EPA 670/2-74-016
March 1974
Environmental Protection Technology Series
RECLAMATION OF
ENERGY FROM ORGANIC WASTE
National Environmental Research Center
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
-------
EPA-670/2-74-016
March 1974
RECLAMATION OF ENERGY FROM ORGANIC WASTE
By
John T. Pfeffer
Department of Civil Engineering
University of Illinois
Urbana, Illinois 61801
Grant No. R-800766
Program Element No. 1DB314
Project Officer
Clarence A. demons
Solid and Hazardous Waste Research Laboratory
National Environmental Research Center
Cincinnati, Ohio 45268
Prepared for
National Environmental Research Center
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
-------
REVIEW NOTICE
The Solid Waste Research Laboratory of the
National Environmental Research Center - Cincinnati,
U.S. Environmental Protection Agency, has reviewed
this report and approved its publication. Approval
does not signify that the contents necessarily re-
flect the views and policies of this laboratory or
of the U.S. Environmental Protection Agency, nor
does mention of trade names or commercial products
constitute endorsement or recommendation for use.
The text of this report is reproduced by the
National Environmental Research Center - Cincinnati
in the form received from the Grantee; new prelimi-
nary pages have been supplied.
ii
-------
FOREWORD
Man and his environment must be protected from the adverse
effects of pesticides, radiation, noise and other forms of
pollution, and the unwise management of solid waste. Efforts
to protect the environment require a focus that recognizes the
interplay between the components of our physical environment--
air, water, and land. The National Environmental Research
Centers provide this multidisciplinary focus through programs
engaged in
• studies on the effects of environmental
contaminants on man and the biosphere, and
• a search for ways to prevent contamination
and to recycle valuable resources.
In an attempt to solve the problems involved in solid
waste disposal, this study applied an anaerobic fermentation
process to the production of methane from an organic fraction
of urban refuse.
A. W. Breidenbach, Ph.D.
Di rector
National Environmental
Research Center, Cincinnati
iii
-------
ABSTRACT
The continued pressure for a clean burning fuel has created such
a demand for natural gas that the gas reserves in this country are
being severely depleted. The mounting quantities of urban solid
wastes presents a significant source of supplimental energy if
a practical means of extracting the energy can be developed. A
process employing microorganisms capable of fermenting organic
compounds to methane and carbon dioxide has been used for decades
to stabilize the organic sludges generated by water pollution
control processes. The biological mechanism has been extensively
studied, elucidating the nutritional, environmental and operational
requirements.
This study applied the anaerobic fermentation process to the
production of methane from the organic fraction of urban refuse.
Shredded domestic refuse from which the inorganic fraction was
separated was used as a substrate. Raw sewage sludge was added
to the substrate in proportion to the rate at which it is produced
by a population producing a given quantity of refuse. The
quantity and quality of gas produced, the rate of gas production,
the solids reduction, nutritional requirements and operating
problems were evaluated in a laboratory system operating at
temperatures ranging from 35°C to 60°C.
The results of the laboratory study together with published data
on both capital and operating costs of refuse shredding, refuse
separation, reactor volume, reactor mixing, reactor heating and
residue dewatering were used to analyze the economics of the
process. This analysis indicated that methane can be produced
by anaerobic fermentation of organic refuse at a cost that would
permit the sale of the gas at a price that is competitive with
the current energy costs.
iv
-------
TABLE OF CONTENTS
Page
Page
xi
CONCLUSIONS
RECOMMENDATIONS xiii
INTRODUCTION 1
Solid Waste Disposal 1
Energy Shortage 2
Refuse as a Source of Fuel Gas 2
Purpose of the Study 5
THEORETICAL CONSIDERATIONS 7
Basic Fundamentals of Anaerobic Digestion 7
Anaerobic Digestion of Domestic Refuse 12
Cellulose Reduction in Anaerobic Digestion 15
Bacterial Decomposition of Cellulose 17
Gas Composition 19
EXPERIMENTAL PROCEDURE 27
Experimental Reactors 27
Substrate 30
Analytical Procedures 31
Dewatering of Digested Solids 34
RESULTS 37
Characterization of the Raw Refuse 37
Phase I 39
Solids Destruction - Phase I 43
Phase 2 - Mesophilic Temperature Studies 46
Gas Production and Composition - Phase 2 53
Phase 3 - Thermophilic Temperature Studies 57
Gas Production and Composition - Phase 3 62
Cellulose Decomposition 67
Dewatering of Reactor Residue 69
Characteristics of the Separated Inorganic Component 71
Digestion of Hydropulped Refuse 72
PROCESS EVALUATION 73
Materials Balance for Process 73
Unit Processes - Shredding 78
Unit Processes - Separation 80
Unit Processes - Anaerobic Fermentation 83
Economic Evaluation of the Anaerobic Fermentation 89
Process
REFERENCES 101
APPENDICES 105
-------
LIST OF FIGURES
Number ^
1 Future Natural Gas Supplies and Requirements 3
2 Anaerobic Stabilization of Complex Organics 7
3 Variation of Volatile Solids Destruction with 13
Detention Time
4 Structure of Beta 1,4 Cellulose 16
5 The Relationship Between Gas Composition, Temperature, 25
pH and Alkalinity
6 Schematic of Experimental Units 28
7 Laboratory Reaction Units (Battery Containing Units ' 29
1 -4)
8 Effect of Raw Sewage Sludge Addition on Refuse 41
Digestion
9 Variation in Gas Production Under "Equilibrium" 49
Conditions - 35°C
10 Variation in Gas Production Under "Equilibrium" 52
Conditions - 40°C
11 Variation in Gas Production Under "Equilibrium" 54
Conditions - 45°C
12 Gas Production at Mesophilic Temperatures 56
13 Variation in Gas Production Under "Equilibrium" 58
Conditions - 50°C
14 Variation in Gas Production Under "Equilibrium" 61
Conditions - 55°C
15 Variation in Gas Production Under "Equilibrium" 63
Conditions - 60°C
16 Gas Production at Thermophilic Temperatures 65
17 Effect of Temperature on Gas Production 66
18 The Effect of Temperature and pH on the Maximum 68
Rate of Cellulose Utilization
VI
-------
FIGURES (Continued)
Number Page
19 Relationship of Cellulose Destruction with Retention 70
Time at 35°C
20 Mass-Volume Balance for Conversion of Refuse to 74
Methane
21 Power Requirements for Hammer-mills and Shredders 79
22 The Effect of Temperature and Retention Time on 91
Reactor Slurry Concentration
23 Heating Requirements for Operating at Various 92
Temperatures and Feed Slurry Concentrations
24 Residue Dewatering Costs Under Various Operating 93
Conditions
25 Net Benefits Resulting at Various Temperatures and 96
Retention Times, Feed Slurry - 20% Solids
26 Net Benefits at Optimum Temperature for Various 97
Retention Times and Feed Slurry Concentrations
27 Gas Production Costs for Various Temperatures and 99
Feed Slurry Concentrations
Vll
-------
LIST OF TABLES
Page
Number —a-
1 Growth Constants and Endogenous Respiration Rates 9
2 Chemical Components of a Sewage Sludge 15
3 pH Effect on Density of Cellulolytic Bacteria 18
4 Summary of Hydrolysis Rate of Cellulose in Aerobic 20
Fermentation
5 Vapor Pressure of Water at Various Temperatures 22
6 Effect of Temperature on K] of Carbonic Acid 22
7 Henry's Constant for Various Temperatures 23
8 Test Cloth Specifications 35
9 Composition of Refuse (Initial Determinations) 37
10 Experimental Determinations on the Feed Refuse 38
11 Suimiary of Operating Conditions at 35°C - Phase 1 42
12 Composition of Gas Produced - Phase 1 43
13 Apparent Volatile and Total Solids Reduction 44
14 Apparent Reactor Gas Production Per Unit Volatile 44
Solids Destroyed
15 Fixed Solids Balance 45
16 Adjusted Volatile Solids Destruction 46
17 Reactor Loadings - Phase 2 47
18 Reactor Operating Conditions - 35°C 48
19 Reactor Operating Conditions - 40°C 50
20 Reactor Operating Conditions - 45°C 53
21 Gas Composition and Production - 35°C 55
22 Reactor Operating Conditions - 50°C 57
viii
-------
TABLES (Continued)
Number Page
23 Reactor Operating Conditions - 55°C 59
24 Reactor Operating Conditions - 60°C 60
25' Gas Composition and Production - 60°C 62
26 Effect of pH and Alkalinity on Gas Composition 64
27 Filterability of Reactor Residue 71
28 Composition of a Typical Domestic Refuse 73
29 Shredding Cost for 5 Inch Grate Size (Gondard Mill) 80
30 Annual Costs of Combinations of Mills and Work 81
Shifts (Tollemache Mill)
31 Annual Cost of Combinations of Mills and Work 81
Shifts
32 Gas Production (Cu Ft/Lb Dry Solids) 84
33 Volatile Solids Destruction (Percent) 84
34 Operation and Capital Cost - Plant Capacity of 100 94
Tons per Day of Dry Refuse - 20 Percent Field Slurry
35 Alkalinity and Volatile Acids Determinations - 35°C 111
36 Alkalinity and Volatile Acids Determinations - 40°C 112
37 Alkalinity and Volatile Acids Determinations - 50°C 112
38 Alkalinity and Volatile Acids Determinations - 55°C 113
39 Reactor Slurry Concentrations Under Various 123
Operating Conditions - Percent Solids
40 Mixing Costs for Various Temperatures and Retention 124
Times - $1000 per Year
41 Heating Costs for Various Operating Conditions - 125
$1000 per Year
ix
-------
TABLES (Continued)
Number Page
42 Dewatering Cost for a Feed Slurry Concentration of 126
20 Percent Solids - $1000/Year
43 Total Production Costs for All Operating Conditions - 127
$1000/Year
44 Net Benefits Accrued Under All Operating Conditions - 129
$1000/Year
45 Gas Production Costs Under Various Operating 131
Conditions - tf/1000 cu ft
-------
CONCLUSIONS
An effective microbial system was developed for producing methane
from organic refuse. As long as due consideration was given to
the nutritional and environmental requirements of the micro-
organisms, the system functioned very well.
The addition of raw sewage sludge to the reactors markedly
improved the biological conversion even though adequate major
nutrients, nitrogen and phosphorus, were supplied separately.
Inadequate alkalinity was present in the reactors to maintain
the pH above 6.6. It was necessary to add caustic for pH control.
The caustic requirements decreased with increasing temperature
and retention time.
At 35°C, the gas production was 1.4 and 3.05 scf per pound of
refuse added at,4 and 30 days retention time respectively. At
60°C, the gas production was 4.23 and 4.96 scf per pound of
refuse added at 4 and 30 days retention time respectively.
The methane content of the dry gas produced varied from 54 to
70 percent. The methane content decreased with increasing
retention time and increasing temperature.
A temperature of 42.5°C yielded the maximum gas production in
the mesophilic temperature range while the maximum gas production
in the thermophilic temperature range occurred at 60°C.
An economic analysis of the unit processes, excluding shredding,
separation and disposal costs, for a potential system processing
100 tons per day of dry refuse yielded the following conclusions:
1. For all combinations of operating conditions, amortization
of capital accounted for a major portion of the total
annual costs. For feed slurry concentration of 20
percent, capital amortization accounted for 44 to 69
percent of the total costs.
2. At high reactor slurry concentrations and long retention
times, mixing costs account for approximately one-fifth
of the total annual costs.
3. The heating costs at the 60°C temperature accounted for
39.3 percent of the total annual cost for a high feed
slurry concentration and short retention time. The
heating costs at 35°C accounted for less than 10 percent
of the total annual costs.
XI
-------
4. The maximum net benefits in the mesophilic range occurred
at the higher feed slurry concentrations, but net
benefits were relatively insensitive to reactor retention
times.
5. The maximum net benefits in the thermophilic range
occurred at the maximum feed slurry concentration and
the minimum retention time tested.
6. The cost of producing gas by this process was 11.0 and
7.8 cents per 1000 scf at the optimum mesophilic and
thermophilic temperature respectively with a feed slurry
concentration of 20 percent solids.
The shredding and separation processes substantially increase the
cost of producing gas. Using a cost for these processes of $2.00
per ton, the production cost of the gas is increased by 27 and
24 cents per 1000 scf for the optimum mesophilic and thermophilic
temperature respectively.
A benefit will accrue from the disposal of the residue. Processing
100 tons of refuse as received will yield 73.4 to 89.4 tons of
residue that will require only one-third the landfill volume
required of the unprocessed refuse.
xii
-------
RECOMMENDATIONS
The requirements for an effective biological conversion of the
refuse to methane have been elucidated. The major work that
needs to be undertaken involves an evaluation of the physical
processes accompanying the biological process. Therefore, it is
recommended that a large scale pilot study be undertaken to
evaluate the physical processes. This study should develop the
necessary design parameters for the shredding and separation of
the raw refuse, mixing and heating requirements for the reactor,
residue dewatering and disposal, and gas purification. A
simulation model should be developed to predict the operation
of the system under various operating conditions such that an
optimum or near optimum design of a full scale system can be
obtained.
Upon completion of an economic evaluation of the pilot study, a
decision can be made regarding the economic attractiveness of
the process. Based upon the economic analysis conducted in
this study, the process has potential for profitable operation.
These analysis should be verified and/or adjusted with a pilot
scale study.
xm
-------
INTRODUCTION
Among the many challenges confronting the world as a whole, and
this country in particular, are two seemingly unrelated problems.
The first and perhaps the more obvious of the two is a problem
of solid waste management. In particular, it is a problem of
solid waste disposal. The second is the decline in energy
resources as evidenced by the increasing shortage in natural
gas reserves. The development of a process to convert organic
refuse into methane would provide at least a partial solution to
both of these problems.
Solid Waste Disposal
In solid waste management programs disposal always plays a major
role. This is a result of the ever-increasing amounts of solid
waste generated in this country. On a per capita per day basis
it was estimated that the urban refuse contribution alone amounted
to seven pounds in 1967 (Eliassen vt aJL.y 1969). The figure
increases to ten pounds when industrial solid wastes are included.
The problem is indeed monumental, particularly when one considers
the increased urbanization of the population of the United States.
Existing techniques for the disposal and processing of solid
wastes consist primarily of landfill and incineration. Composting
is used on a very limited basis. In a broad sense all current
disposal practices are concerned with the return of the waste
material to the land in either of two forms. In the raw form
solid wastes are disposed of in landfills. In the processed
form, incinerator residue must also find its final repository in
landfills. Where composting is practiced, it too is returned to
the land as a soil additive.
Metropolitan areas are finding the disposal problem ever more
difficult as nearby landfill sites become exhausted. Composting
suffers the disadvantages of high production costs as well as a
very limited market for the final product. As a result it has
fallen into disfavor and few communities practice it any longer.
Therefore, with present technology the landfill must be the
ultimate recipient of the generated solid wastes. As the land
areas available for use as landfill sites become more and more
distant from the source of solid waste generation, the costs
attributed to the transportation of these wastes increase. It
follows that volume and weight considerations become very important.
Incineration currently provides reductions in the quantity of
refuse up to 70 or 80 percent depending upon the refuse composi-
tion. This process is expensive and is likely to become even
more so as air pollution standards become more stringent. A
process achieving reductions in weight and volume at economics
competitive with incineration, and not suffering its air pollution
control limitations, would be very attractive indeed.
-------
Energy Shortage
The total energy demand in the United States rose from 45
quadrillion BTU in 1960 to over 60 quadrillion BTU in 1970 and
is expected to top the 90 quadrillion BTU mark by 1980. By the
year 2000, this energy demand is forecasted to reach 200 quad-
rillion BTU (ChemicjaJt and Engineering New6, 1970a). The share
of natural gas (methane) in this energy market was about 31 per-
cent in 1968 (Chenu.co£ and Engineering NewJA, 1970b). This total
energy demand, and in particular that on natural gas, cannot
continue without exhausting the available fuels in the not unfore-
seeable future. With the exception of natural gas, natural resources
are currently abundant but not unlimited and efforts should be
made to conserve these resources.
The outlook for continued expansion of natural gas consumption is
not particularly good. Gains in production of gas between 1947
and 1968 averaged more than six percent a year with the production
growing from 5.6 to 19.3 trillion cubic feet. During the same
period reserves increased on an average of 2.5 percent annually
or from 165 to 282 trillion cubic feet. The reserves-to-product
ratio has declined from 29.5 to 14.6 years (CkwicaJL and
Engineering Mewa, 1970b). By the end of 1969 the reserves were
down to 275.1 trillion cubic feet while production rose seven
percent to 20.7 trillion cubic feet (Chem/ico-2 and Engi.ne.vu.n9
Newa, 1970c). Even with improvements to reserves it is estimated
that production of gas in the 48 contiguous states will be
limited to about 27 trillion cubic feet a year for 1974 to 1990.
With gas from Alaska, U. S. production could reach more than
35 trillion cubic feet a year by 1990 and hold at this level for
nearly ten more years (Chemical and Engineering Wewi6, 1970b).
Figure 1 illustrates the forecasted future situation in natural
gas requirements and supplies.
Even with optimistic views on reserves and future production,
demand for gas will remain greater than additions to reserves.
An increase in additions to reserves will only give lead time to
develop other sources of gas—imports from Canada, imported liquid
natural gas and coal gasification (Chemical and Engineering Newd,
1970b).
Refuse as a Source of Fuel Gas
In wastewater treatment anaerobic digestion has been used for a
long time to effectively reduce the quantity of organic sludges
and to transform them into stable, easily dewaterable residues.
The reduction is accomplished by the biological conversion of the
organic solids to methane and carbon dioxide. This anaerobic
process has a very low efficiency of biological energy conversion.
-------
T
50
0)
0)
40
30
20
Production in contiguous 48 states
Production in 48 states plus Alaska
Requirements
1968
1972
1976
1980
1984
1988
1992
Figure 1. Future Natural Gas Supplies and Requirements
(ChejuicaJt and Eng-tnee^/tng MewA, 1970b)
-------
Most of the energy of the substrate is lost. The gas produced has
a high calorific content and is a valuable end product with
potential for reclamation. The heat value of the gas produced
from the digestion of sewage sludges ranges from 7000 to 9000 BTU
per pound of organic matter destroyed, depending on the nature
and composition of the solids to be digested.
Various studies have demonstrated that organic refuse, particu-
larly garbage, is amenable to anaerobic decomposition as long
as the proper environmental conditions of temperature, pH,
absence of oxygen, etc. and adequate nutrients are maintained.
These will be reviewed in more detail later. Organic refuse is
composed basically of the same compounds, carbohydrates, proteins
and fats, as sewage sludge though in different relative propor-
tions. There appears to be no reason why refuse cannot undergo
adequate anaerobic decomposition. Proper environmental condi-
tions must be maintained, the deficient nutrients supplied and
toxicity problems eliminated.
In addition to providing a means of energy conservation this
process provides a significant reduction in the quantity of
waste material for ultimate disposal. Assuming that about 70
percent of domestic refuse is biodegradable and that 70 percent
of the biodegradable portion is gasified in an anaerobic process,
the resulting overall reduction in weight would amount to about
50 percent. This will greatly reduce transportation costs to
the landfill sites as well as extend the life of the sites.
Inherent in this is the assumption that the solids in the digester
effluent will dewater easily.
Moreover, this residue exhibits a very important advantage over
the present solid wastes that are placed in landfills. Untreated
refuse has not been stabilized biologically and is in a process
of decay. When placed in a landfill it decomposes and can pollute
the groundwater if adequate preventive measures are not taken.
It can also cause substantial settlements in the fill area that
may render it unsuitable for any kind of structural support and
real estate development. These problems are far less severe
with digester residue since it has been biologically stabilized.
The development of systems of this type will be a big contribu-
tion to the overall problem of solid waste management.
Assuming the degree of conversion mentioned above, a per capita
contribution of domestic refuse of seven pounds per day and a
heat value for the refuse digester gas equal to that of sewage
digester gas, the per capita contribution of energy would range
between 24,500 BTU and 31,500 BTU daily. With an average value
of 28,000 BTU per capita per day, this translates to about 2.0
quadrillion BTU per annum for the whole population of the United
States, about 3.3 percent of the total 1970 United States energy
-------
demand. This will satisfy about 11 percent of the 1970 demand
for natural gas. In terms of dollars, using a fuel value of
50 cents per million BTU, this means a fuel value of about one
billion dollars per year. This, of course, is a gross figure and
does not include the cost of the refuse conversion. In addition
this gas would be produced in the urban areas, eliminating the
need for long distance transportation.
Purpose of the Study
This study was designed to evaluate the potential for reclaiming
methane gas from organic refuse by means of anaerobic fermentation.
The mechanism of methane fermentation has been studied extensive-
ly, elucidating the constraints that must be satisfied for the
system to function effectively. The problems likely to be
encountered in the fermentation of organic refuse were unknown.
It was the purpose of this laboratory study to define these
problems and develop information to evaluate the feasibility of
using this process on a large scale.
Studies were undertaken to determine the response of the methane
fermentation process to external variables such as temperature,
retention time, pH and feed slurry concentration. Potential
operational problems were identified. The environmental and
nutritional requirements for the fermentation of urban refuse
were evaluated. The quality and quantity of gas production were
determined for various operating conditions. From the data
collected it was possible to evaluate the practicality of
employing this process as a means of reducing the quantity of
refuse for disposal while reclaiming a useful product, methane.
-------
THEORETICAL CONSIDERATIONS
Basic Fundamentals of Anaerobic Digestion
Anaerobic treatment of complex organic materials is considered to
be basically a two-stage process as indicated in Figure 2. In the
first stage, a group of faculative and anaerobic bacteria, known
as the "acid formers," act upon the complex organics and change
the form of complex fats, proteins and carbohydrates to simple
soluble organic materials. The end products of this first stage
conversion are primarily short chain organic acids, also known
as volatile acids, and small amounts of bacterial cells. This
stage accomplishes no stabilization of the waste material but it
is an essential prerequisite for the second stage in which the
actual stabilization of the waste matter occurs. It places the
organic matter in a form suitable for the second stage of
treatment.
CO,
CO,
Complex
Organics
1
Acid
Formation
Organic
Acids
I
Methane
Formation
CH.
+H
C0«
Figure 2. Anaerobic Stabilization of Complex Organics
In the second stage, the short chain organic acids are acted
upon by a strictly anaerobic group of bacteria, termed the
"methane formers," and are converted to gaseous end products,
methane and carbon dioxide. The methane formed in this second
stage, being insoluble in water, escapes from the system. It
can be collected for use as a fuel. The carbon dioxide involved
partially escapes in the gaseous form and partially goes into
solution. It is in this second stage that stabilization occurs
through the removal of oxygen demanding material in the form of
methane gas. Cell production is also minimal compared to aerobic
processes. This is a direct result of the high energy content of
the products, in particular methane (McKinney and Conway, 1957).
This is an advantage as the amount of solids requiring ultimate
disposal is minimized by eliminating any significant microbial
protoplasm production in the process of stabilizing the organic
material.
-------
There are several different groups of methane formers with each
group characterized by its ability to ferment only a specific
number of compounds. Therefore, several different bacteria are
required for stabilization of the organic material (McCarty,
1964a). The most important of these methane formers, those
utilizing acetic and propionic acids for substrates, grow quite
slowly and for most organic compounds are rate limiting at the
lower retention times. For sewage sludge digestion, Pfeffer
(1968) found that in systems operating with solids retention
times of approximately ten to fifteen days, the rate limiting
step is methane fermentation. When such systems are operated at
solids retention times greater than fifteen days, the rate
limiting step then becomes the hydrolysis of organic solids.
Tracer studies have indicated the major sources of methane
(McCarty, 1964a). One source is the direct cleavage of acetic
acid into methane and carbo'i dioxide. The methyl carbon of
acetic acid, together with its three hydrogen atoms, is
converted intact into methane gas. The carbonyl carbon is
converted to carbon dioxide. Most of the remaining methane is
formed from the reduction of carbon dioxide. Carbon dioxide,
functioning as an electron accepter, is reduced by hydrogen atoms
enzymatically removed from organic compounds. The availability
of carbon dioxide .in such a step is never a rate limiting factor
as there is always a large excess of it from the bicarbonate
buffer system that is present in anaerobic systems (McCarty,
1964a).
Organic destruction in anaerobic treatment is directly related
to methane production and vice versa. Buswell and Mueller (1952)
developed Equation 1 to predict the quantity of methane from a
knowledge of the chemical composition of the waste:
CnHaV <«-$-!> H20 *(£ - f + £) C02 <•
(M-|)CH4 (1)
McCarty (1964a) showed that the theoretical methane production
from the complete stabilization of one pound of BODL or COD was
5.62 cubic feet at standard temperature and pressure. From this
he forwarded Equation 2 for estimating methane production from
waste strength:
C = 5.62 (eF - 1.42A) (2)
8
-------
where: C = cubic feet of Cfy produced per day (STP)
e = efficiency of waste utilization
F = pounds of BODi added per day
A = pounds volatile biological solids produced per day.
The efficiency of waste utilization, e, should not be confused
with the stabilization efficiency. It is a factor designating
the efficiency of the conversion of the organic waste to both
the gaseous form and to biological solids. Whereas the former
form represents waste stabilization, the latter conversion
represents no such thing. The pounds of volatile biological
solids produced per day, A, can be estimated from Equation 3
(McCarty, 1964d):
A = a
H
+ b(SRT)
where: a = growth constant
b = endogenous respiration rate
SRT = solids retention time in days.
Values for a and b are shown in Table 1 for various organic
compounds (Speece and McCarty, 1964).
Table 1. Growth Constants and Endogenous Respiration Rates
Substrate Growth Constant Endogenous Respiration
a Rate, b
Fatty Acid
Carbohydrate
Protein
0.054
0.240
0.076
0.038
0.033
0.014
The percentage of added BODi which is stabilized, S, is given by
Equation 4 (McCarty, 1964d):
q _ 100C
^ " 5.62 F
= 100(eF - 142 A)
-------
The efficiency of waste stabilization is related to the solids
retention time. As the solids retention time decreases the
relative proportion of active cells washed out of the system
increases. If the solids retention time falls below a certain
limit, the microorganisms responsible for anaerobic digestion will
be washed out faster than they can reproduce themselves and the
result is failure of the process. The minimum rentention time
is dependent on the temperature and the types of microorganisms
in the system under consideration. This latter is dependent on
many factors, one of which is the type of substrate being utilized.
Though it is possible to operate near the minimum retention times,
efficiencies are low and process dependability is poor. Solids
retention times of at least two and a half times the minimum are
recommended (McCarty, 1964d).
While there are many different methane forming bacteria, there
are also many different acid forming bacteria. For attaining
good efficiencies of waste stabilization, a proper balance among
all these different organisms is required. The establishment and
maintenance of this balance dictates operation under optimum
environmental conditions. These are discussed in brief in the
next paragraphs.
Temperature is a very important operational parameter in anaerobic
fermentation process. As temperatures increase, rates of reaction
proceed much faster and this results in more efficient operation
and lower retention time requirements (McCarty, 1964b). Two
optimum temperature levels have been established. In the
mesophilic level the temperatures range from 30°C to 37.5°C and
in the thermophilic level, they range from 49°C to 51°C.
Although the rates of reaction in the thermophilic level are
much faster than those in the mesophilic level, the economics of
most sewage sludge digestion systems have dictated operation in
the mesophilic level or lower (McCarty, 1964b). This stems from
the fact that the methane requirements to maintain thermophilic
temperatures in most digesters are excessive and uneconomical.
This, in turn, is a result of the inability to thicken the feed
sludges sufficiently such that the organic loading and the
resulting methane production per unit digester volume are
sufficiently high to make such an operation economically attrac-
tive.
Another environmental requirement for anaerobic treatment is the
maintenance of anaerobic conditions in the digester. The methane
formers are strict anaerobes and even small amounts of oxygen can
be quite detrimental to them. This necessitates, in most cases,
a closed digestion tank which is also convenient because collection
of the methane produced is facilitated.
10
-------
The third environmental requirement for optimum operation is that
for a proper pH. McCarty (1964b) reports that anaerobic treatment
can proceed quite well with a pH varying from about 6.6 to 7.6 with
an optimum range of about 7.0 to 7.2. Beyond these limits anaerobic
digestion proceeds with decreasing efficiency until at a pH of
6.2 and lower, the acid conditions become quite toxic to the methane
formers and waste stabilization comes to a virtual halt.
Control of pH should be exercised when the pH appears likely to
drop below a value of 6.6. This is done by the addition of an
alkali. In sewage sludge digestion the use of lime for such
control has been the most widespread, but because of its many
advantages over lime, sodium bicarbonate has lately been receiving
increasing attention as a substitute for lime for pH control.
The bacteria responsible for waste conversion and stabilization in
the anaerobic process require nitrogen, phosphorus and other
materials in trace quantities for optimum growth. Therefore,
another important environmental condition is the presence of the
required nutrients in adequate quantities. Municipal waste sludge
usually contains all the required nutrients in adequate quantities
but other substrates, industrial and solid wastes in particular,
might not. If the nutrients are not present in the required
quantities, they must be either added or supplemented.
McCarty (1964a) calculated the nitrogen and phosphorus require-
ments based on an average chemical formulation of biological cells
of C5Hg03N. This yielded a nitrogen requirement of about 11
percent of the cell volatile solids weight and a requirement for
phosphorus equal to about one-fifth of this figure. Although
these requirements should theoretically be based on the fraction of
waste removed during treatment rather than on the waste added,
it is good practice to base them on waste additions (McCarty,
1964a). Other elements having stimulatory effects at low
concentrations include, but are not limited to, sodium, potassium,
calcium, magnesium and iron (McCarty, 1964c). All of these
preceding elements exhibit inhibitory effects at higher concen-
trations.
The fifth and final environmental requirement for successful
anaerobic treatment is that the waste be free from toxic materials.
This is particularly true for concentrated organic wastes which,
though normally are more susceptible to anaerobic treatment, are
also more likely to have high or inhibitory concentrations of
various materials ranging from inorganic salts to toxic organic
compounds.
Some alkali and alkaline earth-metal salts above certain concen-
trations exhibit degrees of inhibition and toxicity. The threshold
levels vary depending on whether these metals act singly or in
11
-------
combination. Certain combinations have synergistic effects,
whereas others display antagonistic effects.
Ammonia, particularly when in the ammonia form, is inhibitory when
present in high enough concentrations. At concentrations between
1,500 and 3,000 mg/£ and a pH greater than 7.4, the ammonia concen-
tration can become inhibitory. At concentrations above 3,000 mg/£,
the ammonium ion itself becomes quite toxic regardless of pH
(McCarty, 1964c).
Other common forms of toxicity include those of sulfides, heavy
metals and toxic organic materials. Concentrations of soluble
sulfide varying from 50 to 100 mg/£ can be tolerated in anaerobic
treatment with little or no acclimation, whereas concentrations up
to 200 mg/£ can be tolerated with some acclimation (McCarty, 1964c).
Low soluble concentrations of copper, zinc and nickel salts are
associated with most of the problems of heavy metal toxicity in
anaerobic treatment. Also, there are many organic materials that
exhibit inhibitory effects. These range from organic solvents to
many common materials such as alcohols and long chain fatty acids
in high concentrations (McCarty, 1964c).
It is well to know that microorganisms usually have the ability to
acclimate to some extent to inhibitory concentrations of most
materials if the process is acclimated to the inhibitory substance.
Also, it should be recognized that only materials in solution can
be toxic to biological life (McCarty, 1964c). Control of toxicity
or inhibition can, in general, be achieved by one or more of three
ways: 1) the removal from the waste stream or inactivation of
toxic materials by such means as chemical precipitation, 2) the
dilution of the waste stream below the "toxic threshold" of the
toxicity causing material, and 3) the addition of an antagonistic
material.
Anaerobic Digestion of Domestic Refuse
As mentioned earlier, the solids retention time is one of the most
important factors determining the degree of waste stabilization.
Increasing solids retention times result in increasing stabilization
efficiencies and higher gas production. Figure 3 (Rankin, 1948)
illustrates the relationship between solids retention time and the
degree of stabilization achieved for sewage sludge. However,
especially where the waste is dilute, high retention times become
associated with excessively large digestion tanks and also with
excessively high thermal requirements for heating. At a certain
minimum concentration, the anaerobic process, as conventionally
known, ceases to be economically feasible and either the anaerobic
contact process or a different method of treatment is employed.
The anaerobic contact process is based on exactly the same
principles as its conventional counterpart, except that solids
12
-------
Ill
10 20 30 40 50
Retention Time - Days
60 70
Figure 3. Variation of Volatile Solids Destruction
with Detention Time (Rankin, 1948)
13
-------
return is practiced. This way, the bacteria are not lost with the
effluent but are maintained in the system. This also permits the
maintenance of high solids retention times at low liquid retention
times.
No such special arrangements need be made when digesting organic
solid wastes. The moisture content of the feed slurry can be
adjusted to the required level as it only involves the addition of
water in the desired amounts. Therefore, this process will
capitalize on one advantage refuse has over sewage sludge, i.e. a
high solids content. It will allow for high loadings on the
systems while still maintaining the solids retention times so
necessary for stabilization of the organic material. With
substrates rich in proteins, concentrated feed slurries pose
problems. As the concentration of the solids in the feed slurry
increases, so does the ammonia concentration and the alkalinity.
If the volatile acids remain low due to effective methane fermen-
tation, the high alkalinities will cause an increase in the pH of
the system. This rise in pH causes a shift to the right in the
following equilibrium equation:
NhJ j NH3 + H+ (5)
If this happens, inhibition becomes related to the ammonia concen-
tration. As ammonia is inhibitory at a much lower concentration
than the ammonium ion (McCarty, 1964c), process failure could result
as explained in a previous section. This, however, is not expected
to be a problem with domestic refuse. With garbage assuming an
ever-decreasing role in the composition of domestic refuse, it would
be indeed fortunate if enough proteins were present to even
partially satisfy the nutrient requirements for nitrogen.
Golueke and McGauhey (1967) estimated that about 67 percent of the
refuse that can be converted biologically was found to be paper.
Garbage accounted for only 12 percent of the biodegradable refuse.
The total nitrogen content of urban refuse is extremely small unless
some industrial solid waste is contributing nitrogen. With this
proportion of carbohydrates and proteins in the feed, the system is
not expected to operate at high pH values or be faced with an
ammonia toxicity problem.
Concern has been expressed over the rate of biological breakdown of
such materials as celluloses, hemicelluloses and lignins. With the
exception of lignins which are known to be very resistant to
biological decomposition, it is expected that this will not pose
any serious problems. The percentage of lignineous materials in
typical domestic refuse, it is felt, will be small and restricted
to certain types of wood and newsprint (Golueke, 1970).
14
-------
Cellulose Reduction in Anaerobic Digestion
That cellulose hydrolysis could be the rate limiting step in
digestion of wastewater sludges was postulated by Maki (1954) in his
study of pure and mixed cultures of cellulolytic bacteria removed
from a municipal digester. The characteristics of the undigested
sludge which he used (see Table 2) is similar to that of organic
refuse. Therefore, it might be expected that the hydrolysis of
cellulose is the overall rate limiting step. This assumption
was based on the fact that acid production from the hydrolysis and
fermentation of cellulose was slower than the conversion of these
products to methane and other end products.
Table 2. Chemical Components of a Sewage Sludge
Component % Dry Weight
Hemicellulose - 6
Cellulose 34.5
Lipids 14
Protein 19
Ash 34
Source: Maki (1954)
Kinetic studies of the anaerobic digestion process by Lawrence and
McCarty (1967) and Andrews e* at. (1964) stated that methane
production was the overall rate limiting step. Pfeffer (1968)
pointed out that at solid retention times greater than ten to fifteen
days, hydrolysis of the complex organics becomes the rate limiting
step. As domestic refuse has as high or higher concentrations of
complex organics as the typical sludge on which these conclusions
were based, it is logical to assume that hydrolysis of the complex
organic materials will be the rate limiting step at shorter solids
retention times.
A recent study by Chan and Pearson (1970) on the hydrolysis of
cellulose indicated that hydrolysis of insoluble cellulose to
soluble cell obiose appeared to be the rate limiting step in the
anaerobic decomposition of cellulose. They showed that this rate
could be accurately- characterized by the Michaelis-Menten kinetic
model and gave the different kinetic constants and coefficients that
apply to the fermentation of cellulose. They also concluded that
not only is the cellulose hydrolysis to cellobiose the limiting step
15
-------
of the cellulose hydrolysis process, but also is the rate limiting
step in the overall cellulose fermentation process.
The structure of cellulose, shown in Figure 4, indicates that the
basic unit is the disaccharide cellobiose. This structure shows
a recurrency period of 10.25 angstroms which is double that of a
glucose unit using Haworth's pyranose ring structure.
)fT
H
H OH
COoOH
O-i
C02OH
H OH
Figure 4. Structure of Beta 1,4 Cellulose
Cellulose exists in nature as the fiberous material constituting
the cell wall of plants. The cellulose molecule varies in size with
its source. The degree of polymerization, number of glucose units
within a molecule, varies from 700 for cellulose from wood to nearly
4000 from cotton. The average molecular weight of cellulose is
400,000 corresponding to 2500 glucose units. Cellulose is not a
homogenous chemical compound. Celluloses are all built up by the
same substances but they may differ in average chain length,
distribution of chain lengths and macromolecular structure.
Additionally, the types of non-cellulosic substances contained
within the cellulose structure may differ.
Reese and Levinson (1952), while studying the enzymatic hydrolysis
of cellulose, proposed the following pathway for cellulose
decomposition.
Native
Cl
Linear
Chains
Cx
•—
Cellobiose
Glucose
Little was known about the mechanism of C,. More recent works by
Reese and others indicate that there is kill little quantitative
knowledge about this step C-j. Its presence is assumed due to the
lack of ability of certain organisms to utilize native cellulose
yet being able to attack linear cellulose. Cx represents various
cellulases. These enzymes break linear cellulose down primarily
to cellobiose. Other compounds have been reported to have been
produced by cellulases1 hydrolysis of linear cellulose. These
16
-------
other products include cellotriose, cellotetrose, glucose and
disaccharides other than cellobiose. The mode of hydrolysis, as
described by Reese (1956), is assumed to be both random attack
of linkage and removal of end units.
Tests run by Reese e* aJL. (1950) indicated the effect of pH on
various cellulases. Tests were performed on filtrates from various
cellulolytic bacteria indicating highest activity at pH values
between 5.0 to 7.0. Similar tests to show temperature effects
indicated continual increase in cellulase activity up to 55°C.
Bacterial Decomposition of Cellulose
Mesophilic Bacteria
Von Tappeiner (1884) first showed the importance of mesophilic
bacteria on hydrolysis of cellulose. His works were associated with
cellulolytic bacteria in the cattle rumen. Omelianski is reported
to be the first to isolate a cellulolytic bacteria in 1902. His
studies of by-products of cellulose decomposition indicated the
release of methane and carbon dioxide from one organism and only
hydrogen and carbon dioxide from a second organism. Subsequent
observers have not reported by-products similar to Omelianski's
second pure culture.
Pringsheim (1917) first found cellobiose and glucose to be inter-
mediates in the decomposition of cellulose. Enebo (1949) found
the same pathway for the thermophilic bacteria.
Heukelekian (1927) studied the effect of pH on cellulose digestion
by measuring the decomposition rate in limed and unlimed digesters.
Without liming the pH fell from 5.8 to 4.2 in seven days and then
gradually increased. Buffering with lime to a pH of 7.4 yielded
73 percent cellulose destruction within seven days. This compares
to 30 percent removal in the unlimed digester. Heukelekian
concluded that lime stimulates cellulose decomposition.
Another study by Heukelekian (1928) showed the effect of cellulose
content on gas production. He also studied the effect of lignin,
hemicellulose and complex celluloses on digestion. Digestion of
high grade, pure cellulose toilet tissue gave gas production of
0.57 liters of gas/gm of cellulose. Lower gas production with
higher percentages of carbon dioxide was found for lower grade
toilet tissue and newspaper. A noticeable lag time for initiation
of gas production caused Heukelekian to conclude that establishment
of proper flora to be the rate limiting step in cellulose decompo-
sition.
Hungate (1950) isolated mesophilic bacteria from several differet
sources. Optimum temperature for bacteria isolated from the rumen
17
-------
of cattle was found to be 38°C to 40°C. Isolation of three strains
from sewage sludge classified the bacteria as gram negative rods
and diplorods, extremely variable in size, particularly when
growing on cellulose. Optimum- temperature appeared to be 38°C.
Sampling from different stages in the two stage anaerobic digestion
process found no cellulolytic bacteria in the raw sludge. Cell
counts by the dilution method indicated that cellulolytic bacteria
concentration increased from the influent of the first stage to
the effluent in the second.
Maki (1954) isolated ten strains of cellulolytic bacteria from the
Pullman, Washington wastewater plant digester. Cellulose accounted
for over 50 percent of the solids in the digester. The digester
varied in pH range of 6.2 to 7.8 with 6.8 being average. Cell
counts on the digester sludge indicated up to 106 organisms per
liter. Studies on the pure cultures indicated three strains to
be highly cellulolytic. Comparison of cellulose removal by mixed
culture yielded a rate two and a half times greater than that of
the most cellulolytic pure culture. This implies that there is
synergistic action of by-product removal of the wastes produced
by cellulolytic bacteria by non-eellulolytic bacteria. Maki
suggests that the rate of fermentation of cellulose to be the rate
limiting step in wastewater sludge digestion.
Dubos (1938) studied the effect of pH and availability of nitrogen
on the density of cellulose decomposing bacteria in soil. Table 3
shows the optimum pH to be 6.5 to 8.2 by cell count. However, pH
had little effect on cellulolytic bacteria density when nitrogen
as ammonium sulfate was added.
Table 3. pH Effect on Density of Cellulolytic Bacteri,
pH of Number pH of
Soil Cellulolytic Bacteria Soil
(Bacteria/gm)
Number
f2!lulolytic Bacteria
(Bacteria/gm)
8.7
8.5
8.2
7.5
7.0
0 _
2.5 x 105
2.5 x 107
2.5 x 107
2.5 x 107
6.5
6.0
5.2
4.5
4.5
2.5 x 107
2.5 x 105
0
0
0
18
-------
Thermophilic Cellulolytic Bacteria
McBee (1948) summarized characteristics of thermophilic cellulolytic
bacteria. Morphologically, he described them as "slender, often
slightly curved, gram negative, spore-forming rods." All pure
cultures are obligate anaerobes. Optimum temperature range is 55°C
to 65°C. Growth occurs in pH range of 6.4 to 7.4. Growth and fer-
mentation occurred on cellulose, cellobiose, xylose and hemicellulose
in tests on pure cultures.
Enebo and Pehrson (1960) studied the thermophilic digestion of
cellulose from a pulping factory wastewater seeded with domestic
sludge. Rapid breakdown of cellulose was reported at a ratio of
cellulosic waste to domestic sludge of three to one. At a one to
one ratio, 85 percent of the cellulose was removed in only eight
days. This was achieved at a two percent initial solids concen-
tration. The cellulose concentration decreased from 9900 mg/£ to
1500 mg/£. Optimum temperature was between 55°C and 60°C. A
significant feature of the pulp factory waste was the high lignin
content. The cellulose measurement procedure was not adequately
explained.
Chan and Pearson (1970) summarized the reported hydrolysis rates
shown in Table 4. Comparison of these rates shows that there may
be only little significance in increase of hydrolysis due to
thermophilic digestion. However, consideration must be given for
differences in systems, cultures and substrates utilized in these
uncoordinated studies. Synergistic relationships at the microorga-
nism level are important as shown by Stranks (1956). His method
of by-product removal was dialysis. In mixed culture systems this
by-product removal would be done by other organisms. Wastewater
sludges have been shown to contain significant quantities of
cellulose. Therefore, the known higher rate of digestion at
thermophilic temperatures would suggest higher cellulose destruction
rates.
Gas Composition
In the conversion of carbohydrates to carbon dioxide and methane,
equal volumes of each gas are produced. This reaction is shown in
Equation 6. However, all of the carbon dioxide is not released as a
(C6H1Q05)X + xH20 -»• x C6H1206 •* 3x CH4 + 3x C02 (6)
gas but enters into reactions with water and hydrogen ion. Micro-
organisms can deaminate biodegradable protein producing ammonia
which reacts with water according to Equation 7. The hydroxide
produced by this reaction reacts with carbon dioxide to form
19
-------
Table 4. Summary of Hydrolysis Rate of Cellulose in Anaerobic Fermentation
IM
O
Authority
Maki
Heukelekian
McBee
Stranks
System Initial
and Cellulose
Culture Concentration
Batch, mixed
2 pure cul-
tures from 2,000
sewage, 38°C,
mesophilic
Batch, mixed
culture from
sewage, 25°C, 3,120
mesophilic
Batch, pure
culture from (1 ) 744
soil and
manure, 55°C, (2) 2,980
thermophilic
Batch, mixed
culture from
rumen, 60°C, 41,200
thermophilic
Cellulose pH
Material
Whatman's
#1 filter
paper 6.8
Cellulose
in
sewage 7.4
sludge
Absorbent
cotton
Whatman's
#2 filter
paper 6.5
Hydrolysis
Rate
(mg/£-day)
(1) 260
(2) 660
142
(1) 149
(2) 426
11,400
-------
NH3 + HOH * NH4+ + OH" (7)
bicarbonate ion as shown in Equations 8 and 9. Therefore, the protein
content of the substrate will significantly effect the quantity of
carbon dioxide actually released from solution as well as the
bicarbonate buffer capacity of the system.
C02 + HOH J H2C03 * H+ + HCOg (8)
H2C03 + OH" J HC03 + HOH (9)
The carbon dioxide incorporated in the bicarbonate ion is removed
from the reactor in the liquid phase rather than the gas phase.
There are two factors related to this washout of carbon dioxide.
The concentration of alkalinity will control the washout rate at
a given retention time. The retention time, or the liquid
through-put rate, will also influence the washout rate. Therefore,
for a given substrate, digestion at shorter retention times will
produce a gas having a higher methane content.
The bicarbonate ion is not the only ion contributing to the
alkalinity or buffer capacity. The organic acids and acid salts
also contribute to the alkalinity. Equation 10, originally developed
by Pohland and Bloodgood (1963) and modified by McCarty (1964b),
can be used to calculate the bicarbonate alkalinity.
BA = TA - (0.85)(0.833) TVA (10)
where: BA = bicarbonate alkalinity, mg/£ as CaC03
TA = total alkalinity, mg/£ as CaCO^
TVA = total volatile acid concentration, mg/£ as acetic
acid
The mg/£ of acetic acid is converted to the equivalent alkalinity
as CaC03 by the 0.833 multiplier. The 0.85 factor accounts for the
fact that only 85 percent of the volatile acid alkalinity is
measured by titration of total alkalinity to pH 4. This equation
also assumes that there is no significant concentration of other
materials such as phosphates, silicates, or other acid salts which
also produce alkalinity.
The above analysis permits an accurate evaluation of the actual
bicarbonate ion concentration in a digestion system. Since the
normal operating pH in these systems is 7.0 or less, the carbon
21
-------
dioxide-bicarbonate equilibrium can be combined with the solubility
of carbon dioxide to predict the gas composition under various
operating conditions. One must also include the effect of the
vapor pressure of water. This increases substantially at the
thermophilic temperatures. These data are shown in Table 5. The
added water vapor in the gas reduces the partial pressure of carbon
dioxide which in turn changes the bicarbonate alkalinity equilibrium.
Table 5. Vapor Pressure of Water at Various Temperatures
Temperature °C 30 35 40 45 50 55 60
Vapor Pressure—mmHg 31.8 42.2 55.3 71.9 92.5 118 149.4
Source: Handbook otf Chem
-------
Table 7. Henry's Constant for Various Temperatures
Temperature °C
K x 10-7
30
0.139
40
0.173
50
0.217
60
0.258
Source: Handbook otf ChwtiA&iy and Pky&i.cA, 40th ed., CRC
Publishing Company
in a significant reduction in alkalinity. A decrease in the
solubility of carbon dioxide by a factor of 0.85 in going from 30°C
to 60°C in conjunction with an increase in the vapor pressure of
water by a factor of 4.7 produces a significant reduction in the
carbon dioxide in solution. This effect is not offset by the changes
in the carbon dioxide-bicarbonate relationship with increasing
temperature.
The effect of bicarbonate alkalinity, pH and temperature can be
calculated from the following equation. This equation will yield
18 x 10"3 [H+] [HCO~]
where: PCQ = Partial pressure of C02 in gas phase - mm Hg
K = Henry's constant for
[H ] = Hydrogen ion concentration - mole/£
LHCOj] = Bicarbonate ion concentration - mole/£
K = Dissociation constant for carbonic acid
the partial pressure of carbon dioxide in a gas which also will have
a substantial quantity of water vapor. The gas will be dried prior
to analysis or combustion. The partial pressure of carbon dioxide
will increase as a result of the elimination of the water vapor.
Equation 13 will give the partial pressure of carbon dioxide in a
dry gas. This value represents the carbon dioxide composition of
the gas as determined by a gas analysis.
(PCO ) =
C02 dry H20 2 wet
23
-------
Temperature has a significant effect on the carbon dioxide-bicarbonate
system. Increasing the temperature significantly increases the
partial pressure of carbon dioxide for a given pH and alkalinity.
These data are shown in Figure 5. At the higher temperatures,
considerably more carbon dioxide will be lost from che system in
the gas phase. For a given pH, a lower alkalinity will occur at
the higher temperatures. More of the carbon dioxide will be lost
in the gas rather than be discharged from the system as bicarbonate
alkalinity. With more carbon dioxide being discharged in the gas
phase, the caustic requirements for pH control would be significantly
reduced.
24
-------
50
40
30
20
10
0)
OJ
Q.
o
50
40
30
*j
| 20
I
01 10
I o
a
(O
o
50
40
30
20
10
0
60°C
50°C
40° C
1.0 2.0 3.0 4.0
Bicarbonate Concentration - g/1 as CaC03
5.0
Figure 5. The Relationship Between Gas Composition,
Temperature, pH and Alkalinity
25
-------
EXPERIMENTAL PROCEDURE
Experimental Reactors
Eight reactors of plexiglass construction were built. Each reactor
measured 10.5 in. in diameter and 18 in. in height, and had a volume
of approximately 24 liters (Figure 6). Except for the reactor top,
which was bolted down to the reactor wall and sealed with pliable
Permatex, all plexiglass was fused together with ethylene dichlonde.
Each reactor was fitted with a stainless steel helical worm stirrer
equipped at the bottom with a scraper extending almost the whole
reactor diameter. A 1.25 in. Bunting oil impregnated bronze
bearing fixed to the reactor bottom was used for stirrer alignment.
A Garlock Mechanipak seal maintained an airtight joint between
the extending 1.25 in. stirrer shaft and the plexiglass top. It
also transmitted the weight of the stirrer shaft and the
incremental force required for proper operation of the Mechanipak
seal to the plexiglass top of the reactor. A Boston Gear flanged
cartridge, connected to the stirrer shaft above the Mechanipak
seal served for proper alignment of the stirrer shaft and also
for transmitting the incremental pressure over and above that
supplied by the stirrer weight for proper Mechanipak seal operation.
A 1.25 in. rubber stoppered opening at the top of each reactor
served as an inlet for feeding purposes. A 0.25 in. diameter
plexiglass nozzle extending 2 in. above the reactor top served
to link each reactor to a Tygon tubing gas line leading to a
Sargent-Welch precision wet test gas meter. A 1 in. diameter,
1.5 in. long plexiglass nozzle at the bottom of each reactor
leading to 1.25 in. diameter Tygon tubing served for effluent
withdrawal. A double pipe clamp fixture was used to control the
effluent flow through the Tygon tube. Figure 7 is a pictorial
view of one battery of the reactors.
Units 1 through 4 were built and operated before units 5 through
8 were completed. The outlet nozzles on the former were located
on the cylindrical plexiglass wall about 1 in. above the plexiglass
bottom of the reactor. Because of plugging experiences with this
type of effluent arrangement, units 5 through 8 featured effluent
nozzles extending downward from bottom of the reactor. Reactors
5 through 8 had baffles located in the side walls to improve the
mixing by eliminating the circular motion induced by the mixing
mechanism.
Each four reactors, 1 through 4, and 5 through 8, formed a battery
and had their own framework of Uni-Strut construction and electric
motor. The stirrer shafts extending through the reactor tops in
each battery were connected by miter gears to an overhead shaft.
This latter, in turn, was connected to an electric motor. The
27
-------
ro
CO
•*•> •*• •*• •w-vir
Mechanlpak Seal
To Gas Meter
Reactors 1-4
10 1/2 in.
Reactors 5-8
Figure 6. Schematic of Experimental Units
-------
Figure 7. Laboratory Reaction Units (Battery Containing
Units 1 - 4)
29
-------
electric motors were 0.25 horsepower, 60 rpm, Dayton Gearmotors.
Both batteries were housed in a constant temperature room in which
the temperature could be varied from 35°C to 60°C. All reactors
were tested under pressure for gas tightness before operation and
the operating volume'constituting the reacting contents in each
reactor was 15 liters. The stirrers were operated in a continuous
fashion at 60 rpm.
Substrate
In the fall of 1969, a trip was made to the Center Hill Research
Facility, Environmental Protection Agency, in Cincinnati, Ohio,
for the purpose of obtaining a large, homogeneous amount of
"typical" domestic refuse that would last the duration of the
experimental runs. There, the contents of a collection vehicle
serving a residential area were shredded. By inspection, it was
determined that the size of the shredded refuse particles was about
one inch. It was also noticed that the refuse contained large
percentages of dried leaves and branches and paper but contained
very little garbage. Also, appreciable amounts of foam rubber,
plastic, nylon, styrofoam, etc. were present. No attempt was made
to segregate the broken glass and pieces of metal from the refuse
before shredding. These latter were also found in appreciable
amounts.
Eight 55 gal. drums of refuse were filled, sealed, and all but
one placed in cold storage at -20°C. The drum not placed in cold
storage served the purpose of providing refuse for waste character-
ization and later on, for feed to the reactors. Drums were taken
out of cold storgge one by one as the need for additional amounts
of refuse arose. The contents of each drum were allowed to thaw
and much of the broken glass and metals was removed by hydraulic
separation. Early experiences indicated that the presence of
large quantities of broken glass and metal caused excessive wear
on the moving parts and frequently caused the mixing mechanism
to jam. For a large scale application, it would be possible to
design a reactor to resist the wear from these abrasive materials.
However, it would be desirable to remove them prior to the digestion
step. These materials are inert to biological decomposition and
as such there is no point in introducing them into the reactor.
The refuse obtained on the initial trip was inadequate to complete
the study. It was necessary to return to the Center Hill Laboratory
for an additional supply of refuse in the spring of 1972. This
refuse was found to have the same characteristics as the original
sample.
Hydraulic separation of the broken glass and metals from the refuse
was done in small batches sufficient for about one week's operation
of the reactors. A very simple technique was used for the hydraulic
30
-------
separation. A 30 gal.plastic container was half filled with refuse
and the balance of the container was filled with tap water. The
refuse and water were manually mixed and allowed to stand for a
short period. Styrofoam, foam rubber, plastics and other floating,
biologically inert material were removed by hand. The glass,
metals and grit settled to the bottom of the container. The
degritted refuse was manually placed on a screen which allowed
the bulk of the water to drain from the refuse. Although this
process was not 100 percent efficient in removing the inert
materials, it did reduce the operational problems experienced
prior to this degritting step.
The degritted refuse was stored in a covered container in a
constant temperature room maintained at 2°C. Frequent moisture
determinations were made on this degritted refuse and sufficient
amounts of tap water added to obtain the desired solids content in
the feed slurrry. This feed slurry was added daily on a batch
basis. Prior to addition of the slurry, an equivalent volume of
reactor contents was withdrawn. This effluent was used for
analytical and other determinations. Although the feeding
procedure was conducted with care, some air was entrained in the
reactor vessel. Gas analysis showed that the oxygen was quickly
consumed and, because of the large volume of the reactor contents,
no oxygen toxicity was observed.
Nutrients were added directly to the digesters whenever the need
for nutrient addition arose. Nitrogen was fed in the form of
NfyCl and the concentration of ammonia nitrogen was maintained in
the 200 to 500 mg/t range. Whenever nitrogen was introduced to a
digester, phosphorus in the form of KoHP04 was also added such
that a nitrogen to phosphorus ratio of five to one was maintained.
The pH of the reactors was controlled by the addition of NaOH.
The pH was maintained in the range of 6.7 to 7.0. When ever the
pH dropped to near 6.6, enough NaOH was added to increase the pH
to a more desirable level. By experience it was determiend how
much alkali was needed to maintain the pH in the desired range.
NaOH was added directly to the daily feed slurry, resulting in a
relatively constant pH value.
Analytical Procedures
PH
pH measurements of the contents of the reactors were made on a
daily basis throughout the duration of this study. These readings
were made as soon after withdrawal of the samples as possible
using a Beckman Electromate pH Meter.
31
-------
Alkalinity
Digester effluent was centrifuged in a Sorvall Superspeed RC-2
at 6000 rpm (6 = 5,860) for 15 to 20 minutes. The supernatant was
then used for the alkalinity determination which was carried out
according to the procedure described in Standard Me£kodt> (1965).
Instead of using methyl orange to indicate the titration end
point as described in Standard Method*, the titration was assumed
to be complete at a pH of 4.5. All alkalinity reported are in
mg/£ as CaCOj.
Volatile Acids
A two-stage centrifugation procedure was employed. Samples were
first centrifuged as described for alkalinity determinations. The
supernatant was further centrifuged in the same centrifuge, using
a different size head at 15,000 rpm for 15 minutes (G = 27,000).
The resulting supernatant was then analyzed for volatile acids
(total) using the Column Partition Chromatography Method as described
in Standard Method* (1965). All values are reported in mg/l as
acetic acid.
Cellulose and Non-Cellulose Carbohydrates
The feed and effluent were regularly analyzed for cellulose and
non-cellulose carbohydrates. The Anthrone Method, as described
by Viles and Silverman (1949), Gaudy (1962), Snell and Snell
(1961), and Koehler (1952),was used for these determinations.
This proved to be a very sensitive test requiring much patience
and practice. For a step-by-step description of the procedure
followed, refer to Appendix A.
Gas Analysis
The gases generated were analyzed for methane and carbon dioxide
using a Model 25 V Fisher Gas Partitioner. This instrument is a
gas chromatograph designed for the quantitative determination of
substances which are in the gaseous state at room temperature.
The gas partitioner was coupled to a Model SR Sargent Recorder.
The heights of the peaks are used analytically to determine the
concentrations of the various gases in the mixture. This is
done through comparisons with peak heights obtained with a
standard gas mixture.
In this study, a standard gas mixture containing 40 percent
carbon dioxide and 60 percent methane by volume was used. The
resulting percentages of carbon dioxide and methane in the digester
gas were then normalized to 100 percent. A fresh standard was
prepared for each day on which gas analyses were conducted. Gas
32
-------
analyses were not conducted as frequently as desired due to
operational problems with the partitioner.
Fixed and Volatile Solids
Fixed and volatile solids determinations were made on a weekly
basis on both the feed and digester effluent. These determinations
were carried out according to the procedure described in Standand
Method* (1965) for sludge and bottom sediments with one modifi-
cation. No water bath was used prior to drying in an oven.
To assure as representative a sample as possible, samples were
blended in a Waring Commercial Blender and then poured back and
forth in a vigorous manner from one container to another prior
to placement in the evaporating dishes. To assure that the loss
of volatile solids in the determination of fixed solids is held
to a minimum, samples were dried at 70°C and at 103°C, as
recommended by Standard Method* (1965), and the results compared.
It was concluded that no significant loss of volatile matter were
encountered with the higher temperature which was subsequently
used for all fixed solids determinations.
Ammonia Nitrogen
Weekly determinations of ammonia nitrogen were made on the digesting
contents of all units for the purpose of assuring that sufficient
concentrations of nutrients were being maintained for optimal
digestion. Ammonia nitrogen concentrations were maintained
through this study at a minimum level of 100 mg/£ and a maximum
level of 500 mg/£ as nitrogen. In this manner adequate nutrient
levels were maintained without raising the ammonia nitrogen level
to toxic proportions.
Samples were first centrifuged as described for alkalinity determina-
tions, and the supernatant used for carrying out the analysis for
ammonia nitrogen according to the Distillation Procedure described
in Standard Method* (1965).
Other Analytical Methods Used for Refuse Characterization
These analyses were conducted according to the analytical
techniques recommended by the American Public Works Association
(Muru.cJ.pat. Rerfuae ViApotaJL, 1966) with the following exceptions.
The BTU value of the refuse was determined with a Parr-Bomb
Calorimeter using the technique recommended by the manufacturer.
The carbon, hydrogen and nitrogen analysis was conducted on an
F & M Scientific CHN Analyzer. The carbon and hydrogen values
were within the sensitivity range of this instrument. However, the
nitrogen values were too low to be detected. The wet Kjeldahl
33
-------
analysis was used to determine the nitrogen content of the refuse.
The chemical oxygen demand (COD) of the refuse was conducted
according to S^andoAd Me^tocia (1965).
Dewatering of Digested Solids
A major step in the processing of organic refuse for energy
recovery and for volume reduction is dewatering of the digested
solids. The filterability of the digested solids was examined
using a filter test leaf technique. This technique has been
described by Eckenf elder (1966). The procedure followed in this
study is described in Appendix B. As the test leaf and filter
cloths used were Eimco products, this company's literature was the
one mostly utilized and its instructions followed.
To obtain the various sludge concentrations investigated for
dewaterability in this study, the collected digester effluent
slurry was allowed to thicken by gravity to the desired concentra-
tion.
Although laboratory tests showed that the cake was completely
formed in five seconds, 30 seconds was selected for form time
for all test leaf runs, this time being considered the minimum
practical time for plant scale installations. No washing of the
product cake was practiced and the drying time was varied from
one to 2.5 minutes in increments of half a minute.
The laboratory vacuum outlets served as a vacuum source and
provided a static vacuum pressure of 18 in. of mercury throughout
the filter test leaf runs.
Attempts to keep the sludge well agitated during cake formation
were made using special magnetic stirring rods. However, these
attempts all failed and were consequently abandoned. The procedure
followed consisted of thoroughly agitating the sludge mixture with
a stirring rod immediately prior to immersing the test leaf. Due
to the fact that cake formation was very rapid, in the order of
five seconds, this method of agitation is believed to be satisfactory.
Suspended solids determinations of the filtrate were carried out
according to the procedure described in S£a.ndan.d M&tkodA (1965).
The Eimco test leaf used was circular in design and had a
diameter equal to four inches. Filter cloths were fixed to the
test leaf with an adjustable clamp band. The first two or three
tests with a new cloth were discarded as they are not an accurate
indication of fabric performance.
Three filter test cloths, all 1/1 plain weave, and made of polyethe
lene, were utilized in these dewaterability determinations. All
34
-------
were Eimco products. Table 8 summarizes the pertinent information
for each of these three test cloths.
Table 8. Test Cloth Specifications
Test Cloth
Number
Eimco Style
Number Material
Thread
Weave Count
1 PO-801 HF Polyethelene 1/1 Plain 112 x 48
2 PO-801 RF Polyethelene 1/1 Plain 105 x 40
3 PO-808 Polyethelene 1/1 Plain 40 x 23
35
-------
RESULTS
Characterization of the Raw Refuse
Upon acquisition of the refuse, analyses were run for the
purposes of defining the characteristics and chemical composition
of the raw refuse. These analyses followed hand separation of the
metals and glass, and are summarized in Table 9.
Table 9. Composition of Refuse (Initial Determinations)
Number of
Determination Determinations Average
Volatile total
solids
Owl 1 Uw
Lipids
Chemical Oxygen
Demand
Calorific value
Carbon
Hydrogen
Ni trogen
(CHN analyzer)
Nitrogen
(wet Kjeldahl)
Phosphorus
Total
carbohydrates
Cellulose
3
3
5
2
2
2
2
4
5
5
5
81.8%
6.2%
0.982 gm/gm
7510 BTU/lb
dry solids
44.1%
5.7%
f\Ol
0%
0.68%
t\at
0%
52.75%
35.8%
Range
81.6-82.1
5.98-6.36
0.943-1.021
7410-7610
43.68-44.50
5.625-5.750
• • ™
0.666-0.700
51.5-53.5
35.0-37.0
NOTE: The composition of non-cellulose carbohydrates is the
difference between total carbohydrates and cellulose.
During the course of the study, analyses were routinely made
on the raw feed refuse. The analyses included total volatile
solids, cellulose, and non-cellulose carbohydrate determinations.
These analyses were made on the refuse as fed to the reactors and
are summarized in Table 10.
It can be seen by a comparison of Tables 9 and 10 that differences
exist between the two results. The metals and glass in the
initial analyses had been removed by careful hand sorting as
opposed to the less efficient hydraulic separation technique
employed for preparation of the feed .for the reactors. This
accounts for the refuse solids data shown in Table 9 being
37
-------
Table 10. Experimental Determinations on the Feed Refuse
Number of Average Standard
Determination Determinations Result Deviation
Total volatile solids
Cellulose
Non-cellulose carbohydrates
25
44
46
73 %
22.3%
1.0%
±1.70 %
±0.63 %
±0.113%
81.8 percent volatile as opposed to only 73 percent for the data
in Table 10.
The cellulose determinations in the earlier series of experi-
ments were made on samples from which most of the fixed solids and
non-combustibles had been removed. The samples used for the latter
determinations were identical in composition to reactor feed and
contained an average 73 percent volatile solids. When this
difference in the volatile solids content of the sample is taken
into account, the results for cellulose presented in Tables 9 and
10 are comparable, within acceptable experimental and refuse
characteristic variation. However, the non-cellulose carbo-
hydrate determinations cannot be reconciled even if differences
in the total volatile solids content of the samples are taken
into consideration. If the samples used in the initial analyses
had contained 73 percent volatile solids instead of 81.8 percent,
the adjusted non-cellulose carbohydrate content would be
substantially greater than the one percent non-cellulose carbo-
hydrate content of the reactor refuse feed.
The non-cellulose carbohydrates measured by the procedure
described earlier consist mainly of starches, sugars, etc. These
organic compounds are easily biodegradable and are metabolized
by a host of aerobic, anaerobic and facultative microorganisms.
It is possible that subsequent to the removal of the refuse drums
from cold storage, sufficient biological activity occurred to
metabolize these compounds. Because of the relatively long
service life of each drum, sufficient time was available for
breakdown of the starches and sugars even at the low moisture
content of the refuse in the drums. Growth of fungi was noticed
on the refuse. Short-chain organic acids, alcohols, aldehydes,
etc. are known to be intermediaries in the anaerobic decomposition
of glucose. Under anaerobic conditions, the starches and sugars
would have been transformed into a form not measurable by the
non-cellulose carbohydrate procedure and may partially explain the
discrepancy in the two results. Aerobic stabilization may also
have occurred, generating carbon dioxide and water from these
compounds.
38
-------
The non-cellulose carbohydrates measured by the anthrone procedure
are water soluble. As such, a large portion of these compounds
would be expected to go into solution during the hydraulic
separation procedure used for preparation of the feed for the
reactors. Since the majority of the water used for hydraulic
separation was discarded, most of the solubilized non-cellulose
carbohydrates were undoubtedly discarded. This would also contrib-
ute to the discrepancies in the non-cellulose carbohydrate data
as seen in Tables 9 and 10.
Refuse constituents not analyzed for include, among others,
hemicelluloses and lignins. Hemicellulose, a cellulose-like
carbohydrate is only partially measured by the cellulose procedure
described earlier. The pentose portion of the hemicelluloses,
together with lignin and its derivatives, various organic resins,
pitch, proteins, plastics, rubber, nylon, etc., undoubtedly
constituted the balance of the refuse volatile constituents not
shown in Table 9.
Phase I
The first phase of this study was designed to determine if refuse
could be effectively converted to methane gas by the anaerobic
fermentation process. Upon completion of the construction of the
reactors, the experimental work was initiated. Two problems became
apparent soon after the reactors were started. The first was
mechanical in nature. The shredded refuse contained significant
quantities of glass and metal. This material caused excessive
mechanical problems and it became necessary to remove it prior to
digestion. A simple hydraulic separator was employed to remove
over 90 percent of the glass and metal. Most of the dust and
other fine inert particles were not removed by this separation
step. This material is not biodegradable and simply occupies space
in the reactor. High mixing velocities and associated high power
requirements would be necessary to prevent the deposition of this
material in the reactors. Consequently, an early decision was
made to separate essentially all of the glass and metal from the
shredded refuse before introducing it to the reactors.
The second problem was one of acclimation of the microorganisms
to the refuse. Actively digesting sludge was obtained from the
local water pollution control plant to use as a seed for the
reactors. The initial operation of the process was poor. Rapid
stabilization did not occur as was expected. Careful control of
the pH and organic acids prevented failure of the system. After
a four week acclimation period, the process began to respond; the
organic acid level began to decrease and gas production was
substantially increased. The reactors continued to efficiently
digest the refuse for the balance of this phase of the project.
39
-------
Refuse is known to be nutrient deficient, having an excessively
high carbon to nitrogen and carbon to phosphorus ratio. Active
biodegradation requires supplemental nutrients. Inorganic salts
of nitrogen and phosphorus were added to the reactor feed.
Effective conversion of the refuse to methane was obtained in
this system.
Every municipality has quantities of wastewater sludge that must
be processed. This sludge contains nitrogen and phosphorus
as well as micronutrients. The effect of the addition of small
quantities of raw sewage sludge on the process was evaluated.
The effect is shown in Figure 8. All four reactors were receiving
50 grams of refuse (dry solids) per day. Two of the reactors
were also receiving three grams (dry solids) of raw sewage sludge.
The gas production from the reactors with sewage sludge addition
was disproportionately high-.-r than the other two. At a 15-day
retention time, the addition of sewage sludge increased the gas
production 24 percent. The addition of sewage sludge at the
30-day retention time only increased the gas production by 12
percent. However, the gas production at both retention times was
significantly greater than that which would be accounted for by
the increased loading of six percent resulting from the addition
of three grams of sewage sludge. All subsequent studies included
the addition of sewage sludge to all reactors.
Immediately after the initiation of the feeding schedule, problems
were encountered with maintaining the contents of the various
reactors at a pH greater than 6.6. Initially this was not
unexpected in view of the high concentrations of short-chain
organic acids that very quickly developed. However, even after
the reactors achieved stable operating conditions and the concen-
tration of the short-chain organic acids dropped to 100 to
300 mg/£ as acetic acid, it was still necessary to add caustic
in order to hold pH above 6.6. This was accomplished initially
with the addition of sodium bicarbonate. Because of the release
of COg upon addition of NaHC03 to the low pH reactor contents,
surges in gas production were experienced after addition of
NaHCOo for pH adjustment. To eliminate this problem, NaOH was
used for the balance of the study. Enough caustic was added to
raise the pH of the particular digester to 6.7 to 6.8. During
this phase of the study no further additions of alkalinity were
made until the pH of the digester under consideration dropped
to near 6.6. Due to their higher rate of washout, the caustic
requirements were observed to be higher at the lower retention
times. The caustic requirements at various retention times were
determined in Phase 2.
Once the systems stabilized, the alkalinity values for the various
reactors remained relatively stable for the duration of the six week
40
-------
18
16
i"
"
1 10
c
o
3 8
to
ta
30 Day Retention Time
O 15 Day Retention Time
•— Refuse Only
Refuse Plus Sewage Sludge
\
\
\
\
I
5 6
Time - Weeks
10
Figure 8. Effect of Raw Sewage Sludge Addition on Refuse Digestion
-------
experimental run. As can be seen in Table 11, alkalinity increased
with an increase in the concentration of feed solids but, even
then, the observed values are considerably lower than typical
alkalinity values observed in properly operating municipal sludge
digesters. It should be noted that the alkalinities of the reactors
receiving 100 mt of raw primary sewage sludge in addition to refuse
were higher than those receiving only refuse.
Table 11. Summary of Operating Conditions at 35°C - Phase 1
Reactor
No.
1
2
3
4
5
6
7!
8*
Retention
Time- days
30
15
30
15
30
15
30
15
Volatile
Solids
Feed -
gms/day
8.76
8.76
18.25
18.25
36.50
36.50
38.70
38.70
Gas Production - STP
£/day
3.13
2.79
6.31
5.44
12.89
10.88
14.40
13.55
scf/lb V.S. Add.
5.71
5.10
5.54
4.77
5.65
4.77
5.95
5.61
Alkalinity
mg/£ as
CaC03
1275
1200
1500
1400
1700
1750
1900
2000
Reactors receiving three grams of raw sewage sludge solids
(100 m£ of sludge) per day.
The need for the addition of caustic to prevent a precipitous
drop in pH, coupled with the low alkalinities present in the
reactors, clearly demonstrates that the characteristics of the
refuse were such that adequate buffer did not develop from its
decomposition. The nitrogen content of the refuse was very low
indicating that very little protein was present. The deamination
of protein is a major source of alkalinity in anaerobic fermentation.
A summary of the results of Phase 1 is shown in Table 11. These
data were collected over a six week period during which the system
was at equilibrium. The gas production from the digestion of
refuse only averaged 4.84 and 5.63 SCF at a 15-day and 30-day
retention time, respectively. As would be expected, the longer
retention times resulted in a higher gas yield. The addition of
sewage sludge did increase the gas production indicating some
benefit associated with this material. The cause for this increase
is probably related to two factors. Sewage sludge contains a host
42
-------
of micronutrients in a form readily available to the microorganisms,
Also, the continued introduction of the wide variety of micro-
organisms present in sewage sludge may promote the development of
a more optimum culture for stabilization of the organic material.
Whatever the reason, there is an advantage in adding sewage sludge
when digesting refuse. Since this sludge is readily available
in most, if not all, communities, it would be a logical addition
in the digestion of refuse.
The composition of the gas produced during Phase 1 was measured
and these data are shown in Table 12. These values are for the
Table 12. Composition of Gas Produced - Phase 1
Reactor Number
Methane %
Carbon Dioxide %
61
39
62.
37.
5
5
58.
41.
5
5
60
40
58
42
58
42
59.
40.
5
5
60
40
dry gas. There is very little variation in the quality of the
gas produced by these reactors. Since the composition of the gas
is controlled by the pH-alkalinity relationship, these data
reflect the variation in these two parameters. The addition of
caustic for pH control can be expected to change the quality of
the gas produced by this process.
Solids Destruction - Phase 1
The solids concentrations fed the various reactors and the
measured effluent concentrations from these reactors along with
the percent volatility of the influent and effluent solids are
shown in Table 13. All values are six-week averages during which
time the units were in equilibrium. During this period, the
effluent solids exhibited relatively constant quantitative and
qualitative characteristics. Also shown in Table 13 are the
apparent reductions in volatile and total solids.
Based on the volatile solids destruction efficiencies and the
six-week average daily gas productions, the volume of gas produced
by the various reactors is shown in Table 14. The values are
expressed at standard temperature and pressure.
The values observed in Table 14 range from 8.41 to 10.92 cu ft/lb
volatile solids destroyed. With the exception of the 10.92 cu ft/lb
43
-------
Table 13. Apparent Volatile and Total Solids Reduction
Parameter
Reactor
#1 #2 #3 #4 #5 #6 #7 #8
Feed solids cone. 2.4 1.2 5.0 2.5 10.0 5.0 10.6 5.3
%
Volatility of 73 73 73 73 73 73 73 73
feed solids,%
Effluent solids 1.00 0.73 2.19 1.32 3.34 2.18 3.00 2.16
cone., %
Volatility of 62.3 64.1 61.8 67.5 72.7 74.6 75.6 74.6
effl. solids, %
Apparent volatile 64 47 63 51 67 55 71 58
solids red., %
Apparent total 58 39 56 47 67 56 72 59
solids red., %
Table 14. Apparent Reactor Gas Production Per
Unit Volatile Solids Destroyed
Production
Units #1 #2 #3 #4 #5 #6 #7 #8
£/gm 0.555 0.683 0.550 0.583 0.529 0.538 0.526 0.601
volatile
solids
destroyed
cu ft/lb 8.88 10.92 8.80 9.30 8.46 8.60 8.41 9.60
volatile
solids
destroyed
volatiles destroyed that reactor 2 exhibited, all values are well
below the theoretical gas production of 11 cu ft/lb volatile
solids destroyed. This wide variation in gas production per mass
of volatile solids destroyed is not justifiable. The refuse feed
solids introduced to all digesters were of identical chemical
composition and anaerobic fermentation, as such, should yield nearly
44
-------
identical gas productions per unit mass destroyed. A major reason
for these variations is associated with the problem of obtaining
a representative sample of the reactor effluents. A true solids
balance could not be obtained on the reactors. This is evidenced
by the fact that not all the fixed solids fed the reactors could
be accounted for in the effluents from these reactors. This point
is illustrated by the data in Table 15.
Table 15. Fixed Solids Balance
Influent Fixed Effluent Fixed Fixed Solids
Digester No. Solids - g/day Solids - g/day Lost - g/day
1
2
3
4
5
6
7
8
3.24
3.24
6.75
6.75 -
13.50
13.50
14.30
14.30
1.89
2.51
4.18
4.29
4.57
5.50
3.70
5.50
1.35
0.73
2.57
2.46
8.93
8.00
10.60
8.80
The loss of fixed solids in the digestion process is not an
indication of biological breakdown since these solids are not
significantly changed in this process. The problem is attributed
to two considerations. First, the nature of the refuse made
sampling very difficult. Second, visual observation showed a
significant amount of material entwined with the mixing mechanism.
Since one cannot account for all of the fixed solids, it is also
logical that one cannot account for all of the volatile solids
either. The apparent destruction of volatile solids which are
not actually converted to gas results in lower gas production values
on a per unit mass of volatile solids destroyed basis. This
explains the low values of gas production shown in Table 14.
Reactor 2 appeared to have the best solids balance with only an
average of 0.73 gms of fixed solids not accounted for on a daily
basis. The calculated gas production of reactor 2, at 10.92
standard cu ft/lb volatile solids destroyed is the highest among
all reactors, and closely approximates the predicted production of
11 standard cu ft/lb solids destroyed. It, therefore, may be
concluded that a gas production of 11 standard cu ft/lb of
volatile solids destroyed is a realistic value for the refuse
used in this study.
45
-------
Since the gas production should be constant and equal to about
11 cu ft/lb of volatile solids destroyed for all reactors, and
because of the difficulties encountered in obtaining a true solids
balance, the destruction of volatile solids can be calculated on
the basis of gas production. These calculated reductions are
shown in Table 16. They are more uniform and are typical of what
may be expected from sewage sludges. Reflected in these values
also are the increased volatile solids destuctions in the reactors
receiving sewage sludge in addition to refuse. Such increases
in the destruction of volatile solids are accompanied by increased
gas yield.
Table 16. Adjusted Volatile Solids Destruction
Digester No. Volatile Solids Destruction, %
1
2
3
4
5
6
7
8
52
46 ,
50 '
43
51
43
54
51
The difficulty encountered in obtaining a true solids balance
precludes the expression of gas production in terms of volatile
solids destroyed. All remaining gas production data will be
expressed in terms of volatile solids added.
Phase 2 - Mesophilic Temperature Studies
The results from the first phase of the study indicated that
conversion of the organic refuse to methane was indeed possible.
No inhibition of the digestion process was experienced with the
refuse used in this study. It was necessary to add major nutrients
as the refuse was essentially devoid of nitrogen and phosphorus.
It was observed that the addition of sewage sludge to the feed
stock significantly improved the biological breakdown of the refuse.
It was also observed that the natural buffering capacity of the
process was inadequate to maintain the pH at a satisfactory
level. It is necessary to add caustic to the reactors to maintain
the pH in excess of 6.6.
46
-------
With these observations as a background, a second series of
experiments were undertaken to establish the rates of stabilization
of the refuse at 35°C. The feed stock for the reactors was the
same as previously employed with the exception that all reactors
received 50 mi of raw sewage sludge. The organic load to all units
was held constant at 50 grams of dry refuse plus 50 m£ of sewage
sludge (1.5 grams dry solids). This produced a loading of
0.21 Ib solids per cubic foot of reactor capacity. The loadings
of the units are given in Table 17.
Table 17. Reactor Loadings - Phase 2
Retention Time Feed Solids Concentration %
Reactor No. Days TotalVolatile
1
2
3
4
5
6
7
8
4
4
6
8
10
15
20
30
1.37
1.37
2.10
2.74
3.43
5.15
6.86
10.30
0.94
0.94
1.44
1.87
2.35
3.53
4.70
7.06
As with the previous series, feeding was done daily on a batch
basis. There was initial concern about the shock effect of with-
drawing and adding one-fourth of the volume of reactors 1 and 2.
However, they were not adversely affected by this operation. The
shift to these lower retention periods was made gradually using
the active digesting units from the previous series of experiments.
The transition to these retention times was smooth and the organic
acids remained below 1000 mg/l in all units.
After equilibrium was established, the average operating conditions
for a five-week period were as given in Table 18. Frequent
alkalinity and volatile acid determinations were made during this
run. The results are shown in Appendix C, Table 35. Because of
the relative minor variation in these analyses, they were
conducted much less frequently in the remaining runs. As long
as reactor operation was normal, these analyses were conducted
simply to determine the level of the parameter in each reactor.
The pH was controlled by the addition of caustic which also
contributed to the alkalinity. The caustic requirements are
expressed as meq per 15 liters per day. There is an excellent
47
-------
Table 18. Reactor Operating Conditions - 35°C
Reactor
Number
1
2
3
4
5
6
7
8
pH
6.77
6.78
6.78
6.75
6.75
6.77
6.81
6.87
Alkalinity
mg/£ CaCOj
1400
1460
1700
1840
1900
2300
2600
3000
Volatile Acids
mg/£ Acetic
420
420
380
324
278
282
279
330
Caustic Added
meq/day
93
93
59
56
56
38
38
38
correlation between alkalinity and retention time. The long
retention times are associated with a high concentration of
solids, both suspended and dissolved. The natural buffer capacity
at the longer retention times is much greater as evidenced by
the decrease in the amount of caustic required to maintain the pH
at an acceptable operating level. This consumption of caustic may
mitigate against the use of short retention time in the practical
application of the process. It should be indicated that the
caustic consumption will be a function of the chemical character-
istic of the refuse being processed. Ammonia is one of the primary
ions in developing a natural buffer in the anaerobic process.
Therefore, any refuse, such as a typical municipal refuse, that
is very low in nitrogen may be expected to exert a significant
demand for additional buffer.
The use of actual refuse as a substrate was a cause of substantial
variation in the data collected. Figure 9 shows the variation in
daily gas production for 33 consecutive days when the reactors
were considered to be operating under equilibrium conditions. These
values are the meter readings and they have not been adjusted to
standard temperature and pressure. Reactors 4 and 8, having longer
retention times, had less variation in the daily volume of gas
generated. Some of the variations can be attributed to variation
in feed. When all three curves show similar trends, this trend
would result from the feed. Only 50 grams of dry solids (shredded
domestic refuse) were added to each reactor. In selecting the
refuse feed for daily feeding from the storage container, it was
possible to introduce variations. In a similar manner, preparation
of a batch of refuse by hydraulic separation could result in
variations in feed characteristics. This would result in
increases or decreases in gas production over a several day period.
48
-------
20
IO
I
sc
0>
o 10
*J
U
TJ
I/I
Reactor 8
Reactor 2
I
10
15 20
Time - Days
25
30
35
Figure 9. Variation in Gas Production Under "Equilibrium" Conditions - 35°C
-------
Variations between reactors can also result from the selection
and weighing of the individual 50 grams for each reactor.
Additional variation may be expected from reactor instability,
in particular the units operating at the short retention times,
Other problems such as incorrect reading of the gas meters can
add to the variation. The gas meters recorded cumulative gas
productions, so a high reading followed by a low reading, or the
reverse, may be due to meter reading error.
Upon completion of the 35°C test conditions, the temperature of
the system was gradually increased to 40°C. The equilibrium
operating conditions for a five-week period at 40°C are given
in Table 19.
Table 19. Reactor Operating Conditions - 40°C
Reactor
Number
1
2
3
4
5
6
7
8
Retention
Time
Days
4
4
6
8
10
15
20
30
PH
6.77
6.77
6.78
6.78
6.78
6.80
6.83
6.87
Alkalinity
mg/£ CaC03
1244
1319
1595
1780
1982
2356
2554
3087
Volatile
Acids
mg/£ Acetic
278
240
222
298
405
642
733
866
Caustic
Added
meq/day
80.3
80.3
70.3
52.0
49.0
36.0
23.0
19.0
The pH and caustic additions were recorded daily. However, the
frequency at which the alkalinity and volatile acid analyses were
conducted was reduced considerably from the 35°C study. These
individual results are given in Appendix C, Table 36. There was
no significant variation in these parameters during the steady
state condition. Therefore, only a few analyses were conducted
for the purpose of fixing the level of these parameters under
these specific conditions.
The pH was maintained at the same level as for 35°C. However, the
caustic required to maintain this pH level was significantly less
than that required at the lower temperature. The alkalinity was
slightly lower in the 40°C study, especially at the shorter
retention times. The longer retention times show essentially
50
-------
the same level of alkalinity at both temperatures. However, the
volatile acids in reactors 5 through 8 were significantly higher
than reactors 1 through 4. These organic acid salts do add to
the alkalinity. Therefore, correcting the alkalinity for that
contributed by the organic acids would give the actual alkalinity
due to the bicarbonate ion, which would be substantially less than
at the 35°C temperature.
The reduction in caustic requirements is a result of the temperature
effect on the carbon dioxide-bicarbonate relationship as previously
discussed. More carbon dioxide is being removed from the system
in the gas phase rather than in the liquid phase as bicarbonate
ion.
The higher level of organic acids measured in reactors 5 through 8
can only be related to the differences in mixing between these
two sets of reactors. The baffles on the sides of these reactors
appeared to have an adverse effect on mixing when the concentration
of the solids in the reactors was high. Visual observations showed
areas in the reactor that were very poorly mixed. This poor
mixing is apparently responsible for the higher level of organic
acids found in reactors 5 through 8.
This problem was not experienced in all cases. The volatile
acid data in Table 18 do not show any significant variation between
the two sets of reactors. Therefore, this mixing problem is not
exclusively related to the reactor characteristics. It was
complicated by the buildup of material on the mixing mechanism.
The plastics, cloth and other fibrous materials would entwine
around the mixer. It was necessary to remove this material
periodically. However, the reactors haying a concentrated feed
slurry experienced a rapid buildup of this material. Consequently,
the problem was most pronounced in reactors 5 through 8.
This observation does suggest that mixing conditions must be
given careful consideration in a large scale system. The degree
of mixing required is based upon maintenance of homogeneity within
the reactors. The rate limiting step appears to be the hydrolysis
of the fibers. Therefore, mass transfer considerations do not
enter into the mixing considerations.
The variation in gas production for 34 consecutive days is shown
in Figure 10. Reactor 2 shows variations similar to that found
at 35°C. However, the reactors with the longer retention times
and higher solids content showed somewhat greater variations,
particularly near the end of the period. Some of these variations
may be due to the feed variations. Mixing in reactor 4 and 8 may
be a cause for the more extreme variations. These variations are
not unreasonable when one considers the characteristics of the
substrate employed.
51
-------
CJI
20
01
10
o
•o
X >
Reactor 8
\
Reactor 2
s \
1
1
15 20
Time - Days
25
30
35
Figure 10. Variation in Gas Production Under "Equilibrium" Conditions - 40°C
-------
When the temperature was increased to 45°C, the alkalinity and
caustic requirements decreased slightly. These data are shown in
Table 20. The average for the alkalinity and volatile acids
represent only four analysis. Because of the inhibition expected
and observed at this temperature, only limited data were collected.
The volatile acids remained at a low value at this temperature
even though the gas production decreased substantially. The
gas production will be discussed in detail later. The acids
produced from the breakdown of the organic material were
efficiently fermented. It appears that the reduced gas production
resulted from a decrease in the breakdown of the complex organic
materials.
Table 20. Reactor Operating Conditions - 45°C
Reactor
Number
1
2
3
4
5
6
7
8
Retention
Time
Days
4
4
6
8
10
15
20
30
PH
6.73
6.67
6.74
6.78
6.78
6.81
6.80
6.82
Alkalinity
mg/l CaC03
1269
1214
1312
1612
1752
2002
2335
2410
Volatile
Acids
mg/l Acetic
143
531
116
81
171
282
220
221
Caustic
Added
meq/day
77.3
78.6
57.0
50.7
42.1
35.7
27.0
20.0
The variation in gas production is shown in Figure 11. As with
the previous temperature studies, considerable day-to-day
variation was experienced. Similar trends were exhibited by all
three reactors shown, suggesting that much of the variation is due
to the variation in feed characteristics. There are some variations
between individual reactors which can be attributed to the factors
previously mentioned.
Gas Production and Composition - Phase 2
The gas production and composition is shown in Table 21. There
is a very rapid increase in the quantity of gas produced for a
given loading with increasing retention time until about 15 days.
At this retention time gas production continues to increase, but
at a very slow rate. While a 30-day retention time will yield
more gas, it may be of questionable value to operate at these
53
-------
en
20
15
0)
§ 10
•r-
U
1
a.
M
-------
Table 21. Gas Composition and Production - 35°C
Reactor Retention Time Methane Carbon Dioxide scf Gas
Number Days % % Ibs V.S. Add.
1
2
3
4
5
6
7
8
4
4
6
8
10
15
20
30
69.7
69.8
64.3
58.6
57.2
53.8
53.4
53.8
30.3
30.2
35.7
41.4
42.8
46.2
46.6
46.2
2.02
2.02
2.91
3.36
3.84
4.23
4.28
4.39
longer periods due to the substantial increase in capital costs
for larger reactors.
The composition of the gas was not particularly encouraging at
the longer retention times as shown in Table 21. In the units
receiving large amounts of caustic, a substantial amount of the
carbon dioxide produced reacted with the caustic to produce
bicarbonate alkalinity.
Since gas production was low in these units, enough carbon
dioxide could be absorbed in the liquid to significantly lower
the carbon dioxide content of the gas. At the high gas production
rates, this effect would not be significant because of the larger
quantity of gas produced as well as the lesser amount of caustic
added.
The effect of temperature on gas production is shown in Figure 12.
A significant increase in gas production results from increasing
the temperature from 35°C to 40°C. This effect is observed at
all retention times. A continued increase in temperature to 45°C
causes a reduction in gas production. The optimum temperature
for the mesophilic micro-organisms appears to be near 40°C. The
temperature was gradually increased when the 40°C study was
completed. A temperature increase of 0.5°C to 1.0°C every week
was used to determine the maximum mesophilic temperature. Gas
production increased until a temperature of 43°C was exceeded.
The gas production significantly decreased as the temperature
was increased. Therefore, it was concluded that 40°C to 42°C was
the optimum temperature for mesophilic fermentation of organic
refuse.
55
-------
Q> C
•O 3
C~ 4
o
V)
U o
3
•o
e
o.
I/I
Q f.
"" ""7s*?3
5 10 15 20 25
Reactor Retention Time - Days
Figure 12. Gas Production at Mesophilic Temperatures
30
35
56
-------
Phase 3 - Thermophilic Temperature Studies
This phase extended the temperature into the thermophilic range.
After completion of the mesophilic temperature studies, the
temperature was increased to 50°C. All temperature changes were
made gradually in order to avoid temperature shock. The reactors
had been operating for about two months at 45°C. Even with the
gradual increase in temperature, there was a lag in the initiation
of thermophilic digestion. After a four week period the gas
production was stabilized.
After the acclimation period, data were collected on the operation
of these reactors at 50°C. The average conditions during this
six week period are shown in Table 22. The individual analyses
from which the average alkalinity and volatile acids were
calculated are given in Appendix C, Table 37. As with the
previous studies, the pH was controlled by the addition of caustic.
Increasing the temperature to 50°C reduced the alkalinity and
caustic requirements to levels below those found in the mesophilic
range (see Tables 18, 19 and 20). The reactors operated
effectively at this temperature without any problems.
Table 22. Reactor Operating Conditions - 50°C
Reactor
Number
1
2
3
4
5
6
7
8
Retention
Time
Days
3
4
6
8
10
15
20
30
Alkalinity
pH i
6.75
6.78
6.83
6.81
6.82
6.85
6.86
6.89
mg/£ CaC03
1156
1154
1251
1394
1544
1778
2082
2343
Volatile
Acids
mg/£ Acetic
292
210
187
257
218
325
302
402
Caustic
Added
meq/day
92
87
54.7
52.5
45.0
22.5
15.0
12.5
The gas production was substantially increased, but was extremely
variable. Figure 13 shows 35 consecutive days of gas production
for these reactors. During the initial period of equilibrium,
the gas production increased substantially with retention time.
All reactors exhibited variations that can be attributed to
variation in feed characteristics. Reactor 2 was considerably more
erratic during the first 15 days. Toward the end of the period,
the gas production from reactor 2 was essentially the same as
57
-------
Q
£
tt)
2(
1!
CJl
00
u
-o
o
310
yi
/ \Reactor 8
/v
*
Reactor 2
I
I
I
I
I
1
10
25
30
15 20
Time - Days
Figure 13. Variation In Gas Production Under "Equilibrium" Conditions - 50°C
35
-------
the gas production from reactor 4. The difference among all
three reactors was considerably reduced. It may be that additional
time was required to acclimate the shorter retention time reactors.
It does appear that the longer retention times yield little
additional gas production at the higher temperature.
Upon completion of the data collection at 50°C, the temperature
was gradually increased to 55°C. The average operating conditions
for this eight week period are shown in Table 23. The individual
analyses used to calculate the average alkalinity and volatile
acids are given in Appendix C, Table 38. The average alkalinity
Table 23. Reactor Operating Conditions - 55°C
Reactor
Number
1
2
3
4
5
6
7
8
Retention
Time
Days
3
4
6
8
10
15
20
30
PH
6.85
6.85
6.85
6.85
6.85
6.85
6.90
7.00
Alkalinity
mg/l CaC03
1050
1300
1395
1515
1435
1545
1490
1885
Volatile
Acids
mg/£ Acetic
112
58
64
79
208
215
298
332
Caustic
Added
meq/day
63.7
58.5
37.4
30.0
23.7
10.7
3.7
1.1
was reduced by the temperature increase. The caustic required
to maintain the pH was substantially reduced, especially at the
longer retention times. Reactors 7 and 8 operated at a pH of
6.9 and 7.0, respectively, and required a limited quantity of
caustic to maintain this pH. It is probable that these reactors
could have operated successfully without the addition of any
caustic.
The possible effect of mixing can also be seen from the level of
volatile acids found in reactors 5 through 8 (see Table 23). These
reactors were not as well mixed as the other four. The organic
acid concentration was three to four times greater than that in
reactors 1 through 4. These reactors with the shorter retention
time could be expected to have higher organic acid levels because
of the shorter time available for the biological stabilization
to take place. This higher organic acid level would not signifi-
cantly decrease the gas production, but it would suggest that
some factor may be limiting the biological process. Since the
59
-------
degree of mixing and feed slurry concentration were the only
differences between these two sets of units, it must be concluded
that these differences were affecting the gas production.
The variation in daily gas production is shown in Figure 14. The
variation in feed characteristics appears to be a major cause in
the variability of the gas production since all three reactors
show essentially the same variation. The gas production from
all three reactors was essentially the same. Therefore, it
appears that there is little increase in gas production with
increasing retention times at the higher temperature.
The thermophilic studies were completed with the 60°C temperature
run. This cut off was selected because it is near the upper
limit of the thermophilic temperature range. Studies of mixed
cultures at temperatures of 65°C (149°F) would expect thermal
inactivation of many organisms. The previous studies of anaerobic
digestion show a complete inactivation at 65°C (Golueke, 1958).
The average operating conditions for this seven week study are
shown in Table 24. The data from which the average alkalinity was
computed was based on only two analyses. Only one analysis for
volatile acids was conducted, indicating that they were within
the normal range. Since no operating problems were encountered,
additional determinations were not made because of the limited
project staff available during this period of the study.
Table 24. Reactor Operating Conditions - 60°C
Reactor
Number
1
2
3
4
5
6
7
8
Retention
Time
Days
3
4
6
8
10
15
20
30
PH
6.8
6.8
6.8
6.9
6.9
6.8
6.8
6.8
Alkalinity
mg/£ CaC03
1030
1140
1530
1600
1920
1900
1900
2030
Caustic
Added
meq/day
9.14
7.85
6.43
5.57
3.00
1.86
0.86
0.29
The alkalinity again decreased with the temperature increase. The
caustic requirements were substantially reduced. It would appear
that a very limited amount, if any, caustic would be required for
60
-------
10
25
30
15 20
Time - Days
Figure 14. Variation 1n Gas Production Under "Equilibrium" Conditions - 55°C
-------
pH control at the 60°C temperature. The pH levels were above
the minimum recommended value of 6.6 in all reactors.
The daily gas production is given in Figure 15. As with the other
runs, considerable variation was experienced in the gas production.
During the first 18 days, all these reactors were producing
essentially the same quantity of gas. There was considerable
variation between the reactors during the balance of the time
period. There was no obvious reason for this variation. Part
of it may be attributed to variation in the feed.
Gas Production and Composition - Phase 3
The gas production and composition for 60°C is given in Table 25.
While the gas production greatly exceeds that of the mesophilic
temperature range, the methane content is slightly lower, ranging
from 55 to 62 percent as compared to a 53 to 70 percent range at
35°C. With lower bicarbonate concentrations in the liquid phase,
more carbon dioxide will appear in the gas phase. The methane
content of the gas will decrease with increasing temperatures.
Table 25. Gas Composition and Production - 60°C
Reactor
Number
1
2
3
4
5
6
7
8
Retention Time
Days
3
4
6
8
10
15
20
30
Methane
%
62
58
57
57
57
56
55
55
Carbon Dioxide
%
38
42
43
43
43
44
45
45
scf Gas
Ibs V.S. Add
1.96
6.17
6.75
6.16
6.64
6.61
6.90
7.20
The gas composition is primarily controlled by the alkalinity,
temperature and pH. The data in Table 26 show the relationship
between pH, alkalinity and gas composition at 60°C. The data
represent the conditions on the specific day on which this set of
analysis was conducted. The data from reactors 1 and 2, and 5 and
6 show that an increase in pH of 0.15 to 0.2 units will increase
the methane in the gas by about 1.5 percent. In a similar manner,
data from reactors 3 and 7 show that a substantial increase in
alkalinity of 1000 mg/£ as CaC03 may increase the carbon dioxide
62
-------
40
CJ
1 To ~^R 20 25 30 35"
Time - Days
Figure 15. Variation In Gas Production Under "Equilibrium11 Conditions - 60°C
-------
Table 26. Effect of pH and Alkalinity on Gas Composition
Reactor
Number
1
2
3
4
5
6
7
8
PH
6.60
6.80
6.90
6.80
6.85
7.00
6.90
7.05
Alkalinity
mg/l CaCOj
1105
1125
1572
1600
1920
1895
2600
2030
Methane
%
53.5
55.0
51.9
50.2
52.3
53.5
49.1
53.7
Carbon Dioxide
%
46.5
45.0
48.1
49.8
47.7
46.5
50.9
46.3
in the gas by only about three percent. The gas composition is
much more sensitive to pH shifts than to changes in the alkalinity.
This should be expected as a change in pH represents a change in
the log of the hydrogen ion concentration.
Figure 16 shows the gas production of 50°C, 55°C and 60°C. The
data at the 50°C temperature yielded a reasonably good curve fit.
However, the data at 55°C have more variation. The gas data for
60°C provides a somewhat better fitting curve than the 55°C data.
However, variations such as evidenced by the 60° data were
inherent in this study because of the small reactors and the small
daily substrate addition.
Mixing may have been a factor in the variation in the gas production.
Reactors 1 through 4, operating with retention times of three to
eight days respectively, had significantly better mixing. This
was a result of a more dilute feed as well as a different mixing
system. The first four data points at 55°C in Figure 16 fit the
curve well as shown by the long dashed line. The solid line is
an attempt to obtain the curve of best fit for all 55°C data
points. These data do show a significant increase in gas production
at 55°C, particularly at the shorter retention times. Increasing
the temperature to 60°C provides an additional increase in the
gas production. Again, the increase is larger at the shorter
retention times. Retention times substantially less than 30 days
will yield essentially all of the gas that can be generated by
the system.
The improved gas production at 55°C and 60°C does not aggree with
data reported by Golueke (1958). These data are shown in Figure 17.
The 55°C and 60°C temperature produced less gas than the 50°C
64
-------
•o
0)
•o
o
' 4
c
o
•r—
O
•i
2
CL.
vt
0
10
15
20
25
30
Reactor Retention Time - Days
Figure 16. Gas Production at Thermophilic Temperatures
35
65
-------
CTi
30
40
50
Temperature - °C
60
70
Figure 17. Effect of Temperature on Gas Production (After Golueke, 1958)
-------
temperature. This may have resulted from too rapid an increase
in the operating temperatures. In the Illinois study, temperature
increases were accomplished at the rate of 1.5°C to 2.0°C per week.
This provided adequate time for the microorganisms to acclimate
to the increased temperature.
Cellulose Decomposition
The importance of cellulose destruction in the conversion of
refuse to methane has previously been discussed. A series of
experiments were devised to determine the effect of temperature
and pH on this initial rate of cellulose destruction. These
studies were conducted in flasks operated anaerobically on a
batch basis. The initial slurry concentration was five to six
percent. This slurry consisted of raw refuse plus tap water. A
starter slurry was added at an amount equal to 25 percent of the
total flask contents. This starter solution provides the necessary
cell count to insure cellulose hydrolysis. It was obtained from
the reactors being fed the domestic refuse. The temperature of
the tests were matched with the reactor temperatures so that the
starter was acclimated to both the substrate and temperature.
Before each temperature run, sufficient quantities of refuse,
water and starter were mixed to insure the same initial conditions
for each pH level tested at that temperature. This material was
stored at 4°C for use in subsequent tests at various pH levels.
The Anthrone technique was used to determine the cellulose
concentration. The pH was controlled by buffering with sodium
bicarbonate and acid, or the addition of sodium hydroxide at the
higher pH levels. The pH was constantly monitored with combination
electrodes installed in the flasks. Necessary base was added as
frequently as required to maintain the pH within ±.1 unit of the
desired pH level.
The results of this study are shown in Figure 18. TH" maximum
rate of cellulose utilization was obtained "rom tht ie of
the plot of cellulose remaining versus time. The time interval
between measurements of the cellulose content in the flasks was
24 hours during the first days of the test and 48 hours there-
after. The effect of temperature may not be as accurate as
desired due to variability in the starter solution for different
temperature runs. However, the pH effect is accurate s-'r.e all
pH levels at a given temperature were initiated from the same
substrate and starter. As can be seen fra,. this figure, the
cellulose utilization rate was markedly improved at the elevated
pH levels. A pH of 7.5 was the maximum tested. Operating a
continuous system in excess of 7.5 would ha»e a significant caustic
demand which would markedly increase the operating costs.
Cellulose destruction was monitored in the continuous reactors.
The pH of the reactors was maintained at 6.7 ± 0.1. The data
67
-------
2000
1500 —
S
Si 1000
0)
o
50( —
65
Figure 18. The Effect of Temperature and pH on the Maximum Rate of
Cellulose Utilization
68
-------
shown in Figure 19 were obtained at the 35°C temperature. Even
at the lowest retention time, significant hydrolysis of the
cellulose occurred. At these short retention times, only traces
of noncellulose carbohydrates could be detected. The conversion
of the hydrolysis products of cellulose to methane and carbon
dioxide was rapid and complete with only small quantities of
organic acids remaining.
The 30-day retention time produced a 74.4 percent reduction in
cellulose. Again the conversion of the products of cellulose
hydrolysis to gas was complete. The rate limiting step in the
digestion of refuse appears to be the hydrolysis of the complex
organic solids. Data from studies on the digestion of sewage
sludge show a significant increase in the organic acid concentra-
tion at a retention time of four days as compared to longer
retention periods. However, the organic acids were essentially
constant in all digestion units. This would suggest that the
organic acids were metabolized as rapidly as they were produced
at all retention times used in these experiments.
The effect of increased temperatures on cellulose utilization was
simply to increase the percent utilization at the lower retention
times. At 40°C, the cellulose utilization was 60 percent at the
four day retention time, increasing to 70 percent at the 30-day
retention time. From these data it would appear that the cellulose
remaining in the system is resistant to biodegradation. The
available cellulose is utilized rapidly in the thermophilic
temperature range.
Dewatering of Reactor Residue
A significant step in the processing of organic refuse for energy
recovery will be the dewatering of the digested sludge. This
residue must be adequately processed for disposal. The filter-
ability of the digested solids was examined using a filter test
leaf technique. A small filter section was immersed in the
effluent from the reactors with a vacuum applied for a 30 second
form time. The test leaf was withdrawn from the slurry and
allowed to dry with the vacuum continuing for 1.0 minute. The
filter test cloths were 1/1 plain weave polyethelene with different
size pores. A form time of 30 seconds was selected because that
would be the minimum practical time. However, the lab tests
showed that the cake was completely formed in about five seconds.
Since no chemical conditioning was employed in these tests, it
may be possible to significantly increase the filter yield by
adding chemicals or polymers.
The filter yield at the lower feed slurry concentrations was
quite low but was markedly improved at the higher feed slurry
69
-------
100
£ 80
-------
concentrations. These data are shown in Table 27. The digested
solids appeared to thicken readily. However, with the small
volumes of slurry available in the laboratory no meaningful
thickening data could be obtained. It does appear that signifi-
cant improvements in the filter yield can be obtained with thicker
slurries. It should be mentioned that the solids concentration
in the effluent from the reactor receiving a feed containing ten
percent solids varied between 3.0 and 3.5 percent solids. There-
fore, one can expect a rather dilute slurry discharged from the
reactors.
Table 27. Filterability of Reactor Residue
Feed Slurry Solids Concentration - %
Test Leaf 3.5 6.8 9.3
Filter Yield - Ibs/ft2-hr
1
2
3
4.5
5.0
5.0
13.2
13.5
12.5
19.0
19.2
17.5
Moisture Content - %
1 75.0 73.0 75.0
2 75.0 77.0 75.5
3 77.0 72.0 76.0
The thickening and dewatering characteristics of this digested
material will have to be evaluated on a larger scale. The use
of chemicals and/or polymers to improve the dewatering character-
istics should also be examined. It does appear from these
limited tests that this material has good dewatering potential.
Characteristics of the Separated Inorganic Component
The material removed by the hydraulic separation process consisted
primarily of glass, sand and broken ceramics. It also contained
some aluminum and ferrous metals. A disposal method for this
material must be provided. Particle size was highly variable
ranging from particles passing through a 60 mesh sieve to
particles an inch or so in size. The hydraulic separation was
effective in separating only inert material. The material
contained less than ten percent volatile solids.
71
-------
The bulk density of this material was found to be 2080 Ib/cy.
The material, as separated from the refuse, was oven dried and no
attempt was made to compact the material except for that occurring
through normal handling. The weight of a known volume was taken
to determine the density. The same material was found to have
an average specific gravity of 2.35. As previously mentioned,
glass, sand and grit accounted for a major portion of the separated
inorganics. A refuse having more metals may generate an inorganic
fraction somewhat different than that produced from this refuse.
Digestion of Hydropulped Refuse
The refuse used in this study was shredded to a nominal particle
size of one inch. The rate limiting step in conversion of the
organic refuse to methane is the hydrolysis of the fibers. This
rate is surface area dependent. The surface area in turn can be
related to the particle size. In order to determine if hydropulped
refuse would improve the conversion rate, a quantity of hydro-
pulped refuse was obtained from the Black Clawson demonstration
project at Franklin, Ohio. The separated material was obtained
from this facility in a dry form.
Visual examination of the hydropulped material indicated that
fibers had been separated initially. However, in the drying
process, the fibers tended to form mats. Therefore, the apparent
particle size was not significantly smaller than the refuse being
used in the study. Upon addition to the reactor, the fibers
may have separated readily. An analysis of this material indi-
cated that much of the inorganic material had been removed from
the refuse. The material was 87.7 percent volatile.
The hydropulped refuse was used as the substrate for the reactors
for about two weeks at the end of the study. The temperature was
60°C. The reactors did not show any noticeable difference in
the gas production or operating conditions. The gas production
exhibited the usual variation, but the average for the period
was the same as when the substrate was the shredded refuse. The
long retention times and mixing in the reactor may tend to negate
any benefit of the hydropulping. In the reactor, fiber separation
is very complete. It would appear that, in fact, the hydropulping
operation is accomplished in the reactor. Therefore, increased
rates of conversion should not be expected with the dry hydro-
pulped feed. This substrate did not contain the organic material
that may have been leached from the refuse during hydropulping.
Therefore, a direct feed of the hydropulped slurry to the reactor
may yield a different response than the dried feed. This would
be a result of the presence of the leached organics rather than
a smaller fiber.
72
-------
PROCESS EVALUATION
Based on the results of this research, an analysis of the process
potential can be made. This process evaluation has been divided
into two sections, materials balance and production costs for
the product gas. The material balance relates to the impact this
process would have on the solid waste disposal problem, both
refuse and sewage sludge disposal. The production cost is
indicative of the potential economic benefits that can be obtained
from this system by producing a usable fuel gas.
Materials Balance for Process
A materials balance was performed on this process using a refuse
of a composition shown in Table 28. Figure 20 shows the tons and
volume of material flow through the process. This was developed
in the following manner. Assuming a 100 ton/day raw refuse
received, 78 tons of volatile solids plus 22 tons of inerts are
Table 28. Composition of a Typical Domestic Refuse
% of Total by Weight
Component (Wet Weight)(Dry Weight)
Paper
Leaves
Wood
Synthetics
Cloths
Garbage
Combustibles
Glass
Metal
Ashes, stone, etc.
Total moisture content
48.0
9.0
2.0
2.0
1.0
16.0
78.0
6.0
8.0
8.0
—
35.0
5.0
1.5
2.0
0.5
8.0
52.0
6.0
8.0
6.0
28.0
fed to the size reduction. The volatile solids contain 26 tons
of water plus 52 tons of dry solids. An additional two tons
of moisture is associated with the inerts. Assuming a density
of 300 pounds per cubic yard as dumped from the collection
vehicle, the volume of the refuse would be 667 cubic yards.
The only change resulting from the size reduction would be to
perhaps increase the density to 500 pounds per cubic yard with
a volume of 400 cubic yards passing to the hydraulic separation
unit.
73
-------
Refuse - As
Rece I ved
100 tons
Compos 1 1 1 on - tons
Glass - 6.0
Metal - 8.0
Inerts - 8.Q
Volatile - 78
Size
Reduction
(Reduce size
for separation
and improved
biodegradatlon)
Hydraulic
Separation
(Moisture added.
% moisture
increased to
70%)
So"ds 100 tons 100 tons^
667 cv *" W)0 cy j
(Moisture adds D<>/ Cy /
26 tons to vol.
sol ids weight
and 2 tons to
inert weight)
Assume
Anae rob I c
Fermentation
CM. production -
427,000 cu ft
Vol. sol. dest.
= 37.8 T
Solids
Dewa taring
(Reduce water
from 92% In
reactor to 70%
In cake)
180 tons " 238.7 tons_
225 cy 28i| cy
•1
90% of
glass, metal
and Inerts
removed
,
Sludge add.
Dry solids - 2.9 T
Water - 93.6 T
Volume - 115 cy
Watc
175
Solids
Transport
(Cake hauled to
disposal site)
Sanitary
Landfill
(Conventional
sani tary
landfill for
res 1 due)
63.7 tons 63.7 tons ,_
75.8
,r
tons
cy /y-o ey
Solids from
hydraulic
separation
25.7 tons
17.3 cy
Water -x^^ Dry solids - 18
105.7 tons tons,@ 30%
moisture - 25.7
tons
17-3
cy
Figure 20. Mass-Volume Balance for Conversion of Refuse to Methane
-------
Assuming the separation of the inerts is 90 percent efficient,
then 18 tons of dry solids would be removed from the process stream
at this point. The total weight would be dependent upon the amount
of water retained with the solids. Laboratory measurements of the
bulk density of this residue obtained from Cincinnati yielded a
value of 2080 pounds per cubic yard. The specific gravity of the
individual particles averaged 2.35. Because of the void space in
the granular material, the retained water would probably not
increase the volume. Therefore, the volume of this residue would
be 17.3 cubic yards per 100 tons of refuse processed.
If the refuse discharged from the separation process has a
moisture content of 70 percent, the water required in this step
can be determined. A total of 28 tons is added with the original
refuse. The 18 tons of inerts contain 7.7 tons of water at 30
percent moisture content. The 54 tons of solids discharged to
the reactor will require 126 tons of water. To satisfy the water
requirement, 105.7 tons of water would be added at this point.
The tonage of wet refuse added to the reactor would be 180 tons
per 100 tons of raw refuse processed. Assuming a specific gravity
of 1.0, the volume of this feed would be 225 cubic yards. In
addition to the refuse, sludge from the water pollution control
plant would be added at this point. The volume of sludge would
be 115 cubic yards consisting of 2.9 tons of dry solids plus
93.6 tons of water. The following calculations were used to
develop the solids load to the reactor.
1. Population producing 100 tons of refuse/day at 7 Ib/ca-day
t2ni x 200p_l
2. Primary sludge production— 60 percent efficiency in
primary sedimentation
0.22 l x 28,500 ca x 0.6 = 3780 of total solids
100 t2ni x 200p_lbs x l_cady_ . 28j500 population
Secondary sludge production— 30 percent BOD reduction in
primary
0.2 __ x (1 - 0.3) = 0.14 Ib -— to secondary
ca-day ca-day treatment
Assume 0.5 Ib of suspended solids produced per Ib
BOD5 in an activated sludge process.
Pounds of suspended solids from secondary
Ib BOD
C lh r r
x °'5 -
75
-------
4. Total dry solids generated at water pollution control
plant
3780 + 2000 = 5780 Ibs/day at three percent solids
Ibs wet sludge = 5780/0.03 = 193,000 Ibs
5. Solids loading to the reactor
Dry Solids Water
Refuse - 54 tons Refuse - 126 tons
Sludge - 2.9 tons Sludge - 93.6 tons
Total 56.9 tons Total 219.6 tons
6. Solids content of reactor feed
* So11ds ' 69 ' 20*
At 80 percent moisture content, mixing of the reactor may require
a significant amount of energy. However, the destruction of the
volatile solids will reduce the solids content of the reactor
substantially.
7. Volatile solids to reactor
Refuse - 52 tons
Sludge - 2.9 tons at 70% volatile solids - 2 tons
54 tons
From the laboratory data, gas production was 4.4 scf per pound of
volatile solids added at a 30-day retention time and at 35°C. This
would correspond to a volatile solids reduction of about 40
percent. Operation at 60°C and 30-day retention time yielded a
gas production of 7.2 scf per pound of volatile solids added with
a corresponding solids reduction of 70 percent. Therefore, when
operating at 60°C, 37.8 tons of organic solids would be converted
to carbon dioxide and methane. Gas analysis showed this gas to
contain 55 percent methane.
8. Methane production from volatile solids destruction
at 60°C
2000 x 7.2 x 0.55 = 427,000 scf
The conversion of these solids to gas reduces the solids content
in the reactor. Any mixing considerations must be based on the
solids concentration in the reactor rather than on the concentration
of solids in the feed slurry.
76
-------
9. Solids concentration in the reactor - 30 day retention
time
Tons of Solids Remaining Per Day
Volatile solids = 54 - 37.8 = 16.2 tons
Fixed solids =56.9-54 = 2.9 tons
Total = 19.1 tons/day
Tons of water added per day = 219.6
Solids content in digester = V. x 10° = 8
The effluent from the digester consists of 219.6 tons per day of
water plus 19.1 tons per day of solids or a total of 238.7 tons
per day at a moisture content of eight percent. The specific
gravity of this slurry would be approximately 1.0. Therefore, the
volume of slurry to dewater would be 284 cubic yards per day.
Prior to landfilling, this slurry must be dewatered. Laboratory
filter test-leaf studies suggest that a moisture content of 70
percent is achieved without the use of chemicals to condition
the sludge. At 70 percent moisture, the 19.1 tons of dry solids
would produce a cake weighing 63.7 tons per day. The water,
175 tons, would have to be processed for disposal, discharged to
the water pollution control plant, or up to 105.7 tons could be
recycled to the process.
Because the specific gravity of the solids in the dewatered slurry
would be about 1.1 to 1.3, the specific gravity of the dewatered
material would be approximately 1.0. Therefore, the volume of this
material for landfilling would be 75.8 cubic yards. However, a
reduction of the moisture content to 60 percent would leave a
final volume for disposal of only 56 cubic yards having a weight
of 47.7 tons. Adding to this the 17.3 cubic yards of inert residue
from the hydraulic separation step, the volume for final disposal
would be between 74 and 81 cubic yards per 100 tons of original
'refuse. This contrasts to the perhaps 667 cubic yards of loose
refuse without any processing. The degree of compaction that can
be achieved by landfilling the process residue is not known.
Assuming a 20 percent reduction in volume due to compaction, the
volume of the compacted process residue would be between 59 and
65 cubic yards per 100 tons of refuse processed. This can be
compared to the volume required by the unprocessed refuse.
Assuming compaction of the refuse produces a density of 1000
pounds per cubic yard, the 100 tons of refuse plus digester
residue from 2.9 tons of sewage solids would require 206 cubic
yards of fill space. Therefore, only 31.6 percent of the landfill
volume is required after processing.
77
-------
Unit Processes - Shredding
Size reduction is a well known process, having been applied
extensively in many industrial operations. In recent years, some
of these units have been modified for use in reducing the particle
size of the refuse prior to processing by various means.
Drobny at aJL. (1971) evaluated the cost associated with the
processing of municipal refuse. The power requirements for
several installations are shown in Figure 21. A probable power
requirement for shredding of municipal refuse is approximately
10 HP-hr/ton. The power requirement is a function of particle
size. The following product size factors, developed from
experimental data, show the increased power required for smaller
size particles. The size of product required will shift the
power requirements from near 10 HP-hr/ton for a six-inch size
particle to approximately 25 HP-hr/ton for a one-inch size
particle.
Particle Size Product Size Factors
6 inch 1.00
4 inch 1.39
2 inch 1.64
1 inch 2.38
Actual cost data for a Gondard Mill pilot operation is shown in
Table 29 (Porter 1970). This shredding process was used in
conjunction with the Madison, Wisconsin landfill project. The
author states that these costs data are high because certain
segments of the plant were over designed and the refuse feed
and product removal equipment were not sized to use the mill at
capacity. Additional labor was required because of inadequacies
in the design. Also, the mill was only operated for 5.33 hours
per day rather than seven hours at a capacity of nine tons per
hour. Adjustment of the costs for a mill operating for a 7-hour
shift per day at capacity reduced the cost to $5.05 per ton for
the 5 inch size particle. The major cost reduction was in
labor, being reduced from $3.46 per ton to $1.89 per ton.
Incremental cost decreases occurred in all categories except
hammer wear and mill maintenance. Major cost factors were
amortization and labor for this system.
Additional costs reduction in the pilot plant cost can be
expected. Installation of a Tollemache Mill at this pilot plant
produced a shredding cost of $4.92 per ton. This cost is based
upon operation of the mill only 5.3 hours per day rather than
seven hours.
78
-------
12
O Hammermills
A Shredders
10
o
o
I
J-
QJ
§
Q.
1
I
10 20 30
Capacity - Tons/Hour
40
50
Figure 21. Power Requirements for Hammer-mills and
Shredders (Drobney & aJL.% 1971)
79
-------
Table 29. Shredding Cost for 5 Inch Grate Size (Gondard Mill)
(Period of Operation—June 1968 - May 1969)
Cost Item $/Ton % Total Cost
Labor
Amortization
Power
Utilities
Hammer wear
Mill maintenance
General supplies
Other
Total
3.46
2.80
0.30
0.32
0.15
0.08
0.17
0.19
TA7
45.2
37.4
4.0
4.3
2.0
1.1
2.3
2.5
100.0
A prediction of cost for one and/or two shift operation and
different number of mills is summerized in Table 30 (Ham at
a£., 1971). In this calculation, the size reduction equip-
ment is hammer-mills with 15 tons per hour capacity each. They
are to be operated seven hours per shift and 245 days per year.
Where one man can monitor two mills, the cost ranged from $2.42
per ton for a minimum number of mills and one shift work to
$1.21 per ton for a plant working two shifts, with capacity of
about 200,000 tons per year.
In a similar analysis, Porter (1970) projected the cost
associated with the Gondard Mill based on the number of mills
and one or two shift operation. The predicted cost is presented
in Table 31. The cost in this calculation is $1.96 to $3.95
per ton. These values are highly dependent on operational
conditions and characteristics of the system. Applying these
analysis to another system requires careful study. However,
improvement in the system is expected in the future and the
cost may be reduced. Also operating the system at high
capacity and making full time use of the mills provides
less cost. A value of $1.5 to $2.5 per ton seems to be relatively
reasonable for a cost analysis.
Unit Processes - Separation
Most of the processes involving recovery of solid waste components
require a separation step. Many different systems have been
developed for various purposes, particularly for industrial
application. The efficiency with which separation is affected
and the degree to which the desirable materials can be segregated
80
-------
Table 30. Annual Costs of Combinations of Mills and Work Shifts
(Tollemache Mill)
One shift
Daily tonnage
Milling plant
operation cost
Depreciation
Total cost
Annual tonnage
Cost, $/ton
Two shifts
Daily tonnage
Milling plant
operation cost
Depreciation
Total cost
Annual tonnage
Cost, $/ton
Table 31. Annual
One shift work
Total cost, $/yr
Annual tonnage
Cost, $/ton
Two shift work
Total cost, $/yr
Annual tonnage
Cost, $/ton
1
105
40,500
21,600
62,100
25,700
2.42
210
76,000
24,300
100,300
51 ,400
1.95
Cost of
1
60,900
1 5 ,400
3.95
100,300
30,900
3.24
Number of Mill
2
210
59,500
38,000
97,500
51 ,400
1.89
420
114,700
41 ,300
156,000
102,800
1.52
s Operated
3
315
72,200
50,900
123,100
77,100
1.60
630
139,200
55,900
195,100
154,200
1.26
4
420
91,900
64,200
156,100
102,800
1.57
840
178,100
70,400
248,500
205,600
1.21
Combinations of Mills and Work Shifts
Number
2
95,800
30,900
3.17
153,400
61 ,800
2.48
of Mills
3
120,500
46,400
2.60
190,600
92,700
2.05
4
152,700,
61 ,800
2.48
242,500
123,600
1.96
81
-------
are significant factors contributing to the cost and marketability
of the recovered product. Until recent years hand sorting was the
major means of separation when the objective was basically for
salvage purposes. Hand sorting is no longer a feasible technique
for separation because of low salvage price, low efficiency of the
system, labor costs and human fallibility. Drobny et oJL. (1971)
report a cost of $3.50 to $5.00 per ton of newsprint separated
when using low cost labor.
There is inadequate information to evaluate the effectiveness of
mechanical separation processes for solid waste. Experimental
data need to be collected to determine interrelationships among
such properties as particle size and distribution in the mixed
feed, separation efficiency, capacity, moisture content and
material type. A complete separation of all components in the
refuse is seldom possible. Mechanical separation is used to
divide a mixture into two or more fractions where each group is
uniform in a certain property. Solids can be graded based on
differences in dimensions, shape or specific gravity. In special
cases, the roughness of the surface, col or .or magnetic properties
of the particle can be employed. In general, two types of
separation techniques can be used, i.e. wet or dry process. For
specific cases, an investigation should be undertaken to select
the optimum separation technique. Wet techniques are not used
prior to a dry recovery process because of the dewatering and
drying costs involved.
The characteristic of each separation unit process can be found
elsewhere (Rietema and Verver, 1961). However, there is a
universal lack of cost and operational data. An important part of
the cost involved is auxiliary items. Hence the sum of the unit
costs may not exactly represent the total cost. Also separation
and size reduction should be considered together from the particle
size view point. It is clear that different materials in mixed
solid waste are reduced to different sizes by the shredding step.
Screening may present better efficiency for separation of some
components, particularly for such items as glass and grit.
Using a digester for anaerobic fermentation of refuse with character-
istics as shown in Table 28 requires the separation of inert
materials. The inert materials account for about 22 percent by
weight of the solid waste. The primary disadvantages of these
materials are the volume they occupy in the reactor and the
added materials handling problems. The abrasiveness of the inert
material also will cause excessive mechanical wear on the equip-
ment. These materials may also be the source of toxic substances
and decrease the organisms' activities. The initial investigations
in this study indicated that the biological mechanism was not
inhibited by using unseparated refuse feed. However, the
82
-------
accumulation of potentially toxic materials in the reactors
should be avoided.
With the exception of plastics, the specific gravity of the inert
materials is high. This property can be advantageously employed
in the separation process. Hence for similar size particles, an
air or hydraulic separation process offers a feasible technique
for separation of the various components. Also, the use of
screens prior to the separator may introduce a more efficient
system by removing much of the fine grit and glass particles. For
gravity separation, water is usually preferred to air because air
classification requires a more narrow size distribution range.
Also, particle shape has a more significant effect on the air
separation technique. Additional problems such as a uniform
airflow make this separation technique difficult.
The problem with hydraulic separation is the feed concentration.
Because most of the techniques are applicable for low slurry
concentrations, a thickening or dewatering step is required prior
to the reactor in order to take advantages of the concentrated
feed for achieving significant retention times with small reactor
volumes.
Based on the degree to which the desirable materials can be removed,
some technique such as a "stoner" can be employed. The power
required is about 1 HP per ton and the amortized capital cost
varies $0.10 to $0.15 per ton (Drobny at at., 1971). If the
recovery of inert material is required, a more efficient system
should be introduced. The cost of magnetic separation is about
$13.5 per ton of material recovered or about $0.90 per ton of
refuse. Balistic separation also provides a classification into
two groups, organic and inorganic materials. However, more
research and pilot plant study are required for the evaluation
of the separation system. For this initial investigation, a
cost of about $0.50 per ton may be assumed. This is in the middle
of the reported costs, but these costs are highly dependent upon
the type of system and separation efficiency required.
Unit Process - Anaerobic Fermentation
From economic considerations, the most important variables in the
anaerobic process are temperature and detention time. The other
factors influencing the anaerobic process, such as pH, are assumed
to be controlled without any economic impact on the process. These
two factors were assumed to be the more significant variables and
the optimum conditions are defined in terms of temperature and
detention time. The laboratory data obtained in this study were
used as input to the model.
83
-------
Gas production for various temperatures and detention times is
shown in Figures 12 and 16. Table 32 was developed from these
figures. The variation between two consecutive points is taken
as a linear function. The loading rate for which these data were
obtained was 0.21 pounds of dry solids per cubic foot per day.
The gas production was found to be 11 scf per pound of volatile
solids destroyed. This was the basis for construction of Table 33.
Table 32. Gas Production (Cu Ft/Lb Dry Solids)
Temperature (°F)
95.0
104.0
105.8
107.8
109.4
113.0
118.4
127.4
132.8
140.0
Retention Time (OaysJ
4
1.40
2.33
2.49
2.59
2.58
2.08
2.64
3.33
3.74
4.23
8
2.30
2.98
3.09
3.14
3.09
2.55
3.14
3.88
4.20
4.49
10
2.58
3.20
3.25
3.33
3.26
2.70
3.33
4.07
4.35
4.57
15
2.86
3.60
3.71
3.74
3.67
3.05
3.57
4.29
4.54
4.79
20
2.95
3.85
3.95
3.95
3.82
3.14
3.82
4.51
4.75
4.93
30
3.05
4.10
4.20
4.20
4.09
3.36
4.06
4.70
4.88
4.96
Table 33. Volatile Solids Destruction (Percent)
Temperature (°F)
95.0
104.0
105.8
107.8
109.4
113.0
118.4
127.. 4
132.8
140.0
Retention
4
17.4
29.0
31.0
32.2
32.1
25.9
32.9
41.5
46.6
52.7
8
28.6
37.1
38.5
39.1
38.5
31.7
39.1
48.3
52.3
55.9
10
32.1
39.8
40.5
41.5
40.6
33.6
41.5
50.7
54.2
56.9
Time (Days)
15
36.6
44.8
46.2
46.6
45.7
38.0
44.4
53.4
56.5
59.6
20
36.7
47.9
49.2
49.2
47.6
39.1
47.6
56.2
59.1
61.4
30
38.0
51.0
52.3
52.3
50.9
41.8
50.6
58.5
60.8
61.8
84
-------
No solids recycling was employed and a complete mixed system was
assumed where the outflow solids concentration is equal to the
solids concentration in the reactor. This assumption can be
employed to calculate the slurry content of the outflow based on
the volatile solids destroyed in the reactor.
A computer simulation technique was employed to calculate the cost
of producing gas. The total cost of the system is divided into
two categories, capital and operational cost. The following
discussion presents the cost analysis of the important factors
which contribute to the total cost.
Capital Cost
The cost of the reactor is an important factor in the cost
analysis. This cost relationship can be expressed in terms of
the volume of the reactor. The volume of the reactor based on
fixed feed concentration is equal to:
V-5-t (14)
Lp
where: p = density of feed stream, Ib/cu ft
V = volume of reactor, cu ft
M = Ibs of solid refuse fed per day
C = feed concentration, %
t = detention time, days
If a constant loading rate is used, then C in Equation 14 can
be presented as
C=(F)(t)(V) (15)
where: F = loading factor, Ibs/cf/day
However, the model permits both kinds of reactor volume design,
i.e. a fixed or variable loading factor.
The cost formula developed by Smith (Eckenfelder and Adams, 1972)
was selected from the existing formulae and a construction index
value of 1.23 was used to update the cost into January 1973. This
index was obtained from sewage treatment cost index (Ewg-tnae^cng
Wew6 Reco-td, 1972). Use of the ENR Construction Cost Index
introduces a higher value than the sewage treatment cost index.
85
-------
CC = V(1.65 + 17/V0'87) (16)
where: CC = total capital cost in $1000, January 1973
V = reactor volume in 1000 cf
A life of 25 years and a current value of interest rate of six
percent were used. A factor of 0.07823 for amortization of the
capital cost corresponds to this project life and interest rate.
Operating Costs
The operating costs are separated according to the different
functions. The most important operating costs are heating, mixing
and residue dewatering.
Heating. Heating costs are an important design factor.
Elevating the temperature to increase the rate of solids destruction
requires more heat energy. However, a reduction in the heating
requirements can be obtained by operating at a shorter detention
time. This, in turn, leads to a lower capital cost. The heat
must be adequate to increase the feed stream temperature to the
reactor temperature level and also cover the heat loss from the
system.
The heat required to elevate the temperature of the feed stream
is proportional to the mass flow rate and the difference between
the feed stream and the reactor temperature. However, a part of
this heat can be recovered from the discharged flow by use of a
counter current heat recovery system. An extreme condition for
an initial temperature of 35°F for the feed slurry was selected
for the economic analysis.
HI = S(c)(T1 - TQ) (17)
where: S = feed stream, Ibs/hr
c = specific heat constant, BTU/lb-°F
T, = reactor temperature, °F
T = feed stream temperature, °F
HI = heat required, BTU/hr
An important part of the heat loss results from the surface heat
loss from the reactor. A cylindrical digester with a height equal
to the diameter was assumed in order to achieve the minimum surface
86
-------
area. Equation 18 is used to calculate the heat loss from the
surface area.
Q = lU.A.^ - Tj) (18)
where: Q = heat loss, BTU/hr 2
U. = heat loss per unit area per hour, BTU/hr-ft -°F
T^ = outside temperature, °F
A = surface area, sq ft
Operating the digester at high temperature introduces heat loss
by evaporization of the water which leaves the system as a gaseous
product. This heat requirement can be calculated from Equation 19.
Q = xw (19)
where: \ = unit latent heat, BTU/lb
w = weight of the material changing state, Ib
Q = heat loss, BTU
The actual volume of a gas generated by water vapor is obtained
from Equation 20.
vo = 16-4 v i (20)
where: P = vapor pressure of water at T,
W •
P = atmospheric pressure, 29.6 in. Hg
V-V = volume of the water vapor
An empirical relation was developed to present the heat loss
versus gas production. Equation 21 shows this relationship.
QL = 1 + 0.16 (T1 - 32) (21)
where: T, = reactor operating temperature, °F
Q. = heat loss, BTU/cf gas produced
87
-------
The metabolism of organic compounds releases heat energy. The heat
released in aerobic systems is substantial and raises the temper-
ature of the system. However, in anaerobic systems the metabolic
heat released is low at about 1.6 BTU/lb refuse. Most of energy
released from the organic material is in the form of methane gas
which can be recovered by using it as a fuel.
Mixing Energy. Mixing is a relatively important operational
factor. THe criterion for mixing is to keep the organic materials
in suspension in order to increase the contact between the substrate
and microorganisms and improve the fermentation process. Mixing
should be adequate to prevent deposition of organic materials and
to keep the reactor contents relatively homogenous. The energy
required for mixing is a function of the volume, solids concen-
tration, the type of the solid material and the nature of the
mixing equipment. Unfortunately, limited data are available on
mixing requirements for anaerobic processing of concentrated
organic refuse. Studies of high rate municipal waste digestion
reported that a circulating velocity of 1.5 to 2 fps at the floor
is necessary in order to prevent any substantial amount of
deposition of material. The minimum velocity required is dependent
upon the characteristics of the particles. Minimum velocities as
high as 2.5 fps have been reported (WPCF, 1966). When operating
at high slurry concentrations, continuous mixing is important
because it is sometimes difficult to resuspend the deposited
materials.
There are several mixing techniques that have been used for
anaerobic fermentation processes. Sludge recirculation, gas
circulation, pumping and mechanical draft-tube mixing are the
most common. Each process has some advantages and disadvantages.
In order to determine the proper mixing conditions and the
most economical mixing equipment, more study seems to be required.
Also, the type of the size reduction and separation processes may
affect the mixing requirements. Based upon current practice for
digester mixing, a value of 0.185 HP per 1000 cubic feet of
reactor volume was used for four percent slurry concentration
(McMahon, 1965). The following equation was used to calculate
mixing power required for different slurry concentrations.
P = 0.185 x x V (22)
where: P = power requirements, HP
C = reactor slurry concentration, %
V = reactor volume, 1000 cu ft
-------
Reactor Residue Dewaten'ng. The dewatering operation is related
to the slurry concentration and the characteristics of the material
resulting from the anaerobic fermentation of the refuse. For
this reason, operating costs for dewatering are taken as part of
the total operating cost. The filter yield was assumed to be
linearly proportional to the slurry concentration. The power
requirement was estimated to be 1/8 HP per square foot of filter
area. The laboratory filterability studies found higher filter
yields from the concentrated slurry. Therefore, this model requires
more filter area for the lower slurry concentrations which result
from operation of the system at high temperatures and long
retention times. However, the mass of solids available for
dewatering is reduced under these conditions.
Additional studies of the dewatering step are necessary. When
operating with a relatively low slurry concentration, a thickening
step prior to dewatering of the residue may be advantageous.
However, for the purpose of this analysis, the dewatering costs
were based upon the characteristics of the reactor effluent. The
filter area is the controlling factor in determining the capital
cost of the filter. Equation 23 shows the relationship used to
determine the capital cost for the filter. The area requirements
were determined from the reactor effluent slurry concentration
using filter loadings found from the laboratory studies.
Log C = 0.65 - 0.66A (23)
where: C = cost, $100 per square foot of filter area
A = filter area, 100 square feet
Economic Evaluation of the Anaerobic Fermentation Process
A computer simulation technique was employed in order to evaluate
the process and to find the least cost condition for gas production.
As previously mentioned, temperature and detention time were taken
as major variables. The program for this simulation is shown in
Appendix D. The input data are given on the first page of
Appendix D.
The concentration of the reactor slurry is controlled by the
concentration of the feed slurry and the quantity of solids
converted to gas. Table 39 in Appendix E shows the calculated
slurry concentration in the reactor at various feed concentrations,
temperatures and retention times. As would be expected, the
minimum solids concentration in the reactor for a given feed slurry
concentration is achieved at the highest temperature and longest
detention time. The effect of temperature and retention time on
the reactor slurry concentration is graphically displayed in
89
-------
Figure 22. These data represent only those conditions resulting
from a feed slurry concentration of 20 percent solids. This
figure shows the effect of the efficiency of converting the
volatile solids to gas.
The energy required for mixing is a function of reactor volume and
reactor slurry concentration. The mixing costs shown in Table 40,
Appendix E, are calculated on the basis of the reactor slurry
concentration and the reactor volume determined by the retention
time. The maximum mixing energy required is at the lowest
temperature and longest retention time. The cost of power for
mixing ranges from $2,350 per year at 140°F and 4.0 days retention
time to $20,700 per year at 95°F and 30 days retention time These
calculations are based on a system sized to process 100 tons per
day of dry refuse. Based on the refuse shown in Table 28, this
would be equivalent to 139 tons per day of refuse as received.
The heat requirements are primarily controlled by the reactor
operating temperature. Retention time, as it relates to reactor
size, also influences the heat requirements. The heat required
to maintain the temperature under various conditions is given in
Table 41, Appendix E. These calculations are based on a 75 percent
recovery of the heat in the reactor effluent slurry. It is obvious
that the minimum heat requirements are at the lowest temperature
and shortest detention time. Increasing feed concentrations decrease
the heat requirements for two reasons. Smaller reactors are
required for a given refuse quantity and retention time resulting
in less heat loss from the reactor surfaces. Also, increased
feed slurry concentrations reduce the mass of water that must be
heated when passing through the system. The heat requirements for
a 100 ton per day (dry refuse) system operating with a feed
slurry concentrations of 6 to 20 percent solids are shown in
Figure 23. The heat requirements are shown for a system operating
at the four day retention time. The feed slurry concentration is
the major factor in heat losses. The dashed line shows the heat
required to maintain the temperature if the retention time is
increased from four to eight days when the feed slurry concentration
is 20 percent solids.
The dewatering costs are primarily controlled by the feed slurry
concentration. For a given slurry concentration, the change in
reactor slurry concentration at different retention times and
temperatures is small in comparison to the concentration variation
range resulting from the substantial variations in feed slurry
concentrations studied. Table 42, Appendix E shows the dewatering
costs for a 20 percent feed slurry. The effect of the feed
slurry concentration is shown in Figure 24. In this figure,
the solid lines represent the dewatering costs for the 100 ton
(dry refuse) per day system operating at four days retention time.
90
-------
20
Concentration of Feed Slurry
V)
•o
o
t/J
18
+J
§ 16
i.
O
4J
O
«J
(1)
14
12
Reactor retention time - days
10
90
100
110 120
Temperature - °F
130
140
Figure 22. The Effect of Temperature and Retention
Time on Reactor Slurry Concentration
91
-------
120
Feed slurry
concentration
110 120
Temperature - °F
Figure 23. Heating Requirements for Operating at Various
Temperatures and Feed Slurry Concentrations
92
-------
4.0
3.5
103! Feed Slurry
S 3.0
o
o
J2 2.5
-------
The dashed lines represent the system operating at a 30 day
retention time. The dewatering costs are slightly higher for the
longer retention times due to the more dilute slurry. The
calculated filter yield is lower for the more dilute slurry.
Since the same volume of water must be filtered, the cost increases
slightly for the longer retention times. This figure shows that
dewatering costs are primarily a function of feed slurry concen-
tration. Temperature and retention time have very little effect
on this cost item.
The costs discussed above can be combined to determine the operating
and total costs for the anaerobic fermentation of refuse. These
do not include the preparation costs associated with shredding and
separation nor disposal costs for the dewatered residue. Table 34
presents a breakdown of these cost factors. It is evident that
the capital cost is a significant cost contributor under all
conditions.
Table 34. Operation and Capital Cost - Plant Capacity
of 100 Tons per Day of Dry Refuse - 20 Percent
Field Slurry
30 days, 95° F
Item
Heating
Mixing
Dewatering
Total operation
Capital cost
Total
Annual
Cost
$1000
6.59
20.70
1.58
28.87
64.90
93.77
Percentage
7.0
22.0
1.7
30.7
69.3
100.0
4 days, 140°F
Annual
Cost
$1000
9.38
2.35
1.61
13.34
10.50
23.84
Percentage
39.3
9.8
6.7
55.8
44.2
100.0
Operating at the minimum temperature (95°F) and maximum retention
time generates a total annual cost of $92,770 for a system receiving
a feed slurry with 20 percent solids. The capital costs account
for 69.3 percent of these costs. For these conditions, mixing
costs are also high, accounting for 22.3 percent of the total
costs. Operation of the system at the maximum temperature and
minimum retention time shifts the costs downward. The capital
cost contribution is reduced to 44.2 percent of the $23,840 annual
cost. The mixing cost is considerably reduced. At this high
temperature, the heating cost assumes a much higher proportion of
94
-------
the cost, accounting for 39.3 percent of the total annual costs.
The total annual costs for all operating conditions investigated
are given in Table 43, Appendix E.
The total annual costs become meaningful only when they are
compared to the benefits. The net benefits are calculated from
the quantity of gas produced. The value of the gas produced was
calculated on the basis of selling the gas for $0.45 per 1000 cu ft.
Based upon a heating value of 600 BTU per cubic foot, the energy
would sell for $0.75 per million BTU. The maximum benefits accrue
under conditions that yield the maximum gas production. The
temperature and retention time are the controlling factors in this
analysis. The net benefits have the production costs subtracted
from the value of the gas produced. The net benefits for all
conditions analyzed are given in Table 44, Appendix E. Figure 25
is a plot of the net benefits against temperature for various
retention times. This plot is for a feed slurry concentration of
20 percent solids. The maximum net benefits are obtained when the
system operates at the minimum retention time and maximum temper-
ature. Examination of this figure shows the maximum benefits are
obtained at 140°F and a retention time of four days. In the
mesophilic temperature range, the maximum benefits are obtained at
108°F with a retention time of eight days.
The effect of retention time and feed slurry concentration on the
net benefits at the optimum thermophilic and mesophilic temper-
ature can be seen in Figure 26. Slurry concentration is a major
factor in determining the net benefits under all operating conditions.
When operating at 108°F, the retention time is not a major factor
in net benefits when operating with a feed slurry concentration
of 15 to 20 percent. A retention time between 4 and 20 days
yields essentially the same net benefits. The cost of residue
disposal is not included. Therefore, the total system net
benefits would probably be at the 20 day retention time. This
will yield less residue for disposal. Other factors such as
operating stability would suggest operation at the longer retention
times. Since one of'the major purposes of this process is methane
production, it would be logical to operate under those conditions
that yield the maximum gas without significantly reducing the net
benefits.
In the thermophilic range, the curves for 15 and 20 percent feed
slurry concentrations are not as flat. However, the change in net
benefits between four and ten days retention time is small. When
the cost of residue disposal is included, the optimum retention time
may well shift to the longer times. As with mesophilic temperature,
operating stability and maximum gas production may warrent operating
in the range of 10 days retention time. Since the cost of pH control
has not been included in this analysis, the lower caustic requirements
95
-------
120
100
80
o
o
o
£ 60
0)
0)
CO
40
20
Legend
—— 4 days
— 8 days
— 15 days
-— — -20 days
— - — .—30 days
1
1
90
100
110 120
Temperature - °F
130
140
Figure 25. Net Benefits Resulting at Various Temperatures
and Retention Times, Feed Slurry - 20% Solids
96
-------
120
80
40
3
o
0)
0)
-40
-80
-120
108°F
- 140°F
12 18 24
Reactor Retention Time - Days
30
Figure 26. Net Benefits at Optimum Temperature for
Various Retention Times and Feed Slurry
Concentrations
97
-------
at the longer retention time may also shift the net benefits
slightly to the longer retention period.
Although the net benefits associated with thermophilie operation
are about 50 percent greater than when operating in the mesophilic
temperature, it is not clear that the thermophilie range will be
optimum. Operating at the higher temperatures may require an
increased operation and maintenance cost. Also, these costs assume
75 percent recovery of the heat content of the reactor effluent.
Recovery of only 25 percent of the heat contained in the reactor
effluent increases the heat required by about 60 percent. This
can contribute a significant cost increase under certain operating
conditions. With a feed slurry of 20 percent, the heating costs
would increase $5,600 per year at 140°F operating temperature.
This would reduce the net benefits associated with the thermo-
phi lie temperature range. Many of these factors must be investi-
gated on a large scale system to determine the exact optimum
operating conditions.
Mixing costs associated with the digestion of sewage sludge have
been applied to this system. Increased power requirements for
mixing could be significant at the longer retention times. At a
30-day retention time, mixing accounts for about one-fifth of the
total costs. Increasing the power requirements from 0.185 HP
per 1000 cu ft to 1.0 HP per 1000 cu ft for a four percent reactor
slurry would increase the total annual cost from $93,770 to
$185,000 for the conditions shown in Table 34. A similar analysis
of the four day retention time shows that the total annual costs
increase from $23,800 to $34,190. This is a 43 percent increase
as compared to a 100 percent increase at the 30-day retention time.
Conversely, a reduction in power requirements at the higher slurry
concentration could significantly reduce the mixing costs. There-
fore, due consideration must be given to obtaining better data on
the mixing power requirements.
The cost of producing gas is shown in Figure 27. This figure
shows the minimum cost per unit of gas produced for each temper-
ature and feed slurry concentration. This curve shows the minimum
cost of gas production is at 140°F. Table 45, Appendix E, presents
the gas cost for all operating conditions. For the 15 percent
feed slurry concentration, this cost was 9.6 cents per 1000 cu
ft of gas. In the mesophilic temperature, the minimum cost was
13.7 cents per 1000 cu ft. At a feed slurry concentration of
20 percent, the mesophilic and thermophilie costs were 11.1 and
7.8 cents per 1000 cu ft respectively. Increasing the feed
concentrati on to 25 percent decreases the cost of the gas to
9.5 and 6.7 cents per 1000 cu ft. As previously indicated, the
power required for mixing may significantly alter the operating
conditions that yield the maximum net benefits.
98
-------
60
50
o
o
o
40
30
g
I 20
IS
10
90
100
Feed slurry concentration
I
no r
Temperature - °F
130
140
Figure 27. Gas Production Costs for Various Temperatures
and Feed Slurry Concentrations
99
-------
Operation at the apparent optimum conditions will produce a gas
at a cost of eight to ten cents per 1000 cu ft. If this gas is
marketed at $0.75 per million BTU, a return of $2.50 to $3.00
per ton of dry refuse can be expected. If a shredding and
separation cost of $2.00 per ton is deducted, there is still a
revenue of $0.50 to $1.00 per ton of dry refuse recieved. On the
basis of the refuse characteristics given in Table 28, the revenue
would be $0.36 to $0.72 per ton of refuse as received. Therefore,
by intercepting the refuse on its way to the disposal site and
recovering the gas, an income of $0.36 to $0.72 per ton can be
recognized.
The above benefits have not included benefits relating to land
disposal of the refuse. Processing for methane recovery reduces
the quantity of solids to be disposed. The inert fraction
plus the dewatered reactor residue yield about 80 tons of residue
per 100 tons of refuse as received. However, this residue is more
dense and will require only about one-third the volume of landfill
required for unprocessed refuse. There will be definite cost
reductions both for transportation to the disposal site as well
as disposal costs. Because these costs are highly dependent upon
local conditions, they were not included in this cost analysis.
An additional cost factor not included is the cost for the disposal
of the sludge from the water pollution control plant. These costs
may range from as low as $15 per ton of dry solids to as high as
$60 per ton of dry solids. Again these costs are highly dependent
upon local conditions. A system processing 100 tons per day of
refuse (as received) would also process 2.9 tons of dry sludge
solids from the water pollution control plant. The cost credit
for processing this sludge could range from $43.50 to $174 per
100 tons of refuse processed. This would add a $0.435 to $1.74
cost credit to the $0.36 to $0.72 per ton revenue generated from
the gas production. Therefore, the net benefits that can accrue
from this system may range from $0.80 to $2.46 per ton of refuse
processed. As stated previously, this figure does not include
any allowance for reduction in disposal costs. Also, no credit
has been claimed for any salvage value of the metals that may be
separated from the refuse.
100
-------
REFERENCES
Andrews, J. F., Cole, R. D., and Pearson, E. A., "Kinetics and
Characteristics of Multistage Methane Fermentations,"
Univ. of California, Berkeley, SERL Report No. 64011,
December 1964.
Buswell, A. M., and Mueller, H. F., "Mechanisms of Methane
Fermentation," InduA&iial and Engine.vti.ng Chw&iy, 44,
550-552, 1952.
Chan, D. B., and Pearson, E. A., "Comprehensive Studies of Solid
Wastes Management-Hydrolysis Rate of Cellulose in Anaerobic
Fermentation," Univ. of California, Berkeley, SERL Report
No. 70-3, October 1970.
Ckm
-------
Enebo, L., and Pehrson, S., "Thermophilic Digestion of Domestic
Sewage Sludge and Cellulose Materials," Ao&x Po^ec/i.
Scand., 2V, 36, 1960.
En3, 7, 17-22,
January-February 1962.
Golueke, C. G., "Temperature Effects on Anaerobic Digestion of
Raw Sewage Sludge," Sewage and InduA&ujaJt Wa6/te4, 30, 1225,
1958.
Golueke, C. G., and McGauhey, P. H., "Comprehensive Studies of
Solid Wastes Management. First Annual Report," Univ. of
California, Berkeley, SERL Report No. 67-7, May 1967.
Golueke, C. G., "Comprehensive Studies of Solid Wastes Manage-
ment. Third Annual Report," Univ. of California, Berkeley,
SERL Report No. 70-2, June 1970.
Ham, R. K., Porter, W. K., and Reinhardt, J. J., "Refuse Milling
for Landfill Disposal, Pufa&tc Wo*fc6, December 1971.
Heukelekian, H., "Decomposition of Cellulose in Fresh Sewage
Solids," lnduA&Ua£ and En3
-------
McCarty, P. L., "Anaerobic Waste Treatment Fundamentals - Part
One, Chemistry and Microbiology," Puhtic. WoAfea, 95, 107-112,
September 1964(a).
McCarty, P. L., "Anaerobic Waste Treatment Fundamentals - Part
Two, Environmental Requirements and Control," Public. Wotfei,
95, 123-126, October 1964(b).
McCarty, P. L., "Anaerobic Waste Treatment Fundamentals - Part
Three, Toxic Materials and Their Control," Public. WoAka,
95, 91-94, November 1964(c).
McCarty, P. L. , "Anaerobic Waste Treatment Fundamentals - Part
Four, Process Design," Public. Woifca, 95, 95-99, December
1964(d).
McKinney, R. E. , and Conway, R. A., "Chemical Oxygen in Biological
Waste Treatment," Sewage and lndtu>&u.at Wat>teA, 29,
1097-1106, October 1957.
McMahon, T. E. , "Design of a Mechanically Mixed Digester," The.
KuJLtztLn OfJ EnQ4.ne.vU.ng and M.c.hite.ctuA.e.t Wo. 54, Univ.
of Kansas, Lawrence, 1965.
Municipal Rerfu6e VaJL, 2nd Ed., American Public Works
Association, Chicago, Illinois, 1966.
Pfeffer, J. T., "Increased Loadings on Digesters with Recycle of
Digested Solids," Journal orf DicuLvi PolfaUon Control
FedeAo£t
-------
Reese, E. T. , and Levinson, H. S. , "A Comparative Study of the
Breakdown of Cellulose by Microorganisms," PkyAioiogia.
PJtan&vtum, 5, 345-366, 1952.
Reese, E. T., "Enzymatic Hydrolysis of Cellulose,"
, 4, 39-45, 1956.
Rietema, K. , and Verver, C. G. , [Eds.], CycAonu Jin
Elsevier Publishing Co., Amsterday, 1961.
Snell, F. D., and Snell, C. T. , Colownzfrtic. Mvtkods, oft
Volume, 3, D. Van Nostrand, Princeton, New Jersey, 1961.
Speece, R. E., and McCarty, P. L., "Nutrient Requirements and
Biological Solids Accumulation in Anaerobic Digestion,"
Advance*
-------
APPENDIX A
DETERMINATION OF CELLULOSE AND NON-CELLULOSE CARBOHYDRATES
Apparatus
Waring blender
hotplate or bunsen burner
filtration apparatus
250 mi sidearm erlenmeyer flasks fitted to vacuum system
25 mi Gooch crucibles with asbestos pads
32 mm filtration holders for Gooch crucibles
Spectronic 20 set at 620 mu
1 cm tubes for Spectronic 20
volumetric pipets: 1 m£, 2 m£, 4 mi
graduated pi pet: 1 me in 1/10
ice bath
stirring rods
Reagents
60 percent (by volume) H2S04 - cool
anthrone reagent = 1 g anthrone/ 1 cone. f^SO^ must be between
four hours and nine days old
dextrose solution - 100 mg anhydrous dextrose/ 1 H20
Procedure
Preparation of Gooch crucibles with mats
Stir about 10 g asbestos into 1 i H20
Pour this mixture through the crucibles until a mat about
1/8 in. thick is formed. Make a new asbestos mixture when
the pads start to be made of coarse fibers.
Wash the pads with 60 percent H2S04 and then with H20
Dry overnight in a 70° oven and store in a dessicator
Standards
Standards must be prepared at the time the samples are
tested because the anthrone reagent is unstable.
Put H20 and dextrose solution into 11 colorimeter tubes in
the amounts indicated:
105
-------
Tube H20 m£ Dextrose mi
1 2 0
2 1.8 .2
3 1.6 .4
4 1.4 .6
5 1.2 .8
6 1.0 1.0
7 .8 1.2
8 .6 1.4
9 .4 1.6
10 .2 1.8
11 0 2
Separate the tubes before adding anthrone, because the color
development is very sensitive to heat.
Add 4 ml anthrone reagent rapidly to each tube. If a safety
pipette filter is used, use the smaller bulb to expel the
reagent. Stir well immediately after addition. Be sure that
there is no yellower region at the bottom of the tube.
Allow each tube to air cool for 10 minutes, then cool in an
ice bath.
Wipe each tube carefully to remove water drops and finger-
prints.
Do not put ice-cold tubes into the colorimeter. They will
fog up.
Let the colorimeter warm up for about 20 minutes before
standardizing.
Tube 1 is the reference blank. Use it to set 100 percent
transmittance on the colorimeter. Restandardize the
colorimeter during the test.
Determine the transmittance of each standard tube and plot
transmittance against ug carbohydrate on semi log graph paper.
106
-------
Samples
It Is advisable to do this test in duplicate, because
variation is easily introduced into the procedure.
Dilute the sample in H20 until a 2 mi aliquot contains between
10 and 200 ug carbohydrates. If the undiluted sample is very
thick or heterogeneous, use a blender to mix the first
dilution.
Boil 50 mi of the diluted sample for 15 minutes.
Allow to air cool, then bring back to its original 50 mi
volume with h^O.
Filter this solution through the Gooch crucibles into
erlenmeyer flasks. The filtrate will contain all non-cellulose
carbohydrates which have been broken down in boiling. The
residue will contain cellulose. Set aside this first filtrate.
Use clean erlenmeyer flasks for the second filtration. Add
about 20 mi of exactly 50 mi of 60 percent H2S04 to the
crucibles. With a stirring rod, gently scrape the residue
from the surface of the asbestos pad, and mix the residue
into the acid.
Allow the mixture to digest 15 to 30 minutes, then draw the
acid in the crucible and the balance of the 50 mi into the
flask.
Put a 2 mi aliquot of the first filtrate in a colorimeter
tube. Follow the procedure given for the standards for the
addition of anthrone, mixing and cooling. Determine ug
non-cellulose carbohydrate in the 2 mi aliquot from the
colorimeter and the standard transmission line.
Put 2 mi HoO and 0.5 mi of the second filtrate into a
colorimeter tube. Mix by gentle shaking and allow it to
air cool. Follow the procedure given for the standards for
the addition of anthrone, mixing and cooling. Determine ug
cellulose present in the 0.5 mi aliquot from the colorimeter
reading and the standard transmission line.
107
-------
APPENDIX B
FILTER TEST LEAF PROCEDURE
1. Determine the percent solids in the slurry before starting
test work and record.
2. With the test leaf tubing valued off, read the static vacuum
pressure and record in inches mercury.
3. Place two liters or more of sludge in an open beaker big
enough to accommodate the test leaf.
4. Immerse the fabric-covered test leaf in the agitated slurry
(about two inches below surface). Apply full vacuum at time
zero and filter for the predetermined form time. Record
form time and vacuum.
5. Remove the leaf from the slurry after the form portion of
the cycle is completed and allow air to be pulled through
the cake and leaf for the remainder of the cycle (dry time).
Record dry time and vacuum reading.
6. At the end of the drying cycle shut vacuum off and allow
tubing to drain completely. Record the amount of filtrate
obtained.
7. Place the filter cake on a dry tared container, weigh and
record.
8. Dry the cake in an oven at 103°C, reweigh and record cake
dry weight.
9. Calculate filter yield in pounds per hour per square foot
of area, and percent moisture in the cake formed.
10. Determine the suspended solids concentration of the filtrate.
108
-------
Appendix C
Table 35. Alkalinity and Volatile Acids Determinations - 35°C
o
ID
1
1530
1550
1150
1280
1350
1240
1260
1240
-
1120
1305
1150
1215
1190
1190
1260
1350
1395
1395
1350
1285
1350
1260
2
1620
1550
1215
1280
1325
1280
1260
1380
1305
1210
1260
1215
1190
1280
1240
1460
1510
1485
1395
1390
1420
1485
1305
Alkalinity -
3
1820
1910
1410
1665
1640
1510
1570
1560
1530
1530
1485
1418
1370
1440
1350
1530
1485
1530
1530
1415
1400
1575
1370
4
1840
1935
1730
1935
1960
1910
1890
1960
1945
1870
1775
1730
1690
1755
1685
1800
1850
1820
1755
1708
1760
1800
1820
• mg/£ Ca C03
5
1910
-
1910
1935
2040
2040
2050
2040
2200
2140
1910
1890
1935
1960
2070
2140
2040
2050
2040
2060
2050
2120
2070
6
2280
2360
2250
2000
2360
2500
2280
2808
2680
2510
2420
2360
2320
2520
2340
2740
2540
2480
2410
2340
2340
2380
2290
7
2520
2590
2680
2500
2630
2700
2650
2640
2740
2790
2660
2650
2658
2725
2810
2930
2810
2810
2790
2790
2320
2820
2790
8
2850
3110
2920
3130
3280
3130
3240
3370
3400
3510
3310
3260
3350
3370
3150
3330
3500
3370
3460
3440
3420
3550
3440
Volatile
1
424
484
644
292
256
244
670
646
557
480
414
454
418
558
475
470
380
327
2
408
390
358
586
281
281
364
5Z5
452
420
414
406
372
337
360
543
299
405
3
408
_
210
293
159
195
182
232
183
288
260
615
418
325
232
399
342
250
Acids
4
440
563
160
464
122
171
195
183
183
264
331
522
248
290
313
554
348
260
- mg/f. Acetic
5
344
148
232
171
134
244
183
281
242
262
442
360
395
313
349
318
205
200
6
188
465
359
220
159
232
159
134
183
240
262
278
371
372
360
338
353
211
7
282
422
865
293
146
220
354
281
183
312
331
349
278
418
545
294
174
264
R
376
267
179
263
208
98
278
390
281
300
497
487
255
418
626
410
353
211
-------
Table 36. Alkalinity and Volatile Acids Determinations - 40°C
1
1190
1260
1239
1330
1210
1200
1310
1215
2
1170
1260
1282
1330
1350
1315
1260
1370
Alkal
3
1418
1690
1640
1710
1330
1700
1465
1580
inity -
4
1556
1730
1775
1780
1780
1790
1580
1670
Table 37.
1
1360
1080
1170
878
1290
2
1280
1060
1180
968
1280
Alkal
3
1620
1040
1090
945
1560
inity -
4
1730
1320
1260
1120
1540
mg/l CaC03
5 6
1578 1910
1930 2340
2000 2205
1980 2480
2020 2450
1930 2290
1730 1980
1730 2160
Alkalinity
mg/£ CaC03
5 6
1560 1760
1330 1760
1620 1710
1500 1570
1710 2090
Volatile Acids - mg/£
7
1665
2540
1757
2740
2860
2860
2070
2140
8 1
2115 240
2790 188
2930 249
2970 505
3180 170
3270 565
2050
2140
and Volatile Acids
2
278
200
177
293
293
550
3
192
177
194
298
218
640
Determinations
4
229
216
216
555
205
465
- 50°
5
426
497
475
430
218
336
C
Volatile Acids - mg/£
7
1930
2280
2140
1980
_
8 1
2040 -
2560 -
2430 292
2340
_
2
150
271
210
3
59
204
296
4
288
229
256
5
133
254
267
Acetic
6
655
730
610
650
580
585
Acetic
6
239
350
388
7
1390
906
825
550
650
665
7
199
330
378
8
1115
911
1355
224
975
900
8
218
482
505
-------
Table 38.
Alkalinity -
1
990
1120
1080
1200
890
960
1120
2
1370
1460
1330
1400
1160
1120
1250
3
1530
1600
1540
1550
1220
1150
1180
4
1530
1690
1550
1640
1510
1360
1340
Alkal
inity and Volatile Acids Determinations -
• mg/£ CaCO.,
5
1530
1390
1460
1560
1560
1340
1210
6
1580
1550
1430
1520
1730
1610
1390
7
1510
1640
1430
1440
1550
1500
1380
8
1580
1920
1840
2000
2000
1970
1
102
92
143
Volatile
2 3
39 67
69 71
65 -
55°C
Acids - mg/£ Acetic
4
70
92
58
99
5
146
262
216
6
210
290
164
198
7
350
360
236
237
8
552
339
223
216
-------
APPENDIX D
PROGRAM INPUT DATA
MIXING POWER REQUIRED HP/1000 CU FT AT 4 PERCENT CONCENTT. = 0.185
POWER COST CENTS/KWH = 1
EXTERNAL TEMPERATURE DEGREES F. = 45
INLET SLUDGE TEMPERATURE DEGREES F. =35
HEATING COST CENTS/1000 CU FT = 55
SLUDGE DEWATERING POWER REQUIRED HP/SQ. FT OF FILTER AREA = 0.125
GAS VALUE CENTS/1000 CU FT = 45
PLANT CAPACITY TONS PER DAY = 100.0
-------
1 DIMENSION
? DIMENSION GPCft,10>,VSn(6,lO>,SC<6»10),HL(6»10>,SH<6,10),WV(6,10>
3 DIMENSION TH(6,1(M»HC(6»10)*FVC6»10)»FAC6»10)»DC(6»iO)»TCf6'10)
4 DIMENSION GF(b>I 0 ) ,KBf&»*°VMP(6*10)
5 DIMtNSIUN FJ<7>
A HEAL M,NH,MP
7 FJ(1)=3.U
8 FJ(2)s6.0
9 fj(3)*10.0
10 FJ(«)=15.0
11 Fj(5)=20.0
12 FJ(6)s?5.0
13 ^J(7)=30.0
14 plslVB.O
15 H?sb.06
16 P3=0.4b
17 FOsO.U
16 Tl=3b.O
19
20
21 NEADCb»lUl)(DAV(lD)»lD«l»6)
22 RrAD(5»102)((^h(IO*lT)*lD=t»6>*IT=l»10)
23 READC5»102)((VSD(10*1T)»IDs!,«)•!T«l»10)
24 HRlTt(6,307)
25 HR1T§.(6,201 )
2.6 WRI Ft(6»IDsl>6)* IT = l* 10)
28 WRlTE(6,30b)
29 HRITL(6»201 )
30 MR1TL(6»202)
31 tiRHL(6>312)( TFM(1T)»(VSD( ID*11 )»IUsl»6)»lTsl,1
3? DD 6U J=l»7
-------
33 un iu in=i*6
34 FC(IU)=FJ(J>
35 00 50 1T = U10
36 SC(IU*IT)«(1,-0.007JO*VSD(ID,I I))+FCCID>
37 50 CONTINUE
38 10 CONTINUE
39 DO 20 1D=1,6
40 TsDAY(lD)
41 VOL (1D)=M*1 /FC( ID)/6ii.b*1.0E2
42 VsVOL(lD)
43 Vl=V*1tOE-3
44 ACC(JU)r0.07B^*(l.6b*Vl+17.*Vl**0.13)*ltOE'3
45 UP 30 IT=1*10
46 MP(IU*1T)=SC(1C»IT)*11,95*V/4.0*1.Ot-9
47 T?sTtM(IT)
48 HL(IU*lT)=36.*V**0.67*(T2-TO)*1.0E-6
49 TaDAY(lD)
5C SH( IU,lT)sV*62.5*(-li'-Tl )/T*0 ,?b* 1 . Ot-6
51 lisGP(ID*U)
52 WV(IU, IT ) = ( ( 1 .ii840.16*( T2-35.0) )*M*1,OL-6*G)
53 TH( IO»IT) = HL( IOIT) + SH(1D'IT) + *VCID»IT)
54 HC(IU,IT)=Pl*lHCIO»n )*1 .OE-6
55 SC1=J>C(TU,1T)
56 IF(SCI.LI .1.6) FY1-1.
57 JF(SCl.GE.l.b) FYl=2.34*SCl-3.2
56 FY( !U»lT)sFYl
59 M(IU»IT)B2.6*V/T*SC(IU»IT)*1.0E-2/FYl
60 DC(IU,IT)*P2*tA(ID»IT)*1.0F-6
61 TClsMP(lU,IT)+Hl.( IU,IT)*UC( I 0, I T ) +ACC ( I U )
62 lC(Il)»lT)sTcl
63 OF(IU*IT>S1C1/K/G*1.OE11/360.
64 NB(IU,lT)=P3*M*G*360,*1.0E-9-TCl
-------
65 30
66 20 CONTINUE
67
68 WRIIE(6>600) fJ(J)
69 WRITK(6,201) IDAY(IU),1DM»6)
70 WRITi(o»2BO)(*C(lO)f10x1,6)
71 WRlTt(6,20b)
72 WR1IE(6>600) FJ(J)
73 WR11L(6*201) I DAY( ID ) »IU = l*6)
74 WRI1L(6*202)
75 WRITE<6,203) (TFMC1T)*CSC(ID»If)•ID=1»6)»IT=l»10)
76 WPITt(6»206)
77 WKI1E(6*600) FJ(J)
76 HR1TE(6»201 ) C DAY ( ID ) , 1D = U 6 )
79 WRITL(6,2HO)(VPL(IO)»lDsl»6)
80 WRITE(6»207)
81 WR1FE(6*600) FJ(J)
82 MRITL(6,201) (DAY(1U),1U=1,6)
63 WRIIL(6*280)(ACC(IU)>IU=1>6)
8.4 WRITL(6,206)
85 WRIIt(6,600) FJ(J)
86 WRITt(6,201) (RAY(ID),IU=1,6)
67 WPIfEC6*202)
88 WRITt(6,203) ( TEM( 1T)•(MP(IP,IT)*IOsl,6)* ITsl> 10)
89 WRI7L(6,209)
90 WRI(t(6»600) FJ(J)
91 NR11L(6,201) (DAY(lU)»JUxl«6)
92 WRITL(6,202)
93 WRIU(6,203) C TfM( 1 T ), ( HL ( ID* I T )> Il)=l » 6), I Tsl » 10)
94 WRITL(6>210)
95 WK11E(6*600) FJ(J)
96 WRITt(6,201) IDAY(ID)* 1U=1•6)
-------
97 WRITEC6,202)
98 WRITE(6,203) ITFMCIT)»CSHC10*IT)>ID«l»6)»IT«l>10 )
99 WRllE(6,2tl)
100 WRITEC6.600) FJCJ)
101 WRITEC6.201) IDAYf10),IDsl,6 )
102 WRITEC6,202)
103 WRITE(6,203) tTEMCiT)*CHVCID,IT)> ID«1»6)*IT«l»10 )
104 WRITE(b,212)
105 WRITE(6*600) FJCJ)
106 WRITE(6,201) (DAY(1U),1D=1,6)
107 WRITE(6,202)
106 WRITE(6.203) CTEMC11 ),CTHC 10, IF)»10 = 1>6)*ITB!»10 )
109 WRITE(6,213)
110 WRITEC6/600) FJCJ)
111 WRITE(6,201) (DAY(1U),1D=1,6)
112 WRITE(6,202)
113 WRITE(6,203) ( TFM( 1T),CHCC ID , IT),10=1,6),IT = l,10 )
114 WPITE(6,214)
115 WRITE(6*(SOO) FJCJ)
116 WRITE(6,201) (DAY(10),I Del,6)
117 WR1TE(6,202)
118 WRITE(6,203) ( TEM(1T), ( FY( I D,IT),ID=1,6)* IT = 1, 10 )
119 WRITE(6,215)
120 WRITE(6»600) FJCJ)
121 WR1TE(6,201) IDAYC10),1UB1,6 )
122 WRITEC6.202)
123 WRITE(6,203) (TEM(IT),(FA(ID*I^),ID=1,6),ITz1,10)
124 WRITE<6,216)
• ic mOffp/^.Af\/\\ C* I / I %
1 t J n>*AI|L\O^O«/^IJ * \J \ *J /
126 MR1TEC6.201) ( D A Y C IU ) , I l) = l , 6 )
127 WRITE(6,202)
128 WRITE(6,203) ( UM( 1T), ( DC(ID»IT)* ID=1»6),IT = 1,10 )
-------
129 WPITEC6.217)
130 WRITE(6*600) FJ(J)
131 WRITEC6,201) CpAY(lU),10el,6)
132 WRITt(6,202)
133 WRITE(6,203) ( TfM( 1 1 ) » ( TC ( ID* II ) * ID*1 > 6 )* I Tel » I 0 )
134 WRITE(6,200)
135 WRITE(6>600) Fj(J)
136 &R1TEC6.201) < pA Y( 1U ), IU = 1 , 6 >
137 WRITE(6,202)
138 WRITt(6,203) ( TEM( 1 T ) » ( GF ( I D* 1 1 ) • I D«l * 6 )» I Tel » 10 )
139 WRI1E(6«300)
UO WRITE(6»600) FJ(J)
1«1 WR1TL(6.201)
1*3 WRITE(6>203) ( TEM( 1 T } , ( NB( ID, I T )* ID = 1 »6 )* ITsl * 10}
144 60 CONTINUE
145 100 FOHMAT(lUX,10h7.2>
146 101 FORMATC IOX,6F ".0)
147 102 FORMAT(10X*6F^.3)
148 ?00 FDRMAT(lHl*///»«9x»2bHGAS COST *CEN1S PER 1000 CF '// )
149 ?01 FORMAT(1H , 18X, 21 HKi TENT ION TIME ( DAYS ) * 6(F 1 0. 1 * 3X), // )
150 ?02 FORMATUH >19X,J4HTEMP (DEG. F) )
151 ?03 FORMAT(1H , 21 X, F6 . 1 , 12X, E 10 . 3, JX, E10. 3, 3X, E 1 0. 3* 3X, E10. 3, 3X
1 *Ei0.3,3x*E10.3,3x /)
152 ?0fl FORMAT(lHl»//»flOX»33HFEED CONCENTRATION ( PERCENT) ///)
153 ?05 FORMAT(1H1«//*40X»30HSLURRY CONTENT ( PERCENT ) ///)
154 ?0ft FORMAT(1H1,//*40X»29HDIGESTER VOLUME ( CU FT ) ///)
155 ?07 FORMAT(lHl,//»40x,4bHAMORTlZED CAPITAL COST ( MlLL.S PER YEAR
I///)
156 ?08 FOPMAT(lHl*//*40x*36HMlXlNfi POWER (MILL. S PER YtAR ) ///)
157 209 FURMAT(1HI,//,40XMUHHEAT LOSS .
1 23H(MILL. BTli PER DAY ) ///)
-------
158 210 FORMAT(1H1,//,<*OX,1!5HSLUDGE HLATING *
1 23HCMILL. BTll PER DAY > ///)
159 211 FORMAT(1H1,//,40X,12HWATER VAPIJR »
\ 23HCMILL. flfU PER DAY ) ///)
160 212 FORMAT(1HI,//>40X»4JHTOTAL HEAT REQUIREU (MILLC B1U -&R OAO- //
161 213 FORMAT(1H1,//,40X/ J5HHtATlNG COST (MILL. I PtR YEAR ) ///)
162 211 FORMAT(1H1,//,40X,46HFILTER YIELD (LB PER SQUARE FEET PER HOUR
1) ///)
163 21S FURMAT(lHl,//,40X,39HKtQUIRED FILTER AREA (SQUARt FEET ) ///)
164 ?16 FORMAT(lHl*//*«0X.39HnEwATERlN& COST (MILL. * HER YEAR ) ///)
165 ?17 FDRMAT(lHl*//*40X,3aHTOIAL COST (MILL. S PER YtAR ) ///)
166 ?BO FORMAT(1H ,41X»Fl0,«, 3x»E10. ^»3X»E10.4,3x»El 0.4>3x»El 0.4,3x
1 »E.10.4»3X /)
167 300 FnRMAT(lHl,///,49X»32HNtT SENFFlT IN MILL. S P£K YtAR // )
168 302 FPRMAT(1H *19X,14HFEMP (DEG. F) MOF6.1 //)
169 307 FORMAT(1H1*//,20X,47HTABLE 7 GAS PRODUCTTONB (CU FT / LB SOLID
1 ) )
170 30fl Fl)RMAT(lHl,//,5?Ox»«9HTABLE 8 yHLATIuE SOulD uEbTRUCTION (
INT ) )
171 312 FORMATC1H * 21X,F6.1*1?X*6F7.3 )
172 600 FORMAT(1H ,39y, 20HFEED CONCENTRATION a *F6.1»3x*HH (PERCENT)
1 ///)
173 WRITE(6*400)
174 400 FORMAT(lHl)
175 STOP
176 END
**WARNING** FORMAT STATEMENT 302 is UNREFLRENCLD
SENTRY
-------
Appendix E
Table 39. Reactor Slurry Concentration Under Various
Operating Conditions - Percent Solids
Temperature
°F
4
Reactor Retention Time •
8 10
15
Feed Slurry Concentration -
95.0
104.0
107.8
109.4
113.0
118.4
127.4
132.8
140.0
5.24
4.73
4.59
4.59
4.87
4.56
4.18
3.96
3.69
4.75 4.59
4.38 4.26
4.29 4.18
4.31 4.22
4.61 4.53
4.29 4.18
3.88 3.78
3.71 3.63
3.55 3.51
4.40
4.04
3.96
4.00
4.34
4.06
3.66
3.53
3.39
Feed Slurry Concentration -
95.0
104.0
107.8
109.4
113.0
118.4
127.4
132.8
140.0
8.73
7.88
7.65
7.66
8.11
7.60
6.97
6.60
6.15
7.91 7.66
7.29 7.09
7.15 6.97
7.19 7.04
7.69 7.55
7.15 6.97
6.47 6.30
6.18 6.04
5.92 5.85
7.33
6.73
6.60
6.66
7.23
6.76
6.10
5.88
5.65
Feed Slurry Concentration -
95.0
104.0
107.8
109.4
113.0
118.4
127.4
132.8
140.0
13.1
11.8
11.5
11.5
12.2
11.4
10.5
9.9
9.2
11.9 11.5
10.9 10.6
10.7 10.5
10.8 10.6
11.5 11.3
10.7 10.5
9.7 9.5
9.3 9.1
8.9 8.8
11.0
10.1
9.9
10.0
10.8
10.1
9.2
8.8
8.5
• Days
20
6% Solids
4.39
3.90
3.85
3.92
4.29
3.92
3.54
3.41
3.31
10% Solids
7.32
6.50
6.41
6.53
7.15
6.53
5.90
5.69
5.52
15% Solids
11.0
9.8
9.6
9.8
10.7
9.8
8.9
8.5
8.3
30
4.34
3.77
3.71
3.77
4.17
3.78
3.44
3.34
3.29
7.23
6.28
6.18
6.28
6.95
6.31
5.73
5.56
5.49
10.8
9.4
9.3
9.4
10.4
9.5
8.6
8.3
8.2
119
-------
Table 39. Continued
Temperature
°F
95.0
104.0
107.8
109.4
113.0
118.4
127.4
132.8
140.0
4
17.5
15.8
15.3
15.3
16.2
15.2
13.9
13.2
12.3
Reactor Retention Time - Days
8
Feed
15.8
14.6
14.3
14.4
15.4
14.3
12.9
12.4
11.8
10
15
Slurry Concentration
15.3
14.2
13.9
14.1
15.1
13.9
12.6
12.1
11.7
14.7
13.5
13.2
13.3
14.5
13.5
12.2
11.8
11.3
20
- 20% Solids
14.6
13.0
12.8
13.1
14.3
13.1
11.8
11.4
11.0
30
14.5
12.6
12.4
12.6
13.9
12.6
11.5
11.1
11.0
Table 40. Mixing Costs for Various Temperatures and Retention
Temperat
°F
95.0
104.0
107.8
109.4
113.0
118.4
127.4
132.8
140.0
ure
4
3.34
3.01
2.93
2.93
3.10
2.91
2.67
2.52
2.35
Reactor Retention Time
8
6.05
5.58
5.47
5.50
5.88
5.47
4.95
4.73
4.53
10
7.32
6.78
6.66
6.73
7.22
6.66
6.02
5.78
5.59
15
10.5
9.7
9.5
9.6
. 10.4
9.7
8.8
8.4
8.1
- Days
20
14.0
12.4
12.3
12.5
13.7
12.5
11.3
10.9
10.6
30
20.7
18.0
17.7
18.0
19.9
18.1
16.4
16.0
15.7
120
-------
Table 41. Heating Costs for Various Operating
Conditions - $1000 per Year
Temperature
°F
4
Reactor Retention Time - Days
8 10
15
Feed Slurry Concentration
95.0
104.0
107.8
109.4
113.0
118.4
127.4
132.8
140.0
11.8
14.1
15.0
15.4
15.8
17.3
19.6
21.0
22.9
13.0 13.5
15.4 15.9
16.3 16.8
16.6 17.2
17.1 17.7
18.7 19.3
21.2 21.9
22.7 23.4
24.6 25.3
14.4
17.0
18.0
18.4'
18.9
20.6
23.3
24.9
26.9
Feed Slurry Concentration
95.0
104.0
107.8
109.4
113.0
118.4
127.4
132.8
140.0
7.5
9.1
9.7
10.0
10.2
11.2
12.8
13.9
15.2
8.5 8.8
10.1 10.5
10.7 11.1
10.9 11.3
11.2 11.6
12.3 12.8
14.1 14.6
15.2 15.7
16.5 17.0
9.5
11.3
12.0
12.3
12.5
13.7
15.7
16.8
18.2
Feed Slurry Concentration
95.0
104.0
107.8
109.4
113.0
118.4
127.4
132.8
140.0
5.3
6.6
7.1
7.2
7.3
8.1
9.4
10.3
11.4
6.1 6.4
7.4 7.7
7.9 8.2
8.0 8.4
8.1 8.4
9.0 9.4
10.5 10.9
11.3 11.7
12.3 12.8
7.0
8.4
8.9
9.1
9.2
10.2
11.7
12.6
13.7
20
- 6% Solids
15.1
17.9
19.0
19.3
19.9
21.7
24.6
26.2
28.3
- 10% Solids
10.0
12.0
12.7
13.0
13.2
14.6
16.6
17.8
19.2
- 15% Solids
7.4
9.0
9.5
9.7
9.8
10.8
12.5
13.4
14.5
30
16.4
19.5
20.6
21.0
21.7
23.7
26.7
28.4
30.6
10.9
13.1
14.0
14.2
14.5
16.0
18.2
19.4
20.9
8.1
9.9
10.5
10.6
10.8
12.0
13.7
14.6
15.8
121
-------
Table 41. Continued
Temperature
°F
4
Reactor Retention Time
8
10
15
Feed Slurry Concentration -
95.0
104.0
107.8
109.4
113.0
118.4
127.4
132.8
140.0
4.2
5.3
5.7
5.8
5.8
6.6
7.7
8.4
9.4
4.9
6.0
6.4
6.5
6.6
7.4
8.6
9.3
10.2
5.2
6.3
6.7
6.8
6.8
7.7
9.0
9.7
10.6
5.6
6.9
7.4
7.5
7.5
8.3
9.7
10.4
11.4
- Days
20
30
20% Solids
6.0
7.4
7.8
7.9
8.0
8.9
10.3
11.1
12.1
6.6
8.1
8.7
8.8
8.8
9.9
11.4
12.2
13.1
Table 42. Dewatering Cost for a Feed Slurry Concentration
of 20 Percent Solids - $1 OOP/Year
Temperat
Op
95.0
104.0
107.8
109.4
113.0
118.4
127.4
132.8
140.0
ure
4
1.55
1.57
1.57
1.57
1.56
1.57
1.59
1.60
1.61
Reactor Retention Time
8
1.57
1.58
1.58
1.58
1.57
1.58
1.60
1.61
1.62
10
1.57
1.59
1.59
1.59
1.58
1.59
1.61
1.62
1.62
15
1.58
1.59
1.60
1.60
1.58
1.59
1.61
1.62
1.63
- Days
20
1.58
1.60
1.60
1.60
1.58
1.60
1.62
1.63
1.64
30
1.58
1.61
1.61
1.61
1.59
1.61
1.63
1.63
1.64
122
-------
Table 43. Total Production Costs for All Operating
Conditions - $1000/Year
Temperature
°F
4
Reactor Retention Time - Days
8 10
15
Feed Slurry Concentration
95.0
104.0
107.8
109.4
113.0
118.4
127.4
132.8
140.0
51.8
54.0
54.9
55.3
55.8
57.2
59.5
61.0
63.0
83.8 99.4
85.8 102
86.7 102
87.1 103
87.8 104
89.2 105
91.5 107
93.0 109
94.9 111
138
140
141
142
143
144
146
147
149
Feed Slurry Concentration
95.0
104.0
107.8
109.4
113.0
118.4
127.4
132.8
140.0
33.2
34.6
35.2
35.4
35.7
36.6
38.1
39.0
40.3
53.7 63.7
54.9 64.9
55.5 65.5
55.7 65.7
56.3 66.4
57.1 67.1
58.5 68.4
59.3 69.3
60.5 70.4
88.4
89.5
90.0
90.3
91.3
91.9
93.0
93.8
95.0
Feed Slurry Concentration
95.0
104.0
107.8
109.4
113.0
118.4
127.4
132.8
140.0
24.1
25.1
25.5
25.7
25.9
26.6
27.7
28.4
29.3
38.9 46.1
39.8 46.9
40.1 47.3
40.3 47.5
40.8 48.0
41.3 48.5
42.3 49.4
42.9 50.0
43.7 50.8
63.7
64.4
64.7
65.0
65.9
66.2
66.8
67.4
68.2
20
- 6% Solids
177
179
180
180
182
182
185
186
188
- 10% Solids
113
114
114
115
116
116
117
118
119
- 15% Solids
81.5
81.6
82.0
82.3
83.6
83.5
84.0
84.5
85.3
30
254
255
256
256
259
259
261
262
264
162
162
162
163
165
165
165
166
168
117
116
116
117
119
118
118
119
119
123
-------
Table 43. Continued
Temperature
°F
4
Reactor Retention Time •
8 10
15
Feed Slurry Concentration -
95.0
104.0
107.8
109.4
113.0
118.4
127.4
132.8
140.0
19.6
20.4
20.7
20.9
21.0
21.6
22.5
23.1
23.9
31.6 37.3
32.2 37.9
32.5 38.2
32.6 38.3
33.0 38.8
33.4 39.2
34.2 39.8
34.7 40.3
35.4 41.0
51.4
51.8
52.1
52.3
53.1
53.3
53.7
54.2
54.8
• Days
20
20% Solids
65.7
65.5
65.8
66.1
67.3
67.1
67.3
67.7
68.4
30
93.8
92.7
92.9
93.3
95.2
94.5
94.3
94.6
95.4
124
-------
Table 44. Net Benefits Accrued Under All Operating
Conditions - $1 OOP/Year
Temperature
°F
4
Reactor Retention Time •
8 10
15
Feed Slurry Concentration -
95.0
104.0
107.8
109.4
113.0
118.4
127.4
132.8
140.0
-6.4
21.5
29.0
28.3
11.6
28.4
48.4
60.2
74.0
-9.2 -15.8
10.7 2.2
15.0 5.5
13.0 2.9
-5.2 -16.1
12.6 3.0
34.2 24.6
43.1 32.3
50.6 37.5
-45.5
-23.6
-20.0
-22.7
-43.8
-28.2
- 7.1
- 0.4
- 5.8
Feed Slurry Concentration -
95.0
104.0
107.8
109.4
113.0
118.4
127.4
132.8
140.0
12.1
40.9
48.8
48.2
31.7
48.9
69.8
82.2
96.8
20.8 19.9
41.6 38.8
46.2 42.4
44.4 39.9
26.3 21.1
44.7 40.8
67.2 63.4
76.7 71.7
85.0 77.6
4.3
27.2
31.2
28.6
7.5
23.8
46.0
53.3
60.2
Feed Slurry Concentration -
95.0
104.0
107.8
109.4
113.0
118.4
127.4
132.8
140.0
21.2
50.4
58.4
57.9
41.5
59.0
80.2
92.8
108
35.6 37.5
56.8 56.8
61.6 60.6
59.8 58.2
41.8 39.5
60.4 59.4
83.4 82.5
93.2 91.0
102 97.3
28.9
52.3
56.5
53.9
33.0
49.5
72.2
79.7
87.0
• Days
20
6% Solids
-81.4
-53.9
-51.6
-56.3
-79.8
-58.7
-38.4
-32.1
-28.2
10% Solids
-17.6
11.0
13.7
9.1
-14.4
7.5
28.8
35.8
40.5
15% Solids
14.1
43.2
46.0
41.4
18.2
40.3
62.1
69.4
74.4
30
-155
-122
-120
-124
-150
-128
-109
-104
-104
-63.5
-29.1
-26.4
-30.5
-56.3
-33.3
-13.2
- 8.1
- 6.9
-17.8
17.1
20.0
16.0
- 9.7
13.6
34.2
39.6
41.2
125
-------
Table 44. Continued
Temperature
°F
95.0
104.0
107.8
109.4
113.0
118.5
127.4
132.8
140.0
4
25.8
55.1
63.2
62.7
46.4
64.0
85.4
98.1
113
Reactor Retention Time •
8
Feed
43.0
64.4
69.3
67.5
49.6
68.3
91.5
101
110
10
15
Slurry Concentration -
46.3
65.8
69.7
67.3
48.6
68.7
92.0
101
107
41.2
64.8
69.1
66.6
45.7
62.4
85.3
92.9
100
• Days
20
20% Solids
29.9
59.2
62.2
57.6
34.4
56.7
78.8
86.2
91.4
30
5.0
40.2
43.2
39.2
13.6
37.1
58.0
63.5
65.3
126
-------
Table 45. Gas Production Costs Under Various Operating
Conditions - 1/1000 cu ft
Temperature
°F
4
Reactor Retention Time -
8 10
15
Feed Slurry Concentration -
95.0
104.0
107.8
109.4
113.0
118.4
127.4
132.8
140.0
51.4
32.2
29.5
29.8
37.2
30.1
24.8
22.7
20.7
50.6 53.5
40.0 44.1
38.4 42.7
39.1 43.8
47.8 53.3
39.4 43.7
32.8 36.6
30.7 34.7
29.3 33.6
67.1
54.1
52.4
53.6
65.0
56.0
47.3
45.1
43.3
Feed Slurry Concentration -
95.0
104.0
107.8
109.4
113.0
118.4
127.4
132.8
140.0
33.0
20.6
18.9
19.0
23.8
19.3
15.9
14.5
13.2
32.4 34.3
25.6 28.2
24.5 27.3
25.1 28.0
30.7 34.2
25.2 28.0
20.9 23.3
19.6 22.1
18.7 21.4
42.9
34.5
33.4
34.2
41.6
35.7
30.1
28.7
27.5
Feed Slurry Concentration -
95.0
104.0
107.8
109.4
113.0
118.4
127.4
132.8
140.0
24.0
15.0
13.7
13.8
17.3
14.0
11.5
10.5
9.6
23.5 24.8
18.5 20.3
17.8 19.7
18.1 20.2
22.2 24.7
18.3 20.2
15.1 16.8
14.2 16.0
13.5 15.4
31.0
24.8
24.0
24.6
30.0
25.7
21.6
20.6
19.8
• Days
20
6% Solids
83.3
64.4
63.1
65.5
80.3
66.3
56.8
54.4
52.9
10% Solids
53.3
41.0
40.2
41.7
51.3
42.3
36.1
34.5
33.6
15% Solids
38.4
29.4
28.8
29.9
37.0
30.4
25.9
24.7
24.0
30
116
86.3
84.6
87.1
107
88.6
77.1
74.7
74.1
73.9
54.9
53.7
55.4
68.3
56.4
48.9
47.3
46.9
53.1
39.2
38.4
39.6
49.0
40.3
34.9
33.7
33.5
127
-------
Table 45. Continued
Temperature
°F
95.0
104.0
107.8
109.4
113.0
118.4
127.4
132.8
140.0
4
19.5
12.2
11.1
11.2
14.0
11.3
9.4
8.6
7.8
Reactor Retention Time •
8
Feed
19.1
15.0
14.4
14.7
18.0
14.8
12.2
11.5
10.9
10 •
15
Slurry Concentration -
20.1
16.4
15.9
16.3
20.0
16.3
13.6
12.9
12.5
25.0
20.0
19.3
19.8
24.2
20.7
17.4
16.6
15.9
• Days
20
20% Solids
30.9
23.6
23.1
24.0
29.8
24.4
20.7
19.8
19.3
30
42.7
31.4
30.7
31.7
39.4
32.3
27.9
26.9
26.7
128
-------
TECHNICAL REPORT DATA
(Please read {attractions on the reverse before completing)
I REPORT NO.
EPA-670/2-74-016
2.
3. RECIPIENT'S ACCESSIOIWNO.
4 TITLE AND SUBTITLE
RECLAMATION OF ENERGY FROM ORGANIC WASTE
5 REPORT DATE
1974 - issuing date
6. PERFORMING ORGANIZATION CODE
7 AUTHOR(S)
John T. Pfeffer
8. PERFORMING ORGANIZATION REPORT NO
9 PERFORMING ORG tNIZATION NAME AND ADDRESS
Department of Civil Engineering
University of Illinois
Urbana, Illinois 61801
10. PROGRAM ELEMENT NO.
1DB2314
11. CONTRACT/GRANT NO
R-800766
12. SPONSORING AGENCY NAME AND ADDRESS
U.S. Environmental Protection Agency
National Environmental Research Center
Office of Research & Development
Cincinnati, Ohio 45268
13. TYPE OF REPORT AND PERIOD COVERED
Final
t4 SPONSORING AGENCY COOE
19. SUPPLEMENTARY NOTES
16 ABSTRACTThis study applied the anaerobic fermentation process to the production
of methane from the organic fraction of urban refuse. Shredded domestic refuse from
which the inorganic fraction was separated was used as a substrate. Raw sewage
sludge was added to the substrate in proportion to the rate at which it is produced
by a population producing a given quantity of refuse. The quantity and quality
of gas produced, the rate of gas production, the solids reduction, nutritional
requirements, and operating problems were evaluated in a laboratory system operating
at temperatures ranging from 35 C to 60 C. The results of the laboratory study
together with published data on both capital and operating costs of refuse shredding,
refuse separation, reactor volume, reactor mixing, reactor heating, and residue de-
watering were used to analyze the economics of the process.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATl Field/Croup
*Waste disposal, Waste, Organic chemistry,
*Gases, Temperature, Costs, Economic analy-
sis, Operating costs, Cellulose, *Fermenta-
tion, Fuels, Natural gas, Energy, Anaerobic
processes, *Methane, Sewage, Sludge, Sub-
strates, Nutritional requirements
*Solid waste disposal,
*Resource recovery, Cel-
lulose degradation,
*Urban solid wastes,
Shredded refuse, Solids
reduction, Reactor volume
Reactor mixing, Reactor
heating. Residue devater
13-B, 7-C
B DISTRIBUTION STATEMENT
ReJLease to public
19 SECURITY CLASS (ThisReport)
UNCLASSIFIED
21 NO OF PAGES
143
-129-
20. SECURITY CLASS (Thispage/
UNCLASSIFIED
22 PRICE
EPA Form 2220-1 (9-73)
OU.S.Government Printing Office- 1974 — 757-581/5307
Region 5-11
------- |