Resources, Conservation and Recycling 40 (2003) 111

The effect of organic loading rate on the

performance of UASB reactor treating
slaughterhouse efuent
Ayoob Torkian

, A. Eqbali, S.J. Hashemian

Institute of Water and Energy, Sharif University of Technology, P.O. Box 11365-688,
Azadi Avenue, Tehran, Iran
Received 24 January 2002; accepted 24 January 2003
Abstract
Organic loading rate (OLR) is an important parameter signicantly affecting microbial ecology and
characteristics of UASB systems. In this study, UASB performance was evaluated in a 1000 1 reactor
receiving feed from a traditional medium-size slaughterhouse. The initial seed for granules formed
earlier was froma mesophilic municipal anaerobic digester sludge with a VSS content of 29 g l
1
. The
temperature of inuent was adjusted by an inline thermostat around 33

C. The reactor was started

with an OLR of 5 kg SCODm
3
d
1
with gradual increase to 10 kg SCOD m
3
d
1
over a 2-week
period. Examination of VSS data showed that on the average 89.3 11.3% of bioparticle mass was
present at the lower 30% of the reactor height. Under steady state conditions, experiments were con-
ducted at OLRs of between 13 and 39 kg SCOD m
3
d
1
and hydraulic retention times (HRT) of
27 h. Removal efciencies in the range of 7590% were achieved at feed SCOD concentrations of
30004500 mg l
1
. A reduction in removal efciency to as low as 67% could have been related to a
combined effect of high OLR and low HRT. Up to 300 l of methane were produced per kilogram of
SCOD removed at OLR values of less than 30 kg COD m
3
d
1
but methane production rate seemed
to decline to below200 at higher OLRvalues. No sign of cell washout was observed at high OLRs and
sludge loading rates (SLR) of up to 2.7 kg SCOD kg
1
VSS d
1
. Elimination capacity of the reactor
consistently increased from 9 to 25 kg SCOD m
3
d
1
corresponding to 12 kg SCOD kg
1
VSS d
1
.
Solids retention time (SRT) calculations for the reactor indicated a range of 3.3 days at high upowve-
locity of 1 mh
1
to 60.3 days at lowupowvelocity of 0.33 mh
1
s during different phases of the study.
2003 Elsevier B.V. All rights reserved.
Keywords: Upow anaerobic sludge blanket; Industrial wastewater; Slaughterhouse; Upow velocity; Anaerobic
treatment

Corresponding author. Tel.: +98-21-601-5118; fax: +98-21-601-2983.

E-mail address: torkian@sharif.edu (A. Torkian).
0921-3449/$ see front matter 2003 Elsevier B.V. All rights reserved.
doi:10.1016/S0921-3449(03)00021-1
2 A. Torkian et al. / Resources, Conservation and Recycling 40 (2003) 111
1. Introduction
The slaughterhouse industry poses a signicant environmental impact by discharging ef-
uent to receiving waters containing high concentration of biodegrad-able organic matter.
Aerobic processes are not regarded as a suitable treatment option because of high energy
requirements for aeration, limitations in liquid-phase oxygen transfer rates, and large quan-
tities of sludge production. Traditional anaerobic processes are also limited by low rates of
organic matter removal, long hydraulic retention times (HRT), accumulation of excessive
residual organic matter and intermediate products, and large reactor volume requirements.
Recent developments in anaerobic treatment processes, especially high retention of biomass
in the reactor, has made it possible to decouple solids retention time (SRT) and hydraulic
residence time in high-rate anaerobic reactors. This has resulted in increased treatment ef-
ciency of these processes and gradual but steady improvement of the common perception
that anaerobic processes are not suitable for treatment of various industrial efuents.
The upow anaerobic sludge blanket (UASB) process is one of the recently developed
high-rate systems. It has been widely adopted for treatment of medium to high-strength
industrial wastewaters (Lettinga and Hulshoff Pol, 1991; Fang et al., 1995). Recent re-
search studies indicate feasibility of this process to treat domestic efuents as well (Behling
et al., 1997; Singh and Viraraghavan, 2000). The key feature of this system is the micro-
bial aggregation into a symbiotic multilayer structure called a granule. Improved process
knowledge and operational details on formation and retainment of stable granules has made
high loading possible, resulting in a more sustainable operation of these systems.
Awide range of organic and hydraulic loading rates has been reported in the literature for
UASBreactors, depending on the substrate used and the quality and quantity of the microbial
community. Syutsubo et al. (1997) reported a COD loading of 30 kg COD m
3
d
1
with a
CODremoval efciency of 85%at sludge loading rates (SLRs) of up to 3.7 g CODg
1
VSS
d
1
for thermophilic reactors (Syutsubo et al., 1998). Organic loading rates (OLR) of up to
104 kg COD m
3
d
1
have been reported for anaerobic digestion of sugar substrate under
thermophilic conditions (Wiegant and Lettinga, 1985). According to Soto et al. (1997),
excellent stability and high treatment efciency was achieved with hydraulic residence
times as low as 2 h at an OLR of 6 kg COD m
3
d
1
, the percent COD removals being 95%
(30

C) and 92% (20

C).
Slaughterhouse wastewater contains high amounts of organic matter with a soluble frac-
tion in the range of 4060%. The suspended and colloidal components in the form of
fats, proteins, and cellulose can have an adverse impact on the performance of UASB re-
actors, leading to deterioration of the microbial activity and washout of active biomass
(Lettinga et al., 1997; Nez and Martnez, 1999). This may limit the operation to OLRs of
46 kg COD m
3
d
1
(Lettinga and Hulshoff Pol, 1991). Ruiz et al. (1997) reported sludge
oatation and increased efuent solids concentration at OLR values higher than 5 kg COD
m
3
d
1
. Others (Sayed et al., 1988; Sayed and De Zeeuw, 1988) have shown satisfactory
treatment of slaughterhouse efuent with OLR values as high as 11 kg COD m
3
d
1
at a
process temperature of 30

C; Borja and Banks (1994) reported COD removal efciencies

of 6499% at OLR values of 1217 kg COD m
3
d
1
. Higher OLR values of up to 45 kg
COD m
3
d
1
have been reported only for hybrid reactors using a combination of UASB
reactor and a bentonite packing as a biomass support (Borja et al., 1995).
A. Torkian et al. / Resources, Conservation and Recycling 40 (2003) 111 3
In this study, the effect of loading rate on UASB reactor treating slaughterhouse efuent
was investigated. This is an important parameter and only limited information is available
about the steady-state performance of UASB reactors under high OLRs. Biomass gradi-
ent along the height of reactor and methane production rate during different operational
conditions were also examined.
2. Materials and methods
2.1. Experimental setup
A 1000 l effective volume square (50 50 cm) Plexiglas pilot used in this study is
shown in Fig. 1. It was set up downstream of a medium-sized traditional slaughterhouse.
A perforated piping system was used at the bottom of the reactor to ensure homogenous
distribution of ow into the reactor. Nine sampling ports (20 and 30 cm apart at bottom
and top, respectively) were provided to quantify sludge characteristic at different elevations
along the reactor. The temperature of inuent was adjusted by an inline thermostat prior to
reactor entry. No recirculation of efuent was practiced.
2.2. Feed
The wastewater stream from a traditional slaughterhouse used in this study consisted of
efuent from a combination of several stages. It included blood from killing operations,
wash waters from stomach and intestines, and wastewater from the refrigerated chambers
Fig. 1. Schematic diagram of UASB system (1, feed tank; 2, ow control weir; 3, recycle pump; 4, inuent
distribution; 5, sampling taps; 6, water seal; 7, gas meter). See text for operational details.
4 A. Torkian et al. / Resources, Conservation and Recycling 40 (2003) 111
Table 1
Wastewater characteristics of UASB reactor at different periods of study
Parameter Range Average Std. Dev.
BOD
5
(mg
l1
) 9141917 1748 541
SCOD (mg l
1
) 22584956 3799 429
TCOD (mg l
1
) 326514285 6037 1092
PPO
4
3
(mg l
1
) 726 17 12
NNH
3
(mg l
1
) 35104 89 50
Temperature (

C) 2736 33.3 2.8

PH 6.87.8 7.2 0.3
Alkalinity as CaCO
3
(mg l
1
) 12081713 1351 181
VFA as acetic acid (mg l
1
) 309565 440 124
and toilets. There was no separation of efuent from these operations and because of the
inherent nature of the process, characteristics varied at different times (Table 1). Addition of
nutrients was not deemed necessary since wastewater characteristics indicated an adequate
concentration of essential proteins and trace elements. No dilution or recycling of feed was
made in the beginning or at any of the phases of the study.
2.3. Operation
The reactor contained granulated sludge formed previously in the reactor. The initial
seed was from a mesophilic municipal anaerobic digester sludge with a VSS content of
29 g l
1
. The temperature of inuent was adjusted by an inline thermostat prior to reactor
entry. Slaughterhouse efuent was pumped into a reservoir from the main slaughterhouse
sewer containing composite efuent from different units. After separating inert particles in
a cyclonic grit chamber, efuent was pumped into a container at the top and then fed by
gravity into the inuent distribution line of thereactor.
Two schemes of operation were selected. In the rst three phases of the study, feed
reservoir was lled at different times during the day to allow different concentrations to
be investigated at constant HRT. In the subsequent two phases, both OLR and HRT were
changed simultaneously by increasing inow to the reactor. Temperature was maintained
around 33

C. There was no need to externally regulate pH of the reactor since it remained

relatively constant throughout the study period.
2.4. Analytical methods
Routine analyses including soluble (ltered sample with a 0.45 m pore size glass mi-
crober lter) and total BOD
5
and COD, alkalinity, nitrogen, and phosphorus were per-
formed using procedures outlined in Standard Methods (APHA, 1985). Samples were cen-
trifuged prior to volatile fatty acid analysis using distillation method. Gas evolution was
measured by a cumulative gas ow meter located downstream of a water trap and analyzed
by Schimadzu (5A with molecular sieve and carbon active columns and FID and ECP de-
tectors) gas chromatograph. Most of the parameters were monitored daily during the start
up phase and every other day during the normal operations.
A. Torkian et al. / Resources, Conservation and Recycling 40 (2003) 111 5
2.5. Experimental design
The experimental protocol was designed to examine the effect of different OLRs on the
operational (e.g. efciency of COD removal) and performance (e.g. volumetric and micro-
bial elimination capacity as dened in the next section below) indicators. All experiments
were performed under steady state conditions. The attainment of the steady state was veried
by checking whether the mean of the efuent characteristics for the last two measurements
done within 5HRT were remaining relatively constant. All the performance and operation
results reported are the average values of at least two measurement data.
2.6. Operational and performance parameters
Operational and performance parameters include OLR, SLR, elimination capacity, and
detention time. Loading rates can be looked at from the pollution indicator, empty reactor
bed volume, and microbial mass. OLR takes into account the liquid ow rate and contam-
inant concentration and is dened as the mass of pollutant introduced in a unit volume of
UASB reactor per unit time (e.g. kg COD m
3
s
1
). As such, this parameter integrates
reactor characteristics, operational characteristics, and bacterial mass and activity into the
volume of media. SLRor food to microorganismratio (F/M) integrates contaminant concen-
tration and microbial mass and is the mass of pollutant applied to a unit mass of microbial
mass per unit time (e.g. kg COD kg
1
VSS d
1
).
Elimination capacity is related to OLR and SLR in that it is dened as the fraction of the
organic load biodegraded in a unit volume of the UASB reactor or a unit mass of microbial
mass. This parameter can be expressed either volumetrically (EC
V
, kg pollutant removed
per unit volume of reactor per day) or on the basis of microbial mass (EC
m
, kg pollutant
removed per unit mass of microorganisms in the reactor per day).
Methanogenic activity (MA) can be expressed on the basis of pollutant (liter biogas
produced per unit mass of pollutant removed, MA
scod
) or on the basis of microbial mass
(liter biogas produced per unit mass of microbial population, MA
vss
).
Mass loading rate (kg m
3
d
1
), SLR (kg kg
1
d
1
), and elimination capacity
(kg m
3
d
1
or kg kg
1
VSS d
1
) were determined using the relationships between in-
uent and efuent contaminant concentration, efuent ow rate, the effective volume of
UASB reactor, and applying appropriate conversion factors as follows:
OLR =

Q
V
r

C
in
(1)
SLR = Q

C
in
VSS

(2)
EC
V
=

Q
V
r

(C
in
C
out
) (3)
EC
m
=
EC
V
VSS
(4)
6 A. Torkian et al. / Resources, Conservation and Recycling 40 (2003) 111
MA
scod
=
V
CH
4
Q(C
in
C
out
)
(5)
MS
VSS
=
V
CH
4
Q(VSS)
(6)
where Qis the efuent owrate (m
3
h
1
); V
r
, the effective volume of reactor bed (m
3
); VSS,
the microbial concentration of the reactor (mg VSS l
1
), V
CH
4
, the volume of biogas pro-
duced per day (l d
1
); and C
in
and C
out
are the contaminant concentrations (mg SCODl
1
)
in the inuent and efuent stream, respectively.
3. Results and discussion
3.1. Startup
The startup of the reactor was rapid because the system had been adapted to the slaugh-
terhouse efuent previously. The reactor was started with an OLR of 5 kg SCOD m
3
d
1
to keep the initial loading rate below approximately 50% of the intended loading after the
start-up period (Lettinga et al., 1997). The loading rate was gradually increased over a
2-week period to 10 kg SCOD m
3
d
1
.
3.2. Steady state performance
The total and soluble COD of the feed and of the efuent during the operation period,
and the results for different organic and hydraulic loading rates along with performance
indicators are presented in Table 2.
3.2.1. Removal efciency
The performance of UASB reactor based on soluble COD removals at various upow
velocities and OLRs is shown in Fig. 2. At the initial three phases of the study, V
up
was maintained relatively constant at 0.330.35 mh
1
while OLR was increased from
around 10 to 18 kg SCOD m
3
d
1
. As illustrated in the gure, SCOD removal efcien-
cies showed an increasing trend from a low 62% to a maximum of 92%. At the begin-
ning of each phase where OLR was increased, there was a corresponding decrease in re-
moval efciency but the system recovered shortly and adapted to the new conditions with
time.
At the beginning of phase 4 of the study, OLR was increased to 27 kg SCOD m
3
d
1
.
Upowvelocity was also increased to 0.57 mh
1
to further promote the selective process in
the cultivation of more active biomass (Campos and Anderson, 1992). The system behavior
was similar to earlier stages in that a transient decrease in performance was observed but
the system performance reached the same conditions existing before the change. The 50%
increase in OLRand 80%increase in V
up
did not seemto have any adverse effect on organics
removal and SCOD removal efciency reached 93%.
In the next phase of study, OLR was increased another 30% to above 40 kg SCOD
m
3
d
1
and V
up
to 1 mh
1
. Removal efciency was drastically decreased to below 70%
A. Torkian et al. / Resources, Conservation and Recycling 40 (2003) 111 7
Table 2
Summary of the conditions during the operation period of the UASB reactor
Variable Unit Phase of study
1 2 3 4 5
Time day 138 3964 6591 92112 113136
Upow velocity, V
up
Mh
1
0.33 0.34 0.35 0.57 1.0
Hydraulic res-
idence time,
HRT
H 7.1 6.8 6.7 4.1 2.3
SRT day 60.3 23.4 14.0 14.4 3.3
SCOD in Mg l
1
3143 661 3695 662 4153 364 4288 564 3290 722
TCOD in Mg l
1
8201 3937 5719 1280 5256 589 5495 622 5514 1469
TSS Gl
1
11.6 2.2 10.9 6.8 11.6 1.3 12.8 1 18.3 1.2
VSS g l
1
10.2 1.9 9.9 6.7 10.5 1 11.9 1.2 14.9 0.9
VSS out g l
1
0.09 0.09 0.16 0.13 0.36 0.20 0.24 0.14 0.59 0.36
SCOD removal % 76 9 75 12 85 6 85 8 68 8
TCOD removal % 78 14 73 11 77 15 83 7 68 10
OLR kg SCOD m
3
d
1
13 2.9 16.7 3.3 17.4 1.1 27.4 4.8 39.5 9
SLR kg SCOD kg
1
VSS d
1
1.3 0.2 1.7 0.3 1.7 0.1 2.4 0.3 2.7 0.6
Elimination ca-
pacity, EC
v
kg SCOD m
3
d
1
9.5 1.8 12.6 3.6 15 1.7 25 4.8 27 6.9
Elimination ca-
pacity, EC
M
kg SCOD kg
1
VSS d
1
1.0 0.2 1.3 0.4 1.4 0.2 2.1 0.4 1.8 0.5
Methanogenic
activity,
MA
vss
l kg
1
VSS 222 32 347 66 458 54 464 58 395 68
Methanogenic
activity,
MA
scod
l kg
1
SCOD 213 21 254 54 283 23 201 38 199 36
and there was no indication that a recovery was to ensue. As a result, OLR was decreased
to try a more gradual increasing trend but V
up
was maintained around 1 mh
1
. Variation of
OLR in the 3040 kg SCOD m
3
d
1
did not improve SCOD removal efciencies from the
65 to 68% range. As such, OLR value of about 30 kg SCOD m
3
d
1
was regarded as the
upper limit for satisfactory performance for this type of wastewater under the conditions
of this study. Considering the fact that HRT at this phase was low at 2.3 h, the decrease
in performance could have also been attributed to insufcient time available for substrate
transfer from the liquid to biomass.
The results obtained in this study showed better performance when compared with the
values of 92% SCOD removal at 5.2 kg SCOD m
3
d
1
and HRT of 1.2 day (Ruiz et al.,
1997) and 93.4% COD removal at 20.8 kg SCOD m
3
d
1
and HRT of 0.5 days for this
type of wastewater under similar operating conditions and 87% SCOD removal at 30 and
HRT of 7.2 h for alcohol distillery wastewater under thermophilic conditions (Syutsubo
et al., 1997). This could have been due to a combination of factors including lack of blood
separation in the slaughterhouse operations, high solids content, and long adaptation of
granules to the slaughter-house efuent prior to this study.
8 A. Torkian et al. / Resources, Conservation and Recycling 40 (2003) 111
Fig. 2. Variation of SCOD removal efciencies (%) at different OLRs (kg SCOD m
3
d
1
) and upow
velocities (mh
1
).
3.2.2. Sludge loading rate
Fig. 3 illustrates performance of the reactor at different SLRs. As shown in the gure, the
SLR practiced in this study ranging from 1 to above 2.5. This was in line with the recom-
mended range of 0.11 kg COD kg
1
VSS d
1
for anaerobic processes (Ndon and Dague,
1997). Another important aspect of the performance is to prevent anaerobic microorgan-
ism washout and to provide a margin of safety under transient inhibitory conditions in the
reactor. This is assured by maintaining a minimum value of biological SRT even at low
hydraulic residence times. SRT calculations for the reactor indicated a range of 3.3 days at
high upow velocity of 1 mh
1
to 60.3 days at low upow velocity of 0.33 mh
1
s during
different phases of the study. This was within the recommended range of 410 days to
prevent washout of hydrolytic anaerobic bacteria (Eastman and Feguson, 1981) for cases
where hydrolysis of insoluble organic matter is the rate-limiting step (Parkin and Owen,
1986) and 2.55 days for soluble wastewaters containing acetate as the primary organic
(Stronach et al., 1986) constituent.
3.2.3. Methane production
Fig. 4 illustrates MA based on microbial capacity and SCOD conversion. The gure
shows more uctuation of methane produced on the basis of unit SCOD removed than unit
VSS mass. This may be due to seasonal variability of biological degradability of efuent
and potential presence of various organic and inorganic materials inhibiting treatment per-
formance (Kroeker, 1979). Table 1 shows a steady increase in methane production capacity
up to an OLR of 27 kg SCOD m
3
d
1
. As OLR was increased, MA
vss
(at 25

C and 1 atm)
increased to a maximum of 283 l kg
1
VSS at organic load of 17.4 kg SCOD m
3
d
1
.
From there on, incremental increase in MA
vss
declined and eventually decreases to 199 l
CH
4
kg
1
VSS d
1
at an OLR value of 39.5 kg SCOD m
3
d
1
corresponding to SLR
A. Torkian et al. / Resources, Conservation and Recycling 40 (2003) 111 9
Fig. 3. SLRs (kg COD kg
1
VSSd
1
) applied and microbial elimination capacities (kg SCODkg
1
VSS d
1
) at
different HRTs (days).
of 2.7 kg SCOD kg
1
VSS d
1
. This apparent instability was manifested in Fig. 4 by the
widening and erratic behavior of MA
vss
and MA
scod
curves at high organic loads.
Attachment of gas bubbles is a usual problem of ordinary UASB systems at high OLR
values leading to biomass suspension and cell washout as methane production rate increases.
Even though the system experienced a lower efciency at high OLR values, there was no
drastic increase in efuent VSS. The maximum efuent solids concentration of 590 mg
VSS l
1
observed at the highest OLR studied, was around 3.3% of the reactor biomass
concentration. The fact that no special gas separation system was used in the enlarged
Fig. 4. Methane production per unit mass of biomass (l CH
4
kg
1
VSSd
1
) and SCOD removal
(l CH
4
kg
1
SCOD) throughout the study period.
10 A. Torkian et al. / Resources, Conservation and Recycling 40 (2003) 111
Fig. 5. Prole of solids content along the reactor height.
settling zone suggested good granule stability and characteristics. This was in conformity
with the data on the prole of sludge behavior along the reactor height during the study.
3.2.4. Sludge gradient along the reactor height
As illustrated in Fig. 5, a distinct stratication of solids was maintained through the
experimental period with larger solids (granules) settling down to lower part of the reaction
zone and smaller ones in the upper part. Solids concentration at sampling ports 1 and 2
had a range of 4151 and 3252 mg l
1
, respectively. Examination of VSS data showed
that on the average 89.3 11.3% of bioparticle mass was present at the lower 30% of the
reactor height (sampling ports 13) and the remaining aggregates were suspended due to
the mixing by owing liquor and rising gas bubbles. Reports in the literature indicate that
cell washout is attributed exclusively to sludge blanket erosion (De Zeeuw, 1987) that is
selective for well-aggregated granules. The combined effect of high substrate load and good
granule characteristics along with the physical selection brought about by high OLR and
upow velocity played a positive role in maintaining stable and efcient solids in thelower
part of the reactor.
4. Conclusions
The results of this study showed slaughterhouse wastewater can be satisfactorily treated
by means of high-rate anaerobic processes, specically with the use of USAB reactor. High
SCODremovals of between 75 and 90%at OLRs of 1330 kg CODm
3
d
1
were achieved
in this study. Indication of erratic behavior was observed at organic loads higher than 30 kg
CODm
3
d
1
. There was no sludge washout even at OLRvalues above 30 kg CODm
3
d
1
at HRT values as low as 2.3 h. Methane yields of 200280 l CH
4
kg
1
SCOD
removed
were
in the same order of magnitude as the rates achieved in earlier studies.
A. Torkian et al. / Resources, Conservation and Recycling 40 (2003) 111 11
References
APHA. Standard Methods for the Examination of Water and Wastewater, 16th ed. Washington, DC: American
Public Health Association.
Behling E, Diaz A, Colina G, Herrera M, Gutierrez E, Chacin E, Fernandez N, Forster CF. Domestic wastewater
treatment using a UASB reactor. Bioresour Technol 1997;61(3):23945.
Borja R, Banks CJ. Performance and kinetics of an upow anaerobic sludge blanket (UASB) reactor treating
slaughterhouse wastewater. J Environ Sci Health 1994;A29:206385.
Borja R, Banks CJ, Wang Z. Performance of a hybrid anaerobic reactor, combining a sludge blanket and a lter,
treating slaughterhouse wastewater. Appl Microbiol Biotechnol 1995;43:3517.
Campos CMM, Anderson GK. Effect of liquid upow velocity and the substrate concentration on the start-up and
steady-state periods of lab-scale UASB reactors. Water Sci Technol 1992;25(7):4150.
De Zeeuw WJ. Granular sludge in UASB reactors. In: Lettinga G, Zehnder AJB, Grotenhuis TC, Hulshoff Pol
LW, editors. Proceedings of the GASMAT-Workshop, Granular Anaerobic Sludge: Microbiology and Tech-
nology. Wageningen, Netherlands: Pudoc, 1987:13245.
Eastman JA, Feguson JF. Solubilization of particulate organic carbon during the acid phase of anaerobic digestion.
J Water Pollut Control Fed 1981;53(3):35266.
Fang HHP, Li YY, Chui HK. UASB treatment of wastewater with concentrated mixed VFA. ASCE J Environ Eng
1995;121(2):15360.
Kroeker EJ. Anaerobic treatment process stability. J Water Pollut Control Fed 1979;51:71821.
Lettinga G, Hulshoff Pol LW. UASB-process design for various types of wastewaters. Water Sci Technol
1991;24(8):87107.
Lettinga G, Field J, van Lier J, Zeeman G, Hulshoff Pol LW. Advanced anaerobic wastewater treatment in the
near future. Wat Sci Technol 1997;35(10):512.
Ndon UJ, Dague RR. Effects of temperature and hydraulic retention time on anaerobic SBR treatment of
low-strength wastewater. Water Res 1997;31(10):245566.
Nez LA, Martnez B. Anaerobic treatment of slaughterhouse wastewater in an expanded granular sludge bed
(EGSB) reactor. Water Sci Technol 1999;40(8):99106.
Parkin GF, Owen WF. Fundamentals of anaerobic digestion of wastewater sludges. J Environ Eng Div Am Soc
Civil Eng 1986;112:867920.
Ruiz I, Veiga MC, de Santiago P, Blzquez R. Treatment of slaughterhouse wastewater in a UASB reactor and an
anaerobic lter. Bioresour Technol 1997;60:2518.
Sayed S, De Zeeuw W. The performance of a continuously operated occulent sludge UASB reactor with slaugh-
terhouse wastewater. Biol Wastes 1988;24:199212.
Sayed S, Van Campen L, Lettinga G. Anaerobic treatment of slaughterhouse waste using a granular sludge UASB
reactor. Biol Wastes 1988;21:1128.
Singh KS, Viraraghavan T. Performance of UASB reactor at 6 to 32

C in municipal wastewater treatment. Water

Quality Res J Can 2000;35(1):11324.
Soto M, Ligero P, Vega A, Ruiz I, Veiga MC, Blazquez R. Sludge granulation in UASB digesters treat-
ing low strength wastewaters at mesophilic and psychrophilic temperatures. Environ Technol 1997;18(11):
113341.
Stronach SM, Rudd T, Lester JN. Anaerobic Digestion Processes in Industrial Wastewater Treatment, Berlin:
Springer, 1986.
Syutsubo K, Harada H, Ohashi A, Suzuki H. Effective start-up of thermophilic UASB reactor by seeding
mesophilically-grown granular sludge. Water Sci Technol 1997;36(67):3918.
Syutsubo K, Harada H, Ohashi A. Granulation and sludge retainement during start-up of a thermophilic-UASB
reactor. Water Sci Technol 1998;38(89 part):34957.
Wiegant WM, Lettinga G. Thermophilic anaerobic digestion of sugars in an upow anaerobic sludge blanket
reactors. Biotechnol Bioeng 1985;27:16037.