United States
                    Environmental Protection
                    Agency

                    Research and Development
Municipal Environmental Research
Laboratory
Cincinnati OH 45268

EPA-60o7S2-84-141 ~Sept7?984
                                                                                          i /
&EPA         Project Summary
                    Landfill Gas  Production  from
                    Large  Landfill  Simulators
                                  i
                    Larry W. Jones, Robert J. Larson, and Philip G. Malone
                      A study was conducted to investigate
                     gas production rates and composition
                     in municipal solid waste (MSW). Im-
                     proved monitoring methods were used
                     to corroborate and  add to  previous
                     studies. A completely automated gas-
                     monitoring system was used on four
                     sanitary landfill simulators  (lysimeters
                     or test cells) of two different sizes.
                      Gas was produced in four phases: an
                     aerobic phase,  a nonmethanogenic
                     anaerobic phase, an unstable methano-
                     genic phase, and a stable methanogenic
                     phase. The last stage was just being
                     reached as the experiment was termi-
                     nated.
                      The automated gas-measuring system
                     and  the gas-chromatograph-based,
                     gas-analysis system used in the study
                     both  functioned  satisfactorily. Gas
                     samples were collected in an  all-metal
                     collection system, as plastic and glass
                     vessels proved unsatisfactory.
                      The two sizes of test cells produced
                     very similar volumes and compositions
                     of total gas, but the small cells produced
                     more methane and less hydrogen than
                     the large cells. Relatively high, consis-
                     tent levels of nitrogen were found in the
                     gas from this study. This factor could
                     pose  serious problems regarding the
                     use of this gas for energy.
                      The study demonstrated  that the
                     conditions present in the average MSW
                     landfill  are not  ideal for maximum
                     production of methane. Further studies
                     are needed on the effects  of environ-
                     mental and nutritional factors in me-
                     thane production.

                      This Project Summary was developed
                     by EPA's Municipal Environmental Re-
                     search Laboratory, Cincinnati, OH, to
                     announce key findings of the research
project that is fully documented in a
separate report of the same title (see
Project Report ordering information at
back).

Introduction
  Many investigators have reported the
production of combustible gas from mu-
nicipal solid  wastes (MSW) placed in
anaerobic environments such as landfills.
But to exploit MSW as a possible fuel
source, gas produced during its decompo-
sition must be accurately analyzed and
measured. This  study uses improved
monitoring methods to corroborate and
add to previous investigations of MSW
gas production rates and  composition.
The study also examines problems related
to scaling of large laboratory simulation
tanks. Two sizes of test cells were  in-
cluded to allow broad comparisons of the
data obtained here with those of other
studies. The project uses a completely
automated gas monitoring system on four
sanitary landfill simulators (lysimeters or
test cells).


Methods and Materials

Test Cells
  All gases were collected from four (two
each of two sizes) carefully sealed MSW
test cells. The cells were cylindrical steel
(7.25-mm rolled plate) tanks. Each of the
two  smaller  test  cells  had  an  inside
diameter (ID) of 0.91 m and was 1.83 m
high; each of the two larger cells had an
ID of 1.83 m and was 3.66 m high (see
Figure 1). All interior surfaces were coated
with coal-tar epoxy to protect the walls
from corrosion and the contents from
contamination. Cells were loaded with a
30.5-cm layer of clayey-sandy soil, a 3-

-------
                                                  Deionized
                                                    Water
             Deionized
              Water
                I
    Washed^
    Pea
    Gravel
 Thermistor •—-
       61 cm
          1

Polypropylene^
   Beads  ~~
            91.4 cmOD
f|&feife?-
             So/l'Lihe'r-
               PVC
              Valve
                      122cm
                                     Washed
                                     Pea
                                     Gravel
               .22.9 cm       61 cm
                7.6 cm         *_».
                   Thermistors  T
                            61 cm
                183 cm


              ,30.5 cm         ~
                     Polypropylene
              61 cm  Beads
         10.8 cm
Figure 1.    Schematic diagram of the municipal solid waste test cells.
cm  layer of polypropylene beads (6.25-
mm diameter), a layer of MSW (0.79 m3
for the small  cells and 6.42  m3 for the
large cells) from noncommercial collec-
tion routes in Warren County, Mississippi,
and a 7.6-cm  layer of washed pea gravel
to help disperse the  leaching fluid'over
the surface. A perforated diffusion pan
was bolted to the inside of the test cell top
as an additional aid, and a headspace (23
cm for the small cells and 84 cm for the
large cells) was left below the test cell top
plate. After the test cells were loaded and
all monitoring devices were  tested, the
cells were sealed by welding the steel lids
to the tops and pumping metal  sealant
into a machined groove on the undersides
of the lids.
  The test cell profiles were designed and
constructed to simulate 1 - and2-m-thick
cores of MSW taken from a municipal
landfill. Thetest cells resembled medium-
density cells in a sanitary landfill contain-
ing unprocessed  wastes in  the  humid
eastern United States.


Gas-Flow Measuring and
Sampling Systems
  The gas-flow measuring and sampling
systems consisted of gas probes in each
tank that collected the gas produced by
the decaying waste, a gas-flow measuring
and logging device,  and a system that
allowed the extraction of an uncontami-
nated gas sample  for analysis.
                             Three gas probes were installed in each
                            test cell. Each consisted of a perforated
                            0.6-cm copper tube coated inside and out
                            with coal-tar epoxy to prevent corrosion
                            by leachate. The copper tubing was fitted
                            into a perforated polyvinyl chloride pipe to
                            protect the tubing during waste compac-
                            tion.
                             The gas-flow measuring system oper-
                            ates using a  pressure-controlled flow.
                            Components of the system include a
                            differential pressure switch/gauge, three
                            solenoid valves (normally closed), one
                            linear mass gas f lowmeter, one controller,
                            seven shut-off valves, and one pressure
                            gauge. A data logger was used to record
                            gas production data from all  four test
                            cells.
                             Two  sampling  methods were tested
                            during the first 18 days of  test cell
                            operation—one using flexible plastic bags
                            and the other using  stainless steel cyl-
                            inders. The gas-bag system had to  be
                            abandoned because of oxygen diffusion
                            into the bags. The system  with stainless
                            steel cylinders was used through the
                            remainder of the experiment; it used grab
                            samples collected from  the test cells in
                            glass gas bulbs through the gas-measur-
                            ing system with the same pumping
                            procedure used for the gas bags. Shortly
                            after the change to grab samples, the O-
                            rings on the gas  bulbs began to leak,
                            requiring the use of an all-metal system
                            using a compressor and  steel storage
                            tank. After this  all-metal system was
installed, the oxygen levels were at  o
below detection limits.
  Gas  samples were analyzed on a
Perkin-Elmer Sigma  3* gas chromato-
graph(GC) for oxygen, hydrogen, nitrogen,
carbon  dioxide, methane, and water
vapor.

Data Reduction and
Presentation
  The raw data tapes from the data logger
were read by a Martek Model 421 -DRS
magnetic tape reader and printed on a
paper tape printer. Another magnetic tape
reader transferred the data to a Hewlett-
Packard Model 9830 computer through a
Martex  Model  421-12-DRS  computer
interfacing  model.  The data were then
verified, reorganized, and stored in raw
form on magnetic tapes in an  array that
allowed easy access and manipulation by
the computer. The raw data were reduced
by engineering units and corrected for the
nonlinear  output from the thermistors
and barometric pressure sensors and for
variations in gas composition.  Programs
have been developed to present the final
data in any of several forms.

Results
  The relative amounts and kinds of gases
found in this study correspond with those
reported for the theoretical  stages  of
bacterial succession in common  anaer-
obic digesters: (1) an aerobic phase, (2) a
nonmethanogenic anaerobic phase, (3)
an unstable methanogenic phase, and(4)
a stable methanogenic phase. Gases pro-
duced in the last  phase  are  the most
desirable since they are usually 45%  to
60% methane, with the balance i"-vng
easily removed carbon dioxide.
  The aerobic or first phase consisted of
the rapid uptake of any residual oxygen in
the MSW and the release of nearly equal
amounts of carbon dioxide. This phase
apparently took place before the first gas"
sampling, 24 to 48 hours after the cells
were  sealed.  Only very  low  levels  of
oxygen were found in these first samples.
This result could be due to the high initial
temperature of the waste, which would
result in high  metabolic rates for the
microbiological flora.
  The anaerobic second phase was indi-
cated by the production of large volumes
of carbon dioxide.  This phase was also
well under way  in the first 2  or 3 days
after cell sealing. The total gas output of
the cells (mainly carbon dioxide and
'Mention  of trade names or commercial products
 does not constitute endorsement or recommend
 tion for use                           '

-------
nitrogen) decreased dramatically over the
next  100 to 150 days as  the  MSW
approached water saturation and leachate
production.  Hydrogen production was
noted in all of the test cells 60 to 90 days
before methane was  detected in any of
them.
  The  third  phase (unstable methano-
genic) began for three of the four test cells
at about the  time that leachate began to
be produced, between days 250 and 300.
Methane evolved slowly and somewhat
variably over the remainder of the experi-
ment, with the gas output attaining 10%
to 25% methane only after nearly 4 years
of fermentation.
  The  fourth  phase (stable  methane
production), with its constant 40% to 60%
methane  concentrations,  was  never
reached by any of the cells over the 1500
days of the study. But all of the cells were
increasing in methane production when
the experiment was terminated. In only
one of the small cells  did the percentage
of methane in the released gas increase
continuously, reaching 26% by the end of
the experiment and still increasing.
  The  absence  of a stable methane
production phase may have been caused
by less-than-optimum conditions  in the
leachate such as low  nitrogen-to-carbon
ratios  and low total  volatile acids as
substrates for the methanogenic bacteria.
Other factors may have been the lower-
than-optimum  temperatures in the un-
insulated  test cells, high levels of toxic
metals or  organics, or the acidic pH's of
the leachates. Stable methane production
is not uncommon after only a few weeks
under ideal circumstances; but develop-
ment times of 100 to  300 days are more
common,  and longer  times  are not un-
usual.
  The total volume of gas produced varied
only about 12% between the small and
large cells (17.75  and 20.1  ml/kg dry
MSW per day, respectively). Compositions
of the total gases evolved were also quite
consistent for the major gases from all
test  cells  (70.5% and 69.6% for carbon
dioxide and 19% and 21 % for nitrogen in
the small  and  large cells, respectively).
These values represent the first 630 days
of the experiment.  During  the  same
period, fermentative  activity  in the test
cells  was quite consistent.  The  same
consistency  is shown for the first  100
days when about half the total gas was
produced.
  The relatively high,  consistent levels of
nitrogen found in the  gas from this study
could pose serious problems regarding its
use for energy. These levels could have
resulted from denitrif ication, or they could
have been due to contaminating air drawn
into the test cell. Further study of these
causes and careful monitoring of nitrogen
levels  is needed in future  studies of
methane production.
  The amount of flammable  gases pro-
duced  varied considerably within both
sizes of test cells. Hydrogen was the only
gas that had consistent variations be-
tween the small and large test cells, and
this  result may be  accounted for  by
chance. The smaller cells consistently
produced much less hydrogen than the
larger cells over the first 630  days of the
experiment, possibly because of the larger
methane production in the smaller cells.

Conclusions
  The use of an automated, gas-measur-
ing system  effectively  monitored  the
volume of gases released from four landfill
simulation test cells. Results show that
conditions present in the average MSW
may not be ideal for maximum methane
production,  but  that  very appreciable
amounts of methane(and carbon dioxide)
can be expected from the average landfill
over extended periods. Further studies of
the effects of environmental  and nutri-
tional factors  on the  time required to
develop stable methanogenic conditions
and bacterial populations are needed so
that  conditions in the landfill can  be
modified to  maximize or minimize me-
thane production, depending  on the use
to be made of the site.
  The full report was submitted in fulfill-
ment of I nteragency Agreement No. EPA-
IAG-D4-0569 by the USAE Waterways
Experiment Station under the sponsorship
of the U.S. Environmental  Protection
Agency.
   Larry W. Jones. Robert J. Larson, and Philip G. Ma/one are with USAE Waterways
     Experiment Station. Vicksburg. MS 3918O.
   Robert E. Landreth is the EPA Project Officer (see below).
   The complete report, entitled "Landfill Gas Production from Large Landfill
     Simulators. "(Order No. PB 84-235 779; Cost: $ 13.00. subject to change) will be
     available only from:
          National Technical Information Service
          5285 Port Royal Road
          Springfield. VA 22161
          Telephone: 703-487-4650
   The EPA Project Officer can be contacted at:
          Municipal Environmental Research Laboratory
          U.S. Environmental Protection Agency
          Cincinnati, OH 45268
                                    •tl U S GOVERNMENT PRINTING OFFICE, 1984 - 759-015/7822

                                        3

-------
United States                       Center for Environmental Research                                           BULK RATE
Environmental Protection              Information                                                        POSTAGE & FEES P
Agency                            Cincinnati OH 45268                                                        ppA
                                                                                                      PERMIT No. G-35


Official Business
Penalty for Private Use $300
                                     LOU  W  TILLE.Y
                                            N  V  EPA
                                     230  S  OEAR90HM  ST
                                     CHICAGO  IL

-------