Bioresource Technology 82 (2002) 65±71

E€ects of temperatures and organic loading rates on

biomethanation of acidic petrochemical wastewater
using an anaerobic up¯ow ®xed-®lm reactor
Hardik Patel, Datta Madamwar *

Post-Graduate Department of Biosciences, Sardar Patel University, Vallabh Vidyanagar, 388120 Gujarat, India
Received 24 July 2001; received in revised form 26 July 2001; accepted 5 August 2001

Abstract
The e€ect of temperature and organic loading rate on the rate of methane production from acidic petrochemical wastewater
without neutralization was investigated by continuously feeding an anaerobic up¯ow ®xed-®lm reactor. The temperatures selected
for the studies were 25, 37, 45 and 55 °C. Organic loading rate (OLR) for each temperature was varied from 3.6 to 21.7 kg COD
m 3 d 1 . Best performance with respect to COD and BOD reduction, total gas production and methane yield was obtained with the
reactor operating at 37 °C. OLR could be increased to a maximum of 21.7 kg COD m 3 d 1 with 90±95% COD and BOD reduction
and methane yield of 0.450 m3 kg 1 COD d 1 added. The reactor operating at 55 °C gave the highest methane yield of
0:666 m3 kg 1 COD d 1 at an OLR of 6 kg COD m 3 d 1 . This decreased to 0.110 m3 kg 1 COD d 1 when the OLR was increased
to 18.1 kg COD m 3 d 1 . The reactor operating at 45 °C gave a maximum methane yield of 0:416 m3 kg 1 COD d 1 added at an
OLR of 6 kg COD m 3 d 1 . On further increasing the OLR to 9 kg COD m 3 d 1 , COD reduction was 89%, however, methane
yield decreased to 0:333 m3 kg 1 COD d 1 added. The highest methane yield of 0:333 m3 kg 1 COD d 1 added at an OLR of 6 kg
COD m 3 d 1 was obtained with reactors operating at 25 °C. These studies indicate potential rates of methane production from
acidic petrochemical wastewater under di€erent temperatures. This provides a guideline for various kinetic analyses and economic
evaluation of the potential feasibility of fermenting acidic wastewater to methane. Ó 2002 Elsevier Science Ltd. All rights reserved.

Keywords: Anaerobic bioreactor; Biomethanation; Fixed ®lm; Petrochemical wastewater; Temperature; Organic loading rate

1. Introduction anaerobic digestion is not applied in the petrochemical

It has become increasingly important to prevent the
pollution of limited water resources by providing ade-
quate treatment of e‚uents from industrial sources
(Desai et al., 1994). Two concurrent problems facing our
country are the disposal of wastes generated by indus-
tries and the need for new sources of fuel to supply the
energy needs of society (Pfe€er, 1974). Anaerobic di-
gestion o€ers an excellent opportunity for both energy
conservation and pollution control considerations (Hwu
et al., 1997; Patel et al., 1998; Wheatley, 1990). Major
advantages of this process are the low operational costs,
high energy eciency and process simplicity as com-
pared to other waste treatment methods (Lettinga, 1995;
Patel et al., 1995). However, despite these advantages,
*
Corresponding author. Tel.: (O) +91-2692-26899 (R) +91-2692-
34877; fax: +91-2692-36475.
E-mail address: datta_madamwar@yahoo.com (D. Madamwar).
industry largely due to problems of slow reactions,
which require longer hydraulic retention time (HRT)
and poor process stability in conventional reactor de-
signs. In order to solve these problems and to develop a
better methanogenic process more focus is directed to-
wards developing more suitable high-rate anaerobic bi-
oreactors (Parkin and Speece, 1983; Parkin et al., 1983).
One type of such a con®guration is the up¯ow ®xed-®lm
anaerobic bioreactor, which is capable of retaining ac-
tive biomass in the reactor without the need for biomass
recirculation.
A number of petrochemical wastewater components
are fermented to methane and attached growth systems
have shown considerable advantages (Britz et al., 1983;
Chou et al., 1978; Lettinga et al., 1999; Nel and Britz, 1986;
Nel et al., 1984; Ramakrishna and Desai, 1997; Sharma et
al., 1994; Vartak et al., 1997). In a previous study (Patel
and Madamwar, 2000), we investigated the role of ap-
propriate support materials, which allowed the develop-

0960-8524/02/$ - see front matter Ó 2002 Elsevier Science Ltd. All rights reserved.
PII: S 0 9 6 0 - 8 5 2 4 ( 0 1 ) 0 0 1 4 2 - 0
66 H. Patel, D. Madamwar / Bioresource Technology 82 (2002) 65±71
ment of a speci®c methanogenic population adapted to a
low pH environment for the biomethanation of petro-
chemical wastewater. Among the various carrier mate-
rials tested, bone charcoal (charcoal of bones) exhibited
the best results in terms of biogas and methane yield with
the highest COD removal. There is growing interest in
maximizing methane extraction for energy recovery from
acidic wastes such as petrochemical wastewater. One of
the most important parameters for anaerobic treatment
of wastewater is operating temperature, which selects the
dominant bacterial ¯ora and determines growth rate.
The mesophilic range is traditionally used (Varel et al.,
1980) since it is generally thought that maintaining a high
temperature is uneconomical whereas degradation
within the psychrophilic range is too slow. However, the
thermophilic range now merits consideration because
reaction rates are considerably higher than in the meso-
philic and psychrophilic ranges (Jimeno et al., 1990;
Kotsyurbenko et al., 1993; Lepisto and Rintala, 1996,
1999; Romero et al., 1988; Sa¯ey and Westerman, 1992;
Varel et al., 1980; Vartak et al., 1997; Wiegant et al.,
1985) and thus the HRT necessary to reduce organic load
by a certain percentage is shorter. As such, no detailed
study appears to have been made so far on the e€ect of
temperature and HRT on methane generation from low
pH petrochemical industrial wastewater. Therefore, this
study was undertaken with the aim of maximizing the
generation of methane from low pH petrochemical in-
dustrial wastewater by optimizing temperature and or-
ganic loading rate.
2. Methods
2.1. Fermentor
Laboratory scale anaerobic up¯ow ®xed-®lm glass
column reactors were used in the present studies with the
following speci®cations: reactor inner diameter 4.5 cm,
reactor height 120 cm., media height 108 cm, total vol-
ume (without bedding material) 1.7 l, working volume 1
l. Reactors were packed with bone charcoal as a support
material of size 125 mm2 . The speci®c surface area and
pore speci®c volume of the material was 53:35 m2 g 1
and 0:244 cm3 g 1 of support material, respectively.
Approximately 660 g of support material was used in
each reactor.
2.2. Inoculum development and reactor operation
Bio®lm was allowed to develop using 3% (w/v) cattle
dung slurry as the initial inoculum. Cowdung slurry
contained approximately (w/v) 3% COD, 0.34% volatile
fatty acid (VFA), 3% total solids (TS), 2% volatile solids
(VS), 0.06% total potassium and nitrogen (TKN) and a
pH of 7.6. Reactors were operated at di€erent temper-
atures. Temperatures selected were 25, 37, 45 and 55 °C.
During inoculum development, facultative anaerobic
organisms utilized the organic matter present in the
slurry as substrate and created anaerobic conditions
suitable for the growth of strict anaerobes. Establish-
ment of the active biomass on the support media was
based on the constant gas production from utilization of
organic matter present in the cowdung slurry and the
change in the inoculum color from yellowish brown to
black. In reactors operated at 45 and 55 °C, depletion in
gas production was observed between 15 and 20 days
and reactors at 37 °C required approximately 30±35
days whereas 75±80 days were required at 25 °C. This
may have been due to depletion of organic matter in the
slurry. The slurry was slowly replaced by alternate day
feeding with a petrochemical wastewater of pH 2.5 as
substrate without neutralization at an HRT of 30 days
for approximately 50 days. Subsequently reactors were
operated by continuous feeding of the petrochemical
wastewater starting with a HRT of 30 days slowly
changing to the desired HRT with an increase in organic
loading rate (OLR) and allowing steady-state condition
to eventually be achieved. Steady-state condition is de-
®ned as conditions prevailing in the reactor at which
constant COD reduction of the outlet liquid and biogas
production occurs at a ®xed OLR or HRT (for example,
reactors operating at 37 °C and 15 days HRT during
steady-state conditions showed 98  0:5% COD reduc-
tion and total biogas production of 2:0 
0:4 m3 m 3 d 1 with OLR of 3.6 kg COD m 3 d 1 ).
This indicated acclimatization of a methanogenic pop-
ulation for low pH and to the toxic components of the
petrochemical wastewater. All reactors were operated
on a continuous basis at the desired retention time
(HRT) for 60 days after reaching steady-state condi-
tions. Petrochemical wastewater was continuously
(OLR as given in Tables 2±5) fed into the reactor in an
upward direction at the required rate using a peristaltic
pump (Gilson, Minipuls 3, Model M312, France).
2.3. Substrate
Acidic petrochemical wastewater was used as a sub-
strate. The wastewater was received from the e‚uent
plant of the Gujarat State Fertilizer Company of Baroda
in India. Composition and characteristics of the petro-
chemical wastewater are given in Table 1.
2.4. Analytical methods
Biogas production was measured by the displacement
of acidi®ed saturated salt solution (pH 2.5), making
corrections for atmospheric pressure and temperature
(STP). Gas composition was determined with a Sigma,
Baroda (India) (Model M505) gas±liquid chromato-
graph equipped with a 2 m stainless steel column packed
 H. Patel, D. Madamwar / Bioresource Technology 82 (2002) 65±71 67

Table 1 60 °C for 2 min and raised to 120 °C at the rate of

Composition and characteristics of the petrochemical wastewater 8 °C min 1 and further increased to 250 °C at the rate of
Components Concentration 12 °C min 1 ). The injector and detector temperatures
Formic acid 46:60 g l 1 were 200 and 250 °C, respectively. Identi®cation and
Acetic acid 46:60 g l 1 percentage of di€erent fatty acids and aromatic com-
Cyclohexanol 0:3 g l 1
pounds were based on comparison of retention times
Cyclohexane 0:5 g l 1
Cyclohexanone 0:3 g l 1 and peak areas of unknown with standard compounds.
Phenol 0:360 g l 1 Pure compounds (Analytical Grade Reagent, Ranbaxy
Total nitrogen (as N2 ) 0:050±0:212 g l 1
Lab., India) were used as standard.
1
Total phosphate (as P) 0:102±0:227 g l The speci®c surface area and porosity of the support
1
Total sulfate (as SO4 ) 0:150±0:242 g l
material were determined on a Micrometry's BET Sur-
Chlorides 2±3 mg l 1
Iron 0:05±0:80 mg l 1 face Area Analyzer (Model No. 2375, USA). Experi-
Heavy metals 0:15±0:20 mg l 1 ments were carried out in quadruplicate and averages and
(Co, Ni, Mo, Cr) standard deviations (S.D.) were calculated from the data.
Oil and grease 12±13 mg l 1
COD 55±60 g l 1 2.5. Statistical analysis
BOD 30±32 g l 1
Volatile fatty acids 93±95 g l 1
Total solids 20±300 mg l 1 Values reported are averages of the steady-state data
Total acidity 45±46 g l 1 of reactors operated at given temperatures in quadru-
pH 2.5±2.7 plicate (four reactors per temperature). The Microsoft
Excel (Release 2000) under Windows 98 operating sys-
tem (ICON Software Technologies, Baroda, Gujarat,
with Porapak R (80±100 mesh) at 40 °C and a thermal India) was used to carry out the statistical calculations
conductivity detector. Nitrogen was used as a carrier gas given in Tables 2±5. The values represented in the tables
at a ¯ow rate of 30 ml min 1 . The injector and detector are the limits (2 S.D. limits) within which the observed
temperatures were kept at 125 °C. Feed and reactor data lies. The 2 S.D. from the mean covers 95% of the
e‚uent samples were routinely analysed for pH, COD, observed data. The observed mean e€ect is signi®cant
BOD, volatile fatty acid (VFA), alkalinity, total solids, with a P value less than 0.037.
total nitrogen, total phosphates and O/R potential ac-
cording to standard procedures (APHA, 1995).
VFAs and phenolic compounds were determined us- 3. Results
ing the same gas±liquid chromatograph equipped with
¯ame ionization detector and 3 mm diameter, 3 m 3.1. E€ect of organic loading rate and hydraulic retention
stainless steel column packed with 10% SE 30 [(60/80 time on reactors performance at di€erent temperatures
mesh), Sigma, Baroda (India)]. Nitrogen served as the
carrier gas at a ¯ow rate of 30 ml min 1 and the column To demonstrate the mutual e€ect of OLR and HRT
temperature ranged from 60 to 250 °C (programmed at on reactors performance at di€erent temperatures,

Table 2
Values of an anaerobic up¯ow ®xed-®lm reactor e‚uent operated at 25 °C with bonechar as carrier material at varying OLR and HRTs under
steady-state conditions
HRT (days)
15 12 9 6a
1
OLR …kg COD m 3 d † 3.60 4.50 6.00 9.00
COD reduction (%) 98  1.20 97  1.50 95  1.50 ND
BOD …g l 1 † 0.50  0.05 0.90  0.15 1.30  0.25 ND
Total biogas production …m3 m 3 d 1 † 2.50  1.00 2.20  0.20 3.70  0.50 1.10  0.10
Total methane …m3 m 3 d 1 † 1.20  0.40 0.98  0.12 1.50  0.25 ND
Total carbon dioxide …m3 m 3 d 1 † 0.12  0.04 0.75  0.12 1.00  0.25 0.85  0.10
Total hydrogen …m3 m 3 d 1 † ND ND ND 0.10  0.03
Methane yield …m3 kg 1 COD d 1 added† 0.20 0.27 0.33 ND
Total alkalinity …as CaCO3 mg l 1 † 700  50 680  50 500  75 ND
pH of the e‚uent 8.70  0.40 8.50  0.20 7.60  0.50 4.60  1.00
Total solids as biomass …mg l 1 † 2.80  0.30 2.30  0.20 5.70  0.50 12.50  0.50
Total VFAs …g l 1 † 0.12  0.01 0.12  0.02 0.50  0.10 15.60  2.00
Total phenolics …mg l 1 † ND 47  05 110  12 150  10
ND ± not detected.
a
Values obtained from samples after complete failure of the reactor as a result of overloading substrate.
68 H. Patel, D. Madamwar / Bioresource Technology 82 (2002) 65±71

reactors were operated under varying HRT and OLR at

16.43  1.50
00.50  0.10

14.27  3.00
each temperature as shown in Tables 2±5. Reactors op-

0.30  0.10
0.10  0.04

3.50  1.50
5.00  1.00
45  3.00

125  50
erating at 25 °C worked eciently at 9 days HRT with an

27.20
OLR of 6 kg COD m 3 d 1 reaching a 95  1:5% COD

ND

ND
ND
1.5a

reduction. Methane yield was 0.333 m3 kg 1 COD d 1

added. Total alkalinity of 500  75 as CaCO3 mg l 1 was
observed. A sudden decrease in total gas production was
observed when HRT was reduced from 9 to 6 days with
11.76  1.50
1.46  0.20

9.40  1.00
1.90  0.55

7.50  0.50
4.40  0.50
0.34  0.50
1700  200
Values of an anaerobic up¯ow ®xed-®lm reactor e‚uent operated at 37 °C with bonechar as carrier material at varying OLR and HRTs under steady-state conditions

95  1.00

an increased level of carbon dioxide (Table 2). The pH of

21.70

the reactors e‚uent decreased to 4:6  1:0, containing

0.45
ND

ND
2.5

15:6  2:0 g 1 VFA and 150  10 mg l 1 phenolics. This

led to souring and failure of the reactors. Reactors op-
erated at 45 and 55 °C exhibited better process stability
with respect to COD reduction and total gas production
1.47  0.50
9.80  1.50
7.50  0.50
2.00  0.70

8.00  0.50
3.90  0.20
0.33  0.50
1614  150
95  1.50

at 6 and 3 day HRT( Tables 4 and 5). At 45 °C a maxi-

mum methane yield of 0:416 m3 kg 1 COD d 1 added
18.10

0.43
ND

ND

was observed at 9 days HRT. Even though better per-

3

formance was observed at 6 days HRT, methane yield

was low when compared to 9 days HRT. Total gas
ranged from 0.009 to 0:330 m3 m 3 d 1 and H2 was de-
1.33  0.30
5.00  1.00
3.80  0.70
1.00  0.50

7.80  0.10
3.90  0.40
0.27  0.05
1700  200

tected at each OLR at 45 °C. The reactors operated at 55

95  0.60

°C performed well at 3 days HRT giving 77  3% COD

9.00

0.43
ND

ND

reduction but methane yield decreased to 0:110 m3 kg 1

6

COD d 1 added in comparison to 0.666 and

0:555 m3 kg 1 COD d 1 at 9 and 6 day HRT, respec-
tively. Here hydrogen gas concentration ranged from
Values obtained from samples after complete failure of the reactor as a result of overloading substrate.
0.34  0.15
3.30  0.50
2.24  0.90
1.00  0.40

8.60  0.30
3.75  1.00
2933  130

0:05  0:02 to 0:20  0:04 m3 m 3 d 1 . Total gas pro-

98  0.50

duction decreased to 4:0  1:5 m3 m 3 d 1 at 3 days

6.00

0.37
ND

ND
ND

HRT from 11:0  1:5 m3 m 3 d 1 at 6 days HRT. At

9

each HRT of reactors operating at 45 and 55 °C the total

gas produced contained a signi®cant quantity of carbon
dioxide, which directly a€ected total alkalinity in the
0.34  0.15
2.50  0.50
1.69  0.60
0.66  0.30

8.30  0.30
3.80  0.50
2300  110

reactors. On further reducing the HRT from 6 to 3 days

98  0.50

in the reactors operating at 45 °C, a drop in COD re-

4.50

0.37
ND

ND
ND

duction to 38 % of wastewater was observed with a VFA

12

concentration of 87  10 g 1 in the e‚uent. During

failure of reactors total gas produced contained only
carbon dioxide …0:650  0:20 m3 m 3 d 1 † and hydrogen
HRT (days)

…0:300  0:10 m3 m 3 d 1 †. Similar observations were

0.34  0.10
2.00  0.40
1.36  0.20
0.50  0.20

8.50  0.20
3.35  1.00
2725  150
98  0.50

noted with the 55 °C reactors on reducing HRT to 2.5

3.60

0.37
ND

ND
ND

day. Sudden souring and failure of reactors was ob-

15

served. The best performance was observed with the 37

°C reactors when stability and wastewater stabilization
Methane yield …m3 kg 1 COD d 1 added†

of the reactors were examined (Table 3). The reactors

Total biogas production …m3 m 3 d 1 †

operated continuously for a longer period of time at 2.5

Total alkalinity …as CaCO3 mg l 1 †
Total carbon dioxide …m3 m 3 d 1 †

days HRT with an organic load of 21.7 kg COD

Total solids as biomass …mg l 1 †

m 3 d 1 . Reactors operated under varying temperatures

Total hydrogen …m3 m 3 d 1 †
Total methane …m3 m 3 d 1 †

exhibited di€erent critical HRTs. Performance of reac-

OLR …kg COD m d †
1

tors was a€ected by reducing HRTs below critical op-

Total phenolics …mg l 1 †

eration at all temperatures.

3

COD reduction (%)

Total VFAs …g l 1 †
pH of the e‚uent

ND ± not detected.

3.2. E€ect of feed pH on reactors performance at di€erent

BOD …g l 1 †

temperatures
Table 3

There are instances where acidic wastewaters are

treated by acclimatized anaerobic organisms to lower
a
 H. Patel, D. Madamwar / Bioresource Technology 82 (2002) 65±71 69

Table 4
Values of an anaerobic up¯ow ®xed-®lm reactor e‚uent operated at 45 °C with bonechar as carrier material at varying OLR and HRTs under
steady-state conditions
HRT (days)
15 12 9 6 3a
OLR …kg COD m 3 d 1 † 3.60 4.50 6.00 9.00 18.10
COD reduction (%) 96  2.00 96  1.50 93  1.50 89  1.50 ND
BOD …g l 1 † 1.10  0.60 1.10  0.40 2.00  0.40 3.30  0.50 ND
Total biogas production …m3 m 3 d 1 † 3.10  0.4 3.50  1.0 5.50  1.0 8.00  0.5 1.00  0.2
Total methane …m3 m 3 d 1 † 1.40  0.20 1.70  0.50 2.50  0.40 3.00  0.20 ND
Total carbon dioxide …m3 m 3 d 1 † 1.00  0.10 0.98  0.30 1.80  0.30 3.00  0.20 0.65  0.20
Total hydrogen …m3 m 3 d 1 † 0.01  0.001 0.17  0.03 0.20  0.10 0.33  0.20 0.30  0.10
Methane yield …m3 kg 1 COD d 1 added† 0.38 0.38 0.41 0.33 ND
Total alkalinity …as CaCO3 mg l 1 † 600  55 400  75 425  25 475  25 ND
pH of the e‚uent 8.80  0.20 8.40  0.20 8.30  0.10 8.30  1.00 4.40  1.00
Total solids as biomass …mg l 1 † 2.20  0.10 2.90  0.50 3.30  0.60 6.60  0.20 58.90  10
Total VFAs …g l 1 † 0.29  0.05 0.32  0.01 1.42  0.14 2.33  0.15 87.00  10
Total phenolics …mg l 1 † ND ND 85  10 150  75 350  50
ND ± not detected.
a
Values obtained from samples after complete failure of the reactor as a result of overloading substrate.

Table 5
Values of an anaerobic up¯ow ®xed-®lm reactor e‚uent operated at 55 °C with bonechar as carrier material at varying OLR and HRTs under
steady-state conditions
HRT (days)
15 12 9 6 3 2.5a
3 1
OLR …kg COD m d † 3.60 4.50 6.00 9.00 18.10 21.70
COD reduction (%) 98  1.00 98  1.00 97  1.00 94  1.00 77  3.00 ND
BOD …g l 1 † 0.45  0.10 0.45  0.10 0.90  0.05 2.00  1.00 7.00  2.00 ND
Total biogas production 2.70  0.50 5.00  1.00 9.00  2.00 11.00  1.50 4.00  1.50 1.20  0.20
…m3 m 3 d 1 †
Total methane …m3 m 3 d 1 † 1.20  0.20 3.00  0.60 4.00  0.90 5.00  0.70 2.00  0.50 ND
Total carbon dioxide …m3 m 3 d 1 † 1.00  0.20 1.60  0.30 1.00  0.27 4.00  0.50 1.50  0.20 0.68  0.5
Total hydrogen …m3 m 3 d 1 † ND 0.05  0.01 0.07  0.01 0.07  0.01 0.05  0.02 0.20  0.04
Methane yield 0.33 0.66 0.66 0.55 0.11 ND
…m3 kg 1 COD d 1 added†
Total alkalinity …as CaCO3 mg l 1 † 550  80 400  50 400  50 45050 400  70 ND
pH of the e‚uent 8.70  0.30 8.50  0.10 8.50  0.30 8.30  0.40 7.00  0.50 5.50  0.50
Total solids as biomass …mg l 1 † 3.00  1.00 4.80  0.50 4.80  0.50 5.20  0.30 14.60  4.00 20.00  5.00
Total VFAs …g l 1 † 0.13  0.02 0.15  0.05 0.37  0.05 0.28  0.09 19.00  5.50 55.00  10.00
Total phenolics …mg l 1 † ND ND ND ND 89  10 230  45
ND ± not detected.
a
Values obtained from samples after complete failure of the reactor as a result of overloading substrate.

pH and harness the alkalizing capabilities of methano-
genic activities by their retention in the bioreactors
(Ramakrishna and Desai, 1997; Patel and Madamwar,
2000). When the e€ect of feed pH (2.5) on performance
of reactors at di€erent temperatures was examined, it
was observed that reactors at temperature of 37 °C
performed best at an HRT of 2.5 days. E‚uent total
alkalinity and pH of reactors were 1700  200 CaCO3
mg l 1 and 7:5  0:5, respectively. Total alkalinity of
e‚uent with reactors operating at 25, 45 and 55 °C
never exceeded 1000 mg l 1 , but remained in the range
400  50 to 700  50 CaCO3 mg l 1 . According to
McCarty (1964) bicarbonate alkalinity in the range
2500±5000 mg l 1 provides sucient bu€ering capacity
so that production of high concentrations of volatile
acids result in only a minimal pH decrease. High carbon
dioxide content in total gas produced was detected at a
feed pH of 2.5 at each OLR, which directly a€ected total
alkalinity of the e‚uent. The reactors working at 25 °C
exhibited alkalinity of 500  75 CaCO3 mg l 1 and pH
of 7:6  0:5 above critical HRT at a feed pH of 2.5.
Reactors working at 45 °C exhibited alkalinity of
475  25 CaCO3 mg l 1 and pH of 8:3  1:0 at a critical
HRT of 6 days whereas reactors at 55 °C exhibited
alkalinity of 400  70 CaCO3 mg l 1 and pH of 7:0 
0:5 at critical HRT of 3 days. The bu€ering capac-
ity apparently was lost due to imbalance between the
ratio of VFA and total bicarbonates produced resulting
in reactors souring on reducing HRT below critical
HRT.
70 H. Patel, D. Madamwar / Bioresource Technology 82 (2002) 65±71
3.3. Reactor failure and recovery
All the reactors at di€erent temperatures showed
di€erent critical HRTs, below which complete digester
failure was realized. Recovery of the reactors required
much longer time periods. Reactors operating at 25 and
45 °C returned to normal operation when fed with fresh
cowdung slurry possessing total alkalinity of
3500  200 CaCO3 mg l 1 and pH of 7.5. Reactors were
not recovered when fed with e‚uent slurry having a pH
of 8.0 and a total alkalinity of 2000  250 CaCO3
mg l 1 . Reactors at 37 and 55 °C returned to normal
operation by feeding the same e‚uent slurry on con-
tinuous basis at an HRT of 12 days for 30±35 days. The
reactor at 25 °C took 45±50 days for recovering while
the reactor at 45 °C required 15±20 days for recovery.
All the reactors at di€erent temperatures performed well
above a critical HRT after recovery.
4. Discussion
The results of these studies demonstrates that reac-
tors operating at 37 °C were stable at an OLR of 21.7 kg
COD m 3 d 1 with a 95  1:0% reduction in COD and
BOD, higher methane yield and stable alkalinity. At 37
°C total gas produced contained much less CO2 in
comparison to the CO2 concentration in reactors
working at 25, 45 and 55 °C, which in turn a€ected total
alkalinity of the reactor. The ratio of CH4 :CO2 observed
at 37 °C ranged from 2.24 to 4.94 compared with that of
1.2±1.8 in reactors at other temperatures. The high CO2
levels in the gaseous phase at elevated temperatures
might be due to the e€ect of the high vapor pressure of
water (Pfe€er, 1974). The addition of water vapor in the
gas reduces the partial pressure of CO2 , which in turn
changes the bicarbonate alkalinity equilibrium.
For a given pH of 7.0, the ratio of bicarbonate ion to
carbonic acid (CO2 ) increases from 4.9 at 30 °C to 5.19
at 60 °C (Pfe€er, 1974). It was also observed that tem-
perature a€ects this equilibrium resulted in a substantial
reduction in alkalinity at higher temperatures. Therefore
more CO2 is lost in the gas phase, because of the de-
crease in CO2 solubility by a factor of 0.85 as temper-
ature shifts form 30 to 60 °C. This is in conjunction with
an increase in the vapor pressure of water by a factor of
4.7, which reduces CO2 considerably (Pfe€er, 1974). In
solution this e€ect is not o€set by the change in the CO2 -
bicarbonate relationship with increasing temperature.
Therefore, a much lower bicarbonate alkalinity will be
associated with a given pH at elevated temperatures
(Pfe€er, 1974). In psychrophilic conditions low bicar-
bonate alkalinity was observed which was due to the
slow growth rate of methanogens. This led to an im-
balance between the ratio of methanogenic organisms
and OLR. To overcome this problem either feed pH
should be increased to the required level or biomass
should be recycled into the reactor to reduce reactor
acidi®cation (McCarty, 1964).
No hydrogen was detected in total gas produced at 25
and 37 °C, however, presence of hydrogen was detected
at all HRTs at 45 and 55 °C, respectively, but did not
a€ect reactor performance above critical HRTs. Below
the critical HRT, total gas produced contained a high
quantity of hydrogen which might be a€ecting reactors
performance. Increasing the loading rate above the
critical HRT, methanogenic organisms apparently were
no longer capable of consuming hydrogen produced by
fermentative bacteria and thus leading to an increased
hydrogen partial pressure. This may have led to accu-
mulation of VFAs, hydrogen and carbon dioxide in the
reactors causing souring and failure (Nyns, 1985).
Failure of the reactors under psychrophilic conditions
during overloading presumes that in this condition the
half velocity constant of substrate utilization has been
reported both to increase (Ks ) and to decrease (Km )
(Kettunen and Rintala, 1997). The number of acido-
genic and hydrogen utilizing methanogens expressed as
logarithm values appears to decrease in linear fashion as
the fermentation temperature decreases. This indicates
that the formation of methane by the acetate-utilizing
methanogens would be insigni®cant compared to the
hydrogen utilizing methanogens (Cha et al., 1997). It
was also reported by Cha et al. (1997) that the number
of acetate-utilizing methanogens at 30 °C are reduced
when the temperature was decreased. In our study op-
timum performance was obtained with the reactors at 37
°C. The reactors were able to withstand a high OLR of
21.7 kg COD m 3 d 1 and HRT of 2.5 days with a feed
pH of 2.5. Below this HRT the reactor failed which
might be due to high biomass accumulation, i.e. in-
creasing bio®lm thickness leading to a decrease in e-
ciency because substrate was not able to penetrate
through the depth of the biomass (Kennedy and Van
Den Berg, 1985). Reactors failure might also be due to
the organic and hydraulic loading rates, which ap-
proached and exceeded the washout rate (i.e. critical
hydraulic retention times) at which the compounds will
be ¯ushed out of the reactors un-utilized, along with
VFA build-up inside the reactors vessel creating adverse
conditions for anaerobes (Kennedy and Van Den Berg,
1985; Cobb and Hill, 1991).
5. Conclusion
Based on the studies presented here, it is apparent
that a certain compensation for changing temperature is
possible among anaerobic bacteria but overall activity is
directly proportional to temperature (Westerman, 1996).
Best performance was observed in mesophilic condi-
tions, from which it was concluded that the optimum
 H. Patel, D. Madamwar / Bioresource Technology 82 (2002) 65±71
temperature for the growth of methanogens under the
conditions used for the reactors in this study appears to
be 37 °C. However reactors could be operated even
under psychrophilic and thermophilic conditions once
the organisms were acclimatized to such temperatures.
Acknowledgements
Thanks are due to University Grants Commission,
New Delhi, for the ®nancial support.
References
American Public Health Association, 1995. Standard Methods for
Examination of Water and Wastewater. American Public Health
Association, Washington, DC.
Britz, T.J., Meyer, L.C., Botes, P.J., 1983. Anaerobic digestion of
petrochemical e‚uent. Biotechnol. Lett. 5, 113±118.
Chou, W.L., Speece, R.E., Siddiqui, R.H., 1978. Acclimation and
degradation of petrochemical wastewater components by methane
fermentation. Biotechnol. Bioeng. Symp. 8, 391±414.
Cha, G.C., Chung, H.K., Chung, J.C., 1997. Suppression of acidogenic
activities due to rapid temperature drop in anaerobic digestion.
Biotechnol. Lett. 19 (5), 461±464.
Cobb, S.A., Hill, D.T., 1991. Volatile fatty acid relationship in
attached growth anaerobic fermenters. Amer. Soc. Agric. Eng. 36,
2564±2571.
Desai, M., Patel, V., Madamwar, D., 1994. E€ect of temperature and
retention time on biomethanation of cheese whey-poultry waste-
cattle dung. Environ. Pollut. 83, 311±315.
Hwu, C.S., Molenaar, G., Gartho€, J., Van Lier, J.B., Lettinga, G.,
1997. Thermophlic high rate anaerobic treatment of wastewater
containing long chain fatty acids: impacts of reactor hydrodynam-
ics. Biotechnol. Lett. 19 (5), 447±451.
Jimeno, A., Bermudez, J.J., Conovas-diaz, M., Manjon, A., Iborra,
J.L., 1990. E€ect of feed procedure and di€erent temperatures on
anaerobic trickling ®lters: Intermediate consumption pro®le. Pro-
cess Biochem. 4, 55±60.
Kennedy, K.J., Van Den Berg, L., 1985. Anaerobic down-¯ow
stationary ®xed-®lm reactors. In: Moo-Young, (Ed.), Compre-
hensive Biotechnology, vol. 4. Pergamon Press, Oxford, pp. 1027±
1049.
Kotsyurbenko, O.R., Nozhevnikova, A.N., Kalyuzhnyi, S.V., Zavar-
zin, G.A., 1993. Methane digestion of cattle manure at low
temperatures. Microbiology 62, 761±772.
Kettunen, R.H., Rintala, J.A., 1997. The e€ect of low temperature (5±
29 °C) and adaptation on the methanogenic activity of biomass.
Appl. Microbiol. Biotechnol. 48, 570±576.
Lettinga, G., 1995. Anaerobic digestion and wastewater treatment
systems. Antonie Van Leeuwenhoek 67, 3±28.
Lepisto, R., Rintala, J., 1996. Conversion of volatile fatty acids in an
extreme thermophilic (76±80 °C) up¯ow anaerobic sludge blanket
reactor. Bioresour. Technol. 56, 221±227.
Lepisto, R., Rintala, J., 1999. Kinetics and characteristics of 70 °C,
VFA growth, UASB granular sludge. Appl. Microbiol. Biotechnol.
52, 730±736.
71
Lettinga, G., Rebac, S., Parshina, S., Nozhevnikova, A., Van, L.,
Jules, B., Stams, A.J.M., 1999. High-rate anaerobic treatment of
wastewater at low temperatures. Appl. Environ. Microbiol. 65,
1696±1702.
McCarty, P.L., 1964. Anaerobic waste treatment fundamentals:
environmental requirements and control. Public Works 10, 123±
126.
Nel, L.H., De Haast, J., Britz, T.J., 1984. Anaerobic digestion of a
petrochemical e‚uent using an up¯ow anaerobic sludge blanket
reactor. Biotechnol. Lett. 6, 741±746.
Nyns, E.J., 1985. In: Rehm, H.J., Reed, G. (Eds.), Biomethanation
Processes, vol. 3. VCH Verlagsgesellschaft GmbH, Germany,
Weinheim, pp. 207±266.
Nel, L.H., Britz, T.J., 1986. The in¯uence of di€erent substrate pH
values on the performance of down¯ow anaerobic ®xed bed reactor
treating a petrochemical e‚uent. Biotechnol. Lett. 8, 293±298.
Pfe€er, J.T., 1974. Temperature e€ects on anaerobic fermentation of
domestic refuse. Biotechnol. Bioeng. 16, 771±787.
Parkin, G.F., Speece, R.E., 1983. Attached versus suspended growth
anaerobic reactors: response to toxic substances. Water Sci.
Technol. 15, 261±289.
Parkin, G.F., Speece, R.E., Yang, C.H.J., Kocher, W.M., 1983.
Response of methane fermentation system to industrial toxicants.
J. Water Pollut. Control Fed. 55, 44±53.
Patel, P., Desai, M., Madamwar, D., 1995. Biomethanation of cheese
whey using anaerobic up¯ow ®xed ®lm reactor. J. Ferment. Bioeng.
79, 398±399.
Patel, P., Patel, C., Madamwar, D., 1998. Anaerobic up¯ow ®xed ®lm
bioreactor for biomethanation of salty cheese whey. Appl.
Biochem. Biotechnol. 22, 1±10.
Patel, H., Madamwar, D., 2000. Biomethanation of low pH petro-
chemical wastewater using up¯ow ®xed-®lm anaerobic bioreactor.
World J. Micro. Biotechnol. 16, 69±75.
Romero, L.I., Sales, D., Cantero, D., Galan, M.A., 1988. Thermo-
philic anaerobic digestion of winery waste (Vinasses): kinetics and
process optimization. Process Biochem. 8, 119±125.
Ramakrishna, C., Desai, J.D., 1997. High rate anaerobic digestion of a
petrochemical wastewater using biomass support particles. World
J. Micro. Biotechnol. 13, 329±334.
Sa¯ey Jr., L.M., Westerman, P.W., 1992. Performance of a low
temperature lagoon digester. Bioresour. Technol. 41, 167±175.
Sharma, S., Ramakrishna, C., Desai, J.D., Bhatt, N.M., 1994.
Anaerobic biodegradation of a petrochemical wastewater using
biomass support particle. Appl. Microbiol. Biotechnol. 40, 768±
771.
Varel, V.H., Hashimoto, A.G., Chen, Y.R., 1980. E€ect of temper-
ature and retention time on methane production from beef cattle
waste. Appl. Environ. Microbiol. 40, 217±222.
Vartak, D.R., Englar, C.R., McFarland, M.J., Ricke, S.C., 1997.
Attached-®lm media performance in psychrophilic anaerobic
treatment of dairy cattle wastewater. Bioresour. Technol. 62, 79±
84.
Wiegant, W.M., Claassen, J.A., Lettinga, G., 1985. Thermophilic
anaerobic digestion of high strength wastewater. Biotechnol.
Bioeng. 27, 1374±1381.
Wheatley, A.D., 1990. Anaerobic digestion-industrial waste treatment.
In: Wheatley, A.D. (Ed.), Anaerobic Digestion: A Waste Treat-
ment Technology. Elsevier, London, pp. 171±213.
Westerman, P., 1996. Temperature regulation of anaerobic degrada-
tion of organic matter. World J. Micro. Biotechnol. 12, 497±503.