3
—
5.7
Nuclear Test
Wet Density
kilonewtons/cu m
(Ib/cu ft)
20.7 (131.5)
21.0 (133.5)
™ _ —
21.2 (135.0)
—_ ___
20.8 (132.4)
20.8 (132.4)
—
21.0 (134.0)
— — —
20.8 (132.1)
„__
20.8 (132.1)
— — __
20.7 (132.0)
__
20.8 (132.6)
__
20.8 (132.6)
— — —
20.8 (132.7)
— — __
20.8 (132.7)
—
20.8 (132.7)
—
20.9 (133.3)
*Device used to expedite water content and density determinations.
131
-------
TABLE 1. (CONCLUDED)
Test
No.
W16
W17
W18
W19
W20
Soil Type
Concrete sand
Clay (loess)
Gravelly sand
Concrete sand
Limestone
Concrete sand
Clay (loess)
Limestone
Concrete sand
Clay (loess)
Limestone
Clay (loess)
Limestone
Oven Water Content
Soil Use percent
Cover
Bedding
Subgrade
Cover
Subgrade
Cover
Bedding
Subgrade
Cover
Bedding
Subgrade
Bedding
Subgrade
3.5
18.0
5.5
3.5
3.5
23.2
3.5
18.1
18.1
Troxler Nuclear Test
Wet Density
Water Content kilonewtons/cu m
percent (Ib/cu ft)
6.1 21.0 (133.8)
15.7 (100.0)
15.7 (100.0)
15.7 (100.0)
15.7 (100.0)
Data Collected
The number of punctures were recorded
for selected levels of load repetitions.
The general condition of the membrane was
noted such as scuffed or wrinkled, etc.,
and photographs were made to illustrate
these conditions. The number of loading
cycles, passes, revolutions, plus the
total weight applied, and tire pressure
also were recorded. The density and water
content of the subgrade material along
with the CBR of the clay bedding material
were controlled. In the pneumatic tire
model test, the rut depth, soil upheaval,
and cover over the membrane also were re-
corded at regular intervals.
TEST RESULTS
Pneumatic Tire Load
After the tire had traversed back and
forth in the same path for the desired
number of passes, the rut depth, shoulder
upheaval, and the depth of sand over the
membrane were measured. When the tire had
tracked the test specimens, the collar
and sand were removed and the membrane
inspected for punctures. The test number,
membrane designator, passes, rut depth,
membrane depth below rut, number of punc-
tures in the membrane, subgrade deforma-
tion, and soil data are given in Tables 1
and 2.
A minimum of three tests per membrane
was performed on each soil subgrade and
cover material. These test results indi-
cate that punctures occurred to some of
the membranes from the subgrade rocks and
the membrane needed some bedding or pro-
tection from the subgrade. Hence, tests
W14, W15, and W16 were repeats of the W2,
W3, and W4 tests for M2 membrane, except
the top 2.5 cm (1 in.) of the gravelly
sand subgrade was replaced with a lean
clay (loess). The clay was used as a bed-
ding material to prevent the gravelly sub-
grade from puncturing the membrane during
traffic tests.
Test W17 was conducted first on a
crushed limestone base that readily pro-
duced punctures in the membrane. Then the
test was repeated using 2.5 cm (1 in.) of
a lean clay (loess) as the top layer of
the subgrade. The 2.5 cm (1 in.) of lean
clay was used above the limestone in tests
132
-------
TABLE 2. TRAFFIC DATA FROM WHEEL LOAD TESTS
Mem- Test
brane jj0
M2 Wl
M2 W2
M2 W3
M2 W4
M2 W5
M2 W6
M3 W7
y
M3 W8
M3 W9
M4 W10
MA Wll
M4 W12
M3 W13
Number
of
Passes
10
30
100
30
100
300
30
100
300
30
100
300
100
300
500
100
300
500
30
100
300
30
100
300
30
100
300
30
100
300
30
100
300
30
100
300
30
100
300
Rut
cm
3
3
3
3
3
2
2
3
2
2
2
3
3
2
3
2
3
2
2
3
2
2
2
2
2
3
2
2
2
2
2
2
2
2
3
1
3
2
—
.18
.81
.18
.18
.96
.39
.84
.33
.69
.99
.69
.18
.18
.99
.48
.54
.18
.84
.99
.18
.06
.69
.84
.39
.39
.48
.69
.84
.54
.21
.54
.54
.84
.54
.18
.57
.18
.54
Depth*
(in.)
(1
(1
(1
(1
(1
(0
(1
(1
(1
(1
(1
(1
(1
(1
(1
(1
(1
(1
(1
(1
(0
(1
(1
(0
(0
(1
(1
(1
(1
(0
(1
(1
(1
(1
(1
(0
(1
(1
—
.25)
.50)
.25)
.25)
.56)
.94)
.12)
.31)
.06)
.18)
.06)
.25)
.25)
.18)
.37)
.00)
.25)
.12)
.18)
.25)
.81)
.06)
.12)
.94)
.94)
.37)
.06)
.12)
.00)
.87)
.00)
.00)
.12)
.00)
.25)
.62)
.25)
.00)
Membrane
Depth Subgrade
Below Rut Deformation
cm (in.) cm (in.)
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
63
46
63
79
63
63
79
63
63
79
46
46
63
63
46
63
46
63
90
46
63
79
63
46
79
63
46
79
46
46
79
63
46
63
63
63
63
46
63
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
(0.
25) —
18) —
25) —
31) —
25) —
25) -
31) —
25) -
25) —
31) -
18) —
18) --
25) -
25) —
18) —
25) —
18) —
25) —
37) —
18) —
25) —
31) —
25) --
18) —
31) —
25) —
18) —
31) —
18) —
18) —
31) —
25) —
18) —
25) -
25) —
25) —
25) —
18) -
25) -
No. of
Punc-
tures Remarks
0
0
4 1 puncture from top
down
0
0
10
0
1
9
0
3
8 3 or 4 holes from top
down
1 Undisturbed subgrade
6 of test W4 used;
2 geotextile under
membrane
0 Top 2.5 cm (1 in.) of
0 subgrade material
0 was in lean clay
0
1
5
0
0
7
0
1
1
0
0
0
0
0
0
0
0
1
0
1
11
(Continued)
*Depth from original surface to bottom of rut.
133
-------
TABLE 2. (CONCLUDED)
Number
Mem- Test of
brane No. Passes
M2
M2
M2
M2
M2
M2
M2
W14 300
500
1000
W15 300
500
1000
W16 300
500
1000
W17 10
30
100
W18 50
150
300
W19 50
150
300
W20 300
500
Rut
cm
3.48
3.66
3.18
2.69
3.96
4.14
3.96
3.81
3.81
2.84
2.54
4.75
4.45
5.87
6.35
3.18
3.18
2.84
—
—
Depth
(in.)
(1.37)
(1.44)
(1.25)
(1.06)
(1.56)
(1.63)
(1.56)
(1.50)
(1.50)
(1.12)
(1.00)
(1.87)
(1.75)
(2.31)
(2.50)
(1.25)
(1.25)
(1.12)
—
—
Membrane
Depth
Below Rut
cm (in.)
0.46
0.63
0.46
0.46
0.46
0.79
0.46
0.46
0.46
1.27
0.63
0.63
1.57
1.42
1.27
0.63
0.63
0.15
0
0
(0.18)
(0.25)
(0.18)
(0.18)
(0.18)
(0.31)
(0.18)
(0.18)
(0.18)
(0.50)
(0.25)
(0.25)
(0.62)
(0.56)
(0.50)
(0.25)
(0.25)
(0.06)
Subgrade
Deformation
on (in.)
1
0
0
0
0
1
1
1
1
1
1
5
7
8
.27
.63
.90
.63
.79
.27
.57
.27
.90
—
.57
.90
.08
.62
.89
—
—
—
—
—
(0
(0
(0
(0
(0
(0
(0
(0
(0
(0
(0
(2
(3
(3
.50)
.25)
.37)
• 25)
.31)
.50)
.62)
.50)
.75)
—
.62)
.75)
.00)
.00)
.50)
—
—
—
—
—
No. of
Punc-
tures
0
2
3
0
0
0
0
0
2
3
19
30
0
0
2
0
0
0
0
0
Remarks
Membrane wrinkled in
ruts
Rocks 0.79 cm
(0.31 in.) , 1.1 cm
(0.44 in.) , and
0.63 cm (0.25 in.)
below loess at 300,
500, and 1000 passes
respectively
Clay too wet
Same as test W19 ex-
cept no concrete
sand cover used
W18 and W19. Where the clay was used, no
punctures were observed at 150 passes;
whereas, in test W17, 36 punctures were
recorded at 100 passes without the clay.
The clay in test W18 had 22.2 percent
water and a CBR of 1.8 that permitted
excessive rutting, and after 300 passes,
two punctures were observed in the M2
membrane. Test W19 was conducted simi-
larly to test W18 except the water content
of the clay was 18.1 and the CBR was 21.
The higher CBR prevented excessive rutting
and no punctures were observed in the M2
membrane after 300 passes as compared to
36 punctures recorded at 100 passes in
test W17 without the clay bedding. Test
W20 was conducted on the same subgrade as
test W19 (not reworked) with no sand cover
on the membrane (the tire was in direct
contact with the M2 membrane), but only
two tracking lanes were used and no punc-
tures were observed after 500 passes.
Test W5 was conducted on the same sub-
grade (not reworked) as test W4. A geo-
textile (material Bl) was placed on the
top of the subgrade to see how effective
it was as a bedding material to prevent
punctures in the membranes. The membrane
was placed directly on top of the geotex-
tile and 3.2 cm (1.25 in.) of protective
sand was placed on top of the membrane.
The geotextile reduced the number of punc-
tures at 100 and 300 passes from 3 to 1
and 8 to 6, respectively. The lean clay
used in test W6 produced the first labora-
tory test results that indicated lean clay
used as bedding placed over coarse sub-
grade composed of angular particles will
reduce or prevent pneumatic-tired traffic
134
-------
from producing punctures to a membrane
liner.
Gyratory Shear
A small simple-to-prepare sample is
required for this machine, and the loading
and operation are relatively easily and
quickly performed. Extensive testing and
previous experience with this laboratory
equipment made it appropriate for use in
this study. The initial membrane test
samples were prepared with a subgrade that
consisted of a well-graded crushed lime-
stone (Figure 7) except the material re-
tained on the 1.9-on (0.75-in.) sieve was
not used. The rough irregularly shaped
limestone was used with the expectation of
producing punctures in the membrane after
a small number of revolutions, as no bed-
ding was used in these initial tests. Re-
sults of these tests are shown on Table 3.
Tests Gl and G5 were duplicates used to
determine reproducibillty of test data.
These tests were found to produce reason-
ably close results as 20 and 17 punctures
were counted after 10 revolutions. The
same test materials were used in test G6
except the top 2.5 cm (1 in.) of the lime-
stone was replaced with lean clay. After
500 revolutions, no punctures were
observed in the M2 membrane specimens;
whereas, 20 and 17 punctures had occurred
previously after 10 revolutions in tests
Gl and G5, respectively.
Tests G2, G3, and G4 were conducted on
the M4 membrane for comparison with test
results on the M2 membrane. After 10, 30,
and 100 revolutions, there were 0, 1, and
0 punctures, respectively, in the M4 mem-
brane. Test G7 was identical to test G6
except there was no sand cover placed over
the membrane (the steel upper head of the
gyratory machine was in direct contact
with the membrane). After 500 revolu-
tions, there were no punctures in the M2
membrane.
Plate Loading
Results of the plate loading tests are
presented in Table 4. Four tests were run
on the M2 membrane, and three each on the
M3 and M4 membranes. The initial test
consisted of 1000 cycles of loading which
resulted in 3 punctures of the M2 membrane
on the SP subgrade. The cover material
and liner were removed and a geotextile
placed on the undisturbed subgrade of
test PI. The liner was placed on top of
the geotextile and covered with 6.4 cm
TABLE 3. GYRATORY TEST RESULTS OF MEMBRANES PLACED ON
CRUSHED LIMESTONE (1.9 CM (0.75 IN.))
Test
Number
Gl
G2
G3
G4
G5
G6
Cover Soil
Concrete
Concrete
Concrete
Concrete
Concrete
Concrete
sand
sand
sand
sand
sand
sand
Revolutions
10
10
30
100
10
500
Membrane
M2
M4
M4
M4
M2
M2
Number
of Holes
20
0
1
0
17
0
Numerous
Numerous
Numerous
Layered
Remarks
abrasions noted
abrasions noted
abrasions noted
sample from bottom to top
G7
None
500
M2
as follows: 5.1 cm (2 in.)
limestone plus 2.5 cm (1 in.)
lean clay at 18 percent mois-
ture, plus M2, plus 5.1 cm
(2 in.) concrete sand at
3.5 percent moisture
Layered samples from bottom to
top as follows: 10.2 cm
(4 in.) limestone, plus 2.5 cm
(1 in.) lean clay at 18 percent
moisture, plus M2
135
-------
TABLE 4. PLATE LOADING TEST DATA
Mumbe r Numbe r
of of
Test Membrane Cycles Holes
Remarks
PI
P2
M2
M2
300
1000
500
1000
One puncture caused by a sharp rock one by a flat rock,
and the other in a generally smooth area appeared to
occur from the top in a downward direction
Geotextile placed over undisturbed subgrade of test PI
that produced 3 holes at 1000 cycles
P3
P4
P5
P6
P7
P8
P9
P10
M2
M2
M3
M3
M3
M4
M4
M4
400
1000
400
1000
300
1000
400
1000
500
1000
400
1000
500
1000
1000
3 Two holes caused by a large flat rock under the membrane.
5 Holes at 400 cycles did not increase in size
7
8 Several holes increased in size at 1000 cycles
0
7
0
3
0
0
0
0
0
0
0
(2.5 in.) of sand. After 1000 cycles of
loading, no punctures appeared in the mem-
brane indicating that the geotextile was
effective in protecting the membrane.
Test PI was duplicated in tests P3 and P4
to provide more than one test point.
These tests produced punctures at 400
cycles indicating that the punctures in
test PI may have occurred shortly after
the 300 cycles at which data was obtained.
Membrane M3 sustained several punc-
tures at 1000 cycles in tests P5 and P6,
but not in P7. There also were no punc-
tures at the intermediate number of cycles
where data was taken. Membrane M4 sus-
tained no punctures in tests P8, P9, and
P10 at 1000 cycles of loading.
In all of these tests, the water con-
tent and dry density of the gravelly sand
subgrade was controlled at an average of
6.8 percent and 19.5 kilonewtons/cu m
(124.1 Ib/cu ft), respectively. The cover
sand had an average water content of 3.9
in all plate load tests.
ANALYSIS OF DATA
Pneumatic Tire Load
As indicated in Table 2, nine tests
were run on the M2 membrane over an SP
136
-------
subgrade. Four tests (W1-W4) were repeti-
tive having similar test conditions and
results. Each of these tests sustained
no punctures at 10 passes of the wheel
load, a minimum number of punctures at
100 passes of the wheel load, and several
punctures at 300 passes of the wheel load.
The number of punctures seemed to be
directly related to the number of rocks
in contact with the membrane. Therefore,
in order to prevent puncture of the mem-
brane, it was necessary to provide a pro-
tective layer of material that would pre-
vent membrane from coming in contact with
the rocks in the subgrade. Two different
methods were tried, one being placement of
a nonwoven geotextile between the membrane
and the SP subgrade, and the other being
the placement of a layer of clay soil be-
tween the membrane and subgrade.
Test W5 shows the test results using
the geotextile as a bedding material above
the subgrade. As can be seen in Table 2,
punctures occurred at all pass levels
where observations were made. However, a
comparison of punctures at the different
pass levels indicates that the geotextile
was effective in reducing the number of
punctures.
Test W6 shows the test results using
2.5 cm (1 in.) of lean clay as a bedding
material above the subgrade. The lean
clay was very satisfactory in that no
punctures occurred in the M2 membrane at
any level where observations were made.
To further demonstrate the effectiveness
of the clay bedding, three more tests were
run at higher pass levels using the M2
membrane. These tests (W14-W16) also
showed no punctures at 300 passes of the
wheel load. However, a few punctures de-
veloped at 500- and 1000-pass levels in
test W14 and at 1000 passes in test W16.
Test W15 produced no punctures at 1000
passes. These tests again show the ef-
fectiveness of a clay bedding for at least
300 wheel load passes.
Since the clay material had been
effective on the gravelly sand subgrade,
it was decided to test it on a very severe
condition using crushed limestone as a
subgrade. Test W17 shows the severity of
the crushed limestone on the membrane in
that 36 punctures were produced at 100
passes. Tests W18-W20 were then run using
the lean clay as a bedding material over
the limestone and below the membrane. In
all three tests, the lean clay prevented
punctures up to 300 passes except for test
W18 which sustained 2 punctures due to the
clay being too soft. Test W20, which had
no cover over the membrane, went up to
500 passes without any punctures. These
tests also demonstrated the effectiveness
of the lean clay in preventing punctures
even under very severe conditions. For
maximum protection of the membrane, the
lean clay should be compacted adequately
to prevent rutting (water content on dry
side of optimum).
The initial tests run on the M2 mem-
brane were repeated using the M3 membrane
and the M4 membrane. These tests con-
sisted of the concrete sand cover over the
membrane and the gravelly sand subgrade
under the membrane. Tests on the M3 mem-
brane produced results similar to the M2
membrane; therefore, additional tests were
not considered necessary using the geo-
textile or the clay bedding. The test re-
sults using the M4 membrane were signifi-
cantly better than tests on the H2 or M3
membrane with only one puncture being pro-
duced in the three tests at 300 passes.
Since this is the stronger of the mem-
branes, it was considered that the bedding
requirements for the M2 also would be
satisfactory for the M4. The pneumatic
tire load test would be a useful test for
determining cover and bedding requirements
using available site soils and candidate
membranes.
Gyratory Tests
Results of the gyratory tests using a
crushed limestone subgrade are shown in
Table 3. This was considered a severe
test for the membrane materials, and as
expected, numerous punctures were produced
in the M2 membrane when placed on the
crushed stone with the concrete sand cover
as shown by tests Gl and G5. The number
of punctures was reduced to zero when the
membrane was protected from the crushed
stone by a 2.5-cm (1-in.) layer of lean
clay as indicated by tests G6 and G7.
Tests on the M4 membrane showed that it
had greater resistance to puncture, as
only one puncture occurred during three
different tests (G2, G3, and G4). Numer-
ous abrasions were noted on the M4, and
indications were additional cycles may
have produced punctures from these
abrasions.
137
-------
Plate Loading Tests
Results of the plate loading tests are
shown in Table A. The plate tests did not
develop failures in the membrane as had
been observed in previous field tests or
as produced by pneumatic tire load and
gyratory tests. The plate loading tests
indicated that the fabric prevented punc-
tures to the membranes, but in field tests
and the model tire tests, the fabric was
not as effective in preventing punctures.
COMPARISON WITH FIELD TESTS
At the present time, only general ob-
servations have been made in comparing the
laboratory test results with field test
results. In the field tests reported in
the Seventh Annual Research Symposium,
punctures in the membranes occurred re-
gardless of the thickness of cover over
the membranes. Punctures probably oc-
curred because the membranes were directly
on top of the granular subgrades and were
in contact with gravel particles. Where
the sand and sandy silt subgrade was
used, the number of punctures was re-
duced and was zero for some membranes.
These results compare favorably with
the results of the pneumatic tire tests
conducted in the laboratory which showed
that separating the membrane from the
granular material by a lean clay will
prevent or reduce punctures in the mem-
brane. In the laboratory tests, the
fabric bedding material prevented or
reduced the number of punctures in all
tests. There also was some indication in
the field tests that the use of a fabric
under the liner may protect the liner from
puncture, although not all field tests
indicated this.
CONCLUSIONS
Based upon the results of this study,
the following conclusions are considered
warranted:
a_. The pneumatic wheel load tests
would be the most useful test for
determining cover and bedding
criteria using available site
soils and candidate membranes.
_b. Liner material must be protected
from the top by a cover material
and from the bottom by a bedding
material.
c^. The cover material placed above a
liner as well as the bedding mate-
rial should be a soil classified
as a clay, silt, or sand and
having a gradation similar to the
clay or concrete sand used herein.
Material should have no particles
larger than 0.9 cm (0.37 in.).
_d. The 2.5 cm (1 in.) of lean clay
bedding material was effective in
preventing puncture of the liner
material by the gravel subgrade.
e_. Use of the geotextile as a bedding
material reduced the number of
punctures and use of a thicker
geotextile may prevent punctures
from occurring.
138
-------
APPLICATIONS OF GEOTEXTILES IN LAND WASTE DISPOSAL SITES
R. C. Horz, V. H. Torrey
U. S. Army Engineer Waterways Experiment Station
Vicksburg, Miss. 39180
ABSTRACT
This paper provides those involved with the land disposal of solid and hazardous
waste with an introduction to geotextiles. It is part of an overall program to provide
design criteria, preferred construction methods and basis for cost comparisons for the
use of geotextiles in land disposal of solid and hazardous waste. A brief history of the
use of geotextiles within the Corps of Engineers is presented and an overview is given of
the types of geotextiles available. The functions of geotextiles are described and spe-
cific applications of these and related products to solid waste disposal sites are
suggested.
BACKGROUND
The use of synthetic fabrics in engi-
neering construction works in the United
States appears to have started in 1958
in Florida where a woven fabric was sub-
stituted for a graded filter blanket in
the construction of shore protection.
Throughout the 1960's, the U. S. Army
Corps of Engineers (CE) used woven fabrics
in increasing amounts, primarily in flood
control levee slope protection systems and
as natural filter material substitutes
wrapped about perforated pipes in trench
drain systems. The increasing usage of
such products by CE Districts stimulated
conduct of the research by Calhoun at the
U. S. Army Engineer Waterways Experiment
Station (WES), published in 1972, directed
at developing design guidelines and CE ac-
ceptance criteria for use of woven fabrics
in CE projects. Calhoun1s work included
the single nonwoven fabric on the market
in the United States at that time (Phil-
lips Petromat), but he did not attempt to
establish criteria for nonwoven materials.
In March 1972, the Office, Chief of Engi-
neers (OCE) issued Civil Works Construc-
tion Guide Specification CE-1310, "Plastic
Filter Cloth," which provided official
guidance to CE Districts for acceptance
of only woven-type fabrics. Subsequent
to the guide specification issuance, how-
ever, the use of nonwoven fabrics in
engineering construction increased dramat-
ically. Consequently, WES was authorized
to develop acceptance criteria for nonwoven
fabrics. This work was incorporated into a
new Civil Works Construction Guide Specifi-
cation, CW-02215, "Plastic Filter Fabric,"
issued in October 1975 and revised in
November 1977. The CE is currently using
fabrics in a variety of applications in-
cluding the lining of trench drains, re-
taining soil beneath riprap slope protec-
tion, wrapping perforated collector pipes,
lining drainage blankets, reinforcing the
foundations under dredge spoil dikes and a
barrier to intrusion of pavement subgrade
soils into base or subbase materials.
For the purpose of this paper, a geo-
textile is defined as any synthetic tex-
tile product that can be or is made to be
incorporated in a civil engineering struc-
ture. In addition, plastic grids and
strip drains incorporating geotextiles as
an integral part of the product are in-
cluded. In addition to the term "geotex-
tile" the terms "filter fabric," "geotech-
nical fabric," and "engineering fabric"
are in common use in the United States.
The term "membrane" is used in Great
Britain to refer both to geotextiles (per-
meable membranes) and to impermeable plas-
tic films either with or without internal
fiber reinforcement. "Geofabric" is some-
times also seen in the literature.
139
-------
There has been a rapid growth in the
literature on geotextiles in the last five
years. Some major sources of information
about geotextiles and their uses are:
a. Construction and Geotechnical
Engineering Using Synthetic Fab-
rics, by Robert Koerner and
Joseph P. Welsh, 1980.
b. Membranes in Ground Engineering,
by P. R. Rankilor, 1981.
c. "Evaluation of Test Methods and
Use Criteria for Geotextile Fab-
rics in Highway Applications,"
Federal Highway Administration
Report No. FHWA/RD-80/021, 1980.
d. International Conference on Use
of Fabrics in Geotechniques,
Ecole Nationale des Fonts et
Chaussees, Paris, 1977.
e. Proceedings, Second International
Conference on Geotextiles, Au-
gust 1-6, Las Vegas, Nev., Vol-
umes I-III, Industrial Fabrics
Association International, 1982.
Despite the rapidly expanding use of geo-
textiles, their use and performance in
applications to solid and liquid waste
disposal sites have gone essentially un-
documented in the open literature.
PURPOSE AND SCOPE
The purpose of this paper is to pre-
sent a general overview of the types of
synthetic fabric products available for
use in controlled sanitary landfill con-
struction or in remedial measures for un-
controlled site cleanup. In addition,
specific geotextile applications are sug-
gested. The final report on this project
will present additional details including
design criteria for various applications
and preferred construction practices.
Therefore, the final report will serve as
a handbook suitable for use in various
waste management activities.
FABRIC TYPES AND CONSTRUCTION
Fabric Materials
Geotextiles are currently being made
from polypropylene, polyester, polyethy-
lene, nylon, polyvinylidene chloride, and
fibreglass. Polypropylene and polyester
are by far the most used of these materi-
als. The physical properties of all these
materials can be varied considerably, de-
pending on the additives used in composi-
tion and on the methods of processing into
filaments and subsequent incorporation in
the finished fabric. The tensile strength
of polyester and polypropylene filaments
can be increased greatly by drawing or
stretching the filaments while they are
cooling from the molten state. The resis-
tance to degradation from ultraviolet
light varies among the polymers, but all
can be made relatively resistant to dete-
rioration by the incorporation of addi-
tives during manufacture.
The long-term durability in actual
applications is not yet established for
geotextiles, although resistance to many
soil environments can be assumed from
tests on the polymers when used in other
applications, such as sheathing for buried
cables, pipes, etc. The commonly used
plastics seem to have considerable resis-
tance to biological and chemical degrada-
tion in ordinary environments. However,
the environment at solid waste landfills
is potentially much more severe than the
typical conditions to which geotextiles
are exposed. In some applications, the
fabric will be subject to contact with
leachate from the fill or from volatile
constituents that may have a deleterious
effect on the fabric polymers or bonding
resins. It is therefore necessary to
consider the potential for chemical and
biological deterioration when planning
long-term application in solid waste
fills.
Fabric Construction
The woven type of construction is
the most expensive, but tends to produce
fabrics with relatively high strengths
and moduli (ratio of tensile stress to
tensile strain) and relatively low elonga-
tions at rupture. The modulus, however,
depends greatly on the orientation. When
woven fabrics are pulled on a bias to the
direction of the fibers, the modulus de-
creases dramatically, although the ulti-
mate breaking strength may increase. Woven
construction also produces fabrics with a
relatively simple pore structure and nar-
row range of pore sizes or openings between
140
-------
fibers. Woven fabrics can be composed of
monofilaments or multifilament yarns. The
monofilament woven fabrics are generally
used for filter and some reinforcement
applications. The multifilament woven
fabrics are usually restricted to rein-
forcement applications because these
fabrics are expensive and have the highest
strengths of all fabrics. A special type
of monofilament is the slit film or ribbon
filament fabric. The fibers for this
fabric are thin and flat, rather than
round or nearly round in cross section,
and are made by cutting sheets of plastic
into narrow strips. These fabrics are
relatively inexpensive among the woven
fabrics and are primarily used for separa-
tion, i.e., preventing the intermixing of
two materials such as aggregate and fine-
grained soil. Another special type of
multifilament fabric is one made from slit
(fibrillated) film filaments.
The knitting process is not widely
used in the making of geotextiles. Only
two fabrics are known to be made by this
process; one is designed for unidirec-
tional soil reinforcement and the other
for surface erosion protection. With
knitted fabrics, the strength properties
can be varied unidirectionally or multi-
directionally. The advantages of the
knitting process are that it is less ex-
pensive than weaving and it is possible to
knit tube shapes.
Nonwoven fabrics comprise those fab-
rics which are formed by some process
other than weaving or knitting. These
fabrics can be made from either continuous
filaments or from staple fibers. In
either case, the fibers are laid on a sup-
porting screen or belt to form a mat and
then bonded using one of the processes
described below. The orientation of the
fibers is usually more or less parallel
with the plane of the fabric, but depend-
ing on the process, the fibers can be ran-
domly oriented in the plane or can be
given preferential orientation. In some
processes, staple fibers can be given an
orientation perpendicular to the plane of
the fabric so that the fiber ends are ex-
posed at the surface of the mats. In the
spunbonding process, filaments are ex-
truded, drawn to increase the polymer ori-
entation, and hence strength, and laid on
the moving belt or screen as continuous
filaments to form the mat which is then
bonded by any one of the processes
described below.
a. Needle punching. Bonding by nee-
dle punching involves pushing
many barbed needles through the
fiber mat normal to the plane of
the fabric. The process causes
the fibers to be mechanically
entangled. The resulting fabric
has the appearance of a felt mat.
The size of the needles deter-
mines the sizes of the largest
openings in the fabric, but the
complex structure of the matted
filaments determines the overall
permeability and pore character-
istics. Needle-punched fabrics
can be formed by needling a sin-
gle mat thickness or by needling
several mat thicknesses together.
b. Heat bonding. In this process
the mat is laid and then the fi-
bers are bonded at some or all of
the points where fibers cross one
another. This can be done by in-
corporating fibers of the same
polymer type but having different
melting points in the mat or by
using heterofilaments, that is,
fibers composed of one type of
polymer on the inside and cov-
ered or sheathed with a polymer
having a lower melting point.
With heterofilament bonding, up
to 100 percent of the fiber
crossover points can be bonded,
the number of bonds being con-
trolled by the percentage of
heterofilaments in the mat.
c. Resin bonding. Resin is intro-
duced into the fiber mat, coating
the fibers and bonding the con-
tacts between fibers.
d. Comb ina t ion bond ing. Sometimes
a combination of bonding tech-
niques is used to facilitate
manufacture or confer desired
properties. Since needle punch-
ing tends to leave the fibers
relatively free to move with re-
spect to each other, heat or
resin bonding can be combined
with it to increase the dimen-
sional stability of the fabric.
Very open woven fabrics are also
sometimes coated to fix filament
positions.
141
-------
Combination fabrics are fabrics which
combine two or more of the fabrication
techniques previously described. The most
common combination fabric is a nonwoven
mat that has been bonded by needle punch-
ing to a woven scrim. The nonwoven mat
may be on one or both sides of the woven
backing.
Special products are lumped under the
general category of geotextiles but are
either not properly called fabrics in the
same context as used above or are second-
ary products which incorporate fabrics.
Some of these are described below.
a. Grids or meshes. Included here
are the heavy construction grids
or meshes of polyethelene used
mostly in soil reinforcement ap-
plications. Some of these are
drawn during manufacture to ori-
ent the molecular structure to
provide excellent tensile
strength. Another product is a
thick, open mat composed of
crimped and tangled monofilament
fibers, heat-bonded at the con-
tacts to form a three-dimensional
net particularly suited for eros-
ion control. These first two
products are also being sand-
wiched between nonwoven fabrics
to create self-contained subsur-
face drainage layers. Another
special product suited for sur-
face erosion control, particu-
larly during establishment of
vegetative cover, is a polypropy-
lene open-knitted netting inter-
laced with strips of paper.
Still another newer product is
formed as a mat of very strong
composite webbing or polyesther
filaments sheathed in
polyethylene.
b. Strip drains. An expanding use
of fabrics is in the construction
of strip drains heretofore pri-
marily used as vertical drains to
accelerate the consolidation of
soft soils. Typically, the strip
drains are inserted into the soil
using special equipment. Several
commercial sources are available
in the United States and Canada
for this technique. The actual
details of fabrication of strip
drains vary slightly from one
manufacturer to another, but
nearly all drains consist of a
continuous channeled plastic
core, 9 cm (3.5 in.) to 10 cm
(4 in.) wide, surrounded by a
fabric or occasionally a heavy
paper envelope. The plastic core
holds the envelope open while the
channels in it provide a conduit
for the movement of fluid out of
the soil. Strip drains are pro-
duced in continuous rolls up to
140 m (450 ft) in length. In the
typical installation as vertical
drains to accelerate consolida-
tion as part of a preload fill
technique, the strip drain roll
is mounted on the specialized in-
sertion equipment much like a
spool of thread on a sewing ma-
chine and the drain is "stitched"
within a mandrel into the soft
soil to the required depth. The
drain is then cut off a foot or
so above the ground surface after
the mandrel is withdrawn and the
insertion machine is moved to the
next locaton to repeat the opera-
tion. Such a procedure permits
economical placement of a large
number of drains over an area at
close spacing (generally 0.9-m
(3-ft) to 1.8-m (6-ft) grid on
centers). After all drains are
in place over the specified con-
struction site area, a layer of
pervious soil (sand) is placed
around and over the protruding
drain ends. The area is then
covered with a sufficient thick-
ness of additional random fill to
somewhat exceed stresses expected
to be imposed by the structure to
be built. After consolidation
(settlement) of the soft soil is
sufficiently complete, the sur-
charge fill is removed, and se-
lect compacted fill placed, if
required, to bring the site to
grade.
c. Coated fabrics. The foregoing
discussion has dealt only with
products which can be classified
as permeable to liquids and
gases. Fabrics are, however,
used extensively as reinforcement
for impermeable synthetic mem-
branes. These membranes have
been used as liners for water,
142
-------
wastewater, and other liquid/
slurry lagoons. These membranes
also have applicaton in the
capping/covering of solid waste
disposal sites, but heretofore
their higher cost has discouraged
their general use. However, sev-
eral less expensive products have
come on the market recently which
can be considered as hybrids be-
tween fabrics and membranes.
These differ from conventional
reinforced membranes in that an
impermeable film of either ethy-
lene vinyl acetate or polyethy-
lene is bonded to one side of a
nonwoven fabric partially pene-
trating the fabric thickness.
Because the bonding systems are
only partially penetrating and as
yet unproven, these materials may
not be suitable for liquid con-
tainment, but they could function
well as water shedding elements
in a cap/cover design. Cer-
tainly, they may offer greater
strength and puncture resistance
than the unreinforced membranes
now most widely used.
SOME SUGGESTED USES OF GEOTEXTILES
FOR LAND WASTE DISPOSAL SITES
Filtration
This is probably the most commonly
employed function of both woven and non-
woven fabrics. The fabric must meet the
basic criteria required of natural graded
filters, i.e., those of stability and per-
meability. To meet stability require-
ments, the fabric opening sizes must be
such that particles of the soil being re-
tained or filtered will not pass through.
The permeability criterion which must be
satisfied concurrently with .that of sta-
bility is that the fabric must also be
capable of freely draining the soil re-
tained.
Subsurface interceptor drains may be
utilized upslope from waste fills to in-
tercept and divert groundwater from enter-
ing the waste and becoming contaminated
or, in some cases, they may have applica-
tion downslope from a waste site to inter-
cept contaminated groundwater for con-
trolled removal and treatment. These two
applications are shown in Figure 1.
Filter fabrics may be placed between the
in situ soil and the pervious trench back-
fill soil to prevent movement of the in
situ soil into the trench while permitting
FABRIC-LINED TRENCH
> GROUND-
\ WA TER
J FLOW
GRAVEL-FILLED TRENCH'
a. Upslope interceptor drain
'IMPERVIOUS MEMBRANE.
WASTE
GROUND -WATER LEVEL
-GRAVEL-FILLED TRENCH
b. Downslope effluent interceptor drain
Figure I. Subsurface interceptor drains
using fabrics.
the free flow of fluid. Perforated col-
lector pipes are utilized near the bottom
of the trench to facilitate removal of
fluid. If the most economical pervious
trench backfill material is sufficiently
fine in gradation such that it would enter
the collector pipe openings, the pipe can
be wrapped with filter fabric to prevent
that intrusion.
The filtration functions for subsur-
face drainage blanket systems are the same
as for.irench drains described above ex-
cept orientation is essentially horizon-
tal. The purpose of the blanket is to
drain fluid from overlying material or to
prevent the rise of fluid from below into
an overlying material.
The filtration of fabrics used in
surface drainage ditch lining applications
is most needed for cover soils which are
finer grained cohesionless sands or silts.
The fabric is used to line the ditch and
rock or gravel is placed upon the fabric
to secure it in place, to shield the
143
-------
fabric from ultraviolet light, and to dis-
sipate the energy of the flowing water.
The sizing of the rock or gravel and the
ditch dimensions are, of course, based on
anticipated flow velocities and quanti-
ties. This application, as shown in Fig-
ure 2, was used successfully at the Wind-
ham, Connecticut, municipal waste landfill
site where the cover soils of cobble,
gravel, and sand mixtures were suffering
erosion along concentrated runoff paths
which had exposed the cap membrane in
several places. The fabric in this ap-
plication does aid in erosion control but
by means of its filtration action pre-
venting cover soil particles from being
moved into and through the open work rock
or gravel above.
stresses will not compress the product
sufficiently to destroy its drainage
capability.
VERTICAL DRAINAGE PANEL-
GROUND-
WATER
FLOW
Figure 3. Upslope drainage trench.
FABRIC
.IMPERVIOUS
MEMBRANE
WASTE
Figure 2.
Drainage
Permanent slope protection
using fabric beneath rock.
The filtration functions described
above are in association with drainage
functions. For applications described
here, the fabric or combination fabric-
grid-net product not only performs filtra-
tion, but also serves as a conduit for
fluid being drained. Thick, needle-
punched nonwoven and combination fabrics
or special products specifically designed
for this purpose are most commonly used.
Figure 3 illustrates the use of
trench drains and blanket drains as a sub-
stitute for backfilled trenches as pre-
viously shown in Figure 1. Drainage blan-
ket use is shown in Figure 4. It is par-
ticularly noted that strip drains offer an
excellent and less expensive alternative
to the use of perforated pipe as collec-
tors within a cover drainage blanket or a
pond underdrain system. In any of these
uses, care must be taken in selecting the
fabric, combination, or special product to
be sure that it can carry the required
discharge and that any anticipated
WASTE
-FABRIC
Figure 4. Fabric drainage layer.
Venting
The most common use of fabric for gas
venting to date has been to underlay a
synthetic membrane liner for a liquid
storage facility. In some cases, organic
matter is present beneath the linings of
these reservoirs. As the organic matter
decays, methane gas is released and, if it
has no way to escape, can cause the liner
membrane to balloon up through the reser-
voir liquid. A fabric installed between
the liner and soil provides an avenue for
the release of these gases. Fabric can be
given a similar use in conjunction with
covering a waste site. A fabric having
good permeability in the plane of the fab-
ric could be placed over the waste but be-
neath the cover soil or membrane and would
provide a conduit for waste-generated
gases. At intervals, a riser pipe could
be installed to release the gas as it col-
lected in the fabric. An alternative de-
sign would incorporate fabric strip drains
to replace the perforated pipe now used in
144
-------
vertical gas collector pipes. Figure 5
shows these applications.
WASTE
a. Gas vent using fabric layer under cover soil
J// U it//1/
IMPERVIOUS
WASTE
• VENT
COVER SOIL
-STRIP DRAIN
b. Gas vent using strip drain
Figure 5. Gas venting using geotextiles.
In some cases it is necessary to in-
tercept gas as it migrates horizontally
from a site. In these cases, trenches
lined with fabric and filled with gravel
could act as gas interceptors similar in
construction to the types of trenches
used to intercept and drain water flows
except that standpipes would lead from the
trenches to dissipate accumulated gas.
Reinforcement
Geotextiles can be used in various
applications where reinforcement or sup-
port is needed. A fabric can be placed
beneath an impermeable synthetic membrane
to act as a cushioning and load distribu-
tion layer to reduce the incidence of
puncturing. This fabric could help pre-
vent excessive deformations and rupture
of the membrane as the fill settled un-
evenly or as cracks formed in the fill
beneath the membrane.
Allied with this application is the
application where fabric is placed over an
impermeable membrane to protect it from
damage during construction operations.
The fabric may also reduce the possibility
of cover soil sliding from the sloping
surface of the cover membrane and exposing
the membrane to sunlight damage or damage
from other causes. The fabric would in
this case be reinforcing the soil cover by
increasing shear resistance between soil
and membrane. A fabric placed over a
sloping membrane cover could prevent cover
soil from sliding off a membrane as slope
angle increased during fill settlement.
This would also permit the protection of
steeper slopes using a synthetic membrane
than is presently possible. Fabrics may
also be used for reinforcement by embed-
ding a fabric or special plastic grid in
the cover soil. This would act as a tens-
ion member and reduce or prevent cracking
of the cover soil caused by uneven settle-
ment or localized surficial sliding of
cover soil on the steeper slopes.
Erosion Control
Geotextiles may be used for erosion
control in temporary or permanent applica-
tions. In one temporary application the
fabric is used to prevent sheet erosion
in newly seeded slopes. A fabric spe-
cially designed for this application is
anchored on a prepared and seeded slope
and holds soil in place until seeds germi-
nate and vegetative cover is established.
Fabrics used for this purpose are designed
to disintegrate after the vegetation has
had time to establish itself.
In other temporary erosion control
applications, fabrics are used in the con-
struction of silt fences and brush bar-
riers. The purpose of these temporary
structures is to reduce the amount of soil
loss from a construction site due to sur-
face erosion in storm runoff. The silt
fence or brush barrier serves to reduce
the velocity of the running water to allow
suspended solids to settle out of suspen-
sion and also to provide a physical fil-
tering action to remove suspended parti-
cles. Figure 6 shows these applications.
Sediment will eventually build up behind
the barrier and additional barriers can
be built atop the accumulated sediment.
Since these barriers are temporary, the
accumulated sediment is usually regraded
and stabilized by vegetation or by other
means after construction is complete.
Silt fences or brush barriers are not usu-
ally used where runoff volumes are large.
145
-------
SLOPING DITCH BOTTOM
TRAPPED SEDIMENT
• FABRIC FENCE BURIED IN
SOIL A T BASE
a. Silt fence
SLOPING DITCH BOTTOM-.
Figure 6.
TRAPPED SEDIMENT
-FABRIC BARRIER BURIED IN
SOIL AT BASE
b. Brush barrier
Temporary erosion control
measures.
Geotextlles are frequently used for
permanent erosion control applications.
In the most common of these applications,
fabric is used beneath a riprap, gravel,
concrete blocks, or gabions to prevent the
loss of finer soil from a slope surface
while also serving the filtering drainage
function of allowing relief of hydrostatic
pressures from within the slope. Fre-
quently, a layer of gravel or crushed
stone is placed between the fabric and the
stone to protect it from damage during
placing of the cover stone and to minimize
exposure to sunlight and other environ-
mental degradation. A geotextile is often
used in areas of concentrated water flow
such as culvert outlets and swales where
the fabric is usually staked in place or
covered with a layer of gravel or stone.
The same approach can be applied in
solid waste cover soils. Here, the fab-
ric would serve to hold the fine-grained
cover soil in place while the gravel would
act to protect the fabric from environ-
mental exposure.
Fabric is also used between a rock
buttress or gabion and soil slope to allow
dissipation of hydrostatic pressure while
preventing soil from sloughing into and
eventually washing through the buttress
stone.
Separation
A fabric can be used to keep two ma-
terials separate. In construction of ex-
pedient roads over soft soils, for ex-
ample, fabric is often placed over the
soft subgrade soil and gravel or crushed
stone pushed onto the fabric. With suf-
ficient thickness, the gravel or crushed
stone provides a stable base for vehicular
traffic. In the separation application,
the ability of the fabric to pass water is
usually of secondary importance and may
not be necessary at all. In the capping
of waste lagoons or areas where mucky
waste must be covered, a fabric may be
rolled out onto the surface of the waste
and covered with soil or aggregate. This
would provide a firm working surface on
which to place the capping layers of
either synthetic membrane or soil. In
such applications, some benefit is de-
rived from the ability of the fabric to
strengthen the cover system. The mecha-
nism by which this is done is not well
understood, but it is believed that fabric
is put in tension by the deforming or rut-
ting of the soil beneath and this allows
tensile load to be taken by the fabric.
A special case of separation is the
application of coated fabrics used as
water barriers or shingling for the cover
of a solid waste site as shown in Fig-
ure 7. In the applications shown, it is
not necessary that seams between the fab-
ric panels be sealed if the cover is de-
signed to eliminate pooling of water.
a. Coated fabric as cover shingling
b. Coated fabric as used in fill interior
drainage system
Figure 7. Use of coated fabrics for pre-
vention of effluent migration.
146
-------
SUMMARY
The use of geotextiles in waste
management facilities has yet to reach its
full potential. Additional research will
have to be performed before these materials
are accepted and used. Techniques will
have to be developed and evaluated to
provide design criteria for the user
community. Work is underway which will
provide this data but in the meantime
geotextiles deserve consideration in the
design, construction, maintenance, and
closure of land disposal facilities.
147
-------
POTENTIAL MECHANISMS AND REMEDIES FOR CLOGGING OF
LEACHATE DRAIN SYSTEMS
Jeffrey M. Bass, Raymond M. Cornish, John R. Ehrenfeld,
Stephen P. Spellenberg and James R. Valentine
Arthur D. Little, Inc.
Cambridge, MA 02140
ABSTRACT
Preliminary results are presented from a study of the potential for clogging of
leachate collection systems and possible preventative and mitigative remedies. Some
important mechanisms for clogging are iron deposition, calcium carbonate incrustation,
formation of biological slimes, and various physical mechanisms. Preliminary results of
a laboratory analysis of a sample of partially cemented pea-gravel from the leachate
collection system in Boone County Field Site (Test Cell #1) indicate iron, silica
(quartz), calcium and magnesium as the primary constituents in the cement. The remainder
of the study will address the reason for the clogging at Test Cell #1, the potential for
clogging of leachate collection systems at existing facilities, the expected effec-
tiveness of preventative and mitigative remedies to clogging, and recommendations for
further research.
INTRODUCTION
Current EPA Permitting Standards for
hazardous waste disposal facilities (40CFR
264), promulgated 26 July, 1982, contain
the following requirements for leachate
collection and removal systems in single
and double—lined waste piles and landfills
(264.251(a)(2), 264.301(a)(2)):
• leachate depth over the liner must
not exceed 30 cm (1 ft);
• a leachate collection and removal
system immediately above the liner
must be designed, constructed,
maintained, and operated to
collect and remove leachate;
• construction materials must be
chemically resistant to the waste
managed and leachate expected to
be generated, and must be of
sufficient strength and thickness
to prevent collapse under expected
loading; and
• the system must be designed and
operated to function without
clogging through the scheduled
closure of the landfill.
In addition, Part 264.303(b)(4) requires
that while in operation a facility "must
be inspected weekly and after storms to
detect evidence of... the presence of
leachate in and proper functioning of
leachate collection and removal systems,
where present."
To assume safe and legal operations
at hazardous waste landfills and waste-
piles, therefore, leachate collection
systems must maintain flow capacity over
the expected life and closure of the
facility. Little is known, however, about
the potential for failure of these sys-
tems. Experience in the related field of
agricultural drainage systems indicates
that the service life of such drains is
rather short; repairs are expected to be
required after about twelve years of use
(1). A recent study of 18 leachate
collection systems found that 17% had
148
-------
known problems with clogging. However, it
was also found that most operators assumed
that as long as leachate was being col-
lected, the system was functioning proper-
ly. In many cases, leachate depths were
not even monitored (1). Since failure of
the leachate collection system can be
defined as a leachate buildup on the liner
exceeding 30 cm in depth, many operators
would not know if failure had occurred.
It is the responsibility of the
Regional Administrator in permitting a
facility to specify design and operation
conditions to ensure that the leachate
depth over the liner does not exceed the
30 cm limit. Estimates of the potential
for clogging of leachate collection and
removal systems and means to reduce that
potential through better design or through
in-place remedial techniques are therefore
needed.
OBJECTIVE AND APPROACH
The objective of this project was to
aid RCRA personnel in determining whether
or not current design requirements for
leachate collection and removal systems
are adequate. The emphasis was to develop
a basis for estimating the potential for
drain c]ogging at hazardous waste disposal
facilities and determine a set of prac-
tical preventive and mitigative remedies.
To accomplish this, information was
collected from three basic areas. First,
the available literature on clogging
mechanisms and drain systems was reviewed.
This included a limited number of primary
sources on leachate collection system
design and clogging, and sources from
related areas such as agricultural drain-
age and septic system maintenance. In
addition, available information on leach-
ate collection at existing municipal and
hazardous waste landfills, including
leachate characteristics was collected.
Second, a cemented sample of pea gravel
from the drainage envelope at the EPA test
landfill in Boone County, KY was analyzed
to determine the character and mechanism
of the cementation. The results of this
analysis are discussed below. Finally,
original data was collected to determine
current experience with leachate collec-
tion systems and to estimate the potential
for clogging of existing drain systems and
the effectiveness of proposed remedies.
At the time this paper was written,
much of the literature review and a
preliminary analysis of the cemented
gravel sample had been completed. Some of
the important preliminary results are
presented below.
CLOGGING MECHANISMS
Clogging mechanisms can be defined as
the occurrences or natural processes which
inhibit the flow of leachate to or through
the leachate collection system. As stated
above, a system is considered to be
clogged if the leachate depth over the
liner exceeds 30 cm. Some important
clogging mechanisms, as determined by a
recent EPA study of clogging in leachate
collection systems (1) are:
• calcium carbonate incrustation;
• iron deposition;
• formation of biological slimes;
and
• physical mechanisms.
Calcium carbonate (CaCO_) incrusta-
tion occurs by a mechanism similar to that
seen in the natural formation of stalac-
tites. When a solution containing appre-
ciable concentrations of calcium and
bicarbonate ions is subjected to a de-
crease in pressure (or in partial pressure
of carbon dioxide [P__ ] in equilibrium
with heat solution) then the bicarbonate
equilibria (2HC03^=fcC02 + C03 +
H20) will favor release of CO,, and
formation of carbonate ions. These
carbonate ions combine with calcium ions
(Ca+2) to form a CaCO- precipitate once
the CaCO. solubility product is exceed-
ed. This precipitate can build up on the
collection pipe or cause cementation of
the surrounding drain envelope.
The potential for CaCO- incrusta-
tion is primarily a function of:
• total or bicarbonate alkalinity;
• calcium hardness;
• concentration of bicarbonate and
calcium ions in relation to the
solubility of CaCo ;
149
-------
• pH; and
• changes in pressure or partial
pressure of CO,,.
Iron deposition can occur from a
number of complex processes involving
either Fe(II), Fe(III) or both oxidation
states. Important types of iron deposits
include ochre (a general term for precipi-
tates of iron, including Fe(OH),), iron
sulfide (FeS), and various colloidal
complexes. Factors which influence iron
deposition of ochre and sulfide are given
in Table 1. The presence of various
complexing agents complicate the formation
of iron deposits. _ Soluble complexing
agents (e.g., Cl , Br~, F , CN~,
I ) tend to keep iron species in solu-
tion, thereby reducing their availability
for the iron deposition processes.
Alternatively, some organic complexing
agents (e.g., tannins, humic acids, and
aromatic hydroxyl compounds such as
phenols) may form colloidal iron complexes
in water. These would act as sites for
the accumulation of ferric (Fe )
compounds and subsequent trapping of
particulate matter (1). Iron deposits can
restrict leachate flow by clogging the
inside of pipes or causing cementation or
clogging of the materials surrounding the
pipes (2) .
The formation of biological slimes
can occur when slime-producing bacteria or
organisms are present under favorable
conditions. In general, the formation of
biological slimes is dependent on the
presence of bacteria together with the
appropriate nutrients, growth conditions
and energy sources.
In particular, Vitreoscilla, a
filamentous slime forming organism, and
Pseudomonas, a common soil bacteria, are
known clogging agents when iron is not
present. Enterobacter is also known to
contribute to the clogging of the area
abutting drain envelopes (1). Studies of
the biological clogging of sand (3,4)
indicate that the following factors are
thought to contribute to biological
clogging:
• carbon to nitrogen ratio in the
leachate;
• rate of nutrient supply;
TABLE 1
FACTORS CONTRIBUTING TO IRON DEPOSITION
Factors
pH
redox. potential
temperature
Fe +2 concentration
total sul fides
soil type
Ochre
74-95
"10+20 mV
20 - 30°C
-
organic - favorable
silt and clay loams —
unfavorable
FeS
4-8
<0 mV
20 - 40°C
reducing
>2 ppm
• concentration of the slime mater-
ials (polyuronides) ;
• temperature; and
• soil moisture.
Physical mechanisms which clog
leachate collection systems include pipe
collapse due to loading, hydraulic removal
of fines in the soil (scouring or piping)
surrounding the drain envelope, deterior-
ation of the pipe material, and deposition
of suspended solids. The principle
factors which contribute to physical
clogging are:
• soil grain size;
• leachate characteristics
- production rate
- volume
- head
- velocity;
• pipe characteristics
- size
- material;
design of joints,
envelope, filter;
outlets,
• pipe placement, depth and loading;
• depth to watertable; and
• maintenance and inspection.
150
-------
12
i C / > „ r
18
; ~- a% tiop»
r,;^;/.
V-'vVj..."-..
(o)
As-built Construction Details
Cross section at observation bulkhead
^ ^ /—Refuse penetration into ""i
(b)
Cross section as exposed
8 feet from observation bulkhead
FIGURE 1. LEACHATE COLLECTION SYSTEM, Test Cell #1 (Ref. 5)
151
-------
FIGURE 2. Cemented Mass of Pea Stone
PRELIMINARY CHARACTERIZATION OF A
CEMENTED DRAIN SECTION
INTRODUCTION AND APPROACH
When Test Cell #1 of the Boone County
Field Site was dismantled in September,
1980 after nine years of testing, a
section of partially cemented gravel was
discovered in the drain envelope extending
from 6.5 to 13.5 feet from the collection
sump (bulkhead). Cross-sections of the
Test Cell as constructed and as exposed
are given in Figure 1. The discovery of
the cemented section was significant
because the Test Cell, along with four
others, was constructed to provide a
better understanding of the processes and
related environmental effects that occur
in sanitary landfills (5) . It is there-
fore important to determine whether the
causes of cementation are unique to the
condition at the small-scale test land-
fill, or whether they are common to
sanitary and hazardous waste disposal
landfills in general.
The gravel sample consisted of a
small amount of loose, rounded pea-stone
plus two or three large masses of similar
stones firmly held at contact points
between the stones by a thin layer of
red-brown cement. The largest of these
cemented aggregates was a flat, disc
shaped mass approximately 12 cm across and
5 cm in thickness. It appeared to be
graded or classified with large, individ-
ual stones of 1 to 2 cm diameter on one
side, and smaller stones of 0.5 to 1.0 cm
on the other. A photograph of the two
larger masses of material is shown in
Figure 2.
Two approaches were used in the
initial analysis of the gravel sample.
The first involved a physical analysis of
the cement material itself, including
scanning electron microscopy, optical
microscopy, and x-ray diffraction and
fluorescence analysis. The second in-
volved a more general chemical analysis of
the mass to determine the primary chemical
constituents in the cemented sample.
The results of these microscopic
studies lead to the preliminary conclusion
that the cementing agent for the small
stones is likely a co-precipitated mixture
of calcium carbonate and an insoluble iron
hydroxide. This combination, along with
the silica, constitutes the principal
components of the cement. However, aside
from a few isolated crystals, calcium
carbonate is not observed microscopically
152
-------
as a separate phase and is not present in
the x-ray diffraction pattern. Since the
calcium is present in a substantial
quantity, if it were present as pure
calcium carbonate we would expect to
detect the crystalline form (either
calcite or aragonite) by this technique.
This ambiguity suggests that the calcium
and iron may well have co-precipitated in
an amorphous form and, as such, are not
"seen" by x-ray diffraction.
RESULTS OF CHEMICAL ANALYSIS
In a preliminary test it was deter-
mined that dilute hydrochloric acid would
not affect the gravel but would solubilize
the material holding it together. To
determine the composition of the cemented
gravel material, a portion of the sample
was treated with a known excess of acid
and separated into acid-soluble and
insoluble fractions. The soluble fraction
was subjected to qualitative emission
spectrographic analysis to identify the
principal metal species which were then
quantified by emission spectrometry. The
insolubles were separated into size
fractions and weighed.
A known weight (100 grams) of the
cemented gravel was treated with a known
amount of standardized hydrochloric acid
and boiled for thirty minutes to effect
complete dissolution of the cement mater-
ial and to drive off the carbon dioxide
formed. The insolubles were separated by
decantation and filtration, dried, sieve-
sized and weighed.
A portion of the filtered solution
was analyzed for residual acidity by pH
titration (to pH 5.0) using standardized
sodium hydroxide solution, to measure the
alkali content (acid consumption) of the
dissolved material.
RESULTS OF PHYSICAL ANALYSIS
The raicrostructure of the cemented
material as seen by the scanning electron
microscope (SEM) is shown in Figure 3.
Accompanying the SEM micrograph in Figure
3 is the spectrum of elements detected by
the energy dispersive x-ray analysis
system (EDS) attached to the SEM. EDS
analysis of the cement material from
various points on the sample indicated
calcium and iron as the principal elements
present. (Elements lighter than sodium
are not detected). In addition, in the
sample shown in Figure 3, a trace amount
of manganese was detected.
A larger amount of the red cement was
isolated for analysis of the crystalline
content by x-ray diffraction and elemental
analysis by x-ray fluorescence. X-ray
diffraction indicated silica (quartz) as
the only crystalline material present.
X-ray fluoresence, which detects the
presence of elements heavier than aluminum
in atomic number, indicated a relatively
very strong signal for calcium and iron, a
weak signal for phosphorous, zinc, sulphur
and silica, and a very weak signal for
manganese and potasium.
Using optical microscopy under the
petrographic (polarizing) microscope the
reddish cement was observed to consist of
three major constituents:
• silica;
• a colorless, apparently crystal-
line phase with calcium carbonate
present; and
• an as yet unidentified phase
consisting of an aggregated
cluster of small red particals
comlngled with colorless crystal-
line particles of similar size
range. The unidentified aggre-
gates reacted vigorously with
dilute acid, evolving large
volumes of gas and leaving opaque
red particles without the comin-
gled crystalline phase.
The elemental composition of the
major constituents in the cemented layer
was determined by atomic emission spectro-
scopy. An emission spectrum was obtained
using a Spectrospan III Direct Current
Argon Plasma Optical Emission Spectro-
photometer for qualitative analysis. This
revealed Ca, Fe, Mg, and P as major
constituents in solution, and Mn, Cr, Na,
Ba, Si, Cu, and Sr present at lower
levels. The major constituents, along
with Mn, were quantified on Spectrospan
III Spectrometer using the method of
standard additions for each element
analyzed. The results of these analyses
are shown in Table 2.
153
-------
Ca Mn Fe
FIGURE 3. Red Cement Microstructure with Trace of
Manganese in EDS Spectra
154
-------
TABLE 2
RESETS OF CHEMICAL ANALYSES OF CEMENTED MATERIAL
Weight sample taken
Weight insolubles reniaing
Apparent sample dissolved
Alkalinity (acid consurapti
- 100 468
= 96.958
* 3.530 c
olved material =
60 3 mill equivalent
Elemental composition of dissolved material
Element Weight Found
Ca
Fe
Mg
Mn
P
0 851 g
0 551 g
0.103 g
0.020 g
0.240 g
Size distribution of insolubles
Size Weight Found
94.015 g
>2.0 millimeters
(#10 Sieve)
<2 0 mm, >0.84 mm
(through #10, on #20)
<0.84 mm, > 0.25 mm
(through f'20, on #60)
<0.25 mm
(through ^60)
0.640 g
0 933
PRELIMINARY FINDINGS
The chemical and physical analysis
data are being evaluated currently, and
thus the findings at this point must be
regarded as preliminary.
The "cement" is principally a cal-
cium-iron-magnesium product containing
significant proportions of carbonate (gas
evolution) and phosphate. In addition, a
relatively large proportion of fine silica
appears to be dispersed in the cement.
The lack of significant x-ray dif-
fraction patterns suggests that the cement
is an amorphous material rather than
composed of discrete crystalline phases.
While little can be said yet about the
clogging mechanism(s) at work here,
carbonate incrustation is likely to have
contributed. The role of iron is not yet
clear - it may have been an active agent
in precipitate formation, or may only be
present as discrete oxide (red, Fe_0-?)
particles which have been carried along.
Some additional work on chemical and
physical characterization is planned with
particular emphasis on elucidating the
role of the fine silica particulate and
the distribution and role of iron within
the mixture. In addition, limited tests
for the presence and distribution of
organic materials are planned.
CONCLUSION
A number of the most important
questions in the study remained to be
answered when the paper was written,
including:
• Why did the cementation occur in a
localized section of drain envel-
ope at Test Cell #1?
• Based on typical conditions and
operating practices, is a similar
or other form of clogging expected
to occur at existing sanitary
landfills or hazardous waste
disposal facilities?
• What remedial measures will be
effective in correcting expected
clogging problems?
• What further research is necessary
in this area?
These questions will be addressed based on
analysis of theoretical clogging mechan-
isms, the actual clogging mechanism(s) at
Test Cell #1, and further laboratory
analysis and literature review. Final
results will be presented at the 1983
Research Symposium and will be available
in a published final report.
REFERENCES
Young, et al., Clogging of Leachate
Collection Systems Used in Hazardous
Waste Land Disposal Facilities. GCA
Corporation, prepared for USEPA,
Hazardous Waste Management Division,
Office of Solid Waste, Draft White
Paper, June 1982.
Ford, Harry W. , Characteristics of
Slime and Ochre in Drainage and
Irrigation Systems. In: Transac-
tions of the ASAE Vol 22, No. 5, pp
1093-6, American Society of Agricul-
tural Engineers, St. Joseph, MI 1979.
Avnimelech, Y and Z Nevo, Biological
Clogging of Sands. In: Soil Sci-
ence, Vol 98, pp 222-26, 1964.
155
-------
4. Kristianson, Rolv, Sand-Filter
Trenches for Purification of Septic
Tank Effluent: 1. The Clogging
Mechanism and Soil Physical Environ-
ment. In: Journal of Environmental
Quality, Vol. 10, No. 3, pp 353-7,
1981.
5. Wigh, Richard J., Boone County Field
Site: Final Report (Vol I) and
Appendices (Vol II). Municipal
Environmental Research Laboratory,
Office of Research and Development,
USEPA, Cincinnati, OH, undated draft.
156
-------
ANALYSIS AND FINGERPRINTING OF UNEXPOSED AND EXPOSED
POLYMERIC MEMBRANE LINERS
Henry E. Haxo, Jr.
Matrecon, Inc.
Oakland, California 94623
ABSTRACT
A plan is presented for analyzing polymeric membrane liners for waste storage and
disposal impoundments before and after laboratory or fjilot-scale exposure and field ser-
vice. These analyses can be used to fingerprint a material and to follow the changes
that take place in a polymeric membrane liner during exposure to waste. They can also be
used to determine components of a waste liquid that are absorbed and are aggressive to
polymeric liners.
This analysis plan includes determination of volatiles, extractables, specific
gravity, ash and crystallinity of polymeric liners. The plan also includes gas chroma-
tography and infrared analysis of the extractables (and possibly of the organic volatiles)
and thermogravimetric analysis of the liner. Also suggested is the use of pyrolysis gas
chromatography, which can be performed directly on unexposed and exposed liner materials.
Typical analytical results for unexposed and exposed liners are presented.
INTRODUCTION
Because of the wide range of composi-
tions and constructions of flexible poly-
meric membrane liners that are currently
available and being developed for lining
waste impoundments, analysis and finger-
printing of the membranes is needed for a
number of purposes. For example, a liner
manufacturer needs to test his sheeting as
new polymers are used and as new compounds
and constructions are developed. He also
needs tests to control the composition of
the liner being manufactured.
The analysis of
liner at the time of
for three purposes:
characterizing and
a polymeric membrane
placement can be used
first, as a means of
identifying the specific
sheeting; second, as a baseline for moni-
toring the effects of exposure on the liner;
and third, to assess the aggresive ingredi-
ents in the waste liquid to determine
chemical compatibility.
During exposure to waste liquids,
polymeric liners may change in composi-
tion in various ways that may affect their
performance and result in actual failures.
Polymeric materials may absorb water,
organic solvents and chemicals, organo-
metallic materials, and possibly some inor-
ganics if the liners become highly swollen.
On the other hand, the extractable materials
in the original liner compound may be
leacneu out and result in stiffening and
even brittleness on the part of the liner
membrane. The solid constituents of a
polymeric compound (which include carbon
black, inorganic fillers, and some of the
curing agents) will be retained in the liner
compound, as will the polymer of which the
liner is made (particularly if the polymer
is crosslinked). If organic materials are
similar to the liner in solubility and
hydrogen bonding characteristics, the liner
may swell excessively. Some thermoplastic
157
-------
lining materials may even dissolve, when in
contact with some solvents.
The objective of this paper is to
present an analytical methodology that can
be used to fingerprint and identify liner
materials and to give a baseline for
assessing the changes in composition of
these materials when they are under test or
in service. Also presented are the analy-
tical procedures for testing exposed liner
materials to determine these changes. Data
on representative liner materials, before
and after waste exposure, are presented.
POLYMERS USED IN MEMBRANE LINER MANUFACTURE
Polymers used in the manufacture of
lining materials include rubbers and
plastics differing in polarity, chemical
resistance, basic composition, etc.,
and can be classified into four types:
- Rubbers (elastomers) that are gen-
erally crosslinked (vulcanized),
- Plastics that are generally unvul-
canized (such as PVC),
- Plastics that have a relatively
high crystalline content (such as
the polyolefins), and
- Thermoplastic elastomers that do not
need to be vulcanized.
Table 1 lists the various types of
polymers that are used and indicates whether
they are used in vulcanized or nonvulcanized
form and whether they are reinforced with
fabric. The polymeric materials most fre-
quently used in liners are polyvinyl chlo-
ride (PVC), chlorosulfonated polyethylene
(CSPE), chlorinated polyethylene (CPE),
butyl rubber (IIR), ethylene propylene
rubber (EPDM), neoprene (CR), and high-den-
sity polyethylene (HOPE). The thickness of
polymeric membranes for liners ranges from
20 to 120 mils, with most in the 20 to 60-
mi1 range.
TABLE 1.POLYMERIC MATERIALS USED IN LINERS
Use in liners
Thermo-
Fabric
reinforcement
Polymer
Butyl rubber
Chlorinated polyethylene
Chlorosulfonated polyethylene
Elasticized polyolefin
plastic
No
Yes
Yes
Yes
Vulcanized
Yes
Yes
Yes
No
With
Yes
Yes
Yes
No
W/0
Yes
Yes
Yes
Yes
(partially crystalline)
Elasticized polyvinyl chloride Yes
Epichlorohydrin rubber Yes
Ethylene propylene rubber Yes
Neoprene (chloroprene rubber) No
Nitrile rubber Yes
Polyethylene (partially crystal-
line) Yes
Polyvinyl chloride Yes
No
Yes
Yes
Yes
No
No
Yes
Yes
Yes
Yes
Yes
No
Yes
No
Yes
Yes
Yes
Yes
Yes
158
-------
Most polymeric lining materials are
based on single polymers, but blends of two
or more polymers (e.g., plastic-rubber
alloys) are being developed and used in
liners. Consequently, it is difficult to
make generic classifications based on
individual polymers in the liners, even
though one polymer may predominate. Blend-
ing of polymers introduces the long-range
possibility of the need for performance
specifications, but long-term liner per-
formance in the field cannot presently be
completely defined by current laboratory
tests.
The basic compositions of the different
types of compounds are shown in Table 2.
The crosslinked rubber compositions are
usually the most complex because they con-
tain a crosslinking system that requires
more ingredients (e.g., the sulfur system).
Thermoplastics, except for CSPE compounds,
contain no curatives. Although supplied as
thermoplastic membranes, CSPE liners con-
tain inorganic crosslinking chemicals that
allow the compound to crosslink slowly over
time during service. Crystalline materials
have the simplest composition and generally
consist of polymer, a small amount of carbon
black for ultra-violet protection, and
antidegradants.
Of the various liner components (except
for the polymer), the following are poten-
tial extractables:
Small amounts in the original polymer
(i.e., stabilizers and antidegrad-
ants)
Oils and plasticizers
Antidegradants added to the compound.
Organic constituents of the sulfur
crosslinking system (e.g., vulcaniza-
tion accelerators and activators).
These ingredients are extracted
termination of extractables.
in the de-
Most of the polymeric membrane liners
currently manufactured are based on unvul-
canized or uncrosslinked compounds and thus
are thermoplastic. Even if the polymer in
the vulcanized form is more chemically
resistant (such as CPE and CSPE), it is
generally supplied unvulcanized because it
is easier to obtain reliable seams and to
make repairs in the field. Thermoplastic
polymers can be heat-sealed or seamed with
a solvent or bodied solvent (a solvent con-
taining dissolved polymer to increase the
viscosity and reduce the rate of evapora-
tion). Crystalline sheetings, which are
also thermoplastic, are seamed by thermal
welding or fusion methods. Information on
individual polymers and liners is presented
in the EPA Technical Resource Document on
liners (Matrecon, 1982).
TABLE 2.BASIC COMPOSITIONS OF POLYMERIC MEMBRANE LINER COMPOUNDS
Type of polymeric compound
Component
Crosslinked Thermoplastic Crystalline
Polymer or blends (alloys)
Oil or plasticizer
Fillers:
Carbon black
Inorganics
Antidegradants
Crosslinking system:
Inorganic system
Sulfur system
100
5-40
5-40
5-40
1-2
5-9
5-9
100
5-40
5-40
5-40
1-2
(*)
100
0-10
2-5
1
*An inorganic curing system that crosslinks over time is incorporated
in CSPE liner compounds.
159
-------
ANALYSIS AND FINGERPRINTING OF UNEXPOSED
POLYMERIC LINING MATERIALS
Analyses of liners run before and after
exposure to different environments include:
- Volatiles
- Ash
- Extractables
- Gas chromatography
- Thermogravimetric analysis
- Differential scanning calorimetry
if liner material is crystalline
- Specific gravity
The following subsections describe the
tests performed on unexposed polymeric
linings.
Volatiles
The volatile fraction is represented by
the weight lost by an unexposed specimen of
the liner on heating in a circulating air
oven at 105°C for 2 hr. Polymeric compo-
sitions generally contain a small amount of
volatiles (<1.0%), usually moisture. The
recommended specimen is a disk cut from the
membrane.
The volatiles test can also be used to
determine the direction of the grain that
has been introduced in the membrane during
manufacture. By identifying the orientation
of the 2-in. disk specimen with respect to
the sheeting at the time the specimen was
died out, the grain direction can be iden-
tified. The grain direction must be known
so that tensile and tear properties can be
determined in machine (grain) and transverse
directions. Upon heating in the oven at
105°C, sheeting with a grain will shrink
more in the grain direction than in the
transverse direction (Figure 1).
Volatiles need to be removed before
determining ash, extractables, and spec-
ific gravity. Ash and extractables are
reported on a dry basis (db). Volatiles
contents of representative membrane liners
are presented in Table 3. Monomeric plas-
ticizers that are generally used in PVC
compositions have a limited volatility and
can slowly volatilize at 105°C. Thus the
air oven test must be limited to 2 hr.
As received
After air oven heating
2hr. at 105°C
Figure 1.Machine direction determinations.
A standard test for the volatility of
plasticizers in PVC compounds is performed
in accordance with ASTM D1203. In this
test, activated charcoal is used to absorb
volatilized plasticizer.
Ash
The ash content of a liner material is
the inorganic fraction that remains after a
devolatilized sample is thoroughly burned at
550°±2b"C. The ash consists of (1) the
inorganic materials that have been used as
fillers and curatives in the polymeric
coating compound, and (2) ash residues in
tne polymer. Different liner manufacturers
formulate their compounds differently, and
determining the ash content can be a way to
"fingerprint" a polymeric liner compound.
The residue obtained by ashing can be
retained for other analyses (such as metals
content) needed for further identification
and for providing a reference point to
determine trace metals that may have been
absorbed by the liner. The test method
160
-------
described in ASTM D297, Section 34, is
generally followed in performing this
analysis. Ash contents of representative
membrane liners are presented in Table 3.
Extractables
The extractable content of a polymeric
sheeting is the fraction of the compound
that can be extracted from a devolatilized
sample of the liner with a solvent that
neither decomposes nor dissolves the poly-
mer. Extractables consist of plasticizers,
oils, or other solvent-soluble constituents
that impart or help maintain specific
properties such as flexibility and proces-
sability. A measurement of extractable
content and an analytical study of the
extract can be used as part of the finger-
printing of a sheeting.
During exposure to a waste, the ex-
tractable constituents in a liner may be
removed and result in property changes. At
the same time during exposure, the liner
might absorb nonvolatilizable constituents
from a waste. Measuring the extractable
content of unexposed lining materials is
therefore useful for monitoring the effects
of exposure. The extract and the extracted
liner obtained by this procedure can be used
for further analytical testing (e.g., gas
chromatography, infrared spectroscopy, ash,
thermogravimetry, etc.) and fingerprinting
of the liner.
The procedure for extraction generally
follows ASTM D3421, "Extraction and Analysis
of Plasticizer Mixtures from Vinyl Chloride
Plastics". Also see ASTM D297, "Rubber
Products-Chemical Analysis", paragraphs
16-18.
Because of the wide differences among
the polymers used in liner manufacture, a
variety of extracting media must be used.
TABLE 3. ANALYSIS OF UNEXPOSED POLYMERIC MEMBRANE LINERS*
Base polymer,
Polymer specific gravity
Butyl rubber
Chlorinated polyethylene
Chlorosulfonated poly-
ethylene
Elasticized polyolefin
Epichlorohydrin rubber
Ethylene propylene rubber
Neoprene
Polybutylene
Polyester elastomer
Polyethylene (low density)
Polyethylene (high density)
Polyethylene (high density)
alloy
Polyvinyl chloride
0.92
1.16-1.26
1.11
0.92
1.27-1.36
0.86
1.25
0.91
1.17-1.25
0.92
0.96
0.95
1.40
Compound
Specific
gravity
1.206
1.17b
1.360
1.362
1.377
1.433
1.343
0.938
1.490
1.173
1.122
1.199
1.503
1.480
1.390
0.915
1.236
0.921
0.961
0.949
1.275
1.264
1.231
1.280
1.308
Volatiles,
%
0.45
0.46
0.10
0.00
0.05
0.84
0.51
0.15
0.63
0.38
0.50
0.31
0.76
0.19
0.37
0.12
0.26
0.18
0.12
0.11
0.11
0.09
0.05
0.31
0.03
Extractables
%
10.96
11.79
7.47
9.13
1.49
3.77
5.50
7.27
23.41
31.77
18.16
10.15
13.43
21.46
2.74
2.07
0.49
2.09
33.90
37.25
38.91
35.86
25.17
, Asn,
%
5.25
4.28
14.40
12.56
17.37
33.95
3.28
0.90
4.49
6.78
5.42
0.32
12.98
13.43
4.67
0.08
0.38
0.13
0.46
0.32
6.20
5.81
3.65
6.94
5.67
*Source of some of the data
"""Multiple figures represent
Haxo et al. (1982).
materials from different manufacturers.
161
-------
Table 4 lists the recommended solvents for
extraction of membrane liners of each
polymer type.
Typical values for the extractables in
polymeric membranes are given in Table 3.
Gas Chromatography
Gas chromatography can be used to find
the level of the plasticizer (e.g., diethyl-
hexyl phthalate (DEHP), a dioctyl phthalate)
compounded into a PVC liner material. Fig-
ure 2 shows the quantification of DEHP in
the solvent extract of a PVC liner material.
The weight percent of the DEHP in the liner
can then be calculated, assuming the extrac-
tion to be 100% efficient. A typical
procedure is summarized below.
A weighed sample of liner is extracted
with an appropriate solvent. The extract is
evaporated to dryness over a steam bath.
The dry residue is redissolved in solvent
and brought to an accurately known volume.
Following the development of appropriate
chromatographic conditions, injection of
this solution into the instrument will
separate it into chemically pure components
characterized by different retention times.
An injection of a DEHP standard solution
will identify the retention time of the DEHP
component. Comparison of the peak height
X
LU
I
~J
>-
02
O Standards
-f- Sample
PEAK HEIGHT IN CM
Figure 2.Gas chromatograph determination
of the diethylhexyl phthalate
content in an extract of a PVC
membrane. Column: 6'xl/8" 3% 0V
101 on Chromosorb WHP. Tempera-
ture: 200°-300°C at 8°C/min. He
carrier gas: at 30cc/min.
(area) data obtained from the injection of
equal volumes of the extract solution and
from quantitatively prepared standard
solutions allows the interpolation of the
DEHP concentration in the extract solution
(Figure 2).
TABLE 4. SOLVENTS FOR EXTRACTION OF PuLYhERIC MEMBRANES*
Polymer type
Extraction solvent
Butyl rubber (IIR)
Chlorinated polyethylene (CPE)
Chlorosulfonated polyethylene (CSPE)
Elasticized polyolefin
Epichlorhydrin rubber (CO and ECO)
Ethylene propylene rubber (EPDM)
Neoprene
Nitrile rubber (vulcanized)
Nitrile-modified polyvinyl chloride
Polyester elastomer
High-density polyethylene
Polyvinyl chloride (PVC)
(HOPE)
Thermoplastic olefinic elastomer
Methyl ethyl ketone
n-Heptane
Acetone
Methyl ethyl ketone
Methyl ethyl ketone or acetone
Methyl ethyl ketone
Acetone
Acetone
2:1 blend of carbon tetrachlo-
ride and methyl alcohol
Methyl ethyl ketone
Methyl ethyl ketone
2:1 blend of carbon tetrachlo-
ride and methyl alcohol
Methyl ethyl ketone
*Because lining materials can be sheetings based on polymeric aloys mar-
keted under a trade name or under the name of only one of polymers, this
list can only be taken as a guideline for choosing a suitable solvent for
determining the extractables. Once a suitable solvent has been found, it
is important that the same solvent be used for determining the extract-
ables across the range of exposure periods.
162
-------
Pyrolysis gas chromatography
Pyrolysis gas chromatography is an
alternative method for measuring a plas-
ticizer in liner materials. In this tech-
nique, a small, weighed sample of liner is
heated very rapidly to a temperature suf-
ficient to volatilize all of its organic
components. The plasticizer and other
lower-molecular weight organics will be
driven off as chemically unchanged vapors.
The polymer will undergo pyrolysis, or high
temperature decomposition, and will vola-
tilize as lower-molecular-weight organic
compounds. The resulting volatiles may be
separated and quantified by gas chroma-
tography as previously described, and the
plasticizer content of the liner may be
calculated.
This method has the strong advantage of
not requiring extraction of the liner
sample, but it may not be as reliable a
means of quantification because of the very
small sample size and the large number of
components that must be separated by the gas
chromatograph.
Thermogravimetric analysis (TGA)
TGA is a thermal technique for asses-
sing the composition of a material by its
loss in weight on heating at a controlled
rate in an inert or oxidizing atmosphere.
For example, when a material is heated in an
inert atmosphere from room temperature to
600°C at a controlled rate, it will vola-
tilize at different temperatures until only
carbon, char and ash remain. The intro-
duction of oxygen into the system will burn
off the char and carbon black. Thus from
the weight-time curve which can be related
to weight and temperature, the amounts of
volatiles, plasticizer, polymer, carbon
black, and ash can be calculated. In some
cases, thermogravimetric analysis can
replace measurements of the volatiles, ash,
and extractables contents discussed above.
The TGA curve and the derivative of the TGA
curve can thus be used as part of a finger-
print of a polymeric composition. This
technique is described by Reich and Levi
(1971).
A Perkin-Elmer TGS-2 thermogravi-
metric system, consisting of an analyzer
unit, balance control unit, heater control
unit, and first derivative computer, is used
in our laboratory. Temperature control is
supplied by the temperature controller on
the Perkin-Elmer DSC-2 (Differential Scan-
ning Calorimeter). A double side-arm fur-
nace tube was used to allow rapid changing
of the atmosphere from inert (N2) to oxi-
dative (N2/02 mixture). For the oxida-
tive atmosphere, N2 purge is maintained
through the analyzer unit head, and 02
is introduced at the upper side arm where
it mixes with the N2 to burn the carbon
black and any carbonaceous residue that
forms during the pyrolysis of the polymer.
Use of the double side-arm furnace tube
shortens the turnaround time because it
eliminates the need to flush the analyzer
head completely to remove 02 between runs,
as would be necessary if 02 were intro-
duced through the head. A dual pen recorder
Perkin-Elmer Model 56 allows a simultaneous
display of thermocouple temperature in the
furnace and the change in weight of the
specimen or the first derivative of the
change in weight.
An example of the TGA procedure for the
analysis of a polymeric liner is described
below.
A 5-mg specimen of the liner was placed
in the balance pan and weighed in a nitrogen
flow of 40 cc/min. The instrument was
adjusted to give a 100% full-scale deflec-
tion for the weight of the sample, so the
percent of weight change can be read
directly from the chart.
The specimen was heated to 110°C and
held there for 5 min to determine whether
measurable volatiles were present; it was
then heated from 110° to 650°C at a rate of
20°C/min in a nitrogen atmosphere. The
specimen was held at 650°C until no more
weight loss occurred, usually 2 to 3 min,
after which it was cooled to 500°C and 02
was introduced at a rate of 10 cc/min with
an N2 flow rate of 30 cc/min.
Typical thermograms for HOPE and EPDM
appear in Figures 3 and 4, respectively.
Analyses of a variety of polymeric membrane
liners are presented in Table 5.
Differential Scanning Ca^orimetry (DSC)
DSC is a thermal technique for measur-
ing the melting point and the amount
of crystallinity in partially crystalline
polymers such as the polyolefins polyethy-
lene, polypropylene, and polybutylene.
This technique measures the heat of fu-
sion of a crystalline structure; it can
also give an indication of the modification
of crystalline sheeting with other polymers
163
-------
12 16 20 24 28 32 36 40 44
TIME, MINUTES
Figure 3.TGA of an unexposed black HDPE liner. The plots of sample weight and temperature
as a function of time are shown. Under an N2 atmosphere, the black HDPE sample
lost approximately 95.5% of its mass as hydrocarbons were evolved. The carbon
black added as an ultraviolet light absorber remained as a carbonaceous residue
and was not volatilized until it was oxidized when oxygen was allowed into the
system.
% Original Weight of Specimen \ /
12 16 20 24 28 32 36 40 44 48 52 56 60
TIME, MINUTES
Figure 4.Thermogravimetric analysis of an unexposed EPDM liner membrane. The dotted line
shows the temperature program and the solid line shows the percent of the original
specimen weight. At 46 minutes the atmosphere was changed from nitrogen to air
to burn the carbon black.
164
-------
TABLE 5.THERM06RAVIMETRIC ANALYSIS OF POLYMERIC MEMBRANE LINERS
Polymer type
Volatiles, Polymer,
Ui I or pi as- Carbon Ash,
ticizer, % black, % %
Butyl rubber
Chlorinated
polyethylene
Chlorosul fonated
polyethylene
Elasticized
polyolefin
Epichlorhydrin
rubber
Ethylene propylene
0
0
0.4
1.0
0.9
0.1
45.0
72.2
71.3
53.9
49.3
47.7
58.1
93.1
49.3
12.2
7.6
9.1
13.9
1.5
3.2
5.5
1.7
8.2
37.1
5.3
6.5
21.0
45.6
45.2
9.8
4.0
37.7
5.7
14.9
13.1
10.8
2.6
3.0
26.5
1.2
4.8
rubber
Neoprene
Polyethylene (high
density)
Polyvinyl chloride
0.1
0.2
1.0
0
0
0
0
0
0
0
30.8
33.5
42.3
44.0
97.9
95.6
97.0
48.7
46.0
51.0
32.9
23.2
10.7
10.7
U
0
0
38.2
42.1
35.0
30.9
35.5
34.9
33.8
2.1
4.2
1.8
6.2
7.8
7.0
5.3
7.6
11.1
11.5
...
0.2
1.2
6.9
4.1
7.0
by alloying. Thus this type of analysis can
be used as a means of fingerprinting crys-
talline polymeric liner materials (partic-
ularly high-density polyethylene) and
assessing the effects of aging and exposure
to wastes. This technique is described by
Boyer (1977) and Ke (1966).
The differential scanning calorimeter
used in this work was the Perkin-Elmer
Model DSC-2C, equipped with an Intracooler I
subambient temperature accessory to provide
an operating temperature range of -40 to
725°C.
The equipment is used to characterize
the thermal transitions of materials
such as melting, boiling, and changes in
crystal structure. A sample and reference
cell are provided. A weighed sample is
placed in the sample holder; the reference
cell is generally run empty to provide an
absence of thermal transitions. The two
cells are simultaneously heated or cooled so
that the average cell temperature follows a
preset program. When the sample undergoes
a thermal transition, an endothermic or
exothermic reaction will occur. The
change in power required to maintain the
sample cell at the same temperature as the
reference cell is recorded as a deflection
of the recorder pen. The recorder plots the
temperature (°C) versus the differential
energy flow (meal/sec) required to maintain
the sample cell temperature. An endothermic
transition such as melting is shown as a
positive peak; an exothermic reaction such
as crystallization is shown as a negative
peak. The magnitudes of these peaks and the
temperatures at which they occur are charac-
teristic of the material analyzed.
An example of the use of the DSC to
determine the polyethylene crystallinity in
an HOPE liner is shown in Figure 5.
165
-------
u
HI
CO
<
oc
o
LL
I-
LU
X
Endotherm
396 K = Melting Point
~ Exotherm
n
0 5 meal/sec
380 390 400
TEMPERATURE, KELVIN
Figure 5. DSC determination of the poly-
ethylene crystal 1inity in an
HOPE liner. The x-axis is the
temperature which was raised at
5°C/min. The y-axis is cali-
brated in meal/sec, or rate of
energy flow. A positive deflec-
tion of the plot indicates that
the sample is absorbing energy
(e.g., during melting). The
amount of energy absorbed during
the melting process may be
determined by calculating the
peak area and relating it to the
peak area resulting from the
melting of an Indium standard of
known weight. The energy absorb-
ed is termed the "heat of fusion"
( Hf). Assuming that Hf for
the fully crystalline polymer is
known, the degree of crystal lin-
ity of the sample can be deter-
mined as a simple ratio.
Specific gravity
Specific gravity is an important
characteristic of a material and is gen-
erally easy to determine. Because of dif-
ferences in the specific gravities of the
base polymers, specific gravity of the liner-
compound can give an indication of the com-
position and identification of the polymer.
Specific gravities of base polymers and of
different liner compounds based on them
are presented in Table 3. They show the
the differences among polymers and the
variations in compounds from one manu-
facturer to another.
ASTM Method D792, Method A-l, and D297,
Section 15.1.2, Hydrostatic Method, are
generally used in performing this test.
ANALYSIS AND FINGERPRINTING OF EXPOSED
POLYMERIC MEMBRANE LINERS
During service, several processes can
take place to change the composition of a
liner and thus affect physical properties
and possibly performance. First, the liner
can absorb waste liquids including water,
various organics, and possibly some inor-
ganic substances. The composition of the
absorbed chemicals will probably reflect
that of the waste liquid, but it will vary
depending on the relative solution charac-
teristics and the waste liquid constituents.
The absorbed organics can be both volatile
and nonvolatile. The total amount generally
softens the liner, resulting in possible
loss of tensile strength and other mechan-
ical properties, loss of elongation, and
increased permeability. While the organic
constituents are being absorbed, some of the
extractables such as plasticizer and oils
may migrate out of the compound, either by
evaporation, dissolution in waste liquid, or
biodegradation. Severe loss of plasticizer
could result in stiffening and loss in
mechanical properties. Losses in properties
can occur also through UV degradation and
oxidation of the polymer if the liner is
exposed to the weather. Figure 6 schemati-
cally represents the compositions of a
polymeric liner before and after exposure
and after extractions.
In assessing the long-term effects of
exposure to wastes and the other conditions
in a disposal facility, it is necessary to
know which waste constituents have affected
the lining material and what degradation may
have taken place. An analysis of exposed
liners is most useful to determine these
factors. Figure 7 illustrates a proposed
plan of analysis that we have used in our
analyses of exposed materials, recovered
from the laboratory or the field.
The various tests that are performed on
exposed membranes are the same as those
discussed in the previous section for
unexposed lining materials, with mod-
ifications and special precautions required
in each test for the exposed materials.
These individual tests are described in the
following subsections.
166
-------
PLASTICIZER
-^1 - — """ —
FILLER
POLYMER
XL OR TP
ORIGINAL
LINER
\
%
, WASTE , .
' LIQUID
PLASTICIZER
— —
FILLER
POLYMER
1
EXPOSED
LINER
% WEIGHT INCREASE
' • WASTE .
1 LIQUID ,
PLASTICIZER
~~_p: -xr
FILLER
POLYMER
VOLATILES
REMOVED
_». VOLATILES
{
PLASTICIZER
—
FILLER
POLYMER
VOLATILES
AND
EXTRACTABLES
%VOLATILES %EXTRACTABLE
• Moisture • Nonvolatile organi
• Organics
EXTRACTABLES
Figure 6.Schematic presentation of changes in composition of a polymeric liner compound on
exposure to a waste liquid, on removal of volatiles, and on extraction with
an appropriate solvent.
H20 + volatile organics
Plastiazer
Polymer
Carbon black
Ash
TGA = thermogravirnetrie analysis
GC - gas chromatography
IR = infrared spectroscopy
AAS - atomic absorption spectroscopy
CHONS - carbon, hydrogen, muogen, oxygen, and sulfur determination
Figure 7.Plan for the analysis of exposed polymeric lining materials.
167
-------
Volatiles
The initial analytical test generally
performed is a determination of the vola-
tiles content, which gives an indication of
the amount of waste liquid that has been ab-
sorbed by the liner. As an approximation,
the weight increase can be calculated by
dividing the percent of volatiles by the
percent of nonvolatiles. For example, if
the volatiles are 15%, and the nonvolatiles
are 85%, the percent increase in weight
based on the original liner would be 17.6%.
Inasmuch as the volatiles contain
both water and organic components, it
is desirable to separate these two. A
study was made to dehydrate the speci-
mens of the exposed liner to remove the
water. A series of desiccants was stud-
ied at room temperature and 50°C. Four
days of exposure of the exposed liner at
50°C in a small individual desiccator
containing calcium chloride removed the
moisture from the 2-in. disk specimen
without removing organic constituents.
After removal of the water, a 2-hr heating
of the specimen in a circulat-ing air oven
removed the remaining organic volatiles.
Collection of the organic volatiles and
determination of composition by gas
chromatography would be desirable but
has not been attempted to date. The
water and the organic volatiles equal
the volatiles obtained in bypassing the
desiccator exposure. Such a bypass removes
the moisture by exposure at room temperature
for 1 week in moving air followed by heating
in a circulating air oven for 20 hr at 50°C
and 2 hr at 105°C. We have found, however,
that the highly swollen CPE liner may take
up to 6 days at 50°C to come to constant
weight. The time required to remove
volatiles depends on the thickness and
permeability of the liner and the care taken
to avoid the "skin" that forms on the
surface of the specimen when using too high
an initial temperature to devolatil ize.
After the volatiles are removed, the
exposed materials can be subjected to
the other tests, including specific gravity,
extractables, ashing, etc.
Total volatiles can also be determined
through the use of TGA which is discussed
below.
Extractables
Extractables of exposed materials
will probably differ from the original
values because of the loss to the waste
liquid and absorption of nonvolatile
organics (e.g., oils). After the vol-
atiles have been removed from the liner,
the extractables are determined by the same
method used on the unexposed liner materi-
als. Examples of extractable contents after
exposure are given in Table 6. If the liner
has been in contact with wastes containing
nonvolatile constituents, the extractables
recovered may be greater than the original
values. Analysis of the extractables by gas
chromatography and infrared analysis may
give an indication of the nonvolatile
organics that were absorbed. The analysis of
the extractables will give an indication of
the constituents of the waste that are
aggressive to the liner, as they are the
constituents that have been absorbed. They
may show up in minor amounts in a waste
analysis, but because of their chemical
characteristics such as solubility parame-
ters and hydrogen bonding, they may be
scavenged by the polymeric liner.
Ash
Thp ash content of an exposed membrane
liner is determined after the volatiles have
been removed from the specimen. As in the
case of the unexposed membrane, the exposed
liner is ashed in a muffle furnace at
550°C. The ash value usually differs from
that of the unexposed material, depending on
how many nonvolatile organics were lost or
gained during the exposure period. If plas-
ticizer is lost, the value will increase
because of the nonash content of the plasti-
cizer. Also, if any organic metal compounds
are absorbed by the liner, they will show as
an increased ash content. A comparison of
the elemental analysis of the ash with that
of the original liner will determine whether
any absorption of metal species occurred
during the exposure. No such absorption has
been observed, but metal organics might be
absorbed by a liner. In general, the in-
organics that make up the ash are retained
in the liner and maintain a constant ratio
with respect to the polymer content, as
polymer is generally not dissolved by the
waste liquid.
Thermogravimetric analysis
The TGA analysis can be used to give a
quick analysis of the composition of an
exposed polymeric membrane liner. The test
is run similarly to that of the unexposed
material, except that care must be taken in
handling the small specimens of exposed
168
-------
TABLE 6.EXTRACTABLES OF DIFFERENT POLYMERIC MEMBRANE LINERS
BEFORE AND AFTER VARIOUS EXPOSURES
Polymer
Butyl rubber
Unexposed
11.79
Type of exposure
by waste
Aromatic oil
Spent caustic
Oil 104
Extractables
after exposure
27.23
10.87
40.34
Chlorinated
polyethylene
9.13
Chlorosulfonated
polyethylene 4.08
Elasticized
polyolefin 5.50
Ethylene prop-
ylene rubber 23.64
Neoprene 21.46
Polyvinyl
chloride 33.90
Oil 104
Roof exposure
Aromatic oil
Oil 104
Slop water
Oil 104
Aromatic oil
Slop water
Aromatic oil
Oil 104
Spent caustic
Aromatic oil
Oil 104
Slop water
Aromatic oil
Roof exposure
Pesticide
Oil 104
17.00
5.99
59.81
15.92
3.70
20.74
23.37
2.96
38.35
43.45
22.89
58.47
23.85
17.63
40.55
26.27
35.38
17.95
liners that contain the volatiles. These
volatiles can be easily lost. Figure 8 is a
thermogram of an exposed PVC membrane liner
that had absorbed more than 7% of the waste
liquid, which was predominantly water. The
results are shown on the figure, but the
losses show the effect of the char formation
of the PVC when it is heated in a nitrogen
atmosphere. Chlorinated polymers lose HC1
and leave a char for which correction must
be made in the calculations of the polymer
content. These calculations have been made
on the figure, and the results compare fa-
vorably with those obtained by direct analy-
sis of the volatiles, extractables, and ash.
Specific gravity
The specific gravity of exposed mem-
brane liners is also determined after the
membrane specimen has been thoroughly de-
volatilized. Three steps must be taken to
avoid the formation of bubbles in the liner
mass during a direct devolatilizing process
(which would affect the specific gravity
results). First, the specimens must be
allowed to "dry" at room temperature in an
air stream in a hood until constant weight
is achieved, which could take several days.
Second, they must be placed in an oven at
70°C for 16 hr; and third, they must be
heated at 105°C for 2 hr.
The specific gravity of an exposed
membrane can differ from that before ex-
posure, depending on how much of the origi-
nal extractable material was lost and how
much material from the waste was absorbed
during exposure. The procedure followed was
ASTM D297 using the hydrostatic method.
169
-------
1000
800
w goo
rr
LU
400
200
110°C
% Original Weight
of Sample
X 650°C \
\ X
X
490°C
300°C
160%
-Temperature
Nitrogen Atmosphere
40
60 80
TIME, MINUTES
100
120
140
100
80
I-
I
(3
60 LU
40 S
CC
O
Figure 8. TGA of an exposed polymeric PVC liner.
Weight loss A
Weight loss B
Weight loss C
Weight loss D
7.0% volatiles = moisture + possible organics.
60.2% = plasticizer + HC1 from the polymer (PVC).
16.0% = residual polymer.
10.0% = carbonaceous polymer residue + carbon black,
Residue weight E = 6.8% = ash.
Composition of the exposed liner as received calculated from above data.
Values by direct analysis are shown in parentheses.
Volatiles 7.0% (7.9)a
Polymer (PVC) 44.7%
Plasticizer 34.1% (32.2% as extractables)3
Carbon black 7.4%
Ash 6.8% (6.4)a
aBy direct analysis.
SUMMARY AND CONCLUSIONS
A protocol is presented for the analysis
of polymeric membrane lining materials
before and after exposure to waste liquid.
The results of the analysis can be used
as a fingerprint of the unexposed liner and
as a baseline for assessing the changes that
occurred during exposure to a waste liquid.
The analysis of an exposed liner
furnishes information regarding the change
in the composition of the membrane and the
chemical materials that are actually absorb-
ed during exposure. This latter information
indicates the constituents of the waste that
are aggressive to the lining material.
Thermal analysis and gas chromatography
are particularly useful in the analysis
and fingerprinting of polymeric membranes.
The use of gas chromatography coupled
with mass spectroscopy needs to be investi-
gated to determine the specific chemicals
that are absorbed by a polymeric liner.
The use of this analysis may lead to
techniques which will better predict the
service life of a flexible membrane liner.
ACKNOWLEDGEMENTS
work reported in this paper was
performed under Contracts 68-03-2134
("Evaluation of Liner Materials Exposed to
Leachate"), 68-03-2173 ("Evaluation of Liner
Materials Exposed to Hazardous and Toxic
Wastes"), and 68-03-2969 ("Long-term Testing
of Liner Materials") with the Municipal En-
vironmental Research Laboratory of the U. S.
Environmental Protection Agency, Cincinnati,
Ohio.
170
-------
The author wishes to thank Robert E.
Landreth, Project Officer, for his support
and guidance in these projects.
REFERENCES
ASTM Standards, American Society for Testing
and Materials, Philadelphia, PA.
D297-81, Rubber Products - Chemical
Analyses: Section 15, Density (Hydro-
static Method); Section 34, Fillers,
Referee Ash Method
D792-66, Specific Gravity and Density of
Plastics by Displacement; Method A-l
for Testing Solid Plastics in Water,
Sections 6-12.
D1203-67, Volatile Loss from Plastics
Using Activated Carbon Methods.
D3421-75, Extraction and Analysis of
Plasticizers from Vinyl Chloride
Plastics
Boyer, R. F. 1977. Transitions and Relaxa-
tions. In: Encycl. Polymer. Sci.
Technol. Supplement, Vol. 2. pp
745-839.
Haxo, H. E. 1983. Liner Materials Exposed
to Hazardous and Toxic Wastes. Final
Report. Contract 68-02-2173. U. S.
Environmental Protection Agency,
Cincinnati, OH. In preparation.
Haxo, H. E., R. M. White, P. D. Haxo, and M.
H. Fong. 1982. Liner Materials
Exposed to Municipal Solid Waste
Leachate. Final Report. Contract
68-03-2134. U. S. Environmental
Protection Agency, Cincinnati, OH. In
press.
Ke, B. 1971. Differential Thermal Analy-
sis. In: Encycl. Polymer. Sci.
Technol. Vol 5, pp 37-65.
Matrecon, Inc. 1982. Lining of Waste
Impoundment and Disposal Facilities.
SW-870, Second Edition. U.S. Environ-
mental Protection Agency, Washington,
DC. In press.
Reich, L., and D.
gravimetric
Polymer Sci,
1-41.
W. Levi.
Analysis.
Technol.
1971. Thermo-
In: Encycl.
Vol 14. pp
171
-------
EVALUATION OF A WASTE IMPOUNDMENT LINER
SYSTEM AFTER LONG-TERM EXPOSURE
by
Susan A. Roberts
Malcolm Pirnie, Inc.
White Plains, New York
Nancy A. Nelson
and
Henry E. Haxo, Jr.
Matrecon, Inc.
Oakland, California
ABSTRACT
Samples of a 20-mil polyvinyl chloride (PVC) liner in use for 9 years in an industrial
sludge lagoon that contained a calcium sulfate slurry were collected and tested. The
lagoon was scheduled to be excavated and relined because of mechanical damage to the liner;
therefore, samples were available for studying the effects of long-term exposure of lining
materials to wastes.
Eighteen samples of the liner were collected from the berm, side slopes, and bottom
of the lagoon. Nine samples were selected for analysis and physical testing to assess the
effects of directional orientation of weather exposure and the effects of sludge exposure.
They included samples from the north, south, east, and west sides, from under sludge at
different depths, samples of factory and field seams, and any anomalous samples. Measure-
ments were made of tensile strength, tear strength, puncture resistance, hardness, seam
strength, volatiles content, and extractables content.
Results of the testing show that exposure of the liner to the weather had the greatest
effect on the PVC liner, with the samples facing south being most severely affected.
Judging by their low extractables, the weather-exposed samples lost plasticizer; they had
low elongation at break values, high hardness values, and some were brittle. The samples
exposed to the sludge or soil had higher extractables and elongation at break values and
lower hardness values than weather-exposed specimens. Two samples taken from the lagoon
bottom had comparable properties to those of unexposed 20-mil PVC liners. The factory
seams were in good condition. A sample of the field seam, however, contained an unbonded
section and failed in the glue line in shear testing, indicating an inadequate seam.
INTRODUCTION
A considerable number of laboratory
tests and pilot-scale studies are currently
being conducted to assess the effects on
polymeric membrane liners from exposure to
industrial wastes. However, limited inform-
ation is being collected to assess the
performance of liners at industrial sites
and lagoons (1,2). The scarcity of inform-
ation is partly due to the short time period
during which liners have been in use in
waste impoundments and the problems associ-
ated with obtaining liner samples from im-
poundments.
The purpose of this study, conducted
by Malcolm Pirnie, Inc. and Matrecon, Inc.
was to provide field verification of labo-
172
-------
ratory testing methods of polymeric liners
and to evaluate site-specific effects of
hazardous waste disposal on the performance
of a polymeric liner.
In August and September 1982, Malcolm
Pirnie, Inc., collected samples of a 20-mil
polyvinyl chloride (PVC) liner from an in-
dustrial sludge lagoon. The lagoon was
scheduled to be excavated and relined;
thus it was possible to obtain samples of
the exposed PVC liner. The liner samples
were tested to determine properties of the
liner after a service period of 9 years.
BACKGROUND
The site is located in the northeast
United States. The lagoon was constructed
in 1973, and a PVC liner was installed.
The lagoon, which is approximately 85
ft wide, 240 ft long, and 9 ft deep, has
a generally north-south orientation and a
side slope of approximately 2:1. The
PVC liner was fabricated by the supplier in
two pieces, which were seamed in the field
by the installer. This field seam runs the
length of the lagoon and bisects the 85-ft
side as shown in Figure 1. The factory
seams run parallel to the field seam. All
seams appear to have been made using a
bodied solvent. In the corners of the
lagoon, the liner was folded into place.
An anchor trench was not used to secure
the liner over the berm; instead, the upper
edge of the liner was folded and soil was
placed on it. The PVC liner was designed
to be covered during the lagoon operation;
although a sand layer covered it initially,
it sloughed off the top of the berm.
The lagoon was used as a settling
basin where a slurry waste was pumped in,
the solids were allowed to settle, and the
liquid was pumped back into the plant.
When the lagoon was filled, at approximately
1- to 2-year intervals, the contents were
removed and landfilled onsite. A clam-
shell was used to scoop out the contents,
except for a thin layer of sludge that was
left on the liner to avoid damaging it. The
liner was reported to have been snagged on
occasion and punctured by the clamshell.
The waste disposed of in the lagoon is
a calcium sulfate sludge, the result of the
neutralization of sulfuric acid with lime.
The sludge also contains ammonia and chlor-
ides. The sludge is a fine-grained, gen-
erally greyish-white moist solid. In the
lower portion of the lagoon, the pore spaces
in the sludge are completely filled with
a precipitation/ground-water/waste liquid.
This is termed the saturated portion of the
sludge. In the upper portion of the lagoon,
the pore spaces in the sludge are not always
completely filled with the liquid. This
is termed the unsaturated portion of the
sludge. The location of the unsaturated/
saturated zone interface has moved up and
down during the operation of the lagoon.
However, the sludge at the bottom of the
lagoon probably remained saturated through-
out the life of the lagoon. The pH of the
lagoon liquid varied from 8 to 9 at the
times the liner samples were collected. A
general analysis of the waste liquid in the
lagoon is presented in Table 1; this is
probably the best representation available
of the liquid in contact with the inside
of the PVC liner in the saturated sludge
portion of the lagoon slope.
TABLE 1.
GENERAL ANALYSIS OF SLUDGE LIQUID
Item
pH
Specific
tance,
Chlorides
Sulfates,
Ammonia,
Amount
conduc-
pmhos
, mg/L
mg/L
mg/L
15
4
15
1
8.0 -
,000 -
,500 -
,000 -
,000 -
9.0
60,000
5,000
20,000
7,000
The lagoon is underlain by glacial
deposits that overlie fractured crystalline
bedrock. The glacial deposits consist of
outwash material made up of well-graded to
poorly graded sands and silts, underlain by
till made up of unstratified, poorly sorted
sands, silts, and gravel with some cobbles
and boulders. A natural organic/peat layer
overlies part of the glacial material, and
fill is present in some areas around the
lagoon. The water table, or the top of the
ground-water surface, was approximately 6
to 8 ft below the ground surface in the
area of the lagoon; thus the bottom of the
lagoon would have been in the ground water.
Leakage from the lagoons appears to have
contributed to a ground-water mound in the
vicinity of the lagoon. In addition, sea-
sonal ground-water level fluctuations
resulted in alternate wetting/drying
cycles of the soil on the outside of the
liner.
173
-------
APPROXIMATE
EDGE OF
SATURATED
ZONE
rAPPROXIMATE LOCATION
OF FIELD SEAM
LAGOON EDGE -,
/ q
APPROXIMATE LINER EDGE
r-ArrKUAiMm Lihth
®
LAGOON
©
BOTTOM
t
N
-APPROXIMATE
WASTE EDGE
v_
15
15
30
SCALE IN FEET
Figure 1 (See figure legends list)
174
-------
PROCEDURES
SAMPLING
Samples of the PVC lagoon liner were
obtained according to the procedure devel-
oped by Matrecon, Inc. The samples were
rolled up and wrapped with aluminum foil,
then placed in a polyethylene bag and boxed
for shipment. Six liner samples were taken
from the upper edge of the lagoon in August
1982. Subsequently, 12 additional liner
samples were taken from the lagoon sides
and bottom in September 1982, when the
lagoon was being excavated. Matrecon
diagrammed and photographed the liner samples
upon receipt; they were then resealed to
minimize loss of any volatiles before analy-
sis and testing. In addition, samples of
the sludge above the PVC liner and of the
soil below it on the lagoon bottom were
collected in September 1982.
The lagoon liner sampling locations
are shown in Figure 1. Two techniques were
used to obtain the samples. The samples
located in the upper portion of the lagoon,
i.e., the weather-exposed samples and the
unsaturated zone samples, were cut out
directly from the side of the lagoon. The
samples located in the lower portion of the
lagoon were taken out by the clamshell, then
cut to size.Use of the clamshell was neces-
sary since the lower portion of the lagoon
was filled with a mixture of saturated
sludge and ground water, and direct access
was impossible.
Field observations of the liner samples
are in Table 2.
Ten of the 18 liner samples were se-
lected for testing based on the following
criteria:
o A sample with minimal or no exposure
to sun or sludge to use as a possible
control, since no retained sample was
available. This sample was taken from
the upper edge where the liner had been
covered with soil. It is possible that
this sample was exposed to sludge.
o Samples exposed to the weather from
north, south, east, and west exposures.
These samples were taken either from
liner material completely exposed to
the weather or from material half
exposed and half in the waste. The
samples were chosen on the basis of
evidence of possible degradation, such
as embrittlement, tautness, and any
other indications of anomalies in the
liner.
o Samples representing a vertical cross-
section of the lagoon from the top edge
to the bottom. Samples were taken from
the berm, from the slope, and from the
bottom. The berm samples were primarily
weather-exposed. The slope samples were
taken from the upper waste zone, where
the waste is saturated. Figure 2 shows
an idealized cross-section of the
lagoon.
o Samples with field and/or factory
seams.
TESTING
The following samples were selected for
testing:
Sample 1 - Upper and lower halves.
Sample 3 - Upper half.
Sample 4 - Upper and lower halves,
Sample 5 - Seamed area.
Sample 6 - Upper half.
Sample 7 - Seamed area.
Sample 11 - Seamed area.
Sample 12 - Upper and lower halves.
Sample 15 - Seamed area.
Sample 17 - Seamed area.
Standard tests were used to characterize
the PVC liner samples (3). The samples
were removed from their sealed bags and
physically tested on an "as received" basis.
Physical testing included tensile strength
(ASTM D412), tear strength (ASTM D624),
puncture resistance (Federal Test Method
Standard No. 10IB, Method 2065), and hard-
ness (ASTM D2240). Grain direction of all
samples was determined by heating marked
discs,- shrinkage occurs in the machine di-
rection. Where available, seams were tested
in shear (ASTM D816) and in peel (ASTM
D1876). Volatiles were removed from the
samples in two steps. First, moisture was
removed by placing samples over calcium
chloride (CaCl2) in a desiccator at 50°C
for 4 days. Then the samples were heated
at 105°C for 4 to 20 hr with a 2:1 carbon
tetrachloride (CC14): methanol (CH3OH)
solution. Ash (ASTM D297) was also run
on devolatilized samples.
RESULTS
Results of the testing are summarized
175
-------
TABLE 2. FIELD OBSERVATIONS OF LINER SAMPLES
Sample Location in lagoon Seams Condition of Samples
1 North side of lagoon Factory seam in both halves Upper half cracked, brittle
Upper half weather-exposed
Lower half exposed to unsaturated sludge Lower half supple
2 West side of lagoon, northern corner Factory seam in upper half Upper half brittle, ripped, shows discoloration
Upper half weather-exposed
Lower half exposed to unsaturated sludge Lower half supple
3 West side of lagoon Factory seam in upper half Upper half brittle, cracked, grass growing
Upper half weather-exposed tniough hole in liner
Lower half exposed to unsatvirated sludge Lower half is supple
4 South side of lagoon Factory seam IP both halves Upper half brittle, cracked
Upper half weather-exposed Unadhered field seam in upper
half
Lower half exposed to unsaturated sludge Intact field seam in lower half Lower half supple
5 East side of lagoon Factory seam Discolored, supple
Completely exposed to weather
6 East side of lagoon Factory =;eam in upper half Upper half brittle, cracked
Upper half weather-exposed
Lower half exposed to unsaturated sludge Lower half supple
7 East side of lagoon near lagoon bottom No seam Dimpled, overlies hydrogen-sulfide-smelling,
Completely under saturated sludge peat-like black soil
8 Bottom sample from southern ecd of lagoon No seam Dimpled, supple
9 East side of lagoon near lagoon bottom No seam 0_-pled, supple, overlies similar peat-like
Completely under saturated sludge black soil as #7
10 South side of lagoon near lagoon bottom No seam Two creases in sample, underlain by similar
Completely under saturated sludge peat-like black soil as #7
11 Under north berm-folded portion Factory seam Dry, fairly stiff
Exposed to soil, same possible exposure
to sludge
12 North side of lagoon on slope Factory seam in upper half
Upper half in unsaturated sludge
Lower half in saturated sludge Lower half dimpled, supple
13 North side of lagoon on slope No seam Dimpled, supple
Completely under saturated sludge
14 Bottom samples from lagoon center Dimpled, supple
15 Factory seam in #15
16 Bottom samples from northern end of Dimpled, supple
lagoon
1? Factory seam in #17
18 Sample adiacent to #11 No seam Dryi fairly stiff
176
-------
APPROXIMATE GROUND-WATER LEVEL
(FLUCTUATES)
UNSATURATEO SLUDGE
(Fluctuates)
SATURATED SLUDGE
14) (15) (16) (17)
3036
VERTICAL SCALE IN FEET
Figure 2 (See figure legends list)
-------
in three tables. Table 3 presents data on
the samples exposed to the weather arranged
in order of decreasing exposure severity—
i.e., south-facing exposure being most
severe, followed by west-facing, then east-
facing,with north-facing exposure being the
least severe. In Table 4, the data are pre-
sented in a vertical cross-section arranged
from the top of the lagoon to the bottom.
Table 5 contains data on all the seam test-
ing.
DISCUSSION
WEATHER-EXPOSED SAMPLES
Samples of the liner exposed to weath-
ering were generally brittle and dry. Dur-
ing the sample collection, the brittleness
was evident through cracking and splitting
of the membrane. However, it was possible
to roll up the liner samples, possibly
because of the warm weather. Localized
shrinkage of the liner had occurred, as
evidenced visually by its tautness. Ten-
sile test specimens from the top of sample
1 cracked along the edges as they were
died out. An attempt was made to avoid the
cracks by dieing specimens from a warmed
sample, but these also cracked. Tensile
and tear strenath values reflect this
brittleness.
The brittleness of the weather-exposed
samples,probably resulting from the loss
of plasticizer, is shown by low values
for extractables and elongation at break
and high hardness values (see Figures
3 through 5). Only sample 5 , which was
taken near the discharge pipe, was not
brittle. Sample 6, located adjacent to
sample 5, was brittle and is a better ex-
ample of west-facing exposure. The
arrangement of samples on Table 5 (with the
exception of sample 11), from expected
worst conditions to mildest, holds true for
the test values. Sample 5 is not considered
to be representative of a weather-exposed
sample because of exposure to newly dis-
charged material.
SLUDGE-EXPOSED SAMPLES
Liner samples from the main body of
the lagoon were exposed to:
o Saturated sludge on the bottom.
o Intermittent cycles of saturated and
unsaturated sludge on the slopes when
the liquid level within the sludge fluc-
tuated.
o Intermittent cycles of weather and
sludge at the waste/weather interface
at the top of the slope.
During the sample collection, it was
noted that the liner material in the samples
exposed to saturated waste showed an imprint
of the underlying soil and some swelling,
whereas samples exposed to unsaturated
waste were flat. Samples taken on the low-
er portion of the slopes at the southern
end of the lagoon were underlain by black,
peak-like granular soil with a hydrogen
sulfide odor.
The uppermost liner samples on the
lagoon slope at the waste/weather interface
intermittently exposed to sludge (the lower
portion of samples 1 and 4) showed similar
properties to those of the liner sample
exposed to the soil (sample 11). These
two waste/weather samples had the lowest
extractables content and elongation at
break values and the highest S-100 values
of all sludge-exposed samples. Volatiles
for these samples (1 and 4) were 5.10% and
3.14% water and 1.41% and 0.52% organic
volatiles after 20 hr at 105°C, respectively
(Figures 3 through 5).
Samples tested from lower on the lagoon
slope, which were exposed to saturated or
unsaturated sludge, had similar properties
(Table 4). The volatiles content of these
latter samples was higher than those of the
weather-exposed samples, ranging from
5.60% to 8.62% water, and from 1.54% to
1.96% organic volatiles. They had lower
hardness values (see Table 4), lower S-100
values, and higher elongation at break
values than the weather-exposed samples
(Figures 4 and 5).
The two samples from different loca-
tions on the lagoon bottom (samples 15 and
17) had similar water contents—3.42% and
2.49%, respectively—and their organic
volatiles measured 0.85% and 0.94%, respec-
tively. The samples had equal hardness.
These samples had the highest extractables
and the lowest hardness of all tested
samples.
SOIL-EXPOSED SAMPLE
The soil-exposed sample (sample 11)
was collected as a possible control. Be-
cause of the nature of the cleaning opera-
178
-------
TABLE 3. WEATHERED SAMPLES OF PVC LINER FOR CALCIUM SULFATE SLUDGE LAGOON
Sample number
Item
Lagoon location
Exposure conditions
Average thickness, mils
Analytical properties:
Volatiles, % over desiccant (ar)
At 105°C 20 hr (ar)
Extractables , % (db)
Ash, % (db)
Physical properties (ar) :
Tensile at break, psi:
Machine
Transverse
Elongation at break, %:
Machine
Transverse
Tensile set, %:
Machine
Transverse
S-100, psi:
Machine
Transverse
S-200, psi:
Machine
Transverse
Tear strength, ppi :
Machine
Transverse
Puncture resistance :
Gage, mils
Stress, Ib
Elongation, in.
Hardness , shore D
5-sec reading
11
N-side
Undersoil
18.8
0.75
0.34
26.50
7.54
2,940
3,135
260
285
165
165
2,345
2,280
2,700
2,715
535
580
19.8
37.6
0.44
42
1 Upper
N-side/s-facing
Weather
16.0
1.37
1.78
18.91
8.19
2,430
3,015
5
25
1
15
235
170
16.7
9.2
0.20
53
6 Upper
E-side/w- facing
Weather
15.6
3.68
0.28b
20.23
2,500
2,470
25
20
20
3
295
327
16.8
16.2
0.35
54
5
E-side/w- facing
Weather
19.3
1.04
2.18
30.52
6.45
3,310
3,205
205
225
125
130
2,810
2,585
3,240
3,405
560
495
18.3
33.1
0.42
42
3 Upper
W-side/e- facing
Weather
16.0
4.26
0.92'
20.54
7.50
3,390
3,155
80
55
50
50
515
355
17.4
25.2
0.26
52
4 Upper
S-side/N-facing
Weather
18.2
1.56
0.72
21.91
8.41
2,925
2,795
115
70
90
50
2,895
2,645
595
430
16.0
26.4
0.28
51
ar = as received; db = dry basis.
bAt 105° for 4 hr.
CTests performed on samples as received.
-------
TABLE 4. ANALYSIS OF SAMPLES FROM A VERTICAL CROSS-SECTION OF PVC LINER FOR CALCIUM SULFATE SLUDGE LASOON
Item
Lagoon location
11
N-side
Exposure conditions Undersoil
Average thickness, mils
Analytical properties:
Volatiles, % over desiccant (ar)
At 105°C 20 hr(ar)
Extractables, %(db)
Ash, %(db)
c
Physical properties:
Tensile at break, psi:
Machine
Transverse
Elongation at break, *:
Machine
Transverse
Tensile set,%:
Machine
Transverse
S-100, psi:
Machine
Transverse
S-200, psi:
Machine
Transverse
Tear strength , ppi :
Machine
Transverse
Puncture resistance:
Gage, mils
Stress, Ib
Elongation, in.
Hardness, shore D
5-sec reading
ar = as received; db = dry
At 105° for 4 hr.
18.8
30.75
0.34
26.50
7.54
2,940
3,135
260
285
165
165
2,345
2,280
2,700
2,715
535
580
19.8
37.6
0.44
42
basis.
1 Upper
N-side
Weather
16.0
1.37
1.78
18.91
8.19
2,430
3,015
5
25
1
15
235
170
16.7
9.2
0.20
53
1 Lower
N-side
Weather/waste
interface
20.3
5.10
1.41
21.50
7.90
2,915
2,575
220
220
125
110
2,350
2,090
2,830
2,460
520
355
20.4
31.6
0.33
51
Sample
4 Lower
S-side
Weather/waste
interface
20.8
3.14
0.52
26.10
9.32
2,780
2,495
275
295
105
130
1,885
1,720
2,350
2,065
495
415
21.1
40.1
0.64
36
number
12 Upper
N-side
Unsaturated
waste
21.2
8.62
1.96
31.02
6.89
2,610
2,315
300
310
95
100
1,635
1,340
2,105
1,750
345
330
21.2
35.2
0.66
32
12 Lower
N-side
Saturated
waste
21.6
8.27
1.60
32.71
6.67
2,505
2,335
295
330
70
90
1,500
1,205
2,010
1,630
330
310
21.6
32.3
0.71
29
7
W-side
Saturated
waste
20.4
5.60
1.54
32.44
7.13
2,605
2,440
280
300
85
90
1,740
1,580
2,210
2,010
380
360
20.2
31.8
0.63
34
15
Center
Lagoon
bottom
19.8
3.42
0.85b
34.98
7.08
2,460
2,320
310
315
70
80
1,265
1,245
1,805
1,700
260
270
19.7
26.1
0.69
28
17
N-end
Lagoon
bottom
20.7
2.49
0.94
35.15
6.69
2,760
2,550
265
290
60
70
1,625
1,415
2,280
1,975
325
295
20.5
32.6
0.67
28
Tests performed on samples as received.
-------
TABLE 5. DATA ON SEAMS IN PVC LINER FOR CALCIUM SULFATE SLUDGE LAGOON
00
Sample number
Item
Exposure conditions
Breaking factor , ppi
transverse
Factory seam strength:
Shear, ppi
Locus of failure
% of breaking factor
Factory seam strength:
Peel , ppi
Locus of failure
Field seam strength:
Shear , ppi
% of breaking factor
Field seam strength:
Peel i ppi
Locus of failure
] -Upper 1-Lower
Weather Weather/waste
interface
48.3 49.7
43.9 46.8
SE-BRKa SE-BRK
91 94
27.0 21.3
AD AD
3 -Upper 4 -Upper
Weather Weather/waste
interface
52.1 51.7
40.3 38.4
SE SE
77 74
26.1 25.5
AD AD
4-Lower
Weather/waste
interface
52.5
44.2
SE
84
25.0
AD
"
SE— AD
87
13.0
AD
5 7 11 12-Upper 15 17
Weather Saturated Soil Unsaturated Bottom Bottom
waste waste
62.0 49.4 54.3 48.7 45.6 52.4
51.8 44.8 53.8 47.2 40.0 43.7
SE SE SE SE SE SE
84 91 99 97 88 83
26.6 21.1 25.8 23.5 17.7 20.7
AD AD AD AD AD AD
SE = failure at seam edge; AD = failure in seam adhesion; SE-BRK = failure at either seam edge or in liner; SE-AD = failure at either seam
edge or in seam adhesion.
-------
SATURATED
WASTE, BOTTOM
40 -
30 -
SOIL-
COVERED
26.5
20 -
10 -
DISCHARGE
PIPE
30.5
UNSATURATED WASTE,
SLOPE
31.0
WEATHERED-EXPOSED
18.9
20.2
20.5
21.9
26.1
21.5
SATURATED
WASTE, SLOPE
32.4
35.0
35.2
#11 #1u #6u I3u #4u #5 #1L #4L #12u #12L #7 115 #17
I I
0 FT.
10 FT.
DEPTH OF SAMPLE
Figure 3 (See figure legends list)
-------
400-
SOIL
COVERED
300 -
200 -
100 -
CO CO
WEATHER-
EXPOSED
68
23
UNSATURATED
WASTE, SLOPE
SATURATED
WASTE, SLOPE
DISCHARGE
PIPE
285
305
313
290
SATURATED
WASTE, BOTTOM
313
278
#11 #1u #6u #3u #4u #5 IIL #4L #12u #12L 17 115 117,
0 FT.
10 FT.
DEPTH OF SAMPLE
Figure 4 (See figure legends list)
-------
4000 -
DISCHARGE
WEATHER - EXPOSED PIPE
3000 -
SOIL
COVERED
2313
2000 -
1000 -
SAMPLES
DID NOT
ACHIEVE
100%
ELONGATION
UNSATURATED
WASTt, SLOPE
2770
2698
2220
1803
SATURATED
WASTE, SLOPE
WASTE, BOTTOM
1488
1353
1255
'520
#11 #11) #6U #3U #4U #5 #1L #4L #12U #12L #7 #15 #17.
0 FT.
10 FT.
DEPTH OF SAMPLE
Figure 5 (See figure legends list)
-------
tions at the lagoon, however, the sample
may have been exposed at some point to the
weather and possibly to the sludge. The
low extractables content and high modulus
values (S-100) indicate that the sample
is likely to have undergone moderately
severe exposure conditions; therefore it
could not be used as a control.
SEAMS
Factory seam samples from different
levels in the lagoon and one sample of the
field seam were collected. All seams appear
to have been made with a bodied solvent.
The factory seams are approximately 1 in.
wide and the field seam approximately 2
in. wide. All seams appear to have
been made in the machine direction; thus
testing across the seam constitutes a trans-
verse direction test with respect to the
sheeting. All factory seams were observed
to be in good condition. Seam strength
tested in shear of the factory seams gener-
ally was good,ranging from 74% to 99% of
the breaking strength of the liner material.
The brittle, weathered samples and the
samples from the bottom of the lagoon
had the lowest seam strength in shear; the
sample exposed mainly to soil ( No. 11)
had the best strength. All of the seams
tested in peel failed in the adhesive or
"glue line." The seams peeled apart, leav-
ing traces of bodied solvent on both sides.
The lowest peel strength, 17.7 ppi, was
measured in samples from the bottom of the
lagoon. The highest, 27.0 ppi, was from a
weather-exposed sample.
The field seam collected had a section
that had not been properly bonded. This
appeared to be a "holiday" or unbonded
section. The bonded section showed good
shear strength at 87% of the breaking
strength of the material. This is higher
than the factory seam shear strength for
the same sample, but it does not take
seam width into account. The peel strength,
however, was low at 13.0 ppi, which is
approximately half the peel strength of the
factory seam.
CONCLUSIONS
1. Results from field testing confirm
previous laboratory and pilot-scale
studies. Properties of PVC liner
samples from a sludge lagoon vary
considerably, depending on sample
location,and the condition of samples
reflects the degree to which they are
waste- or weather-exposed. However,
neither the properties nor their vari-
ability were unexpected in light of
results from previous laboratory test-
ing (4) .
Exposure to the weather was the most
significant source of degradation of the
liner. The sample on the north side
(south-facing) showed the most severe
effects. Samples of the liner under
the sludge, which underwent minimal
or no exposure to the weather (satur-
ated sludge and lagoon bottom samples),
showed normal properties for PVC liners.
Samples exposed to the weather showed
evidence of shrinking and had low val-
ues for thickness. These observations
coupled with the physical deterioration
indicate a loss of plasticizer. Compar-
ing the extractables contents of the
samples as an estimate of the plasti-
cizer content with 30% (which approxi-
mates the amount normally compounded
in PVC sheeting), it is clear that the
weathered samples have lost plastici-
zer. The samples on the lagoon bottom,
however, have retained plasticizer and
physical properties.
This PVC liner, after 9 years of expo-
sure under the waste, has comparable
testing values to unexposed 20-mil PVC
sheeting, according to the specifications
shown in Table 6. Therefore it appears
that the type of sludge being impounded
did not affect the integrity of the PVC
liner. However, if a retained sample
of the original PVC liner were avail-
able, these results could be expressed
as changes in properties as a result of
exposure or percent retention of unex-
posed 20-mil PVC sheeting, as reported
in the literature.
The factory seams in this liner main-
tained their integrity during the long-
term exposure, but the field seams did
not. All factory seams tested in shear
failed at the seam edge or elsewhere in
the sample, reflecting the strength of
the liner. The field seam, however,
had some failure in the seam adhesive
itself as well as seam edge failure.
When tested in peel, which is a more
severe test, all seams failed in the
glue lines. The type of seam failure
in shear in the fielcT-seam and the
evidence of a "holiday" 'indicate an
inadequate seaming operation and empha-
185
-------
TABLE 6. PROPERTIES OF UNEXPOSED 20-MIL PVC MEMBRANES AND SAMPLES EXPOSED TO
SLUDGE ON LAGOON BOTTOM
Unexposed
Physical properties
Average thickness, mils
Tensile strength, psi:
Machine
Transverse
Elongation at break, %:
Machine
Transverse
Tensile set, %:
Machine
Transverse
S-100, psi:
Machine
Transverse
S-200, psi:
Machine
Transverse
Tear strength (Die C) , ppi:
Machine
Transverse
Puncture resistance:
Stress, Ib
Elongation, in.
Thickness, mils
Hardness, Durometer A points
5-sec reading
#17b
20
2,640
2,520
270
290
68
77
1,260
1,130
2,080
1,850
350
315
25.8
0.69
20
76A
#19
22
2,780
2,260
330
340
97
105
1,150
1,060
1,890
1,590
295
275
24.0
0.71
21
72A
membranes
#67
22
3,020
2,765
385
415
192
207
1,250
1,110
1,820
1,585
340
295
27.8
0.68
22
75A
#88
20
3,395
2,910
325
335
102
101
1,870
1,600
2,610
2,190
465
470
28.6
0.56
80A
Exposed membranes
#15
20
2,460
2,320
310
315
70
80
1,265
1,245
1,805
1,700
260
270
26.1
0.69
20
72A
#17
21
2,760
2,550
265
290
60
70
1,625
1,415
2,280
1,975
325
295
32.6
0.67
21
71A
Source: Reference 5, p. 50.
Liner number. Unexposed liner numbers are Matrecon inventory numbers; exposed liner
numbers are sludge lagoon sample numbers.
'Tests performed are tensile strength, ASTM D412; tear strength, ASTM D624; puncture
resistance, FTMS 101B, Method 2065; and hardness, ASTM D2240.
-------
size how critical field seaming is to a
successful lined impoundment.
Mechanical puncturing, inadequate pro-
tection from exposure to the weather and
poor field seaming contributed to the
deterioration and failure of the liner
material.
FIGURES LEGEND LIST
Figure 1 - Plan view of the lagoon showing
locations of sampling and the
field seam. Liner made of 2
sheets seamed on the field.
The sheets are folded in the
corners.
REFERENCES
1. Haxo, H.E. Effect on liner materials of
•long-term exposure in waste environments.
In: Proceedings of the Eighth Annual
Research Symposium: Land Disposal of
Hazardous wastes. EPA-600/9-82-002,
U.S. Environmental Protection Agency,
Cincinnati, Ohio, 1982.
2. Emcon, Inc. Field verification of lin-
ers. U.S. Environmental Protection
Agency, Cincinnati, Ohio (in press).
3. Matrecon, Inc. Lining of waste impound-
ment and disposal facilities. SW-870
revised edition, GPO No. 055-00000231-2,
U.S. Environmental Protection Agency,
Washington, D.C., 1983. 448 pp.
4. Haxo, H.E. Testing of materials for
use in lining waste disposal facilities.
In: Hazardous Solid Waste Testing,
First Conference. R.A. Conway and B.C.
Malloy, eds. ASTM Special Technical
Publication 760. American Society
for Testing and Materials, Philadelphia,
Pennsylvania, 1981.
5. Haxo, H.E., R.S. Haxo, and T.F. Kellogg.
Third interim report: Evaluation of
liner materials exposed to leachate.
EPA-600/2-79-038, U.S. Environmental
Protection Agency, Cincinnati, Ohio,
1979.
Figure 2 - Idealized cross-section of
lagoon showing depth at which
samples were taken. Samples
1-6 were taken in August 1982
and Samples 7-18 in September
1982.
Figure 3 - Percent extractables of exposed
samples arranged in
approproximate order from
lagoon top to bottom.
Weather-exposed samples are
arranged in decreasing
severity from south facing to
west, east, then north facing.
Figure 4 - Percent elongation at break
values for exposed samples
arranged in approximate order
from lagoon top to bottom.
Weather-exposed samples are
arranged in decreasing
severity from south facing to
west, east, then north
facing. Data are the averages
of values obtained in machine
and transverse directions.
Figure 5 - Stress at 100% elongation in
psi for exposed samples
arranged in approximate order
from lagoon top to bottom.
Weather-exposed samples are
arranged in decreasing
severity from south facing to
west, east, then north
facing. Data are the averages
for values obtained in machine
and transverse directions.
187
-------
EVALUATIONS OF TIME-DOMAIN REFLECTOMETRY AND ACOUSTIC
EMISSION TECHNIQUES TO DETECT AND LOCATE LEAKS
IN WASTE POND LINERS
J. L. Davis
M. J. Waller
B. G. Stegman
R. Singh
EarthTech Research Corporation
Baltimore, MD 21227
ABSTRACT
There is a great need to find reliable, non-destructive monitoring techniques to determine if a
significant leak exists in hazardous waste impoundment liners before the leachate seriously
damages the groundwater regime. There is also a need to find techniques that can econom-
ically locate precisely the area where a leak is occurring in the liner material.
The phase one part of this contract identified and evaluated a wide range of techniques that
might be applied to leak detection in waste impoundment liners (Waller and Davis, 1981). Two
of the techniques recommended for immediate further study and evaluation were acoustic
emission monitoring and time-domain reflectometry.
This paper describes the laboratory and field studies carried out by EarthTech Research
Corporation to evaluate the potential usefulness of acoustic emission monitoring and time-
domain reflectometry for detecting and locating leaks from waste pond liners.
The acoustic emission monitoring experiments showed that acoustic sounds, at fre-
quencies up to 500 Hz, are emitted when water flows turbulently at rates between 0.3 and 1
cm/sec through a sand and pea gravel. Turbulent flow may occur in the sand at flow rates
down to 0.1 cm/sec. The amplitude of the acoustic sounds is 100 times greater than the
ambient background noise level that is likely to occur at a liquid impoundment pond. The
acoustic sounds, at a frequency of 450 Hz, have an attenuation of 0.5 dB/m at a depth of one
meter in a lake. Using the existing equipment we might detect and determine the location of
leaks with a flow rate of 1 cm/sec at a range of about 3 m from a leak.
The time-domain reflectometry experiments showed that a leak with dimensions equal to or
larger than 0.5 the spacing between the parallel transmission line conductors can be detected
by this technique. The maximum conductor spacing tested was 2 m, but larger spacings
between the conductors are feasible.
Further tests in the field under controlled leak conditions are necessary to determine the
potential of both the acoustic emission monitoring and the time-domain reflectometry tech-
niques for detecting and locating leaks in a waste pond liners.
INTRODUCTION increasingly threatened by poor management
practices both in its removal for industrial
Historically the groundwater in this and agricultural use and by a lack of
country has been considered to be too control over waste disposal practices.
plentiful to be damaged by man. Over the Contaminants that leak from waste disposal
past decade, the groundwater has been sites percolate into the groundwater and
188
-------
eventually into our water-supply reservoirs.
Federal regulations over the past ten years
ha^e resulted in the reduction of waste
Discharge into surface waters, but the land
k,as become a major depository of hazardous
waste materials.
Each year at least 35 million metric
tons of hazardous wastes are put into
impoundments lined with natural or
man-made materials. The liners act as
barriers to the movement of the hazardous
materials out of the site into the ground-
water, but there is little evidence at
present as to the effectiveness of the
different liner materials as barriers. At
present we monitor the groundwater quality
from wells placed hydrologically down
stream. Thus the hazardous materials must
be in the groundwater before they are
detected. It is clear that significant damage
to the groundwater may occur before the
waste materials are detected.
At present there are no proven
methods that can determine the source of a
leak in a liner before it reaches the
groundwater. There is a great need to find
reliable non-destructive monitoring techni-
ques to determine if a significant leak
exists in an impoundment liner before the
leachate seriously damages the groundwater
regime. There is also a need to find
techniques that can economically locate the
area precisely where a leak is occurring in
the liner material.
The Phase One part of this EPA
contract identified and evaluated a wide
range of techniques that might be applied to
leak detection in waste impoundment liners
(Waller and Davis, 1981). Two of the
techniques that were recommended for
immediate further study and evaluation were
acoustic emission monitoring and time-
domain refiectometry.
ACOUSTIC EMISSION MONITORING
Acoustic emission monitoring (AEM)
techniques have been used for at least 50
years to evaluate stresses in rocks and
metals and also to monitor earthquakes
around the earth. Its application to soil
structures is more recent and especially for
monitoring fluid flow through soils. The
technique has been used to measure tur-
bulent flow of water through earth dams as
reported by Esmiol (1971), the Bureau of
Reclamation (1971), Crook and Coxen (1976),
and Koerner et al. (1981).
Koerner et al. (1981) have carried out
experiments in the laboratory showing that
water flowing through soils does emit sound
waves. We also reported similar findings at
last year's symposium. We feel that sounds
are emitted when water flows turbulently
through soils. Normally, turbulent flow does
not occur when water flows through soils,
but if there is a sufficient head or poten-
tial, as in a liquid waste impoundment site,
non-Darcian flow in a permeable soil can
occur. Thus acoustic emission monitoring
techniques might have application for
detecting the source of a leak from a waste
impoundment site. Possible rapid attenuation
of the sounds produced by the leachate
flowing through the soils and high ambient
background noise levels will make the AEM
technique impractical in a solid waste site.
It may be practical, however, in a liquid
impoundment site where a leak of hazard-
ous materials is potentially much more
hazardous.
The objectives of the acoustic
emission monitoring aspect of this project
are to determine the spectral characteristics
of the sounds emitted when water flows at
different rates through different soils. The
attenuation rates of these sounds in dif-
ferent materials are also to be determined.
TIME-DOMAIN REFLECTOMETRY
Time-Domain Refiectometry (TDR) is
less well-known and certainly more recent
than acoustic emission monitoring. Time-
domain refiectometry is a wide-frequency
bandwidth short-pulse length measurement
technique that is sensitive to the high
frequency (10^ to 10^ Hz) electrical prop-
erties of the materials in and around the
conductors of a parallel transmission line.
Typically, the TDR technique has been used
by electrical and communication engineers to
determine the condition of their transmission
lines and the location of any faults in the
lines.
Over the past decade the TDR tech-
nique has been shown to have practical
application for measuring and monitoring the
volumetric water content in soils indepen-
dant of soil type, density, and temperature
(Topp, Davis and Annan, 1981 and 1982 [a]
189
-------
and [b]). The high frequency electrical
properties of liquid water are about twenty
times greater than the electrical properties
of dry geologic materials. The Phase II,
Part 2, report to EPA reviews the TDR
technique as applied to geologic materials.
Leak materials from waste impound-
ment site liners, in the soil below, are likely
to have different electrical properties than
those of the host soil material. TDR can
measure the electrical property variations in
the material along a transmission line and
thus the TDR technique offers significant
potential of detecting leaks from waste
dump sites that have transmission lines
placed under the liner.
The TDR technique is similar to the
better known DC grid technique in that
wires must be placed under the liner, but
that is the extent of the similarity. The DC
grid technique requires wires to be laid in a
cross-grid under the liner. The leachate
must corrode the wires and thus break an
electric circuit before detection occurs. Two
wires running perpendicular must be corroded
before the location of the leak can be
determined. The TDR technique only
requires that parallel wires in one direction
be placed under the liner. The TDR system
actually detects the leachate between the
wires and not just at the wires. If a wire
should happen to break, then the TDR system
can still be used whereas this is not the
case in the DC grid system. Advantages of
the DC grid technique compared to the TDR
technique are that longer lengths of wires
can be used and the instrumentation is less
complex. These advantages occur at the
cost of much less information.
The objectives of the TDR aspect of
this project are to determine the size of an
anomaly that simulates a leachate in a soil
that can be detected as the transmission
line conductor spacing is varied.
This paper will briefly describe the
method we used to determine the potential
usefulness of the acoustic emission monitor-
ing and time-domain reflectometry for
detecting leaks from waste impoundment site
liners before the leachate reaches the
groundwater. The results of the experiments
will be given and discussed and then
conclusions will be drawn and recommenda-
tions offered on each technique. The AEM
experiments will be discussed first and the
TDR experiments next.
AEM - EXPERIMENTAL METHOD
The laboratory experiments were
designed to determine the amplitude and
frequency characteristics of sounds produced
when water flows at different rates through
soils. The test arrangement as shown
schematically in Figure 1 consisted of a 4 m
polyvinylchloride (PVC) duct with a diameter
of 30 cm in which water could flow at
different rates through different soils at
various lengths. A piezoelectric hydrophone
was placed in the soil and the sounds caused
by the water flowing through the soil were
amplified, filtered, processed and displayed,
as shown in Figure 2. A digital spectrum
analyzer capable of carrying out a fast
Fourier transform was used to process the
data and display the amplitude and fre-
quency characteristics of the sounds.
A great number of tests were carried
out to determine the spectral characteristics
of the equipment and the noise produced
when water flows at different rates through
the apparatus having no soil. Great care
was also taken to reduce any unwanted
sounds from all sources other than from the
water flowing through the soil itself. The
methods used are discussed in the Phase II,
Part 1, report submitted to EPA September,
1982.
Experiments were also carried out to
determine the attenuation of sounds over the
frequency spectrum observed in the labora-
tory experiments. Most of the attenuation
measurements were carried out in the field,
either in a lake or in soils. The acoustic
sensor was moved continuously away from
the acoustic source, rather than being
stationary. The same electronic receiving
equipment was used as described above, plus
an audio frequency sweep oscillator and a
loud speaker were used.
Two soils were tested in the labor-
atory. The first was a 20/30 Ottawa sand
with an average grain size of 0.7 mm. The
second soil was a pea gravel in which 90
percent of sample fell between 2.5 and
5 mm, with the remaining portion between 5
and 1 mm.
The height of the soil column was
varied at three levels - 70,100, and 150 cm.
190
-------
30cm Diameter
PVC Duct
4m Long
Water Level
Site Tube
To Amplifiers, Filters & Display
ACOUSTIC
INSULATOR
WATER
SOIL
< Microphone
?(•— 145 Mesh Grid
Funnel
TO RESERVOIR
Figure 1. A schematic diagram of the experimental test
arrangement.
WESTON AEM
I '
L
AMP.
--T-
_i_ j
BANDPASS
FILTER
1
1
J
NICOLET
SPECTRUM
ANALYZER
X-Y
PLOTTER
RECORDER
MICROPHONE
(HYDROPHONE)
Figure 2- A block diagram of the electronic equipment used to detect the acoustic
sounds emitted when water flows through soils.
191
-------
The microphone was placed 15 cm above the
bottom of the soil column in each test. The
water level varied from about 3 m above
the soil to just above the soil in each of
the experiments, thus falling water head
levels occurred during the tests. The flow
velocities in the soils varied from 1 to
0.3 cm/sec which is equivalent to 10 GPM
to 3 GPM in the 30 cm diameter column.
AEM - RESULTS AND DISCUSSION
Water Flow Rate Measurements
Figure 3 shows the water flow rates
versus head level when there was no soil in
the column and when there is sand at 70,
100 and 150 cm length in the column. The
flow rates measured for each condition
measured in the column were repeatable to
+/-GPM.
All the flow rates are essentially
linear between 10 and 4 GPM in the 30 cm
diameter column. The flow through the
soils is turbulent at these flow rates. The
flow becomes laminar at flow rates of less
than 1 GPM through the soils in the
column. We, therefore, would not expect the
linear response that we observe between 4
and 10 GPM on Figure 3. The reason for
the observed response is that the flow rates
are being primarily limited by the plumbing
of the test apparatus and not by the soils in
the column itself. We did not carry out a
detailed analysis of the elements restricting
the flow in the apparatus.
Acoustic Amplitude in Pea Gravel and Sand
vs. Flow Rate
Figure 4 shows the amplitude of the
sounds produced when water flows through
70 cm of pea gravel and 70 cm of
uncompacted sand and 70 cm of compacted
sand. The background noise amplitude when
water flows through the column with no soil
is also shown.
A significant portion of this back-
ground noise is produced when the water
flows turbulently in the narrow section of
the funnel. The flow in the open 30 cm
diameter column is laminar, but the velocity
increases by about 200 times in the narrow
part of the funnel and the flow becomes
turbulent.
We note that there is a significant
increase of sound amplitude as we pass
water through the pea gravel compared to
uncompacted sand. Little or no significant
sound is produced when water passes through
the well-graded compacted sand compared to
the system background noise level. We
shall discuss the significance of these
observations in the following paragraphs.
Source of Acoustic Sounds
The above results in the compacted
and uncompacted sand lead us to postulate
that the movement of the soil grains with
respect to each other is the major source of
acoustic sounds. In the compacted sand the
grains are placed tightly together and this
cannot move.
The significant noise amplitude
observed when water passes through the pea
gravel which has a wide variation of grain
sizes also points to the above conclusion
that the grains hitting each other is a major
source of acoustic sounds. It seems likely
that the finer grain particles are moving
through the larger pore spaces set up by the
larger grain particles and striking other
particles.
We also note from Figure 4 that when
water flows through the pea gravel or
uncompacted sand the sounds fall into the
background noise level at about 4 GPM in
the 30 cm diameter column. This flow rate
is about 2 to * times greater than that
which we calculate to be necessary for
laminar flow in the soil. It is obvious that
acoustic sounds can only occur when the
flow is turbulent.
Acoustic Spectra in Soils vs. Flow Rate
Figure 5 shows the spectra of the
sounds emitted when water flows at dif-
ferent rates through the pea gravel. The
frequency spectra for the compacted and
uncompacted sand is essentially the same,
only the amplitude is different. No sounds
above the background noise level were
detected at frequencies above 1 KHz up to
20 KHz. The low frequency roll-off of
acoustic sounds is primarily due to the low
frequency filter function of the equipment
which had a cut-off frequency of about
80 Hz.
192
-------
D 70 cm oŁ sand
A 100 cm of sand
• 130 CD oŁ sand
TOTAL HEAD, (METRES)
9 10
FLOW RATE,CPM
193
-------
We note from Figure 5 that most of
the sound energy lies at or below 500 Hz
with a peak at 200 Hz. Part of frequency
response observed is due to resonant effects
of the PVC column. We are primarily
interested in the rate of change of ampli-
tude of the sounds and this is a maximum
at about 200 Hz. We do not have at
present an explanation for this frequency
response caused by water flowing through
soils.
Assuming that a similar response is
likely to be observed in the field means
that we can reduce the bandwidth of our
monitoring system and thus improve our
sensitivity to the sounds produced by water
flowing through the soil and reduce the
amplitude of the background noise level.
Simply this means that we can improve the
response from our wanted signal relative to
the unwanted background noise level signal.
Figure 6 shows the acoustic amplitude
at 300 Hz when water flows at different
rates in the 30 cm diameter column through
70, 100 and 150 cm of uncompacted sand.
We note that there is a significant increase
in sound amplitude as the length of the sand
is increased. (This result was essentially the
same at frequencies above 150 Hz for all
the soils tests). Ln contrast there is no
change in the amplitude of the sounds
emitted at frequencies below 150 Hz when
water flows through different lengths of
sand.
It appears that as the number of
sound sources increases, i.e., the height of
the soil column increases, the greater the
sound amplitude at the microphone. Further,
this indicates that the attenuation of sounds
at frequencies between 200 and 500 Hz
must be low. The attenuation may increase
at frequencies below 150 Hz where the
amplitude of the sounds does not vary due
to a significant increase in the attenuation
characteristics at these frequencies. This
result is not expected from our experience
with seismic sounding and we have no
explanation for our observations at this
time.
Attenuation Measurements in Water
Attenuation measurements of sound
waves were carried out in a lake. The
loudspeaker which was powered by an audio
frequency sweep oscillator operating at
frequencies from 20 to 500 Hz was moved
away from the hydrophone.
Figure 7 shows how the sound
amplitude at 400 Hz decreased as the
separation between the loudspeaker and the
hydrophone was increased. The top line
shows the response if there is no attenuation
and only the decrease of amplitude due to
spherical spreading is considered. The
slope of the lower line shows that the
attenuation in the lake water was 0.5 dB/m.
This attenuation value is about five orders
of magnitude greater than the attenuation
value expected for sound waves at 100 Hz
in water. We suspect that much of the
energy is being lost at the water surface.
The lake was 2 m deep and our tests were
carried out at a depth of 1 m. Further, we
noticed that the attenuation increased to
about 1 dB/m when the tests were carried
out at a depth of 0.2 m. The tests in the
lake water do tend to simulate the condi-
tions that might be found in a liquid waste
impoundment site.
The background noise level observed
in the lake which was located in the city
of Columbia, MD, was 0.1V increasing to
0.5 V from time to time depending on the
cultural activity in the area and the waves
on the lake surface. The observed back-
ground noise level is about the same as that
observed in the laboratory. These background
noise levels should be similar to the
background noise levels that will be found
at a waste disposal pond site.
The results of the experiments carried
out in the laboratory and in the field show
that acoustic sounds can be detected at
frequencies below 500 Hz when water flows
at velocities of greater than 0.3 cm/sec in
coarse-grained soils. The amplitude of the
sounds emitted at these velocities will be
equal to or greater than the background
noise levels likely to be found at most field
sites. The attenuation of the sounds emitted
when water flows turbulently through soils
was not determined explicitly in this study,
but a range of up to 3 m the leak and the
acoustic sensor should be possible in the
field. The range may be increased with the
aid of signal processing techniques such as
background removal.
194
-------
I I I ( I I I 1 I
0 0
456
FLOW RATE, 6PM
Figure 6. Shows the acoustic amplitude at 300 Hz when water flows through the 30 cm
diameter tube at different rates through 70, 100 and 150 cm of sand The
background noise level is shown as a reference level
50
40-
30-
ATTENUATION AT 450 HZ.
SPHERICAL SPREADING
SPHERICAL SPREADING
ZERO ATTENUATION
0.5
ATTENUATION - 1 5 dB/FT
(0.5 dB/H)
I 2
SEPARATION, FEET
I 1 I I ' > I I 1 I
4 5 6 7 8 9 10 12 15 2O
Figure 7. Attenuation of Sound in water at 450 HZ.
195
-------
AEM - CONCLUSIONS
There are a number of conclusions
that can be made based on the results of
this work. The most significant conclusions
are that:
I. Acoustic sounds are emitted
when water flows turbu-
lently through soils,
2. The amplitude of the
acoustic sounds increases as
the flow rate increases,
3. The amplitude of the sounds
emitted is equal to or
greater than the amplitude
of the background noise
level likely to be found at
a liquid impoundment site
at a flow rate of 0.3
cm/sec.,
4. For the same flow rates the
amplitude of the sounds
emitted increases as the
grain size variation in the
soil increases,
5. For the same flow rates the
amplitude of the sounds
emitted decreases as the
density of the soil increas-
es,
6. There is significant power in
the sounds emitted at fre-
quencies up to 500 Hz with
a peak amplitude at about
100 Hz when water flows
turbulently through soils,
and,
7. The attenuation of sounds at
400 Hz is 0.5 dB/m at a
depth of 1 m below the
water surface in a lake.
These conclusions indicate that
acoustic emission monitoring of sounds is a
potential method for detecting the source of
a leak from a liquid waste impoundment
site.
Further work is necessary to verify
the usefulness of the acoustic emission
monitoring technique under controlled
conditions at a model pond site where leaks
can be simulated and controlled. Experi-
ments at such a pond will help us to
determine the range of leak rates that can
be detected acoustically and also the
practical range in the pond that a leak can
be detected. Tests are also necessary at the
existing liquid waste impoundment pond to
monitor the background noise levels at
different locations over long periods of
time. After these tests it should be
possible to give a practical design of an
acoustic emission monitoring system for
detecting leaks in liquid waste impoundment
ponds.
TDR - INTRODUCTION
The TDR experiments were designed
to determine the size of an anomaly,
simulating a leachate plume in soil that
can be detected using parallel transmission
lines with various spacings between the
conductors.
There are many factors that affect the
high frequency electrical properties of
materials measured by the TDR technique.
Some of these factors are:
1. The magnitude of the change of
electrical properties of the anomaly
compared to the host material along the
transmission line,
2. The physical size of the anomaly
in the transmission line,
3. The electrical loss properties of
the host material in the transmission line,
*. The source power and receiver
sensititivity of the TDR electronics,
5. The TDR pulse length or the pulse
rise time in the case of a step pulse, and
6. The spacing of the parallel trans-
mission line conductors.
The first three factors listed are
related to the electrical properties of the
materials in the ground and the last three
factors are related to the TDR equipment.
It is clear that there are a number of
factors interrelated to each other that
affect the TDR response. In our studies to
determine the usefulness of the TDR
technique to detect leaks from waste site
liners we can hold some of the geologic and
equipment factors such as Numbers I, 3, 4,
and 5, constant and determine the TDR
responses as Numbers 2 and 6, i.e., the size
of the anomalies and the spacing of the
conductors in the transmission line are
varied.
We shall describe the experimental
methods used in the laboratory and in the
196
-------
field to evaluate the TDR technique for
detecting simulated leaks in soils. The
results of the experiments will be given and
discussed, and a number of conclusions will
be made.
TDR - EXPERIMENTAL METHOD
A schematic diagram of the equip-
ment used is shown in Figure 8. The
electronic equipment used for these tests
was a commercially available Tektronic's
Model 1502 cable tester. This is a portable
field unit that has a pulser with a pulse
step rise time of 10~^ seconds, a sampling
receiver with a sensitivity of 2x10"^ volts
and a display unit using either an interval
oscilloscope, or an internal graphic recorder
or with an output for an external X-Y
plotter.
The balanced parallel transmission
lines used had several different types and
sizes of conductors depending on the
location of the experiments. The conductor
material and diameter was found to have
little effect on the TDR response. We
wanted to determine the effect on the TDR
response when the spacing between the
conductors was varied. The spacing that at
which we ran tests varied in increments
over the range of 10 to 200 cm.
A special connector arrangement to
connect the TDR unit to the conductors at
the different spacings was designed to get
the maximum signal transfer from the TDR
unit on and off the transmission line. The
connector assembly consisted of a 50 ohrn to
200 ohm Balun transformer and then two
metal sheets to connect the transformer to
the transmission line conductors. The metal
sheets were designed to provide a gradual
impedance transition from the transformer
to the transmission line at the different
spacings.
The test setup for the transmission
lines in the laboratory consisted of 6 m
long, 5 cm diameter aluminum tubes being
supported on a wooden frame on which the
conductor spacings could be easily varied
from 10 cm to 200 cm. The leaks were
simulated by using boxes(15 cmp filled with
dry sand. The number of boxes and their
position relative to the conductors could be
varied.
In the field the test setup consisted of
a box 2.5 x 2.5 m filled 60 cm deep with
sand. The transmission line conductors were
made from 1.2 cm diameter copper tubing
placed at fixed spacings of 10, 20, 30, 50,
100, 150 and 200 cm. The transmission
lines were placed about 45 cm below the
sand surface. The leaks were simulated by
adding about 1 liter of water every ten
minutes. One liter of water saturates
approximately a (15 cm)3 volume of sand.
The size of the anomalies used in the field
were therefore similar to the boxes used in
the laboratory. Both these simulations were
equivalent to a leachate saturated soil in a
soil host material having a volumetric water
content of 20 percent.
The electrical losses in the air and in
the sand were very low to the TDR pulse
signal used. We feel that this arrangement
was a satisfactory "first try" model of an
actual field setup of a TDR leak detection
system placed under a waste pond liner. We
shall continue by presenting and discussing
the results of the test carried out.
TDR - RESULTS AND DISCUSSION
A number of tests were carried out
to determine the size of a simulated leak
that could be detected as the separation
between the conductors of the transmission
line was varied. The first series of tests
measured the effect that increasing the
conductor spacing had on the TDR pulse rise
time. Next we determined the change in
the TDR response as the length and height
of the anomaly in the transmission line was
varied. Finally, we looked at the variation
of the TDR response as the width and the
position of the anomaly were varied within
the transmission line.
TDR Pulse-Rise Time vs. Conductor Spacing
Increasing the spacing between the
conductors of a parallel transmission line
typically causes increased attenuation,
especially to the higher frequencies in the
pulse. We carried out a series of tests to
determine just how the pulse-rise time
increases as the transmission line conductor
spacing increases. Figure 9 shows how the
TDR signal response time at the open
circuit end of the transmission line increases
as the separation between the conductors
increases. We note that the relationship is
linear over conductor spacing variations of
197
-------
10 to 200cm for a step pulse with an input
rise time of 10'10 seconds, and thus having
frequencies with wavelengths down to 10cm
in air. The increasing pulse rise time
response means that the size of an anomaly
in the line must also increase. This
experiment indicates that the pulse-rise time
increases linearly as the conductor spacing
increases over a wide range relative to the
wavelengths in the signal. A similar
response was observed when the transmission
lines were placed in moist sand.
Physical Size of Anomaly vs. Conductor
Spacing
This series of tests was carried out
to determine the change to TDR response as
the length of the anomaly along the
transmission line and the height of the
anomaly in the direction perpendicular to
the plane along the conductors was varied.
Figure 10 shows a plan and side views of
the anomaly in the transmission line.
Figure 1 1 shows four TDR data
responses obtained with a conductor spacing
of 40 cm, and the length of the anomaly
along the line is increased from 15 cm to
90 cm (1 to 6 boxes). The height and
width of the anomaly was kept the same.
The top TDR data trace is that observed
when there is no anomaly in the trans-
mission line. Note that as the anomaly
length is increased from 15 to 30 cm, both
the amplitude and time extent of the
response increases. Further increase of the
length of the anomaly only increases the
time extent of the TDR response, and not
the amplitude.
Figure 12 shows the increase of the
time duration of the TDR response as the
length of the anomaly is increased. The
open circle line is the response observed
when the height of the anomaly is equal to
the spacing between the transmission line
conductors and the solid dot line is the
response observed when the height of the
anomaly is increased to 1.5 times the line
spacing. There is little variation in the TDR
response when the height of the anomaly is
increased. There is a significant increase in
the time duration of the response as the
length of the anomaly along the line
increases.
It is more useful at this stage to
reference the size of the anomaly to the
transmission line conductor spacing, and
Figure 13 shows how the TDR signal ampli-
tude varies compared to the length relative
to the line spacing. The laboratory trials
are shown with a dashed line while the field
trials are shown by the solid line. For the
laboratory trials, the width and height of
the anomaly was held equal to the conduc-
tor spacing.
Width and Position of the Anomaly in a
Transmission Line
In the first series of these experi-
ments we kept the width of the anomaly
equal to the spacing between the conduc-
tors. We shall now keep the length and
height of the anomaly constant relative to
the spacing, and vary the width of the
anomaly and the position of the anomaly
relative to one of the conductors.
Figure 14 shows how the TDR signal
amplitude varies as the width of the
anomaly relative to the spacing varies for
two cases first when the anomaly is
centered between the conductors (X/S=0.5),
and then when the anomaly is centered
about one of the conductors (X/S=0).
We observe that the TDR signal
amplitude does not decrease as rapidly when
the anomaly is centered about a conductor
as the signal amplitude does when the
anomaly is centered between the conductors.
The response is greatest when the anomaly
is touching both conductors.
Figure 15 shows how the amplitude of
the TDR response varies as the length of
the anomaly relative to the spacing varies
for anomalies that have a width equal to
one-third the spacing located at different
positions relative to the conductors. Once
again we note a very significant decrease of
the TDR signal amplitude as the boxes are
moved away from a conductor. As ex-
pected, the TDR response as the length of
the anomaly along the transmission line is
increased is very similar to that shown in
Figure 13 even though the anomaly is
narrower than the conductor spacing. The
tests carried out in the sandbox have similar
results, and the field data is shown as a
solid line on Figure 13. The agreement is
quite good.
These experiments show us that the
worst case for obtaining a good TDR signal
198
-------
TIME-DOMAIN REFLECTOMETER
CONNECTOR
PARALLEL
TRANAHISSION
LINE
Fijjure 8. A schematic diagram of the TOR System.
75 100
Spacing, cm
125 150
Figure 9. Shows how the relative slope of the TDK sigral from an open circuit
PLAN VIEW_
,,,..., zt T
TRANSMISSION"^ e"9 XIS ' | f •: } s
LINE ,_ I I- «---! g —_-
CONDUCTORS "" y" •> * °
/ I
L-><
Boxes ,
rr/i . {
L-U-J }
PROFILE VIEW
LO Plan and side views of the boxes filled with sand placed between the conductors of
a transmission line.
-------
oot
AMPLITUDE AXIS
a. <
is
rs
RELATIVE AMPLITUDE, PERCENT
RELATIVE DURATION
1 1-
-------
LENGTH = SPACING, S
HEIGHT = S/3
WIDTH/SPACING , y/S
WIDTH/SPACING=l/3
rflDTH/HElGHT =
H0° )•
C ( D Lost in the Noisel
response is when the anomaly lies between
the transmission line conductors without
actually touching either conductor. This is
an important observation. The TDR can
detect anomalies whose dimensions are at
least one-half the spacing between the
conductors. The TDR signal amplitude will
be equal to or greater than 20 percent the
maximum signal amplitude which occurs
when the transmission line penetrates an
anomaly with similar electrical properties,
but which is physically very large compared
to the transmission line conductor spacing.
It must be noted that this result is based1
upon the condition that the attenuation to
the signals along the line are negligible.
Further tests are necessary to determine the
actual effect when electrical losses occur
along the transmission line.
TDR - CONCLUSIONS
TDR is a versatile high resolution
technique for detecting variations of electri-
cal properties in the ground. Since the
electrical properties of leachates in soils
are likely to be different from the elec-
trical properties of the soil host material,
the TDR technique should be a good method
for detecting small leaks from waste pond
liners. There are many factors which affect
the TDR response when transmission lines
are placed in the ground. This study was to
determine how the size of a leak that can
be detected by the TDR technique varies as
the spacing between the conductors of the
transmission line is varied. Clearly we
should like to detect very small leaks using
as few transmission lines as possible at a
site.
A number of conclusions can be made
based on the TDR experiments carried out
in this project.
1. The TDR pulse signal rise time
increases as the spacing between the
201
-------
conductors of the transmission line increas-
es. In other words, the size of the leachate
leak between the conductors must increase
as the spacing between the conductors is
increased.
2. The electrical length of the
anomaly along the length of the transmission
line should be equal to or greater than the
TDR pulse length in the anomaly.
3. The TDR signal amplitude varies
slightly as the height, i.e., the dimension
whose axis is perpendicular to the plane of
the conductors of the anomaly in the
transmission line varies.
it. The amplitude of the TDR signal is
greatest when the anomaly touches both
conductors of the transmission line. The
signal amplitude decreases significantly as
the width of the anomaly relative to the
spacing of the conductors decreases. The
signal amplitude decreases less rapidly if the
anomaly is touching one of the conductors.
These conclusions lead us to general-
ize that a cubic shaped anomaly whose sides
are at least 0.5 the spacing between the
conductors will have a TDR signal amplitude
of about one fifth the amplitude which
would occur if the transmission line pene-
trates an anomaly which was very large
compared to the spacing of the transmission
line conductors. This generalization will be
affected by a number of factors that we
have not included in our studies.
The electrical loss properties of the
materials likely to be found in the field will
reduce the TDR signal amplitude; thus the
size of the anomaly will have to be in-
creased to increase the reflected TDR
signal amplitude. The electrical property
contrast of the leachate saturated soil
relative to that of the host soil under the
waste impoundment site may not be as
great as the electrical property contrast we
used in our models; thus the reflected signal
amplitude may not be as large in the waste
site environment. Acknowledge of the
electrical contrast between leachate in soils
and background moisture will be necessary
before a TDR system can be properly
designed.
The models we used in these
experiments only tested the TDR response in
transmission line whose maximum spacing
between the conductors was two meters.
Larger spacings between the conductors are
possible using the same TDR unit. Experi-
ments to determine the maximum spacing of
the conductors still need to be carried out.
We need to test the TDR system using a
series of different spacings between the
conductors of the transmission lines placed
in a sandy soil under a lined pond filled
with water. Controlled leaks in the liner
should be used to test the ability of the
TDR system to detect leaks of different
sizes, positions, and rates a round the trans-
mission lines. We also need to determine
the high frequency electrical properties of
some leachate saturated soils. The answers
to these recommendations will help us to
design a practical TDR monitoring system
for detecting leaks from waste pond liners.
REFERENCES
Crook, D. E., and Coxon, R. E., 1976
Proceedings of the Second Congress on Large
Dams, Q.45, R.30, Mexico, pp. 527-540.
Davis, 3. L., Singh, R., Stegman, B. G., and
Waller, M. 3., 1982. Innovative concepts for
detecting and locating leaks in landfill liner
systems, Part I: Acoustic Emission Monitor-
ing. Report to EPA, Cincinnati, Oh.
Contract No. 68-03-3030.
ibid; Part II: Time-Domain Reflectometry
Esimol, E. E., 1971. Field evaluation of the
micro-acoustic detector at Senator Wash
Reservoir. Report to U.S. Bureau of
Reclamation, Denver, CO, 30 July 1971.
Koerner, R. M., McCabe, W. M., and Lord,
A.E., 1981. Acoustic emission behavior and
monitoring of Soils. ASTM Symposium on
Acoustic Emission in Geotechnical Engi-
neering Practice, Detroit, Mich. Publication
No. 750, pp 93-141.
Waller, M. 3., and Davis, 3. L., 1981.
Assessment of innovative techniques to
detect landfill liner failings. Report to
EPA, Cincinnati, OH, Contract No. 68-03-
3209.
Topp, G. C., Davis, 3. L., and Annan, A.P.,
1980. Electromagnetic determination of soil
water content: Measurements in coaxial
transmission lines. Water Resources
Research, Vol. 16, No. 3, pp 574-582.
Topp, G. C., Davis, 3. L., and Annan, A.P.,
1982. Electromagnetic determination of soil
water content using TDR: I: Applications to
wetting fronts and steep gradients. Soil
Science Society of America 3ournal. Vol. 46.
No. 4. pp. 672-678.
ibid;II. Evaluation of installation and
configuration of parallel transmission lines,
pp 678-684.
202
-------
PILOT SCALE VERIFICATION OF A LINER LEAK DETECTION SYSTEM
Wendell R. Peters
David W. Shultz
Department of Geosciences
Southwest Research Institute
San Antonio, Texas
ABSTRACT
Flexible membrane materials have been used as liners at landfills and surface im-
poundments to inhibit fluids from contaminating surrounding water resources. These
materials will act as an electrical insulator in such installations. Therefore, any
leak in the liner will form a detectable electric current path by which a leak may be
detected and located. The primary objective of this project is to develop a method to
detect and locate leak paths in membrane liners.
A summary of the two- and three-dimensional electrical modelling of liners with
leaks, using computers and physical scale models is presented. Search techniques deve-
loped have led to hardware and planned pilot scale field studies. The system developed
for use at landfills and surface impoundments is discussed. Examples of equipotential
plots showing leak dependent surface effects are included.
INTRODUCTION
An important aspect of hazardous waste
treatment and disposal in landfills or sur-
face impoundments is the prevention of
surface and groundwater contamination by
fluids containing hazardous constituents.
Relatively impervious flexible membrane
liners have been used to establish the
facility boundaries and to prevent fluids
from migrating into the surrounding water
resources.
Research and evaluation projects are
underway to investigate the effectiveness
of waste containment liner materials,
including the long-term deterioration of
liners exposed to various waste products.
Early landfill liners consisted of clay
hardpans, and wood or metal barriers. Im-
proved liners have been developed more
recently using flexible polymeric materials
to provide lower liquid permeability and
longer waste containment lifetimes.
While these newer liner materials are
demonstrating improved resistance to pro-
longed waste exposure, no methods are
presently available to nondestructively
test their physical integrity in service
environments. A tear or rupture in a liner
system will allow fluid to migrate from the
facility and thus violate the original
intent of the liner. The work reported
herein has been directed towards developing
a nondestructive electrical system which
can be used for detecting and locating leaks
in waste containment liners while the faci-
lity remains in service.
Since the philosophy of impervious
liners is to contain rather than absorb or
filter contaminants, the physical charac-
teristics of the liner materials will
usually differ significantly from the
underlying soil and the contained hazardous
waste. In particular, liners made of im-
pervious plastics and rubbers will exhibit
very high electrical resistance which will
act as an electrical insulator between the
internal and external liner surfaces. If
the liner is physically punctured or
separated so that fluid passes through the
liner, the electrical conductivity of the
fluid and saturated underlying soil will
form a detectable electric current path
through the liner by which the leak may
be revealed. Electrical methods are
203
-------
attractive because they can be applied on
the surface of the landfill or fluid im-
poundment. The methods are completely non-
destructive in operation and, by means of
automated field equipment, can be made very
efficient and cost-effective for facility
monitoring operations. The resulting
method, when fully developed, will be use-
ful for surveying existing as well as new
facilities where electrically resistive
membrane liners are installed.
Two- and three-dimensional computer
models were used to examine the current
distribution in the cross sections of
several simulated liner geometries having
different leak locations. In addition to
these analytical model studies, a three-
dimensional physical scale model was also
constructed using a plastic liner in a
wooden frame. Soil was placed inside the
model to simulate a landfill. Water was
used to simulate a fluid impoundment. Elec-
trical potential distributions on the sur-
face of the soil and water were measured
to determine the effects of punctures in
the liner. The effects of multiple leaks
as well as those caused by subsurface
anomalies contained within the liner were
studied. Based upon these application
concept studies, the most promising elec-
trical testing methods were identified
and subsequent project activities were
directed towards the design and fabrication
of an appropriate experimental field in-
strumentation system. Computer software
has also been developed and tested to allow
on-site analysis of field data.
EXPERIMENTAL METHOD
Figure 1 shows the basic electrical
testing concept where one current elec-
trode is located away from the facility
and a current path exists through a leak
penetration in the liner as well as over
the buried edge of the liner. This figure
illustrates that when a leak penetration
is present in the liner, current flow be-
tween electrodes located inside and out-
side the facility will follow two paths,
namely, through the leak and over the
buried edges of the liner. Since surface
potentials are directly related to the cur-
rent distributions in the vicinity of the
search electrode, they can be used to
locate the leak.
The first numerical model study used
a general purpose circuit simulation com-
puter program (SPICE). Circuits containing
resistors, capacitors, inductors, and
voltage and current sources were simulated
with the program. A two-dimensional re-
sistor network model designed to simulate
a membrane liner was modelled using the
program.
Figure 2 illustrates a contour plot
showing the two-dimensional potential
distribution around a liner without leaks.
The current injection point is at the top
of the figure with the other reference
electrode connected to a conducting path
along the bottom and sides of the cross
section. The outline of the liner has
FIGURE 1. LEAKING LINER. ARBITRARY SEARCH
ELECTRODE POSITION
FIGURE 2. COMPUTER-MODELED VERTICAL CROSS-
SECTION OF WASTE LINER WITH
NO LEAKS
been sketched in the figure and is repre-
sented by the three-sided trapezoidal
figure in the center of the plots. The
equipotential lines showing the voltage
distribution patterns were computer gene-
rated. The current flow paths are at
right angles to the equipotential lines
and were sketched in by hand. Figure 2
illustrates the current flow over the edges
of the liner and into the surrounding earth.
204
-------
In parallel with the computer ana-
lysis work described above, a three-dimen-
sional physical scale model was constructed
in an outdoor environment where ground is
used as the soil underlying the physical
scale model. Figure 3 shows the model and
the instrumentation van.
Surface voltages were measured with
potential measurement electrodes located in
the basin. The potential reference elec-
trode was fixed in position near one corner
of the model. The exploration measurement
electrode was composed of many electrodes
mounted on a fiberglass beam at 6.35cm
(2.5 in.) spacings. Most of the data were
acquired using a polar grid with the cur-
rent injection electrode at the center.
The leaks in the bottom of the liner
were generated by driving a 1.524cm
(0.5 in.) diameter copper-clad steel rod
through the liner and into the soil. This
rod provided a good conducting path between
the water and the soil without loss of
water.
EXPERIMENTAL RESULTS
Figure 4 illustrates the results of
one of the computer generated two-dimen-
sional analyses of a liner having a leak
in the bottom. The equipotential lines
which appear in the figure were generated
and plotted automatically by the computer
program described earlier.
FIGURE 4. COMPUTER MODELLED VERTICAL CROSS-
SECTION OF A WASTE LINER WITH A LEAK IN
THE BOTTOM
This result shows that, as predicted,
when a leak penetration is present in the
liner, current flow between electrodes
located inside the outside the liner will
follow two paths, namely, through the leak
and around the buried edges of the liner.
Note the non-symmetry of the equipotential
lines terminating on the surface of the
landfill above the liner. The voltage
gradient is clearly steeper along the sur-
face on the side of the current injection
electrode above the leak. Asymmetries
such as were investigated in subsequent
work to develop field measurements which
could locate liner leaks.
Figures 5 and 6 show examples of
equipotential contours measured using the
physical scale model. For all plots, the
current injection point is located at the
center of the polar coordinate system.
In Figure 5, a single leak was located at
an arbitrary position in the bottom.
Figure 6 clearly shows the ability of the
technique to distinguish between multiple
leaks.
Many data acquisition runs were con-
ducted using different combinations of
water and soil depth, distance between cur-
rent inejction point, leak position, number
of leaks, and conductive anomalies in the
water. All of the tests resulted in similar
types of equipotential plots in which the
leak position could be accurately inferred.
To verify the system performance prior
to conducting a survey in the field, a full
scale verification program is being imple-
mented. A one-acre lined test facility has
been constructed on the SwRI grounds. A
two-phased effort is planned for testing
the system using this facility. Phase One
will involve testing the system with water
in the test facility. Phase Two tests will
use an earth fill over the liner. The
intent of each phase is to gather additional
data to predict the usefulness of the sur-
vey system at lined fluid storage impound-
ments and landfills. Details of both the
verification phases are presented below.
PHASE I - WATER FILL VERIFICATION
Figure 7 shows the pattern of leaks
which are being used in the bottom of the
impoundment. The leaks have been induced
in the liner by a short length of 12-inch
diameter pipe penetrating the liner. An
open pipe establishes a leak through the
liner; a closed pipe re-establishes the
integrity of the liner at the location.
For the proposed tests outlined below, no
more than two leaks will be opened at a
time.
205
-------
FIGURE 3. THE PHYSICAL SCALE MODEL WITH MEASUREMENT ELECTRODES IN PLACE.
-------
LEAK
12-29-81
FIGURE 5. EQUIPOTENTIAL PLOT OF VOLTAGES ON THE SURFACE OF AN
EARTH FILLED LINER WITH A SINGLE LEAK
CURRENT INJECTION
1-5-82(1)
FIGURE 6. EQUIPOTENTIAL PLOT OF VOLTAGES ON THE SURFACE OF AN
EARTH FILLED LINER WITH TWO LEAKS
207
-------
o
m
IMPOUNDMENT
FLOOR
7
\
\
\ \
\ \ \
\ \ \
\ \ \
•L31
\ \ \
©-
/
\ ^ / /
\\\\ I //
Nv\N\\l///
* Current Injection Points
© Liner Leaks
FIGURE 1. CURRENT INJECTION AND LEAK POINTS FOR THE ELECTRICAL RESISTIVITY SURVEY
208
-------
The leaks to be studied are labeled
in the figure as LI, L2, L3, L4, and L5.
Leak LI is selected to be in the center of
the model and leak L2 is half way between
the center and the edge. L3 and L5 were
chosen to study the effects of leaks near
the edge of an impoundment. Leak L4 was
chosen primarily to be paired with leak
L2. When viewed from position "A", Leaks
L2 and L4 are separated by an angular dis-
tance of ten degrees. Linearly they are
13 feet apart.
Current injection points near the edge
of the impoundment floor are labeled "A",
"B" and "C". Each current injection point
will be used to study the effects of all
leak points taken one at a time. Leak
point L2, L4, and LI, L4 will be studied
as pairs, simulating two leaks in close
proximity to each other.
The surface potentials will be
measured using a single electrode on a
winch mounted cable. A winch mounted
odometer is used to monitor the distance
to the measurement electrode cable. As
the measurement electrode is pulled across
the surface of the impoundment, the com-
puter used in the data acquisition system
will trigger potential measurement based
on the preprogrammed recording increment
and the odometer readings. The data will
be acquired on a polar grid with ten
degree increments between radials. For
data acquisition and for plotting the re-
sults, the current injection electrode is
the center of the polar grid.
Polar data for all of the leak com-
binations listed above will be taken at
current injection points. All measurements
will be recorded at water depths of one,
three and five feet. Thus for the seven
leak orientations, the three current in-
jection points and the three water depths,
a total of 63 data acquisition runs will be
made.
Software written for the HP9825 field
desktop computer will provide real time
data processing, digital recording and
graphic display for the field data. To
provide hard copy graphics information,
a HP7245 thermal plotter/printer will be
used with the computer. The computer in-
terfaces to the odometer and signal con-
ditioning circuits through a multiprogrammer.
Contained in the multiprogrammer are circuit
modules to interface with the odometer
electronics and a 12 bit A/D converter for
digitizing the potential measurements.
PHASE 2 - SOIL FILL VERIFICATION
For this phase, the impoundment will
be drained and filled with two feet of soil.
Leak paths LI, L2, and L4 which were used
in the water study will also be used with
the soil fill. As with the water study,
the leaks will be induced in the liner by
a short length of PVC pipe penetrating
the liner. To insure electrical continuity
through the pipe, a solid core of moist
earth will be maintained in the pipe.
In the soil fill study, leak points
LI and L2 will be studied as single leaks.
L2 together with L4 will be used to study
a liner with two leaks in close proximity
to each other. Current injection points
"A", "B", "C" will again be used as the
center of the polar data acquisition
array.
A cable system better suited to working
on land will be used to collect data in the
soil fill study. To measure soil potentials,
a cable system with 24 electrodes and pre-
amps will be spread out in an equally
spaced linear array across the surface.
Circuits in the data acquisition system will
address the preamps one at a time to read
and record all potentials. After recording
data from one position of the cable, all
cables and electrodes are moved to the next
position on the same type of polar grid
used to record the data over water. Ten
degree increments will be used between
radials.
Polar data for all leak combinations
listed above will be taken at current in-
jection points "A", "B", and "C". Only
one soil depth of two feet will be used
however. The number of leak combinations
and current injection points will result
in nine complete data cquisition runs.
The software for the cable data acqui-
sition system will be slightly different
than that used with the single electrode
winch system. After acquiring the data
and recording it, however the real time
data processing and display is identical
to that used for the water impoundment
study.
209
-------
FIELD TESTING OF THE SYSTEM
location.
After successful completion of these
tests, Southwest Research Institute plans
to locate at least two actual field sites
for additional testing.
The field tests conducted at landfill
and surface impoundment facilities will
likely duplicate the test procedures out-
lined above. Slight modifications may be
required, depending upon site conditions.
The same cable systems, hardware and soft-
ware will be used that has been developed
and tested in the field scale model.
In selecting test sites, technical
factors for consideration are the size
including depth and surface area, the type
of liner material and current waste manage-
ment procedures such as sump removal of
accumulated leachate. Since the method
depends upon a conductive path through the
liner, removal of leachate could remove
the desired electrical shunt. A minimum
of one landfill and one fluid impoundment
will be selected.
Other parameters to be considered are
the age of the site, type of waste material
deposited in the landfill, the resistivity
nonuniformity of the fill volume, and the
type of fluid stored in the case of an im-
poundment .
Information on known or suspected
liner leaks will be helpful in verifying
the survey results since purposely intro-
duced liner penetrations will probably not
be practical or permissible.
As with the scale field model, data
will be acquired on a polar grid pattern.
In the full scale field testing, the angular
spacing between radials and the linear
spacing between potential measurements
will be determined by data from the test
facility and the depth of the landfill or
surface impoundment selected for survey.
Realtime data processing and display
will help dictate the direction and pattern
of the search over the landfill or surface
impoundment. A typical survey will enable
the field crew to quickly cover the com-
plete waste liner surface to rapidly
identify and tag zones with possible leaks.
When such a zone is found, the survey reso-
lution will be increased to provide the
maximum delineation of the susepcted leak
The desktop computer, portable
generator current source, signal condi-
tioning electronics and all other supporting
equipment will be transported to the test
site in a Department of Geosciences Mobile
laboratory.
CONCLUSIONS
Project results to date have demon-
strated the ability of the electrical
survey system to accomplish the goal of leak
detection and location on a small scale.
The planned system testing program uti-
lizing the full scale lined facility will
further define performance capabilities
and establish the confidence level of survey
results. These results will be evaluated
to form the basis for a decision to test
the system at two field sites.
ACKNOWLE DGEMENT S
The research which is reported in this
paper is being performed under Contract
No. 68-03-3033 "Investigation of Electrical
Technqiues for Leak Detection in Landfill
liners" with the Environmental Protection
Agency, Municipal Environmental Research
Laboratory, Cincinnati, Ohio. The support
and guidance of Mr. Carlton Wiles, Project
officer, is gratefully acknowledged.
210
-------
DEVELOPMENT OF LINER RETROFIT CONCEPTS FOR SURFACE IMPOUNDMENTS
John W. Cooper
David W. Shultz
Southwest Research Institute
San Antonio, Texas 78284
INTRODUCTION
Surface impoundment facilities such
as pits, ponds, and lagoons are used ex-
tensively throughout the United States to
store, treat and dispose of hazardous
wastes. These impoundments usually are
designed to contain fluids, utilizing
native materials or liners and in the
past, many were constructed with little
concern for the prevention of seepage from
the facility. In fact, many such faci-
lities were specifically designed to seep
water into the ground as a means of dis-
posal. Unfortunately, many of these faci-
lities are presently causing groundwater
contamination problems.
With the establishment of regu-
lations passed under authority of the
Resource Conservation and Recovery Act
(RCRA) legislation, retrofitting in-
service disposal and treatment impound-
ments may be required to prevent continued
seepage. Retrofitting with impermeable
liners may accomplish this objective.
Liners may also be required to protect
against wind and water erosion, another
objective of RCRA regulation require-
ments .
Retrofitting an in-service impound-
ment with a flexible membrane liner
generally implies the emplacement of
liner material on the bottom and sides of
a facility to mitigate leakage without
removing any in situ fluid (hazardous
material) in the impoundment. Such action
may be desirable as a temporary measure
while a new impoundment is being construc-
ted; or as a permanent step to upgrade an
untrustworthy facility. As RCRA regu-
lations are implemented and groundwater
studies confirm contamination from seeping
impoundments, information about retro-
fitting techniques will be needed to
assist owners and operators in upgrading
existing facilities.
The main purpose of a membrane liner
system in a fluid impoundment is to pre-
vent loss of liquid to groundwater re-
sources, but a liner can also prevent
bank erosion due to wave and wind action.
Flexible liner membranes are manufactured
using a variety of polymeric materials
such as polyethylene and polyvinyl
chloride. In the construction of new
impoundments, the sheets of liner mem-
brane materials are installed over com-
pacted subgrade and connected together
using a variety of seaming methods. The
most common methods are heat sealing and
solvent bonding. The sheet is attached
at the top of the impoundment perimeter
using an anchoring system, usually an
earthen anchor trench. The finished
liner serves as a relatively impermeable
continuous barrier covering the earthen
impoundment.
The implementation and enforcement
of RCRA regulations concerning hazardous
waste impoundment will create a need for
owners/operators to cease the contamina-
tion of groundwater resources due to
seepage. In many cases it may be desirable
or even necessary to make design modifi-
cations while the facility is in service,
while in others it may be necessary to
temporarily reduce seepage while a faci-
lity is closed or a new one is constructed.
Retrofitting of membrane liner systems
offers a potential way for owners of such
hazardous waste fluid impoundments to re-
duce or stop seepage.
211
-------
Unfortunately, retrofitting existing
impoundments has not been routinely per-
formed. Many questions need to be
answered before this approach to sealing
an existing impoundment can be imple-
mented. Information concerning the size
of facilities which can be retrofitted,
equipment needed, the most efficient
retrofitting system, labor requirements
and costs needs to be developed.
This paper reports on an ongoing
project that Southwest Research Institute
is conducting for the U.S. Environmental
Protection Agency for the development and
demonstration of systems to retrofit
existing liquid surface impoundment faci-
lities with synthetic membrane. The pro-
ject objectives will be stated, the work
accomplished described, and future work
outlined.
PROJECT OBJECTIVES
The primary objective of this project
is to determine the feasibility of retro-
fitting an existing surface impoundment
with an effective membrane liner system
without taking the facility out of ser-
vice. Anticipated project results are
the following:
* Identification of one or more tech-
nically feasible retrofit liner
installation techniques.
* Demonstration of the feasibility of
each system on a pilot scale.
* Preparation of a full scale demon-
stration plan if the systems prove
feasible at pilot scale.
The results of the project will deter-
mine the feasibility of retrofitting
existing surface impoundments with a mem-
brane liner system. Conceptual approaches
to retrofitting different types of im-
poundments will be developed and evaluated.
The most promising approaches will be
attempted at a pilot scale to verify or
deny their feasibility. From these
activities, the important procedural and
physical parameters to consider when re-
trofitting will be established. Limi-
tations of each approach will be defined,
e.g. which approach is best suited for a
particular impoundment geometry. A plan
will be developed for full scale im-
plementation of those retrofitting
systems shown to be feasible at pilot
scale. Potential sites will be iden-
tified where the retrofit approaches could
be implemented.
The most significant benefit which
can be expected as a result of successful
completion of this project will be the
development of a promising technique(s)
which could be used to halt the release of
hazardous fluids into the environment.
Preventing or stopping the contamination
of surface and/or groundwater may be
important to protect drinking water sup-
plies. If a surface impoundment con-
taining hazardous wastes is found to be
seeping contaminated fluids, retrofitting
the facility with an effective membrane
liner system should provide mitigation.
This project should provide information
needed about retrofitting in order to
ascertain the feasibility of full scale
utilization.
LOAD ANALYSES
A theoretical stress/strain analysis
has been performed for the various retro-
fit membrane materials. Mechanical pro-
perties and coefficient-of-friction and
failure data for different types of liner
material were obtained and used during
the analyses. A load analysis program has
been written to calculate the stresses on
liner materials resulting from installa-
tion using the pull-thru concept for retro-
fitting. The program is set up to run
on a HP9325A desktop computer. Required
input data is as follows:
* length of the first sheet to be pulled
into the pond;
* the footage of liner covering the end
slopes of the pond;
* the footage of liner covering the side
slopes of the pond;
* coefficients of friction values for
soil and water;
* the tensile strength of the liner
material at elongation;
* the thickness and density of the liner;
and
212
-------
* the length and width of the pond.
This program will calculate these
stresses for any impoundment geometry and
liner material. Table 1 presents cal-
culations for four generic liner materials
and three different sizes of impoundments.
The pond geometry varies as does the size
of the panels to be seamed together as the
material is pulled into the water.
These calculations provide a very
gross estimate of the forces to be expec-
ted. Critical factors yet to be answered
are, (1) can the maximum force required
to pull the liner be distributed evenly
along the edge of the liner; (2) what are
the coefficient of friction values for
each material and how can they be reduced;
and (3) what effect will weighting the
draw bar to keep the liner on the bottom
have on the stress distribution to the
liner. The values shown in Table 1 are
likely more accurate for the pull-over
retrofit concept than the pull-thru con-
cept. Note that the calculated tensile
values for all cases do not exceed the
tensile at elongation values of the
various materials.
CONSIDERATIONS IN RETROFITTING LINERS
A study has been made of considera-
tions in retrofitting liners. Two general
approaches to retrofitting ponds have
been analyzed. These are pulling the liner
along the contour of the pond and pulling
the liner on top of the fluid, with sub-
sequent sinking of the liner to conform
to the contour. Each approach has posi-
tive and negative features. The following
discussion presents methods for deployment,
advantages and problems and possible miti-
gating actions. Common features and prob-
lems are identified.
PULL-THRU CONCEPT
Method of Deployment
* Assembly - The liner will be assembled
by one of three methods:
(1) All of the membrane panels are laid
out, seamed and then fan folded.
(2) The panels are laid out, seamed
and then fan folded.
(3) The panels are laid out and seamed
one at a time.
* Pulling - The leading edge of the
assembled panel will be wrapped around
a heavy catenary cable or chain or clamped
between sections of heavy steel plates to
distribute the pulling load along the
edge of the panel. A battery of cable
pullers or winches would be mounted at
the opposite end of the impoundment to
pull the panel into and across the basin.
Bottom Contour and Side Slope Following
While new impoundments with membrane
liners generally have smooth flat bottoms
to facilitate proper placement of the
liner, existing unlined impoundments will
most likely not have smooth flat bottoms
or straight side slopes since that would
not have been a design requirement at the
time they were built. Thus, even with
the use of rollers at the bottom of the
side slopes and in the corners, it will be
difficult to prevent bridging by the liner
as it is pulled across depression in the
bottom or irregularities along the sides.
Large pockets of liquid would then be
trapped beneath the liners. The impli-
cations of this entrapped liquid are dis-
cussed in a later section. However,
drag caused by the digging in of the
leading edge of the liner, or by the build
up of mud or sludge balls along this edge,
can greatly increase the required pulling
force or cause the cable or load distri-
buting panel to fail.
Mitigating actions to reduce bridging
and entrapment of liquids will be by the
frequent use of rollers to keep the liner
on the bottom of depressions or irregu-
larities and by the careful design of
the leading edge of the load distributing
panel to prevent digging in. A slight
rise in the weighted leading edge will
cause the liner to dig through the
sludge but not into the harder underlying
soil. If the impoundment is filled with
filamentous sludges (as might be the case
for some sewage wastes) it may not be
possible to place the liner by dragging
because of balling of the sludge.
Tearing
Existing impoundments are likely to
have protrusions that will cause ex-
cessive stress concentrations in the
213
-------
liner and faillure by tearing. Irregu-
larities in the bottom, rocks, pipes
and other debris, known or more likely un-
known and unseen, can cause ripping. Be-
cause the ripping is likely to occur
beneath the surface of what is perhaps a
murkey liquid the tear may not be noticed
or detectable.
Mitigating actions may be to search
the impoundment with a drag wire survey
to locate and/or remove the larger pro-
trusions, by reducing friction forces so
that the localized tension forces at the
protrusion are less, and thus the near
bouyant liner might ride over it without
tearing, or by using a tougher fiber rein-
forced liner that is less likely to tear
on small protrusions.
FLOAT AND SINK CONCEPT (PUMP-OVER)
Method of Deployment
* Assembly - The liner is assembled using
one of the methods described for dragging,
however, the end that is pulled across the
impoundment is doubled back over a stiff
hose, or sections of rods, for a suffi-
cient distance so that when the liner is
deployed, the end will fall several feet
short of the "shoreline" when fully de-
ployed. Clamps around the liner and hose
or rods are then installed to permit pul-
ling the liner without excessive strain.
* Placement - The liner will be pulled
across the impoundment in a fashion similar
to the pull thru concept, however, in this
method of placement, the loading edge of
the liner and the pulling lines will be
kept as light as possible so that the
liner would remain on or near the surface
during the deployment period.
* Sinking of Liner - The deployed liner
will be lightly held in position during
the sinking phase. If the liner has a
density greater than the liquid it will
slowly sink. As it sinks, the liquid
underneath the liner would escape and
flow over the top at the open water margin
between the folded end of the liner and
the far shore line. If the liner is made
of bouyant material, e.g., HOPE, ballast
such a coarse sand would be used to sink
the liner. The ballast would be placed
on the liner starting at the initial de-
ployment end. Only a small amount of
ballast should be required. Only 2%
pounds per 100 square feet should sink
80 mil HOPE. Similar to the dense liner,
the displaced water will move towards the
gap at the far end as the liner is slowly
covered with ballast. When the liner has
completely sunk, the doubled back flap is
opened up and pulled towards the far end
to close the gap. One end of the gap will
be left down until the very end to let any
remaining liquid escape. For the pump-
over concept, fluid would be pumped onto
the liner rather than allowing it to flow
over the leading edge.
Advantages
Placement of the panel will cause
minimal stress to the liner since only the
dry land friction should be of any mag-
nitude and, as discussed earlier, this can
be mitigated. There should be negligible
tearing since the liner is not dragged
across potential tearing protrusions.
Holing of the liner should be limited to
localized punctures by very sharp pro-
trusion. Since the liner sinks under
minimum tension, it should settle into
most bottom depressions and side irregu-
larities.
SPECIFIC PROBLEMS AND MITIGATING ACTIONS
Sinking and Liquid Recirculation
Even with the provision of a folded
back flap at one end to permit the liquid
beneath the liner to recirculate to the
top, proper sinking of the liner may be
difficult to achieve. The liner will
probably not sink uniformly by itself but
will have to be constrained around the
perimeter as it sinks. Otherwise, the
liquid underneath may attempt to re-
circulate to the surface at locations
other than planned with the liner settling
to the bottom possibly out of position.
This could result in excessive folds in
the liner or lack of sufficient coverage
around the side slopes. Also since the
tension in the linear during placement
will be less than for dragging, extra
care will have to be taken to reduce
damage or misplacement due to wind or
wave action. Pumping of fluid would allow
placement of the water at the end of the
facility opposite from the leading edge.
This may facilitate controlled sinking
of the liner.
214
-------
COMMON PROBLEMS AND MITIGATING ACTIONS
Entrapped Liquids and Sludges
Whether by pulling or sinking, some
liquid or sludge will be trapped between
the liner and the bottom of the impound-
ment. The specifics of the type of waste
and the underlying soil conditions will
determine what actions should be taken
to accommodate these entrapments.
Non-Gas Generating Entrapments
If the entrapped liquid or the under-
lying soil does not generate gases, it may
be acceptable to leave the liquid and not
attempt to remove it. If the underlying
soil is permeable enough, and the water
table low enough, the entrapped liquid
will percolate away. If the underlying
soil or entrapped sludge is not permeable,
the bubble of liquid will remain. There
is then the danger of ripping or perfor-
ation of the liner due to the repeated
working of the fabric over the bubble
under wave induced or other hydraulic
motions.
Methods to remove entrapped bubbles
of liquid may include the following:
* One-way valves could be widely distri-
buted over the liner. In the placement
mode they will assist in the rapid and
uniform sinking of the liner and would
permit the entrapped liquids to escape.
Detrimental aspects of their use will
include clogging by the sludge, complexity
of installation and susceptibility to
damage, and the likelihood of leakage,
albeit a small amount due to grit lodging
on the valve faces.
* The impoundment could be lined with an
array of drain pipes with the inlets
covered with porous sand, gravel and/or
a geotechnical filter membrane. Entrapped
liquids would then be pumped out. The
placement of the array in an existing im-
poundment may be difficult and complex.
There will also be susceptibility to drain
blockage by sludge.
* By the use of either ballast or
weighted rollers, entrapped liquids could
be rolled out. Starting in the middle
or deepest point of impoundment, ballast
is placed on the liner to establish an
outward moving radial front of the dis-
placed entrapped liquid. The entrapped
liquid is then allowed to escape to the
surface around the perimeter. Rollers
operating in a sequence of outward moving
radial rolls can similarly work the en-
trapped liquids towards the edges and
then the surface.
Gas Generating Entrapments
Many waste liquids, sludges and
underlying soils may generate gases.
Bubbles of gas under the liner can raise
the liner to the surface or at least
separate it from the bottom and make it
susceptible to wear and failure from
hydraulic motions or induced stresses.
In some cases, if the waste material and
underlying soil are permeable enough,
the entrapped gas will escape by itself.
However, if the gas cannot escape, then
mitigating actions will have to be
taken using the methods for removing
entrapped liquids or other techniques:
* If clogging of the valves can be pre-
vented and if leakage is acceptable, one
way valves can be effective in the removal
of entrapped gas. The escaping gas may
also unclog the valve.
* Similarly, if clogging can be pre-
vented, drain fields can be used to remove
gases.
* Rolling out or ballasting will not be
an acceptable method for gas removal un-
less only a very small quantity of gas
is generated for a short period after the
retrofit of the liner.
* Non-clogging geotechnical filter mem-
branes that can block sludge or sediment
while passing liquids and gases can be
installed under the retrofitted liner.
They will then conduct the entrapped
liquids and gases to the surface or to
either one way valves or a drainage field
where the gas can escape. Uncertainties
with using permeable membranes include
the long term blocking of the pores of the
material and compression of the pores due
to the overlying load of the waste
material.
Sludge and Contaminated Soil Removal
The float and sink method will trap
all, and the dragging method much, of the
solids in the sludge under the retrofitted
215
-------
liner. This sludge and any contaminated
soil under the impoundment can leak un-
desirable trapped liquids or generate gas
for many years. In some cases, the leach-
ate may be able to be pumped out from peri-
pheral wells. However, if this practice
is determined to be unacceptable it may
be necessary to completely remove the
sludge and contaminated soil. The size
of many impoundments and other constraints
will preclude draining the impoundment and
temporarily storing the waste material in
another impoundment. Thus, the solid
material will have to be removed in situ
and prevented from settling before the
retrofitted liner can be placed. Two
methods of dredging and liner installation
might be used to achieve this objective.
Both methods require that the waste
material in the impoundment has the pro-
perty that suspended solids will settle
out under gravitational forces so that
there is a clear liquid overlying a sludge
blanket. Except for some colloidal sus-
pensions, most aqueous wastes have these
properties. Oily wastes may have oil or
foam on the surface but there is usually
a clear liquid underneath.
* Dragging - As the liner is pulled along
the bottom and into position, a dredge
will remove the sludge and contaminated
soil immediately ahead of the liner. The
dredged spoil is initially pumped and dis-
charged at a far corner of the impound-
ment. There the suspended sediment settles
to the bottom and the clear liquid re-
circulates back into the impoundment. It
may be necessary to install a floating
silt curtain around the discharge to con-
fine the dredged material until it has
settled out. When approximately half the
liner is dragged into position, the dis-
charge will be moved so that the spoil
is deposited at a far corner of the lined
part of the impoundment. Care will have
to be taken to widely spread the spoil
over the liner since it will still be
moving into position and thus, it will be
desirable not to create abnormal stresses
in the liner by weighting it down with
sediment. If there is a very large mass
of heavy settled sludge, this method of
installation may not be feasible.
* Placement - In this method, part of the
impoundment is sectioned off by a floating
curtain similar to those used in con-
trolling oil spills. The bottom edge
of the curtain is weighted and lies on
the bottom of the impoundment as a seal.
The sediment from the sections is dredged
and deposited far behind the curtain.
The clear supernatent is permitted to
recirculate over the top of the curtain.
After dredging, the liner is placed
using the methods discussed earlier. The
curtain is then moved to enclose the next
adjacent section. After completion of
dredging of this section, the liner is
placed with an overlap over the previously
placed liner. With sufficient overlap,
acceptable small leakage rates can be
achieved. Similarly to the procedure
for the pull-thru method of installation,
when the impoundment is half lined the
spoil discharge point is moved to a remote
position over the retrofitted liner. Since
in the placement method of installation
the liner is not moved over the bottom,
the mass of the settled sediment at a
particular place is not as critical using
this method. While small road trans-
portable cutter head or dust pan dredges
can be used to move the sludge, con-
sideration should be given to pneumatic
dredges that have been developed for
dredging contaminated sediments. This
type of dredge entrains considrably less
liquid and the discharged spoil tends to
settle out with a considerably smaller
surface plume.
* Folds - As was discussed earlier, the
topography of impoundments in which retro-
fitted liners will be placed will most
likely be quite irregular. In installing
the liner, it will be very difficult, if
not impossible, to prevent folding of the
liner. Where the liner is above the liquid,
as around the sides of the impoundments,
folds can be cut out and the liner patched.
However, when folds occur in the liner
beneath the surface of the liquid waste
it may not be possible to remove the
folds using these methods. One, the fold
may not be detectable if covered by sludge
or murky waste. Two, it may not be
possible to patch the liner underwater.
If the fold is left unattended, the liquid
confined in the fold may drain out by
itself or may be withdrawn using the
methods for removing liquids and gases
discussed earlier. However, the base
of the fold may seal the fold thus trapping
the liquid in the fold. Similar to the
entrapped bubble, the material in the
fold will work under various hydraulic
actions which may ultimately cause the
liner to wear or rip. If the fold
216
-------
matrial just wears enough so the entrapped
liquid can escape, the fold may collapse
and the two overlapping halves may form
a seal. If the fold material tears, the
two halves of the fold may move apart
and an appreciable leak path may be
created. Thus, a liner material that
wears and leaks may be preferable to one
that doesn't wear but tears. Folds in
the subsurface part of the retrofitted
liner may have to be accepted. Perhaps
in areas where folds are likely to be
encountered, a second layer will be in-
stalled to reduce leakage through failed
folds.
DOUBLE PERFORATED LINER WITH PERMEABLE
GEOTECHNICAL FILTER MEMBRANE
For the many reasons discussed in
this presentation, the probability of
having leakage after the installation of
a retrofitted liner is quite high. Proper
design, selection of materials and place-
ment can result in an installation with
greatly reduced leakage but there is still
likely to be some. If this less than per-
fect performance is acceptable then the
utilization of two layers of liners that
are intentionally perforated may lead to
a superior installation system. Aspects
of a two liner system are:
* Installation - If the perforated
liner is installed using the placement
method it will sink more rapidly than a
non-perforated liner and will require
less control around the perimeter. Any
residual trapped liquid will be displaced
as the liner sinks to the bottom. Liquid
trapped in folds will also readily leak
out, permitting the fold to collapse.
in the installation of buoyant liners,
porous ballast such as sand or gravel
would not block the perforations. If the
second layer of the perforated liner is
laid at right angles to the first,
likelihood of folds lying on top of each
other and of being parallel is reduced.
* Perforations - Perforations in the
liner would not be uniformally placed
but would be located in a pseudo-random
pattern. That is, there might be one
hole randomly placed in each square foot.
With this random placement there would
not be a row of holes in one liner
aligning over a row of holes in the other
liner.
* Leakage - If the lower layer is lying
firmly on the bottom and the facing sur-
faces of the two liners are clean than
the double perforations would make an
effective one-way valve or hydraulic
diode. Either due to negative buoyance
or added ballast the two layers will be
pressed together blocking any leakage.
If there is any trapped liquid or gas at
a higher pressure than the overlying
liquid, the upper layer will lift, per-
mitting the gas or liquid to escape
through the perforations in the upper
layer. Once the fluid has escaped, the
pressure is reduced and the upper layer
will collapse resealing the bottom layer.
Since, with many randomly placed holes,
some holes in one layer will align up with
holes in the other layer there will be
some leakage. However, the area of the
perforations will be a very small part of
the total area so leakage would be very
small for a double layer system. A 3/8-
inch hole in every square foot of each
liner would permit the upward flow of all
trapped liquids, yet would have a porosity
of less than 10~3. A double layer would
then have a porosity of less than 10~6.
When combined with existing porosity of
the impoundment bottom there should be
very little leakage.
* Clogging - the efficiency of a per-
forated double layer retrofitter liner
system is dependent upon keeping the per-
forations open and the facing surfaces of
the liner clean. Blockage of the per-
forations will not permit the trapped
fluids to escape and grit caught between
the two layers might create a leakage
path. If a liquid and gas permeable
geotechnical filter membrane is installed
under the perforated liners, sludge and
grit underneath would not move up and
block the perforations or contaminate the
facing surfaces of the liners. A very
effective retrofitted liner system would
then be achieved.
* Gas - Gas trapped beneath the liner
system or subsequently generated in the
underlying sludge or soil will be able
to escape through the filter membrane to
a perforation and then to the surface.
If there is a heavy or impermeable waste
fill in the impoundment, the gas will
move laterally until it can escape through
an unblocked perforation or until it
reaches the surface around the perimeter
of the impoundment.
217
-------
LABORATORY SCALE STUDIES
A 10 foot by 20 foot scale model
impoundment has been constructed and is
being used to make qualitative studies or
retrofitting liners. The model was con-
structed outside on the ground, lined with
10 mil black polyethylene sheeting for
water retention and then filled with an
inch or so of sand to simulate an un-
lined impoundment. The sand was also
banked up the sides and stabilized with
cement.
At the time of writing only one
series of experiments has been undertaken:
the floating and sinking by pump-over of
a buoyant membrane liner. A 4 mil poly-
ethylene liner was pulled over and floated
on the impoundment, overlapping both the
sides and ends. Water was pumped from one
end and released at the opposite end at
the same time as sand ballast was being
put on the liner to sink it. The
following observations were then made:
* The liner would not sink to the bottom
even with the ballast and water added.
Archimedean buoyance forces and tension
in the liner from friction forces around
the perimeter kept the liner afloat.
* The liner was moved back several inches
from the pumping end of the impoundment
and a fold worked back towards the
ballasted end. When the fold was adjacent
to the ballast the liner would then sink
to the bottom.
* With no longitudinal tension the
ballasted section of the liner will fold
under the remaining buoyant liner. Light
tension will keep the size of this fold
manageable.
* An advantage of floating and sinking
is that the liner is not dragged over the
bottom. However, the liner will move a
little along the sides of the impoundment
as it is sunk. Provision will have to be
made to accommodate this movement with-
out overstressing the liner.
ground of Southwest Research Institute.
It allows field testing of retrofit con-
cepts under controlled conditions and in
the absence of hazardous fluids. This
is necessary in the early stages of
system development to avoid risk of in-
jury to field crews.
FUTURE WORK
Following further laboratory scale
studies retrofitting concepts will be
selected for demonstration in the pilot
scale facility. The pilot scale demon-
stration will be conducted and then eval-
uated. A retrofit system field test plan
for an actual hazardous waste site will
then be developed.
ACKNOWLEDGEMENT
The work reported on in this presen-
tation has been accomplished by
Southwest Research Institute under Grant
No. CR 809719 from the Municipal Environ-
mental Research Laboratory of the U.S.
Environmental Protection Agency.
PILOT SCALE DEMONSTRATION
A pilot scale demonstration facility
has been designed and built to test the
retrofit system on a larger scale. The
facility has been constructed on the
218
-------
LONG TERM IMPACTS OF HAZARDOUS WASTE CODISPOSAL
IN LANDFILL SIMULATORS
James J. Walsh, P.E.
W. Gregory Vogt
SCS Engineers
Covington, Kentucky 41017
Riley N. Kinman, Ph.D., P.E.
Janet I. Rickabaugh
University of Cincinnati
Cincinnati, Ohio 45221
ABSTRACT
This project was initiated to simulate landfill decomposition and contaminant re-
lease under controlled conditions. The research objectives were to determine effects on
solid waste decomposition by the addition of different water infiltration rates, by pre-
wetting the refuse mass, and by codisposing municipal refuse with sewage sludge and
various hazardous wastes. To this end, 19 landfill simulators were constructed in late
1974 and early 1975. Each test cell was loaded with approximately 3.0 metric tons of
municipal refuse. Selected cells received quantities of hazardous waste additives. Main-
tenance and operation of these landfill simulators has continued to this date and cells
are monitored for temperature, gas quantity and quality, and leachate volume and compo-
sition. This paper presents seven years of results for the hazardous waste codisposal
cells and describes leachate concentration histories using a semi-empirical leaching
model.
INTRODUCTION
As described by the Resource Conser-
vation and Recovery Act (RCRA) regulations
promulgated by the U.S. Environmental
Protection Agency (EPA) in May 1980, gen-
erators of less than 1,000 kg of hazardous
waste per month may treat or dispose of
their wastes without complying with the
full RCRA program. That is, if small
quantity generators comply with the regu-
lations for acutely hazardous waste and
on-site hazardous waste accumulation,
they are then exempt from RCRA require-
ments. The hazardous waste of these gen-
erators is typically sent to municipal
landfills which may not be designed for
disposal of such wastes. As a result,
the possibilities of chronic environmental
problems have prompted public scrutiny
of hazardous waste codisposal in municipal
landfills.
Leachate generation from municipal
landfills is the result of moisture from
percolating rainfall and surface water,
from infiltrating ground water, or from
the refuse itself as delivered. Since
the chemical composition of the leachate
varies depending on the original waste
mass, the degree of compaction, the pene-
tration of air and water, and the age
of the landfill, research designed to
test these variables will yield better
methods of prediction for leachate gene-
ration and contaminant release.
This research project, motivated
by the need to describe landfill decom-
position and contaminant release under
controlled conditions, addresses the im-
pacts of hazardous waste codisposal
in landfill simulators. The overall
objectives were to determine the effects
on solid waste decomposition by the ad-
dition of different water infiltration
rates, high initial water content,
sewage sludge, calcium carbonate, and
six industrial hazardous wastes. The
purpose of this paper is to present the
219
-------
data acquired after seven years of project
operation and to evaluate long term impacts
associated with the disposal of hazardous
wastes into municipal landfills.
BACKGROUND
The project was initiated in 1974
to simulate the decomposition of municipal
solid waste and to demonstrate effects
on that decomposition process by the addi-
tion of extra precipitation, pH buffering
compounds, sewage sludges, and industrial
wastes. A total of 19 landfill simulators
were constructed according to the program
design in Table 1. Four test cells, con-
taining only municipal refuse to serve
as controls, received different water
infiltration rates. Three other test
cells also served as controls but are
located in the laboratory and operated
at room temperatures. Low, medium, and
high quantities of sewage sludge were
added to Test Cells 5, 6, and 7, respec-
tively. Test Cell 8 received calcium
carbonate as a pH buffer additive. As
shown in Table 1, selected industrial
wastes and sludges were added to six test
cells and an initial water addition was
made to the municipal refuse of Test Cell
11. The operation of Test Cell 15 (polio
virus addition) was discontinued in the
early stages of the project. The 19 test
cells for this program were constructed,
loaded, and completely sealed in November
1974 and April 1975.
The project site for this study is
the EPA Center Hill Laboratory in Cincin-
nati, Ohio. .Fifteen of the landfill simu-
lators are located outdoors and below
the ground surface. As indicated in
Figure 1, these fifteen are arranged in
a horseshoe alignment with a central leach-
ate collection well. The remaining four
test cells are located above ground and
inside the high-bay area of the Center
Hill Laboratory facility.
METHODS
Initial activities included test cell
construction, loading, and( waste charac-
terization. On-going operations include
water infiltration, leachate drainage,
and monitoring of cell gases, leachate,
and temperatures. Individual test cells
consist of steel tubes 1.83 m in diameter,
3.6 m high, and 4.76 mm thick. Steel
sidewalls were coated and sealed with
coal-tar epoxy prior to placement. The
outside test cells were placed on concrete
slabs with layers of fiberglass cloth
lining the bottom sidewall (0.3 m) to
ensure a water and gas seal. The interior
test cells had welded steel bases.
Provisions for leachate drainage were
installed in all cells. Leachate drainage
for the exterior cells is through connec-
tive piping to a central leachate collec-
tion well. Interior cells are placed
TABLE 1. PROGRAM DESIGN
Test
Cell
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Initial Loading Additive
Control
Control
Control
Control
Sewage Sludge
Sewage Sludge
Sewage Sludge
Calcium Carbonate
Petroleum Sludge
Battery Waste
Water
Electroplating Waste
Inorganic Pigment Waste
Chlorine Production Brine Sludge
Polio Virus
Control*
Solvent-Based Paint Sludge*
Control*
Control *
Annual
Infiltration
203 nrn
406 mm
610 mm
813 mm
406 mm
406 mm
406 mm
406 mm
406 mm
406 mm
406 mm
406 mm
406 mm
406 nm
Discontinued
406 mm
406 mm
406 mm
406 mm
Construction/
Loading Date
Nov. April
1974 1975
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
These test cells located inside the Center Hill Laboratory high bay area.
All others located outside and below ground surface.
220
-------
EXTERIOR TEST CELLS
INTERIOR TEST CELLS
INSTRUMENTATION
BUILDING
©|® © ©
©
©
19 18
L
LEACHATE
COLLECTION
WELL
BUILDING WALL
Figure 1. Test Cell Location Plan.
on concrete blocks with leachate drains
mounted into the cell bases.
The test cell loading operation began
with a 15.2 cm thick layer of silica gravel
spread on the bottom of each cell. Refuse
was delivered to the site and added to
each test cell in eight 0.3 m thick lifts.
Each lift was compacted with a wrecking
ball. Sludges and other additives were
applied at the top of all but the first
lift. Temperature probes were installed
atop the second, fourth, and sixth lifts
in each cell. Gas probes were installed
atop the second, sixth, and cover lifts.
After loading the waste mass, a layer
of silty -clay cover soil followed by
a layer of pea gravel were added to the
test cells. A water distribution line
was installed and steel lids were then
welded into place. A cross section of
a typical test cell is shown in Figure 2.
Test cell loading features have been
summarized in Table 2. As indicated,
all cells received approximately 3,000
kg of municipal refuse (wet weight). Se-
lected cells received from 68 kg to 2,039
kg of waste additives mixed in proportionate
amounts for each lift except the bottom
lift. For example,. Test Cell 12,
loaded with 11.5 percent (dry weight ba-
sis) of additive, received 1,190 kg
of electroplating waste. Thus, total
•CTTLCHIMT INDICATOR
tWHT TUIE
2.4 M yUMICIPAL HIP US!
TIMPtftATUftt PHOBEft —
TIMPIIUTUIME PftOU
Figure 2. Test Cell Cross Section.
221
-------
TABLE 2. TEST CELL LOADING FEATURES
Refuse
Solids Content (%
of wet weight)
Wet Weight (kg)
Dry Weight (Kg)
Mix Ratio (additive
as % of total
addi ti ve/ref use
mass - dry weight
10 b?sis)
ts> Additive - Type
Solids Content (%
of wet weight)
Uet Weight (kg)
Dry Weight (kg)
Total
Sol ids Content (%
of wet weight)
Wet Weight (kg)
Dry Weight (kg)
Volume (m3)
Density
Wet weight (kg/m3)
Dry Weight (kg/m3)
1
61 8
3025.0
1869.5
--
--
--
--
61 .8
3025 0
1869.5
6.4
472 7
292.1
2
61.8
2989 0
1847 2
--
--
—
--
--
61 8
2989.0
1847 2
6 4
467.0
288.6
3 4
61.8 61.8
3007 0 3002.0
1858.3 1855 2
-_
-_
_-
__
__
61.8 61.8
3007.0 3002.0
1858.3 1855.2
6 4 6.4
469.8 469 1
290.4 289.9
5
61 8
3001.0
1854.6
0.4
Sewage
SI udge
12 0
68 0
8.2
60.7
3069.0
1862.8
6.4
479.5
291.1
61.8
2919.0
1803.9
1.3
Sewage
SI udge
12 0
204.0
24.5
58.5
3123.0
1828.4
6.4
488.0
285 7
61.8
2964.0
1831.8
4.3
Sewage
SI udge
12 0
680.0
81.6
52.5
3644.0
1913.4
6.4
569.4
299.0
61 .8
2994 0
1850 3
4.2
Cal c lum
Carbonate
90.0
90.7
81.6
62 6
3084.7
1931.9
6.4
482.0
301 9
Test C
69 6
3001 .0
2088.7
13.2
Petroleum
Sludge
21.0
1518.0
318 8
53.3
4519.0
2407.5
6 4
706.1
376 2
ells
69.6
2998 0
2086.6
6.2
Battery
Waste
10 7
1291.0
138.1
51.9
4289.0
2224.7
6.4
670.2
347.6
61.8
2924.0
1807.0
Water
0 0
1293 2
0.0
42 8
4217 2
1807.0
6.4
658 9
282.3
61.8
3048.0
1883.7
11 5
Electro-
plating
Waste
20.5
1190.4
244.0
50.2
4238.4
2127 7
6.4
662.3
332.5
61 .8
3006 0
1857.7
27.0
Inorganic
Pi gment
Waste
48 3
1420.6
686.1
57 5
4426 6
2543.8
6.4
691.7
397.5
61.8
3015.0
1863.3
45.4
Chi on ne
Prod. Brine
SI udge
75 9
2038.7
1547 4
67.5
5053 7
3410.7
6.4
789.6
532.9
61 8
2996.0
1851.5
_.
_-
61 8
2996.0
1851 5
6.4
468.1
289.3
61.8
2998.0
1852 8
39 5
Solvent
Based Paint
Sludge
75 3
1604 0
1207.8
66 5
46U2.0
3060.6
6 4
719.1
478.2
69.6 69 f>
3000.0 3012 0
2088 0 2096 4
__
69 5 69 b
3000 0 3012.0
2088 J 2096 4
64 6.1
468 8 470. b
326 3 1
-------
wet weight for the test cells ranged be-
tween 2,989 kg and 5,054 kg. With waste
compartment volumes of 6.4 m3, the final
densities achieved for all cells ranged
from 467 kg/irr and 790 kg/m^.
To closely simulate in-field land-
fill conditions, all test cells are oper-
ated on a strict monthly schedule of mois-
ture application. This infiltration sche-
dule is presented in Table 3. As shown,
Test Cells 1, 3, and 4 receive approxi-
mately 200, 600, and 800 mm of annual
infiltration, respectively. All remaining
test cells receive the 406 mm annual in-
filtration rate and simulate average pre-
cipitation/percolation conditions in the
Midwest U.S. The annual rates are ad-
ministered on a monthly basis, but no
infiltration is applied in the usually
dry months of January, July, August, and
September. High amounts of infiltration
are applied in the normally wet months
of March, April, and May.
Physical data recording, sampling,
and analyses are performed on 18 test
cells due to the discontinuance of Test
Cell 15. The monitoring program is pre-
sented in Table 4 and indicates the ex-
tensive analytical efforts conducted on
leachate samples. Leachate samples are
collected and analyzed for 25 parameters
each month; with an additional 8 para-
meters each quarter, and one parameter
every six months. As shown in Table 4,
additional parameters were monitored
during the course of the project. However,
these parameters were not detected at
some point since 1974 or were not deemed
useful for describing landfill behavior
and thus monitoring was discontinued.
RESULTS
Due to the large analytical program
for the project, the data cannot be fully
evaluated in this presentation. The
authors present the following results
obtained from two municipal refuse control
cells (Nos. 2 and 4) and the six hazardous
waste codisposal cells. These six cells
contain the following industrial additives:
Test Cell 9, petroleum sludge; Test Cell
10, battery waste; Test Cell 12, electro-
plating waste; Test Cell 13, inorganic
pigment sludge; Test Cell 14, chlorine
production brine sludge; and Test Cell
17, solvent-based paint sludge.
Refuse and Waste Composition
The refuse was characterized by phy-
sical and chemical composition prior to
test cell loading. The refuse physical
composition was determined by hand surveys
for both loading operations (November
1974 and April 1975). Table 5 indicates
the results from the sorting procedures
based on the 11 refuse categories. Results
from the two surveys did not statistically
differ (P<0.05) with the exception of
the garden waste category. Paper was
the major component (41.8 percent by
weight) of the municipal refuse sample.
Total
TABLE 3. INFILTRATION SCHEDULE
Quantity Per Month (Liters)
Month
January
Feb rua ry
March
April
May
June
July
August
September
October
November
December
203 mm
Annual Rate*
0
48.45
97.09
97.09
97.09
48.45
0
0
0
48.45
48.45
48.45
406 mm
Annual Rate*
0
96.90
194.17
1 94 . 1 7
194.17
96.90
0
0
0
96.90
96.90
96.90
610 mm
Annual Rate#
0
145.72
291.45
291.45
391.45
145.72
0
0
0
145.72
145.72
145.72
813 mm
Annual Rate**
o
194 17
388 34
388.34
388 34
194 17
o
0
0
194 17
194.17
1 94 . 1 7
533.52
1,067.01
1,702.95
2,135.87
Test Cell 1.
Test Cells 2, and 5-19.
Test Cell 3.
Test Cell 4.
223
-------
TABLE 4. MONITORING PROGRAM
Med 1 urn
Refuse
Industrial Wastes
Gas
Test Cell/Air
Leachate
Pa ramet er
Composition
Percent Moisture
Chemical Oxygen Demand
Total Kjeldahl Nitrogen
Total Phosphate
Inorganic Carbon
Orga ni c Ca rbon
Manganese
Potassium
Magnes i urn
Calci urn
Sodi urn
Copper
Nickel
Arsenic
Sel en l urn
Mercury
Lead
Beryl 1 i urn
Cadml urn
1 ron
Zinc
Chromi um
Lipids
Ash
Crude Fiber
Sugar
Starch
Asbestos
Percent Moisture
Total Solids
Total Volatile Solids
Copper
Nickel
Arseni c
Selenium
Mercury
Lead
Beryl 1 i um
Cadmi um
Iron
Chromi um
Cyanide
Tin
An t i mony
Vanadium
Titanium
Boron
Chi oride
Asbestos
Methane (CH^)
Carbon Dioxide (C02)
Ni t rogen ( N2 )
Oxygen (0?)
Hydrogen (H^)
Vol ume
Temperature
Air Pressure
Vol ume
Tempera ture
PH
Special Conductivity
Oxidation-Reduction Potential (ORP)
Chemical Oxygen Demand (COD)
Total Organic Carbon (TOC)
Alkalinity
Maxl mum
Frequency
Once, Initially
Once, Initially
Once. Initially
Once, Initially
Once, Initially
Once, Initially
Once, Initially
Once, Initially
Once, Initially
Once, Initially
Once, Initially
Once, Initially
Once, Initially
Once, Initially
Once, Initially
Once, Initially
Once, Initially
Once, Initially
Once, Initially
Once, Initially
Once, Initially
Once, Initially
Once, Initially
Once, Initially
Once, Initially
Once, Initially
Once, Initially
Once, Ini tial ly
Once, Initially
Once, Initially
Once, Initially
Once, Initially
Once, Initially
Once, Ini tial ly
Once, Initially
Once, Initially
Once, Initially
Once, Initial ly
Once, Initially
Once, Initially
Once, Initially
Once, Initially
Once, Initially
Once, Initially
Once, Initially
Once, Initially
Once, Initially
Once, Initially
Once, Initially
Once , Initially
Monthly
Monthly
Monthly
Monthly
Monthly*
Daily
Daily
Daily
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
224
-------
TABLE 4. MONITORING PROGRAM (Continued)
Med1 urn Parameter
Leachate Hardness
Total Kjeldahl Nitrogen (TKN)
Total Solids (TS)
Dissolved Solids
Total Volatile Solids (TVS)
Sulfur (S)
Chloride (Cl )
Boron (B)
Cyanide (CM)
Asbestos
Phenol
Iron (re)
Copper (Cu)
Calcium ( Ca )
Magnesium (Mg)
Sodi urn ( Na )
Potassium (K)
Zinc (Zn)
Cadmium (Cd)
Manganese (Mn)
Mercury (Hg)
Scandi urn (Se)
Beryllium (Be)
Arsenic (As)
Vanadium (V)
Nickel (Ni)
Ortho-Phosphate (0-PO.)
Sulfide (S=) *
Selenium (Se)
Total Phosphorus ( POj)
Volatile Acids
Total Chromium (Cr)
Lead (Pb)
Tin (Sn)
Total Col i form
Fecal Coliform
Fecal Streptococci
Ti tanium (Ti )
Antimony (Sb)
Oil and Grease
Aluminum ( Al )
Si 1 ver (Ag)
Ma ximum
Frequency
Monthly*
Monthly
Monthly
Monthly*
Monthly
Monthly*
Monthly
Monthly*
Monthly*
Monthly*
Monthly
Monthly
Monthly
Monthly*
Monthly*
Monthly*
Monthly*
Monthly
Monthly
Monthly*
Monthly
Monthly*
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Quarterly
Quarterly
Oua rterly
Quarterl y
Quarterly
Quarterly
Quarterly
Quarterly
Semi - Annual ly*
Semi - Annual 1 y*
Semi - Annua 1 ly*
Semi -Annual 1 y*
Semi -Annual 1 y
*Analysis discontinued during monitoring period.
TABLE 5. REFUSE PHYSICAL COMPOSITION
Component
Paper
Garden Waste
Metal
Food Waste
Glass
Plastic, Rubber, Leather
Textiles
Fines*
Ash, Rock, Dirt
Diapers
Wood
November
1974
41.0
16.0
8.3
7.7
8.0
6.5
3.9
3.3
2.4
1.9
1.0
Wet
April
1975
43.8
10.9
7.9
9.1
7.8
8.0
3.1
2.7
2.8
1.6
2.0
Weight [%J
Compos i te
Mean+
41.8
16.6
8.4
7.9
7.8
6.9
4.1
3.1
2.9
2.4
1.7
Standard
Deviation
4.6
6.0
1.4
1.9
1.5
1.8
1.6
0.7
1.3
1.0
1.0
* Material passing through a 25-mm (1.0 in.) sieve.
+ Values are composites of eight lifts from each of 19 test cells.
225
-------
The industrial hazardous wastes were
characterized for chemical composition
to determine the loading inputs of various
parameters. Results for these parameters
are presented in Table 6 and significantly
high concentrations are noted in selected
cells for copper, arsenic, selenium, mer-
cury, lead, iron, chromium, cyanide, tin,
and asbestos. Sub-samples of the indus-
trial wastes were also obtained for mois-
ture content determinations. As shown in
Table 5 and derived for Table 2, the in-
dustrial wastes ranged from 24 to 89 per-
cent moisture. In particular, the chlo-
rine production brine sludge and the sol-
vent-based paint sludge were very thick
additives with total solids contents of
approximately 75 percent at the time of
loading.
Leachate Volume
Leachate samples are withdrawn once
a month prior to water addition. The
volume of leachate is used in conjunction
with the water added volume to determine
the moisture balance and field capacity
of each cell.
Municipal refuse is able to retain
a certain quantity of moisture in addi-
tion to the initial moisture content.
This retention quantity can be determined
when the initial refuse moisture plus
the moisture added causes a flow of leach-
ate. Figure 3, which illustrates the
moisture balance for Test Cell 2, deter-
mines the moisture retained by the dif-
ference of the two plots. The moisture
retained, also termed the field capacity,
is the maximum moisture content which
the solid waste can retain in a gravi-
tational field without producing continu-
ous downward percolation. A simplified
equation used to define the moisture bal-
ance in refuse mass at any time is:
Moisture Retained = (initial
moisture) + (moisture added) -
(moisture leached)
One component of this moisture balance
equation is the quantity for moisture
leached or leachate volume. Figure 4
depicts the cumulative leachate volumes
measured for the test cells with different
infiltration rates. The plots illustrate
that increasing infiltration rates yield
greater cumulative leachate volumes. Ap-
proximately 15,000 liters of leachate
or moisture have passed through the refuse
mass in Test Cell 4 in about 2,600 days.
Table 7 summarizes moisture balances
for selected test cells at the time of
TABLE 6. WASTE CHEMICAL COMPOSITION
Industrial Waste (Test Cell)
Parameter (Units)
Copper (ppm)
Nickel (ppm)
Arsenic (ppm)
Selenium (ppm)
Mercury (ppm)
Lead (ppm)
Beryllium (ppm)
Cadmium (ppm)
Iron (ppm)
Chromium (ppm)
Cyanide (ppm)
Tin (ppm)
Antimony (ppm)
Vanadium (ppm)
Titanium (ppm)
Boron (ppm)
Chloride*
Asbestos*
Total Solids*
Total Volatile Solids*
Moisture*
Petroleum
SI udqe
(9)
3,500
23
1.0
26.0
10.6
182
4.8
0.50
5,560
125
1.0
NA
NA
NA
NA
7.20
2.35
3.00
21 .00
31.00
79.00
Battery
Waste
(10)
1,125
32
72
180
4.80
3.48*
1.8
29.0
2,950
155
4.2
6,800
1.32*
120
NA
8.10
1.12
208
10.75
7.94
89.25
Electro-
plating
Waste
(12)
100
35
460
4.50
14.7
267
0.25
38.5
1.37*
1.56*
460
NA
NA
NA
NA
19.0
1.35
23.0
20.47
8.98
79.53
Inorganic
Pigment
Waste
(13)
110
10
3.4
16.0
7.60
120
20.2
10.5
1,000
0.50
3.4
NA
NA
40
NA
28.5
10.0
45.0
48.25
22.25
51.75
Chlorine
Production
Brine Sludge
(14)
125
65
14.5
16.5
227
697
ND
0.70
2,000
5.00
14.5
NA
NA
NA
ND
1.70
20.0
110
75.89
1.17
24.11
Solvent-
Based Paint
Sludge
(17)
2.0
0.5
12 8
7.60
16.7
12.6
ND
0.50
150
75.0
12.8
NA
NA
NA
NA
11.4
0.75
9.00
75.25
55.31
24.75
* Values reported as percent by wet weight.
+ Fibers/100 gram.
226
-------
Is"
*"<"..
W IB- 5-°
2 2
111 'o 4.0
> CO
!*,..
i?
CUMULATIVE MOISTURE
INCLUDING INITIAL MOISTURE
too 1100 1*00 jooo
TIME SINCE CELL CONSTRUCTION (days)
Figure 3. Moisture Balance for Cell 2.
400 BOO 1200 KOO 2000
TIME SINCE CELL CONSTRUCTION (days)
Figure 4. Leachate Volumes vs. Infiltration Rate.
227
-------
TABLE 7. MOISTURE BALANCE SUMMARY
To Continuous Leaching
Test
Cell
2
4
9
10
12
13
14
17
Loading Variable
Control
Control
Petroleum Sludge
Battery Waste
Electroplating Waste
Inorganic Pigment Waste
Chlorine Prod. Brine Sludge
Solvent-Based Paint Sludge
Annual
nm
406.4
812.8
406.4
406.4
406.4
406.4
406.4
406.4
Infiltration
Liters/kg
0.577
1.150
0.443
0.479
0.501
0.419
0.313
0.348
No. of
Days
476
163
680
218
163
224
163
476
Initial
Moisture
(L/kg)
0.62
0.62
0.88
0.93
0.99
0.74
0.48
0.50
Moisture
Retained
(L/kg)
1.43
1.51
1.45
1.13
1.38
1.11
0.72
1.01
To 12/31/81
No. of
Days
2,596
2,596
2,460
2,460
2,596
2,596
2,596
2,596
Moisture
Retained
(L/kg)
2.16
2.87
1.62
1.52
1.32
1.41
0.78
0.97
continuous leaching and after seven years
of monitoring. The annual applied infil-
tration is shown on a 1iters-per-kilogram
basis to indicate that moisture input
rates varied according to test cell loading
variables. For instance, though most
cells receive the same actual input rate
(406.4 mm/year), they receive effectively
different input rates when presented as
a function of dry weight.
With the exception of Cell 9, the
hazardous codisposal cells began to leach
at earlier dates than the control cell
(No. 2) and retained less moisture (lowered
field capacity) at the time of continuous
leaching. The low field capacities of
these test cells are partially due to
the inability of the waste additives them-
selves to retain moisture. Thus, in Test
Cells 14 and 17, where proportionately
larger quantities (dry weight basis)
of wastes were added to the refuse, the
overall field capacities are generally
lower since moisture retained is deter-
mined on a liters/dry weight kilogram
basis.
After field capacity has been achieved,
one liter of moisture should theoretically
be leached for each one liter of moisture
added (i.e., input = output). Thus, the
moisture retained should stabilize after
the onset of continuous leaching. As
indicated in Table 7, several test cells
show stable field capacities after seven
years. However, this moisture retention
value increased significantly for the
control cells and certain industrial waste
additives. Notably, Test Cell 2, con-
taining only municipal refuse, increased
its moisture retention to 2.16 I/kg from
1.43 I/kg.
The effect of channeling is one pos-
sible reason for the field capacity in-
creases. Channeling exists when moisture
added to a waste percolates directly
through it, failing to fill all the voids
before leaching. Some voids which remain
in the waste at the onset of continuous
leaching may have gradually filled with
moisture over time. The simulated land-
fills do not behave theoretically and
some downward movement of moisture occurs
before field capacity is attained in a
particular refuse layer. Thus, field
capacity was assumed prematurely for some
cells and thought to be attained prior
to actual field capacity.
Leachate Composition
Summaries of leachate quality analyses
are presented in Figures 5 through 24.
More than seven years of leachate data
have produced significant and predictable
trends for many parameters. Due to the
extensive leachate quality data base,
concentration histories are evaluated
for both the exterior control test cells
(Nos. 1, 2, 3, and 4) and the hazardous
waste additive cells (Nos. 9, 10, 12,
13, 14, and 17). As previously discussed,
the municipal solid waste cells receive
different infiltration rates and thus
provide an indication of the effects of
moisture exchange on basic landfill be-
havior and decomposition. The supporting
figures express parameter concentration
228
-------
as a function of cumulative leachate
volume. The horizontal axis is in cumu- v
lative liters of leachate per kilogram
of dry solid waste and serves to normalize
the various test cells by accounting for
the different loading densities. Since
all the cells have identical depths and
volumes, and similar densities of com-
pacted refuse and computed field capacities,
the summation of leachate volumes is re-
lated directly to the number of moisture
exchanges in each simulated landfill.
That is, when the amount of leachate col-
lected, by weight (liter = kilogram),
equals the amount of solid waste (dry
weight) present in the test cell, then
one moisture exchange has passed through
that cell. In subsequent Figures 5 through
24, the x-axis refers to the number of
moisture exchanges through the waste mass
since historical leachate concentration
changes are probably more related to leach-
ate volume collected (or moisture ex-
changes) than to time.
The introduction of industrial wastes
into the solid waste test cells was an
attempt to provide empirical information
on the practice of codisposing refuse
and selected industrial wastes. Chemical
composition of the six wastes (petroleum
sludge, battery waste, electroplating
waste, inorganic pigment waste, chlorine
production brine sludge, and the solvent-
based paint sludge) at the time of loading
indicated that these wastes contained
significant amounts of metals and other
potential pollutants. Thus, leachate
loadings of pollutant levels were expected
to be higher for these cells relative
to controls.
Leachate pH for all cells generally
remained in the 5.0 to 6.5 range through-
out the study period. Levels were basi-
cally unaffected by waste loading additives
with the exception that some industrial
waste cells had pH levels slightly above
the municipal refuse cells. Early in
the project, leachate pH appeared some-
what erratic followed by slow prolonged
rises with time. However, several cells
showed distinct rises to the range of
6.8 to 7.2, optimal for vigorous anaerobic
activity. In particular, the interior
test cells attained optimal pH levels
which may be attributed to the higher
test cell temperatures and thus, greater
biological activity.
Figure 5 presents leachate pH for
test cells receiving proportional infil-
tration rates. Initial levels dropped
to about 5.0 to 5.25 in the four cells
and gradually increased to 5.5 for Cells
1 and 2, and to about 6.8 for Cell 3.
Test Cell 4, which receives the greatest
infiltration rate (813 mm/yr), exhibited
a similar increasing slope to pH 6.3.
This trend of increasing pH is an impor-
tant indicator of stable anaerobic de-
composition and the lessening influence
or build-up of volatile acids within the
system. Adequate buffer capacity is ob-
viously present in Cells 3 and 4 and ap-
pears to be a function of approximately
4 moisture exchanges (horizontal axis).
The pH levels in Cell 4 are somewhat less
than expected and may indicate a dilution
of the buffering system. This dilution
effect is supported by the field capacity
calculations presented in Table 7. In
upcoming years, pH levels are expected
to increase substantially in Cells 1 and
2 as decomposition continues as a function
of moisture exchange.
The chemical oxygen demand was deter-
mined on monthly samples from the test
cells and no discernable trends could
be noted to distinguish effects between
the municipal refuse controls and the
cells containing industrial waste additives.
COD concentrations declined from initial
peak values of over 60,000 mg/liter to
relatively low levels (less than 15,000
mg/1iter).
Figure 6 illustrates concentration
release plots for COD in Cells 1, 2, 3,
and 4. As shown, initial COD levels were
quite varied, ranging from 60,000 mg/liter
to 90,000 mg/liter. Patterns for COD
declines appear exponential and relative
to moisture exchanges through the test
cells. At about 5.0 cumulative leachate
volumes, leachate strength as measured
by COD levels is greatly diminished (250
mg/liter).
Alkalinity is a measure of the leach-
ate's capacity to neutralize acids. Leach-
ate alkalinity concentrations exhibited
the repetitive trends seen for other leach-
ate parameters of significant declines
from early peak levels at the time of
field capacity. Alkalinity levels were
not influenced by the various industrial
wastes (in Test Cells 9, 10, 14, and 17).
229
-------
0.0 1.0 ».0 S.O 4.0 (.0 «.0
CUMULATIVE LEACHATE VOLUME (l/kg dry solid waste)
Figure 5. Leachate pH vs. Infiltration Rate.
90000.
O>
E
O aoooo.
LU
IU
O
O
5 18000.
HI
O
S.O 1.0
CUMULATIVE LEACHATE VOLUME (l/kg dry solid waste)
Figure 6. Leachate Chemical Oxygen Demand vs. Infiltration Rate.
230
-------
However, alkalinity levels were slightly
higher for the electroplating waste (Cell
12) and slightly lower for the inorganic
pigment waste (Cell 13). Generally, con-
centration histories of the industrial
waste test cells showed greater alkalinity
fluctuations than the other test cells.
Figure 7 presents alkalinity concen-
trations for Cell Nos. 1 through 4 and
indicates similar patterns with respect
to moisture exchanges through the refuse
mass. Test Cell 4 declined to stable
alkalinity concentrations (about 3,000
mg/liter) after only 3.0 moisture exchanges
and may support the fact that the buffer
capacity of this cell was partially washed
out by the high infiltration rate. At
the end of the current monitoring period,
Test Cell 1 remained at the highest al-
kalinity level of these control cells
due to attaining only 2.0 cumulative leach-
ate volumes (I/kg) since November 1974.
Leachate samples were analyzed monthly
for Kjeldahl nitrogen and results are
illustrated in Figure 8. Similarly-timed
peak concentrations were recorded for
the four calls (at field capacity), but
peak values of TKN were less for Test
Cell 1 at 1,350 mg/liter. TKN concentra-
tion histories shown in Figure 8 indicate
that the release of TKN is a function
of moisture exchanges through the refuse
mass rather than time. At the end of
the monitoring period, Test Cell 1 shows
high TKN levels relative to Test Cells
2, 3, and 4. After about 6.0 moisture
exchanges, TKN concentration in Cell 4
stabilizes at 20 mg/liter. As decompo-
sition progresses, Cells 1, 2, and 3 are
expected to show similar behavior.
Leachate iron measurements were con-
ducted for the control and hazardous waste
additive test cells throughout the seven
year monitoring period. The concentra-
tion histories for leachate iron are shown
in Figures 9, 10, and 11 and indicate
great variations among the test cells.
Of the major leachate quality parameters
studied on this project, iron alone did
not exhibit the same general peaking (at
field capacity) and gradual concentration
declines with cumulative moisture through-
out. Instead, iron concentrations in-
creased, decreased, or remained stable
as a function of test cell additives and
waste decomposition.
Figure 9 depicts iron concentration
histories for Cells 1, 2, 3, and 4 and
provides a perspective of the variability
of the iron data. Generally, iron levels
in these cells are stable to slightly
increasing and range from 200 mg/1 to
700 mg/1. Test Cells 1 and 3 show similar
behavior and differ from the patterns
shown in Cells 2 and 4. The variable
behavior may be partially due to pH levels.
As shown earlier in Figure 5, pH levels
were also slightly increasing in Cells
1 and 3.
Figures 10 and 11 illustrate leachate
iron levels found in the control (Cell
2) and the six industrial waste cells.
Levels found in the battery waste cell
(No. 10) were similar to Cell 2. The
other five codisposal cells show elevated
iron values and erratic release curves.
Cells 9 (petroleum sludge), 12 (electro-
plating waste), 13 (inorganic pigment
waste), and 14 (chlorine production brine
waste) all show increasing iron concen-
tration released in the leachate as cumu-
lative moisture throughput increases.
Additional influences on long-term iron
release patterns from the test cells need
to be evaluated.
Plots for leachate cadmium levels
are shown in Figure 12, 13, and 14. Most
of the test cells behaved similarly, with
characteristic peak concentrations at
field capacity reaching 0.30 mg/liter.
Rapid declines in cadmium concentrations
were also apparent as ultimate levels
were in the low parts per billion (less
than 0.05 mg/1). Figure 12 is presented
for the exterior control cells receiving
low to high water infiltration rates (Cells
1, 2, 3, and 4). As shown, the plots
are characteristic of numerous leachate
parameters. Cadmium concentrations are
not clearly a function of cumulative leach-
ate volume or of actual time. Peak con-
centrations are significantly lower for
Cells 1 and 2 and thus these cells attain
low cadmium values rapidly. However,
Test Cells 2, 3, and 4 appear to stabilize
(level out) at about 3.0 moisture exchanges.
Since most cadmium levels in the leachate
remain above the recommended drinking
water standard of 0.010 mg/liter, further
monitoring of the test cells is needed
to assess the release patterns of cadmium
at low levels.
231
-------
O)
i.e 4.0
CUMULATIVE LEACHATE VOLUME (l/ kg dry solid waste)
Figure 7. Leachate Alkalinity vs. Infiltration Rate.
1.0 4.0
CUMULATIVE LEACHATE VOLUME (l/kg dry solid waste)
Figure 8. Leachate Kjeldahl Nitrogen vs. Infiltration Rate.
232
-------
9
.•S »oo.
X
0)
O
oc
CELL 1
CELL 2
CELLS
CELL 4
0.0 1.0 t.O J.O 4.0 t.O «.0 T.O
CUMULATIVE LEACHATE VOLUME (l/kg dry solid waste)
Figure 9. Leachate Iron vs. Infiltration Rate.
/ MUNICIPAL REFUSE (2)
/ PETROLEUM SLUDGE (9)
BATTERY WASTE (10)
ELECTROPLATING
0.0 0.7 1.4 1.1 2.8 >.» 4.1 4.« «.«
CUMULATIVE LEACHATE VOLUME (I/kg dry solid waste)
Figure 10. Leachate Iron for Industrial Waste Cells.
233
-------
MUNICIPAL REFUSE (2)
INORGANIC PIGMENT
WASTE (13)
CHLORINE PROD. BRINE
SLUDGE (14)
SOLVENT-BASED PAINT
SLUDGE (17)
0-0 0.7 1.4 2.1 J.« 1.5 4.2 4.1 5.6
CUMULATIVE LEACHATE VOLUME (I/kg dry solid waste)
Figure 11. Leachate Iron for Industrial Waste Cells.
CUMULATIVE LEACHATE VOLUME (l/kg dry solid waste)
Figure 12. Leachate Cadmium vs. Infiltration Rate.
234
-------
A/.
MUNICIPAL REFUSE (2)
PETROLEUM SLUDGE (9)
BATTERY WASTE (10)
ELECTROPLATING WASTE (12)
0.0 0.7 1.4 2.1 2.0 3.6 4.2 4.t *.«
CUMULATIVE LEACHATE VOLUME (I/kg dry solid waste)
Figure 13. Leachate Cadmium for Industrial Waste Cells.
O)
E
S
a
o
MUNICIPAL REFUSE (2)
INORGANIC PIGMENT WASTE (13)
CHLORINE PROD. BRINE SLUDGE (14)
SOLVENT-BASED PAINT SLUDGE (17)
0.0 0.7 1.4 2.1 >.< 3.1 4.2 4.»
CUMULATIVE LEACHATE VOLUME (I/kg dry solid waste)
Figure 14. Leachate Cadmium for Industrial Waste Cells.
235
-------
Figures 13 and 14 present leachate
cadmium results for the same industrial
waste cells and cadmium concentration
histories behaved similarly to the control
(Cell 2). Only the battery waste cell
(No. 10) and the chlorine production brine
sludge (No. 14) appeared to significantly
differ. Results for these two cells indi-
cate that greater cadmium levels were
found from the time of continuous leaching
to about 1.4 cumulative leachate volumes.
Current cadmium levels for all industrial
waste cells are below 0.030 mg/liter.
Leachate nickel was measured for
the seven year monitoring period and the
data indicated similar concentration his-
tories for most of the test cells. With
two exceptions, all the cells resembled
the plot for Test Cell 2 (control), shown
in Figures 15, 16, and 17. Peak concen-
trations were predictably at field capa-
city and ranged from about 2.0 mg/liter
to 3.0 mg/liter. Nickel concentrations
quickly declined in the leachate samples
to about 0.2 mg/liter after 3.0 cumulative
leachate volumes.
Leachate nickel concentration pre-
sented in Figures 16 and 17 indicate some
differences according to test cell addi-
tives. As might be expected, the electro-
plating waste (Cell 12) contributed the
greatest amount of nickel into the leachate
solution. Peak levels attained were high-
est in this cell and in Cell 14, contain-
ing the chlorine production brine sludge.
All nickel plots show steep concentration
declines from initial high levels, charac-
teristic of landfill leachate behavior.
The remaining cells (Nos. 9, 10, 13, and
17) were roughly similar to the plot for
Cell 2.
Further studies are needed to assess
influences from industrial additives on
landfill behavior. Certainly, dramatic
increases in leachate concentrations as
contributed by industrial wastes will
affect leachate treatment designs. How-
ever, for all leachate parameters moni-
tored in this study, industrial waste
influences on leachate concentrations
were always less than one order of magni-
tude. This relationship occurred even
in the chlorine production brine sludge
cell (No. 14) and the solvent-based paint
sludge (No. 17), where the industrial
additives were loaded at approximately
40 percent (dry weight basis) of the total
waste mass within the cells.
LEACHATE MODELING
Qualitative trends of landfill be-
havior in simulated landfills have been
well documented by various researchers
[1, 2, 3, and 4]. However, the quanti-
tative prediction of leachate composition
and contaminant release can provide a
more useful description of the ongoing
processes of solid waste decomposition
and provide an important method for fore-
casting leachate strength and potential
ground water hazards. Recent efforts
[5, 6, 7, 8, and 9] have yielded concep-
tual mixed reactor models which mathe-
matically describe leachate behavior in
simulated landfills. The purpose of this
section is to assess the semi-empirical
equation developed by Wigh [7] using leach-
ate composition data from the Center Hill
project. As part of this evaluation,
Test Cell 4 is mathematically described
for eleven leachate parameters to demon-
strate that the equation can provide a
good approximation of leachate quality
for simulated landfills.
Wigh's equation is based on two first
order reactions and expresses contaminant
concentration as a function of cumulative
leachate volume, maximum concentration,
and two rate constants. The general pre-
dictive equation describing the concen-
tration change with time is:
_ -(q/V - k)t
-
Modification of Wigh's equation allows
the expression of contaminant concentra-
tion as a function of moisture throughput,
or cumulative leachate volume:
-(q/V - k)t
Where: C = concentration
Cc = maximum concentration
e = mathematical constant
q = leachate flow rate (1/kg/yr)
V = volume of water in the refuse
at field capacity (I/kg)
k = empirical rate constant (yr)
t = cumulative leachate volume
(I/kg)
236
-------
D>
Ł,..
UJ
CELL 1
CELL 2
CELL 3
CELL 4
CUMULATIVE LEACHATE VOLUME (l/kg dry solid waste)
Figure 15. Leachate Nickel vs. Infiltration Rate.
* 4.0
I
A
; \
MUNICIPAL REFUSE (2)
PETROLEUM SLUDGE (9)
BATTERY WASTE (10)
\ ELECTROPLATING WASTE (12)
0.0 0.7 1.4 2.1 J.« >.« 4.1 4.i ».«
CUMULATIVE LEACHATE VOLUME (I/kg dry solid waste)
Figure 16. Leachate Nickel for Industrial Waste Cells.
237
-------
MUNICIPAL REFUSE (2)
INORGANIC PIGMENT
WASTE (13)
CHLORINE PROD. BRINE
SLUDGE (14)
SOLVENT-BASED PAINT
SLUDGE (17)
o.o o.r 1.4 1.1 ».« j.» 4.1 *.» s.t
CUMULATIVE LEACHATE VOLUME (I/kg dry solid waste)
Figure 17. Leachate Nickel for Industrial Waste Cells.
This leachate modeling effort is
based on the above predictive equation.
For the parameters monitored in this pro-
ject, most peak concentrations were mea-
sured at the time of continuous leaching.
Therefore, the value for C0, selected
as the maximum concentration, usually
occurred at the time of field capacity.
A careful determination of C0 is neces-
sary. Initial leachate concentrations
(at the time of continuous leaching) may
be highly variable and extreme values
may bias the model. In addition, this
approach assumes that C0, the maximum
concentration, occurs at initial leach-
ing. This assumption allows the model
to capture the general decrease from
high concentration changes.
The leachate flow rate, q, is as-
sumed to be constant with time and is
determined by plotting cumulative leach-
ate volume versus time for the specific
test cell. Figure 4 graphically plots
total leachate generation for Cell 4.
The slope of the plot was obtained by
least squares analysis to provide the
empirical constant, q, in liters/kilo-
gram dry refuse/year. From this pro-
cedure, q was determined to be 1.17
liters/kg/year.
V was calculated for Test Cell 4
from initial moisture contents and field
capacity determinations as previously
shown in Table 7. V is the total mois-
ture retained in the test cell at field
capacity divided by the total kilograms
of dry refuse as loaded into the test
cell. As noted in Table 7, moisture re-
tention values for Test Cell 4 increased
during project operation. Thus, the value
V may range from 1.51 to 2.87 liters/
kilogram. For this current modeling ef-
fort, V was selected at 1.51 liters/
kilogram dry weight.
The empirical rate constant, k, was
determined by the following procedure
for each modeled leachate parameter. A
straight line equation was developed by
applying the log function to both sides
of the modified predictive equation.
From this, the slope of the line becomes
-(q/V - k)/q, which can be solved by least
squares regression analysis. That is,
the log of concentration (C) is plotted
against cumulative leachate volume on
the x-axis. With the slope determined
and since q and V are known, k is deter-
mined by subtraction, where:
k = slope(q) + (q/V)
Determination of the predicted leach-
ate concentration history curve for a
given parameter becomes straightforward.
The appropriate C0, q, V, and k values
are established for a specific test cell
and development of the predictive concen-
238
-------
tration, C, is by cumulative leachate
volume data points.
Comparison of the predictive equation
with the actual concentration histories
measured in Test Cell 4 are shown in Figures
18 through 28. Test Cell 4 receives the
highest water infiltration rate, and there-
fore, has the most developed concentration
curves. Reasonably good fit to the data
is evident except for total phosphorus
(Figure 22), total solids (Figure 24),
and iron (Figure 25). For both total
phosphorus and total solids, the curve
for the predictive equation yielded high
estimates for leachate concentrations
when compared with actual results. Con-
versely, leachate iron in Test Cell 4
remained significantly above the predic-
tive curve through 7.0 cumulative leach-
ate volumes.
Computed rate equations are listed
in Table 8 for each of the parameters
in Figures 18 through 28. The first or-
der equations can be used to determine
the predicted leachate strength at any
given cumulative leachate volume.
SUMMARY
This project simulated the decompo-
sition of landfilled municipal solid waste
and tested effects on that decomposition
process due to the addition of calcium
carbonate, sewage sludge, and industrial
wastes to specific test cells. In pre-
vious lysimeter studies, decomposition
patterns are most frequently measured
by leachate composition since the repe-
titive shape of leachate concentration
curves has been well documented. For
the Center Hill project, comparable re-
sults are reported for leachate parameters
as well as for gas composition and gene-
ration.
More importantly, though, the experi-
mental program at Center Hill is unique
in its attempt to demonstrate landfill
behavior. The uniqueness of the research
experiment has developed because it con-
tains the following:
• Long-term monitoring effort
• Controlled moisture input/output
program
0 Extensive analytical efforts, both
initial and ongoing
• Sufficient control cells
• Duplicate test cells
• Cells constructed to measure gas
production
• Nutrient (sludge) and buffer addition
• Industrial waste codisposal
Each of the above serve to provide
a progressive understanding of the pro-
cesses that occur within a sanitary land-
fill and a better assessment of the re-
lated environmental effects. The interim
results presented in this paper address
the timing of leachate stabilization under
controlled conditions, and determine some
TABLE 8. RATE EQUATIONS FOR TEST CELL 4.
Parameter Rate Equation*
Specific Conductance - 17,500e~°'276t
COD • 75,000e~°-72t
TOC > 30,000e~°-64t
Alkalinity = 13,000e~°'33t
Total Phosphorus - 325e~°'56t
TW = l,700e~°'53t
Total Solids = 55,000e~°-44t
Iron = 210e-°-°9t
Zinc = lOOe'1'3'
Cadmium = 160e~°'44t
Nickel - 1.9e~°'53t
Units Rate Constant, k
imcrosiemans 0.45
mg/liter 0.07
mg/lite 0.03
mg/lite
mg/lite
mg/lite
mg/lite
mg/lite
mg/lite
ug/lite
mg/lite
0.39
0.12
0.15
0.26
0.67
0.75
0.26
0.15
* Rate equation constants based on cell-specific conditions.
t - Cumulative leachate volume (liters/kilogram dry solid waste).
239
-------
PREDICTIVE EQUATION
TEST CELL 4
CUMULATIVE LEACHATE VOLUME (I/kg dry solid waste)
Figure 18. Predicted Specific Conductance Concentration.
PREDICTIVE EQUATION
TEST CELL 4
1.0 M 1.0 4.0 «.0 1.0
CUMULATIVE LEACHATE VOLUME (I/kg dry solid waste)
Figure 19. Predicted COD Concentration.
240
-------
»000-
.•Ł >oooo
~
01
2 24000 •
o
00
CC
1«000<
i
OC 11000.
0
1
O aooo-
\ PREDICTIVE EQUATION
\ TEST CELL 1
\ \
\ \
\ \
\ \
v\
^v
v\^
s>^^
^^X-.^
1.0 1.0 t.O 4.0 1.0 *
CUMULATIVE LEACHATE VOLUME (I/kg dry solid waste)
Figure 20. Predicted TOC Concentration.
Cl 12000-
<
_l
PREDICTIVE EQUATION
TEST CELL 4
-T
T.O
T™
1.0
-TT"
J.O
—r~
4.0
CUMULATIVE LEACHATE VOLUME (I/kg dry solid waste)
Figure 21. Predicted Alkalinity Concentration.
241
-------
PREDICTIVE EQUATION
TEST CELL 4
CUMULATIVE LEACHATE VOLUME (I/kg dry solid waste)
Figure 22. Predicted Total Phosphorus Concentration.
PREDICTIVE EQUATION
TEST CELL 4
1.0 1.0 9.0 4.0 i.O «.0
CUMULATIVE LEACHATE VOLUME (I/kg dry solid waste)
Figure 23. Predicted TKN Concentration.
242
-------
PREDICTIVE EQUATION
TEST CELL 4
CUMULATIVE LEACHATE VOLUME (I/kg dry solid waste)
Figure 24. Predicted Total Solids Concentration.
PREDICTIVE EQUATION
TEST CELL 4
"C »oo.
0
Ul
E eoo-
1.0 J.O 1.0 4.0 «-» «-«
CUMULATIVE LEACHATE VOLUME (I/kg dry solid waste)
Figure 25. Predicted Iron Concentration.
243
-------
N
PREDICTIVE EQUATION
TEST CELL 4
s^^.^
0.0 1.0 2.0 1.0 4.0 f.0 t.0
CUMULATIVE LEACHATE VOLUME (I/kg dry solid waste)
Figure 26. Predicted Zinc Concentration.
PREDICTIVE EQUATION
TEST CELL 4
0.0 1.0 LO 1.0 4.0 «-0 «-0
CUMULATIVE LEACHATE VOLUME (I/kg dry solid waste)
Figure 27. Predicted Cadmium Concentration.
244
-------
01
•** 3.0,
UJ
s
Z 2.0 4,
PREDICTIVE EQUATION
TEST CELL 4
'•• >•<> >.0 4.0 «.<> «J
CUMULATIVE LEACHATE VOLUME (I/kg dry solid waste)
Figure 28. Predicted Nickel Concentration.
of the long-term impacts associated with
the disposal of selected hazardous wastes
into municipal solid waste. An evaluation
of current findings is presented below:
1. Leachate quantities collected from
six hazardous waste codisposal cells
and municipal refuse control cells
indicated that codisposal cells pro-
duce leachate at earlier dates than
the control cell, and that the Co-
disposal cells retain less moisture
(lowered field capacity) for the dura-
tion of the study period. These re-
sults suggest that precipitation will
not be as readily retained in an
actual codisposal landfill for wastes
having physical characteristics simi-
lar to those used in this study, and
that earlier leachate production
should be expected.
2. Long term leachate concentration his-
tories for comparing cell performances
were expressed as a relationship of
cumulative leachate volume rather than
time. Concentrations of metals (e.g.,
iron, cadmium, and nickel) were higher
in leachates from industrial waste
codisposal cells but leachate strength,
as measured by COD, pH, and alkalinity,
was similar for the codisposal and
municipal refuse control cells. Iron
behaved differently than other leach-
ate constituents and showed gradual
increases in concentration during
the study period for the municipal
refuse control cells. For codis-
posal Cells 9, 10, 13, 14, and 17,
iron concentrations remained signi-
ficantly higher over the controls.
Leachate concentrations generally
peaked within a few months of the
achievement of field capacity. Typi-
cal for many of the parameters was
the peak concentration occurring at
field capacity, followed by slow de-
clines thereafter. This occurrence
of initial peak concentrations in
conjunction with earlier leachate
production in hazardous waste co-
disposal test cells suggests that
on-site leachate treatment systems
for codisposal landfills should be
capable of receiving the high strength
leachates during the early stages
of landfill operation.
The controlled moisture infiltration/
leachate collection program coupled
with the extensive leachate quality
data base has provided empirical
evidence that leachate strength de-
creases as a function of moisture ex-
change through the refuse. Since
245
-------
leachate strength is a primary indi-
cator of landfill stabilization and
decomposition, the data show that the
initial loading of the six industrial
hazardous waste did not significantly
impact the decomposition of municipal
refuse.
5. Test Cell 14 (chlorine production
brine sludge) exhibited higher peak
concentrations and elevated leachate
concentration histories most often
among the six industrial additive
cells.
6. Long term leachate concentration
trends were developed for Test Cell
4 which received the greatest amount
of infiltration (813 mm/year). A semi-
empirical model, incorporating para-
meters for the major factors influ-
encing landfill decomposition and
leaching, was calibrated with data
from Test Cell 4. The validity of
the model will be tested with data
from other test cells and should be
tested against comparable field data.
Further efforts as part of this re-
search program will include expansion of
applied forecasting techniques. The quan-
titative prediction of leachate composi-
tion and contaminant release can provide
a more useful description of the ongoing
processes of solid waste decomposition
and provide important methods for fore-
casting leachate strength and potential
ground water hazards. Further batch-type
predictive techniques will be developed
which rely not only on moisture flow
through the refuse mass, but also on dilu-
tion/mixing and fundamental transport
phenomena which occur within the landfill
environment.
REFERENCES
1. Pohland, F.G. Sanitary Landfil1 Sta-
bilization with Leachate Recycle and
Residual Treatment. U.S. Environ-
mental Protection Agency, Report No.
EPA-600/2-75-043. 1975.
2. Chian, E.S.K., and F.B. DeWalle.
Effects of Moisture Regimes and Tem-
perature on Municipal Solid Waste
Stabilization. In: Fifth Annual Re-
search Symposium on Municipal Solid
Waste: Land Disposal, U.S. Environ-
mental Protection Agency, Report No.
EPA-600/9-79-023. 1979.
3. Pohland, F.G., et al . Pilot-Scale
Investigations of Accelerated Landfill
Stabilization with Leachate Recycle.
In: Fifth Annual Research Symposium
on Municipal Solid Waste: Land Dis-
posal, U.S. Environmental Protection
Agency, Report No. EPA-600/9-79-023.
1979.
4. Walsh, J.J., et al. Demonstration
of Landfill Gas Enhancement Techniques
in Landfill Simulators. In: Pro-
ceedings from Seventh Annual Sympo-
sium on Energy from Biomass and
Wastes, Institute of Gas Technology.
January 1983.
5. Qasim, S.R., and J.C. Burchinal.
Leaching from Simulated Landfills.
Journal of the Water Pollution Con-
trol Federation, 42(3), 1970.
6. Raveh, A., and Y. Avnimelech. Leach-
ing of Pollutants from Sanitary Land-
fill Models. Journal of the Water
Pollution Control Federation, 51(11).
1979.
7. Wigh, R.J. Boone County Field Site,
Final Report. U.S. Environmental
Protection Agency, Cincinnati, Ohio.
1982.
8. Straub, W.A., and D.R. Lynch. Models
of Landfill Leaching: Moisture Flow
and Inorganic Strength. Journal of
the Environmental Engineering Divi-
sion, ASCE, 108(2). 1982.
9. Straub, W.A., and D.R. Lynch. Models
of Landfill Leaching: Organic Strength.
Journal of the Environmental Engi-
neering Division, ASCE, 108(2). 1982.
ACKNOWLEDGEMENTS
The work upon which this paper is
based is being performed under Contract
No. 68-03-2758 with the U.S. Environmental
Protection Agency, Solid and Hazardous
Waste Research Division in Cincinnati,
Ohio. Early project work was performed
by Systems Technology, Inc., under EPA
Contract No. 68-03-2120. The authors of
this paper would like to express their
appreciation to Ms. Norma M. Lewis (Pro-
ject Officer) and Messrs. Dirk R. Brunner,
Douglas C. Ammon, and Stephen C. James.
246
-------
APPLICATION OF SOLIDIFICATION/STABILIZATION
TECHNOLOGY TO ELECTROPLATING WASTES
Philip G. Malone, Larry W. Jones, and J. Pete Burkes
USAE Waterways Experiment Station
Vicksburg, Mississippi 39180
ABSTRACT
Electroplating wastes are generally composed of fine-grained metal hydroxides gener-
ated from the treatment of spent plating baths, rinse water, and liquid from cleaning and
etching operations. A number of solidification/stabilization systems have been applied
to these residues in order to develop a waste material with superior containment proper-
ties. Systems for developing new, high-temperature, stable chemical compounds have found
application for immobilization of specific metals. Low-temperature silicate polymers
provide a lower degree of containment. Organic polymer systems that do not involve
acidification have proved to be beneficial. Thermoplastic encapsulation systems provide
excellent containment. Analyses of water leachates from crushed pozzolan-Portland and
pozzolan encapsulated materials showed higher concentrations of some metals than were
obtained from water leaches of the untreated wastes. An X-ray diffraction study of
pozzolan-Portland and pozzolan encapsulates indicates no crystal chemical involvement of
the waste in the setting and hardening process. Containment in these materials comes
from encasing the metal hydroxides in the cementitious matrix.
INTRODUCTION
Electroplating typically involves
surface cleaning and surface etching,
plating, and rinsing the plated parts.
Each step generates a series of acid or
alkaline solutions that contain poten-
tially toxic metals in solution (such as
cadmium, chromium, copper, nickel, lead,
and zinc). The high metal concentrations
are derived from spent processing solu-
tions, rinse water or "drag out," cleanup
from spilled materials, and corrosion of
plating tanks and piping. Wastewater from
an electroplating operation also can con-
tain cyanides from stripping and plating
and organic solvents, oil, and grease from
stripping and metal cleaning operations.
The waste streams from plating opera-
tions are typically segregated and treated
as three separate types of liquid wastes:
a) cyanide-containing wastes,
b) chromium-bearing waste,
treated with FeSO^, S02, or NaHSO
reduce chromium to the trivalent
c) acidic wastes.
The cyanide-containing waste are
generally treated by alkaline chlorina-
tion to oxidize cyanide. The chromium-
bearing solutions are acidified and
to
:orm.
When reduction is complete the chromium
is precipitated as Cr(OH)- using lime
slurry. The acid waste stream is gen-
erally neutralized with lime. Each
process generates a lime-neutralized
metal hydroxide sludge that must be
handled as a hazardous waste.
The Environmental Protection Agency
has recently published new regulations
to improve effluent quality from metal-
finishing facilities (30). These regula-
tions may be expected to increase the
amount of sludge requiring disposal.
Metal hydroxide sludges have generally
presented problems in disposal due to the
potential for dispersal and dissolution
(4).
247
-------
The purpose of this paper is to re-
view the work to date on solidification/
stabilization processes applied to elec-
troplating waste (metal hydroxide sludges)
and to point out technical problems and
complications associated with handling
these materials. The discussion in this
paper will focus on the basic chemistry of
electroplating waste and reactions in-
volved in generic types of waste solidifi-
cation and should not be interpreted as
indicating the performance to be expected
from any proprietary process. Proprietary
processes frequently include specific
adaptations of general solidification/
stabilization systems discussed here and
may perform very differently from the
systems discussed.
Physical, Chemical and
Minerological Characteristics
of Electroplating Waste Sludges
The physical characteristics of a
typical electroplating wastewater treat-
ment sludge are summarized in Table 1. A
typical grain size distribution is pre-
sented in Figure 1.
TABLE 1. PHYSICAL CHARACTERISTICS OF A
TYPICAL ELECTROPLATING WASTEWATER
TREATMENT SLUDGE
Characteristic
Value
% less than 200 mesh
Median grain size (mm)
Percent solids (%)*
Liquid limit (%)
Plastic limit (%)
Plasticity index (%)
USCS classification
Specific gravity
Dry unit wt (kg/cu meter)
Void ratio
Coefficient of
permeability (cm/sec)
95
0.015
36.7
107
58
49
MH
2.70
514.18
4.67
1.69 x 10~5
*Average of two.
The electroplating sludge samples re-
ported in the literature generally have
fine particle sizes, low permeability, low
strength, and high compressibility (1).
These properties generally make them un-
desirable for inclusion as aggregate in
concrete where strength and grain size of
U.S. STANDARD SIEVE
200 MESH
O
cc
UJ
UJ
u
oc
100
80
60
40
20
i.
ELECTROPLATING WASTE
CLAY^
O
UJ
20 g
40 oc-
UJ
CO
oc
60 <
80 Z
100 10 1 0.1
GRAIN SIZE MM
0.01
100
u
oc
111
Figure 1. Grain size distribution of a typical electroplating
waste sludge (after Bartos and Palermo, 1977).
248
-------
aggregates must be regulated to assure
that the aggregate does not prevent
strength development.
In general, the percent of material
passing 200 mesh must not be more than
3 percent of the total aggregate. Clay
in lumps presents special problems in con-
crete and is often limited in specifica-
tions to 1 percent of the mix (12).
Dewatered sludges typically have over
90 percent of the material finer than
200 mesh and contain lumps of clay-sized
particles. They are unsuitable for use
as standard aggregate, and make poor con-
struction materials for dikes or embank-
ment s.
The chemical characteristics of sev-
eral electroplating sludges are summarized
in Table 2. These wastes are varying mix-
tures of fine-grained metal hydroxides
along with salts from materials added to
neutralize the liquid wastes.
X-ray diffraction examination of
powder mounts of the sludge showed the
characteristic diffraction peaks for gyp-
sum (CaSO,' 2H«0) and traces of hydromica
(clay) and quartz. The metal hydroxides
are amorphous. No peaks for calcium
hydroxide, a commonly-used neutralizing
agent, were found. X-ray analysis of
material that was leached for two years
with distilled water saturated with carbon
dioxide showed a slight decrease in dif-
fraction peaks associated with the gypsum.
The metal hydroxides remained amorphous
and no formation of metal carbonates or
other phase transformations were detected
(15).
Solidification/Stabilization Tech-
niques Employed with Electroplating Wastes
Electroplating wastes are a mixture,
consisting primarily of amorphous metal
hydroxides, that has a clay-like consis-
tency. The majority of solidification/
stabilization processes evaluated for the
disposal of electroplating waste do not
attempt to change the nature of the com-
pounds in the sludge; they are intended to
isolate the metal hydroxides from leaching
TABLE 2. CHEMICAL COMPOSITIONS OF ELECTROPLATING WASTEWATER TREATMENT SLUDGES
Sample
Reference
Constituent
As
Be
Cd
Cr
Cu
Fe
Hg
Ni
Pb
Zn
Cl
CN
A B
(25) (8)
460**
0.25
38.5** 8000
15600** 105000
100 32000
13700**
23.0
35 10000
267
—
13500
460
C
(28)
24
840
1800
150000
95000
—
0.53
6100
1400
18000
—
D
(11)
—
—
580
65000
100000
10000
—
—
600
7900
—
*All analyses are in mg/kg on a dry weight basis.
**Maximum concentrations observed in sludge samples analyzed,
249
-------
water. Other processes attempt, with
varying degrees of success, to produce new
insoluble, durable compounds that retain
the potentially toxic metals from leaching
after shallow land burial. In some cases,
especially where possible low-temperature
silicate or carbonate reactions are in-
volved, it is difficult to declare which
chemical and physical mechanisms are in-
volved. From the work done to date at
Waterways Experiment Station and other
laboratories, it is possible to reach some
general conclusions as to the mechanism
for waste containment and the applicabil-
ity of different techniques to electro-
plating wastes.
Production of New Insoluble Compounds
A number of high temperature modifi-
cations of heavy metal (especially chro-
mium) sludges have been reported by Inoue
and others (10) and by Shoto and Hattori
(23, 24), and Hattori, Shoto, and Nagaya
(6). The formation of new crystalline
phases has been documented using X-ray
diffraction and solubility studies.
Uvarovite (Ca Cr Si 0 ) and Cr^ were
found in X-ray diffraction patterns of
solidified chrome plating wastes. For a
complex waste containing cadmium, copper,
nickel, lead, and zinc it may be necessary
to develop a number of very specific sili-
cates to provide carriers for the various
metals. Containment would be very effi-
cient if total chromium-to-garnet conver-
sion were accomplished.
Rousseaux and Craig (21) reported
testing a low-temperature silicate process
for immobilizing heavy metals in electro-
plating wastes. The process involves the
addition of a siliceous reagent to an acid
(pH = 1.5) metal solution. The pH is then
raised by the addition of caustic and a
polymer forms. The polymer mixture is
hardened with the addition of Portland
cement. Extraction procedure testing
showed that with the exception of cadmium,
no heavy metals leached in excess of the
regulatory criteria. The exact mechanism
involved in this technique was not dis-
cussed although the process description
indicates a new siliceous polymer is
formed.
Formation of low temperature silicate
polymers that may include heavy metals
have also been suggested as possible im-
mobilization systems in other processes
involving mixing soluble silicates, elec-
troplating wastes, and Portland cement (7,
22). Additional work is required to
demonstrate the binding mechanisms in-
volved in these systems.
Low temperature reactions between
soluble silicates and metal salts (other
than alkali metals) or metals in solution
can generate a variety of precipitates
that are non-crystalline, non-homogeneous,
and difficult to characterize. When dis-
solved metal salts and soluble silicate
are mixed, the composition of the precipi-
tate depends on the proportions of the
reactants available at the point of mix-
ing. The composition of the first part of
a precipitate differs from the last. Even
if the pH is kept low during mixing to
prevent metal hydroxide formation, the
soluble silica will be continuously poly-
merizing and the proportion of metal bound
to the silica depends on the extent of
silica polymerization that occurs before
the metal binding occurs. The metal con-
tent is even more poorly controlled if
the pH i;- :ot kept low. Metal silicates
precipitate at pH's only slightly lower
than the pK for corresponding hydroxide
formation, therefore, the reaction yields
metal hydroxide or polymeric basic metal
hydroxide that cannot combine in any
orderly way with the precipitating silica.
The resulting material may not contain any
definite compounds and the metal hydrox-
ides precipitate along with the silica by
mutual coagulation (9, 31).
Disordered gel structures offer
greater potential for ion migration than
crystalline materials. Diffusion reac-
tions in silica gel are a common method
for production of specific crystalline
compounds (5, 16). The binding of metals
on the surface of silica gel is similar
chemically to the bonding in hydroxocom-
plexes in metals. This has been shown by
examining pH's at which various metals
adsorb to silica gel surfaces. The pH
drop for release of metals from silica gel
treated waste may in part be related to
the pH change required for adsorption. If
this is so, the order of release with low-
ering pH for some of the metals would be
Cd, Cu, Pb, and Fe. In the extractant
analysis from Rousseaux and Craig (21) the
concentration of the metals always follows
the order Cd, Cu, and Pb in decreasing
concentration, although this is not the
order of concentrations in the waste
250
-------
(Table 3). Dissolution of metal complexes
on silica gel may be a major system of
metal release in these silica gel mate-
rials.
Organic compounds (metal stripper or
cleaners) are generally not compatible.
with concentrated soluble silicate solu-
tions. Polar organic solvents can dehy-
drate or separate from an aqueous phase.
In hydrophobic organic solvents, silicate
solutions separate into the aqueous
phases. A few organic compounds such as
glycerin or ethylene glycol are miscible
in silicate solutions (17). Organic com-
ponents in electroplating waste would
rarely be chemically bound in silica gel-
based treated containment products.
Encapsulation Systems
Encapsulation systems used with elec-
troplating wastes depend on the isolation
of the metal hydroxide waste in some
organic or inorganic matrix that prevents
contact with the surrounding environment
(especially surface water or groundwater).
The scale of encapsulation and the mate-
rials involved are usually used to
classify the treatment process as macro-
or microencapsulation. The microencap-
sulation systems usually involve coating
of individual particles of precipitate.
Macroencapsulation involves coating large
aggregates of particles as a single unit
with a coating that is very much thicker
than a single particle diameter.
The major macroencapsulation system
that has been developed and tested with
electroplating waste employs a fused poly-
ethylene jacket on a core of polymer-
cemented wastes. The proposed waste
blocks would be on the order of one meter
on a side with a 0.5 cm jacket (13). Two-
year leaching tests conducted with the
material have demonstrated its effective-
ness.
Microencapsulation systems have been
developed with a variety of materials, in-
cluding organic polymers, thermoplastic
materials, pozzolan-Portland cement mix-
tures, and pozzolan cements. Most micro-
encapsulation systems do not involve any
reaction between the matrix and the waste;
although in the case of silicate-based
systems this has never been completely
TABLE 3. COMPARISON OF CONCENTRATIONS OF SELECTED METALS IN BLENDED
ELECTROPLATING WASTE AND IN EP EXTRACTANTS FROM SILICATE
POLYMER TREATED BLENDED ELECTROPLATING WASTE*
Calculated Concentration in Blended Electroplating Waste
Cd Cu Pb
Untreated Waste
2890**
7100
1710
Concentrations Found in EP Extracts from Treated Blended Waste
Run No.
1
2
4
5
5.60
2.30
0.10
2.70
0.28
0.64
0.06
0.36
0.013
0.016
0.006
0.017
*From Rousseaux and Craig (21).
**A11 values are in mg/liter.
251
-------
demonstrated and reactions between waste
and pozzolan have been considered possi-
ble.
Organic polymer solidification sys-
tems were originally developed for nuclear
waste disposal and in some cases require
that acidic reagents be used to ini-
tiate polymerization (for example,
urea-formaldehyde systems). For some
wastes this is not a problem; but it
is with metal hydroxide wastes. Test-
ing done with systems where the sludge
pH was lowered to 2.8 has demonstrated
that a major loss of metal hydroxide
waste occurs (15). Other problems
with organic polymer encapsulation
systems can result from undesirable
oxidation-reduction reactions between
wastes and any water soluble polymeriza-
tion promoters and initiators used in
the system. In polyester-based systems
promoters are reducing agents (such as
cobalt naphthenate) that can react with
any chromate (Cr ) present in the plat-
ing waste. Initiators are commonly
strong oxiders such as methyl ethyl per-
oxide or hydrogen peroxide that oxidize
reduced metals (especially Cr ) in the
waste. Undestroyed cyanide in electro-
plating wastes was found to complex metals
in the polymerization promoter (cobalt
naphthenate) used in a polyester system
that was tested. This complicated the
polymerization process and additional pro-
moter was needed to complete solidifica-
tion (26, 27).
Thermoplastic microencapsulation sys-
tems contain the solid particulate waste
by mixing the sludge in a heated polymeric
material, such as asphalt or polyethylene.
The polymer system coats each grain and
prevents contact between the waste and any
surrounding groundwater or surface water.
Asphalts and asphalt blends are generally
the least expensive materials for waste
incorporation. Concentrations of heavy
metals measured in EP extracts and dis-
tilled water leachates prepared from
asphalt microencapsulated electroplating
waste are given in Tables 4 and 5.
TABLE 4. RESULTS OF EP TESTING OF TREATED AND UNTREATED TOXIC METAL SLUDGES
Solidification
Treatment As Cd Cr Cu Hg Pb Ni Se Zn
Cement- based
A
B
Average**
<0.01*
<0.01
<0.01
<0.
0.
0.
0002
0006
0004
0.308
0.299
0.304
0.153
0.150
0.152
<0.
<0.
<0.
0002
0002
0002
<0.001
<0.001
<0.001
0.006
0.005
0.006
<0.010
<0.010
<0.010
<0.05
<0.05
<0.05
Pozzolan-based
A <0.01 0.0022 0.554
B <0.01 <0.0002 0.540
Average <0.01 0.0012 0.547
0.104 <0.0002 <0.001 <0.003
0.068 <0.0002 <0.001 <0.003
0.086 <0.0002 <0.001 <0.003
<0.010 <0.05
<0.010 <0.05
<0.010 <0.05
Asphalt-based
A
B
Average
Untreated
A
B
Average
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.
<0.
<0.
0.
0.
0.
0002
0002
0002
643
912
778
<0.001
<0.001
<0.001
0.350
0.341
0.346
0.007
0.007
0.007
2.72
2.50
2.61
<0.
<0.
<0.
<0.
<0.
<0.
0002
0002
0002
0002
0002
0002
<0.001
<0.001
<0.001
<0.001
0.004
0.002
<0.003
<0.003
<0.003
0.427
0.441
0.434
<0.010
<0.010
<0.010
<0.010
<0.010
<0.010
<0.05
<0.05
<0.05
2.83
3.01
2.92
*A11 values are in mg/liter.
**Values below the limit of detection were considered to be at the limit of detection in
calculating averages when above detection limit results were obtained on the duplicate
samples.
252
-------
TABLE 5. RESULTS OF DISTILLED WATER LEACHING OF TREATED
AND UNTREATED ELECTROPLATING WASTE
Solidification
Treatment
As
Cd
Cr
Cu
Hg
Pb
Ni
Se
Pozzolan-based
A 0.033 0.0027 5.67 1.06
B 0.013 0.0021 6.87 1.06
Average 0.023 0.0024 6.27 1.06
<0.0002 <0.001
<0.0002 <0.001
<0.0002 0.001
Zn
Cement-based
A
B
Average**
0.
0.
0.
043*
043
043
0.0036
0.0035
0.0036
2.68
2.86
2.77
1.18
1.39
1.28
0.
0.
0.
0002
0005
0004
<0.001
<0.001
<0.001
0.075
0.077
0.076
<0.010
<0.010
<0.010
<0.050
0.104
0.077
0.038 <0.010 <0.050
0.052 O.010 <0.050
0.045 <0.010 <0.050
Asphalt-based
A
B
Average
Untreated
A
B
Average
<0
<0
<0
0
0
0
.01
.01
.01
.056
.078
.067
-------
In tests where this has not been done,
asphalt microencapsulation has been much
less successful in immobilizing zinc, cop-
per, and cadmium (20).
Microencapsulation systems using
pozzolan-Portland cement and pozzolan
cement have been widely used to produce
better containment of nuclear, industrial,
and military wastes. The general mecha-
nism proposed has involved the mixing of
wastes and unhydrated cementitious mate-
rials and water. The gelling of the
cement encases the particles of waste in a
network of silicate fibrils formed during
hydration.
Solidifying Portland cement or pozzo-
lanic cement materials produces a mixture
of gel, semicrystalline, and crystalline
structures, with calcium silicate gel as
the major constituent. Other reaction
products are crystalline hydrates formed
from calcium silicate, calcium aluminates,
calcium sulfoaluminates, and calcium sul-
phoferrites (3, 19). Some of the crystal-
line calcium silicate hydrate phases have
variable composition and can accept mag-
nesium, sodium, potassium, chromium, or
manganese in the crystal matrix (18).
Pozzolan-Portland cement- and
pozzolan-encapsulated electroplating
wastes were examined to determine if any
new crystalline phases or changes in the
X-ray lattice spacings of standard crys-
talline phases from hydrated cement
(suggesting metal substitutions) could be
detected. The two treated wastes used in
this study each consisted of approximately
16 percent electroplating wastes (Sample D
in Table 2) on a dry weight basis (11).
Traces of the X-ray patterns are
shown in Figures 2 and 3. Photos of
crushed grain mounts are shown in Fig-
ures 4 and 5. No new crystalline phases
were observed. Phases typical of carbon-
ated hydrated Portland cement pastes
(Table 7) were found in the pozzolan-
Portland treated sample, but no changes in
lattice spacing that could be attributed
to metal substitutions were noted. Simi-
larly in the pozzolan encapsulate, no evi-
dence could be found for involvement of
the waste in the crystalline phases that
were observed. In both treated wastes,
the presence of particles of amorphous
metal hydroxides were noted in microscopic
examination.
The wastes examined by X-ray diffrac-
tion were also subjected to Extraction
Procedure (EP) testing and distilled water
leaching as described for the asphalt
waste. The results of the EP extract
analyses and the distilled water leach
analyses are presented in Tables 4 and 5.
The slurry pH's for the wastes are given
in Table 6.
In the EP testing where the waste was
in the form of a monolithic cylinder,
waste containment was improved for all of
the potentially toxic metals examined ex-
cept for chromium in the pozzolan-based
treatment system. Chromium may have been
added in the reagents used for solidifica-
tion.
The chromium and copper levels in the
distilled water leachates are higher in
leachates from the treated wastes than in
the untreated wastes. The reason for this
is unknown. The slurry pH's observed with
the treated wastes were higher than in the
untreated wastes; but in both cases the
pH's were closer to the pH's of minimum
solubility for these metals as hydroxides
(8.5 for chromium and 9.0 for copper) than
was the pH for the untreated waste. The
slurry pH alone would have suggested that
lower, not higher, concentrations of chro-
mium and copper should be observed in the
microencapsulated wastes.
A similar problem was noted with
silicate polymer-cement treated wastes
(21). Water extract of solidified waste
with Portland cement and silicate polymer
showed higher concentrations of cadmium
and chromium than some samples that were
similar except no Portland cement was
added. The reason for this was not under-
stood.
Radioisotope tracer studies of low
concentrations (1.5-2.0 percent of the
solid treated wastes) of metal hydroxide
sludges encapsulated in concrete (Portland
Type I cement) showed an early release of
metal ions into a water leach and later
resorptlon (14). The X-ray diffraction
work reported in the present study sug-
gests that the resorption process that was
noted may be coprecipitation of the heavy
metals with calcium as carbonates in cal-
cite or vaterite.
Reactions between carbonate in
solution and calcium sulfoaluminates
254
-------
ssz
3 oo
o> c
3 i-l
rt fD
(Ti ro
3 .
O
01
XI X
en I
C i-l
Q, H-
t-h
CD Hi
M i-j
ro ai
o o
rf rt
i-l H-
O O
13 3
M
(B XI
3 rt
CW rt>
§3
cn i-h
rt i-l
(B O
O
N
N
O
3
13
O
3
CL
INTENSITY INCREASES
QUARTZ
CALCITE
. MAGNETITE + ETTRINGITE
ETTRINGITE
. ETTRINGITE + HEMATITE
VATERITE
INTENSITY INCREASES
CALCITE
ETTRINGITE
VATERITE
CALCITE + ETTRINGITE
ETTRINGITE
VATERITE + ETTRINGITE
QUARTZ
S -
» ETTRINGITE
CALCITE
CALCITE
CALCITE
ETTRINGITE
ETTRINGITE
ETTRINGITE (?)
QUARTZ
CALCITE + ETTRINGITE
CALCITE (?) + ETTRINGITE
CALCITE + MAGNETITE
QUARTZ
ETTRINGITE
QUARTZ + CALCITE
-------
9JZ
ro nj
3 H-
O 05
» C
13 i-i
tn ro
M W
(U .
rt
i-(
M W
rc >
-------
Figure 4. Photomicrograph of crushed pozzolan-Portland
cement encapsulated electroplating waste. The spherules
are siliceous material from fly ash; the fine transparent
crystals are hydrated Portland cement; the large amorphous
masses are metal hydroxides. Photo was made with polarized
light, under crossed nicols with a gypsum plate.
Figure 5. Photomicrograph of crushed pozzolan-encapsulated
electroplating waste. The spherules are siliceous material
from fly ash; the amorphous masses are metal hydroxides.
Photo was made with polarized light, under crossed nicols
with a gypsum plate.
257
-------
TABLE 7. PHASES IN POZZOLAN-PORTLAND CEMENT AND
POZZOLAN-TREATED ELECTROPLATING WASTES
Pozzolan-Portland Cement Treated Waste
Phases Identified
Probable Source
Ettringite
Tetracalcium aluminate
carbonate-11-hydrate
CaLcite
Vaterite
Quartz
Magnetite
Hematite
Calcium silicate hydrate
Ettringite
Gypsum
Calcium sulfate hemihydrate
Mullite
Quartz
Magnetite
Hematite
Pozzolan Treated Wastes
Hydrated Portland cement
Hydrated Portland cement
Hydrated Portland cement
Hydrated Portland cement
Fly ash
Fly ash
Fly ash
Reaction of fly ash and lime
Reaction of fly ash and lime
Waste
Fly ash
Fly ash
Fly ash
Fly ash
Fly ash
(ettringite) can cause the precipitation
of calcium carbonate. This is a common
aging reaction in concrete and causes a
gradual increase in the amount of calcite
and decrease in the amount of ettringite
(2). This type of reaction may account
for some of the metal release and repre-
cipitation observed in waste-Fortland
cement mixtures. Additional work on Port-
land cement and pozzolan encapsulates is
needed to further understand the interac-
tion of heavy metal hydroxides and gelling
cement hydrates. The results reviewed
here do not suggest Portland cement is
beneficial to waste containment when water
leaching is expected.
SUMMARY
Electroplating wastes are a complex
mixture of material produced by metal
cleaning, etching, and plating. The major
persistent contaminants associated with
these wastes are potentially toxic metal
hydroxides. The amorphous metal hydrox-
ides of cadmium, chromium, copper, lead,
nickel, and zinc in these wastes can
become significant pollutants if improper
disposal techniques are employed.
Several solidification/stabilization
techniques have been developed or adapted
to assure control of these wastes. New
insoluble, non-reactive compounds have
been developed for disposal by shallow
burial. Systems that are proposed or in
use include:
a) Development of silicate mineral
analogs that are chemically stable under
disposal conditions. The crystalline
product must generally be specific for the
element to be disposed.
b) Development of low-temperature
silicate polymers that can include a vari-
ety of potential metal contaminants in a
258
-------
gel-like system. Non-crystalline poly-
meric materials are, however, generally
more leachable than crystalline compounds.
Encapsulates of metal hydroxides have been
prepared from a variety of materials.
a) Macroencapsulates produced from
polymer-bonded and polyethylene jacketed
wastes have shown excellent containment
characteristics in testing programs.
b) Microencapsulates produced from
organic polymers have been unsatisfactory
for electroplating waste when acidifica-
tion has been used to initiate polymer
formation. Better systems that can main-
tain the wastes as hydroxides have been
developed.
c) Thermoplastic microencapsulates
have shown excellent containment charac-
teristics for electroplating wastes.
d) Data developed from testing
Portland-pozzolan and pozzolan encapsul-
ates indicate that metals in the waste are
aot involved in the crystal chemistry of
the cement or pozzolan setting and harden-
ing process. Analyses of water leachates
of crushed wastes indicate the treated
waste materials may release more metal
than the untreated waste.
ACKNOWLEDGEMENT
This study was funded by the Environ-
mental Protection Agency, Municipal Envi-
ronmental Research Laboratory, Solid and
Hazardous Waste Research Division, Cincin-
nati, Ohio under Interagency Agreement,
EPA-IAG-D4-0569. Robert E. Landreth is
the EPA Program Manager for this research
area.
REFERENCES
1. Bartos, M. J., Jr., and M. R. Palermo.
1977. Physical and Engineering Prop-
erties of Hazardous Industrial Wastes
and Sludges. EPA-600/2-77-139.
U.S. Environmental Protection Agency,
Cincinnati, OH. 89 pp.
2. Bogue, R. H. 1955. The Chemistry of
Portland Cement. Reinhold Publ.
Corp., New York, NY. 793 pp.
10.
Christensen, D. C., and W. Wakamiya.
1980. A solid future for solidifica-
tion/fixation processes, pp. 75-90.
In: Pojasek, R. B. (ed.) Toxic and
Hazardous Waste Disposal, Vol. 4.
Ann Arbor Science Publ., Ann Arbor,
MI. 313 pp.
Deutsch, Morris. 1972. Incidents
of chromium contamination of ground
water in Michigan, pp. 149-159.
In: Pettyjohn, W. A. (ed.) Water
Quality in a Stressed Environment.
Burgess Publ. Co., Minneapolis, MN.
309 pp.
Fisher, L. W., and F. L. Simons.
1926. Applications of colloid chem-
istry to mineralogy, Part II, Stud-
ies of crystal growth in silica
gel. American Mineralogist, 11:200-
206.
Hattori, M., E. Shoto, and K. Nagaya.
1980. Solidification of heavy metal-
containing sludges by heating with
silicates, pp. 141-154. In:
Pojasek, R. R. (ed.) Toxic and
Hazardous Waste Disposal, Vol. 3.
Ann Arbor Science Publ., Ann Arbor,
MI. 205 pp.
Hess, T. L. 1980. Ultimate disposi-
tion for small volumes of noncombus-
tible hazardous wastes generated by
U.S. Army installations, pp. 155-
164. In: Pojasek, R. B. (ed.)
Toxic and Hazardous Waste Disposal,
Vol. 3. Ann Arbor Publ. Inc.,
Ann Arbor, MI. 205 pp.
Houle, M. J., and others. 1977.
Effect of municipal landfill leachate
on the release of toxic metals from
industrial waste, pp. 139-148. In:
Banerji, S. K. (ed.) Management of
Gas and Leachate in Landfills. EPA-
600/9-77-026. U.S. Environmental
Protection Agency, Cincinnati, OH.
297 pp.
Her, R. K. 1979. The Chemistry of
Silica. John Wiley and Sons, New
York, NY. 866 pp.
Inoue, Y., and others. 1981. Heavy
metal stabilization by synthetic min-
erals. Kankyo Kagaku Tokubetsu
Kenkyu Hikokushu B90-R33-5 1.
259
-------
11. Jones, L. W., and P. G. Malone.
1982. Disposal of treated and un-
treated electroplating waste in a
simulated municipal landfill.
pp. 294-314. In: Shultz, D. W.
(ed.) Land Disposal of Hazardous
Waste. EPA-600/9-82-002, U.S. Envi-
ronmental Protection Agency, Cincin-
nati, OH. 549 pp.
12. Lea, F. M. 1956. The Chemistry of
Cement and Concrete. St. Martin's
Press Inc. New York, NY. 637 pp.
13. Lubowitz, H. R. , and others. 1977.
Development of a Polymeric Cementing
and Encapsulating Process for Manag-
ing Hazardous Wastes. EPA-600/2-77-
045, U.S. Environmental Protection
Agency, Cincinnati, OH.
14. Mahoney, J. D., and others. 1981.
Radiochemical studies of the leaching
of metal ions from sludge bearing
concrete, pp. 65-69. In: Third
Conference on Advanced Pollution Con-
trol for the Metal Finishing Indus-
try. EPA-600/2-81-028, U.S. Environ-
mental Protection Agency, Cincinnati,
OH. 144 pp.
15. Malone, P. G., R. B. Mercer, and
D. W. Thompson. 1978. The effec-
tiveness of fixation techniques in
preventing the loss of contaminants
from electroplating wastes, pp. 130-
142. In: First Annual Conference on
Advanced Pollution Control for the
Metal Finishing Industry, EPA-600/
8-78-010, U.S. Environmental Protec-
tion Agency, Cincinnati, OH. 150 pp.
16. McCauley, J. W., and R. Roy. 1974.
Controlled nucleation and crystal
growth of various CaCO_ phases by the
silica gel technique. Amer.
Mineralogist, 59(9+10)-.947-963.
17. PQ Corporation. 1979. Sodium Sili-
cates, Liquids/Solids Bull. 17-103.
The PQ Corp., Valley Forge, PA.
15 pp.
18. Ramachandran, V. S. 1976. Calcium
Chloride in Concrete. Applied Sci-
ence Publ. Ltd., London. 216 pp.
19. Roberts, B. K., and C. L. Smith.
1979. Microencapsulation: simpli-
fied hazardous waste disposal.
pp. 103-118. In: Pojasek, R. B.
(ed.) Toxic and Hazardous Waste Dis-
posal, Vol. 1. Ann Arbor Science
Publ., Ann Arbor, MI. 407 pp.
20. Rosencrance, A. B., and R. K.
Kulkarni. 1979. Fixation of Toby-
hanna Army Depot Electroplating Waste
Samples by Asphalt Encapsulation Pro-
cess. Technical Rept. 7902, U.S.
Army Medical Research and Development
Command. Ft. Detrick, MD. 23 pp.
21. Rousseaux, J. M., and A. B. Craig,
Jr. 1981. Stabilization of heavy
metal wastes by the Soliroc Process.
pp. 70-75. In: Third Conference on
Advanced Pollution Control for the
Metal Finishing Industry. EPA-600/
2-81-028, U.S. Environmental Protec-
tion Agency, Cincinnati, OH. 144 pp.
22. Salas, R. K. 1979. Disposal of
liquid wastes by chemical fixation/
solidification—The Chemfix process.
pp. 321-348. In: Pojasek, R. B.
(ed.) Toxic and Hazardous Waste Dis-
posal, Vol. 1. Ann Arbor Science
Publ., Ann Arbor, MI. 407 pp.
23. Shoto, E., and M. Hattori. 1980.
Studies on the stabilization/solidi-
fication of chromium-containing
sludge, pp. 125-140. In: Pojasek,
R. B. (ed.) Toxic and Hazardous
Waste Disposal. Vol 3. Ann Arbor
Science Publ., Ann Arbor, MI.
205 pp.
24. Shoto, E., and M. Hattori. 1981.
Study of solidification of chromium-
containing sludge by heating with
silicates. 1. Relation between
chromium-extraction and firing tem-
perature. Kankyo Gijutsu, 10(22):
292-296.
25. Streng, D. R. 1977. The effects of
industrial landfill leachates and
gas. pp. 41-54. In: Banerji, S. K.
Management of Gas and Leachate in
Landfills. EPA-600/9-77-026, U.S.
Environmental Protection Agency,
Cincinnati, OH. 297 pp.
26. Subramanian, R. V., and R. Mahalingam.
1979. Immobilization of Hazardous
Residuals by Encapsulation. Washing-
ton State Univ., Pullman, WA, Report
NSF/RA-790046. NTIS.PB 296 642.
260
-------
27. Subramanian, R. V., and R.
Mahalingam. 1979. Immobiliza-
tion of hazardous residuals by
polyester encapsulation, pp. 247-
296. In: Pojasek, R. B. Toxic
and Hazardous Waste Disposal,
Vol. 1, Ann Arbor Science Publ.,
Ann Arbor, MI.
28. Thompson, D. W. 1979. Elutriate
Test Evaluation of Chemically
Stabilized Waste Materials. EPA-
600/2-79-154, U.S. Environmental
Protection Agency, Cincinnati, OH.
62 pp.
29. U.S. Environmental Protection Agency.
1980. EP Toxicity Test Procedure,
Appendix II. Federal Register
45(98):33127-33129.
30. U.S. Environmental Protection Agency.
1982. Proposed NSPS, effluent lim-
its, pretreatment rules for electro-
plating, metal finishing point source
categories. Federal Register,
47:38462-38478.
31. Vail, S. G. 1952. Soluble Sili-
cates, Vol II, Reinhold Publ. Co.,
New York, NY. 669 pp.
261
-------
ANALYZING THE COSTS OF REMEDIAL ACTIONS
Edward J. Yang and James D. Werner
Environmental Law Institute
Washington, D.C.
Abstract
The Comprehensive Environmental Response, Compensation, and Liability Act of 1980 (Superfund)
explicitly requires consideration of cost-effectiveness in the selection of government response
actions. Preliminary results from 23 case studies on expenditures of past site responses to
uncontrolled hazardous waste sites are presented. These case studies represent a wide range of
remedial technologies, response frameworks and response costs. The sites were selected to document
information relevant to Superfund responses. The results here include total response costs, an
example of a cost section of a case study, and a comparison of costs for excavation, transportation
and disposal for 5 sites. Costs for similar technologies appear to vary greatly among the sites due to
the particular characteristics of each site. However, understanding the factors that influence costs
provides a useful tool for planning cost-effective remedial responses.
INTRODUCTION
In response to growing concern over
public health and environmental threats from
uncontrolled hazardous waste sites, Congress
passed the Comprehensive Environmental
Response, Compensation, and Liability Act of
1980, or the so called "Superfund." A major
component of implementing Superfund is to
conduct remedial actions to prevent or mitigate
the long term migration of hazardous
substances. Remedial actions are also distinct
from other parts of the Superfund, such as
removal, in that cost is a concern in formu-
lating the remedial response. In fact, the Act
calls for the response to be "cost-effective"
although, as the National Contingency Plan
points out, concern for cost effectiveness does
not take precedence over public health, welfare
and the environment. It merely helps to select
the least cost alternative from among
adequately effective options. For example,
cost estimation and cost-effectiveness analysis
are key ingredients to the formulation of the
feasibility study which underlies the selection,
design and implementation of a remedial
action. Also, Superfund response action costs
must be considered in the fund balancing
section of the law. Given the number of
Superfund sites it will be expensive to conduct
detailed field investigations on all sites to
match the reduction in risk and the cost of each
site response. A cost estimation model using
data on past remedial actions can approximate
the required expenditures by extrapolating an
estimate from a preliminary description of the
site and its problems. The accuracy of the
estimate will increase as more information is
gathered on each site.
In addition there are two primary needs
for cost analysis in managing remedial actions.
First, the size of the remedial expenditure will
require careful accounting simply from the
point of good management practice. As the
nation becomes more and more budget
conscious, costs of government programs will be
subjected to more scrutiny. Superfund, being a
high exposure program, will undoubtedly be
examined for its financial management.
Second, because the state-of-the-art of
remedial technology is still evolving quickly,
there is also the need to compare expenditures
across regions and sites continuously.
Contractors will invariably charge costs which
contain significant differences reflecting the
differentials in the effectiveness of technology
and marketing practice. This unstable
environment will make the selection of cost-
effective remedial actions difficult. A
continuously updated documentation of
expenditures will significantly increase the
262
-------
management capability of the Superfund
program.
This paper presents some of the interim
research results of a project carried out by the
Environmental Law Institute, in conjunction
with JRB Associates, on one aspect of cost
analysis, namely the collection and analysis of
actual expenditures on past remedial actions.
The methodology section will describe site
selection, data collection and categorization.
Given the fact that the research is ongoing and
not all the empirical results are in, this
presentation will present a summary of total
expenditures by sites, an example of site-
specific expenditure analysis and a comparison
of off-site disposal costs across the sites.
A few general limitations should be
noted at the outset. First, the expenditures
were documented on different levels of
aggregation, depending on the type of contract
and accounting procedures. Results are
presented on a comparable level as much as
possible, but some accuracy may have been lost
in that process. Second, the sites may not be
representative of the universe of existing
uncontrolled hazardous waste sites, not only of
the type of site problems, but also of the cost
of the technologies used. Third, hidden costs
borne in-house by corporations and government
agencies are often not fully accounted for.
Finally, market dynamics, such as increasing
competition and regulatory requirements have
affected response costs over time, which
reduces comparability between sites.
METHODOLOGY
Site Selection
The following information sources were
used to find potential uncontrolled hazardous
waste sites for case study consideration:
o 1981 SCS survey
o EPA Hazardous Material Incidence
Reports
o EPA Regional Offices
o State lists of hazardous substance spills
and release
o Remedial Action Contractors
o Trade Associations
o Professional Contacts
From these sources, 23 sites were
selected using the following criteria:
o Data Availability
o Useful remedial technology
o Post 1975
o Remedial actions completed
Telephone calls and/or letters were used
to ascertain the status of the sites and the
availability of data. In practice, the
availability of data was the most common
limiting factor. For this reason, the sites may
not be representative of the universe of existing
uncontrolled hazardous waste sites and their
remedial costs.
Roughly equal numbers of private and
government financed remedial actions were
chosen, at 12 and 11 sites, respectively. The
variety of remedial technologies includes
excavation and removal, slurry walls, sub-
surface drains, solidification and ground water
extraction, treatment (carbon, biological and
chemical) and reinjection. As of January 1983
the actual draft case study reports have been
completed, but the comprehensive analysis
section is still being prepared.
Collection of Data
Cost data were generally collected in
three stages:
1. Interview preparation
2. Interview and file search
3. Follow-up and refinement.
Before the site visit, readily available
cost data were obtained through the mail in
preparation for the interviews. Researchers
also used available information about work
performed at the site to prepare questions and
organize the work into useful unit operations
and component tasks.
Interviews and file searches during the
site visit were the primary sources of cost
data. All invoices, memoranda, letters,
proposals and contracts were photocopied. This
information was usually supplemented by
interviews with participating personnel who
often recalled numbers that confirmed invoices,
263
-------
helped aggregate related activities, or related
dates with activities.
The last phase of data collection was the
site visit follow-up. Phone calls and
correspondence were used to refine and verify
the data. Again, site contacts were helpful in
organizing related tasks, and specifying what
was or was not included in a contractors'
general invoices.
Categorization of Data
The data base resulting from this search
consists primarily of three forms:
1. Invoices
2. Reports and correspondence
3. Interviews.
Cost data were constructed into
functional categories by two primary means and
supplemented with a variety of sources. The
categorization method depended on the type of
data available from the source. The first
method was the aggregation and summing of
specific costs on invoices to determine the cost
for a general operation such as slurry wall
construction. All costs related to a particular
task were totalled to derive the cost of the
operation. Component costs of the total are
described and dissected in the text. Unit costs
for items like disposal or slurry mixture were
multiplied with the volumes given in as-built
engineering reports to double check the
numbers.
The second method was the correlation
of weekly invoice summaries of general cost
categories with the activity occurring during
that period to estimate the cost of an
operation. Weekly invoice summaries were
often categorized by items such as labor,
equipment, transportation and disposal. These
categories were broken down according to the
operation to which its function contributed and
the proportion of the weekly total used for that
operation. For example, excavation might have
been performed for 100% of a given week, and
50% of the labor and 30% of the equipment
were devoted to this operation, with the
remainder devoted to analytical work. The cost
would be further broken down if the daily
reports showed that transportation and disposal
occurred during two of the seven days included
on the invoice for which X % labor and
equipment were used for loading and analysis.
These invoice-based methods were
supplemented with other sources such as
interviews, reports, correspondence and
contracts. When aggregating specific invoice
items, particular items were sometimes
explained by references to file material or the
site contacts. The breakdowns of general
invoice summaries according to daily reports
were often refined by interviewing or corres-
ponding by mail with site contacts about the
execution and timetable for particular
operations. In-house costs were estimated,
when available, by totalling company time
sheets for the appropriate project. In-house
costs were the most difficult cost to obtain,
usually because of the lack of available records.
RES ULTS
Of the 23 case study cost sections, 11
were based on invoices or invoice summaries, 6
were based on interviews with corporate
officials and engineers, and three were based on
reports and correspondence from contractors.
One report (Richmond Sanitary Services) was
based on estimates using daily progress reports
and the 1^83 Means Construction Cost Data
manual because the ditch construction was
performed using non-invoiced in-house
resources. The sources for individual cost
sections are discussed in the "Project Cost"
section each report.
The following three preliminary cost
result sections are presented below:
o Total project costs
o Specific site involving both capital and
operation and maintenance costs for
groundwater treatment
o Comparison of excavation, transpor-
tation and disposal costs.
Again, these results are preliminary and are
intended here to show the range of observations
in the data base rather than provide a
comprehensive summary. This limitation is
probably especially relevant to the total project
cost section.
In specific sites,
categories were observed:
two general cost
o Containment/collection/treatment
o Excavation, transportation & disposal
264
-------
The containment/collection/treatment costs
consisted of several subtasks such as capping,
cut-off walls and treatment system
construction, as well as operation and
maintenance costs for the systems. The "Cost
and Funding" section of Anonymous Site X is
presented below (Table 1) to show these types
of cost elements, which are described in more
detail in a separate section of the case study .
The costs for excavation, transportation and
disposal were often found to be the largest
single cost category for the sites involving that
response. A chart is used to compare to
compare results of this important cost
category.
Total Cost by Site
The total cost for each of the 23 sites is
shown in Table 2 and the distribution of totals is
shown in Figure 1. The average cost per site
was about $1.4 million ($1,394,463.10), but the
standard deviation of $2.2 million ($2,205,344)
shows a wide variation in the costs. The total
site costs ranged from $23,000 to $10.3
million. Sixteen of the 23 sites were between
$500,000 and $1.5 million.
The variation on the total costs shows
not only the variation in the actual site costs,
but reflects artifacts in the data base as well.
The low end of the total cost range includes a
site that involved significant in-house costs that
were not accounted for in record keeping, and
could not be included in the total. The site at
the high end of the range includes significant
disposal costs that may be considered part of
the work necessary for ongoing facility
operation. The inclusion or exclusion of aquifer
restoration work significantly affected total
costs. Many sites involved only removal of the
source of contamination but not the contamin-
ated ground water; and of those sites where
ground water restoration projects occurred, a
wide variety of levels of effort and success
were encountered.
COST AND FUNDING OF ANONYMOUS SITE X
Source of Funding
The owner of Anonymous Site X (Anon X)
paid for all project costs. The state did not
seek any compensation from the owner for the
monitoring costs.
TABLE 1. SUMMARY OF COST INFORMATION ANONYMOUS SITE X
Task
A. Surface Sealing and
Drainage (a)
1. Asphalt paving
2. Betonite cap
with grave)
3. Curbs and
gutters
B. Oil Recovery
System (c)
1 . French Drain (d)
2. Operation and
Maintenance of
Treatment System
TOTAL Project Cost
quantity
-.
135,000 sq.ft.
(12542 m2)
156,000 sg.ft
(14,493 m^)
4,125 feet
(1,255 m)
length:
210 ft. (63m)
depth:
22.5 ft. (7m)
1,000-1,500 gallons
(3,785-5,678 1/montn
—
Actual Expenditure
Subtotal: S734.100
($155,000 (b)
($177, COO (b)
($132,300)
Subtotal: 5346,200
($337,000)
550,000/year
$1,530,300
Unit Cost
51.15/sq.ft
(12.36/m2)
S1.13/sq.ft.
(12.21/m2)
S32/foot
(S105/m)
$l,605/foot
(S5,264/m)
$2.70-4.16/gallon
(SO. 73-1. 10/1)
--
Period of
1981
1981
1981
1981
1981
1981
1982
1981
given in Table 2.
(b) Cost includes half of $42,COO cost
for grading.
(c) Subtotal also includes other costs given
TableJ.
(d) Excluding sump cost.
265
-------
TABLE 2. TOTAL COSTS OF SITE RESPONSES
Site •Urs
1 . Anonymous A
2. Anonvrrous o
3. Setter 3nte
4. Biocra'-t
5. Chemical Petals
Industries
5. Cne"ucai Recovery
Syste-s
7. Collece 3oint
3. Fairc".ilc ^eouolic
9. Sal IJD
10. General -lsc"nc
1 1 . Goose ran
12. i i ;•' Drun
!3. .-oustcn Cremical
14. .-owe, I-c.
15. riarty ; G."C
15. 'i.W. :;autre
17. 3ccicental Cnemical
13. Qjanta Pesources
19. Sicr.Tiona Sanitary
20. Strc'jcs;urg
2! . Traranei 1 Craw
22. 'Jniv. of !da.no
23. /ertac Cr.emicai
Prinarv Resoonse "ecnnoloav
'Jaste '/later disposal , cut-off \/all
SuDS'jrTace drain
Subsurface Cram
Ground .vater collection, aiodearaaation
Removal, disoosal. caooina
Cut-off wall, subsurface drain
Removal, cnssosal. sol iaif lean "n
Removal, aisscsai, caooinq
^eiio^ai. disposal, : n situ n.-^e t-eatrent,
caooi^a
Suosurface, cram, casoin, r'jn-orf contra!
Se"iovai jisDosai. 3rcund water collection,
carcon treatrrent
Removal, aisoosai, aeration, cacoing
Removal, disposal, oioceqraaation
Remove!, disposal, ground water recovery
Removal, disposal, aeration, capping
Suosurface Cram
Ground .later collection, carcon treatment
Removal, disposal, capping
Removal, disposal, -./aste water treatment
Cut-off '.;all
Slurry nail, recovery well
Solidification, capping
Removal, 'soosal
Removi, disposal, Suosurrace arain,
caooirg, cjt-orf wa 1 1
"3tal Site Cost
S 10.3 mil ion
S263.2!?
5 23.C20
5926,133
5341,349
51.423 mil ion
51.75 -11 1 1 ion
5450. ;:3
5610,445
51 . 53 m 1 11 on
55.1 11 1 1 i on
SI .245 mi 1 1 icn
S709.423
S470.C30
S419.250
5 79,110
S2.65 million
S2.255 m 1 1 icn
$111,036
$594,500
$427,527
5174,397
$143, C30
* See cc^oiete description of site responses
in E.LI/JR3 Case Studies, 1983.
266
-------
FIGURE 1. DISTRIBUTION OF TOTAL SITE COSTS
15
14-
13-
12
11
10.
34 56 7 d
TOTAL COST OF SITE RESPONSE (ID MILLIONS S)
10
11
Selection of Contract
No information is available on the
contractor selection process.
Project Cost
The total construction costs for surface
sealing and drainage, and the oil recovery
system, was about $1.6 million (see Table 2).
This cost does not include preliminary study and
design work. No cost information available for
this work, which included monitoring well
installation, sampling and analysis, and repair,
as well as design and planning. The cost
information given below is based on interviews
with company engineers and on invoices.
SURFACE SEALING AND DRAINAGE
The sum cost of $734,000 for surface
sealing and drainage includes the variety of
expenditures shown in Table 3. The $43,000
cost of grading is split between asphalt and clay
capping costs for the purpose of calculating unit
costs. At a cost of $155,000, the unit cost for
the 135,000 square feet (12,542 m2) of asphalt
pavement was $1.15/square foot ($12.26/m2).
At a cost of $177,000, the unit cost for the
156,000 square feet (14,493 m2) of bentonite
capping with a gravel cover was $1.13/square
foot ($12.21/m2). The 4,150 feet (1,256 m) of
curbs and gutters was determined by dividing
the total cost of $132,800 by the unit cost of
given of $32/foot ($105/m).
267
-------
TABLE 3. EXPENDITURES FOR SURFACE SEALING/DRAINAGE
Equipment mobilization
Equipment demobilization
(including cleaning)
Paving
Perimeter fencing
Drainage system pipe
Curb and gutter
Manholes
Monitoring equipment at end of
of manhole for surface sealing
Supervision (on-site contractor and
consultant oversight
Soil removal
Clay sealing; top rock
Subtotal
$ 26,000
$ 3,600
$134,000
$ 6,000
$150,000
$132,800
$ 4,700
$ 12,000
$ 20,000
$ 47,000
$156,000
$734,100
($10/foot, $33/m)*
($18/foot, $59/m)*
($32/foot, $105/m)*
($280 each)
($23/yd3, $30/m3)
* Unit costs are as-built
OIL RECOVERY AND TREATMENT SYSTEM
The total of $846,200 for constructing
the oil recovery and treatment system includes
items shown in Table 4. At a total cost of
$337,000 for the French Drain, the unit cost
(based on arm lengths of 60,70 and 80 feet
(18,21 and 24 m) of constructing a total of 210
feet (63 m) of 20-25 feet (6-8 m) deep trenches
was about $1,605/ linear foot ($5,264/m). This
cost excludes expenditures for the sump and
other related costs.
The total Operation and Maintenance (O.
<5c M.) cost for treating about 1,000-1,500
gallons (3,785 - 5,678 I)/ month excludes
equipment amortization and about two hours/
week of the plant manager's time to perform
batch treatments. At a total O. & M. cost of
about $50,000/year, the unit cost for this initial
rate of treatment is $2.70 - 4.16/gallon ($0.73 -
1.10/1).
COMPARISON OF UNIT COSTS
A sample of the unit costs for
excavation, transportation and disposal is shown
in Table 5. The site characteristics are listed
for each site to describe some of the factors
that may influence costs. The unique elements
of each site are described in the text of each
case study.
Since disposal costs were found to
depend on the nature and extent of
contamination, the waste at each site presented
in Table 2 is generally for soil contaminated
with non-PCB chlorinated organics. Disposal
costs of PCB contaminated soil were typically
much higher and are described in the EL1/JRB
case studies report. One of the case studies,
involved 12 separate disposal costs for the 13
different waste streams identified on-site.
These waste streams were segregated according
to physical and chemical characteristics, which
affected handling characteristics and regulatory
requirements for disposal.
268
-------
TABLE 4. OIL RECOVERY AND TREATMENT SYSTEM ITEMS
Equipment mobilization and
and demobilization
Sump, including sheet piling
Water treatment system, oil/water
separator, pumps and plumbing
Tank farm building, plumbing modifi-
cations on existing tank farm to receive
material before treatment to test for
treatment need
Sanitary sewer system modifications,
and monitoring equipment to discharge
treated effluent to EBMUD, plus project
management
Electrical and instrumental oil
recovery system
French Drain System
Subtotal
$ 65,000
$ 85,000
$129,200
58,000
$ 65,000
$117,000
$337,000
$846,200
Excavation, Transportation and Disposal
The average cost for excavation, trans-
portation and disposal for the four sites was
$153.82 - 179.02 (standard deviation $72.17 -
87.77). All excavation and transportation costs
are at least an order of magnitude higher than
the SCS estimates. These differences may
reflect the secondary costs associated with
these operations such as safety procedures,
sampling and inspection. For example, a $5/
hour surcharge is typically added to labor costs
for using protective gear, and split spoon
sampling of each truck load is typically
performed. Further discussion of the factors
affecting excavation, transportation and
disposal costs as well as capital and operation
and maintenance costs will be given in the
analysis section of the final case study report.
costs across types of sites as well as across the
technologies used to respond to the sites. Two
general observations can be made on the
complexities of analyzing the costs of site
responses at uncontrolled hazardous waste
sites. The first, and perhaps most important, is
that a unique combination of a finite set of
factors influences the costs of site responses to
varying degrees. When these factors are clearly
identified and described, the variance in costs
for similar actions can be explained. For the
same reason, cost figures, isolated from these
factors, are not useful; future cost estimation
will have to incorporate these factors in some
way. Second, future documentation of remedial
action costs should look for consistency and a
format that will be useful to decision making in
remedial actions. Such improvements can
greatly enhance the effectiveness of future
remedial actions.
CONCLUSION
The preliminary results on the costs of
site responses to uncontrolled hazardous waste
sites presented above provide a useful range of
actual costs as well as a pratical guide to the
variety of realistic cost considerations. The
data base's usefulness is both limited and
enhanced by the fact that it shows a range of
269
-------
TABLE 5. COMPARISON OF UNIT COSTS FOR
EXCAVATION, TRANSPORTATION AND DISPOSAL
Activity
Excavation
'ransportation
Jisposal
TOTAL
Quantity of
Material
Type of
Material
lontimnants
Level of con-
tamination
(on-site soil )
Health Risk
Environmental
Risk
Excavation
Depth
Gallup
S14.02/ra3
573.96/Mt
(800 km)
S44.09/Mt
S132.07/Mt
3, 647 /lit
drums/soil
solvents,
chlorides, heavy
metals
cadmium:
22 ug/g
magnesium:
61 ug/g
Moderate to low
(presently not
a drinking
water supply)
Moderate threat
to recreational
fishing stream
1 - 4 m
S I T
Houston, Mo.
S40-S90/m3
$26/Mt
(273 km)
S48/Mt
5114-164/Mt
2,015 m3
Wet soil
PCP
negligible
surface
water
10 - 15 cm
E 5
Goose Farm, U.J.
S 44-90/m3
$63/Mt
(70S km)
$44/Mt
$151-207/Mt
3,900 m3
Drums mixed
soil, wastes
Various
Ground and
surface
water
5 - 8 m
Howe, Inc., Minn.
S10.00/m3
S37.37/Mt
(225 km)
S10-25/Mt
S57. 37-72. 37/Mt
1,988 m3
ice, snow,
lime, sandy soil
Pesticides
Ground and
surface
water
30 - 60 cm
Martv's
S148.50/Mt
(825 km)
$92/Mt
5301.50/Mt
804 m3 soil 151
drums excavated
470 Mt soil;151
drums disposed
Soil, drums
Mixed solvents,
paint sludges
Still Bottoms.
PCB oil
1 ug/g-l,000ug/g
ground water
fire, explosion
1 - 5 m
REFERENCES
Environmental Law Institute/JRB Associates.
"Case Studies and Cost Analysis of Responses to
Uncontrolled Hazardous Waste Sites." 1983.
Means, Robert Snow, Inc. "Building
Construction Cost Data" 1983. Means
Publishing, Kingston, MA.
SCS Engineers, Inc. "Remedial Actions at
Uncontrolled Hazardous Waste Sites: Survey
and Case Studies." 1981.
SCS Engineers, Inc. "Costs of Remedial
Response Actions at Uncontrolled Hazardous
Waste Sites" 1982.
270
-------
COSTS OF REMEDIAL ACTIONS AT UNCONTROLLED HAZARDOUS WASTE SITES
James J. Walsh, John M. Lippitt,
Michael P. Scott, Anthony J. DiPuccio
SCS Engineers
Covington, Kentucky 41017
ABSTRACT
Subsequent to the passage of the Superfund Act, the U.S. Environmental Protection
Agency and other regulatory agencies were faced with the dilemma of selecting cost-
effective remedial actions for uncontrolled hazardous waste sites. This paper summarizes
two documents being developed to aid regulatory officials in their decision-making process
on Superfund site cleanup. The first document outlines the steps to be taken in costing
alternative remedial action unit operations, and in combining these unit operations to yield
costs for an entire remedial action scenario. Unit operation costs for each of 21 dif-
ferent remedial actions are presented in this paper. Examples are provided of the costing
methodology for applying remedial actions at each of landfills and surface impoundments.
This first document, however, relied primarily on information available from standard
construction references and past applications not dealing specifically with hazardous
waste. This was done since the literature was found lacking back in 1980, and since costs
from actual Superfund operations were not yet available. The purpose of this second
document is to collect remedial action cost data that has recently become available from
site cleanup contractors and regulatory authorities. The emphasis on this second pro-
ject is to be placed on the incremental costs associated with worker health and safety.
INTRODUCTION
In December 1980, The U.S. Congress
passed legislation entitled "The Compre-
hensive Environmental Response, Compensa-
tion, and Liability Act", which is also
known as Superfund. Superfund provides
the U.S. Environmental Protection Agency
(EPA) with the legislative mandate and
monetary base to assist in the elimination
of public health hazards posed by uncon-
trolled hazardous waste sites. Section
105 of the Superfund legislation requires
the EPA to investigate the costs of
remedial/cleanup actions at uncontrolled
waste sites. Specifically, Item 2 of Sec-
tion 105 requires the development of cost
ranges for various types of remedial ac-
tions.
Responsibility for implementing Super-
fund actions and responding to uncontrolled
hazardous waste sites rests primarily with
the EPA Office of Emergency and Remedial
Response (OERR). At the request of OERR,
the EPA Office of Research and Development
(ORD) has performed work on remedial ac-
tion technologies and costs. SCS Engineers
was contracted by ORD to review, update,
compile, and integrate existing data on
remedial action costs at hazardous waste
sites.
This paper provides a review of two
SCS projects undertaken to determine
remedial action costs. The basic approach
taken in developing these costs was to
identify the individual remedial action
unit operations which can be combined to
271
-------
form an entire remedial action plan, or
scenario. Individual unit operations can,
in turn, be broken into several cost com-
ponents which lend themselves to easy
costing. By determining the costs asso-
ciated with these component cost items,
the total unit operation cost can be
developed. Subsequently, an entire reme-
dial action plan for a site can be costed
by summing the costs of the remedial
action unit operations.
The first project, "Costs of Remedial
Actions at Uncontrolled Hazardous Waste
Sites", compiled and classified existing
cost data in terms of scope, location,
and time frame. A methodology was then
provided for computing the costs of reme-
dial action unit operations. The final
report was generated in early 1981. The
document was intended as a compilation
of previous, more general reports generated
in prior years. Because of the lack of
actual Superfund efforts that had been
performed to that time, this report relied
primarily on standard construction cost
references. In some cases, the costs
incurred at similar, commercial applica-
tions of such remedial actions (i.e., at
other than actual, uncontrolled hazardous
waste sites) were utilized.
The costs included in the document
above can be described as "base construc-
tion costs". Admittedly, they may not
sufficiently address the added cost en-
countered on some unit operations due to
the handling of hazardous materials. How-
ever, remedial actions are now being imple-
mented at initial Superfund sites. Addi-
tionally, numerous contractors desiring
to acquire Superfund-related work have
purchased the necessary equipment, and
performed research on the technologies
to be applied and costs to be encountered.
Thus, it is thought that more accurate
remedial action cost information can now
be obtained from these sources. This new
cost data should adequately address concerns
for worker health and safety, and their
impacts on total site costs.
REMEDIAL ACTIONS
A total of 21 remedial action unit
operations were identified and costed in
the course of the first project. They
were derived chiefly from earlier EPA
reports on the subject of remedial actions.
[1, 2, 3] As shown in Table 1, they have
been generally classified into 4 areas:
(1) surface water controls, (2) ground
water controls, (3) gas migration controls,
and (4) waste controls.
Table 1 identifies a total of 6 sur-
face water controls, with 5 applicable
to landfills and a slightly different
set of 5 applicable to impoundments. Sur-
face water controls are usually considered
short-term measures and can be effective
in controlling imminent pollution dangers
and discharges to surface waterways.
Often however, they need to be used in
conjunction with ground water controls
as described subsequently.
A total of 8 ground water controls
are identified in Table 1. These controls
are generally considered to be more long-
term in nature than are surface water
controls. But while curtailing the more
long-term environmental impacts such as
ground water pollution, they can also
be expected to require more sophisticated
technologies, longer lead times, and lar-
ger funding levels.
TABLE 1. REMEDIAL ACTION UNIT OPERATIONS
Remedial Action Unit Operations
Applicability
Landfills Impoundments
Surface Water Controls:
1. Regrading and Surface Water Diversion
2. Surface Sealing with Asphalt
3. Revegetation
4. Cutoff Trenches
6. Basins and Ponds
6. Containment Berms
Ground Water Controls:
7. Slurry Trench
8. Grout Curtain
9. Sheet Piling Cutoff Wall
10. Grout Bottom Sealing
11. Underdrains
12. Well Point System
13. Deep Well System
14. Well Injection System
Gas Migration Co^ntroljj
15, Gas Venting Trench
16. Gas Extraction Wells
Waste Controls:
17. Treatment of Contaminated Water
18. Chemical Fixation
19. Chemical Injection
20. Excavation and Reburial
21. Leachate Recirculation
272
-------
Ground water controls can generally
be classified into active and passive sys-
tems. Slurry trenches, grout curtains,
sheet piling walls, grout bottom sealing,
and underdrains are all considered passive
systems, and operation and maintenance (O&M)
requirements are usually low or non-exis-
tent. Active systems include well point
systems, deep well systems, and well in-
jection systems. In addition to high capi-
tal costs, active ground water controls
usually require on-going O&M efforts such
as electrical service, pump maintenance
and replacement, and possibly treatment
of contaminated ground water.
Both gas migration controls were found
to be applicable to each of landfills and
impoundments. Organic materials buried
in landfills or closed surface impoundments
often generate combustible methane gas.
In many cases, combustible gases have been
found to migrate laterally from such facili-
ties, posing a risk of explosion or fire
in nearby structures.
Gas migration controls can also be
either active or passive. Gas venting
trenches are passive systems and can be
installed by excavating trenches around
facility perimeters. Gases migrating la-
terally from the site are intercepted by
these trenches and vented to the atmosphere.
Venting trenches are usually limited to
where gases are confined within 20 ft of
the ground surface. Where greater migra-
ting depths are encountered, active gas
migration controls are required. These
systems consist of large diameter extraction
wells, header pipes, and central blower
facilities. Extracted gases are vented
or flared to the atmosphere.
A final category of remedial action
unit operations is identified as waste
controls. While all 5 were found applicable
to landfills, only 4 were deemed appro-
priate for impoundments. Treatment of
contaminated water was meant to describe
the removal of low level contamination
from polluted runoff (as opposed to heavily-
contaminated waste materials). Chemical
fixation and injection can be effective
in stabilizing waste masses and preventing
further emission of contaminants. "Exca-
vation and reburial" is usually an expen-
sive proposition; the health and safety
hazards affiliated with removal of the
hazardous (and often unknown) waste ma-
terial may exceed the hazards of leaving
the waste in-place. Lastly, leachate
recirculation can be a temporary mea-
sure to treat or dispose of contami-
nated runoff.
BASE CONSTRUCTION COSTS
Costing Methodology
As mentioned previously, the thrust
of this first project was to develop typi-
cal costs for remedial action unit opera-
tions. In addition, all costs developed
were examined for their sensitivity to
geographic location and facility type
and size. For geographic location, costs
for each remedial action unit operation
were computed at 3 separate cost levels:
"Upper U.S.", "Lower U.S.", and "Newark,
NJ". For facility type and size, remedial
action costs were examined at 2 types
of facilities (i.e., landfills and im-
poundments), and 5 scales of operation.
This paper will concentrate on the third
(or medium) scale of operation.
In determining remedial action unit
operation costs, each operation first
had to be broken down into specific cost
components. For example, cost components
for bentonite slurry walls were found
to include bentonite material, transport,
trench excavation, etc. In turn, cost
components could be further defined through
the use of cost subcomponents, with spe-
cific costs assigned to labor, materials,
equipment, and supplies.
On this project, assigned costs were
in terms of mid-1980 dollars for the U.S.
upper and lower city averages in the con-
tinental 48 states, as well as for the
example location of Newark, NJ. After
costs were assigned to each cost sub-
component and thence component, conceptual
design cost estimates were accumulated
for each remedial action unit operation.
Allowances for overhead and contingencies
were applied, and total and unit life
cycle costs computed.
In most cases, 1980 editions of Means
and Dodge construction cost guides [4,
5] were used to obtain the needed costs.
Regional adjustment indices presented
in the Dodge guide were used to modify
the cost estimates for geographical dif-
ferences. These indices were applied
273
-------
to obtain revised material and labor costs
for the "Upper U.S.", "Lower U.S.", and
"Newark, NJ", estimates. The indices were
not applied to equipment costs since it
was assumed that these costs are somewhat
constant throughout the nation.
Once all the components within a given
unit operation had been costed, the costs
were summed giving a subtotal capital cost
for each unit operation. As indicated
in Table 2, an overhead allowance of 25
percent was then applied to this subtotal,
followed by a contingency allowance varying
between 10 and 40 percent. The subtotal,
capital cost was then added to these over-
head and contingency allowances to obtain
the unit operations' total capital cost.
This method was then applied across all
5 scales of operation.
Efforts then proceeded in costing
O&M impacts on the various unit operations.
First, an assessment was made as to O&M
requirements over an estimated 10-year
life cycle. These 0$"1 requirements were
then priced in terms of 1980 dollars, in-
flated to an appropriate value in the year
they were to be incurred, and lastly, back-
discounted to determine the present value
in 1980. Inflation and discount factors
are summarized in Table 2. As noted, it
was assumed that capital costs were fully
incurred in the first year of operation;
amortization or financing charges have
not been included in this analysis. Also,
when work on this project was initiated
in 1980, an 11.4 percent discount rate
seemed reasonable, and was used. However,
it is now likely that Superfund cost esti-
mates will use the 10 percent discount rate
recommended recently by the Office of Man-
agement and Budget.
TABLE 2. COSTING METHODOLOGY SUMMARY
Capital Costs
Base Year
Overhead
Contingencies
Terms
O&M Costs:
Base Year
Term
Future Inflation
Discount Rate
Mid-1980
25%
10 - 40%
Fully incurred at time = 0;
not amortized
Mid-1980
Annual intervals for 10 years
7.1%*
11.4%
* Except electricity at 9.2
Site Profiles
As described previously, site pro-
files (or hypothetical disposal sites)
were developed for landfills and impound-
ments (i.e., wastewater pits, ponds, and
lagoons). Each of these was portrayed
at 5 different scales of operation. Only
through the use of these site profiles
could the individual unit operation cost
estimates be computed.
Resulting site profiles were con-
figured to conform to somewhat uniform
sets of design criteria and environmental
conditions. For both landfills and im-
poundments, the selected scale of operation
was developed in terms of total surface
area. This emphasis on surface area is
appropriate for uncontrolled waste sites.
Daily input volume would be inappropriate
since the facilities are no longer active;
total storage volume would be inappropriate
since the depths of these facilities are
often unknown.
Table 3 delineates the operational
and environmental conditions developed
for the 5 landfill and 5 impoundment pro-
files. The range of sizes for landfills
was developed based on data presented
in References [6] and [7]. In addition,
it was assumed that the surface area for
each landfill was square. Cut slopes
were designed at 2:1, with fill slopes
at 3:1.
Table 3 also demonstrates the various
scales of operation and environmental
conditions for impoundment profiles. The
range of scale sizes was developed from
data in References [8] and [9],
Unit Operation Costs
Selected unit operation costs have
been included in this paper as Tables
4, 5, and 6. As indicated, Table 4 pro-
vides costs for each of the 21 remedial
action unit operations assuming a geo-
graphic location of "Lower U.S.", and
the medium-scale of operation for each
of the example landfill and impoundment.
Table 5 assumes a geographic location
of "Upper U.S." at medium scales of opera-
tion for landfills and impoundments. Table
6 assumes a geographic location of Newark,
NJ, also at medium landfill and impound-
ment sizes.
274
-------
TABLE 3. SITE PROFILE DATA
Overall Dimensions
Scenario
Number
Landfill:
1
2
3
4
5
Impoundment •
1
2
3
4
5
Surface
Area
(ha)
1
3
5
13
19
0.01
0.03
0.32
1.55
9.54
Depth of
Waste
(m)
11
12
13
14
15
0.5
1.0
2.0
4.0
6.0
Storage
Volume
(n.3)
40,000
200,000
400,000
1,200,000
2,000,000
44
342
6,362
46,575
381 ,428
Depth
To Grouncf
Water
M
4
4
4
4
4
0.5
1.0
2.0
2.0
2.0
from Original Grade
To Waste "
Volume
(m)
5
5
5
5
5
4
4
4
4
4
To
Bedrock
(m)
15
15
15
15
15
15
15
15
15
15
TABLE 4. REMEDIAL ACTION COSTS - LOWER U.S.
Life Cycle Costs ($/umt)
Remedial Action Unit Operation
Surface Water Controls:
1. Regrading and Surface Water Diversion
2. Surface Sealing with Asphalt
3. Revegetation
4. Cutoff Trenches
5. Basins and Ponds
6. Containment Berms
Ground Water Controls:
7. Slurry Trench
8. Grout Curtain
9. Sheet Piling Cutoff Wall
10. Grout Bottom Sealing
11. Underdrains
12. Well Point System
13. Deep Well System
14. Well Injection System
Gas Migration Controls:
15. Gas Venting Trench
16. Gas Extraction Wells
Waste Controls-
17. Treatment of Contaminated Water
18. Chemical Fixation
19. Chemical Injection
20. Excavation and Reburial
21. Leachate Recirculation
Unit of Measure
Site area, ha
Site area, ha
Site area, ha
Trench length, m
Site area, ha
Berm volume, m^
Well face area, m2
Wall face area, m^
Wai 1 face area , m2
Site area, ha
Pipe length, m
Intercept face area, m^
Intercept face area, m^
Intercept face area, m2
Site perimeter, m
Site perimeter, m
Flow rate, 1/d
Site area, ha
Waste volume, m->
Waste volume, m3
Site area, ha
Medium-Size
Landfill
16,300
67,300
14,300
15
647
61
937
73
5,296,000
357
107
29
1,760
347
148
3
82,500
85
117
19,700
Medium-Size
Impoundment
35,200
48,402
3,890
15
__
4
60
343
95
1,003,000
1,550
321
114
109
347
148
4
82,500
85
389
-"
As indicated by these tables, unit
operation costs are presented in terms
of dollars per unit most appropriate for
the respective remedial actions. For ex-
ample, large-scale surface water control
techniques such as revegetation and re-
grading are presented in terms of dollars/
hectare. Passive ground water control
techniques (or barriers) have been pre-
sented in terms of dollars/square meter
of wall face in the vertical plane.
Geographic and Size Variations
To demonstrate how individual reme-
dial action unit operation costs have
been derived, cost calculations for each
of bentonite slurry trenches and well
point systems have been included in
Tables 7 and 8. Table 7 identifies all
the cost components which comprise ben-
tonite slurry trenches. Capital costs
include geotechnical investigation, slurry
275
-------
TABLE 5. REMEDIAL ACTION COSTS - UPPER U.S.
Life Cycle Costs ($/unit)
Remedial Action Unit Operation
Unit of Measure
Medium-Size
Landfill
Medium-Size
Impoundment
Surface Water Controls:
1
2
3
4
5
6
Regrading and Surface Water Diversion
Surface Sealing with Asphalt
Revegetation
Cutoff Trenches
Basins and Ponds
Containment Berms
Site area, ha
Site area, ha
Si te area , ha
Trench length, m
Site area, ha
Berm volume, m3
19,900
92,700
18,100
20
1,028
--
52,400
70,705
5,340
20
--
6
Ground Water Controls.
7
8
9
10
11.
12.
13
14.
Gas
15.
16
Slurry Trench
Grout Curtain
Sheet Piling Cutoff Wall
Grout Bottom Seal ing
Underdrains
hell Point System
Deep Well System
Well Injection System
Migration Controls:
Gas Venting Trench
Gas Extraction Wells
Wall face area, m?
Mall face area, m2
Wai 1 face area , m2
Si te area , ha
Pipe length, m
Intercept face area, nr
Intercept face area, m2 •
Intercept face area, m2
Site perimeter, m
Site perimeter, m
103
1,880
108
10,224,000
416
153
37
1,785
396
268
108
649
135
1,754,000
1,960
398
149
141
396
268
Waste Controls:
17.
18.
19.
20.
21.
Treatment of Contaminated Water
Chemical Fixation
Chemical Injection
Excavation and Reburial
Leachate Recirculation
Flow rate, 1/d
Site area, ha
Waste volume, m3
Waste volume, m3
Site area , ha
4
145,000
142
120
24,000
8
145,000
142
401
TABLE 6. REMEDIAL ACTION COSTS - NEWARK, NJ
Remedial Action Operation
Surface Water Controls:
1. Regrading and Surface Water Diversion
2. Surface Sealing with Asphalt
3. Revegetation
4. Cutoff Trenches
5. Basins and Ponds
5. Containment Berms
Ground Water Controls:
7. Slurry Trench
8. Grout Curtain
9. Sheet Piling Cutoff Wall
10. Grout Bottom Sealing
1 1 . Underdra ins
12. Well Point System
13. Deep Well System
14. Well Injection System
Gas Migration Controls:
15 Gas Venting Trench
16. Gas Extraction Wells
Waste Controls:
17. Treatment of Contaminated Water
18. Chemical Fixation
19. Chemical Injection
20. Excavation and Reburial
21. Leachate Recirculation
Unit of Measure
Site area , ha
Site area, ha
Site area, ha
Trench length, m
Site area , ha
Berm volume , m3
Wei 1 face area, m,.
Wai 1 face area, m.
Wall face area, m
Site area, ha
Pipe length, m ^
Intercept face area, m..
Intercept face area, m.
Intercept face area, m
Site perimeter, m
Site perimeter, m
Flow rate, 1/d
Site area , ha .,
Waste volume, m.
Waste volume, m
Site area, ha
Medium-Size
Landfill
19,006
83,610
17,015
19
909
--
92
1,643
95
8,923,000
399
139
34
1,777
384
247
4
129,500
128
119
22,700
Medium-Size
Impoundment
48,700
62,600
5,000
18
--
7
94
618
121
1 ,588,000
1 ,850
379
139
132
384
247
7
129,400
128
266
--
276
-------
TABLE 7. REMEDIAL ACTION COST DERIVATION - SLURRY TRENCH AT MEDIUM-SIZE LANDFILL
Cost Components
Capital Costs-
Geotechnical Investigation
Slurry Trench Excavation (10,800 m3)
(7ZO m x 15 m x 1.0m)
Bentonite, Delivered (419 tonnes)
Capital Cost Subtotal
Overhead Allowance (25 percent)
Contingency Allowance (30 percent)
Total Capital Costs
O&M Costs:
Monitoring
Sample Collection (13 d/yr, 96 hr/yr)
Analysis
- Primary & Secondary Parameters
- 12 background/yr
- 12 downgradient/yr
24 samples/yr
Total O&M Costs
Total Costs:
Total Life Cycle Cost (over 10 Years)
Average Life Cycle Cost* (per m')
Lower U.S.
$ 3,850
347,760
27.830
$379,440
94,860
1 1 3j830
$588,130
$ 760
7,900
$ 8,660
$661 ,400
$ 61.22
Costs (total*
Upper U.S.
$ 6,520
588,710
74,200
$ 669,430
167,360
_200 ,830
$1,037,620
$ 1 ,570
7,900
$ 9,470
$1 ,117,450
$ 103.47
')
Newark, NJ
$ 5 , 720
516,670
69,570
$591,960
147,990
117,590
$917,540
$ 1,370
7,900
$ 9,270
$995,690
$ 92.19
* For a 10,800 m2 wall face at a medium size landfill (5.41 ha).
TABLE 8. REMEDIAL ACTION COST DERIVATION - WELL POINT SYSTEM AT MEDIUM-SIZE IMPOUNDMENT
Cost Components
Capital Costs:
Geotechmcal Investigation
Drill Rig Rental (I day)
Well Points (26 m)
Well Point Fittings (5)
Pump, Centrifugal (1)
Capital Costs Subtotal
Overhead Allowance (25 percent)
Contingency Allowance (25 percent)
Total Capital Costs
O&M Costs:
Monitoring
Sampling 12 days (96 hr/yr)
Analysis (24 samples/yr)
Electricity (10,000 kwh)
Total O&M Costs
Total Costs:
Total Life Cycle Cost (over 10 years)
Average Life Cycle Cost* (per m^)
Costs tc
Lower U.S. Upper U
ta »)
S. Newark, NJ
$ 9,500 $ 19,500 $ 16,950
280 470 420
1,040 1,760 1,540
40 70 60
1 ,600 1 ,600 1 ,600
$12,460 $ 23,400 $ 20.570
3,120 5,850
3,120 5,850
$18,700 $ 35,100
$ 760 $ 1,570
7,900 7,900
500 500
$ 9,160 $ 9,970
$96,280 $119,510
$ 321 $ 398
5 140
5,140
$ 30,850
$ 1 ,370
7 900
500
$ 9,770
$113,580
$ 379
* For a 300 m2 intercept face area at a medium size site (0.468 ha).
277
-------
trench excavation, acquisition and delivery
of bentonite, an overhead allowance, and
a contingency allowance. O&M costs are
slight and include only sample collection
and analysis. O&M costs were assumed to
be incurred at face value in the first
year, and slightly inflated values for
each of the 9 subsequent years. These
values are then back-discounted to a pre-
sent value and added to capital cost to
derive total life cycle cost. Unit life
cycle costs can then be derived. For slurry
trenches, costs ranged between $61 and
$103 per square meter of wall face at the
medium-size landfill.
Slurry trench economies of scale are
demonstrated in Figure 1. As shown, costs
do not vary significantly for the 5 scales
of landfill operations examined here. In
fact, for each of the geographic locations
delineated in this figure, costs varied
by less than $10 per square meter depending
on scale of operation. Conversely, slurry
trench construction costs appear to be
quite sensitive to geographic location.
This was seen as largely a function of
proximity to bentonite supplies.
WALL FACI AIIIA IN U * (1OOO-»>
The derivation of unit operation costs
for well point systems has been delineated
in Table 8. Unlike the previous example
for a landfill application, this example
applies a well point system to a surface
impoundment. Cost components include
geotechnical investigation, drill rig
rental, installation of well points and
fittings, pump installation, and overhead
and contingency allowances. O&M consi-
derations include sample collection, ana-
lysis, and electricity. Average life
cycle costs ranged between $321 and $398
per square meter of intercepted face area.
As shown in Figure 2, differences
in scale of operation are shown to be
more significant for well point systems
than for bentonite slurry walls. While
the medium-scale varied between $321 and
$398 per square meter, the total variation
of this application over all 5 scales
of operation is between $100 and $800
per square meter. Considerable savings
in large scale operations are mostly due
to "once-incurred" costs such as drill
rig mobilization, centrifugal pumping
facility, and geotechnical investigation.
3
> 200-
IHTIRCIPT FACI ARIA (• ')
Figure 1. Geographic and Size Variations
Landfill Slurry Trench Costs.
Figure Z. Geographic and Size Variations
Impoundment Well Point System.
278
-------
INCREMENTAL HEALTH AND SAFETY COSTS
Data Collection Approach
As described previously, outputs of
the first project were sometimes found
deficient since they do not adequately
reflect safety and health impacts on re-
medial action costs at uncontrolled sites.
Rather, costs presented can best be de-
scribed as "base construction costs" since
they were derived chiefly from standard
construction cost manuals rather than costs
available on applications at uncontrolled
hazardous waste sites. Recently however,
Superfund activities have proceeded at
specific sites and bids for these projects
are now available. In addition, various
cleanup contractors have prepared them-
selves for Superfund work. As such, they
have developed methodologies for costing
cleanup operations and may have specific
cost guides to be made available to the
authors of this report.
This project was conceived to identi-
fy the incremental costs to remedial action
unit opeations due to safety and health
considerations. These incremental costs
are to be limited to worker exposure con-
siderations; health and safety concerns
as they may impact on local population
are not being taken into account.
As a first step on this new project,
a total of 10 health and safety cost items
were identified. These were developed
from a combination of previous regulatory
experience, extensive observations of ha-
zardous waste sites, discussions with con-
tractors, and literature reviews. [10,
11]. Table 9 contains a listing of these
10 health and safety cost considerations.
Obviously not all of these considerations
are applicable in every case. However,
all eventually have an impact within the
universe of Superfund activities.
After identifying and defining various
health and safety consideration items de-
lineated above, an effort was made to cate-
gorize health and safety precautions into
several specific groups. Fortunately,
the EPA Emergency Response Team has re-
cently developed 4 levels of personal pro-
tection as part of its Interim Standard
Operating Safety Procedures. [12] These
levels of personal protection take into
account each of respiratory and skin pro-
tection, and are defined in Table 10.
As work proceeded subsequently, con-
tacts were made with hazardous waste site
cleanup contractors, regulatory officials,
and consulting firms involved in related
EPA projects. The information obtained
was primarily general cost information
on equipment and materials. Specific
prices with reference to suppliers and
manufacturers were provided for equipment
and clothing costs.
Determining the health and safety
cost for actual site cleanups proved to
be difficult. The number of variables
that entered into the various health and
safety considerations made comparisons
of costs difficult. Additionally, infor-
mation on costs has traditionally not
been collected in a format that enables
health and safety costs to be distin-
guished from other costs. Lastly, in-
formation available from bid documents
does not provide sufficient detail and/or
uniformity to identify and compare spe-
cific health and safety cost items.
To resolve these difficulties, ha-
zardous waste site scenarios (i.e., fic-
tional site models) were developed to
be representative of 3 basic types of
facilities: (1) subsurface burial, (2)
surface impoundments, and (3) above-grade
storage facilities. Whenever possible
these scenarios were developed based on
actual cleanup operations either completed,
in progress, or planned for the future.
This approach was adopted partially to
ensure that the scenarios would reflect
realistic conditions, and partially to
assist in identifying organizations and
individuals experienced in particular
types of cleanup operations.
Each scenario is composed of a number
of distinct unit operations. For each
scenario, site characteristics (e.g.,
topography, size, weather conditions,
hydrogeology, etc.) are defined to provide
a detailed profile of the site. Similarly,
the characteristics of the waste at the
site are defined so that the levels of
protection can be determined for each
unit operation. The levels of personal
protection used for the project are those
defined by the EPA Emergency Response
Team (as delineated earlier).
279
-------
TABLE 9. HEALTH AND SAFETY COST CONSIDERATIONS
1.
Personal Protection;
a. Levels of protection
• foot • head
• hand • respiratory
t eyes • hearing
6. Decontamination
a. Personnel
b. Protective gear
7. Site Security:
d. Monitoring equipment
e. Required facilities, equipment,
f. Change rooms/facilities
body
entry/comnunications
c. Mobile and stationary equipment storage and operation stations
2. Medical Services/Surveillance:
a. Medical examinations
• baseline physical •
• baseline follow-up •
• respirator program
b. On-site
unscheduled due to illness or accident
follow-up to highly toxic exposures
t first aid (explosions, animal/
insect bites, falls, exposures,
etc.)
• emergency communication
• medical personnel
• rescue equipment
c. Off-site
t transportation (ambulance) •
• medical facilities •
• fire department
3. Personnel Trainingj
a. Waste handling
b. Emergency procedures
c. CPR and first aid
d. Protective gear/equipment
e. Monitoring equipment
f. Communication equipment
g. Special/seasonal equipment
4. Manpower Inefficiencies-
a. Restricted mobility
b. Waste handling procedures
c. Monitoring requirements
5. Record Keeping:
a. Management requirements
b. Labor union requirements
c. Governmental
d. Medical
emergency showers/eye washes
emergency medical facilities/
equipment
periodic or continuous
monitoring while working
life squad
coordination with health/medical
services and authorities
h. Rehearsals
i. In-house training
j. Outside programs
k. Safety and health practices
1. Entry and exit procedures
m. Technical orientation - basic
sciences
d. Buddy system
e. Pre- and post- work activities
f. Heat stress
e. Training
f. Work history
g. Site safety/maps
h. Manifests, receipts, and permits
a. Restricting access by personnel
or vehicles
b. Signs and tags
c. Security personnel
8. Insurance:
d. Security systems/equipment
e. Coordination with law
enforcement agencies
a. Comprehensive general liability d. Specific coverage
b. Environmental impairment liability e. State and federal requirements
c. Workman's compensation
9. Emergency Preparedness.
a . Fire fighting
t chemicals
• extinguishers
b. Spil 1 containment /control
• absorbents/chemicals
• oil booms/containment devices
10. Hazard Assessment:
t turnout gear
containers and collection
equipment
a. Sample packaging and shipping requirements
t special containers
• special packing materials
• labels, markings, and placards
• authorized transporters
b. Monitoring equipment
• oxygen detectors
t combustible gas detectors
• organic vapor analyzers
c. Sampling equipment
• disposable sampling equipment
• special material construction
d. Analytical costs (field and contract laboratories)
• qualitative t quantitative
e. Review and interpretation of data
t establishment of levels of • assessment of contaminant
protection migration
• source identification
• chain-of-custody
• manifests and bills of lading
• waste product information sheets
radiation detectors
field test kits
specialized laboratory equipment
specialized equipment
-------
TABLE 10. LEVELS OF PERSONAL PROTECTION
Level A Requires full encapsulation and protection from any
body contact or other exposure to materials. Toxic
by inhalation or skin absorption.
Level B Requires self-contained breathing apparatus (SCBA).
Cutaneous or percutaneous exposure to unprotected
areas of the body (i.e., neck and back of head)
is within acceptable exposure standards.
Level C Requires protection for low level concentrations
in air. Hazardous constituents may be known.
But exposure of unprotected body areas (i.e.,
head, face, and neck) is not harmful.
Level D Requires minimal safety equipment. No identified
hazards present. But conditions may be monitored
and minimal safety equipment made available.
Contractors were identified with par-
ticular experience in each type of scenario.
Cost information is now being solicited
from selected contractors based on their
particular experience, availability of
qualified personnel, and their willingness
to provide the information. Information
initially requested is a breakdown of
health and safety costs (using the cost
component table included previously as
Table 9). Costs are to be detailed by
these contractors at each of the 4 levels
of personal protection, as well as the
base construction cost level. This ap-
proach was intended to obtain incremental
health and safety costs for the specified
site, waste, and pollution conditions.
To increase the amount of data obtained
from each scenario, further information
is being sought on health and safety costs
for each unit operation under different
waste conditions. For example, if the
scenario under consideration includes "ex-
cavation and reburial for dioxin wastes
with a personal protection Level A" addi-
tional costs are also being developed for
the same operation involving less hazardous
wastes requiring personal protection at
each of Levels B, C, and D. A further
important refinement is the effect of vari-
able site conditions, particularly weather
conditions. Incremental costs associated
with extremes of heat and cold are also
being investigated as an extension of the
scenarios.
The data collected from the scenarios
will be used to establish health and safety
incremental cost factors for each unit
operation identified at each of the 4 pro-
tection levels defined in the project.
The incremental cost factors will enable
adjustment of remedial action construction
costs determined using the procedures
established in the first project. The
adjusted cost estimate will incorporate
increased health and safety costs associ-
ated with the anticipated hazard condi-
tions on the site in question.
Bid Cost Example
As discussed previously, some dif-
ficulty has been encountered in extracting
data useful for this project from existing
Superfund site bid documents. For example,
the format for bid documents on two Super-
fund site cleanups (designated herein
as Sites A and B) contain different levels
of detail for bid items to be costed.
For Site A, 15 bid items were listed.
For Site B, a total of 62 bid items were
listed. Table 11 contains the bid quote
submitted for Site A. Table 12 contains
selected bid items (containing some of
the health and safety costs) and the total
project cost estimate for 3 of the bids
submitted on Site B.
Obviously, tremendous variation is
seen in the line items submitted from
various firms. For example, bids for
insurance, bonds, and permits shown for
Site A in Table 11, show a range from
about $1,500 and $60,000. Obviously,
there are differences in interpretations
as to the requirements of insurance, bonds,
and permits. In addition, it is entirely
possible that some contractors are allo-
cating these costs at least in part to
other bid items. Despite the variations
noted above, it is interesting to note
the relative uniformity of bottom-line
costs. Although individual line items
may vary by as much as an order of magni-
tude, the 11 bids received varied within
a tighter range of about $880,000 to
$1,800,000.
Preliminary Cost Data
From data collected to date, some
preliminary cost estimates have been de-
veloped. Table 13 is a compilation of
the estimated personal protection equip-
ment cost at the 4 levels of hazard. Costs
are shown to range from a high of about
$1,800 for Level A protection, to a low
of only $260 for Level D protection. In
281
-------
TABLE 11. BID LINE ITEMS FOR SUPERFUND SITE A
Bids
A
B
C
D
E
F
G
H
I
J
K
Average
1
Insurance
Bonds,
Permits
$18,143
$13,625
$21,500
$16,000
$19,000
$ 1,500
$26,420
$20,000
$11,250
$10,000
$60,000
$19,767
Other
Project Start-
up and Site
Services
$
$
$1
$
$
$
$
$
$
$
$
$
193,762
236,628
,048,679
600 ,000
552,000
446,300
190,000
992,248
475,581
101,600
185,000
456,527
3
Site
Preparation
$161,680
$ 80,310
$ 34,958
$345,835
$290,000
$ 60,000
$130,000
$296,500
$187,431
$112,100
$207,753
$173,324
4
Liquid
Material
Disposal*
$135,927
$493,072
$165,590
$199,312
$266,812
$154,677
$173,250
$ 33,025
$203,960
$211,272
$336,247
$215,740
5
Solid
Material
Disposal*
$
$
$
$
$
$
$1
$
$
$
$
$
534,640
423,392
536,683
522,760
532,000
659,334
,280,200
161,950
297,522
442,340
241 ,000
511,984
6
Salvage
of
Motors
($ 2,327)
($ 4,3,10)
($ 3,590)
($ 431)
($ 4,310)
($ 8,620)
($ 4,310)
($ 1,077)
($ 8,620)
$ 2,004
($43,100)
($ 6,958)
7
Sal vage
of
Scrap
Steel
$ 9,400
($ 4,000)
($ 5,000)
($ 200)
($ 2,000)
($ 4,000)
($ 3,000)
($ 2,000)
($ 2,000)
$ 2,000
($ 4,000)
($ 982)
8
Disposal
of Drummed
Material Cur-
rently on Site
$ 1 ,080
$ 1,045
$ 1,179
$ 1,440
$ 4,200
$ 720
$ 1,680
$ 4,200
$ 1 ,308
$ 900
$ 4,500
$ 2,024
Total
$1,052,305
$1,239,762
$1,799,999
$1,684,716
$1,657,702
$1,309,911
$1,794,240
$1,511,000
$1,166,432
$ 882,216
$ 987,400
$1,371,426
Summation of five bid items.
Summation of four bid items.
-------
TABLE 12. SELECTED BID LINE ITEMS FOR SUPERFUND SITE B
Item
Insurance, Bonds, and Permits
Medical Examinations
Security and Communications Center
Laboratory
Emergency Medical Facility
Personal Hygiene Facility
Equipment Storage Area
Safety Officer
Custodian
Security Officer, 8 hr Nornul Working Day
Personal Safety Equipment and Protective Clothing
Preliminary Staging and Sampling Area
Loading and Waste Consolidation Area
Repack Drums
Overpack Drums
Equipment Decontamination Facility with Clanfier Unit
7 ft High Chain Link Fence
22 ft Wide Double Leaf, Vehicle Chain Link Gate
Collection and Handling of Washwater
Total Project Bid*
Estimated
Quantity/
Unit
1/L.S.
35/person
1/L.S.
1/L.S.
1/L.S.
1/L.S.
1/L.S.
100/mandays
100/mandays
100/mandays
35/person
1/L.S.
1/L.S.
700/ea
700/ea
1/L.S.
1 ,200/ft
3/ea
250/drum
Unit
$
$
$
$
$
$
$
$
$
$
$
$
27
5
54
5
24
6
1
19
$121
$
$
$
$
$
$
74
Bid
Price
,760
400
,670
,139
,611
,203
,812
206
206
129
,067
,471
,109
45
63
,223
22
959
83
11
Amount
$ 27,760
$ 14,000
$ 5,670
$ 54,139
$ 5,611
$ 24,203
$ 6,812
$ 20,600
$ 20,600
$ 12,900
$ 37,345
$ 19,471
$121,109
$ 31,605
$ 44,065
$ 74,223
$ 26,904
$ 2,880
$ 20,762
$1,492,186
Bid
Unit Price
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
$
60,000
3,300
11,987
34,650
8,952
53,575
4,029
240
132
8
2,000
9,537
54,314
20
65
35,163
17
1,037
188
02
Amount
$ 60
$115
$ 11
$ 34
$ 8
$ 53
$ 4
$ 24
$ 13
$
$ 70
$ 9
$ 54
$ 14
$ 45
$ 35
$ 70
$ 3
$ 46
$1,661
,000
,500
,987
,650
,952
,575
,029
,000
,200
800
,000
,537
,314
,000
,500
,163
,700
,110
,893
,079
Bid
Unit Price
$ 39,925
$ 476
$ 12,285
$ 36,768
$ 7,155
$ 37,928
$ 1,748
$ 300
$ 194
$ 80
$ 2,475
$ 9,368
$ 28,509
$ 25
$ 90
$ 40,000
$ 15
$ 545
$ 88
#3
Amount
$ 39,925
$ 16,674
$ 12,285
$ 36,768
$ 7,155
$ 37,928
$ 1,748
$ 30,000
$ 19,400
$ 8,000
$ 86,625
$ 9,368
$ 28,509
$ 17,500
$ 63,000
$ 40,000
$ 18,588
$ 1,635
$ 21,992
$1,717,008
Total includes 43 other line items not directly related to health and safety considerations, and not shown here.
-------
TABLE 13. PRELIMINARY COSTS -
PERSONAL PROTECTION EQUIPMENT
TABLE 14. PRELIMINARY COSTS -
OTHER HEALTH AND SAFETY ITEMS
Level A Cost = $1,854
Fully-encapsulating, chemical-reslstant suit
Pressure-demand, self-contained breathing apparatus
Gloves, inner, chemical-resistant
Gloves, outer, chemical-resistant
Boots, inner, chemical-resistant, steel toe and shank
Boots, outer, chemical-resistant, disposable
Hard hat
vel B Cost = $1,136
Items
Chemical-resistant clothing - jacket, bib overalls
Pressure demand, self-contained breathing apparatus
Gloves, inner, chemical-resistant
Gloves, outer, chemical-resistant
Boots, chemical-resistant, leggin
Boots, inner, chemical-resistant, steel toe and shank
Boots, outer, chemical-resistant, disposable
Hard hat
'el C Cost = $736
Sam-coated disposable suit
Full-face air-purifying canmster respirator
-- front belt mounted
-- back belt mounted
Gloves, inner, chemical-resistant
Gloves, outer, chemical-resistant
Boots, chemical-resistant, leggin
Boots, chemical-resistant, steel toe and shank
Boots, outer, chemical-resistant, disposable
Escape mask
Hard hat
vel D Cost = $260
Disposable coveralls
Gloves, outer, chemical-resistant
Boots, chemical-resistant, steel toe and shank
Escape mask
Hard hat and face shield
some respects, protection stipulated for
Level D may not differ substantially from
typical protective gear applied at normal
construction sites.
Other preliminary costs for miscel-
laneous health and safety items have been
included in Table 14. These numbers were
compiled based upon extensive interviews
with cleanup contractors and other officials
and consultants knowledgeable in the field.
CONCLUSION
In current decision-making processes,
concern for cost-effective use of funds
is a major issue in political, economic,
and budgetary considerations. The limi-
tation of available Superfund money to
address the problems associated with un-
controlled hazardous waste sites necessi-
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
Baseline Physical
Medical Coverage
Personal Protection (includes
clothing and monitoring equipment)
Decontamination (capital)
Decontamination {operational and
disposal )
Site Security (general)
Site Security (guard service)
Training - new employee
Training - after 6 months
Record Keeping
Manpower Inefficiencies - losses
(primarily affected by temperature
extremes and level of protection)
Insurance - Liability
Insurance - Workman's Compensation
Insurance - Bonding
$1 50/person
3 to 10% of base pay
$4,100 to $5,000/person
$12,000 to $15,000/site
$100 /day
1 to 5% of bid price
$10/hour
5 to 10% of man-hour
1 to 5% of man-hour
1 to 5% of working hours
25 to W%
$1 to $2/$100 of sales
$1 to $25/$100 of wages
20 to 100% of bid price
tates cost-effective evaluation and se-
lection of remedial actions. This need
has been reflected in the requirements
of the Superfund Act to provide methods
for an analysis of the relative costs
of alternative remedial actions. The
studies discussed in this paper were de-
signed to assist in fulfilling these re-
quirements.
The primary result of these studies
will be a costing methodology which can
be consistently applied to the various
unit operations and reflect variations
due to geographic location, facility size,
facility type, and level of personal pro-
tection taken to accommodate health and
safety concerns. The resulting cost es-
timates will enable regulatory officials,
consultants, and contractors to (1) com-
pare costs for alternative unit operations
performing the same function, and (2)
budget costs for combinations of unit
operations comprising complete remedial
action plans. Other useful applications
could include:
• Design of a specific unit operation.
For example, cost data generated
by this project could help determine
the location of a bentonite slurry
wall. Should it be placed on-site
284
-------
where excavation could require in-
stallation at Level C personal pro-
tection? Or would it be more cost-
effective to move further downgra-
dient where a longer wall would be
required to contain the dispersed'
plume, but personal protection would
not be required?
• More study vs start of cleanup ac-
tions. For example, should remedial
actions proceed immediately without
the benefit of detailed waste analy-
ses, assuming that "worst-case" Level
A protection is required? Or should
further sampling and analyses be per-
formed if there is a possibility that
results will allow a lessening of
personal protection to Levels B or
C?
ACKNOWLEDGEMENTS
The projects described in this paper
were performed under EPA Contract Nos.
68-01-4885 (Directive of Work No. 12) and
68-03-3028 (Work Assignment No. 14). The
authors would like to thank the EPA Pro-
ject Monitors, Messrs. Douglas C. Ammon
and Donald E. Sanning of the EPA Municipal
Environmental Research Laboratory, Solid
and Hazardous Waste Research Division in
Cincinnati, Ohio for their support.
REFERENCES
1.
A.W. Martin and Associates, Inc.
Guidance Manual for Minimizing Pol-
lution from Waste Disposal Sites.
EPA-600/2-78-142. U.S. Environmental
Protection Agency, Cincinnati, OH.
1978.
ORB Associates, Inc. Remedial Action
for Waste Disposal Sites: A Decision
Makers Guide and Technical Handbook.
U.S. Environmental Protection Agency,
Washington, D.C. 1980.
3.
4.
SCS Engineers. Draft Manual for
Closing or Upgrading Open Dumps.
Environmental Protection Agency,
Washington, D.C. 1980.
U.S.
5. L.A. McMahon. 1980 Dodge Guide to
Public Works and Heavy Construction
Costs. McGraw-Hill, New York, NY.
1979.
6. Fred C. Hart Associates, Inc. Analy-
sis of the Technology, Prevalence,
and Economics of Landfill Disposal
of Solid Waste in the United States -
Volume II. U.S. Environmental Pro-
tection Agency, Washington, D.C.
1979.
7. SCS Engineers. Study of On-Going
and Completed Remedial Action Pro-
jects. U.S. Environmental Protec-
tion Agency, Cincinnati, OH. 1980.
8. Geraghty and Miller, Inc. Surface
Impoundments and Their Effects on
Ground Water Quality. U.S. Environ-
mental Protection Agency, Washington,
D.C. 1978.
9. SCS Engineers. Surface Impoundment
Assessment in California. U.S. En-
vironmental Protection Agency,
Washington, D.C. 1980
10. National Safety Council. Accident
Prevention Manual for Industrial
Operations, 425 North Michigan Avenue,
Chicago, Illinois 60611. Seventh
Edition, 1974.
11. National Safety Council. Fundamen-
tals of Industrial Hygiene. 425
North Michigan Avenue, Chicago,
Illinois 60611. 1977.
12. United States Environmental Protec-
tion Agency, Emergency Response
Team Interim Standard Operating Safety
Procedures. 1982.
Robert S. Means Company, Inc. Building
Construction Cost Data for 1980. 1979.
285
-------
DEVELOPMENT OF A FRAMEWORK FOR EVALUATING
COST-EFFECTIVENESS OF REMEDIAL ACTIONS AT
UNCONTROLLED HAZARDOUS WASTE SITES
Ann E. St.CIair
Michael H. McCloskey, P.E.
Radian Corporation, Austin, Texas
James S. Sherman, P.E.
Radian Corporation, McLean, Virginia
1.0 INTRODUCTION
Background
Many uncontrolled hazardous waste sites
across the nation are currently being
"cleaned up" or are slated for remedial
action implementation in the near future.
Part of these remedial actions are being
funded by the parties responsible for the
site, part by state governments, and
others by the federal government under the
Comprehensive Environmental Response,
Compensation and Liability Act (CERCLA).
In any remedial action plan at hazardous
waste sites, potential remedial alterna-
tives must be identified and the most
appropriate one selected.
The formality required in this identifi-
cation/selection phase may depend on the
source of funding for the remedial action.
CERCLA requires that remedial action con-
ducted with Superfund monies be demon-
strated to be the most cost-effective
alternative that adequately protects human
health and the environment. Whether or
not state agencies or private contractors
use a cost-effectiveness assessment per
se, some methodology is necessary for
selecting the most appropriate plan from a
list of potential alternatives.
The Problem
Systematic assessment of remedial action
alternatives for uncontrolled hazardous
waste sites presents several unique chal-
lenges. The most significant of these
challenges is the absence of a uniform,
well-defined clean-up objective to which
the remedial alternatives can be compared
on a consistent basis. In the National
Contingency Plan (NCP), EPA defines the
cost-effective alternatives as "the lowest
cost alternative that is technologically
feasible and reliable and which effective-
ly mitigates and minimizes damage to and
provides adequate protection of public
health, welfare, or the environment."
Although this guidance mandates considera-
tion of alternatives on a case-by-case
basis, it does little to pinpoint what
will be considered an acceptable level of
cleanup.
Remedial alternatives may also show wide
variances in timeframes over which they
are effective; in many cases, this time-
frame is unknown. Cost-effectiveness
evaluation of remedial alternatives must
incorporate and effectively address a num-
ber of factors which are subjective, dif-
ficult to quantify and for which a common
metric, such as dollars, cannot be
established.
Radian Corporation, under contract to
EPA's Municipal Environmental Research
Laboratory in Cincinnati, Ohio, is devel-
oping a methodology for assessing cost-
effectiveness of remedial action alterna-
tives. The project, initiated in June
1982, will be completed in 1983.
Objectives and Criteria
The basic objective in developing a
methodology for assessing cost effective-
ness is to provide for consistency in
286
-------
decision-making. The methodology should
be simple to apply, requiring only minimal
instructions for its use. In this regard,
the analysis should be based on the small-
est possible number of independent vari-
ables that address relevant concerns.
The methodology should not require vast
information concerning either site condi-
tions or the remedial alternatives being
considered; however, the method should
allow for consideration of newly obtained
information as it becomes available. Also,
the methodology should be conducted with a
precision that is consistent with the
degree of knowledge about the problem or
expected results. Finally, the methodol-
ogy should incorporate a means for deter-
mining the sensitivity of the analysis to
judgements made in applying it.
Organization of Paper
This paper addresses the methodology
developed for assessing cost effectiveness
of remedial action options at Superfund
sites. It should be kept in mind that the
paper presents the generic framework for
assessing cost-effectiveness of remedial
actions. This framework must be tailored
to address site specific conditions.
The paper introduces alternative methods
of evaluating cost effectiveness followed
by a step-wise description of the metho-
dology chosen. The primary emphasis of
the paper is on the technical aspects of
an evaluation matrix and the procedure for
using it.
2.0 RELATED APPROACHES
Risk assessment, cost/benefit analysis,
cost-effectiveness analysis, and trade-off
matrices are types of alternatives evalua-
tion which have been used in assessment of
environmental controls. Certain aspects
of these evaluation techniques are appli-
cable to cost-effectiveness assessments of
remedial action plans.
Risk assessments can be used to deter-
mine which options reduce risk to an
acceptable level; however, extensive and
detailed information is needed. It would
be necessary to define the highest accept-
able risk which is likely to be very con-
troversial. Cost/benefit analysis could
be used to quantify for each option the
costs and benefits. If a common standard
of measuring benefits could be developed,
the lowest cost option could be considered
the most cost-effective. Classic cost-
effectiveness evaluation (as defined in 40
CFR, September 4, 1973) could be used to
compare benefits of competing options but
the procedure does not readily conform to
a simple analytical format. As is the
case with cost/benefit evaluation, com-
paring options that have different impacts
(such as containment versus cleanup) is
very difficult. The tradeoff matrix
approach rates options against each other
in terms of cost and effectiveness. The
procedure requires a team of experts to
provide the ratings.
3.0 METHODOLOGY DESCRIPTION
A form of trade-off matrix is presented
here as the selected methodology for eval-
uating remedial actions for hazardous
waste sites. In this methodology, various
alternatives are rated against each other
according to various measures of effec-
tiveness or cost. Weighting factors are
applied to the individual measures, the
measures are rated for each alternative,
and a final score calculated for each al-
ternative. An example of the cost-effec-
tivenss matrix is given in Figure 1.
The matrix is to be completed by a team
of experts and used in conjunction with a
sensitivity analysis. The matrix requires
little "reeducation" of the evaluator
because it requires only that various
alternatives be rated according to the
evaluator"s knowledge of the situation. It
is as usable for situations in which few
data are available as for those in which
extensive background data exist. In fact,
it can be used at more than one point in
the project to update assessments as more
data become available. Sensitivity of the
ranking of alternatives to changes in
ratings can easily be measured.
The Evaluation Team
The application of a trade-off matrix
requires an evaluation team with expert
knowledge of the remedial actions being
considered, the present impacts of the
site, potential impacts from the site or
the remedial actions, and the affected
community.
For most sites, an appropriate team
would include a ground-water hydrologist
or hydrogeologist, an engineer trained and
experienced in remedial actions (most
often a civil engineer), a health impacts
specialist or chemist knowledgeable of the
substances involved and their health
effects, a surface water hydrologist, and
a community impacts specialist, usually a
geographer or sociologist.
287
-------
All*mativ«i
Walghllno Factors
Encompasalno Sturry Wall
wllhCap
Downgradlani Slurry Wall
with Extraction WeMa,
Traat with Carbon
Enaction Walla.
Treat wllh Carbon
Extraction VWUt
Ti«al with BJox
Coil
UAMUTM
i -i 1
Mi *
In 1 1
8 Si S M
Sllt-liKtoiMiKtonI Eltactlniwu Muniw
i ih iHi
ail ii! ft! iilil
Slto-SprcHIc EllKUvMMu
MMMITM
Ii
1 W
H
FIGURE 1
Initial Screening
In most remedial action plans, several
technologies are combined into an overall
remedial action. The Cost-Effectiveness
Methodology can be used to identify a few
highly cost-effective alternatives. How-
ever, as an initial step the user should
screen the list of candidates to eliminate
obviously or marginally inappropriate
alternatives.
To reduce the number of alternatives to
a workable number, the user is to consider
the factors presented in the National
Contingency Plan. These factors are:
• cost,
• environmental effects during imple-
mentation,
• protection of public health, welfare
and the environment, and
• feasibility and reliability of the
alternative.
Any alternative whose costs far exceed
(e.g., by an order of magnitude) the cost
of other alternatives is to be excluded.
Any alternative that would cause signifi-
cant adverse impacts during implementa-
tion, or that after being implemented
would not protect the public health,
welfare, and environment is also to be ex-
cluded, as is any alternative that is not
feasible for the site-specific conditions.
4.0 INPUT TO THE COST-EFFECTIVENESS
MATRIX
As a first step, the matrix itself must
be constructed (with site-independent and
site-specific effectiveness measures and
weighting factors) and cost data for the
alternatives compiled.
Information Needs
The evaluation team must be familiar
with the available information on the
site. The types and amount of data avail-
able will vary widely from site to site,
depending on the amount of previous study
of the site. However, any site-specific
or regional data that can be used in
assessing the feasibility, impacts, effec-
tiveness, risks, and cost of remedial
action should be collected, specifically
information about environmental condi-
tions, waste information, nature and
degree of risk, public issues, and insti-
tutional factors.
Site-Independent Effectiveness Measures
The standards against which the various
alternatives are to be rated are called
"effectiveness measures." Site-independ-
ent effectiveness measures are those which
apply to almost all remedial action situa-
tions. These measures include the
following:
• How clean will the site be after the
remedial action plan is implemented?
• How long will clean-up take?
• How well will the remedial action
reduce health and environmental
impacts?
• What are the impacts of performing
the remedial action?
288
-------
• What can the site be used for after
clean-up?
• Is this technology proven and feasi-
ble?
• What is the risk of failure?
• What would be the impacts of failure?
Site-Specific Effectiveness Measures
For most sites, use of only the site-
independent measures will allow a suitable
effectiveness analysis. However, in
certain cases, conditions or clean-up
objectives unique to the specific site may
dictate factors considered significant
enough to be used as distinct
effectiveness measures. These measures
are to be as independent of the
site-independent and other site-specific
measures as possible. Site- specific
measures can be used in addition to the
site-independent measures or in place of
one or more of them.
For example, if a lake downstream of an
abandoned site has been contaminated by
seepage from the site and if an objective
of the remedial action is to restore the
usefulness of the lake, the degree to
which the remedial action alternatives
meet this objective may be added as a
measure of effectiveness.
Cost
The cost estimates must account for
remedial action construction costs and
annual operation and maintenance cost,
since these types of costs are
specifically identified in CERCLA. A
procedure for estimating costs can be
found in the EPA report: A Standard
Procedure For Cost Analysis of Pollution
Control Operations, Volumes I and II,
EPA-600/8-79-018a and b.
Social Costs
In addition to the direct monetary costs
there are social costs associated with a
remedial action at a hazardous waste site.
Though the public may regard the action
as beneficial to society, it should be
anticipated that some people,
organizations, or units of government will
perceive some aspects of the remedial
action as affecting them negatively.
These negative impacts, "social costs,"
can be either monetary or nonmonetary
(psychological).
Social costs are very difficult to fac-
tor into any methodology used in selecting
a remedial action because they are diffi-
cult to calculate and vary widely by site-
specific considerations. However, these
costs do exist and must be considered.
Five categories of social impacts have
been identified as a starting point for
this evaluation. These categories are ise
useful constructs when looking at the
social costs of remedial action. They are
are:
• psychological impacts,
• sociological impacts,
• political impacts,
• legal impacts; and
• organizational impacts.
Weighting Factors
Weighting factors were identified in the
description of the matrix as a way to
assign relevant importance to the several
cost and effectiveness measures discussed
previously. Because the values of weight-
ing factors can change the results of the
cost-effectiveness comparison of alterna-
tives as drastically as can any other in-
put, the parties-at-interest in the clean-
up should agree on their values. These
weighting factors typically range from 0.1
to 1.0 and are multiplied by the ratings
that each alternative has been assigned
relative to each measure of effectiveness.
In this way a high weighting factor,
appropriate for a measure considered very
important, increases the effect of that
particular measures rating on the final
outcome. Conversely, low weighting
factors are appropriate for less important
measures.
5.0 EVALUATION PROCEDURE
There are fifteen distinct steps
involved in conducting the cost-effective-
ness assessment using the method outlined
above. These steps are:
Step 1. Compile background data on the
site.
This information should include
descriptions of the types and
volumes of waste disposed, geo-
logic aspects of the site (e.g.,
depth to groundwater, soil
types, lithology and depth to
to confining stratum), and any
other data that may exist.
Step 2. Identify appropriate remedial
actions.
Based on the outputs of Step 1,
viable remedial action alterna-
tives are identified. There is a
possibility that some or most of
these alternatives will have been
defined before the evaluator
becomes involved in the assess-
ment process.
289
-------
Step 3. Identify effectiveness and cost
measures.
The measures against which the
various alternatives will be
assessed are selected by the
evaluator. The generic measures,
because of their nature, will
probably be identified for the
evaluator. The site-specific
measures will depend on the site
characteristics, waste type,
surrounding environment, and
alternatives being considered.
The cost measures will typically
consist of construction and oper-
ation and maintenance costs, but
under some circumstances other
cost measures may be included.
Step 4. Assign weighting factors.
The relative significance of all
the measures of cost and effec-
tiveness is established, typi-
cally by a consensus of the eval-
uators through a Delphi process.
Appropriate weighting factors are
then assigned to the measures to
reflect their importance.
Step 5. Rate alternatives.
Each alternative is rated (e.g.,
1 to 5) relative to its ability
to satisfy each measure of effec-
tiveness. The evaluators will
depend primarily on their exper-
tise and knowledge of the situa-
tion to make appropriate judge-
ments. Additional technical
information may be provided to
the evaluation team prior to the
rating exercise.
Step 6. Examine ratings for "outliers".
Examine the inputs from the eval-
uation team to identify any indi-
vidual ratings that seem grossly
inappropriate or indicate a
potential misunderstanding or
extreme inslghtfulness by an
individual evaluator. If such
ratings are identified, consult
the supplier of each rating to
establish why that input was pro-
vided. Do not change any rating
unless the supplier agrees that
the rating was inappropriate and
supplies a substitute. If the
rating is appropriate, share the
information that led the supplier
to use that rating with the other
evaluators and give them oppor-
tunity to change their ratings.
Step 7. Calculate final ratings.
The final rating for each measure
of effectiveness for each alter-
native is then calculated by
multiplying the ratings by the
appropriate weighting factor.
This step enables the evaluator
to factor the relative importance
of the various measures of effec-
tiveness into the analysis.
Step 8. Sum the final ratings for each
alternative.
The sum of all the effectiveness
ratings for each alternative is
calculated. This step involves
summing all the final ratings in
a row, each row corresponding to
a different alternative.
Step 9. Calculate average of all ratings
sums.
Calculate the average effective-
ness rating sum for each alterna-
tive.
Step 10. Estimate costs.
The costs for construction, oper-
ation and maintenance, and any
other identifiable costs, are
estimated. These estimates may
be developed independently by the
evaluator or they may be derived
from available engineering
studies or standard references.
Step 11. Calculate cost ratings.
Cost ratings are calculated by
expressing the cost estimates in
millions of dollars (e.g., a cost
estimate of $753,000 has a corre-
sponding cost rating of 0.753).
Appropriate weighting factors may
be applied as required. Ratings
for each cost measure (e.g., con-
struction, and operation and
and maintenance) should be calcu-
lated.
Step 12. Calculate final cost ratings.
Calculate final cost ratings by
multiplying the cost ratings by
the appropriate weighting factor.
Step 13. Sum of the cost ratings.
Sum cost ratings for each alter-
native as was done for the effec-
tiveness ratings.
Step 14. Calculate cost-effectiveness
scores.
The cost-effectiveness "score"
for each alternative is calcu-
lated by dividing the sum of the
effectiveness measures (Step 8)
by the sum of the cost ratings
(Step 13) for each alternative.
The results of this step provide
the final overall cost-effective-
ness score for each alternative.
290
-------
The alternative with the highest
scores are those that are most
cost-effective.
Step 15. Recalculate trade-off matrix to
check sensitivity.
There will be instances in which
it will be desirable to assess
how the overall cost—effective-
ness rankings change if certain
elements within the trade-off
matrix change. A recalculation
of the affected scores will give
an indication of the sensitivity
of those scores to the changes.
Elements of the trade-off matrix
that may change are costs, indi-
vidual ratings or weighting fac-
tors. The use of such a sensi-
tivity analysis to assess the
changes in overall cost-effec-
tiveness rankings of the alterna-
tives should be considered an
essential element of the metho-
dology.
Selection of Preferred Alternative
An example completed matrix is presented
in Figure 2. While the cost-effectiveness
scores (far right-hand column) are widely
different for Options 2 and 3, the differ-
ences between 3 and 5 may or may not be
significant. While in principal the
highest score is the most cost-effective
option, a sensitivity analysis would
probably confirm that scores close
together are essentially equal.
6.0 CONCLUSIONS
The trade-off matrix offers a flexible
but structural approach to cost-effective-
ness evaluation of remedial action alter-
natives at uncontrolled hazardous waste
sites. The technique can be used in widely
varying situations, but can be tailored to
meet site-specific needs. Because of its
simplicity, the technique is not so time-
consuming as to preclude examination of a.
relatively large number of alternatives.
The approach is easily understood by non-
technical persons and does not require
extensive evaluator education. By relying
heavily on the judgements of experts, the
method is not dependent on extensive site
data, although a good data base undoubt-
edly improves the confidence level of the
evaluation. In addition, the methodology
readily allows examination of the sensi-
tivity of the analysis to judgement or
assumptions made in the evaluation pro-
cess.
A possible drawback to this approach to
cost-effectiveness evaluation is that the
technique may be subject to evaluator
bias. However, care in selection of the
evaluation team and use of the Delphi
technique to deal with outliers in the
ratings can minimize this problem. A dis-
cussion of the ranking rationale can be
supplied to add credibility to the system.
The method is not mathematically rigor-
ous, which may affect its credibility in
some cases. In general, the information
AHwiullv**
Weighting Factor*
UporwHam Dallaclkm Wall
and Cap OVM (Mapoaal Araa
EHcavala Waalaa and
OlapoM In Sacura Undllll
WllnWaOarromOiapoaai
and Ptuma Araa and
tend Walar to POTW
Pluma Araa with Slurry
Wall A Cap; Cottocl Gt W
With Walla A Sand to POTW
EncompaM Olapoaal Araa
Wllh Slurry WaH 1 Ptaca
Cap ovar Dlapoaat Araa
COH
MtUWtt
Pi i
S 51 8 „
1 0
81
6 0
21
133
65
1 2
0
0
41
41
0
(1
6 0
62
1 7«
65
0
1
2.
5 0
2.6
4 . '.
3.0
if1 -I
it ! ii
0 5
1 1
2.0
1.0
1 4
1 2
0 6
1.9
2 6
2 3
2.0
1 9
0 4
1 0
1.8
0.,
1.3
1.0
etlKlhmu
JIfj
ill fi
o.t
2.0
1 2
2 2
1 7
1.8
0
• Muunt
h
1 1
i i
7
2.4
1 1
2 4
1 8
2.0
1.1
3 3
3.5
3 5
2.9
3.5
Silt Sp«llk Ellttlinwu
M«MU»t
0.3
0 8
1 3
1 2
1 2
0.9
.4 9
18 5
16.1
16 5
15 3
It
11
a H
H
18 4
3 1
26.0
9.5
23 5
FIGURE 2
291
-------
base and prediction capability supporting
decisions about remedial actions are in-
sufficient to warrant rigorous quantita-
tive evaluation. To limit evaluation of
sites to those with sufficient hard infor-
mation to support rigorous quantitative
analysis could prevent implementation of
the NCP.
BIBLIOGRAPHY
Cost Effetiveness Analysis Guidelines,
40 CFR Part 35, September 4, 1973.
Dower, Roger C. and Maldonado, An
Overview: Assessing the Benefits of
Environmental, Health and Safety
Regulations, prepared for the U.S.
Regulatory Council, May 1981.
Engineering Science Inc., Water Quality
Management Planning Methodology for Muni
cipal Waste Treatment Needs Assessment,
prepared for Texas Department of Water
Resources, March 1977.
ICF Incorporated, RCRA Risk/Cost Policy
Model Project, Phase 2 Report, Office of
Solid Waste, U.S. EPA, June 15, 1982.
Kufe, Charles and Kilpatrick, Michael,
Rating the Hazard Potential of Waste
Disposal Facilities, JRB Associates and
U.S. EPA.
Municipal Environmental Research Labora-
tory, Handbook, Remedial Action at Waste
Disposal Sites, EPA-62516-82-006, June
1982.
National Economic Research Associates,
Inc., The Business Roundtable — Air Qual-
ity Project — Cost-Effectiveness and
Cost-Benefit Analysis of Air Quality Regu-
lation, Volume IV, New York, November
1980.
Pound, Charles E., Crites, Rondald W.
and Griffes, Douglas A., Costs of Waste-
water Treatment by Land Application,
EPA-430/9-75-003, June 1975.
Proceedings of National Conference on
Risk and Decision Analysis for Hazardous
Waste Disposal, Baltimore, Maryland,
August 24-27, 1981.
Radian Corporation, Cost-Effectiveness
of New Source Performance Standards,
prepared for Office of Environment, U.S.
Department of Energy, January 1981.
SCS Engineers, Cost-Effectiveness of
Remedial Actions at Uncontrolled Hazardous
Waste Sites, prepared for U.S. EPA, Solid
and Hazardous Waste Research Division,
MERL, July 1981.
Unites, Dennis, Possidento, Mark and
Housman, John, Preliminary Risk Evalau
tion for Suspected Hazardous Waste Dis-
posal Site in Connecticut, TRC,
Connecticut 208 Program, and Connecticut
Department of Environmental Protection.
292
-------
SURVEY AND CASE STUDY INVESTIGATION OF REMEDIAL ACTIONS
AT UNCONTROLLED HAZARDOUS WASTE SITES
S. Robert Cochran
Marjorie Kaplan
Paul Rogoshewski
Claudia Furman
JRB Associates
McLean, Virginia 22102
ABSTRACT
JRB, under contract with the EPA, is in the process of conducting a nationwide survey
of uncontrolled hazardous waste sites to identify and examine the various types of reme-
dial action technologies which have been implemented or which are in progress or are
proposed as cleanup techniques. At selected sites, detailed case studies will be per-
formed to document the specific reasons for the success or failure of applied remediations
technologies and to determine the limitations and applicability of these technologies in
other restoration activities or situations.
The results of the case study investigation will be'used to provide an information
transfer between those who have experienced the implementation of a remedial action and
those contemplating or presently assessing or pursuing some form of site cleanup. The
anticipated audience, the survey findings, and case study reports include members of
industry and commerce, state agencies, local authorities, and the EPA. Case study reports
will be structured so that they will provide detailed data on the remediation techniques
employed, the circumstances and conditions in which they were implemented, their apparent
effectiveness in correcting or controlling the problem, and their potential uses in other
natural environments and remedial action situations. Ultimately the survey and case study
results will quantify the number and type of reported uncontrolled releases of hazardous
substances that have undergone a certain degree of remediation; will provide a standard
for comparison when assessing or deciding on a plan for remediation; will identify cleanup
technologies which may warrant further research; and most importantly will provide a forum
in which others can learn from past miscalculations and successes.
INTRODUCTION
With the passage of the Comprehensive
Environmental Response, Compensation, and
Liability Act (Superfund) and the publica-
tion of the National Contingency Plan, the
U.S. Environmental Protection Agency (EPA)
has developed a systematic methodology for
conducting remedial actions at uncontrolled
hazardous waste management facilities.
Prior to the passage of Superfund, EPA
relied on two laws for providing assistance
for remedial actions at uncontrolled
hazardous waste sites, the Clean Water Act
(Section 311) and the Resource Conservation
and Recovery Act.
It is clear from past experiences such
as Love Canal that short term and long term
environmental and health-related hazards
exist when inadequate technologies are used
during the handling and disposal of hazard-
ous materials. Presently the Solid and
Hazardous Waste Research Division of EPA is
involved in the development of existing and
novel technologies for use in the remedia-
tion of hazardous materials released to the
environment. Because new remediation
293
-------
techniques are continually evolving and
known technologies are constantly being
adapted for remedial action use, the
alternatives available for cleanup continue
to expand. In order to assess the effec-
tiveness and limitations of these remedial
action techniques, EPA is conducting case
study examinations of technologies which
have been employed in field situations.
Project Description
JRB, under contract with the EPA, is
in the process of developing an inventory
of uncontrolled hazardous waste sites to
identify and examine the various types of
remedial action technologies which have
been implemented, are in progress, or
are proposed as cleanup techniques. In
addition, detailed case studies will be
performed at selected sites in order to
document the specific reasons for the
success or failure of applied reraedia-
tions technologies and to determine the
limitations and applicability of these
technologies.
The results of the case study investi-
gations will provide information to those
contemplating, presently assessing, or
pursuing some form of site cleanup. The
anticipated audience includes members of
industry and commerce, state agencies,
local authorities, and the EPA. Case study
reports will be structured so that they
will provide detailed data on the remedia-
tion techniques employed, the circumstances
and conditions in which they were imple-
mented, their apparent effectiveness in
correcting or controlling the problem, and
their potential uses in other natural
environments and remedial action situa-
tions. Ultimately the survey and case
study results will quantify the number and
type of reported uncontrolled releases of
hazardous substances that have undergone
some degree of remediation; will provide
for comparison when assessing or deciding
on a plan for remediation; will identify
cleanup technologies which may warrant
further research; and most importantly will
provide a forum in which others can learn
from past experiences.
Survey Method
The initial survey task was to compile
an updated list of uncontrolled hazardous
waste sites which included both ongoing and
completed remedial actions. The sources of
information used to compile this list
included: in-house literature review of
selected documents; data from EPA Municipal
Environmental Research Laboratory (MERL)
and Office of Emergency and Remedial
Response (OERR); remedial action programs
and site studies; EPA Headquarters and
Regional contacts with knowledge of reme-
dial actions projects, including work at
Superfund sites; Department of Defense
(DOD) contacts with knowledge of restor-
ation work at military bases; State
environmental and health agencies involved
in site remedial actions; and industrial
and pertinent trade association contacts
involved in spill responses or hazardous
substances cleanups.
The inventory began by conducting a
review and updating data on the 199 sites
identified by SCS Engineers (1), and the
data contained within the list of 114 Top-
Priority Superfund Sites (2). Simultane-
ously, in-house files and publications,
including the Groundwater Newsletter,
Hazardous Waste Reporter, and Hazardous
Materials Report, were reviewed to identify
any sites not contained in the SCS Report
or Superfund list. Once the data review
was completed, EPA regional, state, and
local parties identified as being knowl-
edgeable in hazardous site remediation
activities were contacted to supplement
information collected from the internal
survey and to expand the site data base.
Knowledgeable parties contacted included
EPA Regional Emergency Response Coordina-
tors, Regional Land Disposal Branch per-
sonnel, Regional On-Scene Coordinators,
State On-Scene Coordinators, consulting
contractors, site managers and operators,
Department of Defense, and state and local
officials.
Cooperation was also solicited from
trade associations representing industries
which may have been involved in hazardous
waste management, generation, or trans-
portation and from firms specializing in
the design and implementation of remedial
actions at hazardous materials spills sites
and waste management facilities.
Approximately 300 additional sites
were identified. Data collected on the
sites included the name and location of
the site, the remedial actions implemented
or planned, the type of waste management
294
-------
practices used at the facility, the waste
types/contaminants present, the availa-
bility of engineering cost data, and the
ease of access for a case study. A clear
majority of the sites surveyed were
obtained from two data sources, the SCS
Report (1) and the list of 114 Top-Priority
Superfund Sites (2). However, contacts
made during discussion with responsible
parties involved in remediation activities
at the Superfund sites and SCS sites often
led to discovery of sites not previously
listed.
The results of the DOD effort indicate
that remediation activities within the
armed forces are in the initial stages.
DOD has established a phased approach for
conducting site restoration activities
within all branches of the Armed Forces.
Presently, each branch of DOD is proceeding
at individual rates and in most cases have
conducted initial site assessments but have
not initiated any site restoration
activities.
Contacts with several cleanup firms,
consultants specializing in remedial alter-
native design, and trade associations, led
to the discovery of several sites. How-
ever, in most cases client confidentiality
agreements hindered full cooperation in
identifying sites. It is reasonable to
assume that the many private sector clean-
ups are completed but are not reported.
Therefore these are not included.
Once the site survey activities were
completed, criteria were established to
select candidate sites for detailed case
study analysis. These criteria included:
o Availability for field survey
activities
o Availability, accessibility, and
completeness of remedial action
cost and engineering data
o Type of remedial action technology
implemented, so that a range of
remedial action techniques were
investigated
o Type of waste management practice,
so that a wide range of tech-
nologies common to hazardous waste
management were studied
o Types of waste and contaminants
present at the facility to ensure
that a variety of waste streams and
pollutants were included
o Hydrogeologic setting, so that a
variety of settings were repre-
sented
o Geographic location to provide a
nationwide distribution of sites.
Approximately 20 sites have been
identified for case study analysis and
field surveys for these sites are presently
being performed.
Survey Results
Three hundred and sixteen sites have
been identified with some form of com-
pleted, ongoing, or planned remediation
activity relative to the uncontrolled
releases of hazardous substances to the
environment (Table 1). This is a substan-
tial increase in the number of sites
identified in a similar survey conducted in
1980 (3). Two key factors can explain the
increased number of identified sites; these
are:
o The data base used in the 1982 sur-
vey was an expanded version of the
1980 data base, including contact
with the Department of Defense,
pertinent trade associations,
industrial contacts, and cleanup
contractors, and
o Additional data relative to
hazardous waste sites has been made
available through the EPA Field
Investigation Team activities over
the past two years and there has
been incresed awareness by state
and local officials relative to
site discovery and identification.
Although the number of identified
sites has increased by 147 the percentage
of sites located in the individual states
has not changed appreciably. It is also
apparent that a majority of sites identi-
fied are concentrated in the heavily
industrialized regions of the country.
This corresponds well with the economics
and convenience of disposing of waste
materials near the generating source.
295
-------
TABLE 1. STATE LOCATION OF REMEDIAL ACTION SITES IN 1980 AND 1982
State
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
Number of
Sites '80
2
0
3
2
3
3
4
2
7
4
0
0
8
3
1
2
5
3
2
1
5
11
3
0
4
5
0
0
1
10
0
14
7
6
4
0
0
16
4
3
0
10
3
1
0
2
0
1
3
1
Percent of Total
Identified
Nationwide '80
1.2
0
1.8
1.2
1.8
1.2
2.4
1.2
4.1
2.4
0
0
4.7
1.8
0.5
1.2
3.0
1.8
1.2
0.5
3.0
6.5
1.8
0
2.4
3.0
0
0
0.5
5.9
0
8.3
4.1
3.6
2.4
0
0
9.4
2.4
1.8
0
5.9
1.8
0.5
0
1.2
0
0.5
1.8
0.5
Number of
Sites '82
2
0
6
4
9
5
11
4
26
5
0
0
14
4
2
3
6
9
3
3
11
13
8
3
10
5
0
1
6
17
3
29
8
6
5
3
0
23
5
5
2
10
7
2
1
4
5
2
5
1
Percent of Total
Identified
Nationwide '82
0.6
0
1.9
.3
2.8
1.6
3.5
1.3
8.2
1.6
0
0
4.4
1.3
0.6
1.0
1.9
2.8
1.0
1.0
3.5
4.1
2.5
1.0
3.2
1.6
0
0.3
1.9
5.3
1.0
9.2
2.5
1.9
1.6
1.0
0
7.3
1.6
1.6
0.6
3.2
2.2
0.6
0.3
1.3
1.6
0.6
1.6
0.3
Total 50 States
169
316
296
-------
Table 2 is a compilation of the vari-
ous types of disposal methods employed at
the sites identified during the survey.
The data collected from approximately 30 of
the sites does not allow an accurate
assessment of the disposal method used.
Therefore, these sites were eliminated from
Table 2 (4).
Waste management practices and con-
taminant releases identified during the
survey include land disposal, drum and tank
storage, incineration, and treatment, sub-
surface injection, spills and illegal
dumps. Data indicate that 78 percent of
the facilities where remedial actions were
either completed, ongoing, or planned can
be associated with three waste management
technologies: landfilling, drum storage,
or surface impounding. This association is
to be expected since these technologies
have been the most common methods over the
years for managing and disposing of hazard-
ous substances. The relatively low per-
centage of incinerators and injection wells
identified in the survey can be attributed
to the limited applicability of these tech-
nologies for treating and disposing of a
wide spectrum of hazardous materials, and
to their limitations for use in broad
geographic and environmental regions. The
low percentages of illegal dumps identified
can be associated with the expected
unavailability of data necessary to suffi-
ciently characterize these facilities.
The remedial technologies used at
the surveyed sites included capping and
grading, removal and offsite burial,
groundwater pumping, chemical and bio-
logical treatment, and containment and
encapsulation (Table 3). However, 76 per-
cent of the remediating efforts identified
involved four techniques: capping/grading,
drum removal, contaminant removal, or con-
taminant treatment. In most instances,
these techniques correlate well with the
high percentage of facilities identified as
practicing landfilling, drum storage, or
surface impounding as a waste management
method. This correlation is based on the
following factors:
o Most remedial actions to date have
been directed at controlling the
immediate threat, i.e. removal of
the waste material and contaminated
soil, surface impoundment pumping
and removal, or drum removal
o Technologies such as grading/
capping, contaminant removal, and
drum removal are in most cases
relatively unsophisticated and
inexpensive remedial activities
when compared with other remedial
options
o Removal of the contaminant source
is the most likely initial step in
performing a staged facility
cleanup
o Complete removal of the source of
contamination is the most direct
method of reducing or eliminating
continued releases of contaminants
to the environment.
TABLE 2. TYPES OF DISPOSAL METHODS
Disposal Methods
Number of Sites
Percent of Total
Landfill
Drum Storage
Surface Impoundment
Spills
Incinerator
Injection Wells
Illegal Dumps
121
67
104
25
10
8
39
32
18
28
7
3
2
10
TOTAL
373
100%
*In some cases, more than one disposal method has been used at an individual
site.
297
-------
TABLE 3: TYPES OF REMEDIAL ACTION EMPLOYED
Remedial Action
Number of Sites
Percent of Total
Capping/ Grading
Drum Removal
Contaminant Removal
Groundwater Pumping
Groundwater Containment
Contaminant Treatment
Encapsulation
Dredging
Gas Control
Incineration
Lining
59
56
70
22
23
48
8
5
3
3
7
TOTAL
304
19
18
23
7
8
16
3
2
1
1
2
100%
*In some cases, more than one remedial action technique has been used at an
individual site.
Less often used methods of site reme-
diation include encapsulation, dredging,
incineration, groundwater containment or
pumping, and subsurface gas control.
Several reasons exist for the limited use
of these methods as remediation techniques,
these include:
o Constraints based on site specific
conditions such as waste type, area
of contamination, and media con-
taminated
o Present level of technological
development relative to proven
field use and successful applica-
tion in real world situation
o Effects of economic and institu-
tional factors which limit the
availability of funds, or require-
ments for specific approaches to be
used in resolving the cleanup
problem (i.e., emergency cleanup
vs. long term cleanup).
Based on the information gathered on
the restoration activities of relatively
new sites there has been a shift in the
approach historically used to plan and
implement remedial actions. In more recent
times, emphasis has been placed on
thoroughly understanding the dynamics of
the contamination and then developing and
implementing the remedial action program.
The shift to more systematic approaches in
planning remedial actions can be attributed
to more comprehensive regulations relative
to hazardous waste management and the
assignment of strict site cleanup liabili-
ties and responsibilities (i.e., Superfund
and the Resource Conservation and Recovery
Act), and also to lessons learned from
successful and unsuccessful cleanup methods
and techniques employed in the past.
Case Studies
Once the case study sites were
selected, field activities were conducted
to collect additional information to ensure
the development of accurate and complete
case history reports. Field visits
included trips to the sites as well as
meetings with the appropriate Federal,
State, and private parties involved in the
site remediation. The following is a brief
synopsis of three case study investiga-
tions .
Trammel Crow Site - Dallas, Texas
The first case study was a former
Texaco oil refinery which operated in
Dallas, Texas from 1915-1945. When the
refinery ceased operation, it was sold to a
298
-------
metal recycling firm which reclaimed any
valuable scrap metal on site. The property
was left vacant for twenty years after the
scrapping operation until it was sold in
1980 to the Trammell Crow Corporation for
industrial development. At the time of
purchase, five waste sludge pits with over
five million gallons of still bottoms and
other petroleum refinery wastes were
located on the site. The five ponds were
not all the same size and did not contain
the same materials. One pond, the largest,
contained approximately 3.5 million gallons
(16,600/yd ) of crude oil tank bottoms con-
sisting of 50 percent carbonaceous mate-
rial, 35 percent water, and 15 percent ash.
The carbonaceous portion of the sludge was
made up of equal portions of asphaltenes
and paraffins. A second contained approxi-
mately 10,000/yd of hard coke/slag mate-
rial believed to be coke cinders from the
petroleum refinery cracking process. The
remaining three ponds contained approxi-
mately 1.5 million gallons of sedimentation
or oxidation pond oily residues.
The waste ponds were located above a
Trinity Clay soil which extends to a depth
of 20 to 45 feet below the ground surface.
The Trinity Clay is a moderately alkaline
soil of low permeability (less than 4.2 x
10 cm/sec). This in turn, overlies an
Eagle Ford shale which extends to a depth
of approximately 400 feet below the sur-
face. The Trinity Clay and Eagle Ford
shale formations form an impermeable
barrier between the surface and the under-
lying aquifer, therefore the potential for
groundwater contamination is low.
The wastes were tested for toxicity
using the EPA required extraction procedure
defined by RCRA. The sludges were not
found to be hazardous. However, the
sludges in the ponds had to be treated
and/or removed before property development
could take place. Although the waste oil
was not a RCRA hazardous waste, it was
classified under Texas law as Class II
industrial waste and could not be put into
a municipal landfill. The costs of trans-
porting the sludge to the closest indus-
trial waste landfill were much greater than
the alternative devised: solidification
and disposal of the sludge in an approved
site. Although the remedial approach was
used for a nonhazardous waste oil, it is
apparent that the same process could also
be applied to similar sites where oil waste
sludges are defined as hazardous wastes
under RCRA.
The technology chosen was a sludge
solidification process using cement kiln
dust available locally. Instead of using
only fresh kiln dust (which is limited in
supply because of outstanding contracts for
the material) to mix and solidify with the
sludge, a stale kiln dust was also used.
The stale dust, which is 38 percent
moisture and stockpiled in large quanti-
ties, is less expensive and is a very
effective solidifying agent (5).
The solidification procedure began
with the three smaller ponds using stale
cement kiln dust. A landfill was excavated
next to the three ponds. Stale cement kiln
dust was delivered to the bottom of the
landfill and leveled by bulldozer into a 6
to 12 inch layer. A backhoe lifted the
sludge out of the pits and it was then
placed on top of the kiln dust in the land-
fill. The bulldozer then mixed the dust
and the sludge into one foot layers with
3:1 and 2:1 dust to sludge ratios. A
pulverizing mixer was then driven over the
one foot layer to completely homogenize the
mixture. Each layer was air dried for
approximately one day and then compacted
and field tested to ensure proper compac-
tion. This procedure continued until the
contents of all three ponds were solidified
in the landfill.
For the largest pond, solidification
was somewhat different as the sludge was
more liquified and the pond itself was
several thousand feet from the landfill.
Because of its liquified nature, the sludge
was solidified using both fresh and stale
kiln dust. The procedure for solidifying
the sludge incorporated the same layering
process used for the three small ponds,
however, the sludge was treated prior to
placement in the on-site landfill. Fresh
cement kiln dust was blown into the sludge
pond using a ratio of 1.5:1 dust to sludge.
A backhoe mixed the dust to form a semi-
solidified sludge. The semi-solidified
sludge was transported to the on-site land-
fill, unloaded onto a bed of stale kiln
dust, and mixed as previously described.
As each side of the pond was excavated, it
was backfilled with clean dirt until all of
the sludge had been removed and solidified
in the on-site landfill. The coke/slag
material was also removed from the
299
-------
remaining pond and mixed in with the sludge
in the on-site landfill.
This entire process took approximately
41,000 tons of kiln dust. This was much
less than originally anticipated because
the stale kiln dust solidified with the
sludge better in the field than in the
laboratory. The projected cost of $500,000
had been based on an estimated use of
75,000 tons of kiln dust, however, the
project cost only $377,528. This was much
less than the alternative of off-site
disposal which was estimated to cost
$1,500,000.
Presently, the site is completely
solidified, graded over, and covered with
vegetation. It is awaiting sale for future
development.
Goose Farm Site - Plumsted Township,
New Jersey
The Goose Farm site is an abandoned
hazardous waste drum burial site that was
partially cleaned up during the one-year
period from August 1980 to November 1981.
The drum burial site is located in Plumsted
Township, Ocean County, New Jersey in a
region known as the Pinelands, a unique
ecological area characterized by acidic
sandy soils and low-lying forests of pine
and oak.
From about 1945 to the mid seventies,
Thiokol Corporation (a manufacturer of
ammunition, rocket fuel, plastics, and
organic fibers) dumped bulk and con-
tainerized hazardous wastes into a pit
about 300 feet long by 50 feet wide by
15 feet deep, under contract with the site
owner. January 1980, township officials
informed the New Jersey Department of
Enviromental Protection (DEP) of the
existence of hazardous wastes at the Goose
Farm site. Shortly, thereafter, DEP con-
ducted a preliminary investigation of the
site, which revealed the discharge of con-
taminated groundwater as a surface seep to
a nearby stream. During the next six
months, site activities included sampling
of the stream and groundwater from 17 moni-
toring wells, conducting metal detector and
resistivity surveys, and reviewing regional
geological data and well driller logs.
A report assessing the situation at
Goose Farm was submitted in June 1980,
which described a contaminated groundwater
plume about 200 feet wide originating from
the drum burial area and moving into the
stream where it was discharging. The plume
was also believed to be moving downward
because of the geology in the area. Pre-
liminary testing was inconclusive, but
suggested potential contamination of up to
60 feet in depth beneath the site, with
the majority of contamination occuring
within a depth of 40 feet. Monitoring of
the upper groundwater and surface seepage
indicated the presence of a large variety
of inorganic and organic solvents, and
pesticides. Levels of contaminants were
highest for methylene chloride, benzene,
and toluene, in the 100 ppm range. Total
organic carbon ranged from 1,600 to 17,000
ppm. The investigation also indicated that
the plume was not an immediate danger to
any drinking water wells in the area, but
if uncorrected could cause widespread con-
tamination of the lower strata, a
moderately used aquifer.
The State established the following
objectives and hired an oil and hazardous
waste cleanup contractor to carry out the
following:
o Obtain data to determine the extent
of contamination of the surface
water and groundwater
o Contain the discharge to the stream
by pumping and treating the
groundwater
o Remove the source of pollution,
i.e. buried wastes and contaminated
soil
o Further groundwater removal and
treatment near the source area.
Temporary containment measures (such as the
installation of an open or gravel-filled
cut-off trench) were suggested but were not
implemented.
Cleanup efforts consisted of con-
structing a well-point treatment system to
prevent the contaminant plume from dis-
charging into the stream; and excavating,
segregating, and treating the buried waste
materials in the disposal area. The well-
point system was located between the
disposal area and the stream and was
oriented in such a way as to intercept the
300
-------
contaminant plume. The well-point system
consisted of about 400 feet of six inch
aluminum header pipe with 100 well-points
spaced about every 7.5 feet. The well-
points were installed to a depth of 22 feet
by water jetting. The system was pumped at
a rate of about 50,000 to 75,000 gallons
per day to contain migration of the con-
taminants .
The collected groundwater was routed
to a treatment system consisting of the
following:
o A vacuum receiver which volatilized
about 20 percent of the total
organic carbon (TOC) in the stream
o Carbon adsorbers to treat the
gaseous stream of the vacuum
receiver
o A clarifier which reduced suspended
solids and removed about 10 percent
of TOC in the stream
o Multi-stage carbon absorbers which
removed about 70 percent of the
initial TOC.
The final effluent after treatment had a
total organic carbon concentration of about
54 mg/1, a reduction of 96.6% to 99.7% in
TOC. After treatment, the effluent was
spray irrigated.
Concurrently waste removal operations
were carried out in the pit area. During a
45-day period, over 4,880 drums and con-
tainers were excavated, analyzed, secured,
and segregated. A backhoe and two
specially designed drum grapplers were used
to complete removal operations. Salvagable
drums were overpacked and stored on-site.
Badly degraded drums were tested for acid-
base compatability and emptied into con-
crete holding tanks prior to disposal. In
addition, about 3,500 tons of contaminated
soil were excavated and stored temporarily
awaiting disposal.
After the drum pit area had been
excavated, a second well-point system was
installed in the pit area. Groundwater was
pumped to the treatment system and the
effluent was injected into the pit area via
injection well-points to flush contaminants
from the underlying soils. The extraction
well-points were first installed in the
unsaturated zone to a depth of 7 to 13 feet
and later lowered into the water table to
collect groundwater contamination.
The groundwater treatment system was
shut down in March 1981 before treatment
was complete because of operating expenses.
Secured drums, bulked liquids, and contami-
nated earth were stored on site from March
1981 to November 1981, awaiting funds for
final disposal. In November 1981, 4,400
tons of waste were transported in 200
trucks loads to an approved waste disposal
site. In addition, 12 drums of PCB's were
transported to an approved hazardous waste
incinerator.
Stroudsburg Site - Stroudsburg,
Pennsylvania
The Stroudsburg site is located in the
borough of Stroudsburg, Monroe County,
Pennsylvania, along the shores of Brodhead
Creek at the site of a historical coal
gasification plant once operated by
Stroudsburg Gas Co.
From the late 1800"s to the early
1900"s, one of the disposal methods prac-
ticed on site for the disposal of coal tar
residuals was the injection of the waste
material into the ground through an injec-
tion well located in the northwestern
quadrant of the plant property. The well
was constructed such that the waste
products were injected into the gravel
alluvium that underlies the plant area,
approximately 20 feet below the surface.
In 1947, Pennsylvania Power and Light
Company PP&L) purchased several parcels of
land from Stroudsburg Gas Co., most of
which were located along the shores of
Broadhead Creek. In October 1980, during
maintenance construction of the flood con-
trol levees along the western shore of
Brodhead Creek, a black, odorous material
was observed seeping from the base of the
dike at several locations along the side of
the stream.
The incident was reported to the
Pennsylvania Department of Environmental
Resources (DER), Bureau of Water Quality
and investigations commenced to determine
the nature and extent of the contamination.
A preliminary assessment of the situation
was made in March 1981 and at this time the
DER requested the assistance of the EPA in
the further investigation of the problem.
The investigative field studies that
301
-------
followed were conducted by DER, EPA and the
Pennsylvania Fish Commission (PFC) and
involved the following the areas:
o The hydrogeology of the site area
o The impact of the coal tar on
stream quality and its biological
community
o The erosional behavior of the
stream.
The remedial actions that were taken
at the Stroudsburg site involved several
technologies. The initial remedial
response to the problem was conducted by
EPA in April 1981 under section 311 of the
Clear Water Act (Public Law 92-500) and
involved the installation of filter fences,
sorbent booms and inverted dams. These
were temporary structures were constructed
to prevent direct and immediate coal tar
seepage into Brodhead Creek.
As field investigations continued, it
was concluded that actions permissible
under section 311 (i.e., emergency, oil
spill response actions) would not suffi-
ciently remedy the pollution problem at the
Stroudsburg site. A more reliable and
permanent remedial system was necessary.
Beginning on November 9, 1981, funds were
appropriated under Superfund.
In November 1981, PP&L began installa-
tion of a recovery well system to remove
the coal tar from the stratigraphic depres-
sion existing behind the west bank levee.
An estimated 35,000 gallons of coal tar had
accumulated in the depression. The project
was completed in late spring of 1982 and
consisted of four well clusters each con-
taining four wells. Presently only one
central well of one of the clusters is in
operation and the recovery rate is less
than had been anticipated. The original
recovery rate predicted was approximately
100 gallons per day. The present rate is
between 20-25 gallons per day. This
situation was primarily caused by an
overestimation of the quantity of
concentrated coal tar present. Much of
what exists is actually a mixture of coal
tar and water, which was not realized until
the wells had already been installed.
Approximately 7,500 gallons of pure coal
tar (95%) have been recovered to date. The
recovered residue is sold to Allied
Chemical in Detroit, MI, where it is used
as fuel.
Concurrent with PP&L's project, EPA
constructed a cement-bentonite, slurry
trench, cut-off wall along the west bank
levee. The containment wall is 648 feet
long, one foot wide and 17 feet deep. The
downstream end of the barrier is hori-
zontally keyed into an impermeable curtain
formed by pressure grouting. The upstream
end is keyed into the sheet piling wall
that exists below the concrete flood wall.
The slurry wall extends over the area of
observed seepage and passes vertically
through the gravel layer that bears the
contaminant and is keyed two feet into the
underlying sand stratum. A total of eight
monitoring wells were installed for
monitoring slurry wall performance and
sampling groundwater in the area. Four
wells are located on either side of the
wall.
In January 1982, following slurry wall
construction and the installation of the
monitoring well system, the contaminated
material was excavated, drummed, and
disposed of at a secure landfill. Through-
out all the activities undertaken at the
Stroudsburg site, materials meant for
disposal were packed in drums and stored on
site until appropriate disposal sites were
located.
Acknowledgements
The authors wish to acknowledge the
assistance of Stephen James, U.S.
Environmental Protection Agency, in
obtaining data and developing research
methodologies which made this paper
possible. This study was performed under
EPA Contract No. 6S-03-3113. Task 39-2.
302
-------
REFERENCES
1. SCS Engineers. "Survey Results and Recommend Case Studies Study of On-Going and
Completed Remedial Action Projects." Prepared for U.S. Environmental Protection
Agency. June 9, 1980.
2. "EPA Announces First 114 Top-Priority Superfund Sites." Environmental News, U.S.
Environmental Protection Agency. Washington. D.C., October 23, 1981.
3. Neely, N.S.. Gillespie. D.P., Schauf, F.J., and Walsh. J.J. "Remedial Actions at
Hazardous Waste Sites: Survey and Case Studies." EPA 430/9-81-05, U.S.
Environmental Protection Agency. Washington. D.C. January 1981.
4. Neely, N.S., Gillespie. D.P., Schauf, F., and Walsh, J.J. "Survey of On-Going and
Completed Remedial Action Projects." Proc. Management of Uncontrolled Hazardous
Waste Sites. Washington. D.C. October 15, 1980.
5. Morgan. D.J., Novoa, A., and Halff. A.A. "Solidification of Oil Sludge Surface
Impoundments with Cement Kiln Dust." Preliminary Draft. Albert A. Halff
Associates, Dallas. Texas.
303
-------
FIELD SCHEME FOR DETERMINATION OF WASTE REACTIVITY GROUPS
Preliminary Investigation
Ursula Spannagel , Richard Whitney, Dean Wolbach
Acurex Corporation
Mountain View, CA 94042
ABSTRACT
Remedial action at hazardous waste disposal sites requires procedures to predict the
potential consequences of mixing waste materials from separate sources. Previous
procedures have assumed a prior knowledge of the chemical composition of the waste
materials. The need for small scale mixing as a safeguard prior to large scale mixing is
emphasized even when the chemical composition of two waste materials indicates
compatibility.
A test scheme has been devised, based upon standard qualitative test procedures, which
will enable workers in the field to determine the gross chemical composition of waste
materials. Chemical composition information obtained from these tests is used to classify
waste materials into reactivity groups, and thus predict compatibility characteristics.
The test scheme includes a field test kit, a series of flow diagrams, and a manual for
using the flow diagrams and test procedures. In addition, a simple device to observe the
effects of mixing two hazardous waste materials has been assembled and is included in the
field test kit.
The test scheme is broken into six procedure sets. Each of the procedure sets consists of
a series of simple tests and acts as a step in the classification of an unknown waste
material. The test procedures have been verified experimentally by applying the tests
directly to a group of 57 compounds, chosen to represent a wide variety of hazard
classes. The information provided in this manuscript is in support of the U.S.
Environmental Protection Agencies implementation of PL-96-510, Comprehensive Environmental
Response Compensation and Liability Act.
INTRODUCTION
During remedial action at an abandoned
waste disposal site, it is frequently
necessary to ascertain whether waste
materials from two sources can be mixed
safely for purposes of bulking,
recontainerization, shipment, and ultimate
proper disposal. The history of hazardous
waste management contains numerous accounts
of disasters resulting from the mixing of
incompatible wastes. Proper management of
hazardous waste therefore requires test
procedures for identification of potential
reactivity hazards so that consequences of
mixing of hazardous wastes can be predicted
and undesired reactions resulting from the
mixing of two wastes can be avoided.
The document "A Method for Determining
the Compatibility of Hazardous Wastes" (1)
classifies wastes by chemical class and/or
general reactive properties and lays out a
sequence of activities to determine the
compatibility of two wastes. The key
elements of this document are a
compatibility chart and a flow diagram for
its use. However, the aforementioned
document presupposes a knowledge of the
composition of the waste materials. This
information is often available from
generators of hazardous waste, but during
304
-------
remedial action at an abandoned site, it is
seldom available. The need to test mix two
wastes on a small scale is emphasized even
if the compatibility chart indicates
compatibility.
PURPOSE
It is the purpose of this study to
establish sequences of simple test
procedures that will enable workers in the
field to classify hazardous wastes
according to their chemical reactivity when
no prior knowledge is available regarding
their composition. The chemical reactivity
classifications (reactivity groups) are
then used to predict which waste materials
can be mixed safely by applying the
compatibility chart in reference 1. A
field test kit will be assembled which
contains all necessary equipment and
reagents for performing the reactivity
tests. The kit will include a simple
device to conduct actual mixing tests on
waste materials shown to be compatible
according to the classification scheme.
Because of the environment in which
these classification test procedures will
be utilized, there are certain desirable
limitations on the schemes, test
procedures, and equipment to be used.
Limitations that are being specifically
addressed during the course of this study
are as follows:
» The procedures must be safe and
easy to perform by non profes-
sionals. To maximize the
safety of the test procedures,
only small amounts of the test
materials are being used, and
materials of extreme reactivity
are identified early in the
testing sequence. The simplicity
of the test procedures will be
verified by using nonprofessionals
to perform the tests in the later
part of the program.
• The procedures must give
definitive, objective results.
Test materials are analyzed as
unknowns to verify objectivity of
observation.
• The equipment must be standard,
readily available, portable, and
require minimal utility hookups,
since the tests will be applied in
remote field locations. For this
reason, the test equipment is
entirely self-contained (i.e.,
requires no utility hookups,
etc.).
• The procedures will need to be
performed as rapidly as possible,
since numerous waste samples will
require testing. For this reason,
abbreviated classification schemes
are proposed as a time saving
measure.
• The test procedures must be able
to be performed with slightly
restricted manual dexterity, since
they may be performed in areas
where protective clothing is
requi red.
• The test procedures must be
reproducible.
SCOPE
A set of 57 test compounds were
selected from a listing (Appendix 2 of
Reference 1) of materials likely to be
found in hazardous waste disposal
facilities. The compounds were selected to
represent as broad a range of chemical
reactivity as possible. All but one of the
reactivity groups from the compatibility
chart (1) is represented by the 57 test
compounds. A list of the test compounds
and assigned reactivity groups are shown in
Table 1.
Test procedures were identified from
literature sources (2, 3, 4, 5, 6) and
arranged into test schemes to classify
waste materials into reactivity groups.
The test procedures are applied to the
58 test compounds. In the development of
the test schemes, each of the test
procedures was applied to the test
compounds and results were compared to
those expected. The test schemes and
procedures were then applied to the 57 test
compounds anonymously and the results were
used to classify the "unknown" test
compounds into reactivity groups.
All equipment and reagents used in the
test schemes is being assembled into a
field test kit. A manual for the use of
the test schemes, including detailed
305
-------
TABLE 1. TEST COMPOUNDS FOR HAZARDOUS WASTE COMPATIBILITY TESTING
Acrolein
Aldicarb
N-Butyl acrylate
Diacetone alcohol
Etnyl acrylate
Fluoroboric acid
Hexaf luorophosphonc acid
Hydrofluoric add
Hydroxyacetophenone
Lannate
Mercaptobenzothiazole
Methyl methacrylate
Peracetic acid
Polypropylene
Polysul fide polymer
Propylene oxide
Vinyl acetate
Selenium
TEPA
Ziram
Cadmium
Chromium
Copper
Lead
NicKe!
Ethylene glycol
monomethyl ether
Styrene
Bromoxym ]
Dipicryl a mine
Barium
Sod i urn
Barium iodide
Barium oxide
Picric acid
Acetamide
Aminopropionitnle
Aminothiazole
Benzoyl peroxide
Calcium hypochlonte
Decane
1-decene
Dimethyl ami noacobenzene
Diphenamid
Hexane
1-hexene
Malathion
Malonic mtnle
Naphthalene
Parathion
Toluene ditsocyanate
12345
X
X
X
X
X
X
X
X
X
31331
6 7 8 9 10
X
X
X
X
X
X
X
X
X X
X
34321
11 12 13 14 15
X
X
X
X
X
X
X
X
X
X
12413
fie
16 17 18 19 20
X
X
X
X
X
X
X
X
X
X
X
X
2 3 Z 2 S
activity Group Number
21 22 23 24 25
X
X
X
X
X
X
X X
X
X
X
2 1 4 12 0
(RGN)
26 27 28 29 30
X
X
X
X
X
X
X
X
X
X
X
X
X
33422
31 32 33 34
X
X
X
X
X
X
X
3311
101 102 103 104 105
X
X
X
X
X
X
X
X
X
X.
X
X
X
25522
106 107
X
X
X
0 1
-------
instructions for conducting the test
procedures and directions for use of the
flow diagrams, is included in the kit
also.
A mixing device has been assembled and
is being evaluated for its ability to
detect reactivity by gas and heat
evolution. The design of the device is a
simplified version of similar devices
reported earlier in the literature (6, 7).
At the conclusion of the study, the
field test kit will be employed in a 1-week
field evaluation at the EPA Combustion
Research Test Facility located in
Jefferson, Arkansas. The tests will be
applied to several "real" wastes during the
field evaluation.
DISCUSSION OF TEST PROCEDURES
Test Procedures
A series of test procedures has been
assembled which can be used to classify
hazardous wastes based upon their gross
chemical composition. These test
procedures are based upon established
qualitative chemical tests (2, 3, 4, 5).
Specific functional group tests have been
identified for all but four of the 41
reactivity groups (RGN's) listed in
Reference 1. A complete listing of tests
by reactivity group is presented in
Table 2. Details of the test procedures
are lengthy and therefore do not warrant a
full description here. Instead, procedural
references are also given in Table 2.
However, general comments concerning some
nonspecific test procedures warrant brief
discussion. Personal protective equipment
required for these procedures will be
utilized.
Physical examination of the waste
material reveals the phase (liquid, sludge,
solid) and may give clues about the
material present (elemental metals, for
example). Paper tests are conducted to
test for pH (pHydrion paper) oxidative
potential (iodide-starch paper) and
reductive potential (methylene blue paper).
Liquids are tested by direct application to
the test paper. Solids are tested by
applying the solid to wet paper, by
applying an aqueous solution of the
material to the test paper, or by applying
a solution of the material in organic
solvent to the paper, drying the solvent,
and wetting the paper with water.
(CAUTION: Care should be taken when
wetting an unknown solid with water.)
Ignition (flame) tests are performed
using a small crystal or droplet of
material on a nichrome wire. The material
is held near the flame, then gradually
brought into the flame until a response is
noted. Larger amounts are tested using a
porcelain spoon where further testing is
indicated.
Solubility-reactivity tests are
performed by placing a small amount (a few
drops or equivalent) of the waste material
in the depressions of a porcelain spot
plate. These portions of waste are treated
with a few drops each of water, hydrochlo-
ric acid, nitric acid, sulfuric acid,
3M sodium hydroxide, methanol, acetone,
hexane, and toluene. Concurrent with these
tests, liquid wastes are tested for the
presence of water with anhydrous copper
sulfate, and tests are performed for
sulfide (lead acetate paper) and cyanide
(Table 2).
Functionality tests are used as a
final assignment of waste materials into
reactivity groups, based upon the
functional groups present. The tests are
established qualitative organic and
inorganic procedures, and are referenced in
Table 2. (Warning: The limits of
detection of these tests in some cases may
not be sufficient to identify trace
quantities of materials, e.g., cyanide
concentrations that could be hazardous to
personnel).
The procedures have been organized
into a visual examination phase and six
"procedure sets (PS)", as outlined below:
• PS 1 — pH and redox tests to
identify acids, caustics,
oxidizing and reducing agents
• PS 2 — Solution-reactivity and
special functionality tests to
determine reactivity and/or
solubility in a series of acids
and solvents and identify the
presence of water, sulfide and
cyanide
• PS 3 — Flame tests to identify
combustible (organic) and
307
-------
TABLE 2. SUMMARY OF SPECIFIC TEST PROCEDURES
RGN
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Reactivity
group name
Acids, mineral ,
nonoxidizing
Acids, mineral ,
oxidizing
Acids, organic
Alcohols and glycols
Aldehydes
Amides
Amines, aliphatic and
aromatic
Azocompounds, diazo
compounds and
hyd razi nes
Carbamates
Caustics
Cyanides
Dithiocarbamates
Esters
Ethers
Fluorides, inorganic
Hydrocarbons, aromatic
Halogenated organics
Isocyanates
Ketones
Name of test*
pH
pH, oxidation
pH, flame
Vanadate-quinol inol
Dini trophenyl hyd razi ne,
Schiff's test
Oxamide test
p-Ni trobenzened i azoni urn
tetraf 1 uoroborate
Dimethyl ami nobenza Id ehyde,
glutaconic aldehyde
Di phenyl carbohyd razide
PH
Chloramine-T
Copper chloride-acetic
acid
Hydroxamate
Iodine test
Zirconium alizarinate
Friedel -Crafts test +
ferrox test
Sodium fusion
• • •
Di ni trophenyl hyd razi ne ,
Reference
Paper
Paper
Paper
4,
2,
4,
2,
4,
4,
4,
P-
P-
P-
P-
P-
P-
P-
Paper
5,
4,
4,
2,
4,
2,
2,
2,
2,
P-
P-
P-
P-
P-
P-
P-
173
• •
P.
test
tests
test
172
264-65
264-65
237
666
269
390
test
380
304-05
214
267
420
233-34
231-32
, 177-180
•
264-65
20 Mercaptans and other
organic sulfides
Schiff's test
Lead acetate
2, p. 274
(continued)
308
-------
TABLE 2. (continued)
RGN
21
22
23
24
25
26
27
28
29
30
31
32
33
34
101
Reactivity
group name
Metals, alkali and
alkaline earth,
elemental and alloys
Metals, other
elemental in the form
of powders, vapors, or
sponges
Metals, other
elemental, and alloys
as sheets, rods,
moldings, drops, etc.
Metals and metal
compounds, toxic
Nitrides
Nitri les
Nitro compounds
Hydrocarbon,
aliphatic, unsaturated
Hydrocarbon,
aliphatic, saturated
Peroxides and hydro-
peroxides, organic
Phenols and cresols
Organophosphates,
phosphothioates and
phosphodithioates
Sul fides, inorganic
Epoxides
Combustible and
Name of test*
H20, HC1 , phosphomolybdic
acid
Phosphomolybdic acid
Phosphomolybdic acid
Zinc sulfide,
ammonium sulfide
. . .
Oxamide test
Tetrabase fusion
Baer's test, ferrox test
Ferrox test
Potassium iodide - starch
Ferric chloride test,
p-Nitrobenzenediazonium-
tetrafluoroborate
Sodium fusion
Lead acetate
. . .
Flame test
Reference
3, p.
3, p.
3, p.
3, p.
•
4, p.
4, p.
2, p.
2, p.
2, p.
4, p.
2, p.
2, p.
2, p.
Paper
• •
2, p.
362
362
362
84
• •
264-65
295
162
231-32
231-32
615
228,
237
173 ff
test
•
160-61
102
flammable materials,
miscellaneous
Explosives
Flame test
2, p. 160-61
(continued)
309
-------
TABLE 2. (concluded;
RGN
103
104
105
106
107
Organic
Organic
All
Reactivity
group name Name of test*
Polymerizable . . .
compounds
Oxidizing agents, Iodide, starch
strong
Reducing agents, Methyl ene blue
strong
Water and mixtures Copper sulfate
containing water
Water reactive Water spot test
substances
Other Test Procedures
All RGN's Ferrox test
All RGN's with Sodium fusion
heteroatoms (N, 0,
P, S, X)
All RGN's Flame test
Reference
. . .
Paper test
Paper test
2, p. 161
Spot test
2, p. 231-32
2, p. 172 ff
2, p. 160-61
*WARNING: The detection limits of these tests, in some cases, may not
be sufficient to identify trace quantities of materials.
310
-------
extremely heat-sensitive
(explosive) materials (additional
"indicators" of composition may
also be obtained)
• PS 4 -- Ferrox and sodium fusion
tests to provide elemental
analysis for organic waste
materials
• PS 5 -- Functional group tests for
specific identification of organic
functional groups (RGN's)
• PS 6 -- Functional group tests for
specific identification of
inorganic RGN's
The procedure sets have been organized
into a master scheme for examination of
unknown waste materials. The organiza-
tional sequence of procedure sets into a
general scheme and the organization of
tests into procedure sets are discussed in
subsequent sections.
Procedures Sequence
The test procedures are organized into
a scheme for the classification of
unknown hazardous wastes into reactivity
groups. Preliminary procedures classify
the material as acid, base, oxidizing,
reducing, or water-reactive, and primarily
organic or inorganic. The results of
preliminary tests are used to give
direction to subsequent testing procedures
which are used for further classification,
ultimately into reactivity groups. The
scheme is organized in a manner such that
materials with high reactivity or unusual
hazards are identified early in the testing
sequence. In some cases, information
gained in the early stages of testing can
be used to classify the material and
eliminate the need for further testing, or
can provide cautions to be applied in
subsequent testing activities (i.e.,
explosive hazard).
The sequence in which test procedures
are carried out is shown schematically in
Figure 1. The testing sequence begins with
a visual examination of the waste material
to classify it as a solid, sludge, slurry,
or liquid. This examination may give a
strong indication of its identity (i.e.,
metal).
Depending upon the phase of the waste
material, preliminary testing is initiated.
If the waste is a liquid, slurry, or
sludge, PS 1 is carried out first. The
rationale for beginning with PS 1 is that
the chance for extreme hazards due to
acidity and/or redox potential may be
greater for liquids than for solids and
these hazards are most easily identified.
If the material is found to be strongly
acidic and/or oxidative, the need for
further testing is virtually eliminated
(strong acids and oxidizing agents are
incompatible with most other classes of
wastes).
The test procedure for liquids
continues with solution-reactivity tests
and flame tests to determine the material's
solubility/reactivity characteristics and
to identify the waste as organic or
inorganic. For organics, functionality
tests (PS's 4 and 5) are performed to
classify the waste into specific RGN's.
For inorganics, PS 6 is conducted to
identify specific RGN's.
For solids, the flame test (PS 3) is
carried out first, followed by the
solution-reactivity tests (PS 2). This
sequence is followed to identify explosive
and unusual reactivity hazards (i.e.,
sulfide and cyanide) early in the testing
sequence. From PS's 1 and 2, it can be
determined if the waste is organic or
inorganic. For organics, PS's 4 and 5
tests are performed to categorize the waste
into specific RGN's. Inorganics are
classified using PS 6.
Procedure Sets
A number of test procedures are
employed in the course of hazardous waste
compatibility testing. For the sake of
organization these have been broken into
"procedure sets" which are discussed in
detail in the following paragraphs.
PS 1 — pH and Redox Tests
Early in the testing sequence, pH,
oxidation, and reduction tests are
performed on a waste material or its
aqueous solution by means of test papers.
These tests identify some of the most
reactive of the RGN's (those with the most
311
-------
1
r
PROCEDURE SET 6
1
r
PROCEDURE SET 1
i
PROCEDURE SET 2
\
F
PROCEDURE SET 3
1
1
LIQUID
1
i
PROCEDURE SET 1
\
FILTER
\
SOLID
1
r
PROCEDURE SET 3
*
PROCEDURE SET 2
i
F
1
r
PROCEDURE SET 4
o
START
STOP
CONTINUE
PROCEDURE
RESULT
Figure 1. Sequence of procedure sets.
-------
limited compatibility characteristics).
Specific RGN's identified are as follows:
• RGN 1 — Nonoxidizing mineral acids
• RGN 2 ~ Oxidizing mineral acids
• RGN 3 ~ Organic acids
• RGN 10 ~ Caustics
• RGN 104 — Strong oxidizing
agents
• RGN 105 — Strong reducing agents
PS 1 is shown in Figure 2. It should
be noted that classification into
RGN's 1, 2, or 104 virtually eliminates the
need for further testing since these RGN's
are incompatible with most wastes. In
addition, classification into RGN's 10 and
104 or 10 and 105 has similar
consequences.
PS 2 — Solution-Reactivity and
Special Functionality Tests'
This set of test procedures is
conducted to determine the reactivity of
the waste material with water, acids, and
bases, and the solubility of the waste
material in aqueous acid, base, and four
organic solvents. PS 2 is shown in
Figure 3. The first procedure is the
treatment of the waste material with water.
At this point, most water-reactive
materials are identified (and further
testing is suspended). The solubility of
the material in water is noted. Liquids
are tested for the presence of water
(RGN 106) and all waste materials are
tested for sulfide and cyanide (RGN's 33
and 11, respectively). Finally, the
material is treated with an aqueous base, a
series of acids, and four solvents. The
following RGN's are determined in PS 2:
• RGN 11 — Cyanides
• RGN 33 — Inorganic sulfides
• RGN 106 -- Mixtures contaning
water
• RGN 107 — Water-reactive
substances
In addition, clues are given regarding the
gross identity (reactive, inert, organic,
inorganic) of the waste material.
PS 3 — Flame Tests
Observation of the behavior of a
material upon ignition can provide a great
deal of insight regarding its composition.
Results (observations) upon ignition are
characterized by one of the following
descriptions:
• Burns violently
• Burns (with or without smoke)
• Produces a colored flame, but does
not burn
• Melts but does not burn
• Evaporates or sublimes but does
not burn
Figure 4 shows the classification
observations. While these observations are
somewhat subjective, after the training
period by an experienced chemist most
observers will be able to make a
distinction between organics, inorganics,
and free metals using the flame test.
RGN 107 (explosive) is determined directly
by means of the flame test.
PS 4 -- Ferrox and Sodium Fusion
The classification of a material as
primarily organic or inorganic is made in
PS's 2 or 3. PS 4 is conducted if it is
determined that the material is organic,
and provides elemental analysis
information. The flow diagram for PS 4 is
shown in Figure 5.
This procedure set consists primarily
of two tests: sodium fusion with specific
tests for nitrogen, sulfur, phosphorus, and
halogen (x); and the ferrox test to
identify organic materials containing
oxygen. In addition, specific optional
tests are included to differentiate between
classes of hydrocarbons.
The waste material is first subjected
to fusion with sodium metal. The metal is
thus decomposed, and heteroatoms form
313
-------
START
STOP
CONTINUE
PROCEDURE
RESULT
CLASSIFY
INTO RGN
Figure 2. Procedure set 1 -- pH and redox tests.
-------
( DISSOLVES 1
ftCID SOLUBLE \ / ACID SOLUBLE
SOLVENT SOLUBLE I ^SOLVENT INSOLUBL
Figure 3. Procedure set 2 -- solution/reactivity and special functionality
tests.
-------
START
STOP
CONTINUE
PROCEDURE
RESULT
CLASSIFY
INTO RGM
INDICATOR
(SOLIDS ONLY)
VERY BRIGHT OR
COLORED WITH SMOKE
AND/OR RESIDUE
BURNS
( BLUE FLAME >j
(. (NO SMOKE) )
_^
\PROBABLY
CONTAINS ONLY
CARBON, HYDROGEN
OXYGEN
PROCEED
WITH CAUTION
OR
STOP
HERE
EVAPORATES OR
NO APPARENT
RESPONSE
(SUBLIMES)
FUSES
(DOES NOT BURN
OR EVAPOWTE)
Figure 4. Procedure set 3 -- flame tests.
-------
(OPTIONAL)
A1C13,
(CONDUCT IF
KMn04
X NEGATIVE)
1
Figure 5. Procedure set 4 -- ferrox and sodium fusion tests.
-------
sodium sulfide (sulfur), sodium cyanide
(nitrogen), sodium phosphate (phosphorus)
and/or sodium halide (halogen). Positive
tests for sulfur, nitrogen, or phosphorus
are followed by appropriate specific
functional group tests (PS 6). If tests
for sulfur, nitrogen, or phosphorus are
negative, the ferrox test is conducted to
determine the presence of oxygen. A
negative ferrox test indicates the material
is a halogenated hydrocarbon (RGN
17- determined by sodium fusion) or a
hydrocarbon. Specific optional tests are
included to distinguish between aromatics
(RGN 16), alkenes (RGN 28), and alkanes
(RGN 29). Since compatibility character-
istics of RGN's 16, 28, and 29 are similar,
it is likely that tests to distinguish
these three RGN's are unnecessary.
PS 4 provides for the classification
of five RGN's specifically. These are as
follows:
• RGN 17 — halogenated organics
• RGN 32 -- organophosphates,
phosphothioates, and
phosphodithioates
• RGN 16 -- aromatic hydrocarbons
• RGN 28 -- unsaturated aliphatic
hydrocarbons
• RGN 29 — saturated aliphatic
hydrocarbons
Elemental composition information is
provided which is utilized in PS 5,
discussed in the following paragraphs.
PS 5 — Organic functionality Tests
Elemental analysis information
obtained in PS 4 is used as a starting
point for PS 5. This set of procedures
includes specific tests for 17 reactivity
groups, and is outlined in Figure 6. The
PS 5 contains three major parts: tests for
functionalities containing sulfur,
nitrogen, and/or only oxygen.
If the ferrox test (PS 4) is positive,
tests are conducted for functional groups
containing oxygen. Three of these
(RGN's 3, 30, and 34) are very reactive but
information already obtained in PS's 1
and 2 is used in the classification into
these RGN's. Other reactivity groups
containing only oxygen (RGN's 4, 5, 13, 14,
19, and 31) are less reactive and have
similar compatibility characteristics.
Separate optional tests are provided for
determining the presence of these RGN's,
but they may be eliminated and materials
with positive ferrox test placed into a
composite group representing all of these
RGN's.
If the sodium fusion results indicate
the presence of nitrogen in the waste
material, tests for functional groups
containing nitrogen are conducted. Like
the RGN's containing only oxygen, three
RGN's (8, 9, and 18) are much more reactive
than the others. Tests are therefore
conducted for RGN's 8, 9 and 18 and
optional tests for RGN's 6, 7, 26, and 27
are provided.
If sulfur is present (from the sodium
fusion), RGN's 12 and/or 20 are indicated.
These can be distinguished by individual
(optional) tests. The compatibility
characteristics of these two RGN's are
similar, however, and conducting tests to
distinguish them is probably not
necessary.
PS 5:
In summary, 17 RGN's are determined in
• Oxygen functional groups
RGN 3 — Organic acids
RGN 4 — Alcohols and glycols
RGN 5 — Aldehydes
RGN 13 — Esters
RGN 14 — Ethers
RGN 19 — Ketones
RGN 31 — Phenols and cresols
RGN 34 — Epoxides
RGN 30 — Peroxides
• Nitrogen functional groups
RGN 6 — Amides
RGN 7 — Amines
318
-------
OPTIONAL PROCEDURES
O START
STOP
CONTI
IMUE
PROCEDURE
1
™E
1
__l__
") (POSITIVE j
1
1
|
RON 14
~j
FERRIC
CHLORIDE
TEST
1
^__J___^^
Q POSITIVE J
1
1
DIPHENYLCARBO-
HYDPAZIDE
TEST
R6N 31
'
OIPHENOLCARBO-
BENZALDEHVDE
TEST
1
COPPER
CHLORIDE
I
1
s —
( POSITIVE
T
1
RGN 12
(OPTIONAL)
)
I
LEAD
ACETATE
1
~ — L_^^
( POSITIVE J
1
RGri 20
(OPTIOJ-.L)
GLUT«CON!C
ALDEHYDE
TEST
OXAKIDE
TEST
1
1
1
P-NITROBENZENE-
DIAZONIUH
TETRA-
FLUOROBERATE
1
_-L
TETRABASE
TEST
1
_J
POSITIVE ) ( POSITIVE
POSITIVE \ / POSITIVE) ( POSITIVE
Figure 6. Procedure set 5 — organic functionality tests.
-------
RGN 8 —
Azo compounds,
diazo compounds,
and hydrazines
RGN 9 — Carbamates
RGN 18 — Isocyanates
(no test)
RGN 26 -- Nitriles
RGN 27 — Organic nitro
compounds
Sulfur functional groups
RGN 12 ~ Dithiocarbamates
RGN 20 —
Mercaptans and
other organic
sulfides
PS 6 — Inorganic Functionality Tests
PS 6 includes tests for metals and
other inorganic species not identified in
PS's 1 and 2. These tests are outlined in
Figure 7.
Solids are tested for metals using
phosphomolybdic acid. The form of the
metal (sheets, powder, etc.) is determined
by inspection. Alkali metals are
determined by their water reactivity. If
the material is not an elemental metal it
is dissolved (using HN03 if necessary).
This solution (or liquid waste) is tested
for the presence of heavy metal compounds
(RGN 24) and for fluorides (RGN 15).
The reactivity groups determined in
PS 6 are the following:
• RGN 15 — Inorganic fluorides
• RGN 21 — Elemental alkali and
alkaline earth metals
• RGN 22 — Elemental metals in the
form of powders, vapors,
or sponges
• RGN 23 —
Elemental metals in the
form of sheets, rods,
mouldings, drops, etc.
RGN 24 -- Toxic metals and metal
compounds
It should be noted that a test has not
been identified for RGN 25, nitrides, but
these compounds fit into the explosives
class (RGN 102), determined in PS 2.
Determination of Compatibility
After completion of the appropriate
test procedures, a waste material is
classified into one or more reactivity
groups, based upon the test results. At
this point, the compatibility chart (1) is
consulted to determine the potential
effects of mixing with wastes of other
classes. In cases where compatibility is
indicated, wastes are mixed on a small
scale as described in the following
paragraphs.
Construction of a Test Mixing Device
Compatibility between binary pairs of
waste materials must be demonstrated by
mixing on a small scale even if
compatibility is indicated by the
compatibility chart (1). Devices have been
reported (6, 7) which are used to evaluate
the effects of mixing two materials. One
such device (6) employs a modified Dewar
flask and another (7) employs a Parr
mini reactor. The requirements of such a
device are the following:
• A means to introduce materials
into a closed container
• A stirring device
• Pressure and temperature sensing
devices
A simple apparatus has been
constructed from readily available
laboratory glassware. It is made from a
large side-arm test tube, fitted with
stirring and sensing devices. The stirrer
is a mechanical stirring shaft that can be
turned by hand. A digital thermometer and
a bubbler to detect temperature and
pressure changes are attached. A diagram
of the device is shown in Figure 8.
The test device shown in Figure 8 has
the advantages of portability and ease of
operation over those described earlier. In
addition, unusual effects of mixing can be
observed through the transparent container.
320
-------
SOLID
ROM VISUAL
EXAMINATION
OR
PS 2 OR 3
SHFETS, RODS,
DROPS, MOULDINGS
o
START
STOP
CONTINUE
PROCEDURE
O
RESULT
CLASSIFY
INTO RGN
Figure 7. Procedure set 6 — inorganic functionality tests.
-------
I
Digital
thermometer
Teflon
tube
Figure 8. Apparatus for test mixing
hazardous wastes.
EXPERIMENTAL RESULTS
Tests have been performed on all of
the selected compounds listed in Table 1
for pH flame, oxidation, reduction and
solution-reactivity (corresponding to
PS's 1 through 3). A generalized summary
of pH, oxidation, and reduction test
results is presented in Table 3. A
generalized summary of solution-reactivity
test results is presented in Table 4. All
results represent those expected, except
that certain oxidizing agents bleach
colored test papers.
Sodium fusion and ferrox tests have
been performed for organic test compounds
listed in Table 2. Results for all
compounds are as expected, based upon the
molecular structure of compounds involved.
The sodium fusion and ferrox results are
summarized in Table 5. It should be noted
that excess sulfur tends to interfere with
tests for phosphorus and halogen (due to
precipitation of metallic sulfides). This
interference was removed by acidifying the
fusion extract and boiling briefly to expel
fusion extract and boiling briefly to expel
hydrogen sulfide. In two cases, namely,
aminothiazole and mercaptobenzothiazole,
tests for cyanide (nitrogen) gave weak
positive responses, but a positive
thiocyanate test was produced. A footnote
was therefore added to the test procedure
to test for thiocyanate in the presence of
sulfur.
Results of functionality tests are
presented in Table 5. In all, over 450
individual tests were performed with
sixteen false positive observations and
six false negative results.
CONCLUSION
A series of published analytical test
procedures have been compiled for use in
categorizing hazardous waste into
reactivity groups, using a set of
flowcharts. To date, testing of 57
compounds using test procedures in all of
the six procedure sets has been completed,
with no significant anomalous results. In
addition, several mixtures and "blind"
samples have been tested. Demonstration of
the applicability of the test scheme to
real waste samples is to be completed.
Specific test procedures have been
identified for all but four reactivity
groups listed in Reference 1. These four
reactivity groups are the following:
• RGN 18 Isocyanates
• RGN 25 Nitrides
• RGN 34 Epoxides
• RGN 103 Polymerizable compounds
A continuing effort is being made to
identify appropriate functional group tests
for these four classes of materials.
Two concerns have been expressed
regarding the applicability of these test
procedures to field operations at hazardous
waste sites. The first concern is the
safety of personnel conducting the tests.
Safety is maximized by utilizing very small
quantities of test materials, and by the
use of appropriate protective clothing and
devices (i.e., safety shield). Prior to
conducting the sodium fusion tests, concern
about its safety was expressed. It was
322
-------
TABLE 3. GENERAL SUMMARY OF FLAME, pH, REDOX TEST RESULTS
General
compound
class
Flame test
pH
Oxidation
(Kl-starch)
Reduction
(methylene
blue)
Metals and
metal salts
Acids --
inorganic
Aliphatic
hydrocarbons
and oxygen
CPDs
Colored flames; Cd,
Cr, Pb, Ba, Na burn
brightly, others do
not burn; 003 burns
very brightly
Generally do not
burn; fluoboric acid
bright green; most
give yellow (Na)
flame
Benzoyl peroxide
burns violently,
others have blue
flames -- no
smoke — no residue;
alkenes yellow
flames
003 - low pH 003 - positive Active
test Ca(OCl )2 metals
give NR
Ca(OC1)2
positive test Ca(OCl)2
bleaches
paper
Na and Ba
react to give
pH -11
pH
NR
No response
acids pH 3-5
Benzoyl
peroxide gives
positive
response,
others NR
NR
Aromatic
hydrocarbons
Organosulfur,
phosphorus,
and nitrogen
compounds
Explosives
(pi crates)
Yellow flames, black
smoke
Burns with white -
yellow smoke
Burns violently
("sparkles")
NR
NR
Ami no Cpds
pH -10
pH «5 for
picric acid
NR NR
NR NR
NR NR
NR -- no observable response
323
-------
TABLE 4. GENERAL SUMMARY OF SOLUTION-REACTIVITY TEST RESULTS
Compound class
Hydrocarbons
Esters, ethers,
acids, aldehydes and
ketones
Nitro compounds and
picrates
Amines
Isocyanates, nitriles
Mercaptans
P-compounds
Epoxides
Reactive olefins
(polymerizable)
Inorganic acids
Metallic elements
Oxidizing agents
Salts and oxides
H20
NR
NR
DISS
NR or DISS
NR
NR
NR
DISS, gas
NR
DISS
NR except
Na and Ba
(gas evol . )
DISS
DISS or NR
HC1
NR
DISS
DISS
DISS
NR
NR or DISS
NR or DISS
DISS, gas
DISS
DISS
Slow gas
evolution
DISS
Ca(OCl)2, fumes
DISS, fumes or
NR
HN03
NR
DISS (color
change)
DISS
DISS
(discolored)
Vigorous RXN,
brown fumes
Brown fumes
Brown fumes
Violent
spattering
OISS or NR
DISS
Gas, brown
fumes (some
metals violent
reaction)
DISS
Ca(OCl)2, fumes
DISS, fumes or
NR
H2S04
NR
DISS (turn
brown)
DISS
NR or darkens
NR or darkens
NR or darkens
Darkens
Vigorous RXN
DISS or darkens
DISS + fumes
Slow RXN (gas
evolution)
DISS
Ca(OCl)2, fumes
DISS, fumes or
NR
NaOH
NR
NR or DISS
DISS, red
color
NR
NR
DISS or NR
DISS or NR
DISS or NR
NR
DISS
NR except
Na and Ba
(gas evol . )
DISS
DISS or NR
Polar
solvents
MeOH, acetone
DISS, acetone
NR, MeOH
DISS or NR
DISS or NR
DISS or NR
DISS or NR
DISS or NR
DISS or NR
DISS or NR
DISS or NR
DISS or NR
NR except Na
and Ba (gas
evol .)
003, violent
Ca(OCl)2, NR
NR
Nonpolar
solvents
hexane, toluene
DISS (except
polypropylene)
DISS or NR
DISS or NR
DISS or NR
DISS or NR
DISS or NR
DISS or NR
DISS or NR
DISS or NR
NR
NR
NR
NR
PJR = No observable response
DISS = Dissolves
-------
TABLE 5. SUMMARY OF FUNCTIONALITY TEST RESULTS
Test name
Parameter
Positive False False
Tests results positive negative
conducted expected results results
Sodium fusion
Cyanide/thiocyanate
Sulfide
Phosphate
Halide
Ferrox test
PH
Vanadium oxinate
Dinitrophenylhydrazine
Schiff s test
Hydroxamate
Iodine test
Ferric chloride
Potasium iodide-starch
Oxamide
Di ni trochl orobenzene
Dimethyl ami nobenzal dehyde
Glutaconic aldehyde
Diphenylcarbohydrazide
Tetrabase fusion
Copper chloride
Lead acetate
Baeyer's test
Friedel -Crafts test
Zirconium alizarinate
Phosphomolybdic acid
Sulfide
Organic nitrogen cpds
Organic sulfur cpds
Organic phosphorus cpds
Organic halogen cpds
Organic oxygen cpds
Acids (organic)
Alcohols
Aldehydes and ketones
Aldehydes
Esters
Ethers
Phenols
Peroxides
Amides and/or nitriles
Amines
Azo compounds
Hydrazines
Carbamates
Nitro compounds
Dithiocarbamates
Mercaptans and other
organic sul fides
Unsaturated hydrocarbons
Aromatic hydrocarbons
Inorganic fluorides
Elemental metals
Heavy metal compounds
40
40
40
40
40
40
40
40
40
40
40
40
40
22
22
22
23
22
22
22
22
13
3
10
16
16
18
8
2
2
10
3
3
4
1
5
8
2
2
5
4
1
1
1
4
2
3
4
3
5
10
16
0
0
0
0
1
2
0
0
0
0
7
2
0
0
1
0
0
3
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
1
0
0
1
0
0
3
0
0
0
0
0
0
0
0
Total
755
16
325
-------
found to be satisfactory, however, even
with heat sensitive compounds such as
picric acid and benzoyl peroxide.
The second concern is the length of
time required to perform enough tests to
completely classify a waste material. In
response to this concern, test procedures
for certain reactivity groups will be
labeled as optional, based on similarity of
compatibility characteristics between key
reactivity groups. As an example, for the
purpose of compatibility testing, it is
probably not necessary to distinguish
between aliphatic and aromatic hydro-
carbons. Even if optional tests are
included, it is anticipated that a complete
set of tests (sufficient for categori-
zation) can be performed on a waste
material in 1 to 2 hours, as a worst case.
Mixing tests are expected to take
15 minutes maximum. By using the instruc-
tion manual being written as part of the
test kit, it is anticipated that analysts
without a significant chemistry background
can learn to use the test procedures safely
and effectively with minimal training.
Feigl , F. 1966. Spot Tests in Organic
Analysis. Elsevier Publishing Company,
New York, New York.
Rand, H. C., Greenberg, A. E., and
Taras, M. J., 1978. Standard Methods
for the Examination of Water and
Wastewater, 14th ed., American Public
Health Associaion, Washington, DC.
Flynn, J. P. and H. E. Rossow. 1970.
Classification of Chemical Reactivity
Hazards. National Technical Information
Center, Alexandria, Virginia.
Simmons, B. P., I. Tan, T. H. Li, R. D.
Stephens, and D. L. Storm. 1982. A
Method for Determining the Reactivity
of Hazardous Wastes. U. S.
Environmental Protection Agency,
Cincinnati, Ohio. (Preliminary)
ACKNOWLEDGMENTS
This effort was supported under EPA
Contract No. 68-02-3176; Task 38. The EPA
project officer is Naomi Barkley (Municipal
Environmental Research Laboratory-
Cincinnati). Significant technical
background was contributed by Richard
Carnes, EPA-IERL, Cincinnati. A special
acknowledgment is due the Acurex Technical
Publications staff.
REFERENCES
1. Hatayama, H. K., J. J. Chen, E. R. de
Vera, R. D. Stephens, and D. L. Storm.
1980. A Method for Determining the
Compatibility of Hazardous Wastes.
EPA-600/2-80-076, U. S. Environmental
Protection Agency, Cincinnati, Ohio.
2. Cheronis, N. D. and J. B. Entrikin.
1960. Semimicro Qualitative Organic
Analysis. Interscience Publishers,
Inc., New York, New York.
3. Feigl, F. 1972. Spot Tests in
Inorganic Analysis. Elsevier
Publishing Company, New York, New
York.
326
-------
DEMONSTRATION OF THE BLOCK DISPLACEMENT METHOD
AT WHITEHOUSE, FLORIDA
Thomas P. Brunsing
Foster-Miller, Inc.
350 Second Avenue
Waltham, Massachusetts 02254
ABSTRACT
Foster-Miller, Inc. has conducted a demonstration of a new remedial action
technique for isolating chemical waste sites. This demonstration was car-
ried out adjacent to the Whitehouse Oil Pits in Whitehouse, Florida, and a
report of preliminary conclusions was presented before the National Con-
ference on Management of Uncontrolled Hazardous Waste Sites in Washing-
ton, DC in November, 1982. This paper presents updated data interpre-
tations and conclusions drawn from more detailed data analysis.
The Block Displacement Method (BDM) is intended to place a low permeability
barrier around and under a large block of earth by means of a combination
of drilling, notching, fracturing, and slurry injection culminating in the
upward displacement of the entire earth mass. There are two separate
processes: one for perimeter barrier construction, the second for bottom
barrier construction. These two processes impact each other and the nature
and sequence of their application is geologic-dependent.
In demonstrating the process at Whitehouse, Florida, horizontal fractures
were extended at a depth of 25 ft from seven injection holes within a
60 ft diam block. These fractures coalesced to form a separation extend-
ing to a semi-fractured perimeter.
The block was then displaced upward as much as 12 in. by injection of
approximately 2,000 ft3 of bentonite slurry. Upward displacement was
monitored by standard survey techniques during the injection phase. Fol-
lowing displacement completion core drilling and geophysical surveys were
conducted to verify the integrity of the bottom barrier.
INTRODUCTION
The Block Displacement Method
(BDM) is a new method proposed for
complete in situ isolation of contam-
inated earth, and in so doing, plac-
ing an "impermeable" barrier at the
bottom and sides of the mass. The
concept attracted interest at EPA as
a potentially cost-effective alterna-
tive for remedial, and possibly emer-
gency action. The purpose of the
program discussed herein was to de-
velop the technique through a test
and demonstration program. The pro-
gram was configured to demonstrate
the feasibility of the BDM and to ob-
tain further data for its development
and future implementation. A review
of the preliminary project conclu-
sions was presented before the Na-
tional Conference on Management of
Uncontrolled Hazardous Waste Sites,
in Washington, D.C., in November,
1982. This paper presents an updated
analysis of the results of the field
work, along with our recommendations
for further concept development.
327
-------
Actual costs of the design and
construction associated with the con-
duct of this project have not been
segregated as this is being written;
resources expended within the various
phases of this work will be reported
at the Symposium presentation.
The BDM is of particular value
in strata where unweathered bedrock
or other impermeable continuum is not
sufficiently near the surface for a
perimeter barrier alone to act as an
isolator. The barrier material
should be compatible with in situ
soil, groundwater, and leachate con-
ditions. The barrier is formed by
pumping slurry composed of local or
imported soil, bentonite and water,
into a series of notched injection
holes. The resulting barrier com-
pletely encapsulates the earth mass
or block. A typical BDM barrier is
shown in Figure 1.
SLURRY
INJECTION
.» t
«^
FRACTURED BEDROCK
POS1TIVE SEAL THROUGH
INJECTED BENTONITE
MIXTURE
BOTTOM BARRIER
Figure 2. Creating Separation to
Induce Displacement
Development. The test site was cho-
sen to be in uncontaminated ground
adjacent to a contaminated site, thus
facilitating future application at
that particular waste site. Based on
a review of the initial list of 114
Superfund designated sites, and on an
elimination process based on weather,
geology, litigation status, and state
interest, the Whitehouse site near
Jacksonville, Florida was selected
for the BDM demonstration.
The Whitehouse site is rela-
tively flat terrain composed of
marine sediments of silty sand in ex-
cess of 100 ft overlaying limestone
bedrock. The silty sand was judged
to be sufficiently stratified, see
Figure 3, to complement the block dis-
placement process.
Figure 1. BDM Barrier in Place
A perimeter separation is first
developed and then surcharged (if
necessary) to ensure a favorable hor-
izontal stress field in the formation
(Figure 2).
The scale of the "block" for the
test and demonstration program was
selected to be 60 ft in diameter and
23 ft deep. These dimensions were
considered to be large enough to rep-
resent significant geologic features
and behaviors, yet small enough to be
managed within the resources availa-
ble from EPA's Office of Research and
HOLE T2
200 100
„ STANDARD PENETRATION
, RESISTANCE (BLOUS)
Figure 3. BDM Demonstration Geology
328
-------
Groundwater level in the area
varies from 2 to 5 ft of depth and
local drillers indicated that a thick
hard pan layer existed at a depth of
approximately 20 ft.
FIELD OPERATIONS
The field operation involved
three distinct operations:
• An explosive blast in
three adjacent test holes
to determine perimeter
fracture performance.
• Preparation of the block by
drilling and notching 32
perimeter and seven slurry
injection holes
• Fracturing both bottom and
perimeter and lifting the
block.
Following trial blasts on three
test holes spaced 3 ft and 6 ft apart
respectively, a final perimeter
pattern of 6 in. diam holes on 6 ft
spacing was drilled (Figure 4).
These holes were each notched from
top to bottom with a 15 in. span
notching tool (Figure 5), following
tests with three earlier notching
tool designs.
32 NOTCHED LATERAL
SEPARATION HOLES
3-1/2 in. O.D,
PIPE
3-1/2 in. O.D
PIPE
-HI--3/8 in.
UNSUCCESSFU.
D1SC NOTCHER
SUCCESSFUL
DIAMOND NOTCHER
Figure 4. BDM Perimeter and Injec-
tion Hole Spacing
Figure 5. Perimeter Hole Notchers
Seven injection holes were
drilled and cased with 25 ft long
6 in. diam PVC tubing which was
cemented in place at the base. The
casing extended to a depth of 23 ft
leaving 2 ft of casing exposed above
ground.
Horizontal notches approximately
2 ft in diameter were cut on all
seven injection holes using a slurry
jet notching tool.
In parallel with bottom notch-
ing, an 18 in. diam, 5 ft high con-
crete forming tube was placed over
each drilled and notched perimeter
hole and filled with high density
slurry, (Figure 6). Upon culmination
of injection hole bottom notching all
32 perimeter holes were line loaded
with prima cord explosive (50 grains/
ft) and blasted simultaneously. Con-
necting fractures were observed at
the surface between adjacent peri-
meter holes (Figure 7).
BDM system analysis indicated
horizontal separation should initiate
from the bottom notches at an injec-
tion hole pressure of approximately
20 psi. Initial attempts at injec-
tion required pressures in excess of
329
-------
Figure 6. Surcharged Perimeter Holes
Ready for Explosive Blast
Separation coalescence between
injection holes was observed after
approximately 500 gallons of slurry
was pumped into the central injection
hole. Pressure response was docu-
mented at intermediate injection hole
well heads followed by observance of
slurry flow up the perimeter holes at
multiple points around the block.
Perimeter flow was characterized by
small flows from approximately 10 of
the concrete forming tubes. During
the course of block displacement
several of these tubes were replaced
with 10 to 20 ft long plugs to impede
excessive flows at the perimeter.
Figure 7. Surface Fracture Induced
by Perimeter Blast
40 psi to induce slurry flow indicat-
ing that renotching of each injection
hole followed by immediate slurry
injection was required.
Bottom notched separation propa-
gation was finally realized by in-
creasing notch diameter to approxi-
mately 4 ft using a slurry jet in air
medium to cut the formation (Fig-
ure 8). Each notching operation was
followed by immediate slurry injec-
tion to propagate a local separation.
Renotching and local propagation
was carried out in all six intermedi-
ate injection holes. This procedure
was then applied to the central in-
jection hole followed by steady state
slurry pumping.
Figure 8. Slurry Jet in Air Notching
Operation
Following separation coalescence
block displacement was observed over
most of the block surface. Block
displacement then proceeded over a
2 week period by pumping approxi-
mately 2 yd/hr alternately into each
injection hole. The resulting upward
displacement of the block is shown in
Figure 9.
DISCUSSION OF DATA AND RESULTS
Displacement of the block was
monitored by recording the change in
elevation of 16 fixed rods located
as shown in Figure 10. The numbers
330
-------
'•^^MBHF
Figure 9. Maximum Upward Displace
ment at Surface due to
Block Lift
in parentheses are the total recorded
lift in inches at the corresponding
positions on the block. Elevations
were read through a surveyor's level
located 50 ft outside of the block
perimeter. Additional survey points
beyond the block boundary were moni-
tored during a portion of the lift
operation to verify that no lift was
occurring outside of the perimeter.
Survey points 0 through 6 correspond
to rods located adjacent to (approx-
imately 3 ft from) injection holes.
The remaining 9 survey points 3p
through 30p correspond to rods
located just inside the perimeter.
These survey point numbers corres-
pond to the closest perimeter
drilled hole numbered progressively
from 1 to 32 in a counterclockwise
direction.
A total of approximately
2000 ft3 of bentonite slurry was in-
jected during the block lift opera-
tion. A daily tabulation of slurry
injected during the lift operation
was maintained by counting the number
of 1-1/2 yd barrels of slurry mixed
and verifying this count with a daily
bag count of powered bentonite used.
A comparison of volume of slurry
pumped to volume of block displaced
(Figure 11), indicates an initial
excess of slurry pumped followed by
a consistent correlation once steady
state pumping was achieved.
60 ft
a
X?P
3p
(3.0)
1»(5.7)
2 •(12.0)
in
• (8.1)
(13)*
) (4.2)
•°- w 18p
•o- -•- -o-
30p "•
(2.3)"
'6 »)
(3.8) $
(2.8) ^
. 25pf
(6 7) (J
22p '
(2.8) ^
MEASURED
VOLUME OF
SLOCK LIFT 1000.
(«3)
BLOCK LIFT VOLUME
VERSUS
PUMPED SLURRY VOLUME
500 1000 1500 2000
RECORDED VOLUME OF SLURRY PUMPED (ft3)
Figure 11. Comparison of Slurry
Pumped to Volume Dis-
placed during Block Lift
KEY (DIMENSIONS IN INCHES)
1,- NEAREST HOLE *
'5'7L-ACCUMULATED iLIFT
A—LOCATION BOTTOM
^"^ SLURRY SAMPLES
[11.5] THICKNESS OF
SLURRY
•CORRECTED FOR
ROD RELOCATION
Figure 10. Recorded Block Lift and
Verified Seal Thickness
Data from surface topographic
surveys and accumulated daily lift
records were combined to give two
profiles which inferred the final
block position shown in Figure 12.
In total, the block was displaced up-
ward in excess of 12 in. at its high-
est point and tilted approximately
1 deg from horizontal. A crescent-
shaped portion of the block outside
331
-------
intermediate injection hole #3, but
inside the originally planned block
perimeter, was sheared free of the
upward-moving block and did not move
significantly. The block area near
the planned perimeter lagged the main
portion of the block by 3 to 6 in. in
upward displacement. This lag was
present around nearly the entire cir-
cumference, except for the stationary
crescent.
LBOTTOM SLURR* BARRIE
SECTION A-A
tion of slurry pumping to determine
the integrity of the bottom seal.
Obtaining undisturbed soil samples
of the soft slurry material bounded
by hardpan on top and unconsolidated
sand beneath was a difficult task.
In eight attempts, three acceptable
samples were obtained at the loca-
tions shown in Figure 10. Sample
No. 8, shown in Figure 13, indicates
a well defined boundary of separation
between injected clay slurry and the
overlaying native sand. Traces of
slurry migrated into fractures in
the underlaying sand.
Figure 12. Final Block Displacement
The crescent-shaped shear zone,
tilting of the block and perimeter
displacement lag were all attributed
to an incomplete fracturing and free-
ing of the block around the perim-
eter. Follow completion of the block
lifting operation, 4 to 6 ft deep
trenches were excavated through the
perimeter and shear zone to try to
determine causes of resistance to
perimeter displacement.
Tap roots as large as 12 in. in
diameter and typically on 6 ft cen-
ters were found extending from ap-
proximately 5 ft below surface to the
upper boundary of the hardpan layer.
In addition, perimeter fractures
filled with gelled slurry extended
down only a few feet before becoming
unmapable hairline fractures.
Thin-walled tube soil samples
were taken 4 weeks after discontinua-
Figure 13. Core Sample #8 Showing
Slurry Barrier Placed
in Native Sand at
23 ft Depth
TECHNICAL ASSESSMENT OF DEMONSTRATION
The BDM demonstration yields
three distinct technical conclusions:
• Based upon recovery of core
samples from soil depths
where bentonite clay slurry
had been injected, the
presence of bottom barrier
material where expected was
inferred.
• The perimeter barrier con-
struction technique incor-
porated as part of the BDM
process at Whitehouse was
332
-------
unsatisfactory; conven-
tional perimeter construc-
tion methods should be used
in this and similar geo-
logic materials.
• Data were collected which
confirmed the applicability
of construction materials,
equipment, and field pro-
cedures used for bottom
barrier emplacement at a
specific site.
Based on this experience, it is
estimated that the BDM could be
applied to a variety of geological
conditions, provided that careful
attention is given to the specific
properties of the earth in which the
bottom seal will be placed. Soil
pore sizes must be small enough to
preclude excessive leakoff of slurry.
Close proximity to solution channels
or fractures should be avoided. Also
discontinuities such as boulders and
bedrock intrusions should be avoided.
Finally, careful consideration must
be given to exploit existing aniso-
tropy and in situ stress conditions
in order to successfully create the
bottom separation and seal.
Foster-Miller concludes that the
BDM is feasible and that the White-
house test, and the associated data
acquired, showed sufficient demon-
stration of the key elements of the
process to warrant further develop-
ment and implementation. We also
conclude that the process can be
carried out in other areas where ap-
propriate geology and soil properties
exist, and where evaluation of these
and other site-specific conditions
shows a high probability for success.
Finally, we conclude that the BDM is
a process whose success depends on a
number of site-specific variables,
such as soil gradation, stratifica-
tion, presence and extent of strati-
graphic discontinuities, presence
and nature of geologic fractures or
solution channels, proximity and
nature of utilities and foundations,
and site access. Correct engineer-
ing decisions made with consideration
of these and other variables are
essential.
Based upon the experience ob-
tained at Whitehouse, Florida, we
recommend that application of the BDM
should be based upon a thorough site
evaluation and preliminary design
phase. This should be followed by a
small pilot program on-site to verify
the success of notching and vertical
separation at the base of an injec-
tion hole. This fracturing would
subsequently be expanded underneath
the entire planned block. A site
performance verification program
which includes monitoring of the
barrier effectiveness should be im-
plemented during and following block
displacement.
REFERENCES
1. "User Guide for Evaluating Re-
medial Action Technologies,"
U.S. Environmental Protection
Agency; to be published.
2. deary, J.M. , "A Method for
Displacing Large Blocks of
Earth," U.S. Patent No.
4,230,368, February 1979.
3. Foster-Miller, Inc., Block
Displacement Method Field
Demonstration and Specifica-
tions, Final Report, to be
issued to the Solid and
Hazardous Waste Research
Division, MERL, Office of
Research and Development,
USEPA, Cincinnati, OH under
JRB Contract 60-03-3113, Task
37-1.
4. Brunsing, T.P., and W.E.
Grube, Jr., "A Block Displace-
ment Technique to Isolate Un-
controlled Hazardous Waste
Sites," presented before the
National Conference on Manage-
ment of Uncontrolled Hazardous
Waste Site, Hazardous Materials
Control Research Institute,
Silver Spring, MD, 1982.
333
-------
THE IMPACT OF TOP-SEALING AT THE WINDHAM CONNECTICUT LANDFILL
Rudolph M. Schuller
Alison L. Dunn
William W. Beck, Jr.
SMC Martin Inc.
Valley Forge, PA 19482
ABSTRACT
The landfill in Windham, Connecticut was operated from the late 1940s
until 1979. In 1976, it was shown to be generating leachate and to be
contaminating ground water moving downgradient toward a public water
supply reservoir. In 1978-1980, SMC Martin designed and implemented a
closure plan for the landfill which included regrading, installation of an
impermeable PVC top seal, and revegetation. Total closure costs amounted
to $0.44 per square foot. Investigations of the structural integrity of
the top seal were conducted in 1980. These investigations indicated a
minimal number of small punctures occurred in the top seal membrane.
Spring recharge was effectively intercepted by the membrane and diverted
from the landfill. The area of the plume of contamination was found to
have been reduced. It is concluded that the top seal installed over the
landfill has effectively reduced generation of leachate. Water erosion of
the surface soils resulted after heavy spring rains but was corrected
before damage to the membrane occurred.
INTRODUCTION
SMC Martin, under contract to
U.S. EPA (#68-03-2519) developed
and implemented a program of reme-
dial action at an inoperative waste
disposal site and monitored the
effects of such action on water
quality. Based on a number of
physical and institutional criteria
(landfill size, accessibility,
representativeness; existence of a
serious pollution problem; availa-
bility of basic engineering and
geologic data for the site; avail-
ability of co-funding; freedom from
litigation), an extensive evalu-
ation of more than 400 landfills
was conducted and the Windham,
Connecticut landfill selected as
the most suitable for implemen-
tation of the project.
After a preliminary inves-
tigation of site conditions, sev-
eral possible remedial actions were
evaluated. The principal purpose
of remediation was to reduce the
generation of leachate at the
landfill and minimize its impact on
surface and ground-water quality.
The methods considered included
surface diversion ditches, a slurry
trench cutoff wall, a subsurface
drainage system or a network of
wells designed to lower the water
table, and an impermeable top seal.
This last was found to be the most
334
-------
practical and cost effective for
the site and was applied to the
landfill in conjunction with re-
grading and revegetation. This
paper reviews physical conditions
at the landfill and the specifics
of landfill closure and discusses
water quality improvements that
resulted from remediation.
PHYSICAL SETTING AND EXISTING
CONDITIONS
Windham Landfill is located in
east-central Connecticut, just
north of the City of Willimantic
{Figure 1), on a site framed by the
Windham Airport to the south and
southeast, Mansfield Hollow reser-
voir to the east and northeast, and
Willimantic reservoir to the west
(Figure 2). Both reservoirs are
situated on the Natchaug River.
The stage in Mansfield Hollow
reservoir, operated by the U.S.
Army Corps of Engineers for flood
control, is generally 6 meters
(20 feet) higher than in Williman-
tic reservoir, a public water
supply.
Subsurface materials in the
area consist of granite gneiss
bedrock overlain by more than
100 feet of fine-grained, poorly-
sorted stratified glacial drift.
Surface topography is highly ir-
regular, marked by the hummocks and
kettles characteristic of glacial
terrain. Many surface depressions
between the landfill and Williman-
tic reservoir are filled with small
ponds.
Ground water occurs under
water-table conditions in the
unconsolidated drift at a depth of
7.5 to 9 meters (25 to 30 feet)
near the landfill. Two types of
material were distinguished during
drilling: a buff-colored, poorly-
sorted medium to fine-grained sand
and gravel near the surface, and a
gray, fine-grained sand and silt
occurring below a depth of about
7.5 meters (25 feet). These materi-
als are highly heterogeneous and
only moderately permeable. An
aquifer test performed in July 1978
yielded values of iransmissivity
between 1 and 10 m /day (10 and
100 ft /day). Water is recharged
to the area primarily by infiltra-
tion of precipitation and occasion-
ally by seepage from Mansfield
Hollow reservoir. It flows north-
west and west and discharges into
Willimantic reservoir.
The Windham Landfill began as
an open-burning dump adjacent to a
pond in the 1940s. The pond was
eventually filled in and the wastes
covered with 5 to 6 m (15-20 ft) of
demolition materials. Landfill
operation continued in excavated
trenches west of the old fill. In
1978, the site consisted of two
mounded fill areas separated by a
drainage pipe for the airport,
occupying a total area of 10 hec-
tares (25 acres).
SITE INVESTIGATIONS PRIOR TO
CLOSURE
An initial investigation was
conducted at the Windham Landfill
site in 1975 and 1976 under the
direction of the Connecticut Depart-
ment of Environmental Protection
(Griswold and Fuss, Inc., 1976).
Further background data were ga-
thered by SMC Martin during a
drilling and sampling program from
1277 to 1979 (Emrich and Beck, 1979),
Results from this phase of the
study indicated that the thickness
of fill underlying the older east-
ern section of the site reached
15 m (50 ft) with 6 m (20 ft) of
refuse occurring below the water
table. The newer western section
had a total thickness of 10 m
(33 ft) occurring entirely above
the water table. A plume of con-
tamination, defined by means of an
electrical resistivity survey and
ground-water sampling, was found to
emanate from the landfill and
extend west-northwest and down-
gradient to the Willimantic reser-
voir. Chemical analyses of ground
water from two wells sampled in
May 1979, one upgradient and one
downgradient from the landfill, are
reproduced in Table 1.
335
-------
Figure 1. Location Map for the Windham Landfill
MANSFIELD HOLLOW
RESERVOIR
O WELL
9 SUCTION LYSIMETER
Figure 2. Site Map for the Windham Landfill
336
-------
TABLE 1
CHEMICAL PARAMETERS OF GROUND WATER
UPGRADIENT AND DOWNGRADIENT FROM THE LANDFILL
BEFORE CLOSURE (MAY 16, 1979)
Parameters
Well #1
(Upgradient)
Well #2
(Downgradient)
Temperature (°C) 11
pH 7.2
Specific Conductance
(umhos) 200
Total Solids (mg/1) 368
Alkalinity (mg/1) 25
Chloride (mg/1) 15
Total Organic Carbon (mg/1) 6.2
Iron (mg/1) 2.83
Manganese (mg/1) 0.23
Sodium (mg/1) 3.7
Copper (mg/1) 0.03
Lead (mg/1) 0.03
12
6.4
1000
738
376
90
28.0
176.00
2.05
45.0
0.03
0.03
LANDFILL CLOSURE: METHOD AND COSTS
The closure plan developed by
SMC Martin for Windham Landfill was
designed to effectively seal out
vertical seepage into the landfill
and meet ground-water requirements
of the Connecticut Department of
Environmental Protection. The plan
included regrading of the landfill,
excavation of surface diversion
ditches around the landfill, instal-
lation of an impermeable cover over
the entire surface of the landfill
and under the diversion ditches,
and revegetation. The landfill
cover actually consists of four
layers (Figure 3): an intermediate
sand and gravel cover immediately
above the regraded waste materials,
a 10-cm (4-in) sand bed designed to
protect the overlying membrane, a
20-mil flexible PVC membrane, and a
45-cm (18-in) final cover of sand
and gravel. Sewage sludge and
composted leaves were subsequently
harrowed into the final cover and
seeded with grass. The closure was
begun in July 1979 and completed in
March 1980.
The costs involved in the
closure of the Windham Landfill
have been summarized in Table 2.
Total costs amounted to approxi-
mately 44C per square foot, with
the cost of the liner representing
80 percent of the total. It should
be noted that sand and gravel used
in the intermediate and final
covers were obtained free of charge
from nearby borrow areas.
TOP SEAL INTEGRITY
In April 1980 an investigation of
selected areas was conducted at the
site to determine the membrane's
integrity. Seven excavations (test
pits) were made by hand through the
cover material to the membrane in
both the old and the new sections
of the landfill. It was determined
that the membrane maintained its
integrity during the placement of
the cover material. Although there
were numerous indentations up to
4 cm (1-1/2 in) deep in the mem-
brane, there was no evidence of
puncture. These indentations were
caused by the final cover material
which consisted of both sand and
gravel.
The thickness of the cover
material on the eastern (old)
337
-------
Pond 3
Pond 5
(C
O
o:
UJ
CONTROL POINT
SPECIFIC CONDUCTANCE CONTOUR
(micromhoi / cm)
Figure 3. Distribution of Specific Conductance, May 16, 1979
-------
TABLE 2
CLOSURE COSTS FOR THE
WINDHAM, CONNECTICUT LANDFILL
Item
Costs, in Dollars
Total Per Acre
Per Square
Foot
1. Grading and
Compaction 105,000
2. Sand Blanket 48,000
3. Liner 219,000
4. Final Sand and
6. TOTAL
474,000
4,200
1,920
8,760
18,960
0.096
0.044
0.201
Gravel Cover
5. Seeding
80,000
22,000
3,200
880
0.073
0.020
0.435
landfill varied from 10 to 46 cm
(8 to 18 in). On the western
section of the landfill (new) final
cover varied from 38 to 46 cm
(15 to 18 in). During the place-
ment of the cover material over the
seal, truck traffic patterns had to
be established to assure a logical
and systematic completion of the
project. Several areas of high
truck traffic were selected for
investigation. The results of this
investigation showed no visible
punctures in the seal.
In May 1980, the integrity of
the membrane was again inspected by
staff members from EPA and the U.S.
Corps of Engineers, Waterways
Experiment Station, Vicksburg, MS.
Three additional test pits were dug
on the two sections of the landfill.
Sections of the membrane seal were
removed for testing and the membrane
was patched using similar material
and solvent welds. Several breaks
in the membrane were observed in
one section when it was held up to
the light. The maximum break was
one-quarter of an inch in length.
All breaks were produced by punc-
tures from the top down.
It was concluded from these
studies that even in areas of
heaviest truck traffic there were
only a few breaks in the membrane
and these breaks did not signifi-
cantly reduce the effectiveness of
the membrane to stop the movement
of water in the cover material.
Therefore, it was concluded that
the majority of the infiltration
was moving down to the membrane and
then moved horizontally across the
surface of the membrane discharging
at the edges of the refuse into a
land drainage swale.
During the first year severe
water erosion occurred on the steep
back slopes and in the swale between
the two landfills after heavy
spring rainfalls. This condition
was investigated and corrective
actions taken (Lutton, et al., 1981).
The assessment concluded that water
was washing away the fine grained
sand and that there was no apparent
damage to the membrane liner.
339
-------
MONITORING
The monitoring network in-
stalled in and around the landfill
during the course of this project
is depicted in Figure 2. It in-
cludes the following:
1. Twenty-eight monitor
wells in and around the
landfill to obtain ground-
water samples and ground-
water level measurements.
2. Three pairs of suction
lysimeters installed in
each of the two landfill
sections and upgradient
from the landfill to
obtain samples of soil
moisture and leachate
from two depths in the
unsaturated zone.
3. Four pan lysimeters,
including three in the
landfill and one in sand
and gravel, to check on
the rate of vertical
infiltration.
4. Two surface ponds down-
gradient from the landfill
to obtain samples for
analysis and additional
water-level measurements.
From February to July 1979,
water levels were measured weekly
and samples taken monthly in ex-
isting wells and lysimeters. From
July 1980 to May 1982, water level
measurements have been made and
samples taken on a bimonthly basis.
Temperature, pH and specific conduc-
tance were measured on-site in all
samples. Laboratory analyses were
made for nine other parameters:
total solids, alkalinity, chloride,
total organic carbon, iron, manga-
nese, sodium, copper, and lead.
The degree of water quality
improvement which has occurred was
illustrated in the comparison of
Figures 3, 4 and 5 as discussed by
Beck, et al (1981). In these
figures, values of specific conduc-
tance measured on three different
dates have been contoured to depict
the plume of contamination spread-
ing west-northwest from the landfill.
Background values of specific
conductance, measured at a well
south of the airport, range from
100 to 400 umhos/cm. Ground water
exhibiting a specific conductance
higher than 500 umhos/cm has been
assumed to be contaminated by
leachate. Between May 1979 and
May 1981, there was a 30 percent
reduction in the area of contami-
nated ground water, but the degree
of contamination beneath the eastern
section of the landfill, where
waste materials where partially
saturated with ground water, was
still very high. By November 1981,
however, the plume had receded even
further and conductivities measured
beneath the eastern half of the
landfill had decreased by 30 percent
from their May 1979 values. The
leachate plume, as depicted by the
500-umhos/cm contour line, had
moved closer to the reservoir,
which is the direction of ground-
water flow.
Since the preparation of these
figures the water quality has
stabilized in the vicinity of the
landfill. Recent results have
shown that the area of contamina-
tion has not changed significantly
during 1982.
CONCLUSION
By May 1982, water quality in
the area surrounding Windham Land-
fill had been monitored for
40 months, including 26 months
since top-sealing of the landfill
had been completed. During that
time, reductions in constituent
levels were observed in wells
downgradient from the landfill, and
the plume of contamination emanating
from the fill appeared to recede in
area and is now stabilizing.
ACKNOWLEDGMENTS
This paper addresses one phase
of a multiphase project being
conducted by SMC Martin under
U.S. EPA Contract No. 68-03-2519,
340
-------
PondS
PondS
KEY
CONTROL POINT
SPECIFIC CONDUCTANCE CONTOUR
(micromh os/cm)
Figure 4. Distribution of Specific Conductance, November 12, 1981
Pond 5
KEY
CONTROL POINT
SPECIFIC CONDUCTANCE CONTOUR
(micromhos/cm)
Figure 5. Distribution of Specific Conductance, May 19, 1981
341
-------
Donald E. Sanning, Project Officer.
Other phases involved the selection
of an abandoned waste disposal site
for study; the production of
Guidance Manual for Minimizing
Pollution from Waste Disposal Sites
(EPA-600/2-27-142), a comprehensive
discussion of remedial measures and
estimates of their costs; the
design and implementation of remed-
ial neutralization procedures at
the Windham, CT Landfill, and the
implementation of a monitoring
program to determine the effective-
ness of the procedures.
BIBLIOGRAPHY
Beck, Jr., William W., Alison L.
Dunn, and Grover H. Emrich, 1982.
Leachate Quality Improvements After
Top Sealing. MERL/SHWRD U.S. EPA
Hazardous Waste Symposium,
EPA-600/9-82-002, 464-474 p.
Emrich, Grover H. and William W.
Beck, Jr., 1979. Remedial Action
Alternatives for Municipal Solid
Waste Landfill Sites, MERL/SHWRD
U.S. EPA Hazardous Waste Symposium,
EPA-600/9-79-023, p. 324-342.
Emrich, Grover H., William W.
Beck, Jr., and Andrews L. Tolman,
1980. Top-sealing to Minimize
Leachate Generation. MERL/SHWRD
U.S. EPA Hazardous Waste Symposium,
EPA-600/9-80-010, p. 274-283.
Emrich, Grover H. and William W.
Beck, Jr., 1981. Top-sealing to
Minimize Leachate Generation -
Status Report, MERL/SHWRD U.S. EPA
Hazardous Waste Symposium,
EPA-600/9-81-002b, p. 291-297.
Griswold and Fuss, Inc., 1976.
Ground-Water Impact Study at the
Solid Waste Disposal Area, Town of
Windham, Connecticut. Report
prepared for the State of Connecti-
cut, Department of Environmental
Protection, Solid Waste Section,
54 p.
Lutton, R. J., V. H. Torrey III,
and J. Fowler, 1982, Case Study of
Repaiing Eroded Landfill Cover.
In: Proceedings of the Eighth
Annual Research Symposium: Land
Disposal of Hazardous Waste, Ft.
Mitchell, KY. U.S. Environmental
Protection Agency, EPA 600/9-82-002,
1982, Cincinnati, OH pp. 486-494.
Sanning, Donald E., 1981. Surface
Sealing to Minimize Leachate Gener-
ation at Uncontrolled Hazardous
Waste Sites. In: Proceedings of
Conference on Management of Uncon-
trolled Hazardous Waste Sites,
Washington, DC, October 28-30,
1981. Hazardous Materials Control
Research Institute, Silver Spring,
MD, pp. 201-205.
342
-------
EVALUATION OF REMEDIAL ACTION ALTERNATIVES-
DEMONSTRATION/APPLICATION OF GROUNDWATER
MODELING TECHNOLOGY
F. W. BOND
C. R. COLE
Battelle, Pacific Northwest Laboratories
Richland, Washington
D. SANNING
U.S. Environmental Protection Agency
Cincinnati, Ohio
INTRODUCTION
The LaBounty landfill was an
active chemical waste disposal site
from 1953 to 1977. During this
period, it is estimated that over
181,000 cubic meters (6.3 million
cubic feet) of chemical waste were
disposed in this 4.86 hectare
(12.0 acre) site located in the
flood plain of the Cedar River in
Charles City, Iowa. Compounds that
are known to be dumped in
significant quantities include
arsenic, orthonitroaniline (ONA),
1,1,2-Trichloroethane (TCE),
phenols, and nitrobenzene. Disposal
ceased in December 1977 following
the discovery that wastes from the
site were entering the river. Since
that discovery, the LaBounty
landfill has been under intensive
investigation.
One aspect of these
investigations was the ground-water
modeling effort described in this
paper. The overall objective of
this modeling effort was to
demonstrate the use of modeling
technology for evaluating the
effectiveness of existing and
proposed remedial action
alternatives at the site. The
primary emphasis of the project was
on technology demonstration which
included:
<» modeling of the ground-water
system and prediction of the
movement of contaminants,
« development of criteria to
determine which water sources
impacted by the site will
require remedial action and the
level of action necessary to
insure risks are at acceptable
levels, and
« the evaluation of the costs and
effectiveness of various
remedial action alternatives.
DESCRIPTION OF THE STUDY AREA
The LaBounty landfill site is
located on the floodplain of the
Cedar River just South of Charles
City, Iowa. Charles City is in
Floyd County which is situated in
northeastern Iowa. The site sits on
alluvial sand and gravel overlying a
bedrock formation (Cedar Valley
formation). Glacial tills have been
found along the west side of the
site above the Cedar Valley
formation. The Cedar Valley for-
mation can be divided into upper and
lower units separated by a shale
layer which acts as a confining
layer at the site (1,2).
343
-------
The hydrology of the LaBounty
site is somewhat complex in that
regional potential contours for the
upper Cedar Valley aquifer and the
shallow alluvial ground-water system
indicate that the river acts as the
major discharge area. In order to
accurately represent the site with a
model it is necessary to simulate
this vertical movement.
MODEL DEVELOPMENT
Two computer codes were used to
model the LaBounty landfill study
area: FE3DGW (3) and CFEST (4).
The FE3DGW code is a three-
dimensional, finite-element, ground-
water flow code that can be used to
predict ground water potentials and
water flow paths and travel times.
The CFEST (Coupled Flow, Energy, and
Solute Transport) code is an
extension of FE3DGW that predicts
the movement of contaminants with
the convective dispersion equation
for contaminant transport.
In order to better characterize
the ground-water system at the
LaBounty site, a two-stage modeling
approach was used. The first stage
consisted of developing a coarse
grid regional hydrologic flow model,
and the second stage consisted of
developing a flow and contaminant
transport model for the immediate
vicinity of the LaBounty site. The
purpose of the regional model was to
establish hydrologic boundary con-
ditions for the local model.
Regional Flow Model Implementation
The coarse grid regional model
covers an area of 6.3 km by 4.8 km
(approximately 3 miles by
4 miles). The regional boundaries
were chosen such that the equal
potential contours could be assumed
nearly vertical with depth, or at a
distance far enough away that errors
in the assigned potentials did not
influence local model boundary
conditions.
The finite element grid pattern
used to represent the large region
is shown in Figure 1. The local
model finite element grid boundaries
superimpose onto the regional
element pattern at the site.
Five different material types
(layers) were simulated in the
regional model. The lower layers,
upper and lower bedrock and shale,
are present over the entire model
region. The alluvial layer exists
along the Cedar River from a point
north of Charles City southeast to
the model boundary (Figure 1). A
combined fill and till layer was
simulated in the area of the
LaBounty Site. The structural top
and ground-water potential for the
layers were taken from data
contained in Munter (2). Hydraulic
conductivities (K) of the layers
were initially set at average values
obtained from several sources
(1,2,5,6,7) and the vertical to
horizontal permeability ratio
(Kz/Kxy) was set at 0.1.
Initially it was assumed that
10% of the total precipitation of
900 mm (35 in.) (7) recharged the
ground-water system. This was the
only source of recharge considered.
Local Flow Model Implementation
The local model covers the area
just in the immediate vicinity of
the LaBounty site (Figure 1) and is
685 m (2,250 ft) on a side. The
Cedar River forms the boundary to
the north, east, and south while the
western boundary was arbitrarily set
to the west of the fill material.
The finite element grid pattern
used to represent the small region
is also shown in Figure 2. The same
materials simulated in the large
region were simulated in the small
region except that the fill and till
were modeled separately.
The structural top and bottom
and potentiometric surfaces for the
shale and bedrock layers were
defined the same as in the regional
model. More detailed structure and
potential surfaces were prepared for
the local model alluvium, fill and
till layers with data obtained from
Munter (2).
344
-------
Local Model Area
Alluvial Layer ESSSSil
FIGURE l" Le3ion Region Model Finite Element Grid
•FIGURE 2' Local Ground-Water Flow Model Finite Element Grid
345
-------
Stage data for the Cedar River
were extrapolated to obtain held
river elevations along the model
boundary. The same permeabilities
and recharge rate used in the
regional model were initially used
in the local model.
Local Contaminant Transport Model
Implementation
The CFEST code was used to
simulate the movement of arsenic
from the landfill to the Cedar River
in the local model region and to
test the effectiveness of the
proposed remedial action
alternatives in controlling the
arsenic plume. Since the ground-
water flow portion of CFEST is
identical to FE3DGW, the final
calibrated FE3DGW data set was used
directly in CFEST to predict ground-
water flow. The additional inputs
required by the CFEST code to
simulate contaminant transport were
longitudinal (D,) and transverse
(Dj) dispersivity and initial
contaminant concentration levels.
Initially D, and DT were set at
30.0 m and 0.5 m, respectively.
Arsenic was chosen for use in the
calibration and remedial action
simulations because it has been
extensively monitored and its source
is fairly uniformly distributed over
the area of the fill material.
It is thought that arsenic
enters the ground-water flow system
from two sources:
1) precipitation infiltrating
through the fill material, and
2) leaching due to fill material
being below the water table. The
location of the saturated portion of
the fill was determined from the
literature (2,5) to be in two places
as shown on Figure 3. For modeling
purposes these two sources were
simulated by holding the down-
gradient surface nodes of the
saturated fill layer at an arsenic
concentration of 500 mg/1. Water
infiltrating through the cap was
also assigned an arsenic concentra-
tion of 500 mg/1.
MODEL CALIBRATION
The flow model calibration
process consisted of matching
observed potentials with model
predicted potentials for both the
regional and local models.
Calibration was accomplished by
first calibrating the regional model
and then testing the local model
with a consistent set of input data.
This alternating process was carried
out until a satisfactory agreement
between measured and model-predicted
heads was achieved in both models.
The final input parameters
obtained for the local model were
then used in the contaminant
transport model. These parameters
were not changed in the contaminant
transport calibration process.
Parameters related to contaminant
transport were adjusted to obtain a
reasonable match between field-
measured and model-predicted
contaminant concentrations.
Flow Model Calibration
The parameters adjusted in the
flow model calibration process were
the hydraulic conductivities (K),
the ratio of vertical to horizontal
hydraulic conductivities (KZ/KX),
the specific storage, and the rate
of recharge. After several cali-
bration runs, a good match was
achieved between measured and model-
predicted potentials for both the
regional and local models. The
final calibrated hydraulic
conductivities and storages used in
the flow models are listed in
Table 1. More detailed recharge
calculations estimated that about
14% of the precipitation (126 mm/yr)
reached the ground water.
Contaminant Transport Model
Calibration
One of the first changes made
was to change the node and element
grid to allow for a more accurate
description of the location of the
arsenic source and better represent
346
-------
FIGURE 3. Contaminant Transport Model Finite Element Grid.
Crosshatched area indicates area determined to
be below the water table.
TABLE 1. INPUT VALUES USED IN THE FINAL CALIBRATED FLOW MODELS
Layer
(Small )
1*
2
3
4
5
6
Material
Fill
Aluvium
Till
Upper Cedar Valley
Shale
Lower Cedar Valley
Specific
K* Storage
(m/day) (1/m)
0.5
10 and 2
5xlCT4
2.0 and 4.0
2.0 and 5xlO"3
2.0
0.0025
0.01
0.0025
io-4
io-4
io-4
KZ/K
Ratio
0.1
0.1
0.1
0.1
C.I
0.1
* Where two values of K are given, the first number listed
covered the largest percentage of the total area covered
by that material.
t The fill and till layers were combined into one layer in
the regional model with a K of 10 m/day.
347
-------
the spread of the resulting arsenic
plume. The finite element grid used
in all subsequent calibration runs
and in the remedial action cases is
shown in Figure 3.
Different values of longitu-
dinal and transverse dispersivity
were tested in order to match model -
predicted with measured arsenic
concentration levels. The results
of these test runs indicated that
longitudinal and transverse
dispersivities of 5.0 m and 0.5 m,
respectively, gave the best
results. The initial concentration
level for arsenic of 500 mg/1 was
not changed in the calibration
process.
The last step in the
calibration process was to determine
the location of the arsenic source
(5). As stated above, it was
suspected that arsenic enters the
flow system from either infiltration
alone, or a combination of
infiltration and ground-water
leaching where fill material is
thought to be below the water
table. Both cases were tested by
comparing model-predicted arsenic
concentrations and river
loading quantities against field-
measured levels.
After running both cases for
6000 days it became apparent that
the case of infiltration alone would
never achieve the arsenic levels
that have been measured at the
site. The results indicate that
part of the fill material must
indeed be below the water table and
that the majority of the contaminant
transport is originating in the
saturated ground-water system and
not in the unsaturated flow system
as a result of infiltration.
Based on these findings the
final calibrated model, hereafter
referred to as the base case,
included both sources of arsenic,
infiltration and groundwater
leaching. A contour map of the base
case arsenic concentrations after
6000 days (near equilibrium) is
shown in Figure 4. This final
calibrated contaminant transport
model base case became the initial
condition from which all the
remedial action cases were
initiated.
The final calibrated transport
model predicted arsenic mass outflow
rate to the river was 25.2 kg/day
(55.4 Ib/day). The average mass
outflow rate as measured by the
monitoring system at the site is
22.7 kg/day (50.0 Ib/day).
ASSESSMENT OF REMEDIAL ACTION
ALTERNATIVES
Seven remedial action
alternatives were identified for
potential application to the
LaBounty landfill site and were
modeled with the CFEST model; clay
cap, upgradient cutoff wall,
downgradient cutoff wall, limited
excavation, limited bottom lining,
stabilization, and pump and treat.
The clay cap alternative has already
been implemented at the site.
Contaminant transport model
parameters were adjusted to simulate
arsenic migration to the river for
each alternative. The alternatives
were assessed by first quantifying
their effectiveness on reducing
contaminant concentrations and Cedar
River loading rates. These results
were compared with the results for
the clay cap alternative to obtain a
measure of effectiveness.
The next step was to estimate
the capital and annual operating and
maintenance costs associated with
each alternative. Given the
effectiveness and cost, the final
step was to make a cost-effective-
ness comparison of the alternatives.
The base case, as discussed
earlier, was run for 6000 days
(16.4 years). Each remedial action
case was run for an additional
4500 days (12.3 years) beginning at
the end of the 6000-day period.
Clay Cap
The clay cap alternative was
modeled by reducing the infiltration
rate by a factor of 100 in the area
of the cap. The magnitude of the
decrease was estimated using a
348
-------
7. .... CONCENTRATION
.75.00— CONTOUR
Scale(m)
200
FIGURE 4. Arsenic Concentrations (mg/1) at the Top of the Alluvium
for the Base Case of Combined Infiltration and Disposal
Below the Water Table
simple compartmental model and a
permeability for clay of 3 x
10 m/day. The rate of arsenic
injection from infiltration was also
reduced by a factor of 100 to
0.08 kg/day (0.04 lb/day).
The results of this run
indicated a continued but slight
rise in predicted arsenic loading of
the Cedar River 27.5 kg/day
(60.4 lb/day) after 4500 days, again
illustrating that the portion of the
waste disposed below the water table
is the major source of the
contamination problem at LaBounty.
The clay cap was assumed to be in
place for simulation of the
remaining six remedial action
alternatives.
Upgradient Cutoff Wall
The objective of the upgradient
cutoff wall is to reroute incoming
ground water such that it bypasses
waste contaminated soil. This can
only work if the wall extends
downward to an impermeable layer or
incorporates some means of transport
to carry off water mounded behind
the wall. Since the dolomite base
rock at Charles City is
transmissive, the latter would be
required. For the purposes of this
evaluation, it was assumed that a
subsurface drain is constructed
upgradient from the cutoff wall to
carry uncontaminated water away from
the site.
The effectiveness of the
upgradient cutoff wall was simulated
by introducing a line of elements
with low permeability upgradient of
the landfill (Figure 5). The
upstream set of surface nodes along
the low permeability cutoff wall
were treated as held potential
boundary nodes to simulate the
349
-------
^-Direction of Grwmdwoter Row
"•Sluny Trench
Subwrfac* Drain
TicudiMnt System
FIGURE 5. Upgradient Cutoff Wall Remedial Action Alternative
upgradient drain. The flux to these
held nodes was calculated to
determine the required capacity of
the subsurface drain associated with
this alternative.
The results of this analysis
indicate that new equilibrium is
reached in about 5 years with Cedar
River loading rate reduced to
14.9 kg/day (32 Ibs/day) after
4500 days.
Downgradient Cutoff Wall
The downgradient cutoff wall is
intended to intercept contaminated
ground water and then route it via
drains for treatment. The
effectiveness of the downgradient
cutoff wall was simulated in the
model with a line of impermeable
elements downgradient from the site
(Figure 6). The upstream set of
surface nodes formed the simulated
drain and were treated as a held
potential outflow boundary. The
flux to these nodes and arsenic
concentration levels were predicted
to determine the capacity of the
subsurface drain and the treatment
facility.
Results of this simulation
indicate that new equilibrium river
loading rates were achieved after
approximately 3 years, 2 years
faster than for the upgradient
case. The new Cedar River loading
rate predicted at the end of
4500 days was 7.66 kg/day
(16.9 Ib/day).
Limited Excavation
The intent of the limited
excavation approach is to remove
those portions of the waste below
the water table and to backfill with
clean fill.
Simulation of the limited
excavation case with the model
involved removing the source term
350
-------
NORTH
OOWNGRADIENT CUTOFF WALL
Direction of Groundwoter Flo*
Suny Trendi
Subsurface Drain
Treatment System
SCflLE CMTTERSJ
200
FIGURE 6. Downgradient Cutoff Wall Remedial Action Alternative
associated with wastes disposed
below the water table and changing
the permeability in the area where
the fill material was removed to
that of the alluvium. Figure 3
indicates the approximate size of
the areas below the water table.
Results of this simulation
indicate that this alternative
provides a more delayed response in
terms of river loading rates but
that with time, much lower arrival
rates could possibly be achieved.
Model results indicated that after
4500 days predicted Cedar River
loading has decreased to 8.0 kg/day
(17.6 Ib/day).
Limited Bottom Lining and
Stabilization
These two approaches are very
similar in that they are both
isolation techniques, either through
the use of a liner or fixation
method. Only those portions of the
waste below the water table would be
lined or stabilized. Presumably,
this would reduce water flux and
hence minimize further contributions
from these portions of the landfill
to the overall contaminant plume.
Stabilization presents a challenge
to current technology since in situ
fixation has not been perfected.
Both of these alternatives were
modeled in an identical manner.
The total source mass of waste
remained the same as the base case
but the leach rate was reduced.
This was simulated in the model by
changing the permeability in the
saturatedj-fill area to that of a
clay (10~b m/day).
The results for these
alternatives are very similar to
those of the limited excavation
alternative. As with limited exca-
vation, the reduction in Cedar River
loading is more gradual but it has
the possibility of eventually
351
-------
reaching much lower river loading
rates. After 4500 days a Cedar
River loading of 8.14 kg/day
(17.9 Ib/day) was predicted.
Pump and Treat
In this isolation method,
removal is accomplished through use
of a series of withdrawal wells
(Figure 7). The water is
subsequently treated to eliminate
contaminants and either discharged
back to the river or reinjected to
assist in directing the
contaminanted plume toward the
wells.
Pumping rates were estimated at
1806 m /day (0.477 mgd). As with
the downgradient cutoff wall,
predicted concentrations for
treatment during the first year is
higher (71 ppm) but it soon levels
off at about 34 ppm. Predicted
Cedar River loading rates have
pretty much equilibrated after
5 years and were predicted to reach
10.12 kg/day (22.3 Ib/day) after
4500 days.
The capital cost and annual
operating cost for all the
alternatives are given in Table 2.
It is important to note that for all
cases these costs should be
considered only as rough estimates,
since they are based on preliminary
conceptual designs for each
alternative and they do not reflect
local cost differences.
Discussion of Results
A comparison of arsenic mass
outflow rates to the river for the
base case and the seven remedial
action alternatives as predicted by
the model is shown in Figure 8. The
contaminant transport model very
clearly indicated that the major
source of ground-water contamination
arises from that portion of the
arsenical sludge buried below the
water table.
Model results indicate that
only the clay cap and the upgradient
cutoff wall would probably fail to
bring river concentrations (under
low flow conditions) below drinking
water standards. The upgradient
cutoff wall, however, could probably
be significantly improved through
optimization studies with the
transport model. Optimization
studies were not performed for any
of the alternatives.
The remaining five alternatives
all lowered river concentrations
below drinking water standards for
Cedar River low flow conditions;
however, the selection of the best
alternative is still very much sub-
jective. The downgradient cutoff
wall and pump and treat alternatives
show a quick response in reducing
contamination levels but they
approach a higher Cedar River
loading limit. On the other hand,
the bottom lining, excavation, and
stabilization alternatives do not
respond as quickly but have the
potential to achieve much lower
Cedar River loading rates and
aquifer contamination levels. Cost
effectiveness of the various reme-
diation alternatives is an important
part of any decision making pro-
cess. As indicated in Table 2,
limited excavation (on site
reburial) and bottom lining are
the least cost alternatives
(excluding the upgradient cutoff
wall which failed to meet the level
of contamination criteria). When
comparing costs per unit reduction
in predicted Cedar River arsenic
concentrations (low river flow
conditions) or the cost per unit
reduction after 4500 days in arsenic
contaminant mass in the saturated
aquifer system, the limited
excavation (on site reburial) is
most cost-effective, followed by
limited bottom lining and pump and
treat. In both cases, stabilization
is the least cost effective,
followed by limited excavation
(offsite reburial) and the
downgradient cutoff wall.
CONCLUSIONS
A hydrologic flow and transport
model of the Charles City LaBounty
landfill was developed, calibrated
and used to evaluate seven remedial
action alternatives for the LaBounty
site. This modeling study has
352
-------
FIGURE 7. Pump and Treat Remedial Action Alternative
353
-------
TABLE 2. ESTIMATED COST FOR EACH OF THE SEVEN REMEDIATION ALTERNATIVES
Remediation Alternative
Capital Cost
(dollars)
Annual Operating Cost
(1982 dollars)
Upgradient cutoff wall* 835,000
and treatment
Limited excavation** 1,000,000
(on site reburial)
Limited bottom lining 1,348,000
Pump and treatment 1,431,000
Downgradient cutoff 2,569,000
wall and treatment
Limited excavation 16,000,000
(off site reburial)
Stabilization 16,787,000
99,000
35,000
67,000
168,000
114,000
35,000
35,000
* If optimized, costs for this alternative are likely to be
higher since diversion of more water will most likely be required.
** Assumes that waste removed from below the water table would
be reburied on site above the water table and capped with clay.
reaffirmed the importance of
considering the regional hydrologic
system when modeling the hydrology
and transport for a more local
system like LaBounty, and
illustrates the value of ground-
water modeling in developing an
understanding of a complex flow and
transport system. In the process of
modeling the LaBounty site, it was
demonstrated that boundary
conditions for the local model
cannot be arbitrarily set and still
obtain reasonable estimates for
travel time and river discharge
rates. The contaminant transport
model developed for the LaBounty
site, while not a perfect indicator
of observed contamination movement,
is certainly an adequate indicator
for the purpose of general
understanding and for the evaluation
of the effectiveness of various
remedial action alternatives.
354
-------
COMPfifllSON OF EFFECTIVNESS OF REMEOIflL flCTlON flLTEWWTIVES flT LflBOUNTY
60
t'
Limited Bottom
Lining/Stabilization
I960
1990
HrPOTHETICflL TIME (YEflflS)
FIGURE 8- Predicted Effective of the Various Remediation
Alternatives in Reducing the Arsenic Loading
of the Cedar River.
REFERENCES
1. Eugene A. Hickok and Associates. "Hydrologic Investigation, LaBounty
Site, Salsbury Laboratories, Charles City, Iowa." For: Department of
Environmental Quality, Des Moines, Iowa, 1981.
2. Munter, J. A. Evaluation of the Extent of Hazardous Waste Contamination
in the Charles City Area. Iowa Geologic Survey, Iowa City, Iowa, 1980.
3. Gupta, S. K., C. R. Cole and F. W. Bond. Methodology for Release
Consequence Analysis - Part III, Finite-Element Three Dimensional Ground-
Water (FE3DGW) Flow Model. Formulation Program Listings and User's
Manual, PNL-2939, Pacific Northwest Laboratory, Richland, Washington,
1979.
4. Gupta, S. K. , C. R. Cole, C. T. Kincaid and G. E. Kaszeta. Description
and Application of the FE3DGW and CFEST Three-Dimensional Finite-Element
Models, LBL-11566, Lawrence Berkeley Laboratory, University of California,
Berkeley, California, 1980.
5. Eugene A. Hickok and Associates. "Soil Characteristics, LaBounty Site,
Salsbury Laboratories, Charles City, Iowa." For: Department of
Environmental Quality, Des Moines, Iowa. 1977.
6. Layne-western Company, Inc. Monitoring well System Hydrologic Evaluation,
LaBounty Landfill Site. Charles City, Iowa, 1980.
7. NOAA (National Oceanic and Atmospheric Administration). Loca 1
Climatological Data. Monthly Summary for Charles City and Waterloo,
Iowa. Department of Commerce, 1979, 1981.
355
-------
SLURRY TRENCH CONSTRUCTION OF
POLLUTION MIGRATION CUT-OFF WALLS
Philip A. Spooner
Roger S. Wetzel
JRB Associates
8400 Westpark Drive
McLean, Virginia 22102
Dr. Walter E. Grube, Jr.
U.S. Environmental Protection Agency
Municipal Environmental Research Laboratory
Solid and Hazardous Waste Research Division
26 W. St. Glair Street
Cincinnati, Ohio 45268
ABSTRACT
A technical handbook evaluating the use of slurry trench construction techniques for
controlling pollution migration has recently been completed. This work, funded by the
U.S. Environmental Protection Agency's Office of Research and Development, is intended to
provide guidance to evaluators of proposed remedial actions that include subsurface
barriers emplaced by slurry trenching. This paper updates the overview presented before
the National Conference on Management of Uncontrolled Hazardous Waste Sites, in
Washington, D.C., in November 1982. The technical aspects of slurry trench construction
as it is applied to hazardous waste site groundwater control are discussed.
The materials used for trench excavation are described, along with their functions. Also
described are the materials used to backfill the trench, forming the final cut-off wall.
Emphasis is placed on the two most common backfill materials, soil-bentonite and
cement-bentonite; their characteristics, and what is known of their compatibility with
hazardous wastes. The way in which various cut-off wall configurations can be applied to
a site are discussed and illustrated. Construction procedures for a typical
soil-bentonite slurry wall are outlined, and examples of completed wall costs given.
INTRODUCTION
Over the last two decades, cut-off Conference on Management of Uncontrolled
walls emplaced by slurry trenching have Hazardous Waste sites in Washington, D.C.,
been installed at many solid and hazardous in November 1982. Following is an update
waste facilities. An overview of a draft of data interpretation and project ac—
Technical Handbook on construction of complishements since the November report.
cut-off walls using slurry trench tech- This Handbook was prepared for EPA's
niques was presented at the National Office of Research and Development to
356
-------
compile the scientific and technical
parameters which need to be evaluated when
considering this type of construction for
isolation of hazardous chemicals in near-
surface groundwater regimes. It contains
sections which present the theory of
slurry and backfill function, applica-
bility of cut-off walls to curb ground-
water migration, site investigations
needed preliminary to construction, actual
design and construction parameters, wall
performance monitoring and maintenance,
and presentation of major cost elements.
Slurry trenching is a method by which
a continuous trench is excavated (by
backhoe or clam-shell grab) under a slurry
of bentonite and water. This slurry, or
more correctly, colloidal suspension,
supports the trench walls and allows
excavation to great depths with no other
means of support. Once a trench has been
excavated under slurry to the required
depth and length, it is back-filled to
form a continuous, nearly impermeable
barrier, or cut-off wall, to groundwater
flow. In some cases, the slurry is a
mixture of water, cement, and bentonite
(CB backfill) which hardens in place to
form the final barrier. In other
instances, this trenching technique is
used to place soil-bentonite (SB) cut-off
walls, where the backfill is composed of
the excavated soil with small amounts (1
to 470) of bentonite added. In some
instances, borrowed soil materials, such
as additional fine grained materials, or
gravel must be added to meet the require-
ments of lower permeability of higher
strength. Another variation of this tech-
nique, seldom used for pollution migration
control, is the use of pre-cast or
cast-in-place concrete diaphragm walls,
which are installed when both pollution
migration cut-off and structural support
are required. Because they are seldom
used, diaphragm walls are not further
discussed here. Table 1 lists some
typical slurry walls installed for pol-
lution control.
Slurry Materials and Function
The material used in the trenching
slurry is a commercially available product
called bentonite, sodium bentonite, or
Wyoming bentonite. Bentonite is con-
sidered by many to be a clay but is in
fact, a rock, composed mostly of the clay
mineral montmorillonite, with smaller
amounts of other clays and various me-
tallic oxides. Montmorillonite crystals
contain negative charges which attract and
adsorb cations. In most bentonites, the
montmorillonite is saturated with calcium
ions: however, the products which are
used for slurries and drilling muds are
the Wyoming bentonites which bentonite is
generally highly colloidal and plastic,
and has the ability to swell many times
its volume and to form thixotropic gels in
water
Many commercially available bentonite
are modified by the addition of polymers,
peptizers or other chemicals. This
practice has been adopted in order to
increase the amount of mud (slurry)
yielded by lower grade bentonite deposits,
and, in the case of polymer extended
products, to increase piping resistance in
liner applications.
The slurry used in trenching is
composed of clean water mixed with from 5%
to 7% premium grade bentonite by weight 2.
Make-up water should be low in dissolved
salts and hardness, and neutral or
slightly basic in pH. When introduced into
the trench, the slurry should be fully
hydrated with an "apparent viscosity" of
from 15 to 20 centipoise and a density of
approximately 65 pounds per cubic foot,
(pcf)3'1". Where field practice uses the
Marsh cone funnel to estimate slurry
viscosity, a reading of at least 40
seconds is necessary to assure adequate
filter cake formation and sufficient
density to support the trench 2.
As stated earlier, the primary role
of the slurry is to keep the trench open
during excavation. A number of mechanisms
have been suggested for trench support,
including hydrostatic force exerted by the
slurry, formation of a filter cake on the
trench sides, and compaction of the earth
on either side of the trench during
excavation5. Of these three, the first two
are the most important, and are closely
related5'6.
In order for the force exerted by the
weight of the slurry to support the
trench, the force must be exerted on the
soil grains and not just on the soil pore
fluids. The formation of a low perme-
ability filter cake on the trench sides,
caused by slurry and water loss from the
trench into the ground, eventually seals
357
-------
LOCATION
Indiana
Illinois
Louisiana
Pennsylvania
New York
New Hampshire
Wisconsin
New Jersey
Saskatchewan
Florida
Ohio
Florida
WALL
COMPOSITION
Soil-Bentonite
Soil-Bentonite
Soil-Bentonite
Cement-Ben tonite
Cement-Ben tonite
Soil-Bentonite
Soil-Bentonite
Soil-Bentonite
Soil-Bentonite
Soil-Bentonite
Soil-Bentonite &
Cement- Ben tonite
Cement— Bentoni te
SIZE*
(ft.2 x 1000)
500
13.5
30
1
77.3
204
98.4
10
53.4
58
25
10
APPROXIMATE
COST
($/ft.2)
3
4
2.5
2
5.5
5
2.8
6
4
2.6
10
4
POLLUTANT
CONTAINED
Inorganics
Ash Disposal
Methane Gas
Ethylene
Bichloride
PCB
Oils
Jet Fuel
Mixed Organics
Solid Waste
Landfill Leachate
Phenols
Uranium Mill
Tailings
Methane Gas
Landfill Leachate
Acid Mine
Drainage
Radioactive
Kerosene
STATUS
On-going
Completed
Completed
Completed
Completed
On-going
Completed
Completed
Completed
Completed
Completed
Completed
"These walls range from 2 to 4 feet in width and 15 to 100 feet in depth.
TABLE 1. Example Slurry Walls Installed For Pollution Control
-------
off the soil pores, allowing the hydro-
static force of the slurry to be trans-
ferred more directly to the sides. Filter
cake formation is a function of clay type,
clay concentration in the slurry, and
formation time and pressure. Perme-
abilities of filter cakes qhave been
measured as low as 2.3 x 10 cm/sec2'6.
Additional trench support can be expected
by the plastering effect of the filter
cake holding the soil grains together, and
by gelation of the slurry in the soil
pores into which it infiltrates 5 . The
contribution of the filter cake to the
final cut—off wall is examined later in
this paper.
Backfill Materials and Function
There are three major types of slurry
trench backfill: soil-bentonite (SB),
cement-bentonite (CB), and concrete. The
choice of backfill is dependent on spe-
cific site characteristics and the re-
quirements of the completed cut-off wall.
SB and CB cut-off walls are discussed
below.
Soil-Bentonite Cut-Off Walls
Soil-bentonite (SB) cut-off walls are
often backfilled using the soil materials
excavated from the trench. These soils, or
soils with more favorable properties
borrowed from another source, are mixed
with small amounts of slurry from the
trench until they become a homogenous
paste. The recommended consistency for
the backfill is a mixture that will stand
on a 5 to 10:1 slope (as measured by slump
cone) with a water content of 25% to 35%.
The minimum amount of bentonite in the
backfill mixture will range from 0.5% ot
2% depending on the initial water conent
of the excavated soil2.
The permeability of the backfill is
very much dependent on the gradation of
the soil used, as well as on the bentonite
content and intial water content of the
backfill. The lowest permeabilities are
obtained when the backfill is well graded
and contains an appreciable amount (i.e.,
10-207, or more) of fines (particles
passing a 200 mesh sieve) an especially
plastic fines2. Figure 1 shows the extent
to which fines and plastic fines can
affect backfill permeability. Figure 2
shows the contribution of added bentonite
to the permeability of the backfill. In
cases where the excavated soils are not
well suited for backfill, borrowed soils
may have to be added to meet the design
criteria. For example, if low com-
pressibility is called for in the design,
and the trench materials lack coarser
particles, additional sand or gravel may
be needed. If low permeability is of
greater concern than low compressibility,
addition of borrowed coarser materials
would be unnecessary. If fines are
lacking in the trench soils, and low
permeability backfill ( 1x10 cm/sec) is
called for, borrowed material could be
required. Bentonite can make up only a
portion of the plastic fines in the
backfill mixture due to its high com-
pressibility. In brief, gradation of the
backfill is a trade off between perme-
ability and compressibility as illustrated
in Figures 1 and 3.
The permeability of the final wall is
a function of both the filter cake
permeability and the backfill perme-
ability. Because the backfill is less
permeable than the surrounding soil it is
possible that the downgradient filter cake
is forced into the soil matrix under the
hydraulic gradient across the wall. This
gradient would act to force the upgradient
filter cake into the backfill, but given
proper backfill gradations (i.e., suf-
ficient fines) it would remain intact 2 .
Figure 4 illustrates the relative con-
tribution of the backfill and one filter
cake to the final wall permeability. This
figure also shows that the contribution of
the filter cake is greater for more
permeable backfill and less for more
impermeable backfill. It also illustrates
that, even with very permeable backfill,
the maximum permeability of a SB cut-off
wall should be on the order of 1x10
cm/sec due to the contribution of one
filter cake alone.
Cement-Bentonite Cut Off Walls
Cement-bentonite (CB) cut-off walls
are of two major types: self-hardening
slurries, which both support the trench
and are allowed to harden in-place; and
replacement slurries which are introduced
into the trench following excavation using
a bentonite-water slurry for trench
support. Replacement slurries are used in
very deep excavations c- with time con-
suming rock key-ins where a se If-hardeni r"r\
slurry would thicken or set befci ~
359
-------
80
70
60
50
40
30
20
10
0
, Plastic Fines
Non-Plastic or Low
Plasticity Fines
0 NT8
1
10-
10-
10-
10-
10"'
10-
10-
10-
Well Graded
Coarse Gradations
(30-70% +20 Sieve)
w/10 to 25% NP Fines
Clayey Silty Sand
w/30 to 50% Fines
Poorly Graded
Silty Sand w/
30 to 50% NP Fines
0 12345
% Bentonite by Dry Weight of SB Backfill
Figure 2. Relationship Between Permeability
and Quantity of Bentonite Added to SB Backfill
(After D'Appolonia and Ryan 1979)
360
-------
80
70
60 -
50
40 -
X
20
• Plastic Fines
O Non-Plastic Fines
I-D Compression
Stress Increment
0 5 to 2 0 kg/cm2
• Plastic Fines
O Non-Plastic Fines
Isotropic Compression
Stress Increment
05 to 02 kg/cm2
04 .06 08 10
Compression Ratio, Cc/1 + CQ
Figure 3. Compressibility of SB Backfill
Related to Fines Content of Mix
(After D'Appolonia. 1980)
k . Wall Permeability
ks . Cake Permeability
ku = Backfill Permeability
tb = Backfill Thickness
tc - Cake Thickness
tb
10-' 10~6 10-5 10-
Backfill Permeability, k,,, cm/sec
Figure 4. Theoretical Relationship Between Wall Permeability
and Permeability of the Filter Cake and Backfill
(After D'Appolonia and Ryan 19791
361
-------
excavation was complete. The self-
hardening slurries are used in relatively
shallow trenches (< 70') where their rapid
set should not interfere with excavation.
Typically, a self-hardening CB slurry
contains (by weight) from 4 to 7%
bentonite, from 8 to 25% type I Portland
cement, and from 68 to 8870 water 7.
If too much bentonite and cement are
used, the slurry may be too thick to work
with. If too little cement is used, no
set will be achieved; and if too little
(or poor quality) bentonite is used, the
cement will settle out of the slurry,
causing excessive water bleeding. Figure
5 illustrates typical CB slurry com-
positions. The actual composition of the
slurry is determined during the wall's
design.
At the time the CB slurry is intro-
duced into the trench, it typically has a
density of about 68 PCF and a Marsh
viscosity ranging from 30 to 50 seconds 7'8.
After excavation, but before set, it
can have densities of around 78 pcf and
will be too viscose to pass a Marsh fun-
nel7.
Filter cakes formed by CB slurries do
not have the same properties as those of
bentonite slurries, yet still serve to
support the excavation. A CB filtrate
cake can have a permeability orders of
magnitude higher than bentonite (i.e.,
1-4x10 cm/sec), and filter losses around
20 times as large 7 . This is due to the
calcium in the cement causing flocculation
and aggregation of the bentonite. Ex-
cessive filtrate loss can cause problems
-..'ith the excavation by thickening the
slurry (by dewatering through filtrate
loss) and by lowering the level of slurry
in the trench.
Set times for the slurry can be
controlled to some extent by controlling
the amount of cement, and through addition
of additives. Rapid set times, however,
are more easily achieved than retarded set
times. The additives used most suc-
cessfully are ground blast furnace slag
and fly ash, which replace a portion of
the cement in the mixture. Slag has been
shown to reduce cut-off permeability,
bleed vat>T, slu-ry shear strength, and
set time. Fly ash can aid in resistance
to chemical attack and can increase set
time and bleed resistance7- Another
method used to reduce bleed water from the
curing slurry, is to ensure that the
slurry is mixed in a high shear mixer
starting with a fully hydrated bentonite
slurry 7.
Self-hardening (CB) slurries can
offer a number of advantages in wall
construction. This method is well suited
for situations where the trenched material
is unsuited for use as backfill or where
there is insufficient area for SB backfill
mixing. Also, because the slurry hardens
in place, the wall can be built in
sections or panels (by excavating a tie-in
between completed sections). This method
allows for more flexible construction
schedules, and can be used in more extreme
topography 8 . Some disadvantages of this
methods include the need for a greater
amount of more expensive slurry, a rela-
tively higher permeability than SB walls,
less resistance to chemical attack, and in
contaminated environments, the need to
dispose of contaminated trench spoils 8 .
COMPATIBILITY
The presence of organic or inorganic
compounds in the groundwater can have a
detrimental effect on the bentonite slurry
used during wall construction as well as
the ability of the finished wall to
contain pollutants. These chemicals can
affect the physical/chemical properties of
the bentonite and the backfill material,
and can lead to failure of the wall either
during construction or during its oper-
ational lifetime. Thus, before a slurry
wall is conidered as an appropriate
remedial response, the effects of the
leachate on the bentonite slurry and on
the finshed wall must be determined
through testing. This testing procedure
will provide the information that is
necessary to select the proper grade of
bentonite and the type of backfill materi-
al that should be used in construction of
the slurry wal1.
The following outlines the types of
effects that chemical have on bentonite
and backfill material, and the laboratory
tests that should be conducted in order to
establish the potential effectiveness of
the slurry wall.
362
-------
WATER
Non-Setting
Slurries
BENTONITE
Bleeding
Slurries
% Water
•CEMENT
Figure 5. Cement-Bentonite
Slurry Compositions
(After Jeffens, 1981)
363
-------
Chemicals can affect the physical and
chemical properties of the bentonite and
backfill material, leading to:
o Flocculation of the slurry
o Reduction of the bentonite's
swelling capacity
o Structural damage to the bentonite
or other backfill material.
In general, flocculation of the
bentonite slurry can be caused by chem-
icals, such as electrolytes and metals,
that act to reduce the natural repulsive
forces between the hydrated clay parti-
cles. From a. practical standpoint, the
exact mechanisms are unimportant.
However, if the slurry in the trench
flocculates, due to chemicals in the
make-up water or the soil, it can lose its
ability to support the trench, leading to
collapse 9. This problem can usually be
avoided by adequate site investigation and
testing.
Numerous organic and inorganic com-
pounds can, through a variety of mecha-
nisms, cause bentonite clay particles to
shrink or swell. All of these mechanisms
affect the quantity of water contained
within the interspacial layers of the clay
structure. Many chemicals can reduce the
double layer of partially bound water
surrounding the hydrated bentonite, thus
reducing the effective size of the clay
particles10'11. This effect can increase
the permeability of the backfill. Proper
testing and backfill composition however,
can prevent reduced swelling from having a
detrimental effect on the final cut-off by
measuring the permeability increase caused
by the reduced swell and designing the
backfill to account for it 16.
Strong organic and inorganic acids
and bases can dissolve or alter the
bentonite or soil in the backfill materi-
al, leading to large permeability in-
creases 10 . Alumina and silica, two of the
major components of bentonite, are readily
dissolved by strong acids or bases,
respectively. For example, when four
types of clay were exposed to acetic acid,
significant soil piping was reported to be
caused by dissolution of soil components
by acid ". This led to a significant
increase in permeability within the four
clay types. Proposed backfill materials
should be tested to determine whether
their permeability is adversely affected
by contact with leachates or groundwater
from the proposed site. Materials suffer-
ing from a predefined significant adverse
affect should be eliminated from consider-
ation at the site.
Compatibility testing is done using
actual site leachate or groundwater and
the proposed slurry and backfill. The
tests are not all standardized or applied
throughout the industry. In general, the
tests include:
o Apparent slurry viscosity
o Filter press test
o Permeability.
Viscosity testing indicates the
ability or inability of the slurry to form
a thixotropic gel in the presence of the
contaminants. The filter press test gives
a rough indication of the ability of the
slurry to form a filter cake in a
contaminanted trench. Together these
tests can tell the design engineer whether
problems with trench support might occur.
Permeameter tests of the proposed
backfill mixture, using the site leachate
as the permeant, are essential in de-
termining the expected cut-off ef-
fectiveness. Following wall installation
permeability testing of the actual back-
fill material can be used to see if the
design permeability was achieved. Care
must be taken in collection of the
leachate to ensure that the samples used
in the tests are representative of the
site conditions.
SLURRY WALL APPLICATIONS
Slurry walls have been, and are
being, applied to a variety of waste site
situations. The composition, dimensions,
and placement of the walls are, of course,
designed on a site specific basis. As
noted above, wall composition is dependent
on a number of factors including site
lay-out, site chemistry and strength
requirements. The following is a brief
discussion of the factors affecting wall
dimensions and placement with respect to
the sources of pollution.
The dimensions of a slurry wall are
essential design parameters, especially
for depth and length. Wall width is
largely a function of the excavation
364
-------
equipment used, and generally ranges from
2 to 5 ft 1D. The depth of the wall is
most dependent on the behavior of the
contaminants and the site stratigraphy.
In most situations, it is necessary to key
the wall into a low permeability under-
lying strata (aquiclude) such as bedrock
or a clay layer. In some cases, such as
with methane, or immiscible fluids less
dense than the groundwater, such as some
petroleum fractions, (so-called "floating
contaminants"), a "handing wall" can be
used. Hanging walls are not keyed into an
aquiclude but are placed only deep enough
to intercept the top several feet of the
water table, thereby allowing capture of
floating contaminants 13 .
The length and shape configurations
of a slurry wall are largely dependent on
the location of the pollution source and
any resultant contamination plume, as well
as the intended role of the wall. In
general, a wall can be placed surrounding
a source, or upgradient or downgradient
from the source with respect to ground-
water flow.
Circumferential wall placement is
employed in situations where more complete
containment is desired. As shown in
Figure 6, the wall surrounds the source
and prevents migraiton into and out of the
site. Care must be taken to prevent wall
destruction by thorough compatibility
testing and/or by preventing contaminant-
wal1 contact .
Upgradient placement can divert
groundwater flow around a pollution
source, thereby protecting that ground-
water from contamination and reducing the
amount and rate of leachate generation.
Care must be taken in the design to ensure
that groundwater does not flow around the
wall ends or rise to overtop the wall.
Where upgradient drains are included to
divert groundwater flow around a contami-
nated area, the cut-off increases
assurance that the hazardous waste site is
isolated from adversely affecting water
quality. Figure 7 illustrates upgradient
placement and barrier drain combination.
Downgradient placement, shown in
Figure 8, can be used in situations where
the amount of flow into a source is very
low. In this practice, the slurry wall
serves to capture the contamination plume
so that it can be collected (by wells or
drains) and treated. Again care must be
taken to ensure that contaminated water
does not have a destructive effect on the
wall (or is kept from contact with it),
and that the contamination does not
overtop the wall.
It should be noted that slurry walls
are rarely the sole type of engineered
remedial measures. Most often, additional
measures such as extraction wells,
leachate collectors, and surface seals are
used in conjuction with slurry walls as
part of the total remedial effort.
SITE CHARACTERIZATION
To design and install a successful
slurry wall, extensive site character-
ization, especially with respect to sub-
surface conditions, is necessary. The
following is a summary of the major site
characteristics relevant to slurry walls.
Site investigations for construction
projects are frequently accomplished in
two phases, a preliminary phase, and a
design phase. A preliminary investigation
will reveal many of the physical con-
straints that would affect slurry
trenching operations. Among the most
important considerations are:
o Topographic restrictions
o Buried or overhead obstructions
o Equipment access
o Sufficient work area.
These factors, and their interrelation-
ships must be identified and considered
very early in the design phase. Ad-
ditional, general information will be
collected on the subsurface conditions,
which will allow the design engineer to
select the most likely construction
method(s).
The in—depth or design phase inves-
tigation will be more involved and more
narrowly focused, centering on developing
design parameters. For slurry walls, the
principal focus will be on complete
geologic and geohydrologic and, in some
cases, geophysical characterization of the
site. This is accomplished through a
series of soil borings, initially general
in nature and becoming more detailed as
variability and anomalies are assessed.
Among the more important parameters
365
-------
Figure 6. Cut-away Cross-section of Circumferential Wall Placement
Figure 7. Cut-away Cross-section of
Upgradient Placement with Drain
366
-------
_^bJt_
To Treatment s|urry
Extraction Well Wall
Figure 8. Cut-away Cross-section of Downgradient Placement
367
-------
produced by this investigation are:
o Soil distribution and nature
o Aquifer geometry and gradients
o Groundwater and leachate chemis-
try.
The accuracy and completeness of the
design phase investigations are of prime
importance to the success of a slurry wall
project and should never be under
estimated. Thorough site characterization
can allow problems to be anticipated and
planned for before they become detrimental
to the project.
SLURRY WALL CONSTRUCTION
This section presents a general
overview of the construction of a typical
SB slurry wall. The major elements
discussed are site preparation, exca-
vation, slurry preparation and control,
backfill mixing and placement, quality
control, and cost.
The first phase of slurry wall
construction is to prepare the site for
work. This will vary considerably in
nature and extent from site to site, but
will involved such tasks as leveling the
trench line, locating and/or moving over-
head and underground obstructions,
identifying storage areas and haul routes,
and constructing impoundments for slurry
hydration and storage. In cases where
groundwater is very near the surface, a
construction platform is often built along
the trench so that the slurry can be kept
well above the water table. Extra care in
site preparation can help maintain con-
struction schedules by identifying
problems and planning for their reso-
lution.
Excavation of the trench is ac-
complish using either a hydraulic backhoe
or a cable or Kelly Bar mounted clam-shell
grab. Backhoes are usually used for
depths under 50 ft, however, in recent
years, modified or "extended stick" back-
hoes have reached 73 ft. Below these
depths, clam-shell grabs have successfully
installed walls over 300 ft 13 . If a
linear wall is being installed, back-
filling can begin as soon as excavation
has reached the point beyond the
calculated backfill slope (see Figure 9).
For a continuous, circumferential wall,
however, the entire trench is often
excavated before backfilling is begun.
Slurry is usually prepared on the
site as close as practical to the trench.
The bentonite is weighed and mixed with
water in a high shear or venturi mixer.
From the mixer, the slurry is pumped to a
pond to circulate while hydration occurs.
The hydration time is a function of
bentonite quality, mixer shear, and pond
circulation, and can vary considerably.
From the hydration pond, the slurry is
usually pumped to a storage pond before
introduction into the trench. This
procedure allows new slurry to be mixed
and hydrated and provides a reserve supply
of slurry in case the slurry level in the
trench should drop suddenly when an
unexpected pervious zone is encountered.
Most slurry trenching operations are
accompanied by a simple mobile laboratory
for testing new bentonite shipments, and
for testing the slurry at various times
during construction. The slurry is gener-
ally tested for viscosity (Marsh), density
(mud balance), sand content, and pH. In
many cases, filter loss and gel strength
are also measured. Testing of the slurry
is usually done after mixing, as it is
introduced into the trench, and again as
backfill is placed (a sample is taken from
the trench bottom near the leading edge of
the backfill). If the slurry in the trench
is found to be too dense, additional new
slurry is added to bring it back within
specifications .
Backfill for a SB wall is usually
mixed alongside the trench using a
bulldozer. If additional soil or dry
bentonite is needed it is spread evenly
over the trench spoils and tracked and
bladed with the bulldozer. During the
mixing, slurry from the trench is added to
the backfill mixture to form a homogenous,
plastic paste I0. When the backfill has
reached the specified composition and
consistency, it is ready to be placed in
the trench.
Backfill placement is begun using a
clam-shell grab to lower the backfill to
the trench bottom. This continues until a
slope of backfill extends to teh surface
(see Figure 9). Thereafter, the backfill
is pushed into the trench with a bulldozer
and allowed to flow down the slope. In
some cases, the slope may be established
368
-------
Fiour* 9 Cross-section of Slurry Trench, Showing Excavation
and BackfiHing Operations
369
-------
by the excavation equipment, and initial
backfill placement by a clam-shell is not
required. If excavation and backfilling
are well coordinated, the level of slurry
in the trench should not fluctuate drasti-
cally.
As with any construction operation,
slurry trenching requires that certain
quality control procedures be followed.
These procedures may differ from job to
job, but are aimed at ensuring trench
continuity (width and depth), backfill
composition and placement, and slurry
quality 8 .
Trench continuity is extremely im-
portant to the finished wall because
unexcavated material or improper key-in
can greatly affect final permeability by
leaving "holes" in the wall. This is
mainly checked by observing the excavation
operation and the material being brought
to the surface. In shallow excavations,
it is possible to sound the trench using a
rod to "feel" for unexcavated material.
Quality control of backfill mixing
and placement is second in importance to
excavation but is essential nonetheless.
Mixing is checked mostly by observtion but
samples are taken for laboratory analysis
at various times during the job to
determine wheather they meet specifi-
cations. The control of backfill placement
is focused on the slump of the initial
backfill, and the slope thus formed. Care
most be taken to ensure that no pockets of
slurry are trapped in the backfill during
placement.
Control of the slurry properties
often receives the most attention, but is
not of extreme importance to teh final
wall. Control procedures, outlined above,
center on ensuring that the slurry is of
sufficient density to support the trench,
but not so dense that it is not easily
displaced by the backfill. Evidence
indicates that as long as the slurry is at
least 15 PCF less dense than the backfill,
no problems should occur2.
The cost of a complete slurry wall
will vary a great deal depending on depth,
ease of excavation, and materials in-
volved. Table 2 shows the range of costs
of SB and CB walls for different depths
and media. _Although the costs shown are
in 1979 dollars, they remain valid
examples. Decreases in the cost of
bentonite, and increased excavation ef-
ficiency have offset inflationary in-
creases .
CONCLUSIONS
The construction of cut-off wall
barriers, using slurry trench techniques,
to control the groundwater around landfill
waste sites has enjoyed rapidly increasing
popularity. This paper summarizes the
"Technical Handbook on Slurry Trench
Construction for Pollution Migration Con-
trol," which EPA's Office of Research and
Development has prepared. It includes
detailed discussions of the theory, design
construction, and performance monitoring
practices which are available within the
engineering profession today.
Individuals within Federal, State, or
local governments, or from private orga-
nizations who are responsible for re-
viewing the technical aspects of 'slurry
trench cut-off walls for pollution mi-
gration control, can gain from this
Handbook a familiarity with the site
characterization, construction parameters,
and performance expectations associated
with these barriers. The evidence
gathered to date indicates that a properly
designed slurry wall, installed by a
competent contractor, can be effective in
long-term pollution control. Unfortu-
nately, most of the slurry wall instal-
lations to date have been in the private
sector and little objective monitoring
data are available. Until such data are
assembled and critiqued, it remains to be
seen just how effective, and how long-
term, this remedial measure is in control-
ling the spread of contaminated ground-
water .
REFERENCES
1) Grim, R.E., "Clay Minerology," McGraw-
Hill Book Co., New York, 1968.
2) D'Appolonia, D.J., "Soil-Bentonite
Slurry Trench Cut-Offs," Journal of
the Geotechnical Engineering Division,
ASCE Vol. 106, No. GT4, April 1980.
3) LaRusso, R.S., "Wanapum Development
Slurry Trench and Grouted Cut-Off,"
Grouts and Drilling Muds in Engi-
neering Practice, Butterworth's
London, 1963.
370
-------
SOFT TO MEDIUM SOIL
N < 40
HARD SOIL
N 40-200
OCCASIONAL BOULDERS
SOFT TO MEDIUM ROCK
N > 200 SANDSTONE, SHALE
BOULDER STRATA
HARD ROCK
GRANITE, GNEISS, SCHIST*
SOIL-BENTONITE
WALL
DEPTH
< 30
FEET
2-4
4-7
4-8
6-12
15-25
DEPTH
30-75
FEET
4-8
5-10
5-8
10-20
15-25
DEPTH
75-120
FEET
8-10
10-20
8-25
20-50
50-80
CEMENT-BENTONITE
WALL
DEPTH
< 60
FEET
15-20
25-30
20-30
50-60
30-40
95-140
DEPTH
60-150
FEET
20-30
30-40
30-40
60-85
40-95
140-175
DEPTH
> 150
FEET
30-75
40-95
40-85
85-175
95-210
175-235
--'NOMINAL PENETRATION ONLY
FOR STANDARD REINFORCEMENT IN SLURRY WALLS ADD $8.00 PER SQ.FT.
FOR CONSTRUCTION IN URBAN ENVIRONMENT ADD 2521 TO 50* of Price
TABLE 2. Approximate Slurry Wall Costs As A Function Of Medium And Depth
(In 1979 Dollars Per Square Foot)
(After Ressi di Cervia)
-------
4) Ryan, C.R., "Technical Specifications,
Soil-Bentonite Slurry Trench Cut-Off
Wall," GEO-CON Inc., undated.
5) Veder, C., "Excavation of Trenches in
the Presence of Bentonite Suspensions
for the Construction of Impermeable
and Load-bearing Diaphragms," Grouts
and Drilling Muds in Engineering
Practice, Butterworth's, London, 1963.
6) Nash, J.K.T.L., and Jones, G.K., "The
Support of Trenches Using Fluid Mud,"
Grouts and Drilling Muds in Engi-
neering Practice, Butterworth's,
London, 1963.
7) Jefferis, S.A., "Bentonite-Cement
Slurries for Hydraulic Cut-Offs,"
Proceedings of the Trench Inter-
national Conference on Soil Mechanics
and Foundation Engineering, Vol. I,
Rotterdam, June 1981.
8) Ryan, C.R., "Slurry Cut-Off Walls;
Methods and Applications," Presented
at GEO-TEC '80, Chicago, IL, March 18,
1980.
9) Xanthakos, P.P., "Slurry Walls,"
McGraw-Hill Book Co., New York, 1979.
10) D'Appolonia, D.J., and Ryan, C.R.,
"Soil-Bentonite Slurry Trench Cut-Off
Walls," Presented at Geotechnical
Conference, Chicago, IL, March 26,
1979.
11) Anderson, D., and Brown, K.W.,
"Organic Leachate Effects on the
Permeability of Clay Liners," Land
Disposal: Hazardous Waste, Proceeding
of the Seventh Annual Research Sym-
posium, U.S. EPA, MERL, Cincinnati,
OH, 1981.
12) Anderson, D., Brown, K.W., and Green,
J., "Effects of Organic Fluids on the
Permeability of Clay Soil Liners,"
Land Disposal: Hazardous Wastes, Pro-
ceedings of the Eighth Annual Research
Symposium, U.S. EPA, MERL, Cincinnati,
OH, 1982.
13) Case Slurry Division, Case Inter-
national Co., "Case Slurry Wall Cons-
truction Notebook," 139 p Roselle, IL,
undated.
14) Ressi di Cervia, A.L., "Economic
Consideration in Slurry Wall Appli-
cations," Proceedings of a Symposium
on Slurry Walls for Underground Trans-
portation Facilities, Cambridge, MA,
August 30-31, 1979.
372
-------
HANDBOOK FOR EVALUATING REMEDIAL ACTION TECHNOLOGY PLANS
John R. Ehrenfeld and Jeffrey M. Bass
Arthur D. Little, Inc.
Cambridge, MA 02140
ABSTRACT
Remedial action technologies are designed to reduce human exposure to hazardous waste
and decrease the contaminants to acceptable levels by either containing hazardous
materials in place or removing the intrinsic hazard by decontaminating or physically
removing the hazardous substances.
This paper discusses a document which provides an outline of technical information that
potentially could be used to evaluate long-term remedial action plans for controlling or
treating wastes or leachates at uncontrolled hazardous waste sites. The document,
"Handbook for Evaluating Remedial Action Technology Plans", is not a design manual, nor
does it contain rules or regulations pertaining to remedial actions. The intended
audience includes those involved in the review of preliminary engineering reports or final
design of remedial action at the waste sites.
INTRODUCTION
The development of remedial action
plans and the ultimate implementation of
those plans follows a series of steps set
out in the National Contingency Plan
(NCP)(1). Early in the process, alterna-
tive technological approaches are first
identified and then screened through a net
of increasing stringency regarding cost
and effectiveness. The U.S. Environmental
Protection Agency (EPA) has developed a
substantial body of information on the
various technological alternatives, most
recently culminating in the publication
"Handbook for Remedial Action at Waste
Disposal Sites"(2). This Handbook con-
tains extensive information about many
potential technologies and is organized to
assist planners and engineers in selecting
or rejecting approaches.
As the plans evolve and get closer to
implementation, they must be examined for
technical and economic feasibility and
conformance with the guidelines in the
NCP. The evaluation process is carried
out by many parties: the EPA, state
agencies, responsible disposers and
facility operators, and the public at
large. This paper describes a complemen-
tary handbook (3) to the earlier publica-
tion that supports the evaluation process.
BACKGROUND
The Comprehensive Environmental
Response, Compensation, and Liability Act
of 1980 (CERCLA, also commonly known as
Superf und), was passed late in 1981 to
provide for (1) cleanup and emergency
response for hazardous substances released
into the environment, and (2) cleanup of
inactive hazardous waste disposal sites.
Section 105 of the Act directs the Presi-
dent to prepare a National Contingency
Plan (NCP) to establish procedures and
standards for responding to releases of
hazardous substances. The NCP was re-
quired to include, among other things,
methods for evaluating (including analyses
of relative cost) and remedying any
releases or threatened releases of hazar-
dous substances, and means of assuring
that remedial measures are cost
effective.
373
-------
After receiving comments on several
drafts, EPA published the NCP in final
form July 16, 1982 (40 CFR 300). The NCP
describes procedures to develop and
implement plans for remedying releases of
hazardous substances. The NCP implements
requirements in CERCLA and in the Clean
Water Act (CWA).
The NCP process incorporates a number
of judgments and decisions based on
technical considerations. These judgments
and decisions occur at several steps along
the process and are made by the lead
agency (EPA or a state depending on the
existence and nature of the agreement
between EPA and the state), private
responsible parties developing remedial
action plans, consultants and engineers
supporting the above interests, and other
interested parties.
The key parts of the NCP concerning
remedial action are contained in Subpart
F, which identifies the state role (Section
300.62) and a phased procedure for respond-
ing to the release of hazardous substances
(Sections 300.63 to 300.70). Figure 1
illustrates the flow of the phases. The
major focus of this report is on Phase VI
Remedial Action (Section 300.68).
Figure 2 illustrates the Phase VI process
in detail and indicates the steps asso-
ciated with technical criteria for which
the information in this document has been
developed and organized.
The first several steps are designed
to elucidate the nature of the problems at
a site, to determine the major courses of
action, and to develop a series of poten-
tially cost-effective alternatives dic-
tated by those actions. The latter steps,
beginning with the initial screening of
alternatives, are designed to select the
cost-effective alternative using the set
of criteria specified in the NCP.
The criteria include several that are
related to technical features of the
conceptual design in relationship to
site-specific characteristics (the factors
identified with an asterisk in Figure 2) .
Judgments as to the degree which proposed
plans conform to the criteria must be
based on project-specific designs and
analyses. But there are a number of
general, technology-specific data and
engineering considerations which can be
used to evaluate the plans, determine
their appropriateness, and ensure that
they have been developed from sound
engineering principles using reasonable
cost estimates.
More specifically, the report pro-
vides data to support the analysis and
evaluation of:
o Feasibility, including:
- acceptability (relevance to the
particular project)
- implementability or
constructability
o Effectiveness, including
reliability
o State of Development
o Cost
REPORT ORGANIZATION
The report includes a general dis-
cussion of remedial approaches to remedy
problems in five environmental media:
o groundwater/leachate;
o surface water;
o soil;
o waste materials; and
o air.
These categories are used to organ-
ize the discussions of 20 control tech-
nologies, i.e., remedial approaches to
limit the extent of contamination and
potential exposure and 28 treatment
technologies, i.e., approaches that
decontaminate the sources of potential
exposure. Table 1 lists the technologies
discussed in the report (3).
Each individual discussion includes
highly summarized information on the items
below:
o Description — A brief, qualita-
tive discussion of the technology
and the principles on which it is
based.
374
-------
FIGURE 1
HAZARDOUS SUBSTANCE RESPONSE SEQUENCE (40 CFR PART 300)
PhiMl
Discovery or
Notification
(300 63)*
Phased
Preliminary
(30064)
Head Agency
PtWMlll
Immediate
Removal
(300.65)
MUM IV
Evaluation and Determination
of Appropriate Response
Planned Removal and
Remedial Action
(30066)
PtMMV
Planned
Removal
1300.67)
"Refers to the section of the NCR in 40 CFR part 300.
-------
FIGURE 2
DETAILED SEQUENCE-PHASE VI-REMEDIAL ACTION (40 CFR PART 300.68)
s to Determine
/ Appropriate Extent of
V Remedial Action
\^
Cost*
Effects of the alternative*
Acceptable engineering practice*
Refinement and specification
Use of established technology*
Detailed cost
Engineering implementation or
constructibility*
Technical effectiveness compared
to other alternatives*
Analysis of adverse impacts
Lowest cost
Technical feasibility and reliability*
Mitigates and minimizes damage,
provides adequate protection
* I ndicates where information in this report may be useful.
**Referenced to Section of 40 CFR Part 300.
-------
TABLE 1
REMEDIAL ACTION TECHNOLOGIES
CONTROL TECHNOLOGIES
TREATMENT TECHNOLOGIES
GRQUNDWATER CONTROL TECHNOLOGIES
Slurry Walls
Grout Curtains
Sheet Pile Cutoff Walls
Block Displacement
Groundwater Pumping
Subsurface Drains
SURFACE WATER CONTROL TECHNOLOGIES
Dikes
Terraces
Channels
Chutes and Downpipes
Grading
Surface Seals
Vegetation
Seepage Basins and Ditches
SOIL AMD HASTE TECHNOLOGIES
Excavation
Drum Handling
Encapsulation
Dewacering
AIR CONTROL TECHNOLOGIES
Pipe Vents
Trench Vents
MONITORING TECHNOLOGIES
Monitoring Wells
Others Discussed
BIOLOGICAL TREATMENT
Activated Sludge
Surface Impoundments
Rotating Biological Discs
Trickling Filters
Land Treatment
CHEMICAL TREATMENT
Neutralization
Precipitation
Reduction (For Cr)
Wet Air Oxidation
Chlorination (For Cyanide Only)
Ozonation
PHYSICAL TREATMENT
Reverse Osmosis
Equalization/Detention
Ion Exchange
Carbon Adsorption
Stripping
Sedimentation
Dissolved Air Flotation
Filtration
DIRECT TREATMENT
In Situ Leachate/Groundwater Treatment
In Situ Physical/Chemical Treatment
On-Site Physical/Chemical Treatment
Solution Mining (Extraction)
Biodegradation
Solidification/Stabilization
Incineration
Thermal Oxidation Systems
Carbon Adsorption For Air Emissions
o Status — A measure of the availa-
bility of the technology and
degree to which it has been
demonstrated for hazardous waste
remedial actions.
o Feasibility and Effectiveness — A
discussion of the technical
factors important in understanding
and reviewing the technology.
This part gives background neces-
sary for evaluating technology
design plans and identifies areas
of particular concern. For
example, the effect of waste
constituents on backfill and
slurry materials is crucial to the
effectiveness of a slurry wall.
This part also contains a short
discussion of special precautions
to be taken in using the tech-
nology and any limitations of the
technology being discussed.
o Design Basis — This portion
includes a summary of the major
factors which determine the
performance of treatment technol-
ogies. This part, which is unique
to the treatment technologies is
used to separate basic design
information from the technical
concerns of implementing leachate
treatment technologies. Where
possible, equations relating
design parameters to desired
performance and site conditions
are provided in summarized form.
Otherwise, the relationships are
described in qualitative terms.
o Principal Data Requirements — The
principal site-specific data
necessary for the design and
evaluation of the technology are
noted. The data noted must be
obtained in the site investigation
or in laboratory based studies.
377
-------
The evaluation criteria or per-
formance related factors most
directly related to each major
data requirement are noted. A
summary of _ the data requirements
common to all of the technologies
is shown in Tables 2 and 3.
o Elements of Cost Review
Information is provided for
analyzing technology cost esti-
mates. Each cost discussion is
divided into three parts. Compon-
ents lists the major components of
cost, including those which
involve initial construction and
capital costs, and those which
involve ongoing operation and
maintenance (O&M) costs. Major
Factors lists the characteristics
of the technology as designed for
a given site (e.g., size, material
availability, pretreatment require-
ments) that most affect overal]
cost. Data provides information
on unit and total costs associated
with the technology. Cost data
provided has been updated from
available sources to 1982 dollars
using several conventional in-
dices.
EXAMPLE
The following example, on precipita-
tion, a leachate treament technology, is
adapted from the report to indicate the
nature of the information included.
Precipitation
Description
Precipitation is a widely used (in
industrial practice), relatively low-cost
physical chemical technique in which the
chemical equilibrium of a waste is changed
to reduce the solubility of the undesired
components. These components precipitate
out of solution, as a solid phase, often
in the form of small or even colloidal
particles, and are removed by one of
several possible solids removal tech-
niques. Precipitation is most commonly
used to treat heavy metals-containing
wastes.
Status
Conventional, demonstrated.
Feasibility and Effectiveness
General Features
Precipitation is induced by one of
the following means:
o adding a chemical that will react
with the hazardous constituent in
solution to form a sparingly
soluble compound;
o adding a chemical to cause a shift
in solubility equilibrium, reduc-
ing the solubility of the hazard-
ous substance; or
o changing the temperature of a
saturated or nearly saturated
solution in the direction of
decreased solubility.
Chemical additives are the most often
commonly used. Typical reagents are:
o sodium hydroxide, sodium sulfate
o lime (Ca (OH) )
o Iron salts, iron sulfide, ferric
sulfate
o phosphate salts (especially for
heavy metals such as As, Cd, Cr,
Cu, Hg, Ni, Pb, Zn)
o Alum (A12 (S04)3)
The theoretical removal limits for
many metal species is very low, particu-
larly with sulfide precipitants. Figure
3 shows theoretical curves as a func-
tion of waste pH. Some organic species—
for example, aromatic compounds and phtha-
lates—can also be treated. Removal in
practice often is one to two orders of
magnitude less than the theoretical limit.
Complexing agents, such as cyanide or
EDTA, compete with the precipitant and may
hold the species in solution.
Conventional precipitation processes
are performed in the following three
steps:
1. rapid mixing of precipitating
chemicals and wastewater;
378
-------
TABLE 2
CONTROL TECHNOLOGY DATA REQUIREMENTS
• = Useful Data
• = Required Data
REMEDIAL ACTION
GROUNDWATER
slurry wall
sheet pile
pumping
subsurface drams
SURFACE WATER
dikes - runoff
terraces
channels
chutes and downpipes
grading
surface seal
vegetation
Seepage basin/ditches
SOIL AND WASTE
excavation
drum handling
encapsulation
AIR
pipe vents
trench vents
Geography
•5
Topography
Accessibility of Site or Materi
Vegetation Characteristics
*
;
» • *
• •
* i •
• •
» » •
• •
• •
* •
Geology
Rock Type
Structural Characteristics
Thickness of Strata
Depth to Impermeable Strata
Hydraulic Conductivity
It*
•
«
*
• •
*
Soils
§
1 f * E
lUfii
• • • • •
• »»«>• *
• • « •
• ••»»•»
• ...»•*
Climatology
1
Precipitation (rainfall paramel
Evapotranspiration
Storm Characteristics
Wind Characteristics
Temperature
Air Quality
• •
9 ^
• *
• •
• • • •
• • » •
• • » •
Groundwater
Depth to WatertaMe
Potentiomatnc Surfaces
Direction and Rate of Flow
Recharge Quantity
Aquifer Characteristics
Chemistry
Infiltration
» • »
* • *
•
*
•
« •
* •
Surface Water
Runoff
Depth of Flow
Drainage Area
Flood Characteristics
Sedimentation
* * *
# • #
* * *
« «
* *
Waste
Descrip-
tion
Chemical Characteristics
Physical Characteristics
Disposal Practices
»
«
III
* » »
» » *
» »
• * *
• * *
Reg
Regulations
*
+
*
«
Other
fc
O
•
I
*
*
4
»
*
*
After Enos, 1980
-------
TABLE 3
TREATMENT TECHNOLOGY DATA REQUIREMENTS
TREATMENT
TECHNOLOGY
Biological Treatment
Activated Sludge
Rotating Biological DISC
Trickling Filter
Surface Impoundment
Land Treatment
Chemical Treatment
Chemical Oxidation
Alkaline Chtormation
Ozonation
Chemical Reduction
Neutralization
Prectprtation
Ion Exchange
Wet Air Oxidation
Physical Treatment
Carbon Adsorption
Density Separation
Sedimentation
Flotation
Filtration
Reverse Osmosis
Stripping
Equalization/Dentention
In Situ Treatment
Biological Methods
Chemical Methods
i
1
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
z
X
X
X
X
X
5
IL
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
i
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
I
«
X
Ul
8
s
•0
I
1
Acidity E
X
X
X
X
X
X
X
8
X
X
X
X
X
X
Q
8
X
X
X
X
X
X
8
X
X
X
X
X
X
X
X
s
X
X
X
X
X
X
X
X
X
X
X
X
B
X
X
X
X
1
X
X
X
X
X
X
X
X
X
Cyinidi
X
X
X
X
X
X
X
X
e
1
X
X
X
X
X
X
Ł
5
f
X
1
X
X
Clinutt
X
X
X
X
X
X
X
Ł
1=
1
i
X
X
X
Soil CEC
X
X
X
1
a.
I
•f
e
Oxidttio
Ł
i*
3
1
-1
X
X
X
X
X
X
X
X
e
2
§
|
0
Ł
X
X
X
X
X
4
1
|
I
3
•s
i
X
X
X
1
1
1
•s
1
X
X
X
X
X
X
X
X
X
X
X
X
X
X
-------
Figure 3.
SOLUBILITY OF METAL HYDROXIDES AND SULFIDES
(Source Ghassemt, el al , 1981)
10°
KT1
I 10-6
5
o
« io*
»o,s
7
pH
2. slow mixing of treated wastewater
in a flocculation tank to allow
settleable floes to form; and
3. sedimentation of solids in a
clarification tank.
The solids are removed by either:
o sedimentation, which separates the
phases by the gravitational
settling of the precipitate to the
bottom of the sedimentation tank;
o filtration, which separates the
phases by passing the precipita-
tion effluent through a granular
or cloth barrier, retaining the
particles and allowing the clear
effluent to pass through; or
o centrifugation, which separates
the two phases in an enclosed
vessel using certrifugal force to
cause the solids to migrate
through the liquid.
Special Precautions and Limitations
As noted, removal can be limited in
the presence of complexing agents in the
wastes. This problem can generally be
eliminated by:
o using a sulfide precipitation
agent;
o breaking up the metal complex by
altering pH to either a basic or
acidic extreme and adding a
substitute cation to tie up the
complexing agent when the pH is
readjusted to precipitate the
metal; or
o using insoluble starch xanthate as
a precipitation agent (not widely
used).
The sludge produced by precipitation
should be considered hazardous unless
laboratory tests show otherwise.
Design Basis
The major design factors are:
o effluent criteria;
o leachate flow; and
o concentration of precipitable ions
in the leachate.
Based on the wastewater analyses and
solubility curves for the species to be
removed, laboratory tests are designed to
determine optimum precipitation conditions
and chemical requirements to satisfy
effluent criteria.
A mixing tank is sized based on the
leachate flow and precipitation chemical/
leachate contact time required. Gener-
ally, contact time ranges from 10 to 60
minutes. Flocculation tank sizes are
based on flow and retention time (typi-
cally 30 to 60 minutes). Clarification
tank size is based on laboratory experi-
ments to determine the settling rate and
the leachate flow.
Principal Data Requirements
Leachate analysis (reagent choice,
capacity);
381
-------
o precipitable constituents;
o Interfering species (i.e., cyan-
ide, EDTA, etc.);
Leachate daily average and variation
on flow (capacity);
Treatability study (capacity, reagent
choice, and rate);
o optimum precipitation conditions;
o settling rate;
o sludge production rate;
o leachate flow;
o wastewater analyses of precipi-
table constituents;
o wastewater analyses for constit-
uents that interfere with precipi-
tation (i.e., cyanide, EDTA,
etc.); and
o sludge production rate.
Elements of Cost Review
Components
Construction and Capital—
o tanks
o pumps
o mixers
0 & M~
o chemicals
o electricity
Major Factors
o process volume
Data
Sample costs for several different
capacity precipitation systems are given
in Table 4.
ACKNOWLEDGEMENT
The preparation of this handbook was
performed under Contract No. 68-01-5949
for the U.S. Environmental Protection
Agency. The handbook will be available
from the National Technical Information
Service and the Solid and Hazardous Waste
Research Division of EPA in Cincinnati,
Ohio.
REFERENCES
Federal Register, July 16, 1982.
National Oil and Hazardous Substances
Contingency Plan 47:31180.
EPA, Technology Transfer, 1982.
"Handbook for Remedial Action at
Waste Disposal Sites," EPA-625/-
6-82-006.
EPA, Technology Transfer, 1983.
"Handbook for Evaluating Remedial
Action Technology Plans," In Press.
Ghassemi, Masood, et. al., 1981.
"Feasibility of Commercialized Water
Treatment Techniques for Concentrated
Waste Spills," TRW Environmental
Engineering Division, Redondo Beach,
Calif. Report No. PB82-108440 or EPA
600/2-81-213.
Arthur D. Little, Inc., 1976. "Physi-
cal, Chemical and Biological Treat-
ment Techniques for Industrial
Wastes," Report to U.S. EPA, Office
of Solid Waste Management Programs.
382
-------
TABLE 4
PRECIPITATION. FLOCCULATION. AND SEDIMENTATION COST ESTIMATES
AS A FUNCTION OF SIZE
Basis: -Wastewater from Automotive Plating Operation
-Flowrate = 410,000 gpd, 350 day/year, 24 hour day
-Zinc Concentration = 11J mg/1
Capital Investment - $487,600
Treatment System Size (Wastewater Flowrate)
Capital Investment
Variable Cost
Labor
Maintenance (4% Inv.)
Chemicals
Quicklime
Coagulant Aid
Sulfuric Acid
Electrical Energy
Total Variable Cost
Fixed Cost
Capital Recovery
Taxes and lusurancu
(2% Inv.)
Total Fixed Cost
Total Annual Cost
Unit Cost ($/1000 Gal)
82,000 gpd
$171,700
72,400
6,900
400
1,400
220
1,200
$ 82,500
27,500
3,400
$ 30,900
$113,400
$ 3.95
410,000 gpd
$487,600
109,800
19,500
1,800
7,300
1,100
6,000
$145,500
78,000
9,800
$ 87,800
$233,300
$ 1.63
2,050,000 gpd
$1,388,300
165,100
55,500
8,900
36,300
5,500
29,800
$301,100
222,100
27,800
$249,900
$551,000
$ .77
Source: ADL, 1976
383
-------
COLLECTION OF INFORMATION ON THE COMPATIBILITY OF GROUTS WITH HAZARDOUS WASTES
Gary E. Hunt, Virginia E. Hodge, Pauline M. Wagner, and Philip A. Spooner
JRB Associates
McLean, Virginia 22102
Herbert R. Pahren
EPA Municipal Environmental Research Laboratory
Cincinnati, Ohio 45268
ABSTRACT
This paper presents information on compatibility of grouts with different classes of
chemicals. This information can be used for testing and selection of the type of grout to
be utilized at a specific waste disposal site based on the compounds found in the
leachate.
Twelve different types of grout are included in this study, based on their availabil-
ity and utilization in waterproofing and consolidation projects. Sixteen general classes
of organic and inorganic compounds are also identified that encompass the types of chemi-
cals which could be found in leachate from hazardous waste disposal sites. The known
effects that each of the chemical classes would have on the setting time and durability of
each grout have been identified through an in—depth information search. These effects are
presented in matrix form. Since compatibility data are not complete for each grout type,
predictions based on reaction theory are made, where possible, for silicate and organic
polymer grouts, to make information available on general compatibility trends.
The selection of a grout for a specific waste site will depend on its injectability,
durability, and strength. These factors relate site hydrology, geochemistry, and geology
to grout physical and chemical properties.
INTRODUCTION
Grouting is a technique by which a
fluid material is pressure injected into
soil or rock to reduce water movement
and/or impart increased strength. Grouts
accomplish this through their ability to
penetrate soil voids and to gel or solidify
in place. This technique has for many
years been widely used for soil consolida-
tion and groundwater control in tunneling,
dam construction, and other building pro-
jects. However, its use at hazardous waste
sites as a groundwater or leachate control
measure has been limited. While all of the
grout application techniques developed by
the construction industry can be utilized
at disposal sites, the ability of grouts to
withstand the adverse chemical conditions
found at these sites requires further
investigation.
The types of application techniques
that can be used at disposal sites include:
o curtain grouting,
o jet grouting,
o area grouting,
o contact grouting, and
o large void sealing (8).
All of these techniques can reduce water or
leachate movement by reducing soil/rock
permeability.
384
-------
A grout curtain is an underground
cut-off or barrier formed by injecting
grout through grout pipes placed down bore
holes in a specific pattern, typically in
triple rows of primary and secondary holes.
Starting with the primary holes, grout is
injected in "pillows," working upward or
downward in stages. This procedure forms
rows of primary columns which, when set,
are tied together by injecting the second-
ary columns (5). Figure 1 illustrates the
wall or curtain formed by the coalescing
columns.
Jet grouting is a method of cutting an
opening in soil or soft rock using a high
pressure nozzle (3). Once an opening has
been made, using either water or grout as
the cutting fluid, grout sets in the voids
to form a low permeability zone. Using
this technique, it is theoretically possi-
ble to emplace a grout liner beneath a
waste site. Although this may require
jetting through a hole drilled down through
the wastes, advances in directional drill-
ing techniques may soon make this method
more practical.
Area grouting is a low pressure tech-
nique used to form a grout "blanket" in
near surface soil materials. This is
accomplished by injecting grout into a
series of closely spaced injection holes
placed in a grid pattern. This is usually
performed to reduce infiltration, prevent
erosion losses, or stabilize soils for
heavy machinery bases (1). Area grouting
may be adaptable to in situ waste immobili-
zation depending on the compatibility of
wastes with grouts.
Contact grouting is low pressure
injection of grout to seal surface cracks
or other voids. This technique may be used
to patch cracked concrete in dams or dikes
to reduce leakage, and to seal the sides of
excavations to reduce water infiltration
(1). Applications of contact grouting to
long term waste site remediation are
limited.
Grouts have long been used for sealing
large voids in rock and other materials,
mainly in conjunction with dam or tunnel
projects (1). Adaptations of this method
to waste disposal sites could include
forming a tie-in between a cut-off wall and
highly fractured or weathered bedrock or
for sealing leaks in aquatards caused by
exploration bore holes or improperly in-
stalled wells. Figure 2 illustrates the
use of grout to key in the bottom of a
cut-off wall.
While grouting is a well proven con-
struction technique for groundwater control
that can be applied to disposal sites, the
extreme chemical conditions found at these
sites could reduce the effectiveness of the
soil seal. While a great deal of informa-
tion has been published on the physi-
cal/chemical properties of grout, very
little data are available on the effect of
chemicals on grout performance. Conse-
quently, this study centered on collecting,
organizing, and analyzing existing publish-
ed and unpublished information on the
compatibility and durability of grouts with
various classes of chemicals.
GROUT TYPES AND THEIR CHEMICAL INTERACTIONS
The grouts currently used for soil
consolidation and groundwater control are
particle suspensions, polymers, and emul-
sions. These materials are generally
water-based solutions of sufficiently low
viscosity to allow them to penetrate rock
and soil voids .
Suspension grouts are the most common
type of grout; they constitute approximate-
ly 95 percent of the grout used. Cement,
clay, and clay-cement grouts are in this
category. These materials are usually the
more viscous of the available grouting
materials and have the largest particle
size. Thus, these grouts are best used in
the grouting of rock or coarse material.
Polymer grouts represent almost 5
percent of the grout used. This class of
grouts includes both silicate and organic
polymer grouts with silicates constituting
almost all of this category. Organic
polymer grouts include acrylamide, ure-
thane, phenolic, urea-formaldehyde, epoxy,
and polyester grouts. Polymer grouts have
low viscosities, initially, and thus can be
used in fine grained, cohesionless soils,
as well as for secondary treatment of
coarse soils and rock fissures.
Emulsion grouts consist of either
bitumenous emulsions or asphalts. These
grouts can be used to seal soils, fill rock
cavities, or construct thin cut-off walls.
385
-------
Semicircular
Grout Curtain
Source: 9
Figure 1: Semicircular Grout Curtain Around Waste Site
Cut-off Wall
Source: 8
Figure 2: Bottom Key Grouting
386
-------
However, they are limited in use.
Chemical interactions with grouts may
be either positive or negative. Positive
interactions tend to improve the character-
istics of the set grout (e.g., decrease
permeability, increase strength). Negative
interactions will change the charac-
teristics or behavior of grout such that it
no longer fulfills its intended purpose
which is to prevent migration of contami-
nants. These negative grout/chemical
interactions take three general forms:
o prevent formation of grout matrix,
o destroy/interfere with grout
matrix, and
o prevent sealing of grout
with soil/rock.
Chemicals, as well as groundwater
flow, can delay the set time of the grout
such that a gel matrix never forms or is
substantially weakened. Many of the
grouts, particularly the organic polymer
grouts, consist of several reactants as
well as additives to catalyze or control
the reaction rate. Any of these substances
may preferentially react with chemicals in
the groundwater (or leachate) as well as
soil components. Such reactions can alter
the set time of the grout through changing
the mole ratio of reactants (by destroying
reactive substances) or destroying addi-
tives that control the reaction and are
generally present in very small quantities.
In addition, the grout may be diluted by
groundwater following injection and this
can prevent gel formation in some grouts
(e.g., phenolics).
Once the grout has set and the matrix
has formed, chemicals may interact with the
matrix in such a way that permeability
increases. First, the grout matrix and its
components can shrink or swell in reaction
to chemicals as well as water. This
swelling or shrinkage corresponds to water
moving into or out of the matrix; such
water movement can result in irreversible
cracking of the grout which increases its
permeability. The second interaction
mechanism involves dissolution or leaching
of grout matrix components which ultimately
creates piping or channeling within the
matrix and increases its permeability.
This leaching may occur through hydrolysis
of matrix components or through chemical
dissolution (e.g., solvents dissolving
bitumen). The third mechanism involves
reaction of chemicals with the grout matrix
components. Such reactions can destroy the
matrix structure, thus increasing its
permeability. For example, chemicals can
react with polymer grouts to destroy either
the polymer chains, the crosslinks between
polymer chains, or both; as these bonds are
broken, the polymer matrix is destroyed and
chemical solutions may penetrate the grout.
Finally, chemicals in the soil can
prevent formation of a seal between the
grout and the soil or rock. Proper sealing
may also be affected by dehydration of the
grout. Dehydration can cause the grout to
shrink from the surrounding soil or rock;
this may also occur during setting or
"drying" of the grout. In some instances
this shrinkage is irreversible and perma-
nent voids are formed.
GROUT/CHEMICAL COMPATIBILITY
Through a detailed evaluation of
available information on the effects of
chemicals on grout performance, a series of
matrices were developed that summarize and
define the compatibilities of grouts with
various chemical groups. This information
was gathered through contacts with repre-
sentatives from universities, industries,
trade associations, and government agencies
along with a detailed review of the
published literature sources which address-
ed not only grouting-related publications
but also references on waste fixation/so-
lidification, cement chemistry, landfill/-
lagoon liner performance, slurry wall
construction, and polymer chemistry as well
as performance specifications for cement,
coatings, and tile grouts.
From the information obtained, three
matrices were develped which summarize the
known and predicted compatibilities of
grouts with various chemical groups. These
matrices provide a step-wise analysis of
the data moving from general to specific
information that has been found. It must
be-noted, though, that most of the informa-
tion detailed the effects of pure chemicals
or did not specify a concentration. While
leachates generally contain low levels of
compounds, there can be instances where
grouts will come in contact with high
387
-------
concentrations of chemicals, such as organ-
ic solvent lenses within the groundwater
system; thus, information on pure chemicals
is relevant to grout/chemical interactions.
Chemical Groups
In order to address the types of
chemicals that may come into contact with
grouts used in a disposal site situation,
the universe of chemicals was divided into
16 basic groups. The choice of these
groups was not meant to be all inclusive
but, rather, a representative choice of the
types of chemicals which could be found in
landfills. These chemical groups are as
follows:
o alcohols and glycols,
o aldehydes and ketones,
o aliphatic and aromatic
hydrocarbons,
o amines and amides,
o chlorinated hydrocarbons,
o ether and epoxides,
o heterocyclics,
o nitriles,
o organic acids and acid chlorides,
o organometallics,
o phenols,
o esters,
o heavy metal salts and complexes,
o inorganic acids,
o inorganic bases, and
o inorganic salts.
The organic categories were chosen by
functional group or structural character-
istic; the inorganics are essentially
divided into the classical acid/base/salt
categories. The functional grouping for
the organics is useful because, although
their physical properties and solubilities
may differ, the interaction of the particu-
lar functional group with other groups
remains the same. For example, all amines
may not be soluble in water, but they all
can be halogenated. In addition, one
chemical may fit in more than one category
if it has more than one functional group.
For example, p-aminobenzoic acid is both an
amine and an organic acid. The amine and
acid moieties will chemically act indepen-
dently of one another although the reactiv-
ity may be modified to some extent by the
presence of the other group.
Known Compatibilities
Based on the information search, a
matrix was developed which defines the
known interactions between grouts and six
general chemical classes: acids, bases,
polar solvents, nonpolar solvents, heavy
metals, and salts. Much of the available
data on grout compatibilities refer to the
effect of general classes of chemicals,
such as salts or solvents, and not specific
chemicals. This matrix, presented in Table
1, utilizes an alphanumeric compatibility
index which shows the effects that a
chemical class will have on the set time of
a grout and its durability once it is set.
The numerical portion of the index defines
the effect that the chemical classes will
have on grout set times, as follows:
1. no significant effect,
2. increase set time (lengthen
or prevent from setting), or
3. decrease set time.
The alphabetic code relates to the durabil-
ity of the grout in the presence of chemi-
cals, as follows:
a. no significant effect,
b. increase durability,
c. decrease durability (destructive
action begins within a short time
period), and
d. decrease durability (destructive
action occurs over a long time
period).
A question mark on the matrices indicates
where information is not available for one
or the other of the codes, or both.
It should be noted that the matrix
codes only address changes in set time or
durability as a result of exposure to
chemicals. The codes do not address the
specific mechanisms that lead to changes in
set time or durability nor do they address
mechanisms other than chemical action. For
example, bacterial action can destroy some
grouts, thus the bacteria affect the
durability of these grouts; this form of
grout destruction is not addressed by the
matrix.
Table 2 presents a more detailed
matrix, where the chemical groups are more
388
-------
Table 1: Interactions Between Grouts and Generic Chemical Classes
Grout Type
Chemical Group
Portland Cement
o
c
e
m
o
x
a
O
Polymers
c
a
Ł.
Acid
1d
Id
3a
2c
2c
Base
Id
1aT
2c
3d
2c
Heavy Metals
la
Non-Polar Solvent
'd
2c
'd
Polar Solvent
2c
Inorganic Salts
2c
2a
2d
3a
KEY Compatibility Index
Effect on Set Time
1 No significant effect
2 Increase in set time (lengthen or prevent from setting)
3 Decrease in set time
Effect on Durability
a No significant effect
b Increase durability
c Decrease durability (destructive action begins within
a short time period)
d Decrease durability (destructive action occurs over a
long time period)
* Except sulfates, which are ?c
t Except KoH and NaOH, which are Id
* Except heavy metal salts which are 2
§ Non-oxidizing
•• Modified bemomte is d
' Data unavailable
389
-------
Table 2: Interactions Between Grouts and Specific Chemical Groups
^v Grout Type
Chemical Group ^v
Organic Compounds
Alcohols and Glycols
Aldehydes and Ketones
Aliphatic and Aromatic
Hydrocarbons
Amides and Amines
Chlorinated Hydrocarbons
Ethers and Epoxides
Heterocyclics
Nitnles
Organic Acids and Acid
Chlorides
Organometallics
Phenols
Organic Esters
Inorganic Compounds
Heavy Metal Salts and
Complexes
Inorganic Acids
Inorganic Bases
Inorganic Salts
Bitumen
Portland Cement
1
>
•D
C
«
I
Clay (Bentonitel
Clay-Cement
Silicate
Polymers
Acrylamide
Phenolic
Urethane
Urea
formaldehyde
>
X
o
Q.
UJ
Polyester
'a
?d»
?d
7
?d
7
7
7
?a
7
?d
7
>d
7a5D
?a
?d
?d
7
2a
7
2d
?
7
7
1d
7
Id
7
>d
7
2'
7
2d
7
7
7
Id
?
7
7
?d
?d
'd
7
7
7
?d
7
?d
7
?d
7
'd
7
7
7
7
7
7
7
?d
7
?d
?
2c
1d
la
2c
2a
1a
1a'
2a
'd
?c"
>c«-
2d
2c
'c
'd
?d*
7
7
7
7
7
7
7
7
'a
7
7
1a
'd
'a
?a
7
~>a
>a
>a
7
2a
7
7
7
7
3a
?d*
7
>d
7
7
7
?a
7
2a
7
3a
?d
'a
3?
'a
?a
7
7
2a
7
?C
7
7
7
2a
7
2a
?d
7
7
1a
7
7
7
?a
?
?d
7
'd
7
7
7
?d
7
7
7
3'
3a
2c
3?
2'
2c
3d
3d
7
?a5
'd
3a*
7
2c
'd
'd
'a
Id
2c
'a
7
?a*
'a
'a
?a
7
'd
7
'd
7
7
7
'd
7
7
7
7
?a*
?d
'a
KEY: Compatibility Index
Effect on Set Time
1 No significant effect
2 Increase in set time (lengthen or prevent from setting)
3 Decrease in set time
Effect on Durability
a No significant effect
b Increase durability
c Decrease durability (destructive action begins within
a short time period)
d Decrease durability (destructive action occurs over a
long time period)
* Except sulfates, which are 'c
t Except KoH and NaOH, which are Id
t Low molecular weight polymers only
§ Non-oxidizing
it Non-oxidizing, except HF
^ Except concentrated acids
• Except aldehydes which are la
* Except bleaches which are 3d
* For modified bentonites, ?d
? Data unavailable
390
-------
specific. The 16 chemical classes describ-
ed previously are divided into two catego-
ries: organic and inorganic. This matrix
provides a more detailed understanding of
the types of chemicals that may have a
detrimental effect on grouts. Like the
general matrix, Table 2 does not contain
any predictions or estimations of chemi-
cal/grout interactions. Only direct com-
patibility data are presented.
Predicted Polymer Grout/Chemical Compati-
bility
Because of the missing information
encountered in preparing Tables 1 and 2,
compatibility predictions were made for the
silicate and organic polymer grouts. Other
grout types were not evaluated due to the
lack of information regarding their actual
solidification chemistry. Predictions were
made only where no data currently exist.
The predictions are only indicative based
on chemical structure and reaction chemis-
try. These predictions are the effects a
chemical might have on a grout and do not
necessarily represent the true field ef-
fects. Table 3 presents the matrix of
predicted polymer grout/chemical inter-
action.
In order to make these predictions, a
number of assumptions were made regarding
the characteristics of landfills, ground-
water/leachate, and polymer grouts. Land-
fills were defined as having the following
properties :
o high salt content,
o organic compound concentrations
around 1 percent, although some may
be in the ppm range,
o metal ions individually do not
exceed 1 percent, and
o pH ranges from moderately acidic
(pH 3) to moderately basic (pH 11).
The groundwater may be considered a
multicomponent dilute solution. It was
further assumed that interactions between
the components do not occur, but that
interactions between the grout and each of
the separate components may occur. These
reactions were considered to be free of
interference from the other components.
Because groundwater has no turbulence and
is slow moving, it is possible that
chemicals which are insoluble in water may
form a layer either on top or on bottom of
the groundwater depending on their densi-
ties. Although these layers would be
expected to be thin due to the low
concentration of contaminants, the point of
contact with the grout would be essentially
"pure" contaminant, rather than a very
dilute solution. This point of contact may
then represent a special case and be a
point at which an incompatibility may
result in a critical weakening of the grout
structure. For example, if a thin layer of
organic solvents formed on top of the
groundwater and the grout was solubilized
or reacted with these solvents, the struc-
ture would be breached at that point. Once
breached, the solvents could then penetrate
more deeply resulting in a general weaken-
ing or alteration of the grout matrix.
One further assumption is made with
respect to organic polymer grouts. The
reaction of the organic polymer grout and
its curing agent is not complete and a
quantity of the unreacted polymer will
remain. This assumption reflects the fact
that even under highly controlled condi-
tions, the stoichiometry of such reactions
is ill—defined, and in a field situation,
the conditions are not only uncontrolled
but many times are undefined. The unre—
acted resins are assumed to be trapped in
the polymer matrix with the reactive sites
randomly oriented. These resins will
appear on the surface as well as the
interior of the grouting wall and will be
available for reaction with chemicals
brought into contact with the surface. A
certain amount of migration from the
interior to the exterior of the matrix is
also expected to occur. Such migration is
thought to occur slowly, so the possibility
of reactions between the grouted structure
and the surroundings may continue for an
extended period of time. However, the
predicted compatibilities are made for the
totally reacted grout and do not take into
account exposure of unreacted resins to the
various chemicals.
The predicted grout/chemical inter-
actions address a number of reactions
between chemicals and grout constituents or
the set grout. These interactions include:
o destruction of grout reactive
components,
391
-------
Table 3: Predicted Grout Compatibilities
x^ Grout Type
Chemicel Group N.
Organic Compounds
Alcohols and Glvcols
Aldehydes and Ketones
Aliphatic and Aromatic
Hydrocarbons
Amides and Amines
Chlorinated Hydrocarbons
Ethers and Epoxides
Heterocyclics
Nitnles
Organic Acids and Acid
Chlorides
Organometallics
Phenols
Organic Esters
(norganic Compounds
Heavy Metal Salts and
Complexes
Inorganic Acids
Inorganic Bases
Inorganic Salts
Silicate
Polymers
Acrylamtde
u
1
0.
I
Ł
3
1
S *
5Ł
o
Q.
LU
Ł
>
o
a.
la
la
Id
3a
1d
la
Id
la
1-
la
1a
-
1-
-
1-
3d
1-
1-
1-
3-
-
3a
la
7
3b
-
-
3b
la
la
la
3-
-
-
7
-
1a
1-
-a
1-
1-
1a
la
2-
-
2-
7
i
-
la
-
1a
la
-
la
la
->
1-
1a
1 —
1a
1
la
la
la
-
la
1a
7
1-
1a
1 -
3a
1-
1a
la
la
1-
y>
\1
1d
-a
-
-
-d
-
-
-
-
-
2-
3-
-
-
-
-
-
-
-
-
-
3-
1-
-
-
3'
1-
1 -
3'-
KEY Compatibility Index
Effect on Set Time
1 No significant effect
2 Increase in set time (lengthen or prevent from setting)
3 Decrease in set time
Effect on Durability
a No significant effect
b Increase durability
c Decrease durability (destructive action begins within
a short time period)
d Decrease durability (destructive action occurs over a
long time period)
* If metal salts that are accelerators
** If metal is capable of acting as an accelerator
? Data unavailable
392
-------
o solubilization of grout, and
o alteration of grout structure.
Destruction of catalyst, activator,
inhibitor, or other additives can inhibit
polymerization and prevent proper setting
of the grout. Chemicals in the groundwater
may undergo reactions with major grout
components that are competitive with the
polymerization reaction; this may also
delay or prevent setting of the grout.
Reactions of chemicals with the set
grout can solubilize portions of the grout.
Such reaction products may be soluble in
water, organics, or both. Such dissolution
can increase the permeability of the
grouted area.
Chemical reactions with grouts and
their reactants can alter the structure of
the grout which in turn affects its
stability and permeability. Ion exchange
processes result in grout matrix components
being replaced by substances in the ground-
water. Such reactions can increase or
decrease the crosslinking within the grout
matrix. Similar results may occur through
competitive reactions of chemicals with
grout reactants. These reactions involve
direct insertion of molecules into the
grout matrix and can alter the matrix
structure by creating voids in addition to
altering the crosslinking which changes the
properties of the grout.
GROUT SELECTION
The success of a grouting operation
will depend on the selection of a proper
grouting material for the specific area to
be treated. The selection of a grout will
not only depend on its compatibility with
the leachate, but also on the physical/-
chemical properties of the grout and the
geological characteristics of the site. If
the grout is not able to penetrate the soil
voids or does not possess enough strength
to withstand the hydrostatic pressure
present at the site, then its chemical
compatibility is of little consequence.
Thus, the properties of the grout must be
matched with the geological properties of
the disposal site. This can be accomplish-
ed by a step—wise analysis of three basic
grout properties:
o injectability,
o strength, and
o durability.
By comparing each property to the condi-
tions present in the geological structure,
the proper type of grout can be selected.
The steps in this selection process are
illustrated in Figure 3.
Injectabi1ity
The injectability of a grout is con-
trolled either by its viscosity or particle
size. This property will dictate the
grout's ability to penetrate a soil/rock
structure. The lower the viscosity, the
finer the voids that can be penetrated.
Also, the smaller the particle size in
suspension grouts, the smaller the voids
that can be penetrated.
For all grouts, the closer the viscos-
ity is to that of water (1.0 cP), the
greater the penetration power. Grouts with
a viscosity less than 2 cP, such as many of
the chemical grouts, can penetrate strata
with permeabilities less than 10 cm/sec.
Higher viscosity grouts, like particulate
and some chemical grouts with a viscosity
greater than 10 cP, can only penetrate
coarse strata having permeabilities greater
than 10~ cm/sec (7,4).
Strength
Once a grout has set in the voids in
the ground, it must be able to resist
hydrostatic forces in the pores that would
tend to displace it. This ability will
depend on the mechanical strength of the
grout and can be estimated by the grout's
shear strength (10). Shear strength of a
grout will depend not only on its class,
but also on its formulation. Thus, a class
of grouts, such as silicates, can possess a
wide range of mechanical strengths depend-
ing on the concentration and type of
chemicals used in its formulation. The
strength of the gel, then, can be adjusted,
within limits, to the specific situation.
Durability
For permanent control of groundwater or
leachate movement, the grout, once it is
injected, must be able to set at the proper
rate and, once set, not deteriorate over
393
-------
All
Grout
Types
Injectability
Strength
Durability
Rejected
Grouts
\
Rejected
Grouts
Ca ididate
G routs
T
Rejected
Grouts
Figure 3: Grout Selection Process
394
-------
time. The set time can be affected by a
miscalculation of the grout's formulation,
dilution of the grout by groundwater, being
washed away in areas of high groundwater
flows, or changes caused by chemicals
contained within the grouted strata.
Deterioration of the grout can be caused by
water or chemicals structurally changing
the grout. Also, removal of water from the
grout matrix through dessication or syner-
esis can lead to shrinkage of the grout.
All of these factors can weaken a grout
leading to increased permeability.
Other Factors
Another grout selection factor that in some
cases might be considered is the toxicity
of the grout's components and the solidi-
fied grout. This factor will be important
if the aquifer with which the grout comes
in contact is a potential drinking water
source. Both the chronic and acute effects
of the compounds and the solidified grout
must be considered.
The acute effects, as measured by oral
LD values, have been determined for many
of the compounds and set grouts. The
values for set grouts are usually very
high, for example over 15,000 mg/kg in
laboratory rats for silicate grouts (10).
The compounds used in formulating the
grouts, though, usually have a lower LD^ ,
for example 150 mg/kg in laboratory animals
for the acrylamide monomer (2).
The evaluation of chronic or long term
exposure to the compounds used in grout and
examination of the types of chemicals that
could leach out of the set grout are
current fields of investigation. Several
of the compounds used in grout formulation,
such as formaldehyde and toluene diisocyan-
ate, could be potential carcinogens.
The cost of the grouting operation is
also a relevant factor that can be consid-
ered in selecting grouting as an appropri-
ate remedial action. This should include
the costs of the materials as well as the
injection costs. Current costs for grouted
soils range from $100/yd to $500/yd of
grouted soil, depending on the type of
grout used and the characteristics of the
soil (10).
CONCLUSION
While compatibility of a grout with
the chemicals present at the disposal site
is important, there are also a number of
environmental and geological considerations
that must be addressed in the selection
process. All of these factors must be
considered and any one could eliminate a
grout from usage. Unfortunately, informa-
tion on grout/chemical compatibilities is
very limited and incomplete. The form or
source of the available information did not
allow for the evaluation of the testing
procedures or the results. This led to
conflicting data and incomplete informa-
tion. As a result, the grout/chemical
compatibility matrices cannot be used as a
direct guide to grout selection, but
instead provide information on the general
compatibility trends and identify areas
where more laboratory testing is needed.
ACKNOWLEDGEMENTS
The work which is reported in this
paper was performed under Contract Number
68-03-3113 Task 40-3, "Collection of Infor-
mation on the Compatibility of Grouts with
Hazardous Waste," with the U.S. Envi-
ronmental Protection Agency, Municipal
Environmental Research Laboratory, Cincin-
nati, Ohio.
REFERENCES
1. Bowen, R. 1981, Grouting in Engineering
Practice. 2nd ed. John Wiley and Sons,
NY. pp. 218-239.
2. Berry, R.M. 1982, Injectite-80 Poly-
acrylamide Grout. In: Proceedings of
the Conference on Grouting in Geo—
technical Engineering, American Society
of Civil Engineers, NY. pp. 394-402.
3. Caron, C. 1982, The State of Grouting
in the 1980's. In: Proceedings of the
Conference on Grouting in Geotechnical
Engineering, American Society of Civil
Engineers, NY. pp. 346-358.
4. Guertin, J.D., and W.H. McTigue. 1982,
Volume 1. Groundwater Control Systems
for Urban Tunneling. FHWA/RD-81/073.
Goldberg, Zonio & Associates, Inc.
395
-------
Prepared for: U.S. Department of
Transportation, Federal Highway Adminis-
tration. Washington, DC. pp. 77-101.
5. Hayward Baker Company, EarthTech Re-
search Corporation, and ENSCO, Inc.
1980, Chemical Soil Grouting. Improved
Design and Control. Draft. Federal
Highway Administration, Fairbank Highway
Research Station, McLean, VA. pp. 5-14.
6. JRB Associates. 1982, Handbook:
Remedial Action at Waste Disposal Sites.
EPA 62S/6-82-006. U.S. EPA Municipal
Environmental Research Laboratory.
Cincinnati, OH. pp. 124-131.
7. Sommerer, S., and J.F. Kitchens. 1980,
Engineering and Development Support of
General Decon Technology for the DARCOM
Installation Restoration Program. Task
1 Literature Review on Groundwater
Containment and Diversion Barriers.
Draft. Atlantic Research Corporation.
Prepared for: U.S. Army Hazardous
Materials Agency, Aberdeen Proving Gro-
und, MD. pp. 35-65.
8. Spooner, P.A., R.S. Wetzel, C.E. Spoon-
er, C.A. Furman, E.F. Tokarski, and G.E.
Hunt. 1982, Draft Technical Handbook on
Slurry Trench Construction for Pollution
Migration Control. JRB Associates.
Prepared for: U.S. EPA Municipal
Environmental Research Laboratory, Cin-
cinnati, OH. pp. 3-8.
9. Spooner, P.A., G.E. Hunt, V.E. Hodge,
and P.M. Wagner. Collection of Informa-
tion on the Compatibility of Grouts with
Hazardous Wastes. JRB Associates.
1982, U.S. EPA Municipal Environmental
Research Laboratory. Cincinnati, OH.
pg. 2-1.
10. Tallard, G.R., and C. Caron. 1977,
Chemical Grouts for Soils. Volume II.
Engineering Evaluation of Available
Materials. FHWA-RD-77-51. Soletanche
and Rodio, Inc. Prepared for: U.S.
Department of Transportation, Federal
Highway Administration, Offices of
Research and Development. Washington,
DC. pp. 232-234.
11. Tiedemann, H.R., and J. Graver. 1982,
Groundwater Control in Tunneling.
Volume 2. Preventing Groundwater
Intrusion
Intrusion in Completed Tra isportation
Tunnels. FHWA/RD-81/074. J icobs Asso-
ciates. Prepared for: U.S. Department
of Transportation, Federal Highway
Administration. Washington DC. pp.
87-111.
396
-------
REMEDIAL ACTION RESEARCH: WASTEWATER LAGOONS
Elly K. Triegel
Joseph R. Kolmer
ABSTRACT
Remedial action research at hazardous waste sites requires assessment of
existing conditions and the design of a program which is effective in abating
current contamination and preventing future releases. This paper discusses
the efforts currently underway at the Louisiana Army Ammunition Plant (LAAP)
to close two inactive wastewater lagoons. The lagoons consist of a small
cadmium electroplating lagoon, and a pink water lagoon containing TNT-laden
wastewater and sludge. Characterization of the site conditions is near
completion, and laboratory evaluation of sludge and water treatment tech-
niques is in the initial phases. Tasks within the research project include
(1) treatment of the sludges, wastewaters and possibly soils to acceptable
contaminant levels, (2) development of safe materials handling procedures for
the TNT ana RDX sludge in the pink water lagoon, and (3) long-term monitoring
and evaluation of the effectiveness of the treatment procedures. Research
activities are being closely coordinated with the U.S. Army Toxic and
Hazardous Materials Agency.
INTRODUCTION
The principal goals of this re-
search effort are to implement and
assess the effectiveness of remedial
actions at hazardous waste lagoon
sites. Closure measures will be mon-
itored at such sites in order to
evaluate the effectiveness and ap-
propriateness of site investigation
techniques and closure methods.
Work to date on this project has
included site investigations at an
abandoned hazardous waste landfill
(Kolmer, 1981) and at the inactive
wastewater lagoons in an U.S. Army
ammunitions packing plant. In ad-
dition, several methods of lagoon
closure have been reviewed, and labo-
ratory evaluations of some of these
techniques are currently underway.
BACKGROUND
Site investigation activities
related to the development of a re-
medial action research program are
near completion at the Louisiana Army
Ammunition Plant (LAAP) , located
east of shreveport, La. (Figure 1) .
This work is being coordinated with
the U.S. Army Toxic and Hazardous Ma-
terials Agency (USATHAMA). The la-
goons to be closed at the LAAP site
include a small electroplating waste
lagoon, inactive since 1964, and a
lagoon used for the disposal of
2 ,4 ,6-Trinitrotoluene (TNT) andHexa-
hydro-1,3,5-Trinitro-1,3,5-Triazine
(RDX) contaminated process water
("pink water"). The principal effort
in this research project will be
directed to suitable closure and con-
tainment of the pink water lagoon,
which is one of sixteen such lagoons
at LAAP.
Groundwater quality analyses
have been conducted in the area of the
two lagoons by the U.S. Army Environs
mental Hygiene Agency (USAEHA, 1980).
This program detected TNT, RDX and a
number of other nitroaromatics near
397
-------
LOCATION OF THE LOUISIANA / RMY
AMMUNITION PLANT (LAAP)
FIGURE I
398
-------
the pink water lagoons. Two mon-
itoring wells were also installed in
a clay stratum near the electro-
plating waste lagoon during the same
drilling program. The concentration
of metals near the electroplating
lagoon in the groundwater was found
to be within drinking water stan-
dards, except for slightly elevated
levels of lead. The geologic units
sampled are not used as a drinking
water supply. No movement of con-
taminants beyond the property bound-
ary has been detected in the wells
south (downgradient) of the pink wa-
ter lagoons.
SITE INVESTIGATIONS UNDER THIS RE-
SEARCH CONTRACT
Field measurements of the thick-
nesses and volumes of sludge and
wastewater in each lagoon were made
to determine treatment volumes.
These results are summarized in Table
1. It was found that the greatest
thickness of sludge occurred very
near the point of discharge of wastes
into the lagoon, indicating a rapid
settling of coarser particulates in
the water. Concentration dis-
tributions of several of the con-
taminants in the sludges are similar
to the sludge thickness contours,
possibly indicating a preferential
association of these compounds with
the coarser particulates. Tetryl,
RDX, ana 2,4-Dinitrotoluene (DNT)
show this type of distribution. The
concentration of 2,4,6-TNT did not
vary in a systematic manner with the
sludge depths.
Chemical analyses of the ground-
water, lagoon water, sludge, and
soils underlying the lagoons were
performed to identify the conta-
minants present, and to characterize
certain parameters which could affect
the treatment method employed. Some
of these results are summarized in
Table 2.
Soil samples were also col-
lected from a soil boring near the
electroplating lagoon, and from
three shallow monitoring wells near
the pink water lagoon. These sam-
ples were tested for geotechnical
properties related to the remedial
work. Soils near the electro-
plating lagoon consist of at least
7 meters of red clay to silty clay.
Samples from the depth equivalent
to the bottom of the lagoon con-
tained cracks and slicken-sides
which increase the permeability (as
measured in the laboratory) to ap-
proximately 1.4 x 10 cm/sec. Im-
mediately below this sample, the
soil is similar, except that cracks
were not observed and the measured
permeability is approximately 5 x
10 cm/sec. Scarification and
recompaction of the cracked surface
clay should provide a suitable lin-
ing material for the disposal of
the solidified sludge in the cur-
rent lagoon area. Soils near the
pink water lagoon consist of fine
sands and silty clays. The geo-
technical properties of these soils
will be reviewed at a later date as
part of the final remedial plan to
determine their effect on the clo-
sure design.
Three monitoring wells, in-
stalled near the corners of the pink
water lagoon, will be used for long-
term monitoring of groundwater qual-
ity. Soil samples from beneath the
TABLE 1: APPROXIMATE VOLUMES OF SLUDGE AND WASTEWATER
Lagoon
Electroplating
Pink Water
Wastewater
Volume
510 m3
(135,000 gal)
3550 m3
(938,000 gal)
Depth
1.4 m
(4.5 ft)
1 m
(3 ft)
Sludge
Volume
59.5 m3
(15,700 gal)
365 m3
(97,000 gal)
Depth
0.06 to 0.50 m
(0.2 to 1.5 ft)
0.06 to 0.90 m
(0.2 to 4.5 ft)
399
-------
TABLE 2; SUMMARY OF ANALYTICAL RESULTS, PINK WATER AND ELECTROPLATING LAGOONS
Parameter
Pink Water
2,4,6-TNT
Tetryl
RDX
2,4-DNT
2,6-DNT
Nitrobenzene
1,3-Dini-
trobenzene
Moisture
PH
Reactivity
Electroplating
Cd, total
Cr, total
Cu, total
Pb, total
Ni, total
Zn, total
Moisture
pH
Sludge
200 to 56,700 ppm
41.8 to 2100 ppm
611 to 3740 ppm
.825 to 40.1 ppm
.106 to .901 ppm
1.090 to 9.14 ppm
.287 to 7.02 ppm
49% to 92%
7.1 to 7.5
No
2300 to 9200 ppm
7 to 1910 ppm
27 to 244 ppm
66 to 133 ppm
37 to 1370 ppm
181 to 5900 ppm
46% to 58%
8.5
Soil
Directly Beneath the Lagoon
46.2 to 67.7 ppm
NA
NA
.101 to .180 ppm
.173 ppm
1.78 ppm
.290 ppm
17% to 21%
7.5 to 8.6
NA
16 to 46 ppm
151 to 230 ppm
0.3 ppm
110 to 141 pm
52 to 113 ppm
72 to 73 ppm
20% to 25%
9.2 to 9.3
Lagoon
Water
774 to 998 ppb
NA
NA
1.1 ppb
1.1 ppb
11 ppb
1.8 ppb
NA
7.69
NA
.016* ppm
<. 025* ppm
<.025* ppm
<.20* ppm
<.10*ppm
<. 015* ppm
NA
7.69
Groundwater
2900 to 6400 ppb
NA
NA
113 to 185 ppb
10 to 19.5 ppb
11 ppb
135 to 173 ppb
NA
5.63 to 5.64
NA
.0045 ppm
.026 to .074 ppm
.0068 to .0387 ppm
.0204 to .147 ppm
.0207 to .0741 ppm
.0992 to .603 ppm
NA
7.31 to 7.32
*L)ata irom an analysis performed by the U.S. Army Environmental Hygiene Agency (May 15, 1981 correspondence).
NA: not determined
% moisture = (g water/g sample) x 100
-------
pink water lagoon are being subjected
to a deionized water leaching pro-
cedure to assess the need to treat the
soils along with the overlying
sludge.
REMEDIAL ACTION AT THE ELECTROPLATING
WASTE LAGOON
A number of different sludge
treatment techniques were evaluated
on a preliminary basis. Three of
these processes were chosen for labo-
ratory evaluation of the solidified
sludge: Chemfix, Sludgemaster, and
asphalt encapsulation. Splits of the
same sample of the electroplating
wastes were sent to the three vendors
for solidification and preparation of
proposals. Chemfix declined to bid
on the project on the basis of infor-
mation developed during this period.
The Sludgemaster and asphalt encap-
sulated samples were analyzed for (1)
total metal content, (2) leachate
quality generated under the EP tox-
icity protocol, and (3) leachate
quality generated using a deionized
water leaching procedure (Battelle
method). Results are summarized in
Table 3. Both processes produced
leachates of acceptable water qual-
ity. The Sludgemaster process was
chosen on the basis of cost and time
required to complete the solidifi-
cation. Treatment of the water cur-
rently in the lagoon will not be
required, since the water is of suit-
able quality for direct discharge.
It is anticipated that remedial
action work at this lagoon will be
implemented in January and February
1983. Subsequent work at the elec-
troplating lagoon will consist of
monitoring of the quality of the
leachate generated in the field and
comparing it with results from labo-
ratory leaching procedures.
REMEDIAL ACTION AT THE PINK WATER
LAGOON
The specific tasks required for
achievement of the research objec-
tives at the pink water lagoon in-
clude: (1) screening and handling or
removal of the sludge from the la-
goon, (2) treatment and ultimate dis-
posal of the sludge, (3) monitor-
ing/laboratory testing of the treated
sludge and leachate, and (4) carbon
treatment of the pink water and re-
generation of the activated carbon
using mobile units. Preliminary
evaluation of techniques to achieve
these goals has been completed. Pre-
vious research by USATHAMA and USAEHA
(Wentsel et al. 1981, USAEHA, 1973)
has been reviewed as part of this
assessment. Three processes were
chosen for further evaluation, as
described in more detail below (pink
Water Sludge Treatment Options).
The final choice of the reme-
dial technique will depend upon a
number of considerations. As in
the electroplating lagoon project,
both deionized water and acidic
leaching of the treated sludge will
be carried out in the laboratory in
order to evaluate the effectiveness
of the treatment process. De-
ionized water is considered to be
more representative of actual field
conditions, while the acidic ex-
traction (Toxic Extraction Pro-
cedure or TEP) is related to regu-
latory requirements for demon-
strating the nonhazardous nature of
the treated sludge with respect to
drinking water parameters. Leach-
ate quality will also be compared
to current surface water discharge
criteria for total nitrobodies to
evaluate the effectiveness of the
process in controlling leaching of
the explosives and nitroaromatics.
Other selection considerations
include the applicability of the
technique to other wastewater la-
goons, the feasibility of im-
plementing and evaluating the meth-
od, the research benefit to be
gained from implementation of the
technique, lack of duplication of
on-going USATHAMA research, costs,
and the availability of mobile e-
quipment.
In treating the sludge from the
pink water lagoon, preference will be
given to techniques which destroy the
aromatic ring without the formation
of undesirable by-products. Fixation
or encapsulation processes which
401
-------
TABLE 3: CHEMICAL ANALYSES OF SOLIDIFIED SLUDGE SAMPLES*
A. Asphalt
1. Total Metals, Solidified Sample (ug/g)
2. E.P. toxicity
a. Leachate concentration (ug/1)
b. % retained in solid
3. Battelle (distilled water)
a. Leachate concentration (ug/1)
b. % retained in solid
B. Sludgemaster
1. Total Metals, Solidified Sample (ug/g)
2. E.P. toxicity
a. Leachate concentration (ug/1)
b. % retained in solid
3. Battelle (distilled water)
a. Leachate concentration (ug/1)
b. % retained in solid
C. RCRA Leachate Criteria (ug/1)
As
<1.8
<54
**
-<54
**
4.6
<2
**
<2
**
5,000
Ba
87
4
99.9%
43
99.9%
13.5
<50
100%
<50
100%
100,000
Cd
551
19
99.9%
10
100%
494
•C20
100%
<20
100%
1,000
Cr
50
<9
100%
< 9
100%
47.4
<40
100%
<40
100%
5,000
Pb
33.8
•••39
100%
<39
100%
7.7
7
98.2%
7
98.2%
5,000
Hg
0.64
<25
100%
<25
100%
«i0.1
1.3
**
0.9
**
200
Se
not detected
note: dilution ratio in leaching tests variea from 100 g/2 liters (EP tox) to 100 g/1 liters (Battelle)
-------
solely act to reduce leaching of the
nitroaromatics would be considered
only as a second choice. The pro-
cesses currently under consideration
for this lagoon include (1) an en-
capsulation process which evolves
heat from chemical reactions, (2)
thermoplastic encapsulation, to which
heat may be added, and (3) biological
degradation using mixed cultures or
engineered microorganisms. These
techniques will be tested on a lab-
oratory and/or pilot scale prior to
final selection and implementation.
A more complete description of the
sludge and water treatment methods is
presented below.
PINK WATER SLUDGE TREATMENT OPTIONS
SLUDGEMASTER TECHNIQUE
Developers of this technique
refer to the method as a "chem-
thermo-coagulation" process. Heat
is obtained from chemical energy
released during the combination of
premixed chemicals with the wastes,
resulting in an increase in tem-
perature and removal of substantial
quantities of water in the form of
steam. The treated wastes form a
non-expansive material with low
plasticity, high compactability and
low permeability. The vendor has
had experience in the treatment of
a variety of organic and mixed
wastes. Mobile equipment is avail-
able. This process will also be
used at the electroplating lagoon.
General characteristics ot
this system are (1) the ability to
process organics and wastes con-
taining impurities (such as soil
particles, waxes, etc.), (2) good
leachate quality for the treated
products of a wide variety of waste
types, (3) the generation of high
temperatures (> 200 F) and steam in
the process, which may result in
the destruction of the aromatic
ring compounds (TNT, DNT, etc.) and
(4) ready availability of mobile
equipment and chemicals. Suita-
bility, advantages, and disadvan-
tages of this process at the pink
water lagoon have not yet been
evaluated. Such evaluation would
involve laboratory analysis of the
heat of reaction and chemical char-
acteristics of the product ma-
ter ial.
THERMOPLASTIC ENCAPSULATION
Coating of waste particles with
asphalt or other thermoplastic ma-
terials has been used in the past,
primarily for waste processing in the
nuclear industry. In this process,
the thermoplastic material is heated
and mixed with the liquid or dry
wastes using a twin-screw extruder or
other mixing equipment. The ex-
truder/mixer serves the dual purpose
of reducing particle size of the
wastes while mixing and coating it
thoroughly with thermoplastic mat-
erial. Coating of potentially ex-
plosive particles should reduce the
propagation pathway of any explosive
reaction which might occur in an
individual particle. Similar ex-
truder and mixer systems have been
used successfully in the manufac-
turing of explosives. A number of
vendors of the twin-screw extruders
and mixers exist in this country.
If a twin-screw extruder is
used, the temperature profile along
the screws (several feet in length)
can be closely controlled. The res-
idence time in each section of the
screw can be varied by changing the
screw configuration and the speed of
rotation. If a batch mixer is used,
the operation may be stopped at any
point to perform procedures such as
the decanting of excess water. In
either case, water and volatile or-
ganics are evaporated and collected
in a condensor system or removed
under a vacuum. The necessity for
controlling any air emissions would
be evaluated in the laboratory test-
ing phase. The process generally
results in a substantial volume re-
duction, due to the loss of water, and
a product which is resistant to
weathering and cracking.
Currently, an evaluation of the
types of mixers and asphaltic mat-
erial readily available on the market
is being conducted. A broad range of
asphaltic materials with different
403
-------
handling characteristics and physical
properties (e.g. melting point, vis-
cosity) are available. These charac-
teristics are being matched with mix-
ing equipment requirements in order
to evaluate (based on published in-
formation) the optimum system for
mixing, coating and heating the pink
water sludge in a safe manner. This
evaluation will be completed in Janu-
ary 1983 and a decision made as to any
subsequent laboratory work. phase
two, if conducted, would consist of
laboratory evaluation of the effec-
tiveness of the system, safety, de-
sign of controls, and a differential
thermal analysis of TNT breakdown
temperatures.
Characteristics of thermoplastic
encapsulation are (1) ability to pro-
cess a very wide range of substances
(limited only to corrosion and ab-
rasion of the extruder), (2) good
leachate quality of processed wastes,
(3) the ability to reach and maintain
high temperatures, with a high degree
of control, and (4) successful use of
the equipment in processing explosive
compounds.
BIODECONTAMINATION
USATHAMA and others have con-
ducted literature reviews and labo-
ratory experiments related to the
decomposition of TNT and RDX in bio-
logical treatment systems (Isbister
et al., 1980, Osmond and Andrews,
1978, McCormick et al., 1981, Kaplan
and Kaplan, 1982). Results indicated
that, in one-step biological systems,
RDX is degraded but the TNT ring
either is not broken or forms un-
desirable by-products. it has been
observed, however, that in the la-
goons TNT tends to convert to other
aromatics (such as 1,3,5 trinitro-
benzene) which may be more readily
degraded. A multi-step process, con-
verting TNT to intermediates (trini-
trobenzene, nitrobenzoic acid, etc.)
which can then be degraded to CO_ and
water, has not been previously in-
vestigated. Formation of degradable
intermediates may be feasible by us-
ing controlled chemical reactions, by
mixed cultures of mutated micro-
organisms, or by alternating treat-
ment conditions.
A substantial amount of research
and field experience has been gained
in the use of mixed microfcial cul-
tures (particularly in the degrada-
tion of spilled organics) since the
publication of the USATHAMA research
on TNT degradation. Mixed cultures
may successfully result in biodecon-
tamination by providing different
strains for each reaction step in the
process, where one type alcne would
not succeed. These cultures are
developed using a combination of
microbial mutation and selection of
species which are able to survive by
metabolizing the waste substrate.
A preliminary laboratory study
is currently underway to study mic-
robial breakdown of TNT. Agar plates
are sprayed with extract .from the
sludge from the study lagoor and in-
noculated with either (a) soil mic-
robes found at the LAAP TNT lagoons
and cultured in the labora:ory, or
(b) a manufactured bacterial product
(Hydrobac") containing adapted mutant
microbes with a capability for de-
grading a number of organic chemi-
cals. Test conditions are described
by the matrix in Table 4. Nitrogen
starvation (all other nutrients sup-
plied) will be tested in an effort to
force the microbes to use the nitro-
aromatics as a nitrogen source. The
pH 8.5 case is being evaluated since
TNT is somewhat less stable chemi-
cally at higher pH levels. Normal pH
and nutrient additions will also be
assessed in case ideal conditions are
needed to initiate and sustain growth
until TNT degradation occurs . Tolu-
ene additions act to provide a come-
tabolite (i.e. a structure similar to
TNT) , which the microbes can adapt to
more readily. Surfactant additions
act to increase surface areas for
microbial reactions.
At the end of a two week period,
the plates will be inspected for
microbial growth. Those showing sig-
nificant growth will be analyzed for
TNT degradation by comparing the ex-
tract from areas containing growth
with no-growth areas of the same
plate. If TNT loss is observed, a
soil degradation experiment which
more closely approximates fisld con-
ditions will be implemented.
404
-------
TABLE 4: AGAR TEST MATRIX, BIODEGRADATION STUDY
Hydrobac"
Toluene
Surfactant
Field Culture
Toluene
Surfactant
NH--N Added
pH = 7.0
X
X
X
X
pH = 8.5
X
X
X
X
Nitrogen starved
pH = 7.0
X
X
X
X
pH = 8.5
X
X
X
X
Characteristics of microbial
systems include (1) reduced materials
handling (in-place treatment may be
feasible) , (2) low costs due to low
equipment costs and ease of opera-
tion, (3) high system mobility, and
(4) lower volumes of by-products and
residuals which must be landfilled
(assuming complete degradation is
possible).
MOBILE WATER TREATMENT SYSTEM
The mobile treatment system has
been developed and will be operated
by the EPA Oil and Hazardous Mate-
rials Spills Branch at Edison, New
Jersey. The system consists of units
for flocculation, sedimentation, fil-
tration, and carbon adsorption. De-
pending upon the treatment require-
ments, contaminated water can be
pumped into a settling tank where
flocculation and sedimentation occur.
The clarified fluid is then passed
through multi-media filters before
entering the carbon adsorption col-
umns. Sludge is removed from the
sedimentation tank and can be stored
for treatment with the rest of the
lagoon sludge. Treatment schemes can
be varied as needed. If required,
additional storage tanks are provided
for filter backwashing or temporary
storage of unprocessed materials.
The Mobile Physical-Chemical
Treatment Trailer is mounted on a
13.7-m (45-ft) trailer and incor-
porates three multi-media filters,
three pressure carbon columns (which
may be used in parallel or in series) ,
pumps, piping, controls, and a 100-kW
diesel generator. A support trailer
is equipped with additional pumps,
fittings, and several collapsible
rubber tanks, which permit the treat-
ment trailer to be located up to 150-m
(500-ft) from the lagoon. Contami-
nated water can be processed at flow
rates up to 2,270 1pm (600 gpm).
MONITORING PROGRAM
The monitoring well network
has been installed at the pink
water lagoon, and will be used in
conjunction with the previously in-
stalled wells to determine baseline
conditions and water quality over a
period of time. Leachate col-
lection systems will be installed
in both lagoons at the time of the
treatment of the sludges. The
leachate collection system at the
M-4 lagoon will consist of a gravel
blanket, sloped such that leachate
can be collected in a sump at the
low point of the disposal area. The
pink water lagoon system will be
designed after selection of the
remedial technique.
405
-------
Post closure monitoring of the
effectiveness of the remedial action
will continue for a period of three
years. The design of the monitoring
and analytical plans will be de-
pendent upon the nature of the reme-
dial plan implemented. Destruction
of explosive and toxic compounds in
the treatment process would reduce
the amount of monitoring required at
the pink water lagoon. In that case,
effectiveness of the system will be
evaluated primarily by laboratory
analysis of the product for nitro-
aromatics and/or other toxic ma-
terials. The objectives and scope of
testing will be formulated jointly by
the EPA and U.S. Army.
SUMMARY
The LAAP project, currently in
the laboratory evaluation phase, will
provide information on site inves-
tigation techniques, remedial action
implementation and long-term moni-
toring. Laboratory testing related
to the pink water lagoon and im-
plementation of remedial action at
the electroplating lagoon will be
completed in early 1983.
The long-term commitment at this
site will allow the controlled in-
stallation of remedial measures and
sufficient post-installation moni-
toring to adequately assess the per-
formance of the entire remedial sys-
tem. The controlled, rather than
emergency, nature of these inves-
tigations will result in a more or-
dered and complete evaluation of the
various aspects which must be con-
sidered in the closure of an inactive
Disposal site.
REFERENCES
and
En-
1. isbister, J.D., R.C. Doyle
J.F. Kitchens. 1980,
gineering and Development Sup-
port of General Decon Tech-
nology for the DARCOM Instal-
lation Restoration Program, Task
6; Adapted/Mutant Biological
Treatment, U.S. Army Toxic and
Hazardous Materials Agency,
Edgewood, Md.
Kaplan, D.L. and A.M. Kaplan.
(1982) Thermophilic Transfor-
mation of 2,4,6-TNT in Com-
posting Systems,Nstick/TR-
82/015, U.S. Army Natich Res.
and Dev. Lab.
Kolmer,J.R. 1981,Investigation
of the LiPari Landfill using
geophysical techniques, inProc.
of the 7th Annual Research Sym-
posium, U.S.E.P.A.
McCormick, N.G., J.H. Cornell,
A.M. Kaplan (1981) The Biode-
gradation of Hexahydro-1,3,5-
Tr initro-1,3 ,5-Tr xazine (RDX) ,
Natick/TR-81/020, U.S. Army Na-
tick Research and Development
Lab.
Osmon, J.L. and C.C. Andrews
(1978) The Biodegradation of TNT
in Enhanced Soil and Compost
Systems, ARLCD-TR-77032U.S. Ar-
my Armament Research and De-
velopment Command, Dover, N.J.
U.S. Army Environmental Hygiene
Agency. 1973 , Carbon Column Sys-
tem Removal Efficiency Study,
StudyNo. 24-033-73/74,Aberdeen
Proving Ground, Md.
U.S. Army Environmental Hygiene
Agency. 1980,Geohydrologic Con-
sultation No. 31-24-0152-80,
Louisiana, Army Ammunition
Plant,Shreveport,La., Aberdeen
Proving Ground, Md.
Wentsel, R.S., S. Sommerer,
J.F.Kitchens. 1981,Engineering
and Development Support of Ge-
neral Decon Technology for the
DARCOM Installation Restoration
Program, Task 2; Treatment of
Explosives Contaminated Lagoon
Sediment -phase 1: Literature
Review and Evaluation, U.S. Army
Toxicand Hazardous Materials
Agency, Edgewood, Md.
Wentsel, R.S., S . Sommerer , J.F.
Kitchens. 1981, Engineering and
Development Support of General
Decon Technology for the DARCOM
Installation Restoration pro-
406
-------
gram, Task 2; Treatment of
Explosives Contaminated Lagoon
Sediment -Phase II; Laboratory
Evaluation, U.S. Army Toxic and
Hazardous Materials Agency, Edge
wood, Md. 21010.
407
-------
A LITERATURE SURVEY OF THREE SELECTED HAZARDOUS WASTE DESTRUCTION TECHNIQUES
Danny D. Reible and David M. Wetzel
Department of Chemical Engineering
Louisiana State University
Baton Rouge, Louisiana 70803
ABSTRACT
In order to provide direction to the LSU Center for Hazardous Waste Research, a
literature survey of current and proposed means for the destruction of hazardous wastes
was conducted. As part of this study, three processes were identified as being
applicable to spill and dump site wastes in which the wastes are present in
concentrations in excess of 1 percent but are sufficiently dilute to preclude
incineration without separation. These applicable processes, in-situ biological
degradation, co-treatment in industrial facilities, and wet-air oxidation, are described
and potential research areas to prove their viability are outlined.
INTRODUCTION
Largely neglected in the past, the
problems of proper containment and/or
complete destruction of hazardous
chemical wastes have, in recent years,
jumped to the forefront of the nation's
environmental problems. The passage of a
law creating the "Superfund" for
hazardous waste cleanup, and the
emotional response of the people affected
by such contamination problems as Love
Canal in New York, are indicative of the
national concern over hazardous chemical
wastes.' As a response to these problems,
the EPA has funded a significant amount
of research into both fundamental and
applied problems of waste generation,
containment and destruction. As part of
this research program, the EPA has funded
the development of hazardous waste
research centers at three universities.
The newest of these centers is located at
Louisiana State University, Baton Rouge,
Louisiana. Each of the centers is
charged with the responsibility of
conducting the fundamental research
necessary to support the development of
hazardous waste control or destruction
strategies. At this, the beginning of
the second year of operation of the LSU
Hazardous Waste Research Center, the
specific direction this research effort
should take are still under development.
In order to provide a basis for selection
of the center's direction or directions,
a literature survey and analysis were
undertaken whose objective was
identification of the most feasible waste
destruction processes and the research
required to prove viability. At the
outset, emphasis of this study was upon
the ultimate disposal or destruction of
hazardous waste rather than containment
or separation. In addition, due to the
expense and relative maturity of
incineration processes, the emphasis of
this study was on alternative (i.e.,
alternative to incineration) or novel
means of waste destruction. This report
is a summary of some of the results of
this study.
Summarized in this paper are
processes that exhibit, at least to the
authors, significant potential for the
disposal of wastes representative of
spills or abandoned dump sites. Thus the
waste will be assumed to contain a
variety of chemical compounds at
concentrations of a few percent.
Incineration would be appropriate for
concentrated wastes, while a variety of
processes (see full report, [22]) would
be appropriate for dilute wastes (taken
to imply waste concentration <1%). This
408
-------
paper will identify destruction processes
that can be employed on predominantly
organic wastes in the intermediate
concentration range.
The identification and assessment of
individual processes were based on
literature reports of existing research.
A computerized search of NTIS, Chemical
Abstracts, EPA documents, and other
environmentally related databases was
instituted, and appropriate book, journal
articles or reports identified and read.
In addition, potential processes were
identified and discussed with the
Chemical Engineering Faculty at LSU, as
well as government and industrial
visitors to the LSU Center for Hazardous
Waste Research. Information was also
gathered from papers presented and in
discussion with conference attendees at
the Cleveland and Los Angeles national
meetings of the American Institute of
Chemical Engineers.
As a result of these discussions and
surveys, the following processes were
judged to exhibit the most short-term
potential for the destruction of spill
and dump site wastes:
1) biological degradation via land
treatment,
2) co-treatment of waste in
industrial processes,
3) wet-air oxidation, including
supercritical process.
Other processes investigated which were
judged appropriate for some waste types
included UV-assisted ozonation,
chlorinolysis, microwave destruction, and
molten sodium destruction. Due to
physical and economic limitations on
production of ozone, ozonation is
apparently not a viable destruction means
for wastes present at concentrations in
excess of approximately 1 percent [4].
Chlorinolysis is applicable to
halogenated wastes and apparently most
appropriate as an end-of-pipe technology
[4,18]. Microwave and molten sodium
destruction of waste require
significantly more research before their
applicability can be proven but, at the
present time, appear to be limited to
small amounts of concentrated wastes
[4,18].
BIOLOGICAL DEGRADATION VIA LAND TREATMENT
Land treatment, or soil incorpora-
tion, is an ancient method of soil
fertilization [11], which is currently
used to dispose of 25-30% of the sludge
produced by municipal biological
treatment processes [8], and perhaps as
much as half of all refinery wastes [20].
The application of these techniques for
the in-situ detoxification of a spill
site has been investigated by Thibault
and Elliot [19] and Johnson, et al. [7].
These investigators showed that
degradation of the spilled chemicals is
possible through application of the
appropriate microorganism(s) and the
necessary nutrients, and by providing
aeration through plowing. The
decomposition process is related to that
which occurs in composting of wastes in
piles, windrows or silos [4]. The
appropriate microorganism can be selected
on the basis of cataloged treatability
[17], or developed for the specific waste
by adaptation, mutation or genetic
engineering [14]. The nutrients required
include nitrogen and phosphorus, which
are not normally available at sufficient
concentrations at spill sites [19], and
oxygen for the microorganisms through
aeration. The significant advantages of
land treatment of contaminated soils are
summarized in Table 1.
Table 1. ADVANTAGES OF IN-SITU
BIOLOGICAL DEGRADATION OF SPILL AND
DUMP SITE WASTE
No soil removal or transportation
costs
Minimal operation and maintenance
costs
Proven technology for municipal
sludge disposal and other wastes
In order to apply j.and treatment
techniques, however, several problems
must be solved. The disadvantages or
potential problems associated with the
in-situ biological degradation of spill
and dump site wastes are included in
Table 2. Table 3 suggests appropriate
research that may allow solution of these
problems. Improvements in the rate and
overall effectiveness of microbial
destruction by genetic engineering is
currently an active area of research
[7,14,17,19]. Proper aeration of the
soil is an important constraint on the
409
-------
Table 2. DISADVANTAGES OF IN-SITU
BIOLOGICAL DEGRADATION OF SPILL
AND DUMP SITE WASTES
Expenses of microorganism development
(if required)
Treatment time of weeks to months
Generally incomplete destruction of
the waste, especially for certain
compounds
Difficulty of effective soil aeration
Requires containment of the waste
during treatment
Potential air emissions problem
Public relations problems due to lack
of an immediate response
Table 3. APPROPRIATE RESEARCH FOR
IMPROVING IN-SITU BIOLOGICAL DEGRADATION
OF SPILL AND DUMP SITE WASTES
Investigation of methods for
enhancing oxygenation of the soil
(e.g. peroxide addition)
Development of genetically engineered
microorganisms suitable for rapid and
complete destruction of the spilled
compounds.
Development of spill containment
methods.
effectiveness of land treatment [19].
Recent research has indicated that
hydrogen peroxide addition to the soil
may provide an oxygen source without
mechanical aeration [5]. Containment
methods must eliminate ground-water
contamination, leaching of the soil, and
air pollution problems due to waste
volatization [13,20]. This is a difficult
problem in view of the length of time
currently required to treat wastes via
biological degradation. Resolution of
these problems, however, will yield an
effective means of destroying hazardous
wastes without site removal.
CO-TREATMENT OF HAZARDOUS WASTES IN
INDUSTRIAL FACILITIES
In an attempt to reduce the cost of
separate treatment of hazardous wastes,
co-treatment in industrial facilities has
been suggested. The presumption is the
through the appropriate choice of process
and means of introduction, the wastes
will be destroyed without significantly
affecting the process. Such a waste
destruction process would have the
benefits listed in Table 4.
Table 4. ADVANTAGES OF CO-TREATMENT
OF HAZARDOUS WASTES IN INDUSTRIAL
FACILITIES
Existing facilities
Essentially complete destruction in
appropriate process
Potentially little effect on existing
operation
Perhaps the best example of a co-
treatment process is the burning of
chemical wastes as fuels in cement kilns.
Cement kilns (wet process) are designed
to produce a calcium silicate based
substance called "clinker". Typical
operating conditions are a maximum
temperature of 1400-1500°C, a solids
residence time of 1-4 hours and a gas
residence time of 10 seconds or more
[10]. Under these conditions, volatiles
in the slurry are driven off, pyrolyzed
and then oxidized, and the remaining
slurry dried and calcined. Calcination
is a solidification process that occurs
through heating without significant
interaction with the gaseous phase [4].
Although basically an incineration
process, cement kiln co-treatment is
included here because the primary
objective remains production of the
clinker and not destruction of the
wastes. Lauker [10] has indicated,
however, that essentially complete
destruction is possible due to the high
temperatures and long residence times,
and the heat of combustion may even
result in a fuel savings to the kiln
operator.
Although cement kiln waste
destruction is the most studied process
for co-treatment, other industrial
calcination processes, such as the coking
of heavy petroleum residues and tars, may
be appropriate. Petroleum residue coking
is conducted at 415-450°C and 3-6 atm
during a semi-continuous process,
delayed coking, and at 480-565°C and
atmospheric pressure during fluidized bed
coking [4] . The principal high-value
outlet for coke is in electrodes for
aluminum production, a use which requires
low metals and sulfur content [21]. The
destruction of hazardous wastes in
refinery coking units may be limited by
the low temperatures of the process and
by the metals and sulfur content
requirements of the product coke.
410
-------
Aaother refinery alternative for co-
treatment is hydrotreating or hydrofining.
Hydrocracking is a high pressure (100 atm)
and moderate temperature (300-400°C)
process used to convert heavy gas oils
into lighter, more valuable products
(e.g. gasoline) [21]. Hydrofining
describes a number of processes that are
designed to hydrogenate, but not crack,
organic feedstocks [21]. Its primary
application is in the desulfurization,
denitrogenation and demetalization of
poor quality crude oils [l] and
synthetic (i.e. coal and shale based)
oils [12]. In many cases, synthetic
crude oil poses many of the same fouling
and pumping problems that are also posed
by hazardous waste mixtures. It may
prove necessary, however, to physically
separate water and inert solids from the
waste before introduction to the hydro-
treating unit (which would also make
incineration an appropriate destruction
method).
Other co-treatment processes for
specific waste streams have been
suggested [4]. All of the possible co-
treatment processes, however, have the
potential disadvantages listed in Table
5. In order to offset these
disadvantages, appropriate areas for
future research are indicated in Table 6.
It is believed, however, that the most
substantial barriers to widespread use of
co-treatment is resolution of the social
and legal implications of third party
treatment in facilities not specifically
designed for wastes.
Table 5. DISADVANTAGES OF CO-TREATMENT
OF HAZARDOUS WASTES IN EXISTING
INDUSTRIAL FACILITIES
Handling and transporation of wastes
required
Physical separation of waste may be
required for a particular process
Question of liability for waste in
the event of accidents
Little incentive for operators of
existing facility
Potential contamination of existing
product
WET-AIR OXIDATION
One of the major costs of the
incineration and co-treatment processes
is for energy. This cost can be greatly
Table 6. APPROPRIATE AREAS OF RESEARCH
FOR IMPROVING CO-TREATMENT OF HAZARDOUS
WASTES IN EXISTING INDUSTRIAL FACILITIES
Identification of appropriate
processes for particular wastes
Accurate assessment of destruction
and/or conversion efficiency and
associated pollutant emissions
Identification of form and
significance of product contamination
in particular processes
reduced by not vaporizing the waste
material, and by carrying out the
reaction at reduced temperature. The wet-
air oxidation process, or oxidation of
organics by oxygen dissolved in water
either at subcritical [15,16] or
supercritical [2,9] conditions provides
both of these reaction condition
advantages. The subcritical conditions
are typically below 300°C and 20
megapascals (MPa), while the super-
critical reaction is carried out above
374°C and 22 MPa.
The only reactant required, other
than the waste itself, is air; and the
only solvent, water. Thus no additional
hazards are introduced in the treatment
process. The advantages of this process
are listed in Table 7, and the dis-
advantages in Table 8.
Table 7. ADVANTAGES OF WET-AIR OXIDATION
OF HAZARDOUS WASTES
Energy efficient - avoids phase
changes
Does not require additional chemicals
Proven commercially - Zimpro process
Table 8. DISADVANTAGES OF WET-AIR
OXIDATION OF HAZARDOUS WASTES
Organics exhibit varying degrees of
reactivity.
Complete destruction requires
substantial temperature and pressure,
or catalyst.
Unproven in three phase systems -
air, water-organic, soil.
Some compounds difficult to destroy.
As the current disadvantages of the
process include incomplete destruction of
daughter compounds and refractory
materials [6], the reaction rates of
these materials provide an active area
411
-------
for research [3]. Disposable catalysts 4.
and free radical production from the
waste itself are possible solutions to
this problem.
In the long run, the process must be
applicable to destruction of hazardous 5.
waste from spill or dump sites. We
envision contaminated soil being
introduced directly into the process.
The soil provides a third phase which
complicates the process by adding a mass 6.
transfer step. The combined desorption-
reaction problem should be an area of
active research.
Appropriate areas of research for
wet-air oxidation are summarized in Table 7.
9. We believe this process has great
potential for treatment of hazardous
wastes at the site of a spill or dump.
The detoxified soil would then be a more
easily accepted material, whether it
remains at the site or is transported to
a new location.
Table 9. APPROPRIATE AREAS OF
RESEARCH FOR IMPROVING WET-AIR 8.
OXIDATION OF HAZARDOUS WASTES
Identification of reaction conditions
necessary to achieve complete
destruction of each type of waste
Evaluation of non-thermal means of
enhancing degradation rate or
effectiveness, including addition of
catalysts, oxidizing agents, or free-
radical initiators
Demonstration of oxidizing capability
in the presence of an inert solid
phase
Fundamental investigation of mass
transfer-reaction in three phase
systems (solid-liquid-gas).
REFERENCES
1. Aalund, L.R., Chevron to tame tough
crude at Pascagoula with rapid
desulfurization and coking. Oil
and Gas Journal, March 30, 1981.
2. Anon., Using Supercritical Water to
Destroy Tough Wastes. Chemical Week, 13.
p 26, April 21, 1982.
3. Baillod, C.R., B.M. Faith and 0.
Masi, Fate of Specific Pollutants
During Wet Oxidation and
Ozonation, Environmental Progress,
1, pp 217-226 (1982).
Berkowitz, J.B., J.T. Funkhouser and
J.I. Stevens, Unit Operations for
Treatment of Hazardous Industrial
Wastes.
Noyes
9.
10.
11.
12.
Park Ridge, N.J.
Data Corporation,
(1978).
Conway, R.A., Union Carbide
Corporation, P.O. Box 8361, South
Charleston, W. Va. 25303. Personal
Communication (1982).
Day, D.C., R.R., Hudgins and P.L.
Silveston, Oxidation of propionic
acid solutions. Canadian Journal of
Chemical Engineering, 51, pp 733-740
(1973).
Johnson, L., J. Kauffman and M.
Krupka, Application of mutant
bacteria for the in-situ treatment
of land based organic chemical
spills and contaminated groundwater.
Summer National Meeting of the
American Institute of Chemical
Engineers, Cleveland, Ohio, August
29-September 1, 1982.
Jones, J.L., P.C. Bomberger, F.M.
Lewis, and J. Jacknow, Municipal
sludge disposal economics.
Environmental Science and Technology,
U, 10, pp 968-972 (1977).
Josephson, J., Supercritical Fluids.
Environmental Science and Technology,
16, pp 548A-551A (1982).
Lauber, J.D., Burning chemical
wastes as fuels in cement kilns.
Journal of the Air Pollution Control
Association, 3:2, 7, pp 771-777.
Maugh, T.H., Hazardous wastes
technology is available, Science,
204, 1, pp 930-933 (1979).
O'Rear, D.J., R.F. Sullivan and B.E.
Strangeland, Catalytic upgrading of
H-coal syncrudes. Presented at the
Symposium on Coal Liquids Upgrading,
179th ACS Meeting, Houston, Texas,
March 23-28, 1980.
Overcash, M.R., K.W. Brown, L.E.
Devel and W.L. Nutter, Use of land
treatment technology for ultimate
disposition of organic hazardous
materials spills. Control of
Hazardous Material Spills,
Proceedings of the 1980 National
Conference, pp 393-397.
412
-------
14. Pierce, G.E., Genetically engineered
microorganisms for degradation of
hazardous organic wastes. Summer
National Meeting of the American
Institute of Chemical Engineers,
Cleveland, Ohio. August 29-September
1, 1982.
15. Randall, T.L., Wet Oxidation of
Toxic and Hazardous Compounds,
Zimpro Technical Bulletin 1-610.
16. Randall, T.L. and P.V. Knopp,
Detoxification of Specific Organic
Stustances by Wet Oxidation. Journal
of the Water Pollution Control
Association, 52, 8, pp 2117-2130
(1980).
17. Sims, R. and M. Overcash, Catalog of
Organic Chemicals - behavior in
terrestial systems, Biological and
Agricultural Engineering, North
Carolina State University (1979).
18. Sworzyn, E.M. and D.G. Ackerman,
Interim guidelines for the disposal/
destruction of PCB's and PCB items
by non-thermal methods. EPA Contract
No. 68-02-3174, USEPA, Washington,
D.C. 20460 (1981).
19. Thibault, G.T. and N.W. Elliot,
Biological detoxification of
hazardous organic chemical spills.
Control of Hazardous Material Spills,
Proceedings of the 1980 National
Conference, pp 398-402.
20. Thibodeaux, L.J. and S.T. Hwang,
Landfarming of petroleum wastes -
modeling the air emission problem.
Environmental Progress, !_, 1, pp
42-46 (1982).
21. Voorhies, A., Jr., Petroleum Refining
Technology, Department of Chemical
22. Wetzel, D.M. and D.D. Reible,
Research Planning Study - Alternate
Methods for Hazardous Waste
Treatment/Destruction. Final Report
to LSU Hazardous Waste Research
Center, Louisiana State University,
Baton Rouge, Louisiana (1982).
413
-------
DRUM HANDLING PRACTICES
AT ABANDONED SITES
Roger Wetzel
Kathleen Wagner
JRB Associates
McLean, Virginia
Anthony N. Tafuri
U.S. Environmental Protection Agency
Municipal Environmental Research Laboratory
Edison, New Jersey
ABSTRACT
The objectives of this project are to document drum handling techniques, identify
research and development needs, and develop new or improved techniques. The aspects
of drum handling that were addressed include the following: detecting and locating
drums; determining drum integrity; excavating and staging; segregation by waste type;
consolidation or overpacking; temporary storage and shipping and decontamination.
A number of approaches or combination of equipment and techniques are applied to
a site based primarily on considerations for safety of field personnel, potential
for uncontrolled releases of wastes, site specific factors affecting equipment perfor-
mance and costs for alternative approaches.
INTRODUCTION
Land disposal of drums containing
hazardous wastes has been a widespread
practice throughout the United States.
Many problems at abandoned waste disposal
sites have been attributed to drum dis-
posal. The results of a 1980 survey of
remedial actions at 169 sites conducted by
the United States Environmental Protection
Agency (USEPA) indicates that about one
quarter of all abandoned sites have major
drum related problems •
Past drum disposal practices have
resulted in many instances of groundwater,
air, soil and surface water contamination.
At some sites explosions and fires have
resulted from reactions of incompatible
wastes which leaked from drums.
The cleanup of waste sites containing
drums poses several hazards to field
workers and to the environment. In many
instances the integrity of the drums is
poor and they are prone to rupture and
leakage. The drum contents are frequently
unknown and it is not unusual to find
drums containing incompatible wastes in
the same pit or trench. The hazard of
cleaning up these waste drums is exacer-
bated in many instances, by the unsuitable
site conditions in which they are found.
Drums containing hazardous wastes have
been found in a variety of environments
including wetlands; the banks of rivers
and streams; and highly industrialized,
confined areas.
Because of the safety and environ-
mental hazards involved in handling drums
and the widespread efforts currently
underway to clean up hazardous waste
sites, the USEPA Solid and Hazardous Waste
414
-------
Research Division has undertaken a re-
search effort to evaluate drum handling
practices and improve upon the existing
equipment and methods where safety and
effectiveness are seriously lacking.
As a means of documenting the state
of the art and identifying research needs,
a manual reviewing the applicability,
advantages and disadvantages of equipment
and methodologies for handling drums was
developed . This manual addresses the
following activities:
o Detecting and locating drums
o Determining drum integrity
o Excavation and on-site transfer of
drums
o Recontainerization and consoli-
dation
o Storage and shipping.
The selection and implementation of
equipment and methods for handling drum
related problems is made on a site-
specific basis. Site-specific factors
which affect the suitability of various
equipment and methods include: the number
of drums present at a site, depth of
burial, accessibility of the site, inte-
grity of the drums, site drainage, and
types of waste present.
The typical sequence of activities
involved in a site cleanup operation can
logically be divided into five phases:
I. Preliminary Site Assessment
II. Field Investigation and Drum De-
tection
III. Development of a Cleanup Strategy
IV. Site Cleanup
V. Temporary Storage and Transporta-
tion
during this phase to reduce safety hazards
and the cost and time required for
subsequent activities. Background data on
the site can often provide a great deal of
insight on the types of wastes, condition
of the drums and site specific variables
which will influence the selection of
equipment.
As with Phase I, Phase II, Field
Investigation and Drum Detection, relies
upon a considerable amount of field work
that has generally been completed by the
time the drum handling operation is begun.
Drum detection using aerial photography,
geophysical surveying and soil sampling
are examples of activities that are
conducted in Phase II.
Phase II, Development of a Cleanup
Strategy, provides an overview of the four
major variables which effect the selection
of equipment and methods for handling
drums:
o Safety precautions
o Protective and mitigative measures
for environmental releases
o Site specific conditions as they
influence equipment selection and
performance
o Cost.
Phase IV considers equipment and
methods for site cleanup after drum
removal in terms of their applicability,
safety, advantages, limitations and re-
lative costs. Equipment and methods for
temporary storage and transportation of
wastes from drum cleanup operations are
considered in Phase V.
The drum handling manual focuses on
equipment, methods and protocol that are
used for activities undertaken during
Phase III through V. Phase I, the
Preliminary Site Assessment is generally
completed by the time the drum handling
operation begins and is not considered in
detail. Nevertheless, the cleanup con-
tractor, and Federal and state officials
will need to draw on information gathered
The drum handling manual emphasizes
both safety measures and measures for
preventing or mitigating environmental
releases. Tables 1 and 2 summarize these
measures.
Although there are a variety of
equipment types and methods available for
handling drums at abandoned waste sites,
there are several areas where the safety
415
-------
Drum Handling Activity
Potential Safety Hazard
Safety Precautionm
Looting Drums
Determining Drum
Integrity
Drum Excavation and Handling
Unknown location and contents of • In conducting geophysical surveys, use
drums can lead to unsuspected hazards hand held instruments rather than
vehicle mounted systems where drums
are close to the surface
• Conduct soil sampling only after the
geophysical survey is completed to
minimize the possibility of puncturing
drums
• Use non-sparking tool* for sampling
• Use direct-reading, air monitoring
equipment to detect hot spots where
contamination may pose a risk to
worker safety
Process of visual inspections require • Any drum which is critically swollen
close contact with drums of unknown should not be approached; they should
content be isolated using a barricade until
the pressure can be relieved
remotely
* Approach drums cautiously, relying on
air monitoring equipment to indicate
levels of hazards which require
withdrawal from working area or uae of
additional safety equipment
Exposure to toxic/hazardous vapors: • Where buried drums are suspected,
rupture of drums, conduct a geophysical survey before
using any construction equipment to
minimize the possibility of rupture
t Use the drum grappler where possible
and cost effective, to minimize close
contact with the druma
• If the grappler is not available, pump
or overpack drums with poor integrity
• Use non-sparking hand tools and non-
sparking bucket teeth on excavation
equipment
• Where slings, yokes or other
accessories must be used, workers
should back away from the work area
after attaching the acceasory and
before the drum is lifted
• Critically swollen drums should not be
handled until pressure can be relieved
• Use plexiglas shields on vehicle cabs
• Use "morman bars" which fit over the
teeth of excavation buckets to prevent
drum puncture
416
-------
Drum Handling Activity
Potential Safety Hazard
Safety Precautions
Drum Excavation
(Con't)
Use direct-readingi air Monitoring
equ Lpotent when in cloae proximity to
drums to detect any hot spot*
Drum Opening
Release of toxic, hazardous
vapors, rupture of drums
Staging Recontainerization
Mixing of incompatible wastes
Storage and Transportation
Mixing of incompatible wastes
» Use remote drum opening Methods
where drums are unsound
i Conduct remotely operated drum opening
from behind a barricade, or behind a
plexiglas shield if backhoe nounted
puncture is being used
* Isolate drum opening from other
activities to minimize a chain
reaction if an explosion or reaction
did occur
> Use only non-sparking hand tools if
drums are to be opened manually
' Remotely relieve the pressure of
critically swollen drums before
opening
1 Clean up spills promptly to minimize
mixing of incompatible materials
• Allow adequate spacing during staging
of drums to provide rapid exit in
case of emergency
* Perform on-site compatibility testing
on all drums
• Drums which are leaking or prone to
leakage or rupture should be
overpacked or the contents transferred
to a new drum before staging
• Clean up spills promptly to avoid
mixing of incompatible wastes
• Intentional mixing of incompatible
wastes such as acids and bases should
be performed under controlled
conditions in a reaction tank where
temperature and vapor release can be
monitored
» Monitor for hot spots using direct-
reading air monitoring equipment
* Segregate incompatible wastes using
dikes
* Maintain weekly inspection schedule
• Allow adequate aisle space between
druBB to «1 low rapid ex it of worker a
in case of emergency
Storage and Transportation
(Con't)
* Clean up spills or leaks promptly
• Ensure adherence to DOT regulations
regarding transport of incompatible
wastes
417
-------
TABLE 2. MEASURES FOR MINIMIZING ENVIRONMENTAL RELEASES DURING DRUM HANDLING
Potential Environmental
Problem
Preventive Measures
Groundwater Contamination
00
Surface Water Contamination
Air Pollution
• Improve site drainage around the drum handling area and minimize run-on
and run-off by constructing a system of dikes and trenches
• Where groundwater is an important drinking water source, it may be
necessary to hydrologically isolate the work area using well-point
dewatering
• Use liners to prevent leaching of spilled material into groundwater during
drum handling, durm opening, recontainerization and decontamination
• Use sorbents or vacuum equipment to clean up spills promptly
• Locate temporary storage area on highest ground area available; install an
impervious liner in the storage area and a dike around the perimeter of the
area; utilize a sump pump to promptly remove spills and rainwater from
storage area for proper handling
• Construct dikes around the drum handling and storage areas
• Construct a holding pond downslope of the site to contain contaminated
run-off
• Use sorbents or vacuum equipment to promptly clean up spills
• Design the dikes for temporary storage area to contain a minimum of 10% of
total waste volume; ensure that holding capacity of storage area is not
exceeded by utilizing a sump pump to promptly remove spills and rainwater;
for proper handling
• Avoid uncontrolled mixing of incompatible wastes by 1) handling only one
drum at a time during excavation; 2) isolating drum opening operation from
staging and working areas; 3) pumping or overpacking leaking drums and
4) conducting compatibility tests on all drums.
-------
TABLE 2. MEASURES FOR MINIMIZING ENVIRONMENTAL RELEASES DURING DRUM HANDLING
(CONT'D)
Potential Environmental
Problem
Preventive Measures
Air Pollution
(Cont inued)
Fire Protection
• Promptly reseal drums following sampling
• Any drum which is leaking or prone to rupture or leaking, promptly overpack
or transfer the contents to a new drum
• Utilize vacuum units which are equipped with vapors scrubbers
• Where incompatible wastes are intentionally mixed (i.e., acids and bases
for neutralization) in a "compatibility chamber" or tank, releases of
vapors can be minimized by covering the tank with plastic liner
• Use non-sparking hand tools, drum opening tools and explosion proof pumps
when handling flammable, explosive or unknown waste
• Avoid uncontrolled mixing of incompatible waste by 1) handling only one
drum at a time, 2) pumping or overpacking drums with poor integrity, 3)
isolating drum opening, and 4) conducting compatability testing of all
drums
• Use sand or foams to suppress small fires before they spread
• Avoid storage of explosives or reactive wastes in the vicinity of buildings
• In a confident area, reduce concentration of explosives by venting to the
atmosphere
• Cover drums which are known to be water reactive
-------
of the operation may be improved by
additional research and development. As a
result of discussions with cleanup con-
tractors, equipment manufacturers, and EPA
and state officials, additional research
and development is recommended in the
following areas:
o Remote drum opening methods
o Protocols for handling lab packs
and gas cylinders
o Recommendations for minimum com-
patibility testing
o Inventory of major equipment used
for drum handling
o Further evaluation of the use of
non—destructive methods for deter-
mining drum integrity.
The next phase of this project will
develop equipment to satisfy one or more
of these needs.
DRUM HANDLING ACTIVITIES
Detecting and Locating Drums
Detecting drums at an abandoned site
involves the use of historic and back-
ground data on the site, aerial photogra-
phy, geophysical surveying and sampling.
Background data should be carefully ex-
amined since it can minimize the cost and
increase the safety of subsequent activ-
ities. Aerial photography involves the
use of historic aerial photographs to show
changes in the site over time, such as
filling in of trenches, or mounding of
earth. It also includes the use of current
aerial imagery (usually color or infra-
red) to show spills, seepage or changes in
vegetation which may indicate the presence
of drums.
The applicability, reliability and
cost effectiveness of geophysical survey
methods are highly dependent upon site
specific characteristics. Magnetometry is
generally the most useful survey tool for
detecting drums. Metal detectors may be of
value if drums are close to the surface.
Ground penetrating radar is extremely
sensitive but is easily subject to inter-
ference. Electromagnetics and electrical
resistivity are often used together to de-
termine the boundaries of leachate plumes
rather than the location of drums, al-
though electromagnetics has been used for
locating drums. Frequently a combination
of geophysical survey methods is recom-
mended. The results of any geophysical
survey must be verified by sampling.
Determining Drum Integrity
Determining drum integrity is one of
the most important and difficult aspects
of the cleanup operation. Excavation and
subsequent handling of unsound drums can
result in spills and reactions which may
jeopardize worker safety and public
health.
The method generally used for deter-
mining drum integrity is to make a visual
inspection of the drum surface for corro-
sion, leaks, swelling and missing bungs.
This approach requires close contact with
the drums, but is the only effective
method available for determining drum
integrity. A variety of non-destructive
testing methods (i.e. eddy current, X--
rays, ultrasonics) have been investigated
as tools for determining integrity, but
all have been found to have serious
drawbacks or limitations. Use of ultra-
sonics or eddy current, for example,
requires that the surface of the drum be
relatively clean and free from chipping
paint in order to get accurate readings.
Drums which have been buried are fre-
quently covered with soil, residue and
chipping paint and cannot be safely and
easily cleaned. Also, the integrity of
the underside of the drum cannot be
determined with these methods without
lifting it.
The safety of workers in close
contact with the drums must be protected
by the use of appropriate safety gear and
equipment. A variety of direct- reading,
air monitoring equipment is available to
determine radioactivity, oxygen levels,
levels of explosives, and concentrations
of toxic gases. This equipment generally
has remote sensing capabilities and alarms
to warn field personnel of impending
danger as they approach the drum.
420
-------
Excavation and On-Site Transfer
of Drurns
Some waste disposal sites may require
certain pre-excavation activities to im-
prove site access for heavy equipment, to
drain a site to provide a more stable work
surface, or to hydrologically isolate the
site to prevent surface water and ground-
water contamination.
Drum excavation is generally accom-
plished by using a combination of con-
ventional excavation, lifting and loading
equipment such as backhoes, front end
loaders and bobcats, but with special
equipment modifications or accessories
adapted to hazardous waste sites. The
most valuable piece of equipment for
handling drums is the barrel grappler
(Figure 1), which is a modified crawler-
mounted backhoe with a rotating grapple
head. The grapple attachment can rotate
360 degrees along a given plane and is
hydraulically self adjusting in grip
radius so that it can grab and lift
various sizes of containers as well as
containers which are slightly dented or
bent .
Other attachments include nonsparking
buckets to prevent explosions, morman bars
to cover the teeth of backhoes to avoid
puncturing drums, plexiglas safety shields
for vehicle cabs, and drum lifting attach-
ments such as nylon yokes and metal
hoist s.
Equipment used in the excavation and
staging of drums must be suitable for
digging, grabbing, lifting, loading and
manipulating drums. Complete handling of
a drum related problem usually requires a
combination of equipment, particularly
where the grappler is not available.
Where drums are found to be leaking
or structurally unsound, the drum should
be overpacked or the contents transferred
to a new container in order to avoid
spills or releases which could jeopardize
worker safety. Where the grappler is
available, it is generally not necessary
for the worker to be in contact with drums
and spills, thus rupture of unsound drums
may not be a threat to worker safety in
this instance.
Drum Opening, Sampling and Compatibility
Once a drum has been excavated, a
series of activities are undertaken which
lead to final disposal or treatment of the
wastes. The first step is to segregate the
drums based on their contents (liquids,
solids, gases, etc.). Next the drums are
opened, sampled to determine compatibility
and resealed. Drum opening and sampling
should be performed in an isolated area in
order to minimize the possibility of
explosion and fires if drums spill or
rupture. Drum opening tools include
non-sparking hand tools, wrenches,
deheaders, remotely operated plungers,
debungers (Figure 2) and backhoe attached
spikes (Figure 3). Where remotely operated
methods are used, further protection
should be provided by conducting the
operation from behind a plexiglas shield.
Measures should be taken to contain and
mitigate spills which occur during drum
opening. In general, remote drum opening
is recommended where drum integrity is
poor or the wastes are suspected of being
highly toxic.
Compatibility testing should consist
of -rapid, on-site procedures for segrega-
ting incompatible wastes based on such
factors as reactivity, solubility, pre-
sence of oxidizers, water content, etc. ^ .
Drums containing similar wastes are conso-
lidated in order to reduce the costs and
time involved in cleanup and trans-
portat ion.
Recontainerization or Consolidation
There are a number of options avail-
able for consolidating wastes or for
recontainerizing them, if consolidation
is not possible. In general, use of
vacuum trucks is recommended where there
are a minimum of 30 to 40 drums which can
be consolidated.
Industrial strength vacuum units are
available for consolidating both solids
and liquids. Skid mounted vacuum units
are available for areas inaccessible to
trucks or where the number of drums to be
consolidated is small. When using vacuum
equipment, care must be taken to avoid
incompatible waste reactions which can
421
-------
Figure 1. Barrel Grappler, Removing Drums from Pit Exavation
(Courtesy of O.H. Materials Co., Findlay, Ohio)
422
-------
Figure 2. Pneumatic Bung Wrench: Attachment to Drum and
Remote Operation Setup (3)
423
-------
Figure 3. Backhoe Spike (Nonsparking) Puncturing Drum
Held by Grappler
(Courtesy of O.K. Materials Co., Findlay, Ohio)
424
-------
result in costly damage to the vacuum
cylinder. Where incompatible waste reac-
tions are possible, wastes should be
allowed to react under carefully control-
led conditions in a "compatibility" cham-
ber prior to transfer to the vacuum truck.
If drums containing wastes cannot be
consolidated because of incompatibility or
where the number of drums is too small to
make consolidation economical, they can be
overpacked or their contents transferred
to new drums for final disposal. This is
frequently a more costly method of final
disposal since drums and overpacks are
bulky to transport. In weighing the
relative costs of overpacking or consoli-
dating wastes using vacuum equipment, the
costs of decontaminating the vacuum equip-
ment after use should also be included.
The final step in the cleanup process
is decontamination. Equipment is rinsed in
the contaminated work area and the conta-
minated rinse water is collected for
treatment, if it is determined that it
will contribute to groundwater or surface
water contamination. Contaminated soils
may be combined with sludges and solids
and/or bulked, packed in drums, or treated
on-site depending upon the volume of soils
and the nature and concentration of
contaminants. Empty drums are also decon-
taminated in some cases depending upon the
nature of the wastes which were stored in
them. This process may be facilitated by
a drum shredder. If the number of empty
drums is small, they can be crushed and
overpacked. If the drums are not con-
sidered hazardous they can be crushed and
loaded in bulk onto vans or flat—bed
trucks. Decontamination of field workers,
protective clothing and sampling equipment
is essential part of the final cleanup
process. Procedures for decontamination
can be found in several publications
including a Hazardous Waste Site Manage-
ment Plan ^, and the EPA Safety Manual for
Hazardous Waste Site Investigations ^.
Storage and Shipping
In some instances it may be necessary
to store drums on-site temporarily until
funds become available for final treat-
ment/disposal or until a suitable disposal
site is found. If drums are to remain
on-site for a period of three or more
months, then requirements for RCRA per-
mitted facilities should be followed.
These requirements include the following:
o Use of dikes or berms to enclose
the storage area and to segregate
incompatible waste types
o Installation of a base or liner
which is impermeable to spills
o Sizing of each storage area (con-
taining compatible wastes) so that
it is adequate to contain at least
10 percent of the total waste
volume in the event of a spill
o Design of the storage area so that
drums are not in contact with
rainwater or spills for more than
1 hour
o Weekly inspections.
Transportaion and final disposal of wastes
from abandoned sites is governed by
Department of Transportaion regulations as
well as by state laws, disposal facility
requirements, and in the case of highly
toxic materials, by the Toxic Substances
Control Act (TSCA).
CONCLUSIONS
The selection and implementation of
equipment for drum handling is made on a
case by case basis based upon the
following considerations:
o Safety of the field workers
o Protection from environmental re-
leases which could jeopardize
public health
o Site-specific factors which effect
equipment performance
o Costs.
There are a number of precautions which
can be taken to protect field personnel.
Many of these measures such as use of
425
-------
protective clothing, medical surveillance,
communications, etc. apply to all waste
investigations and cleanup operations and
are not unique to drum handling. Other
safety measures, such as use of non-
sparking drum opening tools, or isolation
of the drum opening operations are unique
to drum handling.
In addition, precautions must be
taken to minimize environmental releases.
These include both measures for preventing
spills, leaks and incompatible reactions
as well as mitigative measures for con-
taining and controlling spills or reac-
tions which do occur.
Site specific variables such as
accessibility of the site, the number of
drums, site drainage, drum integrity,
etc., effect equipment performance and
safety, and must be considered in selec-
ting and implementing drum handling
methods. Table 3 summarizes the major
site variables as they influence drum
handling activities.
The final factor affecting the selec-
tion of equipment is costs. Because the
number of drums, their contents and
integrity are generally not known with any
certainty at the outset of the cleanup
operation, costs for cleanup can only be
approximated. Costs can be minimized,
however, by the selection of the most
cost-effective contractor and equipment
types and by close management of the
cleanup operation.
ACKNOWLEDGEMENTS
This project was sponsored by the
USEPA's Municipal Environmental Research
Laboratory - Ci, Oil and Hazardous
Materials Spill Branch, Edison, New Jersey
under contract No. 68-03-3113. The
content of this publication does not
necessarily reflect the views or policies
of the USEPA, nor does mention of trade
names, commercial products or organiza-
tions imply endorsement or recommendation
by the US Government or JRB Associates.
JRB Associates expresses their
appreciation to Mr. Anthony Tafuri and Mr.
Frank Freestone, USEPA Oil and Hazardous
Materials Spills Branch for providing
technical direction for this project.
Technical assistance from Mr. James Walker
and Mr. Robert Panning, OH Materials,
Findlay, OH; Mr. William Bloedorne, Wizard
Drum Tools, Milwaukee, WI; Mr. Gregory
Heath, Peabody Clean Industries, East
Boston, MA; and Mr. Joseph Mayhew,
Chemical Manufacturers Association,
Washington, D.C. is also gratefully
acknowledged.
REFERENCES
1. Neely, N.
Hazardous
Case
et al
Waste
Studies.
. Remedial Actions at
Sites.
EPA
Survey
430-9-81-05.
and
U.S.
Environmental
Washington, D.C., 1981.
Wetzel, R., and Wagner, K. "Drum
Handling Practices at Abandoned Waste
Sites." U.S. Environmental Protec-
tion Agency, Municipal Environmental
Research Laboratory. Draft Report.
1982.
Blackman, W.C., Jr., et al. Enforce-
ment and Safety Procedures for Evalu-
ation of Hazardous Waste Sites.
Proceedings, USEPA National Con-
ference on Management of Uncontrolled
Hazardous Waste Sites, Washington,
D.C. October 15-17, 1980.
Chemical Manufacturers Association,
Hazardous Waste^ Site Management Plan.
Washington, D.C. 1981.
USEPA. Safety Manual for Hazardous
Waste Site Investigations. Office of
Occupational Health and Safety and
the National Enforcement Investiga-
tion Center, 1979.
426
-------
9th ANNUAL SYMPOSIUM ATTENDEES LIST
Ackerman, Donald G. Jr. - PhD.
TRW Inc.
Energy & Environmental Div.
23900 Hawthorne
Suite 200
Torrance, CA 90505
Adams, William R., Jr. - Mgr.
NUSCorp.
Park West Two
Cliff Mine Road
Pittsburg, PA 15275
Adaska, Wayne
Senior Engineer
Portlant Cement Assocn.
Skokie, IL 60077
Ainsworth, Brian
Schlegel Lining Tech., Inc.
200 S. Trade Center Parkway
P.O. Box 7730
The Woodlands, TX 77380
Aittola, Jussi-Pekka - Mgr.
Stusvik Energiteknik AB
611 82NYKOPING, Sweden
Akers, Karol - Engr.
VA Dept. of Health
Division of Solid
& Hazardous Waste Mgmt.
Richmond, VA 23219
Alanddin, Mohammad
KY Division of Waste Mgmt.
Ft. Boone Plaza
ISReillyRoad
Frankfort, KY 40601
Aldridge, Wayne C.
Post Buckley Schuh & Jerrigan Inc.
889 N. Orange Ave.
Orlando, FL 32801
Allen, James L.
E.I. du Pont de Nemours & Co.
3500 Grays Ferry Avenue
Philadelphia, PA 19146
Allen, Richard
United Catalysts Inc.
P.O. Box 32370
Louisville, KY 40232
Alterman, Wayne
Certified Consultants
23 Dellforest Ct.
The Woodlands, TX 77380
Alther, George R.
International Minerals
&Chem.Corp.
17350 Ryan
Detroit, MI 48212
Ammon, Douglas C.
Hydrologist
US EPA
26 W. St. Clair
Cincinnati, OH 45268
Ammons, James T.
US Army Corps of Engineers
P.O. Box 1600
ATTN: ED-CS
Huntsville, AL 35807
Anderson, David C.
K. W. Brown & Assoc.
707 Texas Avenue South
Suite 202D
College Station, TX 77840
Andrews, Douglas - Pres.
Andrews Engineers, Inc.
1320 South Fifth
Springfield, IL 62703
427
-------
Andruacola, Evans
Sales Manager
Trane Thermal
Brook Road
Conshohocken, PA 19428
Appleton, H.
Hyton Engineering Co.
282 Maitland Avenue
Teaneck, NJ 07666
Arden, Mildred B.
Dept. of Environ. Protection
ISReillyRoad
Frankfort, KY 40601
Ardiente, Editha M.
Chemical Engineer
US EPA Region V
230 S. Dearborn St.
Chicago, IL 60604
Armstrong, Katherine
Development Engineer
Monsanto Research Corp.
P.O. Box 32
Miamisburg, OH 45342
Austin, David S.
Tech. Associate
Eastman Kodak Co.
Kodak Park Division
Rochester, NY 14650
Austin, J.A.
Supervisor
Mobil Chemical Co.
One Greenway Plaza, #1100
Houston, TX 77046
Balentine, Jack
Evir. Marketing Coordinator
Catalytic Inc.
1500 Market St.
Philadelphia, PA 19002
Ball, Roy O.
Environmental Resources Mgmt.
200 S. Prospect
Park Ridge, IL 60068
Banerji, Shankha - Prof.
University of Missouri
2037 Engineering Building
Columbia, MO 65211
Banker, Michael R.
Allen &Hoshall, Inc.
2430 Poplar Ave.
Memphis, TN 38112
Barkley, Naomi P.
US EPA
26 W. St. Clair
Cincinnati, OH 45268
Barr, William T.
Eastman Kodak Co.
E.T.S. 8th Floor B-23 Kodak Park
1669 Lake Avenue
Rochester, NY 14650
Bass, Jeffrey
Arthur D. Little, Inc.
15 Acorn Park
Cambridge, MA 02140
Bath, Thomas D.
Consultant
2555 M Street, NW
Washington DC 20037
Baugh, Thomas
Defense Property Disposal
Service
74 N. Washington DPDA-HET
Battle Creek, MI 49016
Beecher, Norman
Assoc. Prof. Tufts Univ.
Medford, MA02155
Beggs, Thomas W.
JACA Corp.
550 Pinetown Road
Ft. Washington, PA 19034
Belk, James D. - V.P.
Welker& Assoc., Inc.
P.O. Box 937
328 Roswell St.
Marietta, GA 30061
Beltis, Kevin J.
Environmental Chemist
Arthur D. Little, Inc.
15W. 315 B. Acorn Park
Cambridge, MA 02140
428
-------
Benz, Edward Paul
Project Geologist
Paulus, Sokolowski & Sartor
67 Mountain Blvd. Ext.
Warren, NJ 07060
Beranek, Jr.
President
Beranek Associates
7442 Countrybrook Drive
Indianapolis, IN 46260
Berkowitz, Jorge H.
Bureau Chief
Dept. of Environ. Protection
8 E. Hanover St.
Trenton, NJ 08625
Bernson, Laurence
Air Monitoring Section
US EPA, Region 2
Woodbridge Ave.
Edison, NJ 08837
Betts, Stephen C.
Principal Associate
PRC Consoer Townsend
404 James Robertson Pkwy.
Nashville, TN 37219
Betzhold, Fred C. - Mgr.
Goodyear Tire & Rubber Co.
Dept. 100D
1144 East Market Street
Akron, OH 44316
Bipes, Roger L.
Asst. Site Director
E.M. Scinece
2909 Highland
Cincinnati, OH 45212
Birch, Richard F.
Manager
Envir. Ex-Cell-o Corp.
2155 Coolidge
Troy, MI 48084
Bizzoco, Francis A.
Envir. Engr. Office
Dept. of Army HQDA
(DAEN-ECE-G)
Washington, DC 20314
Black, Michael I.
Environmental Engr.
US EPA
26 W. St. Clair
Cincinnati, OH 45268
Blaha, Frank
University of Wisconsin
1218 Engineering Bldg.
1415 Johnson Dr.
Madison, WI 53713
Blake, Peter J.
Toxic Waste Containment, Inc.
53 D Street, SE
Washington, DC 20003
Blaney, Ben
Physical Scientist
US EPA
26 W. St. Clair
Cincinnati, OH 45268
Blanz, Robert E.
Deputy Director
Pollution Control & Ecology
P.O. Box 9583
Little Rock, AR 72219
Blick, Clifton T.
Environmental Control
E.J.DuPontCo.
P.O. Box 2042
Wilmington, NY 28402
Bodocsi, Andrew - Assoc. Prof.
University of Cincinnati
MailLoc. #71
Cincinnati, OH 45221
Boje, Rita Rae
Indiana State Board of Health
1330 W.Michigan St.
Indianapolis, IN 46206
Bond, Rick
Res. Engineer
Battelle
P.O. Box 999
Richland, WA 99352
429
-------
Borner, Alan J. - Exec. Dir.
Environmental Hazards Mgmt.
Institute
Box 283,45 Pleasant St.
Portsmouth, NH 03801
Boschuk, John Jr. - V. P.
Geotechnical Division
Orbital Eng., Inc.
1344FifthAve.
Pittsburg, PA 15219
Boszak, Gary P.
Facilities Engr.
GMAD Warren
30009 Van Dyke
Warren, MI 48090
Bothnor, Carl H.
Env. Engr.
ARMCO, INC.
P.O. Box 600
Middletown, OH 45043
Bowen, Russell V.
Staff Engineer
ESE, Inc.
P.O. Box ESE
Gainesville, FL 32602
Brausch, Leo M. - Mgr.
Project Development
D'Appolonia
10 Duff Road
Pittsburgh, PA 15235
Breeding, David C.
Asst. Prof.
Walters State Community College
Morristown, TN 37814
Bridges, James S.
US EPA
Industrial Envir. Res. Lab
26 W. St. Clair
Cincinnati, OH 45268
Bridle, T.R.
Environment Canada
P.O. Box 5050
Birlington, Ontario
Brogard, JohnN.
Env. Engr.
US EPA, Region II
26 Federal Plaza
New York, NY 10278
Brooks, Barry R.
Marketing Mgr.
Energy Inc.
P.O. Box 736
Idaho Falls, ID 83401
Brooks, JohnG.
Env. Supervisor
KY Dept. EPA
400 E. Gray St.
Louisville, KY 40299
Broshears, Robert E.
NSF Fellow
Vanderbilt University
Box 6304 B
Nashville, TN 37235
Bross, Ray A.
Engineer
City of Cincinnati, MSD
1600 Gest St.
Cincinnati, OH 45204
Brown, David S.
Mgmt. Mktg., Wyo-Ben, Inc.
1242 N. 28th St.
P.O. Box 1979
Billings, MT 59103
Brown, Donald
Consultant
NY-TREX, Inc.
3969 Congress Parkway
Richfield, OH 44286
Brown, K.W.
Texas A & M University
Dept. of Soil & Crop Sciences
College Station, TX 77843
Brunsing, Thomas P.
Program Mgr.
Foster-Miller, Inc.
350 Second Ave.
Waltham, MA02154
430
-------
Budde, W.L.
US EPA
26 W. St. Clair
Cincinnati, OH 45268
Budzin, Gerald
Bendix Aircraft Brake & Strut.
3520 Westmoor
P.O. Box 10
Southbend, IN 46624
Buewick, John A. - Dir.
Dev. of Environ. Affairs
Tufts University
Packard Hall
Medford,MA02155
Buice, J.E.
Process Specialist
Dow Chemical Co.
Bldg. A-1107
Freeport, TX 77541
Burkart, Joseph K.
Mech. Engr. MERL/USEPA
5995 Center Hill Road
Cincinnati, OH 45268
Burke, Kim K.
Attorney
Taft, Stettinius & Hossister
1800 First National Bank Center
Cincinnati, OH 45202
Butler, Larry-V.P.
Disposal Operations
US Pollution Control, Inc.
2000 Classen Ctr., Ste. 320
S. Oklahoma City, OK 73106
Butts, Charles
GeoEngineering, Inc.
100 Ford Rd.
Bldg. 3
Denville, NJ 07866
Butt, Karl
Regulatory Analyst
JRB Associates, Inc.
8400 West Park Dr.
McLean, VA 22102
Calouche, Samir I.
Chemist
Virginia State Dept. of Health
109 Governor St.
Richmond, VA 23219
Campbell, H.W.
Environment Canada
Wastewater Tech. Ctr.
P.O. Box 5050
Birlington, Ontario
Carcone, Eugene A.
Dir. of Loss Control Field Service
Utica Mutual Ins. Co.
P.O. Box 530
New Hartford, NY 13413
Carfora, Stephen J.
NY EPA
Division of Waste Mgmt.
120Rt. 156
Yardville, NJ 08620
Carlson, Diane, M.
State of Michigan
WQD
P.O. Box 30028
Lansing, MI 48909
Carnes, Richard A.
Environmental Scientist
US EPA, Combustion Research Fac.
Jefferson, AR 72079
Carter, Patricia
Dir. of Public Relations
Russell & Associates, Inc.
2387 W. Monroe St.
Springfield, IL 62704
Carver, Val B.
Process Engineer
Trade Waste Incineration
#7 Mobile Dr.
Saugey,IL 62201
Cashell, Margaret M.
Civil & Environmental Engr.
University of Cincinnati
703 Rhodes Hall #71
Cincinnati, OH 45221
431
-------
Cassidy, Paul
USEPA(WH-565E)
4th & M Streets, SW
Washington, DC 20460
Castaldini, Carlo
Project Engineer
Acurex Corp.
485 Clyde Avenue
Mountain View, CA 94042
Castle, Paul M.
General Mgr.
W.C.Meredith Co., Inc.
P.O. Box 90456
East Point, GA 30364
Chehaske, John T. - Mgr.
Engineering & Monitoring
Engineering Science
10521 RosehavenSt.
Fairfax, VA 22030
Cherry, Kenneth F. - Mgr.
Clayton Environmental
Consultants, Inc.
25711SouthfieldRd.
Southfield, MI 48075
Childs, Kenneth A.
Advisor Site Remediation
Environment Canada
Ottawa, Ontario K19 K8
Cavalcanti, Fernando - A.P.
Combustion Engr.
Union Carbide Corp.
P.O. Box 8361
South Charleston, WV 25303
Cawley, William A.
Acting Director
US EPA, IERL-CI
26W.St. Clair
Cincinnati, OH 45268
Chadbourne, Dr. John F.
Director of Environmental Affairs
General Portland Inc.
P.O. Box 324
Dallas, TX 75221
Chamberlin, Leland E.
Sprtdt. Envir. Activities
Harrison Radiator Div. GMA
200 Upper Mountain Road
Lockport, NY 14094
Chapman, Wayne
General Manager
NY-TREX, Inc.
3969 Congress Parkway
Richfield, OH 44286
Chawla, Ramesh C.
Associate Professor
Chemical Engineering Dept.
Howard University
Washington, DC 20059
Cho, Hak K.
Envir. Eng.
US EPA Region V
230 S. Dearborn St.
Chicago, IL 60604
Clark, Ann W.
Envir. Eng.
Rohm and Haas Co.
Box 584
Bristol, PA 19007
Clark, Lynn A.
The BF Goodrich Company
Chemical Group
6100 Oak Tree Boulevard
Cleveland, OH 44131
Clark, Scott
Dept. of Environmental Health
University of Cincinnati
Mail Loc. #56
Cincinnati, OH 45267
Clarke, W.H.
Partner, Henderson and Bodwell
3476 Irwin Simpson Road
Mason, OH 45040
Clarke, William H. - Exec. V P.
Pioneer Equities, Inc.
One Wheaton Center
Suite 1801
Wheaton, IL 60187
432
-------
Clarke, William J.
President
Geochemical Corp.
162 Spencer Place
Ridgewood, NJ 07450
Claunch, C. Kenneth
President
921 Greengarden Rd.
Eric, PA 16505
Clear, Jack
Envir. Research Group, Inc.
117 N. First Street
Ann Arbor, MI 48104
Clyde, Robert
Chemical Engineer
Consultant
Box 983
Asheville, NC
Cochron, S. Robert
JRB Associates
8400 West Park Dr.
McLean, VA 22102
Coffman, Glenn N.
Sr. Engineer
Law Engineering Testing Co.
2749 Delk Road SE
Marietta, GA 30067
Coia, Michael F.
Roy F. Weston, Inc.
Weston Way
West Chester, PA 19380
Cointreau, Sandra - Pres.
Solid Waste Mgmt.
Consulting Services, Ltd.
RRl-585 Old Shortwoods Rd.
New Fairfield, CT 06810
Coker, David M.
Env. Control Engineer
Aluminum Co. of America
1501 Alcoa Building
Pittsburgh, PA 15211
Collins, T. Leo
Mgr. Env. Quality
General Electric
Nott St.
Schenectady, NY 12309
Cook, RogerS.
KY Division
Air Pollution Control
ISReillyRd.
Frankfort, KY 40601
Cooke, Marcus
Director
Battelle-Columbus
505 King Ave.
Columbus, OH 43201
Cooper, John
Southwest Research Institute
P.O. Drawer 28510
San Antonio, TX 78284
Costello, Richard J.
P.E. NIOSH
4676 Columbia Pkwy.
Cincinnati, OH 45202
Coulter, Royal
Peoria Disposal Co.
Hazardous Waste Landfill
Peoria, IL 61604
Cowan, Dr. Bruce M.
Project Manager
A.M. Kinney, Inc.
2900 Vernon Place
Cincinnati, OH 45219
Cox, Gary R. - Engr.
Rockwell International
Hanf ord Operations
P.O. Box 800
Richland, VA 99352
Coxe, Edwin F.
Associate Vice President
Reynolds, Smith & Hills
P.O. Box 4850
Jacksonville, FL 32201
Craig, Alfred B. Jr
US EPA
Industrial Envir. Res. Lab
26 W. St. Clair
Cincinnati, OH 45268
433
-------
Crawford, Douglas M.
University of Cincinnati
MERC-SHWRD
26 W. St. Clair St.
Cincinnati, OH 45268
Crawford, James G.
Vice President
AAA Environmental Industries
5544 W. Forest Home Ave.
Milwaukee, WI53220
Crider, Al A.
American Excel-Ltd.
1 E. Main St.
P.O. Box 510
S. Vienna, OH 45369
Crocket, Alan - Mgr.
Environmental Sciences
EG&G Idaho
P.O. Box 1625
Idaho Falls, ID 83401
Cross, Terr L.
VA Health Dept.
109 Governor St.
Richmond, VA 23219
Crumbliss, Ralph - Mgr.
Sales & Marketing
Gulf seal Corp.
601 Jefferson
Houston, TX 77002
Grumpier, Eugene P.
Chemical Engineer
US EPA
401 M Street, SW
Washington, DC 20460
Curran, Mary Ann
Chemical Engineer
US EPA
26 W. St. Clair
Cincinnati, OH 45268
D'Aquila, Margaret M.
Technical Sales Specialist
Mead Compuchem
550 W. Brompton
Chicago, IL 60657
Dalberto, Alfred
PA Dept. of Environmental
Resources
P.O. Box 2063
Harrisburg, PA 17120
Daniel, David E.
Asst. Prof.
University of Texas
Dept. of Civil Energy
Austin, TX 78712
Dantin, Elvin J.
Hazardous Waste
Research Centers
Louisiana State Univ.
Baton Rouge, LA 70803
Davis, J.S.
Senior Geophysicist
EarthTech, Inc.
6655 Amberton Drive
Baltimore, MD 21227
Dawson, Donna
Envir. Specialist
NY EPA
120Rt. 156
Yardville, NJ 08620
Degler, Gerald H.
Sen. Engr.
Bowser-Morner
P.O. Box 81
Dayton, OH 45440
Deiss, Richard A.
Richard A. Deiss & Assoc.
R.D.I
Alden Street Extension
Meadville, PA 16335
Dell, Lee
Principal
Dell Engineering
146 South River Ave.
Holland, MI 49423
Dellinger, Barry
Group Leader
University of Dayton
Research Institute
Dayton, OH 45469
434
-------
Delph, Larry - Coordinator
Environmental Protection
Lexington Health Dept.
650 Newtown Pike
Lexington, KY 40508
DePorter, Gerald L.
Los Alamos National Lab
P.O. Box 1663
Group LS-6, MSK495
Los Alamos, NM 87545
Desai, Harish
Illinois EPA
Div. of Air Pollution Control
2200 Churchill Rd.
Springfield, IL 62704
Determann, James
KY Division of Waste Mgmt.
Ft. Boone Plaza
ISReillyRd.
Frankfort, KY 40601
Detweiler, Roy R.
Mgr. Env. Affairs
DuPontCo. Biochemicals
Barley Mill Plaza #7
Wilmington, DE 19898
DeVault, CleveN.
Ohio State Univ.
2070 Neil Ave.
Room 470, Hitchcock Hall
Columbus, OH 43210
Dial, Clyde J. - Dir.
US EPA
Energy Pollution Control Div.
26 W. St. Clair
Cincinnati, OH 45268
Dickinson, Robert H.
Coordinator Corp.
Westraco Corp.
299 Park Avenue
New York, NY 10171
Djafari, Dr. Sirous Haji - Mgr.
D'Appolonia Waste
Management Services
10 Duff Road
Pittsburgh, PA 15235
Dods, David A.
Graduate Student
Vanderbilt Univ. Box 6304-B
Nashville, TN 37235
Donaldson, Robert T.
Massachusetts Dept.
of Environ. Qlty Eng
One Winter St., 8th Fir.
Boston, MA 02108
Douglas, Jeff M.
Civil Engineer
American Fly Ash Co.
606 Potter Rd.
Des Plaines, IL 60016
Downey, Robert A.
Geologist
Indiana State Board of Health
1330 W.Michigan
Indianapolis, IN 46220
Downey, Thomas W.
Sr. Env. Spec.
NY EPA, Div. of Waste Mgmt.
120 Rt. 156
Yardville, NJ 08620
Doyle, John D.
P.E. Section Chief
Dept. of Nat. Resources
P.O. Box 1368
Jefferson City, MO 65102
Drobny, Neil L.
President
ERM-Midwest, Inc.
4621 Reed Road
Columbus, OH 43220
Duffala, Dale S.
Env. Scientist
Black & Veatch
P.O. Box 8405
Kansas City, MO 64114
Duke, John
Engineer
Procter* Gamble
7162 Reading Rd.
Cincinnati, OH 45222
435
-------
DuRoss, Frank B.
Oneida Asbestos Removal Inc.
333 South St.
Utica, NY 13501
DuRoss, James F.
Vice President
Oneida Asbestos Removal Inc.
333 South Street
Utica, NY 13501
Dzindzeleta, Mercedes - Pres.
Energy & Environmental
Mgmt., Inc.
P.O. Box 422
Racine, WI 53401
Edwards, Linda E.
Admin. Manager
WMS Dames & Moore
644 Linn St., Ste. 501
Cincinnati, OH 45203
Edwards, Stuart
Senior Engineer
Dames & Moore
1150 W. 8th
Cincinnati, OH 45203
Ehrenfeld, John R.
Arthur D. Little, Inc.
15 Acorn Park
Cambridge, MA 02140
Eicher, Anthony R.
Chemical Engineer
IT Corp
3333 Vine St., Ste. 204
Cincinnati, OH 45220
Eide, Allan
Minnesota Waste Mgmt. Board
123 Thorson Community Center
7323 58th Ave. North
Crystal, MN 55428
Eisen, Paul
Wapora, Inc.
21 IE. 43rd St.
New York, NY 10017
Eith, Anthony W.
Sr. Project Engineer
Orbital Engineering, Inc.
1344 Fifth Ave.
Pittsburgh, PA 15219
Elam, David L.
Project Scientist
Harmon Engineering
Auburn Industrial Park
Auburn, AZ 31830
Eliades, Nick
Fort Motor Co.
17000Oakwood
RMF3016
Allen Park, MI 48101
Ellwood, Theodore R.
Industrial Hygienist
IT Corp
3333 Vine St., Ste. 204
Cincinnati, OH 45220
Ely, RobertG.
Watkins & Associates, Inc.
446 E. High St.
P.O. Box 951
Lexington, KY 40588
Emrich, Grover H.
SMC Martin Inc.
900 W. Valley Forge Road
P.O. Box 859
Valley Forge, PA 19482
Erdmann, Fred W.
Soil & Materials Engrs., Inc.
11325 Reed Hartman Hwy.
Suite 134
Cincinnati, OH 45241
Ericson, Franklyn A.
The Upjohn Co.
7171 Portage Rd.
6101-41-0
Kalamazoo, MI 49001
Esposito, Dr. R.G.
Union Chemical Co., Inc.
P.O. Box 423
Union, ME 08862
436
-------
Ewing, Tom
3130 Bishop St.
Cincinnati, OH 45220
Ezell, James D. - Mgr.
SLTC
1640Antioch
Antioch,TN37013
Falconer, Kathleen L.
Senior Engineer
EG&G Idaho, Inc.
P.O. Box 1625
Idaho Falls, ID 83401
Farhoudi, Koorosh
Division of Pollution Center
Ft. Boone Plaza
18ReillyRd.
Frankfort, KY 40601
Farnsworth, Alan
Schlegel Corp.
P.O. Box 23113
Rochester, NY 14692
Favero, David - EPS
IL Env. Protection Agency
2200 Churchill Rd.
Springfield, IL 62706
Feeley, James A.
Hydrologist
TX Dept. of Water Resources
P.O. Box 13087
Austin, TX 78711
Feldmann, John L.
KY Division of Waste Mgmt.
7964 Kentucky Drive, Ste. 8
Florence, KY 41042
Fennelly, Dr. Paul - Mgr.
Envir. Measurements Dept.
GCA/Technology Division
213 Burlington Rd.
Bedford, MA 01730
Finucane, Matthew D.
University of Pennsylvania
Nursing Education Building
420 Service Drive/82
Philadelphia, PA 19104
Fish, Douglas K.
DuPont Co.
PPD/ELD
Barley Mill Plaza
Wilmington, DE 19707
Fisher, Gerald E.
Materials Eng.
E.I. DuPont
105224 Terry Trail
Hinsdale, IL 60521
Flaig, James J.
Vice President
The H.C. Nutting Co.
5802 Beechnut Drive
Cincinnati, OH 45230
Flanigan, Jack
Plant Supt.
Cloudsley Co.
470 W. Northland Rd.
Cincinnati, OH 45240
Fleischman, Marvin - Chrm.
Dept. of Chemical
& Environ. Eng.
Univ. of Louisville
Louisville, KY 40292
Flett, Gregory
Eder Associates
Consulting Engrs.
85 Forest Ave.
Locust Valley, NY 11560
Flood, Jared W.
Environmental Engineer
US EPA
401 M Street, SW
Washington, DC 20460
Fochtman, Ed
Manager
Chemical Waste Mgmt., Inc.
3003 Butterfield R.
Oak Brook, IL 60521
Foss, C.B.
Bulk Petroleum Operations
Crowley Maritime Corp.
One Market Plaza
San Francisco, CA 94105
437
-------
Foster, George R.
Los Alamos National Lab
P.O. Box 1663
MS-K495
Los Alamos, NM 87545
Foushee, Roy
Lexington KY Health Dept.
650 Newtown Pike
Lexington, KY 40508
Fowler, Charles F.
Health and Safety Coordinator
Versar, Inc.
Building 45
Jefferson, AK 72079
Fowler, David E.
128 Greenlawn Ave.
Findlay, OH 45840
Fox, Robert D.
IT Enviroscience
312 Directors Drive
Knoxville, TN 37923
Fralinger, Albert A. - SES
Sr. Environ. Specialist
NJ EPA, Div. of Waste Mgmt.
RD#l,Rt. #1
Vincentown, NJ 08088
Freeman, Henry M.
US EPA
Industrial Envir. Res. Lab
26W.St. Clair
Cincinnati, OH 45268
Fuchs, Robert E.-V.P.
Environmental
Consultants, Inc.
391 Newman Ave.
Clarksville, IN 47130
Fuehrer, John G. II
Fuehrer Associates
345 W. Main Street
Ephrata, PA 17522
Garcia, L.H.
Chem. Engr.
US EPA, IERL-CI
26 W. St. Clair
Cincinnati, OH 45268
Gardner, George D.
NUS Corp.
Part West Two
Cliff Mine Road
Pittsburgh, PA 15275
Gashlin, Kevin
Sr. Envir. Specialist
NY EPA
120 Rt. 156
Yardville, NJ 08620
Gaynor, Charles T. II
Manager
Thomsen Associates
105 Corona Ave.
Groton, NY 13073
Gaynor, Ronald K.
Services & Safety
US Ecology, Inc.
9200 Shelbyville Road
Louisville, KY 40222
Gee, John
University of Wisconsin
1218 Engineering Bldg.
1415 Johnson Drive
Madison, WI53713
Georgevich, Maurice
Ind. Hyg.
NIOSH
4676 Columbia Pkwy.
Cincinnati, OH 45226
Gerbracht, Elmer K.
Technical Director
ACTS Testine Labs, Inc.
3900 Broadway
Buffalo, NY 14227
Ghia, Jay R. - Mgr.
Hazardous Waste Program
Harza Engineering Co.
150 S.Wacker Drive
Chicago, IL 60606
Gilbert, George
KY Division of Waste Mgmt.
Ft. Boone Plaza
ISReillyRoad
Frankfort, KY 40601
438
-------
Gilder, Cindy
US EPA, Region I
JFK Federal Bldg.
SWPB
Boston, MA 02203
Gillespie, Dennis P.
SCS Engineers
211 Grandview Dr.
Suite 315
Covington, KY 41017
Gillespie, Elizabeth
KY Division of Waste Mgmt.
Ft. Boone Plaza
ISReillyRd.
Frankfort, KY 40601
Givens-Reynolds, Louise C.
VA Toxics Roundtable
P.O. Box 89
Salem, VA 24153
Glysson, Eugene A.
Prof, of Civil Engineering
Civil Engr. Dept.
Univ. of Michigan
Ann Arbor, MI 48109
Gohara, WadieF.
Development Engineer
Babcock & Wilcox Co.
Barberton, OH 44203
Goldkamp, William
Univ. of Missouri-Columbia
Dept. of Civil Engineering
Columbia, MO 65211
Gore, William D.
Vanderbilt University
Box 6304 B
Nashville, TN 37235
Gorman, Paul
Chem. Engr.
Midwest Research Institute
425 Volker Blvd.
Kansas City, MO 64110
Gorski, Mitchel R. Jr.
Sales Manager
ThermAH, Inc.
P.O. Box 1776
Peapack, NJ 07977
Goshee, Gary B.
Environmental Engineer
US EPA, Region I
JFK Federal Bldg.
Boston, MA 02203
Gould, Cliff
Env. Prot. Spec. IL EPA
5817 Doe Circle
Westmont, IL 60559
Gower, Mike
Service Rep.
Gabriel & Associates
1814 N. Marsh Field Ave.
Chicago, IL 60622
Gracey, Charles M.
Sr. Engr. Spec.
Aerotect Liquid Rock Co.
P.O. Box 13222
Sacramento, CA 95813
Graham, John L.
University of Dayton
Research Institute
KL 101C
Dayton, OH 45469
Grant, Kevin D.
Mgr. Regulatory Affairs
SCA Chemical Services
5 Middlesex Ave.
Somerville, MA
Grauvogel, Lawrence W.
Project Engineer
Cole Associates, Inc.
221 IE. Jefferson Blvd.
South Bend, IN 46615
Gredell, Thomas R.
Environmental Engr.
MO Dept. of Nat. Resources
P.O. Box 1368
Jefferson City, MO 65102
Green, Mark
Environmental Staff Engr.
Rust Internal Corp.
P.O. Box 101
Birmingham, AL 35201
439
-------
Hall, Robert R.
Senior Staff Engr.
GCA/Technology Division
213 Burlington Rd.
Bedford, MA 01730
Haller, Ray C. - Pres.
RayHallerlnc.
Consulting Engineer
363 Bennington
Indianapolis, IN 46227
Hamlin, Wm. E.
Sales Mgr.
Arizona Refining Co.
P.O. Box 1453
Phoenix, AR 85001
Handyside, Thomas A.
Vice President
City Disposal Systems, Inc.
1550 Harper
Detroit, MI 48211
Hanson, Eric G.
Consultant
P.O. Box 750151
New Orleans, LA 70175
Harmon, Carl B.
Principal Engineer
Watkins & Assoc., Inc.
446 E. High St.
Lexington, KY 40508
Harris, Judith C. - Mgr.
Chemical & Food Sciences
Arthur D. Little, Inc.
15-311 Acorn Park
Cambridge, MA 02140
Hartley, Robert P.
Physical Scientist
US EPA, MERL/SHWRD
26W.St. Clair
Cincinnati, OH 45268
Hawfield, Robert A.
Post, Buckley, Schuh
& Jernigan, Inc.
P.O. Box 106
Cola, SC 29202
Haxo, Henry E. Jr.
President
Matrecon, Inc.
2811 Adeline St.
Oakland, CA 94608
Hayes, Joe R.
Dept. of Environmental
Resources
Rt. 2, Box 225
Bernville, PA 19506
Hazelwood, Douglas
Associate
A.T. Kearney, Inc.
P.O. Box 1405
Alexandria, VA 22313
Hedden, Kenneth F.
Environmental Engr.
EPA Env. Res. Lab.
College Station Rd.
Athens, GA 30613
Heitz, Michael W.
Engineer
Metro Sewer District
1600 Gest St.
Cincinnati, OH 45204
Held, William M.
Staff Engineer
SCS Engineers
211 Grandview Dr.
Covington, KY 41017
Henz, Don J.
Assoc. Dir. of Engrg.
Pedco Environmental, Inc.
11499 Chester Rd.
Cincinnati, OH 45246
Herbst, Dr. Richard P.
Env. Coordinator
Exxon Minerals Co.
P.O. Box 4508
Houston, TX 77210
Herning, Leland P.
Envir. Engr.
Gulf Oil Corp.-Cinn. Ref.
P.O. Box 7
Cleves, OH 45002
440
-------
Herrman, Jonathan G.
US EPA
26W.St. Clair
Cincinnati, OH 45268
Hersh, Stewart
Vice Pres.
KVB Inc.
175 Clearbrook Rd.
Elmsford, NY 10523
Hess, Connie
Hess Environmental
Services, Inc.
6497 Oak Park Dr.
Memphis, TN 38134
Higgins, GregM.
Project Mgr.
Systech Corp.
245 N. Valley Rd.
Xenia, OH 45355
Hill, Ronald D.
US EPA
26 W. St. Clair
Cincinnati, OH 45268
Hillard, Ray L.
American Cyanamid Co.
Bound Brook, NJ 08801
Hinkley, David R.
Director Special Services
Central Hudson ODE Corp.
284 South Avenue
Poughkeepsie, NY 12603
Hoad, George E.
University of Connecticut
U-37
Storrs, CT 06268
Hogan, John C.
Armco, Inc.
703 Curtis Street
Middletown, OH 45043
Holberger, Richard
MITRE Corp.
1820 Dolley Madison Blvd.
McLean, VA 22101
Holroyd, Louis V.
Univ. of Missouri
8 Res. Pk. Dev. Bldg.
Columbia, MO 65211
Holton, Gregory A.
Oak Ridge National Lab
P.O. Box X
Oak Ridge, TN 37830
Home, Jim
NUS Corp.
900 Gemini
Houston, TX 77058
Hornig, Arthur W.
Thuyard Research
3303 Harbor Blvd.
Suite C-8
Costa Mesa, CA 92626
Horton, James F.
MERL/US EPA
5995 Center Hill Rd.
Cincinnati, OH 45268
Horz, Raymond C.
USAE Waterways Experiment
Station
P.O. Box 631
Vicksburg, MS 39180
Houthoofd, Janet N.
US EPA
26 W. St. Clair
Cincinnati, OH 45268
Howell, S. Gary
US EPA
Industrial Envir. Res. Lab
26 W. St. Clair
Cincinnati, OH 45268
Hu, Alan
Rochester Institute of Technology
Rochester, NY 14623
Hubbard, Allen P.
ESE, Inc.
P.O. Box ESE
Gainesville, FL 32602
441
-------
Huddleston, Walter E.
Mason & Hanger Co.
P.O. Box 30020
Amarillo, TX79177
Hyland, James
Dana Corp.
8000 Yankee Rd.
Ottawa Lake, MI 49267
Huffman, George L.
US EPA Center Hill Facility
5995 Center Hill Rd.
Cincinnati, OH 45268
Huggins, Andrew
SynCo Consultants Inc.
12 Bank Street
Bank Street Center
Summit, NJ 07901
Hull, John
Hull Consulting
2726 Monroe St.
Toledo, OH 43606
Hunninen, Katherine
Univ. of Cincinnati
3223 Eden Ave.
Cincinnati, OH 45267
Hunt, Gary
JRB Associates
8400 Westpark Dr.
McLeon, VA 22102
Hurley, Steve
Dept. of Navy
200 Stovall St.
Alexandria, VA 22332
Huston, Arthur C.
Washington Works
E.I. du Pont de Nemours & Co.
P.O. Box 1217
Parkersburg, WV26102
Hutzler, Neil
Michigan Technological Univ.
Dept. of Civil Engineering
Houghton, MI 49931
Hyams, Richard W.
Lockwood, Kessler & Bartlett
One Aerial Way
Syosset, NY11791
lannuzz, Alphonse Jr.
NJDEP
1259RI. 46
Parsippany, NJ 07054
luliucci, Robert L.
Sun Chemical Corp.
4605 Esk Ave.
Cincinnati, OH 45232
Irwin, J. Andrew
O'Brien & Gere Engineers
1304 Buckley Rd.
Syracuse, NY 13221
Jackson, C.S.
US EPA, Region IX
215 Fremont
San Francisco, CA 94105
Jackson, Danny R.
Research Scientist
Battelle-Columbus
505 King Ave.
Columbus, OH 43201
Jacobs, Philip W.
Daily & Assoc.
Engineers Inc.
3716 W. Brighton Ave.
Peoria, IL61615
Jacobson, Laurie
Research Technician
10N.E. 17th St.
Rochester, MN 55901
Jaeger, Ralph R.
Monsanto Research Corp.
Mound Lab
Miamisburg, OH 45342
Jahns, Ronald W.
Illinois EPA
2200 Churchill Rd.
Springfield, IL 62706
442
-------
Jakobson, Kurt
US EPA
401 M Street, SW
Washington, DC 20460
James, Ernie
Stauffer Chemical Company
20720 S. Wilmington Ave.
Long Beach, CA 90810
James, Ruby H.
Southern Research Institute
2000 9th Avenue, South
Birmingham, AL 35255
Janik, David S.
University of Cincinnati
Cincinnati, OH 45221
Jansen, Joe
Missouri Dept. of Natural Res.
P.O. Box 1368
Jefferson City, MO 65106
Jargowsky, Lester W.
Monmouth County Health
Department
Route 9 and Campbell Court
Freehold, NJ 07728
Jarrett, Paul T.
Vanderbilt University
487 Clairmont Place
Nashville, TN 37215
Jaspers, Gregory
Camargo Associates
1329 East Kemper Road
Cincinnati, OH 45246
Jepsen, Christopher P.
Technical Supervisor
American Colloid Co.
SlOOSuffieldCt.
Skokie, IL 60077
Jessee, Gene
Monsanto Company
SOON. Lindbergh
St. Louis, MO 63167
Jett, Morris E.
Schlegel Lining Tech., Inc.
200 S. Trade Center Parkway
P.O. Box 7730
The Woodlands, TX 77380
Jhaveri, Vidjut
Groundwater Decontamination
Systems, Inc.
12 Industrial Park
Waldwick, NJ 07463
Johannesmeyer, Herman
Univ. of Missouri-Columbia
Dept. of Civil Engineering
Columbia, MO 65211
Johnson, Charles E.
Procter & Gamble Co.
Invorydale Technical Center
Cincinnati, OH 45217
Johnson, J.S.
CIBA-Geigy Corporation
Ardsley, NY 10502
Johnson, Larry D.
US EPA
Research Triangle Park, NC 27711
Johnson, Thomas M.
Illinois Geological Survey
615E.PeabodySt.
Champaign, IL 61820
Johnston, Eileen L.
Environmental Educator
505 Maple Avenue
Wilmette, IL 60091
Johnstone, John
Corp. Dir. Eng.
Knowlton Bros.
105 W. 45th
Chattanooga, TN 37410
Jones, Curtis
Kentucky Department for
Environmental Protection
400 E. Gray St.
Louisville, KY 40202
443
-------
Jones, Larry W.
Veritec Corporation
P.O. Box 8791
Knoxville, TN 37996
Joyce, William F.
Stauffer Chemical Company
Eastern Research Center
Dobbs Ferry, MN 10522
Julovich, Peter
111. Inst. of Tech. (3)
7405 W. 400 N.
Michigan City, IN 46360
Jung, Kim E.
R.L. Wurz Company
P.O. Box 223
West Chester, OH 45069
Jungelaus, Gregory
Midwest Research Institute
425 Volker Blvd.
Kansas City, MO 64110
Kachmarsky, Dennis J.
Finkbeiner, Pettis
& Strout, Ltd.
4405 Talmadge Road
Toledo, OH 43623
Kamphake, Lawrence
US EPA
26 W. St. Clair
Cincinnati, OH 45268
Kelley, Mike
Waterways Experiment Station
P.O. Box 631
Vicksburg, MS 39180
Kelly, Ben
US Army Corps of Engineers
20 Massachusetts, NW
Washington, DC 20314
Kenning, Todd
Peoria Disposal Company
Hazardous Waste Landfill
Peoria, IL 61604
Kerho, S.E.
KVB, Inc.
18006Skypark
P.O. Box 19518
Irvine, CA 92714
Kessler, Kimberly A.
Geotechnical & Materials
Consultants, Inc.
1341 Goldsmith
Plymouth, MI 48170
Kim, Y.J.
Chemical Engineer
US EPA Region V
230 S. Dearborn
Chicago, IL 60604
Kimbrough, Charlotte W.
Sverdrup Technology, Inc.
Box 884
Tallahome, TN 37388
King, Lawrence P.
Babcock & Wilcox
1562BeesonSt.
Alliance, OH 44601
Kingsbury, Bob
Regional Sales Manager
American Colloid Co.
P.O. Box 696
Laconia, NH 03246
Kingsbury, Garric L.
Research Triangle Institute
P.O. Box 12194
Research Triangle Park, NC 27709
Kinman, Riley N.
Dept. of Civil Engineering
University of Cincinnati
Cincinnati, OH 45221
Kithany, Subhash S.
Owens-Corning Fiberglas
Granville, OH 43223
Kjaelland, Bob
Ky Division of Waste Mgmt.
Ft. Boone Plaza
ISReillyRoad
Frankfort, KY 40601
444
-------
Klee, Albert J.
US EPA
Industrial Envir. Res. Lab
26 W. St. Clair
Cincinnati, OH 45268
Kleinhenz, Ned J.
Mgr. R & D Systech Corp.
245 N. Valley Rd.
Xenia, OH 45385
Klinger, Larry M.
Sr. Dev. Engineer
Monsanto Research Corp.
P.O. Box 32
Miamisburg, OH 45342
Klint, Steen
TRICIL LTD.
#1, Corunna, Ontario
Canada NON-160
Kmet, Peter
Wisconsin - DNR
Box 7921
Madison, WI 53707
Knauss, James D.
Dames & Moore
2551 Regency
Lexington, KY 40503
Kneuer, Paul R.
Executive Vice President
Norlite Corp.
628 S.Saratoga St.
Cohoes, NY 12047
Knowles, C.L. Jr.
Olin Corp.
120 Long Ridge Road
Stamford, CT 06904
Knowles-Porter, C.
Dames & Moore
350 W. Camino Garden Blvd.
Boca Raton, FL 33432
Knudsen, Dennis R.
Naval Surface Weapons Center
Dahlgren, VA 22448
Koczwara, Margaret K.
University of Cincinnati
Cincinnati, OH 45221
Koines, Arthur T.
US EPA
401 M Street, SW
Washington, DC 20460
Kolpa, Ronald L.
Iowa Department of
Environmental Quality
DesMoines, IA 50319
Kondas, Andrew
PADER
250 Kossman Bldg.
Pittsburgh, PA 15222
Koutsandreas, John D.
US EPA
401 M Street, SW
Washington, DC 20460
Kramlich, John C.
Energy and Environmental
Research Corp.
18 Mason St.
Irvine, CA 92714
Krueger, John A. - Dir.
Dept. of Environ. Protection
State House Station #17
Augusta, ME 04333
Kuhn, Donald J.
SLC Consultants/Constructions
Box 603
North Tonawanda, NY 14120
Kuhn, Eric C.
Proctor Davis Ray Engineers
800 Corporate Drive
Lexington, KY 40503
Kush, George-V. P.
Environmental Affairs
SCA Chemical Services Inc.
5 Middlesex Ave.
Somerville,MA02145
445
-------
Ladd, Donald M.
USAE Waterways
Experiment Station
Geotechnical Laboratory
Vicksburg, MS 39180
Laird, Duncan
USDA, Nat'l. Monitoring Lab
P.O. Box 3209
3505 25th Avenue
Gulf port, MS 39503-1209
Lambert, Martha E.
Research Asst.
University of Cincinnati
MailLoc. #71
Cincinnati, OH 45221
Landreth, Robert E.
US EPA
26 W. St. Clair
Cincinnati, OH 45268
Lanford, Ed
Virginia Department of Health
109 Governor St.
Richmond, VA 23219
Lanier, William S.
US EPA
Research Triangle Park, NC 27711
Larsen, Deborah J.
Arthur D. Little, Inc.
15 Acorn Park
Cambridge, MA 02140
Lasrzal, Kenneth
Spaulding Fibre Co., Inc.
310 Wheeler
Tonawanda, NY14156
Laswell, Bruce
SRW Associates, Inc.
2793 Noblestown Road
Pittsburgh, PA 15203
Lauch, Elizabeth
Purdue University
323 Engineering Admin.
W. Lafayette, IN 47907
Lauer, William F.
Clinton Bogert Associates
2125 Center Avenue
Fort Lee, NV 07024
LaVake, Myron
Monmouth County
Health Department
Route 9 & Campbell Road
Freehold, NJ 07728
Lawson, Louis R. Jr.
Oldover Corp.
P.O. Box 27211
Richmond, VA 23261
Lee, C.C.
US EPA
26 W. St. Clair
Cincinnati, OH 45268
Lee, Fred
E-Three Inc.
P.O. Box 155
Getzville, NY 14068
Lee, Kun-chieh
Union Carbide
P.O. Box 8361
Charleston, WV 25303
Lee, Louis W.
US EPA
Municipal Envir. Res. Lab
26 W. St. Clair
Cincinnati, OH 45268
Lee, Lyon Y.
General Motors
3044 W. Grand Blvd.
Detroit, MI 48202
Lefke, Louis W.
US EPA
Municipal Envir. Res. Lab
26 W. St. Clair
Cincinnati, OH 45268
Lenz, Vicki S.
US Ecology
P.O. Box 7246
Louisville, KY 40207
446
-------
Leonard, Hannah H.
Kentucky National Research
& Env. Protection Cabinet
ISReilly Road
Frankfort, KY 40601
Lewis, NormaM.
SHWRD
68 W. St. Clair
Cincinnati, OH 45244
Lewis, Timothy A.
USGS
Reston, VA 22092
Lichtkoppler, Frank
Area Extension Agent
Ohio State University
99 East Eric St.
Cincinnati, OH 44077
Licis, Ivars J.
US EPA
26 W. St. Clair
Cincinnati, OH 45268
Lindsey, Alfred W.
US EPA
4th & M Streets, SW
Washington, DC 20460
Lindsey, Jerry V.
Rhone Ponbeve, Inc.
Mt. Pleasant, TN 38474
Lingle, Stephen A.
US EPA
401 M Street, SW
Washington, DC 20460
Lippitt, John M.
SCS Engineers
211 GrandviewDr.
Ft. Mitchell, KY
LiPuma, TerranceA.
Vice President
Engineering Science
10521 Rosehaven St.
Fairfax, VA 22030
Logan, Thomas J.
US EPA
Research Triangle Park, NC 27711
Logue, Edward R.
Maine EPA
Augusta, ME 04333
Loper, John C.
University of Cincinnati
Mail Loc. #524
Cincinnati, OH 45267
Lough, Chris
Pope-Reid Associates, Inc.
245 East 6th St. ,#813
St. Paul, MN 55101
Lowry, William F.
NJDEP
R.D. 1, Route 70
Vincentown, NJ 08088
Lu, David W.
Southwestern Ohio Air Pollution
Control Agency
2400 Beekman St.
Cincinnati, OH 45214
Lubowitz, H.R.
US EPA
13414 Prairie Avenue
Hawthorne, CA 90250
Lukey, Michael E.
Vice President
Engineering Science
10521 Rosehaven St.
Fairfax, VA 22030
Lynch, Maurice A. Jr.
Consultant
Lotepro Corp.
4248 Ridge Lea Road
Amherst, NY 14226
Lytle, Steven A. - Mgr.
Soil & Material Engineers
11325 Reed Hartman Hwy.
Suite 134
Cincinnati, OH 45241
447
-------
MacDonald, Alison E.
9 Green tree Dr.
Phoenix, MD 21131
Madden, Terry B.
University of Cincinnati
P.O. Box 23125
Cincinnati, OH 45223
Maffet, Vere - Mgr.
Research & Development
UOP Inc.
10UOP Plaza
DesPlaines, IL60016
Magelssen, L. Scott
Union Carbide
Box 8361
S. Charleston, WV 25303
Mahon, Joseph
Groundwater Decontamination
Systems, Inc.
12 Industrial Park
Waldwick, NY 07463
Malanchuk, Myron
EPA
26 W. St. Clair
Cincinnati, OH 45268
Males, Eric H.
ICF, Inc.
1850 K Street, NW #950
Washington, DC 20006
Malone, Philip G.
US Army Engineers
Vicksburg,MS39180
Manning, Richard E.
Environmental Health
Colt Ind.
430 Park Ave.
New York, NY 10022
Mansfield, Charlie
Texas Instruments, Inc.
P.O. Box 1443
Mail Station 680
Houston, TX 77001
Mappes, Thomas E.
Cabot Corp.
Kokomo, IN 46901
Markey, Margaret
Georgia Environmental
Protection Division
270 Washington St., SW
Atlanta, GA 30334
Marreko, Thomas R.
University of Missouri
Columbia, MO 65211
Marti, Luz R.
USDA Southeast Watershed
Res. Lab. CPES
P.O. Box 946
Tifton, GA31794
Martin, Ed
US EPA
401 M Street, SW
Washington, DC 20460
Martin, Elwood E.
Chemical Engineer
US EPA, OWPE WH527
401 M Street, SW
Washington, DC 20460
Mason, Howard B.
Acurex Corp.
485 Clyde Avenue
Mountain View, CA 94042
Mason, Sam
Engineering Consultant A & S
5404 Peachtree Rd.
Atlanta, GA 30341
Matula, Richard A.
LA State University
College of Engr.
Baton Rouge, LA 70803
Maugham, Robert Y.
EG&G Idaho
P.O. Box 1625
Idaho Falls, ID 83401
448
-------
May, James H.
US Army Engineer
Waterways Experiment Station
P.O. Box 634
Vicksburg, MS 39180
Mayer, Richard J.
The PQ Corp.
P.O. Box 840
Valley Forge, PA 19482
Mayne, Yolande C.
League of Women Voters
113 Marshall St.
Yellow Springs, OH 45387
Mayo, Francis T.
US EPA
26 W. St. Clair
Cincinnati, OH 45268
Mays, Kirk
Aramco Services Co.
1200 Smith #1426
Houston, TX 77002
McAdams, J.W.
Mobil Chemical Co.
211 College Rd. E.
Princeton, NJ 08540
McBath, Audrey
US EPA, IERL-C
26 W. St. Clair
Cincinnati, OH 45268
McCabe, Mark
GCA/Technology Division
213 Burlington Rd.
Bedford, MA 01730
McCormick, Robert J.
Acurex Corp
5658 Shadyhollow
Cincinnati, OH 45230
McCune, Harold E.
Armco Inc.
P.O. Box 600
Middleton, OH 45043
McDonnell, Gerald M.
G.M. Assoc., Inc.
8838 Spinnaker Ct.
Indianapolis, IN 46256
McDonough, James F.
Head Civil & Env. Engineer
University of Cincinnati
Mail Loc. #71
Cincinnati, OH 45221
McGuire, Jerry N.
Monsanto Co.
800 N. Lindbergh
St. Louis, MO 63167
McKelroy, Rodney G.
NUS Corp.
900 Gemini
Houston, TX 77058
McLean, Loren A.
G.D. Searle Labs
4901 Searle Pkwy.
Skokie, IL 60077
McMahill, William F.
Univ. of Missouri
1020 Engineering Bldg.
Columbia, MO 65211
McNiel, Terrance J.
Michigan Dept. of
Natural Resources
P.O. Box 30038
Lansing, MI 48909
Meckes, Mark C.
Defense Property Disposal
Service
74 N. Washington DPDS-HET
Battle Creek, MI 49016
Melberg, John M.
Federal Cartidge Corp.
9th & Tyler St.
Anoka, MN 55303
Menzies, E. Miranda
Dames & Moore
1259 Garden Circle
Cincinnati, OH 45215
449
-------
Merrick, Nelson J.
Aluminum Corp. of America
1501 Alcoa Bldg.
Pittsburgh, PA 15219
Merrill, Richard S.
Acurex Corp.
485 Clyde Avenue
Mountain View, CA 94042
Meshkat, Masoud
University of Kentucky
Lexington, KY 40503
Metis, Dennis M.
University of Kentucky
Lexington, KY 40503
Meyer, G. Lewis
US EPA, Radiation
401 M Street, SW
Washington, DC 20460
Michblsbd, Donald L.
Virginia Tech.
Blacksburg, VA 24061
Michels, Otis E.
Daily & Assoc. Engineers, Inc.
3716 W.Brighton Ave.
Peoria.IL 61615
Mihelaraleis, Joseph L.
University of Cincinnati
MailLoc. #71
Cincinnati, OH 45221
Miles, Mark
Eastman Kodak Co.
Kodak Park Division
Rochester, NY 14650
Milke, Mark
University of Wisconsin
1218 Engineering Bldg.
1415 Johnson Drive
Madison, WI 53713
Millan, Renato C.
Wisconsin Dept. of
Natural Resources
101 S. Webster St.
Madison, WI 53707
Miller, Caryle B.
Dept. of Navy, Bldg. 212
CHESNAVFACENGCOM
Washington Navy Yard
Washington, DC 20374
Miller, David H.
Jones & Laughlin Steel Corp.
900 Agnew Road
Pittsburgh, PA 15211
Miller, Jo E.
University of Cincinnati
Cincinnati, OH 45221
Miller, Marvin P.
Battelle-Columbus
505 King Ave.
Columbus, OH 43201
Miller, Vern F.
Bow Valley Ltd.
2465 S. Estes Ct.
Lakewood, CO 80227
Mills, H. Doyle
Ky Environmental Protection
ISReillyRd.
Frankfort, KY 40601
Minkarah, Issam A.
University of Cincinnati
Mail Loc. #71
Cincinnati, OH 45221
Miullo, Nathaniel J.
US EPA Region 8
1860 Lincoln St.
Denver, CO 80925
Mixon, Forest O.
Research Triangle Institute
P.O. Box 12194
Research Triangle Park, NC 27709
Monnot, Donald R.
Envirodyne Engineers, Inc.
12161 Lackland Rd.
St. Louis, MO 63141
450
-------
Monter, Louis A.
Senior Vice President
CLOUDSLEY CO.
470 W. Northland Road
Cincinnati, OH 45240
Moore, Ben
Engineer
B.H.S., Inc.
R.R. l,Boxl!6-F
Wright City, MO 63390
Moore, Charles
Ohio State Univ.
Dept. of Civil Engineering
2070 Neil Ave.
Columbus, OH 43210
Moore, William P.
Rohm and Haas Co.
Engineering Division
Box 584
Bristol, PA 19007
Moran, Brian V.
US Army Corps of Engineers
Washington, DC 20314
Morekas, Georgeann
Duke University
301 Swift Ave. #17
Durham, NC 27705
Morrel, Mark
Fred C. Hart Assoc.
1110 Vermont Ave., NW #410
Washington, DC 20005
Morrison, Allen M.
Civil Engineering Magazine
345 E. 47th St.
New York, NY 10017
Mostara, Radmand
Penn. Dept. of Envir. Res.
Solid Waste Mgmt.
90 E. Union St.
WilkesBarre, PA 18701
Murphy, William
Sr. Env. Scientist
Pope-Reid Assoc., Inc.
245 E. 6th St.
St. Paul, MN 55101
Murray, John E.
SCA Services
3850 Lower Valley Pike
Springfield, OH 45506
Nandan, Shri
US Pollution Control, Inc.
Suite 320 South
2000 Classen Center
Oklahoma City, OK 73106
Nathan, M.F.
Crawford & Russell
17AmpliaPl.
Stamford, CT 06904
Nechvatal, Michael
Illinois EPA
2200 Churchill Rd.
Springfield, IL 62706
Nelson, Nancy Ann
Matrecon, Inc.
P.O. Box 24075
Oakland, CA 94608
Newhof, Thomas
Prein & Newhof
3000 E. Beltline NE
Grand Rapids, MI 49505
Newland, Jim
Minnesota Waste Mgmt. Board
123 Thorson Community Center
7323 58th Ave. North
Crystal, MN 55428
Ney, Ronald E. Jr.
EPA OSW
401 M Street, SW
Washington, DC 20460
Neyer, William
NLS
P.O. Box 39158
Cincinnati, OH 45239
Nielsen, David M.
National Water Well Assoc.
500 W. Wilson Bridge Road
Worthington, OH 43085
451
-------
Noel, Melbourne A.
President
M.A. Noel Consulting Inc.
5646 N. Kenneth Ave.
Chicago, IL 60646
Nowick, Henry
Monsanto Company
730 Worcester St.
Springfield, MA 01151
Nunes, Sary
Peoria Disposal Co.
Peoria, IL 61604
Nutini, David L.
General Mgr.
RNK Environmental, Inc.
P.O. Box 17325
Covington, KYI 7325
Oberacker, Donald A.
Sr. Mechanical Engineer
US EPA
26 W. St. Clair
Cincinnati, OH 45268
O'Bryan, Glenn A.
Regional Eng.
SCA Services, Inc.
P.O. Box 34457
Louisville, KY 40232
O'Connell, Wilbert
Sr. Res. Scientist
Battelle-Columbus
505 King Ave.
Columbus, OH 43201
O'Conner, John T.
Univ. of Missouri-Columbia
Dept. of Civil Engineering
Room 1047 Engineering Bldg.
Columbia, OH 45211
O'Donnell, Francis R.
Oak Ridge National Lab
P.O. Box X
Oak Ridge, TN 37830
Ogle, Gilbert
Sr. Staff Engineer
TRW, Inc.
8301 Greensboro Dr.
McLean, VA 22102
Ogren, Curtis W.
Chemical Waste Mgmt., Inc.
Louisville, KY 40204
Olexsey, Robert A.
US EPA
26 W. St. Clair
Cincinnati, OH 45268
Ombalski, Stephen D. - Pres.
Ombalski Consulting
Engineers, Inc.
166 West End Avenue
Somerville, NJ 08876
Oppelt, E. Timothy
US EPA
26 W. St. Clair
Cincinnati, OH 45268
Osborne, J. Michael
Three M Company
P.O. Box 33331
Bldg. 21-2W
St. Paul, MN 55133
Osheka, J.W.
Environmental Engineer
PPG Industries, Inc.
One Gateway Center
Pittsburgh, PA 15222
Ostergren, Mark
Business Analyst
Babcock & Wilcox Co.
P.O. Box 2423
N. Canton, OH 44720
O'Sullivan, Colleen
Environmental Specialist
Hillsboro EP Comm.
1900 9th Ave.
Tampa, FL 33605
Otermat, A.L.
Res. Chemist
Shell Dev. Co.
P.O. Box 1380
Houston, TX 77001
Otte, Les
US EPA
401 M Street, SW
Washington, DC 20460
452
-------
Padden, Thomas J.
US EPA
401 M Street, SW
Washington, DC 20460
Pahren, Herbert R.
US EPA
26 W. St. Clair
Cincinnati, OH 45268
Palmer, Charlene R.
H.C. Nutting Company
4120 Airport Road
Cincinnati, OH 45226
Park, James E.
University of Cincinnati
Cincinnati, OH 45221
Parker, Beth L.
Duke University
Durham, NC 27706
Pastene, A. James
Env. Eng.
Union Carbide Corp.
P.O. Box 8361
S. Charleston, WV 25064
Paul, Linda S.
NUS Corp.
Park West Two
Cliff Mine Road
Pittsburgh, PA 15275
Payme, John L.
Branch Manager
Soil & Material Engineers
11325 Reed Hartman Hwy.
Cincinnati, OH 45241
Peacock, James A. - Mgr.
Evans Products Co., Paint Div.
P.O. Box 4098
1516 Cleveland Ave. SW
Roanoke, VA24016
Pendleton, Kenneth A.
K.A. Pendleton Company
10760 Thornview Drive
Cincinnati, OH 45241
Pennefill, Roger A.
US Nuclear Regulatory
Commission
Mailstop 623-SS
Washington, DC 20555
dePercin, Paul R.
Chem. Eng.
US EPA
26 W. St. Clair
Cincinnati, OH 45268
Perry, William H.
NIOSH
4676 Columbia Pkwy.
Cincinnati, OH 45226
Perona, Louis J.
County of LaSalle
707 Etna Road
Ohawa.IL 61350
Person, LeRoy S.
US Nuclear Regulatory
Commission
Washington, DC 20555
Peters, James A.
Monsanto Res. Corp.
1515 Nicholas Road
Dayton, OH 45418
Peters, Nathaniel
University of Kentucky
Lexington, KY 40506
Peters, Wendell
Senior Research Engineer
Southwest Research Institute
P.O. Drawer 28510
San Antonio, TX 78284
Pettyjohn, Wayne A.
Oklahoma State University
Geology Department
Stillwater, OK 74078
Pickering, Edward W.
University of Massachusetts
Amherst
47 Holyoke St.
Northampton, MA 01060
453
-------
Pierce, George E.
Sen. Res. Microbiologist
BattelleMem.Inst.
505 King Ave.
Columbus, OH 43201
Piontek, Keith
Univ. of Missouri-Columbia
1613 Wilson
Columbia, MO 65201
Pitz, Edward
E-Three, Inc.
P.O. Box 155
Getzville, NY 14068
Poff, Timothy A.
NLO, Inc.
P.O. Box 39158
Cincinnati, OH 45239
Pohland, Dr. Fred
Georgia Institute of
Technology
Atlanta, GA 30332
Polakovic, Nolton
The Bureau of Air Pollution
Control
P.O. Box CN-027
Trenton, NJ 08625
Polcyn, Andrew
Environmental Science
and Engineering, Inc.
11665 Lilburn Park Road
St. Louis, MO 63146
Potzman, Dennis W.
Wyo-Ben, Inc.
1242 N. 28th St.
P.O. Box 1979
Billings, MT 59101
Powell, Bruce
GMC-Harrison
P.O. Box 824
Dayton, OH 45401
Price, Richard H.
Hess Env. Services
Memphis, TN 38134
Prohaska, JohnW.
Pedco Environmental, Inc.
11499 Chester Road
Cincinnati, OH 45246
Puch, A.B.
Atlantic Richfield Pet. Prod.
400E.Sibley
Harvey, IL 60426
Pufford, Bob
Minnesota Waste Mgmt. Brd.
123 Thorson Community Cntr.
7323 58th Ave. North
Crystal, MN 55428
Quehee, Shane
University of Cincinnati
Cincinnati, OH 45267
Raines, J. Walter
E.I. du Pont de Nemours & Co.
100 W. lOSt.
Montchanin 5625
Wilmington, DE 19898
Randolph, Brain W.
Research Asst.
University of Cincinnati
Cincinnati, OH 45221
Ransom, Randall
Environmental Specialist
Dow Corning Corp.
3901 S. Saginaw,
Midland, MI 48640
Rawe, James M.
Dept. of Civil Engineering
University of Cincinnati
Cincinnati, OH 45221
Redding, Peter M.
Vanderbilt University
Box 6304-B
Nashville, TN 37235
Reed, John C.
Illinois EPA
2200 Churchill Road
Springfield, IL 62706
454
-------
Rehme, Elmer W.
Ohio EPA
7 E. 4th St.
Dayton, OH 45402
Reible, Dan
Asst. Prof, of Chem. Engr.
Louisiana St. Univ.
Baton Rouge, LA 70803
Reich, Andrew R.
University of Alabama
720 20th St., S
Birmingham, AL 35294
Reshkin, Mark
Assoc. Dir. for Envir. Res.
Indiana University, NW
3400 Broadway
Gary, IN 46408
Reuter, Steve P.
Engineer
Indiana State Board of Health
1330 West Michigan St.
Indianapolis, IN 46206
Rich, Charles A.
C.A. Rich Consultant
708 Glen Cove Ave.
Glen Head, NY 11545
Richardson, Jean
Jean Richardson & Assoc.
2709 S. 20th St.
Birmingham, AL 35209
Rickabaugh, Janet I.
University of Cincinnati
Cincinnati, OH 45221
Rickman, Bill
Manager
G.A. Technologies
P.O. Box 85608
San Diego, CA 92138
Riggenbach,Jack D.
ERM-Inc.
P.O. Box 357
West Chester, PA 19380
Riley, Boyd T.
Rycon Inc.
690 Clinton Springs
Cincinnati, OH 45229
Roberto, Gerard
University of Cincinnati
Cincinnati, OH 45204
Roberts, Susan A.
Geologist
Malcolm Pirnie, Inc.
2 Corporate Park Dr.
White Plains, NY 10602
Robine, Deith
National Audubon Society
950 Third Avenue
New York, NY 10022
Robinson, Shird
SCA Services, Inc.
2105 Outer Loop Road
Louisville, KY 40219
Roetzer, James F.
Envirosphere Company
Two World Trade Center
New York, NY 10048
Rogers, Charles J.
US EPA
Industrial Envir. Res. Lab
26 W. St. Clair
Cincinnati, OH 45268
Rohrer, William L.
Pope-Reid Assoc., Inc.
245 E. 6th St.
St. Paul, MN 55101
Rollins, Dixon
Sr. Engineer
New York Dept. of Envir. Con.
6274 E. Avon Lima Road
Avon, NY 14414
Ronayne, Michael
University of Kentucky
Lexington, KY 40506
455
-------
Rosebrook, Dr. Donald D.
Institute Scientist
Gulf South Research Inst.
P.O. Box 14787
Baton Rouge, LA 70898
Ross, Richard D.
Trofe Incineration Inc.
Pike Rd.
Mt. Laurel, NY 08054
Ross, Robert W. II
US EPA Combustion Res. Facility
NCTR Building #45
Jefferson, AR 72079
Rothenstein, Cliff L.
US EPA
401 M Street, SW
Washington, DC 20460
Roulier, MikeH.
US EPA
26 W. St. Clair
Cincinnati, OH 45268
Rowe, Walter
General Portland Company
P.O. Box 324
Dallas, TX 75240
Rubey, Wayne S.
University of Dayton
300 College Park Ave.
Dayton, OH 45469
Ruby, Mike - Asst. Prof.
Dept. of Civil Engineering
University of Cincinnati
MailLoc. #71
Cincinnati, OH 45221
Ruggles, Archie Jr.
Project Engineer
Mason & Hanger Company
P.O. Box 30020
Amarillo, TX79177
Rupa, Edward
JRB Associates
8400 Westpark Drive
McLean, VA 22102
Russell, Charles
Russell & Associates, Inc.
2387 W.Monroe St.
Springfield, IL 62704
Russell, Dwight
Texas Dept. of Water Res.
P.O. Box 13087
Capitol Station
Austin, TX 78711
Salloum, John D.
Environmental Protection Service
Ottawa, Ontario
Canada
Samkow, Willard
Hilton Davis Chem. Co.
Cincinnati, OH 45222
Sampayo, Felix F.
Johnes & Henry Engineers Ltd.
Toledo, OH 45606
Sanning, Donald E.
US EPA, SHWRD
26 W. St. Clair
Cincinnati, OH 45268
Santoro, David S.
EA Engineering/Ecological
Analysts, Inc.
100 TechneCenter Drive, Ste. 212
Milford, OH45150
Santoro, Joseph D.
EA Engineering/Ecological
Analysts, Inc.
100 TechneCenter Drive
Milford, OH 45150
Sappington, Douglas L.
Consultant
3944 Windgap Avenue
Pittsburgh, PA 15204
Sargent, T.N.
Engineering-Science
Suite 590
57 Exec. Park S.
Atlanta, GA 30329
456
-------
Sarro, William F.
US EPA, Region I
JFK Federal Bldg.
Boston, MA 02203
Sauer, Richard E.
US Ecology, Inc.
9200 Shelbyville Rd.
Louisville, KY 40222
Sauter, S. Jay
Orange Cnty. Solid Waste Dept.
P.O. Box 14413
Orlando, FL 32857-4413
Saw, Chin C.
Int'l. Paper Co.
P.O. Box 797
Tuxedo Park, NY 10987
Sawdey, Norman J.
GMC, Harrison Radiator
Dept. 507
P.O. Box 824
Dayton, OH 45402
Sawyer, Charles J.
Syntex, Inc.
3401 Hillview Ave.
Palo Alto, CA 94304
Schaefer, Betty M.
Senior Chemist
Wilson Nolan, Inc.
2809 NW Expressway, Ste. 290
Oklahoma City, OK 73112
Schaefer, Phillip T.
Zimpro, Inc.
Military Rd.
Rothschild, WI 54474
Scheben, Jackie A.
Tech. Sales Rep.
Cecos International
4879 Spring Grove Ave.
Cincinnati, OH 45232
Schmidt, Edgar H.
Ontario Waste Management Corp.
2BloorSt. W., llth Floor
Toronto, Ontario
Canada M4W 3E2
Schmidt, Richard K.
Gundle Lining Systems, Inc.
1340E. RicheyRd.
Houston, TX 77073
Schneider, Carl A.
Vanderbilt Univ.
Box 2231 StationB.
Nashville, TN 37235
Schoenbeck, Melvin
El Dupont
Elastomers Lab
Chestnut Run
Wilmington, DE 19898
Schomaker, Norbert B.
US EPA
26 W. St. Clair
Cincinnati, OH 45268
Schraub, Tony
Acurex Corp.
485 Clyde Avenue
Mountain View, CA 94042
Schreiber, Dale E.
Poe & Assoc. of Tulsa, Inc.
10820 E. 45th St.
Tulsa, OK 74145
Schroeder, Paul R.
USAE Waterways Experiment
Station
P.O. Box631SESEE
Vicksburg, MS 39180
Schroy, Jerry M.
Monsanto Co.
800 N. Lindbergh Blvd.
St. Louis, MO 63011
Schuller, Rudolph M.
SMC Martin, Inc.
900 W. Valley Forge Rd.
P.O. Box 859
Valley Forge, PA 19482
Schumann, Charles E.
Southwestern Ohio Air
Pollution Control Agency
2400 Beekman St.
Cincinnati, OH 45214
457
-------
Schwartz, William G.
Hazardous Waste Landfill
1113 W. Swords
Peoria, IL61604
Seeker, W. Randall
Energy & Envir. Res.
18 Mason
Irvine, CA 92714
Sehgal, S.B.
Geotechnical & Materials
Consultants, Inc.
1341 Goldsmith
Plymouth, MI 48170
Servis, David B.
Procter-Davis-Ry Engineers
800 Corporate Drive
Lexington, KY 40503
Sferra, PasqualeR.
US EPA
Industrial Envir. Res. Lab
26W.St.Clair
Cincinnati, OH 45268
Shack, Pete A.
Phoenix Environmental
Cons. Inc.
P.O. Box 121555
Nashville, TN 37212
Shafer, Joseph D.
Indiana State Board of Health
1330 W.Michigan
Indianapolis, IN 46206
Shaub, Walter M.
National Bureau of Standards
Washington, DC 20234
Shelley, Philip E.
EG&G WASC, Inc.
2150 Fields Rd.
Rockville, MD 20850
Shen, AlmonM.
Shell Environmental
Group, Inc.
4930 N. Penn. St.
Indianapolis, IN 46205
Sherman, J.S.
Radian Corp.
8500 Shoal Creek Blvd.
Austin, TX 78758
Shugart, Steven L.
Mayes, Sudderth & Etheredge
1785 the Exchange
Atlanta, GA 30339
Shultz, Dave W.
Southwest Research Institute
P.O. Drawer 28510
San Antonio, TX 78284
Shuster Ken
EPA (WH 565 E)
401 M Street, SW
Washington, DC 20460
Simms, Ben M.
Mason & Hanger Co.
P.O. Box 30020
Amarillo, TX79177
Sims, Ronald C.
Utah Water Research Lab
Utah State University
Logan, UT 84322
Singh, Rajiv
EarthTech, Inc.
6655 Amberton Drive
Baltimore, MD 21227
Skinner, Donna I.
Regional Hydrogeologist
P.A.D.E.R.
1012 Water St.
Meadville, PA 16335
Skinner, Peter N.
NYS Attorney General
Rm. 239
Justice Bldg.
Albany, NY 12224
Smalley, Carolyn J.
Day & Zimmermann, Inc.
Kansas Army Ammunition Plant
Parsons, KS 67357
458
-------
Smith, Garrett A.
US EPA Region II
26 Federal Plaza
New York, NY 10278
Smith, John M. - Pres.
J.M. Smith Consulting
Engineers
7373 Beechmont Ave.
Cincinnati, OH 45230
Snow, Brad L.
Engineering Services Div.
Kerr-McGee Corp.
P.O. Box 25861
Oklahoma City, OK 73125
Socash, Stephen M.
Dept. of Environmental Res.
1012 Water St.
Meadville, PA 16335
Sokol, John Z.
Clyde E. Williams & Assoc.
1843 Commerce Dr.
South Bend, IN 46628
Speed, Nicholas A.
Brown & Caldwell
1501 N. Broadway
Walnut Creek, CA 94596
Spigolon, S.J.
Memphis State University
Dept. of Civil Engineering
Memphis, TN 38152
Spooner, Philip
JRB Associates
8400 Westport Drive
McLeon, VA22102
Springer, Charles
University of Arkansas
Dept. of Chemical Engineering
Fayetteville, AR 72701
Staley, Laurel
US EPA
26 W. St. Clair
Cincinnati, OH 45268
Steffens, Chuck
Caterpillar Tractor Co.
East Peoria Plant
Bldg. KK-1
East Peoria, IL 61611
Stephan, David G.
US EPA
26 W. St. Clair
Cincinnati, OH 45268
Sterling, Harry J.
Dept. of Civil Engineering
University of Kentucky
Lexington, KY 40506
Stern, David A.
GCA/Technology Division
213 Burlington Rd.
Bedford, MA 01730
Stidham, Leslie N. - SEC Rep.
American Thermoplastics Corp.
1235 Kress St.
Houston, TX 77020
Stoddart, Terry L.
US Air Force/Environics Lab
5724 Ivy Rd.
Panama City, FL 32404
Strachan, William M.
OHIO EPA
7 East 4th St.
Dayton, OH 45402-2086
Stutsman, Mark J.
US EPA
26 W. St. Clair
Cincinnati, OH 45268
Sugarman, Peter
N.J. Dept. ofEnvir.
Protection Agency
P.O.BoxCN-029
Trenton, NJ 08625
Suhrer, F.C.
US EPA
215 Fremont
San Francisco, CA 94105
459
-------
Sullivan, Daniel E.
Roy F. Weston, Inc.
Weston Way
West Chester, PA 19380
Swartzbaugh, Joseph T.
General Manager
Systech Corp.
245 N. Valley Rd.
Xenia, OH 45385
Talak, Anthony
PADER
1012 Water St.
Meadville, PA 16335
Tally, John
Engineer Director
NIOSH
Cincinnati, OH 45226
Tamplin, Judy
GA Envir. Protect. Div.
270 Washington St., SW
Atlanta, GA 30334
Tang, Harry K.
Planning Research Corp.
Kennedy Space Center
(PRC-1217)
Orlando, FL 32899
Tansill, S.P.
R. Stuart Royer & Assoc., Inc.
P.O. Box 8687
Richmond, VA 23226
Taylor, David R.
S-CUBED
P.O. Box 1620
LaJolla, CA92129
Thompson, Joe D.
EG&G
P.O. Box 1625
Idaho Falls, ID 83401
Thompson, Steve R.
Divisional Vice President
Browning Ferris Ind.
P.O. Box 3151
Houston, TX 77071
Thrasher, Stephen
Bowser-Morner, Inc.
420 Davis Ave.
Dayton, OH 45403
Tibbetts, Stephen
Union Chemical Co., Inc.
P.O. Box 432
Union, ME 04862
Tite, Joseph L.
Consulting Engr.
Joseph Tite Co.
P.O. Box 366
Michigan City, IN 46360
Tong, Peter
US EPA Region V
230 S. Dearborn
Chicago, IL 60604
Totts, David
Sr. Envir. Specialist
NY EPA
120Rt. 156
Yardville, NJ 08620
Trapp, John H.
MSD DF Greater Cincinnati
1600GestSt.
Cincinnati, OH 45204
Trembley, J.W.
LAN, Inc.
1500 City West
Houston, TX 77042
Trenholm, Andrew
Midwest Research Institute
425 Volker Blvd.
Kansas City, MO 64110
Triegel, Elly K.
Woodward-Clyde Consultants
5120 Butler Pike
Plymouth Meeting, PA 19462
Tsai, Kuo-Chun
Asst. Prof.
University of Louisville
Dept. of Chem. & Env. Eng.
Louisville, KY 40292
460
-------
Tsang, Wing
Research Chemist
National Bureau of Standards
Washington, DC 20234
Tseng, Louis H.
Environmental Elements Corp.
7249 National Drive
Hanover, MD 21076
Turgeon, Marc P.
EPA OSW
401 M Street, SW
Washington, DC 20460
Tussey, Robert C.
Kenvirons, Inc.
P.O. Drawer V
Frankfort, KY 40602
Twilley, Clinton
RETECH Associates, Inc.
861 Corporate Dr.
Suite 200
Lexington, KY 40503
Tyler, Scott
Battelle PNL
Box 999
Richland, WA 99352
Tyndall, M. Frank
Howard Needles Tammen
& Bergendof f
P.O. Box 68567
Indianapolis, IN 46268
Tyro, Michael J.
General Motors Corp.
EASBldg.,GMTechCtr.
Warren, MI 48090
Ullrich, ArlieJ.
Consultant
Eli Lilly & Co.
307 E. McCarty St.
Indianapolis, IN 46285
Uhl, Michael E.
Peake Operating Co.
Charleston National Plaza #423
Charleston, WV 25301
Underwood, Edward R.
US Ecology, Inc.
9200 Shelbyville Road
Louisville, KY 40222
Vanderveld, Ronald J.
DeTox, Inc.
One Wheaton Center
Suite 1801
Wheaton, IL 60187
Vakili, Hassan
Virginia Dept. of Health
109 Governor St.
Richmond, VA 23219
VanderMeulen, Joseph
Legislative Service Bureau
Michigan State Legislature
Allegan St., Farnum Bldg.
Lansing, MI 48913
Veith, Jim E. - Sr. Eng.
Soil & Material Engineers
11325 Reed Hartman Hwy.
Suite 134
Cincinnati, OH 45241
Velez, Victor G.
EBASCO Services, Inc.
2 World Trade Center
New York, NY 10463
Velie, Margaret M.
US EPA
401 M Street, SW
Washington, DC 20460
Velzy, Charles O.
Charles R. Velzy Assoc., Inc.
355 Main St.
Armonk, NY 10504
Vidmar, Kevin P.
Vanderbilt Univ.
Box 6304 B
Nashville, TN 37235
Vogt, W. Gregory
Staff Scientist
SCS Engineers
211 Grandview Drive
Covington, KY 41017
461
-------
Volk, David R.
PADER
850 Kossman Bldg.
Pittsburgh, PA 15222
Vollstedt, Thomas
Zimpro, Inc.
Military Rd.
Rothschild, WI 54474
Wagner, Douglas H.
V.P. Operations
Solidtek
Box 888
Morrow, GA 30260
Walker, E.G.
Burns and Roe
650 Winters Ave.
Paramus, NJ 07652
Walls, James T.
Hamilton Co. Farm Bureau
2870 Markbreit Ave.
Cincinnati, OH 45209
Walsh, James
SLS Engineers
211 Grandview Dr.
Covington.KY 41017
Watkin, Andrew T.
Vanderbilt University
3102-B Wellington Ave.
Nashville, TN 37212
Watkins, David R.
Physical Scientist
US EPA
26 W. St. Clair
Cincinnati, OH 45268
Walz, Arthur
US Army Corps of Engineers
20 Mass, NW
Washington, DC 20314
Webb, George C.
Geotechnical Engineer
H.C. Nutting Co.
4120 Airport Rd.
Cincinnati, OH 45226
Webb, Thomas E.
Amer. Elec. Power Svc. Corp.
P.O. Box 487
Canton, OH 44708
Weinberger, Lawrence P.
The Aerospace Corporation
955 L'Enfant Plaza, SW
Suite 4000
Washington, DC 20024
Weishaar, Michael F.
Monsanto
SOON. Lindbergh
St. Louis, MO 63167
Weiss, Albert
Weiss Pollution Control
41001 Grand River
P.O. Box 505
Novi, MI 48050
Werner, James
Environmental Law Institute
1346 Connecticut Ave., NW
Washington, DC 20036
Werner, Steven I.
Occidental Chemical Corp.
360 Rainbow Blvd. S.
Niagara Falls, NY 14302
Westbrook, Clifton W.
Research Triangle Institute
P.O. Box 12194
Research Triangle Park, NC 27709
Westfall, Brain A.
US EPA
Industrial Envir. Res. Lab
26 W. St. Clair
Cincinnati, OH 45268
Wetzel, David
Asst. Prof, of Chem. Eng.
Louisiana State Univ.
Baton Rouge, LA 70803
Wetzel, Roger
JRB Associates
8400 Westpark Dr.
McLean, VA 22102
462
-------
Whittle, George P.
University of Alabama
Dept. of Civil Engineering
P.O. Box 1468
University, AL 35486
Whitmore, Frank
Versar, Inc.
US EPA Combustion Research Facility
Jefferson, AR 72079
Whitney, Richard R.
Acurex Corp.
485 Clyde Avenue
Mountain View, CA 94042
Wickline, Bob
Virginia Dept. of Health
109 Governor St.
Richmond, VA 23219
Widmer, Wilber
Dept. of Civil Engineering
University of Connecticut
Box U-37
Storrs, CT 06268
Wiggans, Kenneth E.
US Army, USAEHA
Aberdeen Proving
Ground, MD 21010
Wigh, Richard J.
Regional Services Corp.
3200 Sycamore Ct., #2B
Columbus, IN 47203
Wiles, Carlton C.
US EPA MERL
SHWRD
26 W. St. Clair St.
Cincinnati, OH 45268
Williams, Charles E.
McBride-Ratcliff & Assoc.
8800 Jameel
Suite 190
Houston, TX 77040
Willis, Dudley L.-P.E.
Resources Recovery, Inc.
108 Briar Lane
Newark, DEI 9711
Withiam, James L.
D'Appolonia Consulting Eng.
10 Duff Road
Pittsburgh, PA 15235
Witnrow, William
111. Pollution Control Board
309 W.Washington
Chicago, IL 60606
Wolbach, C.D.
Acurex Corp.
485 Clyde Avenue
Mountain View, CA 94042
Wolf, Fred L.
Kester Management Services
4274 Miramar Drive
Toledo, OH 43614
Wolfe, Doug
McCoy & McCoy, Inc.
85 E. Noel Ave.
Madisonville, KY 42431
Woodley, Ralph
Burke Rubber Company
2250 S. Tenth St.
San Jose, CA 95112
Worm, Brenda
Hazardous Waste Research Ctr.
CEBA3418
Baton Rouge, LA
Yaar, A.
C.E. Williams & Assoc.
1843 Commerce Drive
South Bend, IN 46614
Yalcin, Acar
Louisiana State University
Baton Rouge, LA 70803
Yang, Edward
Environmental Law Institute
1346 Connecticut Ave., NW
Washington, DC 20056
Yare, Bruce S.
Yare and Associates, Inc.
24 S. 77th St.
Belleville, IL 62223
463
-------
Young, Bob
DELCO MORAINE
1420 Wise. Blvd.
Dayton, OH 45401
Zak, Clarence
Sunbeam Equipment Corp.
180 Mercer Street
Meadville, PA 16335
Zaninelli, Linda
NY EPA
120Rt. 156
Yardville, NJ 08620
Zimmerman, R. Eric
ESCOR, Inc.
1845 Oak St.
Northfield, IL 60093
Zitkovic, John J.
O.H. Materials Co.
P.O. Box 551
Findlay, OH 45840
Zlamal, Frank
Slurry Systems Division
of Thatcher Eng.
7100 Industrial Ave.
Gary, IN 46406
Zralek, Robert L.
Director of Civil Systems
Waste Management Inc.
3003 Butterfield Rd.
Oak Brook, IL 60521
Zykan, James Jr.
President
B.H.S.,Inc.
R.R. l,Box!16-F
Wright City, MO 63390
464
*USGPO: 1983-659-095-0750
-------
|