United States
Environmental Protection
Agency
Risk Reduction
Engineering Laboratory
Cincinnati, OH 45268
Research and Development
EPA/600/S2-90/037 Sept. 1990
&EPA Project Summary
Effect of Recycling
Thermophilic Sludge on the
Activated Sludge Process
T.B.S. Prakasam, S. Soszynski, D.R. Zenz, C. Lue-Hing, L. Blyth, and G.
Sernel
A full-scale investigation was
undertaken at the Hanover Park Water
Reclamation Plant (WRP) operated by
the Metropolitan Water Reclamation
District of Greater Chicago to study
whether the net sludge production
from the WRP can be reduced by
implementing a scheme reported by
Torpey et al. This scheme involved
the recycling of sludge withdrawn
from a thermophilic digester into the
aeration tanks of an activated sludge
system.
The Hanover Park WRP, which has a
design flow capacity of 45,400 m3 per
day (12 mgd), was split into a control
section and an experimental section.
The control and experimental sec-
tions were operated as nitrifying ac-
tivated sludge systems, except that
thermophilic drawoff was recycled in-
to the aeration tanks of the experi-
mental section from a digester
system, which consisted of meso-
philic and thermophilic digesters
operated in series.
In contrast to the results reported
by Torpey et al., a significant reduc-
tion in the net sludge production was
not observed as a consequence of
recycling thermophilic sludge into
the aeration tanks of the Hanover
Park WRP.
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 a separate report of
the same title (see Project Report
ordering information at back).
Introduction
In 1987, the Metropolitan Water
Reclamation District of Greater Chicago
(herein called "the District") disposed of
or used 251,000 tonnes (277,000 tons) of
dry sludge solids produced at its seven
sewage treatment facilities. Costs of
treatment and disposal of sludge were a
major fraction of total operating costs.
Consequently, the District has a
continuing interest in investigating
processing approaches that reduce
volume and mass of sludge produced in
wastewater treatment.
A novel approach for reducing sludge
production has been reported by Torpey
and coworkers(l) and demonstrated at
the Rockaway Sewage Treatment Plant in
New York City. In this process, sludge
from a mesophilic digester is processed
through a thermophilic digester and then
a portion of the thermophilic digester
drawoff is recycled through the aeration
tanks of an activated sludge system. The
remainder of the thermophilic drawoff is
passed through a rethickening and
elutriation tank. The authors reported that
the volume and quantity of the sludge,
when compared with that produced by
the activated sludge plant using meso-
philic digestion with no recycle, was
reduced by about 60%. The reduction in
sludge quantity was attributed to
additional destruction of solids achieved
in the thermophilic digester and by the
oxidation of some of the solids contained
in the thermophilic sludge recycled to the
-------
aeration tanks of an activated sludge
system.
Torpey's findings have been supported
by further work by New York City's
Department of Environmental Protection.
Carrio and coworkers (2) reported in 1985
that recirculation of digested sludge to
the activated sludge aerators reduced
total sludge production at the Newton
Creek, Bowery Bay, 26th Ward, and
Oakwood Beach wastewater treatment
plants.
In view of the potential advantage of
substantially reducing sludge production,
the District undertook a demonstration at
its Hanover Park WRP of Torpey's con-
cepts; that is, in a plant using the
activated sludge process, digest the
sludge first by mesophilic and then by
thermophilic digestion, and then divide
the digested sludge into two portions,
one for discharge and one for recycle to
the activated sludge aeration. The
Hanover Park WRP was selected for the
study because it was easily divided into
two nearly identical modules, which
would provide a control section for com-
parison with the experimental section.
The following were the main objectives
of the study conducted at the Hanover.
Park WRP:
1.To demonstrate on a full-scale
whether retrofitting an existing
activated sludge system with a
mesophilic-thermophilic (M-T)
digestion scheme and recycling the
drawoff from the thermophilic
digester into the aeration tanks of an
activated sludge system would
reduce the volume and quantity of
volatile solids produced when
compared with the volume and
quantity of volatile solids produced
from a conventional activated sludge
system using mesophilic digestion
and no recycling of digested sludge.
2. To evaluate the quality of the
effluents obtained under the various
phases of the M-T digestion scheme
and compare these with the results
obtained under the conventional
mode of operating an activated
sludge system.
3.To determine the dewatering
characteristics of the sludges
resulting from the recycling of
thermophilic digester drawoff and
compare them v/ith those of the
drawoff derived from a conventional
mesophilic digester.
4. To evaluate the odor characteristics
of the sludge derived from the M-T
digester sequence and compare
them with those of the mesophilic
digester drawoff.
5. To compare the removal of heavy
metals observed in a conventional
activated I sludge system with the
removal observed using the recycle
of the slucJge from the M-T digestion
scheme. I
Procedures
The Hanover Park WRP with a nominal
dry weather capacity of 45,400 m3 per
day (12 MGJD), employs primary
treatment, secondary treatment by the
activated sludge process, followed by
sand bed filtration. The sludge is
anaerobically digested, pumped to
lagoon storage, and ultimately used on
an adjacent farm. A schematic diagram of
the plant is shown in Figure 1.
Wastewater | enters the treatment plant
through coarse bar screens, and flows to
primary settling tanks. There are two
settling tanks'for each of four aeration
batteries (A,B,C, and D). Two aeration
tanks comprise a battery. Primary influent
flows are 3,780 m3 per day (1.0 MGD) for
each aerator pf Battery A, 7.57 m3 per
day (2.0 MGD) for each aerator of Battery
B, and 5.68 m3 per day (1.5 MGD) for
each aerator of Batteries C and D. There
is one clarifier for each of the eight
aerators. Excess activated sludge is
wasted to the primary tanks. After the
secondary effluent from the clarifiers is
chlorinated with sodium hypochlorite it
then flows to rapid sand filters. Sludge is
digested in [four circular digesters
equipped with floating covers. The
digesters normally are maintained
between 29 and 35°C and are contin-
uously mixed. The digesters are 12.2 m
(40 ft) in diameter with a variable depth of
between 4.4 m (14.5 ft) and 5.9 m (19.5
ft). Digested sludge is pumped to storage
basins and is [applied to the sludge farm
in the summer.
During the [experimental program the
plant was divided into two similar
modules. Batteries A and B, the control
section, were operated in the
conventional activated sludge mode with
mesophilic digestion of the mixed
primary and waste activated sludge
withdrawn frorn the primary tanks. The
sludge was digested at 35°C in Digesters
1 and 2 operated in series. Both
digesters wer^ heated and continuously
mixed. |
The experimental section consisted of
Batteries C and D and Digesters 3 and 4.
Digester 3 received the mixed sludge
from the primary tanks and was operated
at 35°C. The sludge from Digester 3 was
fed in series jb Digester 4 where it was
digested at 53°C.
The incoming wastewater flow was
divided between control and experimental
sections to give similar hydraulic
retention times in the activated sludge
aerators. After appropriate time periods to
ensure steady state, increasing portions
of the thermophilic sludge from the
experimental section were recirculated to
the aerators to study the effects on the
activated sludge process and sludge
production. There were five operating
phases: phase 3—no re-circulation, phase
3a-10% recirculation, phase 3b--30%
recirculation, phase 3c~40% recircula-
tion, and phase 3d--50% recirculation.
Analyses were conducted by standard
methods. An appropriate quality assur-
ance program was carried out to ensure
accuracy of all analytical measurements.
Results
The entire experimental program
including modification to the plant eind
cleaning of digester took almost 4 yr. The
program was divided into three main
phases. In Phase 1, the plant was divided
into the control and experimental
sections, the thermophilic digester was
started up, and steady state operation
was achieved. In Phase 2, background
data on performance of the control and
experimental sections were obtained.
Phase 3, the experimental phase of the
program, was subdivided into Phases 3a
to 3d. In these phases, recirculation of
thermophilic sludge withdrawn from
Digester 4 and recycled to aeration
Batteries C and D was increased in steps:
zero in Phase 2 to 10%, 30%, 40%, and
50% of its output.
Thermophilic Digester Operation
The thermophilic digester (Digester 4~
the second digester in the experimental
section) was operated successfully at
53°C. Figure 2 shows the period of
startup and initial operation. Instability
occurred at about 55°C. Although
operation may have been possible at
55°C, temperature was reduced to 53°C
to reduce system sensitivity. Except for
occasional volatile acid excursions (see
Figure 2), operation at 53°C was
uneventful. During the data collection
period that followed, volatile acids leaving
Digester 4 ranged from 89 to 233 mg/L.
Volatile acids from the second mesophilic
digester in the control section (Digester
2) ranged from 18 to 86 mg/L.
The processing of approximately half of
the sludge output of the plant through the
thermophilic digester as a final digestion
step created problems at the sludge
farm. Numerous complaints about
unpleasant odors were received from
-------
I
NWIS-20A
f
42"RCP
to
ox. pond
to
ox. pond
to lagoons
or farm
Notation:
NWIS 20, 20A: Northwest interceptor sewers; W.W.: wet well; S. BOX: splitter box; P.S.: primary sludge; WAS: waste activated
sludge; R.S.: returned sludge; ML: mixed liquor; P.E: primary effluent; S.£: secondary effluent; OX. POND: oxidation pond
Figure 1. - Schematic of Hanover Park Water Reclamation Plant.
-------
Figure 2. Volatile acids and temperature profile of digester 4.
persons living near the sludge farm
during the experimental program as
contrasted to practically no complaints
when only mesophilically digested sludge
was being discharged. The odor of
thermophilically digested sludge evi-
dently was either more offensive or more
identifiable than the odor of
mesophilically digested sludge.
Fate of Volatile Solids
About 2% to 5% more volatile solids
were destroyed in the experimental
section digesters than in the control
section's mesophilic digesters when
thermophilic sludge drawoff was recycled
(up to 40% of the volume withdrawn) into
the aeration tanks of the experimental
section. When the thermophilic sludge
drawoff recycle rate to the aeration tanks
was increased to about 50% of the
volume withdrawn, the volatile solids
destruction in the experimental section
digesters was found to be lower by about
12.5% than that observed in the control
section digesters. At this high
recirculation, the experimental section
digesters were overloaded. Total resi-
dence times in control and experimental
section digesters were respectively 21.5
and 8.8 days at 50% recycle. A volatile
suspended solids (VSS) balance around
the aerator-clarjifier revealed a net VSS
production for both control and
experimental systems. Scatter was too
great to allow [a quantitative conclusion
about the influence of recirculation of
digested solids Ion VSS production.
When masses of VSS in the waste
activated sludge (WAS) streams from
control Batteryj B were compared with
those from experimental Batteries C and
D, recirculatiori of up to 40% of the
meso-thermophilically digested sludge
did not increase the mass of WAS per
unit volume of wastewater treated.
The difference in the total digested
sludge ultimately disposed from the
control and experimental sections is not
statistically significant. Table 1 shows the
magnitude of the differences for Phase 2
and Phases 3 a-c. Recirculation of up to
30% of the thermophilic sludge produced
a negligible [effect on total sludge
disposed per u|nit volume of wastewater.
At 40% recirculation, the experimental
section produced 13% less sludge, which
was not statistically significant. At 50%
recirculation (npt shown in Table 1), the
experimental unit produced more sludge
than the contrbl. This result is atypical
because the experimental digesters were
overloaded at this high recycle.
Effluent Quality and Plant
Performance
The overall performance of the control
and experimental sections of the Hanover
Park WRP was comparable during all the
phases of the study. The recycling of
thermophilic sludge drawoff, even at the
rate of about 50% of the volume
withdrawn from the digester into the
aeration tanks of the experimental
section, did not have an adverse effect on
the secondary effluent quality.
The overall BOD removals of the
control and experimental sections were
comparable and were in the range of
91.7% to 97.2% and 94.2% to 97.7%,
respectively, during the various phases of
the study. The corresponding secondary
effluent BOD values for these sections
were in the range of 3 to 9 and 3 to 6
mg/L, respectively. Recycling of up to
50% thermophilic sludge had no
noticeable effect on BOD removal.
The overall suspended solids removal
ranged from 83.9% to 94.2% for the
control section and from 89.4% to 94.7%
for the experimental section. The
secondary effluent suspended solids
concentration in these sections were
comparable and were in the ranges of 6
to 15 mg/L (control) and 5 to 10 mg/L
-------
Table 1.
Average Volatile Solids Production Sent to Farm During the Various Experimental Phases
Phase
(Recirculation)
2(0%)
3a(10%)
3b(30%)
3c(40%)
Sludge Production (Ib/mg)*
Control (Q
372
340
274
288
Experimental
(E)
364
334
265
250
Difference
C-E
8
6
9
38
(C-E)(100)
C
2.2%
1.8%
3.3%
13.1%
* Ib/mg XO.J2 =
^ or mglL
(experimental). Again, recycling of
digested sludge had no noticeable effect
on percent removal.
The percent removal of NH4-N was
within the range of 91.0 to 98.6 percent in
the control and experimental sections.
That the NH4-N concentration of the
secondary effluent was less than 1.2
mg/L in both of these sections indicated
that the recycling of thermophilic sludge
did not have an adverse effect on
nitrification.
The recycling of thermophilic sludge
drawoff into the aeration tanks of the
experimental section did not have a
discernible influence on the sludge
volume index (SVI) values of the mixed
liquors. However, there was too much
scatter in SVI values for any firm
conclusions on the effect of recycling.
A comparison of the mixed liquor
dissolved oxygen concentration and
oxygen uptake rate data for the control
and experimental section aeration tanks
revealed no significant trends. Scatter in
the results prevented drawing any firm
conclusions on the effect of recycling on
these variables.
The recycling of thermophilic sludge
into the aeration tanks of the
experimental section did not seem to
alter the microbiological profile of the
mixed liquor when compared with that of
the mixed liquor of the control section
aeration tanks.
The soluble concentration of Zn, Cd,
Cu, Cr, Ni, Pb, and Hg remained
approximately the same in the digester
feeds and draws; this indicated that no
significant solubilization of these metals
occurred during mesophilic or thermo-
philic digestion. However, the soluble
concentration of iron in the digester feed
of the control and experimental sections
decreased significantly as it underwent
digestion in the primary digester. No
further resolubilization of this metal
occurred in the secondary digesters.
Dewaterability
Capillary suction time (CST) measure-
ments at various -polymer dosages
indicated that mesophilic sludge required
a lower polymer dosage than did the
thermophilic sludge (10 vs. 22.5 kg/dry
tonne) to achieve the minimum CST that
was possible. The thermophilic sludge,
however, exhibited a higher floe strength
than did the mesophilic sludge.
Pilot scale centrifuge studies confirmed
that the thermophilic sludge required a
higher polymer dosage than did the
mesophilic sludge. At optimum polymer
dosages, these studies also indicated that
the mesophilic sludge approached 100%
solids capture whereas the thermophilic
sludge approached a maximum of 96%
solids capture. The lower solids capture
with thermophilic sludge probably
resulted from the higher concentration of
fine particles in it than in the mesophilic
sludge.
The percent cake solids achieved in
pilot-scale centrifuge studies appeared to
be similar at any given polymer dosage
for the mesophilic and thermophilic
sludges.
Discussion and Conclusions
The recirculation of meso-
thermophilically digested sludge to the
activated sludge aerator caused a minor
reduction in sludge mass discharged at
Chicago's Hanover Park WRP that was
not statistically significant. There were no
adverse effects on quality of treated
wastewater.
Although the Torpey Process did not
degrade the quality of the treated
wastewater, it did not produce the
desired reduction of mass of solids that
must be discharged to the environment.
This result contrasts sharply with results
obtained at New York City plants. One
possible reason for the disagreement is
that the New York City plants use high
rate activated sludge processes. The
waste biological sludge from these plants
could very well be substantially reduced
in mass by continued recirculation
through the anaerobic/aerobic processes
in Torpey's scheme.
In Torpey's work, sludge was elutriated
before disposal. This process is not used
in Chicago. It also could be responsible
for loss of suspended solid material that
would not occur at the Hanover plant.
Recirculation of sludge at rates above
40% overloaded the experimental
section's digesters at the Hanover WRP.
Although treated wastewater quality was
not degraded, further increases in
recirculation rate would inevitably cause
this effect. An existing facility should
have excess digestion capacity before
the Torpey Process can be adopted.
The objectionable odor of the
thermophilically digested sludge caused
so many complaints that use of this
process as the terminal digestion step is
ill-advised if sludge is to be utilized on
farmland with close neighbors.
Recommendations
Based on the lack of effect on sludge
mass and the increase in digestion
capacity required, the Torpey process is
not recommended for Chicago's
conventional rate activated sludge plants.
Nor is thermophilic digestion as the
terminal sludge digestion process
-------
-------
-------
recommended if the sludge is to be used
at a site with nearby neighbors.
The full report was submitted in
fulfillment of Cooperative Agreement No.
CR-811925 by the Metropolitan Water
Reclamation District of Greater Chicago,
Chicago, IL, under the sponsorship of the
U.S. Environmental Protection Agency.
Literature Cited
1. Torpey, W. N., J. Andrews, and J. V.
Basilico, J. Wat. Poll. Contr. Fed., 56,
62 (1984)," Effect of Multiple Digestion
on Sludge." !
2. Carrio, L. A., A. R. Lopez, et al., J.
Wat. Poll. Contr. Fed., 57, 116 (1985),
"Sludge Reduction by In-Plant
Modification: New York City's
Experiences."
T.B.S. Pra!
------- |