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

-------