mesophilic and thermophilic temperature co-phase anaerobic digestion compared with single-stage...
TRANSCRIPT
Water Research 38 (2004) 1653–1662
ARTICLE IN PRESS
*Correspond
51-410-4415.
E-mail addr
0043-1354/$ - se
doi:10.1016/j.w
Mesophilic and thermophilic temperature co-phase anaerobicdigestion compared with single-stage mesophilic- and
thermophilic digestion of sewage sludge
Young-Chae Song*, Sang-Jo Kwon, Jung-Hui Woo
Division of Civil and Environmental System Engineering, Korea Maritime University, 1, Dongsam-Dong, Yeongdo-Gu,
Busan 606-791, South Korea
Abstract
The performance of thermophilic and mesophilic temperature co-phase anaerobic digestions for sewage sludge, using
the exchange process of the digesting sludge between spatially separated mesophilic and thermophilic digesters, was
examined, and compared to single-stage mesophilic and thermophilic anaerobic digestions. The reduction of volatile
solids from the temperature co-phase anaerobic digestion system was dependent on the sludge exchange rate, but was
50.7–58.8%, which was much higher than 46.8% of single-stage thermophilic digestion, as well as 43.5% of the
mesophilic digestion. The specific methane yield was 424–468mL CH4 per gram volatile solids removed, which was as
good as that of single-stage mesophilic anaerobic digestion. The process stability and the effluent quality in terms of
volatile fatty acids and soluble chemical oxygen demand of the temperature co-phase anaerobic digestion system were
considerably better than those of the single-stage mesophilic anaerobic processes. The destruction of total coliform in
the temperature co-phase system was 98.5–99.6%, which was similar to the single-stage thermophilic digestion. The
higher performances on the volatile solid and pathogen reduction, and stable operation of the temperature co-phase
anaerobic system might be attributable to the well-functioned thermophilic digester, sharing nutrients and intermediates
for anaerobic microorganisms, and selection of higher substrate affinity anaerobic microorganisms in the co-phase
system, as a result of the sludge exchange between the mesophilic and thermophilic digesters.
r 2004 Elsevier Ltd. All rights reserved.
Keywords: Anaerobic digestion; Thermophilic; Mesophilic; Co-phase; Sewage sludge
1. Introduction
Single-stage mesophilic completely mixed anaerobic
digestion has been widely used for the reduction in
volume of organic sludge from wastewater treatment
processes, and for obtaining energy in the form of
methane gas. Here, the mesophilic digestion usually
requires over a 20-day retention time, but is not so
efficient in the reduction of volatile solids and the
deactivation of pathogenic organisms. To overcome
these limitations, interest in thermophilic digestion,
using the higher metabolic rate of thermophilic micro-
ing author. Tel.: +82-51-410-4417; fax: +82-
ess: [email protected] (Y.-C. Song).
e front matter r 2004 Elsevier Ltd. All rights reserve
atres.2003.12.019
organisms, has increased [1–4]. Although better perfor-
mance in the reduction of volatile solids and
deactivation of pathogenic organisms can be obtained
from thermophilic digestion, the effluent quality and
ability to dewater the residual sludge are poor, and
require additional energy to heat the digester [2,3,5].
Especially, thermophilic digestion is a little more
sensitive to operational conditions, such as temperature,
and the organic loading rate, as well as to the
characteristics of the influent sludge [5,6]. Generally,
anaerobic processes can be characterized from the
digestion environments, microorganisms and process
configuration, and each process has its unique advan-
tages. According to previous studies [7–10], two-phase
or two-stage anaerobic processes showed good perfor-
mance in the effluent quality, methane yield, volatile
d.
ARTICLE IN PRESS
C-Meso
Motor
Co-phase
C-Thermo
Feed sludge
Biogas
Biogas
35oC
55oC
Motor
S-Thermo
Biogas
Feed sludge55oC
S-Meso
Motor Biogas
Feed sludge
35oC
Motor
(a)(c)
(b)
Fig. 1. Schematic diagrams of the temperature co-phase
anaerobic digestion system (a), the single-stage mesophilic-
(b), and the thermophilic anaerobic digestion processes (c).
Y.-C. Song et al. / Water Research 38 (2004) 1653–16621654
solid reduction and process stability. This implies that
the performance of an anaerobic process could be
improved with the proper combination of the anaerobic
process characteristics. Recently, the temperature-
phased anaerobic digestion (TPAD) process, which
consists of thermophilic- and mesophilic digesters in
series, has been studied in order to incorporate the
advantages of both mesophilic- and thermophilic diges-
tion [11–14]. The TPAD process could be operated at
higher loading rates compared to single-stage processes
[12,13,15] and was better in the deactivation of
pathogenic organisms [13,15] and in its capability for
absorbing shock loadings like other two-stage or two-
phase anaerobic processes [8]. However, the first-stage
thermophilic digester of the TPAD process is still
sensitive to environmental conditions, and has possibly
an influence on the overall performance of the process as
well as the second-stage mesophilic digester. In addition,
the degree of maximum volatile solids’ reduction and
specific methane yield obtainable from the TPAD
process were not much different from that of single-
stage anaerobic processes with enough solid retention
time [12,15].
The purpose of this research is to test a new
configuration of anaerobic digestion, which consists of
a combination of mesophilic and thermophilic diges-
tions, for the more efficient sewage sludge stabilization.
For this, the performances of the single-stage completely
mixed mesophilic and the thermophilic digestions were
examined to clarify their unique characteristics, and
those of a newly conceived temperature co-phase
anaerobic digestion, which simultaneously uses two
temperature biochemical phases via the sludge exchange
between a spatially separated mesophilic and a thermo-
philic digester, were studied and compared with single-
stage anaerobic processes for sewage sludge.
2. Materials and methods
2.1. Experimental apparatus
The schematic diagrams of the anaerobic digestion
systems used for the experiments are shown in Fig. 1.
The temperature co-phase anaerobic digestion system
(a) consisted of a flow through mesophilic digester
(13.6 L) and a retention thermophilic digester (5 L). The
feeding of the influent sludge and the discharging of
the stabilized sludge from the co-phase system were
carried out at the mesophilic digester. An inter-circula-
tion line was installed to exchange the digesting sludge in
the spatially separated two temperature digesters. For
the single-stage anaerobic digestion systems, a comple-
tely mixed type mesophilic digester (b) and the same
type of thermophilic digester (c) were used, with effective
volumes of 12.2 and 5L, respectively. All the digesters
used in the experiments were made of transparent acrylic
tubes with conical style bottoms. A thermostat was
connected to a temperature sensor, which was inserted
into the digester. The outer wall of the digester was
coiled with an electrical heating material and covered
with an insulating material. The temperature of the
sludge during the digestion was maintained by the
thermostat at 3572�C for the mesophilic- and 5572�C
for the thermophilic conditions. The content of the
digester was completely mixed by a blade connected to a
vertical motor. There was a gas sampling port, with an
n-butyl rubber stopper, on the digester cover for the
analysis of the biogas composition. The biogas produc-
tion was monitored with a water displacement gas
collector, which was connected to the digester head
space. The water in the gas collector was acidified
with sulfuric acid and saturated with salt to prevent
the resolution of the biogas. During the experiments, the
wastewater sludge was uniformly supplied to the
digesters 4–8 times a day, using a peristaltic pump
equipped with a timer.
2.2. Experimental methods
The sewage sludge was obtained from a municipal
wastewater treatment plant in B metro city, and was
stored at 4�C in a refrigerator for less than 2 weeks prior
to its use as feed. The sludge was screened with a 1mm
sieve to prevent clogging problems during the transfer
from the feed tank to the digesters. The anaerobic
digestion sludge was taken from a single-stage meso-
philic anaerobic digester for sewage sludge at B metro
city. This sludge was also screened with a 1mm sieve to
remove impurities, and after analysis of the initial
characteristics was used as the inocula for the start-up
of both the anaerobic digestion systems. Table 1 shows
ARTICLE IN PRESS
Table 1
Characteristics of feed sludge and inocula and the experimental conditions
Characteristics pH VS (g/L) SCOD (g/L) Alkalinity
(mg/L as CaCO3)
NH4+-N (mg/L) Total coliform
(colonies/100mL)
Feed sludge 6.9370.3 25.0877.78 6.4571.5 24267521 406750 1� 106–4� 105
Seed sludge 7.5 9.9 1.8 — — —
Process Run HRT (days) SRT (days) Sludge exchange rate (L/d) OLR (gVS/L/d)
Meso- Thermo-
Co-phase I 21 2.38 1.0 7.35Q 0.8670.29
II 21 7.60 5.0 1.67Q 0.8670.10
III 21 4.68 2.5 3.33Q 1.0870.18
S-Meso — 20 20 — 1.4370.18
S-Thermo — 10 10 — 2.9070.37
Q: Flow rate of feeding sludge (L/d); S-Meso and S-Thermo: single-stage mesophilic and thermophilic anaerobic processes.
Y.-C. Song et al. / Water Research 38 (2004) 1653–1662 1655
the average characteristics of the feed sludge and
inocula, and the experimental conditions for the single-
stage and the co-phase anaerobic digestion systems. The
characteristics of the feed sludge, such as volatile solids,
chemical oxygen demand (COD) and alkalinity, varied
widely during the experiments.
During the experiments, the hydraulic retention times
(HRTs) and average organic loading rates for the single-
stage processes were 20 days and 1.43 gVS/L/d for
mesophilic digester, and 10 days and 2.90 gVS/L/d for
thermophilic digester, respectively. For the temperature
co-phase anaerobic digestion, the system HRTs were 21
days, and the various sludge exchange rates between the
mesophilic and the thermophilic digesters were 7.35,
3.33 and 1.67 times that of the inflow rate of the feed
sludge. Corresponding to each sludge exchange rate, the
sludge retention times (SRTs) for the thermophilic
digesters were the same as those of the HRTs, but were
reduced to 2.38, 4.68 and 7.60 days, respectively, for the
mesophilic digesters.
2.3. Analysis and calculations
To monitor the performance of the anaerobic
processes, digesting sludge samples were taken from
the single-stage and co-phase anaerobic digestion
systems daily. According to the Standard Methods
[16], the pH and alkalinity were measured daily and the
total solids (TSs), volatile solids (VSs), total COD
(TCODs) and soluble COD (SCODs) were analyzed
twice weekly. The volatile fatty acid (VFA) was also
measured daily by the titration method proposed by
Anderson and Yang [17], and the composition of the
VFA was analyzed twice weekly by a high-performance
liquid chromatography (DX-500), equipped with an
Aminex HPX-87H column (300� 7.8mm) employing
ultraviolet detection. The biogas production was mon-
itored daily, and its composition was analyzed by a gas
chromatography (Gow-Mac Series 580), equipped with
a stainless-steel column (Porapak Q, mesh size 807100,
6 ft� 1/800), employing thermal conductivity detection.
After the stabilization of the anaerobic processes, the
NH4+-N was measured according to the Standard
Methods [16]. The concentrations of sulfate and ortho-
phosphate were analyzed with ion chromatography
(DX-500) with an Ion-Pac column (AS 144� 250mm)
and a conductivity detector. The total coliform content
of the digested sludge was also measured using the
membrane filtration method, by counting the colonies
after 36 h of incubation at 35�C [16]. During the
experiments, the steady state indicating stable anaerobic
digestion was determined from the stability of certain
parameters, including pH, VS, the VFA-to-alkalinity
ratio and methane content of the biogas. The specific
hydrolysis rate, which was based on the unit particulate
COD (PCOD) loading rate, was estimated by the PCOD
removal rate per unit volume of the digester. The mean
values and standard deviations of the experimental
results during the steady-state operations of the anae-
robic processes were determined to compare the diges-
tion performances by the compensation of the variations
in the characteristics of the feed and digesting sludge.
3. Results and discussion
3.1. Single-stage mesophilic- and thermophilic digestion
During the 70 days of operation of the single-stage
anaerobic processes, the alkalinity level of the thermo-
philic digestion process was higher than that of the
mesophilic process, as shown in Fig. 2(a). It is well
known that the alkalinity in an anaerobic digestion can
be generated from the degradation of nitrogenous
ARTICLE IN PRESSY.-C. Song et al. / Water Research 38 (2004) 1653–16621656
organic compounds, sulfate reduction, release of ortho-
phosphate and an increase of VFAs [18–20]. In this
study, the ammonia nitrogen from the thermophilic
digestion process was 860mg/L, which was higher than
the 630mg/L of the mesophilic process (Table 2).
However, significant differences in the sulfate and
ortho-phosphate were not observed in the single-stage
mesophilic or thermophilic anaerobic digestions. This
indicates that the activity for the degradation of
nitrogenous organic compounds under the thermophilic
Time (day)0 15 30 45 60 75
pH
5
6
7
8
S-MesoS-ThermoFeed sludge
Alk
alin
ity(m
g/L)
as
CaC
O3
0
2000
4000
6000
8000
10000(a)
(b)
Fig. 2. Alkalinity (a) and pH (b) in the single-stage anaerobic
digestion systems.
Table 2
Effluent quality and performance of the single-stage mesophilic and t
Process pH Alkalinity
(mg/L as CaCO3)
NH4+-N
(mg/L)
SO
(m
S-Meso 7.6770.1 64127545 630 36
S-Thermo 8.0870.1 68757546 860 30
Process VFA VSs
Total
(mg HAc/L)
C2:C3:C4
(%)
Concentration
(g/L)
R
(%
S-Meso 579797 97.6:2.4:0 16.1872.7 43
S-Thermo 15877302 26.4:73.6:0 15.3471.18 46
conditions was higher than that under the mesophilic
conditions [19–21]. The pH value of the influent sludge
gradually decreased from 7.2 to 6.8 during the operation
period, as shown in Fig. 2(b). However, the pH value of
the mesophilic process increased from 7.2 to around
7.67, and was stable at this value. The pH of the
thermophilic process was generally higher, at 8.08, than
that of the mesophilic process. This was as a result of the
higher alkalinity of the thermophilic anaerobic digestion
process. The increased alkalinity, and thus pH, from the
degradation of nitrogenous compounds in our experi-
ments is in agreement with previous studies [22,23].
The SCOD level of the thermophilic process was a
little more dependent on the change in the influent
characteristics, and was generally higher than that in the
mesophilic process, as shown in Fig. 3(a). At a steady
state, the mean values of SCODs were 2555 and
5240mg/L, for the mesophilic and thermophilic pro-
cesses, respectively (Table 2). The VFA level in the
thermophilic process was generally higher than that
in the mesophilic process, which was consistent with
the SCOD data (Fig. 3(a)). This clearly shows that the
mesophilic digestion was superior to that of the
thermophilic digestion in terms of the effluent quality,
which can be explained by the low substrate affinity of
some thermophilic organisms [2,3,5,6]. The main com-
ponent of the VFA in the mesophilic process was
acetate, but in the thermophilic process it was propio-
nate (Table 2). From the literatures [2, 5, 6], the higher
level of propionate in the thermophilic digester occurred
under higher hydrogen partial pressures, and the acetate
from higher organic loading rate conditions. In this
study, the accumulation of propionate in the thermo-
philic digester was probably due to the wide variation in
the influent characteristics. This indicates that acetogens
and hydrogenotrophs under the thermophilic condition
are more sensitive to changes in the environments. The
VFA contents of the SCOD were around 22.7 and
30.3%, for the mesophilic and thermophilic digestion
processes, respectively. These were mainly as a result of
he thermophilic digestion processes
42�
g/L)
PO43�
(mg/L)
SCOD
(mg/L)
Total coliform
reduction (%)
.2 42.4 25557493 66.7
.4 39.5 524071404 99.7
Methane
eduction
)
Specific hydrolysis
rate (g PCOD/L/d)
CH4 (%) Specific yield
(mL/g VSrem)
.578.4 0.451 64.772.6 451.1745.0
.875.4 0.520 63.671.8 416.0767.1
ARTICLE IN PRESS
Time(day)
0 15 30 45 60 75
VF
A/A
lkal
inity
0.0
0.2
0.4
0.6
VF
A(m
g/L
as H
Ac)
0
1500
3000
4500
6000
SC
OD
(mg/
L)
0
2000
4000
6000
8000
10000
S-MesoS-ThermoFeed sludge
(a)
(b)
(c)
Fig. 3. SCOD (a), VFA (b) and the VFA-to-alkalinity ratio (c)
in the single-stage anaerobic digestion processes.
VS
(g/L
)
0
10
20
30
40
50
60
70
80
VS
red
uctio
n(%
)
10
20
30
40
50
60
70
80
Feed sludgeS-MesoS-ThermoS-Meso(%)S-Thermo(%)
(a)
Time (day)
0 10 20 30 40 50 60 70
Spe
cific
met
hane
yie
ld(m
lCH
4/g
VS
rem
oved
)0
200
400
600
800
1000
Met
hane
con
tent
(%)
0
10
20
30
40
50
60
70
80
90S-Meso(mL)S-Thermo(mL)S-Meso(%)S-Thermo(%)
(b)100
Fig. 4. VSs (a) and biogas (b) in single-stage anaerobic
processes.
Y.-C. Song et al. / Water Research 38 (2004) 1653–1662 1657
the higher level of propionate under thermophilic
condition.
The VFA to alkalinity ratio for the two single-stage
anaerobic systems were monitored to compare the
buffering capacities for the rapid change of pH
(Fig. 3(C)). It has been reported that the buffering
capacity was sufficient when the VFA-to-alkalinity ratio
was maintained below 0.4 [11]. In this study, the ratio of
the mesophilic process was almost constant at around
0.1, with the exception of the start-up period. For the
thermophilic anaerobic digestion process, this ratio was
a little unstable during the early stage of the process
operation. However, as the steady-state conditions were
approached, this value decreased to around 0.2 with
some fluctuations. The slightly higher VFA-to-alkalinity
ratio of the thermophilic process was primarily as a
result of the higher VFA concentration. This indicates
that the single-stage mesophilic anaerobic digestion
could have better buffering capabilities than the
thermophilic digestion.
The reduction in VSs during anaerobic digestion is
generally equal to the total amount of VFA converted
from the volatile dissolved solids. The dissolved VSs are
produced from the hydrolysis of suspended VSs, and the
VFA is finally converted to methane gas. The reduction
in the VSs can be expressed as the sum of the residual
VFA and the methane gas produced from the anaerobic
digester. Therefore, the hydrolysis of particulate organ-
ics has a significant influence on the reduction of VSs. In
this study, the levels of VSs were constantly maintained
throughout the operations at 16.18 g/L for the meso-
philic and 15.34 g/L for the thermophilic digester,
despite the wide variation in the influent characteristics
of the feed sludge, as shown in Fig. 4(a). Therefore, the
VS reduction of the thermophilic digester was consider-
ably dependent on the feed sludge characteristics. This
implies that the single-stage anaerobic digestion pro-
cesses have large potentials for the stable reduction of
VSs, which are not influenced by the temperature
conditions. However, the specific hydrolysis rate in the
thermophilic process during the steady state was
0.520 gPCOD/L/d, which was a little higher than the
0.451 gPCOD/L/d for the mesophilic process (Table 2).
This led to the higher reduction of volatile solids of
ARTICLE IN PRESS
Time (days)
Alk
alin
ity (
mg/
L as
CaC
O3)
0
1000
2000
3000
4000
5000
6000
0 20 40 60 80 100 120
pH
6
7
8
C-MesoC-ThermoFeed sludge
(a)
(b)
Run I Run II Run III
Fig. 5. Alkalinity (a) and pH (b) in the mesophilic/thermophilic
co-phase anaerobic digestion system.
Y.-C. Song et al. / Water Research 38 (2004) 1653–16621658
46.8% in the thermophilic digester compared to the
43.5% of the mesophilic digester. Maibaum et al. [3]
reported that the difference in the degradation rates of
solid substrates under the thermophilic and mesophilic
conditions become significant in relation to the decrease
in the retention time [3].
As shown in Fig. 4(b), the average methane content
of the biogas from the mesophilic process was just a
little higher, at around 64%, than that of the thermo-
philic process. This was probably a result of the reduced
solubility of carbon dioxide under the thermophilic
condition [22]. In previous studies, the methane content
of the biogas was mainly affected by the types of
substrate, rather than the temperature conditions,
for the anaerobic digestion [4,22,24]. However, the
specific methane yield of the mesophilic process, based
on the removed VS, was a little more sensitive to the
influent characteristics of the feed sludge, indicating the
higher capacity of mesophilic methanogens for coping
with the variation of the influent characteristics com-
pared to the thermophilic methanogens. The average
specific methane yield of the thermophilic process was
lower, at 416mLCH4/gVSremoved, than the
451mLCH4/gVSremoved of the mesophilic digester
(Table 2). This was presumably due to the higher
maintenance energy of the anaerobic thermophilic
microorganisms [5,6,25], as well as the higher hydrogen
content of the biogas [22].
Interest in the deactivation of pathogenic organisms,
for the production of Class A biosolids from the
digested residual sludge, from wastewater sludge during
anaerobic digestion has significantly increased in recent
years. In this study, the percentage of deactivation of
total coliform in the thermophilic digester was 99.7%,
which was much higher than the 66.7% in the
mesophilic digester (Table 2). The higher deactivation
of pathogenic organisms in the thermophilic process was
probably a result of the combined effects of the long
retention time of 10 days, under higher levels of free
ammonia and VFA, as well as the thermophilic
conditions. [14].
3.2. The mesophilic and thermophilic co-phase anaerobic
digestion
The alkalinity levels of the temperature co-phase
mesophilic and thermophilic digesters were influenced
by the alkalinity variation of influent sewage sludge, as
shown in Fig. 5(a). The average level of alkalinity in
the co-phase thermophilic digester was around 3500–
4700mg/L as CaCO3, which was a little higher than
3300–4400mg/L as CaCO3 in the co-phase mesophilic
digester. The higher alkalinity under the thermophilic
condition was similar to that of the single-stage
anaerobic processes, as shown in Fig. 2(a), and reflects
the higher degradation activity of nitrogenous organic
compounds, such as proteins, under the thermophilic
condition [22,23]. The pH levels of the co-phase
mesophilic and thermophilic digesters in the early stage
of the operation were a little unstable due to the wide
variation in the feed characteristics, as shown in
Fig. 5(b). On the 35th day, the pH level in the mesophilic
digester was stable at 7.5–7.6, which was similar to that
in the single-stage mesophilic anaerobic process. The pH
in the co-phase thermophilic digester during the initial
stage of the operation increased to over 8.0, which was
similar to the single-stage thermophilic digester. How-
ever, at the stable stage, the pH in the thermophilic
digester was slightly decreased to 7.7–7.8, which was a
more favorable pH condition for the thermophilic
anaerobic bacteria [8,26]. This was a result of the
exchange of the thermophilic sludge with the lower
alkalinity sludge from the mesophilic digester [27]. The
influence of the sludge exchange rate on the pH of the
co-phase system was not observed apparently. However,
the alkalinity levels of the co-phase mesophilic and
thermophilic digesters were slightly increased in relation
to the increase of the sludge exchange rate from 1.67Q to
over 3.33Q. This was a result of the enhanced degrada-
tion of nitrogenous compounds due to the extended
portion of the sludge passing through the thermophilic
digester at higher sludge exchange rate [22,23,26].
ARTICLE IN PRESS
Time (day)0 20 40 60 80 100 120
SC
OD
(m
g/L)
0
2000
4000
6000
8000
10000
VF
A(m
g/L
as H
Ac)
1000
2000
3000
4000
C-MesoC-Thermofeed
(a)
(b)
Run(I) Run(II) Run(III)
Fig. 6. VFA (a) and SCOD (b) in the temperature co-phase
anaerobic digestion system.
Y.-C. Song et al. / Water Research 38 (2004) 1653–1662 1659
Fig. 6(a) shows the VFA trends of the mesophilic and
thermophilic co-phase anaerobic digestion system. After
35 days of the operation, the levels of VFA in the co-
phase thermophilic digester became stable, as well as
that of the mesophilic digester, and were not influenced
by the wide change in the influent characteristics. At
stable state, the VFA levels in the co-phase thermophilic
digester were 339–679mgHAc/L, which were lower than
436–795mgHAc/L in the mesophilic digester for all
rates of the sludge exchange (Table 3). This indicates
that the thermophilic digester of the co-phase system
was stable and well functioned and the affinity of the
thermophilic sludge on VFA was quite higher than that
of the sludge from the single-stage thermophilic digester
(Fig. 3(b)). This seems to suggest that the higher
substrate affinity methanogenic bacteria were selected
and dominated in the co-phase thermophilic digester by
the sludge exchange between the mesophilic and
thermophilic digesters. In the case of the co-phase
mesophilic digester, the VFA level was also less than
that of the single-stage mesophilic digester, at 1.67Q and
3.22Q of the sludge exchange rates. However, at 7.36Q
of the sludge exchange rate, the VFA was increased to
around 795mgHAc/L, indicating that the methanogen-
esis was the rate limiting step in the overall anaerobic
reactions due to the reduced solid retention time of 2.38
days in the mesophilic digester. The main VFA
component of the co-phase mesophilic digester was
acetate, as the single-stage mesophilic process (Tables 2
and 3). However, in the co-phase thermophilic digester,
the propionate content was considerable, at 7.35Q of the
sludge exchange rate, as the single-stage thermophilic
process. This higher propionate content at the higher
sludge exchange rate in the co-phase thermophilic
digester was probably related to the higher hydrogen
partial pressure [5,22].
Fig. 6(b) shows the SCOD levels of the mesophilic and
thermophilic temperature co-phase digestion systems.
At steady state, the levels of SCOD in the co-phase
mesophilic and thermophilic digesters were 2100–2200
and 1700–2900mg/L, which were less than those of
single-stage mesophilic and thermophilic processes,
respectively. The good effluent quality in the SCOD
was mainly attributable to the low VFA in the co-phase
mesophilic and thermophilic digesters, probably as a
result of the higher methanogenic activity and higher
affinity of the anaerobic sludge on VFA in the co-phase
system.
Fig. 7(a) shows the VFA-to-alkalinity ratio required
to evaluate the buffering capacity of the temperature co-
phase anaerobic digestion system. At 7.35Q of the
sludge exchange rate, the VFA-to-alkalinity ratios were
stable, at 0.19 for the mesophilic- and 0.16 for the
thermophilic digester, except for the early stage of the
operation. As the sludge exchange rates were reduced to
3.33Q and 1.67Q, the VFA-to-alkalinity ratios were a
little more reduced and stabilized. These indicate that
the buffering capacity in the temperature co-phase
anaerobic system was sufficient for sewage sludge
digestion, as with the single-stage mesophilic anaerobic
processes [11]. The higher buffering capacity in the co-
phase thermophilic digester was attributable to both
higher alkalinity level from the enhanced degradation of
nitrogenous compounds and lower VFA level by the
higher substrate affinity of methanogens. The higher
buffering capacity in the co-phase thermophilic digester
also contributed to good buffering capacity in the
mesophilic digester through the sludge exchange be-
tween the mesophilic and thermophilic digesters.
Generally, the mesophilic anaerobic digestion requires
over a 20-day retention time to stabilize the wastewater
sludge, which was due to the slow growth rate of the
mesophilic anaerobic bacteria [1]. However, as presented
in Table 1, the SRTs of the mesophilic digester of the
temperature co-phase anaerobic digestion system were
varied between 2.38 and 7.60 days according to the
sludge exchange rate between the mesophilic and
thermophilic digesters. The overall specific methane
yields were as good as the single-stage mesophilic
anaerobic process, although some portion of the overall
yield was from the thermophilic digester of the co-phase
digestion system. Especially, at 7.36Q of the sludge
exchange rate, the SRT of the mesophilic digester was
ARTICLE IN PRESS
Table 3
Performance of the temperature co-phase anaerobic digestion system
Content Run I Run II Run III
Meso- Thermo- Meso- Thermo- Meso- Thermo-
Exchange rate (L/d) 7.35Q 1.67Q 3.33Q
pH 7.5370.07 7.7370.09 7.5270.06 7.7470.07 7.5570.06 7.7570.13
Alkalinity
(mg/L as CaCO3)
43087271 45207239 33557320 35137373 43577118 46217215
SCOD (mg/L) 21337148 22557193 21787168 29757103 2182788 1745775
VFA mg/L as Hac 795766 6977128 4467150 3397179 436768 364742
C2: C3: C4 (%) 86.1:10.2:3.7 77.4:20.5:2.1 67.7:32.3:0 93.8:6.2:0 83.8:6.6:9.6 96.7:3.3:0
Specific hydrolysis
rate (g PCOD/L/d)
Each digester 0.098 0.035 0.286 0.019 0.225 0.014
Overall 0.537 0.524 0.566
VSs Concentration (g/L) 10.2070.27 9.9770.32 8.8170.28 8.2070.18 10.1070.44 9.9370.12
Reduction (%) 58.472.7 50.773.2 58.872.0
Specific methane
yield (mLCH4/
gVSremoved)
459.7747.4 468.2734.7 424.2743.1
Destruction of
total coliform (%)
99.6 98.5 99.1
Y.-C. Song et al. / Water Research 38 (2004) 1653–16621660
only 2.38 days, but the specific methane yield was higher
at 460mLCH4/g VSremoved. This shows the possibility
that there are some temperature facultative anaerobic
microorganisms that are highly active under mesophilic,
as well as thermophilic conditions in the temperature co-
phase anaerobic digestion system with sludge exchange.
This is different from the widely held view that two
distinct groups of anaerobic bacteria exist at the
mesophilic and thermophilic temperature regimes. In a
previous study [13], it has been reported that they are the
temperature-tolerant anaerobic bacteria from the anae-
robic biomass activity tests under mesophilic and
thermophilic conditions.
The hydrolysis rate in the anaerobic digestion of
sewage sludge is an important parameter to evaluate the
process performance due to the higher content of the
particulate organics in the sewage sludge. The overall
specific hydrolysis rate of the co-phase anaerobic
digestion system, estimated as the PCOD removal rate
per unit volume based on the PCOD loading rate in the
digester, was 0.537 g PCOD/L/d at 7.35Q of the sludge
exchange rate, which was higher than 0.520 gPCOD/L/d
of single-stage thermophilic anaerobic digestion process
as well as 0.451 g PCOD/L/d of single-stage mesophilic
anaerobic digestion process (Table 3). However, the
specific hydrolysis rate for the thermophilic digester in
the co-phase digestion system was 0.035 gPCOD/L/d,
which was less than 0.098 gPCOD/L/d of the mesophilic
digester, indicating that most of the particulate organics
were hydrolyzed in the mesophilic digester. This is quite
different from that of the single-stage anaerobic
processes, where the hydrolytic activity of the thermo-
philic digester was higher than that in the mesophilic
digester. These interesting results might arise because the
intermediates including hydrolytic enzyme, alkalinity
and other nutrients were easily produced from the stable
and well-functioned thermophilic digester, and were
then transferred into the mesophilic digester through the
sludge exchange process to make the favorable meso-
philic condition for the hydrolytic enzyme, or anaerobic
microorganisms. When the sludge exchange rate was
decreased to 3.33Q and 1.67Q, the overall specific
hydrolysis rates were varied to 0.566 and
0.524 gPCOD/L/d. Independent of the sludge exchange
rate in the co-phase digestion system, however, the
specific hydrolysis rate of the mesophilic digester was
higher than that of the thermophilic digester.
The VSs in the co-phase mesophilic and thermophilic
digesters were stable, which were not influenced by the
VSs variation of the influent sludge, as shown in Fig.
7(c). The reduction of volatile solids was around 51% at
1.67Q of the sludge exchange rate, but increased to
around 59% when the sludge exchange rate was
increased to over 3.33Q as presented in Table 3. In the
literature [15], the reduction of volatile solids obtained
from TPAD process for waste-activated sludge was
about 50% at 28 days of SRT, which was around 10%
higher than that of the single-stage mesophilic digester.
In this study, the reduction of volatile solids that could
be obtained in the co-phase digestion system was over
15% higher than that of the single-stage mesophilic
digester and around 12% higher than the single-stage
thermophilic digester. The enhanced performance on the
VS reduction obtained from the temperature co-phase
ARTICLE IN PRESSS
peci
fic m
etha
ne y
ield
(m
L C
H4/
g V
S re
mov
ed)
0
200
400
600
800
1000
1200
Time (days)0 20 40 60 80 100 120
VS
(g/
L)
0
20
40
60
VS
red
uctio
n (%
)
0
20
40
60
80
100
C-MesoC-ThermoFeed sludgeVS reduction(%)
VF
A/A
lkal
inity
0.0
0.5
1.0
1.5
2.0
2.5
Run I Run II Run III(a)
(b)
(c)
Fig. 7. VFA-to-alkalinity ratio (a), specific methane yield (b)
and VSs (c) in the temperature co-phase anaerobic digestion
system.
Y.-C. Song et al. / Water Research 38 (2004) 1653–1662 1661
anaerobic digestion system was mainly attributable to
the higher hydrolytic activity of the mesophilic digester,
which was a result of the sludge exchange between the
mesophilic and thermophilic digesters. On the other
hand, the additional energy for the sludge exchange and
for heating the thermophilic digester in the co-phase
digestion system is possibly required. However, the
additional energy requirements could be compensated
by the advantages of the co-phase digestion system
including higher reduction of volatile solids, better
effluent quality and process stability and increased
methane production, compared to the single-stage
mesophilic- or the thermophilic processes.
The destruction of total coliform in the co-phase
anaerobic digestion system was increased from 98.5% to
99.6% in relation to the increase in the sludge exchange
rate, as presented in Table 3. This shows that the ratio of
sludge passing through the thermophilic digester in the
temperature co-phase system plays an important role in
the destruction of total coliform, rather than the
retention time, under the thermophilic condition.
4. Conclusions
From the studies on the characteristics of the single-
stage mesophilic- and thermophilic anaerobic processes,
the treating of sewage sludge, and the mesophilic and
thermophilic temperature co-phase anaerobic digestion
systems, the following conclusions were made. The
single-stage mesophilic anaerobic digestion in terms of
the specific methane yield, effluent quality and process
stability was superior to the thermophilic digestion, but
both VS reduction and total coliform destruction from
the single-stage thermophilic digestion were higher than
those of the mesophilic digestion. The performance of
the mesophilic and thermophilic co-phase anaerobic
digestions was dependent on the sludge exchange rate
between the mesophilic and thermophilic digesters, but
the advantages of single-stage mesophilic and the
thermophilic anaerobic digestions could be obtained
from the temperature co-phase anaerobic digestion
system. The effluent quality in terms of SCOD and
VFA, specific methane yield and process stability, which
could be obtained from the temperature co-phase
anaerobic digestion, were better than those of the
single-stage mesophilic anaerobic digestion. The patho-
gen destruction was similar to those of the single-stage
thermophilic digestion, but the reduction of volatile
solids was much higher than that of the single-stage
thermophilic digestion. These higher performances of
the temperature co-phase anaerobic digestion might be
mainly attributable to stable and well-functioned
anaerobic thermophilic digester, selection of the active
and higher substrate affinity of anaerobic microorgan-
isms and sharing the nutrients and intermediates for
anaerobic microorganisms, as a result of the sludge
exchange between the mesophilic and thermophilic
digesters.
Acknowledgements
This work was supported by the academic research
program (Grant No. 2001-N-BI03-P-04) of the Korea
Energy Management Corporation.
References
[1] Aoki N, Kawase M. Development of high performance
thermophilic two-phase digestion process. Water Sci
Technol 1991;23:1147–56.
ARTICLE IN PRESSY.-C. Song et al. / Water Research 38 (2004) 1653–16621662
[2] Fang HHP, Chung DWC. Anaerobic treatment of
proteinaceous wastewater under mesophilic and thermo-
philic conditions. Water Sci Technol 1999;40(1):77–84.
[3] Maibaum C, Kuehn V. Thermophilic and mesophilic
operation of an anaerobic treatment of chicken slurry
together with organic residual substances. Water Sci
Technol 1999;40(1):231–6.
[4] Zabranska J, Stepova J, Wachtl R, Jenicek P, Dohanyos
M. The activity of anaerobic biomass in thermophilic and
mesophilic digesters at different loading rates. Water Sci
Technol 2000;32(9):49–56.
[5] Kim M, Ahn YH, Speece RE. Comparative process
stability and efficiency of anaerobic digestion; mesophilic
vs. thermophilic. Water Res 2002;36:4369–85.
[6] van Lier JB. Limitation of thermophilic anaerobic waste-
water treatment and the consequences for process design.
Antonie van Leeuwenhoek 1996;69:1–14.
[7] Azbar BN, Speece R. Two-phase, two-stage and single-
stage anaerobic process comparison. J Environ Eng 2001;
127(3):240–7.
[8] Azbar N, Ursillo P, Speece RE. Effect of process
configuration and substrate complexity on the performance
of anaerobic processes. Water Res 2001;35(3):817–29.
[9] Roberts R, Son L, Forster CF. Thermophilic/mesophilic
dual digestion system for treating waste activated sludge.
J Chem Technol Biotechnol 1999;74:445–50.
[10] Schafer PL, Farrell JB. Advanced anaerobic digestion
performance comparison. Proceedings of the 75th WEF
Annual Conference and Exposition (WEFTEC 2002),
Session 46. Chicago: WEF; 2002.
[11] Zhao Q, Kugel G. Thermophilic/mesophilic digestion of
sewage sludge and organic wastes. J Environ Sci Health
1996;A31(9):2211–31.
[12] Song YC, Park SH, Lee JS. Enhanced anaerobic stabiliza-
tion of sewage sludge using TPAD process. J Korean Soc
Civil Eng 2001;21(6B):705–13.
[13] Vandenburgh SR, Ellis TG. Effect of varying solids
concentration and organic loading on the performance of
temperature phased anaerobic digestion process. Water
Environ Res 2002;74(2):142–8.
[14] Ferran B. Two-phase anaerobic digestion of municipal sewage
sludge optimization of the pathogen destruction. Proceedings
of the 75th WEF Annual Conference and Exposition
(WEFTEC 2002), Session 46. Chicago: WEF; 2002.
[15] Han Y, Sung S, Dague RR. Temperature-phased anaero-
bic digestion of wastewater sludge. Water Sci Technol
1997;36(6–7):367–74.
[16] American Public Health Association (APHA). Standard
methods for the examination of waste and wastewater,
18th ed. Washington, DC: APHA, AWWA, 1992.
[17] Anderson GK, Yang G. Determination of bicarbonate and
total volatile acid concentration in anaerobic digesters
using a simple titration. Water Environ Res 1992;64(1):
53–9.
[18] Capri MG. Marais GvR. pH adjustment in anaerobic
digestion. Water Res 1975;9:307–13.
[19] van Haandel AC. Influence of the digested COD concen-
tration on the alkalinity requirement in anaerobic diges-
ters. Water Sci Technol 1994;30(8):23–34.
[20] Munch EV, Greenfield PF. Estimating VFA concentra-
tions in prefermenters by measuring pH. Water Res
1998;32(8):2431–41.
[21] Sanchez E, Borja R, Weiland P, Travieso L, Martin A.
Effect of temperature and pH on the kinetics of methane
production, organic nitrogen and phosphorus removal in
the batch anaerobic digestion process of cattle manure.
Bioprocess Eng 2000;22:247–52.
[22] Gallert C, Winter J. Mesophilic and thermophilic anaero-
bic digestion of source sorted organic wastes: effect of
ammonia on glucose degradation and methane produc-
tion. Appl Microbial Biotechnol 1997;48:405–10.
[23] Yu HQ, Fang HHP, Gu GW. Comparative performance
of mesophilic and thermophilic acidogenic upflow reactor.
Process Biochem 2002;38:447–54.
[24] Ahn JH, Forster CF. A comparison of mesophilic and
thermophilic anaerobic upflow filter. Bioresearch Technol
2000;73:201–5.
[25] Borja R, Martin A, Banks CJ, Alonso V, Chica A. A
kinetic study of anaerobic digestion of olive mill waste-
water at mesophilic and thermophilic temperatures.
Environ Pollut 1995;88:13–8.
[26] De Beer D, Huisman JW, Van den Heuvel JC, Ottengraf
SPP. The effect of pH profiles in methanogenic aggregates
in the kinetics of acetate conversion. Water Res 1992;
26(10):1329–36.
[27] Romli M, Greenfield PF. Effect of recycle on a two-phase
high-rate anaerobic wastewater treatment system. Water
Res 1994;28(2):475–82.