Land Disposal: Municipal Solid Waste
Proceedings of the 7th Annual Research Symposium
Held at Philadelphia, Pennsylvania on
March 16-18, 1981
Southwest Research Inst.
San Antonio, TX
Prepared for
Municipal Environmental Research Lab.
Cincinnati, OH
Mar 81
U.S. DEPARTMENT OF COMMERCE
National Technical Information Service
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing
1. REPORT NO.
EPA-600/9-81-002a
2.
4. TITLE AND SUBTITLE
Land Disposal: Municipal Solid Waste
Proceedings of the Seventh Annual Research Symposium
3. 3
5. REPORT OATS
March 1981
6. PERFORMING ORGANIZATION CODE
7. AUTHOPUS)
Edited by David Shultz
Coordinated by David Black
8. PERFOR UNO ORGANIZATION REPORT NO
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Southwest Research Institute
P. 0. Drawer 28510
San Antonio, TX 78284
10. PROGRAM ELEMENT NO.
BRD1A DU109
11. CONTRACT/GRANT NO.
68-03-2962
12. SPONSORING AGENCY NAME AND ADDRESS
Municipal Environmental Research Laboratory—Gin. ,OH
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
13. TYPE OF REPORT AND PERIOD COVERED
Final - 9/20-3/81
14. SPONSORING AGENCY CODE
EPA/600/14
15. SUPPLEMENTARY NOTES
Project Officer Robert E. Landreth, 684-7871
16. ABSTRACT
The Seventh Annual SHWRD Research Symposium on land disposal of municipal
solid waste and industrial solid waste and resource recovery of municipal solid
waste was held in Philadelphia, Pennsylvania, on March 16, 17, and 18, 1981. The
purposs of the symposium were (1) to provide a forum for a state-of-the-art review
and discussion of on-going and recently completed research projects dealing with
the management of solid.and industrial wastes; (2) to bring together people concerned
with municipal solid waste management who can benefit from an exchange of ideas
and information; and (3) to provide an arena for the peer review of SHWRD's overall
research program. These proceedings are a compilation of papers presented by the
symposium speakers. The technical areas covered in the Land Disposal: Municipal
Solid Waste are gas and leachate production, treatment and control technologies
and economics. The areas covered in Land Disposal: Hazardous Wastes are hazardous
waste characterization, transport and fate of pollutants, hazardous waste containment,
land treatment of hazardous wastes, hazardous waste treatment, uncontrolled sites/
remedial action, and economics. Municipal Solid Waste: Resource Recovery include
the areas of equipment and processing, recovery and use of materials, environmental
aspects and economics/impediments and special studies.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
Leaching Collection, Hazardous Materials,
Disposal, Treatment, Soils, Groundwater
Pollution, Waste Treatment, Methane, Gases,
Linings
Solid Waste Management
Sanitary Landfills
Hazardous Waste
Leachate
13B
13. DISTRIBUTION STATEMENT
Release to public
19. SECURITY CLASS (This Report)
llnrl acci.fi ed
21. NO. OF PAGES
SS (This page)
22. PRICE
EPA Form 2220-1 (9-73)
-------
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. Scientifc and Technical AssessmentReports (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-600/9-81-002a
March 1981
LAND DISPOSAL: MUNICIPAL SOLID WASTE
Proceedings of the Seventh Annual Research Symposium
at Philadelphia, Pennsylvania, March 16-18, 1981
Sponsored by the U.S. EPA, Office of Research & Development
Municipal Environmental Research Laboratory
Solid and Hazardous Waste Research Division
Edited by: David W. Shultz
Coordinated by: David Black
Southwest Research Institute
San Antonio, Texas 78284
Contract No. 68-03-2962
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.
-------
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 concerning the land disposal of municipal solid waste.
Francis T. Mayo
Director
Municipal Environmental
Research Laboratory
111
-------
PREFACE
These Proceedings are intended to disseminate up-to-date information
on extramural research projects concerning the land disposal of municipal
solid 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.
The papers in these Proceedings are arranged as they ware presented
at the symposium and have been printed basically as received from the
authors. They do not necessarily reflect the policies and opinions of the
U.S. Environmental Protection Agency. Hopefully, these Proceedings will
prove useful and beneficial to the scientific community as a current
reference on the land disposal of municipal solid wastes.
IV
-------
ABSTRACT
The Seventh Annual SHWRD Research Symposium on land disposal of municipal
solid waste, hazardous waste, and resource recovery of municipal solid waste
was held in Philadelphia, Pennsylvania, on March 16, 17, and 18, 1981. The
objectives of the symposium were (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 solid and hazardous wastes; (2) to bring
together people concerned with municipal solid waste management who can
benefit from an exchange of ideas and information; and (3) to provide an arena
for the peer review of SHWRD's overall research program. These proceedings
are a compilation of papers presented by the symposium speakers.
The symposium proceedings are being published as three separate docu-
ments. In this document, Land Disposal: Municipal Solid Waste, three technical
areas are covered. They are as follows:
(1) Gas- and leachate production
(2) Treatment and control technologies
(3) Economics
-------
TABLE OF CONTENTS
SESSION A - MUNICIPAL SOLID WASTE
Page
Overview: Current Research on Land Disposal of Municipal Solid
Waste vii
Session A-l. Gas and Leachate Production
Leachate Production and Management from Municipal Landfills:
Summary and Assessment ,, , 1
Leachate Production by Landfilled Processed Municipal Wastes .... 18
Recovery of Fecal-Indicator and Pathogenic Microbes from Landfill
Leachates 37
Wastewater Treatment Plant Residual Landfilling: A Critical
Review 55
Leachate and Gas from Municipal Solid Waste Landfill Simulators ... 67
Gas Production in Municipal Waste Test Cells 94
Field Verification of Landfill Methane Movement and Methane
Control Systems 104
Planting Trees and Shrubs in Landfill Cover Soil .......... 116
Collection of Representative Water Quality Data from Monitoring
Wells 126
Session _A-2. Treatment and Control Technologies
Assessment of Liner Materials for Municipal Solid Waste Landfills . . 138
Field Verification of Liners . 163
Effect of Sanitary Landfill Leachate on the Activated Sludge
Process 170
Containment of Heavy Metals in Landfills with Leachate Recycle . . . 181
Session A-3. Economics
Optional Cost Models for Solid Waste Disposal ... 19J"
Summary of Landfill Research Boone County Field Site 211
vi
-------
CURRENT RESEARCH ON LAND DISPOSAL OF MUNICIPAL SOLID WASTES
Norbert B. Schemaker
John V. Klingshirn
Michele A. Gualtieri
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 municipal solid and hazardous waste management,
including both disposal and processing. This research is being directed towards new and
improved systems of municipal solid and hazardous waste management, development of tech-
nology, determination of environmental effects, and collection of data necessary for the
establishment of disposal and processing guidelines.
Division activities in the area of municipal solid waste research have in the past
related to storage, collection, transport, processing, resource recovery, disposal, and
waste-as-fuels programs. However, over the years the municipal solid waste research has
been de-emphasized. As a result storage, collection and transport research activities
have been non-existent for several years. Recently, resource recovery, waste-as-fuels
and disposal have been de-emphasized so that in FY 1981 no new funds will be available for
these research areas.
The current municipal solid waste disposal research program has been divided into
three general areas: (1) Pollutant Predictions for Current Landfill Techniques, (2)
Alternatives to Current Landfill Disposal Techniques, and (3) Remedial Action for Min-
imizing Pollutants from Unacceptable Sites.
The research activities currently funded under these three general areas have been
classified into five categories shown below:
1. Leachate Forecasting
2. Controlled Decomposition
3. Co-Disposal
4. Manuals of Practice
5. Economic Assessment
INTRODUCTION
Increasing amounts of waste residuals government agencies. Their problems are
are being directed to the land of disposal complex, involving legislation, economics,
in landfills. At the same time, there is and public attitudes as well as technology;
increasing evidence of environmental damage additionally, comprehensive information on
resulting from improper operation. The landfilling techniques and protection of
burden of operating landfills and coping the local environment is not readily
with any resulting damages falls most available.
heavily on municipalities and other local
vu
-------
The municipal waste disposal program
initiated by SHWRD was designed to docu-
ment and evaluate the potential adverse
environmental and public health effects
that could result from application of
waste disposal methods without proper
precautions for leachate and gas management.
This research strategy, encompassing state-
of-the-art documents, laboratory analysis,
bench and pilot studies, and full-scale
field verification studies, is at various
stages of implementation. The waste dis-
posal research program was intended to
develop and compile a research criteria
data base for use in the development of
guidelines and standards for waste rasidual
disposal to the land as mandated by the
recently enacted legislation entitled
"Resource Conservation and Recovery Act
of 1976" (RCRA). This intention was
actively pursued through FY 1980. At this
time, it was decided that other research
priorities (i.e., hazardous wastes) re-
quired additional financial resources and
this resulted in a de-emphasis of the
municipal solid waste (MSW) land disposal
program. This de-emphasis reduced the
funding for MSW land disposal to zero in
FY 1981 and FY 1982. As a result of this
action, no new research starts are planned
and only no cost continuation research
activities are being pursued. The current
research program in this area will termi-
nate in FY 1982. As a consequence, the
Intention of developing guidance documents
for MSW residuals disposal has not been
totally achieved. Some best engineering
judgement documents have been developed as
relates to pollutant generation and control,
but a total wrap-up and summation of all
aspects of MSW land disposal technology
will not be achieved. Certain areas as
co-disposal, leachate treatment, gas pro-
duction, migration and control are still
largely unexplored.
LEACHATE FORECASTING
The objective of this program is to
develop a technique to predict the volu-
metric production of leachate and to pre-
dict the quality or composition of leachate
from selected wastes. Achievement of this
objective will allow the necessary input
data to assessment of natural soil/leachate
interactions, an assessment of potential
environmental and public health adverse
effects, and scheduling and selection of
control processes to avoid unreasonable
adverse effects.
One recently completed effort (1) is
the monitoring of simulated landfill test
cells constructed at a research field
installation of the US EPA at Walton,
Kentucky. This field site known as the
"Boone County Field Site" (BCFS) was con-
structed to evaluate the pollution poten-
tial associated with the sanitary landfill
method of municipal solid waste dipsosal.
Leachate and gas sampling and analysis,
temperature and data has been collected
and published in an interim report enti-
tled "Boone County Field Site Interim
Report, Tests Cells 2A, 2B, 2C and 20",
EPA-600/2-79-058. A final report will be
published in 1981 and will include results
from all the simulated test cells.
A second completed effort (1) compiled
a description and critical analysis of
landfill leachate information to advance
the state-of-the-art for management of
landfill leachates from municipal solid
wastes. The report will provide practical
information on the generation, significance,
and costs for controlling of landfill
leachates for use by design engineers and
landfill operators. This report will be
published in the near future after techni-
cal reviews are completed.
Several other on-going and completed
efforts (1) relating to leachate and gas
migration and movement are discussed in
another paper being given at this symposium
entitled "Current Research on Land Disposal
of Hazardous Wastes".
CONTROLLED DECOMPOSITION
The objective of this program is to
develop techniques which may be used to
control the quality and volume of leachate
production. Qua!ity control is achieved
by slowing or accelerating the decomposi-
tion processes. Volumetric control of
leachate production is achieved by im-
proving infiltration control, and develop-
ment of liner and leachate treatment
technology.
Haste Leachability
The objective of the initial effort
(1) which is completed, was to confirm
laboratory studies of the leachate recycle
concept with larger prototype test cells
and to elucidate information on mass flux
of gas and leachate components with par-
ticular attention to the effect of evapo-
-------
transpiration on the rates and quantities
of leachate.
Two doubly-lined test cells received
identical volumes of water and weight of
shredded solid waste. Gas was monitored for
COj. CH^, other constituents and total
volume. Leachate was monitored for BOD,
COD, TOC, total and Individual volatile
acids, alkalinity, acidity, pH, nitrogen
and phosphorus, chlorides, sulfides or
sulfates, and pertinent heavy and alkaline
earth metals.
Mass flux was determined for the con-
taminant monitored and analyses made to
interpret and control the stabilization
process. Recommended design and control
procedures for leachate containment and
recycle was then developed. An economic
assessment was also performed. The final
report will be published in 1981.
A second completed effort (1) investi-
gated the effects of preprocessing municipal
refuse prior to landfiTHng. The prepro-
cessing procedures considered were shredd-
ing, baling, and a combination of both
techniques. Leachate volume and composi-
tion of the gases produced were assessed.
The study was accomplished through the use
of landfill simulators buried in the ground.
The final report will be published in 1981.
A third completed effort (1) investi-
gated the health and environmental signif-
icance of the persistence of fecal strepo-
cocci found in leachate from landfilled
municipal refuse. This study was carried
out in two phases.
The initial phase verified microbial
analytical methods and determined the pres-
ence of study organisms in a variety of
leachates. Samples of leachate from dif-
ferent sources of landfilled waste repre-
senting different stages of waste decompo-
sition and different operational conditions
were assayed for microbial and chemical
content. Microbial assays included total
aerobic and anaerobic plate counts, Indi-
cators of fecal pollution, selected bac-
terial pathogens and anaerobes and the
major fungi of pathogenic significance.
Chemical analyses were used to describe
the leachate environment.
The second phase investigated the
relationship between the extent of waste
decomposition and the microbial population
dynamics. Three experimental lysimeters
containing municipal refuse, municipal
refuse and sewage sludge, and hospital
waste were constructed at 16" of net
infiltration per year. This final report
will be published in 1981.
The objective of the fourth effort
(1) which is currently on-going, is to
assess the impact of leachate from a
sanitary landfill on a conventional acti-
vated sludge process for treatment of
municipal wastewater. The evaluation will
emphasize the effect of leachate on process
performance and effluent quality and its
effect on the economics of process
operation.
Two parallel conventional activated
sludge processes are being operated at a
nominal flow of 20 gpm and the influent
is effluent from the primary settling
basins at the Mill Creek Sewage Treatment
Plant. Leachate is added to the influent
flow of one activated sludge process, the
other process functions as the experimental
control.
During the first phase of the project
the biological processes were operated at
low sludge ages such that nitrification
is not established. The second phase of
the project will be similar to the proce-
dure followed during the first phase
except sludge ages in both systems will
be increased in order to maintain complete
nitrification.
The fifth effort (2) which is com-
pleted, evaluated the aesthetic factors
associated with land disposal of milled
solid waste in a variety of particle sizes
without dally cover.
Four experimental test plots were
constructed containing four different
particle size distributions. The variables
evaluated were:
o effect of wind velocities on refuse
displacement
o differential settlement
o initial density and effects of time
o vegetative growth
o vectors and wildlife.
The results of this study will be
published in a final report entitled
-------
"Effects of Particle Size on landfilled
Solid Waste: Cold Climate Studies" in 1981.
Li ners/Hembranes/Admixtures
The liner/membrane/admixture technol-
ogy is being studied to evaluate suita-
bility for eliminating or reducing leachate
from landfill sites containing municipal
wastes.
The initial effort (3) for liner
materials being investigated under the
municipal solid waste program included six
admixed materials and six flexible mem-
branes. The admixed materials were:
o 2 asphalt concretes, varying in
permeability
o 1 soil asphalt
o 2 asphalt membranes, one based on an
emulsified asphalt and the other on
catalytically-blown asphalt
o 1 soil cement
The six flexible membranes are:
o Butyl rubber
o Ethylene propylene rubber (EPDM)
o Chlorinated polyethylene (CPE)
o Chlorosulfonated polyethylene
(HYPALON)
o Polyethylene (PE)
o Polyvinyl chloride (PVC)
The results of the 1-year exposure have
been discussed in a report entitled
"Evaluation of Liner Materials Exposed to
Leachate - Second Interim Report", EPA-600/
2-76-255, September 1976. The exposure of
these liner materials has'been extended to
a 5 month period. Data on gas migration,
water permeability (as In soils), osmotic,
and finger print analysis" was collected for
individual polymeric materials. Interim
results were published in a report entitled
"Liner Materials Exposed to Leachate -
Third Interim Report", EPA-600/2-79-038.
The final results of this completed effort
will be published in 1981 in a report
entitled "Evaluation of Liner Materials
Exposed to Leachate - Final Report".
In the second completed effort (3),
field verification of liners was undertaken,
The objective of this task was to obtain
samples of lining materials from existing
sanitary landfills and to evaluate and
compare the physical properties of the
sample specimens to original specifications
so that length of service may be estimated.
Permission to utilize four landfills
was obtained. Information compiled on
these landfills includes: length of time
since closure of each portion, type of
lining material, relative amounts of
municipal and non-municipal solid waste
placed in the landfill, and methods of
operation.
In the testing and evaluation portion
of this effort, changes in the physical
properties of the lining materials were
determined and in the cases of selected
soil and admixed materials, chemical
changes with depth were determined. The
results of this effort are being prepared
into a report for publication in 1981.
A third effort (3), currently under-
way, will assess preplacement procedures
used in constructing lined facilities.
Actual field procedures utilized in a)
preparing the supporting structure and
b) placing the various liner materials
common to projects requiring positive
control of fluid loss will be assessed.
Subgrade preparation and liner
placement activities have been observed
at fifteen sites to date. These sites
include landfills, wastewater impoundments,
and potable water reservoirs.
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 various toxic elements in
leachate thereby posing a threat to potable
groundwater supplies. Because the envi-
ronmental effects from landfill ing result
from not only the soluble and slowly sol-
uble materials placed in the landfill but
also the products of chemical and micro-
biological transformation, these trans-
formations should be a consideration in
management of a landfill to the extent
that they can be predicted or influenced
by disposal operations.
One completed effort (1) reviewed
methods for disposal of Wastewater Treat-
-------
merit Plant (WTP) residual by landfill ing.
Nitrate nitrogen, total dissolved solids,
and trace metals appear to represent the
greatest pollution hazard to water re-
sources for nearly all types of residuals.
Co-disposal (WTP sludge/refuse) techniques
were found to be the most cost effective.
The WTP sludge-only trench disposal and
co-disposal methods appear to be the most
environmentally acceptable. This report is
being finalized for publication in 1981.
A second effort (1) is currently
underway to determine the fate of heavy
metals released from industrial sludges in
municipal refuse during landfill decomposi-
tion stabilization. Leachate recirculation
is being used to accelerate the degradation
processes. An evaluation of this leachate
recirculation will be made.
A third on-going effort (1) is a study
of factors influencing the effect of ad-
mixing industrial sludges and sewage sludge
with municipal solid wastes. A combination
of municipal solid waste and various solid
and semi-solid industrial wastes along with
municipal digested primary sewage sludge
were added to several field lysimeters.
All material flows were measured and char-
acterized for the continuing study and
related to leachate quality and quantity,
gas production and microbial activity. A
report describing the results from the
activity will be published in 1981.
A fourth effort (4) examined potential
effects of co-disposal using municipal
solid waste landfill leachate to extract
leachate pollutants from industrial wastes
and study the leachate movement in soil
columns. A sequential batch leaching and
soil adsorption procedure was developed to
evaluate soil type reactivity to attenuate
leached contaminants. These results will
be published 1n a report 1n 1981,
MANUALS OF PRACTICE
The initial effort (5), which is com-
pleted, investigated the environmental
impact of special types of landfills. Vol-
ume reduction (millfill and balefill) and
alternate methods of waste disposal (hill-
fill and strip mine landfill) were evaluated
and compared with standard sanitary landfill
methods. The relative environmental impacts
of these methods on water resources were
determined along with site characteristics
that contribute to or prevent pollution.
Projections were also made for the probable
usefulness of each method in geographical
areas other than those studied. The
results of this effort will be reported
in 1981 in a final report publication
entitled "Environmental Impacts of
Special Types of Landfills".
A second effort (5) recently com-
pleted, was undertaken to determine: a)
if current sampling methods produce samples
that are representative of water contained
in the aquifer being monitored; b) if
groundwater samples collected in the field
must be treated on location or if they can
be brought back to the laboratory for
treatment without altering their chemical
nature; and c) which sampling and preser-
vation techniques should be accepted as
standards for groundwater sampling.
Six landfill monitoring wells were
studied using four different pumping
techniques and thirteen different sample
preservation procedures. Chemical analyses
of samples from each pumping method and
preservation procedure were conducted for
fifteen constituents.
The results of this completed project
will be published in 1981 in a final
report entitled "Groundwater Sampling and
Sample Preservation Techniques".
The third effort (3) recently com-
pleted was undertaken to determine which
tree species can best maintain themselves
in a landfill environment; to investigate
the feasibility of preventing landfill
gas from penetrating the root zone of
selected species by using gas barrier
techniques', and to identify those factors
which are most important in maintaining
adequate plant growth on completed sanitary
landfills. Ten replicates of nineteen
woody species were planted on a ten year
old completed sanitary landfill and five
gas barrier systems were constructed. Of
the nineteen species planted, black gum
proved most tolerant and honey locust least
tolerant to anaerobic landfill conditions.
Of the five gas barrier systems tested,
three proved effective in preventing
penetration of gas into the root systems
of the tested species.
The results of this study are cur-
rently being prepared for presentation
into a report for publication in 1981.
-------
ECONOMIC ASSESSMENT
The overall objective of this
research activity is to assess the con-
siderations involved in solid waste
disposal and the various alternatives
available within an economic framework.
One on-going effort (6) is evaluating
the relative importance of the various
factors affecting the cost of sanitary
landfilling of solid waste and providing
methods for evaluating the cost of combina-
tions of waste processing and sanitary
landfilling. This will identify the least
cost disposal alternative under local
conditions.
CONCLUSIONS
The laboratory and field research pro-
ject efforts discussed here reflect the
overall SHWRD effort in municipal solid
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 con-
tacting the Project Officer referenced in
the text. Inquiries can also be directed
to the Director, Solid and Hazardous Waste
Research Division, Municipal Environmental
Research Laboratory, U.S. Environmental
Protection Agency, 26 West St. Clair Street,
Cincinnati, Ohio 45268. Information will
be provided with the understanding that it
is from research in progress and that con-
clusions may change as techniques are
improved and more complete data become
available.
PROJECT OFFICERS
All the Project Officers can be con-
tacted through the Sol id and Hazardous
Waste Research Division (SHWRD), whose
address is shown above. The telephone
number for all Project Officers is:
513/684-7871.
1. Mr. Dirk R. Brunner
2. Mr. Stephen C. James
3. Mr. Robert E. Landreth
4. Dr. Mike H. Roulier
5. Mr. Donald E. Sanning
6. Mr. Oscar W. Albrecht
XII
-------
LEACHATE PRODUCTION AND MANAGEMENT FROM MUNICIPAL LANDFILLS: SUMMARY AND ASSESSMENT
James C. S. Lu
Robert D. Morrison
Robert J. Stearns
Calscience Research, Inc.
Huntington Beach, California 92647
ABSTRACT
This report is a review, analysis, and evaluation of current literature and infor-
mation on municioal landfill leachates. The overall objective is to provide documentation
of practical information, current techniques, comparative costs, and additional research
needs on the generation, composition, migration, control, and monitoring of leachates
from municipal landfills. Due to the sizeable amount of information available on the
subject matter only the conclusions are presented in this report.
INTRODUCTION
A major environmental impact resulting
from the disposal of municipal solid waste
into a sanitary landfill occurs when water
passing through the waste accumulates
various contaminants. This percolate or
leachate may enter surrounding surface
waters or underlying groundwaters and
seriously degrade the water quality. An
extensive body of literature is available
which describes the phenomenon responsible
for municipal landfill leachate generation,
composition, migration, control, and mon-
itoring. The objectives of this study
are to (1) clarify the understanding of
leachate production and management through
critical analysis of existing information,
(2) develop practical information., tech-
niques, and comparative costs which can
be used by design engineers and site
operators, and (3) identify areas where
additional research may be expected to
yield practical information. This infor-
mation is presented in terms of five
subject areas:
Leachate generation;
Leachate composition;
Leachate migration;
Leachate control; and
invironmental monitoring.
LEACHATE GENERATION
Factors Affecting Leachate Generation
In general, leachate volume generation
is determined by the following four con-
ditions:
0 Availability of water;
• Landfill surface conditions;
• Refuse conditions; and
• Characteristics of the underlying
soil.
The major factors which affect the above
conditions and subsequently affect the
leachate generation are summarized in
Figure 1.
Factors affecting the availability of
water include direct precipitation, surface
runoff, groundwater intrusion, irrigation,
water from refuse decomposition, and com-
paction of liquid waste or sludge co-
disposal refuses. Of these water sources,
primary contributor is direct precipitation.
Precipitation in adjacent areas can be
channeled onto the landfill site. If refuse
is placed in or just above the groundwater
table, a groundwater mound may saturate the
refuse. Considerable quantities of
leachate will result. Irrigation applied
-------
CONDITIONS
FA"
rt
A. [SOURCES or LEACHATESJ—
Precititaf.cn
I Surface P.unof f I
Groundwater [_
Intrusion |
Irrigation)
Rainfall: Amount; Inten-
sity, Frequency; Duration.
Sno*'ali: lenperature;
wind Speed; Snowoack
Characteristics; Site
Conditions; Antecedent
Conditions; Rainfall on
Sno*.
Surface topography (e.g..
size, shape, slope, orien-
tation, elevation, and
surface conf iouration);
Cover material; Vegetation;
Permeability, Antecedent
soil and refuse moisture
condition; precipitation.
Underflo« direction, rate
and location.
-{ Flo* rate and volume.
Refuse
Decomposition
Availability and pH level
of moisture; temperature;
presence of oxygen; «ge;
composition, particle size;
'and mixing of the refuse.
Type and amci/nt; noisture
content; moisture holding
capacity; compaction.
I Evapotrans-
1 piration
•) Surface Runoff)
-) In Ml t ration ]
Temperature; Wind; Humidity;
Atmospheric pressure; Co»er
material; Soil nolsture
content; Vegetation; Solar
radiation; Refuse character-
istics and decomposition
rate.
Surface topography; Cover
material; vegetation;
Permeability; Antecedent
soil and refuse moisture
condition; precipitation.
Cove' material; surface
topography; Vegetation;
Underdrainage conditions;
Surface runoff; Evaporation.
C. REFUSE CONDITIONS
D.
Moisture
Ret^nt ion
1 Initia
| field
'. moisture content;
cawdtv.
URDiBLYlNS SOIL
CONDITIONS
' "' to l A t tun
hoisture 1
Retention)
Permeability; M.olsture j
content; Uniformity and
thickness of the layer.
Ir.itial moisture
Field Cioadty.
content;
P * . •
PerPtd £ i 1 i '. >' . ff.G
CC"t?n"', L'ri'Crr
tr.-.; unp-,; o' sci
isture
ity and
1 layers.
Figure 1
Factors affecting leachate generation
-------
to the landfill surface may also contribute
to leachate generation. This is especially
true if the completed landfill is used for
recreational purposes. Water from refuse
decomposition is another potential contrib-
utor. Co-disposal of liquid wastes or
sludge to municipal landfills may also
contribute to leachate generation.
Water reaching the landfill surface
may either evaporate or transpire, infil-
trate through the landfill surface, or
leave the site as surface runoff. Water
entering the cover soil may only change
the soil to field capacity (field capacity
is defined as the maximum moisture content
which a soil or a solid material can retain
in a gravitational field without producing
continuous downward percolation) during
transmittal. Pathways available to the
water and the distribution of water among
them will depend principally upon the
landfill surface conditions. Surface
conditions which may affect leachate gen-
eration include vegetation, soil permea-
bility, cover material (type, dimension,
compaction, etc.), surface topography,
slope, temperature, humidity, and wind
speed above the landfill.
Once field capacity of the surface
cover material is attained, leachate will
percolate through the refuse. Retention
and transmission characteristics of the
refuse control the percolation rate.
Leachate may be channelled through the
refuse and dispersed by refuse intermediate
cover layers, or may seep through the pores
of the refuse. If channelling does not
occur leachate will not be produced in a
landfill until at least a portion of the
refuse reaches field capacity. Any add-
itional moisture will then cause leachate
movement. Underlying soil conditions can
modify both the rate and amount of leachate
generation when soils underlying and sur-
rounding the site have lower permeabilities
than the cover soils and refuse.
The quantity of leachate produced may
also differ considerably with management or
operating practices, depending on whether
the leachate is viewed as a long- or short-
term problem (1). Operational factors may
include cover material handling, watering
prior to compaction, daily variation in
compaction and cell construction, and
variation in waste composition (e.g. munic-
ipal refuse plus sludge or industrial
wastes; milled or unmilled refuse).
Concepts,and Techniques Describing Leachate
Generation
Numerous mathematical methods have
been used for quantitative estimation of
the volume of leachatt generated from land-
fills ( 7, 9,12,13,16,21,22). Although
these approaches vary, they rely upon
water balance or water budget principles.
The water balance method is based on: (1)
a one-dimensional flow model and conser-
vation of mass relationships among various
components of the leachate sources, and
(2) the retention and transmission charac-
teristics of the refuse and cover soil.
The generalized water balance relationships
are shown in Table 1 and illustrated in
Figure 2.
Equation (1) in Table 1 illustrates
the water balance relationship of the
landfill surface. Water contributed by
precipitation (W ), surface runoff (WSR),
or irrigation (W?B) will either become
surface runoff (R/, or infiltrate into the
cover soil (I). A portion of the infil-
trated water leaves by evapotranspiration
TABLE 1.
GENERALIZED WATER BALANCE EQUATION AT A
MUNICIPAL LANDFILL SITE
"S«
I • R
Input water fron precipitation
Input Mater fron surrounding surface runoff
Input Mater fron irrigation
Infiltration
Suffice runoff
PCRj • I - E
IS.
PER, • I - E - 4S$
4SR
•U )
.(} )
' "»$ ' «„ ' ««
where: PER^ and PER^ • Percolation In soil and refuse respectively
W. • Water contributed by solid waste decomposition
iSs • Change in moisture storage in soil
iSn • Change In nolsture storage 1n refuse
E • Evapotranspiratlon
and L • PERp » «GU (4)
"here: L • Leachate generation
vru " InPut Mater fron underflow
-------
(PRECIPITATION
+ IRRIGATION)
(EVAPOTRANSPJRATIOli)
COVER SOIL
(IEACHATI)
Figure 2.
Municipal landfill water balance
while a portion will recharge the cover
soil. Once the field capacity of the cover
soil is attained, vertical percolation
(PERp) will occur. The quantity of perco-
lation through the cover material can be
calculated by using equation .(2). At first
the percolate will be absorbed by the
refuse. As the landfill refuse reaches
field capacity, refuse percolate (PERR) is
generated in quantities described by
equation (3). The percolate will eventu-
ally evolve as refuse leachate (L). If
groundwater intrusion (^py) occurs,
leachate generation estimates can be
modified using equation (4).
Because leachate generation mechanisms
are site specific, any estimation method
requires knowledge of site topography,
geology, hydrology, climatology, and
meteorology. The many poorly known factors
in the water balance equations often re-
quire empirical, rational, experimental
methods. These different approaches re-
flect the wide variety of concepts
describing leachate generation. In this
report, only those techniques which are
appropriate for municipal landfills are
discussed.
Figure 3 shows the techniques appli-
cable for landfill leachate volume
estimation. As shown by this figure, a
combination of two hundred and forty (240)
different methods are available for
leachate generation calculation. Data
needed for each method is listed in the
first column of Figure 3.
The calculation (Figure 3) involves
two major steps: (1) quantification of
leachate generation factors (precipitation,
surface runoff, infiltration, etc.) and (2)
water balance calculations. Details of the
calculation procedures are presented in the
original report (11).
Discussion
The practicality and accuracy of the
leachate generation techniques were eval-
uated by comparing calculated results with
field leachate measurements. Five munici-
pal landfill sites representing different
geographical areas and site conditions
were selected. Because of the incomplete
data available (e.g. some evaluations
require lysimeter or field testing data),
this study evaluated only twenty-five (25)
methods. They consist of combinations of
the following techniques:
o Precipitation: field measurements;
o Surface runoff: rational method,
and curve number method;
-------
04TA srscc
CALCL'!.AT10.'.S
"onth! v or
-------
• Infiltration: estimation from
surface runoff, and ASCE method;
and
• Evapotranspiration: pan evapora-
tion adjusted, Blaney-Morin
equation, Blaney-Criddle equation,
Thorthwaite equation, and
Thornthwaite table.
Evaluation results show that accu-
racies of the methods range from 1.32 oer-
cent to 5389 percent, depending on the
models used and the site conditions
selected. Among the 125 individual cases
(25 methods x 5 sites) studied, 54 cases
(about 43 percent) are underestimated
for leachate generation, and 71 cases are
overestimated.
No one method estimated leachate
generation with reasonable accuracy for
various site conditions. The following
methods are found to achieve better
leachate estimation (less than 2501 of
average accuracy):
• Use Perry's table for runoff
estimation, calculate infiltration
from surface runoff, and use the
Blaney-Morin equation for evapo-
transpiration estimation;
• Use the Salvato table for runoff
estimation, calculate infiltration
from surface runoff, and use the
Blaney-Criddle equation for
evapotranspiration estimation;
» Use the ASCE rational method for
runoff estimation, calculate infil-
tration from surface runoff, and
use the Thornthwaite equation for
evapotranspiration estimation;
• Use Perry's table for runoff esti-
mation, calculate infiltration from
runoff, and use the Thornthwaite
Tables for evapotranspiration
estimation;
• Use Perry's table for runoff esti-
mation, calculate Infiltration from
runoff, and use the Blaney-Morln
equation for evapotranspiration
estimation;
The results suggest the following
methods may achieve less than 100 percent
error for the listed areas:
• Perry's rational method: northwest,
midwest, and south;
• Salvato rational method: northwest,
midwest, and south;
* Thornthwaite method (equation or
table): east and midwest;
» ASCE rational method: east;
• ASCE infiltraton method: northwest
and west;
• Pan evaporation method: midwest
and south;
a Blaney-Morin method: northwest and
midwest; and
9 Blaney-Criddle method: northwest
and west.
However, the described comparisons are
based.on limited field data. Part of the
error may actually result from field moni-
toring: gathering representative leachate
from a landfill is difficult. No landfills
were designed specifically for the purpose
of evaluating leachate generation models.
Complete information for the purpose of
leachate generation estimation is not
available. For example, the runoff co-
efficients used in these example calcula-
tions were selected from tables used for
other engineering or agricultural purposes
and may not be suitable for landfill
surface conditions. In order to accurately
assess leachate production models .addition-
al research designed for landfills is
required.
LEACHATE COMPOSITION
Chemlcaland Mlcroblological Composition
Figure 4 illustrates concentration
ranges of leachate chemical composition
results obtained through a comprehensive
literature survey. As shown by Figure
4, concentration levels of leachate
constituents vary considerably. They may
range from trace levels for certain
parameters to 90,000 mg/1 for others.
Identical constituents in different tests
also demonstrated variations in the
concentration levels of up to four orders
of magnitude difference.
Several classes of organic compounds
have been identified in landfill leachates
( 2, 3, 4, 8,10,17,18). In general, they
can be classified into three groups:
(1) fatty acids of low molecular weight,
(2) humic, carbohydrate-like substances of
high molecular weight, and (3) fulvic-like
substances of intermediate mc-1ecular weight
Researches have revealed that organic
constituents in landfill leachates are
largely a factor of landfill age. For
relatively unstabilized landfills, results
-------
"
ID-1
tf
i
jitf
ID
1
UD
1
^
1
"" *% *S t**.
§
^
S
§
V*
§s
^
^
§s
^
^
roc
i
IK>
,.•«»-
I
_ I
SS
1
vss
I
1
I
M>
irf1
a?
IB2
10
1
,1
Hwffl
G&,
I*S3
1
1
w/mmw///mm'/////////^
t
*
if*
i
O-P
i
MV
»r
i
1M>
i
i
-
t,
•,
Wr
1
fl
c
hH
1
1
Tffl
10"
1)5
rf
.10
.01
I , , I
_
c*
1
1
N*
«
ss
s
S
»
»
SS
«
»
s
s
SS
^
K
1
*
1
"l
\
mf*^
to
•»
I-fi
SS
w
INS
SS
«
^
^
^
^
^
ZN
1
a?
10
i
§
.m
nrn
-
Cu
*IIM.
«
SS
S
1
h
Co
."•H-M
1
1
«*»•
«
ss
S
»
S
«
§§
§
•<.
sc
I
-
«
' CD
I
•,
!
f*
tL
•MM
1
1
FIGURE 4 . RANGE OF MUNICIPAL LANDFILL LEACHATE CONSTITUENT CONCENTRATIONS
-------
105
«
9
fyjM) NOIWH
8 10
3
*
&C,
I-
I"'
m
[
tuswm
WCXSOH IHITS)
I
-
*150
•120
«60
*30
0
-60
§ -«
3
J-120
£
-ISO
-180
-210
-
•
-
-
.
-
1
1
sg
^
•
"
^
i I H I
1 2 3 H
FIGURE ^
5%%%5%%2%%^3
II lit
56789
(CONCLUDED)
10
show that up to ninety (90) percent of the
soluble organic carbon can be accounted
for as short-chain volatile fatty acids;
acetic, propionic, and butyric acids
being present in greatest concentrations.
The next largest fraction was usually
fulvic acids with a relatively high
density of carboxyl and aromatic hydroxyl
groups. For relatively stabilized land-
fills, volatile fatty acids were decreased
over time and fulvic-like fractions were
increased.
Little information exists concerning
the microbiological composition and the
survival of pathogenic microorganisms
contained in landfill leachates. Bacteria
most frequently isolated from the landfills
are species of the genera Bacillus,
Corynebacterium, and Streptococcus. Other
common bacteria are species of
AchromQbacter.. Acinetobacter,
Ae ron omo nas, C1ostr1diurn, Enterobacter, .
Listeria, Hicrpcpccus, Horaxella,
Neisseria, Pseudomonas, Serratia, and
Staphylococcus. Fungi isolated from the
experimental landfill leachates include
Allescheria boydii, Cephalosporium. sp.,
Fusari urn sp. . PenTci 1 1 lum sp . , Sepedoni urn
sp., and yeast~ Very little information
is available in the literature on viruses
present in landfill leachates. Enteric
viruses have been identified in leachates
from several landfills. Reports of other
viruses found in landfill leachates are not
conclusive.
Factors Affecting Leacha'te -C
Leachate composition is a function of
numerous factors including those inherent
in the refuse mass and landfill location,
and those created by engineers and site
operators. The physical, chemical, and
biological properties of refuse are
largely uncontrollable, although shredding
of refuse changes leachate composition to
some degree. Similarily ambient air tem-
peratures and rainfall are unalterable
characteristics of a landfill site. Con-
versely, the rate of water application,
refuse permeability, refuse depth, and
temperature within the refuse bed can be
regulated.
Among the factors evaluated, landfill
age is found to be the most related factor
affecting leachate composition in the field
condition. Existing data shows that other
factors may have relationships with leach-
ate composition only under the controlled
conditions (e.g. when evaluating one fac-
tor, other factors are constant). A first-
order rate equation finds application for
expression of the relationships between
landfill age and upper or lower concentra-
tion boundaries of leachate composition.
Figures 5 to 8 are examples of such rela-
tionships.
Leachate Composition Models
The complex nature of landfill sta-
bilization and the interaction of landfill
design and operational variables create
a highly unpredictable leachate character.
The quantitative prediction of leachate
composition and contaminant flux in a
municipal landfill has received little
attention. Wigh (23) suggested an
empirical equation to express the leachate
concentration based on cumulative leachate
volume. The equation is applicable only
to test cells in which cumulative leachate
volume can be easily measured. Qasim and
Burchinal (14) developed a concept which
-------
Figure 5
104
Landfill Age vs. Leachate BQD,
in 40 so
«6£(»«J
FIGURE 7. Landfill Age vs. Leachate Organic
Nitrogen Concentration
FIGURE 6. Landfill Age vs. Leachate IDS
280
FIGURE 8. Landfill Age vs. Leachate
Chromium Concentration
-------
assumed chloride to be a surrogate para-
meter of landfill leachate composition.
Comparison of the empirical data with
theoretical estimates were accurate during
the. initial study phase, but underesti-
mated chloride concentrations as leaching
continued.
In this study, empirical equations
based on landfill age were developed to
estimate leachate composition. The
equations differ from previous modeling
efforts in that leachate concentration
histories from actual field experiments
are described rather than batch type
laboratory and pilot scale experiments.
Due to the complexity of the leachate
composition and incompleteness of the
data, only concentration maxima or minima
(e.g. boundaries) could be solved. Re-
sults of equations and rate constants are
shown in Table 2. .However, this approach
failed to indicate trends for concentra-
tion boundaries for nitrite, nitrate,
phosphate, and some trace metals. Add-
itional concentration history data may
reveal empirical relationships and
appropriate factors representing leachate
composition.
TABLE 2.
RATE EQUATIONS AND RATE CONSTANTS
FOR LEACHATE CONSTITUENT
CONCENTRATION HISTORIES
ftttc Equation
1« tt CfiixsUnt. k
Kwctfen
Orti*r
100, • JSMO."0-"*1 »9/1
COO * 880QO»" ' ag/1
TOC • MOOOe"*'*** »9/l
TVS • HOOdi""'"" »9/l
SP-COHO- JOBSOt"0'101 . «j/l
TDS • law*"0'""1 «g/l
Orj-ll • IJOt*0'"** «9/l
w4-» • uooot-°'m q/i
KO. f K0, ~R * ft,*. rt.i.
Totjl P • g.i. «.i.
so. • isaoot"8-0"* «9/i
Ui ' . IWOO.-0-"" .9,1
Cl M?fc"0-°*M »9/1
Cl ZMOOt"6'"** •«/!
m «.i. !>..>,
Cr MOf"0'901 ,,/1
Cu 10t"°-ZOt »|/1
C4 so.-0-l'" us/,
f« . .
«« . .
Pb . .
Z" . .
1* . .
t.m
0.19J
o.»
0.18S
0.10
0.07!
c.ias
0.10
n.ft.
R.a
0.07S
o.on
O.MS
».OM
fi.t
0.90
0.20
0.1ZS
.1
,*.
.,
.1.
. i
1
.ft.
.1.
.1.
.A .
.1 .
n.t
' ts C»C03
1 <*Qt «*t11
LEACHATE MIGRATION
The multitude of reactions that are
operative when contaminants enter a soil
can be qualitatively identified with
reasonable assurance, but quantitative
data relating to specific mechanisms
are not available. Considerable effort
has been placed on the development of
sophisticated leachate migration models
(19). However, due to the highly complex
leachate/soil environments, no leachate
migration model exists that can simulate
all of the physical, chemical, and bio-
logical processes occurring in a typical
landfill system. In the following pages,
conclusions of the leachate migration
studies are presented, me original
contract report provides detailed dis-
cussion of these studies,
Leachate/Environment Interactions
Soil Properties—
The soil properties most useful in
predicting mobility of leachate contami-
nants are: (1) texture {clay content);
(2) content of hydrous oxides (Fe, Mn, and
Al); and (3) type and content of organic
matter. Relative mobilities of leachate
contaminants through soils are quite vari-
able. However, a qualitative prediction
is possible if the important soil physical
and chemical properties are known.
Migration Mechanisms—
The most important physical properties
of a soil system relative to leachate
migration are diffusion and dispersion,
dilution, physical sorption, and straining.
Chemical mechanisms which can regulate the
leachate/soil interactions include precip-
itation/dissolution, sorption, complexa-
tion, ion exchange, and redox reactions.
Biological reactions affecting contaminant
migration are numerous. The following
major effects can be attributed to micro -
bial activities: redox effects, mineral-
ization, immobilization, precipitation/
dissolution, and complexation.
Migration Trends of Contaminants--
General parameters—General parameters
studied include pH, redox potential, and
alkalinity. pH levels in leachates are
usually acidic because of biodegradation
reactions. Low leachate pH levels can
10
-------
usually be increased when leachate per-
colates through the underlying soils,
mainly by the dissolution of carbonate,
oxide, or sulfide solids. The redox
condition of leachates from active bio-
degradation landfills is usually reducing.
Soluble sulfides produced in the landfill
is the main species which controls the
redox level. In the leachate/soil environ-
ment the fate of alkalinity is mainly
controlled by the precipitation/dissolution
of calcite in the soil.
Organics—Byconversion, surface
sorption, and metal-organic complexation
were found to be the major mechanisms
for the migration of organics in the
leachate/soil system. These factors may
either increase or decrease the organics
in leachates. Extensive data on the
fate of organic matter in leachate/soil
systems indicate that the migration of
organic matter is largely site specific.
Major ions—Possible mechanisms for
regulating the levels of major ions are
solubilization and ion-exchange effects.
In leachate/soil systems, chloride, and
sodium were found to be relatively un-
attenuated, while calcium and potassium
were moderately attenuated.
Nutrients—In the leachate/soil sys-
tem, adsorption is considered one of the
most important mechanisms for controlling
movement of ammonia-nitrogen. The nitrate-
nitrogen is relatively mobile and not
retained by ion-exchange processes. The
most important mechanisms governing the
phosphorus migration are solubilization,
sorption, and biological effects. The
soluble phosphate levels in the neutral
leachates are most likely controlled by
Ca5OK (PQ.)3(s) and regulated by complex-
atTon ions. Because of this phenomenon,
the calcium ion concentration or alkalinity
level in leachates will be the dominant
factor to affect the soluble phosphate
levels.
Trace Metals--The movement of trace
metals in the leachate/soil system is
extremely complex. Not only are there
numerous controlling factors, but many
unknowns beyond present state-of-the-art
knowledge are involved. The most important
mechanisms that influence mobility of trace
metals are solubilizatlon, sorption,
complexation, and dilution. Because of
these reactions, metal concentrations in
leachates may either be increased or
decreased through the underlying soils.
Chlorinated hydrocarbons--Sorption
is believed to be the most significant
factor to regulate the levels of chlorinat-
ed hydrocarbons in the leachate/soil sys-
tem. Clay minerals, iron and manganese
hydrated oxides, and organics are
responsible for the decrease of chlorinated
hydrocarbons in leachates.
Migration Models
Models available for predicting
leachate migration can be classified as
descriptive, physical, analog, and
mathematical (19). For each of these
four types of models, the following
distinctions are found: empirical or
conceptual, stochastic or deterministic,
static or dynamic, and spacial dimension-
ality of the model. Among the above
varieties, the conceptual-mathematical
models appear to be the most effective for
leachate migration prediction. Such models
are currently available with differences
between the models arising from the
number of simplifications made during the
derivation of the basic equations (mass
transport and flow equations) method of
solving the equations, and stipulated
boundary conditions. Over a hundred of
such mathematical models are available from
the literature.. However, as yet, no
model exists that can simulate all of the
physical, chemical, and biological pro-
cesses that are operative in a typical
waste disposal system.
LEACHATE CONTROL
Technologies available for municipal
leachate control can be classified into
the following four types:
• Leachate volume control;
• Leachate composition control;
• Leachate collecton; and
• Leachate treatment and ultimate
disposal.
Leachate Volume Control
Leachate volume control refers to the
control of leaching fluids into or out
of the fill. Two approaches are generally
practiced: groundwater control and surface
water control. Available control tech-
niques include the following:
11
-------
• Groundwater control measures
e Surface water control
(1) Groundwater/leachate isolation
(i) Artificial liners
(ii) Bottom sealers
(iii) Slurry trench
(iv) Grout curtain
(v) Sheet piling cutoff wall
(2) Leachate plume control
(i) Dra i n s
(ii) Hell point system
(iii) Deep well system
(1) Contour grading and surface
water diversion
(2) Surface sealing
(3) Revegetation (Land Stabiliz-
ation)
Tables 3 to 5 summarize the suitable site
conditions, limitations, costs, and
perceived effectiveness for the use of the
above control methods.
Leachate Composition Control
The major objective of leachate com-
position control is to reduce the strength
TABLE 3. GROUNDWATER CONTROL MEASURES
Control Tocnolon
Opting* Jit. Conditions
LtattatlOM
Cotti
Artificial Liner CrwndMiter noar to Uw refuse/tell
Interface. Underlying soils pntesi
• aoderotflj repld to »or> rapid poneMbllltj
1.4 i 10*> ca/soc U .1.4 i lO'J ca/toc.
•oojitrot lalUllatlon prior U landfill Ing.
Eipoctod life not ntaklltkcd. *t»tng
grcmdMtor my rapurt Uw lleor.
»r bi donoood d«rtog landfill Ing activities
OKO Installed, oxtrOBtljT difficult to
$1.00 to JS.OO
Intel Iod/j4 *
totton Sealer
Slvrry Trench
Sro«t Certain
Sao* 11 for artificial liner tni ahra <
detailed •ndor>U*d1of of tlw toll Ii
•nr to turfaco fraaduitor wd tadrock
condition. Otkor condition italUr to
then for on ortlficUl Mnr.
too* •< for » •rtlflcltl lloor •• mil «»
• ooir U wrfoco bodrock condition
Shott rlll»9 Cutoff IM> «i for jrout curtain.
Mil
Otullod «ndor«t«ad1o| of toll propartln trt
roailrcd. Only erutlul prior to >»«dftlll««.
Coitl of >klpplB9 koatonlto. l>uon oipnod U klfh ttronfth
Iwcktu nor occxr.
Cou14or*d.off«ct1«t la Mill idtk
fr .1C"5 ca/uc. Cetti oojr tat pronlbUUi
for lBr«o tctlo ow. WfUcnH to •»«> Wit
Intoorltjr of tko tail .
Carrot too potential *nd problooi idtk drtilng
tin thoot tkrmah reckjr will, oblllt, U
mlnuin latofrltjr of tko piling.
II .M U M.OO .
par IntallM ft *
ttW to t4K par
t.ttslliN) linear foot.
U4f U US7 par
tmullad |np
UU to »S*J par
tnttillod llaoor ft.
TABLE 4. LEACHATE PLUIIE C.O;lTn.OL
Control Tochnlgut flptlngn I Hi Condition
UnlUtloni
Econonlci
CffKtlTOIWII
Or* I.I
•Mr'to-twfMO aroond
•itrr; tolli vhlck Aro otilly
o>c»it>d, conilitont ground
Mtfr )»•! vltk lltt)*
fluctMttoa.
Greundiator oojt tat too iov for offoct1>9
«!• cf dnlnt. Jolll mf not b>
aacmblt to rvcovitlon. Aro< roo^lrod
for drclMor nty bt too 10rg« on fro*
for practical dralmot purpoioi.
U00-$800/icr« (ffoctlvt ntao «i»d In con-
Junction olth othtr control
approocntl lucli AI Hntri.
Moll point ijntna SOM at for drain
Uiwlly offoctUo for ihalln
doMtorlng pvrpoiai (
-------
TABLE 5. SURFACE WATER CONTROL
Control TfChntquft
I Sill Condition*
Ptrcilvid Ifftctlttnttt
Contour CridtA9 and
Virfact Itttr
Olvtrtton
Surfice Stlltno,
landfill ttabllliid; tufflcttnt
barro* ana for Mil txcavatlon
available nearby.
Borrow Mttrlll available:
lindflll Mi been ttlhll Hid
and/or grided.
St.Milled fill I ft.). N1n1«u»
of slllleawnl ind alt control
•OJilpMnt available at tM
lite.
Available toll cover tnould
be pretcnt. Differential
tettleoent 1s exttntlve
requiring large volgMt of
tell.
Suitable cover cattrial Mr
net be retdllr available.
tat venting Mr alto be
OecoBpotltlon oitts «4r b«
toilc to veflUtton If not
venttd or control lid.
(126.000 to 1242,000 *« tffecltvt control Mature gttd
for a 10 icrt both during landfill tug ind onct
landfill, the fill Mi been ttabtllied.
I1M.OOO to Kit.ZOO
for a 10 tcrt tltl*
.lit to .IX per
•crt.
CffKtltf If MttlaMnt It ilnlHl
and «Mn uted In conjunction .Ith
otMr control •eotiiret.
Pr«vldl» I buffer to •olttalre tntir*
ll>9 tM fill, ind rtductt Irotlon. It
•ott iffeetlvt Mhcn «BolO]r*d In. con-
junction Hlth t«v«r«l control atAMrit
tltid for • cli
tl.OOO.
and contaminant flux of leachate.
can be done through:
This
• Control of refuse stabilization
rates by construction and operation
features;
• Leachate recirculation; or
• Addition of municipal) sludge,
industrial wastes, or selected
sorbents.
Landfill construction and operation fea-
tures which are significant in influencing
leachate composition are the physical
characteristics of refuse, including part-
icle size (shredding) and density (com-
paction and baling); rate of water appli-
cation; landfill depth or lift height; and
landfill temperature (which can be reg-
ulated to some extent through cover materi-
al, refuse density, and lift height). In
general, higher water application rate,
smaller particle size, and higher temper-
ature increase the refuse stabilization
rates: and therefore, favor the reduction
of leachate composition. However, the
effectiveness, application range, costs,
and practicality of employing this method
is less studied.
Leachate recirculation offers several
advantages including the natural accelera-
tion of refuse stabilization and the
absence of leachate treatment systems.
Cited disadvantages are the high capital
and maintenance costs, and the need to
carefully monitor the leachate application.
Three types of leachate recirculation
processes are recognized: spray irriga-
tion, overland flow, and subgrade
irrigation. The suitable site conditions,
limitations, economics, and perceived
effectiveness of these techniques are
illustrated in Table 6.
Co-disposal of dewatered sewage sludge
with municipal refuse can increase the
refuse stabilization because of pH buffer,
increased microbial population, and
moisture content. Disadvantages include
the potential for pathogens, organics,
and metals in leachate, and odor problems.
Receipt of alkaline and nutrient
bearing industurial wastes may accelerate
stabilization and provide control over
leachate composition. A distinct lack of
data is evident for this co-disposal tech-
nique.
Laboratory studies have demonstrated
the effectiveness of various natural and
synthetic materials in removing contami-
nants from leachates. Selected materials
included bottom ash, fly ash, vermiculite,
illite, ottowa sand, activated carbon,
Kaolinite, natural zeolites, activated
alumina, and cullite. The results have
13
-------
TABLE 6. LEACHATE REC1RCULATION AND COLLECTION TECHNOLOGIES
OotlMM SU* Condition*
tffecUvtnct*
tetchUt toll*
Stelio* Dr*tfii
Nigh tv*00fit1on rcttt for ttw iltt;
lUt «»9iUttOfl «*n tJUbUlbtd. Sltf
io*lOi*r li*f BlMMtlt* pentHag runoff, itc.
mnd **«n*BllUjr vUit vxtfttug 1«cf«tf
eoffccclOft, tprty •oufpnmf, iMt 41tc*t*
ffif rtCjKlt: csUtWKt of nB0t*« fr«ffi
How, f*«or«Bl« Mil propwrtlfl
trrtajtlon teHcclten and IMC Kit* dtlt
i/tic-. ««iit» for chi fin.
lev twtfroci ctwtfJtfo**; teweftvicwt
Sh*tlo* Well ^oJflt S?it« ttMr*ta**vrf4cc »rtt«»4u*wr ce««ttlaiis
iy*Ct1C*l enty In dry «onthi; LipK«l UB.900
Ml Mithtr condttlcHis •1nS»tl«l {ftoltd $tt).
«ff«ct1i*«i*ss. haplnq HoiUi1 ITHJ
ant or durtay tht
*r aonths erlwn trtlUient
^ro*td*i *>
tr«t*0»>t
Und U •»
«rVJ
i* *!*<•
Uretly
Ctd br
UwCy H Itn thin for
Irrigation or o**f-
ror ocit
UrM)(l11t
pilr onct
ritctrle*
ffKtlVf
costs *
DM.
Bit prior t6
cylt to *"*-
til 1-xrM»r;
S300 10 SB06 prr
Set L«C*Wt* PllMM
Control
Rtprestflts 8fl ffffKtlvf
coll«ct)oft Method vfitn
See l«*ttote fliHi
* Cofltrol
for oottauH »sc
indicated that no single sorbent of those
examined can significantly reduce the con-
centration of all constituents in leachate
to acceptable levels. The results, how-
evar, indicate that sorbent combinations
can be used to reduce all the contaminants
to acceptable levels. More pilot or field
studies are necessary to identify the
practicality, limitations, costs, and
effectiveness of this technique.
Leachate Collection
Shallow interceptor drains and well
systems are most commonly used for leachate
collection. The equipment and techniques
are identical to that described for leach-
ate volume control. Table 6 gives the
suitable site conditions, limitations,
economics and perceived effectiveness of
these techniques.
Leachate Treatment and Ultimate Disposal
Some researches ( 1, 5 ) have shown
that an extended aeration activated sludge
plant can accept up to five (5) percent by
volume of leachate without seriously
impairing effluent quality. Others (1,15)
have questioned the ability of treatment
plants to remove the variety of chemicals
found in leachate. A number of studies
have been conducted which examine the
treatment efficiency of unit processes
designated specifically for landfill
leachate. Chain and DeWalle (6) per-
formed an extensive literature survey
dealing with treatment methods for
landfill leachates. They concluded that
newly formed landfill leachate is best
treated by aerobic or anaerobic bio-
logical treatment processes. Leachate
from stabilized landfills is best treated
with physical-chemical treatment processes.
Figure 9 summarizes the COD removal
efficiencies for various demonstrated unit
treatment systems. Among the methods
studied, the trickling filter, the aerated
lagoon, and anaerobic digestion processes
appear to be most effective. Substantial
reduction in inorganic contaminants
generally cannot be achieved when utilizing
those processes as described previously
(except anaerobic filter). Figure 10
indicates those treatment alternatives
that have been tried for ammonia and
heavy metals. Air stripping and activated
14
-------
AERATED tAGOQH
ANAEROBIC DIGESTION
£0 CARBON (COI.)
"ACT, c (BATCH)
AIM* OK LlnE COAGULATION
ALUK 08 U«£ COAGULATION
Lint COAGULATION
OXIDATION (OZONATION)
OXIDATION (CULMINATION)
, REVERSE OSnOSIS
> AND
AtSATlON
SO 60 10
* COD SENCVAL
1 ACTIVATED CAB80K
• REVERSE OSMOSIS
,__ STRONG'BASE AN[01 EXCHANGE RESIN
«EAK SAM 1—
0JONATIQS
ANIOM EXCHANGE RESIN
LIME COAf.uLAfEON
AERATED LAGOON TREATMENT *
CALCIUM. MVPOCHiwiTE COAGULATION
100 C? 50 n K 0
S COD BtMOVM. FOLLWING INITIAL MOlMICAl tMATMNT
Figure 9. COD removal efficiencies for the
treatment of raw leachates
sludge show high reduction in ammonia con-
centrations, while lime coagulation appears
to be the optimal alternative for heavy
metal removal. Combinations of chemical/
physical, and biological processes were
found most effective in treating a variety
of landfill leachates. Field scale treat-
ment facilities are seldom employed and
studied. Additional research is there-
fore suggested.
LEACHATE MONITORING
The Environmental Protection Agency
has published a manual titled "Procedures
Manual for Groundwater Monitoring at Solid
Waste Disposal Facilities" (EPA/530/SW-611)
which provides a detailed guidance for
leachate monitoring. This study comple-
ments the material presented in the manual
with further information which is useful
to landfill designers and site operators.
Three topic areas are Included:
• Monitoring in the zone of aeration;
• Monitoring in the zone of satura-
tion; and
• Approaches and considerations for
monitoring.
Figure 10. % removal of NH,, Fe» Zn, Ni, Cd,
and Cu by varioijs treatment
alternatives
The first two sections address available
monitoring methods, costs, and the advan-
tages and limitations of the monitoring
equipment for these two hydrologic systems.
The final section discusses the selection
of sampling areas, monitoring frequency,
monitoring parameters, and sample collec-
tion and preservation technology. Detailed
information is provided in Lu et al. (ll).
REFERENCES
1. Boyle, W. C., S. Ho, and R.K. Ham.
1974. Chemical Treatment of Leachates
from Sanitary Landfills. Journal
WPCF, 46(7).
2. Burrows, W. 0. and-R. S. Rowe. 1975.
Ether Soluble Constituents of Landfill
Leachate, Jour, Water Poll. Control
Fed., 47 (5), 921.
3. Chlan, E. S. K., and F, B. Dewalle.
1977. Evaluation of Leachate Treat-
ment. Vol. I. Characterization of
Leachate. U.S. Environmental Protec-
tion Agency, Municipal Environmental
Research Laboratory, EPA-600/2-77-186a»
210 pp.
1$
-------
4. Chain, E.S.K., and F.B. Dewalle. 1977. 13.
Characterization of Soluble Organic
Matter in Leachate. Environmental
Science and Technology, 11 (2), 158-
163.
5. Chain, E.S.K., and F.B. DeWalle. 1977. 14.
Evaluation of Leachate Treatment,
Volumn II. Biological and Physical-
Chemical Processes. EPA-600/2-77-1866.
U.S. Environmental Protection Agency,
Cincinnati, Ohio. 15.
6. Chain, E.S.K. and F.B. DeWalle. 1976.
Sanitary Landfill Leachates and Their
Treatment. Journal of Environmental
Engineering Div. -ASCE. 102(EE2): 411- 16.
433.
7. Duvel, W.A., Jr., R.A. Atwood, W.R.
Gallagher, R.G. Knight, and R.J.
McLaren. 1979. FGD Sludge Disposal
Manual. Research Project 786-1, 17.
Electric Power Research Institute,
Palo Alto, Ca.
8. Enners, L.V. 1977. Mineralization of
Organic Matter in the Subsoil'of a
Waste Disposal Site: A Laboratory 18.
Experiment. Soil Science, 126 (1):
22-27.
9. Fenn, D.G., K.J. Hartley, .T.V. DeGeare.
1975. Use of the Water Balance Method
for Predicting Leachate Generation 19.
from Solid Waste Disposal Sites.
EPA/530/SW-168, U.S. EPA, Cincinnati,
Ohio.
10. Khare, M. and N. C. Dondero. 1977.
Fractionation and Concentration of 20.
Volatiles and Organics on High Vacuum
System: Examination of Sanitary Land-
fill Leachates. Environmental Science
and Technology. 11(8):814-819.
11. Lu, J. C. S., 0. Stearns and R. D. 21.
Morrison. 1980. Leachate Production
and Management From Municipal Land-
fills: Summary and Assessment. Vol-
umn II: References and Appendices.
EPA Contract No. 68-03-2861, U.S.EPA,
Cincinnati, Ohio.
12. Lutton, R. J., G. L. Regan, L. W. 22.
Jones. 1979. Design and Construction
of Covers for Solid Waste Landfills.
EPA-600.2-79-165, U. S. Environmental
Protecton Agency, Cincinnati, Ohio
pp. 249.
Purushottam, D., G. R. Tamke, and C.
M. Stoffel. 1977. Leachate Product-
ion at Sanitary Landfill Sites. J.
Environmental Engineering Division.
ASCE, 103(EGG):981-988.
Quasium, S. R. and J. C. Burchinal.
1970. Leaching of Pollutants from
Refuse Beds. Jour. San. Eng. Div.,
ASCE, 96(SAl):49-58.
Reindl, John. 1977. Landfill Course:
Managing Gas and Leachate Production
of Landfills. Solid Waste Manage-
ment.
Remson, I., A. A. Fungaroli, A. W.
Lawrence. 1968. Water Movement in
an Unsaturated Sanitary Landfill. J.
of Sanitary Engineering Division
Proceedings of ASCE, 94(542):307-317.
Robertson, 0. M., Toussaint, C. R.,
and Jerque, M. A. 1974. Organic
Compounds Entering Ground Water From
A Landfill, EPA Report 660/2-74-077,
EPA, Washington, D.C.
Robinson, H. D. and P. J. Maris.
1979. Leachate From Domestic Waste:
Generation, Composition, and Treat-
ment. A Review. Water Research
Centre, Tr 108. pp. 38.
Roy I. Wosten, Inc. 1978. Pollution
Prediction Technqiues For Waste
Disposal Siting. EPA-SW-162C, U.S.
Environmental Protection Agency,
Cincinnati, Ohio. 440 pp.
Teiner, R. L., J. D. Keenan, and A.
A. Fungaroli. 1979. Demonstrated
Leachate Treatment Report on a
Full-Scale Operating Plant. U. S.
EPA SW-758.
Thornwaite, C. S. and 0. R. Mather.
1957. Instructions and Table For
Computing Potential Evapotranspira-
tion and the Water Balance Public-
ations in Climatology, Lab. of Clima-
tology, Drexel Inst. of Tech., 10(3):
185-311.
U.S. Environmental Protection Agency
1974. Summary Report: Municipal
Solid Waste Generated Gas and Leach-
ate (Review Draft) Solid and Hazard-
ous Waste Research Lab., National
Environmental Research Center,
16
-------
Cincinnati.
23. Wigh, Richard 0. 1975. Interim Sum-
mary Report, Boone County Field Light-
Test Cell I. U. S. Environmental
Protection Agency, Cincinnati, Ohio.
17
-------
LEACHATE PRODUCTION BY LANDFILLED PROCESSED MUNICIPAL WASTES
James M. Kemper
Ralph B. Smith
Systems Technology Corporation
245 North Valley Road
Xenia, Ohio 45385
ABSTRACT
The purpose of the project was to evaluate waste processing methods
used prior to landfilling solid wastes by comparing the leachates produced
by wastes that were processed in different ways. Five lysimeters {buried
concrete landfill simulators) were loaded with municipal solid wastes that
were: (1) shredded and baled, (2) baled, (3) baled and saturated with
water, (4) shredded, and {55 unprocessed, respectively. Leachate samples
were collected to determine moisture balances and leachate pollutant
concentrations in order to evaluate the processing methods. After equal
volumes of water were applied to all five lysimeters, the baled wastes were
found to produce large quantities of dilute leachate, while the shredded
waste produced smaller quantities of concentrated leachate. Gas composi-
tion data from the lysimeters was inconsistent but did not indicate any
obvious differences among the waste processing methods.
Introduc t ion
The objective of this project
was to study the leachate produced
by landfilled municipal solid waste.
Five lysimeters (simulated land-
fills) were used to compare the
leachate produced by refuse which
was processed in five different ways
before placement into the lysimeter
cells. The operating conditions are
the same for all five .lysimeters
(except moisture saturation for
Cell 3) so that if the emplaeed
wastes are assumed to be identical,
any identical, any differences in
leachate emanating from the
lysimeters should be due to
differences in processing of the
refuse prior to placement in the
cells,
The processing procedures of
the refuse in the lysimeters are as
follows:
Cell 1 Baled, shredded refuse
Cell 2 Baled, unshredded refuse
Cell 3 Baled, unshredded refuse
saturated with water
Cell 4 Shredded refuse
Cell 5 Unshredded, unbaled refuse
(control)
This report will compare the
refuse processing methods by
observing their effects on moisture
balances, leachate composition, and
gas compositions for the lysimeters.
The cells have been monitored
continuously since January 1975,
and monthly data taken through
May 1980 is presented and used for
graphical representations and
predicting of trends.
Lysimeter Operation
The five landfill cells are
located in Franklin, Ohio. They
are buried concrete cells containing
municipal refuse which had been
preprocessed in different ways.
The refuse was placed in the
lysimeters as described in the
following table.
18
-------
Cell Refuse
1 Baled, shredded
2 Baled, whole
3 Baled, whole,
saturated with
water
4 Unbaled, shredded
5 Unbaled, whole
Wet Weight
11,700 kg
11,700 kg
11,700 kg
10,800 kg
9,700 kg
The refuse wet weights for
Cells 1 through 3 were calculated
using average bale weights. The
cells were instrumented and plumbed
for temperature monitoring, gas
sampling, leachate collection, and
water addition.
Water additions were made
monthly to Cells 1 through 5
according to a schedule designed to
reflect the rainfall patterns of the
midwestern region of the United
States. The schedule is presented
in Table 1.
Leachate collection are
performed via plumbing connecting
each lysimeter to a central
instrumentation cell. Leachate
samples are taken monthly and
analyzed for composition. Leachate
is also quantitatively drained from
the cells each month for water
balance analyses.
Moisture Balance
Landfill Cell? 1, 2, 4, and 5
have been operated on a monthly
water addition-volumetric leachate
collection basis since January 1975.
Water additions are made each month
according to the schedule, and all
leachate is drained each month also.
Cell 3 has been operated during
this time on a saturation basis. In
the initial stages of the project,
only the minimum volume of leachate
required for analyses was withdrawn
each month. After the refuse became
saturated with water, a volume of
leachate equal to the volume of
water added the next month was
removed each month. Since no
evaporation or transpiration occurs
within the sealed cells, the
moisture content of Cell 3 is
assumed to remain constant.
When refuse is landfilled, it
will absorb moisture which
percolates through the soil cover.
The refuse will continue to absorb
water until its field capacity is
reached. At this point, the amount
TABLE 1. SCHEDULE OF WATER ADDITION
Month of year
Centimeters
equivalent
water
Water to be applied
gallons
liters
January
February
March
April
May
June
July
August
September
October
November
December
0
5.54
11.08
11.08
11.08
5.54
0
0
0
5.54
5.54
5.54
0
104,
209
209
209
104.
0
0
0
104.
104.
104.5
0
396
791
791
791
396
0
0
0
396
396
396
19
-------
of water infiltrating the landfill
should equal the amount of leachate
coming from it. But this is not
exactly what happens in an actual
landfill situation. Fenn's
definition of apparent or practical
field capacity more closely
describes what actually occurs in
nature. Fenn(2) defines field
capacity as the maximum moisture
content which a soil (or refuse) can
retain in a gravitational field
without producing continuous
downward percolation. In this
project, "field capacity" is assumed
to have been reached when continuous
leachate production has occurred.
The total moisture content of
a cell can be calculated by sub-
tracting the volume of leachate
removed from the volume of water
added. Since these cells are
enclosed, evaporation, transpira-
tion, and runoff do not occur. The
apparent moisture retained in'the
cell is calculated as follows:
M 1000 (R - L)
G
where
M = apparent moisture retained
(mi/kg),
R = water added (liters),
L = leachate removed (liters),
G = kg dry solid waste.
A summary of the moisture
balance data for Cells 1, 2, 4,
and 5 is presented in Table 2.
These data indicate that the waste
processing procedures had an effect
on the start of continuous leachate
production. The cells containing
wastes that were processed (1, 2,
and 4) started producing leachate
before Cell 5 (untreated waste).
Cells 1, 2, and 4 began producing
leachate in the fifth, second, and
the fifth, second, and fourth
months, respectively; while Cell 5,
the unbaled whole refuse, only began
producing leachate in the sixth
month.
It is also shown that Cells 4
and 5, containing unbaled refuse,
have retained substantially more
moisture than Cells 1 and 2,
containing baled refuse. The
TABLE 2. MOISTURE BALANCE SUMMARY
Moisture retention data (mi/kg refuse)
Date
12/75
12/76
12/77
12/78
12/79
5/80
Cell 1
(5/75*)
306
406
520
737
854
906
Cell 2
(2/75*)
316
426
548
774
851
911
Cell 4
(4/75*)
475
672
814
1,131
1,198
1,225
Cell 5
(6/75*)
579
797
968
1,299
1,309
1,316
* Date continuous leachate production began.
20
-------
moisture retained values for
Cells 1 and 2 are very similar in
magnitude. The unprocessed refuse
of Cell 5 has retained more moisture
than that o£ Cell 4 (shredded
waste). The two cells containing
unshredded waste (2 and 5) have
retained more moisture than their
counterparts containing shredded
waste (1 and 4).
The results indicate that
baling of refuse has an inhibitory
effect on its moisture retention
capability. The tightly-packed
bales are more difficult for water
to infiltrate than loose refuse, and
so absorption of water proceeds at a
slower rate. Edens(l) also found
that baled waste had absorbed little
moisture 1 year after emplacement
and after 8 months of leachate
outflow. Core samples of the baled
waste contained only 1 percent more
moisture than what was originally
present.
Shredding of refuse is also
shown here to decrease moisture
retention, but to a much lesser
degree than baling. Fungaroli(3)
reported that milling (shredding) of
refuse increases greatly the
saturated field capacity, but our
shredded waste appears to have not
reached field capacity yet. When it
does reach field capacity, it may
well retain more moisture"than the
unprocessed waste. Fungaroli also
noted that increasing the density of
milled refuse (by making smaller
particles) increases the amount of
that refuse remaining at a moisture
content less than field capacity.
This results in dry parcels within
the refuse mass, and this may have
occurred to some extent in our
shredded refuse.
The moisture balance figures
also show that the rate of moisture
absorption by the wastes in Cells 4
and 5 has slowed considerably
compared to the baled wastes in
Cells 1 and 2. Cell 5 retains only
a little more moisture at May 1980
than it did at December 1978, and
Cell 4's rate of moisture uptake is
only slightly more than Cell 5's in
the past year. This indicates that
the nonbaled wastes of Cells 4
and 5 are nearing their field
capacities with Cell 5 perhaps
having already reached its field
capacity.
Fungaroli found that the field
capacity of compacted nonbaled
refuse in his lysimeter was
1380 mi/kg dry refuse. This
number is somewhat larger than our
values for nonbaled refuse, and it
suggests that our nonbaled refuse is
near field capacity. Other field
capacity values for nonbaled refuse
have been reported as 350, 358, and
400 mm/m refuse depth, which convert
to 1,150; 1,140; and 1,260 mi/kg,
respectively (Wigh[61).
Leachate Data
Leachate is collected and
drained from the lysimeters each
month through drain lines at the
bottom of each cell. The volume of
leachate is measured, and a sample
from each cell is analyzed for
biological and chemical parameters.
The parameters measured each month
are acidity, alkalinity, pH,
conductivity, oxidation reduction
potential, total solids, total
volatile solids, total organic
carbon, total kjeldahl nitrogen,
chemical oxygen demand, sulfide,
total phosphorus, and the metals
iron, copper, cadmium, zinc,
nickel, chromium, and lead. Every
third month the leachates are
analyzed for volatile acids,
biochemical oxygen demand, and
chloride.
Because the wastes in each of
the five lysimeters were
mechanically processed differently
before "landfilling," the cells
differ in water retention ability.
The cells have not produced equal
volumes of leachate even though
equal volumes of water are added to
them each month. The greatest
amounts of leachate have emanated
from Cells 1, 2, and 3 containing
baled waste. Much smaller leachate
volumes have drained from Cell 4
(shredded waste), and the least
volume of leachate has emanated from
Cell 5 (unprocessed waste). But the
volume of leachate produced alone is
not indicative of its pollution
21
-------
potential. The leachates from
Cells 1 and 2 contain less pollu-
tants per liter than the leachate
from Cell 4. The concentrations of
the measured pollutants in the
leachate and also the. cumulative
masses of pollutants in the leachate
are found to be the least in Cells 1
and 2 and the greatest in Cell 4.
The differences in waste
composition among the cells and the
differences in leachate volumes
produced necessitate the use of some
basis to compare leachate
pollutants. The measure of leachate
quality chosen for our purposes is
the total mass of specific pollutant
emanating from each cell per
kilogram of dry solid waste. This
is calculated by multiplying the
measured concentration of pollutant
times the volume of leachate,
divided by the total mass of dry
solid waste placed in the test cell.
Since the cells do not produce
equal volumes of leachate in the
same month, the cumulative masses of
pollutants from each cell can be
compared on a basis of leachate
volume or on a basis of time. In
comparing mass removals of a
pollutant on a leachate volume
basis, the cumulative mass removed
by a standard volume of leachate is
compiled for each lysimeter. The
same volume of leachate is used for
all the cells, so the cell producing
the most concentrated leachate will
have the largest mass removal.
Comparing mass removals on a time
basis does not take into account the
volumes of leachate emanating from
the cells; it only considers the
total mass of pollutant removed in a
given time period. More pollutants
may be removed in a small volume of
concentrated leachate than in a
large volume of dilute leachate.
Table 3 contains the cumulative
masses of leachate parameters
removed from the lysimeters on an
equal leachate volume basis. The
numbers shown are the total masses
of parameters removed in
12,401 liters of leachate. This is
the total volume of leachate which
has thus far emanated from Cell 5.
It is seen that Cell 4 (shredded
waste) has the greatest mass
removals of 4 out of the 5
biological parameters (alkalinity,
total solids, TOC, TKN, and COD),
and has the second largest mass
removal of TKN. This finding was
expected since shredding of refuse
makes infiltration of the water more
rapid, accelerating the decomposition
process. Conversely, baling of
refuse inhibits water infiltration
and produces a more dilute leachate.
Cells 1 and 2, containing baled
refuse, have lower mass removals of
the biological parameters than the
TABLE 3. CUMULATIVE MASSES OF LEACHATE PARAMETERS PER KG
DRY REFUSE ON A BASIS OF EQUAL LEACHATE VOLUME
Parameter
Cell 1 Cell 2 Cell 3 Cell 4
Cell 5
Alkalinity, mg/kg
Total solids, mg/kg
TOC, mg/kg
TKN, mg/kg
COD, mg/kg
Iron, mg/kg
Copper, yg/kg
Cadmium, yg/kg
Zinc, mg/kg
Nickel, yg/kg
Chromium, ug/kg
Lead, ug/kg
5,450
10,500
6,040
241
9,230
273
149
60
13
539
246
598
6,960
15,200
9,950
411
23,200
714
152
76
20
789
304
654
9,420
21,200
13,300
514
36,000
1,180
490
78
10
1,060
305
564
17,200
40,200
21,700
712
50,000
1,940
152
112
200
1,950
284
856
9,280
24,200
16,500
729
34,600
540
128
84
42
759
167
577
22
-------
other three cells. Cell 3 (baled,
saturated) has mass removals roughly
equivalent to those of Cell 5
(unprocessed) for the biological
parameters. It may be that the
saturation of the waste in Cell 3
has overcome the inhibition of
moisture infiltration normally
produced by baling, thus causing
more leaching of pollutants than
would usually be expected from baled
refuse. The mass removal values for
Cells 3 and 5 are generally in
in between the values for Cells 1
and 4 and Cells 2 and 4.
Before comparing the cumulative
masses of metals leached from the
cells in Table 3, it should be noted
that most of these metals have been
detected in only trace amounts in
the monthly leachate samples. This
means that any analytical error in
the detection of these metals would
have a greater effect on the calcu-
lated cumulative mass leached than
an error in the analysis for a
leachate parameter which has a high
concentration, such as TOC or COD.
Any variations in the metals composi-
tion of the wastes among cells would
then cause a large effect in the
composition of the leachate. In
many instances, the concentration of
a metal in the leachate was below
the detection limit of the method.
Rather than taking the concentration
as zero mg/i, the detection limit
itself was taken as the concentra-
tion, and the total mass of metal
leached was calculated from that
value. This practice gives us an
artificially high number for the
cumulative mass removal, but it is
better to do this than to assume
that no metal is leached by a
large volume of leachate, simply
because the metal cannot be
detected.
Just as Cell 4 had the greatest
mass removals for the biological
parameters, it has the greatest mass
removals of five of the seven
metals. The high relative numbers
for iron and zinc especially point
out the enhanced rate of leaching
caused by shredding of the waste.
In general, the metals results do
not show the inhibition of leaching
by baling of the waste since Cells 1
and 2 have mass removals
approximately the same as those
of Cell 5 (control). Only the
mass removals of zinc from
Cells 1 and 2 are clearly
indicative of a reduced rate of
leaching when compared to the
cumulative zinc emanated from
Cell 5. The great deal of
variability in the metals results is
probably caused by the large error
possibilities associated with the
detection of very low concentrations
of copper, cadmium, nickel,
chromium, and lead in the leachates.
Since the amounts of these metals
are low in the waste placed in the
lysimeters, much variation may be
present from cell to cell in the
quantity of metal originally
available for leaching. This makes
comparisons among the cells for
these parameters difficult if not
impossible.
The plots of cumulative masses
of biological parameters leached vs.
cumulative leachate volume (Figures 1
through 5) further indicate the
effects of the waste processing
procedures. In every case, the
shredded refuse (Cell 4) has had
more mass leached than the waste
processed by other means. The
shredding has increased the surface
area of the waste and enhanced
moisture infiltration, providing
more sites for biological and
chemical activity. The baled wastes
of Cells 1 and 2 have had less
cumulative masses removed than the
control (Cell 5) showing the inhibi-
tion of moisture infiltration and
leaching caused by baling of the
waste.
Looking again at the cumulative
masses of pollutants leached from
the cells on an equal leachate
volume basis (Table 3), it is
interesting to compare the data from
Cells 1 and 2. Both cells contain
baled waste, but the waste in Cell 1
was shredded before baling. A
greater mass of every pollutant has
leached from Cell 2 than from Cell 1.
More than twice as much COD has
leached from Cell 2 than from Cell 1.
It appears that baling of shredded
waste produces bales that are even
more resistant to leaching than
23
-------
200-
TOTAL ALKALINITY
CM
O
x 160
(K
O
3
5 120
v.
w
(K
§ 80
i
40-
CELL 4
20 40 60 80 100 120
CUMULATIVE LEACHATE VOLUME (Liters) xlO2
Figure 1. Total alkalinity - mass flow vs. cumulative lo.-ichate volume.
140
160
-------
400-
N
o
x 320-
o:
o
o
^ 240-
CO
2E
<
o:
o
-, 160
80-
TOTAL SOLIDS
CELL 4
CELL 3
20 40 60 80 100 120
CUMULATIVE LEACHATE VOLUME (Liters) xlO2
140
160
Figure 2. Total solids - mass flow vs. cumulative lenchate volume.
-------
TOTAL ORGANIC CARBON
CELL 4
20 40 60 80 100 120 140
CUMULATIVE LEACHATE VOLUME (Liters) xlO2
Figure 3. Totn.1 organic carbon - macs flow vs. cumulative .le--i.chatc vnl.ume.
160
-------
100
o
x
3
*
80
60
to
< 40
o:
o
20-
TOTAL KJELDAHL NITROGEN
CELL 4
CELL 5
CELL 3
20 40 60 80 100 120
CUMULATIVE LEACHATE VOLUME (Liters) x I02
140
Figure h. Total kjeldahl nitrogen - mass flow vs. cumulative leachate volume.
-------
400-
o
x 320
35
re
o
5 240-
CO
1
-, 160
80 H
CHEMICAL OXYGEN DEMAND
CELL 4
CELL 5
CELL 3
CELL 2
CELL /
20 40 60 80 100 120 140
CUMULATIVE LEACHATE VOLUME (Liters) xlO2
FIf.ure 5- Chemix-al oxygen demand - mass flow vs. cumulative leachute volume.
160
-------
bales of nonshredded waste.
Reducing the size of refuse
particles allows them to "fit
together" more compactly in bales,
reducing air space between particles
more than simple baling. This
process would further retard
movement of water in and solids out,
resulting in a lower rate of
leaching for baled shredded waste
than for baled waste. The increased
rate of leaching of Cell 2 over
Cell 1 is also readily evident in
the plots of cumulative masses
leached versus cumulative leachate
volume.
One of the metal parameters
deserves special mention in our
discussion because of its high rate
of leaching from Cell 4 as compared
to the other lysimeters. The
cumulative mass of zinc leached from
Cell 4 is about one order of
magnitude higher than that of the
other cells. The presence of zinc
alkaline batteries in the waste may
be responsible for this phenomenon,
according to Hentrich.(S) The
batteries contain zinc salts which
would normally be sealed within,
unless the battery were shredded.
The zinc salts would then be
released and available for leaching.
Such batteries would also be
expected to be shredded in the waste
of Cell 1, but the baling of the
waste would hinder leaching of the
salts by making the waste more
impermeable to water. Rapid
leaching of zinc in Cell 1 has not
been observed. The dispropor-
tionately large significance of this
one component of the waste stream
points out how minor variations in
the compositions of wastes can have
great impacts on the character of
the leachates emanating from them.
However, these variations only
assume great importance when the
parameter being considered is
present in a very small quantity in
the refuse.
Gas Composition Data
The biological decomposition
of municipal solid waste occurs in
the lysimeters in the same manner
as in conventional landfills. This
transformation of organic materials
takes place in three stages,
according to Ham.(4) The first
stage is aerobic decomposition in
which organic matter is degraded to
CC>2f water, and partially degraded
residual organics. This process is
rapid and produces heat. The C02
produced dissolves into any water
present, decreasing the pH.
The second stage is decomposi-
tion by facultative acid forming
bacteria. Carbon dioxide is
produced rapidly with little heat
production. Partially degraded
organics, especially organic acids,
are formed. The pH is lower than
in the aerobic stage, which helps
dissolve inorganics.
The third stage is decomposi-
tion by anaerobic methane forming
bacteria. This process occurs
slowly and efficiently, producing
C02» water, and methane with
little heat. Because the decomposi-
tion rate in this stage is slow and
the waste is becoming more stable in
nature, the CO2 content of the gas
decreases and the moisture pH
increases, reducing the amount of
dissolved inorganic material. This
stage is called the methanogenic
stage because methane is produced.
These three stages of biolog-
ical decomposition in landfills are
not exclusive of each other.
Normally, two or three of them occur
simultaneously in different
locations within the landfill. The
methane formers are symbiotic with
the acid formers, practically
assuring that some of the organisms
present in the facultative stage
are also present in the methanogenic
stage.
Figures 6 through 10 show time
histories of the compositions of
gases collected from the test cells.
The graphs indicate inconsistencies
both in time and from cell to cell.
An anaerobic, methane producing
situation is indicated for Cell 1,
while Cell 5 is shown to have
produced little methane. Cells 2
through 4 have produced methane in
varying amounts.
29
-------
CELL 1
m HETHflNE
A CRRBON oiexior
+ NITflQCEH
X OXTCEN
1975
8C7
JUM
1976
FEB
1977
OCT
JUN
1978
FE8
ect
1979
JUM
1980.
Figurs 6. Cell 1 - Gas composition vs. time.
-------
CELL 2
o_
A NETHANE
2 CARBON 01flXIDE
+ NITROGEN
X OKTGEN
OC1 JUN FEB OCT JUN
1975 1976 1977 1978
FEB
OCT
1979
JUII
1980.
Figure 7• Cell 2 - Gas composition vs. time.
-------
CELL 3
m WETHflNE
A cnneoN DIOXIDE
4. NITflOCfN
X OXtCEN
1975
JUN
1976
1977
JUN
1978
CCT
1979
Figure 8. Cell 3 - Gas composition vs. time.
-------
o
o
CELL 4
0 ME7MRNE
A CflflBON 01OK IDE
j. NI7ROCEH
X OXTCEM
1975
8C1
JUM
1976
FfB
1977
oct
JUN
1978
FEB
8CT
1979
JUN
1980.
Figure 9- Cell li - Gas composition vs. time.
-------
o
o
CELL 5
m HETHBNE
A CRRiON DIOXIDE
4. NITROGEN
X OXTGEN
1975
JUN
1976
FEB
1977
OC7
JUN
1978
FEB
0CT
1979
JUM
1980.
Figure 10. Cell 5 - Gas compostion vs. time.
-------
It seemed unbelievable that no
portion of the large mass of waste
in Cell 5 had become anaerobic and
produced methane. The possibili-
ties of blockages or leaks in the
gas collection system of Cell 5
were investigated, and the gas col-
lection probes within the cell were
found to have been broken off by
settlement of the refuse. Where the
gas collection lines went through
the walls of the cell, they now met
settled clay and gravel which was
placed originally over the refuse.
New gas collection probes were in-
stalled further below the old ones,
into the refuse mass. Samples taken
after this were as much anaerobic as
those from the other cells
("5 percent 02), but still only
traces of methane were detected.
The pH of the leachate from Cell 5
has remained below 6, while the
leachates from the other cells are
all above 6. This leads us to
believe that the refuse in Cell 5 is
being maintained in a state of acid
fermentation, with only few methane
forming bacteria present. This
would explain the lack of methane
production, but just why Cell 5
alone is acting this way is unknown.
In general, the gas composition
data do not indicate any obvious
effects of the different refuse
processing procedures. It is still
unknown what effects baling and
shredding have on landfill gas
production.
Discussion of Results
The analysis of the data
collected in the course of the
project from the beginning until now
gives rise to the following
conclusions.
The processing step of baling
has been shown to enhance leachate
production both by decreasing the
time period before continuous
production of leachate and by
increasing the overall volume of
leachate produced. The cells
containing baled waste produced
leachate about 2 months before the
cells containing unbaled waste and
also produced more leachate over the
entire period of the study. Baling
reduces the moisture-retention
ability of the waste, as does
shredding; but shredding enhances
moisture infiltration at the same
time. Baling has a much greater
positive effect than shredding on
the volume of leachate produced.
Baling was seen to result in
the formation of leachate that
contains relatively low
concentrations of pollutants, while
shredding of the waste causes
leachate that has high concentra-
tions of pollutants. A comparison
of the mass flows of pollutants fror
the cells on a basis of equal
leachate volume shows the decreasing
effect of baling and increasing
effect of shredding on pollutant
concentrations in the leachate.
Taking together the leachate
volumes and pollutant concentrations
resulting from baled vs. shredded
waste, it is evident that the two
processes produce opposite results.
Baling of waste results in large
volumes of dilute leachate
emanating from a landfill with a
long period of time required for
stabilization of the landfill.
Conversely, shredding of waste
results in the formation of smaller
volumes of highly concentrated
leachate, and a shorter period of
time passes before landfill
stabilization.
The leachate data from the
baled waste in a saturated '
environment is coincidentally very
similar to the data from the waste
in the control lysimeter. It may be
that these two factors, baled waste
and saturated environment, have
mutually canceling effects when
present in combination. This
finding might be of interest in
designing landfills in areas where
the water table is high.
The data also show that
shredding of waste before baling
inhibits leaching even more than
baling alone. If the slowest
possible rate of leaching and
longest time period before landfill
stabilization were desired,
shredding and baling both could be
employed.
35
-------
The only conclusion that can
be drawn from the gas composition
data is that there is a great deal
of variability both among the cells
and over time in each cell.
References
1. Edens, M. H. Hydrogeologic Study
of a Landfill Site Consisting
of Highly Compressed Solid
Waste. Master's Thesis,
University of Minnesota, 1978.
2. Fenn, D. C., K. J. Hanley, and
T. V, DeGeare. Use of the
Water Balance Method for
Predicting Leachate Generation
from Solid Waste Disposal Sites.
EPA/530/SW-168, U.S.
Environmental Protection Agency,
Cincinnati, Ohio, 1975.
3. Fungaroli, A. A., and R. L.
Steiner. Investigation of
Sanitary Landfill Behavior.
EPA-6001 2-79-0533, U.S.
Environmental Protection Agency,
Cincinnati, Ohio 1979.
Ham, R. K., et al. Recovery,
Processing, and Utilization
of Gas From Sanitary Landfills.
EPA/600/2-79/001, U.S.
Environmental Protection Agency,
Cincinnati, Ohio, 1979.
Hentrich, Jr., R. L., J. T.
Swartzbaugh, and J. A. Thomas.
Evaluation of Landfilled
Processed Municipal Wastes.
Annual Report, EPA Contract
No. 68-03-2598.
Wigh, R. J. Boone County Field
Site Interim Report: Test
Cells 2A, 2B, 2C, and 2D.
EPA-600/2-79-058, U.S.
Environmental Protection Agency,
Cincinnati, Ohio, 1979.
36
-------
Recovery of Fecal-Indicator and Pathogenic Microbes
from Landfill Leachate
J.A. Donnelly and P.V. Scarplno
University of Cincinnati
Cincinnati, Ohio 45221
and
D. Brunner
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
ABSTRACT
With ever-increasing amounts of municipal and commercial solid waste (garbage,
sewage sludge, hospital waste, etc.) entering sanitary landfills and open dumps, the
potential exists for human and animal pathogens in these wastes being leached out by
rainwater to contaminate ground and surface waters. To determine if this leaching water
(i.e. leachate) could contain human pathogens, leachates from commercial and large scale
experimental landfills and laboratory lysimeters containing various solid wastes were
studied.
Microblal assays on active landfills demonstrated that the leachate contained high
concentrations of Gram-negative rods, some of which were Identified as Enterobacter
aerogenes, Citrobacter sp., Salmonella sp., Serratia marcescens. Proteus sp. and Aeromo-
nas sp. The Gram-positive bacteria present were Clostridium perfrinqes, Streptococcus
faecal is. S_. fa eel urn, £. yjridans, and Mycobacterium sp. ATI of these bacteria may be
present in human feces, and several are pathogenic for humans. Landfills Inactive for 2
to 6 years had very low concentrations of Gram-negative rods. The most abundant microor-
ganism present were Streptococcus faecal is. Corynebacterium sp., Bacillus sp., yeasts
and molds. Pathogens found in these old landfills were ListeHa monocytogenes, Acineto-
bacter sp., Horaxella sp., and AHescherla boydii. Newly constructed laboratory lysi-
meters containing a variety of municipal wastes, demonstrated very low col 1 form levels
in the leachate after 4 months. The streptococcal levels remained higher, but eventu-
ally dropped below the detection level after lh years. However, the laboratory lysime-
ters containing sewage sludge continued to discharge fecal microorganisms, both coliform
and streptococci, even after two years. Fecal-Indicator bacteria were isolated in the
solid waste from lysimeters 2 years old, and from a 9 year old fullscale landfill .
These results demonstrate that fecal-Indicator bacteria and some pathogenic forms do not
die as rapidly in landfills as previously reported but survive for extended periods of
time with the potential to contaminate ground and surface waters when released 1n leach-
ates from the landfills.
INTRODUCTION leachates from sanitary landfills. The
ever increasing use in the United States
This research is concerned with the of landfills as disposal sites for gar-
potential contamination of water supplies bage; other municipal waste as pet feces,
by pathogenic microorganisms contained in animal remains, and disposable baby diapers
37
-------
containing human infant feces; respir-
atory waste In paper tissues; hospital
wastes; and raw or partially treated
sewage sludges poses a potentially signif-
icant public health hazard. This is
especially true when consideration is
given to the possible contamination of
adjacent ground and surface waters by
microbial migration through the landfill
site and the adjacent soil.
Specific information on the survival
of infectious agents in sanitary land-
fills is essentially nonexistent. Also,
few studies exist which characterize
(i.e. identify) the specific microbial
populations occuring in the leachates
from landfills. Several investigators,
however, have analyzed landfill leachates
for total coliforms, fecal coliforms, and
fecal streptococci. The survival of such
fecal Indicator-bacteria in leachate
suggests the presence of microbial patho-
gens in the leachate [See Scarpino
(22,23) for a discussion of the microbial
indicator concept]. One of the first
such studies on landfills was reported by
Carpenter and Setter (5) who found a mean
of 10/100 ml levels of Escherichia
col i-type coliforms in New York City
landfills that were less than two years
old. Sobsey (25) found both total coil-
forms, and fecal coliforms in several
inactive landfills, indicating that
coliforms remain alive for an indefinite
length of time after landfills have been
closed.
Specific microorganisms were identi-
fied from old landfills by Cook et al.
(6). They found Escherichia coli,
Serratia marcescens, Streptococcus fae-
cal is. Bac i11 us. Pseudomanas, Achromo-
bacter, F1 a v o ba ct.eri u m, Lactobacillus,
Sarcina, Prop ionibacterium, and Clostrf-
dium. They, also, noted the presence of
many sapropnytic molds such as Asperg.il-
JUjsj, Penicillium, Fusarium and others.
Other investigators have monitored leach-
ates from a variety of commerical and
experimental landfills to determine
densities of total coliforms, fecal
conforms and fecal streptococci. For
example,Blannon and Peterson (3) looked
for fecal coliform and fecal strepto-
coccus levels in leachates from an
experimental landfill containing 435 tons
(wet weight) of residential solid waste.
Biochemical characterization of 60 strep-
tococcal isolates showed that 33.3" were
indicative of fecal contamination from
warm-blooded animals, and included S.
faecal is biotypes II and III (12 strains!",
£. durans (3 strains), and S_. equinus (5
strains).
Thus, the objective of this investiga-
tion was to evaluate the pathogenic micro-
bial presence in leachates, considering
also the fecal-indicator microbes, especi-
ally the fecal streptococci. In this
manner we hoped to obtain information as
to the health hazard associated with
landfill leachates. Specifically this
research included:
1. The determination of bacterial and
fungal pathogens in leachate from
several existing commercial and
experimental landfills.
2. The characterization of the species
of fecal streptococci persisting in
landfill leachates to ascertain their
significance relative to the patho-
genic presence in the same leachates.
3. The construction of small-scale
laboratory experimental lysimeters to
study microbial population changes in
leachates.
The experimental work described in
this report is in two phases, each varying
with the landfill sites used. The first
phase involved analyses on the active and
inactive landfill sites built before 1977.
The second involved a study of 6 newly
constructed lysimeters, each containing
different solid wastes.
MATERIALS AND METHODS
Sampling Sites
A list of the landfill sampling sites
and their characterization are given in
Table 1. The last site mentioned in this
table is further described in Table 2. A
full description and construction details
of these sites were reported by Scarpino
et al. (24). Of all of the sites listed in
Table 1, only the commercial landfill
received waste daily and was thus active,
while the rest were inactive.
Sampling Procedure
Leachate obtained from the commercial
landfill was collected at the point where
it flowed from the surface of the landfill
slope. Occasionally, leachate was ob-
tained from horizontally placed steel
-------
TABLE 1. LANDFILL SAMPLING SITES
Landfill
Commercial ,
Full-Scale.
Operational
Boone
County,
Kentucky
Boone
County,
Kentucky
Center Hill,
Cincinnati,
Ohio
Univ. of
Cincinnati ,
Cincinnati,
Ohio
Leachate
or
Sol id Waste
Abbreviation
CL
BC 1
BC 2
through 5
CH 1
through 19
A
through F
TABLE
Dimensions
and
Placement
700 Acres
in Soil
40' x 140'
x 8.5'
in soil
12' x 6'
Steel
Cylinders
1n Soil
12' x 6'
Steel
Cylinders
2' x 2'
in Steel
55 gallon
Drums
Solid Waste Added Date
Prepared
Type Amount
Municipal, Millions 1954
Hospital, of
Commercial Tons
Municipal 435 Tons 1971
Municipal 2 Tons 1972
Municipal , 2 Tons 1974,
Sewage Sludge, 1975
Industrial
Municipal. 150 Ibs 1978
Sewage Sludge,
Hospital
Date of
Last Waste
Deposit
Dally
(Active)
June 1971
(Inactive)
August, 1972
(Inactive)
Nov. 1974
S April, 1975
(Inactive)
August. 1978
(Inactive)
2. DESCRIPTION AND PURPOSE OF LABORATORY LYSIMETERS
Number
A
Contents**
Sewage
Sludge
Purpose
Microbial control for the sewage
sludge
Municipal Solid
Waste Plus
Sewage Sludge
Municipal Solid
Waste
Municipal Solid
Waste
Municipal Solid
Waste Plus
Hospital Waste
Hospital Waste
used 1n Lysimeter B.
To determine the Impact of municipal
sewage sludge additions on the rates
of decomposition and formation of gas
and leachate. Also, to elucidate
mlcroblal changes in the leachate.
Control on mlcroblal numbers and
species.
Same as Lysimeter C.
To determine the Impact of pathogens
on municipal solid waste and pathogen
ability to survive landfill conditions.
Mlcroblal control on the hospital waste
used 1n Lysimeter E.
0208.2 liters 1n each 55 gallon drum.
Lyslmeters maintained at 20 C in a constant temperature room.
39
-------
pipes extending deep within the landfill.
In all the remaining sites listed in
Table 1, the leachate was obtained from
sampling ports. All leachate samples were
allowed to drain into glass flasks,
previously gassed with nitrogen. These
flasks were iced immediately, taken to
the laboratory, stored at 4 C, and pro-
cessed in 6 to 24 hours.
Solid waste samples taken from
landfills were immediately placed in
sterile plastic bags, and these in turn
were placed in nitrogen-gassed jars to
minimize oxidation of the sample. Those
samples to be used for the recovery of
anaerobes were placed immediately under
anaerobic conditions at the sampling
sites using the GasPak (BBL) system.
All jars were kept in an ice chest during
their transportation to the laboratory.
There the samples were kept at 4 C and
processed within 24 hours.
Concentration and Digestion of Samples
for Mycobacterlal Isolation
To isolate mycobacteria from lea-
chate on a consistent basis, it was
necessary to both concentrate and digest
the samples. The concentration process
involved filtering a liter of sample
leachate through diatomaceous earth
filters. After filtration, the diato-
maceous earth was removed, mixed with
phosphate buffer, and allowed to settle
out. The supernatant liquid was then
digested with n-acetyl-L-cysteine and
NaOH, followed by centrifugation. This
sediment was inoculated onto Lowenstein-
Jensen slants and Middlebrook 7H10 Agar,
and incubated at 37 C for 2-4 weeks. The
digestion procedure destroyed the sapro-
phytes, and allowed the slower growing
and resistant mycobacteria to grow.
Sample Homogenization
The solid waste was homogenized in
preparation for the enumeration of its
bacterial and fungal contents. The solid
waste was weighed, diluted 1:10 with
phosphate buffer, and minced in a Waring
blender for 15 seconds at high speed.
Microbial Enumeration Procedures
The leachate samples were diluted
with phosphate buffer and then plated
using the spread plate technique, as
follows: 0.1 ml of the leachate dilutions
were each placed on separate agar plates,
and the inoculum was spread over the
surface of the agar with a sterile glass
rod. Replicate bacterial plates were
made, and all were incubated at 37 C for
48 hours and then counted. Fungal agar
plates were incubated at ambient tempera-
ture for one week and counted.
Blood agar was used to obtain both
the aerobic and the anaerobic bacterial
counts. The GasPak(BBL) system was used to
produce the anaerobic conditions for the
anaerobic plates. The total bacterial
count was made with brain heart infusion
or Standard Methods agar. The differen-
tial agars used for the microbial plate
counts were: eosin methylene blue (EHB)
for the Gram-negative rods; KF-streptococ-
cal for the streptococci; tellurite for
nonsporing Gram-positive bacteria; mycosel
for dermatophytes; inhibitory mold for
systemic fungi; and Sabouraud for yeasts
and molds. These procedures are consis-
tent with those found in Microbiological
Methods for Mom'torina the. Envi ronment (4)
and Manual of Clinical Microbiology Q5).
The most probable number (MPN) tests
(1, 11) were used to determine the total
coliforms, fecal coliforms, and the fecal
streptococcal levels instead of the mem-
brane filter method, as the former method
gave higher counts (11). The coliform MPN
test procedures were changed during the
Phase II experiments from that of Standard
Methods (1) to that of Allen et al. (2).
Allenet al. had noticed that large
numbers of bacteria in environmental
waters often interferred with lauryl
sulfate broth gas production. The subse-
quent inoculation of all of the cloudy
tubes from the lauryl sulfate broth,
without regard to gas production, i.nto
brilliant green bile broth and EC medium
resulted in significant increases in the
numbers of positive tubes, and a resultant
increase in the MPN counts.
Enrichment Procedures
Enrichment procedures for the isola-
tion of Salmonella and Shigella were a
part of the leachate analysis. The media
used for these isolations included tetra-
thionate, selenite F and GN broths. Five
milliliters to 50 ml of leachate were
added to 50 ml of the broths. After 24
and 48 hours incubation at 37 C, these
40
-------
media were streaked on XLD or Hoekton
enteric agars. The above procedures also
conform to those found in Microbiological
Methods for Monitoring the Environment
(4) and Manual of Clinical Microbiology
(15).
Identification of Microorganisms
The Identification procedures used
for the various microorganisms isolated
from landfills were: Gram-negative rods,
Wolf et al. (27), Lennette et al. (15),
Koneman et al. (12) and MacFaddin (16);
for Gram-positive rods, Cowan (7) and
Lennette et al. (14); and for fungi
Koneman et al. (13) and Lennette et al.
(14). On occassions, the API system was
used to extend the identification beyond
the procedures given above. Also, sero-
logical slide tests were used to confirm
for suspected Salmonella, Shigella. and
Klebsiella.
In addition to the above procedures,
extensive identification (I.e. specia-
tion) of the streptococci isolated from
solid wastes and leachates were made using
two procedures. One (9) was used to
place the streptococci tentatively Into
Lancefield groups A.B.D, etc. These
tests included hemolysis, bacitracin
sensitivity, hippurate hydrolysis, 'bile
-esculin hydrolysis, and growth 1n 6.5%
NaCl broth. Representative cultures were
confirmed with Lancefield grouping using
the serological antisera in Streptex
tests (Burroughs-Welcome Co., Research
Triangle Park, North Carolina.) The
second method identified the Group D
streptococci as species by the procedure
of Pavlova, et al. (18). This latter
procedure involved some 15 biochemical
characteristics.
Leachate Inhibition Procedure
Leachate Inhibition assays were made
to determine the degree to which old
lyslmeter leachate Interferred with
microbial growth, particularly with
respect to Gram-negative rods. Definite
concentrations of the test bacteria were
added to leachate obtained from the
number 19 Center Hill lyslmeter. These
test solutions were mixed, diluted, and
plated on differential agars. Such agars
were useful in that they allowed the test
bacterial growth while preventing the
leachate bacterial growth. The Gram
-negative rods were counted after 24 hours
at 37 C. and the streptococci were counted
after 48 hours. A growth response less
than that appearing in the control indica-
ted leachate inhibitory action.
Mycobacterium Antibiotic Sensitivity Tests
Antibiotic sensitivity tests were
used to separate suspected isolates into
groups for identification purposes. These
bacteria were maintained on Middlebrook
7H9 agar. They were then streaked for
isolation, transferred to 7H9 broth, and
plated on 7H10 agar containing antibiotic
discs. Then they were packed into plastic
bags and incubated at 37 C for 3 weeks.
The resistant growth was recorded.
RESULTS OF PHASE I EXPERIMENTS
Experimental Approach
The initial work on this project was
designed to characterize the bacteria and
fungi present in commercial and experimen-
tal large-scale landfills. As many
species of bacteria and fungi were present
in the leachate, 1t was necessary to use a
variety of isolation and identification
methods. Most bacterial species grew on
enriched media such as blood agar and
Standards Methods agar. However, differen-
tial or selective media allowed only one
kind of bacteria to grow and inhibited the
rest. Bacteria and fungi measured by
these two types of procedures were aerobic
and anaerobic bacteria, Enterobacteria-
ceae. pseudomonads, streptococci, nonspore-
forming Gram-positive rods, molds, yeast,
total col iforms, fecal conforms, and
fecal streptococci. Isolates were picked
from these media and identified.
Microscope Examination of Streptococcus
MPN Tubes
Fecal Streptococci. MPN counts were
first determined on Center H111 landfill
leachates, and are reported 1n Figure 1.
To confirm that these results were caused
by streptococci, a Gram stain was made on
those ethyl violet azide tubes giving a
positive reaction, i.e. the appearance of
the purple button in the bottom of the
tube. The microscopic examination of the
stained purple growth showed that these
positive results were not always caused by
cocci. Indeed, many positive results were
caused by Gram-positive rods. Hereafter,
all MPN tubes were examined mlcroscopi-
41
-------
107
1 106-
y
0 5
O f*
* fc
* 0 .4
*• o 10-
(0 *-
- x 3
« 2 tO-
rn CL
22 2
1O.
1
10-
S -Gram-positive rods
observed
Q -Gram-posit I ve cocci
- observed
S r Sewage Sludge
C= CaCO,
I
I = Industrial Sludge
i
X
X
,
N
P
» *
V i
•.
..
••
^
••
••
I
-i
x
X
X
X
1
-.
• •
"
**
r~i r~1
•
•
*"
••
••
""
1 234 5678 9 1O 11 12 13 14 15 1817 18 19
Center Hill Test Cells
Figurel. Fecal St reptococcal Presence In Cent er
H III Experimental Field Lysimeters
cally for Gram-positive cocci, and strep-
tococcal levels were ultimately determ-
ined based on coccal presence in positive
ethyl violet azide tubes. The signifi-
cance of these false-positive results is
that high counts of streptococci have
been reported from these landfills, when
this may not always be in fact the case.
In additon, the ethyl violet azide
tubes to which 10 ml and 1 ml aliquots of
Teachate had been added contained fewer
or no Gram-positive cocci than those
found at higher dilutions. We interpre-
ted this to mean that leachate toxicity
was present. It was apparent that as the
leachate was diluted out, the toxic agent
was also diluted out, because no inhibi-
tion qf Gram-positive cocci growth was
noticed at the higher dilutions. Thus,
the toxicity present in the leachate gave
false-negative test results and indicated
lower concentrations of streptococci than
were actually present.
The Enumeration ofleachate Microorgan-
isms
The results of the most probable
number (MPN) tests, and agar plate counts
of leachate microorganisms from several
landfills are shown in Table 3. The
older landfills had lower counts and very
low fecal-indicator bacterial levels. The
counts of the Boone County Landfill, Cell
l(BC-l), were much lower than those found
by Blannon and Peterson in 1974. The high
eosin methylene blue levels in Table 3
were not always due to Gram-negative
rods, but were due to Bacillus sp. or to
fungal colonies. Moreover, the microbial
levels indicated that the leachate (col-
lected at the landfill surface) obtained
from the fresh solid waste in the commer-
cial landfill had a higher concentraton
of Gram-negative rods and fecal strepto-
cocci in it than the older landfill
leachate obtained from the pipes. These
higher fecal-indicator bacterial concen-
42
-------
TABLE 3. MICROBIAL COUNTS FROM LANDFILL LEACHATE
Quantitative
Tests
MPN/lOOml
Total Conforms
Fecal Coliforms
Fecal Streptococci
Agarjlate counts, ^FU/100ml
Blood-aerobic 7.
Blood-anaerobic 8.
Brain heart 4.
Eosin Methylene Blue 3.
Inhibitory Mold** 1.
KF Streptococcal
Hycosel*' 2.
Si
Sabouraud 4.
Tellurlte 5.
Landfills (Lysimeters)
Boone
BC-1
u
(6 years)9
<30
<30 ,
1 x 10^
3 x 105
0 x 105
0 x 105
0 x 105
7 x 104
< 103
0 x 103
0 x 105
0 x 104
County
BC-2
(5 Years)*
OO
<30
*20
5.0 x 104
< 103
1.0 x 104
1.0 x 104
7.0 x 103
-ao3
103
2.0 x 106
6.0 x 103
Center HJ11
CH-19
(2 Years)*
"30
-------
trations, indicated that fecal pollution
was present in the landfill, Leachate
from the commercial landfill pipes which
extended deep within the old part of the
landfill demonstrated low bacterial
levels, whereas leachate arising out of
the ground from recently deposited wastes
contained microbial levels a thousand-
fold higher.
The Inhibition of Bacteria by leachate
The inhibition of the streptococci by
the leachate in the lower dilutions of the
fecal streptococcal MPN tests, described
earlier, indicated that some toxicity was
present. It seemed probable that the
leachate toxicity could be responsible for
decreasing plate counts and inhibiting
some pathogen growth. To determine the
presence and the extent of leachate toxic-
ity, known bacterial species were exposed
to leachate, and immediately plated out to
determine the number of survivors. Selec-
tive agars were used to differentiate the
test organisms from the leachate bacteria.
The test bacteria used in this study were
pathogens or pathogen indicator bacteria.
They were K1ebsiell a pneumoniae. Salmonel-
la typhimuriuni. Pseudomonas aeruginosa.
and Streptococcus faecalis. These bacter-
ia after exposure to the leachate had
concentrations of 10/100 ml, while the
controls, which were not exaosed to the
leachate, had levels of 10/100 ml or
higher. These results demonstrated that
the leachate inhibited the microbial
growth of all four test bacteria. Further-
more, there is evidence in the literature
to indicate that surpressed or inhibitory
growth may also be a result of secondary
stress caused by the selective media, in
addition to the primary stress of leachate
exposure [Postgate (20)].
Microbial Identification
The examination in the fall and
winter months of leachates from the com-
mercial and experimental field landfills
revealed that the most frequently isolated
microbes were Bacillus sp., Streptococcus
sp., Corynebacterium sp., and yeast.Tn
the early summer following the spring
rains, numerous bacteria were found in all
of the field landfill leachates, and these
results are presented in Table 4. Some of
the microorganisms in this table are known
pathogens and can be found in fecal
material. Many are saprophytes. All of
these bacteria and fungi were compared to
the U.S. Public Health Service's Classifi-
cation of Etiolojnc.Agents on the Basis of
Hazard (21). whichincludes alldisease-
producing microorganisms. This classifica-
tion is based on the pathology and infec-
tivity of the microorganisms. Persons
working with class one agents need take no
special precautions in handling these
microbes, whereas those working with class
two agents need to take precautions to
prevent from being inoculated by contam-
inated material containing these patho-
gens. Class three agents require that the
investigator use special precautions to
avoid direct contact with these patho-
gens, as contact will usually cause
disease. Of the microrganisms found in
Table 4, Listen a monocytogens, Salmonel
JA sp,, Clostridium perfringens. Mycobac-
terium sp., Moraxella sp., and Acineto-
bacter sp. are"class two bacteria. No
class three microorganisms were found in
these studies. Al1esche ri a boyd i1, while
not on the etiologic list, is a pathogen
which causes madura foot abscesses.
The Mycobacterium listed above could
not always be found in leachate. so a
concentration process was developed to
improve the infrequent isolations. For
this process one-half to one liter of
leachate was filtered through diatoma-
ceous earth. The diatomaceous earth was
extracted with phosphate buffer, diges-
ted, centrifuged and inoculated onto
Lowensteln-Jensen medium and spread on
Middlebrook 7H10 agar plates. The Myco-
bacterium grew out on this medium within
2 to 3 weeks at 37 C. All eleven lea-
chate samples produced Mycobacterium
colonies. The levels isolated varied from
2 to 33/100 ml of leachate with an aver-
age of 10/100 ml. To separate these
bacterial strains into groups for identi-
fication purposes, they were inoculated-
onto antibiotic media containing refam-
pin, isoniazid, p-aminosalicylic acid,
streptomycin and ethambutol, From the
nineteen strains tested in this fashion
seven sensitivity patterns emerged
indicating possibly that seven different
strains of mycobacteria had been isola-
ted. Preliminary work has been started
on their identification. In comparison
with a known ATCC strain of Mycobacterium
fortuitum, which grew out on Middlebrook
7H10agar in 2 days, these isolated
strains grew slower, taking just under a
week to grow out. In addition, a yellow
44
-------
TABLE 4. PARTIAL IDENTIFICATION OF MICROORGANISMS ISOLATED
FROM SPECIFIC LEACHATE SAMPLES.
Leachate Source
(Sampling dates)
Commercial
Landfill
(6/12/79)
Commercial
Landfill
(6/9/80)
Microorganisms Identified
Bacteria
Hycobacterlum sp.
Proteus rettgeri
Mycobacterlum sp.*
Salmonella sp.*
Clostrldium perfrlngens
dtrobacter sp.
Enterobacter aerogenes.
Fungi
Boone County
Cell 1.
(5/3/78)
Center H111
Cell 19.
(5/31/78)
Bacillus sp. *
Listeria monocytoqens
Aeromonas sp.
Streptococcus sp.
Moraxella sp.*
Nelsseria sp.
Corynebacterium sp.
Pseudomonas sp.
Streptococcus sp.
Acinetobacter sp.
Bacillus sp.
Yeast
Allescheria
Yeast
Fusarium
PenicilHum
boydii
*=Listed as class 2 1n the CDC Classification of Etlologic Agents on the
Basis of Hazard (21)
strain was isolated a year ago, but the
recent strains were cream colored.
Mycobacteria are considered class two
agents on the basis of hazard as de-
scribed by the CDC classification, while
fl. tuberculosis, related to those isola-
ted from the leachate, is considered a
class three agent. This particular spe-
cies has not been Isolated from leachate,
but there 1s every Hkehood that 1t ts
present because of the presence of other-
mycobacterla and high fecal-Indicator
levels 1n fresh landfill leachate samples.
RESULTS OF PHASE II EXPERIMENTS
Experimental Approach
This phase of the research investiga-
ted the survival of pathogenic micro-
organisms 1n leachates from experimental
landfills (lyslmeters) containing munici-
pal solid waste, sewage sludge, hospital
waste, and combinations of municipal solid
waste and sewage sludge or hospital waste.
Sewage sludge and hospital waste are
produced in great quantities in our commu-
nities, are potentially hazardous, and are
frequently placed Into landfills throughout
the United States. The Impact of all of
these wastes on the disease potential of
landfill leachates 1s not known. Land-
fills containing such wastes could produce
leachates carrying pathogens Into the
environment via the groundwater or surface
waters. It ,1s, therefore, critical to
know the duration of pathogenic mlcroblal
survival 1n landfills and their leachates.
Accordingly, the lyslmeter studies of
Phase II were designed to evaluate the
survival characteristics of pathogenic
microorganisms and their Indicators. Of
special Interest was the spedation of
streptococcal isolates, because they
represented a different type of pathogen-
Indicator 1n that the source of the waste,
i.e. from animal, human, avian, plant, or
Insect sources, could also be ascertained.
45
-------
Finally, the examination of buried solid
waste from old landfills or lysimeters for
the presence of pathogens and pathogen-
(fecal)-indicators would show the extent
of their survival and their potential for
groundwater and surface water pollution.
The Enumeration and Identification of
Microbes Found in Solid Wastes
Samples of waste placed in the six
lysimeters were first analyzed for their
microbial numbers. Identification of the
microbes isolated included streptococcal
speciation. Densities were determined by
agar plate counts and the most probable
number (MPN) procedures. The results,
using several agar plate counts including
aerobic and anaerobic blood, EMB, and
Sabouraud were similar, i.e. about 10
microorganisms per gram of waste. Sewage
sludge samples were characterized using
KF-Streptococcal and tellurite igars.
Plate counts on both media were 10 /gram
of sludge. The hospital waste microbial
levels using these last two media were 1
to 2 logs higher than those found with
sewage sludge, whereas the municipal
waste was 3 logs higher. All three of
these wastes contained total col 1 forms,,
fecal coliforms, and fecal streptococci
at levels 10 to 103MPN/g of waste.
The pathogens present in these wastes
were identified as Klebsiella sp., Salmo-
nella sp., Mina sp., Here!lea sp., Morax-
ella sp.. and Pasteurella hemolyticum. Of
the 25 pathogens isolated from these
wastes, 6 came from sewage sludge, 9 from
the hospital waste, and 10 from the
municipal waste.
Microbial Densities in Leachates
The determination of plate counts
and MPN levels of microorganisms in the
six lysimeter leachates were made weekly,
biweekly, monthly and then four times a
year. Although the data extends from the
first leachate sample, which appeared in
some cases at the 4th week after lysi-
meter construction, to the 104th week
(two years), only the concentration
levels for the 13th and the 104th weeks,
the latter placed in parenthesis, are
shown in Table 5. The Gram-negative rod
concentrations, which had been high in
the solid waste, had dropped to a low
level by the 13th week. Furthermore the
results taken at the 104th week show even
lower microbial levels. The fecal-
indicators were gone (below the detection
level); most or all of the aerobic and
anaerobic blood agar, Sabouraud and
tellurite plate counts were lower; while
the Standard Methods agar levels were
unchanged. Thus, the Gram-negative rods
(e.g. Salmonella, Klebsiella. other
coliformTi etc.) died within the first
three months, while the streptococci
were <10/100 ml at the end of the first
year. In addition, the nonsporing Gram-
positive rods and fungi had decreased in
concentration.
Other assays not shown here demon-
strate that Gram-positive sporeforming
rods such as Clostridium perfrinqens and
Bacillus sp. survived. This population
change over the 2 years indicates that
many pathogens and pathogen-indicators
present in the solid waste had died in
the lysimeters. However, not depicted by
these results was the occasional release
of leachate from the sewage sludge lysi-
meter containing relatively high levels
of fecal-indicator bacteria, such as 2.8
x 10J, 4.9 x 101, and 5.4 x 10 MPN/100
ml of total coliforms, fecal coliforms,
and fecal streptococci, respectively,
after 2 % years.
The low concentration of fecal-
indicator bacteria in leachates from most
lysimeters after 3 months and certainly
after 2 years suggest that these bacteria
die quickly in landfills. However, the
survival and continued discharge of high
concentrations of fecal-indicator bacteria
in leachates from the sewage sludge lysi-
meter 2 years after its construction
demonstrates the potential for intermittant
contamination by this waste for an extens-
ive time period.
Total Coliform & Fecal Coliform Densities
The total coliform and fecal coliform
MPN levels in the leachate from all lysi-
meters had disappeared by the 9th week
after lysimeter construction, indicating
that all the coliforms had died. However,
the substitution at the 13th week of the
Standard Methods (1) MPN test to the Allen
et al. (2) modification indicated that
fecal and/or total coliform MPN levels
were indeed present in leachates from both
lysimeters A and B. These lysimeters
contained sewage sludge, and sewage sludge
plus municipal waste, respectively. The
total coliform MPN levels for lysimeter A
reached a peak of 2.2 x 10/100 ml at the
-------
TABLE 5. MICROBIAL CONCENTRATIONS IN LEACHATES OBTAINED FROM THE LABORATORY LYSIMETERS
AT THE 13th AND THE 104th WEEK AFTER CONSTRUCTION
Quantitative
Tests
MPN/100 ml
Total col i form
Fecal Collform
Fecal Streptococci
Agar Plates, CFU/100 ml
Blood -aerobic
Blood -anaerobic
Standard Methods
Sabouraud
Tellurite
A
20 .
(<2)f 1
B
86
(<2)
<20 ^20
4.9 x
1.9 x
(1.2 x
6.3 x
(1.2 x
1.4 x
(1.0 x
2.2 x
(3.1 x
1.1 x
(4.0 x
102
2)
10J
108
10B)
108
108)
106
106)
106
106)
1.6
1
1.2 x
(1.3 x
1.1 x
(3.3 x
5.4 x
(1.4 x
4.9 x
(4.2 x
1.6 x
(2.9 x
X<104
106
106)
106
106)
106
106)
105
10$
104)
uc
c
*
Laboratory Lysimeters
D
40
-------
15th week and had measurable levels lasting
through the 35th week. The lysimeter B
leachate contained total coliform levels
of 8,0 x 10 /100ml at the 13th week, and
decreased to < 2/100 ml thereafter. MPN
levels in leachates from lysimeters C and
F remained at < 20/100 ml and did not
change. Thus, the two lysimeters contai-
ning sewage sludge continued to show
coliforms in their leachates for longer
periods of time than those lysimeters
without sewage sludge. As described
above, lysimeter A continued to yield
fecal bacteria for 2 years after its
construction. The very sensitive proce-
dure of Allen et al.(2) detected stressed
or small numbers of bacteria, while the
less sensitive procedure for Standard
Methods (1) did not allow their growth or
their detection.
The Identification of Streptococci Groups
and Species
Blannon and Peterson (3) suggested
that the speciation of streptococci
isolated from landfill leachate would
indicate the natural origin of the strep-
tococci, and this in turn would indicate
the origin of the solid waste. The
natural origins of all streptococcal
species are presented in Table 6.
The identified streptococci from the
lysimeters are shown in Table ?. The
first group of streptococci Identified
here were isolated from the wastes added
to the lysimeters when they were construc-
ted, and from the six lysimeter lea-
chates. The lysimeter wastes, both
municipal and hospital, contained equal
numbers of Group D and viridans strepto-
cocci. Most of the Group D were identi-
fied as Streptococcus faecal is subsp.
1 iquefaciens. which is of plant or insect
origin. This latter microbe has no
health significance. On the other hand,
the viridans group does come from man,
and this indicates the human source of
the waste.
Streptococci isolated from the
lysimeter leachates were similar to those
found in the solid waste. Host of the
streptococci from Lysimeter A were
viridans, which originates from man.
Lysimeter B leachates contained S.. faecium
anc^ !.• JJuTans. These both originate from
the feces of man and animals. Some of the
streptococci isolated from the leachates
produced by municipal solid waste lysi-
meters C and D were Group D streptococci,
but not enterococci. These cocci proba-
bly came from animal feces , such as
pets, horses and cows.
A year later additional grouping and
speciation were done on the surviving
streptococci from these lysimeters and
landfills. The results are shown in
Table 8. After a year and a half, the B
and E lysimeters still retained viable
human or animal Streptococcus faecal Is,
as did the Boone County lysimeters land-
fill after 8 years. The fecal origin of
the commercial landfill is also evident
from the presence of large numbers of
Group D enterococci, Streptococcus faecal is,
and viridans streptococci. Since these
streptococci were able to survive for 8
years under landfill conditions, so may
pathogens of equal resistance.
The final experiment on the six
laboratory lysimeters was the determin-
ation of surviving pathogen and fecal-
indicator concentrations in the solid
waste within each lysimeter. Two lysi-
meters have been examined, although the
results from the D lysimeter only is
presented in Table 9. The waste samples
were taken at 10 on intervals. It may be
recalled that the leachate from D lysi-
meter after 2 years (lable 5) had no fecal
conforms and only 10/100 ml fecal strep-
tococci. Yet in the solid waste within
the lysimeter, measurable levels of the
fecal-indicators were obtained. The
survival of these bacteria were related to
the pH of the solid waste, as the micro-
bial counts decreased with an increase in
acidity. In addition, the Boone County
experimental landfill's solid waste was
analyzed after 9 years at various levels.
Four soil controls from a nearby site were
taken at the same depth levels -as those
from the landfill, and analyzed in a
similar manner. The results of these
studies in Table 9 demonstrate that the
solid waste's coliform and streptococci
levels in the landfill differ signifi-
cantly from those in the control sod
samples. Moreover, the solid waste and
leachate in the bottom part of the land-
fill had significant levels of fecal-
indicator bacteria present. Apparently,
these bacterial strains had survived in
the waste for 9 years. It will be
recalled from Table 3 that the fecal-
indicator levels in the leachate after 6
48
-------
TABLE 6. THE HUMAN. ANIMAL & VEGETABLE ORIGINS
Of STREPTOCOCCAL GROUPS
*
Group
A
B
Not Groups A, B or D
(i.e., Groups C, G, F)
D, not enterococcus
Enterococcus (Group 0)
V1r1dans
Species
S^. pyogenes
S.. agalactiae
S.. dysgalactiae
S.. equi
S. equisimilis
S. zooepidemicus
S. anginosus
£. bovis
S. equinus
S. faecal is
S. faecal is subsp.
1 Iguefaciens
S. faecal is subsp.
zymoqenes
S^. faecium
S.. durans
S. sallvarlus
S. mitis
Origin*
man
cattle, man
cattle
horse
man, animal
animal
man
animal
horse
man, animal
plants
Insects
man, animal
man, animal
man, animal
man
man
S. aylum
fowl
Lancefield Groups are A through Q.
.u
From Wilson & Miles (26), except for S.. faecal is subsp. liquefaciens
which 1s from Geldreich, et al. (10).
49
-------
TABLE 7. STREPTOCOCCAL GROUPS AND SPECIES ISOLATED
FROM SOLID WASTE USED IN THE LYSIMETERS
AND FROM THE LEACHATE AFTER CONSTRUCTION
Streptococcal Groups
Group A
Group B
Not Groups A, B or D
(I.e., Group C, G and/or F)
Number of Streptococcal Isolates From:
**
Municipal Hospital Lysimeters
Solid Waste Solid Waste A B C D E F
1 # ######
1 # ######
2 1 #«####
Group D (enterococcus)
S_. faecal is subsp.
liquefaciens
S. faecalis
S. faecium
S. durans
TOTAL Group D
Group D (not enterococcus)
Viridans Group
S. salivarius
TOTAL Viridans Group
3
1
1
f
5
1
3
2
5
5 f
# 1
$ 4
1 f
5 5
2 2
4 6
1 1
5 6
*
1
9
7
17
1
2
#
2
# # 9
* # 3
4 # 7
3 # 1
5 6 4
# f 1
i # #
f f 1
10
8
,
4
"22
1
1
#
1
The Streptococcal groups were identified by the procedure of Facklam, et al. (9)
and the species were identified by the procedure of
Pavlova, et al. (18).
**b
The content of lysimeters are: A=sewage sludge; B=municipal solid waste plus
sewage sludge; C=municipal solid waste; D=municipal solid waste; E=municipal solid
waste plus hospital waste; F=hospital waste.
*None detected.
50
-------
TABLE 8. THE PRESUMPTIVE IDENTIFICATION OF STREPTOCOCCAL GROUPS
AND SPECIES ISOLATED FROM LANDFILL LEACHATES.
Strepto- Sensitiv-
coccal ity to
Sources Bacitracin
Lysimeter
Leachates:
UC-E
UC-B
BC-2
CH-14
Commercial
Landfill ,
Leachates (CL)ff:
CL, Sample 1
CL, Sample 2 +
CL, Sample 3
CL, Sample 4
CL, Sample 5
Streptococcus
faecal is
ATCC 19433
Growth on:
Hip- Bile 6.5% Number
purate Esculin NaCl
Test
+ + 14
+ + 2
+ + 12
+ + 10
+ + 16
+ . 2
+ 2
3
11
+ 7
3
•»• - 2
+ 2
-i- + 4
+ > 1
Total 91
Isolates:
Identification
D-enterococcus
(6 S. faecal is)
D-enterococcus
(2 S. faecal is)
D-enterococcus
(12 S. faecal is)
D-enterococcus
D-enterococcus
(15 S. faecalis)
D-enterococcus
(2 S. faecal is)
D-enterococcus
Viridans
D-enterococcus
D-enterococcus
Viridans
D-nonenterococcus
D-enterococcus
D-enterococcus
D-enterococcus
*=Facklam et al. (1974)
**=Pavlova et al . (1962),
Stanard Methods (1976), Bordner et al .
(1978)
#=0btained from the surface of the landfill slope.
-------
TABLE 9 FECAL INDICATOR BACTERIAL PRESENCE IN MUNICIPAL SOLID WASTE
FROM A TWO YEAR OLD LABORATORY LYSIMETER AND A NINE YEAR
OLD LANDFILL
Type of Sam
Sample At
Des
Lev
UC Lysimeter D
Solid Waste
Solid Waste
Solid Waste
Leachate
Boone County Landfil
Top Soil
Solid Waste
Solid Waste
Solid Waste
Solid Waste
Leachate
Decomposed
Waste at
Plastic Liner
iples Collected MPN/100 grams of Solid Waste and soil pH
These or MPN/100 ml Leachate of
cending
els
10 cm
20 cm
30 cm
40 cm
1 BC-1
46 cm (1% ft.)
152 cm (5 ft.)
214 cm (7 ft.)
259 cm (83$ ft.)
274 cm (9 ft.)
304 on (10 ft.)
320 cm (10*5 ft.)
Total
Coli form
1.3 x 103
2.2 x 103
2.0 x 101
<• 2
2.4 x 105
1.1 x 102
2.4 x 103
1.6 x TO5
5.4 x 104
9.2 x 103
3.5 x 104
Fecal Fecal
Coli form Strepto-
coccus
9.0 x 101 7.1 x 104
< 20 2.4 x 103
•^20 2.0 x 101
<2 *2
3.5 x 103 9.8 x 105
< 20 2.0 x 104
2.0 x 101 6.0 x 104
2.3 x 102 3.3 x 105
? **
3.3 x 10^ ND
3.5 x 102 1.1 x 102
4.9 x 102 3.5 x 104
Sample
7.1
6.2
5.2
5.2
7.0
6.0
5.4
7.4
7.6
7.0
6.9
Control -Soil near Boone County Landfill
Top Soil
Soil
Soil
Soil
46 cm (1h ft.) _>
152 cm (5 ft.)
213 on (7 ft.)
259 cm (8% ft.)
2.4 x TO5
2.3 x 102
5.0 x 101
X
5.0 x 101
* 20 2.1 x 103
-------
years had no total and fecal col iform 4.
bacteria present but had 10/100 ml of
fecal streptococci. However, the high
levels of microbes in the Boone County
landfill solid waste suggests that fecal-
indicator bacteria, and perhaps pathogens
related to them, survive for longer periods
of time. Finally, the fecal-indicators
survived in higher concentrations in the 5.
landfill than in the lysimeters. This may
indicate a difference 1n the waste storage
conditions, such as the volume of waste,
temperature, humidity, pH, age of the
waste, etc. 6.
CONCLUSIONS
This work shows that some pathogenic
bacteria have been isolated from landfill
leachate, and suggests that with improved 7.
methods, more may be found. In addition,
leacnate from a 2 year old sewage sludge
lysimeter continued to produce leachate
intermittently, containing significant
levels of fecal-indicator bacteria. 8.
Finally, these same bacteria were demon-
strated in municipal solid waste and
hospital waste from lysimeters that were
2 years old, and also from a fullscale
landfill that was 9 years old. These
results demonstrate that fecal-indicator
bacteria do not die as rapidly as previ-
ously reported, but survive for extended
periods of time with the potential to
contaminate ground and surface waters
when released from the landfill. 9.
REFERENCES
1. American Public Health Association.
1976. Standard Methods for the
Examination of water and Wastewater,
14th ed. American Public Health 10.
Association, N.Y.
2. Allen, M.J., R.H. Taylor and E.E.
Geldrelcn. 1976. The Impact of
exesslve bacterial populations on
collfortn methodology. Am. Water
Works Assoc., Water Quality Techno!- 11.
ogy Conference, San 01ego, Califor-
nia. December 6 and 7, 1976.
3. Blannon, J.C. and M.L. Peterson.
1974. Survival of fecal conforms
and fecal streptococci in sanitary
landfill. News of Environmental
Research in Cincinnati, April 12,
1974. U.S. Environmental Protection . 12
Agency, Cincinnati, Ohio. 4pp.
Bordner, R., J. Winter, and P.
Scarpino. 1978. Microbiological
Methods for Monitoring the Environ-
ment, Water and Wastes.
EPA-600/8-78-017. U.S. Evironmental
Protection Agency, Cincinnati, Oh.io.
pp. 338
Carpenter, L.V. and L.R. Setter.
1940. Some notes on sanitary land-
fills. Amer. J. Pub! Health, 30:
385-393.
Cook, H.H., D.C. Cromwell, and H.A.
Wilson. 1967. Microorganisms 1n
household refuse and seepage water
from sanitary landfills. Proc. W.
Va. Acad. Sci., 39: 107-114.
Cowan, S.T. 1974. Cowan and Steel's
Manual for the Identification of Me-
dical Bacteria. 2nd ed., Cambridge
University Press, New York.
Engelbrecht, R.S., M.J. Weber, P.
Amirhor, D.H. Foster, and D.
LaRossa. 1974. Biological Properties
of sanitary landfill leachate, In:
J.F. Milana, Jr. and B.P. Sagik
(eds.) Virus Survival in Water and
Waste Water Systems, Water Resources
Symposium, Number Seven. Center for
Research 1n Water Resources, Univer-
sity of Texas, Austin pp. 201-217.
Facklam, R.R., J.F. Padula, L.G.
Thacker, E.G. Wortham, and N.J.
Sconyers. 1974. Presumptive identi-
fication of Group A, B, and D strep-
tococci ,
107-113.
Appl. Microbial.,
27:
Geldreich, E.E., B.A. Kenner, and
P.W. Kabler. 1964. Occurence of
conforms, fecal conforms, and
streptococci on vegetation and
Insects. Appl. Microblol., 12;
63-69.
Glotzbecker, R.A. and A.L. Novello.
1975. Polio virus and bacterial
indicators of fecal pollution 1n
landfill leachates. News of Envi-
ronmental Research 1n Cincinnati,
January 31, 1975. U.S. Environmen-
tal Protection Agency, Cincinnati,
Ohio. 4 pp.
Koneman, E.W., S.D. Allen, V.R.
Dowel 1, Jr., and H.M. Sommers. 1979.
53
-------
Color Atlas and Textbook of Diagnos-
tic Microbiology. J.B. Lippincott
Co., Philadelphia.
13. Koneman, E.W., G.D. Roberts and S.F.
Wright. 1978. Practical Laboratory
Mycology. 2nd ed. Williams and
Wilkins, Baltimore.
14. Lennette, E.H., A. Balows, W.J.
Hausler, Jr., and J.R. Truant. 1980.
Manual of Clinical Microbiology. 3rd
ed, American Society for Micro-
biology, Washington, D.C.
15. Lennette, E.H., E.H.- Spaulding and
J.P. Truant. 1974. Manual of Clini-
cal Microbiology. 2nd ed., American
Society for Microbiology, Washington,
D.C.
16. MacFaddin, J.F. 1976. Biochemical
Tests for Identification of Medical
Bacteria. The Williams & Wilkins
Co., Baltimore, Maryland.
17. Majou, J. 1976. MPN-Most probable
number. In: M.L. Speck (ed.)
Compendium of Methods for the
Microbiological Examination of Foods.
APHA Intersociety/ Agency of
Committee of Methods for the Micro-
biological Examination of Foods,
Washington, D.C., pp 152-162,
18. Pavlova, M.T., F.T. Brezenski, and
W. Litsky. 1972. Evaluation of
various media for isolation, enumera-
tion and identification of fecal
streptococci from natural sources.
Health Lab. Sci., 9:289-298.
19. Pohland, E.G., and R.S. Engelbrecht.
1976. Impact of Sanitary Landfills:
An Overview of Environmental Factors
and Control Alternatives. Prepared
for American Paper Institute. 82 pp.
20. Postgate, J.R. 1967. Viability
measurements and the survival of
microbes under minimum stress.
Adv. in Microbial Physiol. 1: 1-23.
21. Public Health Service. 1976. Clas-
sification of Etiologic Agents on
the Basis of Hazard. Dept. of
Health, Education, and Welfare,
Center for Disease Control, Atlanta,
Georgia.
22, Scarpino, P.V. 1971. Bacterial and
viral analysis of water and waste
water. In: L.L. Ciaccio (ed.)
Water and Wastewaster Pollution Hand-
book. Vol 2, Marcel Dekker, New
York. pp. 639-761.
23. Scarpino, P.V. 1975. Human enteric
viruses and bacteriophages as indica-
tors of sewage pollution. In:
A.L.H. Gameson (ed.) Discharge of
Sewage from Sea Outfall. Pergamon
Press, Oxford, England, pp. 49-60.
24, Scarpino, P.V,, J.A. Donnelly and D.
Brunner, 1979. Pathogen content of
landfill leachates. In: M.P.
Wanielista and J.S. Taylor,_{eds.)
Proceedings of the Fifth Annual
Research Symposium Municipal Solid
Waste: Land Disposal, Solid and
Hazardous Waste Research Division,
USEPA, Cincinnati, Ohio. pp. 138-167.
25. Sobsey, M.D. 1978. Field survey of
enteric viruses in solid waste land-
fill leachates. Amer. J. Public
Health 68:858-864.
26 Wilson, G.S. and A. Miles. 1975.
Topley and Wilson's Principles of
Bacteriology, Virology and Immunity.
Vol. 1, 6th ed. The Williams and
Wilkins Co., Baltimore, Maryland.
pp 719.
27. Wolf, P.1., B. Russell and A. Shimoda.
1975. Practical Clinical Micro-
biology and Mycology: Techniques and
Interpretations. John Wiley & Sons,
New York.
54
-------
WASTEWATER TREATMENT PLANT RESIDUAL LANDFILLING: A CRITICAL REVIEW
Robert J. Stearns
James C. S. Lu
Robert D. Morrison
Calscience Research, Inc.
7261 Murdy Circle
Huntington Beach, CA 92647
ABSTRACT
A literature review was conducted to critically evaluate the land-filling of wastewater
Treatment Plant (WTP) residuals. Current practices, adverse environmental and public
health impacts, beneficial uses, available control technology and management options
are described and evaluated. Existing literature revealed a lack of information regarding
the environmental and public health impacts related to WTP residual landfilling.
Epidemiological information regarding the effects of WTP residual landfilling was absent.
WTP residual/refuse co-disposal and sludge-only narrow trench alternatives represent
the most environmentally acceptable of the landfilling methods. For the most part,
co-disposal practices pose greater environmental public health risks than conventional
municipal refuse landfills, the extent of which is largely a factor of residual treatment
and landfill management. Apart from organic and microbiological components, leachates
from WTP residual landfills (alone or co-disposal) are not unlike municipal refuse landfill
leachates. It is apparent that the practice of residual disposal in municipal refuse
landfills is not only cost-effective, but can be properly managed to provide protection
of the environment and public health.
INTRODUCTION
Disposal of Wastewater Treatment Plant
(WTP) residuals have historically occurred
by:
1. Various land burial procedures
2. Direct discharge to oceans
3. Internment in evaporative (or
percolation) type lagoons and
ponds
4. Incineration
5. Agricultural utilization
6. Spray irrigation
7. Land and/or forest application
Bans on ocean dumping of residuals and the
high costs of incineration have stimulated
interest in landfilling. Recent amendments
to the Clean Water Act and Solid Waste Act
require that landfilling of these residuals,
either alone or with refuse, does not pose
an unreasonable risk of adverse environ-
mental or public health impacts.
The purpose of this study was to
critically evaluate the landfilling of WTP
residuals for the purposes of:
a) describing current disposal
practices
b) determining adverse/environmental
and public health impacts
c) describing available control
technology
d) evaluating management options
Landspreading and beneficial use of sludge,
such as for agricultural uses, was not con-
sidered in this evaluation.
Information from a variety of sources
was selected on the basis of its signifi-
cance and relevance. These sources
included a computer search of 15 commer-
cially available data bases, general
references, solid waste management repre-
sentatives, and landfill sites personnel
where studies had been conducted. Where
information was deficient, conclusions
55
-------
were drawn based on experience, scientific
judgement, and inference, and are subject
to testing and confirmation.
RESULTS AND DISCUSSION
Types of Residuals
Since the landfilling of WTP residuals
is primarily constrained by their physical
properties (e.g. moisture content) and
environmental/public health considerations,
this report will classify the residuals
according to those treatment processes
which represent distinct changes in the
physical, chemical, or pathogenic nature
of the residual.
In general, residuals can be charac-
terized according to whether they are
treated or untreated. Untreated residuals
are represented by:
t grit, screenings, and floatables
• raw, primary sedimentation sludge
* raw, waste activated sludge
The extent to which a residual is
stabilized prior to landfilling may have
a pronounced affect upon leachate and gas
generation and composition, the pathogenic
nature of the residual, the quantity and
physical properties of the residual, and
residual odor. Advantages and disadvan-
tages of the major stabilization process
include:
1. Anaerobic and aerobic digestion-
reduces total sludge mass and thus
reduce cost of disposal; reduces
pathogens; produces a sludge of
relatively poor mechanical dewater-
ing characteristics.
2. Lime treatment-reduces odor and odor
production potential in sludge; re-
duces pathogen levels; improves
dewatering and settling characteris-
tics; lowers concentrations of soluble
phosphate, ammonia nitrogen and total
Kjeldahl nitrogen, and heavy metals.
3. Chlorine oxidation-inactivates
pathogens:.reduces odor; improves
dewaterability of digested sludge;
chlorine-stabilized sludges pose
substantial risks of highly toxic
leachates because of its high acidity
and potential for releasing chlorinated
hydrocarbons and heavy metals.
4. Heat treatment-inactivates pathogens
and can reduce sludge mass; heat
treatment cannot reduce odor.
Because of potential leachate gener-
ation, landfilling of WTP residuals
usually requires that residuals be dewa-
tered prior to disposal. Dewatering
residuals from a 5 to 20 percent solids
concentration reduces volume by three-
fourths and results in a damp solid.
Various residual conditioning methods may
be employed which are designed to improve
the dewaterabil ity of WTP residuals, and
can provide disinfection, affect odors,
alter the wastewater solids destruction
or addition. Common methods include:
1. Inorganic chemical conditioning—the
addition of Time; ferric chloride, or
ferrous sulfate as flocculant aids;
inorganic chemical conditioning in-
creases the sludge mass by 15 to 30
percent, but can be beneficial because
of its stabilization effects.
2. polyelectrolyte conditioning—little
additional sludge mass is produced
3. Thermal conditioning—the solids con-
centration of the cake and thus its
dewaterability is nearly double that
of the previous two conditioning
alternatives; process sterilizes
the residual.
4. Elutriation—the process of washing
the alkalinity out of anaerobically
digested sludge, improving its
dewaterability.
High temperature processes, such as
incineration or wet air oxidation, sign-
ificantly reduce residual volume and weight
(95%), thereby, reducing disposal require-
ments. Additionally, these processes can
destroy or reduce toxics that otherwise
create adverse environmental impacts.
Current Practices of WTP Residual
filling ~~
Land-
Landfilling approaches for WTP resi-
duals are usually classified according to
three major categories:
• Residual-only Trench
56
-------
Narrow Trench
Wide Trench
« Residual-only Area Fill
Area Fill Mound
Layer Fill
Diked Containment
* Co-disposal
Residual/Refuse Mixture
Residual/Soil Mixture
The residual-only trench method involves
the excavation of a trench into which the
residual is placed. The excavated soil
is subsequently placed over the residual
as a cover. Area fill methods involve
residual placement upon the original
ground surface. In general area fills
require residual mixing with a soil bulk-
ing agent in order to provide sludge
stability. Residual-only area fills are
constructed in locations where a high
groundwater table or nearrto-surface bed-
rock conditions prevail. Cover soil needs
place additional demands upon onsite or
imported soil needs.
Co-disposal is the practice of depos-
iting residuals at a conventional sanitary
landfill. Co-disposal practices offer
significant cost savings due to the shared
capital and operational expenses with
sanitary landfills ( i , z ,25 ). Co-
disposal of soil and residual is used as
a cover for municipal refuse landfills.
POTENTIAL ENVIRONMENTAL IMPACTS
Impacts on Water Quality
Because of the toxic and sometimes
pathogenic nature of WTP residuals, its
placement into a landfill may lead to the
movement of highly contaminated leachates
out of the fill and into the surrounding
soil and subsurface aquifers. Figure 1
depicts the concentration ranges of various
chemical and biological constituents of
leachates from (1) municipal refuse (2)
admixed municipal refuse and WTP residuals
and (3) WTP residuals alone. In the latter
cases, the residuals were either untreated
or stabilized. Segregation of disposal
methods according to types of residuals
was not attempted since many studies failed
to indicate the degree of residual
treatment, and since many landfills accept-
ed assorted residuals. Documented invest-
igations of WTP residual landfill leachate
composition are minimal, considerably less
than for municipal landfills. The major
research was performed by Lofy et al. (11),
Stone (24), Johansen et al. ( 9), Emcon
and Associates (4), Pickard (16), Walker
et al. (26), Charlie et al. (3), and
Sikora et aj. (20).
Figure 1 indicates that concentration
ranges of leachates constituents of munic-
ipal refuse landfills and WTP residual/
refuse co-disposal are similar for the
major organic indicators, total solids,
nutrients, major ions, and pathogens.
Maximum concentrations of heavy metals
and pH, however, differ. Leachates of
municipal refuse exhibit a lower minimum
pH, and higher maximum concentrations of
Fe, Zn, and Cd than co-disposal methods.
This evidence indicates a possible buff-
ering effect provided by WTP residuals
with regard to municipal refuse leachates,
decreasing the solubility of various heavy
metals.
With the exception of pH, COD and
certain heavy metals, Figure 1 shows no
significant difference in the quality of
leachates of residual only and residual/
refuse co-disposal landfills. The pH of
the residual only disposal method was found
to exist within a relatively narrow range
(6.2 - 8.3) in comparison to residual/
refuse co-disposal methods (4.5 - 10).
Interestingly, maximum concentrations of
iron and zinc in residual only leachates
are nearly one order of magnitude less than
co-disposal leachates. It should be noted
that comparatively few studies are repre-
sented in the WTP residual leachate data;
further data is necessary to verify these
results.
Concentrations of WTP residual leach-
ate constituents, whether landfilled alone
or with municipal refuse, pose adverse
impacts to the beneficial uses of receiving
waters. Of primary concern are specific
organics (e.g. toxic refractories such as
PCB, pesticides, etc.), heavy metals, and
pathogens. Table 1 identifies the poten-
tial groundwater contaminants associated
with various common WTP residuals, assuming
improper landfill management. A comprehen-
sive review of the water quality impacts
posed by WTP residual leachates is pre-
sented by Lu et al. (12).
-------
10 -
tSS SP. COMD.
"^(KMg/CH)
NOTE! LEGEND PRESENTED
0« LAST PAGt OF THIS
FIGURE
w-
10*
».
10*
ID
1
KOTC: LEGEND PRESEHTtt OK LAST PAGE OF THIS FIGURE
It
/ — v
Ci
/jfc-
I
TOTAL
f—
V *
— •
-~S
>
G
^_
—
If
k
—\
n
-
-
-
10"
10'
10
1
c
*}
ll
I
0.1
m
FE
r
-
-\
r~
•M • •
NOTE: LEBEKD PRESENTS) OK LAST CASE
OF THIS FIWRE
-\
Cu
1 :
1 i
"N
f~
i *
i
-\
/
•I •*
r
--
^ in
/ —
.,.-
ins
10*
O T
i
1
10
i
F, a
-/—
-
«•' r. STB
"~v /—
IP.
PH
n / A
1 I I 1 1 1 1
156? eg 10.
{ . . - - — -li- — K -i- • -|
LEGEND: 1 1 Q
SLUDGE MUKI . SLUDGE/
Obit REFUSE REFUSE
ONLY CO-DISPOSAL _
•ATM QUALITY- CB1TM1A
IPA PU1LIC WATER SUPPLY
EPA AGRICULTURAL ClSRISATIO
— — IP* WISH I«TI« (AOUATIC
Lift)
EPA MARINE KATCR (AQUATIC
LIFE)
BID STATE OF NEK 10KI, "
MISSOURI GROUND««IER
Figure 1. Comparison of leachates fpom residual-only, residual/refuse, and municipal
refuse landfills.
-------
TABLE 1. POTENTIAL GROUNDWATER CONTAMINANTS OF VARIOUS WTP RESIDUALS*
Type of Residual
Contaminants
Type of Residual
Contaminants
Liquid - unstabilized
Gravity thickened primary,
WAS and primary, and WAS
Flotation thickened primary
and WAS, and WAS without
chemicals
Flotation thickened WAS with
chemicals
Thermal conditioned primary
or WAS
Liquid - stabilized
Thickened anaerobic digested
primary and primary, and
WAS
Thickened aerobic digested
primary and primary, and
WAS
Thickened lime stabilized
primary and primary, and
WAS
Dewatered - unstabilized
Vacuum filtered, lime
conditioned primary
Org.,HM, TDS, NOg, Path.
Org.,HM, TDS, NOj, Path.
Org., HM, TDS, Path.
Org., HM, TDS, N03
HM, TDS, N03
HM, TDS, N03
HM. TDS
Org., HM. TDS, NO-, Path.
Dewatered - stabilized
Drying bed digested and TDS
lime stabilized
Vacuum filtered, lime HM, TDS, NO,
conditioned digested
Pressure filtered, lime HM, TDS, NO,
conditioned digested
CentHfuged, digested and HM, TDS, NO,
lime conditioned digested
Heat dried
Heat dried digested
High temperature processed
Incinerated dewatered
primary and primary, and
HAS
Wet-a1r oxidized primary
and primary,.and WAS
HM, TDS, N03
HM
Org,, HM, TDS, NOj, Path.
* Resulting from improper WTP residual landfill management
WAS - Waste-activated sludge
Org.- Organics
HM - Heavy Metals
TDS - Total Dissolved Solids
NO, - Nitrate-nitrogen
Path. - Pathogens
-------
Impacts on Air Quality
Noxious gases, including H~S and var-
ious niercaptans, are commonly associated
with WTP residuals and represent a pri-
mary source of adversity with regard to
residual disposal near population centers.
Like sanitary landfills, carbon dioxide and
methane comprise as much as 90 percent of
the gases produced at residual/refuse
co-disposal sites.
Volatilization of organic constituents
can occur at residual landfill sites. The
vaporized contaminants of most concern,
due to their toxicity, are halogenated
organics and aromatic hydrocarbons. Aside
from pesticides, available information on
the volatilization of these compounds is
limited. Spencer and Claith (23) have
suggested that volatilization of surface
deposits and soil incorporated pesticides
plays a major role in the dissipation of
pesticides from solid surfaces. The
potential volatibility of pesticides and
other organic compounds is related to the
inherent vapor pressure of the compound,
but the actual volatilization will depend
upon environmental conditions. Shen
and Tofflemire suggest that at least 99%
reduct
-------
£_
jSludat UMflll 1
>L
—1.
)
_!_
Vtttor Transport:
rlltl, BotttntJ.
ma airai
Direct Syr'ace
w«trr Runoff
Bunoff froai
Splllagl at
Sltl
Sptllaot freai
transport Viklclcf
I
k**£S,
ration]
^
[soil inaBetHtattwi) * (ii«-p«i»a>«_[ 1
\
Mroiol s
IM
Vapori
Surface
tutir
X^
-> — •
Srownd-
**ftf
\
Sypf«t
,- — "
I. lln.r (arttflcal or irnttwtlck llptnMaftlt cov«r loll
Surface *at«r-co«trel
1. N*lRten«ncff of aUallnt pN
3. Safety locks on iructt, clotttf vcfttctts
4. Surfacv wattr control iwaiur*: t.c,, dtvtrtton dltcfcct,
tU: tr«*tMllt «U>1 Hut
S. Staw as 4: fnstaUatlon of eovtr soil;
C. Cow 10H dUlnfuunt stoekpllM iludft rMucttoi
of tsposurt of sluda*
7. Colttctlon and trtau»nt/r*cyc1<; utlllii o«*rlan
flow prlnclplts for trtauwftt
a. 5«M K ;.
Figure 2, Simplified contaminant pathways chart and commonly employed control measures.
generated by municipal wastewater process-
ing systems ( 7,10,15,21,22).
WTP residual landfill design and
operational procedures govern the avail-
ability of pathways for contaminant trans-
mission. Trenching of residuals provides
several inherent advantages toward minimiz-
ing contaminant transport from disposal
sites. These include:
• Absence of premlxing with soil
reduces available contact time with
the environment
t Stockpiling is seldom employed
* Erosion of residual is less pro-
nounced due to its placement below
ground level.
Conversely, area fill operations (area fill
mound, area fill layer, and diked contain-
ment} have several design features con-
ducive to contaminant transmission via
surface runoff. These Include:
• Continuous slumping and mound
movement
• Erosion of mound or dike failure
(in the case of diked containment)
a Potential of runoff from stockpiled
residuals during wet weather
periods.
WTP residual/refuse co-disposal operations
pose risks of contaminant transmission via
surface runoff due to the logistics of
refuse and residual arrival at the site
(I.e. stockpiled residuals).
Available Control Techno!ogy
Measures to control the transport of
pollutants from WTP residual landfills
are presented in Figure 2. Many of the
control measures are presented in detail
in the Process Design Manual for Municipal
Sludge Landfills (18). Leachate control,
including leachate volume control, col-
lection, and treatment is the subject of
much recent research, particularly for •
sanitary landfills. Because WTP residual
only landfill and co-disposal landfill
leachates are of similar composition to
municipal landfill leachates, control and
treatment measures should apply in either
case. A comprehensive review of leachate
control and treatment measures Is provided
by Lu et al. (13).
Management Options
The most significant features govern-
ing the selection of a WTP residual land-
filling method are:
* Residual percent solids
• Residual characteristics (stabiliz-
ed or unstabilized)
• Hydrogeology (deep or shallow
groundwater or bedrock)
• Groundslope
-------
Because landfill ing of WTP residuals is
constrained by environmental impacts, pub-
lic health impacts, costs, and the avail-
ability of land, the selection process must
estimate the potential magnitude of each of
these factors with regard to each landfill-
ing alternative. Table 2 shows the results
of these efforts. Of the WTP residual
landfilling alternatives, the residual only
trench and residual/refuse co-disposal
methods represent significant advantages in
terms of costs, environmental factors
(odors, vectors, and aerosols), potential
public health impacts, and potential for
future land use at disposal sites.
Cost rankings in Table 2 were based
on total site costs, i.e. capital plus op-
erating costs. Co-disposal methods can, in
some instances, provide substantial cost
savings in residual disposal inasmuch as
dewatering may not be required. Stone (24)
estimated that 0.6 to 1.8 Ibs of liquid can
be added for every 1.0 Ib of dry weight
solir" waste before complete saturation is
reached. Based on an initial refuse mois-
ture content of 65 percent (which repre-
sented a "worst case" condition in Stone's
study), the maximum possible mix ratios for
residual/refuse co-disposal methods are
listed in Table 3 . These values indicate
the quantity of residual that can be
applied to municipal refuse without ex-
ceeding its moisture retention capacity.
Of course, rainfall and other sources of
moisture will reduce the absorbtive capac-
ity of refuse and may significantly
decrease the ratios indicated. Field
testing is necessary to verify these values.
Residual dewatering is imperative to
many landfill designs and operations from
a construction standpoint. Table 4 out-
lines the important physical character-
sties of residuals of different solids
content in terms of workability. Other
residual characteristics which determine
suitability for landfilling include:
• The anticipated composition and
strenth of leachates
a The anticipated rate and volume of
leachate production
• The anticipated composition and
volume of gas produced
• The extent of stabilization
• Residual settlement
TABLE 2. CRITERIA INFLUENCING THE SELECTION OF A RESIDUAL LANDFILLING PRACTICE
Current UTf Residual Landfilling Practices
Residual Only » Residual Only Trench
Narrow Trench
Wide Trench
• Residual OnW Area Fill
"••. Area Fill Mound
Layer Fill
Diked Containment
Residual/Refuse • Residual Spread With Refuse.
Co-Disposal Covered With Soil
• Trench, Covered With
Refuse and Soil
• Residual Dumped on Refuse
Covered With Soil
• Residual Dried in Piles, Spread
with Refuse, Covered With Soil
i Residual Mixes witn boil, ana
Applied To Refuse
1
I
c
—1
H
K-L
H-L
H
L
N
H
H
H
Soil Bulking Agent Required II
Nn
No
Yes
Yes
Yes-No
No
No
NO
NO
Yes
Residual Solids Content
Recotimended (I)
> 15
i 20
i 20
2 15
i 20
> 3
i 3
> 3
> 3
> 20
VI
1
M
H
H
H
4-M
L
L
L
L
1-Ll
nv Iron mental
Factors
n
A 0
i!
i
L L
• H
1 H
H H
H H
Ll
H H
M H
O
°
M
H
M
M
H
H
M
M
M
K
)
B
1
U
*.
1
L
t
I
L
H
H
M
M
K
Aerosol!
1
M
M'
M
H
H
L
L
H
M
i
>
•
)
u
•** o
I C
Ol b
h- ' u
I
H
H
I
H
H
H
M
M
M
>
t
>
»
v- wi
J C
> J
1
K
H
I
H
H
M
K
L
H
I
«J
> Oi
i —
>
—
11
M
L
(
1
L
L
L
L
L
L
Key: H - High Potential (Comparatively)
-------
TABLE 3. THE MAXIMUM POSSIBLE RESIDUAL/
REFUSE MIX RATIOS FOR RESIDUALS OF VARIOUS
SOLIDS CONTENT*
Residual Percent
Solids Content
Maximum Residual/
Refuse Mix Ratio
5
10
15
20
25
30
1/5
1/5
1/4
1/4
1/3
1/3
* Based on refuse absorption capacity of
0.6 Ibs of water per Ib of dry wt solid
waste, and initial refuse moisture content
of 65 percent
* Weight ratio
The composition and strength of WTP resid-
ual leachates has been discussed. It is
speculated that the various types of resid-
uals can Influence the rate of leachate
generation through the effects of permea-
bility. Residuals of low permeability can
retard or reduce the flux of leachate from
the fill; a more permeable residual may
hasten leachate flux. Figure 3 qualita-
tively compares leachate flux among various
types of residuals based on measured
specific resistance* values. Coagulated
* Specific resistance can be defined as
the pressure required to produce a unit
rate of flow through a residual cake.
raw and digested residuals are more per-
meable than non-coagulated raw, digested,
and activated residuals. The coagulated
residuals, being agglomerates of smaller
particles, allow fluid to pass through more
readily than do the non-coagulated resid-
uals. The water retention characteristics
of the non-coagulated fines could be a
dominant factor in leachate production.
Thus, coagulation processes which tend to
increase the dewaterabiHty of WTP resid-
uals may present a disadvantage from the
standpoint of leachate generation.
The specific gas production from
anaerobic digestion of raw primary WTP
residuals generally ranges between 12 to
17 ftj per Ib of volatile solids destroyed.
Of this gas, between 42.5 and 75 percent
may represent methane (6 ). Complete
microblal degradation of one pound of
municipal refuse under optimal conditions
yields approximately 4-5 ft3 of methane
(8,14). Potential methane production is
therefore greater for raw primary residuals
than for municipal refuse. Figure 4 pre-
sents the methane production potential for
various WTP residuals based on the average
percent volatile content of the residuals
(5).
The stabilization of WTP residuals
prior to landfill ing is highly site-
specific. Unstabilized residuals are
usually unsuitable for landfilling because
of obnoxious odors. A comparison of
leachate constituent concentrations for raw
and digested residuals revealed no apparent
differences, though the data used was pri-
marily from residual/refuse co-disposal
landfill sites. The reduction in residual
pathogen populations afforded by stabili-
zation is important from a public health
TABLE 4. OPERATIONAL CONSTRAINTS OF RESIDUAL MOISTURE CONTENT
Percent Solid*
Workability and Application
1 - 15
15-20
28 - 32
• Frt« flowing fluid
co-disposal operations only
• «H1 not support cover Mterlal, l.t. soil sinks to the
bottom of sludge.
Appllcab • for narrow trenching due to soil bridging, an
operatic s wnere soil bulking 1s provided.
• Solid co
soils st
Sludge 1 . ., ...
for sludge only trenching, aree-f1.il operations, or iludgi/
soil mixtures to act as final cover over coapleted landfills.
e Sludge capable of supporting eoulpnent. Sludge still spread*
out evenly (n a trench when dropped froa atop. Applicable
for all sludge only operations
e Sludge -111 not spread out evenly in a trench when dropped
froa atop. Applicable to area-fill mound and layer appli-
cations
tent Is high enough to support soil cover, though
11 have a tendency to sink Into the sludge.
Incapable of supporting equipment. Applicable
63
-------
Figure 3, Relative rate of leachate pro-
ducible based on residual permeabilities
standpoint.
Settlement due to volume reduction
in landfilled residuals is important from
an environmental as well as planning per-
spective. Settlement can promote cracking
and create fissures in cover material,
thus leading to an increase in potential
odor and toxicant transmission and surface
infiltration. Substantial verticle and/
or horizontal displacement may occur which
may influence future lift placement oper-
ations or land use. Residuals with low
solids content (15-20%) can be expected
to settle more than residuals with a
higher solids content (>28%). Residuals
Figure 4. Comparison of WTP residuals
specific gas production
and
FigureS, Consolidation settlement for
residuals of different solids content
may dewater through evaporation or infil-
tration (into porous soils), and as
moisture is lost, compaction and settling
results. Based on original and final
void ratios, the estimated total consol-
idation settlement of WTP residuals of
various sol Ids content are listed 1n
Figure 5 . From Figure 5, a residual
of 8 ft depth and 25 percent solids con-
tent can settle 2.5 ft, representing a
settlement of over 30 percent. Settlement
rates as calculated closely match actual
findings by Slkora (20).
REFERENCES
1. Anon, R.B. 1978. Co-disposal of Sludge
with Refuse; Progress, Problems and
Possibilities. Sludge Magazine, 1 (2):
pp. 22.
2. Brunner, D.R., D.J. Keller, and 0.
Daniel. 1972. Sanitary Landfill Design
and Operation. PB-227-565. 50 pp.
3. Charlie, W.A., R.E. Wardwell, and O.B.
Andersland. 1979. Leachate Generation
from a Sludge Disposal Area. J. Env,
Eng. Div. ASCE. 105 (EES):947-960.
4. Emcon Associates. 1976. Sonoma County
Solid Waste Stabilization Study. EPA-
530/SW-120C. U.S. Environmental Protec-
tion Agency, Cincinnati, Ohio.
64
-------
5. Golueke, C.G., and P.H. McGauhey. 1970.
Comprehensive Studies of Solid Waste
Management, First Annual Report. U.S.
Public Health Service, Bureau of Solid
Waste Management, SW-3RG.
6. Herpens, H. and E. Herpens. 1966. Im-
portance of Production and Utilization
of Sewage Gas. KWG-KOHL-enwosse-Stoff-
gase.72 (18).
7. Hickey, J.L.S., and P.C, Reist. 1975.
Health Significance of Airborne Micro-
organisms from Wastewater Treatment
Processes, Part I. Summary of Investi-
gations, Part II. Health Significance
and Alternatives for Action, JWPCF,
47 (12):2741.
8, Jackson, A.G. 1975. Evaluation of San-
itary Landfill Gas and Leachate Pro-
duction. Systems Technology Corp.,
First Annual Report.
9. Johansen, O.J. and D.A. Carlson. 1976.
Characterization of Sanitary Landfill
Leachates. Water Research, Vol. 10,
pp. 1129-1134.
10. King, E.D., R.A. Mill, and C.H. Law-
rence. 1973. Airborne Bacteria from
an Activated Sludge Plant. J. Env.
Health. 36 (1):50.
11. Lofy, R.T., K.T. Phung, R.P. Stearns,
and J.J. Walsh. Investigation of
Groundwater Contamination from Sub-
surface Sewage Sludge Disposal,
Volume 1. EPA No. 68-01-4166. U.S.
Environmental Protection Agency, Wash-
ington, O.C.
12. Lu, J.C.S., R.D. Morrison, and R.J.
Stearns. 1980. Critical Review of
Wastewater Treatment Plant Residual
Disposal by Landfill ing. Draft report.
U.S. Environmental Protection Agency,
Cincinnati, Ohio.
13. Lu, J.C.S., R.D. Morrison, and R.J.
. Stearns. 1980. Leachate Production
and Management from Municipal Land-
fills; Summary and Assessment. Draft
report. U.S. Environmental Protection
Agency, Cincinnati, Ohio.
14. Merz, R.C. 1964. Investigation to
deterninge the quantity and quality
of gases produced during refuse de-
composition. California State Water
Quality Control Board, Sacramento,
Calif.
15. Napolitano, P.J. and S.R. Rowe. 1966.
Microbial Content of Air Near Sewage
Treatment Plants. Water and Sewage
Works, 113 (12)480-483.
16. Pickard, B.J. 1974. Landfilling Milled
Refuse Mixed with Digested Sewage
Sludge. M.S. Thesis. University of
Madison, Wisconsin.
17. Schmidt, J., D. Pennlngton, and J.
McCormick. 1975. Ecological Impact of
the Disposal of Municipal Sludge Onto
the Land. In: Proceedings of the 1975
Conference on Municipal Sludge Manage-
ment and Disposal. Anahlem, Calif.
18. SCS Engineers. 1978. Process Design
Manual for Municipal Sludge Landfills.
EPA-625/1-78-010. U.S. Environmental
Protection Agency, Cincinnati, Ohio.
19. Shen, T., and T.J. Toffleniire. 1980.
Air Pollution Aspects of Land Disposal
of Toxic Wastes. J. Env. Eng. Div.
ASCE, pp.211-226.
20. Sikora, L.J., W.D. Surge, and P.S.
Price. 1978. Chemical and Microbiolog-
ical Monitoring of a Sludge Entrench-
ment Site. In: Proceedings of the
First Annual Conference of Applied
Research and Practice on Municipal and
Industrial Waste. Madison, Wisconsin.
21. Sorber, C.A., H.T. Bausum, S.A. Schaub,
and M.J. Small. 1976. A Study of Bac-
terial Aerosols at a Wastewater Irri-
gation Site. JWPCF., 48 (10): 2367.
22. Sorber, C.A., S. Schaub, and H.T.
Bausum. 1974. An Assessment of a
Potential Virus Hazard Associated with
Spray Irrigation of Domestic Waste-
waters. In: Virus Survival 1n Water
and Wastewater Systems. Univ. of Texas
Press, Austin, Texas, pp. 241-252.
23. Spencer, W.F. and M.M, Claith. 1977.
Transfer of Organic Pollutants Between
the Solid-Air Interface. In: Fate of
Pollutants in the Air and Water Envir-
onments, Part I., I.H. Saffet, ed.
John WHey and Sons, New York, N.Y.
pp. 107-109.
65
-------
24. Stone, R. 1973. Sewage Sludge Disposal
in a Sanitary Landfill. EPA SW 71D.
U.S. Environmental Protection Agency,
Cincinnati, Ohio.
25. Sussman, D. 1977. More Disposal Oper-
ations Mixing Sewage Sludge and Mun-
icipal Solid Wastes. Solid Waste
Management Refuse Removal Journal,
Vol. 20, No. 8.
26. Walker, J.M., W.D. Burge, R.L. Chaney,
E. Epstein, and J.D. Menzies. 1975.
Trench Incorporation of Sewage Sludge
in Marginal Agricultural Land. EPA-
600/2-75-034. U.S. Environmental
Protection Technology Series.
66
-------
LEACHATE AND GAS FROM MUNICIPAL SOLID WASTE LANDFILL SIMULATORS
James J. Walsh
SCS Engineers
Covington, Kentucky
Riley N. Kinman
Janet I. Rickabaugh
University of Cincinnati
Cincinnati, Ohio
ABSTRACT
This study was initiated in 1974 to simulate the behavior of various types of land-
fills. The objective of the program was to determine the impact on solid waste decomposi-
tion from (1) varying moisture infiltration rates, (2) adding pH buffering compounds,
(3) prewetting the wastes, (4) varying ambient temperatures, and (5) codisposing refuse
with sewage sludge and various industrial wastes. To this end, a total of 19 landfill
simulators were constructed in late 1974 and early 1975. Each was loaded with approxi-
mately 3.0 metric tons of municipal refuse. Small quantities of sewage sludge,
industrial wastes, or other materials were added to 12 of these simulators. Maintenance,
operation, and monitoring of these landfill simulators was then begun and has continued
to this date. Water additions are made on all simulators on a prescribed schedule to
simulate rainfall conditions in the Midwest U.S. Gas and leachate are collected,
quantified, and analyzed. Refuse, soil, and ambient air temperatures are recorded.
This paper addresses data derived from the municipal-solid-waste-only test cells
over an approximate six year period from November 1974 to August 1980. Moisture balances
in each of these cells are quantified and evaluated. Six leachate analytical parameters
have been selected, plotted, and scrutinized. Landfill gases are also examined with
regard to their composition and-generation rate.
INTRODUCTION
As directed by the Resource Conserva-
tion and Recovery Act of 1976 (RCRA), an
inventory of all solid waste disposal sites
across the nation is now being conducted by
solid waste agencies in each of the 50
states. As part of the inventory process,
state regulatory officials are required to
both (1) identify all such disposal facili-
ties in their state, and (2) evaluate these
sites for compliance with EPA's "Criteria
for the Classification of Solid Waste
Disposal Facilities". These Criteria were
published in the September 13, 1979 Federal
Register, and require an assessment of the
environmental adequacy of these facilities
against each of eight criteria. Sites
found to be in compliance with the
Classification Criteria will be classified
as "sanitary landfills" and likely allowed
to continue operation. Sites found in
non-compliance will be classified as "open
dumps" and either closed immediately or
allowed to upgrade their operations to that
of compliance in accordance with a state-
imposed upgrading schedule not to exceed
five years.
Among the eight criteria, the one
likely to cause the most serious compliance
problem is the one dealing with ground
water. For while promulgation of the RCRA
Criteria is a large step forward in the
regulatory area, the technical state-of-the-
art for ground water protection undoubtedly
requires additional work. It is generally
conceded by most researchers that insuf-
ficient information is known in the areas
of solid waste decomposition and contaminant
67
-------
generation/migration, the very areas that
impact on ground water protection. If RCRA
regulations in the ground water sector are
to be enforced wisely, more will need to
be known on these subjects in order to
mandate appropriate upgrading actions
against existing "open dumps", and to
ensure appropriate design of future sani-
tary landfills.
This project was motivated by a need
for such information. A broad objective
of the program was to study solid waste
decomposition and contaminant release at
various types of landfills. Specific
objectives are to determine:
1. The effect of different water infiltra-
tions on solid waste decomposition.
2. The effect of sewage sludge additions
on solid waste decomposition.
3. The effect of pH buffer addition on
solid waste decomposition.
4. The effect of adding six selected
industrial wastes on solid waste
decomposition.
5. The effect of initial water addition
on solid waste decomposition.
6. The survivability of polio-virus in
landfills.
7. The effect of different air and
surrounding soil temperatures on
solid waste decomposition.
8. The ability to duplicate monitoring
data from two test cells constructed
and operated under similar conditions.
TEST CELL CONSTRUCTION
The project site for this study is the
U.S. Environmental Protection Agency (EPA)
Center Hill Laboratory in Cincinnati, Ohio.
A total of 19 test cells were constructed
for this program in November 1974 and
April 1975. Fifteen of these test cells
are located outdoors and below the ground
surface. These 15 are arranged in a
horseshoe alignment. The remaining four
test cells are located inside the high-bay
area of the Center Hill Laboratory
facility, and are abpve-ground, located
on a concrete slab. (See Figure 1)
Individual test cells consist of
steel tubes 1.83 m in diameter, 3.66 m
high, and 4.76 mm thick. Steel sidewalls
were coated with coal-tar epoxy as a
EXTERIOR TEST CELLS
INSTRUMENTATION
BUILDING
INTERIOR TEST CELLS
BUILDING WALL'
Figure 1. Test cell location plan
moisture and gas seal. The outside test
cells were placed on concrete slabs in an
excavated area._Soil was then backfilled
around the sidewalls of these 15 cells to
within 0.3 m of the top of each steel tube.
Several layers of fiberglass cloth were
placed inside each cell covering the
concrete base and extending 0.3 m up the
sidewall to further ensure a water-tight
and gas-tight seal. Interior test cells
were placed atop steel bases which were
welded onto the bottom of the test cell
tubes.
Provisions for leachate drainage were
installed in all cells. Exterior cells
had a small depression in the concrete
slab and connective piping to a leachate
collection well, located in the center of
the horseshoe layout of test cells. This
well serves as a central leachate collec-
tion point for all exterior cells and is
also used as a ground water drawdown to
prevent pressure and infiltration of
ground water into the test cells. Interior
cells are mounted on concrete blocks and
leachate drains attached below the cell
bottoms.
All test cells were then readied for
waste loading. To minimize the exposure
time and, thus more closely simulate actual
landfill conditions, all test cells were
completely loaded and their tops welded
shut within seven days of waste delivery.
Initially, a 15.2 cm thick layer of silica
gravel was applied at the bottom of each
cell. This layer serves as a reservoir
for leachate and prevents refuse from
clogging the bottom drain. Silica gravel
was selected for its inert nature to
68
-------
prevent any chemical reaction with the
collected leachate.
Refuse was then delivered to the site
and added to each test cell in eight 0.3 m
thick increments or "lifts". Each lift
was compacted with a wrecking ball.
Sludges and other materials added to
selected cells were applied at the top of
each lift in proportionate amounts (except
the first lift to avoid premature leaching).
Temperature probes were installed atop the
second, fourth, and sixth lifts in each
cell. Gas probes were installed atop the
second and sixth lifts.
Upon 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. Steel lids were then welded into
place at the top of each test cell,
providing an air- and water-tight seal.
A cross section of a typical test cell is
shown in Figure 2.
IJJL-
| L-
CAS PROBE
SETTLEMENT INDICATOR
SIGHT TUK
ATER
DISTRIBUTION RIN8
>.
TEMPERATURE PflOBE
•A3 PROBE
TEMPERATURE PROBE
-. -f
ITED INPUT CONNECTOH
iAS PROBE CONNECTOR
~MPERATURE PROBE
tt» • SILTY CLAT
C.4 •> MUNICIPAL REFUSE
O.ISB SILICA CRAVEL
"((CRETE SLM
LEACHATC ORAM
Figure 2. Cross-section of typical
test cell.
A summary of the program plan
covering all 19 test cells has been
included as Table 1. This tabulation
indicates the construction data, location,
additive type, and annual infiltration rate
for each of the 19 cells. As mentioned
previously and indicated in Table 1,
sludges, industrial waste, and other mate-
rials were added to selected cells. The
intent of these additions was to allow an
investigation of what effect codisposing
these materials with municipal refuse
would have on solid waste decomposition
processes. More specifically, this
arrangement provided test cells which ful-
filled the earlier objectives as follows:
1. Different water infiltration rates:
Test Cells 1, 2, 3, and 4.
2. Sewage sludge addition: Test Cells
5, 6, and 7.
3. pH buffer addition: Test Cell 8.
4. Six selected industrial sludge addi-
tions: Test Cells 9, 10, 12, 13, 14,
and 17.
5. Initial water addition: Test Cell 11.
6. Survivability of polio-virus: Test
Cell 15.
7. Different ambient air and soil tempera-
tures: Test Cells 2 and 16.
8. Experimental replication: Test Cells
16, 18, and 19.
TABLE 1. DESIGN, CONSTRUCTION, AND
OPERATIONAL FEATURES OF ALL TEST CELLS
Test
Cell
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Initial Loading Additive
„
..
..
Sewage Sludge*
Sewage Sludge*
Sewage Sludge*
Calcium Carbonate
Petroleum Sludge
Battery waste
Hater
Electroplating waste
Inorganic Pigment waste
Chlorine Prod. Brine Sludge
Polio-virus
._
Solvent Based Paint Sludge
--
~*
Annual
Infil-
tration
203 on
406 on
607 on
813 on
406 on
406 on
«06 on
406 on
406 on
406 on
406 on
406 on
406 on
406 on
406 on
406 on
406 on
406 on
406 on
Construction/
Lo ding Date
Ho
Location 19
Outside
Outside
Outside
Outside
Outside
Outside
Outside
Outside
Outside
Outside
Outside
Outside
Outside
Outside
Outside
Inside
Inside
April
4 1975
x
x
Inside x
Inside x
• Low, medium, and nigh quantities of sewage sludge were added for
Test Cells 5, 6, and 7. respectively.
69
-------
Before loading, two representative
samples were collected from each lift in
each cell for purposes of waste characteri-
zation, chemical analysis, and moisture
content determination. First, a total of
11 refuse categories were established and
the waste characterization sample from
each lift sorted in accordance with .these
categories. Subsequently, this data was
averaged across all eight lifts in each
cell to determine the waste composition
of each test cell. Results for the solid-
waste-only test cells investigated herein
are presented in Table 2. These results
indicate that the waste composition is
relatively consistent from cell to cell,
and in fact is truly representative of
residential/commercial solid waste in the
U.S. After completion of the sorting
exercise for each lift, the waste from
similar sort categories of all lifts and
test cells were combined and a sample
collected from each sort category for a
chemical analysis. The results of this
analysis are reflected in Table 3. Finally,
a moisture content determination was
performed on the sample from each test cell
lift which was extracted for this purpose.
The results of this determination were again
averaged across each test cell, yielding
some of the figures shown in Table 4.
As shown, moisture contents were found
to range from 32 to 35 percent on the
municipal-solid-waste-only test cells (Cells
2, 3, 4, 16, 18, and 19). Considering a
measured in-place wet weight of approxi-
mately 3,000 kg, the dry weight of these
cells was found to range from 1,946 to
2,042 kg. Considering a waste compartment
size of 6.41 m3, an initial refuse density
can be calculated. On a wet weight basis,
refuse densities were found to range
between 466 and 470 kg/m3. On a dry weight
basis, densities ranged between 304 and
319 kg/m3.
OPERATION AND MONITORING
To more closely simulate actual in-
field landfill conditions, all test cells
are operated in accordance with a strict
monthly schedule of moisture application.
This infiltration schedule has been
summarized in Table 5. As shown, Test
Cells 3 and 4 receive approximately 600 and
800 mm of annual infiltration, respectively.
The remaining test cells (Cells 2, 16, 18,
and 19) receive approximately 400 mm of
annual infiltration.
In order to simulate actual precipi-
tation/evaporation conditions in the
Midwest U.S., this annual rate is adminis-
tered on a periodic (monthly) basis, but
in non-equal quantities. As indicated, no
infiltration is applied in the usually dry
months of January, July, August, and
September. High amounts of infiltration
are applied in the normally wet months of
March, April, and May. Low-rate applica-
tions are made in the remaining transition
months.
Physical data recording, sampling, and
analysis are performed on all 19 test cells;
a summary of these activities on the
municipal-solid-waste-only cells has been
included in Table 6. To summarize this
schedule, gas volumes and refuse and air
temperatures are determined on interior
test cells each day. Refuse, soil, and air
temperatures on outside test cells are
determined bi-weekly. Gas composition is
determined for both interior and exterior
test cells monthly. Leachate samples are
collected each month, their volumes
recorded, and representative samples
prepared. These samples are then
analyzed for 22 parameters each month, with
an additional 12 parameters each quarter
and seven parameters every six months.
Analyses are performed at the Center Hill
TABLE 2. REFUSE CHARACTERIZATION FOR SELECTED TEST CELLS [2]
Test
Oil
2
3
i
16
IB
19
Food
Uaste
6.3
11.0
5.9
7.3
6.2
fi.5
Garden
Haste
21.5
30.2
22.6
16.3
9.6
11.9
Paper
f
41.5
34.9
37.8
41.3
46.2
43. ?
Plastic,
Rubber.
Leather
5.4
8.3
6.4
10.1
6.1
7.5
Textiles
4.5
3.3
4.1
3.8
3.3
2.7
Hood
2.2
0.3
2.0
0.8
3.9
1.9
Metal
8.5
8.6
5.9
8.9
9.3
7.<
Glass
9.9
5.9
8.8
6.0
8.4
10.0
Ash,
Rock,
01 rt
3.3
3.4
1.4
1.3
3.0
1.9
Diapers
1.2
2.3
3.2
2.7
2.9
3.0
Mnet
3.5
2.4
3.5
2.8
3.4
3.6
Average*
Std.De».*
8.1
2.2
15.7
5.3
41.8
4.5
6.9
1.8
4.1
1.5
1.8
1.0
8.3
1.4
7.8
1.4
2.9
1.3
2.4
1.0
3.2
0.7
• For all 19 test cells.
70
-------
TABLE 3. REFUSE CHEMICAL ANALYSIS FOR ALL TEST CELLS [3]
««•»—«
cost
ns»
LUld*
AM
Cnrfu (Ibor
Txul carton
iMrfjala cj|W«
OtlMklC • tt r«rt> nr
II n
L*.<1CD.
0.040
0.111
4.4|
1.31
11. 1
3.1)
U.4
7.W
3.40
4.1
1.40
l.t
4.1
0*3*0
in
10.1
177
11)
2.11*
4.01*
0.101*
J2.4
10.1
10.71
11.32
1.1
3.1*
p cm
•llllo*
nrtb.,.
0.710
0.111
1.11
2.21
14.0
U.7
44.)
0.740
41.1
*0. 1
<0.1
4.1
0.2)
141
0.1
l.M
7)0
171
lU
0.110*
4.14
1.11
111
4*. 11
1.1
1.47
br ei9ht basis (kg/m )
(.41
466
304
6.41
469
305
6.41
468
305
6.41
467
304
6.41
468
317
6.41
470
319
71
-------
TABLE 5. MONITORING PROGRAM FOR SELECTED TEST CELLS
Media
Refuse/Soil/Air
leachate
Determination
Volume
Temperature
Air Pressure
Tenperature
Air Pressure
Volume
Temperature
otfti organic ^J^o [ «*
PH
Oxidation Reduction Potential (ORP)
AIL 1 ' . tOnOU"artC*
n*- h V /nth on \
iron (Fe)
Sulfur (S)
Chloride (C1)
Mercury (Hg)
Beryllium (Be)
Selenium (Se)
Cyanide (CN)
Phenol
Arsenic (As)
Mexavalent Chromium (Cr6)
Nickel (Hi)
Asbestos
Fecal Colifonn
Fecal Streptococci
Beryllium (Be)
Titanium (Ti)
Vanadium (V)
Antimony (Sb)
Frequency
Daily
Bi-weekly
Bi-weekly
Dally
Dally
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
Monthly
n 1
Quarterly
Quarterly
yuane y
ftu t 1
Quarterly
Quarterly
-------
Laboratory using facilities and equipment
assigned specifically to this project.
Quality assurance program procedures are
utilized to ensure the validity of sampling
results. Analytical 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
As stated previously, leachate and
gas data and evaluations in this paper will
concentrate on those cells designated as
"municipal-solid-waste-only" (i.e., cells
having received no waste additives). The
first task then will be to evaluate the
relation between moisture added and
moisture leached in each of the six
subject test cells. Data on moisture
added and leached plotted as "interative
liters of water per kilogram of solid
waste versus time" is presented in Figures
3 through 8. As indicated, Test Cells 2,
16, 18, and 19 all receive approximately
400 mm of infiltration per year. Test
Cells 3 and 4 receive 600 and 800 mm per
year, and thus their plots are proportion-
ally higher.
Several observations can be made from
the data presented in these figures. First,
it is obvious that infiltration can be
added for a period of time without signifi-
cant quantities of leachate being generated.
Second, although leachate volumes increase
and decrease from month to month in
general accordance with the variation in
infiltration volumes added, leachate
volume changes appear to lag corresponding
infiltration changes by about one to two
months. Third, even after the onset of
leachate generation in measurable quantities,
infiltration continues to be added each
month at quantities in excess of subsequent
leachate volumes.
Infiltration and leachate volume data
on a cumulative basis is presented for
Test Cells 2, 3, 4, 16, 18, and 19 in
Figures 9 through 14, respectively. The
upper plot on each of these respective
curves reflects in liters/kilogram the
sum of cumulative moisture added (via
infiltration) plus the initial moisture
content (i.e., inherent in the waste as
delivered). The lower plot in each of
these figures reflects the cumulative
quantity of moisture leached. Again it is
obvious that the cumulative amount of
moisture leached is considerably less than
the moisture added for any point 1n time.
140
O.S3
cr
i
«. &3)
\ O.\l
INFILTRATIOM 2 4OO
r
900 too too 1200 uoo taoo zioo 7*00
TIM£ C0">vi SIWOE. CfiU. COMSTSOCTIoT)
Figure 3. Moisture added/leached at Test Cell 2.
73
-------
LOO -I
ioo
Z4OO
Figure 4. Moisture added/leached at Test Cell 3.
IMFH.TRKOOW a 800
INFIUTHATIOIO ADDED
KlOliTURC L£A04£b
Figure 5. Moisture added/leached at Test Cell 4.
-------
a«-
ftSO-
5
2
INRUTRKTIOM 3 400 WA/
tioc '5oo igoo Iioo 14oo
Figure 6. Moisture added/leached at Test Cell 16.
I
I.M
(J.fcT-
050
INFICTRATIOM « 400
300 MO 900 [loo i5oo 1800 Ztoo tAao
TIME £o*VS SIMC£ CCU. COMSTCUCTIOW^
Figure 1, Moisture added/leached at Test Cell 18.
-------
1.00
0.83
§ 0.47 -
I
IMFILTRATIOU S 400 MM/Yff
IWFILTKATlON KDO6D
x-'
/T^W^
300 (00 »00 000 1300 "BOO tlOO
ceu.
Figure 8. Moisture added/leached at Test Cell 19.
CUMUUkTiv/t
o Soo too vra ixoo woo laoo zioo
TlMt (D*vVS SIMCt CEU. O3N6T1?UC.TIOM)
Figure 9. Moisture balance for Test Cell 2.
76
-------
7.8
INRL.TRA.T10M 3 60O MM/YI?
JOO WO too 1100 ISOO I8OO tlOO
CUMULATIVE
LHACMID
Figure 10. Moisture balance for Test Cell 3.
7.8 ,
I JOO WO 900 1100 ISOO 1900 tlOO
Tine (HAYS SIMCC. ceu. a)>aaTl^JCTlOM')
Figure 11. Moisture balance for Test Cell 4.
77
-------
i
s
0
joo MO «x» 1100 tseo woo tioo
TIME (DAVS siNce CEAO,
Figure 12. Moisture balance for Test Cell 16.
7.8
t
I
^ M
tu
S
1.3
o
o
INFH.TUATIOM a 40O MW/VR
300
TIME
*OO I2OO ISOO 1*00 2IOO
SIMCC cett
Figure 13. Moisture balance for Test Cell 18.
78
-------
CUMULATIVE WOliTUCt
UCACHCb
JOO MO *X> 1100 1300 1800 HOO
TIME. (c*Y5 SINC6 CfiU. OOUSTRUCTIOM^
Figure 14. Moisture balance for Test Cell 19.
This, of course, is due to the ability of
refuse to retain a certain quantity of
moisture. Generally, if the initial
moisture of the refuse plus the moisture
added does not exceed this retention
ability, leachate will not appear. This
quantity is known as the field capacity
and is technically defined as "the
maximum moisture content which a soil (or
solid waste) can retain in a gravitational
field without producing continuous down-
ward percolation". [4] Generally, the
release of significant quantities of
leachate marks the initiation of this
continuous leaching and the point at which
field capacity is attained. Of course,
some leachate may be emitted before this
point; but generally these are small
quantities and can be followed by times
at which no leachate is emitted despite
the addition of infiltration water.
As shown in Figures 9 through 14,
field capacities were attained at times
ranging from 339 to 540 days for the 400
mm/year test cells (i.e., Cells 2, 16, 18,
and 19). Although it receives a higher
infiltration rate of 600 mm/year, Test
Cell 3 also reached field capacity within
this time range at 480 days. Test Cell 4
initiated leaching at 150 days, more
appropriate considering its much higher
infiltration rate of 800 mm/year.
Moisture conditions present in each of
the above described test cells at the onset
of continuous leaching are detailed in
Table 7. The equation used to define the
water balance in each cell at that time is:
Moisture Retained = (Initial Moisture) +
(Moisture Added) - (Moisture Leached)
As indicated in Table 7, the moisture
retained in each cell at the point of
initial continuous leaching (i.e.,
attainment of field capacity) was found to
exist within a fairly tight range of 1.02
to 1.31 I/kg for all but Test Cell 18 which
appears to have initially achieved field
capacity at 0.85 I/kg. (Note that all mass
values from this point on are on a dry-
weight basis.) From this data it is appar-
ent that attainment of field capacity is
dependent upon the moisture retained (ex-
pressed as liters of moisture retained per
kilogram of dry solid waste); the time it
takes to reach field capacity does not
appear to have any direct relation to
either the moisture application rate or
initial moisture content by themselves.
Presumably, after field capacity has
been reached, one liter of moisture should
be leached for each one liter of moisture
added. Thus, the moisture retained should
stabilize after the onset of continuous
79
-------
TABLE 7. MOISTURE BALANCE SUMMARY
Moisture to Conpnuoi
Initial Moisture
Test
Cell
2
3
<
16
18
19
Annual
or.
406
607
813
406
406
406
Infiltration
1/*Q
0.55
0.82
1.09
1.10
1.05
1.05
No. Of
Days
540
480
ISO
476
339
395
MoiSturt
(l/k9)
0.54
0.54
0.54
0.54
0.47
0.46
Added
(l/kg)
0.49
0.80
0.70
0.65
0.38
0.62
Moisture
LeacneO
(I/kg)
0.01
0.03
0.05
0.01
0.00
0.00
Moisture
Retained
(l/kg)
1.02
1.31
1.19
1.18
0.65
1.10
Moisture to
Initial Hoisture
No. Of
Days
2.099
2.099
2,099
2.099
1.962
1.962
Moisture
(I/kg)
0.54.
O.S4
0.54
O.S4
0.47
0.48
Added
(I/kg)
3.69
5.52
7.37
3.24
2.87
2.85
Date
Moisture
Leached
(I/kg)
2.32
3.36
5.39
2.13
1.80
1.90
Moisture
Retained
(I/kg)
1.91
2.70
2.52
1.65
1.54
1.43
leaching. However, this has not been the
case. As shown in Figures 9 through 14,
the gap between "initial moisture plus
cumulative moisture added" and "cumulative
moisture leached" has stabilized somewhat
with the onset of initial continuous
leaching. However, the size of this gap
(or the moisture retained) appears to be
growing slightly larger with time. This
gap is quantified in Table 6. As shown,
the six solid-waste-only test cells now
retain between 1.43 and 2.70 I/kg.
These figures demonstrate a substantial
increase for Test Cells 3 and 4. While the
moisture retained in Test Cells 2, 16, 18,
and 19 remains below 2.00 I/kg, measurable
increases in the moisture retained have
also occurred in these cells.
The reason for these increases cannot
be fully explained. One possible reason
may be the effect of channeling.
Channeling exists when moisture added to
a waste percolates directly through it,
failing to fill all the voids before
leaching. Some voids remaining in the
waste at the onset of continuous leaching
may now gradually be filling with moisture,
rather than letting the infiltration water
channel directly through. Leaking test
cell drains for the. exterior lysimeters
is another possible explanation, particu-
larly for Test Cells 3 and 4 considering
their high retained moisture figures.
While no drain line leaks have been
witnessed for these two test cells in the
past two years, visible leaks have existed
for some of the other test cells in this
program. Because Cells 1 to 15 are all
located outside, below the ground surface,
some leaking is possible for all test
cells, and it might go undetected. The
effect of such leaking would allow leachate
quantities to escape unmeasured, decreasing
the measured quantity of moisture leached,
and thus increasing the measured quantity
of moisture retained to a figure higher
than is actually the case.
The above explanation may be plausible
for Test Cells 3 and 4. However, no leaks
have ever been observed for the interior
Test Cells 16, 18, and 19. Increased
moisture retentions for these test cells
thus may have to be explained by the
channeling theory described above. Alter-
natively, it has been theorized that some
moisture may be escaping as condensate in
daily gas volumes collected and vented.
However, an investigation was recently
conducted to quantify the moisture content
in the gas and to determine its impact
on the overall moisture balance. This
impact was determined to be insignificant
at less than one percent.
LEACHATE QUALITY EVALUATION
Of the numerous leachate analytical
parameters determined on this study, five
have been selected for indepth investiga-
tion in this presentation. In addition,
only data from Test Cell 2 (400 mm/year
infiltration), Test Cell 3 (600 mm/year
infiltration), and Test Cell 4 (800 mm/year
infiltration) will be addressed in this
section. Test Cells 16, 18, and 19 all
receive the same amount of infiltration and
it was the intent of this section to de-
velop trends as related to varying
infiltration rates.
It is appropriate to note that graphs
included herein are plots of mass leached
(cumulative kilograms of contaminant per
kilogram of dry solid waste) as a function
of leachate volume (cumulative liters of
leachate/kilogram of dry solid waste).
Plots of mass leached vs time were consi-
dered, since these might be more under-
80
-------
standable to the designer. However,
histograms collected to this end were
observed to be somewhat irregular due to
the seasonal fluctuations in applied
moisture. As a result, trends demonstra-
ted in the ensuing leachate plots were not
as readily discernable on the histograms.
The effect of plotting contaminant masses
leached against cumulative leachate
volumes has the effect of smoothing out
these fluctuations, normalizing them for
irregularities in applied moisture. To
make the plots in this section more useful
as a design aid, it is suggested that
the reader consult Figures 9 through 14
to determine the expected leachate volume
for a given infiltration rate and time.
Once expected leachate volume is computed,
contaminant release can be determined from
the ensuing plots in this section.
Total Organic Carbon
The release of TOC appears to be
approaching an asymptote (Figure 15) at
which only nominal amounts of TOC will
subsequently be released. This trend is
most apparent for Test Cell 4 (800 mm/year
infiltration) which has yielded the
greatest cumulative leachate volume to date.
The curve for the 400 mm/year test cell
exhibits a slightly higher trend and thus,
greater TOC released for a given leachate
volume. This tendency for a greater TOC
release rate with a lower infiltration rate
is not supported by data from the 600 mm/
year test cell however, as the mass
release trending for this test cell is
the lowest of the three.
Despite these differences, the overall
trending of these three TOC curves is quite
similar, with the cumulative TOC leached for
a cumulative leachate volume being somewhat
the same regardless of the infiltration
rate. Thus, the effect of a higher infil-
tration rate is merely to accelerate the
release process with time. Ultimately, it
would appear, each cell will have similar
overall TOC release. As shown in Table 3,
the TOC in the solid waste when loaded
in the test cells was determined to be
about 22 percent on a dry-weight basis.
Since approximately 40 g/kg of TOC has been
leached by the 800 mm/year test cell to
date, roughly 19 percent of the initial
TOC has been leached by the 800 mm/year
cell to this point in the project. This
percentage leached is illustrated on the
right vertical axis in Figure 15. Since
the overal1 trend of the TOC curve appears
to be asymptotic, it could be presumed
that ultimate TOC leaching would not be
much larger than this 19 percent value.
Much of the balance of the initial TOC may
remain in the solid waste, never made
available for leaching. Of course, it
should be pointed out that not all organic
a
s
a
1
tf)
^ I
o
i
• 12 <0
3
800 MM/V8
400 HM/YR
4<30 MM/YR
i.o io *,o 3.0
LCACHED (f, Jt HtO/k) SOUO
ka
Figure 15. Mass leached: TOC.
81
-------
carbon present in the waste at the time of
loading is released in a liquid form; some
of the organic carbon can be converted via
microorganisms into gases such as carbon
dioxide and methane.
A summary of masses leached to date
(for TOC as well as other contaminants) has
been included in Table 8. As shown, the
800, 600, and 400 mm/year cells have
leached a cumulative total of 40, 30, and
23 g/kg (respectively) to date. These
figures represented about 19, 14, and 11
percent of the TOC content found to exist
in these cells at the time of initial
loading.
A summary of concentrations for the
various leachate contaminants addressed in
this paper has been included in Table 9.
As indicated, TOC concentrations attained
maximum levels of between 31,000 and 37,000
mg/1 early in the project. Since that~time,
TOC concentrations have gradually decreased
to levels now between 1,000 and 2,000 mg/1.
TOC concentrations are currently as low as
1,000 mg/1 only for the 800 mm/year test
cell, indicating that it has matured
faster than the two lower infiltration
rate cells, both having TOC concentrations
no lower than 2,000 mg/1.
Chemical Oxygen Demand
COD removal (Figure 16) for the 400
rim/year cell exhibits the highest trajec-
tory, although it correlates very closely
to the trajectory for the 800 mm/year cell.
The trend for the 600 mm/year cell is again
lower than these two other cells and at
times veers substantially o.ff the trajec-
tory established by the 800 and 400 mm/year
cells.
Nonetheless, all three cells generally
follow a similar asymptotic trending, and
the slope for each of the curves can be
said to be similar to that found for the
TOC curve. In general, however, it appears
that the COD curve is still climbing at a
faster rate than that for TOC. This is to
be expected since initially TOC is one of
the primary contributors to the COD deter-
mination. However, inorganics also contri-
bute to COD and as TOC tapers off in the
future, inorganics may become the driving
force behind COD in the leachate. The
cyclical nature of the leachate pH (as
indicated in plots contained in earlier
papers on this project) is especially con-
ducive to the continued appearance of COD
in leachate. When pH values dip low and
carbon dioxide is available in the absence
TABLE 8. SUMMARY OF CONTAMINANT MASSES LEACHED
Mass Leached to Date Mass Leached to Date
(cunulatl»e gAg of solid «astel (as I In Initial solid waste)
«00 mm/600 mm/800 mm/4,00 mm/600 mm/ 800 mm/
Parameter year* year* yearl year* year* year!
TOC
coo
rm
TS
Fe
Ni
• Test
« Test
1 Test
Cell
Cell
Cell
0
2.
•3.
4.
?3
73
1.9
56
0.8
.0013
30
BO
1
SB
1
.7
.3
0.0013
40
95
2.3
57
1.4
0.0019
111
N/A
761
61
91
121
141
N/»
661
71
161
121
19:
N/A
901
81
171
IB:
N/A • Not applicable.
TABLE 9. SUMMARY OF CONTAMINANT CONCENTRATIONS IN LEACHATE
Maximum Approximate Concentration to Date
(rnq/1)
Minimum Approximate Concentration to Date
(mg/1)
Parameter
400 mm/
year*
600 t
year*
BOO i
yeart
400 mm/
year*
600
year
<\r 1100 mm/
year*
TOC
:OD
TKN
TS
re
K\
31,000
78.000
2,100
64.000
650
2.6
37,000
64,000
1.600
46,000
1,000
2.0
32,000
73.000
1,700
56.000
500
1.9
37,000
76.000
2,400
64,000
1,000
2.6
2,000
13,000
200
7,000
200
0.1
Z.OOO
4.000
200
7,000
50
0.1
1.000
1.000
100
3,000
100
0.1
1.000
1,000
100
3.000
50
0.1
• Test Cell 2.
• Test Cell 3.
f Test Cell 4.
*• Extremes for three test cells combined.
82
-------
5 iio
•?
a
I "
u
M
10 3.0 44 to
UKACHCD Oc { HtO/kj SOLJO
Figure '16. Mass leached: COD.
fc.O
of oxygen, ferrous iron and other similar
inorganic contributors to COD can be
created and held in solution, soon to be
manifested in the leachate.
Table 8 indicates, that the 800, 600,
and 400 mm/year cells have leached a total
COD mass of about 95, 80, and 73 g/kg of
solid waste to date. COD concentrations
(Table 9) initially ranged between 31,000
and 37,000 mg/1 for the three cells, but
have gradually decreased to a range of
1,000 to 2,000 mg/1. Again, recent COD
concentrations reflect the maturity of each
cell's decomposition process, with current
COD levels running at 1,000, 4,000, and
13,000 mg/1 for the 800, 600, and 400
mm/year cells, respectively.
Total Kjeldahl Nitrogen
TKN leached by Test Cells 2, 3, and
4 has been plotted in Figure 17. Again the
release exhibits an asymptotic trending .for
all three infiltration rates. The TKN plots
for different cells cross paths for the
first time; nonetheless, the 400 mm/year
cell continues to have the highest overall
trending, again followed in order by the
800 and 600 mm/year test cells.
Overall the deviation among the three
cells appears to be greater than that found
for TOC or COD. In addition, as shown on
the right axis of this plot, from 68 to 90
percent of the TKN has been calculated to
have been released in the leachate to date.
This value is much higher than that seen in
the release of other parameters, and is
expected to be so since conditions are
ideal early in solid waste decomposition
for biological denitrification as a result
of high temperature, acidic pH, and the
presence of organics. Of course, whether
up to 90 percent has actually been
released is highly questionable. It is
possible that the initial TKN reading in
the solid waste is erroneous due to faulty
analysis or compilation of a solid waste
sample that was not truly representative
of the waste mass at large.
As per Table 8, a total of 2.3, 1.7,
and 1.9 grams of TKN per kg of solid waste
has been leached by the 800, 600, and 400
mm/year test cells. Thus, for the first
time a greater contaminant mass has been
leached by a test cell receiving less
infiltration (i.e., the 400 mm/year cell
has emitted 1.9 g/kg versus only 1.7 g/kg
for the 600 mm/year test cell). Since the
400 mm/year cell has emitted less leachate
to date, obviously TKN concentrations in
its leachate must be relatively high. This
is demonstrated by the TKN concentrations
reflected in Table 9. As indicated,
initial concentrations of TKN were as high
as 2,400 mg/1 for the 400 mm/year cell,
83
-------
£
$ 14
O
§ *»\
p
a
UI
u
U
as
<
s*
V)
n
I
.e yo *c 5.0
UEMCHCD (S 5 HsO/k^ SOUO VJA<
Figure 17. I'tess leached: TKN.
versus only 1,600 and 1,700 mg/1 for the
600 and 800 mm/year cells. Currently,
however, TKN concentrations in leachate
from the 400 mm/year cell have returned to
the range of TKN concentrations from other
cells (i.e., 200 mg/1 versus 200 and 100
mg/1 for the 600 and 800 mm/year cells).
Total Solids
Total solids released (Figure 18) also
reveals an asymptotic trending. As for all
previous plots, the 400 mm/year test cell
is the highest followed in order by the
800 and 600 mm/year cells. The curve for
the 400 mm/year test cell revealed a
slightly greater divergence,'-with the 800
and 400 mm/year cells showing much closer
correlation. Nevertheless, the overall
trend for all three cells is similar; the
asymptotic trending of these curves
indicate that solids released is decreasing
with time. As shown previously in Table 4,
approximately 2,000 kg of dry solid waste
were placed in each of the three cells
being investigated herein. Thus, for the
ue>o«.D (jf. Jf
Figure 18. Mass leached: Total Solids.
84
-------
800 mm/year test cell, approximately 8
percent of the available solids in that
test cell have been leached to date. The
asymptotic trending of the curve would
indicate that ultimate release of solids
with time would not be much greater than
this amount.
Table 8 indicates that a total of 57,
48, and 46 g of total solids have been
leached for each kilogram of dry solid
waste for the 800, 600, and 400 mm/year
cells. These quantities represent about
8, 7, and 6 percent respectively of the
initial dry weight of solid waste loaded
into the test cells. Total solids con-
centrations (Table 9) have gradually
decreased with time. Initial solids
concentrations ranged between 46,000 and
64,000 mg/1. Although the total solids
leached to date was demonstrated to be
greater for the 600 mm/year cell than the
400 mm/year cell, these figures were shown
to be quite close. This fact may be
reflected in the much higher initial
concentrations of total solids as
indicated in Table 9 for the 400 mm/year
cell. Currently, total solids concentra-
tions range between 3,000 and 7,000 mg/1.
The mature stage of solid waste decomposi-
tion for the 800 mm/year cell is reflected
by its solids concentration being in the
lower part of this range.
Iron
Unlike previous parameters which have
exhibited an asymptotic curve for mass
released, iron release appears to be some-
what linear with time (Figure 19). As per
previous curves, the 400 mm/year cell
retains the highest trajectory. This time,
however, the 800 mm/year cell retains the
lowest trajectory (unlike the middle tra-
jectory it held in previous plots). Lastly,
although the three plots appear to correlate
well over the first half of the monitoring
period, recent trends have shown a greater
divergence between the 600 and 800 mm/year
cells. In fact, total iron mass released
by the 600 mm/year cell is only slightly
less than that for the 800 mm/year cell
despite the fact that its leachate volume
is only about two-thirds of that for the
800 mm/year eel 1.
As demonstrated in Table 8, total mass
leached for these three cells is 1.4, 1.3,
and 0.8 for the 800, 600, and 400 mm/year
test cells. . Based on the determination of
iron content in the solid waste made at the
onset of the project, these quantities
amount to 17, 16, and 9 percent of the
total iron in the wastes at that time. Due
to the linear trending of all three plots,
however, it would appear that iron will
continue to leach at concentrations found
1.10
g
i o*
Mfl/YR
17
•!
ii g
„ 1
2
1
7
5
3
i
o
O
!
i
Figure 19. Mass leached: Iron.
85
-------
in the past. Thus, it would appear that
the total iron mass leached will ultimately
run much higher than the 17 percent found
to date. Maximum iron concentrations in
leachate (Table 9) have ranged between 500
and 1,000 mg/1 for the three test cells.
The 1,000 mg/1 concentration found for
iron in the 600 mm/year cell is much
higher than the maximum levels encountered
in the other two cells and may be a
partial explanation for its relatively
high trajectory. Minimum iron concentra-
tions in the leachate from these cells have
ranged from 50 to 200 tng/1. Unlike for
previous parameters it should be noted that
there has been no gradual decrease in iron
concentration with time.
Nickel
Mass leached for nickel is demonstra-
ted in Figure 20. As for most of the
previous plots, an asymptotic trending is
observed. Again, the high, middle, and
low trajectories represented the 400, 800,
and 600 inn/year cells, respectively. As
can be seen in Figure 20, recent trends
have caused the 400 mm/year test cell to
dip below the trajectory for the 800
mm/year test cell.
As shown in Table 8, nickel mass
leached to date has ranged from 0.0013
to 0.0019 g/kg of solid waste. The .total
nickel mass leached for the 400 and 600
mm/year cells appears to be essentially
the same despite that the 400 mm/year cell
has leached only two-thirds the volume
leached by the 600 mm/year cell. To date,
18 percent of the nickel found in the
initial solid waste has been leached by
the 800 mm/year cell, while about 12 per-
cent has been leached for each of the 400
and 600 mm/year cells.
As would be expected from an asympto-
tic trending, nickel concentrations have
gradually decreased with time. As reflec-
ted in Table 9, early concentrations
ranged from 1.9 to 2.6 mg/1. A 2.6 mg/1
maximum occurred in the 400 mm/year cell
and may be a partial explanation for its
initially high trajectory on a mass leached
basis (Figure 20). However, nickel levels
are now relatively low and essentially the
same for all three test cells at about
0.1 mg/1.
TEMPERATURE EVALUATION
As stated previously, refuse tempera-
tures are recorded on a regular basis for
all test cells. Temperature plots for
two selected cells have been included in
Figure 21. Note that Test Cell 2 is an
exterior cell placed below the ground
surface, while Test Cell 19 is an interior
cell placed above ground and maintained at
room temperature.
Results for both cells demonstrate increased
temperatures soon after construction. This
trend is more obvious for the interior test
cell with its highest temperatures reaching
e
VI
O
3
0.0016 •
2 O.OOIZ -
Q 0.0006 -
0.0004-
400 MM/YR
400 KtA/YR
1.0
Z.O
3.0
4.0
3.0
to
Figure 20, Mass leached: Nickel
86
-------
300 too wo itoo 1500 1000 iioo 2400
TIME ([O*YS SINCE ceu. cowaTRUcnon^
Figure 21. Refuse temperatures.
about 27°C within 50 to 100 days after
initial loading. Thereafter, temperatures
average about 20°C, although they are
subject to large fluctuations. From obser-
vation, these fluctuations appear to be
somewhat seasonal with highest temperatures
occurring in the wintertime. This can be
explained by their placement at a higher
location within the Center Hill Laboratory's
high-bay area and exposure to warmer drafts
from the heating system than is experienced
in the summer.
The exterior test cell also appears
to have climbed in temperature shortly
after placement but the peak was reached
within 25 to 50 days at only 15°C. There-
after, temperatures appear to drop to about
8°C and may be partially explained by its
location below the ground surface and its
construction date of November 1974.
Because of its relatively small mass,
refuse temperatures are readily affected
by soil temperatures. Also, since it was
constructed in November 1974, It was
shortly thereafter exposed to the cold
wintertime temperatures. Thereafter,
temperatures appear to range between 4 and
18°C. Again, a seasonal fluctuation is
noted; unlike the Interior test cells,
however, temperature lows occur in the
winter, temperature highs in the summer.
GAS EVALUATION
Gas samples are extracted from all
test cells on a monthly basis for composi-
tional analysis. In addition, gas quantity
data is recorded for interior Test Cells
16, 18, and 19 on a daily basis. Because
both quality and quantity data is available
for only the interior cells, our attention
here will be redirected to these three
cells.
Plots of gas composition vs time for
Cells 16, 18, and 19 have been included as
Figures 22, 23, and 24, respectively. As
shown, compositional data is available for
each of these cells for periods ranging
from 5 1/4 years (for Test Cells 18 and 19)
to 5 3/4 years (for Test Cell 16).
Carbon dioxide is the predominant gas
created in the early phases of solid waste
decomposition. Theoretically, carbon
dioxide is generated as the predominant gas
during both aerobic decomposition and the
facultative (or acid-forming) phase of
anaerobic decomposition. Later, as the
methanogenic phase of anaerobic decompo-
sition is established, methane becomes the
predominant gas although carbon dioxide
concentrations should continue to comprise
from one-third to one-half of the gas volume.
87
-------
joo *oo 900 neo i»oo BOO 1100
TIME (DAYS SINCE, ceu. COMSTRUCTIOM')
Figure 22. Gas composition at Test Cell 16.
aoo too 100 ttoo isoo (Boo
TIME COA.YS siwct
Figure 23. Gas composition at Test Cell 18.
8S
-------
88-
c
J>
&
to-
• IB
TIME (DAYS SINCS. ceu. COMSTROC.TIOM'')
Figure 24. Gas composition at Test Cell 19.
2
0
I-
5i
The plots contained 1n Figures 22, 23,
and 24 generally support these theories.
Highest carbon dioxide readings were
recorded quite early in the decomposition
process with readings attaining between
60 and 90 percent for all test cells within
1 to i 1/2 years after construction.
Thereafter, carbon dioxide concentrations
appear to decrease as methane concentra-
tions increase. However, carbon dioxide
concentrations of 40 to 60 percent are
generally still maintained in all three
test cells, even five years after cell
construction.
As was anticipated, methane concentrations
were initially low, averaging no more than
five percent over the first two to three
years of the study period. After about
2 1/2 years, however, methane concentra-
tions were seen to steadily increase in
all test cells and have how attained
readings between 50 and 60 percent in all
three cells. Typical atmospheric gases
(nitrogen and oxygen) consume the balance
of gas compositions and were found to
contribute up to 80 percent of the gas
through the first year before declining
steadily to levels between 0 and 20 per-
cent. Other gases (e.g., hydrogen) were
also detected over the time period shown,
but were in trace concentrations of less
than one percent and, therefore, not shown.
Overall, 1t appears that the trends noted
in this study for gas composition have
generally followed that specified by
decomposition theory. The trends shown in
Figures 22 through 24 appear to generally
follow the theoretical gas production
patterns reported by Rovers and Farquhar,[5]
Plots of gas quantity with time (ml
gas/kg of solid waste/day) for Cells 16,
18, and 19, have been included as Figures
25, 26, and 27, respectively. First, it
should be noted that the gas volumes shown
are not necessarily for a precise 24-hour
period and, in some cases may represent
anywhere from 12 to 48 hours. Nevertheless,
these plots provide good data on the over-
all range and trending of gas generation to
date. As shown, gas volumes for all three
cells appear to range from 0.5 to 2.0 ml/
kg/day for the first 3 1/2 years. There-
after, daily gas volumes have increased
substantially, ranging between 1 and 50
ml/kg/day. Although the ranges of gas
generation are similar for all three cells,
average gas generation rates are not.
Cells 16 and 18 have had average generation
rates of about 25 ml/kg/day. Average rates
1n Cell 19 are considerably lower at about
7 ml/kg/day. The dramatic increase 1n
daily gas volume for all three cells at the
1,200 day point 1s somewhat unexplained;
It should be pointed out that a transition
in sampling personnel was made at that
time and this transition may be wholly
accountable for the increased volumes
recorded. Nonetheless, we have no evidence
to suggest that a difference in sampling
methodology is involved.
89
-------
to
4
P
I
o
i
I
40
16 •
soo too wo 1200 isoo ieoo zioo 1400
TIME ^Wkvs siwce ceu. "
Figure 25. Daily gas volumes at Test Cell 16.
too too «eo ttoo isoo isoo too ttoo
Figure 26. Daily gas volumes at Test Cell 18.
90
-------
tt-i
—-^^ytM^WSw^^^u /
a too uo too itoo 1100 IMP two »«o
TIME 0>**s siKice. ceu. COUSTRUCTION')
Figure 27. Daily gas volumes at Test Cell 19.
Plots of cumulative gas generation to
date for each of these three test cells are
contained in Figures 28, 29, and 30. Over-
all trendings for cumulative gas generated
in Cells 16, and 18 are almost identical.
Total gas volume achieved is about 18 I/kg
of solid waste over 5 3/4 years for Test
Cell 16, and about 14 I/kg over 5 1/4 years
for Test Cell 18, The overall trending
and total cumulative gas volume encountered
in Test Cell 19 is considerably less than
that for the other cells. This, of course,
reflects the lower daily average rates
seen in Figure 27, Total cumulative gas
volume generated by this Cell was calculated
at about 3 I/kg in the 5 1/4 years since
its construction.
is
too too noo 1900 IMO *ioo t4«o
TIME
-------
O SOO MO 900 ItOO 1900 ItOO tlOO 14OO
TIME (DA-TS SINCE ceu. CONSTTTOCTION")
Figure 29. Total gas generation at Test Cell 18.
§
\n
o 900 too too itoo isoo 1000 xioo t4oo
TIME (t*vi SINCE CELL COMSTROCTIOM")
Figure 30. Total gas generation at Test Cell 19.
92
-------
ACKNOWLEDGEMENTS
The work upon which this paper is
based was performed pursuant to Contract
No. 68-03-2758 with the U.S. Environmental
Protection Agency. The authors would like
to express their appreciation to Mr. Dirk
R. Brunner, Project Officer, for his
assistance.
REFERENCES
1. Swartzbaugh, Joseph T.; Robert C.
Hentrich; Gretchen Sabel. "Evaluation
of Landfilled Municipal and Selected
Industrial Solid Wastes." U.S. EPA
Contract No. 68-03-2120. June 1977.
p. 15.
2. Streng, D.R. "The Effects of the
Disposal of Industrial Waste Within a
Sanitary Landfill Environment."
Residual Management by Land Disposal,
Proceedings of the Hazardous Waste
Research Symposium. February 2-4,
1976. pp. 67-68.
3. Walsh, J.J.; R.N. Kinman. "Leachate
and Gas Production under Controlled
Moisture Conditions." Proceedings of
the Fifth Annual Research Symposium
on Land Disposal of Solid Wastes.
March 26-28, 1979. p. 45.
4. Fenn, Dennis G.; Keith J. Hanley;
Truett V. DeGeare. "Use of the
Water Balance Method for Predicting
Leachate Generation from Solid Waste
Disposal Sites." October 1975. p. 4.
5. Rovers, F.A.: G.J. Farquhar. "Gas
Production During Refuse Decomposition."
Water, Air and Soil Pollution. 2-483.
1973.
-------
Gas Production in Municipal Waste Test Cells
Richard A. Shafer, Robert J. Larson, Philip G. Malone, and Larry W. Jones
U. S. Army Engineer Waterways Experincnt Station, Vicksburg, Mississippi 39180
ABSTRACT
Data on landfill gas volumes and gas composition have been collected from four municipal
solid waste test cells for 636 days. The average gas production for the four cells for the
636-day period was 18.32 ml/kg (dry wt) of refuse/day. Gas production averaged over 636
days was only 30% of the rate recorded for the first 100 days. First hydrogen and then
methane was detected in the decomposition gases from each cell. Methane concentrations
as high as 5.8% have been recorded. The larger test cells used in this study are pro-
ducing 10% more gas per kg refuse than the smaller cells. The higher gas production is
thought to be due to a higher operating temperature in the larger cells.
INTRODUCTION
Landfills can be considered uncontroll-
ed, unmixed, anerobic digesters. Work that
has been done on landfill gas generation
both in the laboratory and in the field
indicates that many factors such as refuse
composition, moisture content, degree of
compaction, pli, temperature, nutrient, and
metal content can affect the rate of decom-
position and gas production in a landfill
(1,6). Landfill gas production is also
controlled by changes in the microbial
flora that occur as oxygen is depleted in
the decomposing refuse. All of the factors
listed above can be thought of as affect-
ing or controlling the microbial activity
that occurs in the test cells. The
sequence of bacterial communities may be
the most critical factor related to gas
production. Most organisms involved in
anaerobic decomposition with gas genera-
tion operate in an interdependent fashion
with one organism using the metabolic pro-
ducts of another for essential nutrients
(8).
The present study was developed to
verify gas production data obtained in
other refuse decomposition studies and to
examine problems related to scaling of
large laboratory simulation tanks. Data
are presented on gas composition, and gas
volumes for the first 636 days of opera-
tion. The data are compared with results
from other published studies.
METHODS AND MATERIAL
The experimental setup and analytical
procedures used in this investigation are
described in detail in Myers and others (7).
Two different sized gas-tight steel test
cells were used. The larger tanks have an
inside diameter of 1.83 m and a height of
3.66 m. The smaller tanks have an inside
diameter of 0.91 m and a height of 1.83 m.
The tanks were packed with unbailed, un-
shredded municipal waste from residential
collection routes in Warren County, Missis-
sippi. The composition of the refuse is
given in Figure 1. Details of the loading
and operation of the test cells are given
in Table 1. Cross-sections of tanks are
shown in Figure 2. Infiltration of precipi-
tation was simulated by adding 1.27 cm of
deionized water to each;tank ,per..vefik. This
mimics the infiltration that could occur in
a landfill in the humid eastern U.S. Lea-
chate is drained from the tanks and sampled
at monthly intervals.
An automatic gas measuring system re-
corded the volume of gas produced and the
temperature and barometric pressure at in-
tervals throughout the 636-day experiment
(Fig. 3). Details of the measuring and
data reduction system are given in Myers
94
-------
COMPOSITION OF MSW
GARDEN
WASTES 0.5%
GLASS 8%
TEXTILES 3%
DIAPERS 3%
PLASTIC,
RUBBER, ETC. 9%
0.5%
FOOD WASTES 1%
PERCENTAGE BASED ON DRY WEIGHT
Figure 1. Composition of refuse.
TABLE 1. SUMMARY OF DATA ON TEST CELLS
Inside diameter (m)
Height (m)
Wee weight of refuse (kg)
Dry weight of refuse (kg)
Density of refuse (kg/m-3)
Cover Material
Underlying soil
Surface area of refuse (m^)
Weekly moisture addition (cm)
Volume of water added at (1)
10
0.91
1.83
322
264
405
gravel
clayey sand
0.65
1.27
8.3
Test
15
1.83
3.66
2555
2096
400
gravel
clayed sand
2.63
1.27
33.4
Cells
16
1.83
3.66
2555
2096
400
gravel
clayey sand
2.63
1.27
33.4
22
0.91
1.83
318
261
405
gravel
clayey sand
0.65
1.27
8.3
weekly interval
95
-------
WASHED f
PEA
GRAVEL
THERMISTOR
KltYPRWYLEHE
KAOS
r.son
Figure 2. Cross-section of refuse test cells.
and others (7). Gas samples were taken
periodically and analyzed on a Perkin-
Elmer Sigma 3 gas chromatograph for, oxygen,
hydrogen,nitrogen, carbon dioxide and water
vapor. Gas analyses discussed here Include
only those for hydrogen and methane.
RESULTS AND DISCUSSION
Quantity of Gas Produced
The total gas production for landfill
simulators is given in Table 2. The aver-
age gas production for all four test cells
was 18.32 ml/kg (dry wt)/day. This is
only 30 percent of the average rate record-
ed for the first 100 days.
These gas production figures are higher
than values reported In other studies In-
volving test cells maintained for compara-
ble lengths of tine. Chian and DeWalle (2)
reported a maximum rate of gas production
of 15 ml/kg (dry wt)/day for a 400-day ex-
periment using smaller (208 1) test cells.
In data running from 912 to 1280 days into
a simulation experiment Walsh and Kinman
reported that the highest gas production
figure was only 1.0 ml/kg (dry wt)/day (9).
Unfortunately no data are available for the
first 600 days of the experiment due to
difficulties encountered in sealing the cells.
Most test cells vill underestimate gas pro-
duction due to leakage or equipment failure.
Actual production of gas In a fill has been
estimated at 22-45 iil/kg/day (3). The
higher gas production rates in this study
may reflect the increased accuracy produced
by the automated gas measuring system used
in the present equipment,
Gas Composition
In an earlier report covering the first
100 days of test cell operation it was
noted that; oxygen levels in all four test
cells declined rapidly and that the percent-
age of CO2 increased in all four tanks.
Nitrogen levels declined gradually over the
first 100 days due to sweeping of the tanks
by C02 produced in decomposition. No hydro-
gen or methane was noted in this early part
of the study.
Both methane and hydrogen appeared in
all of the test cells (except no. 16) by
-------
DATA ACQUISITION
SYSTEM
TEST
CELL
h-
Figure 3. Schematic of gas data acquisition system.
TABLE 2. GAS PRODUCTION FROM TEST CELLS FOR 636 DAYS OPERATION
Test Cells No.
10
15
16
22
Total
Gas
(1)
2928.2
25937.2
26018.2
2813.2
Methane
(1)
15.1
86.3
16.4
17.1
Hydrogen
(1)
4.8
196.0
223.4
3.8
97
-------
the 250th day of operation. The daily
average production (taken over ten-day in-
tervals) of hydrogen and methane for each
of the tanks are given in Figures 4-7. All
of the test cells showed the typical gas
production pattern proposed by Farquhar and
Rovers (5) with appearance of hydrogen pre-
ceding the development of methane. In all
cases hydrogen production has dropped off
as methane production increased. This is
due to the consumption of hydrogen in
reduction of carbon dioxide to produce
methane (6). In the test cell 16 where
hydrogen production was highest methane
production was lowest, suggesting inhibit-
ion of carbon dioxide reduction (Table 3)
was occurring.
Gas analysis were routinely made at 30
day intervals, peak methane and hydrogen
levels reported here are spot samples of
gas from the head space of the tank on the
sampling date. The highest methane con-
centration was noted as 8.9% in tank 22
after 625 days. The highest hydrogen con-
centrations (5.8%) appeared in tank 16
after 282 days. The maximum methane and
hydrogen concentrations compare well with
data obtained by Engineering-Science (A).
in instrumentation on an actual landfill.
Typical analyses of gas from the bottom of
the test landfill (Probe 33 A) showed 7.3%
methane after 648 days of decomposition.
Hydrogen concentration at this same probe
was up to 5.5% after 358 days decomposi-
tion. While a great deal of variation
exists in landfill gas composition data the
results from the test cells are reasonable
when compared to the composition of actual
landfill gas.
Scaling of Experiment - Two sizes of test
cells were incorporated into the gas genera-
tion program so that the effects of any
possible scaling problem could be noted.
The different tank designs allow the
smaller tanks to saturate more quickly due
to the shorter column of refuse involved
and the smaller tank should equilibrate
more rapidly with surrounding temperatures.
At this stage in the experiment, both of
the larger tanks show a slightly higher
average gas production rate (V10% larger)
than the smaller tanks. Thermistor data
indicate that the larger test cells are at
a slightly higher temperature than the
small test cells. Temperature differences
may be due to the slower heat dissipation
from the larger cells or the thermal strati-
fication in the building. The higher
temperatures could account for more rapid
decomposition and greater gas production.
SUMMARY AND CONCLUSIONS
Data on gas composition and gas volumes
generated from four landfill test cells
have been collected for 636 days. The
average gas production for the 636-day
period was 18.32 ml/kg (dry wt)/day. This
is only 30% of the average production rate
observed during the first 100 days of
operation. Methane and hydrogen have both
appeared in the gas emitted from the test
cells. Hydrogen appeared in each tank
before methane was detected. The highest
methane concentration observed to date was
8.9%. The highest hydrogen concentration
was 5.8%.
TABLE 3. GAS PRODUCTION FOR TEST CELLS CORRECTED FOR WEIGHT OF REFUSE
Test Cell No.
Volume of
Total Gas
Produced
per kg (dry wt)
of refuse (1)
Volume of
Methane
Produced
per kg (dry wt)
of refuse (1)
Volume of
Hydrogen
Produced
per kg (dry wt)
of refuse (1)
10
15
16
22
11.09
12.34
12.41
10.78
0.06
0.04
0.01
0.06
0.02
0.09
0.11
0.01
98
-------
0.14
-. 0.12
_!
| 0.10
H
o
O
S
<
o
Ul
<
(C
UJ
o.oa
o.oe
0.04
0.02
100 200 300 400
DAY MIDPOINT
500
600
700
Figure 4. Dally average hydrogen and methane production (over 10-day intervals) for
Teat Cell 10.
-------
4.5,-
4.0
3.5
g2.5
o
cc
0.
m
2-0
UJ
1.5
1.0
0.5
Hydrogan —
Methane
100 200
300 400
DAY MIDPOINT
500 600 700
Figure 5. Daily average hydrogen and methane production (over 10-day intervals) for
Test Cell 15.
100
-------
3.5
-. 3-0
-i
i 2.5
u
3
0
g 2.0
0.
(9
u 1-5
1
u
< 1.0
0.5
0
• Hydrogen — — —
1
- !
.
-
. t
• i
A
!\
\ .
1 1
!hV
1 ' ij ' 1 • /^ .yA
IV 7 %/ \^\
1 I i It . 1 .Tf Y X . 1 . 1
0 100 200 300 400 500 600 70
DAY MIDPOINT
Figure 6. Dally average hydrogen and methane production (over 10-day intervals) for
Test Cell 16.
101
-------
2.7
2.4
~ 2.1
_i
? 1.8
o
Q 1.5
O
K
a.
to 1.2
<
CO
111
§0.9
0.6
0.3
0
Hydrogen
'Methane •
IOO 200 300 40O
DAY MIDPOINT
500
600
700
Figure 7. Daily average hydrogen and methane production (over 10-day intervals) for
Test Cell 22.
102
-------
The larger test cells are producing
10% more gas than the smaller cells (on a
ad/kg of dry refuse/day) basis. This
effect is thought to be due to the increas-
ed microbial activity in the larger tank
due to slightly higher temperatures.
The tanks are continuing to operate in
a satisfactory manner. The gas yields
measured are in agreement with data from
similar experiments in test cells and land-
fill field investigations.
ACKNOWLEDGEMENTS
This report is part of a continuing
research program on solid and hazardous
waste disposal, which is now being con-
ducted by the U. S. Army Engineer Waterways
Experiment Station and funded by the U. S.
Environmental Protection Agency, Municipal
Environmental Research Laboratory, Solid
and Hazardous Waste Research Division,
Cincinnati, Ohio, under Interagency
Agreement EPA-IAG-DA-0569. Robert E.
Landreth is the EPA program' manager for
this work.
References
McCarty, P. L. 1964. The methane
fermentation. In: Heukelekian, H. and
N. C. Dondero. Principles and Applica-
tions in Aquatic Microbiology. John
Wiley and Sons, New York, New York.
pp. 314-343.
Myers, T. E. and others. 1979. Gas
production in sanitary landfill simula-
tors. In: Proceedings of the Fifth
Annual EPA Research Symposium. EPA-
600/0-79-023a. pp. 58-73.
Toerien, D. F. and W. H. J. Hattingh.
1969. Anaerobic Digestion. I. The
Microbiology of Anaerobic Digestion.
Water Research 3:385-416.
Walsh, J. J. and R. N. Kinman. 1979.
Leachate and gas production under con-
trolled moisture conditions. In:
Proceedings of the Fifth Annual EPA
Research Symposium. EPA-600/9-79-023a.
pp. 41-57.
1. Chen, K. Y., and F. R. Bowerman. 1974.
Mechanisms of leachate formation in
sanitary landfills. In: Yen, T. F.,
Recycling and Disposal of Solid Wastes.
Ann Arbor Science Publishers, Inc.,
Ann Arbor, Michigan, pp. 348-367.
2. Chian, E. S. K., and F. B. DeWalle.
1979. Effect of moisture regimes and
temperature on MSW stabilization. In:
Proceedings of the Fifth Annual EPA
Research Symposium. EPA-600/0-79-023a.
pp. 32-40.
3. DeWalle, F. B., E. S. K. Chian, and E.
Hammerberg. 1978. Gas production from
solid wastes in landfills. J. Environ.
Engr. Div., Amer. Soc. Civil Engr.
104(EE3): 415-432.
4. Engineering - Science, Inc. 1965. In-
Situ Investigation of Gases Produced
from Decomposing Refuse. Publ. 31,
California State Water Quality Control
Board, Sacramento, California. 211 pp.
5. Farquhar, G. J. and F. A. Rovers. 1973.
Gas production during refuse decomposi-
tion. Water, Air, and Soil Pollution
2(1): 483-490.
103
-------
FIELD VERIFICATION OF LANDFILL METHANE MOVEMENT
AND METHANE CONTROL SYSTEMS
Ron McOmher
Charles Moore
The Ohio State University
Columbus, Ohio 43210
ABSTRACT
This paper describes field studies being conducted to verify computer models developed to
predict methane transport around sanitary landfills and to provide design support for
methane migration control systems.
Introduction
This paper presents a summary of the
work completed through December 1, 1980 on
the Field Verification of Methane Movement
Functions and Methane Control Systems for
Landfills study awarded by the Negotiated
Contracts Branch of the U.S. Environmental
Protection Agency under Contract No.
68-03-2849. The contract/award is being
monitored by the Ohio State University
Research Foundation (OSURF) under the
direction of Dr. Charles A. Moore who
serves as principal investigator. The
OSURF designation for the project is RF
712366 U.S. EPA. This report is organized
to discuss the progress of the site
selection, field verification, and analyt-
ical (computer) simulation portions of the
study and includes a discussion of work
currently planned for completion of the
study.
Site Selection
The Field Verification of Methane
Movement Functions and Methane Control
Systems for landfills proposal specifies
that three landfill sites be selected for
field verification of the computer models
developed at Ohio State University (OSU)
for gas flow within soil. An additional
site is to be selected for field verifica-
tion of the relative effects of methane
migration control systems as predicted by
the computer model. This site may or may
not be one of the three sites mentioned
above. At the time of this report, two
sites have been selected for field verifi-
cation of the computer model. The criter-
ion for selection of these sites are dis-
cussed in the following paragraphs.
Criteria for Selection of Landfill Sites
The landfill sites selected for study
were evaluated according to the following
criteria:
1. surface of the soil surrounding
landfill should not be sealed,
2. cohesionless soil should surround
the landfill,
3. homogeneous soil should surround
the landfill,
4. the groundwater table should be
below the landfill,
5. at least three acres should
adjoin the landfill,
6. no burning of waste should have
occurred,
7. access for constructing gas con-
trol facilities should exist, and
8. power should be available.
In all, 40 landfill sites were con-
sidered, from which two sites have been
selected, There were several additional
sites that rated very high, and the final
104
-------
selection from among the several sites was
based uoon achieving geographical distri-
bution. The following sections present
detailed information for each of the cho-
sen sites.
Lees Lane Landfill, Louisville, Kentucky
This site was presented by the firm
of Stearns, Conrad and Schmidt (SCS) Engi-
neers. The Lees Lane Landfill is located
on a 125-acre tract 1n Jefferson County,
Kentucky as shown in Figure 1. The site
is located along the Ohio River and has
4600 feet of river frontage. The balance
of the landfill perimeter includes 800
feet of property line abutting Borden Inc.
(a chemical manufacturer); 3500 feet of
property line abuttlnq Riverside Gardens
(a residential development); and 3500 feet
of property line abutting Louisville Gas
and Electric (a power plant). As shown in
Figure 1, a floodwall right-of-way fringes
the property line abuttlnq Riverside Gar-
dens. The site was first used as a soil
borrow pit in the early 1950's. Excava-
tions to depths of 40 feet proceeded from
the northern to the southern end of the
site over a period of many years.
Waste was first received at the site
in 1952. Waste was deposited in excavated
areas and deposition generally proceeded
alonq with the excavation operation (north
to south). From 1952 until 1955 the site
served as the primary waste disposal site
for the City of Louisville and Jefferson
County. In 1955 an incinerator was opened
by the City of Louisville, and all waste
generated by the City was hauled via muni-
cipal haulers to the incinerator site.
The Lees Lane Landfill continued as the
primary waste disposal site for Jefferson
County until 1962. During this period,
private haulers serving residences and
Industries 1n the County hauled collected
waste to the site. Much of the Industrial
waste was from an area located five miles
northwest of the landfill where the rubber
Industry predominated.
In 1975, receipt of wastes at the
Lees Lane Landfill was discontinued.
Filling proceeded to original grade 1n the
northern two-thirds of the property.
Although excavations to depths of 40 feet
had proceeded to the southern property
lines, very little waste had been depo-
sited in the "southern one-third of the
property and this remains an excavated pit
to this day. All waste deposited at the
site was ultimately covered with native
soil material.
Riverside Gardens is a residential
development located east of Lees Lane
Landfill (Figure 1). In March 1975, sev-
eral flash fires were reported around
homes in this development. The City-
County Health Department monitored several
homes and methane levels up to 20 percent
were detected.
The Health Department drilled a ser-
ies of four monitoring wells in Riverside
Gardens in an east-west line perpendlclar
to the landfill boundary. These wells are
known as M-l, W-2, W-3, and W-4, and are
located as shown in Figure 1. The percent
and pressure of methane gas in these wells
were measured in March and April, 1975.
Analysis of the data indicated that the
gas had moved through the subsurface an
average of 765 feet and a maximum of 860
feet from the landfill boundary.
In April 1975 regular monitoring of
these gas wells ceased. However, an
inspection was made of these wells in
December 1977 by SCS Engineers. At that
time, the valve of Well W-2 was opened and
an audible pressure release was noted.
Since one (and perhaps all) of these wells
continued to be under pressure, generation
of landfill gases is presumed to be con-
tinuing. Further, since these gases are
migrating large distances, 1t 1s likely
that the gas generation rate 1s relatively
high.
In June, 1978 SCS Engineers was
awarded a contract by the Jefferson County
Department of Public Works to monitor the
presence of landfill gases 1n subsurface
areas Immediately surrounding the Lees
Lane Landfill.
At the onset of the project 14 moni-
toring wells were drilled. All wells were
outside of the fill area, approximately
300 feet from the edge of the refuse. The
boring logs for these wells reveal that
the soil 1s a relatively homogeneous mix-
ture of medium sand with some gravel and
smaller quantities of silt and clay.
Groundwater 1s generally at a depth of 50
feet below the ground surface.
103
-------
« EXISTING MONITORING
WELL
LOUISVILLE CAS AND EL-CTRIC
Fiqure 1. Site Map of Lee's Lane Landfill, Louisville, Kentucky
106
-------
The results of the gas monitoring
program completed by SCS at Lees Lane
Landfill promoted the Jefferson County
Health Department to install a gas migra-
tion control system along the eastern por-
tion of the site. Subsequently, the
county contracted with SCS to design and
install the system. The system has been
operating at Lees Lane since October 1980.
The Lees Lane Landfill exhibits good
conformance with the site selection crite-
ria listed above.
1. The ground surface over the'
landfill 1s not sealed. There is no
standing water on the site and all refuse
fill areas are covered only by low-lying
vegetation and grass. Most of the area
surrounding the landfill is grass-covered
or wooded.
2. The soil is predominantly of med-
ium sand with some gravel and lesser
amounts of clay and silt . Therefore, the
soil 1s relatively coheslonless.
3. The soil 1s relatively homogen-
eous to a depth of 50 feet. (The landfill
extends to a depth of 40 feet.) Textural
changes consist of spatial variations 1n
clay and silt content. Granular soils
predominate throughout the site.
4. firoundwater is approximately 10
to 15 feet below the bottom of the land-
fill. The ground water is not 1n a con-
fined aquifer, and there are no perched
water tables.
5. The land located to the west of
the landfill is accessible and has an area
greater than three acres. It 1s possible
to locate an area of the landfill which
was filled within two years time and which
was closed from eight to 25 years ago.
6. According to County officials,
burning was not an operational regularity
at the site and when fires did occur, they
were quickly extinguished.
7. A gas control system has been
Installed at the site. This system may be
turned on or off during the present study
to allow the examination of gas migration
from an Initial "day zero" simulation.
3. Power 1s available at the site.
Mlssissauga Landfill, City of Mlssissauga,
Regional Municipality of Peel, Ontario,
Canada
This site was presented by Morrison
Beatty Limited, Etobicoke, Ontario. The
site is located within the regional muni-
cipality of Peel in the southeast limits
of the City of Mississauga at latitude 43*
31' N. and longitude 79° 37' U. The land-
fill is approxmately 100 acres in size and
is bordered by Springbank Road to the
north, Mississauga Road to the west, North
Sheridan Way to the south and Queen Eliza-
beth Way to the east as shown on Figure 2.
Private residences adjoin the site to the
east while apartments are contiguous with
the site on the west.
The landfill was operated by Berrlll
and Trustrum Limited, a private contrac-
tor. It served the City of Mississauga
which has- a population of over 200,000.
The refuse fill consists of approximately
80 percent domestic refuse and garbage,
and 20 percent commercial wastes and
industrial refuse. No liquids, sludges or
hazardous wastes have been disposed at the
site and the burning of refuse was not
practiced.
The landfill 1s located 1n two worked
out sand and gravel pits. The smaller of
the two sites is 10 acres 1n area and
about 35 feet deep. Its capacity 1s on the
order of 150,000 tons, and it was filled
between April 1971 and September 1972.
The larger site is approximately 90 acres
1n area and extends from 30 to 50 feet
below the ground surface. Filling began
in 1966 and was completed in March 1980.
The first area to be filled was the nor-
theast corner of the site near Springbank
Road South and Queen Elizabeth Way. Fil-
ling proceeded in a general south west
direction. The area near Springbank Road
North was filled between 3 and 4 years
ago. It 1s in this area that the field
verification study is being conducted.
The Mlssissauga landfill site is
developed within the Iroquols Plain physi-
ographic region. This region 1s charac-
terized by a gently sloping sand plain
that averages about 2 miles in width, par-
alleling the present Lake Ontario
shoreline. The sand deposit marks the
shoreline of post-glacial Lake Iroquois.
The sand and gravel deposits at the
site were laid down 1n the form of a bar-
rier beach where a river emptied into Lake
Iroquols. The deposits are indicative of
those laid down in moderate to fast flow-
Ing waters. There 1s very little silt or
107
-------
LEGEND
» • CM "Of 'OP1
D n »«SIiM f»|TC»
• O mi 'nil* 0™lt
QU 'JCLtntO" T
Gas Monitoring
Program
MISSISSAUGA
LANDFILL SITE
LOCATION PLAN
O.MCT m IM-m
Figure 2. Site Map of M1ss1ssauga Landfill, Ontario, Canada
-------
clav in the material. The deposit is pre-
dominantly medium to coarse sand with len-
ticular deposits of gravel. There is a
zone of cemented gravel which is generally
encountered in the upper 9 feet.
Off-site methane migration was first
monitored in 1971 when explosive concen-
trations were detected in the basements of
several houses near the fill. Severe
vegetation stress also occurred in yards
contiguous with the landfill. The prelim-
inary monitoring showed the 50 percent
methane contour was more than 80 feet from
the landfill boundary. Detectable methane
concentrations were monitored in gas
probes 200 feet from the landfill.
Under the direction of the Regional
Municipality of Peel, a gas interceptor
system was installed in 1972 along Spring-
bank Road South and Queen Elizabeth Way to
protect the affected properties. Ini-
tially, this system was run on an Inter-
mittent basis. However, 1n 1977 a burner
unit designed to flare, off the gases
exhausted from the system was Installed.
The interceptor system has run continu-
ously since that time.
This system was extended in stages as
the landfill operation expanded. Gas
probes were also installed along the
interceptor system to monitor Its effec-
tiveness. The locations of these probes
are shown on Figure 3. The collection
system along Springbank Road North was
completed in September of 1977. This sys-
tem has run continuously since February of
1979.
Morrison and Beatty has monitored the
site for the Region of Peel since explo-
sive methane concentrations were first
detected in 1971. Static pressure read-
ings and percent combustible gas data were
recorded. The freguency of this monitor-
ing varied over the years, but typically
data was taken 4 or 5 times during the
winter months and twice per month during
the summer. Results of gas chromato-
graphlc analyses are also available for
some of the probes.
Officials with the Regional Munici-
pality of Peel have planned an additional
gas Interceptor system which will be
installed in the spring of 1981. This
system will run along the western and
southern edges of the fill. Any probes
that will be installed to gauge its effec-
tiveness may also be of use to this veri-
fication project.
The Mississauga Landfill Site exhi-
bits good conformance with the site selec-
tion criteria listed previously.
1. The ground surface of the site is
not sealed. However, the long winter sea-
son usual in the landfill area may freeze
the upper soil layer and effectively seal
the ground surface.
2. The soil is predominantly sand
with some silt, clay and gravel mixed in.
Therefore, the soil is relatively cohe-
sionless.
3. The soil is relatively homogene-
ous to a depth of 55 feet. (The Landfill
extends to a depth of 47 feet). Textural
changes consist of spatial variations in
clay and silt content. Granular soils
predominate throughout the site.
4. Groundwater 1s approximately 5 to
10 feet below the bottom of the landfill.
The ground water 1s not in a confined
aquifer. No perched water tables exist at
the site.
5. The land located to the north
(see Inset, Figure 2) and west of the site
is accessible and has an area greater 3
acres.
6. According to our information,
burning was not an operational regularity
at the site.
7. A gas control system has been
installed at the site. This system may be
turned on or off during the present study
to allow the examination of gas migration
from an Initial "day zero" simulation.
The Canadian Ministry of the Environment
and the Region of Peel have consented to
cooperate fully with the "day zero" simu-
lation.
8. Power 1s available at the site.
Selection of Additional S1te(s)
The selection of the third site for
inclusion in the field verification por-
tion of this project is currently proceed-
ing. We are investigating seven sites 1n
the Denver, Colorado Metropolitan Area for
inclusion in this study. The Department
of Energy 1s currently funding methane
research in the Denver area, specifically
within Adams County and Commerce City,
Colorado, and it is hoped that the inter-
109
-------
LEES
LANE
LANDFILL
Figure 3. Location of Monitoring Wells at Lee's Lane Landfill
no
-------
est in methane research in Colorado would
contribute to the progress of this study.
We anticipate that the third site for
study will be selected by January 1, 1981.
As mentioned previously, one site is
to be selected for evaluation of methane
migration control systems. At this time,
we anticipate that one of the three
sites used for the field verification of
the computer model will also be used for
this portion of the study. We discussed
the possibility of using the Misslssauga
site with representatives of the U.S. EPA
and the Canadian Ministry of Environment
1n Cincinnati late 1n July. However, no
decision- regarding this matter has yet
been made. We are considering the migra-
tion control study in our Investigation of
possible landfill sites 1n Colorado. We
believe that this will provide an alterna-
tive for comparison with the Ontario site.
Field Verification of Methane Mnvpmpnt
Predictions
The field verification of computer
predictions of methane movement 1s 1n pro-
gress. Gas concentration monitoring wells
have been Installed at Both the Lees Lane
and Misslssauga sites. SCS and Morrlson-
Beatty are currently monitoring gas con-
centrations at each of the respective
sites on a periodic basis. As we antici-
pated in the Technical Proposal, the moni-
toring wells were Installed within the
boreholes required for subsurface Inves-
tigation at each site to realize a cost
savings. A discussion of the results of
completed work at each site is contained
within the following paragraphs.
Lees Lane Landfill, Louisville, Kentucky
Ten nests of monitoring wells were
Installed 1n the period June 24 through
July 15, 1980 along Lees Lane at the loca-
tions Indicated on Figure 4. Each nest
consisted of gas probes Installed at
depths of approximately 15, 30 and 50 feet
1n three adjacent boreholes. Exceptions
are nest V-l which consisted of three
probes Installed in a single borehole at
depths of 5, 10 and 18 feet and nest V-6
wh,1ch consists of three probes 1n a single
borehole at a depth of 10 feet.
The wells for the probes were drilled
to the desired depth with a hollow stem
auger. Probes were then installed gener-
ally by the following procedure:
1. A 1 1/2-inch PVC tubing was
placed down the center of the auger. The
tubing was slotted every 3 Inches within
the 3 foot section at the desired probe
depth. The slotted section was wrapped
with cloth towels to prevent clogging of
the slots. In addition, tubing in deep
wells was slotted at the end to allow a
measurement of groundwater fluctuations.
2. The auger was pulled completely
from the well. Some caving of natural
materials occurred 1n the deep wells.
3. Backfill consisting of the auger
spoil was added around the tubing to fill
the well up to the bottom of the three
foot slotted section.
4. Three feet of pea gravel was
placed around the slotted section of the
tubing.
5. A three foot bentonlte plug layer
was placed above the pea gravel.
6. The well was backfilled to within
6 feet of grade with auger spoil and
another 3 foot bentonlte layer was placed
above the auger spoil.
7. The well was brought to grade by
placing concrete above the bentonlte plug.
A cast iron pipe approximately 18 Inches
long was placed around the PVC tubing and
into the concrete to provide protection
for the well Installation. Both the PVC
tubing and cast Iron pipe were capped.
In addition to the ten nests
Installed for this study, two gas monitor-
Ing wells Installed previously by SCS
Engineers will be monitored during the
field Investigation. These wells are
designated on Figure 4 as wells 1-6 and
III-4.
The subsurface conditions encoutered
during the Installation of gas monitoring
wells at Lees Lane Landfill were somewhat
variable.
The soils consisted generally of 7 to
15 feet of clayey silt underlain by granu-
lar soils. The granular soils consisted
of interlensed medium dense sllty sand,
medium dense sand, and medium dense sand
and gravel. The granular soils extended
to the maximum depth explored of 62 .feet.
Ground water was encountered generally at
depths of 50 to 58 feet below the ground
ill
-------
TEMPORARY I—1
FANHOUSE oT
200 mm dw.
HEADER
CITY OF
MISS1SSAUGA
LANDFILL SITE
i-ao i-8o
9-77 !,
v »••-•
8-7?
v
3OO mm dia
.HEADER
7-T7
4-00 9 •
5-80
6 ?T
v
I
1
LEGEND
O GAS WELL
*~7wOAS PROil
6"8°»6A8 PROlECntit
O W 8O"»>
PEEL
PROPfSITIf
6-so 7-so a-eo
* * *
BOUNDARY
LINE
Figure 4. Location of Monitoring Wells at M1ss1ssauga Landfill
112
-------
surface. Test Boring V-l, located within
the landfill, encountered 5 feet of clayey
silt fill underlain by municipal waste
fill to a depth of 18 feet.
M1ss1ssauga Landfill, City of Mlssissauga,
Ontario
Eight nests of monitoring wells were
Installed in the period June 10 through
June 13, 1980 at the locations indicated
on Figure 5. Each nest consisted of gas
probes installed at depths of approxi-
mately 15, 30, and 50 feet in three adja-
cent boreholes. Nests 1-80, 2-80 and 3-80
were located within the existing landfill.
The wells for the probes were drilled
to the desired depth with a hollow stem
auger. The probes were then installed by
the following general procedures:
1. The auger was pulled 6-12 Inches
and 3/4-Inch PVC tubing was placed Inside
the auger. The tubing was slotted every 3
Inches a distance of 3 feet from the
bottom of the tubing in shallow (15 foot)
and medium (30 foot) wells. In deep wells
the slotted distance was 6 feet to allow
for measurement of groundwater fluctua-
tions.
2. Silica sand was poured down the
auger and the auger slowly raised 2- to 4
feet, allowing the sand to filter through
the auger and Into the well. This
procedure was continued until the sand
covered the slotted portion of the PVC
tubing.
3. The auger was pulled another 2 to
4 feet as bentonlte in pellet or powder
form was poured down the auger. Enough
bentonlte was added to yield a 3 foot plug
over the sand.
4. The auger was then pulled com-
pletely from the well. This usually
resulted in 1 to 4 foot of caving of the
natural materials.
5. At the end of each day, all wells
Installed that day were backfilled to
within 2 feet of grade with a cement and
bentonlte grout poured from a mixer truck.
A 2 1/2-foot galvanized pipe was then
placed around the PVC tubing and the well
was brought to grade with concrete
(sacrete). Both the PVC tubing and
galvanized pipe were capped.
The subsurface conditions encountered
during the installation of gas monitoring
wells at Mlssissauga Landfill were some-
what variable. Subsoils encountered within
the exploratory borings consisted gener-
ally of interlensed deposits of sllty and
gravelly sands. These granular soils
extended to the maximum depth explored of
55 feet. Results of penetration resis-
tance tests indicate the granular soils
may be classified as medium dense to
dense. Groundwater was encountered at
depths on the order of 50 feet. Boreholes
1-80, 2-80, and 3-80, located within the
landfill, encountered approximately 5 feet
of fill underlain by refuse. The refuse
was approximately 47, 14 and 5 feet thick
in Boreholes 1-80, 2-80, and 3-80; respec-
tively. Granular soils were encountered
below the refuse in each hole.
On September 22, 1980 the gas Inter-
ceptor system at the Mlssissauga site was
turned off. Personnel from OSU and Morri-
son Beatty have been monitoring the move-
ment of methane on a weekly basis since
September. This simulation 1s scheduled
to end in mid-December. Results of the
simulation will be reported at a later
date.
Gas Sampling
In accordance with the proposal for
this study, the soil atmosphere at the
landfill sites is being sampled on a per-
iodic basis. The samples obtained are
returned to the OSU Laboratory for gas
chromatograph analysis. The results of
the chromatograph analyses will be used to
verify and/or calibrate the field explo-
simeter readings of methane concentration.
Computer Simulation of Methane Migration
The analytical portion of this study
consists primarily of computer model simu-
lation of methane migration 1n the soil
surrounding each of the subject land-
fills. The Input required for the simula-
tion Includes subsurface soil conditions,
soil properties, and the geometric con-
figuration of the landfill and surround-
ing soil. During the Autumn of 1980, the
geometry of the Mlssissauga site was used
1n a parametric study of the computer
411
-------
(iALVANI/tO PIPE WITH CAP
3/4 -INCH PVC TUBING
W/CAPS BOTH ENDS
|
I
V
\
V
>v^^
— -~^.
NO SCALE
'*- 1
••fci
_J^«
A
X
X
'
X
x-
X .
X
A
x. .
A
x-
X
• \
A
X
* X
A .
x
x"
X
A.
x, .
Sr
->
. X-
A .*<
V
x'
>0
•^ x
X
:.•.'•:
S-
if"
p^
3
«» '.
• >
L<=_
X •
• X
A .
. s
a> >
'
x •
X
• x
' ' A
.x'
.X
x'
X
A .
x .
.A
X .
X
• X
A .
^
tx*
£
-JL.
f,
« .
X ft
• x
£ '
X .
V
X >
X
X N
x
V V
n ji
DA
1
3
3
EXISTING GROUND SURFACE
3 FOOT CONCRETE COLLAR
BACKFILLED WITH CEMENT
AND BENTONITE GROUT
1 TO 4 FEET OF NATIVE
MATERIAL DUE TO CAVING
3 FOOT BENTONITE PLUG
3 FOOT SLOTTED PVC
.SURROUNDED BY SILICA
SAND
Figure 5. Typical Construction of a Methane Monitoring Well
114
-------
model for dlffusional flow. The results
of this study will be reported at a later
time.
Summary and Conclusions
To date (December 1980) field Instru-
mentation has been Installed at two field
sites and initial computer simulation stu-
dies have been completed. Currently, data
are being acquired at the two field sites,
and the verification of the computer simu-
lation Is being accomplished.
The work remaining to be done on
Phase III of the contract consists of the
selection of a third verification site.
The site selection and Implementation of
Phase IV, the verification of control sys-
tems, is 1n its initial stages.
115
-------
PLANTING TREES AND SHRUBS IN LANDFILL COVER SOIL
Edward F. Oilman, Franklin B. Flower,
Ida A. Leone, Hieheal F. TelBOB and
John J. Arthur
Cook College, Rutgers University
New Brunswick, New Jersey 08901
ABSTRACT
During the past six years, vegetation growth on completed refuse landfills has been
investigated through a nation-wide nail survey, follow-up field visits and carefully
monitored field and greenhouse research. Soil gas contamination, surface settlement, soil
compaction, thin, poor quality cover soil, low moisture, high soil temperatures and toxic
components in the soil matrix have adversly affected tree and shrub growth on landfills.
Vegetative reclamation projects were successful only when these limitations were overcome.
No species were tolerant of high landfill gas concentrations (C02 > 25%, CH4 > 30?).
However, some species such as Japanese black pine (Pinus thunbergii) were more tolerant of
the landfill than others. Shallow rooted species grew better-than deep rooters. Small
planting stock (< 1m tell) produced more leaf and stem tissue than large trees (> 2m
tall). Constructing barriers to gas migration significantly improved plant growth.
Introduction
Sanitary landfill has been demon-
strated to be the least expensive environ-
mentally acceptable means of waste disposal
available to date, purportedly possessing
the attributes of neatness and safety in
addition to relatively low cost. Whereas
such sites may have originally been located
at considerable distances from residential
areas, rapid urban and suburban development
in the United States has caused many once
remote dumping grounds to come within
developed areas. As such they provide an
attractive source of much needed land for
many purposes. Although conversion to
recreational areas or other non-structural
usage has long been considered an accep-
table end for completed landfill sites, the
urgent need for space and for increased tax
revenues has caused many municipalities to
eye completed landfills for commercial use
as well. In rural areas, intensifying land
use has resulted in attempts to use com-
pleted landfills for growing commercial
crops.
The composition of landfill refuse
varies considerably depending on its origin
be it municipal, industrial, incineration
material or sewage sludge. The organic
content of solid waste collected from
homes, schools, commercial establishments
and industries generally ranges from 50 to
75% on a weight basis. Most of these or-
ganics are biodegradable and can be broken
down into simpler compounds by both aerobic
and anaerobic micro-organisms. The rate at
which this occurs in a landfill is'reported
to be a function of (a) permeability of
cover material (b) depth of garbage (c)
amount of rainfall (d) moisture content of
the refuse (e) putrescibility of the refuse
(f) compaction (g) pH (h) age and (i) pres-
ence or absence of toxic materials.
When the refuse is initially deposited
in the landfill, there is enough oxygen
present to support a population of aerobic
bacteria. This stage lasts from one day to
many months (6). The literature indicates
C02, and H20 to be the principle products
formed in aerobic decomposition (5). The
116
-------
depletion of soil oxygen results in a
decrease in the aerobic and an increase in
the anaerobic population. During anaerobic
decompostion, the possibility exists for
production of a wide range of gases and
liquids. However, the literature indicates
COZ, H2, CH4, H2S, and N2 to be the
predoninant gases vrith C02 and CH4 making
up the largest portion of the soil gas
atmosphere (4)* These gases can migrate
into the cover soil over 'refuse and
displace the 02 normally present in the
soil air spaces.
There has been a considerable anount
of work done concerning the effects of
excess C02 in the root zone on different
plant species. In 1914, Noyes saturated
soil around tomato and corn plants with C02
(12). Both species died within two weeks,
but there was no irreversible damage to the
soil. A good deal of variation in tol-
erance to C02 between species has been
found. Cotton seedlings grown in hy-
droponlc solutions (10) were able to ex-
hibit optimum growth with 10$ C02 pres-
ent, provided at least 7'.5% 02 was also
present. Thirty to 4056 C02 in the root
zone of cotton seedlings 'was found to se-
verely reduce root growth in hydroponic
solutions. Red and black raspberry (13)
were killed when their roots were exposed
to 10? C02.
Chang and Loomis in 1945 (5) conducted
a survey of the literature and found that
although sone plants could survive 02 con-
centrations in the root zone as low aa 1 to
2%, moat plants would function normally at
02 concentrations ranging from 5 to 10$.
There is also a good deal of variability in
tolerance to low 02 in the root zone among
different species of plants. One-tenth
percent 02 in the flooded root zone of
apple trees resulted in the death of the
trees (2). Tomato plants grown in solution
culture exhibited narked reduction In
growth and ability to take up potassium
when exposed to 5$ 02 in the root sone
(15)« Sour-orange seedlings la sand-
solution culture given 1.5$ 02 in the root
zone for seventeen weeks did not grow, and
seedlings receiving 4.6 to 5*1$ 02 grew
half as well as the controls (9). Rice
plants have been reported to grow as veil
in solution culture having less than 1$ 02
in the root zone as control plants (15)-
Landfills are being converted into
usable recreation areas at a very rapid
rate. Since this trend is likely to con-
tinue for aany years because of the prev-
alence of landfills and the need for more
urban space, we will continually learn more
about how to overcome problems associated
with growing plant material on landfills.
Ve should not be satisfied with the infora-
ation contained in this report, for nany
questions remain unanswered. However,
these pages will set the ground work for
future investigations and provide a
guideline for vegetating completed landfill
sites at the present time.
Objectives
This study was designed to investigate
the problems associated with growing woody
vegetation on completed refuse landfill
sites. The objectives of these investiga-
tions are to:
1. Determine the extent of vegetation
growth problems throughout the U.S.
2. Determine why certain reclamation
efforts have been successful and others
unsuccessful.
5. Screen tree and shrub species for
tolerance to landfill conditions.
4. Identify the factors which limit
vegetation growth
5. Determine methods for overcoming
the factors limiting plant growth.
6. Draw a set of recommendations for
planting treses and shrubs on completed
landfills.
Evaluation of Plant Growth on Landfills
Throughout the United States
A mail survey of about 1,000 individ-
uals, presumed to be knowledgeable of the
vegetation associated with operating the
completed landfills throughout the con-
tinental United States and territories, was
conducted for the purpose of determining
the status of landfill vegetati.on growth.
Of the 500 people responding, approximately
75$ were unaware of any problems or repor-
ted none. Twenty-five percent reported
problems on landfills and seven percent
reported problems with vegetation adjacent
to landfills.
Using reports received through the
mail survey, landfills for site visits were
selected to represent the nine major United
States climatic regions as defined by
Trewarthsa (14). About 60 individual land-
fills were visited, and comparisons of the
quality of soil atmospheres were made in
117
-------
the root zone of healthy specimens and
individuals of the same species that were
dead or dying. Almost invariably, where
soil atmospheres contained high con-
centrations of landfill gases, vegetation
was adversely affected. It -was also found
that in 40$ of the cases, there was a
discrepancy between conditions reported in
the nail survey and those found in on-site
visits to the landfills.
A wide variety of landfill reclamation
projects were found throughout the country.
They ranged from no program at all but
where volunteer vegetation had colonized
the site to a very well managed botanical
garden containing over 2,000 plant species.
Trees and shrubs were planted on some dump-
sites and never irrigated; these projects
failed. Others were more successful
because they were managed on a permanent or
part-time basis. One key to success cer-
tainly included establishing a management
program having provisions for irrigation,
fertilization, lining and pest control.
Little variability in the magnitude of
landfill gas production and consequent
vegetation damage was observed among the
different climatic regions, except for ihe
arid area (southwestern Arizona) where
concentrations of combustible gas and car-
bon dioxide were found to be somewhat leas
than in the eight other regions, fhis waa
presumably due to lack of rainfall.
Soil characteristics observed to in-
crease in landfill gased soils, include
content of moisture and available amaoniun-
nitrogen, iron, manganese, zinc, and cop-
per. Increased availability of theae el-
ements is believed to be due to the highly
reduced conditions in the soils and the
activity of anaerobic microorganisms. Soil
pH was found to approach neutrality in
gassed soils, fhis was' probably due to the
presence of organic acids produced during
anaerobic decomposition of the buried
refuse.
Factors Affecting Plant Growth on Landfills
and Their Possible Correction
Gas
A survey of soil atmospheres on twenty
completed sanitary landfills throughout the
United States revealed that combustible gas
(methane), CO? and 02 readings were con-
centrated in low and extremely high per-
centage categories rather than being evenly
distributed among all of the categories
(7). The combustible gas readings were
most extreme in this respect, with 86.2? of
the samples containing less than 10% com-
bustible gas by volume, whereas, 12.3? of
the samples contained 25% or more com-
bustible gas and only 1.4% of the samples
had combustible gas concentrations between
10 and 24.9S. This bimodel distribution is
probably due to the tendency of refuse-
generated gas to well up in specific areas
rather than uniformly over the entire land-
fill site. This could be due to the fact
that certain areas on the landfill are -less
restrictive of gas flow and act as chimneys
for gas release, or to some characteristic
of the refuse.
fhis tendency of gas to occur in
isolated areas in the cover material could
be useful when vegetating these sites. By
locating the areas where the gases are
present, high concentrations can be avoided
and the loss of expensive trees and shrubs
can be minimized.
Cover Soil Thickness
The amount and quality of soil cover-
ing the refuse has frequently been found to
be inadequate for vegetation growth, proba-
bly because of costs and availability.
Such deficiencies need to be corrected in
areas where you wish to grow deep rooted
woody vegetation. Generally, shallow soil
areas support poor vegetation growth.
Because of the excessive cost of covering
an entire landfill with deep, rich soil, it
is suggested that consideration be given to
mounding soil in areas where deeper rooted
woody vegetation is to be planted. If no
measures have been taken to prevent gas
migration from the refuse into the final
cover soil, the placement of a thick ( 0.3
m) clay barrier beneath the soil mound
should be considered. These methods will
be described in detail subsequently.
Cover Soil Quality
Soil used for cover is frequently
below recommended minimum standards for
plant growth. It is often low in noisture
and nutrient content and mey have the wrong
pH. These problems must be overcome in
order to stimulate good plant growth.
118
-------
Toxic Compounds in Cover Soil
Tests of soils containing high con-
centrations of anaerobic gases frequently
reveal abnormally high contents of
ammonium-nitrogen and manganese. The con-
tent of heavy metals such as iron and zinc
may also be considerably higher than normal
in soils which have been anaerobic for an
extended period of time. Elevated levels
of these soil compounds can cause toxic
symptoms on many plant species. These
symptoms can include stunted growth and
death.
Soil Compaction
Cover soils frequently become highly
compacted during the closing operations of
a landfill site due to the use of large
heavy earth moving equipment. High soil
compaction has been shown to adversely
affect growth of trees and shrubs on com-
pleted landfills and other areas . Compac-
tion is also a problem in farm fields,
where tractor compacted soil can inhibit
root development. For planting in highly
compacted soil, bulk densities above 1.7-
1.8 g/cc, several provisions can be used
which will promote better root growth.
These include disking organic matter into
the soil, adding gypsum or rototilling on
small areas.
Settlement
Refuse and soil settlement will create
an undulating surface frequently causing
water to accumulate in low areas during
rainy and irrigation periods. Trees,
shrubs and grass in low areas may even-
tually die if water remains for extended
periods and/or rainfall is frequent. At
the same time it may not be possible to
adequately irrigate the elevated areas.
Undulating greens will not be tolerated in
most golf courses and, therefore, prov-
isions should be made to prevent green
settlement. Operation of farm equipment
may be hindered in settled areas of
cropland. On the other hand, in extremely
dry climates, plant growth may be greatly
enhanced in the settled areas due to in-
creased soil moisture in these areas.
Soil Moisture
The nature of the landfill soil strata
(i.e. consisting of a meter or less of soil
cover over many meters of refv.a) and,
perhaps, the higher soil temperatures
generally found on landfills help promote
drying of the cover soil. Higher rates of
leaf transpiration observed on trees grow-
ing in landfill soil nay also contribute to
severe soil desiccation. Since moisture
content is significantly lower in landfill
soil, plants on landfills will require a
greater amount of irrigation in order to
exhibit growth comparable to non-landfill
areas.
Soil Temperatures
Temperatures in anaerobic soils are
frequently higher (from a degree or two to
perhaps 20 or 30 C in extreme cases) than
nearby aerobic soils. The reason for these
high soil temperatures is not clear. High
soil temperatures have been detected on
occasion around dead trees, shrubs and
grass and evidently they have an effect on
growth in landfills. However, this problem
is generally little cause for concern com-
pared to the previously mentioned more
frequent occurring limiting factors.
Screening Species for Landfill Tolerance
Gas Tolerance
One desirable characteristic of the
plant material would appear to be tolerance
to elevated levels of carbon dioxide and
methane and low levels of oxygen. Unfor-
tunately, our research and the literature
indicates that there does not appear to be
any species which can tolerate high levels
of these gases (e.g. C02 25%, CH4
50^). However, there are several other
criteria, identified during these in-
vestigations which will greatly enhance
vegetation growth.
Rooting Depth
Tree and shrub species which enjoy a
shallow root system were found to be sig-
nificantly more adaptable to landfill sites
than species requiring a much deeper root
system (Table l). The- deeper roots are
subjected to higher concentrations of land-
fill gases and lower concentrations of 02.
Some species can avoid this gas environment
better than others by producing a shallow
root system. Observations at the South
Coast Botanical Garden on a former 87-acre
landfill site in Palos Verdes, California,
showed that shallow-rooted plants seldom
are affected by landfill gases, but on some
119
-------
occasions there has been root damage to the
deeper rooted larger trees~~snd shrubs (7).
TABLE 1. MEAN HOOT DEPTH*FOR SEVERAL
SPECIES ON LANDFILL AND CONTROL
PLOTS
TABLE 2. MEAN SHOOT LENGTH OF SUGAR MAPLES
IK LANDFILL AND CONTHOL IRRIGATED
AND NON-IRRIGATED AREAS
M
Species
Landfill
Control
Japanese Black Pine 7.8 9.3
Norway Spruce - 5«1 4.2
Hybrid Poplar
(rooted cuttings) 6.3 13.6
Honey Locust 8.3 16.6
Green Ash 9.3 14-7
Hybrid Poplar
(saplings) 8.5 12.8
Each -value is the mean of 2 replicates.
"Species are arranged from most tolerant
to least tolerant of landfill conditions.
This emphasizes the need for more
frequent irrigation in landfill soils plan-
ted with woody vegetation than comparable
non-landfill areas. In our studies at the
Edgeboro Landfill in New Jersey, five of
six species excavated for extensive root
studies on the landfill plot produced a
significantly larger portion of their root
system in the top soil layers than trees
growing in the nearby non-landfill control
area. Figure 1 shows that hybrid poplar
growing on the landfill plot has very shal-
low roots, whereas, figure 2, shows the
much deeper root system in the non-landfill
area. Because many roots were growing so
close to the surface and the top several
centimeters of soil regularly dry out for
extended periods in the temperate zone of
North America, landfill soils must be ir-
rigated more frequently than nearby non-
landfill areas to ensure good vegetation
growth. Table 2 shows that sugar maples
grew significantly better on an irrigated
plot than on a non-irrigated area at the
Edgeboro Landfill.
Flood Tolerance
In greenhouse lysimeters, red maple, a
flood tolerant species was found to be more
tolerant also of soil conditions simulating
landfill gas than sugar maple, which is
sensitive to flooding, if the trees were
irrigated regularly. This suggests that
Area1
Landfill
Control
Irrigated
Hon-irrigated
20.1b+
15-4a
24. 6c
23. 2c
Each value was computed from measurements
on 30 trees.
M
Thirty 2-year old seedlings were planted
in each of 4 areas: Landfill and control
irrigated and non-irrigated areas.
Values followed by different letters are
significantly different at p<.01.
flood tolerant species nay grow well on
landfills because of their ability to with-
stand low soil 02 cover, provided that they
are adequately irrigated. In field expe-
riments on the Edgeboro experimental land-
fill plot, flood tolerant species (i.e.
green ash, honey locust, American sycamore,
red maple) grew very poorly on the landfill
compared to species not tolerant of sat-
urated soils. The soil was significantly
dryer on the landfill plot than on a nearby
non-landfill control. Perhaps the former
species would have fared better on the dump
site if they had been supplied with ad-
ditional irrigation. Much more research is
needed to state with confidence that flood
tolerant species will withstand landfill
conditions if supplied with enough water.
Growth Sate
There is evidence that slow • growing
trees are more tolerant to landfill condi-
tions than rapidly growing species. Rapid
growers generally draw more moisture from
the soil than slow growers and would,
therefore, require more irrigation than the
latter in order to maintain growth compara-
ble to that on a non-landfill area.
However, if one is not concerned about the
relative growth on the landfill compared to
a noc-landfill area, then a faster growing
tree may be more desirable, to provide a
more quickly produced vegetative cover. It
has been demonstrated in our studies that
species classified as fast growers produce
more absolute growth on a landfill than the
120
-------
slow growers if they are regularly ir-
rigated during the first three years after
planting.
Si.ee at Maturity
Another factor to consider when selec-
ting species for landfills is size of the
tree at maturity. If the cover soil is
relatively shallow (0.5 m) then it is best
to plant a tree which remains relatively
small at maturity, or risk tree toppling
during high velocity wind storms. If a
deeper soil cover is used (1 m or more),
then the risk of windthrow will be
diminished because the root system has
adequate soil space to produce anchor
roots, provided, landfill gases are kept
out of the cover soil and the trees have
adequate moisture.
Constructing Barriers for Preventing
tamination of the Root 2one
Con-
The migration of decompositional gases
into the root zone of grass, shrubs and
trees will adversely affect their sur-
vivability. It appears that probably the
most expensive but effective procedure
available to increase the changes of ad-
equate vegetation growth on former sanitary
landfills is by active extraction of gases
from the refuse layers. Conceptually, this
reduces the chances for gas contamination
of the cover soil where vegetation is to be
planted. If this procedure ia not prac-
tical, you should consider the placement of
gas barriers between the refuse and the
roots to prevent migration of gaaes into
the root zone.
Trenches
Placing a thick (0.3 m) layer of im-
permeable clay over the final layer fol-
lowed by an adequate amount (0.6 m or
greater) of fertile soil has been shown to
be effective in preventing the upward
migration of landfill gases into the root
zone of woody trees and shrubs. This tech-
nique has functioned successfully both on a
large scale in California (an area covering
approximately one acre)and on a smaller
scale in New Jersey. A (0.3 m) layer of
gravel placed over the refuse overlain with
plastic sheeting and vented with vertical
PVC pipes (Figure j) has also apparently
successfully prevented gas contamination of
.tree roots growing in soil above this bar-
Figure 3. Gas barrier system at bottom of
trench showing gravel, plastic
sheet and vent pipes.
rier. Only one replicate of each method
was included in the experiments. Future
studies should include two or more
replicates.
Hounds
An alternate method of excluding land-
fill gases may consist of mounding soil
only in areas where deeper rooted veg-
etation (trees and shrubs) is to be plan-
ted. The mound may be underlain with a
clay layer or plastic (at least 10 mil) to
prevent upward migration of decompositional
gases. Mounding has been successful in
California (Figure 4) and in New Jersey
Another method, although untested, is
construction of a saucer lined with clay or
plastic (or some other relatively gas im-
pervious material) with a "U" tube in-
stalled et the bottom of the saucer (Figure
6). The "U" tube would allow accumulated
water to drain out the bottom while preven-
ting gases from backflushing into the
saucer. The "U" tube-liner junction must
be made gastight.
Proper functioning of all the
aforementioned gas-barrier techniques relys
122
-------
Figure U. Acacia tree growing in a 1 m
high soil mound on the Alemeda
Golf Course constructed over a
completed landfill.
Soil line
Clay
Water line
Figure 5. Saucer-lined barrier system
on the assumption that the gas aaal will
remain intact. Since refuse decomposition
will probably continue for many scores of
years, refuse settlement is likely to also
continue for some time. Cracks or breaks
in the gas barrier seal resulting from such
settlement may allow decompositional gases
to migrate through the barrier. Despite
the apparent obstacles, these systems are
likely to greatly increase chances of suc-
cessful vegetation establishment.
Size of Planting Stock
Trees planted when small (<1 m tall)
showed significantly better growth on the
landfill than those which were planted when
larger than 2 m tall, regardless of species
(Table 3). Our data have shown that this
is related to the ability of a small tree
to adapt its root system to the adverse
environment in the cover soil by producing
roots closer to the surface (i.e. away from
the higher landfill gas concentrations
deeper in the soil), whereas, roots of
larger trees start much deeper and cannot
grow to the surface before being killed by
the gases. Our data shows that, in fact,
by the time the large hybrid poplar sapl-
ings adjusted to the landfill by producing
a shallow root system, the smaller spec-
imens which started with a shallow root
system had grown to a larger size than the
saplings.
Effects of Hycorrhizae Inoculation on
Growth
Tree
Hycorrhizae fungi, when in association
with plant roots, have been shown to
greatly increase water and nutrient uptake
(ll). This symbiotic association has been
successfully used in coal strip mine
reclamation. Mycorrhizae may also aid in
successfully establishing vegetation on
completed dumpsites since landfill cover
soil frequently has a poor capacity for
holding nutrients and water. Spore and
mycelium inoculated soil has been tested
for its ability to promote mycelium
development on trees in landfill cover
soil. Our results indicate that both forms
of inoculum may be viable alternatives to
planting trees in un-inoculated soil. The
spore inoculum appears to be more suited
for higher gassed areas. The reasons for
this are unclear at the present time. The
roots may be inoculated directly just
before planting, thus increasing the
likelihood of successfully establishing the
beneficial mycorrhizae relationship.
Landfill Gas Effects on Tomato Plants
Carbon dioxide contamination in the
root zone of tomato plants in solution
123
-------
TABLE 3. SHOOT GROWTH ON SMALL AND LARGE
SPECIMENS OF FIVE SPECIES ON
LANDFILL AND CONTROL PLOTS
Species Size
Small
Pin oak
Large
Small
Green ash
Large
Small
Honey locust
Large
Small
Sugar Maple
Large
Small
Hybrid poplar
Large
Landfill as
% control
76.2*
63.0
67.8
44.3
45.0
81.4
68.4
45.5
92.8
39.5
* Calculated from at least five trees
per plot.
culture was toxic when the concentration of
C02 averaged 17.0? during the experimental
period. When C02 concentrations averaged
8.8? or less no symptoms were observed,
indicating that there was a threshold level
between 9 and 17? at which C02 became toxic
under these experimental conditions.
Exposing the roots of tomato plants to
43? CH4 for an 8-day period resulted in no
measurable adverse effects, vhereas, a
12-day exposure resulted in a decline of
the tomato plants concomitant with a
decrease in 02 concentration in the culture
vessels.
No interaction was observed among 02,
C02 and CH4 when they occurred together in
the root zone in terms of symptom develop-
ment on tomato plants. When C02 .was held
at 27?, 02 at either 5-5 or 16? caused
similar symptom development. Plants expo-
sed to 34 to 38? C02 alone exhibited the
same symptom development as plants exposed
to 34 to 38? C02 with 43? CH4.
Recommendations
1= Consult the most recent publications on
vegetating dump sites before starting a
planting program.
2. Choose a soil with a high clay content
for the primary cover material to help
prevent gas migration into the root
zone. A 0.7 n thick layer of loamy
textured soil spread over this 0.3 m
thick clay layer is likely to promote
good root growth and provide for a
successful reclamation project.
3. The clay layer functions best if the
clay is highly compacted. On the other
hand, one should avoid compacting the
soil which will support the roots of
trees and shrubs (i.e. the top 0.5 m of
cover material).
4. Consider discing organic matter into
the cover soil if compaction reaches
1.6 g/cc or greater.
5.
8.
9.
Avoid soils which are in the anaerobic
condition (i.e. soils which contain no
oxygen, have a high C02 and CH4 content
and smell very putrid).
Check at least the nitrogen, potassium,
phosphorus, manganese, iron, copper and
zinc levels in the cover soil before
planting. Amend the soil according to
soil test results. Consider adding
lime or sulfur to the soil if the pH is
lower than 5.5 or higher than 7.0.
Select plant species which have a natu-
rally shallow root system, are vol-
unteer species, have the ability to
develop a mycorrhizae relationship 01
do not attain large size (>15 m) at
maturity.
Plant trees when they are as young as
possible, preferably 2-3 years old.
Avoid planting highly
tible vegetation.
disease suscep-
10. Construct barriers to vertical gas
migration in areas where valuable trees
and shrubs will be planted. These
barriers may consist of plastic sheet-
ing (at least 10 mil preferably
thicker), clay (at least 0.3 m thick)
or other relatively gas impermeable
material.
124
-------
11. Extracting gaa from the refuse may help
prevent upward migration into the root
zone, thus increasing the chances of
successfully vegetating the landfill.
12. Irrigate plants on a former landfill
more frequently than on non-landfill
areas.
References
1. American Public Works Association.1966
Municipal Refuse. 2nd Ed. Public
Administrative Service, Chicago,
Illinois, pp. 128-132 , 134, 135-
2. Boyton, D. and 0. C. Compton. 1943*
Effect of Oxygen Pressure in Aerated
Nutrient Solution on Production of New
Roots and on Growth of Roots and Tops
of Fruit Trees. In: Proceedings of
the American Society of Horticultural
Science, 42: 53-58.
3. Chang, H. T. and Loomis. 1945. Effect
of C02 on Absorption of Water and
Nutrients by Roots. 'Plant Physiol.,
20:220-232.
4. Coe, J. J. 1970. Effect of Solid
Waste Disposal on Ground Water Qual-
ity. J. Amer. Pub. Works Aaaoc.
62:776-783.
5. Farquhar, G. J. and F. A. Rovers.
1968. Gas Production During Refuse
Decomposition. Public Works. 8:32-36.
6. Flawn, P. T. 1970. Environmental
Geology. Harper and Row, Inc. New
York, NY, p. 150.
7. Flower, F. B., I. A. Leone, E. F.
Oilman and J. J. Arthur. 1979. A
Study of Vegetation Problems As-
sociated With Refuse Landfills. EPA
Publication 600/2-79-128.
8. Gilman, E. P., F. B. Flower and I. A.
Leone. 1981(in press) iCritical Fac-
tors Controling Vegetation Growth on
Completed Sanitary Landfills. EPA
Publication.
9. Girton, R. E. 1927. The Growth of
Citrus Seedlings as Influenced by
Environmental Factors. California
Univ. Pub. in Agricultural Sciences.
5:83-112.
10. Leonard, 0. A. and J. A. Pinkard.
1946. Effects of Various 02 and C02
Levels on Cotton Root Development.
Plant Physiol. 21:18-36.
11. Marx, D.H. 1975. Mycorrhizae and Es-
tablishment of Trees on Stri pained
Land Ohio Journal of Science, 75: 288-
297.
12. Noyes, H. A. 1914. The Effect on Plant
Growth of Saturating the Soil with
C02. Science. 40:792.
13. Rajappan, J. and C. E. Boyton. 1956.
Responses of Red and Black Raspberry
Root Systems to Differences in 02 C02,
Pressures and Temperatures. Proc of
the Amer. Soc. Hort. Set.
75:402-500.
14. Trewartha, G.T. 1968. An Introduction
To Climate. McGraw-Hill Inc., N.Y.
Forth Edition.
15. Valmis, J. and A. R. Davis. 1944.
Effects of Oxygen Tension on Certain
Physiological Responses of Rice, Bar-
ley and Tomato. Plant Physiology.
18:51-65- .
125
-------
COLLECTION OF REPRESENTATIVE WATER QUALITY DATA FROM MONITORING WELLS
James P. Gibb
Illinois State Water Survey
Rudolph M. Schuller and Robert A. Griffin
Illinois State Geological Survey
Champaign, IL 61820
ABSTRACT
Monitoring wells at six sites of widely varying geologic and hydrologic conditions
were selected to study monitoring well sampling and preservation techniques. The goal
was to develop a sampling protocol that would yield water samples that are representative
of the aquifer being sampled. The effects of four pumping mechanisms, time and rate of
pumping, well flushing, and preservation techniques on the water quality of the collected
samples were studied. Pumping tests and multiple sampling experiments provided data on
which recommended sampling protocols and sample preparation, preservation, and storage
procedures are based.
The size and accessibility of the individual well, its hydrologic and chemical
character, the chemical constituents of interest, and the purpose for monitoring all
affect the selections of the pumping mechanism and procedures for sample collection.
Chemical parameters, including pH, Fe, and Zn, are more sensitive to the above variables
than other parameters.
Recommended sampling procedures are: (1) conduct a pump test to determine hydrologic
properties on which to base sampling frequencies, and time and rate of pumping prior to
sampling; (2) use a peristaltic or submersible diaphragm type pump when site location
permits; (3) use a bailer in more remotely located well sites; (4) measure pH, specific
conductance, oxidation-reduction potential, and alkalinity at the time of sample
collection; (5) make field measurements in a closed cell that does not expose the sample
to atmospheric conditions; (6) filter samples through a 0.45 vm pore sized membrane
immediately after collection; and (7) store and preserve samples according to the U.S.
EPA (1979) recommended protocols.
INTRODUCTION
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 monitor-
ing the effects of waste disposal facili-
ties on groundwater. Water sample collec-
tion and preservation techniques have been
established by several different labora-
tories and agencies in an attempt to insure
that water samples delivered to the labora-
tory are chemically representative of the
water sampled. However, there is consider-
able controversy between laboratories,
agency policies, and researchers concern-
ing proper techniques of sampling from
monitoring wells and the appropriate pro-
cedures for preserving the original chemi-
cal character of the samples. If monitor-
ing wells and water samples are to provide
the performance yardstick of disposal
facilities design and operation, the sig-
nificance of the various sampling proce-
dures and preservation techniques must be
determined.
126
-------
The three principal purposes of this
study were: 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 water quality; and 3) to deter-
mine which sampling and preservation tech-
niques should be accepted as standards for
monitoring well sampling. In order to
fulfill these objectives, the following
specific goals were set:
1. Determine the hydrologic properties of
the materials tapped by each monitor-
ing well studied.
2. Determine a pumping scheme for each
well to obtain water samples repre-
sentative of aquifer water.
3. Collect a series of samples from each
well using four different pumping
methods.
4. Determine the effects that the pumping
mechanism, time and rate of pumping,
and preservation techniques have on
the water quality of samples collected.
5. Recommend monitoring well sampling
procedures and sample preservation
techniques for specific chemical con-
stituents.
Because of the enormous quantity of
data produced by the above scheme, certain
sampling practices had to be accepted. All
of the samples were preserved according to
the U.S. EPA's (5) recommended procedures.
Also, all of the samples were stored in
linear-polyethylene bottles. These
practices were well documented in the
literature and are not detrimental to a
sample's integrity for short time periods.
During the collection, preparation, pres-
ervation, and storage of Che samples, no
new procedures or equipment were used.
METHODS
Sito Salection and Doscriptiona
Monitoring walls at six sitoo in Che
Stata of Illinois wero selected for study.
These sices represent a wido variety in
hydrologic conditions and aquifer chemical
characteristics. Two sites aro active
sanitary landfills; two ara inactive sani-
tary landfills; one is a secondary zinc
smelter; and one is a disposal site for a
hog processing plant. The wells were
installed prior to this study. Five are
cased with PVC pipe, either 3.81 or 5.08 cm
(I'-S or 2 in.) in diameter. The sixth well
is cased with 5.08-cm (2-in.) diameter
galvanized iron. The well depths range
from about 5 to 10 m (16 to 30 ft) and have
nonpumping water levels from 0 to 5 m (0 to
16 ft) below land surface.
For this paper, data collected at
Site 6 are presented. Site 6 is an in-
active landfill located in northeastern
Illinois. Filling by the trench and fill
method began in September 1952 and was
completed by November 1966 [Hughes et al.
(1)]. Household and garden refuse was the
major component of the fill, but small
amounts of spent battery acid, construc-
tion debris, and sewage sludge were also
buried at the site.
The well chosen for study is MM 63
constructed in 1968 by Hughes et al. (1).
The well is 3.81 cm (l*s in.) in diameter
and 5.12 m (16.8 ft) deep. It is cased
with PVC pipe and has a 30.5-cm (1-ft)
screen at depths from 4.82 to 5.12 m (15.8
to 16.8 ft). The well reportedly pene-
trates the landfill refuse and is completed
in a 1.22-m (4-ft) sand layer immediately
beneath the refuse.
Pumping Equipment
Four pumping methods were selected for
collecting samples from the monitoring
wells: a peristaltic pump, an air-lift
system, a nitrogen-lift system, and bailing.
For conducting pumping tests and collect-
ing samples by mechanical means, two dif-
ferent pumps were used. In wells where
pumping lifts were within suction lift
capabilities, a Masterflex 7545 variable
speed drive unit equipped with a 7015 peri-
staltic pump head was used. For wells
where pumping lifts were beyond suction
lift capabilities, a diaphragm type pump
modeled after the Middleburg pump was con-
structed and used. This typa of pump is
available commercially. The flow rate and
lift capability from this type of pump is
controlled by varying the frequency of
alternately applied and released pressure
on tho diaphragm and by regulating the
pressure at which it operates. The pump
constructed for this project was 45.72 cm
(18 in.) long, 3.18 cm (Ik in.) in diameter,
and delivered water from 12.19 m (40 ft) at
a rate of 3,500 ml/min (.92 gal/min).
127
-------
.The air-lift and nitrogen-lift pumping
systems used an apparatus similar to that
shown in figure 1. A 1.27-cm (*s-in.) diam-
eter rigid PVC discharge pipe and a0,635-ca
C't-in.) plastic airline were used for some
sampling runs. For others a 0.952-cm
(3/8-in.) diameter flexible polyvinyl dis-
charge line and 0.635-cm (k-in.) plastic
airline were used. All air-lift sampling
runs used a four-cylinder electric driven
air compressor. For the operation of the
citrogeo-lift pumping mechanism, 8.5 a
(300 ft ) capacity cylinders of compressed
nitrogen gas were used to pressurize the
identical apparatus used in the air-lift
pumping experiments. . A 2.54-cm (1-in.)
diameter stainless steel bailer, 0.914 m
(36 in.) long, was constructed for bailing
of samples. This bailer retrieves a sample
of about 300 ml (10 fl oz).
To provide a basis for comparison, the
vertical point of sample collection from
within the well for all pumping mechanisms
was the mid-point of the well screen. The
pump intakes for the mechanical, air-lift,
and nitrogen-lift systems were set in each
well at the mid-point of the screen. When
bailing, the bailer rope was marked such
1/2" cap-^
U.27em)
1/2" Tee-
(8.08 cm! 2" cap—-
2" casing —*
1/2" discharge
pipe
1/2" pipe
" cap
hole
• Shrader valve
bentonite
slurry
airline
7" bore hole-*-
(17.78 cm!
slotted PVC
well casing
(7.62cm) 3"
•">/
~ X
s *•
•/Mi
§
%°i^
a o
00
oo
CO
°0
D °
I
tSjo
"X,
1
1
I
1
•«.
n
^S
;'"•:
B
=^
v. ^\
/ ,
'v'l
:*>"•'*''
66*
f°°°
°o|
0
o o
o°0
B°°o
= ^5/^^S°o
(0.635 on)
1 6" sand(isi4cm)
12" (3048 art
•* — gravel
Figure I. Typical air- and nitrogen-lift
pumping mechanism.
that the bottom of the bailer would be
lowered to the same point in the well (the
mid-point of the screen) during each suc-
cessive bail.
Sample Collection and Preparation
Each site was sampled once a month by
one of the four pumping mechanisms. This
sampling scheme was used to study the
effects of the puaping mechanisms on sample
quality, the effects of well flushing, and
the effects of filtration. The frequency
of sampling allowed ample time for the
wells to recover from pumping during the
previous sample collection. Samples were
taken from the initial water in storage,
designated as the zero well volume, and at
h, 1, 1*5, 2, 4, 6, 8, and 10 well-volume
intervals. One well volume is defined as
the amount of water occupying the well
casing before pumping is initiated.
Measurement of pH, Eh (oxidation-
reduction potential), specific conductance,
and alkalinity were made on unfiltered
aliquots of each of the well volumes or
partial well volumes sampled. The sample
aliquots were subdivided into three por-
tions by filtering through a 3.0 vm, _
0.45 urn, or 0.22 utn pore sized Millipore
membrane. Satorius plastic filter holders
fitted for 47 mm diameter filters were used
with compressed nitrogen gas overpressure
to speed filtration. Each filtered sub-
sample was then further subdivided into
samples for total organic carbon (TOO,
cation, and anion analysis. All filtering
was conducted in the field immediately
after each sample was collected.
All samples were stored in linear
polyethylene bottles that had been washed
with dilute nitric acid and rinsed
copiously with distilled water. Detailed
procedures used for sample preservation are
listed in U.S. Environmental Protection
Agency (5).
ChemicalAnalytical Methods
l»
The Jarrell-Ash Atom Comp 975 direct
reading emission spectrophotometer (ICP),
which has an inductively coupled argon
plasma source, was used for the deter-
mination of Al, As, B, Ca, Cd, Cr, Co, Fe,
Mg, Pb, Se, and Zn. Na and K were deter-
mined by atomic absorption while Cl~, F~,
pH, and Eh were measured electrometrically
with specific ion electrodes and a platinum
128
-------
electrode for Eh measurements. Sulfatt" wns
determined turbldimetrically. Nitrate
analysis was accomplished by the cadmium
reduction method and TOC's were measured
with an Oceanographic International 0524B
total carbon system.
Field measurements of pH and Eh were
conducted with an Orion 407A field model
pH meter. Specific conductance (EC) was
measured with a YSI model 33 S-C-T meter
and alkalinity was measured by titration
to the methyl orange end point.
RESULTS
The proper procedures for collecting
water samples from monitoring wells must
be selected on the basis of the hydrology
of the site, the physical characteristics
of the well, the chemistry of the water to
be sampled, and the parameters of interest.
A comprehensive understanding of the
responses of the well and water sample to
the choice of pump, selected pumping rate,
tiiae of pumping, and appropriate preserva-
tion techniques is essential.
Pump Test Analyses
Controlled pumping tests were con-
ducted in an attempt to determine the
hydraulic characteristics of the materials
the wells were finished in. An understand-
ing of the yield potential of the well and
the transmisslvlty of the "aquifer" being
sampled is essential to devising a proper
sampling scheme. Without knowledge of
these factors, intelligent decisions can
not be made concerning the pumping rate or
the length of time the well should be pump-
ed prior to collecting a "representative"
sample.
Traditional analyses of pump test data
use the equations derived by Theis (4) and
Jacob (2). One of the basic assumptions
made in deriving those equations is that
all of the water pumped from a well during
the pumping test comes from the aquifer and
none from storage within the well. Since
this condition is not always fulfilled in
practice, particularly for low yielding
wells commonly used for monitoring, the
Theis and Jacob equations are inappropriate
for describing the behavior of water levels
during pumping for most monitoring wells.
Papadopulos and Cooper (3) presented
an equation describing the discharge from
a pumped well which lakes into account the
volume of water removed from casing storage.
The drawdown (s) inside the well is ex-
pressed as . .
s = (6.875 Q/T) F(u,a)
(1)
where
s = drawdown in the well in neters
at time t,
Q » pumping rate in 1/sgc,
T = transmlsslvity In m /day, and
F(u,«) « well function as defined by
Papadopulos and Cooper.
Drawdown values calculated from equa-*-
tion 1 differ significantly from those from
the Theis and Jacob equations during the >
early portion of the pumping test when a |
relatively high percentage, of the discharge
comes from casing storage. During the
later stages of the pumping test when only
a negligible quantity of water is obtained
from casing storage, the equations produce;
equivalent results. j
A pumping test was conducted on the j
Site 6 well on July 24, 1979. The well was
pumped with a peristaltic pump at rates
from 765 to 620 ml/mln for a period of 2 j
hours. Recovery measurements were taken j
for 10 minutes after pumping stopped. With
the Papadopulos and Cooper method of
analysis, an aquifer transmissivity of ;
about 2.48 m /day (200 gpd/ft) was deter-
mined . ;
Once the transmissivity of the aquiferi
tapped by a monitoring well has been deter-
mined, the percent of water coming from the,
aquifer and that from storage can be deter-
mined as a function of pumping time.
Figure 2 illustrates the percent o'f water '•
pumped that Is coming from aquifers of
different transmissivities at a pumping
rate of 500 ml/mln. As expected, the
higher percentages of aquifer water are
derived from the higher yielding wells
(larger T values). Once a T value of
62.0 m /day (5,000 gpd/ft) is reached,
larger values of T make little or no dif-
ference. Thus, if the yield capability
of the well is sufficiently large (T>5,000
gpd/ft), the effects of water from storage
become negligible for 5.08-cm (2-in.)
diameter wells and can be Ignored when
establishing a sampling protocol.
129
-------
SITE6-DUP
Q = 500 mL/min
T - 2.48 m2/day
Q = 500 rtiL/min
DIAMETER - 5.08cm
10 15 20
TIME, minutes
Figure 2. Percent of nquifer water versus
time for different transmissivlties.
10 15 20
TIME, minutes
Figure 1. Percent of aquifer water versus
time for different well casing diameter.
Additional calculations illustrate
the effect of the diameter of the well
on the percent of aquifer water removed
at a constant pumping rate and trans-
missivity (see figure 3). For a T of
2.48 m2/day (200 gpd/ft), well diameters
larger than 5.08 cm (2 in.) have a sig-
nificant effect on the percent of aquifer
water pumped, particularly during the very
early stages of pumping. Results from
analyses of pump test data provide'the
basic hydrologic information needed to
permit a better understanding of the '
changes in chemical quality of water likely.
to occur as the sample is pumped from a
monitoring well us a function of pumping
t ime.
Effects of Pumping Mechanisms
The pumping mechanisms selected for
this study are all commonly used methods
for collecting water samples from moni-
toring wells. The sites were sampled using
peristaltic, air-lift, nitrogen-lift, and
bailing mechanisms. For discussing the
affects of pumping mechanism on sample
chemistry, only data for samples filtered
through 0.45 urn pore size membranes are
presented.
Probably the single most important
factor affecting the chemistry of ground-
water is pH. Therefore, anything that
alters the pH of the groundwater samples
is likely to alter the chemistry of Its
constituents. On-site pH measurements at
all six sites using the four pumping
mechanisms were analyzed. The data illus-
trate two possible effects of the type of
pumping mechanism on sample chemistry. For
comparison, all of the pH data for each
site were normalized to the "stabilized"
pH values obtained for the 6, 8, and 10
well volume samples collected by the
peristaltic pump at that site. The norraal-
' ized pH values represent the relative dif-
ferences in pH between samples collected
using the peristaltic pump and the other
pumping mechanisms. The pH data from
samples collected with the peristaltic pump
were used as a base because the poristrtJ t ir
pump works in n closed system and is be-
lieved to induce the least amount of sample
alteration due to exposure to atmospheric
conditions. These delta pH values are
plotted in figure- 4 for each sample col-
lected.
Data from samples collected using the
peristaltic pump at all six sites are
grouped at a normalized pH value of 100.
The fluctuation in pH of samples collected
during the early stages of pumping
(0 through 4 volumes) probably was due to
the mixture of water stored in the well
casing and aquifer water pumped into the
well. As a higher percentage of aquifer
water was pumped (volumes 6 through 10),
the pH values of the samples stabilized.
130
-------
irt
im
X
o
I no
r
0 M
A P«r1it«lt1c PUEP
0 Alr-l'ft (V Plp«j UJ'on) _
• NUroq«r.-)lft (i,' plot)
D »lr-lm (J/8' plp«) iCtWml
• Nitrogen-lift (3/8* pipe)
X Bttltr
WJWUjpititfjjpt.
• . 3*** •
4 6
VOtUtES Of «A
Figure 4. Effects of pumping mechanisms
on pH values.
The pH values converged and presumably
reflect the pH of water representative of
the aquifer.
Two other distinct groupings of data
also are shown in figure 4. When pumping
the wells with air-lift and nitrogen-lift,
two different diameter discharge pipes were
used, a 1.27-cm (1-in.) and a 0.952-cm
(3/8-ln.). Data obtained from samples col-
lected by either air or nitrogen, with a
0.952-cm discharge pipe, appear to stabil-
ize at a normalized pH value of 110. Data
obtained from samples collected using
either air or nitrogen with the 1.27-cm
discharge pipe appear to stabilize at a
normalized pH value of 116. In both
groupings, larger fluctuations in pH values
during the early stages of pumping were
noted, presumably due to the mixing of
stored and aquifer water, as with the
mechanically pumped samples.
Marked pH value changes were noted
throughout the duration of sampling with
the air- and nitrogen-lift mechanisms. It
is postulated that the bubbling air or
nitrogen rising through the water was
stripping dissolved CO,, that was in excess
of atmospheric pressure from the water.
This resulted in increased pH values over
those measured in samples pumped with the
peristaltic pump. Smaller changes were
noted in the samples collected using the
0.952-cm discharge pipe. This probably
occurred because less air or nitrogen was
needed to pump the same quantity of water
than with the 1.27-cm discharge pipe. The
gas-to-water ratios for the samples col-
lected using the 0.952-cm 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
diameter pipe were much higher (30 to 40)
and appear to have effectively stripped the
samples of most of their excess dissolved
TABLE 1. AVERAGE pH VALUES FOR THE SIXTH,
EIGHTH, AND TENTH WELL VOLUMES COLLECTED
AT EACH SITE BY EACH MECHANISM
Site
Pump
*1.27-cm (b-in.) id inlet line
tNo data
t.952-cm (3/8-in.) id inlet line
PH
1
2
3
4
5
6
Peristaltic pump
Air-lift*
Nitrogen-lift*
Bailer
Peristaltic pump
Air-lift*
Nitrogen-lift*
Bailer
Peristaltic pump
Air-lift*
Nigrogen-liftJ
Bailer
Peristaltic pump
Air-lift*
Nitrogen-liftt
Bailer
Peristaltic pump
Alr-liftt
Nitrogen-liftt
Bailer
Peristaltic pump
Air-liftt
Nitrogen-liftt
Bailer
4.5
5.4
5.3
5.2
7.0
8.0
8.1
NDt
7.1
8.2
7.6
7.3
7.2
8.2
7.8
7.0
6.8
7.6
7.5
6.8
7.1
7.8
7.6
7.0
.
131
-------
.C02, resulting in pH values of 8.2 to 8.3.
This can be compared with pH values of 6.9
to 7.0 when the peristaltic pump was used
to collect samples. The measured pH values
are listed in table 1.
Data from samples collected at Site 5
tiding an air-lift pumping mechanism dramat-
ically illustrate the effect of discharge
pipe size on pH as shown in figure 4.
During the first stage of pumping (volumes
0 through 2), a 1.27-cm discharge pipe was
used. To increase the pumping rate, the
1,27-cm pipe was withdrawn and a 0.952-co
pipe Inserted between volumes 2 and 4.
The immediate drop In pH shown In figure 4
confirms the important effect of the gas-
to-water ratios on the pH of water samples
collected with gas-lift devices.
Because of the effect of the pumping
mechanism on sample pH, it was expected
that concentrations of some chemical
parameters also would be affected. Figure 5
shows plots of dissolved Fe concentrations
determined for each of the partial or full
well volumes sampled by each of the pumping
mechanisms for Site 6. Iron is extremely
susceptible to change in'solution concen-
tration due to change in pH or during
oxidation-reduction reactions within a
groundwater system. The soluble Fe con-
centrations in the samples taken with an
air-lift mechanism remain almost constant
through the 10 well volumes (approximately
0.60 mg/1). The initial Fe concentrations
in the samples collected with the peri-
staltic and nitrogen-lift mechanisms were
4.0 and 1.7 mg/1, respectively. However,
after pumping 4 well volumes by the three
mechanisms, the resultant dissolved Fe con-
centrations in the samples were nearly
identical. This would indicate that the
groundwater in the aquifer, volumes 4
through 10, possessed a lower level
(approximately 0.60 mg/1) of Fe than the
storage water in the well casing. Similar
results were noted for Zn ,
There was no obvious effect on Ca, K,
Mg, Mn, or Na due to the choice of pumping
mechanism. There is an absence of any
trend which would Indicate one mechanism
would be better than another when sampling
for any of these constituents. The same
general patterns of concentration Increase
or decrease with well volumes pumped occur
regardless of the pumping mechanism used,
The data should be Interpreted with caution
because the variation in constituent con-
7 —
I
SITE 6"- OUP
PERISTALTIC
• AIR-LIFT
•NITROGEN-LIFT
-BAILER
4 6
WELL VOtUMiS PUMPED
10
Figure 5. Effects of pumping mechanism on
iron concentrations at site 6 as a function
of well volumes pumped
centrations in the samples collected by
the different pump mechanisms may also
result from seasonal fluctuations in
groundwater chemistry.
To eliminate the possible effects of
seasonal variations in site geochemistry,
an additional sampling routine was used
for Site 6. Ten well volumes were pumped
with a peristaltic pump and a sample was
taken. The peristaltic pump was quickly
withdrawn from\the well, replaced by the
air-lift system, and another sample was
collected. Compressed air was replaced
with compressed nitrogen gas and another
sample was taken. Finally, the gas-lift
mechanism was removed, and a bailer was
used and a fourth sample collected. The
results were supportive of the conclusion
that changes in sample chemistry were
132
-------
induced by the pumping mechanism used and
that the differences observed previously
were due to changes in groundwater composi-
tion resulting from seasonal variations.
Effects of Well Flushing
The primary consideration during the
collection of groundwater samples from
monitoring wells is to obtain a sample
that is representative of the groundwater
in the aquifer. The recommended procedure
is to flush the monitoring well to remove
the stagnant water held in storage in the
well casing or to pump until a high per-
centage of aquifer water is being received.
Storage water is defined as the water that
doesn't come into contact with the flowing
groundwater. Acknowledging the existence
of site specific variables of geology,
hydrology, transttissivlty values, and
chemistry between waste disposal sites,
the extent of well flushing required will
be different for each well. As demon-
strated in figure 2, a specific volune of
water pumped from one monitoring well
before sample collection may be sufficient
to produce a representative sample of the
aquifer, but at another site the same
volume may be Insufficient or perhaps
result in overpumping. Overpumping may
introduce groundwater from a distant source
that could dilute or concentrate certain
constituents and result In erratic data.
Site 6 possessed the most unique
hydrogeologlc settings of the six sites.
The well was placed directly beneath the
refuse at a municipal landfill. The
landfill also possesses a relatively thin
soil cover which allows for relatively
rapid recharge of seasonal variations in
groundwater chemistry at the site.
Groundwater within a landfill also
has an extreme heterogeneous chemical
nature, dependent upon Its Immediate
environment within the landfill. Thus,
sampling from a well placed directly Into
or beneath the refuse could result in
sanples exhibiting large fluctuations In
chemical composition.
The plots of Na, K, Mg, and Ca versus
well volumes pumped for site 6 (figure 6)
show only slight changes In concentration
of these constituents while pumping.
Figure 7 depicts the change in Fe concen-
tration during the pumping of 10 well
volumes when the well was sampled In both
300
250
200
Sodium
100
50
0 2 4 8 8 1C
WELL VOLUMES PUMPED
Figure 6. Site 6; Na, K, Mg, and Ca con-
centrations versus volumes pumped
(peristaltic pump). "
August and February with a peristaltic
pump. The decrease In Fe during pumping
in August was felt to represent an aquifer
under oxidizing conditions where most of
the Fe had precipitated. The trend of
increasing Fe in the February samples
would indicate that the aquifer (or
refuse) was under a reduced environment
with more Fe In solution. The presence
of a frozen soil cover during winter,
creating a closed system and not permitting
the Introduction of oxidants, could explain
the reducing environment. The sample col-
lected during August reflect the effects
of well flushing (volumes 0 through 4),
the oxidizing environment of the summer
months (the lower overall Fe concentra-
tions) , and the heterogeneous nature of
water contained in refuse (volumes 6
through 10). It appears that it would be
very difficult to obtain reproducible
"representative" samples of water from
133
-------
0 2 4 E 8
WELL VOLUMES PUMPED
Figure 7. Site 6: Fe concentrations
versus volumes pumped for two
sampling periods (peristaltic pump).
this monitoring well. There are too many
uncontrollable factors affecting the
chemical characteristics of water contained
in the refuse.
Sulfate, F~, and Cl~ determinations
also were made on samples collected by
peristaltic pump. Data show some initial
fluctuation in the Cl~ values before a
stable concentration was obtained. Over-
all, the data suggest that the Cl~ con-
centrations found in the storage waters.
were similar to that in the aquifers. The
effect of pumping on Cl~ concentrations is
of concern since Cl~, along with pH,
temperature, and specific conductance, has
often been used to indicate when represent-
ative groundwater samples have been
obtained. The Cl~ values vary only
sl.ightly with volumes pumped as do many
of the other principal constituents. This
suggests that Cl~ would not be a good
indicator of when representative aquifer
water has been obtained.
The chemical data from this portion
of the study verified the theoretical
ratios of aquifer to stored water pre-
dicted during pumping. One group of
elements, Na, K, Mg, Ca, and Cl, is rela-
tively insensitive to time of pumping or
well flushing. Another group of elements,
Fe, Mg, Zn, Cd, Cu, As, Se, and B, is
sensitive to the effects of well flushing.
These constituents apparently undergo
chemical changes when stored in the well
casings. These constituents along with
temperature, pH, Eh, and conductivity could
be candidates as indicators of when rep-
resentative samples have been obtained.
The selection of the most suitable indica-
tor should be based on the hydrologic and
chemical characteristics of each individual
site.
Effects of Sample Preparation,
Preservation, and Storage
The effort to collect a representative
groundwater sample from an aquifer via a
monitoring well would be futile if the
sample chemistry changed between the time
of collection and analysis. Proper sample
preparation, preservation, and storage can
help prevent such changes from occurring.
Adequate procedures have been developed for
sample preservation and many investigations
into the proper vessels for sample storage
have also been completed. This phase of
the project was primarily concerned with
whether these prescribed procedures for
preparation and preservation of samples
should be applied on-site immediately after
sample collection or in the laboratory some
time later.
Four-liter samples taken at the 10th
well volume from Site 6 during the February
samplings were used for the storage
analysis. Calcium, K, Mg, Mn, and Na show
little change in chemical composition in
response to storage time. The changes in
pH, Fe, and Zn, however, were immediate.
During the 7 hours between collection of
the samples and return to the laboratory,
where a subsample of the larger volumes was
taken, these parameters underwent signifi-
cant change. Almost all of the dissolved
Fe and Zn present in the samples was lost
during the first 7 hours of storage. This
loss could be due to adsorption of the
134
-------
particulate natter onto the wall of the
storage container. The most probable
cause in reduction of the Fe concentrations
is precipitation.
The effects of filter pore size on
sample chemical composition also were
studied, Subsamples, filtered through a
3.0, 0.45, or 0.22 urn pore sized filter,
were taken from each well volume or partial
well volume collected during the monthly
sampling routines. The analyses of these
subsanples were then compared to deter-
mine the effect of membrane pore size on
constituent concentration in the filtered
samples.
These data show that for Ca, Mg, Na,
K, and Mn the membrane pore size used for
sample preparation does not affect their
concentration in solution. However, the
Fe and Zn concentrations showed a definite
effect fron filter pore sizes. The
highest Fe and Zn concentrations were found
in the samples filtered through the 3.0 urn
pore sized membrane. The next highest con-
centrations were found in the samples
filtered through the 0.45 urn filters. The
lowest concentrations of Fe and Zn were
found in samples filtered through the
0.22 urn filters. The samples were very
turbid and colloidal material was observed
to pass through the 3.0 um membrane but
not the 0.45 or 0.22 urn pore sized
membranes. Subsequent acidification of
the samples for preservation resulted in
leaching or dissolution of Fe and Zn from
these colloid particles.
CONCLUSIONS
The results of this study show that
collecting "representative" water samples
from monitoring wells is not a straight-
forward or easily accomplished task. Each
monitoring well-aquifer system combination
has its own individual hydrologic and
chemical character that oust be considered
when planning a sampling protocol. The
selection of the type of sampling device;
sample preparation, preservation, §nd
storage; and sampling procedures all must
be tailored to the size and accessibility
of the individual well, its hydrologic and
chemical character, the chemical constit-
uents of interest, the time of year, and
the purpose for monitoring.
It has been demonstrated that meaning-
ful pump tests can and should be conducted
on monitoring wells. The analyses of time-
drawdown data from constant rate pump tests
using the equations developed by Papadopulos
and Cooper (3) yielded realistic trans-
missivity values for che materials tapped
by the monitoring wells studied. Aquifer
transmissivities can be used to project
time-drawdown relationships at various
pumping rates and to determine the percent
of "aquifer water" contained in the total
amounts pumped at any given time. This
relationship can and should be used as a
guide for obtaining "representative" water
samples from monitoring wells.
The transmissivity values are also
invaluable when attempting to determine
times of travel of groundwater within the
aquifer or materials being monitored. By
applying appropriate hydraulic gradients,
porosities, and transmissivities, the bulk
velocities of groundwater movement can be
estimated. These velocities can be used
to project the rate of migration of pollu-
tants detected in the monitoring well and
to describe the geometry and dynamics of
the contaminate plume. The velocity of
regional groundwater flow also should be
used to determine realistic and economic
frequencies for sample collection. In
very "tight" materials (low transmis-
sivities) the rates of groundwater move-
ment may be on the order of 0.5 to 2 m per
year (1 to 5 ft), If monthly sampling is
prescribed in a case such as this, the
same "slug" of water may be, la effect,
sampled for 2 or 3 months consecutively.
Chemical data from samples collected
with the four types of pumps used in this
study indicated that peristaltic pumps and
bailing yielded comparable data and caused
the least changes in chemical quality of
water delivered to the surface. Air- and
nitrogen-lift pumping mechanisms increased
the pH of water samples during pumping and
therefore altered the chemical concentra-
tions of pH sensitive constituents. Iron
and zinc were shown to be particularly
sensitive to the use of these types of
pumps.
The effects of flushing out a well,
or pumping the well for a period of time
to insure collection of a "representative"
sample, also have been effectively docu-
mented. In most cases, the water stored
in the well casing was of different
chemical quality than that contained in
the aquifer. Usually, the oxidizing
133
-------
environment in the well was sufficiently
different from that of the aquifer to
create a shift of chemical species in
solution.
To insure that a "representative"
aquifer sample is collected, the well
should be pumped until a high percentage
of "aquifer water" is obtained. The length
of tine of pumping will depend on the rate
of pimping, well diameter, and transssis-
sivity of the aquifer being sampled. With
these factors, the aquifer water percent-
ages with time can be calculated and used
as a guide for determining the appropriate
pumping time before a sample should be
collected. Monitoring pH while pumping
with a peristaltic pump or bailer appears
to be a reasonable field check for assur-
ing that a representative aquifer sample
is being collected.
Data collected from the same moni-
toring wells at different times of the
year show that significant seasonal varia-
tions in chemical quality can occur. These
changes can be related to varying rates
of recharge and changes in oxidation-
reduction conditions in relatively shallow
aquifer systems. Awareness of the possi-
bilities of these types of seasonal varia-
tions is essential to understand and
properly interpret the significance of
changes in chemical results with time.
Changes in field measured pH or specific
conductance .values from one sampling period
to another should be expected and are not
causes to abandon established sampling
procedures.
Data from the sanple preparation,
preservation, and storage portions of the
study show that, when chemical concentra-
tions for certain constituents are desired,
samples definitely should be filtered in
the field at the time of collection.
Chemical constituents sensitive to pH
changes can be affected within 7 hours if
not filtered and preserved at the time of
collection. The constituents most sensi-
tive were Fe and En.
In addition, the selection of filter
pore sizes used to filter samples can
affect the chemical analyses results.
Mineral constituents associated with clay
particles increased significantly when
the samples were filtered through 3.0 urn
pore size membranes. Samples filtered
through 0.45 v pore size membranes had
only slightly higher clay related mineral
values over those filtered through 0,22 urn
pore size membranes. For practical pur-
poses, the differences in chemical compo-
sition were small and the use of the
0.45 VB pore size membrane appears satis-
factory. Use of the small pore size
filters was more time consuming and
resulted in filter clogging problems,
particularly for turbid samples.
RECOMMENDATIONS
The purpose or philosophy of a pro-
posed groundwater monitoring program must
be agreed upon and used as the basis for
locating and designing monitoring wells
and establishing applicable sampling pro-
cedures. This study was designed to
evaluate practical sampling techniques and
sample preparation, preservation, and
storage procedures. The basic goal was to
develop procedures and protocols to obtain
chemical results representative of water
contained in the materials being sampled.
.On the basis of the results of this
study, the following recommendations are
made.
1) A brief, 2 or 3 hour, pumping test
should be conducted on each monitoring
well to be sampled. Analyses of the
pump test data and other hydrologic
information should be used to determine
the frequency at which samples will be
collected and the rate and period of
time each well should be pumped prior
to collecting the sample.
2) The general rules of thumb for pumping
4 to 6 well volumes will in most cases
produce samples representative of
aquifer water. For aquifers with
unusually high transmissivitles,' pump-
ing for periods long enough to remove
these volumes of water may induce
migration of water from parts of the
aquifer remote from the monitoring
well, The calculation of percent
aquifer water with time provides a more
rational basis for determining the
length of pumping. Samples should be
collected in the minimum time required
to produce "representative aquifer
water."
3) A controlled sampling experiment,
similar to those in this study, prefer-
ably using a peristaltic or submersible
136
-------
diaphragm type pump, should be conducted
to accurately determine the chemical
quality of the aquifer water and to
verify the chemical response of the
monitoring well to pumping as predicted
from the pump test data. Once the
chemical character and responses of
the monitoring systems have been deter-
mined, key chemical constituents for
routine sampling can be selected.
4) On the basis of the sensitivity of the
selected "indicators", a choice of pumps
for routine sampling can be made. The
use of air- or nitrogen-lift pumping
mechanisms should be restricted to pH
insensitive chemical constituents.
Although this study dealt with In-
organic constituents, the data suggest
that these types of pumping mechanisms
probably would also strip volatile
organic compounds from the water
during pumping. The peristaltic or
.submersible diaphragm pumps and the
bailer are recommended for most
applications.
5) The monitoring well should be pumped
at a constant rate for a period of
time that will result In delivery of
at least 95 percent aquifer water. The
rate and time of pumping should be
determined based on the transmissivity
of the aquifer, the well diameter, and
the results of the sampling experiment.
6) Measurements of pH, Eh, and specific
conductance should be made at the time
of sample collection. These measure-
ments should be made within a closed
cell which will prevent the sample
from coning into contact with the
atmosphere. All samples should be
promptly filtered through a 0.45 um
pore size membrane and preserved
according .to recommended U.S. EPA
procedures for the chemical constit-
uents of interest,
7) Strict adherence to the established
sample collection procedure and pres-
ervation protocol is essential to
produce results that can be compared
with other results from the same well
or other wells in a given study.
REFERENCES
1. Hughes, G. M, R. A. Landon, and
R. N. Farvolden. 1971. Hydrogeology
of Solid Waste Disposal Sites in
Northeastern Illinois. U.S. Envi-
ronmental Protection Agency, Washing-
ton, B.C. SW-12d.
2. Jacob, C. E. 1950. Flow of Ground
Water. Engineering Hydraulics, edited
by H. Rouse, John Wiley & Sons, New
York.
3. Papadopulos, I. S., and H. Cooper.
1967. Drawdown In a well of large
diameter. Water Resources Research
(First Quarter), 3(1):241-244.
4. Theis, C. V. 1935. The relation
between the lowering of the piezo-
metric surface and the rate and
duration of discharge of a well
using ground-water storage. Trans,
Am. Geophys. Union, 16:518-524,
5. U.S. Environmental Protection Agency.
1979. Methods for Chemical Analysis
of Water and Wastes. EPA-600/4-49-G20,
117
-------
ASSESSMENT OF LINER MATERIALS FOR MUNICIPAL SOLID WASTE LANDFILLS
M. A. Fong
H. E. Haxo, Jr.
Matrecon, Inc.
Oakland, California 94623
ABSTRACT
The results of the effects of exposure of 12 lining materials for 56 months to
municipal solid waste (MSW) leachate under conditions designed to simulate those that
exist at the bottom of a MSW landfill are presented and discussed. The principal liner
materials in this study were:
Four admix materials:
Paving asphalt concrete
Hydraulic asphalt concrete
Soil asphalt
Soil cement
Two asphaltic membranes:
Bituminous seal
Emulsified asphalt on a nonwoven fabric
Six flexible polymeric membranes based upon
the following polymers:
Butyl rubber
Chlorinated polyethylene (CPE)
Chlorosulfonated polyethylene (CSPE)
Ethylene propylene rubber (EPDM)
Low-density polyethylene (LDPE)
Polyvinyl chloride (PVC)
In addition to the 12 primary liner specimens, 42 samples of membrane materials were
also exposed to the leachate as coiled strips buried in the sand above the primary liner
specimens.
The 56 months of exposure to the leachate did not increase the permeability of any
of the liners. However, the exposure resulted in losses in the compressive strength of
the admix liner materials and in softening of the asphaltic materials. It also resulted
in swelling of most of the polymeric membranes and in losses in the. physical properties
of some of the membranes. Several seams lost strength, but the heat-sealed seams,
as a group, retained their strength best.
Among the flexible membranes, the low-density polyethylene, a partially crystalline
polymer, sustained the least change during the exposure period; however, the low-density
polyethylene membrane was thin and susceptible to easy puncturing and tearing which could
cause problems during installation and probably cause problems during service under most
conditions. The thermoplastic membranes, i.e., chlorinated polyethylene and Chlorosul-
fonated polyethylene, tended to swell most. The vulcanized rubbery liner materials,
i.e., butyl and EPDM, cnanged little during the exposure period, but had the lowest
initial seam strength.
138
-------
This paper also presents the results of immersing specimens of 28 different poly-
meric membrane liner materials in MSW landfill leachate for 31 months. These 28 liners
include variations in polymer, compound, thickness, fabric reinforcement, and manufac-
turer. The polymeric materials include butyl rubber, chlorinated polyethylene, chloro-
sulfonated polyethylene, elasticized polyolefin, ethylene propylene rubber, neoprene,
polybutylene, polyester elastomer, low-density polyethylene, plasticized polyvinyl
chloride, and polyvinyl chloride plus pitch.
INTRODUCTION
This engineering research project was
undertaken in 1973 with the primary
purpose of assessing the effects of
exposure to leachate from municipal
solid wastes upon a variety of liner
materials. At the time the project
was initiated, the lining of landfills
with low-permeability materials appeared
to be a feasible means of preventing the
leachate generated in a landfill from
migrating into and polluting surface and
groundwaters. Considerable data existed
in 1973 as to the service lives of lining
materials for water containment and
conveyance, but there was concern as to
the long-terra durability and effectiveness
of various liner materials in contact with
MSW leachate.
Interim results on this project have
been reported as they were available
(Haxo and White, 1974; Haxo, 1976; Haxo
and White, 1976; Haxo, 1979; Haxo et al,,
1979). The effects of one year of exposure
were comparatively mild (Haxo and White,
1976) and the data were inadequate
to form_a basis for making long-range
projections of service lives. Consequent-
ly, the exposure periods were extended
from the original two-year period to 56
months and the scope of the project
was expanded to include studies on addi-
tional materials and an investigation of
simpler methods of assessing the effec-
tiveness and durability of the various
liner materials.
This paper reports on the Investi-
gations of polymeric membrane Hner
materials and particularly on the effects
of the total Immersion of these materials
for 31 months In leachate produced in the
same simulated landfills used in this
project. Water absorption is also dis-
cussed. This paper summarizes the find-
ings with respect to exposure testing for
the entire project,
METHODOLOGY AND EXPERIMENTAL DETAILS
The basic approach followed in
this investigation was to expose a variety
of materials to MSW leachate under condi-
tions which simulate those that would be
encountered by a liner in real service.
The effects of the exposure were deter-
mined by measuring appropriate properties
of the respective liners as a function of
exposure time. All lining materials,
except compacted soils and clays, were
considered; the latter were studied in
other projects.
Two sets of 12 primary specimens were
placed in exposure In 24 simulators and
retrieved and tested at two exposure
times. The principal liner materials
that were selected for the exposure study
included a wide range of materials that
were used at the time for water contain-
ment and conveyance. They consisted of
admix materials, 2 spray-on asphalt
membranes, and 6 commercial polymeric
membranes that were available in 1973.
The conditions at the bottom of a
landfill that must be simulated for the
generation of leachate and for the ex-
posure testing of liner materials include
the following features:
- Municipal solid waste.
- A porous soil cover on the liner
on which waste can be dumped
without damaging the liner and
through which leachate generated 1n
the fill can drain.
-The liner, a material of low
permeability, which can vary
considerably in thickness.
- Compacted sand on which the Hner
has been placed. The soil has been
graded to allow drainage of the
leachate.
- Uncompacted native soil.
- Sroundwater level, usually well
below the landfill.
139
-------
The design of the simulated landfill
and leachate exposure apparatus is illus-
trated in Figures 1 and 2.
••••-:
-------
143
1974
ILAPSSDTIMi
Figure 3. Average solids contents of the leachate produced 1n the simulators, November
1974 - August 1979. The data for November 1974 - November 1975 are the
averages for the leachate from 24 simulators. Twelve simulators were dis-
assembled in November 1975 and, consequently, the data for December 1975 -
July 1979 are the averages for the leachates from the 12 remaining simulators.
I
10
ELAPSED TIME
Figure 4. Average total volatile acids content (TVA), as acetic acid, of the leachate
produced in the simulators, November 1974 - August 1979. The data for November
1974 - November 1975 are the averages for the leachates from 24 simulators.
The data for December 1975 - July 1979 are the averages for the leachates from
12 simulators.
141
-------
RESULTS OF 56 MONTHS OF EXPOSURE OF LINERS
TO MSW LEACHATE
Retrieval of Liners from Simulators
The first set of 12 simulators were
dismantled at the end of one year of
operation and the samples of the liner
materials were recovered and tested (Haxo
and White, 1976; Haxo, 1977). The columns
of waste of two simulators were lifted off
the bases after 43 months of operation and
the liner specimens, both asphalt mem-
branes, in the bases were inspected.
Also, the strip specimens buried in the
sand above the primary liner specimens
mounted in the bases of the two simulators
were recovered and tested. One of the
asphalt membranes, the emulsified asphalt,
was sampled and the hole repaired with a
cement grout. New strip specimens
were buried in the sand and the simulators
were returned to operation. All 12 of the
simulators in the second set were dis-
mantled after 56 months of operation and
the liner specimens, both the primary and
the buried, were recovered and tested.
The epoxy seals in several of the bases
had deteriorated allowing leachate .to
bypass the liners and enter the bases. In
these cases, the permeability of the
liners could not be assessed. As most of
the epoxy seals withstood the leachate, it
is believed that for those cast rings
which failed, the mix ratios of the
two-part resin were not correct. Before
removing or cutting the exposed liner
specimens, they were leak-tested by
flooding the surfaces and slightly pres-
surizing the spaces below and any air
bubbles were noted. All specimens were
kept moist until tested.
Admix Liner Materials
Test results on the four admix liner
materials are presented in Tables 2 and
3.
Hydraulic Asphalt Concrete
The hydraulic asphalt concrete
specimens were disks 22 inches in di-
ameter and 2 4 inches in thickness. The
concrete was hot mixed and compacted
in forms outside the bases. The specimens
were sealed into the bases with an epoxy
resin (Haxo and White, 1974; Haxo and
Wfoite, 1976). The concrete contained 9%
asphalt of 60-70 penetration grade.
The aggregate was Watsonville granite
proportioned to meet 0.25 inch maximum
gradation for dense-graded asphalt con-
crete. The original voids ratio of the
concrete was 2.9%.
At the time the simulators were
opened, sand above the liner had hardened
in an area essentially coinciding with the
area of the glass fabric which covered the
drain tube. A dark material had been
deposited along both sides of the drain
tube, which had sunk into the liner.
The entire epoxy ring was light-colored
and appeared to be in good condition. No
leaks were found when the liner and ring
were flooded and pressure applied from
below. Six two-inch diameter cores were
cut. The bottoms of the cores and the
gravel underneath appeared dry.
The unconfined compressive strength
of the hydraulic asphalt exposed for
56 months was lower than that exposed
for 12 months; the strength at 56 months
was only 6% of the strength of an unex-
posed sample and 13% at 12 months (Table
2). There was no measurable flow through
a core of the exposed liner during a
four-day run 1n the back-pressure perme-
ameter, so the coefficient of permea-
bility has been reported as less than
10"9 cm-s'l. Asphalt extracted
from the top of the liner was harder than
the original extracted ashpalt, and only
slightly harder than that extracted from
the liner exposed for only one year.
There is some question regarding the high
viscosity of the asphalt that was extract-
ed from the bottom part of the liner.
This value is being checked. Absorption of
water by the asphalt binder affects the
compressive strength values of the
concrete which are determined without
drying of the specimens. Water absorption
does not influence the viscosity of the
asphalt as the absorbed water is removed
during the extraction and recovery pro-
cess.
Paving Asphalt Concrete
The paving asphalt concrete specimens
were prepared in a similar fashion to the
hydraulic asphalt concrete specimens.
This concrete was hot mixed and contained
7% asphalt (60-70 penetration grade).
The aggregate was also Watsonville granite
proportioned to meet 0.25 inch maximum
142
-------
TABLE 2. PROPERTIES Of AOMIX LINER MATERIALS
Properties after 56 months of exposure
to HSU leachate:
Thickness, Inches
Density. g/cm3 (Ib/ft*)
Voids ratio, vol. voids/vol. solids.
Mater content, g water/ 100 g dry wt.
Compresslve strength, psl
Strength retained of original, %
Leachate collected below liner during
monitoring period:
Total cumulation, kg
Week leakage began
Coefficient of permeability', cm/sec:
Initial value
After one year exposure to land-
fill Ieachate9 (Individual
determinations)
After 56 months exposure to land-
fill Ieachate9 (Individual
determinations)
Paving
asphalt
concrete
2.2
2.417 (150.9)
I 5.3
1.06
258
gb
20th
1.2 x ID'8
7.4 x 10-'
9.3 x 10-9
3.0 x 10-9
<10-9
Hydraulic
asphalt Soil Soil
concrete cement asphalt
2.4 4.5 4.0
2.385 (148.9) ... 2.019 (126.0)
3.6
0.96 ... 6.51
172 1182' 26
6b 62= 2b
3.24<< none 12.1
1st
3.3 x 10-9 j.s » io-6f 1.7 x i0-3
<3.0 x 10-10 1.5 „ io-8h 1.3 „ 10-8
3.5 x 10-9 4.0 x 10-'1 2.8 x lO'8
<10-9 1.2 x 10-* e.6 x 10'8
2.7 x 10-5J 2.7 x 10-'
4.3 x 10-'k
1.2 x 10-5<
'Tested exposed specimen contained 10* cement.
''Original specimen molded from unexposed sample stored 12 months.
'Original specimen tested on unexposed molded specimen 12 months after molding, specimen contained
12S cement.
don dismantling of simulator, leak Has found to be In epoxy seal.
'Back pressure permeameter tests made on cores cut from compacted liner specimen.
^Molded test specimen. Cores satisfactory for testing could not be cut from the compacted liner specimen.
jlValue for core specimens cut from liner.
"Top section of Core 3.
'Bottom section of Core 4.
•Mop and bottom sections of Core 1, which included a leak detected before coring.
*Top and bottom sections of Core 4 where leaks were detected before coring.
'Top and bottom sections of Core 3.
-------
TABLE 3. CHANGES IN PROPERTIES OF ASPHALT IN ADMIX MATERIALS AND MEMBRANES DURING 12 AND 56 MONTHS EXPOSURES TO MSU LANDFILL LEAC1IATE
Type of asphalt
Viscosity at 25'C, sliding plate
vtscometer, HP:
At shear rate of 0.05 sec'*:
Original
After 12 months
After 56 months
Change
Change, »
Penetration at 25'C:
Original
After 12 months*
After 56 months6
Change
Voids ratio (volume voids/volume
solids x 100):
Original
After 12 months
Change
Softening point, 'C :
Original
After 12 months
After 56 months
Asphalt concrete
Whole
core Top Bottom
60-70 pen paving grade
asphalt
S.ja
a!a '... "...
9.1 10.7
+3.7 -4.0 +5.6
+73 +78 +110
44«
36
35 33
-8 -9 -11
6.4
4.2
-2.1
. , . ... ...
Hydraulic asphalt concrete
Whole
core Top Bottom
60-70 pen paving grade
asphalt
2.1«
3!3 '..'. .'.'.
4.2 13. 5C
-1.2 -2.1 +11.4
+57 +100 +548
62«
52
47 30
-10 -15 -32
2.9
1.9
-1.0
... ... ...
... ... ...
Soil asphalt
Whole
core Top Bottom
Slow curing liquid as-
phalt SC-800 grade
0.02°
0.04
... 0.005 0.012
+0.02 -0.015 -0.008
+100 -75 -40
538b'e
390
1020 668
-148 +482 +352
10.4
26.1
+15.7
... ... ...
Bituminous
seal
Catalytic ally
blown asphalt
8.5
10.04
... 17.4
+1.9 +8.9
+22 +104
36
34
27
-2 -9
89
89
101
Asphalt
on non-
wovcn fabric
Emulsified
asphalt
4.5" ...
2.9
3.1
-1.6 1.4
-36 -31
45b,e
55 .'.'.'
0.53
+9 +7
... ...
'Calculated from penetration data.
bAsphalt extracted from unexposed specimens stored 12 months.
C0ata questionable and being checked.
dAsphalt before mixing was 68 penetration. I.e., ca. 1.7 MP.
Calculated from viscosity data.
-------
gradation for dense-graded asphalt.
The specimens were disks 2.2 inches
thick and 22 inches in diameter. They
were compacted hot into forms outside the
bases and sealed into the simulator bases
with epoxy adhesive (Haxo and White, 1974;
Haxo and White, 1976).
On opening the simulator above the
liner, it was found that much of the sand
had hardened with a dark cementing agent.
Some of the hardened sand had loose sand
under it. About two-thirds of the epoxy
ring was in very poor condition; swollen,
flaking, soft and very dark; however, no
leak was found when the liner and ring
were flooded and pressure applied below.
Six two-inch diameter cores were cut. The
bottoms of the cores and the gravel
underneath appeared dry. The permeability
of the asphalt concrete decreased during
the exposure period. The viscosity of the
asphalt increased significantly.
Compressive strength of the asphalt
concrete liner after 56 months of exposure
was only 9% of the strength of the unex-
posed concrete of the same composition.
After 12 months of exposure, the liner had
retained 15% of the strength of the
unexposed liner, thus indicating continued
deterioration of the asphalt concrete.
There was no measurable flow through a
core of the exposed Uner during a three-
day run in the back-pressure permeameter.
Since this time interval would be adequate
for determining a coefficient of permea-
bility of 10"^ cm-s~l, we report less
than 10~9 cm-s~l for this liner. The
properties of the asphalt extracted
from it were very similar to those mea-
sured after 12 months of exposure, indi-
cating that very little change to the
asphalt occurred during the additional
exposure. Asphalt extracted from a
concrete specimen of the same original
batch after 56 months of exposure to air
and weather had a viscosity of 37.0 MP at
25*C and a calculated penetration of 21.
Thus the exposure in the simulator was
much less severe on the asphalt per se
than normal weather, which probably
reflects the anaerobic environment at the
bottom of a landfill.
Soil Asphalt
The soil asphalt specimens were four
inches thick; and were compacted directly
in the bases. The soil asphalt consisted
of 100 parts soil mixed with 7.0 parts of
liquid asphalt, Grade SC-800 (Haxo and
White 1974; Haxo and White, 1976). A ring
seal of epoxy resin was placed at the top
of the compacted liner to prevent possible
bypassing of the liner by the leachate at
the edge of the specimen.
When the simulators were dismantled,
sand was scraped aside just enough to
examine the condition of the Uner, which
was very soft and had the appearance of
stiff mud. Black, soft concretions
were around much of the rim and near the
drain tube. Also, spots of concretions
were scattered over the surface of the
liner. The fabric had been cut just past
the end of the drain tubes, and the Uner
was indented, apparently by the rim of the
simulator during removal. The epoxy ring
appeared to be in good condition, though
part appeared to contain too much, gravel.
A black deposit along the drain tube
was sticky and tarry. In the pressure
leak test, leaks were found in the Uner
and along the epoxy rings. All of which
were within three to four inches of the
rim. The leak areas were mostly from
areas of oily concretions on the surface.
The liner' material lost almost all of its
compressive strength during the exposure.
At the time of compaction, the unconfined
compressive strength measured 1218 psi, at
the end of a year It measured 15 ps1, and
at 56 months it measured 26 psi.
A core cut from the center was
dry. Samples of gravel from the bottom of
the core holes appeared wet.
When a core, which came from an area
where leaks were detected, was tested in
the back-pressure permeameter, it was
deformed by the confining pressure used in
the permeameter. The value determined for
coefficient of permeability, approximately
10~7 cm-s'l, may not be reliable
because of deformation of the specimen.
Asphalt extracted from the bottom
half of a core was slightly less viscous
than the original SC-800 grade liquid
asphalt, and asphalt extracted from the
top half of the core was much lower in
viscosity than the original. This effect,
and the presence of a small amount of oily
material 1n the sand covering the Uner,
leads to the conclusion that the waste in
the simulator must have contained a
145
-------
low-viscosity, nonvolatile, oily sub-
stance which contaminated and softened
the liner.
Soil Cement
The soil cement specimens were
compacted directly in the bases and sealed
with a ring of epoxy resin. These speci-
mens were 4.5 inches in thickness and were
made of 95 parts of soil, 5 parts of
Kaolinite clay, 10 parts Portland cement,
and 8.5 parts water which were thoroughly
mixed before compaction (Haxo and White,
1974; Haxo and White, 1976).
The sand layer was very shallow
because of the thickness of the liner.
When opened, it was found that only a
small amount of sand was hardened,
mostly along a line extending perpendic-
ular from the middle of the drain tube to
the rim. There was a dark deposit along
the drain tube, and most of the slots were
plugged. The epoxy ring appeared to be in
good condition. The liner was hard, and
similar 10 Portland cement concrete.
One small leak, about one inch from the
rim, appeared in the liner when it was
flooded and pressure applied from below.
Six two-inch diameter cores were cut.
Core 1 was cut at the location of the
above leak, and included the leak. All
cores from this liner were recovered in
two or three pieces, and many showed
horizontal voids, possibly indicating
boundaries between lifts. Core 1, close
to the rim, cut into the lower rim and
broke off, though there was some soil
cement below that point. The hole from
the first core did not drain. Water from
the other five cores holes ran out the
bottom drain. The bottoms of all the
cores and the gravel underneath all
appeared wet, though that- may have been
due to the water used for coring.
Compressive strength of the soil
cement liner after 56 months of exposure
to leachate was the same as for the liner
specimens recovered and tested after one
year of exposure. Permeability tests of
the top and bottom sections of Core 1,
which included a leak detected before
coring, gave K values of 1.2 x 10"^ and
2.7 x 10~^ cm-s"^, approximately an
order of magnitude more permeable than the
molded specimen tested originally. For
Core 3, a similar value was determined for
the bottom section, but the top was less
permeable, 4.3 x 10"' cm-s'l. The
latter value is in the same range as the
values determined for cores of the liner
tested after one year {Haxo and White,
1977).
Asphaltic Membranes
Results on the exposure tests of
asphaltic membranes are presented in
Tables 3 and 4.
Bituminous Seal
The bituminous seal membrane was cast
in place at 425*F on a sand bed covering
the aggregate in the base of the simulator
It was a catalytically blown canal lining
asphalt and was approximately 0.30 inch in
thickness (Haxo and White, 1974; Haxo and
White, 1976). A ring of epoxy resin was
formed around the top edge of the liner.
On examination at the time the
refuse and sand were removed, the canal
lining asphalt membrane had some areas in
which the liner was weak, "cheesy", and
cracked on bending; other areas were still
tough and flexible. When the liner was
examined in the laboratory, after 36 days
of storage in a sealed polyethylene bag,
the differences were less pronounced. Two
weak areas were identified, one near the
rim and one at the free end of the drain
tube. The weak areas were not as tender
as when examined originally. Samples from
the each of the above weak areas and from
"normal" area were removed for determi-
nation of volatile loss and viscosity.
Volatile loss of the sample from the
weakest area was about three times that
of the other two areas. Viscosity of the
asphalt from all three areas was approxi-
mately equivalent. Since the sample was
heated in the preparation procedure for
viscosity determination, any water and
volatile organic materials were driven
off. Reversible effects caused by the
presence of water or other volatiles,
therefore, are not detected in the vis-
cosity test.
The asphalt membrane liner tested
after 12 months of exposure had not
changed in softening point and had in-
creased in viscosity only slightly from
the original asphalt. After 56 months
of exposure, both softening point and
viscosity had increased, indicating that
the asphalt had hardened during exposure,
so far as its properties in the dry state
were concerned. The viscosity increase of
146
-------
TABLE 4. ASPHALT MEMBRANE LINERS AFTER 56 MONTHS EXPOSURE TO MSW LANDFILL LEACHATE
Bituminous Seal Catalytical-blown Asphalt
Weak Weak
(near drain tube) Normal (near rim)
Fabric and
Avg asphalt emulsion
Volatiles, %
47 days at 23*C 1.49
2 hours at 110'C 1.42
Softening point, "C (*F)
Tests on extracted asphalt:
Viscosity at 25'C, MP
At 0.05 S-l 18.5
At 0.001 s-1 162
Shear susceptibility 0.56
Penetration at 25"C
(calculated from viscosity) 27
1.26
1.46
101 (214)
17.8
161
0.55
27
3.71
4.55
15.8
152
0.58
28
17.4
158
0.56
27
7.32
7.95
3.1
3.4
0.03
53
the liner exposed to leachate was less
than for a sample removed from the upper
surface of the same lot of asphalt re-
maining in the fiberboard carton stored
alongside the simulators since they were
constructed, showing that exposure to air
caused a greater increase in viscosity
than did exposure to leachate. While the
measurements made in the dry state indi-
cate good retention of properties, the
very tender, "short" and "cheesy" condi-
tion observed when the liner was first
removed causes some concern for the
ability of the Uner to withstand minor
earth movements when it is in the satu-
rated condition existing beneath a
landfill.
Asphalt Emulsion on Nonwoven Fabric
The emulsified asphalt on nonwoven
fabric was an asbestos filled anionlc
asphalt1c emulsion on a nonwoven polypro-
pylene fabric. This membrane was prepared
by the supplier 1n sheet form. Disk
specimens 22 Inches 1n diameter, were
mounted 1n the bases and sealed with epoxy
resin. The Hner was 0.30 Inch thick.
The Hner of non woven polypropylene
fabric coated with asphalt emulsion had
absorbed nearly 8% water (or other vola-
tlles) during 56 months of exposure
to leachate vs. almost 5% for the Hner
tested after one year exposure. Viscosity
of the extracted asphalt was lower than
the original, but differed very little
from the viscosity after one year.
A patch of a nonshrinking cement
grout, applied to fill the hole left where
a sample was cut from the liner at 43
months, was in good condition, well sealed
to the edges of the hole, and appeared to
be functioning satisfactorily. No seepage
of leachate was collected below the liner
during the 13 months that the patch was in
place.
Polymeric Membranes
Primary Test Specimens
The six polymeric membranes that were
exposed as primary liners 1n the bases of
the simulators were:
Butyl rubber
Chlorinated polyethylene (CPE)
Chlorosulfonated polyethylene
Ethylene propylene rubber
Low density polyethylene (LOPE)
Polyvlnyl chloride (PVC)
(CSPE)
(EPDM)
In view of the Importance of seams in
field performance of a membrane Hner, a
seam was Incorporated across the middle of
each specimen. The changes in the physi-
cal properties and seam strengths of the
147
-------
six primary polymeric membrane specimens
after 56 months of exposure to HSW
leachate are presented in Table 5. The
retention of original values of selected
properties are presented in Table 6.
Overall, the changes in the physical
properties of the membranes resulting from
56 months of exposure were relatively
minor. All of the membranes softened to
varying degrees during the first 12
months, as shown by hardness data in Table
6, probably due to the swelling by the
leachate. In the interval of time to 56
months, the PVC, CSPE, and CPE membranes
rehardened slightly, possibly indicating,
in the case of the PVC membrane, loss
of plasticizer and, in the case of the
CSPE and CPE membranes, crosslinking of
the polymers. They all recovered most of
their tensile properties that were lost
due to the initial softening. These
materials were all thermoplastic and
unvulcanized.
Of the six polymeric membranes, the
LDPE film best maintained original proper-
ties during the exposure period (Table
6). It also absorbed the least amount of
leachate. However, this membrane, which
was 10 mils in thickness, probably has too
low a puncture resistance for use in
lining a landfill. This deficiency was
also observed in the handling of the
sheeting which was used to line the steel
pipes in which the ground refuse was
compacted. "
The butyl and EPDM liners changed
somewhat more in physical properties than
did the LDPE during the exposure period.
In particular, they maintained their
stress-strain properties. Some softening
did occur in the 44 months interval
between the removal of the first set of
liner specimens at 12 months and the
second set at 56 months.
The fact that, except in the bases in
which the epoxy resin sealing ring de-
teriorated, no leachate appeared below the
membrane liners during the exposure period
indicates at very low permeability of the
polymeric membranes.
The strength of the seams of the PVC,
the CSPE, and the CPE membrane liner
specimens sustained substantial losses,
with the losses greater for the CSPE and
CPE membranes. The seam shear strength of
the butyl and EPDM liners initially
increased, but then decreased in strength,
The LDPE retained best its original seam
strength throughout the period; it was
heat-sealed and the test values were low
(the failure was just outside of the seam,
thus indicating potential weakness at that
point).
Huch better seam strength values were
obtained with the buried strip test
specimens, as discussed in -the next sec-
tion.
Strip Membrane Specimens Buried in the Sand
above Primary Specimens
In addition to the six primary
membrane specimens, 42 smaller specimens
of membrane liners and other polymeric
compositions were buried in the sand above
the primary specimens and exposed to
leachate. These additional specimens
allowed us to make comparisons between
materials from different sources, differ-
ent constructions and different thicknes-
ses, as well as exposure to one side and
both sides of the test specimens. During
most of the exposure period, a head of at
least a foot of leachate was maintained on
the surface of the primary liner; conse-
quently, these specimen were essentially
immersed in the leachate with both sides
of the strips in direct contact with
leachate. The buried specimens were in
the form of spirally coiled strips,
2 1/8" x 20". At one end of each a lap
adhesive joint approximately 2" x 2" was
incorporated which could be tested in
peel and in shear. Thus, various adhesive
bonding systems could be tested.
The buried specimens included the
following compositions and variations:
1. Samples of four of the six primary
liner materials which were mounted in
the bases of the simulators with the
same adhesive systems1 plus additional
adhesive systems.
2. Additional liner materials of the same
six polymers, but varying in source,
thickness, and fabric reinforcement.
3. Membranes of additional polymers which
are potentially useful as liners, i.e.
neoprene, polybutylene, and polypropy-
lene.
148
-------
TABLE 5. PROPERTIES OF POLYMERIC MEMBRANES AFTER 56 MONTHS OF EXPOSURE TO LEACHATE GENERATED (N SIMULATED HSW LANDFILL
Polymer
Matrecon number
Nominal thickness, mils
Type of compound0
Thread count, ends per inch (epi)
Analyses:
Volatile*
Extractables (after
volatile! removed)
Solvent
Physical properties:
Tensile strength of fabric, ppl
Tensile strength of polymer, pst
Elongation dt break, X
Set dt break, J
S-100, psl
S-200. psi
S-300, pst
Tear strength. Die c, ppl
Hardness:
Duro A, Instant reading
10 sec. reading
Duro 0, Instant reading
10 sec. reading
Puncture strength:
Thickness of specimen, mils
Stress, Ib
Elongation, in.
Seam strength:
Sean system
Shear, ppl
Locus of failure'
Peel:
Average, ppl
Maximum, ppl
Locus of f«1lurac
Direction
of test
Machine
Transverse
Machine
Transverse
Machine
Transverse
Machine
Transverse
Machine
Transverse
Machine
Transverse
Machine
Transverse
Machine
.Transverse
Butyl
7
63
VZ
2.37
9.76
KEK
...
...
1490
1440
420
385
12
11
320
350
720
780
1160
1210
185
185
57
51
...
...
64
50.0
1.25
Cement
8800
(lab)
17
AO-LS
3.4
3.8
AO
CPE
12
31
TP
7.61
5.09
nC7H16
...
...
2270
1650
360
410
160
160
1060
510
1460
820
1910
1220
180
160
75
70
• • •
...
37
51.8
0.98
Solvent,
50 THF:
SO Toluene
(lab)
10
AO-LS
2.9
4.5
AO
CSPE
6R*
34
TP
8x8
13.9
3.35
OMK
37
35
1890
2330
220
245
50
65
800
845
1670
1980
...
...
185
220
74
70
• • •
...
37
58.2
0.86
Cement
280Z
(lib)
17
AO
1.8
3.2
AO-LS
EPOM
18
51
VZ
5.74
28.27
HEX
...
...
1520
1360
380
370
5
7
350
305
860
735
1240
1200
135
145
54
51
> * -
...
49
41.5
1.19
Solvent 374*
Cement MA6
(factory)
18
AO
7.1
7.6
AO
LOPE
21
11
TP
1.95
3.37
MEK
2610
2560
470
610
350
470
1390
1050
1500
1150
1760
1240
450
360
81
81
34
26
9.8
17.1
1.24
Heat
seal
(Ub)
11
SE
12
12
SE
PVC
17
20
TP
2.08
34.39
ecu*
CH3OH
. • *
...
2920
2560
340
335
55
68
1330
1130
1940
1680
2620
2330
300
270
72
70
...
...
22
31.3
0.84
Solvent 6079*
L-1552
(factory)
22
AD-LS
5.6
7.5
AO-LS
*Fabr1c reinforced.
bVZ • Vulcanized; TP • Thermoplastic.
cLocus of failure code:
AO • Failure «1th1n the adhesive.
AO-LS • Failure between adhesive and Hner surface.
SE • Failure at seam edge.
149
-------
TABLE 6. RETENTION OF PHYSICAL PROPERTIES AFTER 56 MONTHS OF EXPOSURE TO LEACHATE GENERATED
IN SIMULATED MSW LANDFILL
Polymer
Matrecon number
Nominal thickness, mils
Type .of compound
Thread count, ends per inch (epi)
Tensile strength of fabric, X ret.
Tensile strength of polymer, X ret.
Elongation at break, X ret.
Set after break, X ret.
S-100, X ret.
S-200, % ret.
S-300, X ret.
Tear strength, X ret.
Puncture strength:
Stress, X ret.
Elongation, X ret.
Hardness, Duro 0:
Duro A, pt. diff. inst. reading
10 second reading
Direction
of test
Machine
Transverse
Machine
Transverse
Machine
Transverse
Machine
Transverse
Machine
Transverse
Machine
Transverse
Machine
Transverse
Machine
Transverse
Butyl
7
63
VZ
...
103
99
86
112
125
164
91
121
94
128
94
121
103
103
112
102
-2
0
CPE
12
31
TP
...
92
79
120
70
80
70
87
98
80
98
78
102
67
67
110
94
-10
-7
CSPE
6R
34
TP
8x8
99
100
107
145
92
82
64
82
81
94
97
137
...
58
77
169
151
-5
-5
EPDM
18
51
VZ
...
101
94
90
78
38
78
100
87
113
97
111
93
75
80
105
83
-3
-3
LDPE
21
11
TP
...
154
99
147
70
198
70
109
102
102
111
105
111
108
100
128
163
NTC
NTC
PVC
17
20
TP
...
Ill
102
126
88
81
88
106
100
93
90
85
85
121
122
-9
-6
aR = Fabric reinforced.
bType of compound = Vulcanized (VZ)
CNT = no test.
or Thermoplastic (TP).
-------
4. A series of five pieces of gasket
sheeting of different rubber compo-
sitions and molded slabs of two
thermoplastic rubbers.
5. At the time that two simulators were
lifted from the base to inspect the
conditions of the asphaltic membrane
liner specimens, 10 strip specimens of
newly-obtained membranes were placed
in the sand above the asphalt liners
and were exposed for the last 13
months of the test.
Selected results of the tests of the
buried strip specimens are presented in
Tables 7 and 8.
The changes in physical properties of
four of the buried specimens are compared
with those of the same materials as the
primary specimens in Table 7. Inasmuch as
the buried specimens contacted the leach-
ate on both sides, the effects of the
exposure to leachate could be expected to
be greater than that encountered by the
primary specimens which contacted the
leachate from only one side. The results
given in Table 7 show that, in three of
the four buried specimens, the absorption
of leachate was greater than that of the
primary specimens. In the fourth case,
butyl, the absorption was essentially the
same.
with respect to the effects upon
physical properties, the results varied
among the four materials. Overall,
however, the buried samples tended to
soften more than did the primary samples.
The major differences between the two
exposures were in seam strengths of some
of the membranes. This probably reflects
the differences in the preparation of the
test specimens. For example, the adhesion
of the buried CSPE sample was signifi-
cantly better than that of the primary
sample; on the other hand, the reverse was
true of the EPDM specimens.
Although there is variation in the
results, the data indicate that specimens
immersed in a landfill can indicate the
effects of leachate and can be used as a
measurement of the effects of leachate on
liners.
There are variations, however, among
the liners based upon a given type of
polymer, as can be seen by inspection of
the data for PVC membranes (Table 8).
As can be seen from the data, the PVC
membranes vary considerably in initial
composition and properties. For example,
the plasticizer contents of five PVC
membranes, as measured by extractables,
varied from 32.3 to 38.9%. In tensile
strength, the variation among 9 different
PVC membranes was from 2435 to 3400 psi
and elongation ranged from 200 to 385X.
The retention values also varied consid-
erably among the different PVC membranes,
indicating a difference possibly in the
migration of the plasticizer out of the
compounds. These results indicate the
variation among the different PVC mem-
branes and show that recognition of this
should be given in their selection for
specific applications.
Of particular interest, are varia-
tions in the leachate absorption by
different liners of the same type polymer.
In the case of the PVC membranes, the
absorption ranged from 0.37X to 6.75%. In
the case of the CSPE membranes, swelling
ranged from 8.7% to 21%. EPDM varies from
5.IX to 9% in leachate absorption.
IMMERSION STUDY OF POLYMERIC MEMBRANES
The comparable results of the first
year of exposure tests of the buried
strips of the membrane materials with
those of the primary liners, indicated
that the complete immersion of a specimen
in leachate would be a feasible method of
exposing liners for test purposes (Haxo,
1979).
The availability of leachate gen-
erated in the simulated MSW landfills made
it possible to use, for immersion testing,
the same fluid that was used in the
exposure testing in the simulators of the
primary and buried specimens. This
gave us the opportunity to develop
correlation among three types of exposure
conditions.
Consequently, a study was undertaken
to evaluate membrane liner materials only,
as admix materials do not lend themselves
to tests of this type. The basic design
of the immersion system was discussed in
detail previously (Haxo, 1977) and is
shown in Figure 5. Three sets of 28
specimens of each membrane were immersed
in the leachate so that they could be
removed in sets from the system after
three exposure intervals. For this
study, the specimens were removed and
151
-------
TABLE 7.
COMPARISON OF PROPERTIES OF PRIMARY AND BURIED SPECIMENS AFTER 56 MONTHS OF EXPOSURE TO LEACHATE
IH THE HSU LANDFILL SIMULATORS
Polymer
Matrecon numoer
Nominal thickness, mils
Type of compoundb
Thread count, ends per Inch (ep1)
Type of specimen
Analyses:
Volatile* (2 h at 105'C), X
Extractables (after
volatiles removed), X
Solvent0
Physical properties'1:
Tensile strength of fabric, pp1
Tensile strength of polymer, psi
Elongation at break, X
Set at break, X
S-100, psi
S-200, psi
S-300, psi
Tear strength, Die C, ppi
Hardness:
0
Duro A, Instant reading
10 sec. reading
Puncture strength:
Thickness of specimen, oiUs
Stress, Ib
Elongation, in.
Sean strength:
Bonding system
Shear, pp1 :
Locus of failure
Peel:
Average, pp1
Maximum, ppi
Locus o* failure0
Butyl
7
63
VZ
Primary Buried
2.37 2.02
9.76 9.79
KEK
1490 1510
420 435
12 16
320 300
720 680
1160 1090
185 185
57 65
51 S3
64 66
50.0 38.0
1.25 1.14
Cement
8800
d«b)
17 39.5
AO-LS AD
3.4 3.0
3.8 4.2
AD AD-LS
CPE
12
3!
TP
Primary Buried
7.61 10.1
5.09 5.73
nC7H16
2270 2205
360 295
160 165
1060 1165
1460 1665
1910 2185
180 215
75 71
70 64
37 38
51.8 49.3
0.98 0.97
Solvent,
50 THF:
SO Toluene
P«b)
17 na
AD na
2.9 2.0
4.5 2.6
AD AD
CSPE
6R«
34
TP
6 x 8
Primary Buried
13.9 17.0
3.35 3.84
am.
37 41
1980 2620
220 200
50 53
800 1245
1670 2620
185 205
74 70
70 67
37 40
S8.2 57.4
0.86 0.84
C merit
2802
(lab)
10 6.92
AO-LS SE
1.8 16.0
3.2 22.8
AD-LS AD
EPDM
18
51
VZ
Primary Buried
5.74 6.47
28.3 30.0
HEX
1520 1505
380 495
5 15
350 295
860 655
1240 975
135 175
54 55
51 47
49 58
41.5 41.5
1.19 1.42
Solvent 374+
Cement MAG
(factory)
18 14.9
AD AD
7.1 0.4
7.6 0.7
AD AD
*Fabric reinforced.
bVZ • Vulcanized; TP • Thermoplastic.
cSol vent: MEK • Methylethyl ketone; DMK • Dimethyl ketone
"Tensile properties all 1n machine direction.
eLocus of failure code:
AD - Failure within the adhesive.
AD-LS • Failure between adhesive and liner surface.
SE * Failure at seam edge.
Acetone;
n-heptane.
152
-------
TABLE 8. PROPERTIES OF PVC (CH8RAME SPECIMENS BEFORE AND AFTER EXPOSURE TO IEAOMTE IN NSW LANDFILL SIMULATORS
Hatreton nufcer
Simulator mater
Nao
-------
LEACHATEIN
LEACHATE OUT
SPECIMENS
COVER DETAIL
SPECIMENS /
ATTACH TO HOOKS
NOTE:
PLASTIC WELD
WEALS CONTAINER
CROSS SECT ION
LEACHATE IN-*.
-LEACHATE OUT
POLYETHYLENE TANK
Figure 5. Individual polyethylene immersion tank, showing method of holding specimens and
the tank and outlet for the leachate.
tested at intervals of eight, nineteen,
and thirty-one months. The leachate from
the 12 remaining simulators was collected
and blended biweekly and slowly pumped
through the cells holding the exposure
specimens.
The tests performed on the lining
materials before exposure and at three
subsequent intervals were:
- Weight
- Dimensions
- Tensile strength, in machine and
transverse directions, ASTM 0412.
- Hardness, ASTM D2240.
- Tear strength in machine and
transverse directions, ASTM
D624, Die C.
- Puncture resistance, FTM 101B,
Met nod 2065.
- Volatiles at 105*C, ASTM 0297.
Selected.- results are summarized in
Tables 9 and 10.
The buried and primary specimens
generally swelled equally in the first 12
months that they were exposed in the
simulated landfills. In many cases,
effects of eight months of immersion were
equal to the 12 months of exposure in the
simulated landfills. On the basis of the
initial data, it appears that the immer-
sion in the leachate is quite comparable
to the exposure in the simulated sanitary
landfills (Haxo, 1979).
In Table 9, the data on swelling of
the liners during the immersion in
leachate are arranged by polymer type.
The range of values for a given polymer is
indicated for eight, nineteen, and thirty-
one months. The data show that the
154
-------
TABU 9. SUMNARY OF THE EFFECTS OF IMMERSION OF POLYMERIC NEHBRANE LINERS IN LEACHATE FOR B, 19, and 31 MONTHS
Liner material
Butyl rubber
Chlorinated polyethylene
Chlorosulfonated polyethylene
Elastic* zed polyolefln
Ethylene propylene rubber
Neoprene
Polybutylene
Polyester elastomer
Low density polyethylene
Polyvinyl chloride
Polyvinyl chloride * pitch
Number
of
liners
in
tests
1
3
3
1
5
4
1
1
1
7
1
Leachate absorption. <
B mo. 19 mo. 31 mo.
1-2
8-20
13-19
0.1
1-13
1-19
0.1
2.0
0.6
1-3
6
3.5
26-28
16-27
2.0
1-12
3-32
O.B
1.9
1
1-6
B
25
25-28
19-32
8
8-24
5-88
-7
16
3
4-24
14
Tensile strength,
X original
8 mo.
90-97
80-115
82-124
86-94
90-01
69-100'
96-99
99-115
110-180
91-110
92
19 mo.
89-94
81-106
95-132
96-107
86-93
60-102
94-95
52-94
92-149
91-111
88-93
31 mo.
92
78-106
103-138
98-106
94-113
1 68-105
84-97
81-90
118-161
87-117
101-104
Elongation,
X original
B mo.
104-106
64-135
97-107
91-92
76-138
82-103
96-97
101-108
96-181
98-139
109-133
19 mo.
99
76-108
77-94
102
83-146
S'b76-I04
96
92-94
67-192
100-129
94-117
31 mo.
90-92
71-103
69-86
96-98
88-138
78-146
86-89
80-96
100-168
79-120
80-103
Change In hardness,
Duro A, points
8 no.
0
-5 to 1
-20 to -4
0
-1 to t2
-11 to »5
-3
-4
-7
-2 to tl
-2
19 OB.
0
-8 to -2
-26 to -5
-2
-2 to -1
-12 to *3
-5
-6
-10
-6 to tl
*1
31 n».
-1 to *2
-11 to -1
-21 to -3
-1
-3 to *5
-ia to *4
-3
-3
-9 to -8
-6 to »3
*1 to »7
aData on fabric reinforced neoprene liner number 42 were not Included.
bUnexposed material broke at 150X elongation.
-------
TABLE 10. RETENTION OF MODULUS3 OF POLYMERIC MEMBRANE LINER MATERIALS ON IMMERSION
IN LANDFILL LEACHATE
Retention
Liner S-200, psi
Polymer
Butyl rubber
Chlorinated polyethylene
no.
44
12
38
86
Chlorosulfonated polyethylene 3.
Elasticized polyolefin
Ethylene propylene rubber
Neoprene
Polybutylene
Polyester elastomer
Polyethylene
Polyvinyl chloride
Polyvinyl chloride + pitch0
aAverage of stress at 200%
6"
85
36
8
18
83b
91
41
9
37
42b
90
98
75
21
11
17
19
40
67
88
89
52
elongation
Unexposed 31 mo.
685
1330
1205
810
745
1550
1770
970
690
760
845
855
1040
1235
1635
1190
3120
2735
1260
2120
1965
1740
1720
1700
2400
2455
1020
(S-200) measured
670
1260
1250
1080
420
. • •
2300
1100
790
880
880
840
1090
940
1620
1370
3310
2735
1340
2080
1850
1950
1790
1990
2420
2580
of original value, %
8 mo.
85
85
89
98
54
114
77
104
126
110
107
91
100
79
100
ios
101
102
106
87
80
89
91
92
79
96
85
in machine and
19 mo.
90
89
90
106
46
134
108
109
124
109
107
92
99
77
100
il4
101
98
102
85
84
94
91
105
88
95
86
31 mo.
98
95
104
133
57
195
130
113
114
116
104
98
105
76
99
iis
106
100
106
98
94
112
104
117
101
105
transverse di-
rections.
bFabric reinforced.
CS-100 values given; original and subsequent exposed specimens failed at less than
200% elongation.
156
-------
swelling varies from polymer to polymer.
The table also shows the variation of
the retention of tensile strength and
elongation, and changes in membrane
hardness. Also, within a given polymer
type there is variation due to compound
differences. CPE, CSPE, and neoprene
showed the greatest variations within a
polymer type. On the other hand, the
seven PVC liner specimens varied only a
few percent. In all cases, except the
polybutylene, the swelling values for 31
months of immersion were higher than the
previous values.
The effect on the modulus (stress at
200% elongation) of the same materials is
shown in Table 10. The effects appear to
be related to the degree of swell which
the specimens have undergone. Those
specimens which swell little vary little
in tensile strength and modulus. Again,
the PVC membranes have a low spread in
values and retained their original tensile
and modulus. Overall, the polyolefins,
such as polyethylene, polybutylene, and
elasticized polyolefin exhibit the lowest
swelling and highest retention values.
The effect of immersion in leachate
up to 31 months appears to have a rela-
tively mild effect upon most of the
membranes as shown for tensile strength
and retention versus immersion time in
Figure 6.
The changes in properties, e.g.
tensile strength, appear to be generally
related to the amount of swell which the
membranes have undergone during the
immersion. This is illustrated in
Figure 7, in which retentions of tensile
strength have been plotted against the
leachate absorption for the liner
materials.
WATER ABSORPTION STUDY
Since MSW leachate is primarily water
which, in itself, can affect the proper-
ties of polymeric materials particularly
on prolonged exposure, water absorption
studies were conducted using ASTM Test
Method 0570 (ASTM, 1975). Absorption
tests were run at room temperature and at
70°C. The objective of these studies
was to determine the correlation of
swelling by leachate with that by water.
Also, water is a standard reference fluid
for swelling tests. In addition, 1t was
desired to determine whether the swelling
of the liner materials would plateau, that
is, come to equilibrium on extended
immersion times. Results of water absorp-
tion of 11 membrane liners on immersion in
water up to 186 weeks, both at room
temperature and at 70°C are presented in
Table 11. The order of the results for
the two temperatures were generally the
same; however, one of the major di'fferen-
ces was with one of the PVC liners (No.
11), which yielded the lowest swell at
room temperature, but was sixth at 70"C.
The liner materials that had the lowest
swell were the polyvinyl chloride, the
elasticized polyolefin, and the ethylene
propylene rubber; those that swelled the
most were the membranes based on neoprene,
CSPE, and CPE. The CPE and the EPOM
essentially reached equilibrium swell at
70"C in the 100 weeks. A PVC liner
appeared to have reached equilibrium
swell; however, it hardened because of
loss of plasticizer. It appears that
several of the other liner materials,
particularly the neoprene and CSPE, will
continue to swell for considerable lengths
of time.
DISCUSSION
Asphalt-Containing Lining Materials
The 56 months of exposure of lining
materials to MSW leachate indicate that
there may be problems in the use of
asphaltic materials for the lining of
municipal solid waste landfills. In this
study four forms of asphalt were included:
- 60-70 Penetration-grade paving
asphalt used in hydraulic and
pavement concretes
- A catalytically-blown asphalt used
in a membrane
- A slow-curing liquid asphalt used
in soil asphalt
- An emulsified asphalt used in an
asphalt membrane
The permeability of the various
asphaltic liner materials did not increase
and the changes in properties of the ex-
tracted asphalts were small compared
with with the changes that occurred in air
aging of the same asphalt; nevertheless,
all five liner materials containing
asphalt changed significantly in the
properties of the liners themselves,!.e:
1J7
-------
a?
i%
LU
sr
tn
LU
UJ
I-
O
z
o
p
z
LU
UJ
e
130
100
BO
120
100
80
110
100
LOPE NO. 21
NEOPRENE NO. 42
I I I I I
CPf NO. 12
I 1 I I i
90 -
EPDM NO. 8
I I I I I
rNEOPneNENO.37
100
90
110
100
90
110
100
r PVC NO. 11
J i
PVC NO. 67
1 I
I
1000
CSPE NO. E
NEOPRENE NO. 9
CPE NO. 38
I I I I I
EPDM NO. 18
I I I 1 I
r NEOPRENE NO. §0
I i i i I
PVC NO. 17
I I I I I
PVC NO. 88
I
CSPE NO. 3
PVC & PITCH NO. 52
CPE NO. 86
I I I I I
EPDMNO.41
I 1 I I 1
POLYBUTYLENE NO. 98
PVC NO.
I I I I I
PVC NO. 89
I
EPDM NO. 83
CSPE NO. 85
II I I I
PDM NO. 81
POLYESTER NO. 7S
ELASTICiZED
POLYOLEFINNO, 36
PVC NO, 40
1000 C
EXPOSURE TIME, DAYS
1000
1000
Figure 6. Retention of tensile strength of the individual polymeric membranes as a func-
tion of immersion time in landfill leachate. Tensile strength values based
upon the average data obtained in the machine and transverse direction. Liner
numbers and initial tensile strength for each liner are shown. Data are given
for 8, 19, and 31 months.
1S8
-------
I
a
z
LU
_l
If)
m
2
£
o
u.
O
ai
U
DC
Ul
a.
150
125
100
75
1
KEY
O 8 MONTHS IMMERSION
• 19 MONTHS IMMERSION
31 MONTHS IMMERSION
CR9
_L
J_
10 20
LEACHATE ABSORPTION, %
30
Figure 7, Retention of average tensile strength of membrane liner materials during
immersion in landfill leachate as a function of swelling by the leachate.
1. There was major absorption of leachate
oe water which resulted in significant
decreases in the compressive strength
of the concretes and the soil asphalt.
2. The canal-lining asphalt, based
upon catalytically-blown asphalt,
absorbed water or leachate and became
cheesey and lost elongation.
3. The emulsified asphalt absorbed
more water or leachate than did
the catalytically-blown asphalt,
but it appeared to retain its strength
because of the presence of the fabric.
The greater absorption may be due to
the presence of the asbestos and water
soluble salts.
In this study, the exposure period
was 56 months; however, Indications
are that these asphaltic-based liners
would continue to change with prolonged
exposure. The complete loss of strength
and elongation can be expected to have
adverse effects upon the functioning
of lining materials, particularly if any
ground movement takes place.
There are no ready explanations for
the deterioration of all of the asphaltic
lining materials. Possibly, a high salt
concentration In the asphalt may cause
swelling by osmosis. The effect on
concrete may reflect the particular
aggregate that Is used, although the
"atsonvllle granite has not been found to
cause "stripping", which is a problem
encountered with many aggregates when used
in road building. The continuous and
prolonged exposure of these asphaltic
materials to water and to leachate is a
condition which is not normally encounter-
ed in many asphaltic products. Even in
the case of water reservoir linings, the
exposure is often Intermittent. Also,
some asphalt may be more resistant to
exposure to MSW leachate, but specifica-
tions would have to be developed to be
certain that the correct types are used.
Soil Cement Liners
The exposure of soil cement to the
MSW 1-eachate did not appear to have
any adverse effects upon the soil cement
Uner specimens. Actually, the soil
159
-------
TABLE 11. WATER ABSORPTION OF SELECTED MEMBRANE LINER MATERIALS AT ROOM TEMPERATURE AND AT 70'Ca
Water absorbed, ( by weight
Polymer
Butyl rubber
Chlorinated polyethylene
Chlorosulfontted polyethylene
Elastic Ized polyolefln
Ethylene propylene rubber
Neoprene
Pclyestar elastomer
Polyvlnyl chloride
Liner
number
57
77
6
36
8
26
43
82
75
11
59
1
Meek
0.82
1.63
3.44
0.39
0.50
1.20
3.80
2.43
1.07
1.29
1.59
11
weeks
3.22
5.53
6.97
O.S2
1.30
1.84
13.62
8.29
1.05
1.10
2.34
At room
44
weeks
4. SO
10.2
10.9
0.0
1.S6
1.49
37.8
18.5
0.67
0.70
2.43
temperature
100
weeks
6.4
12. $
16.3
4.5
2.25
3.85
75.1
32.1
1.31
1.25
2.98
159
weeks
10.92
16.85
20.81
0.93
3.48
3.61
67.20
4.40
1.61
1.74
4.53
186
weeks
10.76
16.51
19.42
0.38
35.3
3.61
63.11
52.48
0.36
-0.21
1.55
1
week
4.62
15.9
22.1
0.36
1.11
1.44
14.1
8.11
1.28
5.59
4.87
11
weeks
17.54
58.4
131.0
0.45
3.55
4.52
107.0
47.4
1.10
12.13
8.25
At 70'C
44
weeks
53.9
140.0
245.6
0.57
10.8
11.20
240.0
191.4
0.72
39.2
24.0
100
weeks
103.2
1/9.3
370.5
8.7
17.8
17.4
(b)
295.0
0.22
87.4
25.5*
159
weeks
...
178.89
438.16
-4.19
12.59
21.08
(c)
349.95
0.79
126.31
26.17
186
weeks
141.3
117.80
471.40
-8.48
18.53
22.60
id)
(d)
164.28
26.86
•ASTH 0570-63 specimens 1x2 Inches 1n delonlied water.
bSpec1mens began to disintegrate between 44th and 69th weeks.
cSpec1mens have become hard. Indicating loss of plastlclzer.
d!51 weeks of exposure.
Specimen disintegrated.
-------
'cement probably gained in strength and
decreased in permeability during the
exposure. The soil cement liner, after
preparation and cure, could not be cored
and it was necessary to use our laboratory
molded specimens for determining the
properties of unexposed soil cement.
These specimen probably gave high results.
The exposed liner specimens, after 12 and
56 months, yielded good cores which could
be tested for permeability and strength.
It must be recognized that soil cement is
brittle and the specimens that were
exposed in this project were only 22
inches in diameter. Large areas of soil
cement, such as used as bases for highway
pavement, crack every few feet which
would eliminate the use of soil cement for
the lining of large water disposal sites,
which can not tolerate major cracking of
a 1ining.
Polymeric Membrane Lining Materials
Butyl Rubber
The butyl rubber sheeting was vulcan-
ized and unreinforced and had a nominal
thickness of 63 mils. During the 56
months of exposure, it absorbed somewhat
over 2.7% by weight of leachate and did
not appear to change in thickness. The
physical properties of the sheeting
changed little when the data in the
machine and transverse directions are
averaged. The differences between the
machine and transverse directions are not
consistent. Generally, the properties 1n
the machine direction changed downwards
and, in the transverse direction,In-
creased .
During the exposure, puncture re-
sistance increased. The seam strength,
however, was poor at the start, but did
not change during the course of the
exposure. The bonding method was a
low-temperature vulcanizing cement.
Chlorinated polyethylene (CPE)
The CPE Hner was thermoplastic and
not crosslInked, had an initial thickness
of 31 mils, and was unreinforced. During
the 56 months of exposure it absorbed 6.7X
by weight of leachate, but the absorption
had levelled off. The properties general-
ly decreased with the exposure, although
the values are not large. However, of the
six membranes which were tested, this mem-
brane showed the greatest change with ex-
posure time. Also, seam strength
exhibited a substantial decrease in both
peel and shear.
Chlorosulfonated polyethylene (CSPE)
The CSPE sheeting also was thermo-
plastic, was fabric reinforced, and had
an Initial thickness of 32-36 mils.
During the exposure, it had the greatest-
absorption of leachate and when tested at
the end of the exposure period had a
volatiles content of almost 14%. As with
the other liners, the absorption appeared
to have levelled off. The tensile
strength of the fabric, which was reported
to be nylon, did not change during the
exposure period. The base compound,
however, increased in tensile strength,
puncture resistance, and modulus. On the
other hand, the tear strength decreased.
The strength of the seams, which were made
with a bodied cement, decreased sub-
stantially during the exposure. This
results was not consistent with that
obtained on the buried strips which
yielded much higher values.
Ethylene Propylene Rubber (EPOM)
The sheeting of EPDM was based on a
vulcanized compound. It was unsupported
and had a nominal thickness of 51 mils.
After 56 months of exposure, the
sheeting had a volatiles content of 5.75X,
which is equivalent to an absorption of
6.1%. The extractables, after removal
of the volatiles, fell from 31.855 to
28.3SL This quantity of extractables is a
compounding oil, a portion of which was
dissolved in the leachate. The effect
of the exposure upon physical properties
was minor. There was essentially no
change in tensile strength, puncture
resistance,and only modest loss in
elongation and tear. None of these losses
was near 5Q%. The bond strength was low
at the beginning and appeared to have
Improved during the exposure period.
Low-density polyethylene (LOPE)
The LDPE sheeting had a thickness of
11 mils. It was used to-line the simu-
lators as well as being used as a test
specimen in the base of two of the
simulators.
This material had a low volatiles
content but, during exposure, 1t appeared
to have absorbed some leachate. With
161
-------
respect to the physical properties, there
was almost no change during the exposure
period. There was & slight increase in
puncture resistance which was compara-
tively low due to its low thickness.
Polyvinyl chloride (PVC)
The polyvinyl chloride sheeting that
was tested as a primary specimen had a
thickness of 21 mils and was unsupported,
Contnercial sheeting ranges from 10 to 30
mils.
During 12 months of exposure, this
sheeting appeared to absorb 3.555! by
weight of leachate and, after 56 months,
the absorption had dropped to 2.08%. At
the same time, the PVC lost plasticizer.
The effect of the exposure on this sheet-
ing was minor. The tensile strength and
elongation remained unchanged; however,
the S-200 stress decreased at 12 months
and increased at 56 months, possibly
reflecting the difference in absorbed
leachate. This effect was also noted in
hardness and tear strength. The strength
of the seam, which was factory prepared
using a proprietary cement, was low,
but improved slightly during the exposure
period, from 4 to 5.6 pounds per inch in
the peel mode.
Changing Materials and Technology
The polymeric membranes that were
selected when the project began were all
commercially available at the time.
However, several of these are no longer
available, either because the companies
which regularly supplied the materials are
no longer manufacturing the sheeting or
have changed their formulations and
construction. Also, the seaming tech-
niques and materials are undergoing
changes. Consequently, .the data presented
must be used with discretion when applied
to present materials.
ACKNOWLEDGEMENTS
The work reported in this paper
was performed under Contract 68-03-2134,
"Evaluation of Liner Materials Exposed to
Leacha.te", with the Municipal Environ-
mental Research Laboratory of the
Environmental Protection Agency,
Cincinnati, Ohio.
The authors wish to thank Robert
E.. Landreth, Project Officer, for his
support and guidance in this project.
REFERENCES
American Society for Testing and Ma-
terials. 1975. ASTM D570-63 (1972).
Test for Water Absorption of Plastics
- Part 35. Philadelphia, PA.
Haxo, H.E. 1976. Assessing Synthetic
and Admixed Materials for Lining Land-
fills. In: Proceedings of Research
Symposium; Gas and Leachate from Landfills
- Formation, Collection, and Treatment.
EPA-600/9-76-004. U.S. Environmental
Protection Agency. Cincinnati, Ohio.
Haxo, H.E. 1977. Compatibility of Liners
with Leachate. In: Management of Gas and
Leachate and Landfills, Proceedings of
Third Annual Municipal Solid Waste Re-
search Symposium. EPA-600/9-77-026. U.S.
Environmental Protection Agency.
Cincinnati, Ohio. (PB-Z72-595).
Haxo, H.E. 1979. Liner Materials Exposed
to MSW Landfill Leachate. In: Municipal
Solid Waste Land Disposal, Proceedings of
the Fifth Annual Research Symposium.
EPA-600/9-79-023a. U.S. Environmental
Protection Agency. Cincinnati, Ohio.
Haxo, H. E. R. S. Haxo, and T. F.
Kellogg. 1979. Liner Materials Exposed
to Municipal Solid Waste Leachate.
Third Interim Report. EPA-600/2-79-
038. U.S. Environmental Protection
Agency. Cincinnati, Ohio.
Haxo, H. E., and R. M. White. 1974.
Evaluation of Liner Materials Exposed to
Leachate, First Interim Report. EPA
Contract 68-03-2124, unpublished.
Haxo, H. E., and R. M. White. 1976.
Evaluation of Liner Materials Exposed to
Leachate, Second Interim Report. EPA-600/
2-76-255. U. S. Environmental Protection
Agency, Cincinnati, Ohio. 53 pp.
Haxo, H. E., R. M. White, and M. A.
Fong. Final Report on EPA Contract 68-
03-2134. Evaluation of Liner Materials
Exposed to Leachate. In preparation.
Katrecon, Inc. 1980. Lining of Waste
Impoundment and Disposal Facilities.
SW870. U. S. Environmental Protection
Agency, Washington, DC. 385 pp.
162
-------
FIELD VERIFICATION OF LINERS
Assessment of Long-term Exposed Liner Materials
From Municipal Solid Waste Landfills
John G. Pacey
Charles G. Brisley, Jr.
R. Lee Oooley
EMCON ASSOCIATES
San Jose, California 95112
ABSTRACT
The objective of this partially completed project is to study the physical and chemi-
cal condition of clay, asphalt and PVC liners subjected to continued long-term exposure of
leachate and decomposing refuse. This paper discusses the procedures followed in
selecting the initial test sites, the development of sampling methodology to obtain
samples from liner materials buried under refuse fills and from the soils beneath the
liner, and the institutional and legal difficulties encountered. Physical and chemical
testing of collected polymeric samples is still in progress by EMCON Associates and its
subcontractor, Matrecon, Inc.
INTRODUCTION
Of prime importance in disposing of
municipal solid wastes is the protection
of ground water and surface water from the
environmental impact of leachate—liquid
that has percolated through waste and has
extracted dissolved or suspended materials
from it, thereby becoming contaminated.
Using impervious lining materials to
intercept and control leachate offers a
promising means of reducing or eliminating
water degradation from this source, and is
recommended by the EPA for consideration
as a control mechanism.
Traditionally, naturally occurring
clay materials have been used to control
leachate from disposal facilities with an
abundant supply of native impermeable
material. However, frequently it was
found that the native material was not as
impervious as had been assumed, there were
areas of higher permeability than antici-
pated, the production of .leachate was
higher than anticipated, the quality of
leachate was worse than predicted, or the
site was in a more sensitive ground-water
system than the investigations had deter-
mined. Furthermore, available future
sites may not be as impervious as desired.
It is therefore imperative that alter-
native containment materials be available
to permit disposal of waste products and
assure containment of leachate.
Over the past 30 years we have seen
the development of many chemically formu-
lated products which are, for all prac-
tical purposes, impermeable to liquid and
gaseous flows. Among these are: poly-
ethylene (PE), polyvinyl chloride (PVC),
chlorinated polyethylene (CPE), chlorosul-
fonated polyethylene (CSPE), butyl rubber,
and various asphalt mixtures.
Chemical and hazardous waste disposal
and storage facilities have been subject
to increasing review and public scrutiny.
Recently they have also been brought under
severe, but properly deserved, regulatory
review similar to that afforded sanitary
landfill facilities.
163
-------
The issues associated with the selec-
tion and use of lining materials include,
but are not limited to the following:
compatibility of liner material
with the waste products and
environment.
aging characteristics of the
liner
construction techniques
economics
A performance issue with liners is
the possibility of degradation after pro-
longed exposure to leachate. Leachate may
chemically or physically attack liner
materials, thus impairing their integrity.
Unfortunately there presently exists no
standard method of testing to predict the
life of various lining materials. There
is strong circumstantial evidence that
liner life can exceed 20 years; research
to date on liner materials exposed to
landfill leachate has indicated only mini-
mal changes in the properties of liner
materials. Clearly, however, more
information is needed.
The primary objective of this project
is to obtain lining sample specimens from
existing landfills to determine the
changes in the properties of the lining
material as a function of age and to
validate data being developed through
laboratory research. Specifically, the
project consists of two phases, with four
subtasks each. Phase I will be a study of
landfill liners at disposal sites having
clay, asphalt, or PVC liners.
Originally four sites were selected
by EPA for Phase I of the study. These
sites were selected from approximately
30 candidate sites identified by another
EPA contractor. Preliminary approval had
been secured from the site owners by EPA
for use of these sites in the research
project; however, the contract provided
that other candidate sites could be
considered.
If the results of Phase I demonstrate
significant changes in liner characteris-
tics (original over aged), then Phase II
may be initiated. In Phase II, up to six
additional sites will be selected for
sampling and testing. The subtasks are:
(1) obtain final permission from site
owners to conduct sampling, (2) obtain
liner samples and repair the lining
material after extraction, (3) test and
evaluate the samples, and (4) compare
original liner material with the extracted
samples and produce a final report.
Knowledge gained by this study will
be valuable to designers, owners and
operators of future landfills, as well as
to regulatory agencies. It is believed the
results will provide valuable assurance
that lined refuse disposal facilities can
be constructed and operated without
inflicting damage to the environment.
STUDY APPROACH
It was recognized at the start of the
project that the recovery of liner samples
in a sensitive environment would not be an
easy task. The liners had been installed
initially to prevent degradation of the
underground water resources. The sampling
techniques therefore included the replace-
ment or repair of the liner so as to con-
tinue the protective features.
Prior to proceeding with the task of
obtaining final permission to utilize
specific sites, a review was made of
available information on the four selected
sites and other sites throughout the
country. The ideal site would be one with
complete data and records regarding:
(1) the design, installation and perfor-
mance of the liner material, (2) methods
of site operation, (3) types, age and
thickness of waste placed over the liner,
and (4) occurrence, quantity and character
of leachate. Other selection criteria
included: information on soils and hydro-
geology beneath the liner, variety of
types and thickness of liner .material,
interest and cooperation of regulatory
agencies, and attitude of site owner/
operator. Telephone contacts were made
with the originally selected site
owner/operators to obtain more data and to
seek their continued interest in the
project. During this process one of the
owners declined to participate.
Using the above criteria, the review
of site information and screening of
owner/operator attitudes resulted in our
recommending four candidate sites as being
most beneficial to the goals and objec-
164
-------
tives of the study - only one of the
originally named four sites remained as a
candidate. Later, another site had to be
substituted when a municipally-owned site
declined to participate just before the
site visits were to occur.
Task 1 - Obtain Final Permission
Each candidate site was visited by
Mssrs. John G. Pacey (EMCON) and Henry
Haxo (Matrecon) to familiarize themselves
with the site characteristics and explain
proposed liner sampling and repair plans
to the owners/operators. Site information
and operating plans were obtained, where
available; geotechnical and design reports
were reviewed; photographs were taken; and
persons knowledgeable about site opera-
tions were interviewed. Particular
emphasis was placed on obtaining specifics
about the liner material used; types, age,
and thickness of wastes placed over the
liner; methods of operation; hydraulic
head on the liner; quality and history of
leachate in contact with the liner; and
the availability of excavating and
drilling equipment.
Discussion also entailed identifying
regulatory agencies, permit requirements,
public relations, release of information
and legal and liability constraints.
These items are discussed in greater
detail elsewhere in this report.
Based on data obtained from these
initial site visits, the sampling and
repair methodology was developed. In all
cases the exact depth to the liner was
unknown, which meant that the excavations
or drilling operations would have to pro-
ceed slowly to explore for the liner with-
out destroying the integrity of the liner
material. Telephone conversations were
held with drilling contractors in the
vicinity of the various candidate sites to
discuss the feasibility of the suggested
sampling and repair techniques.
After the sampling and repair proce-
dures were developed and approved by the
EPA Project Officer, the final details
were explained to the site owner and
written approval was requested. The
responses and conditions encountered at
this stage of the work are discussed
later.
Task 2 - Obtain Sample and Repair
The intent of this task is to obtain
samples of the various in-place liner
materials and the natural soils beneath
the liner to ascertain any movement of
leachate into or across the liner
materials. Soil samples are also to be
obtained from similar soils outside the
influence of the landfill to serve as
background data for comparison purposes.
Samples are to be of sufficient size to
conduct the tests identified in Task 3.
Samples of the original lining material
and/or specifications are to be obtained
where possible.
In addition, grab samples from
various depths of the refuse are to be
obtained to record the general condition
of the refuse. Leachate, if present, will
also be sampled.
After the liner sample has been
obtained, repairs must be made to the
liner to prevent the downward movement of
any present or future leachate.
Sampling Methodology - Clay Lining
In sampling a clay liner, the pro-
posed methodology involves the use of a
hollow stem auger (Figure 1), boring to a
point approximately 5 feet above the
anticipated depth of liner. A split spoon
sampler will be driven beyond the auger in
18-inch increments, with the auger
following every advancement. The sampler
will be withdrawn after each incremental
advancement to ascertain the refuse/clay
interface. The auger will then be ad-
vanced to a depth of approximately 1 foot
into the liner. Shelby Tubes will be
driven into the clay to obtain samples to
a depth 10 feet below the refuse. After
sampling, bentonite will be placed in the
boring as the auger 1s withdrawn to
prevent escape of any leachate above the
liner.
Sampling Methodology - Asphalt Lining
The asphalt lining at the selected
site is reported to have an 8-inch soil
cover. The proposed sampling procedure
(Figure 2) uses a hollow stem auger and
split spoon sampler, as previously dis-
cussed, to identify the depth of soil
cover over the liner. Once this is deter-
165
-------
IX"
• hollow etain ougtr
's
^. s
Step
• Soil eovtr
Rafuw fill
Laoehat* l***l
downward
Sup S
• Clay litiar
1
)'
r
•K
" ' V_
Standard >ptit »poon
er cb»or>cing
(Jo»n«or(J
Step 2
I :•:;: ,v-.<
1 •- -• - >>«••«
• ,x- °
/**" * *
' < ^x
^P- <|
k
kH
- 5 < ''
<> ^ '
^. «. '
X*"
«
' ' i-L:
<•- -^
\
^ ^,
«, ,
) •- /• ^. ,
////yxy/t'/j
1
N
5
1^
II
N
II
M
ij
I]
fi
II
M
II
It
11
M
ij
II
I)
H
It
II
m^»^'y-y-
'"' \ ' }
.*> ~ - •-
l^f
' ' S -
1
i. , x-
. S ^ '
w \
» ^v ' s
t. >v -^
If 1 " "
il «
1 - - ^
iif ~*~
*i///S///// s
Toin»,
^
W////Z'///"
Step 4
FIGURE 1. Proposed sampling and repair procedure for clay liner
166
-------
8 hollow stim auger
M*$^fef<
.
*
"Q
?>
ty:
N N . '
( ^ .
* i-
r' r- \ '
K.
IN
' ^ f
U " /• '
^ \ ^
X- . '
/ x '
f • * 1 '
,. ^ <
f . , x <
* f^~
;>i^^||rr~8otlcoy'r
"• ' «
/ ) /•
^ _ ' M
i
>y_i _ ~
>o ^ '
s , C f ' .
/ X- /"^ .
s V- '
/"" *~
» / x . ^ f
. ^ •*- *-^
" . ^ ^ 'ft
\ - '-' '.-+±
_»
*•
c - -
' s—D\»9\ae»d leochate -
*" ./ refute miiture
/ '•
V '
< ~ /.
T ^ • ;
S." ^ ~
tmi^m™ ^JMS^ osm®ia£fSSSi*£S3iM — ^jjrriwT^T^^^^BagiSaBiaaa szs;' ssssss. ws,*vi^£&f'JiuMmt^^^^^^U
•'.'• 4't'^'-'.'-'-'-:' \^Sf\a tpoon tompier . Sandy toll '•'''.*'•". .•"••-•.•.' ••'.'• .y-TPumped benfpnlteVconcrete
' •;:'.'.•:/•..••.•.'•••.•••.:.••.•••.'•.'•: •'.•'r • •-.- ••.•::•.-•...: (thick grout ).•.•;•::>'••••'
Step 1 Step 2
( ^
"
i C-^-4"core barrel ., / <
^rr - - - ^ i <• ^
I
f
" I ) "
^ e s~
V
r ^f— Pumped bentonite-
^^x concrete beneath liner
^ r- "" ^ f
• v.:v- •'•'•'.•: 77." ~ ••^T?'- •:[:•':'.:•$
-^•Soii, .eornpto.;.- > :.-;.'-^l:: ;.•:.» j
:*:v';;}1 •''"'•' *" -i'-JV:!.
.l^TT^iTT-'
|--::vi:-.->"
NOTE. ^
Flusft or tolldlfy leoehote above bentonlte - NOTE:
concrete plug prior to liner penetration. Continue to pump ae rode are ilovrfy
withdrawn. Remove eating on completion.
Step 3 Step 4
FIGURE 2. Proposed sampling and repair procedures for asphalt liners
167
-------
mined, a 6-inch casing is installed in the
hole to within 1 foot of the liner. A
bentonite-concrete slurry is injected into
the casing to provide sealing against any
leachate present. This slurry will be
allowed to rise in the casing until its
level equals or exceeds the leachate level
in the refuse. Any leachate above this
bentonite-concrete plug shall be flushed
out of the casing or be solidified before
proceeding. A 4-inch core barrel is
drilled through the plug to retrieve the
liner sample. Soil samples will be taken
with a split spoon sampler to a depth of
10 feet below the liner. Repairs will be
made by pumping bentonite-concrete slurry
in the hole up to the top of the pre-
viously placed plug. Casing will be with-
drawn.
Sampling Methodology - PVC Lining
One of the two sites selected is a
relatively shallow fill; sampling was
therefore accomplished by excavating
through the refuse to the liner with a
small backhoe. Shelby Tubes were driven
into the subsoil to obtain samples at
depths below the liner. The other site
was a sludge pond. Excavation at this
site was made with a front-end loader.
Shelby Tubes were also used for the soil
samples.
Task 3 - Testing and Evaluation
This work is to determine any changes
in the physical and chemical properties of
the lining materials over time. The
characteristics and properties of the
liner soils will be compared with the
characteristics and properties of the
background soils to determine the physical
and mechanical changes with depth.
Polymeric material testing is to
include water permeability, thickness,
tensile strength and elongation at break,
hardness, tear, strength, creep, water
absorption or extraction, puncture resis-
tance and density.
Tests on asphaltic materials will
include permeability, density and voids,
water swell, compressive strength and
viscosity.
Chemical tests on soil samples beneath
the liners will analyze for pH, Hg, Pb,
Zn, Cd, Fe, Cl, COD, Na, NH4, K and Mg.
The intent of this testing is to develop
adsorption data. Soils will also be sub-
jected to the following physical tests:
permeability, density and voids, water
swell, and compressive strength.
Limited chemical analysis will be
performed on leachate samples.
Task 4 - Comparison and Reporting
After testing of sample specimens has
been completed, and original lining
material specifications have been
reviewed, the two sets of data will be
compared. Where appropriate, comparison
with the results of previous and ongoing
EMCON/Matrecon studies will be made. The
objective is to determine the potential
serviceable life of the lining material.
PERFORMANCE OF WORK
In the early discussions with site
owners/operators in the Fall of 1979, it
became obvious that obtaining final per-
mission would be difficult and time con-
suming. The ability to meet local and
state regulations and the upcoming RCRA
"inventory" of sites was uppermost in the
minds of each operator. Their sensi-
tiveness to any potential for adverse
publicity was understandable. Some
owners/operators of both privately and
municipally owned sites refused to par-
ticiate, although they were interested in
the results of the study. Others were
willing to participate provided the data
would be kept confidential and not be
available for use against them by the
public or regulatory agencies. Since
these conditions could not be fully
assured, the time allotted for. the work
effort has been extended and the work
redirected to gain limited knowledge of
liner performance. To date only two sites
have been used to obtain liner and soil
samples. Types of polymeric liners
sampled are polyethylene (PE), chloro-
sulfonated polyethylene (CSPE), chlori-
nated polyethylene (CPE) and polyvinyl
chloride (PVC). In addition, a sample
from a clay material placed above a poly-
168
-------
meric liner is being tested. The
materials sampled have been exposed to an
adverse environment and, possibly,
leachate for up to eight years.
Laboratory testing is underway, and
the results should be available early
1n 1981.
CONCLUSIONS AND RECOMMENDATIONS
Based on a visual examination of the
liners exposed during sampling, there were
no obvious failures of the polymeric
materials, except for some tears
apparently made while uncovering the
samples. It should be recognized that
those areas sampled are small when com-
pared to the total lined area. Since the
laboratory work is not yet complete, no
conclusions can yet be made regarding
physical and chemical changes to the
11 hers.
.. • •. The ability to sample liner materials
is hindered by the same regulations that
encouraged the liners to be placed 1n the
landfill. The owner/operator wants legal
and financial guarantees that: (1) any
adverse Information developed remain con-
fidential, (2) his business Image not be
damaged in any way, (3) he receive full
liability protection from sampling and
excavation activities and accidents,
(4) repairs be made to satisfy his
requirements and those of the regulatory
agencies, and (5) deficiencies found
during the investigation not be used
against the site by EPA, local regulators
or the public. In summary, the
owners/operators believe their sites to be
successful and active business ventures,
presumably operating in compliance with
all existing laws. Participation 1n this
type of project holds the operation up to
public scrutiny and places even the best
operation in possible jeopardy from weak
or incomplete media coverage and unfounded
lawsuits. In effect, the advice from
their legal counsels is that there is
little to gain and a lot to lose by par-
ticipation in the study.
Added to these concerns are the con-
sultants' liabilities in contracting for
work to be performed by local- drilling
companies performing unique procedures for,
the first time. There is also the problem
and cost of proving whether or not any
subsequent contamination of the under-
ground water resources was the fault of
the sampling and repair work, the owner's
activities, or some extraneous condition
or activity. The consultant may also be
subjected to the costs of third party
legal actions beyond his control.
A team effort is needed if we are to
obtain Uner samples at landfills which
contain leachate. The owners/operators,
consultants, and subcontractors and other
participants need to be protected from
unfounded legal actions and financial
risks affecting their livelihood. Perhaps
the EPA can find ways to guarantee con-
fidentiality of information, and provide
funds to help correct deficiencies dis-
covered and to indemnify participants from
legal actions. It is doubtful that any
private landfill owner/operator and most
public entities will grant permission to
sample liners without such incentives.
169
-------
EFFECT OF SANITARY LANDFILL LEACHATE ON THE ACTIVATED'SLUDGE PROCESS
Michael D. Cummins
U. S, Environmental Protection Agency
Cincinnati, Ohio 45268
ABSTRACT
Synthetic sanitary landfill leachate was spiked into a conventionally
designed activated sludge wastewater treatment plant and the process control
problems evaluated. The synthetic leachate simulated high organic strength,
20,000 mg/1 COO, execreted from a young landfill. The monitoring of the
activated sludge process included oxygen uptake rate, mix liquor solids,
specific oxygen utilization, sludge settling characteristic, and sludge
production. The limiting factor in treating leachate was found to be sludge
bulking.
INTRODUCTION
It is estimated that the citizens of
the United States dispose of over 500
billion pounds of municipal refuse each
year. The major method of disposal is by
sanitary landfill. It is estimated that in
the United States there ^exists 250,000
acres of active sanitary landfill. There
are many problems associated with sanitary
landfills, one of which is leachate. As
water passes through a landfill, soluble
material is extracted from the refuse and
forms leachate. The leachate contains or-
ganic acids and heavy metals. If the
leachate were released directly to the
environment, it would be equivalent to
dumping 2 billion pounds of BOD per year
into the nations streams and rivers. The
leachate, If released untreated into the
environment, would obviously cause fish
kills and pollute the public water sup-
plies.
In 1976 Congress inacted the Resource
Conservation and Recovery Act. This
placed pressure on sanitary landfills to
collect and treat leachate. The EPA-
Office of Solid Waste is developing al-
ternative methods of disposing of leach-
ate. One treatment method under invest-
igation is to discharge leachate into a
municipal sewer system and to treat the
leachate with the municipal sewage at the
Municipal Sewage Treatment Plant,
OBJECTIVES
The objectives of this project are to
evaluate the impact of sanitary landfill
leachate on a conventional activated
sludge process and identify cause and
effect relationships between the leachate
addition and process performance.
-------
CHARACTERISTICS OF SANITARY LANDFILL
LEACHATE
The organic constitute of sanitary
landfill leachate results from the bio-
logical decomposition of organic refuse
within a landfill. It has been proposed
that the biological decomposition process
is composed of four phases. (1) The first
phase is characterized by the decomposition
of high molecular weight humic carbohydrate
material. The second phase is character-
ized by the biological decomposition of
free volatile fatty acids and the produc-
tion of compounds containing the carbonyl
group (C = 0) and amino acids. The third
phase is characterized by the decomposition
of the compounds containing the carbonyl
group and amino acids and the production of
high molecular weight carbohydrate com-
pounds. The fourth phase is characterized
by the slow decomposition of the carbohy-
drates produced in the third phase. The
organic material present in leachates from
landfills in the first and second phase are
readily treated by biological processes.
Leachates from landfills in the third and
fourth phase generally require both bio-
logical and physical-chemical treatment.
This research evaluates problems en-
countered in a conventionally designed act-
ivated sludge plant receiving leachate. For
this research, the biologically degradable
high organic strength leachate produced
from a landfill in the first and second
phase was selected. This leachate is char-
acterized by COD's in the range of 20,000
mg/1 with acetic and propionic acid the
major components. Amino acids and com-
pounds containing the carbonyl group are
also present.
EXPERIMENTAL APPROACH
The experimental apparatus consisted
of two parallel 40 liter per minute act-
ivated sludge processes, a leachate storage
tank, and the associated support equipment.
One process functioned as the control sys-
tem, the other as the experimental system.
A predetermined amount of leachate was
metered into the experimental system and
the process allowed to equilibrate. After
the stabilization period the processes were
monitored at a quasi steady-state for four
days, after which time the leachate dose was
increased.
Experimental Apparatus
Both activated sludge systems were
designed using the guidelines from "10
state standards" (3). Each of the act-
ivated sludge units contained three com-
pletely mixed passes as shown in Figure 1.
The liquid volume of each pass was 6030
liters. Because the objectives of this
research were to evaluate the conventional
design and operating parameters of the
activated sludge process, particular atten-
tion was given to process control. The air
flow capacity of the aeration system for
each pass was designed with a safety factor
of two. This safety factor was to be
FIGURE 1. Process control diagram
171
-------
evaluated as the potential limiting para-
meter in treating leachate. For this
reason, the dissolved oxygen was controlled
by two cascade PI control loops. The outer
control loop monitored the dissolved oxygen
in the aerator and determined the desirsd
air flow. The inter control loop adjusted
the air flow control valve to deliver the
desired air flow. This control system
maintained the dissolved oxygen in the mix
liquor at 2.0 ± 0.1 mg/1 (5% accuracy).
The sludge return control system was
designed to be adjustable from 10 to 40
liters per rain. (25% to 100% of main flow).
The pumping system was designed to deliver
the desired flow ± 0.5 liter per minute (1%
accuracy). This was accomplished using a
P.I. control loop which monitored the act-
ual return flow, desired return flow, and
determined the pump RPM.
The primary effluent flow (influent to
the activated sludge process) was con-
trolled by opening the influent valve to the
first pass every 30 seconds, allowing 20
liters to flow into the first pass, and then
closing the influent valve. This procedure
was found to deliver 40 ± 0.1 liter per
minute (0.2% accuracy) to each of the two
parallel units.
This control system was implemented
using a Texas Instruments process con-
troller model number PM 550. The control
system was found to satisfactorily maintain
parallel operation of the two activated
sludge processes during a six month shake-
down period.
The operating parameters of the
activated sludge process were as follows:
Sludge Return 10 liter per min.
SRT 1.5 day
Aeration Time 6 hours (includ-
ing sludge recycle)
Secondary Clarifier
Hydraulic Loading 15 m^ per m2 day
Synthetic Leachate
The first series of studies were
conducted using a synthetic leachate.
The synthetic leachate was designed to
simulate the volatile acids, proteins,
oxidation reduction potential, COD, and
pH of sanitary landfill leachate. The
following mixture was used.
0.15 H sodium acetate
0.15 M acetic acid
0,05 M glycine
0.008 M pyrogallol
The sodium acetate and acetic acid simu-
lated the volatile acid and COD, and buf-
fered the pH at 5. The glycine simulated
the proteins and amino acids. The pyro-
gallol simulated the oxidation-reduction
potential, the COD of this mixture was
22,000 mg/1 which simulated a strong
leachate produced from a young (1 to 5
yr.) landfill. This mixture was spiked
into the 1st pass aerator of the experi-
mental system at 100, 200, and 400 mg/1
COD. Each spike level was run for one
week. This provided several days for the
hydraulic and biological systems to read-
just and four days of quasi-steady -state
operation.
Sampling and LabAnalysis
Samples were collected from the mix
liquor and return activated sludge on
four hour intervals and analyzed for
centrifuge volume, settling volume, and
oxygen uptake rates. Two-day composite
samples were analyzed for TKN, NHj, NOs,
Total P, COD, TOC and VSS.
CONCEPTUAL DATA ANALYSIS
The major evaluation parameters are
the aeration requirements, specific oxygen
utilization rate, sludge production, and
sludge settling characteristic. Labora-
tory studies (2) indicated that sludge
bulking occurred at high organic loadings.
However, air flows and dissolved oxygen
levels in laboratory units typically do
not simulate full-scale units. Thus, the
experimental design required close moni-
toring and control of the above parameters
to simulate full-scale operation.
Aeration requirements
The aeration requirements to maintain
the dissolved oxygen at 2 mg/1 was moni-
tored by orifice plates and differential
pressure transducers. The actual air flow
was compared to the maximum capacity air
flow.
Specific Oxygen Utilization Rate
The specific oxygen utilization rate
(SCOUR) was computed from the oxygen up-
172
-------
take rate of the mix liquor and return
sludge divided by the respective volatile
suspended solids of the sample. This value
represents the oxygen utilization rate per
gm of active bacteria. In theory, this
value would be high in the mix liquor as
the bacteria decompose the organics and
low in the return sludge as the bacteria
undergo endogenous respiration. Further,
the SCOUR of the mix liquor in both the
experimental and control system would re-
main the same as the organic load increased
in the experimental system. This 1s be-
cause the strain at bacteria would remain
the same. Only the bacteria mass would
change. Similarly, the SCOUR of the return
sludge would remain the same.
Sludge Production
The mass of sludge produced per mass of
COD applied to the system should remain the
same.
Sludge Settling Characteristics
The laboratory studies indicated that
sludge bulking occurred at high organic
loadings. This research monitored the
sludge mass flow rate to the clarifier and
the sludge volume index (SVI) to determine
if the sludge mass, or SVI, or both caused
bulking in the clarifier. The monitoring
would also determine at what organic load-
ing the bulking occurred.
DATA ANALYSIS
The data analysis consisted of plot-
ting the five point moving average of the
laboratory analysis conducted every four
hours.
Figures 2 and 3 illustrate the speci-
fic oxygen utilization rate (SCOUR) of the
mix liquor and return activated sludge for
the control system and spike system. The
SCOUR in these plots are expressed as mg of
oxygen utilized per hour by one gm of VSS.
The VSS, for these figures, were estimated
from the centrifuge volume. Figures 2 and
3 illustrates that the SCOUR in the spike
system and control system generally
paralled each other. The exceptions were in
the mix liquor following an increase in
leachate spike. After the leachate spike
was increased the biomass responded with an
increase in growth rate which resulted in
increasing the SCOUR to 70mg/gm-hr. After
about 40 hours of growth the SCOUR returned
to the normal range of 40-50 mg/gm-hr.
Figure 3 illustrates that the SCOUR of the
return activated sludge ranged between 20-
25 mg/gm-hr. This indicates that the return
sludge was stabilized thus demonstrating
that the aeration capacity was not exceed-
ed. These two figures illustrate that the
conventional design parameters for aeration
capacity and mix liquor retention time are
capable of stabilizing a leachate spike up
to 400 mg/1 COD.
Figure 4 presents the sludge volume
index (SVI) of the control and spiked sys-
tems. Just after the 100 mg/1 COD spike was
started the SVI of the spiked system in-
creased to 350 ml/gm. This was as expected
due to the increase in organic loading. The
SVI was then seen to decline during the 200
mg/1 and 400 mg/1 COD spikes. This indi-
cated that the spiked system was adjusting
to the organic loading and the sludge
settling characteristics were improving.
However, during the 400 mg/1 COD spike the
mass of bacteria within the spiked system
increased as shown by the centrifuge volume
in Figure 5. This increase in solids offset
the improvement in sludge settling. The
result was an uncontrolled increase in
sludge volume as shown in Figure 6. At 470
hours the sludge volume filled the clari-
fier and flowed over the weirs thus "bulk-
ing" the clarifier.
Table 1 presents the average value of
the control system and the average values,
after the first 40 hours, of the three spike
runs for various parameters. The data
verify that the bacterial mass (VSS) in-
creased in response to the organic loading
(Influent COD) to maintain a food to mic-
roorganism ratio of 2.4 to 2.5 gm COD-
applied/gm VSS-day. The oxygen uptake rate
of the mix liquor increased from 30 mg 02/1-
hr in the control system to 75 mg 02/1-hr in
the 400 mg/1 COD spike system. This
increase did not overload the aeration
system, however, the aeration system was
observed to be at maximum capacity.
RESULTS
These three short runs were performed
to determine what problems would occur in
the activated sludge process when it is
organically overloaded by sanitary landfill
leachate. The activated sludge process was
found to treat a synthetic leachate spike
of 100 and 200 mg/1 COD but not 400 mg/1
COD. The process "bulked" at 400 mg/1 COD
173
-------
JOO
iachaie .5>pike_icdj.
.200 m g/ 1
i&? 20
100
150
200
250 300 350
Pun fours ("r)
^00
450
500
FIGURE 2. Mix liquor SCOUR with and without leachate spike
550
600
100
BO
60
bo
E
- 20
100
Leachate Spike (C OD)
200 mg/1
100
150
200
250 300 350
Run Hours (Hr)
400
450
500
550
600
FIGURE 3. RAS SCOUR with and without leachate spike
-------
500
400
5
, w 300
+
100
O)
E
3
I
O)
O
o
g-
3
X
's
100 mgC
Leachate Spike (COD)
200 mg/1 ^
'w^\ x-W""\
V 'v"x--
400 mg/1
100
150
200
250 300 350
Run Hours (Hr)
400
450
500
550
600
FIGURE 4. Mix liquor SVI with and without leachate spike
100
150
200
250 300 350
Run Hours (Hr)
400
450
500
550
600
FIGURE 5. Mix liquor centrifuge volume with and without leachate spike
-------
V
E
1000
800
fe > 600
||| ^00
C
E
o
to
200
*
100 mg/1
"""1
^''"""N f """"*••, *•*
V
200 mgfl
r-x
^'"" '
___ 400 mg/1
^./
B ulking
0 50 100 150 200 250 300 350 400
Run Hours (Hr)
450
500 550
600
FIGURE 6. Mix liquor settling volume with and without leachate spike
-------
Table 1. Analysis of Leachate Spikes
Influent: COO (mg/1
TSS (mg/1)
Mix Liquor: TSS (gm/1)
VSS (gm/1)
Non VSS (gm/1)
Cent. Volume (X)
Settling Volume (ml/1)
0? Uptake (mg/l-hr)
SVI (ml/gm)
SCOUR Mg 02/gm VSS-hr)
f :M (gm COD App/gm VSS-day)
Solids Production (gm TSS/gm COO App)
RAS: TSS (gm/1)
VSS (gm/1)
NON VSS (gm/1 )
Cent. Volume (X)
02 Uptake (mg/l-hr)
SCOUR (mg 02/gm VSS-hr)
RAS TSS:ML TSS Ratio
Clariflers:
Sol Ids Loading (Kg/M2-day)
Sludge Volume Loading (m-vm^-day)
Sludge Volume Flow:
To Clarlfier (X Q)
From Clarlfier (RAS + HAS)(XQ)
Effluent: COD (mg/1)
TSS (mg/1)
Control
670
290
1.55
0.85
0.70
2.0
215
38
140
45
2.5
0.4
6.6
3.6
3.0
9.5
80
22
4.2
29
4.0
27
30
108
14
Leachate Spike
(mg/1 COD)
100 200 400
700
290
1.68
1.00
0.68
2.8
735
46
440
46
2.4
0.4
8.0
4.8
3.2
14
115
24
4.8
31
14
92
30
135
10
870
290
1.94
1.15
0.79
3.1
705
55
360
48
2.4
0.4
8.4
5.0
3.4
17
150
30
4.3
36
13
88
30
85
8
1070
290
2.28
1.45
0.83
4.5
877
75
385
52
2.4
0.4
10.9
7.0
3.9
25
180
26
4.8
42
16
110
30
200
86
-------
spike due to an increase in sludge volume.
The major problem was a combination of
increase in sludge mass and degradation in
sludge settling characteristics that re-
sulted in the sludge volume overloading
the clarifier.
Sludge Settling Characteristic
The sludge settling characteristic were
monitored using SVI. Upon starting the in-
itial 100 mg/1 COD spike the SVI increased
from the normal 100 to 200 ml/nig range to
350 ml/gm. The SVI was seen to improve
throughout the 560 hour test period. The
length of this test period was not suffi-
cient to restablize the SVI. Thus, the
higher loading may have been treatable if
the activated sludge system was given time
to acclimate at the lower organic loadings.
Aeration Requirements
The oxygen uptake rate at the mix
liquor in the 1st pass of the experimental
system increased to 75 mg 02/1-hr compared
to 38 mg 02/1-hr in the control system. The
conventional design for an aeration systems
treating Cincinnati Mill Creek Sewage was
capable of meeting this increase require-
ment. However, aeration systems designed
to treat- a lower strength sewage than the
above may not have the reserve capacity.
The increase in oxygen uptake rate was seen
to be 10 mg Og/l-hr per 100 mg/1 COO spike.
This increase in oxygen uptake rate should
be considered in evaluating other aeration
systems.
Specific Oxygen Utilization Rate
The specific oxygen utilization rate
(SCOUR) was seen to remain relatively con-
stant in the mix liquor and return sludge.
The SCOUR generally ranged from 40 to 50 mg
02/gm VSS-hr in the mix liquor and ranged
from 20 to 25 mg Og/gm VSS-hr in the return
sludges.
Solids Production
The mass of solids produced per COD
applied was seen to remain unchanged at 0.4
gm TSS/gm COD as the organic loading in-
creased.
RECOMMENDATIONS:
The following topics require further
research:
Determine if the SVI does recover if
the activated sludge system in given
time to acclimate.
o Continue the above research using
leachate from a sanitary landfill.
. Evaluate different process control
strategies in treating the high
organic loadings.
REEERENCES
1. Chian, E. S. K., and F. B. DeWalle.
1977. Evaluation of Leachate Treat-
ment Volume I: Characterization of
Leachate. EPA-600/2-77-18a, U. S.
Environmental Protection Agency,
Cincinnati, Ohio 45268.
2. Chian, E. S. K. and F. B. DeWalle.
1977. Evaluation of Leachate Treat-
ment Volume II: Biological and
Physical-Chemical Processes. EPA-
600/2-77-186b, U.- S. Environmental
Protection Agency, Cincinnati, Ohio
45268.
3. Recommended Standards for Sewage
Works. 1978 Edition. Health
Education Services, Inc., P. 0. Box
7126, Albany, N.Y. 12224.
178
-------
CONTAINMENT OF HEAVY METALS IN LANDFILLS
WITH LEACHATE RECYCLE
Frederick G. Pbhland, Joseph P. Gould,
R. Elizabeth Ramsey, Bruce J. Spiller and Winston R. Esteves
School of Civil Engineering
Georgia Institute of Technology
Atlanta, Georgia 30332
ABSTRACT
The presence, transport and ultimate fate of heavy metals in landfill leachate are
discussed with respect to the opportunities for physical-chemical interaction within the
landfill environment. Most leachates were found to contain high concentrations of dis-
solved ions and a broad array of organic and inorganic ligands which, coupled with the
anaerobic reducing conditions of the landfill, favored the removal of heavy metals pri-
marily by precipitation as sulfides. This removal was enhanced by the increased
stabilization rates and filtering action promoted by the recirculation of leachate
through the landfill mass.
These observations were fortified by data obtained from analysis of leachate samples
from four simulated landfill lysimeter columns operated under conditions of leachate
collection and recycle and with codisposal of residential-type refuse and metal plating
sludge. Utilizing equilibrium concepts modified by considerations for ionic strength and
by activity corrections, the potential for precipitation and/or complexation of heavy
metals and the profound impact of sulfides on these reactions were described.
INTRODUCTION
Most landfills receiving waste
materials and an influx of moisture will
provide unusual chemical environments for
generation of leachate containing high
levels of dissolved, ionic species and a
complex array of organic compounds contain-
ins a variety of functional groups. In
addition, depending on a number of factors,
leachates can exhibit highly oxidizing or
highly reducing tendencies which may
further influence the potential hazard of
heavy metal contaminants. Since heavy
metals such as copper (Cu), cadmium (Cd),
nickel (Ni) and mercury (Hg) are considered
priority pollutants, knowledge of their
chemical behavior and relative mobility or
reactivity in landfill leachates is of
major concern. This paper will provide
additional data and evaluation concerning
the influence of chemical activity and
cocplexation on the behavior of heavy
.-netals in landfill leachates.
LEACHATE SOURCE
While a small amount of data derived
from literature reports of other investi-
gators have been analyzed herein, the bulk
of the data have been obtained from studies
by the authors. The leachate used in this
evaluation was obtained from four simulated
landfill lysimeter columns being used to
study the codisposal of metal plating
sludge with municipal refuse under the in-
fluence of leachate containment and recycle.
Accordingly, one column contained residen-
tial-type refuse, the other three, refuse
plus varying amounts of sludge containing
zinc (Zn), chromium (Cr), cadmium (Cd) and
nickel (Ni) from a local electroplating
Industry. The columns are operationally
similar to the units indicated in Figure 1
and have been described in more detail
elsewhere with some preliminary observa-
tions regarding metal solubility and
oxidation state conditions 1»2. Operation
of these lysimeters with leachate recycle
179
-------
TABLE 1. SELECTED LEACHATE CHARACTERISTICS FROM LYSIHETER COLUMNS*
Parameter
Analyses
Column 1
Column 2
Column 3
Sample obtained 300 days after loading the columns.
Control without metal sludge. Column 3:
Column 4
PH _j_
Conductivity, unho cm
Chlorides, mg/l
COD, mg/i
TOO, mg/i
Ortho-Phosphorus, mg/i
TKN, mg/£as N
Ammonia, ng/i as N
Sulfide, mg/£
ORP, E , mV
Cd, mgh
Ni, Bg/t
Volatile Acids, ag/i as
Total Carboxyl, mg/t as
Aromatic Hydroxyl, ag/t
CH»COOH
CinCOOH
as Tan^c
Ainino Acids, ng/i as Leucine
7.0
2,900
420
890
300
7
60
6
0,9
-240
0.1
0.1
250
1,400
40
200
6.2
4,000
470
5,120
1,800
3
100
6
0.9
-120
0.2
0.4
4,100
6,500
60
500
6.2
6,000
520
8,100
2,000
6
80
6
0.9
-70
0.6
0.5
3,800
7,600
80
700
6.4
5,100
680
5,400
1,900
6
90
5
0.8
-60
1.3
0.8
4,800
3,200
70
800
Column 1:
Column 2:
Test unit with 33.6 kg of
metal sludge.
Column 4:
Test unit with 65.8 kg of
aetal sludge.
Test unit with 135.2 kg of
aetal sludge.
Lsachate Rectrculalion
Line
Leachate Distribution
System
op Soil
Aggregate
Sot Id Waste/Industrial
te Mixture
Corrugated Steel
Pipe
Leech ate Collection
Sump
Support Structure
TEST UNIT
CONTROL UNIT
Figure 1. Simulated Landfill Lyainetcr Coltsans with Lcachate Recycle.
180
-------
has been continued during the past year
and leachate samples have been analyzed
for inorganic constituents as well as
several of the organic functional groups.
ANALYTICAL METHODS AND DATA
The leachate from the columns has been
analyzed for inorganic components by con-
ventional methods. Metals have been ana-
lyzed by digestion and atomic absorption
spectroscopy, chloride and sulfide by
electrochemical techniques, nutrients by
Autoanalyzer methods, and ORP, pH and con-
ductivity by the usual instrumental meth-
ods. Analysis of organic constitutents
included direct aqueous injection gas
chronatography for volatile organic acids,
the photometric ninhydrin method for amino
acids3, the dinitrophenylhydrazine method
for carbonyl compounds^, the folin-Dennis
method for aromatic hydroxyl groups^ and
the hydroxylamine colorimetric method for
carboxyl groups**. A typical range of
analytical results for this leachate after
300 days since loading the columns is given
in Table 1; overall changes in selected
analytical results are indicated in
Figures 2 through 8.
The equilibrium constants used for
analysis herein have been compiled from
Sillen 7»8. Pertinent constants for the
various inorganic species chosen are
summarized in Table 2.
CHEMICAL ACTIVITY
Chemical activity is, in effect, a
measure of apparent rather than actual ion
concentrations. Moreover, representing as
it does the result of interactions among
dissolved ionic species, activity is a
function of the ionic strength of the so-
lution and is expressed as dimensionless
activity coefficients. These coefficients,
which represent that fraction of the actual
ion concentration which exists as active
free ions, are calculated by means of
several semiempirlcal equations and are' In-
variably less than one. Therefore, the
activity coefficient provides a measure of
the extent to which an ionic species will
participate in a given chemical reaction.
Obviously, if an activity coefficient
deviates significantly from unity, simple
concentration-based chemical calculations
will be seriously In error.
Typical activity coefficients in non-
saline natural waters and many dilute
wastewaters tend to be sufficiently near
unity that the impact of activity on
chemical behavior in these systems can
generally be neglected. By contrast, the
high dissolved solids content of most land-
fill leachate results in activity co-
efficients substantially smaller than unity
particularly in the case of polyvalent ions,
Therefore, it is appropriate to explore the
effect of ionic strength on the chemical
reactivity of leachate constituents.
Computation of Ionic Strength in Leachates
Computation of activity coefficients
is dependent on a knowledge of the ionic
strength of the medium involved. Class-
ically the ionic strength is calculated on
the basis of a comprehensive qualitative
and quantitative inventory by means of
Equation 1
9
V - % Z (C.Z. )
1 i i
where: y - ionic strength
(1)
•C. • concentration of species, i
Z. - charge of species, 1
Such a comprehensive inventory of ion
concentrations is relatively simple to
obtain in natural and drinking waters as
well as in some wastewaters. Landfill
leachates, on the other hand, present a
far less accommodating matrix in which to
carry out the necessary analyses. Such
factors as high levels of competing sol-
utes, intense color and lack of any com-
plete qualitative description of leachate
composition renders the foregoing method
of ionic strength determination subject to
very significant levels of uncertainty.
(Indeed, high levels of Ionic strength will
tend to make measurements, such as those in-
volving specific ion electrodes suspect).
Therefore, the work reported herein has
been based on well established correlations
between Ionic strength and the easily
measured property of electrolytic conduct-
ance. 10-12.
The linear relationship between ionic
strength and conductivity, y • 1.6 x 10 x
conductivity, demonstrated Independently by
Llnd10 and by Russell12, was used with
leachate conductivity data collected over
a period of nearly 400 days. The resulting
181
-------
TABLE 2. SELECTED EQUILIBRIUM CONSTANTS FOR INORGANIC SPECIES
Species
Constant
Log of Value
K,S
2
H,S
2
4.
NH.
4
NiS
Ag2S
Hg S
2
Hg (OH)
2 2
Cu (CO ) (OH)
3 32 2
Cr(OH)3
AgCl ~
2
HgCl4~2
Hg.Cl,
Cu(NH3)A*2
Hg(NH ) +2
3 2
Hg (KH ) +2
2 32
2n(NH3)/2
K
a^
R
7
A
K
a
K
S
0
K
K
B
O
K
s
o
K
s
o
K
s
o
6 ,
2
BA
B2
B4
8 .
2
8.
2
"<
- 7.0
-13.7
- 9.3
-23.8
-50.2
-52.5
-23.2
-45.4
-30.5
+ 5.2
+15.9
+12.3
+13.0
+17.8
+17.5
+ 9.3
Note: 6 IB Che constant for Che reaction between a metal, M,
n
and a ligand, L, or M + nL
ML K -8
n n
B&O
I I I
J I I
J I (
I I
40 IftC lid 700 J JO MO J*0
_ ^ ^ O&V* '
TO9C WBCI LOADUM .
Figure 2. COD as a Function of Time since Loading Columns.
182
-------
Figure 3. Sulfide Concentration as • Function of Time since Loading Columns.
two
IMO
r
5"*
MO
MO
I I I
1111
1 I
I I I I I I I I I I I I I
Figure 4.
0 JO100 ISOpO 350 MO MO
TMt MMCI IOAOMM.1UVI
Chloride Concentration as a Function of Time since Loading Columns.
Figure S. Cadaium Concentratioo as a Function of Tiae »ince Loading Columns.
183
-------
r-.'W,
Figure 6. Nickel Concentration as a Function of Time since Loading Columns.
too IM
Teas omen
I I I I I I
»wno.l&rt *x> "°
Figure 7. Oxidation Reduction Potential as a Function of Time since Loading Columns.
I I I I
to 71 KX IIS ISO 17) 100 JJS ?W J7« JOC tu UO 174
Teao Macs MAomft. BAVO
Figure 8. pH as a Function of Tine cince Loading Columns.
184
-------
Ionic strengths are plotted in Figure 9.
The data for Columns 2 through 4, contain-
ing the metal sludge, have been averaged
since no obvious trends of differences
were apparent betveen them. Therefore, It
was concluded that the resulting ionic
strength data, while certainly not as
precise as those which might be calculated
on the basis of exact analytical deter-
minations, were probably subject to less
error than might be expected from any
specific analyses in the complex leachate
matrix.
Further examination of Figure 9 shows
some interesting trends which will be re-
flected in the calculated activity co-
efficients. In all columns, the trend of
leachate ionic strength with time followed
the same pattern; high initial values were
followed by a steady decrease to minimum
values 30-50 percent of the initial value
after about 200 days. Following this
minimum, a slow increase in ionic strength
was observed with recovery of a substantial
fraction of the decrease of the ionic
strength by Day 400. For .Columns 2 through
4, there was a further suggestion of a »
decrease in the Ionic strength after Day
350; this decrease did not appear for
Column 1.
While an unequivocal explanation for
the observed trends cannot be advanced, it
seems likely that the initial decrease in
V was a response to the washout of easily
mobilized ions such as sodium (Na+),
chloride (Cl~) and sulfate (SO ~2) combined
with such factors as the conversion of
sulfate to sulfide under Increasingly re-
ducing conditions consequenced by anaerobic
biological activity. The subsequent pre-
cipitation of the sulfide as heavy metal
sulfides would tend to withdraw signifi-
cant Ionic strength from solution. The in-
crease in ionic strength after 200 days
may have been due to some reduction in
leachate volume due to evaporation and/or
the gradual release of less mobile ions
from the landfill materials as they decom-
posed. In addition, the generation of
volatile carboxyllc acids by biological
activity at pH levels over six would also
add anlons and contribute to the ionic
strength. Other products of the decom-
position of organic matter such as amino
acids and phenolic compounds probably also
contributed to the rising trend in ionic
strength.
The late decrease in leachate Ionic
strength In Columns 2 through 4 may have
been the result of a corresponding de-
crease In pH as shown in Figure 8. This
decrease became especially pronounced after
250 days. While the hydrogen ions make a
negligible direct contribution to u at pH
values higher than four, by altering the
distribution between the ionized and
neutral forms of acids and bases, small
changes in the concentration of hydrogen
ions will be reflected in major changes in
p.- In this specific case, the observed
decrease in pH from =6.2 to -5.5 when
applied to a total volatile acid concentra-
tion of 2,200 mg/4. as acetic acid is
sufficient to account for 30 percent of
the decrease in u based simply on con-
version of acid anions to neutral acids.
Similar pH dependant transformations of
other less fully defined acidic species
might well have accounted for the remainder
of the decrease in y. It is Interesting
to note that Column 1, which experienced
no important decrease in leachate pH after
about 200 days, also experienced no late
decrease in ionic strength.
Another factor likely having an
important impact on ionic strength is bio-
logical activity in the columns. An exam-
ination of Figure 9 indicates that, the
control column showed a smooth decrease in
ionic strength to a stable level after
about 160 days, suggesting that biological
activity had largely ceased in Column 1.
On the other hand, Columns 2 through 4
displayed an initial decrease In ionic
strength followed by an apparent retarda-
tion between 60 and 160 days. This period
of retardation probably reflects a
corresponding inhibition of biological
activity by the heavy metal sludge. The
increase in ionic strength In Columns 2
through 4 after 200 days probably reflects
to some extent an upswing In biological
activity resulting In increased cbncentra-
tlons of acid anions in the leachate.
Finally, increased acid production In
Columns 2 through 4 leads to the very late
drop in pH 'and corresponding decrease in
ionic strength.
Examination of the ionic strength
data yielded little concrete evidence of
systematic trends among the columns to
which the metal sludge was added. All
three of these columns produced leachate
with ionic strength values which were
consistently significantly higher than la
185
-------
f"
r
tM IM 800 It) MO «M
Figure 9. Ionic Strength as a Function of Time since Loading Columns.
I'4
I I I I I I I I I I I I I I I
I I I I I I I
1H 900 UO
TGtl MOCI LOAOOCO, <^0
Figure 10. Activity Coefficients for Mono-, Dl- and Trivalent lor.s as a Function of
Time since Loading Columns.
I I I I I I I I I I I I I I I
O.K
9
*•• ColMa t-4
TSBI »«SC«
••»•
Figure 11. Activity Coefficients for Div&lent Ions as a Function of Time since Loading
Columns .
186
-------
the control column. This was expected due
to the addition of Ionic contributions
originating with the waste metal plating
sludge.
Estimation of Activity Coefficientsin
Leachate
The estimation of activity eoeffic-r
ients, especially In systems of very high
Ionic strength, presents special diffi-
culties. The most common expression for
activity coefficients, the DeBye-Hlickel
equation' as given in Equation 2 applies
only up to values of tt of 5 x 10"-%.
-log
0.5 Z
2 1/2
coefficients deviate markedly from unity.
Thus, ions with a charge of one have co-
efficients of approximately 0.83, divalent
ions have y values of about 0.43 and tri-
valent ions have coefficients in the range
of 0.1 to 0.2. It is also apparent from
Figure 10 that y is quite insensitive to
variations in leachate composition.
Figure 11, an expanded scale plot of
activity coefficient data for a divalent
ion, provides a more detailed representa-
tion of the trends which do occur.
The activity coefficients display the
expected inverse relationship to ionic
strength as shown in Figure 9.
where Y« * the activity coefficient of
ion, X.
Z, - the charge of the Ion
V • the ionic strength
Since leachates usually have ionic .
strengths well in excess of 5 x 10 M,
this equation is Inapplicable for such
systems. However, two extended approxima-
tions are available which afford estimates
of Y at substantially higher values of y.
These are the extended BeBye-Huckel ex-
pression' given in Equation 3 and usuable
to y values of approximately 0.1 - 0.15 |i,
0.5
(3)
1 + V1/2
9
and the Davies expression given in
Equation 4 and applicable up to y- 0.6M.
i/a
(4)
-log tl - 0.52,
Since* in the system under study, the ionic
strengths measured la the leachates were
consistently near or below 0.1 M, the ex-
tended DeBye-Hiickel expression was used
for calculation of Yt« Values of f. for
ions of charge one, two and three are
shown as a function of time in Figure 10.
Figure 11 shows the trends of Y as a
function of time for divalent ions on a
more detailed scale.
As is shown in Figure 10, the activity
Impact of Activity Corrections on Heavy
Metal Solubilities
The Implications of the observed
values of the activity coefficients to-
ward the solubility of heavy metals In
leachate are substantial. For example,
it has been demonstrated that the solu-
bility of trivalent chromium in leachate
is controlled by hydroxide equilibria.2
so
(5)
This expression will be modified by
activity coefficient considerations to
so
[Cr+3J[OH~]3 -
OH
(6)
- (7)
Applying values of activity coefficients
reasonably representative of the metal
sludge laden columns as shown la Figure 10,
i.e., YCr+3 " 0.15 and tOR~ " 0.83, leads
to the following corrected expression for
the solubility of trivalent chromium.
fCr*3] [OH~]3 - 11.7 K
so
(8)
Thus, on the basis of tbese considerations,
1C would be expected that the true chromiim
concentration la these leachate systems
I*?
-------
vould be higher than those predicted by
concentration consideration alone by a
factor of -12.
During periods when eulfides were
present in the leachate samples as pre-
sented in Figure 3, sulfide solubilities
would be expected to be similarly effected
then,
by activity considerations. If the
the sulfide system is expressed as
C for
+[HS
r-2
] (9)
an equation for the concentration of
sulfide corrected for activity will take
the form
[ S 1 - C (
C
K K
al a
Again from figure 10, with
YH+ - 0
icEivity
(10)
0.83
on
and Ys-2 - 0.43, the effect of ac
the actual sulfide concentration in
leachate can best be Illustrated by means
of a pC-pH diagram. Figure 12 shows a pC-
pH diagram comparing the concentration of
eulfide ion as a function of pB when
activity is included with the concentration
calculated neglecting activity. As in-
dicated, the sulfide concentration is con-
sistently higher by a factor of almost four
when activity corrections are made. In
situations where metal ions are of concern,
such increased availability of S~* could
dramatically influence their mobility and
ultimate fate since the importance of the
sulfide ion as a precipitant for most toxic
metal ions has been clearly shown.
Correcting metal sulfide solubilities
for activity can also be approached in a
similar fashion. Consider as an example
the solubility of nickel in the presence of
sulfide.
+2
~2
>+2[Ni]-r-2[S~] - K _- 1.7 x 10
"24
Ni
BO
(ID
Again using Yc-2 » Y .,.+2 -0.43
or,
[Ni+2] [ S"2] - 9.19 x 10~24 (12)
pNi+2 « 23.04 - pS"2 (13)
Using the corrected sulfide concentrations
from Figure 12 in this expression yields
the curve shown in Figure 13.
It should be noted that while activity
corrections applied to the solubility
product of nickel sulfide will result in
an increase in the solubility the eleva-
tion of sulfide concentrations shown in
Figure 12 has the effect of decreasing the
nickel solubility. As a result of these
opposing tendencies, the corrected as
opposed to the uncorrected solubility dia-
grams correspond fairly closely under the
conditions chosen. Other systems will have
to be considered on a case by case basis.
Similary, under oxidizing rather than
reducing conditions, where sulfide would
tend to control the metal solubility,
precipitation of copper as carbonates will
be subject to major modification as a
result of activity factors. For example,
one major solid copper carbonate has the
formula Cuj(CO-)2(OH)2. Using the activity
coefficients employed previously, the
solubility product for this species,
corrected for activity,
V* I^lVco.
(14)
would be higher by a factor of approximate-
ly 100. A factor of this magnitude might
well have a major Impact on the chemical
behavior and mobility of copper ions in a
landfill leachata.
While the values of conductance used
herein are fairly typical of landfill
leachates, other researchers have observed
leachates with conductances approaching
20,000 y mhos cm~l. Conductances of this
magnitude indicate ionic strength values
which approach 0.3 M_ and values of -y of
0.74 for nonovalent ions, 0.30 for divalent
ions and 0.065 for trivalent ions. Con-
stants of this magnitude would have a
substantial impact on leachate chemistry.
For example, the solubility of chronduE as
the hydroxide would be Increased by a
188
-------
M «•
It
14
i r
___ Corrected tor
_._ Uncorr«et«d
C,» 0 02M
i i i
• * 7 •
PH
II
i—r
Correct** !•! *cll«|lf
—•-Uncorr«ete«
CfO.OlH
I l 1 I L
Figure 12. Concentration of Sulfide lona as a Function Figure 13. Concentration of Nickel in Solution in
of pH Equilibrium with Sulfide as a Function of pH
-------
factor of approximately 40 times while the
solubility product of Cu3(CO,)2(OH). would
increase about 750 times. Thus, tmaer the
more extreme conditions which might exist
in leachates, activity can be expected to
distort simple concentration based
equilibrium calculations substantially.
Failure to recognize and compensate for the
differences between activity and concentra-
tion in the analysis of leachates may lead
to serious errors in evaluation and inter-
pretation of leachate characteristics.
CHMICAL COMPLBCAT10R
Heavy metals dissolved in aqueous
systems exist not as the free ions but in
combination with other chemical species in
the form of complexes. In complexes,
metal ions combine with non-metallic com-
pounds called ligands by means of coordi-
nate—covalent bonds. For example, a
cupric ion forms a complex with four mole-
cules of the ligand water as shown.
+2
While complexation in natural waters al-
most exclusively involves either water or
the hydroxide ion as the ligand, leachates,
being abundantly provided with such ligands
as chloride, ammonia, phosphate and sul-
fate as well as an array of organic com-
pounds provide conditions under which com-
plexation must be considered in evaluating
the transport and fate of toxic metal ions.
The complexation of heavy metals in
landfill leachates is an area in which
even more work remains to be done. Indeed,
on a conceptual basis, it is easier at this
point to eliminate possible complexing
agents from positions of importance in
determining heavy metal speciation. The
greatest source of uncertainty is the
rather incomplete knowledge regarding
specific composition of leachates.
Inorganic Ligands
Some fairly reliable information is
available on inorganic complexing agents.
The most likely candidates as Inorganic
complexing agents in landfill leachates
are chlorides, sulfates, phosphates and
ammonia.
Effect of Sulfide
Of extreme importance to complexation
is the impact of sulflde solubility
equilibria on the levels of complex which
can exist in the presence of even very
small concentrations of sulflde. This is
a consequence of the very low solubilities
of heavy metal sul fides. In essence,
sulfide very effectively competes with
most complexing agents. Consequently,
many heavy metals will precipitate as
sulfides rather than remain in solution as
complexes. For example, the concentration
of the Hg-CNH-K"*"2 complex in the absence
of sulfide will be controlled by a reaction
with a pK of 5.7
Hg2(OH)2(s)
20H~
while in the presence of sulfide ions, the
equivalent reaction will be
Hg2S(s)
~2
and have a pK of 34.9. Clearly, this
reflects a massive difference in solu-
bilities. Even at sulfide concentrations
as low as 10" 12 molar, Hg2(NH.j) *2 levels
would be predicted to be on the order of
10" 30 molar or less. Concentrations this
low are virtually without meaning beyond
suggesting that the rate at which pre-
cipitation occurs might control solu-
bility of Eg*2 in the presence of sulfide
ions. Otherwise, in those cases in which
inorganic complexes might be significant,
i.e., for Hg"1"2 and Hg+2, the metal sul-
fide solubility equilibria are of such
magnitudes that, in the presence of sul-
fide, metal complexation should be of
little or no importance. Precipitation/
filtration of the metals as the solid
sulfides will control the metal be-
havior totally particularly where leachate
recycle serves to accelerate these pro-
cesses .
Chloride
While, as a general rule the chloride
ion is a rather weak complexing agent, its
effectiveness in complexing the two forms
of mercury is quite high.
190
-------
1 oimiiiiii
Hg2(OH)2(s)^Hg2 + 20H~
Hg2(OH)2(s)^Hg2(OH)+ + OH~
Hg2(OH)2(s) ^SHg2(OH)2(Aq)
Hg2(OH)2(s) + 2CJ^Hg2Cl2+201
pK - 23.2
pK - 15.5
pK - 6.0
1" pK - 10.9
figure 14: Distribution of Hg2 Species
as a function of pH in the presence of
0.019 M chloride (680 mg/1)
Hg"*"24-
HgCl
-2
•»• 2C1,
logS 4 - 15.9
log 8, - 12.3
Thus, in the absence of sulflde, solubility
of Hg*2 will be controlled by the solu-
bility of the neutral Hg.Cl, species at pH
values below about 9 in a system contain-
ing as much as 680 mg/1 of chloride (Table
1). The resulting pC-pH diagram for this
ays tea is shown in Figure 14.
The shaded area of this graph
presents the solubility region for Hg.
In this leachate. As can be seen, below
a pH of about 9 the major soluble species
Is the neutral dlchloride complex IgjCl,.
Thus, in this instance, chloride complexa-
tlon is potentially of major significance
la determining the behavior of a toxic
metal.
Sulfate
Although sulfate is abundantly present
in many landfill leachates, examination of
the literature suggests that, relative to
other available ligands, sulface is simply
too poor a complexing agent to be a major
factor in heavy metal complexation.
It sust be recognized however, that,
while »ulfate has very little Importance
as a direct completing agent, its signif-
icance as a precursor for sulflde under
the Influence of anaerobic conditions and
microbial activity cannot be minimized.
Phosphate
Polyphosphatea tend to be very potent
ligands for many heavy metals. However,
it seems likely that phosphates will be of
only minor importance in most landfill
leachates. Following a rapid early dis-
charge of high levels of phosphate with
values approaching 30 mg/1 of phosphorous,
phosphorous levels in the control column
leachate decreased rapidly to values in
the vicinity of 2 mg/1 followed by a slow
steady increase to values on the order of
6 mg/1 at 300 days (Table 1). The columns
carrying the metal sludge showed no early
peaks of phosphate breakthrough, due most
likely to binding with heavy metals and
the abundance of calcium in the waste
sludge! final levels were similar to those
detected in leachate from the control
column. Since pH values in the leachates
were consistently in the range of 5 to 7,
the dominant phosphate species to be __
expected would be the H2?04~ and HPO," .
An examination of literature regarding
phosphate complex formation constants in-
dicates that, with the exception of the
ferric ion, these species are very weak
ligands for any metals likely to be
present and of concern in typical leachatea
This poor complexing ability combined with
the low (<10~3M) concentrations of phos-
phate In most leachates assures that
phosphates will be of relatively little
significance In control of metal behavior
in landfill leachates.
Ammonia
Except for the control column which
displayed a rapid early washout of ammonia,
ammonia levels were consistently at the
level of ^0,4 mm tt HH--H throughout the
191
-------
studies. This lav value is rendered even
lover by the pH which assures that the
effective ligand, NH,, will be present at
only the 0.55 uM level with the balance
being present as the ammonium ion. Count-
ering this is the exceptional ability
ammonia has to complex certain of the
metals of interest, i.e.,
Cu+2 «•
+2
Hg +
Hg,+2 +
Zn*2 +
Cu(NH_)
3' 4
Hg(NH-),
+2
'+2
4-2
6^
log 6 .
ioge]
logB,
- 13.0
- 17.8
- 17.5
- 9.3
Considering these systems relative to
metal hydroxide solubility in the absence
of sulfide demonstrates that only the
Hg+2 and Hg+2 will complex to a signifi-
cant extent with ammonia under the
conditions normally present in leachates.
For each of these metal species, ammonia
complex concentrations in the range of 10
ujl to 10 mM can be predicted to exist in
equilibrium with the observed ammonia
concentrations. This suggests that, in
the absence of sulfide, ammonia may have
a major impact on the behavior of any
mercury which migh be present in landfill
leachates. On the other hand, copper and
zinc complexes with ammonia will be very
minor, approaching maximum concentrations
of no more than one micromolar at the
highest pH levels.
Carboxylic Acids
Data from volatile acid analyses and
the hydroxamic acid test for carboxyl
groups suggest that high levels of car-
boxylic acids in leachates from the sub-
ject columns were present (Table 1). Since
with the exception of the analysis for
volatile acids, identification of the
specific species responsible for the car-
boxyle group levels is unavailable,
efforts at drawing conclusions concerning
the degree of complexation expected from
these compounds are unlikely to yield much
information of value. However, low mole-
cular weight monocarboxylic acids such as
acetic and propionic acids will generally
play little role in heavy metal complexttaa;
the log values of the metal complex forma-
tion constants for such metals as copper
and cadmium are generally samller than two.
Llgands having such small constants will
simply be unable to compete significantly
with other dissolved substances in
•leachates. Low molecular weight dicar-
boxylic acids such as oxalic and malonic
acids are much more effective ligands;
log values are as high as seven for some
metals. While constants of this magnitude
are substantial, data regarding the con-
centrations of these or other such acids
in leachates are lacking.
Amino Acids and Peptides
Functional group analyses using the
ninhydrin method for amino acids following
hydrolysis indicated that the leachates
studied had amino acid levels on the order
of 5 - 10 mM. If these values represented
the free amino acids, the corresponding
level of complexing potential would be sub-
stantial since, for such elements as
copper and nickel, the logs of the complex
formation constants are on the order of
15 or high enough to be significant. How-
ever, it is more likely that much of this
amino acid was present as polypeptides of
varying size. Since the formation con-
stants tend to decrease as the number of
amino acid units increases ^' , it is
probable that the log of the formation con-
stant for the actual chemcial entities
present might be as low as about 10 for
copper and nickel. Even at this level,
amino acids and polypeptides could play a
significant role in the chemistry of many
heavy metals in leachate. This role would
again be greatly reduced in the presence
of sulfide.
Phenolic Substances
Aromatic hydroxyl compounds, as
measured by functional group analysis, are
major constituents of leachate. Since many
of these phenolic-type compounds have a
substantial ability to complex heavy metals
they could be of great significance in
evaluation and interpretation of heavy
metal behavior in leachates. For instance,
copper forms strong complexes with low
molecular weight phenolic compounds such
as salicylic acid and catechol.
Cu(Salicylic
Cu(Catechol)"
log 62 - 15.7
log B - 41.2
Unfortunately, the nature of phenolic sub-
stances is not well understood beyond the
192
-------
likelihood that they exist as moderate to
high molecular weight substances of the
types generally referred to as humlc sub-
stances. This weakens efforts at drawing
conclusions based on smaller compound
equilibria. However, it is obvious that
the complexing ability of many phenolic
compounds for heavy metals is high and
that the humic substances will share this
complexing capacity. Indeed, some evidence
exists 15 that the logs of the formation
coos cants for heavy metal-fulvic acid com-
plexes are on the order of 4-20. On this
basis, It seems probable that the behavior
or heavy metals in leachates will also be
strongly linked to their Interactions with
aromatic hydroxyl compounds.
SUMMARY AND CONCLUSIONS
Studies on the removal of heavy metals
from landfill leachates have been extended
to further consider the importance of
ionic strength, chemical activity, com-
plexatlon, precipitation and filtration
when the landfill is operated under con-
ditions of leachate containment and re-
cycle. Examination of leachates obtained
from the codisposal of residential-type
refuse with metal plating sludge, indicated
that ionic strengths of these and many
other leachates were sufficiently high so
that equilibium - based predictions of re-
action potential would need to be corrected
for activity. Application of activity
coefficients to the reactions describing
chemical behavior of important leachate
constituents resulted in major modification
is such factors as solubility and the
availability of sulfides for interaction
with heavy metals.
Chemical activity was determined to
be responsive to variations in ionic
strength and pK of the leachate in a pre-
dictable and eonsistant manner and to re-
flect the nature of the concomitant
biological tranformationa occurring ia the
leachate and landfill mass. Moreover,
chemical characterization of the leachate
provided strong evidence that chemical
complexation is potentially a significant
factor influencing the transport and fate
of heavy metals. Several inorganic and
organic Uganda were Identified at
sufficiently high concentrations to exert
a strong influence on the solubility of
one or more of the heavy metals.
Finally, the present studies rein-
forced previous indications that, with
leachate recycle and with even very small
concentrations of sulfides, most heavy
metals will undergo precipitation as sul-
fides and removal by filtration. There-
fore, the use of leachate containment and
recycle not only accelerates stabilization
processes and the establishment of bio-
logically mediated reducing conditions
favorable for sulfide formation, but
provides an in situ physical-chemical
process for Immobilization of heavy metals
sad reduction of potential for external
environmental impairment. Moreover, from
the results of these studies, it may be
possible for many landfills to receive and
detoxify certain types of industrial
wastes without causing irreversibe damage
to either the stabilization process in-
digenous to the landfill or to the
surrounding environment.
REFERENCES
1. Pohland, I. C. 1975. Sanitary Landfill
Stabilization with Leachate Recycle
and Residual Treatment. EPA-600/2-75-
043, 0. S. Environmental Protection
Agency, Cincinnati, Ohio. 105 pp
2. Fohland, ?. G. and Gould, J.P. 1980.
Stabilization of Municipal Landfills
Containing Industrial Wastes. In:
Proceedings of the Sixth Annual
Symposium. Disposal of Hazardous
Waste. EPA-600/9-80-010. 0. S.
Environmental Protection Agency,
Cincinnati, Ohio 242-253
3. Moore, S. and Stein, W.R. 1948.
Photometric NInhydrin Method for Use
In the Chromatography of Aad.no Acids.
Jour. Biol. Chem. 176, 367.
4. Lappin, G.R. and Clark, L.C. 1951.
Colorimetric Method for Determination
of Traces of Carbonyl Compounds
Anal. Chem, 23, 41.
S. Foils, 0. and Denis, H. 1912. OB
Phosphoeusgstlc-Phosphomolybdic
Compounds as Color Reagents, Jour.
Biol. Chem. 12, 239.
6. Montgomery, H.A.C., et. al.. The Rapid
Colorimetric Determination of Traces
of Carbonyl Compounds, Anal. Chem. 23,
41,
193
-------
7. Sillen, L. G. and Martell, A.E. 1964.
Stability Constants of Metal-Ion
Complexes. London: The Chemical
Society, Special Publication No. 17,
753 pp.
8. Sillen, L.G. and Martell, A.E. 1964.
Stability constants of Metal-Ion
Complexes, Supplement. London: The
Chemical Society. Special Publication,
No. 25, 860 pp.
9. Snoeyink, V.D. and Jenkins, D. 1980.
Water Chemistry, John Wiley & Sons,
New York, 463 pp.
10. Lind, C.J. 1970, Specific Conductance
as a Means of Estimating Ionic
Strength, U. S. Geol. Surv. Prof.
Paper 700D 272-280.
11. Langlier, W. F., 1963. The Analytical
Control of Anti-Corrosion Water Treat-
ment. Jour. Amer. Water Works
Assoc. 28, 1500.
12. Russell, L.L. 1976. Chemical Aspects
of Groundwater Recharge with Waste-
waters Ph.D. Thesis, University of
California at Berkeley.
13. Stumm, W. and Morgan, J.J. 1970.
Aquatic Chemistry: Wiley Interscience,
NY 583 pp.
14. Chian, E. S.K. and DeWalle, F.B. 1977.
Evaluation of Leachate Treatment
EPA-600/2-77-186a U. S. Environmental
Protection Agency, Cincinnati, Ohio
210 pp.
15. Reuter, J.H. and Perdue, E.M. 1977.
Importance of Heavy Metal-Organic
Matter Interactions in Natural Waters
Geochim. et Cosmocbim. Acta 41, 325.
194
-------
OPTIONAL COST MODELS FOR SOLID WASTE DISPOSAL
J. Hudson, P. Deese, R. Burke
Urban Systems Research and Engineering, Inc.
ABSTRACT
This paper presents our findings from an analysis of 45 landfills, and associated
transfer station, balers, shredders, and transportation networks. The analysis of the
site attempted to determine how much it costs to build and operate a landfill and which
factors have the greatest impacts on those costs. The studied landfills ranged in size
from under 100 to over 5,000 tons per day and were located across the continental United
States.
A primary concern of the study was to determine the reduction in landfill costs from
waste baling or shredding. Analysis of the data indicates that the savings incurred at
the landfill, because of prior baling or shredding, does not compensate for the added
cost of the more sophisticated processing facilities for the average case. On the other
hand, baling or shredding may be feasible in a situation where component costs for the
entire landfilling system are exorbitantly high.
Another important finding is that landfilling is only a small portion of total costs.
On average, both the haul and the processing components of the system are larger than the
landfilling component. The following paper presents the findings from our analysis, the
implications of those findings and a methodology designed for use by local communities in
determining their most cost-effective landfill system.
How much does it cost to build and
operate a sanitary landfill? Which factors
have the greatest impacts on cost: size,
location, ownership, depth to groundwater,
baling or shredding?
Those two questions were the basis
for a recent study of landfill costs,
performed by Urban Systems Research and
Engineering under contract to the Environ-
mental Protection Agency, Municipal
Environmental Research Laboratory. The
study analyzed costs at 45 landfills, and
associated transfer stations, balers,
shredders, and transportation networks.
The landfills ranged in size from under
100 to over 5,000 tons* per day and were
*"1'ons" in this report are metric tons
1000 kg; cost per ton is about 91% of the
cost per ton.
located across the continental U.S. Data
on current landfill capital and operating
costs has been used to develop estimating
procedures, which may be useful in evaluat-
ing alternatives for your next landfill
site. This article describes the sites,
the data collection procedures, the results
and how the results can be used.
1.0 The Sample of Landfills
In choosing sites, five basic condi-
tions had to be met on available data:
o the system had to have, and use
scales,
o the site had to have been in
operation at least one year,
193
-------
the operator had to be willing
to provide us with both operational
and cost data,
the bulk of the waste stream
had to be residential/commercial,
and
resource recovery, if any, could
only be source separation, and
hand-picking.
Since the results were expected to be
clearest in dollars per ton, the overriding
prerequisite was the existence of scales
and weight records. In addition, because
of RCRA, only landfill sites which state
regulatory agencies considered to have at
least acceptable (and preferably outstanding)
performand records were selected.
A major concern addressed by the study
was any reduction in landfill costs from
waste baling or shredding. For that reason,.
the original research design called for up
to one-third of the sites to be balefills
and one-third to be shredder waste fills.
However, there are only 12 balefill and
nine shredded-waste landfills in the
country which meet the five conditions
listed above. Another objective of the
original research plan was to collect a
significant amount of data from sites land-
filling less than 100 metric tons per day.
Primarily because of the lack of scales at
small sites, the samele is more reoresenta-
tive of medi\nn-scale or larger operations.
After hundreds of phone calls and ex-
haustive use of various site surveys, 45
acceptable sites were selected. Figure 1
shows their location, and Tables I and II
provide a general classification by type
and size.
i s
FIGURE 1: LOCATION OF THE 45 CASE STUDY .SITES
196
-------
TABLE I. SIZE DISTRIBUTION, BY PROCESSING TYPE, OF CASE STUDIES
(in metric tons per day)
Size Type
0-100 100-300 300-600 600+ Total
Direct Haul to Landfill
12
Shredder Processing
24
Baler Processing
Transfer Station
Total:
3 3
1 1
8 11
4 2 12
6 4 12
13 13 45
TABLE II. NAMES OF NON-CONFIDENTIAL SITES
Direct Haul
to
Landfill
Shredder
Processing
Baler
Processing
Transfer
Station
Ft. Lauderdale, FL Onondega County, NY Portland, ME
Hamden, ME
Grunderville Landfill
Warren, PA Guilford County, NC Atlanta, GA
New Gloucester, ME
Spadra Landfill, Los
Angeles County, CA Appleton, WI
Miramar Landfill
San Diego, CA
New Castle, DE
Pullman, WA
Oyster Bay, NY
Smithtown, NY
Columbia, SC
Mission Canyon
Landfill, Loa
Angeles County, CA
Carson City, NV
Marion, MA
Sacramento, CA
Pensecola, FL
Shawnee County, JN
Lycoming County, PA
197
-------
2-0
Collection and Organization
Data were collected from all sites in
January through April, 1980. In an average
case, staff from USR&E were on site for
1-2 days, observing the operation and
collecting financial and operational
data. This was not a detailed engineering
study. No tine and motion analyses were
conducted nor were actual measurements
of cell size of dozer horsepower made.
Instead, the focus was on equipment and
labor allocations and costs. For opera-
ting costs, the most recent fiscal year was
used; for capital costs, data were collec-
ted on the actual equipment in use, age of
the equipment, and purchase prices.
In all cases, the data were organized
according to the disposal process, as shown
in Figure 1. It was hypothesized that, for
example , baler costs would depend on capac-
ity, transportation costs on distance, and
landfill costs on whether the waste was
baled or not. To test these hypotheses,
it was necessary to allocate the costs to
the various parts of the disposal process.
By doing so, the study could differentiate
between costs for a baler and costs for the
transfer truck that took the bales to the
landfill.
The only unusual allocation is the
differentiation between "landfill construc-
tion" and "landfill ing." Landfill sites
cone in all types, . Site preparation nay
range from "none" to "two liners, hundreds
of acre- feet of excavation, a leachate
collection system, and 10 monitoring wells."
These landfill construction costs are ex-
tremely site- specific. TCie costs of
actually placing and covering the waste on
a prepared site, landfilling. are less
site specific, except where cover must
be purchased off-site.
Obviously, landfill construction and
landfilling nay interact. Site preparation
is often phased, occurring simultaneously
with landfilling of waste. Frequently,
equipment is used for landfilling on peak
days and for landfill construction during
off peak tines. Still, with the help of
the site operators, it proved possible to
allocate the costs; this greatly improved
the results.
The final topic to be addressed under
this section is standardization. To com-
pare costs from site to site, it was
necessary to adjust for a wide variation
in labor and capital costs. Mage rates
varied considerably (see Table III) and
fringes and overhead varied even more.
Typical values were chosen for each, and
standardized labor costs for all sites
were calculated using actual hours and
typical hourly rates.
It was more difficult to standardize
capital costs, because of the diversity in
equipment and differences in accounting
practices. However, design lives, deprecia-
tion schedules, and interest rates were set
for fixed and mobile equipment and buildings.
tie assumed no depreciation on land. For
older equipment, either new quotes were
obtained for comparable items or the ENR
Construction Costs Index was used to convert
actual costs to estimsted costs for July
1979 purchase. Standard present worth
tables provided the equivalent annual cost
used in the analysis. These mean assump-
tions for standardization are shown in
Table IV.
3-0 'Descriptive Results
Table V shows the costs for the differ-
ent steps in the process, along with the
relevant sample sizes. The sample sizes
indicate the sites for which there were
complete data collected for each particular
component. In some cases there is complete
data for some components while there was
insufficient data on another component.
For example, it is difficult to estimate
landfill construction costs when all the
excavation is done using whatever depart-
ment has free at the time. On the other
hand, the landfilling data may be adequate.
Based on Table V, it is obvious that
landfilling is usually only a small portion
of total costs. Consider a complete hypo-
thetical system: collection vehicles haul
to a transfer station, larger vehicles
continue to the landfill, and the waste is
landfilled there. If all'costs are at the
average for all sites, the total is $28.95
per metric ton, with landfill accounting
for only $6.88. For this average system,
a baler would add $17.58 but the transfer
station would be replaced; the charge would
increase processing costs by $7.83. Even
if landfill costs were reduced to zero
because of the baler, the systems costs
would increase to $29.90. A baler would
obviously be a bad choice in this particular
situation. On the other hand, if all costs
are "maximum" rather than average, a baler
would be a good investment.
198
-------
TABLE III. 45-SITE AVERAGE WAGE RATES (WITHOUT FRINGES)
Collection
Large-vehicle transport
Processing
Landfilling
Landfill construction
Average
Over
All Sites
5.92
5.74
5.62
5.66
5.83
Lowest Highest
Value Value
3.50 10.62
3.57 8.87
2.90* 8.21
3.27 8.82
4.20 9.33
Sample
Size
20
31
28
42
27
*Below minimum wage.
TABLE IV. ASSUMPTIONS FOR STANDARDIZATION OF LABOR AND CAPITAL COSTS
Labor Assumption;
Standard hourly wage
Standard hourly fringe benefits
Capitalization Assumptions;
Moving equipment
Stationary equipment
Buildings
$ 5.33
33%
5 years 910%
10 years 98%
20 years 98%
199
-------
Vs OOST IN DQLIARS PER METRIC TON
Haul in collection vehicles
collection
operating
Total:
Transfer stations
capital
operating
Total:
Shredder
capital
operating
Total !
Baler
capital
operating
Total!
Haul in large vehicles after
processing
capital
operating
Total:
landfilling
capital
operating
Ratals
Landfill construction
capital
operating
Total:
Sample
Size
20
20
20
10
10
10
7
8
7 •
11
11
11
20
18
18
37
36
36
24
23
23
Average
1.18
5.67
6. 65
2.53
7.22
7.75
4.95
5.63
11.06
9.58
8.00
17,58
2.15
3.30
5.48
1.27
3.20
4.46
1.34
1.14
2.42
Standard
Deviation
.79
3.57
4.69
4.28
10.60
14.23
2.54
2.52
3.75
5.82
4.68
9.50
3.26
2.39
4.68
.90
3.47
3.89
2.09
1.95
3.94
Minimum
.08
.40
.47
0
.33
.83
2.00
2.31
6.49
1.45
2.31
6.07
.43
.95
1.74
.01
.40
.84
.02
.02
.07
Maximum
3.41
15.21
18.62
14.58
32.55
47.12
8.44
9.48
17.93
19.76
17.75
31.61
15.29
10.97
19.03
3.89
16.04
19.92
7.54
7.82
14.97
200
-------
The costs presented in Table V are
averages. The landfill costs include
normal landfills, balefills, and shredded
waste landfills, as well as sites with
direct haul or transfer. Analysis of costs
in a complex situation is aided by the
use of statistical methods to develop a
series of cost curves. Those analyses have
been completed and the results are available
in the main report, and are discussed in
Section 4.0. Here, we will simply summar-
ize a few of our main findings:
o not surprisingly, the main
variable affecting cost of haul
in collection vehicles is the
amount of time spent in haul
on an average day.
o for both haul in the collection
vehicles and haul in the larger
•transfer vehicles, increasing
vehicle capacity reduced costs.
A 10 percent increase in
vehicle size led to a 6-7
percent decrease in costs,
across all sites. '
o baling and shredding had little
effect on transportation costs
per ton. The major factors
affecting transportation
costs—distance, travel time,
and total system level of
operations—were much more
important.
o Shredders and balers were, as
expected, more expensive than
transfer stations, both in
capital and operating costs,
correcting for other factors
such as size. However, the
estimate for shredders is
probably low: in general,
the shredders were less clean,
and apparently less well-
maintained, than the other
facilities. In addition, one
shredder was not included in
the models because it had
been down 325 days in the
preceding year as a result of
an explosion. To get a good
estimate of shredder costs, it
is necessary to look over a
very long period. At the mean
capacity (400tpd) and 1 shift
of actual operation, the model
estimated costs of:
—baler S 13.03
—shredder 9.34 .
—transfer station 3.39
Figure 3 shows cost curves.
o large processing facilities did
not appear to be much, if any,
cheaper per ton than small ones.
However, maintenance and hours
of equipment operation did have
an effect on costs. Increasing
the hours of equipment opera-
tion by 10 percent reduced
costs by 6 percent. Increasing
the hours by going from one
to 1.5 shifts of equipment
operation per day would reduce
operating costs per ton by 30
percent. In essence, as long
as there is waste to process,
it makes the most sense to keep
the equipment operating.
o with respect to landfilling,
there were several key variables
(see Figure 4). Increases in
capacity reduced costs con-
siderably, with a 200 ton per
day facility 45 percent less
expensive per ton than a 50 ton
per day facility. Baling and
shredding also helped; a bale-
fill was 43 percent less
expensive per ton than a con-
ventional landfill, while an
uncovered shredded waste fill
was 70 percent less expensive to
run. The extra processing costs
do lead to savings in landfilling.
Note, though, that the shredded
waste fills also seemed to
operate under significantly
lower standards, and the costs
may underestimate new system
costs'. Two other factors were
important: phasing of construc-
tion and the price paid for
cover. All else being equal,
having to buy off-site cover
more than doubled the cost of
landfilling. Phasing construc-
tion—prearing the next area
while filling the current one
reduced landfilling costs,
probably because it allowed
more efficient use of equipment.
However, this was not a particu-
larly large or important variable.
201
-------
HAUL IN
COLLECTION
VEHICLES
HAUL IN
TRANSFER VEHICLES
LANDFILLING
LANDFILL
CONSTRUCTION
FIGURE 2: Disposal Process
202
-------
.18.19
18-
16-
14-
12-
10.
l$/to !
2-
TRANSFER STATION
I
4
i
8
1
12
Equipment Operating Hours Per Day
FIGURE 3: Processing Cost
16
203
-------
5.50-
5-
I$/ton It
4-
3-
2-
.523
\M
1.28
I
SO
I
100
200
Conventional
Landfill
2.18
Baled Waste
Shredded Wast*
400
TONS PER DAY
*Does not include construction; ^imed construction and on-site cover assumad.
FIGURE 4s Landfilling Cost
304
-------
o we could not model landfill
construction. The coats varied
too widely and were influenced
by a large number of site-
specific characteristics. An
engineering cost estimate
for the site preparation
is recommended as the best
approach.
Several variables showed no signifi-
cant effect including!
o ownership (public/private)
o weather (amounts of rain and snow)
o shredder particle size
o bale size
As known, these factors can be impor-
tant in specific situations. However, as
measured at these sites, they seen to be
generally less important than the variables
discussed above.
Two of the sites deserve special
mention, because of their costs. Both were
small landfills in an eastern state with
stringent groundwater protection regulations.
Both required extensive liners, monitoring
systems, clay cover, and other special treat-
ment. They were the best examples in our
sample of landfills which would meet strin-
gent environmental requirements under RCRA.
Our average cost for landfill construction
at other sites in the 50-150 ton per day
category was $1.34. These two sites had
costs of about §14 per ton, though the
landfilling costs once construction was
completed were about the same as for other
sites in that size range.
4.0 Modeling Results
Table VI shows the models. . All are
in log-log form, so {the coefficients can be
viewed as elasticities of cost with respect
to the other variables. The list of vari-
ables is provided at the bottom of the
table, and regression statistics are in-
cluded. The fits are not especially good,
but the models do explain a significant
portion of the variance. The standard
errors for each coefficient are also not
small, partly because of the small sample
size and measurement problems, but all the
major coefficients are significantly
different from zero, based on t-test. As
planning models, these results should be
sufficient and useful.
5.0 Osing_ the Results
Many system operators face a common
problem. Their current landfill is being
filled, and there is no obvious next site.
The options available may includei
o stretching the life of the
current site through processing,
better compaction, or resource
recovery
o making a future site more
acceptable through processing
o going further away to find an
acceptable site
o entering into regional agreements
involving large landfills or
resource recovery facilities
Costs are not the only concern in
evaluating these options, but they are one
of the most important. The toold developed
as a result of this study and presented
here may be helpful in determining whether
a particular option bears further con-
sideration. Although detailed evaluation
will require good, site/specific engineer-
ing cost estimation, rough comparisons are
useful when the number of choices is large
and the data are limited.
Consider a hypothetical systems —200
tons per day, direct haul to landfill, SO
km round trip. Now, assume that the system
operator has five options:
o make no changes
o increase collection truck
size 10 percent
o build a transfer station
o build a baler
o build a shredder
We're interested in the cost/ton for
each of the options under reasonable
assumptions; these have been computed and
are shown in Table VII. Note that both
baling and shredding increase processing
costs and reduce landfilling costs. We
have assumed that a balefill or shredded
205
-------
TABLE Vis MODELING RESULTS
Type
Model
Statistics
Haul in collection
capital;
operating:
totals
vehicles
In(HCOSTC)'
Processing
capital
operating:
total:
Landfilling
capital:
opera-tings
total:
'3.46
(.90)
1.431ji(TIMFR) -,7081n
(.33) (.31)
(TONLD)
ln(HCOSTO)-S.49 •«• 1.761n(TIMFK) -.6861n
(.80) (.29) (.28)
(TONLD)
In(HCOST) -5.56 + 1.661n(TIMFR) -.6881n
(.80) (.29) (.28)
(TOHLO)
R = .54
F{17,3) - 9.95
R - .67
F(17,3) -17.48
R = .65
F(17,3) =16.11
Transfer in larger vehicles
capital:
operating:
totals
ln(TCOSTS)=7.00+.1381n(TIME)-.8351n(TONS)
(1.80) (.21) (.21)
ln(TCQSro)-4.88+.3731n(TIME)-.6861n(TONS)
(1.71) (.20) (.20)
ln{TCOST> -6.86+.2871n(TIKE)-.7921n
-------
HCOST*
TIMFR
TONLD
TOOST
TIME
TONS
LOADS
PCOST
BALER
SHREDDER
CAP
MRS
LCOST
TOKDAY
COVER
PHASED
haul cost per ton; suffix:(c) = capital (o) = operating
fraction of vehicle and crew time spent hauling waste
haul vehicle capacity, tons per load
transfer in large vehicles, cost per ton
round trip travel time, minutes
total tons transferred per year
transfer vehicle loads per year
processing cost, $/ton
is the waste processed by a baler (0 = no, 1 » yes)
is the waste processed by a shredder (0 = no, 1 = yes)
design capacity of processing facility, tons per hour
actual processing equipment operating hours, last fiscal year
landfilling cost, S/ton
average landfill operation, tons per day
is cover purchased off site? (o • no, 1 = yes)
is landfill construction phased with landfilling? (0 = no,
1 - yes)
•NOTE: all models estimated in English units. Standard errors are in ( ) under
coefficients
.TABLE VII. COST PER TON UNDER VARIOUS OPTIONS FOR A 200 TPD HYPOTHETICAL.OPERATION
DIRECT TO LANDFILLING WITH PROCESSING AK2- TRANSFER
Increased
Current Vehicle Size Transfer Baling Shredding
Haul in collection vehicles 16.80
13.74
7.35 7.35 7.35
Processing
3.09 11.85 8.49
Haul in transfer trucks
4.64 4.64 4.64
Landfilling
4.'21
4.21
4.21 2.40 1.19
Landfill construction
2.78
2.78
2.78 2.78 2.78
TOTAL:
23.79
22.73
22.07 29.02 24.45
207
-------
waste landfill have the same environmental
requirements as a conventional landfill, so
that landfill construction costs are tin-
changed. We have also assumed, for the
moment, that balefills and shredded waste
landfills have the same density as conven-
tional landfills.
Increasing collection vehicle size,
within reason, can reduce costs by reducing
tripsi transfer stations nay, in some cases,
save inoney. But, in this simple analysis,
installation of a baler or shredder would
increase costs a great deal.
The question is then raised, would the
increased cost of operating a shredder or
baler be offset by more efficient use of
the existing landfill. We have referred to
landfill construction as "site-specific"
and suggested getting estimates for each
particular facility. Such estimates can
provide a cost/acre for site preparation.
To determine construction costs per ton of
waste, landfilled density must be estimated.
Fox example, assume a new site can
be prepared with construction costs of
SlO/ton of conventional waste. If using
bales increases in-place density by 25
percent, the construction cost would drop
to only $8/ton. In areas where baling or
shredding increases total waste per acre —
either from better packing or reduced
cover—these sophisticated technologies may
make sense. The data on in-place densities
in our sample was limited, at best, This
is the area where more field analysis would
be useful.
Now consider the next site to be
developed, some years hence. It will
undoubtedly be more expensive than the
current one. It is likely to be a long
distance away, subject to more stringent
regulations, and hard to find. If that site
is going to be much more expensive, it may
make sense to' try to stretch the useful
life of the current site, through baling or
shredding. Present value analysis using
the cost curves presented in pur report can
be used to estimte financial benefits of
delayed investment in the new site. However,
the political benefits of delay may be even
more important.
Based on the data from 45 sites, we
believe that transfer stations are often
justified, but that baling or shredding
before landilling will seldom save money
over conventional techniques. The advantages
of these more sophisticated waste processing
technologies are in extending the useful
life of a landfill site and reducing demand
for cover material.
Me also believe that the techniques
developed in the study, and briefly des-
cribed here, can be useful in investigating
options for the solid waste disposal system.
To aid that use, our final report has been
formated as a handbook with examples and
work sheets showing how to adapt an actual
situation for comparison with our results.
Acknowledgements
This work was performed under EPA
Contract #68-03-2868 for the Municipal
Environmental Research Laboratory {Cincinnati,
Ohio). Oscar Albrecht was the project
officer. The final report, entitled "Optional
Cost Models for land. Disposal of Municipal
Solid Waste" will be available through the
National Technical Information Service.
Obviously the results are the authors, and
do not necessarily reflect the vievs arid
policies of EPA.
208
-------
SUMMARY OF LANDFILL RESEARCH
BOONE COUNTY FIELD SITE
Richard J. Wigh
Regional Services Corporation
Columbus, Indiana 47201
Dirk R. Brunner
USEPA
Cincinnati, Ohio 45268
ABSTRACT
Five municipal waste test cells were constructed at the Boone County Field Site
during 1971 and 1972. Two were field-scale, and three were small-scale cells con-
structed for the purpose of performance comparison. All cells were monitored for
temperature, gas composition,settlement, and leachate quantity and characteristics
until closure in August, 1980, This paper presents preliminary results from the
analysis of over eight years of the leachate quantity and quality data. It must be
emphasized that the results are from relatively shallow (less than 2.5 meters) batch-
type cells.
INTRODUCTION
The Boonc County Field Site consisted
of a ten acre tract located 8 tan west of
the City of Walton, Kentucky. Available
facilities at the site included an office
trailer, a pole barn, a truck scale, and
a weather station.
Elevations at the research site range
between 213 and 244 m above sea level.
Surficial soils at the site are predomin-
antly a lean clay, classified by USDA as
Nicholson silt loan. Rubbly limestone
mixed with thin beds of soft calcareous
shales of the Falrview formation underlie
the shallow soil. The mean annual pre-
cipitation in the area is 927 mm. Monthly
normal mean temperatures range from 0 deg.
C in January to 24.4 deg. C. in July.
The objectives of test cells 1, 2A,
2B, 2C, and 2D included:
(1) Analyze the amount and character-
istics of leachate.
(2) Analyze the composition of gases
present In the cells.
(3) Evaluate a clayey silt soli as
an Impervious liner for leachate control.
(4) Evaluate the behavior of a field
scale test cell, 2D, as compared to simi-
larly constructed small-scale test cells,
2A, 2B, and 2C.
TEST CELL DESCRIPTION
Test Cell 1 consisted of a 45.4 m
long by 9.2 m wide trench type sanitary
landfill cell. The trench was excavated
•with vertical side walls and ramps on both
ends sloping approximately 1:7. The cent-
er 15.3 m of the trench was sloped approx.
7 percent to the transverse center line as
shown In Figure 1. A 30 mil (.76 ran) syn-
thetic liner (Hypalon) was centered in the
base of the cell to prevent any leachate
from migrating below the cell. Directly
above the liner a slotted collection pipe
209
-------
Observation Unit
11 t
u — - —
1r^
I!
50'
-Leachate Collection
Pipes
11491 J
30
PLAN VIEW
2' Soil Cover
Compacted Solid Waste
18" Clay
30 Mil. Synthetic Liner
Leachate Pipes (See Figure 3)
LONGITUDINAL SECTION
Observation Unit "~
Soil Cover
30'
I
Leachate Pipes
TRANSVERSE SECTION
Figure 1. Design of cell no. 1.
210
-------
was installed along the transverse center
line. Above this pipe and the synthetic
liner a 45.7 cm thick clayey silt soil
liner was placed.
A second slotted collection pipe was
installed in a trench in the soil liner.
Short-circuiting to the lower pipe was
prevented by lining the base and sides of
the trench with a polyethylene strip.
Both pipes drained into an observation
unit beside the cell. The area around
both pipes was backfilled with clean sili-
ca gravel. Details of the drainage scheme
are shown in Figure 2.
During June 1971, 395 metric tons or
residential refuse was placed in the test
cell at a dry density of 429 kg/cubic
meter. Moisture, temperature, gas and
settlement probes were placed within the
cell as it was constructed. Approximately
.6 m of the compacted clayey silt (CL)
soil was placed over the refuse as final
cover.
Four additional test cells containing
municipal solid waste were constructed
during August 1972. Three of the cells,
2A, 2B, and 2C were enclosed in identical
cylindrical steel pipes, 1.83 m in dia-
meter and 3.66 m long. The fourth test
cell, 2D, was an 8.53 m square field-scale
cell constructed to compare performance
with the small-scale cells.
The steel pipes were coated with cold
tar epoxy and placed vertically in an ex-
cavation on reinforced concrete pads.
Slotted pipe in a trough in the concrete
pad was used to collect leachate. Earth
backfill was placed around the pipes to
within 150 mm of the tops. Refuse was
placed in the pipes in 90-135 kg incre-
ments and compacted by dropping a 135 kg
weight from a height of approximately 1.2
m above the refuse. Temperature and gas
probes were placed at several levels with-
in the cell during refuse placement. 300
mm of compacted soil was placed above the
refuse. Then 300 mm of pea gravel was
placed on the soil to allow rapid perco-
lation of rainfall and to minimize evap-
oration.
Cell 2D was constructed in an excav-
tion 8.53 m square and 3.20 m deep. A 30
mil (.76 nm) CPE liner was placed above
a shaped sand bed in the base of the cell
and extending up the sidewalls. A slotted
PVC pipe was placed along the center line
of the base of the cell for leachate col-
lection and gravity drainage to the col-
lection well. The entire base of the cell
and liner was covered with 300 mm of sili-
ca sand. Plywood sheets were placed
against the synthetic liner on the side-
walls for protection from puncture and
tearing during cell filling.
A 7 tonne bulldozer was lowered into
the cell by a crane for compaction of
refuse. Temperature and gas probes were
placed during the filling at locations
shewn in Figure 3. A 300 mm layer of com-
pacted soil cover was placed over the 2.44
m of refuse. A berm system, consisting of
150 mm high triangular-shaped clay berms,
was hand constructed on top of the soil
cover to promote uniform percolation of
rainfall into the refuse cell.
Samples were obtained from the resi-
dential refuse being placed in all cells
for moisture and composition studies.
Chemical analysis of the refuse was done
for a number of parameters for Test Cell 1.
Statistical analysis indicated that any
performance variations within cells 2A,
2B and 2C could not be attributed to the
differences in the composition or the ori-
ginal moisture. A summary of the cell
data is shown in Table 1.
LEACHATE QUANTITY
Test Cell 1
Leachate was initially collected from
both the upper and lower pipes in Test
Cell 1 approximately two months after
construction. Cumulative leachate volume
is shown in Figure 3 together with the
volume of leachate predicted by using the
water balance method. Leachate collected
was 27% of the precipitation after 6.5
years.
In computing the water balance it was
assumed that the monthly mean temperatures
for the Boone County site were the same as
that reported for Covington, Kentucky.The
runoff coefficients for the wet and dry
seasons were chosen as .17 and .13. Assum-
ing a soil moisture storage capacity of
125 mm for the final cover, 90 mm of water
were required to bring the final soil cov-
er to field capacity. An additional
106,000 liters of water were required to
211
-------
6 mil
Plastic Liner
I Edge of
mi x" Bulkhead
Silica
Gravel
30 mil Synthetic Liner
LONGITUDIN&L VIEW OF CQUECTIGN SYSTEM
6 mil Plastic Liner
30 mil
Synthetic Liner
8,8% Slope 8.b£ Slope
SECTION A-A
Figure 2. Leachate collection system.
212
-------
T
o
* 1.400
14
Jj 1,200
«-l
B
| 1>(m
3 800
g
i 600
p
400
200
O- Leachate volume T.C, #1
A- Leachate volume estimated by water balance
t
A
O A
O A
O A
O /^ jjj\
0 A A -
O
o A
0 A
g A
W3 1974 1975 1976 1977 1978
Figure 3. Cumulative leachate volume
1979
-------
TABLE 1. SUMMARY OF CELL DATA
Cover soil classification
Depth of soil cover, m
ry
Surface area of refuse, m
Maximum depth of refuse, m
Mass of refuse, kg (dry)
Dry density of refuse, kg/m
Moisture content of refuse, 7. wet wt.
1
CL
.60
432.3
2.56
286,000
429
27.6
2A
CL
.30
2.627
2.56
2,046
304.3
22.5
Test
2B
CL
.30
2.627
2.56
2,113
3U.1
27.1
Cell
2C
CL
.30
2.627
2.56
2,135
317.6
24.1
2D
CL
.30
72.83
2.44
72,450
407.7
31.8
bring the entire cell and the soil liner
to field capacity based on an estimated
refuse field capacity of 330 mm/m of
refuse.
The water balance calculations were
reasonably accurate in predicting the
quantity of leachate, with only 167. dif-
ference after 6.5 years. If average eva-
potranspiration values had been used,
rather than ones computed from the actual
climatic conditions experienced, the dif-
ference would have been 43% leachate pre-
dicted than actually collected. This
large difference indicates a need for
leachate volume calculations in designs to
be based on extreme weather conditions as
well as average values.
Soil Liner
The quantity of leachate collected
from both the upper and lower pipes is
listed in Table 2. The quantity of
leachate from the lower pipe was equal to
or greater than that volume from the upp-
er pipe until January, 1972. This was
caused by leaving the valve closed on the
upper pipe except for weekly sampling,
thereby inducing sufficient head to cause
leakage into the lower pipe. After Dec-
ember 20, 1971, the upper pipe was allowed
to drain freely through an S-trap into a
sump in the observation unit, thereby
greatly reducing the hydraulic head and
consequently reducing the flow from the
lower pipe.
The percentage for the leachate col-
lected from the lower pipe decreased
over the years due to the change in col-
lection method, and/or a decrease in per-
meability of the soil liner, to a point
where more than 997. of the leachate was
collected from the upper pipe.
From Table 2, a tendency can be not-
ed that a greater percentage of the total
flow comes from the lower pipe during the
months of low total flow. For example,
during the period of December 1973-Novem-
ber 1974, 467. of the flow in the lower
pipe occurred during the six months when
only 317. of the total flow occurred. Ap-
parently during periods of high total
flow, the soil liner is saturated, and
with greater head existing at the refuse-
soil interface the leachate flows princi-
pally to the upper pipe.
Soil liner-drain efficiency calcu-
lations (5) predict greater than 907. of
the leachate should pass through the soil
liner. This indicates that the synthetic
liner and the soil liner are functioning
together and that the larger percentage of
the leachate collected in the upper pipe
is not due to the soil liner alone. The
synthetic liner blocks any deep percola-
tion forcing flow along the refuse-soil
interface and resulting in the high per-
centage of leachate collected in the
upper pipe.
214
-------
TABLE 2. SEASONAL DISTRIBUTION OF COLLECTED LEACHATE
1
Months
June -Aug.
Sept. -Nov.
Dec. -Feb.
Mar. -May
June -Aug.
Sept. -Nov.
Dec. -Feb.
Mar. -May
June-Aug.
Sept. -Nov.
•: Dec. -Feb.
Mar. -May
' June-Aug.
Sept. -Nov.
Dec. -Feb.
Mar. -May
June-Aug.
Sept. -Nov.
Totals
Year
1971
1971
1971-1972*
1972
1972
1972
1972-1973
1973
1973
1973
1973-1974
1974
1974
1974
1974-1975
1975
1975
1975
Upper & Lower Pipe
Volume - Liters
223
376
2789
7944
1552
19,608
54,964
95,157
53,802
24,835
81,321
61,336
22,565
41,035
61,869
96,325
16,378
31.063
673,152
Lower Pipe
Volume-Liters
107
243
539
661
66
880
395
121
178
97
50
131
72
80
51
67
106
81
3925
7. Through
Lower Pice
48
65
19
8
4
4
0.7
0.1
0.3
0.4
0.1
0.2
0.3
0.2
0.1
0.1
0.6
0.2
.58%
* Beginning of freely drained condition on January 1, 1972
215
-------
Test Cells 2A. 2B. 2C, 2D
The experimental design called for
the input of approximately 500 mm of pre-
cipitation each year into all of the cells.
Average annual rainfall at the site is in
excess of 900 mm so the cells were period-
ically covered, the cylinders with caps
and 2D with nylon reinforced Hypalon.
After June, 1979 the test cells were left
uncovered all the time.
Leachate was initially collected from
each test cell on the date and at the cum-
ulative rainfall quantities in Table 3.
The early appearance of leachate from 2D,
indicative of some short-circuiting, also
occurred in Cell 1.
TABLE 3. INITIAL COLLECTION OF LEACHATE
Test Cell
2A
2B
2C
2D
Date Cumulative
Leachate Precipitation(mm)
Collected
6-15-73
2-13-73
6-19-73
9-25-72
724
588
765
51
Figure 4 shows the quantity of leach-
ate collected from each test cell with pre-
cipitation. Test Cell 2C produced very
little leachate during the reporting peri-
od in comparison to 2A and 2B. A test
boring in the cell did not show any free
water stored in the cylinder. It is pos-
sible that a leak developed at a welded
joint near the surface of the soil cover
and very little of the precipitation actu-
ally entered the refuse mass.
The quantities of leachate collected
from 2A and 2B varied only slightly. By
the end of 1979, leachate collected from
2D was twice that per unit of surface area
from 2A or 2B.
LEACHATE QUALITY
Leachate samples were initially ana-
lyzed on a bi-weekly schedule. Monthly
samples were obtained in the later phases.
Concentration data was reduced to weighted
mean concentrations at the approximate time
at which equal intervals of leachate flow
were recorded. This normalized the data
for each cell so that concentration his-
tories were based on a parameter accounting
for the varying leachate quantities collec-
ted from each cell. Concentration and
mass removal histories normalized by dry
refuse weight are presented in Figures
5-16.
A summary of peak concentrations and
the leachate volume on the date of the
sample is presented in Table 4 for selec-
ted parameters. It is notable that all
but a few of the peak concentrations for
Test Cell 2A occurred within a short time
span starting with the onset of leachate
production. This was the time period dur-
ing which or shortly after field capacity
was achieved. Apparently this peak con-
centration period results from the initial
water contact with the refuse when the sup-
ply of the leachable substances and the
contact time are high.
For Test Cell 2B the time span for
peaks was somewhat longer than 2A, rang-
ing over a longer time period and greater
volume of leachate collected. Test Cell
2B did begin leachate production 4 months
earlier than 2A and before field capacity
was reached so this range might actually
have coincided closely with 2A.
For the field-scale cells, 1 and 2D,
the same general trend of most peak con-
centrations occuring during a certain time
span is noted. It does not appear though
that the peak concentrations for 2D occur-
red as early as for 2A, 2B and Test Cell 1.
If peaks did occur earlier, then the high
concentrations in the leachate must have
been either reduced by significant dilut-
ing leakage from the sides of the cell or
by channeling through the cell. The latter
situation would result in field capacity
not actually having been achieved until
somewhat after estimated water requiremerts
had been met, possibly during the time and
leachate volume range when peak concentra-
tions were recorded. Dilution was indica-
ted by the lower magnitude of almost all
of the peak concentrations of 2D as com-
pared to 2A and 2B.
Leachate Composition Comparability
One of the primary objectives of the
later test cells was to evaluate the be-
havior of a field-scale test cell, 2D,
compared to similarly constructed small-
scale test cells 2A, 2B and 2C. The con-
cept was to determine whether similarity
existed between individual small-scale
cells as well as similarity between the
216
-------
t*
6,000
5,000
4,000
3,000
2,000
1,000
A-—
2A
2B
2C
2D
1,000 2.000 3,000
CUMULATIVE PRECIPITATION, MM
Figure 4. Laachate volume, cells 2A, 2B, 2C, 2D
47000
5,000
-------
to
00
TABLE 4. PEAK CONCENTRATIONS
Parameter
PH
COD
BOD
Kjeldahl-N
Ammonia-N
Orthophosphate
Sulfate
Alkalinity
Acidity
Conductivity
Total Solids
Sodium
Potassium
Chloride
Iron
Magnesium
Manganese
Calcium
Zinc
Hardness
Test
Cone.*
7.07
37,500
30,000
700
552
61
1,160
8,870
3,620
12,200
23,600
1,049
1,950
1,749
614
374
184
2,320
104
7,500
Cell 1
Lea. Vol. #
4.1
0.3
2.1
0.4
0.3
0.04
0.3
0.3
0.3
0.2
0.3
0.4
0.7
0.6
1.5
0.2
0.04
0.2
0.2
0.3
Test
Cone.*
7.0
41,869
79,120
1,242
947
82
1,280
8,963
5,057
16,000
36,252
1,375
1,893
2,260
1,492
411
58
2,300
67
6,713
Cell 2D
I*a. Vol.tf
0.5
0.9
0.5
0.5
0.4
0.5
0.7
0.7
1.0
0.4
0.4
0.5
0.5
2.4
0.5
0.4
0.6
0.3
1.9
Test
Cone.*
6.2
57,330
62,560
1,560
1,035
390
2,215
11,535
6,720
17,000
46,484
1,900
2,225
2,335
1,547
486
109
2,280
150
7,067
Cell 2A
Lea. Vol.#
0.2
0.2
1.0
0.2
0.5
0.02
2.2
0.2
0.06
0.2
0.2
0.2
0.1
0.2
0.2
0.2
0.02
0.1
0.1
0.06
Test
Cone.*
6.0
61,600
72,220
1,897
1,185
185
2,775
13,880
6,843
18,000
45,628
1,700
2,939
2,343
2,902
617
115
4,000
360
10,575
Cell 2B
Lea. Vol.#
0.3
0.04
0.7
0.3
0.3
0.1
1.9
0.01
0.2
0.2
0.2
0.2
0.3
0.2
0.04
0.3
0.04
0.04
0.1
0.04
*. Concentration in mg/1 except for pH and conductivity (micromhos/cm)
it. Lcachate volume - liters per dry kilogram of refuse
-------
50
SB
o
1
40
30
20
10
o--
2A
2B
2D
234
CUMULATIVE LEACUATE VOLUME - I/kg of dry refuse
Figure 5. COD concentration history
-------
100
111
to
3
>u
SI
M
o
60
W
I
o
u
80
60
40
20
2A
2B
O 2D
A 1
O-
. O
1 23 4
CUMULATIVE LEACHATE VOLUME - I/kg of dry refuse
Figure 6. COD maaa removal
-------
10
ao
i
M
en 6
Bl
O--
_ 2A
_ 2B
- 2D
- -1
-O
234
CUMULATIVE LEACHATE VOLUME - I/kg of dry refuse
Figure 7. Hardness concentration history
-------
20
18
„ 1*
I
* 1*
•y 12
o
oo
,*
te
i
S
10
f Li
—^trir.--o-- -
23 4
CUMULATIVE LEACHATE VOLUME - I/kg of dry refuse
Figure 8. Hardness mass removal
-------
1.6
1.2.
o.a
0.4 .
O
234
CUMULATIVE LEACUATE VOLUME - I/kg of dry refuse
Figure 9. Iron concentration history
-------
5,0
o
I
55
i
4.0
3.0
2.0
1.0
o—
_ 2A
_ 2B
_ 2D
_ 1
234
CUMULATIVE LEACHATE VOLUME T I/kg of dry refuse
Figure 10. Iroji mass removal
-------
500^
§
§
u
400
300
200
100
O--
_ 2A
_ 2B
_ 2D
_ 1
O
234
CUMULATIVE LEACHATE VOLUME - I/kg of dry refuse
Figure 11. Magnesium concentration history
-------
-------
2.5
s
1
M
a
2.0
1.5
A.O
.5
=3 ^
1234
CUMULATIVE LEACHATE VOLUME - I/kg of dry refuse
Figure 13. Chloride concentration history
-------
4.or
3.0
to
.44
60
I
U
Q
2.0
1.0
A A A
rrrrQ- O-
O
234
CUMULATIVE LEACHATE VOLUME - I/kg of dry refuse
Figure 14. Chloride mass removal
-------
1.2
1.0
.8
.6
.4
.2
O--
_2A
— 2B
_ 2D
- 1
O
1 2 3 4
CUMULATIVE JLEACHATE VOLUME - I/kg of dry refuse
Figure IS. Sulfate concentration history
-------
K
o
0)
a
I?
•o
<*4
o
00
tie
l
2.0
1.5
1.0
Q
123 4
CUMULATIVE LEACHATE VOLUME - I/kg of dry refuse
Figure 16. Sulfate mass removal
-------
small-scale cells and the field-scale
ceil, 2D. It was hoped that the small-
scale cells would be adequate models of
Che large scale cell so that future re-
search efforts might utilize the smaller
cells for prediction of field behavior.
It was determined that on the basis
of a 10 component analysis of composition
that variations in performance between
cells 2A, 2B, and 2C could not readily
be attributed to variation in refuse com-
position. Unit densities of 2A, 21, and
2C were similar, ranging from 392-431
kg/m3. The refuse depth was the same for
all small cells. Water input to each
cell was assumed to be the same.
The initial refuse composition for
cell 2D was assumed similar to that of
the small cells. The density was some-
what greater at 598 kg/m^ at a slightly
higher moisture content. The possible
effect on leachate concentration of this
greater density is not known. The input
of water was controlled the save as for
the small - scale cells. Cumulative leach*
ate production was actually greater than
the precipitation. Refuse depth was
slightly less in 2D at 2.44 m whereas
in the small cells the depth was 2.56 m.
For the six parameters presented
here, there is a strong similarity in
concentration histories for 2A and 2B
over the leachate volume range examined*
For COD, Cl, sulfate, Mg and hardness,
the concentration has dropped in an ex-
ponential manner to less than 10% of the
peak values recorded. Iron has been re-
duced to about 50% of peak, indicating a
slow release of the mass available for
leaching.
Statistical comparison of 2A and 2B
concentration histories with an analysis
of differences technique (1) was not mean-
ingful. Future efforts for comparison
will be attempted with a goodness of fit
test, such as the Kolmogorov (3).
Total mass removed plots indicate
similar trends for all parameters when
comparing 2A and 2B. This would be ex-
pected since the leachate volumes and
concentration histories were similar.
For COD, the field-scale cells cumulative
mass removals lag those of 2A and 2B and
show very little mass removal after 3
liters/kg of cumulative leachate. Con-
centrations for these cells have dropped
to 1-27. of the highest values. Dilating
side wall leakage has probably also con-
tributed to the low mass removals. Other
than for Test Cell 1 iron, the cumulative
mass removals for the field-scale and the
small-scale cells are quite similar, even
though the concentrations recorded from
the field-scale cells were generally low-
er in the initial phase of the experiment.
The mass removals will also be compared
with a goodness of fit test in the future
to determine whether the small-scale cells
were statistically comparable to the fieU-
scale cells.
MATHEMATICAL DESCRIPTION
OF
LEACHATE CONCENTRATION HISTORY
The repetitive shape of the leachate
concentration curves and similar volumes
at peak concentrations for many of the
parameters Indicated that the weighted
mean concentration history curves night
be mathematically described. The need
for such a descriptive model has been
documented by Brunner (2).-—A forecasting
technique would provide a significant im-
provement in the design and regulatory
review processes and could be used to de-
termine optimal operational practices.
Forecasting would be extremely useful in
evaluation and design of leachate treat-
ment needs and the suitability of the hy-
drogeologic setting.
The concentrations in the leachate
are dependent on many factors, not all of
which are understood or known, especially
the magnitude of influence the factors
exert on resultant concentrations. Im-
portant variables might be the initial
mass of a substance readily available for
leaching, decomposition of refuse within
the fill and subsequent additional mass
availability, the pH, solubility limits,
the rate of water throughput, decompo-
sition
-------
centration is to consider the refuse mass
as a single well-mixed reactor (7). Such
a system is depicted in Figure 17.
V,C,G
q, c
Figure 17. Well-mixed Reactor
C± and C are the water input and
leachate concentrations. q< and q are
water input and leachate flow rates re-
spectively. V is the total water con-
tained in the refuse system and G is the
internal mass generation rate.
Writing a mass balance on the
reactor yields:
- ,C
or the time rate of change of mass within
the system equals mass inflow less outflow
plus internal generation. This can be
further simplified by assuming that the
incoming concentrations are negligible,
and that the volume of water within the
system is constant or q^ - q and V is a
constant. This yields:
VdC
dt
- qC + G
(2)
To solve the differential equation, an
expression for G is necessary. One op-
tion is to consider the mass generation
as a first order reaction as is commonly
done for biochemical decomposition (8).
This expression is simply G = kVC, or the
rate of mass generation is proportional
to the amount of mass present, with k
being the rate coefficient. The differ-
ential equation becomes:
VdC
dt
qC + kVC
(3)
which when solved for C=CO at t-o yields:
C - C0e
Thus, a general expression for the
concentration change with time with t=o
at field capacity is a simple exponential
dependent on the initial concentration
(C0), the leachate flow rate q, total
volume of water in the refuse at field
capacity V, and an empirical rate constant
k.
For most of the parameters studied,
peak concentrations were recorded when
field capacity was being achieved. Co
was therefore selected as that peak con-
centration for each parameter. V can be
calculated for each test cell from initial
moisture contents and field capacity de-
terminations. The leachate flow rate (q)
must be constant with time to be valid in
the equation. This was determined by plot-
ting leachate collected with time as shown
in Figure 18. The resultant constants for
each test cell are listed in Table 5.
TABLE 5. TEST CELL CONSTANTS
Test Cell
q*
1
2A
2B
2D
.58
.58
.51
.67
.75
1.15
1.04
.92
* Units are liters/kg dry refuse/year
# Units are liters/kg dry refuse
The concentration history curves for
2A and 2B as compared to the general ex-
pression with k selected for best visual
fit are shown in Figures 19-24. Reasona-
bly good fit was obtained for all the par-
ameters except iron. It is interesting to
note that the mass transfer rate coeffi-
cient (k) for chloride was zero, indicat-
ing that all mass present in the refuse
was available for leaching initially or
so readily removeable that it was inde-
pendent of biochemical reactions within
the test cells. Magnesium and sulfate
had low constants (.15) and iron the high-
est (.4). Hardness had a rate coefficient
of .25, perhaps influenced by the slow
iron removal rate.
232
-------
41
IB
-------
50
o
M
H
§
8
H
30
20
10
1 234
CUMULATIVE LEACHATE VOLUME - I/kg of dry refuse
Figure 19. COD Concentration-comparison.
-------
00
I
1
o
g
234
CUMULATIVE LIAGHAfE VOLUME - I/kg of dry refuse
Figure 20. Hardness concentration comparison.
-------
55
O
M
i
W
U
Z
O
CJ
S5
O
Q
H
1.6
1.2
.8
1234
CUMULATIVE LEACHATE VOLUME - I/kg of dry refuse
Figure 21. Iron concentration comparison.
-------
500
400
§
o
300
200
100
2 3 4
CUMILATIV1 LEACHATE VOLUME - g/kg of dry refuse
Figure 22. Magnesium concentration comparison.
-------
2.5
00
I
•z
o
w
u
§
u
u
Q
O
M
fi?
2.0
1.5
1.0
.5
1234
CUMULATIVE LEACHATE VOLUME - I/kg of dry refuse
Figure 23. Chloride concentration comparison.
-------
1.2
00
I
§
§
o
1.0
.9
.8
.6
.4
.2
2A
2B
— — Eon. 4
2 34
CUMULATIVE LEACHATE VOLUME - I/kg of dry refuse
Figure 24. Sulfate concentration comparison.
-------
Mass Removal
Integrating the concentration times
the flow rate over infinite time (Cqdt)
yields a general expression for total
leachable mass.
(5)
The total mass available estimated from
this expression for 2A and 2B and the
amount removed at 4.5 I/kg of dry refuse
are shown in Table 6. There is reasona-
bly good agreement between these values
and the shape of the mass removal curves.
TABLE 6. 2A-2B MASS REMOVALS
AT 4.5 lAg OF LEACHATE
TABLE 7. 2D CONCENTRATIONS (mg/1)
AT FIELD CAPACITY
Parameter
COD
Hardness
Iron
Magnesium
Chloride
Sulfate
k
.25
.25
.40
.15
0
.15
Total * Mass
Est. Mass Removed
2A
104 96.6
12.8 11.2
6.6 3.3
.7 .52
2.1 2.5
1.6 1.4
2B
97.8
13.4
3.0
.'62
2.2
1.5
* q/kg of dry refuse
As was noted earlier, the peak con-
centrations of 2D at field capacity were
lower than those of 2A-2B. Likely causes
were thought to be diluting sidewall leak-
age and short-circuiting within the cell.
Another explanation would be the higher
flow rate, less contact time, resulting
in reduced concentrations. If the total
available masses/unit weight for each cell
are considered equal and the mass trans-
fer constants do not change for the .high-
er flew rates then the ratio of the peak
concentrations for 2A-2B and 2D can be ob-
tained using Equation 5. The results are
shown in Table 7.
Parameter
Est. Cone. Actual Cone.
COD
Hardness
Iron
Magnesium
Chloride
Sulfate
33,000
4,030
490
340
1,760
810
34,500
5,000
880
330
1,730
870
Application to Field Situation
It must be emphasized that Equation
4 is only applicable to batch-type sy-
stems. If, after further comparison to
test cells of more variable conditions
there is still reasonable agreement, the
general expression could possibly be ap-
plicable to field conditions. Shown in
Figure 25. are two concentration curves,
one for the single cell or batch system,
and the other the estimated COD concen-
tration for a single lift landfill oper-
ated for 10 years. If proven valid by
field studies, an expression of this type
would provide a valuable tool for siting,
design and operation of landfills.
SUMMARY AND CONCLUSIONS
Five sanitary landfill test cells
were constructed at the Boone County
Field Site during 1971-72. Three of
these were small-scale cells containing
approximately 2100 kg of dry refuse. The
two field-scale cells, 1 and 2D, contained
266,000 and 42,450 kg of dry refuse. The
test cells were closed in August, 1980.
Preliminary findings from analysis
of part of the data include:
1. Leachate collected from Test Cell
1 was 27% of precipitation after
6.5 years. Water balance calcu-
lations were reasonably accurate
in predicting the quantity of
leachate, provided actual cli-
matic data rather than average
was used. This demonstrates the
need for analyzing a variety of
240
-------
50
\
1.0 YEAR LANDFILL OPERATION
q/v - 0.5k -.25C•52
M
1
§
o
1
30
20
10
\
One year
Operation
\
\
V
\
Operation discontinued
after 10 years.
4 8 12 16
TIME - years
Figure 25. Projected landfill leachate COD concentrations.
20
-------
weather conditions when using the water
balance method as a design tool.
2. The efficiency of the soil liner
in Test Cell #1 could not be e-
valuated because of interference
from the underlying synthetic
liner.
3. Leachate concentration histories
showed similar trends for all
cells, with most parameter peaks
occuring at field capacity.
After 4-5 I/kg of refuse of
leachate had been collected,
contaminant concentrations were
generally less than 107. of peak
values.
4. From 56-104 grams of COD, 11-13
grams of hardness, 1.2-1.5 grams
of sulfate, .52-.62 grams of
magnesium, 2.0-2.4 grams of
chloride and 1.9-3.6 grams of
iron per kg of dry refuse had
been leached from the cells at
the close of the study. Mass
removal rate trends for all but
iron were tending towards no
further removal.
5. A simple exponential equation
provided an adequate description
of the concentration histories
of 5 of the 6 parameters examin-
ed. Comparison of predicted to-
tal mass removals with actual
mass removals showed reasonable
agreement. The theory should be
applied to other test cells and
field conditions to verify ap-
plicability.
ACKNOWLEDGMENT
The test cell data analysis was per-
formed under contract with the Solid and
Hazardous Waste Research Division of the
Municipal Environmental Research Labora-
tory in Cincinnati, Ohio. Project Of-
ficers for the contracts were Richard A.
Carnes and Dirk R. Brunner.
REFERENCES
1. Boone County Field Site Interim
Report. 1979. EPA-600/2-79-058,
U.S. Environmental Protection
Agency, Cincinnati, Ohio 45268
2. Brunner, D. R. 1979. Forecast-
ing Production of Landfill
Leachate. In: Municipal Solid
Waste: Land Disposal. Proceed-
ings of the Fifth Annual Research
Symposium. EPA-600/9-79-023a,
U.S. Environmental Protection
Agency, Cincinnati, Ohio 45268
3. Conover, W. J. 1971. Practical
Nonparametric Statistics. John
Wiley & Sons. New York.
4. Interim Summary Report: Boone
County Field Site - Test Cell
#1. 1976. U.S. Environmental
Protection Agency, Cincinnati,
Ohio 45268 (unpublished)
5. Moore, C. A. 1980. Landfill and
Surface Impoundment Performance
Evaluation. SW-869, U.S. Envi-
ronmental Pro ection Agency,
Washington D.C. 20460
6. Phelps, D. H. Solid Waste Leach-
ing Model. University of British
Columbia, (unpublished)
7. Straub, W. A. 1980. Development
and Application of Models of
Sanitary Landfill Leaching and
Landfill Stabilization, RP #259.
Dartmouth College, Hanover, NH
03755
8. Weber, W. J. 1972. Physicochem-
ical Processes for Water Quality
Control. Wiley-Interscience,
New York.
242
------- |