Batch and continuous biooxidation of sulphide by Thiomicrospira sp. CVO: Reaction kinetics and stoichiometry

ARTICLE IN PRESS WAT E R R E S E A R C H

40 (2006) 2436– 2446

Available at www.sciencedirect.com

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

Batch and continuous biooxidation of sulphide by Thiomicrospira sp. CVO: Reaction kinetics and stoichiometry S. Gadekar, M. Nemati, G.A. Hill Department of Chemical Engineering, College of Engineering, University of Saskatchewan, 57 Campus Drive, Saskatoon, Sask., Canada S7N 5A9

ar t ic l e i n f o

A B S T R A C T

Article history:

Aqueous phase biooxidation of sulphide by the novel sulphide-oxidizing bacterium

Received 10 December 2005

Thiomicrospira sp. CVO was studied in batch and continuous systems. CVO was able to

Received in revised form

oxidize sulphide at concentrations as high as 19 mM. Sulphide biooxidation occurred in two

18 March 2006

distinct phases, one resulting in the formation of sulphur and possibly other dissolved

Accepted 12 April 2006

sulphur compounds rather than sulphate, followed by sulphate formation. The specific

Available online 30 May 2006

growth rate of CVO in the first and second phases were 0.17–0.27 and 0.04–0.05 h
1,

Keywords:

respectively. Nitrite accumulated in the culture during the first phase and was consumed

Sulphide-laden water

during the second phase. The composition of end-products was influenced by the ratio of

Sulphide biooxidation

sulphide to nitrate initial concentrations. At a ratio of 0.28, sulphate represented 93% of the

Nitrate reduction

reaction products, while with a ratio of 1.6 the conversion of sulphide to sulphate was only

Thiomicrospira sp. CVO

9.3%. In the continuous bioreactor, complete removal of sulphide was observed at sulphide

Continuous bioreactor

volumetric loading rates as high as 1.6 mM/h (residence time of 10 h). Overall sulphide

Kinetic modelling

removal efficiency decreased continuously upon further increases in volumetric loading rate. However, the volumetric removal rate increased until a maximum value of 2.4 mM/h was obtained at a loading rate of 3.2 mM/h. The corresponding sulphide conversion and residence time were 76% and 5.6 h, respectively. As expected from the high ratio of sulphide to nitrate loading rates (1.7–1.9 mM/h), no sulphate was formed in the continuous reactor. Using the experimental data the value of maximum specific growth rate, saturation constant, decay coefficient, maintenance coefficient and yield were determined to be 0.36 h
1, 1.99 mM sulphide, 0.0014 h
1, 0.078 mmol sulphide/mg ATP h and 0.018 mg ATP/ mmol sulphide, respectively. & 2006 Elsevier Ltd. All rights reserved.

1.

Introduction

Biogenic production of H2S in oil and gas reservoirs, is a major cause of contamination of oil products with sulphur compounds (McComas and Sublette, 2001). During secondary oil production, water is often injected into the reservoir to maintain pressure and to enhance the recovery of residual oil. The water which is coproduced with oil contains sulphides as a result of the activity of sulfidogenic bacteria Corresponding author. Tel.: +1 306 966 4769; fax: +1 306 966 4777.

E-mail address: [email protected] (M. Nemati). 0043-1354/$ - see front matter & 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.watres.2006.04.007

(Gevertz et al., 2000). The removal of sulphide from the produced fluids is essential due to the toxic and corrosive nature of sulphides, and for preventing emission of sulphur oxides upon combustion of fuels. Options for the treatment of sulphide include physicochemical and biological processes. Claus, Alkanolamine, Lo–Cat and Holmes–Stretford are typical physicochemical processes for removal of sulphide (Iliuta and Larachi, 2003; Krishnakumar et al., 2005). Physicochemical processes are in general energy and cost intensive.

ARTICLE IN PRESS WAT E R R E S E A R C H

2437

4 0 (200 6) 243 6 – 244 6

Nomenclature

ms

D K kd KS KSX

n rs substrate utilization rate (mM sulphide/h) S substrate concentration (mM sulphide) X biomass concentration (mg ATP/L) YX=S yield coefficient (mg ATP/mmol sulphide) mg, mm, mnet gross, maximum and net specific growth

dilution rate (h
1) constant (mM sulphide)
1 decay coefficient (h
1) saturation constant (mM sulphide) constant (mM sulphide/mg ATP L
1)

maintenance ATP h) constant

coefficient

(mmol

sulphide/mg

rates (h
1)

Biological processes, by contrast, operate at atmospheric pressure and low temperatures without the need for expensive chemicals and catalysts. Additionally, biological processes offer high removal efficiencies even with low sulphide concentrations and can be used for the treatment of small volumetric flows (Henshaw and Zhu, 2001). Biological treatment of sulphide-containing streams are categorized as direct and indirect methods. The indirect methods rely on the oxidizing power of ferric iron for conversion of sulphide to elemental sulphur and catalytic activity of iron-oxidizing bacteria for the regeneration of ferric iron ( Pandey et al., 2004; Park et al., 2005),while in the direct approach photoautotrophic or chemolithotrophic sulphide-oxidizing bacteria are used (Henshaw et al., 1998; Henshaw and Zhu, 2001). The simpler nutritional and energy requirements has made the chemolithotrophs a favourable candidate for biooxidation of sulphide (Sublette and Sylvester, 1987; Sublette, 1987; Lee and Sublette, 1993; Sublette et al., 1998; Elias et al., 2002; Alcantara et al., 2004; Cytryn et al., 2005). Despite the significant amount of research on biooxidation of sulphide, the sensitivity of bacteria to high levels of sulphide has remained a main technical barrier to widespread application of this technology. Recent work on the microbiology of oil fields has led to the isolation and identification of a novel sulphide-oxidizing bacterium designated as Thiomicrospira sp. CVO (referred to as CVO for the remaining part of this paper) from a Canadian oil field (Gevertz et al., 2000). CVO is a microaerophile, chemolitothroph which uses either sulphide or sulphur as an electron donor and nitrate or nitrite as an electron acceptor (Gevertz et al., 2000). The capability of CVO to use nitrate as a terminal acceptor makes it a suitable candidate for oxidation and removal of sulphide under anaerobic conditions (i.e. in situ removal of sulphide from oil reservoirs). Preliminary studies with CVO have concentrated mainly on the physiological and microbiological aspects, and the role which this bacterium plays in the control of souring in western Canadian oil reservoirs (Nemati et al., 2001; Hubert et al., 2003). Greene et al. (2003) reported that CVO was able to oxidize sulphide at concentrations up to 15 mM, a level significantly higher than that demonstrated by Thimicrospira denitrificans, Arcobacter sp. FWKOB and Thiobacillus denitrificans strain F. Using a microbial enrichment dominated by CVO for the treatment of a gaseous stream containing H2S, McComas and Sublette (2001) achieved a sulphide removal rate of 3 mmol H2S/h g biomass. The enrichment was shown to be more tolerant of extremes in pH, elevated temperature and salinity.

Considering the limited amount of work on biooxidation of sulphide by CVO and the important role which this bacterium could play in ex situ treatment of sulphide-laden waters, as well as in situ removal of sulphide in oil reservoirs under anaerobic conditions, further work is clearly required to explore the potential of this novel biocatalyst. The present paper describes the results of a kinetic study on biooxidation of sulphide by CVO. The data collected in the batch system provides insight regarding the effects of sulphide and nitrate concentrations on the activity of the bacteria and composition of sulphide biooxidation end products, while kinetic data generated in the continuous system have been used to establish the underlying principles governing the kinetics of microbial growth and sulphide biooxidation.

2.

Materials and methods

2.1.

Microbial culture and medium

A pure culture of Thiomicrospira sp. CVO (NRRL-B-21472), was kindly provided by Dr. G. Voordouw, University of Calgary, Canada, and was used in this study. A modified form of Coleville Synthetic Brine (CSB; Gevertz et al., 2000) containing 2–3 mM sodium sulphide and 10 mM nitrate was used for the growth and maintenance of CVO. The modified CSB made with reverse osmosis water contained: 120 mM NaCl; 2.75 mM MgSO4  7H2O; 1.6 mM, CaCl2  2H2O, 0.38 mM NH4Cl, 0.2 mM, KH2PO4, 8.3 mM NaCH3COO  3H2O, 10 mM KNO3, 22.6 mM NaHCO3, 50 mM tris (hydroxymethyl) methylamine and 0.5 ml of trace element solution. The trace element solution contained: 0.5 mL concentrated H2SO4 (98%), 13.5 mM MnSO4  H2O, 1.75 mM ZnSO4  7H2O, 8.1 mM H3BO3, 0.1 mM CuSO4  5H2O, 0.1 mM Na2MoO4  2H2O, 0.19 mM CoCl2  6H2O and 3.6 mM FeCl3. All medium components, except sodium sulphide, were combined and the pH was adjusted to 7.0, using 2 M HCl. Serum bottles (125 mL total volume) containing 100 mL medium were purged with nitrogen gas for 5 min, sealed and autoclaved for 30 min at 121 1C. Filter sterilized stock solution of Na2S  9H2O (1 M) was added aseptically to the medium to a final concentration of 2–3 mM and pH was readjusted to 7.0. A stock culture of CVO was used as inoculum (10% v/v). The cultures were maintained at 22 1C and subcultured on a weekly basis.

ARTICLE IN PRESS 2438

WAT E R R E S E A R C H

2.2.

Experimental systems and procedures

2.2.1.

Batch experiments

40 (2006) 2436– 2446

The effect of initial sulphide concentration on the activity of CVO was studied in 250 mL flasks each containing 200 mL of modified CSB medium with 10 mM nitrate and 6.3, 8.7, 12.8 or 16.2 mM sulphide. A 2-day-old CVO culture was used as an inoculum (10% v/v). All experiments were carried out at 22 1C. Concentrations of sulphide, sulphate and nitrite were monitored during the course of the experiments. ATP concentration (as an indication of biomass concentration) was monitored in two experiments (6.3 and 8.7 mM initial sulphide). To assess the effect of available nitrate on the composition of end products, a number of experiments were carried out using modified CSB containing 20 mM nitrate and 5.5, 13 or 22 mM sulphide. The extent of abiotic oxidation of sulphide was assessed by conducting two sets of experiments using sterilized modified CSB medium (10 mM of nitrate and 4.6 or 20 mM sulphide) with no inoculation. All other conditions were similar to those described earlier.

liquid medium (520 ml) was pumped into the sterilized bioreactor. The bioreactor was inoculated by injection of 60 ml of a 2-day-old CVO culture. To maintain the anaerobic conditions, filter sterilized nitrogen was introduced into the bioreactor headspace at a low flow rate. Initially the bioreactor was operated batch-wise. Once complete sulphide removal was achieved, the bioreactor was switched to continuous mode and medium was pumped into the bioreactor at a flow rate of 8.7 ml/h. The flow rate of the feed was then increased stepwise. At each flow rate sufficient time was allowed for the establishment of steady state conditions. The stepwise increase in flow rate of the feed continued up to a value of 134.5 ml/h. A substantial decrease in conversion of sulphide and ATP concentration was observed at this flow rate and as a result higher flow rates were not tested. Concentrations of sulphide, sulphate, nitrite and effluent pH were monitored on a daily basis. Concentration of ATP at each flow rate were determined after establishment of steady state conditions. In all cases, analysis of the samples was repeated three times.

2.3. 2.2.2.

Analytical procedures

Kinetic study in continuous bioreactor

In order to determine the kinetics of microbial growth and sulphide biooxidation by CVO, a set of experiments was conducted in a continuous flow bioreactor. The experimental set-up (Fig. 1) consisted of a glass bioreactor (580 ml working volume), placed on a magnetic stirrer, a multi-speed peristaltic pump, 5 l sterile plastic medium bags and an effluent container. Modified CSB medium containing 20 mM sulphide (average: 17.870.8 mM) and 10 mM nitrate was used as the feed. Medium was prepared in 5 l Erlenmeyer flasks, purged with filter sterilized nitrogen gas for 2–3 h and sterilized for 30 min at 121 1C. The cooled medium was then pumped aseptically to a plastic culture medium bag. A concentrated solution of sodium sulphide (1 M) was then added to the medium and the pH of the mixture was adjusted to 7. The

The concentrations of sulphide, sulphate and nitrite were determined using spectrophotometric methods described elsewhere (Cord-Ruwisch, 1985; APHA, 1992). Biomass concentration was monitored indirectly by measuring the ATP concentration (Shuler and Kargi, 2002). Presence of sulphur particles made the direct measurement of biomass in terms of cell number or dry weight impractical. Measurement of ATP was performed using a BD MoonlightTM 3010C Luminometer, Turner Designs ATP releasing reagent with phosphate inhibitor (Biochemical Diagnostics) and rLuciferase/rLuciferin reagent (Promega) that were added (50 ml of each) to 50 ml of sample at room temperature. The luminescence of the resulting mixture was measured and correlated with ATP concentration using standard solutions of ATP.

Plastic medium bag

Nitrogen gas

Filter Gas vent

Peristaltic Pump

Bioreactor

Effluent container

Fig. 1 – Schematic diagram of the experimental set-up used for continuous biooxidation of sulphide.

ARTICLE IN PRESS WAT E R R E S E A R C H

3.

Results and discussion

3.1.

Batch experiments

4 0 (200 6) 243 6 – 244 6

In the control systems maintained under aseptic conditions, initial decreases in concentrations of sulphide were observed. With 4.3 mM sulphide, the initial decrease amounted to 8% of the total sulphide present, while with 20 mM sulphide a 15% decrease in sulphide concentration was observed. No significant changes in sulphate and nitrite concentrations were observed throughout the abiotic experiments. The initial decrease in sulphide concentration could be attributed to either transfer of sulphide from the liquid phase to the head space gas or abiotic oxidation of sulphide. The results of biooxidation of sulphide at initial concentrations of 6.3, 8.7, 12.8 and 16.2 mM and with 10 mM nitrate are shown in Fig. 2. With 6.3 mM of sulphide (Fig. 2A), the sulphide concentration reached to a negligible value in less than 13 h (removal rate: 0.5 mM/h). Oxidation of sulphide led to production of nitrite and increases in ATP concentration. Turbidity of the culture increased and suspended particles were observed in the liquid. The concentration of sulphate, however, did not change significantly, an indication that sulphur or other soluble sulphur compounds (i.e. thiosulphate) were the main products of CVO metabolism. Biooxidation of sulphide to elemental sulphur in the presence of nitrate could occur through two different reactions as shown below: 
þ HS
þ NO
3 þ H ! S þ NO2 þ H2 O;

(1)

 þ HS
þ 0:4 NO
3 þ 1:4 H ! S þ 0:2 N2 þ 1:2 H2 O:

(2)

The stoichiometry of Reaction 1 indicates that oxidation of each mole of sulphide results in formation of one mole nitrite. Using the experimental data, sulphide biooxidation and nitrite production rates were determined to be 0.50 and 0.51 mM/h, respectively. This together with a constant sulphate concentration during the biooxidation of sulphide, and the presence of suspended particles in the culture implied that during this phase sulphide was mainly oxidized to elemental sulphur with concomitant reduction of nitrate to nitrite via Reaction 1. Formation of sulphate was observed as soon as the sulphide concentration reached a negligible level. ATP concentration continued to increase and turbidity of the culture decreased, indicating the oxidation of sulphur to sulphate. Nitrite concentration increased initially and then decreased. Oxidation of elemental sulphur to sulphate could occur through three different reactions as outlined below: 2

þ þ S þ 3 NO
3 þ H2 O þ 3 H ! SO4 þ 3 NO2 þ 5 H ,

(3)

2
þ þ S þ 1:2 NO
3 þ 0:4 H2 O þ 1:2 H ! SO4 þ 0:6 N2 þ 2 H ,

(4)

2
þ þ S þ 2 NO
2 þ 2 H ! SO4 þ N2 þ 2 H .

(5)

Based on the experimental data collected during sulphate formation, the rates of sulphate production, nitrite production followed by nitrite reduction were 0.17, 0.39 and
0.33 mM/h, respectively. A comparison of the experimental rates reveals that the nitrite production rate was less than

2439

that expected from the stoichiometry of Reaction 3 and as a result Reactions 3 and 4 must have occurred simultaneously. The nitrite reduction rate observed during the latter stage of sulphate formation was almost twice as high as that of the sulphate production rate which is consistent with the stoichiometry of Reaction 5. The growth rate of bacteria during the initial phase of sulphide oxidation was significantly faster than that observed during the formation of sulphate, with the value of specific growth rates being 0.27 and 0.04 h
1, respectively. With 8.7 mM sulphide, oxidation of sulphide occurred at a rate of 0.57 mM/h, with a corresponding nitrite reduction rate of 0.59 mM/h, again consistent with the stoichiometry of Reaction 1. Sulphate production occurred at a relatively constant rate of 0.19 mM/h as soon as the sulphide concentration decreased to a negligible level (Fig. 2B). The production rate of nitrite (0.18 mM/h) was lower than that expected from the stoichiometry of Reaction 3, while the nitrite reduction rate of 0.34 mM/h was close to that expected from Reaction 5. The specific growth rate of bacteria during the oxidation of sulphide and formation of sulphate were 0.17 and 0.05 h
1, respectively. With higher initial sulphide concentrations long lag phases were observed before initiation of sulphide biooxidation. Sulphide removal rates of 0.7 and 0.43 mM/h were measured in the cultures initially containing 12.8 and 16.2 mM sulphide, respectively (Figs. 2C and D). The production rate of nitrite during the oxidation of sulphide and nitrite reduction rate during the formation of sulphate were significantly lower than that expected from Reactions 1 and 5. In the cultures initially containing 5.5, 13 and 22 mM of sulphide and 20 mM nitrate, the concentration profiles for sulphide, nitrite and sulphate were similar to those observed in the cultures containing 10 mM nitrate (Fig. 3). The increase in initial sulphide concentration led to a longer lag phase in the bacterial activity. Furthermore, with 5.5 mM sulphide, the concentration of produced nitrite remained at a high level (maximum 9 mM) for an extended period. As observed with 10 mM nitrate, at high sulphide concentrations (13 and 22 mM), formation of sulphate initiated before complete removal of sulphide. In order to establish a criterion for controlling the composition of end-products, the extent of sulphate formation was related to the initial ratio of sulphide to nitrate concentrations. The information compiled in Fig. 4 shows that at low values of sulphide to nitrate ratio, sulphate is the main product and as this ratio is increased the conversion of sulphide to sulphate is decreased. For instance, at a ratio of 0.28, sulphate constitutes 93% of the reaction products, while with a sulphide to nitrate ratio of 1.6 the conversion of sulphide to sulphate is only around 9.3%. This is consistent with information in the literature regarding the effects of oxygen on the oxidation state of sulphur compounds during the aerobic biooxidation of sulphide. Using a mixed culture consisting of Thiobacilli species, Janssen et al. (1995) found that at sulphide loading rates up to 2.33 mM/h and oxygen concentrations below 0.1 mg/l both sulphur and sulphate were produced, while under highly limited oxygen conditions thiosulphate was the main product. Alcantara et al. (2004) used the ratio of oxygen to sulphide loading rates (Rmt) to control the biooxidation of sulphide in a recirculation reactor

ARTICLE IN PRESS 2440

WAT E R R E S E A R C H

40 (2006) 2436– 2446

10

0.12

(A)

8

0.09

6 0.06 4

0.00

0 10

0.12

(B)

8

0.09

( ) ATP (mg L-1)

0.03

2

6

( ) Sulphide, ( ) sulphate and ( )nitrite (mM)

0.06 4 0.03

2 0

0.00 0

10

20

30

40

50

20 (C)

16 12 8 4 0 20

(D)

16 12 8 4 0 0

50

100

150

200

250

300

350

400

Time (h) Fig. 2 – Profiles of sulphide, sulphate, nitrite and ATP concentrations in the batch cultures of Thiomicrospirs sp. CVO with 10 mM nitrate and various initial sulphide concentrations: (A) 6.3, (B) 8.7, (C) 12.8, and (D) 16.2 mM.

system. Three stages of partial sulphide oxidation producing sulphur for Rmt of 0.5–1.5, complete oxidation of sulphide to sulphate for Rmt from 1.5 to 2, and low sulphide oxidation at Rmt below 0.5 were distinguished. Using T. denitrificans strain D-4, Wang et al. (2005) demonstrated that both sulphide

concentration and the influent sulphide to nitrate ratio were key factors. The suitable levels for maintaining sulphur as the main product were suggested to be less than 9 mM sulphide and an influent sulphide to nitrate ratio in the range 1.6–2.5, respectively.

ARTICLE IN PRESS WAT E R R E S E A R C H

2441

4 0 (200 6) 243 6 – 244 6

24 (A) 20 16 12

( ) Sulphide, ( ) sulphate and ( )nitrite (mM)

8 4 0 24 (B) 20 16 12 8 4 0 24

(C)

20 16 12 8 4 0 0

50

100

150 200 Time (h)

250

300

350

Conversion to sulphate (%)

Fig. 3 – Profiles of sulphide, sulphate and nitrite concentrations in batch cultures of Thiomicrospira sp. CVO with 20 mM nitrate and various initial sulphide concentrations: (A) 5.5, (B) 13, and (C) 22 mM.

100 75 50 25 0 0

0.5 1 1.5 Ratio of initial sulphide to initial nitrate concentrations

2

Fig. 4 – Effect of initial sulphide to initial nitrate concentrations ratio on the conversion of sulphide to sulphate in the batch cultures of Thiomicrospira sp. CVO; data derived from batch cultures with 10 mM (K) and 20 mM (J) nitrate.

Information regarding the oxidation state of sulphur compounds during biooxidation of sulphide by CVO is rather contradictory. Gevertz et al. (2000) reported that with 10 mM nitrate and 0.5–1 mM, sulphide CVO transformed 80–100% of the sulphide to soluble sulphur compounds. With higher sulphide concentrations (2–3 mM) less than 15% of sulphide was transformed to soluble sulphur compounds. Using 10 mM nitrate and sulphide at concentrations in the range 2–10 mM, Greene et al. (2003) reported that sulphide at concentrations below 4.6 mM was completely oxidized to sulphate by CVO, while at higher sulphide concentrations formation of sulphate was not observed. The results of the present study indicate that the initial ratio of sulphide to nitrate concentrations is a suitable criterion for controlling the oxidation state of produced sulphur compounds by CVO. Partial biooxidation of sulphide to sulphur instead of complete oxidation to sulphate could facilitate the permanent removal of sulphur

ARTICLE IN PRESS 2442

WAT E R R E S E A R C H

40 (2006) 2436– 2446

from the contaminated stream and would positively impact the feasibility of the process through the requirement for lower quantities of nitrate.

3.2.

Kinetic study in the continuous bioreactor

The steady-state profiles of residual sulphide, ATP, sulphate and nitrite concentrations observed at various dilution rates are shown in Fig. 5. Applying dilution rates in the range 0.05–0.1 h
1 led to increases in ATP concentration from the initial value of 0.15 mg/l to a maximum value of 0.31 mg/l. Further increase in dilution rate led to a decrease in ATP concentration. The residual concentration of sulphide was negligible for dilution rates up to 0.1 h
1 but increased continuously to a value of 13.6 mM at the highest applied dilution rate of 0.23 h
1. The concentration of sulphate remained relatively constant at 3.1570.14 mM (close to the sulphate content of modified CSB medium) over the entire range of applied dilution rates, indicating that sulphur was probably the main product. This was consistent with the data compiled from the batch experiments (Fig. 4), since the ratio of sulphide to nitrate loading rates was in the range 1.7–1.9. Oxidation of sulphide led to formation of nitrite, with high nitrite concentrations observed at low dilution rates in the range 0.02–0.1 h
1. The production rate of nitrite at all dilution rates was lower than the rates expected from Reaction 1, indicating that oxidation of sulphide occurred via both

Reactions 1 and 2. The complete conversion of sulphide (99–100%) was observed at sulphide volumetric loading rates as high as 1.6 mM/h (residence time: 10 h). Volumetric removal rates of sulphide reached a maximum as volumetric loading rate was increased with the highest volumetric removal rate of 2.4 mM/h obtained at a loading rate of 3.2 mM/h (Fig. 6). The corresponding sulphide conversion and residence time were 76% and 5.6 h, respectively. At the highest applied volumetric loading rate of 4.6 mM/h, the observed conversion and removal rate were 31% and 1.4 mM/h, respectively.

3.3.

Modelling of the kinetic data

Using the material balances for biomass (ATP) and substrate (sulphide) it can be shown that in a continuous system and under steady state conditions: D ¼ mnet ¼ mg
kd , rS ¼

(6)

1 dX þ mS X, Y X=S dt

(7)

and dX ¼ mnet X. dt

(8)

The experimental data were fit to Eq. (7) and the values of yield and maintenance coefficient were determined to be

20

1.0

0.6 10 0.4

( ) Sulphate (mM)

5

0.2

0

0.0

10

10

8

8

6

6

4

4

2

2

0 0

0.05

0.1

0.15

0.2

( ) ATP (mg L-1)

0.8

15

( ) Nitrite (mM)

( ) Sulphide (mM)

Model prediction

0 0.25

Dilution rate (h-1) Fig. 5 – Steady-state profiles of sulphide, sulphate, nitrite and ATP concentrations as a function of dilution rate in the continuous bioreactor fed with modified CSB medium containing 17.870.8 mM sulphide and 10 mM nitrate.

ARTICLE IN PRESS WAT E R R E S E A R C H

2443

4 0 (200 6) 243 6 – 244 6

2.5

100

2

80 1.5 60 1 40 0.5

20

Model prediction

0

( ) Sulphideremoval rate (mM h-1)

( ) Sulphide conversion (%)

120

0 0

1 2 3 4 Sulphide volumetric loading rate (mM h-1)

5

Fig. 6 – Overall conversion and volumetric removal rate of sulphide as a function of sulphide volumetric loading rate in the continuous bioreactor fed with modified CSB medium containing 17.870.8 mM sulphide and 10 mM nitrate.

Table 1 – Biokinetic parameters determined for various unstructured, non-segregated models Coefficients mm (h
1) KS (mM sulphide) K (mM sulphide)
1 KSX (mM sulphide/mg ATP l) kd (h
1) n Sum of residual squares (based on m) Sum of residual squares (based on S) P value result from F testb

Monod

Tessier

Moser

Contois

0.22 0.63 — — 0 — 0.0025 109a 0.26

0.19

0.36 1.99 — — 0.0014 0.46 0.0012 15 —

0.21 —

1.29 — 0 — 0.0050 240a 0.1

1.81 0 — 0.0031 82a 0.4

a

The value of residual sulphide concentration for dilution rates above the critical value was assumed to be equal to concentration of sulphide in the feed (wash-out conditions). b All expressions were assessed against the Moser expression.

mnet ¼

mm S
kd KS þ S

Monod expression;


 mnet ¼ mm 1
e
KS
kd mnet

m Sn ¼ m n
kd KS þ S

mnet ¼

mm S
kd KSx X þ S

Tessier expression;

Moser expression;

(9) (10) (11)

0.25 Calculated specific growth rate (h-1)

0.018 mg ATP/mmol sulphide and 0.078 mmol sulphide/mg ATP h, respectively. The net specific growth rate can be related to substrate concentration, using a variety of unstructured, non-segregated models. In this work the experimental data were fit to Monod, Tessier, Moser and Contois expressions and the value of various coefficients were determined (nonlinear regression, Table 1):

0.2

0.15

0.1

0.05

0 0 Contois expression:

(12)

The actual values of specific growth rate and the theoretical values calculated from each expression have been compared in a parity chart (Fig. 7), revealing that the Moser expression

0.05 0.1 0.15 0.2 Experimental specific growth rate (h-1)

0.25

Fig. 7 – Parity chart comparing the actual and theoretical values of specific growth rate as calculated by Monod (J), Tessier (n), Moser (E) and Contois (&) expressions.

ARTICLE IN PRESS 2444

WAT E R R E S E A R C H

40 (2006) 2436– 2446

predicted the experimental data with the highest accuracy. To further examine the accuracy of each expression, the sum of residual squares (Eq. (13)), for specific growth rate and residual sulphide concentration were calculated for each expression and are included in Table 1. A comparison of the calculated values again shows that the Moser expression has the best fit. Assessing the Moser expression against the other expressions by an F test also resulted in small values for P, an indication that the differences between the values predicted by Moser and other expressions were statistically significant: SUM ¼

n  X

yi
actual
yi
predicted

2

.

(13)

Furthermore, based on the maximum specific growth rate determined from Monod, Tessier and Contois expressions, the critical dilution rate was in the range 0.19–0.22 h
1, indicating that application of higher dilution rates would result in wash out of the cells. In practice, this was not the case and microbial activity was observed at dilution rates above the critical values predicted by Monod, Tessier and Contois expressions. To show the goodness of fit, the value of various coefficients from the Moser expression were used in Eqs. (7), (8) and (11) and the residual concentration, conversion and volumetric removal rate of sulphide, as well as ATP concentration were calculated as a function of dilution rate or

i¼1

Table 2 – Removal rate of sulphide by photoautotrophs and chemolithotrophs as reported in the literature in recent years Reference

Microbial culture

Oxygen source (electron acceptor)

Reactor configuration

Sulphide removal rate

Main endproduct

Henshaw et al., 1998

Chlorobium limicola

CO2

0.13 mM/h

Sulphur

Prosthecochloris aestuarii

CO2

54.5 mmol g/h cell

Sulphur

Henshaw and Zhu, 2001

Chlorobium limicola

CO2

8.7 mM/h

Sulphur

Syed and Henshaw, 2003

Chlorobium limicola

CO2

44 mM/h

Sulphur and sulphate

Sublette and Sylvester, 1987 Sublette, 1987

Thiobacillus denitrificans Thiobacillus denitrificans Activated sludge biomass

NO
3

Continuous flow photosynthetic reactora Continuous flow photosynthetic reactora Fixed film continuous flow photosynthetic reactorb Fixed film continuous flow photosynthetic reactorb Batch reactora,c

5.4-7.6 mmol g/h cell

Sulphate

Continuous flow reactora,d Fluidized bedb

15.1–20.9 mmol g/h cell

Sulphate

1.8 mM/h

Sulphur and sulphate

Pig manure

O2

1.2 mM/h

Sulphur

Thiobacilli consortium Thiobacillus denitrificans Mixed culturee

O2

Packed-bed biofilterb Recirculation systema Reverse fluidized bed loop reactorb Fluidized bed reactorb Batch reactora,c

4.5 mM/h

Sulphur or sulphate Sulphur and sulphate -

Takashima et al., 2000

Annachhatre and Suktrakoolvait, 2001 Elias et al., 2002 Alcantara et al., 2004 Krishnakumar et al., 2005 Cytryn et al. 2005 McComas and Sublette, 2001

Present workf

a

Enrichment dominated by Thiomicrospira sp. CVO Thiomicrospira sp. CVO

O2 O2

O2 O2 and NO
3 NO
3

NO
3

Continuous flow reactora

34 mM/h 1.6 mM/h 2.9–3.1 mmol g/h cell (1.5-1.6 mM/ h)

Sulphate

10.5 mmol g/h cell (2.4 mM/h)

No sulphate formed

Freely suspended cells. Attached cells. c A gas stream with a composition of 0.5–1% H2S, 5% CO2 and balance of N2 injected into the reactor containing medium and bacteria. d Two gas stream with a composition of 0.9–1.1% H2S, 5% CO2 and balance of N2, and air supplemented with 5% CO2 injected into the reactor containing medium and bacteria. e Thiomicrospira denitrificans, filamentous Thiotrix genus and sulphide-oxidizing symbionts from Gammaproteobacteria. f Specific oxidation rate calculated assuming that 1 mg ATP corresponds to 1 g dry-weight cell (Shuler and Kargi, 2002). b

ARTICLE IN PRESS WAT E R R E S E A R C H

4 0 (200 6) 243 6 – 244 6

volumetric loading rate. The resulting profiles, which are included as solid lines in Figs. 5 and 6, indicate that the theoretical values are generally in agreement with the experimental data. The literature regarding the kinetics of growth and sulphide biooxidation by CVO is very limited and prior to the present work no data for biokinetic parameters have been reported for this bacterium. The maximum specific growth rate of 0.36 h
1determined for CVO is close to those reported for Thiobacilli species (0.1–0.2 h
1), while the saturation constant of 2 mM is significantly higher than the reported values of 0.1–0.3 mM for Thiobacilli (Alcantara et al., 2004). Assuming that 1 mg of ATP corresponds to 1 g of biomass (Shuler and Kargi, 2002), the yield coefficient of 0.018 g biomass/mmol sulphide determined for CVO is relatively close to a value of 0.007 g biomass/mmol sulphide reported by McComas and Sublette (2001) for a microbial enrichment dominated with CVO, and 0.012 g biomass/mmol sulphide determined by Sublette and Sylvester (1987) for T. denitrificans. Table 2 summarizes the biological removal rates of sulphide as reported by various researchers. For ease of comparison, where possible, the original removal rates have been recalculated in terms of a consistent unit of mM/h. As can be seen, both chemoautotrophic and photoautotrophic bacteria have been utilized for the purpose of sulphide biooxidation, either as freely suspended cells or attached biomass. In general the biological sulphide removal rates reported for the chemoautotrophs are comparable or faster than those achieved with photoautotrophs. The simpler nutritional and energy requirements make chemoautotrophs a more attractive biocatalyst for oxidation of sulphide. T. denitrificans is one of the chemoautotrophs which has been frequently used for the purpose of sulphide biooxidation. A comparison of the data presented in Table 2 indicates that the maximum specific sulphide removal rate obtained in the present work by CVO is higher than that reported for T. denitrificans with nitrate as electron acceptor, and lower than that under aerobic conditions. It should be pointed out that the possible chemical oxidation of sulphide in the presence of oxygen or stripping of sulphide by the flowing air could be a contributing factor in higher removal rates reported for the systems studied under aerobic conditions. Considering the ability of CVO in oxidizing sulphide at concentrations as high as 19 mM (based on the results of present work), and the tolerance of a wider range of pH, elevated temperature and salinity makes CVO a favourable biocatalyst for biooxidation and removal of sulphide in a variety of applications.

4.

Conclusions

The results of the present study revealed that CVO is capable of oxidizing sulphide at concentrations as high as 19 mM under anaerobic conditions. Biooxidation of sulphide by CVO results in formation of sulphur or sulphate, and the ratio of sulphide to nitrate initial concentrations (or the ratio of sulphide to nitrate loading rates in a continuous system) is the main factor which can be used to control the oxidation state of the end products. The kinetic data generated in the continuous system was used to determine the biokinetic

2445

parameters for the growth of CVO and biooxidation of sulphide for four growth models, indicating that the Moser expression represents the biokinetic behaviour best. Finally, the maximal volumetric removal rate of 2.4 mM sulphide/h obtained with CVO is higher than that reported for anaerobic oxidation of sulphide by Thiobacillus denitrificans but lower than that under aerobic conditions. Tolerance of high sulphide concentrations, wide range of pH, higher temperatures and salinity are characteristics of CVO which makes it a favourable biocatalyst for biooxidation and removal of sulphide. The enhancement of sulphide removal rate through utilization of CVO immobilized cells is a possibility which is currently under investigation.

Acknowledgements This work was supported by Discovery Grants from the Natural Sciences and Engineering Research Council of Canada (NSERC) to MN and GAH. Provision of a New Opportunity Fund to MN by Canada Foundation for Innovation is greatly appreciated. R E F E R E N C E S

Alcantara, S., Velasco, A., Munoz, A., Cid, J., Revah, S., Razo-Flores, E., 2004. Hydrogen sulfide oxidation by a microbial consortium in a recirculation reactor system: sulfur formation under oxygen limitation and removal of phenols. Environ. Sci. Technol. 38, 918–923. American Public Health Association (APHA), 1992. Standard Methods for the Examination of Water and Waste Water, 18th ed. Washington, DC. Annachhatre, A.P., Suktrakoolvait, S., 2001. Biological sulphide oxidation in a fluidized bed reactor. Environ. Technol. 22 (6), 661–672. Cord-Ruwisch, R., 1985. A quick method for determination of dissolved and precipitated sulfides in cultures of sulfatereducing bacteria. J. Microbiol. Meth. 4, 33–36. Cytryn, E., Minz, D., Gelfand, I., Neori, A., Gieseke, A., De Beer, D., Van Rijn, I., 2005. Sulfide-oxidizing activity and bacterial community structure in a fluidized bed reactor from a zerodischarge mariculture system. Environ. Sci. Technol. 39, 1802–1810. Elias, A., Barona, A., Arreguy, A., Rios, J., Aranguiz, I., Pen˜as, J., 2002. Evaluation of a packing material for biodegradation of H2S and product analysis. Process. Biochem. 37, 813–820. Gevertz, D., Telang, A.J., Voordouw, G., Jenneman, G.E., 2000. Isolation and characterization of strains CVO and FWKO B, two novel nitrate-reducing, sulfide oxidizing bacteria isolated from oil field brine. Appl. Environ. Microbiol. 66, 2491–2501. Greene, E.A., Hubert, C., Nemati, M., Jenneman, G., Voordouw, G., 2003. Nitrite reductase activity of sulfate-reducing bacteria prevents their inhibition by nitrat-reducing, sulfide-oxidizing bacteria. Environ. Microbiol. 5 (7), 607–617. Henshaw, P.F., Zhu, W., 2001. Biological conversion of hydrogen sulphide to elemental sulphur in a fixed-film continuous flow photo-reactor. Water Res. 35 (15), 3605–3610. Henshaw, P.F., Bewtra, J.K., Biswas, N., 1998. Hydrogen sulphide conversion to elemental sulphur in a suspended-growth continuous stirred tank reactor using Chlorobium limicola. Water Res. 32 (6), 1769–1778. Hubert, C., Nemati, M., Jenneman, G.E., Voordouw, G., 2003. Containment of biogenic sulfide production in continuous

ARTICLE IN PRESS 2446

WAT E R R E S E A R C H

40 (2006) 2436– 2446

up-flow packed-bed bioreactors with nitrate or nitrite. Biotechnol. Progr. 19 (2), 338–345. Iliuta, I., Larachi, F., 2003. Concept of bifunctional redox ironchelate process for H2S removal in pulp and paper atmospheric emissions. Chem. Eng. Sci. 58, 5305–5314. Janssen, A.J.H., Sleyster, R., van der Kaa, C., Jochemsen, A., Bontsema, J., Lettinga, G., 1995. Biological sulphide oxidation in a fed-batch reactor. Biotechnol. Bioeng. 47 (3), 327–333. Krishnakumar, B., Majumdar, S., Manilal, V.B., Haridas, A., 2005. Treatment of sulphide containing wastewater with sulphur recovery in a novel reverse fluidized loop reactor (RFLR). Water Res. 39, 639–647. Lee, C., Sublette, K.L., 1993. Microbial treatment of sulfide-laden water. Water Res. 27, 839–846. McComas, C., Sublette, K.L., 2001. Characterization of a novel biocatalyst for sulfide oxidation. Biotechnol. Progr. 17, 439–446. Nemati, M., Jenneman, G.E., Voordouw, G., 2001. Mechanistic study of microbial control of hydrogen sulfide production in oil reservoirs. Biotechnol. Bioeng. 74 (5), 424–434. Pandey, R.A., Malhotra, S., Rajvaidya, A.S., Sharma, S., Peshwe, S., Raman, V.K., Bal, A.S., 2004. Chemo-biochemical desulphurization of various gaseous streams on bench scale. Water Air Soil Pollut., 1–17.

Park, D., Lee, D.S., Joung, J.Y., Park, J.M., 2005. Comparison of different bioreactor systems for indirect H2S removal using iron oxidizing bacteria. Process. Biochem. 40, 1461–1467. Shuler, M.L., Kargi, F., 2002. Bioprocess Engineering, Basic Concepts, second ed. Prentice-Hall Inc., Englewood Cliffs, NJ, pp. 158–160. Sublette, K.L., 1987. Aerobic oxidation of hydrogen sulfide by Thiobacillus denitrificans. Biotechnol. Bioeng. 29, 690–695. Sublette, K.L., Sylvester, N.D., 1987. Oxidation of hydrogen sulfide by Thiobacillus denitrificans: desulfurization of natural gas. Biotechnol. Bioeng. 29, 249–257. Sublette, K.L., Kolhatkar, R., Raterman, K., 1998. Technological aspects of the microbial treatment of sulphide-rich wastewaters: a case study. Biodegradation 9, 259–271. Syed, M.A., Henshaw, P.F., 2003. Effect of tube size on performance of a fixed-film tubular bioreactor for conversion of hydrogen sulfide to elemental sulfur. Water Res. 37, 1932–1938. Takashima, T., Nishiki, T., Konishi, Y., 2000. Anaerobic oxidation of dissolved hydrogen sulfide in continuous culture of the phototrophic bacterium Prosthecochloris aestuarii. J. Biosci. Bioeng. 89 (3), 247–251. Wang, A.J., Du, D.Z., Ren, N.Q., van Groenestijn, J.W., 2005. An innovative process of simultaneous desulfurization and denitrification by Thiobacillus denitrificans. J. Environ. Sci. Health 40 (10), 1939–1949.