United States
Environmental Protection
Agency
Risk Reduction
Engineering Laboratory
Cincinnati, OH 45268
Research and Development
EPA/60QVS2-89/001 Aug. 1989
v°/EPA Project Summary
Comparative Evaluation of
Mesophilic and Thermophilic
Anaerobic Digestion
Irwin J. Kugelman and Vincent G. Guida
A study comparing anaerobic
digestion at mesophilic and
thermophilic temperature was
conducted. In the first phase of the
study, operation under temperature
transition in 750-mL lab-scale
digesters was studied. Systems
seeded with domestic sewage
sludge, but subsequently fed a
synthetic sludge, were operated at
20- and 30-day detention times at
35 °C. The temperature was raised to
55°C at rates varying from 0.258C to
2.5°C per day in duplicate, parallel
units. Regardless of temperature rise
rate, as soon as the temperature
exceeded 45°C methane production
was retarded. The units were held at
45 "C until recovery occurred. Once
recovery of methane production
ability occurred, transition to 55 "C
took place with little Incident except
for minor difficulty at 51 °C. Data
analysis indicated that rate of
temperature rise had little effect on
the total time required to obtain
stable operation at 55 "C; detention
time had a minor effect with longer
detention times yielding superior
results.
The organisms that function under
thermophilic conditions appear to be
present in mesophilic sludge but are
not active at low temperature. When
thermophilic conditions are brought
about, the thermophilic organisms
will multiply and reach an adequate
level in several weeks.
A temperature drop study was also
conducted during the first phase of
the project from 55°C to as low as
47.5°C. No adverse effect was
observed until the temperature was
reduced to less than 50«C. Washout
of methane bacteria seemed to occur
when the temperature was suddenly
dropped below 50 °C and the system
detention time was less than 20 days.
These data indicate that thermophilic
digester failure due to a loss of
thermal input should not be a
problem, unless the heat source is
not restored.
Comparison of operation at 55°C
vs 35 °C under steady state at
detention times ranging from 7.5 to
30 days with the chemically defined
feed Indicated that based on effluent
volatile acids level, mesophilic
operation was superior and that this
superiority was greater at the lower
detention times.
In the second phase, steady-state
operation was conducted in larger 75-
L digesters on a feed of raw primary
sludge. Long-term, steady-state
performance data were obtained at
49.50C and 35-C. Two hydraulic
detention times, 25 and 15 days, were
used. Attempts to operate at
temperatures above 50 °C resulted in
poor performance (propionic acid
levels well above 1,000 mg/L).
Operation of both the mesophilic
and thermophilic units was within
parameter ranges normally
considered satisfactory- Mesophilic
operation was slightly superior to
thermophilic operation as indicated
by total and volatile solids breakdown
and by gas production. Mesophilic
breakdown of carbohydrate and oil
and grease was superior.
Thermophilic breakdown of organic
-------
nitrogen was superior. Dewaterability
of sludge produced under mesophilic
conditions was significantly better
than sludge produced under
thermophilic conditions.
The results obtained here indicate
no advantage to operation of
anaerobic digestion at thermophilic
temperatures.
This Project Summary was
developed by EPA's Risk Reduction
Engineering Laboratory, Cincinnati,
OH, to announce key findings of the
research project that is fully
documented in two separate reports
of the same main title, one on each
phase of the project (see Project
Report ordering information on back).
Introduction
The purpose of this study was to
conduct a comparative evaluation of the
performance of anaerobic digestion
systems under different temperature
regimes. The temperature regimes
chosen were those most commonly used
in field installations (i.e., mesophilic 35°C
and thermophilic 50°C to 55°C).
Evaluation of performance was in terms
of a number of parameters including:
stability of operation, degree of waste
stabilization, dewaterability of the
digested sludge.
The work has been divided into two
phases. The first deals primarily with the
operation of anaerobic digestion systems
under situations of temperature transition.
The second deals with differences
between system performance at steady
state in the two temperature regimes.
The primary reason for conducting
temperature transition studies revolves
around the question of how to startup a
thermophilic digestion system. The most
efficient procedure is to seed with sludge
from an operational thermophilic system.
However, this material may not be readily
available. In such a case it will be
necessary to convert a mesophilic
anaerobic digestion system to a
thermophilic system. Because methane
forming microorganisms are known to be
quite sensitive to all environmental
conditions, it was felt that a rapid change
in temperature may not yield positive
results. On the other hand, if a very low
rate of adjustment of temperature is
used, the anaerobic treatment system
may be out of operation for a long period
of time. Consequently, an evaluation was
made here into the effect of temperature
rise rate when converting from
mesophilic to thermophilic operation.
Another question that was addressed in
this phase of the study was the effect of
temperature decreased on the
performance of thermophilic systems. It
is generally considered that thermophilic
systems are more sensitive than
mesophilic systems. Thus, small
reductions in temperature may have
severe consequences in thermophilic
system performance. This part of Phase I
of the study incorporated experiments in
which the temperature was intentionally
and rapidly reduced to determine the
magnitude of the effect on the
thermophilic anaerobic systems.
During Phase I, limited steady-state
data on system performance were
collected at 55 °C and 35 °C to compare
performance at these two temperatures.
In Phase I a chemically defined complex
substrate synthetic sludge was used.
Much more extensive steady-state data
were collected during Phase II in which
parallel steady-state operation was
conducted over periods of several
months on a feed of raw primary sludge.
Evaluation of performance in Phase II
was based on measurement of total gas
production, methane production, COD
destruction, grease destruction,
carbohydrate destruction, organic
nitrogen destruction, total and volatile
solids destruction and sludge
dewaterability. Two periods of steady-
state operation were intensively
monitored. One period lasted almost 6
months during which the hydraulic
detention time was maintained at 25
days. After a short transition period of 2
weeks at a 20-day detention time, a
second period of steady-state operation
at a 15-day hydraulic detention was
carried out for a 2-1/2-month period.
Procedures
Phase /
The major goal of this phase of the
study was to determine the effect of rate
of temperature change on system
performance during conversion from
mesophilic conditions to thermophilic
conditions. The basic plan used was to
set up a series of bench scale units at
35 °C, operate them to a steady state and
then raise the temperature at various
rates while maintaining as long as
possible a normal feeding pattern. Units
were operated at more than one
detention time to determine the effect of
this parameter (20 and 30 days).
A series of 750-mL bench scale
anaerobic digestion systems were set up
using digested sewage sludge from the
anaerobic digester at the Allentown, PA,
Sewage Treatment Plant. The units
operated on a batch feed and withdr
basis such that the HRT and SRT in
unit was identical.
A total of 20 units were set up, 1
each detention time used. In each si
10, units were operated in duplicate.
provided for four pairs of replicate i
at each detention time in which
temperature would be raised, plus
pair as controls. The control units v
maintained at 35°C. The four ratei
temperature rise used were 1°C, <
3.5°C, and 5°C per feeding period.
All of the units were set up i
temperature-controlled room maintai
at 35 ± 0.1 °C. The units in which
temperature was to be raised were
addition immersed in temperati
controlled water baths. Before actu
conducting the experiments detai
below, the necessary settings on
thermostats were manually calibral
Eight water baths were used; e.
contained a pair of bench scale anaerc
units.
As indicated above, these units w
originally started with digested slud
However, chemically defined comp
feed was used in the studies. The ft
chosen was a commercial prod
"Carnation Instant Breakfast"* suspenc
in whole milk. It was chosen becai
when mixed with whole milk in
recommended proportion (1 envelop
fluid oz of milk), it had a f
carbohydrate, and protein content sim
to that of raw sludge. In addition,
contained most of the organic a
inorganic nutrients required
microorganisms. The feed was,
addition, supplemented with seve
inorganic materials for which metha
fermenting organisms have a hi
demand (Fe, Co, Ni). Prior to starting t
study, the units were operated for seve
months to ensure washout of the origii
seed material.
After the washout period w
complete, the temperature change peril
was initiated. The volume of fe<
required for the detention times us<
corresponds to 25 mL/day at a 30-d<
HRT and 37.5 mL/day at a 20-day HRT.
was felt that these volumes were t<
small to be fed accurately. Thus durii
this study, the 30-day HRT units were ft
50 mL once every 2 days and the 20-d;
HRT units were fed 75 mL once every
days. It was indicated above that nomin
temperature changes per feeding perk
"Mention of trade names or commercial prodifl
does not constitute endorsement or recommendati
for use.
-------
ere to be used. Thus a feeding period
^presents a period of 2 days.
In conducting the temperature rise
study, the following procedure was used.
At the beginning of a feeding period, the
thermostat in the water bath of the
appropriate units was adjusted based on
the prior calibration and the normal
once/2-day withdrawal and feed took
place. At the end of the 2-day period, a
comparison was made between the gas
production of thep units whose
temperature was raised and that from the
control units (those which were
maintained at 35 8C). If gas production
was close to that of the control unit the
temperature was again raised and .the
unit was given a normal feeding. If gas
production was significantly less than in
the control unit, the temperature was not
raised and the units were not fed. The
units were not fed again until gas
production data indicated that most of the
previous feed had been converted to
methane and carbon dioxide. The
temperature was not raised again until
most or all of the feed added during the
last feeding was consumed in the
standard 2-day period. Each time a
withdrawal was made from a unit, the
digester mixed liquor was analyzed for
H, alkalinity, and volatile acids. Daily
as production was determined; and at
the end of each feeding period, gas
analysis for methane and carbon dioxide
was conducted. Alkalinity, pH, and
volatile acids were monitored by the
procedures given in Standard Methods;
gas analysis was conducted using a
Fisher Gas Partioner.
Over a 3-month time period, the
temperature of each unit except the
controls was raised to 55°C. At that time
it was decided to continue operation of
these units at 55°C and 35°C,
respectively, to establish a comparison
between operation at these temperatures
with this feed and at these two detention
times.
Subsequent to this steady-state
period, a study was conducted in which
the temperature was suddenly dropped
by various amounts while normal
operation, i.e., feed and withdrawal, were
maintained. This study ascertained the
effect that a sudden loss of heat supply
could have on the operation of a
thermophilic digester.
Only some of the 16 units being
maintained at 55 °C were used in the
temperature drop study. The remainder
was used in a study in which the
itention times of the 55 °C and the 35 °C
units were reduced to as low as 7.5 days.
The purpose of this study was again to
obtain a comparison between operation
of units in parallel at the two
temperatures but at much lower detention
time.
Thus, this study contained three
separate periods:
temperature rise rate study
temperature drop study
steady state at 55 °C and 35 °C at
various detention times.
Phase II
Raw primary sludge was periodically
collected from the Allentown, PA, Sewage
Treatment Plant. This is a typical
municipal treatment plant serving a large
metropolitan area. The sludges at this
plant are separated so that the primary
sludge contains little or no secondary
sludge. Sludge was collected as it was
being pumped from the primary tanks to
the sludge handling area.
Sludge collected once per week was
transported to Lehigh University
Environmental Studies Center and kept
under refrigeration at 4°C until used.
Upon being brought to the laboratory, the
sludge was sampled for total solids
analysis. It was decided early in the
study to maintain a constant total solids
concentration in the feed. The original
target was 4%, but after 1 month of
sludge feed to the units this value was
changed to 3.5%.
Two identical anaerobic reactors were
used in this study. Each reactor was
fabricated of plexiglass in the shape of a
rectangular parallelepiped. A false bottom
sloped toward the outlet pipe eliminating
dead spaces. The total volume of each
reactor was 75 L At 25-day detention
time, the liquid volume was 50 L; at 15-
day detention time, the volume was 45 L.
Gas recirculation was used to keep the
reactors mixed. Cumulative gas was
recorded on a wet tip gas meter. Each
reactor was kept in a walk-in
temperature-controlled room (± 0.1 °C).
Feed of raw sludge and withdrawal of
digested sludge was conducted once per
day on a batch basis.
Periodically, digested sludge removed
from each unit was analysed for pH,
alkalinity, volatile acids, ammonia-N.
organic-N, grease, COD, carbohydrate,
and total and volatile solids, and capillary
suction time (CST). Analysis was started
within an hour after sample withdrawal,
except for sludge dewatering tests that
were carried out on sludge cooled to
25 °C. All sludge samples, except those
portions used for dewatering tests, were
ground in a Waring blender to reduce
errors from lumps when making dilutions.
Gas samples were collected in 100-mL
glass sampling bulbs connected to a
reservoir containing acid-salt solution.
The reactors were initially filled with
digested mesophilic sludge from the
Allentown, PA, Sewage Treatment Plant.
The thermophilic unit was seeded with
sludge saved from Phase I and sludge
from a thermophilic digester in New York
City. Feed was first Carnation Instant
Breakfast (which was used in the Phase I
study), then glucose and whole milk, and
finally raw sludge. Operation of the
thermophilic unit was somewhat erratic
until 25 mg/L of yeast extract was added.
Subsequent to this addition, thermophilic
operation was satisfactory. It was found
necessary to add 25 mg/L of yeast
extract to the thermophilic unit to ensure
good operation. To ensure identical
treatment, this quantity of yeast extract
was added to the mesophilic unit as well;
although it was not needed. When raw
sludge feed was started, the mesophilic
unit temperature was 35 °C and the
thermophilic unit was slightly above
50 °C.
Results and Discussion
Phase I
Temperature Rise Rate
The data collected during the
temperature rise rate study are much too
extensive to be presented here; reference
to the full report is recommended. It was
found, however, that regardless of
temperature rise rate and/or detention
time, a similar pattern of results was
obtained. System performance was
satisfactory until the temperature was
raised above 45"C, at which time severe
retardation took place. The retardation
was primarily in the methane
fermentation but some inhibition of acid
formation was observed. After a period of
dormancy, gas production started and
eventually reached normal levels. During
this dormant period, feeding was
restricted to prevent pH failure. After a
normal level of gas production was
achieved, the temperature increase was
begun again. The temperature was raised
to 55 °C with little difficulty except for a
minor discontinuity between 50°C and
52 °C.
The major question explored in this
part of the research effort was what is the
best procedure for converting a
mesophilic anaerobic digester to a
thermophilic unit. Table 1 presents a
summary of pertinent data that can be
-------
used to address this question. For each
temperature rise rate and detention time,
the following data are presented: time in
days to first reach 55°C, time in days to
achieve stable operation at 55°C, time in
days to the first temperature inhibition,
time in days to be able to return to a
normal 2-day feed pattern after inhibition,
time in days to the next temperature
change after the first temperature
inhibition, and the number of periods
during which feeding took place until
55°C was achieved. The figures
presented in the table except for the first
and last columns represent days after the
first change from 35°C. The following
major points can be deduced from this
table and other data in the full report.
1. The slower the temperature rise, the
longer it takes to reach 55 °C.
2. The rate of temperature rise, has
only a minor effect on the time to
reach stability at 55°C.
3. Regardless of the rate of
temperature rise, a major retardation
occurs whenever the temperature
exceeds 45 °C.
4. The effect of this retardation is lower
at low temperature rise rates.
5. At high temperature rise rates, once
acclimation to the thermophilic
conditions occurs, few problems are
manifest.
6. At low temperature rise rates, small
periods of acclimation are needed as
the unit proceeds from 45 "C to
55 °C.
7. At low temperature rise rates, the
unit can be fed a high percentage of
time during the transition procedure.
8. At high temperature rise rates, the
unit can be fed only a small
percentage of time during the
transition period.
9. There seems to be a zone of minor
retardation in the 50 °C to 52 °C
range.
10. Temperature changes adversely
affect the methane bacteria to a
much greater extent than the acid
formers.
11. Temperature effects are not
instantaneous. During the first feed
period after 45°C is exceeded,
almost normal operation occurs.
Retardation is manifest during the
next feeding period.
12. Temperature effects are magnified
at lower detention time.
Overall, it does not seem that the rate
of temperature rise greatly affects the
total time for conversion to thermophilic
conditions. A decision on the rate to use
will depend on another major factor, i.e.,
what can be done with the raw sludge
during the conversion process. If there is
no alternate sludge-handling procedure,
the slow rate of rise should be used, as
for the most part, feeding of the digester
can continue. If an alternate sludge-
handling procedure exists, the rapid rise
method would achieve operation at 55 °C
in a somewhat shorter period of time.
Since operation is satisfactory between
35°C and 45°C, perhaps a combination
procedure would be best: a rapid rise to
45 °C followed by a slow rise to 55 °C.
As previously indicated, once all of the
units reached 55 °C, it was decided to
maintain them at that temperature to use
in the temperature drop study or for
gathering steady-state operational data at
thermophilic temperatures vs mesophilic
temperatures. Consequently, the 16 units
were maintained at 55 °C, half at 30-day
detention time and half at 20-day
detention time for the next 3 months. This
ensured that prior to any other studies
using these units that they had been
through at least three detention times of
operation at 55°C. At the end of this
period some of the units were used in a
temperature drop study which is detailed
below. The detention time in the other
units was reduced to 15, 10, and 7.5
days. Simultaneously the temperatures in
the 35 °C units was lowered to the same
detention times. This allowed for
comparison of operation of units
operating over a range of detection times
at the two different temperatures.
Temperature Drop Study
To determine the effect of a
temperature drop on the thermophilic
system, the temperature was dropped
from 55°C to the new setpoint over night
and then maintained at the new setpoint.
The normal feed and withdrawal pattern
was maintained. The performance of
these systems was then monitored for
the next 25 days. Units with detention
times of both 20 and 30 days were used.
Temperatures were reduced to 52.5°C,
50°C, and 47.5°C. Controls were run with
units maintained at 55 °C and 35 °C.
The results at 30-day detention time
were that effluent volatile acids remained
low in all units and were not
distinguishable from the control values.
There was some early tendency for the
volatile acids to increase in the unit
whose temperature was reduced to
47.5°C. However, this trend was soon
reversed. The data for the 20-day
detention time units exhibited a similar
situation, except that both the 50°C and
47.5 °C units exhibited an early tendency
for volatile acids to rise. The trend
reversed for the 50 °C unit, but not fc
47.5°C unit. Figure 1 is a plot of the
from the 20- and 30-day detention
units in which the temperature
dropped to 47.5°C. This plot ch
shows the significant difference
resulted. The pattern of rise in vo
acids exhibited by the 20-day detei
time unit is typical of a wash
phenomenon. That is, the tempers
reduction does not appear to intei
with the microorganisms ability
degrade the substrate, but for si
reason their ability to reproduc<
impaired. It is possible that the 30
detention time unit would also go in
failure mode if the run had b
continued for a longer period of time.
Despite the problem when
temperature was dropped to 47.£
these data indicate that operatic
problems from failure of the hea
system of a thermophilic digester shi
not be a major concern. The only adv<
situation which was manifest took sev
weeks to develop. It is unlikely that
heating system of a digester could no1
repaired in a few days. In addition,
rate of temperature drop utilized h
was extreme compared to what we
occur in the field. Field digestion syste
are massive tanks with high h
capacity. It is unlikely that the diges
temperature would drop by more tl
1 °C per day even if a complete failure
the heating system occurred in
middle of the winter.
Comparison of Mesophilic and
Thermophilic Operation
Table 2 presents a summary of
operation of the units referred to above
detention times ranging from 30 days
7.5 days. Data presented are efflui
volatile acids averaged over a minimi
of three detention times of operation
can be seen that the performance
similar at the longer detention times I
that at shorter detention times t
mesophilic systems were clearly super
to the thermophilic units. It has oft
been suggested that operation
thermophilic temperatures will provi
better breakdown of organics ai
consequently, more methane. The d<
here do not support this suppositic
Performance of the mesophilic units w
always superior to that of tl
thermophilic units.
Phase // (
The Phase I data had indicated th
mesophilic operation seemed better thi
-------
Table 1. Summary of Performance During Temperature Transition from Mesophilic to Thermophilic Conditions
Temp. Change
Detection Time
5.0-30
5.0-20
3.5-30
3.5-20
2.0-30
2.0-20
1.0-30
1 .0-20
Time to
Reach
55°C
52
56
54
48
54
60
64
74
Time to
Reach
Stability
at 55°C
52
72
54
56
62
68
64
74
Time to
First Major
Inhibition
6
6
10
8
14
12
24
22
Time to
Resume 2 -day
Feed Pattern
38
48
44
34
26
48
36
48
Time Till
Next Temp.
Change
52
56
48
46
42
52
42
50
No. of
Feeding
Periods
to55°C
13
13
13
14
23
21
30
27
thermophilic operation with a chemically
defined complex substrate. In Phase II,
reactors were fed sludge from an actual
treatment plant to determine if any
difference would be seen. For a 6-month
period, the reactors were operated at a
25-day detention time. No change in the
operation of the mesophilic digester was
necessary during that period. However, a
significant change in the operation of the
thermophilic digester was made after 3-
1/2 months of operation. At that time, the
temperature was reduced from slightly
above 50°C to slightly below 50°C. It
was found that with the temperature
above 50°C, the volatile acids were high
(1,500 mg/L, 90% propionic); but when
the temperature was dropped to 49.5°C,
the volatile acids level was similar to that
in the 35°C unit (200 mg/L). Gas
production and COD data were in
accordance with the high levels of volatile
acids in the thermophilic unit; that is,
when the thermophilic temperature was
> 50 °C, gas production was lower and
COD was higher in the thermophilic vs
the mesophilic unit.
Several times the temperature was
cycled between 50.5°C and 49.5°C.
Each time the rise in temperature
retarded the thermophilic unit and the
temperature reduction yielded better
performance.
The operation at a 25-day detention
time was maintained for 2-1/2 months
after the temperature reduction. At that
time the detention time was reduced to
20 days. After 2 weeks the detention time
was converted to 15 days, which was
maintained for the next 2-1/2 months.
The results are discussed below in terms
of the parameters measured.
pH, Alkalinity, Volatile Acids, and
Gas Production
Raw sludge pH ranged from 5.4 to 5.8,
alkalinity from 600 mg/L to 2,100 mg/L,
and volatile acids from 1,000 mg/L to
3,000 mg/L, which are typical values.
Thermophilic sludge pH was 7.3 to 7.5,
while mesophilic sludge pH was 7.0 to
7.2. Thermophilic alkalinity was also
slightly higher than mesophilic alkalinity,
4,300 to 5,100 mg/L vs 3,800 to 4,600
mg/L. When the thermophilic temperature
was greater than 50°C (discussed
above), thermophilic sludge volatile acids
were high; however, they were almost
always <300 mg/L once the temperature
was below 50°C. During the conversion
from 25-day detention time to the 15-day
detention time, thermophilic volatile acids
rose to 550 mg/L for a few days but soon
dropped again. The mesophilic volatile
acids were always in the range of 100 to
200 mg/L. Gas production was always
similiar in the two units, with mesophilic
production only slightly higher (0% to
5%) than thermophilic production.
Methane fraction in each unit averaged
close to 60%, ranging from 57.5% to
64.5%. The operation of both units based
on these traditional parameters was
normal except for the period when the
thermophilic temperature was above
50 °C.
Total and Volatile Solids -
Digested Sludge
Total and volatile solids were always
slightly higher in the thermophilic unit
than the mesophilic unit. The differences
were most pronounced during the period
when the thermophilic unit was at a
temperature above 50°C. Indeed, near
the end of each steady-state period,
solids levels in both units were almost the
same. Overall volatile solids destruction
was higher in the mesophilic unit than the
thermophilic unit.
COD
Prior to lowering the thermophilic
temperature below 50 °C, the total COD
of the thermophilic unit was consistently
higher than that of the mesophilic unit.
After the temperature reduction (at 25-
day detention time), the COD of both
units was similar. After reduction of the
detention time, the COD of the
thermophilic sludge was initially higher
than that of the mesophilic sludge. Much
of this difference was eliminated by the
end of the 15-day detention time
operation.
The soluble COD data indicate that the
thermophilic unit had a soluble COD from
1,400 to 1,700 mg/L higher than in the
mesophilic unit. The difference was
greater when the thermophilic unit was
operated at temperatures above 50°C.
Even under the best of thermophilic
operation, a defined difference in soluble
COD was observed.
Nitrogen
The ammonia-N was always higher
and the organic nitrogen was always
lower in the thermophilic unit. Since the
TKN of both units was the same, these
data indicate superior organic nitrogen
conversion to ammonia-N in the
thermophilic unit. The difference was
higher at 25-day detention time then at
15-day detention time. As indicated
previously, because of the higher degree
of conversion of organic nitrogen to
-------
to
BJ
800
700
600
500
^ 400
o
V)
.•o 300
o
200
100
20-Day Detention Time
30-Day Detention Time
\
I I I I t I I I
8 12 16 20
Time in Days From Temperature Drop
24
Figure 1. Comparison of mean volatile acids concentration in units at 20- and 30-day
detention times subjected to temperature drops from 55°C to 47.5 "C.
Table 2. Steady State Performance
of Thermophilic and
Mesophilic Anaerobic
Treatment Systems
HRT
Effluent Volatile Acids, mg/L
S5°C
35'C
30
20
15
10
7.5
100- 150
150- 250
200- 300
300- 600
1000-1200
50-100
75-100
100-150
100-200
200-300
ammonia-N, the pH and alkalinity were
always higher in the thermophilic than in
the mesophilic unit.
Oil and Grease
The reduction of oil and grease was
higher in the mesophilic unit than in the
thermophilic unit. This difference was
especially pronounced when the
thermophilic unit was operated above
50°C, because volatile acids are
measured as grease and oil in this
particular analytical test. There was little
change in the difference between the
units that could be ascribed to the
reduction of the detention time to 15
days.
Carbohydrate
Sludge carbohydrate measurements
were only conducted during the 15-day
detention time operation. Total
carbohydrate reduction was higher in the
mesophilic unit than in the thermophilic
unit. The difference ranged from 500 to
1,200 mg/L.
Sludge Dewaterability
Throughout most of the study,
measurements were made on the
dewaterability of the effluent from each
unit. Preliminary tests were performed,
using three methods of dewatering: the
-------
CST test, the Buchner funnel filtration
test, and a batch centrifugation test. It
was found that the latter two tests were
very difficult to run unless the sludge was
conditioned with a coagulant. The CST
test, however, gave reasonable
measurements with and without the
addition of coagulants. Thus, the CST
test was used to characterize the
difference between both types of sludge.
The results indicate that the thermophilic
sludge was more difficult to dewater in an
unconditioned state than the mesophilic
sludge. CST for mesophilic sludge
(unconditioned) was 350 to 450 sec,
while values for unconditioned
thermophilic sludge were 500 to 800 sec.
Visual observation of the sludge clearly
indicated better separation under gravity
conditions for the mesophilic sludge. In
addition, the sludge supernatant was
more turbid for the thermophilic sludge,
indicating that the size of digested sludge
particles in the thermophilic sludge was
smaller than in the mesophilic sludge.
When the sludge was subjected to
centrifugation without conditioners
present, the thermophilic centrate was
more turbid than that from the mesophilic
sludge, although the depth of the solids
pool was almost the same for both
sludges.
In addition to the unconditioned tests,
some tests were conducted in which
sludge was conditioned with ferric
chloride and/or lime. Ferric chloride
conditioning had a significant affect on
the CST values of both sludges. The
addition of lime to a sludge already
conditioned with ferric chloride had little
effect as did the use of lime alone. The
dose of ferric chloride required to reduce
the CST to below 50 sec for mesophilic
sludge was 20% less than that required
for thermophilic sludge.
A summary of the steady-state results
in terms of volatile solids destruction,
COD destruction (direct measurement
and methane production), organic
nitrogen destruction, grease and oil
destruction, and carbohydrate destruction
is given in Table 3. Somewhat better
performance was achieved at both
detention times by the mesophilic unit
except for nitrogen breakdown.
Overall Conclusions
1. Rate of temperature change had
little effect on the total time required
to reach stable operation at 55 °C
during conversion from mesophilic
to thermophilic conditions.
2. The transition from mesophilic to
thermophilic conditions seems to
occur at 45°C, as operation was
interrupted (methane production
retarded) for one to several weeks
when this temperature was reached.
3. Eventual recovery occurs indicating
that thermophilic organisms exist in
mesophilic sludge but are dormant
at low temperature.
4. At longer detention time, operation
temperature transition effects are not
as severe.
5. When the temperature is rapidly
reduced from 55°C, no effect on
operation is manifest until the
temperature is reduced below 50°C
and the detention time below is 20
days.
6. The temperature drop effect is not a
reduction in the ability of the
methane bacteria to process
substrate but in the ability to
reproduce.
7. Based on effluent volatile acids,
steady-state operation at 35 °C was
slightly superior to operation at 55 °C
at long detention times and clearly
superior at short detention times,
with a chemically defined complex
substrate as the feed.
8. With a raw sludge substrate, steady-
state operation above 50 °C was
characterized by poor performance.
Volatile acids, especially propionic
acid, were above 1,000 mg/L.
Breakdown of various sludge
components was less than under
mesophilic conditions. Conse-
quently, long-term steady-state data
collection was obtained at 49.5 °C in
the thermophilic region.
9. In terms of pH, alkalinity, volatile
acids, and methane production the
long-term steady-state performance
at 49.5°C and 35°C was satisfactory
at 25-and 15-day HRT.
10. Under all conditions the performance
of the mesophilic system was
slightly superior to the thermophilic
system.
11. At detention times of 15 and 25
days, significantly higher breakdown
of carbohydrate and oil and grease
were achieved in the mesophilic
unit.
12. At both those detention times,
significantly higher breakdown of
organic nitrogen occurred under
thermophilic conditions.
13. At both those detention times,
slightly higher destruction of total
and volatile solids and COD
occurred under mesophilic
conditions.
14. Performance of both systems was
also better, in terms of breakdown of
raw sludge components, at 25-day
detention time.
15. The soluble COD of the thermophilic
sludge was always at least 1,000
mg/L higher than for the mesophilic
sludge.
16. At both detention times with the 50-L
digesters, sludge dewaterability was
significantly better, as measured by
the CST test, under mesophilic
conditions.
17. Sludge dewaterability for both
temperature systems could be
significantly improved by
conditioning with ferric chloride.
Higher doses were required for the
thermophilic sludge.
18. Lime, both alone and with ferric
chloride, had little effect on sludge
dewaterability.
19. Detention time had little effect on
sludge dewaterability.
20. Thermophilic sludge odor was more
disagreeable than that from
mesophilic sludge, even when
volatile acids were low.
21. Thermophilic operation at
temperatures above 50 °C was better
with chemically defined substrate
than with raw primary sludge, but
mesophilic operation was better than
thermophilic operation regardless of
feed.
22. These data indicate that operation of
anaerobic digestion at thermophilic
conditions has no advantage over
operation at mesophilic conditions.
A report on each phase was
submitted in fulfillment of Cooperative
Agreement CR811022 by Lehigh
University under the sponsorship of the
U.S. Environmental Protection Agency.
-------
Table 3. Summary of Performance of Mesophilic and Thermophilic Digesters
% Removal
Detention
Time
25
25
15
15
Unit
Thermo.
Meso.
Thermo.
Meso.
Volatile
Solids
51.7
57.2
44.7
47.0
COD
D.M.
52.6
52.9
44.2
49.9
COD
Gas
53.2
63.1
53.3
56.8
Organic
Nitrogen
59.4
50.2
44.9
27.2
Grease
&Oil
65.0
71.2
59.3
67.4
Carbohydrate
-
-
56.2
68.8
Irwin J. Kugelman and Vincent G. Guida are with Lehigh University, Bethlehem,
PA 18015.
Harry E. Bostian is the EPA Project Officer (see below).
The complete report consists of two volumes, entitled "Comparative Evaluation
of Mesophilic and Thermophilic Anaerobic Digestion:"
Phase I. Temperature Transition Studies," (Order No. PB 89-151 3931 AS; Cost:
$15.95, subject to change).
Phase II. Steady State Studies," (Order No. PB 89-151 4011 AS; Cost: $15.95,
subject to change).
The above reports will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Risk Reduction Engineering Laboratory
U.S. Environmental Protection Agency
Cincinnati, OH 45268
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
BULK RATE
POSTAGE & FEES PAID
EPA
PERMIT No. G-35
Official Business
Penalty for Private Use $300
EPA/600/S2-89/001
------- |