Springer 2006

Biotechnology Letters (2006) 28: 175181

DOI 10.1007/s10529-005-5332-7

Enhancement of sulphide production in anaerobic packed bed bench-scale

biolm reactors by sulphate reducing bacteria
Maria Teresa Alvarez1,2, Tania Pozzo1 & Bo Mattiasson1,*
1

Department of Biotechnology, Center for Chemistry and Chemical Engineering, Lund University, PO. Box
124S22100, Lund, Sweden
2
Instituto de Investigaciones Farmaco Bioqumicas, Facultad de Ciencias Farmaceuticas y Bioqumicas,
Universidad Mayor de San Andres, Casilla Postal 3239, La Paz, Bolivia
*Author for correspondence (Fax: +46-46-2224713; E-mail: bo.mattiasson@biotek.lu.se)
Received 20 September 2005; Revisions requested 5 October 2005; Revisions received 10 November 2005; Accepted 11 November 2005

Key words: biolm, Poraver, pumice stones, sulphate-reducing bacteria, sulphide production

Abstract
Two biolm reactors, using pumice stone and Poraver as biolm supports, were run, based on the optimization of sulphide production using a factorial design. The maximum H2S concentrations reached were
10 and 15 mM, respectively, both being appropriate for metal precipitation in euents. The set-up of the
pumice stone biolm reactor is suitable for application in the mining area in the Bolivian Andean region,
where this material is widely available. The use of specic primers for sulphate-reducing bacteria groups
permits the identication of the sulphide-producing bacteria present in biolms.

Introduction
Environmental-friendly processes based on biological sulphide production by sulphate-reducing
bacteria (SRB) have become an alternative for
the removal of heavy metals (Jong & Parry
2003). Sulphate-reducing bioreactors are used in
environmental biotechnology, as a method of
treating mine leachates and industrial wastewaters. The anaerobic bioprocess for sulphate
reduction is based on microbial utilization of
electron donors, such as organic compounds
(carbon sources), and sulphate as terminal electron acceptor for sulphide production. Concomitantly to sulphide production, there is an increase
in pH due to the production of HCO3) by the
microbial oxidation of the electron donor (Webb
et al. 1998).
Various sulphate-reducing bioreactors have
been constructed with the aim of developing
high-rate immobilized biomass reactors, such as
the upow anaerobic xed lm reactor (UAFFR)

(El Bayoumy et al. 1999), packed lter reactors

and uidized-bed reactors (Kaksonen et al.
2003). Dierent forms of inorganic supports have
been used as carriers for the SRB biomass, e.g.,
ceramic carriers in UAFFR (Glombitza 2001)
and silica sand, porous glass particles and plastic
pall rings in anaerobic packed bed reactors
(Kolmert et al. 1997, Steed et al. 2000, Jong &
Parry 2003).
The present work was carried out to optimize sulphide production. Two important issues
were addressed: dependency on the concentrations of electron donor and acceptor and/or increase in the available biolm SRB biomass
growth surface. The optimization of sulphide
production was based on a factorial design
with three different concentrations of lactate
and sulphate. Subsequently, the inuence of
biomass carriers such as pumice stones and
Poraver was studied through the establishment
of packed bed bioreactors for high sulphide
production.

0.1
0.5
0.3
0.1
0.1
0.1
0.2
0.2
0.3

3.9
9.5
7.2
2.8
4.2
4.7
5.8
2.8
4.3
2
1
1
1
3
2
1
4
3

97
65
32
97
64
31
99
63
33
12 1
14 3
10 3
92
12 1
10 3
10 0
80
81
100
67
33
100
67
33
100
67
33
46
46
46
32
32
32
18
18
18
1
2
3
4
5
6
7
8
9

a
Higher and lower concentrations of sulphate and lactate than the ones detailed in modied Postgate C medium (sulphate = 32 mM and lactate = 67 mM) were studied. The
selected concentrations of sodium sulphate (46 mM) and lactate (67 mM) resulted in the appropriate curved surface in the response surface graph. Further increases in their
concentrations showed a negative eect on hydrogen sulphide production (Data not shown).
b
The oxidation of dierent concentrations of lactate is proportionally related to the production of acetate and propionate, as the main products of its degradation amongst the
other VFAs produced.

0
0
0
0.2
0.1
0
0.02
0.05
0.05

0.2
0.2
0.1
1.0
0.7
0.3
1.0
1.0
0.2
42 5
22 3
8.0 1
47 1
28 3
10 1
39 1
33 1
12 1
1
2
2
0
1
1
0
0
1

Propionate
(mM)
Acetate
(mM)

VFAs productionb
Produced Sulphide
(mM)
Consumed Lactate
(mM)
Removed SO42)
(mM)

Initial
Lactate (mM)

Two biolm reactors (Figure 1) were used to enrich the sample consortium for sulphide production. Experiments were carried out in lab-scale
anaerobic packed-bed reactors (180 ml working
volume). Poraver [1020 mm diam. porous glass
beads (Poraver GmbH, Germany)] and pumice
stone (cut in squares 10
15
15 mm) were
used as biomass carriers. Biolms were established on both carriers over a two-month period
by recirculating modied Postgate C medium.

Initial SO42
(mM)

Anaerobic packed bed reactors

Variablesa

The experiments (see Table 1) were performed in

triplicate in 100 ml serum asks (vials) lled with
80 ml of modied Postgate C medium, inoculated with the enriched sample and incubated at
20 C. The maximum sulphide concentrations
were analysed using response surface methodology. This allowed ecient tting and checking of
the second-degree polynomial model which was
used to estimate the predicted response. STATISTICA v.6 (StatSoft Inc., USA) was applied for
the analysis of statistical data. All results were
analysed with factorial ANOVA (p 0.05).

Run

Optimization of sulphide production

Table 1. Eects of sulphate and lactate concentrations on sulphide and volatile fatty acids production.

A sample of sludge and degraded organic matter

was collected from an efuent from a local brewery. Microorganisms were enriched by adding
0.5 g sample to 80 ml sterile modied Postgate
medium in 100 ml sterile bottles ushed with
O2-free N2 according to the Hungate technique
(Miller & Wolin 1974).
Postgate C medium (Widdel & Bak 1991) was
used with some modications. One litre of medium contained: 0.5 g KH2PO4, 1 g NH4Cl, 60 mg
MgSO47H2O, 60 mg CaCl26H2O, 0.3 g trisodium citrate, 4 mg FeSO47H2O and, unless
otherwise stated, 4.5 g Na2SO4 and 6 g lactate
(80%) were used. The pH was adjusted to 7.5
with KOH. The preparation of the media in
batch experiments (factorial design) and in continuous-culture experiments was done under
anaerobic conditions.

i-butyrate
(mM)

Microorganisms and culture conditions

34
26
19
32
27
20
36
24
17

n-butyrate
(mM)

Materials and methods

0.4 0
0.4 0

0.1 0

176

177
Feed
Sampling
port

Sterile
filter

purication of genomic DNA were performed

according to Pospiech & Neumann (1995).
Nested PCR

Recirculation
loop

Medium

Fig. 1. Schematic gure of the set-up of the reactors. The

glass columns were tted with sample ports and the system
set-up allowed recirculation of the medium. They were lled
with 20 ml of the enriched sample mixed with 160 ml of modied Postgate C medium. Thereafter, they were left undisturbed as batch cultures until sulphide production was
observed as formation of FeS precipitates. The used liquid
medium from the continuous reactors was collected by overow in a container connected to a sterile lter (0.2 lm pore
size) allowing air to exit when the waste entered.

The 16S rDNA was amplied and puried as reported previously (Quillaguaman et al. 2004).
Direct PCR amplication with the six group-specic PCR primers (Daly et al. 2000) was
attempted on these eubacterial PCR amplication
products (DNA nested amplication). Reactions
were carried out as follows: 94 C for 4 min, then
30 cycles of 94 C for 20 s, 55 C for 20 s, 72 C
for 90 s, and a nal extension for 72 C for
7 min. Each reaction tube (20 ll) contained:
16.5 ll membrane-ltered water 2.0 ll Buer
STR 10
(Promega), 0.15 ll of each primer
(50 pmol/ll), 0.05 ll BSA (10 mg/ml BioLab),
0.15 ll Taq polymerase (50 U/ll Fermentas)
and DNA template (approx. 100150 ng).

Results
The feed ow was established at 1.8 ml/h, corresponding to a retention time of 100 h.
Samples and analysis
H2S determination
Dissolved sulphide was measured immediately
after sampling to prevent its oxidation, using the
turbidimetric method reported by Cord-Ruwisch
(1995), based on the precipitation of copper
sulphide.
Determination of sulphate reduction
Sulphate in the medium was measured by the
turbidimetric method described by Kolmert et al.
(2000), based on the precipitation of barium
sulphate.
Volatile fatty acids analysis
Volatile fatty acids (VFAs) were analyzed by
HPLC as described by Parawira et al. (2004).
Genomic DNA extraction and purication
Genomic DNA was extracted from leachate samples of the anaerobic bioreactors. A pre-treatment of the sample was performed according to
Daly et al. (2000). Subsequently, extraction and

Optimization of sulphide production

Variations in sulphate and lactate concentrations
had a strong effect on volumetric production of
sulphide. Table 1 describes the 32 factorial design
used for optimization of hydrogen sulphide and
VFAs production due to the reduction of sulphate and the utilization of lactate. The investigated minimum and maximum levels of variables
were 1846 mM for sulphate and 33100 mM for
lactate, respectively. The analysis of the results
obtained in this experimental design by means of
ANOVA showed that the eects of sulphate and
lactate are statistically signicant in respect to
hydrogen sulphide production. The results obtained t the following equation, where X is sulphate concentration (mM) and Y is lactate
concentration (mM):
Sulphide [mm] 2:77  0:28X 0:16Y
0:0083X2  0:0025XY  0:0007Y2

A three-dimensional response surface was plotted

based on the maximal results of the factorial
design (Figure 2). The highest concentration of
sulphide reached in these batch cultures

178
Anaerobic packed bed reactors
According to the results obtained in the factorial
design, three conditions were established in the
bioreactors (Table 2). Condition C was demonstrated to be most suitable for sulphate removal
and sulphide production when using Poraver as
biomass carrier. Under these conditions, the daily sulphate removal eciency achieved was 63%
(30 mM) and sulphide production was 15 mM. In
the case of pumice stone, the results obtained
were similar for conditions A and C with a sulphide production of 9 mM and the eciency of
sulphate removal was 69% and 63% respectively.
The production of sulphide was stable for both
biolm reactors during approximately 70 days of
continuous culture. The metabolism of lactate
during continuous culture operation was also
incomplete, as was observed in batch cultures.

Fig. 2. Response surface of hydrogen sulphide production by

means of dierent lactate and sulphate concentrations.

Nested PCR
Nested PCR amplications carried out with the
total 16S rDNA using SRB-group specic primers according to Daly et al. (2000), determined
the presence of Desulfovibrio-, Desulfotomaculumand Desulfobulbus-like bacteria for both biolm
reactors.

(ca. 10 mM) was achieved at 46 mM sulphate and

67 mM lactate (Table 1 and Figure 2). Under all
the conditions studied, lactate was partially
metabolized and, the metabolic products were
mainly acetic, propionic and butyric acids, besides carbon dioxide and water.

Table 2. Sulphate removal, lactate consumption, hydrogen sulphide and VFAs production in biolm reactors using pumice stones
and Poraver as biomass supports.
Concentration(mM)

Biomass support
Pumice stone

Lactatea
Sulphatea
Lactateb
Sulphateb
Sulphide
Acetatec
Propionatec
i-butyratec
n-butyratec
a

Poraver

Condition A

Condition B

Condition C

Condition D

67
32
64
22
8.8
34
12
0.04
1.6

33
46
31
23
6.8
26
2.1
0.05

67
46
66
29
9.1
27
10
0.01
0.9

67
32
63
23
9.8
32
13
0.07
1.2

2
0
0.3
2
1
0
0.2

1
2
0.2
0
0.2
0

1
1
0.2
1
2
0
0.1

3
2
0.3
3
2
0
0.1

Condition B
33
46
32
24
10
11
0.2
0.15

1
1
0.2
1
0.1
0

Condition C
67
46
66
30
15
19
6.0
0.04
0.2

1
2
0.1
2
0.1
0
0.1

Concentrations in the feed.

Consumed concentrations.
c
Volatile fatty acids production.
Results are presented as an average of those obtained in each condition.
Condition A consisted of the modied Postgate C medium, when the concentrations of sodium sulphate and lactate were 32 and
67 mM, respectively. Condition B consisted of 46 and 33 mM and nally, condition C was based on 46 and 67 mM of sodium sulphate
and lactate, respectively.
b

179
Discussion
The optimization of sulphide production was
studied with the aim of developing a high producing and robust process for later application in
heavy metal precipitation. This process should be
suitable for application in areas where continuous monitoring is difcult, such as the Bolivian
Andean region. A high production of sulphide
could sustain precipitation of heavy metals even
after dilution several times. Such a process would
be suitable for treatment of waters with high metal content. Based on the response surface analysis
(Figure 2), it is clear that higher concentrations of
sulphide can be obtained at other concentrations
of lactate and sulphate than those described in the
modied Postgate C medium. Thus, optimized
conditions were established in the biolm reactors.
Biolm reactors have proven successful in
environmental biotechnology. Furthermore, the
SRB are grown on a porous support and the
high cell density contributes to a robust system
with high and stable productivity. This is
achieved when the biolm thickness is such that
the conversion becomes diffusion controlled. To
set up such a reactor, smooth starting conditions
(low dilution rates) should be chosen in order to
avoid a wash-out of SRB. Therefore, the application of a reactor with immobilized microorganisms appears to be the set-up of choice for
continuous operation (Kolmert et al. 1997,
Glombitza 2001). Research and strategies have
been carried out to improve biomass retention by
combination of dierent reactor design and/or
carrier materials. For example, lactate was used
as electron donor in a uidized bed reactor using
a silicate mineral as biolm support (Filtralite,
Norway), the maximum sulphate reduction rate
was up to 2 g/ld (21 mmol/ld) achieving 20
85% of sulphate removal during 120 days of
continuous working conditions and obtaining a
sulphide
production
rate
of
300 mg/ld
(8.82 mmol/ld) at a HRT of 16 h (Kaksonen
et al. 2003).
On the other hand, when biolm reactors
packed with pool lter sand (Commercial Minerals Limited, Melbourne, Australia), were applied
for sulphate removal using sodium lactate as electron donor, the efuent showed a minimum of
357 mg sulphate/l (3.72 mM) by day 7, achieving
83% of sulphate removal at a HRT of 16.2 h.

The sulphide concentration in the euent ranged

between 15 and 134 mg/l (0.44 and 3.94 mM)
throughout the 14-day period (Jong & Parry
2003). According to Kolmert et al. (1997), the
packed bed reactor proved to be the most ecient, where the biolm established on Poraver
was very robust in producing sulphide. This procedure was considered to be the most suitable for
scaling up. The conversion rate of sulphate to sulphide in this packed bed bioreactor was reached
at a HRT of 4.35 h which corresponds to a load
of 5.3 g SO4/ld (55 mmol/ld). The highest obtained sulphide concentration was 14 mM, but the
biolm became inhibited, so adjustments in working conditions allowed the system to regularly
produce approximately 10 mM of sulphide. In the
present study of the optimum conditions found in
the factorial design, the highest production of sulphide was also achieved with Poraver as carrier
material (15 mM), conrming the potential of this
material for biomass growth. Due to the costs of
this system and taking into account that this process is intended for implementation in Bolivia, it
is important to consider the pumice packed bed
reactor as a feasible option. The H2S production
achieved in this biolm reactor, 9 mM, was stable
and is comparable with the systems described
above (Figure 3 and Table 2). In addition, pumice stone is widely available in the Andean region.
The most important electron donors for
microbial sulphate reduction are H2, acetate, lactate and propionate (Hao et al. 1996). Lactate
has been described as a suitable carbon source,
but it was reported to be degraded only partially
by SRB resulting in release of acetate (Montogomery et al. 1990) which may therefore cause
high chemical oxygen demand (COD) in the
euents (Glombitza 2001). According to the
reaction of lactate to acetate conversion in terms
of sulphate reduction, the following equation
should be considered:

2CH3 CHOHCOO SO2
4 ! 2CH3 COO
2HCO
3 H2 S

When two moles of lactate are oxidised, two moles of acetate and two moles of bicarbonate are
formed, while one mol of sulphate is reduced to
one mol of sulphide. However this was not ob-

180
35

14

30

12

25

10

20

15

10

0
0

30

60

[Sulphide] mM

[Removed sulphate] mM

(a)

0
90 120 150 180 210 240 270 300

35

20
A

30

18
16

25

14
12

20

10

15

8
6

10

[Sulphide] mM

40

[Removed sulphate] mM

Time (days)

(b)

4
2

30

60 90 120 150 180 210 240 270 300

Time (days)
Fig. 3. Sulphide production and sulphate reduction in packed
bed bioreactor. (a) Pumice stones and (b) Poraver were used
as biomass carriers. s = Removed sulphate. n = Sulphide.
The packed bed bench-scale glass bioreactors needed approximately 25 days to reach good working conditions. Once
established, the system was very robust. According to the results obtained in the factorial design, three conditions were
established on the bioreactors. Condition A, consisted of the
modied Postgate C medium, when the concentrations of sodium sulphate and lactate were 32 and 67 mM, respectively.
Condition B, consisted of 46 and 33 mM and nally, condition C, based on 46 and 67 mM of sodium sulphate and lactate, correspondingly. Samples for analysis were taken in
duplicate.

Since it is a system that starts from a mixed culture, it is not clear that the microbial consortia
are the same in the two reactors, although the
presence of Desulfobrivibrio-, Desulfotomaculumand Desulfobulbulos-like cells was determined in
both bioreactors as contributing sulphidogenic
bacteria. There may be an advantage to having
several sulphide producing bacteria present in the
bioreactors, since they may have slightly dierent
living conditions, and this may help to stabilize
the performance of the reactors. Still, one should
be careful to draw conclusions from the observations. However, both reactors gave constant concentrations of sulphide during prolonged
operation. This means that the biolm concept
helps in forming a reactor system with good
operational stability.

Acknowledgements
The authors want to thank specially the support
of Sida/SAREC in the development of the project Microbial Diversity of Poopo Lake and
Desaguadero river basin in La Paz, Bolivia and
the nancial support from Swedish Institute to
support the studies of Tania Pozzo. Britte Marie
Pott is specially thanked for giving advice and
help with the bioreactors. We are also grateful to
Dr. Roberto Romero from Analytical Chemistry,
Center for Chemistry and Chemical Engineering
Lund University (Sweden) for his critical assistance in the statistic analysis.

References
served experimentally in some of the conditions
studied. The results obtained here for both reactor systems showed the incomplete oxidation of
lactate in terms of sulphate reduction, having
acetate and propionate as main metabolic products. The production of these acids increased
while the conversion of lactate proceeded. It is
also important to note that acetate production
under conditions B and C established on the
Poraver-packed bed bioreactor is approximately
half of what is produced under the same conditions in the pumice stone-packed bed bioreactor.

Cord-Ruwisch R (1985) A quick method for the determination

of dissolved and precipitated suldes in cultures of sulfatereducing bacteria. J. Microbiol. Meth. 4: 3336.
Daly K, Sharp RJ, McCarthy AJ (2000) Development of
oligonucleotide probes and PCR primers for detecting
phylogenetic subgroups of sulfate-reducing bacteria. Microbiology 146: 16931705.
El Bayoumy MA, Bewtra JK, Ali HI, Biswas N (1999) Sulde
production by sulfate reducing bacteria with lactate as feed
in an upow anaerobic xed lm reactor. Water Air Soil.
Poll. 112: 6784.
Glombitza F (2001) Treatment of acid lignite mine ooding
water by means of microbial sulfate reduction. Waste
Manage. 21: 197203.

181
Hao OJ, Chen JM, Huang L, Buglass RL (1996) Sulfatereducing bacteria. Crit. Rev. Env. Sci. Tec. 26: 155187.
Jong T, Parry DL (2003) Removal of sulfate and heavy metals by
sulfate reducing bacteria in short-term bench scale upow
anaerobic packed bed reactor runs. Water Res. 37: 33793389.
Kaksonen AH, Riekkola-Vanhanenc ML, Puhakkaa JA (2003)
Optimization of metal sulphide precipitation in uidized-bed
treatment of acidic wastewater. Water Res. 37: 255266.
Kolmert A, Henrysson T, Hallberg R, Mattiasson B (1997)
Optimization of sulphide production in an anaerobic
continuous biolm process with sulphate reducing bacteria.
Biotechnol. Lett. 19: 971975.
Kolmert A, Wikstrom P, Hallberg K (2000) A fast and simple
turbidimetric method for the determination of sulfate in
sulfate-reducing bacterial cultures. J. Microbiol. Meth. 41:
179184.
Miller TL, Wolin MJ (1974) A serum bottle modication of the
Hungate technique for cultivating obligate anaerobes. Appl.
Microbiol. 27: 985987.
Parawira W, Murto M, Read JS, Mattiasson B (2004) Volatile
fatty acids production during anaerobic mesophilic digestion

of solid potato waste. J. Chem. Technol. Biotechnol. 79:

673677.
Pospiech A, Neumann B (1995) A versatile quick prep of
genomic DNA from Gram-positive bacteria. Trends Genet.
11: 217218.
Quillaguaman J, Hatti-Kaul R, Mattiasson B, Alvarez MT,
Delgado O (2004) Halomonas boliviensis sp. nov., an alkali
tolerant, moderate halophile bacterium isolated from soil
around a Bolivian hypersaline lake. Int. J. Syst. Evol.
Microbiol. 54: 721725.
Steed VS, Suidan MT, Gupta M, Miyahara T, Acheson CM,
Sayles GD (2000) Development of sulfate-reducing biological process to remove heavy metals from acid mine
drainage. Water Environ. Res. 72: 530535.
Webb JS, McGinness S, Scott HML (1998) Metal removal by
sulphate-reducing bacteria from natural and constructed
wetlands. J. Appl. Microbiol. 84: 240248.
Widdel F, Bak F (1991) Gram-negative Mesophilic Sulfatereducing bacteria. In: Balows A, Truper HG, Dworkin M,
Harder W & Schleifer KH, eds. The Prokaryotes, New York:
Springer-Verlag, pp. 33523378.