ARTICLE IN PRESS

Chemical Engineering Science 65 (2010) 4771–4781

Contents lists available at ScienceDirect

Chemical Engineering Science

0009-25

doi:10.1

Abbre

DS, dissn Corr

E-m

rob.van

journal homepage: www.elsevier.com/locate/ces

Kinetic analysis of biological sulphate reduction using lactate as carbonsource and electron donor: Effect of sulphate concentration

Oluwaseun O. Oyekola n, Robert P. van Hille, Susan T.L. Harrison

Centre for Bioprocess Engineering Research, Department of Chemical Engineering, University of Cape Town, Private Bag, Rondebosch 7701, South Africa

a r t i c l e i n f o

Article history:

Received 11 February 2010

Received in revised form

3 May 2010

Accepted 12 May 2010Available online 21 May 2010

Keywords:

Acid mine drainage

Bioprocessing

Bioreactors

Environment

Kinetics

Sulphate reducing bacteria

09/$ - see front matter & 2010 Elsevier Ltd. A

016/j.ces.2010.05.014

viations: VSRR, volumetric sulphate reduction

olved sulphide; VSLR, volumetric sulphate lo

esponding author. Tel.: +27 21 650 4021; fax

ail addresses: seunoyekola@gmail.com (O.O. O

hille@uct.ac.za (R.P. van Hille), sue.harrison@

a b s t r a c t

This study investigated the effect of feed sulphate concentration on the kinetics of anaerobic sulphate

reduction by a mixed SRB culture, using lactate as the sole carbon source and electron donor. Chemostat

cultures were operated across a range of residence times (0.5–5 d) and feed sulphate concentrations

(1.0–10.0 g l�1). Similar phenomena were observed at feed sulphate concentrations of 1.0 and

10.0 g l�1 with the volumetric sulphate reduction rate increasing linearly with increasing volumetric

sulphate loading rate. These reactors were characterised by higher specific volumetric sulphate

reduction rates with maximum values of 0.24 and 0.20 g h�1 g�1. Contrastingly, the reactors fed with

sulphate concentrations of 2.5 and 5.0 g l�1 showed distinctly different trends in which the volumetric

sulphate reduction rate passed through a maximum at the dilution rates of 0.014 and 0.021 h�1,

respectively, followed by a decline with further increase in sulphate loading rate. The maximum specific

volumetric sulphate reduction rates observed were 2–6-fold lower than those observed at 1.0 and

10.0 g l�1 feed sulphate concentrations. Profiles of specific volumetric sulphate reduction rate and

biomass concentration suggested a shift in lactate utilisation from oxidation to fermentation at high

dilution rates, implying a change in the dominant components of the microbial consortium. The data

suggest that population structure was influenced by lactate affinity and dissolved sulphide

concentration. The trends observed were attributed to the greater ability of lactate oxidisers to

scavenge lactate under limiting concentrations of the substrate and their greater resilience to dissolved

sulphide species in comparison to lactate fermenters.& 2010 Elsevier Ltd. All rights reserved.

1. Introduction

With increasing global population and industrial activities,generation of wastewaters is rapidly increasing (Cenni et al.,2001). Examples of such wastewaters posing a great danger to theecosystem include those emanating from the following indus-tries: pulp and paper (Thompson et al., 2001; Hulshoff Pol et al.,1998), chemical, metallurgical and mining (Baker and Banfield,2003; Johnson and Hallberg, 2003; Kuo and Shu, 2004) andfertilizer (Ghigliazza et al., 2000). The effluents from theseindustries include sulphate-rich and heavy metal-laden streams.The mining industry is a major part of the industrial sector inSouth Africa and the activities of this industry are accompanied bythe generation of sulphate and metal-laden acid mine drainage(AMD), generated on subsequent weathering of waste rock,tailings and surfaces exposed on mining. When dischargeduntreated, it poses a threat to the quality of fresh water resources

ll rights reserved.

rate; TS, total sulphide;

ading rate.

: +27 21 650 5501.

yekola),

uct.ac.za (S.T. Harrison).

and consequently the well-being of humans and the environment atlarge. The AMD generated causes the acidification and contamina-tion of the ground water with heavy metals (Naicker et al., 2003). Itsmanagement thus requires attention (Naicker et al., 2003).

Sulphate, a major pollutant, increases the salinity of receivingwater bodies, which in turn, reduces the availability of potableand usable water (Pulles et al., 1995). Therefore, one of the majoraims of developing technology for AMD treatment in semi-aridSouth Africa, where water is a limited natural resource is toreduce the sulphate concentration to acceptable levels. Anaerobicbiological sulphate reduction (BSR) has attracted intensiveresearch in recent years due to the advantages it offers over theother AMD treatment technologies. In this treatment process,sulphide and bicarbonate are produced by sulphate-reducingbacteria (SRB) in the presence of a suitable electron donor andcarbon source. The bicarbonate alkalinity neutralises acidity whiledissolved metals are precipitated by the sulphide. These reactionsare summarised in Eqs. (1)–(3) (Drury, 1999):

ElectrondonorþSO2�4 -HS�þHCO�3 ð1Þ

Me2þþHS�-MeSkþHþ ð2Þ

HCO�3 þHþ-H2OþCO2 ð3Þ

ARTICLE IN PRESS

O.O. Oyekola et al. / Chemical Engineering Science 65 (2010) 4771–47814772

where Me2 + represents the metal ion, such as Zn2 +, Cu2 +, Pb2 +

and Ni2 +.Anaerobic biological sulphate reduction technology has been

employed effectively under controlled and optimised conditionsat both laboratory and full-scale. The use of two full-scalesulphidogenic technologies, namely, BioSUREs and Thiopaqs

processes has been widely reported (Boonstra et al., 1999;Corbett, 2000; Rose et al., 2000; Hulshoff Pol et al., 2001; vanHouten et al., 2006; Whittington-Jones et al., 2006). The provisionof a suitable carbon source and electron donor for this processremains a challenge. Lactate is a potential carbon source andenergy donor, offering advantages in the BSR process. It supportsthe growth of a wide spectrum of sulphate-reducing bacteria,encouraging microbial diversity and consequent treatment sys-tem resilience (Kaksonen et al., 2004; Oyekola, 2008). Usinglactate as a substrate for BSR has been reported to decreasesulphide toxicity (a major threat to biological treatment technol-ogy) (Kuo and Shu, 2004). Furthermore, the complete oxidation of1 mole of lactate results in the production of 3 moles ofbicarbonate alkalinity as opposed to other simple organicmolecules, such as ethanol, which on complete oxidation yields2 moles of bicarbonate, with concomitant reduction of 1.5 molessulphate in both cases (Kaksonen et al., 2004).

In the bioremediation of AMD and other sulphate-laden waste-waters using SRB, sulphate reduction is a critical sub-process(Eq. (1)) (Baskaran and Nemati, 2006). Chemostat studies usingacetate and ethanol as electron donors for suspended mixed SRBconsortia, and lactate-fed immobilised SRB cells, have shown that thesulphate reduction reaction kinetics are influenced by feed sulphateconcentration (Erasmus, 2000; Moosa et al., 2002; Baskaran andNemati, 2006). In this current work, the kinetic study was extended tothe use of lactate as the sole carbon source for suspended mixed SRBcells in a continuous system. Kinetics studies across a broad spectrumof substrates underpin the fundamental knowledge base for a designof an efficient BSR process. Most kinetic data reported for microbialsulphate reduction using lactate as carbon source use pure culturesand are incomplete. However, mixed cultures are generally employedin treatment of wastewaters by BSR. In this paper, we describe theeffects of sulphate concentration and its volumetric loading rate, andresidence time on the kinetics of anaerobic sulphate reduction by alactate-fed, mixed SRB culture.

B

C

D

EG

H

J

A

I

F

Fig. 1. Schematic diagram of experimental set-up: (A) thermoregulated waterbath;

(B) anaerobic bioreactor; (C) overhead stirrer; (D) overflow port; (E) gas vent;

(F) sampling port; (G) feed inlet; (H) feed pump; (I) pH meter; (J) pH probe.

2. Materials and methods

2.1. Microorganisms and growth medium

A mixed culture of SRB, adapted to growth on lactate, wasobtained from the laboratory of Prof. John Duncan (RhodesUniversity, South Africa). Modified Postgate B medium, in whichlactate formed the sole carbon source and electron donor, was usedas the growth medium (Postgate, 1984). Bromo-ethane-sulphonic-acid (BESA) (3.2 g l�1) was added to the culture at the enrichmentstage, prior to the culturing of the continuous reactors, to inhibitmethanogenic activity (Visser, 1995). The medium containing1.0 g l�1 of sulphate had the following composition in 1 liter madeup with deionised water: 0.5 g KH2PO4; 1.0 g NH4Cl; 2.0 gMgSO4.7H2O; 1.0 g Na2SO4; 1.0 g yeast extract; 3.7 ml 60% sodiumlactate; 0.3 g sodium citrate. All reagents were analytical grade.Sodium sulphate and sodium lactate were used as the sulphate andlactate sources, respectively. Lactate was added at 120% of thestoichiometric requirement based on the feed sulphate concentra-tion, assuming its metabolism according to Eq. (4). The mediumwas heat sterilised (121 1C, 15 psig, 20 min).

2lactateþSO2�4 -2acetateþ2HCO�3 þHS�þHþ ð4Þ

2.2. Experimental set-up and procedures

Chemostat studies were carried out using 1 liter anaerobicstirred tank reactors, equipped with overhead stirrers drivingRushton impellers (D¼50 mm, D/T¼0.5, C¼50 mm) for contin-uous mixing at 400 rpm (tip speed of 1.05 m/s) (Fig. 1). Thereactors were initiated in batch culture using a 10% inoculum.Constant temperature was maintained at 35 1C while pH wasmaintained at 870.2 using 1 M NaOH. After the establishment ofa viable microbial population, the reactors were changed tocontinuous operation. In the continuous mode, the medium wasintroduced into the bioreactor using a variable speed peristalticpump (ISMATECs), while the effluent was discharged by gravitythrough a U-shaped overflow tube (Fig. 1). Steady-state wasachieved at each flow rate. Steady-state conditions were assumedto be established when both the residual sulphate and bacterialconcentrations varied by o10% after a period of operation equalto three retention times since system perturbation.

In order to investigate the effects of sulphate concentrationand its volumetric loading on the kinetics of biological sulphatereduction, four separate experimental runs were carried out. Thefeed sulphate concentrations used were 1.0, 2.5, 5.0 and 10.0 g l�1

at a constant lactate excess of 20% on a stoichiometric basis. Thereactors were operated at dilution rates in the range of 0.0083–0.083 h�1 (residence times of 0.5–5.0 d). The inoculum was welladapted for each feed concentration prior to the continuousoperation. For experiments with feed sulphate concentrations 1.0and 2.5 g l�1, the stock culture was used as the inoculum. Thebioreactor operating at 5.0 g l�1 was inoculated with culture fromthe reactor fed with 2.5 g l�1 sulphate, while the 10.0 g l�1 feedsulphate experiment was inoculated with culture taken from theexperiment with media containing 5.0 g l�1sulphate. A moredetailed description of the experimental set-up and procedure isprovided elsewhere (Oyekola, 2008; Oyekola et al., 2009).

2.3. Analytical methods

Sulphate concentration was quantified as barium sulphateusing a turbidimetric method (APHA, 1975). Total dissolvedsulphide was estimated spectrophotometrically at 670 nm,following the color development of methylene blue resultingfrom the reaction between the sulphide and the colorimetricreagent, N,N-dimethyl-p-phenylenediamine sulphate in acidmedium (Cline, 1969). Acetate, propionate and lactate concentra-tions were determined using high performance liquid chromato-graphy (HPLC, Beckman, System Gold), with a UV detector(Detector no. 168) and a glass lined Wakosil column, using amethod adapted from Moosa et al. (2002). Phosphoric acid

ARTICLE IN PRESS

O.O. Oyekola et al. / Chemical Engineering Science 65 (2010) 4771–4781 4773

(20 mM, pH 2.5) acted as the mobile phase. Prior to injection,samples were acidified using 4 N H2SO4 and filtered usingsterilised 0.22 mm Expresss PES membrane filters (Millipore,Corp.). Bacterial concentration was estimated as dry mass.Samples were centrifuged in pre-dried and pre-weighed micro-fuge tubes (80 1C, 48 h). This was followed by the washing ofpellets obtained using 1� phosphate-buffered saline (PBS) anddrying of washed sample pellets (80 1C, 48 h).

3. Results and discussion

3.1. Steady-state profiles

The steady-state profiles of bacterial concentration, sulphateconversion, lactate conversion and volumetric sulphate reductionrate (VSRR) as a function of volumetric sulphate loading rate (VSLR)and dilution rate for the different feed sulphate concentrations(1.0–10.0 g l�1) are shown in Figs. 2a–4a. The correspondingconcentrations of residual lactate, residual sulphate, acetateproduced and propionate produced are shown in Figs. 2b–4b.

3.1.1. Feed sulphate concentration of 1.0 g l�1

The steady-state profiles at a feed sulphate concentration of1.0 g l�1 are shown in Fig. 2. As shown in Fig. 2a, within the rangeof dilution rates 0.0083–0.042 h�1 (residence time: 5.0–1.0 d), theVSRR increased linearly up to a value of 0.036 g l�1 h�1

, with anincrease in the volumetric sulphate loading rate, while sulphateconversion to sulphide remained relatively stable at 8670.5%.The corresponding lactate conversion across this range of dilutionrates was 9970.6%. Fig. 2a shows that the bacterial concentrationwas low and gradually decreased with increasing dilution rate.An average bacterial concentration 0.22670.031 g l�1 wasmaintained over the operation range of this study. This corre-sponded to a maximum specific VSRR of 0.16 g h�1 g�1 repre-senting an efficient microbial sulphate utilisation. On reduction ofthe residence time to 0.5 d (volumetric loading rate¼0.083 g l�1

h�1), the bacterial concentration was reduced to 0.169 g l�1. Theincrease in the volumetric sulphate reduction rate to a maximum

0.000

20

40

60

80

100

120

0

20

40

60

80

100

120

0.000

0.005

0.010

0.015

0.020

0.025

0.030

0.035

0.040

0.045

0.0

0.1

0.2

0.3

0.4

0.5

0.60.00

Sulp

hate

con

vers

ion

(%)

Volumetric sulphate loading rate (gl-1h-1)

Sulphate conversion

Lac

tate

con

vers

ion

(%)

Lactate conversion

Volumetric sulphate reduction rate

Vol

umet

ric

sulp

hate

red

ucti

on r

ate

(gl-1

h-1)

Bacterial concentration

Bac

teri

al c

once

ntra

tion

(gl

-1)

Dilution rate (h-1)

(a)

0.02 0.04 0.06 0.08

0.02 0.04 0.06 0.08

Fig. 2. Steady-state kinetics of continuous reactor with a feed sulphate concentratio

pH 870.2. Kinetic data presented are (a) bacterial concentration, sulphate conversion

residual sulphate, acetate produced and propionate concentrations.

value of 0.041 g l�1 h�1 was not proportional to the increase inthe volumetric loading rate; hence, sulphate conversion decreasedto 50% (Fig. 2a). Despite the reduction in the sulphate conversion,the conversion of lactate remained at 99%. Based on stoichio-metric estimation and the amount of sulphate reduced, theexpected residual lactate concentration at this residence time was1.31 g l�1. However, a lower concentration of 0.029 g l�1 wasdetected. This indicates that the sulphate conversion was notlimited by lactate addition. This observation also suggests that,while sulphate reducers oxidised lactate at a rate near theirmaximum, another group of microorganisms, characterised byhigher mmax values for lactate utilisation, was able to proliferatedue to increased lactate availability at high volumetric loadingrate of lactate (sulphate). Consequently, there was no build-up inthe residual lactate. Hence, the decline in sulphate conversioncorrelated with the washout of some SRB species from the reactorwhen operated at a dilution rate greater than their mmax andselection of resilient SRB strains. These phenomena have beenpreviously described by Oyekola (2008).

Complete utilisation of lactate was observed at dilution rates0.0083–0.014 h�1 (residence time: 5–3 d). Beyond this range, lowresidual lactate levels (0.0014–0.029 g l�1) were recorded(Fig. 2b). Further, decreasing acetate production, an indication ofdecline in lactate oxidation (Eq. (4)), was congruent withincreased levels of propionate and residual sulphate (Fig. 2b).Propionate production is an indication of lactate fermentation(Eq. (5)) (Heimann et al., 2005):

3lactate-acetateþ2propionateþCO2þH2O ð5Þ

These trends suggest that lactate oxidation was predominantat low dilution rates while increasing dilution rate favoured thelactate metabolism via the fermentative pathway. The phenom-enon of metabolic shift with increasing dilution rate is describedin detail by Oyekola et al. (2009). This result is consistent with thefindings of Baskaran and Nemati (2006). They showed thatsulphate conversion of a lactate-fed chemostat culture, receivinga feed sulphate concentration of 1.0 g l�1, decreased from 100% to58% as the sulphate volumetric loading rate was increased from0.001 to 2.85 g l�1 h�1. This was attributed to reduced lactate

0.0

0.2

0.4

0.6

0.8

1.0

0.0

0.5

1.0

1.5

2.0

2.5

3.0

0.0

0.5

1.0

1.5

0.00

0.01

0.02

0.03

0.04

Res

idua

l sul

phat

e co

ncen

trat

ion

(gl-1

)

Volumetric sulphate loading rate (gl-1h-1)

Residual sulphate concentration

Ace

tate

pro

duce

d co

ncen

trat

ion

(gl-1

)

Acetate produced concentration

Propionate produced concentration

Pro

pion

ate

prod

uced

con

cent

rati

on (

gl-1

h-1)

Residual lactate concentration

Res

idua

l lac

tate

con

cent

rati

on (

gl-1

)

Dilution rate (h-1)

(b)

0.00 0.02 0.04 0.06 0.08

0.00 0.02 0.04 0.06 0.08

n of 1.0 g l�1 and lactate feed concentration of 2.2 g l�1, operated at 35 1C and

, lactate conversion and volumetric sulphate reduction rate; (b) residual lactate,

ARTICLE IN PRESS

O.O. Oyekola et al. / Chemical Engineering Science 65 (2010) 4771–47814774

oxidation. In the current study, consistent low levels of residualsulphate (0.13–0.14 g l�1) were maintained in the volumetricloading rate range of 0.0083–0.042 g l�1 h�1. Beyond this range,at the highest volumetric loading rate imposed (0.083 g l�1 h�1),a significant increase of residual sulphate concentration to0.5 g l�1 (50% sulphate conversion) was observed. Formation ofacetate and dissolved sulphide correlated with these trends(Fig. 2b and Table 1). The concentrations of these products ofbiological sulphate reduction steadily declined with increasingvolumetric sulphate loading rate at dilution rates 0.0083–0.021(residence time: 5.0–2.0 d). Beyond this range, the concentrationsof acetate and dissolved sulphide remained relatively constant at0.970.1 g l�1 and 0.09070.0029 g l�1, respectively.

Despite the high sulphate conversion maintained at dilutionrates 0.0083–0.042 h�1, there was a consistent decrease insulphide levels within this range. This can be attributed to theloss of some of the sulphide produced. The measured sulphideconcentration was the total amount of sulphide in the liquidphase. This underestimated the equivalent total sulphide that

Table 1Comparison of expected sulphide (TS) and assayed sulphide (DS) concentrations acros

RT (d) D (h�1) Calculated total sulphide (TS) and assayed dissolved sulphide (

So (g l�1)

1.0 2.5

DS TS DS TS

0.5 0.083 0.087 0.165 nd nd

1.0 0.042 0.089 0.285 0.014 0.067

1.5 0.028 0.093 0.287 0.027 0.096

2.0 0.021 0.103 0.288 0.033 0.162

3.0 0.014 0.134 0.288 0.059 0.348

3.5 0.012 nd nd nd nd

4.0 0.010 0.184 0.289 0.131 0.453

5.0 0.0083 0.184 0.289 0.189 0.452

nd: not determined.

0.000

20

40

60

80

100

120

0

20

40

60

80

100

120

0.000

0.005

0.010

0.015

0.020

0.0

0.1

0.2

0.3

0.4

0.5

0.60.00

Sulp

hate

con

vers

ion

(%)

Volumetric sulphate loading rate (gl-1h-1)

Sulphate conversion

Lac

tate

con

vers

ion

(%)

Lactate conversion

Volumtetric sulphate reduction rate

Vol

umet

ric

sulp

hate

red

ucti

on r

ate

(gl-1

h-1)

Bacterial concentration

Bac

teri

al c

once

ntra

tion

(gl

-1)

Dilution rate (h-1)

(a)

0.01 0.02 0.03 0.04 0.05

0.02 0.04 0.06 0.08 0.10 0.12

Fig. 3. Steady-state kinetics of continuous reactor with a feed sulphate concentration

870.2. Kinetic data presented are (a) bacterial concentration, sulphate conversion, lacta

sulphate, acetate produced and propionate concentrations.

would be expected from the amount of sulphate reduced (Table 1).The observation of sulphur biofilms on the inner walls of some ofthe reactors and the evolution of sulphide from the gas vent in thereactor set-up used in this study is associated with the significantloss of volatile sulphide. Baskaran and Nemati (2006) suggestedthat a reactor set-up having a large head space and continuouslymixed at 400 rpm, as in the current study, would facilitate theevolution of sulphide produced from the liquid phase.

3.1.2. Feed sulphate concentrations of 2.5 and 5.0 g l�1

Similar phenomena were observed in the experiments withfeed sulphate concentrations of 2.5 and 5.0 g l�1. Hence, thekinetic profiles at a feed sulphate concentration of 2.5 g l�1 arepresented herein to typify the trends observed in these twoexperiments. At a feed sulphate concentration of 2.5 g l�1, thevolumetric sulphate reduction rate increased to an optimumvalue of 0.015 g l�1 h�1 at a low dilution rate of 0.014 h�1

(residence time: 3 d) (Fig. 3a). The corresponding sulphate

s the experimental conditions investigated for BSR in chemostat culture.

DS) concentrations (g l�1) at different feed sulphate concentrations (So)

5.0 10.0

DS TS DS TS

nd nd nd nd

0.022 0.099 0.312 0.72

0.088 0.254 0.312 0.90

0.232 0.513 0.352 1.04

0.369 0.665 0.557 1.10

nd nd nd nd

0.519 0.924 0.559 1.29

0.529 0.970 0.622 1.34

0.0

0.5

1.0

1.5

2.0

2.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

0.0

0.5

1.0

1.5

2.0

2.5

0.0

0.2

0.4

0.6

0.8

1.0

Res

idua

l sul

phat

e co

ncen

trat

ion

(gl-1

)

Volumetric sulphate loading rate (gl-1h-1)

Residual sulphate concentration

Ace

tate

pro

duce

d co

ncen

trat

ion

(gl-1

)

Acetate produced concentration

Propionate produced concentration

Pro

pion

ate

prod

uced

con

cent

rati

on (

gl-1

)

Residual lactate concentration

Res

idua

l lac

tate

con

cent

rati

on (

gl-1

)

Dilution rate (h-1)

(b)

0.00 0.01 0.02 0.03 0.04 0.05

0.00 0.02 0.04 0.06 0.08 0.10 0.12

of 2.5 g l�1 and lactate feed concentration of 5.5 g l�1, operated at 35 1C and pH

te conversion and volumetric sulphate reduction rate; (b) residual lactate, residual

ARTICLE IN PRESS

O.O. Oyekola et al. / Chemical Engineering Science 65 (2010) 4771–4781 4775

conversion was 42%. The maximum bacterial concentration andsulphate conversion achieved were 0.35 g l�1 and 54%,respectively. These values were obtained at the residence timeof 5 d. Similarly, at a feed sulphate concentration of 5.0 g l�1,maximum bacterial concentration and sulphate conversion of0.46 g l�1 and 58%, respectively, were achieved at residencetime of 5 d. A sharp decline in the volumetric sulphate reductionrate was accompanied by a gradual decrease in the bacterialconcentration with the increasing volumetric loading rate(residence time: 3–1 d). Across the range of dilution ratesstudied, the lactate conversion was maintained at 100% despitethe significant decline in the sulphate conversion as thevolumetric loading rate was increased. These observationssuggest the presence of non-SRB lactate utilisers, which becamepredominant as the dilution rate was increased. Competition bythese microorganisms could have led to the reduced availabilityof lactate for SRB growth; hence, there is a decline in thevolumetric sulphate reduction rate (Fig. 3a). This observation wasconsistent with the concomitant sharp increase in theconcentrations of propionate and residual sulphate observedwith increasing volumetric sulphate loading rate (Fig. 3b).

Even though the stoichiometric amount of lactate in themedium containing 2.5 g l�1 sulphate was 2.5-fold times thanthat present in the reactor receiving a feed sulphate concentrationof 1.0 g l�1, the acetate production in these two reactors wascomparable (Figs. 2 and 3b). Furthermore, similar lactate conver-sion was recorded for these two reactors, yet the sulphateconversion was lower in the reactor fed with 2.5 g l�1 sulphate(Figs. 2 and 3a). This indicates a lower fractional lactateconversion by oxidation in the reactor fed with 2.5 g l�1 sulphate.These observations further show that lactate was utilised via asecondary metabolic pathway. As shown in Figs. 2 and 3a, in spiteof higher bacterial concentrations observed in the bioreactoroperating with 2.5 g l�1 sulphate, the sulphate reduction ratewas lower than at 1.0 g l�1. Maximum volumetric sulphatereduction rates of 0.042, 0.015 and 0.032 g l�1 h�1 were observedat 1.0, 2.5 and 5.0 g l�1, respectively. Similarly, maximum specific

0.00

20

40

60

80

100

120

0

20

40

60

80

100

120

0.00

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0.10

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.00

Sulp

hate

con

vers

ion

(%)

Volumetric sulphate loading rate (gl-1h-1)

Sulphate conversion

Lac

tate

con

vers

ion

(%)

Lactate conversion

Volumetric sulphate reduction rate

Vol

umet

ric

sulp

hate

red

ucti

on r

ate

(gl-1

h-1

Bacterial concentration

Bac

teri

al c

once

ntra

tion

(gl

-1)

Dilution rate (h-1)

(a)

0.1 0.2 0.3 0.4 0.5

0.01 0.02 0.03 0.04 0.05

Fig. 4. Steady-state kinetics of continuous reactor with a feed sulphate concentration

pH 870.2. Kinetic data presented are (a) bacterial concentration, sulphate conversion

residual sulphate, acetate produced and propionate concentrations.

volumetric sulphate reduction rates of 0.24, 0.044 and0.11 g h�1 g�1 were recorded for 1.0, 2.5 and 5.0 g l�1, respectively.

3.1.3. Feed sulphate concentration of 10.0 g l�1

Substrate inhibition was significant at a 10.0 g l�1 feedsulphate concentration as reflected by the low sulphate conver-sion of 40% observed at the lowest volumetric loading rate of0.0417 g l�1 h�1 (residence time: 5 d) (Fig. 4a). Nevertheless, thisrelatively low sulphate conversion translated into a significantremoval of sulphate (2.2–4.0 g l�1) and production of sulphide(0.31–0.62 g l�1) (Table 1). The gradual decline in the sulphateconversion with increasing loading rate was accompanied by asteady lactate conversion of 6071.8%. This indicated that aportion of the electrons released from lactate metabolism waschannelled via other pathways than sulphate reduction. This wasconsistent with a steady increase in the concentrations ofpropionate and residual sulphate (Fig. 4b). The volumetricsulphate reduction rate increased linearly across the range ofdilution rates studied, with the maximum value of 0.090 g l�1 h�1

recorded at the residence time of 1 d. The corresponding sulphateremoval efficiency was 22%. High bacterial concentration (0.44–0.59 g l�1) was maintained over the operation range of thisexperiment. This could have been responsible for the sustainedenhancement of the volumetric rate of sulphate reduction and thegradual decrease in the sulphate conversion observed over theentire period of bioreactor operation. The specific volumetric ratesincreased in the range 0.05–0.20 g�1 h�1 g�1 with increasingdilution rate. These values were comparable to those observed inthe reactor receiving a feed sulphate concentration of 1.0 g l�1.There was a slight increase in the residual lactate over the entireperiod of bioreactor operation; high lactate concentration(8.970.4 g l�1) was maintained.

Decline in the acetate concentration with increasing volu-metric loading rate was similar to the trend exhibited by thesulphate conversion (Fig. 4a and b). There was a consistentincrease in both the concentrations of residual sulphate and

0.0

1.5

3.0

4.5

6.0

7.5

9.0

0.0

1.5

3.0

4.5

6.0

0.0

1.5

3.0

4.5

6.0

0.0

2.5

5.0

7.5

10.0

12.5

15.00.00 0.01 0.02 0.03 0.04 0.05

Res

idua

l sul

phat

e co

ncen

trat

ion

(gl-1

)

)

Volumetric sulphate loading rate (gl-1h-1)

Residual sulphate concentration

Ace

tate

pro

duce

d co

ncen

trat

ion

(gl-1

)

Acetate produced concentration

Propionate produced concentration

Pro

pion

ate

prod

uced

con

cent

rati

on (

gl -1

Residual lactate concentration

Res

idua

l lac

tate

con

cent

rati

on (

gl-1

)

Dilution rate (h-1)

(b)

0.0 0.1 0.2 0.3 0.4 0.5

of 10.0 g l�1 and lactate feed concentration of 22.2 g l�1, operated at 35 1C and

, lactate conversion and volumetric sulphate reduction rate; (b) residual lactate,

ARTICLE IN PRESS

O.O. Oyekola et al. / Chemical Engineering Science 65 (2010) 4771–47814776

propionate produced. On the other hand, the high concentrationsof dissolved sulphide over the entire period of bioreactoroperation (Table 1) indicated substantial lactate oxidation in thisexperiment.

3.2. Dependency of biological sulphate reduction kinetics on feed

sulphate concentration, residence time and sulphate loading rate

3.2.1. Volumetric sulphate reduction rate

Although the results shown in Table 2 suggest that themaximum volumetric sulphate reduction rate may be influencedby the feed sulphate concentration, a consistent trend is notfound. In Fig. 5, the volumetric sulphate reduction rate is reportedas a function of the dilution rate and the volumetric sulphateloading rate across the feed sulphate concentrations used. Similartrends of the response of the volumetric sulphate reduction rateto the loading rate of sulphate were exhibited by the reactorsreceiving 1.0 and 10.0 g l�1 sulphate (Fig. 5).

Across the range of the dilution rates studied, the VSRRincreased with increasing VSLR in these two experiments. On thecontrary, in the reactors receiving media containing 2.5 and5.0 g l�1 sulphate concentration, the volumetric sulphate reduc-tion rate increased to a maximum at low dilution rates (0.0083–0.014 h�1, residence time: 5–3 d) (Fig. 5a). Further increase of theloading rate beyond 0.035 and 0.104 g l�1 h�1 in the reactors fed

Table 2Effect of feed concentration and volumetric sulphate loading rate on maximum volum

Feed sulphateconcentration (g l�1)

MaximumVSRR (g l�1 h�1)

CorrespondingSpecific VSRR (g h�1 g�1)

1.0 0.041 0.24

2.5 0.015 0.044

5.0 0.032 0.11

10.0 0.090 0.20

0.000.00

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0.10

Vol

umet

ric

sulp

hate

red

ucti

on r

ate

(gl-1

h-1

)

Dilution rate (h-1)

1.0 g l-1

2.5 g l-1

5.0 g l-1

10.0 g l-1

O

F

0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09

Fig. 5. (a) Effect of feed sulphate concentration and dilution rate on volumetric sulphat

lactate oxidation (O) and fermentation (F) when So¼2.5 and 5.0 g l�1. (b) Effect of feed s

reduction rate.

with 2.5 and 5.0 g l�1, respectively, led to a decline in thesulphate reduction rate (Fig. 5b). In addition, the maximumvolumetric sulphate reduction rates achieved in the reactorsreceiving feed sulphate concentrations of 1.0 and 10.0 g l�1 werehigher than those obtained at 2.5 and 5.0 g l�1 (Table 2). This isfurther emphasised when specific VSRR is considered. Clearly twodifferent environments have been generated. An environment ofhigh specific VSRR (0.20–0.24 g h�1 g�1) at 1.0 and 10.0 g l�1 feedsulphate concentration and an environment characterised by lowspecific VSRR of 0.044–0.11 g h�1 g�1 at 2.5 and 5.0 g l�1 feedsulphate concentration.

The results obtained at residence times 3–5 d, in the presentstudy agree with the observations by Moosa (2000). On studyingthe effect of feed sulphate concentration in an acetate-fedchemostat, it was revealed that VSRR increased with increasingfeed sulphate concentration in the range 1.0–10.0 g l�1. Further,Hansford et al. (2007) reported that increasing the feed sulphateconcentration in the range 1.0–10.0 g l�1 enhanced the volu-metric sulphate reduction rate in an ethanol-fed chemostat.

Contrary to the observations reported in the suspended-cellsystems discussed above, in immobilised cell-systems fedwith lactate, increasing feed sulphate concentrations, in the range1.0–5.0 g l�1, negatively influenced the reactor performance asmeasured by the volumetric sulphate reduction rate. This wasattributed to sulphide toxicity. Although similar feed sulphateconcentrations were employed, the higher sulphate reduction

etric sulphate reduction rate.

CorrespondingVSLR (g l�1 h�1)

Correspondingresidence time (d)

Correspondingsulphate conversion (%)

0.083 0.5 50

0.035 3.0 42

0.104 2.0 31

0.417 1.0 22

0.000.00

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0.10

Vol

umet

ric

sulp

hate

red

ucti

on r

ate

(gl-1

h-1

)

Volumetric sulphate loading rate (gl-1 h-1)

1.0 g l-1

2.5 g l-1

5.0 g l-1

10.0 g l-1

0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45

e reduction rate. Dotted line distinguishes between regions of the predominance of

ulphate concentration and volumetric sulphate loading rate on volumetric sulphate

ARTICLE IN PRESS

0.00

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0.10

0

Residual sulphate concentration (gl-1)

Vol

umet

ric

sulp

hate

red

ucti

on r

ate

(gl-1

h-1)

1.0 gl-1

2.5 gl-1

5.0 gl-1

10.0 gl-1

3 6 9

Fig. 6. Dependency of volumetric sulphate reduction rate on residual sulphate

concentration.

O.O. Oyekola et al. / Chemical Engineering Science 65 (2010) 4771–4781 4777

efficiency associated with the immobilised cell-systems resultedin sulphide concentrations inhibitory to SRB growth (Baskaranand Nemati, 2006). The maximum VSRR was reduced from1.7 g l�1 h�1 at a feed sulphate concentration of 1.0 g l�1–0.7 g l�1 h�1 at a feed sulphate concentration of 5.0 g l�1. Thecorresponding maximum sulphide concentrations were 0.6 and1.1 g l�1 at feed sulphate concentrations 1.0 and 5.0 g l�1,respectively.

A batch study by Mohanty et al. (2000) showed that sulphatereduction rate decreased with increase in feed sulphate concen-tration from 1.3 to 3.6 g l�1. This was associated with sulphatetoxicity. On the contrary, our study did not follow a particulartrend with respect to the dependency of maximum volumetricsulphate reduction rate on feed sulphate concentration (Table 2).This can be attributed to the fact that lactate supports the growthof a wide variety of microorganisms (SRB and non-SRB). Hence,varying operational conditions may alter the predominance ofdifferent reactions and different microbial population members.

The three possible effects of substrate on the systemsinvestigated in the current study are discussed in the followingorder: sulphate toxicity, sulphide toxicity and volatile acidtoxicity.

(1)

Substrate toxicity: According to White and Gadd (1996), theinhibitory effect of residual sulphate on the BSR in a lactate-fed sulphidogenic system was attributable to its effect on theoperating pH and redox potential. An increasing concentra-tion of residual sulphate increases the redox potential andreduces the operating pH. An increased redox potential wouldselect for non-SRB lactate utilisers in the mixed consortia usedsince SRB are known to thrive at low negative redox potentials(Postgate, 1984; White and Gadd, 1996). In the current study,residual sulphate was recorded in all the reactors across theoperating conditions studied (Figs. 2b–4b). This indicated thatthe microbial activity was not limited by the feed sulphateconcentrations. Additionally, the reactors were maintained atpH 8.070.2. Hence, possible sulphate toxicity by lowering thereactor pH was eliminated. Sulphide, known to result innegative redox potential (Postgate, 1984), was detectableacross the operation range except at the range of high dilutionrates in the reactors receiving 2.5 and 5.0 g l�1 sulphate(Table 1). It can therefore be postulated that in the reactorswith feed sulphate concentrations 2.5 and 5.0 g l�1, lowsulphide levels (0.014–0.088 g l�1) at high dilution rates(0.021–0.042 h�1) could not counteract the positive effect ofresidual sulphate on redox potential. It thus follows that theincrease in residual sulphate may be expected to raise theredox potential at low sulphide concentrations, encouragingthe prevalence of non-SRB lactate utilisers at high dilutionrates. This was postulated to occur in the reactors fed at 2.5and 5.0 g l�1 sulphate. However, in the reactors with feedsulphate concentrations of 1.0 and 10.0 g l�1, high dissolvedsulphide levels were expected to reduce the redox potential.Fig. 6 supports this hypothesis.

(2)

Sulphide toxicity: Previous investigations of acetate-fed sul-phidogenic reactors have shown that 50% growth rateinhibition for a mixed consortium of SRB occurred atHS� concentrations between 0.4 and 1.04 g l�1 (totalsulphide¼0.57–1.11 g l�1) as the pH was varied between7.2 and 8.5 (Visser, 1995; O’Flaherty et al., 1998). Results froman investigation by Kuo and Shu (2004) revealed that acetate-and butyrate-fed anaerobic systems exhibited inhibition bydissolved sulphide (DS) concentrations of 0.15–0.2 g l�1

(0.06–0.075 g l�1 of undissociated hydrogen sulphide), whilethe system depending on lactate operated steadily at elevated

concentration ranges of 0.2–0.4 g l�1 of DS (0.1–0.15 g l�1

undissociated hydrogen sulphide). Additionally, a chemostatstudy of acetate–sulphate systems by Moosa and Harrison(2006) showed that the VSRR decreased with increasing totalsoluble sulphide concentration in the range 0.75–1.45 g l�1,corresponding to an undissociated sulphide concentration of0.07–0.16 g l�1.At pH 8.070.2 maintained in the current study, HS� wasexpected to be the predominant sulphide species. As shownin Table 1, concentrations of sulphide detected (0.014–0.23 g l�1) within the range of dilution rates 0.021–0.042 h�1 (residence times: 2–1 d) in the reactors with feedsulphate concentrations 2.5 and 5.0 g l�1 were below theconcentrations reported to result in inhibition. This indicatedthat the SRB growth rate was not inhibited by sulphide at thehigh dilution rates for these experiments. High sulphideconcentrations which fall within the range at which 50%sulphide inhibition occurs, reported by O’Flaherty et al.(1998), were only detected at low dilution rates 0.0083–0.010 h�1 for the reactor receiving feed sulphate concentra-tion of 5.0 g l�1 and throughout the operation periods for thereactor operated with media containing 10.0 g l�1 sulphate.The investigation by O’Flaherty et al. (1998) showed thatsulphide inhibition is species specific. Hence, level of inhibitionexperienced by each system also depends on the dominantbacterial group(s) under different operating conditions.

(3)

Volatile fatty acid (VFA) toxicity: The inhibitory effect of lactateon SRB growth kinetics has not been reported in the availableliterature. However, acetate in its undissociated form isknown to inhibit SRB activity at low pH values (r6) (Reiset al., 1992). In the current study, the reactor pH wasmaintained at 8.070.2. Owing to this, it is hypothesised thatthe VFA concentrations in the bioreactors were not respon-sible for the decline in the metabolic activity of the SRB.

Against this background, the following hypotheses are presentedto explain the contrasting trends in the profiles of the VSRR inresponse to VSLR across the feed sulphate concentrations studied:

(1)

Shift in lactate metabolism: It may be hypothesised that lactateoxidation which occurs concomitantly with sulphate reduction

ARTICLE IN PRESS

Spec

ific

VSR

R (

gh-1

g-1)

Fig.Dott

oxid

Table 3Effect of feed sulphate concentration on sulphate conversion.

Feed sulphateconcentration(g l�1)

Maximumsulphateconversion(%)

Correspondingvolumetric sulphateloading rate(g l�1 h�1)

Correspondingresidence time(d)

1.0 87 0.0083–0.010 5–4

2.5 54 0.021–0.026 5–4

5.0 58 0.042 5

10.0 40 0.083 5

O.O. Oyekola et al. / Chemical Engineering Science 65 (2010) 4771–47814778

was the predominant reaction throughout the operationperiods in both reactors receiving feed sulphate concentrationsof 1.0 and 10.0 g l�1. On the contrary, this reaction was onlyprevalent at residence times of 5–3 d (D¼0.0083–0.014 h�1) inexperiments with media sulphate concentrations of 2.5 and5.0 g l�1 (Fig. 5). Within this range, the volumetric sulphatereduction rate was positively influenced by the volumetricsulphate loading rate, in the range 1.0–10.0 g l�1. This can besupported by the sulphide levels and the concentration ofacetate produced relative to the inlet sulphate and lactateconcentrations (Table 1; Figs. 2b–4b).

(2)

Influence of sulphide concentration on microbial populations:High levels of sulphide could have selected for active lactate-oxidising SRB in experiments with media concentrations of10.0 g l�1 sulphate, while the non-SRB lactate fermenterswere inhibited. An investigation by Icgen and Harrison (2006)showed that the growth of certain groups of SRB (Desulfo-

bacter, Desulfotomaculum and the Desulfovibrionaceae group)was enhanced by increased sulphide concentrations, up to0.75 g l�1, added into the feed medium. At pH 8.070.2maintained in this study, HCO�3 and HS� are the predominantbicarbonate and sulphide species, respectively. The effect ofsulphide inhibition on SRB at this working pH should beminimal (Kaksonen, 2004). On the contrary, other anaerobicmicroorganisms are known to be strongly inhibited bysulphide (Kalyuzhnyi et al., 1997; O’Flaherty et al., 1998).

As shown in Fig. 3a, at feed sulphate concentrations of2.5 g l�1, decrease in bacterial concentration was gradual despitethe marked decline in VSRR with increase in the volumetricsulphate loading rate. Significant propionate production, anindication of the activity of lactate fermenters, was also evident(Fig. 3b). This suggests that while SRB metabolic activity wasreduced or lost, a significant microbial population of non-SRB wasmaintained by the metabolism of lactate. This observation issupported by the profiles of the specific volumetric sulphatereduction rate as a function of dilution rate (Fig. 7). Specific VSRRwas reduced in the region where fermentation was prevalent forthe experiments with feed sulphate concentrations of 2.5 and5.0 g l�1. This further supports the postulation of significant shiftsin the dominant microbial population from lactate oxidisers to

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0

Dilution rate (h-1)

F

O

1.0 g l-1

2.5 g l-1

5.0 g l-1

10.0 g l-1

0.02 0.04 0.06 0.08 0.1

7. Dependency of specific volumetric sulphate reduction rate on dilution rate.

ed line distinguishes between regions of the predominance of lactate

ation (O) and fermentation (F) when So¼2.5 and 5.0 g l�1.

lactate fermenters as the dilution rate was increased for these twochemostat cultures.

On the other hand, there was a consistent increase in specificVSRR throughout the experiments with feed concentrations of 1.0and 10.0 g l�1 sulphate. This implies that despite the decrease inthe total microbial concentration, significant SRB activity wasretained. The shift in lactate metabolic pathway and the responseof the mixed population to varying operating conditions aredescribed in detail elsewhere (Oyekola et al., 2007, 2009).

3.2.2. Sulphate conversion

As depicted in Table 3 and Fig. 8, an increase in feed sulphateconcentration has a negative influence on the sulphateconversion. The maximum sulphate conversion achieved at eachfeed sulphate concentration was at the lowest dilution rateapplied (Table 3). In contrast to the other reactors, in the 1.0 g l�1

sulphate-fed reactor, high sulphate conversion of 8670.5 % wasmaintained over a wide range of dilution rates (0.0083–0.042 h�1). As the feed sulphate concentration was increased inthe range 2.5–10.0 g l�1 decreasing sulphate conversion withincreasing loading rate of sulphate was evident. Fig. 8a indicatesthat there is no consistent relationship between lactateconversion and sulphate conversion as the feed sulphateconcentration was varied between 1.0 and 10.0 g l�1.

Alternatively, across the feed sulphate concentrations studied,at low dilution rates, the sulphate removed remained relativelyconstant despite increasing acetate production. At high dilutionrates there was a marked decrease in sulphate removal accom-panied by a slight decrease in acetate production (Fig. 8b). Thisphenomenon is attributable to the production of acetate fromother non-SRB microorganisms. Lactate is known to be fermentedto acetate and propionate by some non-SRB lactate fermenterssuch as Veillonella parvula, Propionibacterium spp. and Propioni-

bacterium freudenreichii (Eq. (5)) (Seeliger et al., 2007). Thispattern is congruent with increasing propionate productionwithin this range of dilution rates (Fig. 8c).

Similar to the results obtained in the current study, Baskaranand Nemati (2006) demonstrated that maximum sulphateconversion of 100% was achieved when the feed sulphateconcentrations were 1.0 and 2.5 g l�1 in lactate-fed immobi-lised-cell systems. Increasing the feed sulphate concentration to5.0 g l�1 led to a decline in the sulphate conversion to 90%.Furthermore, a study by Kalyuzhnyi et al. (1997) showed thatsulphate conversion was maintained at 80% when the influentsulphate concentration of an ethanol-fed UASB reactor was below1.34 g l�1 h�1 at a constant hydraulic retention time. Above thisfeed sulphate concentration, the sulphate removal efficiency wasreduced to 62%. This was associated with decrease in pH to 6.8and the resulting increased levels of undissociated sulphide. Sincethe pH was kept constant at 870.2 in this present study, theeffect of undissociated sulphide on the sulphate conversion wasnegligible. Previous studies have shown that sulphate conversion

ARTICLE IN PRESS

0

10

20

30

40

50

60

70

80

90

100

0

Lactate conversion (%)

Sulp

hate

con

vers

ion

(%)

1.0 g l-1

2.5 g l-1

5.0 g l-1

10.0 g l-1

Lowest D Highest D Highest D

0.000

0.005

0.010

0.015

0.020

0.025

0.030

0.035

0.040

0.045

00.020.040.060.080.1

Acetate produced (mol I-1)

Sulp

hate

rem

oved

(m

ol l-1

)

1.0 g l-1

2.5 g l-1

5.0 g l-1

10.0 g l-1

0.000

0.005

0.010

0.015

0.020

0.025

0.030

0.035

0.040

0.045

00.020.040.060.08

Propionate produced (mol lI-1)

Sulp

hate

rem

oved

(m

ol l

-1)

1.0 g l-1

2.5 g l-1

5.0 g l-1

10.0 g l-1

20 40 60 80 100 120

Lowest D

a

Fig. 8. (a) Relationship between sulphate conversion and lactate conversion; (b) Dependency of sulphate removed on acetate produced; (c) Dependency of sulphate

removed on propionate produced.

O.O. Oyekola et al. / Chemical Engineering Science 65 (2010) 4771–4781 4779

is limited by the stoichiometry of lactate metabolism (Oyekolaet al., 2009).

3.2.3. Bacterial concentration

As shown in Table 4, increased feed sulphate concentration inthe range 1.0–10.0 g l�1 resulted in increased bacterial con-centration. Mohanty et al. (2000) reported that the sulphatereduction rate was positively influenced by biomass con-centration in a batch culture. A 4-fold increase of the biomassconcentration resulted in a 1.5-fold increase of the sulphate

reduction rate (Mohanty et al., 2000). In the current study,maximum bacterial concentrations were obtained at the lowestdilution rate (residence time¼5 d) in all experiments. This issimilar to the findings of Moosa et al. (2002), which showed thatat a given dilution rate the bacterial concentration increased withincrease in the feed sulphate concentration.

Further, the specific VSRR profiles shown in Fig. 7 suggests thatdespite the loss in bacterial sulphate reduction activity in theexperiments with feed sulphate concentrations of 2.5 and 5.0 g l�1,at high dilution rates, the bacterial growth was maintained via analternative lactate utilisation pathway. High propionate production

ARTICLE IN PRESS

Table 4Effect of feed sulphate concentration on bacterial growth at residence time 5 d.

Feed sulphateconcentration(g l-1)

Maximum bacterialconcentration (g l�1)

Corresponding volumetricsulphate loading(rate g l�1 h�1)

1.0 0.267 0.083

2.5 0.350 0.10

5.0 0.458 0.21

10.0 0.588 0.42

O.O. Oyekola et al. / Chemical Engineering Science 65 (2010) 4771–47814780

suggested the proliferation of non-SRB lactate fermenters in theseexperimental runs. It has been suggested that lactate-fed SRB’s areprone to competition from other microorganisms when present inmixed cultures (Laanbroek et al., 1982). Should competition occurunder the experimental conditions used, the bacterial concentra-tions measured would account for both SRB and non-SRB. A previousfinding by Oyekola et al. (2009) showed that lactate fermentationwhich is independent of sulphate reduction was the prevailinglactate degradation pathway under these conditions (feed sulphateconcentration¼2.5 and 5.0 g l�1; residence time¼0.5–3 d).

4. Conclusions

It was evident from this study that volumetric sulphateloading rate, mediated through both sulphate concentration inthe feed and dilution rate significantly influenced the kinetics ofBSR. The results suggest a link between the BSR processstoichiometry and the kinetics. Incomplete oxidation of lactate,which supports sulphate reduction, was dominant in the range offeed sulphate concentrations of 1.0–10.0 g l�1 and residencetimes of 5–3 d. Hence, the volumetric sulphate reduction ratewas enhanced under these conditions and a correlation of VSRR toVSLR was observed. At feed sulphate concentrations of 1.0 and10.0 g l�1, there was a consistent increase in the volumetricsulphate reduction rate with increasing volumetric sulphateloading rate across the residence time range 5–0.5 d. This wasattributed to the predominance of lactate oxidation reactionunder these conditions (Oyekola et al., 2009). The significantdecrease in VSRR observed for feed sulphate concentrations of 2.5and 5.0 g l�1 at high dilution rates (residence times of 1–2 d), wasnot commensurate with a decrease in bacterial concentration.This observation implied a shift in the primary metabolic pathwayfrom lactate oxidation to fermentation. This was corroborated bythe specific VSRR data which exhibited trends similar to that ofthe VSRR. Despite the higher bacterial concentrations recorded atfeed sulphate concentrations of 2.5 and 5.0 g l�1, relative to thevalues observed at feed sulphate concentration of 1.0 g l�1, higherVSRR and specific VSRR values, with increased dilution rate wereobserved in the latter. Data obtained from this study suggest thatlactate fermenters were inhibited by high sulphide concentrations(0.3–0.6 g l�1) obtained during operation at a feed sulphateconcentration of 10.0 g l�1. Consequently, the oxidative pathwayfor lactate utilisation dominated.

Acknowledgements

The authors gratefully acknowledge the financial support fromthe National Research Foundation (NRF), Claude Leon Foundation,the University of Cape Town (UCT) and the South African ResearchChairs Initiative of the Department of Science and Technology andthe National Research Foundation during the course of thisresearch.

References

American Public Health Association (APHA), 1975. Standard Methods for theExamination of Water and Wastewater Fourteenth ed. APHA, New York.

Baker, B.J., Banfield, J.F., 2003. Microbial communities in acid mine drainage. FEMSMicrobiology Ecology 44, 139–152.

Baskaran, V., Nemati, M., 2006. Anaerobic reduction of sulfate in immobilized cellbioreactors, using a microbial culture originated from an oil reservoir.Biochemical Engineering Journal 31, 148–159.

Boonstra, J., van Lier, R., Janssen, G., Dijkman, H., Buisman, C.J.N., 1999. Biologicaltreatment of acid mine drainage. In: Amils, R., Ballester, A. (Eds), Biohydro-metallurgy and the Environment Toward the mining of the 21st century.Proceedings of the International Biohydrometallurgy Symposium IBS’99, SanLorenzo de El Escorial, Madrid, Spain, June 20–23. Part B, Molecular Biology,Biosorption, Bioremediation. Amsterdam, Elsevier, pp. 559–67.

Cenni, R., Janisch, B., Spliethoff, H., Hein, K.R.G., 2001. Legislative and environ-mental issues on the use of ash from coal and municipal sewage sludge co-firing as construction material. Waste Management 21, 17–31.

Cline, J.D., 1969. Spectrophotometric determination of hydrogen sulfide in naturalwaters. Limnology and Oceanography 14, 454–458.

Corbett, C.J., 2000. Bioremediation of acid mine drainage using sewage sludge. MScThesis. Department of Biochemistry and Microbiology, Rhodes Universiy,Grahamstown, South Africa.

Drury, W.J., 1999. Treatment of acid mine drainage with anaerobic solid substratereactors. Water and Environmental Research 71, 1244–1250.

Erasmus, C.L., 2000. A preliminary investigation of the kinetics of biological sulphatereduction using ethanol as a carbon source and electron donor. Masters Thesis,Department of Chemical Engineering, University of Cape Town, South Africa.

Ghigliazza, R., Lodi, A., Rovatti, M., 2000. Kinetic and process considerations onbiological reduction of soluble and scarcely soluble sulfates. Resources,Conservation and Recycling 29, 181–194.

Hansford, G., Gopal, H., Harrison, S., van Hille, R., Icgen, B., Jacobs, T., Lewis A.,Moosa S., Pillay, V., 2007. An extended investigation into the mechanism andkinetics of bacterial sulphate reduction. WRC Report No. 1251/1/07.

Heimann, A.C., Friis, A.K., Jakobsen, R., 2005. Effects of sulfate on anaerobicchloroethene degradation by an enriched culture under transient and steady-state hydrogen supply. Water Research 39, 3579–3586.

Hulshoff Pol, L.W., Lens, P., Stams, A.J.M., Lettinga, G., 1998. Anaerobic treatment ofsulphate-rich wastewaters, microbial and process technological aspects.Biodegradation 9, 213–224.

Hulshoff Pol, L.W., Lens, P.N.L., Weijma, J., Stams, A.J.M., 2001. New developmentsin reactor and process technology for sulfate reduction. Water Science andTechnology 44, 67–76.

Icgen, B., Harrison, S., 2006. Exposure to sulfide causes populations shifts insulfate-reducing consortia. Research in Microbiology 157, 784–791.

Johnson, D.B., Hallberg, K.B., 2003. The microbiology of acidic mine waters.Research in Microbiology 154, 466–473.

Kaksonen, A., 2004. The performance, kinetics and microbiology of sulfidogenicfluidized-bed reactors treating acidic metal- and sulphate-containing waste-water. Thesis for the Degree of Doctor of Technology, Tampere University ofTechnology, Tampere, Finland.

Kaksonen, A.H., Plumb, J.J., Franzmann, P.D., Puhakka, J.A., 2004. Simple organicelectron donors support diverse sulfate-reducing communities in fluidized-bed reactors treating acidic metal- and sulfate-containing wastewater. FEMSMicrobiology Ecology 47, 279–289.

Kalyuzhnyi, S.V., de Leon Fragoso, C., Rodriguez Martinez, J., 1997. Biologicalsulfate reduction in a UASB reactor fed with ethanol as the electron donor.Microbiologiya 66, 687–693.

Kuo, W.-C., Shu, T.-Y., 2004. Biological pre-treatment of wastewater containingsulfate using anaerobic immobilized cells. Journal of Hazardous Materials 113,147–155.

Laanbroek, H.J., Abee, T., Voogd, I.L., 1982. Alcohol conversions by Desulfobulbuspropionicus Lindhorst in the presence and absence of sulphate and hydrogen.Archives of Microbiology 133, 178–184.

Mohanty, S.S., Das, T., Mishra, S.P., Roy Chaudhury, G., 2000. Kinetics of SO2�4

reduction under different growth media by sulphate-reducing bacteria.BioMetals 13, 73–76.

Moosa, S., 2000. A kinetic study on anaerobic sulphate reduction: effect of sulphateand temperature. Ph.D. Thesis, Department of Chemical Engineering, Uni-versity of Cape Town, South Africa.

Moosa, S., Harrison, S.T.L., 2006. Product inhibition by sulphide species onbiological sulphate reduction for the treatment of acid mine drainage.Hydrometallurgy 83, 214–222.

Moosa, S., Nemati, M., Harrison, S.T.L., 2002. A kinetic study on the anaerobicreduction of sulphate, part I, effect of sulphate concentration. ChemicalEngineering Science 57, 2773–2780.

Naicker, K., Cukrowska, E., McCarthy, T.S., 2003. Acid mine drainage arising fromgold mining activity in Johannesburg, South Africa and environs. Environ-mental Pollution 122, 29–40.

O’Flaherty, V., Mahony, T., O’Kennedy, R., Colleran, E., 1998. Effect of pH on growthkinetics and sulphide toxicity thresholds of a range of methanogenic, syntrophicand sulphate-reducing bacteria. Process Biochemistry 33, 555–569.

Oyekola, O.O., van Hille, R.P., Harrison, S.T.L., 2007. Effect of sulphate concentrationon the community structure and activity of sulphate-reducing bacteria.Advanced Materials Research 20-21, 513–515.

ARTICLE IN PRESS

O.O. Oyekola et al. / Chemical Engineering Science 65 (2010) 4771–4781 4781

Oyekola, O.O., 2008. An investigation into the relationship between processkinetics and microbial community dynamics in a lactate-fed sulphidogenicCSTR as a function of residence time and sulphate loading. Ph.D. Thesis,Department of Chemical Engineering, University of Cape Town, South Africa.

Oyekola, O.O., van Hille, R., Harrison, S.T.L., 2009. Study of anaerobic lactatemetabolism under biosulphidogenic conditions. Water Research 43, 3345–3354.

Postgate, J.R., 1984. The Sulphate-Reducing Bacteria second ed. CambridgeUniversity Press, UK.

Pulles, W., Howie, D., Otto, D., Easton, J.,1995. A manual on mine water treatmentand management practices in South Africa. Water Research Commission,Report No. TT 80/96.

Reis, M.A.M., Almeida, J.S., Lemos, P.C., Carrondo, M.J.T., 1992. Effect of hydrogensulphide on growth of sulphate-reducing bacteria. Biotechnology andBioengineering 40, 593–600.

Rose, P., Pletschke, B., Whiteley, C., 2000. Complex organic carbon compounds as electrondonors for sulphate reduction—the Rhodes ‘‘Biosure’’ process in the treatment ofmine drainage wastewaters. In: Proceedings of the Technology Transfer Workshopon Biological Sulphate Removal of the Anaerobic Processes Division of the WaterInstitute of South Africa, 22 and 23 August, Pretoria, South Africa.

Seeliger, S., Janssen, P.H., Schink, B., 2007. Energetics and kinetics of lactatefermentation to acetate and propionate via methylmalonyl-CoA or acrylyl-CoA. FEMS Microbiology 211, 65–70.

Thompson, G., Swain, J., Kay, M., Forster, C.F., 2001. The treatment of pulp andpaper mill effluent, a review. Bioresource Technology 77, 275–286.

van Houten, B.H.G.W., Roest, K., Tzeneva, V.A., Dijkman, H., Smidt, H., Stams, A.J.M.,2006. Occurrence of methanogenesis during start-up of a full-scale synthesisgas-fed reactor treating sulfate and metal-rich wastewater. Water Research 40,553–560.

Visser, A., 1995. The anaerobic treatment of sulfate containing wastewater.Doctoral Thesis, Wageningen Agricultural University, Wageningen, TheNetherlands.

White, C., Gadd, G.M., 1996. Mixed sulphate-reducing bacterial cultures forbioprecipitation of toxic metals, factorial and response-surface analysis of theeffects of dilution rate, sulphate and substrate concentration. Microbiology142, 2197–2205.

Whittington-Jones, K., Molwantwa, J.B., Rose, P.D., 2006. Enhanced hydrolysis ofcarbohydrates in primary sludge under biosulphidogenic conditions. WaterResearch 40, 1577–1582.