vvEPA
United States
Environmental Protection
Agency
Municipal Environmental Research
Laboratory
Cincinnati OH 45268
Research and Development
Disposal of
Hazardous Waste
EPA-600/9-80-010
March 1980
Proceedings of the
Sixth Annual
Research Symposium
Do not remove. This document
should be retained in the EPA
Region 5 Library Collection.
-------
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-600/9-80-010
March 1980
DISPOSAL OF HAZARDOUS WASTE
Proceedings of the Sixth Annual Research Symposium
at Chicago, Illinois, March 17-20, 1980
Cosponsored by Southwest Research Institute and
the Solid and Hazardous Waste Research Division
U.S. Environmental Protection Agency
Edited by:
David Shultz
Coordinated by:
David Black
Southwest Research Institute
San Antonio, Texas 78284
Grant No. R807121
Project Officer
Robert E. Landreth
Solid and Hazardous Waste Research Division
Municipal Environmental Research Laboratory
Cincinnati, Ohio 45268
MUNICIPAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
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 constitute
endorsement or recommendation for use.
i1
-------
FOREWORD
The Environmental Protection Agency was created because of increasing
public and governmental concern about the dangers of pollution to the health
and welfare of the American people. Noxious air, foul water, and spoiled
land are tragic testimony to the deterioration of our natural environment.
The complexity of the environment and the interplay between its components
require a concentrated and integrated attack on the problem.
Research and development is the first necessary step in problem solution;
it involves defining the problem, measuring its impact, and searching for
solutions. The Municipal Environmental Research Laboratory develops new and
improved technology and systems to prevent, treat, and manage wastewater and
the solid and hazardous waste pollutant discharges from municipal and community
sources; to preserve and treat public drinking water supplies; and to minimize
the adverse economic, social, health and aesthetic effects of pollution. This
publication is one of the products of that research—a vital communications
link between the researcher and the user community.
These proceedings present the results of completed and ongoing research
projects covering the disposal of hazardous wastes.
Francis T. Mayo
Di rector
Municipal Environmental
Research Laboratory
-------
PREFACE
These proceedings are intended to disseminate up-to-date information on
extramural research projects dealing with the disposal of hazardous wastes.
These projects are funded by the Solid and Hazardous Waste Research Division
(SHWRD) of the U.S. Environmental 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 ten sessions includes papers dealing with major
areas of interest for those involved in hazardous waste disposal technology.
The papers are printed here basically as received from the authors. They
do not necessarily reflect the policies and opinions of the U.S. Environmental
Protection Agency or Southwest Research Institute. Hopefully, these proceed-
ings will prove useful and beneficial to the scientific community as a current
reference on the disposal of hazardous wastes.
IV
-------
ABSTRACT
The sixth Solid and Hazardous Waste Research Division Research Symposium
on treatment of hazardous waste was held at the Conrad Hilton Hotel in Chicago
March 17-20, 1980. The purpose of the symposium was two-fold: (1) to provide
a forum for a state-of-the-art review and discussion of ongoing and recently
completed research projects dealing with the management of hazardous wastes
and (2) to bring together people concerned with the disposal of hazardous
waste who can benefit from an exchange of ideas and information. Bound in two
volumes, Treatment and Disposal, the proceedings of the symposium are published
to provide a copy of all papers in the order presented. In this document, the
Disposal volume, the following seven technical areas are covered:
(1) Waste sampling and characterization
(2) Transport and fate of pollutants
(3) Pollutant control
(4) Co-disposal
(5) Landfill alternatives
(6) Remedial actions
(7) Thermal destruction techniques.
-------
TABLE OF CONTENTS
Page
OPENING SESSION
Current Research on Land Disposal of Hazardous Wastes
Norbert B. Schomaker, Bonnie L. Rittenhouse
U.S. Environmental Protection Agency
SESSION I: WASTE SAMPLING AND CHARACTERIZATION
Analysis of Hazardous Waste
Robert D. Stephens, Emil R. de Vera
Hazardous Materials Laboratory 15
Hazardous Waste Compatibility
Howard K. Hatayama, P. E., Robert Stephens, Ph.D.,
Emil R. de Vera, James J. Chan, David L. Storm, Ph.D 21
Monitoring Well Sampling and Preservation Techniques
James P. Gibb, Illinois State Water Survey
Rudolph M. Schuller, Robert A. Griffin
Illinois State Geological Survey ........ 31
The Utility of Extraction Procedures and Toxicity Testing With Solid Wastes
C.W. Francis, Environmental Sciences Division
M.P. Maskarinec, Analytical Chemistry Division
J.L. Epler, Biology Division
O.K. Brown, Environmental Sciences Division
Oak Ridge National Laboratory 39
Chemical Speciation of Flue Gas Desulfurization Wastes
James C.S. Lu, Calscience Research, Inc. 46
SESSION II: TRANSPORT AND FATE OF POLLUTANTS
Interpreting Results From Serial Batch Extraction Tests of Wastes and Soils
Martin J. Houle, Duane E. Long
Department of the Army, Dugway Proving Ground . . 60
Disposal and Removal of Halogenated Hydrocarbons in Soils
Robert A. Griffin, Sheng-Fu J. Chou
Illinois State Geological Survey 82
vii
-------
TABLE OF CONTENTS (Cont'd)
Page
Movement and Biological Degradation of Large Concentrations
of Selected Pesticides in Soils
J.M. Davidson, P.S.C. Rao, Ti-TSE Ou
Soil Science Department, University of Florida 93
Influence of Leachate Quality on Soil Attenuation of Metals
W.H. Fuller, A. Amoozegar-Fard, E.E. Niebla, M. Boyle
Department of Soils, Water and Engineering
The University of Arizona 108
SESSION III: POLLUTANT CONTROL
Predicting Percolation Through Waste Cover by Water Balance
Richard J. Lutton, Geotechnical Laboratory
U.S. Army Engineer Waterways Experiment Station 118
Effect of Organic Chemicals on Clay Liner Permeability
A Review of the Literature
K.W. Brown, David Anderson
Texas Agricultural Experiment Station
Texas A&M University 123
Assessment of Liner Installation Procedures
David W. Shultz, Michael P. Miklas, Jr.
Southwest Research Institute 135
Interaction of Selected Lining Materials with Various Hazardous Wastes - II
H.E. Haxo, Jr.
Matrecon, Inc 160
Assessment of Processes to Stabilize Arsenic-Laden Wastes
Jaret C. Johnson, Robert L. Lancione
JBF Scientific Corporation 181
Field Investigation of Contaminant Loss From Chemically Stabilized Sludges
Larry W. Jones, Philip G. Mai one, Tommy E. Myers
U.S. Army Engineer Waterways Experiment Station 187
SESSION VI: CO-DISPOSAL
Leachate From Municipal and Industrial Waste Landfill Simulaters
Riley N. Kinman, University of Cincinnati
James J. Walsh, SCS Engineers 203
v i i i
-------
TABLE OF CONTENTS (Cont'd)
Page
Chemically Stabilized Industrial Wastes in a Sanitary Landfill Environment
Tommy E. Myers, Norman R. Francingues, Douglas W. Thompson,
Philip G. Mai one
U.S. Army Engineer Waterways Experiment Station 223
Stabilization at Municipal Landfills Containing Industrial Wastes
Frederick G. Pohland, Joseph P. Gould
School of Civil Engineering
Georgia Institute of Technology 242
SESSION VII: LANDFALL ALTERNATIVES
Optimization of Land Cultivation Parameters
K.W. Brown, L.E. Deuel, Jr., J.C. Thomas
Texas Agricultural Experiment Station
Texas A&M University 254
Field Verification of Land Cultivation/Refuse Farming
Joan B. Berkowitz, Sara E. Bysshe, Bruce E. Goodwin,
Judith C. Harris, David B. Land, Gregory Leonardos,
Sandra Johnson, Arthur D. Little, Inc 260
SESSION VIII: REMEDIAL ACTIONS
Top-Sealing to Minimize Leachate Generation
Grover H. Emrich, William W. Beck, Jr.,
Andrews L. Tolman, SMC-MARTIN 274
SESSION IX: THERMAL DESTRUCTION TECHNIQUES
A Research Program in Waste Management Technology for Carbon Fibers
Richard A. Carnes, Laura A. Ringenbach
U.S. Environmental Protection Agency 284
-------
CURRENT RESEARCH ON LAND DISPOSAL OF HAZARDOUS WASTES
Norbert B. Schomaker
Bonnie L. Rittenhouse
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 proaram 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 Minimizina Pollutants from Un-
acceptable or Inoperative Sites.
5. Co-Disposal
6. Uncontrolled Sites/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: ReA&aAch SympoAiwn
on GaA and Lzachate. ^nom Land^jJtJtA: forma-
tion, Co££e.ction and T-tea/fmeni, March 25-26,
1975, Rutgers, State University of New
Jersey; and Schomaker, N.B., Current
Research on Land Disposal of Hazardous
Wastes: KeA-LduafA Monogemewxt by Land
ViApoAaJt.: Ptioce.edA.ngA o<{ the. HazaJidouA
WaAte. Raieotc/i SympoA
-------
Current Research on Land Disposal of Muni-
cipal Solid Wastes: Land Disposal: PA.O-
ceecttng.4 aft the, f^th Annual RuAiatick Sym-
posium March 26-28, 1979, University of
Central Florida.
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 due. Study ofi Ha.zaAdou& (Da&teA
Input Into UOAA 1 la.nd^WU> - 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. This effort
. is being published in a report entitled
/ Sampt&iA and Sampling ?n.oc.e.dunu £01 Haz-
{-M.dou& Wa&te. Sfie.am!> - EPA 600/2-80-018,
January 1980. This effort has been expanded
to include development of a list of reactive,
wastes and test methods for those wastes
thought to cause serious problems.
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 (2) relates to the
compilation and evaluation of current leach-
ing tests methods. In this study various
available leaching tests are described and
the methods utilized are evaluated in
relation to their adaptability to field
(1) Parenthesis numbers refer to the pro-
ject officers, listed immediately following
this paper, who can be contacted for addi-
tional information.
-------
conditions. Also, as a part of the sam-
pling technique activity, analysis of
wastes has been developed as a four level
scheme, which begins with field characteri-
zation followed by three levels of pro-
gressively more detailed and instrument
intensive laboratory analysis. The results
of this effort are discussed in a report ;
entitled Compilation and Evaluation o£ ,-
Laa.cM.nQ Tut Method*, EPA-600/2-78-095, }
May 1978.
A second effort (2) will determine
which sampling and preservation techniques
should be accepted as standards for ground-
water sampling. Other objectives of the
study are to determine if current sampling
methods produce samples representative of
water contained in the aquifer being moni-
tored and if groundwater samples collected
in the field can be treated on location or
if laboratory treatment is required. Six
landfill monitoring wells will be studied
using four different pumping techniques and
thirteen different sample preservation
procedures.
A third effort (2), is an in-house
activity to determine the capability of
existing analytical procedures to quanti-
tate priority pollutants in various waste
landfill leachates. Preliminary efforts
will obtain leachates from four landfill
sites, two municipal solid wastes and two
hazardous waste sites, to determine speci-
fic priority pollutants. Data will be
analyzed to resolve analytical methodology
applicable to leachate samples.
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.
Validation of a Standard Leaching
Test (SLT) was the initial effort funded
in part by SHWRD (2). Existing leaching
tests were evaluated for those elements
that were of special benefit to the
development of an SLT. This information
has been published in a report described
in the "Standard Analytical Techniques"
section above. Three candidate SLT's
were chosen for further testing. The
additional effort has been published ,
in two reports entitled Compcutiton o& /
Jkti&e. Watte. Le.acking Tetti,, Exe.c.utive.
SurmaJLtj, EPA-600/8-79-001, May 1979
and Companion ofi Thine, Watte. Le.ac.king }
Tettt, EPA-600/2-79-071, July 1979. /
Also a
W
^
May 1971, declTc~ated to the rapid/
dTs~semination of information relative
to solid waste "Extraction Procedure"
(EP) portion of Section 3001 of RCRA
has been published (2).
A second effort (2) has been identi-
fied as the "Extraction Procedure" (EP)
test will establish a data base on toxicity
of leachates from wastes and establish rec-
ommended toxicity of leachates from wastes
and establish recommended toxicity test
protocols in order to establish criteria
defining acute and/or chronic hazardous
wastes. Work has been initiated on four
waste categories: municipal refuse,
Chicago sewage sludge, arsenic sludge and
power industry residuals.
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
(I.A.E.A.) for leach testing immobilized
radioactive waste solids, this effort has
been published in a report entitled -5-
Elutniate. TeAt Evaluation o£ Chemically y
Stabilize.d Watte. MateMal& - EPA-600/2-79- >
154, August 1979. Translucent plexiglass
columns were utilized and observations of
flow patterns as well as possible biologi-
cal activity were made. Five industrial
sludges and five Flue Gas Desulfurization
(FGD) sludges were investigated. This
effort is basically completed and the
final report is being prepared. An
interim report entitled Pollutant Pote.ntial ,
o£ Row and Cke.mical£y FXxed HazatdouA >
InduAttiial WaAteA and flue. Gal, Vetulfiusu.- j)
zation Sludge*, EPA-600/2-75-182, July
1975, 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 leach-
ing conditions 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 lea-
chate. A major consideration in the
leaching behavior 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 com-
position 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 and to compensate for extract absorbed
by the soil. The laboratory leaching pro-
cedure 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 research
activity is to develop procedures for using
soil as a predictable attenuation medium
for pollutants. Not all pollutants are
attenuated by soil, and in some cases, the
process is one of delay so that the pol-
lutant is diluted in other parts of the
environment. Consequently, a significant
number of the research projects funded by
SHWRD are focused on understanding the pro-
cess and predicting the extent of migration
of contaminants (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, i
M.H., Attenuation o^ Lea.ch.ate. ?oflutant&
by Soili,, presented at the Management o£
GOA and L&achate. in Land&ill&: Proceeding.*
0|{ the. Jhind. Annual Municipal Solid Wa&te.
ReAeasich Sympo&ium, March 14-16, 1977,
University of Missouri.
Bibliography and State of the Art
The initial effort (4) in this area ^
resulted in a completed bibliography en-
titled, A EibliogJiaphy, Volume. I: Selected
Metal* EPA-600/9-79-024a, August 1979,
and Volume. 2: Pesticide* EPA-600/9-79-
024b, August 1979.
A second effort (4) consisted of
a review 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$ Sele.cte.d Metal*, Aafaea-toA, and Cyanide.
in Soil*: Application* to Wa&te. Vi&po&al
PlobtemA, EPA-600/2-77-022, April 1977.
The document presents a critical review of
the literature pertinent to biological,
chemical, and physical reactions, 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 (4) examined 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^
o& land^M Le.ach.ate. Pollutant ktte.nuati.on
by SoW> EPA-600/2-78-158, August 1978.
The second effort (4) in this area
studied 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 Atte.ntuation o& Pollutants '.
Hi-nesiall,, EPA-600/2-78-157 August 1978.
The third effort (5) relates to model-
ing movement in soils 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
evaluated efficiency of vertical chimney
vents, barriers and forced convection
systems.
A fourth effort (4) relates to the
use of large-scale, hydrologic simulation
modeling as one method of predicting con-
taminant movement at disposal sites. The
two-dimensional model that was used suc-
cessfully to study a chromium contamination
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 presently needs a sub-
stantial amount of input data, it appears
promising for determining contaminant trans-
port properties of field soils and, even-
tually, predicting contaminant movement
using a limited amount of data.
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 procedure was developed that
allowed improved quantitative measurement
of PCB's in aqueous solutions. The initial
results of this study is currently being
published in a report entitled Mte.nuation '
ok WaAteA Soluble. PCB'4 by Easith MateMaU. !-;
A sixth effort (4) relates to an
evaluation 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 labor-
atory 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 aijr-fi_Vled_popsjJty and was
therefore greatly" reduced by increased
soil compaction and increased soil water
content.
A seventh effort (4) relates to the
adsorption, movement and biological degra-
dation 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 aque-
ous solubility limit of each pesticide.
The mobility of each pesticide increased
as its concentration in the soil solution
-------
phase increased. Pesticide degradation
rates and soil microbial populations gen-
erally declined as the pesticide concentra-
tion in the soil increased. The results of
this effort will be published in a report
entitled kd&onption, Movement and Elolo-
g-ical. Vigfiadcition o£ Lotge Canc.e.ntneuticmA
oj( Selected Peiitcxxfei Jin SoUA.
An eighth effort (4) was recently
initiated to determine how accurately the
EPA Gas Movement Model predicts the maxi-
mum distance that methane gas will move
through soils adjacent to landfills and how
accurately this model will predict the
relative effectiveness of control systems
(e.g., trenches, wells, barriers) for mini-
mizing methane gas movement.
A ninth (4) effort relates to the in-
fluence of chemicals on the permeability of
clay soils. The objective is to develop
a guidance manual identifying wastes that
increase the permeability of soils at a
disposal site. The study will also develop
a test procedure for predicting whether a
specific waste and soil will react to in-
crease the permeability.
A tenth (4) effort relates to the
determination of the attenuation mechanisms
and capacity of selected clay minerals and
soils for hexachlorocyclopropentadial
(HCCPD) and "hex" wastes. Also effects of
caustic-soda brine on the attenuation and
solubility of HCCPD and the development of
a chemical model to predict HCCPD migration
through soil will be pursued. Except for
the measurement of the salting coefficient
of HCCPD, the approach will be the same as
described in the fifth effort listed above.
Field Verification
Limited field verification is being
conducted. The initial effort (3) 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 beneath
the landfill site. The sites represent
varying geologic conditions, recharge rates,
and age, ranging from a site closed for 15
years to a site currently operating. In-
dividual site characteristics were identi-
fied, and sample analyses necessary to
determine the primary pollutant levels in
the waste soils and groundwater were deter-
mined. The results of this effort are
discussed in a report entitled
and ?hy&
-------
being performed in the controlled lab stu-
dies previously discussed in the section on
pollutant transport.
Li ners/Membranes/Admi xtures
The liner/membrane/admixture tech-
nology (3) 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 investigated
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 (12.0 in)
- Hydraulic asphalt concrete (2.5 in)
- Modified bentonite and sand (5.0 in)
- Soil cement with and without surface
seal (4.0 in)
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 h.ave
been exposed for more than 3 years to the
following six classes of hazardous wastes
which utilized ten specific types of wastes:
strong acid; strong base; waste of satu-
rated 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 wastes 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 Mat&u.a£& Exposed
to Haza/tdouA and JOXA.C. SfudgeA TiM>t
Re.povt, EPA-600/2-77-081, June 1977.
A second effort (3) relates to a
state-of-the-art of landfill impoundment
techniques. The 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 obtained
from various manufacturers, suppliers and
installers and contains analyses of liner
compatibility with industrial wastes.
The results of this effort have been
published in a report entitled, State.-o^- ,
Study ^01 land^-UUL Impoundment -
EPA-600/2-78-196, Dec. 1978. >
A third effort (3) 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 SO 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. Consequently, methods of
lining such disposal sites must have a low
unit cost. It is desirable that materials
be easy to apply or install. Because of
these considerations, the number of poly-
meric 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 prefabri-
cated 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 perform 12
and 24 month exposure tests. The final re-
port is entitled F£ue GOA Cleaning Sludge
' Le.ac.kate./ Lin&i Compatibility Investigated:
' Interim R&poit, EPA-600/2-79-136, Aug. 1979.
A fourth effort (3) 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
Line.U fan. Sanitany Landfii-llA and Che.mic.al
and HazandouA Watte. VJj>pot>al
9-78-005, May 1978.
EPA-600/
A fifth effort (3) relates to a study
to assess actual field procedures utilized
in (1) preparing the supporting subbase
soil structure for liners and (2) placing
the various liner materials common to pro-
jects requiring positive control of fluid
loss. In addition, problems and solutions
will be identified which can be avoided by
proper preplacement and placement procedures.
To perform this study, a study team pro-
vided with a checklist, obtained from manu-
facturer's recommendations for placement,
are visiting a variety of liner installation
projects to observe first-hand the placement
construction and procedures actually used
to place the linings. Differences between
the recommended practices and the actual
procedures used, if any, will be questioned
and reviewed with the placement contractor.
Chemical Stabilization
Chemical stabilization is achieved
by incorporating the solid and liquid
phases of the waste into a relatively
inert matrix which is responsible for in-
creased physical strength and which pro-
tects the components of the waste from dis-
solution by rainfall or by soil water. If
this slows the rate of release of pollutants
from the waste sufficiently so that no seri-
ous 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
(3) 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 withaat least one of seven
separate fixation processes and subjected
to leaching and physical testing.
The lab and field studies have been
completed. The laboratory leaching test
data for the first six months includes:
methods for physical and chemical analyses,
documentation of various sludge fixation
processes, and physical and chemical data
on the sludges. The results have been
compiled and discussed in a report entitled
Pollutant Potential o& Row and Chemically
F-txed HazatdouA InduAttial (Ua&teA and flue.
Ga& deAulfiustization Sludged - Interim
Kepont, EPA-600/2-76-182, July 1976.
The laboratory physical program test
included physical properties (grain size
distribution, atterbug limits, specific
gravity, volume-weiqht-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,
Pky-b4.cal and Engineering PnopeitieA o{> the.
HazandouA JnduAtnial WaAteA and SludgeA,
EPA-600/2-77-139, August 1977.
The second effort (3) relates to a
survey and identification of solidification/
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
results have been published in a report
entitled Susivey o^ Solidification/'Stabi-
lization Technology fan HazandouA
-------
EPA-600/2-79-056, July 1979.
The third effort (3) relates to a
series of field verification studies to
verify success and which pollutants have
been immobilized at landfills receiving
stabilized hazardous wastes. Four sites
where stabilized industrial wastes and three
sites where stabilized flue gas cleaning
sludges 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 process. Two of the industrial
waste sites contained auto assembly (metal
finishing wastes, one site contained electro-
plating wastes and the fourth site contained
refinery sludges. The physical properties
of soils under the disposal sites were af-
fected little, if at all, by the disposal
operation. The results are reported in the
publication entitled The- E^e.ct& o& flue.
Ga& Cleaning Ua&te. on GnoundaateJi Quality
and Soil ChaJiacteAi&ticA , EPA-600/2-79-164,
August 1979.
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 (6) 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.ve£cpmejit o£ a
Polymeric. Cementing and Encapsulating
Ptocew {,01 Managing HazandouA ltia&teJ>,
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 be-
ing evaluated is thf
encapsulation of SITU
Moisture Infiltrati
The initial ef
identification of t
merits of cover soil
various soil types
ments. Of primary
mization of moistui
landfill cover soils. Other criteria as-
sociated with landfill cover soils, i.e.,
infiltration, gas venting, vegetation, soil
erosion, rodent burrowing trafficabi1ity
will also be identified. The characteristics
of the various soil types are reviewed in-
dividually and in combination with other
soil types to determine the most suitable
type of cover material for use in meeting
the desired functional requirements for a
given disposal site. The initial results
from this study have been published in a
report entitled CoveJi &on. Solid Wa&te.
Cont>i.deJiation& £01 Design and
, EPA-600/2-79-136, Aug. 1979.
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 (4) 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 (5) was evaluated in several
laboratory studies using the following
physical-chemical processes: chemical
precipitation, activated carbon adsorption,
and chemical oxidation. This study dis-
cussed in two reports entitled Evaluation
o£ Lejachate. TJieatmesit: Volume. T. - Ckanac.-
tesuJ>ticA o({ Leac/iote, EPA-600/2-77-186a,
September 1977 and Evaluation o£ Lzachate.
-------
Volume. I - ChaxacteJii&ticA o£
JL, EPA-600/2-77-186a, September 1977
-valuation oft Le.ac.hate. Tle.atme.nt: Vol-
. II - B^olog-ic-al and PhyA-ic.al-Che.mic.al
/tocewea, EPA-600/2-77-186b, Nov. 1977.
A second effort (4) relates to a labor-
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. How-
ever, the increased water contamination
from solubilization of iron seems to rule
out use of iron oxides until further work
is conducted.
A third effort (4) relates to a lab-
oratory evaluation of ten natural and syn-
thetic materials (bottom ash, flyash, ver-
miculite, illite, Ottawa Sand, activated
aluminia, cullite) for the removal of
contaminants in the leachate and liquid
portion of three different industrial
sludges (calcium fluoride sludge, petroleum
sludge, metal finishing sludge). Results
of the laboratory experiments indicate sor-
bent capacity is a function 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. The information discussed in a report
entitled Sox.be.nti> {,01 Ftu.oJu.de., ttttal F.OI-
i^hing and PetH.ole.um Sludge. Le.ac.hate. Con-
taminant Control, EPA-600/2-78-024, March
1978.
Biological Treatment
Various unit processes for biological
treatment have been investigated in the
laboratory for only municipal refuse
leachates. This effort (5) has investi-
gated the process kinetics, the nature of
the organic fraction of municipal refuse
leachate, and the degree of treatment that
may be obtainable using conventional waste-
water treatment methods. The biological
methods evaluated were the aerobic 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.
CO-DISPOSAL
The overall objective of the co-disposal
activity is to assess the impact of the
disposal of industrial waste materials with
municipal solid waste. Concern has been
voiced that the addition of industrial
waste may result in the occurrence of var-
ious toxic elements in leachates and there-
by pose a threat to potable groundwater
supplies. Because the environmental effects
from landfilling result from not only the
soluble and slowly soluble materials placed
in the landfill but also the products of
chemical and microbiological transformations,
these transformations should be a consid-
eration in management of a landfill to the
extent that they can be predicted or in-
fluenced by disposal operations.
The initial effort (5) involves a
study of the factors influencing (1) the
rate of decomposition of solid waste in a
sanitary landfill, (2) the quantity and
quality of gas and leachate produced during
decomposition, and (3) the effect of admix-
ing industrial sludges and sewage sludge
with municipal refuse. A combination of
municipal solid waste and various solid
and semi-solid industrial wastes was added
to several field lysimeters. All material
flows were measured and characterized for
the continuing study and related to leachate
quality and quantity, gas production, and
microbial activity.
The industrial wastes investigated
were: petroleum sludge, battery production
waste, electroplating waste, inorganic pig-
ment 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. The results of this initial effort
have been reported in a paper entitled
The. Ef$|Jec£& o£ InduAfvial Sludger on
Land&M Le.ac.hate. and Ga&, Pioc.e.e.dingA -
National ConfieAe.nce. on Vtbpo&al o&
R&&-idueA on Land, September 1976, pp. 69-
76. The updated results of this effort
is being published in a report entitled
Ptlot Sc.ale. Evaluation ofi SanitaAy Land-
10
-------
Muru.cu.pal and Indii&tsiLal
A second effort (4) to assess the
potential effects of co-disposal involves
the leaching of industrial wastes with
municipal landfill leachate as well as
water. Leachate from a municipal solid
waste (MSW) landfill was used to extract
the five industrial wastes and to study
movement of their components in the soil
columns. MSW leachate dissolved much greater
amounts of substances from the wastes and
apparently increased the mobility of these
substances in the soil columns relative to
the dissolution and mobility observed when
deionized water was used as a leaching
solution. The municipal landfill leachate
is a highly odorous 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. A
sequential batch leaching and soil adsorp-
tion procedure has been developed that pro-
vides information comparable to that from
soil column studies but in a much shorter
time.
A third effort (4) relates to chemi-
cally treated and untreated industrial
wastes being disposed of in a simulated
municipal refuse landfill environment.
Large lysi meters, six foot in diameter by
twelve feet high, are being utilized to
determine the difference in leachate qua-
lity 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.
UNCONTROLLED SITES/REMEDIAL ACTION
An ongoing study by OSW has identified
incidents of well contamination due to
waste disposal sites. Seventy-five to 85
percent of all sites investigated are con-
taminating ground or surface waters. In
order to determine the best practical tech-
nology and economical corrective measures
to remedy these landfill leachate and gas
pollution problems, a research effort has
been initiated (2) to provide local munici-
palities and users with the data necessary
to make sound judgments on the selection
of viable, in-situ, remedial procedures
and to give them an indication of the cost
that would be associated with such a pro-
ject. This research effort is being pur-
sued at a municipal refuse landfill site
at Windham, Connecticut. The site was
selected from a candidate site list of 17
which took into account appropriate site
selection criteria. This effort consists
of three phases. Phase I will be an engi-
neering feasibility study that will deter-
mine on a site specific basis the best
practicable technology to be applied from
existing neutralization or confinement
techniques. Phase II will determine 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. The
engineering feasibility study has been com-
pleted and a report entitled Guidance Man-
ual fan \\LYiimizLnQ VoULuAion fifiom Wa&te.
ViApotal Situ, EPA-600/2-78-142, August
1978 has been published. This guidance
document emphasizes remedial schemes or
techniques for pollution containment. The
remedial schemes discussed in this document
are:
- Surface Water Control
o surface sealing
o revegetation
- Groundwater Control
o bentonite slurry-trench cutoff
wall
o grout curtain
o sheet piling cutoff wall
o bottom sealing
- Plume Management
o extraction
o injection
o leachate handling
- Chemical Immobilization
o chemical fixation
o chemical injection
- Excavation and Reburial
The scheme currently installed and
monitored for the Connecticut MSW site is
a surface capping technique which could
be followed by a leachate extraction scheme
if required.
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,
11
-------
the alternatives currently being considered
are: (1) deep well injection, (2) under-
ground mines, and (3) land spreading. The
overall objective of the landfill alterna-
tives study is to determine the feasibility
and beneficial 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 Review and
Aj>&eA&me.nt o& Die.p-d'ell Injection o{, Haz-
afidouA Wa&te., Volume* I-1V, 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
hydro!ogic characteristics of the injection
and confining intervals are considered
along with the physical, chemical and bio-
logical compatibility of the receiving
zone with the wastewater to be injected.
Design and construction aspects of injection
wells are presented along with recommended
preinjection testing, operating procedures,
and emergency precautions. This effort is
discussed in a report entitled An Iwtsio-
ductcon to the. Te.chnof.OQii of, SubAuA.^ac.e.
Wa&teiaateA Inje.ct4.on, 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 environmentally acceptable manner if
they are properly treated and containerized.
This effort is discussed in a report entitled
Evaluation of, HazandouA Wa&te. Emplacement
AJI MZxed Opening^, EPA-600/2-75-040,
December 1975.
A second effort (4) related to an
economic 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 underground
storage facilities. This effort is dis-
cussed in a report entitled CoAt A&&eA&ment
jjoi the. Emplacement o£ Wazatdoui Material*
-in a Sa.it 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-
ti on for many years by pharmaceuti cal,
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 adsoprtion,
dilution, biodegradaticn, 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 oj$
lndu&t>u.al Ua&teA and Municipal Solid Wa&teA:
S tote-oft-the.-Ant Study, Volume I - Tech-
nical Summony and LiteAatune Susive.y, EPA-
600/2-78-140a, August 1978 and Statz-o^-
tha-Afut Study, Volume. II - f-ie.ld Investi-
gation and Ca&e. Studies, EPA-600/2-78-14Cb
August 1978.
The second effort (4) is a combination
laboratory, greenhouse and field study to
determine the fate and mobility of wastes
in soil for the purpose of developing cri-
teria for use in the design, management
and monitoring of land cultivation disposal
operations. Decomposition rate, applica-
tion rate, plant survival and growth, pol-
lutant 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
12
-------
streams versus operational parameters.
This matrix of information will be used to
develop design and guideline criteria.
The fourth effort (8) relates to the
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. The
initial results of the study are described
in a report entitled State.-o£-the.-Mt
Re.polt: PteticAde. Vl&poAaJt ReAzaAch, EPA-
600/2-78-183, August 1978.
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 in-
novation, 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 (8)
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 techniques.
The costs attributable to any technique
comprise the control costs, and the environ-
mental costs and benefits together deter-
mine the net benefits associated with the
threat scenarios. This effort has been
received in draft report form for processing
into a final report entitled Economic.
knaly&iA ofa HazakdouA Wadtz Management
klteAnativeA: Methodology and Vemon&tia-
tion.
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 im-
pact on the promulgation of sludge disposal
regulations. The Phase I portion of this
effort has been published in a report
entitled Data Ea&e. fan. Standard*/Re.gula£ion&
Ve.velopme.nt fat Land ViApo&at o& flue. Ga&
Cleaning Sludge*, EPA-600/7-77-118,
December 1977.
The Phase II economic draft report (8)
has been received and it is being processed
into a final report publication to be
entitled Economic. Impact Analytic o£ Alter-
native, flue. Gat> VeAul6usu.zation (FGP) Sludge.
Vl&po&al Re.gu£ation& on the. Utility Indu&tky.
CONCLUSION
The laboratory and field research pro-
ject 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 are asso-
ciated 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. Robert E. Landreth (SHWRD)
513/684-7871
4. Dr. Mike H. Roulier (SHWRD)
513/684-7871
13
-------
5. Mr. Dirk R. Brunner (SHWRD)
513/684-7871
6. Mr. Carl ton C. Wiles (SHWRD)
513/684-7881
7. Mr. Charles J. Rogers (SHWRD)
513/684-7881
8. Mr. Oscar W. Albrecht (SHWRD)
513/684-7881
14
-------
ANALYSIS OF HAZARDOUS WASTES
Robert D. Stephens and Emil R. de Vera
Hazardous Materials Laboratory
California Department of Health Services
Berkeley, California
ABSTRACT
A paper on the characterization of hazardous waste would have never been included in
such a symposium as this five years ago. However, since that time the impact of toxic and
otherwise hazardous industrial waste products had become an important public and govern-
mental concern.
The first major step toward solution
of this problem was taken in late 1976 with
the passage of Public Law 9^-580, the
Resource Conservation and Recovery Act
(RCRA). This far reaching law, mandates
the U.S. Environmental Protection Agency
(EPA) to develop regulations for minimum
standards for those who generate, transport
and dispose of hazardous waste. In the
words of Douglas Costle, EPA Administrator,
"We do not underestimate the complexity and
difficulty of our proposed regulations.
Rather, they reflect the large amounts of
hazardous wastes generated in our diverse
society. The regulations will affect a
large number of industry... as well as la-
boratories, pesticide applicators, and
waste transporters. The agency estimates
that approximately 270,000 waste generating
facilities, 10,000 waste transporters, and
many thousands of waste disposal and pro-
cessing facilities will be regulated."
In the two years since the passage of
RCRA, EPA has been working to develop these
regulations, and on December 18, 1978, the
core of the regulations was proposed in the
Federal Register.
The phrase "core of the regulations"
is used because the December 18th issue
of the Federal Register covers what is
called "Subtitle C" of RCRA. Subtitle C
is directed toward the identification and
characterization of hazardous wastes. The
regulations in this section of the law
outline biological and chemical testing
methodologies by which these complex indus-
trial and other toxic waste products may be
examined to evaluate their potential hazard.
This highlights a key point in the
regulation and management of hazardous
waste. Such activities must be founded on
an adequate understanding of waste compo-
sition. The generation of waste composi-
tion data by the laboratory is in many
cases a formidable problem. However, if
proper procedures are to be developed to
assure safe handling, transportation, re-
cycle, or disposal of waste, as well as
assessment of public health and environ-
ment impacts, waste chemical composition
must be known.
Our laboratory is attempting to devel-
op a variety of methodologies to attach
this problem. These include many of the
modern instrumental techniques such as gas
chromatography (GC) , high pressure liquid
chromatography (HPLC) , atomic adsorpition
spectrophotometry (AAS), X-ray fluores-
cence (XRF), and gas chromatography/mass
spectrometry (GC/MS). In addition to these
techniques, we have found that thin layer
chromatography (TLC) is widely applicable
to waste analysis, and has certain definite
advantages over some of the more "equipment
intensive" methods. Many laboratories
which must provide data on waste composi-
tion do not have the financial resources to
obtain expensive sophisticated instrumenta-
tion, TLC in these laboratories Is an
ideal alternative.
15
-------
The approach taken by our laboratory
has been one of a phased, or mult Neve)
characterization. These phases are:
TABLE 1. (continued)
Method
Applications & Limitations
Phase I
Field Tests
Temperature test
Phase II Laboratory Hazard Assessment
Tests
Phase IN Laboratory Sample Preparation
and Screening Tests
Phase IV Laboratory Confirmatory Tests
An analytical scheme based on these
four phases is currently under development.
To date, the greatest emphasis has been
placed on phases Ml and IV, and they will
be the primary subject of this paper.
TEST PHASE I (FIELD TESTS)
This phase is intended for field as-
sessment of the physical and chemical pro-
perties of the wastes. This includes the
measurements of the properties of wastes
that rapidly changes with time such as pH,
and other properties that can cause injury
to field personnel or immediate damage to
public health and the environment such as
generation of toxic gases, radioactivity,
flammabi1ity, or explosivity. Other pro-
perties such as odor and color that may
contribute to the preliminary characteri-
zation of the wastes are also noted. Table
1 lists the suggested field tests to be
performed.
Detector tubes
Radioactivity
Bloassay test
TABLE 1. SUGGESTED FIELD TESTS
Elevated temperature of a sample considerably above
ambient indicates possibly exothermic reaction is
taking place. Special precautions must be observed
In handling the waste to avoid accidents. Tempera-
ture measurements are conveniently taken with a
thermometer.
A low pH is indicative of the presence of strong
acids; high pH that of strong alkali. This informa-
tion indicates that the waste is hazardous and must
be treated as such. The Information also alerts the
analyst to test for strong acid and bases. pH Is
measured In the field with a pH paper or a portable
pH meter.
A number of detector tubes for gases are commercial-
ly available. These tubes provide quick tests for
the presence or absence of most common toxic gases.
One single tube can detect the presence of more than
20 gases. To Identify a gas, a certain volume of
the gas is aspirated through a detector tube specific
for the gas. The concentration of the gas is propor-
tional to the length of the color change in the tube.
The waste is hazardous when a toxic gas is detected
to be generated from it. Accuracy of most detector
tubes range from 20 to 50%.
Radiation from waste is measured with a radiation
counter capable of detecting alpha, beta and gamma
radiations. Portable Gelger counters are most useful.
When a waste Is found to contain radioactive material,
the proper regulatory agency should be notified for
immediate action.
TABLE 2. TEST PHASE I I METHODS
ApplIcatIons
Applications & Limitations
Color test Certain compounds and metallic ions have charactels-
tic colors. These colors may serve as Indicators of
the predominant components In the waste. This method
is very subjective.
Odor test Odors of certain gases and other volatile substances
are very distinctive. This information can assist
In determining the group or classes of compounds in
the wastes. This method Is subjective. No instrument
has yet been invented to measure odor.
(continued)
Total acidity
Total alkalinity
Explosivity test
Waste extraction test
Whole and/or separated phases of 1iquid waste
mixtures are used for this test. The 96-hour
LCjn of the waste must be determined on fathead
minnows (pimephales promelas). If a waste dilu-
tion of 100 mg or less per liter of water kills
on-half or more of the population of the test
fish In 96 hours, the waste is hazardous.
Whole and/or separated organtc phase of 1iquId
waste mixture Is determined by ASTH D56-70,
Standard of Test for Flash Point by Tag Closed
Tests. Any liquid waste with a Flash Point of
100°F or less is flammable, thus hazardous.
This is a measure of the ability of the waste to
neutralize a strong base. High acidity indicates
the presence of strong mineral acids and/or
hydrolyzable salts of mineral acids.
This is a measure of the ability of the waste to
neutralize a strong acid. High alkalinity indi-
cates the presence of mostly hydroxyt ions, and
partly carbonate and bicarbonate ions.
This test is used to determine whether the waste
is sensitive to shock or detonates on Impact.
An explosivity meter or impact apparatus for
explosivity can be used. Explosive wastes are
hazardous.
This test is used to determine the teachability
of the components of a solid waste. If the total
soluble concentration of any component exceeds
the soluble threshold limit concentration for
that substance, the waste is hazardous.
16
-------
TEST PHASE II (LABORATORY HAZARD ASSESSMENT
TESTS)
The objective of this test phase is
to confirm some of the field data and to
obtain more information about the hazard-
ous properties of the waste sample under
controlled laboratory conditions. Tests
included in this phase are bioassay, flash
point, explosivity, total acidity, total
alkalinity, waste extraction test and
toxic gas generation. Small portions of
the whole or phase separated sample is
used for the test. Results from these
tests, in some cases, is adequate to
characterize a waste as nonhazardous,
hazardous or extremely hazardous. Further
analysis of the waste sample is continued
when isolation and quantitation of the haz-
ardous components are required. Table 2
outlines the suggested Test Phase II
methods.
TABLE k. TESTS FOB AQUEOUS PHASE
Procedure
Residue at 550 C
Acidity
Alkalinity
Beilstein test
Spot tests
Redox potential
An ion analysts
TLC
Volatiles up to 95°C
Extraction with solvent
Purpose
To determine presence of metals and
metallic compounds
To determine presence of strong acids,
and corrosivity
To determine presence of strong alkali,
and corrosivity
To determine the presence of dissolved
organochlorides
To determine functional groups of water
soluble organic compounds, and simple
organic and inorganic compounds
To determine presence of strong oxidizing
or reducing agents
To determine presence of anions such as
S0i,= , Ho~3» cl~- CN~' ctc-
To separate dissolved organic and certain
inorganic compounds
To collect volatiles with boiling points up
to 95°C
To separate solvent soluble compounds from
others
TEST PHASE III (LABORATORY SAMPLE PREPARA-
TION AND SCREENING TESTS)
The initial step in this phase is the
high speed centrifugal ion of the liquid
waste mixture at 7,000 to 17,000 RPM to
separate the aqueous, organic and solid
phases. The separated phases are then sub-
jected to qualitative compositional ana-
lysis using various procedures to obtain
such information as the identify of organic
functional groups and chemical classes, and
that of certain suspected and more commonly
TABLE 3. TEST Oil OOSAHIC PHASE
Purpose
Beilstein test
Spot tests
Flash point
Bioassay
UV
III
TLC
Column chromatography
Distillation up to 95°C
Residue at 550°C
To detect presence of organic chlorides
To detect organic functional groups and
simple compounds
To determine flammability
To determine toxicity to organism
To detect presence of aromatic and
unsaturated compound^
To determine molecular structures
To separate and preliminarily identify
components
To separate components according to
molecular weight
To collect low boiling distillates for
Identification
To determine presence of organometals
and metals
TABLE 5. TESTS F0» SOLIDS AHD SOLID PHASE
Procedure
TLC
Volatiles up to 95°C
Extraction with solvent
Residue at 550°C
Acidity
Alkalinity
Beilstein test
Spot tests
Redox potential
See
See
See
See
See
See
See
See
Purpose
Table
-------
occurring simple organic and inorganic com-
pounds. Separation procedures such as thin
layer and column chromatography or distil-
lation are also used to isolate groups or
individual components. TLC, particularly,
is a simple, fast and inexpensive separa-
tion procedure that can be used when very
little information about the waste is
available. The isolated components are
further examined with the compositional
test procedures (i.e., U.V., I.R., Bellstein
before Test Phase IV procedures are
applied. In some cases, the separated
phases of the sample are subjected back to
the Test Phase II procedures for assessing
whether the separated phases are more
hazardous than the whole sample. Other
times the separated phases are directly
analyzed by the Test Phase IV procedures
without performing any additional Test
Phase II procedures. This usually happens
when information is desired only on one or
two suspected components of the waste
e.
A solid waste sample in this test
phase is extracted with a solvent or sol-
vent combinations by dissolution or by
using a Soxhlet extractor. Aqueous and
organic solvent solutions of the solid
sample may also be prepared and subjected
to the same test procedures as used for the
separated aqueous and organic phases. The
solid sample can also be subjected back to
Test Phase 11 procedures for bioassay and
waste extraction test and.or directly ana-
lyzed for metallic elements by AA, XRF or
inductively-coupled plasma (ICP) atom ana-
lyzer in Test Phase IV.
After conducting Test Phase III proce-
dures, usually enough information is known
about the waste components. In some cases,
some components are presumptively or close
to being identified. When further charac-
terization of the components are desired,
the Test Phase III information serves as a
guide in selecting the proper instrumental
methods and optimal operating parameters
that can be used in Test Phase IV. For
instance, when low boiling chlorinated
hydrocarbons are identified in a sample,
one most certainly would select to use a
GC with an electron capture detector.
Test Phase III procedures prescribe the
Test Phase IV procedures to be used.
Tables 3, k, and 5 show the different
suggested procedures for Test Phase III.
TEST PHASE IV (LABORATORY CONFIRMATORY
TESTS)
This phase of the analytical scheme
consists of the state-of-the art instrumen-
tal analytical methods. These methods in-
clude atomic absorption (AA). spectrophoto-
metry, gas chromatography (fiC;, gas chroma-
tograph/mass spectrometry (GC/MS), high
pressure liquid chromatography (HftSj,'
X-ray fluorescense spectrometry (XRF) and
inductively-coupled plasma (ICP) atom anal-
ysis. These methods are expensive and
complex requiring skilled and trained ana-
lysts to operate and maintain them as well
as to interpret the data generated. Des-
pite their high costs and sophistication,
these methods are necessary in the analysis
of complex mixtures such as hazardous
wastes. The specialized methods can be
used, in some instances, to analyze waste
samples directly or with minimal sample
pretreatment or preparation. In most cases,
however, due to multiple inherent inter-
ferences, the waste samples are first sub-
jected to clean-up, isolation, fractiona-
tion or partition procedures such as in
Test Phase III. Components of wastes which
have been presumptively identified in Test
Phase III are definitively identified and
quantJtated with these methods.
ANALYTICAL APPROACHES
Different analytical pathway will be
included in the analytical scheme to cover
most types of wastes. A sample submitted
to the laboratory for analysis is first
visually examined. Its physical state,
color and other characteristics are noted.
These data in addition to field test
results, if any, and the background infor-
mation usually furnished by the sample
collector, provide the basis for the
analytical strategy. The analytical scheme
will be consulted and the best possible
analytical approach is selected. For
example, the scheme provides the following
analytical pathways for liquid/solid waste
mixtures:
(1) Mixture -
(2) Mixture -
->-AA or XRF or ICP
-v Centrifuge at ?-l?K RPM
- organic, aqueous and
sol id phases
(3) Organic phase + f lash point
(k) Organic phase -*GC and/or HPLC; and/or
GC/MS
-------
r%
(5) Organic ^fase^TLC -*• GC and/or HPLC;
and/or GC/MS
(6) Organic phase -"distil 1, 95°C + CO and/
or (5); and/or (7); and/
or (8)
(7) Organic phase->-TLC and/or Beilstein
test; and/or spot test;
and/or UV; and/or 1R;
and/or column chromato-
graphy GC and/or •>GC/MS
(8) Organic phase
^residue at 550°C -*• AA
and/or XRF; and/or ICP
(9) Aqueous phase ->acidity/alkal inity ->•
Beilstein test; and/or
spot test; and/or redox
potential; and/or anions
analysis; and/or vola-
tiles; and/or organic
solvent extraction
(10) Solid phase
•pathways (11) to (15)
(for solids) (not inclu
ded nor discussed in
this paper)
Analytical pathway (l) is chosen when
analysis for metals is required. Instru-
mental methods such as the AA, XRF or ICP
are used for the analysis. Wet chemical
methods can also be used for analyzing the
individual metals, but presently are sel-
dom used because they are often time con-
suming and suffer from poor reproducibi11-
ty. The AA is the least expensive of the
three instruments, but metals can only be
analyzed only one at a time. The XRF and
ICP are very expensive, however, up to 60
metallic elements can be analyzed simulta-
neously.
Analytical pathway (2) is to separate
the different phases of a mixed sample. A
portion of the sample is centrifuged at
7,000 to 17,000 RPM. The separated phases
are isolated and analyzed according to
pathways (3) to (10).
The first analysis that might be per-
formed on the organic phase is flash point.
This is indicated in analytical pathway
(3). Flash point is determined by the Tag
Closed Cup procedure. When the flash
point exceeds the hazardous flammability
standard, further analysis of the sample
may not be necessary.
-
^YI_^^-J
i^s &<*/•
"~«T-..'C
-------
In analytical pathway (5), thin layer
chromatography (TLC) is first employed
before instrumental analysis is attempted.
In many cases, components of hazardous
waste samples are separated by TLC. Dif-
ferent TLC systems as well as the applica-
tion of different developing or chromogenic
sprays on the separated components often
indicate the presence of distinctive types
of compounds. Comparison of the Rf values
or the developed color of the TLC spots
with that of standards often presumptively
identify the compounds. The tentative
identifications are further confirmed by
either GC, HPLC or by GC/MS. The TLC in-
formation narrows down the selections, of
the instrumental analytical parameters.
For example, if by TLC systems a sample
contains mostly chlorinated hydrocarbons,
immediately the instrument of choice will
be most likely the GC with electron capture
detector. If high-boiling organics are
indicated, confirmatory analysis can most
likely be achieved by using the HPLC.
Preliminary analysis by TLC in most
cases saves considerable analysis time and
cost. It should be used for quick screen-
ing of the components of samples.
In analytical pathway (6), the distil-
late at 95°C is first collected. The dis-
tillate and/or the higher boiling fraction
are analyzed by either or combinations of
pathways (4), (5), (7) or (8). The distil-
lation segregates the lower-boiling compo-
nents thus enhances faster identification.
The lower-boiling fraction may contain com-
pounds with lower molecular weights as well
as those responsible for flash points below
37.8°C.
In analytical pathway (7) several
simple tests are first performed before the
instrumental analysis. The preliminary
test provide information such as the pre-
sence of functional groups or classes of
compounds in the sample. Often times ten-
tative identification of compounds is
achieved by these tests. Table 3 outlines
the information that results from these
tests. The preliminary information obtain-
ed serves as the basis for selecting the
analytical instrument and/or confirmatory
operating parameters such as the examples
given in (4) for further analysis of the
sample.
Analytical pathway (8) requires ashing
of the sample at 550°C. Appearance of a
residue indicates the presence of organo-
metal in the sample. The residue or sample
is then analyzed by AAS, XRF or ICP for the
metal(s).
The separated aqueous phase of the
sample is analyzed using pathway (9). This
pathway consists of several tests that may
be performed. Combinations or all of the
indicated tests might be necessary to.ana-
lyze the sample. The measurement of pH is
more accurately done with a pH meter.
Acidity/alkalinity is determined by titra-
ting a portion of the sample with standard
base or acid to pH 3-7- High acidity/
alkalinity values indicate the presence of
strong base or acid. Positive Beilstein
test usually indicates the presence of
chlorinated hydrocarbons. Spot tests can
show the presence of functional groups,
classes of compounds and dissolved com-
pounds. Redox potential can indicate the
presence of highly ionized electrolytes.
An ion analysis can identify negative ions
such as sulfates, nitrates, cyanides and
others. Water soluble organics can be se-
parated from the aqueous phase by distil-
lation, volatilization or by organic sol-
vent extraction. Other phases are analyzed
by different routes through the analytical
scheme.
Analytical pathways intended for
solids (11 to 15) and liquids (16 to 18)
will also be included in the analytical
scheme. These will not be discussed in
this paper.
This report is part of a more complete
work for the preparation of an analytical
procedural manual for the analysis of
hazardous wastes. The work has been sup-
ported by the U.S. Environmental Protection
Agency, Municiapl Environmental Research
Laboratory, Cincinnati Ohio, Richard A.
Carnes, Project Officer.
OH,
20
-------
HAZARDOUS WASTE COMPATIBILITY
Howard K. Hatayama, P. E.
Robert D. Stephens, Ph.D.
Emil R. de Vera
James J. Chan
David L. Storm, Ph.D.
ABSTRACT
This report describes a method for determining the compatibility of the binary
combinations of hazardous wastes. The method consists of two main parts, namely:
1) the stepwise compatibility analysis procedures, and 2) the hazardous wastes com-
patiblity chart. The key element in the use of the method is the compatibility
chart. Wastes to be combined are first subjected through the stepwise compatibility
procedures for identification and classification, and the chart is used to predict
the compatibility of the wastes on mixing.
INTRODUCTION
The method of determining hazardous
waste compatibility briefly described
herein is an attempt to provide those in
the waste management industry and the
regulatory agencies with a systematic
approach to reducing the risk to the
public health and the environment of
handling, processing, transporting and
disposing hazardous wastes. Many ser-
vice accidents have resulted from the
mixing of incompatible hazardous wastes.
These often violent chemical reactions
have resulted in damage to public and
private property, injury and sometimes
death to individuals, and disposal of
hazardous materials in the environment.
These accidents most often occur because
waste handlers have inadequate knowledge
of the chemical compositions of the
wastes, or are unaware of how chemical
components of different waste types
interact.
The objectives of the method are
to:
1) Present the chemical reactions which
are likely to produce signficant
hazards to waste handlers and the
environment.
2) Present a listing of chemical classes
based on molecular structure and
chemical reactivity which typically
occur in wastes.
3) Provide guidelines for estimating which
chemical classes occur in specific
wastestreams.
4) Provide a method for estimating the
potential consequences of mixing of
different classes of wastes.
This paper very briefly describes the
method which was presented to the U.S.
Environmental Protection Agency, Municipal
Environmental Research Laboratory in the
form of a manual or handbook (Hatayama,
et al.) under a grant entitled, "Hazardous
Waste Sampling, Analysis, and Compatibility
Study." The contract officer for this work
is Richard A. Carnes.
The manual should be considered an
interim report on the study of hazardous
waste interactions. Many shortcomings in
the available data were revealed during
the course of its development. These are
21
-------
currently being address by laboratory in-
vestigations. Development of a field
method for testing hazardous waste com-
patibilities is also underway.
The best available knowledge of case
histories of accidents, wastestream
composition and chemical reaction conse-
quences was used in developing this method.
Department of Health Services files
(California Department of Health Services3),
U.S. EPA Hazardous Waste Disposal Damage
Association (Manufacturing Chemists Asso-
ciation^), and other sources. Wastestream
composition data was derived from the
California Liquid Waste Hauler Record
(California Department of Health Services^),
U.S. Environmental Protection Agency's
hazardous waste surveys (Assessment) of
Industrial Hazardous Waste Practices .Chem-
ical reactivity data was derived from
textbooks, and sources dealing primarily
with hazardous chemical interactions such
as the Handbook of Reactive Chemical
Hazards (Bretherick2), the Fire Protection
Guide on Hazardous Materials (National
Fire Protection Association^), Dangerous
Properties of Industrial Materials (Sax10) ,
and others.
HAZARDOUS WASTE COMPATIBILITY CRITERIA
Many types of hazardous wastes, are
extemely reactive. In combination with
other wastes or other materials, such
wastes may react to cause very undesirable
and uncontrollable consequences. These
adverse consequences include: 1) heat
generation (reaction code H). 2) fire (F),
3) innocuous gas N£ or CC>2» such as gen-
eration which can cause pressurization
(G), 4) toxic gas, such as HCN or H2S,
generation (GT), 5) flammable gas, such
as H2 or C2H2, generation (GF), 6)
explosion due to extremely vigorous
reactions or reactions producing enough
heat to detonate unstable reactants or
reaction products (E), 7) violent poly-
merization resulting in generation of
extreme heat and sometimes toxic and
flammable gases (P), 8) solubilization
of toxic substances including metals (S).
These consequences often cause injury
and sometimes death to individuals in
the area. They result in the disposal of
extremely hazardous materials which
threaten the environment and public health.
CAUSES OF ACCIDENTS
The full manual, "A Method for
Determining Hazardous Waste Compatibility"
(Hatayama, et al.5) documents twenty-four
incidents involving adverse consequences
from mixing of incomaptible hazardous
wastes. This is not a complete list by
any means but it provides a basis for the
above mentioned compatibility criteria and
for general observations relating to causes
of such incidents. Three primary causes
have been identified.
The first primary cause is the insuf-
ficiency or inaccuracy of information about
the waste. Hazardous wastes are often
complex mixtures of chemicals. In order
to define them accurately, laboratory
analysis is often required. This is
expensive and frequently not performed.
Waste generators often do not maintain
adequate records of the components of their
wastestreams. In some cases, information
about certain wastestreams are deleted or
altered to reduce the cost of disposal.
In other instances, the properties of
some waste change with time and temperature
thereby producing more hazardous and
unknown components. Persons handling the
wastes often have insufficient or in-
accurate information on which to base
transportation and disposal decisions.
The second primary cause of accidents
is indiscriminate handling of the wastes.
Often supposedly "empty" containers
actually contain hazardous residues which
react adversely with the materials added.
Haulers, uninformed of hazardous chemical
interactions, often "top-off" their loads
on the way to the disposal site (Calif-
ornia Department of Health Services3).
This often initiates chemical reactions
in the tank truck which result in violent
22
-------
HAZARDOUS WASTE COMPATIBILITY CHART
EBSJWF
i
2
3
4
S
6
7
t
9
10
II
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
2*
29
30
31
32
33
34
101
102
103
104
IOS
1M
107
REACTIVITY CROUP NAME
Acids. Mineral. Non-ox idlilng
Acids. Mineral, Oxldi/iii|
Acids. Orpnk
AkoholsindClycub
Aldehyde!
Amides
Amines. Aliph.ilic .ind Aromatk
A/II Compounds, l)i.i/ti Compuunds, and ll)drazincs
Carhamalrs
Cnnltn
Cyanides
Dilhi.H.irli.im.ik".
F.MCP,
Ethers
Fluorides, lumgantc
ll)ilroc.irbons. Aromatic
llalii£en.ilcd Organic*
Isocyanates
Kclones
Mcrranlaits and Oilier Organic Sulfidcs
Metals. Alkali and Atkahnc i"jrth. elemental
Mel.i)s. Other rieinenljl 4 Allo)s js Powders, Vapors, or Sponges
Metals. Other Clcinenl.il 4 Allujs as Sheets. Rods. Drops. Moldings, ele.
Metals antl Metal Compounds, Toxic
Nitrides
Nitrites
Nilro Compounds, Organic
Hydrocarbons, Aliphatic. Unsaturatcd
Hydrocarbons, Aliphatic. Saturated
Peroxides and Hydropcroxldes. Organk
Phenols and Cresob
Orianophosphalcs. Phoipholhioales . PhosphodilMoiln
Suindn. Inorpnk
Epoxldes
Combustible and Flammable Materials. Miscellaneous
Explosira
Poiymniubte Compounds
Oxldiiltif Atents. Slron|
Reducing Agents. Sirenf
Water ind Mixtures Containing Wittr
Water Rracltn Subslincn
1
H
HP
H
H
HC
Hc
H
CT
CF
H
H
H
CT
HCT
Hc
H
CT
CF
HF
CF
"F
\
s
CF
HF
VF
H
"c
H
Gl
Cf
"P
HC
HE
fH
CT
H
CF
H
2
C
H
HF
H
F
HCT
HCT
GT
HCT
H
CT
CF
H
HF
H
F
CT
"F
'F
GT
HF
\T
Oh
II,
CF
"F
\
s
'FE
H
'Fci
HF
«F
HE
HF
CT
HF
CT
"p
\
HE
'H
HF,
H
3
H
P
HP
H
H
G
H
CT
CF
VT
CT
"c
01-
CF
S
HCF
H
CT
HP
HE
'H
CT
CF
4
H
C
HP
Oh
"l
'*F
HF
HP
HF
CFr
5
H
H
H
Ol
01
JlE
'''H
H
H
HC
H
U
HF
CP
"F
6
Oh
II
S
"F
01
"H
7
U
"CT
HP
Oh
II
S
"CT
HP
«F
o^iT
'~-^rt &
12
U
CFC1
WH
%
u
u
"F
Hr.T
13
Cl
II
°FH
HF
HF-
HF
REACTIVE!
II
12
13
s*&>*.
23
-------
HAZARDOUS WASTE COMPATIBILITY CHART (Continued)
Read t> it) Code
14
HF
DO
14
IS
16
_.
HF
17
H
"t
II
E
II
F
Gi-
ll
"E
HCT
HE
18
II
or
H
Gt
II
u
H
HP
H
''„,
Wi.
HC
19
H
GF
II
GF
H
E
'F
4H
NOT MIX WITH ANY
15
16
17
IK
19
20
GF
II
II
«F
GF
H
H
FGT
HP
"F
CT
"H
21
E
"r
GFE
HE
CF
H
H
HP
\
HE
PH
%
°>H
22
HE
"c
HP
HE
)
H
\
°KH
23
HE
>
H
HF
24
S
"c
HP
E
'H
s
CHEMICAL OR WASTE
20
21
22
23
24
25
GF
H
H
CFF
GFp
CF
H
HP
V
E
PH
'FE
°KH
26
"CT
HF
CT
HCF
H
F
G
CT
CF
E
P
S
U
Fx.implc
HF
hCT
27
HE
HE
MATERIAL!
25
26
27
28
"p
H
F
29
H
10
H
U
H
CT
HP
«F
CT
HE
'«
H
C
"E
EXTREMELY
28
29
30
Consequences
Meal generation
Fire
Innocuous and non-fl.imm.ibk- gas genera lion
Toxic gas generation
Fl.unm.iblo g.is gcncfjlmn
Explosion
ViuVnt put) mcri/atiun
So iibili/jhon (if loxic substances
M.-I) be h.i/.-mlom but iniknuwn
Mont gcncratinn, fire, .uid toxic gas gcnerntion
3!
32
33
Hp V Hp 34
101
H£ H£ H£ H£ ,02
fH 'H HE '05
"'\AA\"*\r»»
CFH^, " "H "k i ^ "*
COF <& '«
31 32 33 34 101 102 101 104 105 1 106 107
24
-------
WORKSHF.FT FOR DF.TFRMINATION
OF HAZARDOUS \VASTC COMPATIBILITY
Waste A
Waste B
Name
Date
25
-------
and disastrous consequences. Rough hand-
ling of waste containers have resulted
in the rupture or leakage of highly re-
active materials (California Department
of Health Services^). Inadvertant mixing
of materials known to be incompatible have
resulted in numerous accidents.(Manufact-
uring Chemists Association.^) Inadequate-
ly designed chemical treatment processes
for purposes of detoxification of or
resource recovery from hazardous waste has
resulted in uncontrolled reactions. (U.S.
Encivronmental Protection Agency).
The third primary cause is indiscrim-
disposal. Bulk wastes which were known to
be incompatible with already ponded wastes
have been indiscriminately disposed of to
the ponds at disposal sites. Some wastes
are incompatible with the composition of
the disposal area such as concentrated
nitric acid and refuse in a sanitary
landfill or acid in an abandoned salt mine.
Containerized wastes, irrespective of con-
tents, often are co-disposed and hazardous
reactions result when the containers
rupture or leak due to corrosion. (Calif-
ornia Department of Health Services3).
BASIS FOR THE METHOD
The principal assumption underlying
the method is that waste interactions are
due to the reactions produced by the pure
chemicals in the wastes. Included in this
assumption is the condition that the chem-
icals react at ambient temperature and
pressure and that their reactivities are
uninfluenced by concentration and synergis-
tic and antagonistic effects. By this
assumption, the compatibility of a com-
bination of wastes can be predicted by the
reactivities of the chemical constituents
in the respective wastes.
Available data indicate that hazard-
ous wastes are ill-defined, complex mixture
generated by a great variety of sources.
No two types of wastes appear to be iden-
tical. Even a single process appear to
produce different types of wastes. Lab-
oratory analyses of wastes are not easily
obtainable and are often very cursory due
to high costs and the complexity of analy-
tical methods required. Characterization
of the wastes by analysis of the processes
and the materials used appear to give
inaccurate descriptions of the resulting
wastes. The data indicate that each waste
is unique and their individual reactivi-
ties may be best assessed by identifying
their respective chemical constituents.
This information supports the pure chem-
ical approach used in determining the
reactivities of the wastes in the develop-
ment of the compatibility method.
HAZARDOUS WASTE COMPATIBILITY CHART
The key to this method is the Hazard-
ous Waste Compatibility Chart (Figure 1).
It represents a two dimensional matrix
of forty-one functionality, (Reactivity
Group Number (RGN) 1 through RGN 34),
and general chemical reactivities (RGN
101 through RGN 107). The more than one
thousand compounds considered in the deve-
lopment of this method are classified into
one or more of these reactivity groups.
The potential consequences of mixing
compounds from one reactivity group with
compounds from another are denoted by
reaction codes in the square representing
the combination of their RGN's. The
reaction codes (RC) and consequences are
detailed in a Table to the right of the
matrix. These codes represent the Hazard-
ous Waste Compatibility Criteria discussed
earlier.
Many reaction squares contain multiple
RC. These indicate that several consequ-
ences are predicted from the combination
of two RGN. The order of the codes corres-
ponds to the order in which the consequen-
ces are expected to occur. This is true
for all multiple RC except for GTGF,
where the gas that is produced is both
toxic and flammable.
Many reaction squares contain the RC-
U for "unknown". In these cases, the
literature indicates that an adverse reac-
26
-------
tion between members of these RGN may occur.
But, the consequences cannot be predicted
at this time and actual experimental work
may be necessary.
Blank reaction squares represent com-
binations which appear to be compatible.
The literature indicates that adverse
reaction consequences are not expected.
The number of RGN 107, "Water Reactive
Substances" are considered to be extremely
reactive that they should not be combined
with any waste chemical or waste material.
Although some neat chemicals are combined
specifically with water reactive materials
to preserve or protect them from decompo-
sition, waste chemicals typically contain
other constituents which may alter their
protective chracteristics.
SUPPLEMENTS TO THE CHART
The chart_is_augmented by four appen-
dices which are essential to determining
the compatibilities of real wastes. The
first is a list of chemical substances.
This appendix lists are the compounds with
appropriate synonyms which were considered
in developing this manual in alphabetical
order. It also lists in an adjacent column,
the reactivity groups assigned to each
compound. This appendix is used to determ-
ine the RGN(s) of a known chemical consti-
tuent of a waste.
The second appendix is a list of Waste
Constituents by Chemical Class and Reacti-
vity. It categorizes the compounds in the
List of Chemical Substances into reactivity
groups based on chemical functionality such
as alcohols or carbamates, and on general
chemical reactivity such as strong oxidiz-
ing agents and strong reducing agents.
These reactivity groups correspond to those
on the chart. This appendix is used to
obtain the RGN(s) of a hazardous waste when
its chemical constituents are known by
chemical classes, molecular functionality
or chemical reactivities. An example is
if a waste is identified as containing
alcohols, ketones, and water.
The third appendix is an Industry
Index and List of Generic Names of Waste-
streams. The Industry Index list names
of industries in general terms alphabet-
ically with their corresponding Standard
Industrial Classification (SIC)(Standard
Industrial Classification Manual) code
numbers. This index is used to determine
the general industrial class of a waste
generator. The SIC code number obtained
then allows entry to the List of Generic
Names of Wastestreams which contains the
names of typical wastestreams generated by
industry. Such names include Citrus Pectin
Wastes, Chromate Printing Wastes, Acetylene
Manufacturing Sludges, etc. Each waste is
assigned RGN(s) based on hazardous chemical
constituents known or expected to be pre-
sent in the wastes. This appendix is used
to determine the RGN of Wastestreams when
their compositions are not known specifi-
cally but are identified by their generic
or common names.
The fourth appendix is a List of In-
compatible Binary Combinations of Hazard-
ous Waste Reactivity Groups and the Poten-
tial Adverse Reaction Consequences. This
appendix describes in detail the potential
adverse reaction consequences predicted in
the Hazardous Waste Compatibility Chart.
It is written for the hazardous waste
manager who is well versed in reaction
chemistry and can be used to make a more
refined assessment of the compatibility of
two waste types.
GENERAL PROCEDURES FOR DETERMINING
HAZARDOUS WASTE COMPATIBILITY
The basic requirement for using this
method is to classify the chemical compo-
nents of a waste into one or more of the
forty-one reactivity groups. Each RGN of
one waste is then matched with each RGN of
another waste to determine this compatibi-
lity. There are five general steps in the
procedure:
Step 1 Obtain as much information as
possible about the history and
composition of the wastes. Such
27
-------
in many aspects of the management of hazard-
ous wastes. It will be useful to the waste
generators in identifying and segregating
their wastes for disposal; to the trans-
porters for segregating, combining, and/or
proper containerizing of the wastes; to the
site operators for determining co-burial of
containerized wastes in the same cell or
co-ponding of bulk wastes; to the regulatory
agencies for determining suitability of
sites for disposal of certain wastes; and
to those who perform chemical treatment of
the wastes for purposes of detoxification
or resource recovery to present possible
uncontrolled reactions.
This method cannot be used to predict
all the potential incompatible reactions of
any two given wastes, and neither can it
furnish information on all hazardous waste-
streams due to the tremendous variety of
waste types, constituents, and character-
istics. Additionally, the method does not
address ternary combinations of incompatible
hazardous wastes.
SUMMARY AND CONCLUSIONS
An extensive review of the literature
and surveys of hazardous wastes management
practices has shown that adverse reactions
can result from the mixing or combination
of incompatible hazardous wastes. These
reactions have been categorized into twelve
classes on the basis of reaction products
with the potential of causing public health
and environmental damage. The twelve
classes are: 1) heat generation, 2) fire,
3) gas formation, 4) formation of toxic
fumes, 5) generation of flammable gases,
6) volatilization of toxic or flammable
substances, 7) formation of substances of
greater toxicity, 8) production of shock
and friction sensitive compounds, 9) pres-
surization in closed vessels, 10) solubili-
zation of toxic substances, 11) dispersal
of toxic dusts, mists and particles, and
12) violet polymerization.
Three primary causes of the combination
of incompatible wastes were identified,
namely:
1) insufficiency or inaccuracy of inform-
ation about the wastes
2) indiscriminate handling of the wastes,
and
3) indiscriminate waste disposal practices.
In order to present and/or minimize
the chances of combining incompatible
hazardous wastes and to avoid the resulting
adverse reactions, it was determined that
a method of determining wastes compati-
bility is necessary. Such a method was
developed for the binary combinations of
wastes types. A compatibility method
addressing ternary or more combinations
was considered, but found to be unwieldy.
In the binary method the potential for
occurrence of any one of the twelve
identified reactions was taken as an
indication of incompatibility. The determ-
ination of the occurrence of incompatible
reactions was based on the assumption that
the waste reactions are results of pure
chemical components of the wastes reacting
at ambient temperature and pressure. These
assumptions are made primarily for reasons
of simplification, however, it is believed
that they are justified in view of most
disposal and transport situations.
The development of the stepwise pro-
cedures for the compatibility method re-
quired the assignment of waste components
into reactivity groups based on molecular
functionality and reactivity character-
istics. Using this procedure, it was found
that the reactivity group(s) of the com-
ponents of one waste paired with the
reactivity groups of another waste could
predict the potential occurrence of certain
incompatible reactions. A two dimensional
graphic display was determined as the best
method for presenting the reactivity groups
and allowing for intergroup pairing. This
resulted in the development of the compati-
bility chart presented in Figure 1. Color
coding of group pairings can be included
to aid in rapid determination of potential
incompatibilities.
A primary conclusion which was reached
from this work was that there is a dearth
28
-------
information can usually be obtained
from records of the waste producers,
the manifests that accompany the
wastes and examination of the pro-
cesses that produced the wastes.
When no information is available,
collect representative samples
of the wastes and submit them for
analysis. The analysis shoud
provide information on the specific
chemical constituents or classes of
compounds in the wastes.
Step 2 Starting with one waste, Waste A,
List the names of or the classes
of compounds found in the waste,
or list its generic name on the
vertical axis the Worksheet for
Determination of Hazardous Waste
Compatibility (Figure 2). The
composition of a waste is Known
Specifically when the constituents
are listed by chemical names such
as ethylene glycol, sodium nitrate,
etc. The composition is Known Non-
specifically by Classes when the
constituents are identified only by
chemical classes or reactivities
such as alcohols, caustics, mer-
captans, etc. The composition is
Known Nonspecifically by Generic
Name when the waste is classified
as spent caustic, tanning sludge,
copper plating waste, etc.
Step 3 When the composition of Waste A is
Known Specifically by chemical
names, consult the List of Chemical
Names to obtain the RGN's for each
chemical constituent. These RGN
are then noted on the Worksheet.
If a compound is not on the list,
a synonym can be found in various
chemical references (Merck**,
Hawley"). When a suitable synonym
cannot be found, the RGN of the
component may alternatively be
determined based on its chemical
class or reactivity.
When the composition of the waste
Step 4
is Known Nonspecifically by Classes,
consult the List of Waste Constitu-
ents by Chemical Class and Reacti-
vity to determine the corresponding
RGN.
When the composition of the waste
is Known Nonspecifically by Generic
Name, go to the Industry Index and
List of Generic Names of Wastes to
obtain the corresponding RGN are
noted on the worksheet.
Repeat Steps 2 and 3 for the second
waste, Waste B, and note the infor-
mation on the horizontal axis of
the Worksheet.
Step 5 Consult the Hazardous Waste Compati-
bility Chart (Figure 1) and note
the Reaction Codes (RC) between
all binary combinations of RGN of
Waste A and Waste B. If any RC
corresponds to any binary combin-
ation of RGN between Wastes A and
B, then Wastes A and B are incom-
patible and should not be mixed.
Details of the potential adverse
reactions can be obtained from the
List of Incompatible Binary Combin-
ations of Hazardous Wastes Reacti-
vity Groups and the Potential
Adverse Consequence.
SCOPE, APPLICATION AND LIMITATIONS OF THE
METHOD
The method provides a systematic
approach for determining the compatibility
of most binary combinations of hazardous
wastes produced by industry and agriculture.
The method also provides a list of compounds
known or expected to be present in hazardous
wastes. Lastly, the report classifies the
compounds as well as the wastes into chemi-
cal reactivity groupings and lists the
potential adverse reaction consequences of
most incompatible binary combinations of
the groupings.
This method will be useful reference
29
-------
of information about the reactivities of
chemicals in the complex matricies of
wastes. Many factors assuredly do greatly
influence waste component reactions. Among
these are temperature, catalytic effects of
dissolved or particulate metals, soil re-
actions and reactions with surfaces of
transport vehicles or containers. The
simplified compatibility methodology which
has been developed in this study, however,
should provide a useful aid to persons
involved in generating, transporting,
processing, and disposing of hazardous
wastes if reasonable precaution is taken
in its use.
REFERENCES
1. Assessment of Industrial Hazardous
Waste Practice Series. 1974-1976.
U.S. Environmental Protection Agency,
Washington D.C.
2. Bretherick, L. 1975. Handbook of
Reactive Chemical Hazards. CRC
Press, Inc., Cheveland, Ohio.
3. California Department of Health Ser-
vices. 1978. Vector and Waste Manage-
ment Section Files. Berkeley, Los
Angeles, and Sacramento, California.
4. California Department of Health Ser-
vices. 1978. California Liquid Waste
Hauler Record. Berkeley, CA.
( 5. Hatayama, H.K., J. J. Chan, E. R.
\ de Vera, R. D. Stephens, and D. L.
; Storm, 1980. A Method For Determining
*\ Hazardous Waste Compatibility. U.S.
Environmental Protection Agency,
\ Cincinnati, Ohio.
6. Hawley, G.G. 1971. The Condensed
Chemical Dictionary. 8th Edition.
Van Nostrand Reinhold Company,
New York, Cincinnati, Toronto,
London, Melborne.
7. Manufacturing Chemists Association.
1962, 1970, 1976. Case Histories
of Accidents in the Chemical Industry.
Washington, D.C.
8. Merck and Company, Inc. 1976. The
Merck Index. 9th Edition, Rahway,
N.J.
9. National Fire Protection Association.
1975. Fire Protection Guide on
Hazardous Materials. Boston, MA.
10. Sax, I.N. 1968. Dangerous Properties
of Industrial Materials. 3rd Edition.
Van Nostrand Reinhold Company,
New York.
11. U.S. Environmental Protection Agency.
1976. Hazardous Waste Disposal
Damage Reports. Washington,D.C.
30
-------
MONITORING WELL SAMPLING AND PRESERVATION TECHNIQUES
James P. Gibb
Illinois State Water Survey
Urbana, Illinois 61801
Rudolph M. Schuller
and
Robert A. Griffin
Illinois State Geological Survey
Urbana, Illinois 61801
ABSTRACT
Water sample collection and preservation techniques have been established by several lab-
oratories and agencies to insure that water samples received are chemically representative
of water contained in the aquifer being monitored. However, there is considerable con-
troversy between laboratories, agency policies, and researchers concerning the proper
sampling techniques from monitoring wells and the appropriate preservation procedures of
the samples for various chemical constituents. The development of recommended sampling
procedures and preservation techniques for certain types of wells and specific chemical
constituents is the principal goal of this study. Samples have been collected from six
study sites using four different pumping mechanisms and various sample preservation tech-
niques. This paper describes the experimental procedures being used and presents pre-
liminary results based on work to date. Data collected thus far demonstrate that the
length of pumping (time), type of pumping mechanism used, and size of membranes used to
filter the samples all affect the chemical composition of water collected from monitoring
wells.
INTRODUCTION
Regulatory agencies are charged with
the assignment of regulating the disposal
of waste to insure that the environment is
not adversely affected. To accomplish this
task, it is necessary for these agencies to
set design and operational standards based
on available technology to minimize poten-
tial pollution. The operating disposal
facilities must then comply with these
standards. They also must monitor the
effects of their operation on the surround-
ing environment. The use of wells or
piezometers for collecting water samples
and water level data has been, and probably
will continue to be, the traditional method
for monitoring the effects of waste dis-
posal facilities on groundwater.
Much research has been conducted to
develop analytical laboratory techniques to
detect the low levels of various constit-
uents set forth in water quality standards.
Water sample collection and preservation
techniques have been established by several
different laboratories and agencies in an
attempt to insure that water samples deliv-
ered to the laboratory are chemically rep-
resentative of water contained in the
aquifer being monitored. However, there
is considerable controversy between labora-
tories, agency policies, and researchers
concerning proper techniques of sampling
from monitoring wells and the appropriate
procedures for preserving the samples for
various chemical constituents. If monitor-
ing wells and water samples are to provide
the performance yardstick of disposal
31
-------
facilities design and operation, the ques-
tion of significance of the various sampling
procedures and preservation techniques
should be determined.
This report is the result of work
performed on a project partially funded by
the U.S. Environmental Protection Agency,
Cincinnati, Ohio, Grant No. R-806304-01-0.
This report should be considered prelimi-
nary and subject to reinterpretation.
Purpose
The three principal purposes of this
study are: 1) to determine if current
sampling methods produce samples that are
representative of water contained in the
aquifer being monitored; 2) to determine if
groundwater samples collected in the field
must be treated (filtered and acidified) on
location, or if they can be brought back to
the laboratory for treatment without alter-
ing their chemical nature; and 3) to deter-
mine which sampling and preservation tech-
niques should be accepted as standards for
monitoring well sampling.
Specific Objectives
1. Determine the hydrologic properties of
the materials tapped by each monitoring
well studied.
2. Determine a pumping scheme for each
well to obtain water samples represent-
ative of aquifer water.
3. Collect a series of samples from each
well using four different pumping
methods and various preservation tech-
niques.
4. Determine the effects of time of pump-
ing, pumping rate, pumping mechanism,
and preservation techniques on the
chemical results of samples collected.
5. Recommend sampling procedures and pres-
ervation techniques for monitoring wells
for specific chemical constituents.
SITE SELECTION
Monitoring wells at six sites have
been selected for study. These sites rep-
resent a cross section of hydrologic condi-
tions and chemical characteristics of
aquifer water. Two sites are active land-
fills; two are inactive landfills; one is a
secondary zinc smelter; and one is a hog
processing plant. The wells at all sites
were in existence prior to this study and
all but one was constructed by persons
other than the authors. Five are cased
with PVC pipe either 3.81- or 5.08-centi-
meters (1^- or 2-inches) in diameter. The
sixth well is cased with 5.08-centimeter
(2-inch) diameter galvanized iron. The
well depths range from about 5 to 10 meters
(16 to 30 feet) and have non-pumping water
levels from 0 to 5 meters (0 to 16 feet)
below land surface.
Factors considered in selecting the
sites included: 1) accessibility to the
monitoring wells so that pumping tests and
sampling could be accomplished without un-
due hardship; 2) the physical characteris-
tics of the monitoring well and geology of
the location; 3) the potential yield capa-
bility of the materials tapped by the moni-
toring well; and 4) the chemical quality of
water obtained from the monitoring wells.
PUMPING EQUIPMENT AND TESTS
Pumping Methods and Equipment
Egur^methods of obtaining samples from
the monitoring wells are being used. These
include samples collected by mechanical
pumping means, an air-lift system, a
nitrogen-lift system, and bailing. Most
monitoring wells examined to date were 1^-
inch or 2-inch diameter PVC wells. The
small diameters create rather severe
limitations on the selections of pumping
apparatus.
The mechanical pump used was a Master-
flex 7545 ^variable "speed drive unit equipped
with a model 7015 peristaltic pump head.
When operated on 115 volts (a portable
generator is used in the field) , it is
capable of producing from about 50 to 1000
milliliters per minute (0.013 to 0.264
gallons per minute) .
mAi fU.
• Tjf- *v For air^j-ift and nitrggen-lif t pumping
systems an apparatus similar to that shown
in Figure 1 was used. For some sampling
runs, a 1.27-cm (%-inch) diameter rigid PVC
discharge pipe and 0.635-cm (%-inch)
plastic airline were used. For others a
0.952-cm (3/8-inch) diameter flexible poly-
vinyl discharge line and 0.635-cm (%-inch)
plastic airline were used. For all air-
lift pumping a four-cylindar electric-
driven air compressor was used.
For the nitrogen-lift pumping runs,
8.5 cubic meter (300 cubic foot) cylindars
of nitrogen were used. A pressure regula-
tor and Matheson type flowmeter were used
32
-------
1/2" cap
(1.27cm)
1/2" Tee
(5.08 cm) 2" cap
2" casing
1/2" discharge
pipe
7" bore hole -*-
(17.78cm)
slotted PVC
well casing
(7.62cm) 3"
\,
— \
\ -"
' o o
a o
o o o
*
0 O
00
0 O
00
°0
o °
I
°o°
--
1
=
=
=
1
=
"x-
n
^^-
§
=
^=
=^
-- A
/ '
S. '
!.>'.-.
ooo'
o o
0 0°
CO
°0 0
°0°
°0°
O o
o"
0 0
0°0
°0
o o
J
bentonite
slurry
1/4" airline
(0.635 cm)
I 6" sand (15.24 cm)
12" (30.48cm)
E = 0°0 -«—gravel
Figure 1. Typical Well and Air-lift Pumping Mechanism
to control the pressure and flow rate to
the gasline.
Experience gained during the first
year of the project concerning air-lift
and nitrogen-lift pumping has demonstrated
the need for more detailed experimentation
with this type of pumping system. Labora-
tory data will be collected during the next
year to more accurately define 1) the flow
rates possible with varying discharge pipe
size, 2) the amount of air or nitrogen
required for given physical settings and
desired water discharge rates, and 3) the
limitations of this type of pumping mech-
anism.
A 2.54-cm (1-inch) diameter stainless
steel bailer, 0.914-meters (36-inches) long
was constructed to bail the test wells.
This bailer retrieves a sample of about
350 milliliters.
Pump Test Results
Three pumping tests have been success-
fully completed to date. Discussion in
this paper will be limited to the results
of the pumping test at Site 1.
Site j.—Well No. 12 at the Sandoval
Zinc Company site is a 5.08-cm (2-inch)
diameter well 4.25 m (14.0 feet) deep.
The well is cased with PVC pipe and has
slots sawed in the bottom 60.96 cm (2 feet).
The casing extends 54.87 cm (1.8 feet)
above land surface.
A pumping test was conducted on this
well on January 16, 1979. The well was
pumped with a Jabsco self-priming pump at
rates varying from 1750 to 1000 ml/min
(0.462 to 0.264 gpm) for a period of 15
minutes before the pump broke suction.
Water level recovery data were collected
for a period of 30 minutes after pumping
stopped. Discharge measurements were made
with use of a graduated cylinder and stop
watch. Water level measurements were made
with a steel tape.
Two different approaches have been
taken in analyzing all pump test data.
First, the techniques developed by
Papadopulos and Cooper were applied to
determine an aquifer transmissivity. In
the case of the Sandoval test, the adjusted
recovery data were used to determine an
aquifer transmissivity of about 3600 m /day
(200 gpd/ft). The aquifer transmissivity
is used to compare the relative water
yielding character of each site.
The second approach used in analyzing
the pump test data is much more simplistic
and problem oriented. Adjusted recovery
data were examined at the end of each time
increment during the recovery portion of
the test to determine the rate of water
delivered to the casing. This rate was
then related to the average drawdown over
the time increment of interest. Regression
analysis relating aquifer yield (Qa) to
total water level drawdown (Tdd) for the
Sandoval pump test data yielded the rela-
tionship Qa=109.7+110.7 (Tdd). This rela-
tionship can be used to simulate total
drawdown in the Sandoval well for different
pumping rates.
In addition to being able to develop
theoretical drawdown curves from this re-
lationship, the amount of water coming from
the aquifer can be calculated for each time
increment. These amounts, represented as
percentages of total (Qt) pumpage [(Qa/Qt)
*100], are presented in Figure 2 for pump-
ing rates of 2000, 1000, and 200 ml/min
(0.528, 0.264, and 0.005 gpm). Based on
these calculations for the Sandoval well,
it appears that a pumping rate of 200 ml/
min should produce water samples containing
a higher percentage of aquifer water in the
F
e o
33
(7 /
-------
100
15 20
TIME, minutes
Figure 2. Site 1: Aquifer Yield Curves
shortest period of time. This statement is
based on the assumption that the pump in-
take is located in the screened portion of
the well near the bottom so that minimal
mixing of aquifer and stored water from the
casing will occur.
SAMPLE COLLECTION AND ANALYSIS
The techniques employed in the storage
and preservation of water samples collected
for environmental assessment will directly
affect the chemistry of those samples.
Many investigators have studied the effects
of storage and preservation of aqueous
samples. Rattonetti3 studied the effects
of preservation on surface waters and con-
cluded that filtration and subsequent acid-
ification were needed to maintain sample
integrity. Subramanian et al.7 looked at
the storage of both synthetic and natural
water samples held in several different
types of containers at various pH's over
periods of time. Their study concluded
that acidification to a pH <1.5 and storage
in linear polyethylene containers were most
effective in avoiding the loss of metals
from solution. Struepler,5 Smith,5 and
Shendrikar et al.1* are among the many in-
vestigators who have studied the adsorption
characteristics of trace metals on the
surfaces of various types of container
walls.
These studies all were conducted with
either natural surface waters or synthetic
water samples whose constituent concentra-
tions were in the mg/1 range. Samples
collected in this study, however, were taken
from monitoring wells at industrial or
municipal refuse sites where the constituent
concentrations were considerably greater.
In addition, these groundwater samples were
being held under anaerobic conditions. This
created additional preservation problems
that may not be encountered with surface
waters.
Another sampling decision was that oi
which pore sizes of filter membranes should
be used. The use of 0.45 ym pore size
cellulose acetate membranes has been widely
accepted as the standard for sample pre-
paration. However, in the case of water
samples with high turbidity, the 0.45 urn
membrane may be impractical for on-site
filtration because of the slow filtration
rate. A larger pore-sized membrane may be
more applicable for these samples, but may
also allow for the passage of clay-sized
materials through the membrane. Acidifica-
tion of these samples could result in the
solubilization of particulate matter or
leaching of trace metals from the clays
which would produce high constituent con-
centrations that are nonrepresentative.
Kennedy et al.1 also showed that certain
clay sized particulate material can pass
through a 0.45 ym membrane.
While pumping at each site, samples
were taken from the initial water in storage
and at %, 1, 1^, 2, 4, 6, 8, and 10 well
volume intervals. Aliquots for analysis
were taken from each of the well volumes
or partial well volume samples collected.
While on site, pH, Eh (oxidation-reduction
potential), specific conductance, and
alkalinity measurements were made on the
unfiltered samples immediately after they
were collected. The aliquots were then
subdivided into three portions and filtered
through either a 3.0 ym, 0.45 ym, or a
0.22 ym pore sized Millipore@membrane
using Satorius@plastic filter holders.
[Mention of trade names should not be con-
sidered as an implied or direct endorse-
ment of the products mentioned. They are
given for the convenience of the reader and
do not indicate preferential treatment by
the Illinois State Water or Geological
Surveys.] Nitrogen was used as a pressure
source to speed filtration and avoid de-
gassing of the sample as may occur if a
vacuum is applied. Each filtered subsample
34
-------
was then further subdivided into samples
for analysis of cations and anions. The
samples for cation analysis were preserved
immediately after being filtered by addi-
tion of concentrated nitric acid to a pH
<1.5. The samples for anion and nitrogen
analysis were stored in an ice chest in the
field and later at 4°C in the laboratory.
At Site 5 additional sample volumes were
filtered through the 0.45 ym membrane for
total organic carbon (TOC), nitrate and
ammonia analysis. All of these samples
were stored in linear polyethylene bottles.
Samples for TOC analyses were preserved by
addition of sulfuric acid.
The Jarrell-Ash Atom Comp 750 induc-
tively coupled plasma unit was employed for
sample analysis of Al, As, B, Ca, Cd, Cr,
Fe, Mg, Mn, Pb, Se, and Zn. Sodium and K
were determined by atomic adsorption.
Chloride, fluoride, pH, and Eh were analyzed
by ion selective electrode and sulfate was
determined colorimetrically.
CHEMICAL RESULTS
Approximately 3500 samples and over
42,000 chemical analyses have been collected
and analyzed to date. The data from these
analyses are being evaluated in an effort
to explain the variations in chemical con-
centrations with pumping or time, different
pumping rates, pumping mechanisms, and dif-
ferent filter pore sizes. Additional data
also are being collected to determine the
effects of storage on sample integrity.
For this paper discussion will be
limited to the effects of time of pumping,
pumping rates, pumping mechanisms, and
filter pore size on the chemical results of
the collected samples.
Probably the single most important
factor affecting the chemistry of ground-
water is pH. Therefore, anything that
alters the pH of the groundwater samples
may alter the chemistry of its constituents.
During sampling, extreme care was taken to
obtain the most reliable pH measurements
possible.
On-site pH measurements at all six
sites using the four pumping mechanisms
have been analyzed. The results illustrate
two possible effects of the type of pumping
mechanism on sample character. For com-
parison, all pH data for each site were
normalized to the "stabilized" pH values
obtained for 6-, 8-, and 10-well volume
samples collected by the mechanical pump
at that site. The "normalized" pH values
therefore represent the relative differences
in pH between mechanical pumping and the
other pumping mechanisms. This was done
to give a standard basis of comparison for
all six sites. These normalized pH values
are plotted in Figure 3 for each sample
collected.
Data from samples collected by mechan-
ical pump at all six sites are grouped along
a normalized pH value of 100. The fluctua-
tion in pH during the early stages of pump-
ing (0 through 4 volumes) probably was due
to the mixture of water stored in the well
casing and aquifer water pumped into the
well during this portion of sampling. As
a higher percentage of aquifer water is
pumped (volumes 6 through 10) the pH values
begin to stabilize. The pH values converge
and are thought to reflect the "true" pH of
water in the aquifer.
Two other distinct groupings of data
also are shown in Figure 3. During the
pumping of samples by air-lift and nitrogen-
lift, two different discharge pipe sizes
were used, a 1.27-cm (^-inch) and 0.952-cm
(3/8-inch). Data obtained from samples
collected by either air or nitrogen, with
0.952-cm (3/8-inch) discharge pipe, appear
to stabilize at a normalized pH value of
EXPLANATION
A Mechanical pump
Air-lift (y pipe) (127cm)
* Nitrogen-lift (V pipe)
D Air-lift (3/8" pipe)(0952
• Nitrogen-lift {3/8" pipe)
X Bailer
46
VOLUMES OF WATER PUMPED
Figure 3. Effects of Pumping Mechanisms on pH Values
35
-------
110. Data obtained from samples collected
using either air or nitrogen with the 1.27-
cm (%-inch) discharge pipe appear to sta-
bilize at a normalized pH value of 116. In
both groupings, larger fluctuations in pH
values during the early stages of pumping
were noted because of the mixing of stored
and aquifer water, as with the mechanically
pumped samples.
However, marked pH value changes were
noted throughout the duration of sampling.
It is postulated that the bubbling air or
nitrogen passing through the water is
stripping dissolved CC>2 that was in excess
of atmospheric pressure from the water.
This results in increased pH values over
those obtained by peristaltic pump. Smaller
changes were noted in the samples collected
using the 0.952-cm (3/8-inch) discharge
pipe. This was thought to occur because
less air or nitrogen is needed to pump the
same quantity of water than with the 1.27-
cm (h-inch) discharge pipe. The gas to
water ratios for the samples collected
using the 0.952-cm (3/8-inch) diameter
pipe varied from about 3.4 to 9.5, with the
higher ratios generally resulting in larger
pH changes. The gas to water ratios for
the samples collected using the 1.27-cm
(%-inch) diameter pipe were much higher
(30.0 to 40.0) and appear to have effec-
tively stripped the samples of most of
their excess dissolved C02, resulting in
pH values of 8.2 to 8.3. This can be com-
pared with pH values of 6.9 to 7.0 when
the peristaltic pump was used to collect
samples.
Data from samples collected at Site 5
using an air-lift pumping mechanism dramat-
ically illustrate the effect of pipe size
on pH as shown in Figure 3. During the
first stage of pumping (volumes 0 through
2), a 1.27-cm (%-inch) discharge pipe was
used. To increase the pumping rate, the
1.27-cm (*5-inch) pipe was withdrawn and a
0.952-cm (3/8-inch) pipe inserted between
volumes 2 and 4. The immediate large drop
in pH shown in figure 3 confirms the
important effect of the gas to water ratios
on the chemistry of water samples collected
with these types of devices.
Results of pH measurements at the six
study sites suggest that air-lift or
nitrogen-lift pumping mechanisms do alter
the chemistry of the collected sample.
These effects could be minimized by proper
design and operation of the pumping device
for collecting water samples in certain
physical settings (depths, water levels,
and pumping lifts). However, since the
changes in pH are significant in terms of
some chemical constituents, these types of
pumping mechanisms may not be desirable for
collecting water samples in many cases.
The metals (Fe, Zn, Cu, Pb, and Cd) appear
to be particularly affected by the use of
these pumping mechanisms, as any changes in
pH are likely to affect the concentrations
of metals in solution. The alkaline earth
materials on the other hand appear to be
somewhat insensitive to the effects of pH
change induced by the different pumping
mechanisms employed.
Figure 4 illustrates the change in
iron concentrations with volumes of water
pumped (time) at Site 5 for all four pump-
ing mechanisms. The differences in iron
content noted between samples collected
mechanically and by nitrogen-lift princi-
pally are the result of pH change. The
increased pH values of the samples col-
lected by the nitrogen-lift mechanism
probably is causing some soluable iron
compounds to precipitate and fall out in
the well casing or be filtered from the
collected samples. More extreme effects
are noted for samples collected by air.
24
20
16
W
- 12
SITE 5
A Mechanical pump
O Air-lift pump
• Nitrogen-lift pump
X Bailing
46g
VOLUMES OF WATER PUMPED
Figure 4. Effects of Pumping Mechanisms on Iron
Concentrations at Site 5 as a Function of Well
Volumes Pumped
10
36
-------
Additional soluable iron was apparently
oxidized and similarly removed, thus re-
sulting in even lower iron concentrations
than found in samples collected with
nitrogen. The small differences in iron
concentrations of samples collected mechan-
ically and by bailing may be the result of
seasonal fluctuations in iron concentrations.
Similar trends also were noted for
other metals (Zn, Cu, Pb, and Cd). The
effects on iron were more dramatic at the
sites studied because of the relatively
high concentrations and the anaerobic
nature of landfill leachates and the anaer-
obic digestion ponds at the hog processing
plant (Site 5).
Comparison of chemical analyses for
the samples filtered through the three
pore sizes of membranes indicated that,
for certain constituents, the concentration
was dependent upon the pore size employed.
Figure 5 illustrates the effect of filter
pore size on iron content of samples col-
lected at Site 6 using the nitrogen-lift
pumping mechanism. In all cases the higher
iron concentrations were found in those
samples filtered through the 3.0 ym pore
size filters. The concentrations of the
samples filtered through the 0.45 ym and
0.22 ym membranes were comparable to each
other. However, there was a trend toward
SITE 6
NITROGEN-LIFT SAMPLES
0.22
0.45 urn
I
I
4 6
VOLUMES OF WATER PUMPED
10
Figure 5. Effects of Filter Pore Size on Iron Concentrations
of Samples Collected at Site 6 using a Nitrogen-lift
Pumping Mechanism
slightly lower values for samples filtered
through the 0.22 ym membranes. This trend
would indicate that some fine-sized mate-
rial was passing through the 0.45 ym mem-
brane. In comparing the concentrations of
other metals (Zn, Cd, Cu, Mn, and Pb) there
also was a trend towards decreasing con-
centrations with decreasing pore size.
However, the concentrations of these metals
are near the detection limit of the ana-
lytical procedures employed, making positive
conclusions difficult.
Because of the complexity of the
environmental systems being studied, it is
difficult to determine the exact cause of
the filtration effect. It is possible that
the increased concentrations in the fil-
trates were due to the passage of fine clay
particles, or possibly due to the passage
of precipitates that formed prior to fil-
tration and were later dissolved when the
samples were acidified for storage.
SUMMARY
Data gathered during the first year of
this project has demonstrated that the
length of pumping, type of pumping mechanism
used, and the size of membranes used to
filter samples all affect the chemical
composition of water collected from moni-
toring wells. The relative magnitude of
their effects appears to be site specific
and related to: 1) the yield potential of
the sampled well; 2) the depth, diameter,
and water level of the monitoring well;
3) the rate at which the well is pumped;
4) the general chemical character of the
water being sampled; and 5) the specific
elements that are to be monitored or ana-
lyzed.
During the next year of the project
an attempt will be made to quantify as many
of these variables as possible. Additional
sampling and analyses will be undertaken to
determine the effects of storing samples
for various periods of time before they are
filtered and analyzed. Additional sampling
runs also may be undertaken to test hypoth-
eses developed during the analyses of data
collected to date.
Recommendations relative to specific
sampling procedures and preservation tech-
niques for certain types of wells for
specific chemical constituents also will
be made in the future.
37
-------
ACKNOWLEDGMENTS 3.
This project is partially supported by
the Solid and Hazardous Waste Research
Division of the Federal Environmental
Protection Agency (Grant R806304-01-1) 4.
under the general guidance of Donald E.
Banning and Marion R. Scalf, Project
Officers. Special thanks are given to
Michael O'Hearn and William Bogner,
Assistant Hydrologists, Illinois State
Water Survey, for their assistance in run-
ning pumping tests and collecting samples. 5.
Thanks also is given to Kenneth E. Smith,
Associate Chemist, Illinois Natural History
Survey, for permitting and supervising the
use of the Jarrell-Ash Plasma Atom Comp 750.
REFERENCES 6.
1. Kennedy, Vance C., Gary W. Zellweger,
and Blair F. Jones. 1974. Filter pore
size effects on the analysis of Al, Fe,
Mn, and Ti, in water. Water Resources
Research, 10(4): 785-790.
7.
2. Papadopulos, Istavros S. and Hilton H.
Cooper. 1967. Drawdown in a well of
large diameter. Water Resources Re-
search (First quarter), 3(1): 241-244.
Rattonetti, Anthony. 1976. Stability
of Metal Ions in Aqueous Environmental
Samples. National Bureau of Standards
Special Publication 422, p. 633-648.
Shendrikar, A. D., V. Dharmarajan, H.
Walter-Merrick, and P. W. West. 1976.
Adsorption characteristics of traces
of barium, beryllium, cadmium, manganese,
lead, and zinc on selected surfaces.
Analytica Chim. Acta, 84: 409-417.
Smith. A. E. 1973. A study of the
variation of pH on the solubility and
stability of some metal ions at low
concentrations in aqueous solution.
Part 1. Analyst, 98: 65-68.
Struempler, Arthur W. 1973. Adsorp-
tion characteristics of silver, lead,
cadmium, zinc, and nickel on borosili-
cate glad, polyethylene, and poly-
propylene container surfaces. Anal.
Chem., 45(13): 2251-2254.
Subramanian, K. S., C. L. Chakrabarti,
J. E. Sueiras, and I. S. Maines. Pre-
servation of some trace metals in
samples of natural waters. Anal. Chem.,
50(3): 444-448.
38
-------
THE UTILITY OF EXTRACTION PROCEDURES AND TOXICITY TESTING WITH SOLID WASTES
C. W. Francis, Environmental Sciences Division
M. P. Maskarinec, Analytical Chemistry Division
J. L. Epler, Biology Division
D. K. Brown, Environmental Sciences Division
Oak Ridge National Laboratory
Oak Ridge, Tennessee 37830
ABSTRACT
Solid wastes from diverse origins were extracted using the proposed EPA extraction
procedure (EP). These extracts were characterized chemically and tested for toxicity
using bioassays for aquatic, phytotoxic, and mutagenic characteristics. None of the EP
extracts appeared to be mutagenic, however, the negative responses may be due to the
inability of the acidic, aqueous EP to extract nonpolar organics and/or possible sorption
of these compounds on the type of filter used in the EP. Several of the extracts were
toxic in the phytotoxicity and/or aquatic test systems. Acetic acid in the EP interfered
in the phytotoxicity and aquatic toxicity bioassays, making interpretation difficult.
INTRODUCTION
To describe the potential environmen-
tal hazards of a solid waste, under Sub-
title C of the Resource Conservation and
Recovery (RCRA) of 1976 (PL 94-580), an
extraction procedure (EP) was proposed by
EPA in the December 18, 1978, Federal
Register (43 FR 58956). The EP was estab-
lished as an integral part of a screening
protocol for the characterization of solid
wastes as 'toxic' or 'nontoxic'. Concen-
trations of eight elements in the extract
(10 times the EPA Primary Drinking Water
Standard of As, Ba, Cd, Cr, Pb, Hg, Se, and
Ag) are a primary basis for classification.
Ideally, the protocol should be applicable
to a diverse spectrum of wastes covering a
wide range in physical and chemical charac-
teristics. It should estimate the quantity
of potential contaminants released as a
function of time and disposal conditions.
To be widely accepted, it should be rela-
tively simple, rapid, inexpensive, and
easily interpreted.
To assist the EPA in developing these
protocols for identifying solid wastes that
may pose a potential hazard to human health
and environment, the Oak Ridge National
Laboratory (ORNL) has conducted studies on
the EP extract from seventeen solid wastes
of diverse origin. Using these extracts,
analytical procedures have been developed
to chemically characterize and separate the
organic and inorganic constituents in the
waste. The EP extracts have been used in
bioassays developed to test possible
toxicity to aquatic organisms and plants.
A battery of three bioassays diagnostic for
detecting changes in genetic activity were
used to serve as an indicator of chronic
hazards relative to possible mutagenic and
carcinogenic effects of the EP extract.
The purpose of this manuscript is to
describe the development and rationale for
some of the procedures used to test the
toxicity of the extracts in the various
biological test systems and to evaluate the
applicability and compatability of the EP
for such a testing protocol. None of the
EP extracts of the seventeen solid wastes
showed detectable mutagenic activity;
although activity was found in a direct
assay of an As-contaminated groundwater.
To assess the relevance of the EP, samples
of solid waste were taken from the landfill
believed to be the source of the As in the
groundwater sample, and extracted according
to the EP protocol.
39
-------
MATERIALS AND METHODS
A truly multidisciplinary team was
necessary to evaluate properly the toxicity
of extracts from solid wastes. It included
environmental scientists, responsible for
assessing the relevance of the extraction
procedure and testing the effects of the
extract on aquatic and terrestrial
organisms, analytical chemists, responsible
for detection, quantification, separation,
and identification of the organic and in-
organic toxic constituents, as well as a
team of microbial biologists and geneticists
trained in the development and interpreta-
tion of bioassays to detect indicators of
mutagenicity and carcinogenicity. Each
group has directed considerable effort into
developng the necessary methods to accom-
plish their tasks. Thus, the following will
be a short description of the protocol used
by each; detailed description is available
in the final report "Toxicity of Leachates"
to be issued soon as a joint ORNL and EPA
publication (Epler, et. al.3).
Extraction Procedure
The solid wastes were extracted using
the proposed EP published in the December
18, 1978, Federal Register (43 FR 58956).
Several vessels [plexiglass, Type 316 stain-
less steel, and glass) with certain modifi-
cations in engineering designs were used.
Data recorded during the EP consisted of:
1) initial pH; 2) final pH; 3) amount of
acetic acid added; and 4) electrical
conductivity of the final extract. The
final solid and liquid phases were separated
by vacuum filtration (Millipore 0.45 ym pore
size, type HA filter). Aliquots of the
extract were distributed to the appropriate
groups for chemical characterization and
biotesting (aquatic and phytotoxic effects).
Chemical Characterization
Chemical characterization included
identifying and quantifying the chemical
constituents in the EP extracts and con-
centrating the organic components in the
EP extract for mutagenesis biotesting.
A macroreticular resin sorption technique
(XAD-2 resin provided a direct 100-fold
increase) was used to concentrate the
organic constituents from the EP extracts.
These were separated into nonpolar and
polar fractions using a Florisil column.
The nonpolar fraction was further divided
on an alumina column using various eluting
solutions (hexane, 6:1 hexane:benzene, 2:1
hexane:benzene and acetone) into a PCB-
pesticide-monoaromatic-diaromatic-parafin
fraction, a diaromatic fraction, a poly-
aromatic fraction, and a heteroaromatic
fraction. The EP organic concentrates
tested for mutagenesis were collected on
the XAD-2 resin and eluted using cyclo-
hexane. Atomic absorption spectroscopy
(flame!ess graphite furnace) was used to
determine the inorganic constituents in the
EP extract.
As an alternative to the EP, a steam
distillation procedure was developed to
extract greater quantities of organics from
selected solid wastes than those observed
with the EP. The solid waste (100 g) in
250 ml of triply distilled water is refluxed
so that the steam and an organic solvent
(immiscible in water), which is heated in
another vessel, co-condense; the aqueous and
non-aqueous phases separate and return to
their respective vessels. In this manner,
organics in the solid wastes entrained in
the steam can be separated from the aqueous
phase without direct extraction of the waste
with an organic solvent, and without filtra-
tion.
Aquatic Toxicity
The test organism, Daphnia magna, was
used in acute (48 h) and chronic testing
(28 d) procedures. Results of the acute
tests, expressed in LCso's, represented
the concentration of the EP extract that
was lethal to 50% of the D_. magna during
the 48 hours of exposure. Chronic testing
was carried out only at EP extract dilu-
tions of 1:100 and 1:1000. The extracts
were neutralized to pH 7 before testing.
To evaluate the influence of acetic acid
added in the EP, controls containing
equivalent amounts of acetic acid and
neutralized to 7.0 were used in conjunction
with well-water controls.
Phytotoxicity
Short-term root elongation tests (48
and 72 h) using radish and sorghum and
longer term seedling tests (2 and 3 weeks)
using wheat and soybeans (response in dry
weight of root and shoot) were used to
evaluate the toxicity of EP extracts to
plants. Extracts were diluted to various
concentrations with distilled water for the
root elongation tests and with a balanced
nutrient solution for the seedling tests.
40
-------
Mutagenicity
Three bioassays diagnostic for genetic
activity were used to test the EP extracts
and XAD-2 concentrates of the EP extracts
for mutagenecity. They were: 1) the
Salmonel1 a/Microsome, 2) the Saccharomyces
canr/his* dual assay, and 3) the Repair
Assay (Ames, McCann, and Yamasakil; McCann
et. al.4; Larimer, Ramey, Lijinsky, and
Epler5).
RESULTS AND DISCUSSION
Extraction Procedure
Preliminary extractions were carried
out in vessels made of plexiglass and Type
316 stMnless steel; both designed as
described in Federal Register, Vol. 43,
No. 243. The plexiglass vessel was elimi-
nated due to possible sorption of organics
on the vessel walls as well as its incom-
patibility in testing solid wastes con-
taining high concentrations of organic
solvents. The stainless steel unit design
was found to be unsatisfactory because of
binding between the stirring blades and
solid particles that caused overloading
and eventual malfunction of the stirring
motor. A modified version of the extrac-
tion equipment was designed and fabricated
at ORNL. The new design integrated the
type 316 stainless steel vessel with a
high-torque, low rpm stirring motor with
the stirring rod centered on a conical
bearing at the bottom of the vessel.
Sample grinding was the main problem
encountered with this design. A stirrer
with Teflon blades proved to be a satis-
factory alternative.
Contamination by the Stainless Steel
Vessel — Our data indicate that use of a
stainless steel vessel likely contaminates
EP extracts with Cr and Ni. For instance,
distilled water blanks acidified to pH 3.5
with nitric acid contained significantly
more Cd, Cr, Cu, Pb, Ni, and Pb than the
same water elevated to pH 10 with NfyOH.
Acid washed sand (solid:water ratio 1:20)
EP extracts contained significantly more Cr
and Ni than distilled water blanks (Fig. 1),
indicating that the abrasive action of
course particles grinding against the vessel
walls will likely enhance dissolution of Cr
and Ni from the stainless steel vessel.
Extraction Variability - Each of the 17
solid wastes were extracted four times with
ORNL-OWG 79-14339
1 FROMMILLIPORE MILLI-0 WATER
PURIFICATION SEPTEM
2 FROM WATER COLLECTING FLASK
3 AFTER AUTOCLAVING WATER
4 FROM EXTRACTOR-INITIAL
5 FROM EXTRACTOR -1 HOUR
6 FROM EXTRACTOR-3HOURS
7 FROM EXTRACTOR-7 HOURS
8 FROM EXTRACTOR-24 HOURS
9 AFTER FILTERING
2345678
WATER COLLECTION POINTS
Figure 1. Higher concentrations of Cr and
Ni were observed in EP extracts using acid
washed sand than distilled water in a stain-
less steel extraction vessel with Teflon
stirrer.
• f)u?*
-------
the £P and analyzed separately for Cr, Ni,
and Ca. The coefficient of variation [CV =
(s/x) (100)] ranged from 7 to 83, 10 to 94,
and 2 to 29% for Cr, Ni, and Ca, respec-
tively. There did not appear to be a
relationship between extraction variability
and waste type. The variability associated
with the analysis of Cr, Ni, and Ca at the
reported concentrations needs to be
verified. Considering the major constraints
involved in interpreting what is a toxic
concentration, (Is 10 times the primary
drinking water standard really toxic in
the environment?) the errors associated
with the extraction variability appear to
be a minor problem at CV's less than 100%.
Toxicity of Extracts
'•foxicity" of an EP extract is defined
in this manuscript as 1) any extract in
which the concentration of As, Ba, Cd, Cr,
Pb, Hg, Se, or Ag was greater than ten
times the EPA Primary Drinking Water Stan-
dard or 2) any extract that showed a
positive test for toxicity in any of the
bioassays (aquatic, phytotoxicity, or muta-
genicity). A summary of the results is
presented in Table 1. Omitted from this
Table is the analysis for organic compounds.
The following organics were not detected
(limit of detection) in any of the extracts:
PAH's (2 mg/n), volatile organics (0.1
mg/i.), and pesticides (0.1 to 1 mg/£).
Quantities of PCB's detected ranged from
0.2 to 1.0 mgA but they probably repre-
sented contamination rather than originating
from the waste. A concentration of 320 mg/£
of o-nitroaniline was measured in the As-
groundwater sample and represented the only
organic compound found in any of the samples
that could be considered "toxic" based on
RCRA classification.
Only four of the 17 solid wastes would
not comply with the currently proposed RCRA
regulations (43 FR 58950) on toxicity.
These wastes (No.'s 2, 3, 8, and 9) had
excessive metal concentrations in the EP
extract. The others that showed positive
signs of toxicity by the bioassay tests
would only be considered toxic or hazardous
if adequate tests could be developed that
would reflect accurately the characteristics
of that waste. One of the major objectives
in this work was to identify the constraints
in bioassay protocols using EP extracts so
that reliable bioassays can be developed
TABLE 1. TOXICITY OF EP EXTRACTS FROM SOLID HASTES
Bioassay
Solid Waste
Metals Aquatic Phytotoxicity Mutagenicity
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
Soybean Process Cake
Metal Processing Waste
Plater's Waste
Raw Shale
Retorted Shale
Dye Waste
Textile Waste
Municipal Sewage Sludge
Power Plant No. 1 - Fly Ash
Cr
Cd, Cr
Cd
Cd
Power Plant No. 2
Fluidized Bed Residue
Gasification Bottom Ash No. 1
Gasification Bottom Ash No. 2
Gasification Bottom Ash No. 3
As-Contaminated Groundwater
Bottom Ash
Scrubber Sludge
Treated Scrubber Sludge-
Fly Ash
As, Cd
+ = toxicity; - = nontoxicity.
Is not an EP extract but a groundwater sample furnished by EPA and concentrated on an
XAD-2 column before testing for mutagenicity.
42
-------
and used to effectively determine the
hazards of solid wastes.
None of the EP extracts appear to be
mutagenic based on these bioassays. Short-
term assays often show negative results
with certain heavy metals and classes of
organic compounds known to be biologically
toxic. However, in this case, the primary
reason for the observed negative responses
in the mutagenicity bioassays may be the
inability of the EP to extract organics
from the solid wastes. The acidic aqueous
character of the EP would not favor the
removal of nonpolar organics, particularly
PAH's which are highly water insoluble and
sorb tenaciously to solid surfaces in the
suspension. Likewise, some carbon-14 label-
ed PAH's are known to be sorbed (>90%) on
the filters used in the separation of the
extract (Brown et. al.^). Thus, final
judgment on the mutagenicity of these solid
waste leachates should be deferred until
scientific evidence indicates that the
present EP or an alternative EP is free
of artifacts peculiar to the EP itself.
Environmental, biogeochemical conditions
expected in groundwater from a landfill
containing these wastes should be selected
for in future modifications of the EP
The phytotoxicity tests showed positive
signs of toxicity for 12 of the 17 EP
extracts (Table 1). The waste was clas-
sified as toxic if statistically significant
(P<0.05) reduction in growth (either in the
root elongation or seedling growth tests)
was observed with any of the plant species
(radish, sorghum, wheat, or soybean) at
any concentration of the EP extract (0.5
to 100%). A major and very serious
constraint in interpretation of the phyto-
toxicity tests is the toxicity of acetic
acid to plant growth. Growth reduction in
seedling and root elongation tests were
noted at 2.5 x 10'3 M acetic acid. The EP
protocol requires adjustment to pH 5.0
using acetic acid during the 24 h extraction
period (maximum amount of acetic acid
allowed, 10~' M). Thus, this phytotoxity
bioassay reflects not only the inherent
toxicity of the waste but in many instances
the quantity of acetic acid required to
maintain a pH of 5.0 during the EP. It
appears that before phytotoxicity bioassays
can be implemented as a reliable protocol
to test toxicity of EP extracts, problems
inherent in using acetic acid will have to
be overcome.
Acetic acid in the EP also interferes
in the interpretation of the aquatic bio-
assay. In this case, the acetic acid can
reduce as well as enhance toxic responses.
For example, acetate is an excellent carbon
source for bacterial growth which in turn
is a food supply for [L magna, the aquatic
test organism. Thus, acetic acid additions
in the EP can stimulate as well as reduce
the growth and reproduction of JD- magna.
As with the phytotoxicity assays, problems
associated with using acetic acid must be
overcome before reliability can be placed
on the aquatic assays.
As-Contaminated Groundwater
The As-contaminated groundwater
appeared to be very toxic judging from the
concentrations in the inorganic and organic
analyses as well as from results of all the
bioassays. To assess the significance of
an EP, two solid waste samples were re-
trieved from this landfill along with
additional groundwater samples during the
fall of 1979. These samples of solid waste
taken from the landfill are not intended to
represent a scientifically well-defined
sampling protocol describing the physico-
chemical character of the waste or its
distribution in the landfill, but merely a
first order estimate of the waste in its
present form. Groundwater was sampled from
three wells at increasing distances from
the landfill.
Three extracting procedures were used
to compare the dissolution of nitroaniline
from the two solid waste samples (Table 2).
The acidic, aqueous EP extracted as much
nitroaniline as either the soxhlet or
steam distillation procedures. Nitro-
aniline, though, is a highly water soluble
organic compound and, as such, represents
a best possible case for the EP. Earlier
comparisons have indicated the EP to be
considerably less effective than these two
procedures in the extraction of high-
molecular weight, low water soluble organic
compounds. The low recovery of As in the
aqueous phase of the steam distillate likely
reflects the loss of organic arsenicals to
the methylene chloride. These samples were
expected to contain various phenylarsenic
acids. The concentration of nitroaniline
in the EP extract was significantly lower
than the concentrations observed in the
well water. Well water concentrations
decreased with increasing distance from
43
-------
TABLE 2. COMPARISON OF SOLID WASTE EXTRACTION PROCEDURES AND CONCENTRATIONS
IN THE EP EXTRACT RELATIVE TO THAT IN GROUNDWATER NEAR LANDFILL
Extraction
Procedure
Soxhlet
Steam Distillation
EP
Nitroaniline
As
Quantity Extracted
mg/g
0.96 + 0.17
0.90 + 0.04
1.04 + 0.07
0.96 + 0.00
3.06 + 1.62
Groundwater and
EP
Concentration
yg/ml
Well No.
Well No.
Well No.
EP
1
2
3
1320
680
440
( 51 + 1 >
151
157
92
153 + 81
Methylene chloride used as the solvent in the soxhlet and as the organic solvent in the
steam distillation procedure; nitroaniline determined in organic phase of steam distilla-
tion procedure while As was determined in the aqueous phase. Averages and standard
deviation from two separate solid waste samples.
the landfill, approximately 40, 60, and 80
meters for No. 1, 2, and 3 wells, respec-
tively. Arsenic concentration in the well
water did not vary as much with distance
and was close to that observed in the EP
extracts. With the limited data available,
further comparisons regarding the signifi-
cance of these concentrations with that
observed in the EP extracts are simply
speculation. However, it does represent
the first step in assessing the usefulness
of an EP.
CONCLUSIONS AND RECOMMENDATIONS
.
^ Acetic acid interferes in bioassays
f°r aquatic and phytotoxic effects.
An extractor vessel made of glass or
Teflon should be used in the EP ratfier~than
stainless steel to avoid possible heavy
metal contamination (Cd, Ni, and Cr).
To eliminate the sorption of PAH's or
other highly water-insoluble organics on
the filter membrane, the suggested filtering
procedure in the proposed EPA EP should be
modified or replaced with an alternative
method.
A comprehensive evaluation should be
made correlating the properties of the EPA
proposed EP extract (or an alternative EP)
and authentic landfill leachates.
ACKNOWLEDGMENTS
The research described was sponsored
jointly by the U.S. Environmental Protection
Agency (IAG 78-DX-0372) and the Office of
Health and Environmental Research, U.S.
Department of Energy under contract W-7405-
eng-26 with the Union Carbide Corporation.
Appreciation is expressed to staff of the
Analytical Chemistry Division, Oak Ridge
National Laboratory, M. Ferguson,
E. Burnett, and W. Griest; the Biology
Division, Oak Ridge National Laboratory,
F. Larimer, and T. Rao; and the Environ-
mental Sciences Division, Oak Ridge National
Laboratory, N. Edwards and B. Parkhurst for
their research contributions to the "Toxic-
ity of Leachates" project which was the
basis for material presented in this paper.
44
-------
REFERENCES
1. Ames, B. N., J. McCann, and E. Yamasaki,
1975. Methods for detecting carcinogens
and mutagens with the Salmonella/mam-
mal ian-microsome mutagenicity test.
Mutation Res. 31:347-364.
2. Brown, D. K., P. Lowry, S. E. Lindberg,
S. E. Herbes, and J. M. Coe. Sorption
of PAH's from aqueous solutions filtered
through membrane and glass fiber fil-
ters. (In preparation).
3. Epler, J. L., F. W. Larimer, T. K. Rao,
E. M. Burnett, W. H. Griest, M. R.
Guerin, M. P. Maskarinec, D. K. Brown,
N. T. Edwards, C. W. Gehrs, R. E.
Millemann, B. R. Parkhurst, B. M. Ross-
Todd, D. S. Shriner, H. W. Wilson, Jr.
Toxicity of leachates. Joint report
ORNL and EPA. (In press).
4. McCann, J., E. Choi, E. Yamasaki, and
B. N. Ames, 1975. Detection of carcino-
gens as mutagens in the Salmonella/
microsome test: Assay of 300 chemicals.
Proc. Natl. Acad. Sci. USA 72:5135-5139.
5. Larimer, F. W., D. W. Ramey, W.
Lijinsky, and J. L. Epler, 1978. Muta-
genicity of methylated N-nitrosopiperi-
dines in Saccharomyces cerevisiae.
Mutation Res. 57:155-161.
45
-------
CHEMICAL SPECIATION OF FLUE GAS DESULFURIZATION WASTES
James C. S. Lu
Calscience Research, Inc.
7261 Murdy Circle
Huntington Beach, California 92647
ABSTRACT
In this study, a thermodynamic equilibrium model for evaluating the chemical speciation of
constituents in FGD wastes was evaluated. This model was also used to characterize the
distribution, migration trends, stability fields, and concentration levels of constituents
in FGD wastes. The suitability and accuracy of the model was verified by comparing calcu-
lated results to analytical data and certain scientific considerations. The model was
then utilized to determine the effects of various operational or chemical changes in the
FGD system or sludge treatment system on the concentration and chemical form of the im-
purities of interest.
INTRODUCTION
Conventional environmental impact
assessment of flue gas desulfurization (FGD)
sludge disoosal includes chemical analysis
and identification of the total concentra-
tions of constituents in the sludge and
its leachate. However, environmental and
public health effects of FGD waste disposal
depends upon which chemical forms or species
of the constituents are released to sur-
rounding waters, and not necessarily on
their total concentration.
Thermodynamic modeling provides a
feasible means of obtaining contaminant
species information in FGD wastes. A ther-
modynamic model can also be used to predict
the migration trends of the constituents as
the FGD waste ages, as well as to estimate
the final constituent concentration in the
FGD leachate (aged wastewater) without
conducting expensive field monitoring.
Modeling is also useful for predicting the
effects of operational and chemical changes
in the FGD wastes.
Many available techniques can be used
to construct and interpret a chemical
thermodynamic model (Brinkley2, Brinkley^,
Butler4, Crerar5, Crerar^, Feldman',
HelgesonS, Helgeson^, Helgenson'O,
Helgesonll, Karpov12, Lu14, Morel I6,
Naphtalil7, and White2
-------
actual mathematical equilibrium model in-
volves a series of simultaneous equations
which describe the various interactions
among components of the system. In order
to solve these equations simultaneously,
the information on metal and ligand species,
overall formation constants, solubility
products (and/or Henry's constants), and
activity coefficients must be compiled from
the literature. A computer solution is
necessary, as the expanded equations number
in the hundreds. The resultant nonlinear
equations are solved by Newton-Raphson
iteration.(see Table 1)
Because the chemical composition of
FGD waste can vary over an extremely wide
range, this study focused on speciation at
the lowest levels (ionic strength (I) =
0.05) and the highest levels (I = 0.8).
All possible distributions of species are
expected to be within this range. The
general concentration ranges of constitu-
ents in FGD sludges and leachates are
listed in Table 2.
SPECIATION OF CONSTITUENTS IN FGD WASTES
Fresh FGD Leachates
Modeling speciation of fresh FGD
leachates require two simplifying assump-
tions: (1) equilibrium conditions among
the soluble species can easily be reached;
and (2) the rates of nucleation and dis-
solution of the solid species are extremely
low. Thermodynamic modeling of fresh FGD
leachates can therefore be performed as if
no new solids were formed or dissolved in
the system. Included in the model were
20 important metals,13 important ligands
and 155 possible complexes.
Calculation results indicate that the
major ions (i.e., calcium, magnesium, po-
tassium, and sodium) exist mainly as free
ions. However, trace metals are complexed
considerably in fresh FGD leachates. The
speciation of calcium and cadmium at both
low and high ionic strengths (I = 0.05 and
I = 0.8) is shown in Figures 1 to 4 as
examples. In general, chloride complexes
may become the predominant species for cad-
mium, copper, lead, mercury, and zinc;
borate complexes may become the predominant
species for copper and lead; sulfite com-
plexes may become the predominant species
for cadmium and iron; and hydroxide com-
plexes may become the predominant species
for the trivalent metals, e.g., chromium
and iron, mercury and zinc. In fresh FGD
leachates, arsenic and selenium exist pri-
marily as arsenate and selenite species.
The predominance of a given species can be
affected significantly by the pH level of
leachate. The ionic strength (or, more
specifically, the soluble levels of the re-
lated ligands) also plays an important role
in the speciation of most constituents.
Aged FGD Wastes
The speciation of constituents in the
solid and soluble phases of aged FGD wastes
was computed with the assumption that the
equilibrium condition among all the soluble
and solid species had been reached. Due
to the long contact period, it is generally
quite possible that equilibrium conditions
between solid and liquid phases can be
reached in the aged FGD wastes. The calcu-
lated results for two elements (calcium
and cadmium at I = 0.05)are shown in Figures
5 to 8 as examples. Results of other ele-
ments at selected conditions are summarized
in Table 3.
Results show that sulfur dioxide re-
moved from the flue gas reacts to form
CaS04-2H20(s) and CaSOs-y^Ots) in the FGD
sludge. In the aged sludge, carbonate
solids may become the predominant species
for cadmium, calcium (when pH is greater
than 7), copper, lead (at pH greater than
about 9), manganese (at pH greater than
about 7.5), and zinc (at high ionic
strength, and pH around 8). Hydroxide
solids are the predominant species for
chromium, iron, cadmium (at pH greater than
9), magnesium (at pH greater than 9), manga-
nese (at pH greater than about 9), and zinc
(at low ionic strength, and pH greater than
about 9) in the aged sludge. Arsenic, mer-
cury, and selenium exist primarily as
elemental metals in the aged sludge. Alu-
minium forms predominantly phosphate solids
at low pH, and oxide solids at high pH. In
aged sludge, the molybdate and silicate
solids are usually the predominant species
for lead and zinc, respectively.
The predominant soluble species of con-
stituents in the aged FGD leachates are
similar to those found in fresh FGD
leachates. However, the concentrations of
these soluble species are generally de-
creased through aging, due to the nature of
solids formed. The predominant soluble
species, and their concentrations for each
individual constitutent at two different
ionic strengths, are shown in Table 3. In
most cases, the predominant species alone
47
-------
TABLE 1. GENERAL MODELS USED FOR SPECIATION CALCULATION
[M(i)mL(J)J -
• fM(i)pL(j)q
d) [L(j)f]
YMU(T) • YLV(J) [M(Df]
£ ^ ^ RM(i)pL(j)
g c d R
Z Z Z RM(i)L(j)v = 1
1=1 u=l v=l u v
k 1 h
f m=l n=l j=l
h a b
Z Z Z P[M(1)L(J)]
-;-•] r)= | Q=] r H
J i -• i -^
h c d
+ Z Z Z u[M(i)L(j) ]
j=l u=l v=l
k 1 g
[L(j)T] = L(j)f + Z Z Z n [M(i)mL(j)n]
m=l n=l 1=1
+ Z Z Z q[M(1)L(j)]
1=1 p=l q=l L p q
(continued)
48
-------
TABLE 1. (concluded)
9 c d
+ £ E I V [M(i) L(j) ]
i=l u=l v=l
where:
[M(i)ml-(j)n] = concentration of complex M(i)mL(j)n (in Molar)
[M(i)f] = free concentration of ith metal (in Molar)
D-U)f] = free concentration of jth ligand (in Molar)
[M(i)y] = total concentration of ith metal in the system (in Molar)
[M(DpL(J)q]
and = concentration of solids M(i) L(j) and M(i) L(j) based
[M(i) L(j) ] on the solution volume (in Molar)
RM(i)pL(j)q
and = mole fraction of solid or gas species for metal or ligand
RM(i)uL(j)v SOlidS
i = metal species
j = ligand species
g = total number of metals
h = total number of ligands
k = maximum number of metals M(i) coordinate ligands L(j)
1 = maximum number of ligands L(j) coordinate metals M(i)
a,b, c and d = positive integers showing maximum number of the composition of
metals or ligands in the solids or gases
B(i,j)nm = overall formation constant of complex M(i)mL(j)n
Yx = thermodynamic activity coefficient of soluble species x, and
f = thermodynamic activity coefficient of solid (or gas) species x
(in this study, assume f = 1).
A
K = solubility products or Henry's constants.
49
-------
TABLE 2. CONCENTRATION RANGES OF CONSTITUENTS IN
* References: Bornstein ; Leo
FGD SLUDGES AND LEACHATES*
TO "ic "| Q "in
- - Lunt ; Princiotta ; and SCS .
Constituents
Aluminum
Arsenic
Beryllium
Boron
Cadmi urn
Calcium
Chromium
Cobalt
Copper
Iron
Lead
Magnesium
Manganese
Mercury
Molybdenum
Nickel
Potassium
Selenium
Silicon
Silver
Sodium
Tin
Vanadium
Zinc
Carbonate
Chloride
Fluoride
Sulfite
Sulfate
Phosphate
PH
Ionic Strength
Concentration in Scrubber Liquors Total Concentration (Sludge
mg/1 (except pH) Molar & Liquor) (Molar)
0.03-0.3
<0. 004-0. 3
<0. 002-0. 14
8.0-46
0.004-0.11
520-3,000
0.01-0.5
<0.19-0.7
<0. 002 -0.2
0.02-8.1
0.01-0.4
3.0-2,750
0.09-2.5
0.0004-0.007
0.91-6.3
0.05-1.5
5.9-32
<0. 001-2. 2
0.2-3.3
0.005-0.6
14-2,400
3.1-3.5
< 0.001-0. 67
0.01-0.35
41-150 (as CaCO,)
•J
420-4,800
0.07-10
0.8-3,500
720-10,000
0.03-0.41
3.04-10.7
—
10-5795_1Q-C95
10-7.27_10-5.40
10-6.65_10-4.81
10-3.13_10-2.37
10-7.44.106.01
1 RQ 1 1 ?
10 " -10
10-6.72.10-5.02
10-5.77.104.92
10-7.50.105.50
10-6.45.10-3.84
10-7.32_10-5.71
10-3.91_1Q-0.95
1Q-5.79_10-4.34
10-8.70_]0-6.46
— 4. 1 1 — 4. 1 ft
10 ' -10
10-6.07.10-4.59
1Q-3.82_10-3.09
10-7.90_1Q-4.56
10-5.15_1Q-3.93
10-7.33_10-5.25
o on r\ QQ
10" " -10
10-4.58.10-4.53
10-7.71.10-4.88
10-6.82.10-5.27
1Q-3.39_10-2.82
10-1.93_10-0.87
10-5.43.10-3.28
10-5.00.10-1.36
10-2.12_10-0.98
10-6.50.10-5.36
10-3.04_10-10.7
0.05-0.80
10-5.95_1Q-4.95
10-3'88-10"3'87
10-3.97.10-3.91
10-2.57.10-2.21
10-4.97_1Q-4.97
, n 19 n ?i
1Q u. I»_IQU.^I
10-4.03.10-3.99
10-2.87.10-2.86
10-4.18.1Q-4.16
10-0.57.10-0.57
10-4.69.10-4.65
10-3.91.10-0.95
10-3.46.10-3.41
10-5.83.10-5.74
,«-3 95 ,,,-3.80
10 -10
10-4.06_1Q-3.95
10-1.89.101.87
10-4.58_10-4.27
10-5.15.10-3.93
10-4.56_10-4.48
-1 36 -, -0 83
10 -10
10-3.06_1Q-3.06
10-3.45.10-3.43
10-3.58.10-3.57
10-0.20_10-0.20
10-1.93_1Q-0.87
io-2-2-io-2-17
10-0.24.10-0.21
10-0.45_]0-0.35
10-6.50_1Q-5.36
10-3.04_1Q-10.7
0.05-0.80
50
-------
Fiqure 1. Speciation of Ca in fresh FGD leachate at
I •= 0.05, £ra.] = 10"1-89M.
Figure 3. Speciation of Cd in fresh FGD leachate at
I =• 0.05, fcdT] • 10~7'44M.
Flnure 2. Speciation of Ca in fresh FGD leachate at
I = 0.8, £aT] = 10'1>12H.
11
Figure 4. Speciation of Cd in fresh FGD leachate at
I = 0.8, [CdT] = 10'6'01M.
51
-------
11
Figure 5. Speciation of soluble Ca in aged FGD waste
at 1=0.05, original [CaT] = 10°''9M
3£
K^
~12
o
16
20
24
7
pH
J_
-10
,-5
,-9
,-13
11
Figure 7. Speciation of soluble Cd in aged FGD wastes at
I - 0.05, original |cd = 10"4'97M
100
.1 40
20
FREE Ca (aq)
CaC03 (S)
Fioure 6.
5 7 9 n
pH
Primary distribution of Ca in aaed FG^) wastes
at I = 0.05, original CaT = 10°'19t1
Figure 8. Primary distribution of Cd in aged FGD wastes
at I - 0.05, original CdT
10-4'97M
52
-------
TABLE 3. PREDOMINANT SPECIES OF CONSTITUENTS IN AGED FGD SLUDGE
Ionic
Constituent Strength
Al
As
~ 0.05
0.8
0.05
_ 0.8
[0.05
0.8
Ca
Cr
~ 0.05
0.8
~ 0.05
_ 0.8
Predominant Solid Species
pH • 5
A1(H2P04)(OH)2
A1(H2P04)(OH)2
As0
As0
CdC03
CdC03
CaS03.lsH20,
CaS04.2H20
CaS03.'sH20,
CaS04.2H20
Cr(OH)3
Cr(OH)3
pH ' 7
A1(H2P04)(OH)2
A1(H2P04)(OH)2
As0
As0
CdC03
CdC03
CaS03.ijH20,
CaS04.2H20
CaS03.»sH20,
CaS04.2H20
Cr(OH)3
Cr(OH)3
pH - 9
A1203.3H20
A1203.3H20
As0
As0
Cd(OH)2
Cd(OH)2,cdC03
CaC03,CaS03.>sH20,
CaS04.2H20
CaC03,CaS03.isH20,
CaS04.2H20
Cr(OH)3
Cr(OH)3
Predominant Soluble Species
pH > 5
A1F2*(6.04)
A1F2+(5.05)
H2AS04'(8.03)
Vs04~{?.51)
Cd+Z(5.23)
CdCl+(5.12)
Ca*2(.21)
Ca+2(.25)
CrOH+2(4.13)
Cr(OH)2+(5.0)
pH - 7
A1(OH)3(6.26)
A1(OH)3(6.89)
HAs04"2(11.23)
HAs04"2(10.87)
Cd+2(6.03)
CdCl*(5.13)
Ca+2(.53)
Ca+2(.32)
Cr(OH)2*(4.76)
Cr(OH)2+(4.72)
pH » 9
A1(OH)3(5.95)
A1(OH)3(5.36)
HAs04'2(8.82)
HAs04"2(10.91)
Cd(S03)2'2(7.72)
CdC10H(6.07)
Ca*2(2.19)
Ca+Z(2.0)
Cr{OH)4'(4.03)
Cr(OH)4'(3.99)
-------
TABLE 3. (Continued)
Ionic
Constituent Strength
Cu
Fe
Pb
,
Hn
.
[0.05
0.8
•" 0.05
0.8
[0.05
0.8
~ 0.05
_ 0.8
~ 0.05
0.8
[0.05
0.8
Predominant Solid Species
pH - 5
Cu2C03(OH)2
Cu2C03(OH)2
Fe(OH)3
Fe(OH)3
PbMo04
PbHo04
__t
._t
.-+
_.t
Hg°
H9°
pH « 7
Cu2C03(OH)2
Cu2C03(OH)2
Fe(OH)3
Fe(OH)3
PbMo04
PbMo04
_.t
__t
—t
—t
H9°
H9°
pH • 9
Cu2C03(OH)2
Cu2C03(OH)2
Fe{OH)3
Fe(OH)3
PbMo04
PbHo04.PbC03
Mg{OH)2
Hg(OH)2
HnC03
«n(OH)2,HnC03
H9°
H9°
Predominant Soluble Species
pH - 5
CuB(OH)4*(15.38)
CuB(OH)4+(14.99)
Fe(OH)2+(7.16)
FeS03*(6.98)
Pb+2(5.80)
A
PbCl (5.67)
Mg+2(3.91)
Mg+2(0.95)
Mn+2(3.49)
Mn+2(3.56)
HgCl2{22.1)
HgCl3-(19.9)
pH » 7
Cu(B(OH)4)2(16.78)
Cu(B(OH)4)2(16.09)
Fe(OH)2+(9.16)
Fe(OH)2+(9.12)
PbB(OH)4*(5.82)
PbB(OH)4+(5.44)
Mg+2(3.92)
Mg+2(.95)
Mn+2(3.49)
Mn+2(3.56)
HgCl2(20.4)
HgCl3-(18.2)
pH • 9
Cu(B(OH)4)2(16.
Cu(B(OH)4)2(16.
Fe(OH)4(10.07)
Fe(OH)4'(8.96)
Pb(B(OH)4)3-(7.
Pb(B(OH)4)3'(5.
Mg+2(4.16)
Mg(1.13)
MnS04(4.10)
Hn+2(4.33)
Hg(OH)2(17.9)
HgC10H(17.0)
9)
4)
14)
55)
-------
TABLE 3. (Concluded)
Predominant Solid Species Predominant Soluble Species
Ionic
Constituent Strength pH = 5 pH » 7
0.05 — f — +
K
t t
L 0.8
~ 0.05 Se° Se°
Se
_ 0.8 Se° Se°
F 0.05 ~* --*
Na
L 0.8
[0.05 --f ZnS103
0.8 ZnS103 ZnS103
PH " 9 PH - 5
--* K*(1.89)
K+{1.87)
Se° HSe03"(28,6)
Se° HSe03"(28.6)
--t Na+(1.36)
--f Na+(0.83)
Zn(OH)2 Zn+2(3.63)
ZnS103,ZnC03. Zn+2(3.84)
Zn(OH)2
pH = 7
K+(1.89)
K+(1.87)
Se03=(18.2)
Se03=(18.2)
Na*(1.36)
Na+(0.83)
Zn+2(3.65)
Zn+2(4.06)
pH = 9
K+(1.93)
K+(1.91)
Se03*(6.19)
Se03'(6.19)
Na+(1.37)
Na+(.85)
ZnS04(5.67)
Zn(OH)2(5.9)
Note: ^Values In parenthesis indicate the -log molar concentration.
If one species accounts for less than 50 percent of the total concentration,
then more than one species will appear for each condition.
— Indicates that there Is no stable solid or that the stable solid Is In complex forms (e.g., complex silicates).
-------
will account for a major portion of the con-
centration of each constituent in FGD
leachates. Therefore, knowing the predomi-
nant solid and soluble species, the total
soluble concentration of a constituent in
FGD leachate can be easily calculated with-
out the aid of the computer.
MODEL VERIFICATION
The thermodynamic model was verified
by checking the model results against both
analytical data and certain theoretical con-
siderations.
Evaluations of the model in relation
to analytical data, was performed by com-
paring the known soluble concentrations of
constituents in aged FGD wastes to those
predicted by the model. As summarized in
Table 4, the calculated results for aluminum,
arsenic, boron, cadmium, cobalt, copper,
iron, manganese, mercury, potassium, sele-
nium, sodium, and zinc, either approach or are
very close to the concentration levels ex-
perienced in the field. For other elements
(specifically calcium, chromium, fluoride,
lead, and magnesium), the model was not as
effective. The low levels of calcium pre-
dicted by the model are due primarily to the
interaction of calcite with the Ca^+-C032-
and Ca^+-S0^~ complexes in the model. The
high levels of chromium and lead calculated
by the model are due to the inclusion of
hydroxide and carbonate complexes in the
model. For fluoride and magnesium, the dis-
crepancy may be caused by certain unsuitable
solids included in the model. The discrep-
ancies may be due to (1) errors in the
stability constants and/or activity coeffi-
cients; (2) the effects of other mechanisms,
such as adsorption by hydroxide solids or
clay minerals; and (3) the effects of kin-
etic constraints.
An evaluation of the thermodynamic
model was also performed according to scien-
tific considerations. In general, the model
results behave in accordance with basic
chemical and thermodynamic principles,
including the effects of changing pH, Eh,
and ligand levels.
EFFECTS OF FGD SYSTEM AND SLUDGE VARIABLES
ON CHEMICAL SPECIATION
For the purpose of selecting a sludge
treatment or disposal procedure, it is use-
ful to observe the possible beneficial or
adverse effects of operational or chemical
changes in an FGD system on sludge specia-
tion. The chemical changes studied here
include those of pH, ionic strength, chlo-
ride concentration, borate concentration,
sulfate concentration, and sulfite oxida-
tion. The operational changes studied were
limited to the addition of lime, silicates,
hydrogen sulfide, phosphates, and magnesium
to the FGD system. Due to the sizable
amount of results, details will not be de-
scribed here. Following sections only give
the qualitative descriptions of the results.
A change in pH can influence the direc-
tion of the alteration processes, e.g.,
dissolution, precipitation, adsorption, and
complexation, in any chemical system. In
general, a pH increase in the FGD waste
system tends to dissolve more elemental
constituents, e.g, As°(s), Hg°U), and
Se°(s), and to transform some of the car-
bonate, phosphate, or other solids into
hydroxide solids, thus affecting the concen-
tration of soluble constituents. A pH
change may also affect the ligand concen-
trations, and thereby change the concentra-
tion of soluble constituents.
The overall effects of pH on the total
constituent concentration depend on the
solubility constants of the new solids
formed, the new ligand concentrations, and
the formation constants of the complexes.
For example, a high pH level can increase
total soluble mercury and selenium, and yet
decrease most of the other bivalent trace
metals. For trivalent metals such as
chromium and iron, the minimum soluble con-
stituent concentrations occur in the neu-
tral pH region.
Although a change in ionic strength in
the FGD waste can affect the stability con-
stants, its effect on the soluble levels of
constituents, or on the stability fields of
various solids, are usually negligible if
their related ligand levels are unchanged.
The soluble chloride concentration of the
FGD waste is a very important factor in
determining the total soluble level of cad-
mium, copper, lead, mercury, and zinc.
Variations in borate concentrations have
an impact primarily on total soluble copper
and lead concentrations. The soluble sul-
fate concentration may affect the total
soluble calcium, magnesium, cadmium, and
zinc concentrations. In general, if the
total soluble levels of the above-mentioned
ligands (e.g., chloride, borate, and sul-
fate) are known, the total soluble metal
concentrations in the aged FGD leachates
can be approximated without extensive
56
-------
TABLE 4. VALIDITY OF THE THERMODYNAMIC MODEL FOR THE
PREDICTION OF FGD SLUDGE SPECIATION*
Constituent
Al
As
B
Cd
Ca
Cr
Co
Cu
F
Fe
Pb
Mg
Mn
Hg
K
Se
Na
Zn
Validity of
Model
Excellent
Good
Excellent
Excellent
Not applicable
Not applicable
Good
Excellent
Not applicable
Good
Not applicable
Not applicable
Excellent
Excellent
Good
Good
Good
Excellent
Reason for Discrepancy
Form strong CaCOo(s) when pH > 7
Form strong Cr3+-OH~ complexes
Solubility controlling solid unknown
Form strong Pb2+-C032~ & Pb -OH" complexes
Solubility controlling solid unknown
* Based on comparison of modeling results with Kansas City Power and Light FGD sludge
analysis.
computation.
With regard to operational changes,
sulfite oxidation may reduce the concen-
tration of sulfite complexes and increase
the concentration of sulfate complexes, but
will have very little impact on the total
soluble concentration of most metals. The
most significant effect of sulfite oxida-
tion is the transformation of CaSOs-^OU)
to CaS04-2H20(s) or CaC03(s), depending on
pH levels. This transformation may affect
the soluble levels of arsenic, mercury,
and selenium if the redox potential is
controlled by sulfate/sulfite species.
The addition of lime to the FGD sludge
has been employed in pozzolanic fixation
processes for the purpose of improving the
engineering properties of the dewatered
sludge. However, the model shows that lime
addition may have an adverse effect on
constituent solubility. The addition of
lime to FGD wastes may reduce the total
soluble levels of certain constituents such
as arsenic and manganese. However, the
total soluble levels of most other trace
toxic metals, such as cadmium, chromium,
copper, lead, mercury, selenium, and zinc,
57
-------
increase in aged FGD wastes following lime
addition. This may actually increase the
potential for environmental damage, should
the concentration increase outweigh the
dilution factor decrease which results
from permeability reduction.
The addition of silicates may reduce
the total soluble aluminum and zinc con-
centrations, but other elements studied
are virtually unaffected.
Phosphate addition will only reduce
two soluble major ions (calcium and mag-
nesium) while increasing the soluble cad-
mium level. Phosphate itself is also a
water pollutant, so the addition of phos-
phates is not recommended for the treatment
of FGD leachates.
Hydrogen sulfide addition may reduce
the soluble concentrations of trace metals
substantially. This operational change,
however, may not be desirable for an FGD
system for two reasons: (1) hydrogen sul-
fide itself is a pollutant, and (2) the
diffusion of oxygen into the sludge, fol-
lowed by the oxidation process, will even-
tually return the soluble metals to their
original concentrations.
Magnesium has been shown to improve
the efficiency of wet FGD systems; the use
of high magnesium reagents could therefore
become commonplace. The model shows that,
in general, the magnesium addition will not
significantly affect the total soluble lev-
els of most constituents.
CONCLUSIONS
Thermodynamic modeling of chemical
speciation in FGD waste has shown that waste
constituents can exist in a wide variety of
chemical forms or species. The predominance
and concentration of any particular chemical
species are influenced by chemical factors
such as pH, Eh, ionic strength, and total
concentrations of ligands and metals in the
system. Although the FGD chemical systems
are extremely complex, the speciation of
their elemental constituents can be quanti-
fied by calculation with reasonable ac-
curacy.
Knowledge of the relative distribution
of constituent species in the FGD system
is useful for (1) the evaluation of general
toxicity, and (2) predicting the migration
of the constituent in the environment.
When assessing the potential impacts of FGD
sludge leachate on groundwater, examination
of data from aged FGD wastes is most appro-
priate. Most in-situ FGD sludges have a
low permeability (10~4 to 10-10 cm/sec)
which provides months to years of contact
time between leachate and sludge. During
this period, various chemical species in
the FGD waste (either in the solid or sol-
uble phases) would gradually approach
equilibrium. Unfortunately, there is a
lack of documented information relating
to the chemical species present in aged FGD
waste. Therefore, the thermodynamic model
can be useful for prediction both the con-
centrations of various species, and the total
soluble concentrations of constituents in
aged FGD waste. The background required
for the calculation need include no more
than the total levels of the constituents
in the fresh FGD waste. This thermodynamic
approach could provide a considerable cost
saving over the traditional field survey.
The thermodynamic model discussed here can
also be used to predict solid or soluble
species changes, and changes in the levels
of total soluble constituents caused by
operational or chemical factors.
The thermodynamic model employed in
this study was found to be inaccurate when
predicting the speciation of calcium,
chromium, fluoride, lead, and magnesium.
The disparity may have been caused by sev-
eral factors, including adsorption by
various solids or the kinetic constraints
of the reactions. The speciation of other
constituents, such as aluminum, arsenic
cadmium, boron, cobalt, copper, iron, man-
ganese, mercury, potassium, selenium,
sodium, and zinc, showed very close corre-
lation with the analytical results. More
study is therefore suggested to (1) verify
the model against different types of FGD
wastes, and (2) include more of the con-
trolling factors in the model.
REFERENCES
1. Bornstein, L.J., et al. 1975. Reuse
of power plant desulfurization waste
water. Prepared for Pacific Northwest
Water Laboratory, National Environmen-
tal Research Center, U.S. Environmen-
tal Protection Agency, Corvallis,
Oregon, Grant No. R802853-01-0.
58
-------
2. Brinkley, S.R., Jr. 1946. Note on the
conditions of equilibrium for systems
of many constituents. J. Chem. Phys.,
14:563-564.
3. Brinkley, S.R., Jr. 1947. Calculations
of the equilibrium composition of
systems of many constituents. J. Chem.
Phys. 15:107-110.
4. Butler, J.N. 1964. Ionic equilibrium:
A mathematical approach. Addison-
Wesley, 547 pp.
5. Crerar, D.A. 1975. A method for com-
puting multicomponent chemical equili-
bria based on equilibrium constants.
Geochimica et Cosmochimica Acta,
39:1375-1384.
6. Crerar, D.A., and G. M. Anderson. 1971.
Solubility and solvation reactions of
quartz in dilute hydrothermal solutions.
Chem. Geol. 8:107-122.
7. Feldman, H.F., W.H. Simons, and D.
Bienstock. 1967. Calculating equili-
brium compositions of multiconstituent,
multiphase, chemical reaction system.
U.S. Bureau of Mines, RI 7257, 22 pp.
8. Helgeson, H.C. 1964. Complexing and
hydrothermal ore deposition. McMillan
Co., New York, pp 56-64.
9. Helgeson, H.C. 1968. Evaluation of
irreversible reactions in geochemical
processes involving minerals and aqueous
solutions - I. Thermodynamic relations.
Geochem. Cosmochim. Acta, 32:853-877.
10. Helgeson, H.C. 1970. A chemical and
thermodynamic model of ore deposition
in hydrothermal systems. Mineral
Soc. Amer. Spec. Paper 3:155-186.
11. Helgeson, H.C., et al. 1970. Calcula-
tion of mass transfer in geochemical
processes involving aqueous solutions
Geochim. Cosmochim Acta,34:569-592.
12. Karpov, I.K. L.A. Kazmin. 1972. Calcu-
lation of geochemical equilibria in
heterogeneous multicomponent systems.
Geochem. Int. 9:252-262.
13. Leo, P.R. and J. Rossoff. 1976. Control
of waste and water pollution from power
plant flue gas cleaning systems: First
annual R and D report. EPA-600/7-76-018.
U.S. Environmental Protection Agency.
14. Lu, J.C.S. 1976. Studies on the long-
term migration and transformation of
trace metals in the polluted marine
sediment-seawater system. Ph.D. Thesis,
University of Southern California.
pp. 184-233.
15. Lunt, R.R. et al. 1977. An evaluation
of the disposal of flue gas desulfuri-
zation wastes in mines and the ocean:
Initial assessment. Prepared for U.S.
Environmental Protection Agency, Office
of Research and Development, Washington,
D.C. 20460. Contract No. 68-03-2334.
16. Morel, F. and J.J. Morgan. 1972. A
numerical method for computing equili-
bria in aqeuous chemical systems.
Envir. Sci. and Tech. 6:58-67.
17. Naphtali, L.M. 1959. Complex chemical
equilibria by minimizing free energy.
J. Chem. Phys. 31:263-264.
18. Princiotta, F.T. 1975. Sulfur oxide
throwaway sludge evaluation panel
(SOTSEP), Vol II: Final report - Techni-
cal discussion. EPA-650/2-75-010-b,
U. S. Environmental Protection Agency.
19. SCS Engineers. 1977. Compilation of
data base for the development of
standards/regulations relating to land
disposal of flue gas cleaning sludges.
Report submitted to U.S. Environmental
Protection Agency, Ohio,
Contract no. 68-03-2352.
20. White, W.B. et al. 1958. Chemical
equilibrium in complex mixture. J.
Chem. Phys. 28:751-755.
59
-------
INTERPRETING RESULTS FROM SERIAL BATCH EXTRACTION TESTS OF WASTES AND SOILS
Martin J. Houle and Duane E. Long
Department of the Army, Dugway Proving Ground, Dugway, Utah 84022
ABSTRACT
It has been established that a graded serial batch extraction method is very useful
for studying the Teachability of industrial wastes and for determining the retention
characteristics of soils. A correlation between waste and soil extraction volumes and the
time of leaching in columns or in the field allowed developing a method for the accelerated
testing of wastes and soils. Proper interpretation of the results yields a real-time
analysis of what is occurring in the waste and underlying soils as the leaching progresses.
Applications of the data and methods of presenting it in graphical and tabular form are
demonstrated and discussed.
BACKGROUND
CONTROLLED EXPERIMENTS ARE NECESSARY
Samples taken in the field have some
value for making predictions of waste
Teachability and the migration of hazardous
ions through soil. However, the obser-
vations are limited to only those wastes
which have actually been dumped and even
then to just certain combinations of waste
and soil types, waste-to-soil ratios, etc.
In the field, variables are not readily
changed to determine their effect and this
lack of flexibility limits the examination
to relatively few factors. Many variables
are completely out of the control of the
experimenter. Thus, predictions made from
these types of studies may apply to a
relatively limited range of industrial
situations.
Laboratory studies have the advantage
of allowing more control over a wider range
of experimental conditions. But they also
have the potential disadvantage that signifi-
cant field variables may be excluded because
of the difficulty in simulating complex
field conditions or because of improper
experiment design. It is necessary to
include those variables active in the field
which have a statistically significant
effect upon the leaching of substances from
a waste and/or the passage of these
substances through soils, and to exclude or
minimize variables that might be introduced
by the experimental setup. If all of the
controlling factors are not varied to
establish their effects and if they are not
included at proper levels, the empirical
equations derived from this data may not
be reliable in making predictions for
situations where these variables are
involved. It may be necessary to run
preliminary experiments to see which vari-
ables have significant enough effects to
merit including them in the final experi-
mental plan. (Theoretical models may be
of some help in identifying potential
factors, but they are not reliable enough
for making predictions in complex physical -
chemical situations without experimentally-
determined constants, coefficients, and/or
exponents capable of compensating for an
incomplete knowledge of the controlling
variables and their interrelationships.)
There are two principal laboratory
approaches for measuring the Teachability
of wastes and the migration of hazardous
substances through soil. These are batch-
wise extractions and the continuousTy-Teached
column method, both of which are discussed
below.
CONTINUOUSLY LEACHED COLUMNS VERSUS BATCH
EXTRACTIONS
60
-------
The migration of chemical substances
through soil is usually studied in the
laboratory using columns packed with soil
to a predetermined bulk density (usually
approximating the field density of the
undisturbed soil). These soil columns are
challenged with a solution extracted from
a waste by water or by some other solvent
such as municipal landfill leachate, or the
soil is treated with simple solutions
containing the ion under study. A useful
configuration is shown in Figure 1 along
with illustrative plots of the data
obtained.
Continuously-leached column experi-
ments provide information as to the ability
of a soil to remove chemical substances
from a waste extract. However, an important
limitation of this method is the time and
effort required to obtain and analyze a
sufficient number of samples to make pre-
dictions 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
relatively short-term column studies cannot
be relied upon to describe what will occur
during years of leaching.
When setting up experiments of this
type, an investigator is faced with the
problem of selecting values for each
experimental parameter such as leaching
solvent flow-rate, head pressure, soil
Waste
Waste Effluent Volume -
Soil
Soil Effluent Volume
Figure 1. Continuously Leached Columns and
Associated Output Plots
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 experi-
ment may not be very useful for making
general predictions. In addition, column
experiments are cumbersome and do not
readily lend themselves to changes in the
levels of the experimental parameters.
Thus, they are slow and relatively limited
in applicability.
If an experimental approach were
available which is more rapid and more
versatile than the usual column leaching
methods, a wider range of environmental
conditions could be investigated, thereby
more completely describing the behavior of
a waste deposited on a soil. This would
also make it more practical to use factorial
experiment designs. Factorial experiments
allow making predictions without sacrificing
reliability even in the presence of inter-
action between multiple variables. (Inter-
action exists when the effect of one factor
is dependent upon the level another factor.
This introduces error into the results of
classical, vary-one-factor-at-a-time experi-
mentation). A fast method would also allow
making timely determinations, on demand,
for each specific situation.
In previous studies of the Teachability
of certain metals from a number of industri-
al wastes and the migration of these metals
through soils, this laboratory used continu-
ously-leached columns. A batch method was
employed to rapidly screen soils for their
ability to remove these metals from the
waste leachate (1). 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 the leaching of both waste
and soil could provide much of the same
information obtained from column studies.
However, no adequate information was avail-
able as to the correlation between batch
extractions and continuously-leached columns
of industrial wastes and soils. Therefore,
samples of the industrial wastes used in the
continuously-leached column studies were
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 (2) even though substantially
greater amounts of water were present
(200 to 4,800 percent versus less than 50
61
-------
percent). 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 was experimentally much
simpler, and it was concluded that this
approach would permit the rapid investi-
gation of the effect of a wide variety of
environmental factors such as freeze/thaw
and drying/resaturating cycles and similar
variables that would be difficult to include
in column studies.
Other investigators have used batch
soil methods to study the removal of
certain chemicals from waste extracts or
municipal landfill leachate (3-6) and
obtained results that compared reasonably
well with column experiments. However,
their experiments either did not allow for
the changes in the waste extract composition
as the waste depleted, for the further
change as the extract contacted each
increment of soil, and/or for the continu-
ally changing conditioning of each increment
of soil, a change which depends both upon
the leaching time and the soil depth.
CORRELATING CONTINUOUS AND BATCHWISE
LEACHING^2)
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 samples, 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 kind or
amount of soil will change the scale of the
cumulative volume axis when pore volume is
employed. Figure 2 is an example showing
the difference obtained with different pore
volumes (in this case 40 and 60 mililiters).
The corresponding total volume in milli-
liters is appended for comparison.
It often is not practical or possible
to determine a pore volume for a waste
because of its physical form (heterogeneous
suspension, liquid, etc). This problem was
circumvented by using the soil column pore
volume as the measure of liquid volume
through the waste. It allowed correlating
the waste-column output with the soil-column
results in a given set of experiments.
However, instead of using the soil pore
volume as the principal plotting parameter,
it is much more flexible to plot the
0
0
0
6
2 4
200
ib
6 8
400
Cum.
15
10
600
Vol.
20
12 14
800
25
16
1000
PV =
PV =
Total
4O
60
mis
ml
ml
Figure 2. Differences in Scales used to
Plot Cumulative Volume.
observed concentration of a chemical in an
extract versus the cumulative milliliters
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 differ-
ent designs of experiments. It normalizes
the results so they can be more readily
correlated to a range of field conditions.
The areas under the curves represent the
total weight of a chemical extracted per
gram of waste or soil. The weights thus
obtained can be used to calculate attenu-
ation or penetration factors for the soil.
o>
a.
o
o
0 20 40 60 80 IOO 120
Cum. Vol., ml/g
Figure 3. Normalization of Cumulative
Volume using Milliters Per Gram
62
-------
These considerations make it possible
to correlate batch and continuously-leached
column experiments. Batchwise extractions
can be related to continuously-leached
columns by recognizing that continuous
leaching is equivalent to running a series
of discrete extractions spaced by the fre-
quency of collecting the effluent sample.
Figure 4 shows that the concentration of a
substance in the periodic column samples
can be plotted to represent the average for
that sampling period. Thus, samples from
the continuous leaching of a column corres^
pond to sequential 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
graded in size, as indicated by the ex-
traction volumes pictured in Figure 5. A
small sol vent-to-sol ids ratio should
probably always be employed for the first
extractions. This is usually when the
concentration 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 maximum. Greater
dilutions would reduce this, possible
affecting the solubility of other com-
ponents. After the more soluble components
have been extracted, the sol vent-to-sol id
ratio can be greatly increased, thus reduc-
ing the total number of extractions required.
Obviously, the further along the cumulative
milliliter per gram axis the extraction
o
o
20 ' 40 ' 60 80
Cum. Vol., ml/g
100 120
o
O
K *-_!
20 40 60 80
Cum. Vol., ml/g
100 120
Figure 5, Graded Serial Batch Extractions.
volumes extend, the longer the period of
column leaching the batch work is equivalent
to,
ACCELERATED TESTING
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. The time required for a volume
of liquid to penetrate a gram of soil at a
selected pore water velocity can be cal-
culated from Equation 1 : (7;
where T
o>
time, days,
effective pore volume per
Figure 4. Correlation of Batch with
Continuously Leached Columns.
gram soil , cm,
Pj, = soil bulk density, g/cm3,
v = pore water velocity,
cm/day.
This equation was used to prepare Table 1
which applies to the sequence of extraction
volumes (the accelerated testing schedule)
used in this report. To choose the correct
penetration time from the table, the pore
water velocity of the liquid in a soil
underlying a specific waste-disposal site
must be calculated or determined experi-
mentally. (The flowrate-determining factor
will often be the penetrability of the layer
of waste). This table also applies to the
63
-------
TABLE 1, CORRELATION BETWEEN EXTRACTION VOLUME AND PENETRATION TIME (7)
Extraction
Number
1
2
3
4
5
6
7
Water
Added,
(ml/q)
2
3
6
12
24
48
96
Cumul ,
Vol.,
(ml/g)
2
5
11
23
47
95
191
@10~'*
0
1
3
7
15
30
61
Equivalent
cm/sec
.65
.62
.56
.45
.2
.8
.9
Days of Penetration*
@10-5cm/sec @1
6.5
16.2
35.6
74.5
152.
308.
61 9. (1.7 yr)
in-6
65
162
356
745
1520
3080
6190
cm/ sec
'. (2
. (4
. (8
.(16
.0
.2
.4
.9
yr)
yr)
yr)
*At the specified pore water velocity through a soil having an effective pore volume of
0.223 ml/g.
leaching of waste because the volume of
liquid passing through a column of waste
will ordinarily be about the same as through
the soil beneath it. Similar tables can be
calculated for soils having other pore
volumes and for other velocities.
MATERIALS AND METHODS
WASTE SAMPLES
Of the industrial wastes which have
been examined by this graded serial batch
extraction method, data from two wastes of
divergent characteristics were chosen to
demonstrate here the utility of this pro-
cedure for the accelerated testing of
wastes and soils, and to show ways the
results can be interpreted.
The hydrofluoric acid production waste
discussed here is the residue from treating
fluorspar with concentrated sulfuric acid.
The dry waste is primarily calcium sulfate
with a small amount of calcium fluoride and
is strongly acid. It contains 4,900 micro-
grams of fluorine per gram.
The secondary zinc smelter sludge is
the residue from the air pollution control
scrubber. The amounts of potentially
hazardous elements in this waste are shown
is Table 2. Their concentrations were
followed in each of the waste and soil
extracts; the data from cadmium is used
in the illustrative example.
WASTE AND SOIL EXTRACTION METHOD
A sequence of seven extracts was made
from each waste. Ordinarily a sample of
waste was dried to determine moisture con-
tent, then sufficient undried sample was
weighed to give 300 grams dry weight.
TABLE 2. ANALYSES OF SELECTED ELEMENTS IN
SECONDARY ZINC SMELTING SLUDGE
Element
Concentration
(microgram/gram)
Beryll ium
Boron
Cadmium
Chromium
Copper
Lead
Magnesium*
Nickel
Zinc
0.82
57.6
54.5
14.8
1,270.
68,200.
920.
360.
383,000.
*Magnesium was measured in order to correct
the lead obtained bv argon plasma emission
spectrophotometry(''.
(Drying the sample could affect hydrated
species and drastically reduce the solubi-
lity.) Appropriate volumes of water were
added for each extraction to produce the
liquid-to-sol id ratio given in the second
column of Table 3. (If a waste has super-
natant water, the volume of the water should
be considered as part or all of the first
extract.) The sample 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. At the end of the extrac-
tion 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 milliliters of
the filtrate was withdrawn for analyzing
the ions of interest and filtered through
64
-------
TABLE 3, SPECIFICATIONS FOR SERIAL BATCH EXTRACTIONS
(8)
Extract-
ion
Number
1
2
3
4
5
6
7
Water
Added
(ml/g)
2
3
6
12
24
48
96
Volume of
Water (ml)
Extracting
300 g Waste
600
900
1,800
3,600
7,200
14,400
28,800
Volume of
I
60 g
Soil
120
180
360
720
1,440
2,880
5,760
Filtrate Onto a Soil (ml)
II
30 g
Soil
60
90
180
360
720
1,440
2,880
III
15 g
Soil
30
45
90
180
360
720
1,440
a 0.5 v Millipore filter to remove fine
particles which might have by-passed the
filter paper and could dissolve when the
sample was acidified (after measuring
conductance pH, and fluoride ion, one
percent 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 6
outlines the sequence of operations.
In the procedure detailed here, the
liquid-to-solid ratio was continually
increased (Figure 5) to further accelerate
the testing»the volume of each extraction
after the second one was made double the
one before, which redoubles the time rep-
resented 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-solid ratios, followed by
a very large one. However, this would
have to be checked for each kind of waste
by comparison with the more conservative
series of extractions utilized in Table 3.
Such a procedure would allow rapid simula-
tion of long leaching periods and could be
useful in the routine monitoring of vari-
ations in waste composition and leach-
ability. (2)
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
extract left over to challenge the next
soil batch at the same liquid-to-solid
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 3.
Although the soil equilibrates in six
hours or less (9), 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 sets of samples progressing
smoothly without gaps in the series. After
filtering the soil extract, an aliquot was
refiltered through a Millipore filter 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.
The waste composition changes as
components are leached from the waste.
Each succeeding portion of extract will
therefore generally have a different
composition as shown by the upper two
histograms of Figure 7, Besides being
challenged by a changing solution, the
soil's ion-removal characteristics contin-
ually change with time as the soil becomes
conditioned and loaded by the passage of
waste extracts. Since each portion of
waste is changed by passage through soil ,
the conditioning each succeeding segment of
soil receives is different and each segment
therefore may remove different portions of
the various ions present in the waste
extract. So although the soil segments
start out the same, in effect they become
65
-------
99
Extraction # 2
-i
(D
o
o
13-
OJ
-3
-i
£75
-s
01
O-
n>
a.
oo
n>
-s
ca
a>
o
/^ —
s s»'
XT- "*
V i/
Refilter portion
through Hillipore
and analyze.
Z_
** ft «B 1
I 2||g Mix b
' ' G J M periodically g
s— r
J
Filtrate: Add 45 ml
V A^
Refilter portion
through Hillipore
and analyze.
, /I-———
a f ° £ P ml^f
? i f • s * i s s
t 57 SH^
J
V. ?
*££ xt— S
SL-"g si? 1
• ?&^:
V r
Filter portion through
Hillipore for analysis.
ft Hix
ff periodically
k
PS1? £5
•• s • f «
:x_5u gi
v /^
Refilter portion
through Millipore
and analyze.
r
~l
Hix >
f periodically 3
i
4
ftsr MIX
IT R periodically
^1
"§. *
9
Filtrate: Add 180 ml to filtered soil batch.
< ? ff B ? I i
V >^
Refilter portion
through Hillipore
and anlayze.
f
]
f •* a
ft Mix C
g periodically i
i
^
;
•
!
^
3
^
^
* | » *
«& 7
PS ^ »
• sr
*
v /^
Filter portion of remainder through
Hillipore and analyze.
H Mix
J periodically
!
i
1
t
*
1
if
ft
ff
Filtrate: Add 120 ml to 60 g of each type of soil.
W
IT
S
H
H
*4 I O > < w C
p tr H * °
i » >. s *•
« P 5"S
s s Is
D. 10
0 ml aliquot through Hillipore and analyze.
aily to find equilibration tune.
*\
s |
u w
4 P
P 1
H W
H
SH W
in
|!f
IP'S
' "re
-------
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 pfck
up a specific ion from a waste solution 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 sol-
ution. If extract samples were taken with-
in a layer of soil, it would be possible to
study this dynamically-changing situation.
This can be accomplished by placing sampling
ports in the side of a soil column, as shown
in Figure 7. The same results can be
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
sampling between batches of soil allows
Water In
Waste
Soil
Waste Extract
Soil Sect I Extract
Soil Sect 1C Extract
Soil Sect HI Extract
Figure 7. Challenging Multiple Soil Segments
with Successive Extracts of
Waste.
perceiving what is happening within a bed
of soil and provides data which could per-
mit extrapolating 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 conditions. It is re-emphasized
that batchwise testing also yields its
information in a small fraction of the time
required by columns or field studies.
UTILIZATION OF BATCH EXTRACTION DATA
A variety of calculations can be per-
formed using the results from the graded
serial batch extraction experiments on
wastes and soils. Table 4 lists those
which were done for this report. In Table
4, W refers to the waste extract, and I,
II, and III, identify the extraction from
the first, second, and third batches of
soil, respectively. The subscripts affixed
to these symbols identify the number of
the extraction in the series of seven
employed. The resulting character refers
to the amount of a chemical species found
in the extract, expressed in terms of micro-
grams per gram of waste or soil . Each
calculation utilizing these quantities is
explained in the following paragraphs.
Calculating the Weight of a Substance
Leached from Waste
The weight of a substance leached per
gram of waste can be calculated by multi-
plying the concentration of the substance
observed in the extract by the volume of
water or other extraction liquid used,
divided by the weight of waste being ex-
tracted. In consistent units: microgram/
milliliter x milliliter/gram = microgram/
gram waste. Thus, the weight extracted by
each extraction in the series listed in
Table 3 was obtained by multiplying the
observed concentration in micrograms/
milliliter by the extraction volumes listed
in the second column of the table.
Since one percent concentrated nitric
acid was added to keep ions in solution,
and ten percent by volume of lithium nitrate
solution was added to suppress interference
from alkaline metals and earths in the
analysis by argon plasma emission spectro-
photometer, the observed concentration must
by adjusted to obtain the actual concent-
ration. (The correction for the lithium
addition was accomplished by programming the
concentrations printed out by the spectro-
67
-------
TABLE 4. CALCULATIONS MADE FROM THE SERIAL BATCH EXTRACTION DATA
i
00
Extraction
Number
Layer
I-HI
I+II+III
III
I-MI
I-HI + III
Amount
Amount
Penetrating j Retained
Averaged this
Raw Data Extract
ug/ml ug/g ug/g
] r
Cumulative 1
Total !
Challenge
ug/g
Out In-Qut
(In) !
III,
III,
n,-in,
ni2 w2-ni2
E In
Cumulative Total
Retained
ug/g
E(In-Out)
Fractionjetained.
From "[From Cumulative
this iTotal Challenge
Extr. i
Distribution Coefficients
In-Put I £( In-Put)
In I In
Pene-
tration [ Including Soil From Solution Only
Factor Distrib.^Slope, Distrib.j Slope,
1 Ratio Degrees Ratio Degrees
E(I"-Qut)+Si)i1
Out
wr'i
Vi
wrni
Vnii
wz-iz
wr'i
Ml
IIj
Ij
III,
III,
Eqn. 6
Eqn. 8
Eqn."8
Tan" Eqn.
Tan" Eqn.
Tan Eqn.
Eqn. 6
with
Eqn. 8
with
II =0
Tan
Tan
-1
Eqn.=8 Tan
with *
Eqn. 9
Tan" Eqn.
Eqn. 9 Tan
with I , -i-
II0.0 »
Eqn. 10 Tan" Eqn. Eqn.10 Tan
10 with I , *
II .II1}
=0° °
APPLY ABOVE EQUATIONS
-------
photometer.) The corrected raw data from
replicate extractions Is averaged and posted
as column C, Table 4, Multiplying each
waste-sample concentration by the appropr-
iate above-mentioned factor yields the
corresponding result in column D.
Calculating the Height of a Substance
Penetrating and Retained on Soil
The weight of a substance extracted
per gram of soil is calculated for each
extraction in the same manner as described
above for waste. The multiplying factors
remain the same because the extraction
volumes in Table 3 were adjusted for all
the soil batches to maintain the same
millIHter/gram as for the waste.
The amount penetrating a batch of
waste or soil becomes the challenge to
the next batch. Thus, the yg/g out of
the W-| becomes the yg/g in for I]. The
weight of a substance coming in to a batch
of soil minus the weight out equals the
weight retained by the soil during the
extraction. In this way, the values
entered in column D, Table 4 are used
to produce column E. The fraction retained
by a soil from an extract is calculated by
dividing the results in column E by the
weight coming in, as indicated 1n column
H.
Whenever a calculated value has a minus
sign, it means that the soil either gave
up some of that substance which it had
previously picked up from the waste
leachate, or it gave up some originally
present in the soil before being exposed
to that waste. If the original analyzed
concentrations were near the detection
limit, then, because of the normal deter-
ioration of the precision of an assay near
its detection limit, less significance can
be assigned to the corresponding derived
values and to the appearance of a minus
sign.
A useful property of the fraction
retained from the cumulative total chall-
enge is that when the soil is yielding the
element of interest and it is desired to
know how many time greater the amount
exiting is greater than the challenge,
change the minus sign on the value in
column I to a plus and add 1,00. Thus,
a fraction of -8.05 retained from the
total challenge means that 9.05 times as
much of that element was given up as was
present in the total challenge.
The results of passing one waste ex-
tract through the three soil batches can be
presented as in the histogram of Figure 8.
The height of the histogram bar labeled W
represents the mass 1n micrograms of a
substance extracted per gram of waste.
This is the challenge to the first batch
of soil , which represents the top layer in
a bed of soil. The height of the bar
labeled I shows the concentration of the
species penetrating the batch representing
the next layer of soil, and the difference
in height between I and W is the amount of
the species retained per gram of soil.
The cumulative sum of the challenges
to a given batch (layer) of soil is obtained
by summing up the results for the successive
extractions that have challenged that batch.
Thus, as seen in column F of Table 4, the
cumulative sum of the challenge to soil
I IP is I-i + Ip. The cumulative total
retained can oe calculated in a similar
manner, as indicated in column G, and used
to determine the fraction retained from the
total challenges, as recorded in column I.
These values are of particular interest in
studies of soil capacity.
o
in
o>
o
c
o
O
Out of
Waste
Retained
on I
Penetra-
ting I
Retained
on H
Penetra-
ting IT
------- 1
I
Retained
on
Penetra-
ting HI
W
Waste
in
Soil Sections
Figure 8. Histogram Showing the Penetration
and Retention of a Species by
Sofl for One Set of Extractions.
69
-------
Calculating Penetration Factors
The quantity of a substance penetrating
a batch of soil divided by the amount of
challenge (both found in column D) can be
defined as the Penetration Factor (the
reciprocal of the Attenuation Factor), which
is found in column J of Table 4. This is
the fraction penetrating the soil and it
can also be viewed as the decimal percent.
Multiplying a challenge by the Penetration
Factor gives the amount penetrating. If
the Penetration Factor is greater than 1.0,
it indicates that the soil is either yield-
ing material previously held up during the
passage of extracts, or that some is being
displaced out of that originally present in
the soil.
The penetration factor calculated in
column J is for the corresponding extraction
only and is not cumulative. The cumulative
penetration factor can be obtained by
subtracting the cumulative fraction retained,
column I, from 1.00.
Calculating Distribution Coefficients
(10)
The chromatographic distribution co-
efficient, K, is defined as the concent-
ration of a species in the solid phase
divided by the concentration in the liquid
phase. (11) This distribution ratio (columns
K and M of Table 4) is the slope of the
line showing the relationship between the
concentration adsorbed on the soil and the
concentration remaining in the solution.
The angle of the slope at the point is the
arctangent of K. (The angles are tabulated
in column L and N.) At low concentrations,
the relationship is usually linear in simple
systems, but K typically decreases with
increasing amounts of solute, i.e., as the
sites are occupied by increased amounts of
the substance being studied. Large K's
show a high relative retention. Experiments
to determine K are normally done at constant
temperature because K is temperature
dependent (changing the temperature changes
the position of a point on the plot showing
the amount of a species removed from the
solution by the soil) and so the curves
are called isotherms.
Temperature is not the only parameter
which affects the distribution between
phases. The pH, the ionic strength, the
presence of competing ions, the previous
history of the solid, its current surface
energy and effective surface area, in short,
the total conditions must be uniform because
each of these things can affect the distri-
bution ratio. So, iso-conditons are needed,
not just iso-therm, while experiments are
run if this kind of plot is to be obtained.
If the conditions are changed to some other
level, a distinctly different set of points
can be obtained. (If conditions are allowed
to vary at random, then randomly located
points can be expected; if conditions are
varied according to some plan, like a
factorial experiment design, then their
effect on the distribution coefficient can
be examined. For the results to apply to
a field situation, field conditions must be
simulated in the experimental array.)
These lines are called adsorption
isotherms, but more than adsorption can be
involved in removing ions from solution.
Besides dispersion and dipole forces,
hydrogen bonds and weak covalent bonds,
acid-base interactions and complex formation,
strong covalent bonding responsible for
chemisorption, precipitation, and even
mechanical filtration, all could contribute
to the removal of a substance from solution.
So what are called adsorption isotherms
really are plots of removal or retention in
iso-conditions, that is, they are plots of
temporary retention under a given set of
conditions. Even irreversibility is cond-
ition-dependent. An insoluble compound may
become soluble as the conditions change,
e.g., as the extraction progresses or as
the soil changes.
The distribution coefficient, K, is not
for adsorption alone, but for all equilibria
causing retention or displacement. K can
be computed on several different bases; as
an experiment is started, the concentration
of a substance being studied will probably
be zero in the solid phase if a chromato-
graphic substrate is being worked with, but
in soil work, any compounds present in the
soil which contain the ion(s) being studied
are a potential source that can contribute
to its concentration in the leachate solution
passing through the soil. The waste leachate
is a potentially powerful solubilizer and
displacer of components in the soil, i.e.,
the soil can act as a reservoir of Pb, Cd,
Ti, Zn, etc., which the waste leachate can
cause to bleed off and carry on down to
ground-water. A negative slope to the reten-
tion isotherm shows the soil is giving up
the ion, acting as a source. The waste is
the cause of this kind of pollution, but
not the source—the soil is the source.
70
-------
The distribution coefficient, K, can be
based on a unit volume of solid adsorbant,
or a unit weight, which is employed in this
report. The concentration expressed in
microgram/gram soil can refer to the number
of micrograms of ion removed from a given
solution challenging the soil, it can
represent all that the soil has removed from
a series of solutions (as in column M of
Table 4), it can be a total which also in-
cludes all that was originally present in
the soil and therefore potentially available
to the equilibria (as used in column K), or
the total concentration in the soil can
include just that portion known to be soluble
in or displaced by the solution challenging
it.
The concentration in the solution is
usually expressed as weight per unit volume,
such as micrograms per milllliter, and the
resulting units for K are mill inters per
gram, as shown in equation (2):
= yg/g x ml/iag = ml/g. (2)
yg/ml
yg/ml x ml/g = yg/g.
(3)
Then, as seen in Equation (4) the ratio, K,
becomes a properly dimension!ess constant
as a result of employing this more funda-
mental relationship:
(ml/g) " yg/g (4)
(ug/ml)
As previously, discussed, the conditions
for retention can be expected to differ with
soil depths because each layer of soil is
challenged by a different solution matrix.
The conditions also will change with time
as the leaching of the waste progresses and
generates a solution of chanqlnq composition.
So that this latter effect could be examined,
equations were derived to calculate the
distribution coefficient for every succeed-
ing extraction of the waste. The distribu-
tion of a species between the solid and the
liquid phases will be the sum of the amount
originally on the soil and the amounts
retained from each of the successive waste
extracts, all divided by the amount in the
solution leaving that batch of soil. Thus,
the distribution ratio resulting from n
extracts passing through soil batch I can
be computed by the following equation:
I + (H.-I.) + (H,-I,) + ... (H -I
(5)
(6)
* Slope of adsorption isotherm,
where,
I = concentration of a species originally
on the soil , yg/g,
But, even though it is the liquid that is
being analyzed, it is the soil that is being
studied. The concentration is measured as
micrograms per mllliliter, but of more direct
importance 1s the weight of soil contacted
by this solution. If the concentration
observed in the liquid, micrograms/mllliliter,
is multiplied times the total amount of
liquid put on the soil, expressed as mini-
liters/gram, Equation (3) shows that the
concentration in the solution is obtained
in terms of micrograms per gram of soil
contacted.
I.
^
W,
concentration of the species in the
iVn extract passing through soil
batch I, yg/g, and,
concentration of the species in the
ith extract out of the waste, yg/g.
The concentrations of the solutions after
equilibrium has been established (the out-
put concentrations) are used in these
calculations instead of the starting (the
input) concentrations.
The distribution ratio will ordinarily
be different for the second batch of soil
because it is being challenged and condi-
tioned by a different solution — a solu-
tion which has been modified by passing
through the first batch of soil. The
distribution coefficient for the second
soil batch can be calculated for each
extraction using the following formula:
n0+ (1,-n,) * »riiz) * ... (in-nn) (7)
K = - -
n n
"
i - 1
• Slope. (g)
71
-------
The distribution coefficients for each
extraction passing through the third batch
of soil are calculated in a similar manner
using the differences in concentration
between the second and third soil batches.
The experiments generally will be done
with the same kind of soil in each batch,
so the concentration of a species originally
in the batch of soil, I0= II0= III0. If
only the amount of a species removed from
the solution is to be considered in the
distribution coefficients being calculated,
neglecting that which is originally present
in the soil, set I , II , and III all
equal to zero.
The effect of different soil-to-waste
ratios can also be checked in this experi-
ment by calculating distribution coeffic-
ients for the waste extract challenging
the first two soil batches and then all
three batches:
i + ii
(w- - ii,)
(9)
1/2 II
for a 2:1 soil-to-waste ratio, and
i + ii + nr +
000
.(10)
for a 3:1 soil-to-waste ratio.
The effect of changes in the extract
due to passage through the soil also can
be obtained by calculating:
U1
(K - HI,)
(11)
1/2 III
and comparing it with the results from
K(H-II)n (Ec!uation 9)-
Since a distribution coefficient is
the slope of the retention curve (the
"isotherm") at the point represented by
the numerical values of the numerator and
denominator in the ratio, the angle whose
tangent is K can be depicted on a graph by
an appropriately oriented line segment.
This seemingly would allow drawing a. curve
of retention under iso-conditions (an
"isotherm"). But an examination of the
experimental data reveals, as postulated
when designing this experimental approach,
that there is a considerable change in K
for most ions as the leaching of the waste
progresses and as the waste passes down
through the soil. This shows that in the
real field situation, or in experiments
•which approximate it, constant conditions
for adsorption will not exist during the
presentation of the challenge. This means
that the remainder of the curve cannot be
determined. Only a family of line segments
will be obtained from the K values calculated
for each waste and soil extract, with each
line segment representing the distribution
ratio under a different set of conditions.
(An example is depicted in Figure 13, back
where the data are discussed.)
INTERPRETATION OF RESULTS FROM
INDUSTRIAL WASTE EXTRACTION
The samples of wates were collected
from a hydrofluoric acid manufacturing
plant, and a zinc secondary-refining
operation. These wastes were examined
for the water-extractibility of certain
inorganic ions during a series of seven
extractions. The resulting leachates were
then applied, in sequence, to three batches
(corresponding to three layers) of each of
three different clay soils. The data for
Chalmers soil is used here to demonstrate
the interpretation of results. Analysis of
the solutions for pH, conductivity, and
concentrations of specified ions before and
after contact with the soils allowed observ-
ing the effect of a soil on the leachate,
as well as seeing how later changes in the
leachate affected the retention of an ele-
ment on a soil . No attempt was made to
investigate the mechanisms responsible for
the soil and waste leachate interactions.
The wastes discussed as examples
represent a reasonably wide range of viaste
characteristics from very soluble to
relatively insoluble, and from extremely
acidic to near neutral. These practical
applications provided data which could be
presented in a variety of ways.
First, a short table is used to indicate
the Teachability of the element of interest
in from a waste. (Tables 5 and 6 are examples-)
72
-------
TABLE 5. LEACHABILITY OF HYDROFLUORIC ACID PRODUCTION HASTE
Element Initial
Cone.
(yg/ml )
When Concentration Levels Off Total Weight
Cone.
(yg/ml )
Extr.
Nr.
Extr. Vol.
(ml/g)
Equiv.
Days
Extracted
(yg/g waste)
Percentage
Extracted
970.
2.4
190
620
2,760.
56.
Measurement
Initial
Final
Estim.Tot. Extr.
(11 equiv/g)
Conduct, (y mho)
PH
33,300.
2.2
1786. 4480.
3.7
TABLE 6. LEACHABILITY OF SECONDARY ZINC SMELTER SLUDGE
Element
Cd
Pb
Ni
Zn
Initial
Cone.
(yg/ml )
116.
9.5
0.66
605.
When Concentration Levels Off Total Weight
Cone.
(yg/ml )
0.08
7.5
0.6
21.
Extr.
Nr.
7
7
5
7
Extr. Vol,
(ml/g)
190
190
47
190
, Equiv,
Days
620
620
150
620
, Extracted
(ug/g waste)
410.
1,540
5
11,600
Percentage
Extracted
75.
2.2
1.4
3.0
Measurement
Initial
Final
Estim.Tot.Extr.
(y equiv/g)
Conduct, (y mho)
PH
9,524.
6.4
148.
6.0
886.
The column labeled Initial Concentration
refers to the concentration in the first
extract. Next, the point in the series of
extractions when the concentration in the
extract levels off is identified. The
equivalent leaching time is taken from
Table 4 and is offered as a comparative
index of the rate of leaching. The next
column in the table gives the total weight
of the element extracted per gram of waste.
This is calculated up through the extraction
prior to the one in which the lower detec-
tion limit (LDL) was reached or up through
the seventh extraction if samples remained
above the LDL. The last column in the table
gives the percentage this weight represents
of the total weight of an element in the
waste.
The bottom portion of these waste
Teachability tables gives the conductance
and pHinthe first and seventh waste extracts.
The specific conductance was measured for
each sample because it is a convenient
indicator of the solubilization of mater-
ials from a waste or soil. Multiplying the
specific conductance by 0.01 yields an
estimate of the number of microequivalents
of dissolved solids per milliliter for many
waters (12). Multiplying the microequiv-
alents /mill il iter by the sample volume
gives the total microequivalents of dissolved
ionic species present in the extract. The
figure given in the table is the cumulative
sum calculated from all seven extracts.
Although its accuracy will vary depending
upon the equivalent conductances of the
mixture of ions present in the solution,
this figure provides an estimate of the
amount of waste that dissolved.
For each element of interest a graph
was prepared showing its Teachability from
the waste. (See for example Figures 9 and
73
-------
10.) The y-axis gives the concentratton,
mlcrograms/milliliter, of the element found
in the extract. The cumulative volume scale
on the x-axis is accompanied by a scale
giving the calculated correlation with time,
showing the equivalent number of years of
exposure to a source of water which is
moving through the waste Into the underlying
soil at a pore water velocity of 1 x 10-5
cm/sec. The lower detection limit for the
assay is indicated to the far right for
guidance in evaluating the significance
of the histogram bar height. Above the
histogram bar is printed the total weight
of the element liberated by that extraction.
This is expressed as micrograms per gram of
waste, which is equivalent to grams per
metric ton of 2204.6 pounds. If and when
the lower detection limit is reached, no
weight is given because the result in this
region are undefined: they can be anywhere
in the range from zero to the weight
calculated from the detection limit.
Each waste characterization histogram
is followed by a set of histograms (one
for each extraction) which compares the
fraction of that element retained from the
waste leachate by the three soils for three
different son-to-waste ratios (1:1, 2:1,
and 3:1). The height of the histogram
bar and the value printed on 1t show the
fraction retained from the cumulative total
challenge up to the point represented by
that extraction (calculated as in column I,
Table 4). If the soil gives up that element
then the negative value is printed under
the space corresponding to the appropriate
layer of soil. The soils tested are
designated by the letter C for Chalmers,
D for Davidson, and N for Nicholson soil.
For example, the results of taking the
leachate from the extraction of zinc
secondary refining sludge, and challenging
the soil batches are given in Figure 11,
The first bar in the first histogram
of extract 1 (EXTR. 1) shows that 54 percent
of the cadmium present in the first extract
had been retained by Chalmers soil at a 1:1
soil-to-waste ratio. When this resulting
solution was placed on the second batch of
soil, a total of 86 percent of the cadmium
was removed by passage through what is now
a 2:1 soil-to-waste ratio. The solution
from the second batch was placed on the
third batch of soil. A total of 97 percent
of cadmium was retained after passage
through what is now a 3:1 soil-to-waste
ratio. As the extraction proceeded and the
soil batches were placed in contact with the
subsequent waste extracts, less and less of
the total cadmium extracted from the waste
was retained by the soil. By the seventh
extract not only had all the cadmium pre-
viously retained by the first soil batch
at 1:1 been released but some of the cadmium
normally present in the soil also had been
released, as shown by the-6 percent. Even
the second and third soil batches released
significant quantities of the previously-
retained Cadmium by the seventh extract,
and could conceviably have released all of
it if the leaching had proceeded further
or if a different waste extract was then
used to challenge the soils to simulate
co-disposal.
The histograms of Figure 11 provide
a direct comparison of the ability of three
soils to attenuate one ion from the
particular waste as the leaching proceeded.
The soil's behavior toward that ion would
be expected to be different if the ion were
present in a waste of different composition.
A soil modified by prior exposure to waste
of the same or different composition would
also behave differently.
In addition to the above-mentioned
waste characterization curves and the
summarizing soil-retention histogram, a
series of tables and graphs were prepared
that detail, for each kind of soil, the
results obtained for each element extracted
from a waste. A table of values calculated
from the batch extraction data in the
manner shown in Table 8, was printed out
for each soil. From this data, sets of
histograms can be prepared that give the
weight, micrograms per gram waste or soil,
of element observed in the extracts from
the waste (designated by histogram bar W)
and from each batch of soil (I, II, III)
for each of the seven extractions. This
latter histogram thus shows the amount of
element penetrating or released from the
soil. Figure 12 is an example showing
cadmium leached from zinc secondary-refining
sludge and passing through Chalmers soil.
The analytical detection limits are indic-
ated to the right of this type of histogram,
but here the position of the arrow specifies
micrograms of the ion per gram of waste or
soil to be consistent with the units on the
histogram. (The liquid-to-sol id ratio was
kept the same for W, I, II, and III so that
results could be expressed as micrograms
per gram of etther waste or soil.) Because
74
-------
c o
-------
EXTR
2 10.
O
EXTR
5 in.
O
51
HU
112111 I 1 7! 31 117111
C D N
C - Chalmers
D - Davidson
N - Nicholson
1 1 -'} Jl 11 7 ! 31 112111
C D N
jug/g
o ug/g
EXTR
1 81
EXTR
5 51
EXTR
2 §
EXTR
6
101 83.2
rH""u
EXTR
3
8-1
J07 97.0 79.5
w i i nr
EXTR
4
w i n m
Figure 11. Comparing Fraction Cadmium
Retained by Soils from Zinc
Secondary-Refining Sludge
Leachate.
the detection limits are given as a weight
of element per weight of waste or soil,
they increase with each succeeding extrac-
tion, because increasing volumes of solu-
tion are used to extract a fixed weight of
waste or soil. This is not apparent in
Figure 12 because of the scale required to
depict the high concentration of cadmium.)
For this reason the histogram bar height
will increase as the extraction proceeds,
for the case when the output concentration
becomes constant, because a given concent-
ration in a larger volume represents a
greater weight. So it is necessary to note
that these soil histograms are expressed
in terms of weight of element per unit
weight of waste or soil. (The corresponding
concentrations in micrograms per milliliter
can be obtained from the accompanying tables.
The histogran identified as EXTR.l in Figure
12 shows that the 1st time the sample of
zinc secondary-refining cludge was extracted,
the solution from W contained 232 micrograms
cadmium per gram of waste. This solution was
nixed with the first batch (I) of soil. (A
flowchart of the serial batch extraction
procedure is shown in Figure 6 for the first
two extractions.) When the solution was
filtered, it contained 107. micrograms
Figure 12. Weight of Cadmium from Zinc
Secondary-Refining Sludge on
Chalmers Soil.
cadmium per gram of soil contacted by the
solution; the difference between W and I
had been retained by the soil. The solution
filtered from soil batch I was mixed with
soil batch II for a predetermined length of
time. After filtering II, analysis of the
filtrate showed that soil batch II had
reduced the solution concentration to 33.3
micrograms cadmium per gram of soil contacted.
This corresponds to the concentration pene-
trating a second layer of soil . The
difference, 107 minus 33.3 or 74 micrograms
cadmium per gram, was retained by the soil.
The solution from II was mixed with soil
batch III and the solution concentration of
the resulting filtrate was 6.25 micrograms
cadmium per gram. Of the initial 232. micro-
grams cadmium per gram in the first extract,
only 6.25 micrograms cadmium per gram
penetrated (i.e., was not retained by) the
three soil batches. The weight of waste and
soil were chosen so that this corresponds
to the penetration through the amount of
soil equivalent to a 3:1 soil-to-waste ratio.
Although analytical variations will be
responsible for some of the differences with-
in sequences, the progressing waste extrac-
tion and the passage of the resulting solu-
tion through the soil continually changes
the environment and the soil. Thus, what
76
-------
TABLE 7. CADMIUM FROM ZINC SECONDARY-REFINING SLUDGE ON CHALMERS SOIL.
AMT.PENETR. AHT.RETD, CUH.1Q1, CUM,TOT, FRACTION REID.
DISTRIBUTION COEFFICIENTS
EXT,
NR, LAYER
i tf
I
II
III
I+II
I+II+III
2 U
I
II
III
I+II
I+II+III
3 U
I
II
III
I+II
I+II+III
4 U
I
II
III
I+II
I+II+III
5 W
T
4
I]
III
I+II
I+II+III
6 U
I
II
III
I+II
I+II+III
? U
I
II
III
I+II
I+II+III
,UG/ML
115,9(1
S3. 68
16,47
3,13
33.71
27.73
11,51
5.86
5,67
5,28
c oc
1.77
1,47
2,37
1.31
,71
,34
.80
,60
.25
.22
1.26
1.41
°5
.08
i.ll
1,01
,93
UG/G
231.79
107.36
33,33
6,26
101,13
93,17
34.54
17.57
34.03
31.21
31.51
16,60
17,78
28,48
15,76
8,48
8,12
19.15
14,30
6,06
10.42
60,60
67,87
12.12
7.27
106.66
?6,96
79.51
THIS EXT,
iJG/G
124.43
74,03
27,97
99,23
75.18
17.95
48.63
16.97
33.29
27.85
2.82
-,30
20,91
1,26
7.81
-10,79
12,73
7,27
,97
3,07
-11,03
4,85
8.24
-3, 89
,69
-50.18
-? °?
55.75
-28.72
-.57
-99.38
9.70
17.45
-44,84
-24, 08
CHALLG,
UG/C
231,79
107,36
33,33
115.90
77,26
332,92
190.54
67.87
166.46
118.97
366,95
221,75
99,38
183,47
122,32
384,64
256,23
115,14
192,32
128,21
392.76
269,38
129,44
196,38
130,92
403.19
329,98
197.31
201.59
134,40
410.46
436,63
294,27
205.23
136,82
RETD.
JL'C/G
124,43
74,03
27,07
99,23
75,18
142,38
122.66
44,04
132,52
103,03
145.20
122.36
64,94
133,78
110,84
134,42
135,89
•n •»
134,75
113,91
123,39
139,94
80,46
131,66
114.59
73,21
132. 6S
136.21
102.94
114.83
-26,17
142,36
153,66
58,09
89,95
THIS
EXTR,
,54
.69
,81
,86
,97
,18
.58
,49
.66
,83
,08
-.01
.66
,07
,69
-,61
,45
.46
,11
,52
-1,36
,'J5
,58
-.76
,25
-4,81
-.13
,82
-5,51
-.16
-13,67
.09
,18
-12,33
-9.93
TOTAL
CHALLG.
,54
,69
,81
,36
,97
,43
,64
.65
.80
,93
,40
,55
it.
i U_f
.73
,91
,35
,b4
,63
,70
00
, W J
11
,52
,62
,67
,88
,18
,40
.69
,51
.35
-.06
,33
,52
,28
.66
PENETR,
FACTOR
.46
11
1 Ul
,19
.14
,03
,82
.42
.51
,34
.17
.92
1.01
,34
,93
,31
1.61
,55
.54
,89
,48
2,36
,75
.42
1.76
.75
5.81
1,12
,18
6.51
1.16
14,67
.91
,82
13.33
10.93
INCL SOIL
RATIO
1,16
2,24
4,40
6,01
36,73
1.72
1 C1?
W 1 -II
1 Cl
i. , .* J
7 73
17,84
4,67
3,90
6,17
8,55
31,78
4,74
S.il
y.57
*'; 07
40,81
6.47
0 OO
13,36
18.55
5? , 47
1.22
l,9i
11.28
3,06
28,60
DEC.
49,33
65,91
77.20
30,56
88,44
59,80
74.33
68,47
82,63
86,79
77.91
75,61
80,80
83,33
86.20
78.08
83,37
83,35
86.68
88,60
81,21
84,19
85.72
86.91
89,80
58.57
62,99
84.93
71,92
88,00
-.24-13,53
1,47
1,94
1.2°
3.45
55.83
62,72
50,63
73,84
SOLN ONLY
RATIO DEG,
1,16 4?, 21
2,22 65.76
4,32 76,97
5.95 80,47
36,02 88,41
1,71 59,71
3.55 74,27
2.51 6S.24
7,67 82,57
17.59 86.75
4,65 77.87
7 00 1C CL
ij i UU r J t -*iJ
6.12 80,73
8,49 83,28
31,35 88,17
4,72 78.04
8, 57 83,35
3,51 83,30
17,10 86,65
48,29 88,58
6,44 81,18
9.78 84,16
13,28 85,69
18,41 86,89
56,73 88,99
1,21 50,38
i 1 95 62 9i
11,21 84,92
3,03 71,75
28.22 87,97
-.25-13,79
1,4? 55,74
1.93 62.64
1,20 50.15
3,39 73,58
77
-------
happens in one batch is not necessarily an
indication of what to expect fn the next,
(This unpredictability is expressed by the
varying concentrations and changing slopes
of the isotherm segments shown in Figure
13.) In many cases a chromographic type
"peak" can be clearly seen to move though
soil batches I, II, and III, In other
cases the peak remains "submerged" and is
discernable only as a wave of lowered
retention progressing through the soil as
the extractions are continued.
180
"5> 60
d 140
O
(ft
in 120
UJ
S 100
o 60
z
T> 6°
O
Z 40
O
(-
< 20
I 0
O
0-20
-40
V
20 40 60 BO 100
CONCENTRATION Cd IN LIQUID PHASE, ug/g Soil
Figure 13, Segments of Adsorption
"Isotherms" Showing the Effects
of Soil Depth (Layers I to III)
and Extent of Leaching
(Extractions 1 to 7) on Cadmium
Distribution.
CONCLUSIONS
If the interest in obtaining equations
which describe a complex system is not for
theoretical purposes but is for practical
applications like predicting a response
resulting from known levels of given vari-
ables, then obtaining purely empirical
equations should be a more straight-forward
approach. But caution is still required if
the equations are to describe a field
situation. Equation parameters should be
derived from experiments that include all
of the factors having significant effects.
The design of the experiment should be
such that the scope of its applicability
and its relation to an actual field
situation will minimize the number of
assumptions needed in the corresponding
prediction equations. This will improve
the validity of the predictions. For
example, the changes in K show that the
experiments should employ an actual changing
waste leachate so that predictions will not
have to be based on the assumption of a
constant composition or so that the nature
of the changes will not have to be assumed.
The closer the laboratory simulation
approaches a field situation, the fewer
additional assumptions that will have to be
applied and the more reliable the predictions
likely will be. Rapid experimentation which
is flexible enough to simulate each new
field situation even makes case-by-case
experimental determinations practical, thus
reducing to a minimum the dependence upon
the ability of an equation to extrapolate.
Except for helping to identify potential
experimental factors, it is our opinion that
mathematical models should not be intro-
duced too soon in an attempt to construct
an expression meant to predict the behavior
of a complex physical-chemical system.
Experimental examination of the effects of
impressed variations can (if the experiments
are suitably designed) take into account the
effect of unidentified (unknown) factors.
Afterward, in the application of these
results, mathematical compensation may be
introduced to correct for those factors
which were not included in the experimental
set-up (such as upward capillary flow,
horizontal spreading, etc). Additional
experimentation may be required to quantitat'e
these effects. The final description of the
predictions should be in a mathematical
framwork which yields output weight or
concentration in terms of time, volume, etc.,
upon plugging in values for all factors which
differ from those employed in the experi-
mental determinations or those effects which
otherwise need correction (like experiments
on a saturated system being applied to
predictions on unsaturated soil).
In this connection, some comments are
in order concerning the nature of the pro-
posed graded serial batch extraction pro-
cedures :
a. Batch extractions have been con-
sidered to be of zero dimensionality. (13)
The batch extraction method i_s_ independent
of soil dimension in one sense, but the
relation employed in the design of the
experiments reported here yields results
expressed per unit volume of soil, i.e.,
in three-dimensional space. In the
application of these results to the field,
the effect of horizontal spreading (how the
space occupied by the volume changes with
78
-------
time) can be taken into account.
b. Even though an extraction can be
classified as static with respect to liquid-
front movement, the sequential extractions
proposed here simulate a stepped dynamic
situation so a steady state need not be
assumed for prediction equations in which
these data are applied. The proposed
batch extraction experiments thus should
provide very useful input data for a
variety of transport models. (Predictions
are better based on laboratory simulations
than upon completely mathematical "simul-
ations".) Additional soil batches (i.e.,
more than three) could be used to better
define what happens with increasing pene-
tration depth into the soil.
c. The serial batch experiments pro-
vide direct information for saturated-only
transport models. Corrections must be
inserted if there is a significant diff-
erence caused by unsaturation. (Comparison
between batch and column showed no diff-
erence for most ions studied.) (2) The rate
of movement could also differ greatly with-
out causing significantly different adsorp-
tion equilibria, i.e., the K values may be
sufficiently alike so long as equilibrium
is attained.
d. The calculated time equivalencies
listed in Table 1 are for the idealized
case of a uniformly-packed bed of soil
(similar to that prepared in a column or
for a lagoon lining). Inhomogenities in
composition (such as steaks of sand in a
bed of clay) or fissures produced by, e.g.,
rotten roots, will not only greatly increase
the liquid front velocity in those zones
but the resulting channeling can signifi-
cantly reduce the effective mass of soil
contacted by the waste extract. For example,
if channeling is bad, 70 percent of the
liquid may leak through and contact but a
small fraction of the soil, while only the
remaining 30 percent of the waste leachate
would be available to percolate through
the bulk of the soil. Corrections can be
readily included for this kind of deviation
from ideality if their relative magnitudes
can be estimated for the site of interest.
Another situation that would affect the
liquid-front velocity presumed for a given
bed of soil, one which would require
separate flow-rate column to detect, is
the case where the waste leachate itself
affects the flow through the soil, either
by plugging the pores and reducing the flow,
or by affecting the soil structure and
drastically increasing the flow-rate (this
was observed with flue gas waste on
Davidson soil (14)).
e. It is recognized that in the field
very slow processes may contribute to the
net retention or even change the conditions
under which retention is occurring (like
micro-biological modification of the
leachates and/or soils). The relative net
effect of the slow to the fast processes
during the time of contact will determine
their significance. If slow processes have
a significant proportional effect, it may
not be possible to accelerate the testing
by reducing the contact time below the
residence time calculated for a given liquid
front velocity. However, it is also not
desirable simply to wait for equilibrium
to be established if this requires longer
than the field residence time for a slug of
liquid of a given composition.
A change in the distribution coef-
ficient shows a shift in the equilibrium.
The change in K could be plotted versus pH
or hydrogen ion concentration, the concen-
tration of other ions, or against the
measured values observed or calculated
for the slope, or other responses of con-
cern could be regressed against the level
of selected experimental factors to see if
a simple relation exists between a response
and a factor, (i.e., test for or derive a
relationship between the measured or cal-
culated parameters and the experimental
variables.) But, it is necessary to be
careful when drawing conclusions in this
way. What is being done with the data to
relate factors (variables) may be analogous
to classical vary-one-factor-at-a-time
experimentation. If so, the conclusions
can be very far off if interactions exist
between the factors being plotted or
examined numerically. The only way to take
interaction into account is to run the
experiments as factorial experiments and
then derive the relations between the
statistically significant factors and
interactions using regression analysis.
The effect of many kinds of variables can
be determined by relatively small pertur-
bations superimposed upon the total simulated
field conditions, as by using the method of
additions to study the effect of the con-
centration of Ca++ or other ions. Other
factors, like temperature, surface area,
soil type, etc., can be readily included.
A number of factors can be included
79
-------
simultaneously with factorial experiments.
(Models might be of help to point out
possible-important factors that should be
included in the experiments, but mental
imagery, not mathematical formulations,
ultimately provides the input for deciding
what variables to include in experiments.)
If the number of potential factors is large,
it may be desirable to run screening
experiments like main-effect factorial
experiments first, possible followed by
fractional factorials.
Even with the best kind of designs,
the magnitude of the effect of only those
factors purposely varied can be learned
from the experiment, but some other factors
may have an effect and even interact with
the experimental factors. But if any
unidentified factors can be kept at the
same levels as they occur in the field
(such as by using the same soil throughout
the experiment because some unknown soil
properties may be significant factors),
then the effect of the known factors will
be correctly estimated even if interaction
does exist between them and the unident-
ified factors.
The graded serial batch extraction
procedure provides the flexibility reli-
ability, speed, and ease of interpretation
needed to provide data for the derivation
of empirical equations which contain a
minimum of pitfalls for predicting the
movement of hazardous substances through
soil.
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.
REFERENCES
1. Houle, M. J., Grabbe, R,, Soyland, J.,
Bell, R,, and Lee, H., "Migration of
Hazardous Substances Through Soil,"
December 1974, Nine Month Progress Report,
Prepared for Solid and Hazardous Waste
Research Division, Municipal Environmental
Protection Agency, Cincinnati, Ohio 45268,
Contract Number EPA-IAG-04-0443, pages 20,
28, 33, 40, 46.
2. Houle, M. J,, Long, D., Bell, R.,
Weatherhead, D. C,, Jr., and Soyland, J.,
"Correlation of Batch and Continuous
Leaching of Hazardous Wastes," Proceed-
ings of a National Conference About
Hazardous Waste Management, February 1977,
San Francisco, California, Collins ed.,
Prepared for US Environmental Protection
Agency, In Press.
3. Liskowitz, J. W., et al, "Evaluation of
Selected Sorbents for the Removal of Con-
taminants in Leachate from Industrial Sludges,"
Residual Management by Land Disposal, Pro-
ceeding 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 Environ-
mental Research Laboratory, US Environmental
Protection Agency, Cincinnati, Ohio 45268,
page 162,
4. Farquhar, G, J., and Rovers, F. A.,
"Leachate Attenuation in Undisturbed and
Remolded Soils," Gas and Leachates from
Landfills: Formation, Collection, and
Treatment, Proceedings of a Symposium at
Rutgers University, 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, page 54.
5. Rovers, F. A., Mooji, H., and Farquhar,
G. J,, ".Contaminant Attenuation-Dispersed
Soil Studies," Residual Management by Land
Disposal, Proceedings of the Hazardous
Waste Research Symposium, February 1976,
The University of Arizona, Tucon, 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, page 224.
80
-------
6. Griffin, R. A., Frost, R. R,, and SMmp,
N. F., Ibid, pages 259-268,
7. Long, D. E., Houle, M. J,, Weatherhead,
D. C,, Jr., and Ricks, G. K., "Migration of
Hazardous Substances Through Soil, Part III,
Development of a Serial Batch Extraction
Method and Application to the Accelerated
Testing of Seven Industrial Wastes",
November 1978, Draft Report Prepared for
Solid and Hazardous Waste Research Division,
Municipal Environmental Research Laboratory,
U.S. Environmental Protection Agency,
Cincinnati, Ohio 45268, Contract Number
EPA-IAC-04-0443.
8. Houle, M. J., Long, D. E., "Accelerated
Testing of Waste Leachability and Conta-
minant Movement in Soils", Land and Disposal
of Hazardous Wastes, Proceedings of the
Fourth Annual Research Symposium, San Antonio,
Texas, March 1978, Municipal Environmental
Research Laboratory, US Environmental
Protection Agency, Cincinnati, Ohio 45268,
EPA-600/9-78-016, pages 152-168,
9. Griffin, R. A., and Shimp, N. F.,
"Attenuation of Pollutants in Municipal
Landfill-Leachate by Clay Minerals",
Solid and Hazardous Waste Research Division,
Municipal Environmental Research Laboratory,
US Environmental Protection Agency,
Cincinnati, Ohio 45268, EPA-600/2-78-157,
August 1978.
10. Long, D. E., and Houle, M. J., "The
Use of a Graded Serial Batch Extraction
Procedure to Evaluate Contaminant Movement
in Soils," 33rd Annual Northwest Regional
Meeting, American Chemical Society, Seattle,
Washington, June 1978.
11. Heftman, E., Editor, Chromatography,
Van Nostrand Reinhold, New York, Third
Edition, page 47.
12. Standard Methods for Examination of
Water and Waste Water. APHA, AWWA, WPCH,
13th Edition, 1971, page 323.
13. Van Genuchten, M. Th., "Simulation
Models and Their Application to Landfill
Disposal Siting; A Review of Current
Technology", Land Disposal of Hazardous
Wastes, Proceeding of the Fourth Annual
Research, Symposium, San Antonio, Texas,
March 1978, Municipal Environmental Research
Laboratory, US Environmental Protection
Agency, Cincinnati, Ohio, 45268, EPA-600/9-
78-016, pages 191-214.
14. Houle, M. J., Long, D. E., Weatherhead,
D. C., Jr., Bell, R. E., Ricks, G. K.,
Soyland, J. E., Griffiths, L., "Migration
of Hazardous Substances Through Soil Part II
Flue-Gas Desulfurization and Fly-Ash Wastes",
November 1978, Draft Report Prepared for
Solid and Hazardous Waste Research Division,
Municipal Environmental Research Laboratory,
U.S. Environmental Protection Agency,
Cincinnati, Ohio 45268, Contract Number
EPA-IAG-04-0443, page 357.
81
-------
DISPOSAL AND REMOVAL OF HALOGENATED HYDROCARBONS IN SOILS
Robert A. Griffin and Sheng-Fu J. Chou
Illinois State Geological Survey
Urbana, IL 61801
ABSTRACT
In this report, the adsorption, mobility, and degradation of polybrominated biphenyls (PBBs) and hexachloro-
benzene (HCB) in soil materials and in a carbonaceous adsorbent were studied, together with the solubilities of
PBBs and HCB in waters and landfill leachates. The aqueous solubilities of both materials were low ( < 16 ppb), but
solubilities were higher in river water and landfill leachate than in distilled water. The solubilities can be directly
correlated with the level of dissolved organics in the waters. By using a soil thin-layer chromatography technique, it
was found the PBBs and HCB were immobile in all soils when leached with deionized water and landfill leachate;
they were highly mobile in all soil materials when leached with organic solvents.
Freundlich adsorption isotherm plots of PBBs and HCB sorption on soils and on a carbonaceous adsorbent
yielded straight and nearly parallel lines. All regression lines generated had a coefficient (r2) of at least 0.98, which
indicated an excellent fit of the data to the Freundlich equation. PBBs and HCB were found to be strongly adsorbed
by the carbonaceous adsorbent and by soil materials, with HCB being adsorbed to a greater extent than PBBs. The
adsorption capacity and mobility of PBBs and HCB were highly correlated with the organic carbon content of the
soil materials.
In a soil incubation study, it was found that PBBs and HCB persisted for 6 months in soil with no significant
microbial degradation. Because of their low water solubilities, strong adsorption, and persistence in soils, these
two compounds are highly resistant to aqueous phase mobility through earth materials; however, they are highly
mobile in organic solvents.
INTRODUCTION
Polybrominated biphenyls (PBBs) and hexachloro-
benzene (HCB) are halogenated hydrocarbons. They
belong to a class of aromatic halogenated organic com-
pounds; these compounds are chemically similar to the
more commonly known contaminant polychlorinated
biphenyls (PCBs) and polychlorinated napthalene (PCN).
In 1973 PBBs were accidently added to livestock
feed in place of magnesium oxide.9 >2 8 >40 This inci-
dent has been called the most costly and disastrous
contamination ever to occur in United States agri-
culture.2 8 The contamination had a catastrophic impact
on the Michigan livestock industry and resulted in
thousands of farm animals being killed and buried.
Altogether approximately 29,800 cattle, 5,920 swine,
1,470 sheep, and 1.5 million chickens were killed and
buried by the end of 1975.9
The commercial production of PBBs began in
1970. Approximately 13.3 million pounds of PBBs were
produced in the United States from 1970 to 1976.3 8
About 11.8 million pounds of this total was hexa-
bromobiphenyl (fireMaster BP-6, FF-1); the remaining
1.5 million pounds consisted of octabromobiphenyl
and decabromobiphenyl. At the present time, no PBBs
are being imported in commercial quantities. On the
other hand, the export of PBBs from the United States
to Europe has increased during the past several years
and totaled 805,000 pounds in 1976.3 8 The major
uses of PBBs were for the production of flame re-
tardant resins of acrylonitrite, bertadiene, and styrene
for business machines, electrical housings, textiles, and
other materials.4 3' '4 All of these uses were dis-
continued in late 1974 as result of the Michigan
incident.14
HCB is both a starting material and a by-product of
the chemical industry.16 Quinlivan, Ghassemi, and
Santy3 9 estimated that 3,909 metric tons of HCB are
generated annually in the United States. Approximately
82
-------
240 metric tons are used in industry; the rest is dis-
posed of by incineration, or as wastes on the land.
There are two major uses of HCB in industry: as a
peptizing agent in the manufacture of styrene and
nitroso rubbers for tires, and as a fungicide for agri-
cultural seed treatment.3 9 HCB is present in industrial
waste as a by-product in several manufacturing pro-
cesses, mainly in the commercial production of various
chlorinated solvents such as carbon tetrachloride, per-
chloroethylene, dichloroethylene, and trichloroethylene.
HCB has also been present as an impurity or by-product
in the production of several pesticides such as dacthal,
mirex, simazine, atrazine, propazine, and pentachloro-
nitrobenzene.3'
Mismanagement of HCB-containing wastes has
resulted in several episodes of environmental con-
tamination in the United States and abroad. The best
documented case of an HCB episode occurred in south-
ern Louisiana.6 >7
Many researchers have reported the effects of
PBBs and HCB on organisms such as fish, birds, and
mammals including humans.24' 41> 35> 3 7 > 4 2' 23>
26,27,45, 12,36,25,31,46 Both PBBs and HCB have
a high bioaccumulation factor in fish.2 4 >3 2 The major
problem with regard to PBBs and HCB intake by man is
that he resides at the top of most food chains. There-
fore, man can take in substantial amounts of PBBs and
HCB even though only trace levels are present in meat,
or in the water of streams and lakes. PBBs and HCB
have been identified as a significant hazard to human
health as well as to the environment;27' 8 therefore,
their disposal has caused great concern. Little informa-
tion is presently available, however, concerning the
mechanisms of attenuation of PBBs and HCB in earth
materials, or concerning the possibility of ground-
water contamination by PBBs and HCB being leached
from landfills.
The limited amount of information presently
available indicates that PBBs and HCB have a fairly
strong affinity for soil. The mechanisms of transport
of PBBs and HCB in the biosphere, and the mechanisms
of attenuation in earth materials are unknown. Data on
the factors affecting PBBs and HCB attenuation by soil
materials, solubility in waters and landfill leachates, and
mobility in soils would provide a useful basis for deter-
•mining waste treatment methods, for predicting PBBs
and HCB migration under landfills, and for selecting and
designing future disposal sites.
The purposes of this project are: (a) to conduct a
literature review of information on the attenuation of
PBBs and HCB in soil materials; (b) to measure the
solubility of PBBs and HCB in waters and landfill
leachates; (c) to measure the adsorption capacity of
selected soils and a carbonaceous adsorbent for PBBs
and HCB; (d) to quantitatively evaluate the effects of
adsorbent composition and of organic solvents on
adsorption, and mobility of PBBs and HCB; (e) to use
this data to develop a mathematical model that will
assist in the prediction of PBB and HCB adsorption.
This research is supported in part by Grant
R-804684-01-0, from the U.S. Environmental Pro-
tection Agency, Municipal Environmental Research
Laboratory, Solid and Hazardous Waste Research
Division, Cincinnati, OH 45268. This is a report of
work conducted as part of that grant, and the results
should be considered preliminary and subject to re-
interpretation if future results require it. Mention of
trade names does not constitue endorsement.
METHODS AND MATERIALS
PBB and HCB Materials
FireMaster BP-6 (lot #6244A) was obtained from
Michigan Chemical Corporation and was used without
further purification. The product contained more than
30 isomers of polybrominated biphenyl (PBB) as
identified in our laboratory using a SE-30 WCOT glass
capillary gas chromatographic column and an electron
capture detector (BCD). Several major isomers were
identified: 2, 2', 4, 5, 5'-penta-; 2, 3', 4, 4', 5-penta-;
2, T, 4, 4', 5, 5'-hexa-; 2, 2', 3, 4, 4', 5'-hexa-; 2, 3',
4, 4', 5, 5'-hexa-; 2, 2', 3, 4, 4', 5, 5'-hepta-; and 2, 2',
3, 3', 4, 4', 5, 5'-octabrominated biphenyl.43'29' 13
Uniform '4C-labeled-PBB (lot #872-244) was
used in some studies and was synthesized and purified
according to our specifications by New England Nuclear
Corporation, Boston, Massachussetts. This product
contained the two major isomers of fireMaster BP-6,
approximately 65 percent 2, 2', 4,4', 5, 5'-hexabromo-
biphenyl, and 35 percent 2, 2', 3, 4, 4', 5, 5'-hepta-
bromobiphenyl. The specific activity of the '4 C-PBB
was 9.34 mCi/mmole.
Hexachlorobenzene (HCB) was purchased from
Aldrich Chemical Company, Inc., Milwaukee, Wisconsin.
This product was recrystalized from distilled-in-glass
hexane, and this process was repeated several times
until the purity reached nearly 100 percent.
Uniform '4 C-labeled-HCB (lot #852-058) was
purchased from New England Nuclear Corporation,
Boston, MA. The specific activity was 35.5 mCi/mmole
(.125mCi/mg).
83
-------
Adsorbents
Six earth materials and a carbonaceous adsorbent
representing a wide range in characteristics were selected
as adsorbents. The materials being studied are: Hough-
ton muck, Catlin sil, Flanagan sicl, Ava. sicl, Bloomfield
Is, Ottawa silica sand, and Ambersorb^ XE-348. The
carbonaceous adsorbent was selected because of its
excellent performance characteristics in the removal
of organic compounds at low concentrations.
Waters and Leachates
Distilled water, deionized water, Sugar Creek water,
Blackwell landfill leachate, and DuPage landfill leachate
were selected for use in the solubility study. Leachates
were centrifuged through a continuous flow centri-
fugation apparatus (model JCF-Z, Beckman Instru-
ments) at approximately 17,000 rpm, prior to passing
through a 0.22 yum pore size Millipore cellulose acetate
membrane. The Sugar Creek water was also passed
through the 0.22 jum membrane before use.
Analytical Development
In general, PBBs are determined quantitatively
by comparing gas chromatographic (GC) response
patterns of a multicomponent environmental sample
with the GC-response patterns of commercial PBBs.
In practice, quantitative results are usually achieved
by comparing the integration of the GC peaks obtained
from the unknown mixture with the integration of the
GC peaks obtained from a standard PBB mixture.
A PBB mixture contains a large number of isomers,
and it was not possible to obtain a standard containing
all isomers observed in this work. Therefore, the re-
sponse of each PBB isomer relative to the response of
an internal standard (hexabromobenzene) was used for
quantification.
To accomplish this, an assumption was made that
the response of the flame ionization detector (FID) was
the same for all of the isomers of fireMaster BP-6.
Such an assumption is plausible because each isomer
has an equal number of carbon atoms. Thus, the per-
centage of each PBB isomer relative to the internal
standard in a mixture of PBBs of known concentration
was determined by GC using an FID. From these per-
centages and the same internal standard the response
factors for individual PBB isomers were calculated for
the electron capture detector (BCD). The BCD was used
because it is specific to halogenated compounds and
gives much lower detection limits for these compounds
than can be achieved using FID. The results of the
analysis showed that, in all cases, the overall standard
deviation of error for PBB analysis ranged from 2.75
to 3.76 percent.
For the quantitative analysis of HCB, the same
technique was used, except tribromobenzene was used
as an internal standard. The standard deviation of
error for 0.03 ppm and 0.05 ppm standards was less
than 1 percent.
The samples and standards were analyzed on a
Perkin-Elmer Sigma I gas chromatograph (GC) equip-
ped with electron capture detection and flame ioniza-
tion detection capabilities. The conditions for the gas
chromatographic analysis for PBBs and HCB are shown
in Table 1.
Solubility Studies
In the study of solubilities of PBBs and HCB in
waters and leachates, 0.5 gm and 0.1 gm of fireMaster
BP-6 or HCB were placed in 1-liter Erlenmeyer flasks.
The flasks were then filled with distilled water, de-
ionized water, Sugar Creek water, DuPage leachate,
or Blackwell leachate. The mixture in each flask was
agitated vigorously with a magnetic stirring bar coated
with teflon. Water samples were collected after 2, 4,
and 7 days for PBBs and 2, 7, and 30 days for HCB.
Solubility measurements were conducted by
filtering 250-mL aliquots of PBB- or HCB-water solution
through PBB- or HCB-saturated Millipore membranes
(0.
TABLE 1. CONDITIONS FOR GAS CHROMATOGRAPHIC
ANALYSIS
Conditions
PBBs
HCB
Column
Injector
temperature
Column
temperature
Detector
temperature
Carrier gas
6 ft x 2 mm I.D. glass
column (or stainless
column for FID). 3
percent SE-30 on 80/100
mesh chromosorb WHP
270 C
250 C
300 C
6 ft x 2 mm I.D. glass
column, 5 percent
OV-17 on 80/100 mesh
chromosorb WHP
250 C
200 C
300 C
Methane/argon (helium Methane/argon, flow 35
for FID), flow 40 mL/min mL/min
84
-------
A separate study had shown that the adsorption
capacity for PBB or HCB of the cellulose acetate mem-
brane used to filter the PBB or HCB solution could be
saturated by soaking the membrane in a fresh SOO^nL
aliquot of PBB- or HCB-saturated water each day for
3 days, followed by flushing with 200 mL of PBB or
HCB saturated water.
Two 100-mL aliquots of the filtrates were then
extracted with 10, 5, and 5 mL of water-saturated
hexane. Hexabromobenzene or tribromobenzene
(0.25 us) was added as an internal standard to the PBB
or HCB extracts, respectively. The extracts were then
concentrated to I'.O mL and analyzed by GC. A re-
presentative chromatogram is shown in Figure 1.
The solubilities of major PBB isomers and HCB in
waters and leachates are shown in Tables 2 and 3
respectively. The results show that the solubility of
PBBs and HCB was very low. The average solubilities
of PBBs in distilled water, deionized water, Sugar
Creek water, DuPage leachate, and Blackwell leachate
were 0.057 A
-------
TABLE 3. SOLUBILITY OF HCB IN WATERS AND LEACHATES
Waters and
leachates
Distilled water
Deiomzed water
Sugar Creek water
Du Page leachate
Blackwell leachate
Concentration (pg/D*
2 days
1.77
1.83
2.43
4.17
4.58
7 days
1.71
1.67
2.35
4.04
4.29
30 days
1.75
1.78
2.22
4.14
4.47
•Each value is the mean of 2 replicates.
A similar finding on PBBs had also been reported
elsewhere.30 These results indicate that PBBs were
greater than 200 times and HCB greater than 2.5 times
more soluble in landfill leachates than in "pure" waters.
The type of dissolved organic matter is also important
in determining how soluble the compound will be in
a given water. These factors should be taken into
account when attempting to predict the migration of
these compounds from waste disposal sites.
Adsorption Studies
Equilibrium adsorption studies were carried out
by shaking known volumes of PBB and HCB solutions
with varying weights of soil materials and carbonaceous
adsorbent at a constant temperature of 22° C. The rate
of adsorption of PBBs and HCB by soil materials and
400-
300-
40-
30-
PBB IN ETHANOL
PBB IN HEXANE
PBB IN
ETHYL ACETATE
001 002 0.04006 01 0.2 0.304 06
CONCENTRATION OF PBB (PPM)
ISGS 1979
Figure 2. Freundlich adsorption isotherms of PBB and 14C-PBB adsorp-
tion on Ambersorb XE-348 from organic solvents at 22°C ± 1°C.
Open symbols indicate analysis by GC.
30.000
20,000
40°0
- 3000
5
x 2000
1000-
400
300-
MUCK
CATLIN
~l 1—I I I I I I—
346 10
—I 1 1—I I I I I I
20 30 40 60 100
CONCENTRATION OF PBB IN
BLACKWELL LEACHATE (PPB)
ISGS 1979
Figure 3. Freundlich adsorption isotherms of PBBs on three soil materials
from Blackwell leachate at 22°C.
carbonaceous adsorbent was rapid. Equilibrium con-
ditions were achieved in less than 4 hours. The adsorp-
tion data were fitted by linear regression to the log
form of the empirical Freundlich adsorption equation:
log£=logK+ 1/nlogCeq
where x = ^g of compound adsorbed, m = weight of
adsorbent (g), Ceq = equilibrium concentration of the
solution (jug/mL or ng/mL), and K and 1/n are
constants.
The use of this relation allows quantitative pre-
dictions of PBBs and HCB adsorption—by a given
adsorbent—over the concentration range of water-,
leachate-, and organic solvent-soluble PBBs and HCB.
Freundlich adsorption isotherm plots of PBBs
and HCB sorption on Ambersorb XE-348 and on each
soil type yielded straight and nearly parallel lines. The
adsorption isotherms for PBBs are shown in Figures 2
and 3. Similar isotherms were obtained for HCB sorp-
tion. Values of K and 1/n were obtained from the
calculated linear regression equations as the intercept
and slope, respectively, of the adsorption data. The
calculated Freundlich constants are shown in Tables 4
and 5. All regression lines generated had a coefficient
(r2) of at least 0.98, which indicates an excellent fit of
the data to the Freundlich equation.
PBBs and HCB were found to be strongly adsorbed
86
-------
TABLE 4. FREUNDLICH K (/Jg/g), 1/n, AND REGRESSION CO-
EFFICIENT, r2 FOR THE ADSORPTION OF 14C-PBB
AND 14C-HCB FROM VARIOUS ORGANIC SOL-
VENTS BY AMBERSORB XE-348
Freundlich parameters
K
1/n
Solvents
Ethanol
Hexane
Ethyl acetate
PBB HCB PBB HCB PBB HCB
578 2690 0.64 0.71 0.9982 0.9945
389 625 0.88 0.69 0.9995 0.9989
98 804 1.03 0.80 0.9929 0.9977
Ethyl dichloride ND 228
ND
0.70
ND 0.9924
ND - Not determined.
by Ambersorb XE-348 and by soil materials. In all
studies, HCB had a greater adsorption tendency than
PBBs. The adsorption of PBBs and HCB for the three
soils followed the series, muck > Catlin > Ava.
The PBB adsorption constant (K) plotted as a
function of TOC gave a very high correlation. Similar
results for PCBs were also reported elsewhere.3 3 The
calculated linear regression relation was:
K = 64.92+17.57 TOC
r2 = 0.999
A similar result was also found for HCB on the same
adsorbates. The HCB adsorption constant (K) in log
form plotted as a function of TOC also showed a very
high correlation. The calculated linear regression re-
lation was:
log K = 2.01+0.10 TOC
r2 = 0.999
The adsorption of hexane- and ethanol-soluble
TABLE 5. FREUNDLICH K (M9/g),1/n, AND REGRESSION CO-
EFFICIENT (r2) FOR THE ADSORPTION OF PBB AND
14C-HCB BY VARIOUS SOILS FROM AQUEOUS
SOLUTION
Freundlich parameters
1/n
Soils
PBB HCB PBB HCB PBB HCB
HoughtonMuck 361 4193 1.99 1.12 0.9918 0.9985
Catlin si) 144 276 1.89 0.99 0.9820 0.9906
Avasicl 88 136 1.77 0.93 0.9790 0.9880
.£0-1 It ict
14 C-PBB and '4 C-HCB bjMCatlin and muck soils was
also investigated. The results showed that virtually no
adsorption took place. This resultjstrongly indicates
tfiat potential migration of PBBs and HCB in a landfill
could o_ccur_if the PBB ajjcTHCB "waste's were dissolved
in organic solvents or iforganic solvents we're disposed
ofin the same landfill-location.
Mobility Studies: Determination by Soil TLC
The technique of determining pesticide mobility
in_soils by soil TLC was introduced by Helling and
Turner.20 Since the introduction of the technique,
the mobility of a large number of pesticides in a variety
of soils have been tested.2''2 2 >2 3 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 methods used to prepare soil TLC plates has
been reported previously by Griffin.19 In these studies,
the 14C-PBBs, 14 C-HCB, and non-14C-labeled-PBBs
and HCB were spotted 2 cm from the base and leached
12 cm with an acetone, acetone/water mixture (1:1,
v/v) and DuPage leachate; they were also leached
10.5 cm with water, methanol, and dioxane. For the
fireMaster BP-6 mixture, the plates were leached 10.5
cm with acetone/water (1:1, and 8:2, v/v), acetone,
and carbon tetrachloride. The plates were immersed in
0.5 cm of the various solvents in a closed glass chamber
and were removed when the wetting front reached the
12 cm or 10.5 cm (distance) line. The soil plates were
removed and air dried. Soil increments of 1 cm, starting
1.5 cm above the base of the plate, were scraped off.
The soils were placed in a glass centrifuge and/or test
tubes and extracted with suitable organic solvents.
The concentration of '4 C-labeled compounds was
determined with standard liquid scintillation techniques.
With non-14 C-labeled compounds, the concentration
in the extracts was measured by GC analyses.
The Rf values are shown in Table 6. The results
show that under the conditions tested, PBBs and HCB
were immobile in these earth materials when leached
with water and DuPage leachate, but were highly mobile
when leached with organic solvents. The mobility of
PBBs and HCB in earth materials was directly pro-
portional to the solubility of the compounds in the
leaching solvents and indirectly proportional to the
soil organic matter contentTThelnoBility of PBBs and
HCB in the tested earth materials was: Ottawa sand >
Bloomfield sand > Ava soil > Flanagan soil > Catlin
soil. A similar finding for PCBs was reported else-
where. 18 It was also observed that PBBs were extremely
mobile in Catlin soil when leached with carbon tetra-
chloride.
ll^
-------
TABLE 6. MOBILITY OF PBBs AND HCB IN SEVERAL SOIL MATERIALS LEACHED WITH VARIOUS SOLVENTS AS MEASURED BY
SOILTLC
Soil
materials
Catlin sil
Flanagan sicl
Ava sicl
Bloomfield Is
Ottawa s
Water
PBB HCB
0.00 0.00
ND
0 00 0.00
0.00 0.00
Du Page Acetone/water
leachate (1 1,v/v) Methanol Acetone Dmxanp
PBB
0.00
000
0.00
0.00
000
HCB PBB
0.00 0.01
0.00
0.00 0.02
0.00
0.00 0 1 7
HCB PBB
0.00 0.40
0.60
0.01 0.61
0.86
0.02 1.00
HCB PBB
0.40 0.69
0.44
0.48 0.76
0.72
0.99 0.99
HCB PBB
0.45 1 .00
1 00
0.80 1 .00
1 .00
1 .00 1 .00
HCB
0.99
1.00
1.00
1.00
1.00
•Computed from statistical peak analysis of data by using values of first moment for grouped data.
ND = Not determined.
PBBs and HCB are non-polar by nature and have
very low solubility in water. Measurements of solubility
of PBBs have yielded values of 0.497 and 16.892 ppb in
river water and leachate, and the solubility for HCB to
be 2.22 and 4.47 ppb in river water and leachate;
however, PBBs and HCB are very soluble in organic
solvents such as dioxane, carbon tetrachloride, acetone,
methanol, etc. It is interesting that the mobilities of
PBBs and HCB in earth materials were dramatically
reduced when leached with an acetone/water mixture
(Table 6). Separate studies have shown that the mobility
of the major PBB isomers increased when the percentage
of water decreased in the acetone/water mixtures.
Another study also demonstrated that no PBB or HCB
was retained in soil columns when ethanol, containing
14C-PBB or 14C-HCB, was percolated through the
columns. This data again confirmed our previous
findings on soil TLC plates. It showed that nearly
100 percent of ' 4 C-PBBs and '" C-HCB were recovered
from soil columns when percolated with organic solvent.
This data agreed with our adsorption studies, which
showed no significant amount of organic-soluble PBBs
or HCB were adsorbed by soils.
This finding is significant in the disposal of PBBs
and HCB wastes. To prevent potential migration of
PBBs and HCB in a landfill, PBB and HCB wastes and
organic solvents should not be disposed of in the same
landfill location, and neither compound should be
allowed to come in contact with leaching organic
solvents.
Degradation Studies
The susceptibility of PBBs and HCB to biodegrada-
tion was evaluated using Catlin silt loam soil. Thirty
grams of air-dried soil, which had passed through a
2 mm sieve, was placed in 250-mL Erlenmeyer flasks.
One-third of the set of flasks were sterilized with
propylene oxide. Sterilization was confirmed by inocu-
lating a loopful of soil on a nutrient broth agar slant
and finding no microbial growth after one week of
incubation. Half of the non-sterilized flasks were inocu-
lated with 5 mL of mixed culture,1 ' which had pre-
viously been found to degrade water-soluble PCBs.
In the soil incubation studies, known amounts of
PBBs or HCB were distributed dropwise on the soil. All
soil samples were moistened with sterilized deionized
water to approximately 35 percent of field capacity,
then sealed with a sterilized rubber stopper wrapped in
aluminum foil. Samples were then incubated in the dark
at 22°C ± 2°C for 0, Y2, I, 2, and 6 months.
Three replicates of each treatment were extracted
with a suitable organic solvent after the indicated
period. The PBB or HCB extracts were each concen-
trated to 100 mL, and 1 mL of this extract was then
diluted to 3 mL by adding 0.75 /ug of hexabromo-
benzene or tribromobenzene in solvent as internal
standards. The diluted extracts were then analyzed
byGC.
The recovery of PBB components and HCB after
incubation in sterilized and non-sterilized Catlin soil are
shown in Tables 7 and 8. The data clearly show that
PBB and HCB persisted for 6 months in soil with no
significant microbial degradation. The same pattern of
persistence holds for PBB incubation with mixed
cultures added, which had previously shown effective
degradation of water-soluble PCBs.' ' This data agreed
with previous findings.2 9 >3° The persistence of PBBs
and HCB is also consistent with evidence reported for
PCBs which shows that the more heavily chlorinated
isomers (penta or greater) are resistant to degradation,
88
-------
TABLE 7. RECOVERY OF PBB ISOMERS AFTER INCUBATION OF 10 ppmw IN CATLIN SOIL,* IN PERCENTAGES
Incubation
time
(months)
0
'/2
1
2
6
0
'/2
1
2
6
Sf
82.1
80.9
79.5
73.4
73.1
89.9
85.1
840
825
81.9
Peak 1
N-S
83.2
80.4
80.0
74.5
75.3
Peak 6
92.9
89.8
88.7
84.3
83.1
,#
83.7
81.5
81 7
76.1
74.2
90.4
86.1
84.5
82.3
82.8
S
84.5
81.3
79.8
74.3
73.2
91.7
86.9
84.7
82.9
80.9
PeakS
N-S
85.3
85.5
81.3
75.0
75.4
Peak 7
93.1
88.4
84.0
83.6
84.0
I
84.6
87.5
81.5
75.9
75.6
94.2
89.5
87.7
85.9
84.3
S
89.5
803
81.0
80.4
79.3
90.8
85.3
84.1
82.0
83.0
Peak 4
N-S
91.3
87.7
84.8
86.1
86.8
PeakS
91.6
87.9
86.6
86.4
87.2
I
90.5
88.1
86.8
88.3
85.7
92.5
88.3
87.1
86.2
86.4
S
91.2
84.9
83.3
81.9
80.5
91.9
87.7
87.8
85.6
84.7
Peak5
N-S
93.8
88.9
87.8
84.5
83.0
Peak 9
94.1
89.1
89.7
86.8
84.7
1
925
86.9
85.4
84.0
83.3
93.3
88.4
91.9
87.4
85.1
•Each value is the mean of three replications
,S = Sterilized soil
JM-S = Non-sterilized soil
I = Inoculated mixed culture
though many lesser chlorinated components are meta-
bolized. '' s'4 4' '' Of interest is the significant loss
of extractability with time of all PBB isomers (Table 7)
and HCB (Table 8) in the sterile as well as the non-
sterile treatments. The loss of extractability of PBBs
and HCB from soil is probably due to sorption, masking,
or volatilization during long-term incubations.
We examined the fate of photodegraded PBB
because the higher brominated forms are readily de-
graded by ultraviolet light to lesser brominated forms
that could be more toxic. It can be reasoned that the
PBB on PBB-amended soil surfaces might be subjected
to some photodegradation; however, this does not
appear to have been a significant reaction in the long-
term incubation studies since the ratios of peaks 1:5:9
found in most of the PBB-amended soil extracts were
(5.3:100:18.5), which did not vary significantly from
the PBB standard (5.3:100:18.6).
TABLE 8. PERCENT OF HCB RECOVERED AFTER INCUBATION
OF 3.33 ppmw IN CATLIN SOIL
% of HCB remaining in soil
Treatment
Sterilized
Non-sterilized
0 month
85.68*
89.04
1 month
8278
84.08
2 months
79.75
82.24
6 months
63.26
64.74
*Each value is a mean of 3 replicates.
"J f^'
For a liquid culture study, a culture of mixed
bacteria was used that was isolated from a Hudson
River sediment from the Fort Miller disposal site in New
York.3 4 The predominant organisms found in the
mixed culture were Alkaligenes odorans, Alkaligenes
denitrificans, and an unidentified bacterium.1 ] The
culture was tested to determine if these organisms
could metabolize any of the PBB isomers or HCB when
present in a mineral growth medium:17
KN03 1.0 g/L FeCl3 0.02 g/L
MgS04 0.2g/LNaCl 0.1 g/L
CaCl2 0.1g/LK2HPO4 1.0 g/L
The deionized water used to make up this medium was
saturated with PBBs and/or HCB.
For these studies, 10 mL of a 3-day-old culture
were added to 200 mL of mineral medium with PBB
and/or HCB-saturated water. The solutions were
shaken at 23° ± 2°C on a rotary shaker for 0, 2, and 4
weeks. Part of the samples were used as a control with-
out adding inoculum. After an indicated period of time,
two samples of each treatment were extracted with
water-saturated hexane, then concentrated to 1 mL
prior to GC analysis.
The results showed that no PBB or HCB meta-
bolites were found in PBB- and HCB-saturated mineral
89
-------
solutions after 4 weeks. Apparently these mixed cul-
tures cannot degrade fireMaster BP-6 or HCB.
Because PBBs are not degraded, are not leached
in aqueous solutions, are not taken up by plants, B
and are not volatilized (because of their low vapor
pressure), we expect PBB to be a rather permanent
component of contaminated soils. The same conclusion
is probably true for HCB if the wastes are well-covered
with compacted wet soil to prevent volatilization
losses1 s in landfill sites.
REFERENCES
1. Ahmed, M, and D. D. Focht. 1973. Oxidation of
polychlorinated biphenyls by achromobacter PCB.
Bull. Environ. Contam. Toxicol. 10, p. 70.
2. Anderson, H. A., R. Lilis, I. J. Selikoff, K. D.
Rosenman, J. A. Valcinkas, and S. Freedman.
1978a. Unanticipated prevalence of symptoms
among dairy farmers in Michigan and Wisconsin.
Environmental Health Perspective 23, p. 217-226.
3. Anderson, H. A., E. C. Holstein, S. M. Daum,
L. Sarkozi, and I. J. Selikoff. 1978b. Liver func-
tion tests among Michigan and Wisconsin dairy
farmers. Environmental Health Perspective 23,
p. 333-339.
4. Avrahami, M., and R. T. Steele. 1972. Hexachloro-
benzene I. Accumulation and Elimination of HCB
in sheep after oral dosing. N. Z. J. of Agric.
Res. 15, p. 476-481.
5. Baxter, R. A., P. E. Gilbert, R. A. Lidgett, J. H.
Mainprize, and H. A. Vodden. 1975. The degrada-
tion of polychlorinated biphenyls by microorga-
nisms. Science of the Total Environment 4, p. 53.
6. Booth, N. H., and J. R. McDowell. 1975. Toxicity
of hexachlorobenzene and associated residues in
edible animal tissues. Journal of the American
Medical Association 166, p. 591-595.
7. Burns, J. E., and F. M. Miller. 1975. Hexachloro-
benzene contamination: Its effects in a Louisiana
population. Arch. Environ. Health. 30, p. 44-48.
8. Cam, C., and G. Nigogosyan. 1963. Acquired
toxic porphyria cutanea tarda due to hexachloro-
benzene. Journal of the American Medical Associa-
tion 183, p. 88-101.
9. Carter, L. J. 1976. Michigan's PBB incident:
Chemical mixup leads to disaster. Science 192,
240 p.
10. Chou, S. F., L. W. Jacobs, D. Penner, and J. M.
Tiedji. 1978. Absence of plant uptake and trans-
location of polybrominated biphenyls (PBBs).
Environmental Health Perspectives 23, p. 9.
11. Clark, R. R., E. S. K. Chian, and R. A. Griffin.
1979. Degradation of polychlorinated biphenyls
by mixed microbial cultures. Applied and Environ-
mental Microbiology 37, p. 680-685.
12. Cromartie, E., W. L. Reichel, L. N. Locke, A. A.
Belisle, T. E. Kaiser, T. G. Lament, B. H. Mulhern,
R. M. Prouty, and M. Swineford. 1975. Residues
of organochlorine pesticides and polychlorinated
biphenyls and autopsy data for bald eagles. 1971-
1972. Pestic. Monit. J. 9, p. 11-14.
13. Dannan, G. A., R. W. Moore, and S. D. Aust.
1978. Studies on the microsomal metabolism and
binding of polybrominated biphenyls (PBBs).
Environmental Health Perspectives 23, p. 51-61.
14. Di Carlo, F. J., J. Seifter, and V. DeCarlo. 1978.
Assessment of the hazards of polybrominated
biphenyls. Environmental Health Perspectives 23,
p. 351-365.
15. Farmer, W. J., M. S. Yang, and J. Letey. 1977.
Land Disposal of Hexachlorobenzene Wastes:
Controlling Vapor Movement in Soil. Contract
no. 68-03-204. Final report to U. S. Environ-
mental Protection Agency. Univ. of California,
Riverside, California.
16. Gilbertson, M., and L. M. Reynolds. 1972. Hexa-
chlorobenzene (HCB) in the eggs of common
terns in Hamilton Harbour, Ontario. Bull. Environ.
Contam. Toxicol. 7, p. 371-373.
17. Gray, P. H. H., and H. G. Thornton. 1928. Soil
bacteria that decompose certain aromatic com-
pounds. Zentralblatt Bacteriologie Parasitenkunde,
Abteilung II. v. 73, p. 74.
18. Griffin, R. A., A. K. Au, and E. S. K. Chian. 1979.
Mobility of polychlorinated biphenyls and
decamba in soil materials: Determination by soil
thin-layer chromatography. Proceedings of the 8th
National Conference and Exhibition on Municipal
Sludge Management, p. 183-189.
19. Griffin, R. A., F. B. DeWalle, E. S. K. Chian,
J. H. Kim, and A. K. Au. 1977. Attenuation of
90
-------
PCBs by soil materials and char wastes. Proceedings
of the Third Annual Municipal Solid Waste Re-
search Symposium. U. S. Environmental Protection
Agency. EPA-600/9-77-026, p. 206.
20. Helling, C. S. and B. C. Turner. 1968. Pesticide
mobility: Determination by soil thin-layer chroma-
tography. Science 162: 562-563.
21. Helling, C. S. 197la. Pesticide mobility in soils
I. Parameters of thin-layer chromatography. Soil
Sci. Soc. Am. Proc. 35: 732-737.
22. Helling, C. S. 1971b. Pesticide mobility in soils
II. Applications of soil thin-layer chromatography.
Soil Sci. Soc. Am. Proc. 35: 737-743.
23. Helling, C. S. 1971c. Pesticide mobility in soils
III. Influence of soil properties. Soil Sci. Soc.
Am. Proc. 35: 743-747.
24. Hesse, J. L., and R. A. Powers. 1978. Polybromi-
nated biphenyl (PBB) contamination of the Pine
River, Gratiot, and Midland Counties. Michigan
Environmental Health Perspectives 23, p. 19-25.
25. Holden, A. V. 1970. International co-operative
study of organochlorine pesticide residues in
terrestrial and aquatic wildlife. 1967-1968.
Pesticide Monit. J. 4, p. 117-135.
26. Humphrey, H. E. B., and N. S. Hayner. 1975.
Polybrominated biphenyls: An agricultural inci-
dent and its consequences. II. An epidemiological
investigation of human exposure. Trace Subst.
Environ. Health. 9, p. 57.
27. Isbister, J. L. 1977. PBBs in human health. Clin.
Med. 84, p. 22.
28. Isleib, D. R., and G. L. Whitehead. 1975. Poly-
brominated biphenyls: An agricultural incident
and its consequences. I. The agricultural effect of
exposure. Ninth Annual Conf. on Trace Substances
in Environmental Health, Univ. of Missouri,
Columbia, Missouri.
29. Jacobs, L. W., S. F. Chou, and J. M. Tiedji. 1976.
Fate of polybrominated biphenyls (PBBs) in
soils. Persistence and plant uptake. J. Agri. Food
Chem.24,p. 1198.
30. Jacobs, L. W., S. F. Chou, and J. M. Teidji. 1978.
Field concentration and persistence of polybromi-
nated biphenyls in soils and solubility of PBB in
natural waters. Environmental Health Perspectives
23, p. 1.
31. Johnson, J. L., D. L. Stalling, and J. W. Hogan.
1974. Hexachlorobenzene (HCB) residues in
fish. Bull. Environ. Contam. Toxicol. 11,
p. 393-398.
32. Laseter, J. L., C. K. Bartell, A. L. Laska, D. G.
Holmquist, D. B. Condie, J. W. Brown, and R. L.
Evans. 1976. An Ecological Study of Hexachloro-
benzene. U. S. EPA-560/6-76-009. 62 p.
33. Lee, M. C., R. A. Griffin, M. L. Miller, and E. S. K.
Chian. 1979. Adsorption of water-soluble poly-
chlorinated biphenyl Aroclor 1242 and used
capacitor fluid by soil materials and coal chars.
J. of Environ. Science and Health A14,
p. 415-442.
34. Leis, W. M., W. F. Beers, J. M. Davidson, and
G. D. Knowles. 1978. Migration of PCBs by
Ground-Water Transport-A Case Study of Twelve
Landfills and Dredge Disposal Sites in the Upper
Hudson Valley, New York. Conference of Applied
Research and Practice on Municipal and Industrial
Waste, Box 5571, Madison, WI 53705. p. 539-548.
35. McCormack, K. M., W. M. Kluwe, V. L. Sanger,
and J. B. Hook. 1978. Effects of Polybrominated
biphenyls on kidney function and activity of
renal microsomal enzymes. Environmental Health
Perspectives 23, p. 153-157.
36. Medline, A., E. Bain, A. I. Menon, and H. F.
Haberman. 1973. Hexachlorobenzene in rat
liver. Arch. Pathol. 96. p. 61-65.
37. Moorhead, P. D., L. B. Willett, and F. L. Schan-
bacher. 1978. Effects of PBB on cattle. II. Gross
pathology and histopathology. Environmental
Health Perspectives 23, p. 111-118.
38. Neufeld, M. L., M. Sittenfield, and K. F. Wolk.
1977. Market Input/Output Studies. Task IV:
Polybrominated Biphenyls. EPA-560/6-77-017.
39. Quinlivan, S.,M. Ghassemi, and M. Santy. TRW
System Group. 1975. Survey of Methods Used to
Control Wastes Containing Hexachlorobenzene.
Final Report to U. S. Environmental Protection
Agency. Contract no. BOA 68-01-2956.
40. Robertson, L. W., and D. P. Chynoweth. 1975.
Another halogenated hydrocarbon. Environ. 17,
25 p.
91
-------
41. Robl, M. G., D. H. Jenkins, R. J. Wingender,
D. E. Gordon, and M. L. Keplinger. 1978.
Toxicity and residue studies in dairy animals
with fireMaster FF-1 (polybrominated biphenyls).
Environmental Health Perspectives 23, p. 91-97.
42. Schanbacher, F. L., L. B. Willett, P. D. Moorhead,
and H. D. Mercer. 1978. Effects of PBBs on cattle.
III. Target organ modification as shown by renal
function and liver biochemistry. Environmental
Health Perspectives 23, p. 119-127.
43. Sundstrom, G., O. Hutzinger, and S. Safe. 1976.
Identification of 2, 2', 4, 4', 5, 5'-hexabromo-
biphenyl as the major component of flame re-
tardant fireMaster BP-6. Chemosphere. 5, p. 11.
44. Tucker, E. S., V. W. Saeger, and 0. Hicks. 1975.
Activated sludge primary biodegradation of poly-
chlorinated biphenyls. Bull. Environ. Contam.
Toxicol. 14, p. 705.
45. Vos, J. G., H. L. Van Der Maas, A. Musch, and
E. Ram. 1971. Toxicity of hexachlorobenzene
in Japanese quail with special reference to
porphyria, liver damage, reproduction, and tissue
residues. Toxicol. and Applied Pharmacol. 18,
p. 944-957.
46. Zitko, V. 1971. Polychlorinated biphenyls and
organochlorine pesticide in some freshwater and
marine fishes. Bull. Environ. Contam. Toxicol.
6, p. 464-470.
92
-------
MOVEMENT AND BIOLOGICAL DEGRADATION OF LARGE CONCENTRATIONS
OF SELECTED PESTICIDES IN SOILS
J.M. Davidson, P.S.C. Rao, and Ti-Tse Ou
Soil Science Department
University of Florida
Gainesville, FL 32611
ABSTRACT
Equilibrium adsorption isotherms of the non-linear Freundlich type were obtained for
atrazine, methyl parathion, terbacil, trifluralin, and 2,4-D and four soils. Pesticide
solution concentrations used in the study ranged from zero to the aqueous solubility limit
of each pesticide. The mobility of each pesticide increased as the concentration of the
pesticide in the soil solution phase increased. These results were in agreement with the
equilibrium adsorption isotherm data. Biological degradation of each pesticide was mea-
sured by 14C02 evolution resulting from the oxidation of uniformly 14C ring-labeled pesti-
cides, except trifluralin which was labeled at 1'*CF3. Technical grade and formulated forms
of each pesticide at concentrations ranging from zero to 20,000 yg/g of soil were used in
the biological degradation experiments. Pesticide degradation rates and soil microbial
populations generally declined as the pesticide concentration in the soil increased; how-
ever, some soils were able to degrade a pesticide at all concentrations studied, some
soils degraded a pesticide at a low concentration but not a higher concentration, while
others remained essentially sterile throughout the incubation period. Total C02 evolution
was not always a good indication of pesticide degradation. Several pesticide metabolites
were formed and identified in various soil-pesticide systems. The quantities of methyl
parathion, trifluralin and atrazine "bound" to the soil at the end ot the incubation per-
iod were measured and in some cases appeared to be related to types of metabolites formed
during biological degradation.
INTRODUCTION
Because of a continued increase in the
number and quantity of pesticide compounds
being placed on the market, the safe dis-
posal of surplus and/or waste pesticide
materials has become an acute problem (von
Rumker, et a!., 1974). Incineration, en-
capsulation, isolation in underground caves
and mines, chemical stabilization, land
spreading and landfills are some of the
procedures being considered for the dispos-
al of pesticides and other hazardous wastes
(Schomaker, 1976; von Everdingen and
Freeze, 1971; Wilkinson, et al., 1978). Of
these methods, disposal by landfills and
land spreading appear to be more common and
economical (Fields and Lindsey, 1975; Lind-
sey, et al., 1976). Placing hazardous
wastes in the land has come under attack
recently (Atkins, 1972; Rouston and Wil-
dung, 1969) because there is no guarantee
that the hazardous chemicals disposed of
in this manner will not migrate from the
disposal site to potable water supplies.
In general, pesticide applications
associated with agricultural production
have had very little adverse effect on the
soil microbial activity (Cole, 1976;
Hubbell, et al., 1973; Kaiser, et al.,
1970, Newman and Downing, 1958; Roslycky,
1977). However, reports on soil microbial
activity where large concentrations were
used have been contradictory and inconclu-
sive. For example, Ou et al. (1978a) ob-
served 2,4-D (2,4-dichlorophenoxyacetic
acid) degradation at concentrations of
5,000 and 20,000 yg/g of soil (pptn) for
one soil type and no degradation at the
same concentrations for another soil. Soil
respiration and bacterial, fungal and
93
-------
actinomycete populations were significantly
reduced in the soil unable to degrade
2,4-D. They concluded that the physical
and chemical properties of the soil as well
as the 2,4-D concentration were important
factors in governing microbial activity and
pesticide degradation in soils receiving
large pesticide concentrations.
Trifluralin (a,a,a-trifluro-2,6-dini-
tro-N, N-dipropyl-p-toluidine) and atrazine
(2-chloro-4-ethylamine-6-isopropylamino-s-
triazine) are commonly used herbicides.
Trifluralin applications of 3 kg/ha (1.4
ppm) have been shown not to influence soil
bacterial, fungal and actinomycete popula-
tions significantly (Tyunaeva, 1974). How-
ever, when 1.1 kg/ha (0.5 ppm) was applied
per year over a five year period, bacterial
populations were inhibited while fungal and
actinomycete populations were enhanced
(Breazeale and Camper, 1970). When analyt-
ical grade trifluralin was incorporated
into the soil at concentrations of 5,000
ppm, C02 evolution and bacterial popula-
tions were inhibited while streptomycete
populations were stimulated. Stojanovic
et al. (1972) has shown that formulated
trifluralin stimulated C02 evolution and
streptomycete populations while inhibiting
bacterial populations in soil. Atrazine
was shown not to inhibit soil respiration
at concentrations associated with agricul-
tural production (Kaiser, et al., 1970;
Eno, 1962). Cole (1976) and Voets et al.
(1974) have shown that soil bacterial and
fungal populations were not affected at
rates below 4 kg/ha (1.8 ppm). However,
Stojanovic et al. (1972) reported that
atrazine inhibited soil respiration and
bacterial populations at 5,000 ppm but had
no effect on fungal populations.
A thorough understanding of the vari-
ous processes that influence the persis-
tence, retention, and leaching of
pesticides in soils is required to develop
technology for the selection and management
of pesticide disposal sites involving
soils. The fate of pesticides in soils
when applied at concentrations similar to
those associated with agricultural prac-
tices has been well-documented in several
reviews (Bailey and White, 1970; Sanborn,
et al., 1977). However, the direct extrap-
olation of this data base to systems con-
taining large pesticide concentrations,
such as those occurring at or below dispo-
sal sites, may not be feasible (Davidson
et al., 1976).
Laboratory experiments were initiated
to investigate the physical, chemical and
microbiological behavior of five pesti-
cides in four soils when the pesticide was
present at large concentrations. The ob-
jectives of the study were to: 1) Measure
and describe pesticide adsorption in se-
lected soil-water systems over a wide
range of chemical concentrations (zero to
water solubility), 2) Measure mobility and
distribution of pesticides in selected
soils, when applied or initially present
in soil at large concentrations, 3) Mea-
sure chemical and microbial degradation
rate, and identify metabolites produced in
soil-pesticide systems receiving large
pesticide concentrations, and 4) Measure
the influence of large pesticide concen-
trations on soil microbial activity and
respiration rate for selected soil-pesti-
cide systems.
MATERIALS AND METHODS
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 (Enti-
sol) from New Mexico, and Eustis fine sand
(Entisol) from Florida. These soil types
were selected on the basis of their geo-
graphic and taxonomic representation 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 stor-
age and use. Selected physical and chemi-
cal properties of the mineral soils are
given in Table 1.
Pesticides
Five pesticides included in this
study were: 2,4-D (2,4-dichlorophenoxyace-
tic acid), atrazine (2-chloro-4-ethyla-
mino-6-isopropylamino-s-triazine),
terbacil (3-tert-butyl-5-chloro-6-methy-
luracil), methyl parathion (0-0-dimethyl
0-p-nitrophenyl phosphorothioate), and
trifluralin (a,a,a-trifluoro-2,6-dinitro-
N, N-dipropyl-p-toluidine).
Adsorption Experiments
Equilibrium adsorption isotherms for
all soil-pesticide combinations were mea-
sured using the batch procedure. Equilib-
rium was achieved by shaking duplicate
samples of five or ten grams of soil with
94
-------
TABLE 1. PHYSICAL AND CHEMICAL PROPERTIES OF THE SOILS USED IN THIS STUDY
SOIL
Webster
Cecil
Glendale
Eustis
PARTICLE
sand
18.4
65.8
50.7
93.8
SIZE FRACTION
silt
45.3
19.5
16.4
3.0
(%)
clay
38.3
14.7
22.9
3.2
PH (1
Water
7.3
5.6
7.4
5.6
:1 paste)
IN KC1
6.5
4.8
6.5
4.1
CEC
(meq/100 g)
54.7
6.8
35.8
5.2
Organic C
(*)
3.87
0.90
0.50
0.56
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 adsorp-
tion occurred after 48 hours. Following
equilibrium, the test tubes were centri-
fuged at 800 x 6 for ten minutes and the
ll*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 (Rao and
Davidson, 1979) using the miscible dis-
placement technique described by Davidson
et al. (1968). 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 solution. A known volume of
pesticide solution at a desired concentra-
tion was introduced to the soil at a con-
stant flux using a constant-volume
peristaltic pump. After a specific volume
of pesticide solution had been applied, the
pesticide solution was subsequently dis-
placed through the soil column with 0.01 N
CaCl2 at the same soil-water flux. Efflu-
ent solutions were collected in 5 or
10 ml aliquots using an automatic fraction
collector. A pulse of 3H20 (specific acti-
vity 5 nCi/ml) was also displaced through
each soil column to characterize the trans-
port of non-adsorbed solutes. The activity
of 1UC and 3H20 in effluent fractions was
counted by liquid scintillation. All dis-
placements were performed at a Darcy flux
of approximately 0.22 cm/hr to ensure
near-equilibrium conditions for pesticide
adsorption during flow. The total volume
of water held in the soil column was
gravimetrically determined at the end of
each displacement by extruding the soil
from the glass cylinder and oven drying.
The number of pore volumes (V/V0) of solu-
tion displaced through the soil column
was calculated by dividing the cumulative
outflow volume (V) by total water volume
(V0) in the soil column. Effluent pesti-
cide concentrations were expressed as
relative concentrations (C/C0), where C
and C0 are, respectively, effluent and
input concentration. Plots of C/C0 versus
V/Vo are referred to as breakthrough
curves (BTC).
For the leaching or infiltration exper-
iments, air-dry soil was packed into 3.2-cm
diameter lucite cylinders composed of 1-cm
sections supported by a V-shaped container
that permitted observation of the wetting
front position with time. Technical or
analytical grade pesticide was dissolved
in benzene and was spiked with !i*C-labe1ed
compound. The benzene solution was mixed
with air-dry soil (to give 200 or 2,000 pg
of pesticide/g of soil and 10 nCi/g soil)
and the benzene evaporated. In order to
simulate a waste disposal site, the pesti-
cide-spiked soil was packed into the top
1.5 cm of the soil column. Infiltration
of water into horizontal columns of soil
was controlled by maintaining the soil
surface at a negative pressure (-4 cm of
water) using a fritted-glass plate and a
constant-head burette. The fritted-glass
plate apparatus was filled with 0.01 N
CaCl2 and the desired negative pressure
95
-------
applied before it was placed in contact
with the soil. Measurement of time in all
experiments commenced the instant contact
was established between the fritted-glass
plate and soil surface. Water entering the
soil was measured volumetrically using a
constant-head burette connected to the
fritted-glass plate apparatus. Measure-
ments of distance to the wetting front
(zero at the contact plane between soil and
plate) were visually observed. When flow
had proceeded for the desired time (i.e.,
until the wetting front had advanced to
about 30 cm), the water supply was discon-
tinued and the soil column was immediately
cut into 1-cm segments. About one-half of
the soil contained in each 1-cm segment was
oven-dried at 105° C for 24 hours to deter-
mine the gravimetric soil-water content.
The remaining one-half of the soil from
each 1-cm-segment was dried in a vacuum
desiccator over P205 or H2SOit for a 24-
48 hour period. About 0.5-0.7 g of the
desiccator-dried soil was then combusted in
a Packard Model 306B sample oxidizer; the
14C02 evolved by combustion was trapped in
a premixed organic amine-fluor cocktail and
assayed by liquid scintillation. The pes-
ticide concentrations were calculated using
the specific activity (dpm/ug pesticide) of
the pesticide-spiked soil sample. The pes-
ticide concentrations determined in the
above manner represent the sum of adsorbed
and solution-phase concentrations and were
expressed as yg pesticide/g oven-dry soil.
Pesticide Degradation
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 C02-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 leav-
ing each flask was also bubbled through a
KOH solution (0.1 - 0.2N) to absorb the
evolved C02. At frequent intervals, the
KOH solutions were replaced with fresh KOH
solutions and the C02 concentrations
determined by titration. 1'*C-C02 activity
in the KOH solution was determined by
liquid scintillation counting.
Metabolite Identification
Soil treatments and incubation were
essentially the same as for the pesticide
degradation experiments, except 150 g of
soil were used and 2 yCi 11+C-pesticide per
100 g soil were added. Ten gram soil
samples were withdrawn biweekly. The soil
samples were extracted three times with
the appropriate solvent (Table 2). The
20,000 yg/g samples were extracted four
times because of their high pesticide con-
centrations. The extracted soil was then
air dried and stored in a cold chamber
prior to analysis for "bound" 1UC. The
extracts were concentrated to 10 ml on a
rotary evaporator, or in a Danish Kuderna
evaporator, then further concentrated
using a gentle stream of N2 and an aliquot
was assayed for lkC. The extracts were
further concentrated with N2 to one ml and
an aliquot equivalent to 15,000 dpm of
each sample was streaked on a thin layer
plate. The TLC plates were developed and
placed on Kodak on-screen x-ray film
(NS-57) for one month. Radioactive
streaks on the TLC plates were scraped and
the radioactivity eluted with two ml of
solvent (Table 2). The percentage of
radioactivity in each separate radioactive
component on the TLC plates was determined
by liquid scintillation counting.
The unextractable "bound" portion of the
14C-residue was determined by oxidizing
the extracted soil samples in a stream of
02 at 800° C according to the modified
method of Watts (1971). The air dried
extracted soil was placed directly in the
oven at 800° C and the 02 stream which
passed through the soil was bubbled
directly into 15 ml of phenethylamine C02-
trapping cocktail solution. Complete com-
bustion took approximately five minutes.
Then N2 was purged through the system to
eliminate 02 and the trapped 14C02 was
assayed.
96
-------
TABLE 2. SOLVENTS USED FOR PESTICIDE EXTRACTION FROM SOIL
Pesticide
Solvent
2,4-D
Trifluralin
Atrazine
Methyl Parathion
Ether Ethyl
1) Benzene/Ethyl Acetate (3:1)
2) MeOH
MeOH (soxhlet)
Hexane:Acetone (80:20)
RESULTS AND DISCUSSION
Adsorption Experiments
Equilibrium adsorption isotherms were
determined for each soil-pesticide combina-
tion by measuring pesticide adsorption at
five or more concentrations ranging from
zero to the pesticide's aqueous solubility.
All adsorption isotherms considered in this
study, with the exception of 2,4-D and the
Glendale soil, were described by the
Freundlich equation (S = KCN), where K and
N are constants, and S and C are adsorbed
(yM/kg soil) and solution (yM/1) phase
pesticide concentrations. The values of
the Freundlich adsorption constants, K and
N, for each soil-pesticide combination
studied were obtained using a least-square
fit procedure to the adsorption data.
These values are presented in Table 3.
Two important conclusions can be made
based on the data presented in Table 3.
First, the fact that the Freundlich equa-
tion describes nearly all pesticide
adsorption isotherms over a wide con-
centration range suggests that adsorption
sites were not saturated at any concen-
tration considered in this study. The
amount of pesticide adsorbed by the soil
continued to increase, at a decreasing
rate, with each increase in solution con-
centration. The behavior may not, however,
hold for other pesticide adsorbents (Weber
and Usinowicz, 1973). Second, contrary to
a frequent assumption, pesticide adsorp-
tion isotherms are generally nonlinear,
that is, N is greater or less than one
(Table 3). Linear adsorption isotherms
have been generally accepted for low
pesticide concentrations because it
simplified computer simulation modeling
(Davidson et al., 1968; Davidson and
Chang, 1972; Hugenberger et al., 1972;
Kay and El rick, 1967).
Because soil organic carbon content
generally correlates well with pesticide
adsorption, the use of an adsorption
partition coefficient based upon organic
carbon content rather than total soil mass
has been proposed by Lambert (1968) and
Hamaker and Thompson (1971). Using this
procedure, the amount of pesticide adsorb-
ed was expressed as yg/g organic carbon
and the Freundlick constant (KQC) for each
adsorption isotherm was computed. These
values are also presented in Table 3. It
is apparent that the values of KQQ for a
given pesticide are much less variable
(smaller percent CV) among the four soils
studied than are the K values uncorrected
for organic carbon. These results are in
general agreement with the observations of
Hamaker (1975) where the KQC values for a
given pesticide were nearly independent of
soil type. It should be recognized, how-
ever, that other factors such as soil pH,
clay content, and cation exchange capacity
may also play a significant role in deter-
mining pesticide adsorption by soils
(Bailey and White, 1970). On the basis of
the KOC values listed in Table 3, the ex-
tent of pesticide adsorption in soils was
in the order of terbacil < trifluralin
< 2,4-D amine < atrazine < methyl para-
thion.
Column Displacement Experiments
The partial differential equation
generally assumed to describe the move-
ment of pesticides and other adsorbed
97
-------
TABLE 3. FREUNDLICH CONSTANTS CALCULATED FROM EQUILIBRIUM ADSORPTION
ISOTHERMS FOR VARIOUS SOIL-PESTICIDE COMBINATIONS
PESTICIDE
SOIL
Atrazine
Methyl
Parathion
Terbacil
Trifluralin
2,4-D Amine
Webster
Cecil
Glendale
Eustis
Average ± % CV*
Webster
Cecil
Glendale
Eustis
Average ± % CV
Webster
Cecil
Glendale
Eustis
Average ± % CV
Webster
Cecil
Glendale
Eustis
Average ± % CV
Webster
Cecil
Glendale
Eustis
Average ± % CV
9.12
0.84
0.69
0.85
2.87 ± 145
18.67
4.81
6.05
3.30
8.21 ± 86
2.96
0.39
0.42
0.15
0.98 + 135
2.49
0.43
1.31
0.23
1.11 ± 92
7.27
0.84
--
1.14
3.08 ± 118
6.03
0.89
0.62
0.62
2.04 ± 131
13.39
3.95
3.57
2.72
5.91 ± 85
2.46
0.38
0.38
0.12
0.83 ± 130
2.93
0.46
1.60
0.24
1.31 ±94
4.62
0.65
--
0.76
2.01 ± 112
0.73
1.04
0.93
0.79
0.87 ± 16
0.75
0.85
0.61
0.86
0.77 ± 15
0.88
0.99
0.93
0.88
0.92+6
1.15
1.05
1.18
1.06
1.11 ± 6
0.70
0.83
—
0.73
0.75 ± 9
155.8
98.9
124.0
110.7
122.3 + 20
346.0
438.6
714.5
486.4
496.4 ± 32
63.6
42.2
76.0
21.4
50.8 ± 47
75.7
50.7
177.8
43.2
86.8 + 72
119.4
72.2
--
135.7
109.1 ± 30
KM Freundlich constant when solution and adsorbed phase concentrations are expressed
M as uM/1 and yM/kg of soil.
1-Kr Freundlich constant for solution and adsorbed phase concentrations are as yg/ml and
b yg/g of soil [Kg = KM(MW/1 ,000)"!-N], where MW is the pesticide's molecular weight.
tKnr Freundlich constant for solution and adsorbed phase expressed as yg/ml and yg/g of
organic carbon.
*CV is the coefficient of variation, % CV = (standard deviation/average) x 100.
solutes through soils under steady-state
water flow conditions is (Van Genuchten,
et al., 1974):
= D
at u
35
3X"
- V
where t is time (days), D is dispersion
coefficient (cm2/day), x is distance (cm),
v is average pore-water velocity (cm/day),
p is soil bulk density (g/cm3), e is volu-
metric soil-water content (cm3/cm3), and
C and S are solution and adsorbed pesti-
cide phase (yg/ml and yg/g), respectively.
When the adsorption isotherm obeys the
Freundlich equation, the convective-
dispersive solute transport model (Equa-
tion 1) reduces to:
98
-------
- V
3C
3X
where,
R(C) = [1 +
pKNCN'Ve]
[2]
[3]
The retardation term R(C) is a quanti-
tative index of the pesticide's mobility in
that its value is equal to the ratio of the
positions of the adsorbed and nonadsorbed
solute fronts in soil. The value of the
adsorption coefficient K in Equation [3]
for nonadsorbed solutes (e.g., chloride or
3H20) is equal to zero; hence, R(C) = 1.
For adsorbed solutes, R(C) is greater than
one because the value of K is larger than
zero. Thus, larger values of R(C) indicate
reduced pesticide mobility in soils. It
may be noted from Equation [3] that for the
case of nonlinear adsorption isotherms
(N < 1), the value of the retardation term
increases with decreasing solution concen-
tration C, while for a linear isotherm
(N = 1), R(C) is independent of pesticide
solution concentration. Thus, the mobility
of pesticides and other adsorbed solutes
through soils is directly influenced by the
shape of the equilibrium adsorption iso-
therms.
Effluent breakthrough curves (BTC)
were measured for 2,4-D amine at two input
concentrations (C = 50 and 5,000 yg/ml) and
tritiated water (3H20) using columns of
each soil. Tritiated water represents a
nonadsorbed solute and serves as a refer-
ence for the adsorbed pesticides. A shift
of the BTC for adsorbed solutes to the
right of 3H20 BTC is due to an adsorption-
induced retardation. The greater the
righthand shift of the BTC, the greater
the adsorption; thus, a decrease in
mobility. It is apparent from the data
presented in Figure 1 that the mobility of
2,4-D amine in the Webster soil was signi-
ficantly increased as the input concentra-
tion (C0) increased from 50 to 5,000 jig/ml.
Note that for the 5,000 ug/ml input con-
centration, 2,4-D amine was nearly as
mobile as was 3H20. The effect of increas-
ed mobility at high concentration was more
pronounced in the Webster soil than in the
other soils. These column results are con-
sistent with Equation [3] and the measured
nonlinear adsorption isotherms (Table 3)
for 2,4-D amine. Similar results were ob-
tained for other soil-pesticide combina-
tions (Rao and Davidson, 1979).
The position of the BTC for an
adsorbed solute is governed by the nature
of the equilibrium adsorption isotherm
(Equation 3), whereas the shape of the BTC
(i.e., symmetry or lack of it) is defined
by nonlinearity of the adsorption isotherm
and the kinetics of adsorption-desorption
processes. Symmetrical BTC are obtained
when adsorption is an instantaneous pro-
cess and the adsorption isotherm is
linear. For nonequilibrium adsorption
conditions during flow, asymmetrical BTC
are generally obtained (Van Genuchten, et
al., 1974; Rao, et al., 1979). All of the
pesticide BTC measured in this study were
asymmetrical in shape with extensive
"tailing" observed as C/C0 approached 1.0
or zero. Tailing was absent in 3H20
breakthrough curves. The extent of the
asymmetrical shape of each pesticide BTC
exceeded that which could be attributed
to the nonlinear nature of the adsorption
isotherms. Hence, much of the asymmetry
measured for the pesticide BTC may be
attributed to nonequilibrium conditions
which exist in the soil columns during
flow. Rao et al. (1979) presented an
evaluation of two conceptual models where
nonequilibrium during flow was attributed
to either kinetics-controlled or diffu-
sion-controlled adsorption-desorption
processes.
The illustrated increase in pesticide
mobility at high concentrations limits the
usefulness of the present low concentra-
tion data base for developing safe manage-
ment practices for pesticide disposal in
the soil. However, underestimation of
pesticide movement by assuming linear
adsorption isotherms may not be severe for
pesticides with low aqueous solubilities.
Ou et al. (1978) showed that for high
loading rates, up to 20,000 yg 2,4-D/g
soils, there was a significant decrease
in the pesticide degradation rate with a
comcomitant depression of total microbial
activity in the soil. Thus, due to rapid
leaching and minimal microbial decomposi-
tion of pesticides at high concentrations,
the potential for groundwater contamina-
tion with pesticides is increased.
Infiltration Experiments
Pesticide transport in soils during
transient, unsaturated, and one-dimension-
al water flow was investigated using hand
packed horizontal soil columns. In order
to simulate a waste disposal site, the top
99
-------
A'tBS Fr~ R SOI
2, 4 - D Amine
Figure 1. Effluent breakthrough curves for
2,4-D amine (C0 = 50 and 5,000 yg/ml) and
for tritiated water displacement through
Webster soil column.
1.5 cm of each soil column was packed with
pesticide-treated soil (2,000 yg pesti-
cide/g soil), and 0.01 N CaCl2 was infil-
trated into the soil at a constant negative
head (-4 cm of water). During infiltra-
tion, the rate of wetting front advance was
recorded by visual observations. The soil
column was cut into 1-cm segments at the
end of the infiltration. In each soil seg-
ment, the total amount of pesticide (sum of
solution, adsorbed, and solid phases) was
determined by combustion, while the soil-
water content was measured by oven-drying.
The depth (xp) to which the pesticide
front moved in a soil due to water infil-
tration was dependent upon the wetting
front depth (xw)> the pesticide adsorption
isotherm constants (K and N), and the
aqueous solubility (Cs) of the pesticide.
The relationship between these variables
may be expressed as:
C4]
e
where, p is the soil bulk density (g/cm3)
and e is the average soil-water content
(cm3/cm3) in the wetted zone behind the
wetting front. It should be noted that
for the case of a linear adsorption iso-
therm (N = 1), Equation 4 is exact and the
retardation of the pesticide front due to
adsorption is independent of concentration,
while for the nonlinear isotherm case,
Equation 4 is only an approximation.
Assuming that Darcy's law is valid
for unsaturated water flow in soils, the
rate of the advance of the wetting front
is given by (Kirkham and Powers, 1972):
[5]
where, m is a constant and t is time
(min). In the present study, Equation [5]
described (r2 >_ 0.95) the advance of the
wetting front for all soil columns con-
sidered. The values of m and other per-
tinent data for the infiltration
experiments are summarized in Table 4.
Measured pesticide concentration pro-
files and soil-water content distributions
at the end of infiltration in Eustis,
Cecil, and Webster soil columns are shown
in Figures 2 and 3. Because the final
wetting front position in each soil
column was different, for ease of compar-
ison the ordinate in Figures 2 and 3 is
plotted as soil depth relative to the
wetting front depth (i.e., x/xw)- Except
for the 2,4-D-Eustis and the terbicil-
Eustis data, the relative mobilities of
the pesticides are in general agreement
with those anticipated from the equili-
brium adsorption isotherms and pesticide
aqueous solubilities. The measured mobil-
ity of terbacil and 2,4-D in the Eustis
soil (Figure 2) was nearly the same al-
though the adsorption coefficient for
2,4-D is greater than that for terbacil
in Eustis soil (see Tables 3 and 4). The
importance of aqueous solubility is demon-
strated by the atrazine data in the Eustis
soil (Figure 2 and Table 4). Note that
the volume of water infiltrated into the
soil column could solubilize and transport
only about 4% of the total pesticide pre-
sent in the top 1.5 cm segment; thus, most
of the atrazine does not appear to have
moved. The retardation factors, (xw/Xn)>
calculated by Equation [4] are generally
larger than those measured in the infiltra-
tion experiments (Table 4). Similar re-
sults were reported by Wood and Davidson
(1975) for transient-flow studies and by
Davidson and Chang (1972) for saturated
flow experiments. The kinetics of pesti-
cide adsorption-desorption in soils are
not understood well enough at this time
(Rao et a!., 1979) to describe modeling
the nonequilibrium conditions for pesticide
100
-------
TABLE 4. PHYSICAL DATA FOR INFILTRATION EXPERIMENTS
SOIL PESTICIDE C,
xw/x
Volume Pesticide
of water Recovery
applied (%)
(yg/ml)
Eustis
Eustis
Eustis
Cecil
Webster
Atrazine
2,4-D
Terbacil
Terbacil
Terbacil
33
650
710
710
710
(g/cm3;
1.67
1.69
1.64
1.51
1.40
1 (cm) (cm/minis)
21.3
30.0
29.5
28.4
24.1
3.195
3.854
4.039
0.705
0.543
Measured
2.
1.
1.
1.
4.
50
05
11
49
16
Calcu
3.
1.
1.
3.
4.
lated** (ml)
03
89
37
03
55
44
68
63
68
67
>100
>100
91
93
99
* see Equation [5]
**see Equation [4]
adsorption-desorption during transient
soil-water flow. Additional studies are
needed in this area.
Pesticide Degradation
Total C02 evolution (respiration) is
generally a good indicator of soil micro-
bial activity. This procedure was used by
Pesticide Concentration, Jjg/g soil
Stojanovic et al. (1972) to estimate pes-
ticide degradation rates in soil systems
receiving large pesticide concentrations.
A more direct procedure for determining
pesticide degradation, however, is to mea-
sure lkCQ? evolution from uniformly ring-
labeled (^CFs position for trifluralin)
pesticides.
Because 2,4-D degradation was deter-
mined by measuring the evolution of
Soil-Water Content, cmj/cmj
Figure 2. Soil-water content (solid line)
and 2,4-D, terbacil and atra-
zine concentration distribution
in Eustis soil following infil-
tration of water to approximate-
ly 30-cm. Soil was initially
air dry and herbicide was in
top 1.5 cm of soil (2,000 ug/g
of soil).
Figure 3. Soil-water content (solid
lines) and terbacil concentra-
tion distributions in Eustis,
Cecil and Webster soils fol-
lowing infiltration of water to
approximately 30-cm. Soils
were initially air dry and
herbicide was in top 1.5-cm of
soil (2,000 yg/g of soil).
101
-------
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.
At 5,000 ppm, both forms were degraded, but
exhibited lag periods of 10 to 19 days for
the technical and formulated 2,4-D materi-
al, respectively. No significant degrada-
tion occurred before 50 days for
20,000 ppm, and a total of 11.5% and 21.5%
of the technical grade and formulated mate-
rial were degraded after 80 days of incuba-
tion (Ou et al. 1978a,b).
Unlike the Webster soil, 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. Small stimulations
were noted in the total C02 evolution
from the soil treated with 5,000 ppm of
formulated 2,4-D. In contrast to the Webs-
ter soil, very little degradation was ob-
served in the Cecil soil receiving 5,000
and 20,000 ppm. At a rate of 500 ppm,
both forms of 2,4-D were degraded slowly.
Only 10.5% and 6.3% of the technical and
formulated material were degraded, respec-
tively, in the Cecil soil after 80 days of
incubation.
Attempts were made to stimulate 2,4-D
degradation in the Cecil soil receiving
5,000 ppm of 2,4-D. In addition to 2,4-D
(technical grade or formulated) and 14C-2,
4-D, the Cecil soil was amended with vari-
ous nutrient sources. Readily degradable
nutrients such as yeast extract (1%) and
glucose (1%) plus urea (0.5%) did not
stimulate 2,4-D degradation. Only a small
stimulation, if any, occurred with these
treatments with the exception of the treat-
ment with lime plus 2,4-D degrading bacter-
ium (Ou et al. 1978a,b). In this treatment
28.6 and 3.9% of the formulated and techni-
cal grade 2,4-D, respectively, were degrad-
ed in 60 days.
Both technical grade and formulated
parathion at 25 ppm were degraded rapidly
to C02 in the Webster and Cecil soils as
indicated by 11+C02 evolution from uniformly
ring labeled ^C-methyl parathion. More
than half of the 14C-methyl parathion in
each soil was mineralized to 1LfC02 after
10 days of aerobic incubation. At the end
of the 32-day incubation period, 75 to 79%
of the ll4C-methyl parathion in the 25 ppm
treatments was mineralized to 11+C02. Of
the total 11+C-activity added to the Webs-
ter and Cecil soil receiving 25 ppm of
technical grade methyl parathion, 21 and
23%, respectively, could not be extracted
by the organic solvent mixture methanol-
acetone-benzene (1:1:1) and were termed
"bound residues." Similar results have
been reported by Lichtenstein, et al.
(1977).
When the methyl parathion concentra-
tion was 10,000 ppm, less than 0.1% of
14C-methyl parathion was mineralized to
14C02 in 52 days. These findings were
supported by the fact that nearly 100% of
ll*C-activity was extracted from each
10,000 ppm treated soil by the organic
solvent mixture. Total C02 evolution in
the 10,000 ppm treatments was reduced.
The ^C-actiyity remaining in the Webster
and Cecil soil receiving 10,000 ppm of
technical grade methyl parathion was 0.2
and 0.1%, respectively, after extraction
with the organic solvent mixture at the
end of the incubation period.
Atrazine and trifluralin at an appli-
cation rate of 10 ppm were mineralized
slowly in the Webster and Cecil soil. No
more than 5% of 14C-atrazine or ll+C-tri-
fluralin was mineralized to 11+C02 in the
10 ppm treatments after 80 days of aerobic
incubation. Less than 0.1% of 14C-
activity in the 10,000 ppm treatments was
mineralized to ^C02. Unlike 2,4-D and
methyl parathion, microbial activity in
soils generally was not affected by the
addition of 10,000 ppm of atrazine or
trifluralin.
Metabolites and Bound Residues
Only tentative identifications of
atrazine metabolites were made. Struc-
tures are based on thin layer and gas
chromatographic behavior. In the Webster
soil treated with 10 or 1,000 ppm, 4 to
8 percent of the radioactivity was asso-
ciated with a compound or compounds having
a relative Rf similar to that of 2-chloro-
4-amino-6-isopropylamino-s-triazine (0.80)
and/or 2-chloro-4-ethyl-amino-6-amino-s-
triazine (0.88). One to four percent of
the activity was associated with a com-
ponent having a relative Rf indistinguish-
able from that of 2-hydroxy-4-ethylamino-
6-isopropylamino-s-triazine (0.48;
102
-------
hydroxy-atrazine). Less than 2 percent of
the radioactivity was detected at a rela-
tive Rf similar to that of 2-hydroxy-4-
amino-6-isopropyl-amino-s-triazine (0.10)
and 2-hydroxy-4-ethylamino-6-amino-s-
triazine (0.10). In the Cecil soil, all
the radioactivity not associated with atra-
zine had a relative Rf corresponding to
hydroxy-atrazine (0.48). Percentages of
metabolites increased with time and corre-
sponded to a reduction in the level of the
parent compound.
The percentage of unextractable radio-
activity was generally higher at 10 and
1,000 ppm for the Cecil soil than for the
Webster soil; this difference was not ob-
served at the 20,000 ppm treatment. There
was no apparent difference in bound resi-
dues between the technical and formulated
applications at any of the three concentra-
tions studied. The percentage of bound
14C-activity increased with time and as
much as 30% of the lkC detected in the soil
was unextractable. In addition to the
trifluralin added to the Webster and Cecil
soils, four compounds were identified:
(1) a,a,a-trifluoro-2,6-dinitro-N-propyl-
p-toluidine, (2) a,a,a-trifluoro-2,6-
dinitro-p-toluidine, (3) 2-ethyl-7-nitro-
l-propyl-5-trifluoromethyl benzimidazole,
and (4) 2-ethyl-7-nitro-5-trifluoromethyl
benzimidazole. Each compound identified
was identical to authentic standards in
chromatographic behavior and mass spectral
fragmentation patterns. In addition to
the four metabolites described above, two
other unknown compounds were detected
(Wheeler et al., 1979).
No differences in metabolic rates or
pathways could be detected between the
Webster or Cecil soils or between the form-
ulated or technical trifluralin. The per-
centage bound was calculated by measuring
the organic solvent extractable ^Carbon
and the unextractable amount as determined
by combustion of the soil after extraction.
For the 10 ppm concentration, a greater
percentage of technical trifluralin was
bound to Webster soil than to Cecil. Ten
days after application, 10% was bound and
by 35 days it had risen to 30% and after
65 days of incubation, 72% was unextract-
able. The quantity of formulated tri-
fluralin bound in the Webster soil was
only measured after 68 and 84 days incuba-
tion, but showed a similar but somewhat
lesser (45 and 51%) amount of bound
residue than that measured for the tech-
nical material.
Recent reports by Katan et al. (1976)
and Katan and Lichtenstein (1977) show
rapid binding of the parathion amine ana-
logue which suggests that a similar
phenomenon may have occurred with triflur-
alin in the Webster soil. The parent com-
pound represented 90% or more of the
extractable 1HC, while considerable radio-
activity was not extracted. Thus, it is
possible that some metabolic products
were formed and a substantial portion of
them were bound. This would explain why
certain previously reported metabolites
were not detected in the extract or were
only found in small quantities. In an
effort to examine this possibility, tri-
fluralin, mono-dealkylated, and di-dealky-
lated derivatives were incubated in
sterilized Webster soil for four hours.
Percentages of each compound extractable
after four hours were 89, 75 and 57 for
the above compounds, respectively. Thus,
in the Webster soil there was a clear re-
lationship between the amount of non-
extractable material and the substitution
on the amino nitrogen. This is consistent
with the findings of Katan and Lichten-
stein (1977) with amino analogs of para-
thion. If one fortifies Webster soil with
either trifluralin or the di-dealkylated
derivative and immediately perform an
extraction, 93-95% of both compounds were
recoverable. Substantial binding did not
occur for the di-dealkylated derivative
in the sandy Cecil soil. Thus, it is
possible in the case of the Webster soil,
that metabolites containing secondary or
primary amino functional groups became a
part of the "bound" portion of the resi-
due. The absence of the same magnitude of
binding in Cecil soil could be the result
of different kinds and numbers of micro-
bial populations, etc. This could also
influence the formation of microbially
induced amino metabolites in the Cecil
soil.
In the majority of cases, parent
2,4-D was the only radioactive component
isolated from the soil. In a few cases,
2,4-dichlorophenol (based upon thin-layer
chromatography) was detected. It appeared
that more 2,4-dichlorophenol accumulated
at the 500 ppm concentration after 7 to
10 days of incubation than for other con-
centrations and was present to a greater
103
-------
extent in the Webster than in the Cecil
soil. Only the parent 2,4-D, however, was
detected after 29 days of incubation.
IMPLICATIONS OF PROJECT RESULTS WITH
REGARDS TO PESTICIDE DISPOSAL
Two significant conclusions can be
made regarding the adsorption results ob-
tained during this project. First, the
Freundlich adsorption equation described
all pesticide adsorption isotherms con-
sidered for solution concentrations up to
the aqueous solubility of the pesticide.
Thus, the pesticide adsorption sites for
all soils investigated were apparently not
saturated at any concentration considered
in this study. Second, contrary to a fre-
quent assumption, pesticide adsorption
isotherms were not linear, that is, N in
the Freundlich equation was generally less
than one. The nonlinearity of the pesti-
cide adsorption isotherm is an important
observation because it explicitly points
out that pesticides will be more mobile in
soils containing pesticide concentrations
similar to those associated with pesticide
waste disposal sites. This is especially
true for pesticides that are very soluble
in water (e.g., 2,4-D amine).
The increased pesticide mobility at
high pesticide concentrations limits the
usefulness of the available low concentra-
tion data base for developing "safe" man-
agement practices for pesticide disposal
procedures in soils. If a linear adsorp-
tion isotherm is assumed on the basis of
the low pesticide concentration data, one
underestimates the soil depth to which a
pesticide will leach or move for a given
water input. The seriousness of the fail-
ure of the low concentration data base to
describe the true mobility of a pesticide
as it moves toward the groundwater from a
waste disposal site depends upon the water
solubility of the pesticide and nonlinear-
ity of the adsorption isotherm. For
example, the adsorption isotherms for
atrazine and 2,4-D and a Eustis soil were
similar (see Table 3); however, 2,4-D
moved similar to an unadsorbed chemical as
it moved away from a simulated waste dis-
posal site, while the mobility of the
atrazine through the same soil was about
2.5 times less. To have assumed that the
adsorption isotherms were linear would
have resulted in a serious underestimation
of the depth to which each pesticide would
have moved for a given water input.
Many of the pesticides available on
the market today are biodegradable in
soils when applied at low concentrations
(0.5 to 10 kg/ha). However, many of these
same organic chemicals become persistant
when applied to soils at high concentra-
tions. It has been observed that some
soils are able to biologically mineralize
one pesticide (e.g., 2,4-D in Webster
soil), but the same pesticide may be per-
sistant in another soil (e.g., Cecil).
This study clearly points out that the
soil respiration rate of a soil receiving
high pesticide concentrations is not, in
general, a reasonable procedure for mea-
suring pesticide degradation potential.
Also, the apparent persistance of some
pesticides may be further confounded by
the formation of metabolites which are
"bound" (not extractable by recommended
procedures) to the soil and suggest a
greater apparent loss of the original
chemical than what actually occurred.
The contrast between the behavior of
soil environments containing large and
small pesticide concentrations illustrates
the importance and potential for manage-
ment of pesticide waste disposal sites.
Many soils frequently can be altered
chemically and/or biologically to enhance
their potential for biologically degrad-
ing pesticides. Also, microorganisms
capable of degrading specific chemicals at
high concentrations have been identified
and isolated which could be added to a
waste disposal site to enhance the degra-
dation and mineralization of a given
chemical. Because of increased pesticide
mobility at high concentrations, the
chemical may not, however, remain in the
vicinity of desired biological environ-
ment for degradation; thus, it is impor-
tant to manage both the water leaching
rate and biological environment for opti-
mum inactivation and efficiency of a pes-
ticide waste disposal site.
A major limitation in using the
results of this project is that only one
chemical was applied to a soil at a time.
Combinations or mixtures of pesticides
would be the common situation for a pesti-
cide waste disposal site. Waste disposal
sites receiving several pesticides may
fail to function as designed for a speci-
fic pesticide because of interactions
between chemicals and their environment.
The behavior of a pesticide mixture may
or may not be independent or additive, but
104
-------
rather based upon the influence of one pes-
ticide and/or the formulation associated
with a given pesticide. Problems which may
arise owing to the mixing of pesticides in
a waste disposal site should be considered
before site selection and management proto-
cols are defined by the United States
Environmental Protection Agency. This work
should include the evaluation of the major
surfactants used with pesticides and
various formulation chemicals.
REFERENCES
1. Atkins, P.R. The Pesticide Manufactur-
ing Industry—Current Waste Treatment
and Disposal Practices, Water Pollu-
tion Control Res. Series, 12020 FYE
61/72, 1972. 185 pp.
2. Bailey, G.W. and J.L. White. Factors
influencing adsorption, desorption, and
movement of pesticides in soil. Resi-
due Reviews 32:29-92, 1970.
3. Breazeale, F.W. and N.D. Camper. Bac-
terial, fungal, and actinomycete popu-
lations in soils receiving repeated
applications of 2,4-D-dichlorophenoxy-
acetic and trifluralin. Appl. Micro-
biol. 19:379-380, 1970.
4. Cole, M.A. Effect of long-term atra-
zine application on soil microbial
activity. Weed Sci. 24:473-476, 1976.
5. Davidson, J.M., L.T. Ou, and P.S.C.
Rao. Behavior of high pesticide con-
centrations in soil-water systems. In:
Residual Management by Land Disposal:
Proceedings of the Hazardous Wastes
Research Symp. (ed. W.H. Fuller).
EPA-600/9-76-015, U.S. Environmental
Protection Agency, Cincinnati, Ohio,
1976, p. 206-212.
6. Davidson, J.M. and R.K. Chang. Trans-
port of picloram in relation to soil
physical conditions and pore-water
velocity. Soil Sci. Soc. Amer. Proc.
36:257-261, 1972.
7. Davidson, J.M., C.E. Reick, and P.W.
Santelman. Influence of water flux
and porous material on the movement of
selected herbicides. Soil Sci. Soc.
Amer. Proc. 32:629-633, 1968.
8. Eno, C.F. The effect of simazine and
atrazine on certain of the soil
microflora and their metabolic pro-
cesses. Soil Crop Sci. Soc. Florida
Proc. 22:49-56, 1962.
9. Fields, T., Jr. and A.W. Lindsey.
Landfill Disposal of Hazardous
Wastes: A review of Literature and
Known Approaches. U.S. Environmental
Protection Agency, Publication S.W.
15, Washington, D.C., 1975. 36 pp.
10. Hamaker, J.W. The interpretation of
soil leaching experiments. In:
Environmental Dynamics of Pesticides
(eds. R. Haque and V.H. Freed).
Plenum Press, N.Y. 1975. p. 115-131.
11. Hamaker, J.W. and J.M. Thompson.
adsorption. In: Organic Chemicals in
the Soil Environment, (eds. C.A.I.
Goring and J.W. Hamaker). Vol. 1.
Marcel Dekker, Inc., N.Y. 1972.
p. 49-143.
12. Hubbell, D.H., D.F. Rothwell, W.B.
Wheeler, W.B. Tappan, and F.M.
Rhoads. Microbiological effects and
persistence of some pesticide combi-
nations in soil. J. Environ. Quality
2:96-99, 1973.
13. Hugenberger, F., J. Letey, and W.J.
Farmer. Observed and calculated
distribution of lindane in soil
columns as influenced by water move-
ment. Soil Sci. Soc. Amer. Proc.
36:544-548, 1972.
14. Kaiser, P., J.J. Pochon, and R.
Cassini. Influence of triazine herb-
icides on soil microorganisms. Resi-
dues Rev. 32:211-233, 1970.
15. Katan, J., T.W. Fuhreman, and E.P.
Lichtenstein. Binding of [14C]
parathion in soil: A reassessment of
pesticide persistence. Science 193:
891-894, 1976.
16. Katan, J. and E.P. Lichtenstein.
Mechanisms of production of soil-
bound residues of [1>fC] parathion by
microorganisms. J. Agr. Food Chem.
25:1404-1408, 1977.
17. Kay, B.D. and D.E. Elrick. Adsorp-
tion and movement of lindane in
soils. Soil Sci. 104:314-322, 1967.
18. Kirkham, D. and W.L. Powers, Advanced
105
-------
soil physics. Wiley-Interscience,
New York. 1972. 238 pp.
19. Lambert, S.M. Omega (n), A useful
index of soil sorption equilibria. J.
Agr. Food Chem. 16:340-343, 1968.
20. Lichtenstein, E.P., P.J. Katan, and
B.N. Anderegg. Binding of "persis-
tent" and "nonpersistent" 14C-labeled
insecticides in an agricultural soil.
J. Agr. Food Chem. 25:43-47, 1977.
21. Lindsey, A.W., D. Farb, and W.
Sanjour. OSWMP chemical waste land-
fill and related projects. In: Resi-
due Management by Land Disposal:
Proceedings of the Hazardous Waste
Research Symp. (ed. W.H. Fuller).
EPA-600/9-76-015, U.S. Environmental
Protection Agency, Cincinnati, Ohio,
1976. p. 14-31.
22. Newman, A.S. and C.R. Downing. What
pesticides do to soil. 3. Herbicides
and the soil. J. Agr. Food Chem.
6:342-353. 1958.
23. Ou, L.T., O.M. Davidson, and D.F.
Rothwell. Response of soil microflora
to high 2,4-D concentrations on de-
gradation and carbon dioxide evolution
in soils. J. Environ. Quality 7:241-
246, 1978a.
24. Ou, L.T., J.M. Davidson, and D.F.
Rothwell. Response of soil microflora
to high 2,4-D applications. Soil Bio.
Biochem. 10:443-445, 1978b.
25. Rao, P.S.C. and J.M. Davidson. Ad-
sorption and movement of selected
pesticides at high concentration in
soils. Water Res. 13:375-380, 1978.
26. Rao, P.S.C., J.M. Davidson, R.E.
Jessup, and H.M. Selim. Evaluation
of conceptual models for describing
nonequilibrium adsorption-desorption
of pesticides during steady-flow in
soils. Soil Sci. Soc. Amer. J. 43:
22-28.
27. Roslycky, E.B. Response of soil
microbiota to selected herbicide
treatments. Can. J. Microbiol. 23:
426-433, 1977.
28. Rouston, R.C. and R.B. Wildung. Ulti-
mate disposal of wastes to soil. In:
Water (ed. L.K. Cecil). Chem. Eng.
Progress Symp. Series 64.97, 1969.
p. 19-25.
29. Sanborn, J.R., B.F. Francis, and
R.L. Metcalf. The Degradation of
Selected Pesticides in Soil: A
review of published literature EPA
(in press) U.S. Environmental Pro-
tection Agency, Cincinnati, Ohio,
1977.
30. Shomaker, N.B. Current Research on
land disposal of hazardous wastes.
In: Residue Management by Land Dis-
posal : Proc. of the Hazardous Waste
Res. Symp. (ed. W.H. Fuller). EPA-
500/9-76-015, U.S. Environmental
Protection Agency, Cincinnati, Ohio,
1976. p. 1-13.
31. Stojanovic, B.J., M.V. Kennedy, and
F.L. Shuman, Jr. Edaphic aspects of
the disposal of unused pesticides,
pesticide wastes, and pesticide con-
tainers. J. Environ. Quality 1:54-
62, 1972.
32. Tyunaeva, G.N., A.K. Minenko, and
L.A. Pen'kov. Effect of treflan on
the biological properties of the
soil. Agrokhimiya 1974:110-114,
1974.
33. Van Genuchten, M.Th., J.M. Davidson,
and P.J. Wierenga. An evaluation of
kinetic and equilibrium equations for
predicting pesticide movement through
porous media. Soil Sci. Soc. Amer.
Proc. 38:29-35, 1974.
34. Voets, J.P., P. Meerschman, and W.
Verstraete. Soil microbiological
and biochemical effects of long-term
atrazine applications. Soil Biol.
Biochem. 6:149-152, 1974.
35. Von Everdingen, R.O. and R.A. Freeze.
Subsurface Disposal of Waste in
Canada. Tech. Bull. No. 49. Inland
Waters Branch, Dept. of Environment,
Ottawa, Canada, 1971. 19 pp.
36. Von Rumker, R., E.W. Lawless, A.F.
Meiners, K.A. Lawrence, G.L. Kelso,
and F. Horay. Production, Distribu-
tion, Use and Environmental Impact
Potential of Selected Pesticides.
EPA-540/1-74-001, U.S. Environmental
Protection Agency, Cincinnati, Ohio,
106
-------
1974. 439 pp.
37. Watts, R.R. Extraction efficiency
study—examination of three procedures
for extracting C-labeled and unlabeled
residues of organophosphorus pesticide
and carbaryl from bean leaves and
kale. Jour, of the Association of
Official Analytical Chemists 54:953-
958, 1971.
38. Weber, W.J. and P.J. Usinowicz. Ad-
sorption from Aqueous Solution, Tech.
Publication, Research Project 17020
EPA, U.S. Environ. Prot. Agency,
Cincinnati, Ohio. 1973. 187 pp.
39. Wheeler, W.B., G.D. Stratton, R.P.
Twilley, L.-T. Ou, D.A. Carlson, and
J.M. Davidson. Trifluralin degrada-
tion and binding in soils at high con-
centrations. J. Agr. Food Chem. 27:
702-706, 1979.
40. Wilkinson, R.R., G.L. Kelso, and F.C.
Hopkins. State-of-the-Art Report:
Pesticide Disposal Research. EPA (in
press) U.S. Environmental Protection
Agency, Cincinnati, Ohio, 1978.
237 pp.
41. Wolfe, H.R., D.C. Staff, J.F. Arm-
strong, and S.W. Comer. Persistence
of parathion in soil. Bull. Environ.
Contamin. Toxicol. 10:1-9, 1973.
42. Wood, A.L. and J.M. Davidson.
Fluometuron and water content distri-
butions during infiltration: Measured
and calculated. Soil Sci. Soc. Amer.
Proc. 39:820-825, 1975.
107
-------
INFLUENCE OF LEACHATE QUALITY ON SOIL ATTENUATION OF METALS
W.H. Fuller, A. Amoozegar-Fard, E.E. Niebla, and M. Boyle
Department of Soils, Water and Engineering
The University of Arizona
Tucson, Arizona 85721
ABSTRACT
The rate of movement through soil of constituents in waste leachates, whether from
municipal solid waste (MSW) or industrial wastes depends on the nature of: (a) the porous
medium through which solutions move (soil, geologic debris, clays, etc.). (b) the vehicle
transporting the constituent (leachate), and (c) the potential pollutant constituent itself
(Cd, Fe, Ni and Zn). This presentation is concerned with the rate of movement of Cd, Fe,
Ni, and Zn as correlated with certain measurable broad parameters of MSW landfill leachates.
The most prominent, readily measurable parameters have been identified as total organic
carbon (TOC), pH, concentration of soluble common salts, and iron (Fe). The influence of
these parameters on the prediction of movement of Cd, Fe, Ni, and Zn through soils is
discussed and quantitative data on rate of movement presented.
BACKGROUND
The establishment of mathematical
equations for predicting movement of metals
through soil and geologic debris depends on
the identification and quantification of
certain major components of the land dispo-
sal environment, Fuller et al.(1979a,b,c),
Alesii et al.(1978), and O'Donnell et al.
(1977). The major components are (a) the
soil and geologic debris, (b) the leachate
and (c) the polluting constituent, Fuller
(1977) and Fuller et al.(1979a,b). The ef-
fect of the most influential soil factors on
movement of heavy metals has been well re-
searched and reported earlier, Korte et al.
(1976, 1975), Fuller (1977), Fuller (1978),
O'Donnell et al.(1977). However quantitat-
ive data for specific potential polluting
constituents is needed. The purpose of the
research presented here, therefore, is to
extend the original soil formula as a user-
oriented predictive tool to include quanti-
tative data on prominent, readily measurable
MSW leachate characteristics. The cadmium
(Cd) prototype, presented in our report last
year, Fuller et al. (1979c), forms a base to
accomplish this purpose for nickel (Ni),
zinc (Zn) and to a lesser extent iron (Fe).
The suggested major MSW leachate prop-
erties affecting metal attenuation to be
quantified are (a) total organic carbon
(TOC), total soluble inorganic salts
(salt), and iron (Fe). Predictions for
metal movement then are to be based on
regression analyses because models based
on chemical reactions have rarely been
successful.
OBJECTIVES
The objectives, therefore, were to (a)
evaluate the effects of passing MSW leach-
ate through representative soil on the
attenuation of Cd, Fe, Ni, and Zn contained
in the leachate, and (b) relate attenuation
of the metals to specific parameters useful
for establishment of a field-oriented
predictive tool.
MATERIAL
Six soils were perfused with MSW
leachate enriched singly with Cd, Ni, and
Zn at a concentration of 100 ppm metal.
Some soil properties are reported in Table
1 and leachates in Table 2. These soils
have been more fully characterized by
Fuller (1978). They range in pH from 4.2
to 7.8 and clay percentages 1 to 52. The
soils were packed into 10-cm lengths of 5-
cm PVC pipe at known and reproduceable bulk
densities.
108
-------
TABLE 1. SOILS USED IN THE Cd, Fe, Ni, AND Zn ATTENUATION RESEARCH
Soil
Series*
Davidson
Molokai
Nicholson
Fanno
Mohave(Ca)
Chalmers
Ava
^Anthony
Mohave
Kalkaska
Wagram
River
all uvi urn
Soil
Order
Ultisol
Oxisol
Alfisol
Alfisol
Aridisol
Mollisol
Alfisol
Entisol
Aridisol
Spodosol
Ultisol
Entisol
Clay
52
52
49
46
40
31
31
15
11
5
4
1
Silt
23
25
47
19
28
52
60
14
37
4
8
2
Sand
25
23
3
35
32
13
10
71
52
91
88
97
Soil
Paste
PH
6.4
6.2
6.7
7.0
7.8
6.6
4.5
7.8
7.3
4.7
4.2
7.2
Cation
Exch.
Capacity
meq/lOOg
9
14
37
33
12
22
19
10
10
6
2
2
Elec.
Cond. of
Extract
ymhos/cm3
169
1262
176
392
510
288
157
328
615
237
225
210
Column
Bulk
Density
g/cm3
1.40
1.44
1.53
1.48
1.54
1.60
1.45
1.87
1.78
1.53
1.89
1.73
Soil
Surface
Area
m2/g
51.3
67.3
120.5
122.1
127.5
95.6
61.5
49.8
38.3
8.9
8.0
3.6
Predominant
Clay
Mineralst
Kaolinite
Kaolinite, gibbsite
Vermicul ite
Montmorillonite, mica
Mica, montmorillonite
Montmorillonite,
vermiculite
Vermiculite, kaolinite ,
Montmorillonite, mica
Mica, kaolinite
Chlorite, kaolinite
Kaolinite, chlorite
Kaolinite, mica
Oriented on basis of clay content
Listed in order of importance
^
-------
"
'• U\A
v
TABLE 2
MUNICIPAL SOLID WASTE LEACHATE II
CHARACTERISTICS
Constituent
Concen-
tration
pH 6.9
Electrical Conductivity (EC) mmhos/cm 5.1
Total Organic Carbon (TOC) - ppm 2000
Calcium - ppm 91
Magnesium - ppm 35
Potassium - ppm 653
Iron - ppm 39
Manganese - ppm 0.10
Phosphorus - ppm 1.00
Silicon - ppm 21.00
Zinc - ppm 0.46
MSW leachates were enriched to contain
ppm Cd, Ni, or Zn. Iron was evalu-
ated as a natural constituent, not enriched.
Total organic carbon (TOC) effects were
compared by using undiluted (100%), and
water-diluted (10, 25, and 50% original)
leachates. Each dilution was also brought
to levels of approximately 2,000, 1,000,
and 500 ppm of representative salts to eval-
uate the salt effect on Cd, Ni, and Zn
migration. Various salt levels were devel-
oped by adding CaCl2, Mg(N03)2, NaCl, and
KC1 to give the desired concentration at
the same cation ratio as the natural land-
fill leachate used. The treatments appear
in Table 3. Treatment of ^ 500 ppm salt
and 100% leachate was not made since the
natural leachate contained more than 500
ppm inorganic salt.
METHODS
MSW leachate was displaced through the
soil under saturated-anaerobic conditions
similar to that found for most landfills.
The fluxes ranged between 3 to 11 cm/day
as controlled by a peristaltic pump. Efflu-
ent displacements were collected in a frac-
tion collector initially every 2 or 3 hours
depending on the rate of migration of metal.
Both influent and effluent were analyzed
for the metal in question on a daily basis.
The initial leachate was lowered to a pH of
5.2 before metal enrichment, aqueous dilu-
tion and inorganic salt addition. The pH
of the leachate used in this study ranged
between 5.2-5.4.
The soil influent and effluent MSW
leachate was7 monitored for TOC, Ca, Mg, Na
and K according to standard atomic absorp-
tion procedures previously described (Korte
et al., 1976). The Cd, Fe, Ni, and Zn,
trace and heavy metals of the leachates were
determined by atomic absorption spectropho-
tometry according to the U.S. EPA Standard
Procedure (EPA 1974 and 1979).
RESULTS
Iron
The influence of MSW leachate constitu-
ents on the movement of indigenous Fe
through Kalkaska sand is illustrated in
Figure 1 to be significant. Natural Fe of
MSW Leachate II moved through soil1 in di-
rect proportion to the TOC, salts, and iron
concentration where levels of these constit-
uents ranged from 3,700-210, 10,000-2,000,
and 1,000-100 mg/A, respectively. One can
only speculate as to the reason for these
trends. For example, well known sequester-
ing constituents, such as fulvic and humic
acids, are included in the broad category of
TOC. Greater the concentration of these
compounds, the greater is the opportunity
for more rapid migration through soil.
Also, there must be a common mass ion
effect with such high concentrations of
inorganic salts. Therefore, one may expect
a greater rate of migration to be associ-
ated with the greater concentration or mass
of ions in solution. The iron effect is
not clear but one may expect there to be a
competitive effect for adsorption or reten-
tion soil surface sites the greater the
concentration of soluble Fe in the leachate.
Ca dmi urn
The same trend for attenuation of
leachate Fe in soil to be significantly
influenced by TOC, salts, and iron is ex-
hibited with Cd, Figure 2. The higher the
concentration of these three constituents,
the higher was the rate of migration through
Ava silty clay soil. Again, specific mech-
anisms of attenuation are not identified.
Nevertheless, broad parameters such as
these may be sufficiently sharp to be use-
ful tools in model development for pre-
diction of heavy metal migration rates.
1Soils other than Kalkaska sand not shown
because of the space problem associated
with the great volume of data.
110
-------
IRON CONCENTRATION '"I"'"' (C/C0)
CADMIUM CONCENTRATION ^1 (C/CO)
dtto6olujti» « to o M .fc
•
I rf 5
t
"111
- 414
oo
OVOVOOOOOC3
o^joocsooenen
oooooooo
poenpopoenpoeno
(noiotnioicovo
r-j
3
I
m
3
2
o
3-
(D
O.
i—
n>
o>
o
=r
o>
r+
0>
CO
O Ol
a —'
PO — ' — '
— IVD^^J
CTlPOO
— ' rU1
COPOO
OO— '00
471 U1 01 —' PO —•
CO 01 VO -P> CO OO
ro PO —i po 4^ oo
vo
en
co
O O O
Ul U1
o o
o
o
o
o
PO
en
PO
en
PO
en
en en
o o
en
O
o O
O O
ft)
o>
o
3"
a>
c+
n>
o
-s
a
a
o
o
0 £
-«• n>
—i 01
c o
O r+
3 n>
o
o
CO
3s
x
-O -H
rn o
2 >-i m
co o
ac co
m 70
t» 03
O O
O
cz
CO
o
-------
Nickel
As illustrated in Figures 3 and 4 both,
the quantity of organic constituents (TOC)
and common inorganic salt (ION) in the MSW
leachate significantly influenced the rate
of Ni movement through the soils.1 The
higher the concentration of these constitu-
ents, the higher was the rate of movement
and the lower was the Ni attenuation
through the soils. Reducing the TOC to •v
50 and 10% of the original leachate levels
proportionately decreased rate of movement
in the Ava silty clay, Figure 3. This
effect was less pronounced in the Davidson
clay, yet it was significant, Figure 4.
The influence of salt concentration appeared
to be more obvious with the Ava silty clay
than the Davidson clay despite the lower
clay content of the Ava soil. On the other
hand, there is a greater surface area avail-
able for adsorption and other reactions to
occur in the Ava (61.5 m2/g, total soil
surface area) than in the Davidson (51.3
m2/g, total soil surface area) soil. The
higher silt content of Ava silty clay (Ava
60%, Davidson 23%) more than compensated
for the lower clay (< 2p) content (Ava 31%,
Davidson 52%). Furthermore, the higher
UJ
O
1.0
0.8
0.6
0.4
0.2
0
1.0
0.8
j 0.6
: 0.4
0.2
0
i
1.0
as
0.6
OA
0.2
0
AVA
silty clay
I—+— TOC = 1950 mg/l
salt- 2010 "
0 2
6 8 10 12 14 16 18 20 22
/Jf
/fjf —+— TOC = 1190 w
ff salt= 2090
/
ig/l
TOC =
salt'
TOC -
salt=
1280
990
1140
600
12 16 20 24 28 32 36 4O 44
8 16 24 32 40 48 56 64 72 80 88
PORE VOLUME DISPLACMENTS
Figure 3. The influence of organic carbon (TOC) and salt (ION)
concentration in MSW leachate on Ml movement through
Aya sic. """
*Data for other soils not shown due to need
for economy of space.
in —
o
<
oc
1.0
0.8
0.6
0.4
0.2
0
1.0
0.8
0.6
0.4
0.2
0
1.0
0.8
0.6
0.4
0.2
0
DAVIDSON
clay
TOC = 1280 mg/t
salt* 990
TOC •
salt-
1140
600
2 4 6 8 10 12 14 16 18 20 22
PORE VOLUME DISPLACMENTS
Figure 4. The influence of organic carbon {TOC) and salt (ION)
concentration in f'SW leachate on Ni movement through
Davidson c. "^*
hydrous oxides of the Davidson clay did not
appear to fully compensate for the greater
surface area of the Ava silty clay.
The concentration effects of TOC and
salt (ION) on the quantity of Ni retained
per unit weight of soil at Breakthrough
(C/C0 = 1.0) are also indicated in Table 4.
The lower the level of these constituents,
the lower was the rate of metal migration
through the six soils. Particle size dis-
tribution and surface area effects are
indicated to favor greater attenuation in
the finer than coarser soils.
These effects of leachate quality are
reflected further in Ni migration velocity
through soils as reported for Davidson clay,
Ava silty clay, and wagram sand in Table 5.
Sands that are very low in clay such as the
Wagram s do not always conform to the pat-
terns of silty and clayey soils. This is
due primarily to the very rapid movement of
the leachates through the soils and the
very low retention of heavy metal contribu-
ting to a large error factor in measure-
ments.
Zi nc
Zinc reacted to different concentra-
tions of organic carbon (TOC) and inorganic
112
-------
TABLE 4
AMOUNT OF MSW LEACHATE II PASSED THROUGH SOIL COLUMNS AND QUANTITY OF NICKEL RETAINED
PER GRAM OF SOIL AT BREAKTHROUGH
W0% MSW Leachate
Soil
Davidson
Chalmers
Ava
Anthony
Kalkaska
Wagram
1*
Amt.
ml
2047
1954
1932
1152
2098
658
Ret.
yg/g
304
402
496
195
180
65
II
2
Amt.
ml
1583
2525
1647
1624
1437
671
Ret.
yg/g
261
571
334
217
139
38
50* MSW Leachate
3
Amt.
ml
1014
4411
1287
1145
1460
420
Ret.
yg/g
199
551
190
200
130
36
4
Amt.
ml
1274
3276
2779
1973
1937
1488
Ret.
jjg/g
338
779
429
302
259
90
II
25% Leachate II
5
Amt.
ml
1072
2940
3003
1612
1779
452
Ret.
yg/g
300
874
323
258
259
48
6
Amt.
ml
824
4511
1907
1683
1464
2218
Ret.
pg/g
184
632
344
240
195
135
7
Amt.
ml
1561
3423
4887
2205
1231
494
Ret.
yg/g
294
996
597
244
124
32
8
Amt.
ml
5216
7700
7818
5582
2929
808
Ret.
yg/g
640
1366
753
797
395
43
Refers to Leachate II composition of salt, TOC, and Ni as related to MSW Leachate II dilution recorded in Table 4.
TABLE 5
THE VELOCITY OF NICKEL THROUGH THREE SOILS AS INFLUENCED BY MSW LEACHATE II AT VARIOUS TOTAL ORGANIC CARBON (TOC) AND
SALT (ION) CONTENTS AT PORE WATER VELOCITY OF 10 cm/day.
Soil
Davidson
clay
Ava
silty
clay
Wagram
sand
Leachate
*
100
50
10
100
50
10
100
50
10
Salt
ppm
2000
1000
2000
1000
500
2000
1000
500
2000
1000
2000
1000
500
2000
1000
500
2000
1000
2000
1000
500
2000
1000
500
.1
1.33
1.50
2.00
1.17
1.14
1.66
1.22
0.90
1.30
1.06
1.71
0.90
0.69
1.13
0.91
0.43
3.66
7.27
5.64
4.67
4.29
5.67
6.18
4.01
.2
1.26
1.44
1.91
1.13
1.12
1.67
1.18
0.89
1.24
1.00
1.64
0.89
0.68
1.07
0.90
0.40
3.28
6.67
5.32
4.25
4.03
5.36
5.82
3.62
Velocity
.3
1.22
1.40
1.85
1.12
1.12
1.69
1.16
0.88
1.19
0.96
1.59
0.87
0.68
1.03
0.89
0.37
3.27
6.34
5.15
3.96
3.83
5.16
5.60
3.49
in cm/day if C/C0 equals ratio below:
.4
— velocity
1.18
1.37
1.81
1.12
1.11
1.66
1.13
0.87
1.16
0.43
1.54
0.86
0.67
0.95
0.88
0.34
2.84
6.09
4.97
3.68
3.67
5.01
5.42
3.34
.5
in cm/day
1.15
1.34
1.77
1.12
1.11
1.65
1.11
0.87
1.13
0.90
1.47
0.83
0.67
0.91
0.87
0.32
2.67
5.93
4.83
3.52
3.54
4.89
5.26
3.21
.6
1.13
1.30
1.73
1.12
1.11
1.65
1.09
0.85
1.11
0.88
1.43
0.78
0.67
0.90
0.87
0.29
2.47
5.81
4.67
3.34
3.41
4.77
5.11
3.07
.7
1.10
1.26
1.69
1.12
1.10
1.64
1.06
0.80
1.09
0.85
1.40
0.70
0.65
0.88
0.85
0.27
2.28
5.67
4.56
3.16
3.28
4.64
4.97
2.93
.8
1.09
1.20
1.64
1.13
1.07
1.64
1.02
0.70
1.06
0.84
1.36
0.66
0.57
0.78
0.77
0.23
2.05
5.69
4.40
2.90
3.14
4.52
4.80
2.72
.9
1.08
1.12
1.59
1.15
0.93
1.47
0.96
0.64
1.04
0.83
1.27
0.62
0.49
0.76
0.65
0.18
1.68
5.87
4.20
2.64
2.94
4.34
4.58
2.50
salts (ION) the same way as Ni despite the
slower migration rate of In through the
soil, Figure 5. At the lowest concentra-
tion of TOC -v 200 ppm, and salt ^ 500 ppm
in leachate, Zn moved several times more
slowly through soils than at the highest
concentration of ^ 2,500 ppm TOC and 2,000
ppm salts.
Furthermore, calculation of the veloci-
ties of Zn at different ratios of effluent/
influent or C/C0 concentrations according
to the Lapidus and Amundson (1952) model,
reveals and identifiable effect of differ-
ent TOC and salt concentrations on Zn move-
ment, Table 6.
113
-------
1.0
0.5
0
C D
1.0
= 0.5
K
C
N
1.0
0.5
A B
AVA si c
C D
WAGRAM •
16 24 32 40.
8
16 24 32 40
PORE VOLUME DISPLACMENTS
A. 1. 2,000 mg/Z TOC + 2,000 mg/i
salts.
B. T, 2,000 mg/1 TOC + 1,000 mg/l
salts
C. i, 1,000 mg/1 TOC + 2000 mg/2 salts
D. •v 1,000 mg/J. TOC + 1000 mg/l salts
E. % 1,000 mg/i TOC + 500 mg/i salts
Figure 5. The influence of MSH leachate concentration of organic carbon (TOC)
and Inorganic salts (ION) on the movement and breakthrough of In
through three soils.
TABLE 6
THE VELOCITY OF ZINC THROUGH THREE SOILS AS INFLUENCED BY MSW LEACHATE II AT VARIOUS TOTAL ORGANIC CARBON (TOC) AND SALT
(ION) CONTENTS FOR PORE WATER VELOCITY EQUAL TO TO en/day
Soil
Davidson
clay
Ava
silty
clay
Ma gram
sand
TOC
ppm
2510
2620
970
2510
2620
970
2510
2620
970
Salt
ppm
1990
1040
1990
1990
1040
1990
1990
1040
1990
.1
1.13
1.14
0.38
2.58
1.13
0.41
2.36
2.33
2.38
.2
1.07
1.07
0.38
2.50
1.05
0.38
2.32
2.75
2.24
Velocity
.3
1.02
1.02
0.38
2.46
0.99
0.32
2.28
2.27
2.15
in cm/day if
.4
velocity
0.97
0.97
0.38
2.43
0.97
0.29
2.26
2.60
2.08
C/C0 equals ratio
.5
in cm/day
0.94
0.93
0.36
2.40
0.89
0.28
2.25
2.62
2.04
.6
0.90
0.91
0.36
2.37
0.85
0.25
2.22
2.94
1.98
below:
.7
0.85
0.87
0.36
2.36
0.80
0.24
2.18
2.57
1.94
.8
0.79
0.83
0.36
2.35
0.76
0.21
2.14
2.57
1.91
.9
0.68
0.82
0.32
2.32
0.70
0.18
2.07
2.57
1.89
DISCUSSION
Leachate Quality Effects
The purpose of obtaining quantitative
information on leachate factors affecting
metal attenuation is to aid in providing a
suitable base for predicting attenuation
in soils. One type example of our computa-
tions, involving the Lapidus and Amundson
model is provided here. Cadmium is used as
the prototype. All leachate factors, TOC,
ION, and Fe, influenced migration rate of
Cd through six different soils. The fast-
est rate of movement was associated with
the highest concentration and the slowest
rate associated with the lowest concentra-
tion, we (Fuller and Alesii, 1979 b ) have
earlier suggested that
114
-------
1. At relatively low concentrations
of inorganic salts and soluble Fe there is
less competition for adsorption and other
reactive sites on the surfaces of the soil
particles than at high leachate constituent
concentrations.
2. At relatively low concentrations
of TOC there is less competition for ligand
sites in the leachate but also there is
less ligand available. The net effect on
migration will be a complicated synergic
relationship between the soil adsorption
sites, the ligand, and competing ions in
the leachate, and the metal under consider-
ation.
3o Ferrous iron as well as organic
molecules are active electron donors under
anaerobic conditions and at the pH range
involved.
PREDICTING METAL MOVEMENT
Table 7 gives the apparent Cd velocity
through soils. To illustrate the applica-
tion of these equations, suppose the soil
and the leachate have the following
properties:
1. Clay content = 21%
2. Free iron oxide = 1.2%
3. Soil porosity = 0.4
4. Total soluble ions in the leachate
= 0.15%
5. Total organic carbon of the leach-
ate = 0.6%
6. Concentrations of Cd in the leach-
ate = 4 ppm.
Suppose the depth of ground water is
50 meters, the Cd concentration of ground
water is zero and the Cd concentration
limit for the ground water is 1.25 ppm.
If the average infiltration is 1 cm/day,
how long will it take for the ground water
to reach the specified limit?
The pore water velocity is 1/0.4 = 2.5
cm/day and the relative concentration for
Cd would be 1.25/4 =.3J. Substitutions of
the information into equation for S>3 gives
the velocity of Cd at 1.5 ppm concentration.
S ., = [-.05(21)+. 14(1.2)-J4(1.2)(.6+. 15) +
•J .05(21)(.6+.15)]2.5
S , = K025 cm/day /j.
TABLE 7
THE APPARENT Cd VELOCITY THROUGH THE SOIL MAY BE APPROXIMATED BY THE FOLLOWING SET OF
EQUATIONS
S , = [ - 0.06 x clay + 0.16 x FeO - 0.16 x FeO x (TOC + ION) + 0.06 x clay x
(TOC + ION) + 0.81] x V
S 3 = [ - 0.05 x clay + 0.14 x FeO - 0.14 x FeO x (TOC + ION) + 0.05 x clay x
(TOC + ION) + 0.63] x V
S 5 = [ - 0.04 x clay + 0.12 x FeO - 0.12 x FeO x (TOC + ION) + 0.04 x clay x
(TOC + ION) + 0.55] x V
S 7 = [ - 0.03 x clay + 0.11 x FeO - 0.11 x FeO x (TOC + ION) + 0.03 x clay x
(TOC + ION) + 0.48] x V
S g = [ - 0.03 x clay + 0.09 x FeO - 0.11 x FeO x (TOC + ION) + 0.03 x clay x
(TOC + ION) + 0.39] x V
S 95= [ - 0.03 x clay + 0.09 x FeO - 0.11 x FeO x (TOC + ION) + 0.03 x clay x
(TOC + ION) + 0.36] x V
where S represents the velocity of the relative concentration x.
A
115
-------
Therefore, the time required for the con-
centration of 1.5 ppm to reach the 50 meter
depth would be
50x100/1.025 = 4878 day =13.4 years
It will take over 13.4 years for
the concentration of Cd in the ground
water to exceed 1.5 ppm limit.
The set of equations in Table 7 can
also be used to obtain breakthrough curves
at any depth within the profile. To do
this, substitute the soil and leachate
properties into the equations Sj through
S^gs and find the velocity of Cd movement
for the relative concentrations (C/C0) 0.1
through 0.95. Then, to find time, divide
the desired depth to the velocity for each
C/Co and plot the time vs. C/C0 for that
depth. Table 8 gives the velocity of Cd
for each one of the relative concentrations
of Cd for the soil and leachate described
previously. In addition, the time required
for each C/C0 to reach 50 and 80 meter
depths are given. The breakthrough curves
for the two depths are shown in Figure 6.
TABLE 8
VELOCITY VALUES FOR Cd AT GIVEN C/C0 AND THE TIME FOR WHICH THE C/CC
METER DEPTH
REACHES 50 AND 80
c/cc
0.1
0.3
0.5
0.7
0.9
0.95
Velocity (cm/day) 1.36
Time (years)
to reach a 50
1.03
.94
.82
.605
.503
depth of:
meters
80
meters
10.07
16.12
13.30
21,28
14.57
23.32
15.39
24.63
22.64
36.23
27.23
43.57
1.0
0.8
o
N.0.6
0.4
O 0.2
— — — 50 meters
80 neters
10 15 20 25
TIME : Years
30
35
40
45
Figure 6. Breakthrough curves at 50 and 80 meter depths for Cd.
116
-------
It should be stressed that these
equations need further refinements and
field testing before unqualified reliability
can be placed on their use by landfill
designers. We are in the process of such
refinements now,,
ACKNOWLEDGMENTS
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 805731-01 and
the University of Arizona, Soils, Water and
Engineering Department. Arizona Agricultu-
ral Experiment Station Paper No.
REFERENCES
Alesii, B.A., W.H. Fuller, and M.V. Boyle.
1978. Effect of leachate flow rate on
metal migration through soil. J. Environ.
Qua!.(1)9:
American Public Health Association. 1975.
Standard methods for the examination of
water and wastewater. 14th Ed. APHA,
Washington, D.C. p. 191-192.
Bernas, B. 1968. A new method for decom-
position and comprehensive analysis of
silicates by atomic absorption spectrome-
try. Anal. Chem, 11(40) :1682-1686.
Fuller, W.H. 1978. Movement of selected
metals, asbestos, and cyanide in soil:
Application to waste disposal problems.
Solid and Hazardous Waste Res. Rep. EPA-
600/2-77-020, Cincinnati, OH 243 p.
Fuller, W.H. 1977 . The importance of soil
attenuation for leachate control. In:
Waste Management Technology and Resource
& Energy Recovery. Proc. Fifth Natl.
Cong., Dallas, TX, Dec. 7-9, 1976. (Co-
sponsored by National Solid Waste Manage-
ment Assoc. and U.S. Environ. Protec.
Agency.)(EPA SW-22P. 1977. Washington,
D.C. 20402 pp. 297-320.
Fuller, W.H., B.A. Alesii, and G.E. Carter.
19793- Behavior of municipal solid waste
leachate: I. Composition variations. J.
Environ. Sci. Health, A14(6) :461-485.
Fuller, W.H. and B.A. Alesii. 1979b. Behav-
ior of municipal waste leachate: II. In
soil. Environ. Sci. Health, A14(7):559-
592.
Fuller, W.H., A. Amoozegar-Fard and G.E.
Carter. 1979C. Predicting movement of
selected metals in soils: Application to
disposal problems. In: Solid and Hazar-
dous Waste Research Div. Fifth An. Res.
Symp. Municipal Solid Waste: Land Dis-
posal. March 26-28, 1979, Orlando, FL.
MERL, U.S. Environ. Protec. Agency. EPA-
600/9-79-023a, Cincinnati, OH 45268. p.
358-374.
Heilman, M.D., D.L. Carter, and C.L.
Gonzales. 1965. The ethylene glycol
mono-ethyl ether (EGME) technique for
determining soil surface area. Soil Sci.
100:409-413.
Jackson, M.L. 1964. Soil clay mineralogy
analysis, p. 245-294. _In C.I. Rich and
G.W. Kunze (ed.) Soil clay mineralogy.
Univ. of North Carolina Press, Chapel
Hill, N.C.
Kilmer, V.J. 1960. The estimation of free
iron oxides in soils. Soil Sci. Soc. Am.
Proc. 24:520.
Korte, N.E., J. Skopp, W.H. Fuller, E.E.
Niebla, and B.A. Alesii. 1976. Trace
element movement in soil: Influence of
soil physical and chemical properties.
Soil Sci. 122:350-359.
Korte, N.E., J. Skopp, E.E. Niebla, and
W.H. Fuller. 1975. A baseline study on
trace metal elution from diverse soil
types. Water, Air, Soil Pollut. 5:149-
156.
Lapidus, L. and N.R. Amundson. 1952.
Mathematics of absorption in beds. VI.
The effect of longitudinal diffusion in
ion exchange and chromatographic columns.
J. Phys. Chem. 56:984-988.
O'Donnell, D.F0, B.A. Alesii, J. Artiola-
Fortuny and W.H. Fuller. 19770 Predic-
ting cadmium movement through soil as
influenced by leachate properties. U.S.
EPA Report.
U.S. Department of Agriculture. 1954.
Saline and alkali soils. Agric. Handbook
No. 60. U.S. Government Printing Office,
Washington, D.C.
U.S. Environmental Protection Agency. 1979.
Methods for chemical analysis of water
and wastes. Environ. Monitor. Support
Lab. 0
-------
PERCOLATION THROUGH WASTE COVER BY WATER BALANCE
U.
Richard J. Lutton
Geotechnical Laboratory
S. Army Engineer Waterways Experiment Station
Vicksburg, Mississippi 39180
ABSTRACT
A basic part of designing a new cover or evaluating the effectiveness of an exist-
ing cover on waste is a means of estimating the amount of water that percolates through.
The water balance method appears to be suitable for approximate design despite some short-
comings in modeling the actual process.
INTRODUCTION
This paper summarizes a portion of a
study on cover for solid and hazardous
waste by the U. S. Army Engineer Waterways
Experiment Station (WES). From 1977 to
1978, the study was concentrated on assem-
bling and developing guidance on the de-
sign and construction of cover for solid
waste. It was concluded from those ef-
forts that a reliable tool for predicting
percolation was needed for engineering
soil cover design. This paper reviews the
utility of one such predictive tool.
Background
The principal products of the WES
study to date are four reports and
papers.(1-4) The first report is an in-
terim presentation of the subject which
was distributed for external review.
Every conceivable function of soil and
alternative covers for solid waste was ad-
dressed, and recommendations for the sel-
ection of material and the design of cover
were made. The interim report also pro-
vided quick response to EPA's desire for
background as an aid to preparing guide-
lines. A separate, short paper(2) summa-
rized the general approach of this first
phase of the study.
The next effort of this study was
toward completing and publishing a compre-
hensive report on design and construction
of soil and alternative covers. The state-
of-the-art in soil mechanics and soils
construction was used to expand the interim
report and a final report has now been
published.(3)
In the future, as cover construction
becomes more and more guided by engineer-
ing design, an analytical tool for predict-
ing the amount of water percolating through
a cover will be needed. The background
and techniques for analysis of and
design for percolation through cover were
summarized at the Fifth Research Sympo-
sium. (4) As a part of further efforts to
improve the capability for predicting per-
colation the present paper offers a criti-
cal review of the most convenient predic-
tive method, i.e., by water balance.^5'
Water Balance Method
Water balancing^) amounts to record-
ing and analyzing the excesses and defi-
ciencies of water in soil (Figure 1) and,
thus, is fundamentally sound as a tool for
predicting percolation through a soil
cover. Water input as precipitation is
fairly well known from records at numerous
weather stations, but the other major
factors, surface runoff, evapotransipra-
tion, and soil water storage, are critical
because of their greater degrees of un-
certainty.
Figure 2 shows graphically the water
balance in monthly increments for 2 ft
of loamy soil cover at Cincinnati,
118
-------
EVAPOTRANSPIRATION
/DO
, 00 f
PERCOLATION
Figure 1. Schematic diagram of water
balance in cover soil
ACTUAL
EVAPOTRANSP/RA TION
JfMAMJJASOND
Figure 2. Water balance for Cincinnati,
Ohio(5)
Ohio. Also see Table 1. The monthly
rainfall means are based on a 25-year rec-
ord. The basic curve, for infiltration,
was obtained by subtracting a reasonable
runoff percentage from each monthly mean
precipitation amount. The actual evapo-
transpiration curve is obtained by adjust-
ing the potential evapotranspiration esti-
mates (derived according to Thornthwaite &
Mather (°') for the availability of water
in storage; thus in dry months soil water
in storage decreases and may cause actual
evapotranspiration to be less than poten-
tial levels.
APPRAISAL OF METHOD
The water balance method may be ex-
amined according to its sensitivity to
variations in the major factors and accord-
ing to its correspondence to the actual
movement and distribution of water in soil
in nature.
Conceptual Modeling
The water balance method is vulnerable
to criticism in regard to failures in model-
ing the actual infiltration and percolation
processes; e.g. the method makes no attempt
to model changing infiltration capacity rate
(Figure 3) which is known to vary as a func-
tion of time (for a constant source of water
at the ground surface). Instead of func-
tional relationships, the water balance
method provides for an accounting in which
increments of water are added and subtracted
for the chosen time interval. Thus, the
water that infiltrates over a period of a
month, for example, is simply the difference
between total precipitation falling in that
one-month period and a given percentage of
that precipitation considered to run off
the surface. Similarly, the water sub-
tracted as evapotranspiration is generalized
to a total amount for the one-month time
increment. The validity of the water bal-
ance method therefore depends on the valid-
ity of simple monthly additions and sub-
tractions to represent very complicated
processes that vary from day to day and
within a collection of storm events.
In a similar manner the ultimate in-
crement of water removed from the system
as percolation is calculated as a remainder
after an addition from precipitation and
f = fc+(f0-fc)e'
,-ct
INFILTRATION
CAPACITY CURVE
TIME
Figure 3. A general infiltration rate
function
119
-------
TABLE 1. WATER BALANCE DATA FOR CINCINNATI, OHIO
(5)
Parameter
M
0 N D Annual
PET
P
R/0
R/0
I
I-PET
5LNEG (I-PET)
ST
AST
AET
PERC
0
80
0.17
14
66
+66
150
0
0
+66
2
76
0.17
13
63
+61
150
0
2
+61
17
89
0.17
15
75
+58
150
0
17
+57
50
82
0.17
14
68
+18
(0)
150
0
50
+18
102
100
0.17
17
83
-19
-19
131
-19
102
0
134
106
0.13
14
92
-42
-61
99
-32
124
0
155
97
0.13
13
84
-71
-132
61
-38
122
0
138
90
0.13
12
78
-60
-192
41
-20
98
0
97
73
0.13
9
64
-33
-225
33
-8
72
0
51
65
0.13
8
57
+6
39
+6
51
0
17
83
0.13
11
72
+55
94
+55
17
0
3
84
0.17
14
70
+67
150
+56
3
+11
766
1025
154
872
+106
658
213
The parameters are as follows: PET, potential evapotranspiration;
P, precipitation; CR/Q surface runoff coefficient; R/0, surface runoff;
I, infiltration; ST, soil moisture storage; AST, change in storage; AET,
actual evapotranspiration; PERC, percolation. All values are in millimeters
subtractions for runoff and evapotranspira-
tion. For realistic simulation of the ac-
tual process, it would be better to use a
drainage function. In this way the impor-
tant permeability characteristics of the
soil could be involved. Otherwise, the
soil permeability does not enter into the
processes except in an indirect effect
on the quantity of surface runoff, i.e.
clayey soil does not transmit water as
rapidly as sandy soil but instead pro-
motes more surface runoff.
Sensitivity
A sensitivity analysis of the water
balance method helps to pinpoint strengths
and weaknesses for design purposes. A
somewhat poorly known factor in water bal-
ancing is the evapotranspiration. The un-
certainty inherent in using such calculated
factors as input reduces somewhat the con-
fidence that can be placed in the water
balance method. Potential evapotranspira-
tion calculated by different methods, how-
ever, seems to agree within a few tens of
percent and the method cannot be condemned
on this basis for approximate design work.
Table 2 presents some values of potential
evapotranspiration calculated directly
from pan evaporation data for comparison
to the values in Table 1 based on Thorn-
thwaite and Mather. The correspondence
is reasonably good.
The soil's capacity for storing water
is the second major factor in the water
balancing procedure that needs to be eval-
uated. Actual measurements of the capacity
of various soils in the remolded condition
and configuration of cover are scarce. At
the present time this input into the water
balance method is usually taken from some
authoritative source in the existing lit-
erature. There can be considerable dis-
crepancies among authorities in the actual
values to be used, and in addition the root
depth and condition of vegetation exert
modifying influences. Consequently the
available water storage space, i.e. between
the field capacity and the wilting point,
are indicated to vary by as much as a
factor of three. In either case it has
been shown(3) that the result as
120
-------
TABLE 2. POTENTIAL EVAPOTRANSPIRATION BY TWO METHODS
Potential Evapotranspiratlon, PET (mm)
Method JFMA M J J ASOND Annual
Adjusted
Pan
Measure-
ment*
75 102 105 119 91 61 45 -
Thorn-
thwaite
Estima-
tion** 2 6 11 37 90 107 115 124 72 56 32 - 652
*For Versailles, Indiana in 1975
**For Cincinnati, Ohio in 1975
percolation predicted by the water balance
method is not greatly affected by major
changes in water storage capacity in the
soil. Table 3 shows that increasing the
storage capacity of a cover soil 3-fold
may have only a modest effect on the pre-
dicted percolation.
One of the most critical decisions in
water balancing concerns the handling of
surface runoff. In this regard the choice
of the appropriate runoff coefficient is
most critical. Table 4 shows for the
Cincinnati example^) the effect of a range
of runoff coefficients. It can be seen that
the predicted percolation is quite sensi-
tive to the choice of this coefficient, and
the runoff coefficient may be the most im-
portant part of the whole design procedure.
Incidentally, this conclusion focuses at-
tention on the importance of designing the
cover configuration for efficient and rapid
runoff. Thus, the slope of the cover in
general should be sufficient to promote
rapid rill and sheet runoff and the main
ditches, on somewhat gentler slopes, should
be adequate for conveying the concentrated
runoff beyond the edges of the landfill.
USEFULNESS OF METHOD
This study has confirmed the usefulness
of the water balance method within certain
constraints. It has been found that the
method is simple but basically sound and,
therefore, could be used for making design
decisions. The range of variable conditions
usually encountered when working with natu-
ral materials such as soil generally limits
analytical design tools to providing conser-
vative estimates of stability, adequacy, and
so forth.
TABLE 3. PERCOLATION FOR RANGE OF STORAGE CAPACITIES
Water
Storage
Capacity,
ST (mm)
300
150
100
75
Percolation,
J
38
66
66
66
F
61
61
61
61
M
57
57
57
57
A
18
18
18
18
M
0
0
0
0
J
0
0
0
0
J
0
0
0
0
PERC (mm)
A
0
0
0
0
S
0
0
0
0
0
0
0
0
0
N
0
0
0
0
D Annual
0 174
11 213
38 240
56 258
At Cincinnati, Ohio (see Table 1)
121
-------
TABLE 4. PERCOLATION FOR RANGE OF RUNOFF COEFFICIENTS
Runoff
Coeffi- J
cient,*
CRO
Percolation, PERC (mm)
M
0 N
Annual
0.51,0.39 08 27 000000000 35
0.34,0.26 20 48 42 40000000 0 114
0.17,0.13 66 61 57 18 0 0 0 0 0 0 0 11 213
At Cincinnati, Ohio
*Separate coefficients were used for wet and dry seasons
This review had indicated that expect-
able variabilities in evapotranspiration and
soil water storage as used in water balanc-
ing have only modest effects on predicted
percolation. On the other hand the runoff
coefficient affects the predicted percola-
tion in an important manner. Therefore the
selection of the runoff coefficient should
receive a great deal of attention. It might
even be advisable to establish runoff-
infiltration test plots to measure runoff
and to compute runoff coefficients for
available cover soils. Otherwise the de-
signer is limited to using existing runoff
coefficients or tables of coefficients from
the literature where the variability is
considerable.
REFERENCES
1. Lutton, R. J. and Regan, G. L., "Selec-
tion and Design of Cover for Solid
Waste, Interim Report," U.S. Environ-
mental Protection Agency, Interim Re-
port, Cincinnati, Ohio, May 1978,
153 pp.
Lutton, R. J., "Selection of Cover for
Solid Waste," in Land Disposal of Haz-
ardous Waste, U. S. Environmental Pro-
tection Agency, Report EPA 600/9-78-016,
Cincinnati, Ohio, August 1978,
pp 319-325.
Lutton, R. J., Regan, G. L., and Jones
L. W., "Design and Construction of
Covers for Solid Waste Landfills,"
U. S. Environmental Protection Agency,
Report EPA-600/2-79-165, Cincinnati,
Ohio, August 1979, 250 pp.
Lutton, R. J., "Soil Cover for Control-
ling Leachate from Solid Waste," in
Municipal Solid Waste: Land Disposal,
U. S. Environmental Protection Agency,
Report EPA-600/9-79-023a, Cincinnati,
Ohio, August 1979, pp 234-240.
Fenn, D. G., Hanley, K. J., and
DeGeare, T. V., "Use of the Water Bal-
ance Method for Predicting Leachate
Generation from Solid Waste Disposal
Sites," U. S. Environmental Protection
Agency, Report SW-168, Cincinnati,
Ohio, 1975, 40 pp.
Thornthwaite, C. W. and Mather, J. R.,
"Instructions and Tables for Computing
Potential Evapotranspiration and the
Water Balance," Publications in Clima-
tology, Vol X, No. 3, Drexel Institute,
New Jersey, 1957, pp 185-311.
122
-------
EFFECT OF ORGANIC CHEMICALS ON CLAY LINER PERMEABILITY
A Review of the Literature
K. W. Brown and David Anderson
Texas Agricultural Experiment Station
Texas A&M University
Soil & Crop Sciences Department
College Station, TX 77843
ABSTRACT
Permeability is the primary criteria for evaluating the suitability of clay soils
for lining of waste impoundments. A review of available information has revealed the
near-complete lack of knowledge about the possible impact typical waste impoundment
contents have on the permeability of clay liners. The aims of this review have been to
develop a list of organic chemicals found in waste impoundments; develop a list of clay
minerals used to line impoundments, and review possible mechanisms for the failure of
these liners.
INTRODUCTION
Proposed regulations for disposal of
hazardous waste (USEPA, 1978) rely on the
permeability of a clay soil liner as the
primary criterion for judging the liners
effectiveness at preventing the movement
of toxic leachates into water below or
adjacent to a waste impoundment. No
single review is available that summarizes
and interprets the available information
on the possible influence of organic
liquids on the permeability of clay
liners. This report gathers and summa-
rizes available information to allow
better identification of cases where
organic liquids may alter the permea-
bility of clay liners. The basic
approach has been to identify failure
mechanisms and the organic liquids most
likely to cause the failure. The
information has been collected from a
wide range of disciplines including soil
chemistry, soil mineralogy, soil physics,
chemical engineering, environmental
engineering, petroleum engineering and
geology. Related subject matter found
useful included research reports on deep
well waste injection, secondary oil
recovery, soil water repellency, and clay-
organic chemistry. Possible failure
mechanisms discussed are catagorized as
(1) Dissolution (2) Volume changes
(3) Soil piping and (4) Miscellaneous.
ORGANIC CHEMICALS CONSIDERED
Essentially all available literature
on the behavior of organic chemicals in
soils relates to systems where water is
the dominating fluid (Goring et al.,
1972) . Water is viewed as the carrier
fluid and the organic chemical is in
trace quantities. In the case of clay
liners in direct contact with
concentrated organic fluids, the various
equations for organic adsorption by
clay minerals have limited usefulness.
A study is now underway to attempt
correlation of various chemical proper-
ties with possible failure mechanisms.
Table 1 is a list of selected organic
fluids and the properties most likely
to relate to their behavior in clay
liners. Organic fluids placed in waste
impoundments cover the spectrum of
chemical species ; the list is an attempt
to select the most prevalent and
representative of these compound types.
The groupings of the organic chemicals
which will be considered are acids,
bases, neutral-polar and neutral-nonpolar.
123
-------
TABLE 1. SELECTED CHEMICALS AND TJiEIR PHYSICAL PROPERTIES
ORGANIC
Forms
AC id
aclu
Phenol
Base
Aromatic
Ar-tne
Alky!
AT- me
HP tero-
cvcl'.c
A-rlne
CH1MCAL
Example
s
Phenol
Aniline
Dodecyloammoniuii
Pyricline
3P(°C)
118
182
184
ND
115
m
17
43
-6.2
ND
-42
Sn. Gr.
1.049
1.072
1.021
ND
.981
Mol
60
94
93
ND
79
. Wt
.05
.11
.13
.0
Dif
Cc
20°C
ND
ND
6.89
N'D
ND
•lectric Water Sol.
instant R"1/!
25°C 60^0
ND
ND
ND
ND
12.3
ND
9.1
NU
ND
ND
infinite
82
34
ND
ND
Pressure
at 20°C
nm/Hg
14.4
0.2
0.3
ND
14.0
Neutral-Polnr
Alcohol
Aldehyde
Clycol
Al kvl
HjJlde
Ke tone
Neutral-No
Alk.mf
Aronatic
AlVyl
benzene
Methan.il
Butyr Aldehyde
Ethyl ene Glyeol
Chloroform
Ethyl MutUv v'ctoT\e
n Polar
H»ptane
Iteupenc
X ylene
65
75
198
61
80
98
80
139
-98
-99
-13
-63
-6R
-9!
5.5
-47
.792
.817
1.108
1.491
.805
.683
.878
.868
32
72
62
119
72
100
78
106
.02
.10
.07
.38
.11
.21
.11
.16
ND
ND
ND
ND
ND
ND
2.28
2.37
32.6
13.4
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
8
353
0.003
1.780
0.198
92.0
71.0
0.05
160.0
77.5
53.0
76.0
6.5
CLAYS USED FOR CLAY LINERS
Clay Minerals - Introduction
Identification of clay minerals and
their interaction with organic compounds
has been accomplished by the use of Xray-
diffractometry, and Infra-red spectroscopy
(Greenland, 1965; Little, 1966; Theng,
1974). The clay minerals most often used
for the lining of waste impoundments are
Kaolinite, Illite or Montmorillonite.
Considerations of upmost importance are
the changes on the interlayer spacing and
the consequential volume changes exhibited
by each of these clays when exposed to
organic chemicals that are likely to
appear in a waste impoundments leachate.
Each clay mineral is briefly described
here along with a characterization of its
interlayer properties. Table 2 gives
the relative size of the three clay
minerals being studied and also gives
typical values for several other
properties of these clays.
Kaolinite
Non-expansive
1:1 Lattice
Illite
Non-exp ans ive
2:1 Lattice
Montmorillonite
Expansive
2:1 Lattice
UNIT
Contracted
(nm)
200.0
20.0
2.0
PARTICLE THICKNESS Charge Per Surface Area
Hydrated* A Volume formula Wt. M /gm
(nm)
202.0 1% 0 8
22.0 10% 1.0 80
6.0 200% .5 800
Pure Mineral
Exchange Cap.
meq/lOOg
Cation Anion
10 pH
dependant
15 pH
dependant
100 pH
dependant
<5
* Four water layers adsorbed for each available basal surface.
124
-------
.
Kaolinite -
Kaolinite interactions with organic
chemicals have been the subject of
several studies (Conley et al., 1971;
Olejnik et al., 1970; Carr et al. , 1971;
German et al., 1969). Kaolinite particles
have a small negative surface charge.
Adjacent layers of this mineral are
strongly held together by hydrogen
bonding. For this reason, kaolinite
exhibits little interlayer expansion and
adsorption of organic chemicals is largely
confined to external crystal surfaces.
Edges of the crystals exhibit broken
bonds which give rise to a small number of
highly pH dependant positive or negative
charges.
Illite -
Illite has been shown to interact
with organic chemicals (Macintosh et al.,
1971; Greenland et al., 1965; Grim et al.,
1947). Interlayer fixation by potassium
ions usually permits no expansion of the
space between illitic interlayers.
Macintosh et al., (1971) however pointed
out that one organic cation could replace
these fixed potassium ions. This case
is probably rare and exists only because
the organic cation (dodecyclammonium)
closely approximates the size and charge
of potassium. Adsorption of organic
chemicals will thus occur largely on the
external crystal surfaces of this clay.
Illites surface area , cation exchange
capacity (CEC), and shrink-swell are
intermediate between that of kaolinite
and montmorillonite. The charge density
on illite is the highest for the clays
considered but due to its neutralization
with non-exchangeable potassium ions, its
CEC does not reflect this.
Montmorillonite -
The interaction between organic
chemicals and montmorillonitic clays has
been extensively studied (Parfitt, et al.
1968; Bradley, 1945; Hoffmann, et al.,
1961; Greenland, 1963; Greene-Kelley,
1945; Theng, et al., 1967). This mineral
has the smallest unit particle diameter
of the three clay minerals considered.
It also readily adsorbs polar or
positively charged organic species on its
interlayer surfaces. For these reasons,
montmorillonite exhibits the greatest
CEC, surface area, and shrink-swell
-*?.
potential. Montmorillonite can adsorb
over 200% of its solid phase weight and
consequently has the capacity for large
shrinkage if its interlayer water is
displaced by other fluids that yield a
lower interlayer spacing. Montmorillo-
nite has an intermediate charge density,
but since its cations are exchangeable,
its exhibits the highest CEC.
PERMEABILITY OF CLAY LINERS TO ORGANIC
CHEMICALS
Clay soil permeability (K) is a
numerical value representing its ability
to transmit fluid. From Darcy's Law for
liquid flux through a porous media, it
can be seen that a soil K value is
independant of the volume of soil, the
volume of fluid passing, and the
hydraulic gradient moving the water:
J =
.«,
Flux of a fluid
Flow (cm-Ysec)
i 2
/cm -sec
where;
J
Q
K = Permeability coefficient (cm/sec)
H = Hydraulic head cm and
A = Crossectional area of flow (cm^)
and since:
Q = |
where:
V
T
then:
K
Volume of fluid (cm ) and
Time (sec)
V
ATH
Any change in the value of K will
represent a change in the permeability
of the media. For convenience of
comparing the flow of fluids other than
water, it is desirable to separate out
the properties of the fluid which
influence its movement.
K
where:
n
v
ATH
(n)
gY
Viscosity of the fluid
(gram/cm-sec)
density of the fluid (g/cm3)
gravitational constant
Viscosity normalizes a fluids
resistance to flow due to its cohesive-
ness, while the fluids density value
normalizes the effect of gravity on its
flow.
125
-------
As part of the current study, the
influence of selected organic chemicals
on the permeability of clay liners is
being tested. Central to these tests
has been the development of a constant
pressure permeameter suitable for
permeability measurements on compacted
clay liners subjected to organic
chemicals.
FAILURE MECHANISMS
Clay Dissolution
Dissolution of a clay liner is
brought about by an infiltrating chemical
that dissolves the exposed surfaces of
a pore or channel. Either organic acids
or organic bases may solubilize portions
of the clays structure. Acids have
been reported to solubilize aluminum,
iron, alkali metals and alkaline earths
while bases will dissolve silica (Grim,
1973). Since clay minerals contain both
silica and aluminum in large quantities,
they are susceptible to partial dissolu-
tion by either acids or bases.
Clay Dissolution by Acids -
Pask, et al., (1945) boiled several
clays minerals in acid and found the
percent solubilization of alumina was 3%
from kaolinite, 11% from illite and
greater than 33% from montmorillonite.
Grim (1968) found the solubility of clays
in acid "varies with the nature of the
acid, the acid concentration, the acid-
to-clay ratio, the temperature and the
duration of treatment." He also found
that the action of an acid on clay was
enhanced when the acid had an anion
about the same size and geometry as a
clay component. This would permit even
veak acids to dissolve clays under some
conditions.
Hurst (1970) found that the permea-
bility of geologic formations could be
increased by pumping in acetic or formic
acid. Johansen, et al., (1951) reported
flow increases for water wells following
their treatment with a solution containing
citric acid. Grubbs, et al., (1972)
cited acid waste as the probable causal
agent in the permeability increase of
carbonate-containing minerals. Xray
diffraction studies of the four clay
minerals injected with acid waste showed
them to be dissolved or completely
altered. Diffraction peaks showed the
most variability with montmorillonite
clays.
Acidization is the name used for
the process of permeability increase by
acid mineral dissolution. This process
is widely used to increase the
permeability and hence the productivity
of oil wells (Sinex, 1970).
An everpresent source of organic
acids in waste impoundments are anerobic
decomposition by-products. These include
acetic, propionic, butyric, isobutyric
and lactic acid. Anerobic decomposition
will yield the carboxylic acid deriva-
tives of whatever organic fluids are
placed in the impoundment.
Material that incrusts at the base
of wells used to inject waste usually
consist of calcium, magnesium and iron
carbonates, along with imbedded sand and
clay particles. In order to remove the
carbonate compounds, they must be dis-
solved and then held in solution
against precipitation forces. Dis-
solution is usually accomplished with a
strong acid. At this point, calcium
will reprecipitate (as calcium sulfate in
the presence of sulfuric acid) unless it
is chelated and remove by a flowing
fluid. Chelating agents effective at
preventing reprecipitation of various
carbonates are Citric acid, Tartaric
acid, and Glycolic acid (Anonymous, 1977),
Clay Dissolution by Base -
Bases also have been implicated in
the dissolution of clay liners. Haxo,
Jr., (1976) found in preliminary tests
that Bentonite liners allowed the
passage of both strong acids and strong
bases in a short period of time.
Nutting (1943) showed even extremely
dilute solutions of alkali to be
effective at removing silica from
smectites by dissolution from the
crystal lattice. As part of this study,
the effects of three organic bases on
clay liners are being investigated.
Volume Changes -
Volume changes in clay liners occur
when there is a change in the water
content of the clay. Adsorption of
water on external surfaces occurs with
all three clay minerals. For a given
change in water content, the magnitude
126
-------
of volume change is dependant on the
clay mineral type, the arrangement of
the clay particles, the size of the clay
particles, the surface area per unit
weight of the clay, and on the kind and
proportion of cations adsorbed to the
clay. From Table 2, it can be seen
that montmorillonite (and to a lesser
degree illite) may cause problems
associated with changes in the volume
of clay liners. Two contracted lattice
sheets of montmorillonite have a 2 nm
thickness. The adsorption of water on
montmorillonites interlayer surfaces
gives this type of clay soil the
potential for greater than 200%
difference in volume between the
dehydrated and hydrated state.
On the other hand, Gieseking (1939)
reported that montmorillonite lost its
ability to swell when exposed to organic
cations.
Extraction of Interlayer Water -
Extraction of interlayer water
causes the shrinking and associated crack-
ing exhibited by montmorillonitic soils
(Baver, 1972; Grim, 1968). Cracking is a
result of the clay undergoing three
dimensional shrinkage. Where the rate of
water extraction is not uniform, cracks
will form in wet soil (Young, et al.,
1975). The water content of a clay liner
will change if an organic leachate dis-
places the water from the clay liner. To
more fully understand how much water will
be displaced by a given organic fluid, it
is necessary to first understand the
nature and forms water takes in a clay
liner.
Consider a case in which the clay
particles are initially surrounded by
multiple layers of water. The thickness
of the water between adjacent
montmorillonite lattice sheets would
effect the plasticity, interparticle
bonding, compaction and water movement
within a clay liner. These properties
change as the thickness of the interlayer
water changes (Young, et al., 1975).
Examination of the forces holding water
to the interlayer surfaces in montmori-
llonite will assist in predicting the
effects of an intruding organic leachate
on the interlayer spacing.
It is widely believed that water
layers immediately adjacent to montmori-
llonite interlayer surfaces are non-
liquid, hexagonally structured, and held
much more strongly than water layers
further out from the surface (Grim, 1968).
The thickness of this structured water
varies depending on the adsorbed cation.
Montmorillonite has a structured water
thickness of 1 nm or four water layers
per clay surface where calcium is the
adsorbed cation. Sodium montmorillonite
has a structured water thickness of .75
nm or 3 water layers on its surfaces
(Grim, et al., 1945). Structured water
should then yield an interlayer spacing
of 1.5 nm and 2.0 nm for sodium and
calcium montmorillonite respectively.
Glaeser (1949) found a similar relation-
ship to hold for montmorillonite where
acetone was the interlayer fluid.
Glaeser found the interlayer spacings
to be 1.25 nm and 1.51 nm for a sodium
and calcium montmorillonite respectively
after the clay was dehydrated and
subsequently exposed to acetone.
These surface-bound layers of water
are held strongly to the clay. They
represent however only a part of the
interlayer water. The water layers
further out from the clay surfaces are
held in place by hydrogen bonds. The
hydrogen bonded water extends back to the
structured water layers anchored to the
clay surface. These outer layers of
water would easily be displaced by an
intruding fluid. If the fluid lacked
water's large dipole moment or its
ability to hydrogen bond, a decrease in
interlayer spacing would probably result,
since fewer layers of organic molecules
would be retained. If the intruding
fluid had a higher affinity for the clay
surface than the structured surface
layers of water, large decreases in
interlayer spacing should be possible,
since some of the more tightly bond water
would be removed.
Clay Wettability -
When a fluid adsorbs to a surface,
that surface is said to be wetted by the
fluid. Clay surfaces in a clay liner
are initially wetted with water. If a
permeating organic fluid has a higher
affinity for the clay surface, the clay
may become organic-wet. Petroleum
engineers commonly refer to surfaces with
oil adsorbed as being oil-wet (Raza,
et al., 1968). Water and oil are said
to compete with each other for solid
surfaces in oil reservoirs. A
127
-------
quantitative measure of the preferential
wettability of a clay surface for water
and an organic fluid can be represented
as the difference between the water-
clay and organic-clay interfacial
energies as represented by the Young-
Dupre equation (Admas, 1941).
(y org:c) - (y w:c) = (Yorg:wXCos 9)
where:
Y org:c = interfacial energy
between the organic and
Y w:c = interfacial energy between
------ - --- - — o
clay (dynes/cm)
j-nterfacial energy UCLWCCH
the water and clay (dynes/cm)
Yorg:w= interfacial energy between
the organic and water (dynes/
cm)
6= angle at the organic relay
water interface measured
through water (degrees)
There is no direct method for
measuring either the organic relay or the
waterrclay interfacial energy (Raza, 1968).
Their difference, however, is equivalent
to the product of the water:organic inter-
facial energy and the cosine of the waterr
clayrorganic contact angle (6).(Figure 1).
The contact a n ^ 1 e ( •;) m>'.
cne wetting " 1 LI 1 d 13 1 .^.
che contact- .in^le vere
D n — k
90°
Displacement of Interlayer Water -
If an intruding organic leachate
displaces the anchored water layers,
there will be no forces holding the
balance of the interlayer water against
gravitational drainage forces. In this
case montmorillonite interlayer spacing
(d(001))would decrease from »2 nm to
whatever the interlayer spacing value
would be for the newly adsorbed organic
fluid. Table 3 is a list of the inter-
layer spacings for montmorillonite after
exposure to organic fluids. There is an
abundance of xray diffraction data for
clay minerals with correlations to the
interlayer cation, dehydration temperature
and the immersion fluids pH, dielectric
constant and concentration (Grim, 1968;
Theng 1974; Barshad, 1952). The useful-
ness of this data is limited, however,
because as a matter of standard procedure,
the clay minerals are initially
dehydrated. In order to more closely
simulate the situation in a clay liner,
there needs to be a series of xray
diffraction studies on hydrated clay
minerals. First, xray diffraction data
is needed for the clay liner saturated
with water as it would be prior to the
impoundment of a waste. Secondly, xray
diffraction data is needed for the same
clay liner after organic leachates have
permeated the clay. If a substantial
interlayer spacing decrease is observed
for the second set of xray diffraction
data, shrinkage cracks may be anticipated
for that particular liner-waste combina-
tion.
In a waste impoundment, the affinity
an organic leachate has for the hydrated
clay surfaces will determine if it will
displace the water. Since the clay
surface is negatively changes, organic
leachate components that are positively
charged or uncharged but polar will have
an affinity for clay surfaces. Since
the water on the clay's surface is several
layers thick, water solubility will also
improve an encrouching fluids access to
the clay surface. The strength with
which water is held to clay surfaces will
vary with the cations adsorbed to the clay
and th>. clay's charge density. These
factors may change from one place to
another in a clay liner. In order to
predict interlayer spacing and permea-
bility of inplace clay liners, a series
of xray diffraction studies should be
undertaken on the range of liner-waste
combinations likely at a given site.
After a leachate is in place in a liner,
the charge density of the clay will
effect the resulting interlayer spacing.
WeissC1963) found the interlayer spacing
after exposure to alkylammonium ions to
be 1.3 nm, 1.9 nm and 2.76 nm for
montmorillonites with low, medium, and
high charge density respectively.
It should be understood that
128
-------
TABLE 3. INTERLAYER SPACINGS OF MONTMORILLONITE-ORGANIC COMPLEXES
Compound
Benzene
Benzene
Me than ol
Ethanol
Ethanol
N-Butanol
N-Butanol
N-Butanol
N-Decanol
Paraffin
Paraffin
H20
Me thy Ethyl
Ketone
2.1M 6 Alanine
Interlayer Spacing
Ca. Sat. Na Sat.
nm
1.0
1.52
1.71
1.70
1.70
1.52
1.45
1.45
3.68
1.45
.99
1.92
1.73
2.05
.99
.99
ND
1.34
1.35
1.32
1.32
1.32
ND
ND
ND
1.87
ND
ND
Dielectric
Constant
20°C
2.3
2.3
32.35
25.00
25.00
17.70
17.70
17.70
5.0
2.10
2.10
78.5
18.85
150.0
DiPole
Moment*
0
0
1.66
1.70
1.70
1.66
1.66
1.66
1.7
0
0
ND
2.74
ND
Dehydrated
temp.
°C
250
150
250
170
250
20
170
250
250
20
250
250
250
250
* (x W~*-° Electro Static Units)
From Barshad (1952).
organic leachates may cause either
shrinking or swelling of clay liners.
Barrier (1978) reported swelling of
montmorillonite clays as a result of
exposure to several forms of acetonitrite,
xylene, cyclopentane, alcohols,glycoIs and
ketones. A liner which has swelled and
heaved may loose its intergrity during
heaving, or it may shrink later when
water replaces inbibed organic chemicals.
Soil Piping -
Underseepage as the result of soil
piping is an ever present danger in
earthen dams. Mansur, et al. , (1956)
describes piping as "the active erosion.. .
of soil from below the ground surface
which occurs as a result of substratum
pressure and the concentration of seepage
in localized channels." Jones (1978)
found the early stages of piping develop-
ment to be associated with vertical
contrasts of the structure and permea-
bility in soils. Soil piping was also
associated with shifts in a soils pore
size distribution toward macropores with
no corresponding change in total porosity.
A reactive fluid may enlarge the surface
area of a pore by dissolution of the
pore wall and by the eventual dissolu-
tion of the soil matrix between two
pores. While a fluid's reactivity is
reduced by its action on the pore wall,
the size increase in the pore will
increase the rate of delivery of the
fluid to the pore. In this manner, any
difference in the pore size distribution
of a clay liner may be magnified with
time. Schechter et al., (1969 found
that wormhole formation was the result
of a reactive fluids preferential flow in
larger pores. He went on to say that a
quasi-equilibrium is reached where
further growth in a pore is limited by
the rate of difffusion of the reactive
fluid.
Mitchell (1975) found that quick
clays were often associated with the
presence of organic compounds possessing
strong dispersing characteristics.
Quick clays act as viscous fluids and
hence have no structural strength. Quick
clays are susceptable to erosion caused
be seepage. Seepage by reservior waters
129
-------
into dams have been reported to have
caused dispersive piping and eventual
tunneling all the way through earth dams.
Tunneling was reported to occur in soils
with a local permeability of 1 x 10~5
cm/sec.
Differential solution and subsequent
leaching, especially with calcareous
sediments, was reported to result in the
formation of channels, sink holes, and
cavities (Mitchell, 1975).
Cedergren (1967) reported that
differential leaching of limestone,
gypsum and other water soluble mineral
components can lead to development of
solution channels that get larger with
time and substantially increase
permeability.
He warned not to underestimate the
importance of minor soil and geologic
details on the permeability of soil
formations as they cause the majority of
failures in dams, reservoirs and other
hydraulic structures.
Cedergren (1967) concluded that
most failures caused by seepage can be
placed in two catagories:
(1) those that are caused by soil
particles migrating to an
escape exit, causing piping
and erosional failures;
(2) those caused by uncontrolled
seepage patterns which leads
to saturation, internal flood-
ing, and excessive seepage.
Crouch (1978) found that so called
tunnels, tunnel-gullies, or pseudokarsts
will develop in dispersive soils where
the soil-colloid bond strengths are low
compared to the energy of water flowing
through the soil. He found dispersive
soils or those with low structural
stability have been associated with
tunnel erosion throughout the world.Other
factors found to. be related to tunnel
erosion were ESP (Exchange sodium per-
centage) , soil cracks, low permeability
and hydraulic gradients.
In a study of the variables effect-
ing piping, Landau, et al., (1977)
noted a strong interaction between the
chemical composition of the eroding
water and compaction water content.
Ion concentration seemed to have little
effect on soil piping suseptibility for
mixed illitic and kaolinitic clay loam
compacted dry of optimum. For the same
soil compacted wet of optimum, soil
piping suseptibility was highly related
to ion concentration in the eroding
water. Landau (1977) postulated that the
wet of optimum compaction produced a more
parallel orientation of the soil particles
which increased the effect of osmotic
repulsion. Consequently the ion
concentration of the soil water would
have a relatively greater effect on
dispersive forces in a soil compacted wet
of optimum. When low ion concentration
eroding water is combined with wet of
optimum compaction, Laudau, et al., (1977)
reported an exceptionally low resistence
to internal erosion. These findings are
felt to be especially important due to
the long standing practice of compacting
clay liners wet of optimum to produce
the minimum initial permeability
possible for a given compactive effort.
Piping involves the slaking of soil
particles. Slaking is defined as the
disintegration of unconfined soil samples
when submerged in a fluid. Moriwaki,
et al., (1977) investigated the disper-
sive slaking of sodium and calcium
saturated kaolinite, illite, and
montmorillonite. All the clays slaked
by dispersion when saturated with sodium
with the process proceeding faster with
sodium-kaolinite and sodium-illite.
Sodium-illite swelled slightly while
sodium-montmorillonite's dispersion was
preceded by extensive swelling. Sodium-
kaolinite underwent no visible swelling
while dispersing. For the calcium
saturated clays, illite dispersed much
more slowly while the rate of dispersion
increased for kaolinite an
-------
MISCELLANEOUS
Failure Mechanisms
There are a variety of situations
that may increase the permeability of clay
liners beside those three discussed in
detail here. The phenomena causing the
permeability increase may not be fully
understood but they are presented here
for their possible usefulness in future
research.
Miller, et al., (1975) found the
permeability of a soil to increase as
water flushed out an earlier application
of a surfactant.
Grubb, et al., (1973) found that
methyl alcohol increased the permeability
of a core previously injected with oil-
base wastes. He also noted the use of
solvents, organic acids, surfactants,
alcohols, and emulsion breakers for
permeability enhancement in deep well
injection operations.
Letey, et al., (1962) observed an
increase in the infiltration rate with
time for water-repellent soils. He felt
this was due to the progressive neutrali-
zation of the soils water-repellency as
the depth of infiltration increased.
In a later study, Letey, et al.,
(1975) found that permeability increased
with time if there is a substance in the
soil that would dissolve into the water
and decrease its surface tension.
Brant (1968) found an increase in
the water permeability after a soil was
treated with 4-t-butyl Catechol. He
postulated that the increase was due to
the soil matrix being rendered more
stable to water flow yielding a decrease
in the migration of soil particles.
Watson (1968) found surfactants acted
to stabilize soils against dispersion and
swelling, thereby preventing a decrease
in permeability values at certain
surfactant concentrations.
Wolstenholme (1977) stated that
solvents of low viscosity are "by their
very nature" leachable and able to
extract organic components from otherwise
dry waste. Lower viscosity would
significantly increase the rate of a
fluid's movement through a clay liner.
REFERENCES
1. Adams, N. K. 1941. The Physics
and Chemistry of Surfaces. Oxford
University Press, London.
2. Anonymous, 1977. Ground Water Manual:
A Water Resources Technical Publi-
cation. U.S. Dept. of Interior,
Bureau of Reclamation pp. 457-460.
3. Barrier, R. M. 1978. Zeolites and
Clay Minerals as Sorbents and
Molecular Sieves. Academic Press.
N.Y. p. 497.
4. Barshad, I. 1952. Factor effecting
the interlayer expansion of
vermiculite and montmorillonite with
organic substances, pp. 176-182.
SSSAP. 1952.
5. Baver, L. D., W. H. Gardner and
W. R. Gardner. 1972. Soil Physics.
John Wiley & Sons, Inc., N.Y.
6. Bradley, W. F. 1945. Molecular
associations between montmorillonite
and some poly-functional organic
liquids. J. Am. Chem. Soc. 67:
975-981.
7. Brant, G. H. 1968. Water movement
in hydrophobic soils, pp. 91-115.
In Proceedings of Water Repellent
Soils Symposium. Univ. of Calif.
at Riverside. May 6-10, 1968.
8. Carr, R. M. and H. Chih. 1971.
Complexes of halloysite with organic
compounds. Clay Minerals 9: 153-166.
9. Cedergren, H. R. 1967. Seepage,
Drainage and Flow Nets. John Wiley
& Sons, Inc., N.Y.
10. Conley, R. F. and M. K. Lloyd. 1971.
Adsorption studies on kaolinites.
II. Adsorption of Amines Clays and
Clay Minerals 19: 273-282.
11. Crouch, R. J. 1978. Variation in
structural stability of soil in a
tunnel eroding area. Chap. 34.
131
-------
pp. 267-274. In Modification of
Soil Structure. Edited by
W. W. Emerson, R. D. Bond,
-A. R. Dexter, John Wiley & Sons,
Inc.
21. Grim, R. E., W. H. Allaway and
F. L. Cuthbert. 1947. Different
clay minerals with some organic
cation. J. Am. Chem. Soc. 30:
137-142.
12. German, W. L. and D. A. Harding.
1969. The adsorption of aliphatic
alcohols by montmorillonite and
kaolinite. Clay Minerals 8: 213-227.
13. Gieseking, J. E. 1939. Mechanism
of cation exchange in the
Montmorillonite-Beidellite-Nontronite
type of clay minerals. Soil Sci. 47
pp. 1-14.
14. Glaeser, R. 1949. On the mechanism
of formation of montmorillonite-
acetone complexes. Clay Minerals
Bulletin No. 3 pp. 89-90.
15. Goring, C.A.I, and J. W. Hamaker.
1972. Organic Chemicals in the
Soil Environment. Vol. I & II.
Marcel Dekker, Inc., N.Y.
16. Greene-Kelly, R. 1955. Sorption
of aromatic organic compounds by
montmorillonite. Part I. Orienta-
tion Studies. Trans. Faraday Soc.
51. pp. 412-424.
17. Greenland, D. J. 1963. The
adsorption of polyvinyl alcohols by
montmorillonite. J. Colloid Sci:18
pp. 647-664.
18. Greenland, D. J. 1965. Interaction
between clays and organic compounds
in soil. I. Mechanisms of inter-
action between clays and defined
organic compounds. Soils and
Fertilizers 28: pp. 415-425.
19. Greenland, D. R., R. H. Laby,
J. P. Quirk. 1965. Adsorption of
amino acids and peptides by
montmorillonite and illite. Part II.
Physical adsorption. Trans, of the
Faraday Soc. ol pp. 2024-2035.
20. Grim, R. E. and F. L. Cuthbert.
1945. The bonding action of clays.
I. Clays in green molding sands.
Illinois State Geologic Survey
Report of Investigation No. 102.
22.
23.
24.
Grim, R. E. 1953. Clay Mineralogy
p. 296. Me Graw Hill Book Co. Inc.,
N.Y.
25.
26.
27.
28.
Grim, R. E.
pp. 434-436.
Inc., N.Y.
1968. Clay Mineralogy
Me Graw Hill Book Co.
Grubbs, D. M., C. D. Haynes,
T. H. Hughes and S. H. Stow. 1972.
Compatibility of subsurface
reservoirs with injected liquid
wastes. Nat. Resources Center, Univ.
of Alabama, Report No. 7?i...
Grubbs, D. M., C. D. Haynes and
G. P. Whittle. 1973. Permeability
restoration in underground disposal
reservoirs. Nat. Resources Center,
Univ. of Alabama. Report No. 773.
Gum, R. E., W. H. Allaway,
R. L. Cuthbert. 1947. Reaction of
different clay minerals with some
organic cations. J. Am. Chem. Soc.
30: pp. 137-142.
Haxo, Jr., H. E. 1976. Evaluation
of selected liners when exposed to
to hazardous wastes. Proc. of the
Hazardous Waste Research Symposium.
EPA-600-9-76-015.
Hoffmann, R. W. and G. W. Brindley.
1961. Adsorption of ethylene
gylcol and glycerol by montmori-
llonite. Am. Mineralogist 46:
450-452.
29. Hurst, R. E. 1970. Chemical find
growing use in oil fields. Chem
& Engineering News. 48. March 9,
p. 10.
30. Johansen, R. T., J. P. Powell and
H. N. Dunning. 1951. The use of
non-ionic detergent and citric acid
for improving cleanout procedures
of water-input wells in secondary
oil-recovery projects. Bureau of
Mines, Information Circular No. 7797.
132
-------
31. Jones, J. A. A. 1978. Soil pipe
networks: Distribution and discharge.
Cambria. Vol. 5 pp. 1-21.
32. Landau, H. C. and A. G. Altschaeffl.
1977. Conditions causing piping in
compacted clay. In Dispersive Clays,
Related Piping, and Erosion in
Geotechnical Projects. ASTM STP 623.
J. L. Sherard and R. S. Decker eds.,
pp. 240-259.
31. Letey, J., J. F. Osborn,
R. E. Pelishek. 1962. Measurement
of liquid solid contact angles in
soil and sand. Soil 3ci. 93. pp.144-
153.
34. Letey, J., J. F. Osborn, N. Valoras.
1975. Soil water repellency and the
use of non-ionic surfactants.
Contribution No. 154. Calif. Water
Resources Center.
35. Little, L. H. 1966. Infra-red
Spectra of Adsorbed Species.
Academic Press. London.
36. Macintosh, E. E., D. G. Lewis, and
D. J. Greenland. 1971.
Dodecyclairanonium - mica complexes -
I. Factors affecting the exchange
reactions. Clays and Clay Minerals
19: 209-218.
37. Mansur, C. I., R. I. Kauffman. 1956.
Underseepage, Mississippi river
levees. St. Louis District.
Proceedings Paper 864. Am. Soc. of
Civil Engineers.
38. Miller, W. W., N. Valoras, J. Letey.
1975. Movement of two non-ionic
surfactants in wettable and water
repellent soils. SSAP 39: 11-16.
39. Mitchell, J. K. 1975. Fundamentals
of Soil Behavior (Series in Soil
Engineering) John Wiley & Sons, Inc.,
N.Y.
40. Moriwaki, Y. and J. K. Mitchell.
1977. The role of dispersion in the
slaking of intact clay. Dispersive
Clays, Related Piping, and Erosion
in Geotechnical Projects. ASTM STP
623. J. L. Sherard and R. S. Decker
eds.
41. Nutting, P. E. 1943. The Action of
some aqueous solution in Clays of
the Montmorillonite Group. U.S.
Geological Survey Prof. Paper 197F
pp. 219-235.
42. Olejnik, S. A., M. Posner and
J. P. Quirk. 1970. The inter-
calation of polar organic compounds
unto kaolinite. Clay Minerals 8:
421-434.
43. Parfitt, R. L. and M. M. Mortland.
1968. Ketone adsorption on
montmorillonite. SSSAP 32: 335-363.
44. Pask, J. A. and B. Davis. 1945.
Thermal Analysis of Clays and Acid
Extraction of Alumina from Clays.
U.S. Bureau of Mines Technical
Paper 664 pp. 56-78.
45. Raza, S. H., L. E. Trieiber and
D. L. Archer. 1968. Wettability of
Reservoir Racks and its Evaluation.
Producers Monthly. Vol. 32 No. 4.
pp. 2-7. April 1968.
46 Schechter, R. S., J. L. Gidley. 1969.
The change in pore size distribution
from surface reactions in porous
media. ALChE Journal. Vol. 15 No.3
pp. 339-350.
47. Sherard, J. L. and R. S. Decker.
1977. Summary-evaluation of
symposium in dispersive clays.
Dispersive Clays, Related Piping,
and Erosion in Geotechnical Projects.
ASTM STP 623. J. L. Sherard and
R. S. Decker eds.
48. Sinex, Jr., H. E. 1970.
Dissolution of a Porous Matrix by
Slowly Reacting Flowing Acids.
Masters Thesis in Civil Engineering,
Univ. of Texas at Austin.
49. Theng, B. K. G., D. J. Greenland and
J. P. Quirk. 1967. Adsorption of
alkylammonium cations by
montmorillonite. Clay Minerals. (7).
1-7.
50. Theng, B. K. G. 1974. The
Chemistry of Clay Organic Reactions.
Halsted Press. John Wiley & Sons,
Inc., N.Y.
133
-------
51. USEPA, Federal Register, Monday
December 18, 1978. pp. 58946-
59028.
52. Watson, C. L. 1968. Hydraulic
conductivity of soil as influenced
by surfactants, pp. 163-169. In
Proceedings of Water Repellent
Soils Symposium. Univ. of Calif.
at Riverside. May 6-10, 1968.
53. Weiss, A. 1963. Organic
derivatives of Mica-type layer
silicates. Angew. Chem. 2 pp. 134-
143.
54. Wolstenholme, R. M. 1977. Disposal
of solvent waste, pp. 138-145.
In the Second Solvent Symposium
held by the University of
Manchester Institute of Science
and Technology.
55. Young, R. N. and B. P. Warkentin.
1975. Soil Properties and
Behavior. Geotechnical Engineer-
ing 5, Eisevier Scientific Publ.
Co., N.Y.
134
-------
ASSESSMENT OF LINER INSTALLATION PROCEDURES
David W. Shultz
Michael P. Miklas, Jr.
Southwest Research Institute
6220 Culebra Road
San Antonio, Texas 78284
ABSTRACT
Southwest Research Institute is presently conducting a study to identify current methods
and equipment used to (1) prepare supporting subgrade and (2) place liners at various im-
poundments in the United States. Subgrade preparation and liner placement activities have
been observed at fifteen sites to date. The sites selected have included landfills, waste-
water impoundments, and potable weter reservoirs. Information obtained during each site
visit included:
(1) Methods and equipment used to prepare the subgrade upon which the liner
is to be placed;
(2) Methods and equipment used to place the liner material;
(3) Special problems encountered and their solutions; and
(4) Important characteristics which must be considered during design and
construction of an impoundment facility.
Various aspects of subgrade preparation and liner placement are discussed herein. Photo-
graphs depicting construction and placement activities are presented.
INTRODUCTION
The use of surface impoundments and
landfills to store, treat and/or dispose of
unwanted materials has been and continues
to be common practice for industry and
municipal agencies since these types of
facilities have proven to be cost effective
solutions to treatment and disposal prob-
lems. Recent studies have shown that the
use of such facilities can result in sub-
surface migration of hazardous materials
into groundwater resources.1'2
The Resource Conservation and Recovery
Act of 1976 and proposed EPA regulations
will require positive control of subsurface
migration of contamination from many of
these facilities where hazardous materials
are stored.3 Future treatment and/or dis-
posal sites will likely have to be designed
to prevent groundwater contamination. It
is likely that impermeable liners with high
integrity will find increased usage in the
future.
The proper planning, design and con-
struction of surface impoundments and/or
landfills designed to contain hazardous
wastes involve numerous steps, including
the following: (1) defining facility
function and geometry; (2) selection of a
liner material which is compatible with
the material to be stored or treated; (3)
planning suitable subgrade preparation,
proper liner installation, adequate seep-
age monitoring and collection provisions;
and (4) developing appropriate post-
installation operation and maintenance
planning.
This study, which addresses subgrade
preparation and liner placement, has the
following project objectives:
135
-------
(1) To identify current subgrade
preparation procedures and
equipment used to build surface
impoundments and landfills;
(2) To identify methods and equip-
ment utilized to install various
liner materials; and
(3) To identify special problems
which should be considered dur-
ing the planning, construction,
and operation of such facilities.
The following generic types of liners
are intended to be included in this study:
(1) compacted native soils (clays); (2)
admixes (asphalt, concrete and soil cement);
(3) polymeric membranes (rubber and plastic
sheeting); (4) sprayed-on linings; (5) soil
sealants; and (6) chemical absorbants.
Construction sites where many of these
materials are being incorporated into the
facility design have been identified and
visited. Field notes from interviews, along
with photographs, are used to document ob-
servations at each visited facility.
In this paper, preliminary findings
relating to subgrade preparation and liner
installation are presented and discussed.
It has not yet been possible to visit sites
representative of each of the generic liner
types listed earlier. Photographs illus-
trating many of the points discussed are
included in the text.
APPROACH
First, a list of sites which could be
visited by a field crew was developed. The
development was accomplished by contacting
liner material manufacturers, fabricators,
installation contractors and national,
state, and local governmental agencies.
Manufacturers, fabricators, and installers
are most aware of ongoing and/or planned
jobs, particularly where bids have been re-
quested. The authors were directed to
specific installers who had more up-to-date
scheduling information for their particular
interests. Arrangements were made for site
visits with the owner of the facility, and
the agreement of the installer.
All pertinent information about each
site is kept on a master scheduling list
which is updated continually as project
schedules change. Field visits are sched-
uled and conducted concurrently with on-
going site identification. Every effort is
made to identify and select sites represen-
tative of the generic liner types included
in the project plan.
To date, fifteen sites have been
studied. The type of facility visited and
liner installed are as follows;
Liner Type
CPE
HCPE
CSPE
CSPE/PVC
Compacted Clay
Soil Sealant
CSPE
CPE
CSPE/Clay
PVC
EPDM
PVC
Soil Sealant
CSPE
PVC
Facility Type
Wastewater Storage
Evaporation
Evaporation
Evaporation
Evaporation
Landfill
Potable Water Storage
Evaporation
Landfill
Spill Containment
Wastewater Treatment
Landfill
Water Reservoir
Potable Water Reservoir
Landfill
SUBGRADE PREPARATION
Introduction
Subgrade is defined as the support
medium upon which a liner, either man-made
or natural materials, is placed. The sub-
grade is critical to the ability of the
liner to perform its intended function;
i.e., minimizing fluid seepage. The fol-
lowing sections discuss two important
aspects of subgrade preparation for a lined
impoundment, compaction and soil steriliza-
ti on.
Compaction
Adequate compaction of subgrade is
critical to the integrity of any lined im-
poundment, as the subgrade serves as the
liner support. Various pieces of standard
construction equipment are used when a
flexible liner is installed. Size require-
ments depend upon the magnitude of the job
and the degree of compaction required.
Figure 1 shows a small sheep's foot roller
being pulled by a bulldozer. This is suit-
able for use up steep slopes where the
total acreage to be compacted is small.
Figure 2 shows a vibratory roller and com-
pactor being used to smooth the floor of a
water reservoir impoundment. This size and
type of equipment is typical within the
136
-------
construction industry.
Most facility designs specify a sub-
grade compaction to a standard proctor den-
sity or modified proctor density. In the
event that native soils require the addi-
tion of moisture to achieve the designated
soil compaction densities, equipment must
be used to supply water to the soil. Fig-
ure 3 shows the use of a watering trailer
to apply water over a large area. These
watering vehicles are commonly found around
surface mining operations to control dust
and are readily adapted for use in applying
water to subgrade.
Many installers do not use rollers to
finish and smooth the surface of the sub-
grade. Oftentimes, bulldozers or other
earthmoving equipment are used as a compac-
tion device, producing a rough surface tex-
ture. The surface must then be smoothed
prior to placement of a flexible liner.
Figure 4 shows a surface scraper being used
to finish surface preparation prior to
placement. This type of equipment removes
much of the surface rubble and, depending
upon the soil type, can provide very ade-
quate subgrade for liner placement.
In the event that earthmoving equip-
ment used for excavation leaves a rough
subgrade, further work is required to pre-
pare the subgrade for liner placement.
Figure 5 shows the type of surface which
can be left by earthmoving equipment. This
surface texture is not considered desirable
for placement of flexible liners and should
be smoothed and further compacted. Often-
times, the subgrade will require a large
amount of rework prior to liner placement.
This may be due to the addition of soils
from off-site resulting in incompatibili-
ties in soil textures. Figure 6 shows a
disc used to mix two types of soils to pro-
duce uniformity prior to compaction. Discs
are also used to mix soil sealants such as
bentonite with the native soils prior to
compaction and subsequent sealing.
In many parts of the country, a common
problem with subgrades is the presence of
rocks. Rocks may cause problems because of
their size and/or because of jagged edges.
Most installers believe that if a liner is
placed over rocks, the probability of fail-
ure is increased. Figure 7 shows workmen
making a final inspection to remove any
rocks which might be present in a finished
subgrade. These men are working ahead of
the liner placement crew in an effort to
insure that a very smooth subgrade is ob-
tained.
Soil Sterilization for Vegetation Control
In many parts of the United States,
there are certain grasses and other plants
which are extremely tenacious and which can,
in the growing process, puncture a liner.
Figure 8 illustrates a 'salt grass1 which
is growing up through a flexible liner be-
ing installed in the western United States.
This area had been treated previously with
an herbicide; unfortunately, the herbicide
used was not effective against grasses com-
mon to the arid regions of the western
United States, but, rather, was designed
for use against grasses more common to the
humid eastern United States. The end re-
sult was that not all the seeds were killed,
the grass grew up through the liner and, as
shown in Figures 9 and 10, either individ-
ual patches were placed on each pin-prick
created by the grass, or in the case where
large areas of liner were affected simi-
larly, areas of 300 to 500 ft sq were en-
tirely replaced with new liner materials.
It is imperative that the proper regional
herbicide be used in a particular geographic
area. Generally, it is best to apply the
herbicide and then wait for a few days be-
fore placing the liner directly over the
area which was treated.
Figure 11 shows a typical herbicide
application at a liner installation. At
this particular site, a problem with a
'brush grass' occurred after application of
the herbicide. The liner was placed over
the treated area immediately, without wait-
ing for a few days to pass. The grass
punctured the liner and repairs were re-
quired. The installer believes that by
placing the liner immediately after appli-
cation, the growth process of the plant was
enhanced due to the warmth and moisture
which collected beneath the liner.
PLACEMENT
Introduction
The following sections discuss various
aspects of liner placement which usually
require consideration during facility plan-
ning and construction. Comments are based
upon field observations as well as discus-
sions with installers and contractors.
Discussions mainly involve flexible membrane
137
-------
liners due to the fact that the majority of
sites identified and visited to date have
installed this generic type.
Site Storage
Consideration must be given to proper
storage of liner materials after they are
delivered to the site. If a covered area
cannot be provided, many manufacturers
recommend that the liner material be cov-
ered with a light or reflective plastic to
help prevent possible degradation due to
sunlight and massive heat buildup. Figure
12 shows boxes of fabricated panels of
flexible liner material being stored at a
job site. Note that some of these have
light colored plastic covers. The impor-
tance of this protection during storage at
the site is a function of the type of liner
being stored, anticipated weather, length
of time the material will be stored on the
site and ambient temperatures. If extremes
of temperatures are anticipated, the liner
should be stored under a protective shelter.
Securing Liners
Various problems arise in the field
when attempting to secure liners either to
the subgrade or to penetrations or other
concrete structures which are found within
the confines of a surface impoundment or
landfill. Figures 13, 14 and 15 show a
method of attaching liners in their anchor
trenches, utilizing concrete. Figure 13
shows the liner coming up the side slope
and extending through the slit trench.
Concrete is then poured over the liner.
Figure 14 shows the concrete as it has been
poured to secure the liner. Figure 15
shows a finished portion of a fresh-water
reservoir in which concrete was used in the
anchor trenches.
Many anchoring trenches do not utilize
the concrete method. Rather, fill material
is placed over the liner which extends into
the trenches. The trenches are generally
of two varieties, the slit-type as shown
previously and a blade-cut type which has a
steep edge adjacent to the liner slope and
a gently sloping edge away from the liner
slope, almost like a truncated V. The
final result of the blade type of installa-
tion is similar to the slit trench in that
fill material is merely placed over the
liner after it has been placed into the
trench.
Figure 16 shows a concrete base to
which liner must be attached. Of interest
here, of course, is the differential rais-
ing of the concrete on one side of a crack
which has developed in this concrete im-
poundment, necessitating lining by flexible
material. Before this impoundment was
lined, filter-fabric was placed in double
layers over cracks such as these to ensure
that the liner placed over the sharp edges
of these cracks would not be damaged by
future movement. When attempting to in-
stall liner on concrete, it is necessary to
use batten strips of one form or another.
Figure 17 illustrates a method of attaching
a liner at the top of the concrete impound-
ment utilizing batten strips. Figures
shown later show similar methods of attach-
ing PVC batten strips to liner via expand-
ing bolts.
Bentonite
Bentonite, a hydrous aluminum silicate,
is a colloidal clay of the montmorillonite
group. It swells in water. The use of a
bentonite is common in areas where compati-
bility or soil tests show that the proper
application will lower premeability to the
desired level. The method of applying ben-
tonite varies. One method which has been
used is to place 100-lb bags on a grid pat-
tern covering the entire subgrade. A large
piece of equipment such as a disc is then
driven over the entire area. The disc rup-
tures the bags of bentonite and hopefully
spreads the material evenly. One problem
with this approach is that if the disc is
run only in one direction, the bentonite
may not be spread evenly. Figure 18 shows
the result of working the soil in only one
direction. The location of each bentonite
bag can be seen, and the uneven distribu-
tion of the bentonite is apparent as
lighter areas trending up and down the side
slope. If the 'disc in' method of soil
sealant application is to be used, it is
important that the equipment be pulled in
both directions to assure even distribution
of the soil sealant. Figure 19 shows a
specially designed rake which has been suc-
cessfully used to blend bentonite with soil.
This rake is especially effective when the
soil is sandy in nature and void of rocks
or lumps of clay. The rake is dragged be-
hind a tractor or bulldozer and will mix
bentonite down to approximately 3 inches
below grade.
138
-------
A common method of application uses a
standard soil spreader attached to the back
of a dump truck. The truck, loaded with
bentonite, is then driven over the impound-
ment surface. Bentonite can also be ap-
plied pneumatically when large areas re-
quire application. The bentonite is pneu-
matically forced through an applicator.
This method requires the use of more expen-
sive equipment and finds limited use.
Seaming
A common problem for field installa-
tion crews of flexible lining material is
steep side slopes. When walking on steep
side slopes the potential for injury of
workmen due to falls is greatly increased.
To allow field crew personnel to accomplish
field seaming and anchoring, special accom-
modations must be developed. Figure 20
shows a simple 2x4 (wrapped with a filter
fabric for cushioning) and ropes tied to it.
This board is secured around the top of the
berm, providing the field crew footing when
placing the batten strip shown on the right,
and when seaming the liner.
Poor quality seams result from the use
of the wrong adhesive systems or the place-
ment of liner and subsequent seaming occur-
ring during adverse field conditions. Ad-
verse field conditions include cold temper-
atures, excessive moisture and high winds.
Figure 21 shows a field crew attempting to
use a solvent adhesive in temperatures
which are too cold for the specific type of
adhesive system to work properly. The
seams which resulted from the activity
shown did not meet quality control inspec-
tion criteria and had to be redone. The
use of heat guns in this particular case
might have allowed the liner material to be
pre-heated hot enough to allow the adhesive
system to set. As a result of the time and
dollar losses, a different adhesive system
was later adopted and successfuly utilized
at this particular installation.
It is very important that a proper ad-
hesive system be used for a given climate
or set of weather conditions at a job site.
The installation contractor needs to be
aware of potential problems with certain
systems in regard to cold and excessive
moisture and should have a backup adhesive
system which can be used to provide ade-
quate seams when the desired system is not
working. When weather conditions get too
bad (i.e., temperatures <50°F, rain falling,
standing water, and/or winds in excess of
20mph), seaming should be halted.
The geometry of a particular impound-
ment can cause an excessive amount of liner
material to accumulate in corners. Figure
22 shows such accumulation. This material
must be cut out in the corner and a suit-
able seam made. Figure 23 also shows the
use of patching compounds on the concrete
panels prior to placement of the flexible
membrane liner. Most liner installers
recommend placement of panels such that
field seams run perpendicular to the toe of
the slope or up and down side slopes. Some
installers violate this general recommenda-
tion when installing materials which are
reinforced and can be field seamed easily.
The field seams reach complete strength
rapidly and panels may be placed such that
field seams parallel the toe of the con-
taining embankment. Figure 24 illustrates
the placement of panels where field seams
parallel to the toe result. This type of
placement may stress the field seam before
a complete cure of the adhesive is realized.
Inadequate adhesion due to the applied
stress can cause 'fishmouth' to occur and
eventually result in seam failure. It is
best to avoid this type of field seaming.
Figure 25 shows a large wrinkle at a
seam edge. This wrinkle is probably a re-
sult of improper spotting and stretching of
the panel. These kinds of wrinkles demand
additional corrective procedures, thus slow-
ing field seaming activities. To avoid
this type of problem, this panel should be
laid as smooth as possible prior to com-
mencement of seaming.
A flexible membrane liner panel placed
in a field may weigh from 2,000-5,000
pounds. It is important when moving this
panel that the field crew does not physi-
cally stretch the material, especially at
the edge to be seamed. Figure 26 shows the
use of plastic dowels which allow the crew
members a means to grab onto the liner
without stretching it. The liner will be
rolled around the dowel and the dowel will
serve as a handle. This simple technique
to avoid stretching and to ennance the
ability of the workmen to move the liner is
very effective and used by many installers.
Field seaming often requires the use
of heat guns or lighting which means that a
power source must be available. Remote
field installations may not have suitable
139
-------
power. Installers must therefore make pro-
visions to supply their own power.
Figure 27 shows a small portable gen-
erator which can be moved onto the liner
itself and can follow a field crew provid-
ing adequate electricity for seaming opera-
tions. Of course, multiple generators can
be used as required.
Liner panels can be as long as 200-300
feet. With large, heavy panels it is dif-
ficult to maneuver the panel material ade-
quately in order to minimize seams or
wrinkles. Figure 28 shows a simple way to
secure the liner as the field crew moves
down the panel removing wrinkles. The two
men shown standing on the liner are there
to keep the liner from sliding over the
subgrade while the crew removes the
wrinkles by pulling the line taut without
stretching the material.
Invariably, when seaming many thou-
sands of feet of liner material, wrinkles
will occur along the seams. This situation
is shown in Figure 29. Careful seaming
procedures will be needed to assure seam
integrity when wrinkles such as these are
present. Excessive wrinkles are an endemic
problem when subgrades are rough or the
geometry of the site causes unusually
shaped or various size panels to be used.
In Figure 30 one solution to the excessive
wrinkles is to pull them into one flap, cut
them out, lap seam and patch the entire
area.
Depending upon the size of an install-
ation, one aspect of impoundment construc-
tion can be on-going while liner installa-
tion is proceeding. Figure 31 shows a
large impoundment being filled with ef-
fluent while liner placement continues
elsewhere. The entire area shown in this
slide has been lined and soil cover applied.
Liner installation crews are at work else-
where on this site, but they are not shown
on this photo. In many instances, the im-
poundment facility is placed into service
prior to the final completion of liner
placement.
During field seaming, it is important
that a good foundation be provided upon
which to work a field seam. A hard, flat
surface (the best for rolling and seaming)
is normally not present in most installa-
tions. Therefore, a portable board device
can be used to provide a suitable surface
upon which seaming can be done. Figure 32
shows such a board. A rope is attached to
one end-, the board is approximately 12"
wide and 10' long and can be slid along be-
neath the field seam as the act of seaming
progresses. Most field seams require some
type of rolling or smoothing, especially if
an adhesive is applied. A board such as
this provides an excellent, homogeneous
surface for such activities. The pad shown
in Figure 33 serves as a kneeling pad for
the field crew and is designed to reduce
fatigue.
Oftentimes, the geometry of a site will
result in the occurrence of excess material
during the installation process. In Figure
34 the liner materials can be seen over-
lapping, resulting in an excess of material.
This excess was inevitable in this particu-
lar job as the circular top perimeter was
larger than the circular perimeter at the
toe. In order to properly excise the ex-
cess material and still provide an adequate
overlap seam, a vertical cnalk line down
the side slope was established (see Figure
35). The liner material was cut along this
line and the excess removed to be discarded
or used elsewhere.
Seal ing Around Penetrations
Penetrations through liner materials
require adequate sealing to prevent migra-
tion of fluid through the liner at the
point of penetration. Oftentimes, boots
are constructed which fit over a penetra-
tion and seal to the liner. Figure 36
shows a small boot that was placed around a
grounding cable which penetrated through
the liner. Such a boot, which was con-
structed in the field, can be sealed to the
liner material using an adhesive system
similar to that used for field seaming.
Oftentimes for larger structures these
types of boots may be fabricated in a fac-
tory and shipped to the site for installa-
tion.
Penetration of a liner by any designed
means should be avoided if at all possible.
If penetrations exist it is imperative that
an excellent seal be achieved around any
liner penetrations. Figure 37 shows an in-
let pipe penetrating through the side slope
of an impoundment. The black/dark liner
shown around the pipe is being placed prior
to the installation of a 'boot' which will
surround the entire pipe. This 'boot' is
shown in Figure 38. Note the use of a
140
-------
stainless steel strap mechanically sealing
the boot around the perimeter of the pipe
at the impoundment end. This boot, when
sealed to the liner, will prevent the seep-
age of fluid into the subgrade adjacent to
the pipe. It is advisable that the in-
staller familiarize himself with the number
and types of penetrations ahead of time in
order to develop adequate plans to seal the
liner around the penetrations to minimize
leaks.
Figures 39, 40 and 41 illustrate a
method for securing liner material to and
around a concrete penetration in a soon-to-
be lined impoundment. A small pad is
shown in Figure 39. This pad is secured
to the concrete base by the use of a recom-
mended mastic. Liner is then slit and
placed over the penetration and abutted
against the concrete structure. More mas-
tic is applied to ensure a good seal be-
tween the concrete and the lining material.
On top of this, as shown in Figure 41, are
placed steel batten strips through which
anchoring bolts will be emplaced. Though
this particular operation was viewed in the
field, it is probably not as reliable as
other methods of securing lining around
penetrations. Leak paths have been left
around the edges of the penetration and the
concrete itself may be a pathway for mate-
rials to migrate out of the lined impound-
ment when it is complete.
Wind Damage
When installing membrane liners in the
field, the installers must be aware of the
potential damage caused by heavy winds.
When winds exceed 20 miles/hour, the em-
placement of membrane lining materials be-
comes extremely tedious and the probability
of damage resulting from the wind's lifting
of the liners, much as a sail is filled by
the wind on the ocean, is enhanced. The
use of weights to hold lining materials
down is common within the industry. Gen-
erally, the weights are bags filled with
sand or, in the case shown in Figure 42,
actual rolls of liner material placed over
the installed layers. Figure 43 shows rows
of sand bags which have been laid adjacent
to the edge to be seamed in the field.
These sand bags serve two purposes; to
hold the liner down in heavy winds and to
hold the liner in place as the seaming
process goes on. In some instances, it is
impossible to stop liner materials from be-
ing lifted by the winds when the winds
become too strong. Figure 44 shows a reser-
voir being lined. The reservoir has a con-
crete rim to which the liner material was
attached. After the liner was attached to
the concrete rim, the liner was left over-
night during which a heavy wind arose,
lifted the liner, tugged it and eventually
ripped it away from the uppermost concrete
edge to which it was sealed. Figures 45,
46 and 47 illustrate the damage caused by
the wind when the sand bags emplaced on a
liner in the field proved to be insuffi-
ciently heavy to hold the liner in place
when a heavy wind arose overnight. Tearing
of the liner material is evident, as is a
stretching of the liner material. The dam-
age shown in these slides required consid-
erable retailoring of the liner material in
the field at great expense of both time and
money.
Occasionally, even when very large
panels have been laid and seamed together,
the wind will find a crevice beneath the
panel and will enter that crevice, producing
a bubbling-like effect which will then be
wafted by the wind and may eventually, if
allowed to continue, weaken the liner in
place. Figure 48 shows the occurrence of
such a wafting effect with a liner which is
partially covered by soil materials. Two
or three large bubbles of air are beneath
it, and are migrating and moving rapidly.
Continued movement of this type may weaken
field-seams and actually tear the liner
material.
Another means of anchoring liners is
shown in Figure 49. At this site, land-
fill materials were used to anchor the
liner once it was in place. Large piles of
sand are shown, weighting down the under-
lying liner material.
There is probably no way to adequately
insure that wind damage will not occur when
winds in excess of 30 miles/hour blow in an
area in which lining by membrane materials
is progressing. The best way to avoid any
wind damage is to first anchor materials as
well as is feasible in the field arrange-
ment. Any loose edges which are exposed in
the direction from which the wind normally
blows should be extremely well anchored.
Do not leave exposed edges on the side
slopes. The wind will have a good fetch
across the bottom of an impoundment and
will migrate up the side slope with a
swirling motion, thus being able to pick up
liner along an edge which is bared along
141
-------
the side slope. When planning field-lining
activities for a given day, attempt to
leave as few as possible of these bare
edges which the wind can pick up.
In the event that any liner damage
does occur due to wind action, repair is an
immediate necessity. Tears will have to be
repaired, stretched material will undoubt-
edly have to be replaced, and the efforts
of the placement team will be confined for
a period of time to redoing that portion of
the liner which was damaged.
In the case of some jobs which are con-
fined to concrete tanks and other struc-
tures, the swirling effect of the wind due
to the Venturi-like shape of the contain-
ment area may present a serious problem to
the eventual correct placement of liner
materials. This problem can be lessened or
avoided in some instances by using batten
strips such as those shown in Figure 50.
These strips are being placed over the
liner material to hold it in place because
of excessive swirling winds within the cir-
cular concrete structure which is being
lined. This batten arrangement was devised
in the field precisely to control the ef-
fects of the swirling wind prevalent at
this particular installation.
Soil Cover on Liners
In the case of liners such as PVC, it
is necessary to place a soil cover over the
liner to protect the liner from the bad ef-
fects of ultraviolet radiation on the
plasticizers and plastic materials within
the liner. Interestingly enough, liners
such as Hypalon, HPE and others have soil
covers placed on them not because of their
susceptibility to sunlight, but because the
soil cover acts as a stabilizing agent,
holding down the liner beneath. Figure 51
shows a wheeled vehicle which is pushing
soil cover over a liner. Figure 52 shows a
dump truck which is being driven over a
liner to deliver soil cover.
There is a question as to whether the
use of heavy equipment on liners will
eventually cause the integrity of the liner
beneath to be lessened. One installer uses
the equipment shown in Figures 53 and 54 to
transfer soil cover material to the top of
the liner without ever placing his equip-
ment on the liner itself. The controversy
over the use of heavy equipment on liners
continues, but there are some methods to
alleviate potential liner damage from the
over movement of heavy vehicles. For
example, the contractor can vary the
ingress/egress locations for the heavy
equipment onto the liner, or he can limit
the absolute weight of the equipment which
is allowed to work on the liner. A third
way is to use tracked vehicles or bulbous
tires with relatively low weight ratios
when working over the liner materials.
Figure 55 shows a liner which has be-
come degraded and cracked and pulled away
from the subgrade as a result of the soil
cover being washed away from the liner and
subsequent weakening from the ultraviolet
radiation of the sun. This frame illus-
trates what the use of soil covers will
avoid. It is interesting that this parti-
cular site originally had a soil cover, but
the soil was washed away at this point by
the activity of wind generated waves moving
across the surface of the impoundment, lap-
ping on the edge, carrying away the soil
cover. In some instances, extreme measures
such as rip-rapping or wave-deflectors may
be required in order to ensure that the
underlying liner is protected by a continu-
ally existing overlying soil cover.
Quality Assurance
One of the problems at a given site is
the maintenance of high quality in the
workmanship which is progressing at the
site. Figure 56 shows a machine which is
spreading water over the dusty roads over
which heavy equipment moves near the site.
This may seem a trivial matter, but the ab-
sence or lessening of dust is actually a
very important condition to strive for at a
site. If at all possible, dust should be
kept nonexistent. The presence of dust
will inhibit the sealing capabilities of
various types of adhesive materials. When
it is not possible to absent an area of
dust, extreme care must be taken to clean
seam edges before the actual seaming
process occurs.
A problem which can occur in some
types of materials, actually damaging the
materials themselves, is an effect known as
"blocking." In blocking, the materials
which are packed for transport will seal to
themselves in the presence of excessive
heat or with age. The end result is that
when the liner is unfurled for use in the
field, portions of the liner will stick
142
-------
to itself, and may cause a release of ad-
hesion with the underlying reinforcement
material when it is present. Figure 57
shows a picture of such a 'blocking' condi-
tion on a piece of Hypalon ready to be in-
stalled in the field. This type of block-
ing will generally not occur singly, but
rather will occur along a seam in which the
conditions were ripe for its formation.
Figure 58 shows the condition as it would
appear when repaired in the field. The
three patches in the upper center portion
of the figure show the blocking condition
which was repaired with an overlying seal-
ing patch of Hypalon.
The integrity of seams is also an im-
portant factor to be observed in the field.
Figure 59 shows a seam which contains many
wrinkles; additionally, this seam illus-
trates what happens when solvent evaporates
before actual adhesion occurs. The bubbles
seen in the central portion of the figure
represent points where solvent evaporated
before the overlying liner material ad-
hered to the underlying material. It is
probable that the seam in this area has
less integrity than in an area which is
free of bubbles. If possible, this 'bub-
bling' condition should be avoided by seal-
ing as soon after the application of ad-
hesive as is advised by the manufacturer.
If the 'bubbling' condition develops ex-
cessivly, the seam will either have to be
redone or patched. The manufacturer's
recommendations for sealing should also be
reviewed in an effort to find a method to
stop the undesired bubbles from developing.
when edges are sealed, many times pathways
for leakage develop at wrinkles even though
the adhesive is well applied. When wrinkles
which are potential leakage pathways de-
velop, the wise contractor will ensure that
there will be no leakage by placing a patch
over the potentially offending area. Fig-
ure 60 shows such a patch as does Figure
61. Both these patches are of questionable
quality and may need further evaluation and
eventual replacement. Figure 62 shows the
need for multiple patching which occurs when
numerous wrinkles are present in a layer to
be seamed to another layer which has numer-
ous wrinkles. Each of the wrinkles repre-
sents a leak path which must then be sealed
by the addition of the overlying patch.
This particular figure shows three patches,
one on top of the other. It is not recom-
mended that this type of patching be done
with wrinkles of this sort. Rather, all
attempts should be made to smooth and re-
move the wrinkles; if this is not possible,
it is best to consolidate the wrinkles into
one large wrinkle which can then either be
excised and sealed, or folded over, glued
and sealed with one single patch on top.
Each site will generally have a
Quality Control inspector. It is recom-
mended that Quality Control inspectors
operate in a pre-formulated plan, rather
than a haphazard manner. This means that a
well-defined plan for inspection should be
developed prior to liner installation.
Figure 63 shows the markings left by
Quality Control inspectors as they went
about inspecting the field seams and fac-
tory seams in one installation. This type
of control should generate the best possible
integrity.
SPECIAL DESIGN CONSIDERATIONS
Future Expansion
When a flexible membrane liner is in-
cluded as part of the facility, the design
should incorporate considerations on how
the facility will be expanded in the future.
Haxo ejt aj_ have shown that aging of flex-
ible membrane materials often results in
changes in certain physical properties.
This can be an important factor when con-
sidering adding on new liner material to
existing liner material. Figure 64 shows
a worker uncovering the edge of an existing
liner panel which has been buried for sev-
eral years. In this particular case, a new
liner will be seamed or attached to the
edge shown in Figure 64. Since this mate-
rial has been buried for several years, it
will be difficult at varying locations to
get a good adhesion of the old and new
liner materials. This situation resulted
because proper planning for expansion had
not occurred.
Figure 65 shows the incorporation of a
method to plan for the expansion of a liner
in the future. The figure shows a flap on
top of the existing liner; this flap will
seal the plastic strip shown in the figure
to the existing liner. In the event of
future expansion, the flap will be exposed,
the polyethylene plastic removed and the
new liner material seamed to the bottom of
this flap. This assures a relatively vir-
gin surface for seaming the new liner mate-
rial to the old. Figure 66 depicts the
flap which has been sealed into place.
143
-------
Liner Selection Based on Use
The costs of liner materials often
play a very important role in design. Dif-
ferent polymeric liners have different pur-
chase and installation costs. For large
impoundments, some design engineers are in-
corporating the use of two different liner
materials to help decrease total facility
costs. Figure 67 shows such a design. In
this particular instance, the side slopes
are a reinforced liner material, 36 mil
thickness; the bottom of the impoundment is
a 30 mil material, non-reinforced. The
side slopes have been lined with reinforced
material since the reinforcing fiber lends
greater strength to the material and facil-
itates installation during hot weather.
These two materials are both chlorinated
polyethylene and, therefore, the seaming
techniques at the toe did not require any
unusual adhesive system or bonding tech-
nique. The cost savings in this design re-
sult from the use of a less expensive,
thinner, non-reinforced material on the
floor of the impoundment.
Figure 68 illustrates the use of a
combination of two different generic liner
materials seamed together at the toe of the
impoundment. The material on the bottom is
a PVC non-reinforced material; the material
above is a reinforced Hypalon. This parti-
cular design was utilized to save money.
These two materials are not directly com-
patible to field seaming and a marriage
material was utilized to join them. Figure
68 shows a factory seam. The two materials
shown in Figure 68 vary greatly in in-
stalled cost. The PVC is a much cheaper
material than the Hypalon; the PVC was used
on the floor of the impoundment because the
total area there accounted for the majority
of the square footage of the facility. The
use of the reinforced Hypalon on the side
slopes was decided upon because this mate-
rial is resistant to sun induced aging. In
addition, the reinforced Hypalon provides
greater strength against the damaging ef-
fects of wave action.
Monitoring
Monitoring wells provide a means of
determining if a liner is functioning ade-
quately before any leakage from the liner
can leave the immediate vicinity of an im-
poundment. Figure 69 shows a monitoring
well casing located adjacent to an existing
landfill. This monitoring well will serve
as a source of groundwater samples which
will be taken periodically to detect
changes in groundwater quality. Monitoring
wells are best situated hydraulically down
stream from the landfill site. It is im-
portant that casings like this be protected
against vehicle damage or vandalism while
remaining accessible to sampling crews.
Monitoring systems are an important part of
facility design and will, undoubtedly, be
incorporated more and more in the future.
Surface Covers
One method to prevent the production
of leachate at an existing landfill is to
install a flexible liner, much like an
'umbrella,' over the completed fill to pre-
vent the inflow of rainwater or surface
runoff. Figure 70 shows a completed land-
fill cover in the final stages of the ap-
plication of the earth cover.
Gas Venting
One problem that can result from the
use of a relatively impermeable liner is
the containment of gases below the in-
stalled liner. If the liner is placed over
material known to cause gas production,
such as organics, the potential collection
of gas is increased. When the potential
for gas buildup is present, some provision
must be made to relieve the ensuing gas
pressure. Figure 71 shows an overall view
of a facility where a liner is being placed
for a landfill. Note the white PVC pipe
stands projecting along the crest of the
cover. These are gas vents and will serve
to relieve pressure which might develop
underneath the liner. Such gas venting is
a very necessary control aspect at landfill
installations.
Control of Surface Runoff
Another aspect in impoundment design
is the prevention of surface runoff from
encroaching upon the impoundment perimeter.
This is important for two reasons: (1)
such runoff may tend to violate the in
tegrity of the berms or erode a soil cover
which might be placed on the liner, and (2)
one should minimize the amount of water
contained in the impoundment and therefore
increase the useful life of the facility.
Figure 72 shows the use of a diversion dam
located in the upper left hand corner.
This dam will prevent surface runoff from
flowing down into the impoundment.
144
-------
Interestingly, some facilities with high
evaporation rates actually channel runoff
into their impoundments via surface routes
in order to control the erosive effects.
This is an important design consideration
regardless of the magnitude of the facility.
Dual Liners
When added confidence is required,
particularly for a facility where very
hazardous materials will be stored, it is
desirable to incorporate the use of dual
liners. These may be a combination of
soil, flexible liners, or specially treated
soils. Figure 73 shows a facility incor-
porating two separate liners to prevent
fluid movement into the subsoil. The upper
layer is a soil cover over a flexible mem-
brane liner; the third layer from the top
is another compacted soil underlined with
Fuller's earth. By incorporating two
barriers against fluid loss, the owner of
this facility increases the confidence that
fluid will be contained.
Special Subgrades
A liner is often placed over a dam-
aged or extremely irregular subgrade, which
may be concrete, aggregated materials or
even bedrock. In such instances, the de-
sign and installation must take into con-
sideration the condition of the subgrade;
any cracks that might be present, shifts in
the surface elevation, or irregular, jagged
edges must be ameliorated. Placement of a
liner material over such differential set-
tling or stress induced cracks or extremely
irregular subcrades may cause eventual liner
failure. Figure 74 depicts the placement
of filter fabric over a 4-inch deep crack
caused by differential settling beneath an
existing concrete containment wall. The
fabric material will protect the liner from
tearing by preventing the liner from being
"embossed" onto the concrete surface by the
overlying weight associated with the fluid
head in the impoundment. Note that on the
right, fabric has been placed over the en-
tire concrete surface, while on the bottom
sand has been placed. The piece of fabric
that the two workmen are placing will be an
additional layer of filter fabric and will
be covered again by a second layer. This
particular location had an extreme crack,
thus warranting the use of two layers of
filter fabric.
Another site was visited where the
irregular condition of the bedrock subgrade
demanded that a cushioning layer of filter
fabric be laid beneath the liner. Of
course, the use of filter fabric adds addi-
tional cost of materials and installation
time to a given job, but should increase
the integrity of the given installation.
145
-------
SUMMARY
REFERENCES
As a result of project activities,
lined surface impoundment and landfill con-
struction sites have been visited in order
to observe and assess subgrade preparation
and liner placement procedures. During
these visits, emphasis was placed on iden-
tifying construction and installation meth-
ods which provided acceptable solutions to
the various problems encountered during the
field activities. Photographs have been
presented in this paper to show various as-
pects of facility construction and install-
ation. Important items which should be
considered during construction and install-
ation were discussed. There exists in the
liner industry literature, recommended
acceptable procedures for installation.
Often, the recommended procedures have to
be modified to accommodate special problems
or site characteristics. Accommodation
which insures reliable construction and in-
stallation methods requires an experienced,
knowledgeable installation crew to adapt to
the various field situations. There are
also unacceptable practices which must be
avoided. This paper has attempted to pre-
sent some of the problem areas, identifying
both satisfactory and unsatisfactory solu-
tions.
At the present, the final responsi-
bility for any installation performing its
intended function rests with the owner.
Therefore, the owner must take an active
part in the planning, installation and op-
eration of a lined surface impoundment or
landfill. It is the authors' opinion that
a series of nationally accepted guidelines
are needed, defining methods and procedures
for the installation of various materials.
Included in these guidelines should be a
well-defined, meangingful, quality assur-
ance program which can be employed by a
prospective owner to insure field installa-
tion procedures will provide desirable
results.
ACKNOWLEDGEMENTS
This study is being supported by the
Solid and Hazardous Waste Research Division,
Municipal Environmental Research Laboratory,
U.S. Environmental Protection Agency,
Cincinnati, Ohio (Grant R806645010) and
Southwest Research Institute, San Antonio,
Texas. The authors want to thank Robert E.
Landreth, Project Officer, for his guidance
and support.
Miller, D., Braids, 0., and Walker, W.
1977. The Prevalence of Subsurface
Migration of Hazardous Chemical Sub-
stances at Selected Industrial Waste
Land Disposal Sites. EPA/SW-634,
Environmental Protection Agency,
Washington, D.C. 530 pp.
Miller, D., Braids, 0., and Walker, W.
1977. The Prevalence of Subsurface
Migration of Hazardous Chemical Sub-
stances at Selected Industrial Waste
Land Disposal Sites. EPA/530/SW-634,
Environmental Protection Agency,
Washington, D.C. 529 pp.
U.S. Environmental Protection Agency.
1978. Hazardous Waste Guidelines and
Regulations. Federal Register, Vol.
43, No. 243.
146
-------
Figure 1. Sheeps Foot Roller used for
subgrade compaction.
Figure 2. Vibratory roller used for
subgrade compaction.
Figure 4. Surface scraper used for
final smoothing.
Figure 5. Undesirable subgrade
surface texture.
Figure 3. Large watering vehicle used for
dust control.
Figure 6. Disc suitable for soil blending.
147
-------
Figure 7. Workmen removing small rocks
during subgrade inspection.
Figure 10. Liner replacement due to
"salt grass" damage.
Figure 8. "Salt grass" growing through a
flexible membrane liner.
Figure 11. Application of an herbicide to
the subgrade for soil sterilization.
Figure 9. Liner repair required due to
"salt grass" damage.
Figure 12. Site storage of liner material
showing protective plastic covering.
148
-------
Figure 13. Initial placement of a liner
into an anchor trench.
Figure 14. Anchor trench filled
with concrete.
Figure 16. Differential settlement of a
concrete subgrade.
Figure 17. Stainless steel batten strips
securing a liner to concrete.
Figure 15. Finished anchor trench -
concrete.
Figure 18. Bentonite applied to a soil
where spreading was uneven.
149
-------
Figure 19. An iron "rake" used to blend
bentonite and native soil.
Figure 20. Support board for working on
steep incline.
Figure 22. Excessive accumulation of
liner material.
Figure 23. Patching compounds applied to
concrete panel joints.
Figure 21. Unsuccessful seaming with a
solvent adhesive in cold temperatures.
150
-------
Figure 24. Field seams parallel to
side slope toe.
Figure 26. Plastic dowels being used
to pull liner panels.
Figure 27. Field generator.
Figure 25. Large wrinkle at edge of
liner panel.
Figure 28. Removing wrinkles from panel
151
-------
Figure 29. This seam wrinkle will
require removal.
Figure 30. Repair of seam wrinkle.
Figure 32. Seaming board.
Figure 33. Knee pad used by field crews.
Figure 31. An impoundment being utilized
while liner placement continues.
Figure 34. Material overlap.
152
-------
Figure 35. Use of a chalk line to mark a
liner for removal of excess material.
Figure 36. Field constructed boot for a
liner penetration.
Figure 38. Completed boot installation.
Figure 39. Liner pad around a concrete
penetration.
Figure 37. Pipe penetration being prepared
for installation of a 'boot.'
Figure 40. Pad prior to locating steel
batten strips.
153
-------
Figure 41. Placement of steel battens
around concrete penetration.
Figure 44. Wind damage.
Figure 42. Liner panels used to prevent
wind damage.
Figure 45. Wind damage.
Figure 43. Use of sandbags along a seam.
Figure 46. Wind damage.
154
-------
Figure 47. Stretching of material by wind.
Figure 50. Batten strip to secure liner to
concrete subgrade.
Figure 48. Wafting effect on a liner not Figure bl. Rubber wheeled vehicle used for
secured by a soil cover. placing soil cover.
Figure 49. Use of cover material in place Figure 52. Large dump truck placing
of sandbags. soil cover.
155
-------
Figure 56. Dust control on haul roads.
Figure 53. Equipment to place soil cover.
Figure 54. Equipment to place soil cover.
Figure 57. 'Blocking.'
Figure 55. Results of sun aging and
degradation.
Figure 58. Repair of 'Blocking.'
156
-------
Figure 59. Blistering from solvent
evaporation.
Figure 60. Patch over field seam.
Figure 62. Multiple patching should be
avoided.
Figure 63. Planned quality control
inspection.
Figure 61. Patch over field seam.
Figure 64. Exposing edge of liner for
expansion.
157
-------
Figure 65. Flap to be used for future
expansion of liner.
Figure 66. Flap sealed into place
(shown on left).
Figure 68. Two materials 'married'
together for cost effective design.
Figure 69. Monitoring well casing.
Figure 67. Shows the use of two liner
materials.
Figure 70. Completed landfill cover.
158
-------
Figure 71. Surface cover for a landfill
showing gas vent pipes.
Figure 72. Surface runoff diversion dam
shown in upper left hand corner.
.
.•*§ ^^m - <£ "? i v »*»i c% *• * * * ^
Figure 74. Placement of filter fabric over
differential settling crack in concrete.
Figure 73. Shows use of a membrane
and soil liner.
159
-------
INTERACTION OF SELECTED LINING MATERIALS
WITH VARIOUS HAZARDOUS WASTES - II
H. E. Haxo, Jr.
Matrecon, Inc.
Oakland, California
ABSTRACT
Results of the exposure under several conditions of selected lining materials to
typical hazardous wastes for periods up to three years are presented. These exposure
tests include:
1. One-sided exposure of a square foot of each of 12 lining materials under one
foot of six different wastes.
2. Immersion testing of 12 types of polymeric membrane lining materials in water
and in nine wastes.
3. Tests of nine different thermoplastic materials fabricated into bags and filled
with wastes to assess their compatibility and the permeability of the membranes.
4. Outdoor exposure testing of specimens of membrane materials exposed on racks
and in 12 tubs containing various wastes.
INTRODUCTION AND OBJECTIVES
In view of the nation's great depen-
dence upon groundwater as a source of pot-
able water, the need to protect the ground-
water from pollution by wastes placed on
land for storage or disposal is quite
obvious.
Placing a highly impermeable man-made
layer or liner below the disposal site
offers a means of intercepting the flow of
pollutant from the wastes and thus protect-
ing the groundwater.
This study was undertaken in 1975 to
determine the state-of-the-art of liner
technology and to assess experimentally a
broad range of promising lining materials
which were available at that time. The
principal objectives of the study are:
1. To determine the effects of ex-
posing a selected group of lining materials
to various hazardous wastes over an ex-
tended period of time.
2. To determine the durability of and
the cost effectiveness of utilizing syn-
thetic membranes, various admix materials,
and native soils as liners for hazardous
wastes storage and disposal sites.
3. To estimate the effective lives of
12 lining materials exposed to six types of
industrial nonradioactive hazardous waste
streams under conditions which simulate
those encountered in holding ponds, lagoons,
and landfills.
In this study, the lining materials
that are being studied are a native soil,
two treated bentonite clays, soil cement,
hydraulic asphaltic concrete, an asphaltic
membrane, and a wide range of commercial or
developmental flexible 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
160
-------
polyolefin, and a thermoplastic polyester.
The nine hazardous wastes selected for ex-
posure testing included strong acids,
strong bases, oil refinery tank bottom
wastes, lead wastes from gasoline, satu-
rated and unsaturated hydrocarbon wastes,
and a pesticide waste.
Specimens of the lining materials are
exposed to wastes under a variety of cir-
cumstances which simulate conditions that
might be encountered in actual service.
These tests consist of one-sided exposure,
two-sided exposure, outdoor exposure, and
composite exposure to weather and wastes.
Results have been reported in two pre-
vious symposia: At Tucson, Arizona1, in
1976 and at San Antonio, Texas, in 19782.
Additional details are given in an interim
report on the project .
In this paper we present the current
results of exposure testing of various
lining materials, particularly the poly-
meric membrane materials. Data on the
effects to liners of exposure to wastes of
up to three years duration are included.
EXPERIMENTAL APPROACH AND METHODOLOGY
To meet the principal objectives of
this project, we selected a range of prom-
ising lining materials which have low perm-
eability and exposed them to a series of
typical hazardous wastes under different
conditions which we felt would be aggres-
sive to lining materials. The effects of
the exposure were measured as a function
of time by such means as weight change,
change in physical properties, seepage
through the primary liner specimens and
permeation through membranes.
Details of our approach which is re-
viewed briefly below are given in Refer-
ences 1, 2, and 3.
The lining materials included both
flexible polymeric membranes and various
admixed types, e.g. soils, soil cements,
asphaltic concrete, and membranes. All
were selected on the basis of their having
low permeability to water and having a
history of use in the impoundment of water
and various industrial wastes and chemi-
cals.
Polymeric membranes based upon eight
different polymers were selected for the
exposure testing. The polymers include
butyl rubber, chlorinated polyethylene,
chlorosulfonated polyethylene, elasticized
polyolefin, ethylene propylene rubber,
neoprene, polyester elastomer, and poly-
vinyl chloride. All of the membrane lining
materials were commercial products except
the polyester elastomer which was included
because of its reportedly high resistance
to oily wastes. The admix materials in-
cluded compacted native fine grain soil,
hydraulic asphalt concrete, asphalt emul-
sion membrane, modified bentonite, and
soil cement.
Nine hazardous wastes were selected
from the petroleum, chemical, and pesti-
cide industries for use in this test pro-
gram. They include two strongly acidic
wastes, two strongly alkaline wastes, a
lead waste which was a blend of three lead
wastes, three oily wastes, and a pesticide
waste. These examples were felt to be
typical of the range of wastes and all
except the pesticide waste contain constit-
uents which were felt to be aggressive
toward lining materials. The properties of
these wastes are given in Tables 1 and 2,
taken from Reference 2. In addition to
these wastes, several additional test
fluids were incorporated in the test pro-
gram for assessing membrane liners. The
test fluids included deionized water, a 5%
aqueous solution of sodium chloride, a
highly ionic hazardous waste containing
minor amounts of organic chemicals and a
saturated solution of tributyl phosphate
in deionized water. The latter solution
was kept saturated by means of a reservoir
of tributyl phosphate.
The principal exposure test used in
this project consisted of exposing one
square foot liner specimens in cells, such
as shown in Figure 1, under one foot of
the wastes.
•Top Cover
Epoxy
Coaled
Waste Column
*—11 Gauge Steel
10" x 15 x 12" High
«/ Welded
Z " Flange
Outlet tube with
Epoxy-coattd
Diaphragm
161
-------
TABLE 1. WASTES IN EXPOSURE TESTS
Phases
Type of Waste
Acidic
Alkaline
Lead
Oily
Pesticide
Name
"HFL"a
"HNO-, HF, HOAC"
"Slopwater"a
"Spent caustic"^
"Low lead gas washing" )
) Blend
"Gasoline washwater" )
"Aromatic oil"
"Oil Pond 104"b
"Weed oil"a
"Weed killer"b
Organic
Phase
I
0
0
0
0
10.4
1.5?
100
89.0
20.6
0
Water
Phase
11
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 immersion tests only.
In both primary exposure and immersion tests.
TABLE 2 . WASTES IN EXPOSURE TESTS
pH, Solids, and Lead
Type of waste Name
Acidic
Alkaline
Lead
Oil
Pesticide
"HFL"
"HNO , HF, HOAC"
"Slopwater"
"Spent caustic"
"Low lead gas washing"
"Gasoline washwater"
"Aromatic oil"
"Oil Pond 104"
"Weed oil"
"Weed killer"
pH
Solids, %
Water phase Total
4.
1.
12.
11.
7.
7.
7.
2.
a
5
0
3
2
9
-
-
5
7
2
0
22
22
1
0
ca.
1
0
.48
.77
.43
.07
.52
.32
-
36
.81
.78
Volatile
0.
0.
5.
1.
0.
0.
ca.
1.
0.
9
12
09
61
53
17
-
31
00
46
Lead,
ppm
_
-
-
-
34
11
-
-
-
-
162
-------
This schematic drawing illustrates the cell
designed for the admix specimens. Such
cells were constructed with spacers between
the upper part of the cell, which contains
the test fluid, and the base in which is
placed a coarse silica sand. Membrane
specimens were tested in the same type of
cell, except that a spacer is not needed.
The original program called for 12
lining materials to be exposed to six
wastes; however, it was soon apparent that
many of the combinations of liners and
wastes were incompatible and liner failure
would occur in relatively short periods of
time. A preliminary screening test was
performed on various combinations of liners
and wastes and specific combinations were
then selected for the long term tests. The
selected combinations are shown in Table 3.
Two cells of each combination were set up
to allow for two exposure periods. The
first set of liner specimens were recovered
and tested after approximately one year of
exposure. These results were reported at
the San Antonio Symposium2. The second set
of specimens are being removed and test
results obtained to date are reported in
this paper.
In order to extend the number of
wastes and liners, the tests in the pri-
mary cells were supplemented by the immer-
sion testing of 12 liners in nine wastes.
Specimens of membrane liners were placed in
the additional test fluids given above.
In the immersion test procedure, small
slabs of the lining materials were hung in
the test fluids that were in the primary
cells. Not only did this allow an expanded
exposure test, but it also allowed exposure
of two sides of the liner to the waste.
Such a procedure accelerates the effects of
exposure and allows the specimens hung in
the wastes which stratify to show the ef-
fect of stratification of the waste. This
condition is in contrast to that in the
primary cell in which the liner is at the
bottom of the cell and only one side is
exposed to the waste.
Inasmuch as many of the liners for
ponds and lagoons which are used in the
storage or disposal of hazardous wastes are
exposed to the weather, outdoor exposure
testing of some of the membrane liners was
performed. Two types of tests were in-
cluded. In the first samples of the liner
materials were exposed in an undeformed
condition on a rack and the effects of
weather were observed by removing specimens
at different times and testing and, in the
second, the liners are simultaneously ex-
posed to the weather and to the wastes.
This was accomplished through the use of
small tubs which were lined with the mater-
ial being tested and the tubs filled with
waste. This type of test simulates open-
service and also allows the liner materials
to be draped over sharp corners which simu-
lates some conditions encountered in
service.
As most of the wastes contain water, a
separate study was made of the effect of
water upon polymeric membrane lining ma-
terials at room temperature and at 70°C.
For this test we followed ASTM D751, in
which the water absorption is observed as
a function of time.
A test method developed in our project
on liners for landfills was adapted for
liners for hazardous wastes. In this test
bags of heat-sealable membranes were fab-
ricated and filled with waste or other test
fluids such as salt water. The bags were
sealed and immersed in deionized water.
The permeabilities of the membrane to water
and to pollutants are determined by observ-
ing respectively the change in weight of
the bag and the measurement of pH and elec-
trical conductivity of the deionized water.
Due to osmosis water should enter the bag
and ions and dissolved constituents should
leave the bag.
THE PRIMARY LINER SPECIMENS
Polymeric Membrane Liners
All of the flexible polymeric mem-
branes survived the exposure testing, ex-
cept the polyester elastomer which failed
on exposure to the nitric acid waste. This
material completely lost its elongation and
cracked in the cell at the edges of the
liner. The nitric acid waste caused sev-
eral of the cells to leak due to corrosion
of the steel cell through pinholes in the
epoxy coating at welds between the cell
walls and the flanges.
In the case of the admix materials,
there was seepage through the soil, soil
cement, asphalt concrete, and bentonite
163
-------
TABLE 3. MATERIALS TESTED AS LINERS IN THE PRIMARY EXPOSURE CELLS CONTAINING VARIOUS WASTES
Material
Flexible polymeric membranes :
Liner Type of Extractables,
No . compound^ %
Butyl rubber (Butyl) 57RC VZ 6.4
Chlorinated polyethylene (CPE) 77 TP 9.1
Chlorosulfonated polyethylene (CSPE) 6RC TP 3.8
Elasticized polyolefin (ELPO) 36 TP 5.5
Ethylene propylene rubber (EPDM) 26 VZ 18.2
Chloroprene rubber (Neoprene) 43 VZ 13.9
Polyester elastomer (Polyester) 75 TP 2.7
Poly vinyl chloride (PVC) 59 TP 35.9
Admixes :
Compacted fine grain native soil
Soil cement
Treated bentonite with sand:
A
B
Hydraulic asphalt concrete
Emulsified asphalt membrane
Pesticide
T
T
T
T
T
T
T
T
T
T
T
T
T
T
HN03
T
T
T
T
T
d
T
T
d
d
d
d
T
d
Waste
Spent
caustic
T
T
T
T
T
T
T
T
T
T
d
d
T
T
Lead
T
T
T
T
T
T
T
T
T
T
I
T
T
Oil
104
d
T
T
T
T
T
T
T
T
T
eT
d
d
Aromatic
oil
d
d
T
T
T
d
T
T
T
T
eT
d
d
Identification number assigned by Matrecon.
^Type of compound, i.e. vulcanized (VZ) or thermoplastic (TP).
R = Fabric reinforced.
Combination of liner and waste was eliminated in the screening test.
Spacers not available.
-------
liners. The seepage through the soil did
not contain the waste which had been placed
on the soil. It was the fluid from the
pores within the soil itself. As the soil
contained considerable salt when it was
placed (it had been dredged from San Fran-
cisco Bay), the fluid which seeped also had
a high salt content.
The hydraulic asphalt concrete failed
below the nitric acid waste, principally
because of the incorrect choice of aggre-
gate . The acid consumed much aggregate and
hardened the asphalt considerably eventu-
ally resulting in the concrete leaking.
Another asphalt concrete liner which was
placed beneath the lead waste failed. This
waste contained considerable low molecular
weight oily material which the asphalt
absorbed and became "mushy".
Mostof the cells containing pesticide,
nitric acid waste, spent caustic waste and
lead waste have been dismantled and the
liner specimens have been tested. The
cells containing the two oily wastes are in
the process of being dismantled and the
liners tested.
The membrane liners that were removed
from the cells were photographed, inspect-
ed visually, and subjected to the follow-
ing tests :
- Determination of volatiles and
extractables.
- Tensile properties in machine and
transverse directions.
- Hardness.
- Tear strength in machine and
transverse directions.
- Puncture resistance.
loss in weight of a 2 inch diameter disk
heated for two hours at 105°C. The ex-
tractables were determined on specimens
that had been devolatilized by heating two
hours at 105°C. The appropriate solvent
was used in order to extract the nonpoly-
meric fraction without dissolving the
polymer.
The volatiles test was also used to
verify the direction of the grain of the
membrane which was introduced during man-
ufacture. (On heating polymeric liners
shrink more in the machine direction.)
Some physical properties are significantly
affected by grain direction, a factor which
was not considered in the preparation of
some of the early specimens.
Elongation which is an important
property in the functioning of many rubber
and plastic products appears an important
property for liners. The loss of elon-
gation in service either by loss of plas-
ticizer or by excessive swelling could
result in breakage of a membrane, since it
would not tolerate the extension for which
it was designed. The stress at 100% elon-
gation is a measure of the stiffness or
modulus of a rubber or plastic compound.
It is affected by the amount of swelling
that takes place and by crosslinking or
oxidation that might take place in the rub-
ber. This property is related to hardness.
As shown in Table 4, all of the ma-
terials that were exposed Absorbed vola-
tile material, primarily water, and, in
the case of the lead waste, some low molec-
ular weight hydrocarbons. The two liners
which absorbed the least amount of volatile
material were the elasticized polyolefin and
the polyvinyl chloride membranes. The
rubber liners tended to absorb more; the
neoprene and the chlorinated polyethylene
absorbed the most.
- Seam strength.
The results related to the volatiles
and extractables, ultimate elongation and
modulus, are presented in Tables 4 and 5.
Determining the volatiles and extract-
ables of the exposed liner supplies infor-
mation with respect to the amount of waste
that is absorbed by the liner and the
amount of the original compound that is
leached out during the exposure. The
volatiles were run first by determining the
In most cases, the "extractables", of
the exposed liners were higher than the
extractables of the respective unexposed
liner material. Exceptions were the poly-
vinyl chloride, the chlorinated polyethy-
lene, and, to a minor extent, the neoprene,
all of which had been immersed in lead
waste. The plasticizers in these materials
leached into the wastes resulting in a
lower extractables contents.
The elongation data in Table 5 show
almost complete loss of elongation by the
165
-------
TABLE 4. VOLATILES3 AND EXTRACTABLESb OF PRIMARY POLYMERIC MEMBRANE LINER SPECIMENS
AFTER EXPOSURE TO SELECTED WASTES
Compound data
Liner data
Polymer
Butyl
CPE
CSPE
ELPO6
EPDM
Neoprene
Polyester
PVC
No.c
57R
77
6R
36
26
43
75
59
Extractables ,
Typed %
VZ
TP
TP
TP
VZ
VZ
TP
TP
6.4
9.1
3.8
5.5
18.2
13.9
2.7
35.9
Waste
Unexposed
0.29
0.00
0.29
0.15
0.50
0.45
0.26
0.26
Volatiles, % Extractables, %
and exposure time in days (d) Waste and exposure time in days (d)
Pesticide
1260 d
4.8
7.9
9.7
f
6.3
13.6
2.9
3.6
HN03 Caustic
1220 d 1250 d
11.5 1-4
13.2 2.8
7.2 5.8
5.3 ...f
12.0? 1.3
5.7
7.4h 0.9
...f 1.8
Lead
1340 d
3.5
19.2
11.4
1.5
5.3
17.5
1.7
4.4
Pesticide
1260 d
7.6
9.4
5.4
f
25.2
16.1
5.8
33.4
HN03
1220 d
8.7
10.6
4.6
7.1
22. 8g
13. 5h
f
Caustic
1250 d
7.9
9.1
3.8
...f
24.0
13.7
3.3
35.6
Lead
1340 d
7. 9
7.2
6.0
8.1
26.0
12.2
5.4
22.5
aPercent weight loss after 2h @ 105°C.
bAfter devolatization in air oven for 2h @ 105°C.
cNo. = Serial number of liner set by Matrecon; R = Reinforced with a fabric.
dType = Vulcanized (VZ) or thermoplastic (TP) .
eELPO = Elasticized polyolefin.
fSpecimens still under exposure to waste.
gExposure time = 1150 days.
hfixposure time = 509 days.
-------
polyester elastomer in the nitric acid
waste. The other lining materials re-
tained most of their respective elonga-
tions. The CSPE showed the greatest loss
other than the polyester elastomer, re-
taining 68% of its original elongation.
The S-100 modulus of the neoprene
liner immersed in the pesticide and the
lead wastes showed significant losses.
These losses of modulus are the result of
excessive absorption of water and compo-
nents of the wastes. The pesticide waste
had a low ion concentration which resulted
in high absorption of water. The lead
waste contained oily material which was
absorbed by the neoprene.
There were significant increases in
stiffness (S-100 modulus) by the PVC liner
exposed to the pesticide and the caustic
wastes and by the CSPE which was exposed to
the caustic waste. These changes in stiff-
ness within an exposure period of less than
three and one-half years indicate that, on
extended periods of time, there will be
considerable hardening.
Admix Liner Materials
As shown in Table 3, five types of
admix materials are being studied in this
project:
- Compacted fine-grain native soil.
- Soil cement.
- Modified bentonite in sand.
- Hydraulic asphalt concrete.
- Membrane based on emulsified asphalt
on a nonwoven fabric.
Because of the incompatibility of
some of the wastes with particular admix
materials, several combinations were de-
leted. These deletions are shown in the
table. The only liner material that was
placed below the acid waste was the hy-
draulic asphalt concrete. Neither of the
two oily wastes was placed on the asphalt-
ic liners; however, the lead waste, which
contained a light oily fraction, was placed
on these liners. The performance of the
individual admix liners is discussed below:
Compacted Fine-grain Soil—
All of the wastes, except the nitric
acid waste, were placed above the compacted
fine-grain soil liner. Seepage below all
of the liners took place. The amount of
seepage was measured and the respective pH,
conductivity, and percent total solids were
determined. The following observations are
made with respect to the seepages through
the soil liners:
- The rateof seepage is
10~8 to 10~7
cm sec" which compares favorably with the
permeability of the soil measured in the
laboratory permeameter. There is some
variation in the amount of seepage collec-
ted below the liner which may reflect perm-
eability differences, perhaps due to den-
sity of the soil.
- The fluids being collected after
more than three years of exposure still
continue to be essentially neutral and to
have high solids content (mostly salt) and
electrical conductivity.
- There is a downward trend in solids
content of the seepages collected under the
pesticide and lead wastes, but the seepage
under the spent caustic waste continues to
be 23% solids.
One set of the soil liners has been
removed and tested.
The permeability of a specimen taken
from the cell containing the soil and the
aromatic oil waste was determined using the
standard "back-pressure" procedure. The
sample was collected from a depth of seven
to ten inches below the surface of the
soil, i.e. from that part of the soil which
was not penetrated by the oil. The three
consecutive values obtained were: 1.83
x 10 , 2.43 x 10~8, and 2.60 x 10~8 cm
-1
These figures indicate the low
sec
permeability of the soil, which had a bulk
density of 1.318 g cm~^ and a saturation
degree of 101%.
Trace metals analyses were made of the
soils which were below the lead waste, Oil
104, and the aromatic oil. The testing
included determination of pH and heavy
metal content (cadmium, chromium, copper,
magnesium, nickel and lead) on samples
167
-------
TABLE 5. RETENTION OF ULTIMATE ELONGATION3 AND S-100 MODULUSb OF PRIMARY POLYMERIC MEMBRANE LINER SPECIMENS
ON EXPOSURE TO SELECTED WASTES
Compound data
Liner data
Polymer
Butyl
CPE
CSPE
ELP09
EPDM
Neoprene
Polyester
PVC
NOT
57R
77
6R
36
26
43
75
59
Typed
VZ
TP
TP
TP
VZ
VZ
TP
TP
Extract-
ables, %
6.4
9.1
3.8
5.5
18.2
13-9
2.7
35.9
Original
elongationf %
70
405
235
665
450
320
575
385
Ultimate elongation, % retention
Waste and exposure time in days
Pesticide
1260 d
77
90
88
.. .h
103
84
87
93
HNO3 Caustic
1220 d 1250 d
419 142
88 115
83 68
96 ...h
791 95
95
-------
collected at different depths in the cells. bentonite liners were similar.
With the exception of the liner with
spent caustic, the pH of the Mare Island
soil liner was not significantly altered
by the wastes. The pH of these samples was
in the range of 7.0 to 7.6; the ratio soil:
solution was 1:2 with 0.01 N CaCl being
the equilibration solution.
In the case of the spent caustic, the
pH values were around 9.0 for samples col-
lected in the first two to three centi-
meters, which concurs with our previous
findings that, over the exposure period of
30 months, the wetting front of the wastes
penetrated the soil to a depth of only
three to five centimeters.
The heavy metals distribution, as
indicated by the analysis, shows, in the
case of the lead, only a very shallow con-
tamination of the soil. Similar results
were obtained on all six of the heavy
metals in the case of the soil below the
Oil 104 waste.
Soil Cement—
All of the wastes except the acid
waste were placed on the soil cement liner.
No seepage occurred through the liner dur-
ing the 30 months of exposure.
One set of the soil cement lining
materials was recovered after 625 days of
exposure to the various wastes and the
individual linings have been cored and
tested for compressive strength. In all
cases, compressive strength of the exposed
soil cement was greater than that of the
unexposed material. There was some
blistering of the epoxy asphalt coating
which was applied to one-half the surface
of each specimen.
Modified Bentonite and Sand—
Two types of modified bentonites were
used as liners in ten cells. One type al-
lowed somewhat less seepage than the other.
There was measurable seepage in seven of
the ten cells and one failed allowing the
waste (Oil 104) to come through the liner.
Irrespective ofthe type of waste above
the liner, the quality of the seepage was
not greatly different among the samples
collected. The seepage collected below the
pesticide waste of both types of modified
When the spacers containing the benton-
ite-sand were sampled, it was found that
there had been considerable channeling of
the wastes into these liners. There was no
channeling at the walls of the spacer.
We conclude that this liner is prob-
ably not satisfactory for these types of
of waste. (We prepared these liners as
directed by the suppliers.) The use of a
soil cover on the bentonite layer to pro-
duce an overburden would probably reduce
the channeling effect.
Hydraulic Asphalt Concrete—
Liner specimens of hydraulic asphalt
concrete were placed under four of the
wastes. Excluded were the oily wastes.
This lining material functioned sat-
isfactorily under the pesticide and spent
caustic wastes, but failed beneath the
nitric acid waste. However, the failure
arose primarily from the failure of the
aggregate which contained calcium carbon-
ate; also the asphalt tended to harden
considerably.
In the case of the lead waste, the
asphalt absorbed much of the oily con-
stituents of the waste and became "mushy".
There was some staining of the gravel below
the asphalt liner.
The second cell containing the hy-
draulic asphalt concrete and the lead waste
is still functioning without seepage.
Membrane Based on Emulsified Asphalt and
Nonwoven Fabric—
This membrane was placed under only
three of the six wastes: pesticide, spent
caustic, and lead wastes. The acid waste
was deleted because of the severe harden-
ing the acid waste caused the asphalt, and
the oil wastes were deleted because of the
high mutual solubility of the asphalt and
the wastes.
The asphalt membrane functioned sat-
isfactorily with the pesticide and spent
caustic wastes; however when the cell
containing the lead waste was dismantled
the gravel below the liner was wet and
stained brown. This result indicates that
some seepage took place.
169
-------
IMMERSION TESTING OF THE MEMBRANE LINER
MATERIALS
To expand the matrix of information
relating to the compatibility of membrane
liners and wastes, samples of additional
liner materials were immersed in nine dif-
ferent wastes, including the six wastes
that were used in the primary test program.
This test study was further expanded to in-
clude 22 liner materials and plastics and
13 wastes and test fluids.
In this paper, results on the immer-
sion testing are presented for 12 liner
materials in eight wastes. Data include
exposures of approximately 2.2 years. The
test was designed to immerse specimens of
sufficient size to perform most of the same
tests that had been performed on the pri-
mary samples. Missing from the immersion
study was the seam adhesion test. The
specimens, each six by eight inches, were
weighed and measured prior to exposure and
weighed and measured when removed from the
waste after which they were tested for
volatiles and extractables, tensile, tear,
and puncture strength. The same type of
data as were presented for the primary
liners are reported for the immersion test
specimens. Tables 6, 7, and 8 present data
on weight changes during the immersion, and
"volatiles" and "extractables" of the ex-
posed specimens. Data on the physical
properties, ultimate elongation, and S-100
modulus are presented in Tables 9 and 10.
In all tables, the data are arranged
by waste and liner. Also included in the
tables are data on the liner compounds,
such as the type of compound, that is,
whether it is thermoplastic or vulcanized,
the amount of extractable material in the
original compound, and data for the perti-
nent property measured on the unexposed
lining. The weight gains, extractables,
and volatiles are presented in percent, the
elongation and modulus data are presented
as a retention of the property after the
exposure period.
As shown in Table 6, most of the liner
materials that were exposed in the wastes
gained weight, particularly those that were
immersed in wastes which contained oily
components. Even among the oily wastes the
amount of swelling, or weight change, is
determined by the type of hydrocarbons they
contain, i.e. their molecular weight and
aromaticity. The aromatic oil and the weed
oil caused the greatest degree of swelling.
Both of these wastes are aromatic; further-
more, the weed oil contains low molecular
hydrocarbons. The swelling was so great in
the case of the CPE and CSPE liners that
the specimens disintegrated or swelled to
the point where they could not be effec-
tively tested for physical properties.
In several instances, the PVC liner
specimens actually lost weight. A loss in
weight reflects leaching of the plasticizer
and possibly the loss of absorbed material
by syneresis. Such losses generally result
in reduced elongation and substantially
increased modulus.
Inasmuch as a considerable amount of
the weight of the immersed specimens con-
sists of volatile components, volatility
tests were run, the results of which are
presented in Table 7. Again, the results
show a substantial volatiles content, par-
ticularly in the wastes which are basically
aqueous. The waste which caused the lowest
volatiles in the liners and also the lowest
weight increase was the spent caustic. The
pesticide waste caused comparatively high
weight gain and volatiles content; the
aromatic oil, which resulted in high weight
increases, yielded relatively low volatiles
because it is nonvolatile.
The nonvolatile extractable components
in the immersed liner were determined fol-
lowing the volatility tests with the re-
sults presented in Table 8. Part of the
extractables are from the original compound
and part absorbed from the waste. Compari-
sons can be made with the extractables in
the original compound to determine the
amount of nonvolatile waste the liner ab-
sorbed and the amount of compounding oils
and plasticizers which had leached out of
the compounds. A precise ratio, however,
requires further analysis of the extracted
fraction.
Of particular interest is the signif-
icant loss of plasticizer of the PVC liners
immersed in the lead. Oil 104, and weed oil
wastes. Also of interest is the higher ex-
tractables of most of the liner materials
which were immersed in oil-containing
wastes. The liners that were in the pre-
dominantly aqueous wastes generally did not
lose plasticizer. It can be assumed in
long-term exposure that the soluble compo-
nents of a compound will eventually leach
into the waste and that some waste compo-
170
-------
TABLE 6. ABSORPTION OP WASTE BY POLYMERIC MEMBRANE ON IMMERSION IN SELECTED WASTES
(Data in Weight Percent)
Data on liner
compound
Polymer
Butyl
CPE
CSPE
CSPE
ELPO6
EPDM
EPDM
Neoprene
Polyester
PVC
PVC
PVC
No.3
44
77
6Rd
55
36
83Rd
91
90
75
11
59
88
Type
VZ
TP
TP
TP
TP
TP
VZ
VZ
TP
TP
TP
TP
Extract-
ables, %
11.8
9.1
3.8
4.1
5.5
18.2
23.6
21.5
2.7
33.9
35.9
33.9
Waste and immersion
Pesticide
807 d
1.6
12.7
17.3
15.7
0.5
4.5
20.4
11.4
4.2
5.1
1.0
1.6
HNO3
751 d
3.8
19.9
10.0
10.9
7.6
4.2
50.9
17.4
6.4
22.1
-6.1
28.2
HF
761 d
3.7
12.9
9.0
7.7
1.1
3.1
23.9
12.0
2.0
18.1
0.9
14.3
Spent
caustic
780 d
0.8
1.1
4.3
3.3
0.6
1.6
1.3
1.5
1.5
0.4
-0.9
1.1
time in
Lead
786 d
28.7
118.9
120.7
116.2
17.0
24.8
34.7
59.1
7.4
-1.5
7.4
-5.2
days (d)
Oil Aromatic
104 oil
752 d 761 d
103.9
36.9
49.5
55.0
28.9
26.5
84.7
26.3
8.5
-10.4
-0.5
-9.8
31.2
226.4
105.2
110.5
29.4
19.8
34.2
142.6
16.6
18.5
28.9
14.1
Weed
oil
809 d
64.2
NDC
368.4
347.5
38.1
84.4
76.2
89.3
14.7
15.3
24.7
25.2
aNo. = the serial number of liner set by Matrecon.
bType = vulcanized (VZ) or thermoplastic (TP).
CND specimen was lost; some indication that it dissolved in the waste.
"R = fabric reinforced.
eELPO = jslasticized polyolefin.
-------
TABLE 7. VOLATILES CONTENT OF FLEXIBLE POLYMERIC LINERS
ON IMMERSION IN SELECTED WASTES
(Data in percent loss of weight)
ts>
Waste and immersion
Data on liner
Polymer
Butyl
CPE
CSPE
CSPE
ELPOh
EPDM
EPDM
Neoprene
Polyester
PVC
PVC
PVC
No.b
44
77
6R9
55
36
83Rg
91
90
75
11
59
88
Type0
VZ
TP
TP
TP
TP
TP
VZ
VZ
TP
TP
TP
TP
Extract-
ables , %
11.8
9.1
3.8
4.1
5.5
18.2
23.6
21.5
2.7
33.9
35.9
33.9
compound
Volatiles
(unexposed) ^<
0.28
0.00
0.29
0.42
0.15
0.31
0.34
0.27
0.26
0.15
0.26
0.17
Pesticide
i 807 d
3
8
10
7
0.
4.
15
7 .
i.
30
40
3.
.0
.6
.8
.9
45
89
.6
26
75
.9
.1
51
HN03
751 d
3.8
20.4
10.4
9.1
7.2
3.1
22.6
13.7
28.0
22.6
10.6
23.2
HF
761 d
2.9
8.1
7.6
7.1
1.3
2.8
23.6
10.5
3.8
18.3
3.0
13.1
Spent
caustic
780 d
6.9
1.6
4.2
3.6
0.6
2.2
1.2
2.9
1.3
0.6
1.5
1.2
time in
Lead
786 d
15.3
42.2
3.3
1.1
9.5
13.9
18.5
23.6
6.1
16.1
12.7
11.5
days (d)
Oil
104
752 d
9.8
5.2
20.2
10.0
3.1
8.5
11.8
6.3
2.1
2.7
3.3
4.3
Aromatic
oil
761 d
3.3
NDe
6.1
4.8
1.5
4.4
4.2
3.9
3.4
3.6
4.6
3.5
Weed
oil
809 d
19.2
NDf
50.7
39.7
16.4
24.1
33.4
28.9
9.5
17.1
16.1
24.1
a "Volatiles" content is the loss in weight of a specimen of liner (unexposed or exposed) on air-oven heating for
2 hours at 105°C .
bNo. is the serial number of liner assigned by Matrecon for identification.
cType of compounds: vulcanized (VZ) or thermoplastic (TP) .
^Volatiles content of unexposed specimens of liners.
eND = no data obtained because the sample was too badly deteriorated to test.
^Immersion sample was lost. Some indication that it dissolved in the waste.
?R = fabric reinforced.
hELPO = e_lasticized polyolefin.
-------
TABLE 8. EXTRACTABLES OF FLEXIBLE MEMBRANE LINERS AFTER IMMERSION
IN WASTE AND DEVOLATILIZING IN OVEN FOR 2 HOURS AT 105°C
(Date in weight percent based upon the devolatilized specimen)
Data on liner
compound
Polymer
Butyl
CPE
CSPE
CSPE
ELPO
EPDM
EPDM
Neoprene
Polyester
PVC
PVC
PVC
No.
44
77
6Rd
55
36
83Rd
91
90
75
11
59
88
Type
VZ
TP
TP
TP
TP
TP
VZ
VZ
TP
TP
TP
TP
Extract-
ables , %
11
9
3
4
5
18
23
21
2
33
.8
.1
.8
.1
.5
.2
.6
.5
.7
.9
35.9
33
.9
Waste and immersion
Pesticide HNO
807 d 751 d
11.1
12.0
3.8
4.0
6.2
17.6
25.4
19.2
10.8
35.4 32.3
34.9 23.3
31.7 33.2
HF Spent
caustic
761 d 780 d
10.6
10.2
3.8
3.6
5.1
17.1
22.4
19.6
3.3
32.7
34.0
31.1
10.9
9.2
3.8
4.2
5.5
18.0
22.9
20.8
2.7
33.9
35.8
32.5
time in days (d)
Lead
786 d
17.1
16.6
4.6
3.7
6.9
21.2
30.0
19.2
3.3
17.3
27.9
17.9
Oil Aromatic
104 oil
752 d 761 d
40.3
19.2
16.3
15.9
17.8
22.2
43.5
23.9
6.5
18.0
28.0
14.8
27.2
NDC
45.4
59.8
23.3
27.2
38.4
58.5
16.6
40.6
47.6
38.5
Weed
oil
809 d
15.2
ND°
19.0
16.1
8.1
18.7
25.4
7.6
6.1
21.1
30.1
20.1
No. = The serial number of liner set by Matrecon.
Type = Vulcanized (VZ) or thermoplastic (TY).
Specimen was lost; some indication that it dissolved in the waste.
R = Fabric reinforced.
ELPA =Elasticized gplyolefin.
-------
TABLE 9. PERCENT RETENTION OF S-100 MODULUS BY FLEXIBLE POLYMERIC LINERS ON IMMERSION IN SELECTED WASTES
(Data in Percent of S-100 of the Unexposed Liner)
Waste and immersion
Data on liner
Polymer
Butyl
CPE
CSPE
CSPE
ELPO"
EPDM
EPDM
Neoprene
Polyester
PVC
PVC
PVC
No.b
44
77
6R9
55
36
83B9
91
90
75
11
59
88
Type0
VZ
TP
TP
TP
TP
TP
VZ
VZ
TP
TP
TP
TP
Extract-
ables, %
11.8
9.1
3.8
4.1
5.5
18.2
23.6
21.5
2.7
33.9
35.9
33.9
compound
Original
S-100, d psi
310
900
945
880
875
810
340
560
2585
1280
960
1735
Pesticide
807 d
88
102
124
122
122
94
90
77
111
118
135
100
HNOa
751 d
72
69
67
71
99
84
58
63
-
104
262
70
HF
761 d
93
116
126
113
112
109
100
104
117
106
129
83
Spent
caustic
780 d
91
126
170
167
116
100
100
121
105
121
128
99
time in
Lead
786 d
57
18
88
93
80
44
73
38
92
102
91
97
days (d)
Oil
104
752 d
45
45
88
89
79
54
58
70
95
207
152
172
Aromatic
oil
761 d
56
NDe
73
85
76
58
60
40
76
131
112
123
Weed
oil
809 d
39
NDf
NDe
NDe
57
NDe
67
27
78
52
47
45
aS-100 modulus is the average of the values obtained in each the machine and transverse directions.
bNo. is the serial number of liner assigned by Matrecon for identification.
cType of compounds: vulcanized (VZ) or thermoplastic (TP).
^Original S-100 modulus is the averaged S-100 of the liner membrane before immersion in the waste.
SND = no data obtained because the sample was too badly deteriorated to test.
^Immersion sample was lost. Some indication that it dissolved in the waste.
9R = fabric reinforced.
hELPO = elasticized polyolefin.
-------
TABLE 10. PERCENT RETENTION OF ULTIMATE ELONGATION3 BY FLEXIBLE POLYMERIC LINERS ON IMMERSION IN SELECTED WASTES
(Data in Percent Retention)
Waste and immersion
Data on liner
Polymer
Butyl
CPE
CSPE
CSPE
ELPOh
EPDM
EPDM
Neoprene
Polyester
PVC
PVC
PVC
»0>
44
77
6R9
55
36
83R9
91
90
75
11
59
88
Type0
VZ
TP
TP
TP
TP
TP
VZ
VZ
TP
TP
TP
TP
Extract-
ables , %
11.8
9.1
3.8
4.1
5.5
18.2
23.6
21.5
2.7
33.9
35.9
33.9
compound
Original
elongation, d
445
405
235
280
665
255
490
415
575
350
385
330
Pesticide
% 807 d
109
94
82
79
95
128
101
101
89
103
96
98
HN03
751 d
110
88
126
104
97
135
97
87
1
86
65
86
HF
761 d
93
81
93
94
96
109
86
68
90
86
93
89
Spent
caustic
780 d
94
87
72
74
91
111
106
83
84
97
86
100
time in
Lead
786 d
111
92
65
60
84
157
98
91
101
104
101
105
days (d)
Oil
104
752 d
54
92
79
68
88
145
52
77
96
85
96
78
Aromatic
oil
761 d
102
NDe
77
52
90
155
100
62
87
76
78
82
Weed
oil
809 d
71
NDf
NDe
NDe
78
NDe
56
65
96
107
96
107
aUltimate elongation = the average of the values in both machine and transverse directions.
^No. = the serial number of liner set by Matrecon.
cType = vulcanized (VZ) or thermoplastic (TP).
dOriginal elongation is the averaged ultimate elongation of the liner membrane before immersion in the waste.
eND = no data because sample was too deteriorated to test.
fND specimen was lost; some indication that it dissolved in the waste.
9R = fabric reinforced.
hELPO = elasticized jDolyolefin.
-------
nents will swell the lining materials. The
net effect will determine the overall de-
gree of swell of the lining materials and
the effect upon physical properties, part-
icularly elongation and stiffness.
In Tables 9 and 10 the results of the
retention of elongation and modulus are
presented in comparison with the original
values obtained on unexposed samples of
the respective liners. The effect upon
different liners varies considerably with
the waste, polymer, and liner compound.
Of particular interest are the effects upon
the PVC compounds which vary among the
three PVC's and among the wastes. The ni-
tric acid waste caused a significant in-
crease in the stiffness of one PVC membrane
but a substantial loss in another. Two of
the oils, Oil 104 and the aromatic oil, in-
creased the modulus; weed oil reduced the
modulus. Again, these effects are a func-
tion of the aromaticity of the oil and its
molecular weight.
The retention of elongation, a prop-
erty which appears to relate to the func-
tioning of a liner, varies considerably
with the liner and the waste. In several
cases, the elongation of the liner was re-
duced by almost 50%, a value which in many
rubber products is considered to be crit-
ical to utility. Low retention values were
also encountered with CSPE and EPDM in oily
wastes. The polyester elastomer became
nonserviceable in contact with nitric acid
waste early in the exposure period; it had
an elongation of less than 1% of its orig-
inal elongation within a few days of ex-
posure. One PVC liner lost 35% of its
elongation, which reflects a substantial
loss of plasticizer. PVC's in other wastes
are also showing weight losses which could
be indicative of long-term problems with
this type of liner where high oily wastes
are used. However, if polymeric liners are
placed into a "proper" waste, the loss of
plasticizer could be offset by the absorp-
tion of the waste to yield compatible com-
binations, which would not result in ex-
cessive hardening and loss of elongation.
WATER ABSORPTION OF POLYMERIC LINERS
As most of the wastes contain water
which most polymeric materials absorb to
varying degrees, the water absorption of a
series of liners is being measured for ex-
tended periods of time, in accordance
with ASTM D570. The results of the tests
at 1113 days at room temperature and at
70°C are presented in Table 11. The change
in weight varies considerably with the
material and with the temperature of test.
At room temperature, the weight change
ranges from + 0-9% for elasticized poly-
olefin to + 67% for a neoprene liner ma-
terial. At 70°C, the change was a loss of
4% for the elasticized polyolefin to ex-
cessive swelling and disintegration of the
specimen of butyl liner. Other liners
which showed excessive swelling at the
higher temperatures were the two neoprenes,
the chlorinated polyethylene, and the
chlorosulfonated polyethylene. One of the
two polyvinyl chloride liners in the test
swelled or increased in weight 125%,
whereas the second increased only 26%; how-
ever, the liner which increased the lesser
amount had become quite stiff, indicating
that it had lost plasticizer. Therefore,
in considering the weight gain of a lining
material on exposure to a waste, the loss
of oils and plasticizers within the com-
pound should also be considered. Further-
more , the swelling at the higher temper-
atures , although rapid, cannot always be
used as a measure of the level of weight
increase at lower temperatures. For ex-
ample , the butyl that disintegrated at
70°C, swelled 11% at room temperature, and
the CSPE which swelled almost 440% at 70°C,
swelled about 21% at room temperature. On
the other hand, a neoprene liner swelled
350% at 70°C and 44% at room temperature.
OUTDOOR EXPOSURE OF POLYMERIC MEMBRANE
LINERS
The second of three sets of specimens
of 11 polymeric membranes that were exposed
to the weather have been recovered and
tested after 745 days of exposure. These
specimens are six-inch squares cut from the
sheeting and mounted at 45° on a rack fac-
ing southward on the roof of our laboratory
in Oakland, California. On removal from
the rack, the samples are cleaned, weighed,
and their dimensions measured, after which
selected physical properties are measured.
The effects of the exposure on the ultimate
elongation and the weight of the material
are presented in Table 12. As discussed
above, the retention of elongation is a
measure of the rubberiness of the membrane
that is retained and the change in weight
is an indication of the loss of volatiles,
e.g. plasticizer, and possibly of the de-
gradation of the polymeric compound. All
materials remained rubbery and serviceable,
176
-------
TABLE 11. CHANGE IN HEIGHTS3 OF POLYMERIC MEMBRANE LINERS IMMERSED IN WATER
FOR 1113 DAYS
(Percent weight change)
Liner data
Polymer
Butyl
CPE
CSPE
ELPO
EPDM
EPDM
Neoprene
Neoprene
Polyester
PVC
PVC
Compound data
Immersion
Extract- temperature
No.D Type0 ables, % Room
57R
77
6R
36
8
26
43
82
75
11
59
""Test method ASTM D570.
bMatrecon
ment.
identification
VZ
TP
TP
TP
VZ
VZ
VZ
VZ
TP
TP
TP
Data are
number; "
6.4
9.1
3.8
5.5
23.4
23.0
13.7
13.4
2.7
33.9
35.9
10.9
16.9
20.8
0.9
3.5
3.9
67.2
44.0
1.6
1.7
4.5
70°C
_d
179.9
438.2
-4.2
12.6
21.1
387.7
350.0
-O.P
126.3
26.2
the average of three values.
R" indicates liner has fabric reinforce-
°Type = Vulcanized (VZ) or thermoplastic (TP) .
^Specimen
TABLE
disintegrated .
12. OUTDOOR EXPOSURE OF POLYMERIC MEMBRANE LINERS ON ROOF RACK
Effects upon the Ultimated Elongation3 and Heights
of the Membrane
Specimens
Liner
Polymer
Butyl
CPE
CSPE
ELPO
EPDM
EPDM
Neoprene
Neoprene
Polyester
PVC
PVC
data Thickness,
No/5
57R
77
6R
36
8
26
43
82
75
11
59
mils Typec
34 VZ
29 TP
32 TP
23 TP
62 VZ
39 VZ
34 VZ
60 VZ
7 TP
30 TP
33 TP
Compound
Extract-
ables, %
6.4
9.1
3.8
5.5
23.4
23.0
13.7
13.4
2.7
33.9
35.9
data
Ultimate
elongation, %
75
325
240
675
520
440
330
390
340
365
370
Retention of
elongation, %
343 d 745 d
41 125
100 92
63 52
96 96
94 89
89 94
76 68
80 64
89 88
100 93
92 88
Change
weight
343 d
-1.89
-0.99
+0.64
-0.72
-2. 02
-1.69
-1.80
-0.53
-2.50
-1.36
-1.45
in
/ *
745 d
-2.79
-2.27
+0.91
-1.55
-3.O9
-2.71
-2.96
-1.17
-8.32
-8.40
-6.66
aElongation of the liner measured in the machine direction. Retention of elongation on expo-
sure after 343 and 745 days given in % of the original elongation.
bMatrecon identification number. "R" indicates liner is fabric reinforced.
°Type: Compound is vulcanized or is thermoplastic.
Ultimate elongation of unexposed liner.
177
-------
which indicates that they would be satis-
factory for several years more. However,
the CSPE and neoprene membranes lost sig-
nificant elongation, probably the result of
crosslinking which had taken place during
exposure. All three of these compounds in-
creased substantially in modulus which, in
rubber technology, is related to crosslink-
ing of the rubber. The butyl rubber liner
has given erratic results, probably due to
its low initial elongation, i.e. 75%, and
to the fabric reinforcement.
All of the specimens lost weight ex-
cept that based on CSPE. The greatest los-
ses were incurred by the two PVC specimens
and by the polyester elastomer specimen.
The loss of weight by the PVC specimens is
no doubt caused by loss of plasticizer;
however, the loss did not cause a signif-
icant loss of physical properties. The
loss in weight by the polyester which does
not contain plasticizer may reflect polymer
degradation.
In addition to the small slabs of
liners which are being exposed to the wea-
ther on the roof rack, we have 12 small
tubs lined with various polymeric membranes
and filled with wastes. In this test, nine
different lining materials are being ex-
posed to four different wastes. Of the 12
tubs, only one has failed, a tub lined with
elasticized polyolefin and containing Oil
104 waste. The specimen cracked at a
sharp fold in the area which is exposed at
times to both the oil and the air. This
liner also showed significant swelling.
The butyl liner is showing cracking at
sharp bends. All of the liners containing
oil exhibit some swelling. Also, the poly-
vinyl chloride liner containing the nitric
acid waste is hardening.
Exposure is being continued and the
tubs will be dismantled when they fail and
the specimens tested.
BAG TEST FOR ASSESSING MEMBRANE LINER
MATERIALS
The bag test described above, which
was developed originally for use on the
landfill project, has been applied to a
variety of thermoplastic polymeric materi-
als and the wastes being used in this pro-
ject. The initial tests were made with the
thermoplastic materials because they could
be fabricated into bags with ease by heat
sealing. Some of these bags have now been
exposed more than 1000 days. This test
method continues to be a very promising one
for assessing the long-term performance of
membrane lining materials.
We showed previously that the bags
containing the wastes actually increased in
weight, indicating the flow of water into
the bags through osmosis, as shown in Table
13. The long-term tests now show that some
ionic material is diffusing through the
liners into the deionized water in the
outer bags. Table 14 presents the inter-
polated or estimated times to reach an
electrical conductivity of 1000 ymho for
slopwater and nitric acid, both of which
are concentrated wastes. The data show the
higher permeability of the PVC lining ma-
terials as compared to CPE, CSPE, elas-
ticized polyolefin, and polybutylene. Of
particular interest is the low permeability
of the elasticized polyolefin and the poly-
butylene, both of which are partially cry-
stalline materials. Table 15 presents the
results of thethermoplastic membranes test-
ed with 5% sodium chloride solution. The
data again show the greater permeability of
the PVC with respect to the CSPE and the
elasticized polyolefin, which is the most
impermeable of the three. These bags have
now been taken out of test and physical
properties of the bag wall materials have
been measured. The results show that,
within the 1150 days of exposure, there is
some loss in elongation and an increase in
the stiffness of the membranes.
DISCUSSION
At this point in the project, when we
are recovering and testing specimens which
have been exposed to wastes under some test
conditions for more than three years, we
can make a few observations which we feel
will be helpful in the choice of lining ma-
terials for impounding specific wastes.
There does not appear to be any single
lining material now commercially available
which is suitable for long-term impoundment
of all wastes. Generally, wastes that are
highly ionic, contain salts, strong acids, or
strong bases, can be aggressive to soils,
soil cement, aggregate, hydraulic asphaltic
concrete, and to some of the membranes, par-
ticularly those containing plasticizers.
Wastes which have organic components can be
aggressive toward the membrane liners and
toward asphaltic materials. However, there
178
-------
TABLE 13. RELATIVE PERMEABILITIES OF POLYMERIC MEMBRANE LINING
MATERIALS IN BAG TEST WITH THREE WASTES
Average Flux into the bag m grams per square meter per day x 10~2a
Polymer
CPE
CSPE
ELPOC
Polybutylene
PVC
PVC
Liner
no.
86
85
36
98
19
88
Nominal
thickness, mils
22
33
23
7.5
22
20
HN03
waste
78.2
67.8
2.54
2.98
32.4
64.2
Spent
caustic
26.3
36.3
3.80
7.94
78.8
65.9
Slop
water
190.
49.
18.
13.
325.
118.
7b
2
4d
6
0
8e
aExposure time is 552 days unless otherwise noted.
bBag failed at 450 days.
°Elasticized polyolefin.
dBag failed at 300 days.
eBag failed at 40 days.
TABLE 14. PERMEABILITY OF THERMOPLASTIC POLYMERIC MATERIALS
IN OSMOTIC BAG TEST
Time in Days for Electrical Conductivity of Water in Outer Bag
to Reach 1000 pmho
Wall of inner bag
Liner Thickness, Extractables
Polymer No.a mils %
CPE 86 20
CSPE 85 33
ELPOb 36 22 5.5
PBC 98 7
PVC 19 20 38.9
PVC 88 20 33.9
Haste
] , HNO3
waste
200
500
300
600
70
110
in inner bag
"Slopwater"
420
510
>1000
>1000
200
160
,Matrecon identification number.
Elasticized polyolefin.
Polybutylene .
TABLE 15. BAG TEST OF THERMOPLASTIC
MEMBRANES3
Bags Filled with 5% NaCl Solution
Polymer
Liner No.c
Thickness, mils
Volatiles content of exposed bag wall, *
Change in weight of bag plus waste, %
Change in weight of fluid in bag during exposure,
Conductivity of water in outer bag, pmho
Retention of physical properties, %:
Elongation
S-100
CSPE
6
32
8.7
2.6
% 0.95
585
95
106
ELPO PVC
36 59
23 33
0.38 0.90
0.71 0.38
0.76 0.38
34 4500
100 94
119 120
Exposure time: 1150 days (164 weeks).
Elasticized p_olyolefin.
Matrecon identification number.
179
-------
is considerable variation in the compatib-
ity of wastes and liners. In particular,
the oily wastes cause swelling and loss of
physical properties in the membrane liners
and the asphaltic materials.
It is quite apparent, therefore, that
compatibility studies must be performed
prior to the selection of the liners for
handling specific wastes and that, once the
selection has been made and the site lined,
the waste that is being impounded should be
monitored closely to insure that only prop-
er waste is added. In making these studies,
contact of samples of the liner under con-
sideration and the waste to be impounded
should be as long a time as possible. Our
tests show that, in many cases, no leveling
off with time occurs in the effects of the
waste on the membrane. Such behavior means
that, for long-term exposures, there can be
a continuing change of the liner with time.
In developing a new liner material for
use in the lining of waste disposal sites,
it is quite obvious that the new material
should be tested under a variety of enviro-
mental conditions. Furthermore, it would
appear desirable to consider composite
types of liners involving two or more ma-
terials for use in impounding more aggres-
sive and polluting types of wastes.
In the case of the polymeric membrane
liners, some of the more useful tests and
properties which appear to be promising for
assessing the suitability of a given liner
for use with a given waste are:
1. The measurement of swelling of the
membrane materials in the wastes or in the
leachate of the wastes.
2. The loss of elongation with time of
the polymeric membranes when in contact
with the waste or its leachate.
3. The use of a bag test for measur-
ing the compatibility of the liners and the
waste and the permeability of the liner to
the constituents of the waste.
There is still a need to correlate these
experimental tests with long-term expos-
ures in order that laboratory tests can be
used to assess liners and predict their
long-term performance in service. Such
correlation requires observation of liners
in actual service under extended periods of
time.
ACKNOWLEDGMENTS
The research which is reported in this
paper is being performed under Contract 68-
03-2173, "Evaluation of Liner Materials Ex-
posed to Hazardous and Toxic Sludges", with
the Environmental Protection Agency, Munic-
ipal Environmental Research Laboratory,
Cincinnati, Ohio. The support and guidance
of Mr. R. E. Landreth, Project Officer, is
gratefully acknowledged.
REFERENCES
1. Haxo, H. E., "Evaluation of Selected
Liners When Exposed to Hazardous Wastes,"
in Proceedings of the Hazardous Waste Re-
search Symposium, Residual Management by
Land Disposal, EPA 600/9-76-015, July 1976.
2. Haxo, H. E., "Interaction of Selected
Lining Materials with Various Hazardous
Wastes" in Proceedings of the Fourth Annual
Research Symposium, Land Disposal of Haz-
ardous Wastes, EPA 600/9-78-016, August
1978.
3. Haxo, H. E., R. S. Haxo, and R.M. White,
"First Interim Report - Liner Materials Ex-
posed to Hazardous and Toxic Sludges," EPA
600/2-77-081, June 1977.
180
-------
ASSESSMENT OF PROCESSES TO STABILIZE ARSENIC-LADEN WASTES
Jaret C. Johnson and Robert L. Lancione
JBF Scientific Corporation
Wilmington, Massachusetts 01887
ABSTRACT
C/,
«>wi
O X
£*-
Industrial solid waste containing arsenic was treated with many proprietary and
generic fixation processes to evaluate the processes' ability to retard the leaching of
arsenic. This evaluation was achieved by performing several types of laboratory leaching
tests. Several processes were found to reduce arsenic leaching rates by at least four
orders of magnitude. A concluding discussion shows that fixation can be economically com-
petitive with other legal methods of hazardous waste disposal. Several barriers to i mp 1 £-
mentation are identified.
7
INTRODUCTION
Many industrial residues containing
arsenic are presently held in long-term
storage because of the lack of an accept-
able disposal method. Other similar resi-
dues are disposed of on land, but the
design, construction and operation of
secured landfills that can protect water
resources from arsenic contamination are
often very difficult and expensive. This
paper describes investigations conducted to
determine whether fixation is a useful
technique that can expand the options for
safe disposal of arsenic-containing wastes.
Previously published papers^''^) and
a draft final report'-^) have presented
details of the experimental work. Many of
these details are, therefore, mentioned
only briefly in this paper, which addresses
the more interpretive phases of the work.
EXPERIMENTAL
As the above-cited papers and report
describe, three wastes were selected for
investigation based on a national survey of
arsenic producers and users. The wastes
differed widely from each other in physical
and chemical properties. Only one waste -
filter cake from the purification of food-
grade phosphoric acid - is discussed in
this paper, because the data from that
waste demonstrate several important points.
The waste was treated by a variety of
fixation processes. These included pro-
cesses performed by commercial ventures,
those developed at universities, and some
generic processes that were performed in
the JBF laboratories. Leaching tests were
conducted with all products. This paper
discusses the results for six vendor pro-
cesses. Products were molded by all ven-
dors into cylinders 7.5 cm in diameter and
approximately 20 cm long.
Two types of leaching tests are dis-
cussed here: a long-term shake test and a
modified Extraction Procedure (EP). The
version of the EP was current at the time
of this work, but differs from that later
proposed'^' by EPA for identification of
hazardous wastes under Section 3001 of the
Resource Conservation and Recovery Act
(RCRA).
The shake test was used to
long-term behavior of intact specimens in
contact with water. Shake test steps are
briefly described:
1
Cut a cylindrical slice of approx-
imately 100 g from the fixed mono-
lithic product (usually about 2 cm
thick).
Support the sample so that all
sides are exposed to water in a
glass jar. Add ^-saturated
181
-------
distilled water in the ratio 10 ml
sample. Raw waste tests involved a 50-g
sample and 1 liter water.
3. Oscillate jar at 60 1-in. strokes
/hr.
4. Sample and replace water every
48 hrs. Analyze sample for com-
ponents of interest.
5. Repeat Step (4) for up to two
months.
The version of the EP used was as
follows:
EQUIPMENT
1.
2.
An agitator that, while preventing
stratification of sample and
extraction fluid, also ensures
that all sample surfaces are con-
tinuously brought into contact
with well-mixed extraction fluid.
Equipment suitable for maintaining
the pH of the extraction medium at
a selected value.
PROCEDURE
Weigh a representative sample of
the waste to be tested. Separate
sample into liquid and solid
phases by either centrifugation,
followed by filtration of the
liquid through a 0.4- to 0.5-y
filter media, or by pressure fil-
tration using a 0.4- to 0.5-vi
filter having a surface area of at
least 0.5 cm^/g of sample. Save
the liquid for further use.
Grind the solid material, if
necessary, to pass through a
9.5 mm (3/8") standard sieve.
The solid material is taken and
added to eight times its weight of
deionized water. The pH of the
solution is then adjusted to pH 5
with 1:1 acetic acid or 1 N sodium
hydroxide. Samples are to be
determined electrometrically
following standard calibration
procedures.
Samples are to be maintained at
room temperature during extrac-
tion. They are to be agitated for
24 ± 0.5 hr with pH to be main-
tained during leaching within the
range 4.9-5.2. The preferred
method of maintaining pH is with
automatic titration. If the
necessary equipment is not avail-
able, manual procedures can be
employed.
5. At the end of the 24-hr extraction
period the solution is filtered as
in Step 1. The filtrate is then
adjusted with deionized water so
that its volume is 10 times the
initial weight of solid sample
[v(cc)/w(g)J. The liquid is com-
bined with the original liquid
phase and the solid reextracted
with fresh extractant as in steps
3 and 4.
6. At the end of the second extrac-
tion period the mixture is filter-
ed, the concentration adjusted as
in Step 5, and the liquid phase
combined with that from the pre-
vious separations. This combined
liquid, and any precipitate that
later forms, is designated as the
toxicant extraction procedure
eluate.
pH ADJUSTMENT PROCEDURES
Automated: Follow manufacturer's
instructions as to procedures for instru-
ment calibration and operation.
Manual:
dure to use
manually.
This section prescribes proce-
if extractant pH is maintained
1. Calibrate pH meter in accord with
manufacturer's specifications.
2. Adjust pH of solution to 5.
3. Manually adjust pH of solution at
15, 30 and 60 minutes, increasing
the interval if the pH did not
have to be adjusted more than
0.5 pH units since the previous
adjustment.
4. Continue procedure for a period of
not less than 6 hrs.
5. Final pH after a 24-hr period must
be within the range 4.9-5.2.
182
-------
6. If the conditions of Step 5 are
not met, continue pH adjustment at
approximately 1-hr intervals for a
period of not less than 4 hrs.
Other leaching tests were performed and are
reported elsewhere'^"'); space does not
permit coverage of all results in this
paper.
ANALYSES
Arsenic analyses were performed on a
Perkin-Elmer Model 372 atomic absorption
spectrophotometer. Digestions for assay
analysis of the stabilized residues were
conducted in a mixture of hydrofluoric,
hydrochloric, and nitric acid within a
tightly sealed Teflon®* vessel.
DISCUSSION OF RESULTS
Forms of Presentation
The leaching data from the shake tests
are presented in two forms: the maximum
arsenic concentration (usually found at the
first sampling/water change event), and the
cumulative mass of arsenic leached after
repetitive sampling and analysis. This
second form of data reporting has been used
in studies on other industrial wastes'^),
and on radioactive wastes'"'''. The mass
of pollutant leached is considered in rela-
tion to that initially present in the
sample, and the fractional mass leached in
each time period is expressed
cumulatively. It is also useful for
comparing data from differently con-
figured specimens to normalize the data
with the volume:surface ratio. Therefore,
the parameter
Where
£a = cumulative mass of arsenic leached
n in n periods
A0 = mass of arsenic initially present
in sample
V = sample volume
S = sample apparent surface area
has been used to express leaching behavior
in shake tests. Godbee and Joy( '
showed the usefulness of this parameter in
connection with diffusion theory; they
used the expression
.*
*
where
De = effective diffusivity
t = time
and other symbols have been defined above.
Plotting
la.
versus
tl/2
*Registered trademark of E.I. du Pont
de Nemours and Company, Inc., Wilmington,
Delaware
should yield a straight line with slope
equal to 2(De/ir)1'2 if diffusion of
arsenic to the surface of the solid is the
limiting factor in leaching. The effec-
tive diffusivity can then be computed from
the value of this slope.
Results
Data from the two leaching tests used
are shown in Table 1. For the shake test,
two forms of results are shown: the maxi-
mum arsenic concentration (usually ob-
served in the first sampling) and De, the
effective diffusivity. Several observa-
tions result from review of these data:
1. The extraction procedure on
crushed samples was not well correlated
with the shake test. The least successful
product according to the EP was the most
successful according to the shake test.
2. The two forms of shake test
results are consistent with one another;
inferences about process ranking based on
initial arsenic release were confirmed by
the long-term release rate based on effec-
tive diffusivity, De.
3. The long-term release rate is a
more sensitive indicator of success than
the other parameters shown. While concen-
trations varied by a factor of 100 to 200,
effective diffusivity values varied by a
factor of 17,000.
This last inference is demonstrated by
a diffusion plot, as shown in Figure 1.
183
-------
(CUMULATIVE FRACTION LEACHED)-(VOLUME/SURFACE AREA), cm
id
c:
-s
n>
o>
n>
Q>
-s
O)
— '•
o
CO
S
o>
o
en
rt-
O
O
-J
Ol
-s
(D
n
o
CL
fD
i/>
o
-s
-s
0
o
fD
fl>
l/l
Q.
rt>
i/>
-I
m
101 —
CO cn
101-
00
o
o
Ol
p
b
p
b
Ol
o
IN3
O
O
b
ro
01
p
b
01
O
p
b
O)
en
K < ~° 3)
H > 33 >
>
o
TJ H CO
-I 33 30
m > <~>
3) o
-------
TABLE 1. RESULTS OF LEACHING TESTS
EP Result
Process (mg/x, As)
B 127
C 5.2
F 1.7
G 162
H 97
K 43
Raw Waste 41
* = Granular
Shake Test Results
Maximum As
Concentration De
(mg/i) (cm2/day)
2.1
3.4
1.1
0.2
7.9
4.0
40
1.8xlO~5
1.9xlO-5
2.6xlO-8
l.lxlO-8
1.9xlO~4
*
*
Computations of De were based on the slopes
from this plot.
ECONOMICS
Normally, generators of hazardous waste
will seek to minimize disposal costs within
regulatory constraints on environmentally
acceptable disposal methods. EPA. in its
Proposed Rules on Hazardous Waste'4',
briefly related the concept of fixation to
environmental acceptability in the context
of the Extraction Procedure (EP); it stated
that the Extraction Procedure "is designed
to encourage the chemical or physical 'fix-
ing' of waste so that its constituents are
no longer available to be leached out."
This view of fixation requires complete
success in making the waste non-hazardous.
Whether such success was achieved in this
study is uncertain because the officially
proposed EP differs from that used here.
If one assumes that some processes used
here would have been successful in render-
ing wastes non-hazardous (perhaps with more
tailoring of processes to suit the wastes),
some economic generalizations can be
derived.
Cost estimates provided by fixation
vendors for the processes used in this
study varied widely, were accompanied by
many caveats, and were not correlated with
processes' success. For discussion
purposes, however, a typical approximate
cost can be set at approximately JlOO/kkg
(metric ton) of waste. Literature
values^8' 9» 10) for C0sts of transport
and disposal of hazardous wastes at distant
sites (e.g. 500 miles to an acceptable
site, which is common for industries in New
England) indicate that costs for this
option can exceed ?100/kkg. Thus, fixation
and disposal at a nearby conventional muni-
cipal landfill appears cost competitive.
Another way to view fixation is under
the "degree of hazard" concept. Comments
on the Proposed Rules have frequently dis-
puted the classification of all wastes into
only two groups: "hazardous" and "non-
hazardous". The point has been made by
many commenters that several classes of
"degree of hazard", each with appropriate
disposal requirements, may be more cost
effective and environmentally sound. If
the "degree of hazard" concept were imple-
mented, hazardous waste generators would
have the opportunity to demonstrate that
wastes, even though not totally benign,
could be disposed of with techniques less
sophisticated than those required for very
hazardous wastes. For example, wastes that
had been fixed to reduce their leaching
potential to an intermediate extent might
be disposed of at an intermediate tech-
nology landfill. Cases would probably
arise where this opton would have lower
costs and no more environmental hazard than
either of the options described in the
preceding paragraph.
BARRIERS TO IMPLEMENTATION
Any evaluation of the market potential
of a technology such as fixation requires
an unbiased analysis of a complex set of
technical, economic and other factors. The
foregoing discussion has shown that fixaton
has high potential for expanding the op-
tions for safe disposal of hazardous
wastes. Although these general findings
cannot be applied to every waste/site com-
bination, the fixation option appears suf-
ficiently attractive to stimulate questions
about why it is not used more frequently.
Among the most important reasons for
the lack of widespread use of fixation are:
Delays in official promulgation of
hazardous waste regulations;
Uncertainty about the long-term
stability of fixed products;
185
-------
Uncertainty about future defini-
tions of "hazardous". Disposal is
normally a perpetual commitment,
but industry is concerned that
what is legal today may be found
unwise through future research.
All three of these reasons are based on
inadequate knowledge. Active research pro-
^J iwjrams can improve the state of knowledge,
i \j the basis for sound regulations, and the
V\decisions of hazardous waste generators.
V &
CONCLUSIONS
Commercially available fixation pro-
cesses offer a wide range of effectiveness
in reducing the leaching of arsenic from
waste discussed here. Some processes
reduced leaching rates by more than four
orders of magnitude relative to raw wastes
and some less effective processes. Several
commercial processes can achieve this level
of improvement. Inferences about whether a
given process was "successful" should be
site-specific, in the context of a total
F disposal system design. Leaching tests
with granular samples provide different
inferences from tests with monolithic
samples.
The costs of fixation may be offset by
the opportunity to dispose of adequately
fixed wastes at sites that may not require
sophisticated control and monitoring. For
example, the cost to transport and dispose
of untreated hazardous wastes at distant,
secure disposal sites can be greater than
the cost of fixation and disposal at a
nearby site with less stringent require-
ments.
ACKNOWLEDGEMENT
This work has been performed under
Contract No. 68-03-2503 from the U.S
Environmental Protection Agency. Project
Officer is Donald Sanning.
REFERENCES
1. Johnson, J.C., and R.L. Lancione. 1978.
Laboratory assessment of fixation and
encapsulation processes for arsenic-
laden wastes. In: Proc. of the Fourth
Annual Research Symposium. EPA-600/9/
78-016, U.S. Environmental Protection
Agency, Cincinnati, Ohio. pp. 326-341.
2. Johnson, J.C., R.L. Lancione, and D.E.
Sanning. 1978. Stabilization, testing
and disposal of arsenic-containing
wastes. In: Toxic and Hazardous Waste
Disposal, Vol. 2, (R.B. Pojasek, ed.)
Ann Arbor Science, pp. 201-215.
3. Johnson, J.C., and R.L. Lancione. 1979.
Stabilization, Testing, and Disposal of
Arsenic Containing Wastes. Draft final
report on Contract No. 68-03-2503, U.S.
Environmental Protection Agency,
Cincinnati, Ohio. 172 pp. + App.
4. U.S. Environmental Protection Agency.
1978. Hazardous waste-proposed guide-
lines and regulations and proposal on
identification and listing. Federal
Register, 43(243):58946-5902^
5. Mahloch, J.L. Leachability and physical
properties of chemically stabilized
hazardous wastes. In: Proc. of the
Second Annual Research Symposium.
EPA-600/9-76-015, U.S. Environmental
Protection Agency, Cincinnati, Ohio.
pp. 127-138.
6. Godbee, H.W. and D.S. Joy. 1974.
Assessment of the Loss of Radioactive
Isotopes from Waste Solids to the
Environment. Part I: Background and
Theory. ORNL-TM-4333, Oak Ridge
National Laboratory, 57 pp.
7. Columbo, P., and R.M. Neilson, Jr.
1977. Properties of Radioactive Wastes
and Waste Containers. BNL-NUREG-50664,
Quarterly Progress Report, Oct.-
Dec. 1976, Brookhaven National
Laboratory.
8. U.S. Environmental Protection Agency.
1979. Draft Environmental Impact
Statement for Subtitle C, Resource
Conservation and Recovery Act of 1976.
9. McMahan, J.R. et al. 1976. Hazardous
Waste Treatment and Disposal in the
Pharmaceutical Industry. SW-508, U.S.
Environmental Protection Agency,
Washington, DC, 188 pp.
10. Fields, T., Jr., and A.W. Lindsey.
Landfill Disposal of Hazardous Wastes:
A Review of the Literature and Known
Approaches. EPA/530/SW-165, U.S.
Environmental Protection Agency,
Washington, DC, 1975. 35 pp.
186
-------
7 ,
J r ^^l'
FIELD INVESTIGATION OF CONTAMINANT LOSS FROM
CHEMICALLY STABILIZED SLUDGES
Larry W. Jones, Philip G. Malone, and Tommy E. Myers
U. S. Army Engineer~"Waterways "Experiment Station
Vicksburg, Mississippi 39180
ABSTRACT
The movement of contaminants from treated industrial sludges placed at four disposal
sites was examined. Test borings were made through the treated sludges into the soil
beneath and into surrounding, uncontaminated soils. Measurements of selected chemical
constituents in nitric acid digests from shallow soil samples collected immediately below
the treated sludges and from soil samples collected at comparable elevations in surround-
ing soils were compared using the Mann-Whitney U Test. Soils beneath treated auto
assembly wastes disposed at two locations showed elevated levels of manganese, sodium, and
selenium. Soils beneath treated electroplating waste showed elevated levels of iron,
sodium, mercury, and nickel. At one location where treated oil refinery waste was dis-
posed no significant soil contamination was detected. Even in contaminated soils the
levels of constituents determined were well wi'thin the ranges reported for natural soils.
INTRODUCTION
This report is an extension of
earlier work on the effects of landfilled
treated or solidified wastes or surround-
ing groundwater and soil. An earlier
report covered the investigation of
groundwater chemistry and the physical
properties of the soils at treated indus-
trial waste disposal sites. The current
investigation examines effects of treated
wastes on the chemical composition of
surrounding soils.
In the earlier study groundwater
analyses from four disposal sites where
treated industrial wastes had been placed
showed little contamination that could be
related to the disposed wastes. Waste
containment could be attributed to either
the treatment process or to soil attenua-
tion. The goal of the present investiga-
tion is to determine if increased levels
of potential contaminants from the treated
wastes are present in the soil under the
disposal sites. The presence of elevated
levels of contaminants in the soils is
attributed to release of these materials
from the solidified/stabilized wastes.
Thirty-five million metric tons
(roughly 77 billion pounds) of hazardous
wastes must be disposed in the country
each year. In many cases reuse or incin-
eration is not appropriate and the mate-
rial must be landfilled. Solidification/
stabilization may greatly reduce the po-
tential hazard associated with
landfilling.
In the present investigation, four
sites where treated industrial sludges
containing electroplating, metal finish-
ing or refinery wastes had been placed
are examined. A summary of the major
characteristics of each site is given in
Table 1. All of the sites are located
in humid areas where the production of
leachate from the waste is probable.
None of the landfills are lined or had
any secondary containment facilities.
All of the sites contain treated sludges
that have appreciable metal content
(Table 2). The treatment process used
involved the addition of cementitious
materials to the sludges to produce a
soil-like material. All four sludges
were treated by the same processor.
Soil samples were obtained from
under the disposal sites and from simi-
lar elevations surrounding the site.
187
-------
TABLE 1. SUMMARY OF CHARACTERISTICS OF FOUR SITES SELECTED
Characteristic
Geographic area within the U. S.
General geologic setting
Mean annual rainfall
Mean annual air temperature
Nature of waste
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
Average thickness of cover
Dates of emplacement of fixed
sludge
Type of operation
Site W
Central
Glacial Drift
102 cm
12° C
Paint, putty
B, Cr, Fe, Pb,
Mn, Ni, Zn
None
1.22-3. 05m
(avg. 2.14m)
Sandy clay
3. 05-8. 60m
(avg. 5.60m)
l.lxlO"?cm/sec
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
0.0
1973
Diked fill
Site Y
North Central
Pleistocene -
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
0.0
1974*
Fill
Site Z
South Central
Deltaic -
Fluvial
Deposits
117 cm
21° C
Refinery sludge
Pb, Mn, and
phenol
None
1.83-3. 20m
(avg. 3.79m)
Clay
1.04-6. 63m
(avg. 3.79m)
4.2xlO~ cm/sec
0.5m
1974
Diked fill and
cover
* Fixed waste was placed on the ground in April and May 1974; but, the major portion of the fixed
material was removed to a landfill in Jaunary 1975 when the area was regraded.
The soil samples were digested in con-
centrated nitric acid and the digests
were analyzed for selected metals. A
nonparametric statistical test, the Mann-
Whitney U test ' (one-tailed, level of
significance = 0.10) was used to deter-
mine if the soils directly under the dis-
posal site were significantly higher in
metal content than the surrounding soils.
MATERIALS AND METHODS
Sampling Procedures
Eight to ten borings were put down
to the water table at each site. The
locations of the borings at each dis-
posal area are given in Figures 1-4.
All boring was done with a truck-
mounted, 16.8-cm diam hollow stem auger.
Soil samples were obtained by drilling
to the selected depth with a central
plug in place in the auger. To obtain a
sample the central plug was withdrawn
from the auger and a thin-walled tube
sampler (Hvorslev sampler) or a split-
spoon sampler was lowered into the
hollow-stemmed auger and driven or pushed
into the soil below the auger cutting
tip. This technique is a standard
1 2
system for foundation testing. ' Care
was taken to clean the samplers before
use so as to avoid contamination of
the soil samples. Soil samples were
extruded from the sampler on the surface,
TABLE 2. CONCENTRATIONS OF MAJOR
CHEMICAL CONTAMINANTS IN THE
FIXED SLUDGE AT EACH SITE*
Constituent
B
Cd
Cr
Cu
Fe
Pb
Mn
Nl
Se
Na
Zn
Hg
Site W
-------
Figure 1. Location of borings at site W.
Arrows indicate most probable direction
of groundwater movement
Figure 2. Location of borings at site X.
Arrows indicate most probable direction
of groundwater movement
trimmed to remove any potential cross
contamination, and placed in cleaned
plastic jars. The jars were packed in
ice and shipped to the laboratory for
digestion and analysis.
Samples were obtained at 1 to 2 m
intervals down each boring to a point
below the water table. Particular care
was taken to obtain samples immediately
below the sludge/soil interface and at
the water table. Splits of the soil
samples were used for physical testing,
chemical analysis, and additional geo-
chemical studies.
Analytical Procedures
Approximately 50 g of moist soil
was taken from each sample bottle and
weighed into a clean, tared, 250-ml
Figure 3. Location of borings at site Y.
Arrows indicate most probable direction
of groundwater movement
Figure 4. Location of borings at site Z.
Arrows indicate most probable direction
of groundwater movement
f luorocarbon beaker and 60 ml of 8N
reagent-grade nitric acid was added.
The soil-acid suspension was heated to
95 °C for 45 min and stirred every
15 min. After cooling to room tempera-
ture, the suspension was filtered
through a 0.45-micron membrane filter.
The digested soil was washed in the fil-
ter three times with 20-ml portions of
8N nitric acid. The filtrate was quan-
titatively transferred to a 250-ml volu-
metric flask and brought up to volume
with 8 N nitric acid. The acid digests
were stored in clean polyethylene
bottles. The procedures for the chem-
ical analyses are given in Table 3.
The nitric acid digestion procedure
T
189
-------
TABLE 3. TECHNIQUES USED IN THE
ANALYSIS OF NITRIC ACID DIGESTS
Procedures and/or instrumentation*
Lowest reporting
concentration
(ppm)
Determined with Perkin-Elmer Heated Graphite
Atomizer Atomic Absorption Unit
Plasma Emission Spectrophotometer Model II
Determined with Perkin-Elmer Heated Graphite
Emission Spectrophotometer Model II
Deti
Cd
Cr
Cu
Kg
Graphite Atomizer Atomic Absorption Unit
Same as above
Same as above
Same as above
Determined wi!.h a Nisseisangyo Zetman Shift
Atomic Absorption Spi'ctrophotometer
Determined with a Perkin-Llmer Heated
0.02
0.005
0.0003
0.003
0.003
0.0002
0.005
0.002
0.005
O.OU
takes into solution all metals present
as carbonates or sulfides or which are
adsorbed to clay minerals, iron oxide or
insoluble organic materials. Elements
in non-clay silicate lattices are not
brought into solution.
A separate aliquot of moist soil
was taken to determine moisture content
of the soils. Moisture contents were
used to correct the chemical analyses so
that soil acid digests could be expressed
in milligrams per kilogram dry weight of
soil.
RESULTS AND DISCUSSION
Results of the chemical analyses of
the nitric acid digests of the soil
samples taken at various depths in the
bore holes are given in Tables 4-11 for
all four sites. At sites W, X, and Y,
borings 1 and 2 are experimental borings
through the treated sludge itself
(Tables 4, 6, 8). At site Z, the experi-
mental borings are 1, 2, and 3
(Table 10). The remaining borings at
each site are control holes which were
drilled in the area surrounding the
sludge disposal site (Tables 5, 7, 9,
11). The samples from each boring are
numbered consecutively down the bore
hole - for example, "2C7" represents a
sample from boring number 2, chemical
testing split (C), seventh sample down
the boring from the surface. The eleva-
tion above mean sea level, depth below
the sludge/soil interface (negative
numbers being samples from the treated
sludge or cover material), and the
height of the sample above or below the
watertable are indicated in the tables
for each sample. At site W, the major
materials disposed were sludges produced
in an automobile assembly plant. As can
be seen in samples 1C1, 1C2, 2C1, and
2C2 (Table 4), the sludges differ from
surrounding soils in having elevated
concentrations of all metals determined.
The major potential contaminants are
boron, chromium, iron, lead, manganese,
nickel, and zinc; also, sodium and
selenium. Elevated levels of most con-
taminants were only evident in the
samples from the first meter below the
sludge/soil interface. No soils samples
recovered from borings more than 1 m
below the sludge/soil interface had
metals levels above that observed in the
surrounding uncontaminated soils except
in the case of selenium. Selenium levels
were higher in deep (greater than 1 m)
soils samples in both experimental
borings at site W.
The Mann-Whitney U Test was used to
detect differences in metal contents
between soils collected immediately
below the sludge disposal site and those
taken from similar elevations outside
the disposal area. Significantly ele-
vated levels were found for manganese,
sodium, and selenium in samples from the
sludge disposal area (Table 12).
The waste disposed at site X was
treated electroplating waste. The major
potential pollutants contained in this
sludge were cadmium, chromium, copper,
sodium, and zinc; other constituents
present in the sludge at levels in
excess of that found in the surrounding
soils were boron, mercury, nickel, lead,
and selenium. Several metals (iron,
potassium, and manganese) were much less
prevalent in the treated sludge than in
the local soils. No soils samples were
taken below 1-m depth under the sludge/
soil interface. Significantly elevated
levels of iron, sodium, mercury, and
nickel, were found in the digest of the
190
-------
TABLE 4. ANALYSES OF NITRIC ACID DIGESTS OF SOIL
SAMPLES FROM EXPERIMENTAL BORINGS AT SITE W
Boring and
sample
Elevation (m)
Depth below
sludge/soil
interface (m)
Ht . above water
table (m)
Cone, (mg/kg
dry wt.)
Fe
K
Mn
Ha
B
Be
Cd
Cr
Cu
Hg
Ni
Pb
Se
Zn
1C1
67709.
211.
1063.
361.
50.
0.
6.
367.
40.
0.
95.
66.
0.
1630.
41
.66
65
,76
10
,56
,36
,12
,71
,09
,88
,90
.80
.64
1C2
38777.90
1080.86
779.20
601.98
30.27
0.46
4.39
195.69
41.73
0.08
68.56
17.17
0.86
1184.26
1C3
255.87
0.07
2.49
30192.42
768.45
893.51
317.91
9.40
0.83
3.33
21.41
29.63
0.02
30.28
14.06
0.49
72.24
1C4
254.
1.
1.
13621,
468
336.
107
8.
0.
0
9
16.
0
23,
,94
,00
.56
.28
.35
.76
.29
.99
.38
.30
.94
.51
.01
.31
4.62
0
54
.77
.11
1C5
253.
2.
0.
11407.
S5°.
305.
Ill,
9.
0.
0,
9.
13,
0,
20,
4
0,
47.
.73
.21
.35
.92
•*'
.40
.32
,74
,36
.26
.44
,25
.01
.50
.55
.82
.01
1C6
252.
3.
-0.
13340.
580.
328.
113.
10.
0.
0.
9.
15.
0.
19.
6.
0.
55.
79
15
59
41
90
41
14
80
39
89
68
70
01
92
.71
73
,30
1C7
250.54
5.40
-2.84
14054.88
721.00
384.19
114.08
10.12
0.42
0.13
10.37
15.61
0.01
20.98
7.26
0.86
44.63
(continued)
TABLE 4. (Concluded)
Boring and
sample
Elevation (m)
Depth below
sludge/soil
interface (m)
Ht. above water
table (m)
Cone, (mg/kg
dry wt.)
Fe
K
Mn
Na
B
Be
Cd
Cr
Cu
Hg
Ni
Pb
Se
Zn
2C1
26272.
266.
626.
330.
36.
0.
2.
98.
16.
0.
54,
10.
1.
1048,
.88
,41
,08
.25
.82
.44
.84
.66
.82
,02
,13
.99
.12
.46
2C2
17532.36
m".14
479.58
1439.87
11.88
0.45
2.25
30.99
20.35
0.07
27.40
18.09
0.84
229.73
2C3
256.
0.
4.
15162.
483.
1084.
357.
5.
0.
1.
12.
11.
.51
,08
,31
.99
.10
,29
,92
,31
,54
,55
92
,11
0.03
13.
14.
,92
06
0.46
48.
,22
2C4
255.
0.
3.
20867.
773.
439.
206.
10.
0.
2.
15.
24.
0.
32.
9.
0.
79.
.60
,99
,22
,45
.94
.32
.91
56
,67
,43
99
30
.02
,22
64
.67
,24
2C5
254,
2,
2.
13576,
735.
319.
180.
11,
0,
1
11.
17.
0,
24,
7.
1.
60.
.38
.21
.00
.87
.41
.58
.08
.02
.37
.73
,65
,06
,02
.50
,50
,77
,06
2C6
253.45
3.14
1.07
8662.46
435. M
239.66
102.61
6.96
0.29
1.16
9.31
15.34
0.01
17.95
6.73
0.18
52.65
2C7
251.40
5.55
-1.34
9075.01
ni.?8
293.43
119.60
7.14
0.22
1.13
12.70
11.54
0.00
13.01
5.94
0.21
38.85
191
-------
TABLE 5. ANALYSES OF NITRIC ACID DIGESTS OF SOIL SAMPLES
FROM CONTROL BORINGS AT SITE W
Boring and
sample
Elevation (m)
Ht. above water
table (m)
Cone, (mg/kg
dry wt.)
Fe
K
Mn
Na
B
Be
Cd
Cr
Cu
Hg
Ni
Pb
Se
Zn
3C1
256,
6.
17024.
634.
445.
110.
10.
0.
2.
40.
27.
0.
21.
11.
0.
126.
.84
.16
76
06
.30
.71
23
23
18
15
59
03
17
74
25
64
3C2
255,
5.
16985.
61?.
403.
100.
9.
0.
2.
20.
18.
0.
24.
10.
0.
57.
.93
.25
11
73
24
.59
68
49
04
56
12
01
24
44
24
32
3C3
254.71
4.03
13627.81
546.98
378.62
111.59
10.34
0.34
1.79
17.17
14.27
0.01
19.34
10.98
0.25
53.42
3C4
253.81
3.13
12422.47
645.28
291.06
120.52
10.75
0.29
1.59
17.68
13.16
0.04
17.93
7.42
0.23
50.16
3C5
251,
0.
14003.
6«7.
344.
111.
10.
0.
1.
16.
12.
0.
15.
6.
0.
43.
.20
,52
01
69
.21
70
00
31
43
,31
42
02
20
67
23
68
3C6
249
-1
15313,
897,
420
124,
10,
0,
1.
17,
14.
0.
18,
7.
0.
90
.37
.31
.91
.67
.64
.55
.15
.32
.70
.68
.21
.01
,49
.38
,23
.41
4C6
250.75
-1.02
8674.08
380.32
258.93
100.51
6.39
0.08
1.08
10.02
8.51
0.01
10.27
5.34
0.23
35.59
5C1
256.68
5.63
14464.37
638.49
364.36
141.87
12.80
0.32
2.01
17.68
20.77
0.02
18.68
21.33
0.23
82.43
continued
TABLE 5. (Concluded)
Boring and
sample
Elevation (m)
Ht . above water
table (m)
Cone, (mg/kg
dry wt. )
Fe
K
Mn
Na
B
Be
Cd
Cr
Cu
Hg
Ni
Pb
Se
Zn
5C6
250.60
-0.45
12693.60
584.46
393.56
126.09
11.25
0.23
1.40
14.54
12.01
0.01
15.80
6.82
0.14
36.98
6C1
257.07
2.33
18631.07
667.09
329.14
185.36
28.58
0.77
3.00
23.82
120.67
0.07
23.98
22.20
0.16
158.32
6C5
252.68
-2.06
9209.18
432. ii
299.06
143.85
10.61
0.27
1.18
11.99
9.47
0.01
12.75
14.03
0.18
35.70
7C1
255
2
13957
377
358,
89,
6,
0,
1,
12.
8.
0,
9,
6.
0.
38.
.97
.82
.05
.5''
.99
.13
.76
.36
.50
.17
.21
.02
.28
.73
.08
.51
7C2
255.21
2.06
23965.56
608.69
522.73
79.02
11.98
0.73
2.55
19.67
21.71
0.02
30.61
10.42
0.25
72.59
7C3
253
.99
0.84
13565,
575,
390.
111.
13.
.70
,11
.73
77
.18
0.29
1.
14.
13.
0.
19.
8.
0.
53.
73
86
66
.01
06
60
47
02
7C4
253.07
-0.07
8430.77
373.00
235.49
121.34
11.14
0.19
1.20
12.81
45.21
0.01
18.29
8.99
0.49
83.77
7C5
251.55
-1.60
9744.03
263.86
199.62
94.87
7.64
0.06
1.10
26.04
10.48
0.01
8.21
3.96
0.68
29.15
8C6
250.55
1.33
13705.00
621.41
301.32
109.83
10.70
0.22
1.36
14.22
43.37
0.01
14.97
6.57
0.33
43.37
192
-------
TABLE 6. ANALYSES OF NITRIC ACID DIGESTS OF SOIL
SAMPLES FROM EXPERIMENTAL BORINGS AT SITE X
Boring and
sample
Elevation (m)
Depth below
sludge/soil
interface (m)
Ht. above water
table (m)
Cone, (mg/kg
dry wt.)
Fe
K
Mn
Na
B
Be
Cd
Cr
Cu
Hg
Hi
Pb
Se
Zn
1C1
3494.40
50.74
90.47
952.58
30.73
0.69
109.14
1493.45
1531.79
0.26
42.65
18.19
0.45
413.58
1C2
285.32
0.09
2.65
22579.62
837.14
498.00
664.89
13.12
0.49
2.44
14.14
28.20
0.04
26.28
11.60
0.22
67.47
2C1
4579.29
125.63
128.66
1610.86
40.12
0.90
102.32
1783.04
1772.95
0.17
56.28
21.28
2.12
481.23
2C2
285.33
0.07
2.40
23090.18
950.77
600.54
883.97
15.46
0.70
2.76
18.08
26.56
0.03
21.36
11.89
0.27
84.36
TABLE 7. ANALYSES OF NITRIC ACID DIGESTS OF SOIL
SAMPLES FROM CONTROL BORINGS AT SITE X
Boring and
sample
Elevation (m)
Ht. above water
table (m)
Cone, (mg/kg
dry wt.)
Fe
K
Mn
Na
B
Be
Cd
Cr
Cu
Hg
Ni
Pb
Se
Zn
3C1
285.
1.
22852.
912.
619.
85.
17.
0
3.
42
25.
0
24.
12,
1
72
.35
.97
.51
,40
.06
.27
.14
.78
.34
.42
.84
.02
.43
.79
.02
.48
3C2
284.
1.
35987.
719.
673.
44.
17.
0.
3.
36.
32.
0.
45
07
73
75
07
98
31
,98
14
,28
,17
.03
78.72
12.
0,
30.
,73
,28
,30
3C3
282.
-1.
11.
565.
19
19
.32
90
238.08
154.56
21.
0.
1.
12.
12.
0,
16.
8.
0,
45.
,19
.36
,48
,78
,09
,02
.64
,72
.45
,87
4C1
285.
2.
22466.
572.
599.
95,
9.
0.
7.
76,
93.
45
,08
.78
,06
,77
,01
,06
.46
.99
.85
.43
0.03
20.
13,
1.
77.
.64
.86
.19
.89
5C1
285.
2.
17202.
867.
933.
25.
8.
0.
83
45
,49
,44
.33
,62
,15
.41
2.10
17.
38.
0,
14,
12,
1,
59,
,52
,25
.02
.96
,81
,17
,02
5C2
284.91
1.53
18357.78
637.19
453.61
151.60
27.36
0.34
1.92
15.10
41.10
0.03
24.10
11.84
0.99
98.54
(continued)
193
-------
TABLE 7. (Concluded)
Boring and
sample
Elevation (m)
Ht. above water
table (m)
Cone, (mg/kg
dry wt.)
Fe
K
Mn
Na
B
Be
Cd
Cr
Cu
Hg
Ni
Pb
Se
Zn
6C1
285.56
2.18
14961.01
473.20
578.27
59.68
7.92
0.36
1.50
10.46
25.97
0.02
14.83
10.09
1.14
61.10
6C2
284.66
1.28
40577.16
907.77
814.27
47.20
15.16
1.05
3.33
19.74
34.86
0.02
46.16
18.16
1.04
125.27
7C1
285.52
2.22
13444.79
388.71
364.90
55.79
4.52
0.36
BDL
7.79
35.21
0.01
10.75
10.06
0.48
45.18
7C2
284.59
1.29
23916.70
347.03
487.68
34.70
7.87
0.59
0.03
17.58
20.54
0.03
19.27
12.19
BDL
72.22
8C1
285.55
2.19
20850.39
562.01
754.38
37.91
7.77
0.46
0.44
25.45
22.08
0.02
16.35
13.27
0.52
56.58
9C1
285 . 19
2.11
16176.78
385.75
723.48
105.77
5.93
0.44
5.44
95.95
80.08
0.02
13.73
11.55
0.59
61.24
BDL - Below detection limits.
TABLE 8. ANALYSES OF NITRIC ACID DIGESTS OF SOIL
SAMPLES FROM EXPERIMENTAL BORINGS AT SITE Y
Boring and
sample
Elevation (m)
Depth below
sludge/soil
interface (m)**
Ht. above water
table (m)
Cone, (mg/kg
dry wt. )
Fe
K
Mn
Na
B
Be
Cd
Cr
Cu
Hg
Ni
Pb
Se
Zn
1C1
186.55
1.35
15877.04
331.73
650.13
1800.58
9.05
0.46
0.03
45.28
17.12
0.02
36.68
70.96
0.67
353.02
1C2
185.67
0.47
9403.42
110.63
134.57
262.74
3.60
0.18
BDL
11.11
9.50
0.02
9.36
5.43
BDL
55.96
1C3
184.91
-0.29
20457.00
911.11
299.09
97.13
10.66
0.41
0.04
15.51
20.89
BDL
28.58
8.60
BDL
60.60
2C1
186.85
9.68
24215.85
1114.84
494.34
2147.44
10.51
0.71
0.26
13.66
18.09
0.02
24.08
12.79
0.59
63.24
2C2
185.92
8.75
21591.43
369.45
302.25
369.45
6.52
0.33
BDL
11.85
13.63
0.02
20.68
4.80
BDL
65.54
2C3
185.19
8.02
21366.44
972.08
255.02
97.21
10.78
0.27
0.19
14.48
19.34
0.02
25.26
7.70
BDL
65.16
2C4
183.79
6.62
22986.16
961.24
327.63
109.50
10.62
0.27
0.49
10.91
21.56
0.01
25.62
8.36
BDL
69.54
2C5
181.21
4.04
19866.21
562.88
270.37
71.74
8.02
0.42
BDL
5.99
20.20
0.01
20.69
8.83
BDL
78.58
2C6
179.21
2.04
17554.88
566.45
176.54
75.53
7.55
0.45
BDL
5.81
18.17
0.01
20.06
9.18
BDL
68.08
BDL - Below detection limits.
** = Sludge removed at this site.
194
-------
TABLE 9. ANALYSES OF NITRIC ACID DIGESTS OF SOIL
SAMPLES FROM CONTROL BORINGS AT SITE Y
Boring and
sample
Elevation (m)
Ht. above water
table (m)
Cone, (mg/kg
dry wt.)
Fe
K
Mn
Na
B
Be
Cd
Cr
Cu
Hg
Ni
Pb
Se
Zn
3C1
186.94
9.82
24269.17
763.01
561.66
86.41
11.49
0.65
0.28
12.09
21.79
0.02
31.12
10.11
0.10
65.27
3C2
186.02
8.90
36164.67
562.86
530.03
43.43
13.03
1.20
1.36
14.76
30.05
0.02
34.52
9.75
BDL
80.57
3C3
185,
8,
27242.
1375,
.26
.14
,40
,61
210.36
115.
14.
0.
0.
17.
26.
0.
34.
8.
BDL
68.
,98
,20
.87
,73
,40
,25
,08
,39
,99
78
3C4
184
7,
20723.
731,
225,
93.
.13
.01
.46
.66
.82
.04
10.40
0,
0.
10,
22.
0,
23.
8,
BDL
74.
.52
.22
,19
.16
.20
,98
.46
,35
3C5
181.43
4.31
17832.90
422.36
335.99
90.10
7.48
0.10
1.48
7.47
21.92
0.12
19.46
8.45
BDL
60.72
4C1
186.92
9.53
20436.30
519.28
409.55
229.49
7.87
0.59
1.76
11.47
17.14
0.02
24.65
9.21
0.24
55.86
4C2
185,
8,
25864.
318,
283.
153,
8.
0,
2,
13.
18.
0.
25.
7.
BDL
61,
.99
.60
,15
.48
,75
.45
.59
,82
,08
70
.10
.21
.31
,72
.96
4C3
185.23
7.84
19465.27
489.08
346.25
89.99
9.39
0.44
1.84
10.37
22.16
0.02
27.12
8.80
BDL
63.09
(continued)
TABLE 9. (Continued)
Boring and
sample
Elevation (m)
Ht. above water
table (m)
Cone, (mg/kg
dry wt . )
Fe
K
Mn
Na
B
Be
Cd
Cr
Cu
Hg
Ni
Pb
Se
Zn
4C4
183.87
6.48
20435.79
937.01
280.19
103.52
12.05
0.53
1.99
13.21
23.88
0.02
35.72
10.71
BDL
63.90
4C5
181.43
4.04
17808.00
926.38
325.56
98.03
11.42
0.48
1.65
11.87
25.60
0.01
25.03
7.20
BDL
52.61
4C6
179.23
1.84
17740.80
858.58
242.85
96.77
12.33
0.37
1.80
8.66
19.12
BDL
25.07
9.49
BDL
74.28
4C7
177.16
-0.23
19266.13
628.68
298.27
96.33
9.72
0.30
1.54
8.70
18.13
0.02
22.08
8.45
BDL
77.49
5C1
186.64
9.74
19305.95
807.70
477.90
111.67
12.49
0.62
2.00
15.62
54.95
0.05
23.88
18.93
0.18
149.53
5C2
185.73
8.83
13473.95
294.92
481.46
63.13
5.54
0.47
1.32
9.42
7.28
0.10
12.18
5.65
BDL
46.83
5C3
184.97
8.07
13749.88
344.26
709.88
81.48
6.19
0.55
1.27
0.97
11.06
0.06
13.68
7.13
BDL
46.75
5C4
183.58
6.68
26527.58
700.06
445.77
72.65
11.23
0.59
2.57
10.79
32.65
0.02
37.02
12.11
BDL
132.09
(continued)
195
-------
TABLE 9. (Continued)
Boring and
sample
Elevation (m)
Ht. above water
table (m)
Cone, (mg/kg
dry wt.)
Fe
K
Mn
Na
B
Be
Cd
Cr
Cu
Hg
Ni
Pb
Se
Zn
5C5
181.14
4.24
24518.92
633.10
374.64
67.70
8.69
0.55
1.98
9.42
25.58
0.18
31.86
10.98
BDL
85.63
5C6
179.01
2.11
16899.77
687.07
228.08
44.33
7.76
0.46
1.52
0.02
18.89
0.42
22.74
9.23
BDL
63.72
5C7
176.
-0.
17360.
434.
309.
,86
,04
29
01
08
114.32
14,
0.
2.
9,
26.
0,
30.
8.
.82
23
,94
6C6
179.
2.
13259.
272.
128.
53.
3.
0.
1.
09
,14
11
60
86
78
18
,25
,46
,88 6.78
.21
,22
,09
.74
BDL
134.
.44
16.
0.
18.
6.
,47
,52
,39
.49
BDL
59.
,34
7C5
177.06
-2.01
21145.56
503.15
268.04
68.91
8.18
0.35
5.28
10.52
22.75
0.02
31.08
9.44
0.15
103.70
8C1
186.
9.
22177.
439.
521.
134.
3,
0.
1.
8.
9.
0.
15.
9.
0.
52.
.55
.01
,23
74
,54
.20
.24
,36
,37
.96
.33
.03
.35
.52
.55
.87
8C2
185.62
8.08
10968.34
293.14
179.51
119.39
2.84
0.30
1.73
7.50
9.61
0.02
13.33
5.82
BDL
55.47
8C3
183.45
5.91
14555.71
276.47
347.82
33.87
3.91
0.38
1.84
5.83
17.58
0.01
17.05
6.41
BDL
63.67
(continued)
TABLE 9. (Concluded)
Boring and
sample
Elevation (m)
Ht. above water
table (m)
Cone, (mg/kg
dry wt . )
Fe
K
Mn
Na
B
Be
Cd
Cr
Cu
Hg
Ni
Pb
Se
Zn
8C4
181
3
17877
752
262
64
9
1
2
10
21
0
28
9
0
68
.04
.05
.82
.75
.46
.92
.24
.13
.29
.20
.17
.02
.35
.41
.12
.92
8C5
178.
1.
22857.
48.
343.
53.
10.
0.
1,
10.
28,
.91
.37
,84
.74
.75
.40
,68
.43
.90
.62
,95
0.02
34.
11.
.78
.24
BDL
87
.26
8C6
177.06
-0.48
17456.36
468.10
84.77
45.59
5.87
0.37
1.38
6.70
20.12
0.02
23.83
7.78
BDL
80.67
9C1
186.66
9.82
21687.46
842.40
436.38
673.03
9.23
0.72
1.78
14.20
18.19
0.03
25.84
8.96
BDL
62.96
9C2
185.
8.
11310.
225.
87.
86,
3.
0,
0.
9.
7.
0.
9.
2,
.74
90
70
.24
.89
,78
.62
,36
.60
,79
.98
.02
.97
,85
BDL
39,
.44
9C3
185.
8.
21324.
1028.
471.
96.
11.
0.
1.
14.
19.
0.
26.
00
61
,99
17
19
15
04
.64
.84
,32
32
,02
,59
7.62
0,
59.
,19
,64
9C4
183.63
6.79
20166.66
1100.99
394.55
109.01
11.12
2.77
2.05
12.69
19.62
0.01
28.05
8.72
0.05
66.44
BDL - Below detection limits.
196
-------
TABLE 10. ANALYSES OF NITRIC ACID DIGESTS OF SOIL
SAMPLES FROM EXPERIMENTAL BORINGS AT SITE Z
Boring and
sample 1C1
1C2 1C3 1C4 1C5 1C6 1C? 1C8 2C1 2C2
Elevation (m)
Depth below
sludge/soil
interface (m)
Ht. above water
table (m)
Cone, (mg/kg
dry wt.)
Mn
Na
Be
Cd
Cr
Cu
Kg
Ni
Pb
Se
Zn
5.61 4.09 2.46 1.54 0.32 -0.59 -2.12 -3.64 5.17 3.64
-3.05 -1.53 0.10 1.02 2.24 3.15 4.68 6.20 -3.05 -1.52
5.15 3.63 2.00 1.08 -0.14 -1.05 -2.58 -4.10 4.03 2.50
8230.47 15804.15 24469.34 24149.29 18865.61 16047.70 13435.52 13105.36 7559.94 15823.24
60.80
163.28
1.93
0.48
1.05
12.69
4.98
0.09
4.96
92.43
0.13
19.80
267.84
1007.77
5.00
1.30
2.45
742.93
139.57
10.97
21.74
4751.46
0.54
442.28
614.57
713.23
6.96
1.91
2.84
24.12
15.17
0.01
25.40
36.18
0.86
54.04
291.68
1138.33
7.60
1.92
2.89
22.38
16.76
BDL
25.27
29.12
0.66
68.16
283.44
987.39
6.45
1.66
2.39
18.80
14.27
BDL
20.91
29.42
0.30
57.04
1381.88
755.13
6.75
1.49
2.42
17.41
12.93
BDL
30.47
33.12
0.87
51.69
78.39
597.53
4.22
1.37
1.74
14.52
8.01
0.01
14.30
14.28
0.52
38.96
23.08
241.28
4.24
0.95
1.73
9.22
7.95
BDL
11.87
15.93
0.34
27.28
89.77
162.78
2.11
0.39
1.08
20.58
8.02
0.24
5.60
15.73
0.50
33.20
292.27
448.49
3.81
0.99
1.99
212.58
48.68
2.29
16.81
2328.08
0.78
149.14
TABLE 10. (Concluded)
Boring and
sample
2C3
2C4
3C1
3C2
3C5
3C6
Elevation (m)
Depth below
sludge/soil
interface (m)
Ht. above water
table (m)
Cone, (mg/kg
dry wt.)
Mn
Na
Be
Cd
Cr
Cu
Hg
Ni
Pb
Se
Zn
2.01 1.10 -0.12 -1.04 5.14 3.61 1.98 1.07 -0.15 -1.07
0.11 1.02 2.24 3.16 -3.05 -1.52 0.11 1.02 2.24 3.16
0.87 -0.04 -1.26 -2.18 6.10 4.57 2.94 2.03 0.18 -0.11
21350.54 20433.11 21178.35 19583.95 10935.74 18975.67 14570.30 20138.28 18907.56 15161.72
364.20
860.78
5.51
1.13
2.42
20.12
16.71
0.01
23.94
51.44
0.40
63.45
211.96
966.28
6.10
1.17
2.22
20.84
13.94
BDL
21.75
26.02
0.64
59.94
195.55
748.00
5.98
1.06
2.25
19.67
13.70
BDL
20.44
25.01
0.86
53.69
76.26
681.84
4.94
1.25
1.97
19.79
10.33
0.01
18.58
20.98
0.84
48.76
337.92
383.65
3.31
0.75
1.50
75.68
23.57
0.73
13.43
537.78
0.75
82.44
266.86
2554.94
10.56
1.14
3.69
1926.84
200.16
5.13
35.68
2806.02
1.16
975.22
103.07
1240.95
3.38
1.00
1.59
14.70
11.89
0.02
15.84
18.98
0.37
45.27
149.16
1232.54
5.50
1.08
2.21
16.90
13.89
CI
19.65
21.86
0.52
49.95
228.34
1033.98
5.22
1.05
2.24
16.19
12.88
CI
18.89
26.90
0.48
50.49
120.55
743.47
3.69
1.06
1.78
13.18
0.88
BDL
17.24
19.13
0.55
34.96
197
-------
TABLE 11. ANALYSES OF NITRIC ACID DIGESTS OF SOIL
SAMPLES FROM CONTROL BORINGS AT SITE Z
Boring and
sample
4C1
4C3
4C4
4C5
5C1
5C2
5C3
Elevation (m)
Ht. above water
table (m)
Cone, (mg/kg
dry wt.)
1.99
0.65
1.03
-0.24
-0.14 -1.05
-1.48
-2.58
-2.39 -3.92
2.15
0.99
1.24
0.08
0.02
-1.14
Fe
Mn
Na
Be
Cd
Cr
Cu
Hg
Ni
Pb
Se
Zn
20880.57 18826.87 16938.52 13209.13 16241.00 20992.32 18581.89 20227.31
576.07
860.39
5.40
1.17
2.47
17.18
13.89
CI
23.28
34.63
1.27
54.42
182.58
938.50
5.18
1.10
2.08
14.98
12.68
CI
18.43
23.32
1.07
46.81
136.22
689.19
4,80
0.85
1.88
15.08
9.95
CI
14.41
22.63
0.96
38.52
131.17
720.58
3.43
1.16
1.57
12.32
7.01
BDL
17.20
19.54
0.93
33.41
272.77
536.58
3.33
0.88
1.88
11.87
9.87
BDL
18.56
23.64
0.94
44.66
491.73
679.61
4.02
1.06
2.40
16.61
15.72
CI
23.85
31.58
0.56
55.19
413.56
788.06
4.50
1.08
2.23
15.37
14.01
BDL
22.95
28.82
0.70
52.58
288.16
829.41
5.09
1.15
2.26
19.79
14.96
BDL
21.69
28.77
0.70
58.53
(continued)
TABLE 11. (Continued)
Boring and
sample 5C4 5C5 6C1 6C2 6C3
Elevation (m) -0.90 -2.42 2.06 1.14 -0.08
Ht. above water
table (m) -2.06 -3.58 0.93 0.01 -1.21
Cone, (mg/kg
dry wt.)
6C4 6C5 6C6
-0.99 -2.52 -4.04
-2.11 -3.65 -5.17
15093.02 12363.93 22073.16 20138.13 23213.19 21385.59 21547.15 26061.62
Mn
Ha
Be
Cd
Cr
Cu
Hg
Ni
Pb
Se
Zn
87.74
621.27
2.75
0.55
1.60
15.47
7.73
BDL
9.11
16.28
0.71
34.38
112.23
408.62
2.54
0.79
1.46
11.25
9.91
BDL
14.45
14.16
0.45
31.72
585.03
694.42
4.50
1.07
2.43
18.59
14.84
BDL
24.40
33.84
0.59
60.82
377.34
896.80
5.26
1.20
2.38
20.07
15.15
BDL
23.79
36.84
0.56
61.80
223.12
762.98
6.62
1.02
2.78
27.11
11.43
0.00
17.08
24.43
0.59
50.90
192.19
713.76
4.54
1.14
2.54
20.51
10.38
BDL
16.13
19.81
0.56
46.81
29.53
345.80
5.11
1.04
2.53
20.10
10.25
BDL
16.40
15.86
0.68
54.54
1040.64
301.34
6.03
1.15
3.13
19.84
13.57
0.01
21.16
37.96
0.74
54.00
(continued)
198
-------
TABLE 11. (Concluded)
Boring and
sample
Elevation (m)
Ht. above water
table (m)
Cone, (mg/kg
dry wt.)
7C1 7C2 7C3 7C4 7C5 8C5 9C5 10C2
1.98 1.07 -0.15 -1.06 -2.59 -2.81 -2.55 1.11
0.96 0.05 -1.17 -2.08 -3.61 1.03 6.67 2.71
Fe
26560.80 20633.29 17637.50 19919.02 21003.51 18402.23 22972.47 27836.84
Mn
Na
B
Be
Cd
Cr
Cu
Hg
Hi
Pb
Se
Zn
253.22
1182.38
6.87
1.48
3.15
29.83
15.23
BDL
22.45
26.23
0.55
70.23
176.46
988.23
4.99
1.04
2.60
20.30
12.94
BDL
18.17
22.31
0.83
53.82
269.64
626.64
4.61
0.83
2.25
16.22
9.33
BDL
14.40
24.46
0.80
40.26
182.28
606.46
5.31
2.19
2.59
18.61
10.94
BDL
18.21
25.56
0.84
47.37
379.50
404.74
4.95
1.00
2.77
19.14
12.98
BDL
20.98
20.06
0.84
45.41
497.11
832.02
2.88
0.85
2.54
13.28
8.16
BDL
12.63
29.61
0.79
35.39
222.27
544.08
4.82
1.00
3.14
14.35
8.99
BDL
15.05
32.32
0.70
38.39
283.19
1026.48
6.31
1.44
3.56
19.55
11.50
BDL
16.51
27.11
0.68
53.72
BDL - Below detection limits.
CI * Chemical interference.
TABLE 12. RESULTS OF MANN-WHITNEY
U TEST OF NITRIC ACID DIGESTS OF
SOIL SAMPLES DIRECTLY UNDER THE
DISPOSAL SITES AND AT COMPARABLE
ELEVATIONS OUTSIDE THE DISPOSAL
SITES
Parameter
Measured
Fe
K
Mn
Ha
B
Be
Cd
Cr
Cu
Hg
Ni
Pb
Se
Zn
Sice
W
NS
US
S
s
MS
NS
NS
NS
NS
NS
NS
NS
S
NS
Site
X
S
NS
NS
S
NS
NS
NS
NS
NS
S
S
NS
NS
NS
Site
Y
NS
NS
NS
S
NS
NS
NS
NS
NS
NS
NS
NS
NA
NS
Site
Z
NS
NS
NS
NS
NS
NS
NS
»S
NS
NA
NS
NS
NS
NS
S " Metal level significantly larger under the disposal site at a 0.10 level
of significance.
NS • Metal level not significantly larger under the disposal site at a 0.10
level of significance.
NA - Too feu sampli
.cally.
soil from less than 1 m under the
treated sludge when compared to sur-
rounding soils at similar elevations
(Table 12). Iron is not a prominent
constituent in the treated electroplat-
ing sludge, therefore, the elevated
levels of iron under the waste disposal
site are probably due to chance
variation.
The treated automobile assembly
plant waste (largely paint and putty)
deposited at site Y had been in place
for approximately 9 months before the
material was removed to a landfill and
the area was regraded. Major potential
pollutants from this treated sludge were
chromium, iron, lead, manganese, sodium,
and zinc. Only sodium was found at sig-
nificantly elevated levels In soils
samples taken near the surface from which
the treated sludge had been removed
(Table 12). However, high levels of
chromium, lead, and zinc were found in
samples from the upper level of one of
the experimental holes (1C1, Table 8).
The sodium levels in the soils samples
taken approximately 1 m deeper in both
experimental holes (samples 1C2 and 2C2)
were still slightly above the range of
sodium levels in the surrounding soils.
Sodium levels for soils samples taken
199
-------
from greater depths below the sludge/soil
interface were within the range of sodium
levels in the surrounding soil.
Site Z contains treated refinery
sludge which had been covered with 0.5 m
of clay. Three experimental borings were
drilled through the cover material and
the treated sludge to the watertable
(Table 10). The major pollutants at this
site are chromium, copper, mercury, lead,
and zinc. Slightly elevated levels of
boron, cadmium, and sodium were also
detected in experimental boring No. 3.
The top samples in each of the experi-
mental borings (1C1, 2C1, and 3C1)
reflect the composition of the covering
clay layer and bear little relation to
either the treated sludge sample or the
underlying soils. The second sample from
each boring (sample 1C2, 2C2, and 3C2)
contained a large portion of the treated
sludge as can be noted by the high con-
centrations of the potential pollutants.
However, the levels of all potential pol-
lutants in the third sample down each
experimental hole (samples 1C3, 2C3, and
3C3) are all within the range of these
constituents in the surrounding soils
even though these samples were taken
only 10 to 12 cm below the sludge/soil
interface. In only one case (lead in
sample 2C3) is any sample from an experi-
mental boring above the range observed
in surrounding soils. No significant
loss of pollutants to the soil under the
treated sludge disposal area was detected
when samples from below the disposal pit
were compared to surrounding soils at
similar elevations (Table 12).
The soils at all four sites have
concentrations for the constituents
tested which fall within the typical
concentration ranges for soils in a
natural state in the eastern United
States (Table 13). All of the soils
have concentrations near to or below the
average for such soils. Even in the
cases where significant increases in the
constituents under the treated sludge
were found, these elevated levels were
always well within typical ranges for
eastern soils. For instance, sodium,
which was found at levels of 357, 883,
and 2147 mg/kg dry weight of soil under
the sludges at sites W, X, and Y, re-
spectively, has an average concentration
in eastern soils of 2600 mg/kg—higher
than any "contaminated" sample found.
TABLE 13. CONCENTRATIONS OF SELECTED
CONSTITUENTS IN SOILS IN THE
EASTERN UNITED STATES
Fe
K
Mn
Na
B
Be
Cd
Cr
Cu
Hg
Ni
Pb
Se
Zn
Concer
verage
15,000**
7400
290
2600
32
0.6
<1
36
13
0.096
13
14
0.39
36
itration*
Range
100 - >100,000
50 - 370,000
<2 - 7,000
<200 - 15,000
<10 - 150
<1 - 7
<1 - 4
1 - 100
3 - 100
.01 - 3.4
<3 - 700
<7 - 300
<.l - 1.4
<5 - 400
Concentrations are in milligrams per kilogram dry wt of soil.
Soil analyses are for Eastern U. S. soils as given by Connor and
Shacklette.4
Three of these significantly contami-
nated soils samples taken at the sludge/
soil interface - selenium at site W,
iron and nickel at site Y - had maximum
value less than twice the average soil
concentration and well within typical
ranges. In the "worst case" noted, that
of manganese contamination, at site W,
the highest contaminant level was less
than four times the average, and was
still well within typical ranges for
manganese in eastern soils. Even in the
cases where significant contamination
was found, appreciable degradation of
the soil below that of typical eastern
soil could not be demonstrated.
CONCLUSIONS
The loss of contaminants from
disposal areas containing treated indus-
trial wastes to the soils below the
disposal area has been examined. Analy-
ses of nitric acid digest of soils
samples taken at various depths under
and immediately surrounding four dis-
posal areas has shown only a minor gain
of potential pollutants by underlying
soils. Although elevated levels of a
limited number of constituents which
could be related to the presence of the
treated wastes were detected in the
200
-------
first meter of soil under the sludge/soil
interface, in no case were these levels
higher than the typical range for these
elements in eastern United States soils.
Attenuation of pollutants from the
leaching medium by the underlying soil
does not seem to be a major factor in
maintaining high quality groundwater
under the studied sites.
At one site (site Y) the treated
sludge had been removed after being in
place for less than 9 months and the
site was subsequently regraded. The
presence of the disposal area was still
evident in significantly higher sodium
levels in all soils samples. Some of
the individual soils samples also had
metals concentrations outside the range
of concentrations found in surrounding
soils. In no case was the residual
amount of any measured metal beyond the
range encountered in natural soils for
the eastern United States.
The only site in which no signifi-
cant effect of the treated sludge was
measured (site Z) was unique in that it
had been covered with 0.5 m of clay.
Even though it has the highest mean
annual rainfall for the sites studied,
the clay cover may have been effective
in precluding percolation through the
treated sludge into the soil below since
no contamination of the soil was evident.
The results of this study corrobo-
rate and correlate well with our previous
studies at these four sites on the physi-
cal properties of the underlying soils
and on the chemistry of groundwater
samples under and down dip from the dis-
posal area. No effect of the disposal
areas on the physical properties of
underlying soils samples, when compared
to surrounding soils, was found. None
of the toxic constituents (with the ex-
ception of phenols at site Z) predicted
to be pollution problems on the basis of
their presence in the treated sludges
was found at significantly elevated
levels in the groundwater. No unusual
accumulations of toxic elements in the
soil under the disposal areas were found.
The absence of changes in the physi-
cal properties and the lack of major
changes in chemical composition in the
soils underlying the treated waste dis-
posal areas, and the absence of evidence
of major pollutants in the groundwater
under and down dip from the waste dis-
posal area may be attributed to the short
time available for the transport of the
material out of the treated sludges or
to the efficacy of the treatment process.
The treated waste had been in place for
only 3 to 4 years at the sites studied.
Further work is necessary in order to
determine the long-term efficacy of the
treatment and disposal operations.
ACKNOWLEDGMENTS
This study is part of a major re-
search program on chemical treatment
technology, which is now being conducted
by the U. S. Array Engineer, Waterways
Experiment Station and funded by the
Environmental Protection Agency, Munici-
pal Environmental Research Laboratory,
Solid and Hazardous Waste Research Divi-
sion, Cincinnati, Ohio, under Interagency
Agreement, EPA-IAG-D4-0569. Thanks are
extended to Norman R. Francingues for
technical review. Robert E. Landreth is
the EPA Program Manager for this research
area.
REFERENCES
1. Acker, W. L. 1974. Basic Procedures
for Soil Sampling and Core Drilling.
Acker Drill Co., Scranton, Pennsyl-
vania. 246 pp.
2. American Society for Testing and
Materials. 1978. Annual Book of
Standards, Part 19, Section D420-69,
Society for Testing and Materials,
Philadelphia, Pennsylvania.
pp. 64-68
3. Committee on Interstate and Foreign
Commerce, U. S. House of Representa-
tives, Ninety-Sixth Congress. 1979.
Hazardous Waste Proposal. Committee
Print 96-IFC 31. U. S. Gov. Printing
Office, Washington D. C. 82 pp.
4. Connor, J. J. and H. T. Shacklette.
1975. Background Geochemistry of
Some Rocks, Soils, Plants, and Vege-
tables in the Coterminous United
States. U. S. Geol. Surv. Prof.
Paper 574-F. U. S. Gov. Printing
Office, Washington, D. C. 186 pp.
5. Mercer, R. B., P. G. Malone, and
J. D. Broughton. 1978. Land
201
-------
disposal of Hazardous Wastes, Pro-
ceedings of the Fourth Annual Re-
search Symposium. EPA-600/9-78-016.
U. S. Environmental Protection
Agency, Cincinnati, Ohio.
pp. 357-365.
6. Siegel, S. 1956. Nonparametric
Statistics for the Behavioral
Sciences. McGraw-Hill Book Co.,
New York. 312 pp.
7. U. S. Environmental Protection
Agency, 1979. Statistical Test;
Hazardous Wastes Guidelines and
Regulations. Federal Register,
44(183): 54323-54324.
202
-------
LEACHATE FROM MUNICIPAL AND INDUSTRIAL WASTE
LANDFILL SIMUIATERS
Riley N. Kinman
University of Cincinnati
Cincinnati, Ohio
James J. Walsh
SCS Engineers
Covington, Kentucky
ABSTRACT
In late 1974, a simulated landfill study was initiated to study landfill behavior
under controlled conditions. A total of 19 test cells was constructed to study different
infiltration rates, pH buffering compounds, prewetting the wastes and co-disposing
refuse with sewage sludge and various industrial wastes. Each cell received approxi-
mately 3.2 metric tons of municipal refuse. Twelve received municipal refuse plus small
quantities of industrial wastes and other materials. Water additions were made on all
cells on a preset schedule to approximate Midwest U.S. infiltration rates. Quality and
quantity data were collected on leachate and gas produced.
This paper addresses data collected from leachate and gas analysis for the cells
which contained municipal refuse plus industrial wastes. Data encompass the period from
November, 1974 to August, 1978. Mass releases of leachate constituents were plotted and
studied. Gas quantities and qualities were studied also. Data continue to be collected
and analyzed on this project.
INTRODUCTION
Under the Resource Conservation and
Recovery Act of 1976 (RCRA), an inventory
of all solid waste disposal sites is to be
made across the nation. In addition, all
sites identified by the inventory must be
classified as either a "sanitary landfill"
or an "open dump". Sites found to be in
compliance with the Classification Criteria
for Solid Waste Disposal Sites^' (and thus
posing no significant threat to human
health or the environment) will be classi-
fied as sanitary landfills. Sites found to
be in non-compliance with the criteria will
be classified as open dumps. Open dumps
must be either closed immediately or up-
graded as necessary to mitigate the hazard
identified by the inventory. The closure
and upgrading procedures required will de-
pend in large measure on the degree and
status of the hazard. That is, if the
hazard is large and yet to be fully exer-
cised expensive confinement and/or waste
removal procedures may be dictated. If the
hazard is large but most of the damage is
already done, money would be better
directed toward clean-up than confinement
and/or removal. Lastly, if the hazard is
small, the site may be simply closed and
funds used elsewhere.
A framework to assist in these deci-
sions needs to be established. From
numerous previous studies, data for such
a framework has been generated on the
decomposition processes in landfills and
the release of contaminants from landfills
into the environment. However, many of
these efforts have had relatively short-
lived monitoring terms. In addition, the
variability and unknowns of initial con-
ditions among such studies have made
comparisons of data and compilation of
large data bases difficult and often im-
possible.
This project was motivated by the
need for such a data base. The broad
objective of the program was to study
solid waste decomposition and contaminant
release at various types of landfills.
203
-------
Specific objectives were to determine:
1. The effect of different water
infiltrations on solid waste de-
composition.
2. The effect of sewage sludge addi-
tions on solid waste decom-
position.
3. The effect of pH buffer addition
on solid waste decomposition.
4. The effect of six selected indus-
trial sludge additions on solid
waste decomposition.
5. The effect of initial water addi-
tion on solid waste decomposition.
6. The survivability of poliovirus
in landfills.
7. The effect of different air and
soil ambient temperatures on
solid waste decomposition.
8. The ability to duplicate monitor-
ing data from two test cells con-
structed and operated under simi-
lar conditions.
TEST CELL CONSTRUCTION
The experimental test facility for
this project is located at the U.S. Envi-
ronmental Protection Agency's Center Hill
Laboratory in Cincinnati, Ohio. A total of
19 test cells were constructed for this
effort in November, 1974 and April, 1975.
Fifteen of these are located outdoors and
below the ground surface. These exterior
test cells are arranged in a horseshoe
alignment as shown in Figure 1. The re-
maining 4 cells are located inside the
EXTERIOR TEST CELLS
INSTRUMENTATION
BUILDING
®0[®0©
© fa 0
© O ©
©
O
LEACHATE
COLLECTION
WELL
high bay area of the Center Hill Labora-
tory and are above ground resting on the
concrete slab.
Individual test cells consist of
steel tubes 1.83 m in diameter, 3.66m high
and 4.76 mm thick. Steel sidewalls were
coated with coal tar epoxy as a moisture
seal. The outside test cells were placed
on concrete slabs in a excavated area.
Soil was then backfilled around the side-
walls to within 0.3 m of the top of the
cell. Several layers of fiberglass cloth
were placed inside at the base of these
cells and extended 0.3 m up the sidewall
to provide a watertight and gas tight seal.
The interior test cells were placed atop
steel bases welded into place.
Provisions for leachate drainage were
installed for all cells. Exterior cells
had a small depression in the concrete slab
and connective piping to a leachate collec-
tion well. This well serves as a central
leachate collection point for all exterior
cells. It is also used as a groundwater
drawdown to prevent pressure and infiltra-
tion of groundwater into the test cells.
Interior cells are mounted on concrete
blocks and leachate drains are attached be-
low the cell bottoms.
The test cells were then readied for
waste loading. To minimize the exposure
time and more closely simulate actual land-
fill conditions all test cells were com-
pletely loaded and covered within 7 days in
each loading period. First, a 15.2 cm
thick layer of silica gravel was applied at
the bottom of each cell. This serves as a
reservoir for leachate and prevents refuse
from clogging the drain. Silica gravel was
INTERIOR TEST CELLS
\OK
BUILDING WALL'
Figure 1. Test Cell Location Plan
204
-------
selected to prevent any chemical reaction
with collected leachate. Refuse was then
delivered to the site and added in eight
0.3-m thick increments or lifts. Each
lift was compacted with a wrecking ball to
a density of approximately 270 kg/m3.
Sludges and other materials added to se-
lected cells were applied to the top of
each lift (except the first) in propor-
tionate amounts. No additives were
applied to the first lift to avoid any
premature leaching of moisture. Tempera-
ture probes were installed atop the second,
fourth and sixth lifts in each cell. Gas
probes were installed atop the second and
sixth lifts.
At completion of waste loading a
0.3-m thick layer of silty clay cover soil
was applied atop the waste. An additional
0.3-m thick layer of pea gravel was placed
atop the cover soil, and a gas probe and
water distribution ring installed in the
gravel. A settlement indicator was
affixed to the top of the gravel and
mounted inside a sight glass through the
steel lid on each cell. This lid is
bolted down to the test cell and caulked
to provide an airtight seal. A cross-
section of a typical test cell is shown in
Figure 2.
-SETTLEMENT INDICATOR
• SI8HT TUBE
_WATER INPUT CONNECTOR
ROBE CONNECTC
ERATURE PROB
TEMPERATURE PROBE
TEMPERATURE PROBE
-03m PEA GRAVEL
03m SILTY CLAY
4 m MUNICIPAL REFUSE
0 15m SILICA SRAVEL
CONCRETE SLAB
LEACHATE DRAIN
WASTE CHARACTERIZATION
As mentioned previously, sludges and
other materials were added to selected
cells. These additives are identified in
Table 1 along with other pertinent initial
loading conditions such as weight and
moisture contents. The intent of these
additions was to allow an investigation
of what effects codisposing these
materials with municipal refuse would
have on solid waste decomposition. More
specifically, this arrangement provided
test cells which fulfilled the earlier
objectives as follows:
1. Different water infiltration
rates;Test Cells 1, 2, 3 and 4.
2. Sewage sludge additions: Test
Cells 5, 6 and 7.
3. pH buffer addition; Test Cell 8.
4. Six selected industrial sludge
additions: Test Cells 9, 19, 12,
5.
6.
13, 14 and 17.
Initial water addition:
Cell 11.
Test
Survivability of poliovirus:
Test Cell 15.
Figure 2. Cross-section of Typical Test Cell
7. Different ambient air and soil
temperatures: Test Cells 2 and
and 16.
8. Duplication of monitoring data;
Test Cells 16, 18 and 19.
Before loading, all solid waste for each
lift in each cell was weighed and sorted.
Eleven sort categories were used and the
average composition in each cell was then
computed. A summary of the solid waste
composition of each cell is shown in
Table 2. In addition to categorization,
two samples were extracted from each lift.
An 11.4 kg sample was used for moisture
content determination; these were then
averaged over each cell to determine the
moisture contents shown in Table 1. An
18.2 kg sample was used for chemical
analysis; results were computed for each
sort category. A sample composited accord-
ing to average sort categories in all cells
was also chemically analyzed (see Table 3).
OPERATION AND MONITORING
To more closely simulate actual in-
field environmental conditions, all test
cells are operated in accordance with a
strict monthly schedule. First, gas
205
-------
TABLE 1. TEST CELL LOADING AND OPERATION
Cell Number
Kef use
wet weight (kg)
initial moisture ( 1 )
ury weujht (ky)
Additive
type
wet weight (kg)
moisture content {% of ww)
inl tia) moisture (1 )
dry weight (kg)
Solid Waste Total
wet weight (kg)
moisture content (I of ww)
Utltldl uio'.iture (I )
dry weight (kg)
Top Type
open
open/closed
closed
Annual IntUttdtion
nm
1
1 2 • 3 A
3025 2989 3007 3002
3t> 35 35 35
1056 1043 1049 1048
1969 1946 1958 1954
..
.-
.-
-
3025 2989 3007 3002
35 35 35 35
1056 1043 1049 1048
1969 1946 1958 1954
X X
X 1C
203.2 406!4 609.6 812 8
533.30 1067 01 1602.95 2135.87
5
3001
35
1047
1955
sewage
sludge
68
88
60
8
3069
36
1107
1963
X
406 4
1067.01
6
2919
35
1019
1900
sewage
sludge
204
BE
179
25
3123
38
1198
1925
X
406 4
1067.01
7
2964
35
1034
1930
sewage
sludge
680
38
596
82
3644
45
1632
2012
X
406.4
1067.01
8
2994
35
1045
1949
CaCOj
90.7
10
9
82
3084 7
34
1054
2031
X
406.4
1067.01
9
3001
32
966
2035
petroleum
s 1 udge
1518
79
1199
319
4519
48
2165
2354
X
406.4
1067.01
10
2998
3?
965
20 J3
battery
waste
1291
89
1153
138
4289
49
2118
2171
X
406 4
1067.01
11
2924
35
1020
1094
H20
1243 2
100
1293
0
4217.2
55
2313
1904
X
406 4
1067.01
12
3048
35
1064
1984
electro-
plating
waste
1190.4
80
946
244
4238 4
47
2010
2228
X
406.4
1067.01
13
3006
35
1049
1957
inorganic
pigment
waste
1420.6
51
728
693
4426 6
40
1777
2650
X
406.4
1067.01
14
3015
35
1052
1963
chlorine
prod.
brine
sludge
2038.7
24
491
1548
5053.7
31
1543
3511
X
406.4
.067 01
15
1010
32
969
2041
pol 10-
vlrus
-
--
-
3010
32
966
2041
X
406.4
1067 01
16 17 18
2996 2998 3000
35 34 32
1046 1046 966
1950 1972 2034
solvent
based
paint
sludge
1604 0
25
1206
1206
2996 4602 3000
35 31 32
1046 1444 966
1950 3153 2034
XXX
406.4 406.4 406.4
1067.01 1067.01 1067 01
19
3012
32
970
2042
-•
JOI2
32
970
2042
X
406 4
1067 01
-------
TABLE 2. REFUSE COMPOSITION IN TEST CELLS
(2)
Plastic, As
Ttst Food C*cd«n rubber roc*
1
I
1 1
5 1
6
•j
a
9
10 1
11
12
13
1*
IS 1
16
17
ia
19
.5 13.8 37.1 6
.3 2
.5 41. S 5.
.0 30.2 34
.3 16.6 41.
-i 2
.3 2(
.6 1
.4 i
.2
.7 1
_ 2
!s 2
.0 1
,7
.3 1
.0 1
.2
.9 36.
.3 43.
.1 53.
.0 41.
.2 44.
.2 39.
.5 46.
.7 39.
,1 37.
.4 42.
.3 41.
.6 45.
8.
6.
5.
6.
a.
11.
6.
S.
5.
5.
6.
7.
10.
5.
.8 40.3 6.
.5 11.9 43.2 7.
5>
4.
J-
«!
i.
3.
3.
2,
4.
5.
2.
4,
J.
3.
3.
1.4 I
:.2
.3
2.0
1.1
1.4
1.7
l.l
1.9
4.4
1.6
1.2
1.4
1.5
2.9
.6
6.0 1 4
3.3 3.9
2.7 1.9
3
10.0
2 S 4
1-2 3
2-
1.
1.
2.
1
2.
2.
1,
3.
4 .
2.
2.
lt
2.
3
2
;
2
3
j
1
2
3
15.7
1.1
41.8
4.S
(.9
1.8
4.1
l.S
1.8
1.0
2.1
1.0
Psrcint by wcc vet|hc
TABLE 3. REFUSE CHEMICAL ANALYSIS
(3)
Coajponent
CODi
TICK
Total Phoaphata
Liplds
Ash
Crude flher
local carbon
Inorganic carbon
Organic carbon
Sugar aa Sucroae
Starch
Asbaatoa
Arscnlctt
Selenlusitt
Marcurytt
Uadtt
(erylllusitt
Cad>lusrtt
Irontt
ZllKtt
dtroiluajtt
Manganaaatt
Potaaaiu«tt
Hagnisluutt
CalciusH
Sodlintt
Coppertt
Hickeltt
Moisture
Molaturell
Coeiposition
Gosipositionff
* Percent by dry
t <25.4 n (1.0
\ g COD/g sample
Papar
0.804
0.028
0.048
2.47
92.0
21.7
58.0
4.30
53.7
<0.1
3.40
MAf
<0.1
MA
MA
MA
MA
0.36
375
50.0
(.2
13.1
11.2
160
77.5
9.70
4.5
15.7
it. 7
33.20
42.6
41.62
weight
inch).
Garden
0.315
0.171
3.14
3.04-
36.5
16.6
14.4
4.66
9.74
1.71
7.42
HA
NA
HA
MA
NA
HA
HA
330
106
1.1
194
0.135*
4175
0.830*
185
9.34
15.7
156.4
56.91
10.7
15.77
unless
Metal
0.492
0.022
2.79
0.420
4.85
0.235
4.80
3.40
1.40
<0.1
<0.1
KA
*0.1
HA
NA
HA
MA
20.9
625*
175
15.3
870
1.00
to.s
<0.25
37.0
0.221*
115
8. BO
6.18
12.2
8.21
otherwise
Glaaa
0.011
0.140
0.049
1.54
2.25
0.040
0.750
0.220
0.530
<0.1
<0.1
KA
10.2
HA
KA
NA
HA
2.7
3220
9.75
1.1
15.7
2.70
472
16.2
60.0
2.54
19.0
2.00
1.65
12.2
7.83
Food
0.754
3.09
10.
13.
41.
10.
19.
2.58
17.0
6.08
8.57
HA
HA
HA
HA
NA
HA
NA
505
59.0
1.3
12.2
0.162*
377
0.465*
804
8. 58
12.5
216.5
70.07
3.6
7.56
rustics
rubber,
leather
2.14
1.25
1.40
5.02
182
21.5
15.8
5.75
10.1
<0. 1
3.42
HA
NA
NA
KA
NA
HA
finest
0.935
0.131
1.97
4.85
49.5
6.39
16.4
4.30
12.1
1.18
7.20
HA
1.2
HA
KA
NA
HA
I. lit 4.2
444
118
2.0
12.1
98.7
289
912
143
12.4
32.0
57.04
49.27
8.7
10.91
specified. ••
tt
I!
0.392*
322
13.1
113
135
1.02*
2.11*
400
35.8
33.2
123
49.36
2.9
3.58
Fibers per
Parts per
Plastics,
Ash,
rocxs,
dirt
0.040
0.119
4.48
1.32
19.6
5.85
13.4
7.80
5.60
<0. 1
6.40
HA
3.6
NA
NA
NA
NA
4.5
0.340
181
10.1
177
555
2.63*
4.08*
0.209*
32.6
10.1
30.79
18.52
3.2
3.36
gram.
Billion
rubber.
Diapers
0.720
0.138
2.65
2.26
96.0
13.7
44.5
0.740
41.8
<0,1
<0.1
HA
<0.1
NA
HA
HA
HA
0.25
99.0
343
0.5
5.90
750
279
360
0.110*
4.14
3.36
133
66.28
1.3
2.47
by weight
Wood
0.503
0.228
0.103
1.00
77.9
20.8
51.0
0.380
50.7
<0.1
0.78
HA
<0.1
HA
HA
HA
NA
1.6
0.378
59.4
I.I
50.0
90.0
253
590
572
38.2
27.0
21.43
17.10
2.6
1.99
and leather not
Conpoaite
0.520
0.247
2.32
2.84
25.8
11.3
24.8
3.08.
21.8
3.50
16.2
-------
systems on each cell are controlled by
top positions. As shown in Table 1, tops
may be (1) permanently open as for Test
Cells 9, 10, 12, 13 and 14; (2) perma-
nently sealed closed as for Test Cells
3, 4, 16, 17, 18 and 19; or (3) opened
and closed in accordance with a monthly
schedule as for Test Cells 1, 2, 5, 6, 7,
8, 11 and 15. The schedule of open/
closed gas systems is shown in Table 4.
These gas systems are closed only during
selected months to simulate temporary
infield conditions of frozen or saturated
ground cover.
Physical data recording, sampling
and analysis are performed in accordance
with the schedule in Table 5. To summa-
rize this schedule, gas volumes, refuse
temperatures and refuse settlements are
recorded regularly. Gas samples are
collected each month and analyzed via a
gas chromatograph for five major constit-
uents. Leachate samples are collected
each month, their volumes recorded, and
representative samples prepared. Monthly
samples are then analyzed for 22
parameters. An additional 12 parameters
are determined quarterly and 7 parameters
TABLE 4. MONTHLY CELL OPERATION SCHEDULE
Gas System Schedule
Infiltration Schedule
Month
January
February
March
April
May
June
July
August
Septemoer
October
November
December
Amount per month (liters)
Position on Open/
Closed Cell Tops
Open
Open
Closed
Closed
Qoen
Closed
Closed
Open
Closed
Open
Closed
Closed
Type per
Month
None
Low
High
High
High
Low
None
None
None
LSW
Low
Lou
203. 2 nm
Annual Rate
0
48.52
96 90
96.90
96.90
48.52
0
0
0
48.52
48.52
40.52
406.4
Annual Rate
0
96.90
194 17
194.17
194.17
96.90
0
0
0
96.90
96.90
96.90
609 6
Annual Rate
0
145.72
291.45
291.45
291.45
145.72
0
0
0
145 72
145.72
145.72
812.8
Annual Rate
0
194 17
388.34
388.34
388.34
194.17
0
0
0
194.17
194.17
194.17
1602.95
Secondly, water is applied to each
test cell in accordance with the annual
infiltration rates listed in Table 1. As
shown, with the exception of Test Cells
1, 3 and 4, all test cells receive an
annual water application of approximately
400 mm. Test Cells 1, 3 and 4 receive
annual applications of approximately 200,
600 and 800 mm, respectively. In this
way, the goal of varying the moisture
regimen is realized in Test Cells 1, 2, 3
and 4 (with 200, 400, 600 and 800 mm of
infiltration, respectively). Water is
applied in accordance with the schedule
shown in Table 4. As shown, an attempt
has been made to duplicate infield con-
ditions. Specifically, high applications
are made in the normally rainy months of
March, April and May. No applications are
made in the normally dry months of January,
July, August and September. Low appli-
cations are made in the remaining months.
semi-annually. All analyses are performed
at the Center Hill Laboratory using facil-
ities and equipment used exclusively for
this project. Strict quality assurance
program procedures are utilized to ensure
the validity of sampling results. Ana-
lytical data derived from the project is
punched onto computer cards and entered
into the memory banks of an EPA in-house
computer. This computational facility
can then be used for plotting project data
as an aid in evaluating results.
MOISTURE BALANCE EVALUATION
(A)
A previous paper on this pro3ect
addressed the moisture balance for these
cells. Figure 3 is a plot of the moisture
added and moisture leached for Test Cell 2,
which is the control cell for the exterior
industrial waste loaded cell comparison.
All of the cells loaded with industrial
208
-------
TABLE 5. DATA RECORDING, SAMPLING AND ANALYTICAL SCHEDULE
SAMPLE
Gas
Soil/
Refuse
Leachate
ANALYSIS
CHU
C02
«2
02
«2
Volume
Temperature
Temperature
Settlement
TOC
COO
Total Solids
Total Volatile Solids
PH
ORP
Specific Conductance
Alkalinity
TKN
ortho-POt.
Fe
Cd
s"
Cl
Hg
V
Be
Se
CN
Phenol
As
Cr6
Cu
Pb
N1
Zn
Asbestos
Sn
T1
Total P0»
Organic Ados C2 to C«
Fecal CoHform
Fecal Streptococci
Cr total
As
Phenol
Be
T1
B
V
Sb
TEST CELL
All
All
All
All
All
16, 17, 18, 19
1 - 15
16, 17, 18, 19
All
All
All
All
All
All
All
All
All
All
All
All
All
All
All
All
2, 10. 13
2, 9, 10, 12, 13, 14
2, 9, 12, 13
2. 10, 13
2, 5, 6, 7, 9
2, 14
2, 9, 13
All
All
All
AH
2, 4, 5, 6, 7, 13, 14
All
2. 4, 14
All
All
All
All
All
All
All
All
All
All
All
All
FREQUENCY
Monthly
Monthly
Monthly
Montnly
Monthly
Daily
Biweekly
Daily
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Quart
Quart
Quart
Quart
Quart
Quart
Quart
Quart
Quart
Quart
Quart
Quart
Semi-
Seml-
Semi-
Semi-
Sem1-
Semi-
Hy
rly
rly
rly
rly
rly
rly
rly
rly
rly
rly
rly
nnually
nnually
nnually
nnually
nnually
hnually
Semi- nnually
wastes received the same moisture appli-
cations as Test Cell 2. Test Cell 16 was
the control for Test Cell 17 comparison.
LEACHATE QUALITY EVALUATION
A total of 41 parameters were
measured on leachate samples collected for
this project. Thirteen of these were
selected for discussion here. It is useful
to remember that plots shown depict mass
leached (cumulative Kg contaminant/Kg of
dry solid waste) as a function of leachate
volume (cumulative 1 of leachate/Kg of dry
solid waste).
PH
Figures 4 and 5 are plots of pH as a
function of cumulative leachate volume for
the control Test Cell 2 and Test Cell 12
which received plating wastes. This is an
example of a waste which influenced the
pH of the leachate to be generally higher
than the control. Usually approximately
0.7 units higher than the pH of the con-
trol. Sewage sludge addition did not
appear to cause any appreciable change in
pH of the leachate. Other industrial
wastes which did not appreciably change
leachate pH were the chlorine production
sludge, paint sludge and water softening
sludge. Petrochemical wastes, battery
wastes, and inorganic pigment wastes re-
sulted in noticeable higher pH values
similar to the plating wastes. So Test
Cells 9, 10, 12, and 13 all exhibited
leachate pH values that were noticeably
higher than the control. This would appear
to reflect some toxic effect on the orga-
nisms producing acids in the cells, which
resulted in less acid production and higher
pH values.
Alkalinity
Figures 6 and 7 are alkalinity plots
for the control Test Cell 2 and the Battery
Production Waste Test Cell 10. These
plots illustrate the reduction effects on
alkalinity caused by this industrial waste.
Note that the control plot is a curve which
209
-------
3429 S0.7 «»0
-------
0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00
M3ISTUKE LEMED (I 1 HjOAg SOLID VBSTE)
Figure 6. Mass Leactad: Alkalinity At Test Cell 2
0.50 O.T5 1.00 1.25 1.50 1.75
KlISTuPE LEKHED (£ 1 HjO/kg SOLID WCTE)
Figure 7. Mass Laached: Alfcalijiity At Test Cell 10
appears to be approaching an asymptote, but
the Test Cell 10 plot has a straight line
configuration and the values for alkalinity
are considerably lower than the control.
This could mean that the industrial waste
was retarding decomposition and hence the
production of alkalinity in the period pre-
sented. Test Cells 8, 9, 13 and 17 ex-
hibited this kind of effect. Sewage sludge
in Test Cells 5, 6, and 7 had an opposite
effect. An increase in alkalinity was
noted in this case. Test Cells containing
plating wastes and chlorine brine pro-
duction waste also exhibited this effect of
an increase in alkalinity. Alkalinity is
important in methane production and study
of the plots reveals that there is con-
siderable alkalinity available to withstand
momentary increases in acids without
drastic changes in pH.
Total Solids
Total solids release followed the
asymptotic configuration in every case.
Figures 8 and 9 are plots for Test Cells
2 and 14 respectively. Test Cell 14 had
a much higher release of solids than did
Test Cell 2. Test Cell 12 also had a
greater release of solids than the control.
All of the other cells loaded with sewage
and various industrial waste had a lower
release of solids than the control. Only
the plating wastes cell and the chlorine
production brine waste cell had a release
of solids higher than the control.
Total solids contain both an organic
fraction and an inorganic fraction, both
of which vary with time from initial
placement of the refuse. This makes
211
-------
t.OO 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00
M3I3IURE I£ftOffiD (£ 1 H2OAg SOLID VftSTE)
Figure 8. Mass Leached: Total Solids At Test Cell 2
1 1 1 1 1 1 1 1
0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00
M3ISIURE LEftOffiD (T I HjO/Kg SOLID VRSTE)
Figure 9. Mass Leacted: Total Solids at Test Cell 14
interpretation of why the solids release
from one cell is greater than the solids
release from another cell difficult, but
it appears that in the two cases where the
solids levels were greater than the con-
trol the greater release resulted from a
high contribution of inorganic materials
that were amenable to leaching. In all of
the other cells where total solids release
was less than the control, there appeared
to be a slowdown in the rate of solubili-
zation, probably caused by decreased bio-
logical activity in the cells and de-
creased levels of inorganics readily solu-
ble in water. This suggested that the
wastes were exerting an effect on bio-
logical activity and solids release.
TOG
Total Organic Carbon values were
plotted in Figures 10 and 11 for Test Cell
2 and Test Cell 10 respectively. Test
Cells 8, 9, and 10 released TOC at a re-
duced rate when compared with the control.
Also the release follows a different
pattern as evidenced by the straight line
plot in Figure 12 for Test Cell 10 con-
taining battery wastes. The control, Test
Cell 2, followed an asymptotic release.
Test Cells 12, 13, 14, 17 and the sewage
sludge cells followed a release pattern
similar to the control. Both major
mechanisms of TOC release in the leachate
(.chemical and biological) were probably
affected by the wastes which lowered the
release.
COD
Chemical Oxygen Demand values were
plotted in Figures 12 and 13 for Test Cell
2 and Test Cell 10 respectively. COD was
released at a reduced rate by Test Cells
8, 9, 10 and 13. Release followed a
similar asymptotic configuration as the
212
-------
K>
's'
MBSS IEACHED (I 102mg CODAg SOLID WASTE)
s
y
N
X
\
\
*
MASS LEACHED (I 102mg TOC/kg SOLID TOSTE)
8 X
8-
s-
MASS LEACHED (J 102ng TOC/kg SOLID TOSTE)
,8888
$ 1 r
X
§ 8
\
\
-------
/
0.00
0,25 0.50 0.75 1.00 1.25 1.50 1.75
MOISTURE: LEACHED (i i H2oAg SOLID WSTE)
Figure 13. Mass Leached: COD At Test Cell 10
2.00
control with the exception of Test Cell 10,
which followed a straight line release at
a much lower rate. COD values that are
higher than most parameters on a mg/1
basis result from the many organic and in-
organic substances in the refuse leachate
amenable to chemical oxidation in this
test. Test Cells 8 and 10 (CaC03 and
battery wastes respectively) showed less
TOC release and COD release than the con-
trols. Test Cells 5, 6, 7, 12, 14 and 17
had COD release similar to the control.
TKN
Total Kjeldahl nitrogen release
followed an asymptotic pattern. Figures
14 and 15 represent the control, Test Cell
2, and Test Cell 8 containing the soft-
ening sludge. Note that the release from
Test Cell 8 was significantly lower than
the control. This was characteristic of
Cells 9, 10, 12, and 13. Substances in
these cells may have interferred with the
biological denitrification reactions. All
of the other cells 5, 6, 7, 14, and 17 had
approximately the same release of this
form of nitrogen. Note that the sewage
sludge addition had little effect on this
important leachate parameter.
Metals; Fe, Cu, Cr, Pb, Ni, Cd, Zn, Hg
Eight metals were selected for review
here. In general the release of metals
followed three types of curves or com-
binations thereof. Either a straight line
release or series of straight line
releases was a common configuration.
Another was the familiar curve approaching
an asymptote with a decreasing rate of
increase. The third type was a curve with
an increasing rate of increase. Many
mechanisms work in a landfill to cause the
release of specific metals in the leachate.
It appears that any metal no matter what
its theoretical solubility in water, may
be leached from the refuse over a period
of time.
Figures 16 and 17 depict the mass re-
lease of iron for Test Cell 2 and Test
Cell 14 respectively. The straight line
release in Figure 16 is typical for Test
Cells 5, 6, 7, 8, 9, 10, 12, 13 and 17.
Figure 17 depicts the mass release for
Test Cell 14 and has an increasing rate of
increase type of release. Release from
Test Cells 8, 12, 13, 14 and 17 was less
than the control. These wastes appeared
to slow down the release of iron. Two of
the sewage sludge Cells, 6 and 7, had
greater releases of iron.
Cu
Figure 18 depicts the mass release of
copper from Test Cell 2. All of the Cells
released copper in a similar manner.
Cells 9, 12 and 14 released more, while
Cells 8, 10, and 13 released less. Sewage
sludge in Cells 5, 6 and 7 did not appear
to affect any change in release of copper.
Copper was released from different
locations in the cells at different rates.
Cr
Figure 19 depicts the mass release of
chrome from Test Cell 2. All of the Cells
released chrome in a similar manner. Each
Cell containing industrial waste released
less chrome than the control Cell. Chrome
was present in the leachate in sub-mg/1
214
-------
150.00
S 125.00
ty 100.00 •
|T 75.00-
25.00-
/
X
1 I 1 1 I I 1 I
0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00
MOISTURE LEACHED (I 1 HjO/kg SOLID WSTE)
Figure 14. Mass Leached: TON: At Test Cell 2
62.65
12.53-
r
•
/
0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00
MOISTURE LEACHED (!) 1 t^O/kg SOLID WASTE)
Figure 15. Mass Leached: TKN At Test Cell 8
433.00
216.51-
*•
•
/
y
—i 1 1 1 1 1 1 r
0.25 0,50 0.75 1.00 1,25 1.50 1,75 2,00
MDISTUBE LEACHED (I 1 HjO/kg SOLID HASTE)
Figure 16. Mass Leached: Fe At Test Cell 2
215
-------
v s
s §
QI -"-I
> \3 KJ
t? Q
tn U
rt D
MASS LEACHED (I Irtj CrAfl SOLID WASTE)
S I g I g
MASS LEACHED (S mg CuAg SOLID WASTE)
o o o p p
O O h-> M to
*. 03 N) 1^ O
M S
fftSS LEACHED (I mg FeAg SOLID WASTE
> l£) 4> kD *»
0 kB - s"
^^ •€
V 10"
tf p -
| S ^
E 1 s"
i-1 m
-o"
to
\
\
\
TV
X
x^
\^
v-^
-------
concentrations only. Chrome release frcm
landfill refuse appeared to be a slow
process.
Pb
Figures 20 and 21 depict the mass re-
lease of lead from Test Cells 2 and 12 re-
spectively. Test Cells 8, 9, 10, 13 and
17 had higher releases of lead than the
control. Note in Figure 21 the increasing
rate of increase with time for lead re-
lease.
Ni
Figures 22 and 23 depict the release
of nickel fron Test Cells 2 and 12 respec-
tively. Note that the control was
asymptotic, while the cell containing the
electroplating wastes had different rates
of release and a higher total release to
date. Only Test Cells 12 and 14 released
nickel at higher values than the control.
Test Cells 5, 6, 7, 8, 9, 10, 13 and 17 re-
leased nickel at either the same or lower
values as the control.
Cd
Figures 24 and 25 depict the release
of cadmium frcm Test Cells 2 and 10 respec-
tively. Note the straight line release
pattern and the different rates of release.
Only the battery wastes in Test Cell 10
showed a higher release of cadmium than
the control. Test Cells 5, 6, 7, 8, 9, 12,
13, 14 and 17 released cadmium at either
the same or slightly lower values as the
control.
Zn
Figures 26 and 27 depict the release
of zinc from Test Cells 2 and 17 respec-
tively. Release followed an approximate
asymptotic configuration. All of the Test
Cells released zinc at either the same
rate or a lower rate than the control to
date.
He,
Figures 28 and 29 depict the release
of mercury frcm Test Cell 2 and 14 respec-
tively. Both curves follow an approxi-
mate asymptotic configuration. All of the
Test Cells containing industrial wastes
released mercury at either the same or a
lower rate than the control. Only Test
Cell 14 released more mercury than the
control. Release frcm this cell contain-
ing the chlorine production brine sludge
was over twice as much as from Test Cell
2. This waste would be expected to con-
tain some residual mercury frcm the
mercury cells used in chlorine production.
So the release values were not unexpected.
81.30-1
67.75-
54.20-
7 40.65-
o
27.10-
13.55-
0.00
0.00
0.25
1—
0.50
0.75 1.00 1.25 1.50 1.75
MOISTURE LEACHED (£ 1 H-OAg SOLID WASTE)
Figure 20. Mass Leached: Pb At Test Cell 2
—i—
2.00
217
-------
fftSS LEACHED (I rag NiAq SCUD IftSTE)
h- to * w -J
& 00 U> -O IO
Ck CO tO W O
LESCHED (I mj NiAg SOLID TOSTE)
§
o
K"
"5 O
O
| 9 S"
r^ 1
oo & n M
& M 2"
K, M
ft Jft
g | is"
(-• n
i—
K)
-J
LP
KI
§"
V §'
\ o
\ bi"
\ -q p
V iS " jjc 01 -
X -W H o
x rj -J -
• «i
x s I r
\ s i K"
\rt ~ i-1
g I is-
M
^, .
Ul
to
§H
^••^^^
\
\
Xx
v\
\
\
\
MASS LESCHED (I I0"2mg PbAg SQUD WISTE)
CB»
-
o-
°
X
-------
MASS LEACHED (I mj ZnAg SOLID VBSTE)
MASS LEACHED (I 10 3 rog CdAg SOLID IftSTE)
MASS LEACHED (I Itf 3rog CdAg SOLID TOSTE)
N>
u u u h-i-'wiv ^j ^- ts) rvj
NJ -JNJCT- *. <» 10 ^ 0 £
O Ul O Ul O Dl O OifcvOWOOIO-J OOOOOOO
c
o c
s
Lfl
•T) o
| 5 !5"
K ^
M r>
F B
S M O
" -o
s > T_
> m
ft CO
K ^
" ? r-
H 1 S
K)
*"•
tn
K..
§
8888S8 8 g 8 g. 8 8 8 0§
^v s"
^ X B-
\ 3 P
\ '"HP.
\ a R Di
\ 1 |
\ 1 » £-
" "o
> tn
rt 10
fT ^ h-
O P tn ~
(D W O
*" S
M
o
• _
Ul
N>
§-
J_ . . . . . ^ 5
~~~ "*^^~^-^_
^s"*'»^^ p
** ^-^^ KJ "
V\ 1 8 "'
*x * H o
S S ^
I -r
" K?
2 |C.
a "
ft M
g | s-
^ —
. .
s
10
8
888888
, i i i i .
'v
\
\
\
\
\
v_^__
~~~~— «^
"*-y
\
-------
78.00 _
g 39.00 -
26.00 -
I
L
r 1 1 1—' T " 1 r
0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00
MOISTURE LEACHED (I 1 H2OAg SOLID WASTE)
Figure 27. Mass Leached: Zn At lest Cell 17
51.25.
_£ 30.75-
20.50-
10.25-
0.00
0.00
—I 1 1 1 1 I I r
0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00
MOISTURE LEBCHED (T 1 HjOAg SCUD WASTE)
Figure 28. Mass Leaclled: Hg At Test Cell 2
147.OOH
98.00-
f
.00-
.50-
0.
/
/
1
• ••- | i i i i i I '
DO 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00
Figure 29. Mass Leached: Hg At Test Cell 14
220
-------
GAS EVALUATION
REFERENCES
Gas samples were extracted from all
test cells on a monthly basis for CR^,
O~, CO-, and N- analyses. Gas quantity
data were recorded for the interior cells
16, 17, 18 and 19 on a daily basis.
Figures 30 and 31 depict the quantity of
gas generated at Test Cells 16 and 17.
Gas volumes from Test Cell 17 with the
solvent based paint sludge do not appear
to be greatly different from the control
Test Cell 16.
Figures 32 and 33 depict the methane
composition in the gases from these same
two cells. Percent methane appeared to
be higher in the solvent based paint
sludge cell for most of the time period
with the exception of the initial days of
gas production. It appeared the toluene
xylenes etc. were readily amenable to
conversion to methane.
ACKNOWLEDGEMENTS
The work upon which this paper is
based was performed pursuant to Control
No. 68-03-2758 with the U.S. Environmental
Protection Agency. The Authors would like
to express their appreciation to Mr. Dick
R. Brunner, Project Officer, for his
assistance. The data base used in pre-
paring this paper was obtained from the
efforts of Systems Technology, Inc.,
pursuant to EPA Contract No. 68-03-2120.
The Authors also would like to express
their appreciation to June Schuck for
typing the paper.
1. U.S. Environmental Protection Agency.
"Solid Waste Disposal Facilities,
Proposed Classification Criteria."
Federal Register. February 6,
1978. Part II. pp 4942-4955.
2. Swartzbaugh, Joseph T.; Robert C.
Hentrich; Gretchen Sabel.
"Evaluation of Landfilled Munici-
pal and Selected Industrial Solid
Wastes." U.S. EPA Contract No.
68-03-2120. June, 1977. p. 15.
3. Streng, D.R. "The Effects of the
Disposal of Industrial Waste
Within a Sanitary Landfill Environ-
ment." Residual Management by
Land Disposal, Proceedings of the
Hazardous Waste Research Symposium.
February 2-4, 1976. pp. 67, 68.
4. Walsh, J.J.; Kinman, R.N.; "Leachate
and Gas Production Under Controlled
Moisture Conditions." Municipal
Solid Waste; Land Disposal, Pro-
ceedings of the Fifth Annual
Research Symposium. March 26-28,
1979. p. 46.
f?, 1J 00-
0 792 5 904 2 1015 9 1127.6 1239.3
TIML (DAYS SINCE CELL CONSTRUCTION)
221
-------
357.0 481.2 605.4 729.6 853.8 978.0 1102.2 L226.4 1350.6
TIW (MYS SINCE (T3L CnNSTRJCTICM)
Figure 31. Ru Quantity At Teat Cell 17
152.0 305.2 458.5 611.8 755.0 918.2 1071.4 ]224.6 1377.1
Til* (DAYS SINCE CEU, CTHffrwrTION)
Figure 32. Methane Gas At Test Cell 16
/\
;
/
'/
198,9 356.3 513.7 671.1 828,5 985,9 1103.3 L24O.3 1378,0
TIME (LAYS SINCE CEU. CONffTttCTICN)
Figure 33. Methane Gas At Test Cell 17
222
-------
CHEMICALLY STABILIZED INDUSTRIAL WASTES IN A
SANITARY LANDFILL ENVIRONMENT
Tommy E. Myers, Norman R. Francingues
Douglas W. Thompson, and Philip G. Malone
U. S. Army Engineer Waterways Experiment Station
Vicksburg, Mississippi 39180
ABSTRACT
Simulation of the co-disposal of municipal solid waste (MSW) with untreated industrial
waste and with chemically stabilized industrial wastes in large-scale test cells was ini-
tiated in April 1978. Simulations were conducted in 1.83 x 3.66 m cylindrical test cells
designed to simulate sanitary landfill environments. Three industrial wastes were chemi-
cally stabilized by two commercially available processes. The three types of industrial
wastes were electroplating sludge, chlorine production brine sludge, and calcium fluoride/
sewage sludge. All test cells received deionized water equivalent to an infiltration rate
of 66 cm per year. Leachate samples were collected and analyzed for 10 parameters on a
monthly basis and extensive chemical characterization of leachates for 17 metals was con-
ducted on a quarterly basis. Results of these analyses for the first 9 months of leachate
production are compared to control test cells containing MSW only. Release of toxic
metals from untreated chlorine production brine sludge co-disposed with MSW was observed,
however, chemical stabilization significantly reduced the leaching of toxic metals from
the co-disposed chlorine production brine sludge. The co-disposal of treated or untreated
electroplating sludge with MSW did not affect typical leachate quality. The co-disposal
of MSW with calcium fluoride/sewage sludge apparently improved leachate quality.
INTRODUCTION
Since passage of the Resource Conser-
vation and Recovery Act of 1976 (RCRA),
regulation of the disposal of solid and
hazardous waste has become a top priority
of the EPA. As increasing amounts of solid
and hazardous waste are directed to the
land for disposal, concern about the ade-
quacy of protection for the environment and
human health has played a major role in
establishing regulations and guidelines
under RCRA. Reports of improperly disposed
chemical waste and leaking landfills have
made nationwide news during the past year.
As a result, an extensive hunt or survey
has been launched by the EPA to identify
waste disposal sites for the purpose of
assessing the magnitude and severity of
problems associated with solid and hazard-
ous waste disposal under RCRA.
Oftentimes sanitary landfills have
received industrial wastes at some time
during their operation. Inadequate records
of landfill operation and the absence of
indicators of industrial waste pollution,
have lead to the misconception that pollu-
tion problems from sanitary landfills are
due primarily to contaminants released by
the decomposing municipal solid waste (MSW).
Thus, information on the transformation and
mobility of industrial waste co-disposed
with MSW is needed in order to determine
whether the quality of leachate from sani-
tary landfills is effected.
Previous studies reported by Newton,1
Streng,2 Swartzbaugh et al.,3 and Walsh and
Kinman,1* have dealt with the effects of
untreated industrial sludges on leachate
quality when co-disposed with MSW. Newton
concluded that the presence of industrial
wastes (an oil/water emulsion, metal hydro-
xide sludge, and cyanide waste) except in
the case of the cyanide waste, did not
effect the composition of the MSW leachate.
Other researchers2 '3>Lf concluded that the
addition of industrial wastes evaluated in
their studies appeared to have little
effect on organic, nutrient and other
demand parameters; whereas, metallic ion
223
-------
concentrations within the leachate may be
affected depending on the chemical compo-
sition of the industrial waste itself.
The purpose of this paper is to pre-
sent preliminary findings from a large
scale simulation of the co-disposal of MSW
with untreated industrial sludge and with
chemically solidified/stabilized industrial
sludge. This paper is the first in a
series of reports to be published by the
EPA, presenting the results of this study.
MATERIALS AND METHODS
Test Cell Construction
The eleven test cells used for this
project are housed in a large-scale testing
facility (Figure 1) at the U. S. Army
Engineer Waterways Experiment Station in
Vicksburg, Mississippi. This facility is
temperature regulated to maintain 25 ± 3° C
for the duration (A years) of this project.
The test cells are cylindrical, steel
(6.35 mm rolled plate) tanks with a coal-
tar epoxy coating on all interior surfaces.
The acid- and base-resistant, coal tar
epoxy protects the cell walls from corro-
sion and the cell contents from contamina-
tion. The nine co-disposal test cells and
two control (MSW) test cells used in the
study have dimensions of 1.83 m inside
diameter aid 3.66 m in height. The test
cells are equipped for collection of lea-
chate samples and insitu measurements of
temperature and redox potential.
Test Cell 3rofiles
Figure 2 is a schematic diagram of the
profile of the co-disposal test cell con-
taining chemically stabilized industrial
sludge. This profile, as are all the test
cell profiles, is built on a base of poly-
propylene beads overlain by a 30.5 cm layer
of sandy soil. The soil is 80 percent sand,
4 percent silt, and 16 percent kaolinite
clay. While still at its natural moisture
content of 7 percent, the soil was com-
pacted to fi dry density of 1763 kg/m . Each
layer of chemically fixed industrial sludge
is composed of four cylinders or cores of
fixed sludg.e. Figure 3 is a picture of four
cores of chemically fixed sludge placed over
a lift of the MSW. During the loading opera-
tion MSW WES packed around the cores to a
density of 400 kg/m3 (wet wt). The profile
was completed with a 7.6 cm layer of washed
pea gravel. The pea gravel provides a rela-
tively ineit cover that reduces evaporation
and assists in the distribution of the in-
filtrate. The test cell lid was tack weld-
ed at 15 cir intervals to the flange around
Figure 1. Facility for housing co-disposal test cells,
224
-------
Total Municipal Waste - 7.64 cu, yd.
Total Stabilized Sludge «l.86cu.yd.
Municipal Waste : Stabilized Sludge »4.l'l vol/vol
Figure 2. Schematic diagram of typical co-disposal test cell profile.
Figure 3. Placement of chemically stabilized cores in test cell.
225
-------
the top of the test cell and the small
cracks (1 mm) that occurred between some
welds were filled with metal sealant.
The co-disposal test cell profiles
containing the untreated industrial sludges
are the same as those containing the stabi-
lized cores except that the cores are re-
placed with a layer of untreated industrial
sludge.
The control (MSW only) test cell pro-
file was constructed to resemble a core of
MSW taken from a municipal landfill of
medium density in the humid eastern United
States.
Test Cell Compositions
Waste Descriptions -
Table 1 shows the weights of co-disposed
material and MSW contained in the test
cells.
All MSW was obtained from non-
commercial collection routes in Warren
County, MS, in April 1978. A 500-kg com-
posite sample of MSW was sorted at the time
the test cells were filled and the result-
ing percent composition of the MSW used in
the study is given in Table 2. Table 2
also presents composition data from other
leaching studies.
Three industrial sludges were selected
from the five studied by Mahloch et al8.
This study used the same sludge sources and
sludge numbering system used by Mahloch.
Sludge selection was based on availability,
potential hazard if co-disposed with MSW,
and the potential applicability of chemical
stabilization as a disposal pretreatment
process.
Sludje 200 is an electroplating waste.
The slud;»e generator discharges separately
treated wastes from plating, phosphatizing
and metal cleaning operations to a series
of lagoons. Solids settle in the first
TABLE 1. TEST CELL CONTEMS
Untreated
Test Cell
No.
19
20
21
Sludge
Type
MSW
Kg, Dry Wt
200
800
900
2035
2041
2039
Co-disposed Material
Kg, Dry Wt
704
1678
1303
Process A
14
17
18
200
800
900
2038
2036
2045
1951
2609
2026
Process B
11
12
13
200
800
900
2028
2043
2031
2371
2302
2282
MSW only
15
16
2038
2096
None
None
226
-------
TABLE 2. PERCENT COMPOSITION OF MUNICIPAL SOLID WASTE
USED IN THIS AND OTHER INVESTIGATIONS
Category
Paper
Metal
Plastics
rubber,
leather
Glass
Textiles
Disposable
diapers
Food waste
Wood
Garden waste
Ash, rock,
dirt, fines
This
Study
44.79*
10.82
9.03
7.61
3.08
2.68
0.94
0.49
0.41
20.15
Jackson &
Streng (5)
40.53
8.29
6.52
7.42
4.19
1.78
7.53
0.86
15.32
5.48
Chian
and others (6)
36.5
14.7
2.8
6.8
0.7
—
14.4
—
3.1
14.9
Eifert &
Swartzbaugh (7)
49.6
9.5
6.0
12.0
3.2
1.4
7.3
—
4.6
5.4
* All percentages are on a dry wt. basis.
lagoon which is allowed to evaporate every
summer so that dried sludge can be removed
to a confined disposal area. Chromium and
copper are the primary contaminants in this
bright green, clay-like material.
Sludge 800 is chlorine production
brine waste. The sludge generator manu-
factures chlorine gas and caustic soda in a
mercury cell process. Sludge from clari-
fiers in the chlorine and alkali operations
is removed to a confined disposal area.
The primary contaminants in the tan, granu-
lar material are mercury and various alkali
salts.
Sludge 900 is clacium fluoride/sewage
sludge generated by an electronic manufac-
turer. An acidic fluoride waste stream and
the generator's sewer separately enter the
company's wastewater treatment plant and
there undergo separate treatment. The
sludges that result from physical-chemical
treatment of the fluoride waste stream and
biological treatment of the sewage are com-
bined in a confined disposal area. The
primary contaminants in this light brown
material are calcium fluoride and domestic
sewage sludge.
Bartos and Palermo9 have reported on
the physical and engineering properties of
these sludges and chemical characterization
of the sludges is presented elsewhere
(Myers et al10).
Chemical Solidification/Stabilization
Processes -
Separate cement-based (Process A) and
pozzolan (Process B) processes were used to
treat the industrial sludges selected for
this study (Table 3).
Process A, a patented technology, uses
two additives to produce a hard, concrete-
like material. The additives and waste are
dispersed with water in a slurry prior to
casting and solidification.
Raw sludge was slurried with a speci-
fied amount of water in a rotary pan turbine
227
-------
TABLE 3. FORMULATIONS FOR SLUDGE SOLIDIFICATION/STABILIZATION
Process
Sludge
Additive/Sludge*
Added Watert
percent
200
800
900
1.630
0.64
0.90
52.5
13.9
15.8
B
200
800
900
3.26
2.14
5.44
0.1
6.9
18.5
* Dry weight basis.
+ As percentage of total dry weight.
mixer in 0.38 m3 batches. After the pro-
cessor was satisfied that all the lumps of
sludge had been dispersed and the suspen-
sion was homogeneous, the first ingredient
was added to the batch one scoop at a time.
A second ingredient was added in 40 kg bag
increments. After the additives were mixed
(usually 3-8 min), the discharge gate on
the bottom of the mixer was opened and the
treated sludge dumped into a 0.61 x 0.61 m
cylindrical cardboard (paraffin impregnated)
mold. The molds were mounted on wooden
pallets to facilitate removal by a fork
lift to a storage area for a curing time of
28 days. The molds were labeled and covered
with plastic after being placed in the
storage area. Additive to sludge ratios and
percentages of water used in the stabiliza-
tion of each sludge by Process A are listed
in Table 3.
Process B uses flyash and other addi-
tives to produce a pozzolan type product.
Raw sludge was loaded into the mixer in
the same manner as described for Process A.
After a predetermined weight of sludge was
in the mixer, required amounts of water
were added. While the mixer was running,
32 kg bags of the first additive were dumped
into the mixer, one bag at a time. Then
if additional water was needed to maintain
a slurry, more was added. Other Process B
additives were then added one scoop at a
time. After mixing for approximately 5 min
the discharge gate on the bottom of the
mixer was opened and treated sludge fell
into a pal Let-mounted mold. Treated sludge
was compac:ed in the mold in four layers
using a hand tamp. The surface of each
preceeding layer was scarified before the
next layer was packed. The top layer was
molded to uffect a dome shape for the top
of the cons. After compaction was finished
a plastic cover was taped over each mold
and the mo Lds were set in the storage area
to cure foT 28 days. Two molds were filled
per batch. Sludge weights, additive
quantities, and amounts of water used in
the stabilLzation of each sludge by Process
B are listud in Table 3.
After i;he specified 28-day curing time
eight molds of each sludge type for each
process we^re selected from the available
cores. Fo:r selected cores the molds and
bottom pallets were removed. During the
operation i:he cores were examined for soft
spots, fisisures, scars, and other obvious
signs of poor stabilization or damage. None
were found. For loading into the test
cells the cores were rigged with nylon
nets fastened to iron lifting rings.
Leachate Collection and Analysis
After loading was completed on April
16, 1978, i:he test cells were sealed and
leaching bugan. Infiltration in each test
cell is simulated by applying 66 cm of
deionized water per year (at a rate of 1.27
cm per week). The details of the loading
228
-------
operation, provisions for leaching and
sample collection, and measurement of cell
temperature and redox potential are des-
cribed in a construction report.10
Separate leachate samples fro^ ports
above and below the soil layer were col-
lected. Samples collected on a monthly
basis were analyzed for 10, select para-
meters and samples collected on a quarterly
basis were analyzed for the entire list of
29 parameters shown in Table 4.
Monthly samples were pumped from the
above soil leachate port with a peristaltic
pump and were prepared for analysis as
given in Table 4. Quarterly samples were
collected in a similar manner except that
they were prepared in a glove box under an
inert atmosphere without oxygen reaching
the leachate. This procedure was found
necessary in order to avoid sample deteri-
oration between the time of sample collec-
tion, subsample preparation, and chemical
analysis.
TABLK 4. SAMPLE WORKUP
Subsample
Type
Preparation*
Alkalinity, pH,
conductivity
Chloride
Mercury
Total organic carbon
Cd, Cr, Fe, and Zn
Biochemical oxygen
demand (5 day)
Chemical oxygen
demand
Total volatile acids
Fluoride
TKN, TP
Al, As, B, Be, Ca,
Cu, Pb, Mg, Mn,
Ni, K, Se, Na
Monthly and quarterly
Monthly and quarterly
Monthly and quarterly
Monthly and quarterly
Monthly and quarterly
Quarterly only
Quarterly only
Quarterly only
Quarterly only
Quarterly only
Quarterly only
No preparation; analyze
immediately
Store at 4°C, analyze within
48 hr
No preparation; initiate analysis
with 4 hr by sealing sample
in digestion vessel
No preparation; initiate analysis
with 4 hr by sealing sample in
digestion ampule.
Dilute 100 ml to 250 ml and pre-
serve with 2.5 ml of Ultrex
HNO • analyze within 30 days.
Dilute 1 ml to 500 using dis-
til led water.**
Dilute 0.5 ml to 50 ml using dis-
til led water.**
To pH 21.0 using H?SO, and Thymol
blue indicator; store at 4°C.
analyze within 48 hr.
No preparation; analyze within
8 hr.
Dilute 10 ml to 100 ml; preserve
with I ml of H2SO,. Analyze
within 48 hr.**
Included in Cd, Cr, Fe, and Zn
subsample; analyze within
90 davs.**
*
**
All quarterly subsamples are collected under a He atmosphere.
Performed under He atmosphere.
229
-------
Alkalinity, pH, conductivity, BOD, and
COD were run immediately after sample
collection. Total volatile acid (TVA) and
chloride samples were analyzed within 48
hours of sample collection. Mercury, flu-
oride and total organic carbon (TOG) analy-
ses were performed within 4 hrs of sample
collection. Metals were preserved by
adding Ultrex nitric acid to 1 percent
acid by volume and analyzed within 60 days.
The total Kjeldahl nitrogen and total
phosphorous subsamples were preserved with
sulfuric acid and analyzed within 48 hrs.
The methods used in analyzing the prepared
leachates are summarized in Table 5.
To obtain assurance within the ana-
lytical program a quality control program
including replicate determinations, spiked
samples, and referee samples was imple-
mented. Internal control consists of
replicate determinations and spiked addi-
TABLE 5. METHODS OF ANALYSIS
Parameter
Procedures and/or instrumentatisn*
Lowest reporting
concentration
(ppm)
Specific
Conductance
ALK
Cl
YSI model 31 conductivity bridge
Standard acid titrationt
Destruction of organics using ^SOit and ^02;
0.5
25
10
ymhos
TKN
TP
TOC
Ca
Fe
K
Mg
Mn
Na
Al
As
B
filter; potentiometric titration with AgN03-
Cl electrode, and Metrohm model E53ij
recording titratorj
Technicon II Auto Analyzer, Industrial Method 0.01
376-75W/B and 32974W/B#
Same as above 0.01
Determined with Oceanography International 1
Corp. Model No. 0524B TOC Analyzer
Determined with a Spectrametrics Argon Plasma 0.03
Emission Spectrophotometer Modell II
Same as above 0.003
Same as above 0.05
Same as above 0.03
Same as above 0.001
Same as above 0.03
Same as above 0.01
Determined with a Perkin-Elmer Heated Graphite 0.001
Atomizer Atomic Absorption Unit
Determined with a Spectrametrics Argon Plasma 0.02
Emission Spectrophotometer Model II
(Continued)
* Mention of trade names or commercial products does not constitute endorsement
or recommendation for use.
t Standard Methods for the Examination of Water and Wastewater, American Public
Health Association, New York, 14th Edition, 1975.
| Compilation of methodology for measuring pollution parameters of landfill
Leachate. E.S.K. Chain and F. B. Dewalle. EPA-600/3-75-011 October 1975.
# Technicon Industrial Systems, Tarrytown, New York.
230
-------
TABLE 5. (Concluded)
Parameter
Procedures and/or instrumentation
Lowest reporting
concentration
(ppm)
Be
Cd
Cr
Cu
Hg
Ni
Pb
Se
Zn
BOD5
COD
TVA
pH
F
Same as above 0.0005
Determined with a Perkin-Elmer Heated Graphite 0.0003
Atomizer Atomic Absorption Unit
Same as above 0.003
Same as above 0.003
Determined by standard cold vapor techniquet 0.0002
Determined with a Perkin-Elmer Heated Graphite 0.005
Atomizer Atomic Absorption Unit
Same as above 0.002
Same as above .005
Determined with a Spectrametrics Argon Plasma 0.014
Emission Spectrophotometer Model II
Standard 5-day Biochemical Oxygen Demand, 2.0
unseededt
Standard dichromate/H2SO^ refluxt 50.0
Standard column partition chromatographic methodt 30.0
Electrometric —
Techicon II Auto Analyzer, Industrial Method 0.04
129-71-W
Standard Methods for the Examination of Water and Wastewater, American Public
Health Association, New York, 14th Edition, 1975.
tions to representative samples. Analysis
by standard addition using three separate
levels of addition, each addition level
run in triplicate, was used to assure
quality results from the heated graphite
atomizer atomic absorption unit. The
interlaboratory control program consists
of analysis of performance evaluation
samples submitted by the Environmental
Monitoring and Support Laboratory, U.S.E.P.
A. Cincinnati, OH.
RESULTS AND DISCUSSION
The results from the simulation of the
co-disposal of MSW with untreated indus-
trial waste and with chemically stabilized
industrial waste represent a 9 month leach-
ing period. The data for the quarterly
sample analyses represent 3 sampling
periods. Leaching data for the monthly
parameters analyzed represent 9 sampling
periods. The rationale for selecting the
parameters for monthly analysis was based
on the usual composition of leachate (Fe,
Cl~, TOC) and the individual composition of
the industrial waste being leached. Due to
the preliminary nature of the data, no
statistical analysis was attempted. For
purposes of this report, an attempt was
made to directly compare the limited data
available for each test cell. To provide
illustrations of these comparisons, the
results from the monthly analyses are
presented in graphs as cumulative mass
releases. In this manner, a preliminary
comparison of the leaching data can be made
for the untreated, treated, and control
test cells.
Leachate Quality from Control Test Cells
Factors influencing leachate quality
include degree of compaction of refuse, rate
of infiltration, moisture content, composi-
tion and processing of waste, and the time
231
-------
over which decay has occurred. Table 6 analyses from the control (MSW only) test
compares the range of values for leachate cells in this study to those summarized by
TABLE 6. COMPOSITION OF LEACHATES FROM CONTROL TEST CELLS
Parameter*
Age (months)
COD
BOD
TOC
pH
TVA
Specific Conductance
Alkalinity
TKN
TP
Cl
F
Al
As
B
Be
Ca
Cd
Cr
Cu
Fe
Pb
Mg
Mn
Na
Ni
K
Se
Zn
Hg
ORPt
Control Test
Cells in
This Study
8-17
325,400-76,300
13,000-25,500
7,000-12,000
4.35-5.55
38-12,000
6,000-9,800
1,750-3,820
711-902
13.6-39.2
867-1442
<0. 01-2. 12
1.69-6.08
<. 001-0. 045
6.00-8.38
<.005
655-1,118
0.004-0.028
0.089-0.338
0.004-0.030
229-362
<0. 001-0. 082
13P-178
11.0-18.6
712-942
0.268-0.482
357-615
0.078-0.160
6.75-21.7
<0. 0002-0. 0030
(-450)-(-73)
Chisn and
DeWsllet James
12-193 24-240
40-89,520 79-30,933
81-33,600
256-2,800
3.7-3.5
_.
2,810-16,800 2,050-13,500
0-20,850
0-:.,106 (NH3)
0-] 30
4. 7-2', 467
—
__
__
__
—
60-7,200
0.03-17 0.001-0.073
0.01-0.29
0-9.9 0.004-1.54
0-2320 6.5-1,500
<0. 1-2,0 0.01-0.33
17-15,600
0.09-1?5
60-7,700
—
28-3,770
0.001-0.33
0-370 0.07-33.0
0.0006-0.16
(-220)-(+163)
.
* All values in mg/S, except specific conductance which is measured as micromhos
per centimeter, pH in Ph units, and ORP in millivolts.
+ 23 leachates analyzed by Chi an and
232
-------
Chian^and DeWalle11 and to those listed by
The ranges of analytical results
James
12
from the control test cells are consistent-
ly within the previously reported ranges
for all constituents.
The leachate analyses reported in this
paper were obtained from simulated landfill
test cells during the period from 8 to 17
months after sealing. The leachate samples
are representative of young landfills. The
highest concentrations for BOD, COD, TOG,
TVA, and alkalinity were measured in the
first and second quarterly samples. These
data are typical for young landfill lea-
chates as reported by other investiga-
tors11'12.
Metal concentrations (with the except-
ion of zinc, lead, and beryllium), OKP,
and specific conductance have all tended
to fluctuate randomly. Zinc and lead
concentrations have declined from initial
high values (Table 6) to near the lower
limit of the reported zinc and lead con-
centration ranges. To date, Beryllium
has not been present in detectable
amounts in any of the control test cell
leachates.
Cumulative Mass Releases of Contaminants
Comparisons of the cumulative mass of
contaminants leached from the various test
cells are based on data normalized to the
weight of MSW in the test cell. Increases
in the total mass of contaminant leached
per unit weight of MSW from the co-
disposal test cells over that leached from
the controls can be assummed to be due to
release of contaminants from the material
co-disposed with the MSW. Graphs comparing
the total masses of leached contaminant per
kilogram of MSW for the untreated and
treated co-disposal test cells and the
controls are presented as Figures 4 through
11. The plots for the controls are average
values for the two control test cells.
Plots for the treated sludges were develop-
ed for Process A only due to the limited
data available for Process B treated
sludges. Table 7 and 8 show the cumula-
tive mass releases for the parameters
measured on a monthly and on a quarterly
basis, respectively. Plots of pH for
various leachates are presented as
Figure 12.
Electroplating Waste (Sludge 200) -
Of the 17 metals analyzed for the test
cell containing untreated sludge 200 and
MSW, boron, calcium, magnesium, nickel,
and sodium, have mass releases above con-
trol cell levels. Releases of chromium
and copper, which are characteristic of
Sludge 200, were not detected in concentra-
tions above the control cell releases. The
cumulative mass releases of boron, copper,
nickel, potassium, selenium, and zinc for
the chemically stabilized Sludge 200 were
greater than those released by the control
cells (Table 8). The pattern of Cumula-
tive release of TOC (Figure 7) and other
demand parameters (Table 8) indicates that
neither the co-disposal of untreated or
treated Sludge 200 significantly alters
the character of the MSW leachate. Also,
little difference was observed in the pH
of the untreated and treated Sludge 200
when compared with the controls (Figure 12).
Chlorine Production Brine Waste (Sludge 800) -
Numerous metals, including As, Al, Ca,
Cd, Cu, Hg, K, Mg, Mn, Ni, Na, Pb, and Zn
have been leached from untreated Sludge 800
in appreciable quantities above those
found for the control cells.
As anticipated, significant quantities
of Cl have also been leached from the un-
treated sludge (Figure 5). Figures 8 and 9
show the mass releases of Cd and Hg, res-
pectively. The only metal from untreated
Sludge 800 showing no measurable increase
over the control test cells is iron.
The relative effectiveness of chemi-
cal stabilization on Sludge 800 (compared
to the untreated and control test cells)
is shown in Figures 5, 8, 9, and 10. The
mass leaching of toxic metals such as Cd,
Hg, and Cr have been significantly reduced.
The only rirrnulative mass releases of metals
still above the control test cells after
chemical stabilization are for Ca, K, and
Na.
Little difference has been observed in
the leaching of TP, TKN, COD, BOD, TVA
(Table 8), and TOC (Figure 7) between the
Sludge 800 co-disposal cells and the con-
trol cells. The pH of the untreated,
treated and control cell leachates have
been virtually the same during the initial
9 months of leachate production (Figure 12).
Calcium Fluoride/Sewage Waste (Sludge 900)-
Cumulative mass releases of Cd, Cu,
Mg, Mn, and Se higher than the control test
233
-------
300
S.
2
o
200
£
*
u
o»
1
20
CO
o>
10
0
U= Untreated
T= Treated
.
-
U T
ill
U
"7"| U T .
200 800 900 CONTROLS
Figure 5. Cumulative mass release of Chloride.
234
-------
30
c
N
CO
10
12.0..
CO
6.0
|
3.0
1.5
LEGEND
U= Untreated
T= Treated
U
U
U
200 800 900 CONTROLS
Figure 6. Cumulative mass release of Zinc.
LEGEND
U= Untreated
T= Treated
U
U
U
200 800 900 CONTROLS
Figure 7. Cumulative mass release of Total Organic Carbon.
235
-------
200 _
150
•o
O
o»
I
gioo
*
50
0
U = Untreated
T= Treated
-
-
-
"r
U
2, FU, _
200 800 <)00 CONTROLS
Figure 8. Cumulative mass release of Cadmium.
LEGEND
i
20
o»
X
o»
£
15
(O
o»
^ 10
5
0
U = Untreated
T = Treated
-
-
-
U T
U
T U T
200 800 900 CONTROLS
Figure 9. Cumulative mass release of "tercury.
236
-------
240
200
i_
0
o>
3.
1
150
jglOO
o>
XL
50
0
r LEGEND
U= Untreated
T = Treated
U
T
U
T
U
*
T
200 800 900 CONTROLS
Figure 10. Cumulative mass release of Chromium.
O.I5_
o»
E
0.10
0.05
LEGEND
U = Untreated
T= Treated
U
U
200 800 900 CONTROLS
Figure 11. Cumulative mass release of Fluoride.
237
-------
TABLE 7. CUMMULATIVE MASSES LEACHED PER KG MSW
(MONTHLY PARAMETERS)
Sludge 200
Parameter
Alkalinity (mg)
Chloride (g)
TOG
Cd
Cr
Hg
Fe
Zn
(mg)
(VJg)
(Pg)
(mg
(mg)
(mg)
Controls Untreated Treated
1.14
0.490
29.3
3.94
119
0.6
147
6.8
1.53
0.587
24.9
3.63
130
0.4
100
3.9
1.66
0.605
31.1
4.19
120
0.4
101
14.5
Sludge 800
Untreated Treated
1 . 50
30.11
38.1
199
193
21.7
78
21.1
TABLE 8. CUMULATIVE MASSES LEACHED
1.65
0.284
36.6
7.11
86
0.4
Sludge 900
Untreated Treated
138
PER
5.1
KG MSW
1.73
0.533
11.6
24.7
75
0.4
4
3.7
4.
1.
51.
7.
240
0.
296
19.
08
30
3
25
7
2
(QUARTERLY PARAMETERS)
Parameter
Al (mg)
As (yg)
B (mg)
Be (vg)
Ca (mg)
Cu (Mg)
Pb (pg)
Mg (mg)
Mn (mg)
Ml (ug)
K (mg)
Se (pg)
Na (mg)
TP (mg)
TKN (mg)
BOD (kg)
TVA (g)
COD (kg)
F" (ug)
Controls
1.13
8.25
2.8
<10.0
301.0
6.3
7.51
71.1
6.86
174.0
251.0
49.3
400.0
9.4
381.0
9.33
1.88
14.90
75
Sludge
Untreated
0.906
<12.0
10.33
<6.0
743.0
5.65
2.26
163.0
7.78
263.0
290.0
35.7
673.0
4.1
346.0
8.97
1.49
15.60
61
200
Treated
0.525
<11.0
5.43
<6.0
454.0
12.9
4.64
90.1
6.92
266.0
424.0
87.1
536.0
8.79
681.0
7.92
1.59
17.93
26
Sludge
Untreated
15.1
55.5
5.18
<7.0
751.0
106.0
88.5
235.0
12.94
490.0
512.0
106.0
2130.0
8.01
495.0
10.88
1.23
19.20
<20
800
Sludge
Treated
0
<12
3
<11
493
6
4
83
8
192
416
71
1810
11
406
8
1
15
14
.824
.0
.97
.0
.0
.14
.17
.6
.15
.0
.0
.4
.0
.08
.0
.53
.15
.24
Untreated
0
<8
4
<8
370
25
<0
142
15
122
242
71
213
4
309
5
1
10
102
.848
.8
.54
.0
.0
.0
.88
.0
.0
.0
.0
.9
.24
.0
.59
.10
.12
900
Treated
0.971
12.3
8.69
<17.0
870
9.17
4.99
177
14.59
340
734
66.3
944
17.29
885
21.54
1.94
29.90
30
238
-------
LEGEND
pH 6
4
200 O Untreated
• Treated
800 D Untreated
• Treated
900 A Untreated
Treated
Controls
A
X
Figure 12.
34567
MONTHS
Variation in leachate pH values.
8
cells have been observed for the untreated
Sludge 900 co-disposal test cell. Leaching
of Na and Fe have been noticably lower than
the control cell values. The remaining
10 metals have been leached at approxi-
mately the same levels as that for the
control test cells. The chemically
stabilized or treated Sludge 900 co-
disposal cells have shown a cumulative
release of B, Ca, Cr, K, Mg, Mn, Ni, and
Zn above the control cells. These cumula-
tive mass releases are probably related to
an unexplainable increase in leachate
production from the test cells containing
the treated Sludge 900. Beginning 5
months after the initiation of leachate
production, leachate quantities increased
dramatically to an amount almost equiva-
lent to the amount of infiltrate applied.
Checks are being made to ascertain whether
the test cells have received more water
than originally scheduled or if there
exists an anomally in the leaching charac-
teristics of the test cells. Figure 11
shows the cumulative fluoride leached from
the treated waste test cells. Cumulative
mass releases for both the untreated and
treated Sludge 900 co-disposal cells were
higher than for the control cells. Actual
concentrations of fluoride released have
been less than 3 mg/X. Chloride levels
released from the treated waste are near
the control levels (Figure 5).
TOG (Figure 7) and other demand para-
meters (Table 8) for the untreated Sludge
900 have been considerably less than that
shown for the controls; whereas, the same
parameters from the treated sludge co-
disposal cells have been at or above the
control levels.
The pH of the leachate from the un-
treated Sludge 900 has risen from 5.0 to
7.2 (Figure 12). The significant rise in
pH and concurrent reduction in demand para-
meters tend to indicate an accelerated rate
of stabilization of test cell contents, per-
haps due to overwatering of the cells or to
the availability of the organics from the
sewage sludge.
CONCLUSIONS
Contaminant characteristics of leachate
generated from the control test cells in
this study are consistent with data report-
ed elsewhere.
Release of the major metal contaminants
239
-------
of electroplating sludge when co-disposed
with MSW was not observed.
Untreated chlorine production brine
sludge when co-disposed with MSW, releases
significant quantities of Al, Cd, Cu, Cl,
Hg, Na, and other dissolved solids. Chemi-
cally stabilized chlorine production brine
sludge when co-disposed with MSW signifi-
cantly reduces the mass release of toxic
metals (e.g. Cd, Hg, and Cr) as well as
chlorides.
To date the co-disposal of industrial
wastes with MSW has not significantly
affected the traditional leachate pollu-
tion indicies of BOD, COD, TOC, alkalinity,
pH, and iron with the exception of the
untreated calcium fluoride/sewage sludge.
ACKNOWLEDGMENTS
This study is part of a major research
program on chemical treatment technology,
which is now being conducted by the U. S.
Army Engineer, Waterways Experiment Sta-
tion and funded by the Environmental Pro-
tection Agency, Municipal Environmental
Research Laboratory, Solid and Hazardous
Waste Research Division, Cincinnati, Ohio,
under Interagency Agreement, EPA-IAG-D4-
0569. Robert E. Landreth is the EPA
Program Manager for this research.
REFERENCES
1. Newton, J. R. 1977. Pilot Scale
Studies of the Leaching of Industrial
Wastes in Simulated Landfills. Wat.
Pollut. Control. Vol 76, No. 4, pp.
468-480.
2. Streng, David R. 1977. The Effects
of Industrial Sludges on Landfill Lea-
chates and Gas. In Banerji, S. K. (ed.)
Management of Gas and Leachate in Land-
fills. Proceedings of the Third Annual
Municipal Solid Waste Research Sympos-
ium. EPA-600/9-77-02b. U. S. Environ-
mental Protection Agency, Cincinnati,
OH. pp. 41-54.
3. Swartzbaugh, J. T., Hentrich, R. L. and
Sabel, G. V. 1978. Co-disposal of
Industrial and Municipal Waste in a
Landfill. In_ Shultz, D. W. (ed.) Land
Disposal of Hazardous Wastes. Pro-
ceedings of the Fourth Annual Research
Symposium. EPA-600/9-78-016. U. S.
Environmental Protection Agency,
Cincinnati, OH. pp. 129-151.
4. Walsh, James J. and Kinman, Riley N.
1979. Municipal Solid Waste: Land
Disposal, Proceedings of the Fifth
Annual Research Symposium. EPA 600/
9-79-023a. U. S. Environmental Pro-
tection Agency, Cincinnati, OH. pp.
41-57.
5. Jackson, A. G. and Streng, D. R. 1976.
Gas and Leachate Generation in Various
Solid Waste Environments. In Gas and
Leachate from Landfills: Formation
Collection, and Treatment. EPA 600/
9-76-004, U. S. Environmental Protect-
ion Agency, Cincinnati, OH.
6. Chian, E. S. K., DeWalle, F. B. and
Hammerbert, E. 1977. Effect of Mois-
ture Regime and Other Factors on Muni-
cipal Solid Waste Stabilization. In
Banerji, S. K. (ed.), Management of
Gas Leachate in Landfills. EPA-600/9-
77-026, U. S. Environmental Protection
Agency, Cincinnati, OH. pp. 73-86.
7. Eifert, M. C. and Swartzbaugh, J. T.
1977. Influence of Municipal Solid
Wastes Processing on Gas and Leachate
Generation. In Banerji, S. K. (ed.)
Management of Gas and Leachate in
Landfills. EPA-600/9-77-026, U. S.
Environmental Protection Agency, Cin-
cinnati, OH. pp. 55-72.
8. Mahloch, J. L., Averett, D. E., and
Bartos, M.J.,. Jr. 1976. Pollutant Pot-
ential of Raw and Chemically Fixed
Hazardous Industrial Wastes and Flue
Gas Desulfurization Sludges. Interim
Report. EPA-600/2-76-182, U. S. En-
vironmental Protection Agency, Cincin-
nati, OH. 117 pp.
9. Bartos, M. J., Jr. and Palermo, M. R.
1977. Physical and Engineering Prop-
erties of Hazardous Industrial Wastes
and Sludges. EPA-600/2-77-139, U. S.
Environmental Protection Agency, Cin-
cinnati, OH. 89 pp.
10. Myers, T. E., Malone, P. G., and Duke,
J. C. The Effect of Raw and Treated
Sludge on Leachate Quality in Land-
fill Environments and Gas Production
Rates and Compositions in a Sanitary
Landfill Environment: Construction
Report. USAE Waterways Experiment
Station, Vicksburg, MS. Draft in
240
-------
preparation.
11. Chian, E. S. K. and DeWalle, F. B. 1976.
Sanitary landfill leachates and their
treatment. ASCE, Jour. Env. Eng. Div.
102:EE2.
12. James, Stephen C. 1977. Metals in
municipal landfill leachate and their
health effects. AJPH, 67:5.
241
-------
STABILIZATION AT MUNICIPAL LANDFILLS CONTAINING INDUSTRIAL WASTES
Frederick G. Pohland and Joseph P. Gould
School of Civil Engineering
Georgia Institute of Technology
Atlanta, Georgia 30332
ABSTRACT
The technical feasibility and merit of codisposal of municipal and industrial wastes in
landfills are discussed with specific emphasis on the fate of heavy metals under the
influence of leachate containment, collection and recycle. The results of both pilot-
scale and lysimeter-type landfill investigations are utilized to describe prominent bio-
chemical and physical-chemical phenomena within the landfill environment, to illustrate
probable reaction mechanisms, to develop a control strategy and to indicate a need for
some flexibility in regulations governing landfill management options.
Utilizing residential-type solid wastes alone and in conjunction with a metal plating
sludge, data have indicated the possibility for significant reductions in leachate heavy
metal concentrations and the key roles of redox potential, pH and sulfides, hydroxides
and hydroxy-carbonates in sequestering metals. Present studies are being extended to
determine limits of heavy metal loadings that can be sustained during codisposal without
deleterious effects on the biochemical processes of stabilization within the landfill,
and/or the identity of internally produced or externally added mediators which are
capable of minimizing these effects.
INTRODUCTION
The production of leachate at land-
fills has become a sensitive environmental
issue with respect to potential damage to
ground and surface water resources. This
potential for adverse environmental impact
will vary depending on leachate quantity
and quality as well as the opportunity of
migration through and interaction with the
surrounding soil or geologic formation to
which the leachate is exposed. Moreover,
the nature of the waste deposited and its
relative chemical stability collaborate
to determine leachate characteristics at
any time. Therefore, the amount and
quality of leachate produced is not only
site specific but time dependent thus
voiding attempts to use sweeping gen-
eralizations in interpretive analysis of
leachate problems.
It is the purpose here to review and
discuss some of the phenomena occurring
during waste stabilization within landfills
as promoted by the influx of moisture and
controlled by containment, collection and
recycle of leachate. Initial emphasis
will be placed on the progress of stabi-
lization of municipal residential-type
wastes; this emphasis will then be ex-
tended to include considerations for co-
disposal of residential with industrial-
type wastes.
FORMATION AND CHARACTERISTICS OF LEACHATE
Regardless of waste type, leachate
will be formed when the liquid holding
capacity of the landfill is surpassed. In
the case of most municipal landfills,
leachate is produced when infiltration from
rainfall, surface drainage and/or ground-
water intrusion combine with the inherent
moisture content of the waste to exhaust
its liquid holding capacity. The rami-
fications of such leaching will differ
based primarily upon the type of waste
being handled and the location at which
it has been deposited.
242
-------
Leachate from Residential-type Solid Waste
For a residential-type solid waste
disposal site exposed to leaching oppor-
tunities and operated to permit leachate
containment, collection and recycle, it is
possible to generate data capable of re-
flecting changes in leachate quality with
time. Such information has been presented
for lysimeter experiments^"^ as well as
for larger pilot-scale studies. The
latter efforts are exemplified by recent
studies at the Georgia Institute of
Technology employing two large (3-m square
by 5.2-tn deep) lined cells containing
shredded municipal solid wastes manually
compacted to a density of 319 kg/nr. Final
compacted depths of solid waste were 2.75m
and 2.6m in the open and sealed cells,
respectively. Moisture from incident rain-
fall was allowed to enter the open cell
directly; the other cell was sealed to
permit gas measurements and thus received
an equivalent amount of tap water at each
rainfall event. The other construction
and operational features of the two land-
fill cells are illustrated in Figure 1 and
the characteristics of the shredded wastes
placed in each are included in Table 1.
Preliminary results of these studies
have been presented previously-* and have
since been extended to include a total test
period of 699 days. During this time, the
cells reached field capacity, leachate was
collected and recycled, and analyses were
performed on both leachate and gas samples.
Although some leaching occurred initially
in both cells on an intermittent basis due
to differential moisture movement and up-
take within the solid waste mass, it did
not appear regularly until 383 days after
the waste had been placed. Therefore,
about one year was required to promote
continuous leaching from the cells.
It is informative to compare the
leachate analyses at the time when con-
tinuous leaching (and recycle) began to
those at and after the onset of methane
fermentation. The analyses for the open
cell for these times are indicated in
Table 2 and are considered typical of the
longer-term changes to be anticipated in
leachate quality at a municipal landfill
subject to leaching and having received
residential-type solid wastes. In other
words, a site will normally accumulate
moisture to field capacity, develop an
active microbial community, and release
leachate commensurate with influx of
moisture from rainfall (or groundwater).
The leachate initially reflects develop-
ment of acid fermentation with low pH,
high organic content (COD, BODs, TOC and
TVA), a relative abundance of mobilized
ions, and obvious pollutional character-
istics. Usually by this time, any free
oxygen originally present has been de-
pleted, temperatures have stabilized at
near ambient and the aqueous environment
has become chemically reduced. It is
during this period also that some
hydrogen will be released in the gas phase,
nutrients will begin to be depleted in
support of microbial growth and combined
oxygen species will serve in the oxidation
of complex organic materials to simpler
intermediates such as the volatile
organic acids. The intensity and/or
magnitude of the analytical parameters will
vary but the trends will be essentially
the same from site to site.
In time, methane fermentation initially
restricted to very localized areas, will
TABLE 1. CHARACTERISTICS OF SHREDDED RESIDENTIAL-TYPE SOLID WASTES USED IN
PILOT SCALE INVESTIGATIONS WITH LEACHATE RECYCLE
Parameter
Average Analysis*
Density as placed, kg/m
Moisture content as placed, %
Total Volatile Solids, % dry
Carbon, % dry
Hydrogen, % dry
Nitrogen, % dry
Heating Value, kJ/kg dry
319
33.5
75.5
45.3
5.46
3.33
18,045
*Average of six representative samples.
243
-------
Tempertturt
Prob*
Letchtte DIttrlbutlon Sytttm
Git Port /
Letchtte
Collection
Sytt ten
Rtcyclt Lint
Gtt
Collection
Network
Pltchtrgt
sLmpt
PLAN OF SIMULATED SANITARY LANDFILLS
(not to tcilt)
61 cm Covtr Soil
Concrtte
Block Wtll
Gtt Ports
0.84 cm Sttel
^ Lid
Gtt Collection
30 mil Hy pt I on
Llntr
SECTION VIEW: SEALED CELL (not to tctle)
Figure 1. Pilot-Scale Landfill Cells with Leachate Recycle
244
-------
TABLE 2. CHANGES IN LANDFILL LEACHATE CHARACTERISTICS FROM A
RESIDENTIAL-TYPE SOLID WASTE
Leachate Composition
Leachate Parameter
PH
Total Alkalinity, mg/1 CaCO
COD, mg/1
BOD , mg/1
TOC; mg/1
Total Volatile Acids (TVA), mg/1 HAc
Total Kjeldahl Nitrogen, mg/1 N
Ammonia Nitrogen, mg/1 N
Total Phosphorus, mg/1 P
ORP, mv E
Specific Conductance, ymho/cm
Chloride, mg/1
Sulfate, mg/1
Calcium, mg/1
Magnesium, mg/1
Manganese, mg/1
Sodium, mg/1
Potassium, mg/1
Iron, mg/1
Zinc, mg/1
Copper, mg/1
Cadmium, mg/1
Lead, mg/1
Aluminum, mg/1
Chromium, mg/1
Initial*
5.3
8,300
40,210
20,437
10,938
15,611
825
645
4.93
_
11,850
862
750
2,850
264
73.7
1,175
1,070
900
2.65
ND
0.2
ND
ND
0.4
Intermediatet
6.1
4,283
9,464
5,800
2,957
4,710
374
111
0.83
-15
6,250
505
110
360
153
1.0
650
611
157
0.1
ND
ND
ND
ND
ND
Final//
6.7
1,338
350
88
307
ND
20
9
0.27
-91
1,080
178
ND
260
71
1.6
515
311
39
0.5
ND
ND
ND
ND
0.3
*After onset of acid fermentation with daily recycle.
tAt onset of methane fermentation with daily recycle.
//After methane fermentation with daily recycle.
ND = none detectable.
spread throughout the landfill mass. In
conventional landfills, this will be a
relatively slow process but with leachate
recycle, it will be accelerated due to
the benefits of distribution of methane
formers and nutrients throughout the
landfill mass. Action of the methane
formers will deplete methane precursors
such as the volatile acids and hydrogen
and will increase the quantity of methane
gas produced. Reduction in volatile acids
will also promote an increase in pH and
an opportunity for reactions beneficial to
the immobilization of certain ions.
Finally, organic pollutant characteristics
will be essentially eliminated, the pH
will rise to near neutral, and with leach-
ate recycle, opportunities for physical-
chemical interactions such as precipi-
tation and filtration on and in the
remaining solid waste mass will be en-
hanced. Ultimate leachate concentrations
eventually detected will be a function of
dilution and/or opportunity for sustain-
ing continued biological activity within
the site.
The fact that anaerobic biological
stabilization effectively removes pol-
lutants from leachate is borne out by the
changes in parametric composition indi-
cated in both Table 2 and Figure 2.
Negative ORP values are indicative of re-
ducing conditions required by anaerobic
microorganisms and also are important with
respect to feasibility of reduction of
oxidized species and enhanced opportunity
for interaction, i.e., reduction of
sulfates to sulfides and subsequent
reaction with metals to cause their pre-
cipitation and removal from the leachate.
These latter reactions would be of par-
ticular advantage in landfills containing
potentially toxic heavy metals. The re-
245
-------
Time Since Leacnate Recycle Began, Days
Figure 2. Changes in Leachate Characteristics With Leacnate Recycle
246
-------
duction in the concentration of sulfates,
iron and other metals to the low or non-
detectable levels indicated in Table 2
could well have been a consequence of
such a condition.
Other environmental factors required
for support of active biological sta-
bilization of leachate constituents are
the availability of substrate and nutri-
ents. Normally, leachate from a recently
closed site will contain an abundance of
acceptable carbon species; the appearance
and subsequent conversion of volatile
fatty acids attest to this fact. The
overall disappearance of these acids and
other organic constituents (measured as
BOD5) with time, coupled with the usual
production of methane and carbon dioxide
as shown in Figure 2, also support the
premise that for residential-type solid
wastes, toxicity is not a problem, and
that most organic pollutants in leachate
are removed by anaerobic biological pro-
cesses. However, the data in Table 2
would also suggest the possibility of a
nutrient deficiency in terms of phosphorus
which may have permitted a residual con-
centration of BOD5 to linger. This
premise tends to be substantiated by the
indicated BOD5:N:P ratios of about
100:10:0.3. Therefore in this case,
although nitrogen was in abundance, phos-
phorus was probably limiting.
Leachate from Residential/Industrial Wastes
Analyses on leachate from the co-
disposal of residential and industrial
wastes will contain in addition to normal
leachate constituents, the leachable con-
tent of the industrial waste. As a con-
sequence, the industrial waste contribution
will act as a superimposition of constit-
uents over and above those normally pre-
sent without the influence of industrial
wastes. The effect will again be highly
time dependent and site specific. There-
fore, it is difficult to utilize and
translate leachate quality data from such
sites without prior information on the
many variables alluded to heretofore.
However, if the constituents and environ-
mental conditions are known, some pre-
dictions can be made concerning ultimate
impacts.
To demonstrate and provide a basis of
discussion of some of the phenomena deter-
mining the fate of metal species in land-
fills, lysimeter studies of the codisposal
of residential and industrial solid wastes
with leachate recycle have also been
initiated at the Georgia Institute of
Technology. Four lysimeter columns, used
in previous leachate recycle studies^,
have been refurbished to be operationally
similar to the units indicated in Figure 3.
In this study, residential-type municipal
solid wastes were received in bulk,
separated into categories, and analyzed
for the parameters indicated in Table 3.
At the same time, a treatment sludge from
the electroplating of zinc and chromium
and to a lesser extent, cadmium and
nickel was received and analyzed for its
metal content as indicated in Table 4.
One of the columns served as the control
TABLE 3. CHARACTERISTICS OF BULK RESIDENTIAL-TYPE SOLID WASTES AND SAWDUST
USED IN LYSIMETER INVESTIGATIONS
WITH INDUSTRIAL WASTE CODISPOSAL AND LEACHATE RECYCLE
Solid Waste
Category
Paper
Glass
Plastic and Leather
Diapers
Garden Debris
Food Wastes
Wood
Metals
Fines
Sawdust
Wet Wt.,
%
53.0
13.7
8.4
2.9
1.8
6.2
2.0
5.9
6.1
-
Moisture,
%
11.6
-
2.5
2.1
0.7
75.4
-
-
45.1
7.6
Volatile C,
Solids, % %
87.5
_
98.6
98.2
92.4
92.0
_
_
90.0
88.1
41.57
_
82.88
46.89
45.03
41.76
_
_
40.65
43.41
H,
%
6.76
_
8.89
6.04
5.90
6.61
_
_
5.47
5.94
N, Heating Value,
% kJ/kg dry
0.07
_
0.06
0.55
0.09
1.97
_
__
0.97
0.09
15,114
_
37,932
20,432
16,873
18,966
_
—
15,282
17,961
247
-------
0.91m
Leachate Distribution
System
-Top Soil —
Aggregate
Solid Waste/Industrial
Waste Mixture
Solid Waste
Leachate Recirculation
Line
Corrugated Steel
Pipe
Graded Aggregate
Conical Concrete
Base
Support Structure
TEST UNIT
Leachate Collection
Sump
CONTROL UNIT
Figure 3. Simulated Landfill Lysimeter Columns with Leachate Recycle
and received 400 kg of bulk residential
solid wastes and sawdust without any
metal sludge; the other three columns
received the same amount of residential
solid wastes together with 33.6 kg,
65.8 kg and 135.2 kg of metal sludge,
respectively. The solid waste and/or
metal sludge in each case was mixed with
37.3 kg of sawdust and placed in suc-
cessive layers throughout the columns as
they were being filled. The columns were
then covered with soil, brought to field
capacity with tap water and allowed to
leach as a consequence of the influx of
rainfall and the recycle of leachate back
through the columns.
Although the research investigations
are continuing, preliminary results have
indicated that acid fermentation has set
in, that the pH has been lowered, and that
certain ionic species have been transferred
to the leachate and appear in concen-
trations indicated in Table 5. Although
248
-------
TABLE 4. CHARACTERISTICS OF METAL SLUDGE USED IN LYSIMETER INVESTIGATIONS
WITH RESIDENTIAL-TYPE SOLID WASTES CODISPOSAL AND LEACHATE RECYCLE
Parameter
Analysis
Moisture Content, %
Volatile Solids, %
Zinc, rag/kg dry
Chromium, mg/kg dry
Nickel, mg/kg dry
Cadmium, mg/kg dry
Copper, mg/kg dry
Iron, mg/kg dry
317
21
84.7
24.6
000
000
400
13,100
185
94,000
Note: Metals determined after acid digestion.
TABLE 5. SELECTED INITIAL LEACHATE CHARACTERISTICS FROM LYSIMETER INVESTIGATIONS
OF CODISPOSAL OF RESIDENTIAL-TYPE SOLID WASTES WITH INDUSTRIAL WASTE
AND WITH LEACHATE RECYCLE
Analyses
Parameter
pH
Alkalinity, mg/1 CaC03
COD, mg/1
BOD5, mg/1
Total Volatile Acids, mg/1 HAc
Suspended Solids, mg/1
Dissolved Solids, mg/1
ORP, mv Ec
Sulfate, mg/1
Sulfide, mg/1
Zinc, mg/1
Chromium, mg/1
Nickel, mg/1
Cadmium, mg/1
Copper, mg/1
Iron, mg/1
Column 1
5.68
1,010
11,740
5,900
989
114
6,510
-100
34
1.2
3
0.05
0.10
ND
ND
25
Column 2
5.45
1,190
20,830
10,400
1,320
11
11,200
-80
218
0.03
500
0.28
1.1
1.7
ND
88
Column 3
6.08
1,370
13,450
6,000
874
6
7,830
-90
104
0.03
205
0.20
0.82
1.1
ND
68
Column 4
6.17
1,980
13,830
6,200
1,220
18
10,180
-80
109
0.03
390
0.12
1.2
2.0
ND
62
Column 1: Control without metal sludge.
Column 2: Test unit with 33.6 kg of metal sludge.
Column 3: Test unit with 65.8 kg of metal sludge.
Column 4: Test unit with 135.2 kg of metal sludge.
ND = Nondetectable.
there was some dissimilarity between some
of the initial analyses, probably con-
sequenced by some unavoidable differences
in the bulk wastes received by each
lysimeter column, some of these variations
will likely be moderated by continued
recycle operation. Moreover, because of
the differing amounts of metal sludge
added to the three test columns, some
variations in leachate characteristics
were anticipated. It is too early in the
investigation to speculate on the even-
tual changes in parametric concentrations
caused by the inherent biochemical and
physical-chemical processes within each
column, but it is anticipated that bio-
logically mediated stabilization will
continue at least at the lower metal con-
centrations with the eventual sequence of
events described previously. Accordingly,
249
-------
changes in analytical parameters similar
to those indicated in Table 2 should occur;
of particular interest will be the even-
tual fate of the higher metal concen-
trations with respect to opportunity for
removal under the combined invluences of
leachate recycle and stabilization.
GENERAL DISCUSSION
The results to date on the present
lysimeter studies, augmented by experi-
ences elsehwere, allow some speculation on
the fate of metals in landfill leachates
during biochemical stabilization. Of
particular interest are the pertinent
equilibria, the pH and ORP, and the
availability of various constituent species
for physical-chemical interaction.
It is known that when the impacts of
leachate production are most detrimental,
the pH is low, the chemical environment
has become reduced and a primary deter-
minant of the presence or absence of
various constituents will be their relative
solubilities in the leachate medium. In
a biochemically reduced environment
similar to that indicated in Table 2,
certain previously oxidized species will
be transformed to their reduced counter-
parts. For example, nitrates and sulfates
will typically be reduced to ammonia and
hydrogen sulfide, respectively. These
species ionize in accordance with known
equilibrium relationships, the former as
a weak base, the latter as a weak acid.
As such, they contribute to the buffer
capacity and help to moderate the pH.
In many landfills, the sulfate-
sulfide redox system is of additional
importance because of the role it has to
play in the eventual reaction with and
removal of metals. Typically, when
leachate pH and ORP are low, sulfates will
have been reduced to sulfides and the fate
of many metals will be determined by
relative concentrations and the solu-
bilities of the metal sulfide products.
Figure 4 demonstrates the anticipated
distribution of S04~2- S~2 species as
determined by ORP (EH = Ec - 282mv) and
pH^. The boundary limits for each region
of the figure represent loci of equi-
molar concentration of the two indicated
species separated by the respective
boundaries. It is important to note that,
since the reduction of sulfate to sulfide
=-2
requires eight electrons per mole, a
change of as little as 12mv toward
reducing potentials will cause a 10-fold
increase in the ratio of sulfide to
sulfate species. Moreover, since heavy
metal sulfides tend to be so sparingly
soluble, even at acid pH values wherer
HS or H2S predominate, sufficient S"
will be present in equilibrium with the
HS~ and I^S to encourage precipitation
of most metals. Finally, as S is
removed by precipitation, the sulfide
equilibrium will shift in accordance with
Le Chatelier's principle, to generate
more S~2. In effect, therefore, a very
high percent of the total [H2S] + [HS~]
+ [S ] is available for precipitation
of metal sulfides.
Given that sulfate is present and
that complete reduction to sulfide has
occurred within the landfill environment,
the immobilization of specific metal
constituents will then be controlled by
the relative solubilities of the respec-
tive sulfide species. For the studies
under consideration, Figure 5 indicates
the exceedingly low concentrations of
heavy metals (Zn+2, Pb+2, Cd+ , Cu+2 and
Cr ) to be anticipated upon reaction with
sulfides (0.02 moles/liter) throughout
the broad pH range of 3 to 11. Under the
reducing conditions observed in the
experimental studies but with an increas-
ing pH consequenced by the conversion of
volatile acids to methane in a bio-
chemically active landfill, these solu-
bility profiles progressively become even
lower.
In addition to the preceding, those
metal species that are inevitably present
and/or exist in several oxidation states
deserve special consideration. Iron
(and manganese) is a normal ingredient
of landfill leachates and would exist in
the reduced Fe+2 form. It would also be
precipitated as sulfide under the stated
conditions but somewhat less favorably
than the other indicated metals (In
contrast, mercury would exist as Hg2
but would be highly insoluble as Hg2S.)
On the other hand, chromium would be
reduced to Cr+3 but at redox potentials
(EH) approximating -400mv, a transition
occurs from the trivalent to the divalent
state. Available solubility data suggest
that sulfides would play no role in the
removal of chromium in either state and
that solubility will depend uponhydroxide
250
-------
60
0
-60
> -180
E
~ -300
-450
o
•o
o
-600
U
-750
H2S
HS"
320
200
50
0
-50 -
u
UJ
-200
-350
-500
c
0>
o
0.
01
•o
o
6
8
10
pH
SO? » 9H* »
— HS"« 4H20 logK=34
-~ H2S • 4H20 log K=41
HS*« H* H,S logK=7
SO? « 10H* «8e
Figure 4. Species Distribution Diagram for Sulfate-Sulfide Equilibrium
as a Function of pH and Electrode Potential
equilibrium. In this regard, Cr+3 is
much less soluble than Cr+^ ranging from
200 mg/1 at a pH of 5 to 100 yg/1 at
pH 7 as compared to 100 mg/1 at pH 7 for
the divalent state. Therefore, chromium
would tend to be one of the most mobile
of the heavy metals in landfill leachates.
In those rare cases where sulfide
concentrations are very low or have been
depleted due to their removal or the
possible reinstatement of oxidizing
conditions, the fate of the metals becomes
considerably more complex. With sulfide
concentrations as low as 10~6 molar, the
removal of zinc would no longer be con-
trolled by sulfide and would, like chro-
mium, depend upon hydroxide equilibria.
However, superimposed on this reaction
opportunity would also be the relative
influence of the carbonate-bicarbonate
equilibrium which would in the absence of
sulfide, help determine the probable pre-
cipitating species. For instance, assum-
ing an alkalinity as CaC03 of 2000 mg/1
and a pH of 6.5 in the absence ofsulfides,
251
-------
u
a
10
20
30
40
50
_ Sulfide C, =0.02 M
_J I I I I
7 8
pH
10 11
Figure 5. Solubility Profiles for Selected Metals
as a Function of pH and Sulfide Concentration
the probable precipitate species would be
CdC03, Cu3(C03)2(OH)2, PbC03 and Cr(OH)3;
zinc and nickel would be least likely to
precipitate, probably as respective carbon-
ate and hydroxy-carbonate species if at all
and would therefore be most mobile under
the indicated conditions. However, even
at equilibrium concentrations of dissolved
sulfide as low as 10~° molar, the control
of solubility of such metals as Hg2+^,
Cu+2, Cd+2, Pb+2 and Ni+2 will remain
unequivocally in the domain of the sulfide
system.
SUMMARY AND CONCLUSIONS
In view of the opportunity for
significant attenuation and reduction of
contaminant concentrations in leachate
from landfills receiving both municipal
and industrial wastes, it is possible that
rigid and ultrarestrictive regulations on
landfilling of such wastes may not be
completely justified. This appears to be
particularly true of heavy metals when
codisposed of with municipal wastes in a
controlled manner and with the benefits
of leachate containment, collection and
recycle.
Whether or not metals will be
available for migration from a landfill
depends upon their relative concentrations,
the state of the chemical environment, and
presence of certain companion materials for
physical-chemical interaction. The pH
and reducing nature of most landfill
environments determine the redox couples
that predominate and, in the case of
heavy metals, emphasize the importance of
sulfides in the overall attenuating
process. With such a perspective, it is
252
-------
possible to predict probable reaction
mechanisms as well as those processes which
will be most significant for the ultimate
removal of contaminants from the leachate
medium. Therefore, with operational
control over the leachate, metals pre-
cipitated as sulfides (or hydroxides,
carbonates, or hydroxy-carbonates) will
be further removed with leachate recycle
by filtration and/or sorption within the
landfill mass.
What remains to be determined for
such a landfill management option are the
metal loadings which can be sustained
without deleterious effects on the bio-
chemical processes responsible for the
overall stabilization processes within the
landfill and how such effects can be min-
imized with either restrictions on loading
or judicious selection and/or addition
of an appropriate sequestering agent.
The fact that there are relatively few
documented cases of groundwater contam-
ination with heavy metals, even at land-
fills operated with them for long periods
of time, lends credence to such an
approach to controlled codisposal of mu-
nicipal and selected industrial wastes.
ACKNOWLEDGMENT
The research reported herein was
sponsored jointly by the Georgia Institute
of Technology and the U.S. Environmental
Protection Agency under EPA research
grants R-803953 and R-806498. The
assistance of Messrs. Ronald Benson,
Herbert Timmerman and Bruce Spiller in
the conduct of the research investigations
is acknowledged.
REFERENCES
1. Pohland, F. G. 1975. Sanitary Landfill
Stabilization with Leachate Recycle
and Residual Treatment. EPA-600/2-75-
043, U.S. Environmental Protection
Agency, Cincinnati, Ohio. 105 pp.
2. Mavinic, D. S. 1979. Leachate Treat-
ment Schemes-Research Approach. In:
Proceedings of Fifth Annual Research
Symposium on Municipal Solid Wastes:
Land Disposal. EPA-600/9-79-023a,
U.S. Environmental Protection
Agency, Cincinnati, Ohio. pp. 296-305.
3. Fungaroli, A. A., and R. L. Steiner.
1979. Investigation of Sanitary Land-
fill Behavior. Vol. I: Final Report.
EPA-600/2-79-053a, U.S. Environmental
Protection Agency, Cincinnati, Ohio.
313 pp.
Fungaroli, A. A., and R. L. Steiner.
1979. Investigation of Sanitary Land-
fill Behavior. Vol. II: Supplement to
Final Report. EPA-600/2-79-053b,
U.S. Environmental Protection Agency,
Cincinnati, Ohio. 117 pp.
Pohland, F. G., D. E. Shank, R. E.
Benson and H. E. Timmerman. 1979.
Pilot-Scale Investigations of Accel-
erated Landfill Stabilization with
Leachate Recycle. In: Proceedings of
the Fifth Annual Research Sumposium
on Municipal Solid Wastes: Land Dis-
posal. EPA-600/9-79-023a, U.S. Environ-
mental Protection Agency, Cincinnati,
Ohio. pp. 283-295.
Leckie, J. 0., and J. G. Pacey. 1979.
Landfill Management with Moisture
Control. J. Environmental Engineering
Div., ASCE. 105(EE2):337.
Wigh, R. J., and D. B. Brunner. 1979.
Leachate Production from Municipal
Waste-Boone County Field Site. In:
Proceedings of the Fifth Annual
Research Symposium on Municipal Solid
Wastes: Land Disposal. EPA-600/9-79-
023a, U.S. Environmental Protection
Agency, Cincinnati, Ohio. pp. 74-97.
Summer, J. 1978. Cooperative Programme
of Research on the Behaviour of Haz-
ardous Wastes in Landfill Sites.
Report of the Policy Review Committee,
Department of the Environment, London:
Her Majesty's Stationery Office.
169 pp.
Sillen, L. G. 1964. Stability Constants
of Metal-Ion Complexes. Special Pub-
lication No. 17, The Chemical Society,
London. 753 pp.
253
-------
OPTIMIZATION OF LAND CULTIVATION PARAMETERS
K. W. Brown, L. E. Deuel, Jr., and J. C. Thomas
Texas Agricultural Experiment Station
Texas A&M University
Soil and Crop Sciences Department
College Station, Texas 77843
ABSTRACT
Preliminary studies of the effects of amending soils with API oil-water separator
sludge on the germination and yield of ryegrass were initiated to generate data on
acceptable sludge loading rates and to elicit mechanisms which affect plant responses.
A petrochemical and refinery waste were utilized and each waste was mixed with
each of four soils in varying ratios by volume. The mixtures were planted with rye^
grass (Loliwn multifloxnffn) and emergence and dry matter yields were determined. Eight
plant harvest cycles were completed over a 17 month test period. Soil wettability was
also determined periodically.
Concentrations of petroleum hydrocarbons as low as 2% v/v depressed ryegrass
emergence and yields, apparently through at least two mechanisms. Phytotoxic waste
constituents initially acted to severely diminish plant response. Long term yield
reductions largely resulted from impaired water, air and nutrient relations associated
with recalcitrant hydrophobic hydrocarbons.
The petrochemical sludge suppressed the emergence and yield over a longer period
of time. The suppression was proportional to the amount of each waste applied.
INTRODUCTION
Disposal of industrial wastes is a
chronic problem intensified by the
current awareness of the need for environ-
mentally sound management practices. Of
particular importance is the proper hand-
ling of potentially hazardous materials
which are by-products of the petroleum
industry. An effective method of disposal
should immobilize, degrade, or isolate
the waste compounds such that the
possibility for environmental contamina-
tion is minimized or eliminated.
Landspreading, also called land-
farming, land treatment, and soil
incorporation, is the controlled disposal
of wastes in the surface soil accompanied
by the continued monitoring and manage-
ment of the disposal site. Although this
means of disposal has been used by the
petroleum industry to dispose of process
sludges, information is still needed on
site selection, optimum soil and climatic
conditions, application rates and
scheduling, decomposition products,
potential contaminant emissions and the
persistence of toxic residues.
The aims of this research were part
of a comprehensive study of the land
application of petrochemical sludges.
Greenhouse studies were utilized to
observe the effects of sludge amendments
on the emergence and yield of ryegrass
in order to assess the residual toxicity
of the wastes as a guide to optimum
loading rates.
254
-------
MATERIALS AND METHODS
SOIL PROPERTIES
Soils of varied properties including
mineralogy, texture, organic matter
content and reaction were employed to
represent a range of potential disposal
sites. The four soils chosen were
Bastrop, a fine-loamy mixed, thermic Udic
Paleustalfs; Lakeland, a siliceous,
acid, thermic, coated Typic Quartzipsam-
ments; Nacogdoches, a fine kaolinitic,
thermic, Rhodic Paleudalfs and Norwood,
a fine, silty, mixed (calcareous) thermic,
Typic Udifluvents. The textures of the
four soils were sandy clay loam, sandy
loam, clay and sandy clay for the Bastrop,
Lakeland, Nacogdoches and Norwood soils
respectively. The bulk densities were
1.49, 1.38, 1.4 and 1.44 g cm"3 at
field capacity for each of the four soils
respectively. The cation exchange
capacity of the soils were 27.4, 0.3,
17.2 and 19.6 respectively. The pH of
each soil was 6.86, 6.45, 5.59 and
7.69 respectively.
WASTE CHARACTERIZATION
The waste materials were API oil-
water separator sludges from the petro-
leum industry. A survey of refiners
revealed that this class of waste could
vary widely depending upon crude stocks,
refinery end products, and plant equip-
ment and management. The sludges used
in this study were obtained from a
refinery producing primarily fuels and
lubricants and from a petrochemical
plant. Water contents determined by
distillation were 41.1% and 20.0% for the
refinery and petrochemical wastes
respectively. The solvent extractables
were quantified gravimetrically following
Soxhlet extraction with pentane, benzene
and dichloromethane and were 3.09, 3.29
and 3.23% respectively for the refinery
waste and 19.75, 13.89 and 4.26%
respectively for the petrochemical waste.
Residues of 49.3 and 42.8% respectively
for the refinery and petrochemical wastes
were heated to 7500C and ash was weighed.
Ash contents of the refinery and
petrochemical wastes were 37.8 and 7.65%
respectively. Total sulfur was
ascertained by an idometric titration of
SO- after oxidative combustion and
yielded 0.88 and 0.39% S for the
refinery and petrochemical wastes
respectively. Total N was found by
Nesslerization following ammonification
of all N sources with a H-SO.rK-SO,:
mercuric oxide solution and yielded 0.09
and 0.11% N respectively. Wet combustion
with potassium dichromate and
HjSO,:HJPO, yielded total C contents of
10.0 and 35.0% for the refinery and
petrochemical wastes respectively.
PREPARATION OF GREENHOUSE POTS
Pots 15 cm ID x 15 cm deep were
filled with soil .-sludge mixtures and
seeded. Waste loading rates were 0, 5,
10 and 20 percent v/v for each soil,
and 2,000 cm3 of the thoroughly mixed
soil:sludge was placed in each pot with
three replications per treatment. Soils
were uniformly fertili;:ed by applying
1 g of 17-17-17 per pot incorporated
into the top 2 to 5 cm. This is
equivalent to 650 kg/Ha of the fertilizer.
Soils were lightly watered, and 100
ryegrass (Loliwn multiflomm') seeds were
broadcast on the surface. Irrigation
water applied to pots was from a rain-
water collection system.
SEEDLING EMERGENCE COUNTS
Counts of seed emergence were
performed as one indicator of the toxic
effects of sludge amendments in soils.
Four to 6 days after the planting date,
daily counts of seedling emergence were
begun and continued until no further
increase was apparent. Seedling counts
were not made once successive plantings
no longer indicated a significant
treatment effect on emergence.
HARVESTING
Thirty to forty days after seeding,
the plant material was harvested for
treatment comparisons of dry matter
production. All aerial plant material
was removed by clipping the ryegrass
at the soil surface. Harvested plants
were dried in a 60°C oven and weighed
to determine yield per pot. Soils were
mixed to simulate cultivation and to
arrest further growth activity which
could interfere with later yield results.
Cultivation also insured adequate
255
-------
aeration in order to foster microbial
degradation of the wastes. After drying
for a period of two to 4 weeks, all pots
were replanted with 100 ryegrass seeds
per pot. The yield of all treatments
decreased with time, indicating a
possible depletion of nutrients. Thus a
second fertilizer application was made
just after the seventh harvest.
RESULTS AND DISCUSSION
Emergence Comparisons Between Soils
For both wastes, the Bastrop soil
manifested the highest germination rates
as observed over all planting dates over
the range of sludge loads. Differences
between the remaining soils were minimal.
Since this relationship persisted
regardless of sludge application, this
appeared to be a function of soil
properties.
Loading Rate Comparisons of Emergence
The emergence response to varied
sludge loads was dependent upon the waste
considered. The refinery waste appeared
not to alter or in some cases, possibly
enhanced germination (Figure 1). In
view of the ryegrass seeds' insensitivity
to refinery sludge after the first
planting, further emergence counts were
performed only on petrochemical waste
amended soils.
• BASTROP
A LAKELAND
o NACOGDOCHES
X NORWOOD
10 15
WASTE (%)
zo
Figure 1.
The percent emergence of rye-
grass seedlings in four soils
treated with refinery sludge.
80
7oi
60'
~ 50
UJ
t> ;
| 40
a: i
u
u 30
20
10
• BASTROP
A LAKELAND
o NACOGDOCHES
X NORWOOD
X
10
WASTE (%)
15
20
Figure 2. The percent emergence of rye-
grass seedlings in four soils
treated with petrochemical
sludge.
In contrast, petrochemical sludge-
amended soils significantly reduced
germination in all soils at all three
loading rates (Figure 2). Several
mechanisms may have operated to inhibit
emergence. An increased water
repellency was observed when water was
applied to sludge-amended soils which
may have restricted the water available
for seed inbibition.
Another possibly substantial control
of emergence could have been the toxic
effect of waste constituents. In a
review of the effects of plant exposure
to oils, Baker (1970) cited several
sources which indicated that aromatic
compounds, particularly 12-carbon atom
aromatics, are highly toxic to plants.
Thus, increasing aromatic content could
increase toxicity. Aquatic plants, for
example, were controlled by as little
as 300 ppm light aromatics in water,
and foliar applications of oil containing
10% aromatics severely affected maize
plants. The aromatics contents of the
refinery and petrochemical sludges used
in this study were 3.29 and 13.89% (w/w),
respectively, which contributed to the
soil organics load. The petrochemical
plant produces various herbicidal agents
as well, residues of which may have
been present in the waste.
256
-------
Changes In Emergence With Time
Successive ryegrass plantings on
petrochemical waste-amended soils result-
ed in increasing emergence, such that
after six months the 20% sludge rate
evidenced little emergence depression
below control plantings. Absorption of
water by the soil continued to be poor
after six months while the strong sludge
odor virtually disappeared after two
months. Such qualitative observations
suggest that the toxic compounds present
in the waste are responsible for depress-
ing emergence. Decomposition and
volatilization of phytotoxic compounds
could have reduced the concentrations
during this period so that seeds on the
soil surface no longer suffered adverse
effects.
Soil Comparisons of Yields
Corresponding to the emergence
results, the yields of ryegrass were
found to be consistently highest for the
Bastrop soil. Table 1 compares soil
effects on yields. Differences were
3
w
H
CO
CO
3
0
W
c£
2
O
CO
H
1
W
!_J
H
O
CO
.
1—)
W
^J
3
H
4-1
O
cx
00
\^
•o
o>
•rl
01
i >
W
cfl
^
rH
CO
O
•H
01
43
O
0
tJ
4J
01
ftj
CO
4J
CO
g
M
CO
C
•H
S
rH
•H
O
CO
PQ pq pq
<
PI CM CM r-
CN rH rH O
* S « °
<
-* iH in vD
....
PI CM CM t— 1
CO
01
fi
•o o
a C o -n
o n) T3 o
Vj i-H 60 o
4J O) O S
tn ^ o S-i
CQ Co cd o
« i-4 Z Z
o
T3
CO
1 1
4J
iH
.
0) rH
43 Ol
4J >
CO
-D
•a in
CO
3 co
O 43
rH 4-1
O 4->
<4-i n)
6 K^
£3 i~H
3 4-1
rH C
o n)
o o
•H
CO -H
> c
•H 00
00 -H
CO
C CO
•H ^
4H
CO -H
CO T)
rH 4J
03 o
> c
*
attributable to differences in native
fertility and chemical and physical
properties, may be another criteria in
selecting soils suitable for waste
disposal where plant cover is desired.
Loading Rate Comparisons of Yields
Both the sludges applied effected
significant yield reductions below the
untreated control. Results from the
first harvest where sludges were applied
just prior to planting are plotted in
Figure 3 and 4. Since the yields for
any given sludge rate varied depending
upon which waste was applied, concentra-
tions of one or more constituents of
the wastes rather than the bulk sludges
appeared to regulate plant growth.
Comparison of the 20% refinery waste
applications to the 5% petrochemical
waste treatments suggests the possible
influence of total carbon and/or percent
aromatics on the yield. The percent
carbon reflects the petroleum hydrocarbon
content of the soils. Several
investigators (Carr, 1919; Plice, 1948;
Schwendinger, 1968; Udo and Faymi, 1975)
have concluded that oil-associated yield
depressions were partially due to impair-
ed aeration and to poor water and
nutrient availability. Additionally,
as mentioned in the discussion of
emergence, aromatics present in the
wastes may have been toxic to plants when
present in substantial concentrations.
15
• BASTROP
A LAKELAND
o NACOGDOCHES
X NORWOOD
10
WASTE (%)
15
20
Figure 3. The yield of ryegrass on four
soils treated with refinery
sludge.
257
-------
• BASTROP
A LAKELAND
o NACOGDOCHES
X NORWOOD
20
WASTE
Figure 4. The yield of ryegrass on four
soils treated with petro-
chemical sludge.
Changes In Yield With Time
Degradation of the organic fraction
of the API oil-water separator sludges
apparently controlled the suitability of
sludge-amended soils as a plant growth
medium. An initial lag time, characteri-
zed by lowered yields, was followed by a
period of rapid improvement as toxicity
decreased and soil physical relations
improved with waste degradation.
Apprently, however, recalcitrant oily
materials ultimately stabilized yields
at levels below control yields, and
little further change was noted. After
17 months, ryegrass plants in pots
receiving sludge continued to appear
sickly and showed symptoms of water and
nutrient deficiencies.
Over a 13 month period, yields from
refinery waste-amended soils approached
control values (Table 2) while the higher
rates of petrochemical waste loading
continued to reduce yields by 50% or
more (Table 3). In a landf arming context,
additional petrochemical sludge amend-
ments might increase residual oils in
soils to such an extent that plant
growth could be depressed even further.
s
o
Fn
W
Fd
PH O
*^ s
35 Fd
CO -* >-3
pej co
Ft* ^
O PH
Ed
S fe
Ed W
hH pd
>H
Cd
>
H
<
3
CM
W
jj
5
H
rH
z
4-J
c
o
CJ
UH
0
B-S
T3
rH
CU
•rl
CU
^
•H
CO
rH
K
T)
o
Q
>
^
O
21
CO
Q)
r;
O
0
13
O
0
CO
£3
•a
c
CO
i^
CU
CO
a
o
*H
4J
CO
CO
1
|
\o
CO
1
vO
CM
^.
CT<
7
•"^
1
oo
r-^
1
CM
-d-
a\
1
1
vO
oo
1
VO
CM
1
O\
1
1
vO
00
r~-
1
vO
CM
1
4-1
C
CU
CJ
^l
CU
a,
^ ^ ^C ^
O rH CO CO
O 00 o\ O
rH rH
O
O
o in m CM
O vO -* CM
rH
CO pq
^3 !**«• CO r^
O O r- vo
rH rH
O O
CQ ro
•*^ <^
o ^d" r^- co
O vO CO rH
rH
-------
TABLE 3. RELATIVE YIELDS OF RYEGRASS HARVESTED ON TWO OCCASIONS
FROM PETRO-CHEMICAL SLUDGE-AMENDED SOILS
Relative Yield (% of Control)
Percent
Waste
0
5
10
20
Bastrop
4-26-78
100A*
27 B
3 C
0 C
6-4-79
100A
87AB
55 B
54 B
Lakeland
4-26-78
100A
55AB
1 B
0 B
6-4-79
100A
10 1A
97 A
60 B
Nacogdoches
4-26-78
100A
29 B
1 C
0 C
6-4-79
100A
104A
77 B
46 C
Norwood
4-26-78
100A
2 B
0 B
0 B
6-4-79
100A
56A
44 B
24 B
* Values in a given column followed by the same letter do not differ significantly at
f-Vl£i XL"/ la-IT^I
the 5% level.
REFERENCES
1. Baker, J. M. 1970. The effects of
oils on plants. Environ. Pollut. 1:
27-44.
2. Carr, R. H. 1919. Vegetative growth
in soils containing crude petroleum.
Soil Sci. 8: 67-68.
3. Plice, M. J. 1948. Some effects of
crude petroleum on soil fertility.
Soil Sci. Soc. Am. Proc. 13: 413-416.
4. Schwendinger, R. B. 1968. Reclama-
tion of soil contaminated with oil.
J. Inst. Petroleum 54: 182-197.
5. Udo, E. J. and A. A. A. Fayemi. 1975.
The effect of oil pollution of soil
on germination, growth and nutrient
uptake of corn. J. Environ. Qual.
4: 537-540.
259
-------
FIELD VERIFICATION OF LAND CULTIVATION/REFUSE FARMING
Joan B. Berkowitz
Sara E. Bysshe
Bruce E. Goodwin
Judith C. Harris
David B. Land
Gregory Leonardos
Sandra Johnson
Arthur D. Little, Inc.
Cambridge, Mass. 02140
ABSTRACT
Land cultivation practices were observed at six sites disposing of industrial residuals.
The sites were selected to permit identification and verification of parameters that con-
tribute to the environmental acceptability of land cultivation for a variety of industrial
wastes under markedly different hydrogeological, meteorological and land use conditions.
Landfarming at each site investigated was characterized as a physical/chemical/biological
waste treatment process. Accordingly, information was obtained on: the physical and
chemical composition of the waste fed to the surficial soil treatment system; character-
istics of the treatment system; operational procedures; effectiveness of treatment; poten-
tial environmental impacts; and costs. To assess the effectiveness of treatment, samples
of waste, soil, and waste-soil mixtures were collected and analyzed for organic compound
classes; pH; S04 and CSL ; electrical conductivity; cation exchange capacity; total
Kjeldahl nitrogen; As, B, Ca, Cd, Cr, Cu, K, Mg, Mo, Na, Ni, P, Se, V, and Zn. To assess
the extent of uptake of inorganic waste constituents, vegetation was collected at most
sites and elemental analysis was carried out. Waste inputs to a landfarming system can be
characterized quite precisely; output stream can be determined in principle by comparing
the composition of waste application areas with the composition of suitably chosen control
areas. Treatment processes occurring in the soil system for particular wastes in particular
locations can, at the present time, only be inferred from observed compositional changes.
This points up the need for long-range research on fundamental mechanisms of waste treat-
ment in soil systems; and short-range research on monitoring protocol for landfarming op-
erations.
INTRODUCTION
Land cultivation/refuse farming is a
waste management method that utilizes the
natural capacity of soil systems to effect
physical, chemical, and biological treatment
of certain waste constituents. The practice
of land cultivation of municipal sludge is
well-established and documented.
Within broad limits, the physical and chem-
ical composition of municipal sludges is
relatively constant, and general guidelines
can be provided for successful operations.
In contrast, land cultivation of industrial
sludges is not well documented. Further-
more, the industrial sludges to which land
cultivation is potentially applicable in-
clude a wide variety of types, compositions
and manufacturing sources. Certain oily
wastes have been landfarmed, with apparent
success, by the petroleum industry for at
least 25 years. However, even oily wastes
from different sources and at different
times can differ significantly in proper-
ties that can affect the success of a land
cultivation operation.
260
-------
This paper summarizes observations
made at six sites currently landfarming
different types of industrial sludges in
different areas of the country. The objec-
tive of the field work described was to
identify and verify parameters that con-
tribute to the environmental acceptability
of land cultivation for the management of
some industrial residuals. Costs associ-
ated with landfarming operations at the
sites investigated were also assembled, in
order to provide a reference for compari-
son with costs of other disposal alterna-
tives.
LANDFARMING AS A WASTE TREATMENT METHOD
Landfarming as a waste treatment meth-
od is operationally simple and mechanis-
tically complex. Once a site has been es-
tablished, the basic operations involve
essentially: (1) uniform application of
wastes to the surface or subsoil of a se-
lected site; (2) cultivation of the applied
waste into the soil with a disc or roto-
tiller to assure intimate blending of waste
and soil in the zone of incorporation; (3)
incorporation of fertilizer, lime or other
additives, if necessary, by standard agri-
cultural techniques; (4a) periodic reculti-
vation to assure aerobic conditions in the
soil treatment system; (4b) periodic moni-
toring to determine: the extent of treat-
ment that has occurred; the possible need
for pH adjustment; environmental effects;
and whether wastes can safely be reapplied;
(5) repeat steps (1) through (4b), if no
adverse environmental effects have been
found, and if re-application is appropri-
ate.
Monitoring (step (4b) above) is of
major importance in practical applications,
because the rates and routes of action of
a given surficial soil treatment system on
a particular waste cannot be accurately
predicted at this time. The general types
of treatment that can occur in the upper
soil layers are known. The efficiency and
effectiveness of those treatment processes
on various components of a complex waste
stream are not known. Hence, monitoring is
key to assessing the environmental adequacy
of the landfarming as a waste treatment
method.
Landfarming might be considered among
other alternatives for the management of
an industrial waste stream, if the follow-
ing types of treatment, potentially avail-
able in soil systems, appear to be appli-
cable without adversely affecting the en-
vironment :
• Aerobic microbial decomposition of
organic components of the waste —
Wastes which will biodegrade com-
pletely and relatively rapidly (say
within less than a year) to micro-
bial cell mass, 002 and water are
particularly good candidates, pro-
vided that intermediate products of
degradation either do not leach
and/or volatilize significantly, or
do not adversely impact ground
water and/or air if they do migrate.
• Chemical oxidation and/or hydroly-
sis
• Ion exchange — binding of cations
to unoccupied sites in the soil
matrix or substitution of heavy
metals in the waste for exchange-
able cations (such as Na , Ca ,
Mg++) in the soil. Once the cation
exchange capacity of the soil is
saturated, however, the substitu-
tion mechanism will release alkali
and alkaline earth salts into in-
filtrating waters.
• Precipitation — formation within
the soil system of relatively in-
soluble compounds with little po-
tential for migration either via
leachate or via biological uptake.
• Neutralization — In one of the
site examples given below, an alka-
line lime treatment sludge is used
by a turf farmer for neutralization
of an acid soil.
While these processes are analogous to
those in engineered treatment systems,
their mode of action within a soil-waste
system is dependent on a far greater range
of parameters, and the possibility for
side reactions is legion. There is a very
extensive literature on the subject. While
this paper focusses on waste treatment in
surficial soils, landfarming also provides
an opportunity for recycling of the compo-
nents of certain type of wastes. Some
wastes can have a beneficial impact on soil
structure and soil productivity for plant
growth. However, given the complexity of
the system, consideration must be given
to the potential for unwanted reactions
261
-------
that might impact the environment ad-
versely. In some cases, these might be
suppressed by pretreatment of the waste or
management of the landfarming system.
STUDY METHODOLOGY
Six sites in different areas of the
country were selected for observations of
landfarming practices. Field visits were
made to each site to document operational
practices. Data were obtained, by means of
interviews with plant and landfarming per-
sonnel, on the origin of the wastes; the
character of the sites; the method and rate
of waste application; the procedure for
mixing wastes into the soil; post waste-
addition care; management of the landfarm-
ing process; and costs. Samples of waste,
soil, and waste/soil mixtures were col-
lected and analyzed for inorganic constitu-
ents and organic compound classes in order
to provide information for assessing:
• the rate and extent of waste de-
gradation
• the migration of inorganic spe-
cies, and
• the accumulation of waste consti-
tuents in the soil.
Vegetation was collected at most sites and
analyzed to assess the extent of uptake of
inorganic waste constituents.
Sampling
At each site, samples were taken from
1) the area where waste had been applied
and 2) the best identifiable control area.
The ideal control area would have been up-
wind and upgradient of the waste applica-
tion area, but in all other respects would
have been identical to that area except
that no waste had been applied. Selection
of an appropriate control area was a prob-
lem at the majority of sites for a variety
of reasons. In general, the design and
operation of most sites did not include
provision for a bona fide control area as
described above. More specifically, prob-
lems included substantial variations in
soil characteristics, drainage patterns and
vegetation occurring within small dis-
tances; and uncertainty about the history
of waste application outside the bounda-
ries of the present application site. In
all cases, a "control" area was selected
which necessitated the least compromise in
the requirements stated above. In both
control and application areas, samples were
collected from a 4 x 4 grid covering the
area. Wastes were collected from storage
facilities or other sources, selected so as
to assure that the samples were representa-
tive of the waste that was actually applied.
Core samples of soil and soil waste
mixture were taken with a sampling tube.
The resulting cores were 1" in diameter and
12" in length. These samples were divided
in the field into two subsamples corres-
ponding to 0-6" soil depth and 6-12" soil
depth. Bulk samples were subsequently
coned, quartered and riffled to obtain
amounts of sample appropriate for each
analysis. Vegetation samples were col-
lected insofar as possible from the same
control and application areas from which
soil samples were obtained. Unless noted
otherwise, composite vegetation samples were
prepared from whole plants minus roots.
There was no sampling of surface water, run-
off, or groundwater at any site during the
sampling visits made.
All samples were labeled with codes in
the field, so that the identities of par-
ticipating parties were not apparent from
the label information. Analyses of se-
lected soil, soil-waste mixture, and waste
samples for Total Kjeldahl Nitrogen (TKH)
and Total Organic Carbon (TOC) were per-
formed by Schwartzkopff Microanalytical
Laboratories. All other analyses were con-
ducted at Arthur D. Little, Inc. labora-
tories in Cambridge, Massachusetts.
Analysis Methodology
Inorganic parameters measured included
sixteen elements, T.O.C., T.K.N., two
anions (Ci~ and SOi^), C.E.C., B.C., pH,
and DTPA-extractable metals. The methods
of analysis are summarized in Table 1.
The organic analysis scheme is shown
schematically in Figure 1. The methodology
involved:
(1) Solvent extraction with methylene
chloride;
(2) Quantitative analysis of volatile
and non-volatile organlcs;
(3) Separation of the extracts by
liquid chromatography to divide
262
-------
TABLE 1. METHODS OF ANALYSIS
Parameter
Method
B, Cu, Cd, Cr, Cu, K, Mg, Mo,
Na, Ni, P, Pb, V, Zn
As, Se
Total Organic Carbon (T.O.C.)
Total Kjeldahl Nitrogen (T.K.N.)
Sulfate
Chloride
Cation Exchange Capacity (E.C.E.)
Electrical Conductivity (E.G.)
PH
DTPA-extractable metals
HN03-HC104 digestion, plasma emission
spectroscopy
Hydride evolution, atomic absorption
spectroscopy
Combustion, gravimetric
Digestion, distillation, volumetric
0.1 M LiCl extraction, ion chromatography
H20 extraction, volumetric Hg(N03)2 ti-
tration
Volumetric
Conductivity meter
pH meter
DTPA extraction, plasma emission spec-
troscopy
the organics into three polarity
classes: aliphatic hydrocarbons,
aromatic hydrocarbons and polar
species; and
(4) Qualitative analysis of organic
compound classes present by high
resolution mass spectrometry.
The above species, for which analyses
were done, accounted for 38-69% of the dry
waste weight, depending on the site. Spe-
cies not determined that might account for
the remaining waste weight include Fe, H,
0, Si, AX,, and non-extractable organics.
OBSERVATIONS
The sites selected for field study
represent, by design, a wide diversity of
landfarming operations as actually prac-
ticed today. None of the sites was es-
tablished solely for the purpose of obtain-
ing research data. All of the sites serve
a significant waste management function.
Specific observations at each of the sites,
as waste treatment systems, are summarized
below.
Site A/Organic Chemical Industry Sludge
Physical and Chemical Composition of Wastes
Treated — The waste being landfarmed at
Site A is a secondary wastewater treatment
sludge from a batch organic chemicals manu-
facturing facility (SIC 2865). The waste
composition varies significantly over short
periods of time, but the volume of waste-
water treated is so large that there is
some compositional inertia in the sludge
generated for land cultivation.
The sludge, as applied, contains about
7% solids and 93% water. Inorganic analy-
sis of the solids fraction yielded the re-
sults shown in Table 2. The site is a com-
mercial turf farm, and the waste is being
applied primarily for its lime (calcium)
content. The waste is also seen to be high
in the plant nutrients: potassium, nitro-
gen and phosphorus. The heavy metals ob-
served in highest concentration were zinc
and copper.
263
-------
LC-H
Aliphatic
Hydrocarbons
TCO
(VolatiKs)
GRAV
(Non-Volatiles)
Soil and/or Waste
Sample
Soxhlet
Extraction
Organic Extract
LC Separation
on Silica Gel
LC-A
Aromatic
Hydrocarbons
TCO
(Volatile!)
GRAV
(Non-Volatiles)
Combined Fractions
HRMS
Analysis
Figure 1. Flow Chart of Organic Analysis Scheme
LC-X
Polar
Species
TCO
(Volatiles)
GRAV
(Non-Volatiles)
TABLE 2. SITE A WASTE ANALYSIS - TOTAL METALS. T.O.C.. AND T.K.N.*
Element As B
Ca
Cd Or Cu K Mg Mo N Na Jl
Pb Se
Zn
Concen- 2 10 16.27% 7.4% 4.8 37 2000 1.8% 4.6% 8.9 2.96% 2% 170 1.7% 100 0.3 17 1900
trations
* (in pg/g of dried solids, except as noted)
264
-------
The extractable organics in the waste
amounted to 96 mg/L, characterized as fol-
lows:
Aliphatic Hydrocarbons, non-
volatile (GRAV) 9.5%
Aromatic Hydrocarbons, non-
volatile (GRAV) 3.6%
Polar Organics, volatile (TCO) 1.8%
non-volatile
(GRAV) 85.0%
The HRMS analysis showed mainly low molecu-
lar weight (<150) ions that are probably
fragment ions, representing aromatic C, H,
N; C, H, 0; and C, H, N, 0 compositions.
Since parent ions of higher molecular
weight homologs were not found, it seems
probable that the actual waste components
(from which the fragments are formed) are
of sufficiently high molecular weight to be
non-volatile at the 300°C probe tempera-
ture.
Characteristics of the Site as a Waste
Treatment System — Site A is a turf farm,
principally growing bluegrass. The fact
that the site supports agronomic activity
suggests favorable conditions for biodegra-
dation of waste components amenable to this
process. The soil consists of approximate-
ly three feet of silt loam underlain by
stratified sand and gravel. Waste treat-
ment would be expected to be confined pri-
marily to the relatively shallow surface
loams. The cation exchange capacity of
the unamended surface soils is in the range
of 6.1 - 10.2, which reflects a low to mod-
erate potential for immobilization of heavy
metals (such as copper and zinc) by an ion
exchange mechanism. However, the unamended
soils are also slightly acidic (pH = 5.0 -
5.7), and the increase in pH due to the
lime content of the wastes can also in-
crease the cation exchange capacity of the
soil. The topography of the site is essen-
tially flat, but drainage is good. Hence,
the potential for pooling or flooding,
which could create anaerobic conditions, is
low. Average annual precipitation is about
40 inches, distributed fairly evenly over
the year. This, combined with the high
available water capacity in the surface
soils, would favor chemical hydrolysis and
oxidation reactions for waste components
subject to such treatment. Climate is tem-
perate, with mean summer temperatures of
73°F, mean ^inter temperatures of 29°F, and
a frost-free period of about 200 days.
Operational Procedures — Wastes are trans-
ported from an aerated concrete lined
lagoon at the plant to the turf farm by
tank trucks. Wastes are applied by gravity
flow through a nine-foot spreader bar at
the rear of the tank truck. The amount of
sludge deposited per unit area is a func-
tion of truck speed, and is not well con-
trolled. The application area sampled as
part of this study was 39 acres in extent.
Something of the order of 500,000-2,000,000
gallons of waste had first been applied to
the area in 1976, two years prior to our
sampling visit.
Following application, the waste is
typically allowed to lie on the surface for
1-2 weeks prior to incorporation. Standard
farming techniques are used to mix the
wastes into the soil. Included are plow
harrowing, disc harrowing, reharrowing, dry
screening, fertilization and seeding. Once
an area is seeded, wastes are not re-applied
for 1-1/2 - 2 years, as in the application
area sampled in this study.
Post-application monitoring of the
soil-water mixtures has been largely
limited to pH and fertilizer requirements
needed to grow the crop.
Effectiveness of Treatment — If all of
the waste reportedly applied in 1976 (two
years prior to our sampling) had been uni-
formly incorporated into the top 12" of
soil over the entire application area and
simply accumulated, then observable in-
creases in the application area soil (as
compared to the control) would have been
expected for Ca, C, N, K, Na, P, and Mg.
In fact, among the inorganics, only copper
was found in higher concentration in the
application area soil than in the control.
Organic analysis showed no significant in-
crease of extractable organics in applica-
tion area soils, suggesting that biodegra-
dation may have occurred. On the other
hand, the HRMS data showed the presence of
C, H, N and C, H, N, 0 compounds in both
the waste and the application area, but not
in the control area. Thus some organic
waste components may have persisted in the
soil over the period between application
and sampling.
With the possible exception of copper,
the soil in the application area seems to
have been restored to natural background
conditions at the time of sampling.
Growth of a turf grass crop after waste
265
-------
application may have reduced the concentra-
tions of the primary plant nutrients (N, P,
K) in the soil to background levels. The
lack of accumulation of Ca, Na, and Mg may
be due to the leaching of soluble salts be-
low the zone of incorporation. The in-
creased copper concentration in application
area soils is also reflected in increased
copper concentrations in application area
grasses. Following another year of appli-
cation of waste at the site, observable in-
creases in zinc might be expected. Both
copper and zinc are essential plant (and
animal) nutrients, but can be phytotoxic at
excessive levels. Such levels have not
been reached at Site A, but if further
analysis should indicate continued build-
up, these elements in the waste could be
the parameters controlling the useful life
of the site as a waste treatment system.
Potential Environmental Impacts — Poten-
tial for erosion and for contamination of a
small stream close to the site is minimal,
due to the general topography, the rela-
tively well drained soils, and the growth
of vegetation on the site.
Groundwater is found 15-30 feet below
the surface of Site A, and was not sampled
as part of the present program. The aqui-
fer, however, is considered by State offi-
cials to be of major importance as a drink-
ing water source. If there is any migra-
tion of waste constituents to ground water,
Ns+ is likely to be a leading indicator.
The elevated copper levels observed at
Site A do not appear to have had any ad-
verse effects on agronomic activity. How-
ever, the potential for accumulation of
both copper and zinc could eventually limit
continued use of the waste amended soils
for turf grass production. At present,
there is no established program at the site
to monitor for waste components of poten-
tial environmental significance, At Site
A, these components are principally Cu, Zn
and persistent organics in the upper soil
layers; and sodium below the zone of in-
corporation.
Costs — Since no additional land or
equipment was needed at Site A, other than
that associated with turf farming, the cap-
ital cost associated with landfarming per
se is zero. The annual operation costs,
associated primarily with transportation
and surface application, are about $27,000
or a little over 2c/gallon of wet sludge
applied. It should be noted that these
costs would be increased if a monitoring
program were initiated to track the fate
of potentially problematic waste constitu-
ents.
Site B/Plastics Industry Sludge
Physical and Chemical Composition of Wastes
Treated — The waste being landfarmed at
Site B is a centrifuged activated sludge
from the wastewater treatment system of a
plastics manufacturing facility (SIC 3079).
The sludge, as applied, contains about 8.5%
solids and 91.5% water. Inorganic analyses
of the solids fraction yielded the results
shown in Table 3. The waste is high in or-
ganic carbon, sodium, and the plant nutri-
ents, N, K, and P.
The extractable organics in the waste
amounted to 33 mg/L, characterized as fol-
lows:
Aliphatic Hydrocarbons, vola- 12%
tile (TCO)
non-volatile (GRAV) 46%
Aromatic Hydrocarbons, non-
volatile (GRAV) 7%
Polar Organics, non-volatile
(GRAV) 35%
HRMS analysis showed the extractable or-
ganics to consist primarily of hydrocarbons
with at most one aromatic ring. Fatty
acid/ester mass spectral fragments account-
ed for 24% of the total composition assign-
ments. There were also some indications of
the presence of aliphatic sulfur compounds.
Characteristics of the Site as a Waste
Treatment System — Site B is a 180-acre
plot owned by the waste generator and
leased to local farmers for growing wheat
and corn. Soils in the region consist of
moderately fine textured silty clay loam
and are reported to be well-drained. The
small area sampled as part of this program
appeared to have a higher proportion of
clay soils than did other areas of the
farm. The area sampled had a cation ex-
change capacity of 20.6 - 20.8, and a
slightly alkaline pH of 7.3 - 7.5. The
topography of the area sampled was nearly
level, but rolling hills with 1-3% slopes
were more typical of the rest of the farm.
Average annual precipitation is about 33
inches, with 70% of it occurring in the
266
-------
TABLE 3. SITE B WASTE ANALYSIS - TOTAL.METALS, I.O.C.. AND T.K.N.*
Element As B
Ca Cd Cr Cu
Mg Mo N Ha Ni P Pb Se
Zn
Concen- 4 25 22.93% 660 5 54 60 8200 1600 6.0 2.90% 16.0% 44 6100 50 0.5 4.9 81
trations
(in pg/g of dried solids, except as noted)
growing season. Mean summer temperatures
are about 90°F; average daily minimum win-
ter temperatures are less than 32°F. The
frost free period is typically 180-185 davs.
Operational Procedures — Wastes from the
manufacturing process are stripped of sul-
fides, neutralized, and then sent to acti-
vated sludge secondary treatment where ni-
trogen and phosphorous are added. The
secondary treatment sludge is centrifuged
to increase the solids content in the waste
which is landfarmed. Big Wheels® tank
trucks are used for both waste transport
and application. Sludge is directly in-
jected six to eight inches into the soil
via a three-knife subsurface application
unit attached to the Big Wheels®. Wastes
are applied at the rate of about 7200 gal/
acre-yr.
Wastes are incorporated into the soil
by standard agronomic procedures, and soils
are amended for pH adjustment or with fer-
tilizers, etc., as dictated by the needs of
the crop to be grown.
The soil-waste mixtures are analyzed
by the farmer for pH, N, P, and K about
every two years, and occasionally for or-
ganic matter and micronutrients. More ex-
tensive analyses are carried out by the
waste generator for research purposes. The
generator also analyzes a. representative
sample of the crops grown for Cd, Pb, Zn,
Cu and Ni prior to releasing them for sale.
Effectiveness of Treatment — Wastes had
been applied to Site B for two years prior
to our sampling, with the latest applica-
tion having been made 3 months previously.
If all wastes applied over the history of
the site had simply accumulated, the only
elements for which significant increases
above background could have been observed
are Na, C, and N. In actuality, no sig-
nificant differences were observed among
the elements analyzed between the applica-
tion and control areas. Nitrogen is al-
most certainly lost from the soil via
plant growth, since the nitrogen fertilizer
has been reduced by 20-40 pounds/acre due
to waste application. Organic analysis re-
vealed no significant differences between
application and control areas. The aver-
age electrical conductivity of the upper
12" of the application area (365 p mhos)
was slightly elevated compared to the con-
trol area (173 u mhos).
Wheat plants grown in the application
area were taller than those grown in the
control area. Analysis of grain for inor-
ganic waste constituents showed no differ-
ences between application and control areas.
Potential Environmental Impacts — Poten-
tial for erosion and for surface run-off
contamination of a nearby stream is mini-
mal.
Ground water is reportedly at least
60 feet below the soil surface. Soil char-
acteristics favor the build-up of soluble
salts in the soil over migration to ground
water. However, the potential for migra-
tion of sodium salts to aquifers over the
long-term does exist.
The waste, as noted earlier, is high
in sodium, which is known to have a dele-
terious effect on soil-plant systems. No
impacts have been noted on crops grown to
date. If a problem does arise, gypsum may
be added to the waste to offset the sodium
content or more salt-tolerant crops may be
grown.
Costs — Total capital cost for land, site
approval and equipment at Site B is
$202,000 in 1978 dollars. Annual operating
costs, including capital amortization, come
to $40,500/yr or somewhat less than 3c/gal
of wet sludge treated.
267
-------
Site C/Leather Industry Sludge
Physical and Chemical Composition of Wastes
Treated — The wastes being landfarmed at
Site C are sludges from a wet scrubber in-
stalled for air pollution control of lea-
ther finishing dusts (SIC 3111).
The sludge, as applied, contains about
6% solids and 94% water. Inorganic analy-
ses of the solids fraction yielded the re-
sults shown in Table 4. Major elements
identified, in order of decreasing concen-
tration, are seen to be organic carbon, N,
Cr, Na, Ca, P, K, and Mg.
The extractable organics in the waste
amounted to 2200 mg/L, considerably higher
than the waste farmed at Sites A and B.
The organics in the extract were character-
ized by liquid chromatography as follows:
Aliphatic Hydrocarbons,
tile (TCO)
non-volatile (GRAY)
vo la-
Aromatic Hydrocarbons, vola-
tile (TCO)
non-volatile (GRAY)
Polar Organics, volatile (TCO)
non-volatile (GRAV)
3.6%
32.0%
1.4%
9.0%
0.5%
54.0%
HRMS analysis showed the Site C waste to
include a higher molecular weight range of
hydrocarbons (up to mass 400) than Sites A
and B. Among the hydrocarbons are small
quantities of polynuclear aromatics. The
C, H, N species in the waste are dominated
by a high molecular weight series (C^HsyN,
C22H35N> etc.). The C, H, 0 species in the
waste are primarily fatty acids (both sa-
turated and unsaturated) with a molecular
weight range up to about 400. No organic
sulfur species were detected.
Characteristics of the Site as a Waste
Treatment System — Site C is a farm that
includes wheat, hay and corn fields as well
as a small orchard. Although a variety of
tannery wastes have been landfarmed at the
site since 1961, the particular field
sampled (planted in corn) had had only one
waste application, in the spring, three
months prior to our sampling.
The field sampled consists of deep,
well-drained silt loams. The pH (6.8) is
near neutral and the cation exchange capa-
city (10.3 - 14.2) is moderately high. Al-
though the areas sampled had a gentle slope
(<3%), the rest of the farm consists of
rolling hills with slopes of 3-8%. Aver-
age annual precipitation is about 41 inches,
distributed fairly evenly over the year.
Average daily maximum summer temperature is
about 86°F. Average daily minimum winter
temperature is about 26°F. The frost free
period averages 194 days.
Operational Procedures — Waste is trans-
ported to the site in a fiberglass lined,
1300-gallon tank mounted on a truck. At
the site, the waste is deposited into an
unliried holding basin. Settled sludge is
periodically removed from the basin with a
shovel or front-end loader. It is spread
on surface soils to depths up to 6 inches
by hand, tractor or manure spreader. In-
corporation is carried out via standard
agronomic procedures. Post-application
monitoring is limited to pH and fertilizer
requirements for the farm.
Roughly 3000 gallons of waste had been
incorporated into the 1 acre area sampled,
about 3 months before our arrival.
Effectiveness of Treatment — If all of the
waste applied had simply accumulated, con-
centration increases would have been ob-
servable in the application area soils for
C, N, Cr, Ca, and Na. Increases were ob-
TABLE 4. SITE C WASTE ANALYSIS - TOTAL METALS. T.O.C.. AND T.K.N.*
Ele
As B
Ca
Cd Cr Cu K Mg Mo N Na Ni
Pb
Se V Zn
Concen- 0.6 2.4 44.25% 1200 2.8 5600 14 390 360 1.1 8.10% 1300 1.0 937 54 <0.1 5.2 50
tratlons
(in yg/g of dried solids, except as noted)
268
-------
served for C, N, and Cr. Predicted and ob-
served levels of increase are remarkably
close, as shown in Table 5. This suggests
that no migration of these elements later-
ally or to a depth of 12" had occurred.
The levels of extractable organics in
the application area soils were appreciably
higher than the corresponding levels in the
control soils. Furthermore, levels in the
0-6" depth samples of the application area
were ten times those in the 6-12" depth
samples. This suggests that little degra-
dation had occurred in the three-month
period between waste application and site
sampling. Further evidence is provided by
the close correspondence between the or-
ganics in the waste and the organics in
the application area soils as determined by
HRMS analysis. In view of the high mole-
cular weight of the organics and the short
time elapsed between application and samp-
ling, little degradation would have been
expected.
Potential Environmental Impacts — These
are discussed for Site C as a whole (50
acres) rather than for the small area
sampled (1 acre).
Portions of the site contain poorly
drained subsoils, and a seasonally high
water table that comes within 3 feet of
the surface. Small wetland areas occur at
several locations throughout the site.
Thus the potential exists for wastes to
come into direct contact with the ground
water, but the impact is unclear.
Chromium had clearly accumulated in
the application area soils sampled, and if
present in trivalent form as an insoluble
oxide, little migration would be expected.
On the other hand, corn samples grown in
the application area showed small increases
in chromium levels, compared to corn grown
in the control area. More work is needed
on the form(s) of chromium present.
The fate and potential impacts of the
organic components of the waste cannot be
assessed from the data obtained in this
study, because of the short time the waste
had been in the soil at the time of samp-
ling. Additional sampling and analysis
would have to be done at intervals of sev-
eral months or more to detect possible de-
gradation.
The farmer uses the waste as a supple-
mentary source of nitrogen and as a condi-
tioner to improve the water retention of
the soils. State regulatory authorities
also view the sludge as a useful by-product
rather than as a waste.
TABLE 5. PREDICTED AND OBSERVED CONCENTRATION CHANGES IN
SITE C APPLICATION AREA SOILS
Predicted1
Observed
Element
C
+1.1%
+ .87%
Cr
+130
+250
N
+1900
+ 950
1. Predicted concentration change =
[(Kg solids/acre/yr.) x (ppm of element in waste)]
6
2 x 10 kg soil/acre-ft.
2. Observed mean concentration change =
[(ppmAPN 0-6" + APN 6-12")"(ppmCTL 0-6" + CTL 6-12");
2
269
-------
Costs — Capital costs attributable to
landfarming at Site C are minimal ($2500).
Annualized operating costs come to $46507
yr or a little over 3c/gallon.
Site D/Petroleum Refining Industry Sludge
Physical and Chemical Composition of Wastes
Treated — The wastes being landfarmed at
Site D are sludges from holding ponds,
which are part of the wastewater treatment
system of a petroleum refinery (SIC 3111).
laa sludge, as applied, was reported to con-
tain about 13% solids, 25% oil and 62% wa-
ter. Inorganic analysis of the solids
fraction yielded the results shown in Table
6. The major elemental components in order
of decreasing concentration, are seen to
be: organic carbon, P, Na, N, Ca, Cr, K,
Zn, and Mg. The extractable organics in
the waste amounted to 1100 mg/L, charac-
terized as follows:
Aliphatic Hydrocarbons, non-
volatile (GRAY) 41%
volatile (TCO) 39%
Aromatic Hydrocarbons, non-
volatile (GRAV) 10%
volatile (TCO) 3%
Polar Organics, non-volatile
(GRAV) 7%
HRMS analysis showed the presence of a
variety of aromatic hydrocarbons, including
polynuclear aromatics, small quantities of
oxygen and nitrogen containing species, and
somewhat larger amounts of sulfur-contain-
ing species.
Characteristics of the Site as a Waste
Treatment System — Surface soils on the
site range from silty clay to silty clay
loam, with clays predominant in upper
levels. Soil depths to bedrock vary from
more than 60 inches to as little as 10
inches. Soil pH is on the alkaline side
(7.8 - 7.9), and cation exchange capacity
is high (26.9 - 28.5). This should favor
ion exchange and precipitation reactions,
which would immobilize heavy metals. Rates
of microbial decomposition may be slow due
to the limited rainfall (11-14 inches/yr),
and generally cool climate (88°F, average
daily summer maximum; 10.6°P, average
daily winter minimum;. The frost free per-
iod is 120-125 days.
Operational Procedures — Wastes are trans-
ported by tank truck, from which they are
applied to the site surface by gravity
flow. Roughly 68,000 gallons of waste had
been applied in 1974 to the 1.7 acres that
«rere sampled of the 20 acre site.
Following application, the wet sludge
is allowed to dry out on the ground surface
over a period of several weeks. The dried
wastes are incorporated into the soil to a
depth of 5-6 inches with a roto-till.
Roto-tilling is repeated at frequent in-
tervals, and fertilizer applications have
been made on an experimental basis to in-
crease rates of microbial degradation.
Post-application monitoring includes
periodic analyses for the oil content of
the soil and classes of organic degradation
products.
Effectiveness of Treatment — Significant
increases above background were observed in
the application area soils for organic C,
Cr, Na and Zn. Significant increases in
SOii", CS~ and electrical conductivity were
also observed in application area soils.
Interpretation of the results is compli-
cated by the reported presence of a
saline seep below the application area
soils.
TABLE 6. SITE n WASTE ANALYSIS - TOTAI, METALS., T.O.C.. AND T.K.M.*
Element As
Ca Cd Cr Cu K Mg Mo N Na Ni P Pb Se V Zn
Concen- 105 39 28.24% 1.5% 9.6 1.3% 310 7900 5400 22 2.08% 2.1% 19 3.1% 89 150 39 6500
tratlons
(in pg/g of dried solids, except as noted)
270
-------
The 1973-74 application area 0-6" depth
(APNO-6) and 6-12" depth (APN6-12) samples
were found to contain 3000 mg/kg and 520
mg/kg respectively of extractable organics.
These values are substantially higher than
corresponding values for the control area
of 140 mg/kg and 89 mg/kg. Thus complete
biodegradation had not yet been achieved.
However, the APNO-6 sample was found to be
much richer in polar organics than the!978
waste, and the enrichment is even larger
in the APN6-12 sample. The shift towards
polar organics may reflect oxidative de-
gradation of waste components in the soil
and possibly some downward migration. On
the other hand, the HRMS analysis indicates
that a substantial quantity of aliphatic
and aromatic hydrocarbons from the waste
had not undergone oxidative degradation.
Volatile aliphatic and aromatic hydrocar-
bons appear to have been preferentially
lost from the application area. Non-
volatile hydrocarbons showed a higher aro-
matic to aliphatic ratio in the application
area samples than in the waste, possibly
because of a higher rate of biodegradation
of the aliphatics.
Potential Environmental Impacts — Berms
have been constructed around application
areas to protect against runoff in a 100-
year storm. The surface stream along the
northeast border of the site is very high in
oackground dissolved solids (over 20,000
TDS). Ground water may be as much as 200
feet below the surface, and possibly brack-
ish. Site D is in a relatively isolated
rural area, more than one kilometer from
the nearest area of human habitation. The
site is surrounded by a cyclone fence,
which prevents access by cattle.
Costs — Capital costs for land, site ap-
proval, and site preparation were $62,500.
Equipment is rented, and charged against
operating costs. The latter, including
capital amortization amount to nearly
$13,000/year or a little over 37c/gallon of
waste treated.
Site E/Fabricated Metal Industry Sludge
Physical and Chemical Composition of Wastes
Treated — The wastes being landfarmed at
Site E are neutralized pickling liquor
sludges from a facility producing fabri-
cated metal products (SIC 349).
The sludge as applied, contains about
20% solids and 80% water. Inorganic analy-
sis of the solids fraction yielded the re-
sults shown in Table 7. Among the elements
analyzed, the waste is seen to be relatively
high in Ca (due to lime neutralization), P,
C, K and Mg. Among the elements not anal-
yzed (50% of the dried waste weight), iron
would clearly account for a major fraction.
The waste is purely inorganic.
Characteristics of the Site as a Waste
Treatment System — The soils range from
light silt loam to light silty clay loam.
The area sampled was slightly acid (pH =
5.8 - 6.2), and had a moderate cation ex-
change capacity of 14.9 to 15.3 meq/100 g
of soil. Average annual precipitation is
around 30 inches, with 40% occurring in
summer. The frost-free period is about
150 days.
Operational Procedures — The lime-neutral-
ized pickling acid waste is transported to
the site by tank truck. The waste is
either applied to the soil immediately by
gravity flow through a stove pipe at the
side of the truck, or stored in a clay-
lined storage pond for later application.
The site has been in use for 5 years.
About 600,000 gallons/yr of wet sludge are
applied. The most recent application had
been one day prior to our sampling visit.
Wastes are applied on a weekly basis,
and then incorporated into the soil by a
TABLE 7. SITE E WASTE ANALYSIS - TOTAL METALS, T.O.C., AND T.K.N.*
Element As B
Ca Cd Cr Cu
Mg Mo N Na Ni P Pb Se V Zn
Concen- 0.2 26 0.12% 3.6% 1 110 92 1800 1400 7 <40 92 30 2.2% 26 <0.1 5 12
trations
(in yg/g of dried solids, except as noted)
271
-------
spring tooth harrow after up to three
months of surface application. State regu-
latory authorities require that a phosphate
fertilizer be applied to the soil-waste
mixture annually.
Monitoring is limited to measuring
soil-waste pH three times a year. The re-
sults are used to determine the need for
additional lime.
Effectiveness of Treatment — Roughly
600,000 gallons of waste have been applied
at the site for 5 years, with the latest
application occurring about a day prior to
our sampling visit. Elevated concentra-
tions of C, Ca, N, Na and Zn were expected
in application area soils compared to con-
trols, and were observed. The observed
increases are considerably higher than
those predicted on the basis of simple ac-
cumulation, however, suggesting that the
actual application was larger than esti-
mated by the generator or that the area of
application was smaller than that estimated
by the site operator, or both.
Potential Environmental Impacts — Site E
is located on a levelled hilltop. Berms
have been constructed and the site slope
contoured to minimize erosion and surface
runoff. Nonetheless, erosion gullies were
observed during site visits. Groundwater
is located at least 200 feet below the
surface. Although the moderate cation ex-
change capacity would favor immobilization
of heavy metals, drastic lowering of the
soil pH by application of wastes (from
around 6.0 to around 3.6) would promote
solubilization and migration. Large in-
creases in electric conductivity (from 250
to 4400 u mhos) and S0^= (from 320 to
38,000) in the application area, compared
to the control area, also suggest the pres-
ence of soluble salts which can be leached
downward below the zone of incorporation.
Costs — Capital investment for site prep-
aration and equipment was estimated at
around $3500. Operating costs come to
$29,000/yr or a little less than 5c/gallon.
Site F/Petroleum Refining Industry Sludge
Physical and Chemical Composition of Waste
Treated — The wastes being landfarmed at
Site F consist of API separator bottoms and
other tank bottom wastes from a petroleum
refinery (SIC 2911). The wastes are dif-
ferent in character from those being
handled at Site D.
The sludge, as applied, contains about
34% solids, 55% water and 11% oil. Inor-
ganic analyses of the solids fraction
yielded the results shown in Table 8.
Organic analysis showed the waste to
contain 77 g/L of extractable organics,
characterized as follows:
Aliphatic Hydrocarbons, vola-
tile, (TCO) 20%
non-volatile (GRAV) 31%
Aromatic Hydrocarbons, vola-
tile, (TCO) 11%
non-volatile (GRAV) 27%
Polar Organics, volatile (TCO) 8%
non-volatile (GRAV) 4%
HRMS analysis showed the waste extract to
contain a wide variety of unsaturated and
aromatic hydrocarbons (including polynu-
clear aromatics), and relatively high mole-
cular weight (200-300) sulfur containing
organics, primarily aromatic.
Characteristics of the Site as a Waste
Treatment System — The site is a former
wetland, built up with dredge spoil and
other types of fill material. The pH is
alkaline (7.8 - 8.0) and the cation ion ex-
change capacity is relatively high (22.9 -
31.7 meq/100 g soil). Average annual pre-
cipitation is about 48 inches, distributed
fairly evenly throughout the year. The
site is located in a warm climate (average
TABLE 8. SITE F WASTE ANALYSIS - TOTAL METALS. T.O.C.. AND T.K.N.*
Element As B
Ca Cd Cr Cu K Mg Mo
Na Ni P Pb Se V Zn
Concen- 8 4 46.46% 8600 2 1600 160 1900 3400 10 0.23% 3000 44 750 93 3 26 1100
tratlons
(in Vig/g of dried solids, except as noted)
272
-------
maximum summer temperature of 94°F and
average minimum winter temperature of 43°F)
and is essentially frost-free year round.
Operational Procedures — Waste is trans-
ported by vacuum tank truck and dumped at
one edge of the site. Several telephone
poles in tandem, drawn by a tractor, are
used to spread the waste over the 7 acre
site. Roughly, 1,500,000 gallons/yr of
waste have been applied for over ten years.
Following application, the waste is
allowed to dry for 1-3 weeks. Incorpora-
tion is by rototilling and/or discing.
Further discing is done at monthly inter-
vals.
Post-application monitoring is done
monthly on the soil-waste mixture. Samples
are analyzed for percent oil, percent water,
and for 19 heavy metals.
Effectiveness of Treatment — The site has
been farmed intensively for more than 10
years. Application area soils showed
higher levels of organic carbon, Ca, Cr,
Mo, N, Na, Pb and Zn than control area
soils. Increases were also seen in elec-
trical conductivity, S0^=, and CUT in ap-
plication areas compared to controls. Or-
ganic analysis showed much higher levels of
extractable organics in the application
area than in the control area, but results
are difficult to interpret since wastes
had been freshly incorporated one day prior
to our sampling visit. A slight enrichment
in polar material did indicate that some
oxidative degradation had occurred.
Potential Environmental Impacts — Site F
is located in a flat lowland, of signifi-
cant subsidence. Although very high berms
surround the site, there is a potential for
flooding and washout in the event of an un-
usually severe storm.
Application rates have been exception-
ally heavy for a long period of time, and
the analyses discussed above show signi-
ficant quantities of waste material down to
a depth of 12 inches. Deeper core samples
would need to be analyzed to assess the im-
pact potential on groundwater or the nearby
estuary.
Costs — Capital investment for land, site
preparation and equipment is estimated at
$310,000. Operating costs are about
$208,000/yr or slightly over 4c/gallon.
CONCLUSIONS
No adverse environmental impacts were
positively identified at any of the six in-
dustrial landfarming sites investigated.
However, a number of potential problem
areas were found which should be explored
via continued sampling and analysis of
soil-waste mixtures over time. In sites
handling wastes high in salt, the possi-
bility for migration of sodium below the
zone of incorporation should be checked.
A great deal of additional work is re-
quired on the mechanisms of in situ degra-
dation of organics in soil systems. Such
work should include not only the rate of
disappearance of the principal organic
components of the wastes, but also the
identification and rate of formation of
intermediate products of degradation.
273
-------
TOP-SEALING TO MINIMIZE LEACHATE GENERATION
Case Study of the Windham, Connecticut Landfill
Grover H. Emrich
William W. Beck, Jr.
Andrews L. Tolman
SMC-MARTIN
King of Prussia, PA
ABSTRACT
19406
After the evaluation of more than 400 landfills, the Windham, CT Landfill
was selected for the implementation of a remedial action program. For the
present study, SMC-MARTIN drilled additional wells to determine the thick-
ness of the refuse, the depth of the water table under the landfill, the
depth to bedrock and ground-water quality. An electrical resistivity
survey was conducted in order to define the areal extent of the leachate
plume.
Remedial action alternatives evaluated for this site included regrading,
revegetation, surface sealing, ground-water cutoff, and plume management.
The first three methods proved to be the most cost effective. Suction
lysimeters, pan lysimeters, staff gages, and monitoring wells were in-
stalled to determine water movement into the landfill and the effectiveness
of the remedial measures.
During the summer of 1979, a remedial program was implemented. The site
was regraded, gas vents were installed, and 15 cm (6 in) of sand and
gravel were emplaced with an additional layer of 10 cm (4 in) of fine-
grained sand washings to protect the surface seal. A 20-mil PVC membrane
seal was emplaced and covered with 46 cm (18 in) of sand and gravel into
which composted sewage sludge and leaves were disced. Revegetation will be
accomplished by hydroseeding with a mixture of grasses. Monitoring is
being conducted to determine the effectiveness of the membrane seal.
INTRODUCTION
It has been estimated that
15,000 landfills are now operating
in the United States. The Resource
Conservation and Recovery Act
(RCRA) of 1976 requires that land-
fills not meeting criteria for
solid waste disposal facilities and
practices be closed within five
years. Criteria were promulgated
on September 10, 1979 (44 FR 53438)
on the basis of which as many as
10,000 of the present operations
may be classified as open dumps and
may have to be closed. In the
actions taken for the closure of
these sites and of hazardous waste
sites, it is important to implement
practices that will minimize or
abate the negative environmental
effects that the facilities are
having. In order to develop and
implement a remedial action program
at an inoperative waste disposal
site, SMC-MARTIN was awarded a
contract by the U.S. Environmental
Protection Agency.
There are numerous remedial
action alternatives that can be
used during or after the closure of
landfills and dumps. Some of these
are passive requiring little or no
maintenance after being emplaced.
Others are active and require a
continuing input of manpower and/or
electricity. Techniques for the
reduction or elimination of water
movement into landfills may be
considered in five categories:
surface water control, passive
274
-------
ground-water management, active
ground-water or plume management,
chemical immobilization of wastes,
and excavation and reburial. The
technologies for these approaches
have been widely used in construc-
tion, but have not yet been applied
to landfill closure. For details
concerning remedial action alter-
natives, see the Fifth Annual
Research Symposium of the Solid and
Hazardous Waste Research Divi-
sion. (1)
SITE SELECTION
The initial efforts for the
project included a site selection
process based on criteria estab-
lished by SMC-MARTIN and EPA. The
criteria required that the site
have basic engineering and geologic
data available; be causing a
pollution problem both visible and
of concern to the public; be 4 hec-
tares (10 A) or less in size; be
inoperative; be geographically
accessible to SMC-MARTIN; be able
to provide co-funding; be free of
current litigation; and be repre-
sentative of a typical landfill.
Once the site selection
criteria had been established, SMC-
MARTIN evaluated more than 400
landfill sites in 32 states. Fifty
of these sites were inspected by
SMC-MARTIN personnel. A total of
15 sites were recommended to EPA as
potential candidates for study.
After further evaluation and
discussion with EPA, five candidate
sites were selected. Each of these
sites was visited by both SMC-
MARTIN and EPA and further evalu-
ation narrowed consideration to
three sites. Site investigations
and testing indicated that the
Windham, CT Landfill would be the
site for the implementation of the
project.
WINDHAM LOCATION
The Windham Landfill site is
located north of Route 6 in the
northeastern portion of Connecticut
(see Figure 1). It is immediately
adjacent to the state-owned and
operated Windham Airport (see
Figure 2). Drainage from the site
is provided by the Natchaug River
which flows to the north and west
oVERNON
O MANCHESTER
HARTFORD
GLASTONBURY
STORRS
** €
CONNECTICUT
TOWN OFWINDHAM
Figure 1. Location of the Windham Landfill, Windham, Connecticut.
275
-------
POND.
I
^7 / See*: I *• *OO' Appro*.
Figure 2. Generalized site plan of the Windham Landfill,
Windham, Connecticut.
of the landfill. The U.S. Army
Corps of Engineers' Mansfield
Hollow Dam is situated on the
Natchaug River upstream from the
landfill and provides flood control
for the area. Immediately down-
stream from the landfill is the
Willimantic Reservoir, a public
water supply. The area surrounding
the landfill is extremely irregular
with numerous water-filled depres-
sions.
WINDHAM LANDFILL OPERATION
The Windham Landfill began
operation in the 1940s as an open
burning dump in a water-filled
depression. When the entire
depression had been filled, the
Willimantic Redevelopment Authority
placed approximately 4.6 to 6.1 m
(15 to 20 ft) of demolition wastes
over the solid wastes already
present in the depression to
complete that portion of the
landfill.
Filling continued in the
western portion of the landfill
site initially using the trench
method in which solid waste is
dumped in depressions excavated
above the water table. As filling
progressed, the landfill was
brought above grade and was changed
to an area-type operation. When
completed, the landfill was ap-
proximately 1.5 to 3.0 m (5 to
10 ft) above local ground surface
and occupied approximately 10 hec-
tares (25 A). Medium to coarse-
grained sand and gravel was used as
cover material.
GEOLOGY
The Windham Landfill is situ-
ated in a fine-grained stratified
glacial drift. At the surface, the
drift is composed of buff colored,
medium to fine-grained sand and
gravel with pebbles and some cob-
bles. This material forms a cap
for a fine-grained stratified drift
276
-------
deposit which is predominately buff
to gray, silty, very fine to fine
sand. Depressions (kettles) are
common throughout the area of the
landfill. Kettles result from ice
blocks being buried in outwash from
the glaciers which once covered
this area. As the ice blocks
melted, subsidence resulted, leav-
ing depressions in the ground
surface. The eastern (older) half
of the Windham Landfill originated
as a dump in such a water-filled
kettle.
Bedrock in the area is com-
posed of granite gneiss. The
bedrock surface dips rapidly from
the vicinity of the Windham Airport
to the northwest toward the Wil-
limantic Reservoir. According to
previous studies, the depth to
bedrock ranges from at or near the
surface in the vicinity of the
Windham Airport to 30.5 m (100 ft)
or more in the vicinity of the
Willimantic Reservoir.
GROUND WATER
Ground water occurs under
water table conditions in the
unconsolidated granular deposits of
stratified drift surrounding the
Windham Landfill.(2) Stratified
drift is a highly productive aqui-
fer yielding over 6.3 I/sec (100
gallons per minute, gpm). The
average grain size of stratified
drift may vary widely. Coarse-
grained deposits are found in the
valleys in this part of Connect-
icut. Fine-grained stratified
drift deposits occur in association
with the coarse-grained deposits
yielding appreciably smaller
quantities of water.
Ground water at the landfill
site occurs at depths of as much a?
7.6 m (25 ft). The water table is
exposed in ponds lying between the
landfill and the reservoir. Ground
water moves from the area of the
landfill to the west discharging
into the Willimantic Reservoir.
Mansfield Hollow Lake, north and
east of the landfill, and the area
to the south of the Reservoir
occupied by the airport recharge
the ground water.
SITE INVESTIGATIONS
A previous investigation at
the Windham Landfill site was
conducted under the direction of
the Connecticut Department of
Environmental Resources in 1975-
1976. This study documented the
fact that the landfill was con-
taminating the local ground water
which ultimately discharged into
the Willimantic Reservoir. As part
of this study, a series of eight
wells were drilled around the
landfill (see Figure 3, Wells #1-
8); water samples were collected
from these wells and from bodies of
surface water. Water elevations in
the wells was measured and a water
table map was prepared indicating
the westward movement of ground
water from the landfill toward the
Willimantic Reservoir.
Since the western boundary of
the plume of contamination emana-
ting from the landfill toward the
Willimantic Reservoir was ill-
defined, SMC-MARTIN undertook an
electrical resistivity survey to
further define it and to provide
information for the location of
additional monitor wells. Sub-
sequent to the electrical resis-
tivity survey, four monitor wells
were emplaced to further define the
western boundary of the plume of
contamination, to determine the
depth of contamination from the
landfill site, and to further
assess the hydrology of the site.
The four wells emplaced by SMC-
MARTIN during this phase of the
study (Wells #9, #10, #11, and #12)
were sampled together with selected
wells that had been completed as
part of the 1975-76 study. The
results of chemical analysis
confirmed that contamination
emanating from the Windham Landfill
site was moving toward the Willim-
antic Reservoir.
Prior to developing the pre-
liminary design for remedial
measures for use at the Windham
Landfill site, SMC-MARTIN conducted
additional hydrogeologic investi-
gations. These included the
drilling of Well #13 to conduct an
277
-------
Q I-23-MONITORING WELLS
O 24-29-SUCTION LYSIMETERS
O 3O-33-PAN LYSIMETERS
Figure 3. Location of monitoring points at the Windham Landfill,
Windham, Connecticut.
aquifer test and drilling addi-
tional observation wells (Wells 114
through #22). In this investi-
gation, the nature and thickness of
the refuse and its relationship to
the local ground-water flow system
was assessed in order to determine
appropriate remedial measures,
their design considerations, and
their costs.
The results of this phase of
the drilling indicated that the
Windham Landfill is situated in a
fine-grained stratified glacial
drift. Bedrock in the area, com-
posed of granite gneiss, dips
rapidly to the northwest toward the
Willimantic Reservoir. Figure 4, a
cross section through the landfill
between Wells #9 and #3, shows that
the water table saturates approxi-
mately 6.1 m (20 ft) of the older
portion of the landfill. The
western portion of the landfill is
above the water table. The gray
fine-grained sand is the aquifer in
the vicinity of the landfill.
Ground water flows from the airport
and the Mansfield Hollow Dam to the
landfill and from there toward the
Willimantic Reservoir. A 24-
hour aquifer test was conducted on
Well #13. The influence of the
pumping well declined very rapidly
in the direction of the Mansfield
Hollow Lake since the lake re-
charges the ground water in the
vicinity of the landfill. The
major effect of pumpage was found
in Wells #14 and #15 immediately
adjacent to the pumping well.
Wells #18 and #19 were also af-
fected by the pumping because of
the lower specific capacity of the
refuse through which they were
drilled. Given the lack of re-
sponse from wells located between
the landfill and the airport and
from those located between the
landfill and the ponds, a majority
of the water pumped from Well #13
must have come from the direction
of the Mansfield Hollow Lake.
278
-------
DEMOLITION
WASTE
REFUSE
OUT FINC S*NO
SAND
WWtL
u
I
900
I
eoo
I
MOO
SCALE IN FEE!
Figure 4. Cross-section through the Windham Landfill, Windham, Connecticut.
-------
PROPOSED REMEDIAL MEASURES
The Windham Landfill is
located in fine-grained sands of
moderate permeability; local
surface sands and gravel were used
for cover material. The eastern
half of the landfill extends into
the water table and the western
half of the landfill is above the
water table. Bedrock is at a depth
of greater than 33.5 m (110 ft).
Ground water moves from the Mans-
field Hollow Lake and the area of
the airport through the landfill
and discharges into the Willimantic
Reservoir. A plume of leachate
extends from the landfill toward
the reservoir.
Remedial actions considered
for the landfill included: 1) re-
grading the completed landfill to
increase runoff from the surface,
2) developing diversion ditches to
carry surface water away from the
area of the landfill, 3) construc-
ting a slurry trench cutoff wall to
prevent ground-water movement
through the refuse and to lower the
water table below the refuse, 4)
developing a drainage system to
lower the ground-water table in the
area of the refuse, 5) counter-
pumping to lower the water table
below the refuse, 6) covering the
landfill with impermeable material
to prevent surface water infil-
tration, 7) revegetating the
surface.
The hydrogeologic investi-
gation conducted by SMC-MARTIN
refined the alternative remedial
measures under consideration. A
depth to bedrock of more than
30.5 m (100 ft) below the ground
surface would prevent keying a
slurry trench cutoff wall into the
low-permeability bedrock. Using
transmissivity data gathered during
the aquifer test, the approximate
head loss for slurry trenches of
varying depths were calculated. In
order to effect a 6.1-m (20-ft)
drop in the water table, i.e., to
lower it so that it would no longer
intercept the landfill, would
require a cutoff wall extending
38.1 m (125 ft) into the water
table or a total depth of 44.2 m
(145 ft). Several approaches were
considered in determining the
length of a cutoff wall that would
dewater the saturated portion of
the landfill, reduce ground-water
movement through the refuse, or
isolate the saturated refuse from
the ground water by forming a
stagnation cell in the flow system.
The first method would require a
slurry trench 944.9 m (3100 ft)
long x 44.2 m (145 ft) deep at a
cost of $3 to $4.5 million. The
second method would reduce the flow
of ground water through the refuse
using a trench 548.6 m (1800 ft)
long x 18.3 m (60 ft) deep at an
estimated cost of $530,000 to
$700,000. The third method would
isolate the saturated refuse and
provide a further reduction in
contamination over Method Two by
the construction of a trench
944.9 m (3100 ft) long x 18.3 m
(60 ft) deep at an estimated cost
of $1 to $1.5 million.
Another alternative ground-
water management technique would
employ a ground-water drain. This
method would use the existing
Mansfield Hollow Dam drainage
network by extending a subgrade
drain around the upgradient end of
the landfill. Such a system is
commonly used in agriculture and
industry to lower water tables from
.6 to 2.4 m (2 to 8 ft). In order
for the drainage system to be
effective, the outlet must be at an
elevation lower than the lowest
area to be drained. At the Windham
Landfill, the lowest outlet is in
the northeast corner of the land-
fill at an elevation of approxi-
mately 62.5 m (205 ft). Since the
prevailing ground-water levels
upgradient of the landfill are
between 63.4 and 64.6 m (208 and
212 ft), such a drain would reduce
but not eliminate ground-water flow
through the landfill.
A third method for controlling
ground-water flow through the
landfill, counter-pumping, would
employ the same technique as that
used in the aquifer test but would
involve more wells. It was cal-
culated that a ring of 80 wells,
280
-------
19.8 m (65 ft) deep and surrounding
the landfill on 9.1-m (30-ft)
centers, would lower the water
table approximately 7.0 m (23 ft).
A total of 25.2 I/sec. (400 gpm)
would be generated by such a system
requiring appropriate disposal. An
alternate form of this approach
would use fewer wells to lower the
water table and would form a stag-
nation cell in the landfill. This
would require 18 .3 I/sec (5 gpm)
wells approximately 19.8 m (65 ft)
deep. The wells would be on 18.3-
m (60-ft) centers along the up-
gradient side of the landfill. The
estimated capital cost of the
completed dewatering system would
have been $160,000. Operation and
maintenance costs for such a system
would be approximately $14 to
$15,000 per year. Added to this
would be the cost for the disposal
of the contaminated ground water
intercepted during the pumping;
this could include recycling,
direct discharge, treatment, or
land application.
It was concluded that the
slurry trenches to dewater the
landfill or to restrict the move-
ment of ground water would cost
from $0.5 million to $4.5 million.
A drainage system would not be
feasible because of the depth of
the water. A counter-pumping
program could be designed at a
present worth ranging from
$125,000 to $380,000 plus dis-
posal costs for the leachate.
After discussion with the
Project Officer and the U.S. EPA
staff, it was concluded that the
most feasible remedial action
measures would be to regrade the
landfill, cover it with a membrane
seal, and revegetate the surface.
This would eliminate most infil-
tration of precipitation into the
landfill and would lower the water
table in the landfill.
CLOSURE AND MONITORING
SMC-MARTIN was retained by the
Town of Windham to prepare closure
plans for the landfill that would
meet the requirements of the Con-
necticut Department of Environ-
mental Protection. The closure
plan included the regrading of the
landfill surface, the installation
of a PVC membrane seal over the top
of the entire 10 hectares (25 A)
site, and the application of up to
46 cm (18 in.) of cover material.
Subsequent to the placement of
cover, composted leaves and sewage
sludge are to be applied in order
to supplement the capacity of the
cover material to support vegeta-
tion. The entire area then will be
completely vegetated and the
closure procedure completed (see
Figure 5).
In the fall of 1978 it was
decided that a monitoring program
would be implemented to determine
the response of the refuse and the
underlying ground water to pre-
cipitation events. Suction ly-
simeters were placed in the older
area of the refuse to determine the
passage of moisture (leachate)
through the landfill (Suction
Lysimeters #24 and #25, see Fig-
ure 3). In the newer part of the
landfill, suction lysimeters were
emplaced in the refuse and in the
earth materials below the refuse
and above the water table (Suction
Lysimeters #26 and #27). Back-
ground suction lysimeters were
emplaced adjacent to the landfill
to determine moisture movement
through relatively undisturbed
soils (Suction Lysimeters #28 and
#29). In order to determine the
quantity of moisture moving through
the landfill, four pan lysimeters
were emplaced; three with col-
lection areas of .4m2 (4 ft2) and
a fourth with a collection area of
2.3 m2 (25 ft ) in order to provide
cross correlation of data between
the smaller and the larger col-
lection areas. Pan lysimeters were
installed in the older landfill
(Pan Lysimeter #30), in the newer
landfill (Pan Lysimeters #32 and
#33), and adjacent to the landfill
(Pan Lysimeter #31) for background
information.
In addition to these suction
lysimeters and pan lysimeters,
Well #23 was installed (see Fig-
ure 3) to collect additional
ground-water information immedi-
ately downgradient of the landfill
281
-------
^VEGETATED FINAL COVER
rPVC MEMBRANE SEAL
-REGRADED LANDFILL
DIVERSION
DITCH—,
Figure 5. Typical section through the Windham Landfill.
and to define the limit of con-
tamination from the landfill.
Staff gages were installed on
Ponds #1 and #4 located below the
landfill site to determine if the
ponds are affected by the stage of
the Mansfield Hollow Lake and how
they respond to fluctuations in
ground-water levels.
Between February and June of
1979, wells, ponds, and lysimeters
at the Windham Landfill were
sampled on a monthly basis for
water quality; water levels in the
wells and ponds were measured
weekly as was the volume of water
collected in the pan lysimeters
located on and near the landfill.
Data from this water-sampling
program indicated that leachate of
moderate strength was generated by
the old fill whereas leachate of
somewhat higher strength was
generated by the newer portion.
Leachate was found to affect the
ground water downgradient of the
fill. There is a significant
contrast between the background
water quality measured upgradient
of the landfill and that found
downgradient of the landfill.
Analysis of the results of the
weekly monitoring of ground-water
and surface water elevations and
daily precipitation data have
verified that significant infil-
tration is taking place into the
landfill. As much as 85 percent
of the water passing through the
refuse derives from the infil-
tration of precipitation. The
remaining 15 percent of the water
represents the presence of ground-
water flow through the lowermost
portion of the refuse.
IMPLEMENTATION OF A REMEDIAL ACTION
PLAN
Contracts were let in the
summer of 1979 for regrading the
landfill, laying the seal, and
applying approximately 46 cm
(18 in.) of final cover. Regrading
began in August 1979 and was
essentially completed by December
1979. The subbase for the seal was
excavated from adjacent areas,
spread and compacted. It was
decided that a 10-cm (4-in.)
minimum of fine-grained compacted
sand washings would be placed
282
-------
immediately below the seal. This
material was emplaced during Sep-
tember and October. A 20-mil PVC
seal with solvent weld seams was
placed on the site during the
period of September through Nov-
ember. It was then covered with a
minimum of 46 cm (18 in.) of local
sand and gravel which was carefully
spread in 15-cm (6-in.) layers.
The site had been essentially
completed by mid-December 1979.
Annual grasses were broadcast on it
because of the late completion
date. Perennial grasses will be
hydroseeded in the spring. A
bimonthly program of monitoring was
initiated in November 1979 to
determine the effectiveness of the
seal in preventing the infiltration
of precipitation. The suction
lysimeters, pan lysimeters, and se-
lected wells will be sampled for
the next 24 months.
CONCLUSIONS
The Windham, CT Landfill was
selected to demonstrate remedial
actions to abate or minimize
pollution from closed landfills.
The system of remedial actions used
at this site included regrading,
covering at a minimum depth of
15 cm (6 in.) with local sand and
gravel and a minimum of 10 cm (4
in.) of fine-grained sand washings,
the application of a 20-mil PVC
cover and a final 46-cm (18-in.)
vegetated cover. A monitoring
system has been installed con-
sisting of suction lysimeters and
pan lysimeters to determine the
movement of moisture through the
refuse, a ground-water monitoring
system consisting of wells to
determine fluctuations in the water
table as well as the rate and
movement of leachate in the ground
water, and a surface water moni-
toring system consisting of staff
gages in nearby ponds. Monitoring
of the landfill has continued for
several years establishing complete
baseline data, and will continue
for two years following the instal-
lation of the remedial action
alternatives to determine their
effectiveness.
ACKNOWLEDGMENTS
This paper addresses one phase
of a multiphase project being
conducted by SMC-MARTIN under U.S.
EPA Contract No. 68-03-2519, Donald
E. Banning, Project Officer. Other
phases involved the selection of an
abandoned waste disposal site for
study; the production of Guidance
Manual for Minimizing Pollution
from Waste Disposal Sites (EPA-
600/2-27-142),a comprehensive
discussion of remedial measures and
estimates of their costs; the
design and implementation of
remedial neutralization procedures
at the Windham, CT Landfill, and
the implementation of a monitoring
program to determine the effec-
tiveness of the procedures.
REFERENCES
1. Beck, William W., Jr., 1979.
Remedial action alternatives for
municipal solid waste landfill
sites. In: Proceedings of the
Fifth Annual Research Symposium,
Solid and Hazardous Waste Research
Division, MERL, U.S. EPA.
2. Thomas, M.P., Bednar, G.A.,
Thomas, C.E., Jr., Wilson, W.E.
1967. Water Resources Inventory of
Connecticut, Part 2, Shetucket
River Basin. Connecticut Water
Resources Bulletin No. 11.
283
-------
A RESEARCH PROGRAM IN WASTE MANAGEMENT TECHNOLOGY FOR CARBON FIBERS
Richard A. Carnes and Laura A. Ringenbach
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
In FY 1978 the U.S. Environmental
Protection Agency (EPA) began developing a
research and development program to address
potential problems which may arise from the
release of carbon fibers into the environ-
ment. The program consists of two parts:
Carbon Fiber Characterization and Measure-
ment Technology Development and Carbon
Fiber Waste Management Technology Develop-
ment (Figure 1). The Environmental Sci-
ences Research Laboratory is developing the
program for Carbon Fiber Characterization
and Measurement Technology. We at the
Municipal Environmental Research Laboratory
are responsible for the Carbon Fiber Waste
Management Technology program. Both of
these laboratories are part of EPA's Office
of Research and Development and Richard A.
Carnes is the Agency's Coordinator for
Carbon Fiber Research.
EPA's budget for the first five years
of the carbon fiber program is shown in
Figure 2. By the end of this fiscal year,
it is planned that all funds will be allo-
cated for the characterization and measure-
ment program. The annual funding for the
waste management technology development
cc C
re 01
c e
en 01
n) O
OJ
-------
program is expected to range from $600,000
to $1,000,000 between FY 1979 and FY 1982.
Carbon Fiber Waste Management Technol-
ogy Development program being carried out
by EPA's Municipal Environmental Research
Laboratory was initiated in late FY 1979
with technical contracts awarded in three
task areas. During FY 1980 and 1981
additional technical contracts will be
awarded in four task areas. Long range
plans call for further contract awards in
five task areas in the FY 1982 period and
beyond. The program has three objectives
(1) characterization and assessment of the
carbon fiber problem, including market
penetration studies, (2) development and
demonstration of carbon fiber control tech-
nology (in the broadest sense, including
process changes, etc.) and (3) assessment
of the legal, economic and social impacts
of carbon fiber regulation and control.
The three tasks which were initiated
during late FY 1979 are described below.
A complete assessment will be performed by
Bionetics of existing information on the
environmental implications of the carbon
fiber problem including hazards, ambient
concentrations and geographical distribu-
tions, existing control mechanisms, disposal
techniques and risk assessment. The task
includes several subtasks. Materials for
these subtasks require a close liaison with
all programs in other agencies and an
incorporation of all pertinent information
into the EPA program. The initial subtask
is to perform a literature search to
identify problems encountered by the re-
lease of carbon fiber during handling and
disposal, uses of carbon fibers, potential
health effects, and information related to
the properties, production, manufacture,
and resins applications. Based on this
literature search, several sets of summary
data tables will be prepared. Tables will
summarize the type and number of present
and/or proposed research military aircraft
and transport aircraft which utilize a
carbon fiber. Summary tables will be
developed for information on carbon fiber
manufacturers and their products. The
objective of the second subtask is to
define risk considerations as interpreted
by various Federal agencies, compile their
risk assessment programs, and identify
areas of concern and data requirements.
The third subtask required the preparation
of a directory and locator for principal
individuals participating in carbon fiber
programs (see Appendix 1). Efforts on the
fourth subtask are ongoing. The purpose of
this subtask is to predict the average con-
centrations required to cause failure in
the types of electrical equipment used for
solid waste management; to calculate trans-
fer functions for solid waste management
facilities and enclosures for equipment, and;
to estimate free fiber characteristics at
solid waste locations. Review of present
disposal methods is an important part of
subtask 5 and accordingly a visit has been
made to the refuse incinerator at Saugus,
Massachusetts. Finally, a subtask has been
undertaken to develop scenarios for carbon
fiber life cycles from raw product to
ultimate disposal covering the range of
potential usage in commerce. Additionally,
critical points and potential areas relative
to hazards for both individuals and the
environment will be defined. We anticipate
the completion of all these subtasks early
in 1980.
A separate effort has been undertaken
by Econ of Princeton, New Jersey. The
objective of this effort is the quantifica-
tion of the current and projected uses of
carbon fiber composites in the production
of consumer products, evaluation of the
potential threats to society from accidental
discharge of carbon fibers to the environ-
ment and assessment of the economic trade-
offs associated with the use and/or
restriction in the use of this material.
As this is a 2-year effort the first major
results are not expected until late 1980/
early 1981.
Researchers at the University of
California at Berkeley are working on the
assessment of the effects of carbon fiber
composite materials in solid waste pro-
cessing. This task will involve the use of
a laboratory scale 10 ton-per-hour shredder,
a 3 ton-per-hour continuing system, and
associated equipment to investigate the
effects of processing solid municipal waste
containing projected typical amounts of
carbon fiber wastes. The assessment will
cover (1) comparison of the processed waste
with conventional municipal solid waste,
(2) effects of the carbon fiber waste on
the processing machinery, including wear
and tear and power consumption, (3) evalua-
tion of fugitive dust at various locations
in the process, worker exposure and
possible pathways to the ambient environ-
ment, and (4) preparation of a number of
refuse derived fuel (RDF) and densified RDF
285
-------
samples in order to test wear on equipment
and power consumption in comparison to con-
ventional solid waste.
Four tasks are programmed to be
initiated during FY 1980 or 1981. The task
with the highest priority is research on
incineration. Combustion tests will be
conducted on three types of pilot-scale
incinerators comparing emission character-
istics of conventional municipal solid
waste with municipal solid waste seeded
with carbon fiber material. Emission con-
trol will be the best available technology
for the control of particulate matter for
both types of waste. It is the purpose of
this study to determine if carbon fiber
materials disposed of in the municipal
solid waste sector by incineration will
pose a potential environmental hazard
should incinerator emission contain signif-
icant amount of free fibers. Investigation
will cover changes in the combustion pro-
cess itself, characterization of emitted
carbon fibers and changes in conventional
emissions. Residue from the process will
also be carefully analyzed for carbon
fiber content.
Three other tasks are planned. The
first is an evaluation of carbon fiber
waste impacts on existing and emerging dis-
posal and resource recovery systems. This
will include a study of trends in the use
and application of current and emerging
solid waste management technologies,
identification of those processes or steps
which will receive substantial carbon
fiber impact and identification of those
most likely to cause significant atmospher-
ic emissions of carbon fiber.
The second is an evaluation of
measures to mitigate the impact of carbon
fiber on municipal solid waste technologies
such as source separation, modification of
resins, production changes, modified or
partial bans and labeling.
Another task will evaluate carbon
fiber discharge test results and determine
the adequacy of current and future solid
waste processing, recovery and disposal
technology to eliminate carbon fiber
hazards. This includes identification of
technologies which must be modified or
newly developed to adequately control dis-
charges, and an estimate of the research
and demonstration efforts required.
Tasks to be initiated during FY 1982
or beyond may include:
- Evaluation of the legal, economic
environmental, social and political
impacts of instituting necessary
modifications to current and pro-
jected solid waste management sys-
tems. These impacts will be evalu-
ated in light of the various risk
assessments conducted previously by
EPA and other agencies.
- Evaluation of carbon fiber disposal
demonstration research in three
areas, (1) full-scale incineration
studies, (2) RDF and densified RDF
combustion, and (3) evaluation of a
small particle collection device for
controlling carbon fiber emissions.
EPA is pleased to have the opportunity
to participate in this meeting as it will
help to insure the development of carbon
fiber research programs which are compat-
ible with the on-going and planned programs
of other organizations. We look forward to
working with all of you in the future on
this interesting and important environmental
program.
286
-------
APPENDIX 1
DIRECTORY AND LOCATOR
NATIONAL GRAPHITE FIBER PROGRAM
FEDERAL OPERATION OR TASK LEADERS
ENVIRONMENTAL PROTECTION AGENCY
The Municipal Environmental Research
Laboratory
Richard A. Carnes
U.S. Environmental Protection Agency
Municipal Environmental Research Laboratory
Cincinnati, Ohio 45268
(513) 684-7871
FTS 684-7871
• Agency coordinator for EPA activities
• Carbon fiber waste management and dis-
posal technology development
The Environmental Sciences Research
Laboratory
Dr. J. Wagman
Environmental Protection Agency
Environmental Sciences Research Laboratory
Research Triangle Park, NC 27711
(919) 541-3009
FTS 629-3009
• Coordination of activities for EPA
activity with ESRL
• Carbon fiber emission characterization,
plus development of monitoring instru-
mentation and measurement techniques
Dr. Robert Shaw
Mr. Charles Lewis
Environmental Protection Agency
Environmental Sciences Research Laboratory
MD-47
Research Triangle Park, NC 27711
(919) 541-3149/3154
FTS 629-3149/3154
• Development of specialized measurement
and analysis techniques for carbon
fibers
NATIONAL AERONAUTICAL AND SPACE AGENCY, NASA
NASA Headquarters
Dr. Leonard A. Harris
NASA Headquarters
Codt RTM 6
Washington, D.C. 20546
(202) 755-3261
FTS 755-3261
• Coordinator of NASA activities and
research
NASA Ames Research Center
Mr. Richard Fish
NASA Ames Research Center
Mail Stop 223-6
Moffett Field, California 94035
(415) 965-5991
FTS 448-5991
• Alternate Materials Program. Fiber reten-
tion by char formation or chemical
stabilization of fiber by metal compounds
NASA Langley Research Center
Mr. Robert Huston, Program Manager
Mr. Thomas Bartron, Deputy Manager
NASA Langley Research Center
Mail Stop 231
Hampton, Virginia 23665
(804) 827-2851
FTS 928-2851
• Graphite Fiber Risk Analysis Program
Office
Dr. S.S. Tompkins
NASA Langley Research Center
Mail Stop 188B
Hampton, Virginia 23665
287
-------
Dr. S. S. Tompkins
(804) 827-2434
FTS 928-2434
• Alternate Materials Program. Nonconduc-
tive coatings on fibers, hybrid systems,
fiber modification by intercalation
NASA Lewis Research Center
Dr. J. Serafini
NASA Lewis Research Center
2100 Brookpark Road
Cleveland, Ohio 44135
(216) 433-4000 Ext. 487
FTS 294-6487
• Alternate Materials Program. Fire
resistant resins
DEPARTMENT OF COMMERCE
Office of Basic Industries
Mr. Donald Parsons
Office of Basic Industries
Bureau of Domestic Business Development
U.S. Department of Commerce
Washington, D.C. 20230
(202) 377-4033
FTS 377-4033
(301) 278-3086
FTS 922-3311
(Baltimore)
• Chief JTCG
U.S. Air Force Coordination and Direction
of Activities
Mr. Quentin Porter
Commanding Officer
Rome Air Development Center/RBTC
RAPCIRB
Griffiss AFB, New York 13441
(315) 330-3061
FTS 952-3061
U.S. Army Coordination and Direction of
Activities (Efforts now terminated)
Dr. L. R. Vande Kieft
Director
Ballistics Research Laboratory
USA AVRADCOM Attn: DRDAR-BLT-HN
Aberdeen Proving Ground, Maryland 21005
(301) 278-2632
(301) 278-2528
FTS 922-3311
(Baltimore)
U.S. Navy Coordination and Direction of
Activities
• Carbon fiber data base and dissemination
of information
National Bureau of Standards
Mr. Denver Lovett
National Bureau of Standards
Building 202, Room 216
Washington, D.C. 20234
(301) 921-3828
FTS 921-3828
• Evaluation of consumer electrical and
electronic equipment (also support to
Graphite Fiber Risk Analysis)
DEPARTMENT OF DEFENSE AND DEFENSE CIVIL
PREPAREDNESS
C. E. Gallaher
Naval Surface Weapons Center
Attn: Code CF 56
Dahlgren, Virginia 22448
(703) 663-8136
FTS 937-6011
(Roanoke)
Defense Civil Preparedness Agency
Mr. Thomas Boven
Defense Civil Preparedness Agency
Staff College
Federal Center
Battle Creek, Michigan 49106
(616) 962-6171
FTS 372-6171
Lt. Col. Lawrence Abramson
Director
Ballistics Research Laboratory
USA AVRADCOM Attn: DRDAR-BLC-HN
Aberdeen Proving Ground, Maryland 21005
• Technical information and briefing to
local governments. Fire, police emergency
services procedures for handling a carbon
fiber fire release incident
• Data and record compilation and analysis
288
-------
for carbon fiber fire release incidents
DEPARTMENT OF ENERGY
Mr. Thomas Garrity
Division of Electric Energy Systems
Energy Research and Development Admin.
200 Massachusetts Avenue, NW
Washington, D.C. 20545
(202) 376-4595
FTS 376-4595
• Effects of airborne fibers on power
generation and distribution equipment
DEPARTMENT OF HEALTH, EDUCATION AND WELFARE
Dr. Ralph Zumwalde
U.S. Public Health Service
Robert A. Taft Laboratories
4676 Columbia Parkway
Cincinnati, Ohio 45226
(513) 684-3255
FTS 684-3255
• Carbon fiber environment studies
• Morbidity mortality studies on exposure
to other fibrous materials (continuation
of existing programs as basis for
comparison)
DEPARTMENT OF LABOR
Dr. R. Hays Bell
Department of Labor
OSHA
200 Constitution Avenue, NW
Room N-3651
Washington, D.C. 20210
(202) 523-7031
FTS 523-7031
• Review and comparison analysis relative
to development of a regulation for
carbon fiber exposure in a worker
environment
DEPARTMENT OF STATE
Mr. R. Reckmales
Department of State
OES/APT/SA
Room 4333
Washington, D.C. 20520
(202) 632-5071
FTS 632-5071
• International coordination and dissemina-
tion of carbon fiber related data or
information
DEPARTMENT OF TRANSPORTATION
Transportation System Center, Cambridge,
Massachusetts
Dr. Karl Hergenrother
Transportation Systems Center/521
Kendall Square
Cambridge, Massachusetts 02142
(617) 494-2696
FTS 837-2696
• Carbon Fiber Studies Project Office
The Federal Aviation Administration
Mr. Arnold E. Anderjaska
Federal Avaiation Administration
AWS 120
800 Independence Avenue
Washington, D.C. 20546
(202) 426-8382
FTS 426-8382
• Coordination of accident data involving
carbon fiber on civil aircraft
• Notification of incidents involving carbon
fiber release from civil aircraft
CENTRAL INTELLIGENCE AGENCY
Mr. Chester Schuler
Central Intelligence Agency
OSI
Washington, D.C. 20505
(703) 351-6306
FTS (202) 351-6306
• Coordination of information for the CIA
GENERAL ACCOUNTING OFFICE
Mr. C. Boykin
General Accounting Office
Room 2220 Annex, SSA
Woodlawn, Maryland 21235
(301) 594-4430
FTS 934-4430
• General program review
OFFICE OF MANAGEMENT AND BUDGETS
289
-------
Dr. K. Mohan
Office of Management and Budget
Room 8002
New Executive Office Building
Washington, D.C. 20503
(202) 395-3935
FTS 395-3935
• Allocations of financial resources and
budget planning
OFFICE OF SCIENCE AND TECHNOLOGY POLICY
Col. Wayne Kay, Director
Mr. Benjamin Huberman, Deputy
Office of Science & Technology Policy
Old Executive Office Building
Room 483
Washington, D.C. 20500
(202) 395-3272
FTS 395-3272
• Overall direction, coordination and
control monitor for the Carbon Fiber
Action Plan
290
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1 REPORT NO.
EPA-600/9-80-010
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
DISPOSAL OF HAZARDOUS WASTES
Proceedings of the Sixth Annual Research Symposium
5. REPORT DATE
March 1980 (Issuing Date)
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO
Edited by David Shultz
Coordinated by David Black
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Southwest Research Institute
P.O. Box 28510, 6220 Culebra Road
San Antonio, Texas 78284
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
R807121
12. SPONSORING AGENCY NAME AND ADDRESS
13. TYPE OF REPORT AND PERIOD COVERED
Municipal Environmental Research Laboratory--Cin.,OH
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
Final 10/79-9/80
14. SPONSORING AGENCY CODE
EPA/600/14
15. SUPPLEMENTARY NOTES
Project officer:
Robert E. Landreth, 684-7876
16. ABSTRACT
The sixth annual SHWRD research symposium on management of hazardous waste
was held at the Conrad Hilton Hotel in Chicago, Illinois, on March 17-20, 1980.
The purpose of the symposium was two-fold: (1) to provide a forum for a state-
of-the-art review and discussion of ongoing and recently completed projects
dealing with the management 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 proceedinqs are a compilation of the papers presented
)y symposium speakers. They are divided into two volumes represent!'no the techno-
logies of Treatment and Disposal. The primary technical areas covered are:
(1) Waste Sampling and Characteristics (6)
(2) Transport and Fate of Pollutants (7)
(3) Pollutant Control (8)
(4) Waste Treatment and Control (9)
(5) Pesticide Treatment and Control (10)
Co-Disoosal
Landfill Alternatives
Remedial Actions
Thermal Destruction Techniques
Economics
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
Leaching, Col 1ection, Hazardous Materi als,
Disposal, Treatment, Soils, Groundwater,
Pollution, Waste Treatment
b.lDENTIFIERS/OPEN ENDED TERMS
Solid Waste Management,
Hazardous Waste,
Leachate, Toxic
c. COSATI Field/Group
13B
13. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (This Report}
21. NO. OF PAGES
20. SECURITY CLASS (Thispage)
22. PRICE
Unclassified
EPA Form 2220-1 (9-73)
291
U. S. GOVERNMENT PRINTING OFFICE: 1980-660-235 Region No. 5-11
------- |