oEPA
United States
Environmental Protection
Agency
Municipal Environmental Research EPA-600/9-78-016
Laboratory August 1978
Cincinnati OH 45268
Research and Development
Land Disposal
of Hazardous Wastes
Proceedings
of the Fourth Annu.al
Research Symposium
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/9-78-016
August 1978
LAND DISPOSAL OF HAZARDOUS WASTES
Proceedings of the Fourth Annual Research Symposium held at San Antonio, Texas
March 6, 7 and 8, 1978. Cosponsored by Southwest Research Institute
and the Solid and Hazardous Waste1 Research Division, U.S. Environmental
Protection Agency
Edited by David W. Shultz
Department of Environmental and Resource Engineering
Southwest Research Institute
San Antonio, Texas 78284
Project Officers
Norbert B. Schomaker
Robert E. Landreth
Solid and Hazardous Waste Research Division
Municipal Environmental Research Laboratory
Cincinnati, Ohio 45268
Mildred Smith
U.S. Environmental Protection Agency
Region VI
Dallas, Texas 75201
MUNICIPAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
DISCLAIMER
These Proceedings have been reviewed by the
U.S. Environmental Protection Agency and ap-
proved for publication. Approval does not
signify that the contents necessarily reflect
the views and policies of the U.S. Environ-
mental Protection Agency, nor does mention
of trade names or commercial products consti-
tute endorsement or recommendation for use.
11
-------
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 inter-
play 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 dis-
charges from municipal and community sources; to preserve and treat public drinking water
supplies; and to minimize the adverse economic, social, health and aesthetic effects of
pullution. This publication is one of the products of that research—a vital communica-
tions link between the researcher and the user community.
The proceedings present research aimed at minimizing the impact of direct land dis-
posal of hazardous wastes and provide solutions to specific problems.
Francis T. Mayo
Director
Municipal Environmental
Research Laboratory
iii
-------
PREFACE
These proceedings are intended to disseminate up-to-date information on extramural
research projects dealing with the disposal of solid and hazardous wastes. These projects
are funded by the Solid and Hazardous Waste Research Division (SHWRD) of the U.S. Environ-
mental Protection Agency, Municipal Environmental Research Laboratory in Cincinnati, Ohio.
Selected papers from work of other organizations were included in the symposium to identify
closely-related work not included in the SHWRD program.
The papers in these proceedings are arranged as they were presented at the symposium.
Each of the five sessions includes papers dealing with major areas of interest for those
involved in hazardous waste management and research.
The papers are printed here basically as received from the symposium authors. They
do not necessarily reflect the policies and opinions of the U.S. Environmental Protection
Agency or Southwest Research Institute. Hopefully, these proceedings will prove useful
and beneficial to the scientific community as a current reference on land disposal of
hazardous wastes.
iv
-------
ABSTRACT
The fourth SHWRD research symposium on land disposal of hazardous waste was held at
the St. Anthony Hotel in San Antonio on March 6, 7 and 8, 1978. The purpose of the
symposium was two-fold: (1) to provide a forum for a state-of-the-art review and discus-
sion of ongoing and recently-completed research projects dealing with the management of
hazardous wastes and (2) to bring together people concerned with hazardous waste manage-
ment who can benefit from an exchange of ideas and information. These proceedings are a
compilation of the papers presented by symposium speakers. They are arranged in the order
of presentation. The five primary technical areas covered are
(1) Methods development and economic assessment,
(2) Identification of pollutant potential,
(3) Predicting trace element migration,
(4) Modification of disposal sites and waste streams, and
(5) Alternatives for land disposal.
-------
TABLE OF CONTENTS
Page
OPENING SESSION
Current Research on Land Disposal of Hazardous Wastes
Norbert B. Schomaker, U.S. Environmental Protection Agency.
Industrial Hazardous and Toxic Waste Program of EPA's Industrial Pollution
Control Division—IERL—CI
Leo Weitzman, U.S. Environmental Protection Agency 14
Texas' Solid Waste Management Activities: An Overview
Jay Snow, Texas Department of Water Resources 19
Solid Waste Research Activities in Canada
Hans Mooij, Environmental Impact Control Directorate 20
SESSION I: METHODS DEVELOPMENT AND ECONOMIC ASSESSMENT
Kepone: An Overview
Richard A. Carnes, U.S. Environmental Protection Agency 25
The Development of a Leaching Test for Industrial Wastes
Robert K. Ham, University of Wisconsin 33
Genetic Toxicity Testing of Complex Environmental Effluents
James L. Epler, Oak Ridge National Laboratory 47
A Framework for Economic Analysis of Hazardous Waste Management Alternatives
Graham C. Taylor, Colorado School of Mines 55
Economics of Disposal and the Compilation of a Data Base for Standards/
Regulations of FGD Sludge
John Woodyard, SCS Engineers 74
Laboratory Characterization of the Thermal Decomposition of
Hazardous Wastes
Don Duvall, University of Dayton Research Institute 104
SESSION II: IDENTIFICATION OF POLLUTANT POTENTIAL
Standardized Methods for Sampling and Analysis of Hazardous Wastes
Robert D. Stephens, California Department of Health 112
Characterizing Input to Hazardous Waste Landfills
Richard A. Carnes, U.S. Environmental Protection Agency 122
Co-Disposal of Industrial and Municipal Wastes in a Landfill
Joseph T. Swartzbaugh, Systems Technology Corporation 129
vii
-------
TABLE OF CONTENTS (Cont'd)
Page
SESSION III: PREDICTING TRACE ELEMENT MIGRATION
Accelerated Testing of Waste Leachability and Contaminant Movement
In Soils
Martin J. Houle, U.S. Army Dugway Proving Ground 152
Disposal and Removal of Polychlorinated Biphenyls in Soil
Robert A. Griffin, Illinois State Geological Survey 169
Land Disposal of Hexachlorobenzene Wastes: Controlling Vapor Movement
In Soil
Walter J. Farmer, University of California 182
Simulation Models and Their Application to Landfill Disposal Siting; A
Review of Current Technology
M. Th. Van Genuchten, Princeton University 191
*
Field Verification of Hazardous Waste Migration From Land Disposal Sites
James P. Gibb, Illinois State Water Survey 215
Adsorption, Movement, and Biological Degradation of High Concentrations
of Selected Pesticides in Soils
Jim M. Davidson, University of Florida 233
Modelling Contaminant Attenuation in Soil: Microbial Degradation of
Organic Matter
Graham J. Farquhar, University of Waterloo 245
Use of Pollutant Movement Predictions to Improve Selection of Disposal Sites
Mike H. Roulier, U.S. Environmental Protection Agency 255
SESSION IV: MODIFICATION OF DISPOSAL SITES AND WASTE STREAMS
Interaction of Selected Lining Materials With Various Hazardous Wastes
Henry E. Haxo, Jr., Matrecon, Inc 256
The Use of Liner Materials for Selected FGD Waste Ponds
Z.B. Fry, Department of the Army 273
Use of Limestone to Limit Contaminant Movement From Landfills
Wallace H. Fuller, University of Arizona 282
Pilot Scale Evaluation of Design Parameters for Sorbent Treatment of
Industrial Sludge Leachates
John W. Liskowitz, New Jersey Institute of Technology 299
Selection of Cover for Solid Waste
Richard J. Lutton, Department of the Army 319
Laboratory Assessment of Fixation and Encapsulation Processes for
Arsenic-Laden Wastes
Jaret C. Johnson, JBF Scientific Corporation 326
Encapsulation Techniques for Control of Hazardous Materials
H.R. Lubowitz, TRW 342
viii
-------
TABLE OF CONTENTS (Cont'd)
Page
SESSION IV: MODIFICATION OF DISPOSAL SITES AND WASTE STREAMS (Cont'd)
Field Evaluation of Chemically Stabilized Sludges
Philip G. Malone, Department of the Army 357
SESSION V: ALTERNATIVES TO LAND DISPOSAL
Land Cultivation of Hazardous Industrial Wastes
Tan Phung, SCS Engineers 366
Cost Assessment for the Emplacement of Hazardous Materials in a Salt Mine
B.T. Kown, Bechtel Corporation 377
Assessment of Deep Well Injection of Hazardous Waste
Carlton C. Wiles, U.S. Environmental Protection Agency 397
Developing Practical Methods for Controlling Excess Pesticides
Charles J. Rogers, U.S. Environmental Protection Agency 405
List of Attendees 409
-------
ACKNOWLEDGEMENTS
The Fourth Annual Symposium and these resulting proceedings were made possible by a
cost-sharing grant agreement between the U.S. EPA Office of Research and Development,
SHWRD and Southwest Research Institute. This support, from Mr. Francis Mayo, Director,
Municipal Environmental Research Laboratory, U.S. EPA and Mr. Martin Goland, President,
Southwest Research Institute, is gratefully acknowledged. In addition to the contributors
to these proceedings, the help of the following session chairmen and other contributors is
recognized: Robert L. Stenburg, Donald E. Sanning, Richard H. Carnes, Mike H. Roulier,
Norb Schomaker, and Robert E. Landreth (U.S. EPA, SHWRD, Cincinnati); Martin Goland (SwRI)
for welcoming remarks; Francis Mayo (EPA-Cincinnati) for the symposium introduction;
David Black, Thelma Greene and Rita Hogue (SwRI), and Becky Lawson (EPA) for symposium
organization assistance; Jack Harmon and Jerry Lochbaum (SwRI) for public relations and
press coordination; Gus Garcia (SwRI) for the cover art work; and Jim Pryor and staff
(SwRI) for editorial assistance. Special thanks to Bob Landreth (SHWRD), Project Officer
for his valuable guidance.
-------
CURRENT RESEARCH ON LAND DISPOSAL OF HAZARDOUS WASTES
Norbert B. Schemaker
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 re-
sponsibility for research in the areas of solid and hazardous waste management, including
both disposal and processing. To fulfill this responsibility, the SHWRD is developing
concepts and technology for new and improved systems of solid and hazardous waste manage-
ment; is documenting the environmental effects of various waste processing and disposal
practices; and is collecting data necessary to support implementation of processing and
disposal guidelines mandated by the recently enacted Resource Conservation and Recovery
Act (RCRA). 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:
1. Waste Characterization/Decomposition
2. Pollutant Transport
3. Pollutant Control
4. Pollutant Treatment
INTRODUCTION
The waste residual disposal research
strategy, encompassing state-of-the-art
documents, laboratory analysis, bench and
pilot studies, and full-scale field verifi-
cation studies, is at various stages of im-
plementation, but over the next 5 years the
research will be reported as criteria and
guidance documents for user communities.
The waste disposal research program will
develop and compile a data base for use in
the development of guidelines and standards
for waste residual disposal to the land as
mandated by the recently enacted legislation
entitled "Resource Conservation and Recovery
Act of 1976" (RCRA).
The current waste residual disposal
research program has been divided into
three general areas: (a) Design considera-
tions for Current Landfill Disposal Tech-
niques; (b) Alternatives to Current Land-
fill Disposal Techniques; and (c) Remedial
Action for Minimizing Pollutants from Un-
acceptable or Inoperative Sites.
5. Co-Disposal
6. Remedial Action
7. Landfill Alternatives/Land Cultivation
8. Economic/Environmental Impact
The waste residual disposal research
program has been discussed in previous sym-
posiums. See Schomaker, N.B. and Roulier,
M.H., Current EPA Research Activities in
Solid Waste Management: Re4eoAc/t Siflnpo&ium
on Go6 and Le.adn.atn &iom Land{,iW>: forma-
tion, CofJLe.ction and Iie.aAme.wt, March 25-26,
1975, Rutgers, State University of New
Jersey; and Schomaker, N.B., Current
Research on Land Disposal of Hazardous
Wastes: RuiduaU Management by Land
Vi&poAat'- Pnoc.e.e.din9A o& the. HazandouA
(Ua&te. ReAe.an.ck Sympo&ium, February 2-4,
1976, University of Arizona; and Schomaker,
N.B., Current Research on Land Disposal of
Municipal Wastes: Management o& Ga6 and
Le.achate. in land^UJLt,- ?ioc.e.idinQ& orf the.
Jhind Annual Municipal SoLid Wa&te. Re4e.aAcA
Symposium March 14-16, 1977, Universtiy of
Missouri.
WASTE CHARACTERIZATION/DECOMPOSITION
The overall objective of this re-
search activity is to provide information
on the composition of municipal and hazar-
-------
dous wastes, the compatibility of hazardous
wastes, and aqueous and gaseous emissions
from disposed wastes. Sampling techniques,
analytical methods, and techniques for de-
termining emissions will be developed for
implementing better disposal practices and
waste management.
The initial effort (1) in the hazardous
waste characterization area relates to the
chemical composition, physical character-
istics, and origin of hazardous wastes
delivered to several Class I (hazardous
chemical) landfills in the State of
California. The average concentration,
estimated daily decomposition, and parti-
tioning of 17 metal species in hazardous
waste landfills in the greater Los Angeles,
California area was studied. Mass deposi-
tion rates were determined by calculating
the average concentration of the metal
species in conjunction with approximate
daily volume received at the site. The
effort has been published in a report en-
titled A Coie. Study o& Hazardous, Wa6-tw
Input Into Clu& I Landfall* - EPA-600/2-
78-064, June 1978.
Standard Sampling Techniques
Standard Sampling procedures, in-
cluding collection, preservation, and
storage of samples, do not exist for solid
and semi-solid wastes. Hazardous wastes,
both at the point of generation and the
point of disposal, are not homogenous mix-
tures and, additionally, may range in con-
sistency from a liquid, through a pumpable
sludge, to a nonpumpable solid. Existing
procedures for sampling liquid effluents
and soils will have application but must
be adapted to a variety of circumstances
and, more importantly, field tested exten-
sively before they can be advocated as
"the" way to sample. Experience with samp-
ling procedures is being accumulated as
part of several on-going SHWRD projects.
The initial effort (1) in this samp-
ling technique area relates to an activity
for standardized methods for sampling and
analysis of hazardous wastes. Specialized
sampling methods have been developed for
safe, reproducible, and representation
sampling of such wastes in a wide variety
of physical states, compositions, and
locations. Methods of containment,
handling and custody of waste samples
have also been investigated.
Standard Analytical Techniques
Analysis of the contaminants within
a waste leachate sample is difficult due to
interfering agents. Existing instrumenta-
tion functions well in the analysis of
simple mixtures at moderately low concen-
trations but interference problems can be
encountered for complex mixtures at high
concentrations (1 percent by weight and
greater). In this range the sample cannot
always be analyzed directly and commonly
must be diluted and/or analyzed by the
method of standard additions. Options
are the development of standard procedures
for diluting and accounting for errors
introduced thereby or the development of
instrumentation capable of accurate,
direct measurements at high concentrations
in the presence of potential multiple
interferences. Existing USEPA procedures
for water and wastewaters are often not
applicable. Analytical procedures are
being developed on an as-needed basis as
part of the SHWRD projects. However most
of this work is specific to the wastes
being studied and separate efforts were
required to insure that more general pro-
cedures/equipment would be developed.
The initial effort (1) has resulted
in compilation of analytical techniques
used for contaminant analysis. This effort
has been published in a report entitled
Compilation of, Matkodotogy (Iked. £01 Meoau/i-
-otg Potttution PateuneteAA o£ Sanitary Land-
^JUUL Leachatu, EPA-600/3-75-011, October
1975, and SHWRD is currently finalizing a
collaborative testing study (2) on leachate
analyses. In this study, leachate samples
were sent to 50 laboratories for analysis of
specific parameters. The results will pro-
vide information on detection limits and
precision for contaminants in leachate,
using currently accepted methods developed
for water and wastewater.
Another effort which is also being
finalized, relates to the compilation and
evaluation of current leaching tests
methods. In this study (2) various avail-
able leaching tests are described and the
methods utilized are evaluated in relation
to their adaptability to field conditions.
Also, as a part of the sampling technique
(1) Parenthesis numbers refer to the project
officers, listed immediately following this
paper, who can be contacted for additional
information.
-------
activity, analysis of wastes has been de-
veloped as a four level scheme, which
begins with field characterization followed
by three levels of progressively more
detailed and instrument intensive labora-
tory analysis.
Standard Leaching Tests
In studying the potential environ-
mental impact of contaminants, a standard
leaching test is needed to assess contami-
nant release from a waste. Such a test
must provide information on the initial
release of contaminants from a waste con-
tacted not only by water but also by
other solvents that could be introduced in
disposal. Additionally, such a test must
provide some estimate of the behavior of
the waste under extended leaching. The
Office of Solid Waste (OSW) has funded
the Industrial Environmental Research Lab-
oratory (IERL), USEPA, a study (3) to
procedures for determining whether a waste
contains significant levels of toxic con-
taminants and whether a waste will release
such contaminants under a variety of
leaching conditions.
Validation of a Standard Leaching
Test (SLT) has been initiated and funded
in part by SHWRD.(2) The passage of the
RCRA on October 21, 1976 and the imposing
time restraints, necessitated developing an
Interim Standard Leaching Test (ISLT)
Existing leaching tests are being evaluated
for those elements that may be of special
benefit to the development of an SLT and
at least three candidate ISLT's have been
tested as part of the OSW/IERL project.
In conjunction with the ISLT, a genetic
toxicity test for landfill leachates is be-
ing investigated.(2) This effort is being
performed to rapidly and inexpensively as-
certain the potential mutagenicity haz-
ards of waste materials by examining the
leachates to determine the feasibility of
using short-term genetic assays to predict
chemical mutagens.
Waste Leachability
The characteristics of leachates from
hazardous wastes are being investigated in
several different studies. The initial
hazardous waste leaching effort (4) was
patterned after a method developed by the
International Atomic Energy Agency
(I.A.E.A.) for leach testing immobilized
radioactive waste solids. Translucent
plexiglass columns were utilized and ob-
servations of flow patterns as well as
possible biological activity were made.
Five industrial sludges and five Flue Gas
Desulfurization (FGD) sludges were investi-
gated. This effort is basically completed
and the final report is being prepared. An
interim report entitled Pottutant Potwtial
o£ Row and Chemtco£&/ Rtxed Hazotrfooi
Iwdai;fu.a£ Wa&teA and F£ue GOA VeAutfiuJu,-
zation S&idgu, EPA-600/2-76-182, July 1976
has been published.
Another on-going waste Teachability
effort (5) relates to inorganic industrial
waste where there is no appreciable bio-
logical activity. This testing program was
designed to evaluate leaching and pollutant
release under a variety of leaching condi-
tions which may be encountered in one or
more disposal situations and to develop a
short-term leaching and soil interaction
test. Five different types of leaching
conditions were utilized which employed
three varying pH leaching fluids, deionized
water and municipal refuse leachate. A
major consideration in the leaching be-
havior of wastes is time. Some wastes
will not release appreciable amounts of
contaminants until leaching has removed
salinity or reserve alkalinity from the
waste. Accordingly, the laboratory
leaching test is extended over time.
Successive extracts of a waste, which
change in composition as the waste is de-
pleted, are used to challenge a sequence
of three soil batches that are graded in
size to allow taking samples for analysis
between each step and to compensate for
extract absorbed by the soil. The labora-
tory leaching procedure allows simulating
one to ten years of field leaching and soil
interaction in about two weeks of laboratory
extractions. The test has not yet been
checked against field data.
POLLUTANT TRANSPORT
Pollutant transport studies involve
the release of pollutants in liquid and
gaseous forms from various municipal and
hazardous wastes and the subsequent move-
ment and fate of these pollutants in soils
adjacent to disposal sites. Both labora-
tory and field verification studies at
selected sites are being performed to
assess the potential for groundwater con-
-------
tamination. The studies will provide the
information required to (a) select land
disposal sites that will naturally limit
release of pollutants to the air and water
and (b) make rational assessments of the
need for and cost-benefit aspects of
leachate and gas control technology.
The overall objective of this re-
search activity is to develop procedures
for using soil as a predictable attenua-
tion medium for pollutants. Not all
pollutants are attenuated by soil, and,
in some cases, the process is one of
delay so that the pollutant is diluted
in other parts of the environment. Con-
sequently, a significant number of the re-
search projects funded by SHWRD are focused
on understanding the process and pre-
dicting the extent of migration of con-
taminants (chiefly heavy metals) from waste
disposal sites.
These pollutant migration studies are
being performed simultaneously in the areas
of hazardous wastes and municipal refuse.
Several previous discussions of these
efforts have been presented. See Roulier,
M.H., Attenuation of Leachate Pollutants by
Soils, presented at the Management ojj
GUI, and Leachate. in Landfitltb • PswceedlngA
OjJ the, JhlAd Annual Uuyu.CA.pa£ Solid Wa&te
ReAeaAch Symposium, March 14-16, 1977
University of Missouri.
Bibliography and State of the Art
The initial effort (5) in this area
resulted in a completed bibliography en-
titled Tian&poxt o£ Ha.zaA.dout> Substance*
Through Soil PnoceAAeA - Volume. I: Bibli-
ography ORNL-EIS-74-70 Part I and Tlan&polt
o& HazasidouA SubbtanceA Through Soil
P*.oc.e44£A - Volume. II: Pesticides - ORNL-
EIS-74-70 - Part 2. It is expected that it
will be made available through the national
technical information service (NTIS).
A second effort (5) consisted of a re-
view of information on migration through
soil of potentially hazardous pollutants
contained in leachates from waste materials.
The results have been published in a report
entitled Movement o& Selected Metal*, Aa-
beAtoA, and Cyanide. in Soil*: kpplicationA
to Wa&te Vi&potal Pnoblmb, EPA-600/2-77-022,
April 1977. The document presents a criti-
cal review of the literature pertinent to
biological, chemical, and physical re-
actions, and mechanisms of attenuation
(decrease in the maximum concentration for
some fixed time as distance traveled) of
the selected elements arsenic, beryllium,
cadmium, chromium, copper, iron, mercury,
lead, selenium, and zinc, together with
asbestos and cyanide, in soil systems.
Controlled Lab Studies
The initial effort (5) is examining the
factors that attenuate contaminants
(limit contaminant transport) in leachate
from municipal solid waste landfills.
These contaminants are: arsenic, beryl-
lium, cadmium, chromium, copper, cyanide,
iron, mercury, lead, nickel, selenium,
vanadium, and zinc. The general approach
was to pass municipal leachate as a leach-
ing fluid through columns of well charac-
terized, whole soils maintained in a
saturated anaerobic state. The typical
municipal refuse leachate was spiked with
high concentrations of metal salts to
achieve a nominal concentration of 100 mg/1.
The most significant factors in contaminant
removal were then inferred from correlation
of observed migration rates and known soil
and contaminant characteristics. This
effort will contribute to the development
of a computer simulation model for pre-
dicting trace element attenuation in soils.
Modeling efforts to date have been hindered
by the complexity of soil-leachate chemis-
try. The results of this effort have been
published in a report entitled Investigation
OjJ Land£i£l Leachate PoU.uta.nt kttenuatton
by So-UU, EPA-600/2-78-158, August 1978.
The second effort (5) in this area is
studying the removal of contaminants from
landfill leachates by soil clay minerals.
Columns were packed with mixtures of
quartz sand and nearly pure clay minerals.
The leaching fluid consisted of typical
municipal refuse leachate without metal
salt additives. The general approach to
this effort was similar to that described
in the preceding effort except that (a)
both sterilized and unsterilized leachates
were utilized to examine the effect of
microbial activity on hydraulic conduc-
tivity and (b) extensive batch studies of
the sorption of metals from leachate by
clay minerals were conducted. The results
of this effort have been published in a
report entitled Attentuation orf Pollutant*
-in Municipal Landfill Leachate by Clay
Minvialt>, EPA-600/2-78-157 August 1978
-------
The third effort (6) relates to model-
ing movement in soil of the landfill gases,
carbon dioxide and methane. The modeling
movement has been verified under labora-
tory conditions. This effort has not
focused on the impact of gases on ground-
water, but considers groundwater as a sink
for carbon dioxide. Results to date have
involved design curves and tables which
have been used to successfully evaluate a
gas problem in Minnesota.
A fourth effort (5) relates to the use
of large-scale, hydrologic simulation
modeling as one method of predicting
contaminant movement at disposal sites.
The two-dimensional model that was used
successfully to study a chromium con-
tamination problem is being developed
into a three-dimensional model and will be
tested on a well-monitored landfill where
contaminant movement has already taken
place. Although this type of model pre-
sently needs a substantial amount of
input data, it appears promising for de-
termining contaminant transport properties
of field soils and, eventually, predict-
ing contaminant movement using a limited
amount of data. A list of available simu-
lation models has been developed which
describes the advantages and disadvantages
of these models for simulating landfill
behavior.
A fifth effort (1) relates to organic
contaminant attenuation by soil. This is
our initial effort in organic contaminant
movement in soil. Much more is known about
inorganic contaminant movement in soil be-
cause the analytical techniques for in-
organic materials are well developed and
relatively cheap compared to the time-
consuming analytical techniques for
organic materials. The problem is com-
pounded by the fact that organic contami-
nants are more numerous and more are being
synthesized all the time. PCB is the
organic contaminant currently being in-
vestigated. As a part of the above de-
scribed effort, a gas chromatographic
analytical procedures was developed that
allowed improved quantitative measurement
of PCB's in aqueous solutions.
A sixth effort (5) relates to an evalu-
ation of the conditions that would control
the movement of hexachlorobenzene (HCB)
out of landfills and other disposal/
storage facilities into the surrounding
atmosphere. The potential for volatili-
zation indicates, a need for disposal site
coverings that will reduce the vapor phase
transport of HCB into the surrounding
atmosphere. The volatilization fluxes of
hexachlorobenzene from industrial wastes
(hex waste) were determined using coverings
of soil, water and polyethylene film in a
simulated landfill under controlled
laboratory conditions. Coverings of
water and soil were found to be highly
efficient when compared on a cost basis.
Volatilization flux through a soil cover
was directly related to soil air-filled
porosity and was therefore greatly re-
duced by increased soil compaction and
increased soil water content.
A seventh effort (5) relates to the ad-
sorption, movement and biological degrada-
tion of high concentrations of pesticides
in soils. Equilibrium adsorption isotherms
were obtained for 2,4-D amine, atrazine,
terbacil, and methyl parathion and four
soils from different locations within the
United States. Pesticide solution con-
centrations ranged from zero to the aqueous
solubility limit of each pesticide, The
mobility of each pesticide increased as its
concentration in the soil solution phase
increased. Pesticides degradation rates
and soil microbial populations generally
declined as the pesticide concentration
in the soil increased.
Field Verification
Limited field verification is being
conducted. The initial effort (4) to date
has consisted of installing monitoring
wells and coring soil samples adjacent to
three municipal landfill sites to identify
contaminants and determine their distri-
bution in the soil and groundwater be-
neath the landfill site. The sites repre-
sent varying geologic conditions, recharge
rates, and age, ranging from a site closed
for 15 years to a site currently operating.
Individual site characteristics were iden-
tified, and sample analyses necessary to
determine the primary pollutant levels in
the waste soils and groundwater were
determined. The result os this effort are
discussed in a report entitled Chenu.co£
and Phyt-Lcal Ejects o{ Municipal. laM^UJU
on (indeAJUjAjfiQ So and GwundwateA
EPA-600/2-78-096, May 1978.
A second effort (5) relates to the
vertical and horizontal migration patterns,
of zinc, cadmium, copper, and lead through
-------
the soil and shallow aquifer systems at a
secondary zinc smelter. These patterns
were defined using soil coring and
monitoring well techniques. The migration
of metals that occurred has been limited to
relatively shallow depths into the soil
profile by attenuation processes. Soil
coring was determined to be an effective
investigative tool, but not suitable by
itself for routine monitoring of waste
disposal activities. However, it should
be used to gather preliminary information
to aid in determining the proper horizontal
and vertical locations for monitoring well
design.
POLLUTANT CONTROL
Pollutant control studies are
needed because experience and case
studies have shown that some soils will
not protect groundwater from contaminants.
Even sites with "good soils" may have to
be improved to protect against subsurface
pollution. The overall objective of this
research activity is to lessen the impact
of pollution from waste disposal sites by
technology that minimizes, contains, or
eliminates pollutant release and leaching
from waste residuals disposed of 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 mi-
grates through the soil from landfill
sites.
Natural Soil Processes
The treatment by natural soil processes
of pollutants from hazardous waste and muni-
cipal refuse disposal sites is being per-
formed in the controlled lab studies pre-
viously discussed in the section on pollu-
tant transport.
Liners/Membranes/Admixtures
The liner/membrane/admixture tech-
nology (4) is being studied to evaluate
suitability for eliminating or reducing
leachate from landfill sites of municipal
or industrial hazardous wastes. The test
program will evaluate, in a landfill en-
vironment, the chemical resistance and
durability of the liner materials over
12- and 36-month exposure periods to
leachates derived from industrial wastes,
SOX wastes, and municipal solid wastes.
Acidic, basic, and neutral solutions will
be utilized to generate industrial waste
leachates.
The liner materials being investi-
gated under the hazardous waste program
include five admixed materials and eight
polymeric membranes. The admixed
materials are:
- Asphalt emulsion or nonwoven fabric
(0.3 in)
- Compacted native fine-grain soil (12.0in)
- Hydraulic asphalt concrete (2.5in)
- Modified bentonite and sand (5.0in)
- Soil cement with and without surface
seal (4.0in)
The eight polymeric membrane liners
are:
- Butyl rubber, fabric reinforced (34 mils)
- Chlorinated polyethylene (32 mils)
- Chlorosulfonated polyethylene, fabric
reinforced (34 mils)
- Elasticized polyolefin (20 mils)
- Ethylene propylene rubber (50 mils)
- Neoprene, fabric reinforced (32 mils)
- Polyester elastomer, experimental (8 mils)
- Polymeryl chloride (30 mils)
Specimens of these materials have
been exposed for more than 2 years to the
following six classes of hazardous wastes
which utilized ten specific types of
wastes: strong acid; strong base; waste
of saturated and unsaturated oils; lead
waste from gasoline tanks; oil refinery
tank bottom waste (aromatic oil); and
pesticide waste.
Preliminary exposure tests have been
completed on the various liner materials
in the various waste in order to select
combinations for long term exposures. The
results of these tests along with a dis-
cussion of the overall hazardous waste
liner material program are presented in a
report entitled LineA. MateAi&tb Exposed
to Ha.zM.doui> and TOJU.C Sludge* Tritet
Intzrujn Re.povt, EPA-600/2-77-081, June 1977.
A second effort (4) relates to a state-
of-the-art of landfill impoundment tech-
niques. This literature search surveyed
the use of liner materials in impoundment
sites for the containment of seven general
types of industrial wastes. This data
was supplemented with information ob-
-------
tained from various manufacturers,
suppliers and installers and contains
analyses of liner compatibility with
industrial wastes.
A third effort (4) relates to the types
of materials being tested for use as
liners for sites receiving sludges gener-
ated by the removal of sulfur oxides
(SOx) from flue gases of coal-burning
power plants. The volumes of SOx sludge
generated in any particular place will,
typically, be much greater than those for
other types of wastes, and therefore the
.disposal sites will be large. Conse-
quently, methods of lining such disposal
sites must have a low unit cost It is
desirable that the materials be easy to
apply or install. Because of these con-
siderations, the number of polymeric
membranes included in the study have been
reduced whereas admixed and sprayed-on
materials are being emphasized. A total
of 18 materials are being tested with two
types of Flue Gas Desulfurization (FGD)
sludges. The sludges are from an eastern
coal lime and limestone scrubbed process.
The liner materials consisted of ad-
mixed material, prefabricated membranes,
and spray on materials. The admixed materi-
als consisted of the following: cement;
lime; cement with lime; polymeric bentonite
blend (Ml79); gray powder-guartec (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 follow-
ing: polyvinyl acetate; natural rubber
latex; natural latex; polyvinyl acetate;
asphalt cement; and molten sulphur.
For this above effort, a total of 72
special test cells were constructed to per-
form 12 and 24 month exposure tests. The
12-month exposure tests has been completed
and the report is being prepared.
A fourth effort (4) relates primarily
to the identification and description of
waste disposal sites and holding ponds
which have utilized an impermeable lining
material. Also, three potential excavation
techniques for liner recovery operations
are described and discussed. This effort
has been published in a report entitled
fan. Sanitary Land^Mi, and
and HazoAcfoai Ua&te.
9-78-005, May 1978.
Chemical Stabilization
EPA-600/
Chemical stabilization is achieved
by incorporating the solid and liquid
phases of the waste into a relatively
inert matrix which is responsible for
increased physical strength and which
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
so that no serious stresses are exerted
on the environment around the disposal
site, then the wastes have been rendered
essentially harmless and restrictions on
where the disposal site may be located
will be minimal.
The initial chemical fixation effort
(4) relates to the transforming of the
waste into an insoluble or very low solu-
bility form to minimize leaching. The
test program consists of investigating
ten industrial waste streams, both in the
raw and fixed state. The waste streams
were treated with at least one of seven
separate fixation processes and sub-
jected to leaching and physical testing.
The fixation process and sludge
assignment matrix is shown below.
The lab studies have been completed
and the field studies initiated. The
laboratory leaching test data for the first
6 months includes: methods for physical
and chemical analyses, documentation of v
various sludge fixation processes, and
physical and chemical data on the sludges.
The results have been compiled and dis-
cussed in a report entitled PoUutant
Potential otf Raw and ChemicjaJUtjLj F-txed
HazaAdouA InduA&ual Isla&tu and ffue. Goi
VeAulfiustization Sludge* - InteMm
EPA-600/2-76-182, July 1976.
The laboratory physical program test
included physical properties (grain size
distribution, atterbug limits, specific
gravity, volume-weight-moisture relation-
ships and permeabilities), engineering
properties (compaction and unconfined com-
pression) and durability properties (wet-
dry and freeze-thaw) for both the raw and
fixed sludges. The results of this effort
have been discussed in a report entitled,
Ph.yAA.caJ. and Engineering PiopeJitiu o& ike.
HazaJidou& InduAtsUal Wa&te* and Sludge*,
-------
Sludge Category
FGD, lime process, eastern coal
Electroplating
Nickel /cadmium battery
FDG, limestone process, eastern coal
FGD, double alkali process, eastern coal
FGD, limestone process, western coal
Inorganic pigment
Chlorine production, brine sludge
Calcium fluoride
FGD, double alkali, western coal
A
X
X
X
X
X
X
X
X
X
B
X
X
X
X
X
X
X
X
X
Processes
C D
X X
- -
_ .
_ _
_ _
x
+
-
E
X
_
_
X
X
x
X
F G
X X
_ _
+ x
+ X
X X
'» f\
+ X
X = Sludge actually fixed by processor and placed in column.
+ = Sludge evaluated by processor but not fixed for this study.
- = Sludge not evaluated by processor.
EPA-600/2-77-139, August 1977.
The second effort (4) relates to a
survey and identification of solidifica-
tion/stabilization technology available
in addition to those techniques currently
being investigated. This effort consisted
of examining six major categories of indus-
trial waste fixation systems and discussing
the advantages/disadvantages of each. The
six categories are shown below:
o Cement - based techniques
o Lime - based techniques
o Thermoplastic techniques (including
bitumen, paraffin and polyethylenes)
o Organic polymer techniques
o Self-cementing techniques
Another aspect of this effort consisted of
listing and describing some 13 companies
that solidify or fix hazardous wastes or
sell fixation materials. The survey and
identification have been completed and the
report is being prepared.
The third effort (4) relates to a
series of field verification studies to
verify success with which pollutants have
been immobilized at landfills receiving
stabilized hazardous wastes. Four sites
where stabilized industrial wastes had been
disposed were examined to determine the
effects of stabilized wastes on surrounding
soils and groundwater. The sludges had all
been fixed using the same proprietary proc-
ess. Two of the industrial waste sites
contained auto assembly (metal finishing)
wastes, one site contained electroplating
wastes and the fourth site contained re-
finery sludges. The physical properties of
soils under the disposal sites were affected
little, if at all, by the disposal operation.
A fourth effort (2) relates to a lab-
oratory assessment of fixation and encap-
sulation processes for arsenic-laden wastes.
Three industrial solid wastes that are high
in arsenic concentration have been treated
by generic processes in laboratory and by
proprietary processes at vendors' facili-
ties. Leaching studies on treated wastes
consisting of Shake tests on pulverized
samples and on intact monolithic samples
are being performed to assess the relative
safety of each product for disposal.
A fifth effort (7) relates to encap-
sulating process for managing hazardous
wastes. This study consists of developing
and evaluating techniques for encapsulating
hazardous wastes. Techniques for encapsu-
lating unconfined dry wastes are discussed
in a report entitled Ve.vzlopme.nt otf a.
PofymeAAc. Cejne.ivU.ng and Encapsulating
Ploc&64 fafi Managing HazaidouA Wa&tu,
EPA-600/2-77-045, August 1977. Additional
evaluations are currently being performed
whereby containers of hazardous waste (i.e.,
55 gallon drums) are placed in a fiber
glass thermo setting resin casing and the
casing is covered with a high density poly-
ethylene. Laboratory tests are being per-
formed: to evaluate stresses encountered
during storage, transport and disposal in a
landfill; to leaching by water and HCL, and
-------
to mechanical tests to evaluate the ability
to retain toxic materials. Also being
evaluated is the use of cements for encap-
sulation of small containers.
Moisture Infiltration
Moisture infiltration is the primary
mechanism which can control pollutant gener-
ation from landfills. This technology is
being studied under an initial effort (4)
which is developing a criteria matrix where-
by the various functional requirements of
cover material can be described and evalu-
ated for each of the various soil types
occurring in the U.S.A.
POLLUTANT TREATMENT
The overall objective of the pollutant
treatment studies is to develop chemical,
physical, and biological processes for
treating landfill leachate once it has been
collected and contained at the landfill
site. Treatment by natural soil processes
may be used at some sites where pollutant
retention by soils is sufficient to allow
uncontrolled release of leachate from the
landfill into underlying soils.
Natural Soil Processes
The treatment of pollutants (5) from
hazardous waste and municipal refuse dis-
posal sites by natural soil processes has
been described previously in the section
on pollutant transport.
Physical-Chemical Treatment
The treatment of municipal refuse
leachate (6) was evaluated in several
laboratory studies using the following
physical -chemical processes: chemical
precipitation, activated carbon adsorption,
reverse osmosis, ion exchange, ozonation,
and chemical oxidation. The activated
carbon was quite effective in removing
refractory organics in the effluent of
biological units. Results of the initial
efforts have been reported in a journal
article entitled, Chemical. Treatment otf
Leac.kateA iiam Sanita/iy land^WLk, JWPCF,
Vol. 46, No. 7, July 1, 1974, pp. 1776-1791
and in a report entitled Sanitasiy Landfill.
Stabilization ulitk Ig^gj^g- Rg-g-t/g-le. and
Treatment, EPA-600/2-75-043,
October 1975. The expansion (6) of the
initial efforts to include chemical oxida-
tion, ion exchange and ozonation has been
discussed in two reports entitled Evaluation
o& Leackate. Tti&atmewt'- Volume. I - Ckafiac.-
tesuAtic* of, Leackate., EPA-600/2-77-186a,
September 1977 and Evaluation ofi Leackate.
Tn.eat3ne.nt: Volume. II - Biological, and
Pkyt-ical-Ckemical PAoc.eA4W, EPA-600/2-77-
186b, November 1977.
A third effort (5) relates to a labbr-
atory study of agricultural limestone and
hydrous oxides of Fe. This study was per-
formed to evaluate their use as landfill
liner materials to minimize the migration
of metal contaminants. Preliminary re-
search on limestone and Fe hydrous oxide
liners indicates these materials have a
marked retarding influence on many of the
trace elements. The limestone barrier
showed the migration rate of all 12 metals
studied and it was more effective in re-
tention of some metals than others.
However, the increased water contamination
from solubilization of iron seems to rule
out use of iron oxides until further work
is conducted.
A fourth effort (5) relates to a lab-
oratory evaluation of ten natural and syn-
thetic materials (bottom ash, flyash, ver-
miculite, illite, Ottawa Sand, activated
alimrinia, oillite) for the removal of
contaminants in the leachate and liquid
portion of three different industrial
sludges (calcium fluoride sludge, petroleum
sludge, metal finishing sludge). This in-
vestigation Involves beaker studies to
evaluate the static adsorption capacity of
sorbent materials using maximum background
concentrations of contaminants in the
leachate, followed by column studies to
obtain information regarding the dynamic
adsorption of leachate resulting from a
calcium fluoride sludge. Results of the
laboratory experiments indicate sorbent
capacity is a function of the pH and con-
centration of the particular contaminant
in the leachate with the volume of leachate
that can be treated with maximum removal
being regulated by the velocity through the
bed. The information is discussed in a
report entitled Soxbent* fan fluoride.,
Metal Fini&king and Pe&ioieum Sludge.
Leackate. Contaminant Con&iot, EPA-600/2-
78-024, March 1978.
Biological Treatment
Various unit processes for biological
treatment of municipal refuse leachate
have been investigated in the laboratory.
-------
The results of this initial effort (6) have
been reported in a journal article entitled,
Tteatabitity o& Leachatz &lom Sanitary
LandtiUs, JWPCF Vol. 4, No. 6, June 1974,
pp. 860-872"! TTsecond effort (6) has in-
vestigated the process kinetics, the
nature of the organic fraction of municipal
refuse leachate, and the degree of treat-
ment that may be obtainable using con-
ventional wastewater treatment methods.
The biological methods evaluated were the
anaerobic filter, the aerated lagoon, and
combined treatment of activated sludge and
municipal sewage. Biological units were
operated successfully without prior removal
of the metals that were present in high
concentrations. The results of this effort
are discussed in the same two EPA leachate
treatment reports mentioned under the above
physical/chemical treatment section.
Thermal Decomposition
Treatment by thermal decomposition
relates to the establishment of time-temper-
ature relationships for incinerating pesti-
cides. Specifically, through the test pro-
gram, existing information will be summa-
rized into a state-of-the-art document and
experimenta1 i nci nerati on/decomposi ti on
studies will be conducted on approximately
40 pesticides. A lab scale evaluation/
confirmation study and a pilot scale in-
cinerator study are currently being
performed.
The initial effort (1) related to
the determination of incineration condi-
tions necessary for safe disposal of
pesticides. An experimental incinerator
was constructed and utilized to determine
the time-temperature conditions needed
for the safe destruction of pesticides.
This effort is discussed in a report en-
titled VeteMminatsion o£ Incinerator Oper-
ating Condition* NtceSAary fior Sa.^e. Vis-
po&at o^ Pesticides, EPA-600/2-75-041,
December 1975. This research has been
supplemented by another effort documenting
in detail the various research projects
relating to thermal destruction of pesti-
cides. Efficiencies of combustion, resi-
dence time, and other parameters for safe
incineration were documented. This effort
is discussed in a report entitled Summation
of, Condition* and Investigation &or the.
Complete. Combustion o& Organic. Pesticides
EPA-600/2-75-044, October 1975.
A second effort (1) relates to the
development of laboratory scale methods
for determining the time-temperature
relationships for the decomposition of
pesticides. A specialized laboratory
system incorporating a two-stage quartz
tube was utilized for determining the
thermal destruction properties of pesti-
cides and other hazardous organic sub-
stances. With this system, a small sample
was first converted to the gas phase, then
exposed to high-temperature destruction
conditions in flowing air. With this
system, high-temperature decompositions of
Kepone, Mirex, DDT, and PCB's have been
studied.
This effort has been discussed in two
reports entitled Laboratory Evaluation o£
Higk-TempeAature. Destruction oft Kepone and
Ke£ate.d Pesticides, EPA-600/2-76-299, Decem-
ber 1976 and La.boratory Evaluation of, High.--
Temperature. Destruction o$ Poly
-------
field lysimeters. Material flows were
measured and characterized for the continu-
ing study and related to leachate quality
and quantity, gas composition and produc-
tion, and microbial activity.
The industrial wastes investigated
were: petroleum sludge, battery production
waste, electroplating waste, inorganic
pigment sludge, chlorine production brine
sludge, and a solvent-based paint sludge.
Also, municipal digested primary sewage
sludge dewatered to approximately 20 per-
cent solids was utilized at three different
ratios. Initial results of this effort are
discussed in a paper entitled "The Effects
of Industrial Sludges on Landfill Leachate
and Gas," PnotadUnQ* - National CowrfeAence
on V-i&poAal o& R&>-idute on Land, September
TS76, pp. 69-76.
A second effort (5) to assess the
potential effects of co-disposal involves
the leaching of industrial wastes in the
laboratory with municipal landfill leachate
as well as water. Results to date indicate
that, when compared with water, municipal
landfill leachate solubilizes greater
amounts of metals from the wastes and pro-
motes more rapid migration of metals through
soil. The municipal landfill leachate is a
highly adorous material containing many
organic acids and is strongly buffered
at a pH of about 5. Consequently, it has
proved to be a very effective solvent.
This study examines only the solubili-
zation of industrial waste by landfill
leachate. In the landfill, the opportunity
exists for retention of solubilized materials
by the organic fraction of the municipal
refuse. This will be evaluated in the field
lysimeter study described above.
A third effort (4) relates to chemi-
cally treated and untreated industrial
wastes being disposed of in a simulated
municipal refuse landfill environment.
Large lysimeters, six foot in diameter by
twelve feet high, are being utilized to de-
termine the difference in leachate quality
between the treated and untreated wastes
when mixed with municipal refuse. The in-
dustrial sludges selected for the study were
calcium fluoride, chlorine brine production
and electroplating. The sludges were chemi-
cally fixed by two processors. The material
was allowed to cure as per the processors
recommendations. Leaching data will be cor-
related with the laboratory leaching data to
determine the benefits of co-disposing of
treated wastes.
REMEDIAL ACTION
The Office of Solid Waste (OSW) has
concluded the investigation of 391 damage
cases. Fifteen percent of these cases in-
volved groundwater pollution from hazardous
waste landfills, 25 percent involved ground-
water pollution from indiscriminate dumping
practices, and 40 percent involved leachate
problems. Nine percent or 35 of the 391
damage cases involved well pollution. In
order to assist OSW in resolving this con-
cern, this research activity was created
with an overall objective to determine the
best practical technology and economical
corrective measures to remedy these land-
fill leachate and gas pollution problems.
The initial effort (2) is to provide local
municipalities and users with the data
necessary to make sound judgments oh the
selection of viable, in-situ, remedial
procedures and to give them an indication
of the cost that would be associated with
such a remedial project. This research
effort consists of three phases. Phase I
will be an engineering feasibility study
that will determine on a site specific
basis the best practicable technology to be
applied from existing neutralization or
confinement techniques. Phase II will de-
termine the effectiveness, by actual field
verification, of the recommendations/first
phase study. Phase III will provide a site
remedial guide to local municipalities and
users. A guidance manual is currently being
prepared which will assist the user community
in the selection of available engineering
technology to reduce or eliminate leachate
generation at inoperative landfills. Five
categorical areas are discussed: Surface
water control, groundwater control, plume
management, chemical immobilization and
excavation/reburial.
LANDFILL ALTERNATIVES/LAND CULTIVATION
Waste materials are primarily deposited
in sanitary landfills or incinerated. Be-
cause of concern for environmental impact
and economics, other landfill alternatives
have been proposed. For SHWRD purposes,
the alternatives currently being considered
are: (1) deep well injection, (2) under-
ground mines, (3) land cultivation, and
(4) saline environments. The overall ob-
jective of the landfill alternatives study
is to determine the feasibility and bene-
11
-------
ficial aspects of these techniques by
assessing the environmental impact and
economics.
Deep Well Injection
The first effort (7) consisted of a
review and analysis of available information
related to deep-well injection, and an as-
sessment as to the adequacy of this method
for managing hazardous wastes and ensuring
protection of the environment. The study
provided a comprehensive compilation of
available information regarding the injec-
tion of industrial hazardous waste into
deep wells. This effort has been discussed
in a four volume report entitled, Rev-tew
and M&eA&ment ojj Veep-Well Injection o{>
Hazasidout, Wa&te, Volumes I-IV, EPA-600/2-
77-029 (a-d), June 1977.
A second effort (5) provided detailed
information on the application of deep well
injection technology. Local geologic and
hydrologic characteristics of the injection
and confining intervals are considered
along with the physical, chemical and bio-
logical compatability of the receiving zone
with the wastewater to be injected. Design
and construction aspects of injection wells
are presented along with recommended pre-
injection testing, operating procedures,
and emergency precautions. This effort is
discussed in a report entitled, AM Intro-
duction to the. Technology o& SubAuAface
Wutemten Injection, EPA-600/2-77-240,
December 1977.
Underground Mines
The initial effort (7) consisted of a
review and analysis of information of the
placement of hazardous waste in mine open-
ings. The study assessed the technical
feasibility of storing nonradioactive haz-
ardous wastes in underground mine openings.
The results showed that a majority of the
wastes considered can be stored underground
in an evironmentally acceptable manner if
they are properly treated and containerized.
This effort is discussed in a report entitled
Evaluation ofa HazasidouA Wai>te& Emplacement
in Mined Opening*, EPA-600/2-75-040, Decem-
ber 1975.
A second effort (4) related to an eco-
nomic evaluation of storing non-radioactive
hazardous wastes in a typical room and
pillar type salt mine. The results include
capital and operating costs and a cost
analysis. The cost study was based on a
simulated waste characteristic and con-
ceptual design of the waste receiving,
treatment, containerization, and under-
ground storage facilities. This effort is
discussed in a report entitled CoAt A*4e44-
ment fai the. Emplacement ofi Hazardous
UatViiaU In a Salt Mine, EPA-600/2-77-215,
November 1977.
Land Cultivation
The disposal technique of land culti-
vation, whereby specific waste residues
have been directly applied or admixed into
soils, has been an alternate disposal op-
tion for many years by pharmaceutical,
tannery, food processing, paper and pulp,
and oil refinery industries. The soil en-
vironment can assimilate most types of or-
ganic waste by processes of adsorption, di-
lution, biodegradation, and oxidation. The
initial effort (4) relates to gathering and
assessing available information on land
cultivation of hazardous industrial sludges
with emphasis on characterization of waste
types, quantities, operational technology,
economics and environmental impacts. The
results of this initial effort are discussed
in two reports entitled, Land Cultivation
o$ Industrial Wastes and Municipal. Solid
Wastes: State.-orf-the-Mt Study Volume I -
Technical Summa/iy and Lutefocttune Survey,
EPA-600/2-78-140a August 1978 and State-o^-
the Mt Study Volume II - field Investiga-
tion and Cat>e Studies, EPA-600/2-78-140b
August 1978.
The second effort (4) is a combina-
tion laboratory, greenhouse and field
study to determine the fate and mobility
of wastes in soil for the purpose of de-
veloping criteria for use in the design,
management and monitoring of land culti-
vation disposal operations. Decomposi.-
tion rate, application rate, plant sur-
vival and growth, pollutant runoff and
leachate generation will be obtained in
development of the data base.
The third effort (4) relates to de-
tailed field surveys and limited labora-
tory field experimentation for the purpose
of developing a matrix of industrial or-
ganic and inorganic and municipal waste
streams versus operational parameters.
This matrix of information will be used
to develop design and guideline criteria.
The fourth effort (8) relates to the
12
-------
pit disposal method for excess pesticides
as generated by farmers and agricultural
applicators. Experimental systems have
been developed to investigate the fate of
selected pesticides in isolated micro-pits
under controlled conditions. The chemical
and biological consequences of pit disposal
of dilute insecticides, fungicides, and
rodenticides are being determined.
Saline Environment
This saline environment effort relates
entirely to municipal solid wastes and
therefore will not be discussed in detail
at this symposium. This effort is pri-
mar.ily a paper study detailing the present
environmental and economic status of this
type of disposal option. It will be sup-
plemented by case histories and state regu-
lations and policies in effect for those
states bordering saline waters.
ECONOMIC/ENVIRONMENTAL IMPACT
The use of market-oriented incentive
(disincentive) mechanisms has received very
scant consideration for pollution control
policy in the United States, particularly
in the area of hazardous waste management.
Economic theory suggests that incremental
pricing of waste collections and disposal
would reduce the waste generation rate,
enhance source separation of recyclable
materials, accelerate technological
innovation, and minimize total system cost.
The economic relevance is being
addressed to hazardous waste management in
general and the environmental impact aspect
is being addressed to Flue Gas Cleaning
(FGD) sludge disposal.
In the economic relevance effort (9)
currently being investigated, a methodology
is being developed that permits economic
and social impacts of alternative approaches
to hazardous waste management to be ad-
dressed. The procedure involves generation
of a series of environmental threat scen-
arios that might arise from the use of dif-
ferent hazardous waste management techni-
ques. The costs attributable to any
technique comprise the control costs, and
the environmental costs and benefits to-
gether determine the net benefits associ-
ated with the threat scenarios.
In the environmental impact effort (2)
the problem of FGD sludge disposal to the
land has been addressed. This effort con-
sidered the problem from a potential regula-
tory approach by evaluating the existing
data base and projecting its potential
impact on the promulgation of sludge dis-
posal regulations. This effort has been
published in a report entitled Vcuta. Bo&e
fan StandandA/Re.gu£ation>> Ve.ve£opment fat
Land ViApotal o& ffue. Gat> C£e.a.ning S£udge6,
EPA-600/7-77-lia December 1977.
CONCLUSION
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 Direc-
tor, Solid and Hazardous Waste Research Di-
vision, Municipal Environmental Research
Laboratory, U.S. Environmental Protection
Agency, 26 West St. Clair Street, Cincin-
nati, Ohio 45268. Information will be pro-
vided with the understanding that it is
from research in progress and that con-
clusions may change as techniques are im-
proved and more complete data become
available.
PROJECT OFFICERS
All the Project Officers, except for
Mr. Michael Gruenfeld, are associated with
the Solid and Hazardous Waste Research
Division (SHWRD), whose address is shown
above.
1. Mr. Richard A. Carnes (SHWRD)
513/684-7871
2. Mr. Donald E. Sanning (SHWRD)
513/684-7871
3. Mr. Michael Gruenfeld, Industrial En-
vironmental Research Laboratory, U.S.
Environmental Protection Agency,
Edison, New Jersey 08817
201/321-6625
4. Mr. Robert E. Landreth (SHWRD)
513/684-7871
5. Dr. Mike H. Roulier (SHWRD)
513/684-7871
6. Mr. Dirk R. Brunner (SHWRD)
513/684-7871
7. Mr. Carlton C. Wiles (SHWRD)
513/684-7881
8. Mr. Charles J. Rogers (SHWRD)
513/684-7881
9. Mr. Oscar W. Albrecht (SHWRD)
513/684-7881
13
-------
INDUSTRIAL HAZARDOUS AND TOXIC WASTE PROGRAM OF EPA's
INDUSTRIAL POLLUTION CONTROL DIVISION - IERL-CI
Leo Weitzman § David L. Becker
U. S. Environmental Protection Agency
Cincinnati, Ohio 45268
ABSTRACT
The Industrial Pollution Control Division has responsibility for research into control
of all pollutants (air, water and solid waste) from the following type of industries: (1)
organic chemicals, (2) inorganic chemicals, (3) nonferrous metals, (4) food, (5) wood
and wood products and (6) miscellaneous industries related to these. The program is an
integrated multimedia approach to these problems and ranges from assessment to ultimate
disposal and economical reuse of waste materials. The assessment program identifies poten-
tially hazardous pollutants from these industries and methods of solution are then sought.
This paper discusses the portions of this program relating to the disposal, detoxification
and reuse of hazardous wastes from these industries. The ultimate disposal and resource
recovery portions of the program are given particular emphasis.
INTRODUCTION
The Industrial Environmental Research
Laboratory (IERL) is a new addition to the
office of Research and Development of EPA.
It was formed out of parts of earlier EPA
entities, including a portion of the Indus-
trial Wastewater Treatment Laboratory in
Edison, New Jersey, a portion of the Nation-
al Environmental Research Center in Corval-
lis, Oregon, and other groups throughout
the country. The IERL-CI is one of two
such laboratories, the other being in Re-
search Triangle Park, North Carolina.
These laboratories are unique in the
EPA, in that rather than dealing strictly
w^th air and water pollution or solid waste
generation and disposal, they deal with the
industries and their pollution problems on
a multimedia basis. As a result, the two
laboratories' programs are split along
industry lines rather than medium lines.
For example, the IERL-RTP has responsibil-
ity for the Ferrous Metals Industry. While
the IERL-Cincinnati has responsibility for
the Nonferrous Metals Industry. Similarly,
in the Organics area, IERL-RTP assumes
responsibility for Petroleum Refining while
Cincinnati has responsibility for the pro-
duction of Industrial Organic Chemicals and
related processes. As can be seen, there
is a great deal of overlap between the pro-
gram, so close cooperation between the lab-
oratories is essential.
The IERL, Cincinnati consists of three
Divisions, The Resource Extraction and Hand-
ling Division (REHD), The Energy Systems
Environmental Control Division (ESECD), and
the Division, which I will be talking about
today, - The Industrial Pollution Control
Division (IPCD).
The IPCD is made up of three Branches.
The Organic Chemicals and Products Branch
(OCPB), The Metals and Inorganic Chemicals
and Products Branch (MICE), and the Food
and Wood Products Branch (FWPB). Each
Branch has the responsibility to investi-
gate multimedia pollution from the indus-
tries under its jurisdiction. The program
while considering all the classical pollu-
tants such as COD and BOD in water, and
paniculate, nitrogen oxide, etc. in air,
is mainly concerned with the control of
toxic and hazardous materials in each of
these media. As a result, it is often
14
-------
difficult to separate the solid waste as-
pects of the program, especially when deal-
ing with the measurement of pollutants or
identification of the type of waste.
The IPCD's program is broken up into
three general categories:
1. Industry assessment
2. Control equipment and disposal techno-
logy development and evaluation
3. Control equipment and disposal techno-
logy demonstration
Becasue we were one of the first
groups in EPA to start looking at indus-
tries as units rather than as individual
sources of pollutants, the principal por-
tion of our program to date has been in the
assessment area. These assessments are
multimedia and are generally done on a full
industry. Some of these that have been
done or are in the latter stage of comple-
tion are:
1. Secondary Nonferrous Metals Industry
2. Pharmaceutical Industry
3. Compounding and Fabricating Industries
These industries were chosen partly
because it was felt that they were complex
enough that they should be looked at as a
unit, and partly in order to satisfy the
pressing needs of various EPA program
offices, especially the Effluent Guidelines
Division and the Office of Air Quality
Planning and Standards. When these assess-
ments were started it was quickly realized
that many studies of the individual indus-
tries were either completed or were under
way. Typically, these studies addressed
only one aspect of an industry, such as
air pollution, water pollution, energy, or
regulatory investigations. A review of the
situation led to the decision that an over-
all industry assessment was needed to col-
lect all available information. Over the
succeeding months, past studies were ex-
tended and integrated. Many single media
oriented research, demonstration, and deve-
lopment programs were assembled into a
multimedia approach which then resulted in
the skeleton for the assessment. Next the
data was examined in the light of defining
what does the industry look like, what are
its bounds, what are the major processes
of it, and what are the likely potential
environmental impacts from it. The emphasis
on the last was generally placed on poten-
tially toxic and hazardous substances being
emitted in any one of the three pollutant
media.
Each production process or (in the
case of the very complex industries such as
the Organic Chemicals or Pharmaceutical
Industries) some representative processes
are described in the following terms:
1. Function
2. Feed materials
3. Operating conditions (temperature,
pressure, etc.)
4. Utility requirements
5. Waste streams
6. Pollution control technology
7. Occupational and health effects
Obviously, we would often find that
much of this data, especially the health
and ecological effect data is just not
available. In addition, we found with
some industries that industry secrecy pre-
vented a complete description of processes
involved. In fact, with some industries we
could not even obtain the total tonnages of
product produced because of various confi-
dentiallity requirements of the companies
involved.
The next step of the assessment in-
volves a detailed engineering analysis of
the production processes in the industry.
This analysis is performed for three pri-
mary reasons:
1. To identify obscure discharges
2. To determine if closer attention
should be given to a unit operation
that has traditionally been ignored
or considered only a limited environ-
mental problem
3. To predict the quantity and character
of the discharges in each of the three
media associated with each unit oper-
ation, based on the materials process-
ed production capacity, and process
conditions.
The engineering analysis is conducted
by senior engineering personnel. Often
little quantitative information is avail-
able on the discharges from many industrial
processes. Therefore, a process-by-process
analysis must be made of inputs, outputs,
and operating conditions to determine the
presence of unidentified potentially hazar-
dous substances.
The engineering analysis is followed
15
-------
by an in depth analysis of the same nature,
once again relying on engineering know how
of the senior staff doing the analysis. In
this way the assessment study hopes to
identify the industry and its potential for
pollution.
The study then goes through a series
of reviews by various experts in the field
as well as the industry involved and cul-
minates in the establishment of a computer-
ized data base on the industry. This data
base includes location of plants, products,
(where available), production rates, pollu-
tants involved, etc. Such a data base has
been established for the Organic Chemicals
Industry and the Pharmaceutical Industry
and it will be also established for the
plastics and synthetics, inorganic chemicals
and the pigments and dyes industries.
As can be seen, the assessment work is
quite complicated and can be expensive. The
second phase, the control equipment or dis-
posal technology development phase is of
course the area in which the attendees of
this conference are most interested. This
phase is just getting under way. The IPCD
has identified a number of waste streams
that do need a great deal of attention, and
we now are examining various technologies
for their control and ultimate disposal.
Currently the IPCD's emphasis is mainly in
the development of sound disposal or des-
truction programs for such waste, and where
economical processes for this can be iden-
tified (and economics is the key to this)
recovery processes for organic materials
and heavy metals.
In dealing with a hazardous material
we feel the best choice is actually des-
troying it rather than placing it in a
landfill where it may end up being just
stored for a long period of time. Because
of its common use, incineration is an ob-
vious alternative to landfilling for many
such cases. It is a well established tech-
nique and, when properly done, eliminates
the hazard rather than just storing it for
tomorrow. Unfortunately, as the Agency's
experience with the disposal of Kepone,
PCB's and other such materials and sludges
have shown, there is no real way of know-
ing, prior to actual test burns, that a
given incinerator can safely destroy a
particular waste. A lengthy and expensive
field testing procedure is required for the
incineration of a given hazardous material.
The problem was well illustrated by
the approach being used to dispose of Ke-
pone and Kepone-contaminated residuals.
Initially, little was known about the ther-
mal destruction characteristics of Kepone
so samples were sent to the University of
Dayton Research Institute for a laboratory
scale thermal decomposition analysis of the
material. This work was done by Duval and
Ruby who will be giving a paper on this
later today. Their work determined that at
too low a temperature, Kepone can produce
hazardous substances upon decomposition.
They established that at sufficiently high
temperatures Kepone could be destroyed to
relatively harmless substances.
Having predicted the ideal behavior of
the compound, it was necessary to go to
pilot scale studies. Based on these exten-
sive very complicated studies involving a
great deal of precautions (Dick Carnes will
be discussing this later) a large incinera-
tor was proposed to destroy the wastes.
Even after this incinerator has been built,
it will no doubt require extensive testing'
and modification before it can be used on
an operating basis to destroy the large Ke-
pone stocks stored there (if past experi-
ences are a good indicator).
The problem is that no method of scal-
ing up incinerators exist. Because of a
lack of design criteria, one cannot say that
if one incinerator works that another unit
will do the job as well. While there are
rough guidelines for doing this they are
often inaccurate and their implementation
is very expensive. Such a situation may be
tolerated when a non-hazardous material is
being destroyed, but with hazardous substan-
ces one does not have that maneuverability.
It is necessary to establish whether a giv-
en incinerator can safely destroy a waste
without going through the time and expense
occurred in the above mentioned Kepone work.
This is a long-term program and it is hoped'
that with time we can approach such ideal-
izations .
The program has four parts:
1. Laboratory scale waste and composition
studies
2. Development of general guidelines for
incinerator design similarity criteria
3. Verification of these criteria through
the use of non-hazardous wastes
4. Verification with hazardous waste in-
cinerator in existing facilities
16
-------
The Laboratory scale decomposition
studies are a continuation of the Duval and
Ruby work done under sponsorship of the
Municipal Environmental Research Laboratory
under Dick Games. We have joined the MERL,
in the project with the University of Day-
ton Research Institute so that a Gaschro-
matograph Mass Spectrometer system could be
purchased for the analysis of the decompo-
sition products. The IPCD's main interest
in this project is the establishment of
what we term a "center of expertise" which
would be available to us to respond to
emergency conditions as they arise. For
example, should we be asked to assist
either program office or an industry in
their effort to dispose of a hazardous mat-
erial or waste, we could use the facilities
at UDRI to characterize its thermal decom-
position properties and potential products
of decomposition. It could then be de-
termined what the potential dangers are in
its improper incineration and what temp-
eratures are necessary for its proper in-
cineration.
The second phase of our program, De-
velopment of General Guidelines for Incin-
erator Design, will translate the labora-
tory data to a specific incinerator design.
It will basically establish residence time-
temperature distribution profiles for dif-
ferent units of the same design and will
result in some generalized scale up criter-
ia to enable one to attempt at least to
answer the question "if this material is
properly disposed of in this small inciner-
ator, what dimensions, burner configura-
tions, etc. does a full scale unit need in
order to do the same job." The first phase
of this work is currently being done for
the IERL, Cincinnati by TRW in Redondo
Beach, California. The last two portions
of this program are still in the planning
stages. They are of course dependent on
the results of phase 2 but in general they
will consist of major sampling and analysis
programs to verify the applicability of
whatever generalized design criteria are
established.
As an adjunct to our somewhat theore-
tical, incineration program, the IPCD is
initiating a major program in the destruc-
tion of hazardous materials, especially
chlorinated hydrocarbons, in cement kilns.
This work is based on the fine work done
by Skinner of Environment Canada and Mc-
Donald of St. Lawrence Cement in 1976, '77
on the disposal of chlorinated hydrocarbons
in cement kilns. The work that they did
showed that the process is technically
feasible, however there is an acceptance
problem by local citizens which must be
overcome before the system can be imple-
mented.
This problem is very similar to the
one encountered in the siting of landfills
or other refuse disposal facilities and is
compounded by the fear of substances which
are called "hazardous." The IPCD's program
in this area, which is in conjunction with
the Energy Systems Environmental Control
Division, will be a two-year effort. The
first phase which is now in the final
stages of negotiation, will involve the
burning of a non-hazardous chlorinated hy-
drocarbon in the cement kiln. This will
involve sampling at the exhaust for the
hydrocarbon as well as for the by-products
produced in the partial decomposition mat-
erial as determined by the University of
Dayton Research Institute facility. Once
destruction of this inocuous material is
established, the work will shift towards
more hazardous substances-culminating in
the destruction of PCB's in the kiln.
This project is rather unique in that
it is a joint project between a private
cement company, a state agency, and U.S.
EPA. It also has the interest and full
cooperation, to date, of the County Depart-
ment of Health, and representative of sev-
eral environmental groups in the areas.
By keeping everything "above board" and in
the open and by following procedures laid
out by the Office of Solid Waste in the
selection of sanitary landfills, it is
hoped that citizens acceptance of the tech-
nique will take place.
Another area where the IPCD is getting
involved is the treatment of various indus-
trial wastes, especially sludges for the
purposes of reclamation of useful materi-
als. This objective which has been inves-
tigated by a number of people, has often
been a technical success and an economic
failure. As a result our approach to this
is very cautious. There is no question
that there are many very specific waste
streams which can be reclaimed; because of
our process-by-process approach to the in-
dustries we deal with (as described in the
assessment area description] we are ap-
proaching the reclamation concept on that
basis. An area where such work looks pro-
mising is in the reclamation of ethylene
17
-------
glycol and antimony from polyester still-
bottoms. Currently polyester stillbottoms
are incinerated or disposed of at sea.
However, these materials are quire toxic,
containing large amounts of antimony cata-
lyst used in the process, and an alterna-
tive to these disposal techniques is very
desirable. We were recently approached by
a University who has been in contact with
a company that has developed and patented
a rather simple process by which these
stillbottoms could be depolymerized to the
basic monomiers, and these as well as the
antimony oxide be recovered for reuse.
Because of the nature of these still-
bottoms, they are concentrated in relative-
ly small areas of the country, and the fact
that the process involved is very cheap and
easy to use, the recovery of useful mat-
erials from these looks very attractive.
A proposal has been submitted to us and is
now being evaluated for funding.
Should this project prove successful,
(I have seen an actual demonstration of the
process which is very simple) it will turn
a hazardous waste material into a profit
center and a source of useful raw materi-
als.
Another project is currently ongoing
in which industrial PVC scrap is reclaimed
using a solvent process. The process is
currently being developed on a laboratory
scale and the properties of the recovered
PVC and plasticizers are being evaluated.
Future funding of this project is dependent
on results of the lab scale phase.
In summary, the Industrial Pollution
Control Divisions program for the measure-
ment, reduction, and reuse of hazardous
waste materials is just getting under way.
Rather than being geared to the disposal
of hazardous substances in general it is
aimed at the solution of specific indus-
trial problems as they arise and with an
eye towards the processes producing the
problems, rather than towards the solu-
tion of problems as independent entities.
References:
1. Duvall, D.S. § Rubey, W.A. "Labora-
tory Evaluation of High-Temperature
Destruction of Kepone andRelate3
Pesticides" EPA-600/Z-76-Z99, Decem-
ber 1976.
2. McDonald, L.P. § Skinner, D.J.,
Burning Waste Chlorinated Hydrocarbons
in~a Cement Kiln, Technology Develop-
ment Report #EPA 4-WP-77-2 Water Pol-
lution Control Directorate Environment
Canada, March 1977.
18
-------
TEXAS' SOLID WASTE MANAGEMENT
ACTIVITIES—AN OVERVIEW*
Jay Snow
Texas Department of Water Resources
Austin, Texas
ABSTRACT
The Texas Solid Waste Disposal Act assigns the solid waste regulatory jurisdiction to
the Texas Department of Water Resources for industrial wastes and to the Texas Department
of Health for municipal wastes. The term "industrial solid waste" is defined so as to
include waste materials in liquid form, or suspended in liquids, and effectively, all
other waste produced by a given industrial activity, except emissions to the air and waste-
waters discharged or injected as authorized by Texas Department of Water Resources permit.
Pursuant to this statute, the Texas Water Quality Board adopted an Industrial Solid Waste
Management Regulation, Order 75-1125-1, to address multi-faceted problems existing in the
field of industrial waste collection, handling, storage, and disposal. The Act limits the
agency's authority over on-site disposal of solid waste to the adoption of rules, regula-
tions, and guidelines. However, recent amendments to the Texas Solid Waste Disposal Act
and Department rules require permits for on-site disposal of waste determined to be hazard-
ous pursuant to federal rules.
The regulations established uniform requirements for all solid waste disposal sites, a
shipping control system for off-site disposal, and a reporting system. The concept util-
ized provides general limitations stated in direct terms and leaves industry wide discre-
tion in developing methodology to meet the limitations. Development of this regulation
included consideration of numerous alternatives to the approach adopted. Future develop-
ments will be subject to experience gained in operating under the present regulation and
changes in state and federal solid waste statutes.
*Manuscript of the paper not received in time for publication.
19
-------
SOLID WASTE RESEARCH ACTIVITIES IN CANADA
H. Mooij, P. Eng.
Waste Management Branch
Environmental Impact Control Directorate
Environmental Protection Service
Fisheries and Environment Canada
Ottawa, Ontario
K1A 1C8
ABSTRACT
In Canada, industrial and hazardous waste management, including land
disposal, is generally a private sector responsibility controlled by provincial
government legislation and regulation. To date, the Federal government's role
has been to develop and demonstrate appropriate technology and to develop
guidelines or codes of good practice relating to various waste management
aspects. These activities are conducted on a national basis in co-opertion
with the various provincial govenment agencies.
Current research activities relating to land disposal of industrial and
hazardous wastes are underway in the following study areas:
a) Special waste disposal;
b) Contaminant migration and attenuation;
c) Disposal site evaluation;
d) Landfilling procedures;
e) Waste exchange.
Some of the specific projects being conducted in each of these study areas
by the Waste Management Branch, as part of our landfill research program, are
discussed in this paper.
INTRODUCTION
This presentation deals only with
the research activities relating to
industrial and hazardous waste disposal
as conducted by the Waste Management
Branch of our federal government's
Department of Environment. There are
other research activities being carried
out in Canada at the provincial govern-
ment levels, as well as by various re-
search institutes, universities and
consultants. However, their activities
are not discussed here unless they are
under contract to the Branch.
One of the Waste Management
Branch's responsibilities is to conduct
research activities on a national basis.
Development and demonstration of new or
improved technology, and the subsequent
development of guideline documents,
based on our new understanding of the
state-of-the-art, is undertaken to
assist our provincial governments and
the public in general to come to grips
with the many "high profile" issues of
industrial waste management.
The efforts of the Hazardous Waste
Management Division within the Branch
are directed primarily towards:
land disposal research, with
20
-------
some work also being done on
incineration and waste ex-
change as disposal alterna-
tives ;
inventory work to derive an
accurate picture of waste gen-
eration ; and
guideline or policy develop-
ment.
Current land disposal research
activities may be divided into the
following broad study categories:
- special waste disposal;
contaminant migration and
attenuation;
disposal site evaluation;
landfilling procedures.
Some of the specific projects in each
category will be mentioned.
PCB ATTENUATION IN SOILS
Laboratory scale experiments were
performed at the University of Waterloo
to examine the attenuation of Aroclors
1016 and 1254 in soil. Three soils were
used including sand, sandy silt and
clayey silt.
Techniques were developed for: the
extraction of aroclor from aqueous
samples; the concentration of aroclor in
solvents; and, the measurement of
aroclor concentrations using gas chroma-
tography with electi-on capture detec-
tion.
The experimental programme involved
the use of teflon soil-waste reactors,
which exhibited no apparent aroclor
sorption.
Aqueous working solutions of
Aroclor 1016 and 1254 were prepared.
Unfortunately, the Aroclor 1254 was
poorly soluble in water to the extent
that its aroclor content could not be
distinguished from background interfer-
ence. This served to negate the results
obtained from the subsequent Aroclor
1254 - soil contact experiments.
Aroclor-soil contact experiments
were performed using two working
solutions and three soils. The results
showed that the Aroclor 1016 was removed
from the aqueous phase in excess of 90%
by weight in most cases. The stability
of the Aroclor 1016 removal was tested
in a series of desorption experiments.
The results generally indicated that the
sorption of the Aroclor 1016 was only
slightly reversible under the conditions
used.
The results of our study, and the
indications from other studies that
Aroclors are generally immobile within
the soil system, might indicate that
controlled disposal of PCBs in a munici-
pal landfill, may be a feasible disposal
alternative for PCB containing wastes.
Such a landfill would provide retention
times, allowing for the possible biode-
gradation of the aroclors, and it would
also provide a variety of opportunities
for sorption to occur within the fill
itself. Any PCBs which might eventually
be discharged from the landfill would be
quickly sorbed and thereby retained onto
the underlying soils.
SEPTIC TANK SLUDGE LANDFILLING
Earlier work carried out at the
University of British Columbia on the
effects of adding septic tank pumpings
to municipal refuse indicated that
additions could be beneficial in reduc-
ing both the peak concentrations and the
total mass discharged of trace and heavy
metals in the leachate. This work indi-
cated, however, that relatively large
amounts of septic tank pumpings had to
be added to achieve the desired effect.
The main purpose of the follow-up
research was to determine the optimum
amount of septic tank additions to land-
fill, under varying conditions of net
infiltration, refuse depth, and sludge
additions. The septic tank sludge addi-
tions ranged from a relatively small
amount to an amount approaching the
estimated field capacity of the refuse.
The results have shown that signi-
ficant reductions, both of peak concen-
trations and total mass, of contaminants
discharged can be achieved by adding
sludge to refuse under low precipitation
infiltration conditions. This is par-
ticularly true of most metals. While
the optimum amount of sludge to be added
was not consistent in all cases, the
trend is toward batter reductions at
high loadings, with lower loadings some-
21
-------
times creating a detrimental effect. It
was generally found that a minimum ratio
of sludge addition should be 16,000 mg
septic tank sludge solids per kg. of dry
refuse.
Specific reasons for the preceding
behaviour are difficult to provide for
each of the contaminants. The final
report, soon to be issued, discusses
many possible explanations.
It was also found that no positive
fecal coliform counts were ever obtained
from the leachate, and that increasing
the rate of infiltration tends to count-
eract some of the benefits obtained. In
virtually all cases, self attenuation
within the fill increased with Increas-
ing fill depth.
BORDEN LANDFILL STUDY
The detailed Camp Borden landfill
leachate migration study, discussed at
last year's research symposium, is now
being extended to more specifically
focus on the behavior of certain contam-
inants . In addition, a very extensive
modelling effort is underway using the
excellent field data complemented by
more exact numerical field values of the
various model coefficients.
Although the Borden data was made
available to any modelling group for
their own use, no requests for specific
information have been received. The
University of Waterloo work should,
therefore, be a significant singular
advancement of the state-of-the-art.
MISSISSAUGA LANDFILL STUDY
A 2 year study has been completed
at this 78 acre site near Toronto which
has served as a mixed municipal and
industrial waste disposal site. The
site was well instrumented upgradient,
downgradient, and through the landfill,
and routinely monitored.
Detailed chemical analyses were
conducted including determinations of
pesticide and PCB concentrations. At
this site, the contaminant plume, as
defined by chloride concentrations, has
migrated about 3,000 feet through surfi-
cial sands and gravels.
Two mathematical models were util-
ized to simulate both the flow and mass
transport at the site. The resulting
downgradient concentration profiles were
then used to predict a contaminant load-
ing to a nearby receiving stream up to
the year 2000.
A final report on this work should
be available in the next few months, in
the meantime, the modelling techniques
are already being applied in landfill
design work in Ontario and the U.S.
HAZARDOUS WASTE DISPOSAL SITE STUDY
As part of an evaluation of special
waste management in the National Capital
Area, a chemical landfill site was moni-
tored to evaluate its environmental
impact.
The site was found to be next to an
old landfill, and it is known that toxic
wastes have been discharged at the site
for at least 18 years. The site is used
to dispose of about 10 tons per year of
miscellaneous toxic and hazardous inor-
ganic and organic chemicals.
Although our investigations are
still incomplete at this time, the very
preliminary indications show no evidence
of any nearby drinking water contamina-
tion either from the old landfill or the
special waste disposal operation. The
soils underlying the area are character-
ized as sands and gravels.
PCB SPILL SITE
A study is underway at a spill site
in Regina to investigte the mobility of
the transformer fluid containing PCBs.
Soil samples have been extracted from
the site in a first attempt to determine
the extent of the contaminated soil.
Samples will also be used to con-
duct extraction and resolubilization
tests. Some useful data should be in
hand within four months, and a further
in-depth investigation of PCB migration
at the site will likely result.
PROCEDURES DOCUMENTS
Starting in 1975, the Branch ini-
tiated a series of international round-
table discussions on various technical
22
-------
aspects of landfilling. Participants
were selected from among the top
experts, including researchers, govern-
ment personnel, and consultants in
Canada and the U.S.
Our first session dealt with land-
fill leachate analysis, and was attended
by a handfull of Canadian and U.S.
scientists, including EPA sponsored
contractors. Since then we have had
sessions on ground water and soil
sampling procedures, landfill monitoring
programme design and implementation
procedures, landfill gas control proce-
dures, landfill leachate control proce-
dures, and our next session on landfill
leachate treatment will be held this
April. Each session has been documented
and results in the publication of speci-
fic procedures documents containing
recommendations of a practical nature
based on the consensus of opinion.
To date the documents have been
well received in many countries, and the
popularity of the sessions has increased
greatly.
DISPOSAL GUIDELINES
Specific guideline documents deal-
ing with particular waste disposal
problems are being prepared. We have
developed a draft copy of a guideline
document dealing with wood waste
disposal, as a first effort.
It is known that wood waste land-
fill leachate can be very toxic when
fresh, and some data indicates a contin-
ued toxicity of leachate even from aged
bark, sawdust, and shavings. Special
precautions are therefore necessary in
the disposal of such wastes in bulk
quantities. The guideline document will
address land disposal accordingly, in
light of current waste-soil interaction
information.
Another separate document is being
prepared for arsenic bearing wastes.
This document will provide a waste
generator or discharger information on
the nature of the wastes, its general
behavior in the environment, and recom-
mendations for the land disposal of the
waste, based on the soils capabilities
to attenuate the contaminants at the
site.
This document will be followed by
others. It is anticipated that the land
disposal of PCBs and mercury will be
addressed next.
OTHER RESEARCH ACTIVITIES
Other research activities relating
to industrial and hazardous waste
disposal include:
1) our continued work on develop-
ing a soil-waste matrix proce-
dure for assessing the environ-
mental suitability of a pro-
posed industrial waste disposal
site;
2) a study on the treatment of
landfill leachate discharging
from pulp mill disposal sites,
by leachate recirculation;
3) a study on the treatment of
leachate from a co-disposal
site by aerobic biological
treatment, and
4) a study attempting to develop a
mathematical model to predict
leachate characteristics under
varying landfill operating con-
ditions, using over 65,000
pieces of data routinely col-
lected by several investigators
over a period of almost 5
years.
Aside from land disposal research
activities, our other efforts have been
directed towards testing the burning of
pesticides and PCBs, conducting inven-
tory studies, and developing PCB manage-
ment guidelines.
WASTE EXCHANGE
Last, but not least, is our waste
exchange operation. We have recently
given birth to a nation-wide exchange,
operated by the Ontario Research
Foundation in accordance with the
recommendations for a methodology
contained in a report prepared under
contract to us.
The operation is being sponsored
for a period of two years by ourselves,
after which it should, if successful,
become self-supporting.
The exchange has solicited listings
23
-------
and will be generating bulletins once
every two months. Listings in our
January bulletin are from all parts of
Canada. In addition to listing wastes
available and wastes wanted, the
bulletin also lists services available
for transportation and reprocessing.
During the two year sponsorship
period, the contractor will evaluate
several aspects of a waste exchange
operation/ including the possibility of
making a passive exchange more active,
regionalizing the operation, and
instituting exchanges across the border.
Should you want more information on
the waste exchange operation, or on any
of the other projects briefly mentioned,
please contact the author.
24
-------
KEPONE: AN OVERVIEW
Richard A. Carries,1 Frank C. Whitmore,? Robert L. Stenburgl
TU.S. Environmental Protection Agency, 26 W. St. Clair Street,
Cincinnati, Ohio 45268
2versar, Inc., Springfield, Virginia 22151
ABSTRACT
The serious environmental contamination of the manufacturing facility, sewage treat-
ment works, and nearby environs of Hopewell, Virginia, by the pesticide Kepone required a
sound technical approach to disposal options and a public awareness program stretching
from Hopewell to Toledo, Ohio. Background data were developed for the thermal stability
of the Kepone molecule; this was of direct assistance in selecting the incineration tech-
nology for experimental purposes and also lent supportive dialogue to the public relations
program.
A complex management structure was employed for the Kepone incineration Test. Respon-
sibilities for public information, technical evaluation of data, and development of a test-
ing protocol were divided among a variety of individuals. Health and safety was always
held in the fore of all research experiments and resulted in a completely safe experiment.
A series of incineration experiments was conducted where increasing concentrations of
Kepone were mixed with sewage sludge and exposed to partial pyrolytic conditions in a
rotary kiln followed by h'igh temperature fume incineration and finally caustic scrubbing
with the calculated combustion efficiency always in excess of 99.99 percent.
INTRODUCTION
In the aftermath of the severe en-
vironmental contamination and the employee
medical problems associated with the prod-
uction of the pesticide Kepone in Hopewell,
Virginia, the cry was heard loud and
clear for a concerted effort to safely
dispose of the Kepone and Kepone contami-
nated materials. Several task forces were
established to review and assess the prob-
lem. The Administrator of the U.S. En-
vironmental Protection Agency established
a Kepone Task Force at the Federal level
and the Governor of Virginia established
a Task Force at the state level. The
overall charge of both was to assess the
problem, determine the best available
short-term disposal options available and
to determine the long-term environmental
problems associated with Kepone contami-
nation in the aquatic food chain. The
later charge will take longer to determine
and will have many economic and social
overtones to it before Kepone is com-
pletely removed from the environment.
With the aforementioned goals in mind,
the U.S. EPA, Office of Research and
Development, and Solid and Hazardous Waste
Research Division undertook a laboratory
research study to develop thermal degrada-
tion data for Kepone and related pesti-
cides.(l) Meanwhile, the Virginia Task
Force, through their contractor Design
Partnership, was evaluating disposal
alternatives and planning for the ulti-
mate disposal of all Kepone and Kepone
contaminated materials.(2) After evalu-
ating existing disposal options, labora-
tory data supported the hypothesis that
under controlled combustion conditions
Kepone could be safely destroyed. Design
Partnership recommended that thermal
destruction was the safest, most economi-
cal, and most convenient method of disposal
of these materials.
25
-------
Historically, extrapolating labora-
tory results to full scale operations has
not been generally successful. There-
fore, the recommendation was made to
conduct a pilot scale test burn using
the information gained in the laboratory.
In addition to confirming the lab results,
the pilot scale study could also serve to
define the range of operating parameters
needed for safe disposal of Kepone.
Because of the physical form of the ma-
terial requiring disposal, it was agreed
that the system for these pilot scale
tests necessitated a rotary kiln which,
in conjunction with a temperature above
11000C, could volatize the material and
yield a residence time greater than 2
seconds. In addition, the installation
should have the necessary pollution
control devices to remove the natural
products of organochlorine incineration.
The research facilities of the
Surface Combustion Division of the
Midland-Ross Corporation in Toledo, Ohio
had the required hardware to conduct the
pilot scale testing, and were subsequently
selected for this study.
THE KEPONE INCINERATION TEST (KIT)(3)
While making the necessary arrange-
ments for the KIT it became apparent that
this particular program required a public
information program, a well planned and
supervised health and safety program, and
a technical group to plan, coordinate and
evaluate the KIT results. After much dis-
cussion, the overall program management
structure presented in Figure 1 evolved,
which offered the strongest possible con-
trol over the program by both the responsi-
ble governmental bodies and by the techni-
cal staff assigned to the project. The
plan was for the detailed experiments to be
carried out under the direction of the
Experimental Management Group, imple-
menting stringent safety procedures es-
tablished and monitored by the Health and
Safety Group. A committee of representa-
tives from the various Federal, state,
and local agencies whose responsibilities
included air and water quality control
was organized and designated the Burn
Authority (BA). The results of each ex-
periment were presented to the Burn
Authority who had at least one repre-
sentative on site at all times. The BA
BURN
AUTHORITY
PUBLIC
INFORMATION
HEALTH
AND
SAFETY
EXPERIMENTAL
MANAGEMENT
EXPERIMENTS
FIGURE 1. PROGRAM MANAGEMENT STRUCTURE
26
-------
reviewed results of each test, and was
empowered to decide whether the program
should proceed as scheduled, or be
modified. The results of each experiment
and the BA decision were then transmitted
to the public through the Public
Information Group.
In spite of the obvious complexity
of the management structure just dis-
cussed, it worked quite successfully.
The dedication of the BA members to their
responsibilities to the public and their
sympathy for the stated goals and objec-
tives of the program, were in no small
way responsible for the successful
outcome.
EXPERIMENTAL
A schematic of the physical layout
for the experiments conducted at the Sur-
face Combustion Division is shown in
Figure 2. Main components and their
characteristics are discussed below:
1. Rotary kiln pyrolyzer. The
rotary kiln pyrolyzer, 1.52 m in
diameter and 3.0 m in length, was
fitted with rotary seal charge
and discharge connections so as
to minimize the leakage of gases
into or out of the kiln. The
kiln was heated with one 500,000
Btu/hr gas burner calculated to
maintain a nominal temperature
of 500°C. The kiln, designed for
batch feed, was modified for the
KIT such that sludge mixed with
Kepone could be Moyno pumped into
the kiln at a nominal rate of
45 kg/hr (100 Ib/hr).
2. Fume Incinerator/Afterburner.
The fume incinerator, with a
residence chamber volume of
2.4 m3, was fired by two
KEPONE INJECTION
KEPONE
SOLUTION
SAMPLE
PORT
STACK
BURNER
AIR
AIR
FUEL
WATER
DRAIN
FIGURE 2. KEPONE INCINERATION TEST SYSTEM
27
-------
500,000 Btu/hr gas burners. The
system was designed to operate
at temperatures above 11000C
and still maintain a residence
time for hot gases in excess of
2 seconds.
3. Quench. Cold water was injected
into the bottom of the inciner-
ator as a quench. This promoted
evaporative cooling of the hot
gases in the incinerator.
4. Scrubber. The scrubber, a tower
76cm in diameter, was packed to
a depth of 1.83 m with 5 mm
Intalox plastic saddle packing.
A liquid distributor at the top
of the packed bed caused the
liquid to be evenly distributed
across the packed bed. The mist
eliminator consisted of a 15 cm
deep bed of saddles placed above
the liquid distributor. The gas
flow through the entire system
was by an induced draft fan with
a 0.94m3/sec capacity mounted at
the top of the scrubber. The pH
of the scrubber water was moni-
tored and continually adjusted
with 12% caustic solution to lie
between nine and ten.
5. Sludge Feed System. Kepone con-
taminated sludge was simulated
by mechanically mixing in the
feed tank the appropriate
amounts of Kepone dissolved
in acetic acid and Toledo sludge.
6. Direct Injection Feed System.
The direct injection experiments
were designed to study the feasi-
bility of directly injecting
Kepone containing low Btu fluids
into the incinerator without
passing through the kiln. These
experiments, as well as those
conducted with sludge, were
planned so that in succeeding
experiments the amount of Kepone
could be increased or the specific
set of operational variables
could be altered. The direct
injection experiments were termi-
nated when the results indicated
that another sequence would be
likely to cause the 25 ng/m3
ambient air standard imposed on
this test to be violated.
The health and safety precautions
for this research were well planned so
that all operating personnel would have
sufficient time and space to be isolated
from Kepone contamination. The general
outline of these procedures can be seen
in Figure 3, where the operations and
control area are shown in conjunction
with the kiln and mixing areas.
The operations and control area,
which included the incinerator, the
scrubber, the furnace operational equip-
ment, and the brine and caustic storage
tanks, was enclosed by a wooden frame-
work lined on the inside with heavy
plastic. This area was accessible only
to specifically authorized personnel
through locked doors and was considered
to be minimally contaminated. Operating
personnel in the area were required to
wear disposable coveralls, rubber boots,
and rubber gloves. Upon leaving, it was
required that hands and face be washed
and protective clothing be stored in
appropriate containers.
In the kiln and mixing areas much
more detailed precautions were taken.
These areas were most likely to be exposed
to Kepone and its solutions, thus it was
designated as hazardous. Personnel
working in these areas were required to
wear full protective gear, including a
respirator. Further, when the operator
was handling the acetic acid he was re-
quired to wear a face mask. The only
entrance to this area was through the
change and shower area; the doors shown
to the outside from the mixing room were
emergency doors (Fig. 3).
The change and shower area was pro-
vided to assure proper isolation of the
facility. Contaminated clothing was kept
within the change area and personnel exit-
ing from the kiln and mixing areas had to
shower before dressing in street clothes.
Also, all personnel in the operating area
were required to shower at the end of day.
EXPERIMENTAL RESULTS
The results of the experiments with
Kepone incineration are presented in Table
1. As can be seen, the ambient air
standard of 25 ng/m3 for Kepone, set on an
interim basis, was not violated in any of
the experiments. In fact, most experi-
ments have a combustion efficiency of at
least 99.9999%. Such detailed calcula-
28
-------
EMERGENCY
OBSERVATION
. > _y//w<
L/'CONTAMlNAtEDV''
V////}i\ A LL/'/////>*//
SHOWER CLEAN
. .CONTAMINATED
'
EXIT
SINK p
rP
ENTRANCE ONLY
AMBIENT
OPERATIONS
AND
CONTROL
MINIMAL
CONTAMINATION
^OBSERVATION WINDOWS
FIBERGLASS FILTERS
FOR AIR MOVEMENT
. ENTRANCE ONLY
[ [
CLEAN
CONTAMINATED
MINIMAL CONTAMINAT.O
OFFICE
y
LABORATORY
y
FIGURE 3. EXPERIMENTAL AREA-ISOLATION PLAN
tions were permitted by the extremely
precise analytical support present during
the KIT.
The continuous interaction of the
Burn Authority with the Experimental
Management Group was of great benefit
to the overall program. This form of
program management proved to be a viable
and useful tool that should be considered
in future efforts concerning hazardous
materials that have received adverse
publicity.
The stack was sampled using the EPA
Standard Method Five with an RAC Stack-
sampler® • After each sludge and stack
sample run, samples were analyzed using
accepted procedures. A number of Kepone
standards were run as reference checks
on retention times.
It was calculated that during the
course of the KIT some 1000 individual
analyses were performed at the site by
one chemist and one technical assistant.
CONCLUSIONS
A summary of the results and con-
clusions derived from the KIT program
are presented below:
1. Coincineration of Kepone with
sewage sludge was a safe and
environmentally acceptable
method for disposing of Kepone.
The results of these experiments
indicated that the destruction
efficiency for Kepone is greater
than 99.9999% in a system con-
sisting of a pyrolyzer and a
fume incinerator which operates
at temperatures above lOOOQC and
has a residence time of at least
2 seconds.
2. Direct injection of low Btu solu-
tions of Kepone was not a useful
process. This does not rule out
the possibility of using a high
Btu solution of Kepone and
incinerating it.
3. Adequate safety instruction
coupled with constant monitoring
29
-------
TABLE 1: SUMMARY KEPONE INCINERATION EXPERIMENTS
Experiments Kepone Feed Total Feed Total Kepone
Rate Time Fed
(gms/min) (min) (gms)
Stack Emission
Concentration^!)
(gm/m3)
Total Stack
Output
(gms)
Afterburner
Temp.
Combustion
Efficiency
Injection runs
1
2
3
4
5
Sludge runs
CO
o
7
8
9
10
Totals
acetic acid
only
1-67x10-5 130
1.67x10-5 135
1.67x10-5 100
1.50 115
Toledo sludge 120
only
5.68 315
5.68 240
5.68 220
24.2 165
2.16x10-3
2.25x10-3
1.67
172.5
1789
1354
1241
.3995
8553
NO
5.35x10-8
ND
1
7x10-8
2.4x10-8
2.55x10-8
2.95x10-8
ND
5m
1.76x10-4
ND
1.85x10-4
1.85x10-4
1.53x10-4
1.67x10-4
1.58x10-3
1260
1260
1093
1093
1093
1093
1149
1093
1093
1093
100(3)
100
100
99.99991
99.9999
99.9999
99.9999
99.9999
(1) Measured at stuck conditions, not ambient conditions.
(2) Apparent Kepone peak on chromatogram of filter collected.
of original extract.
The Kepone peak did not appear after base partitioning
(3) To speak of 100 percent combustion efficiency is somewhat unrealistic but is a natural consequence of the defini-
tion for efficiency of combustion in those situations for which there is so small a stack load of Kepone that it
was undetectable. Further, the expression of the combustion efficiency in terms of 6 or 7 decimal places is
justified since even if both the input rate and the emission rate were in error by as much as 25 percent, the
extremes would only effect the combustion efficiency in the sixth decimal place -- again a consequence of the
definition of the combustion efficiency and the very small emission rates that were found.
-------
to assure the proper use of
safety equipment was necessary
to prevent exposure of operating
personnel.
4. The layout and the isolated
location of the facility pre-
vented the escape of Kepone
into the surroundings, and at
the same time made it easy to
clean up after the tests were
completed.
5. Real time analyses of Kepone on
site was highly satisfactory and
of great assistance in the de-
cision making process.
6. The Burn Authority control
led the KIT program very
satisfactorily.
7. A completely informed public
was an ally rather than an
antagonist when experimenting
with material that had pre-
viously received a large amount
of adverse publicity.
Results of the KIT indicated that
land-based incineration technology is
available which will safely and ade-
quately dispose of the presently stored
Kepone and that of the sludge lagoon in
Hopewell. The geometry of a full scale
system would naturally not be the same
as existed at the Surface Combustion
Division. The KIT program was designed
to determine the feasibility of incin-
eration for the disposal of Kepone and
to evaluate data generated from labora-
tory research. To this end, the KIT
was a successful venture.
RECOGNITION OF CONTRIBUTIONS
The authors of this paper are deeply
indebted to the following:
1. Mr. Eric H. Bartsch, Director,
Bureau of Sanitary Engineering,
Department of Health, Virginia,
for outstanding leadership in
coordinating the various activi-
ties and for serving on the Burn
Authority.
2. Dr. Albert J. Klee, Chief,
Processing Branch, Solid and
Hazardous Waste Research Division,
MERL, ORD, EPA, for Support
of the KIT program.
3. Mr. John D. Steele, Operations
Manager, Design Partnership
Consultants, for his coordina-
tion and communications help.
4. Mr. Dale J. Krygielski, Toledo
Pollution Control Agency, for
his help and participation on
the Burn Authority.
5. Mr. Julius Foris, Chief, Divi-
sion of Technical Support
Operations, Office of Air
Pollution Control, Ohio EPA
for his participation on the
Burn Authority.
6. Mr. Karl J. Klepitsch, Jr.,
Chief, Solid Waste Branch,
USEPA, Region V, for his con-
tributions to and participation
on the Burn Authority.
7. Mr. James A. Saunders, Industrial
Hygienist, Virginia DOH, for his
contribution to the health and
safety program that prevented
operating personnel and the en-
vironment from contamination by
Kepone.
8. Dr. Mohamad N. Kkattah and David
Sood of Versar Inc., Springfield
Va., for outstanding analytical
contributions despite extreme
conditions of the laboratory
setting and the northern winter.
9. Others who assisted in a variety
of ways that made the KIT the
success it was. Without every-
one's cooperation, the program
could easily have gone awry and
failed to obtain the necessary
information to allow safe
disposal of Kepone on a full
scale.
REFERENCES
Duvall, D.S., and Rubey, W.A,
"Laboratory Evaluation of High-
Temperature Destruction of Kepone
and Related Pesticides," EPA-600/
2-76-299 December 1976.
"Engineering Feasibility Report:
Destruction of Kepone Contaminated
Waste in the Lagoon located at the
31
-------
Hopewell Sewage Treatment Plant,"
prepared for the Virginia Kepone
Task Force by Design Partnership
May 20, 1976.
3. Bell, B.A. and Whitmore, F.C.,
"Kepone Incineration Test Program,"
draft final report for USEPA Grant
R805112. In press.
32
-------
The Development of a Leaching Test for Industrial Wastes
BY: R.K. Ham
M.A. Anderson
R. Stanforth
R. Stegmann
Department of Civil £ Environmental Engineering
The University of Wisconsin
Madison, Wisconsin 53706
ABSTRACT
Recent legislation dealing with land disposal of industrial or hazard-
ous wastes has focused attention on potential water pollution resulting
from such practices. This paper provides a summary of procedures as well as
some of the background information and rationale which led to development of
a laboratory test procedure to assess the leaching potential characteristics
of a particular waste. The procedure results in information regarding the
types of materials leached from a waste, an estimate of the maximum concen-
tration likely to be observed of these materials, an estimate of the amount
of each material likely to be released per unit weight of waste, and an
indication of the effect of co-disposal with mixed municipal refuse or
other specific wastes on the leaching potential of the waste being tested.
Introduction
Recently an increased awareness
of the potential for ground water
pollution from industrial wastes dis-
posed in landfills has become evi-
dent. Since many wastes will not
produce polluting leachates when
landfilled, there is need for crite-
ria to discriminate between wastes
that will produce hazardous leach-
ates and those that will not. One
such criterion is a short standard-
ized leaching test which subjects
the waste in the laboratory to
simulated landfill conditions or to
test conditions that can be related
to landfill conditions. The test
would evaluate the leaching poten-
tial of the waste under landfill con-
ditions by indicating what constitu-
ents would leach out of the waste,
how much of that material would
leach out and under what conditions
they will leach. This paper sum-
marizes a background study performed
to develop such a leaching test. It
will discuss the philosophy of leach-
ing tests and the test variables af-
fecting the leaching test, review
some of the leaching tests that have
been developed, and discuss briefly
the procedure which evolved from
this work and the interpretation of
leaching test results.
Intensive vs. Quick Tests
Two general approaches can be
used to evaluate the leachability of
waste material: (1) a very inten-
sive study of waste leaching charac-
teristics, or (2) a quick test using
standardized procedures. The inten-
sive study gives more information
about the leaching characteristics
of a waste. Test conditions can be
varied as needed, and the effects of
different variables on the leaching
33
-------
characteristics can be studied.
Such a test takes considerable time,
money and personnel. The standard-
ized test uses only predetermined
testing conditions, and so cannot
show the effects of the different
variables of the waste leaching pat-
tern. It can, however, give useful
information in a short time which
when properly interpreted can give
an indication of the leaching char-
acteristics of a waste. It is much
cheaper, faster, and simpler than the
intensive study.
Wastes generated in large quan-
tities should be subjected to the in-
tensive study, particularly wastes
that are generated at different sites
throughout the country but are of
relatively uniform composition, e.g.
fly ash or scrubber sludge. However,
for the many wastes that are produced
in relatively small amounts, a stand-
ard leaching test is more appropriate.
The snail amount of waste produced
does not justify the expense involved
in the intensive study unless the
waste is of particular concern be-
cause of its characteristics or land-
filling situation.
In an intensive test, factors
affecting the test results will be
varied and their effects observed
and analyzed. In the standard test
these factors will be set beforehand
and their effects on the test
results will not be observed. It is
important to be aware of the test
factors which are set in the stand-
ard test and their potential effect
on the test results so that a proper
interpretation can be made of the
results.
Ideal and Practical Leaching Tests
Ideally a leaching test would
determine four characteristics re-
garding the release of a parameter,
A, from a waste:
1. the highest concentration
of A to be found in the
leachate;
2. the factors controlling this
concentration,
3. the total amount of A
available from a given
amount of waste, and
4. the release pattern of A
with time.
The last characteristic includes the
kinetics of the release, physical or
chemical changes occurring in the
waste as it is leached, any effects
of these changes on the release of A,
and the influence of the waste on
the leachate. Water quality stand-
ards are given in terms of concen-
tration and since many toxic effects
are concentration dependent as are
most chemical reactions, concentra-
tion is of obvious interest. Maxi-
mum release is important when pre-
dicting the total amounts of A that
may be leached from the waste in a
landfill. It is also of importance
when A may be accumulated, whether
due to biological uptake or chemical
processes (e.g. sorption onto soil,
precipitation, etc.). Accumulated
materials may be released at high
concentrations if conditions change.
With the four characteristics deter-
mined by the leaching test, the
potential hazard of a waste can be
evaluated, and the suitability of
landfilling as a disposal technique
assessed. The information from the
test could also be used along with
other information, to design waste
processing or landfilling procedures
so as to minimize the release of A
from the waste.
A standard leaching test, as
defined herein, will not give enough
information to predict completely
the four characteristics mentioned
above; however, it will show the
behavior of a waste under a pre-
scribed set of conditions. These
conditions can then be related to
landfill conditions through modeling
and careful selection of test condi-
tions , and through correlation
studies between test and actual land-
fill results. Correlation studies
serve both to relate the non-modeled
conditions to landfills and to veri-
fy the modeled conditions. With
careful interpretation of results, an
estimation of the behavior of a
waste in a landfill can be made from
standard test results.
34
-------
Since the purpose of the test
is to evaluate the leaching poten-
tial of a waste, it is reasonable to
use an aggressive test to model a
worst case situation. This is a con-
servative rationale. Further test-
ing can then be done, if warranted,
using conditions more typical of the
long-term landfill situation.
A test which extracts only
those components that would be
leached in a landfill and extracts
them in the same pattern that they
would be extracted in a landfill is
needed. A chemical solvent or
series of solutions which would
extract only leachable components
would be ideal. Serial extraction
procedures have been developed for
soils and sediments; however, such
procedures are based on a compre-
hensive understanding of the
extraction process. This under-
standing requires that the chemical
composition of the leached material
and the chemical interactions be-
tween the material and the leach-
ing solution be understood, and
that the composition of the leached
material be fairly consistent. Most
of the wide array of landfilled
wastes do not meet these criteria.
The leaching test must then attempt
to model landfill conditions, so
the results are more readily related
to full-scale landfill situations.
Batch and Column Tests
Two types of tests are commonly
employed for determining the leach-
ing potential of a landfilled
waste—batch and column tests. In
batch tests, a properly prepared
sample of the waste to be tested
is placed in a container along
with leaching media. After a suit-
able period of time, and under con-
ditions specified as being appro-
priate to the test, the elutriate
or leachate is separated from the
waste and analyzed to determine the
material leached from the waste.
Column tests, in which the waste is
packed in a column and the leaching
solution passed through, is a closer
approximation of landfill conditions
than a batch test, at least at first
glance. The column test simulates
both the waste—leachate contact
(except around the column edge) and
the rate of leachate migration
found in landfills. The column test
also is good for predicting the
release pattern of A with time,
since it models the continuous
leaching and long time periods found
in landfills. However, column tests
have several disadvantages, such as
the following:
1. problems arising from chan-
neling and nonuniform pack-
ing,
2. potential unnatural clog-
ging,
3. possibly unnatural biologi-
cal effects,
4. edge effects,
5. long time requirements y and
6. difficulty in obtaining
reproducible results even if
done by experienced lab per-
sonnel.
All of these difficulties, but par-
ticularly the time requirements for
an adequate column test (months to
years), suggest that a batch test be
chosen as the standard testing pro-
cedure. Both batch and column tests
might be used in an intensive study.
Factors in a Batch Test
There are several factors
affecting a parameter's concentra-
tion in the batch test elutriate
which need to be considered in de-
signing a leaching test. These fac-
tors are given in Table 1. O'f par-
ticular interest in the development
of a leaching test are the test
conditions. These are discussed
individually below.
A. Leachate Composition
Industrial wastes can be land-
filled in one of three general cate-
gories of landfills as shown in Table
2. It is clear that the leaching
media composition prior to waste
contact is one of the key variables
in a leaching test, and must be com-
parable to leaching media composition
35
-------
contacting the waste in the actual
landfill for the test to have any
practical significance.
For a monolandfill, distilled,
deionized water or a synthetic rain-
water can be used as a representative
extractant or solvent. Alternatively,
this landfill situation can be
Table 1
Factors Affecting Parameter
Concentrations in a Batch Test
1. Sample treatment prior to test
A. Sampling in industrial plant
and shipment to test labora-
tory
B. Sample treatment in labora-
tory prior to test
1. subsampling procedure
2. sample preparation
(solid-liquid separation,
drying, grinding, etc.)
Test conditions
A. Leaching media composition
B.
C.
D.
E.
F.
G.
Solid to liquid ratio
Time per elution
Number of elutions
Temperature
Agitation technique
Surface area contact
Separation procedure and
analysis after test
modeled by contacting the waste with
a leachate produced by the waste
itself. A sample can be contacted
with leachate resulting from prior
contact between another waste sample
and distilled water, for example.
This simulates what would occur if a
drop of relatively uncontaminated
rainwater passed through a thick
layer of waste in a monolandfill.
For co-disposal with mixed muni-
cipal refuse, both distilled deion-
ized water and a municipal landfill
leachate can be used. The water
simulates rainwater and provides mild
leaching conditions. A municipal
landfill leachate has widely dif-
fering characteristics depending
on the refuse composition and state
of decomposition, dimensions of the
landfill, age, degree of chanelling
of moisture, and both long-term and
instantaneous climatic effects, etc.
Further, even for a given sample of
such leachate, the composition is
very complex, precluding developing
an exact recipe from which leachate
of both reproducible and realistic
composition can be produced. Rather
than attempt to define a standard
landfill, from which leachates repre-
sentative of different landfill ages
would conceivably be obtained, it is
more promising to examine the leach-
ing characteristics of leachate
typical of actively decomposing
municipal landfills, and model a
synthetic leachate on the results.
Such a synthetic municipal landfill
Table 2
Classification of Landfills and Leachate Composition
Landfill Type
Monolandfill
Municipal
Industrial
Waste Landfilled
By itself
With municipal wastes
With other industrial
wastes
Leachate Composition
Controlled By
The waste itself
Municipal refuse
decomposition products
Other industrial wastes
36
-------
leachate has been developed as part
of this study, which simulates
aggressive leaching conditions which
might be obtained by co-disposal
of the waste being tested with
municipal refuse.
The leachate produced in a
heterogenous industrial landfill is
the most difficult to model, since
the material entering the landfill
will likely vary in composition. In
this case, two approaches may be
used in the selection of leaching
media. If the wastes in the land-
fill are known, a synthetic leachate
can be developed based on character-
istics of those wastes. Alterna-
tively, a series of leaching solu-
tions can be used, each emphasizing
a single leaching parameter—i.e.,
acid base, complexer, organic solvent
etc. Results obtained using dif-
ferent leaching solutions would
indicate what types of wastes might
be co-disposed with the waste in
question. For example, a waste
which released large amounts of an
undesirable parameter under acidic
leaching conditions should not be
landfilled with acid or acid pro-
ducing wastes.
The use of distilled water or
other mild leaching solutions
allows the waste to create its own
leaching environment, whereas, a
synthetic leachate or strong chemi-
cal solution essentially controls
the leaching environment. For
example, a waste containing small
amounts of a leachable basic salt
will raise the pH of a distilled
water leachate, and only materials
that are soluble in basic solutions
will be found in the leachate. Con-
versely, use of a synthetic munici-
pal leachate which is heavily buf-
fered, or an acid leaching solution,
will probably neutralize the basic
salt while maintaining an acidic pH.
In the first case the waste con-
trols the pH of the solution, while
in the second case the leaching
media is the controlling factor.
B. Solid to Liquid Ratio
Solid to liquid ratios (or waste
to eluent ratios) used in the test
can have profound effects on test
results. The concentration of a
very soluble parameter will be
directly dependent on the solid to
liquid ratio (S/L). On the other
hand, parameters for which concen-
trations are controlled by solubil-
ity will not show S/L ratio effects,
but rather will have the same con-
centration at all S/L ratios, pro-
vided enough solid is present to
saturate the system. S/L ratios can
also affect concentration if adsorp-
tion or desorption processes are
controlling the concentration.
In a given waste, several chemi-
cal constitutents may be of interest.
These may have different factors
controlling their concentrations,
and so may show different dependen-
cies on the S/L ratio. Several cur-
rently available leaching or elutri-
ation tests start with high_S/L
ratios and saturated conditions,
then decrease the ratio until unsat-
urated conditions are reached. This
procedure could be complicated if
more than one parameter of interest
reached saturation at different S/L
ratios.
The S/L ratio encountered by a
drop of leachate percolating through
a landfill will be very_high, by the
very nature of percolation. If
there were extensive channeling in
the landfill, however, this would
not be true.
The choice of a solid/liquid
ratio for use in the test is based
on practical considerations. A very
high S/L ratio, such as is used in
the saturation test, is most likely
to result in many components being
saturated. This makes it difficult
to estimate the total release of a
component from the waste, since many
elutions will be necessary to elute
the leachable fraction of the com-
ponent. Also, it is often difficult
to obtain enough leachate for analy-
sis with a high S/L ratio. On the
other hand, a very low S/L ratio can
produce very low concentrations of
the parameters of interest, leading
to analytical problems.
An interesting approach to the
37
-------
selection of an S/L ratio is used in
the State of Illinois E.P.A. test
as reported in reference^'. The
ratio to be used for a waste R is
calculated from the equation R =
5.34 D, using the waste density (D)
for the wet waste and a constant
based on 1he average annual rainfall
in Illinois. Thus, the S/L ratio
is based on volume rather than
weight, and can oe readily interpre-
ted in terms of landfill conditions
and annual rainfall. If the results
of a leaching test are to be directly
related to landfill conditions, a
correlation between the annual rain-
fall, the waste density, and the
test S/L ratio will need to be made
in order to determine how much leach-
ate a unit area of waste will contact
per unit time, and thus the time span
to which the S/L ratio used in the
test corresponds. The Illinois test
is interesting in that the correla-
tion is determined before rather
than after the test is performed.
C. Time per Elution
Ideally, either each elution
would allow the parameter of inter-
est to come to equilibrium, or it
would be designed to study the
release kinetics of the parameter.
In practice both situations are dif-
ficult to obtain. Different param-
aters may equilibrate at different
rates. Lee and Plumb^' found four
release patterns in a leaching study
using taconite tailings, as shown in
Figure 1. The time span for their
experiment was 500 days. The experi-
ment used a very low S/L ratio (5 to
35 gm taconite per 10 liters dis-
tilled water), with periodic samp-
ling. Not only did equilibration
times for different parameters vary
widely, but for some parameters a
series of reactions occurred which
produced concentration maxima with
subsequent concentration decreases.
The varieties of release patterns
found make it apparent that no one
sampling time could be chosen which
is the best for each of the release
patterns. Nor can a short leaching
test be assumed to measure equilibri-
um concentrations of a parameter, or
even to determine all the parameters
that would be released from a waste.
Of course, the four patterns found
by Lee and Plumb are not the only
release patterns possible; different
wastes may have different and pos-
sibly unique release patterns.
The selection of an elution time
is arbitrary. Some considerations
do apply, however. The test should
be long enough to allow rapidly
equilibrating species to approach
equilibrium and analytically deter-
minable amount of most species to be
released, yet short enough to mini-
mize biological growth in the test
chamber, secondary effects, and con-
sistent saturation of species of
interest. Biological growth can
produce constantly changing condi-
tions and could make test results
very difficult to interpret. Con-
sistent saturation of chemical
species would require many elutions
to be performed. Finally, the time
chosen should be convenient to per-
sonnel, if possible.
By removing samples at various
times during an elution, one obtains
information about the release kine-
tics of a rapidly equilibrating
specie. However, unless the contact
time between the waste and leachate
in the landfill are known, this
information may be difficult to apply
to a landfill situation. It could
be used to determine the equilibrium
concentration of a parameter. The
additional work involved makes this
determination more appropriate in an
intensive test rather than a stand-
ard leaching test.
D. Number of Elutions
The information obtained from
more than one elution often justi-
fies the extra work involved. Suc-
cessive elutions can indicate the
release pattern of a parameter over
time, and often can give an idea of
the factors affecting the release
of the parameter. Successive elu-
tions are particularly important
when the release of one parameter,
A, is inhibited by the release of
another parameter, B. For example,
imagine a leaching situation with a
38
-------
Type
Example
Asymptotic
Release
Specific Conductance
Alkalinity, Ca,
Mg and others
Exponential
Release
Silica (a slow
hydrolysis step
needed before Si
is solubilized)
Release
followed by
loss from
solution
Co,Zn - loss due
to rising pH in
solution or
absorption back
onto solids
C
o
•H
-M
-p
0)
o
a
o
o
No Release
Several species
Time
Figure 1
Types of Release in Long-Term Leaching Test Mentioned by
Lee and Plumb with Examples from their Leaching Experiment
Using Talconite Tailings^).
39
-------
waste containing a soluble basic
parameter (e.g.,a carbonate) and an
acid soluble, base insoluble com-
ponent (e.g., a trace metal), being
leached by an acidic leachate. The
carbonates in the waste will neutra-
lize the acid leachate until the
carbonates have been leached from
the waste. Incoming acidic leach-
ate will then reestablish acidic
conditions and bring the trace metal
into solution. If only one elation
were used, or if the test were ended
before the acidic pH had been rees-
tablished, the potential for trace
metal leaching would be completely
overlooked. More than one elution
can also sometimes indicate the
factors controlling the release of
a soluble parameter—steady concen-
trations over several elutions may
indicate solubility or desorption
control, whereas, a rapidly falling
concentration indicates washout.
The additional information
obtained from repeated leachings
needs to be balanced against the
extra work involved. The experience
of the authors is that the most use-
ful information is obtained in the
first several leachings. One reason-
able approach is to use a set number
of elutions, say three to five, with
more elutions suggested if an indi-
cator parameter, e.g. pH, has not
returned to a baseline value.
In the discussion above, it has
been assumed that the same waste
sample has been eluted several times
with fresh leachate. An alternative
approach is to use the same leachate
sample to elute several fresh samples
of waste. This procedure provides
information regarding the maxi-
mum concentrations that a parameter
can reach in the leachate, rather
than indicating release characteris-
tics. By using both elution tech-
niques—replacing either the leach-
ing media or the waste in subsequent
elutions—one can obtain consider-
ably more information about the waste
than by using either procedure alone.
E. Temperature
Temperature should have an effect
on the leaching pattern of a waste
due to its effects on solubility
and reaction kinetics. Generally,
however, leaching tests have been
conducted at room temperature, and
the effects of temperature on the
leaching pattern of a waste within
the range of average laboratory tem-
peratures may not be great enough to
Justify specially controlled Temper-
ature conditions. Temperature should
be measured and reported, however.
Occasionally, temperature control
can be very important if the waste
itself is affected by temperature.
For example, if a solid component of
the waste melts at room temperature»
constant temperature conditions are
important.
F. Agitation Technique
An agitation technique which pro-
motes mixing without causing waste
particle or container abrasion is
needed. Agitation is needed to avoid
concentration gradients between the
leachate in contact with the waste
and that at a distance; however,
overly vigorous agitation can cause»
particle abrasion CBoyle, et. al.c ')
and give unnaturally high results.
One agitation technique, used in some
tests, involves shaking for a short
time followed by settling. The
authors found that this procedure
results in the development of signif-
icant concentration gradients between
the settled waste and the leaching
solution, and thus is not aggressive
enough for a good leaching test.
Other methods, such as recriprocal
shaking, wrist action shaking or cir-
cular shaking are more suitable pro-
vided they produce well-mixed sys-
tems and are slow enough so as to
not promote abrasion.
G. Surface Area Contact Between
Waste and Leachate
For some wastes, the amount of
surface area in contact with the
leaching solution can be important
in controlling parameter concentra-
tions in the leachate. For example,
viscous liquid or solid wastes which
are water-impervious but which con-
40
-------
tain water soluble parameters, can
show this behavior. Such species
can be leached from the surface of
the waste, where they are in contact
with the water, but not from the
interior of the waste since the
waste is impervious to water. Dif-
fusion through the waste is generally
too slow for these species to reach
the surface.
If this situation is known to
occur with a particular waste, the
surface area of the waste in the
test should be measured and the
release calculated per unit surface
area as well as per gram. Inter-
pretation of this data, however, may
be difficult unless it is known how
the waste is going to be landfilled
and whether physical breakdown of
the waste occurs with time. The
surface area of a waste may be con-
trolled initially during sample
preparation before the test by grind-
ing, cutting, etc., or by the agi-
tation technique.
Summary of Existing Tests
Several batch leaching tests have
been developed. A survey of some of
the existing tests has been done by
the Mitre Corporation^1 . A summary
of the surveyed tests (plus two
additional tests) is given in Table
3. The table provides both the
range and the frequency at which
values occur within the range for
each of the various test variables
discussed in this section. For
those factors for which the selec-
tion of a value is somewhat arbi-
trary, as in the S/L ratio or the
elution time, the range of values
reported might be given considera-
tion in the specification of values
to be used in a test, and an aver-
age value (median or mode) used.
For other factors (especially the
number of elutions, for example),
average values have little meaning.
The wide variety in all the speci-
fied factors indicates the need
for a standardized test so that
results on different wastes and by
different laboratories would be
comparable.
A Standard Leaching Test
As part of a background study on
leaching tests for the EPA, a stand-
ard leaching test has been devel-
oped by the authors (Ham, et.
al.(4)). The test is summarized
below.
The test utilizes only the sol-
id portion of the waste being stud-
ied. The liquid portion, namely,
that which will pass through a 0.45
micron filter, is analyzed directly
for the components of interest. The
rationale for the separation is that
the liquid component of the waste
can move away from the solid^portion
in the landfill, either due to
gravity or capillary flow, or to
absorption by surrounding materials.
The liquid component of a waste
represents an intrinsic potential
impact on water quality as a result
of waste disposal which is not de-
pendent on external sources of
water or leaching media. The solid
portion remains behind for leaching,
as simulated by the leaching test.
Two elution procedures are used
in the test, as shown in Figure 2,
one in which the waste is replaced
in each elution (Procedure C) and
one in which the leaching media or
eluent is replaced (Procedure R).
Procedure C is designed to estimate
the maximum concentrations of leach-
able species arising in the leachate
as a result of leaching media con-
tact with the solid portion of a
waste, while Procedure R estimates
the amounts of leachable specie's to
be released. The latter may be ex-
pressed with most wastes on a weight
of each constituent released per
unit weight of waste basis, for
example. Both procedures use three
24-hour elutions with appropriate
leaching media. The most common
leaching media are likely to be
distilled, deionized water and
synthetic.municipal landfill leach-
ate, useful for modeling monoland-
fill and co-landfilling with mixed
municipal refuse situations, re-
spectively. If co-disposal with
wastes which are not properly mod-
eled by either of these leaching
41
-------
Table 3
Summary of Existing Leaching Test Variables
(Number of Tests Specifying Each Operating Variable Indicated)
Leachates No.
H-0 (dist, deion, dist-
deion or unspecified) 17
H^O with pH adjustment or
simple acid or base 5
Site specific 1
Acetate buffer 1
Synthetic municipal land-
fill leachate 1
Synthetic natural rainwater 1
Bacterial nutrient media 1
Tests with more than one
leachate 5
solid/
liquid
ratio
Time
per
elution
range 1:1- <
1:500
range 30 min
-10 days
1:4 1:4
4 4
1-
<1 hr 24
hrs
1:5 1:10 >1:10
352
24 hr 48 72 hr
hrs
varied
2
>72 hr
calculated
1
to
"equil. "
No. of range 1-10
elutions
1
15
3
1
5
1
7
1
10
2
Agitation
shaker, stirring £ gas agitation used. Two tests use
short agitation times with extended settling times.
Surface
area
Unspecified
media is involved, appropriate media
should be used. The most likely
media in these situations would be
of suitable controlled pH or a
media obtained by distilled water
contact with the other wastes
involved.
For the first elution, both pro-
cedures use a 1:10 solid to liquid
ratio (dry weight to volume). A
separate sample of the waste is
used to determine drv weight. In
Procedure C, at the end of the first
elution the sample is filtered
and a portion of the leachate
for analysis. Fresh waste is then
added to the leachate. The amount
of waste added is such that the
solid to liquid ratio is 1:7.5 and
1:5 in the second and third elu-
tions, respectively. The test is
42
-------
Determine Moisture Content
Procedure C
Solid/Liquid Ratio 1:10
Separate
Solids Removed
Portion of Liquid
Removed for Analysis
Replace Solids
S/L 1:7.5
liquid
Day 1
solid
f///7/7
Day 2
Separate
Solids Removed
Portion of Liquid
Removed for Analysis
Replace Solids
S/L 1:5.0
Day 3
Separate
Liquid Removed
Procedure R
Solid/Liquid Ratio 1:10
Separate
Liquid Removed and Analyzed
Replace Liquid
S/L 1:10
Separate
Liquid Removed and Analyzed
Replace Liquid
S/L 1:10
Separate
Liquid Removed and Analyzed
Further Leaching of Solids
if Necessary
I
Analyze and Interpret Results
Figure 2
Standard Leaching Test Flow Scheme
43
-------
ended after the third elution. Pro-
cedure R uses fresh leachate on the
same waste sample for each elution,
thus maintaining a constant solid/
liquid ratio of 1:10. More than
three elutions may be used if there
is a reason for the additional elu-
tions (such as variable leachate
composition after each successive
elution). The test is run at
ambient temperatures, unless special
conditions dictate the use of a
specific constant temperature. A
rotating mixer is used in conjunc-
tion with a square sample bottle.
The mixer is tilted to give an
almost-vertical rotation plane. As
the bottles turn, the samples slide
down the square side and turn over
somewhat in the process rather than
simply staying at the bottom as
would be the case in a round bottle.
This shaking technique has been
found to give good mixing with lit-
tle or no abrasion. A 1 revolution
per minute mixing speed is used. A
more detailed description of the
test and its development is given by
Ham, et. al.(4).
Interpretation
Of the test conditions mentioned,
leachate composition and tempera-
ture are most readily related to
landfill conditions, as discussed
previously. Other test conditions
are more difficult to base on land-
fill conditions and are chosen based
on criteria other than landfill
modeling. Since these conditions
can affect a parameter's concentra-
tion in the test leachate, caution
must be exercised in data interpre-
tation. One cannot always transpose
test data to landfill practice
directly. It may be possible to cor-
relate test conditions with landfill
concentrations by running extensive
verification tests, correlating a
waste's behavior in the test with
the behavior of the same waste in a
carefully monitored landfill. Cor-
relation coefficients could then be
developed for parameters and con-
ditions similar to those in the
verification study, and the test
result used to estimate landfill
concentrations.
A short leaching test cannot
completely duplicate the long-term
leaching characteristics of a waste.
The short test might completely miss
a parameter having exponential re-
lease (see Figure 1) or might over-
estimate the release of a parameter
showing a concentration maxima.
These patterns could be seen in a
long-term study, but are difficult
to determine in a short test.
When interpreting test results,
it is important to consider the
physical condition of the landfilled
waste. Of special concern in this
regard would be a waste which is
landfilled in large stable chunks
or with a stable impervious coating
which could behave far differently
in a landfill than in a test in
which it were ground. Although not
specifically tested as part of this
study, it seems reasonable to cut,
crush, or specially make such wastes
to yield particles approximating the
size equivalent to a 1 cm. cube,
for example. This particle size is
small enough to work in the proce-
dure yet large enough to not increase
drastically the surface area per unit
weight waste exposed to leaching.
This is observed by plotting the sur-
face area per volume contained for
cubes of varying dimension.
^
H
i
E
0
•— '
UJ
i
o
UJ
ac.
ou
25
20
15
10
5
0
1 1 1 1 1
^_ __
•
_ _
- -
- -
_ * _
* '
1 1 1 1 1
0 0.5 1.0 1.5 2.0 2.5 3.0
CUBE SIDE DIMENSION
(cm)
An evaluation of the hazardous
nature of a waste must incorporate an
evaluation of the waste's landfill
environment. A waste's hazardous
44
-------
nature is a situation-specific
characteristic. For example, a
waste may be hazardous to an organ-
ism under one set of environmental
conditions yet completely innocuous
under a different set of conditions.
Furthermore, its hazard may be or-
ganism specific, i.e., it may be
hazardous to one organism and not to
another under the same set of condi-
tions. Thus a determination of the
hazardous nature of a waste must
include an evaluation of its effect
on specific organisms (or plants or
animals, etc.) under specific con-
ditions.
One obvious way to interpret the
leachate composition results is
to compare the concentrations of the
various chemical species to some
standard, for example, drinking
water standards. This is dangerous,
however, and is difficult to defend
for the leaching test developed in
this study. Obviously, test re-
quirements could have been adjusted
to yield virtually any concentration
of leachable species in the leach-
ate for analysis. This could have
been done, for example, by use of
different elution times (e.g., 5
minutes instead of 24 hours), or
solid/liquid ratios. The test was
designed to be rapid, aggressive,
and to yield as much information
about the leaching characteristics
of a waste as possible in a rela-
tively short time. It was not de-
signed to provide realistic concen-
trations of the various species for
a specific situation.
One reasonable method for inter-
pretation of test results involves
adjustment of measured chemical
specie concentrations by a factor.
This factor may be based in part on
the amount of leaching media the
waste might contact during the esti-
mated active life of the landfill.
Thus, the cumulative infiltration
expected over a ten-year period at
an appropriate landfill might be
used. This could be further modi-
fied by the amount of waste expected
to be landfilled per unit landfill
area. Another way of developing
interpretive criteria would be to
dilute both the leaching media and
the elutriates obtained by waste con-
tact to the point where no toxic
effects are noted in a toxicity test.
The difference in dilutions required
to provide no toxic effects would be
a measure of the incremental effect
of the leachable constituents of a
waste on the leaching media, and
would thus be an indication of the
hazardous character of the waste.
Once again the difference would
have to be interpreted knowing that
the test is designed to be aggress-
ive, but at least the various base-
level degrees of toxicity inherent
in the selection of the raw leaching
media are taken into account.
In summary, the standard test pro-
vides a rapid indication of which
chemical species are immediately
leached from a waste, and an indica-
tion of the maximum concentrations
of each specie likely to be found in
the leachate. In addition, an
estimate of the amount of each specie
likely to be released per unit weight
of waste is obtained. Finally, the
test can provide valuable informa-
tion with regard to the relative
effect of co-disposal of the waste
in question with other wastes or
mixed municipal refuse. The fact
that the test is not perfect in
predicting the long-term leaching
pattern of a waste, or the precise
concentration of a particular
parameter in a particular landfill,
for example, means that the test
results need to be interpreted with
care.
References
1. Abelson, A. and Lowenbach, W.,
"Procedure Manual for Environ-
mental Assessment of Fluidized
Bed Combustion Processes,"
Mitre Corp., M77-34, January
1977.
2. Lee, G.F. and Plumb, R.H.,
"Literature Review on Research
Study for the Development of
Dredged Material Disposal Cri-
teria," Contract Report D-74-1,
Office of Dredged Material Re-
search, U.S. Army Engineers
45
-------
Waterways Experiments Station,
Vicksburg, Miss., 1974.
3. Boyle, W.C., Ham, R.K., Kunes,
T., Liu, T., and Kmet, P.,
"Assessment of Leaching Poten-
tial from Foundry Process
Solid Wastes," Final Report,
submitted to American Foundry-
men's Society, Des Plaines,
Illinois, August 1977.
U. Ham, R.K., Anderson, M.A.,
Stegmann, R., and Stanforth, R.,
"Background Study on the De-
velopment of a Standard Leach-
ing Test," Final Report on
Grant No. R-804773010, to be
submitted to U.S.E.P.A.,
Washington, B.C., 1978.
Acknowledgement
This study was supported by the
U.S. Environmental Protection Agency
under Grant No. R-80U773010. Mr. M.
Gruenfeld was the Project Officer.
46
-------
GENETIC TOXICITY TESTING OF COMPLEX ENVIRONMENTAL EFFLUENTS
F. W. Larimer and J. L. Epler
Biology Division, Oak Ridge National Laboratory, Oak Ridge, TN 37830
ABSTRACT
We have used the Ames Salmonella histidine-reversion system and other genetic test
systems to assay the mutagenic potential of crude synthetic oils and natural crude oils
as exemplary complex mixtures. Extracts and leachates from particulate matter are
also considered. Mutagenicity data on isolated or suspected organic components are
presented. The results support the use of the short-term genetic tests in examining
crude mixtures and point to the advantages of coupling the bioassays with chemical
fractionation.
INTRODUCTION
The enormous amount of industrial and
technological activity in the modern world
creates a large number of chemical pol-
lutants. The developing synthetic fuel
industry is only one example that could
have significant environmental impact on
man. Although the health effects of chemi-
cals in the environment are being exten-
sively studied, it has become obvious that
complex mixtures represent an unusual
analytical problem.
Our work has emphasized test materials
available from the developing synthetic
fuel technologies(1). However, the pro-
cedures are applicable to a wide variety of
industrial and natural products, environ-
mental effluents, and body fluids. The
general applicability of microbial test
systems has already been demonstrated with,
for example, the use of the assay as a
prescreen for potential genetic hazards
of complex environmental effluents or
products, e.g., tobacco smoke conden-
sates(2), soot from city air<3), fly
ash(4)f and, in our work with synthetic
fuel technologies, oils and aqueous
wastes(5-6)_ other potential hazards
arise from materials leached from raw
materials, spent oil shales, ash/slag and
other particulates.
In addition to obvious toxic effects,
chemical pollutants may also produce car-
cinogenic, mutagenic, or teratogenic ef-
fects whose expression may be divorced in
time from the actual exposure(s). The
research effort described here specific-
ally approaches the question of genetic
hazard; but, as has been pointed out
recently O, certain microbial genetic
assays (for example, the Ames test with
Salmonella) show a high correlation be-
tween positive results in mutagenicity
testing and the carcinogenicity of the
compounds under test. The overall need to
subject environmental chemicals to muta-
genicity testing has been discussed in
By acceptance of this article, the publisher or recipient, acknowledges the right of the
U.S. Government to retain a non-exclusive, royalty-free license in and to any copyright
covering the article.
Research jointly sponsored by the Environmental Protection Agency(IAG-D5-E681 and
IAG-40-646-77) and the Department of Energy under contract with Union Carbide Corporation.
47
-------
the Committee 17 Report on Environmental
Mutagenic Hazards^). Their key recom-
mendation can be summarized as: "screen-
ing should be initiated as rapidly and as
extensively as possible." DeSerres has
discussed the utility of short-term tests
for mutagenicity(9) and the prospect for
their use in toxicological evaluation^0).
He stresses that the data base of know-
ledge on untested environmental chemicals
should be expanded, but cautions that "we
are not ready, however, to extrapolate
directly from data obtained in short-term
tests for mutagenicity directly to man"^1^,
Tests on other organisms must be performed
to validate and reinforce results from
short-term tests. Short-term tests simply
point out potential mutagenic and carcino-
genic chemicals and serve to order priori-
ties for further testing in higher organ-
isms. Our approach has been to develop
validating data in conjunction with ongoing
analysis of synthetic fuels using Salmon-
ella. Table 1 lists the assays currently
being employed.
of histidine mutants that revert after
treatment with mutagens to the wild-type
state (growth independent of histidine).
Both missense mutants and frame-shift
mutants comprise the set, and their rever-
sion characteristics with a potential
chemical mutagen qualify the mechanism of
action. In addition, the detection scheme
yields the highest resolution possible by
the inclusion of other mutations: (a) the
deep rough mutation, rfa, which affects
the lipopolysaccharide (LPS) coat, making
the bacteria more permeable; (b) the
deletion of the uvrB region, eliminating
the excision-repair system(11), and
(c) R factor plasmids, which increase
sensitivity to mutagens.
Mutagenicity Testing. The Salmonella
typhimurium strains used in the various
assays are listed below. All strains were
obtained through the courtesy of
Dr. Bruce Ames, Berkeley, California.
MATERIALS AND METHODS
TA1535 hisG46, urvB, rfa (missense)
TA100 hisG46, uvrB, rfa (missense plus
R factor)
TA1537 hisC3076, uvrB, rfa (frameshift)
The Salmonella tester series is composed TA1538 hisD3052, uvrB, rfa (frameshift)
TABLE 1
SHORT-TERM BIOASSAYS
Test system
Assay
Salmonella
E. coli
Yeast
Drosophila
CHO cells
Human leukocytes
Mouse
Carcinogenesis
his •> his
+
arg ->• arg
gal" -> gal+
his -> his
CAN3 -»• canr
Sex-linked recessive
lethals
6-thioguanine
resistance
chromatid aberrations
dominant lethals
skin painting
48
-------
TA98 his D3052, uvrB, rfa (frameshift plus
R factor)
In the routine screening of fractiona-
ted materials, the two strains TA98 and
TA100 were generally used. Experimental
procedures have been given by Ames
et al.dl). The strain to be treated with
the potential mutagen(s) is added to soft
agar containing a low level of histidine
and biotin along with varying amounts of
the test substance. The suspension
containing approximately 2 X 108 bacteria
is overlaid on minimal agar plates, and
revertants to wild type are counted after
a 2-day incubation. The assay is quanti-
tated with respect to dose (added amount)
of test material and modified to include
treatment with the liver homogenate re-
quired to metabolically activate many
compounds.
Fractions and/or control compounds to be
tested were suspended in dimethylsulfoxide
(DMSO, supplied sterile, spectrophotometric
grade from Schwarz/Mann) to concentrations
in the range 10-20 mg/nil solids. The po-
tential mutagen was in some cases assayed
for general toxicity (bacterial survival)
with strain TA1537. Generally, the
fraction was tested with the plate assay
over at least a 1,000-fold concentration
range with the two tester strains TA98 and
TA100. Revertant colonies were counted
and plotted versus added concentration, and
the slopes of the induction curves were
determined. Positive or questionable re-
sults were retested with a narrower range
of concentrations. All studies were car-
ried out with parallel series of plates
plus and minus the rat liver enzyme pre-
paration^-'-' for metabolic activation.
Routine controls demonstrating the steri-
lity of samples, enzyme or rat liver S-9
preparations, and reagents were included.
Positive controls with known mutagens were
also studied in order to recheck strain
response and enzyme preparations. All
solvents used were nonmutagenic in the
bacterial test system.
Samples. Exemplary oils that have been
tested and their sources are listed below:
(1) coal-liquefaction product from a pro-
cess under development, courtesy of the
Pittsburgh Energy Research Center (Synfuel
A), or Coal A from ORNL repository; (2) a
crude shale-oil sample (B) from the
above-ground simulated in situ oil-shale
retorting process; (3) Louisiana-Missis-
sippi sweet crude oil, courtesy of
Dr. j. A. Carter of the Analytical Chemis-
try Division, Oak Ridge National Labora-
tory.
The authors recognize the possibility
that these samples may bear no relation-
ship to the process as it may exist in the
future, nor should it be construed that
these materials are representative of all
natural crudes, synthetic or shale-oil
processes. They are used here simply
as appropriate and available materials for
the research.
Fractionation. The fractionation scheme
described by Swain et al.d2), as modified
by Bell et al.<13) , has been applied to the
product from the oil-shale process, Synfuel,
and the natural crude. The separation
scheme has been illustrated and detailed
in reference 14. Reagent-grade solvents
and glass-distilled water were used for
extraction and chromatography.
RESULTS AND DISCUSSION
Chemical analyses of synthetic fuel
materials(15-16) ancj predictions based on
work with tobacco smoke conden-
sates(12,17,18)^ give the investigator a
general view of the organic components of
the various fractions. Based on these
predictions, we list a selected group
(Table 2) of organic compounds pertinent
to shale oil and synfuel, and the prelimi-
nary results on mutagenicity in the Sal-
monella histidine-reversion system.
Although a number of the compounds may be
promoters or modifiers of carcinogenesis
(that is, active in co-carcinogenesis,
perhaps in some cases as inhibitors), they
do not appear to be mutagenic.
The major mutagenic components appear
to be heterocyclic nitrogen compounds,
aromatic amines, and polycyclic aromatic
hydrocarbons.
In the investigation of the feasibility
of the coupled analytical-mutagenicity
assay approach, we fractionated the simu-
lated in situ retorting shale-oil sample
the synfuel sample, and the natural crude
"control" oil sample into primary acidic,
basic, and neutral components. Each pri-
mary fraction was then assayed with the
Ames strains. Data are given for the
frameshift strain TA98 with metabolic
activation with enzyme preparations from
49
-------
TABLE 2
MUTAGENICITY OF CHEMICALS KNOWN OR SUSPECTED TO OCCUR IN
SYNFUELS USING SALMONELLA TESTER STRAINS
Chemical name
Acenaphthene
Anthracene
Benz [a] anthracene
Benz [a] anthracene, 7, 12-dimethyl
Benzo [a] f luorene
Benzo[a]pyrene
Chrysene
Fluoranthene
Naphthalene
Phenanthrene
Pyrene
Triphenylene
1,2,3, 4-Dibenzanthracene
1,2,5, 6-Dibenzanthracene
Acridine
Carbazole
Isoquinoline
Quinoline
7-Methylquinoline
8-Methylquinoline
2 , 6-Dimethylquinoline
8-Hydroxyquinoline sulfate
8-Nitroquinoline
8-Aminoquinoline
a-Naphthylamine
2-Acetamido fluorene
Strain Mutagenicity
98, 100 +
98, 100
100 +
100 +
100 +
98 +
100 +
98 +
98, 100
98 +
1537 +
98 +
100 +
100 +
1537 +
100
98, 100
98 +
100 +
100 +
98, 100
100 +
100
1537 +
100 +
98 +
Aroclor 1254-induced rats (Table 3). The
shale-oil and natural crude samples both
contain significant activity in the neutral
fractions and appear to represent compar-
able mutagenic hazards. However, the
shale-oil material appears to contain
additional activity in other fractions,
particularly the BE ether-soluble fraction.
Synfuel is qualitatively similar to shale
oil, although the Bja fraction is the most
active encountered from these samples.
In addition to these exemplary samples,
we have been applying the mutagenicity
assay to a large group of synthetic fuel
applications (Table 4).
The results presented here show that
biological testing-within the limits of the
specific system used-can be carried out
with complex organic materials but perhaps
only when coupled with the appropriate
analytical separation schemes. The pri-
50
-------
TABLE 3
MUTAGENICITY OF OIL FRACTIONS IN SALMONELLA
Fraction
Natural Crude
Synfuel
Shale oil
NaOH
WA
WAS
SAI
SAE
SAw
Bla
B
+
+
Ib
Bw
Neutral
Revertants with frame-shift strain TA98.
- denotes no response over background.
+ denotes approximately 100 colonies over background.
++ denotes approximately 1000 colonies over background (at most
effective volume added).
+++ denotes greater than 1000 colonies over background.
mary use that such combined chemical and
biological work may serve is to aid in iso-
lating and identifying the specific class-
es or components involved.
In these initial feasibility studies,
the purpose has not been to reflect on
whether a relative biohazard exists in
comparison with other materials or pro-
cesses. We feel an extrapolation to rela-
tive biohazard at this point would be, at
least, premature. A number of precautions
are listed below.
A number of biological discrepancies
can enter into the determinations. For
example, the short term assays chronically
show negative results with, e.g., heavy
metals and certain classes of organics.
Similarly, compounds involved in or requir-
ing co-carcinogenic phenomena would pre-
sumably go undetected. Concomitant bac-
terial toxicity can nullify any genetic
damage assay that might be carried out;
the choice of inducer for the liver enzy-
mes involved can be wrong for selected
compounds; and additionally, the applic-
ability of the generally used Salmonella
test to other genetic assays and the vali-
dation of the apparent correlation be-
tween mutagenicity and carcinogenicity
still remains a point of significant
fundamental research.
51
-------
TABLE 4
OIL SAMPLES FOR MUTAGENICITY TESTING
Control
Coal liquifaction
Louisiana-Mississippi Sweet
Natural Crude Mix
Synfuel A (PERC)
Synfuel B (COED)
Separator liquor
Hydrocarbonization Process
Hydrocarbonization liquid
waste
SRC Process
Shale Oil
*
Shale Oil
Product Water
Raw Shale
Spent Shale
Coal Gasification
Gasifier Condensates (PERC)
Low BTU Gasification
Ash, Tar, Particulates
Aqueous
Extract*
In situ retort.
Above ground retort.
The chemical integrity of the sample
may be compromised. The choice of samp-
les, their solubility, and the time and
method of storage are unavoidable variables
which may alter the composition of a
sample. The detection or perhaps the
generation of mutagenic activity may well
be a function of the chemical fractiona-
tion scheme utilized. The inability to
recover specific chemical classes or the
formation of artifacts by the treatment
could well corrupt the results obtained
in addition to the possibility of an
inability to detect the specific biologi-
cal endpoint chosen.
In light of the difficulties pointed
out above, we feel that over-interpretation
of biohazard potential, i.e., rankings of
relative hazard or reliance on negative
results, should be approached with caution.
However, in the context of a prescreen to
aid the investigators in ordering their
priorities, the short-term testing appears
to be a valid testing approach with com-
plex mixtures. Our experience to date with
a variety of comparative assays (Table
5)(l^) supports the contention that short
term testing can be a useful predictor of
potential genetic hazard in higher systems.
52
-------
TABLE 5
COMPARATIVE MUTAGENESIS OF FRACTIONS FROM
SYNTHETIC CRUDE OILS3
Test system
Salmonella
E. coli
Yeast
Drosophila
CHO cells
Human leukocytes
Mouse
Carcinogenesis
Assay Basic fraction Neutral
fraction
his •> his + +
arg -*• arg + +
gal" ->• gal+ + +
his" -»• his"1" + +
CAN8 -> canr + +
SLRL +
6-thioguanine + NT
resistance
chromatid P +?
aberrations
dominant lethal s - P
skin painting p P
Crude
synfuel
NT
NT
NT
NT
NT
NT
NT
+
P
The fractions utilized were generally those from Synfuel A-3, or Synfuel B-2.
+ = mutagenic; - = nonmutagenic; NT = not tested; and P = in progress.
Crude synfuels are generally too toxic to test in most systems.
°Work of J. M. Holland, Oak Ridge National Laboratory, in progress.
1. Klass, D. L. "Synthetic crude oil
from shale and coal", Chem. Technol.
August, 499-510, 1975.
2. Kier, L. D., Yamasaki, E., Ames, B. N.
"Detection of mutagenic activity in cigar-
ette smoke condensates", Proc. Natl. Acad.
Sci. USA, 71:4159-4163, 1974.
3. Ames, B. N., McCann, J., Yamasaki, E.
"Methods for detecting carcinogens and
mutagens with the Salmonella/mammalian
microsome mutagenicity test", Mutat. Res.
31:347-364, 1975.
4. Chrisp, C. E., Fisher, G. L.,
Lammert, J. E. "Mutagenicity of filtrates
from respirable coal fly ash", Science
199: 73-75, 1978.
5. Epler, J. L., Young, J. A., Hardigree,
A. A., Rao, T. K., Guerin, M. R.,
Rubin, I. B., Ho, C.-h., and Clark, B. R.
"Analytical and biological analyses of
test materials from the synthetic fuel
technologies. I. Mutagenicity of crude
oils determined by the Salmonella
typhimurium/microsomal activation system",
Mutat. Res., in press.
6. Epler, J. L., Rao, T. K., Guerin, M.R.
"Evaluation of feasibility of mutagenic
testing of shale oil products and
effluents", Environ. Health Prespect, in
press.
7. McCann, J., Choi, E., Yamasaki, E.
and Ames, B. N. "Detection of carcinogens
as mutagens in the Salmonella/microsome
53
-------
test: Assay of 300 chemicals. Part I".
Proc. Nat. Acad. Sci. USA 72:5135-5139,
1975.
8. Drake, J. W., Abrahamson, S.,
Crow, j. F., Hollaender, A., Lederberg, S.,
Legator, M. S., Neel, J. V., Shaw, M. W.,
Sutton, H. E., von Borstel, R. C., and
Zimmering, S. (Committee 17). "Environ-
mental mutagenic hazards", Science 187:
503-514, 1975.
9. de Serres, F. J. "The utility of
short-term tests for mutagenicity", Muta-
tion Res., 38:1-2, 1976.
10. de Serres, F. J. "Prospects for a
revolution in the methods of toxicological
evaluation. Mutat. Res. 38:165-176, 1976.
11. Ames, B. N., Lee, F. D., and
Durston, W. E. "An improved bacterial
test system for the detection and classifi-
cation of mutagens and carcinogens", Proc.
Nat. Acad. Sci. USA, 70:782-786, 1973.
12. Swain, A. P., Cooper, J. E., and
Stedman, R. L. "Large scale fractionation
of cigarette smoke condensate for chemical
and biologic investigations", Cancer Res.
29:579-583, 1969.
13. Bell, J. H., Ireland, S., and
Spears, A. W. "Identification of aromatic
ketones in cigarette smoke condensate",
Anal. Chem. 41:310-313, 1969.
14. Rubin, I. B., Guerin, M. R.,
Hardigree, A. A., and Epler, J. L.
"Fractionation of synthetic crude oils from
coal for biological testing", Environ.
Res. 12:358-365, 1976.
15. Ho, C.-h., Clark, B. R., and
Guerin, M. R. "Direct analysis of organic
compounds in aqueous by-products from
fossil fuel conversion processes: Oil
shale retorting, synthane coal gasification
and COED coal liquefaction", J. Environ.
Sci. Health, All, 7:481-489, 1976.
16. Guerin, M. R., and Epler, J. L.
"Determining emissions measurements needs
for an emerging industry-advanced fossil
fuels utilization", paper presented at
the First Conference on "Determining
Fugitive Emissions Measurements Needs,"
Hartford, Connecticut, May 17-19, 1976.
17. Bock, F. G., Swain, A. P., and
Stedman, R. L. "Bioassay of major frac-
tions of cigarette smoke condensate by an
accelerated technic", Cancer Res. 29:
584-587, 1969.
18. Patel, A. R., Haq, M. Z.,
Innerarity, C. L., Innerarity, L. T.,
and weissgraber, K. "Fractionation studies
of smoke condensate samples from Kentucky
Reference Cigarettes", Tob. Sci. 18:58-59
1974.
19. Epler, J. L., Larimer, F. w.,
Rao, T. K., Nix, C. E., Ho, T. "Energy-
related pollutants in the environment: The
use of short term tests for mutagenicity
in the isolation and identification of
biohazards", Environ. Health Perspect.,
in press.
54
-------
A FRAMEWORK FOR ECONOMIC ANALYSIS OF HAZARDOUS
WASTE MANAGEMENT ALTERNATIVES
Graham C. Taylor
Research Economist, Colorado School of Mines
Golden, Colorado 80401
(On leave from the University of Denver Research Institute,
University Park, Denver, Colorado 80208)
and
Oscar W. Albrecht
Senior Economist, U.S. Environmental Protection Agency
Office of Research and Development
26 W. St. Clair St., Cincinnati, Ohio 45268
ABSTRACT
This paper describes a methodology for analysing the economic and social effects
of alternative approaches to hazardous waste management. The procedure involves the
generation of a series of environmental 'threat scenarios' that might arise from the
use of different hazardous waste management techniques, and identification of the
'parties-at-interest' to these techniques. By considering the likely attitudes of
these parties-at-interest, a decision-leaker can take into account social impacts as
well as direct economic costs of alternative approaches to hazardous waste control.
INTRODUCTION
The purpose of this paper is to
introduce some concepts and techniques
that have been developed as part of an
ongoing EPA-sponsored study.(1) A
primary objective was to develop a
jcethodology for the economic analysis of
hazardous waste management problems.
This methodology provides an analytical
framework designed to assist a
decision-maker to identify costs and
other Impacts associated with a given
approach to hazardous waste management.
The general approach adopted is
that of cost-risk-benefit analysis, but
the methodology discussed here departs
from the traditional approach in two
ways. Tather than developing expected
values for environmental damages, it
concentrates on identifying and char-
acterizing the threats to man and the
environment which may arise from the use
of various hazardous waste management
techniques. Secondly, from the effects
identified for these management
techniques, it identifies 'parties-at-
interest' and endeavors to predict ways
in which these parties iray respond.
Their responses will, in turn, influence
the outcomes from the use of alternative
approaches to hazardous waste management.
The methodology is in the form of a
general framework that is adaptable to
specific situations. It is intended to
help a decision-maker consider both the
economic and social aspects of hazardous
T?aste management alternatives. It also
55
-------
permits evaluation of the effects of
whatever degree of risk aversion Is
favored.
UNIQUE ASPECTS OF HAZARDOUS WASTE
MANAGEMENT
As a first step towards developing
a methodology for the analysis of
hazardous waste management alternatives,
an economist may ask the question "what
is special about hazardous wastes, i.e.,
how does the economics of the manage-
ment of these wastes differ from that of
other wastes or pollutants?" A clear-
cut answer to this question is difficult
to find, but a definition of hazardous
waste provides some indications of
their special characteristics. A waste
is hazardous if it
. . . poses a substantial present
or potential hazard to human health
or living organisms because such
wastes are lethal, nondegradable or
persistent in nature or because
they can be biologically magnified.
or may otherwise cause or tend to
cause detrimental cumulative effects
[emphasis added]. (2)
Several distinguishing character-
istics relevant to economic analysis
emerge from this definition. First, a
hazardous waste can pose a substantial
threat to man or the environment,
suggesting that hazardous wastes need
more careful management than inert
wastes or conventional pollutants (e.g.,
B.O.D.). Consequently, a management
approach often used for conventional
pollutants, that of permitting dis-
charges that can be assimilated by the
environment,may not be acceptable for
hazardous wastes.
Secondly, hazardous waste manage-
ment will be move concerned with
'threats' or risks* as opposed to
readily anticipated environmental
impacts. If, for example, a paper mill
discharges an organic waste to a river,
for some distance below the discharge
point the level of dissolved oxygen in
the water will fall. We have a
reasonably good Idea of what may happen
to the aquatic life in this zone, and
are relying on natural biological
processes to 'treat' the waste and
render it harmless, providing some local
environmental degradation is acceptable.
On the other hand, if a heavy metal
waste is injected into a saline aquifer
as a means of disposal, the Intention is
that the waste remains in the aquifer
and thereby causes no harm to the
environment. One must, however, be
concerned about threats that may arise
from the use of this disposal method.
The saline aquifer may, for example, be
interconnected with an aquifer used as a
source of fresh water, which thereby
becomes contaminated. Unlike the
comparatively predictable effects from
non-hazardous waste management techniques
it is usually difficult to estimate the
probability that a threat will materialize,
and the magnitude or cost of the
potential damage may also be hard to
predict.
Thirdly, hazardous wastes are often
persistent or non-degraddble. This is
significant because (i) effects may be
irreversible, and (11) time scales of
Interest can be intergenerational.
Injecting a heavy metal waste into an
aquifer may be irreversible. (3)
Should the aquifer later be needed as a
water or mineral resource, or be found to
be interconnected with a fresh water
source, decontamination might not be
feasible. Irreversibllity can also
Introduce 'option value1 and associated
concepts that relate to the benefits of
keeping options available. Thus, it
may be desirable to delay making an
irreversible decision until improved
information is available (e.g., on the
toxicity of a waste proposed for dis-
charge) ; or changing values may mean
that an irreversible decision made today
is a regretted decision in the future.(4)
When today's decisions affect
future generations, it raises the diffi-
cult problem of whether or not
discounting is appropriate. The
National Academy of Sciences pointed up
this difficulty as follows:
There have been long standing
debates as to the appropriateness of
applying a discount rate to effects
on future generations, since any
positive rate of discount will
directly discriminate in favor of
choices that involve bad Impacts on
56
-------
later generations but not on earlier
ones. ... if the discount rate were
5 percent, 100 cases of toxic
poisoning 75 years from now would be
equivalent to about 3 cases today; or
1 case today would be valued the same
ae 1,730 cases occurring in 200 years
.... Clearly, intergenerational
effects of these magnitudes are
ethically unacceptable. . . . Some
other method of ethically weighting
intergenerational incidence of
effects must be devised. . . . there
is no generally accepted method to
weight intergenerational incidence of
benefits and costs. (5)
Fourthly, hazardous wastes include
those that are biologically magnified or
have cumulative effects. These factors
compound the difficulties of estimating
damages due to exposure. It should,
however, be noted that this problem is not
unique to hazardous waste, as pollutants
that are not normally regarded as hazardous
may also have cumulative effects.
A final characteristic of hazardous
wastes (one that is not apparent from the
definition), is that the composition of
wastes can vary substantially, not only
because of different sources, but also
from day to day for a given source. This
complicates treatment to reduce the hazard
and can inhibit resource recovery activ-
ities. Furthermore, it may mean that the
degree of hazard posed by the waste is not
well defined. (6)
To summarize, hazardous wastes are
characterized by strong potential adverse
effects, and their management may involve
irreversible decisions and intergenera-
tional time scales. Because insufficient
information is available, the threats that
these wastes pose to man and the environ-
ment are often difficult to precisely
specify. Waste composition variability,
magnification and cumulative effects
compound the problem.
APPLICATIONS OF COST-RISK-BENEFIT ANALYSIS
Techniques of cost-benefit and risk-
benefit analysis are well developed, and
have already been applied to several
classes of environmental problems*. For
example, cost-benefit analysis has been
applied to air and water pollution control
programs, (10) while cost-risk-benefit (or
risk-benefit) analysis has been used to
compare alternative means of generating
electric power (11).
Cost-benefit analysis is usually used
to determine whether or not a project or
an activity should be undertaken. Such
analysis frequently Involves determining
the optimum scale of activity; i.e., the
project scope at which the net benefit
(total benefits less total costs) is
maximized. Analysis of this type could be
appropriate to deciding whether or not to
create a park or to preserve a natural
environment as a wilderness area.
Where we are dealing with pollution
control, the economic activity that
generates the pollutant is already in
existence. Hence analysis considers only
costs associated with pollution, and the
conventional economic approach to pollution
control becomes that of determining the
optimum level of pollution, and devising a
policy that results in that level. Con-
ceptually (but not in practice) this is
straightforward, and involves controlling
the discharge of the pollutant to the
level which minimizes the total cost,
i.e., Ql in Figure 1.
*The distinction between cost-benefit and
risk-benefit analyses is not usually clear-
cut. The term risk-benefit analysis is
often applied to a category of cost-
benefit analysis in which risks to life and
health are an Important component of the
costs. (7) It would not be applied to an
analysis in which the risks are purely
economic (e.g., where there is construction
cost uncertainty, or where there is doubt
about the magnitude of the project
benefits). Some authors dealing with risks
to life retain the term cost-benefit
analysis (8), while others use the term
cost-risk-benefit analysis to suggest that
the costs Include both conventional costs
and risk-related ones (9).
57
-------
INCREASING POLLUTION
(DECREASING ENVIRONMENTAL QUALITY)
FIGURE 1. CONVENTIONAL APPROACH TO POLLUTION CONTROL
Cost-risk-benefit analysis can examine
whether or not an activity that involves
risk should be undertaken or permitted. It
could, for example, be used to decide
whether or not a particular toxic substance
should be manufactured. Alternatively, the
net benefit of the activity may be assumed
to be positive, and the analysis used to
compare different ways of achieving the
objective. Thus, In assessing alternative
means of generating electric power the
expected numbers of mining and transporta-
tion accidents, etc. have been estimated
and expressed as a cost; but the need for
power has not been questioned. (12,13)
A feature of many cost-risk-benefit
assessments is that they use expected
values to describe the risks. By assigning
economic values to the risks, an expected
value of total cost can be obtained, and
the least cost means of achieving the
specified objective determined. Alterna-
tively, the expected value approach can
allow the optimum level of exposure to
risk (e.g., radiation exposure from mantmo-
graphy (14)) to be determined in a manner
that is analogous to Figure L This is
possible where risks are statistically
well defined; but the approach takes no
account of a decision-maker's risk aver-
58
-------
sion, which may be the critical factor in
policy decisions.
Hazardous vastee ere highly diverse,
and the threats that they pose to man and
the environment can vary considerably
with the circumstances. This makes it
difficult to accurately define the proba-
bility that a threat will materialize, and
to estimate the magnitudes of associated
damages. The conventional economic ap-
proach to pollution control (Figure 1)
implies a single pollutant or some index
of pollution as the independent variable.
Construction of an index representing the
magnitude of threat would be complex. In
any event, for most problems data require-
ments would be enormous. The irethodology
suggested here is irtended to overcome
these difficulties.
DESCRIPTION OF TEL METHODOLOGY
The aspects of the methodology
described here are those related to the
use of threat scenarios and the social
interaction process.
THREAT SCENARIOS
Assessing the damages caused by pollu-
tion is almost invariably more difficult
than determining control costs. According
to Fisher and Peterson, there are four
stages in the assessment of damages from
conventional pollution sources (15), as
shown in the upper part of Figure 2.
Starting with a specified emission or vaete
discharge, the ambient conditions and the
physical effects must be determined before
the dollar damage costs can be estimated.
To extend Fisher and Peterson's model to
hazardous wastes it is necessary to add a
preliminary stage; identification of the
possible threat mechanisms. One could then
model the effects of the threats and arrive
at dollar costs associated with each
threat.
The central concept of the methodology
presented here is to replace the first
three stages (emissions, ambient conditions
and effects) of Fisher and Peterson's
damage model by a 'threat scenario* as
illustrated in the lower part of Figure 2.
The threat scenario vould describe what
might typically happen when a specified
threat became a reality. More than one
scenario could be used to depict different
outcomes from a giver threat mechanise.
While qualitative descriptions are useful,
where possible quantitative data should be
used, based on modelling studies or judg-
APPROACH FOR CONVENTIONAL POLLUTION PROBLEMS
EMISSIONS
E.G. CU. M. OF
RAW SEWAGE
TO RIVER
AMBIENT
CONDITIONS
P.P.M. OF
DISSOLVED
OXYGEN
EFFECTS
THOUSANDS OF
DEAD FISH, ETC.
DOLLAR
COSTS
VALUE OF
DEAD FISH, ETC.
THREAT
MECHANISM
E.G. INJECTED
WASTES INFILTRATE
FRESHWATER AQUIFER
THREAT
SCENARIO
THOUSANDS OF
WELLS CONTAMINATED
OVER SPECIFIED AREA
COSTS & IMPACTS
HEALTH COSTS,
COST OF REPLACEMENT
WATER SUPPLY
ADAPTATION TO HAZARDOUS WASTE PROBLEMS
FIGURE 2. DAMAGES ASSOCIATED WITH HAZARDOUS WASTE MANAGEMENT
59
-------
meats reflecting actual experience with
that type of threat.
A major difficulty with threat
scenarios that are not based on adequate
data, will be to choose & scenario in
which the magnitudes of the effects are
appropriate to the types and quantities of
wastes concerned, and to the local circum-
stances (e.g., geohydrologic conditions).
In some situations, worst case assumptions
could be used to place upper limits on
effects. In evaluating the effect of land-
fill leaching, for example, it could be
assumed that after many years a 'steady
state* is reached in which the leachate
contains the same quantities of non-
degradable toxic elements as enter the
landfill, and that this leachate enters
the local river system without attenuation.
Given the streamflow, the average concen-
tration of toxic elements in the river can
be calculated, and the effect on aquatic
life predicted.
Clearly numerical estimates that do
not adequately reflect local conditions
are unlikely to be accurate. We are,
however, Interested In individuals' reac-
tions to the threats from hazardous waste
management alternatives. These reactions
may reflect past experience with a similar
threat mechanism under different circum-
stances. (16) Responses to threats depend
on perceived magnitudes and probabilities,
which may be at variance with reality.
Hence, the mere identification of threats
ia an Important part of the methodology,
even if the magnitudes of their impacts
and the probabilities of their occurence
are not well defined.
Each threat scenario provides a
largely physical description of what may
happen from the use of a given hazardous
waste management technique. In principle,
each scenario can be translated into
economic terms using established tech-
niques for valuing the effects. (17,18)
In practice, however, It can be difficult
to attribute dollar values to some types
of environmental Impact. Additionally,
there may be major uncertainty ahout the
magnitude of these impacts. Eence, in the
lower part of Figure 2, the term costs and
impacts has been used for the final stage
(I.e., the output) of the threat scenario
approach. This refers to dollar costs
where determinable, and to descriptions
of social Impacts where dollar costs
cannot realistically be assigned.
IBENTIFICATIOK OF THREATS
A key step in analyzing alternatives
is to identify environmental threats
associated with each control technique.
Figure 3 presents a morphological map
that demonstrates in a general way how a
threat evolves from some initiating event
to some physical outcome. (19) Although
the map is not fully detailed, it can be
seen that there are numerous ways in which
threats could develop. It will be neces-
sary to limit the. threats considered in
analysis to those believed to be compara-
tively probable, based on previous exper-
ience or consensus judgment taking local
conditions into account. Table 1 lists
the major techniques that are available
for hazardous waste management, and
identifies some of the more important
threats that may arise from the use of
each technique.
There is some published information
on the environmental threats that may
arise from hazardous wastes. Lazai; et al.,
have analyzed the mechanisms involved in
over 400 'incidents' (i.e., threats that
have materialized) arising from the dis-
posal of hazardous waste to lend. (20)
Ground and surface water pollution have
been by far the most common outcomes of
inadequate disposal methods, followed by
contact poisoning. The EPA has published
detailed descriptions of some of these
incidents (21-27), while other sources
contain additional data on damage from
leachates and injected wastes. (28,29)
Data on transportation accidents are
widely available, and soire statistics on
in-plant accidents have been developed.(3C)
Dawson and RtrecJley have published data on
the probabilities and sizes of spills of
hazardous chemicals, and have also deter-
mined the probability that these spills
will reach surface waters.(31) Fish kill
reports collected and published by the EPA
provide a useful indication of the impacts
on surface waters from spills and
dumping. (32)
There have recently been E number of
attempts to estimate the probability and
effects cf Eon>e infrequent or gradual geo-
logic and climatic events (e.g., earth-
quake, erosion, tornado) that might affect
the disposal of high level radioactive
wastes. (33-35) Schneider ar
-------
INITIATING
EVENTS
-THREAT MECHANISMS
OUTCOMES
o>
ODOUR
OLEFACTORY
INSULT
FIGURE 3. MORPHOLOGICAL MAP OF ENVIRONMENTAL THREATS
-------
TABLE 1. HAZARDOUS WASTE MANAGEMENT TECHNIQUES AND ASSOCIATED THREATS
MAJOR EBVIFOKMEN1AL TKK1ATS
All techniques
Surreptitious dumping to lard, sewer
or vatervay
Accidental flrc/fxplcsion/spillage
leading to ^amage/deetructlcu of
property and life (all forms)
Vnter cortairlratlon
riuonic pcisoring (vcrkere)
The threats teJov, arc in addition to those listed above
COMMEKTS
Threats can occur at generating or
trenti't-rt plants, aluo during
transport
Change lr. waste
stretiT'B
-process change
-source reduction
-vast* separation
To generate wastes that are lees
hazardous
To generate less of hazardous waste
To separate hazardous vaste fror
non-hazardous waste
Resource recovery
-Materials recovery
-energy recovery
Treatment to reduce Treatment nay fail leaving vaste still Failure most likely with biological
trrotirent
Physical and cherical treatment
fey facilitate resource recovery
To imrobillze wastes
hazard
-physical
-cher.ictl
-biological
-thermal
-encapsulation
Land application
hazardous
LaodfiUJOB,
Mine disposal
Lac
OOP Ing
for storage cr
evaporation)
Air pollution
Crop take-up of toxic elements
Soil sterilization
Loacblng/run-off
Odor
Leachiue/run-off
Odor
Leakage to grourdwater
Seepage to f.roundwater
Ovcrflov
Odor and air pollution
Vector could also be a threat
Deep well injection Contamination of usable groundwater
Land movement ai.d (consequent)
property dosage
Ocean duaping
Modification of iiarlne ecosyster
Engineered storage Containment failure
Space disposal
Contamination cf space
Discharge to sever Waste may destroy biological treatment
colonies
Waste ray cause sewage sludge to
becomfe hazardous
''Slugs" of waste nay overload ayatea
Discharge to
waterway
Watte may destroy aquatic life
Some rodificoticn of groundwater
syetetr Is inevitable
Threat is inevitable but extent
difficult to predict
Failure poses save threat as All
Techniques-ace identa
Threat largely unknown
Not hazardous waste oanagenent
techniques, but included because
they nay be Involved accidentally
or surreptitiously
62
-------
none currently exists, and that it will
penetrate a mine disposal site. (36)
THE INTERACTION PROCESS
Figure A presents the authors'
schematic model of the interaction process
involved in hazardous waste management
decision-making. This model is central to
the methodology and demonstrates all the
essential elements used in the analytical
framework, as described below. Policy
objectives are considered to be an
exogenous input to the analysis.
POLICY
LEVEL
POLICY
OBJECTIVES
APPROACHES
TO HAZARDOUS
WASTE MANAGEMENT
TECHNIQUES
FOR THE CONTROL OF
HAZARDOUS WASTE
PHYSICAL
OR
TECHNICAL
LEVEL
I
OUTCOMES
(I.E. WHAT PHYSICALLY
HAPPENS)
ENVIRONMENTAL
IMPACTS
(•THREATS; ETC.)
ECONOMIC AND
SOCIAL EFFECTS
OF TECHNIQUES
^ WASTE
DISPOSITIONS
RESPONSES
OF THE PARTIES-
AT - INTEREST
PARTIES-
AMNTEREST
(THOSE AFFECTED)
SOCIOECONOMIC
LEVEL
ECONOMIC COSTS
A SOCIAL IMPACTS
FIGURE 4. INTERACTION MODEL FOR HAZARDOUS WASTE MANAGEMENT
63
-------
Approaches to hazardous waste management
Each approach to hazardous waste
management represents a general concept or
actual strategy consistent with the policy
objectives to be attained. Approaches can
be quite specific, such as requiring land-
fill disposal of all hazardous wastes, or
prohibiting ocean dumping. Alternatively
approaches can endeavor to influence waste
management by providing economic incentives
for desirable techniques or disincentives
for those considered undesirable. Examples
include tax-breaks for resource recovery,
or effluent charges on certain disposal
techniques.
Techniques for the control of hazardous
waste
From the approaches to hazardous waste
cariagement, it is possible to predict which
control techniques are likely to be used In
practice. The term technique refers to the
physical methods that may be used to elim-
inate or reduce hazardous waste generation,
to treat wastes, or to dispose of them
(refer to Table 1). The linkages between
approaches and techniques may not be clear-
cut, and the approaches are likely to
change and be developed as the analysis
progresses. Therefore a wide range of
techniques should be considered in the
early stages of analysis.
Environmental impacts
The next step is to determine the
environmental impacts. The major impacts
associated with each technique will be
threats. These can range from the very
unlikely (e.g., earthquake destruction) to
virtual certainties. In general, emphasis
should be placed on identifying and char-
acterizing the impacts of the more probable
threats associated with each technique.
If, however, a relatively unlikely threat
poses disastrous consequences, this should
also be considered. Resource consumption
(e.g., energy, land use) may also be a
significant environmental impact.
Economic and social effects
The effects of the techniques on man
are. shown as economic and eccial effects in
Figure A. These effects may be both direct
and indirect (primarily via threats).
Direct effects Include the waste generator's
disposal costs, program administration costs
and other costs including subsidies given
to encourage preferred waste management
techniques.
The major effects that can arise via
environmental threats are listed below.
(1) Destruction or damage to
man-Fade structures.
(11) Damage to human life and
health.
(ill) Destruction or damage to
animals, vegetation and land ecosystems.
(iv) Fish and other aquatic life
kills In surface waters.
(v) Changes in property values.
The first four effects can inflict costs on
various parties ranging from those actually
incurred (e.g., clean-up and repair costs)
to those that reflect aesthetic considera-
tions. Loss of option value (for example,
not preserving an aquifer that may repre-
sent a useful resource in the future) is
another form of cost. In principle there
are ways to attach an economic coot to each
of these effects, but in practice some will
defy quantification, and may be listed as
social impacts (see Figure 4).
Changes In property values are
interesting because they provide a measure
of the external effects of pollution. (37)
They may reflect changes in consumers1
utility due to the physical presence of
pollution, or they rcay be completely
psychological, reflecting concern that some
sort of incident may occur (e.g., that a
landfill may pollute a property owner's
well). The proceeding example illustrates
that it is the way in which individuals
perceive a threat that will determine
their behavior. Thus, a threat does not
have to actually materialize, or even have
a significant probability of occurence, in
order to be important to an Individual,
and analysis should take account of this.
Parties-at-Interest
In order to conduct analysis that
recognizes differing viewpoints, it is
64
-------
useful to identify partiea-at-intereet,
Each party-at-interest consitutes a class
or group of individuals or firms that have
a common Interest and viewpoint on the
outcome of any particular hazardous waste
management alternative. For analysis, the
parties-at-interest must be limited to
those that are significantly affected by
one or more of the alternatives. Gilmore,
et al., who developed this concept, list
four categories of parties-at-interest, as
follows:
• Parties internal to the affected
industry: e.g., owners, stock-
holders, management, employees and
their unions.
• Suppliers and customers of an
affected industry: e.g., vendors
of materials and of services in-
cluding financing, insurance and
advertising, intermediate and
final consumers. A more compre-
hensive listing may be available
from an Input-output analysis.
• Government: e.g., at different
levels, and in different roles.
Includes legislator, executor,
adjudicator, taxer, regulator, and
keeper of economic stability,
social welfare, and national
security.
• Affected bystanders: e.g.,
resources, wildlife, recreation
potential, those concerned vith
aesthetic effects, and those
secondarily involved such as
investors, employees, residents,
end other property owners. (38)
their jobs safer. The effect on the waste
disposal and transport industries will
depend on the process streams following
treatment (+-); they may have safer
wastes to dispose of (+), or there may be
no waste requiring off-site disposal (-).
Local officials are likely to favor
chemical treatment due to the reduced risk
of ar. environmental incident (+), but
water supply authorities might be con-
cerned over the possible discharge of an
undesirable effluent to a river that
constitutes part of a water supply (-).
While water supply authorities and environ-
mentalists are likely to have definite
views on most techniques, the perception
of threats and benefits from chemical
treatment may be remote to most local re-
sidents (no entry in Table 2). Thermal
treatment, on the other hand, could cause
the deterioration of local air quality and
might be viewed negatively by many resi-
dents and property owners (-), while a
technique such as lagooning could pose a
definite threat to local water supply
users (-).
Responses of the parties-at-interest
The parties-at-interest will oppose
or seek to mitigate effects perceived as
being adverse, but will accept favorable
effects, although they may be less likely
to behave aggressively. Parties-at-
interest may respond in ways that have
direct economic consequences; for example,
waste generating firms will seek to pass
some or all increased operating costs on
to their customers, while residents
adversely affected by air pollution might
move to another location.
Table 2 presents a matrix of the
parties-at-interest for each major tech-
nique, and suggests the nature of the
effect that use of the technique has on
each party-at-interest. It may be useful
to examine a few of the entries in Table 2
to understand how the authors' judgments
about the nature of the effects were made.
Consider, for example, a technique in-
volving chemical treatment tc reduce
hazard potential. Management of the firm
generating the waste may have nixed views
about this (+-); it is likely to be
comparatively costly, but treatment should
reduce the risk of an adverse incident.
The firm's workers are likely to favor
treatment (+) because it probably makes
Outcomes
The physical cvtcomea of a specific
approach to hazardous waste management
follow from the responses of the parties-
at-interest and will include environmental
impacts that actually occur, as well as the
disposition of the wastes. Outcomes cculd
include such effects as reduced sales of
products resulting frou higher prices,
changes in property values, etc.
The outcomes provide feedback to the
approaches to hazardous veste management.
A decision-maker can examine the outcores
and associated economic costs and social
65
-------
TABLE 2. MATRIX OF EFFECTS ON THE PARTIES-AT-INTEREST
Party-At-Interest
r- i
Nature of effect
Very favorable
Generally favorable
Mixed, or depends on
circumstances
Generally adverse
Very adverse
Not significant
Hazardous Waste Control
Other
Farties-at-Intereet
Change in waste stream
Resource recovery
Treatment - physical
- chemical
- biological
- thermal
- encapsulation
Land application
Landfill - secure
- ordinary
Mine disposal
Deep well injection
Ocean dumping
Engineered storage
Space disposal
4—
4-
4—
4-
-
4-
-
+
4—
4
-
+
+
-
—
4-
4—
4-
+
4—
4—
+
4-
+
4—
+
+
+
+
+
4—
4—
4-
4-
4-
4-
4-
4-
4-4
-
44-
-
-
4-
4-
4-
-
+-
4-
-
-
4
4-
44-
4
44-
4-
44-
4-
4—
4-
4-
4—
4—
-
-
++
+
++
+-
++
4-
4-
-
4-
4
44-
4-
4
-
-
-
-
-
-
4-
—
-
4-
-
4-
4-
+
+
4-
4-
4-
-
4-
-
4-f
-
-
4-
-
4-
+
+
4-
4—
4-
4-t-
4-
4-
-
44-
—
—
4-
-
-
-
-
—
-
-
-
-
-
-
-
+-
+
4—
4—
-
4-
4
-
4-
-
4
-
4-
4-
4-4-
-
4—
-
-
4-
4-
4-
-
-
-
—
-
4-
4-f
4-
4-
4—
4-
44
-
4-
-
4+
-
—
4-
—
•
-
—
-
-
+
4—
-
-
-
Virgin material suppliers
Chemical suppliers
Farmers
Adjacent mineral interests
rowl nutitere, lonters
Other nations, politicians
Other nations, politicians
Discharge to sewer
Discharge to waterway
+
44-
4-
4-
-
—
—
—
-
-
-
~
1
-
-
-
-
-
'
— (Sever authorities i
-- (Other water recreationlsts
-------
Impacts for each approach, and can then
modify an approach, or select a new ap-
proach that appears to be more desirable.
OTHER ASPECTS OF THE METHODOLOGY
It is beyond the scope of this paper
to fully detail the methodology; however,
a few additional specific features will be
mentioned.
Equity is a normative facet of eco-
nomics, and identification of various
parties-at-interest is particularly useful
for considering the equity of an approach.
By examining the way in which costs and
iupacts fall on different parties-at-
interest, the decision-maker can evaluate
the acceptability of the results. He can
also devise strategies to render a given
approach more equitable by finding ways to
shift some of the costs and impacts between
parties-at-intercst. Examination of the
alternatives for the disposal of a parti-
cular waste might, for example, lead to
the conclusion that economic efficiency
would be served by discharging th.ls waste
to a landfill, but that this would render
the water ir a limited number of wells
unsafe to drink. To make this solution
more equitable, arrangements could be made
whereby the waste generator pays for the
cost of installing and operating an alter-
native water supply. An additional pay-
ment could be made to compensate the well
owners for a loss of psychic value caused
by changing their water supply to an
alternative source.
Intertemporal considerations
As already indicated, there IB some
doubt as to the appropriateness of dis-
counting where irreversible decisions are
involved that have effects over inter-
generational time spans. This probleu has
recently beer addressed in contexts other
than waste management, and that work sug-
gests that the appropriate solution may
depend on the nature of ary altruism that
the present generation extends towards
later generations. (39-41) On the other
hand, discounting ic almost universally
used by industry and is accepted t>y eco-
nomists where decisions are reversible or
have a duration that is limited to one
generation.
A possible practical solution to this
dilemma would be to discount effects that
occur during the first generation (i.e., a
normal project life of two or three
decades) but not to further discount any
effect that occurs or extends teyond this
time. (42) This suggestion is particularly
attractive for use with this methodology as
it eliminates some of the problems assoc-
iated with the timing of threats. In most
cases it is difficult to predict the time
at which a threat will materialize if it
does. Uncertainty may arise because the
initiating event is random, because effects
are cumulative and involve threshold levels,
or simply because the data required to
predict the timing are not available. If
effects that occur after one generation
are all discounted by the same amount,
prediction of their timing becomes
unimportant.
Unfortunately, it is possible only to
use lump sum valuations of effects (e.g.,
a one-time clean-up or replacement cost
for the effect of a threat) if the period
of evaluation is infinite. Any annual
cost (such as an option value or the cost
of perpetual care) continuing for an in-
definite period will have an infinite pre-
sent value if it is not discounted. Thus,
where no lump sum equivalent can be found
to replace a cost that extends indefi-
nitely, the evaluation horizon must be
limited or discounting must be employed.
Flexibility of approach
An attractive feature of the methodo-
logy is that it provides a flexible frame-
work for analysis that can readily accom-
modate inputs from a variety of sources.
Waste management personnel, for example,
may generate a number of threat scenarios
that they consider to be the most relevant
to a given situation. If, however, it
becomes apparent that the public is largely
concerned with some other threat, an
appropriate scenario can be added without
disrupting or contradictirg the previous
work.
DEMONSTRATION OF THE METHODOLOGY
This section provides a sicple hypo-
67
-------
thetical example of the use of the
methodology discussed above.
TEE PROBLEM
You are the administrator of any
agency responsible for hazardous waste in
your area. A firm is installing a new
process which will generate 250,000 cu.m.
per year of an aqueous waste containing
50 parts per million of non-degradable
toxic elements. The firm proposes to dis-
pose of this waste by injecting it into a
saline aquifer some 600 deters below their
premises. They estimate that this will
cost $50,000 per year (including capital
charges) over a 20 year life span.
Tour staff investigates other possi-
bilities, and their findings are as
follows:
(1) The waste stream can be reduced
to 25,000 cu.m. per year with a corre-
sponding increase in the concentration of
toxic elements, at a cost of $20,000 per
year to the firm.
(11) The waste stream can be
treated to provide an effluent that is
acceptable to the municipal sewer. Treat-
ment results in 250 cu.m. per year of a
toxic sludge. The cost of treatment and
effluent charges would be $115,000 per
year.
(ill) There are two landfills that
could accept either the liquid waste
(from (i), above) or the sludge (from
(11)). The 'local landfill' is a publicly
operated landfill located in a wet zone
Immediately adjacent to a river 50 km. from
the generating firm. The 'secure landfill1
is a privately operated chemical landfill
located in a dry zone 360 km. from the
firm. Transport to either landfill would
be by truck.
(iv) Other possibilities for dealing
with the waste (such as resource recovery
or ocean dumping) are not available.
Hence, there are five technically
feasible disposal plans, as follows:
(A) Deep well injection on the firm's
premises.
(B) Sludge sent to the local landfill
(C) Liquid waste sent to the local
landfill.
(D) Sludge sent to the secure
landfill.
(E) Liquid waste sent to the secure
landfill.
THREAT SCENARIOS
The following scenarios represent the
major threats. In-plant accidents under
any plan are expected to have approximately
similar impacts, and hence do not have to
be evaluated.
Threat Scenario I: Water contamination
from deep well injection
Drinking water is obtained from
numerous wells that penetrate a shallow
aquifer in the vicinity of the generating
firm. This aquifer may become contaminated
as a result of some unanticipated inter-
connection with the deep saline aquifer.
The probability of this is unknown. If
this happens, corrective action could be
taken by providing temporary water supplies
to the local residents, and by drilling
several additional wells into the saline
aquifer and counterpumping to reverse the
migration of the waste. The total cost of
this clean-up operation is estimated to be
some $2,400,000; which would be consider-
ably less costly than providing a permanent
new water supply to the local residents.
Threat Scenario II: Leaching from the
local landfill
If the liquid waste were discharged
to the local landfill, it is assumed that
the entire waste would quickly enter the
river. If the sludge were deposited at
this landfill an appreciable proportion of
the toxic elements would probably be re-
tained, but the above assumption could be
used as the worst case. Either sludge or
liquid waste contributes 12.5 tons of toxic
elements per year. The river has a mean
flow of 200 cu.m. per second implying a
toxic element concentration of two parts
per billion (ppb) If the waste was uni-
formly diluted. However, local concentra-
tions are expected to be higher.
The river supports an important salmon
fishery, and experts value a typical year's
68
-------
fishing at $800,000, excluding indirect
effects such as tourist dollars brought
into the region by the fishery. The ex-
perts expect the onset of high fish
mortality to occur at toxic element con-
centrations somewhere between 20 and 100
ppb, but are reluctant to say what ircpact
two ppb would have on the salmon due to
effect variability with duration of expo-
sure, alkalinity and the presence of other
elements. They point out, however, that
there is some evidence that fish avoid
sub-lethal concentrations of toxic ele-
ments; hence the waste could conceivably
decljnate the fishery by discouraging the
salmon from returning to spawn.
Threat Scenario III; Transport accidents
Statistics indicate that 50 accidents
involving waste spills car be expected per
billion kilometers traveled by truck ir
the region. The clean-up coat associated
with a typical accident is estimated to be
$10,000. Serious injuries ard deaths
directly attributable to the properties cf
the waste are expected to be negligible.
Threat Scenario IV; Flash flood at the
gecure landfill
The most likely threat from the
secure landfill is contaminated run-off
froti a flash flood. Such a flood is ex-
pected to occur less than once per hundred
years and damage along the flood path
directly attributable to the toxic
elements is expected to be minimal.
ANALYSIS OF THE ALTERNATIVES
Table 3 presents a comparison of the
alternative plans. The gate fee for all
wastes at the local landfill is $3.00 per
cu.m. At the secure landfill it is $30.00
per cu.m. for the sludge and $20.00 per
cu.m. for the liquid. Transport costs are
$7.00 per cu.m. to the local landfill and
$22.00 per cu.m. to the secure landfill.
As soon as the control costs arc
evaluated, it ie possible to eliminate
plans C and E because these plans are
'dominated' by B and D respectively.
Dominance occurs when both the quantified
costs and non-quantified costs are greater
for one plan than another. Consider plan
B versus plan C. The control costs for B
are $117,500 per year versus $270,000 per
year for C, and detailed evaluation ie not
necessary to show that the potential
environmental damages from B are also less
than from C. There is less possibility
for release of toxic elements from the
sludge (plan B) than from the liquid (plan
C); while B requires less transportation
than C, resulting in fewer accidents. Thus
plan B is clearly preferable to plan C as
both the quantified costs (the control
costs) and the non-quantified costs (the
environmental dan-age potential) are lower
for B than C. Similar arguments apply to
D versus E.
This reduces the number of plans to
be evaluated to three (A, B and D). The
next step is to examine the threats asso-
ciated with each plan and to determine the
nature of the effects of each plan on the
partiee-at-interest. These effects are
summarized in Table 4.
The parties-at-interest most strongly
affected in this example include the water
supply authority, and to a lesser extent
the water users near the plant, who would
be concerned about the threat from deep
well injection (Threat Scenario I). Fish
experts would oppose plan B, although fish-
ermen and related industry might perceive
only a weak threat. Fishing interests
might, however, have an unexpected ally.
If plan A were prohibited, the generating
firm itself could well prefer D over E.
The annual cost of plan B is only $10,500
greater than plan B, while if the firm
opted for L, it could receive adverse pub-
licity if the fish threat (Scenario II)
materialized. In contrast, the finr night
enhance its reputation as a responsible
environmental citizen if it sent its waste
to the secure landfill under plan D. That
the parties-at-interest analysis shows only
negative impacts for plan B, confirms the
unattractiveness of thle plan in relation
to A and D.
"IT'S YOUR DECISION"
The critical choice is between plans A
and r, and involves reduced control costs
and greater damage potential if A is pre-
ferred to D. If plan A is selected, the
present value of the control costs dis-
counted over the 20 year project is
$664,000 less than for plan D. On the
69
-------
TABLE 3. COMPARISON OF ALTERNATIVE PLANS
CONTROL COSTS (dollars)
Ger crater 1d annual costs
-Cn-site operations
-Transport
-Gate fees
Total annual cost
A
Deep Well
Injection Sludge
PLAN
B C
Local Landfill
115,000
1,750
750
117,500
Liquid
20,000
175,000
25,000
270,000
D E
Secure Landfill
Sludge Liquid
115,000 20,000
5,500 550,000
7.500 500.000
128,000 1,070,000"
Present value 20 yearn' costs
discounted at 107 per year
426,000 1,000,300 2,298,000 1,089,700 9,109,500
Other control costs
-Administration and monitoring
J"ot significantly different^
PLAN A PLAN B
FUJI
LKV!ROK>iiNTAl IMPACTS
Ir-fiant accidentt
Hot significantly different
Major threat scenarios
T: V.'ater supplies con- II: Saliecn fishery
tanlnatcd. at risk.
Mitigation cost Direct value
$2,400,000, plus cost $800,000 per year.
of charting to another Indirect effects
plan. locally important.
Probability: not Probability: not
estimated. estimated.
Ill: Transport
accident.
Clean-up cost
$10,000.
Probability:
2xlO~5/year
III: Transport
accident.
Clean-up cost
$10,000.
Probability:
1.4xlO~*/year.
IV: Contamination
via flash flood.
Damage potential
slight.
Probability:
<10"2/year.
OTHER EFFFCTf
Irodui t pricts vould be slightly Mgher and/or generating firm's
net incoci* slightly lower with Plan E or D than with Plan A
other hand, plan A poses the threat of
water contamination (Scenario I) with its
clean-up costs and the need to find an
alternative disposal method if deep well
injection does contaminate the water
supply. The threats from transport acci-
dents (Scenario III) and from flash floods
at the secure landfill (Scenario IV) ap-
pear to be so minor that they can be neg-
lected. Nevertheless it was important to
70
-------
TABLE 4. PARTIES-AT-INTEREST ANALYSIS
The matrix characterizes the expected attitudes of
the major parties-at-lnterest towards each plan.
PartT-at-interest
FLAN
B
Generating firm's
Management
Local water supply users
Water supply authority
Fish **t?c^
Fishermen, fish-related
industry
Environmentalists
T
—. ! *
Key: see Table 2
recognize them, and demonstrate (or obtain
consensus judgment) that they could be
disregarded.
If plan A was adopted, and problems
with the deep well scheme developed after
(say) five years, the additional costs of
taking the correcting action described
under Threat Scenario I and switching to
plan D for the next 15 years would have a
present value of $1,895,000. Ie it worth
risking $1,895,000 to save $664,000? If
these data completely and accurately repre-
sented the choice, a decision-maker who
was not risk adverse would support plan A
if he believed that the probability that
the threat would materialize was less than
35 percent. The probability that the
threat will materialize is unknown. It
may be one in two, 10%, 1%. . . . If,
however, the threat materialized and cor-
recting action as described in Scenario I
failed, a completely new water supply might
have to be brought in at a cost of tens of
millions of dollars. Does society regard
the risk as acceptable, or would it prefer
to avoid risk and accept higher costs that
would be passed on to consumers in the form
of higher product prices or lower divi-
dends? -If the risk is to be taken, Is the
distribution of the risk and the benefits
acceptable, or Is there a way in which it
can be made more equitable?
It's your decision!
NOTES AND CITATIONS
1. U.S. Environmental Protection Agency,
"Economic Analysis of Alternative Hazardous
Waste Controls." Grant Number R804661, to
Office of Research Services, Colorado
School of Mines.
2. U.S. Environmental Protection Agency,
"Report to Congress: Disposal of
hazardous Wastes," U.S. EPA, SV-115, 1974
p. 3.
3. Unfortunately in real life there are
degrees of reversibility and irreversi-
bility, see Frutilla, J.V. and A.C. Fisher,
"The Economics of Natural Environments:
Studies in the Valuation of Commodity ard
Amenity Resources," The Johns Hopkins
University Press, Baltimore, 1975, for
Resources for the Future, pp. 40-59.
4. Fisher, A.C. and F.M. Peterson, "The
Environment in Economics: A Survey,"
Journal of Economic Literature, v. 14,
March 1976, pp. 1-33.
5. National Academy of Sciences, Comnittee
on Principles of Decision Making for Regu-
lating Chemicals in the Environment,
"Decision Making for Regulating Chemicals
in the Environment," Printing and Pub-
lishing Office, National Academy of
Sciences, Washington, B.C., 1975, pp. 177,
178.
6. These statements are based on discus-
sions with representatives of the
hazardous waste management industry.
7. National Academy of Engineering, Com-
mittee on Public Engineering Policy,
"Perspectvies on Benefit-Risk Decision
Making," The National Academy of Engineer-
ing, Washington, D.C., 1972, pp. 3,4.
8. National Acedetry of Sciences, National
Research Council Advisory Committee on the
Biological Effects of Ionizing Radiations,
"Considerations of Health Benefit-Cost
Analysis for Activities Involving Ionizing
Radiation Exposure end Alternatives," U.S.
EPA, EPA 520/4-77-003, 1977.
9. U.S. Atomic Energy Cowmission, "Com-
parative Risk-Cost-Benefit Study of
Alternative Sources of Electrical Energy,"
U.S. AEC, WASH-1224, December 1974.
10. Peskin, II.M. and E.P. Seskln (eds.),
"Cost-Benefit Analysis and Water Pollution
Policy," The Urtan Institute, Washington,
D.C., 1975.
71
-------
11. Barrager, S.M., B.R. Judd and D.W.
North, "The Economic and Social Costs of
Coal and Nuclear Electric Generation: A
Framework for Assessment and Illustrative
Calculations for the Coal and Nuclear Fuel
Cycles," prepared under NSF contract CEP-
75-06564, U.S.G.P.O., March 1976.
12. Ibid.
13. U.S. Atomic Energy Commission, loc.
cit.
14. National Acadeny of Sciences, National
Research Council Advisory Committee on the
Biological Effects of Ionizing Radiations,
op. clt., pp. 143-187.
15. Fisher and Peterson, loc. cit.
16. Attitudinal research has provided
some support for this position. See, for
example, Elovic, P., H. Kunreuther and
G.F. White, "Decision Processes, Ration-
ality, and Adjustment to Natural Hazards,"
Natural Hazards: Local, National and
Global, White, G.F. (ed.), Oxford Univer-
sity Press, New York, 1975, pp. 187-205.
17. Peskin and Seskin, loc. cit.
18. Maler, K.G. and P.E. Wyzga, "Economic
Measurement of Environmental Damage : A
Technical Handbook," Organisation for
Economic Co-operation and Development,
Paris, 1976.
19. The morphological map is used to dis-
play the components that iray be involved in
either fault tree or event tree analysis.
For information on these two types of anal-
ysis, see: Fischhoff, B., "Cost Benefit
Analysis and the Art of Motorcycle Main-
tenance," Policy Sciences, v. 8, 1977,
pp. 177-202.
20. Lazar, E.C., R. Testani, and A.E.
Giles, "The Potential for National Health
and Environmental Damages froo> Industrial
Residue Dispose],|: a paper presented at
the National Conference or. Disposal of
Residues on Land, sponsored by U.S. FPA,
Hazardous Waste Management Division,
September 15, 1976.
21. Gbassemi, M., "Analysis of a Land
Disposal Damage Incident Involving Haz-
ardous Waste Materials: Dover Township, Few
Jersey," U.S. FPA, contract 68-01-2956,
hay 1976.
22. Shuster, K.A., "Leachate Damage
Assessment: Case Study of the Fox Valley
Solid Waste Disposal Site in Aurora,
Illinois," U.S. EPA, EPA/530/SV-514,
June 1976.
23. Shuster, K.A., "Leachate Damage
Assessment: Case Study of the Peoples
Avenue Solid Waste Disposal Cite in
Rockford, Illinois," U.S. EPA, EPA/530/
SW-517, June 1976.
24. Shuster, K.A., "Leachate Damage
Assessment: Case Study of the Sayville
Solid Waste Disposal Site in Islip (Long
Island), Nev York," U.S. EPA, EPA/530/
SW-509, June 1976.
25. U.S. Environmental Protection Agency,
"Hazardous Waste Disposal Damage Reports "
U.S. EPA, EPA/530/SW-151, June 1975.
26. U.S. Environmental Protection Agency,
"Hazardous Waste Disposal Eamage Reports-'
Document No. 2," U.S. EPA, EPA/530/SW-151 I
December 1975.
27. U.S. Environmental Protection Agency,
"Hazardous Waste Disposal Damage Report;
Document No. 3," U.S. EPA, EPA/530/SW-151 3
June 1976. '
28. U.S. Environmental Protection Agency,
"The Report to Congress; Waste Disposal
Practices and Their Effects on Ground
Water," U.S. EPA, January 1977.
29. Reeder, L.R., et al., ''Review and
Assessment of Deep-Well Injection of Haz-
ardous Waste," 4 volumes, prepared under
EPA contract no. 68-03-2013, June 1977.
30. Buckley, J.L. and S.A. Wiener,
"Hazardous Material Spills: A documenta-
tion and Analysis of Historical Data,"
prepared under EPA contract no. 68-03-0317
April 1976.
31. Dawson, G.W. and M.W. Stradley, "A
Methodology for Quantifying the Environ-
mental Risks from Spills of Hazardous
Material," a paper presented at AIChF
Conference, September 8, 1975, in Boston.
32. U.S. Environmental Protection Agency,
"Fish Kills Caused by Pollution in 1973,"
U.S. EPA, EPA-440/9-75-003, (annual scries).
33. Energy Research and Development Admin-
istration, "Alternatives for Long-Term
72
-------
Management of Defense Eigh-Level Radio-
active Waste, Savannah River Plant, Aiken,
South Carolina," ERDA, 77-42/1, v. 1,
May 1977.
34. Claiborne, B.C. and F. Gera, "Poten-
tial Containment Failure Mechanisms and
Their Consequences at a Radioactive Waste
Repository in Bedded Salt in Hew Mexico,"
Oak Ridge National Laboratory,
ORNL-RM-4639, October 1974.
35. Schneider, K.J. and A.M. Platt (eds.),
"High-Level Radioactive Haste Management
Alternatives," U.S. AZC contract
AT(45-1):1830, BNWL-1900, May 1974.
36. Ibid.
37. Fieher and Peterson, loc. cit.
38. Gilmore, J.S., et al., "Environmental
Policy Analysis: Public Policy Inverven-
tion in Inter-Industry Flows of Goods and
Services to Reduce Pollution," Denver
Research Institute, University of Denver,
National Science Foundation grant GI-11,
August 1971, p. 92.
39. Page, T., "Conservation and Economic
Efficiency: An Approach to Materials
Policy," The Johns Hopkins University
Press, Baltimore, for Resources for the
Future, 1977.
40. Krutilla and Fisher, loc. cit.
41. Schulze, V., "Social Welfare Functions
for the Future," American Economist, v. 18,
no, 1, 1974, pp. 70-81.
42. Some justifications for this approach
have been developed by one of the authors
and will be included in the complete
report of this research.
73
-------
ECONOMICS OF DISPOSAL AND THE COMPILATION OF
A DATA BASE FOR STANDARDS/REGULATIONS OF
FGD SLUDGE
John P. Woodyard
SCS Engineers
Long Beach, California
Donald E. Sanning
USEPA
Cincinnati, Ohio
ABSTRACT
Regulations regarding the control of sulfur dioxide emissions from power plants
have suggested wet scrubbing of flue gas as the best means of meeting Best Available
Control Technology emission levels, yet some disagreement exists as to how the resulting
throwaway sludges can be disposed of in an environmentally acceptable fashion. Cognizant
of this problem and its effect in deterring many plants from installing FGD systems, the
EPA has sponsored a series of 19 studies aimed at evaluating the effectiveness of the
various disposal methods. The purpose of this paper is to comment on the data base
established thus far, and to project its potential impact on the promulgation of sludge
disposal regulations.
The data base includes information in the following general areas:
• The technology of FGD sludge disposal;
• Environmental criteria by which to evaluate the
various disposal options; and
• Existing or proposed regulations/standards which have
been or could be applied to the disposal of FGD sludges.
The available methods of FGD sludge treatment and disposal are reviewed. Available
cost information for these methods is summarized. The potential environmental impacts
of FGD sludge disposal are enumerated, with supporting data from ongoing research
efforts. The past and present federal and state regulatory approaches are reviewed as
they relate to each of these impacts. Future regulatory approaches are discussed, with
specific reference to SCS Engineers study findings and recommendations.
INTRODUCTION
The Clean Air Act and its amendments
have established standards of performance
for new fossil fuel-fired steam generators.
Because of the lack of commercially
available alternatives to conventional
combustion technology, most utilities have
chosen to use either low sulfur fuel or
wet scrubbing flue gas desulfurlzation
(FGD) systems to meet S02 emission limita-
tions. A decline in low sulfur fuel use
will result from the most recent Clean Air
Act amendments.
The vast majority of planned and
operational FGD systems employ lime or
limestone as the sorbent. The by-product
sludges from lime/limestone FGD are gener-
ated in large volumes and are typically
disposed of in lagoons or landfills. The
chemical and physical nature of these
waste sludges has prompted EPA to investi-
gate the environmental impact of FGD sludge
74
-------
disposal to land, and to demonstrate and
develop treatment and disposal/utilization
methods which are perhaps more environ-
mentally acceptable.
SCS Engineers, Long Beach, California
was contracted by EPA in 1975 to compile
and analyze the findings of previous and
ongoing FGD sludge disposal research.*
It was recognized at that time that FGD
sludge disposal would some day be regulated
at the federal level. The study was there-
fore structured as a preliminary guideline
development document using the data base
established thus far as the basis for
analysis. This paper summarizes the
SCS findings.
INDUSTRY CATEGORIZATION
Description of Power Generating Facilities
In 1977, there were 1,040 fossil
fuel-fired power plants operating in the
United States, possessing a combined
generating capacity of more than 320,000
megawatts. By 1994, 200 more plants with
an additional 111,041 megawatts of gener-
ating capacity are projected to become
operational. Of this 1994 capacity,
66 percent will fire coal exclusively.
The remainder will use oil, natural gas,
or a combination of oil and coalU).
Under the federal Clean Air Act,
power plants are responsible for emission
control. For control of sulfur oxide
emissions, the two most widely used
methods are:
1. The use of a natural low sulfur
fuel or partially desulfurized
high sulfur fuel; and
2. Flue gas desulfurization (FGD).
The Clean Air Act Amendments of 1977
require the "best technological system of
continuous emission reduction" and the
"achievement of a percentage in..emissions..
from the use of fuels which are not subject
to treatment prior to combustion."(2)
In other words, low sulfur fuels and inter-
mittent controls will not be acceptable as
a means of compliance.
Non-regenerable processes account for
approximately 90 percent of those FGD
systems which are either operational or
under construction. Lime and limestone
systems are the most prevalent, with 75
percent of the scrubbers projected to be
operational by 1980 utilizing these
processes(3). The sodium-lime double-alkali
*"Compilation of Data Base for the Develop-
ment of Standards/Regulations Relating to
Land Disposal of Flue Gas Cleaning Sludges."
EPA Contract No. 68-03-2352; July 1975.
systems thus far have been installed pri-
marily on industrial fossil fuel-fired
boilers. Scrubber systems regenerating
sulfur or sulfuric acid have not yet gained
very much acceptance in the United States.
Only five full-scale projects are still in
the active planning stages(4).
Table 1 lists the total committed FGD
capacity through 1985 and the associated
annual sludge generation.
Description of Industrial FGD Users
Many non-utility industrial scale
fossil fuel-fired steam generators are also
regulated by the new source standards.
Most smaller boiler units are equipped for
firing a variety of fuels and have achieved
compliance by use of low sulfur fuels. A
total of eight non-regenerable industrial
scale FGD systems were on-line in June 1977
in the United States. Non-regenerable
double-alkali FGD is the prevalent system
for industrial steam generators.
Rationale for Subcategorization
The purpose of Subcategorization with-
in an industry category is to take into
account basic differences in waste genera-
tion, treatment technology, and economics
between plants for regulatory purposes.
When discussing Subcategorization for FGD
sludge disposal regulation purposes, some
of the same factors that entered into sub-
categorization during effluent guidelines
development should be considered, but with
several important differences: (1) FGD
sludge is disposed to land, and (2) utility
economics are regulated by the government.
Economic Subcategorization has been neces-
sary in some other industries to prevent
economic inequities and plant closures.
Because costs for power generation are
passed directly to the consumer, it would
be a disservice to the power utility consumer
to create an arbitrary Subcategorization on
the basis of "ability to pay."
The following variables were consid-
ered as rationale for Subcategorization:
1. Size and age of plant;
75
-------
TABLE 1. CUMULATIVE COMMITTED NON-REGENERABLE FGD
CAPABILITY, 1975-1985 (MW)*++
Lime Limestone Lime/ Double Total Total Sludge
Scrubbing Scrubbing Limestone Alkali Capacity Generation**
1975 475 (2) 1,954 (9) 20 (2) 32 (1) 2,481 (14) 1,225
1980 9,445 (23) 12,695 (35) 270 (3) 652 (3) 23,062 (64) 11,200
o>
1985 11,515 (28) 13,445 (37) 797 (4) + 26,409 (72) 13,300
*Numbers in parenthesis indicate number of scrubber units.
+None committed beyond 1980.
**Thousand metric tons of dry solids/year.
++Reference 5.
-------
2. Type of plant or fuel;
3. Type of air pollution control
and FGD equipment;
4. Untreated (raw) waste
characteristics;
5. Disposal site characteristics.
The effects of plant operating vari-
ables upon the characteristics of FGD
sludge are complex and difficult to predict.
In addition, plants may often change fuel
and/or operational procedures, thereby
chanqing the FGD sludge characteristics.
Therefore, it is deemed impractical to
regulate FGD sludge disposal on the basis
of type of plant or in-plant process.
Sludge characteristics also vary substan-
tially within the same plant and alone,
do not provide a practical means of sub-
categcrizing the industry.
Disposal site characteristics vary
substantially between plants and result in
associated differences in potential environ-
mental impact. These characteristics set
a need for differences in regulation
between plants, and are therefore a logical
means of subcategorization. Important
site characteristics in this respect are:
• Present and future land values
and possible uses;
• Groundwater resources (quantity
and quality);
• Disposal site hydrology and mass
transport potential (permeabili-
ties, hydraulic gradients,
impurity attenuation, flooding
potential, etc.);
• Disposal site meteorology
(evaporation and precipitation
rates, etc.);
t Local ecology;
• Geographic location; and
• Acceptable land degradation.
SULFUR OXIDE CONTROL TECHNOLOGY
Utilities and other industries burn-
ing fossil fuels can select from three
general categories of control systems to
achieve compliance with standards:
• Control systems that do not
generate a sludge
As of June. 1977, 49 FGD systems were
in commercial operation in the U.S. Of
these, 41 were installed on utility plants
and 8 on industrial fossil fuel-fired steam
generators. Table 2 differentiates the
number and capacity of these operational
systems by FGD process. This table shows
that the lime/limestone systems account for
34 utility scale systems and for more than
90 percent of the total scrubbing capacity.
Double alkali systems currently dominate
the industrial scale FGD market, although
several of these smaller systems were
conceived as pilot or demonstration
facilities. Lime/limestone systems are
expected to dominate the industry through
1985. An estimated 154 FGD systems are
currently operational or at some phase of
implementation or selection(S).
FGD Systems with Sludge Generation
Primarily nonregenerable FGD systems
are employed by industries and utilities.
Available options for nonregenerable FGD
systems are:
t Lime scrubbing (tail end)
• Limestone scrubbing (tail end)
• Double alkali scrubbing (sodium,
ammonia)
• Sodium carbonate scrubbing
These systems together comprise the de facto
"Best Available Control Technology" under
current New Source Performance Standards.
The first two use lime and limestone,
respectively as the scrubber sorbent, while
the fourth uses sodium carbonate. In all
three systems, the spent sorbent is not
regenerated. In the double alkali system,
sodium salts used as the sorbent are regen-
erated from the scrubber effluent through
lime addition; the calcium-based precipi-
tates are then disposed of. Sludge genera-
tion rates for lime, limestone, and double
alkali systems are on the order of 0.06 to
0.12 tons of dry solids/MWh. The principal
reasons for lime/limestone predominance is
the low sorbent "cost and greater availa-
bility.
FGD systems with sludge generation FGD Systems Hhich Do Not Generate A Sludge
• FGD systems that produce a
marketable product
Regenerate FGD systems can have as
their by-products gypsum, hydrogen sulfide,
77
-------
TABLE 2. SUMMARY OF OPERATIONAL FLUE GAS
DESULFURIZATION SYSTEMS"1"
FGD Process
Uti1i ty Seale
Lime scrubbing
Lime
Lime/alkaline ash
Limestone scrubbing
L imestone
Limestone/alkaline ash
WeiIman-Lord scrubbing
Dilute acid scrubbing
Sodium carbonate scrubbing
Magnesium oxide scrubbing
Industrial Scale
Double alkali scrubbing
Sodium carbonate scrubbing
Number in^
Opera tion
Total MM
Capacity
12
2
18
2
1
2
3
1
7
1
3,897
720
4,267
1 ,390
115
23
375
120
243
150
* For fossil fuel-fired steam generators as of June 1977.
+ Reference 4
sulfur dioxide, sulfur and sulfuric acid.
Hydrogen sulfide and sulfur dioxide are
gases with relatively little commercial
value, while gypsum, elemental sulfur, and
sulfuric acid are readily marketable.
Certain FGD systems that remove and concen-
trate sulfur dioxide can either oxidize to
sulfuric acid or reduce to elemental sulfur.
Other systems directly produce sulfur or
sulfuric acid. Currently, the most promis-
ing systems are:
• Dilute acid scrubbing
• WeiIman-Lord scrubbing
• Magnesium oxide scrubbing
§ Citrate scrubbing
Reqenerable FGD technology producing
elemental sulfur or sulfuric acid has a
proven commercial market in certain areas.
Marketing and utilization of FGD sludges or
gypsum is currently under investigation by
the EPA and other public and private
organizations. Current market conditions,
couplad with potentially higher FGD costs,
have limited the commercial acceptance of
these systems to date.
Lime/Limestone Sludge Utilization
In order to avoid the problems
associated with FGD sludge disposal, re-
search has been in progress in the area of
nonregenerable sludge utilization. Tech-
nical developments for FGD sludge reuse
have largely followed precedents estab-
lished for the utilization of fly ash.
78
-------
The current market for nonregsnerable
FGD sludge is very limited, and disposal is
generally more cost-effective than recovery.
Potential markets for FGD sludge include
use as gypsum, structural landfill, synthe-
tic aggregate, mine void backfill, soil
amendment, and acid mine drainage control
or treatment. Research has shown these
alternatives to be technically feasible
under proper conditions of treatment and
control.
Other Non-Scrubbing Desulfurization Options
The.high cost of wet scrubbing
techniques has prompted many public and
private institutions to seek alternate
means for reducing S02 emissions. Some
of these techniques are:
• Fuel switching from high sulfur
coal to natural gas, oil, or
low sulfur coal
0 Coal cleaning
• Fluidized bed combustion
• Coal gasification
Many of the techniques are still in the
developmental stages, but accelerated
research may, in a few years, promote them
to commercially available status. •
Low sulfur fuel use, although the
predominant means of compliance with
emission standards, will no longer be
acceptable for this purpose in many
instances. New Source Performance Standards
prohibit the use of low sulfur coal for
compliance. The Clean Air Act may also
require some existing sources to use certain
native coals which are higher in sulfur
content than those obtainable elsewhere.
Coal cleaning has been practiced for
many years and will be acceptable as a
means of meeting Clean Air Act requirements.
The predominant coal cleaning technologies
remove pyritic ( organic) sulfur by physi-
cal means, which can amount to a 30 to
70 percent reduction in coal sulfur content.
Chemical means of organic sulfur removal
have been developed and should see rapid
commercial acceptance under the new law.
Fluidized bed combustion and coal
gasification are both promising technologies,
but should see limited full-scale appli-
cation in the next decade.
CHEMICAL AND PHYSICAL PROPERTIES OF
FGD SLUDGE
Many factors affect the chemical and
physical characteristics of FGD sludge.
These factors include:
• Characteristics of the fuel burned;
• Combustion equipment and
operating parameters;
• Particulate collection mode;
• Scrubber operating parameters
(.liquid/gas ratio, slurry reten-
tion time, recirculation, etc.);
t FGD reagent and input water quality.
The impact of these factors could be
controlled with modification in plant
operation. This might be construed to mean
that sludge characteristics can be modified
to suit the disposal method and specific
site characteristics. However, the power
plant and associated gas cleaning equipment
costs are generally much greater than sludge
disposal costs. Overall system cost mini-
mization will therefore generally require
designs which minimize the FGD system cost,
rather than optimize the sludge character-
istics. Sludge characteristics can there-
fore be predicted, to some extent, using
plant operating data. These chemical and
physical characteristics are important to
an effective disposal operation and have
been the subject of most disposal-related
research efforts to date. An extensive
data base is becoming available.
Chemical Characteristics
The by-products of non-regensrable
FGD systems are typically composed of four
major constituents: calcium sulfate hemi-
hydrate and/or dihydrate (due to mixed
crystallization), calcium sulfite hemihyd-
rate, unreacted sorbent, and fly ash. The
relative amounts of these constituents are
determined by the various scrubber and
plant operating parameters discussed in the
previous section. Table 3 shows the wide
variation in sludge solid phase composition
encountered in different FGD systems.
The solid phase of FGD sludge can
also contain a variety of trace metals.
These metals can come from several sources,
including fly ash, sorbent, makeup water or
vapors in the flue gas itself. The metals
which are contained in fly ash generally
79
-------
TABLE 3. COMPOSITION OF SLUDGE FROM OPERATING S02 SCRUBBERS
Facil i ty
Lawrence
Hawthorn 3
Hawthorn 4
Will County 1
Stock Island
La Cygne
Cholla
Paddy's Run 6
Mohave 2
Shawnee 1
Shawnee 2
Phil lips
Parma
Scholz 1A
Utah
Col strip
Scrubber
Sorbent
Limestone
Limestone
Limestone
Limestone
Limestone
Limestone
Limestone
Lime
Limestone
Limestone
Lime
Lime
Dual Alkali
Dual Alkali
Dual Alkali
Lime/Al kal ine
Sludge Composi
CaSO,'*2H00
o i
10
20
17
50
20
40
15
94
2
19-23
50
13
14
65-90
0.2
Ash 0-5
tion (dry
CaS04'2H20
40
25
23
15
5
15
20
2
95
15-32
6
19
72
5-25
82
5-20
basi s) ,
5
5
15
20
74
30-
0
0
0
4-14
3
0.2
8
2-10
11
0
wt percent
Fly Ash Comment
45
50
45
15
1 Oil fired
15
65 14% CaS203-6H20
4
3
20-43
41
60 9.8% CaS30]0
7
0
9
40-70 5-30% MqS04
*Reference 7.
-------
are often dissolved in the waste slurry and
enter the liquid phase; from there they
may precipitate either as pure compounds or
with the sulfate/sulfite crystals.
The liquid phase of FGD sludges is
important due to its potential as leachate.
The sludge liquors typically contain ions
of sulfate, sulfite, chloride, calcium,
magnesium, and various trace chemical
species. The total dissolved solids (TDS)
concentration in FGD sludges is a function
of their equilibrium levels in the scrubber
system, with TDS concentration in excess
of 20,000 ppm being quite common in closed-
loop operations. Table 4 lists the typical
range of values for various species concen-
trations in the liquid phase.
TABLE 4.* CONCENTRATION OF CONSTITUENTS IN SCRUBBER LIQUORS
Ramjo of Constituent Concentrations at
Potential Discharge Points
Constituents
Aluminum
Antimony
Arsenic
Beryllium
Boron
Cadmium
Calcium
Chromium (total)
Cobalt
Copper
Iron
Lead
Magnesium
Manganese
Mercury
Molybdenum
Nickel
Potassium
Selenium
Silicon
Mg/1 (Except pH)
0.03 -
0.09 -
<0.004 -
<0.002 -
8.0 -
0.004 -
520 -3
0.01 -
0.10 -
<0.002 -
0.02 -
0.01 -
3.0 -2
0.09 -
0.0004-
0.91 -
0.05 -
5.9 -
< 0.001 -
0.2 -
0.3
2.3
0.3
0.14
46
0.11
,000
0.5
0.7
0.2
8.1
0.4
,750
2.5
0.07
6.3
1.5
32
2.2
3.3
M
10-5.95_10-4.95
10-6.!3_1()-4.72
-------
TABLE 4* (Continued)
Range of Constituent Concentrations at
Potential Discharge Points
Constituents
Silver
Sodium
Tin
Vanadium
Zinc
Carbonate
Chloride
Fluoride
Sulfite
Sulfate
Phosphate
pH
Ionic strength
Mg/1 (Except pH)
0.005 - 0.6
14 -2,400
3.1 - 3.5
<0.001 - 0.67
0.01 - 0.35
41 - 150
(as CaC03)
420 -4,800
0.07 - 10
0.8 -3,500
720 -10,000
0.03 - 0.41
3.04 - 10.7
M
10-7.33_10-5.25
10-3.21_10-0.98
1Q-4.58.10-4.53
<10-7.71_10-4.88
10-6.82_10-5.27
10-3.39_1Q-2.82
10-1.93_10-0.87
10-5.43_]0-3.28
10-5.00_1(J-1.36
10-2.12.10-0.98
10-6.50_10-5.36
10-3.04_1(J-10.7
. 0.05 - 0.80
* Reference 8.
Physical Characteristics
The physical characteristics of FGD
sludge solids are important to treatment
process operations, disposal methods, and
potential environmental effects. Of
special interest are crystal morphology,
bulk density, viscosity, and compressibi-
lity. Environmental considerations include
permeability, settling characteristics, and
load-bearing strength. Table summarizes
the importance of various sludge properties
to FGD sludge disposal operations.
Crystal Morphology--
Crystal morphology has perhaps the
strongest Influence on FGD sludge physical
characteristics. Calcium sulfite hemi-
hydrate is the predominant solid phase in
most sludges. Sulfite crystals usually
form in single flat plates, although
spherical aggregated forms are also observed.
In contrast, calcium sulfate dihydrate
solidifies in blocky crystals. A limited
amount of sulfate is soluble in the sulfite
crystals, thereby making an unsaturated mode
of operation possible. FGD sludge crystals
are typically between 1 and 200 microns
in diameter and classify as silt and silt
loam under the U.S.D.A. system.
Bulk Density—
The bulk density of FGD sludges is
another important physical characteristic.
Bulk density determines such disposal-
related parameters as compressibility, land-
fill volume requirements, permeability, and
to some extent, compaction strength.
82
-------
Bulk density of FGD sludge varies with
sludge type (sulfate/sulfite), which in turn
is related to the types of scrubber and coal.
Mechanical dewatering of sludge does not
substantially increase bulk density beyond
that of settled sludge. Only high sulfate
sludges with fly ash can exceed the optimal
moisture content; this low dewatering
effectiveness is typical of most FGD sludges.
Permeability—
Permeabillty, another Important physi-
cal parameter, is related to the rate of
flow of liquids through the material under
a hydraulic driving force. This parameter
enters Into the calculation of the mass
transport rate of contaminants from the
disposal site to the groundwater.
FGD sludge permeability is dependent
both upon size distribution and particle
shape. Settled sulfite sludge is generally
less permeable than sulfate sludge due to
the irregular shape of the sulfite crystal
aggregates. Settled and drained FGD sludges
exhibit permeabilities of 10~3 to 10-4
cm/sec; light compaction will decrease this
to 10-5 cm/sec, and even lower permeabilities
have been attained on the laboratory scale
using greater compaction and more extensive
dewatering.
The presence of fly ash will decrease
sludge permeability. The smaller fly ash
particles fill in the sludge interstices,
thereby blocking the movement of fluid
through the media. Permeabilities of about
10-6 cm/sec have been achieved for a sulfate
sludge mixed with fly ash; this is thought
to be the minimum permeability achievable
with untreated sludge.
Compressibility—
The compressibility, or compactibi-
lity, of FGD sludge is an important para-
meter when planning for disposal site
operation and reclamation. The amount of
compaction which can be achieved is depen-
dent upon moisture content, crystal struc-
ture, and compaction force.
Studies of FGD sludge compressibility
are documented in literature; one found
that a sludge with equal quantities of
sulfate and sulfite could be compacted to
1.3 g/cm3 at 77 percent solids, but only
1.2 g/cm3 at 72 percent solids. Similarly,
a'study of double alkali sludge achieved
1.15 g/cm3 at 75 percent solids. Another
study dealing with the effects of sulfate
versus sulfite on compressibility found
that higher sulfate concentrations in the
solid phase reduced compressibility
significantly(lO). Laboratory experiments
by the Aerospace Corporation have achieved
bulk densities in excess of 1.44 g/cm^;
this represents a volume reduction of 25
percent for pure sulfite sludge, but only
7 percent for sulfate sludgeUU.
Load-Bearing Strength—
Load-bearing strength is a partial
function of compressibility, but also
depends heavily upon the moisture content
of the sludge. Sulfate sludge compaction
strength has been shown to increase rapidly
from 2.0 x 105 to 2.0 x 10$ dynes/cm2
below 35 percent moisture content. Since
this moisture content can be achieved by
mechanical dewatering, sulfate sludge can
be made to support men and equipment if
compacted. Japanese experience with high
sulfate sludge (gypsum) production also
indicates that this material will readily
support grading equipment 12).
The load-bearing strength of sulfite
sludge samples has been shown to increase
gradually with decrease in moisture content.
At 30 percent moisture content, the strength
begins to increase markedly corresponding to
the peak viscosity. The load-bearing capa-
city at this point was 2.1 to 2.4 x 10&
dynes/cm2. Even at low moisture content,
however, the sludge may liquify if jarred
or vibrated.
Despite the variation in load-bearing
strength between sludges, certain general-
ities can be made. Light compaction will
result in only a small improvement of
strength. Settlement is unavoidable even
after heavy compaction, although landfllled
sludge settlement may be less than that of
many common soils under these circumstances.
Unconfined compressive strength is quite
low, typically in the range of 7 to 14 x
dynes/cm2.
Settling Characteristics—
The settling and dewatering charac-
teristics of FGD sludge have been observed
to be related to sulfite content. The flat
plate structure of sulfite crystals results
in water retention and a slower settling
rate than the blocky sulfate crystals.
A study of sulfate/sulfite settling
and the associated design considerations has
83
-------
concluded that the percent sulfite oxidation
to sulfate has little effect on the required
thickener diameters, provided the higher
final moisture content can be tolerated when
preparing for vacuum filtration(13). The
primary difference in sulfite sludge set-
tling rates is in the crystal plate size.
Smaller plates entrap more liquor during
the compression stage due to a higher
incidence of inner-particle touching;
thicker sulfite slurries take longer to
reach the compression stage.
TREATMENT AND DISPOSAL TECHNOLOGY
Treatment of FGD Sludges
Three forms of FGD sludge treatment
are considered state-of-the-art technology:
• Dewatering;
• Stabilization (fixation); and
• Forced oxidation.
Each process is designed to reduce disposal
costs and/or reduce the potential environ-
mental impact of disposal to land.
Dewatering--
Dewatering reduces the volume of
sludge, removes and recirculates water, and
improves sludge handling characteristics.
Dewatering processes with demonstrated or
potential applicability to scrubber
sludges are:
Clarifiers
Centrifuges
Vacuum filters
Solar evaporation ponds
Bed dryers
Thermal dryers
The first four processes listed above have
been applied to scrubber sludges, and
operating experience is documented. The
other processes listed have not been applied
to scrubber sludges, but have been used
successfully for other types of industrial
and municipal sewage sludges.
Clarifiers are standard equipment on
most FGD systems. The underflow from these
units can contain 15 to 35 percent solids.
Centrifuges and vacuum filters are gaining
popularity for FGD sludge dewatering.
Dewatering of 50 to 80 percent solids has
been documented using these devices; actual
dewatering efficiency depends upon the
chemical characteristics of the sludge and
the presence of fly ash. Solar evaporation
ponds serve both disposal and treatment
functions. Disposal pond supernatant is
often recirculated, and the settled sludge
ranges from 40 to 65 percent solids in
most applications.
Stabilization--
Stabilization techniques employed
thus far have incorporated such power plant
wastes as fly ash and bottom ash, thereby
consolidating several waste disposal
activities into one activity. Stabilization
can reduce the volume required for disposal
through the mechanical dewatering operations
included in some systems. Since pozzblantic
stabilization improves the load-bearing and
handling characteristics of sludge, such
materials can be used as a fill material
for road base, berms, embankments, and
damsu4). in addition, stabilized sludge
often has less pollution potential than
raw sludge. This is because the stabili-
zation process normally restricts the move-
ment of water through the sludge and/or
chemically binds the contaminants so they
are less readily dissolved by permeating
water.
As of December 1977, at least 15 power
plants employed some form of sludge stabili-
zation. Of these, 7 employ commercial
stabilization processes and/or services
The remainder typically use fly ash, bottom
ash, and/or lime as additives. The product
in these cases, is a physically stable '
material which can be transported like soil
and disposed of as landfill material.
A variety of commercial stabilization
processes and services are available for
FGD sludge treatment. Several of these
services use a proprietary additive or
process. IU Conversion Systems (IUCS) offers
a patented treatment procedure using the
lime-fly ash reaction to produce a struc-
turally stable and low-permeability material
Dravo Lime Company offers a stabilization
service utilizing a proprietary additive,
CalciloxR, which after a curing of 2 to 4
weeks, will solidify either in a landfill
or under water. Other proven stabilization
processes are available commercially for
hazardous waste treatment, but high proces-
sing and additive costs have precluded their
use for FGD sludge treatment.
Each type of stabilization process
results in a material with a different set
of? physical and chemical properties.
Research has been under way for several
years to evaluate existing stabilization
84
-------
processes and to develop new processes.
Thus far, the most extensive studies of
stabilized sludge and its properties have
been completed by the Aerospace Corporation
for EPA(15)S the Army Corps of Engineers
Waterways Experiment Station for EPAU6),
and IU Conversion Systems(17,18).
All of the above research efforts have
concluded that commercial stabilization
improves the physical properties of FGD
sludge. Sludge permeability is typically
reduced by 1 to 2 orders of magnitude,
depending on the physical nature of the
treated sludge. Load-bearing strength is
also improved substantially, particularly
when treating sulfite sludges. Sulfate
sludges (70 to 80 percent solids) have been
shown, in some instances, to possess
similar handling and load-bearing properties
to stabilized sludge, but this is more the
exception than the rule for sludge, as
generated by FGD systems.
Stabilization, as well as providing
a structurally improved product, may also
reduce the rate of mass transport of
soluble contaminants from the FGD sludge.
This can be accomplished through a reduction
in permeability/porosity and through chemi-
cal bonding of contaminants within the
stabilized matrix.
The above-mentioned studies have
shown that stabilization will result in a
significant reduction in pollutant mobility.
Aerospace^9) reported a 50 percent reduc-
tion in TDS loading between the first pore
volume displacements in stabilized FGD
sludge over that encountered for unstabi-
lized sludge; more substantial reductions
were experienced in subsequent pore volume
displacements. While each study has used
a different leaching test, the combined
results seem to indicate that the principal
means of pollutant mobilization from
stabilized sludge is through the initial
surface washing. Further leaching of
contaminants from the sludge mass is limited
by: 0) tne l°w permeability and low
hydraulic driving force (unsaturated) and
(2) some apparent level of complexation or
chemical binding within the sludge matrix
due to stabilization.
Disposal of FGD Sludges--
The general categories of available
disposal options include:
• Ponding (lined or unlined)
0 Landfill
0 Disposal to oceans
0 Disposal to mines (backfill)
The use of unlined disposal ponds is
typically the least expensive method of
sludge disposal. This technique, however,
has been subject to criticism, primarily
due to the potential for sludge contami-
nants to enter the groundwater. Nonethe-
less, a majority of the FGD sludge disposal
systems in operation use unlined ponding
for either intermediate clarifying or
ultimate disposal. The use of unlined ponds
for sludge disposal is expected to continue
pending regulation of the practice.
"Unlined" refers to the use of compacted
indigenous soils as the liner material
in situations where such soils are deemed
adequate by local regulatory agencies.
Pond liners have been required in
several instances. Typical liners are
composed of imported natural clay.
Artificial or synthetic liner materials
have not been used for full-scale FGD
sludge disposal ponds, but considerable
research and cost analysis have been per-
formed for this alternative.
The use of landfills for FGD sludge
disposal is also prevalent. Backfilling of
impoundments is the predominant mode.
Landfill ing requires that the sludge be
either physically stabilized or thoroughly
dewatered. The principal operational
difference between sludge landfills and
ponds is that sludge liquors are recycled
at the plant instead of from the pond; this
results in a significant land saving and
allows for access to remote disposal sites
by truck.
Ocean disposal of FGD sludges is an
option that is perhaps available to non-
regenerable FGD system users with economic
access to the ocean. New ocean disposal
initiatives are discouraged by regulatory
agencies.
The feasibility of FGD sludge disposal
to mines is of particular interest to the
many mine-mouth, coal-fired power plants.
Sludge disposal to mines has the advantages
of: (1) minimal transportation requirements,
when a mine is nearby; (2) minimal incremen-
tal land use; (3) possible mine reclamation;
and (4) serving as a pH amendment for mine
tailings. The principal concerns are
possible interruption of mining operations,
contamination of ground .and surface waters,
85
-------
and fugitive emissions and runoff. The use
of FGD sludge as mine backfill is currently
being studied by several state and federal
agencies(20).
ENVIRONMENTAL IMPACT OF FGD SLUDGE DISPOSAL
Those aspects of FGD sludge disposal
which contribute to potential environmental
impacts include:
• Disposal site characteristics
• Health effects of sludge components
• Safety
• Other ecological considerations
The specific impacts relating to each aspect
are discussed below.
Disposal Site Characteristics
The disposal site characteristics
which Influence the environmental impacts
of FGD waste disposal include:
• Present and projected land use
• Topography
t Hydrology and geology
• Meterology
These site characteristics, when combined
with the sludge characteristics, economics,
and other disposal variables, will determine
the optimal solution to the disposal
problem. The variety of these relationships
indicates that no general solution is
appropriate to all areas.
Present and Projected Land Use--
The present and potential future land
use of a proposed disposal site will
influence the disposal technique to be
utilized. A primary distinction can be
made between urban and rural land uses.
Property values are typically higher in
urban locations than in rural areas. There-
fore, techniques which utilize less land
and/or generate a fill material of adequate
bearing strength would be more desirable
in urban areas.
Potential land use will be at least
partially determined by present methods of
FGD sludge disposal. For example, depend-
ing upon sludge characteristics, pond
disposal of untreated sludge may result in
a site with virtually no bearing strength.
In urban locations, this lack of mechanical
stability would prevent future use of the
site for anything more than the lightest
duty applications.
Topography--
Topography of the disposal site and
adjacent areas influences the potential
environmental impact of the disposal
operations in four ways:
o Rainwater runoff may flow
naturally into the disposal area,
increasing the hydraulic head;
o Topography may allow sludge
released by retention structure
failure flow to nearby surface
waters;
o The relative elevations of the
disposal pond surface, pond
bottom, and groundwater have
a strong influence upon the
mass transport of FGD contami-
nants to the groundwater; and
o Topography can control the visi-
bility of the disposal site
operation.
Hydrology and Geology--
The impact of the FGD disposal site
upon the groundwater resources below the
site is of major concern. Of particular
interest is the mass transport of contami-
nants from the FGD sludge to the ground-
water. This process is a function of:
1. Quality and quantity of leachate
generated by the in-place sludge;
2. The rate at which leachate
permeates through the retention
structure or a pond liner;
3. The rate at which individual
contaminants in the leachate
which escapes the retention
structure travel through the
underlying soil; and
4. The effect of the contaminants
which reach the groundwater
aquifer upon the quality of
the groundwater.
Groundwater quality can range from
highly desirable drinking water to water
that is nonpotable. These variations in
existing groundwater quality can be of
natural or artificial origin. In many areas
of the southwest, it is common to find
shallow, highly saline water due to concen-
tration by evaporation or a preponderance
of exchangeable cations. The water quality
in many of these areas may be worse than the
86
-------
quality of the leachate generated in a
disposal site.
Where potentially useful water under-
lies a disposal area, it is necessary to
consider the mass transport of contaminants
from the site to the water table, in order
to evaluate the potential environmental
significance of any given disposal practice.
The term "mass transport" as 1t 1s used
here, refers to the movement of contaminants
from a disposal site to the surrounding
environment. The rate of mass transport is
perhaps the most Important factor in
estimating groundwater pollution potential.
Different disposal techniques or site
variables affect the mass transport by
changing the controlling parameters. For
example, clay liners 1n the bottom of ponds
will reduce the flow rate through the
bottom of the pond by decreasing the permea-
bility; while locating a pond in a high
water table, low horizontal hydraulic
gradient area will decrease the mass
transport by decreasing the hydraulic
gradient. This will also decrease the flow
rate. Locating the disposal site above
clay and/or organic soils which have high
cation exchange capacity will increase the
retention factor. This will decrease the
mass transport of some contaminants by
slowing cation movement.
Meteorology-
Local meteorology can interact with
the disposal technique being employed
through two primary mechanisms. The first
of these relates to the characteristic
evaporation/precipitation ratio at the site,
and the second to the effects of wind
erosion on the surface of the site 1f the
site is dried.
Evaporation/Precipitation--In areas with
high evaporation rates relative to preci-
pitation rates, significant amounts of water
can be evaporated from a pond. The net
evaporation rate can vary from a fraction of
a meter per year to up to 2.5 meters
annually. A high evaporation rate allows
certain types of solutions to disposal
problems that are not feasible 1n low
evaporation rate areas. Under such climatic
conditions, contaminated leachate can be
disposed of through evaporation.
Wind Erosion—The second possible effect
of local meteorological conditions is
surface wind erosion. If the disposal site
dries out and is not supporting vegetative
cover, wind can erode the surface material
and remove it from the site. The importance
of this phenomenon will vary from area
to area.
Covering a site with soil to estab-
lish a vegetative cover will help stabilize
the site, as is done with dry ash disposal.
Observations of FGD sludge disposal sites
in the southwestern United States indicate
that the surface can also be stabilized by
bottom ash. When both fly ash and bottom
ash are disposed of at the same site, the
wind erodes the loose fly ash from between
the mixture of fly and bottom ash, leaving
a surface covered only by bottom ash.
Bottom ash particles tend to be too large
for wind erosion. Once this bottom ash
surface is formed, the site is stabilized
against further wind erosion. The proper
selection of cover materials is certainly
preferable to sprinklers and other dust
control methods due to limited maintenance.
Health Effects
The dissolved salts, trace metals,
and other contaminants contained in FGD
sludges are all found in man's natural
environment. Through evolution man has
acquired a tolerance for these contaminants
and requires many of them in his diet for
good health. A typical biological response
curve would show an increasingly beneficial
effect with increasing contaminant concen-
trations to a certain level. Beyond this
level there is a tolerance region beyond
which benefits decrease, Injurious effects
begin, and finally, a lethal concentration
is reached. The goal of FGD sludge manage-
ment is not to reduce contaminant levels to
zero, but rather to ensure that the concen-
trations of the contaminants which could
eventually return to man and other biota
are not at harmful levels.
Contaminant Pathways—
The potential health problems posed by
sludge disposal can be approached by exam-
ining contaminant pathways to man. Adding
an FGD system to a power plant effectively
removes trace contaminants from the atmos-
pheric pathway to man and places the
contaminant* instead into useful water.
This converts a regional atmospheric pollu-
tion problem into a potential local water
pollution problem.
Groundwater--The major potential pollution
problem in FGD sludge disposal is the trans-
87
-------
port of leachate contaminants from the
disposal site into local potable groundwater
supplies. Predicting the potential mass
transport of specific contaminants to man
via this pathway is complex and involves
the following major variables:
• The flow rate of the leachate from
the disposal site into the aquifer,
which is dependent upon such
factors as permeability of the
soils and liner (if any) underlying
the disposal site, and the hydro-
static head force upon the leachate;
• The concentration of the specific
contaminants contained in the
leachate generated by the disposal
site, and the changes occurring in
these concentrations due to attenua-
tion effects by the soil during
passage of the leachate to the
aquifer;
• The capability of the aquifer
through actual replenishment to
dilute the leachate reaching it,
i.e., the relative volumetric
rate of: (1) the leachate reach-
ing the aquifer, (2) the replen-
ishment by natural sources such as
horizontal groundwater migration,
(3) the rate of water supply with-
drawal , and finally (4) the extent
of the mixing which occurs in the
aquifer between leachate and
natural source water; and
• The contaminant concentrations of
the natural source water in the
aquifer and their relation to
concentrations of similar contami-
nants in the leachate reaching
the aquifer.
ihe above-listed variables are often inter-
related, complex, and difficult to measure
accurately. Considering, however, the
large size and costs involved in FGD sludge
treatment and disposal, the effort to
evaluate these factors should be cost-
effective in designing the best sludge
management system for a particular site.
Surface waters—A second pathway by which
contaminants contained in FGD sludge may
reach man is by way of surface waters.
Considering the initial concentrations of
the contaminants in question and the dilu-
tion capabilities of the surface waters
adjacent to many power plants, the levels
of contaminants reaching man from sludge
supernatant discharge will be site-specific.
Health Significance of Sludge Contaminants—
The health effects of metals and their
compounds are often difficult to quantify.
Of the heavy metals found in FGD sludge,
five elements (arsenic, lead, mercury,
fluoride, and selenium) represent known
hazards to human health based on existing
evidence. Lead, mercury, and cadmium are
particularly insidious because they can
be retained in the body for relatively long
periods of time, thus functions as cumula-
tive poisons^l). Table 5 provides a
comprehensive summary of the presence of
metals in the environment, their toxicity
to humans, and their half life in the body.
In the liquid phase, trace element
concentrations are important in determining
the potential heal hazards of leaching or
surface water discharge of FGD sludges.
A summary of concentration ranges common to
various sludges and effluents was given in
Table 4. The range of data shown in the
table for sludge leachates/elutriates
represents a wide variety of power plants
and scrubber systems.
The following conclusions can be
drawn regarding the health significance of
leachate and surface water discharges from
disposal systems:
0 All studies agree that the total
dissolved solids in the sludge
liquors exceed drinking water
standards;
0 All studies indicate that the
selenium concentration commonly
exceeds drinking water standards;
0 Selected data indicate that, in
some instances, fluoride concen-
trations may be high;
0 All studies indicate that arsenic,
lead, and mercury (among others)
can exceed drinking water standards
in some instances, although
typically below standards and
readily attenuated by most soils.
Safety considerations associated with
the disposal of FGD sludges can be directed
toward three groups: (a) workers employed
at the site; (b) the general public; and
88
-------
TABLE 5. METALS IN THE ENVIRONMENT AND THEIR TOXICITY*
Metal
Antimony
Arsenic
Barium
Beryl 1 i urn
Cadmi urn
Chromi urn
Copper
I ron
Lead
Manganese
Mercury
Nickel
Selenium
Silver
Tin
Zi nc
* Ref . 21.
Average daily intake
(ng/d )
Food and Water Air
100 1.7
400-900
735 30
Oral dose
produci ng
toxicity (mg}
100
5-50
200
12 0.04
160 7.4
245 1.1
1,325 11.4
15,000 84
300 46
4,400 28.8
25
3
200
50-250
--
--
not known
--
Fatal
dose
( i n g e s t i o n )
100-200 mg
120 mg
1 9
not known
not known
5 g
10 gb
5-10 gc
0.5 g
--
20 mg - 1 gd
600 2.36
62
60-80
7,300 0.6
14,500 16.8
5
60
2,000
--
--
2 gf
--
10 g9
Total
body content
(mg}
7.9
15-20
22
0.3
50
1 .8
72
4,200
120
12
--
10
14.6
1
17
2,300
Whole body
half-life
(days)
38
280
65
180
200
616
80
800
1 ,460
17
70e
667
11
5
35
933
b. Copper sulfate
c. A two-yr
d. Mercuric
old child
salts
e. Methyl mercury
f . Silver n-
trate
g. Zinc/sulfate
-------
(c) livestock and wild animals. These
safety considerations relate to the physical
rather than the chemical properties of
the sludge.
The thixotropic nature of certain
untreated (typically "sulfite") FGD sludges
poses an unusual safety hazard. A pond
filled with such material may develop an
apparently solid surface crust which may
liquify on impact (such as a person jumping
or a vehicle braking on the surface).
Ecological Considerations
FGD sludge disposal operations have
potential for adverse impact on the surround-
ing ecosystem. Of primary concern are:
• The aquatic environment;
• The terrestrial environment;
• Irrigated agriculture and
livestock.
Aquatic Environment--
Sludge disposal may affect the aquatic
environment through either accidental
release to a waterway or through intention-
al discharge to surface or groundwater.
Surface and groundwater discharge is
generally continuous; accidentla discharge
is a random, possibly catastrophic event.
Since large volumes of cooling water
are required to operate a fossil fuel-
fired powar plant, most plants are located
near a large body of water. Installation
of FGD units on these power plants entails
a large number of pipes to carry the
slurries around the scrubber and associated
materials handling systems. Pipelines may
also carry the sludge to disposal areas.
Depending upon system design, a pipeline
break could result in the accidental dis-
charge of the slurry into a watercourse.
Dike failure.at the disposal pond may
release several years' supply of sludge
into a waterway in only a few hours.
It is estimated that the COD in one
300 hectare-meter sulfite sludge pond would
be sufficient to remove all the dissolved
oxygen from virtually any lake or several
hundred miles of a large river. However,
in a lake-type environment, rapid settling
of the sludge would limit the impact to
that of the immediate area of discharge(22).
In addition to accidental discharges
to surface waters, water from an FGD sludge
pond can enter the aquatic environment
through: (1) an intentional direct dis-
charge to surface waters or (2) percolation
of pond water through underlying soils to
groundwater supplies. The possibility of
a direct surface discharge depends upon
details of disposal site design, Percola-
tion rates will again depend upon local
site characteristics and disposal techniques.
Irrigated Agricultural Considerations—
When estimating the potential impact
of FGD sludge leachate or liquor on irri-
gated agriculture, the most important con-
taminants are TDS, selenium, and boron.
None of these is readily absorbed by soils.
These contaminants can therefore enter
the ground or surface waters. After leav-
ing the site, these impurities may enter
an irrigation system. The adverse impact
of these components on agricultural lands
is documented. A high TDS concentration fn
irrigation water requires modification of
irrigation practices in order to prevent
high salinity in the root zone of the plants.
Livestock--
The effects of FGD sludge liquor con-
sumption on livestock could conveivably be
a local problem under unusual circumstances.
The salt tolerance limits of domestic live-'
stock are specific to each animal, although
generally between 3,000 to 12,000 ppm. As
FGD sludge pond supernatant is ordinarily
at the low end of this range, minimal dilu-
tion would render the leachate or super-
natant harmless in terms of TDS.
The metals and toxic anions noted
previously under Health Effects are equally
important here. Metals such as lead and
mercury are accumulated in animals and man
alike, and could present a problem with
livestock. The available chemical form of
each element is important, but further
investigation of speciation is necessary.
Aesthetics and Land Use
The aesthetic impact of an FGD sludge
disposal operation is strongly site-specific.
Local variables include public proximity,
appearance of the operation, and site re-
clamation potential. Untreated sludge
ponds resemble other common ponding operat-
ions, e.g., ash ponds, sewage lagoons, raw
water supplies, storage reservoirs, etc.
Obviously, disposal site location, grading,
landscaping, and access roads should be
selected to minimize public view and proxi-
mity to the site.
90
-------
SELECTION OF CHEMICAL AND
PHYSICAL REGULATING PARAMETERS
Rationale for Chemical Regulating Parameters
Various rationale are available for
establishing chemical regulating parameters.
Because groundwater and surface water pro-
tection are of primary concern, the regulat-
ing parameters for FGD sludge disposal
should be associated with the liquid phase.
The significant chemical characteris-
tics are the liquid phase concentrations of
TDS, boron, fluoride, mercury, lead, and
selenium. The selection of a regulating
parameter from among these components should
satisfy the following criteria:
• Accurate correlation with
environmental impacts;
• Usefulness as an indicator of
the migration of many contaminants,
aside from itself;
• Simplicity of measurement;
0 High measurement accuracy and
reproducibility; and
• Low analytical cost.
Of the contaminants listed, TDS provides
the best indication of overall contaminant
migration, due to limited attenuation and
ion exchange. TDS is also the.simplest
and least expensive contaminant to measure.
In view of these considerations, it appears
that TDS is the best chemical regulating
parameter for FGO sludge disposal. While
TDS should be used to monitor contaminant
migration, concentrations of other contami-
nants should also be measured in the
leachate to confirm the mass transport
calculations.
Rationale for Physical Regulating Parameters
The physical properties of FGD sludges
are important to the following disposal
site considerations:
• Future land use;
• Rate of mass transport of
contaminants from the site, as
a factor of in-place sludge
permeability; and
• Sludge flow during accidental
release (particularly to a
flowing stream).
Physical characteristics relating to these
considerations include load-bearing
strength, viscosity, and permeability. How
many, if any, of these characteristics are
important to a particular site are site-
dependent.
The use of physical "stability" as a
regulating parameter is subjective. While
stabilization may be desirable for site
reclamation in urban areas, the disposal of
an unstable material (slurry or thixotropic
solid) may be acceptable in areas of limited
land use or value. The use of an absolute
measure of stability for regulating all
disposal operations is therefore not
recommended.
The chemical and physical charac-
teristics of FGD sludge set needs for
regulation. Such regulations should
emphasize the site and sludge
characteristics.
ECONOMICS OF FGD SLUDGE DISPOSAL
The FGD sludge disposal cost data
base has expanded rapidly since 1976. This
expansion has been due to both the
completion of several important economic
studies (conceptual) and the publication of
sludge disposal costs for several operating
systems. The principal cost studies have
been performed by the Aerospace Corporation
(1976)(23), TVA (1977)(f4). and Michael
Baker, Jr., Inc. (1977)(25).
The principal factors affecting FGD
sludge disposal economics include:
1. Type of coal;
2. Characteristics of the air
pollution control system (wet
and/or dry collection, type of
sorbent, etc.);
3. Plant size, sludge generation
rate;
4. Distance from plant to
disposal site;
5. Geographic location (land
cost/availability); and
6. Site characteristics.
The following cost estimates are in
all cases based on one or more model power
plants, each of which has subtle but
important differences in the model plant
definition.
Aerospace Corporation (1976)
The Aerospace studies have developed
and updated detailed cost estimates for a
variety of disposal alternatives, including
91
-------
lined ponding (natural or synthetic liner)
and chemical fixation. Their most recent
findings, presented 1n Table 6, show chemi-
cal fixation to be 2 to 3 times as expensive
as ponding with a natural clay Uner. Costs
are based on conditions at a model mid-
western power plant.
Tennessee Valley Authority (1977)
The TVA study of FGD sludge disposal
economics developed comparative cost esti-
mates for four disposal alternatives at
each of 27 model plants (base case plus
variations). Table 7 shows the relative
ranking of the four alternatives for each
model plant, based on annual revenue
requirements. In general, the study found
fixation costs to be competitive with those
of untreated sludge disposal in many cases.
The base case analysis, for example, showed
IUCS fixation to be only 50 percent more
expensive than untreated sludge disposal
(compared to the Aerospace difference of
100 to 200 percent).
The extensive data base developed by
TVA also permitted a detailed sensitivity
analysis to be performed. The relative
effect of 17 case variations on disposal
costs for the 4 alternatives is presented
in Table 8. The most important variables
were shown to be distance to the disposal
site and plant size. Other variables had a
substantial effect on individual processes
which manifested itself in the broad range*
of costs between case variations. For
example, the IUCS and untreated disposal
alternatives were shown to have equal annual
revenue requirements for one case variation
(10 ml to disposal site), and yet another
TABLE 6. SLUDGE DISPOSAL COST RANGES (UNTREATED AND
CHEMICALLY TREATED LANDFILLS, 1000-MW
STATION. 50 PERCENT LOAD FACTOR, 30-YEAR
AVERAGE, JANUARY 1976 DOLLARS)
Disposal
Method
Untreated
Pond
Pond
Chemi-
cally
Treated*3
Base
Material
Q
Natural Clay
Liner'
Indigenous
Soil
$/Ton Sludgea> b
(Dry)
3. 50
5.70-7.80
7.30-11.40
$/Ton Coalb
1. 00
1. 60-2. 20
2. 10-3. 20
Mills /kWhb> Cf d
0. 4
0. 7-1. 0
0. 9-1. 4
510,000 short tons/year average (dry basis) including fly ash.
Coal burned at rate of 0. 88 Ib/kWh, 3% sulfur, 12% ash. 85% SO2
removal, 1.2 CaCO3/SO2 mole ratio.
ft
Land costs at $1000/acre are included (equivalent to $0.25/ton sludgi-
dry).
Disposal within 5 miles of power plant.
e 6
Assumes coefficient of permeability of clay is 1 x 10" cm/sec or
better.
Ponding costs cover range based on low-to-high material costs,
i.e., PVC-20 (low) to Hypalon-30 (high).
"Chemical fixation costs vary, depending on characteristics of the waste
and the disposal process chosen.
*Reference 26.
92
-------
TAIJLE 7* SUMMARY OF TOTAL ANNUAL RliVENUE REQUIREMENTS11
ALL PROCESSES
Untreated
Caea
k
Ba.ee caee
200 HV new
200 HV oletlng, 15 y regaining life
Bunting 25 yr rtwln ng life
Exlatlng 20 yr regain ng life
1500 MV nev
life
life
1500 HV enletlni, IS yr reulnlne.
life
12S aah In coal
20S aeh In coal
2S aulfur In coal
5S lulfur In coal
200 MV, no flyaeh
No flytih
1500 HV, no flyaah
5 mi to dlepoaal
10 •! to dlepolel
Fixation additive rete * variation 1
Lie* ecrubbtnft pioceie
Internvdlete ponding end truck to
landfill
Traneport by truck to dlapoaal area
Clarify (35S tolldt) end pu»p 1 *1
Clarify (35S eollda) and puBp 5 ml
Clarify (35S eolld.) and pu»p 10 .1
Clarify, filter (60S eollda) end
truck 1 »l
Clarify, filter (60S eollda) and
truck 5 mi
Clarify, filter (601 eollda) and
truck 10 el
(SOX of optlBua)
200 HU, constrained pond acreage
(751 of opclM)
Conatralned pond acreage (iOt of
optlamaO
Constrained pond acreage
1500 HV, constrained pond acreage
(7SS of optl»u»)
Pond erttted dtntltyUOS eollda)
Pond imtled den»lty(6Dl eollda)
Unltntd pond
Synthetic pond lining (Sl.SO/yd2)
Synthetic pond lining (S2.SO/yd2)
Synthetic pond lining (S3.50/yd2)
Synthetic pond lining (»4.50/yd2)
Annual
amount,
kS
3,280C
2,014
1^605
1.411
2.906
2,135
2, 1 30
6,746
5,827
4,726
3,969
2,902
3,609
2,639
3,869
1,495
2,289
4,546
5,527
7,504
-
3,136
-
-
3,694
5.195
6,450
3,809
4.925
5.818
2,462
2,031
4,119
3.165
8,985
7,037
3.747
2,949
2,765
3,882
4,253
4,590
4,900
Mill./
kvti
0.94C
1.44
1.32
1.15
1.02
0.8]
0.70
0.61
0.64
0.55
0.45
0.38
0.81
1.03
0.75
1.10
1.07
0.65
0.41
1.58
2.14
-
0.90
-
.
1.06
1.48
1.84
1.09
1.41
1.66
1.76
1.45
1.18
0.96
0.86
0.67
1.07
0.84
0.79
l.ll
1.21
1.11
1.40
Dravo
Annual
aexiunt ,
M
d
6,701°
3,641
1.6)2
1.461
1.190
6,177
5,941
5, 728
14^264
13,821
12,749
12,118
5,924
7,406
5.114
8,007
3.063
5,073
10,448
8,124
9,160
6,052
7 , 5 1 7
t.ite
6.620
.
_
_
-
_
_
-
4.423
4,134
9.109
7.901
20.166
17,290
_
_
_
_
_
_
Mllle/
VVh
A
1.91
2.60
2.6!
2.4)
2.4)
1.11
1.70
(,(,*,
1.12
1.21
1.16
1.69
2.12
1.51
2.29
2.19
1.45
1. 00
2.12
2.67
i.n
v'.i*
1.89
_
_
_
-
_
_
-
3.16
2.95
2.60
J.25
1.94
l.SS
_
_
_
_
_
_
"
IUCS
Annual
aenunt ,
kS
5.291*
3.567
3.699
3,725
5,402
5,430
5, SS9
10.411
10.656
10,684
10,900
4,531
5,971
4,654
.188
.196
.650
.162
.490
7.475
4.994
i, 1 1 1
S.091
-
.
_
_
-
_
_
-
_
„
_
-
_
.
_
_
_
_
_
_
"
Nllle/
kVh
1.51*
2.55
2.64
2.66
J.7J
\ .54
1.55
0.99
1.01
1.02
.04
.30
.71
.33
.77
.43
.33
.78
.85
.14
.43
. S 7
.(.6
_
_
_
_
.
_
_
-
_
.
_
-
_
_
_
_
_
_
_
_
"
Che.flu
Annual
Mount ,
,
6,988
4,529
4.695
4.710
4.856
7,152
7.191
7 359
14,' 162
14,749
14,791
15,053
6,229
7,600
5,915
8,263
1,766
5,184
9,771
8.675
10.00]
13.651
10 ,099
$.100
_
6,698
_
_
.
_
,
-
„
.
_
-
_
„
_
_
_
_
_
_
~
Hlltn/
kVTh
r
2.00
1.24
1.15
1.38
1.47
2.04
2.05
1.37
1.40
1.41
1.41
1. 78
2.17
1.70
2.16
2.69
1.48
0.93
2.48
2.86
1.9(1
2. 89
V.77
_
1.91
_
_
_
_
_
-
_
_
_
-
_
_
_
_
_
~
Raala: Midwant plant location; »ld-19BO operating coat*; 7.00O hr/yr on-streum time.
New 500 HU pl«nt; )0-yr life; roul analyala (by wt) : LSI S (dry haela) , 161 Aah; flyaiin and
SO rewivrd toftvthvi; l»» aaet HSPS, 11 meat one proc**aa with 1,3 atolchlnaetry based on SO removed.
Dtfect pondtnft of I >I elurry; clay-lined pond; 1 ail from ncrubber fa.«lt teleii; 50t aolldB aettled den
Pending of Yi\ aolld* alurry; clay^llned pond; pimp 1 ml to pond fro* aerubber facltlttea; SOZ noltJ
aettled denalty In pond; Created with CaUtlox <7t of dry elud**) and Thtoearbic line (IX of dry *\->
Landfill dlNpo^al of 60? aolida naeertal ; 1 mi to landfill trow acrubber faetlUlea. trucke uaed for
tr ana port o( treated aludge ; treated with Have (41 of dry eludge) .
Landfill dlapoial of 60X aoltda uterlal; L &i (o UndfUl ttom «crubber facllttlea; pipeline u«e4 t
traneport pf thlrkenir under t IBM to diapAaal and t reatn*nt are a, t r*a»t §>e! wl tK Port l»nt ccacnt ( 'X r>f
dry iludge) and nodlu* Ullcma (21 B( dry aLudge).
*Reference 27.
93
-------
TABLE 3* EFFECTS ON TOTAL CAPITAL INVESTMENT AND ANNUAL REVENUE
REQUIREMENT BY CASE VARIATIONS FOR ALL PROCESSES
gree of change from base case
Capital Investment
Case variation*
Plant slie
Rem/tlnlng plant life
Aah in coal
Sulfur In coal
No flyash In sludge
Distance to disposal
Fixation additive rate
scrubbing process
Thickened sludge
Thlrkened sludge and dis-
tance to disposal site
Filtered sludge
Filtered sludge and dis-
tance to disposal site
Constraint of available
pond acreage
Pond 1 In Ing
Truck to disposal
Intermediate ponding
and truck to landfill
Settled density in pond
Vnt rented
Large
Large
Small
Moderate
Moderate
Large
-
Slight
Small
Large
Large
Large
Moderate
Large
-
Small
Dravo
Large
Moderate
Small
Small
Small
Large
Slight
Small
-
-
-
«•
-
-
Large
Large
Il/CS
Large
Slight
Small
Small
Slight
Small
Slight
Slight
-
-
-
-
-
-
-
. . .
Cheraf Ix
Large
Slight
Small
Small
Moderate
Large
Small
Slight
-
-
-
-
-
-
Saall
-
Revenue requirements
Unt reated
Large
Moderate
Small
Snail
Moderate
Large
-
Slight
Small
Large
Small
Large
Moderate
Large
-
Small
Dravo
Large
Small
Small
Moderate
Moderate
Large
Small
Slight
-
-
-
-
-
-
Slight
Large
IUCS
I..ITKC
Slight
Small
Snu 1 1
Small
Large
Slight
Slight
_
_
_
-
-
_
-
-
Uu-mf in
L.irKC
SI ll'.llt
Sm.i 1 )
Sm,i 1 1
Moclo rate
Large
Large
Small
.
_
Slight
_
An arbitrary scale is established with the following rating criteria being applied.
Slight: less than -flOX change from the base case
Small: from ±10% to +20Z
Moderate: from greater than +201 to +40X
Large: greater than + 40Z
*Reference 28.
case (1,500 MW existing, 15 year remaining
life), the IUCS process was 175 percent
more expensive than untreated disposal.
The most important conclusion to be
drawn from the TVA work is that commercial
fixation can be competitive with untreated
FGO sludge disposal in some Instances.
The actual cost of any disposal alternative
is site-specific, so generalizations are
not recommended.
Michael Baker. Jr., Inc. (1977)
The Michael Baker study for EPRI
evaluated the state-of-the-art of FGD sludge
fixation and included detailed cost esti-
mates for untreated sludge disposal (wet
and dry), and Dravo and IUCS fixation.
This study used a single model plant
(similar to the TVA base case) for cost
analysis. The results showed that fixation
is only slightly more expensive than wet
or dry untreated disposal for the model
plant. The 30-year levelized revenue re-
quirements for the respective processes
were as follows:
Wet disposal without fixation $10.90
Dry disposal without fixation $M.'oO
Dry disposal with fixation
(IUCS) $12 40
Wet disposal with fixation
(Dravo) $14.60
It was noted in the study that the figures
are again quite sensitive to site-specific
conditions and requirements, in particular,
the hauling distance to the disposal site,'
cost of fixation additives, and site
geology. Blanket conclusions about the
relative cost of these disposal alterna-
tives were therefore avoided.
Ongoing Studies of FGD Sludge Disposal
ft
Economics
SCS Engineers is currently under
contract with EPA (No. 68-03-2471) to assess
the economic impact of various FGD sludge
disposal regulatory scenarios on the utility
industry. Using the TVA estimates, a
comprehensive list of model plants will be
developed and their disposal costs assessed
based upon variation in the TVA base case.
94
-------
Selected regulatory scenarios will then
be applied to the model plants to determine
the most cost-effective means of Industry
compliance. Results are expected In
late 1978.
TVA 1s also expanding the previous
effort to Include costs for: (1) disposal
of gypsum from FGD sludge-forced oxidation,
and (2) disposal of untreated sludge which
1s dewatered and blended only with fly ash.
REGULATIONS WHICH IMPACT UPON
FGD SLUDGE DISPOSAL
Regulations that have been or could
be applied to residual disposal practices
fall under one or more of the following
classifications:
• Water quality criteria for
various beneficial uses;
fl Comprehensive air quality criteria;
• Solid waste disposal;
• Industrial and hazardous
waste disposal;
• Waste disposal to oceans;
t Waste disposal to mines.
In implementing a residual disposal plan
for FGD-equipped power plants, the power
companies have had to comply with one or
more of the above regulations.
Water Quality Standards
The Federal Water Pollution Control
Act Amendments of 1972 (Public Law 92-500)
established the framework for the water
quality control laws and regulations used
by the various states. Under Public Law
92-500, the states retain the primary
responsibility for water quality protection.
However, the U.S. EPA is authorized to
Intervene, if the states do not enforce the
law. The relevant elements of Public
Law 92-500 are:
t Water Quality Standards Program
• NPDES Program
• Treatment Requirements
• Effluent Guidelines
All states currently operate a water
quality standards program for surface
waters, which is used as a basis for Issu-
ance of discharge permits. These water
quality standards would only be applicable
to FGD sludge liquors if they were dis-
charged to a surface water. The recommen-
ded relating parameters for wastswater
disposal Include:
• Dissolved oxygen (DO)
• pH and acidity/alkalinity
t Suspended and other particulate
solids
0 Heavy metals
In nearly all cases, the state regu-
latory agencies are authorized to protect
both surface and subsurface waters, i.e.,
prevent degradation of groundwater. How-
ever, only a few states have specific
criteria for groundwater protection.
Air Quality Standards
With the promulgation of the New
Source Performance Standards (NSPS) for
Fossil Fuel-Fired Steam Generators (40 CFR,
Part 466, 8/17/71), wet lime/limestone
scrubbers became the anticipated "best
method" of removing S02 from the flue gas
of a power plant.
The NSPS were remanded to EPA by a
U.S. Court of Appeals in 1973, with one of
the allegations involving insufficient EPA
consideration of the sludge disposal
problem. The court decision stated that
the record be remanded for further "con-
sideration and explanation...of the adverse
environmental effects of requiring a 1.2
Ib/million (BTU) standard for those...
plants which must use a lime slurry scrub-
bing system..."
In its Response to Remand (40 FR42045)
EPA concluded that the standards should not
be revised to account for sludge disposal.
It was noted, however, that "EPA considers
permanent land disposal of raw sludge to be
environmentally unsound, because it defi-
nitely degrades on large quantities
of land" (Section C.I. (c)). As an accep-
table solution, both fixation and alterna'te
control systems which do not generate sludge
are considered acceptable. The response
states that "Fixation of the sludge will
greatly reduce its environmental impact,"
as it has a "much lower Teachability" than
untreated sludge "and the leachates...are
potentially less of a disposal problem than
(those) from fly ash."
Subsequent legal decisions (Essex
Chemical, Indiana Public Service) have
failed to resolve the issue of turning an
air pollution problem into a potential water
pollution problem. EPA has not as yet
stated a formal position on FGD sludge dis-
posal other than that implied in the
Remand.
95
-------
Other Waste Disposal Requirements
Federal regulations controlling the
disposal of certain industrial wastes could
be applied, at least indirectly, to the
disposal of F6D sludges. Similarities bet-
ween these waste categories and FGD sludges
are based on one or more common potentially
hazardous constituents and/or a similarity
in the method of disposal. The industrial
wastes discussed below are:
• Slimes from phosphate mining;
• Acid mine drainage;
• Gypsum from phosphoric acid
manufacture; and
• Coal mine tailings.
The principal regulatory emphasis in these
industries is toward proper impoundment
construction due to a history of catastro-
phic structural failures. Groundwater
protection and physical stabilization are
typically not considered in federal or
state regulation of disposal of these
wastes.
Solid Waste
The predominant method for the ulti-
mate disposal of solid waste is the land-
fill. Generally, federal and state regula-
tions governing the disposal of solid waste
apply typically to specific waste classi-
fications (municipal refuse, brush and tree
wastes, demolition and construction waste,
sewage sludge, bulky wastes). The regula-
tion of these facilities and activities
has been the primary responsibility of state
and local agencies.
Federal Mandates--
The recent passage of PL 92-580, the
"Resource Conservation and Recovery Act of
1976" (RCRA), sould have some impact on FGD
sludge disposal regulation by states. The
most applicable portion of RCRA is Subtitle
C, "Hazardous Waste Management," under which
EPA must develop criteria for listing
hazardous wastes. According to RCRA, a
hazardous waste is defined as: "a solid
waste, or combination of solid wastes,
which because of its quantity, concentra-
tion, or physical, chemical, or infectious
characteristics may -
(A) cause, or significantly contri-
bute to, an increase in mortal-
ity or an increase in serious
irreversible, or incapacitating
reversible, illness; or
(B) pose a substantial present or
potential hazard to human health
or the environment when improper-
ly treated, stored, transported,
or disposed of, or otherwise
managed."
Should FGD wastes or any portion
thereof meet these criteria, standards
promulgated under RCRA for waste transport,
treatment, storage, and disposal would
become applicable. The hazardous waste
"definition" under RCRA is not expected
until early 1978.
Section 3001 stipulates that criteria
for "identifying the characteristics of
hazarJous waste, and for listing hazardous
waste" shall be promulgated by early 1978.
Similarly, standards applicable to genera-
tors, transporters, and disposers of
hazardous waste shall be promulgated.
State Regulations--
Examination of the solid waste regula-
tions for the 50 states revealed some nota-
tion of groundwater protection, flood plain
consideration, and site reclamation in
almost every case. The degree of specifi-
city varied significantly, however, as many
states prefer a permit system and reserve
the right to site-specific evaluation.
Many of the power plants employing
FGD have been subject to disposal site
approval by state solid waste agencies.
Site evaluation is usually limited to soils
analysis, including permeability and chemi-
cal analysis. Contacts with state agencies
revealed that FGD site evaluation criteria
were generally excerpted from applicable
landfill requirements, since specific FGD
sludge disposal site evaluation criteria
were not available.
Table 9 summarizes the groundwater, flood
protection, and site reclamation rules for
selected states in which an FGD-equipped
utility is presently located; thede do not
necessarily reflect the fules which were
applied to the FGD systems themselves, but
serve only as an interpretation of the
written regulations. Those state regula-
tions which do not specifically refer to the
given problem are indicated by the word
"none." Specific comments follow concerning
leachate control, flood and drainage control,
and site reclamation.
Leachate from Sludge Disposal--Most states
have adopted the federal recommendation for
protecting groundwater quality beneath solid
waste disposal sites without significant
extrapolation. Some, however, have speci-
96
-------
TABLE 9.*
STATE SOLID WASTE REGULATIONS PERTAINING TO EXISTING FGD SYSTEMS
State
Ariz.
Col.
Groundwater
General
General
protection
protecti on
None
Floods
General
rainfal 1
Rainfal
None
General
1
Reclamati
2' mi-
on
nimum cover
rainfal 1
Fla. Monitoring, possible
leachate collection
111. Monitoring, soil
analysis
Kan. General protection,
soil analysis
Ky. General protection,
soil analysis
Mo. Monitoring, soil
analysis, possible
leachate collection
Mont. General protection,
soi1 analysis
Nev. Stabilize or
neutralize
Penn. Monitoring, soil
analysis, possible
leachate collection
protection
Away from flood
plain
General rainfall
protection
None
100 yr. flood
100 yr. flood
100 yr. flood
None
50 yr. away from
flood plain
protection
General diversion
General rainfall
protection
None
General diversion
20 yr. rainfall
20 yr. rainfall
None
General diversion
2' minimum cover,
33% slope (sides)
2' minimum cover,
50% slope (sides)
State approval
2' minimum cover,
revegetation
2' minimum cover,
33% slope (sides)
revegetation
2' minimum cover
2' minimum cover,
2-4% slope
2' minimum cover,
1-15% slope
*Reference 29
-------
fled restrictions pertinent to site design
and location. Most states which require
groundwater monitoring during operation of
the fill also require the monitoring to be
continued for a specified period after
closure.
Flooding and Drainage Considerations--Most
of the state solid waste disposal regula-
tions address the problem of flooding, while
all regulations mention or specify the fill
slope necessary to divert runoff water.
Flood consideration varies with each
regulation, from simple mention of the
potential danger to specifications regarding
site location based on flood probability.
Most states specify that the site shall not
be located within a 50- or 100-yr flood
plain. Some require encapsulation or other
protection if the site is to be located in
a flood plain. Provisions to eliminate
standing water from landfills would obviously
apply only to stabilized or dried sludge
lagoons and landfills.
Provisions for Site Reclamation--The extent
to which a landfill site is reclaimed is up
to the owner. Many states do, however,
require a certain amount of landscaping and
planting to prevent devaluation of nearby
properties, excessive erosion, or standing
water accumulation. Reclaimed site uses
vary, but most landfills are developed into
recreation areas; the erection of structures
over the fill is generally not recommended
due to structural instability and subsidence.
Regulating parameters such as load-
bearing strength are rarely, if ever,
applied to site reclamation. Some states
require that subsidence, surface erosion,
and cracking be monitored for several years
after closure.
SUGGESTED REGULATORY APPROACHED)
The physical and chemical properties
of FGD sludge set a need for regulation.
Because the impact of these sludge charac-
teristics on the environment will vary
according to certain characteristics of the
site, ah equitable regulatory approach
would be based upon these site
characteristics.
The suggested regulatory approach is
analogous to the procedures being followed
by state regulatory agencies in establish-
ing guidelines and limitations for treated
wastewater point sources. Regulatory
agencies in their review of proposed new
wastewater treatment facilities for point
source discharges go through the following
steps: a
1. The beneficial use of the
receiving water is established;
2. Based upon the beneficial use
established, maximum concentra-
tions for various contaminants
in the treated effluent are set;
3. The applicant responds with a
treatment technology which will
theoretically supply sufficient
treatment to meet the effluent
limitations on a sustained basis;
4. Upon approval of the technology
the regulatory agency will provide
the applicant with specific design
criteria for use in preparing
detailed design plans and speci-
fications; and
5. Finally, the regulatory agency
establishes monitoring, control,
and reporting procedures which
the applicant must implement to
ensure that the effluent limita-
tions established in No. 2 above
are being met.
The above approach is site-specific, as
opposed to a universally applied "one
number" approach and recognizes that local
site conditions are important in establish-
ing best practical control technology. The
approach also provides continuous inter-
action between applicant and regulatory
agency to ensure that environmental protec-
tion objectives are met.
The wastewater effluent limitations
approach outlined above can be adopted to
FGD sludge treatment/disposal guidelines
and limitations. Figure ] shows in a
simplified flow diagram the major steps
involved. If each of the steps shown can
be carefully defined in a procedural manner
it will provide a uniform, organized '
approach to decision-making based upon
site/waste-specific data. The following is
a discussion of each of these steps:
Define Beneficial Uses
In this step, the regulatory agency
with the cooperation of the applicant,
performs an evaluation of the proposed
disposal site to determine what beneficial
environmental resources require protection
The environmental factors which warrant
98
-------
Establish Environmental Benefits
to be Protected at
the Treatment/Disposal Site
Determine Degree of Protection
Necessary to Protect
Environmental Benefits
Categorize Treatment/Disposal System
Capability to Provide Necessary Environmental
Benefit Protection
EstablIsh Monitoring.
Control, and Reporting Procedures
FIGURE 1.* STEPWISE APPROACH TO THE
PROPOSED REGULATORY SYSTEM
*Reference 30
protection during FGD sludge disposal are
the following:
1. Groundwater resources under-
lying the site;
2. Local surface water resources
at elevations lower than the
site; and
3. Future land uses at the site
location.
In order to characterize the ground-
Mater resources, a detailed hydrological
report would be required. The report would
Include: (1) the location and size of
groundwater aquifers, (2) the quality of
each aquifer, and (3) a best estimate of the
existing and/or potential beneficial uses
of each aquifer.
The potential dangers to local surface
waters should be evaluated. This assessment
Mould be based on the following criteria:
• Distance from site to surface
water(s);
• Elevation difference between
site and surface water(s);
• Type of surface water(s), e.g.,
lake, river, etc.;
• Present or projected beneficial
uses of surface water(s); and
t Volume and/or flow rate(s) of
the surface water.
Future land use would be projected to
some point in time to determine the extent
of site reclamation required. The load
bearing capability of ponded, unstabilized
sludge is generally poor. With proper
cover material, light recreational use is
possible, but in essence the pond area is
eliminated from higher use purposes. How-
ever, in some disposal site areas this
situation may be acceptable, because the
land has little projected value.
Categorize Sites According to Degree of
Protection Necessary
Having assessed the acceptability of
the site for FGD sludge disposal, the site
can then be categorized for the purpose of
regulation. Table 10 lists 7 comprehensive
site categories based upon possible bene-
ficial uses. The categories range from
sites having no environmental benefits to
protect (Category 1), to sites which are
unacceptable for FGD sludge disposal
because no practical technique will ade-
quately protect a valuable-groundwater
resource.
The 7 categories selected evolved
from the various combinations of decisions
which would be made based on each site
variable. In order to better demonstrate
the logic of site categorization, Figure 2
shows a conceptual decision tree approach.
Each of the decision points is based on the
following 5 variables:
1. Possible future land use
• possible urban or industrial
development
• little or no probability of
development
2. Groundwater resources
• potentially useful
• do not meet useful standards
or not accessible for use
3. Protection of groundwater by site
• time constants (rate of flow)
• rate constants (mass transport
of contaminants)
4. Surface water protection
• sludges cannot reach water
course
99
-------
TABLE 10-
8
Site
Category
No.
1
2
3
4
5
6
7
Environmental Benefit
Requiring Protection
Land Surface Ground-
Value Water water
Low
Low
High
High
Low
Low
No
Yes
No
No
No
Yes
No
No
No
Yes'3)
Yes
Yes
Protection Method SUe Modification .
Sludge Retention Reduce Mass
Pretreatment Dikes Transport
No
No
Stabilize^1 *
Stabilize
No
No
Site unacceptable regardless of protection methods used
of adequately protecting groundwater resource
Standard
(2}
Extraordinaryv '
Standard
Standard
Standard
Extraordinary
due to impossibility
No
No
No
Yes
Yes
Yes
(1) It is assumed that stabilized sludge will not flow and poses no threat to surface water
regardless of retention dike integrity.
(2) Dual dikes, supplementary interceptor pond, or other failsafe method of preventing
escape of sludge if primary dikes fail.
(3) See narrative for discussion of pollutant mass transfer potential through soil.
*Reference 30.
-------
CATEGORY !
UNSTABLE
CATEGORY
UNSTABLE .
PELIAP1LITY DIKE
LOW FLOW, HIGH
PCL1AH1U1TY DIKE
CATEGORY *,
LOW FLOW
MODIFY SITE TO
LOWER MASS
TRANSPORT
\
XILL
GROUNDWATEn
IS TOS > 10X7
,
WATER
EFJDANC
NO
FACE
.EOEO7
YES
MODIFY SITE TO
LOWER MASS
TRANSPORT
•
•
"ILL
GROUNDWATER
S TDS > 10X7
NO
YES
CATEGORY »
STABLE. LO» FLOW
CATEGORY 7
UNACCEPTABLE
FIGURE 2.* DECISION TREE APPROACH TO DISPOSAL SITE CATEGORIZATION
*Reference 30
-------
• sludges can reach water course
5. Site meteorology
t Net evaporation area
• Net precipitation area
The rationale for selecting the values
shown and the related alternatives is
partially subjective. Zero impurity trans-
port is physically impossible, but accep-
table mobility limits (mass transport) have
not been legally defined. Also, levels of
"acceptable degradation" have not been
defined. Thirdly, the prediction of land
use (and value) has not been defined
legally or socially. Recognizing these
constraints, the categorization approach is
nonetheless valid in concept based upon
the existing data base for F6D sludge
disposal.
Establish Monitoring, Control, and
Reporting Procedures
Once the disposal site had been
subjected to the above criteria, it would
become operational under the appropriate
category and its restrictions. Monitoring,
control, and reporting procedures should be
specified for all FGD sludge disposal sites.
During the categorization procedure,
the potential for environmental impact
from several parameters was evaluated.
These parameters Included:
• Groundwater contamination
• Sludge in-place stability
• Catastrophic sludge release
to surface waters
The monitoring system would be estab-
lished for each of the above parameters
to verify, on a continuing basis, the
validity of the categorization scheme
selected.
REFERENCES
1. Rlttenhouse, R.C., "A Profile of New
Generating Capacity, Power Engineering
80(4):76-82, April 1976.
2. Clean Air Act Amendments, August 7,
1977 (PL 95-95).
3. Hurtar, A.P., Jr., "Flue Gas Desulfuri-
zation and Its Alternatives; The State
of the Art, Argonne National Library,
Argonne, Illinois, November 1974.
4. PED Co - Environmental Specialists,
Inc., "Summary Report: Flue Gas
Desulfurization Systems, EPA 68-02-1321
U.S. Environmental Protection Agency '
December 1976. '
5. SCS Engineers, "Speciation of Contami-
nants in FGD Sludge and Wastewater -
Interim Report," EPA Contract 68-03-
2371, March 1978.
6. SCS Engineers, "Data Base for Standards/
Peculations Development for Land
Disposal of Flue Gas Cleaning Hastes "
EPA Contract No. 68-03-2352, July 1977
Final Report (unpublished). '
7. Ibid.
8. SCS Engineers, "Speciation."
9. Arthur D. Little, Inc., "An Evaluation
of Alternatives for the Disposal of
FGD Sludges, Progress Report No. 4 "
EPA Contract No. 68-03-2334, U.S.
Environmental Protection Agency,
December 17, 1975.
10. Humenick, M.H. and A.J. Huckabee, "SO?
Scrubber Sludge Dewatering," paper
presented at the 1976 Purdue Industrial
Wastes Conference, (unpublished).
11. Leo, P.O. and J. Rossoff, "Control of
Waste and Water Pollution from Power
Plant Flue Gas Cleaning Systems: First
Annual R&D Report," EPA-600/7-76-018,
U.S. Environmental Protection Agency'
October 1976.
12. Rossoff, J. and R.C. Rossi, "Disposal
of By-Products from Non-Regenerable
Flue Gas Desulfurization Systems:
Initial Report," EPA-650/2-7-037a,
U.S. Environmental Protection Agency
May 1974.
13. Humenick and Huckabee.
14. Isaacs, G.A. and F.K. Zada, "Survey of
Flue Gas Desulfurization Systems: Will
County Station, Commonwealth Edison
Company," EPA-650/2-75-0571, U.S.
Environmental Protection Agency
October 1975.
15. Leo and Rossoff.
16. Mahloch, J.L., D.E. Averett, and M.J.
Bartos, Jr., "Pollutant Potential of
102
-------
Raw and Chemically Fixed Hazardous
Industrial Wastes and Flue Gas Desul-
furization Sludges, Interim Report,"
EPA-600/2-76-182, U.S. Environmental
Protection Agency, July 1976.
17. "Stabilizing Waste Materials for Land-
fills," Environmental Science and
Technology, 11(5):436-437, May 1977.
18. Taub, S.I., "Treatment of Concentrated
Waste Water to Produce Landfill
Material," presented at the Inter-
national Pollution Engineering Exposi-
tion and Congress, Anaheim, California,
November 10, 1976.
19. Rossoff, J. et al., "Disposal of By-
Products from Non-Regenerable Flue
Gas Desulfurization Systems: Second
Progress Report," EPA 600/7-77-052,
May 1977.
20. SCS Engineers, Data Base.
21. SCS Engineers, "State-of-the-Art
Evaluation of Health Effects Associated
with Wastewater Treatment and Disposal
Systems, Draft Final Report." U.S.
Environmental Protection Agency,
Research Triangle Park, North Carolina,
February 4, 1976.
22. SCS Engineers, Data Base.
23. Rossoff et al. (1977).
24. Barrier, J.W. et al., "Economics of
FGD Waste Disposal," presented at the
EPA Symposium on Flue Gas Desulfuriza-
tion, Hollywood, Florida, November 1977.
25. Baker, Michael, Jr., Inc., "State-of-
the-Art of FGD Sludge Fixation, Final
Report," EPRI FP-671, January 1973.
26. Rossoff et al., (1977).
27. Barrier, p. 14
28. Ibid, p. 42.
29. Woodyard, J.P. and G.L. Merritt,
"The Regulation of Flue Gas Desul-
furization Sludge Disposal - A
Status Report," presented at the 38th
International Water Conference,
Pittsburgh, Pennsylvania, November
1977.
30. SCS Engineers, Data Base (Appendix A).
103
-------
LABORATORY CHARACTERIZATION OF THE THERMAL
DECOMPOSITION OF HAZARDOUS WASTES
D.S. Duvall, W.A. Rubey, and J.A. Mescher
University of Dayton Research Institute
300 College Park
Dayton, Ohio 45469
ABSTRACT
A specialized laboratory system incorporating a two-stage quartz tube
was utilized for determining the thermal destruction properties of pesti-
cides and other hazardous organic substances. With this system, a small
sample was first converted to the gas phase, then exposed to high-temper-
ature destruction conditions in flowing air. Critical parameters of
temperature and residence time were accurately measured. High-temperature
decompositions of Kepone, Mirex, DDT, and PCB's have been studied.
At this time, a new, more sophisticated closed system is being develop-
ed to conduct thermal destruction studies. In addition to increased
versatility and more refined control of the thermal destruction process, a
gas chromatograph-mass spectrometer with dedicated minicomputer is being
coupled to the system as a tool for identifying and quantifying products of
decomposition.
INTRODUCTION
The problems associated with the
safe disposal of hazardous organic
waste materials have been recognized
for several years. For these
materials, high temperature incin-
eration is being considered as a
method of permanent disposal. How-
ever, there is a serious lack of
basic data concerning the high-
temperature decomposition behavior
of organic materials.
On the laboratory-scale,
accurate measurements can be obtain-
ed of destruction temperature and
residence time. In addition,
thorough analysis of the decom-
position and/or recombination pro-
ducts will definitely show if
controlled high-temperature incin-
eration is a viable disposal route
for a specific waste material.
Consequently, such laboratory-scale
testing should logically precede
pilot-scale studies and the eventual
use of full-size hazardous waste
incinerators.
The University of Dayton Research
Institute, operating under Environ-
mental Protection Agency grants, is
developing laboratory procedures
for studying the high-temperature
decomposition of a wide variety of
organic hazardous waste materials.
The purpose of this presentation is
to discuss the laboratory instru-
mentation and to describe its
practical application.
RATIONALE OF LABORATORY
APPROACH
In order to determine the high-
temperature destruction character-
istics of an organic molecule, it is
necessary to obtain precise informa-
tion on certain crucial parameters.
It is important that the exposure
temperature be accurately deter-
mined, along with the residence
time, i.e., the time interval during
which the molecule experiences the
destructive temperature. Another
critical parameter that must be ful-
filled in such an evaluation is that
104
-------
there be an adequate supply of air
so that the destruction takes place
in an oxidative environment, as
opposed to an atmosphere containing
insufficient oxygen. Lastly, an
important factor is the composition
of the high-temperature environment,
e.g., water content, presence of
other gases, and interfering solid
materials.
The laboratory approach to
establishing a material's high tem-
perature, nonflame destruction
characteristics has certain distinct
advantages. First, one can examine
the undiluted sample without inter-
ferences from other materials.
Second, the composition of the high-
temperature environment can be pre-
cisely established by using com-
pressed air of known quality and
employing in-line filters to remove
water, oil, and other foreign mate-
rials. Next, by using a technique
whereby a small sample is gradually
vaporized and then passed through a
high-temperature zone, an excess of
oxygen is assured, thus avoiding the
possibility of an oxygen-deficient
reaction occurring. Further, it is
possible to evaluate the behavior of
the pure sample on the molecular
level. By vaporizing the sample
prior to its exposure to the high-
temperature environment, one can be
assured, based on the kinetic theory
of gases'1', that the molecules
experience the actual average tem-
perature. Finally, the laboratory
evaluation of a sample's destruction
characteristics can be accomplished
quickly and economically with
minimum environmental risk.
It is important to note that
during incineration of waste mate-
rials, a certain amount of energy
must be applied just to change the
organic compounds from their usual
state (whether it be solid or
liquid) to the gas phase. These
phase transitions, apart from re-
quiring additional energy, also
require an undefined amount of
exposure time to the high-tempera-
ture source. Therefore, it is al-
most certain that in the case of
waste materials, some molecules do
not encounter the prescribed
incineration temperature. In short,
even though the temperature within
a chamber is known, it does not
necessarily mean that all substances
within that chamber have attained
thermal equilibrium. However, if
all of the substances in the
chamber are in the gas phase, one
can safely assume that the individ-
ual molecules have (with sufficient
time exposure) achieved the average
temperature of that environment.
For similar reasons, the
laboratory evaluation of a sample's
residence time is considerably
simplified over that of measuring
residence time in a large scale
unit. When a gas is passed through
a long narrow-bore flow path (in the
laminar flow region), radial dis-
persion can be neglected and the
main factor affecting the residence
time distribution is longitudinal
dispersion. However, for flow paths
that are of large diameter, mixing
chamber, or multichamber configu-
ration, radial dispersion is the
major factor affecting the variation
in residence time of transported
molecules.
The approach for obtaining high-
temperature destruction data on
selected organic wastes utilized a
discontinuous system where the
thermal stressing of the sample and
product analysis were performed
separately. Using this approach,
various organic samples have been
evaluated in a series of tests where
the sample was first vaporized and
then transported through a narrrow-
bore, high-temperature zone by a
controlled flow of air. In this
way, average temperature and resi-
dence time were firmly established.
These tests were conducted at a
series of temperatures ranging from
^300°C to 1000°C. Also, the
effluent from each high-temperature
test was passed through an appro-
priate trapping medium and the
collected fraction was subsequently
analyzed by gas chromatography.
Quartz Tube Apparatus
Figure 1 is a schematic of the
high-temperature quartz tube
105
-------
COMPRESSED AIR, BREATHING OUAUTY GRADE
TWO STAGE PRESSURE REGULATOR
'HYDROPURGE" FILTER
FLO* CONTROL VALVE
PRESSURE TRANSDUCER
SAMPLE HOLDER. PYREX
HEATED INLET CHAMBER
OUART2 TU«E
HEATED OUTLET CHAMBER
EFFLUENT TRAP. TENAX- GC OR CHARCOAL
FLOW METER
Figure 1. Schematic of Quartz
Tube Apparatus.
apparatus that was designed and
assembled for the examination of
pesticides and other organic wastes.
Upstream of a folded quartz
tube, in a furnace capable of 1000°C
is a special sample holder. With
the use of a 10 microliter syringe,
a known quantity of sample can be
readily deposited on the sand
blasted region of the sample holder.
The solvent quickly evaporates,
leaving the pure sample on the rough
Pyrex surface. This device is then
inserted into a chamber upstream of
the high temperature quartz tube.
Next, this chamber is heated to
vaporize the pesticide or other
organic waste. The gas phase sample
is then swept through the high-
temperature tube, and captured in an
effluent trap located at its outlet.
The outlet of the trapping medium
tube is near ambient temperature.
Once the sample molecules, or frag-
ments and products thereof, have
passed through the high-temperature
quartz tube, they are trapped in, or
on, the sorbing medium of the
effluent trap. This trap is then
removed from the high-temperature
apparatus, and a quantitative
standard deposited in the trap
(4.0 yg of n - octadecane in a
50:50 vol mixture of acetone and
benzene). The contents are then
subjected to a programmed temper-
ature gas chromatographic analysis
using a modified Tracor 550 Instru-
ment (Figure 2).
Figure 2. Interior of Modified
Tracor 550.
During this study, different
trapping media were investigated.
Of these, Tenax-GC was found to be
an excellent trapping medium for the
three pesticides and their large
fragmentation products. Tenax-GC,
a porous polymer, is currently
commercially available in 35/60 and
60/80 mesh sizes, and possesses a
desirable feature in that it can be
readily, thermally desorbed^2"6).
Traps prepared with the 60/80 mesh
Tenax-GC particles performed quite
well and did not present an excessive
pressure drop. Also, a 2.0 cm length
of Tenax-GC was sufficient to obtain
quantitative recoveries of the pesti-
cides. However, acetone and benzene
were not quantitatively retained at
the typical trap inlet temperature of
approximately 300°C.
Discussion
To illustrate the success
achieved with the system, a brief
review of the work done with Kepone,
Mirex, and DDT follows. The chemical
structures of these compounds are
shown in Figure 3.
106
-------
KEPONE
MIREX
p.p. DDT
Figure 3. Chemical Structures.
Destruction of Kepone
From the present experimental
data, it was observed that Kepone can
be readily vaporized in flowing air
at temperatures ranging from approxi-
mately 200° to 300°C. It was also
observed that this temperature range
is not destructive to the Kepone
molecule. However, when Kepone was
subjected to temperature above 350°C,
decomposition does occur as is
evidenced by the rapid drop in the
Kepone concentration shown in Fig-
ure 4. Also, it is seen that the
thermal degradation of Kepone at
exposure temperatures less than
600°C is markedly different from that
when gaseous Kepone is subjected to a
temperature exceeding 600°C. For
example, peak Ka, which was identi-
fied as hexachlorocyclopentadiene,
was formed at temperatures below
500°C; however, it was not found
above 600°C. Conversely, Kc, which
Ul
I
B
100
30
10
30
1.0
o
I
CL
g 0.3
o
9°
o
O.I
0.03
001
KEPONE
J L
200
400 600
TEMPERATURE. *C
800
1000
Figure 4 . Thermal Destruction
Plot for Kepone.
was determined to be hexachloro-
benzene, was observed only in the
products of tests conducted above
600°C. K- was not identified.
Effect of Residence Time
The thermal breakdown of Kepone
is pronounced in the 420° to 440 °C
region. Consequently, this temper-
ature region was selected for an
experimental evaluation of the
effect of residence time on the
destruction of Kepone. Accordingly,
three separate tests were conducted
with 40 ug samples of_Kepone at a
carefully controlled T2 of 433°C.
The first test was conducted with a
residence time, tr, Of 1.79 sec.
The tr for the second test was 1.04
sec, while for the third test, tr
was 0.23 sec. These different resi-
dence times were obtained by
appropriately adjusting the air flow
control valve of the quartz tube
107
-------
apparatus. Chromatograms were ob-
tained from the captured effluents.
From these chromatographic data, the
concentrations of emerging Kepone
were plotted versus residence time,
as shown in Figure 5. It is evident
100
80
60
KEPONE
(WT %)
20
433°C
01 03 10 30 10
<«r>
RESIDENCE TIME, SEC
Figure 5. Effect of Residence Time
on Thermal Destruction of Kepone.
from this logarithmic plot that
residence time is a strong factor in
the destruction of Kepone at 433°C.
Comparison of the Three Pesticides
Studies similar to that done
with Kepone were performed with
Mirex and DDT. From compiled GC
data, thermal destruction plots were
prepared for Kepone, Mirex, and DDT.
Figure 6 shows a direct comparison
of the high-temperature destruction
of the three pesticides. Mirex is
the most thermally stable of these
three, with Kepone next, and DDT the
least stable.
Current Work
Past work at the University of
Dayton Research Institute(7'8) has
confirmed that there is a definite
destruction temperature/residence
time relationship in a nonflame mode
when an organic compound is sub-
jected to an oxidative thermal ex-
posure, it follows then that any
zoo
400 600
TEMPERATURE, »C
800
1000
Figure 6. Comparison of Thermal
Destruction of the Three Pesticides.
system designed to evaluate destruc-
tion characteristics must have the
capability of precise control and
measurement of these important
parameters.
The new system now being
developed will employ a two-stage
destruction setup, similar to that
used in our earlier work. However
a major improvement in the handling
and analysis of the decomposition
products will be accomplished by
incorporating into the system a gas
chromatograph-mass spectrometer-data
system (GC-MS-DS) as shown in
Figure 7.
As can be seen from the block
diagram, the in-line system is
closed; therefore, a complete
capture and subsequent analysis of
evolved products can be achieved.
The gas chromatograph shown just
upstream of the mass spectrometer
must of necessity be a complex
separatory device in order to re-
solve individual compounds from the
wide range of expected decomposition
products. High resolution gas
chromatography, employing glass open
tubular columns, will be used to
108
-------
HIGH TEMPERATURE TRANSFER
CAPTURE
OF
EFFLUENT
PRODUCTS
VJJ/7/J/i
r///////
CONTROLLED
HIGH
TEMPERATURE
EXPOSURE
SAMPLE
INSERTION
AND
VAPORIZATION
PRESSURE AND
FLOW REGULATION
COMPRESSED GAS
AND PURIFICATION
ANALYSIS OF EFFLUENT PRODUCTS
IN-LINE
HIGH RESOLUTION
GAS CHROMATOGRAPH
COUPLED
MASS
SPECTROMETER
ON-LINE
DATA SYSTEM
NIH-EPA-MSDC
DATA FILE
Figure 7. Thermal Decomposition Analytical System.
separate and analyze even the trace
level components present in the
effluent. The mass spectrometer
utilizes signals from an electron
impact source and will enhance the
overall analytical sensitivity of
the system as well as the overall
detectability of the various de-
composition products. The GC-MS is
interfaced with a dedicated mini-
computer which gathers, processes,
and stores the vast quantity of data
generated.
With the nonflame version of
this system, both residence time and
exposure temperature can be pre-
cisely controlled. Specifically,
the residence time is determined by
accurately measuring the internal
pressure, temperature, and the gas-
volume throughput for a narrow-bore
quartz tube'8'. The exposure
temperature is determined by
measuring the average ambient
temperature that surrounds the
quartz tube. Again, with this new
system, product analysis will serve
as the basis for evaluating the de-
gree of destruction of the hazardous
wastes at the different exposure
temperatures and residence times.
In our earlier program, the
prime objective was to determine
destruction temperature/residence
time conditions for pesticide
molecules. However, along with the
parent pesticide analyses, a limited
search and identification was made
of various decomposition and re-
combination products which were
heavier than ^150 molecular weight.
In some of our thermal destruction
tests, particularly at intermediate
temperatures, substances of higher
toxicity than the parent molecule
have been formed. Consequently,
with the new system, the entire
range of decomposition products will
109
-------
be captured and analyzed. This
complete spectrum of a sample's
decomposition products could be of
paramount importance to parties
responsible for the environmentally-
safe thermal disposal of hazardous
wastes.
Characteristics of the Thermal
Decomposition Analytical System
The thermal decomposition
analytical system is being designed
with two distinguishing features:
flexibility with respect to the
test variables of interest and
"quick-response" capability when
results are needed promptly. With
this system, the gathering of basic
data can be obtained under con-
trolled conditions in both a non-
flame and flame mode of thermal
decomposition. However, with the
built-in flexibility, parameters
which can be easily varied can also
be well controlled and accurately
measured.
There are numerous advantages in
adopting the closed system approach.
First, the closed system permits the
total collection of decomposition
products. Also, the analytical data
acquisition can be accomplished with
a minimum of error, as the system-
atic errors generally associated
with sample handling in a dis-
continuous technique will be
eliminated. Additionally, this
thermal decomposition analytical
system can be applied to a wide
range of organic hazardous mate-
rials. The system will have the
capability to examine such samples
as fire retardants, agricultural
chemicals, industrial wastes,
environmental contaminants, and a
wide variety of other organic waste
materials. It will be capable of
handling liquid, solid, gaseous,
and polymeric organic materials.
The expansion of analytical
capabilities with the new system
will allow a gathering of informa-
tion that was not previously avail-
able. For example, the effluent
resulting from a high-temperature
gas phase destruction of a multi-
component organic mixture would be
expected to differ from a simple
composite of the effluents from
the individual constituents.
Indeed, in some cases, entirely
different compounds would be formed.
Such important synergistic effects
can readily be examined with this
system.
ACKNOWLEDGEMENTS
The earlier research reported in
this paper was supported by the
Municipal Environmental Research
Laboratory (Grant No. R803540-01-0)
The current work is being cosponsor-
ed by the Municipal Environmental
Research Laboratory and the
Industrial Environmental Research
Laboratory (Grant No. R805117-01-Q)
REFERENCES
1. Kennard, E.H., "Kinetic Theory
of Gases," McGraw-Hill Book Co.
Inc., New York, 1938, Chap. I.* '
2. Novotny, M., Lee, M.L., and
Bartle, K.D., "Some Analytical
Aspects of the Chromatographic
Headspace Concentration Method
Using a Porous Polymer,"
Chromatographia, v. 7, p. 333
1974.
3. Zlatkis, A., Lichtenstein, H.A
and Tishbee, A., "Concentration'
and Analysis of Trace Volatile
Organics in Gases and Biological
Fluids with a New Adsorbent,"
Chromatographia, v. 6, p. 67,
J. y I 3 •
4. Bertsch, W., Chang, R.C., and
Zlatkis, A., "The Determination
of Organic Volatiles in Air
Pollution Studies: Character-
ization of Profiles," J.
Chromatog. Sci., v. 12, p. 175
1974.
5. Bellar, T.A., and Lichtenberg,
J.J., "The Determination of
Volatile Organic Compounds at the
ug/1 Level in Water by Gas
Chromatography," Environmental
Protection Agency Report, EPA-
670/4-74-009, November 1974.
110
-------
6. Leoni, V., Puccetti, G., and
Grella, A., "Preliminary Results
on the use of Tenax for the Ex-
traction of Pesticides and Poly-
nuclear Aromatic Hydrocarbons
from Surface and Drinking Waters
for Analytical Purposes," J.
Chromatog., v. 106, p. 119, 1975,
7. Duvall, D.S., and Rubey, W.A.,
"Laboratory Evaluation of High-
Temperature Destruction of Kepone
and Related Pesticides," Environ-
mental Protection Agency Report,
EPA-600/2-76-299, December 1976.
8. Duvall, D.S., and Rubey, W.A.,
"Laboratory Evaluation of High-
Temperature Destruction of Poly-
Chlorinated Biphenyls and Related
Compounds," Environmental Pro-
tection Agency Report, in press.
111
-------
STANDARDIZED METHODS FOR
SAMPLING AND ANALYSIS OF HAZARDOUS WASTES
R.D. Stephens, D.L. Storm, E.R. de Vera, B.P. Simmons and K.C. Ting
California Department of Health
2151 Berkeley Way
Berkeley, California 94704
ABSTRACT
The management and regulation of hazardous wastes must be based on a sound knowledge
of waste compositions. The complexity and diversity of industrial wastes necessitate the
availability of specialized sampling and analytical procedures. Methods have been
developed for safe, reproducible, and representative sampling of such wastes in a wide
variety of physical states, compositions, and locations.
Analysis of wastes has been developed as a four level scheme, which begins with
field characterization, followed by three levels of progressively more detailed and instru-
ment intensive laboratory analysis. Each level is designed to serve a particular infor-
mational need of a waste management or regulatory program. The first laboratory level
is designed to give rapid information on the general nature of waste composition and
hazardous properties. The second level consists of more detailed but qualitative pro-
cedures for both inorganic and organic components. The third level is designed to give
a more complete set of quantitative data.
Methods for sampling, containment, handling and custody of waste samples as well as
the individual analytical procedures and the rationale for using an analytical scheme
will be discussed.
INTRODUCTION
The proper management of, regulation
of, or resource recovery from industrial
wastes is not possible without an adequate
knowledge of the physical and chemical
characteristics of the waste. This infor-
mation comes from a detailed examination
of the waste producing process, or of the
waste itself. In either case, techniques
are required for sampling and analysis of
such wastes. As mixtures become more and
more complex in physical and chemical
characteristics, the problems of represen-
tative sampling and of analysis increase
dramatically.
Under a 1972 Industrial Waste Control
Act, the California Department of Health
(CDH) initiated a Hazardous Waste Manage-
ment Program in late 1973. By mid 1974,
a chemical support unit was formed to
address the problems of what were hazardous
industrial wastes in all of their diversi-
ties. Almost immediately it became clear
that existing methods of chemical sampling
and analysis were inadequate for this pur-
pose. Many specialized sampling methods
were practiced in various industry seg-
ments designed to meet the problem of
specific product lines. In such situa-
tions, the compositions are relatively
known, and constant which result in greatly
simplified sampling and analytical problems.
Additionally, much has been published on
sampling and analysis of water and waste-
water. Such procedures are designed
primarily for relative homogeneous, dilute
112
-------
aqueous media and are either inapplicable
or used with great difficulty with indus-
trial wastes.
When the National Resource Conserva-
tion and Recovery Act (PL 94-580) was
passed in October 1976, the law mandated
the U.S. Environmental Protection Agency
(EPA) to initiate a program to identify and
list hazardous wastes and criteria. One of
the needs for EPA to initiate the program
was the development of methodology for the
sampling of hazardous wastes. The EPA
awarded the California Department of Health
a research grant to assist in this effort.
Under this grant, the CDH embarked on
a study to develop a handbook for sampl-
ing hazardous wastes. The study was based
on some initial procedures and equipment
designs developed in field operations con-
ducted by the CDH and a literature search.
The guidelines used in choosing the
candidate procedures were the commercial
availability of the sampling equipment,
cost, simplicity in design and resistivity
to chemical reactions. The samplers
should be available in commercial supply
houses, or readily constructed and inex-
pensive enough to be discarded when not
easily cleaned or decontaminated.
The candidate methods and samplers
were subjected to laboratory and field
tests. The laboratory tests consisted of
sampling water with the liquid samplers.
The samplers were then observed for leak-
age and ease of transfer of the collected
samples. The samplers that passed the pre-
liminary screening were further tested
using multiphase liquid mixtures such as
water, used motor oil and insoluble
sludges. In the field tests, the liquid
and solid wastes samplers were tested by
sampling actual wastes in various types of
containers.
A workable scheme for analysis of
hazardous waste must follow the sampling
program. Analysis of such wastes presents
the problem that complete analysis of most
hazardous waste may be so expensive and
time consuming as to be Impractical, where-
as superficial general waste characteris-
tics may not be adequate for proper
management decisions. Consequently, a
multi-level analytical scheme has been
proposed with each level, progressively
more detailed and quantitative in scope.
Use of each level is keyed to informational
objectives of the analysis.
The analytical part of this project
is in its developmental phase and far from
complete. However, the analytical scheme
and several procedural elements of the
scheme will be presented.
SAMPLING
PURPOSES AND GENERAL CONSIDERATIONS
Sampling of hazardous wastes is con-
ducted for different purposes. In many
instances, it is performed to determine
compliance with existing regulations pro-
mulgated by the different regulatory
agencies. In some cases it is conducted
to obtain data for purposes of classifying,
treating, recovering, recycling or deter-
mining compatibility characteristics of
the wastes. Sampling is also conducted as
an important part of research activities.
In general, sampling of hazardous
wastes requires the collection of adequate
and representative samples of the body of
wastes. Sampling situations vary so widely
that no universal sampling procedure can
be recommended. Rather, several procedures
are outlined for sampling different types
of wastes in various states and containers.
These procedures require a plan of
action to maximize safety of sampling per-
sonnel, minimize sampling time and cost,
reduce errors in sampling and protect the
integrity of the samples after sampling.
The following steps are essential in this
plan of action:
1. Determine what should be sampled
2. Select the proper sampler
3. Design the proper sample container
and closure
4. Design an adequate sampling plan
which includes the following:
a) Choice of the proper sampling
point
b) Determination of the number of
samples to be taken
c) Determination of the volumes of
samples to be taken
5. Observe proper sampling precautions
6. Handle samples properly
7. Identify samples and protect from
tampering
8. Record all sample information in
113
-------
a field notebook
9. Fill out chain of custody record
10. Fill out the sample analysis request
sheet
11. Deliver or ship the samples to the
laboratory for analysis.
BACKGROUND INFORMATION ABOUT THE WASTE
Accurate background information about
the waste to be sampled is very important in
planning any sampling activity. The infor-
mation is used to determine the types of
protective sampling equipment to be used
and sampling precautions to be observed as
well as the types of samplers, sample con-
tainers, container closures, and preserva-
tives (when needed) required. Generally
the information about the waste determines
the kind of sampling scheme to be used.
Most often, the information about the
waste is incomplete. In these instances,
as much information as possible must be
obtained by examining any documentation
pertaining to the wastes such as the
hauler's manifest (i.e. Figure 1). When
documentation is not available, information
may be obtained from the generator, hauler,
disposer or processor. The information
obtained is checked for hazardous proper-
ties against standard reference sources by
field personnel.
SELECTION OF SAMPLER
Hazardous wastes are usually complex
multiphase mixtures of liquids, semisolids
or sludges or solids. The liquid and
seraisolid mixtures vary greatly in vis-
cosity, corrosivity, volatibility, explos-
ivity or flanmability. The solid wastes
can range from powders to granules to big
lumps. The wastes are contained in drums,
barrels, sacks, bins, vacuum trucks, ponds
and other containers. No single type of
sampler can therefore be used to collect
representative samples of the different
types of wastes. Table 1 lists most waste
types and the corresponding recommended
samplers to be used.
CONTAINERS
The most important factors to consider
when choosing containers for hazardous
waste samples are compatibility, resistance
to breakage and volume. Containers must
not melt, rupture or leak due to chemical
reactions with constituents of waste
samples. It is, therefore, important to
have some idea of the components of the
waste. The containers must have adequate
wall thickness to withstand ordinary handl-
ing during sample collection and transport
to the laboratory. Containers with wide
mouths are desirable to facilitate transfer
of samples from samplers into the contain-
ers. Also, the containers must have enough
capacity to contain the required volumes
of sample or the entire volumes of samples
contained in samplers. Table 2 indicates
the choice of containers available and
their usage.
SAMPLING PRECAUTIONS AND PROTECTIVE GEAR
Proper safety precautions must always
be observed when sampling hazardous wastes.
In all cases a person collecting a sample
must be aware that the waste can be a
strong sensitizer, corrosive, flammable
explosive, toxic and capable of releasing
extremely poisonous gases. The background
information about the waste should be
helpful in deciding the extent of sampling
safety precautions to be observed and in
the choice of protective equipment to be
used.
For full protection, the person col-
lecting the sample must use a self-contained
breathing apparatus, protective clothing,
hard hat, neoprene rubber gloves, goggles
and rubber boots.
A self-contained breathing apparatus
consists of an air-tight face mask and a
supply of air in pressure tank which is
equipped with a pressure regulator. Pro-
tective clothing consists of long-sleeved
neoprene rubber coat and pants, or long-
sleeved coverall and oil-and-acid proof
apron. In hot weather, the coverall-apron
combination might be preferable. Table 3
lists respiratory protective devices and
their usage. All equipment, except the
respirator, must be properly decontaminated
and washed between uses.
SAMPLING EQUIPMENT
Sampling of hazardous wastes requires
several different types of samplers. Some
sampling equipment is commercially avail-
able, while others must be fabricated.
This report will cover only the composite
liquid waste sampler in detail. Discussions
of the soil samplers, grain and other
samplers as well as pond and storage tank
114
-------
FIGURE 1. CALIFORNIA LIQUID WASTE HAULER RECORD
en
i i i iii i rn
SI ATI WATER MtOU»CK CONTMM. MMHO
STATC Ot»A*TNCNT Of HEALTH
PRODUCE* or VASTI (Mist be tilled by producer)
laea (flint or nea
nek up Mdroaot
Telephone »ia»aor!(
Oraor riacoa IT:
^_^_^_
' I I J T I «e»* <•"•««
•AULB Or VASTI (Nut be filled by hauler)
(Street) (City)
r.O. or Contract aa-.
Tyea of rroc*aa
•hlch rroeucoa llootoa:
DESCRIPTION Or MASTT (Nuat b* filled by producer)
Ch«ck cyf* «f
. Q *Cl« Ml
. D Alk«lt*«
. D
. D
:SS
tMUt «MtU
I. D Teek kottee ooelee
». D Ml
10. Q Drllllni *v4
11. D CeateMeeted Mil
12. Q raaaoif *
13. Q Lataa vai
1*. O I
19. O :
I uM
QotMr
4, ll
ph«fi«llc«, •olvenc* (Hit). ••!«!• (li«t)v
craanlo (Hit), cyanlei)
* [j>o«o * Qto«U Q(la««kl. Qcorroiln
TO
(Street)
rick He: ___
SUM Ueuie neat*
(if xr
., J. MM. I
: cla*elea» *il drlUloe Eoea »e.
Q.tk*>
D-..
The described vaete we* hauled by ee to the diapoaal
facility naaMd below and vaa accepted.
I certify (or declare) under penalty
of perjury that the foregoing la true
and correct. ^^^^^^_^^^_
DISPOSER or VASTI (Mutt be filled by diapoaer)
stt< Aelnu:
The hauler above delivered the deeerlbed Mate to tin* diapoaal facility and
it »ae an acceptable eaterial under the terna of noca requirenenta, state
Department of Health regulation*, and local reetrictiona.
Quantity eeaearei at .Ita (If aeallcaaU): *t*t* (to (If *ny):
aaealla*. Metkae(l):
PI reeewary
Qtroataeet (•feclfy):.
Q«eoooll (aaeclfy): n»eaa
iltatl«n).Cfl&« No.
lutlev arecl»ltetl»n)-c«<« «o.
lll Qltjactlo «11 I—T—I
I I 1
If eaata 1* WI4 for olaeeaal *l**ueai* i
Diapoaal Pate;
city fliul Iccatlx:
1 certify (or declare) under penalty
of perjury that th* foregoing i« true
and correct.
Signature of authorized agent and tltl*
The alt* operator (hall aubeit < legible copy of each coeplated (ecord to the
State Department of Health with nunthly fee report*.
The >«t« i- a««crib»
-------
TABLE 1. CHOICE OF SAMPLERS
Waste Type
Liquids, sludges and
slurries in drums,
vacuum trucks,
barrels and similar
containers
Sampler
Coliwasa
a) Plastic
Liquids and sludges
in ponds, pits or
lagoons
Powdered or gran-
ular solids in
bags, drums,
barrels and similar
containers
Dry wastes in
shallow containers
and surface eoil
Waste piles
Soil deeper than
3 inches
Wastes in storage
tanks
b) Glass
Pond
a) Grain
sampler
b) Sample
trier
Trowel
or
scoop
Waste pile
sampler
a) Soil auger
b) Veihmeyer
sampler
Weighted bottle
sampler
Limitations
Not for containers
> 5 ft. deep. Not
for wastes contain-
ing ketones, nitro-
benzene, dimethyl-
formamide mesityl-
oxlde or tetra-
hydrofuran 1»2
Not for wastes con-
taining hydrofluoric
acid and concentrated
alkali solution
Cannot be used to collect
samples beyond 12 ft.
Dip and retrieve
sampler slowly to avoid
bending the tubular
aluminum handle
Limited application for
sampling moist and sticky
solids and when the dia-
meter of the solids is
greater than V
May incur difficulty in
retaining core sample of
very dry granular mat-
erials during sampling
Not applicable to sampling
deeper than 3 inches.
Difficult to obtain
reproducible mass of
samples.
Not applicable to sampling
solid wastes with dimensions
greater than V'the diameter
of the sampling tube.
Does not collect undisturbed
core sample.
Difficult to use on stony
or rocky or very wet soil.
May be difficult to use on
very viscous liquids.
116
-------
TABLE 2. CHOICE OF SAMPLE CONTAINER AND CLOSURE
Waste Types
All types except pesticide,
chlorinated hydrocarbons
and hydrocarbons
All types except hydrofluoric
acid, concentrated alkali
Photosensitive
wastes
Container and Closure
Linear polyethylene bottles3,
wide mouth with plastic cap,
1000 ml and 2000 ml
Glass bottles^5, wide mouth
with bakelite cap and teflon
film linerc, 1000 ml and 2000 ml
Amber plastic or brown glassd
bottle, 1000 ml and 2000 ml
aNalgene, Cat. Nos. 2104-0032 and 2120-0005 or equivalent
bScientific Products, Cat. Nos. 87519-32 and B7519-64 or equivalent
cAvailable from Scientific Specialties, P.O. Box 352, Randallstown, MD
dScientific Products, Cat. Nos. B7528-050 and 7528-2L or equivalent
TABLE 3. SELECTION OF RESPIRATORY PROTECTIVE DEVICE
Hazard
Oxygen deficiency
Gaseous contaminant:
Immediately dangerous
to life
Not immediately
dangerous to life
Particulate contaminant
Combination of gaseous and
particulate contaminants
immediately danerous to
life
Not immediately
dangerous to life
Respirator
Self-contained breathing apparatus
Hose mask with blower
Self-contained breathing apparatus
Hose mask with blower
Gas mask
Air-line respirator
Hose mask without blower
Chemical-cartridge respirator
Dust, mist, or fume respirator
Air-line resporator
Abrasive-blasting respirator
Self-contained breathing apparatus
Hose mask with blower
Gas mask with special filter
Air-line respirator
Hose mask without blower
Chemical-cartridge respirator
with special filter
Source: "Respiratory Protective Equipment", Bulletin 226,
Safety in Industry, Environmental and Chemical
Hazards, No. 3, U. S. Dept. of Labor, Bureau of
Labor Standards.
117
-------
samplers are given in the Sampling Hand-
book.
COMPOSITE LIQUID WASTE SAMPLER (COLIWASA,
FIGURE 2) DESCRIPTION
The Coliwasa is primarily applicable
to sampling containerized liquid wastes
such as those contained in vacuum trucks,
barrels, drums, or similar containers. Its
design permits the representative sampling
of multiphase wastes of a wide range of
viscosity, corrosivity, volatility and
solid contents. It is simple and easy to
use allowing rapid collection of samples
thus minimizing the exposure of the sample
collector to potential hazards from the
waste during sampling. The sampler is not
commercially available, but it is relatively
easy and inexpensive to fabricate.
The first model of the Coliwasa was
designed and fabricated in the California
Department of Health.3 This sampler was
previously used in the collection of liquid
waste samples in connection with the imple-
mentation of the hazardous wastes program of
the Department. The sampler was also used
in a case study jointly conducted for the
U.S. Environmental Protection Agency by
the University of Southern California and
the Department.^ This first model of the
Coliwasa underwent a number of refinements
and testing. The improved model of the
Coliwasa is shown in Figure 2. The main
Tapered
stopper
T-handle
Locking
block
Stopper rod. PVC,
3/8* O.D.
Pipe, PVC, l-5/8"l.D.,
1-7/8" O.D.
Stopper, neoprene,
' #9 with 3/8Tf SS
or PVC nut & washer
SAMPLING POSITION
CLOSE POSITION
FIGURE 2. COMPOSITE LIQUID WASTE SAMPLER (COLIWASA)
118
-------
parts of the sampler consist of a positive
quick-engaging closing and locking mechan-
ism and a sampling tube.
Plastic and glass types of the Coliwasa
have been constructed. The sampling tube
and stopper rod of the plastic Coliwasa
are made of translucent polyvinyl chloride
(PVC) plastic while that of the glass
Coliwasa are made of borosilicate glass and
teflon plastic, respectively. The materials
for constructing the two types of the Coli-
wasa are available at hardware stores
and supply houses.
SAMPLING PROCEDURES
The two most common containers en-
countered in hazardous waste sampling are
the 55 gallon drum and the vacuum truck.
There are several other sampling situations
encountered which will not be covered here,
but are discussed in detail in a. manual soon
to be issued by EPA. Procedures for use
of the Coliwasa have already been described.
The prime problem confronting the field
personnel is access to the hazardous waste.
Sampling a vacuum truck requires the
person collecting the sample to climb onto
the truck and walk along a narrow catwalk.
In some trucks, it requires climbing access
rungs to the tank hatch. These situations
present accessability problem to the sample
collector who in addition has to wear
usually full protective sampling gear. Pre-
ferably, two persons should perform the
sampling; one person to do the actual
sampling and the other to hand the sampling
device and stands ready with sample container
and to aid in any problems. The sample
collector positions himself to collect
samples only after the truck driver has
opened the tank hatch.
The Coliwasa was specifically designed
to allow for access to drummed waste through
a bung hole. However, sampling of drums
requries extreme caution for they are fre-
quently corroded and/or under pressure. In
addition, many extremely hazardous materials
are stored in drums. Opening the bung of
a drum can produce a spark that might deto-
nate an explosive gas mixture in the drum.
This situation is difficult to predict and
must be taken into consideration every time
a drum is opened. The use of full protec-
tive sampling equipment can not be over-
emphasized when sampling a drum.
HAZARDOUS WASTE ANALYSIS
The analysis of hazardous wastes pre-
sents unique problems for the complexity
of wastes is generally inversely propor-
tional to their value and directly propor-
tional to the cost of analysis. However,
public health and environmental impacts
are unrelated to value. Consequently any
analytical scheme must be flexible and
allow for general characterization of
wastes as well as quantitative analysis.
A four level analytical scheme has been
proposed to provide this flexibility. The
primary elements of the scheme are shown
in Figure 3.
Field Tests Phase is intended for
field assessment of waste physical and
chemical properties. The primary objec-
tives of this phase are two fold. They
are to avoid immediate public health and
environmental damage due to fire, explosion,
or release of toxic gases. In addition,
the tests are designed to give an approxi-
mate verification to composition informa-
tion. Even though several of these tests
are common, well characterized procedures,
difficulties arise due to interferences
from oily phases, solid phases, and excess-
ive temperatures. Details of these pro-
cedures are discussed in the forthcoming
handbook.
Phase I (Initial Laboratory Assessment
and Sample Preparation). The objective of
the initial laboratory phase is for con-
firmation of the field data and to provide
more accurate physical property data pos-
sible under controlled laboratory conditions.
This phase uses 105°C residue and 550°C
residue tests to assess approximate levels
of dissolved inorganic and organic compo-
nents. The final step in this section
involves high speed centrifugation (10-15K
rpm) to separate the organic aqueous and
solid phases.
PHASE II (LABORATORY SCREENING TEST)
The objective of this phase of the
scheme is to provide for relatively rapid,
inexpensive means for qualitative analysis
of the separated phases. Emphasis is placed
on noninstrumental methods where possible.
A key element of this phase of the testing
will involve the use of thin layer chroma-
tography (TLC). This method is particularly
valuable on wastes for which little back-
ground information is available. Procedures
119
-------
FIGURE 3. ANALYSIS SCHEME FOR INDUSTRIAL WASTES
SAMPLE
Ni
O
Odor
Color
Field Tests Test Phase pH
Test Phase 1 Biological Test
Detection tubes for toxic gas
Temperature
Radioactivity
Flammability
Physical Tests Chemical Tests
Laboratory Initial Bioassay for toxicity Flash point Tot
Assessment and Explosivity Tot.
Sample Preparation Conductivity Tot
Phase separation by
centrifugation — .
al acidity/alkalinity
al fixed residue at 550°C
r
i
Test Phase II Aqueous Phase
Laboratory
Screening TLC Ion electrode
Tests
Test Phase III Wet chemistry
AA
Laboratory X-ray fluorescence
Confirmation HPLC
Tests GC
GC/MS
1C
Organic Phase
TLC
Solid Phase
1 1
Aqueous Solution Organic Solvent Solution
\
Infrared Wet chemistry
UV/VIS AA
GC GC
GC/MS GC/MS
HPLC HPLC
X-ray fluoresc
1C
GC/MS
HPLC
UV/VIS
-------
being proposed will involve selected banks
of eluting solvent systems, stationary
phases, and developing sprays which will
identify organic functional groups and
chemical classes as well as certain sus-
pected and more commonly occuring standard
compounds. Additionally, TLC, will provide
for qualitative identification of a wide
variety of inorganic compounds.
PHASE III (LABORATORY QUANTITATIVE ANALYSIS)
This phase of the scheme is a compila-
tion of the state of the art instrumental
methods for analysis of complex mixtures of
organic and inorganic systems. Primary
effort to date has been in the development
of infrared, atomic absorption (AA) and
x_ray fluorescence (XRF) spectroscopic
methods. A new and powerful tool for the
analysis of complex mixtures of metals,
both solid and solution, is the polarized
Zeeman AA spectrometer. The special fea-
ture of this instrument allows for auto-
matic and effective correction for sample
matrix effects and specific interferences.
The use of the polarized Zeeman spectro-
meter has allowed for rapid analysis of
waste samples directly, without sample pre-
paration for a large number of metals.
Another powerful tool for inorganic analysis
is the XRF. XRF has proven particularly
useful on wastes from unknown sources such
as spills, soil samples, and illegal dump-
ing. The method allows for simultaneous
analysis of 40 elements with little or no
sample preparation. New XRF methods cur-
rently under development will make possible
direct 40 element analysis of any homogene-
ous aqueous or organic liquid. It is
recognized that XRF requires specialized
and expensive instrumentation. However,
the rapidity with which complex samples can
be analyzed does justify its use in areas
where large amounts of industrial wastes
are produced.
The next one to two years shall bring
considerable development and refinement of
the proposed analytical methods. Major
effort will be placed toward the analysis
of complex organic mixtures, particularly
for those materials which because of their
toxicity or persistence pose significant
public health and environmental health
hazards.
ACKNOWLEDGEMENT
The authors would like to fully acknow-
ledge the valuable assistance of many co-
workers at the California Department of
Health as well as the financial assistance
of the U.S. Environmental Protection Agency,
Richard A. Games, Project Officer.
REFERENCES
1. Merck Index, 9th Edition, Merck & Co.,
Rahway, N.J. (1968)
2. Chemical Resistance of Excelon R-400,
Thermoplastic Process, Inc., Stirling,
N.J. 07980
3. Stephens, R.D., Hazardous Sampling, Resi-
dual Management By Land Disposal, Pro-
ceedings of the Hazardous Waste Research
Symposium, University of Arizona, Tucson,
Ariz. 85721 (Feb. 2-4, 1976).
4. Eichenberger, B., Edwards, J.R., Chen,
K.Y., and Stephens, R.D., A Case Study
of Hazardous Wastes Input Into Class I
Landfills, Report to EPA, Solid and
Hazardous Waste Research Division, Cin-
cinnati, Ohio 45268 (1977)
121
-------
CHARACTERIZING INPUT TO HAZARDOUS WASTE LANDFILLS
Richard A. Carnes
U.S. Environmental Protection Agency
26 West St. Clair Street
Cincinnati, Ohio 45268
ABSTRACT
The average concentration, estimated daily deposition, and partitioning of 17 metal
species in hazardous waste landfills in the greater Los Angeles, California area were
studied. These sites received a combined estimated 2.3 million liters per day of waste
classified as hazardous requiring disposal at Class I sites. A total of 320 samples was
collected and consolidated to 99 representing 17 industry types.
Mass deposition rates were determined by calculating the average concentration of the
metal species in conjunction with approximate daily volume received at the site. Approxi-
mately 50 percent of the total volume of hazardous wastes sampled was generated by the
petroleum industry. About 35 percent of the volume was equally divided between the chem-
ical and industrial cleaning industries. Approximately 70 percent of the total volume
was in the aqueous phase while 8 percent consisted of an organic liquid phase. Some recom-
mendations concerning the interactions of hazardous wastes with differing soil types and
the resulting effect on leachate formation and potential migration of the toxic metal
species were proposed especially regarding the need for further research.
INTRODUCTION
In recent years, controls which further
restrict emissions to air and discharges to
water, as well as open dumping have increased
the pressure to dispose of waste on land.
Approximately 10 million tons of nonradio-
active hazardous waste are generated annual-
ly with the fPtal increasing by about 5 to
10$ per year!*'About 40% of these wastes by
weight are inorganic, whereas 60% are organ-
ic; some 90% accurs as liquid or semi-liquid.
Over 70% of all hazardous wastes are gener-
ated in the mid-Atlantic, Great Lakes, and
Gulf Coast areas of the United States.(2)
Although sanitary landfill ing has been de-
veloped as a means of disposing of waste
material in general, these sites were not
specifically designed to handle hazardous
wastes. Many hazardous wastes are, never
the less, disposed of at conventional land-
fill sites. The problems associated with
improper land disposal of these wastes, un-
like those of air and water pollution, have
not been widely recognized by the public.
However, as bans on ocean dumping become
more widespread, and pesticide cancellations
increase the tonnages of hazardous wastes
which must be removed from the environment
the concern for safe disposal also grows.
The lack of reliable information has gener-
ated concern about the practice of confined
land disposal of hazardous wastes. Few data
are available on the concentrations of trace
metals, chlorinated hydrocarbons, pesticides
and other toxic constituents of these wastes'
The more comprehensive version of the study
reported on here may be obtained by contact-
ing the author.
EXPERIMENTAL
Liquid wastes were collected at five
major Class I landfills in the Los Angeles
basin: BKK in West Covina; Pacific Ocean
Disposal (POD) in Wilmington; Operating in-
dustries (01) in Monterey Park; Calabasas
(CB); and Palos Verdes (PV).
In any program of sampling hazardous
wastes, the following parameters must be
considered:
1. Phase complexity: Techniques must yield
a representative sample, because hazard-
122
-------
ous wastes appear in all phases: solid,
aqueous, and organic liquid.
2. Access to waste: Hazardous wastes are
usually contained in ponds, vacuum trucks,
barrels, etc.; sampling procedures must
be amenable to all of these.
3. Chemical Reactivity: Many wastes are
either highly corrosive, strong oxidizers,
or, because of their physical nature,
very hard on sampling equipment. These
features place severe demands on equip-
ment design.
4. Safety: Because of the undefined nature
of most hazardous wastes, extensive pre-
cautions must be taken by sampling per-
sonnel .
5. Sampling Containment and Preservation:
The containment and preservation of cor-
rosive, toxic, or volatile samples pres-
ent many problems to sampling and analyt-
ical personnel.
SAMPLING HAZARDOUS WASTE
1. Sampling Teams: Each team consisted of
three people. Two functioned primary
as collectors and one as record keeper.
Based on a review of manifests over a 1
year period, it was estimated that sam-
ples collected during the 2 week survey
by two teams, represented 90% of the
waste types received at the sites over a
1 year period. The remaining 10%, mainly
seasonal types of disposal, could not be
sampled during the course of this study.
2. Sampling Equipment Inventory: Table 1
lists the equipment requirements for
each sampling team.
3. Sampling Procedures: The object of the
program was to obtain representative sam-
ples of all liquid, sludge, and solid
hazardous waste delivered to a disposal
site. The following were stepwise pro-
cedures used for sampling incoming
traffic:
• Sampling personnel, intercepted waste
truck, requested copy of manifest, re-
corded manifest number and appropriate
information on waste sampling form
(Appendix A), and requested driver to
pull off main traffic flow and opened
inspection hatch.
Table 1. Sampling Equlpment
Equipment List
Sampling:
1. Three (3) sample tubes
2. Sample bottles
3. Funnels
4. Tube Cleaners
5. Disposable wipers
6. Drums, 1-55 gal.; 3-5 gal. pails
7. 5 gal.-1,1,1, Trichloro Ethane or
Other Suitable Solvent
8. Spares: tubes
9. Spares: rods
10. Spares: stoppers
11. Ink pens: Mark-on-anything
12. Tool Kit
13. Clip Board
14. Analytical forms
15. First Aid Kit
Personnel: each team
3 protective suits
2 hard hats with shields
boots for each sampler/not necessary
for record keeper
2 respirators
4 pair gloves
2 pair goggles
• Sampling personnel, properly attired
with safety equipment, climbed onto
truck and lowered the specially designed
composite liquid waste sampler (Coli-
wasa) (Figure 1) into the liquid to ob-
tain a representative cross section of
hazardous waste. The waste samples
were then transferred directly to a
one-liter polyethylene container.
• Containers were appropriately numbered
and stored on the sampling truck approx-
imately 4 days before transfer to 4°C
storage.
A schematic of the sample collection, prepara-
tion, and analysis, is presented in Figure 2.
Analytical Methods
1. General Parameters: Solids content (to-
tal, soluble and insoluble), pH, acidity,
and alkalinity were measured in accordance
with procedures described in Standard
Methods. 14th Edition.(3) Mineral acidity
and total alkalinity were determined by
123
-------
potentiometric titration to pH 4. Flash
point was measured using ASTM Standard
Method, D1310-72.(4) Phase distribution
(organic, aqueous, solid) was determined
by centrifugation for 10 minutes at
12,000 RPM; the separated phases were
removed and recorded as a weight percent.
Volatile organics were identified by
distillation of the organic phase at
Volume = .4U(.43 qt.)/ft. of depth
•3/8"PVCrod
72"
60"
.1 7/8" - outer dimensions
1 5/8" - inner dimensions
-Class 200 PVC pipe
FIGURE 1.
I No. 9Vi neoprene stopper
3/8" S.S. nut and washer
COMPOSITE LIQUID WASTE SAMPLER
(CoLiWaSa)
95°C with a Kontes microsteam distilla-
tion unit.
2. Metal Species: Sample preparation is pre-
sented in Appendix B. Seventeen metal
species were analyzed; Be, Na, Ag, Mg, K,
Ca, V, Cr, Mn, Fe, Ni, Cu, Zn, As, Cd, Ba,
and Pb. All analyses were performed on a
double beam atomic absorption spectropho-
tometer equipped with a HGA 2100 graphite
furnace. Na and K were analyzed using
emission flame photometry. The graphite
furnace was employed in the analysis of
As and Cd, toxic elements in low concen-
tration, and Be, V, and Ba which require
special fuel when analyzed by flame meth-
ods. More details on the analytical por-
tion of the project are presented in the
comprehensive report mentioned above.
RESULTS AND DISCUSSION
A total of 320 waste samples were col-
lected, and duplicates (those originating
from the same source were identified and
sorted so that one representative of each
duplicate set would be analyzed.
The calculated mass deposition rates
are based on the total volume inputs esti-
mated by the Class I site operators. The
data indicate that the highest average daily
mass deposition of metal species is gener-
ated by the following industries:
Petroleum
Chemical
As, Be, Ca, Cd, K, Mg
As, Na
Industry Cleaning Pb
Metal Cr, Zn
Misc./Unknown
Ba, Cu, Fe, Mn, Ni
The volume flow and concentration of
soluble toxic metals present a potential
threat to the quality of ground and surface
water supplies. Physical and chemical prop-
erties of the soil which may be affected in-
clude attenuation and field capacity, floc-
culation or dispersion of clay particles,
hydraulic conductivity, infiltration rates;
toxic elements may also accumulate.
Leachate will not be produced until a
sizeable portion of the landfill has reached
field capacity. However, some leachate may
be produced immediately after initially wet
waste is compacted and disposed of or liquid
124
-------
SAMPLE COLLECTION
SAMPLE PREPARATION
, DIGES-
TION
*
DILUTION
, FILTRA-
TION
TO BERKLY STATE DEPARTMENT
OF HEALTH
HIGH SPEED
| CENTRIFUGE
OTHER TEST
'•'I PREPARATIONS
fi-
SAMPLE ANALYSIS
TOTAL TRACE
METAL CONC.
ON TRUCK
DIGESTION
DILUTION
FACTOR
CONC. OF
DISSOLVED
TRACE
METALS
ON TRUCK
RELATIVE VOLUME
'MEASUREMENTS
pH
ORGANICS
FLASHPOINT
ACID/BASE
EQUIVALENTS
FIGURE 2. SCHEMATIC OF SAMPLE HANDLING
is channeled through the fills. If concen-
trations of hazardous wastes are high in the
leachate, the soil attenuation capacity may
be reached relatively quickly. The cation
exchange capacity will vary with the nature
and concentration of ions in solution. Clay
particles may either flocculate or disperse
depending on their state of hydration and the
composition of their exchangeable cations.
Dispersion usually occurs with monovalent
and highly hydrated cations, for example, Na.
Conversely, flocculation occurs at high sol-
ute concentrations and/or in the presence of
divalent and trivalent cations.\*> Because
of various chemical, physical, and biological
processes, hydraulic conductivity may change
as liquid permeates the soil and flows throuah
it. The composition of the exchangeable-ion
complex when, for instance, leachate enters
soil changes because the concentration of
solutes differs from that of solution; this
can greatly alter hydraulic conductivity.^-8)
Further studies on the interactions of haz-
ardous wastes and soil, particularly on the
effect upon hydraulic ocnductivity and par-
ticle size distribution are called for.
From the analysis, we can project the
estimated daily deposition rate according
to species as follows:
Total : Na>Fe>Ca>Zn>K>Mg>Cu>Cr>Ni>Pb>Ba>
Sol ubl e:
Solid:
Na> Fe> Ca>Cu>Zn> K>Cr>Mg>Ni> Pb>Mn>
Ba>V>Cd>As>Ag>Be
Na>Ca>Fe>Mg>Zn>K>Pb>Cu>Cr>Ba>Ni>
Mn>V>Cd>As>Be>Ag
Approximately one-half of the total
volume of hazardous wastes was generated by
the petroleum industry. About 35% was con-
tributed by the chemical and industrial clean-
ing industries. The total volume contribu-
tion is shown in- Table 2.
About 70% of the estimated total volume
input of 2.3 X 16° a/day is in the aqueous
phase and 8% consists of an organic liquid
phase, 16% of which is volatile (B. Pt. less
than 95°C).
125
-------
Table 2. Volume Input Generated
by General Industry Types
Industry Type
% Total Volume
Petroleum
Chemical
Metal
Food
Industrial Cleaning
Misc./Unknown
45.9
17.9
6.0
3.6
17.4
9.2
The results reported on here represent
an overview of the more comprehensive report,
"A Case Study of Hazardous Wastes Input Into
Class I Landfills." This work was sponsored
by the USEPA under Research Grant R803813 to
the University of Southern California.
CONCLUSIONS
The following conclusions are based on
results of sampling and analysis of hazardous
wastes received at the five Class I landfills.
1. Average concentrations, estimated average
daily inputs, and the partitioning of
metal species (soluble and solid phase)
varied widely for the five sites.
2. The data collected for individual sites,
together with the estimated daily volume
input, provide the approximation of mass
deposition rates for selected metal spe-
cies at a specific site.
3. Results show that wastes containing cop-
per, chromium and zinc pose the greatest
threat to possible groundwater and surface
water contamination when the following
factors are considered: (a) total mass
deposition, (b) weight percent in the
soluble phase, and (c) maximum concentra-
tion levels of 14,000-20,000 mg/1.
4. About 8% of total volume input was com-
prised of organic liquid wastes and 162S
of the organic phase had boiling points
of less than 95°C and flash points as low
as 17°C.
5. The combined results for the five Class I
sites is considered a first approximation
of the hazardous waste streams generated
in the greater Los Angeles area.
6. Some manifests collected and evaluated
against the analytical results are highly
suspicious. The results of certain metal
analysis are incompatible with those of
waste stream identification. Continued
refinement of the data along with adequate
support at both State and Federal levels
will prove the manifest system to be quite
workable and informative to regulatory
agencies.
7. The volume flow, concentration, and mass
deposition rate of the toxic metal species
identified from this study should prove
useful in the preliminary selection of
required treatment processes and facili-
ties for hazardous wastes from industrial
activities.
REFERENCES
1. U.S. Environmental Protection Agency,
"Hazardous Wastes," Pub. SW-138, 1975
25pp.
2. Lehman, J.P., "Federal Program for Hazard-
ous Waste Management," Waste Age. September
1974, pp. 6-8 *- v er
3. American Public Health Association,
Standard Methods for the Examination of
Water and Wastewater, 14th Edition^
Washington, D.C. 1976
4. "Standard Method of Test for Flash Point
of Liquids by Tag Open-Cup Apparatus."
ASTM D1310-72, 1973, p.358
5. Jenny, H., and R.F. Reitemeier, "Ion Ex-
Change in Relation to the Stability of
Colloidal Systems," J. Physical ChemUi-».»
39, 1935, pp. 593-604 —«E£
6. Brooks, R.H. et al., "The Effect of Vari-
ous Exchangeable Cations upon the Physical
Condition of Soils," Soil Science Soc
Amer. Proc.. 2jD, 1956, pp. 325-327 ~
7. Quirk, J.P. and R.K. Schofield, "The
Effect of Electrolytic Concentration on
Soil Permeability," J-Spil Science 6
1955, pp. 163-170 ' ' -'
8. Reeve, R.C. et al., "A Comparison of the
Effects of Exchangeable Sodium and Potas-
sium upon the Physical Conditions of
Soils," Soil Science Soc. Amer. Proc 1ft
1954, pp. 130-132 —»
126
-------
APPENDIX A
HAZARDOUS WASTE UNIT
SURVEILLANCE FORM
Sample 01 122 Lab No. Sampling Date 9/12/75
Manifest No. 1411 Time 11:35
Producer Beth. Stl. Corp.
Producer's Address 3300 E. Slauson Ave., Vernon
Hauler Capri Pumping Service
Hauler's Address 3128 Whittier Blvd., Los Angeles, CA
Process Type Steelmaking Waste Type Mud and Water
Chemical Components Concentration Volume
upper lower (units)
EE-203 85% 60%
AL 203 5% 2%
Grease 5% 2%
SIO-2 2% 1%
Brief Physical Description Black Liquid
127
-------
APPENDIX B
SAMPLE PREPARATION
Preparation for filtration and digestion included thorough washing of all labware which
would come in contact with samples. The following washing procedure was used: Scrubbing
with a brush using detergent and industrial water. Rinsing three times with deionized
water. Drying in low temperature oven. Storing in washed polyethylene bags.
Samples were stored at 4°C within the original one-liter plastic containers. After one
week, about one-third of each sample was poured into identical, labeled containers and sent
to the State Department of Health for analysis of organics and determination of percent
liquid and solid volumes. Samples were returned to 4°C storage until aliquots were taken
for filtration and digestion. Samples were kept at room temperature for approximately two
days during this process. One aliquot was poured first through a #1 Whatman Filter and
then passed through a 0.1 nm millipore filter into a sample bottle. Another aliquot (5 ml)
was placed in a teflon beaker and digested with HN03, HF and HClO/j. The resulting liquid
was centrifuged, poured into sample bottles, and diluted.
Sample Analysis
The partitioning of trace metals between those in the soluble phase and those associated
with nonfilterable solids was attained by analyzing the filtrates (0.1 nm) and the acid
digested total sample. Nonfilterable solids are then determined by subtracting dissolved
trace metal concentrations from total concentrations.
128
-------
"CO-DISPCSAL OF INDUSTRIAL AND MUNICIPAL WASTE IN A LANDFILL"
By:
Joseph T. Swartzbaugh, Ph.D.
Robert L. Hentrich, Jr.
Gretchen V. Sabel
SYSTEMS TECHNOLOGY CORPORATION
Xenia, Ohio
ABSTRACT
A research effort was initiated in late 1974 to study the effects of
the co-disposal of selected industrial wastes with municipal refuse in a
sanitary landfill environment. Observations were made utilizing 1.5 tonne
test cells designed to closely simulate sanitary landfill field conditions.
Six types of industrial wastes were mixed with municipal solid waste during
loading. The six types of industrial wastes were petroleum sludge, battery
production waste, electroplating waste, inorganic pigment waste, chlorine
production brine sludge, and a solvent based paint sludge. All. test cells
received infiltration water additions equivalent to 406 mm (16 inches) per
annum. Leachate samples were collected and analyzed on a monthly basis.
The results of these analyses over a two and one half year period were com-
pared with control cells containing only municipal refuse. The observations
made resulted in the identification of a number of potential problem areas
and recommendations for future research in this area.
Concurrent to this leachate study, the composition of gases generated
in simulated landfill environment were investigated. The results of the
gas composition study were noteworthy in that they indicated that methane
production in landfilled refuse varied in a cyclic manner rather than the
straight forward theoretical manner which was indicated in earlier studies.
This apparent cyclic production of methane was possibly caused by a cyclic
dominance of different types of microbial colonies in the landfill. The
state of flux of the microbial colonies varies on a cycle which is dependent
on several possible environmental factors.
INTRODUCTION
Solid Waste disposal has been
a problem for society since its
earliest beginnings. Until recently
the "out of sight, out of mind"
syndrome has been in effect. That
is, no one apparently cared what
happened to their waste once the
refuse hauler took it. With a
recent awakening of environmental
awareness, brought about by a real-
ization that our resources and land
areas are not infinite, the popula-
tion in general is more concerned
with the final disposal of their
waste.
In a non-industrial society
the magnitude of the solid waste
disposal problem was formerly a
function of the size of the popu-
lation. While this is still true
to a degree, the recent trend in
legislation as well as new innova-
tions and refinements of waste treat-
ment processes are bringing to the
129
-------
fore sewage sludge disposal problems
which were heretofore largely
ignored. Add to this the problem of
ultimate disposal of the vast amount
of industrial sludges, many of which
are of a protentially hazardous
nature, which are by-products of
modern technological advancement and
the situation has taken on an
entirely new dimension.
While there are many new treat-
ment processes either in use or
undergoing testing today, the oldest
form of solid waste disposal, namely
landfilling, is still the most
widely used method employed for the
ultimate disposal of wastes.
Congress recognized the need
for solutions to the problems of
modern solid waste disposal and
through various legislation, es-
pecially the Resource Conservation
and Recovery Act of 1976, has man-
dated that research be accomplished
to provide the information needed to
properly design new landfills and to
determine the environmental controls
necessary to protect man from him-
self.
The disposal of refuse through
the utilization of landfilling
techniques presents a number of
potential problem areas which need
to be resolved. These are as
follows:
1. The mechanics of simply
handling the vast amounts of solid
refuse generated by a modern soc-
iety.
2. The geographical location
of new landfill sites.
3. The needs of the design
engineer for useful data with which
to plant environmentally acceptable
landfill sites.
4. The infiltration of ground
v?ater resources through the leaching
processes.
5. The pollution of nearby
surface waters via runoff.
6. The generation of poten-
tially flammable and explosive
gaseous mixtures in a decomposing
landfill.
7. The landfilling of new,
potentially hazardous, industrial
wastes.
The disposal of refuse in a
landfill leads to a number of pro-
blem areas related to the design
and operation of the landfill site.
The major problems historically
associated with landfills have
always been primarily operational
in nature. However, society's
modern day capacity for generating
solid waste has made it necessary
to consider other problem areas
which heretofore were given little
attention. These problem areas
were discussed in the proceeding
paragraphs. In June 1974, SYSTECH
was contracted by the Solid and
Hazardous Waste Research Division
of the USEPA to investigate some
of the problem areas which must be
understood in order to properly
maintain and control the situation
which is occurring within the
landfill itself.
This paper concerns itself with
one phase of that overall study and
embraces the problem of co-disposinq
industrial wastes of various types
with municipal solid waste. For
the purposes of this study sewage
sludge is considered in this cate-
gory. In addition the following
types of bona fide industrial
sludges were considered.
1. Petroleum Sludge
2. Battery Production Waste
3. Electroplating Waste
4. Inorganic Pigment Waste
5. Chlorine Production Brine
Sludge
6. Solvent Based Paint Sludge
Due to the variety and number of
co-disposal systems involved in this
study replication of the test cells
was not possible. Therefore, it must
be noted that this is an observation
130
-------
study snd the recommendations at
this point must be limited to sug-
gestions of possible problem areas
and situations in which further in-
depth work is needed.
FACILITY DESCRIPTION
The experimental landfill test
cells employed for this study are
located at the USEPA Center Hill
Research Facility in Cincinnati,
Ohio. The total facility consists
of fifteen outside cells buried in
the ground, along with a leachate
collection well and an instrumenta-
tion building. An additional four
test cells are located in the
northeast corner of enclosed high
bay area directly adjacent to the
research facility. Of these nine-
teen cells, nine were loaded with
either co-disposed industrial waste
or co-disposed municipal sewage
sludge. It is these nine cells plus
control cells loaded with municipal
solid waste only which are the pri-
mary considerations of this paper.
The test cells proper are 1.8m
(6 ft.) in diameter and 3.6m C12 ft.)
deep. They are constructed of
4.76mm (3/16 in.) steel covered with
coal tar epoxy. The outside test
cells were placed on top of concrete
bases and the excavated area back-
filled with soil to approximately
0.3m (1 ft.) from the top of the
test cells. Several layers of
fiberglass cloth were placed at the
base of the cells and extended up
the interior walls to a height of
approximately 0.3m (1 ft.) and
epoxied in place in an attempt to
insure quantitative leachate and
gas collection. The four interior
cells have steel bases welded onto
the main tube of the cell.
Provisions for draining
leachate have been included in the
construction of the bases of all the
cells. The outside test cells have
a depression in the concrete base
of the test cell with piping to the
central leachate collection well.
The cells located in the enclosed
bay area are mounted on concrete
pedestals allowing the center
section of the base of the cells to
be exposed for leachate collection.
Water additions are made to the
cells by means of a distribution
system composed of perforated tube
formed into a circle. The circular
tubing is connected through the wall
of the test cells to a hose con-
nection where water is introduced
in accordance with a predetermined
rainfall application schedule. The
interior distribution tube is
located within the pea gravel layer
to aid in uniformly distributing the
water addition over the surface of
the clay cover.
Figure 1 shows a cross section
of a completed test cell. At the
base of the cell is a 150mm (6 in.)
layer of silica gravel. Above the
silica gravel were eight lifts of
solid waste, each 0.3m (1 ft.) in
thickness. Temperature probes are
located at the second, fourth, and
sixth lifts, while gas probes are
located at the second and sixth
lifts and in the pea gravel cover.
The observation well for lea-
chate collection is a well 4.8m
(16 ft.) deep and 1.8m (6 ft.) in
diameter with a scaling ladder
welded to the interior wall of the
cell for access purposes. All of
the leachate drains from the exter-
ior cell are connected to this
observation well and valved at this
central collection point.
TEST CELL CONTENTS
The purpose of the study is to
observe the potential effects of
the co-disposal of industrial wastes
with municipal solid wastes on the
landfill environment. The municipal
solid waste used in the study was
obtained from the City of Cincinnati,
Ohio. A brief description of each
of the industrial waste types mixed
with the MSW is given in the follow-
ing discussion (Table 1).
131
-------
CROSS SECTION OF A TYPICAL TEST CELL
T.C. PROBE
GAS PROBE
SETTLEMENT INDICATOR
SIGHT TUBE
WATER INPUT CONNECTOR
ADDITIONAL GAS PROBE
THIS LOCATION ON
CELLS #1,2,3,4,5,6,7,
8,9, 16 AND 17
.3m PEA GRAVEL
.3m CLAY
2.4m. MUNICIPAL REFUSE
Figure 1
150mm SILICA GRAVEL
Cross Section of a Typical Test Cell
132
-------
Table 1
Summary of Additive Materials Placed in
Test Cells
TEST
CELL
5
6
7
9
10
12
13
14
17
ADDITIVE
SEWAGE SLUDGE
SEWAGE SLUDGE
SEWAGE SLUDGE
REFINERY
SLUDGE
BATTERY
PRODUCTION
WASTE
ELECTROPLATING
WASTE
INORGANIC
PIGMENT WASTE
CHLORINE
PRODUCTION
SLUDGE
SOLVENT BASED
PAINT SLUDGE
AMOUNT ADDED PER LIFT ( kg , WET WEIGHT)
1
0
0
0
0
0
0
0
0
0
2
9.7
29.2
97.2
216.9
184.4
170.1
202.9
291.3
229.2
3
9.7
29.2
97.2
216.9
184.4
170.1
202.9
291.3
229.2
4
9.7
29.2
97.2
216.9
184.4
170.1
202.9
291.3
229.2
5
9.7
29.2
97.2
216.9
184.4
170.1
202.9
291.3
229.2
6
9.7
29.2
97.2
216.9
184.4
170.1
202.9
291.3
229.2
7
9.7
29.2
97.2
216.9
184.4
170.1
202.9
291.3
229.2
8
9.7
29.2
97.2
216.9
184.4
170.1
2023
291.3
229.2
MOISTURE
CONTENT
PERCENT BY
WET WEIGHT
88
88
88
79.0
89.2
79.5
51.7
24.1
24.7
WEIGHT ( kg)
WET
WEIGHT
68
204
680
1518
1291
1 191
1420
2039
1604
DRY
WEIGHT
8.20
24.5
81.6
319
152
244
686
1550
1210
In the primary treatment of
oily wastes the wastewater is sent
to a API separating tank, then to an
air flotation unit. All the oily
sludges are then sent to a holding
tcoik where chemicals are added to
removed most of the oil. The sludge
used in this project consists of the
bottoms from the holding tank, and
contains slightly more emulsified oil
than API bottoms sludge.
The battery production waste
was obtained from a manufacturer of
lead/acid storage batteries. This
waste is s. composite from all phases
of a manufacturing operation with
the exclusion of the battery plate
assembly section itself. The com-
posite processing wastes were neu-
tralized and discharged to a
settling pond. The sludge used in
this program was obtained from this
settling pond.
The electroplating waste was
obtained from a large plating firm
which employs the following plating
processes: chromium (Cr), nickel
(Ni), cadmium (Cd), copper (Cu),
iron (Fe), and zinc (Zn). After
employing various treatment pro-
cesses to neutralize and breakdown
the wastes the treated wastes are
pumped to a lagoon for settling.
The settled sludge used in this pro-
gram was a brown soupy material with
high moisture content.
The inorganic pigment waste v/as
obtained from a supplier whose main
product was titanium dioxide. The
wastewaters from the various ir.anufac-
turing operations in this plant are
treated simultaneously. The result-
ing settled sludge was dewatered.
The waste used in our program was
this dewatered sludge.
The chlorine production brine
sludge was obtained from a manufac-
turer which employed the mercury
cell technique for chlorine liber-
ation. The major portion of the
sludge used in the landfill program
(60 to 80 percent) was from the
brine saturator while the remainder
was from i.he clarifier after set-
tling. The sludge used appeared
somewhat moist, brown in color, and
very dense.
The solvent-based paint sludge
used was representative of a paint
133
-------
sludge produced in industries
where a water curtain is used to
capture overspray. The water was
pumped directly to a holding tank
for disposal with no pretreatraent.
The material contained organic
solvents and was extremely viscous,
its color dependent upon the pig-
ments present.
Some notable components of
these sludges were: copper (Cu),
iron (Fe), mercury (Hg) in the
refinery sludge; copper (Cu), iron
(Fe), cadmium (Cd), lead (Pb),
asbestos, tin (Sn), antimony (Sb) in
the battery production waste;
chromium (Cr), iron (Fe), arsenic
(As) , cadmium (Cd) , cyanide, asbestos
in the inorganic pigment waste; and
nickel (Ni), lead (Pb), chloride,
asbestos, mercury (Hg) in the
chlorine production brine sludge.
MOISTURE DATA
The field capacity of refuse is
its absorptive capacity for mois-
ture. Theoretically, once field
capacity is reached in a landfill
environment, the amount of water
infiltrating the landfill should
equal the quantity of leachate
existing the landfill.
Utilizing a definition as given
by Fenn et alT, "field capacity"
is the maximum moisture which a soil
(or solid waste) can retain in a
gravitational field without produc-
ing continuous downward percolation.
In i-he case of this study the
appearance of collectible leachate
volumes on a continuous basis was
indicative that field capacity had
been reached according to the defin-
ition above.
In order to evaluate potential
effects of the co-disposal of indus-
trial wastes in a municipal landfill,
moisture additions were made to the
simulators and were varied monthly
as one would expect the rainfall
pattern to be in the midwestern U.S.
The total moisture added to each cell
was equivalent to the infiltration of
406mm (16 in.) of water per annum.
The moisture retention capacity
of the test cells was determined by
the water balance method. This
method considers the initial moisture
content of the refuse, the moisture
content of the industrial wastes
added, the volume of water added to
the cell as part of the moisture
addition regiment, and the quantity
of leachate removed from the test
cell with time. Utilizing the above
parameters the water retained in the
cell was described as follows:
water retained (ml/kg) =
initial moisture content + water
added - leachate removed
kg solid waste (dry wgt.)
The test cells containing the
various co-disposed industrial
wastes exhibited a high initial mois-
ture content. A significant part of
this initial moisture content was
contributed by the industrial waste
additive. For example, the moisture
content of the petroleum sludge was
approximately 80 percent by wet
weight; the battery waste was approx-
imately 90 percent moisture; the
electroplating waste was approxi-
mately 80 percent moisture; the
inorganic pigment waste was approx-
imately 51 percent moisture. These
industrial waste sludges are typical
in that they have a high moisture
content and would be representative
of sludges obtained by a by-product
of a waste treatment operation
which did not employ mechanical
dewatering techniques. The result-
ing effects of adding the high
moisture content sludges to land-
filled municipal refuse would
appear to be a more rapid attain-
ment of apparent field capacity
(with the attendant onset of
leaching) than is the case with
municipal waste only. (Table 2)
This was true in all cases except
for the solvent based paint sludge
cell which began continuous
leaching at approximately the same
time at the MSW only cell. This
again was more or less expected as
the paint sludge contained about
25 percent moisture while the other
additives contained from two to
four times as much initial moisture.
134
-------
Table 2
Water Balance Summary
Test
Cell
2
5
6
7
9
10
12
13
14
17
18
19
Time To
Reach
F.C.*
In Months
16
16
16
16
13
1
5
7
5
16
11
13
Water Added
nun
516
516
516
516
368
74
184
331
134
516
331
368
1
1357
1357
1357
1357
969
194
485
872
485
1357
872
969
Average
Moisture
Retained
In Cell
Subsequent
To Reaching
F. C. (1)
x 2899
a ±80 (2.7%)
x 2565
a ±81 (3.2%)
x 2764 (3.1%)
a ±87
x 2596 (3.4%)
a ±89
x 3567 (2.2%)
a ±78
x 2554 (2.9%)
a ±75
x 2855 (5.2%)
a ±150
x 2628 (5.7%)
a ±151
x 2402 (4.5%)
a ±107
x 2930 (3.9%)
a ±115
x 2364 (7.2%)
a ±170
x 2452 (3.2%)
a ±78
Average
Moisture
Retained
(ml/kg
Solid Waste)
x 1688
o ±46 (2.7%)
x 1449
a ±46 (3.2%)
x 1745 (3.2%)
a ±55
x 1784 (3.4%)
a ±61
x 1480 (2.2%)
a ±32
x 1407 (2.9%)
o ±41
x 1200 (5.3%)
a ±63
x 1029 (5.7%)
o ±59
x 663 (4.5%)
o ±30
x 892 (3.9%)
o ±35
x 1146 (7.2%)
o ±82
x 1178 (3.2%)
a ±38
* Field capacity is defined as the maximum moisture content
which a soil (or solid waste) can retain in a gravitational
field without producing continuous downward percolation.
(Ref. I) Data is presented in terms of "field capacity"
meaning the initiation of continuous leachate production.
135
-------
LEACHATE DATA
In order to evaluate the
possible effects on leachate char-
acteristics by the co-disposal of
different industrial wastes with
municipal refuse, the following
industrial wastes were edded during
the loading sequence for the fol-
lowing cells:
Cells 5,6,7
Cell 9
Cell 10
Cell ]2
Cell 13
Cell 14
Cell 17
Sewage Sludge
Petroleum Sludge
Battery Production
Waste
Electroplating
Waste
Inorganic Pigment
Waste
Chlorine Production
Brine Sludge
Solvent Based Paint
Sludge
Cells 5, 6, 7, 12, 13, 14, and
]7 were loaded in November 1974.
The leachate data from these cells
are compared with that from cell 2
and cell 16 containing only muni-
cipal refuse which were also loaded
in November 1974. (Figure 2-14)
Cells 9 and 10 were loaded in April
1975 and rre compared to cells 18
and 19, containing municipal refuse
only, which vere also loaded in
April 1975. (Figures 15-19)
The data collected from the
landfill simulators containing
sewage sludge in addition to the
municipal refuse seem to indicate
that such disposal will not present
a major problem. This conclusion
is based on the fact that when com-
paring the cumulative amounts of
the various pollutants produced from
each cell, no generalized effect due
to the addition of varying amounts
of sewage sludge was noted. There
were some individual parametric
differences, however, examination of
the data traced these differences
primarily to sosne channeling effects
which occurred in cell 5 early in
the study.
The data garnered to date from
the cell containing the co-disposed
petroleum sludge indicates some
rather interesting phenomena.
petroleum sludge appears to inhibit
the leaching process. This is pos-
sibly due to a coating action where-
by the decomposition of the waste
and the solubilization of components
of the waste is inhibited. While it
is recognized that this data was
gathered from a single cell, without
replication, it is felt that this
phenomenon should be the subject for
a future in-depth study to investi-
gate the possibility of the coating
breaking down with the subsequent
production of a highly polluted
leachate stream.
Observations made of the data
generated by the cell containing co-
disposed battery production waste
indicate that in most cases the
leachate pollutant concentration was
greater than that found in the con-
trol cell containing municipal
refuse only. This is an indication
of potential problems in the land-
filling of such wastes and more in-
depth studies should be undertaken
before landfilling practices are
approved.
Data gathered from the cell
containing co-disposed electroplat-
ing waste indicate that there is
again an apparent enhancement effect
on leachate strength which is shown
by the increase in concentration of
those leachate parameters which are
indicative of the presence of elec-
troplating wastes. These parameters
are COD, total and dissolved solids,
chlorides, copper, chromium, and
nickel. The possible effects of the
co-disposal of this waste on the
surrounding ground water must be
seriously evaluated prior to dis-
posing of these wastes in a
municipal landfill.
The test cell containing
municipal refuse co-disposed with
inorganic pigment waste provided
data which, during the time period
covered by this paper, exhibited no
great differences between this cell
and the municipal refuse only cell
which was used as a control.
Chlorine brine sludge, on the
other hand, when co-disposed over
136
-------
10000
8000
s
^ 6000
E
4000
2000
I5OOO
CELL 2
CELL 5
CELL 6
CELL?
TOC
ALKALINITY HARDNESS
TOTAL
SOLIDS
DISSOLVED
SOLIDS
COD
Figure 2
MSW/Sewage Sludge Data
400 -
2OO -
CELL 2
CELL 5
CELL 6
CELL?
TKN
Figure 3
CHLORIDE NICKEL
MSW/Sewage Sludge Data
137
-------
100
80
M
i 60
40
20
CELL 2
CELL 5
CELL 6
CELL 7
PHOSPHORUS
IRON
COPPER
ZINC
Figure 4
MSW/Sewage Sludge Data
400
i
200
CADMIUM
Figure 5
CELL 2
CELL 5
CELL 6
CELL?
CHROMIUM
LEAD
MSW/Sewage Sludge Data
138
-------
10000
8OOO •
••
.
6000 -
4OOO -
2000 -
TOC
Figure 6
ALKALINITY
D CELL 2
• CELL 12
ED CELL 13
B CELL 14
3 CELL 17
HARDNESS
MSW/Industrial Waste Data
12000
10000
8000 -
42,800
p
6000 -
4000 -
2000
TOTAL SOLIDS DISSOLVED SOLIDS
N'SW/Industrial Waste Data
139
-------
11320
4OO
.*
g
200 -
CELL 2
CELL 12
CELL 13
CELL 14
CELL 17
MSW/Industrial Waste Data
247
PHOSPHORUS
Figure 9
IRON
CELL2
CELL 12
CELL 13
CELL 14
CELL 17
COPPER
ZINC
MSW/Industrial Waste Data
140
-------
9OO
600
300
CADMIUM
Figure 10
CHROMIUM
SCELL 2
CELL 12
H CELL 13
CELL 17
MSW/Industrial Waste Data
D CELL 16
53 CELL 17
IOOOO
8000
6000
4OOO
ZOOO
TOC ALKALINITY HARDNESS TOTAL DISSOLVED COD
SOLIDS SOLIDS
Figure 11
MSW/Industrial Waste Data
141
-------
400
•'•
200
CELL 16
CELL 17
TKN
Figure 12
CHLORIDE
NICKEL
MSW/Industrial Waste Data
IOO
8O
S
60
40
20
CELL 16
CELL 17
PHOSPHORUS ZINC COPPER IRON
Figure 13 MSW/Industrial Waste Data
142
-------
CELL 16
CELL 17
200
S
•N
9
100
CADMIUM CHROMIUM LEAD
Figure 14 MSW/Industrial Waste Daca
5000
4OOO -
3000 -
2OOO -
1000 -
TOC
Figure 15
ALKALINITY
CELL 9
CELL 10
CELL 18
CELL 19
r~
J
1
I
E
rfU r—
V////////////A
V
HARDNESS
MSW/Industrial Waste Data
143
-------
10000
8000
6000
4000
3000
500
1
*
^ CE
• CE
BCE
s^
ss
$s
$c
X*
§
§
^
§
b
COD TOTAL SOLIDS DISSOLVED SOLIDS
Figure 16 MSW/Industrial Waste Data
D CELL 9
^ CELL 10
H CELL 18
M CELL 19
CELL 9
CELL 10
400
w 3OO
200
100
VZWWZZZZZZZZ®
'
TKN
Figure 17
CHLORIDE
NICKEL
MSW/Industrial Waste Data
144
-------
IOO
20 -
PHOSPHORUS
Figure 18
CELL 9
CELL 10
CELL 18
CELL 19
500
ZINC COPPER IRON
MSW/Industrial Waste Data
D CELL 9
^ CELL 10
• CELL 18
11 CELL 19
400
300
200
100
V///////////////A
CADMIUM
Figure 19
CHROMIUM
LEAD
MSW/Industrial Waste Data
145
-------
the time period discussed in this
report, appears to be a serious
potential hazard to the surrounding
water environment. A comparison of
this cell with the control cell
containing municipal refuse only
indicates that in all parameters the
concentrations of leachate pollu-
tants are generally higher. Many
of these differences are quite
large. Serious consideration of the
situation should be made before
landfilling this waste.
The test cell containing the
solvent based paint sludge additive
(cell 17) when compared to the out-
side cell containing municipal
refuse only (cell 2), exhibited test
parameter concentrations generally
lower than the control cell. This
would appear to indicate that the
solvent based paint sludge has an
inhibitory effect on leachate
pollutant concentration. However,
as cell 2 is located inground in the
outside environment and cell 17 is
located in the high bay building the
noted inhibitory phenomenon could
possibly be more a temperature
effect than a coating effect of the
paint sludge additive. A comparison
with interior municipal refuse only
(cell 16) only serves at this time
to complicate the issue as cell 17
exhibits generally higher cumulative
leachate pollutant concentrations
than does cell 16. It is also impor-
tant to note that in comparing cell
2 (outside) with cell 16 (inside) it
was noted that the inside cell (#16)
was generally lower than the outside
cell (#2) in cumulative leachate
concentrations, thus indicating &
inhibitory effect possibly due to
environment temperature differences.
These observations produce the
following question in regard to th&
MSW/Solvent E.ased Paint Sludge
System. Is the fact that cell 17
generally exhibits higher cumulative
pollutant concentrations than cell
16 but less than cf:ll 2 indicative
of an environmental temperature
effect, the result cf a coating pro-
cess by the paint sludge, or some
combination of the two? The resolu-
tion cf this situation v/ill neces-
sarily be part of the continued
program as findings to date do not
provide satisfactory answers.
GAS DATA
In any study concerning the
landfill environment, the collection
of data on the different gases c;en-
eratec! is desirable though not
always easy when coupled with other
studies. Such was the situation in
this program. Plans were made to
collect information on the effects
of industrial waste co-disposal upon
gas generation. A recurring gas
leakage problem thwarted these
efforts in the majority of cells
located underground. However, the
interior cells were successfully
sealed and gas data was collected
and analyzed. Those results are
presented here not because they ere
associated with t.he co-disposal of
hazardous wastes but because of
their potential interest to landfill
investigators.
In their work on landfill gas
composition Rovers and Farquhar*
predict a composition versus time
curve following a pattern as
shown in Figure 20. This plot is
based upon the "assumptions that,
upon placement, subsequent aeration
of the refuse does not occur and
that conditions within t.he
decomposing refuse are sufficient
to encourage and to sustain CHU
production". This is a valid
assumption in theory but is a
condition which does not necessarily
exist in an actual situation, in
their reports, Rover and Farquhar2
state the infiltration from melting
snow end ice caused an inhibition
of the CHU forming bacteria, the
cause of which was rot known. They
also stated that aeration of the
refuse from air drawn in by the
infiltrating water may also have
disrupted CHU production. The
authors of this study also feel that
the lowered CHU production was very
likely caused by an inhibition of
the CHU forming bacteria. However
data obtained in this study appear'
to indicate that the aeration which
146
-------
LANDFILL GAS PRODUCTION PATTERN - PHASE
Figure 20
TIME
Sanitary Landfill Gas Production Pattern
caused the inhibition had as its
possible source the dissolved
oxygen content of the infiltration
water rather than eny air drawn in
by the water addition. This pos-
sibility is borne out by other
researchers.2'3 A reasonable cor-
relation was noted between water
addition and the cyclic methane
production found in both this study
and another similar cne being con-
ducted et another site in Franklin,
Ohio. The observation as made t.hc.t
the aerobic bacteria systems end
the anaerobic systems which produce
methane in a real world situation
do not necessarily develop in s
hard and fast time frame r.uch as
that postulated in theory by P.overs
and Farquhar2. In fact, these
bacterial systems are in o dynamic
state of flux and are dependent
upon
outside sources of influence
to determine which is dominant at a
civen time. In 1.he case of a. land-
fill this outside influence appears
to be caused to a great degree by
the infiltration of water containing
dissolved oxygen. The availability
of oxygen is the determining factor
of whether biological decomposition
is by aerobic or anaerobic organ-
isms.1* In this study it was con-
cluded that the addition of water
containing a high concentration of
dissolved oxygen caused a shift
from anaerobic to aerobic bacterial
dominance, thereby serving to cause
an overall low concentration of CHu
to be produced. (Figures 21-26)
ACKNOWLEDGEMENT
This project was supported by
the U.S. Environmental Protection
Agency (Contract No. 68-03-2120)
under the direction of Mr. Dirk
Brunner, Project Officer for the
Municipal Environmental Research
Laboratory in Cincinnati, Ohio.
The authors would also like to
thank Mr. Ken Mueller for his valu-
able assistance in the sampling and
analysis of the leachate and gas.
147
-------
55-
50-
o 45-
§ 40-
5
I
35-
30-
25-
15-
10-
5 5-
0
Figure 21
WATER ADDITIONS
CELLS 16 AND 17
100 200 300 400
DAYS FROM LOADING
500
600
Water Addition Data - Cells 16 and 17
N2 A A
co2 • •
CH^ a a
100 200 300 400 500 600 700
TIME (DAYS AFTER LOADING)
Figure 22 Cell 16 Gas Composition Data
148
-------
100
Figure 23
200 300 400 500
TIME ( DAYS FROM LOADING )
N2 4-..-A
C02 • •
CH4 a a
600
700
Cell 17 Gas Composition Data
65-
60-
55-
50-
45
40-
35-
30-
35-
20-
15-
10
5
WATER ADDITIONS
CELLS 18 AND 19
100
200
300
400
500
Figure 24
Water Addition Data - Cells 18 and 19
149
-------
130-
120-
110-
100-
90-
§ 8°-
§ 7°"
g? 50-
40-
30-
20-
10-
\ A—^
' V v
I M » /sT
( / \f\i \ /
/ / ^~3£- — V V^^ .— S-TT^v^
N2 A 4
CO2 • •
CH4 Q J3
-dn
»^-»=-*-=^s
100
Figure 25
200 300
TIME ( DAYS FROM LOADING )
400
500
Cell 18 Gas Composition Data
130 T
120-
110-
100-
90 -
80 •
70-
60 -
o
0 50-
40-
30 -
20-
10 -
0
N2 A 4
C02 • .
CH4 o a
100
Figure 26
200 300
TIME (DAYS FROM LOADING)
400
500
Cell 19 Gas Composition Data
150
-------
REFERENCES
1. Fenn, D. G., Hanley, K. J., and
DeGeare, T. V., "Use of the
Water Balance Method for Pre-
dicting Leachate Generation From
Solid Waste Disposal Sites",
USEPA Report, EPA/530/SW-168,
1975.
2. Rovers, F. A., Farquhar, G. J.,
"Gas Production During Refuse
Decomposition", Water, Air, and
Soil Pollution, 2-483, 1973.
3. EMCON Associates, "Sonoma County
Waste Stabilization Study, USEPA
Report, EPA/530/SW-65d.l, 1975.
4. Sawyer, C. N., McCarty, P. C.,
"Chemistry for Sanitary Engi-
neers", McGraw-Hill Book Co.,
Inc., New York, 1967, 518 pp.
151
-------
ACCELERATED TESTING OF WASTE LEACHABILITY AND CONTAMINANT MOVEMENT IN SOILS
Martin J. Houle and Duane E. Long
Department of the Army, Dugway Proving Ground, Dugway, Utah 84022
ABSTRACT
Sequential batchwise extractions of waste and soil can be used to simulate continu-
ously leached columns. Successive extracts of a waste (which change in composition as the
waste is depleted) are used to challenge a sequence of three soil batches that are graded
in size to allow taking samples for analysis between each step. By the proper choice of
plotting parameters, a time scale can be related to the cumulative extraction volumes.
The described procedure allows simulating years of field leaching in a few weeks of labora-
atory extractions.
INTRODUCTION
The ability of a soil to retard the
movement of chemical substances leached
from a waste is one of the important
factors in designing and selecting a waste
disposal site. Unless the soil can retard
waste leachate movement or remove the
hazardous materials from the leachate,
contamination of ground-water may result.
The chemical and physical composition of a
soil are the primary factors determining
the soil's effectiveness for shielding
ground-water from contamination. However,
the composition of the waste, environmental
factors such as annual precipitation, and
the geology of the area, are also very
important. These factors may be of equal
or greater importance than the soil compos-
ition if a very soluble waste containing
large quantities of toxic substances is
improperly disposed of.
The migration of chemical substances
through soil is usually studied in the
laboratory using columns packed with soil
to a predetermined bulk density. These
soil columns are challenged with a solution
extracted from a waste by water or some
other solvent such as municipal landfill
leachate, or the soil is treated with sim-
ple solutions of the ion under study. (A
useful configuration is shown in Figure 1,
along with illustrative plots of the data
obtained.)
Waste
Soil
Waste Effluent Volume-*
'|' Soil Effluent Volume -»•
1. Continuously leached columns and
associated output plots.
Continuously-leached column experi-
ments provide information as to the ability
of a soil to remove chemical substances
from a waste extract. However, an impor-
tant limitation of this method is the time
and effort required to obtain and analyze
a sufficient number of samples to make
predictions of migration rates and toxic
hazards. This usually requires months and
may even take years, depending upon the
flow rate of the leaching solution through
the columns. The information obtained from
152
-------
relatively short-term column studies cannot
be expected to describe what will occur
during years of leaching.
The desirability of maintaining a
uniform flow through waste or soil columns
is shown by studies which indicate that the
extraction of the waste and the removal of
some compounds from the leachate by the
soil is flow-dependent U, 2) . Maintaining
the flow rate at a constant, preselected
value is usually attempted either by
restricting the flow rate of the effluent
from the soil column, by regulating the
head pressure of the liquid above the
waste, or by using precise, flow-regulat-
ing pumps. The regulation is not easy, so
the columns must be carefully monitored.
(Soils containing a large sand fraction
allow nearly free flow of the waste leach-
ate while soils containing large amounts
of clay and silt require a large head pres-
sure in order to obtain a reasonable flow.)
When setting up experiments of this
type, any investigator is faced with the
problem of selecting values for each exper-
imental parameter such as leaching solvent
flow-rate, head-pressure, soil bulk density,
column diameter, waste-to-soil ratio, etc.
The choice of these values may not all be
entirely arbitrary, but a given set will
yield results which probably apply only to
that particular combination of conditions,
and the experiment may not be very useful
for making general predictions. By having
a more rapid and more flexible experimental
approach, a wider range of conditions can
be investigated within the framework of
factorial experiment designs which do allow
making predictions even in the presence of
interaction between multiple variables.
(Interaction exists when the effect of one
factor is dependent upon the level of
another factor. This introduces error into
the results of classical, vary-one-factor-
at-a-time experimentation.) A fast method
also allows making timely determinations,
on demand, for each specific situation.
Recently, our laboratory investigated
the Teachability of certain toxic metals
from a number of industrial wastes and
studied the migration of these metals
through soils. We used a batch method to
rapidly screen soils for their ability to
remove these metals from the waste leach-
ate^3'- This information was then used to
select soils for more detailed column
studies. While examining the data, it
became apparent that properly designed batch
studies of both wastes and soils could pro-
vide much of the same information obtained
from column studies. However, no adequate
information was available as to the correla-
tion between batch extractions and continu-
ously-leached columns of wastes and soils.
Samples of industrial wastes used in the
previous studies with continuously-leached
columns were therefore leached using a
serial batch extraction procedure. It was
found that the weights of toxic metals
extracted from the wastes batchwise compared
well to the column leaching^'. Besides
this, the results were obtained by the
serial batch extraction method in only a
small fraction of the time required by the
column method. The serial batch method is
experimentally much simpler, and it permits
the rapid investigation of the effect of a
wide variety of additional environmental
factors, such as freeze/thaw and drying/
resaturating cycles.
Other investigators have used batch
soil methods to study the removal of cer-
tain chemicals from waste extracts or muni-
cipal landfill leachate^5-8) and obtained
results that compared with soil columns for
their experimental conditions. However,
their experiments either did not allow for
the changes in the waste extract composi-
tion as the waste is depleted, for the
further change as the extract contacts each
increment of soil, and/or for the continu-
ally changing conditioning of each incre-
ment of soil, a change which depends both
upon the leaching time and the number of
soil increments above the one in question.
This paper reports some further
refinements of the serial batch procedure
for evaluating waste leaching and presents
the extension of the procedure to studying
the removal of metal ions by soil.
CORRELATING CONTINUOUS AND BATCHWISE
LEACHING
The data obtained from continuously-
leached columns may be presented in several
ways. One technique is to plot the concen-
tration of the chemical of interest found
in the waste or soil column sample versus
the cumulative volume through the column.
The common way of expressing the cumulative
volume is to use the cumulative pore volume
calculated for the type and weight of soil
employed. However, changing the type or
weight of soil will change the scale of the
cumulative volume axis when pore volume is
153
-------
employed. Figure 2 is an example showing
the difference obtained with pore volumes
of 40 and 60 mis. The corresponding total
volume in milliliters is appended for
comparison.
10
15
20
25
8
10
12 14 16
PV " 40 ml
PV ' 60 ml
200
400 600
Cum. Vol.
BOO
1000 Total mis
2. Differences in scales used to plot
cumulative volume.
It often is not practical or possible
to determine a pore volume for a waste due
to its physical form (heterogeneous sus-
pension, liquid, etc). This problem is
circumvented by using the soil column pore
volume as the measure of liquid volume
through the waste. It allows correlating
the waste-column output with the soil-
column results in a given set of experi-
ments. However, instead of using the soil
pore volume as the principle plotting
parameter, it is much more flexible to plot
the observed concentration of a chemical in
an extract versus the cumulative milli-
liters of leaching solvent per gram of
waste or soil, as shown in Figure 3. This
makes the scaling independent of soil type,
soil sample weight, and waste-to-soil ratio
and allows the direct comparison of many
different designs of experiments. It nor-
malizes the results so they can be more
readily correlated to a range of field con-
ditions. The area under the curve repre-
sents the total weight of a chemical
extracted per gram of waste or soil.
Batchwise extractions can be related
to continuously-leached columns by recog-
nizing that continuous leaching is equiva-
lent to running a series of discrete
extractions spaced by the frequency of
20 40 60 80
Cum. Vol., ml/g
IOO
3. Normalization of cumulative volume
using milliliters per gram.
collecting the effluent sample. Figure 4
shows that the concentration of the peri-
odic column samples can be plotted to
o
o
4.
0 20 40 60 80 100 120
Cum. Vol., ml/g
Correlation of batch with continuously
leached columns.
represent the average for that sampling
period. Thus, samples from the continuous
leaching of a column correspond to sequen-
tial batchwise extractions by volumes of
extractant equal to the volume passing
through a column between the taking of
samples.
When extracting a batch of waste or
soil, instead of using the same volume of
solvent for each successive extraction, the
sol vent-to-waste or -soil ratios can be
154
-------
graded In size, as indicated by the extrac-
tion volumes pictured in Figure 5. A small
£
o»
o
o
•**•
* «- — !._
20
40
60
8O
IOO 120
Cum. Vol., ml/g
5. Graded serial batch extractions.
sol vent-to-solid ratio should probably
always be employed for the first extrac-
tions; this is usually when the concentra-
tion is changing most rapidly, so smaller
Increments define the curve more accurately.
It also is when the soluble species will be
the most highly concentrated in the extract
and the ionic strength will be at its max-
imum. Greater dilutions would reduce this,
possibly affecting the solubility of other
components. After the more soluble compon-
ents have been extracted, the sol vent-to-
solid ratio can be greatly increased, thus
reducing the total number of extractions
required. The further along the cumulative
millilHer per gram axis the extraction
volumes extend, the longer the period of
column leaching the batch work is equival-
ent to.
Batch extractions are rapid compared
to letting the liquid percolate through a
column. If the volume of liquid used in a
batch extraction can be related to the same
volume of liquid passing through a waste or
soil over a period of time, sequential
batch extractions can be the basis for an
accelerated testing of wastes and soils.
However, the fraction of void space in a
soil (the pore fraction) affects the volume
of effluent passing through a soil per unit
time. The pore fraction can be calculated
from the formula
PF-1--^. (1)
where:
PF = pore fraction,
PL = bulk density, g/cm3, and
p = particle density, g/cm3.
The pore volume is the total volume of void
space in given quantity of soil. This is
obtained by multiplying the volume of the
soil by the pore fraction
PV = (PF)(V),
where: PV = pore volume, cm3, and
V = volume of soil, cm3.
(2)
To find the pore volume of one gram of soil
(which is of interest because the gram is
the basis for the normalization employed in
this paper), substitute the volume occupied
by one gram of soil, as calculated from
(3)
into equation 2, which yields,
PV = 1 -
(4)
(The pore volume may be corrected for the
fact that not all of the liquid contained
in the pores is exchangeable with a mass of
liquid moving through the soil. This
effective porosity is estimated to be 90
percent for most soils^9', so values
obtained from equation 4 should be multi-
plied by 0.90.)
The velocity of a moving liquid front
will be related to the volume displaced by
the front by first calculating the depth
occupied by one gram of soil out of the
total weight contained in a cubic centi-
meter. One gram of soil having a specified
bulk density and a cross-section of one
square centimeter will have a depth
where:
cm,
h = depth, cm.
(5)
The time required for a liquid front to
pass through one gram of soil having a
depth of -x— cm is
155
-------
T =
days,
(6)
where:
T = time, days, and
v = velocity, cm/day.
Since this gives the time required for a
liquid front to pass through one gram of
soil at the selected velocity, i.e., the
time required to fill the pore volume of
one gram of soil, the reciprocal is the
number of times the gram of soil is pene-
trated by a unit volume of liquid. Multi-
plying this by the pore volume for one gram
of soil gives the volume of liquid penetra-
ting the soil at the selected velocity
V -
(7)
The time required for any volume of liquid
to penetrate the one gram at a selected
velocity can be calculated from the
equation
T =
days.
(8)
To illustrate the use of these gener-
alized equations, assume that a soil has a
bulk density of 1.6 g/cm3 and a particle
density of 2.65 g/cm3. (These values are
representative of the clays used in the
experimental work reported in this paper.)
Thus, from equation 1, the pore fraction is
PF = 1 -
1.6
2.65
= 0.40,
and the pore volume per cubic centimeter is
PV = (0.40)(1.0) = 0.40 cm3.
The volume occupied by one gram of soil is
therefore
V = -~- = 0.625 cm3,
and the pore volume for the one gram of
soil is
PV =
Assuming a 90 percent effective porosity,
PV = 0.248 x 0.90 = 0.223 cm3.
If a liquid penetrates this soil at a
velocity of 1 x 10'5 cm/sec (which is 0.864
cm/day), one milliliter of liquid will pene-
trate one gram of soil in
(0.864H0.223) = 3'24 days'
T =
Thus, an extraction with 2 milliliters per
gram of soil is equivalent to 2 x 3.24 =
6.48 days of penetration in the field or in
a column. Table 1 lists the liquid-to-solid
ratios we employed, together with the cumu-
lative volumes and equivalent exposure times
for liquid front velocities of 1 x lO'1*,
10"s, and 10~6 centimeters per second
through a typical clay. This correlation
is displayed graphically in Figure 6.
The velocity of the liquid through a
soil underlying a specific waste-disposal
site must be determined experimentally to
be able to choose the correct correlation.
However, the flow-rate determining factor
will often be the penetrability of the
layer of waste. Table 1 also applies to
the leaching of waste because the volume of
TABLE 1. Correlation Between Extraction Volume and Penetration Time
Extraction
Number
1
2
3
4
5
6
7
Water Added, Cumul.
ml/g ml/q
2
3
6
12
24
48
96
2
5
11
23
47
95
191
Equivalent Days of Penetration3
10-" TO'5 10-6 cm/sec
0.65
1.62
3.56
7.45
15.2
30.8
61.9
6.5
16.2
35.6
74.5
152.
308.
619. (1.7yr)
65
162
356
745 (2.0 yr)
1520 (4.2 yr)
3080 8.4 yr
6190 (16.9 yr)
aAt the specified liquid front velocity through a soil having a pore volume of 0.223
156
-------
Front Vel.
I x ID"4 •»
I « ID'6 —
cm/see
20 40 60 80
Cum. Extraction Vol., ml/g
100 120
10
20
30
1000 2000 3000
Equiv. Leaching Time, days
6. Relating batch volumes to leaching
times for soils having a pore volume of
0.223 ml/g.
liquid through a column of waste will be
the same as through the soil beneath it, as
previously discussed.
THE GRADED SERIAL BATCH EXTRACTION
OF WASTES AND SOILS
The waste composition changes as com-
ponents are leached from the waste. Each
succeeding portion of extract will there-
fore generally have a different composition
as shown in Figure 7. Besides being chal-
lenged by a changing solution, the soil's
ion-removal characteristics continually
change with time as the soil becomes con-
ditioned and loaded by the passage of waste
extracts. Since each portion of waste is
changed by passage through a segment of
soil, the conditioning each succeeding seg-
ment of soil receives is different and each
segment therefore may remove different pro-
portions of the various ions present in the
waste extract. So although the soil seg-
ments start out the same, in effect they
become different soils due to the passage
of the different waste extracts.
The soil removes ions from the waste,
but the waste extract can also displace
ions from the soil. In addition, soil can
pick up a specific ion from a waste solu-
Water In
Batch No.
I
2
i
3
i
4
_j
Waste
Soil
u
S
O
Waste Extract
> 8
o
a
th
Soil Extract
7. The results of challenging soil with successive extracts of waste.
157
-------
tlon of one composition and then give it up
again as the liquid composition changes.
The soil may also give up ions later
because of intervening conditioning of the
soil by the passage of the changing waste
extract solution.
If extract samples were taken within a
layer of soil, it would be possible to
study this dynamically-changing situation.
This can be accomplished by placing sampl-
ing ports in the side of a soil column, as
shown in Figure 8. The same results can be
Water In
Waste
Soil
8.
Woste Extract
Section I Extract
Section H Extract
Section HI Extract
Challenging multiple soil segments with
successive extracts of waste.
attained in a shorter time with far fewer
equipment difficulties by putting waste
extracts on successive batches of soil and
taking a sample after each extraction. A
batch of soil then will represent a segment
of soil from a soil layer.
Normally, the distribution of sub-
stances retained by the soil column is
determined after leaching is concluded by
sectioning and analyzing the soil column.
But, this serial batch approach, with sampl-
ing between batches of soil, allows perceiv-
ing what is happening within a bed of soil
and provides data which could allow extra-
polating to the effect of thicker strata-
something which cannot be done with validity
from experiments with only a single layer,
or from experiments which use simpler condi-
tions. It is re-emphasized that batchwise
testing also yields its information in a
small fraction of the time required by col-
umns or field studies.
APPLICATION OF THE BATCH TECHNIQUE TO
INDUSTRIAL WASTES
Characteristics of Waste and Soils
We have applied this procedure to
seven different industrial wastes so far.
The results from a zinc-carbon battery
waste are presented here to illustrate the
method. This waste consists of broken-open
reject batteries. Table 2 shows the waste
battery composition per 1000 kg of batteries
produced. (Approximately one percent of the
batteries produced are rejected.) The
extract samples were analyzed to determine
the amount of each metal present at every
state of the batch tests.
TABLE 2. Composition of Zinc-Carbon Battery
Waste per 1000 kg Batteries
Produced P9
Constituent
Mercury
Zinc
Zinc Chloride
Manganese Dioxide
Cadmi urn
Lead
Waste Factor
(kg/1000 kq)
0.0073
3.812
0.248
6.147
0.00027
0.00031
Three soils were investigated for their
ability to remove the metals of interest
from extracts of the zinc-carbon batteries.
The soils chosen were Chalmers (a gray,
silty, clay loam from Indiana, a Mollisol),
Davidson (a red clay from North Carolina,
an Utisol), and Nicholson (a yellow, silty,
clay from Kentucky, an Alfisol).
These soils were selected because of
differences in their chemical properties
and clay mineralogy. Table 3 shows that
Chalmers and Nicholson soils have similar
surface areas but significantly different
cation exchange capacities. In addition,
the clay mineral composition (< 2 micron
separates) are much different. The
Chalmers clay composition is largely mont-
morillonite, with small amounts of vermicu-
lite, chlorite, and kaolinite. The Nichol-
158
-------
TABLE 3. Some Physical and Chemical Properties of Soils Used in this Study
Soil
Chalmers
Javidson
Nicholson
Soil
Paste
PH
6.6
6.2
5.0
Cation
Exchange
Capacity,
meq/100 g
26
9
37 (?)
Surface
Area,
m2/g
125.6
51.3
120.5 (?)
Free
Iron
Oxides,
Percent
Texture
Sand,
Percent
3.1 7
17.0 19
5.6 9
Silt,
Percent
58
20
31
Clay,
Percent
35
61
60
aThese values, obtained from the soil we used, show that it differs from Fuller's vV
sample, which had a pH of 6.7. The CEC and surface areas listed are for his sample,
which was 49% clay, so our values for these would be somewhat higher.
TABLE 4. Specifications for Serial Batch Extractions
Extract-
ion
Number
1
2
3
4
5
6
7
Water
Added,
ml/q
2
3
6
12
24
48
96
Volume of
Water, ml ,
Extracting
300 g Waste
Volume of Filtrate Onto a
I
60 g
Soil
600 1 20
900 180
1 ,800 360
3,600 720
7,200 1,440
14,400 2,880
28,800 5,760
II
30 g
Soil
60
90
180
360
720
1,440
2,880
Soil
III
15 g
Soil
30
45
90
180
360
720
1,440
son clay faction is predominately vermicu-
lite with only a trace of mica and kaolln-
ite. In contrast, Davidson soil has a low
surface area and cation exchange capacity.
The clay is predominately kaolinite but
this soil contains a high percentage of
hydrous oxides of iron. It has been shown
that iron oxides play a major role in heavy
and trace metal removal 12 .
The Procedure Employed
A sequence of seven extracts was made
from waste zinc-carbon batteries. Four
batteries were weighed and placed in two-
quart, wide-mouth screw-cap jars. (Although
it was not done with the batteries, ordin-
arily a sample of waste is dried to deter-
mine moisture content, then sufficient
undried sample is weighed to give 300 grams
dry weight. Drying the sample could affect
hydrated species and drastically reduce the
solubility. If the waste had supernatant
water, the volume of the water was
considered as part or all of the first
extract.) Appropriate volumes of water
were added for each extraction to produce
the liquid-to-sol id ratio given in the
second column of Table 4. The bottle was
shaken gently 4 or 5 times daily (Continual
mechanical shaking was not used because of
concern that it might abrade the waste
agglomerates, making them more susceptible
to extraction). The time required to reach
equilibrium was determined by periodically
withdrawing an aliquot for analysis; 24
hours is adequate for most wastes of small
particle size, 72 hours was required for
the first battery extraction, 48 hours for
the remainder. At the end of the
extraction period, the mixture was filtered
under vacuum using a hardened filter paper
(such as Whatman 54) in a Buchner funnel.
An aliquot of approximately 20 mill inters
was withdrawn for metals analysis and
filtered through a 0.5 u Millipore filter
to remove fine particles which might have
by-passed the filter paper to possibly dis-
solve when the sample was acidified. (After
measuring conductance and pH, one percent
159
-------
20 ml aliquot through Hlllifor. and analyu.
dally to find aquilibration tiaa.
60 9 soil
30 ,
Filtrate Add UP Ml to 60 g of .aofa typa of .oil. | Mima:
Fllur portion of
Millipora and analyi.,
through
r
Filter portion throu9h
Millipor. for uialyii..
Add 18OO ml
water, «tc.
Hi* WBll.
1 Etc.
SOIL BATCH
III
15 g .oil ~[
flltrat.; Add 60 ml
v
R«f UtaK portion
through Hillipor*
and anlayxa.
Fj,ltratet Add 18O ml to fjlterad apil batch.
Flltratet Add 30 ml
Rafiltar portion
through Millipora
riltrato; Add 9O ml
Rafilter portion
through
and analyca.
Mix {
periodically |
Filter
throug
Hhatma
t 54.
*
v*x._
h Filtrat*: Add 45 ml 1
Rafiltcr portion
Etc. L< *nd analyca-
through
Whatman
t 54.
[Filter-
ctika of
-^
Mix wall.
Etc.
Filtrate; Refiiter portion
''through Hillipor*
and analyM.
9. Flow chart for graded serial batch extractions.
-------
concentrated nitric acid was added to the
filtrate to inhibit precipitation while
standing.) The solid waste residue was
transferred back to the jar and mixed with
the volume of water specified for the next
batch. The flow-chart of Figure 9 outlines
the sequence of operations.
In the procedure detailed here, the
liquid-to-sol id ratio was continually
increased to further accelerate the test-
ing—the volume of each extraction after
the second one was made double the one
before, which redoubles the time repre-
sented by that extract. With some wastes,
adequate results may be obtainable by
using very large volumes right from the
first (or one or two extractions using
small liquid-to-sol id ratios, followed by
a very large one as shown in Figure 10,
20
40
60
80
100 120
Cum. Vol., ml/g
10. Making a rapid estimate by extracting
with 3, 5, and 100 milliliters per
gram.
but this would have to be checked for each
k'ind of waste by comparison with the more
conservative series of extractions utilized
in Table 4. However, such a procedure
would allow rapid simulation of long leach-
Ing periods and could be useful in the
routine monitoring of variations in waste
composition and Teachability^ .
The filtrate resulting from each
sequential extraction of the waste was
mixed with the first of three batches of
each kind of soil. The weights of soil
used were 60, 30, 15 grams, representing
section I, II, and III, respectively. This
gradation in weight allows taking an
aliquot of the extract for analysis and
having enough left over to challenge the
next soil batch at the same liquid-to-sol id
ratio. Extracting 300 grams of waste
yields sufficient solution to challenge
three different kinds of soil in experi-
ments set up with the proportions stated
in Table 4.
Although the soil equilibrates in 6
hours or less M , each solution was kept
in contact with the batch of soil before
filtration for the same length of time as
used to extract the waste. This was to
keep the samples progressing smoothly
without gaps in the series. After filter-
ing the soil extract, an aliquot was
refiltered through Mil11 pore and saved for
analysis. The appropriate volume of the
remaining filtrate was added to the next
batch of soil. The soil exposed to the
first waste extract was recovered and mixed
with the second waste extract in the series.
This was repeated until the waste had been
extracted seven times and each waste
extract had progressed through all three
soil batches. (An eighth extract could have
been made to further increase the equivalent
leaching time.) This procedure was run in
duplicate. When the volume of liquid
became too large for the two-quart jar, the
sample was transferred to a plastic five-
gallon jug. (Square jugs with the opening
above one side were the easiest to remove
the contents from for filtering.)
Interpretation of Results
Plotting the waste output as in Figure
6 shows the depletion of the waste with con-
tinued leaching. To show the effect of
passing the waste extract through soil, the
histogram of Figure 11 is more useful. It
presents the results obtained from extract-
ing a batch of waste and using the solution
to challenge a succession of three batches
of soil. (Multiplying the observed concen-
tration in micrograms/milliliter by the
individual batch extraction volume per gram,
milliliter/gram, converts the values to
micrograms/gram of waste or soil. Micro-
grams per gr.am is equivalent to grams per
metric ton of 1,000 kg or 2,205 Ibs.) The
height of the histogram bar labeled W
represents the micrograms of. e.g., zinc
extracted per gram of waste. This is the
challenge to the batch representing soil
section I. Because the liquid-to-soil ratio
is kept the same for both the waste and the
161
-------
o
(A
O
O
Out of
Waste
Retained
on I
Penetra-
ting I
1
Retained
on H
Penetra-
ting IE
I
Retained |
on HE i
Penetra-
ting HE
w
Waste
Soil Sections
11.
Histogram showing the percentage and
retention of a species by soil.
soil, W also is equal to the micrograms of
zinc per gram of soil challenged. The
height of the bar labelled I shows the con-
centration of zinc penetrating the batch
representing soil section I, and the differ-
ence in height between I and W is the
amount of zinc retained per gram of soil.
The ratio of (W-I) to W is the fraction
removed by that soil section. Similarly I
is the challenge to II and bar II shows the
penetration through II. If the fraction of
zinc removed by each soil section is differ-
ent, this shows that the removal character-
istics of the soil are affected by condi-
tioning and by changes in the extract. This
can be further studied by comparing the
histograms for the different extractions.
When a histogram bar is higher than the
preceding one, it shows that the soil is
releasing zinc. The soil sections can be
treated in pairs as above, or it can be
considered that a given amount of waste has
challenged three different amounts of soil:
section I, section I + II, and sections I +
II + III. Thus, the fraction removed can
be calculated for three different waste-to-
soil ratios.
The concentration, per gram soil,
of a sorbed species divided by its con-
centration in the solution yields the
distribution coefficient, which is the
slope of the adsorption isotherm. By
calculating a distribution ratio for
each histogram bar, it is possible to
follow the change in slope of the
adsorption isotherm for each serial
extraction (which show the effect of
the changing sample matrix) and for
each different waste-to-soil ratio.
This is of considerable importance for
modeling and for making predictions of
contaminant movement through soils.
(These relations will be developed in
a later paper.)
Although several metal ions were found
in significant concentrations in the
battery waste leachate, zinc will be used
here as the example. (The full discussion
of all of the wastes studied will appear in
a separate report.) Figure 12 presents a
composite plot of histograms which show the
extraction of zinc from battery waste and
its penetration through Davidson and
Nicholson soils. (Chalmers soil gave
nearly the same results for zinc as did
Davidson soil so is not plotted here.)
Going vertically down the figure viewing
any single column shows how the extraction
of zinc and its retention by soil changes
as the leaching progresses. For example,
the W's show the depletion of zinc in the
waste extracts, with an increase in the
fourth batch probably due to the removal of
other highly soluble species in the early
extractions, resulting in an environment
more favorable to the release of the
remaining zinc.
It is seen that Davidson soil is con-
siderably more effective in removing zinc
from waste battery extract than is
Nicholson soil. Although Nicholson soil
has a much higher surface area and cation"
exchange capacity than Davidson soil the
iron oxide content of the Davidson soil is
much higher than the Nicholson. This is
one of the factors influencing the differ-
ence in zinc removal by the two soils. The
ability of Nicholson soil to remove zinc
began to approach the removal capability of
Davidson soil by the fourth in the series
of extractions, either because of a condi-
tioning of the Nicholson soil or because
the zinc is then present in a waste leach-
ate of different composition.
The Nicholson soil histograms for the
second and succeeding extracts show a con-
162
-------
wg Zn/g Soil
40O
338
Extraction
I
0
400
23
2.7 1.3
Extraction
2
Extraction
3
0
400
54
6.0 5.2 2.8
Extraction
4
64
5.3 1.8 2.1
Will
Davidson
ug Zn/g Soil
400i
338
196
135
106
400i
109
73 69
86
400-
0
400-
r\ .
54 65
64
13 12 , 20 t
w i i in
Nicholson
12. The penetration of zinc in battery extract through soils.
163
-------
siderable release of zinc from the second
or third soil batches. This is probably
zinc taken out of previous extracts and
later released by the soil due to the change
in composition of the later extracts and the
different soil history.
Each histogram of Figure 12 could have
been treated as in Figure 13 by calculating
(A
U
c
o
0
338ug/g
Out of
Waste
Retained
on I
I96ug/g
Penetra-
ting I
6lug/g i
Retained {
on 31 I
I35yg/g
Penetra-
ting!
Ret.onJr I
I06ug/g
Penetra-
ting IE
W
Waste
Soil Batches
13.
The penetration and retention of zinc
in battery extract on Nicholson soil.
the amount of zinc retained on the soil.
This data can then be used to calculate the
fraction of zinc removed from the extracts
by the soil, or the fraction released from
the soil as a peak elutes through the soil
batches.
The method of Figure 6, which is
particularly useful for graphing the waste
extraction results, can be applied to the
soils to emphasize the change in each
succeeding extract. The widths of the his-
togram bars in this case represent the
milliliters per gram used to extract each
batch. The heights of the bars can denote
the zinc concentration in ug/g, as depicted
in Figure 14. If the bar heights stand for
the observed zinc concentrations in ug/ml,
the area under each bar is equal to the
jig Zn/g Soil
400
338
Wo«t«
400
Soil Batch
I
Soil Botch
31
13S
109
400 .
Soil Batch
IE
20
25 mlt/g
ptgna 3rd 4«, Extraction
Nicholson
14. Plots emphasizing amounts extracted.
total number of micrograms of zinc
extracted from that batch. (Concentration,
ug/ml, times ml/g equals micrograms zinc
extracted per gram of waste or soil.)
The conductance of every extract was
measured to obtain an Indication of the
total concentration of ionized substances
in the solution. The specific conductances
graphed as column W of Figures 15 and 16
show that the soluble ions in the waste are
rapidly leached out. When placed on batch
164
-------
I of Nicholson and Davidson soils, the ions
in the waste extract either penetrated the
soils or they exchanged or displaced from
the soil a comparable number of ions. The
extracts from soil batch I produced roughly
the same effect upon soil batch II. The
conductance of the second waste extract was
considerably lower than the conductance of
the soil extracts, probably because resi-
dual soluble ions from the first waste
extract were being washed through the soils.
The ions appear to move as a broad peak in
a manner similar to that obtained 1n
elution chromatography.
The pH of the Nicholson soil extracts
were considerably different from the
Davidson soil extracts, as shown in Figures
15 and 16. Even though the first waste
extract had a pH of 7.6, the Nicholson soil
extracts from batches I, II, and III were
quite acid: pH 5.8, 4.5, and 4.7, respect-
ively, for the first extract. In compari-
son, the first Davidson soil extracts were
6.4, 6.3, and 6.4. The pH of the extracts
from both soils were not comparable until
they had been exposed to the fourth waste
extract.
These batch testing experiments were
also extended to determine the limiting
soil capacity for the metals in the waste
leachate. This work will be reported in a
separate paper.
CONCLUSION
It has been demonstrated that the
leaching of a waste by water or other sol-
vent (e.g., the extract from another waste,
such as municipal landfill) can be charac-
terized and the ability of a soil to remove
a chemical species from the waste leachate
can be rapidly evaluated using serial
batchwise extractions. The batch technique
is also far more convenient than columns
for rapidly investigating the effect of
various environmental factors. Particu-
larly if the extraction jars are sparged
with nitrogen, this technique simulates the
saturated anerobic conditions in the field.
(This is considered the condition that
would favor migration of cations through
soil so the procedure may provide a "worst
case" investigation into the suitability
of the soil.) It is recognized that any
effect of the waste leachate upon the soil's
hydraulic properties can not be investi-
gated using the batch method, but some
columns can be set up to provide this data.
(Hydraulic studies are quite simple to
conduct because they use free-flow columns
and do not require close monitoring or
analysis of the soil leachate. Also,
changes in the hydraulic properties caused
by the waste leachate usually appear during
the passage of the first few pore volumes
of waste leachate.) Some columns can also
be set up to follow long-term biological
effects and similar slow changes.
No one method can give all the answers,
but the graded serial batch approach can
quickly provide much of the information
needed to assess the categories of hazards
from a class of wastes and to make decisions
concerning the suitability of soil types for
inhibiting the migration of hazardous
chemicals.
ACKNOWLEDGEMENTS
This study was a part of a major
research program on the migration of hazard-
ous substances through soil, which was con-
ducted at Dugway Proving Ground under the
auspices of, and funded by the Environmental
Protection Agency, Municipal Environmental
Research Laboratory, Solid and Hazardous
Waste Division, Cincinnati, Ohio 45268,
under Interagency Agreement EPA-IAG-04-0443.
Dr. Mike H. Roulier was the EPA Project
Officer and his assistance is gratefully
acknowledged. Special thanks is extended to
Mr. Gordon Ricks who performed the metals
analyses, Mrs. Mae Barcus, Miss Lynnette
Gilmore, and Mrs. Cathy Meikel who performed
many of the extractions and analyses.
Finally, the assistance of Mrs. Karen Zamora
in the preparation of the manuscript is
acknowledged.
REFERENCES
1. Houle, M. J., Long, D., Bell, R.,
Weatherhead, D. C., Grabbe, R., and Soyland,
J., "Migration of Hazardous Substances
Through Soil, Part I, Electroplating,
Chlorine Production, Nickel-Cadmium Battery,
Water Base Paint, and Pigment Wastes,"
October 1977, Draft Report Prepared for
Solid and Hazardous Waste Research Division,
Municipal Environmental Research Laboratory,
US Environmental Protection Agency,
Cincinnati, Ohio 45268, Contract Number EPA-
IAG-04-0443, 258 pages.
2. Korte, N. E., Fuller, W. H., Niebla, E.
E., Skopp, J., and Alesii, B. A., "Trace
Element Migration in Soils: Desorption of
Attenuated Ions and Effects of Solution
165
-------
8.0 i
Extraction
I
6.0
4.0
PH
7.6
5.8
4.5
4.7
.32300
3QOOO
27500
29)500
21100
8.0
Extraction
2 6.0
4.0
7.5
6.0
5.1
4.8
30000 .
9,900
4900
11,200 13*800
8.0 .
Extraction
3
4.0
7.5
6.4
6.2
5.4
301000 .
5000
Extraction
8.0
6.0
4.0
7.4
6.8
6.5 6.4
30/000
960 1020 U20 jj«00
wiirnr w i n
pH of Extract Specific Conductance
Nicholson
15. The pH and conductance of battery extract before and after
passage through Nicholson soil.
166
-------
8.0 !I
PH
Extraction
I
6.0
4.0
7.6
6.4 6>3 6.4
3Q000
eoo
29300
26)000
21,500
8.0
Extraction 6.0 .
2
4.0
7.5
6.3 «•*
30POO .
1Q100
IWOO
4300
Extraction
8.0
6.0
4.0
7.5
6.7
6.5
30000 .
4300
Extraction
8.0 ,
6.0 .
4.0
7.4
6.8
6.7 6.7
30/000
960 970
WOO
winm winut
pH of Extract Specific Conductance
Davidson
16. The pH and conductance of battery extract before and after
passage through Davidson soil.
167
-------
Flux, IN: Residual Management by Land Dis-
posal . Proceedings of the Hazardous Waste
Research Symposium, February 1976, The
University of Arizona, Tucson, Arizona,
Wallace H. Fuller ed., EPA-600/9-76-015,
Solid and Hazardous Waste Research Division,
Municipal Environmental Research Laboratory,
US Environmental Protection Agency,
Cincinnati, Ohio 45268, 270 pages.
3. Houle, M. J., Grabbe, R., Soyland, J.,
Bell, R., and Lee, H., "Migration of Haz-
ardous Substances Through Soil," December
1974, Nine Month Progress Report, Prepared
for Solid and Hazardous Waste Research Div-
ision, Municipal Environmental Research
Laboratory, US Environmental Protection
Agency, Cincinnati, Ohio 45268, Contract
Number EPA-IAG-04-0443, 59 pages.
4. Houle, M., Long, D., Bell, R.,
Weatherhead, D., and Soyland, J.,
"Correlation of Batch and Continuous
Leaching of Hazardous Wastes," IN: Proce-
edings of a National Conference About
Hazardous Waste Management, February 1977,
San Francisco, California, Harvey F. Collins
ed., Prepared for US Environmental Protec-
tion Agency, In Press.
5. Liskowitz, J. W., e_t a]_, "Evaluation of
Selected Sorbents for the Removal of Con-
taminants in Leachate from Industrial
Sludges," IN: Residual Management by Land
Disposal. Proceedings of the Hazardous
Waste Research Symposium, February 1976,
The University of Arizona, Tucson, Arizona,
Wallace H. Fuller ed., EPA-600/9-76-015,
Solid and Hazardous Waste Research Division,
Municipal Environmental Research Laboratory,
US Environmental Protection Agency,
Cincinnati, Ohio 45268, 270 pages.
6. Farquhar, G. J., and Rovers, F. A.,
"Leachate Attenuation in Undisturbed and
Remolded Soils'," IN: Gas and Leachates
from Landfills: Formation, Collection, and
Treatment. Proceedings of a Symposium at
Rutgers University, New Brunswick, New
Jersey, March 1975, Emil J. Genetelli and
John Cirello eds., EPA-600/9-76-004, Solid
and Hazardous Waste Research Division,
Municipal Environmental Research Laboratory,
US Environmental Protection Agency,
Cincinnati, Ohio 45268, 190 pages.
7. Rovers, F. A., Mooji, H., and Farquhar,
G. J., "Contaminant Attenuation-Dispersed
Soil Studies," IN: Residual Management by
Land Disposal. Proceedings of the Hazard-
ous Waste Research Symposium, February 1976,
The University of Arizona, Tucson, Arizona,
Wallace H. Fuller ed., EPA-600/9-76-015,
Solid and Hazardous Waste Research Division,
Municipal Environmental Research Laboratory,
US Environmental Protection Agency,
Cincinnati, Ohio 45268, 270 pages.
8. Griffin, R. A., Frost, R. R., and Shimp,
N. F., Ibid, pages 259-268.
9. Davidson, J. M., Personnel Communica-
tion, Soil Science Department, University of
Florida, Gainsville, Florida 32611, 1975.
10. Versar Inc., "Assessment of Industrial
Hazardous Waste Practices, Storage and
Primary Batteries Industries," January 1975,
Publication SW-102c, Prepared for, Office of
Solid Waste, US Environmental Protection
Agency, Washington, DC 20460, 209 pages.
NTIS, PB-241 204.
11. Fuller, W. H., "Movement of Selected
Metals, Asbestos, and Cyanide in Soil:
Application to Waste Disposal Problems,"
April 1977, EPA-600/2-77-020, Prepared for:
Solid and Hazardous Waste Research Division,
Municipal Environmental Research Laboratory,
US Environmental Protection Agency,
Cincinnati, Ohio 45268, 243 pages.
12. Jenne, E. A., "Controls on Mn, Fe, Co,
Ni, Cu, and Zn Concentrations of Soils and
Water: The Significant Role of Hydrous Mn
and Fe Oxides," IN: Trace Inorganics in
Water, Advan, Chem. Ser. 7,3: 1968, p. 337-
387.
13. Griffin, R. A., and Shimp, N. F.,
"Attenuation of Pollutants in Municipal
Landfill-Leachate by Clay Minerals,"
September 1975, Draft Report Prepared for,
Solid and Hazardous Waste Research Division,
Municipal Environmental Research Laboratory,
US Environmental Protection Agency,
Cincinnati, Ohio 45268, Contract Number 68-
03-0211.
168
-------
DISPOSAL AND REMOVAL OF POLYCHLORINATED BIPHENYLS IN SOIL
Robert Griffin
Illinois State Geological Survey, Urbana, IL
and
Robert Clark, Michael Lee, and Edward Chian
Civil Engineering Department
University of Illinois at Urbana-Champaign
ABSTRACT
The adsorption, mobility, and microbiological degradation of polychlorinated biphenyls
(PCBs) in soil materials were studied relative to land disposal. The mobility of Aroclors
1242 and 1254 and a used capacitor fluid were measured by the soil thin layer chromatog-
raphy technique. PCBs were immobile in all soils when leached with aqueous solvents
(water and landfill leachate); they were highly mobile in all soil materials when leached
with organic solvents.
The rate of adsorption of PCBs by soil materials was found to be rapid, with equi-
librium conditions achieved in less than 8 hr. The adsorption process conformed to the
Freundlich adsorption equation. PCBs were found to be strongly adsorbed by soil materials.
The adsorption capacity and the mobility of PCBs were correlated to the organic carbon
content and surface area of the respective soil materials.
Microbial degradation of PCBs was studied using mixed cultures of PCB-degrading
microogranisms. Aerobic degradation of water-soluble Aroclor 1242 was 92 percent complete
within 20 hr and as high as 98 percent within 10 days. PCB isomers with less than four
chlorines were degraded while those with four or more chlorines were not significantly
degraded. Soil samples collected from field plots where PCB-contaminated sewage sludge
was applied for 7 years were found to contain mainly high chloririe substituted isomers.
The field samples confirmed the mode of microbial degradation found in the laboratory
studies.
INTRODUCTION
Polychlorinated biphenyls, or PCBs,
have been manufactured since 1929. In
the United States, they have been marketed
under the trade name "Aroclor" by the Mon-
santo Company. It has been estimated that
over 800 million pounds of PCBs have been
produced. In 1970, the year of peak U.S.
production, over 85 million pounds of PCBs
were produced in the U.S. More than 57
percent of that amount was in the form of
Aroclor 1242.t1)
The major application of PCBs has
been in large electrical capacitors and
transformers. PCB isomers having four or
more chlorines per molecule are nonflam-
mable and make excellent di-electric mate-
rial for electrical equipment. Their
unique properties led to many other appli-
cations such as hydraulic, vacuum-pump,
and heat-transfer fluids. PCBs have been
used as plasticizers, lubricants, inks,
and pesticide extenders.
Many of these uses led to the release
of PCBs into the environment either acci-
dentally or deliberately. Not until
Jensen(2) reported PCB contamination of
humans in 1966 was notice taken of PCBs in
other environmental samples. In recent
years, the accumulation of PCBs in humans
and animal tissue and their toxic effects
have been well documented.(3-5) PCBS have
been identified as a significant hazard to
human health as well as to the environment;
their disposal has caused great concern.(6)
Incineration, considered the safest
method of disposal of PCB wastes, is costly
and can be a source of PCBs returning to
the environment if operating conditions
are not properly controlled!?) Further-
more, since the major sources of waste PCBs
are large electrical transformers and ca-
pacitors, the satisfactory incineration of
the waste PCBs becomes even more compli-
cated. Land burial, much less expensive
169
-------
and more energy efficient than inciner-
ation, is thus a more desirable altern-
ative for disposing of PCBs. However,
little information is presently available
concerning the mechanism of attenuation of
PCBs in soils or the possibility of ground-
water contamination by PCBs leaching from
landfills.
The water around a sanitary landfill
was analyzed by Lidgett and Vodden, ( '
who found no detectable (<4 ppb) PCBs in
the water. Similarly, Robertson and Li(9)
failed to detect PCBs in ground water using
gas chromatography/mass spectrometry tech-
niques. Moon et al.('' reported that
levels of PCBs in groundwater in the vicin-
ity of eleven sanitary landfills were below
detection (<1 ppb) but that low levels of
PCBs were found in waters from monitoring
wells at several industrial PCB disposal
sites and lagoons. They concluded from
analyses of water and split-spoon soil
samples that PCBs are present in most
leachates from land disposal sites and that
PCBs have a. strong affinity for soil.
Tucker, Litschgi, and Mees(l°) and Griffin
et al.(H) also concluded that PCBs have a
strong affinity for soil and that PCBs were
not readily leached from soil by percolat-
ing water.
The above studies suggest that the
interaction of PCBs with soil particles is
an important attenuation mechanism.
Another potentially important mechansim is
the degradation of PCBs. Koeman(12)fOund
that lower-chlorinated PCB isomers were
found in animals low on the food chain,
whereas higher-chlorinated isomers were
found in animals high in the food chain.
This seemed to indicate that the less-chlo-
rinated isomers were less persistent, i.e.
they were being degraded.
Many microorganisms have been reported
that can degrade PCBs to some degree. De-
gradation has been demonstrated for a soil
fungus, Rhizopus japonicua,(13) for a
species of Nocardla, and for species of
Pseudomon. CW.O) it has been known for
some time that PCBs can be degraded by
birds and mammals.(15,16)
PCBs frequently end up in sewers,
either through spills or dumping. Herbst
et al.(17) studied the fate of two radio-
labeled isomers of PCB, a trichlorobiphenyl
and a pentachlorobiphenyl, in an activated
sludge unit. They found that the isomers
were not degraded and that the major
portion of the PCBs ended up In the acti-
vated sludge. However, Tucker et al.(l8)
reported that Aroclor 1242 was degraded
by 26 percent in an activated sludge unit.
Mihashi et al.(19) reported that PCBs were
degraded by 50 percent in activated sludge
and that the degree of degradation de-
creased as the chlorine substitution in-
creased. PCBs have also been shown to be
degraded by two species of Archromo-
bacter(20»21) and by a Alkaligenes
species(22) isolated from sewage.
The limited information presently
available indicates that PCBs have a
strong affinity for soil and are somewhat
degradable by microorganisms. There is no
evidence that ground water around sites
containing relatively low levels of PCBs,
such as sanitary landfills, has become
contaminated with PCBs. However, surface
and ground waters around some industrial
disposal sites and around lagoons con-
taining relatively large quantities of
PCBs have become contaminated by leaching
PCBs. The mechanisms of transport of
PCBs in the biosphere and the mechanisms
of attenuation in soil are unknown. Data
on the factors affecting PCB attenuation
by earth materials and the microbial de-
gradation of PCBs would provide a useful
basis for determining waste treatment
methods, for predicting PCB migration under
landfills, and for selecting and designing
future disposal sites.
BACKGROUND
The research reported here is supported
in part by Grant R-804684-01, from the
U.S. Environmental Protection Agency, Mun-
cipal Environmental Research Laboratory,
Solid and Hazardous Waste Research Divi-
sion, Cincinnati, OH 45268. Since the
research is currently in progress, the
techniques, results, and conclusions re-
ported here should be considered as tent-
ative and subject to revision and rein-
terpretation.
The purposes of the present project
are: (a) to conduct an extensive liter-
ature review of pertinent information on
the attenuation of PCBs in soil materials;
(b) to measure the adsorption capacity of
selected earth materials for pure PCBs
and PCB wastes; (c) to quantitatively
evaluate the effects of biological de-
gradation, volatilization, time, and ad-
sorbent structure on adsorption of PCBs;
(d) to use this data to develop a mathe-
matical model that will allow prediction
of PCB adsorption and mobility; and (e)
to further develop analytical procedures
that will allow improved quantitative meas-
170
-------
urement of PCBs contained in aqueous solu-
tions.
PCB Materials
Polychlorinated biphenyls (PCBs) is
a generic term applied to certain mixtures
of synthetic organic compounds. These com-
pounds are mixtures of very closely related
isomers and homologs that contain two phenyl
rings with 10 possible chlorine attachments.
The biphenyl structure is shown in Figure
1. PCBs are formed by substituting chlo-
6' 5'
Fig.
1. BIPHENYL STRUCTURE: Positions 2
to 6 and 2* to 6' indicate ten pos-
sible positions for chlorine sub-
stitution. Different amounts of
chlorine substitution form the
various PCBs.
rine atoms for one or more of the hydrogen
atoms at the numbered positions of the bi-
phenyl structure. There are over 200 pos-
sible isomers; about 103 may occur in com-
mercial PCB formulations.
The PCB materials chosen for this
study were Aroclor 1242 and Aroclor 1254
(42 and 54 percent chlorine, respectively).
They were supplied by the Monsanto Company.
The 1*C labeled compounds were prepared by
New England Nuclear Corporation. Since gas
chromatographic traces of the l*C labeled
compounds were identical to those of Aroclor
1242 and Aroclor 1254, respectively, it was
assumed that there were no significant dif-
ferences in the respective compounds and
that the 14C labeled and the Aroclors would
behave similarly in studies of adsorption,
mobility, and microbial degradation.
Also obtained for study was a used
apacitor fluid drained from a burned out
50 KVA capacitor manufactured by Westing-
house in 1966 and originally containing
Aroclor 1242. The capacitor fluid was ana-
lyzed for PCB with standard GC techniques
and found to contain Aroclor 1242. However,
the density of the capacitor fluid was mea-
sured to be 1.304 g/cc, while that for un-
used Aroclor 1242 was determined to be
1 354 g/cc. It is believed that the Aro-
clor 1242 in the capacitor fluid was less
dense due to suspended carbon particles.
These carbon particles gave the fluid a
black color and are presumed to have been
generated during the "burn out" of the cap-
acitor. This capacitor was supplied by
Illinois Power Company and was scheduled to
be landfilled. We believe this fluid is
representative of the type of PCB wastes
that are normally disposed of in landfills.
Adsorbents
Earth materials, representing a wide
range in characteristics, have been se-
lected as adsorbents. The materials being
studied are: Ottawa silica sand; Panther
Creek southern bentonite clay; the soils,
namely Bloomfield Is, Ava sic, Cisne sil,
Flanagan sil, Catlin sil, Drummer sicl,
Weir sic, a calcareous loam till; and
two coal chars. The chars were selected
because of their high adsorption capacity
for organic compounds. Because they are
a waste product of many coal conversion
processes, they have potential use as a
liner material for disposal sites accepting
organic wastes.
Analytical Development
In general, PCBs are determined quan-
titatively by comparing gas chromatographic
(GC) response patterns of a multicomponent
environmental sample with commercial PCBs
(Aroclor) or a mixture of Aroclors. This
technique is limited by the sensitivity
and reproducibility of comparisons of the
large number of peaks produced by the var-
ious PCB isomers. The procedure is further
complicated because the various components
of water soluble PCBs contained in envi-
ronmental samples are not likely to have
the same composition as those in the or-
iginal Aroclor used as a reference compound.
For practical reasons, the quantitation is
usually done by integration of the peaks
and by comparison with standards of known
isomeric composition. This can cause some
error, depending on how well the mixture
of isomers in an unknown sample compares
to a standard.
Because of these problems, we have de-
veloped procedures that allow improved
quantitative measurement of PCBs in aqueous
samples. The main thrust of our studies
has been to improve previous procedures
whereby isomeric mixtures of PCBs were
converted to the fully chlorinated bi-
phenyl, decachlorobiphenyl (DCB), by di-
gestion with SbCls. This procedure has
the advantage of converting all the PCBs
171
-------
to a single peak for Improved quantisation.
The electron capture GC detector Is many
times more sensitive to DCB than it is to
PCBs; thus, the conversion to DCB improves
the sensitivity and lowers the detection
limit for PCBs. Improved perchlorination
procedures have been reported recently by
Griffin et al.HD and Berg et al.<23'
Further improvements in PCB analysis
have been attempted during the project
by experimenting with custom coatings of
glass capillary columns for GC analysis.
The liquid phase used was Emulphor ON-870,
wall coated on 40 meter open tubular glass
capillary columns. The procedures and ap-
pratus were based on the dynamic coating
method of Grob.^24>25)
Generally, liquid phases for capillary
gas chromatography can be divided into ap-
olar, medium polar, and polar phases. Polar
phase columns were notoriously short lived.
Apolar liquid phases (OV-17, OV-101, SE-30,
etc.) were widely accepted as the most ef-
ficient coatings due to their longer life;
these generally are the ones available com-
mercially. However, due to the aromatic
characteristics of PCBs, a polar phase is
needed to achieve maximum separation ef-
ficiency. The merit of the glass surface
pretreatment methods given by Grob(2^»")
is that polar phases such as Emulphor can
be coated on the glass and used for PCB
analysis with increased column life. In
addition, the column can easily be regen-
erated. When a column has deteriorated to
an unusable quality, a solution can be
pulled through the column to wash out the
residual liquid and the column can then be
recoated. In the case of silicone liquid
phases such as OV-17, OV-101, SE-30 etc.,
the regeneration step is not possible.
Thus, capillary columns custom prepared for
PCB analysis have superior separation ef-
ficiency compared to commercially available
columns and have the advantage of being
able to be regenerated.
CURRENT STUDIES OF PCB ATTENUATION
PCB Mobility: Determination by Soil Thin
Layer Chromatography
The technique of determining pesticide
mobility in soils by soil thin-layer chro-
matography was introduced in 1968 by Helling
and Turner.(2°) Since the introduction of
the technique, the mobility of a large num-
ber of pesticides in a variety of soils
has been tested.(2/-Z9) Soll thin_iayer
chromatography, or soil TLC, is a labora-
tory method that uses soil as the adsorbent
phase and a developing solvent (water,
leachate, organic solvent, etc.) in a TLC
system. The system Is relatively simple
and yields quantitative data on the mobil-
ity of organic compounds in soils that ap-
pear to correlate well with trends noted in
the literature.(2°) The results reported
here are mobility data for Aroclor 1242
and Aroclor 1254 on TLC plates made from
sand, three soils, and a coal char. Di-
camba, a pesticide of known high mobility,
was used as an internal standard.
The methods used to prepare soil TLC
plates and to determine the mobility of
PCBs in deionized water has been reported
previously by Griffin et al.^ In ad-
dition to using water as the developing
solvent as reported previously, this paper
reports the results of studies using land-
fill leachate and carbon tetrachloride as
the developing solvent. The leachate was
obtained from the Du Page County sanitary
landfill (well MM63). The details of the
site description and well location were
given by Hughes et al.( Chemical
characterization of the leachate was given
by Griffin et al.^31J The three leaching
solutions—deionized water, carbon tetra-
chloride, and leachate—were each extracted
with hexane and analyzed for trace PCB
using standard GC techniques and were found
to be free from PCB.
Generally, the mobility of PCBs was
measured using radioactive "C labeled
compounds as described by Griffin et al; '
The resulting autoradiographs indicated
the relative movement of the compound,
which was measured as the frontal R, of
the spot or streak. The R£ value is de-
fined as the ratio of the distance the
compound moved relative to the distance
the solvent moved. The Rf value is a
quantitative indication of the front of
PCB movement and a reproducible index of
mobility.
To confirm the data obtained from auto-
radiographs and to obtain data for non-
radioactive PCBs such as capacitor fluid,
a zonal extraction technique was used. An
individual lane of a plate was divided into
12 equal segments starting from 1.5 cm
below the spotting origin of the particular
compount and ending at 10.5 cm above the
origin. The soil material from each seg-
ment on the plate was transferred to a
glass centrifuge tube, and the PCBs were
extracted into suitable organic solvents.
In the case of radiolabeled compounds, the
concentration of compound was determined
with standard liquid scintillation tech-
niques. With non-radiolabeled PCBs, the
172
-------
concentration in the extract was measured
by GC techniques.
Rf values obtained from autoradio-
graphy were compared with those obtained
by zonal extraction. The agreements were
excellent. Autoradiography is especially
satisfactory for soil TLC studies, as it
provides a qualitative picture of movement
(e.g. diffusion, tailing) while allowing
measurement of frontal Rf. Zonal extract-
ion gives a more quantitative picture of
PCB movement. Also detected by the zonal
extraction procedure are tailing, the
origin spot, and the frontal movement of
the compound. However, definition of the
concentration profile is limited by the
length of the soil segment chosen and by
the extraction and analytical efficiencies.
In summary, the two methods gave essen-
tially identical pictures of PCB movement
on TLC plates.
The mobilities of Aroclor 1242, Aroclor
1254, and Dicamba in several earth materials
expressed as frontal Rf values are summar-
ized in Table 1. The data for capacitor
fluid are incomplete and are not shown in
Table 1. However, all the results obtained
to date are identical to those of Aroclor
1242 and will be discussed in that context.
The data show that under the conditions
tested, Aroclor 1242, Aroclor 1254, (and
the capacitor fluid) stayed immobile in
these soil materials when leached with
water or Du Page leachate but were highly
mobile when leached with carbon tetrachlo-
ride. Dicamba showed the reverse trend,
being highly mobile in water and in Du Page
leachate and quite stationary in carbon
tetrachloride.
A closer look at the structure of
Dicamba and PCB will help explain the
mobilities observed. Dicamba is 3, 6-di-
chloro-o-anisic acid having the structural
formula shown below:
With two polar groups—COOH and OCH3,
hydrogen bonding between the water mole-
cules and the carboxyl and methoxy groups
of Dicamba can occur and this increases the
solubility of Dicamba in polar solvents.
Solubility of Dicamba in water is 4,500
ppm. <32>
The PCB structure given in Figure 1
illustrates that PCBs are non-polar in
nature and only very slightly soluble in
polar solvents. Solubility of Aroclor 1242
in water has been determined to be 200
ppb/10) and that of Aroclor 1254, 56 ppbP3}
However, PCBs are quite soluble in organic
solvents such as CgH^, CCl^, CgHg, CH2C12,
and CH3COCH3. Mobilities of Aroclor 1242
and Aroclor 1254 were tested in silica gel
leached with C6H6, C H.,, CH Cl , and
CH3COCH . Consistent with data from soil
plates using CCl^ as the developing solvent,
Rf values of 1.00 were obtained for PCBs
using these solvents on silica gel plates.
It is obvious that mobilities of PCB and
Dicamba in a given soil material are re-
lated to the solubility of PCB and Dicamba
in the solvent with which the soil material
is leached.
The above finding has great signi-
ficance in PCB waste disposal. To pre-
vent potential migration of PCBs in a land-
Table' 1. Mobility of Aroclor 1242, Aroclor 1254, and Dicamba in several earth materials
with various leaching solvents as measured by soil TLC.
Ottawa ailica sand
Catlin Loam CCB
Ava silty clay loam B2
Catlin silt loam Ap
Coal char (1200°F)
Aroclor
1242
.03
.02
.02
.02
.03
H^O
Aroclor
1254
.03
.02
.02
.02
.03
Dicamba
1.00
1.00
1.00
.85
.79
Aroclo
1242
.03
.03
.02
.04
.04
DuFage Leachate
r Aroclor
1254
.03
.03
.02
.04
.04
Dicamba
1.00
1.00
1.00
.90
.80
Aroclor
1242
1.00
1.00
1.00
1.00
1.00
ecu
Aroclor
1254
1.00
1.00
.96
1.00
1 .00
Dicamba
.02
.03
.02
.02
.03
173
-------
Z predicted
Fig. 2. Multiple linear regression of surface area and total organic carbon vs.
adsorption of polychlorinated biphenyl (PCB) by several soils and coal char.
fill, PCB wastes and organic solvents
should not be disposed of in the same land-
fill location,and the PCBs should not be
allowed to come in contact with leaching
organic solvents.
It is interesting to correlate the
mobility of PCBs in CCU with character-
istics of the earth materials, e.g. clay
content, surface area, pH, cation exchange
capacity, organic matter, etc. The zonal
extraction technique yields a quantitative
measure of the quantity of PCB that moves
to the top of each soil TLC plate with the
solvent. Although PCBs have Rf values of
1.00 for all-the soil materials leached
with CC14 in Table 1, the percentage of
PCBs that are retained against leaching
are different for each of the soil mate-
rials. The mobility measured in this
manner is correlated with the soil mate-
rial properties.
Statistical methods were used to cor-
relate soil material properties with the
mobility of PCBs. Multiple linear regres-
sion analyses were performed on the data
for surface area and total organic carbon
of the soil materials vs. the mobility of
PCBs. The following relation was obtained:
Z = -0.59 + 2.85x + 1.15y
where Z » the percentage of Aroclor 1242
retained against leaching by
CC1 on soil TLC plates,
x « the surface area of the sample
in m^/g, and
y = the organic carbon content of
the sample in percent.
A similar relation was found when data
from the adsorption studies, to be re-
ported in the following section, were eval-
uated. The results of a comparison between
Z measured and Z predicted by this relation
are presented in Figure 2. The excellent
linear fit of the data to the theoretical
line indicates that surface area and total
organic carbon content are two major com-
ponents of soil materials that can be
evaluated to allow us to make predictions
of PCB mobility in soils.
PCB Adsorption Studies
Equilibrium adsorption studies were
carried out by shaking known volumes
of PCB saturated water with varying weights
of earth materials at a constant tempera-
ture of 25°C. The rate of adsorption of
PCBs by earth materials was found to be
rapid. Equilibrium conditions were achieved
in less than 8 hr. The adsorption process
conformed to the Freundlich adsorption
equation:
x/m = Kp Cn
where x/m = the amount of PCB adsorption
per unit weight (pg/g) of
adsorbent (soil or char),
174
-------
Kp = the Freundlich adsorption con-
stant which Is an indirect mea-
sure of the adsorption capacity
of the adsorbent,
C = the equilibrium concentration
(yg/Jl) of PCB in solution, and
n " a constant and the value of 1/n
indicates the extent of adsorp-
tion.
The use of this relation allows quan-
titative predictions of PCB adsorption—by
a. given adsorbent—over the concentration
range of water soluble PCBs.
The results of plotting the adsorption
of water-soluble Aroclor 1242 by Ottawa
sand (OS), montmorillonite clay (MC), Catlin
soil (CS), high temperature char (HTC), and
low temperature char (LTC) are shown in Fig-
ure 3, and the constants computed from the
equation are given in Table 2. The value
of r in Table 2 is the linear regression
correlation coefficient for the data plot-
ted in Figure 3.
PCBs were found to be strongly adsorbed
by earth materials. Coal chars had the
highest adsorption capacity, followed by
Table 2. Constants of the Freundlich
isotherm equation
Adsorbent
Ottawa Sand
Montmorillonite Clay
Catlin Soil
High Temp. Char
Low Temp. Char
5.
1.
0.
0.
0.
240
003
686
768
567
0.
0.
1.
1.
1.
190
997
459
302
764
2
0
1
1
7
.994x10"'
.159
.356
.690
.486
0.
0.
0.
0.
0.
731
996
974
924
972
1000-
500-
100-
IOJ
I
n LTC
I
10
50 100
500 1000
Fig. 3. Freundlich isotherm of Aroclor 1242 with different adsorbents at.25°C.
175
-------
5 day blank
5 day degradation
Fig. 4. Comparison of 5-day degradation and 5-day blank chromatograms.
soils, clays, and sand, which had progres-
sively lower adsorption capacities. The
adsorption capacity was found by multiple
linear regression analysis to be correlated
to the organic carbon content and surface
area of the respective earth materials.
Regression coefficients of 0.97 were
obtained, indicating that these were the
two major properties responsible for PCB
adsorption.
Microbial Degradation of PCBs
Because PCBs appear to be somewhat
degradable, attempts have been made in this
study to select and enrich a mixed culture
of soil bacteria that can significantly de-
grade PCBs. We feel that if the mixed cul-
ture, enriched in PCB-degrading organisms,
can significantly degrade PCBs, then bio-
logical treatment units may be feasible to
destroy waste PCBs or PCB wastes could
be inoculated with the culture before land
disposal. The microblal degradation, in
conjunction with other means, can serve as
an important mechanism for attenuating PCBs
and preventing their movement into the bio-
sphere .
Mixed cultures of PCB degrading mic-
robes were isolated from soils and Hudson
River sediments that had previously been
contaminated with PCBs. Much greater act-
ivity of PCB-degrading microorganisms was
found in the river sediment and in soils
that had been previously exposed to PCBs
than in soils that had no previous exposure.
The presence of PCBs in a particular envir-
onment appears to enhance the selection of
microorganisms capable of degrading PCBs.
In the study reported here, a mixed
culture of microorganisms was isolated by
biphenyl enrichment of soil from an exper-
imental farm plot to which PCB contaminated
activated sludge from a municipal waste
treatment plant was applied for 7 years.
We obtained several cores from both the
treated plots and a nontreated plot to
determine if PCBs were being degraded in
the field.
Representative results of laboratory
degradation studies using the sludge-
treated-soil mixed culture and Aroclor
1242 are presented in Figure 4 and Table
3. Figure 4 shows both a chromatogram of
a 5-day control and 5-day degradation;
176
-------
Table 3: Microbial degradation of Aroclor 1242
Peak
number
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
27
28
29
30
31
32
Total
Number
of chlorines
1
1
2
2
2
2
2
3
3
3
2 + 3
3
3
3
3
4
4
4
3
4
4
4
4
4
3
4
4
3
4
After
20 hr.
100
100
74
51
100
98
99
84
49
81
65
87
100
93
75
46
99
54
99
69
37
11
16
16
11
11
41
4
11
92
Percentage
After
5 days
100
100
100
93
100
100
99
88
67
90
71
99
100
98
96
72
99
72
100
"90
42
38
34
39
38
44
81
15
30
97
degradation
After
10 days
100
100
100
99
100
100
99
89
70
90
76
99
100
100
98
75
99
76
100
~90
59
56
49
48
50
42
80
48
48
98
After
15 days
100
100
100
99
100
100
99
89
71
89
78
99
100
100
98
76
99
73
100
~90
56
43
49
67
51
45
85
50
43
98
many peaks are missing or reduced in the
5-day degradation sample.
Table 3 shows the amount of degradation
of each peak when compared to the controls.
The peak number refers to those peaks num-
bered in Figure 4. Peak 1 is an internal
standard used for quantitation and is not
a PCB. The number of chlorines refers to
the chlorine content of a particular PCB
isomer and was assigned to individual peaks
using mass-spectrometer analysis. The
standard deviation of the samples, expres-
sed as percentage, ranged from 1 to 20 per-
cent. Any degradation less than 50 percent
was considered insignificant.
The data presented in Table 3 show that
all of the monochloro-isomers and most of
the dichloro-isomers were degraded in less
than 5 days. Many of the trichloro-isomers
were largely degraded, whereas only a few
of the tetrachloro peaks show these isomers
to have undergone significant degradation.
Because much of the degradation occur-
red in less than 5 days, samples were sub-
sequently run for 20 hours to determine
which peaks were most easily degraded.
The results can be compared to the results
for 5-, 10-, and 15-day degradation (table
3). Both of the major monochlor-isomers
are degraded in 20 hours; however, dich-
loro peaks 4 and 5 are still present to
some degree. Many of the trichloro-peaks
showed much less degradation in 20 hours
than in 5 days. The tetrachloro peaks
showed much less degradation in 20 hours as
compared to the longer degradation times.
177
-------
6- -
5--
4- -
•
DC
2- -
E Soil
D Sludge
LflJUMu,
12 14 16
8 10
Peak Retention Time
—i—
18
Fig. 5. Relative distribution of peaks in soil and sludge extracts.
These results are consistent with previous
evidence that the lower-chlorinated isomers
of PCBs are more easily degraded than the
higher-chlorinated isomers. (^~16'18» 20-22)
These results also were in accord with
the preliminary data obtained from the soil
and sludge extract experiments. Comparison
of the chromatograms of the soil and sludge
extracts with a chromatogram of Aroclor
1242 and Aroclor 1254 illustrated which
peaks were PCBs and which were other com-
pounds. Degradation of PCBs in the sludge-
treated soil could be determined by com-
paring the gas chromatogram of the soil
extract with that of the sludge extract.
Figure 5 shows the results of the soil ex-
tract compared to those of the sludge ex-
tract. The sludge has predominantly low-
chlorine isomers (low retention time),
whereas the soil contains predominantly
high-chlorine isomers (longer retention
time). This indicates that the lower-
chlorinated isomers that were placed in
the soil from the sludge have been degraded,
while the higher-chlorinated isomers, which
are more slowly degraded, have apparently
accumulated in the soil.
Because the results of the soil ex-
traction study are preliminary, we are not
able to estimate quantitatively the amount
of degradation or accumulation in this
soil. Only traces of PCBs were detected
at depths of 6 to 12 inches in the soil
profile, and no PCBs were detected at depths
greater than 12 inches, even though PCB-
contaminated sludge had been applied to
this plot for 7 years. PCBs, therefore,
appear to be highly immobile in this soil.
The conclusion is consistent with the
findings of the mobility and adsorption
studies reported above.
Several points must be clarified when
interpreting the results of this study.
First, a mixed culture of microorganisms
was used that contains at least four
distinct types of microbes. This is in
contrast to the pure cultures of PCB-
degrading microbes used by most other
researchers. Therefore, the degradation
of the PCBs found in this study, more rapid
178
-------
than previously reported degradation, can
be attributed partly to the use of the
mixed culture. For example, in addition
to the PCB-degrading microbes, such a
culture may have other species of micro-
organisms that can degrade the PCB meta-
bolites produced by the PCB-degrading
microbes. The degradation of the meta-
bolites thus leads the way for further
degradation of the PCBs by removing po-
tentially inhibitory products. This mixed
culture approach is closer to the con-
ditions that exist in the environment than
is a pure culture system.
Second, only water-soluble PCBs were
used. Water-soluble and total PCB have
different relative proportions of PCB
isomers.(33«3^ Therefore, 90 percent de-
gradation of water-soluble PCBs is less
than would be achieved for total PCBs.
However, since most potential migration of
PCBs would occur in an aqueous environment,
the degradation of the water-soluble
isomers are clearly the more environment-
ally significant.
Since the mixed culture was enriched
from soil to which PCB-containing sludge
had been applied, and since it appears
that there were already PCB-degrading
microbes present, PCBs should be degraded
to some extent in the environment. This
was indeed found with the soil extracts in
this study. Koeman(^) pointed out that
most PCB residuals in animals yield GC
chromatograms similar to Aroclor 1260, a
rather highly chlorinated series of PCB
isomers. Other researchers'- ' have re-
ported a predominance of the higher-chlor-
inated isomers in environmental samples.
Because Aroclor 1242, a relatively lower-
chlorinated series of isomers, accounted
for the majority of PCBs produced and
marketed, it would appear that the lower-
chlorinated isomers are degraded in the
environment, at least to some extent.
The isolation of microorganisms from soil
that can degrade PCBs readily suggests
the possibility of future inoculation of
wastes containing PCB before disposal or
biological digestion prior to disposal.
ACKNOWLEDGMENTS
The authors gratefully acknowledge
the U.S. Environmental Protection Agency,
Solid and Hazardous Waste Division, Munic-
ipal Environmental Research Laboratory,
Cincinnati, Ohio, for partial support of
this work under Grant No. R-804684-01.
The authors would also like to thank
Dr. Joseph Chou, Dr. Hsi Meng, and Anna K.
Au for assistance with portions of this re-
search, and Dr. T. Kinsley, Department of
Agronomy, University of Illinois, for as-
sistance in collection of soil and sludge
samples.
REFERENCES
1. "PCBs and the Environment," Department
of Health Education and Welfare, NTIS
COM-72-10419, March 1972.
2. Jensen, S., "Report of a New Chemical
Hazard," New Scientist, v. 32, p. 612,
1966.
3. Nelson, N., "Panel on Hazardous Trace
Substances, Polychlorinated Biphenyls-
Environmental Impact," Env. Res., v. 5,
p. 249, 1972.
4. Illinois Institute for Environmental
Quality, Polychlorinated Biphenyls
(PCBs): Health Effects and Recommen-
dations, Env. Health Resource Center
News, no. 20, May 1976.
5. Gustafson, C. G., Ed., Proc. Symp. on
PCB's - Still Prevalent - Still Per-
sistent, 164th American Chemical Soci-
ety Meeting, New York, NY, September
1972.
6. Jordan, D., "The Town Dilema," Environ-
ment, v. 19, no. 2, p. 6, 1977.
7. Moon, D. K., I. W. Leighton, and D. A.
Huebner, "New England PCB Waste Man-
agement Study: New Bedford," Solid
Waste Program, Air and Hazardous
Materials Div., Region I, US-EPA,
J. F. K. Federal Bldg., Boston, MA
02203, 1976.
8. Lidgett, R. A., and H. A. Vodden,
"PCB - The Environmental Problem," In
PCB Conferences, Wenner-Gren Center,
National Environmental Protection
Board, Stockholm, pp. 88-96, Sept. 29,
1970.
9. Robertson, J~. M. , and E. C. C. Li,
"Organic Leachate Threatens Ground-
water Quality," Water and Sewage
Works, pp. 58-59, Feb. 1976.
10. Tucker, E. S., W. J. Litschgi, and
W. M. Mees, "Migration of Poly-
179
-------
chlorinated Biphenyls in Soil Induced
by Percolating Water," Bull. Environ.
Contam. and Toxicol. v. 13, pp. 86-93,
1975.
11. Griffin, R. A., F. B. DeWalle,
E. S. K. Chain, J. H. Kim, and A. K.
Au, "Attenuation of PCB's by soil
materials and char wastes," In "Man-
agement of Gas and Leachate in Land-
fills," S. K. Banerji, Ed., Ecol-
ogical Research Series, EPA-600/9-77-
026, U.S. Environmental Protection
Agency, Cincinnati, Ohio 45268, pp.
208-217, 1977.
12. Koeman, J. H., M. C. T. N. DeBrauw,
and R. H. DeVos, "Chlorinated Bi-
phenyls in Fish, Mussels and Birds
from the River Rhine and the Nether-
lands Coastal Area," Nature, v. 221,
p. 1126, 1969.
13. Wallnofer, P. R., G. Engelhardt, S.
Safe, and 0. Hutzinger, "Microbial
Hydroxylation of 4-chlorobiphenyl and
4, 4'-dichlorobiphenyl," Chemosphere,
v. 2, p. 69, 1973.
14. Baxter, R. A., P. E. Gilbert, R. A.
Lidgett, J. H. Mainprize, and H. A.
Vodden, "The Degradation of Polychlo-
rinated Biphenyls by Microorganisms,"
Sci. Total Environ., v. 4, p. 53,
1976.
15. Sayler, G. S., M. Shou, and R. R.
Colwell, "Growth of an Estvarine
Fseudomonas sp. on Polychlorinated
Biphenyls," Microbial Ecology, v. 3,
p. 241, 1977.
16. Sundstrom, G., 0. Hutzinger, and S.
Safe, "The Metabolism of Chloro-
biphenyls," Chemosphere, v. 5, p. 267,
1976.
17. Herbst, E., I. Scheunert, W. Klein,
and F. Korte, "Fate of PCBs - 14C
in Sewage Treatment - Laboratory
Experiments with Activated Sludge,"
Chemosphere, v. 6, p. 725, 1977.
18. Tucker, E. S., V. W. Saeger, and 0.
Hicks, "Activated Sludge Primary Bio-
degradation of Polychlorinated Bi-
phenyls," Bull. Environ. Contam. and
Toxicol.. v. 14, p. 705, 1975.
19. Mihashi, K., T. Hashinaya, M. Dazai,
J. Ito, and T. Misono, "Biodegrada-
bility of Polychlorobiphenyl," Kogyo
Gijustsuia Biseibutsu Kogyo Gijustsu
Kenkyusho Kenkyu Hokoko, v. 46, p. 65,
1975.
20. Ahmed, M., and D. D. Focht, "Degrada-
tion of Polychlorinated Biphenyls by
Two Species of Achromobacter," Can. J.
Microbiol., v. 19, p. 47, 1973.
21. Ahmed, M., and D. D. Focht, "Oxidation
of Polychlorinated Biphenyls by Achro-
mobacter PCB," Bull. Environ. Contam.
and Toxicol., v. 10, no. 2, p. 70,
1973.
22. Furukawa, K., and F. Matsumura, "Micro-
bial Metabolism of Polychlorinated
Biphenyls: Studies on the Relative
Degradability of Polychlorinated Bi-
phenyl Components by Alkaligenes sp.,"
£. Agric. Food Chem., v. 24, no. 2,
p. 251, 1976.
23. Berg, 0. W., A. V. Rees, and M. S.
Ali, "Simplified Techniques for the
Perchlorination of Polychlorinated
Biphenyls in Environmental Samples,"
Abstracts from the 29th Annual Pitts-
burgh Conference, March 2, 1978,
Clevland, Ohio, p. 579, 1978.
24. Grob, K., "Glass capillary columns in
gas chromatography, a tool and a tech-
nique," Chromatographia, v. 8, p. 423,
1975.
25. Grob, K., G. Grob, and K. Grob, Jr.,
"New improved carbonate deposition
technique for coating wall coated
open tubular columns," Chromatographia.
v. 10, p. 181, 1977. •
26. Helling, C. S., and B. C. Turner,
"Pesticide Mobility: Determination by
Soil Thin-Layer Chromatography,"
Science, 162, pp. 562-563, 1968.
27. Helling, C. S., "Pesticide Mobility in
Soils I. Parameters of Thin-Layer
Chromatography," Soil Sci. Soc. Am.
Proc.. v. 35, pp. 732-737, T9~7l~
28. Helling, C. S., "Pesticide Mobility
in Soils II. Applications of Soil
Thin-Layer Chromatography," Soil Sci.
Soc. Am. Proc., v. 35, pp. 737-743,
1971.
180
-------
29. Helling, C. S., "Pesticide Mobility
in Soils III. Influence of Soil Prop-
erties, " Soil Sci. Soc. Am. Proc.,
v. 35, pp. 743-747, 1971.
30. Hughes, G. M., R. A. Landon, and R. N.
Farvolden, "Hydrogeology of Solid
Waste Disposal Sites in Northeastern
Illinois," U.S. Environmental Protec-
tion Agency, Solid Waste Management
Series, Report SW-12d, 154 p., 1971.
31. Griffin, R. A., R. R. Frost, A. K.
Au, G. D. Robinson, and N. F. Shimp,
"Attenuation of Pollutants in Munic-
ipal Landfill Leachate by Clay Min-
erals: Part 2 — Heavy-metal Ad-
sorption," Environmental Geology
Notes Number 79, Illinois State Geo-
logical Survey, Urbana, Illinois, 1977.
32. Herbicide Handbook, Weed Society of
America, 425 Illinois Bldg., 113 N.
Neil St., Champaign, Illinois, 3rd
Ed., 1974.
33. Haque, R., D. W. Schmedding and V. H.
Freed, "Aqueous Solubility, Adsorp-
tion, and Vapor Behavior of Polychlo-
rinated Biphenyl Aroclor 1254," Env.
Sci. and Tech., v. 8, no. 2, February
1974
34. Griffin, R. A. et al., "Predicting
Organic Contaminant Removal," 2nd
Quarterly Report, US-EPA Grant No.
R-804684-010, Prepared for: Richard
Carnes, Project Officer, Municipal
Environmental Research Lab, US-EPA,
Cincinnati, Ohio.
35. Smith, K., Illinois Natural History
Survey, University of Illinois at
Champaign-Urbana, personal communica-
tion.
181
-------
Land Disposal of Hexachlorobenzene Wastes:
Controlling Vapor Movement in Soils
by
Walter J. Fanner, M. S. Yang, and J. Letey
Department of Soil and Environmental Sciences
University of California, Riverside
Riverside, California
W. F. Spencer
Agricultural Research Service, U.S.D.A.
Riverside, California
and
Mike H. Roulier
Solid and Hazardous Waste Research Division
Municipal Environmental Research Laboratory
U.S. Environmental Protection Agency
Cincinnati, Ohio
ABSTRACT
The disposal/storage of potentially volatile hazardous waste on land suggests the need
for developing information on controlling vapor movement in soils. This information would
be useful for the purpose of limiting vapor phase transport through a soil cover to the
surrounding atmosphere to an acceptable level. Hexachlorobenzene-containing industrial
waste (hex waste) was used as the source of a volatile compound in a laboratory study of
the important soil parameters controlling vapor movement. Volatilization flux through a
soil cover was directly related to soil air-filled porosity and was greatly reduced by
increased soil compaction and increased soil water content. This paper reports on the
application of these research findings in developing predictive equations for use in
designing landfill covers.
The soil has proven to have tre-
mendous capacity for neutralizing
certain of the wastes produced by
society. When properly managed this
waste-handling capability can be util-
ized to the fullest. Likewise when mis-
managed the consequences can be severe,
expensive and long-term. One of the
management tools required for the
utilization of land as a waste disposal
medium is the ability to control
volatilization of wastes in landfill.
This project was designed to gather
laboratory data useful to the"landfill
planner in controlling vapor movement
through the soil cover of a landfill.
The study was carried out using hexa-
chlorobenzene (HCB) from HCB-containing
industrial wastes as a model for the
volatilizing compound.
Hexachlorobenzene is present in
industrial wastes as a by-product of
certain commercial production processes.
A recent EPA report^ 'indicated over
3900 metric tons (8.5 million pounds) of
HCB produced annually. This HCB is con-
tained in over 24,000 metric tons (53
million pounds) of industrial waste, a
significant amount of which is disposed
182
-------
of on landfill. Other methods currently
being used for disposal of HCB-containing
wastes (hex waste) range from temporary
storage in water covered lagoons to more
permanent storage via deep well injection.
Successful incineration in highly
efficient devices has been demonstrated
and is in use. The major source of hex
waste (ca. 90%) is as a by-product in
the commercial production of several
chlorinated solvents such as
perchloroethylene and carbon tetra-
chloride. In addition significant
quantities of HCB are produced as
impurities or by-products in the produc-
tion of chlorine gas and in the produc-
tion of certain pesticides.
HCB appears to be widespread in the
environment. Residues have been found in
soil, wildlife, fish andjEood samples
as well as in humans. *• ' ' ; The source
of this HCB is not always known as the
residues are reported across the United
States as well as around the world.
Widespread contamination of cattle, soil,
plant and humans was discovered in
Southern Louisiana in 1973 near HCB
production facilities and municipal land-
fill facilities which accepted hex waste.
In past Symposia ' ' we have
reported on a number of aspects of land
disposal of hexachlorobenzene wastes.
These included a statement of the prob-
lem, properties of HCB, and experimental
results dealing with transport of HCB in
soil especially vapor transport. Follow-
ing a brief review of earlier reports,
this Symposia will report on the appli-
cation of research results to designing
a landfill cover for reducing HCB
volatilization.
Hexachlorobenzene is a persistant,
water-insoluble, fat-soluble, organic
compound. Because of its low water
solubility (6.2 ug/1), transport in
moving water will be negligible. Its
long term persistance and appreciable
vapor pressure (1.91 x 10~5 mm Hg at
25 C) allows significant volatilization
to occur. The potential for volatiliza-
tion indicates a need for disposal site
coverings that will reduce the vapor
phase transport of HCB into the surround-
ing atmosphere.
Experimental procedures were devel-
oped and evaluated in our laboratory
using a simulated landfill to determine
the parameters necessary to predict HCB
volatilization from hex waste placed
under a soil cover. These procedures
and results have been detailed in a
previous symposium.' ' Briefly it was
found that soil depth and soil air-
filled porosity were the two prime
factors controlling vapor movement in
soil. This relationship between soil
depth, soil air-filled porosity, and
vapor flux is a direct result of the
nature of vapor phase diffusion in soil.
Since HCB is essentially insoluble in
water, diffusion in soil pores will be
the only mechanism available to HCB
molecules for movement. Furthermore,
because of the much greater quantity of
HCB in the vapor phase compared to that
in solution, HCB diffusion will be pri-
marily in the vapor phase. This leaves
us with a diffusion controlled process
involving only air-filled pores. This
will be the process controlling movement
of HCB molecules to the soil surface of
a landfill cover. Once the HCB molecules
reach the soil surface, they will be
carried rapidly away by air currents.
The volatilization or vapor loss of
HCB and other compounds from landfill can
be treated as a diffusion controlled
process. The rate at which compounds
will volatilize from the soil surface and
be lost to the atmosphere will be con-
trolled by the rate at which they diffuse
through the soil cover over the waste.
Assuming no degradation of the compound
and no transport in moving water, the
volatilization can be predicted using
Fick's First Law for steady state dif-
fusion:
-Dg(C2-Cs)/L
(1)
where J is the vapor flux from the soil
surface (ng/cm /day),
D is the apparent steady state
diffusion coefficient (cnr/day),
C2 is the concentration of the
volatilizing material in air
or vapor density at the surface
of the soil (ug/1),
Cs is the concentration of the
volatilizing material in air or
vapor density at the bottom of
the soil layer (ug/1), and
L is the soil depth.
183
-------
The negative sign is present to indicate
that the vapor flux is in the opposite
direction from the vapor concentration
gradient in the soil. Table 1 gives
several conversion factors which may be
useful depending on whether regulatory
agencies specify flux in metric or
English units.
Table 1. Conversion Factors
To convert
atmospheres
centimeters
O
cm /day
feet
g/cra3
kg /ha
mm Hg
ng/cm^
ng/cm
into
mm Hg
ft
mm /week
cm
lbs/ftJ
Ibs/acre
atm
kg /ha
Ibs/acre
multiply by
760
3.28 x 10
700
30.48
62.4
0.89
1.32 x 10~J
10
8.9 x 10
In order to use Equation (1) for
predicting volatilization, the apparent
diffusion coefficient, Dg , must be
evaluated. The investigators, Millington
and Quirk (?) have suggested an
apparent diffusion coefficient which in-
cluded a porosity term to account for the
geometric effects of soil on diffusion.
(2)
where D is the vapor diffusion
° coefficient in air (cm /day),
P is the soil air-filled
porosity (cnr/cm ), and
P is the total soil porosity.
Combining Equations (1) and (2)
yields the following expression:
(3)
This equation will be used in this paper
as the basis for a suggested step-wise
procedure intended to assist a planner in
designing a landfill cover that minimizes
the escape of HCB or other vapors.
Alternatively, an existing landfill cover
may be assessed, using Equation (3), for
its potential for allowing HCB vapor flux
through the surface of the landfill. The
validity of Equation (3) has been experi-
mentally verified for hexachlorobenzene
in our laboratory using a simulated land-
fill and HCB-containing industrial waste
as the volatilizing compound. ^) The
diffusion coefficient in air, Do, for
HCB was found to be 1 x 104 cm /day.
This procedure for assisting in the
design of a landfill cover is only a
suggested procedure and is not an accepted
official EPA procedure. This will be an
example of how research findings can be
used to arrive at a suggested set of
procedures that will assist planners in
designing landfill covers taking advantage
of the best current knowledge available
for reducing vapor flux from a landfill.
DESIGN APPLICATION
In order to use the results of this
study in designing a proposed landfill,
the planner will normally begin with an
acceptable value for HCB flux through a
cover and determine the most efficient
combination of soil porosity and soil
depth to produce the acceptable value.
The establishment of the actual values
for an acceptable flux thru the land-
fill cover is beyond the scope of this
research or that of the landfill designer.
Flux from the soil will be established
through regulations by state or federal
agencies. Alternatively the regulating
agencies may establish acceptable air
concentration values at some specified
distance from the landfill site. In the
latter case the landfill designer may
have to calculate the acceptable flux at
the soil surface using the specified
air concentration and existing wind
dispersion models. '
Once an acceptable flux value has
been established either directly by a
regulatory agency or indirectly by use
of an air dispersion model, the follow-
ing steps can be followed to determine
what soil conditions would limit flux to
this value. The following steps are
based on Equation (3) relating vapor flux
through soil to soil depth and soil
porosity. Initially the soil depth nec-
184
-------
essary to produce the desired flux will
be calculated assuming a dry soil and a
minimum reasonable compaction of the soil
at the landfill site. If this calculated
soil depth is unrealistically high, then
increased compaction and/or an appro-
priate water content will be considered.
Finally a modified soil will be consi-
dered when necessary and a new flux value
calculated. Equation (3) is rearranged
below to allow calculation of L, the soil
depth (cm).
L - -D (P 10/3/P 2>(C -C )/J
o a T 2s
(4)
By assuming that C. is zero, Equation
(4) simplifies as follows:
L - D C P 10/3/JP 2
o s a T
(5)
This is a reasonable assumption since
under actual landfill conditions the
amount of vapor reaching the soil surface
from below hopefully will be small and
will be rapidly dispersed by wind currents
and by diffusion in air. Additionally
the assumption of C equaling zero intro-
duces a safety factor into the calcula-
tion. Any increase in ^ w*ll reduce the
vapor flux from the soil surface.
The diffusion coefficient in air for
HCB has been measured to be 1.0 x 10^
cnrVday. If soil cover is being designed
for some material other than HCB, values
for" diffusion coefficients in air are
available, for a limited number of com-
pounds. For those compounds of inter-
est for which no published values are
available, reasonable estimates can be
made using the following. The vapor
diffusion coefficient of compound A
can be estimated from the known vapor
diffusion coefficient of compound B
by using the following equation:
(6)
where the subscripts A and B denote the
values of the diffusion coefficients for
compounds with molecular weights H. and
>L, respectively.
When the temperature changes from T
to T-, where T is the absolute tempera-
ture (°K), the diffusion coefficient at
temperature T2 can be estimated using
the following equation:
(7)
where the subscripts 1 and 2 denote the
values of the diffusion coefficients at
the temperatures T. and T_, respectively.
The value for the vapor concentration
at the bottom of the soil layer, C , can
be taken as that equivalent to the satur-
ation vapor pressure of the pure compound.
This is true because a waste need contain
only a relatively small amount of a
volatile material to give a saturated
vapor. In the case of HCB, a saturated
vapor density was obtained with only 2%
HCB coated on sand. It is conservative
to assume Cs equal to the saturation
vapor pressure. If the actual vapor
pressure is less than saturation, the
actual vapor flux through the surface of
the soil layer will be less than calcu-
lated. Saturation vapor densities are
calculated from the following using vapor
presure data which are available for a
number of compounds. (*>10)
Cs - pM/RT
(8)
where p is the vapor pressure (mm Hg),
M is the molecular weight of the
compound (ug/mole),
R is the molar gas constant
(1 mm Hg/deg mole), and
T is the absolute temperature (°K).
The value of Cs for HCB is 0.294 ug/1
at 25 C calculated from a vapor pressure
of 1.91 x 10~5 mm Hg.
Based on the above considerations
the following steps are to be followed
in calculating the optimum combination of
soil depth and soil porosity to achieve
the desired vapor flux. A suggested
185
-------
method or sequence for working through
these steps is shown in Figure 1.
1. Base all flux calculations on
total soil porosity assuming a dry soil.
Thus Pa - Pj and Equation (5) is further
simplified to
The addition of any water to the soil by
irrigation or by natural rainfall will
reduce the air-filled porosity and there-
by reduce the vapor flux from the soil
surface. The total porosity of the
soil can be calculated from the soil bulk
density, 0, using the following equation:
D0CsPT4/3/J
(9)
i-
(10)
This not only simplifies the calculations
but introduces another safety margin.
where 0 = soil bulk density (g/cm ) and
p = particle density (g/cm^).
Estimate MAXIMUM density for soil cover and
calculate the corresponding porosity assuming
that all porosity is air-filled (Eqn. 10)
Is data avaiable
to estimate
minimum water
content ? *
•* NO
-YES
Using the porosity corresponding
to Maximum density, us* Eqn. II
and the minimum water content
to calculate air-filled porosity.
Are soils or soil
materials (e.g.
Bentonite) available
for modifying or
substituting for •
on-site soils ?
-NO-
l
r
•i-NO
Using the air-filled
porosity and Eqn. 5,
is the required soil
depth technically and
economically feasible?
YES-
Estimote MINIMUM reasonable density for
soil cover and calculate corresponding porosity
(Eqn. 10) assuming all porosity is air filled.
NO
Using the porosity corresponding
to Minimum density, use Eqn. II
and the minimum water content
to calculate air-filled porosity.
I NO
Using the air-filled
porosity and Eqn. 5.
is the required soil
depth technically and
economically feasible?
YES-
Repeat process for
modified cover material
Soil cover will not limit flux
to acceptable value.
Seek other method of dealing
with waste.
Develop landfill
design plans
»4/ this stage, could consider irrigation or
other treatment to maintain higher water
content (if allowed by regulatory agency)
Figure 1. Flow diagram for predicting depth of soil cover required to
limit vapor flux through soil cover to an acceptable value.
186
-------
The particle density can be measured.
However, the particle density for most
soil mineral material is usually taken as
2.65 g/cm3 (165 Ibs/cu. ft.).
2. Establish a minimum and maximum
soil porosity that is likely to be
achieved through maximum and minimum
compaction of the soil type available at
the proposed landfill site. Standard-
ized procedures such as the Procter
(ASTM D-698) ^;| and modified Procter
(ASTM D-1557) ^ ' tests are available
for estimating the densities that can be
achieved for a given soil with various
degrees of effort. The degree of com-
paction need not be limited to either
minimum or maximum as intermediate
values may be calculated if necessary to
determine the precise limiting density.
3. Using the porosity value estab-
lished in Step 2 above, Equation (9) is
used to determine the depth of soil nec-
essary to attain the acceptable vapor
flux value through the soil surface.
4. If circumstances allow an esti-
mate of long-term soil water contents in
the landfill cover, the planner can re-
fine the flux calculations to give a re-
duced flux due to reduced air-filled
porosity. This refinement required the
use of Equation (5), taking into account
the air-filled porosity as affected by
soil water content. The air-filled
porosity, Pa> is calculated from the
total porosity, PT, and the volumetric
soil water content, 0 , by the following:
5. If the soil depths calculated
in the previous steps prove to be too
great to be technically or economically
feasible then the planner will need to
decide if it is possible to replace or
modify the on-site soil with other soil
materials (e.g., bentonite) to increase
soil bulk density. If soil modification
is feasible than it will be necessary
to go back and calculate a new soil
porosity term, and thus a new flux value,
using the higher bulk density.
6. If none of the above procedures
will produce a soil cover that will limit
the vapor flux to an acceptable value,
other disposal methods may have to be
considered for dealing with the waste.
ASSESSMENT APPLICATION
In the case where an existing land-
fill site has a history of accepting and
disposing of HCB-containing materials or
other waste of similar concern, it may be
desirable to evaluate the cover and to
estimate the HCB vapor flux through the
surface of the soil to determine if the
landfill cover should be redesigned.
Equation ( 9) is rearranged as
follows to calculate vapor flux through
a dry soil cover of depth, L, and total
porosity, PT.
J -
(13)
PT - fl
(11)
The volumetric water content (cm3/cnr) is
determined from the gravimetric soil
water content as follows:
we/
(12)
The same considerations apply to the use
of this equation for calculating flux as
applied to Equation (9 ) for calculating
soil depth:
1. The diffusion
for HCB is 1.0 x 10 cmZ/day.
oefficient in air
For other
compounds, a value for Do can be estima-
ted if the actual value has not been
determined.
where W is the gravimetric soil water con-
tent (g/g), and p is the density of
water (g/cm-*). For practical purposes
the density of water can be taken as one
and the volumetric water content is simply
the gravimetric water content multiplied
by the soil bulk, density, p.
2. The concentration in air at the
soil surface, C2, is assumed to be zero.
As discussed before, this is a reasonable
assumption and introduces a safety factor
into the calculation of a maximum value
for the vapor flux through the soil
surface.
187
-------
3. The concentration in air at the
bottom of the soil layer, C , is the
saturation vapor concentration of the
compound calculated from its vapor
pressure data.
4. In making these calculations a
dry soil is assumed so that Pfl - P^,.
5. If data is available on soil
water content, Equation (5) can be used
to estimate vapor flux.
DISCUSSION
Figures 2 and 3 illustrate the use
of Equation (13) for predicting vapor
PREDICTED HCB VOLATILIZATION
PREDICTED HCB VOLATILIZATION
1.0
o
£
v.
s
0.1
00
u
I
0.01
I
I
I I I I
I i
2.3 Z2 2.1
SOIL BULK
2.0 1.9 1.8 17 1.6 1.4
DENSITY
1.2
Figure 2. Predicted hexachlorobenzene
volatilization fluxes through a soil
cover of various soil bulk densities
and soil thicknesses. The soil is
assumed to be dry in order to yield
a maximum vapor flux through the soil
cover.
flux at different depths of soil cover.
Figure 2 shows the HCB volatilization
fluxes through soil cover with dry soil
at various bulk densities and thick-
nesses. Soil temperature was taken as
25 C. The vapor fluxes shown in Figure
2 are the maximum to be expected through
the soil cover because the addition of
water would reduce the air-filled
o
C.
0,001
Using Eqn. (3)
Assuming Co = 0
0.01 —
0.40
0.30
SOIL WATER
0.20 0.10
CONTENT (g/g)
Figure 3. Predicted hexachlorobenzene
volatilization fluxes through a soil
cover of various soil water contents
and soil thicknesses. Soil bulk
density is 1.2 g/cm3 (74.9 Ibs/cu ft).
porosity and thus reduce the vapor flux
through the soil. Figure 3 shows the
predicted HCB vapor flux through a soil
cover at various soil water contents and
soil depths with the soil bulk density
held constant at 1.2 g/cm2 (74.9 Ibs/
cu. ft.).
These procedures are written for
application to disposal of HCB-
containing wastes. However, these same
methods can be used to aid in designing
landfill covers for other compounds as
well, subject to the following quali-
fications. Hexachlorobenzene degrades
very slowly, if at all. Since HCB is
very slightly soluble (6.2 ug/1) its
transport by moving water is negligible.
For compounds which degrade more
readily or are more mobile in moving
water than HCB, these procedures will
tend to overestimate the actual flux
through the soil cover. These proce-
dures assume no movement of the compound
in water which may be percolating through
the soil profile. If a compound is more
188
-------
soluble in water than HCB, it may eventu-
ally move with the water its actual
mobility being dependent upon the extent
it is adsorbed by the soil materials.
Assuming that net water flow will be
downward in soil, mobility in water will
move the compound away from the surface
of the soil cover. These procedures
assume no decomposition of the material.
Any decomposition which may occur will
serve as a sink for the compound and
will decrease the amount escaping as a
vapor. The procedures presented here,
therefore, will not account for any re-
duction in vapor flux due to decomposi-
tion of the compound or due to transport
of the compound in water.
If hex wastes are to be placed on
land, our work has shown the importance
of minimizing the air-filled porosity
of soil covers. However, our contribu-
tion to the design of a landfill assumes
an intact soil cover is maintained. If
any cracks or other small openings de-
velop in the soil cover, they will
result in an appreciable increase in HCB
flux through the soil cover. The place-
ment of hex waste with any material, such
as municipal solid waste, that are sub-
ject to settlement could cause such
cracking and flux increase.
If hex wastes are placed on land,
consideration should be given to the
need for long-term arrangements for en-
suring the integrity of soil covers.
Calculations assuming no degradation
indicate that HCB placed on land could
continue to volatilize at a maximum
rate for several centuries. The inte-
grity of the soil cover must be main-
tained for this period by preventing
such things as erosion or digging.
In addition to soil as a cover, we
have examined the efficiency of water and
polyethylene film as covers for reducing
vapor movement of HCB. Both water and
polyethylene film have been used in actual
disposal of hex waste. Water lagoons have
been used for the temporary storage of
hex waste and polyethylene films have been
used in conjunction with a soil cover.
In our studies water was highly effective
in reducing volatilization of HCB. Poly-
ethylene film was about as effective as
water. However, the cost of polyethylene
film would preclude its use in layers
thick enough to significantly retard HCB
flux.
Some hex waste were found to contain
a dense organic liquid in which HCB is
highly soluble (23,000 ug/ml). If this
liquid waste drains from the waste, the
potential would be created for substantial
amounts of HCB to move through the soil.
REFERENCES
1. Quinlivan, S., M. Ghassemi, and M.
Santy. "Survey of methods used to
control wastes containing hexa-
chlorobenzene," U.S. Environmental
Protection Agency, Office of Solid
Waste Management Programs,
Washington, D.C. EPA/530/SW-120C,
1976.
2. "An ecological study of hexachloro-
benzene", EPA, Office of Toxic
Substances, EPA 560/6-7-009, April,
1976.
3. "Summary characterizations of select-
ed chemicals of near-term interest,"
EPA, Office of Toxic Substances,
EPA 560/4-76-004, April, 1976.
4. Burns, J. E. and F. E. Miller.
"Hexachlo robenz ene contamina t ion:
Its effects in a Louisiana popula-
tion." Arch. Environ. Health 1975,
v. 30, pp. 44-48.
5. Farmer, W. J., M. Yang, J. Letey, and
W. F. Spencer, "Problems associated
with the land disposal of an organic
industrial hazardous waste contain-
ing HCB." Proceedings of the
Hazardous Waste Research Symposium
on Residual Management by Land
Disposal, Tucson, Arizona, EPA,
Office of Research and Development,
EPA-600/9-76-015, July, 1976.
6. Farmer, W. J., M. Yang, J. Letey, and
W. F. Spencer, "Land disposal of
organic hazardous wastes containing
HCB," Proceedings of the National
Conference on Disposal of Residues
on Land, St. Louis, Missouri, EPA,
Office of Research and Development,
Published by Information Transfer
Inc., Rockville, Maryland, 1977.
7. Farmer, W. J., M. Yang, J. Letey, and
W. F. Spencer, "Land disposal of
hexachlorobenzene wastes: controlling
vapor movement in soil," EPA, Office
of Research and Development (Final
189
-------
Report under EPA Contract No.
03-2014).
68-
9.
"Chemicals in the air: The atmos-
pheric system and dispersal of
chemicals," In Environmental
Dynamics of Pesticides, R. Haque
and V. H. Freed, editors, Plenum
Press, New York and London, 1975,
pp. 5-16.
Handbook of Chemistry and Physics.
R. C. Weast, editor. 53rd ed. CRC
Press, Inc., Cleveland, Ohio. 1973.
10. Spencer, W. F. and M. M. Cliath, "The
solid-air interface: transfer of
organic pollutants between the solid-
air interface," In Fate of Pollutants
in the Air and Water Environments,
Part I, I. H. Suffet, editor. John
Wiley & Sons, New York, 1977, pp.
107-126.
11. American Society for Testing and
Materials, Philadelphia, PA,
"Tests for moisture-density relations
in soils," ASTM-D-698-70 and ASTM-D-
1557-70, v. 19, 1977.
190
-------
SIMULATION MODELS AND THEIR APPLICATION TO LANDFILL DISPOSAL SITING;
A REVIEW OF CURRENT TECHNOLOGY
M. Th. van Genuchten
Department of Civil Engineering
Princeton University
Princeton, N.J. 085^0
ABSTRACT
Predicting potential groundwater pollution associated with the land disposal
of liquid and solid wastes is a highly complex technological adventure. The simultaneous
presence of numerous interactive mechanisms, physical, chemical as well as biological,
makes it difficult to obtain an advanced picture of the possible pollution of a given
waste deposited in a given geohydrological environment. Consequently, many investigators
have resorted to the use of models to forecast the performance of waste disposal sites.
This paper reviews some of the models currently available for such predictions. Main
emphasis is thereby put on a discussion of existing conceptual-mathematical models, since
these models appear the most promising but, unfortunately, also the most complex ones
for evaluating groundwater pollution. An extensive list of available simulation models
is given, and some of the advantages and disadvantages of these models for simulating
landfill behavior are discussed.
INTRODUCTION
Whenever liquid or solid wastes are
deposited on land, leachates may be genera-
ted which can seriously affect the quality
of underlying groundwater systems. Pre-
dicting such groundwater pollution is a
highly complex technological undertaking.
The simultaneous presence of numerous in-
teractive mechanisms (physical, chemical
and biological) makes it very difficult
to obtain a clear picture of the pollution
potential of a given waste when deposited
in a specific geohydrological environment.
Many investigators have therefore resorted
to the use of "models" for predicting the
impact of proposed waste disposal sites.
The purpose of this paper is to discuss
some erf the advantages and limitations of
the use of "models" for such predictions.
Two definitions pertinent to this dis-
cussion are given first.
A waste disposal system (e.g., a
landfill or lagoon) is here identified with
the set of physical, chemical and biologi-
cal processes which act upon a number of
input variables (precipitation-evaporation,
amount and type of waste), and subsequently
convert these into output variables
(amount and type of leachate leaving the
landfill, pollutant concentration in ground-
water, etc.). From a management viewpoint,
the waste disposal system should, in
addition to the disposal site itself, also
include the groundwater aquifer under and
immediately downgradient of the site. A
waste disposal model is defined as a sim-
plified representation of the real waste
disposal system. As a result of simplifica-
tions, different types of models exist. For
example, a scaled-down replica of the
system is as much a model of the system as
is a highly sophisticated mathematical
model based upon a set of partial differ-
entia) equations. Even when an engineer
evaluates a proposed site and uses his
experience to make a decision regarding
the suitability of the site for waste dis-
posal, he most likely applies a "model"
to the site. Obviously, the model in this
case will be highly descriptive in nature
(subjective engineering judgement is used
as a decision tool).
MODEL CLASSIFICATIONS
Models can be classified in several
ways. A possible classification is given
below (for a more general discussion of
models and simulation procedures, see
Maisel and Gnugnoli, 1972, or Fishman,
191
-------
1973).
1. Descriptive models. These models are
expressed in one's "native language"
(Emshoff and Sisson, 1970). A descriptive
model results when an expert does not rely
upon well-defined procedures, but uses
qualitative engineering judgement to
evaluate the suitability of a proposed site
for waste disposal. An obvious advantage
of this type of model is its low cost. The
greatest limitation of a descriptive model,
however, is that its predictions are very
subjective and therefore difficult to
communicate. Different experts may easily
reach different conclusions when inter-
preting the model results.
2. Physical models. Physical models are
those which represent scaled-down versions
of the true conditions (e.g., the globe is
a physical model of the earth).Unfortunatly
only a few physical landfill models have
been constructed in the past (Quasim, 1965;
Pohland, 1975). The laboratory and scaled-
down field landfills built by Drexel
University in cooperation with the
Pennsylvania Department of Health
(Fungaroli and Steiner, 1973) represent
interesting examples. The laboratory land-
fill was operated under controlled environ-
mental conditions, and the field facility
under natural conditions. Although these
scaled-down facilities were constructed
primarily to study landfill behavior as
such, the field site can be viewed as a
physical model for the. landfill later con-
structed in the immediate vicinity of the
experimental site. In fact, the field
facility may still be regarded as a
physical model for other landfills (in
Pennsylvania or elsewhere), provided the
geohydrological environment remains
essentially the same, and similar wastes
and management procedures are applied to
the landfill. Generally, however, extrapo-
lation outside the region of study is
difficult due to the presence of unique
local conditions (wastes, soils, hydrology,
management)•
Although physical models of waste
disposal sites are generally lacking, many
experiments can be conducted to aid the
engineer in making more accurate
predictions. Data may be generated,
either through field or laboratory experi-
mentation, which can be used to assess
the behavior of specific waste constituents.
Such experimentation may include column
leaching studies to determine the rate at
which certain chemicals move through a
given soil, thin-layer chromatography
leading also to estimates of the rate of
migration of certain chemicals, or batch
equilibrium studies to characterize
chemical adsorption onto soils. Unfortu-
nately, this information does not in and
of itself define a waste disposal "model",
and as such cannot be used as a predictive
tool.
While it is obvious that scaled-down
(physical) models can provide useful in-
formation about the type and concentration
of chemicals expected from certain waste-
soil combinations, their utility as a
predictive tool appears doubtful. They are
not only costly to build, but also very
time-consuming to study, especiaMy when
one considers that the different chemical
and biological processes within a landfill
may operate over a period of several years
or decades.
3. Analog models. Analog models employ a
convenient transformation of a given
physical system into another one which
behaves in a similar manner, but which is
more easily measured. The problem under
investigation is then solved in the sub-
stitute state and the answer translated
back into properties of the original
system. Electrical analog models have found
widespread application in groundwater flow
modeling. Here electronic devices and
properties (diodes, varistors, currents,
voltages) are used to simulate groundwater
flow. The analogy here, of course, is
between electrical current and porous media
flow. Because of the high cost of building
electrical analog models for large-scale
field problems, it appears doubtful that
many such models will be constructed in
the near future for simulating groundwater
quality problems.
4. Mathematical models. Mathematical
models use concise mathematical expressions
for description of the physical system.
Generally a set of mathematical equations
is used to describe the relationships
between the various system parameters and
their input and output variables.
Depending upon the method of analysis, this
type of model may range from a few simple
equations (criteria ranking) to thousands
of complex mathematical expressions which
can be solved only through the use of a
digital computer. In the latter case, a
governing set of partial differential
equations is generally derived based on
physical and chemical principles, and
subsequently solved using either analytical
192
-------
or numerical techniques. This modeling
approach will be given more attention
below.
In addition to the above classifica-
tion, several other distinctions between
models can be made, depending upon the
method of analysis defined by the model and
the particular approach used to solve the
model. These classifications, among
others, include:
1. Conceptual vs. empirical models. Models
can be classified as conceptual or empiri-
cal depending upon whether or not the
actual physical processes of the waste
disposal system are considered. Empirical
models are completely based upon observa-
tion and experimentation. The distinction
between empirical and conceptual models is
not always clear. Several models describing
single-ion adsorption, for example, are
empirical in nature (e.g., linear ad-
sorption; the Freundlich isotherm), while
others are based upon physico-chemical
theory (e.g., most cation exchange
equations). When column displacement
experiments are carried out to measure the
migration rate of certain chemicals
through given soils, the approach will be
empirical in nature, even though this
experiment ion will yield parameters which
are required in conceptual models (e.g.,
dispersion coefficients or adsorption con-
stants). The partial differential equations
describing mass transport in porous media,
on the other hand, constitute a conceptual
model because the governing equations are
based upon principles of conservation of
mass, energy and momentum. These equations,
however, often use empirical relations in
their derivation (adsorption, degradation,
Darcy's law for fluid flow, etc.). Certain
writers have used the term "black box" to
indicate the empirical nature of certain
models, while the terms "white box" and
"synthetic model" have been used to des-
cribe conceptual models.
2. Stochastic vs. deterministic models.
In a deterministic model, all input
variables and system parameters are
assumed to have fixed mathematical or
logical relationships with each other. As
a consequence, these relationships com-
pletely define the system (a unique
solution is obtained). Stochastic (or
probabilistic) models, on the other hand,
can take into account the intrinsic ran-
domness associated with any of the system
parameters, or the uncertainties associa-
ted with the many mechanisms operating in
the system, the system parameters them-
selves or the input variables. Randomness
may be present because of spatial varia-
bility or medium heterogeneity. Here,
certain probability distributions (normal,
log-normal, etc.) are generally used to
characterize the statistical behavior of
the parameters in question (see for
example Freeze, 1975)- Uncertainty, on the
other hand, may be present because of
insufficient or unreliable input data, or
because of measurement errors. Uncertainty
may also result from the use of an over-
simplified model. Because of an inadequate
understanding of the basic physical and
chemical processes in the system, different
mechanisms are often lumped together, thus
leading to less well-defined parameters
(the dispersion coefficient is an excellent
example of this). A mean and variance may
then be used to describe the parameter
under uncertainty (Tang and Pinder, 1977).
The use of a stochastic model generally
leads to a confidence interval for each of
the output variables.
3. Static vs. dynamic models. This dis-
tinction depends upon how the time di-
mension is viewed in the model. Static
models are those which evaluate steady-
state conditions, i.e., where the input
and output variables do not change with
time. When the input and output variables
do change with time, dynamic (or transient)
models result. Although static models,
which are much simpler and require less
computational effort than dynamic models,
could be used to describe certain sub-
systems of the waste disposal-groundwater
system (e.g., for description of fluid
flow in and under the disposal site), it
seems that the overall system is a dynamic
one and should be modeled accordingly.
A. Spatial dimensionality of the model.
Although a waste disposal site and the
underlying groundwater reservoir constitute
a three-dimensional system, useful and
accurate information can often be obtained
with models which consider only one or two
spatial dimensions. For example, a one-
dimensional model could be used to describe
the vertical transport of contaminants in
the unsaturated zone between the landfill
and the groundwater table. While consider-
able insight into the mechanisms of con-
taminant transport may be obtained in this
way, the model stops short of providing
accurate information regarding groundwater
pollution of the underlying aquifer. This
is because the effects of leachate dilution
by flowing groundwater cannot be evaluated
193
-------
with such a one-dimensional model. An ex-
ception to this obviously occurs when the
water table lies far below the soil sur-
face and when the evapotranspirat ion equals
or exceeds the average (say yearly)
precipitation. It appears that, in general,
at least a two-dimensional orosa-aeotional
model must be formulated. Two dimensional
models may also be formulated on an areal
basis. Here the system parameters and
input and output variables represent
averaged quantities along the vertical
dimens ion.
Table 1 lists a few example models
and their classification into different
groups. When these models are used to
evaluate the physical-chemical behavior
of constituents present in a waste dis-
posal site, including an evaluation of the
potential pollution of the underlying
groundwater aquifer, the model is said to
simulate the waste disposal system. It is
important to note here that the process
of simulation not only includes con-
struction of a particular model, but also
its use for study of the system under con-
sideration. As will be shown in the next
section, considerable effort has been put
into the development of very sophisticated
groundwater quality models. Much of this
effort will be wasted if these models are
not used for the purpose for which they
were designed (either because of their
dubious practical value, their lack
of testing and field verification, or the
lack of familiarity of the prospective
user with the capabilities of the model).
CONCEPTUAL-MATHEMATICAL MODELS
Of the different types of models
discussed above, conceptual-mathematical
models appear to be the most promising but,
unfortunately, also the most complex ones
for evaluating potential groundwater
pollution from waste disposal sites. Models
in this group are generally based upon a
set of governing equations which are
derived using well-known principles of con-
servation of mass, energy and momentum.
After suitable simplifications, the
governing equations generally reduce to a
set of coupled non-linear, second-order
partial differential equations, one being
descriptive of fluid flow and the others
characterizing the transport of the
different chemical species associated with
the waste leachate. The governing
equations are then augmented by auxiliary
conditions, describing certain system para-
meters (unsaturated hydraulic conductivity,
ion-exchange equations, etc.) and the
initial and boundary conditions imposed on
the system (geometry of the landfill,
aquifer characteristics, precipitation-
evaporation data, location of wells or open
surface water bodies, etc.).
Numerous conceptual -mathematical
models are presently available, the
differences between them being mostly a
result of the number of simplifications
made during derivation of the governing
equations and their method of solution. As
an example, the following partial differ-
ential equations may be used to simulate
the (density independent) transport of
water and dissolved constituents in a
saturated-unsaturated three-d imens ional
medium (such as occurs in a landfill and
the underlying groundwater system)
Mass transport equation.
35 39c 3c
„ _ * . _ * d / K
p + ~
(a) (b)
3x.
(c)
- fe- * *! <•„,*?
i m= I
(d)
± Z2 R. ± Qcf
m=l k k
(f) (g)
Flow equation.
(e)
(D
(I.J = 1,2,3)
(2)
where c, and S. represent solution and
adsorbed concentrations, respectively, of
chemical species k (see Table 2 for a
definition of the other symbols). The
different terms in the transport equation
describe the following processes: (a)
changes in adsorbed concentration, (b)
changes in solution concentration, (c)
diffusion and dispersion effects, (d) con-
vective transport by the fluid, (e) pro-
duction (+) or decay (-) reactions, (f)
additional chemical-soil or chemical-
194
-------
Table 1; Some Example Models and Their Classification
Into Different Groups.
MODEL DEFINITION
On site inspection using
engineering judgement
The Drexel University experi-
mental field landfill
Batch equilibrium study; shaker
test; solid waste evaluation
leachate test (subsystem models
Column displacement studies; thin-
layer chromatography (subsystem
models)
Criteria listing; matrix method
One- dimensional unsaturated
transport model of Bresler (1973)
(subsystem model)
Two-dimensional saturated-unsatu-
rated transport model of Duguid
and Reeves (1976)
Model for groundwater flow and
mass transport under uncertainty
of Tang and Pinder (1977)
TYPE OF MODEL
Descriptive
Physical
Mathematical
Mathematical
Mathematical
Mathematical
Mathematical
Mathematical
Empi r ical
Emp i r i ca 1
Emp i r i ca 1
Emp i r i ca 1
Emp i r i ca 1
Conceptual
Conceptual
Conceptual
-
Deterministic
Deterministic
Deterministic
Deterministic
Deterministic
Deterministic
Stochastic
Dynamic
Dynamic
Static
Dynamic
Static
Dynamic
Dynamic
Dynamic
3D*
3D
OD
ID
OD
ID
2D
2D
*) indicates spatial dimension: 3D ~ three-dimensional, etc.
to
-------
Table 2: Explanation of Symbols Used in the Mass Transport
and Flow Equations.
Symbol Explanation
c. solution concentration of chemical species k (ML );
cf solution concentration of the fluid source or sink (ML );
specific soil moisture capacity, C = n TT— (L );
ient aV1);
pressure head (L);
2-1
D.. dispersion coefficient (L T );
K. . soil hydraulic conductivity (LT );
n porosity (L );
q. volumetric fluid velocity (LT );
Q fluid source or sink term, Q =
-------
chemical interactions (precipitation,
chemical transformations, cation exchange
reactions, volatilization, etc.), and (g)
concentration changes resulting from fluid
sources (+) or sinks (-).
Examination of equation (1) reveals
that the volumetric fluid velocity q.
is required before the transport equation
can be solved. For this it is necessary
to also solve the flow equation. This may
be done once in the case of a steady-state
flow simulation (9h/3t = 0), or repeatedly
in the case of a transient simulation.
Whatever approach is used, the volumetric
fluid velocity is obtained from Darcy's
1 aw as:
_
q .
.,
\. . „ is. ~
IJ 3X. 1 3
(3)
The transport equation is also coupled to
the flow equation through the dispersion
coefficient, D... The magnitude of D..
depends upon tn4 fluid flow velocity,IJq.,
and the soil moisture content, 0 (which
in turn is determined from the pressure
head).
When k=l in equation (1), transport
of only a single chemical constituent is
considered (e.g., chloride, a pesticide,
or a trace metal). Adsorption, if present,
can then be modeled by employing an
equation describing the dependency of the
sorbed concentration, S, on the solution
concentration, c (i.e., through the use of
an appropriate adsorption isotherm).
Several models for adsorption and ion
exchange are available for this purpose.
Table 3 lists some of the most frequently
used (single-ion) adsorption models. These
models have been divided into two categor-
ies: equilibrium models which assume
instantaneous adsorption of the chemical,
and kinetic models which consider the rate
of approach towards equilibrium. Not
listed in the table are those models which
describe competition between two ionic
species, such as the commonly used cation
exchange equations. Except for a few cases
(e.g., model 1.5 in Table 3), generally
two or more transport equations have to be
solved for such multi-ion transport prob-
lems (k=2,3,...). Most of the equilibrium
models in Table 3 represent special cases
of the kinetic models given, and follow
immediately from them by setting the time
derivatives, 3S/3t, equal to zero. All
adsorption models, furthermore, represent
reversible adsorption reactions, except
for model 2.6. This model, used by Enfield
(1971*) to describe orthophosphate
adsorption, represents an irreversible
reaction because it does not allow for any
chemical desorption (the adsorption rate is
always positive).
Once the governing equations and the
initial and boundary conditions are defined,
solutions for the concentration can be
generated by fairly straightforward, albeit
very sophisticated, and tedious mathemati-
cal manipulations. Basically two approaches
are currently available for this purpose:
analytical and numerical methods. These
two approaches are now discussed briefly.
1. Analytical methods. In order to obtain
an analytical solution of the transport
equation (Equation 1), one generally must
assume a constant fluid velocity, a
constant dispersion coefficient, constant
physical parameters and a highly simplified
geometry for the simulated system. Exact,
usually explicit expressions for the con-
centration can then be generated through
the use of integral and differential cal-
culus. Although the advantages of having
analytical solutions are numerous, (i.e.,
ease of application, and low cost of
operation once derived), the necessity
of having to make various simplifying
assumptions severely restricts the appli-
cability of analytical solutions to waste
disposal-groundwater contamination prob-
lems. In spite of these restrictions, it
appears that some of the available two-
and three-dimensional analytic solutions
(Kuo, 1976; Yeh and Tsai, 1976) could be
applied to a few well-defined geohydro-
logical systems, and hence should not be
excluded a priori from consideration. An
interesting application of this approach
is also given by Larson and Reeves (1976),
who describe a transport model which pre-
dicts the simultaneous flow of water and
trace contaminants through a two-dimension-
al, unsaturated and layered soil medium.
2. Numerical methods. While some situations
may lend themselves to analysis using
analytical methods, most field problems
have such complex physical and chemical
characteristics that the flexibility of a
numerical approach is required. When
numerical techniques are used, the partial
differential equations are generally re-
duced to a set of approximating algebraic
equations, which subsequently are solved
using methods of linear algebra. The most
commonly used numerical methods are finite
197
-------
Table 3: Partial List of Equations Used to Describe Adsorption Reactions [k. and k
are constants, k represents a rate constant (T~'), and S. and S represent
initial and finaF (or maximum) adsorbed concentrations, respectively (after
van Genuchten and Cleary, 1978)].
MODEL
1 . Equi 1 ibrium
1.1 (linear)
1.2 (Langmuir)
1.3 (Freundlich)
1.5 (Modified
Kjelland)
2. Non-equi 1 ibrium
2.1 (linear)
2.2 (Langmuir)
2.3 (Freundlich)
2.5
2.6
EQUATION
S = kjC + k2
,_ V
I + k£c
k,
S = kjC 2
-2k S
S - k c e *-
cS
m
H-- k (k.c + k - S)
ot r i i
3S _ , . V ...
3t r 1 + ICc
3S S
f|=kr(k]C2-S)
a<; k9S '2ISS
f=kre2 (k,ce 2 -S)
ss s -s
at r m 2 Sm - S,
35 . \k2
at rc s
REFERENCE
Lapidus and Amundson (1952)
Lindstrom et al (196?)
Tanji (1970)
Ballaux and Peaslee (1975)
Lindstrom and Boersma (1970)
Swanson and Dutt (1973)
Lindstrom et al (1970
van Genuchten et al (I97l»)
Lai and Jurinak (1971)
Lapidus and Amundson (1952)
Oddson et al (1970)
Hendricks (1972)
Hornsby and Davidson (1973)
van Genuchten et al (197^)
Lindstrom et al (1971)
) Fava and Eyring (1956)
Leenheer and Ahlrichs (1971)
Enfield (197*0
198
-------
differences, finite elements, and the
method of characteristics.
2.1 Finite differences. When finite
difference techniques are used, the partial
derivatives in the governing equations are
approximated by appropriate difference
equations. This method has been success-
fully used in a number of groundwater flow
problems, but its application to ground-
water quality studies is limited. This is
mainly due to the inability of this method
to accurately reproduce irregular system
boundaries, and is probably also related
to the possible introduction of numerical
dispersion, (the artificial smearing of a
concentration front) or to the presence of
undesirable oscillations in calculated
concentration distributions (especially
when dispersive transport is small compared
to convective transport). In general,
however, finite difference methods are
numerically the simplest to use and easiest
to program, while they still can yield
accurate results provided the area of in-
terest is subdivided into a sufficiently
fine grid of square or rectangular elements.
Finite difference methods have found
frequent application in the simulation of
one-dimensional unsaturated transport
problems; their application to two-dimen-
sional problems, however, is limited.
2.2 Finite element met hod s. Here, the
dependent variables(such as pressure
head and concentration) are approximated
by a finite series of basis (or shape)
functions and associated, time-dependent
coefficients. The approximating series
are then substituted into the governing
equations and the resulting errors (called
"residuals") minimized through the use of
weighted-residual theorems. In the
Galerkin method the locally defined basis
functions are the same as the weighting
functions. The approximate integral
equations derived in this way are evalua-
ted using the finite element method of
discretization, thereby minimizing computa-
tional effort. The domain of interest is
for that purpose again subdivided into
elements which, unlike for finite differ-
ences, can attain nearly any desired shape
(rectangular, triangular, even elements
with curved sides). Finite element methods
have been successfully applied to a
number of field problems involving mass
transport, although in some cases the
occurrence of numerical oscillations re-
mained a problem. Main advantages of the
finite element method are its flexibility
in describing irregular boundaries, the
ease of introducing non-homogeneous soil
properties and anisotropy, and the possi-
bility of using small elements in regions
of relatively rapid change. Although
the finite element method generally re-
quires somewhat more complex numerical
manipulations than finite differences,
its solutions are often more accurate than
finite differences if applied to the same
net.
2.3- Method of characteristics. The method
of characteristics, as generally used in
groundwater quality simulation studies,
employs a finite difference approach for
solution of the flow equation, while the
transport equation is solved by replacing
it with a set of characteristic equations.
These characteristic equations are obtained
by deleting the convective transport terms
from the governing equation and describing
them separately. One designs for this
purpose a standard finite difference net
and inserts marker particles (or moving
points) into each finite difference cell.
The marker particles are moved through the
network as prescribed by the local fluid
velocities, thereby accurately describing
the effects of the convective transport
terms. The effects of the remaining terms
in the transport equation are then super-
imposed on the updated positions of the
marker particles by making use of the con-
centrations of these moving points and an
appropriate finite difference scheme. The
method is fairly simple in concept, and has
been shown to produce acceptable results
for a wide variety of field problems. An
important drawback of this particular
method is that it is apparently not easy
to program in a general way in two or three
dimensions. The resulting code is, to
some extent, problem dependent.
2,k Other numerical methods. There exists
a variety of other numerical models which
can be applied to groundwater contamination
problems. The most primitive ones are
probably those using a lumped parameter
approach (i.e., models which do not con-
sider any spatial variability in the
dependent variables). In this approach, a
mass balance equation is generally applied
to the entire system, the different input
and output variables being only a function
of time (hornsby, 1973; Gelhar and Wilson,
\3Jk; Mercado, 1976). An approach incor-
porating more of the characteristics of a
distributive parameter scheme may be ob-
tained by directly applying the mass
balance calculations to a number of well-
199
-------
defined spatial cells (elements or layers).
Instantaneous mixing is often assumed
within each cell, and the values of the
dependent variables are identified with
nodes located in the center of each element
(Orlob and Woods, 1967; Tanji, 1970).
A rigorous analysis of this approach shows
that in this way an explicit in time finite
difference approximation of the governing
transport equation is obtained, provided
all significant transport mechanisms are
included in the mass balance equations. A
very similar approach was also followed by
Elzy et al (197^), who applied a "vertical-
horizontal routing model" to the transport
of hazardous waste products from a waste
disposal site. A much more elaborate, but
still somewhat similar model is the
"polygonal finite difference model" of
Hassan (\STt). The two-dimensional elements
here take the form of a polygonal network.
Additional refinement of this method will
eventually lead to an "integrated finite
difference" approximation of the governing
transport equations.
Each of the numerical techniques dis-
cussed above has a number of advantages and
disadvantages associated with its use.
These may be subdivided into factors
affecting the accuracy, efficiency and
accessibility of the particular method.
While important differences in accuracy
and efficiency between finite differences
and finite element methods are known to
exist (Pinder and Gray, 1977), it is still
not quite clear to what degree these
differences are important in the simulation
of large field problems. In fact, it may
very well be that the accuracy of the
numerical technique becomes secondary when
viewed against the uncertainty in the many
input parameters needed in most models.
Furthermore, the efficiency in programming,
the general set-up of the model and its
accessibility to others are as much
factors determining the usefulness of the
resulting program, as is the choice of the
numerical technique.
SURVEY OF EXISTING MODELS
In this section, a compilation is
given of the different types of conceptual-
mathematical models currently available for
possible use in groundwater contamination
problems. The list of models, given in
Table A, is not intended to be complete;
numerous other models are likely to exist,
either published, unpublished or under
development. The main purpose of Table k is
to demonstrate the existence of a wide
variety of models, to characterize their
most important capabilities and limitations,
to identify their method of solution, and
to show some of their applications. The
models are differentiated into four dis-
tinct groups: saturated-unsaturated trans-
port models, saturated-only transport
models, unsaturated-only transport models,
and analytical transport models. Each
group will be discussed briefly.
A. Saturated-unsaturated transport models.
Models in this group are based directly upon
equations (1) and (2), or upon appropriate
simplifications of these equations. The
different models simulate either a three-
dimensional system (Model A4), or two-
dimensional cross-sections. Except for
Model A6, no cation exchange (multi-ion
transport) is considered in any of the
models in this group, although several do
include (single-ion) adsorption and decay.
Saturated-unsaturated transport
models are probably the most appropriate
ones for landfill simulation because
they can consider the unsaturated conditions
in and under the waste disposal site. For
example, aerobic decomposition of hazardous
organic wastes (including pesticides)
and certain oxidation/reduction reactions
could be taken into account in these models.
Also one of the most important "attenuation"
mechanisms, dilution of leachate by
flowing groundwater, can be much more
clearly defined with saturated-unsaturated
transport models'.
Unfortunately, inclusion of the un-
saturated zone also places a considerable
burden on the effective and economical use
of these models. The highly non-linear
nature of the governing saturated-unsatura-
ted flow equation makes its numerical
solution difficult, and generally small
time steps in the algorithm are necessary
to ensure a correct solution. This can
easily lead to extremely high computer
costs when simulations are to be made over
a period of several years (Segol, 1975;
Duguid and Reeves, 197&)- It appears that
several simplifications are possible to
circumvent at least some of these problems.
For example, by using monthly averaged
rain-evaporation data (Duguid and Reeves,
1977; van Genuchten et al (1977) rather
than hourly or daily data, or by assuming
only steady-state flow conditions (Sykes,
1975), one could speed up the calculations
200
-------
considerably. While steady-state flow con-
ditions may be justified in a number of
cases, it seems that predictions of the
amount and quality of leachate interacting
with the groundwater under the waste dis-
posal site could be severely underestimated
when the average evapotranspiration equals
or exceeds the annual precipitation. Ob-
viously, steady-state flow models are also
unable to describe seasonal changes in the
groundwater table.
An important problem associated with
the use of saturated-unsaturated transport
models is the need for additional input
data. For example, the non-linear relation-
ships between the hydraulic conductivity,
the moisture content and the pressure head
must be determined for each soil type
present in the system. Expensive and time-
consuming experimentation is generally
required to obtain these functional rela-
tionships. In addition, and of equal
importance, the difference soil-chemical
interactions occurring in the unsaturated
zone have to be quantified. Many of these
interactions are still not fully understood
and considerably more research in this
area seems necessary before the full
potential of these models can be utilized.
A more detailed discussion of this problem
is given later.
B. Saturated-only transport models. Models
in this group generally ignore the
dynamics of the unsaturated zone between
the waste disposal site and the ground-
water table. Hence, important mechanisms
associated with unsaturated flow and con-
taminant transport cannot be considered,
unless they are represented in an approxi-
mate way through data adjustments. To use
these models, it is necessary to have a
method of quantifying the amount and
quality of leachate reaching the ground-
water table. Such quantification is
necessary for an accurate evaluation of the
important dilution (mixing) processes
operating at and immediately below the
groundwater table. Given that this can be
done beforehand, i.e., in a predictive
way, saturated-only transport models appear
to be useful tools for evaluating possible
groundwater contamination. The need for
describing the unsaturated zone, obviously,
becomes much less when the waste disposal
site is in direct contact with the
saturated zone.
Saturated-only transport models have
found application to a wide variety of
practical field problems, often to cases
where groundwater contamination was ob-
served over a period of time and hence
where calibration of the model to actual
field data was possible. Unfortunately,
most applications have been restricted to
non-interactive (no adsorption) and single-
ion transport. It seems that considerably
more work is necessary to determine the
accuracy of these models when applied to
the transport of adsorbing chemicals.
A special class of saturated-only
transport models is provided by saltwater
intrusion models (Models B19-B22). These
models differ from the other (cross-
sectional) models in this group in that
they consider density dependent flow. As
such they are also directly applicable to
contaminant transport from waste disposal
sites.
C. Unsaturated-only transport models.
Because these models consider only the
unsaturated zone, they cannot be used to
describe contaminant migration in ground-
water systems, However, the different
(mostly one-dimensional) models in this
group are very useful tools for studying
the mechanisms of contaminant transport
in the unsaturated zone, especially the
transient waste-soil interactions
associated with column leaching studies.
Another and very important application of
these models may result when they are used
simultaneously with saturated-only trans-
port models. The unsaturated models can
then be used to predict in an approximate
way the amount and type of leachate
reaching the groundwater table, information
which subsequently can be used as input
for the saturated-only transport models.
Models A2 and A6 represent examples of this
type of approach.
D. Analytical models. As discussed earlier,
analytical models appear to have only
limited application to actual (field)
groundwater contamination problems. Their
application is restricted to those cases
wherein the geohydrology of the area is
very simple (flow in one direction, con-
stant porosity, dispersivity and conductiv-
ity). The different one-dimensional
analytical solutions (Models D2) are again
potentially useful tools for identification
and quantification of waste-soil inter-
actions when used in conjunction with
column leaching studies (quantification of
adsorption and decay constants, dispersion
coefficients, etc.).
201
-------
Table k: Partial List of Published Transport Models for Application
to Groundwater Quality Problems.
model references
geometry
of
model
1)
A. SATURATED-UNSATURATED TRANSPORT
Al. Elzy, et al. (1974)
A2. Perez, et al. (197*0
A3. Sykes (1975)
A*». Segol (1976, 1977)
2D.C
20, C
2D.C
3D.2D
A5. Duguid and Reeves (1976,77) 2D,C
A6. Shaffer, et al. (1977)
A7. van Genuchten, et al.
(1977), van Genuchten and
Pinder (1978)
2D.C
2D.C
math.
sol'n.
method
2)
MODELS.
0
FD
HFE
HFE
LFE
FD
HFE
B. SATURATED-ONLY TRANSPORT MODELS.
Bl. ReddeH and Sunada (1970)
B2. Fried and Ungemach (1971),
Fried, et al. (1970
B3. Bredehoeft and Pinder
(1973)
Bk. Robertson and Barraclough
(1973), Robertson (197*0
B5. Pinder (1973)
2D.A.C
2D.A
2D,A
2D.A
2D,A
HOC
FD
MOC
HOC
HFE
type
of
flow
3)
Tr
Tr
St
Tr
Tr
Tr
Tr
Tr
Tr
Tr
Tr
Tr
type
of
soi 1
M
-
L
L.A
L,A
L,A
L
L,A
A
-
A
A
A
model
chemi-
stry
5)
A,D
1
1
1
A,D
A,C,D
A.D
1
1
1
A,D
1
appl i cat ions/comments
Simple vertical-horizontal routing model for conta-
minant transport from a landfill.
Coupled ID vertical, unsaturated model and 2D
saturated model; describes groundwater pollutior
from agricultural sources.
Contaminant transport from a landfill.
Theoretical formulation only; one-dimensional veri-
fication.
Transport of radionucl ides from a seepage trench;
studied effects of different rainfall boundary
conditions on tritium transport.
Irrigation return flow model; coupled ID, vertical
unsaturated mode), and 2D, saturated mode); un-
saturated model is adapted version of model C3.
Theoretical formulation only; applied to contami-
nant transport from a hypothetical landfill.
Three-dimensional formulation; two-dimensional
appl ication only.
Flow part based on Boussinesq equation; describes
NaCl pollution from large salt dumps into
alluvial aquifer in Northeastern France.
Movement of salt water in lime stone aquifer; pre-
dicted future concentrations and tested effects
of protective pumping.
Model includes effects of linear equilibrium ad-
sorption and linear decay; describes transport
of industrial and low-level radioactive wastes
into Snake River Plain aquifer, Idaho.
Used calibrated model to predict future concentra-
tions of hexavalent chromium seeping from waste
-------
NJ
O
U
B6. Konikow and Bredehoeft .
(1973, 1971*)
\ *•//•* 1 'SI • f
87. Hughes and Robson (1973),
Robson (1971*)
B8. Ahl strom and Baca (1974)
B9. Gupta, et al. (1975)
BIO- Martinez, et al. (1975),
Thorns, et al. (1977)
Bll. Schwartz (1975, 1977)
BIZ. Helweg and Labadie (1976)
B13- Pickens and Lennox (1976)
Bill. Besbes, et al. (1976)
B15- Less! (1976)
B16. Cabrera and Marino (1976)
B17 Konikow (1977)
1) 10 = one-dimensional
2D.A
2D.A
2D,A
3D
2D.A
2D,C
2D,A
2D.C
2/3D,A
2D.C
2D.C
2D.A
HOC
MOC
HOC
HFE
HFE
MOC
MOC
TFE
A/FD
HFE
TFE
MOC
Tr
Tr
St/Tr
Tr
St
St
Tr
St
Tr
Tr
Tr
Tr
A
A
A
L.A
-
A
A
L,A
L
L
L.A
A
1
1
A,C
1
A,D
A.C.D
1
A
1
1
1
1
disposal ponds into underlying aquifer.
Irrigation return flow model; calibrated model was
used to evaluate the effects of different irri-
gation practices on salinity changes in a
stream-aquifer system in southeastern Colorado.
Pollution of a shallow aquifer by seepage from
industrial and municipal wastes; predicted fu-
ture concentrations and tested effects of alter-
native groundwater management practices.
Model considers adsorption and exchange of several
ions.
Simulated rising connate waters through a vertical
fault in a mul t i -layered aquifer system (steady-
state application only).
Simulated groundwater pollution from salt dome
leachates.
Hypothetical study of subsurface pollution by
radioactive wastes (1975); model analysis of a
proposed waste disposal site (1977)-
Adapted version of model B6; used as a cost-
effective salinity management model for stream-
aquifer systems.
Contaminant transport from a hypothetical landfill.
Approximate areal model for multi-layered aquifer;
tested effects of dam construction on salinity
changes in the Kairouan plain, Tunes ia.
Applied to contaminant transport in a heterogeneous
aquifer system.
Theoretical study of contaminant transport in a
stream-aquifer system.
Simulated 30-year history of groundwater pollution .
by chloride from un lined waste disposal ponds.
2) A = analytical A) L - layered medium
2D = two-dimensional FD = finite differences A = anisotropic medium
30 = three-dimensional LFE = linear finite elements .
A = area! (2D) model TFE = triangular finite elements 5) A = adsorption (sing e-ion)
C = cross-sectional (20) HFE = higher order/mixed finite C = cation exchange (multi-ion
elements transport)
1) Tr = transient flow model HOC = method of characteristics D = decay reactions
St = steady-state flow 0 = other methods. ' = inert med.um; no mteract.ons
-------
N)
2
mode} references
B18. Gureghian and Cleary
(1977)
B19. Green and Cox (1966)
B20. Pinder and Cooper (1970)
B21. Lee and Cheng (197*0
B22. Segol, et al. (1975)
geometry
of
model
1)
3D
20, C
2D.C
20, C
2D,C
math.
sol 'n.
net hod
2)
LFE
HOC
MOC
LFE
HFE
C. UNSATURATED-ONLY TRANSPORT MODELS.
Cl. Tanji, et al. (1967a,b),
Tanji, et al. (1972)
C2. Warrick, et al. (1971)
C3- Dutt, et al. (1972)
Cl». Bresler (1973)
C5. King and Hanks (1973,
1975)
C6. Kirda, et al. (1973)
C7. Shah, et al. (1975)
C8. Wood and Davidson (1975)
Davidson, et al. (1975)
C9- Bresler (1975)
ID
ID
ID
ID
ID
ID
ID
ID
2D.C
FD
FD/A
FD
FD
FD
FD
FD
FD
FD
type
of
flow
3)
St
St
St
St
St
St
Tr
Tr
Tr
Tr
Tr
St
Tr
Tr
type
of
so! 1
*0
L.A
-
-
-
A
L
-
L
-
-
-
L
-
-
model
chemi-
stry
5)
A,D
1
1
1
1
A,C
1
A,C,D
1
A.C.D
1
A
A
1
appl icat ions /comments
Contaminant transport from an existing landfill on
Long Island, New York.
Saltwater intrusion model.
Saltwater intrusion model.
Saltwater intrusion model.
Saltwater intrusion model.
Approximate finite difference solution for cation
exchange in field soils.
Approximate analytical solution of transport equa-
tion; applied to field infiltration study using
chloride as tracer.
Model considers nitrogen transformations, nitrogen
uptake by crops, ion exchange and certain chemi-
cal precipitation reactions; applied to cation
movement in field soils.
Describes chloride transport during infiltration
and redistribution; compared results with field
data.
Model includes water uptake by plant roots; applied
to irrigation return flow quality studies.
Applied to anion movement in laboratory soil
col umns.
Model assumes constant dispersion coefficient and
kinetic adsorption; applied to phosphorus
transport in soi I s.
Applied to pesticide transport in laboratory soil
columns; model assumes chemical hysteresis in
the adsorpt ion-desorpt ion isotherms.
Describes two-dimensional transport of solutes
under a trickle source.
-------
CIO. Ungs, et al. (1976)
Cll. Hildebrand (1975),
Hildebrand and Himmelbau
(1977)
C12. Selim, et al. (1976)
C13- Hillel, et al. (1976)
C14. Jury, et al. (1977)
CIS- Gureghian, et al. (1977)
C16. Gaudet, et al. (1977)
C17- Selim, et al. (1977)
C18. Wierenga (1977)
C19- van Genuchten (1978)
ID
ID
ID
ID
ID
ID
ID
ID
lu
ID
D. ANALYTICAL TRANSPORT MODELS.
Dl. Cleary, et al. (1973); Kuo
(1976); Yeh and Tsa i (1976);
Wang, et al . (1977), Cleary,
(1976); among others.
D2. Lapidus and Amundson (1952);
Brenner (1962); Lindstrom,
et al. (1967); Lindstrom and
Boersma (1971, 1973); Cleary1
2,30
ID
FD
FD
FD
FD
FD
FD
FD
FD
FD
HFE
A
A
Tr
Tr
Tr
Tr
Tr
Tr
St
St
Tr
Tr
St
St
and Adrian (1973); Marino (197^); Lindstrom and
Stone (197^3, 197/*b); van Genuchten and Wierenga
(1976); Selim and Mansel 1 (1976); among others.
.
-
-
-
-
L
•
L
-
L
-
-
A
1
A
1
1
A,C,D
1
A
1
A.D
(A,D)
(A,D)
Model considers water uptake by plant roots;
describes chloride transport during infiltra-
tion and redistribution.
Simulated nitrate movement through laboratory
sand columns.
Applied to transport of 2,^-D during infiltration
and redistribution.
Model considers water uptake by plant root system-,
program written in S/360 CSMP language.
Model assumes unit gradient for pressure head in
flow equation; considers water uptake by plants.
Model considers multi-ion transport (ammonium,
nitrite, nitrate) and non-linear adsorption.
Model considers solute exchange between mobile and
stagnant water; applied to chloride transport
in laboratory sand columns.
Applied to transport of chloride and 2,^-D in a
two- layered soil column.
Compares results obtained with transient and
steady-state flow equations.
Describes contaminant transport during infiltration
and redistribution
Various assumptions and applications.
Various assumptions and applications, including:
- Zero- and first-order decay;
- linear equilibrium adsorption;
- first-order kinetic adsorption;
- solute exchange between mobile and
stagnant water;
- decaying boundary conditions.
-------
DISCUSSION
The increased use of land for the dis-
posal of a wide variety of wastes makes it
necessary to have a realistic method for
analyzing the effects of such disposal on
the quality of underlying groundwater
systems. Computer simulation models of the
type discussed in the previous section may
greatly aid in such an analysis, provided
they are used with a clear understanding
of their capabilities and limitations. The
following discussion attempts to define
some of the advantages and limitations
associated with the use of a simulation
model as a site selection tool for waste
disposal systems. No attempt is made to
list all possible advantages and disadvan-
tages of a water quality (or any other)
simulation model; for this the reader is
referred to several recently published
review papers (Coats, 19&9; Weber and
Hassan, 1972; Evenson et a!., 197^1
Holcomb, 1976; Grimsrud et al., 1976).
Many of the conclusions reached in these
studies are equally well applicable to the
present discussion, and hence will not be
restated here.
Perhaps the most important advantage
of the use of a simulation model for site
evaluation is that a quantitative pre-
diction of the potential pollution of
nearby groundwater aquifers can be made.
For example, the shape of the contaminant
plume, if present, can be described by a
simulation model, while types and levels
of certain contaminants at various points
in the system and at different time levels
can be predicted. Such predictions can be
made before actual construction of the
waste disposal site, and hence would pro-
vide decision makers with an advance
picture of the magnitude and extent of the
expected leachate migration from the site.
Of course, when such evaluations are made
before actual development of the site, no
calibration of the model to observed field
data will be possible, and hence a high
degree of uncertainty will be inherent in
the simulation results. This, in itself,
should not limit the usefulness of a simu-
lation model. By making several trial runs,
using system parameters covering a
reasonable range of uncertainty, one can
still delineate the maximum and minimum
extent of future groundwater contamination.
In this way a model (or sensitivity)
analysis of the proposed waste disposal
system is obtained, which may contribute
greatly to an understanding of the future
behavior of the system. An excellent
example of such an approach is given by
Schwartz (1977), who applied a model
analysis to a proposed waste disposal site
for low-level radioactive wastes. The use
of a simulation model for such an analysis
demonstrates that they can be used as tools
for site selection, but that they can never
replace sound engineering judgement. Rather
they can help the engineer in extracting
the maximum amount of information from the
available data.
Simulation models may find useful
application in the effective design and
management of waste disposal sites. Models
can be used to match different types of
wastes and disposal sites in a manner that
would minimize groundwater pollution at
each site. Given a certain waste, different
candidate sites with different soil proper-
ties could be ranked with respect to their
pollution potential. Simulation models,
furthermore, can be valuable tools for
determining key parameters that control
the migration of leachate from a site. The
presence of certain liners or impervious
covers, the type and thickness of reactive
earth materials beneath the waste disposal
site, the thickness of the unsaturated
zone, the local hydrology of the area, etc.,
are all factors determining the pollution
potential of a given site. The effects of
these factors on leachate migration are
most easily assessed using appropriate
(one- or multi-dimensional) simulation
models. In addition, simulation models can
be used to study the optimal design and
location of contaminant retrieval systems.
Once the waste disposal site is constructed
and it is clear that unacceptable pollution
is present, models could be used to deter-
mine the optimum location of wells for
possible recovery of contaminants (pro-
tective pumping).
A very important application of a
simulation model, not directly associated
with the problem of site selection itself,
is in the study of existing landfills (the
model becomes now a research tool). Because
many of the basic soil-physical and
chemical processes operating in a waste
disposal system are not very well under-
stood, simulation of existing landfills
with given waste-soil combinations may lead
to a better understanding of these pro-
cesses. As mentioned earlier, the one-
dimensional transport models are particu-
206
-------
larly useful tools for studying the migra-
tion of contaminants in and under waste
disposal sites. Such applications will un-
doubtedly increase our knowledge of con-
taminant transport (especially of adsorbing
chemicals), and could eventually lead to
the formulation of new theories (e.g.,
regarding certain adsorption mechanisms,
chemical chain reactions, channeling pro-
cesses, etc.)
Unfortunately, a number of limita-
tions associated with the use of computer
simulation techniques for landfill evalu-
ation purposes also exist. Several of these
limitations can be traced back to a general
lack of reliable input data. A successful
simulation is highly dependent upon an
accurate description of the various system
parameters and input variables. In par-
ticular, the two- and three-dimensional
transport models, probably the most
appropriate ones for landfill simulation,
require considerable input data (unsatura-
ted hydraulic conductivities, soil moisture
retention curves, dispersion coefficients,
adsorption and/or exchange coefficients,
decay constants, etc.) It is clear that
the need for such input data will greatly
limit the effective and economical use
of these models. Interestingly, simulation
models may also find here immediate appli-
cation. Because errors in different input
data will not have the same effect on the
calculated results, one can apply a sen-
sitivity (or model) analysis to the pro-
posed waste disposal site to determine
those parameters which have the greatest
effect. More effort can then be spent to
accurately determine these parameters,
while needless work and cost will be
avoided by estimating the remaining ones
only in a very approximate manner. As
suggested earlier by Coats (19&9), interest
in the accuracy of the input data should
be proportional to the sensitivity of
computed results to variations in those
data. If, for example, wide variations in
the unsaturated hydraulic conductivity
curve lead to only minor changes in cal-
culated leachate migration rates, then an
accurate determination of this curve
seems unnecessary (actually, there exist
some indications in the literature that
this indeed may be the case, provided
reliable estimates of the prevailing pre-
cipitation-evaporation rates at the soil
surface can be obtained; see for example,
Jury et al., 1977, or Wierenga, 1977).
Another problem limiting the
immediate application of simulation models
is associated with a lack of understanding
of the many chemical and biological pro-
cesses operative in a typical waste dis-
posal site. These processes include
adsorption-desorption, cation exchange,
precipitation-solution, chemical chain
reactions, oxidation-reduction and bio-
degradation. Because many of these pro-
cesses are still under investigation, their
quantification into reliable mathematical
expressions for use in simulation models
remains doubtful. For instance, it is
known that there are considerable variations
in quality and quantity of leachate as a
function of time, probably due to an inter-
play between such variables as precipita-
tion-evaporation, type and age of the
waste, pH and temperature. Reliable
estimates of leachate generation cannot be
obtained before these interrelationships
have been studied in detail and certain
quantitative expressions have been estab-
lished.
Many of the physical processes
occurring at the waste disposal site and in
the receiving groundwater aquifer are also
simplified in order to complete the re-
quired simulation. For example, hetero-
geneities of the site and the underlying
aquifer (including such processes as
channeling of leachate through the waste,
physical non-equilibrium due to immobile
water and dead-
diffusion and adsorption, fractured flow
in the aquifer, etc.) are often only
included in a very approximate manner,
mostly through adjustments in the value of
the dispersion coefficient. Much research,
also in this area, remains necessary.
CONCLUSIONS
From the discussions in this paper, it
is clear that as yet no model exists that
can simulate all of the physical, chemical,
and biological processes that are operative
in a typical waste disposal system. The
question, however, arises whether or not
such a model should be developed. The com-
plexity of these processes, which operate
in a simultaneous and interactive manner,
are probably such that the program would be
impractical. Assuming for the moment that
the knowledge for such a general model is
available (which it is not), and that the
vast amount of input data for use in the
model can be obtained (which seems unlikely)
207
-------
the resulting program would undoubtedly
become so large and bulky that the cost of
operating it would be immense.
The usefulness of a simulation model
is a direct consequence of the type of
questions being asked (i.e., the model
should be cormens'ura.te with these ques-
tions). Many situations may lend themselves
to analysis without an all-encompassing
model. For example, several currently
available models already possess the capa-
bility to describe the migration of a con-
taminant plume (or of IDS, chloride, BOO,
etc.), and hence could find immediate
application to waste disposal problems.
Considerably more study is necessary for
the more complex cases, such as the migra-
tion of certain trace metals or organic
chemicals (both for single- and multi-ion
transport). The expertise in this area
is available, but it needs to be integra-
ted into a few, relatively accurate, yet
simple conceptual-mathematical models.
This will require the cooperation of
scientists from widely different fields
(soil physicists, soil-chemists, civil
engineers, sanitary engineers, geologists,
mathematicians and computer scientists).
A close cooperation between the computer
analyst and the more experimentally
inclined investigator is especially
needed. A computer analyst, for example,
may construct a very sophisticated, multi-
dimensional, multi-ion transport model,
only to discover that it does not work in
the field or even fails to describe a
single-ion column displacement experiment
involving an adsorbing chemical. An ex-
perimentalist, on the other hand, may
perform numerous time-consuming and ex-
pensive column leaching studies, only to
discover that some simple model could have
predicted many of his observations in the
first place.
The list of simulation models pre-
sented in this paper shows that tremendous
progress in simulation technology has been
achieved in a time span of only a few
years. Much research obviously remains
necessary. This will likely result in a
better understanding of the different
mechanisms operative in waste disposal
systems, and hence will lead to the con-
struction of more efficient and reliable
simulation models.
ACKNOWLEDGEMENTS
Part of this paper was used as con-
tributary material for "Pollution Pre-
diction Techniques for Waste Disposal
Siting - A State-of-the Art Assessment",
prepared by Roy F. Weston (EPA contract
No. 68-01-4369). Th's work was supported
in part by funds obtained from the Solid
and Hazardous Waste Research Division,
U. S. Environmental Protection Agency,
Municipal Environmental Research Labora-
tory, Cincinnati, Ohio (EPA grant No.
R803827-01).
REFERENCES
1. Ahlstrom, S. W. and R. G. Baca, 1974.
Transport model user's manual. Batelle
Pacific Northwest Laboratories, Report
BNWL-1716, UC-70.
2. Ballaux, J. C. and D. E. Peas lee, 1975.
Relationship between sorption and de-
sorption of phosphorus by soils. Soil
Sci. Soc. Amer.Proc. 39:275-278.
3. Besbes, N., E. Ledoux and G. deMarsily,
1976. Modeling of the salt transport
in multilayered aquifers. In System
Simulation in Water Resources, G. C.
van Steenkiste (ed.), North-Holland
Publ. Co., Amsterdam, pp. 229-245.
4. Bredehoeft, J. D. and G. F. Pinder,
1973- Mass transport in flowing ground-
water. Water Resour. Res., 9:194-210.
5. Brenner, H., 1962. The diffusion
model of longitudinal mixing in beds
of finite length, Numerical values.
Chem. Eng. Sci. 17:229-243-
6. Bresler, E., 1973- Simultaneous trans-
port of solutes and water under tran-
sient unsaturated flow conditions.
Water Resour. Res. 9:975-986.
7. Bresler, E., 1975- Two-dimensional
transport of solutes during nonsteady
infiltration from a trickle source.
Soil Sci. Soc. Amer. Proc. 39:604-613.
8. Cabrera, G. and M. A. Marino, 1976.
A finite element model of contaminant
movement in groundwater. Water Resour
Bull. 12:317-335-
9. Cleary, R. W., 1976. Unsteady-state,
multi-dimensional analytical modeling
208
-------
of water quality. j_n Proceeding of the
Conference on Environmental modeling
and simulation, Wayne Ott (ed.)» EPA
600/9-76-016, pp. 434-438.
10. Cleary, R. W. and D. D. Adrian, 1973-
Analytical solution of the convective-
dispersion equation for cation ad-
sorption in soils. Soil Sci.Soc.Amer.
Proc. 37: 197-199-
11. Cleary, R. W., T. J. McAvoy and
W. L. Short, 1973- Unsteady state
three-dimensional model of thermal
pollution in rivers. Water-1972, Amer.
Inst. Chem.Eng.Symp. Series 129, Vol.
69:422-431.
12. Coats, K. H., 1969. Use and misuse of
reservoir simulation models. Soc.Pet.
Tech.: 1391-1398.
13. Davidson, J. M., G. H. Brusewitz,
D. R. Baker and A. L. Wood, 1975- Use
of soil parameters for describing
pesticide movement through soils.
EPA-660/2-75-009. Office of Research
and Development, U.S. Environmental
Protection Agency, Corvallis, Oregon.
14. Duguid, J. 0. and M. Reeves, 1976.
Material transport through porous
media: a finite - element Galerkin
model. Oak Ridge National Laboratory,
ORNL-4928.
15. Duguid, J. 0. and M. Reeves, 1977. A
comparison of mass transport using
average and transient rainfall
boundary conditions. In Finite
Elements in Water Resources, W. G.
Gray, et al (eds.), Pentech Press,
London, pp. 2.25-2.35.
16. Dutt, G. R., M. J. Shaffer and
W. J. Moore, 1972. Computer simulation
model of dynamic bio-physicochemical
'processes in soils. Techn. Bull. No.
196, Agr. Exp. Sta., University of
Arizona, Tucson, Arizona.
17. Elzy, E., T. Lindstrom, L. Boersma,
R. Sweet and P. Wicks, 1974. Analysis
of the movement of hazardous waste
chemicals in and from a landfill site
via a simple vertical-horizontal
routing model. Agr. Exp. Sta.,
Special Report 414, Oregon State
University, Corvallis, Oregon.
18. Emshoff, J. R. and R. L. Sisson, 1970.
Design and use of computer simulation
models. The MacMillan Company, New
York.
19- Enfield, C. G., 1974. Rate of phos-
phorus sorption by five Oklahoma
soils. Soil Sci.Soc.Amer.Proc. 38:
404-407.
20. Evanson, D. E., G. T. Orlob, and
T. C. Lyons, 1974. Ground-water
quality models: what they can and
cannot do. Ground water 12:97-101.
21. Fava, A. and H. Eyring, 1956. Equili-
brium and kinetics of detergent ad-
sorption - a generalized equilibrium
theory, J. Phys. Chem. 65:890-898.
22. Fishman, G. S., 1973. Concepts and
methods in descrete event digital
simulation. John Wiley & Sons, New
York.
23. Freeze, R. A., 1975- A stochastic-
conceptual analysis of one-dimensional
groundwater flow in nonuniform homo-
geneous media. Water Resour. Res.
11:725-741.
24. Fried, J. J. and P. 0. Ungemach, 1971-
A dispersion model for a quantitative
study of a groundwater pollution by
salt. Water Res. 5:491-495-
25. Fried, J. J., J. L. Gamier and
P. 0. Ungemach, 1971. Etude
quantitative d'une pollution de nappe
d'eau souterraine: la salure de la
nappe phreatique dans le department
du Haut-Rhin. Bull. Bu. Rech. Geol.
Min., Sect III, 1:105-115.
26. Fungaroli, A. A. and R. L. Steiner,
1973- Investigation of sanitary land-
fill behavior. EPA-R2/2-7- Office of
Research and Development, U.S.
Environmental Protection Agency,
Cincinnat i, Ohio.
27. Gaudet, J. P., H. Jegat, G. Vachaud
and P. J. Wierenga, 1977. Solute
transfer, with exchange between
mobile and stagnant water, through
unsaturated sand. Soil Sci. Soc.Amer.
J. 41:665-671.
209
-------
28. Gelhar, L. W. and J. L. Wilson, 1974.
Ground-water quality modeling. Ground-
water, 12:399-408.
29. Grimsrud, G. P., E. J. Finnemore,
and H. J. Owen, 1976. Evaluation of
water quality models: a management
guide for planners. EPA-600/5-76-004.
Office of Research and Development,
U.S. Environmental Protection Agency,
Washington, D.C.
30. Green, D. W. and R. L. Cox, 1966.
Storage of fresh water in underground
reservoirs containing saline water.
Phase 1. Completion Report No. 3,
Kansas Water Resour. Res. Inst.,
Manhattan, Kansas.
31. Gupta, S. K., K. K. Tanji, and
J. L. Luthin, 1975. A three-dimen-
sional finite element groundwater
model. Water Resour. Center, Contr.
Series No. 152, University of
California, Davis.
32. Gureghian, A. B. and R. W. deary,
1977- Three-dimensional modelling
of pollutant transport. Final report
208 project, Nassau-Suffolk County
Regional Planning Board, Hauppauge,
New York.
33. Gureghian, A. B., R. W. Cleary and
D. S. Ward, 1977- One-dimensional
modeling of unsaturated pollutant
transport, Final Report 208 project.
Nassau-Suffolk County Regional
Planning Board, Hauppauge, New York.
31*. Hassan, A. A., 1974. Mathematical
modeling of water quality for water
resources management. Volume 1.
Development of the groundwater
quality model. District report,
Department of Water Resources,
Southern District, State of Calif-
ornia.
35. Helweg, 0. J. and J. W. Labadie, 1976.
A salinity management strategy for
stream-aquifer systems, Hydrology paper
paper No. 84, Colorado State Univer-
sity, Fort Collins, Colorado.
36. Hendricks, D. W., 1972. Sorption in
flow through porous media. _|_n_
Fundamentals of transport phenomena
in porous media. Developments in Soil
Science 2, IAHR. Elsevier Publ. Co.,
Amsterdam, pp. 384-392.
37. Hildebrand, M. A., 1975- The trans-
port of nitrate during unsteady water
flow through unsaturated sand media.
Ph.D. thesis. University of Texas,
Austin, Texas.
38. Hildebrand, M. A. and D. M. Himmelbau,
1977- Transport of nitrate ion in
unsteady, unsaturated flow in porous
media. A.I. Ch.E.J. 23=326-335.
39- Hi 1 lei, D., H. Talpaz and H. van
Keulen, 1976. A macroscopic-scale
model of water uptake by a nonuniform
root system and of water and salt
movement in the soil profile. Soil
Sci. 121:242-255.
40. Holcomb, 1976. Environmental modelling
and decision making. The Holcomb
Research Institute for the Scientific
Committee on Problems of the Environ-
ment. Praeger Publ., Inc., New York.
41. Hornsby, A. G., 1973. Predicting
modeling for salinity control in irri-
gation return flows. EPA-R2-73-168.
U. S. Environmental Protection Agency.
42. Hornsby, A. G. and J, M. Davidson,
1973- Solution and adsorbed fluometuron
concentration distribution in a
water-saturated soil: experimental
and predicted evaluation. Soil Sci.
Soc. Amer. Proc. 37=823-828.
43. Hughes, J. L. and S. G. Robson, 1973.
Effects of waste percolation on
groundwater in alluvium near Barstow,
California. ^Underground Waste
Management and Artificial Recharge,
J. Braunstein (ed.), 1:91-129,
American Association of Petroleum
Geologists.
44. Jury, W. A., H. Fluhler and L. H.
Stolzy, 1977- Influence of soil
properties, leaching fraction, and
plant water uptake on solute con-
centration distribution. Water Resour.
Res. 13=645-650.
45. King, L. G. and R. J. Hanks, 1973.
Irrigation management for the control
of quality of irrigation return flow,
EPA-R2-73-265. Office of Research and
Development, U. S. Environmental
Protection Agency, Washington, D.C.
210
-------
46. King, L. G. and R. J. Hanks, 1975-
Management practices affecting quality
and quantity of irrigation return
flow. EPA-660/2-75-005- Office of
Research and Development, U. S.
Environmental Protection Agency,
Corval1 is, Oregon.
47. Kirda, C., D. R. Nielsen and J. W.
Biggar, 1973- Simultaneous transport
of chloride and water during infil-
tration. Soil Sci. Soc. Amer. Proc.
37:339-345.
48. Kohikow, L. F., 1977- Modeling
chloride movement in the alluvial
aquifer at Rocky Mountain Arsenal,
Colorado, Geological Survey Water-
Supply Paper 2044, U. S. Government
Printing Office, Washington, D.C.
49. Konikow, L. F. and J. D. Bredehoeft,
1973- Simulation of hydrologic and
chemical-quality variations in an
irrigated stream-aquifer-system - a
preliminary report. Colorado Water
Resources Circular No. 17, Colorado
Water Conservation Board, Denver,
Colorado.
50. Konikow, L. F. and J. D. Bredehoeft,
1974. Modeling flow and chen>ical
quality changes in an irrigated
stream-aquifer system. Water Resour.
Res. 10: 546-562.
51. Kuo, E.Y.T, 1976. Analytical solution
for 3~D diffusion model. J. Env. Eng.
Div., ASCE 102:805-820.
52. Lai, S. -H and J. J. Jurinak, 1971.
Numerical approximation of cation
exchange in miscible displacement
through soil columns. Soil Sci. Soc.
Amer. Proc. 35= 894-899.
53. Lapidus, L. and N. R. Amundson, 1952.
Mathematics of adsorption in beds, VI.
The effect of longitudinal diffusion
in ion exchange and chromatographic
columns. J. Phys. Chem. 56:984-988.
54. Larson, N. M. and M. Reeves, 1976.
Analytical analysis of soil-moisture
and trace-contaminant transport.
Oak Ridge National Laboratory, ORNL/
NSF/EATC-12.
55- Lee, C. H. and R. T. Cheng, 1974.
On seawater encroachment in coastal
aquifers. Water Resour. Res. 10:1039"
1043.
56. Leenheer, J. A. and J. L. Ahlrichs,
1971- A kinetic and equilibrium study
of the adsorption of carbaryl and
parathion upon soil organic matter
surfaces. Soil Sci. Soc. Amer.
Proc. 35:700-704.
57- Lessi, J., 1976. Simulation numerique
de la propagation d'un polluant dans
un milieu poreux sature. These de
Docteur-lngenieur, Universite de
Mecanique des Fluides de Strasbourg.
Universite Louis Pasteur, Strasbourg,
France.
58. Lindstrom, F.T., R. Haque, V. H. Freed
and L. Boersma, 19&7- Theory on the
movement of some herbicides in soils.
Env.Sci. 6 Techn. 1 :561-565-
59. Lindstrom, F.T. and L. Boersma, 1970.
Theory on the chemical transport with
simultaneous sorption in a water
saturated porous medium. Soil Sci.
110:1-9.
60. Lindstrom, F. T. and L. Boersma, 1971-
A theory on the mass transport of
previously distributed chemicals in
a water saturated sorb ing porous
medium. Soil Sci. 111:192-199-
61. Lindstrom, F.T., L. Boersma and D.
Stockard, 1971. A theory on the mass
transport of previously distributed
chemicals in a water-saturated sorbing
porous medium: isothermal cases.
Soil Sci. 112:291-300.
62. Lindstrom, F. T. and L. Boersma, 1973-
A theory on the mass transport of
previously distributed chemicals in
a water-saturated sorbing porous
medium. III. Exact solution for first-
order kinetic sorbtion. Soil Sci.
115:5-10.
63. Lindstrom, F. T. and W. M. Stone,1974a.
On the start up or initial phase of
linear mass transport of chemicals in
a water saturated sorbing porous
medium. I. SIAM J. Appl. Math. 26:
578-591.
211
-------
6*4. Lindstrom, F.T. and W. M. Stone,
On the start up or initial phase of
linear mass transport of chemicals in
a water saturated sorbing porous
medium. II. Integral equation approach.
SIAM J. Appl. Math. 26: 592-606.
65. Maisel, H. and G. Gnugnol i , 1972.
Simulation of discrete stochastic
systems. Science Research Associates,
Inc. , Chicago.
66. Marino, M. A., 197*». Distribution of
contaminants in porous media flow.
Water Resour. Res. 10:1013-1018.
67. Martinez, J. D., D. H. Kupfer, R. L.
Thorns, C. G. Smith and R. C. Kolb,
1975- An investigation of the utility
of Gulf Coast salt domes for the
storage or disposal of radioactive
wastes. Institute for Environmental
Studies, Louisiana State University,
Baton Rouge, Louisiana.
68. Mercado, A., 1976. Nitrate and
chloride pollution of aquifers: a
regional study with the aid of a
single-cell model. Water Resour. Res.
12:731-747.
69. Oddson, J. K., J. Letey and L. V.
Weeks. 1970. Predicted distribution
of organic chemicals in solution and
adsorbed as a function of position
and time for various chemical and
soil properties, Soil Sci . Soc. Amer.
Proc.
70. Ortob, G. T. and P. C. Woods, 1967.
Water-qua! ity management in irrigation
systems. J. Irr. Drain. Div., ASCE,
93:49-66.
71. Perez, A. I., W. C. Huber, J. P.
Heaney and E. E. Pyatt, 1974. A water
quality model for a conjunctive
surface-groundwater system. EPA-600/
5-74-013. Office of Research and
Development, U. S. Environmental
Protection Agency, Washington, D.C.
72. Pickens, J. F. and W. C. Lennox,
1976. Numerical simulation of waste
movement in steady groundwater flow
systems, Water Resour. Res. 12: 171-
180.
73. Pinder, G. F., 1973. A Galerkin-
finite element simulation of ground-
water contamination of Long Island,
New York. Water Resour. Res. 9: 1657-
1669.
74. Pinder, G. F. and H. H. Cooper, Jr.,
1970, A numerical technique for calcu-
lating the transient position of the
saltwater front. Water Resour. Res.
6:875-882
75. Pinder, G. F. and W. G. Gray, 1977.
Finite element simulation in surface
and subsurface hydrology. Academic
Press, New York.
76. Pohland, F. G., 1975- Sanitary land-
fill stabilization with leachate
recycle and residual treatment. EPA-
600/2-75-043. Office of Air, Land,
and Water Use, U.S. Environmental
Protection Agency, Cincinnati, Ohio.
77- Quasim, S., 1965- Chemical characteris-
tics of seepage water from simulated
landfills. Ph.D. thesis. West Virginia
University, Morgantown, Virginia.
78. Reddell, D. L.and D. K. Sunada, 1970.
Numerical simulation of dispersion in
groundwater aquifers. Hydrology Paper
No. 41, Colorado State University,
Fort Collins, Colorado.
79- Robertson, J. B., 197**. Digital
modeling of radioactive and chemical
waste transport in the Snake River
plain aquifer at the National Reactor
Testing Station, Idaho, U.S. Geologi-
cal Survey, Open-file Report IDO-
22054.
80. Robertson, J. B. and J. T. Barraclough,
1973- Radioactive and chemical-waste
transport in groundwater at National
Reactor Testing Station. Idaho: 20-
year case history, and digital model.
In Underground waste management and
artificial recharge. 1:291-322, Ameri-
can Association of Petroleum Geolo-
gists.
81. Robson, S. G., 197**. Feasibi 1 ity of
water-quality modeling illustrated by
application at Barstow, California.
Water Resour. Invest. Rep. 46-73,
U. S. Geol. Survey, Menlo Park,
Cali fornia.
82. Schwartz, F. W., 1975. On radioactive
waste management: an analysis of the
parameters controlling subsurface
212
-------
contaminant transfer.
51-71-
J. Hydro!. 27:
83. Schwartz, F. W., 1977. On radioactive
waste management: model analysis of
a proposed site. J. Hydrol. 32:257-
277-
84. Segol, G., 1976. A three-dimensional
Galerkin finite element model for the
analysis of contaminant transport in
variably saturated porous media.
User's guide. Department of Earth
Sciences, University of Waterloo.
85. Segol, G., 1977- A three-dimensional
Galerkin-finite element model for the
analysis of contaminant transport in
saturated-unsaturated porous media.
In Finite Elements in Water Resources,
W. G. Gray et al. (eds.), Pentech
Press, London, pp. 2.123-2.144.
86. Segol, G., G. F. Pinder and W. G.
Gray, 1975- A Galerkin-finite element
technique for calculating the trans-
ient position of the saltwater front.
Water Resour. Res. 11:343-347-
87. Selim, H. M., R. S. Hansel 1 and A.
Elzeftawy, 1976- Distributions of
2, 4-D and water in soil during
infiltration and redistribution. Soil
Sci. 121:176-183-
88. Selim, H. M. and R. S. Hansel 1, 1976.
Analytical solution of the equation
for transport of reactive solutes
through soil. Water Resour. Res.
12:528-532.
89. Selim, H. H., J. H. Davidson and
P.S.C. Rao, 1977- Transport of re-
active solutes through multilayered
soils. Soil Sci. Soc. Amer. J. 41:
3-10.
90. Shaffer, H. J., R. W. Ribbens and
C. W. Huntley, 1977- Prediction of
mineral quality of irrigation return
flow. V. Detailed return flow
salinity and nutrient simulation
model. EPA-600/2-77-179e. Office of
research and Development, U. S.
Environmental Protection Agency, Ada,
Oklahoma.
91. Shah, D. B., G. A. Coulman, L. T.
Novak and B. G. Ellis, 1975- A
mathematical model for phosphorus
movement in soils. J. Env. Qua 1.
4:87-92.
92. Swanson, R. A. and G. R. Dutt, 1973-
Chemical and physical processes that
affect atrazine movement and distribu-
tion in soil systems. Soil Sci. Soc.
Amer. Proc. 37: 872-876.
93- Sykes, J. F., 1975- Transport phenom-
ena in variably saturated porous
media, Ph. D. thesis, Department of
Civil Engineering, University of
Waterloo.
94. Tang, D. H. and G. F. Pinder, 1977-
Simulation of groundwater flow and
mass transport under uncertainty,
Advances in Water Resources 1: 25~30.
95. Tanji, K. K., 1970. A computer analy-
sis on the leaching of boron from
strafified soil columns. Soil Sci.
110:44-51.
96. Tanji, K. K., G. R. Dutt, J. L. Paul
and L. D. Doneen, 1967a- Quality of
percolating water II. A computer
method for predicting salt concentra-
tions in soils at variable moisture
contents. Hilgardia 38:307-318.
97. Tanji, K. K., L. D. Doneen, and
J. L. Paul, 1967b. Quality of per-
colating water III. The quality of
waters percolating through strati-
fied substrata, as predicted by com-
puter analysis. Hildardia 38: 319~347.
98. Tanji, K. K. , L. D. Doneen, G. V.
Ferry and R. S. Ayers, 1972. Computer
simulation analysis on reclamation
of salt-affected soils in San Joaquin
Valley, California, Soil Sci. Soc.
Amer. Proc. 36:127-133-
99- Thorns, R. L., C. G. Smith and
J. D. Hartinez, 1977. Denial salt
plumes in ground water. In Finite
Elements in Water Resources, W. G.
Gray et al (eds.). Pentech Press,
London, pp. 2.37-2.49-
100. Ungs, H. , R. W. deary, L. Boersma,
and S. Yingjajaval, 1976. The
quantitative description of transfer
of water and chemicals through soils.
In Land as a waste management al-
ternative. R. C. Loehr (ed.). Ann
Arbor Science, Ann Arbor, Hichigan,
213
-------
pp. 109-137.
101. van Genuchten, M. Th., 1978. Mass
transport in saturated-unsaturated
media. I. One-dimensional solutions.
Research Report 78-WR-ll, Water
Resources Program, Department of
Civil Engineering, Princeton Univer-
sity, Princeton, New Jersey.
102. van Genuchten, M. Th. , J. M. Davidson
and P. J. Wierenga, 1971*- An evalua-
tion of kinetic and equilibrium
equation for the prediction of pesti-
cide movement through porous media,
Soil Sci. Soc. Amer. Proc. 38:29-35-
103. van Genuchten, M. Th. and P. J.
Wierenga, 1976. Mass transfer studies
in sorb ing porous media. I. Analytical
solutions. Soil Sci. Soc. Amer. J.
106.
105.
van Genuchten, M. Th., G. F. Pinder
and W. F. Saukin, 1977. Modeling of
leachate and soil interactions in an
aquifer. J_r^ Management of gas and
leachate in landfills. S. K.
Banerji (ed.), EPA-600/9-77'026.
Office of Research and Development.
U. S. Environmental Protection
Agency, Cincinnati, Ohio, pp. 95-
103.
van Genuchten, M. Th. and R. W.
Cleary, 1978. Movement of sqlutes
in soil: computer-simulated and
laboratory results. In Soil
Chemistry. G. H. BolTfand M.G.M.
Bruggewert (eds.). Chapter 10.
Developments in Soil Science,
Elsevier Publ. Co., Amsterdam.
van Genuchten, M. Th. and G. F.
Pinder, 1978. Mass transport in
saturated-unsaturated media. II.
Two-dimensional solutions. Research
Report 78-WR-12, Water Resources
Program, Department of Civil
Engineering, Princeton University,
Princeton, New Jersey.
107.
108.
B. H.
Wang, S. T., A. F. McMillan and
Chen, 1977- Analytical model of
dispersion in tidal fjords. J.
Hydraul. Div., ASCE 103:737-751.
Warrick, A. W., J. W. Biggar and
D. R. Nielsen, 1971- Simultaneous
solute and water transfer for an
unsaturated soil. Water Resour. Res,
7:1216-1225.
109. Weber, E. M. and A. A. Hassan, 1972.
Role of models in groundwater manage-
ment. Water Resour. Bull. 8:198-206.
110. Wierenga, P. J., 1977- Solute dis-
tribution profiles computed with
steady state and transient water
movement models. Soil Sci. Soc. Amer.
J. 41:1050-1055-
111. Wood, A. L. and J. M. Davidson, 1975.
Fluometuron and water content dis-
tributions during infiltration:
measured and calculated. Soil Sci.
Soc. Amer. Proc. 39:820-826.
112. Yeh, G. and Y. Tsai, 1976, Analytical
three-dimensional modeling of effluent
discharges, Water Resour. Res. 12:
214
-------
FIELD VERIFICATION OF HAZARDOUS WASTE MIGRATION FROM LAND DISPOSAL SITES
J. P. Gibb
Illinois State Water Survey
ABSTRACT
The vertical and horizontal migration patterns of zinc, cadium, copper, and lead
through the soil and shallow aquifer systems at a secondary zinc smelter were defined by
the use of soil coring and monitoring-well techniques. The migration of metals that
occurred has been limited to relatively shallow depths into the soil profile by attenuation
processes. Cation exchange and precipitation of insoluble metal compounds, as a result of
pH changes in the infiltrating solution, were determined to be the principal mechanisms
controlling the movement of the metals through the soil. Increased metals content in the
shallow groundwater system have been confined to the immediate plant site.
Soil coring was determined to be an effective investigation tool, but not suitable by
itself for routine monitoring of waste disposal activities. However, it should be used
to gather preliminary information to aid in determining the proper horizontal and vertical
locations for monitoring-well design. The analysis of water samples collected in this
project generally did not provide an understandable pattern of results. A brief experiment
on monitoring-well sampling indicated the need to develop reproducible sampling techniques
to obtain representative water samples. The failure of some well seals in a highly pol-
luted environment also indicates the need for additional research in monitoring-well
construction.
INTRODUCTION
The primary purpose of this study was
to verify in the field the effectiveness
of glaciated region soils and associated
surface deposits in retaining specific
hazardous chemicals. The study also v?as
designed to develop effective investigative
and monitoring techniques for detecting and
quantitatively evaluating the extent of
groundwater pollution from surface waste
disposal activities.
Special emphasis was placed on defin-
ing: 1) the vertical and horizontal migra-
tion patterns of chemical pollutants
through the soil and shallow aquifer sys-
tems; and 2) the residual chemical buildup
in soils in the vicinity of pollution
sources. In accomplishing these goals, an
understanding was developed for the prac-
tical aspects of core drilling, soil sam-
pling, piezometer installation, and water
sample collection.
Four industrial complexes were select-
ed to study the effects of their waste dis-
posal practices on the soils and shallow
groundwater systems. For the purposes of
this paper, the site where most of the
time, money, and effort was spent will be
discussed. This was referred to as Site A.
Site A is a secondary zinc smelter
located in south-central Illinois. The
plant started operations between 1885 and
1890, initially processed zinc ore, and was
converted to a secondary zinc smelting
facility about 1915. Wastes from the'
smelting operations during the first 85
years principally were heavy metals-rich
cinders and ashes. During the early years
large quantities of cinders were used as
road fill or surfacing in the plant area.
As a result of these disposal practices,
there now is a 1- to 10-foot (0.30- to
3.04-m) thick layer of metals-rich cinders
covering about 12 acres of the plant pro-
perty.
In compliance with dir pollution con-
trol regulations, a scrubber was installed
on the plant stack in 1970. Prior to that
time, wind-blown ash, rich in zinc and
other neavy metals, was deposited on the
plant site and on the surrounding farmland.
215
-------
This source of pollution has now been mini-
mized, but wastewater from the scrubber is
disposed of in a seepage pit constructed on
the cinder materials that form the present
day land surface. Several hundred tons of
high zinc content sludge have accumulated
from the frequent cleaning of this pit and
are being reprocessed for zinc recovery.
Most of the water from the pit infiltrates
into the ground underlying the plant prop-
erty.
PRINCIPAL FIELD TECHNIQUES
The collection of continuous vertical
core samples for geologic study and chemi-
cal analyses, and the construction of pie-
zometers or monitoring wells for water
level measurements and water sampling were
the principal field techniques used in this
study.
Continuous vertical core sampling was
conducted with conventional Shelby tube and
split spoon sampling methods through hollow
stem augers. These dry drilling techniques
were used to minimize the chemical altera-
tion of samples from drilling fluids or
external water sources.
Coring was done with a truck mounted
Central Mining Equipment (CME) 55 and a CME
750 rig mounted on an all-terrain vehicle.
The drilling crew consisted of an equipment
operator and helper, assisted if necessary
by the principal investigators. For the
first few holes drilled, a geologist from
the Illinois State Geological Survey as-
sisted in collecting samples and made pre-
liminary soil identifications for use in
subsequent drilling.
Core Sampling
Shelby tube and split spoon samples
were extruded in the field, cut into 6-inch
lengths, placed in properly labeled wide
mouth glass jars, and subsequently deliver-
ed to the Illinois State Geological Survey
and Environmental Research Analytical Lab-
oratory for processing and analysis. One
6-inch (15.2 cm) length of core from each
5-foot (1.52 m) segment or change in forma-
tion was taken by the drilling contractor
for moisture content determination before
being sent for geologic and chemical analy-
sis.
Core Analysis
Core samples for heavy metals determi-
nations were analyzed at the Environmental
Research Analytical Laboratory with zinc as
a target element. Previous experience in
determining heavy metal contaminants in
soil showed that digestion of a dried soil
sample in 3 N HC1 at slightly elevated
temperatures effectively released the heavy
metals without destructing the silicate
lattice of the soil. The heavy metals so
released were determined primarily by
atomic absorption spectroscopy.
For a number of soil samples, the
multi-element capability of optical emis-
sion spectroscopy was used to determine
cadmium, copper, lead and zinc concentra-
tions.
Tests using atomic absorption measure-
ment of small spot samples from the 6-inch
long samples indicated that they were too
heterogeneous to permit reproducible analy-
sis. Reproducible results were attained by
homogenizing the samples and subdividing
them to sample weight levels of 1 gram.
Well Construction
Analyses of water samples from obser-
vation wells has been the traditional me-
thod for monitoring groundwater pollution.
To demonstrate the effectiveness of this
approach and the relative cost of using
wells compared with coring, a number of
small-diameter (2-inch) (5.04 cm) observa-
tion wells were constructed. Since heavy
metal contaminants were expected, plastic
casing, screen, and pumping equipment were
used.
Observation wells were constructed in
the following manner. A 7-inch (17.78 cm)
diameter hole was constructed, and a 2-inch
diameter PVC pipe (bottom 2 feet (61 cm)
slotted with a hacksaw) was placed in the
open borehole. Gravel was placed from the
bottom of the hole to a level about 1 foot
(30 cm) above the slotted portion of the
pipe. The remainder of the annulus was
filled with bentonite slurry to land sur-
face (see Figure 1).
In late February 1975, engineers from
the American Colloid Company, Volclay Divi-
sion, informed us of a possible problem
associated with using standard sodium ben-
tonite to seal out highly mineralized
216
-------
1/2" discharge
pipe
7" bore hole
slotted PVC ^
well casing °="
3 0°
°»°
Oo
00
o °
1/2" pipe
^1/2" cap
"Nvent hole
Shrader valve
bentonlte
s lurry
1/4" airline
6" sand
12"
-gravel
00
0°0
0
O 9
Figure 1 Typical well and pumping mechanism
waters such as those encountered in our
study. They indicated that the%normal
swelling-sealing capability of sodium ben-
tonite could be so seriously affected by
such waters that well seal failures were
almost inevitable. It was their opinion
that some of our observation wells had al-
ready failed because of this factor. Water
samples from shallow and deep wells were
collected and submitted to American Colloid
with a request for them to design a seal-
ant mixture which would insure no leakage
between aquifers.
In general, they concluded that an ef-
fective seal cannot be achieved with ben-
tonite in an environment that is already
affected by pollutants such as at this
site. They suggested that monitoring wells
be installed prior to disposal of waste
(not always a very practical solution).
Pumping Mechanism
Observation wells were equipped with
individual pumping devices to minimize pos-
sible contamination of samples from other
wells. The pumping devices consisted of a
1/2-inch (1.27 cm) diameter PVC discharge
pipe that extended from above the 2-inch
(5.08 cm) well casing to the bottom of the
well. A tee fitted with short nipples and
removable caps was placed at the top of
this pipe (Figure 1). The cap on the ver-
tical segment could be removed to allow for
water level measurements within the 1/2-
inch pipe. The cap on the horizontal seg-
ment (water discharge outlet) was vented to
permit stabilization of the water level
within this pipe.
A 1/4-inch (0.63 cm) plastic airline
also was installed in eacn well. The air-
line was attached to a Shrader valve at the
top of the well casing and extended the en-
tire depth of the well. The lower end of
the airline was bent up into the bottom of
the 1/2-inch discharge pipe for a distance
of about 3 inches (7.62 cm).
Water was pumped from the wells by re-
moving the cap from the horizontal portion
of the 1/2-inch pipe and applying air to
the system through the Shrader valve.
Pumping from depths as great as 70 feet
(21.3 m) was possible with only a bicycle-
type hand pump. A gasoline powered, 4-
cycle air-compressor capable of delivering
about 5 cubic feet per minute (0.14 cubic
meter per minute) at pressures up to 60 psi
(4.2 kg/sq cm) also was used. An activated
charcoal filter was placed in the airline
from the compressor to insure that air from
the compressor was not introducing airborne
contaminants.
Since most of the wells in this study
generally had a column of water less than
30 or 40 feet (9.14 or 12.19 m) deep, it
was found that operating the compressor at
20 to 25 psi (1.4 to 1.8 kg/sq cm) was most
effective. Higher operating pressures
caused the air bubbles to rise through the
water column instead of pushing slugs of
water in front of them as desired. In the
very shallow wells, less than 20 feet (61m)
deep, the bicycle type hand pump actually
worked more effectively.
Water Sampling
Water level measurements were made and
water samples were collected from each well
once a month. Water samples were collected
in 6-ounce (170 grams) plastic containers
and placed on ice until they were brought
into the laboratory where they were refri-
gerated. Each well was pumped for a period
of time theoretically adequate to insure
217
-------
that all stored water in the well casing
had been removed. The wells were allowed
to recover, and a sample was then collected
from the water that had just entered the
well. This procedure was followed in hopes
that the water sample would be representa-
tive of the water flowing through the
aquifer at the time of collection.
Near the end of the project a brief
experiment was conducted on four wells at
this site to determine if the pumping
scheme just described was necessary or ade-
quate for collecting representative water
samples. Four wells were selected on the
basis of early results of chemical analyses
of water samples collected. Zinc concen-
trations in water from these wells ranged
from 6.2 to 25.9 mg/1, 300 to 790 mg/1,
342 to 850 mg/1, and 12,700 to 21,580 mg/1,
respectively. These values represent fluc-
tuations in zinc content of 76, 62, 60, and
42 percent using the higher values as base
concentrations.
To determine if these fluctuations
were real or a function of the sampling
procedure, the following experiment was
conducted. Water level measurements were
taken in each well and the volumes of water
stored in the 2-inch diameter casings and
screens were calculated. Pumping was ini-
tiated and samples were collected just after
pumping began and at increments of one-half
the total stored water volume until a total
of 5 volumes had been pumped.
Figure 2 illustrates the results of
these tests for the four wells. Percentage
decreases in zinc concentrations from the
first sample to the last ranged from about
45 to 78 percent with the greatest decreases
occurring in the wells with the lowest zinc
concentrations. Reductions in zinc content
at the 1 volume pumped stage (the procedure
followed in our sampling program) ranged
from about 18 to 46 percent. This suggests
that all of the zinc determinations of
water samples collected could be as much as
30 percent higher than the stabilized zinc
content beyond the 5 volume pumped stage.
If the sampling procedure employed varied by
even as little as 50 percent, pumping only
one-half or one and one-half the stored vol-
ume, it could account for as much as 15 to
20 percent of the fluctuations in the sample
results.
Additional experiments of this type
should be undertaken to design satisfactory
sampling procedures for other chemicals.
Further, these experiments should be con-
Figure 2
12345
VniUMES OF STORED WATER PUMPED
Effects of pumping on zinc content of water samples
218
-------
ducted at the beginning of a project as op-
posed to near the end as in this case. The
results of these brief tests indicate the
need for further research in developing
suitable water sampling procedures.
Water Analyses
Water samples for heavy metals analyses
were analyzed at the Environmental Research
Analytical Laboratory with zinc as the tar-
get element. Two electrochemical methods,
anodic stripping voltammetry and pulse pol-
arography, proved to be most effective for
making zinc determinations and screening for
the presence of other metals of interest.
Total mineral analyses were conducted at the
laboratories of the Illinois State Water
Survey by standard procedures.
RESULTS
Altogether, 49 wells at 36 locations
were completed at Site A. Core samples were
taken at each of these locations and at an
additional 23 sites (see Figure 3). Total
well and core sampling footages are about
1309 and 1454 feet (398 and 443 m),
respectively.
Geology
The glacial materials at this site
range in thickness from about 55 feet
(16.7 m) on the east to about 75 feet
(22.8 m) on the west. The stratigraphic
units recognized are essentially uniform in
character and thickness and generally flat
lying across the site. The elevation of the
bedrock surface dips from 449 feet (137 m)
above sea level on the east to 432 feet
(132 m) on the west.
Table 1 summarizes the textual and
mineralogical information for each unit. A
brief description of each stratigraphic unit
is as follows:
(A) Peoria Loess (4 to 6')—Brownish-gray
clayey silt, non-calcareous, with
iron stains.
(D) Roxanna Silt (3 to 4')—Dark brown
clayey silt with up to 34 percent
sand (av. 20 percent), noncalcareous
Glasford Formation
(C) Berry Clay Member (3 to 4')—Dark
gray sandy silty clay with trace
gravel, up to 40 percent sand, non-
calcareous (an accretion gley)
EXPLANATION
. O C3RL HOLE
• WELL
SULE OF FEET
" *"• "4
Figure 3
Location Map - Site A
219
-------
Table 1
Textural and mineralogical data for stratigraphic units
Site A
STAGE
WISCONSINAN
SANGAMONIAN
ILLINOIAN
YARMOUTH I AN
KANSAN
PENNSYLVANIAN
SYSTEM
UNIT
PEORIA LOESS
ROXANA SILT
BERRY CLAY
MEMBER-
6LASFORD
FORMATION
HAGARSTOWN
MEMBER-
GLASFORD
FORMATION
G LAS FORD
FORMATION
TILL
LIERLE CLAY
MEMBER-
BANNER
FORMATION
BANNER
FORMATION
TILL
BONO
FORMATION
AVERAGE
TEXTURE
(4-59-37)
10 samples
(20-47-33)
9 samples
(33-31-36)
SAND INCREASES
TOWARD BASE
13 samples
(46-33-21)
VARIABLE
3 samples
(31-40-29)
SANDIER
NEAR TOP
79 samples
(24-41-35)
8 samples
(25-44-31)
SANDIER
NEAR TOP
40 samples
NO
INFORMATION
AVERAGE CLAY
MINERALOGY (<2y)
M 84.5*
I 11%
8 samples
M 85%
I 8%
7 samples
M 77%
I 13%
13 samples
M 30%
I 51%
3 samples
M 45%
INCREASES
WITH DEPTH
I 402
DECREASES
WITH DEPTH
51 samples
M 45%
I 37%
VARIABLE
6 samples
M 16%
I 55%
VARIABLE
31 samples
M 3%
I 4%
2 samples
CARBONATE
MINERALOGY (<2p)
LEACHED
LEACHED
LEACHED
LEACHED
CONTAINS
CARBONATES
MAY BE
LEACHED
AT TOP
LEACHED
CONTAINS
CARBONATES
MAY BE
LEACHED
AT TOP
LEACHED
(4-59-37) = Average percentage of sand, silt, and clay excluding gravel.
M 64£ = Average percentage of montmorillonitic (expandable) minerals
in clay fraction (<2y).
I 11% = Average percentage of illite in clay fraction.
(D,E) Vandalia Till Member (2 to 8') —
Yellowish brown, oxidized in upper
portion, little yellowish gray at
base
(F) Smithboro Till Member (25 to 31') —
Dark gray sandy silt
(G) Lierle Clay Member (1 to 2')—-Dark
olive brown sandy silty clay, with
little gravel, noncalcareous, soft.
Banner Formation
(H) Unnamed Till Member (1 to 2')—Gray
clayey silty till, sandy, with some
gravel, mottled yellowish brown,
brown and gray in upper part, strongly
calcareous, hard, dry, a few scattered
silt and sand lenses, predominantly
dry Bond
Formation (Pennsylvanian bedrock)—
Greenish-gray shale with abundant
mica; hard, dry
The expandable clay minerals, generally
referred to as montmorillonite (M), make up
more than 80 percent of the clay minerals
within the Peoria Loess, Roxana Silt, and
Berry Clay Member, thereby suggesting a high
base exchange potential in the upper 13 feet
of the materials encountered. The thin,
continuous silty sand zone at the top of the
Vandalia Till would appear to be the only
"permeable" unit to allow groundwater to
travel laterally at any moderate rate away
from the site. Although there is probably
some downward movement of groundwater
220
-------
through the remainder of the Glasford Form-
ation, it would be expected to be extremely
low.
Hydrology
On the basis of the geologic descrip-
tion of this site, it is quite obvious that
there is no significant aquifer present in
the immediate vicinity of the plant site.
The Hagerstown Member, a thin (1 to 2 feet
thick) continuous silty sandy zone, appears
to be the only permeable zone that could
allow for significant lateral groundwater
movement away from the site. To develop
even a domestic water supply from this sand
unit probably would require the construc-
tion of two or more large-diameter bored
wells.
To determine the hydrologic charac-
teristics of this unit a pumping test with
3 observation wells was conducted at well
site 2 on August 12, 1975. Well 2S (S =
shallow well) was pumped for a period of 3
hours at rates from 0.180 to 0.111 gallons
per minute (0.011 to 0.007 liters per
second). Observation wells 1, 2, and 3
were located 7.2, 14.0, and 28.0 feet (2.19,
4.26, and 8.53 m) north of the pumped well,
respectively. All drawdown data were ad-
justed to the final pumping rate (0.111 gpn)
for analysis purposes.
The average computed coefficients of
transmissivity, permeability, and storage
are 285 gpd/ft (3.53 IP /day), 190 gpd/ft2
(7.75 m/day), and 0.00128, respectively.
Water table contour maps were drawn
for each round of water level measurements
made. Figures 4a and b for March 1976 and
November 1975 illustrate the high and low
water table configurations, respectively.
In both instances there is a water table
mound beneath the plant site and movement
of water is in all directions away from the
plant complex. The relatively high permea-
ability of the fill materials at the plant
site, its topographic setting (higher than
the surrounding land), and the liquid dis-
posal activity at the plant all contribute
to the development of this recharge mound.
Water levels for deep wells 5D (D =
deep well), 6D, 7D, and 8D took approxi-
mately 15 months to stabilize. This was
due to the relatively impermeable materials
these wells were completed in. Wells ID,
2D, 3D, ^D, and 9D were completed in sand-
ier units and therefore were reflecting
stabilized water levels within 1 or 2
months. Because of the slow recovery rates
of some deep wells, they were not sampled
monthly as were the shallow wells.
As a result of the problem of not hav-
ing stabilized water levels in some of the
deep wells for the period of record, month-
ly piezometric surface maps were not drawn.
In general, water levels in the deep wells
are higher than those in the shallow wells
indicating a probable upward movement of
water within the glacial drift sequence.
In the immediate plant area, where the
shallow water levels are mounded, the move-
ment of water in the shallow deposits pro-
bably is downward and horizontally while
water in the deeper units probably is mov-
ing upward and horizontally. An average
apparent rate of groundwater movement in
the shallow deposits was calculated for the
immediate plant area with Darcy's equa-
tion v = PI where v = apparent or bulk
velocity, P = permeability, and I = hydrau-
lic gradient. The average rates of ground-
water movement from the mounded area in
March 1976 and November 1975 where 0.17 and
0.19 ft/day (5.2 and 5.8 cm/day) respective-
ly. The actual or effective velocity is
described by Hantushd) as the apparent velo-
city divided by the effective porosity of the
soil or aquifer. The effective porosity is
the portion of pore space in a saturated
permeable material in which flow of water
takes place. Not all of the pore space of
a material filled with water is open for
flow, since part of the voids are filled
with water that is held in place by molecu-
lar and surface water tension forces.
The porosities of the aquifer materials
(Hagerstown Formation) at Site A were mea-
sured to 0.32. On the basis of data pre-
sented by Todd^ , an effective porosity or
specific yield of 0.10 was assumed. With
this value, effective velocities of 1.7 and
1.9 ft/day (0.52 and 0.58 m/day) or 620 to
690 ft/year (189 to 210 m/day) were calcu-
alted for the mounded area of the plant.
These unexpected high rates of move-
ment can be explained by the relatively
steep hydraulic gradients developed beneath
the plant complex. Similar calculations in
areas removed from the influence of the re-
charge mound resulted in average effective
velocities of 0.15 to 0.40 ft/day (0.05 to
0.12 m/day) or 55 to 145 ft/year (16.7 to
44.2 m/yr).
221
-------
EXPLANATION
_ ELEVATION !N FEET
—
>x^v'.r/ f* ^x
y^i/^ y ;<*it> •
I c.5\ j3 o S' * "~"°-NN'
,n . • . V
N •
\ \' I
• fa" V o I mo '".Hi o" o
_A IT^r-V!1 ^s^.
Figure 4
,n, ELEVATION IN FEET
=U<: ABOVE MEAN SEA LEVEL
c o CORE HOIE
• WELL
Water table contour maps for a) March 1976, and b) November
1975 - Site A
222
-------
Chemical Data
Aside from providing data for detailed
geologic description, chemical analysis of
the core samples were conducted to define:
1) the vertical and horizontal migration
patterns of chemical pollutants through the
soil; and 2) the residual chemical buildup
in soils in the vicinity of the pollution
source. Preliminary analyses of core sam-
ples during the early stages of the study
indicated that four elements (zinc, cadium,
copper, and lead) were most likely to be
carried into the soils and groundwater sys-
tem beneath the plant property. As a re-
sult these elements were selected for rou-
tine analytical analyses.
Results of chemical analysis of core
samples from the Site A control hole lo-
cated approximately 3 miles south-southwest
of the plant, and samples from unaffected
soil horizons beneath the plant property
suggest that background concentrations for
the four elements tested should be about 20
to 50 mg/1 for zinc, 0.04 to 1.5 mg/1 for
cadium, 10 to 30 mg/1 for copper, and 10 to
40 mg/1 for lead. There appears to be no
significant chemical variation with depth
or between geologic unit boundaries. Some
zinc levels in isolated Pleistocene soils
were higher.
To outline the limits of migration of
these metals beneath the plant and give an
indication of the effectiveness of the
soils in retaining these metals, a series
of cross sections showing zinc, cadmium,
copper, and lead concentrations of the soil
were prepared. In figure 5 the north to
south cross section illustrates that north
of the railroad tracks rather small quanti-
ties of zinc were found in the upper 3 to 5
feet of the soil profile. Most of the zinc
introduced into this area probably was wind
blown dust and ashes from the plant stack.
The area of greatest accumulation and deep-
est penetration of zinc in the soil occur-
red immediately beneath the plant property.
Two principal sources of pollution, the
cinders covering the plant property and the
scrubber waste water, have resulted in
large quantities of zinc moving into the
soil profile in this area. The effect of
the scrubber wastewater discharge is obvi-
ous in both cross sections. Beneath well
12 the depth of penetration (to the 100 mg/1
boundary) is approximately 28 feet (3.65m).
However, it is interesting to note that
lateral migration due to this activity has
still been very limited. It also is worth
noting that no significant lateral migra-
tion has taken place beyond the two drain-
age ditches bounding the plant on the west
and east. Farther south, beyond the limits
of the cinder covered portion of the plant
property, very limited zinc penetration
also has occurred.
Similar cross sections illustrating
the buildup of cadmium, copper, and lead
are shown in figures 6, 7, and 8. The
general shapes of these cross sections are
similar to those for zinc. The depth of
penetration of cadmium is slightly less
than that of zinc but considerably greater
than that of copper and lead.
The very shallow depths of penetration
of these elements substantiates results of
laboratory studies by Frost and Griffin^3^
indicating the relative immobility of these
metals. In general, the areas of greatest
penetration of the four elements occurred
beneath the scrubber wastewater pit where
wastewater is a significant source of the
metals and the recharge into the soil sys-
tem is greatest. In other areas removed
from the pit, the presence of the cinders
becomes the dominant source and lowland
areas where ponded water accumulates is the
secondary source.
In addition to the direct percolation
of metals-rich water at the plant site, a
significant amount of metals-rich surface
water is running off the plant property and
percolating into the stream beds draining
the plant to the southwest and southeast.
An accumulation of metals-rich cinder-type
sediments in the stream bed was noted. The
retention of metals from the percolating
water by the soil beneath the stream beds
is illustrated in figure 9. The concentra-
tion of metals retained and depth of pene-
tration decreases as the distance away from
the plant site increases.
The mechanisms retaining the metals in
the soil profile at Site A are predominant-
ly cation exchange and precipitation of in-
soluble metal compounds as a result of pH
change. Cation exchange capacity data gen-
erated indicate little variation in the
retention capabilities of the upper geolo-
gic units, the silts, clays, and tills.
Therefore, as metals-rich water percolates
downward through the soil profile, the me-
tals are exchanged preferentially in re-
verse order of their mobility.
223
-------
Figure 5
Soil zinc concentration profiles - Site A
224
-------
WEST
510-
500
UJ
UJ
-1 495
UJ
2 490
SOUTH
g 485
i-
UJ
UJ
"- 510
z
z
« 505
<
-
ul
1
J 500
495
490
485
Figure 6
EAST
i
S29 S28S27 S3
..I
S2 S37 S36
I I I
C5S12 CADMIUM, mg/1
NORTH
CADMIUM, mg/1
Soil cadmium concentration profiles - Site A
225
-------
[WEST
EAST
510r
505
500
495-
490
S29 S28S27 S3
C9
^
100
c, S37 S36
32 | | -
C5S12 COPPER, mg/1
Figure 7
Soil copper concentration profiles - Site A
226
-------
S37 S36
S2 I
| LEAD, mg/1
sis C2Q ss LEAD, mg/1
490
Figure 8
Soil lead concentration profiles - Site A
227
-------
505
500
495
490
LU
o 485
PLANT
.'
SCALE IN FEET
200 400
:•'
ZINC, mg/1
CO
:
i
•
505
500
495
490
485
•PLANT
10
3 ZINC, mg/1
Figure 9
Streambed soil zinc concentration profiles - Site A
Cation exchange capacities (CEC) of
soils measured at Site A range from about 4
to 10 me/100 grams with the larger values
occurring in the shallower soils. If zinc
were transferred onto the available ex-
change sites of the soil, cation exchange
could account for soil zinc concentrations
up to about 3500 mg/1. This value could be
higher according to some soil specialists,
because the measured CEC may be lower than
the original capacity of the soil. Some
researchers maintain that the soil becomes
"poisoned" in the presence of pollutants
and true CEC values cannot be measured.
Aside from that possibility, three
other factors can be used to explain the
difference between the very high soil zinc
concentrations shown in the upper part of
the soil profiles and the values attribut-
able to cation exchange:
1) Some of the very high values ob-
tained for the surface and near
surface samples actually are
chemical results of cinder fill
samples.
2) Immediately beneath the cinder
fill, fine-grained sediments
from the cinders have been
illuviated into the underlying
soil, also resulting in high
zinc values of those samples.
3) Soluble and insoluble salts of
zinc and the other metals may
be temporarily stored in the
aerated zone waiting for even-
tual migration downward with
later recharge events. This
also results in higher zinc
concentrations in soil samples
from this horizon.
228
-------
As the cation .exchange capacity of the
soil is exhausted and sufficient depth is
reached to eliminate the three factors just
noted, the metals buildup in the soil slow-
ly continues to advance deeper into the
soil profile. As the exchange process oc-
curs, calcium and magnesium are released
into the water from the soil and the pH of
the infiltrating fluid is lowered. As the
pH increases the formation of zinc precipi-
tates results and a sharp break or decrease
in soil zinc content is noted. Samples of
water collected after percolating through
the cinder fill materials forming the sides
of the disposal pit had measured pH values
near 5. It can be assumed that the same pH
wrs experienced by water filtering downward
through the cinder fill covering the plant
surface. It is suspected that sulfur con-
tained in the cinders was dissolved to form
a weak sulfuric acid thus creating the low
pH and increasing the mobility of the zinc.
The conclusions pertaining to the me-
chanism of zinc and other metals fixation
drawn from this field study are in agree-
ment with the results of recent laboratory
studies by Frost and GriffinO). They con-
clude that increased removal of metals from
solution occurs "with increasing pH values
and with increasing concentrations of the
heavy metal in solution."
Chemical analysis of water collected
from deep wells at Site A showed less than
0.5 rag/1 zinc. The zinc contents of water
collected from the shallow wells tapping
the Hagarstown sand unit are illustrated in
figures lOa and b. Figure lOa illustrating
data for samples collected September 9,
1976, represents the maximum extent of
pollution, and figure lOb illustrating data
for samples collected August 13, 1975, re-
presents the minimum extent of shallow
groundwater pollution. Because the sampl-
ing procedure was not satisfactory and the
attempt to detect well seal failures was
unsuccessful, further analysis of the water
quality data generated probably is not
worthwhile. As indicated earlier, the sam-
pling procedure used in this study could
account for as much as 40 to 80 percent of
the fluctuations noted between sampling
periods.
To better define the quality of water
in an affected and unaffected area, water
samples were collected from wells 3S and 6S
for total mineral analysis. The results of
these analyses and the analysis of a water
sample from the shallow control hole well
indicate general agreement for 6S (the un-
affected area well) and CH1-S. Increases
in mineral constituents in the affected
area well (3S) are obvious (see table 2).
The data from 3S was as expected and
substantiates the solubility product
calculations. The zinc concentration (750
mg/1) and pH (6.5) are in excellent agree-
ment with the solubility of zinc hydroxide
(800 mg/1 at 6.5 pH).
Evidence of ion exchange also is shown
by the high concentrations of calcium (2AOO
mg/1) and magnesium (893 mg/1) present in
this sample (3S). The cation exchange
positions in soils in this region of Illi-
nois are principally filled with calcium
and lesser amounts of magnesium. Grim' '
indicates that zinc is higher in the mont-
morillonite exchange series than Ca and Mg,
and thus will replace these ions on the
clay structure. According to Griffin(5),
this process releases calcium and magnesium
to the environment even when these are not
part of the original waste stream.
Because of these phenomena, it is re-
commended that total mineral analyses be
conducted on water samples from monitoring
wells where cation exchange is likely to
occur. Increases in one or more of these
constituents (calcium or magnesium) could
be an early warning of the eventual appear-
ance of the more toxic metals. However, to
properly interpret a series of samples the
problem associated with collecting compara-
tive samples must be solved.
Summary
The results of soil coring and water
sampling of wells at Site A have defined
the migration patterns of toxic metals from
this site into the ground and shallow
groundwater system. Cation exchange and
precipitation of metal compounds as a re-
sult of the change in pH of the infiltra-
ting fluid are the principal attenuating
mechanisms influencing the metals movement.
The geologic setting at Site A has
demonstrated the ability to contain high
concentrations of toxic metals over an ex-
tended length of time. The desirability of
this type of geology for disposal of simi-
lar types of pollutants has been clearly
demonstrated.
229
-------
EXPLANATION
C O CORE HOLE
• VEU
SCALE OF f£ET
0 190 200 300
Si
ZINC, mg/1
SEPTEMBER 9, 1976
Figure 10
Zinc content of water from shallow wells - Site A
230
-------
Table 2
Selected total mineral analyses data - Site A
Iron (total)
Manganese
Calcium
Magnesium
Strontium
Sodium
Potassium
Ammonium
Barium
Cadmium
Chromium
Copper
Lead
Lithium
Nickel
Zinc
Phosphate
(filt)
Phosphate
(unfilt)
Silica
Fluoride
Boron
Nitrate
Chloride
Sulfate
Alkalinity
Hardness
Total Dissolved Minerals
Fe
Mn
Ca
Mg
Sr
Na
K
NH
Ba
Cd
Cr
Cu
Pb
Li
Ni
Zn
P04
P°4
SiO
F
B
NO
C1J
so,
(as CaCO,)
(as CaCO,)
SCH1-S
mg/1
2.3
.14
96.4
38.3
.27
75.4
1.0
0.1
<0.1
.00
.00
.00
<.05
.01
<.05
.00
0.1
0.5
15.8
0.3
0.0
0.5
6
140.1
404
398
S-6S
me/1
4.81
3.15
.01
3.28
.03
.01
.01
.17
2.91
8.08
7.96
mg/1
4.7
.44
33.5
13.7
.08
81.9
0.6
0.0
<0.1
.00
.00
.00
<.05
.00
<.05
.04
0.1
0.4
13.2
0.3
0.1
1.2
20
120.1
166
140
me/1
1.67
1.13
3.56
.02
.00
.02
.56
2.50
3.32
2.80
mg/1
0.0
21.01
2400
893
5.25
389
367
156
<0.1
1.19
.01
.04
<.05
.15
2.8
750
0.0
0.0
33.1
0.0
0.5
223
8300
564.0
40
9660
S-3S
me/1
119.76
73.44
.12
16.92
9.39
8.64
.02
.10
22.95
3.59
234.06
11.73
.80
193.20
615
394
13802
The successful application of soil
coring and water sampling from wells has
proven the value of these techniques as
research tools. The complementary data
generated from each technique are necessary
to fully understand toxic metals migration.
COST ANALYSIS
Detailed cost data for collecting core
samples and installing piezometers were
analyzed for all sites in this study. The
cost of routine core sampling in a clay en-
vironment using a professional drilling
crew was $6.50 to $7.00 per foot in 1974.
The cost of constructing piezometers as de-
scribed in this paper was $5.00 to $6.00
per foot in 1974.
In attempting to evaluate the relative
merits of soil sampling versus well install-
ation and water sampling, the costs of ana-
lytical work and the results obtained
should be considered. During the course of
this study the analysis of soil samples for
the four elements of interest, zinc, cad-
mium, copper, and lead, ran about $12.00.
If a single element assay was made it ran
about $7.00 Analysis of water samples for
the same four elements using electric ana-
lytical techniques ran about $7.00. Single
element assays ran about $5.00.
With the cost figures generated in
this study some comparisons can be made.
Assume that a 30-foot deep core and well is
constructed. The cost of coring would be
about $7.00/ft x 30 feet or $210. If ana-
lyses were made for every other 0.5-foot
segment, a total of 30 samples at $7.00
each for the four elements would cost about
$210. The 30-foot deep well at $6.00/ft
would cost about $180. If water samples
were collected monthly for one year and
analyzed for the four elements of interest,
the analytical work would cost about $84.
The total costs of coring and analysis
would be about $420 compared with $264 for
a well and one year of sampling and analy-
sis.
231
-------
Depending on the type of waste to be
disposed of or monitored, soil sampling may
prove to be more effective even though more
costly. If a research type project is
being conducted, coring would prove to be
most useful. The coring and analyses of the
core samples give a better understanding of
the phenomena that are taking place in the
soil. However, due to the costs, coring may
not be practical as a routine monitoring
technique.
On the basis of the results of this
study we recommend:
1) That coring and soil analysis be
used in waste disposal research
projects or in waste disposal
operations when the toxicity of
a waste product warrants the
additional expenditure.
2) That a limited number of core
samples, soil analyses, and
geologic interpretations would
be advisable for evaluating
alternative disposal sites.
Soil analyses provide a better
understanding of soil inter-
action with the waste product
than is otherwise possible, and
geologic interpretation of the
core samples provides better
information for design and loca-
tion of monitoring-wells.
3) That routine monitoring of most
disposal sites should be ac-
complished with wells located
and designed on the basis of
preliminary coring analysis.
Periodic coring and soil
analysis may be worthwhile to
substantiate original soil ef-
fectiveness assumptions.
REFERENCES
1. Hantush, Mahdi S., "Hydraulics of
Wells," "Advances in Hydroscience,"
Vol. 1 edited by Ven Te Chow. Academic
Press, New York, New York, 1964. pp.
284-286.
2. Todd, David K., "Ground Water Hydrolo-
gy," John Wiley & Sons, Inc., New York,
1967. pp. 23-26.
3. Frost, R. R., and R. A. Griffin, "Ef-
fect of pH on Absorption of Copper,
Zinc, and Cadmium from Landfill Leach-
ate by Clay Minerals," Journal of En-
vironmental Science and Health, Part A,
v. 12, in press.
4. Grim, Ralph, "Clay Mineralogy," McGraw
Hill Book Company, Inc., 1953. 384 pp.
5. Griffin, R. A., et al., "Attenuation of
Pollutants in Municipal Landfill Leach-
ate by Clay Minerals, Part 1 - Column
Leaching and Field Verification," En-
vironmental Geology Note 78, Illinois
State Geological Survey, Urbana, Illi-
nois, 1976. 34 pp.
232
-------
Adsorption, Movement, and Biological
Degradation of High Concentrations
of Selected Pesticides in Soils
J. M. Davidson, Li-Tse Ou, and P. S. C. Rao
Soil Science Department
University of Florida
Gainesville, Florida 32611
ABSTRACT
Equilibrium adsorption isotherms of the non-linear Freundlich type were obtained for
2,4-D amine, atrazine, terbacil, and methyl parathion and four soils from different loca-
tions within the United States. Pesticide solution concentrations ranged from zero to the
aqueous solubility limit of each pesticide. The mobility of each pesticide increased as
its concentration in the soil solution phase increased. These results were in agreement
with adsorption isotherm data. Biological degradation of each pesticide was measured by
14*C02 evolution resulting from the oxidation of uniformly 1(tC ring-labeled pesticides.
Technical grade and formulated material (10 to 20,000 vig/g of soil) were used. Pesticide
degradation rates and soil microbial populations generally declined as the pesticide con-
centration in the soil increased.
The use of pesticides continues to
increase despite known and suspected en-
vironmental problems and consequences. As
a result, greater quantities of empty con-
tainers with pesticide residues, unused
compounds which have been banned, and com-
mercial container waste must be disposed
of in an environmentally safe manner. In-
cineration, encapsulation, isolation in
underground caves and mines, chemical sta-
bilization, land spreading and land-fills
are some of the procedures being considered
for the disposal of pesticides and other
hazardous wastes (1, 2). Of these methods,
disposal by land-fills and land spreading
appear to be the most common and economi-
cal (3, 4). Placing hazardous wastes in
the soil has come under severe attack re-
cently (5, 6) because it does not guaran-
tee that the hazardous chemicals disposed
of in this manner will not migrate from
the disposal site or be degraded by
soil microorganisms. An understanding of
the various processes that Influence mi-
crobial degradation, retention, and leach-
ing of pesticides in soils is required be-
fore technology can be used for the selec-
tion and management of pesticide disposal
sites involving soils.
The ability of a soil to degrade
pesticides at low concentrations (less
than 10 ng/g soil) is well established, as
well as the movement of pesticides through
soils at these concentrations (7, 8). How-
ever, the direct extrapolation of this data
to systems containing high pesticide con-
centrations such as those occurring at o-r
below disposal sites may not be feasible
(9). Some pesticides have been reported
to persist in soils following large appli-
cations. After five years of parathion
(0,0-diethyl 0-p-nitrophenyl phosphoro-
thioate) application to fields (30,000 to
95,000 ppm), the lowest residue level was
13,000 ppm (10). The herbicide simazine
(2-chloro-4, 6-bis-ethylamino-s-triazine)
for repeated annual applications of 22.4
kg/ha, persisted much longer than for an-
nual applications of 2.8 and 5.6 kg/ha
(11). Some pesticides have enhanced soil
microbial activities and others have ex-
hibited adverse effects (12, 13).
233
-------
Several studies were initiated to in-
vestigate the physical, chemical, and bio-
logical processes determining the fate of
pesticides in soils when applied at high
concentrations. Details of these studies
have been reported elsewhere (14-21);
therefore, only a summary of the principal
findings and conclusions are presented in
this manuscript. Emphasis is given here
to: (i) the influence of adsorption char-
acteristics on pesticide mobility, (ii)
measured pesticide degradation rates over
a wide concentration range, and (iii) high
pesticide concentration effects on soil
microbial activity as measured by soil res-
piration rates as well as population dynam-
ics of bacteria, fungi, and actinomycetes.
The implications of the results reported
here are discussed with regard to predict-
ing the behavior of pesticides and other
hazardous wastes in or around disposal
sites.
MATERIALS AND IffiTHODS
Soils
Soils used in this study were: Web-
ster silty clay loam (Mollisol) from Iowa,
Cecil sandy loam (Ultisol) from Georgia,
Glendale sandy clay loam (Entisol) from
New Mexico, Eustis fine sand (Entisol) and
Terra Ceia muck (Histosol) from Florida.
These soil types were selected on the basis
of their geographic and taxonomic repre-
sentation of major U. S. soils. Surface
samples (0-30 cm) of.each soil were air-
dried and passed through a 2-mm sieve prior
to storage and use. Selected physical and
chemical properties of the mineral soils,
are given in Table 1. Terra Ceia muck is
characterized by 81% organic matter, 19%
total mineral content, CEC of 350 meq/100 g
and pH of 6.4.
Pesticides
Five pesticides included in this study
were: 2,4-D (2,4-dichlorophenoxyacetic
acid), atrazine (2-chloro-4-ethylamino-6-
isopropylamino-s-triazine), terbacil (3-
tert-butyl-5-chloro-6-methyluracil), methyl
parathion (0-0-dimethyl 0-p-nitrophenyl
phosphorothioate), and trifluralin (a, a,
a-trifluoro-2,6-dinitro-N,N-dipropyl-p-
toluidine). Selected properties of these
pesticides are given in Table 2. Stock
solutions of each pesticide in 0.01 N
CaCl2 were prepared using the technical
grade or formulated material of each pesti-
cide. Stock solutions were made up to the
aqueous solubility limit of each pesti-
cide. A mixture of antibiotics consisting
of Penicillin G and Polymixin B sulfate
was added to all pesticide solutions to
prevent microbial degradation during stor-
age and use. The stock pesticide solutions
were fortified with the appropriate 11+C-
labeled pesticide compound to give specific
activities in the range of 2 - 5nCi/ml.
Adsorption Experiments
Equilibrium adsorption isotherms for
all soil-pesticide combinations were mea-
sured (14) using the batch procedure.
Equilibrium was achieved by shaking dupli-
cate samples of five or tin grams of soil
with 10 ml of pesticide solution in pyrex
screw-cap glass test tubes for 48 hours.
Preliminary experiments had indicated that
no measurable increase in pesticide ad-
sorption occurred after 48 hours. Follow-
ing equilibrium, the test tubes were cen-
trifuged at 800 x G for ten minutes and
the 1^C-activity in one-mi aliquots of the
clear supernatant solution was counted by
the liquid scintillation method. Decreases
in pesticide solution concentration were
attributed to adsorption by the soil. All
adsorption experiments were performed at a
constant temperature (23±1 C).
Column Displacement Experiments
Pesticide movement through water-sat-
urated columns of soil was studied (14)
using the miscible displacement technique
described by Davidson et al. (22). The
air-dry soil was packed in small increments
into glass cylinders (15 cm long, 45 cm2
cross-sectional area). Medium porosity
fritted glass end plates served to retain
the soil in the column. The soil was
initially saturated with 0.01 N CaCl2 solu-
tion. A known volume of pesticide solution
at a desired concentration was introduced
to the soil at a constant flux using a
constant-volume peristaltic pump. After a
specific volume of pesticide solution had
been applied, the pesticide solution was
subsequently displaced through the soil
column with 0.01 N CaCl2 at the same soil-
water flux. Effluent solutions were col-
lected in 5 or 10 ml aliquots using an
automatic fraction collector. A pulse of
3H20 (specific activity 5 nCi/ml) was also
displaced through each soil column to char-
acterize the transport of non-adsorbed sol-
utes. The activity of lkC and 3HaO in ef-
234
-------
TABLE 1. Physical and chemical properties of the soils used in this study.
Base Extractable
PARTICLE SIZE FRACTION (%) pH (1:1 paste) CEC Organic C Saturation Acidity
SOIL sand silt clay Water IN KC1 (meq/100 g) (_%_) (%) (meq/100 g)
CO
01
Webster
Cecil
Glendale
Bust is
18.4
65.8
50.7
93.8
45.3
19.5
16.4
3.0
38.3
14.7
22.9
3.2
7.3
5.6
7.4
5.6
6.5
4.8
6.5
4.1
54.7
6.8
35.8
5t2
3.87
0.90
0.50
0.56
91
31
90
10
5,15
4.68
3.74
4.68
-------
TABLE 2. Properties of pesticides used in this study.
Common
Name
Molecular
Formula
Molecular
Weight
Vapor
Pressure
Solubility
(g/lOOg n.0)
Melting
Point
2,4-D
221.0
Not determined 0.09 at 25°C 135 - 138°c
(technical)^
2,4-D Amlne CiOH13C12N°3 266>1
1 X 10"10 mm 300 at 20°C
Hg at 38°C
85 - 87°C
Trifluralin Ci3Hi6F3N3°4 335<3
1.99 X 10~4 mm < 1 X ~4 at 48.5 - 49 Q°C
UK at 29.5°C 27°c
Terbacil
216.7
4.8 X 10 7 mm 0.071 at 25°C 175 - 177°c
Hg at 29.5°C
Atrazine C0H.,C1NC 215.7
o J.4 j
1.4 X 10~6 mn 0.0033 at 27°C 173 - 175°c
Hg at 30°C
parathion C.IL-.SPNO, 279.2
O 1U O
0.97 X 10~5 mm
Hg at 20°C
.0055-. 0060
at 25°C
36 - 36°C
fluent fractions was counted by liquid
scintillation. All displacements were
performed at a Darcy flux of approximately
0.22 cm/hr to ensure near-equilibrium con-
ditions for pesticide adsorption during
flow. The total volume of water held in
the soil column was gravimetrically deter-
mined at the end of each displacement by
extruding the soil from the glass cylinder
and oven drying. The number of pore vol-
umes (V/V0) of solution displaced through
the soil column was calculated by dividing
the cumulative outflow volume (V) by total
water volume (Vo) in the soil column. Ef-
fluent pesticide concentrations were ex-
pressed as relative concentrations (C/CO),
where C and C0 are, respectively, effluent
and input concentration. Plots of C/C0
versus V/V0 are referred to as breakthrough
curves (BTC).
Pesticide Degradation and Soil Respiration
Each soil was initially wet to a soil-
water content corresponding to 30% of the
0.33 bar soil-water tension and incubated
for one week at 25 C. Following incuba-
tion, each soil was then mixed thoroughly
with a specific quantity of pesticide and
sufficient water was added to bring the
soil up to 0.33 bar soil-water tension.
For pesticide degradation and soil respira-
tion experiments, 100 gms (oven-dry basis)
of each pesticide-treated soil was placed
in a 250 ml Erlenmeyer flask. Special
care was then taken when mixing the pesti-
cide with the soil to ensure a uniform dis-
tribution of the pesticide within the soil.
Technical grade and formulated materials
of each pesticide were used. The flasks
were then connected to a plexiglass mani-
fold and COa-free water-saturated air
passed through the manifold into the flasks
at a flow rate of 10 .ml/min per flask.
For pesticides with a high vapor pressure
the air leaving the flask was passed
through 40 ml of ethylene glycol to absorb
the volatilized pesticide. The air leaving
each flask was also bubbled through a KOH
solution (0.1 - 0.2N) to absorb the evolved
C02. At frequent intervals, the KOH solu-
tions were replaced with fresh KOH solu-
tions and the CO? concentrations determined
by titration. 1^C-C02 activity in the KOH
solution was determined by liquid scintil-
lation counting.
Soil Microbial Activity
For the microbial enumeration exper-
236
-------
iments, 250 g of each pesticide-treated
soil was placed in a 500 ml Erlenmeyer
flask. Ten gram soil samples were
withdrawn weekly from each flask and bac-
terial, fungal, and actinomycete popula-
tions determined using a dilution plate
count method (18-20). The number of micro-
organisms were expressed as colony forming
units (cfu) per gram of soil (oven-dry
basis). Bacteria populations were deter-
mined using a TGY medium consisting of
5 g tryptone, 5 g of glucose, 4 g of yeast
extract, 18 g of agar and 1000 mis of
distilled water. The pH of the medium
after autoclaving was 7.0. Bacterial col-
onies were counted after 48 to 52 hours at
28 C. Fungal populations were counted
using a Rose Bengal-Sterptonycin agar.
The pH of the fungal medium after autoclav-
ing was 6.0. Fungal populations were de-
termined after 60-72 hours at 28 C. Acty-
nomycetes were plated on starch-casein
agar (SC agar) supplemented with antibio-
tics cycloheximide 50 yg/ml and nystatin
50 yg/ml. The pH of the actynomycete
medium after autoclaving was 7.2. The
actynomycete colonies were counted after
12 to 14 days at 25 C.
RESULTS AND DISCUSSION
Adsorption Experiments
Equilibrium adsorption isotherms were
determined for each soil-pesticide combi-
nation by measuring pesticide adsorption at
five or more concentrations ranging from
zero to the pesticide's aqueous solubility
limit (14). An example of the type of data
obtained from these experiments is present-
ed in Figure 1. All adsorption isotherms
measured were described by the Freundlich
equation (S = KC^); where K and N are con-
stants, and S and C are, respectively, ad-
sorbed (yg/g of soil) and solution (yg/ml)
pesticide concentrations. The values of
the Freundlich adsorption constants K and
N for a given soil-pesticide combination
were obtained using a least-squares fit.
These values for each pesticide-soil sys-
tem studied are summarized in Table 3. Be-
cause soil organic carbon content generally
correlates well with pesticide adsorption,
the use of an adsorption partition coeffi-
cient based upon organic carbon content
rather than total soil mass was proposed by
Hamaker and Thompson (23). Using this pro-
cedure, the amount of pesticide adsorbed
was expressed as yg/g of organic carbon
10"
: 2.4-D Amlne
u1°3
z
O
u
o
U 2
m 10
o:
o
CO
Q
10
SOLUTION
10J
CONC.
10" 10=
(.urnol. /I)
Figure 1.
Adsorption isotherm for 2,4-D
amine on Webster and Cecil
soils. Parameters associated
with each line are coefficients
for Freundlich equation.
and the Freundlich constant (Kgc) f°r each
adsorption isotherm was computed,, These
values are presented in Table 3U It is
apparent that the KQC values for a given
pesticide are less variable among the four
soils studied than are the K values uncor-
rected for organic carbon. These results
are in general agreement with the observa-
tions of Hamaker (24) where the KQC values
for a given pesticide were nearly independ-
ent of soil type. It should be recognized,
however, that other factors such as soil
pH, clay content and cation exchange capa-
city may also play a significant role in
determining pesticide adsorption by soils
(7). On the basis of the KOC values listed
in Table 3, the extent of pesticide adsorp-
tion in soils was in the order of terbacil
< 2,4-D amine < atrazine < methyl para-
thionu
Two important conclusions can be made
based on the data presented in Table 3.
First, the fact that the Freundlich equa-
tion describes all pesticide adsorption
isotherms studied over a wide concentra-
tion range illustrates that adsorption
sites were not saturated at any concentra-
tion considered in the study. The amount
of pesticide adsorbed by the soil continued
to increase, at a decreasing rate, with
each increase in solution concentration.
This behavior may not be true for other
pesticide adsorbants (25). Second, con-
trary to a frequent assumption, pesticide
adsorption isotherms in most cases are
237
-------
TABLE 3. Freundlich constants calculated from equilibrium adsorption isotherms for
various soil-pesticide combinations.
PESTICIDE
2,4-D Amine
SOIL
TYPE
Webster
Cecil
Glendale
Eustis
Average ± % CV*
K
4.62
0.65
—
0.76
2.01 ± 112
N
0.70
0.83
—
0.73
0.75 ± 9
Koc
119.4
72.2
_ —
135.7
109.1 ± 30
Atrazine
Webster
Cecil
Glendale
Eustis
Average ± % CV
6.03
0.89
0.62
0.62
2.04 ± 131
0.73
1.04
0.93
0.79
0.87 ± 16
155.8
98.9
124.0
110.7
122.3 ± 20
Terbacil
Webster
Cecil
Glendale
Eustis
Average ± 7, CV
2.46
0.38
0.38
0.12
0.83 ± 130
0.88
0.99
0.93
0.88
0.92 ± 6
63.6
42.2
76.0
21.4
50.8 ± 47
Methyl
Parathlon
Webster
Cecil
Glendale
Eustis
Average ± % CV
13.39
3.95
3.57
2.72
5.91 ± 85
0.75
0.85
0.61
0.86
0.77 ± 15
346.0
438.6
714.5
486.4
496.4 ± 32
* CV is the coefficient of variation, % CV = (standard deviation/average) x 100.
non-linear, that is, N is less than one
(Table 3). Linear adsorption isotherms
have generally been accepted for low pest-
icide concentrations because it simplified
computer simulation modeling (26-28). The
significance of non-linear adsorption iso-
therms with regard to pesticide mobility
in soils is discussed in the following
section.
Column Displacement Experiments
Effluent breakthrough curves were mea-
sured (14) for 2,4-D amine at two input
concentrations (Co = 50 and 5,000 yg/ml)
and tritiated water (3H20) using water sat-
urated Webster and Eustis soil columns
(Figures 2 and 3). Tritiated water repre-
sents a non-adsorbed solute and serves as
a reference for adsorbed solutes (2,4-D
amine). A shift of the ETC to the right of
3HaO ETC is due to an adsorption-induced
retardation. A greater shift of the ETC to
the right indicates increased adsorption
and a decreased mobility. It is apparent
from the data presented in Figures 2 and
3 that the mobility of 2,4-D amine signi-
ficantly increased as the input concentra-
tion of the pesticide increased from 50 to
5,000 ug/ml. Note that the 5,000 ug/ml
2,4-D amine input concentration was nearly
as mobile as the 3H20 in the Webster soil
and as mobile in the Eustis soil. The
difference in mobility between the low and
high 2,4-D amine concentrations was more
pronounced in the Webster than in the
Eustis soil. These column results are
consistent with calculated retardation
terms (9) obtained from the adsorption
data presented in Table 3.
Breakthrough curves for the displace-
238
-------
3
o
~0.8
u
z
o
u Q6
> 0.4
4
H 0.2
WEBSTER SOIL
2,4-D Amine
Figure 2.
) ) I 1
PORE VOLUMES ( V: V )
Effluent breakthrough curves
for 2,4-D amine (Co = 50 and
5,000 ug/ml) and for tritiated
water displaced through water
saturated Webster soil column.
(J-
o
EUSTIS SOIL
ATRAZINE
o .' C.,5>jg/
'
PORE
6 8 1O
VOLUMES (V/V. )
Figure 4.
Effluent breakthrough curves
for atrazine (C0 = 5 and 5,000
yg/ml) and for tritiated water
displaced through water satu-
rated Eustis soil column.
_l
LU
a.
EUSTIS SOIL
2,4-D Amine
,"••'
O 2 4 6 8 10
PORE VOLUMES (V/V.)
Figure 3. Effluent breakthrough curves
for 2,4-D amine (Co = 50 and
5,000 Ug/ml) and for tritiated
water displaced through water
saturated Eustis soil column.
ment of atrazine through a water saturated
Eustis soil column at two input concentra-
tions (C0 = 5 and 50 vig/ml) and 3H20 are
shown in Figure 4. The trend of increased
mobility for higher atrazine solution con-
centrations is evident. However, the dif-
ferences in atrazine mobility (Figure 4)
between the two concentrations were not as
large as those for 2,4-D amine (Figures 2
and 3) for a larger concentration differ-
ence. In this study, atrazine concentra-
tions varied only by ten-fold while for
the 2,4-D amine, concentrations varied by
100-fold. Also, the adsorption isotherm
for 2,4-D amine in Webster soil was more
non-linear than that for the Eustis soil-
atrazine system (Table 3).
Pesticide Degradation and Soil Respiration
Total CC>2 evolution is generally a
good indicator of soil microbial activity.
This procedure was used by Stojanovic et
al. (13) to study soil systems receiving
high pesticide concentrations and from this
pesticide degradation estimated. Because
^002 evolution from uniformly ring-labeled
14C-2,4-D provides a measurement of pesti-
cide degradation, the accuracy of using
total C02 evolution in calculating 2,4-D
degradation can be compared with results
obtained from ll|C02 evolution.
The rates of CC>2 evolution from the
Webster soil receiving various concentra-
tions of technical grade and formulated
2,4-D are given in Figures 5 and 6. A com-
mon characteristic of the C02 evolution
rate at 5,000 and 20,000 ppm of 2,4-D was
the presence of two response peaks. The
30
I
§20
Oppm
SO ppm
SOOppm
5.000ppm
20,000ppm
20
40
Days
60
60
Figure 5.
C02 evolution rate from the
Webster soil receiving various
concentrations of technical
grade 2,4-D.
239
-------
Oppm
o o 50 ppm
SOOppm
5,000 ppm
20,000ppm
80
Figure 6.
C02 evolution rate from the
Webster soil receiving various
concentrations of formulated
2,4-D.
initial increase in C02 evolution rate was
followed by a decline and a second increase.
This was observed for technical grade and
formulated 2,4-D (Figures 5 and 6). The
CO2 carbon produced prior to the appearance
of the second peak was not from the 2,4-D-
carbon because very little ll*C02 was mea-
sured during this period. For formulated
2,4-D, the C02-C in the first peak appeared
to come from the oxidation of the formula-
tion ingredients including dimethyl amine
(Figure 6); and for technical grade 2,4-D,
the C02-C probably originate from impuri-
ties and soil organic matter (Figure 5).
The appearance of the second peak coincided
with the occurrence of 2,4-D degradation
(lkC02 evolution). Assuming all C02-C was
evolved from 2,4-D-C, the extent of degrad-
ation during the entire experimental period
(80 days) and during the second peak period
was calculated. These results clearly il-
lustrate that the majority of the C02-C
evolved during the second peak period was
primarily from the carbon in the 2,4-D.
Carbon dioxide evolution rates were gener-
ally higher from the formulated 2,4-D treat-
ed soil than from the technical grade treat-
ed soil. Therefore, microbial activity in
the soil receiving formulated material was
higher.
Because 2,4-D degradation was deter-
mined by measuring the evolution of 14C02
from uniformly ring-labeled material, it
can be assumed that the 2,4-D was degraded
to its final oxidation products, C02, H20,
and Cl~. At concentrations of 50 and 500
ppm in the Webster silty clay loam, both
forms (technical grade and formulated) of
2,4-D were mineralized rapidly to C02,
H20, and Cl~ within 40 days of incubation
(Figures 7 and 8). At 5,000 ppm, both
forms were degraded, but exhibited lag per-
iods of 10 to 19 days for the technical and
formulated 2,4-D material, respectively.
100
tu 40
o—o SO ppm
•-•500 ppm
a—A 5,000 ppm
»-* 20,000 ppm
20
40
DAYS
60
80
Figure 7.
Percent 14C02 evolved from the
technical grade 2,4-D treated
Webster soil fortified with
14C-labeled 2,4-D.
too
o oSOppm
• , SOOppm
SPOOppm
ZOjOOOppm
Figure 8.
Percent llfCO evolved from the
formulated 2,4-D treated Web-
ster soil fortified with 14C-
labeled 2,4-D.
240
-------
No significant degradation occurred before
50 days for 20,000 ppm, and a total of 11,5%
and 21.5% of the technical grade and for-
mulated material were degraded after 80
days of incubation (Figures 7 and 8),
The C02 evolution rate from the un-
treated Terra Ceia muck soil was approxi-
mately four times higher than that from
the Webster soil and eleven times higher
than that from the Cecil soil. Total C02
evolution from the Terra Ceia muck soil
treated with 5,000 and 20,000 ppm of for-
mulated 2,4-D was high. Some of the C02-C
produced from 5,000 ppm of formulated 2,4-D
in the drganic soil after seven days was
from the 2,4-D-C because significant de-
gradation occurred thereafter. For the
20,000 ppm of formulated 2,4-D treatment,
total C02 evolution was enhanced after 36
days. This was concurrent with the de-
gradation of 2,4-D. The C02 evolution
rate from the Terra Ceia muck treated with
5,000 ppm of the technical grade 2,4-D in-
creased after 20 days of incubation and co-
incided with a period of rapid 2,4-D de-
gradation. Total C02 evolution was in-
hibited throughout the 80 days for 20,000
ppm of technical grade 2,4-D. Similar to
the Webster soil, comparable 2,4-D degrad-
ation results were calculated where total
COz evolution during the second peak per-
iod rather than the entire experimental
period was used.
Unlike the Webster and Terra Ceia muck
soils, total C02 evolution was inhibited in
the Cecil soil which received 5,000 and
20,000 ppm of technical grade 2,4-D as well
as 20,000 ppm of formulated 2,4-D. A small
stimulation was noted in total C02 evolu-
tion from the Cecil soil treated with 5,000
ppm of formulated 2,4-D, but no major de-
gradation of 2,4-D occurred.
Soil respiration rates from Cecil and
Webster soils receiving 10 ppm of techni-
cal grade and formulated trifluralin were
not significantly different from non-treat-
ed soils except during the first week of
incubation (Figures 9 and 10). Only the
Webster soil data will be presented; how-
ever, soil respiration data for other soils
and pesticides are discussed by Ou, et alt
(17-20). Carbon dioxide evolution from the
Webster soil receiving 1,000 and 20,000 ppm
of technical grade trifluralin were not en-
hanced significantly except during the ini-
tial incubation period (Figure 9). Soil
respiration rates from the Webster soil
3 8
O
tn
O)
86
E
V4
c\j
O
u
° — o
Tnfiuralm
(Technical)
Webster
o ppm
1O ppm
1,000 ppm
20.00O ppm
20
40
Days
60
Figure 9.
C02 evolution rate from the
Webster soil receiving various
concentrations of technical
grade trifluralin.
receiving 1,000 ppm of formulated triflu*-
ralin were enhanced initially, but were
not different from the untreated soil dur-
ing a major portion of the incubation per-
iod. For the 20,000 ppm of formulated tri-
fluralin treatment, C02 evolution was sti-
mulated whereas C02 evolution for this con-
centration in the Cecil soil was below the
untreated soil.
Soil ilicrobial Activity
Bacterial populations in the 10 ppm
trifluralin treated (technical grade and
formulated) Webster soil were not signifi-
cantly different from the untreated soil.
This is in agreement with the soil respira-
tion data presented in Figures 9 and 10.
However, bacterial populations in the Web-
ster soil receiving 1,000 ppm of formula-
ted trifluralin were consistently higher
than the untreated soil with the greater
stimulation occurring in the Cecil rather
than the Webster soil. For 20,000 ppra of
technical grade trifluralin, the bacterial
populations in the Cecil soil were stimulated
but not in the Webster soil. The bacterial
populations in the Cecil soil receiving
241
-------
s*
o
»—*
U
CVJ
O
U
B
Trifluralin
(Formulated)
Webster
o—o 0 ppm
•—• 10 ppm
o—a 1,000 ppm
20,000 ppm
_L
20
40
Days
60
Figure 10. COz evolution rate from the
Webster soil receiving various
concentrations of formulated
trifluralin.
20,000 ppm of formulated trifluralin were
initially inhibited, but were enhanced sig-
nificantly after 2 weeks of incubation;
whereas bacterial populations in the Web-
ster soil for the same treatment were stim-
ulated during the first 18 hours and con-
tinued to increase throughout the incuba-
tion period.
Fungal populations in the Cecil and
Webster soils treated with 10 ppm of tech-
nical grade and formulated trifluralin were
generally not significantly different from
the untreated soil. For the 1,000 ppm
treatments, fungal populations in the Cecil
soil receiving technical grade and in the
Webster soil receiving formulated triflura-
lin were not significantly affected, where-
as the Cecil soil treated with the formula-
ted material and the Webster soil with the
technical grade material showed a signifi-
cant effect at the 0.01 and 0.05 levels,
respectively. The fungal populations were
significantly reduced in both soils at
20,000 ppm of formulated trifluralin; where-
as fungal populations in the Webster soil
receiving the technical grade material were
stimulated during the first four weeks.
Actinomycete populations in Cecil and
Webster soils receiving 10, 1,000 and
20,000 ppm of trifluralin (technical grade
or formulation) were not, in general, sig-
nificantly different from the untreated
soils.
CONCLUSIONS AND SUMMARY
The increased pesticide mobility at
high concentrations limits the usefulness
of the present low concentration data base
for developing safe management practices
for pesticide disposal in soils. However,
underestimation of pesticide mobility by
assuming linear adsorption isotherms may
be less severe for pesticides with low
aqueous solutilities. Ou et al. (17, 18)
show that for high pesticide loading rates,
up to 20,000 ug 2,4-D/g soil, there was a
significant decrease in pesticide degrada-
tion rate with an accompanying depression
of total microbial activity in the soil.
Thus, due to rapid leaching and minimal
microbial decomposition at high 2,4-D con-
centrations, the potential for groundwater
pollution exists. The results presented
in this manuscript may also be applicable
when describing the behavior of other haz-
ardous wastes applied to soils.
Microbial populations (especially bac-
teria) in the Webster soil were nearly
twice that in the Cecil soil. The Webster
soil had a higher organic matter and clay
content than the Cecil soil as well as a
neutral pH. High organic matter and clay
contents provide larger surface area which
enhance pesticide adsorption and reduces
the toxicity of the applied pesticide.
The results presented here are not consist-
ent with those of Stojanovic et al (9);
however, this should not be surprising
owing to differences observed in the be-
havior of different soils and pesticide
forms. The results presented in this study
suggest that disposal of trifluralin and at-
razinewastes in soils may not drastically
reduce soil microbial activity, and in fact
may enhance it under some situations. How-
ever, long term effects under field condi-
tions remain unanswered. The fact that the
microbial populations were stimulated by
pesticide application, as inferred from
enhanced soil respiration, does not neces-
sarily mean that pesticide degradation has
occurred. Certain precautions should be
made in evaluating pesticide degradation
data based on soil respiration (i.e., total
C0£ evolution) as the only measurement of
pesticide degradation.
242
-------
LITERATURE CITED
1. van Everdingen, R. 0. and Freeze, R.A.,
"Subsurface Disposal of Waste in Can-
ada," Inland Waters Branch, Dept. ot
Environment, Ottawa, Canada, Tech.
Bull. No. 49, 1971.
2. Schomaker, N. B., "Current Research on
Land Disposal of Hazardous Wastes,"
Residue Management by Land Disposal,
Fuller, W. H. (ed.), Proc. of the
Hazardous Waste Research Symp., EPA-
600/9-76-015, July, 1976.
3. Fields, T., Jr. and Lindsey, A. W.,
"Landfill Disposal of Hazardous Wastes:
A Review of Literature and Known Ap-
proaches," U. S. Environ. Prot. Agen-
cy Publication S. W. 15, 1975.
4. Lindsey, A. W., Farb, D., and Sanjour,
W., "OSWMP Chemical Waste Landfill
and Related Projects," Residue Manage-
ment by Land Disposal, Fuller, W. H.
(ed.), Proc. of the Hazardous Waste
Research Symp., EPA-600/9-76-015,
July, 1976.
5. Rouston, R. C., and Wildung, R. E.,
"Ultimate Disposal of Wastes to Soil,"
Water, Cecil, L. K. (ed.), Chem. Bug.
Prog. Series 64.97, 1969.
6. Atkins, P. R., "The Pesticide Manu-
facturing Industry - Current Waste
Treatment and Disposal Practices,"
Water Pollution Control Research
Series, 12020 FYE 61/72, 1972.
7. Bailey, G. W. and White, J. L., "Fac-
tors Influencing Adsorption, Desorp-
tion, and Movement of Pesticides in
Soil," Residue Reviews, v. 32, pp 29-
92, 1970.
8. Sanborn, J. R., Fracis, B. F., and
Metcalf, R. L., "The Degradation of
Selected Pesticides in Soil: A Review
of the Published Literature," EPA-600/
9-77-022, 1977.
9. Davidson, J. M., Ou, L. T., and Rao,
P. S. C., "Behavior of High Pesticide
Concentrations in Soil-Water Systems,"
Residue Management by Land Disposal,
Fuller, W. H. (ed.), Proc. of the
Hazardous Waste Research Symp., EPA-
600/9-76-015, July, 1976.
10. Wolfe, H. R., Stoiff, D. C., Armstrong,
J. F., and Comer, S. W., "Persistence
of parathion in soil," Bull. Environ.
Contamin. Toxicol., v. 10, pp 1-9,
1973.
11. Clay, D. V., "The persistence and pen-
etration of large doses of simazine in
uncropped soil," Weed Res., v. 13,
pp 42-50, 1973.
12. Grossbard, E., and Davis, H. A., "Spe-
cific microbial responses to herbi-
cides," Weed Res., v. 16, pp 163-169,
1976.
13. Stojanovic, B. J., Kennedy, M. V. and
Shuman, Jr., F. L. "Edaphic aspects
of the disposal of unused pesticides,
pesticide wastes, and pesticide con-
tainers,, J. Environ,, Quality, v. 1,
pp 54-62, 1972.
14. Rao, P. Su C., and Davidson, J. M.,
"Adsorption and Movement of Selected
Pesticides at High Concentrations in
Soils," Water Research, 1978 (sub-
mitted) .
15. Rao, P. S. C., Davidson, J. M., Jes-
sup, R. E., and Selim, H. M., "Evalu-
ation of Conceptual Models for Des-
cribing Kinetics of Pesticide Adsorp-
tion-Desorption During Steady-Flow in
Soils," Soil Sci. Soc. Amer. Jour.,
v. 42, 1978 (submitted).
16. Rao, P. S. C., and Davidson, J. M.,
"Non-Equilibrium Conditions for Sol-
ute Transport in Soils: Flow Inter-
ruption Experiments," Soil Sci., 1978
(submitted).
17. Ou, L.-T., Rothwell, D. F., Wheeler,
W. B., and Davidson, J. M., "The ef-
fect of high 2,4-D concentrations on
degradation and carbon dioxide evolu-
tion in soils," J. Environ. Quality,
v. 7, 1978.
18. Ou, L.-T., Davidson, J. M., and Roth-
well, D. F., "Response of soil micro-
flora to high 2,4-D applications,"
Soil Biol. Biodiem., 1978 (in press).
19. Ou, L.-T., Davidson, J. M., and Roth-
well, "High Concentration Effect of
Herbicides Trifluralin and Atrazine on
Microbial Biota and Soil Respiration,"
Appl., Environ. Micro., 1978 (submitted)
243
-------
20. Ou, L.-T. and Davidson, J. M., "High
Concentration Effect of Methyl Para-
thion on Degradation and C02 Evolution
in Soils," Bull. Environ. Contain. Tox-
icol., 1978 (submitted).
21. Wheeler, W. B., Stratton, G. D., Twil-
ly, R. R., Ou, L.-T., Carlson, D. A.,
and Davidson, J. MM, "Trifluralin De-
gradation and Binding in Soils," Jour.
Agr. Food Chem., 1978 (submitted).
22. Davidson, J. M., Rieck, C. E., and
Santelman, P. W., "Influence of Water
Flux and Porous Materials on the Move-
ment of Selected Herbicides," Soil
Sci. Soc. Aiaer. Proc., v. 32, pp 629-
633, 1968.
23. Hamaker, J. W., and Thompson, J. M.,
"Organic Chemicals in the Soil Env-
ironment," Goring, C.A.I., and Hama-
ker, J. W. (eds.), Marcel Dekker, Inc.
N. Y., v. 1, pp 49-143, 1972.
24. Hamaker, J. W., "Environmental Dynam-
ics of Pesticides," The Interpreta-
tion of Soil Leaching Experiments,
Haque, R., and Freed, V. H. (eds.),
Plenum Press, N. Y., pp. 115-131,
1975.
25. Weber, W. J. and Usinowicz, P. J., Ad-
sorption from Aqueous Solution. Tech-
nical Publication Research Project
17020EPF, U. S. Environ. Prot. Agency,
Cincinnati, Ohio, 1973.
26. Kay, B. D., and Elrick, D. E., "Ad-
sorption and Movement of Lindane in
Soils," Soil Sci., v. 104, pp 314-
322, 1967o
27. Hugenberger, F., Letey, J., and Far-
mer, W. F., "Observed and Calculated
Distribution of Lindane in Soil' Col-
umns as Influenced by Water Movement,"
Soil Sci. Soc. Amer. Proc., v. 36,
pp 544-548, 1972.
28. Davidson, J. M., and Chang, R. K.,
"Transport of Picloram in Relation to
Soil Physical Conditions and Pore-
Water Velocity," Soil Sci. Soc. Amer.
Proc., v. 36, PP 257-261, 1972.
244
-------
MODELLING CONTAMINANT ATTENUATION IN SOIL:
MICROBIAL DECOMPOSITION OF ORGANIC MATTER
S. Soyupak, G.J. Farquhar and J.F. Sykes
Civil Engineering, University of Waterloo,
Waterloo, Ontario, Canada
ABSTRACT
An investigation was undertaken to examine the microbial decomposition of organic
matter in soil. Sealed saturated soil columns received a continuous upflow of laboratory
lysimeter leachate at a Darcy velocity of 1.54 cm/day. After 70 days of operation, the
influent COD of 5700 mg/X. had been reduced by 90% over the 41 cm soil column length.
Methane production was vigorous. A carbon balance across the column showed 94% recovery
of influent carbon. Monod-type equations combined with expressions for advection and
dispersion were used in a model to simulate column effluent COD concentrations. Biological
parameters taken from reports external to this study yielded good simulation under the
operational conditions used. Guidelines were provided for applying the results to other
situations including the disposal of industrial wastes in soil.
INTRODUCTION
The microbial decomposition of organic
matter is a subject which has received little
attention in research on contaminant move-
ment in soil. This is peculiar in view of
the high concentration of organic matter
frequently reported for landfill leachates
and certain industrial liquids released to
soil/groundwater systems for disposal.
The concentration of organic matter
with respect to water pollution has tradi-
tionally been determined by non-specific
measurements such as biochemical oxygen
demand (BOD), chemical oxygen demand (COD)
and total organic carbon (TOC). In some
cases more specific measurements for com-
ponents such as volatile acids and trace
organics have been used. An example of the
latter is polychlorinated biphenyls (PCB).
The concentration of organic matter in
landfill leachate varies from site to site
but in many cases is quite high. Concen-
trations of COD have been reported in excess
of 70,000 mg/X,(l) although this appears to
be near the upper limit for leachates.
Concentrations between 20,000 and 50,000
mg/Jl have been reported on several occasions
(1,2,3,4). These reports have dealt mostly
with simulated landfill conditions and/or
rates of infiltration well above field
rates. This has usually been accompanied
by acid pH in the range of 5 to 7, indica-
ting that the anaerobic processes occur-
ring in the landfill have been weighted
heavily on the side of fermentation with
volatile acid release well in excess of
methane production (5,6).
For in situ landfills receiving nat-
ural precipitation, COD's tend to range
up to 20,000 mg/X, depending on site age
with pH slightly acidic to neutral
(4,7,8). Methane production is often
active at these sites.
As landfills age, leachate COD's tend
to reduce. However, the work of Hughes
et al.(8) showed that leachate COD's in
the 1000"s were common at Northeastern
Illinois landfills ten to fifteen years
after placement. Consequently, the land-
fill represents a significance source of
organic contaminants, for many years.
In contrast however, at many landfill
sites, high concentrations of organic
matter tend not to appear in adjacent
groundwaters. It is clear therefore,
that other attenuating processes work to
245
-------
reduce the concentration of leachate orga-
nics, possible processes being dispersion,
sorption onto soil and microbial decom-
position.
Dispersion during migration reduces
leachate component concentrations, the
extent being represented by changes in
chloride ion (Cl) concentrations down flow.
However, reductions in organic matter
concentrations exceed those observed for
Cl (7,8) and thus other mechanisms are
operative.
From the limited data available
(9,10), the sorption of leachate organic
matter onto soil is not extensive although
it does vary with organic matter and soil
type.
While little evidence is available to
support its existance, the microbial
decomposition of leachate and other organic
matter in soil appears to be a significant
attenuating force. Consequently, this
work was undertaken to investigate:
1. the extent and rate of organic matter
reduction and factors affecting them
2. the types of organic matter reduced
3. the production of gaseous by-products
and
4. the development of models for simula-
tion.
TABLE 1
SOIL COLUMN PROPERTIES
soil depth
soil surface area
DIG
40.6 cm
76. 5 cm2
0.012 mm
1.73
permeability :
(initial)
(120 days)
porosity
organic content
cation exchange
capacity
1.3 xlO~* cm /sec
I.I xlO cm/ sec
0.35
0. 66 •/. i-y weight
2.03 meq /IOO gm
EXPERIMENTATION
Two parallel experiments, R-2 and R-3
were conducted on a laboratory scale using
packed soil columns operated anaerobically
under continuous saturated flow conditions.
The properties of the silty sand soil in
place are shown in Table 1 for column R-2.
Permeability measurements based on constant
head experiments are given before and after
experimentation and show only a slight
reduction resulting from 120 days of micro-
bial activity.
A schematic representation of the
columns is shown in Figure 1. Upflow
conditions under positive head ensured
saturated flow. Liquid addition under
constant head was continued throughout the
experiments. Anaerobiosis was achieved by
sealing the system, maintaining the gases
as produced above all liquid surfaces and
flushing appurtenances with nitrogen after
each feed addition.
FIGURE 1
SCHEMATIC DIAGRAM OF SOIL COLUMNS
LIQUID ADDITION
The liquid used was simulated landfill
leachate produced through moisture addition
to residential solid waste in a laboratory
lysimeter (2). Its properties are summar-
ized in Table 2. It was characterized by
high concentrations of organic matter, in
particular free volatile fatty acids and
an acidic pH which showed it to be typical
of lysimeter leachate (1,3). The free
volatile fatty acids detected consisted
mainly of acetic, propronic and n-butyroc
acids and accounted for 54% of the TOG as
compared to 61% reported by Chian and
DeWalle(l). The balance of the TOC was not
identified but was assumed to consist in
part of additional fatty acids, nitrogen
containing compounds (ORG-N = 147 mg/Jl) in
246
-------
addition to fulvic and humic-like materials
(1).
TABLE 2
LEACHATE PROPERTIES
PH
COO
TOC
free volatile
fatty acids .
acetic
propionic
isobulync
n-bulyric
isovalenc
n-valeric
3-
P04 -P
NHj -N
OR6 -N
cr
5.7
39000 mg/l
II 600 mg/l
3810 mg/l
1640 mg/t
590 mg/J
4880 mg/<
870 mq/t
340 mg/l
mg/l
12030
( 54 % of TOC)
4.7 mq/t
1134 mg/{
147 mg/J
895 mg/1
The stoichiometry and minimum nutri-
tional requirements of methanogenic bac-
teria are shown in Table 3 (5,11) along
with information on COD: total P ratios
for various leachates. The requirement
for P is modest at a COD: P ratio of
1250:1 since much more of the organic
matter is devoted to electron reception
than is the case for aerobic populations.
Notwithstanding however, the leachate of
this study and of certain others as well,
are shown to be P deficient.
In addition, the P requirements above
relate to minimum growth. They do not
address the need for additional P to
increase growth rates above the minimum
level.
In order to encourage growth in this
study, conditions were improved, through
the addition of 40 mg/£ P to the leachate
in the form of phosphate ion (Na2 H PO^ K
H£ PO^). Further encouragement was given
because the P addition also increased
the leachate pH to near neutrality, the
region most suitable for CH^ production.
The buffering of acidic pH's is not
uncommon in soils but the addition of P
is. It was justified however, by the need
TABLE 3
NUTRIENT REQUIREMENTS OF METHANOGENIC
BACTERIA (MCCARTY, O'ROURKE)
Cell Formula
Minimum Yield
Coefficient
COO
COO
N : P
P
* C62.3 HII2.S °37.5 NI15 P
« 0.04 mg VSS/mg COO
' 100 : 0.44 : 0.08
» I25O : I
P AVAILABILITY
Leachate Source
Oupagi LW9B (&);
Mission Canyon( iZ) ;
Baone County ( IZ1-.
Sonoma County ( 4 ) ;
Illinois ( | );
Waterloo ( £ );
landfill, natural H20
" i " "
test cell, " "
landfill, high HZ0
lysimeter
lysimeter
COD : P
6670. 1
468O: 1
630: 1
3420: 1
710- 2250: 1
83CO-. 1
to detect methanogenic activity in the
soil within a reasonable time. After P
addition, the leachate was diluted to
yield a COD - 5700 mg/l to be more rep-
resentative of in situ leachates (7,8).
Prior to the column experiments two
completely mixed reactors (CMR) were
operated anaerobically with daily additions
of the leachate described above. After
four weeks, methane production had been
established with substantial COD reduction.
The biological culture so developed was
used to seed the soil/columns uniformly
with depth at a level of 50 mg/i as
volatile suspended solids (VSS). This was
designed to reduce lag time for methano-
genic culture development in the columns.
Leachate was added to the columns at
a Darcy velocity (q) of 1.54 cm/day.
Measurements of COD, TOD, Cl, P, pH,
nitrogen (N) and gas production were per-
formed on the column influents and effluents
to assess microbial activity.
RESULTS
The results obtained from the parallel
column experiments are presented below.
247
-------
Hydraulic Properties of Columns
Figure 2 shows the normalized tracer
output curve for column R-2. The data
points represent the response to a step
function addition of Cl introduced on the
90th day of operation with q = 1.54 cm/day.
The curve is typical of saturated flow
through porous media.
A least squares analysis was used to
fit the equation
3C
where :
1.
n = porosity
x = axial flow direction, (L)
D = dispersion coefficient in
the x direction, (L2
interstitial velocity,
nv = Darcy velocity,
tracer concentration,
q
C
(L 1
(L t-J)
(M I"3)
to the fjracer data. The fit achieved is
represented by the solid curve for which
the values of v, D and n are shown. The
fitted value of D*. is less than the pre-
dicted by the equation of Van Genuchten
and Wierenga (13);
R-2 and R-3 under conditions of parallel
operation. While the data are somewhat
scattered, they show that the removal of
organic matter in the columns increased
from nil initially to near 90% after 70
days of operation. This was considered to
be excellent removal across a soil column
distance of 41 cm. In spite of the column
having been seeded with 50 rag/a VSS, a lag
of approximately 10 days was experienced
prior to the initiation of CH^ production.
This is not uncommon for microorganisms
placed in new environments. An analysis
of gas production and COD removal showed
that the initial active microbial popula-
tion (X(t=10,x)) was approximately 20 mg/fc
VSS. This indicated that some micro-
organisms had died off or become inactive
during the lag period.
After column R-3 was taken out of
service, attempts were made to measure the
concentration of microorganisms within the
soil. The soil was removed in sections
and analysed immediately for VSS. However
the data from these measurements were too
scattered to permit interpretation.
Gas Production
1.0
o.a
0.*
O.4
o,
THEORETICAL
STEP INPUT
A
~ • TRACER (CHLORIDE /
ION) CONCENTRATION /
• ^S
fill
J*
• • O.I63 cm/lit
D • 1.021 • cn2/hr
• • 0 35
i l l l l 1
0 0.2 0.4 O.« 08
MMENSIONLESS RESIDENCE TIME 10)
1.0 12 1.4 1.6
EXPERIMENT TIME
THEORETICAL RESIDENCE TIME
FIGURE 2
TRACER BREAKTHROUGH CURVE
Organic Matter Destruction
Figure 3 shows the concentrations of
COD and TOC in the effluent from Columns
The data shown in Figures 4 and 5 show
the production of gaseous carbon as methane
(CH4-C) and carbon dioxide (C02-C) in the
soil columns. The scatter in the data was
attributed to flow variations and to the
movement of gas up through the soil prior
to release. The production of CH^ lagged
behind C02 production initially but increa-
sed to a maximum rate after 70 days. At
this time the volumetric gas composition
was 19% C02 and 81% 014.
A carbon balance performed on column
R-2 using averaged values for days 60
through 80 inclusive is presented in
Table 4. Total carbon (TC) input to the
column was measured at 221 mg/day. The
TC output was composed of carbon in solution
in the effluent = 62 mg/day, TC in the
gases produced = 137 mg/day and TC
incorporated into new microbial cells
synthesized = 8 mg/day for a total of
207 mg/day or 94% recovery. The TC incor-
porated into new cells was calculated from
estimates of cell yield and cell composition.
248
-------
60OO
40
TIME (days)
FIGURE 3
EFFLUENT COD AND TOG CONCENTRATIONS FOR R-2 AND R-3
— 140
a
» 120
E
iu 100
H-
<
IE
_ 80
P
0
a. 40
0
I 20
O
0 REACTOR 2
A REACTOR 3
o
0 °
o o o o
0 0
0
~ O O 0
goo °
a
° 49
A4
0
V
o
J1*
,-f2i < i i i i
2 30
5 _
§ |zo
Q ^
§ E
10
"
8 o
o REACTOR 2
A REACTOR 3
A 0 «
'/V A „«
0 «A ,» A .
O"OA oft «
-» .s^"
A
A
20 40 SO 80 100 120
TIME Uoyt)
20 40 80 80 100 120
TIME (day«)
FIGURE 4
CH PRODUCTION
FIGURE 5
C02 PRODUCTION
249
-------
TABLE 4
CARBON BALANCE IN COLUMN
(60th -80th day means)
CARBON IN
I. liquid in
TOC • 210 mg/doy
1C » II mg/doy
TC (IN;
CARBON OUT
I. liquid out
: 221 trig/day
TOC « 18 mg/doy
1C »44 mq/dqy
TC « 62 mg/dqy
2. gas out*
' CH4 « 119 mg/day
C02 * '18 mg/day
TC « 137 mg/doy
3. cell synthesis
cell TC ' 8 mg/dqy
TC (OUT)
207 mg/day
(94V. recovery)
volumetric ratio, COg : CH4 > 19 : 81
Model Development
The reduction of organic matter
expressed as COD was represented by Monod-
type equations given as
dS kSX
dt
dX
dt
K +S
YkSX
K +S
s
- b X
3.
4.
A computer programme was assembled
to solve these equations using the finite
element technique presented by Sykes (14).
Suitability of the Model
The equations presented above were
used to simulate the experimental results
of this study. However, the parameter
values chosen were based on data available
elsewhere. This was done to emplore the
applicability of the model in the absence
of site specific experiments. The bio-
logical parameter values for k, Kg, b and
Y were based on the work of O'Rourke (11)
involving the anaerobic decomposition of
free volatile fatty acids. The dispersion
coefficient DX was taken from the work
of Van Genuchten and Wierenga (13).
The parameter values used are shown in
Table 5.
TABLE 5
S (t « 0, x ) • 5 700 mo/J
'tag * I0 da*»
X (flO, x ) » 20 mg/J
n • 0.4
v » 4.1 cm/day
0 * 3v cm2/day (Van tenuchten and Witr«nga )
Y « 0.04
b • 0.015 days'1
K,» 2800 mg/J
K • 3.6 days"'
(0' Rourke )
(0.04)
(0.015 days'1 )
(2000-10000 mg/f)
(3.6-3.9 days'1)
where:
S • substrate concentration as
COD, (ML~3)
X «• microorganism concentration as
VSS, (ML'3)
k » maximum substrate utilization
rate, (t-1)
K * half velocity coefficient,
(ML'*)
Y - yield coefficient
b = microbial death rate, (t~l)
Equation 3 was combined with Equation
1, to yield;
n || - |-
3t 9x (_
- (nv)
- n
kSX
K +S
s
The values of S(0), v and n were
known from the operating conditions of
the column. Only the establishment of
X(0) required use of the experimental
results because of the microbial die-off
described previously. Thus, no fitting
of the equations to the data was involved.
The results of this effluent COD simula-
tion are shown as the broken line on
Figure 3. Considering the use of para-
meters developed elsewhere, the fit is
good. It supports the expectation that
available information and methodology can
be employed to obtain useful estimates of
organic matter decomposition in soil at
landfill sites.
250
-------
The solid line in Figure 3 shows an
Improved fit after manipulation of some of
the parameter values. Because of the large
number of parameters in the model, manipu-
lation must be based on an understanding of
the operative physical and biological phen-
omenon and the use of constrained nonlinear
parameter estimation techniques.
The model output tends to underpredict
the experimental data as the total COD
removal percentage increases. The higher
than expected effluent COD values have
been attributed to the presence of multiple
substrates in the leachate. The model
assumes a single limiting substrate react-
ion and thus, the use of a multiple sub-
strate model is required.
The multiple substrate concept is
supported by the fact that 54% of the
leachate TOG was in the form of free
volatile fatty acids. These are used
readily by the methanobacteria. The
balance of the organic matter would not be
readily useable by the methanobacteria
and hence would be converted more slowly
resulting in the "tailing-off" of the
effluent COD data.
Sensitivity Analysis
An analysis was performed to determine
the extent to which variations in Kg and
X (t » 0, x) would affect COD reduction in
the soil columns. The other biological
parameters, particularly Y and b were held
constant because they were characteristic
of free volatile fatty acids (11) and not
simply leachate type.
The results of the analysis are shown
in Figure 6. The values for Kg lie with-
in the range of 2000 mg/S, to 10,000 mg/fc
as identified by O'Rourke (11). The var-
iation is substantial and suggests a doub-
ling in the time required for COD removal.
A similar trend is shown for variations in
X(t -'0, x). In this case, the upper
limit of X(t - 0, x) - 20 mg/Jt VSS is the
microbial concentration determined for the
seeded column experiments. The lower
limit is more representative of conditions
which might exist leachate. It corresponds
to a microbial population on the order of
106 cells/100 mi. Although no data
enumerating the methanogenic bacteria in
leachate are available, populations of
other microbial types have been measured
Slot' 6OOO m«/»
y > 4.1 cm/Me
D • 3. cm'/t.c
O.O4
0015 doyt"'
36
2 560 O -
O
Z
O
O
a
o
u
100
4O 60
TIME (days)
FIGURE 6
MODEL SENSITIVITY TO K0 AND X(0)
S
in excess of 106 cells/100 in landfill
leachates (12).
The sensitivity analysis underscores
the need to ensure that the parameter
values be realistic for applications of
the model. It shows as well however, that
even for conservative estimates of para-
meter values, substantial COD removal will
occur within a one year period. In the
context of most landfill operations, this
is a significant observation.
ASSESSMENT
The scope of the work is limited and
thus caution must be exercised in applying
the results to new situations. Some
guidelines for use are identified below.
Substrate Type
The leachate used with its acidic pH,
its high volatile acid content and its P
deficiency was typical of many reported
elsewhere. However the organic composition
of certain leachates may deviate sub-
stantially from the composition used here.
In such cases an- individual assessment of
kinetic parameter values would be required.
An important extension of the work is
the study of the microbial decomposition of
251
-------
organic matter present in other liquids
discharged to the soil for disposal. Be-
yond the assessment of kinetic parameters
is the search for substances inhibitory to
methanogenic bacteria. Examples of
inhibitory concentrations for several
common ions are presented in Table 6.
While such concentrations are not usually
present in landfill leachates, they may be
in industrial liquids to the point where
CH^ production and thus COD removal would
be slowed or even halted.
P Deficiency
It has been shown that a minimum
COD:P ratio is necessary for the growth of
methanobacteria. Concentrations of P
beyond this minimum level may serve to
increase reaction rates the extent of which
must be evaluated. While the addition of
P to enhance microbial activity at
landfills seems difficult to justify, it
may not be in cases of liquid waste
disposal.
Soil Properties
The most important soil properties
with respect to the removal of organic
matter are acid neutralizing capacity,
sorptive capacity particularly for bio-
logically resistant compounds and fluid
flow related properties.
Not only does neutralization reduce
acidity, it may also initiate toxic metal
precipitation both of which will enhance
TABLE 6
INFLUENCE OF SEVERAL IONS ON METHANE FORMING BACTERIA
ION
CONCENTRATIONS WHICH ARE
Stimulatory Inhibitory
Moderately
Refer-
Strongly ence
sodium
potassium
calcium
magnesium
iron
aluminum
copper
zinc
nickel
chromium
ammonia
sulphide
(Na+)
8000
> 12000
> 8000
> 3000
> 2
> 2
> 2
to solubility
limit
> 3000
> 200
(5)
(5)
(5)
(5)
(5)
(5)
(15)
(15)
(15)
(15)
0-6)
252
-------
the growth of methane bacteria. Not all
soils possess this capacity 0-7) and, thus,
site specific measurements are required.
Although the sorption of leachate
organic matter onto soil is generally poor
(9,10) the sorption of specific compounds
may not be. An example is the extensive
sorption of the biologically resistant
compounds, PCB's 0-8,1ft). Further work
on sorption is required specifically with
hazardous organic matter.
With respect to microbial decomposi-
tion,' fluid flow is important in terms of
reaction time per unit distance. As the
rate of flow increases, the active micro-
bial zone will tend to expand. However,
the active zone in this study was only
41 cm to achieve 90% COD removal at a
Darcy velocity of 1.54 cm/day. Thus, even
a ten fold increase in the size of this
active zone would not be a problem at
most landfills.
Clogging of the soil due to microbial
growth has been a subject of concern at
landfill sites. The results of this study
showed a reduction in soil permeability
from 1.3xlO~3cm/sec initially to l.lxlO~3
cm/sec after 120 days of microbial activity.
This was not considered to be serious, but
further investigations over longer periods
are required before significant conclusions
can be drawn.
SUMMARY
The results of this work showed that
extensive removal of leachate organic
matter is possible in soil through the
action of microorganisms. While further
work is necessary to quantify microbial
reaction parameters in response to various
conditions and to provide field verifica-
tions, available information suggests that
microbial decomposition in soil is active
at many landfills. Since leachates tend
to be acidic with much of their organic
matter composed of free volatile fatty
acids, it would appear that methane pro-
duction occurs in the soil where pH is
buffered and not in the landfill.
The microbial processes were expres-
sed in terms of Monod-type equations to be
used as a model along with expressions for
advection and dispersion to simulate COD
reduction in the columns. The biological
parameters used in the simulation were
taken from reports external to the study.
Yet the quality of the simulation was
good. This supported the idea that the
methods and concepts presented in the study
were broadly applicable to landfills and
to other waste disposal systems as well.
Guidelines for application to other
situations were presented.
References
1. Chain, E.S.K. andDeWalle, F. "Treatment
of Leachate From Landfills", First
Annual Report, Solid and Hazardous
Waste Research Laboratory. Cincinnati,
1973.
2. Rovers, F.A. and Farquhar, G.J.
"Infiltration and Landfill Behavior".
ASCE, Journal of Environmental
Engineering Division, 9£, EE5, 1973.
3. Fungaroli, A.A., "Pollution of Sub-
surface Water by Sanitary Landfills,
Vol. 1". US EPA Report, SW-l2rg, 1971.
4. EMCON Associates, "Sonoma County ,
Refuse Stabilization Study, Second
Annual Report, Department of Public
Works, Santa Rosa, 1973.
5. McCarty, P.L. "Anaerobic Waste Treat-
ment Fundamentals, Parts One, Two and
Three". Public Works, September,
October, November, 1964.
6. Farquhar, G.J. and Rovers, R.A. "Gas
Production During Refuse Decomposition".
Water, Air and Soil Pollution, 2_,
Dordrecht, Holland, 1973.
7. Rovers, F.A., Farquhar, G.J. and
Nunan, J.P. "Landfill Contaminant Flux,
Surface and Subsurface Behavior".
Proceedings of the 21st Industrial
Waste Conference, MOE Ontario, June
1973.
8. Hughes, G.M., Landon, R.A. and
Farvolden, R.N. "Hydrogeology of Solid
Waste Disposal Sites in Northeastern
Illinois". Office of Solid Waste
Management Programs, US EPA, Washington
D.C., 1971.
253
-------
9. Farquhar, G.J., "Leachate Treatment
by Soil Methods". Proceedings EPA
Symposium on Management of Gas and
Leachate in Landfills, St. Louis, Ho.,
March 1977.
10. Griffin, R.A., Cartwright, K., and
Shimp, N.F., "Attenuation of Pollu-
tants in Municipal Landfill Leachate
by Clay Minerals". Environmental
Geology Notes, Number 78, Illinois
State Geological Survey, November,
1977.
11. O'Rourke, J.T. "Kinetics of Anaerobic
Treatment at Reduces Temperatures".
Ph.D. Thesis, Stanford Univ., Calif.,
1968.
12. Solid and Hazardous Waste Research
Division, "Municipal Solid Waste
Generated Gas and Leachate". Draft
Summary Report, Cincinnati, 1976.
13. Van Genuchten, M.T. and Wierenga, P.G.
"Mass Transfer Studies in Sorbing Porous
Media I. Analytical Solutions." Journal
Soil Science Society of American, 40,
4, July-August, 1976.
14. Sykes, J.F. "Transport Phenomenon in
Variably Saturated Porous Media". Ph.D.
Thesis, University of Waterloo, Waterloo
Ontario, 1975. '
15. Moore, W.A., McDermott, G.N., Post, M.A.,
Mordia, J.W. and Ettinger, B.A. "Effects'
of Chromium on the Activated Sludge
Processes". JWPCF, 33, 54-72, 1961.
16. Laurence, A.W., McCarty, P.L. and Grain
F.J.A. "The Effects of Sulphides on
Anaerobic Treatment" Proceeding of 19th
Industrial Waste Conference, Purdue, 1964.
17. Fuller, W.H. and Korte, N. "Attenuation
Mechanisms of Pollutants Through Soils".
Proceedings EPA Symposium on Gas and
Leachate From Landfills, Rutgers Uni-
versity, March, 1975.
18. Griffin, R.A., Au, A.K., Chian, E.S.K.,
Kim, J.H. and De Walle, F.B. "Attenuation
of PCB's by Soil Materials and Char
Wastes". Proceedings EPA Symposium on
Management of Gas and Leachate in Land-
fills, St. Louis, Mo., March, 1977.
19. Farquhar, G.J. and Sykes, J.F. "PCB
Attenuation in Soil" WRI Report,
University of Waterloo, February, 1978.
254
-------
USE OF POLLUTANT MOVEMENT PREDICTIONS TO IMPROVE
SELECTION OF DISPOSAL SITES*
Mike H. Roulier
U.S. Environmental Protection Agency
Cincinnati, Ohio
Wallace H. Fuller
University of Arizona
Tucson, Arizona
ABSTRACT
The characteristics of soils in restricting the movement of pollutants are one *actor
in the selection of safe disposal sites. Improvements in the measurements and prediction
of pollutant-soil interactions will allow the selection of disposal sites that are less
likely to contaminate underlying waters and may promote favorable public opinion and rul-
ings from zoning and regulatory groups; factors that significantly impact the approval of
a proposed disposal site. This paper discusses the current status of two studies of pol-
lutant movement modelling and the possible application of their results to disposal site
selection. Although quantitative predictions of nonconservative pollutant movement in
field soils do not appear feasible at present, it is concluded that qualitative predictions
are within present capability and can improve the selection and management of disposal
sites.
*Manuscript of the paper not received in time for publication.
255
-------
INTERACTION OF SELECTED LINING MATERIALS
WITH VARIOUS HAZARDOUS WASTES
H. E. Haxo, Jr.
Matrecon, Inc.
Oakland, California
ABSTRACT
The results'of the first year of exposure of a range of lining materials to selected
hazardous wastes are presented. The lining materials under study in this ongoing experi-
mental research project include a native soil, treated bentonite clay, soil cement, hy-
draulic asphaltic concrete, an asphaltic membrane, and a wide range of commercial or de-
velopmental polymeric membranes. The polymers used in the manufacture of these membranes
are polyvinyl chloride, chlorinated polyethylene, chlorosulfonated polyethylene, ethylene
propylene rubber, neoprene, butyl rubber, an elasticized polyolefin, and a thermoplastic
polyester. The nine hazardous wastes selected for expos lire testing included strong acids
strong bases, oil refinery tank bottom wastes, lead wastes from gasoline, saturated and in-
saturated hydrocarbon wastes, and a pesticide.
Specimens of the linings were exposed to the wastes both by total immersion in the
wastes and one-side exposure as a lining at the bottom of a cell under one foot of waste.
The effects of the wastes on the linings were measured by changes in physical prop-
erties of the lining and, in the case of the polymeric materials, by swelling. Initial re-
sults indicate highly specific reactivity of the linings with respect to the wastes. The
results also indicate the importance of the specific composition of the polymeric linings.
INTRODUCTION AND OBJECTIVES
Confining and controlling hazardous
wastes through the use of manmade impervi-
ous lining materials appears to be a feasi-
ble means of preventing seepage of pollut-
ants in wastes from entering and polluting
the ground water. A great range of materi-
als, differing in permeability to water and
potential wastes, composition, form and
shape, and costs, have been used or are po-
tentially useful for confining a range of
pollutants (1).
The ongoing project on which this paper
is based was undertaken with the following
broad objectives:
1. To determine the effects of expos-
ing a selected group of lining materials to
various hazardous wastes over an extended
period of time.
2. To determine the durability of and
the cost effectiveness of utilizing
synthetic membranes, various admix materials
and native soils as liners for hazardous
wastes storage and disposal ponds.
3. To estimate the effective lives of
12 lining materials exposed to 6 types of
industrial nonradioactive hazardous waste
streams under conditions which simulate
those encountered in holding ponds, lagoons,
and landfills.
This paper presents current results of
the exposure of a wide range of lining ma-
terials to 9 hazardous wastes, water, and
weather. A more complete and detailed re-
port of this work is now being prepared (2).
Details regarding the experimental approach,'
the design and construction of the exposure
cells, test methods, and preliminary test
results are reported in the proceedings of
256
-------
the Tucson meeting (3) and in the First In-
terim Report (4).
EXPERIMENTAL APPROACH AND METHODOLOGY
Our basic approach in meeting the ob-
jectives is to expose specimens of the var-
ious commercial lining materials under con-
ditions which simulate real service using
actual wastes, to measure seepage through
the specimens, and to measure effects of
exposure by following changes in important
physical properties of the respective lining
materials.
In this study, various lining materi-
als have been or are being subjected to 7
types of exposure testing:
- Bench screening tests; small speci-
mens immersed in wastes.
- Primary exposure cells; one-side ex-
posure to waste.
- Immersion of liners in wastes; two-
side exposure.
- Weather test; roof exposure.
- Weather test; small tubs lined with
membranes and containing wastes.
- Water absorption at room temperature
and 70°C.
- Membrane bags containing wastes in
deionized water; one-side exposure.
BENCH SCREENING TESTS
The bench screening tests were per-
formed early in the project (3,4) to de-
termine which combinations of lining mate-
rials and wastes should be installed in the
primary cells for long-term exposure test-
ing. It was recognized that some of the
lining materials would be unsuitable for
the confinement of certain wastes or that
the service lives could be expected to be
short. It was desired to eliminate such
combinations. For these tests small spec-
imens of the lining materials were hung or
placed in contact with the wastes. Swell-
ing and general effects of the exposure
were observed.
PRIMARY EXPOSURE CELLS
The major effort of this research pro-
gram is to assess the various lining
materials in test cells under conditions
which simulate real service. Test cells
were designed to meet this requirement, as
shown in Figures 1 and 2.
The version shown in Figure 1 is suit-
able for thin membrane materials; the ver-
sion shown in Figure 2 is for thicker mate-
rials, up to 12 inches. These are the pri-
mary cells of this research project.
The lining specimens, approximately
1 foot in area, are placed in these cells
under one-foot of waste. All of the speci-
mens except the asphaltic membranes are in
direct contact with the waste. The latter
are covered with 2 inches of silica sand.
This design allows one-side exposure
of the liner to the waste and measurement
of permeation or seepage of the waste
through the liner. Each combination of
waste and liner was installed in duplicate
in order that the samples could be exposed
for two lengths of time. The liners can be
retrieved and appropriate tests performed
to determine the effects of the wastes as a
function of time with the expectation that
an estimate can be made of service life and
durability of the lining materials.
IMMERSION OF LINERS IN WASTES; TWO-SIDE EX-
POSURE:
In order to extend the matrix of liner/
waste information, additional membrane lin-
ing specimens were hung in selected primary
cells and additional wastes. This type of
exposure presents two sides of the liner to
the waste which should accelerate the ef-
fects of the wastes. Weight and area
changes of the specimens and measurements
of physical properties can be made. Fur-
thermore, hanging specimens in the wastes
can allow the various phases or layers of
the wastes to come in contact with the speci-
imens. This is in contrast to the primary
cells in which the lining specimen is at
the bottom of the cell and only the lower
phase of the waste, probably aqueous, di-
rectly contacts the liner.
WEATHERING TESTS: ROOF EXPOSURE ON RACK AND
IN SMALL LINED TUBS
As liners in actual service are often
exposed to the weather, two weathering
tests were included in this project. In
one test small specimens are exposed on a
rack on the roof and removed periodically
257
-------
Top Cover
Epoxy
Coating
Steel Tank
to
s
Caulking
'- Crushed Silica
Outlet tube with
Epoxy-cooted
Diaphragm
Fig. 1 - Erasure Cell for Membrane Liners.
-------
-Top Cover
Epoxy
Coated-
-\
O1
<0
Flanged Steel
Spacer —
Waste
-Neoprene Sponge Gasket
^Epoxy Grout Ring
ADMIX
~Epoxy and Sand
Coating
LINER
Waste Column :
-11 Gauge Steel
10" x 15" x 12" High
w/ Welded
2 " Flange
Outlet tube with
Epoxy-coated
Diaphragm
|>-Glass Cloth
To
Collection
Bag
Fig. 2 - Exposure Cell for Thick Liners.
-------
to determine changes in dimensions, weight,
and hardness. They also can be removed and
their stress-strain properties determined.
In addition, 12 small plywood tubs
were lined with polymeric membrane lining
materials in which seams had been incorpo-
rated. These tubs were filled with the same
wastes as used in the primary and immersion
tests. This type of test simulates open-
air service, such as in ponds, and combines
the weather and waste exposure simultaneous-
ly. These tubs also feature draping of the
liner over sharp corners to simulate a se-
vere condition encountered in the field,
i.e., sharp objects. Materials which are
subject to ozone cracking or creep are sus-
ceptible to failure at such points. The
open exposure can also result in consider-
able wanning of the wastes in summer weath-
er, which can accelerate deterioration.
WATER ABSORPTION AT ROOM TEMPERATURE AND
70°C
As most of the wastes contain water
which can affect various polymeric lining
materials, two series of exposures were un-
dertaken, one at room temperature and the
other at 70 C, the latter to accelerate
water absorption.
MEMBRANE BAGS CONTAINING WASTES
In the ongoing EPA study of liners for
sanitary landfills (5), a test method is
being developed which was applied to the
liners for hazardous wastes. In this meth-
od, small bags of heat-sealable polymeric
membranes are fabricated and filled with
wastes or test fluids. These bags are seal-
ed and immersed in deionized water.
To determine the permeability of the
membranes to water and pollutants, the pH
and electrical conductivity of the deion-
ized water and the change in weight of the
filled bags are determined as a function of
time. At the end of the exposure period,
these bags are emptied and the physical
properties of the liners determined. Thus,
in this test, it is possible for both short
and long periods of time to determine:
- The permeability of the liner to
both wastes and water.
- The functioning of the seams.
- The durability of the material to
one-side exposure to a waste.
This type of bag is shown in Figure 3.
DESCRIPTION OF HAZARDOUS WASTES
A total of 9 hazardous wastes from the
petroleum, chemical, and pesticide indus-
tries were selected for use in this test
program. They included the following basic
types:
- 2 strongly acidic wastes.
- 2 strongly alkaline wastes.
- 1 lead waste (a blend of 3).
- 3 oily wastes.
- 1 pesticide waste.
Data on these wastes are presented in.
Tables 1 and 2. The scheme of waste analy-
sis described by Stephens (6) has been fol-
lowed in general. Six of the wastes were
placed in the primary cells for exposure
testing of liner materials (see Table 1).
The additional 3 wastes were placed in sev-
eral additional cells, but the primary pur-
pose for these was for use in the immersion
tests.
Individual wastes are briefly describ-
ed below. The designation for each waste
is shown in quotes, taken from the barrels
in which the wastes were delivered.
Acidic Wastes
Waste "HN03-HF-HOAC" is a mixed acid
containing nitric, acetic, and hydrofluoric
acid. It was reported to contain approxi-
mately 17% nitric acid. It is a transpar-
ent, light-straw colored material contain-
ing a small amount of suspended solids and
has the viscosity approximately that of wa-
ter. It contains no organic fraction. This
waste was used in both the primary exposure
and immersion tests.
"HFL" is another relatively strong
acid; however, it is far weaker than the
above acid. "HFL" was used in immersion
tests only.
Alkaline Wastes
"Slopwater" is basically a water solu-
tion of caustic. It has a high pH and con-
tains a high percentage of various briny
solids. This waste was used in immersion
tests only.
260
-------
HAZARDOUS WASTE
0)
MEMBRANE UNDER TEST
DEIONIZED WATER
POLYBUTYLENE
Fig. 3 - Bag Containing Waste Immersed in Water.
-------
TABLE 1. WASTES IN EXPOSURE TESTS
Phases
Type of Waste
Acidic
Alkaline
Lead
Oily
Pesticide
Name
"HFL"a
"HNO , HF, HOAC"b
"Slopwater"a
"Spent caustic"*3
"Low lead gas washing" )
) Blend
"Gasoline washwater" )
"Aromatic oil"b
"Oil Pond 104"b
"Weed oil"a
"Weed killer llb
Organic
Phase
I
0
0
0
0
10.4
1.5?
100
89.0
20.6
0
Water
Phase
II
100
100
100
95.1
86.2
89.1?
0
0
78.4
99.5
Solids
Phase
III
0
0
0
4.9
3.4
0
11.0
-
0.5
In both primary exposure and immersion tests.
In immersion tests only
TABLE 2.. WASTES IN EXPOSURE TESTS
pH, Solids, and Lead
Type of waste
Acidic
Alkaline
Lead
Oil
Pesticide
Name
"HFL"
"HN03, HF, HOAC"
"Slopwater"
"Spent caustic"
"Low lead gas washing"
"Gasoline washwater"
"Aromatic oil"
"Oil Pond 104"
"Weed oil"
"Weed killer"
PH
Water phase
4.8
1.5
12.0
11.3
7.2
7.9
-
-
7.5
2.7
Solids, » Lead.
Total
2.48
0.77.
22.43
22.07
1.52
0.32
-
ca. 36
1.81
0.78
Volatile ppm
0.9
0.12
5.09
1.61
0.53 34
0.17 li
-
ca. 31
1.00
0.46
262
-------
"Spent caustic", obtained from the
petroleum industry, is a somewhat less al-
kaline waste. It is transparent, tan-col-
ored, containing white crystalline solids
in suspension. It contains no oily materi-
al and has a strong soapy odor. Its vis-
cosity is much like water; however, when
cool, considerable crystallization occurred
which caused problems in the mixing and
pumping of this waste in the loading of the
cells.
Lead Waste
Three lead wastes were originally ob-
tained, two of which are listed in the
tables and a. third which had a very low-
lead content and was essentially aqueous.
All of the two listed, which contained most
of the lead, were blended and a minor a-
nount of the third was added to make the
volume needed to fill the primary cells.
This blended waste contains considerable
gasoline and suspended solids. Originally
this was not considered to be an oily waste,
as the first samples received were aqueous,
thus pointing out a problem in obtaining
representative samples of wastes.
The blended waste has a low-viscosity
and separates into aqueous and oily phases.
Such a waste, with its low-boiling hydro-
carbons and water, can be expected to cause
damage to organic-type lining materials,
such as the polymers and asphalts.
Oily Wastes
Aromatic oil is a highly aromatic,
high-viscosity oil, such as used in the
manufacture of carbon black. It is an
opaque brown oil containing some suspended
solids. It required heating in order to
be pumped into the test cells.
"Oil Pond 104" is predominantly oil
with some suspended solids. It is a naph-
thenic hydrocarbon and contains no water.
"Weed oil" is an oil-water mixture
with a relatively low-molecular weight oil.
This waste separated with the water phase
on the bottom. It was used in immersion
tests only.
Pesticide Waste
"Weed killer" is an herbicide, and is
predominantly water with a minor amount of
clay or talc as solid carrier. A waste of
this type should have a minimum aggressive
character with respect to linings.
SOIL AND ADMIX LINER MATERIALS
The 5 soil and admix lining materials
which were selected for this test are:
Thickness,
inches
Asphalt emulsion on nonwoven
fabric 0.3
Compacted native fine-grain soil 12.0
Hydraulic asphalt concrete 2.5
Modified bentonite and sand 5.0
Soil cement with seal 4.0
Linings of these materials are pre-
pared on site. The preparation of these
lining materials is described in Refer-
ence 4. These materials all have water
permeabilities less than 10~7 cm/sec (4).
The admix lining materials are being
subjected to appropriate tests from the fol-
lowing list before and after exposure to
wastes:
Permeability
Density and voids
Water swell
Compressive strength
Penetration of asphalt
Viscosity of asphalt
All of the liners were tested ini-
tially for water permeability. The perm-
eability of the lining material to the
waste is determined by the amount of the
wastes which seep through the liner. Sam-
ples can also be taken from the liner after
exposure and tested for permeability to wa-
ter or water vapor.
POLYMERIC MEMBRANE LINING MATERIALS
Eight polymeric membranes, based on
different polymers, were selected for these
exposure tests:
263
-------
Butyl rubber
Chlorinated polyethylene (CPE)
Chlorosulfonated polyethylene
Elasticized polyolefin
Ethylene propylene rubber (EPDM)
Neoprene
Polyester elastomer
(developmental)
Polyvinyl chloride (PVC)
R = Fabric reinforced
Thickness,
mils
34-R
30
34-R
23
37
32
8
33
All of these membrane linings, ex-
cept the polyester elastomer, are commercial
products. The polyester elastomer membrane,
although still in development, was included
because of its reportedly high resistance
to oily wastes.
The physical properties which were
measured on the original materials are:
*Tensile strength
*Elongation at break
*Modulus, S100 and S200
*Tear strength
Hardness
Puncture resistance
Water vapor permeability
Seam strength
Swelling in water and waste
*Tested with and across grain.
Details regarding these materials and
their properties are reported in Reference
4.
After exposure to the wastes the lin-
er materials are subjected to most of the
same tests listed above. Several of the
tests are being run in two directions. This
is necessary because of the grain introduced
in the membranes during manufacture. In
most cases the data in the two directions
of test are averaged for assessing the ef-
fects of exposure.
In addition to physical properties,
the following analytical tests are being
run on individual membrane liners to de-
termine their basic compositions:
- Density
- Ash content
- Extractables
- Polymer content
- Carbon black content
EFFECTS ON LINING MATERIALS
OF EXPOSURE TO HAZARDOUS WASTES
LINERS IN PRIMARY CELLS
Monitoring During the Exposure Period
The overall performance of the cells
during the monitoring period was satisfac-
tory for all of the cells except those con-
taining the highly acidic waste, "HNO -HF-
HOAC". This waste attacked the epoxy coat-
ing of the cells resulting in blistering of
the coating and corrosion of the steel cell.
Failure of the coating occurred primarily in
the flange area at the welds, resulting in
actual loss of the waste. This waste also
deteriorated the polyester elastomer lining
and caused failure of both of the hydraulic
asphaltic concrete liners. The asphaltic
concrete also leaked in one of the "spent
caustic" cells and a cell containing lead
waste. The cell containing the "spent
caustic" was repaired and returned to expos-
ure. The concrete liner in the cell con-
taining the lead waste essentially lost all
physical properties.
Leakage around the gasket occurred on
several cells but they were repairable by
tightening the bolts. Only in a couple of
cases was it necessary to dismantle the cell
and remount the liner.
Dismantling of Cells and Recovery and
Testing of Lining Specimens
The dismantling of the cells was per-
formed in a progressive manner, starting
with the cells containing the pesticide
waste and ending with the cells containing
aromatic oil. This allowed the testing to
be performed on specimens that had been
newly removed from the cells. At the time
of dismantling, the liners were photograph-
ed both in the cell and after removal from
the cell. The individual liners were placed
in polyethylene bags to protect them and to
prevent evaporation of absorbed materials
prior to testing.
All of the membrane liners and the fol-
lowing admix liners were removed:
- Hydraulic asphaltic concrete
- Soil cement
264
-------
TABLE 3. COMBINATIONS OF ADMIX LINER MATERIALS AND HAZARDOUS WASTES
WASTE
Liner Material
Asphalt emulsion
Compacted native soil
Hydraulic asphalt concrete
Modified clay and sand
Soil cement
Thick-
ness,
inches
0.3
12.0
2.5
5.0
4.0
Acidic Alkaline
"HN03 "Spent
HF-HOAC" caustic"
X
X
X X
-
X
Lead
X
X
X
X
X
Oily
Aro- "Pond
matic 104"
_
X X
-
X X
X X
Pest-
icide
"Weed-
killer"
X
X
X
X
X
- Emulsified asphalt on nonwoven
fabric
All of the native soil and modified-clay
sand specimens remain under exposure test-
ing.
Admix Materials
As a result of the preliminary bench
tests, specific combinations of admix lin-
ers and wastes were selected as shown in
Table 3. This selection resulted in the
elimination of combinations of oily wastes
and the asphaltic lining materials and of
acidic and highly ionic wastes and the soil
and clay type linings.
The performances of the lining materi-
als to date are described below:
Hydraulic asphalt concrete—
This lining material has given us the
most difficulties. First, several of the
lining specimens failed between the steel
spacer and the specimen due to the differ-
ence in the coefficient of thermal expan-
sion of steel and asphaltic concrete. This
resulted in the loss of adhesion and by-
passing of the liner by the waste. It was
necessary to cast epoxy walls around each
liner to replace the steel spacers. Second,
four of the asphaltic concrete liners have
leaked, two containing strong acid, one con-
taining spent caustic, and one containing
lead waste. The liners in the cells con-
taining the strong acid failed and were re-
moved. The liner in the cell containing
spent caustic was repaired and put back in
the test and the liner in the cell contain-
ing the lead waste was removed and tested.
The hydraulic asphaltic concrete lin-
ers were cored and the asphalt extracted
from different layers. The asphalt in the
liners from the cells containing the acid
waste hardened considerably, particularly
that in the top layers which were in con-
tact with the waste (Table 4). The asphalt
concrete liner from the cell containing the
lead waste absorbed large amounts of the
organic fraction in the waste and became
very soft, almost a slush. The liner with
the pesticide waste remained in satisfac-
tory condition, retaining its compressive
strength and only showing slight hardening
on aging.
Emulsified Asphalt on Nonwoven Fabric—
During the leak-testing of the cells
with water, considerable delamination and
blistering of the asphaltic materials oc-
curred. Consequently, two inches of silica
sand was placed on these liners. As a re-
sult of the preliminary bench tests, the
emulsified asphalt liner was placed only in
cells containing pesticide, "spent caustic",
and the lead wastes. There was no leakage
in these cells during the exposure period.
The liners were recovered after 469-
487 days of exposure. All appeared to be
in satisfactory condition, although there
was swelling of the liners taken from the
cells containing lead waste and "spent caus-
tic" waste. The relatively low swelling of
this liner below the lead waste which con-
tains volatile hydrocarbons is probably the
result of the sand-water layer that was di-
rectly above the liner.
Soil Cement—
All of the wastes, except the acid
waste, were placed above the soil cement
265
-------
TABLE 4. HYDRAULIC ASPHALT CONCRETE AFTER EXPOSURE TO HAZARDOUS WASTES
Waste - Type
Name
Cell No.
Exposure time, days
a
HAC Concrete
Coefficient of permr
eability, cm sec
Density, g/cm
lb/ft3
Voids content, % vol.
Compressive strength,
psi
% original
Extracted asphalt
None Acidic
"HNO -HF-HOAC"
C-31 C-26
40 199
1.3xlO~8 2.6xlO~8 2.4xlO~
2.322 - 2.294
145.0 - 143.2
4.0 - 5.8
268 162 234
60 87
Lead Pesticide
"Weed Killer"
C-27 C-23
192 569
(b)
2.351
146.8
3.4
21b 237
8 88
Depth from top, cm
Viscosity at 25°C,MP
0-1
2-3 0.3-1 2-3
0-1 2-3
0-1 2-3
at 0.01 s
Shear susceptibility
Penetration at 25°C,
calculated
4.
0.
90
10
48
1.78
0.07
71
3.66
0.09
53
1.22
0.13
87
(d) 1.
0.
78
14
75
2.73
0.11
60
2.55
0.03
58
1.78
0.14
75
Thickness of liner specimen: 6.3 cm (2.5 in.)
bCompressive strength measured using "Pocket Penetrometer" because cores were too
soft to hold shape, so could not be tested. The gasoline absorbed from the waste
evaporated along with the benzene used for extraction of the asphalt from the ag-
gregate, so the viscosity measured does not represent the very low viscosity while
the liner was in place.
°Asphalt from 0 to 0.3 cm was too hard to test.
Asphalt from 0-1 cm was too hard to test.
6 Calculated from absolute viscosity at 25 C.
liners. The acici waste was deleted from ex-
posure to the soil cement because of the
poor resistance of the soil to the strong
acid waste as shown by the bench tests. No
seepage of waste occurred through the soil
cement specimens during the exposure period.
On recovery, they were in satisfactory con-
dition.
In the case of the coating on the lin-
ers exposed to the "spent caustic" and the
pesticide wastes, there were some blisters.
The compressive strengths of the cores taken
from these liners were all satisfactory and
had higher values than specimens which were
not exposed to the waste but kept in poly-
ethylene bags.
Native Soil—
All of the wastes except the acid
waste were placed on the compacted soil lin-
ers. During the course of monitoring, all
of the cells seeped. However, the composi-
tion of these was essentially the same in
all cases, containing large amounts of salt.
This soil had been dredged from the mouth
266
-------
of the Sacramento River and placed on land
at Mare Island. It contains large amounts
of salt from the water.
It is planned to dismantle these cells
in the near future and to determine the mi-
gration of the wastes or their components
through the soil.
Modified Clays--
Pesticide, lead waste, "Pond 104 Oil",
and aromatic oil were placed on liners pre-
pared from modified clays and sand. These
clay liners were not exposed to the acid
and alkaline wastes because of the fail-
ures of these materials with these wastes
in the bench tests. During the monitoring
there was no seepage through these lining
materials. They remain under test but will
be recovered within the next two months
and the migration of the wastes into them
will be determined.
Polymeric Membranes
The effects of the wastes upon the
membrane linings can vary considerably with
the polymer and the waste, as shown in
Tables 5 and 6 by the elongation and hard-
ness of the liners on exposure. Both of
these properties are measures of the rub-
bery character of polymers. Substantial
increases or decreases in either of these
properties can significantly affect the
performance of many rubber products. One
of the objectives of this project is to de-
termine the effects of changes in proper-
ties upon the utility of the liners. Noted
on both of the tables are those combina-
tions of liners and wastes which were not
tested.
The elongation of the polyester liner
in the acid waste dropped to essentially
zero during the exposure resulting in the
failure of this liner. The butyl liner has
a low elongation because it is reinforced
and breaks with the breaking of the fabric.
The values shown are those at which the
tensile was the maximum. In most cases,
there is a drop in elongation; however, in
several cases the elongation increased due
to absorption of a fluid.
The hardness values generally decreas-
ed during the exposure period, particularly
in those liners which were in contact with
oily wastes. There were increases in hard-
ness of the PVC liners exposed to the acid
waste and to the "Pond 104" oil, probably
reflecting the loss of plasticizer.
Immersion Tests of Polymeric Membrane
Liners
In the immersion testing, 12 membrane
liners were exposed to 9 wastes. The 6
TABLE 5. RETENTION OF BREAKING ELONGATION OF POLYMERIC MEMBRANES
ON EXPOSURE TO WASTES IN PRIMARY CELLS
polymer
Butyl
Chlorinated polyethylene
Chlorosulfonated polyethylene
Elasticized polyolefin
Ethylene propylene rubber
Neoprene
Polyester elastomer
Polyvinyl chloride
Liner
No.
57-R
77
6-R
36
26
43
75
59
Original
Value,
%
45
410
245
665
470
320
575
375
Acidic
"HNO3
HF-HOAC"
56*
88
90
99
94
No X
0.4°
83
Alkaline
"Spent
caustic"
56*
105
69
100
98
99
104
103
WASTE
Lead
111
100
106
93
96
77
98
96
Oily
Aro-
matic
No Xb
No X
No X
96
No X
No X
77
91
'Pond
104"
No X
96
102
86
No X
86
95
85
Pest-
icide
"Weed-
killer1
133
99
110
101
96
94
96
101
Elongation at maximum tensile with failure of the reinforcing fabric.
bNo X - No exposure test because of poor performance in bench tests.
CFailed at 323 days.
o = Fabric reinforced.
267
-------
TABLE 6. CHANGES IN HARDNESS OF POLYMERIC MEMBRANES
ON EXPOSURE TO WASTES IN PRIMARY CELLS
Polymer
Butyl
Chlorinated polyethylene
Chlorosulfonated polyethylene
Elasticized polyolefin
Ethylene propylene rubber
Neoprene
Polyester elastomer
Polyvinyl chloride
Liner
No.
57 -R
77
6-R
36
26
43
75
59
Original
Value,
71
84
77
87
58
57
93
72
Acidic
"HNO,
HF-HOAC"
-4
-12
-3
-3
-8
NO X
-15. 5C
+ 12
A Pea line
"Spent
caustic"
-2
-9
-2
-2
-2
+1
-2
-1
WASTE
Lead
-4
-17
-11
0
-1
-17
-3
-3
Oily
Aro-
matic
No X*5
No X
NO X
-10
NO X
No X
-10
-4
"Pond
104"
No X
-34
-32
-16
No X
-20
-10
+4
._
Pest-
icide
"Weed-
killer"
_
-4
-7
-4
+1
-5
-9
-1
-2
Duro A - 10 sec. reading.
No X - No exposure test because of poor performance in bench tests.
Failed in 323 days.
R = Fabric reinforced.
basic polymers are included in the 12. How-
ever, the 12 now include 2 chlorosulfon-
ated polyethylene, 2 ethylene propylene
rubber, and 3 polyvinyl chloride membrane
linings. The original 6 wastes used in the
primary cells are again used in this test,
plus the following:
Acidic waste
Alkaline waste
Oily waste -
"HFL"
"Slopwater"
"Weed Oil"
The effect of the immersion upon the
liners is shown in Table 7, which presents
the increase in weight during the exposure
TABLE 7. SWELLING OF MEMBRANE LINING MATERIALS ON IMMERSION IN WASTES
% Weight increase
Polymer
Innersion time, days
Butyl rubber
Chlorinated polyethylene
Chlorosulfonated poly-
ethylene
Elasticized polyolefin
ethylene propylene rubber
Neoprene
polyester elastomer
Polyvinyl chloride
Liner
No.
-
44
77
6-R
55
36
83
91
90
75
11
59
88
Vulcan-
ized
-
Ves
NO
No
No
No
Yes
?
Yes
No
No
No
No
Acidic
HF
250
2.7
9.4
6.8
5.4
0.3
3.1
16.1
9.6
0.6
10.2
2.8
7.6
Wastes
HNO3
193
1.4
9.3
10.3
7.5
2.7
2.6
18.3
10.9
4.2
16.8
-2.8
19.8
Alkaline Wastes
Slop-
water
193
2.0
1.5
3.8
3.8
17.3
2.7
3.1
0.4
35.1
13.5
-6.4
-13.5
Spent
caustic
238
0.4
0.6
3.3
2.2
0.5
1.3
0.2
0.8
0.6
0.1
-3.0
0.1
Lead
Waste
236
20.1
70.9
83.0
69.6
18.2
23.0
29.3
45.6
7.6
4.4
8.8
2.2
Oily Hastes
Are-
na tic
257
32.3
59.5
51.1
53.2
21.3
15.8
35.3
60.7
17.1
10.7
11.3
7.2
Pond
104
248
96.5
31.6
75.1
58.4
33.5
35.4
80.1
25.8
7.9
-7.7
-1.5
-10.3
Weed
252
70.8
116.7
202.3
210.5
44.2
73.4
79.4
94.8
16.3
10.0
33.4
18.1
Pest-
icide
Herb-
icide
242
0.8
9.6
13.1
12.3
0
3.7
8.1
8.5
2.4
4.0
0.5
2.9
Reinforced
268
-------
TABLE 8. RETENTION OF TENSILE STRENGTH OF MEMBRANE LINER
MATERIALS ON IMMERSION IN "WEED OIL"a
(Percent of original value)
Polymer
Butyl rubber
Chlorinated polyethylene
Chlorosulfonated polyethylene
Elasticized polyolefin
Ethylene propylene rubber
Neoprene
Polyester elastomer
Polyvinyl chloride
Liner
No.
44
77
6-R
55
36
83
91
90
75
11
59
88
Original
Value ,
psi
1520
2235
1730
1715
2595
970
1865
1940
6770
2955
2555
3155
Retention, %
Top
of specimen
16
(b)
7
(b)
26
(b)
30
18
9G
72
50
63
Bottom
of specimen
55
4
43
45
32
39
40
32
81
84
42
68
Immersion time, 252 days.
Too soft to test.
R = Fabric reinforced.
period. As in -cne primary cells, the ef-
fects vary considerably with liner type and
waste. In addition, there is a significant
variation in the effects between materials
of the same polymer. This is particularly
true among the three PVC lining materials.
The hanging of the samples in the
wastes allows the various phases of the
waste to contact the specimens. The effect
upon properties of the top and bottom sec-
tions of the test specimen is shown in
Table 8.
ROOF EXPOSURE
Eleven samples of polymeric membrane
liners were exposed 343 days on the roof of
our laboratory in Oakland, California.
They were removed, cleaned, weighed, meas-
ured, and stress-strain properties deter-
mined. Results of the liner tests are pre-
sented in Table 9. Most of these materials
show a loss in elongation and an increase
in modulus. The liner material which show-
ed the greatest change was the chlorosulfon-
ated polyethylene, which retained 61% of
its elongation and doubled in modulus. This
probably reflects a crosslinking of the
polymer during exposure.
WATER ABSORPTION OF POLYMERIC LINERS
Results of the water absorption tests
run on the same 11 polymeric membrane lin-
ers at room temperature and at 70 C are
presented in Table 10. The immersion time
in this experiment was 308 days. These re-
sults show the great difference in ultimate
swelling of these lining materials by water.
Of particular interest is the low water
swell of the elasticized polyolefin mem-
brane and of these polyvinyl chloride mem-
branes at room temperature.
269
-------
TABLE 9. RETENTION OF PROPERTIES ON ROOF EXPOSURE
OF POLYMERIC MEMBRANE LINING MATERIALS
(Retention in percent)
Polymer
Butyl rubber
Chlorinated polyethylene
Chlorosulfonated polyethylene
Bias tici zed polyolefin
Ethylene propylene rubber
Neoprene
Polyester elastomer
Polyvinyl chloride
Liner
NO.
57-R
77
6-R
36
8
26
43
82
75
11
59
Elongation
at break
125
103
61
96
93
87
76
85
89
111
96
S-100
Modulus
-
116
200
117
137
111
146
107
104
105
111
343 days in Oakland, CA.
R =Fabric reinforced.
TABLE 10. SWELLING OF POLYMERIC MEMBRANE LINERS IN WATER
(Percent weight increase)
Polymer
Butyl rubber
Chlorinated polyethylene
Chlorosulfonated polyethylene
Elasticized polyolefin
Ethylene propylene rubber
Neoprene
Polyester elastomer
Polyvinyl chloride
Liner
No.
57-R
77
6-R
36
8
26
43
82
75
10
59
Room
Temp.
4.5
10.2
10.9
0
1.6
1.5
37.8
18.5
10.2
0.7
2.4
70°C
53.9
140.0
245.6
0.6
10.8
11.2
240.0
191.4
140.0
39.2
24.0
R = Fabric reinforced.
Immersion time: 308 days
270
-------
TABLE 11. SEALED BAGS CONTAINING STRONG ACID WASTE IMMERSED IN DEIONIZED WATER
Polymer
Chlorinated polyethylene
Chlorosulfonated polyethylene
Elasticized polyolefin
Polybutylene
Polyvinyl chloride
Blank
Liner
No.
86
85
36
98
19
88 '
-
Deionized water
PH
4.0
7.7
5.9
3.1
2.5
2.4
4.9
Conductivity
^i mho
132
306
25
296
1970
2285
7
Increase in
weight of bag
containing waste, g.
6.1
8.9
0.5
0.6
5.0
11.3
-
Immersion time, 337 days.
MEMBRANE BAGS CONTAINING WASTES
In this test of the liner membranes
run up to 337 days, the membranes have been
found to be very impermeable. There has
been a slight increase in the conductivity
of the deionized water outside the bags in-
dicating some ions pass through the mem-
branes. On the other hand, the bags have
generally increased in weight,%indicating
the movement of water into the bags due to
osmotic pressure. This is illustrated in
Table 11 for the bags containing strong
acid.
DISCUSSION
Although the results must be consid-
ered preliminary at this point in the ex-
posure testing, it is quite apparent that
some of the hazardous wastes can seriously
affect the physical properties of lining
materials. There is a highly specific in-
teraction between lining materials and
wastes which requires testing of specific
liner and waste combinations for compati-
bility.
In characterizing the wastes, compon-
ents which are aggressive toward linings
should be identified. Organic constituents
tend to have solvent effects upon the or-
ganic polymeric membranes and asphaltic ma-
terials. The effects will depend upon the
specific components and the liner material.
The acidity and ion concentration are par-
ticularly important when soils and clays
are being considered.
The fact that many of the wastes are
heterogeneous and can separate into oily
and aqueous phases poses additional prob-
lems in the selection of lining materials.
At the present time no one lining material
is satisfactory for both of these compon-
ents of the wastes, although some of the
liners achieve a better compromise than
others.
ACKNOWLEDGMENTS
The work which is reported in this pap-
er is being performed under Contract 68-03-
2173, "Evaluation of Liner Materials Expos-
ed to Hazardous and Toxic Sludges", with
the Environmental Protection Agency, Munic-
ipal Environmental Research Laboratory,
Cincinnati, Ohio.
The author wishes to thank
Robert E. Landreth, Project Officer, for
his support and guidance in this project.
The author also wishes to acknowledge Dr.
Clarence Golueke, who is Principal Investi-
gator for the Sanitary Research Engineering
Laboratory, University of California,
Berkeley, who is responsible for the char-
acterization of the individual wastes, and
the efforts of R. M. White, R.S. Haxo, and
John Holliday in carrying out the experi-
mental work involved in this project.
REFERENCES
1. Kays, W.B., "Construction of Linings
for Reservoirs, Tanks, and Pollution Con-
trol Facilities", John Wiley & Sons, Inc.,
N.Y., 1977.
271
-------
2. "Liner Materials Exposed to Hazardous
and Toxic Sludges, Second Interim Report
(in preparation).
3. Haxo, H.E., "Evaluation of Selected
Liners Exposed to Hazardous Wastes," Pro-
ceedings of Hazardous Waste Research Sym-
posium, "Residual Management by Land Dis-
posal," Tucson, Arizona, EPA-600/9-76-015,
July 1976.
4. Haxo, H. E., Haxo, R.S., and White,R.M.,
"Liner Materials Exposed to Hazardous and
Toxic Sludges," First Interim Report, EPA-
600/2-77-081, June 1977.
5. Haxo, H.E., "Compatibility of Liners
with Leachate", Proceedings of 3rd Annual
Municipal Solid Waste Research Symposium,
University of Missouri, EPA-600/9-77-026,
September 1977.
6. Stephens, R.D., "Hazardous Waste Sam-
pling, " Proceedings of the Hazardous Waste
Research Symposium, "Residual Management by
Land Disposal, Tucson, Arizona, EPA-600/9-
76-015, July 1976.
272
-------
THE USE OF LINER MATERIALS FOR SELECTED FGD WASTE PONDS
By
Z. B. Fry, Gen Engr and C. R. Styron III, Res Civil Engr
Soils and Pavements Laboratory
U. S. Army Engineer Waterways Experiment Station
Vicksburg, MS 39180
ABSTRACT
A comparison was made of the compatibility of 18 linear materials and two selected PGD
waste sludges. A total of 72 special test cells was constructed to contain the liners and
the sludges. Devices were installed to collect leachate from the sludge, and the cells
were pressurized to simulate a 30-ft head of sludge. Physical and chemical tests were con-
ducted prior to initiating the investigation, at the midpoint (12 months) and then sched-
uled for the end (24 months). The leachate collected for the 12 months has been analyzed
for content of approximately 20 heavy metals. Also, observations and physical tests were
made of the liner materials. At the midpoint of the investigation, two of the admix liner
materials had disintegrated due to reaction with the FGD sludge. The leachate analysis
from the majority of the other test cells indicated a considerable attenuation of the con-
centration of heavy metal content as compared with the analysis of the original or raw
sludge. Further tests are scheduled during the remainder of the investigation to determine
the cause of this phenomenon.
INTRODUCTION
The utilization of liners for water
storage ponds, irrigation canals, and even
industrial waste ponds is not a new con-
cept. There is, however, a lack of knowl-
edge concerning the compatibility of liner
materials subjected to certain toxic wastes
and the life expectancy of the liners.
The U. S. Army Engineer Waterways
Experiment Station (WES) has been engaged
for a number of years in research of the
chemical and mechanical stabilization of
soils and the development of prefabricated
membrane surfacings and dust control mate-
rials for military applications. On the
basis of the capabilities established and
procedures and materials developed by the
above-mentioned research, the WES has
undertaken, through an interagency agree-
ment with the U. S. Environmental Protec-
tion Agency, a study to assess the
potential feasibility for application of
selected materials to use as liners for
flue gas desulfurization (FGD) waste
areas.
OBJECTIVES
The objectives of the study are as
follows:
a_. To determine the compatibility of
liner materials with FGD sludge
wastes and associated liquors and
leachates.
b_. To estimate the length of life for
the liners.
c_. To assess the economics involved
with the purchase and placement
(to include construction) of
various liner materials.
A subsequent objective required that the
liner materials be subjected to a simulated
30-ft head (depth) of sludge, as would be
expected in prototype storage ponds.
273
-------
SCOPE
This paper presents the liner mate-
rials and sludge wastes selected for the
study, the test devices and procedures
developed, and a comparison of selected
results obtained at the initiation and at
the end of a 12-month period of study.
APPROACH
To accomplish the objectives, primary
consideration was placed on the use of
admixed and/or stabilized in situ mate-
rials; secondly, on spray-on materials;
and finally, limited use of prefabricated
membrane-type materials. This resulted
in a requirement to study a total of
18 liner materials consisting of two
prefabricated membranes, six spray-ons,
and ten admix-type liners. The liner
materials were subjected to two FG-D
sludges. The liners were installed in
exposure cells designed to simulate a
30-ft depth of sludge, which was applied
over a 10-month period of time for this
particular study. The liner materials
were scheduled to be exposed to the sludges
for 12- and 2U-month periods. Prior to
installation in the exposure cells, physi-
cal tests were conducted on the liner
materials and a chemical analysis was
made of the sludges. The physical tests
were scheduled to be made on the liner
materials at the end of the 12- and
2l*-month periods of exposure. The expo-
sure cells were so designed that any
leakage of the liquid from the sludge
was collected and chemical analysis of
such leachate was made as required, i.e.,
when sufficient liquid was collected for
the required analysis. The chemical
analyses of the raw sludge, both solid
and liquid portions, were evaluated for
the presence of heavy metals, which now
number 20 such parameters.
MATERIALS AND METHODS
Liners.—The selection of the liner
materials involved a screening process
based on prior experience with the mate-
rials and additional evaluation primarily
using a test for permeability of the liner.
The test for permeability was readily
applicable for the membrane and spray-on
materials but was modified to some extent
for the admixed materials. The modifica-
tion consisted of conducting the perme-
ability test on admixed samples constructed
in a Harvard miniature test apparatus.
This procedure precluded construction of
rather large samples and thus reduced the
time and materials needed for the tests.
It further provided additional samples
for physical tests. Since a rather large
number of admixed samples were discarded
because of high permeability, new mixes
were prepared with different percentages
of additives. The prior research con-
ducted at the WES for stabilization of
soils required principally an increase in
strength, and permeability was not con-
sidered of great importance.
The final items selected for the two
prefabricated membrane liners are listed
in Figure 1. Although there are hundreds
commercially produced, these membranes were
selected due to ease of placement and
strength, which would possibly prevent
damage during construction.
The spray-on materials selected for
the study are listed in Figure 2. The
materials are basically products of research
at the WES for dust control palliatives and
are the items that exhibited the greater
performance potential.
The admix liner materials are shown in
Figure 3. The materials (additives) are,
except for the Takenaka Aqua-Reactive
Chemical Soil Stabilization System (TACSS)
materials, generally those commonly used
for stabilization and paving purposes and
for which primary consideration was for
cost and ease of placement. The TACSS
materials are relatively new to the United
States but are reported to have been exten-
sively used in Japan and some European
countries. The materials were supplied by
the parent firm, which also provided per-
sonnel to aid in preparation of the admixes.
The types 020 and 025 are two-component
liquids, while the C^OO and ST are medium
gray powders.
FGD Sludges.—The FGD sludges selected
for the study consisted of one from an
eastern coal lime scrubbed process and the
other from an eastern coal limestone
scrubbed process. The sludges were con-
sidered to be representative of what could
be expected from most areas. Also, the two
sludge materials are included in a study of
chemical fixation of hazardous wastes being
conducted by the Environmental Effects
Laboratory (EEL), WES. This provided the
possibility that supporting of additional
274
-------
MATERIAL
TOTAL
LINER
T-16
TYPE
ELASTICIZED POLYOLEFIN
(30 MIL)
MANUFACTURER
THE GOODYEAR TIRE
AND RUBBER COMPANY
BLACK NEOPRENE-COATED REEVES BROTHERS, INC.
NYLON-REINFORCED
FABRIC
FIGURE 1. PREFABRICATED MEMBRANCE LINER MATERIALS
MATERIAL
TYPE
DC A-1295
DYNATECH
FORMULATION 267
UNIROYAL
AEROSPRAY 70
AC 40
SUCOAT
POLYVINYL ACETATE
NATURAL RUBBER
LATEX
MANUFACTURER
UNION CARBIDE
DYNATECH R&D CO.
NATURAL LATEX UNIROYAL, INC.
POLYVINYL ACETATE AMERICAN CYANAMID
ASPHALT CEMENT GLOBE ASPHALT
MOLTEN SULPHUR CHEVRON CHEMICAL CO.
FIGURE 2. SPRAY-ON LINER MATERIALS
MATERIAL
CEMENT
LIME
CEMENT WITH
LIME
TYPE/PERCENT
10%
10%
4% PORTLAND CEMENT,
6% LIME
MANUFACTURER
DUNDEE CEMENT CO.
WILLIAMS LIME CO.
M179
GUARTEC (UF)
POLYMER BENTONITE
BLEND 4%
A LIGHT GRAY
POWDER 4%
DOWELL DIVISION OF
DOW CHEMICAL
GENERAL MILLS
ASPHALTIC
CONCRETE
TACSS 020
TACSS 025
TACSS C400
TACSS ST
11%
1/2-
6%
6%
15%
15%
11% ASPHALTIC CEMENT, LOCAL CONTRACTORS
1/2-IN. AGGREGATE
TAKENAKA CO. (JAPAN)
DISTRIBUTED IN U. S.
BY AIR FRAME MFG. CO.
CALIFORNIA
FIGURE 3. ADMIX LINER MATERIALS
275
-------
information could be obtained from the
chemical fixation study.
Chemical Properties.—The chemical
properties of the raw sludges and subse-
quent leaohates were determined for the
parameters presented in Figure k. The
PARAMETER
ARSENIC (AS)
BERYLLIUM (BE)
CADMIUM (CD)
CHROMIUM (CR)
CYANIDE (CN)
COPPER (CU)
MERCURY (HG)
MAGNESIUM (MG)
MANGANESE (MM)
NICKEL (Nl)
PARAMETER
LEAD (PB)
SELENIUM (SF)
ZINC (ZN)
SULFITE (SO3)
SULFATE (SO4)
BORON (B)
CHLORIDE (CL)
VANADIUM (V)
NITRITE (N02)
NITRATE (NO3)
FIGURE 4. CHEMICAL PROPERTIES EVALUATED
FOR RAW SLUDGES AND LEACHATES
REGULATOR
GAGE
PRESSURE TYGON
PORT-^ XTUBiNG
AIR
PRESSURE
EDGE SEAL
LINER TEST
MATERIAL
FILTER
CHECK VALVE
MANIFOLD
DRAIN PORT
TUBINGX
PLASTIC
LEACHATE—tU CONTAINER
FIGURE 5. TYPICAL TEST DIAGRAM
list basically follows the parameters
being evaluated by EEL for the chemical
fixation study. The chemical analysis
performed was at the level of sensitivity
of that obtainable by the flame AA spec-
trophotometry technique.
Test Cells.—The test cell developed
for exposure of the liner materials to
the FGD sludge is shown in Figure 5. The
cylindrical portion is common PVC pipe
with a wall thickness of 0.3T5 in.
(0.953 cm) and an LD of 11.813 in.
(30.001+ cm). Flanges 1 in. (2.51* cm)
thick were fixed to the top and bottom
of the cylinder. The base, also PVC, is
2.125 in. (5.398 cm) thick and 15 in.
(38.1 cm) square. The base is tapped at
the center to accommodate the drain port,
and a special recess provided to house a
6 in. (15.2k cm) in diameter and 0.25-in.
(0.635-cm) thick porous plastic filter.
The top, also, PVC, is 15 in. (38.1 cm)
in diameter and 0.75 in. (1.905 cm) thick
with a tap provided for pressure attach-
ments. The top, flanges, and base are
drilled to accommodate 0.375-in. (0.953-cm)
bolts for connection purposes. An assem-
bled test cell is shown in Figure 6.
The cells were pressurized through
the top cover by compressed air, passing
through a system of regulators, gages, and
FIGURE 6. TYPICAL ASSEMBLED TEST CELL
276
-------
check valves, and by a manifold to simulate
the desired depth of sludge. The bottom
drain port was connected to a plastic con-
tainer for holding the leachate.
TEST PROCEDURES
The methods for constructing pond
liners for field installations were the
primary consideration in developing test
procedures for the study. In this re-
spect , the soil types encountered would
be of major importance; therefore, the
inclusion of soil material to be used in
conjunction with the liners was a neces-
sary measure.
The soil used with the membrane and
spray-on liners and the asphaltic con-
crete was a nonplastic silty sand with
good drainage characteristics. However,
the soil used for the admix liners was
a slightly plastic clayey silt, which was
readily adaptable for chemical stabiliza-
tion. Both soil types were locally avail-
able but widespread in occurrence and are
considered representative of materials
that would be encountered at disposal
sites.
The sequence of constructing the test
(exposure) cells consisted of assembly of
the base plate and cylindrical part of the
cell. Next, a silicone sealant was applied
at the interface of the two parts, and the
connecting bolts were secured. The filter
was then placed in the recess.
The admix liners were prepared based
on previously determined optimum mixtures
(Figure 3). The material was placed in
the cell and compacted to a thickness of
6 in. (15.2k cm). The thickness was con-
sidered to be the minimum that construction
equipment would normally be capable of
placing in a field installation. The excep-
tion for an admix liner was the asphaltic
concrete that was premolded to a 2-in.
(5.08-cm) thickness and supported by U in.
(I0.l6 cm) of silty sand previously com-
pacted at optimum moisture and density.
The silty sand that was used as a base
for the membrane and spray-on liners was
compacted at optimum moisture and density
in the cells to a depth of 6 in. (15.2U cm).
The membranes were prepared with a seam or
joint along the middle of the specimen, cut
to fit snugly in the cylinder, and placed
on the sand base. The spray-on materials
were measured to provide a placement rate
of 0.75 gal/yd^ (3.1* fc/m2), poured over the
sand base, and smoothed to provide a uni-
form thickness. The materials were then al-
lowed to cure for the required time period.
Following installation of the liners,
a silicone sealant, GE 1200, and primer,
No. 315^, were applied around the periphery
of the liner and on the cell wall. The
result was a triangular-shaped seal that
extended approximately 1 in. (2.5U cm) out
on the liner and also up on the cell wall.
The sealant was then allowed to cure for
the required time.
The FGD sludge was thoroughly mixed
until a liquid state was attained. Approxi-
mately U gal (15.1 A) of sludge was placed
in each cell.
The tops were placed on the cells for
which the silicone sealant was again used
at the interface with the cylinder and
secured with bolts. The cells were placed
on shelves in an area provided with con-
stant temperature and humidity (Figure 7).
FIGURE 7. TEST CELL HOLDING AREA
277
-------
The appurtenant equipment for leachate
collection and pressurization vas connected
to complete the assembly.
A total of 72 test cells were
assembled. This provided 36 cells with
the two FGD sludges and 18 liner mate-
rials with each sludge for a 12-month
exposure period and the remaining 36 cells
for 24-month exposure.
RESULTS
The study plan required that physical
tests be conducted on liner material
specimens prior to their exposure, and
a chemical analysis be made of the raw
sludges.
The 36 cells designated for the
12-month exposure period were dismantled.
Then similar physical tests were con-
ducted, and observations made of the
liner materials. Also, chemical analyses
were made for the leachates collected
during the 12-month exposure period. The
results presented herein will, of neces-
sity, be limited to selected data and
observations because of the massive amount
of data collected.
As the cells were dismantled, it was
readily apparent that the solids in the
sludge had settled and separated from the
liquid portion. The solids were black in
color and had caused discoloration of the
liner surface in most of the cells. Since
the solids were extremely difficult to
remove from the cell, considerable caution
was observed to prevent damage to the liner
materials. In some of the cells, failure
of the interior seal, even though minute,
had occurred and allowed complete discharge
of the liquid portion of the sludges. The
admix materials in which ML79 and Guartec
had been used as additives were almost
completely disintegrated. The reason for
the disintegration has not yet been deter-
mined, but it is assumed to be the result
of a chemical reaction of the additives
with the sludges.
The data considered to be the most
pertinent and meaningful at this stage
of the study were the chemical analyses
of the collected leachate as compared
with that of the raw sludge, especially
the solid portion. It was recognized as
the study progressed that the concentration
of the heavy metals in the collected
leachate attenuated considerably, usually
by several orders of magnitude as compared
with the raw sludge solids. However, the
concentrations determined for the liquid
portion of the raw sludge were approxi-
mately equal to that of the leachate.
In an effort to determine the cause
of such attenuation, two cells were assem-
bled containing 6 in. (15.2U cm) of com-
pacted silty sand without any type of
liner. Sludge was added, one type in one
cell and one in the other, and leachate
collected as for the lined cells. A
chemical analysis of the leachate from
the two "blank" cells indicated that the
concentration of heavy metals was about
the same for the leachate collected from
lined cells and the liquid portion of the
raw sludge.
A comparison of the chemical analysis
of the leachate from five selected liner
materials and the leachate from the "blank"
cells with the raw sludge solids is shown
in Figure 8 for one sludge type and in
Figure 9 for the other sludge type. The
particular liners were selected because of
the low permeability exhibited during the
exposure period. It should be noted that
some of the chemical parameters, i.e.,
magnesium, manganese, and nitrate/nitrite,
have higher concentrations than that of
the raw sludge. This probably can be
attributed to additional accumulation of
the particular elements from either the
soil or liner materials.
SUMMARY
Although the data presented herein
are limited to only a small portion of
data collected, it is considered to be of
significance if only indicating the
attenuation of the concentration of heavy
metals found in the leachate liquid. This
could be construed that a rather high con-
centration is being retained in the sludge
solids or in the liner materials and
underlying soil. The continuation of the
study through the 2l*-month exposure period
may provide some insight for determining
such a reaction. A chemical analysis of
the solid portion of the sludge material
retained in the test cells and of the
underlying soils and liner materials may
possibly clarify the matter.
278
-------
CHEMICAL ANALYSIS, SLUDGE A
Sludge solids and leachate from unlined and select lined test cells, mg/1
1000.000
1DO.OOO
5 10.000
1.000
0.100
0.010
o.ooi
Raw Sludge Solids
O Unlined Silty Sand
O T15
Uniroyal
Aerospray 70
• AC40
Sucoat
As T Be YCd I Cr | Cn * Cu | Hg I Mg | Mn | Ni | Pb | Se J Zn ( SOj ] S041 B | Cl | V |N02
CHEMICALS ANALYZED
FIGURE 8. COMPARISON OF HEAVY METAL CONCENTRATION IN RAW SLUDGE A AND LEACHATE
279
-------
CHEMICAL ANALYSIS, SLUDGE B
Sludge solids and leachate from unlined and select lined test cells, mg/1
1000.000
100.000
10.000
•z 1.000
1
0.100
0.010
0.001
LEGEND
Raw Sludge Solids
O Unlined Silty Sand
O T16
A Uniroyal
D Aerospray 70
• AC40
m Sucoat
As Y Be fCd T Cr I Cn f Cu T Hg | Kg | Mn I Ni | fb \ S.e I Zn I S«3|S04 I B I O | V | N02| Nttj
CHEMICALS ANALYZED
FIGURE 9. COMPARISON OF HEAVY METAL CONCENTRATION IN RAW SLUDGE B AND LEACHATE
280
-------
An interim report describing the study Note: The citation of trade names listed
in detail is being published. The report in this paper does not constitute
will present all the physical data and an official endorsement or approval
chemical analyses plus a few of the fail- of the use of such commercial
ures incurred. products.
281
-------
USE OF LIMESTONE TO LIMIT CONTAMINANT MOVEMENT FROM LANDFILLS
Wallace H. Fuller and Juan Artiola
Soils, Water and Engineering
The University of Arizona
Tucson, AZ 85721
ABSTRACT
Crushed limestone is an effective low-cost landfill liner aid in the migration control
of certain heavy metals from solid-waste leachates. Agricultural limestone is particularly
attractive as a liner because of its (a) wide geographic distribution, (b) particle sizes,
and (c) relatively low cost. Limestone barriers affect the mobility of hazardous constitu-
ents by (a) adsorbing metal ions directly, (b) forming less soluble calcium and carbonate
compounds, and/or (c) raising the pH level of the leachate as it passes through the lime-
stone. Chromium changes solubility through electron change (i.e. Cr"6 to Cr"3) at differ-
ent pH values when suitable electron donors are available. Although the limestone barrier
slowed the migration rate of all 12 metals studied, limestone was more effective in reten-
tion of some metals than others. Practical use of crushed limestone as liner material for
landfills as an aid in reducing the migration rate of potentially toxic metal pollutants
from leachates is suggested.
INTRODUCTION
The successes of Gehm (1944), Charmbury
and Maneval (1967), Jones and Ruggeri (.1969),
Stoddard (1973), and Pearson and McDonnell
(1975) and others, in the use of crushed or
powdered limestone for pollution abatement
management of acid mine waste provides en-
couragement for the use of similar materials
for control of polluting constituents of
other liquid wastes, such as landfill leach-
ates. Almost all of the limestone research
effort for pollution control before the
early 1960's was limited to acid mine-waste
management programs. During the 1970's,
involvement of limestone in pollution con-
trol has rapidly expanded to fields of flue
gas scrubbing. Most of this relates to de-
sulfurization (Nannen, West, and Kreith,
1974). Lime and lime slurries are effective
in the neutralization, adsorption, and pre-
cipitation of sulfur oxides and certain
heavy metals. Land and lagoon disposal of
the spent slurries still is required.
Limestone encasement of industrial
sludges containing high concentrations of
toxic metals such as has been suggested by
Loughry (1972) seems to be a good program
for certain sludge-like wastes. An encasement
could be designed in soil trenches to retain
hazardous polluting metals for a long period
of time.
Using lime to minimize problems with
solid waste landfill leachate, however, in-
volves another requirement. The liquid
phase should be stripped of as much of its
contaminating constituents as possible be-
fore it is allowed to gravitate to lower
depths where it might threaten underground
water sources or enter the food chain.
One method of using lime in landfill
construction would be to incorporate it into
the soil at the base of the excavation. The
literature on lime-soil reactions is exten-
sive and varied in topics, as reviewed by
Herrin and Mitchell (I960), Penrose and
Holubec (1973) and Thompson (1964). These
and other studies compiled by Jarrett (1977),
however, advise against mixing lime with
soils as the most desirable use of limestone
for soil modification in the disposal of
282
-------
solid wastes on land where a "leaky-tub"
effect is required. Lime mixed with soil is
an effective stabilizing agent. Cation ex-
change, flocculation, and agglomeration may
take place but they are not the basic lime-
soil stabilizing reactions. Cementation
that takes place when lime and carbon di-
oxide interact and pozzolanic reactions
(Thompson, 1964) between soil silica and/or
alumina and lime are generally regarded as
the major sources of stabilization of the
lime-soil aggregate. Since solid waste
landfills that are to use the contaminant
retention capacity of the soil will require
a leaky bottom to avoid overflowing like a
bathtub, highly compacted or cemented soils
would be unsatisfactory encasement or liner
structures. Moreover, dissolving of the
cemented mosaics by acid activity of leach-
ate would result in opening up the sur-
rounding site material to stabilized cracks
and fissures through which leachates could
flow freely out of the landfill without
first reacting with either the soil or lime
liner. Another disadvantage of an intimate
association between soil and limestone in
humid climates is that the acid of the soil
would react to consume it directly through
solubilization and neutralization. Lime-
stone has a relatively short life when used
to counteract the acidity of agricultural
soils. Necessary reapplications are rela-
tively frequent, compared to the lifetime
of a landfill.
Quantitative data evaluating the influ-
ence of limestone as a soil liner for land-
fills on the migration rates of potentially
toxic pollutants of municipal solid waste
are not available, except to a limited ex-
tent in a preliminary report by Fuller et
al. (1977). Research was undertaken, there-
fore, to quantify more extensively the ef-
fectiveness of a crushed agricultural lime-
stone layered over soil for controlling the
migration rate of solutions containing se-
lected trace and heavy metals (Be, Cr, Cd,
Fe, Ni, and Zn) and total organic carbon
(TOC) from municipal solid waste landfill
leachate.
MATERIALS
SOILS
Eleven soils representing 7 of the
major orders in the U.S. were collected be-
low the organic-laden top soil. Although
the soils were not collected entirely at
depths as great as many landfills, the
selection provides an opportunity for
studying the effectiveness (in retarding
contaminant movement) of limestone layered
over soils of varying physical and chemical
characteristics. If others are working with
a different soil or soil-like material, they
may determine the characteristics of the ma-
terial and then select the most similar to
the soils in this study as a base for esti-
mating how their material will retain con-
taminants when limestone is used with it.
Some general characteristics of the
soil collected are shown in Table 1. The
clay contents range from 4 to 61%. The clay
minerals of the less than 2 micron separate
of the 11 soils varied widely from largely
montmorillonite-type in Anthony s.l. and
Chalmers si.c.l. to largely kaolinite-type
in Davidson and Molokai clays and Wagram 1.
s. The pH values range from 4.2 for the
Ultisol, Wagram l.s. to 7.8 for the alkaline
Aridisols, Anthony s.l. and Mohave (Ca) c.l.
Some other characteristics of the soils as
packed in the columns are described in Table
2 and in an earlier published report by
Fuller (1977).
MUNICIPAL LANDFILL LEACHATE II*
The leachate used as a vehicle for
carrying the trace elements Be, Cd, Cr, Fe,
Ni, and Zn and total organic carbon (TOC)
came from a simulated municipal solid waste
landfill. Leachate characteristics are re-
ported in Table 3. The trace elements were
"spiked" at concentrations of 100 ppm for
all elements except iron. The Fe research
was done on unspiked leachate containing
variable levels of Fe up to 750 ppm as given
in Table 3. Such concentrations were used
for the convenience in identifying the at-
tenuation effect of the limestone; the indi-
vidual elements differ in their rate of
migration through the limestone and at lower
concentrations the rates of migration were
too slow to allow completion of the study in
a reasonable length of time.
LINER MATERIAL
The crushed limestone is a commercial
product from Cedar Bluff, Kentucky, commonly
used for agricultural soil applications.
Particle size analysis of this 98% pure
Two other leachates, I and III, were used
in another study (Fuller, 1978); the report
of that study may be available before this
paper is published.
283
-------
TABLE 1. GENERAL CHARACTERISTICS OF THE SOILS
Soil
Series
Davidson
Molokai
Nicholson
Fanno
Mohave
(Ca)
Ava
Chalmers
Anthony
Mohave
Kalkaska
Wag ram
Soil
Order
Ultisol
Oxisol
Alfisol
Alfisol
Aridiso]
Alfisol
Mollisol
Entisol
Aridisol
Spodosol
Ultisol
Soil
Paste
pH
6.2
6.2
6.7
7.0
7.8
4.5
6.6
7.8
7.3
4.7
4.2
Cation
Exch
Capac. ,
meq/lOOg
9
14
37
33
12
19
26
6
10
10
2
Elec.
Cond.of
Extract,
ymhos/cm
169
1262
176
392
510
157
288
328
615
237
225
Column
Bulk
Density,
g/cc
1.89
1.44
1.53
1.48
1.54
1.45
1.60
2.07
1.78
1.53
1.89
sand
19
23
3
35
32
10
7
71
52
91
88
Silt
to
20
25
47
19
28
60
58
14
37
4
8
clay
61
52
49
46
40
31
35
15
11
5
4
Texture
Class
clay
clay
silty
clay
clay
clay
loam
silty
clay
loam
silty
clay
loam
sandy
loam
sandy
loam
sand
loamy
sand
Predominant *
Clay Minerals
Kaolinite
Kaolinite,
gibbsite
Vermiculite
Montmorillonite,
mica
Mica,
montmorillonite
Vcrmiculite,
kaolinite
Montmorillonite,
vermiculite
Montmorillonite,
mica
Mica, kaolinite
Chlorite,
kaolinite
Kaolinite,
chlorite
Listed in order of dominance
-------
TABLE 2. SELECTED CHARACTERISTICS OF THE SOIL IN THE COLUMN STUDY WITH MUNICIPAL LANDFILL
LEACHATE II.
Soil
Davidson c.
Molokai c.
Nicholson si .c.
Fanno c.
Mohave (Ca) c.l.
Chalmers si. c.l.
Ava si. c.l.
Anthony s.l.
Mohave s.l .
Kalkaska s.
Wagram 1 .s.
Clay
Content,
%
61
52
49
46
40
35
31
15
11
5
4
Bulk
Density
of Column
1.38
1.35
1.38
1.42
1.53
1.42
1.41
1.73
1.67
1.54
1.75
Pore
Volume,
ml
123
111
119
125
115
120
123
93
94
104
97
Soil,
grams
355
348
357
367
394
366
363
446
431
398
452
Total m2
Surface
Area
18,212
23,420
43,019
44,811
50,235
45,970
22,325
8,831
16,507
3,542
3,616
Porosity
.476
.429
.460
.484
.446
.467
.478
.360
.365
.404
.378
TABLE 3. SOME CONSTITUENTS OF THE NATURAL MUNICIPAL SOLID-WASTE LANDFILL LEACHATE II USED
IN LIMESTONE RESEARCH.
pH EC TC TOC Ca Mg Na K Cd Co Fe Mn N1 Zn
mmhos/cm ppm
5.4 8.8
7225 7007 621 116 320 641 .05 .46 753 9.8 .40 7.6
appears in Table 4 along with two
other limestones for comparative purposes.
Also see Penrose and Holubec (1973) for
other characteristics of limestones.
BATCH STUDIES
A series of "batch" studies was con-
ducted prior to the more definitive column
studies. The batch technique is simpler
and less, expensive so it was used initially
to determine maximum adsorption precipita-
tion interaction between selected trace
elements and limestone. A known weight of
limestone was shaken with a known volume of
landfill leachate for predetermined time
intervals under anaerobic conditions and
C02 tension followed by centrifugation and
analysis of the supernatant using the con-
ventional atomic absorption flame emission.
Total organic carbon was determined by a
Beckman 915A Model Total Organic Carbon
Analyzer.
SOIL COLUMNS
The details of the soil-column proced-
ure have been described by Korte et al.
(1976), and by Fuller (1977). In brief,
soil was compacted to a uniform density in
10-cm lengths of 5-cm PVC pipe (Table 2).
The top 2 cm of the column consisted of the
liner limestone to be evaluated. Some col-
umns were prepared with 5-cm of crushed
limestone but movement of the trace element
was so slow that this thickness was aband-
oned in favor of the 2-cm thickness. To
evaluate the effect of the liner material
independently of the soil, columns were also
prepared in which acid-washed quartz sand
replaced the soil.
285
-------
TABLE 4. PARTICLE SIZE DISTRIBUTION OF AGRICULTURAL LIMESTONE FROM THREE
REPRESENTATIVE KENTUCKY SOURCES
Sieve Size Classes
*
-2.5 to -2.0
-2.0 to -1.5
-1.5 to -1
-1 to -0.5
-0.5 to 0
0 to 0.5
0.5 to 1
1 to 1.5
1.5 to 2.0
2.0 to 2.5
2.5 to 3.0
3.0 to 3.5
3.5 to 4.0
nm
- 5.66
5.66-2.80
2.80-2.00
2.00-1.40
1.40-1.00
1.00-0.71
C. 71-0. 50
0.50-0.355
0.355-0.250
0.250-0.180
0.180-0.125
0.125-0.090
0.090-0.063
0.06-0.050
< 0.050
USDfr
gravel
_^___~-^
very coarse
sand
coarse
sand
medium
sand
fine
sand
very fine
sand
silt + clay
Limestone Source
% of Size Separation
Lexington
0.89
17.3
13.2
15.2
11.4
10.2
7.5
8.3
6.0
3.0
2.4
1.5
1.0
0.79
1.22
Cedar*
Bluff
0
5.1
4.4
6.8
7.8
8.3
9.4
12.5
10.2
6.7
6.3
4.8
6.0
1.9
9.9
Gibsonburg
0
0
0
0
0
0
0
0.3
0.4
1.4
6.8
13.3
13.5
24.7
39.5
Limestone source used in studies reported here.
The solution carrying the spiked trace fill leachate, Figures 1 and 2. The differ-
element being studied was in all cases ence between attenuation in limestone-lined
municipal solid waste landfill leachate and unlined soil can be even much more dra-
maintained under anaerobic conditions using matic than these figures show, since the
CO?. The solution was passed through the effluent concentration (C) of Fe in lime-
column until the effluent concentration of stone-lined soils seldom reached that of the
the spiked trace element was equal to the influent (C0) before the experiments had to
concentration in the influent (breakthrough), be terminated. C/C0 = 1.0 was seldom
achieved where limestone was present. The
RESULTS pronounced prolongation of Fe attenuation
of the effluent/influent concentration ratio
NATURAL LEACHATE CONSTITUENTS between -v 0.8 to 1.0 is Illustrated at the
bottom of Figure 2, where C/C0 is plotted
Iron against number of pore volumes (pv) of solu-
tion in limestone columns alone.
Lining soils with crushed agricultural
limestone significantly delays the migration The cumulative total retention of Fe
of soluble Fe of municipal solid waste land- attributable to the limestone liner in soil
286
-------
o
o
o
o>
0)
0>
or
Mohove Sandy Loam
No Limestone
Limestone
i i i i i i i i i i i
Clay Loam
Mohove (Ca) Clay Loam
i i i i i 1 i i I i I
6 10 14 18 22 26 2 6 10
PORE VOLUME DISPLACEMENT: number
14
18 22
Figure 1. The effect of ground agricultural limestone layered over 6
soils on the retention of Fe from natural municipal solid-
waste landfill leachate.
also is much greater than that of unlined
soil, Table 5. Statistically, the F value
of 96.3 is significant at the 0.01 level.
There also is a significant difference be-
tween the effectiveness of different soils
for retaining leachate Fe. Those soils con-
taining the most clay held onto the Fe more
tightly than the sandier soils. For example,
over 90% of the leachate Fe was retained by
Davidson and Molokai clays by the time C/C0
reached 0.8, whereas the Kalkaska and Wagram
sands retained less than 36% of the Fe at
the same C/C0.
After the concentration of Fe in the
effluents from both lined and unlined soils
stabilized at some level ranging between
C/C0 of 0.8 to 1.0, deionized water was dis-
placed through the columns for an equal num-
ber of pvd's as the landfill leachate to
ascertain the retention stability of the
limestone- and soil-attenuated Fe. The re-
287
-------
1.0
.8
.6
.4
.2
No Limestone
Limestone
0
1.0
.8
.6
o>
-2 .21-
CO
Q o
o:
Chalmers Silty
Clay Loam
I I I I l I I I I l
10 14 18 22 26
10
14 18 22
Quartz Sand
2 cm Limestone (Rep.I)
2cm Limestone (Rep.2)
2 cm Limestone (Rep.3)
I I I I I I l I l l l I l l i l i i i i
0 2 4 6 8 10 12 14 16 18 20
PORE VOLUME DISPLACEMENT: number
Figure 2. The effect of ground agricultural limestone layered over
4 soils and pure quartz sand on the migration rate of Fe
from natural municipal solid waste landfill leachate.
suits as reported in Table 5 show that solid
waste leachate Fe is rather tightly held
against leaching action of water both by the
limestone and soil. Only a small percentage
of Fe (^ 2-7%) appeared in the effluents of
the many water displacements. Even after
leaching the lined and unlined columns of
soil with deionized water there was still a
highly significant difference in retention
of Fe favoring the limestone treatment. The
F value of 87.8 is well within the 0.01 lev-
el of significance.
The total retention of Fe correlates
significantly with the clay content of the
soils, Table 6. Davidson clay held more
than 5 times as much leachate Fe per gram of
soil as the sands, Kalkaska and Wagram. The
9-fold difference between Molokai clay and
the same sands is even more dramatic. Based
288
-------
TABLE 5. EFFECT OF CRUSHED LIMESTONE LINER ON THE RETENTION OF SOLUBLE IRON FROM NATURAL
MUNICIPAL SOLID WASTE LANDFILL LEACHATE PASSED THROUGH 11 SOILS.
Soil
Davidson c.
Molokai c.
Nicholson si.c.
Fanno c.
Mohave (Ca)
Chalmers
Ava si. c.l.
Anthony s.l.
Mohave s.l.
Kalkaska s.
Wagram l.s.
*
Pore
Volume
25
20
18
12
20
18
16
11
16
15
17
Soluble Landfi
With Limestone
Before
H20 Leach
94
97
82
82
83
74
70
51
44
36
34
11 Leachate Fe
Layer- 2 cm
Aftert
H20 Leach
91
96
79
80
82
71
68
48
41
33
31
Retained in Soi
1 Column - %
Without Limestone Layer
Before
HpO Leach
64
71
68
62
62
52
44
41
34
20
15
Aftert
H20 Leach
59
65
65
61
61
51
40
36
31
13
12
Pore volume refers to breakthrough where C/C0 = 1.0 or at peak C/C0 plateau between 0.8
and 1.0.
^Soil columns were leached with deionized water by equal number of pvd as used for landfill
leachate.
TABLE 6. ATTENUATION OF IRON IN NATURAL MUNICIPAL LANDFILL LEACHATE II PASSED THROUGH 11 .
SOILS AS RELATED TO UNIT WEIGHT AND SURFACE AREA.
Soil
Davidson c.
Molokai c.
Nicholson si.c.
Fanno c.
Mohave (Ca) c.l.
.Chalmers si. c.l.
Ava si. c.l.
Anthony s.l .
Mohave s.l.
Kalkaska s.
Wagram l.s.
Clay
Content
%
61
52
49
46
40
35
31
15
11
5
4
Pore
Volume
Displace-
ments*
25
20
18
12
20
18
16
11
16
15
17
Fe Attenuatedt
Total
Fe,
mg
922
1671
1024
899
1061
889
680
383
470
289
251
Fe
mg/g
Soil
2.597
4.803
2.868
2.450
2.962
2.448
1.874
1.491
1.090
0.545
0.555
Fe yg/mz
Surface
Area
51
71
24
20
21
20
30
28
28
61
69
Pore volume displacements necessary to achieve breakthrough (i.e. C/C0 = 1) or at peak
C/C0 plateau between 0.8 and 1.0.
'''mg Fe attenuated = mg Fe added - mg Fe in leachate effluent.
289
-------
on surface area, clay again appears to be a
very effective attenuator compared with sand
or silt. The relatively high values of Fe
held per unit of surface area for Kalkaska
and Wagram sands make it seem that these
sands have unusual surface activities. The
data in these cases are misleading, since
landfill leachate collected under C02 and
displaced through soil in this condition al-
ways will form some precipitation even along
the tubing approaching the soil column from
the reservoir. The precipitation becomes
more prominent at each successive leachate
displacement through the columns. Further-
more, columns of glass beads of relatively
small surface area have been observed to
collect a certain amount of Fe precipitation.
If it were possible to arrive at a quantita-
tive figure for this kind of precipitation
(unrelated to surface area), no doubt sands
would be shown to possess less active sur-
faces than clays per unit area. This is a
small point in view of the great difference
between total surface area per unit weight
of sand and clay.
Total Organic Carbon
The soluble organic carbon constituents
of natural municipal solid-waste landfill
leachate II were not retained to any signif-
icant extent by soil or crushed limestone,
Table 7. Limestone liners do not have
enough effect to be considered as a manage-
ment practice for TOC control in landfill
sites.
METALLIC CATIONS
Be. Cd. Mi and Zn
Because the landfill leachate must pass
through the limestone barrier prior to
contacting the soil, a batch-type experiment
was developed. The influence of time of
contact, ratio of leachate to water (simu-
lated rain water dilution effect), and pH
were studied between the limestone and
leachate alone. The resulting data should
represent maximum reactions between the two
systems. Figure 3 reveals the limestone/
leachate reactions with Cd were significant-
ly dependent on (a) contact time, (b) leach-
ate concentration, and (c) leachate pH
level. Sorting out the precise cause and
effect relationship between these three
factors and attenuation would be difficult
because of their interdependence. From a
standpoint of actual landfill leachate man-
agement alone, this batch-type research is
more academic than practical. However, it
serves to point out possible profitable de-
partures for the more practical research
program.
The effect of crushed limestone placed
as a liner barrier over soils on the reten-
tion of Be, Cd, Cr, Ni, and Zn contained in
municipal solid waste landfill leachate
varies both with the kind of soil and
metal, Table 8. (Breakthrough was not
achieved in every column.) For example, to
achieve the same C/C0 of Be, Davidson clay
(soil 1) required 30 pv in the presence of
limestone and only 10 in its absence. Cad-
mium, on the other hand, was less well re-
tained by limestone when associated with
Davidson clay than Be since the difference
between limed and unlimed soil was only 14
pv. The data make it quite clear that lime-
stone can have a highly significant effect
on the migration rate of the metals tested
even though only a thin liner of 2 cm was
used. Had there been sufficient time to
conduct column studies with limestone layers
thicker than a few centimeters, the barrier
effect of limestone might have been more
dramatic. For example, the 5 cm of lime-
stone used in preliminary experiments was
so effective in retaining the metals,
months went by with little evidence of the
test element appearing in the effluent.
The program for that thickness had to be
abandoned (with the concentrate leachate II
available) because no data were being gen-
erated for comparative purposes. The data
in Table 8, therefore, are relative and
will serve a practical purpose for liner
management if used with this understanding
(i.e., the effectiveness is directly related
to and proportional to the thickness of the
crushed limestone. Conceivably, such a
thickness may be used to keep any toxic
element from migrating for infinite time.
Practical application, however, may not
prove out so well if ideal conditions can't
be maintained.)
ANION-FORMING METAL
Chromium
The influence of limestone liner on Cr
retention is much greater than that for the
other metals tested, Table 8. This occurs
however, at lower pH values of the influent
leachate necessary to keep Cr in solution at
a constant concentration. Migration rate of
Cr through both the limestone-lined and un
lined soil columns was significantly greater
290
-------
TABLE 7 . EFFECT OF GROUND LIMESTONE LINER ON THE RETENTION OF SOLUBLE
ORGANIC CARBON (TOC) CONSTITUENTS FROM NATURAL MUNICIPAL SOLID
WASTE LANDFILL LEACHATE PASSING THORUGH SOIL AND SAND
Pore Volume
Displacement
Number
1
2
3
4
5
6
7
8
10
12
14
16
18
20
1
2
3
4
5
6
7
8
10
12
14
16
18
20
Davidson
4000
5300
6400
6500
6850
6700
6800
6800
6900
7000
7000
7000
--
—
Ava
4600
5100
6000
6800
6750
6700
6700
6800
7000
7200
6900
6900
Molokai
6400
6300
6700
6600
6500
6600
6200
6200
6100
6800
7000
7100
7100
7100
Anthony
7300
6800
6300
6700
7000
7100
6700
6800
6600
6400
7000
7300
7300
7300
TOC - ppm*
Soils
Nicholson
6400
5900
6300
6400
6300
6300
6400
6500
6900
7300
6900
6600
—
--
Mohave
7600
6400
6500
6600
6800
6800
6700
6400
6500
6600
7000
7400
7400
7400
Fanno
6800
6800
6600
6500
6600
6700
6600
6500
7300
7200
7100
—
—
—
Kalkaska
6800
6400
6900
7000
7000
6900
6800
6900
7300
7300
7300
7300
7300
Mohave(Ca)
7400
6300
6600
6600
7200
6900
6200
6400
6500
7200
7300
6700
6700
6700
Wagram
5550
6000
6900
6300
6600
6700
6800
6800
6800
7300
7600
7300
6800
Chalmers
6500
6600
5700
6000
6200
5800
5600
6100
—
--
—
—
—
--
Quartz sand
6800
6200
6400
6800
6900
7000
7000
7100
6900
6900
6900
6900
7000
7000
Natural influent leachate contained ^ 7000 ppm TOC
at pH 2.5 than a 4.0. The difference is less soluble Cr (Artiola and Fuller, 1978).
thought to be due to an e~ effect on the Limestone also contributes to this shift be-
proportion of CrIII to CrVI. Higher pH cause of its effect on pH.
levels favor a shift in concentration of
chromium from the more soluble Cr to the There appears to be a combined influ-
291
-------
1000
900
800
o>
o
700
o
o> 600
0
Ul
CD
CE
500
400
300
200
100
FINAL SOLUTION pH VALUES
\
.5 I 2 4 24 48 .5 I 2 4 24 48
.5 I 2 4 24 48 .5 I 2 4 24 48
CONTACT TIME - hours
\-0 3M hi l'3
RATIO OF LEACHATE TO WATER
o--\
Figure 3. The influence of time of contact, ratio of municipal landfill leachate to water,
and pH in the adsorption of Cd by agricultural limestone
(Cd concentration in all solutions at 100 ppm after water dilution).
ence on retention of Cr between the lime-
stone liner and soil not apparent when
leachate was passed through limestone and
soil independently, Table 9. At Cr C/C0 =
1.0, the difference between soil alone and
limestone alone for Anthony s.l., Kalkaska
s. and Wagram l.s. was 29, 37, and 37 pvd,
respectively. The difference for Mohave s.
1. is even greater. Explanation for the
reactions or mechanism involved is not
available at this time since this effect
occurred with all limestone-lined soils and
irrespective of spiked metal or constitu-
ents of unspiked natural municipal leachates.
LIMESTONE UTILIZATION
The length of time limestone liners
will remain active in landfills determines
their practicality for immobilizing or sta-
bilizing potentially hazardous pollutants.
A balance sheet was, therefore, maintained
to accurately account for the Ca in the in-
fluent and effluent of the soils, while
leachate displacement was going on. Since
the crushed agricultural limestone con-
tained 98% CaCOa, the Ca calculated as dif-
ference between column input and output
minus soil-Ca was converted back to lime-
292
-------
TABLE 8. RELATIVE EFFECT OF CRUSHED LIMESTONE LINER PLACED OVER SOIL ON'THE PREVENTION OF
HEAVY METALS. IN MUNICIPAL SOLID WASTE LANDFILL LEACHATE. FROM MIGRATING.
Pore
Volume
Displacement to Achieve the
With Limestone Layer with Soi
Element
Be
Cd
Cr(pH 2.5)
Cr(pH 4.0)
Fe
Ni
Zn
1*
30
30
35
28
31
40
30
2
29
35
42
49
31
35
25
3
35
40
85
125
35
30
30
4
30
38
88
144
36
35
32
5
30
35
86
111
30
28
25
1
6
25
35
65
109
25
24
30
Same C/C0
No limestone Layer, Soi
1
10
16
10
2
21
20
10
2
16
18
15
21
26
15
11
3
15
28
5
39
10
11
15
4
15
15
10
25
25
15
18
1 Only
5
18
18
10
17
10
7
12
6
10
6
6
15
10
10
16
Numbers refer to: 1 = Davidson c.; 2 = Ava si.c.l.; 3 = Anthony s.l.; 4 = Mohave s.l.;
5 = Kalkaska s.; 6 = Wagram l.s.
TABLE 9. EVALUATION OF THE EFFECTIVENESS OF AGRICULTURAL LIMESTONE ALONE AND LIMESTONE
LAYERED OVER SOILS FOR ATTENUATION OF CHROMIUM IN MUNICIPAL SOLID-WASTE LANDFILL
LEACHATE II*
Series
Davidson c.
Ava si.c.l.
Anthony s.l.
Mohave s.l .
Kalkaska s.
Wagram l.s.
Soil
Clay, %
61
31
15
11
5
4
PH
6.2
4.5
7.8
7.3
4.7
4.2
Ratio of
Cr in
Influent &
Effluent,
C/C0
0.13
0.38
1.00
0.76
1.00
1.00
Pore volume displacement number when
Cr concentration in effluent is at
C/Cn qi
Limestone
Alone
2
5
57
46
57
57
ven in Column Four
Soil
Alone
2
21
39
25
17
15
Soil and
Limestone
28
49
125
144
111
109
Leachate II had a pH value of 4.0, TOC of > 3000 and Fe = 300 ppm.
stone CaC03. The limestone liners were con-
sumed as a result of leachate interactions,
Table 10. The rate of loss depends
primarily on time and pH or acidity of the
landfill leachate. For example, approxi-
mately twice as much limestone was solubi-
lized at 90 pvd when the leachate pH was
2.5 as at 4.0. It took about 150 pvd of pH
4.0 leachate to cause the same loss of lime-
stone as 90 pvd with the pH 2.5 leachate.
With natural (unadjusted) leachate at pH
5.4 utilization of limestone was still less.
Averages for the 6 soils at 20 pvd was 19.1,
8.4, and 1.8 percent for the pH 2.5, 4.0,
and 5.4 leachates, respectively. The util-
ization of limestone alone in a 2-cm layer
placed between pure quartz sand layers by
the pH 4.5 leachate is shown in Table 11 to
be in between that solubilized at pH 4.0 and
5.4 reported in Table 10. The Ca in the
effluent decreased with time and the Fe in-
creased slightly under these conditions.
293
-------
TABLE 10. EFFECT OF MUNICIPAL LANDFILL LEACHATE pH ON THE LOSS OF LIMESTONE USED AS
LINERS OVER SOIL.
Pore Volume
Displacement
Number
1
10
20
30
40
50
60
70
80
90
1
10
20
30
40
50
60
70
80
90
100
no
120
130
140
1
2
3
4
6
8
10
12
14
16
18
20
Davidson
0.4
7.9
15.0
21.1
--
--
_-
__
_-
--
0.1
5.9
9.0
10.5
11.4
--
--
--
__
--
_-
—
—
--
--
0.3
0.6
0.8
0.9
1.1
1.3
1.5
1.6
1.7
1.8
1.9
2.0
Soi
Is
Mohave Ava Anthony
Limestone Loss - %
Leachate pH
0.6
4.1
24.3
19.8
25.0
29.5
33.5
36.8
39.1
39.9
Leachate pH
0
4.8
7.9
11.3
13.9
16.5
18.8
20.9
23.1
25.1
27.1
29.1
30.9
32.8
34.4
Leachate pH 5
0.5
1.0
1.2
1.3
0.6
1.8
2.0
2.2
2.3
2.4
2.5
2.6
*
Kalkaska
•
Wagram
adjusted to 2.5
0.1
14.8
24.0
31.5
—
—
--
--
--
—
adjusted
0.3
6.0
10.0
13.1
15.8
--
--
—
--
--
--
--
—
__
—
.4 (not
0
0
0
0
0.4
0.7
0.9
1.0
1.2
1.3
1.4
1.5
0.9
6.3
14.8
21.4
25.6
32.1
36.0
39.1
40.9
41.1
to 4.0*
0.4
2.9
7.6
10.6
13.9
17.0
20.0
23.1
26.0
29.0
32.0
34.8
37.5
39.9
42.1
adjusted)
0.3
0.5
0.8
0.9
1.2
1.5
1.7
1.8
1.9
2.0
2.1
2.2
0.5
5.4
16.5
22.8
28.9
34.4
38.8
41.5
43.5
45.1
0.3
3.9
7.6
11.3
15.1
18.3
21.9
25.0
28.4
31.5
34.3
36.6
38.6
39.9
2.5
0
0
0.1
0.2
0.3
0.5
0.6
0.8
0.9
1.0
1.0
1.1
0.5
8.8
20.1
26.0
30.9
34.4
31.3
41.0
43.1
44.2
0.4
4.8
8.1
11.3
14.0
16.9
19.6
22.3
25.0
27.6
30.1
32.5
35.2
37.4
40.2
0.1
0.3
0.4
0.5
0.6
0.8
1.0
1.0
1.1
1.2
1.3
1.4
pH was adjusted with HC1 to aid in evaluating leachate pH effects on limestone
utilization or loss.
Original leachate, pH = 5.4.
pH levels of 2.5 and 4.0 fall below those of
most municipal solid-waste landfill leach-
ates reported in the literature (Garland and
Mosher, 1974; and Chain and Dewalle, 1976).
However, the pH of some industrial leachates
may fall well below these values.
294
-------
TABLE 11. IRON AND CALCIUM CONTENT OF
EFFLUENT FROM LIMESTONE LINER-
QUARTZ SAND COLUMNS INFILTRATED
WITH MUNICIPAL LANDFILL LEACHATE
at pH 4.5 and the CONCURRENT
LIMESTONE LOSS.
Pore
Vol ume
Displace- Fe,
merit
1
9
19
28
38
48
60
ppm
1.5
5.0
6.5
4.0
8.0
7.0
9.0
Ca,
ppm
1250
1224
1225
1170
1100
1100
995
Limestone Lost
9
0.2
2.3
4.7
7.0
9.5
12.2
13.8
%
0.25
2.88
5.88
8.75
11.88
15.25
17.25
Average of 3 closely agreeing limestone
columns.
CONSIDERATION IN USE OF LIMESTONE LINERS
Research with soil columns and crushed
limestone has identified factors that must
be considered if a disposal operation is to
be successful in avoiding water contamina-
tion. At some ideal disposal sites, the
soil alone may restrict contaminant move-
ment sufficiently to prevent water contami-
nation. However, it is difficult to identi-
fy, with certainty, an ideal site and even
at ideal sites the use of limestone liners
will be required. For limestone use to be
successful, the following factors should be
considered:
1. Kind and depth of soil selected for
disposal site.
2. Characteristics of the leachate
generated or to be deposited.
3. Characteristics of the specific
potentially hazardous pollutants.
4. Peculiarities of limestone and its
reactions with leachate constituents.
5. The environment (climate, depth to
water table, etc.)
SOIL
Regardless of the characteristics of
the disposal site, inclusion of limestone
alone can slow the migration rate of heavy
metals through soil. Our experiments indi-
cate, however, that limestone varies con-
siderably in degree of effectiveness for
controlling migration of potentially haz-
ardous metals in leachates depending on the
way it is associated with soil as well as
the kind of soil involved. Soil texture is
one of these factors. Limestone appears to
be more effective with clayey than sandy
soils, provided of course the clay is per-
meable to water. Soils unusually high in
clay, however, may clog with degradation
constituents created in reactions with lime-
stone. This brings into focus the factor
of structure or pore space distribution in
soil. Despite the favorable reaction of
the divalent Ca ion (solubilized from the
added limestone) on soil structure, the re-
precipitation of Ca as CaCOa and/or other
reaction products at lower depths in the
soil remains a possibility to be considered.
An open-structured soil, therefore, is fav-
ored over a tightly structure soil as site
material.
Layering or lining soil is preferred to
mixing limestone with soil. Intimate mixing
results in soil stabilization or cementation
which at first will block the movement of
the solution from leaving the site and at a
later date encourage the formation of fixed
channels (holes) through which solutions can
move into the natural soil or geologic de-
bris without first having reacted with the
barrier material. Examples of this problem
are well documented. Where landfills have
been located over thick layers or natural
limestone deposits, the leachates follow
down cracks and holes to some lower depth
virtually unchanged. Moreover, mixing lime-
stone with soil, particularly in humid reg-
ions where soils are acid, results in the
loss of limestone to neutralization of the
highly buffered medium. The life of the
limestone, under such conditions, is greatly
shortened.
One of the encouraging results of the
limestone research is the discovery of a
synergic effect. Attenuation of the leach-
ate metals is greater when soil and lime-
stone (as a liner) are together than the sum
of attenuation of both when they are kept
separate. Such findings lend support to the
practice of lining limestone over soil rath-
er than using other methods of application.
It does not preclude, however, a possible
advantage of adding some limestone directly
295
-------
to landfills at different stages of
deposition of refuse, to neutralize the
acids. Additional research needs to be
undertaken to justify such a practice.
Limestone was observed in the batch studies
to cause some organic material as well as
inorganic elements to flocculate and pre-
cipitate.
LEACHATE
The single most important characteris-
tic of leachate that influences the rate of
metal migration through limestone-lined
soil is acidity (or pH). By reducing acid-
ity (i.e. raising the pH) migration rate
can be lessened several fold. Therefore,
any practice (such as aeration) that re-
sults in raising the pH toward neutrality
may be expected to enhance the effective-
ness of the limestone and prolong its life.
Leachates high in total organic carbon
constituents and soluble salts (or IONS)
possess a greater limestone requirement
than more dilute leachates because of their
known higher buffer capacity. From a prac-
tical standpoint limestone lining alone may
be effective as a barrier to undesirable
migration of metals from dilute, older
leachates. Fresh or highly concentrated
leachate, on the other hand, may be only
temporarily controlled by limestone lining.
Additional control measures may be needed
for full control. Exploratory soil column
studies on a laboratory scale may be neces-
sary to determine what level of leachate
concentration would break through the bar-
rier under the specific environmental con-
ditions of the disposal site (e.g. high
rainfall and acid soil conditions).
LIMESTONE
Variations in particle size distribu-
tion of crushed limestone, within the range
reported in Table 4 did not result in dif-
ferences in the retention of the elements
studied (Fuller et al., 1976). It was con-
cluded that particle size was not critical
as long as particles ranged between sizes
of gravel and coarse sand. Thick layers of
fine particles seriously clog and finally
compact into an impervious layer. There-
fore, coarser size grades should be uti-
lized particularly when (a) leachates are
high in soluble constituents (organic and
inorganic) and (b) when thick layers in
excess of 5 cm (2 inches) are required.
The quality of limestone varies from
place to place. Dolomitic limestone (mag-
nesium carbonate) is less effective than
calcium carbonate limestone. Magnesium car-
bonate reacts more slowly and often incom-
pletely with acids of the leachate and may
form slick, bulky colloids with organic and
inorganic constituents that clog and seal
the disposal lining. Silicates associated
with limestone also may cause infiltration
problems. Since these and other constitu-
ents often found deposited with limestone
may have an adverse effect on the quality of
lining, only high quality CaC03 limestone
should be used.
Thickness of limestone liners requires
a judgment decision. Because leachate com-
position varies so widely and because we
have been unable to quantitatively relate
leachate quality, liner thickness, and con-
taminant retention by mathematical equation,
no single threshold thickness for all land-
fills can be provided. Obviously the 2-cm
layer used in our experiments is too thin
for practical disposal operations. Neces-
sary traffic over such a thin layer would
disrupt continuity and provide escape chan-
nels through the intended barrier. A mini-
mum limit of thickness may be near 13 cm (5
in.) and a maximum near 30 cm (12 in.) de-
pending on particle size distribution.
Field experiments must determine the most
effective thickness under conditions at an
actual field disposal site. Excessive
depths of limestone lining may prove unsat-
isfactory because of compaction, cementation
and reactions that seal the site from solu-
tion Infiltration.
ELEMENT
Specific elements react differently
with limestone. Individual differences,
however, may be modified, even to the extent
of complete masking, by (a) high concentra-
tion of soluble constituents in the leach-
ates that compete for the same reactions and
adsorption sites and (b) dilution effect of
both the associated solid and liquid phases
of the system. Such anion elements as Cr
that undergo valency changes readily, re-
spond more acutely to limestone than the
cations (Be, Cd, Ni, Zn). From a manage-
ment standpoint, limestone use in Cr dis-
posal has a great chance of successful
control and therefore should be given top
consideration.
296
-------
ENVIRONMENT
ACKNOWLEDGMENT
Site selection is an all-important
part of land disposal management. Environ-
mental factors such as depth and nature of
soil and/or geologic debris, at the dis-
posal site, rainfall quantity and intensity,
seasonal temperature means, slope of land,
runoff and erosion hazard, and depth to
underground water must be assessed in re-
lation to their impact on the effectiveness
of the limestone liner use. Design of a
liner must relate to the disposal site en-
vironment.
SUMMARY
These laboratory studies indicate that
the use of crushed limestone linings on the
bottom of municipal solid waste and other
types of landfills will significantly re-
tard the movement of some contaminants into
and through underlying soils. The movement
of metallic cations such as Be, Cd, Fe, Ni,
and In and anions such as Cr in leachate is
retarded; movement of the total organic
carbon (TOC) fraction of municipal solid-
waste leachate is not retarded by limestone
linings. Characteristics of the underlying
soil affect the effectiveness of a lime-
stone liner. Increased protection from a
liner is greater in finer textured soils
(clays, loams, etc.) than in coarser tex-
tured soils (sands, sandy loams, etc.). The
relationship between leachate characteris-
tics, thickness of the limestone liner, soil
characteristics, and the degree to which
contaminant movement is retarded is complex;
a quantitative measure of this relationship
and of the effective life of a limestone
liner could not be developed. Additionally,
due to problems from compaction and plug-
ging, there may be operational limitations
(under field conditions) on the maximum
thickness of a limestone liner. No field
data are available to evaluate this problem.
Using the data in this paper (volume
and characteristics of leachate, limestone
and soil thickness, limestone and soil char-
acteristics) estimates of limestone thick-
ness required at a proposed landfill can be
made, provided information is available on
the soil and the expected leachate charac-
teristics at the site. Although inclusion
of limestone in landfill design will clearly
decrease certain types of environmental im-
pacts, such design estimates will be quali-
tative judgments and should not be expected
to provide an exact degree of protection.
This research was supported in part by
the U.S. Environmental Protection Agency,
Solid and Hazardous Waste Research Division,
Municipal Environmental Research Laboratory,
Cincinnati, OH. Grant No. R 803-988 and The
University of Arizona, Soil, Water and
Engineering Department. Arizona Agricultur-
al Experiment Station Paper No. 245.
REFERENCES
Artiola-Fortuny, J. and W.H. Fuller. 1978.
Effect of crushed limestone on chromium
attenuation in soil (in press). (1978)
Chain, E.S.K. and F.B. DeWalle. 1976. San-
itary landfill leachates and their
treatment. J. Environ. Eng. Division.
ASCE, 102:412-431 (1976).
Charmbury, H.B. and Maneval, D.R. 1967.
Operation Yellowbay, design and econom-
ics of a lime neutralization mine
drainage treatment plant. Preprint No.
67F35, Soc. Mining Enqr. AIME.
Fuller, W.H. 1977. Movement of selected
metals, asbestos, and cyanide in soil:
Application to waste disposal, EPA-600/
2-77-020, U.S. Environmental Protection
Agency, Cincinnati, OH (1977) 243 pp.
Fuller, W.H. 1978. Investigation of land-
fill leachate pollutant attenuation by
soils. EPA-600/0-000. U.S. Environ.
Protection Agency, Cincinnati, OH. 243
pp. (in press). (1978)
Fuller, W.H., C. McCarthy, B.A. Alesii, and
E. Niebla. 1976. Liners for disposal
sites to retard migration of pollutants.
In: Residual Management by Land Dis-
posal. Proc. of the Hazardous Waste
Research Symp., Feb. 2-4, 1976, Tucson,
AZ. W.H. Fuller, ed. EPA-600/9-76-015,
U.S. Environmental Protection Agency,
Cincinnati, OH. 280 pp. (1976)
Garland, G.A. and D.C. Mosher. 1975. Leach-
ate effects from improper land disposal.
Waste Age 6:42-48. (1975)
Gehm, H.W. 1944. Neutralization of acid
wastes with an up-flow expanded lime-
stone bed. Sewage Works J. 16:104-120
(1944).
Herrin, M. and H. Mitchel. 1961. Lime-soil
297
-------
mixtures.
(1961).
Highway Res. Bd. Bull. 304
Jarrett, R.E. 1966. Porous limestone bar-
riers for neutralization of acid
streams. Thesis. The Pennsylvania
State University Library, University
Park, PA. (1966).
Jones, J.B. and Ruggeri, S. 1969. Use of
in situ precipitation techniques—pol-
TutTon control survey. Investment
Dealer's Digest. 24-26 (1969T!
Korte, N.E., J. Skopp, W.H. Fuller, E.E.
Niebla, and B.A. Alesii. 1976. Trace
element movement in soils: Influence
of soil physical and chemical proper-
ties. Soil Sci. 122:350-359 (1976).
Loughry, F.G. 1972. Soil science and geo-
logical principles applied to disposal
of metallic sludges. Proc. First Tech.
Conf. Environmental Conservation. The
Pennsylvania State University.Uni-
versity Park, PA 16802 (1972).
Nannen, L.W., R.E. West, and F. Kreith.
1974. Removal of S02 from low sulfur
coal combustion gases by limestone
scrubbing. Air Pollut. Contr. Assoc.
J. 24:29-39 (1974).
Pearson, F.H., and A.S. McDonnell. 1975.
Limestone barrier to neutralize acidic
streams. J. Environ. Engr. Div. ASCE
Vol. 101, No. EE3:425-439 (1975).
Penrose, R.G., Jr., and Igor Holubec. 1973.
Laboratory study of selfseal ing lime-
stone plugs for mine openings. EPA-
670/2-73-081. ORD, U.S. Environmental
Protection Agency, Washington, D.C.
20460. 217 pp. (1973)
Stoddard, C.K. 1963. Abatement of mine
drainage pollution by underground pre-
cipitation. U.S. Environmental Pro-
tection Agency Report No. EPA-670/2-
73-092 m (Oct. 1973).
Thompson, M.R. 1964. Lime-reactivity of
Illinois soils as it relates to com-
pressive strength. Thesis^ University
of Illinois Library, Univ. of Illinois,
Urbana, IL (1964).
298
-------
PILOT SCALE EVALUATION OF DESIGN PARAMETERS FOR SORBENT
TREATMENT OF INDUSTRIAL SLUDGE LEACHATES
P. CHAN, J. LISKOWITZ, A. J. PERNA,
R. TRATTNER, AND M. SHEIH
ENVIRONMENTAL INSTRUMENTATION SYSTEMS LABORATORY
NEW JERSEY INSTITUTE OF TECHNOLOGY
Newark, New Jersey 07102
ABSTRACT
Based on laboratory lysimeter studies an outdoor pilot scale sized lysimeter was de-
signed and utilized for the sorbent treatment of leachate resulting from a calcium fluoride
sludge.
Results of the laboratory lysimeter experiments indicate sorbent capacity is a func-
tion of the pH and concentration of the particular contaminant in the leachate with the
volume of leachate that can be treated with maximum removal being regulated by the velocity
through the bed. Velocity of the leachate can be influenced by a number of factors such as
bed height, particle size and addition of inert material. Effective treatment of calcium,
copper, cyanide, fluoride and COD was achieved.
INTRODUCTION
The purpose of most waste treatment
processes is to convert the pollutants into
a gas, such as carbon dioxide, or into a
solid which can be readily removed from the
waste streams. In the latter instance, the
end product is a sludge which must be dis-
posed of in an environmentally acceptable
manner. At the present time, ocean dumping
and landfill are the two most common methods
being utilized for the disposal of this
sludge. However, ocean dumping is to be
banned by congressional action in 1981; thus,
landfill will be the remaining receptacle
for these sludges.
Disposal of sludges by landfill ing can
lead to heavy metal, toxic anion and organic
contamination of surface and ground waters
by leachate which results from ground water
seepage or rainwater filtration through
these sludges. In general, this contamina-
tion can be minimized by 1) respective loca-
tion of the landfill so that the natural
clay components in the soil can attenuate
the pollutants in the leachate or 2) lining
the landfill site with an impermeable mem-
brane to prevent groundwater intrusion into
the site, collecting the leachate resulting
from rainwater filtration through the sludge
and treating it by conventional physical-
chemical treatment (i.e., activated carbon
and alumina). While this latter approach
i s expensi ve because of the attendent cost re-
quired for the regeneration of these refin-
ed sorbents , i t provi des for treatment of the
leachate before it leaves the landfill site
and minimizes the dependence on natural
attenuation in soil which is, in general,
considered unreliable.
The identification and utilization of
inexpensive clay and waste product sorbents
which could be used to replace the refined
sorbents (i.e., activated carbon and alumi-
na) would eliminate the need for sorbent
regeneration facilities and provide a sys-
299
-------
tern for treatment of the leachate at the
site that is economically attractive.
A number of investigators have shown
that in leachates and waste streams, organ-
ics (e.g., phenols1, surfactants2, pyri-
dine3'1*, other organics characterized by
COD5), pesticides (e.g., parathion6, DDT,
dieldrin and heptachlor7), herbicides (e.g.,
paraquet8), heavy metals (e.g., lead, cad-
mium, mercury, zinc, mangenese, copper9'10
11>12)13 and toxic anions (e.g., chromium
VI, arsenic and selenium11*) are attenuated
by clay minerals, soils and waste products15.
However, none, have explored the use of fly
ash-clay sorbent mixtures to define the most
effective combination for the removal of
heavy metals, toxic anions and organics from
leachate, identified engineering parameters
required for design of the treatment system
using these sorbent combinations and con-
trolled the background leaching of heavy
metals encountered when fly ash, illite and
kaolinite16 are used as sorbents.
This investigation has been concerned
overall with: (1) defining the clay-fly ash
combinations which are most effective in re-
moving the heavy metals, toxic anions, and
organics present in leachates originating
from industrial sludges; (2) examining the
effect of such factors as pH of the leachate
and velocity of leachate through the soVbent
on removal of contaminants; and (3) estab-
lishing a design approach for this treatment.
Item (1) and part of item (2) have been in-
vestigated earlier. This involved a labora-
tory evaluation of the natural sorbents,
vermiculite, illite, kaolinite, acidic and
basic fly ashes, and zeolite for the removal
of cations, anions and organics under flow-
ing conditions from an acidic petroleum
sludge leachate, a neutral calcium fluoride
sludge leachate, and an alkaline metal fin-
ishing sludge leachate. Activated alumina
and activated carbon were involved in this
study for comparison purposes since these
materials are commonly used for the treat-
ment of industrial waste streams.
The results of this investigation indi-
cate that rather than a single sorbent, a
combination of acidic and basic sorbents
(which induce acidic and alkaline conditions
respectively into the leachate) is required
in a layered system for the removal of all
the measurable contaminants present in the
leachates. These are illite, vermiculite
and zeolite for the acidic leachate, illite,
acidic fly ash and zeolite for the neutral
leachate and illite, kaolinite and zeolite
for the alkaline leachate based upon a com-
parison of their sorbent capacity (total
amount of specific cations, anions or or-
ganics removed by a gram of each sorbent).
The sorbent capacities exhibited by the
natural sorbents for the removal of the
cations, anions and organics in the lea-
chates are comparable to those exhibited by
the refined sorbents (i.e., activated car-
bon and alumina). Also, pH control of the
leachates by combined use of the acidic and
basic sorbents is essential for effective
treatment. The removal of the anions in
the leachates are favored by acidic condi-
tions, the cations by basic conditions and
the organics either by acidic or basic con-
ditions.
It was the purpose of this study to
further define the process controlling
parameters using laboratory lysimeters and
to carry out a pilot scale study for the
treatment of calcium fluoride sludge lea-
chate using the fly ash-clay sorbent combi-
nations. The calcium fluoride sludge lea-
chate was selected for the pilot study
since the levels of fluoride in this lea-
chate were in the 5 to 20 ppm range and
there is no inexpensive treatment process
presently available to reduce the fluoride
concentration to 1 mg/1 or less.
The aims of the laboratory lysimeter
studies which were conducted were: (1)
identify the most effective sorbent combi-
nations for the treatment of the calcium
fluoride sludge leachate; (2) determine
that velocity of leachate through the sor-
bent bed which provides the most effective
removal of fluoride; and (3) to install
outdoor lysimeters and evaluate the most
promising sorbent combination, based on the
laboratory lysimeter studies.
EXPERIMENTAL
The preparation of the industrial
sludge leachates, as well as the analytical
and laboratory lysimeter procedures used in
this study have been discussed previous-
ly16' 17.
In order to evaluate the use of the
clay-fly ash combination for the treatment
of industrial sludge leachates on a pilot
scale, two large vertical lysimeters were
set up outdoors. This outdoor system con-
sisted of an agitator, filtration column,
300
-------
storage, and constant head lysimeter. (See
Figure 1).
Calcium fluoride sludge was collected
from the same source over a period of a year
to take advantage of changes in production
schedules and processes which could lead to
compositional changes in the leachate. In
this manner, the effect of compositional
changes on the removal of the cations, anions
and organics by the natural clay-fly ash
sorbent combinations could be studied.
The preparation of sludge leachate for
the outdoor study was as follows: a sample
of each batch of sludge was dried at 103°C
to constant weight in order to determine its
moisture content. The unaltered sludge was
then mixed with water in a ratio of 2.5 ml
water per gram of dried sludge and mechani-
cally stirred for 24 hours. The resultant
mixture was then filtered through a multi-
media filtration bed. The filter bed, which
was housed in a stainless steel column, con-
sisted of four layers of filter sand and
gravel. The top layer was uniform medium
gravel with 050 = 19-1 mm and a thickness of
7.6 cm; the second layer from the top was a
fine gravel with D8c = 16.8 mm and D^ =
14.2 mm, and a thickness of 7.6 cm; the
third layer was a coarse sand with DSB = 6.3
mm and Di5 = 5.1 mm and a thickness of 10.2
cm; the fourth layer was medium sand with
D85 = 1-47 mm and D-|5 = 1.2 mm and a thick-
ness of 10.2 cm and the bottom layer was #20
- #30 Ottawa sand with DBQ - 0.715 mm and a
Agitator
(Leachate
generation]
Filter
Bed
Storage
thickness of 22.9 cm. This arrangement of
gravel and sand in the filter bed was based
on the results of a series of measurements
made using different sizes of gravel and
sand and was designed to provide an effluent
whose suspended particles would not clog the
bed.
The outdoor lysimeters were constructed
of P.V.C. tubing (56 cm O.D. with a wall
thickness of 4.8 mm and 153 cm length) sup-
ported by lucite plates in a vertical posi-
tion. The general configuration and fea-
tures of the field lysimeter are similar to
those of the laboratory lysimeters previous-
ly described16. The use of identical lysi-
meters provided a measure of the reproduce-
ability of the sorbent system. Both lysi-
meters were packed with the preweighed sor-
bents, sufficient to treat 140 gallons of
leachate. Five to 10 cm of Ottawa sand was
placed below and above the sorbents to pre-
vent disturbing the geometry of the sorbents
during addition of leachate. Leachate was
fed to the top of the column through a
valved manifold which distributed this to
both lysimeters, simultaneously, from the
storage reservoir. The lysimeters were de-
signed for constant hydraulic head. Thus,
overflow from the constant head drain was
collected and pumped back to the storage
reservoir. All tubing in the system was
made of Tygon tubing (9.6 mm I.D.). The
effluent volume was monitored as a function
of time and samples of leachate effluent
analyzed at intervals to determine the con-
Constant Head
Lysimeter
Chemical
Analysis
Figure 1 - Schematic Diagram of Pilot Scale Study
301
-------
centration of all measurable constituents
remaining in the effluent after a known
volume of leachate had passed through the
column. This was continued for three dif-
ferent calcium fluoride sludge leachates
collected from the same source at different
times.
A stainless steel 50 gallon tank
equipped with drainage outlet was ued as
storage tank. This storage tank was loca-
ted between the filter bed and field lysi-
meters to serve as both a reservoir and
overflow receiver.
Results and Discussion
In order to carry out pilot scale stu-
dies on the treatment of calcium fluoride
sludge leachate, it was necessary to iden-
tify the most promising sorbent combina-
tions and determine the velocity of lea-
chate through the sorbent bed which pro-
vides for the most effective removal of
fluoride.
Laboratory Study
Analysis of the leachate used in the
laboratory study is presented in Table 1.
TABLE 1
ANALYSIS OF THE NEUTRAL CALCIUM FLUORIDE
SLUDGE LEACHATE USED FOR OBTAINING
SORBENT COMBINATIONS TO PROVIDE OPTIMUM
TREATMENT
Measured Pollutant
Ca
Mg
Zn
F
CN
COD (Organics)
' Concentration (mg/1j
119
89
0.31
15.5
0.61
36
, Cr, Cu, Fe, Ni and Pb were analyzed for,
but found to be below measurable levels.
In this table are listed the pollutants
whose concentrations in the leachate were
at measurable levels. The leachate was
also analyzed for cadmium, chromium, cop-
per, iron, nickel and lead but these heavy
metal ions were found to be below measur-
able levels.
This leachate was passed through labo-
ratory lysimeters that contained varying
amounts of the two most promising sorbent
combinations17, illite, acidic and basic
fly ashes or illite, acidic fly ash and
zeolite. The sorbent combinations were
placed in lysimeters in a layered system in
the weight ratios of 1:1:1 or 2:2:1, with
the illite being the top layer followed by
acidic fly ash or visa versa and either
basic fly ash or zeolite forming the bottom
layer. The basic fly ash or zeolite was
placed at the bottom to remove the cations
(e.g., copper and zinc) that are initially
leached from the illite and acidic fly ash.
The use of the zeolite or basic fly ash
as a bottom layer minimizes the initial lea-
ching of such pollutants as copper and zinc.
This leaching is responsible for concentra-
tions of 4 ppm copper and 1.7 ppm zinc in
the initial 1.4 liter of effluent collected.
The leaching occurs when only the illite and
acidic fly ash combination is used in the
lysimeter. However, when zeolite was used
in combination with the illite and acidic
fly ash, the above copper and zinc concen-
trations were reduced to 0.3 ppm and 0.6
ppm, respectively, in the initial 1.4 liter
of effluent collected from the lysimeters
(see Figure 2 and Figure 3). Basic fly ash,
when used in combination with the illite and
acidic fly ash, reduces the copper and zinc
concentrations from 4 ppm and 1.7 ppm to
1 ppm and 0.6 ppm, respectively.
The results determined with the use of
illite and acidic fly ash in the top layer
of the laboratory lysimeters indicate that
the sorbent combination of illite, followed
by acidic fly ash and basic fly ash, in the
weight ratios of 1:1: exhibits, in general,
greater sorbent capacity than the sorbent
combination of the acidic fly ash, illite
and basic fly ash in the weight ratio of
1:1:1 (see Table 2). The sorbent capacity
represents the amount of contaminant retain-
ed by the sorbent after the spent sorbent is
extensively washed with water. However,
with the exception of total cyanide, the
sorbent removal capacity exhibited by the
illite:acidic fly ash:zeolite (2:2:1) is the
most effective of the combinations studied
for treating all the measurable contaminants
in the calcium fluoride sludge leachate fol-
lowed by the illite, acidic and basic fly
ashes (2:2:1) combinations.
The velocity of the leachate through
the sorbent bed in the lysimeters also was
found to influence the removal of the con-
302
-------
oo
a
C
O
a
M
iJ
e
SI
o
c
c
a
3
4.0
0.4
>(Concentration of copper in the initial 1.4 1 of lysimeter effluent
collected using only illite and acidic fly ash!
0.3
0.2
u.l
Fa + Z
6 8
Effluent Volune>liter
10
12
1
Figure 2 - Control of Copper Leaching from Illite and Acidic Fly Ash Using Zeolite
-------
2.0,
1.6
e
o
•H
I
O
s
M-l
o
c
o
•rl
3
o
I
1.2
0.8
0.4
(Concentration of zinc in the initial 1.4 1 of effluent
collected using only illite and acidic fly ash)
6 8
Effluent Volume, liter
10
12
Figure 3 - Control of Zinc Leaching from Illite and Acidic Fly Ash Using Zeolite
tamihants in the leachates. It does not
effect the total amount of contaminant that
can be removed by a sorbent (sorbent removal
capacity) but it does define the volume of
leachate that can be treated with maximum
removal of a contaminant. For example, dif-
ferent leachate velocities were obtained
when neutral calcium fluoride sludge lea-
chate was passed through lysimeter that con-
tained different amount of illite. The
fluoride, and COD concentration in the ef-
fluent was monitored until breakthrough was
achieved. The results are shown in Figure
4 and Figure 5. Here, it is seen that as
the velocity of the leachate decreases, the
volume of effluent that contains minimal
amounts of fluoride and COD, increases.
The sorbent removal capacities, how-
ever, are not influenced by the velocity of
the leachate through the sorbent bed. For
example, the different velocities were
found to have no significant effect on the
sorbent removal capacity exhibited by the
illite for fluoride and COD removals (see
Table 3).
An examination of the curves in Figure
4 reveals that the optimum velocity for
treating the largest volume of leachate
with maximum fluoride removal should be
less than 0.042 cm/min. The curve repre-
senting operation at the optimum leachate
velocity should allow the greatest volume
of leachate to be treated with a sharp rise
in C/CQ to breakthrough.
Pilot Studies
Since the combination of illite, aci-
dic fly ash and zeolite (2:2:1) showed the
most promise for treating the neutral cal-
cium fluoride sludge leachate in the labo-
ratory, two large vertical lysimeters were
set up outdoors with sufficient sorbent to
treat 140 gallons of neutral calcium fluo-
304
-------
TABLE 2
REMOVAL CAPACITIES' OF COMBINED SORBENTS IN LYSIMETER FOR
NEUTRAL CALCIUM FLUORIDE SLUDGE LEACHATE
Measured I+Fa+Fb Fa+I+Fb I+Fa+Fb I+Fa+Z
Parameters Description 1:1:1 1:1:1 2:2:1 2:2:1
Ca Sorbent Capacity 000 406
Mg Sorbent Capacity 849 528 515 866
Zn Sorbent Capacity 5.9 7.2 6.1 9.5
F" Sorbent Capacity 110 105 128 148
CrT Sorbent Capacity 1.3 1.5 3.9 1.7
COD Sorbent Capacity 199 133 241 218
Remarks:
(1) Sorbent Capacities are expressed in g of contaminant removal per gram of sorbent
used. I = Illite, Fa = Fly Ash (Acidic), Fb = Fly Ash (Basic), Z = Zeolite.
(2) Cd, Cr, Cu, Fe, Ni and Pb were analyzed and found to be below measurable levels.
ride leachate. The leachate was collected
over a period of a year in order to study
the effect of variations in the composition
in the leachate on the removal process due
to changes in plant operation. The pilot
studies were designed for removal of fluo-
ride present in the leachate down to 1 ppm
or lower. The combination sorbent capacity
for fluoride (see Table 2) defines the
amount of total sorbent required in the
lysimeters. The permeability of the clay
fractions were adjusted by admixing with
inert sand to obtain a leachate velocity of
0.01 cm/min through the sorbent bed in or-
der to insure adequate removal of the fluo-
ride ion. This leachate velocity was se-
lected because it is 4 times smaller than
the .042 cm/min which was shown earlier to
approach the required leachate velocity
needed to treat the largest volume of lea-
chate with maximum removal of fluoride ion.
The results of this study are shown in
Figures 6 through 11.
The concentration of calcium in the
three leachates that was passed through
these lysimeters were 309, 115, and 228 mg/1
(see Figure 6). The ill He, acidic fly ash,
zeolite combination (2:2:1) reduced these
concentrations to effluent concentrations of
approximately 80 mg/JL During the addition
of the initial leachate, poor removal of
calcium was observed. This was probably due
to channelling of the leachate through the
sorbent as a consequence of adding the sor-
bent to the columns in the dry state rather
than in slurry form. However, elimination
of the channelling resulted in calcium con-
centration being rapidly reduced to 80 mg/1
and was observed to be independent of in-
fluent concentration. For the case of cop-
per treatment, the copper concentrations in
the three leachates (0.12, 0.10, and 0.07
mg/1) were reduced to 0.04 mg/1 (see Figure
7). Also, the leaching of copper from the
illite and acidic fly ash (vide supra) is
effectively controlled by the zeolite. The
copper concentration in the effluent from
the leachate initially treated was 0.08 mg/1,
which is well below the influent concentra-
tion of 0.12 mg/1.
The effect of channelling on the copper
removal can again be seen by the sudden rise
in the effluent concentration after about 40
liters of leachate was treated. However,
once the channelling was minimized, the cop-
per concentration in the effluent was again
reduced to 0.04 mg/1 and observed to be in-
dependent of influent concentration. This
channelling effect on removal points out the
need for proper dispersion of the leachate
stream through the bed.
The removal of magnesium by illite,
acidic fly ash, zeolite combination (2:2:1)
showed a dependency upon the influent con-
centration (see Figure 8). Influent magne-
sium concentrations of 29.6 mg/1, 75.2 mg/1
305
-------
1.0
0.3
0.6
0.4
0.2
20
Leachate
Velocity
cm/min
• 0.140
O 0.138
A 0.078
A 0.049
40 60 80 100
Effluent Volume, ml per g of Illite
Figure 4 - Effect of Leachate Velocity on the Removal of
Fluoride in Calcium Fluoride Sludge Leachate
120
TABLE 3
SORBENT CAPACITY EXHIBITED BY ILLITE FOR REMOVAL OF FLUORIDE AND COD
AT DIFFERENT LEACHATE VELOCITIES THROUGH SORBENTS
Leachate Velocity
Through the Bed (cm/min)
.140
.138
.079
.042
Sorbent Capacity for
Fluoride yg/g
190
186
179
175
Sorbent Capacity
for COD yg/g
185
192
198
216
306
-------
CO
s
Leachate Velocity (cm/mln)
Effluent Volume, ml per g of Illite
Figure 5 - Effect of Leachate Velocity on the Removal of COD
in Calcium Fluoride Sludge Leachate
-------
60
E
n)
O
e
o
•H
C
0)
o
C
s
C
(II
300
200
100
Leachate 1
- Influent Co=309 mg/1
A
r
Aw
A
•
A
L
Leachate 2
Lysimeter I
Lysimeter II
Influent Co=115 mg/1
Leachate 3
Influent Co=228 rag/1
40
80
120 160 200 240
Volume of Leachate Treated, Liters
Figure 6 - Calcium Effluent Concentration in Pilot Scale Lysimeter Study
280
-------
. 12
rH
E
3 0.08
"
!
g
w II
§ 2
§ 0.04
c
o
4-1
C
0)
rH
M-l
U-(
I - 1
Leachate 1 .^
Influent Co=0.120 mg/1
^
A*
• A
A. -.A
1^ 9 A
A ^
^•** .%.At;
1 ! 1
n 1 1 n
Leachate 2
Influent Co=0.100 mg/1
• Lysimeter I
A Lysimeter II
• ••
A • AAAj^AA^
i i I
1 1
Leachate 3
Influent Co=0.070 mg/1
£A^ka5t*-
' '
40
80
Volume of Leachate Treated, Liters
Figure 7 - Copper Effluent Concentration in Pilot Scale Lysimeter Study
-------
80
OLE
nt Concentration of MR, mg/1
N> *> O>
0 o 0
3
.H
*M
U-l
td
i T ^| i i 1 n — 1 1 j 1
— Leachate 1 Leachate 2 j Leachate 3 —
1
Influent Co=75.7 mg/lj
j
J
^™ ^™ 1
^L ^fc^B ^^ ^^ ^^ ^ft ^^^^k ^V i ^^^k
• • i «A
A i *
Influent Co=29.6 mg/1^ i
' ^ > S"^*^^"1!^*^ j •
•i^ • ^ ! . Influent Co = 18.
~ ^ | A. A AA_
*~ A Lysimeter II i
1 L L _ , 1 , | , !
mg/1
240
40 80 120 160 200
Volume of Leachate Treated, Liters
Figure 8 - Magnesium Effluent Concentration in Pilot Scale Lysimeter Study
280
-------
16
r- 1
00
. 12
i
b-.
U-l
Q
c
o
•H
4-1
2 8
4_l
c
-------
CO
0.4
0.3
r-H
txO
e
i
2
U
iw
o 0.2
c
o
•H
ID
4J
e
0)
o
g 0.1
u
4J
c
0)
3
^H
U-t
M-l
1 1 1 1
Leachate 1
_
.
Influent Co=0.245 mg/1
. A / A V*?
4~ • A
A
4 • *A
%- *
A« A
> H
, 1 ,
III 1
1
Leachate 2
Influent Co=0.37 mg/1
• Lysimeter I
A Lysimeter II
^k ^k ^k J
1
i I .
1 ' !
Leachate 3
_
_
—
_
Influent Co=0.02 mg/1 —
, *
^^ A A\
• A
€T *~, 0 T**~
40
80 120 160 200
Volume of Leachate Treated (Liters)
240
230
Figure 10 - Cyanide Effluent Concentration in Pilot Scale Lysimeter Study
-------
u
40
30
M
e
0
o
M-J
° 20
c
o
•H
uent Concentral
M
0
r-4
im
•4-1
'
Leachate 1
-Influent Co=24.2 mg/1
A*
"•AA/^
!•
4
1 1 .
|
Leachate 2
Influent Co=26.7 mg/1
• Lysimeter I
A Lysimeter II
•*!***£•«
1 vl f
.1, i i — i — u
— i — i 1 —
Leachate 3
Influent Co=44.8 mg/1
_
-
AAi****iJwr
/** *
.I* ~~
r
_
. 1 ,
200
240
40 80 120 160
Volume of Leachate Treated, Liters
Figure 1 1 - COD Effluent Concentration in Pilot Scale Lysimeter Study
280
-------
and 25.6 mg/1 gave an effluent concentration
of approximately 25 mg/1, 53 mg/1 and 15
mg/1, respectively. One would expect re-
sults similar to that observed for the cal-
dium and copper removals which appeared to
be independent of the influent concentra-
tion. The explanation for these results is
unclear at this time.
Not unexpectedly, effective removal of
fluoride is also achieved with this sorbent
combination since the amount of sorbent and
the leachate velocity used were designed for
fluoride removal. An effluent concentration
of 1 mg/1 was achieved with the influent
concentration which varied from 10.2 mg/1
to 15.3 mg/1 (see Figure 9). Again, it is
shown (as was the case for the calcium and
copper results) that the concentration of
fluoride ion in the treated leachate is in-
dependent of the influent concentration.
This also appears to be true for the re-
moval of the cyanide ion. Where the con-
centration of cyanide in the effluent is
significant (i.e., 0.25 mg/1 and 0.37 mg/1
in the first two leachates) the sorbents
reduce these concentrations to approximate-
ly 0.06 mg/1 (see Figure 10). However, for
the third leachate where the influent con-
centration was extremely low (0.02 mg/1),
no significant removal of cyanide is obser-
ved. The minimum concentration to which
the cyanide can be reduced to with this
sorbent combination appears to be about
0.06 mg/1. However, if the illite, acidic
fly ash and basic fly ash sorbent system
(2:2:1) was used instead of the illite,
acidic fly ash, zeolite combination (2:2:1),
the effluent concentration of cyanide
would probably be significantly lower than
0.06 mg/1. This is because of the greater
sorbent removal capacity achieved with the
non-zeolite combination (3.9 mg/gm) than
with the zeolite combination (1.7 mg/gm)
(see Table 2).
The minimum effluent concentration of
the organics achieved with the illite, aci-
dic fly ash, zeolite combination (2:2:1)
appears to be dependent on the organic in-
fluent concentration. As the concentration
of organics in the influent increases from
24.2 mg/1 up to 44.8 mg/1, the concentra-
tion of organics remaining in the treated
leachate also increases from a low of about
2.5 mg/1 to 18 mg/1 (see Figure 11). The
reason for this behavior is not clear, but
the overall results indicate that the il-
lite, acidic fly ash and zeolite combina-
tion is not only extremely effective in re-
moving the cations and anions, but also the
organics present in the neutral calcium
fluoride sludge leachate.
PROPOSED DESIGN OF A CALCIUM FLUORIDE
SLUDGE LEACHATE TREATMENT SYSTEM
The results of the pilot study indi-
cate that the use of the combination sor-
bent capacity (0.148 mg of fluoride removed
per gram of sorbent used)(see Table 2), and
a leachate velocity of 0.01 cm/min through
the illite, acidic fly ash, zeolite combi-
nation (2:2:1) was effective in treating
all the measurable constituents (i.e., Ca,
Mg, Cu, F, CN and organics) in 140 gallons
of leachate without breakthrough of any of
the constituents occurring. In view of
these results, two approaches are proposed
for the treatment of leachate. This lea-
chate will originate from a landfill (2051 x
205' x 12') that is designed to contain an
estimated 10 years production (27,400 tons)
of calcium fluoride sludge. The 12 foot
depth is presently being used in a storage
pit at the plant where the sludge is genera-
ted.
The first approach (see Figure 12) will
involve lining the sludge pit with an imper-
meable liner to prevent ground water intru-
sion. A 1 foot filter bed similar to that
used in our pilot studies will be placed at
the base of the landfill to remove the sus-
pended solids. The leachate will be collec-
ted at the bottom of this filter bed and
pumped on to an adjacent illite, acidic fly
ahs, zeolite bed (2:2:1). The dimensions of
this bed are 28' x 28' x 9'. The bed con-
tains sufficient sorbent to treat one year's
production of leachate at a rate of7.51/rniri
without ponding and still maintain a maximum
leachate velocity through the bed of 0.01
cm/min. The 7.5 1/min flow rate was deter-
mined by assuming an annual average rainfall
of 40 inches and that all the rainfall which
falls upon the landfill becomes leachate.
The second approach is to line only the
sides of the sludge pit with an impermeable
liner to prevent the escape of leachate from
the landfill and to place at the bottom of
the landfill a 2 foot layer of the illite,
acidic fly ash sorbent combination (2:2:1).
This layer will be covered with a 1 foot
layer of filter media to prevent clogging of
the sorbent bed by the suspended solids in
the leachate (see Figure 13). The 180' x
180' x 2' layer of sorbent combination would
314
-------
m o
o oa M
CN H
205'
PLAN VIEW
1. Illite
2. Fly Ash (Acidic) —
3. Fly Ash (Basic) —
or Zeolite
Sand Filter (I1)
SIDE VIEW
Figure 12 - Design of Sorbent Treatment System
be adequate to treat 10 years of leachate
production containing on the average 10 mg/1
of fluoride at the flow rate of 7.5 1/min.
This approach, however, would be limited to
areas where the ground water table is well
below the landfill so that ground water in-
trusion through the sorbent bed into the
landfill will not increase the rate of lea-
chate production beyond 7.5 1/min.
The sorbent cost using the ill He, aci-
dic fly ash, zeolite combination (2:2:1) is
estimated to be $1.37 per ton of calcium
fluoride sludge disposed of in the landfill.
A price of $10/ton for illite, and $50/ton
for zeolite was used to estimate the sorbent
cost. There is no cost associated with
obtaining the acidic fly ash since it is a
waste product and the utility is presently
paying to have it hauled away.
Although the illite, acidic fly ash,
basic fly ash sorbent combination (2:2:1)
was not evaluated on the pilot scale (be-
cause of the time constraints), based upon
the laboratory studies (see Table 2), this
combination should also be effective for
treating the measurable constituents (with
the exception of calcium) in the calcium
fluoride sludge leachate. The illite, aci-
dic fly ash, basic fly ash combination
offers a far less expensive approach. If
the calcium ion concentration encountered
in this leachate (see Figure 6) presents no
significant problem and its attenuation not
considered, the sorbent cost for disposing
315
-------
205'
in O
o oo
28 L.
PLAN VIEW
Sludge
Sorbents
nff
T
i1
.Sand Filter (I1)
\
~r
V
Sand filter
SIDE VIEW
1. Illite
2. Fly Ash (Acidic)
3. Fly Ash (Basic)
or Zeolite
Figure 13 - Design of Liner Red
of 1 ton of calcium fluoride sludge de-
creases to $0.45. All bed or layer dimen-
sions remain the same. It should be noted
that t'here is a disadvantage at the present
time to using the illite, acidic fly ash,
basic fly ash combination rather than the
illite, acidic fly ash, zeolite combina-
tion. The supply of basic fly ash from the
utility at which it is generated is some-
what limited. The power station generates
more acidic fly ash than basic fly ash.
CONCLUSIONS
As a result of these studies, an ex-
tremely promising inexpensive system has
been developed for the treatment of lea-
chate arising from calcium fluoride sludge
disposed of in a landfill. The combination
of illite, acidic fly ash and zeolite or
illite, acidic and basic fly ashes has been
found to be the most effective in a layered
system for removing cations, anions and
organics in neutral calcium fluoride sludge
leachate.
The order that the natural clays and
fly ashes are used in a layered bed influ-
ences the removal of the cations, anions
and organics in the industrial sludge lea-
chates. Acidic sorbents (e.g., illite and
acidic fly ash), which can initially induce
316
-------
slightly acidic conditions into the lea-
chate, are placed at the top of the layered
system followed by those sorbents which can
induce slightly alkaline conditions in the
leachate. This results in the removal of
the anions before the cations.
Alkaline conditions at the base of the
bed are desirable. This favors the removal
of both the cations in the leachate and the
heavy metal cations initially leached from
specific sorbents (i.e., ill He and acidic
fly ash). Either zeolite or basic fly ash
was found effective in controlling this ini-
tial leaching of heavy metal ions by the
acidic sorbents.
In the design of a sorbent system, the
total amount of a specific cation, anion or
organics which is removed by a sorbent is
indicated by the sorbent removal capacity.
The volume of leachate that can be treated
with maximum removal is regulated by the
velocity of leachate through the sorbent
bed. This leachate velocity can be regula-
ted by sorbent bed height which defines lea-
chate volumetric flow rate through sorbent
bed, under specified hydraulic conditions.
These hydraulic conditions can be controlled
by varying the amount of inert material
added to the clays to regulate their perme-
ability, or varying the particle size of the
sorbents in the bed.
With the exception of magnesium, the
illite, acidic fly ash, zeolite sorbent com-
bination in the weight ratio of 2:2:1 was
found to be effective for treating the mea-
surable contaminants in calcium fluoride
sludge leachate on a pilot scale. A calcium
ion concentration of over 300 mg/1 in lea-
chate was reduced to 80 mg/1, copper ion
concentration of 0.12 mg/1 was reduced to
0.04 mg/1, the fluoride ion concentrations
were reduced from 15 mg/1 to 1 mg/1, the to-
tal cyanide was reduced from 0.37 mg/1 down
to about 0.06 mg/1 and the COD was reduced
from about 45 mg/1 down to 15 mg/1. Magnes-
ium ion concentration was reduced from 76
mg/1 down to only about 53 mg/1. In addi-
tion, with the exception of the magnesium
and the COD, the resultant effluent concen-
trations were found to be independent of in-
fluent concentrations.
Sorbent cost for the illite:acidic fly
ash:zeolite combination (in the weight ratio
of 2:2:1) required for the treatment of the
leachate during a ten year period of working
the landfill has been estimated to be $1.37
per ton of sludge disposed of in the land-
fill. This cost is based upon an annual
rainfall of 40 inches, assuming that all the
rainfall that falls upon the landfill be-
comes leachate. However, this cost may be
reduced to only $0.45 per ton of sludge dis-
posed of in the landfill if the illite:aci-
dic fly ash:basic fly ash combination (in
the weight ratio of 2:2:1) is used. Based
upon the laboratory lysimeter results, the
illite:acidic fly ash:basic fly ash combi-
nation appears promising for the treatment
of the measurable contaminants (with the
exception of the calcium ion in the calcium
fluoride sludge leachate).
REFERENCES
1. Rios, C.B., "Removing Phenolic Com-
pounds from Aqueous Solutions with Ad-
sorbents," U.S. Patent #2,937,142.
2. Bhargava, R., and Khanna, M.P., "Remo-
val of Detergents from Wastewater by
Adsorption on Fly Ash," Indian Journal
Environmental Health 16, 109-120 (1974).
3. Baker, R., and Hah, M., "Pyridine Sorp-
tion from Aqueous Solutions by Mont-
morillonite and Kaolinite," Water Re-
search, 5_, 839-849 (1971).
4. Lub, M., and Baker, R., "Sorption and
Desorption of Pyridine-Clay in Aqueous
Solution," Water Research, 5_, 849-859
(1971).
5. Nelson, M.D., and Quarino, C.F., "The
Use of Fly Ash in Wastewater Treatment
and Sludge Conditioning," J.W.P.C.F.,
42, R-125-135 (1970).
6. Kliger, L., "Parathion Recovery from
Soils After a Short Contact Period,"
Bulletin of Environmental Contamination.
and Toxicology, n, 714-719 (1975).
7. Liao, C.S., "Adsorption of Pesticides
by Clay Minerals," A.S.C.E. Sanitary
Engineering Div. 96_, 1057-1078 (1970).
8. Damanakis, M., Drennan, D.S.H., Fryer,
J.D., Holly, K., "The Adsorption and
Mobility of Paraquet on Different
Soils," Weed Research, 10, 264-277
(1970).
9. Emig, D.D., "Removal of Heavy Metals
from Acid Bath Plating Wastes by Soils,"
Diss. Abst. B, 2661 (1973).
317
-------
10. Griffin, R.A., Cartwright, K., Shrimp,
N.F., Steele, J.D., Ruch, R.R., White,
W.A., Hughes, G.M., and Gilkeson, R.H.,
"Alternation of Pollutants in Municipal
Landfill Leachate by Clay Minerals:
Column Leaching and Field Verification,"
Environmental Geology Notes, 78, 1-34
(1976).
11. Griffin, K.A., Frost, R.R., Au, A.K.,
Robinson, G.D., and Shrimp, N.F., "At-
tenuation of Pollutants in Municipal
Landfill Leachate by Clay Minerals:
Heavy-Metal Adsorption," Environmental
Geology Notes, 79, 1-47 (1977).
12. Bittell, J.G., and Miller, R.J., "Lead,
Cadmium and Calcium Selectivity Coef-
ficients on a Montmorillinite, Illite,
and Kaolinite," Journal of Environmen-
tal Quality, 3, 250-253 (1974).
13. Babich, H., and Stotzky, G., "Reduction
in Toxicity of Cadmium to Microorgan-
isms by Clay Minerals," Applied and
Environmental Microbiology, 3 696-705
(1977).
14. Basu, A.N., "Exchange Behavior of Cop-
15.
16.
per, Manganese, and Zinc Ions," Journal
Indian Society Soil Science, 6, 71-76
(1958).
Fuller, W., McCarthy, C., Alesii, B.A
and Niebla, E., "Liners for Disposal
Sites to Retard Migration of Pollu-
tants," Residual Management by Land
Disposal, EPA - 600/9-76-015 (1976).
Chan, P.C., Dresnack, R., Liskowitz,
J.W., Perna, A., and Trattner, R.,
"Sorbents for Fluoride, Metal Finish-
ing, and Petroleum Sludge Contaminant
Control," Final Report, EPA Grant R803-
717-01.
Chan, P., Liskowitz, J.W., Perna, A.J.,
Trattner, R., and Sheih, M., "Control
of Pollution from Leachates," Proceed-
ing of the 1st Annual EPA/AES Confer-
ence, Jan 17-19, 1978, Orlando, Fla.
This research was supported in part by
EPA Grant R803-717-012 from the Industrial
Environmental Research Laboratories Office
of Research and Development, D. S. Environ-
mental Protection Agency, Cincinnati, Ohio.
17.
318
-------
SELECTION OF COVER FOR SOLID WASTE
Richard J. Lutton
Soils and Pavements Laboratory
U. S. Army Engineer Waterways Experiment Station
P. 0. Box 631
Vicksburg, Mississippi 39180
ABSTRACT
Selection and design of cover for solid/hazardous waste can be accomplished effectively
by engineering planning in four steps. First, identify the functions to be served by the
cover and establish relative importances, e.g., impeding water percolation is more impor-
tant than supporting vegetation, etc. Second, use data on soil properties in terms of the
Unified Soil Classification system or the U. 3. Department of Agriculture system to rate
available soil for effectiveness in the various functions. Third, specify certain design
procedures in the disposal operation, including those necessary for circumventing the defi-
ciencies of the selected cover soil. Placement procedures such as compaction will improve
the soil for certain functions. Elsewhere, the soil properties favorable for one function
are unfavorable for another, e.g., a clayey cover soil impeding water percolation will pre-
vent escape of decomposition gases that may also be desired. Fourth, where a single soil
cannot serve contrasting cover functions, incorporate features such as layering or use
special nonsoil materials and additives.
INTRODUCTION
This paper summarizes the first phase
of a study on cover for solid and hazardous
waste.(l) The most common cover material
has been soil, particularly indigenous soil
at the disposal site. This is not to say,
however, that other materials cannot be
used, and quite likely synthetic and waste
materials will be used in greater amounts in
the future. Although some aspects of soil
as cover have not been thoroughly investi-
gated, much is standardized and well known
from previous experience and practice in
soil mechanics, agriculture, and soils
construction.
BACKGROUND
Preliminary investigations for solid
waste disposal operations are summarized
briefly here, whereas a review of all as-
pects of solid waste disposal is available
elsewhere.(2) During preliminary investi-
gations the potential sites are chosen, the
local geology is determined, and the subsur-
face is explored. Subsurface exploration
may be accomplished by trenching or boring,
and soil samples are usually obtained for
laboratory testing.
One purpose of preliminary site inves-
tigations is to ascertain the subsurface
conditions of each potential site. The
geology should primarily be favorable to
minimizing the environmental impact of the
disposal project on surrounding water sup-
plies, i.e., not significantly altering
water quality. Subsurface exploration
should provide soil samples representative
of the soil types and stratification. After
boring logs have been prepared and the soils
classified, a specific site is ordinarily
selected on the basis of the type of waste
disposal area, suitable topography, favor-
able geological and groundwater conditions,
and availability of adequate amounts of
cover soils/materials with necessary pro-
319
-------
parties for minimizing objectionable envir-
onmental effects.
Geologic and subsurface sampling and
exploration may also be conducted on poten-
tial off-site borrow areas, which might pro-
vide suitable soil for cover. Investiga-
tions should be sufficient to define the
soil types and quantities and natural water
contents. Laboratory studies may also
include tests for classification and
compaction and, if necessary, for permeabil-
ity and strength properties in compacted
samples.
SOIL CLASSIFICATION
Currently, soils used in solid waste
disposal design and construction are classi-
fied according to either the Unified Soil
Classification System (USCS) or the U. S.
Department of Agriculture (USDA) system.
The USCS is engineering oriented; soils are
grouped according to gradation of particle
sizes, to percentages of gravel, sand, and
fines, and to plasticity characteristics.
The USDA system classifies on the basis of
texture only, i.e., percentages of gravel,
sand, silt, and clay. Figure 1 presents
100-0
/ \ USDA TYPE
USCS TYPE
Figure 1. USCS superimposed on USDA
soil chart.\3)
relationships between the two systems based
on laboratory tests on hundreds of soil
samples.
REQUIREMENTS OF COVER ON SOLID WASTE
A basic legal requirement in solid waste
disposal is that a layer of earth be placed
over the compacted solid waste after each
day's operation. Most solid waste disposal
operations also employ a final, thick soil
layer for long-range cover after completion
of the operation. A membrane or other syn-
thetic material is sometimes used in place
of soil. Intermediate covers are also dis-
tinguished for surfaces exposed for long
periods but eventually buried under subse-
quent lifts.
COVER FUNCTIONS
Specific functions of cover on solid and
hazardous waste sites are as follows:
a. Control insect emergence and entrance.
b. Control rodent burrowing.
c. Reduce bird and animal attractiveness.
d. Minimize or maximize water percolation,
e. Minimize water erosion.
f. Minimize or maximize gas movement.
g. Minimize fire hazard potential.
h. Minimize blowing paper.
i. Control noxious odors.
J. Provide sightly appearance.
k. Minimize settlement and maximize
compaction.
1. Provide for vegetative growth.
m. Minimize wind erosion and dust
generation.
n. Resist cold climate deterioration
and operational difficulties.
o. Preserve slope stability.
p. Resist cracking.
Functions k and m concerning trafficability
and wind erosion are discussed later, in
examples of the technical approach.
Among these functions, a few stand out
as primary while the others are usually
secondary. Primary functions are more or
less independent and concern such landfill
processes as movement of water and gas and
susceptibility to erosion. Secondary func-
tions restate primary functions viewed in
a different sense. Resistance to fire, a
secondary function, is contingent upon satis-
factorily impeding air or gas migration, a
primary function. Similarly, slope drain-
age (and side slope stability) is a second-
ary or somewhat different way of looking at
the primary function, assisting water per-
colation. Physical characteristics of the
320
-------
cover that are basic to the primary cover
functions and are also involved in most of
the secondary functions are conductivity to
water, conductivity to gas, shear strength,
and tensile strength (or capillary potential).
FUNCTIONAL PRIORITIES
The functions as listed above are
straightforward, tut in actuality cover
functions are complexly interrelated; for
this reason, designing solid waste cover
involves considerable subjective judgment.
Ranking of priorities is necessary in deter-
mining which functions are most important
at the specific site and in designing fea-
tures to compensate for adverse side effects
on other cover functions. For example, a
cover that functions to impede infiltration
and percolation of water from the surface
will also impede upward movement of decom-
position gases. If gases are to be moved
through the cover to minimize lateral move-
ment, a gas vent system may be required. A
brief hypothetical scenario illustrates the
problem of conflicting functions and sug-
gests the priorities approach recommended
here.
Suppose that it has been determined
that cover for a disposal area receiving
hazardous chemicals must not only serve all
of the usual functions but have special
effectiveness in preventing the generation
of dust in a region prone to high wind
velocity. Suppose that the powdery waste
is damp upon arrival at the area but that
it dries quickly in the semiarid climate.
Finally, suppose that the area is usuaj.±y
subjected to a two-month rainy season in
which much of the 15-in. annual precipita-
tion is concentrated.
First priorities for most of the year
would be wind erosion resistance and dust
control with water infiltration control and
" other functions following in progressively
lower rankings; this is assuming that there
are no other particular hazards, such as
flammability or generation of toxic gas.
Control of animals and asthetic considera-
tions usually fall low in the ranking be-
cause they are insensitive to choice of soil
type. The problem of air pollution by toxic
dust particles is solved by specifying that
the daily cover shall consist of sand, which
by its coarse grain size will resist wind
erosion, and shall be 8 in. or more in thick-
ness to assure complete shielding of an
irregular surface of waste. The infiltration
of rainwater largely takes place in the short
rainy season; therefore, specifications
should require that all intermediate cover
exposed during this rainy period shall
include a 6-in. layer of compacted clay
below 6 in. of sand. Finally, it may be
specified that certain dust palliatives or
water be applied during disposal operations
to prevent drying and dust generation by
traffic erosion.
SELECTION OF COVER SOIL
DETERMINING COVER FUNCTIONS
The ranking of cover function priorities
is a key part of the cover selection proce-
dure, but it can only be accomplished s-ite
specifically. For example, at many landfills
the rate of infiltration of surface water
may be of utmost importance; elsewhere, for
instance in arid regions, minimization of
infiltration and percolation is entirely
subordinate. Long- and short-term functions
should be separated and given some absolute
ranking. For example, it may be difficult
to distinguish the relative importance of
reduction in dust during the operation of a
landfill as opposed to the reduction of po-
tential for spontaneous combustion, where
the latter requires covering and enclosing
cells in clay-rich, but dust-prone, soil.
These decisions are made at the design stage.
Only at this point in the development of a
project can the actual conditions, such as
climate, soil availability, transportation
costs, and population proximity, be evaluated
in the context of generalizations such as
made below.
For each function Judged to be signifi-
cant, soils should be ranked from best to
worst, e.g., for trafficability and wind
erosion in Table 1. Next, it will be neces-
sary to flag those soils that are actually
available for use at the site. These may
include materials brought in from a distance
in preference to indigenous soil. Allowance
may be needed for the high cost of importing
materials.
RATING SOILS
Soils can be rated from I (best) to as
many as IX or more (poorest); examples are
shown in Table 1. Quantitative parameters
are also provided in the table so that the
reader can see the absolute position of a
321
-------
TABLE 1. EXAMPLE SOILS RANKING FOR
FUNCTIONAL PERFORMANCE
Ranking For
uses
Soil
GW, GP
.'•:
SW, SP
•
Description
Gravel
Silty Gravel
Clayey Gravel
Sand
Silty Sand
Clayey Sand
Silt
Silty Clay
Clayey Silt
Clay
Trafficability
(RCI Value)
I (>200)
III (177)
V (150)
I (>200)
II (179)
IV (157)
IX (1014)
VII (III)
VIII (107)
VI (lit5)
Wind Erosion
(% Sand/Gravel)
I (95-100)
III (60-95)
V (50-90)
II (95-100)
IV (60-95)
VI (50-90)
VII (0-60)
VIII (0-55)
IX (0-50)
X (0-50)
particular soil rating. The first column
concerns the go/no go aspect of trafficabil-
ity that may be particularly critical during
disposal and covering operations. The
ratings of I to IX from coarse granular to
fine silty soils are based on a complete set
of rating cone index (RCI) values. The
second data column concerns wind erosion
control, and soils are rated I to X accord-
ing to their sand and gravel content. The
basis for the ratings and quantitative para-
meters (Table l) is reviewed in the next two
sections.
EXAMPLE FUNCTION - TRAFFICABILITY
Trafficability of solid waste cover is
the capacity of the material to support
vehicle traffic. Factors influencing traf-
ficability of cover soils include not only
the many variables that combine to determine
the strength but also the slope, condition
of underlying solid waste, and geometrical
obstacles, such as drainage ditches. Traf-
ficability of a soil is considered adequate
for a given vehicle if the soil has suffi-
cient bearing capacity to support the vehi-
cle and sufficient traction capacity to
enable the vehicle to develop the forward
thrust necessary to overcome its rolling
resistance.
MEASURING SOIL TRAFFICABILITY
It has been demonstrated in military
studies (l) that the go/no go performance
of vehicles can be predicted with reasonable
accuracy using soil strength indexes. Rela-
tive shear strength of soil may be evaluated
with a cone penetrometer. This instrument
employs a 30-deg cone with 1/2-sq-in. base-
end area. The force necessary to push the
cone slowly through the soil is indicated
as the cone index (Cl), from 0 to 300. The
value 300 occurs under a force of 150 Ib.
As traffic crosses moist or wet soil,
the soil is remolded and its strength usually
is changed. Remolding index (Rl) procedures
quantify this change. Five CI values are
read at 1-in. increments to a depth of k in.
in undisturbed soil, and five CI values are
obtained in the same soil after remolding.
The RI is determined by dividing the sum of
the CI readings made after remolding by the
sum of the readings made before remolding.
The rating cone index, defined RCI = CI
x RI where CI is measured undisturbed in the
field, expresses the probable strength of
the specific soil under repeated traffic.
Figure 2 shows the RCI for the various soil
types determined in a large testing program.
Comparing the relationships for RI and RCI
it can be seen that the CI for more or less
undisturbed soil cancels and the RCI largely
represents the trafficability strength of
remolded soil. Accordingly, the RCI is
directly applicable to solid waste cover
where the soil is always in a remolded con-
dition. For coarse-grained soils or clean
sands, CI measurements alone are usually ade-
quate to quantify trafficability.
!' .1
Figure 2. RCI ranges for USCS soil types.
EVALUATING TRAFFICABILITY
The minimum RCI, just sufficient for 50
passes of a specific vehicle, is called the
322
-------
vehicle cone index (VCI). If the soil RCI
is higher than the VCI of a particular vehi-
cle, numerous successful passes can "be ex-
pected along the same straight path, or one
such vehicle can be expected to execute
severe maneuvers without becoming immobilized.
The VCI for individual vehicles within "broad
categories are given in Table 2. Categories
2 through 7 include vehicles commonly used
in waste disposal operations.
TABLE 2. VCI FOR VEHICLE CATEGORIES
_ Vehicle Type
CI Range
ct.rs with '.' w • intact pressure
100 or greater
SPECIAL OPERATING PROCEDURES
In cases where cover trafficability is
marginal or where modifications producing
beneficial trafficability effects are not
compatible with other cover functions, special
design procedures may be followed. Four
procedures are recommended.
Semipermanent Distribution Routes
Lay out the oil-site handling and cover-
ing operation so that vehicles with poten-
tial trafficability problems are confined to
limited paths designed and maintained for
good trafficability. Indigenous or imported
granular soils, base courses, and/or inex-
pensive pavements should be considered to
prepare a distribution route for use during
the operations at one level, for example.
Daily Maintenance
Daily maintenance may be established
and equipment made available to assure that
deterioration in cover trafficability is
rectified before becoming disruptive. As a
first approach, it may be effective to
assign municipal road grading equipment and
a crew to half-day maintenance periods at
intervals of two to five days to add supple-
mental soil, then sprinkle, smooth, and com-
pact. The on-the-site equipment should be
sufficient to take care of rutting and other
trafficability problems that arise during
daily operation. Stockpiles of suitable
additive material (granular, clayey, etc.)
should be kept on hand or be readily available.
Special Wet-season Area
Organize the disposal area in such a
way as to have a standby section available
for use during periods of inclement weather.
This section might be laid out on well-
drained sandy terrain near arterial access
and be composed of single-height waste cells
where likelihood of trafficability problems
is minimal.
Special Waste Compaction
It is anticipated that cover traffica-
bility may become a problem where high
degrees of compaction of solid waste are
not achieved. Deep, immobilizing ruts may
develop as a consequence of spongy waste
subgrade rather than deficiencies of the
cover soil. Attempts to reduce this problem
may be made by increasing the ratio of soil
to waste in the covering operation. Improv-
ing compaction of waste should help, and in
the extreme, bailing of the solid waste
might be considered.
EXAMPLE FUNCTION - WIND EROSION CONTROL
This section follows a procedure^4) for
using wind characteristics along with soil
and vegetation factors to predict soil loss
and as a basis for selecting and designing
cover to control wind erosion. Most studies
on wind erosion and measures for its reduc-
tion have been conducted in the interest of
agriculture where the seriousness on a
regional scale is obvious. Corresponding
erosion potential at solid waste sites is
of much lesser magnitude, but another aspect
of concern arises where the soil is actually
breached by erosion to expose the waste to
potential wind dispersion.
WIND EROSION EQUATION
Soil loss depends on cloddiness, surface
roughness, and surface moisture of the soil;
on amount, kind, and orientation of the
vegetation; on wind velocity or force; and
on the windward distance across the field.
The wind erosion equation (WL'E) expresses
functionally the amount of erosion in tons
per acre per annum, from a given soil sur-
face as A' = f(K', C', L' , T' , \") where
K' = soil erodibility index; C' = climatic
323
-------
factor; L1 = field length along prevailing
wind erosion direction; T1 = soil ridge
roughness factor; and V = equivalent quan-
tity of vegetative cover.
The WEE is potentially useful for deter-
mining acceptability or unacceptability of a
solid waste cover design where coupled with
a criterion for permissible levels of annual
soil erosion. Ways of reducing high erosion
should become apparent when each factor in
the equation is carefully considered sepa-
rately; thus, a manipulation of the orienta-
tion of a landfill might reduce L' and, in
turn, er.osion loss. This and other methods
of design modification are discussed below.
DESIGNING WIND EROSION CONTROL
Design for wind erosion control at solid
waste sites has considerable flexibility in
comparison with control measures applicable
to agricultural fields. Each factor of the
WEE including even C' concerns one or more
variables that are at least locally subject
to manipulation. Techniques for designing
new sites or modifying existing ones are
briefly explained below.
Soil Composition
The susceptibility of soils to wind
erosion is a function of the coarse soil
fraction. Figure 3 gives erosion loss
values corresponding to measured percentages
of grains over 0.8k mm in diameter (for level
terrain). The figure is a basic tool to be
used in design for selecting or modifying
cover soil type on the basis of sieve analy-
ses of those soils available.
300
20 40 60 80
FRACTION >0.84 mm (PERCENT)
100
Knoll and Side Slope Configuration
The vulnerability of knoll-like
configurations on cover can be evaluated
with a factor (Figure 1+) that adjusts erosion
loss percentage by 100 or more in comparison
with erosion loss from a similar flat sur-
face. This factor should be used to esti-
mate the effects of sloping sides of land-
fills facing toward the prevailing winds.
Figure b shows that even gentle windward
slopes may cause major erosion concentration
and, therefore, are places in need of sta-
bilization or protection (withwind barriers).
Thus, it may be reasonable to arrange side
slopes in protected orientation and lee side
positions.
700
600
Z 400
UJ
& 300
s
O 200
-J
Si
100
T
i—r i i r
2 3 4567
WINDWARD KNOLL SLOPE, %
8 10
Figure 3. Wind erosion versus percent coarse
fraction. (**)
Figure k. Knoll adjustment in percent.^
Length-width and Orientation Configuration
L1 is an important variable in the wind
erosion process. On an eroding surface, the
rate of soil flow is zero on the windward
edge and increases with leeward distance
until, on a large surface, the flow reaches
a maximum. Three techniques available
during the designing of a landfill to reduce
L' and soil loss are: orient width parallel
to prevailing wind; use barriers at strate-
gic locations; and use only a small segment
of landfill at one time, keeping the remain-
der protected by vegetation or other means.
Vegetation Cover
The importance of healthy vegetation in
reducing wind erosion can hardly be overem-
phasized. Even the relatively sparse cover
provided by crop stubble has a significant
324
-------
effect in the WEE. Timely seeding and
emergence of vegetation are important in
reducing bare soil exposure time. However,
timing may not be a straightforward matter
since coordination with the growing season,
windy season, and disposal operations plan
must also be assured.
c. Providing for analysis of percola-
tion through cover to allow for estimating
leachate production and for designing effec-
tively in that usually high-priority function.
REFERENCES
Additivea
Since the need for wind erosion protec-
tion at landfills is often short-term and
may be urgent, emergency methods should be
such as can be applied when the need arises
and will be instantaneously beneficial. The
application of chemicals to soil cover is
one of the most effective quick-response
methods for stabilization. Common soil
stabilizers fall into five broad groups:\"'
inorganic salts, inorganic cementing materi-
als, bituminous materials, resinous materi-
als, and polymeric materials. Actually, the
increasing of soil moisture by sprinkling
with water is temporarily quite effective.
Unfortunately, wind erosion is most common
in water-deficient areas where, in addition,
evaporation rates are usually high.
FUTURE DIRECTION OF STUDY
The second phase of this study is pro-
ceeding in the following directions:
a. Revising and refining the interim
report/1^ particularly for operational
practicability.
b. Establishing comprehensive design
and construction techniques applicable after
soil selection.
1. Lutton, R. J. and Regan, G. L., "Selec-
tion and Design of Cover for Solid Waste,
Interim Report," U. S. Environmental
Protection Agency, Interim Report (in
preparation), 1978.
2. Brunner, D. R. and Keller, D. J., "Sani-
tary Landfill Design and Operation,"
U. S. Environmental Protection Agency,
Report SW-65ts, 1972.
3. Meyer, M. P. and Knight, S. J., "Traffi-
cability of Soils, Soil Classification,"
U. S. Army Engineer Waterways Experiment
Station, Technical Memorandum 3-2UO,
Supplement l6, 1961.
Skidmore, E. L. and Woodruff, N. P.,
"Wind Erosion Forces in the United
States and Their Use in Predicting Soil
Loss," U. S. Department of Agriculture,
Handbook 3^6, 1968.
Woodruff, N. P. and Siddoway, F. H.,
"A Wind Erosion Equation," Proceedings
Soil Science Society of America, v. 29,
1965, pp. 602-608.
Sultan, H. A., "Soil Erosion and Dust
Control on Arizona Highways, Part I,
State-of-the-Art Review," Arizona
Department of Transportation, Report RS-
10-lla-l, October 197**.
325
-------
LABORATORY ASSESSMENT OF FIXATION AND ENCAPSULATION PROCESSES FOR ARSENIC-LADEN WASTES
Jaret C. Johnson and Robert L. Lancione
JBF Scientific Corporation
2 Jewel Drive
Wilmington, MA 01887
ABSTRACT
This study is evaluating the effectiveness of several proprietary and generic
processes for the fixation of arsenic in arsenic bearing wastes. Three industrial solid
wastes that are high in arsenic conentration have been treated by generic processes in
the laboratory and by proprietary processes at vendor's facilities. Leaching studies on
the treated wastes are being performed to assess the relative safety of each product for
disposal. Data are presented from shake tests on pulverized samples and on intact mono-
lithic samples. Leachate analyses have shown that some processes have value, but that
great differences exist among the various processes.
INTRODUCTION
Many industrial residues containing
arsenic are presently held in long term
storage because of the lack of an accep-
table disposal method. Other similar
residues are disposed of on land, but the
design, construction, and operation of
sites that can protect water resources
from arsenic contamination are often very
difficult and expensive. Several regions
of the country do not contain many sites
with the needed hydrogeological charac-
teristics, so hauling becomes a major
expense. Long-distance hauling of
hazardous wastes also incurs the risks of
accidental spills.
For these reasons, a study is being
conducted to learn if ways exist or can
be developed that can make arsenic-laden
residues safe for disposal at sites less
restricted than the types of sites
presently required.
A national survey of industrial
users and producers of arsenic-containing
materials indicated that three forms of
arsenic-bearing wastes should be of
primary concern because of their amount in
storage (or amount annually produced) and
their arsenic concentration. These wastes
are:
(1) Residues from the manufacture of
organic arsenical herbicides. Most of
these wastes are presently stored at or
near the facilities producing them.
(2) Filter cake from the production of
food-grade phosphoric acid. Most of
these wastes are presently shipped to
secured landfills.
(3) Flue dusts from nonferrous smelting
operations. A variety of disposal,
storage, and recycling routes is
presently employed by the many faci-
lities producing these residuals.
Waste No. 1 is composed mostly of
sulfate and chloride salts. It contains
approximately 2% organic arsenicals, or
0.9 to 1.0% expressed as As. Waste No. 2
is a yellow, damp (37% moisture) acidic
filter cake. It contains approximately
1.2% As2S3, or 0.7% expressed as As.
Waste No. 3 as used in this study was a
fine white powder containing 90% As_0_,
or 68% expressed as As. Although moat
flue dusts are much lower in arsenic con-
tent, As20, is the major arsenic compound
in all or them, so this material was
selected to represent flue dusts. The
arsenic concentrations stated here are
those provided by the sources of the
wastes, and they have been confirmed by
326
-------
digests and analyses of the actual waste
samples used in this project.
Each of these wastes has been treated
with fixation processes by commercial or
academic developers of those processes.
In addition, generic processes such as
formation of less soluble compounds and
mixing with Portland cement have been
developed in the laboratory. For specific
wastes of the types addressed here, it is
quite possible that an ad hoc fixation
process may be more successful than com-
mercial processes, which are developed and
offered for a broad spectrum of wastes
ranging from radioactive liquids to flue
gas desulfurization sludges. Leaching
tests have been used to evaluate the suc-
cess of each fixation attempt.
The term "fixation" is used through-
out this paper to describe any process or
effect that is intended to, or actually
does, reduce the leaching of arsenic from
treated residues.
This paper presents information on
the fixation of all three wastes involved
in the study. Products of five vendor
processes are discussed, representing a
wide range of approaches to fixation.
All products discussed here are solid and
monolithic; none are granular.
METHODS
The three wastes as described above
were each subjected to five processes. A
coding system for each product is used
here; for example, in the product iden-
tified C155,
C designates the commercially available
process
1 indicates Waste No. 1
55 indicates that the waste loading, or
percentage of waste in the fixed pro-
duct, was 55%. All waste loadings and
other percentages in this paper are
expressed on a wet weight basis.
Two types of leaching tests were
used: an elutriate test and a slow shake
test. The elutriate test, modified from
that developed for assessing dredged
materials(1), was as follows in this brief
summary:
(a) Crush the waste product and screen
through #10 sieve. Use 100 gm of
material that passes the screen.
(b) Mix the granular material from (a)
with 1 £ of C02 - saturated distilled
water.
(c) Shake on a wrist action shaker for
24 hr.
(d) Let stand one hr.
(e) Filter through 0.45 urn filter.
(f) Analyze filtrate
This test was a physically severe
method of assessing behavior of products
should their structure break down with
time in the disposal environment.
The slow shake test was used to
assess the longer term behavior of intact
specimens in contact with water. Products
were molded by all vendors into cylinders
7.5 cm in diameter and approximately 20
cm long. Shake test steps are briefly
described:
(a) Cut a cylindrical slice of approxi-
mately 100 grams from the fixed pro-
duct (usually about 2 cm thick).
(b) Support the sample so that all sides
are exposed to water in a glass jar.
Add CO, - saturated distilled water
in the ratio 10 m£ HO:gram sample.
Raw waste tests involved 50 gm sample
and 1£ water.
(c) Oscillate jar at 60 one-inch strokes
per hour.
(d) Sample and replace water every 48
hours. Analyze sample for components
of interest.
Arsenic analyses were performed on a
Perkin-Elmer Model 372 atomic absorption
spectrophotometer. Digestions were con-
ducted in a mixture of hydrofluoric,
hydrochloric, and nitric acid within a
tightly sealed Teflon vessel.
RESULTS
The leaching data for arsenic are
presented for each waste type separately.
Plots of concentration in the slow shake
test leachate versus time, although not
directly indicative of field conditions,
allow some important inferences to be
327
-------
drawn. For example, if arsenic concen-
tration for a sample were to continue at
a high and stable level, leaching is
probably limited by solubility. If the
concentration drops with time, the avail-
able arsenic in the sample may be becoming
depleted.
Figure 1 shows the slow shake test
leachate arsenic data as a function of
time. Both the raw waste and product
(Fig. 1)
C155 leached high arsenic concentrations
early in the test, followed by a rapid
decrease to low levels. In the case of
the raw waste, all the arsenic in the
sample dissolved. Here one can see the
value of some fixation processes: for
products A170 and F150, leaching has
begun and continued at a much slower rate
than in the raw waste. To compare this
situation to rainfall and percolation
events at a land disposal site, note that
one such event would cause a large and
sudden release of arsenic from unfixed
wastes, while many such events would cause
smaller and slower releases from a proper-
ly fixed waste.
Elutriate test data for Waste No. 1
are shown in Table 1. (Samples from
Process A could not be crushed into granu-
lar form.) Only Process F achieved any
success in this destructive test.
(Table 1)
Leachate arsenic concentrations from
slow shake tests for Waste No. 2 products
are shown in Figure 2. Note that the
(Fig. 2)
early peak for the raw waste is much lower
than that for Waste No. 1 because the
solubility of As-S. is lower than that of
the organic arsenicals. The amount of
arsenic leached from raw Waste No. 2 in
the first ten days (five leachant renewal
cycles) was only 25% of that initially
present. The advantages of the fixation
processes can be seen in that the initial
leachate from the raw waste contained 40
mg/£ arsenic, while the leachate from the
fixed wastes never exceeded 2 mg/J?,.
Table 2 lists elutriate test data for
Waste No. 2. (Table 2)
It is interesting to note that the
samples yielding high pH values in the
elutriate were most successful in retain-
ing arsenic. This behavior is unexpected
because As2S is more soluble at high pH
than at low pH. The more likely reason
for this behavior is that these samples
produced a large amount of clay-sized par-
ticles upon crushing, and that these par-
ticles adsorbed arsenic from the elutriate.
Data from slow shake tests on Waste
No. 3 products are shown in Figure 3. in
this case, the very soluble arsenic from
(Fig. 3)
the raw waste was found in the leachate
early, and in continuing high amounts.
Saturation of the samples' interstitial
water probably prevents depletion of the
available arsenic. A clear distinction
among processes can be seen in Figure 3,
with three orders of magnitude separating
Processes B and F. (Processes C and D
were similar in results to Process B).
The elutriate data for Waste No. 3
do not show similar trends, as Table 3
illustrates. (Table 3)
One of the processes that was least
successful in the slow shake test (Process
D) performed very well in the elutriate
test. The agitation that induces intimate
TABLE I
ELUTRIATE TEST RESULTS
WASTE NO. I
SAMPLE NO. pH S04(mg/l) Cl(mg/l) Mg As % As LEACHED
BUS
CI55
DI46
FI50
RAW WASTE
12.6
7.1
12.6
5.5
5.7
4,320
8,750
4,650
9,630
9,640
24,800
19,400
16,700
186
880
438
120
540
100
100
100
27
100
328
-------
1000
LEGEND:
• RAW WASTE ~\
• AI70
A CI55
T FI50
WASTE NO.
0.0
10 20 30
TIME (DAYS)
Figure 1. Effect of Time on Arsenic Concentration in Leachates:
Slow Shake Tests, Waste No. 1
329
-------
100
LEGEND--
• RAW WASTE
• F260
A B225
T C229
WASTE NO. 2
Figure 2.
20 25
(DAYS)
Effect of Time on Arsenic Concentration in Leachates:
Slow Shake Tests, Waste No. 2
330
-------
lOOOc
100:
o
UJ
-------
TABLE 2
ELUTRIATE TEST RESULTS
WASTE NO. 2
SAMPLE NO. pH S04(mg/l) Cl(mg/l) MgAs % As LEACHED
B225
C229
D228
F260
RAW WASTE
12.4
1.7
12.5
1.5
1.8
<0.2
3.1
1.5
1,350
22
4.4
22
0.04
5.8
-------
TABLE 4
COMPARISON OF ELUTRIATE AND SHAKE TEST RESULTS
WASTE NO. I
SAMPLE NO.
BII8
CI55
0146
FI50
AI70
ELUTRIATE
% As RELEASED
100
100
100
27
SHAKE
CUMULATIVE % As
I DAY
RELEASED
8 DAYS
14.5
42.6
38.2
6.1
3.9
TABLE 5
COMPARISON
OF ELUTRIATE AND SHAKE
WASTE NO 2
TEST RESULTS
SAMPLE NO.
B225
C229
D228
F260
ELUTRIATE
% As RELEASED
0.02
2.9
< 0.005
3.3
SHAKE
CUMULATIVE % As
DAY
RELEASED
8 DAYS
3.2
0.82
3.4
0.31
TABLE 6
COMPARISON
OF ELUTRIATE AND SHAKE
WASTE NO. 3
TEST RESULTS
SAMPLE NO.
B33I
C370
D3I5
F350
ELUTRIATE
% As RELEASED
1.7
6.6
0.01
0.2
SHAKE
CUMULATIVE % As RELEASED
I DAY 8 DAYS
15.4
I.I
6.7
0.004
333
-------
that initially available, and to accumu-
late the fractional mass leached in each
leaching period. It is also helpful for
those who would compare these data with
other data from differently configured
specimens to normalize the fractional
mass with the volume: surface ratio.
A plot, therefore, of the parameter
Z\ . V
where: la "= cumulative mass of arsenic
leached in n periods
A » mass of arsenic initially
0 present in sample
V • sample volume
S - sample apparent surface area
vs. time is shown in Figure 4 for Waste No.
1. Of the processes shown, Process C
(Fig. 4)
shows a very rapid increase in arsenic
leaching, followed by a very slow rate of
release. (A similar fixation process, per-
formed by another vendor whose products
are not discussed in this paper, showed
almost identical behavior). This process
almost immediately released all the
arsenic in the product. The other pro-
cesses shown in Figure 4 behaved more in
accordance with similar plots reported in
the literature (3,4).
Figure 5 is a similar plot for Waste
(Fig. 5)
No. 2. Clearer differences among processes
can be seen, especially from the low
leaching rate of Process F. Performance
of all processes shows significant improve-
ment over the raw waste.
Cumulative releases from products of
Waste No. 3 are shown in Figure 6. The
(Fig. 6)
retention of arsenic by Process F in this
case is three orders of magnitude better
than the raw waste and Process D.
The value of plots such as Figures 4
through 6 is to present a conceptually
easy view of the relative success of the
various processes. To gain a more rigorous
quantitative figure of merit, it has been
shown by Godbee and Joy (4), that a plot
of the cumulative fraction leached versus
the square root of time can be used. From
such a plot, one can take the slope and
calculate an effective diffusion coeffi-
cient for each product with the relation
-
S
1/2
where: De • effective Diffusivity
t - time
and other symbols have been
defined above.
Data are shown this way in Figures
7, 8, and 9 for Wastes 1, 2, and 3 respec-
tively. The effective diffusion coeffi-
cients calculated from these plots are
listed in Table 7.
(Table 7)
The approach to plotting the above
equation should yield a straight line
with slope equal to 2(—) if the under-
lying assumptions are valid. The primary
assumption is that diffusion of arsenic to
the surface of the solid is the limiting
factor in leaching. For Wastes 2 and 3, the
plots on rectangular paper are linear (the
curves on Figures 8 and 9 appear nonlinear
because semi-log paper was used to present
the full range of leaching behavior).
Thus evidence for a diffusion mechanism
is present. For Waste No. 1 (Fig. 7) the
plots of all products on both linear and
semi-log paper are erratic. Therefore
De values were not computed for Waste No.
1 products. The fractional amount of
arsenic leached from Waste No. 1 products
was consistently higher than for the other
wastes in both types of leaching tests.
These observations suggest that Waste
No. 1 requires treatment other than the
processes discussed here to reach a dif-
fusion-limited condition.
The slow shake tests discussed herein
are continuing, so that endurance data on
the products can be produced. In addition,
other products remain to be subjected to
leaching tests. The following conclusions,
therefore, are tentative although they are
well supported by the data presented here.
CONCLUSIONS
(1) Commercially available processes vary
widely in their success in retarding
leaching of arsenic from the wastes
under study.
(2) For Wastes 2 and 3, all processes
appeared to produce diffusion-limited
leaching behavior.
334
-------
WASTE NO. I
UJ
>
o
0.001
LEGEND:
• A170
CI55
• FI50
NOTE:
100% OF ARSENIC IN
RAW WASTE DISSOLVED
WITHIN ONE DAY
10 20 30
TIME (DAYS)
40
Figure 4. Cumulative Leaching Behavior, Slow Shake Tests, Waste No. 1
335
-------
o
UJ
cc
UJ
g
QL
UJ
O
O
UJ
UJ
g
b
tr. 0.01
UJ
I
=>
o
0.001
LEGEND:
• RAW WASTE
• B225
A C229
T F260
WASTE NO. 2
10 20 30
TIME (DAYS)
Figure 5. Cumulative Leaching Behavior, Slow Shake Tests, Waste No. 2
336
-------
0.
o
UJ
cc
8
tr
v>
-v
UJ
3
O
0
UJ
UJ
_J
UJ
I
2
O
0.01
LEGEND:
R RAW WASTE
• A380
A D3I5
C370
F350
- WASTE NO. 3
10 20
TIME (DAYS)
Figure 6. Cumulative Leaching Behavior, Slow Shake Tests, Waste No. 3
337
-------
WASTE NO. I
1.0
-R-
-R—R
u
UJ
o:
UJ
o
a:
(0
UJ
s °-
UJ
Q
UJ
O
o:
u.
UJ
§0.01^-
LEGEND:
R RAW WASTE
• At 70
A BII8
V FI50
• CI55
2
6
TIME172 (DAYSI/2)
Figure 7. Diffusion Plot, Waste No. 1
338
-------
WASTE NO. 2
LEGEND:
R RAW WASTE
• B225
• F260
A C229
-
0.001,
I/* * 1/2 *
TIME' (DAYSI/Z)
Figure 8. Diffusion Plot, Haste No. 2
339
-------
0.
o 0.01
UJ
or
ui
u
if
DC
8
UJ
o .001
o
UJ
UJ
Q
o:
,., .0001
.OOOOL
LEGEND:
R RAW WASTE _J
• A 380
• F350
A C370
V D3I5
1
WASTE NO. 3
I 2345 6
TIME172 (DAYS172)
Figure 9. Diffusion Plot, Waste No. 3
340
-------
TABLE 7
EFFECTIVE DIFFUSIVITIES OF FIXATION PRODUCTS
PROCESS
A
B
C
D
F
De (cm2/day)
WASTE NO. 3 WASTE NO. 3
9.6 X I0~5
6.9 X IO"5
3.0 X IO"8
1.2 X IO"8
1.4 X IO"3
7.5 X IO"4
I.7X IO"4
4.IX IO"10
(3) For all three wastes, the elutriate
test on crushed samples is not a
reliable substitute for, or aid in
predicting, long-term leaching
behavior on monolithic specimens. It
does yield useful information on the
separate question of leaching
behavior in the event that the mono-
lithic structure should break down.
(4) The three wastes examined were quite
different in their response to
fixation attempts, as measured by
leaching -behavior. Waste No. 1
produced the least success with all
processes discussed here.
REFERENCES
(1) Keeley, J.W. and Engler, R.M.,
"Discussion of Regulatory Criteria for
Ocean Disposal of Dredged Materials:
Elutriate Test Rationale and Imple-
mentation of Guidelines," Paper
D-74-14, U.S. Army Engineer Waterways
Experiment Station, Vicksburg, MS,
1974.
(2) Mahloch, J.L., "Leachability and
Physical Properties of Chemically
Stabilized Hazardous Wastes,"
Proceedings of the Hazardous Waste
Research Symposium, Feb. 2-4, 1976,
at Tucson, AZ, Report No. EPA-600/9-
76-015, July 1976.
(3) Colombo, P. and Neilson, R.M., Jr.,
"Properties of Radioactive Wastes and
Waste Containers," Quarterly Progress
Report, Oct.-Dec. 1976, Brookhaven
National Laboratory Report No. BNL-'
NUREG-50664, June 1977.
(4) Godbee, H.W. and Joy, D.S., "Assess-
ment of the Loss of Radioactive
Isotopes from Waste Solids to the
Environment. Part I: Background and
Theory," Oak Ridge National
Laboratory, Report No. ORNL-1M-4333,
February 1974.
ACKNOWLEDGMENT
This work is being performed under
Contract No. 68-03-2503 from the U.S.
Environmental Protection Agency.
341
-------
ENCAPSULATION TECHNIQUE FOR CONTROL OF HAZARDOUS MATERIALS
H. R. Lubowitz* and C. C. Wiles
*
TRW Systems Group
One Space Park
Redondo Beach, California 90278
U.S. Environmental Protection Agency
26 W. St. Clair Street
Cincinnati, Ohio 45268
ABSTRACT
A process using polyolefins and fiberglass to encapsulate containers holding hazardous
wastes was researched, developed and evaluated on a laboratory scale. The process is geared
to reinforce deteriorating containers in order to secure their hazardous consignments in
subsequent handling operations and under environmental conditions existing in a direct dis-
posal site such as a landfill. Polyolefins were employed because they provide a unique
combination of low cost, abundance, chemical stability, mechanical toughness and flexi-
bility. Fiberglass is low cost, prevalent, chemically stable, stiff and load bearing.
Encapsulate test specimens (right cylinders, 3-inch dia. x 4-inch ht.) were fabricated by
fashioning cylindrically shaped fiberglass casings, having walls about 1/50-inch thick, and
then forming 1/4-inch thick jackets by fusion of powdered high density polyethylene (HDPE)
upon the surfaces of the casings. The test specimens were thus rendered seamless. Mainly,
two types of specimens were produced, some filled with sand and some hollow. A set of
sand-filled ones were charged with aqueous solutions of heavy metals (simulating loss of
container contents within an encapsulate) and then examined with respect to their ability
to secure metal contaminants under aqueous leaching conditions. They were found to hold
their contents secure in leaching by water and by a strong solvent for metals such as
dilute hydrochloric acid. Hollow and filler containing specimens were employed to evaluate
mecahnical performance of encapsulates. These, under appreciable mechanical pressure, were
found to maintain their dimensional integrity. Under extreme pressures, they underwent
distortion, yet their jackets did not fail even though their casings were flawed by the
mechanical stresses. The above results indicate the processes' ability to prevent, or
limit to acceptable levels, the release or delocalization of hazardous waste from deterio-
rating containers in manipulation and in final disposition. This paper discusses the
process and provides results of the evaluations. Since the process is viewed as a constit-
uent aspect of a general approach being developed for encapsulation of hazardous materials,
the encapsulation technique applied in earlier work for managing unconfined, particulated,
hazardous wastes is reviewed.
INTRODUCTION
Encapsulation technique being developed waste management schemes are found wanting.
at TRW under EPA sponsorship is addressed to Either contaminants, inherently, cannot be
management of wastes which are potentially "neutralized", e.g., heavy metals, or they
harmful to man and/or the environment, and are difficult to render harmless, as in
in addition, difficult to control by current incineration of highly haloginated organic
means.1»2 For many hazardous wastes current compounds. When heavy metals are "stabilized"
342
-------
by chemical fixation, a pretreatment step
may be required. Yet the fixed metal con-
taminants may exhibit stability under
ambient conditions that are more benign
than those they may encounter in environ-
ments designated for final waste disposition.
Furthermore, pretreatment in many cases must
be tailored to the contaminant type thereby
making complex mixtures difficult to manage.
Under the above circumstances, encapsulation
becomes an attractive alternative technique
for hazardous waste management.
In the management of difficult to con-
trol wastes by encapsulation, a singularly
important feature must be general applica-
bility. It is recognized that means not
applicable directly to essentially all the
outstanding difficult to control wastes
would be of limited utility. For example*
it would be prohibitively expensive and
complex to match waste types with technique
designated appropriate. Nor is it accept-
able to carry out pretreatment of wastes to
render them chemically suitable for a
particular approach. However, concentration
of hazardous contents may be suitable so
that more cost effective use can be made of
encapsulating materials and product manipu-
lation procedures.
The waste types under study are:
• Particulates (Unconfined Wastes)
• Containers (Preconfined Wastes)
• Sludges (Unconfined Wastes)
Work was carried out initially with respect
to encapsulation of particulated wastes.*
The particulates employed were obtained by
dewatering selected sludges. The present
study concerns encapsulation of contain-
erized wastes, i.e., 55-drums and smaller
containers holding hazardous consignments.**
Although experimental work presently is not
being carried out in management of sludges
per se, one objective of the work is to fit
sludges (without dewatering) into the encap-
sulation technique.
Description of previous study of parti-
culated waste encapsulation is given here;
and in addition, new aspects of this work
Contract Nos. 68-03-0089 and 68-03-2037,
Solid and Hazardous Waste Research Divi-
sion, U.S. Environmental Protection Agency,
Cincinnati, Ohio 45268
**
Contract No. 68-03-2483
are provided. Greater emphasis is given to
description of the present study. In this
respect, fabrication of laboratory scale
test specimen encapsulates and evaluation of
products, presently completed, are described
in some detail. This work is scheduled for
completion in August 1978.
NATURE OF WASTE ENCAPSULATES
TRW hazardous waste encapsulates are
characterized by two essential elements:
a stiff, load bearing moiety, and a tough,
flexible, seamless plastic jacket. The
stiff element is geared to provide dimen-
sional stability under mechanical stresses
an encapsulate may encounter in manipulation,
transportation and final disposition such as
a landfill. Jacket secures the hazardous
consignment in the event the encapsulate
encounters inordinate mechanical stresses
causing dimensional distortion that may give
rise to flaws in the load bearing element.
And furthermore, the jacket plays the
important role of sealing the encapsulate
and securing the contaminants vis-a-vis
environmental leaching waters.
The selection of materials and encapsu-
late fabrication technique are interdependent.
But the jacketing materials were fixed
because it was deemed desirable to employ
polyolefins as jackets for encapsulates
because polyolefins are mass produced, low
cost, well characterized and exhibit excel-
lent chemical compatibility and mechanical
performance. On a price/performance these
materials are. unique. Thus the objectives
of encapsulation studies were to establish
ready method for jacket formation using
polyolefins, and therewith to determine con-
comitant means for encapsulate fabrication.
Encapsulation of Particulated Wastes
Encapsulations of waste particulates
were carried out on a laboratory scale by
cementation of particulates with an organic
binder, thus fashioning load bearing agglom-
erates, and then enveloping the agglomerates
with jackets of polyolefinic plastic. Test
specimens were produced in two shapes;
cubic, 3-inches on edge and cylindrical,
3-inch dia. x 4-inch ht. The resin 1,2-
polybutadiene was selected for fashioning
agglomerates. This resin was shown to
exhibit the unique feature of readily con-
solidating particulates having a wide range
of chemical compositions and consistencies.
Consolidation occurred under minimal mechan-
343
-------
ical pressures by chemical reaction in the
presence of waste concentrations up to 97%
by weight yielding agglomerates with
densities over 100 Ibs per cu ft. The
agglomerates were then secured by 1/4-
inch thick jackets of high density polyethyl-
ene (HDPE). (A commercial scale encapsulate,
e.g., a cubic shape two feet on edge, is
expected to weigh about 800 to 1000 pounds
and require about 8% by weight total resin
for its fabrication.) The jackets were
formed by fusing powdered HDPE upon the
surfaces of the agglomerates. This method
of jacket formation was selected and
developed because it required only minimal
mechanical pressures and no molten resin
displacements, thereby simplifying jacket
formation. Jackets were thus rendered
seamless.
The encapsulates were found to exhibit
excellent chemical stability and mechanical
performance. With chemical reaction in
agglomerate formation, 1,2-polybutadiene
yielded binders chemically cross-linked
having backbones characterized by singly
bonded carbon atoms, chemical configurations
that exhibit exceptionally high resistance
to hydrolytic and oxidative degradation and
yield stiff substances. The agglomerates
were indeed found to be chemically stable
and high load bearing. The HDPE jackets
sealed the agglomerates and thus secured
the hazardous contents with respect to de-
localization by aggressive aqueous solutions.
The encapsulates withstood appreciable
mechanical pressure until finally yielding
their dimensional integrity, yet with
dimensional distortion their hazardous con-
signments remained secure due to the tough,
seamless plastic jackets.
Continuing exploratory work was ori-
ented to making agglomerates exhibit more
"plastic" than ceramic yield modes under
extreme mechanical pressures. The incorpo-
ration of powdered polyethylene into the
agglomerate indicated that plastic character
can be imparted in this manner. There were
also excellent indications that polyethylene
was grafted chemically onto the polybutadi-
ene structure in agglomerate formation,
thereby polyethylene when molten remained
homogeneously dispersed in the agglomerates.
This phenomenum gave rise to the con-
cept that agglomerate formation can be
carried out with simultaneous plastic
encapsulation of the particulates thus
negating the need for jacket formation.
However, this apparent advantage may be
counter-balanced in realizing high perfor-
mance contaminant containment by the need
to specify suitable particulate consisten-
cies, a need practically eliminated when
jacket formation is employed. With respect
to jacket formation, the exploratory work
oriented to demonstrate that polyolefins
other than HDPE may be readily utilized.
Thus evidence was garnered that ultra high
molecular weight polyethylene and chemically
cross-linkable polyethylene were indeed
utilizable. Future investigation should
demonstrate utilization of polyethylene
ionomers and polybutadiene treated poly-
ethylenes. The above polymer types, in our
opinion, could lead to two or three jacket
formulations wherewith essentially the
overwhelming portion of hazardous waste may
be securely encapsulated.
Encapsulation of Containerized Wastes
In management of containerized wastes
by encapsulation, it was assumed that the
walls of containers would eventially corrode
and subsequently containers would not be
load bearing. As a result, mechanical
stresses would be transferred to the encapsu-
late walls. In contrast to particulated
waste encapsulates wherein jackets need not
be load bearing, containerized waste jackets
need to provide load bearing character.
Fiberglass was selected as the stiff
element to reinforce polyethylene jackets
because this material exhibits a unique
combination of features: low cost, pre-
valence, chemical stability and mechanical
stiffness. Methods for its use utilize
well developed state-of-the-art techniques.
Spray-up is the preferred method; however,
in the laboratory the convenient method of
hand lay-up was employed.
Fiberglass reinforced polyethylene
encapsulate test specimens were produced in
the laboratory cylindrically shaped, 3-inch
dia. x 4-inch ht., with walls of two-layer
construction. The walls consisted of about
1/4-inch thick polyethylene jackets backed
by 1/50-inch thick fiberglass. (The
thickness and nature of the fiberglass
moiety in a commercial scale encapsulate
would be characterized in future work to
withstand mechanical stresses an encapsu-
late may encounter in real-life manipulation
and final disposition. A commercial scale
encapsulate is envisioned to provide about
344
-------
75 gallons capacity. The wall would consist
of a 1/4-inch thick polyolefin jacket backed
by a 1/16 to 1/8-inch thick chopped fiber-
glass casing. The jacket and the casing
would comprise by percent of the encapsulate
volume about 3.5 and 1.8, respectively.)
The purpose of the laboratory test specimens
was to provide demonstration of high chemical
and mechanical performance for fiberglass
reinforced polyethylene encapsulates.
Some encapsulates were produced con-
taining sand, some containing plastic
microballons, and some hollow. The sand
filled ones were charged with aqueous
solution of heavy metals (simulating loss
of container contents within an encapsulate)
and then examined with respect to their
ability to secure metal contaminants under
aqueous leaching conditions. (Sand or soil
would be used to fill voids between con-
tainers and container/encapsulate walls.
Compaction of containerized wastes rather
than filler utilization was considered to
be potentially desirable because it gives
rise to more effective use of encapsulating
materials and product manipulation methods.
But compaction operations may also require
pollution control procedures which may be
burdensome, hazardous and costly. As a
result, encapsulation technique was initially
tailored to manage hazardous waste containers
as is.) Hollow ones and some sand and micro-
ballon ones were employed to evaluate the
mechanical performance of encapsulates. It
was assumed that a clearer picture of encap-
sulate wall performance would be realized by
more detailed examination of encapsulates
without filler than those with filler.
ENCAPSULATION TECHNIQUE AND ENCAPSULATION
EVALUATIONS
The following sections show (a) labora-
tory procedure for preparing encapsulation
test specimens, (b) performance of specimens
in leaching by aqueous solutions, and (c)
performance under extreme mechanical stresses.
Further evaluations of the test specimens
are underway with completion of studies
scheduled in August 1978. The initial re-
sults indicate that polyolefin/fiberglass
encapsulation of wastes should yield com-
mercial scale, high performance products.
Although cost studies are yet to be completed
it is our opinion that on a cost/performance
basis polyolefin/fiberglass encapsulation
would provide an attractive alternative for
control of hazardous materials existing in
containers.
Fabrication of Polyethylene Encapsulate Test
Specimens
Figure 1 shows construction sand being
poured into the glass/resin casing (cocoon),
and the following figure (Figure 2) shows
the encased sand product.
Fusion of polyethylene was carried out
in the annulus resulting from positioning
an encased product in a mold as shown in
Figure 3. Polyethylene was then placed as
shown in Figure 4. And the polyethylene was
fused as in Figure 5.
The steel mold walls were heated by hot
mold plattens, with the rise in temperature
of the mold walls monitored by a thermal
probe. The changes in phase of the poly-
ethylene, i.e., powdered resin to molten
resin, and then subsequently to solidified
resin with cooling of the mold walls, was
readily observed on the pressure gauge. The
speed of jacket formation was essentially
limited only by the heat exchange rate.
(Commercially, 1/4-inch thick polyethylene
products are fabricated in 30 seconds.)
The specimen stemming from the operation
in Figure 5 is shown in Figure 6; Figure 7
shows the injection of a solution of hazar-
dous chemical agents into the cavity. (The
solution occupies the space between the sand
particles.)
Employing additional polyethylene
powder, the encapsulation of the test
specimen was completed, as shown in Figure
8. (In multiple specimens the following
metals were localized: Ni, Cd, Hg, Cr, Zn,
Cu, Sb, As, Se, and Pb. The specimens were
placed in plastic containers holding ISOOcc
of leaching solution. Plastic in contrast
to glass allows ready agitating and tumbling
of the ensembles.) Figure 9 shows container
holding specimen in leaching solution.
A specimen residing in the container is
• shown in Figure 10. The set of test speci-
mens in leaching is shown in Figure 11.
Preliminary LeachinR Results
Table 1 contains preliminary analytical
determinations of the heavy metal containing
> encapsulates immersed in deionized water.
Table 2 pertains to metals immersed in 0.1N
HC1 leaching solution.
Initial analytical determinations were
performed on a Jarrell-Ash Atomic Absorption
345
-------
FIGURE 1. SHOWS FABRICATION STEP OF
GLASS/RESIN ENCASEMENT OF SAND
'. ' 21 3i ' 41 Si
FIGURE 2. CLASS/RESIN ENCASED PRODUCT
'
FIGURE 3. AN ENCASED PRODUCT CENTERED IN
MOLD FOR POLYETHYLENE ENCAPSULATION
FIGURE 4. SUBMERGING ENCASED PRODUCT WITH
POWDERED POLYETHYLENE
346
-------
FIGURE 5. APPARATUS EMPLOYED FOR FUSING
POWDERED POLYOLEFIN
1 5
'"*
FIGURE 6. A POLYETHYLENE ENCASED PRODUCT
STEMMING FROM THE FIRST FUSION STEP
FIGURE 7. HAZARDOUS CHEMICAL AGENTS BEING FIGURE 8. COMPLETED POLYETHYLENE ENCAPSU-
INJECTED INTO THE TEST SPECIMEN LATED TEST SPECIMEN
347
-------
FIGURE 9. SHOWS CONTAINER TYPE FOR HOLDING
SPECIMEN IN LEACHING SOLUTION
FIGURE 10. SHOWS SPECIMEN RESIDING IN
CONTAINER HOLDING LEACHING SOLUTION
FIGURE 11. SHOWS SET OF SPECIMENS
IN LEACHING SOLUTIONS
Spectrophotometer, model 850. Detailed
analyses are now underway employing instru-
mentation and techniques with lower detection
limits.
Compressive Specimens and Properties
The response mode of cylindrical poly-
ethylene encapsulate specimens were
investigated to increasing mechanical loads
applied to specimens in vertical and lateral
positions. (Figures 12, 13, 14, 15 and 16
show the response of a vertical encapsulate
to progressively increasing compressive
stress.) All specimens were compressed to
40% of their original vertical heights.
The specimens recovered about 90% of their
heights with release of the applied com-
pressive loads. Hollow encapsulates
exhibited two bulges, as shown in Figure 17,
while sand filled ones, showed single bulges
as shown in Figure 18. No explanation is
presently on hand that accounts from these
two observed modes of deformation.
Figures 19 and 20 show the response
modes of encapsulate under lateral com-
pression. Under extreme degree of compres-
sion about 75%, no rupture of the encapsulate
was observed. With release of the applied
load, the specimen recovered to about 70% of
its lateral dimension. The nature of the
encapsulates resulting from lateral com-
pression are shown in Figures 21 and 22.
(The "circles" on the compressed encapsulates
stemmed from the imprints on the faces of
the compressive rods.)
348
-------
TABLE 1. METAL CONCENTRATIONS (PPM) IN DEIONIZED WATER LEACHING SOLUTION*
CO
*>
CD
Metal
As
Cd
Cr
Cu
Hg
Ni
Pb
Sb
Zn
Se
Day** Day Metal Loss*** Comments
90 120 (Calculated)
0.002 (to. 001)
<0.05 0.0006 (±0.0006)
<0.05 0.007 (±0.0005)
<0.05 0.009 (tO. 001)
< 2 0.005 (tO. 0003)
<0.5 0.007 (tO. 003)
<1.0 0.003 (tO. 002)
<1.0 0.040 (±0.002)
<0.05 0.003 (tO. 003)
0.117 (tO. Oil)
1000
ROD These encapsulated waste -concentration values are
within the variation of the background "instru-
mental noise".
810
3190
820
810
1000
860
2070
860
*Change in sensitivity of values reflects employment of more definitive
analytical techniques
**"Day" - Number of days encapsulated waste immersed in deionized water
***"Metal Loss"
Values calculated giving concentrations of contaminates
in leaching solutions assuming catastrophic failure of
specimens.
-------
TABLE 2. METAL CONCENTRATION (PPM) IN 0.1N HC1 LEACHING SOLUTION*
Metal Day**
60
Day
90
Metal Loss***
(Calculated)
Comments
u>
Ol
o
Ni
<0.5 0.010 (10.003)
810
Sb <1.0 0.040 (to.002) 860
Zn <0.05 0.006 (10.003) 2070
These encapsulated waste concentration values are
within the variation of the background "instru-
mental noise".
*Change in sensitivity values reflects employment of more definitive
analytical techniques
**"Day" - Number of days encapsulated waste immersed in HC1 leaching solution
***"Metal Loss" - Values calculated giving concentrations of contaminates
in leaching solutions assuming catastrophic failure of
specimens.
-------
FIGURE 12. ONSET OF BULGING OF ENCAPSULATE
UNDER COMPRESSION STRESS
FIGURE 15. "VASE" LIKE CONFIGURATION
FIGURE 13. BULGING PRONOUNCED
AT SPECIMEN MIDDLE
FIGURE 16. ENCAPSULATE UNDER
EXTREME COMPRESSION
FIGURE 14. FURTHER BULGING
FIGURE 17. SHAPE OF HOLLOW ENCAPSULATE
RESULTING FROM COMPRESSION
351
-------
FIGURE 18. SHAPE OF FILLED ENCAPSULATE
RESULTING FROM COMPRESSION
FIGURE 21. SIDE VIEW OF KNCAfSULATE
AFTER LATERAL COMPRESSION
FIGURE 19. ENCAPSULATE POSITIONED
FOR LATERAL COMPRESSION
FIGURE 22. FRONT VIEW OF ENCAPSULATE
AFTER LATERAL COMPRESSION
FIGURE 20. SIDE VIEW OF ENCAPSULATE
IN LATERAL COMPRESSION
The ability of the encapsulates of this
work to undergo extreme lateral compression
without failure is a feature of these
structures that is, in our opinion, not
characteristic of conventional containers.
(Conventional containers fitted with lids
will not undergo appreciable lateral com-
pression without rupture of the lid-container
wall interface. With respect to maintaining
a water tight configuration, even minimal
degrees of compression would be detrimental.
Figure 23 shows graphically the response
of a hollow encapsulate under compressive
load. The influence of the stiff casing is
clearly noted on the graph by the spiked
portion of the curve which is emphasized in
Figure 24. This is characteristic of brittle
materials such as glass fiber reinforced
epoxy composites. In contrast to catastro-
352
-------
15,000
CO
Ul
00
10,000 -
5,000
15.0 30.0 45.0
DEFLECTION (% OF ORIGINAL SPECIMEN SIZE)
60.0
FIGURE 23. LOAD-DEFLECTION OF HOLLOW ENCAPSULATE
-------
5000
4000 -
3000 -
CO
CJ1
CO
CO
Q
<
O
2000 -
1000 -
5.0 7.5
REFLECTION (% OF ORIGINAL SPECIMEN SIZE)
10.0
FIGURE 24. LOAD-DEFLECTION OF HOLLOW ENCAPSULATE
-------
phic property loss of such materials, the
encapsulates continue to exhibit load
bearing properties due to presence of the
jacket. The remaining portion of the curve
reflects the influence of the jacket. Up to
30% deflection the curve is characteristic
of polyethylene in tension. This is most
likely due to the jacket being subject to
tensile stresses as the casing undergoes
flawing compression. After 30% compression,
the apparent load bearing properties of the
hollow encapsulate increases dramatically.
This may be due to continuing flawing of the
casing while the encapsulate is being com-
pressed thereby introducing more plastic
jacket as load bearing material.
A view of the cross-section of a hollow
encapsulate after compression to about 40%
of its original vertical height is given in
Figure 25. One notices the high degree of
FIGURE 25. SHOWS FRAGMENTATION OF THE CASING
AND FUNCTIONAL INTEGRITY OF THE JACKET
AFTER GREAT EXTENT OF COMPRESSION
FIGURE 26. INNER VIEW OF ENCAPSULATE
AFTER GREAT EXTENT OF VERTICAL COMPRESSION
were no apparent breaks nor was there any
apparent change in its cross-sectional
dimension.
An inner view is given in Figure 27 of
an encapsulate that contained plastic hollow
sphere filler (simulating containers) and
underwent extreme extent of lateral com-
pression. In these cases, appreciable
fragmentation of the casing occurred at the
vertical extremities, while the casing
disposed laterally was relatively unaffected.
One can see the crease in the casing travers-
ing down the center caused by the great
extent of lateral compression. In lateral
compression, as well as in vertical com-
pression, the jackets of the encapsulates
were found to retain their functional
integrity.
fragmentation of the casing. The fragments
stemmed mainly from the vertical inner walls
while the casing remained essentially in-
tact on top and bottom of the cavity. Al-
most all the vertical casing was sheared
from the jacket upon extreme compression of
the encapsulate. The jacket, however, did
not fail.
An inner view is given in Figure 26 of
an encapsulate with sand filler which under-
went extreme vertical compression. In this
case the casing of the filled encapsulate
was not fragmented to the degree seen for
the hollow one. One notices, however, the
casing to be appreciably crazed. Neverthe-
less, the jacket remained intact; there
FIGURE 27. INNER VIEW OF ENCAPSULATE
AFTER GREAT EXTENT OF LATERAL COMPRESSION
355
-------
CONCLUSIONS
Results of the study and evaluations
described in this paper led to the following
conclusions:
• Polyethylene jackets backed by fiber-
glass casings yield encapsulates
exhibiting excellent chemical and
mechanical properties.
• Jackets are readily formed by fusion
of powdered polyethylene onto the
surfaces of casings.
• Thick jackets can be rapidly formed.
• Encapsulates of polyethylene and
fiberglass can be fabricated in the
presence of unconfined aqueous solu-
tions in the cavities. Based upon
the process evaluations, the process
appears to be compatible with a wide
range and condition of containerized
hazardous wastes which may not be
adequately manageable by other
techniques.
• Based upon performance in laboratory
tests, encapsulated containerized
wastes would be expected to provide
a high degree of control over the
release to the environment of un-
wanted quantities of hazardous
wastes tested.
• Based upon literature characteri-
zation of polyolefins, control of an
extensive range of hazardous and
corrosive materials would be expected.
DISCUSSION AND RECOMMENDATIONS
Studies are still in progress with
respect to encapsulation of containerized
wastes with polyolefins and fiberglass. The
trend of the current work indicates that
fusion of powdered polyolefins onto glass
casings give rise to jackets having the
expected properties of polyolefinic resins.
Consequently, the extensive information
supplied by the resin manufacturers for
these plastics can be applied directly to
estimating their performance as jackets of
encapsulated wastes. An objective of future
work should be selection of two or three
polyolefinic resins which are compatible
with the bulk of hazardous waste types and
exhibit high performance chemical and
mechanical properties. In this respect,
HDPE is an excellent candidate material. In
addition, the required nature and thickness
of the fiberglass casings should be defined
wherewith encapsulates may withstand real-
life stresses.
Additional study is recommended to
determine means to fabricate commercial
scale encapsulates of containerized wastes,
particularly 55-gallon drums, under actual*
field conditions. This study should give
rise to an encapsulation unit that is
readily transportable to the site where
containerized wastes reside. At that site
encapsulation of containers should be car-'
ried out. The managed containers are thus
made suitable for transportation to and
sequestering in a direct disposal site.
REFERENCES
Wiles, C.C., and Lubowitz, H.R., " A
Polymeric Cementing and Encapsulating
Process for Managing Hazardous Waste"
Proceedings of the Hazardous Waste Re-
search Symposium, EPA-600/9-76-015,
p. 139, July 1976.
Lubowitz, H.R., Derham, R.L., Ryan, L.E.
and Zakrzewski, G.A., "Development of a"
Polymeric Cementing and Encapsulating
Process for Managing Hazardous Wastes"
U.S. Environmental Protection Agency Re-
port, EPA-600/2-77-045, August 1977.
356
-------
FIELD EVALUATION OF CHEMICALLY STABILIZED SLUDGES
by
R. B. Mercer, P. G. Malone, and J. D. Broughton
U. S. Army Engineer Waterways Experiment Station
Vicksburg, Mississippi 39180
ABSTRACT
Four sites where stabilized industrial wastes had been disposed were examined to
determine the effects of stabilized wastes on surrounding soils and groundwater. All
areas selected for study were in the humid eastern or southern United States where rain-
fall was sufficient to produce abundant leachate. The sludges had all been fixed using
the same proprietory process involving addition of cementitious materials to form a
stable, soil-like product. Two of the industrial waste sites contained auto assembly
(metal finishing) wastes, one site contained electroplating wastes and the fourth site
contained refinery sludges. The physical properties of soils under the disposal sites
were affected little, if at all, by the disposal operation. At one of the auto assembly
waste disposal sites, high background levels of constituents in the groundwater masked
any escape of pollutants from the fixed sludge. Elevated levels of chloride, calcium and
sodium were found in groundwater under and down the groundwater gradient from the disposal
area at the second auto assembly plant. Only elevated levels of sulfate and boron were
found under and down the groundwater gradient from the electroplating waste site. Ele-
vated levels of sulfite, nitrite, cyanide, phenols, and arsenic were noted in groundwater
under and down dip from the disposal area at the refinery site. Toxic metal contamination
in groundwater was not a serious pollution problem at the sites studied at the time of
the survey.
INTRODUCTION
A recent EPA-funded survey of 50
typical land disposal sites that received
large volumes of industrial wastes showed
that 43 of the dumps, landfills or dis-
posal lagoons had caused some degradation
of local groundwater quality (1). The
moat frequently occurring inorganic con-
taminants from industrial solid waste were
selenium, barium, cyanide, copper, and
nickel. Levels of hazardous inorganic
substances that exceeded concentrations
set in Federal drinking water standards
were found at 26 sites. The most common
pollutants found in excessive amounts were
selenium, arsenic, chromium and lead.
Careless disposal of industrial solid
wastes particularly electroplating sludges,
refinery sludges or wastes containing paint
has, in the past, caused severe degradation
of groundwater quality. In Nassau County,
New York, electroplating wastes leached
chromium and cadmium into the local ground-
water forcing surrounding water supply
wells to be abandoned. Sludges containing
paint wastes were buried in an abandoned
sand and gravel pit in Maryland. Monitor-
ing wells showed that the level of hexa-
valent chromium in local groundwater rose
to 7.2 mg/liter; far above levels safe for
human consumption (2), At Alma, Michigan,
refinery wastes (containing phenols) were
buried in a pit in a shallow, glacial
aquifer. The resulting high phenol levels
in groundwater caused the condemnation of
two nearby water-supply wells (3).
The current investigation was under-
357
-------
TABLE 1. SUMMARY OF CHARACTERISTICS OF THE FOUR SITES SELECTED
Characteristic
Geographic area within the U. S.
General geologic setting
Mean annual rainfall
Mean annual air temperature
Major pollutants detected in
sludge analyses
Liner used below fill
Thickness of waste
Nature of material in
unsaturated zone
Thickness of unsaturated zone
Average hydraulic conductivity
below waste
Dates of emplacement of fixed
sludge
Type of operation
Site W
Central
Glacial Drift
102 en
12°C
Faint, putty
B, Cr, F«, Pb,
Mn, Ni, Zn
None
1.22-3.05m
(avg. 2.14m)
Sandy clay
3. 05-8. 60m
(avg. 5.60m)
l.lxlO~7cm/sec
None
0.0
1974
Diked fill
Site X
North Central
Glacial Outwash
93 cm
11°C
Electroplating
Cd, Cr, Cu, Mn,
Na, Zn
None
0.91-1. 22m
(avg. 1.07m)
Sandy clay
2. 16-2. 93m
(avg. 2.41m)
3.45xlO"7cm/sec
None
0.0
1973
Diked fill
Site V
North Central
Flelstpcene -
Lake
Terrace
88 cm
10° C
Paint, putty
Cr, Fe, Pb,
Mn, Zn
None
Thin*
Clayey sand
1.46-11. 80m
(avg. 8.79m)
2.63xlO~4cm/sec
None
0.0
1974*
Fill
Site Z
South Central
Deltaic -
Fluvial
Deposits
117 cm
21°C
Refinery sludge
Pt>, Mn, and
phenol
None
1.83-3. 20m
(avg. 3.79m)
Clay
1.04-6.630
(avg. 3.79m)
4.2xl(f7cm/sec
Clay
0.5m
1974
Diked fill and
cover
* Fixed waste was placed on the ground In April and Hay 1974; but, the major portion of the fixed
material was removed to a landfill in Jaunary 1975 when the area was regraded.
taken to determine if the stabilization
or "fixation" of industrial sludges could
be shown by field investigations to pre-
vent the loss of contaminants into sur-
rounding soil and groundwater. Four sites
where stabilized industrial sludges con-
taining electroplating, paint, or refinery
wastes had been placed were examined.
Borings were made to groundwater at each
site and samples of soil and groundwater
were collected for testing or chemical
analysis. The principal physical charac-
teristics of soil samples beneath and
outside the disposal sites were examined.
A randomization procedure was used to
detect differences between experimental
samples collected under the landfills and
down the apparent groundwater gradient and
control samples collected up the apparent
groundwater gradient. The major features
associated with the four industrial dis-
posal sites studied are summarized in
Table 1. Analyses of the fixed sludges
placed at each site are given in Table 2.
MATERIALS AND METHODS
Sampling Procedures
At each site eight to ten borings
were made to the water table using a
truck-mounted drill rig equipped with a
16.8-cm diameter, hollow-stem auger. The
locations of borings at each site are
given in Figures 1-4. Soil samples were
obtained by removing the central plug from
the auger and forcing an Hvorslev fixed-
piston sampler or a split-spoon sampler
into the soil or sediment. Care was taken
TABLE 2. CONCENTRATIONS OF MAJOR CHEMICAL
CONTAMINANTS IN THE FIXED SLUDGE
AT EACH SITE*
CoutltuMit
•
U
Cr
Cu
F*
Pb
Mn
111
S*
K*
Zn
HS
Sit* U
43
t.S
230
28
47,000
40
150
75
1
150
1300 »
0.05
Sit* X
(-•At)
35
105
1600
1650
4000
20
110
SO
1.3
1281
450
0.20
SU* Y
(••/kg)
10
0.1
30
la
23,500
40
572
30
0.62
2000
200
0.02
Site Z
25
1.3
36
12
8900
215
160
8
0.5
236
45
0.3
On • dry vt b«la.
twtti obtain** fro*
i involv*4 and mn*ljM of iludg* i
358
-------
FIGURE 1. LOCATION OF BORINGS AT SITE W
ARROWS INDICATE MOST PROBABLE DIRECTION
OF GROUND WATER MOVEMENT.
FIGURE 2. LOCATION OF BORINGS AT SITE X
ARROWS INDICATE MOST PROBABLE DIRECTION
OF GROUND WATER MOVEMENT.
to obtain soil samples at the sludge/soil
interface and near or below the water
table. Water samples were obtained by
bailing water from the borings with a
small-diameter tube lowered down the auger
Stem. Soil and water samples were shipped
and stored under refrigeration until they
could be prepared for testing or analysis.
Testing and Analytical Procedures
Soil samples were tested physically
to determine water content, dry density,
permeability, and grain size distribution.
All testing was conducted according to
standard soil engineering methods (4).
Water content was determined by weighing a
fresh soil sample and then drying the
sample at 110°C until a constant dry weight
was obtained. Sample dry densities were
FIGURE 3. LOCATION OF BORINGS AT SITE Y
ARROWS INDICATE MOST PROBABLE DIRECTION
OF GROUND WATER MOVEMENT.
FIGURE 4. LOCATION OF BORINGS AT SITE Z
ARROWS INDICATE MOST PROBABLE DIRECTION
OF GROUND WATER MOVEMENT.
determined by trimming a fresh soil sample
to a known volume, then drying and weighing
the sample. Permeabilities were run using
a constant head peraeameter for coarse-
grained materials and a falling-head test
apparatus for fine-grained soils. Grain
size analyses were conducted by passing the
sample through a standard sieve set. A
separate hydrometer analysis was performed
on a sample passing through a 200-mesh
sieve. The major characteristics of the
soils or sediments, especially the grain-
size analysis and the characteristics of
the fine fraction were used to classify
the soils into standard soil engineering
categories (5).
359
-------
The groundwater samples collected were
filtered through a 0.45-micron membrane
filter and preserved for analysis. Care
was taken to store all samples under re-
frigeration and run the analyses within the
recommended time limits (6). The tech-
niques used for chemical analyses are given
in Table 3.
TABLE 3. TECHNIQUES USED IN THE ANALYSIS
OF GROUNDWATER FILTRATES
Chemical
epeclee Procedural and/or Inacri
Loweet reporting
)
,o. 100-70+
« aa above
Determined with Envlrotech Node] No. DC SO
TOC Anelyler
rial Method
AutoBlier Atonic Abaorptlon Unit
Sana a" above
Determined with Perkln-Elaer Heatei
Atoffticr Atonic Abaorptlon Unit
Determined with Spectranetrtca Argon P]
Emlealon Spectrophotometer Model 11
Perkln-Elmer Atonic Abaorptlon Unit
Determined with a apectrametrlca Argon Plai
Eailaalon Spectrophotometer Model II
Determined with a Perktn-ElBjer Heated Grapl
Atomlier Atomic Abaorptlon Unit
Same aa above
Same aa above
Determined with a Nlaaeleangyo Zceman Shift
Atomic Abeorptlon Spectrophotometer
Determined with a Ferkln-Elmer Heated Graphite
Atomizer Atonic Abaorptlon Unit
0.01
0.01
I
<0.01
0.03
0.003
0.05
0.03
0.001
0.03
«
0.003
0.02
0.005
0.0003
0.003
0.003
0.0002
0.005
0.00?
0.005
o.ou
Public Health Aaaoclatlon, New York, nth Edition, 1971.
RESULTS AND DISCUSSION
Physical Testing of Soils
Existing engineering reports had
shown that at all sites selected, the soil
was uniform over the area of the disposal
site. The goal of the physical testing was
to see if soil characteristics had been
altered in the soils beneath the industrial
disposal operation. Table 4 summarizes the
physical testing data for samples taken
directly below the fixed sludge and for
samples taken at comparable depths outside
the disposal site. In most cases the con-
trast between samples at different loca-
tions at a disposal facility was found to
be very slight. A randomization test (7)
was applied to the data and in only two
cases were the test results found to be so
extreme as to be statistically significant
(93% confidence level). The water content
of sediments beneath the waste at site W
is significantly higher than those outside
the disposal area and the percent fines
(<200 mesh) in samples under the landfill
site X is significantly lower than corre-
sponding samples outside the disposal area.
No consistent pattern of changes in soil
character can be seen. Any changes in
physical properties are occurring in the
fixed waste itself, not in the underlying
soil.
Chemical Analyses of Groundwater
Tables 5-8 summarize the results of
the chemical analyses of groundwater
samples obtained from borings at the four
industrial waste disposal sites. In ex-
perimental borings through the disposal
sites or downdip from the disposal areas
the concentrations of major cations and
anions (Ca, Na, SO^ and Cl) were often high
suggesting contamination from the overlying
waste; yet, the trace element levels were
remarkably low. A randomization test (7)
was used to determine which constituents
were present in consistently high concen-
trations in either experimental or control
samples. The results are summarized in
Table 9.
At site W, only chloride, nitrite,
calcium and sodium showed significant ele-
vated concentrations in the experimental
borings. Nitrite levels, however, were
very close to the limit of detection for
the analytical method used and the small
variation observed does not suggest this
is a major contaminant leaching from the
waste. Concentrations of chloride, calcium
and sodium were all well above the detec-
tion limits and the contrasts between con-
trol and experimental samples was
impressive.
At site X, only sulfate, nitrate and
boron showed significant differences be-
tween control and experimental samples.
At this site, however, high nitrate levels
360
-------
TABLE 4. COMPARISON OF PHYSICAL PROPERTIES OF SOIL SAMPLES COLLECTED DIRECTLY
BENEATH AND AT COMPARABLE DEPTHS OUTSIDE THE DISPOSAL AREAS
Site
W
W
Vi
U
W
W
W
W
x
X
X
x
x
X
X
X
X
Y
Y
Y
Y
Y
Y
Y
Z
Z
Z
z
z
z
z
Boring
no.
1
2
3
A
5
6
7
8
1
2
3
4
5
6
7
8
9
1
2
3
4
b
7
9
2
3
4
5
6
7
8
Position
with respect
to disposal
srea
Inside
Inside
outside
outside
outside
outside
outside
outside
Inside
Inside
outside
outside
outside
outside
outside
outside
outside
Inside
inside
outside
outside
outside
outside
outside
inside
outside
outside
outside
outside
Depth below
surface
(m)
3.23
1.40
3.23
0.21
6.40
3.76
3.26
6.31
1.16
1.43
0.24
0.21
0.24
0.27
0.21
0.24
0.24
0.27
0.21
1.92
0.21
1.13
1.13
1.16
4.39
3, 48
3.63
2.71
3.32
3.32
2.41
- 3.76
- 1.62
- 3.60
- 0.43
- 6.75
- 3.72
- 3.81
- 6.52
- 1.49
- 1.77
- 0.64
- 0.46
- 0.67
- 0.46
- 0.58
- 0.52
- 0.55
- 0.64
- 0.76
- 2.44
- 0.64
- 1.68
- 1.68
- 1.68
- 4.72
- 3. 81
- 3.99
3 02
- 3.63
- 3.72
- 2.80
Dry
density
(8/cc)
1.92
1.73
1.99
1.75
2.13
1.92
2.13
2.05
1.70
1.56
1.78
1.71
1.59
1.88
1.81
1.62
1.69
1.73
1.73
1.75
1.66
1.73
1.57
1.82
1.48
1.43
1.45
1. 38
1.43
1.51
1.60
Water
content
(I)
14.8
16.0
12.2
12.8
9.1
13.9
9.8
10.1
20.5
21.7
18.3
18.2
24.8
14.5
17.4
21.2
20.2
18.6
17.4
19.5
22.8
17.4
22.1
17.3
30.5
32. 8
32.0
34. 8
31.4
28.8
25.4
Hydraulic
conductivity
(cm/sec)
1.
2.
3.
8.
1.
5.
3.
4.
3.
4.
8 x
2 x
2 x
8 x
1 x
1 x
4 x
0 x
0 x
5 x
10"'
10"6
l
-------
TABLE 6. CHEMICAL COMPOSITION OF GROUNDWATER OBTAINED FROM BORINGS AT SITE X
Experimental Borings
Parameters
S04
SO*
Cl3
NO -N
NO,-N
ctr
TOC
Ca
Fe
Kg
Hn
Na
As
B
Be
Cd
Cr
Cu
Hg
Ni
Fb
Se
Zn
Boring
1
272
<1
25
0.10
<0.01
0.05
5
40.00
ND
54.20
0.023
85.00
<0.003
0.04
<0.005
<0.0003
<0.003
<0.003
<0.0002
<0.005
<0.002
0.009
Se
Zn
Boring
1
500
<1
130
0.04
<0.01
<0.01
32
131.00
ND
49.80
0.137
272.00
<0.003
0.06
<0.005
'0.0003
<0.003
ND
<0.0002
0.013
0.003
<0.005
<0.014
Boring
2
950
<1
100
0.24
<0.01
ND
ND
355.00
0.336
119.00
0.550
48.90
0.010
0.05
<0.005
<0.0003
0.004
ND
ND
0.052
<0.002
0.005
0.023
Boring
5
9'JO
<1
130
0.01
'0.01
<0.01
4
338.00
0.107
115.00
0.410
77.00
0.008
0.04
<0.005
0.0006
<0.003
ND
'0.0002
<0.005
0.003
'0.005
<0.014
Boring
6
500
<1
60
1.60
'0.01
ND
ND
266.00
ND
83.60
0.006
28.50
0.006
0.03
<0.005
<0.0003
<0.003
ND
'0.0002
-------
TABLE 8. CHEMICAL COMPOSITION OF GROUNDWATER OBTAINED FROM BORINGS AT SITE Z
Experimental Borings
Parameters
SO.
SO*
Cl3
NO -N
NO,-N
CN
TOC
Phenol
Ca
?e
K
Mg
Hn
Na
As
B
Be
Cd
Cr
Cu
Hg
NI
Fb
Se
Zn
Boring
1
24
<1
585
0.251
0.04
0.02
22
0.06
206.00
0.566
2.80
37.50
1.230
271.00
0.007
0.07
<0.005
0.0028
0.003
ND
0.0004
0.014
0.004
<0.005
ND
Boring
2
23
3
1500
0.08
0.01
<0.01
24
<0.01
349.00
11.300
2.00
71.70
0.622
691.00
0.007
0.15
<0.005
0.1070
0.004
ND
<0.0002
0.028
0.002
<0.005
ND
Boring
3
158
25
1360
<0.01
0.02
0.60
419
6.90
241.00
NO
2.70
72.70
0.103
762.00
0.016
0.07
<0.005
0.0056
0.003
ND
< 0.0002
0.191
0.003
<0.005
0.112
Boring
6
14
3
390
0.13
<0.01
0.01
13
0.02
161.00
ND
1.20
17.60
0.542
249.00
<0.003
0.07
<0.005
0.0048
<0.003
ND
0.0002
<0.005
0.002
<0.005
ND
Boring
7
79
<1
1010
0.02
<0.01
<0.01
11
0.03
362.00
HD
1.80
62.40
1.050
408.00
<0.003
0.06
<0.005
0.0080
0.003
ND
<0.0002
0.006
0.003
<0.005
ND
Boring
»
163
<1
1780
2.35
<0.01
<0.01
17
0.08
354.00
NO
1.50
90.50
0.049
784.00
< 0.00 3
0.12
<0.005
0.0176
0.003
ND
0.0010
<0.005
0.003
<0.005
ND
Boring
4
<8
<1
960
0.01
<0.01
<0.01
29
0.03
226.00
HD
3.00
53.50
0.030
474.00
<0.003
0.08
<0.005
0.0078
0.003
ND
0.0002
<0.005
0.002
<0.005
ND
Control
Boring
5
20
<1
995
<0.01
<0.01
<0.01
21
<0.01
277.00
ND
2.30
62.40
0.155
464.00
<0.003
0.09
<0.005
0.0075
0.003
ND
<0.0002
<0.005
0.002
<0.005
ND
Borings
Boring
9
50
<1
660
<0.01
<0.01
0.01
13
0.01
175.00
ND
1.00
38.00
0.408
412.00
<0.003
0.17
<0.005
0.0016
0.005
ND
0.0003
<0.005
0.003
<0.005
ND
Boring
10
24
<1
1630
0.04
<0.01
<0.01
17
0.03
294.00
NO
1.10
82.90
0.201
691.00
<0.003
0.26
<0.005
0.0151
0.005
ND
<0.0002
0.006
0.003
<0.005
ND
Note: All values are in mg/i.
ND - Not determined.
were found in the control wells and not in
the experimental wells. The control wells
were drilled in a farmed area west and
south of the disposal area and may have
been contaminated by fertilizer leaching
from this agricultural land. Only sulfate
and boron levels increased under the fixed
sludge disposal area.
At site Y, the wells that were bored
on the southwest side of the sludge dis-
TABLE 9. RESULTS OF RANDOMIZATION TEST OF
CHEMICAL ANALYSES OF GROUNDWATER SAMPLES
boring* only.
„ Too [mtt tampl*m above detection lleilt.
MS - tot •laniHcunt «t 94.9« l«v«l.
S - Slgfllflcent at 94-961 level.
HD . Too a*ny •«npl«» not dettrvlned.
posal site were in close proximity to a
coal storage area for the plant boiler
house. Seepage from the coal pile probably
accounts for the very high sulfate levels
in the two upgradlent wells. The plume of
contaminated water continued under the
sludge disposal area making it difficult to
establish which pollutants were contributed
by the fixed sludge. Only boron showed a
statistically significant difference in
concentration between the control and ex-
perimental wells. Boron was higher in the
control wells than in the experimental
wells suggesting it was coming from the
coal storage area, not the sludge disposal
site.
At site Z, the sludge disposal area is
between a tank farm and a small brackish
bay. The combination of these circum-
stances results in high background levels
of many constituents in the groundwater.
The randomization test indicated the
occurrence of significantly higher levels
of sulfite, nitrite, cyanide, phenols and
arsenic in the experimental wells. The
concentrations of chromium observed also
were significant statistically; however,
the levels of chromium observed were very
close to the limit of detection for the
analytical technique used and the chromium
levels were higher in the control borings
than in the experimental borings. The
fixed sludge contained a great deal of oil
and the groundwater samples from under the
363
-------
disposal area smelled strongly of oil and
phenol. The decomposition of these oily
materials may account for the production
of sulfite and nitrite, as well as, the
persistence of cyanide that may have been
present in the refinery sludge. The
phenols and arsenic are probably migrating
directly from the fixed sludge.
Table 10 summarizes the groundwater
analyses on the basis of occurrences of
certain constituents at levels that ex-
ceeded standards for public water sup-
plies (8). The control wells in many cases
contained unacceptable levels of pollutants
as might be expected in areas of constant
industrial activity. In only a few cases
could contaminants be detected at unaccept-
able levels in experimental wells that did
not also appear at elevated levels in con-
trol wells. These cases include the
occurrence of boron at site VJ, sulfate at
site X, iron at site Y and cyanide at
site Z.
CONCLUSIONS AND COMMENTS
Physical Properties
The physical properties of soil sam-
ples from beneath the stabilized industrial
sludges did not appear to be notably dif-
ferent from soil samples collected at
similar depth in the surrounding area. The
disposed material may itself have altered
physical properties; but, there is no in-
dication of any effect on underlying soils.
Chemistry of Groundwater Samples
Changes in groundwater quality attrib-
utable to sludge disposal could be observed
in three of the four sites investigated.
At one of the two auto assembly plants, the
levels of chloride, calcium and sodium
showed significant increases under and down
the groundwater gradient from the disposal
site. At the electroplating facility,
sulfate and boron levels increased under
and down the groundwater gradient from the
disposal site. At the disposal site for
stabilized refinery sludge, increased
levels of sulfite, nitrite, cyanide,
phenols, and arsenic were found in water
samples taken under and down the ground-
water gradient from the disposal sites.
With the exception of phenols at site
Z, none of the toxic constituents predicted
to be pollution problems on the basis of
analysis of the waste material. (Table 1)
were found in significantly elevated con-
centrations in the groundwater. The ab-
sence of these pollutants may be attrib-
utable to lack of time for transport of
these materials to the groundwater, attenu-
ation effects occurring in the soils, or
binding of these constituents in the
fixation/stabilization process. Future
work in this project will concentrate on
analyses of nitric acid digests and dis-
tilled water leaches of soils in and around
the disposal sites to establish metal levels
in the soil as an indication of contaminant
attenuation.
TABLE 10. OCCURRENCES OF CONCENTRATIONS OF CONSTITUENTS IN GROUNDWATER
ABOVE STANDARDS FOR PUBLIC WATER SUPPLIES
Chemical
constituents
so
Cl*
NO -N+NO -N
CN '
Phenols
Fe
Hn
As
B
Cd
Cr
Cu
Hg
Pb
Se
Zn
No. borings
sampled
Control
borings
2
0
0
0
ND
0
3
0
0
0
0
0
0
0
2
0
4
Sice V
Experimental
borings
4
0
0
0
ND
0
3
0
1
0
0
0
0
0
i
0
4
Control
borings
0
0
0
2
ND
ND
4
0
0
0
0
0
0
0
2
0
5
Site X
Experimental
borings
4
0
0
2
ND
ND
2
0
0
0
0
0
0
0
2
0
4
Control
borings
2
0
0
0
ND
0
2
0
0
0
0
ND
0
0
0
0
2
Site Y
Experimental
borings
5
0
0
0
ND
1
4
0
0
0
0
ND
0
0
0
0
5
Control
borings
4
0
0
0
3
ND
3
0
0
1
0
m>
0
0
0
ND
4
Site Z
Experimental
borings
6
0
0
1
5
2
5
0
0
2
0
ND
0
0
0
ND
6
ND - Mot determined.
364
-------
ACKNOWLEDGEMENTS
This study is part of a major research
program on chemical fixation technology,
which is now being conducted by the U. S.
Army Engineer, Waterways Experiment Station
and funded by the Environmental Protection
Agency, Municipal Environmental Research
Laboratory, Solid and Hazardous Waste
Research Division, Cincinnati, Ohio under
Interagency Agreement, EPA-IAG-D4-0569.
Thanks are extended to Douglas W. Thompson
for technical review. Robert E. Landreth
is the EPA Program Manager for this
research area.
REFERENCES
1. Anonymous, "Industrial Dumps Threaten
Groundwater", Engineering News-Record,
October 6, 1977. p. 69.
2. Miller, D. W., DeLuca, F. A. and
Tcasier, T. L. "Ground Water Contamination
in the Northeast States," EPA-660/
2-74-056, Environmental Protection Agency,
Washington, D. C., 1974.
3. Deutsch, M. "Phenol Contamination of an
Artesian, Glacial-Drift Aquifer at Alma,
Michigan, U. S. A.," Proc. Soc. Water
Treatment and Examination, 3.1 (2): 94-100,
1962.
4. U. S. Dept. of the Army. Laboratory
Soils Testing. Engineering Manual
EM 1110-2-1906, U. S. Dept. of the Army,
Washington D. C. 1970.
5. U. S. Army Engineer, Waterways
Experiment Station. The Unified Soil
Classification System. Tech, Memorandum
No. 3-257, v. 1, USAE Waterways Experiment
Station, Vicksburg, MS., 1960.
6. U. S. Environmental Protection Agency,
Manual of Methods for Chemical Analysis^
of Water and Wastes., U. S. Environmental
Protection Agency, EPA-625/6-74-003,
Cincinnati, OH, 1974.
7. Siegel, Sidney. Nonparametric
Statistics for the Behavioral Sciences,
McGraw-Hill Book Co. New York, NY, 1956.
8. National Academy of Sciences, National
Academy of Engineering. Water Quality
Criteria, 1972, R.73.033, U. S. Environ-
mental Protection Agency, Washington D. C.,
1973.
365
-------
LAND CULTIVATION OF INDUSTRIAL WASTES
David E. Ross and Hang-Tan Phung
SCS Engineers
4014 Long Beach Boulevard
Long Beach, California 90807
ABSTRACT
Among the land disposal methods available for industrial wastes, land cultivation has
been practiced by pharmaceutical, tannery, food processing, paper and pulp, and oil
refinery industries, and offers an acceptable means of disposal for other organic wastes
as well. The soil environment can assimilate most types of industrial organic waste by
processes of adsorption, dilution, biodegradation, and oxidation. The main objectives of
this study were to gather and assess available information on land cultivation of indus-
trial sludges with emphasis on characterization of waste types, quantities, operational
technology, economics and environmental impacts. This paper summarizes results from a
state-of-the-art investigation over a 20-month period.
INTRODUCTION
The land cultivation method of disposal
involves several steps: (1) application of
waste onto the surface soil, (2) mixing the
waste with the surface soil to aerate the
mass and expose waste to soil microorgan-
isms, (3) possibly adding nutrients or
other soil amendments, and (4) remixing the
soil/waste mass periodically to maintain
aerobic conditions. This disposal method
is also referred to as landspreading, land
farming, soil incorporation, sludge farming,
and other names.
Land cultivation is being more widely
practiced as one alternative to conventional
or unacceptable waste disposal techniques.
However, very little published data are
available on the land cultivation method,
its benefits, disadvantages, and potential
environmental impacts. Accordingly, the
U.S. Environmental Protection Agency, Solid
and Hazardous Waste Research Division,
Municipal Environmental Research Laboratory,
sponsored a comprehensive state-of-the-art
study of the land cultivation practice.
The project has several basic objec-
ti ves:
t Document the extent of land culti-
vation use and project the future
use
• Identify waste types amenable for
disposal by land cultivation
• Evaluate actual or potential
impacts from land cultivation
• Assess status of state regulations
relevant to land cultivation
activities
• Prepare a conceptual design for a
hypothetical land cultivation
disposal site
366
-------
• Estimate costs to implement and
operate a site.
Information sources included personal
Interviews, published and unpublished
literature, and case study investigations.
The project is completed and final report
publication is expected in mid-1978. This
oaoer presents a summary of report findings
concerning industrial wastewater and sludge
disposal. A companion paper on land culti-
vation of municipal solid waste is pub-
lished in the Proceedings of the Third
Annual Municipal Solid Waste Research
Symposium (1).
LAND CULTIVATION - AN ALTERNATIVE TO
CONVENTIONAL LAND DISPOSAL
Soil is a natural environment for the
deactivation and degradation of many waste
materials through complex physical, physio-
chemical, chemical, and microbiological
nrocesses (2). Land cultivation of waste
is a disposal technique by which wastes are
mixed with the surface soil to promote
these processes, particularly microbial
decomposition of the organic fraction. If
managed properly, the process could be
carried out repeatedly on the surface of a
disposal site. In practice, sludges are
either hauled directly from the wastewater
treatment plant or from an interim storage
laaoon to the disposal site. The sludges
are applied to the land by spraying, spread-
ina or subsurface injection. The field is
then disced or plowed by conventional farm
cultivation equipment.
Proponents of land cultivation claim
that the site could be returned to any other
land use, including agriculture, after cess-
ation of disposal activities. Possible
disadvantages of land cultivation include
the need for relatively large tracts of
land long-term exposure of wastes to the
atmosphere, and impacts on vegetation grown
at the site.
The potential merits of land cultiva-
tion have encouraged many industries and
waste disposal contractors to implement
full or pilot scale operations. Many more
Industries are studying the concept.
Results of this study point to the potential
aDPlicability of land cultivation as a
disposal alternative to conventional
sanitary landfill, incineration, and ponding
for certain waste types, although the method
1s not a panacea for industrial waste
generators.
WASTES SUITABLE FOR LAND CULTIVATION
Table 1 lists those industries that
produce wastes considered amenable for
land cultivation. The suitability of an
TABLE 1. WASTES AMENABLE
TO LAND CULTIVATION
Food and kindred products
Textile finishing
Wood preserving
Paper and allied products
Organic fibers (noncellulosic)
Drugs
Soap and detergents
Organic chemicals
Petroleum refining
Leather tanning and finishing
industrial waste for land cultivation
depends on such characteristics as concen-
trations of chemical constituents in the
soluble as well as insoluble forms; bulk
densities of waste solids; pH, sodium, and
soluble salt contents; as well as flamma-
bility and volatility (3,4). Local
climatic conditions can influence the
viability of this disposal practice.
Excessively high moisture content may
impede oxygen transfer to soil microorgan-
isms, thereby slowing waste degradation.
Also, it is operationally difficult to
mix waste into a muddy soil. Establishment
of a vegetative cover can enhance the
disposal practice through the uptake of any
excess nutrients and water. Lacking
specific information, the Irrigation Water
Quality Criteria (5) and existing proposed
guidelines of heavy metal loading for
sewage sludges can be used as a first
approximation to determine the suitability
of industrial wastewaters and sludges for
land cultivation (4).
Industrial Wastewaters
Land cultivation of wastewaters from
food processing, pulp and paper, textile,
tannery, wood preserving, and pharmaceu-
tical industries has been practiced on a
limited scale (6). At most locations, the
practice is primarily used as a wastewater
treatment method, and little or no effort
has been made to incorporate wastewater
into the soil.
Three application methods are gener-
ally used: overland flow, slow
367
-------
infiltration, and rapid infiltration (7).
It appears that of the three methods, slow
infiltration through spray irrigation offers
the highest degree of disposal reliability
and potential for long-term site usage.
Overland flow may require operational manip-
ulations to realize the same useful site
life as slow infiltration. Rapid infiltra-
tion or groundwater recharge requires
extensive and thorough subsurface site
investigations to ensure that favorable
conditions exist.
Industrial Sludges
The information gathered indicates
that industrial sludges applicable to land
cultivation have been either organic (e.g.,
oil refinery, paper and pulp, and fermen-
tation residues), or treated inorganic
(e.g., steel mill sludge) wastes containing
low concentrations of extractable heavy
metals. When the sludge is applied to
agricultural land, it is sometimes used as
a soil amendment to improve soil character-
istics and/or low-analysis nitrogen ferti-
lizer, although straight disposal is
usually the prime goal.
Among the industrial sludges, oil
refinery wastes have been most extensively
disposed of by land cultivation (8,9). Oil
degradation rates vary, depending on
climate, oil content in the soil, and fer-
tilization. The types of oily wastes that
are disposed by land cultivation include
cleanings from crude oil, slop emulsion,
API separator bottoms, drilling muds, and
other cleaning residues (9). The sludge is
spread to a depth of about 7 to 20 cm by a
track-dozer, and then disced into the soil.
At existing sites, mixing intervals vary
from once per week over several weeks to
twice per year. The practice is strictly
for disposal; no crops or vegetation other
than weeds grow at the sites.
Some industries and disposal firms are
practicing land cultivation of hazardous
industrial sludges on a trial basis. In
most instances, the sludges have been pre-
treated to inactivate or remove the
hazardous constituents in the waste.
Waste Loading Rates
Loading rates generally depend on the
waste's BOD, total dissolved solids, heavy
metal and soluble salt contents, and the
soil's texture and drainage characteristics.
Table 2 indicates that wastes with a wide
range of chemical characteristics have been
land cultivated without apparent problems.
TABLE 2. CHARACTERISTICS OF
LAND CULTIVATED WASTES
Parameter
Maximum
Concentration/Value
BOD
COD
SS
TDS
pH
5,000 ppm
18,000 ppm
3,000 ppm
4,000 ppm
2 - 12
Most state regulations do not pre-
sently control application of waste to
soil based on the concentration of heavy
metals or other potentially toxic consti-
tuents contained in the sludge. Increased
regulation will likely occur, as land
cultivation becomes more common. Overall,
the Sanitary Landfill Criteria promulgated
by EPA for application to all land disposal
facilities will apply to land cultivation
operations.
As noted, soil can often serve as an
effective disposal sink for industrial
wastes. However, if a soil cannot assimi-
late the applied waste, it will become
anaerobic, resulting in nuisance conditions
and failure of the system to effectively
degrade the organic matter. Furthermore,
unless the wastes are detoxified or
decomposed by the soil or weather to non-
deleterious products, the upper soil zone
receiving the wastes eventually could
become loaded to its ultimate capacity. As
a result, disposal activities at the site
will have to be terminated and the site
may be unusable for alternative purposes
for many years.
Volume of Waste Suitable for Land
Cultivation
Figure 1 shows the estimated volume
of both industrial sludge and wastewater
that were suitable for land cultivation in
1975. Projected volumes for 1980 and 1985
are also indicated. These wastes represent
approximately 3 percent of all industrial
sludge and wastewater generated and pro-
jected for generation in the United States.
Thus, while the method could accommodate
specific waste streams, land cultivation
368
-------
1200
1100-
£ 1000
00
00
-------
will not significantly decrease the overall
demand for other industrial waste disposal
methods.
ENVIRONMENTAL IMPACTS
Data concerning environmental contami-
nation from industrial waste land cultiva-
tion are scarce. Improper land disposal of
industrial wastes often goes unnoticed in
the short term because the impacts are
chronic, rather than acute (10). Buried
potentially hazardous compounds may move
slowly and take decades to leach through
soil into surface and groundwater supplies
(10). Similarly, environmental impacts of
land cultivation operations may not be
readily detectable at first application and
mixing. When waste is incorporated into
the surface soil, it triggers a series of
physical, chemical, and biological proces-
ses. Phillips and Nathwani (2) have
reviewed the various mechanisms involved in
soil-waste interactions. Of significance
are the potential impacts on ground and
surface water, air quality, soil, vegeta-
tion, and human and animal health.
Mater Quality Impacts
Wastes applied to surface soils are
susceptible to washout by precipitation
runoff. Land cultivation designs generally
include appropriate drainage control facil-
ities to prevent such occurrences, including
upstream runoff diversion channels and
holding ponds or settling basins to contain
any waste-laden runoff water from precipi-
tation incident on the site. No evidence
of surface water contamination due to runoff
from a land cultivation site is documented
although flooding has been reported. Flood
problems are due to poor site selection
rather than inherent shortcomings of the
disposal method itself.
As with any land disposal operation,
groundwater quality may be impaired if
surface-applied waste constituents move
through the subsurface soil to underlying
aquifers. Soils exhibit tremendous chemi-
cal and biological attenuation and degrada-
tion capacity due to their high reactive
surface area and the available varied
microbial populations. However, if a waste
has soluble or partly soluble constituents,
there is always a risk that it will affect
the quality of subsurface waters. For
example, Adriano, et al. (11) have shown
that the nitrate and phosphate levels in
subsurface waters exceeded public health
standards and environmental guidelines,
respectively, from long-term land treatment
of food processing wastewaters.
The downward movement of heavy metals,
oils, and pesticides in soils is often
restricted, due to low water solubility,
and high retention and degradation by
various soil processes (9,12,13). Other
chlorinated hydrocarbons, phenols, and
detergent components may be present in
varying amounts in the wastes, mostly
depending on the generator industry. It
is believed that such organic compounds
will be eventually decomposed by soil
microorganisms. Precise decomposition
times are not known. Unless the soil is
overloaded with wastes containing large
amounts of these substances, land cultiva-
tion is not likely to pose a serious threat
to groundwater quality.
Air Emissions
Like municipal effluents and sludges,
certain industrial wastewaters and sludges
can also emanate odors when exposed to the
atmosphere, impairing air quality of the
disposal area. If the waste contains
volatile components, land cultivation
practices could increase evaporation of
these components, although much of the
readily volatile fraction would have come
off during waste handling prior to
delivery to a disposal site. Additionally,
during soil incorporation, dust could pre-
sent a health hazard to the personnel on
the site and may be carried off site by
winds. Subsurface injection of the waste
and/or mixing with soil as soon as prac-
tical after deposition can alleviate, but
not always eliminate, odor and evaporation
problems.
Soil Impacts
As part of this study, composite sam-
ples of surface soil and typical native
vegetation were taken from several land
cultivation sites. Samples were also
collected from the corresponding control
sites which had similar soil and vegetation,
but had received no waste. The objective
of the limited sampling program was to
determine the effect of land cultivation
on soil chemical properties and elemental
uptake. Pertinent information on two of
the sites studied is presented on Table 3.
Note that the site age, soil, and sludge
characteristics from the two sites are
markedly different. Site A is managed by
370
-------
TABLE 3. PERTINENT INFORMATION ON STUDY SITES USED IN THE INVESTIGATION
OF VEGETATIVE IMPACTS OF LAND CULTIVATION
Site A
Site B
Area of site
Waste type
Waste volume
Soil Characteristics
Depth of soil/waste
mixture
Site age
Vegetative type
Vegetation sampled
14 ha (35 ac)
Drilling muds
Tank bottoms
1,000 to 1,200 bbl daily
Sandy, well-drained,
slightly acidic
0.9 to 1.2 m (3 to 4 ft)
22 years
Weeds and shrubs are growing
along the perimeter of the
disposal area
Tall grass, golden bush,
ragweed, and ice plant
8.2 ha (20 ac)
API oil/water separator
sludges
Periodic disposal, 185,000
bbl/year
Clayey, poorly drained,
alkaline
15 to 30 cm (6 to 12 in)
5 years
Same as Site A
Nut grass leaves and cocklebur
seeds
TABLE 4
. CHEMICAL CHARACTERISTICS OF THE SURFACE SOILS
FROM CONTROL AND OIL-TREATED PLOTS
Constituents*
PH
EC, mmhos/cm
Oil, %
TKN, %
Org. C, %
P
Na
B
Mn
Ni
Zn
Se
Mo
Cd
Pb
Site A
Control
6.04
0.40
0.006
0.16
ppm
410
110
0.2
35.4
1.5
7.5
0.022
1.3
0.14
4.2
Treated
7.65
4.46
2.28
0.079
2.53
230
280
2.28
55.0
2.5
40.7
0,09
1.1
0.06
5.4
Site B
Control
7.41
2.21
0.080
2.10
ppm
17.5
185
0.2
65
4.8
53.5
0.01
0.6
0.06
212
Treated
7.40
3.91
2.06
0.134
5.10
17.5
375
0.22
71.6
5.3
71.5
0.028
0.55
0.06
242
*Electrical conductivity (EC) and B were measured in the saturation extracts;
other elements in ppm were determined in O.lN^HCl extracts.
371
-------
a disposal company, whereas Site B is lo-
cated within and managed by an oil refinery
plant.
Results of soil analyses (Table 4) show
that land cultivation of oily wastes at
Site A resulted in increased pH, soluble
salt content (expressed as EC), and levels
of total Kjeldahl nitrogen (TKN) and organic
carbon in the sandy soil. Likewise, water-
soluble B and O.lf[-extractable Na, Mn, Ni,
Zn, Se, and Pb also increased. At Site B,
similar increases were noted for soluble
salts, TKN, organic carbon, and 0.1 P[-
extractable Na, Mn, Zn, Se, and Pb. Except
for Pb concentrations at Site B, the ele-
ments analyzed had concentrations which are
within the range typically found in soils.
Impacts on the Food Chain
The most significant potential human
health impact is uptake of waste consti-
tuents by vegetation and subsequent inges-
tion of such vegetation by animals and/or
humans.
At ongoing industrial wastewater land
treatment sites, a cover crop is generally
regarded as an integral part of the system,
useful to improve water infiltration,
remove nutrients, and increase the allow-
able hydraulic loading. In disposing of
industrial sludges by land cultivation,
existing vegetation is usually removed .from
the site before waste application. Weeds
and small bushes will become established in
the disposal plot only if the plot is left
untilled for some time. Available informa-
tion indicates that crops grown on agri-
cultural soils treated with selected indus-
trial wastewaters and sludges do not
accumulate toxic metals in sufficient quan-
tities to adversely affect plant growth.
However, long-term effects of land cultiva-
tion on crop quality and the food chain are
not known.
Land cultivation of wastes can pose
hazards to the food chain through surface
contamination and plant uptake. For
example, pesticide residues have been shown
to accumulate in various crops by these
mechanisms (14), rendering the crops unsafe
for consumption.
The effects of potentially toxic ele-
ments in sewage sludges and effluents on
the food chain have been discussed in the
literature. Cadmium, Cu, and Zn are the
elements commonly posing significant
potential hazards to the food chain through
plant accumulation (15). However, it
should be noted that under improper site
management (overloading, low pH, etc.),
elements such as Pb, Hg, As, Se, Mo, and
Ni, if present in significant quantities in
the wastes that are land cultivated, could
also pose serious hazards to man and ani-
mals consuming the crops.
Limited data are available from green-
house and field investigations conducted
to evaluate the potential adverse effects
of application of some industrial sludges
on the yield and quality of crops.
DeRoo (16) evaluated the use of mycelial
sludges produced by the pharmaceutical
industry in Connecticut as a nitrogen ferti-
lizer and organic soil amendment. He con-
cluded that if the mycelial sludge is
applied repeatedly at high rates (222 mt/ha)
to the same field, the soluble salt concen-
tration and high zinc content in the sludge
may be injurious to plants. Studies with
similar objectives have been conducted
using lagooned paper pulp sludge (Jacobs,
personal communication), cannery fruit
sludge (17), and nylon sludge (Cotnoir,
personal communication). Results indicated
that these sludges would have value as a
low-analysis nitrogen fertilizer, and that
no adverse effects were observed in crops
and soils.
In field studies, the effects of a
refractory metal sludge (18) and steel mill
sludge (Nelson, personal communication) on
the yield and chemical composition of for-
age and grain crops were evaluated. No
adverse effects from heavy metals were
observed. In the case of steel mill sludge,
the growth was stunted at high waste appli-
cation rates (sludge about 20 cm thick),
which was attributed to nitrogen and phos-
phorus deficiencies and poor aeration from
soil compaction. A thorough survey and
chemical analysis of all types of vegeta-
tion at the 10 case study land cultivation
sites was not possible within the project
scope. Some typical and prevalent species
were sampled and analyzed, as highlighted
in Table 5, which summarizes data from
Sites A and B.
Results from these plant analyses
indicate differences in plant uptake due
to differences in site conditions, plant
species, and waste oil treatment methods.
At Site A, ragweed and ice plant grown on
the oil-treated plot contained higher
372
-------
TABLE 5. SELECTED RESULTS OF VEGETATION ANALYSES AT TWO CASE STUDY SITES
Site A*
Site B*
Ragweed
Ice Plant
Nutgrass
Cocklebur Seed
Element
N
P
Na
B
Mn
Ni
Zn
Se
Mo
Cd
Pb
C
~— "~* '
2.52
0.32
2560
102
88
6.9
101.4
0.04
1.2
0.05
3.12
T
i
b
0.87
0.33
ppm —
3375
96
95.4
8.5
190
0.125
0.71
0.05
18.0
C
1.29
0.21
20937
12
96.7
3.8
40.6
0.04
0.31
0.10
3.12
T
. v
0.91
0.21
ppm
36560
12
198
2.0
38.2
0.04
< 0.1
0.21
3.28
C
1.44
0.17
2062
7
63.6
1.9
93.8
0.23
7.1
0.41
61.5
T
%_ __
-._--
1.42
0.11
ppm
6187
15
48.7
6.3
131.9
0.23
9.5
0.41
90.5
C
- °L
3.05
0.29
ppm
1000
28
18.8
3.6
43.8
0.04
4.2
0.15
11.2
T
1.07
0.16
687
14
19.1
3.1
53.1
0.04
8.7
0.15
23.2
C = control plot, T = treated plot
concentrations of Na, Mn, Zn, Se, and Pb
than those from the control plot. These
observations related to the trends in soil
data. Plant N contents were reduced by the
land cultivation treatment, probably due to
immobilization of available N by the soil
microorganisms. This suggests that nitrogen
fertilizer application may be necessary for
an oily waste disposal area if subsequent
growth is planned.
Nut grass and cocklebur grown on the
oil-treated plot from Site B contained
higher concentrations of Zn, Mo, and Pb
than those grown on the control plot.
Overall, concentrations of the elements
analyzed were lower in the cocklebur seed
than the concentrations in nut grass. The
levels of Mo and Pb in both plant species
from the oil-treated plot are worthy of
notice since they have approached the
undesirable level U 10 ppm) for animal
consumption. The concentrations of Pb were
exceedingly high, comparable to the Pb
concentrations in pasture grasses in a lead-
contaminated area near Antioch, California
(Ganje, personal communication). The data
suggest that nut grass and cocklebur have
accumulated significant amounts of Pb,
particularly from the oil-treated plot.
Humans and other animals can also be
affected by land cultivation waste disposal
practices in other ways. For example, site
workers could be injured by contacting
deleterious wastes; unauthorized visitors
could inadvertently contact the waste; and
dust or fumes laden with possibly hazardous
concentrations of toxic chemicals, heavy
metals, or pathogens due to waste applica-
tion may move off site to settle on nearby
untreated fields. Also, poorly operated
land cultivation sites may be aesthetically
unpleasing. These and other adverse impacts
are possible, but no such health threatening
incidents have been documented.
LAND CULTIVATION ECONOMICS
Figure 2 shows the unit costs of
industrial liquid and sludge waste disposal
for five case study sites. Also shown is
the expected range of costs for a hypo-
thetical land cultivation site, estimated
on the basis of conceptual designs prepared
during the project. Unit cost values for
both actual case studies and the conceptual
design include expenditures for land, equip-
ment, labor, interim on-site storage of
waste, and environmental monitoring. Costs
373
-------
ct:
UJ
Q_
CO
O
/il-
ia.
J.O
1C .
Tl •
.IM
10.
ii.
in -
J.U
8.
6.
•
4-
o .
L.
©
PHARMAC
DETERGE!
EUTICAL
4
IT
k
*"*--^
MIXE
e_|
** —
) WASTE
• • •
C
DE
•-^»«i
ONCEPTU
SIGN CO,
/
z_
M
BTS
— M
OIL
©
i
F
(
ULP
10 20 30 40 50 60 70 80 90 95
Figure 2.
WASTE RECEIVED - 1000 M3 PER YR
Costs of hypothetical conceptual design for a
land cultivation site compared to case study costs.
for waste transport to the site are not
included.
Capital costs for the conceptual design
ranged from $262,000 to $746,000 for waste
delivery rates from 1,000 to 4,000 dry
metric tons per year (equivalent to 5,340
to 21,370 gal per year of sludge @ 5 percent
solids).
Costs for disposal of uncontained indus-
trial wastes by sanitary landfill methods 3
generally range from $5.30 to $22.00 per m
($0.02 to $0.08 per gal), depending on loca-
tion, site volume, and other factors. For
industrial waste disposed of in drums or
other containers, costs can be substantially
higher, ranging up to $194 per m3 ($0.72 per
gal). Thus, land cultivation cost is com-
parable with or lower than conventional
disposal methods.
MONITORING
A monitoring program for a land culti-
vation facility is best tailored to specific
waste and site conditions encountered. If
volatile wastes are to be handled, air emis-
sions should be monitored. Any crops for
human or animal consumption grown on waste-
treated soil should be periodically analyzed
to ensure that potentially harmful levels of
contaminants are not building up in edible
portions of plant tissue.
Monitoring for groundwater quality
control may be simpler and less costly at
a land cultivation site as compared to a
sanitary landfill. Since wastes are applied
only to the surface soil, routine analysis
of surface and near-surface soil samples can
indicate if and when any contaminants are
moving away from the zone of waste incor-
poration. Any movement of contaminants can
be detected and remedial action taken well
before groundwater has been contaminated.
Monitoring groundwater via wells is
generally not necessary except to establish
background water quality data prior to site
activation.
374
-------
SUMMARY AND RECOMMENDATIONS
Industrial wastewaters and sludges
that have been land cultivated are primarily
from food processing, oil refinery, paper
and pulp, tannery, and pharmaceutical indus-
tries. The wastes are composed mainly of
organic material and are thus biodegradable.
The land cultivation practice is limited
and will likely remain in applicability to
about 3 percent of all industrial wastes.
Land cultivation is viable only where soil,
climate, waste characteristics, and environ-
mental conditions permit. Depending on
specific conditions, the disposal program
can either be related to agriculture or can
be solely a disposal practice.
Existing state regulations generally
call for consideration of planned land
cultivation projects on a case-by-case
basis. However, with increasing use of
this disposal method and in light of new
EPA criteria that include coverage of land
cultivation activities, many states will
likely implement specific regulations.
Documented environmental impacts of
land cultivation operations concern loading
soil with waste constituents, plant uptake
of heavy metals, and emanation of odors.
These and other potential impacts are con-
trollable by improved operating techniques
and effective monitoring programs. In
particular, routine soil monitoring to
detect any movement of waste constituents
away from the zone of incorporation can
provide an early warning of fugitive con-
tamination. In the face of a substantial
threat to groundwater, all land cultivated
waste at a site could be removed to other
sites.
Land cultivation costs are in the
range of $2 to $18 per m3 ($0.01 to $0.07
per gal) of industrial waste, varying with
waste type, loading rates, site usage, site
area, and other factors. These costs are
generally competitive with other methods
of industrial waste disposal.
Recommendations developed on the basis
of this study include:
1. All wastes should be routinely
analyzed before and during land
cultivation operations to ensure
that an accurate record of applied
material is developed. If
potentially deleterious or incom-
patible waste is delivered for
disposal, it should not be
accepted.
2. As for all land disposal facili-
ties, a site for land cultivation
should be carefully selected to
afford maximum environmental
protection.
3. An engineering design for a land
cultivation site should be prepared
to ensure provision and proper con-
figuration of runoff control, faci-
lities, access roads, waste storage
facilities, personnel safety fea-
tures, and equipment maintenance
sheds.
4. A specific protocol for waste
receipt, spreading, mixing, re-
mixing, monitoring, and trouble
shooting should be developed and
followed as an integral part of
land cultivation operations.
5. A monitoring program is essential.
Routine soil monitoring wherein
surface and subsurface soils are
analyzed for presence of waste
constituents can detect any move-
ment of contaminants toward ground-
water. Air monitoring may also be
necessary.
6. Further research is warranted
into such topics as waste degrada-
tion mechanisms, fate of degrada-
tion products, soil retention
potential, limits of waste loading
in soils, and air quality impacts
of land cultivation.
REFERENCES
1. Phung, T., Ross, D., and Landreth, R.,
"Land Cultivation of Municipal Solid
Waste," Proceedings of the Third Annual
Research Symposium: Management of Gas
and Leachate in Landfills, U.S.
Environmental Protection Agency, EPA-
600/9-77-026, September 1977, pp. 259-
267.
2. Phillips, C.R., and Nathwani, J.,
"Soil-Waste Interaction: a State-of-
the-Art Review," Solid Waste Management
Report EPS 3-EC-76-14, Environment
Canada, October 1976.
375
-------
3. Carlile, R.L., and Phillips, J.A.,
"Evaluation of Soil Systems for Land
Disposal of Industrial and Municipal
Effluents," Report No. 118, Water
Resources Research Institute,
University of North Carolina, July 1976.
4. Epstein, E., and Chaney, R.L. "Land
Disposal of Industrial Waste,"
Proceedings of the National Conference
on Management and Disposal of Residues
from the Treatment of Industrial
Wastewaters, Information Transfer, Inc.,
Rockville, Maryland, 1975, pp. 241-246.
5. National Academy of Sciences - National
Academy of Engineering. "Water
Quality Criteria, 1972," U.S. Govern-
ment Printing Office, Washington, D.C.,
1974.
6. Wallace, A.T., "Land Disposal of
Liquid Industrial Wastes," Land
Treatment and Disposal of Municipal
and Industrial Wastewater, Sanks, R.L.,
and Asano (eds.), Ann Arbor Science
Publishers, Inc., Ann Arbor, Michigan,
1976, pp. 147-162.
7. Hunt, P.G., Glide, L.C., and
Francingues, N.R., "Land Treatment
and Disposal of Food Processing
Wastes," Conference on Land Application
of Waste Materials, Soil Conservation
Society of America, Ankeing, Iowa,
1975, pp. 112-135.
8. Kincannon, C.B., "Oily Waste Disposal
by Soil Cultivation Process," U.S.
Environmental Protection Agency, EPA-
R2-72-100, December 1972.
9. Lewis, R.S., "Sludge Fanning of
Refinery Wastes as Practiced at Exxon's
Bayway Refinery and Chemical Plant,"
Proceedings of the National Conference
on Disposal of Residues on Land,
Information Transfer, Inc., Rockville,
Maryland, 1977, pp. 87-92.
10. Lazar, E.C. "Summary of Damage Inci-
dents from Improper Land Disposal,"
Proceedings of the National Conference
on Management and Disposal of Residues
from the Treatment of Industrial
Wastewaters, Information Transfer, Inc.,
Rockville, Maryland, 1975, pp. 253-257.
11. Adriano, D.C., Novak, L.T., Erickson,
A.E., Wolcott, A.R., and Ellis, B.G.,
"Effect of Long-Term Land Disposal by
Spray Irrigation of Food Processing
Wastes on Chemical Properties of the
Soil and Subsurface Water," Journal
of Environmental Quality, 4:242-248,
1975.
12. Page, A.L., "Fate and Effects of
Trace Elements in Sewage Sludge When
Applied to Agricultural Lands," U.S.
Environmental Protection Agency, EPA-
670/2-74-005, January 1974.
13, Letey, J., and Farmer, W.J., "Movement
of Pesticides in Soils," Pesticides
in Soil and Water, Guenzi, W.D. (ed.),
Soil Science Society of America, Inc.|
Madison, Wisconsin, 1974, pp. 67-97.
14. Nash, R.G., "Plant Uptake of Insecti-
cides, Fungicides, and Fumigants from
Soils," Pesticides in Soil and Water,
Guenzi, W.D. (ed.), Soil Science
Society of America, Inc., Madison,
Wisconsin, 1974, pp. 257-299.
15. Chaney, R.L., "Crop and Food Chain
Effects of Toxic Elements in Sludges
and Effluents," Proceedings of the
Joint Conference on Recycling Municipal
Sludges and Effluents on Land, National
Assoc. State Univ. and Land-Grant
Colleges, Washington, D.C., 1973,
pp. 129-141.
16. DeRoo, H.C., "Agricultural and Horti-
cultural Utilization of Fermentation
Residues," Connecticut Agricultural
Experiment Station, Bulletin No. 750
1975.
17. Noodharmcho, A., and Flocker, W.J.,
"Marginal Land as an Acceptor for
Cannery Waste," Journal of the
American Society of Horticultural
Science, 100:682-685, 1975.
18. Poison, R.L. "Refractory Metals
Processing Waste Utilization on Dayton
Silty Clay Loam." M.S. Thesis.
Oregon State University, Corvallis,
1976.
376
-------
COST ASSESSMENT FOR THE EMPLACEMENT OF
HAZARDOUS MATERIALS IN A SALT MINE
B.T. Kown
Bechtel Corporation
San Francisco, California
R.A. Stenzel
Bechtel Corporation
San Francisco, California
R.T. Milligan
Bechtel Corporation
San Francisco, California
R.E. Landreth
U.S. EPA (MERL)
Cincinnati, Ohio
A.D. Krug
Rockwell Hanford Operation
Richland, Washington
ABSTRACT
The results of an economic evaluation of storing non-radioactive
hazardous wastes in a room and pillar type salt mine are discussed.
The results include capital and operating costs, and a cost analysis.
The cost study was based on a simulated waste characteristic and con-
ceptual design of the waste receiving, treatment, containerization,
and underground storage facilities. The storage facilities were
designed for long-term storage of hazardous wastes in solid form and
in an environmentally safe manner. The cost of storing hazardous
wastes in a salt mine depended considerably on the system capacity,
the waste characteristics, and the storage methods.
INTRODUCTION
The promulgatibn of air and water pollu-
tion control regulations has resulted in
more effective removal of contaminants
from waste streams, especially the hazard-
ous constituents in many industrial efflu-
ent streams. These cleanup activities
have resulted in an increased quantity of
concentrated hazardous wastes that must be
disposed of. Disposal of hazardous wastes
in a manner that isolates them from the
environment is becoming a difficult prob-
lem throughout the country.
In a continuing effort to find a new and
improved method of disposing of or storing
hazardous wastes, EPA has been supporting
a number of studies on the hazardous waste
disposal, including offshore incineration,
secured landfill, chemical stabilization,
encapsulation, and isolation.
One of these studies was "Evaluation of
Hazardous Wastes Emplacement in Mined
Openings" (EPA-600/2-75-040), supported
by the Solid and Hazardous Waste Research
Division (SHWRD) of the Municipal Environ-
mental Research Laboratory, EPA(l). The
study, conducted by Fenix and Scisson,
Inc., provided an assessment of the tech-
nical feasibility and environmental ac-
ceptability of emplacing hazardous indus-
trial wastes in underground mines. The
377
-------
study concluded that storing hazardous
industrial wastes in a room and pillar
typo salt mine would be an environmentally
acceptab11' method of managing hazardous
waste, provided that the recommended pro-
cedures of site selection, treatment, con-
tainer i zat ion, and waste handling are fol-
lowed. In view of this assessment EPA
decided that an economic evaluation of the
concept should he conducted. This econom-
ic evaluation is the subject of this
paper. (-)
The objective of this study then was to
provide KPA with the necessary cost infor-
mation to make sound decisions on future
commitments of resources regarding the
emplacement of hazardous wastes in under-
ground storage facilities.
The study involved four major tasks:
(I) Development of Design and Operating
Criteria
. Typ i ca 1 salt mine
. Characteristics of received waste
. Characteristics of stored material
. Required treatment
(2) Conceptual Design
. Surface facility
. Subsurface facility
(3) Cost Estimates
. Capital costs
. Operat ing costs
(•'>) Economic Analyses of Overall Opera-
t ion
DKSICN BASIS OF THE STUDY
The only known practice of underground
storage of hazardous wastes is at Herfa-
Neurodo in West Germany. At the present
time, that operation is a simple storage
facility without treatment or recontainer-
ization of received wastes. No attempt is
made to convert the wastes to a more stable
form. It was decided that for this study,
a 1resli approach based on the trend of U.S.
wasti' management practices would be devel-
oped .
The characteristics of hazardous wastes
to l)e received, waste treatment and con-
tainer izat ion, and storage concept used in
this study reflect the current U.S. hazard-
ous waste management practice and its like-
ly future changes.
Waste Characteristics
It was apparent from the literature that
at the present time the characterization
and inventory of hazardous wastes are in-
complete. It is also expected that the
characteristics of hazardous wastes (i.e.,
both quantity and composition) will be
changed rapidly as new laws and new manage-
ment programs are implemented.
From these findings, it was concluded
that the hazardous waste characteristics
to be used for this study would be of a
general nature, reflecting a wide range
of waste types, but specific enough to re-
veal the requirements of different treat-
ment and handling methods.
The hazardous wastes were classified into
four groups, each of which requires differ-
ent treatment and handling. These waste
types are defined as follows:
. Type A. Aqueous liquids and slurries
containing dissolved hazardous eleme-
ments, primarily toxic heavy metals.
Type A wastes require chemical treat-
ment before dewatering, containeriza-
tion, and storage. Type A wastes in-
clude four subtypes: chromate waste
(A-l), cyanide waste (A-2), acid/caus-
tic waste (A-3), and nonreactive waste
(A-4).
. Type B. Aqueous and organic sludges
containing solid hazardous elements.
Type B wastes require only pH adjust-
ments and dewatering before container-
ization and storage. Type B wastes
include acid/caustic sludges (B-l),
inorganic sludges (B-2), and organic
sludges (B-3).
. Type C. Inorganic and organic solids
containing solid hazardous elements
requiring only containerization before
storage.
. Type D. Special wastes to be stored
on a temporary basis at customer re-
quest. These wastes will be retrieved
and sent back to the waste generator.
This is included to develop cost data
associated with short-term storage of
the waste and its retrieval.
Study Mine
One of the study tasks was to evaluate
existing salt mines to select one specific
378
-------
typical salt mine suitable for the storage
of hazardous waste. The selected mine was
used as a base from which design and cost
information were obtained. To prevent po-
tential adverse public reaction, the study
mine has not been identified. Due to the
limited time and budget, the selection
procedure was to be based on readily
available information without actual site
surveys of the mines being considered.
Seventeen salt mines were considered.
Based on the preliminary evaluation, three
mines were considered acceptable for stor-
ing hazardous waste without any extensive
mine rehabilitation. Any one of the three
mines could have been selected as a typical
salt mine acceptable for the storage of
hazardous waste. However, one of them had
more available information, and with the
approval of the EPA Project Officer, this
mine was selected for the study.
General Description
The selected mine is conveniently located
near a major metropolitan area with a heavy
industrial output. Direct access to the
mine is provided by several railroads and
interstate highways.
Mining of the selected mine began over 50
years ago in a bed of salt at a depth of
"^r*5--rtfc
approximately 1,400 feet. The seam thick-
ness ranges from 25 to 30 feet. The on-
going monitoring program has established
that the mine is stable, and there have
been no significant changes in the mine
environment since the initiation of mining.
Two shafts provide access to the mine. The
production shaft is a 16-foot diameter,
concrete-lined shaft divided into four com-
partments, two hoisting compartments, each
with a 10 ton capacity, and two service
compartments. The second shaft, a man and
service shaft, is a 4- by 8-foot rectangu-
lar, concrete-lined shaft, divided into two
compartments.
The mining method is conventional room
and pillar. Rooms of 60 feet wide with a
ceiling height of from 18 to 27 feet are
mined at the end of a long gallery, up to
several thousand feet long. At regular
intervals along the gallery are huge pil-
lars of rock salt 60 by 80 feet, which are
the sole means of supporting the roof.
Using this system, about 65 percent of the
salt is extracted. Several feet of salt
are left to preserve the roof and several
inches to preserve the floor.
To get a better perspective of the size
of room and pillar mines, Figure 1 presents
a photo of a main haulway in a mine. The
• »*JM^~
Fig. 1 Main haulway in a salt Mine (courtesy of Int. Salt Co.)
379
-------
pillars are visible on either side of the
haulway. Figure 2 illustrates the typical
room and pillar arrangement. Conditions
are a comfortable 58°F year round with a
relative humidity of 55-56 percent.
i i
o o o a
: a a a a
ana
Fig. 2 Typical room and pillar arrangement
Mine Facilities —
Surface — Surface facility includes the
offices, warehouses, shops, and salt pro-
cessing plant. In addition, there are com-
plete truck and rail loading and unloading
facilities.
Underground — The mine itself contains
very Little in the way of permanent facili-
ties. The area adjacent to the main shaft
has been developed to serve as a shelter in
case of a mine fire. Other facilities in
the mine include an office, a machine shop,
water tanks, and the pump facilities. All
other mine facilities are of a temporary
nature and are frequently relocated as the
production areas change.
Available Storage Space —
The general mine layout is shown in Fig-
ure 3, with mining currently taking place
in one area.
An upper limit on the storage space in
the mine has been estimated from production
figures and mine maps. This, however, re-
flects the void left by mining and must be
reduced to reflect actual conditions in the
mine.
The total open space in the mine desig-
nated for the waste storage is estimated to
be 500,000,000 cubic feet. This is located
in the following areas as designated in
Figure 3.
Area
Haulways and
service areas
Total
Cubic Feet
210,000,000
70,000,000
40,000,000
70,000,000
10,000,000
100,000,000
500,000,000
This volume can be thought of as an 18-
foot deep pit, one mile square. Additional
space is available in the area of current
mining and is being created at the approxi-
mate rate of 15,000,000 cubic feet per year.
The Storage Concept
Early in the study, the concepts of re-
trievable versus non-retrievable storage
were debated. It was decided that, for
this study, underground waste emplacement
should be based on long-term secured stor-
age which makes for more efficient utiliza-
tion of the mine space. Retrieving the
long-term stored waste would be considered
only in an emergency and when no other al-
ternative was available. However, in order
to develop cost data associated with short-
term storage of the waste and its retrieval,
cases were included in which a portion of
the waste is stored temporarily and later
retrieved, i.e., brought back to the sur-
face and shipped.
General Criteria Considered: A number of
general criteria were considered in formu-
lating the storage concept. This includes:
Waste storage win be long-term storage
without planned retrieval of stored
380
-------
^ADVANCEMENT
NJN NEV/ MINIJ
Fig. 3 General mine layout
-------
waste except certain types of waste that
may be stored temporarily.
. Retrieval of the long-term stored waste
will be considered only when an extreme
emergency occurs.
. All wastes brought to the plant will be
converted to the most stable chemical
and physical forms using the best prac-
tical technology.
. Any substance that could generate toxic
fumes, fire, hazardous vapor, or explo-
sive material will be excluded.
. The storage operation will include all
activities from waste receiving at the
gate to waste emplacement in underground
storage cells.
. All wastes brought to the plant will be
stored, except the clean condensate
from the filtrate evaporation.
. There will be no effluent from the
plant, except sanitary wastewater and
clean runoffs.
. The selected storage method will provide
long-term environmental protection with-
out perpetual maintenance of the stored
materials.
. The selected storage method will use the
available space effectively.
. The stored material will be inventoried
to identify the waste type, quantity,
and storage location, in case retrieval
becomes necessary.
These criteria dictate that hazardous
wastes brought to the plant have to be
treated to precipitate any dissolved hazard-
ous constituents and then dewatered. In
addition, if the waste contains a component
that may generate toxic fumes, the waste
must be treated to remove or destroy the
component. Cyanide, for example, must un-
dergo chloride oxidation followed by lime
precipitation and filtration.
The criteria also dictate that the waste
has to be stored in a systematic manner to
allow controlled handling, segregated stor-
age of different wastes, short-term main-
tenance of stored waste, inventorying of
the stored waste, and long-term protection
of the environment. To meet these criteria,
it was decided to containerize the wastes
in open-top, 16-gauge, epoxy-lined 55-gal-
lon steel drums. The waste storage system
and overall operation are summarized in
Figure 4.
In summary, the waste received at the
storage plant will be treated to convert
hazardous elements into relatively insolu-
ble forms, then filtered to remove free
water, and finally containerized in 55-gal-
lon steel drums. The waste in steel drums
will then be hoisted into the mine on pal-
lets, each containing four drums. The pal-
lets will be transferred to the storage
cell where they will be stacked.
THE BASE PLANT CAPACITY AND ALTERNATIVE
CASES
A hazardous waste plant of 1,250 tons per
day (TPD) capacity (as received) was selec-
ted as the base case (Case 1) for this
study. This represents approximately 685
TPD of hazardous waste stored in the un-
derground mine after precipitation and de-
watering. The 685 TPD capacity selected
for the base case represents the capacity
of the existing hoisting system during nor-
mal two-shift operation of the selected
salt mine facility.
To evaluate the sensitivity of plant ca-
pacities on the design of the storage faci-
lity and the costs, two additional plant
capacities — one higher and one lower than
the base case — were also included in the
study. The high-capacity case (Case 2)
receives 1,875 TPD of hazardous wastes and
stores 1,030 TPD, while the lower capacity
case (Case 3) receives 188 TPD and stores
103 TPD. The proportions of the Type A, B
C, and D wastes in Case 2 and Case 3 are
the same as that of the base case.
The 1,030 TPD storage (Case 2) represents
an upper limit based on three-shift opera-
tion of the existing hoisting system oper-
ated at 75 percent of the design capacity.
The 103 TPD storage (Case 3) represents a
low capacity of one-shift operation of all
underground facilities.
During the course of this study, it be-
came apparent that Type A waste requires a
complicated and extensive chemical treat-
ment facility. To evaluate the concept of
receiving only non-liquid (residue) types
of hazardous wastes, a special case (Case 4)
where only Types B and C wastes are handled
is included in this study. Operation of
Case 4 will be the same as that of the base
case, except that it would require no chemi-
cal treatment and thereby reduces the sur-
face activity considerably.
It also became apparent that the contain-
er cost will be a major item of the overall
operating cost, and any storage method that
eliminates the need of containerization
would be very attractive.
382
-------
RECEIVING
UNLOADING
A, B, C, D
HOISTING
A, B, C, D
UNLOADING
A, B, C, D
RAILROAD CARS (TANK, BOX AND DUMP CARS)
TRUCKS (TANK, CONTAINER AND DUMP TRUCKS)
STORAGE
h
CHEMICAL
TREATMENT
A
B
NEUTRALI-
ZATION
B
DEWATERING
A
B
C
C
CONTAINERIZATION
A, B, C
D
O
STAGING
A1, A2, A3, A4, B (ACID), B (ALKALINE).
B (INORGANIC), B (ORGANIC),
C, D STORED SEPARATELY
TYPE A CHEMICALLY TREATED
TYPE B NEUTRALIZED
TYPES A AND B DEWATERED
TYPES A, B AND C CONTAINERIZED
ALL WASTES STORED IN STAGING AREA
WASTES HOISTED INTO MINE
TRANSFER TO STORAGE CELL WASTES TRANSFERED TO STORAGE CELLS
A, B, C, D
STORE
WASTES STORED
MONITORING
Fig. 4 Underground storage of hazardous wastes
383
-------
To explore the possibility of eliminating
containerization, several presently avail-
able waste stabilization methods (solidifi-
cation) were reviewed and one of these
stabilization methods (Case 5) is included
in this study. In Case 5, the non-liquid
wastes (Types B and C) will be dewatered,
mixed with a stabilizing additive, and then
pumped into the storage cell area in the
mine. Tables 1 and 2 summarize the five
cases in terms of the types of waste, the
quantities of waste received and stored,
and storage concepts.
SURFACE FACILITIES AND OPERATION
In the storage concept chosen for the
base case (1250 TPD waste received), a major
portion of the handling operations occur on
the surface before the material is rendered
acceptable for emplacement in the mine. A
list of the material handling steps for
Table 1 Waste Quantity and Treatment
Summary
Waste Received
TPD
Waste Stored
TPD
Storage Concept
Case 1
Case 2
Case 3
Case 4
Case 5
1250
1875
188
600
600
685
1030
103
420
420
Type A treated. Types A & B filtered,
Types A, B & C containerized,
Types A, B, C & D stored.
Sane as Case 1, except total plant
capacity.
Same as Cases 1 and 2, except total
plant capacity.
Sane as Case 1 except, Types A & D
are not included.
Sane as Case 4 except waste is stabilized
(cementized) and stored directly into mine
space.
Table 2 Waste Type Breakdown of Alternative
Cases
Type A
Type B
Type C
Type D
Solids
, TPD Received
TPD Stored
, TPD Received
TPD Stored
, TPD Received
TPD Stored
, TPD Received
TPD Stored
From Wasteuater
Case 1
600
150
400
200
200
200
50
50
85*
Case 2
900
225
600
300
300
300
75
75
130
Case 3
90
22
60
30
30
30
8
8
13
Case 4
0
0
400
200
200
200
0
0
20
Case 5
0
0
400
200
200
200
20
Evaporation, TPD Stored
Total Received
Total Stored
1250
685
1875
1030
188
103
600
420
600
420
*85 TPD wasteuater solids include 65 TPD from Type A and 20 TPD from Type B wastes
384
-------
each waste type and sub-type would require
several pages. The Intent is to give a
brief picture of the surface operations.
The plant surface process facilities are
designed to receive the four types of waste
and temporarily store, treat, containerize
and transfer them to the staging area for
lowering into the mine. A number of auxi-
liary operations are required to support
the main processing functions, e.g., mate-
rials (drums and pallets), and chemicals
handling, process steam generation, elec-
tric power distribution, wastewater treat-
nent and oily waste incineration. Adminis-
trative and maintenance functions are also
part of the surface operation.
Waste Receiving and Processing
Figure 5 illustrates in block diagram
form the main processing steps for each
waste type. All four types are shipped to
the plant by rail and by truck. Type D
wastes (retrievables) are shipped only in
55-gallon drums; the others are shipped
both in bulk containers and in drums.
Six hundred tons of Type A waste are re-
ceived per day. Bulk shipments are pumped
into storage tanks designated for storage
of the particular A Waste subtypes,
(Chromate, cyanide, acid, alkaline and non-
reactive). Drummed wastes are emptied into
1P»4»>0«T
HM(cini«,Mn J »i,»,,,.M| „.,«, I »..!
" -- I 1 •« •'-"•• I -»-^l.^»»«.M
Fig. 5 Block flow diagram of surface operation
385
-------
transfer tanks and pumped to the appropri-
ate storage tanks. A nominal four-day
storage capacity is provided for each waste
subtype. Waste containing hexavalent chro-
mium is reduced with sulfurdioxide to pro-
duce trivalent chromium. Cyanide bearing
waste is oxidized with chlorine to nitro-
gen and CC>2. Acid and alkaline wastes are
neutralized. Heavy metals are then precip-
itated as hydroxides by addition of lime.
The precipitates are dewatered by vacuum
filtration to a minimum 40 percent solids
cake for waste containerization. One hun-
dred fifty tons per day of filter cake are
produced.
Four hundred tons of Type B waste are
received per day. These are also pumped
into storage tanks designated for the par-
ticular B Waste subtypes (acid, alkaline,
neutral inorganic and organic sludges).
Acid and alkaline wastes are neutralized.
Inorganic and organic wastes are kept sep-
arate. All are dewatered or deoiled by
automatic pressure filtration to a minimum
40 percent solids cake for waste container-
ization. One hundred fifty tons of in-
organic filter cake and fifty tons of or-
ganic filter cake are produced each day.
All waste receiving and processing is two
shifts per day operation.
Two hundred tons of inorganic and organic
Type C wastes are received per day. Bulk
wastes are transferred to bulk storage
bins. Wastes in 55-gallon drums are trans-
ferred to the drummed waste storage build-
ing. These solid wastes receive no treat-
ment .
Fifty tons per day of Type D waste are
received in 55-gallon drums. These wastes,
designated for retrieval, are stored in a
building and receive no treatment.
Containerization and Staging
Five hundred eighty five tons of waste
are put in 55-gallon drums during two
shifts of operation each day. About 2,000
drums and 500 wood pallets per day are used
in the containerization facility. Six
nearly-automatic containerization lines are
operated, each in which empty drums on a
conveyer system are filled, weighed, closed,
labelled and palletized. Four-drum pallets
are picked up by fork lift trucks and
transferred to the staging area (building)
near the production shaft.
In the staging building loaded pallets
are stacked four-high in segregated pallet
banks according to the emplacement designa-
tion of waste types, i.e., inorganic (X),
organic (Y) and retrievable (Z). Only one
of these three types is normally lowered
into the mine during a shift. This permits
shipment to only one emplacement area in
the mine during a shift. The staging area
is the interface between surface and sub-
surface operations.
Figure 6 schematically illustrates the
basic surface processing steps through con-
tainerization and staging at the production
shaft.
Effluent Waste Treatment
Aqueous filtrates and other plant waste-
waters are processed in an evaporation-
crystallization-filtration system which re-
covers 85 tons per day of contaminated salt
cake which must be stored in the mine.
Water is recovered as clean condensate for
reuse and ultimate discharge (net water
output). Oily filtrates are burned in a
waste incinerator unit equipped with a two-
stage scrubber for flue gas cleanup.
Surface Site Development
The layout of the surface civil struc-
tures and buildings is shown in Figure 7
for the base case plant. Of the total 17-
acre fenced area, 15 acres are within the
present mine boundary and two acres will
have to be purchased from the local resi-
dents. Except for the production shaft,
the main shaft and some portion of the
railroad tracks, all surface structures
will be new. About 13 acres of land will
have to be cleared, including all of the
existing buildings except the shaft houses.
The surface facilities include all of the
mechanical equipment (piping, electrical,
instrumentation) and the following new
civil structures and buildings:
Truck scale office and pad
Rail scale office and pad
Tank truck unloading platform
Dump truck unloading building
Tank and hopper car unloading building
Box car unloading platform
Drum unloading platform
Chemical unloading platform
Chlorine unloading platform
Storage tank area
386
-------
CTAGIMG
t
FALUTIIIHC
CONTAINERI2ATKW
Fig. 6 Schematic diagram of surface operation
Drum waste storage building
" Waste treatment building
Filtration building
\ Containerization building
Staging building
* Administration building
* safety /medical building
Laboratory
\ Equipment storage building
Warehouse
Shops
Drum cleaning building
" Wastewater collection ponds
* Wastewater treatment and filtration
building
Boiler house
SUBSURFACE FACILITIES AND OPERATION
underground facilities include
ventilation systems, underground
area, haulways, storage cells and
tce buildings. The subsurface opera-
•fTatarts with the transfer of drummed
ci°" s from the surface staging area to
production skip in the same building.
Subsurface operations are depicted in
F1gure 8 and include:
Surface loading and lowering into
the mine
Underground unloading and staging
Hauling to storage zones
. Storage
Storage cell preparation
. Monitoring
Record keeping.
At the surface, a fork lift truck places
four pallets (16 drums) on the skip and a
utilityman secures the load. Upon a
"ready" signal, the loaded skip is lowered
1,400 feet into the mine. A lowering
cycle takes five to seven minutes - 1 1/2
minutes of actual lowering time.
At the base of the shaft, a forklift
takes the four pallets out of the skip and
sets them aside. Another forklift picks
up two pallets at a time and loads then
onto a waiting flat bed. haul truck. A
staging area near the shaft has one-day's
waste storage capacity and can be used if
there is any reason to stop direct haulage
to the storage cells.
Tractor-trailers haul 20-pallet loads
(25 tons) from the shaft area to the stor-
age zones. Round-trip distances to the X,
Y, Z storage zones (see Figure 3) vary from
387
-------
CO
.
-—I
1
1
/
»*Hlt
t1««M*
i MM |
l*»*"l
1 »it i
r ~i
L_.J
t'OOltl
Q
o
o
O
0
O
o
o
O
o
0
o
O
o
Q
O
o
O
o
O
o
"D1
o;
o
O
o
0
o
* M 11. S T A u I N f.
i '
R00u( * ON
Fig. 7 Plot plan of surface facilities
-------
SURFACE LOADING
CO
8
-rt
UNDERGROUND
STAGING
R
RECORD KEEPING
to tyXT^
UNLOADING
HAULING
MONITORING
STORAGE
Fig. 8 Schematic diagram of subsurface operation
-------
a 3,000 foot average to Zone Z to an 18,000
foot average to sections of Zone X.
Storage takes place In any of five zones:
XL X2, X3, Y, Z). Each zone is formed by
many interconnecting rooms (cells) about
60 ft. by 60 ft. by 20 ft. high. Unmined
salt barriers or ventilation stoppings
isolate the zones. In the long term stor-
age zones (X and Y), forklifts unload two
pallets at a time and stack them six pal-
lets high in the cell (Figure 9). The pal-
lets are locked together for additional
stability. An average "full" cell would
hold about 1,200 pallets, including an
allowance for side clearance and monitoring
access.
Figure 10 illustrates the short-term
storage operation - Zone Z. The storage
density is much lower in order to maintain
ease of retrieval. After an initial two-
year buildup of material (24,000 pallets),
the amount of waste sent to storage is
assumed to be offset by a similar amount
retrieved from storage and hoisted to the
surface. Storage and retrieval operations
are conducted concurrently.
A storage cell preparation is a major
underground activity. Preparation involves
removing waste salt and debris, scaling the
roof and ribs, bolting the roof and grading
and brushing the floor. A one-month supply
of prepared rooms is maintained.
Waste monitoring and recordkeeping, al-
though not discussed here, are also very
important underground activities.
Fig. 9 Schematic diagram of long-term storage operation
Fig. 10 Schematic diagram of retrievable storage (Type D) operation
390
-------
Underground Facilities
Mine rehabilitation, installation of a
ventilation system and new underground ser-
vice buildings comprise the bulk of the
civil/structural works. New subsurface
equipment are largely mobile equipment for
material hauling.
Man-Container Storage Case
In Cases 1 through 4, all wastes are
Stored in drums. To eliminate container-
ization, the waste stabilization method
(Case 5) is considered.
As in Case 4, only Type B and C wastes
are received in Case 5. The wastes are
neutralized, dewatered, and mixed with a
cementizing agent before being pumped into
the mine. In the mine, the wet mixture is
pumped into a prepared cell where it is
cured to form a solid mass. The concept of
the waste stabilization and direct storage
into the mine is shown in Figure 11.
ECONOMIC EVALUATION
The capital costs (first quarter 1977
dollars) were estimated based on lists and
specifications of equipment, buildings,
civil structures, and mine rehabilitations.
The capital costs include the following
items:
. Existing mine facilities
Site development
Buildings, civil structures, and
mine rehabilitation
. Equipment, piping, electrical, and
instrumentation
. Engineering service, allowance during
construction and contingency.
Table 3 summarizes the capital cost for
Cases 1, 4 and 5.
Table 3 Capital Cost Summary
Item
WASTE QUANTITY
Received Waste, Tons/Yr
Stored Waste, Tons/Yr
EXISTING MINE
NEW SURFACE FACILITY, DIR. FIELD COST
NEW SUBSURFACE FACILITY, DIR. FIELD COST
Base Case
Case 1
$1000's
375,000
205,500
30,000
28,013
8,229
Case 4
$1000 's
180,000
125,000
30,000
15,101
8,197
Case 5
$1000 's
180,000
126,000
30,000
12,311
8,060
TOTAL INDIRECT FIELD COST
(3 6% of TDFC
ALLOWANCE DURING CONSTRUCTION &
ENG. SERVICE @ 1% of TFC 4- $500,000 +
2,175
1,398
1,222
15% of TFC
CONTINGENCY
-------
CO
CD
NJ
Fig. 11 Schematic diagram of stabilized waste storage operation
-------
The operating costs (first quarter 1977
dollars) were estimated from conceptual
operating plans, including a list of man-
power requirements and an estimation o£
material requirements. The operating costs
include the costs of:
Direct material and labor
. Maintenance material and labor
Overhead material and labor
Taxes and insurance, depreciation,
and long-term liability insurance
Table 4 summarizes the operating cost of
Cases 1, 4, and 5.
Cases 4 and 5 eliminate the liquid waste
(Type A) and the retrievable waste (Type D)
handling. This is reflected in the much
lower costs of the surface facilities.
Table 4 Operating Cost Summary
Item
Base Case
Case 1
Case 4
Case 5
WASTE QUANTITY
Received Waste, Tons/Yr
Stored Waste, Tons/Yr
TOTAL CAPITAL COST
$1000's
375,000
205,500
90,135
$1000's
180,000
126,000
68,853
$1000's
180,000
126,000
64,041
DIRECT OPERATING COST
RAW MATERIALS & UTILITIES
DIRECT LABOR
Operating Labor
MAINTENANCET LAB & MAT.
16,828
3,695
5,208
9,203
2,533
3,400
1,048
1,658
2,635
INDIRECT OPERATING COST
ADMINISTRATION & GENERAL OVERHEAD
1,823
1,273
900
TAXES AND INSURANCE
@ 2% of Plant Cost and $1.10 per ton for
Long Term Liability Insurance 2,215 1,575
OPERATING COST 29,769 17,984
1,479
7,720
$/Ton Received
$/Ton Stored
79
145
100
143
43
61
393
-------
Operating cost (excluding cost of capital
and depreciation) is about $30 million per
year for the base case, of which 45 percent
is the cost of 55-gallon drums and pallets.
The cost of containers (drums and pallets)
alone is $65 per ton of waste stored In the
mine ($35.70 per ton received). There is a
substantial economic incentive to look at
alternative storage concepts that do not
involve containerization.
Case 5, which eliminates the use of con-
tainers, reflects a considerably lower
operating cost in terms of dollars per ton
stored.
Unit Cost Per Ton
To get a comparison of the economics, the
unit cost per ton (total cost per ton in-
cluding return on investment) was estimated
for all five alternatives based on the dis-
counted cash flow net present value method-
ology. The unit costs were computed for
private and government ownerships. The
private ownership is based on a 10 percent
return on investment, 100 percent equity,
and 48 percent income tax, while the govern-
ment ownership is based on a 100 percent
financing at 6 percent cost of capital and
no income tax. Results are summarized in
Tables 5 and 6.
Plant capacity has the expected effect on
the unit cost. As can be seen in Figure 12,
the unit cost drops by about 50 percent as
the plant capacity increases from about
50,000 to about 250,000 tons processed
(received) per year. Thereafter, the effect
on the unit cost is lessened.
Table 5 Waste Management Fee of Alternative Plant Sizes
Item
Base
Case 1
Case 2
Case 3
Waste Quantity
Received, tons/yr
Stored, tons/yr
Total Capital ($1,000)
Economic Life (Years)
375,000
205,000
90,135
30
562,500
309,000
104,075
20
56,250
30,900
61,494
40
Waste Management Fee
Private Ownership
(1)
$/ton of Received Waste
$/ton of Stored Waste
(2)
Government Ownershipv
$/ton of Received Waste
$/ton of Stored Waste
131
240
101
185
117
213
95
173
377
686
233
424
Note: (1) Private ownership assumed 10% return on investment.
(2) Government ownership assumed 6% cost of capital.
394
-------
Table 6 Waste Management Fee of Alternative Storage Methods
Item
Adjusted
Case 1
Case 4
Case 5
Waste Quantity
Received, tons/yr
Stored, tons/yr
Total Capital ($1000)
Economic Life (Years)
229,300
126,000
180,000
126,000
68,853
40
180,000
126,000
64,041
40
Waste Management Fee
Private Ownership
$/ton of Received Waste
$/ton of Stored Waste
Government Ownership
$/ton of Received Waste
$/ton of Stored Waste
210*
382*
164*
298*
179
257
131
187
118
168
71
102
*Adjusted to reflect the cost at 126,000 tons per year.
700
600
600
M
t 300
200
100
-t-
0.1 0.2 0.3 0.« 0.5 0.6
PLANT SIZE, MM TONS
(TON PROCESSED PER YEAR)
0.7 0.8
Fig. 12 Sensitivity of unit cost to
plant size
The sensitivity of unit cost to the cost
of the mine is shown in Figure 13 for a
mine cost up to $100 million. A $70 mil-
lion increase in the mine cost would add
about $40 per ton to the $131 base case
unit cost in this example.
Table 6 shows the unit costs of Cases 4
and 5 and compares them with the base case
costs. Cases 4 and 5 process the same two
types of wastes (Types B and C), but Case 4
would containerize the waste, while Case 5
would stabilize the waste to eliminate the
containers. To compare the unit costs of
these different storage concepts based on
the same underground storage loading, the
base case unit cost was adjusted to reflect
the cost at 126,000 tons per year (420 tons
per day). In the case of government owner-
ship, the Case 4 unit cost (187 per ton
stored) is almost twice of the Case 5 unit
cost ($102 per ton stored). The Case 1
unit cost ($298 per ton stored) is almost
three times the Case 5 unit cost.
The numbers seem to say it will cost a
lot to store hazardous wastes in a salt
mine. The cost can be reduced somewhat by
minimizing the needed surface treatment
facilities (though the waste generators
would incur additional costs for converting
395
-------
180
160 • •
-------
ASSESSMENT OF DEEP WELL INJECTION OF HAZARDOUS WASTE
Carlton C. Wiles
U.S. Environmental Protection Agency
Municipal Environmental Research Laboratory
26 West St. Clair Street
Cincinnati, Ohio 45268
ABSTRACT
In response to specific requirements to assess technologies available for managing
hazardous waste, a review and analysis was completed on available information related
to the injection of hazardous waste into deep wells. Based upon this information,
assessments were made as to the adequacy of this method for managing hazardous wastes
and ensuring proper environmental protection. Additional work provided detailed in-
formation on the application of the technology.
The analysis indicated that geologic and engineering data available for many areas
are sufficient to properly locate, design, and operate a deep-well injection system.
Serious problems, however, have been encountered because of failure to properly use
the data and proven engineering techniques. Information available on saline aquifer
chemistry, and the chemical and microbiological reactions within the receiving saline
aquifer was judged inadequate in many areas.
This paper will summarize the large amount of information generated during these
studies.
INTRODUCTION
The primary objective of this sympo-
sium is to provide up-to-date information
on research being conducted by the Solid
and Hazardous Waste Research Division
(SHWRD), particularly in the area of dis-
posal/management of hazardous wastes via
the land. This presentation deals more
with data gathering and assessment of avail-
able information on deep-wells than with
research. However, the deep-well injection
of industrial and/or hazardous waste is of
interest as a technology for managing liquid
hazardous wastes, and thus a discussion of
SHWRD deep-well activities is appropriate.
In 1974, in anticipation of new legis-
lative requirements in hazardous waste
management, the U.S. Environmental Protec-
tion Agency's (USEPA) Office of Solid Waste
(OSW) initiated programs to assess tech-
nologies available or nearly available for
managing hazardous wastes.1 Subsurface in-
jection of these wastes into deep wells was
considered. The OSW therefore requested the
Office of Research and Development (ORD),
SHWRD, to conduct an assessment of deep-well
injection as an environmentally sound tech-
nology for managing and controlling liquid
hazardous wastes. The agency has completed
inventories of injected wells under various
contracts.2 But the OSW had specific needs
not addressed by these inventories. One
basic need was to have available information
on deep wells assembled in one place. The
important need, however, was to make an
assessment of the environmental consequences
of Injecting hazardous wastes into appropri-
ate subsurface formations. An impossible
taskl Many have expressed real concern over
the impossibility of determining the final
disposition of the waste after injection.
They also expressed concern that no one can
397
-------
guarantee that eventual severe environmental
problems will not result. Others state that
the technology is completely acceptable.
Notwithstanding, SHWRD issued a re-
quest for proposals to conduct the assess-
ment. The contract period of only 4 months
was dictated by the timetable established
for OSW outputs. Some additional outputs
required dealing with well design, and
operation guidelines followed the assess-
ment task. It is important to note that the
assessment and the follow-up work were based
solely on available information—no labora-
tory or field investigations were performed.
This paper summarizes two studies sup-
ported by SHWRD. First, the comprehensive
review, data gathering, and assessment study
conducted by Louis R. Reeder, et. al.,3 and
second, a follow-up report by Warner and
Lehr4 on the technology and best available
practices for proper implementation of
wastewater injection into deep wells.
Considerable interest has been shown in
the assessment results, and some controversy
has arisen. The follow-up work, which will
appear in the second report, has not yet
been published, so the discussion here will
focus on the assessment findings, conclu-
sions, and recommendations.
STATUS OF DEEP-WELL INJECTION
Briefly, deep-well (subsurface) Injec-
tion normally refers to the injection of
liquid waste into selected rock formations
that are below and isolated from fresh water
aquifers. The prefix "deep" does not refer
to any specific depth. The concept of un-
derground injection of liquid waste was
probably started by the oil industry when
they began to inject waste salt brines back
into the subsurface reservoirs from which
the oil and salt brines were produced. The
first controlled use of deep-well injection
for disposal of industrial waste is diffi-
cult to determine. A look at the growth
figures (Figure 1) and the inventories con-
ducted through the years does, however,
provide some information on the historical
aspect of deep-well injection. These in-
ventories show only four such wells con-
structed before 1950, whereas the most
recent inventory (1975) listed 322 indus-
trial and municipal injection wells, of
which 209 were operating.l( The following
list shows their distribution across the
country:
— 300
— 200
D
— 100
_ 50/55 60 65 70 75 80
i I i i i i I i i i i I i i i i I i i i i I i i i i I i i i i I i
Year
Figure 1. Growth of Operating Deep-
Well Injection Systems in the United
States.
Texas 91
Louisiana 65
Kansas 30
Oklahoma 15
Industrialized
North and East
Central States 73
Other States 48
TECHNOLOGY ASSESSMENT
A governing principle of deep-well in-
jection of industrial waste is to dispose of
a maximum amount of hard-to-treat, toxic,
hazardous, or innocuous waste at a minimum
cost, with the least deleterious effects
on the environment. The term "least" may
be applied to an effect such as odor, which
is of relative importance. Assurance that
398
-------
the environmental effects are acceptable is
perhaps the major uncertainty facing the
technology.
In the Reeder study,3 the contractor's
primary task was to make an assessment of
the capability of the technology to protect
the environment by compiling and analyzing
existing data. The waste of concern for
this effort was defined as any industrial
toxic or hazardous waste. The contractor
was directed not to consider oil field
brines, similar brines, or wastewaters un-
less they were determined to present spe-
cial problems. Because the major work was
to be completed in only 4 months, the study
took the following form.
REVIEW
This task usually appears in the con-
tract scope-of-work. In this case, however,
it had tremendous significance to the ulti-
mate output. Since no field or laboratory
investigations were to be done, the Import-
ant assessment had to be based on available
information. The review included many dis-
cussions with experts involved with deep-
well activities, detailed review of avail-
able literature, and discussions with reg-
ulatory officials, State and Federal gov-
erments, and university and other re-
search organizations.
ENVIRONMENTAL AND TECHNOLOGY ASSESSMENT
Based on the reams of information
gathered through the review task, the con-
tractor assessed the ability of a deep-well
injection system to accept and manage haz-
ardous waste in a manner that would insure
protection of the environment. The contrac-
tor was directed to give precedence to en-
vironmental aspects over all other consider-
ations.
Additional details of the assessment
are useful in providing insight as to the
nature of this task. The contractor was
directed to review and technically evaluate
such factors as:
--Detection, monitoring, and control tech-
nology available or being developed to
monitor and control the waste injection
operation and system. (Detection of and
monitoring of faulty operations, breaks,
leakage, and similar problems were consi-
dered.)
--Potential for the injected waste to mi-
grate or be transported from the receiv-
ing reservoir into areas where potenti-
ally harmful environmental effects could
result.
--Geological criteria used to select well
sites.
—Well engineering design, installation,
and operating requirements.
--Waste characterization, pretreatment, and
handling requirements. (This included the
chemical, biological, and physical pro-
perties of the waste. Potential inter-
actions between the wastes and the re-
ceiving formations were considered.)
The review and the assessment tasks
were the most critical ones, but others
were also extremely important in providing
needed input to the assessment task and in
fulfilling OSW needs. Other tasks included
representative case studies depicting ex-
amples of good and bad deep-well injection
practices, compilation and characterization
of hazardous wastes being injected, assess-
ment of regulatory programs and controlling
statutes, compilation of ongoing research
projects, and recommendations for required
research and/ or demonstration activities.
TECHNOLOGY APPLICATION GUIDE
As a follow-up to the assessment study,
the Robert S. Kerr Laboratory, with support
from the Municipal Environmental Research
Laboratory (MERL), implemented and compiled
a report on acceptable practices for deep-
well injection. The effort developed in-
formation on injection well construction
and maintenance that can be used as a guide
and standard for those involved in planning,
design, construction, operation, and aban-
donment of injection wells. The report was
developed in conjunction with the National
Water Well Association.
ASSESSMENT STUDY FINDINGS
A clear presentation and discussion of
findings from the assessment study are dif-
ficult. The assessment was completed based
on available data and some subjective fac-
tors. Add to this the judgment calls re-
quired by the investigators to determine
alternatives in completing the work, and
one realizes the potential pitfalls in mak-
ing a scientific assessment. Nevertheless,
399
-------
large amounts of useful data were generated,
and some important conclusions were reached
by the study. The following is a summary of
the findings for the assessment study. Dis-
cussion is arranged according to the cate-
gories presented in the final report. The
reader is cautioned again to note the waste
of concern for the study.
GEOLOGICAL CONSIDERATIONS
Geological information available is
judged adequate to divide the United States
into the following areas:
--Areas in which deep-well injection
is feasible and may be performed
safely.
--Areas where injection is possible
but may produce detrimental effects
to the environment.
--Areas where injection is either
impossible, marginal, or will de-
finitely have detrimental environ-
mental effects.
Information presented classifies broad
geomorphic areas in the United States as to
apparent geologic compatibility with deep-
well injection. In most cases, the infor-
mation identifies areas feasible for safe
injection. Detailed geological investiga-
tions must still be made, however, for any
given well site within an area rated as
acceptable. Ten criteria are provided for
determining the feasibility of a well site
within an acceptable area. These are:
--Uniformity
--Large areal extent
--Substantial thickness
--High porosity and permeability
--Low pressure
--A saline aquifer
--Separation from fresh water horizons
--Adequate overlaying and underlaying
aquicludes
--No poorly plugged wells nearby
--Compatibility between the mineralogy
and fluids of the reservoir and the
injected waste.
A reservoir properly selected will be
able to safely contain wastes injected into
it, provided the injected volume does not
exceed the available volume of the reservoir
and injection pressures do not exceed criti-
cal formation pressures. There is an im-
portant caveat: these criteria for reser-
voir selection do not take into account
possible destruction of the reservoir inte-
grity by improper operations or by forces
of nature such as earthquakes.
ENGINEERING CONSIDERATIONS
Basically, engineering for well design
and construction is judged available and
adequate to safely handle wastes being in-
jected. The design of any given system
must be based on consideration of the ef-
fects of all factors related to the res-
ervoir, the wastes, and the physical dimen-
sions of the well. In addition, operating
limitations must be established for every
well to ensure that the limits of the equip-
ment, the reservoir, and the confining
beds are not exceeded. Such excesses
could release the waste from the well
bore or the saline aquifer.
CHEMICAL CONSIDERATIONS
If we assume proper well design and
operation, the success of injecting less
hazardous or relatively innocuous chemicals
into deep wells depends on the formation,
the fluids in the formation, and the other
chemicals in the waste streams.
Because waste streams are a mixture of
different chemicals, their activity will be
affected by all the chemicals in the
stream. If acidic waste is not neutralized
before injection or injected into a zone
containing sufficient carbonate to dissi-
pate the acid, it may be transported out of
the injection zone and into a zone contain-
ing usable mineral resources, unless the
host zone is sealed above and below by beds
impervious to acid. Other chemicals may re-
act to form plugging materials that will
damage the formation and prevent migration
of waste away from the well bore. Each
waste stream should be evaluated to deter-
mine whether or not there is a better or
safer method of disposal. If possible,
some of the most hazardous chemicals and
solids should be removed before injection
if their removal and disposal can be done
safely.
Hazardous materials with long persis-
tence periods may cause unacceptable risks
to the environment if disposed of in deep
wells. The long storage periods required
to reduce the chemical hazard to an accept-
able level increases the risk that the waste
will escape from the storage area because
400
-------
of system failure or because of migration
of the fluids.
Long persistence and a very high human
and ecological hazard rating make certain
chemicals unacceptable for disposal in deep
wells. Because these chemicals either do
not degrade or have long persisence times,
they pose a long-term potential hazard to
underground water supplies. These chemicals
include acrolein, arsenic and arsenic com-
pounds, cadmium and cadmium compounds, car-
bon disulfide, cyanides, diazinon and other
pesticides, fluorides, hydrocyannic acid,
hydrofluoric acid, hexavalent chromium com-
pounds, mercury and mercury compounds, and
nitrophenol.
MONITORING
Although monitoring is judged as rela-
tively effective in some operations of deep-
well operations—especially in detecting
equipment malfunction—an inherent weak-
ness of available monitoring systems 1s that
the detection of problems is usually after
the fact. Monitoring wells may adequately
detect waste passage within a reservoir,
but it usually cannot confirm the escape
of waste or vertical fluid movement.
CONTENTS OF THE FOUR-VOLUME REPORT
The four-volume report is divided as
follows:
Volume I (215 pages) contains an ex-
ecutive summary and details on the geologic,
engineering, chemical, microbiological, and
other aspects important in the environment-
al and technological evaluations of inject-
ing hazardous waste into deep wells. Nara-
tive detail is provided on waste charac-
terization, well inventory, related case
histories, research projects, economics of
deep-well systems, legal considerations,
and legislative, and regulatory considera-
tion. Also provided is a detailed glossary
of terms.
Volume II (316 pages) contains Appen-
dices A, B, and C. Appendix A is the Bib-
liography of literature used during the
study. Appendix B is a compilation of
patents related to deep-well systems. Ap-
pendix C provides available characteriza-
tion profiles of chemical wastes being
injected into deep wells.
Volume III (561 pages) contains Ap-
pendix D, which is the contractor's inven-
tory of industrial waste injection wells
through December 31, 1974. The inventory
incorporates information gathered from pre-
vious inventories and supplemented by data
collected from State agencies or individual
operations. Location plates are included
for the Texas and Louisiana wells, and the
calculated radius of invasion is presented
for wells where sufficient data were avail-
able.
Volume IV (413 pages) contains Appen-
dices E, F, G, H, I. and J. Appendix E
presents limited case histories of actual
industrial waste injection well operations
and provides examples of unacceptable and
acceptable operations. Recent research on
microbiological aspects of subsurface in-
jection is discussed in Appendix F. Appen-
dix G is a summary of research related to
deep-well injection, and Appendix H is a
summary of research dealing with treatment
of hazardous wastes. Legislation, regula-
tions, and policies governing deep-well in-
jection operations are provided in Appendix
I. As a result of EPA review of this docu-
ment, there were a number of questions and
issues raised as to the conclusions made
versus those warranted based on available
information. Appendix J was added by EPA
to give readers the benefit of clarifying
information that resulted from that review
and from the contractor's response.
AN INTRODUCTION TO THE TECHNOLOGY OF
SUBSURFACE WASTEWATER INJECTION
This report is in the publication proc-
ess. As the title suggests, it deals with
the technology and its proper application.
From an environmental protection point of
view, this report is an important addition
to the available literature on deep wells..
It was developed by the National Water Well
Association, in conjunction with the EPA.
In addition, to ensure that important views
of the State and Federal governments and
industry be reflected in the document, a
panel of experts guided the development of
the manual. These individuals were all ex-
pert in some facet of subsurface injection
and helped to assure that the resulting
document will stand the critical review of
the profession. The planned use is as a
guide and standard for those involved in
the planning, design, construction, opera-
tion, and abandonment of injection well
systems.
401
-------
A most efficienct method of communicat-
ing what Is contained in the report is to
reproduce portions of the Abstract and In-
troduction to that report as follows:
When wastewater is injected
into deep wells for disposal it
can pose a serious environmental
threat unless the injection proc-
ess is carefully planned and exe-
cuted from start to finish.
Local geologic and hydrologic
conditions must be thoroughly in-
vestigated including such charac-
teristics as structure, stratigraphy,
composition and engineering pro-
perties of the underlying formations.
The nature of injection and confining
intervals will determine the condi-
tions that the injection wells must
meet, or whether such wells are even
feasible. Specific information ac-
quired through the use of cores,
probes, and other tests, will help
to pinpoint areas of potential dif-
ficulty, and should suggest ways
to avoid these problems.
Once an injection site has
been selected, the injection inter-
val must be tested to insure*that
it is physically, biologically, and
chemically compatible with the waste-
water to be injected. Both the in-
jection interval and the wastewater
must be examined to guarantee that
each will remain stable over an ex-
tended period of time. If problems
exist, the wastewater can be treated
to make it more compatible with the
injection interval. Failure to bring
the wastewater and injection inter-
val into compatibility can lead to
excessive corrosion, clogging, well
and plant damage, and may necessit-
ate well abandonment.
The injection well itself must
be carefully designed and construct-
ed to guarantee the safety and inte-
grity of the injected wastewater as
well as of the surrounding forma-
tions and nature resources. When
construction of the injection well is
completed, the well should undergo
final testing to establish records
of baseline conditions for future
reference and comparison. At this
time, operating procedures and
emergency precautions should be
established and approval should be
obtained from the necessary regula-
tory agencies. Only then can full-
scale wastewater injection begin.
An operating injection well
should be monitored throughout its
working life for any changes in in-
jection conditions that may lead to
system failure. An injection well
operator has the responsibility of
knowing what and where the injected
wastewater is, and for keeping adequ-
ate operating records. When an injec-
tion well system permanently ceases
operating, the well must be properly
sealed, and a record describing the
method and date of sealing and the
precise location of the well should
be filed with the proper authorities.
When the guidelines for injection
well operation set forth here are
followed, the safety and success of
this method of wastewater disposal
will be insured.
An Introduction to the Technology
of Subsurface Wastewater Injection has
been prepared to assist engineers, geo-
logists, and others in the tasks of
planning, designing, constructing,
operating, and regulating industrial
and municipal wastewater injection
well systems. It is apparent to any-
one reviewing the literature that a
great deal has been written about this
subject in the past 25 and parti-
cularly the past 10 years. Also,
there is an extensive literature
base in the related fields of petro-
leum engineering and ground water
hydrology that can be applied to
Injection well technology. One pur-
pose of this publication is, therefore.
to provide a summary of selected in-
formation in a form convenient for
use by well operators, engineering
consultants, and regulatory authori-
ties in performing their respective
tasks, so that injection wells may be
used, where desirable, more efficient-
ly and with a minimum potential for
environmental damage.
402
-------
Impetus for development of An_
Introduction to the Technology of Sub-
surface Wastewater Injection was pro-
vided by passage of Public Law 92-500,
the Federal Water Pollution Control
Act Amendments of 1972 and Public Law
93-523, the Safe Drinking Water Act of
1974. Both of these laws contain
specific provisions concerning waste-
water injection wells. Public Law
92-500 requires that in order to
qualify for participation in the Na-
tional Pollution Discharge Elimination
System permitting program, states must
have adequate authority to issue per-
mits which control the disposal of
pollutants into wells. Therefore, it
1s only technically necessary that a
state have the authority to issue or
deny permits to qualify. However, to
do this it is necessary to have a
program for permit evaluation. The
Safe Drinking Water Act requires that
EPA develop regulations for state
underground injection control programs.
The objective of the law is to insure
that underground injection does not
endanger drinking water sources.
This guide is a technical document
intended to complement the required
EPA regulations.
Included in the technical guide
are chapters concerning the units of
measurement used in injection well en-
gineering, the nature of the subsur-
face geologic and hydrologic environ-
ment, the means of acquiring subsurface
geologic and hydrologic data, the
criteria used for injection well site
evaluation, the physical and chemical
properties of wastewater, wastewater
classification, pre-injection waste-
water treatment, injection well design
and construction, the procedures pre-
paratory to injection, well operation
and monitoring, and system abandonment.
The flow of these chapters is approxi-
mately in the order that the material
is used during the planning, construct-
ing, operating, and abandonment of an
Injection system. References used in
the text are listed and appendices
included at the end of each individu-
al chapter.
TO INJECT OR NOT?
A decision to use deep-well injection
for disposal of hazardous wastes must de-
pend on many factors. If one keeps in
mind the nature of the waste under consid-
eration during the assessment study, in
conjunction with the overriding environ-
mental considerations, there appear to be a
number of statements in the report that
contradict the basic conclusion that haz-
ardous wastes can be safely injected. The
following conclusion reached by investiga-
tors is an example:
Deep-well injection systems for
nearly all types of nonhazardous and
hazardous industrial waste are a safe
method of handling the waste if the
systems are properly located, designed,
operated, managed and regulated. Ten-
tatively excluded as being safe for
deep-well injection if containment is
questionable are 13 chemicals and
their related compounds indent!fled in
the Assessment section.
But the following conclusions seem to
contradict this:
Weaknesses appear in limited
areas in chemistry, microbiology, en-
gineering, and geology in relation to
deep-well injection.
The effects of escaping wastes,
or products of waste degradation on
ground-water quality and aquatic life
in surface water is essentially un-
known .
Reactions between waste, formation
water, and formation minerals and the
persistence period of compounds under
varying conditions is not well under-
stood. A better understanding of these
reactions is essential if a good as-
sessment of the overall effect of deep-
well injection can be made.
If one strictly follows the guidelines
established for this study, the stated
shortcomings of deep-well injection would
preclude assessing the technology as envi-
ronmentally acceptable for many hazardous
wastes. Again, however, one should consi-
der that the definition of "environmentally
acceptable" is not clear. It is a relative
matter and will remain so until some abso-
lute criteria are established.
403
-------
There appears to be no black or white
answer. Problems that have developed with
deep-well systems can usually be traced to
failure of making proper use of available
geologic information, ignoring such informa-
tion, or failure to follow proven engineer-
ing design and well-completion practices.
Implementation of proper practices
appear to provide a degree of assurance that
the injected waste will not harm the en-
vironment. Documents such as An Intro-
duction to the Technology of Subsurface
Wastewater Injection will aid this situa-
tion.
Even so, there are many who insist
that, after a waste stream is injected un-
derground, there is practically no control
over what happens to it or where It goes.5
EPA's policy vis-a-vis subsurface in-
jection is provided in the Federal Regis-
ter, August 31, 1976. In summary, this
policy is designed to meet three goals: (1)
To protect the subsurface from pollution
caused by improper injection or from im-
properly located wells; (2) to ensure that
engineering and geological safeguards are
adequate during all phases of subsurface in-
jection operations, including preliminary
investigations, design, construction, opera-
tion, monitoring and abandonment; and (3)
to encourage development and use of better
alternatives. EPA opposes waste injection
without strict controls. The method should
be used only if all other reasonable waste
disposal alternatives have been explored
and found unacceptable. EPA recognizes
subsurface injection only as a temporary
means of disposal, and as new technology
becomes available, a change to more envir-
ronmentally acceptable methods will be re-
quired.
SUMMARY STATEMENT
This paper presents only summary in-
formation contained in two voluminous re-
ports dealing with subsurface injection of
waste. One report3 deals specifically
with hazardous wastes and attempts to as-
sess the technology's ability to dispose
of hazardous wastes without adverse effects
to the environment. The other report1* pro-
vides details on the proper application of
the technology and represents an extremely
useful addition to the literature. In-
terested persons should refer to the full
reports for details.
REFERENCES
1. Disposal of Hazardous Wastes. Report
to Congress. U.S. Environmental Pro-
tection Agency, SW-115. Prepared by
the Office of Solid Waste Management
Programs. 1974.
WAPORA, Inc. Compilation of Industrial
and Municipal Injection Wells in the
United StatelT EPA-520/9-74-020. U.S.
Environmental Protection Agency, Wash-
ington, D.C., 1350 pp. 1974.
Reeder, L. R. , J. H. Cobbs, J. W.
Fields, Jr., W. D. Finley, S. C.
Vokurka, and B. N. Rolfe. Review and
Assessment of Deep-Well Injection of
Hazardous Waste. EPA-600/2-77-029a.
June 1977.
Warner, D. L. and H. L. Day. AH
Introduction to the Technology of
surface Wastewater Injection. U.S.
Environmental Protection Agency Re-
search Grant R-803889. (In press.)
5. Ottinger, R. S., J. L. Blumenthal,
D. F. Dal Porto, G. I. Gruber, M. J.
Santy, and C. C. Shih. Disposal Proc
cess, Descriptions, Ultimate Disposal
Incineration, and Pyrolysis Processes
Recommended Methods of Reduction, Neu
tralization, Recovery or Disposal
Hazardous Waste. U.S.
of
Environmental
Protection Agency, EPA-670/2-73-053c,
Vol. 3, 249 pp. 1973.
404
-------
DEVELOPING PRACTICAL METHODS FOR CONTROLLING EXCESS PESTICIDES
Charles J. Rogers
U.S. Environmental Protection Agency
Municipal Environmental Research Laboratory
26 West St. Clair Street
Cincinnati, Ohio 45268
ABSTRACT
Many hazardous materials are manufactured for specific industrial and agricultural
applications. These materials often are applied for beneficial uses, and are accidentally
spilled, or released through waste streams into the environment.
To reduce the impact of toxic and hazardous materials (whether in liquid, semi-
liquid, or solid form) on the environment, several techniques have been developed and new
promising technologies are continuously being developed. Ongoing research directed to
determining if specially designed disposal pits can be used in an environmentally
acceptable manner to control unwanted pesticides will be discussed.
INTRODUCTION
Promising technologies are continuous-
ly being developed to reduce the impact of
toxic and hazardous materials (whether
liquid, semi-liquid, or solid) on the en-
vironment. Although in the past excess
materials have been disposed of by the
quickest, easiest, and most economical
means available, a recent increase in the
supply of agricultural chemicals and, more
importantly, an increase in concern for the
environment places new emphasis on treat-
ments needed prior to disposal.
Specialized techniques investigated by
the U.S. Environmental Protection Agency
(USEPA) for detoxification of toxic and
hazardous materials include wet oxidation,
chlorinolysis, sulfonation, catalysis, in-
cineration, microwave plasma, encapsula-
tion, and experimental disposal pits that
prevent the loss of unwanted material while
allowing biodegradation. The last, a pro-
mising technique for treatment of pesti-
cides used by farmers and agricultural and
commercial applicators, is discussed in
this paper.
PIT DISPOSAL OF EXCESS HAZARDOUS MATERIALS
Ongoing work at Iowa State University
includes development of a simple technique
for the disposal of pesticides left unused
in ground or aerial application operations.
At present the safety and effectiveness of
the various disposal systems in use by
applicators have not been documented to the
extent that USEPA can recommend this tech-
nique over others.
Since 1967, Iowa State Univeristy has
developed and used a simple plastic-lined
disposal pit at the Agronomy and Agricul-
tural Engineering Research Station near
Ames. In 1973, a more sophisticated con-
crete-lined pit was constructed at the
Horticultural Research Station.
There has been no monitoring of either
pit for effective degradation, possible
loss by volatilization, or possible concen-
tration of hazardous pesticides. To obtain
adequate data on the disposal pits, a USEPA
grant was awarded October 1975 to Iowa
State University—specifically, to evaluate
405
-------
the effectiveness of pits currently in use
and to determine the factors and parameters
for improving the technique.
Some of the pesticides used at the
Horticulture Station and deposited in the
specialized pit are shown in Table I.
TABLE I
CHEMICALS USED AT HORTICULTURE STATION
Small amounts of leftover diluted
materials were deposited.
Insecticides
Carbaryl 50%(sevin) Toxophene E.L.(6 Ib/gal)
Chlordane 50%
Kelthane 35%
Pyrethrum 20%
Cythion 25% W.P.
Guthion 50% W.P.
Lannate 90%
Heptachlor 30.2%
Xylene 44%
Herbicides
Daethai
2,4-D
Paraquat
Benlate
Bravo 6F 54%
Captan
Ami ben 23% W.P.
Tenoran 50% W.P.
Casoron
Fungicides
Ma neb
Zineb
Sulfur
The chemical and microbiological stud-
ies, although incomplete, have revealed
that many of the pesticides disposed in the
pits can still be detected. Chemical analy-
ses show pesticide concentrations in ppm
for the horticultural farm disposal pit as
sampled in March and June 1977.
After approximately a 1-year study of
the pit. the major finding is that there
has been no buildup of pesticide residues.
This conclusion was substantiated by chemi-
cal and microbiological data. It is of
interest to note that microbial numbers
were similiar to what would be expected in
a pit of this type, whether or not pesti-
cides were present. The flora, however,
were less diverse than would be expected in
an open pit. Another predictable finding
was that the pesticides in the pit were
segregated. Samples taken from seven loca-
tions in the pit established that inhomoge-
neity existed.
As would be expected, a wide range of
analytical results were obtained. These
results were due in part to adsorption of
pesticides to particles of soil. For this
reason, the data could not be used to ac-
curately calculate the total amount of
pesticides discharged in the pit. The data
were used, however, to establish trends and
fate of the discharged pesticides.
The microbial flora identified in the
pit consisted almost entirely of aerobic
and facultatively aerobic bacteria; molds
and yeast were present in low numbers and
no strict anaerobes were detected. If a
correlation with pesticide concentration
and bacterial count existed, it was a trend
toward higher counts in the areas contain-
ing greater pesticide concentrations.
OZONE TREATMENT
In an attempt to accelerate the pes-
ticide degradation rates, a study was ini-
tiated to investigate ozone treatment of
pesticides. In a review of the literature,
several references to ozonation were found
and used to establish experimental proto-
col.
In laboratory studies on ozonation,
an ozone generator in the Engineering De-
partment of Iowa State University was used.
The necessary experimental apparatus for
ozonation of pesticide solutions was pro-
cured and attached to this equipment.
Operational conditions were established
for the convenient determination of the
possible reaction of selected pesticides
with ozone. Ozonations were conducted for
two hours at pH 5 and 13.5 for each of the
test pesticides. The procedures allowed
different solvent systems but 1:1 acetone/
H20 was used for most of the tests. Pes-
ticide concentrations normally employed
were 10-2g/«, in 2i solution volumes. Vis-
ual observations of color changes and pre-
cipitate formation, and chemical analyses
for the depletion of the pesticide were
used to determine which donations were
successful.
Visual observations generally provided
a convenient means of determining whether
ozonation had any effect on the degradation
of selected pesticides. With some addition-
al work, a very rapid and useful screening
procedure could be developed. For our pre-
liminary study of ozonation as an aid to
degradation the procedure as now used is
406
-------
satisfactory.
OZONATION OF SELECTED PESTICIDES
In this preliminary work eight chemi-
cals were individually ozonated; aldrin,
propachlor, alachlor, parathion, atrazine,
treflan, endosulfan I and dimethyl aniline.
The results of these tests are summarized
In Table II where the reaction results were
determined by chemical analysis before and
after ozonation at the listed pH values.
Atrazine, alachlor and treflan were
also ozonated at higher concentrations in
ethanol and these solutions were investi-
gated to determine any improvements in
degradation rates.
TEST THE BACTERIOLOGICAL ACTIVITY AFTER
OZONATION OF SELECTED PESTICIDES
Bacteriological sampling of the pes-
ticide dispsoal pit containing alachlor and
atrazine yielded pure cultures of one ala-
chlor degrader and three atrazine degraders.
All the isolates proved to be gram negative,
rod-shaped bacteria except one of the atra-
zine Isolates that is a yeast. Each Iso-
late is able to grow on the pesticide as
Its sole source of carbon and energy.
Quantitative growth studies in liquid
media showed that the alachlor isolate grew
exceptionally well in shake flasks, but the
three atrazine isolates grew as well in
stationary cultures as in shake flasks.
The three atrazine isolates are being
used to examine the effect of ozonation
upon atrazine degradation. Quantitative
growth studies in broth to compare the rates
of growth and the quantity of growth on
ozone-treated and untreated atrazine are in
progress. A preliminary qualitative growth
study suggested that ozonation of atrazine
may produce substances that inhibit the
growth of the three isolates. The organic
quantitative study should clarify this point
for atrazine and other treated pesticides.
MINIPIT STUDY
In order to determine the degree of
chemical alteration of pesticides in the
horticultural disposal pit, a minipit ex-
periment is being conducted as part of the
larger project by personnel from the De-
partments of Agricultural Engineering,
Agronomy, Bacteriology, Botany and Plant
Pathology, Energy and Mineral Resource Re-
search Institute, and Entomology at Iowa
State University.
The minipit experiment consists of
fifty-six (56) thirty (30) gallon polyethyl-
ene garbage cans partially buried 1n the
ground and arranged in seven 8-can rows or
a 7 x 8 grid. Four rows contain herbi-
cides (alachlor, atrazine, 2,4-D, and tr1-
fluralin), with one herbicide per row in
different amounts and concentrations. Two
rows contain insecticides (carbaryl and
parathion), one per row, again in different
amounts and concentrations. The last row
contains mixtures of all six pesticides.
The eight garbage cans per row alternately
contain 300 gram, 0.5% (w/w) and 15 gram,
0,025% (w/w) pesticide samples. The 0.5%
(w/w) concentration represents a reason-
able, practical application concentration.
The 0.025% (w/w) concentration approxi-
TABLE II
RESULTS OF OZONATION TESTS
Compound
aldrin
propachlor
alachlor
parathion
endosulfan I
treflan
dimethylaniline
atrazine
pH=5
Reaction at
positive
negative
negative
negative*
negative
slight
some
some
pH=13.Q
not tested
negative
slight
negative*
positive
positive
positive
positive
Products
dieldrin, unknowns
unknown
unknown
unknown
unknown
"heterogeneous reaction mixture probably responsible for negative results here.
407
-------
mates the dilution one might obtain by
rinsing the tank on equipment used to ap-
ply the pesticide.
Half of the cans contain 0.1% (w/v)
peptone as a nutrient. The contents of
half the cans (some with, some without
nutrients) are aerated using a small com-
pressor, a polyvinyl chloride manifold,
and tygon tubing leading to each fritted
glass aerator. The entire area is pro-
tected by a wire fence and warning signs.
Garbage can lids are kept in place between
sampling periods.
Analyses and evaluations for both the
mini pits and large pit will be completed
by November 1978. It is expected that the
results will show whether a low-cost, im-
proved pit disposal system can be develop-
ed and recommended for wide-scale use in
the disposal of unwanted pesticides.
SUMMARY
The pesticide disposal pit (macro)
located at Iowa State University's horti-
culture station has been in use for seven
years. Water and soil samples were ana-
lyzed during the 1976-77 growing season for
chemical, biological and microbial activity
and content. The major finding, as sub-
stantiated by chemical and microbiological
data, -is that no build-up of pesticide
residues in the pit occurs. Sampling and
analyzing of the pit at various locations
revealed that inhomogeneity of pesticide
distribution existed. Consequently, mini-
pit experiments were designed to determine
trends in pesticide degradation, and the
overall fate of pesticides disposed of in
the pits.
It is expected that at the conclusion
of 1978, sufficient data will be available
for full-scale demonstration of this dis-
posal technique.
REFERENCE
1. U.S. Environmental Protection Agency
Grant No. R804533-02, "Development of
Safe Disposal Methods for Disposal of
Excess Pesticides Used by Farmers and
Agricultural Applicators." (Awarded
October 1975 to Present.)
408
-------
Attendance List
Fourth Annual Research Symposium
Land Disposal of Hazardous Waste
March 6-8, 1978
St. Anthony Hotel
San Antonio, Texas
ABBRUSEATO, Frank A., Tenco Services, Inc., 102 N. Main Street, Bel Air,
Maryland
ADAMS, J.T., Jr., Atlantic Richfield Co., 1403 Antigua, Houston, Texas 77058
ABERLEY, Richard C., Brown and Caldwell, Consulting Engineers, 1501 North
Broadway, Walnut Creek, California 94596
ALBRECHT, Oscar, U.S. Environmental Protection Agency, Solid Hazardous Waste
Research Div., Municipal Environmental Research Lab., 26 West St. Clair
Street, Cincinnati, Ohio 45268
ALBRIGHT, Fred R., Dr., Lancaster Laboratories, Inc., 2425 New Holland Pike,
Lancaster, Pennsylvania 17601
ALOIS, Hussein, Department of Natural Resources & Environmental Protection,
Div. of Hazardous Materials & Waste Management, Pine Hill Plaza,
Frankfort, Kentucky 40601
ALLISON, H.G., Dow Chemical Company, B-2606 Dow Chemical Company, Freeport,
Texas 77541
ANDREWS, Douglas, 1320 South Fifth St., Springfield, Illinois 62703
Andrews Engineers
ARTHUR, L., Hoffmann-La Roche Inc., 340 Kingsland Street, Nutley, New
Jersey 07110
AMSTUTZ, Ray W., Williams Brothers Process Services, Inc., Resource Sciences
Center, 6600 s. Yale Ave., Tulsa, Oklahoma 74136
AVANT, Robert V., Jr., Texas Department of Agriculture, Box 12847, Austin,
Texas 78711
BAKER, stu, Arizona State Dept. of Health Ser., 4600 N. 68th St., #307,
Phoenix, Arizona 85251
BAKER, Weldon M., Texas Department of Health, 1100 W. 49th Street, Austin,
Texas 78756
BATCHELDER, Francis J., American Electric Power Service Corp., 82 Braunsdorf
Road, Pearl River, New York 10965
409
-------
BEAUTROW, Phillip A., Ventura Regional County Sanitation District, 181 South
Ash Street, P.O. Box AB, Ventura, California 93001
BECK, Stephen R., Werner & Pfleiderer Corporation, 160 Hopper Avenue
Waldwick, New Jersey 07463
BEDINGER, C.A., Jr., Southwest Research Institute, 3600 Yoakum, Houston,
Texas 77006
BEEMER, Rebecca L., Dow Chemical Company, P.O. Box 1398, Pittsburg, California
94565
BERN, Joseph, Bern Associates, Three Parkway Center, Pittsburgh, Pennsylvania
15220
BEST, Charles N., Procon, Inc., 30 UOP Plaza, Des Plaines, Illinois 60016
BILLHAM, John A., Dow Chemical of Canada, Ltd., P.O. Box 3030, Sarnia Ontario,
Canada N7T 7MI
BLANCHETTE, Dennis R., Dept. Env. Prot., 122 Washington St., Hartford, Ct
06116
BOLTZ, David G., Bethlehem Steel Corporation, 1831 Levering Place,
Bethlehem, Pennsylvania 18017
BOX, Stephen W., Southwest Research Institute, 3600 Yoakum Blvd., Houston,
Texas 77006
BRNICKY, Larry, U.S. Environmental Protection Agency, U.S. EPA - 1201 Elm,
Dallas, Texas 75270
BROWN, Donald P., Battelle Columbus Labs, 505 King Avenue, Room 11-5028,
Columbus, Ohio 43201
BROWN, K.W., Texas ASM University, College Station, Texas 77840
BROWNING, Larry, Environmental Protection Agency, EPA Water Supply Reg. VI,
1201 Elm Street, Dallas, Texas 75270
BRUSH, Charles P., Koppers Co., Inc., 2818 Koppers Bldg., Pittsburg, Pennsylvania
15219
BUOL, Bill, Chanslor-Western Oil and Development Company, 10737 Shoemaker
Avenue, Santa Fe Springs, California 90670
BUERCKLIN, M.A., Sun Oil Company, Box 2039, Tulsa, Oklahoma 74102
BURD, R.M., Rice Div/NVS Corp., 1910 Cochran Road, Pittsburg, Pennsylvania
15220
BYER, Philip H., Dept. of Civil Engineering, University of Toronto, Toronto,
Ontario, Canada M5S 1A4
410
-------
CALLEGARI, W.A., Am-Tex Corporation, 4134 S. Kirkwood, P. o. Box 725, Alief,
Texas
CALLOWAV, J. Michael, Chemfix, Inc., Post Office Box 1572, 1675 Airline High-
way, Kenner, Louisiana 70063
CAMPBELL, Donald L., Office of Project Manager for Cml Derail & Installation
Restoration, Aberdeen Proving Ground, Maryland 21010
CARLSON, Diane M., Michigan Air Quality Division, Dept. of Natural Resources,
3121 Rayborn Drive, Lansing, Michigan 48910
CARNES, Richard, U.S. Environmental Protection Agency, Solid and Hazardous
Waste Research Div., Municipal Environmental Research Lab., 26 West
St. Clair Street, Cincinnati, Ohio 45268
CARRIER, David, Woodward Clyde Const., 3 Embarcadero Center, San Francisco,
California 94111
CHAFFIN, Raymond L. (Bud), Champlin Petroleum Company, P.O. Box 9365, Fort
Worth, Texas 76107
CHARBONNET, Wilfred H., State of Louisiana, P.O. Box 60630, New Orleans,
Louisiana 70160
CHASE, Eileen, Conservation Chemical Co., Kansas City, Missouri 64108
CHIAN, Edward S.K., Environmental Engineering, University of Illinois, 3217
CEB, Univ. of Illinois, Urbana, Illinois 61801
CHOING, Eduardo, Department of Environmental Quality, P.O. Box 1760,
Portland, Oregon 97207
CHOU, Sheng-Fu., Dr., Illinois State Geological Survey, Natural Resources
Bldg., Urbana Illinois 61801
CLARK, Robert R., University of Illinois, 4129 Civil Engineering Bldg., Univ.
of Illinois, Urbana, Illinois 61820
COLBURN, Edwin, Texas ASM University, Soil & Crop Science Bldg., Texas ASM
Univ., College Station, Texas 77843
COLON, Sol Luis, PPG Industries (Caribe), Box 7572, Ponce, Puerto Rico 00731
COOP, Phillip G., United States Testing Company, Inc., 3765 Premier Cove,
Memphis, Tennessee 38118
COOPER, R.H., Salsbury Laboratories, 2000 Rockford Road, Charles City,
Iowa 50616
CORBIN, Michael H., Roy F. Weston, Inc., Weston Way, West Chester,
Pennsylvania 19380
COUNTERMAN, Paul R., New York State Dept. of Environmental Conservation,
Div. S.W. Mgmt., NYSDEC, 50 Wolf Road, Albany, New York 12233
411
-------
COWLEY, Donald, Arbuckle Regional Development Authority, 319 E. Main,
P.O. Box 95, Davis, Oklahoma 73030
COX, Don P., Temple Industries, P.O. Drawer N, Diboll, Texas 75941
CROWDER, Donald G., 4104 Taft Park, Metirie, Louisiana 70002
DANIELS, Stacy, Dow Chemical Company, Midland, Michigan
DAVIDSON, James M., University of Florida, 2169 McCarty Hall, Soil Science
Dept., Univ. of Florida, Gainesville, Florida 32611
DAVIS, Jack E., Tuscaloosa Testing Labs, Inc., P.O. Box 1094, Tuscaloosa,
Alabama 35401
DeBAENZ, Theodore, Owen & White, Inc., P.O. Box 66396, Baton Rouge,
Louisiana 70896
DERDYNE, Mike, Browning-Ferris Industries, P.O. Box 3151, Houston, Texas 77001
DeROCHE, Lawrence, City of Austin, Office of Environmental Response Mgt., P.O.
Box 1088, Austin, Texas 78701
DeVERA, Emil R., California Dept. of Health, 2151 Berkeley Way, Berkeley,
California 94564
DILLARD, Frank J., Dillard and Associates, 4631 Shetland Lane, Houston, Texas
77027
DOLAN, Daniel R., Dept. of Natural Resources & Environ. Protection, Div. of
Hazardous Material & Waste Management, Pine Hill Plaza, Frankfort,
Kentucky 40601
DRYE, Robert J., North Carolina Dept. of Human Resources, P.O. Box 28407,
Raleigh, North Carolina 27611
DUFFY, John J., The Analytic Sciences Corp., 6 Jacob Way, Reading,
Massachusetts 01867
DUVALL, Don, University of Dayton Research Institute, 4358 Sillman Place,
Kettering, Ohio 45440
DYER, John H., Allied Chemical, P.O. Box 2120, Houston, Texas 77001
DYER, Robert H., Gulf Coast Waste Disposal Authority, 910 Bay Area Blvd.,
Houston, Texas 77058
EMRICH, Grover H., Dr., A.W. Martin Associates, Inc. 900 West Valley Forge
Road, King of Prussia, Pennsylvania 19406
ESCHER, E. Dennis, Penn Environmental Consultants, Inc., 1517 Woodruff Street,
Pittsburgh, Pennsylvania 15220
FARMER, Walter J., Dr., University of California, Riverside, Dept. Soil and
Environmental Sciences, Riverside, California 92521
412
-------
FARQUHAR, Grahame J., University of Waterloo, Waterloo, Ontario, Canada
N2L 3G1
FONTENOT, Martin M., Ciba-Geigy Corp., P.O. Box 11, St. Gabriel, Louisiana
70776
FOSTER, R.C., INCON Industrial Division, Turner Collie & Braden, Inc., P.O.
Box 22123, Houston, Texas 77027
FRY, Z.B., U.S. Army Engineers Waterways Experiment Station, 1315 Baum Street,
Vicksburg, Mississippi 39180
FULLER, Wallace H., University of Arizona, Soils, Water and Engineering,
Tucson, Arizona 85721
FULTON, Joe, City Public Service, P.O. Box 1771, San Antonio, Texas 78296
GARMON, Roland C., Texaco, Inc., 3600 Normandy, Apt. A-9, Port Arthur,
Texas 77640
GARNER, Kenneth L., Kirby Forest Products, Inc., Rt. 1, Box 367, Silsbee,
Texas 77656
GARZA, Noe, City of Edinburg, P.O. Box 1078, Edinburg, Texas 78539
GENETELLI, Emil J., Dr., Princeton Aqua Science, 789 Jersey Avenue, New
Brunswick, New Jersey 08902
GHASSEINI, Masood, Dr., TRW, 1 Space Park Bldg., R-4 Rm 1128, Redondo Beach,
California 90278
GIBB, James P., Illinois State Water Survey, P.O. Box 232, Urbana,
Illinois 61801
GILLEY, William F., Virginia State Department of Health, 109 Governor Street
(Madison Building), Richmond, Virginia 23219
GLENN, B.E., Jr., Texas Utilities Services Inc., 1708 Russell Glen, Dallas,
Texas 75232
GOEBEL, Joseph E., Minnesota Geological Survey, 8121 Long Lake Road, Mounds
View, Minnesota 55432
GOELZ, John C., Sewerage Commission of the City of Milwaukee, P.O. Box 2079,
Milwaukee, Wisconsin 53201
GOLDEN, Dean M., Electric Power Research Institute, 3412 Hillview Avenue,
Palo Alto, California 94303
GOODWIN, Richard W., Researd-Cottrell, P.O. Box 750, Bound Brook, New Jersey
08805
GRAY, Kent, Utah State Division of Health, Bureau of Solid Waste Management,
150 West North Temple, P.O. Box 2500, Salt Lake City, Utah 84110
413
-------
GREEN, Clois L. , Alcoa, Box 472, Rockdale, Texas 76567
GRIEB, Henry E., GTE Sylvania, Inc., 816 Lexington Avenue, Warren,
Pennsylvania 16365
GRIFFIN, R.A., Dr., Illinois State Geological Survey, University of Illinois
Natural Resources Bldg., Urbana Illinois 61801
GUADAGNINO, Nat, Southwest Research Institute, 6220 Culebra Road, San
Antonio, Texas 78284
HAHN, William L., American Colloid Company, 5100 Suffield Ct., Skokie,
Illinois 60076
HAM, R.K., University of Wisconsin, 3232 Engineering Bldg., Madison,
Wisconsin 53706
HANKE, Albert R., Jr., State of Florida, Department of Environmental Regula-
tion, 2600 Blairstone Road, Tallahassee, Florida 32301
HARTBARGER, Kenneth, SCA Services, 1838 North Broadway, St. Louis, Missouri
HARVEY, Carolyn, Union Carbide Corp., P.O. Box 471, Texas City, Texas 77590
HARVEY, Zol B., Jr., Environmental Waste Systems, Inc., 1209 6th Avenue, South,
Texas City, Texas 77590
HAULSEE, Ronald E., ICI Amercias, P.O. Box 208, Goldsboro, North Carolina
27530
HAXO, Henry E., Jr., Matrecon, Inc., P.O. Box 24075, Oakland, California 94623
HAYS, Al, Environmental Science & Engineering, Inc., 5000 E. Ben White Blvd,
(#111), Austin, Texas 78741
HEFLIN, L.S., Calgon Corporation, P.O. Box 1346, Pittsburgh, Pennsylvania 15230
HIBBS, Minor Brooks, Texas Department of Water Resources, 1619 Glencrest,
Austin, Texas 78723
HOLMES, Robert, Redland Purle, Ferry Lane, Rainham Essex, England
Hoppe, P.A., Jr., D. Ralph Caffery & Associates, Inc., 411 Wall Street,
Lafayette, Louisiana 70506
HOULE, Martin J..U.S. Army Dugway Proving Grounds, Chemical Lab. Division,
Dugway, Utah
HSI, Meng, Univ. of Illinois, 502 W. Main St., Urbana, Illinois 61801
HUGHES, Christopher F. , Edison Electric Institute, 1140 Conn Ave., NW,
Washington, D.C. 20036
414
-------
HUGHES, D.L., State of Arkansas, Dept. of Pollution Control & Ecology, 8001
National Drive, Little Rock, Arkansas 72209
HUGHES, F.E., Jr., City of Hopewell, Department of Public Works, 300 North
Main Street, Municipal Building Room 322, Hopewell, Virginia 23860
HUNNEWELL, Dorothy S., Dept. of Environmental Quality Engineering, 600 Wash-
ington, Street, Room 320, Boston, Massachusetts 02111
HUSSAIN, Sajjad, Gutierrez, Smouse, Wilmut & Assoc., Inc., 11171 Harry
Hines/Suite 113, Dallas, Texas 75220
JOHNSON, Charles, Magna Corporation, 2434 Holmes Rd., Houston, Texas 77051
JOHNSON, Jaret C., JFB Scientific Corporation, 2 Jewel Drive, Wilmington,
Massachusetts 01887
JOHNSON, Michael E., Pfizer, Inc., Eastern Point Road, Groton Connecticut 06340
JOHNSON, Robert, Texas Department of Health, 1100 W, 49th Street, Austin,
Texas 78756
JOHNSON, Sandra L., Arthur D. Little, Inc., 20 Acorn Park, Cambridge,
Massachusetts 02140
JOHNSTON, W.R. (Dick), Denis Ranch Company, P.O. Box 638, Vancourt, Texas 76955
JONES, Larry W., Dr., D.O.A., Waterways Exp. Sta., Botany Dept., University of
Tennessee, Knoxville, Tennessee 37916
JURGENSEN, Connie, Forsyth County Environmental Affairs, Dept., Room 206,
Government Center, Winston-Salem, North Carolina 27101
KERRIGAN, James E., Amax Environ. Services, Inc., 4704 Harlan St., Denver,
Colorado 80212
KLIPPEL, Richard W., Calocerinos & Spina Consulting Engineers, 43 Trelign
Drive, North Syracuse, New York 13212
KNUDSEN, Gerald W., North Dakota State Department of Health, 1200 Missouri
Avenue, Bismarck, North Dakota 58505
KOWN, B.T., Bechtel Corp., San Francisco, California
KRAUSE, Rolf G., Shipley Company, Inc., 2300 Washington Street, Newton,
Massachusetts 02162
KULKARNI, Ram K., U.S. Army Med. Bioeng. R&D Lab. Environmental Prot. Res.
Div., 4207 Briggs Chancy Road, Beltsville, Maryland 20705
KUPIEC, Albert R., Penn Environmental Consultants, Inc., 1517 Woodruff Street,
Pittsburgh, Pennsylvania 15220
415
-------
LANCIONE, Robert L., JBF Scientific Corporation, 2 Jewel Drive, Wilmington,
Massachusetts 01887
LANCY, L.E., Lancy Division of Dart Industries, Zelienople, Pennsylvania 16063
LANDRETH, Bob, U.S. Environmental Protection Agency, Solid and Hazardous Waste
Research Div., Municipal Environmental Research Lab., 26 West St. Clair
Street, Cincinnati, Ohio 45268
LARIMER, Frank, Oak Ridge National Laboratory, Comparative Mutagenesis,
Biology Division, Oak Ridge, Tennessee
LATTA, George, Olin Corporation, P.O. Box 2896, Lake Charles, Louisiana 70605
LAWSON, Becky, U.S. Environmental Protection Agency, Solid and Hazardous
Waste Research Div., Municipal Environmental Research Lab., 26 West
St. Clair Street, Cincinnati, Ohio 45268
LEAH, Timothy D., Environment Canada, 26-11 Crown Hill Pi., Toronto, Ontario
M8Y 4C6
LEE, Young I., AWARE Engineering, Inc., 8556 Katy Freeway, Suite 128, Houston,
Texas 77024
LEWIS, M., Hoffmann-La Roche,Inc., 340 Kingsland Street, Nutley, N.J. 07110
LINN, L.G., Jr., State of Alabama Health Dept., 716 West Shawnee Drive, Montgomery,
Alabama 36107
LISKOWITZ, John W., New Jersey Inst. of Tech., 323 High Street, Newark, New
Jersey 07102
LISOT, Larry, NEDLOG Technologies, 12191 Ralston Road, Arvada, Colorado 80004
LISTER, Robert, City Public Service, P.O. Box 1771, San Antonio, Texas 78296
LONGNECKER, Tom, TAMU, TAMU Res & Est Center, Box 10607, Corpus Christi, Texas
78410
LOVELL, John B., Lovell Engineering Associates, 4505 Lake Forrest Drive, N.E.,
Atlanta, Georgia 30342
LOWENBACH, Bill, Mitre Corp., 1820 Dolly Madison Blvd., McLean, Virginia 21001
LUBOWITZ, H.R., TRW Systems, 4060 W. 132nd Street, Apt. B, Hawthorne, Cali-
fornia 90250
LUTTON, Richard J., Department of the Army, Waterways Experiment Station,
Corps, of Engineers, Vicksburg, Mississippi 39180
MACFARLANE, P. Allen, U.S. Environmental Protection Agency, 1846 Ohio, Lawrence,
Kansas 66044
MADSEN, Stuart, University of Texas at Austin, 505 C W 37th, Austin, Texas 78705
416
-------
MALECKI, Ron F., Dow Chemical Co., Michigan Division, Midland, Michigan
48640
MALONE, Philip G., Department of the Army, Waterways Experiment Station,
Corps, of Engineers, Vicksburg, Mississippi 39180
MANUEL, Robert J., ADI, Consulting Engineers, P.O. Box 226, Bay St. Louis,
Mississippi 39520
MARTINIERE, John P., Jr., Roy F. Weston, Inc., 798 Corundum Court, Stone
Mountain, Georgia 30083
MARTIEN, R.F., Vulcan Materials Co., P.O. Box 227, Geismar, Louisiana 70734
MARSH, Don, VT Solid Waste Programs, Box 489, Montpelier, Vermont 05602
MASLANSKY, Steven P., Malcolm Pirnie, Inc., Consult. Env. Engs., 52 Avon Circle,
Port Chester, New York 10573
MATHES, John, John Mathes & Associates, Inc., P.O. Box 330, Columbia, Illinois
62236
MATHEWS, Dean, Texas Municipal Power Agency, 600 Arlington Downs Tower,
Arlington, Texas 76011
MATTHEWS, Michael, Tennessee Valley Authority, 6522 Harbor View Dr., Hixson,
Tennessee 37343
MAXWELL, Shirley F., Georgia Environmental Protection Div., 38 La Rue PI.,
N.W., Atlanta, Georgia 30327
MAYO, Francis T., U.S. Environmental Protection Agency, Solid and Hazardous
Waste Research Div., Municipal Environmental Research Lab., 26 West
St. Clair Street, Cincinnati, Ohio 45268
MEHTA, Anil, Montana Tech Alumni Foundation, Mineral Research Center,
P.O. Box 3708, Butte, Montana 59701
MELLOTT, Philip, Nishna Sanitary Service, Inc., 706 West 2nd, Red Oak, Iowa
51566
MERIFIELD, Paul M., U.C.L.A., 3436 Wade St., Los Angeles, California 90066
MERKEL, James, City Public Service P.O. Box 1771, San Antonio, Texas 78296
METRY, Amir A., Roy F. Weston, Inc., West Chester, Pennsylvania 19380
MEYER, G. Lewis, U.S. Environmental Protection Agency, 9520 Beach Mill Road,
Great Falls, Virginia 22066
MIHELICH, D.L., Williams Bros, Urban Ore, 4344 E. 72 Street, Tulsa, Oklahoma
74136
417
-------
MILLAN, Renato C. , Wisconsin Dept. of Natural Resources, P.O. Box 7921,
Madison, Wisconsin 53707
MILLER, Mike, Div. of Land & Noise Pollution Control, 2200 Churchill Road,
Springfield, Illinois 62706
MITCHELL, Richard P., Espey, Hutson and Associates, Inc., 3010 S. Lamar,
Austin, Texas 78704
MOHN, N.c., Combustion Engineering, Inc., 1000 Prospect Hill Road, Windsor,
Connecticut 06095
MOLES, R.J., Oklahoma State Health Dept., 930 S. Boulevard, Apt. 239, Edmond,
Oklahoma 73034
MOLINA, J.C., Diamond Shamrock Corp., 13 Litchfield Lane, Houston, Texas 77024
MOOIJ, Hans, Environmental Conservation Directorate, Ottawa, Ontario,
Canada K1A OH3
MOORE, William, Missouri Dept. of Natural Resources, 2010 Missouri Blvd.,
Jefferson City, Missouri 65201
MOUTREY, Curtis E., Moutrey S Associates, Inc., 929 NW 72, Oklahoma City,
Oklahoma 73116
MULLEN, Hugh, IU Conversion Systems, Inc., 3624 Market Street, Philadelphia,
Pennsylvania 19104
MULLIN, Patrick B., Mobil Chemical Company, P.O. Box 3868, Beaumont, Texas
77704
MULLINS, J.M., Celanese Chemical Company, Inc., P.O. Box 937, Pampa, Texas
79065
MURPHY, G. Robert, U.S. General Accounting Office, 10020 47th Ave., S.W.,
Seattle, Washington 98146
MURPHY, William J.H., Illinois Inst. for Env. Quality, 309 W. Washington,
Chicago, Illinois 60606
MURRAY, David E., Reitz S Jens, Inc., 111 S. Meramec, St. Louis, Missouri
63105
McCLEARY, Gloria, Schneider Consulting Engineers, 98 Vanadium Road, Bridge-
ville, Pennsylvania 15017
McCLURE, J.E., Dept. of Natural Resources S Environ. Protection, Division of
Hazardous Material S Waste Management, Pine Hill Plaza, Frankfort,
Kentucky 40601
McCOMBS, Don, U.S. Environmental Protection Agency, Region IV, 1537 Idlehour Way,
Tucker, Georgia 30084
418
-------
McGUIRE, Jerry N., Monsanto Company, 800 N. Lindbergh (Mail Code F3EB), St.
Louis, Missouri 63166
McMAHON, Martin E., Phillips Petroleum Co., 10-Cl-Phillips Bldg., Bartles-
ville, Oklahoma 74004
NAIRN, T.M., Jr., Cosden Oil & Chemical Company, P.O. Box 1311, Big Spring,
Texas 79720
NELSON, Joseph, GTE Sylvania, Inc., 60 Boston St., Salem, Massachusetts
01970
NELSON, Richard K., Burke Rubber Co., Div. of Burke Industries, 2250 South
10th Street, San Jose, California 95112
NELSON, R. William, BCS Richland, Inc., 825 Jadwin Avenue, M/S R8-20,
Richland, Washington 99352
NOVAK, Rudy, Hydroscience 9041 Executive Park Drive, Knoxville, Tennessee
37919
O'CONNOR, William A., Hennepin County, 320 Washington Ave., South, Hopkins,
Minnesota 55343
O'LEARY, Laurence, International Joint Commission, 100 Ouellette Avenue,
Windsor, Ontario, Canada N9A 6T3
OTT, Randy R., Onondaga County, Dept. of Drainage & Sanitation, 125 Elwood
Davis Road, N. Syracuse, New York 13212
OZUNA, Nick, AWARE Engineering, Inc., 8556 Katy Freeway, Suite 128, Houston,
Texas 77024
PADDEN, Thomas J., Environmental Protection Agency, 2007 Freedom Lane, Falls
Church, Virginia 22043
PALMER, Linda L., Chevron U.S.A., Rm 1872, 575 Market Street, San Francisco,
California 94105
PARKS, Chris, Dowell Division of Dow Chemical Co., Box 21, Tulsa, Oklahoma
74102
PARKS, John C., Burns £ McDonnell Engineering Co., P.O. Box 173, Kansas City,
Missouri 64141
PATTERSON, Leland W., Tenneco, Inc., Box 2511, Houston, Texas 77001
PERRY, Larry D., N.C. Dept. of Agriculture, Rt. 4, Box 344-E, Zenulon, N.C.
27517
PETROS, James K., Jr., Union Carbide, P.O. Box 186, Port Lavaca, Texas 77979
419
-------
POLLOCK, William, Tippett & Gee, 502 N. Willis St., Abliene, Texas 79603
POUNDS, William F., Dept. of Environmental Resources, Solid Waste Management,
P.O. Box 2063, Harrisburg, Pennsylvania 17120
PITCH, James, AquaAir Labs, P.O. Box 7048, Charlotte, North Carolina 28217
RAPPS, Michael W., Chemical Waste Management, 900 Jorie Blvd., Oak Brook,
Illinois 60521
RATJE, J.D., Shell Oil Co., P.O. Box 2633, Deer Park, Texas 77536
REIMER, Bryce L., Aerojet Services Company, 5242 Ridgegate Way, Fair Oaks,
California 95628
REIMERS, Robert S., Dr., Tulane University, Tulane Riverside Research Labs,
Building A-1, Belle Chasse, Louisiana 70037
RICHARDS, Christina D., Oklahoma State Department of Health, 7833 N.E. 10th,
Apt. 260, Midwest City, Oklahoma 73110
RILEY, John, Environmental Protection Agency, 401 M St., S.W., Washington,
D.C. 20460
RINEBOLD, Gene, Wes-Con, Inc., Box 564, Twin Falls, Idaho 83301
ROSS, David, SCS Engineers, 4014 Long Beach Blvd., Long Beach, California 90807
ROTCHEY, Kenneth S., Environmental Protection Agency, Region VII, 6214 NW
51st Terr., Parkville, Missouri 69151
ROBERTS, Beverly K., IU Conversion Systems, Inc., P.O. Box 331, Plymouth
Meeting, Pennsylvania 19462
ROBERTSON, Duane L., Montana Dept. of Health and Environmental Sciences,
1400 11th Avenue, Suite A, Helena, Montana 59601
ROGERS, Wallace M., S.D. Warren Company, Div. of Scott Paper Co., 89 Cumberland
Street, Westbrook Maine 04092
RDULIER, Mike H., U.S. Environmental Protection Agency, Solid and Hazardous
Waste Research Div., Municipal Environmental Research Lab., 26 West
St. Clair Street, Cincinnati, Ohio 45268
ROURKE, Robert V., University of Maine, 29 Sunrise Terrace, Orono, Maine 04473
ROYAL, Allison, Resource Industries of Alabama, Inc., P.O. Box 1200,
Livingston, Alabama 35470
RUBEY, Wayne A., University of Dayton Research Institute, 621 Damian Street,
Vandalia, Ohio 45377
SABEL, Gretchen, Systems Technology Corporation, 7833 Stonehill Dr., Cincinnati,
Ohio 45230
SAMSONOV, Alexander E., TRW Systems, 4922 Bindewald Road, Torrance, California
90505
420
-------
BANNING, Donald E., U.S. Environmental Protection Agency, Solid and Hazardous
Waste Research Div., Municipal Environmental Research Lab., 26 West
St. Clair Street, Cincinnati, Ohio 45268
SANTINI, Gary S., Allied Chemical Corporation, P.O. Box 1139R, Morristown,
New Jersey 07960
SCHERMER, Joy, University of Missouri, 503 Locust, Columbia, Missouri 65201
SCHNEIDER, Mark H., Delmarva Power & Light, 800 King Street, Wilmington,
Delaware 19899
SCHOFIELD, John T., Stablex Corporation, "Oaksend", Copsem Lane, Oxshott
Surrey, KT 22 ONZ, England
SCHOMAKER, Norbert B., U.S. Environmental Protection Agency, Solid and
Hazardous Waste Research Div., Municipal Environmental Research Lab.,
26 West St. Clair Street, Cincinnati, Ohio 45368
SCHROEDER, Henry C., U.S. Environmental Protection Agency, 1860 Lincoln
Street, Suite 103, Denver, Colorado 80295
SCHULER, Ray, Right Turn, 1007 N. Stratford Rd., Arlington Heights, Illinois
60004
SCHURTZ, Gerald, Kennecott Copper Corp., Salt Lake City, Utah
SCULLIN, Ron, Public Technology, Inc., 1140 Conn. Ave. N.W., Washington, D.C.
20036
SELIM, H.M., Dr., Agronomy Department, Louisiana State University, Baton Rouge,
Louisiana 70803
SERINO, John, IBM Corporation, 1000 Westchester Avenue, White Plains, New
York 10604
SHARP, David A., Battelle Columbus Laboratories, 505 King Ave., Columbus Ohio
43201
SHEFCHIK, Bill, Burns & McDonnell Engineering Co., Inc., P.O. Box 173, Kansas
City, Missouri 64141
SHILLINGTON, Warren, Wes-Con., Inc., Box 564, Twin Falls, Idaho 83301
SHULTZ, Dave, Southwest Research Institute, 6220 Culebra Road, San Antonio,
Texas 78284
SILVIERA, David J., Dr., Battelle, Pacific Northwest Laboratories, Battelle
Boulevard, Post Office Box 999, Richland, Washington 99352
SIMMONS, Donald W., National Steel Corporation, 2800 Grant Building, Pittsburgh,
Pennsylvania 15219
421
-------
SISK, Steven W., U.S. Environmental Protection Agency, 25 Funston Road,
Kansas City, Kansas 66115
SLIMAK, Karen, JRB Associates, 8400 Westpark Drive, McLean, Virginia 22101
SLIVA, Thomas F., General Electric Silicone Products Department, Waterford,
New York 12188
SMALLEY, W.B., General Electric Company, P.O. Box 780, Wilmington, N.Carolina
28401
SMITH, Richard H., Brazos River Authority, P.O. Box 7555, Waco, Texas 76710
SNOW, Jay, Texas Department of Water Resources, Industrial Solid Waste Branch,
Austin, Texas
SPICER, James R., Tennessee Dept. of Public Health (Div. of Solid Waste Mngt),
Suite 320, Capitol Hill Bldg., 301 7th Ave. N., Nashville,Tennessee
37219
STAGG, William W., D. Ralph Caffery & Associates, Inc., 411 Wall Street,
Lafayette, Louisiana 70506
STANSBURY, John, Diamond Shamrock Corp., 1149 Ellsworth Drive, Pasadena, Texas
77051
STEINKRUGER, A.F., DuPont, 722 4th St., Marietta, Ohio 45950
STEPHENS, Robert D., California Dept. of Health, 2151 Berkeley Way, Berkeley,
California 94708
STENBURG, Robert L., U.S. Environmental Protection Agency, Solid and Hazardous
Waste Research Div., Municipal Environmental Research Lab., 26 West
St. Clair Street, Cincinnati, Ohio 45268
STINSON, E. Frank, Stuever & Assoc. Consulting Engineers, 5815 Melton Drive,
Oklahoma City, Oklahoma 73132
STRAIN, Ray E., Black & Veatch, Consulting Engineers, 9709 England, Overland
Park, Kansas 66212
STROM, James R., Gulf Science s Technology Co., 229 Cornwall Drive, Pittsburgh,
Pennsylvania 15238
STUEVER, Joseph H., Stuever & Associates Consulting Egnineers, 5815 Melton Drive,
Oklahoma City, Oklahoma 73132
STYRON, C.R., III, U.S. Army Engineer Waterways Experiment Station, 111 Stone-
wall Road, Vicksburg, Mississippi 39180
SUIDAN, Makram, Daniel Laboratory, Georgia Institute of Technology, Atlanta,
Georgia 30332
SYKES, Jon F., Dr., Dept. of Civil Engineering, University of Waterloo,
Waterloo, Ontario, Canada N2L 3G1
422
-------
TARJAN, Armen Charles, University of Florida AREC, Box 1088, Lake Alfred,
Florida 33850
TAYLOR, Graham C., Colorado School of Mines, 1008 S. Miller Street, Lakewood,
Colorado 80226
THOMPSON, Douglas W., U.S. Army Engineer Waterways Exp. Station, P.O. Box 631,
Vicksburg, Mississippi 39180
THOMPSON, Michael S., U.S. Army Environmental Hygiene Agency, Solid Waste
Management, Aberdeen Proving Ground, Maryland 21010
THOMPSON, Warren, University of Texas, 8.6 ECJ. University of Texas, Austin,
Texas 78721
TIPPLE, Greg, Texas Department of Water Resources, 3303 B. Bonnie Road, Austin,
Texas 78703
TOLSON, Dennis, Environment Ontario, 135 St. Clair Ave. West, Toronto, Ontario
M4V 1P5
TOOTHAKER, Anne M., General Electric Company, 1 River Road, Schenectady, New
York 12345
TRASK, Harry W., U.S. Environ. Prot. Agnecy, OSW/HWMP, 7428 Lanham Lane, Oxon
Hill, Maryland 20022
ULLERY, James, South Carolina Dept. of Health & Environmental Control, 2600 Bull
Street - Sims-Aycock Bldgs., Columbia, South Carolina 29201
ULMAN, Jan, University of Oklahoma, Carson Engr. Center, Norman, Oklahoma 73071
URBANEK, Joseph W., Defense Logistics Agency, U.S. Govt., Rm 4C499, Cameron
Station, Alexandria, Virginia 22314
VAN GENUCHTEN, M. Th., Dr., Dept. of Civil Engineering, Princeton University,
Princeton, New Jersey 08540
VILLAUME, James F., Pennsylvania Power & Light Company, Two North Ninth
Street, Allentown, Pennsylvania 18101
VOWELL, Don L., Panhandle Eastern Pipe Line Company, P.O. Box 1642, Houston,
Texas 77001
WALLACE, Wayne, Occidental Chemical Co., P.O. Box 1185, Houston, Texas 77001
WALRATH, JESS R., Jr., Xerox Corp., 180 Normandy Ave., Rochester, New York
14619
WEITZMAN, Leo, U.S. Environmental Protection Agency, Solid and Hazardous
Waste Research Div., Municipal Environmental Research Lab., 26 West
St. Clair Street, Cincinnati, Ohio 45268
423
-------
WELSH, Stephen K., 3M Company, Envir. Eng. & Pollution Control, P.O. Box 33331,
Bldg. 21-2W(58), St. Paul, Minnesota 55133
WHITFORD, Stuart, Seligmann & Pyle Consulting Engineers, Inc., 3918 Naco-Perrin,
San Antonio, Texas 78217
WICKS, Patrick H., Chem-Nuclear Systems, Inc., P.O. Box 1866 (10602 N.E.
38th Place), Bellevue, Washington 98009
WILBANKS, Doye W., Lone Star Steel, Box 21, Lone Star, Texas 75668
WILES, Carlton C., U.S. Environmental Protection Agency, Solid and Hazardous
Waste Research Div., Municipal Environmental Research Lab., 26 West
St. Clair Street, Cincinnati, Ohio 45268
WILKINSON, Ralph, Dr., Midwest Research Institute, Kansas City, Missouri 64110
WILLIAMS, James Hadley, Dept. of Natural Resources, Div. of Geology & Land
Survey, P.O. Box 250, Rolla, Missouri 65401
WOODYARD, John P., SCS Engineers, Long Beach, California
YOUNG, Edward E., Jr., U.S. General Accounting Office, 16204 Pond Meadow Lane,
Bowie, Maryland 20716
YOUNG, Richard, Kansas Geological Survey, 1930 Avenue A, Campus West, Lawrence,
Kansas 66044
424
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1 REPORT NO. 2.
FPA-600/9-78-016
4. TITLE AND SUBTITLE
. ««in nTCD/ICfll r\C UA7/lDr»AllC L/ACTrC
LAND DIbrUoAL Ur HA^HKUUuo WMblti
Proceedings of the Fourth Annual Research Symposium
7. AUTHOR(S)
David W. Shultz, Editor
g. PERFORMING ORGANIZATION NAME AND ADDRESS
Dept. of Resource and Environmental Engineering
Southwest Research Institute
San Antonio, Texas 78284
12. SPONSORING AGENCY NAME AND ADDR/ESS
Municipal Environmental Research Laboratory--Cin. , OH
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati. Ohio 45268
3. RECIPIENT'S ACCESSION NO.
5. REPORT DATE
August 1978 (Issuing Date)
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO
17-5013
70. PROGRAM ELEMENT NO.
1DC618
11. CONTRACT/GRANT NO.
805544
13. TYPE OF REPORT AND PERIOD COVERED
Symposium March 6-8, 1978
14. SPONSORING AGENCY CODE
EPA/600/14
15. SUPPLEMENTARY NOTES
Robert E. Landreth, Project Officer 513/684-7871
16. ABSTRACT
The fourth SHWRD research symposium on land disposal of hazardous waste was held
at the St. Anthony Hotel in San Antonio on March 6, 7, and 8, 1978. The purpose of th
symposium was two-fold: (1) to provide a forum for a state-of-the-art review and dis-
cussion of ongoing and recently-completed research projects dealing with the managemen
of hazardous wastes and (2) to bring together people concerned with hazardous waste
management who can benefit from an exchange of ideas and information. These proceed-
ings are a compilation of the papers presented by symposium speakers. They are
arranged in the order of presentation. The five primary technical areas covered are
(1) Methods development and economic assessment,
(2) Identification of pollutant potential,
(3) Predicting trace element migration,
(4) Modification of disposal sites and waste streams, and
(5) Alternatives for land disposal.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
COSATi Field/Group
Leaching, Collection, Hazardous Materials,
Disposal, Soils, Groundwater, Pollution,
Permeability, Waste Treatment, Linings
Solid waste management,
Hazardous waste,
Leachate, Toxic
13B
JlSTDISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (This
Unclassified
21. NO. OF PAGES
435
Form 2220.1 (R«». 4-77)
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
425
tm.6ST.oto/ii
------- |