Influence of the agitation rate on the treatment of partially soluble wastewater in anaerobic sequencing batch biofilm reactor

ARTICLE IN PRESS

0043-1354/$ - se

doi:10.1016/j.w

�Correspond16-273-9550.

E-mail addr1Present add

Grande do Sul,

Water Research 38 (2004) 4117–4124

www.elsevier.com/locate/watres

Influence of the agitation rate on the treatment of partiallysoluble wastewater in anaerobic sequencing batch

biofilm reactor

Samantha Cristina Pinhoa,1, Suzana Maria Ratuszneib, Jose AlbertoDomingues Rodriguesb, Eugenio Forestia, Marcelo Zaiata,�

aDepartamento de Hidraulica e Saneamento, Escola de Engenharia de Sao Carlos (EESC), Universidade de Sao Paulo (USP),

Av. Trabalhador Sao-Carlense 400, CEP 13566-590, Sao Carlos, Sao Paulo, BrazilbDepartamento de Engenharia Quımica e de Alimentos, Escola de Engenharia Maua, Instituto Maua de Tecnologia (IMT),

Prac-a Maua 1, Sao Caetano do Sul, Sao Paulo, Brazil

Received 26 March 2004; received in revised form 13 August 2004; accepted 18 August 2004

Abstract

This work reports on the influence of the agitation rate on the organic matter degradation in an anaerobic sequencing

batch reactor, containing biomass immobilized on 3 cm cubic polyurethane matrices, stirred mechanically and fed with

partially soluble soymilk substrate with mean chemical oxygen demand (COD) of 974770mg l�1. Hydrodynamic

studies informed on the homogenization time under agitagion rates from 500 to 1100 rpm provided by three propeller

impellers. It occurred very quickly compared to the total cycle time. The results showed that agitation provided good

mixing and improved the overall organic matter consumption rates. A modified first-order kinetic model represented

adequately the data in the entire range of agitation rate. The apparent first-order kinetic constant for suspended COD

rose approximately 360% when the agitation rate was changed from 500 to 900 rpm, whereas the apparent first-order

kinetic constant for soluble COD did not vary significantly.

r 2004 Elsevier Ltd. All rights reserved.

Keywords: Anaerobic process; Anaerobic sequencing batch reactor (ASBR); Anaerobic sequencing batch biofilm reactor (ASBBR);

Mechanical agitation; Partially soluble wastewater; Particulate organic matter

1. Introduction

Anaerobic sequencing batch reactors have been

extensively studied for presenting several operational

e front matter r 2004 Elsevier Ltd. All rights reserve

atres.2004.08.015

ing author. Tel.: +55-16-273-9546; fax: +55-

ess: [email protected] (M. Zaiat).

ress: UERGS—Universidade Estadual do Rio

Unidade Caxias do Sul, Av. Julio de Castilhos

advantages when compared to other anaerobic reactors

such as good biomass retention, efficient operational

control and no need of biomass separator. These

reactors are especially useful in studies for elucidating

several fundamental aspects of anaerobic degradation

due to the feasibility of their instrumentation that

provide efficient process control. In this sense, many

configurations have been tested and the parameters that

significantly affect the performance of these reactors

have been evaluated such as the food/microorganism

ratio (Dague et al., 1992; Reyes III and Dague, 1995),

d.

ARTICLE IN PRESS

Nomenclature

BA bicarbonate alkalinity (mg CaCO3 l�1)

COD chemical oxygen demand (mg l�1)

CODF filtered chemical oxygen demand (mg l�1)

CODSS suspended chemical oxygen demand

(mg l�1)

CODT total chemical oxygen demand (mg l�1)

k0 and k00 adjustment parameters in the modified

hyperbolic function (rpm�1)

k1app apparent first-order kinetic constant (h�1)

k1Fapp apparent first-order kinetic constant for

filtered COD (h�1)

k1SSapp apparent first-order kinetic constant for

suspended COD (h�1)

k1Tapp apparent first-order kinetic constant for

total COD (h�1)

N agitation rate (rpm)

pH0 initial pH (hydrodynamic study)

pHf final pH (hydrodynamic study)

RSoF initial filtered reaction rate (mg

COD l�1 h�1)

RSoSS initial suspended reaction rate (mg

COD l�1 h�1)

RSoT initial total reaction rate (mg COD l�1 h�1)

SoF initial filtered substrate concentration (mg

COD l�1)

SoSS initial suspended substrate concentration

(mg COD l�1)

SoT initial total substrate concentration (mg

COD l�1)

SR residual substrate concentration (mg l�1)

SRFmax maximum residual filtered substrate concen-

tration (mg l�1)

SRTmax maximum residual total substrate concen-

tration (mg l�1)

t0 delay hydrodynamic time (s)

TVA total volatile acids concentration (mg l�1)

t first-order constant (hydrodynamic study)

(s)

S.C. Pinho et al. / Water Research 38 (2004) 4117–41244118

the reactor’s geometry (Sung and Dague, 1995), the feed

strategy (Droste and Masse, 1995), the agitation or

mixing conditions, and the biomass immobilization (in

granules or biofilms) (Rodrigues et al., 2003; Ratusznei

et al., 2000,2001; Camargo et al., 2002).

A typical operational cycle of an anaerobic sequen-

cing batch reactor comprises of four steps: feed,

reaction, settling, and liquid withdrawal. It was pre-

viously verified that agitation occurring during the

reaction step improves the mass transfer fluxes, thus

improving the overall substrate consumption rate

(Ratusznei et al., 2001). Based on the concept that the

homogenization by mixing is a crucial feature affecting

most bioprocesses, some authors emphasize that the

overall performance of a bioreactor is determined by the

interaction of physical (fluid hydrodynamics, for in-

stance) and biological-dependent parameters (growth

factors and nutrient demand) (Nienow, 1998).

The immobilization of anaerobic biomass on inert

supports represents an important contribution to the

improvement of anaerobic sequencing batch reactor

(ASBR) performance, since it eliminates the settling step

and reduces the total cycle time (Zaiat et al., 2001).

Some authors tested this form of immobilization and

have obtained good results for organic matter degrada-

tion (Hirl and Irvine, 1997; Ratusznei et al., 2000, 2001;

Camargo et al., 2002; Cubas et al., 2004).

Some studies have demonstrated that organic sus-

pended solids can represent the major fraction (up to

85%) of the total chemical oxygen demand (COD) in

domestic sewage (Zeeman et al., 1997), as well as in

many other types of wastewaters, such as those from

agro industrial activities. In such cases, hydrolysis,

which is an essential step in the overall anaerobic

organic matter conversion, is the limiting and proble-

matic step (Eastman and Ferguson, 1981; Hobson,

1987). Therefore, if dissolution of particulate material

could be accelaerated, the overall performance of the

anaerobic process in terms of time and efficiency would

be, probably, expressively improved. In fact, ASBR

containing flocculent or granular biomass has already

been employed to treat wastewaters with a high

percentage of organic suspended solids, namely swine

manure (Dague et al., 1992; Droste and Masse, 1995;

Masse et al., 1997), leachate (Hollopeter and Dague,

1994; Timur and Ozturk, 1999), dairy wastewater

(Dugba and Zhang, 1999) and slaughterhouse waste-

water (Masse and Masse, 2001).

The configuration of anaerobic batch reactor contain-

ing immobilized biomass on polyurethane foam pro-

vided with mechanical stirring has been shown to be

very suitable for the treatment of low-strength

wastewater with low suspended solids concentration

(Ratusznei et al., 2000, 2001, 2003; Cubas et al.,

2004). However, this reactor may also constitute an

alternative for the treatment of partially soluble waste-

waters, because efficient mechanical stirring would

improve the particle suspension and accelerate

the solubilization of suspended organic matter (Pinho

et al., 2004). This acceleration would occur due to the

following factors: (1) it also helps the shearing of larger

particles into smaller ones; (2) stirring improves the

solids–microbial contact, and even the contact between

the solids and the extra-cellular enzymes; (3) it may help

ARTICLE IN PRESSS.C. Pinho et al. / Water Research 38 (2004) 4117–4124 4119

remove soluble concentration gradients around hydro-

lyzing solids.

The hydrodynamic patterns in this reactor are

certainly quite complex, due to the presence of the

polyurethane foam matrices. However, these fluxes that

must be pushed into the foam probably improves the

contact between the biomass and the substrate. In

addition, to promote an efficient mass transfer between

phases, the solids have to be maintained in suspension

and the main parameter associates to the mixing

degree in agitated liquid–solids systems is the agitation

rate (Upadhyay et al., 1994). The main aim of this

work is to evaluate the influence of the agitation rate on

the performance of an anaerobic sequencing batch

biofilm reactor (ASBBR) in removing filtered and

suspended COD from a soymilk-based substrate con-

taining relatively high concentration of suspended

organic matter.

2. Material and methods

The bench-scale ASBBR consisted of an 8 l-cylindrical

flask with useful volume of 6.5 l when the foam was in

place. The biomass was immobilized on 3 cm cubic

polyurethane foam particles (porosity of 95%) disposed

in a perforated basket of 22 cm diameter and 18 cm

height inside the cylindrical flask (bed porosity of 40%).

Three 6 cm diameter propeller impellers provided the

mechanical mixing, displaced 4 cm above the bottom.

33 cm

SludgeDischarge

Effluent

GasVent

Detail1: Basket ContainingImmobilized Biomass

6 cm

Detail 2: PropellerImpeller

18 cm

7 cm

22 cm

Fig. 1. Experimental apparatus: ASBBR with imm

The reactor had its external surface involved by a heat

exchanger (streamer), permitting the process to occur at

a desired controlled temperature (3071 1C). Fig. 1

presents the diagram of the bench-scale reactor.

The sludge inoculum was obtained from a full-scale up-

flow anaerobic sludge blanket (UASB) reactor treating

slaughterhouse wastewater. The immobilization procedure

consisted of the granules maceration before being well

mixed on the polyurethane foam particles. Afterwards, the

sludge was left in contact with the foam particles for 24h

before being placed in the perforated basket.

The substrate was soymilk (NAN, Nestle, Germany)

with a total COD (CODT, mg l�1) of approximately

1000mg l�1, about 45% of which was suspended COD

(CODSS, mg l�1). The reactor was operated in a batch

mode with sequential cycles of 8 h (3 cycles per day),

with feeding and discharge steps lasting for 15 and

5min, respectively. Considering this time of cycle the

organic loading rate (OLR) was calculated as approxi-

mately 2.4 kg CODm�3 day�1. The substrate was kept

at approximately 4 1C to maintain its characteristics and

heated to 30 1C in a heat exchanger (streamer) before

entering the reactor.

COD (mg l�1), total volatile acids concentration

(TVA, mgHAc l�1), bicarbonate alkalinity (BA,

mgCaCO3 l�1), pH and solids analyses were performed

based on the standard methods for the examination of

water and wastewater (APHA/AWWA/WEF, 1998).

Suspended, dissolved and colloidal COD data were

assessed using 1.2 and 0.45mm membranes.

26 cm

Thermometer

HeatExchanger

2 cm

24 cm

Influent

Streamer

obilized biomass and mechanical agitation.

ARTICLE IN PRESSS.C. Pinho et al. / Water Research 38 (2004) 4117–41244120

Initially, the experiments were carried out to deter-

mine the time for complete mixing at different agitation

rates, using water (6.5 l) and inert support without

biomass. After a pulse input of a H2SO4 solution (3M,

1ml), the pH inside the reactor was monitored along

time. The mixing time, i.e., the time required to

homogenize the liquid medium, was determined by

considering the curve obtained from the pulse distur-

bance as a first-order response:

lnpHf � pH

pHf � pH0

� �¼ �

1

tðt � t0Þ: (1)

In expression (1), pH is the pH at a time t; pH0 is the

initial pH; pHf is the final pH; t represents the first-orderconstant and t0 is the delay hydrodynamic time (s).

Assuming that the total mixing time (tmix, in seconds) is

the time required to attain 0.01% of (pHf–pH0), tmixcould be estimated by

tmix ¼ t0 þ 6:91t: (2)

The influence of the agitation rate on the performance

of the reactor was assessed by means of COD temporal

profiles in batch cycles, using agitation rates ranging

from 500 to 1100 rpm. Before the temporal profiles were

obtained, the reactor was subjected to the gradual

increase of the agitation rate (from 0 to 100 rpm) during

20 days to acclimatize the biomass. In this initial phase,

the agitation rate was varied after the observation that

the reactor performance remained stable during 10 to 12

subsequent cycles (8 h each). Afterwards, the reactor

remained, in average, 1 week in each operational

condition (500, 700, 900 and 1100 rpm) before the

temporal profiles were obtained.

3. Results and discussion

3.1. Hydrodynamic assays

The values of pH and the mixing times profiles

obtained in the hydrodynamic experiments, with agita-

tion rates ranging from 200 to 1000 rpm, are shown in

0 10 20 30 40 50 60 70 80 90 100

3.03.54.04.55.05.56.06.57.07.5

pH

time (s)(a)

Fig. 2. (a) Response-curves obtained in hydrodynamic assays with di

(&) 800 rpm and (K) 1000 rpm. (b) Mixing time as a function of the

Fig. 2. As expected, the mixing time decreased as the

agitation rate increased from 200 to 900 rpm. However,

the mixing time estimated at 1000 rpm was longer than

that obtained at 600 and 800 rpm, possibly indicating

that the hydrodynamic behavior under such a very high

agitation rate differs from those previously assayed. In

fact, it was already observed that the performance of the

same type of reactor at 1100 rpm was atypical compared

to the data obtained at lower agitation rates (Cubas et

al., 2004). As a general rule, the data obtained from

experiments with 1100 rpm showed no correlation to the

other operating conditions.

Nonetheless, the mixing times estimated for all the

experimental conditions were extremely low (lower than

90 s) as compared with the cycle time (8 h). Therefore,

the liquid medium can be considered homogeneous for

all the agitation rates tested.

These values of mixing time are useful to be compared

with the apparent kinetic coefficients (k1app), and must be

negligible compared to 1/k1app, as this quantity represents

the time taken by a degradation step.

3.2. Performance experiments

The average values of total, filtered and suspended

affluent COD were 974770, 530772 and

444731mg l�1, respectively. The reactor start-up was

very fast during the acclimatization phase, achieving

75% of total COD removal efficiency in 20 days. The

pH ranged from 7.0 to 8.0 for the influent and 6.5–7.5

for the effluent. The affluent TVA concentration was

2873mg l�1 and the effluent concentrations were rarely

higher than 80mg HAc l�1. The process stability was

confirmed by the effluent BA values (528755mg

CaCO3 l�1), which were similar to the influent ones

(522784mg CaCO3 l�1). After the start-up period, the

effluent COD remained practically constant at 240725,

130715, 110720mg l�1 for total, filtered and sus-

pended COD, respectively.

Fig. 3 shows the temporal profiles of substrate

concentration, expressed as total, suspended and filtered

COD. The colloidal fraction of COD was insignificant in

200 400 600 800 1000

20

30

40

50

60

70

80

90

agitation rate (rpm)(b)

fferent agitation rates: (’) 200 rpm, (J) 400 rpm, (m) 600 rpm,

agitation rate.

ARTICLE IN PRESSS.C. Pinho et al. / Water Research 38 (2004) 4117–4124 4121

this synthetic wastewater, presenting values lower than

60mg l�1 for the total influent COD of 974mg l�1 and

did not exceed 50mg l�1 along the temporal profile.

Therefore, this COD fraction was not considered

important in the analysis of the system, and the filtered

COD fraction was taken as representing the concentra-

tion of the soluble organic matter concentration.

The effect of the agitation rate was evaluated by

adjusting the modified first-order kinetic expression as

presented previously (Cubas et al., 2004):

S ¼ SR þ ðSo � SRÞe�k

app1

t: (3)

In expression (3), t is the time (min); S and So are,

respectively, the substrate concentration (as COD) at t

and t0 (initial time, assumed as zero); SR is the residual

0 1 4 6 7 80

200

400

600

800

1000

1200

tota

l CO

D(m

g.l-1

)

time (h)(a) (

0 1 2 4 6 7 8

0

100

200

300

400

500

susp

ende

d C

OD

(m

g.l-1

)

time (h)(c) (

2 3 5

3 5

Fig. 3. (a–c) Temporal profiles of total, suspended and filtered COD:

Efficiency of CODSS removal along the batch cycle.

Table 1

Parameters from the adjustment of the first-order expression for diffe

Agitation rate (rpm) k1app (h�1)

Total Suspended F

(k1Tapp) (k1SS

app) (

500 0.9770.08 0.8370.16 1

700 1.4070.06 1.5470.09 1

900 2.3470.14 3.8270.19 1

1100 1.2170.02 1.6570.13 0

Average influent total, filtered and suspended COD: 974770, 53077

COD (mg l�1) and k1app is the apparent first-order kinetic

constant. Table 1 presents SR and k1app for each

operating condition estimated by a non-linear regression

analysis (Levenberg–Marquardt—Microcal Origin

6.1s). This regression was carried out using one

temporal profile for each experimental condition.

The value of k1Fapp (apparent first-order kinetic

constant for filtered COD) did not vary significantly

(1.2270.09 h�1), whereas k1SSapp (apparent first-order

kinetic constant for suspended COD) increased about

360% when the agitation rate was changed from 500 to

900 rpm, thus making k1Tapp (apparent first-order kinetic

constant for total COD) to rise as well.

The value of k1Fapp is directly related to the liquid-phase

mass transfer coefficient, improved by the minimization

0 1 2 4 6 7 8

0

100

200

300

400

500

600

filte

red

CO

D (

mg.

l-1)

time (h)b)

0 2 3 4 5 6 8

0

10

20

30

40

50

60

70

80

90

100

rem

oval

effi

cien

cy (

%)

time (h)d)

3 5

1 7

(’) 500 rpm, (J) 700 rpm, (m) 900 rpm and (X) 1100 rpm. (d)

rent agitation rates

SR (mg COD l�1)

iltered Total Suspended Filtered

k1Fapp) SRT SRSS SRF

.1370.13 200718 77722 123713

.3070.08 18578 6674 11877

.2270.12 15978 3673 11078

.8670.05 9573 3576 4977

2 and 444731mg l�1, respectively.

ARTICLE IN PRESSS.C. Pinho et al. / Water Research 38 (2004) 4117–41244122

of the stagnant liquid layer surrounding the bioparticle,

while k1SSapp is related to the hydrolysis step of the overall

conversion rate, which is dependent on the particle size

of the suspended organic matter. It is also important to

point out k1Fapp is, in fact, the result of the difference

between the solubilization of the suspended COD and

the effective consumption of the filtered organic matter.

Hence, it can be supposed that, although the effective

liquid-phase mass transfer flux (represented by k1Fapp) was

not increased, the increase of the agitation rate

improved the solubilization of the organic suspended

solids (represented by k1SSapp), thereby improving the

overall conversion rate and decreasing the batch cycle

time needed to obtain a required efficiency. This

statement is clearly illustrated in Fig. 3(d).

The k1app values estimated for the agitation rate of

1100 rpm did not follow the same trend observed for

lower agitation rate values. This behavior may be

ascribed to the different hydrodynamic pattern observed

previously at the agitation rate of 1000 rpm.

As seen in Fig. 4, the total residual substrate

concentration (SRT) decreased continuously from 200

to 95mg l�1 as the agitation rate was increased. Such an

effluent quality improvement can be credited mainly to

the decrease of the residual concentration of particulate

organic matter from 77 to 36mg l�1 for agitation rates

between 500 and 900 rpm, since the soluble residual

substrate concentration (SRF) remained practically

constant (11777mg l�1) in this range of agitation rate.

So, as discussed previously for k1app, the agitation rate

affected mainly the solubilization of particulate organic

material. However, for agitation rate of 1100 rpm, the

total residual substrate concentration decreased as a

result of the decrease of the soluble residual substrate

concentration.

The influence of particulate dissolution rates and

liquid-phase mass transfer on the cycle time can also be

500 600 700 800 900 1000 1100

40

60

80

100

120

140

160

180

200

SR

(mg.

l-1)

N (rpm)

Fig. 4. Residual substrate concentration (SR) as a function of

the agitation rate (N): (’) total, (m) suspended and (J) filtered

COD.

confirmed by evaluating the reaction rate profile along

the batch cycle. The degradation rate of organic matter

(RS) can be estimated as a function of time from the

mass balance in the batch reactor, as

RS ¼ �dðS � SRÞ

dt: (4)

Fig. 5 shows the reaction rates as a function of time,

based on the first-order kinetic model (Eq. (3)) and the

estimated parameters (Table 1). Higher reaction rates

were found at higher agitation rates, except for

1100 rpm, due to the peculiar hydrodynamic behavior

occurred at that agitation rate. Under this condition, the

liquid streams inside the reactor may have changed,

affecting the overall hydrodynamic pattern or even

causing superficial aeration of the reactor, thus interfer-

ing in the overall degradation rates.

The initial particulate substrate degradation rates

(RSoSS) increased as agitation rate increased (Table 2),

except for 1100 rpm, which confirms its atypical

behavior. As observed previously, this indicates that

the solubilization of particulate organic matter was the

main factor responsible for the enhancement of the

overall reaction rate (RSoT) with the increase of the

agitation rate. On the contrary, the soluble substrate

consumption rate (RSoF) remained practically constant

and no tendency could be observed. The initial

particulate substrate consumption rate (RSoSS) was

lower than the soluble substrate consumption rate

(RSoF) only for the agitation rate of 500 rpm (RSoF/

RSoSS=1.5), whereas RSoSS reached higher values for

agitation rates of 700 and 900 rpm (RSoF/RSoSS=0.92

and 0.33, respectively). Even for the condition of

1100 rpm, the particulate substrate degradation rate

was higher than that observed for soluble substrate

(RSoF/RSoSS=0.63).

4. Conclusions

The results obtained in this study indicate that the

agitation rate plays an important role in the solubiliza-

tion of suspended organic material. The acceleration of

suspended COD degradation with the increase of

agitation rate was probably due to the higher velocity

of shear of larger particles and major contact between

the particulate organic matter and the extra-cellular

enzymes. These statements can be supported by the data

of SRSS, that decreased from 500 to 1100 rpm, and the

stable values of SRF in the same range of agitation rate.

This conclusion is particularly important in the treat-

ment of partially soluble wastewaters. The following

specific conclusions could be drawn:

�

time, indicating that the system can be considered

ARTICLE IN PRESS

0 2 4 6 8

0

250

500

750

1000

1250

1500

17502000

RS

T (

mg

CO

D.l-1

.h-1

)

time (h)(a)

0

100

200

300

400

500

600

RS

F(m

g C

OD

.l-1.h

-1)

time (h)(b)

0

0

200

400

600

800

1000

1200

1400

1600R

SS (

mg

CO

D.l-1

.h-1

)

time (h)(c)

1 3 5 7 0 1 2 3 4 5 6 7 8

1 2 3 4 5 6 7 8

Fig. 5. Organic matter degradation profiles (a) total, (b) filtered and (c) suspended COD; (’) 500 rpm, (J) 700 rpm, (m) 900 rpm and

(X) 1100 rpm.

Table 2

Initial rates of organic matter degradation

Agitation rate

(rpm)

RSo (mg COD l�1 h�1)a

Total (RSoT) Suspended

(RSoSS)

Filtered

(RSoF)

500 751 305 460

700 1104 582 536

900 1907 1558 512

1100 1063 675 426

aThese values of initial velocities were calculated using the

average influent values of COD, i.e., SoT=974, SoF=530 and

SoSS=444mg COD l�1.

S.C. Pinho et al. / Water Research 38 (2004) 4117–4124 4123

well mixed at agitation rates ranging from 500 to

1100 rpm. However, the experiments revealed that the

hydrodynamic behavior under 1100 rpm was atypical

in comparison to the other conditions. This unusual

hydrodynamic pattern influenced the reactor’s per-

formance, probably caused by changes in liquid

streams or even by causing the aeration of the liquid.

�

not only for providing good mixing or improve

liquid-phase mass transfer, but also for improving the

solubilization of particulate organic material with

positive benefits to the organic mater consumption

rates. Furthermore, the residual substrate concentra-

tion decreased with the increase of the agitation rate,

affecting positively the final quality of the treated

wastewater.

�

the increase of the agitation rate from 500 to 900 rpm

due to the increase of the particulate organic matter

conversion rate, while the soluble organic matter

conversion rate remained practically constant.

Acknowledgments

This work was funded by FAPESP—Fundac- ao de

Amparo a Pesquisa do Estado de Sao Paulo, Brazil. The

authors acknowledge the grants received from FAPESP

and CNPq—Conselho Nacional de Desenvolvimento

Cientıfico e Tecnologico, Brazil.

References

APHA/AWWA/WEF, 1998. Standard Methods for the Exam-

ination of Water and Wastewater, 20th ed. American Public

Health Association, Washington, DC.

Camargo, E.F.M., Ratusznei, S.M., Rodrigues, J.A.D., Zaiat,

M., Borzani, W., 2002. Influence of the liquid superficial

velocity on the stability and performance of an anaerobic

sequencing batch reactor containing immobilized biomass.

Braz. J. Chem. Eng. 19, 267–275.

ARTICLE IN PRESSS.C. Pinho et al. / Water Research 38 (2004) 4117–41244124

Cubas, S.A., Foresti, E., Rodrigues, J.A.D., Ratusznei, S.M.,

Zaiat, M., 2004. Influence of the liquid-phase mass transfer

on the performance of a stirred anaerobic sequencing batch

reactor containing immobilized biomass. Biochem. Eng.

J. 17, 99–105.

Dague, R.R., Habben, C.E., Pidaparti, S.R., 1992. Initial

studies on the anaerobic sequencing batch reactor. Water

Sci.Technol. 26, 2429–2432.

Droste, R.L., Masse, D.I., 1995. Anaerobic treatment in

sequencing batch reactors. In: Proceedings of the Interna-

tional Symposium on Technology Transfer, Salvador,

Brazil, pp. 353–363.

Dugba, P.N., Zhang, R., 1999. Treatment of dairy wastewater

with two-stage anaerobic sequencing batch reactor sys-

tems—thermophilic versus mesophilic operations. Biore-

source Technol. 68, 225–233.

Eastman, J.A., Ferguson, J.F., 1981. Solubilization of particu-

late organic carbon during the acid phase of anaerobic

digestion. J. WPCF 53, 352–366.

Hirl, P.J., Irvine, R.L., 1997. Reductive chlorination of

perchloroethylene using anaerobic sequencing batch biofilm

reactors (AnSBBR). Water Sci. Technol. 35, 49–55.

Hobson, P.N., 1987. A model of some aspects of microbial

degradation of particulate substrates. J. Ferment. Technol.

65, 431–439.

Hollopeter, J.A., Dague, R.R., 1994. Anaerobic sequencing

batch reactor treatment of landfill leachate. In: Proceedings

of the 49th Purdue Industrial Waste Conference, Ann Arbor

Press, Chelsea, pp. 277–284.

Masse, D.I., Masse, L., 2001. The effect of temperature on

slaughterhouse wastewater treatment in anaerobic sequen-

cing batch reactors. Bioresource Technol. 76, 91–98.

Masse, D.I., Droste, R.L., Kennedy, K.J., Patni, N.K.,

Munroe, J.A., 1997. Potential for the psychrophilic anae-

robic treatment of swine manure using a sequencing batch

reactor. Can. Agri. Eng. 39, 35–41.

Nienow, A.W., 1998. Hydrodynamics of stirred bioreactors.

Appl. Mech. Rev. 51, 3–32.

Pinho, S.C., Ratusznei, S.M., Rodrigues, J.A.D., Foresti, E,

Zaiat, M., 2004. Feasibility of treating partially soluble

wastewater in an anaerobic sequencing batch biofilm reactor

(ASBBR) with mechanical stirring. Biores. Techn. in press.

Ratusznei, S.M., Rodrigues, J.A.D., Camargo, E.F.M., Zaiat,

M., Borzani, W., 2000. Feasibility of a stirred anaerobic

sequencing batch reactor containing immobilized biomass

for wastewater treatment. Bioresource Technol. 75,

127–133.

Ratusznei, S.M., Rodrigues, J.A.D., Camargo, E.F.M., Zaiat,

M., Borzani, W., 2001. Influence of the agitation rate on the

performance of a stirred anaerobic sequencing batch reactor

containing immobilized biomass. Water Sci. Technol. 44,

305–312.

Ratusznei, S.M., Rodrigues, J.A.D., Zaiat, M., 2003. Operating

feasibility of anaerobic whey treatment in a stirred sequen-

cing batch reactor containing immobilized biomass. Water

Sci. Technol. 48, 179–186.

Reyes III, F.L., Dague, R.R., 1995. Effects of initial seed

concentration on the start-up of the anaerobic sequencing

batch reactor. In: Proceedings of the 50th Purdue

Industrial Waste Conference. Ann Arbor Press, Chelsea,

pp. 449–459.

Rodrigues, J.A.D., Ratusznei, S.M., Camargo, E.F.M., Zaiat,

M., 2003. Influence of agitation rate on the performance of

an anaerobic sequencing batch reactor containing granu-

lated biomass treating low-strentgth wastewater. Adv.

Environ. Res. 7, 405–410.

Sung, S., Dague, R.R., 1995. Laboratory studies on enhance-

ment of granulation in the anaerobic sequencing batch

reactors. Water Environ. Res. 67, 294–301.

Timur, H., Ozturk, I., 1999. Anaerobic treatment of leachate

using sequencing batch reactor and hybrid bed filter. Water

Res. 33, 3225–3230.

Upadhyay, S.N., Rai, N.B., Kumar, V., Shah, Y.T., 1994.

Particle suspension and liquid–solid mass transfer in

mechanically agitated vessel. Rev. Chem. Eng. 10, 1–130.

Zaiat, M., Rodrigues, J.A.D., Ratusznei, S.M., Camargo,

E.F.M., Borzani, W., 2001. Anaerobic sequencing batch

reactors for wastewater treatment: a developing technology.

Appl. Microbiol. Biotechnol. 55, 29–35.

Zeeman, G., Sanders, W.T.M., Wang, K.Y., Lettinga, G., 1997.

Anaerobic treatment of complex wastewater and waste

activated sludge-application of an upflow anaerobic solid

removal (UASR) reactor for the removal and pre-hydrolysis

of suspended COD. Water Sci. Technol. 35, 121–128.