Process Biochemistry 41 (2006) 2118–2123

www.elsevier.com/locate/procbio

Effect of hydraulic retention time on biohydrogen production

and anaerobic microbial community
Zhen-Peng Zhang a,b, Kuan-Yeow Show a,*, Joo-Hwa Tay a,b,
David Tee Liang b, Duu-Jong Lee c, Wen-Ju Jiang d
a
School of Civil and Environmental Engineering, Nanyang Technological University, N1#1a-27, Nanyang Avenue, 639798 Singapore, Singapore
b
Institute of Environmental Science and Engineering, Nanyang Technological University, 637723 Singapore, Singapore
c
Department of Chemical Engineering, National Taiwan University, Taipei 10617, PR China
d
Department of Environmental Science and Engineering, Sichuan University, Chengdu 610065, PR China
Received 20 February 2006; received in revised form 15 May 2006; accepted 16 May 2006

This article is dedicated to Wolf-Dieter Deckwer on the occasion of his 65th birthday.

Abstract
The effects of hydraulic retention time (HRT) on biohydrogen production and its mixed anaerobic microbial community grown with glucose
were investigated in a continuous stirred tank reactor culture. Starting from a HRT of 50 h the culture was acclimated by stepwise HRT reduction
until a steady-state was reached at a HRT of 8 h after 19 days. After that the culture was run with HRTs of 6 h, 8 h, 10 h and 12 h, each lasting 9–15
days. Hydrogen, CO2 and dissolved products were daily measured. The species composition of the culture was ascertained using 16S rDNA genes
separated by denaturing gradient gel electrophoresis (DGGE). While a hydrogen yield of 1.6 mol/mol glucose was found during the first steady-
state at 8 h HRT it stabilized at 1.9 in the following steady-states. The predominant dissolved products were butyrate and acetate in a ratio of about
2.1:1 and amounted to 82–94% of the total products. In the first 8 h HRT period also propionate was found in an amount of 9%. It is concluded that
the increase of the hydrogen yield after transition from 8 h to 6 h HRT is caused by a washout of the propionate producing population during the 6 h
HRT period. In fact the fading of two 16S rDNA gene fragments was noticed in the DGGE profiles at the same time.
# 2006 Elsevier Ltd. All rights reserved.

Keywords: Anaerobic biohydrogen production; Hydraulic retention time; Mixed culture; Microbial community; Continuous stirred tank reactor; Denaturing
gradient gel electrophoresis

1. Introduction from nonrenewable sources such as natural gas thermal cracking

and coal gasification [2]. Anaerobic fermentation route is a
Energy is vital to global prosperity, yet dependence on fossil promising biological process for hydrogen production owing to
fuels as our primary energy source has been reported to be a the fact that hydrogen can be produced continuously at high rate
major cause to global climate change, environmental degradation from renewable organic compounds [3].
and health problems [1]. To exploit alternative energy sources Studies on hydrogen production by anaerobic fermentation
has became a pressing agenda due to recent global energy have been carried out using pure cultures of bacteria such as
concern as a result of skyrocketing demand but relatively limited Enterobacter [4], Rhodopseudomonas [5], Bacillus [6],
resources of fossil fuels. Hydrogen has been recognized as a Citrobacter [5], Escherichia [7] and Clostridium [8] or
perfect, clean fuel because of its inherent oxidation chemistry combined bacteria, i.e., Clostridium and Enterobacter [9] to
with water as the only reaction product, without any greenhouse degrade monosaccharides and disaccharides, cellulose and
gases generated. Nearly 90% of hydrogen production at present, starch in the laboratory-scale studies. The highest hydrogen
however, is generated in centralized thermochemical processes conversion efficiency reported so far is 2.6 mol H2/mol
glucose obtained by Taguchi et al. using Clostridium [10]. On
the other hand, it may be difficult to use a pure culture for
* Corresponding author. Tel.: +65 67905282; fax: +65 67910676. hydrogen production from organic wastes since pure culture
E-mail address: [email protected] (K.-Y. Show). is easily out-competed by various non-hydrogen producers.
1359-5113/$ – see front matter # 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.procbio.2006.05.021
 Z.-P. Zhang et al. / Process Biochemistry 41 (2006) 2118–2123 2119

However, utilizing mixed cultures for hydrogen production

did not attract much attention, until in recent years a
comparable hydrogen yield (
2.45 mol H2/mol glucose) by
mixed cultures rich in Clostridia originated from anaerobic
digesters, composts, soil and other natural sources has been
achieved [11–13].
Continuous stirred tank reactor (CSTR) processes have been
widely used for fundamental studies of continuous anaerobic
hydrogen production due to its well-mixing characteristics to
allow hydrogen-producing bacteria to overcome mass transfer
resistance [14,15]. Nevertheless, many factors are contributing
to the continuous production of hydrogen by the mixed cultures
in a CSTR, including system pH [16], hydraulic retention time
(HRT) [17], temperature [6], carbon source [18] and its strength
[19], metabolic pathways involved [20] and microbial diversity
[21]. HRT is one of the important control parameters affecting
continuous production of hydrogen, since HRT control can
avoid the hydrogen utilization by hydrogen-consumers, such as
the methanogens [22]. However, the reported optimum HRTs
for hydrogen production are rather inconsistent and varies from
8 h for sucrose [23] and 12 h for glucose [24] to 18–24 h for
brewery wastewater [15,17].
The objective of the present study was to investigate the
continuous production of hydrogen and its associated microbial
community of the mixed culture at varying HRTs in a CSTR.

2. Materials and methods

2.1. Seed sludge

Anaerobically digested sludge obtained from a wastewater treatment plant

in Singapore was used as inoculum. The collected sludge was sieved using a
425 mm mesh, and the filtrate was stored in a 4 8C fridge. The seed sludge was
heat treated at 98 8C for 2 h and acid treatment at a pH of 2 for 24 h before
inoculation. Initial concentration of volatile suspended solids (VSS) in the
reactors was determined to be about 3500 mg/L.

2.2. Feed composition

The synthetic wastewater mainly consisting of glucose (10 g/L) was used as
feed, which also contained the following nutrients (mg/L): 200 NH4Cl, 200
peptone, 200 KH2PO4, 30 FeCl36H2O, 20 MgCl26H2O, 10 MnCl24H2O, 5
CoCl26H2O, 6 CaCl22H2O, 5 Cu(NO3)23H2O, 5 ZnSO47H2O and 5
NiSO46H2O.

2.3. Experimental setup and operation

A glass-made CSTR (Biostat B, B. Braun Biotech, Germany) was used in

the present study. It had a working volume of 6 L with an internal diameter of
160 mm and a height of 345 mm. The CSTR was covered with aluminum foil to
avoid sun light and stirred at a constant rate of 280 rpm. The pH of mixed liquid
in the reactor was controlled at 5.5  0.05 with 4 M NaOH and 4 M HCl by
automatic titration. The reactor was operated at a constant temperature of 37 8C
through a water jacket. The feed stored in a 4 8C fridge was pumped into the
reactor by a peristaltic pump (Masterflex L/S, Cole-Parmer, USA). A level
sensor and effluent peristaltic pump were used to control the slurry volume at a
constant value. A 1.16 L gas–liquid separating bottle was connected to the
effluent to measure the biogas carried by the effluent, thus daily amount of
biogas production was equal to the biogas in the effluent plus the biogas released Fig. 1. Variations in: (a) HRT, (b) glucose conversion rate, (c) biogas produc-
directly from headspace, which was measured by a wet gas meter (Ritter TG 05, tion rate, (d) biogas contents and (e) soluble metabolites over time (AA, acetic
Germany). acid; PA, propionic acid; BA, butyric acid; EtOH, ethanol).
2120 Z.-P. Zhang et al. / Process Biochemistry 41 (2006) 2118–2123

After inoculation, 10 g glucose was added to the reactor. The reactor was 5.5 8C and 37 8C, respectively. The reactor mixing condition was
flushed with industrial argon gas for 10 min to create an anaerobic environment
ensured by keeping the variation of VSS concentrations to less
before the start-up of reactor. The reactor was initially operated for 1 day in batch
mode without additional feed supply in order to avoid washout of inactivated than 5% at three different reactor heights, i.e., at 50 mm, 170 mm
biomass, and then converted to continuous operation. The HRTs were controlled and 280 mm from the reactor bottom (data not shown).
in a shortening mode from 50 h to 8 h to start up the reactor. After reaching steady- The initial HRT was set at 50 h for reactor start-up. To
state stage, the reactor was operated at HRTs of 8 h, 6 h, 8 h, 10 h and 12 h. The accelerate the acclimation process of the mixed culture, HRT
reactor was operated for at least one week at each HRT for data collection. Steady- was shortened gradually once the glucose conversion rate
state conditions defined in the present study were considered to be reached when
the variations of products along with biomass concentration were less than 10%. reached 95% (Fig. 1a and b). A dynamic increase in biogas
production was observed during the start-up stage (Fig. 1c).
2.4. Analytical methods This is probably attributed to the fact that the microorganisms
are adapting to the new environment resulted from the HRT
Biogas contents: The biogas composition was analyzed by a Micro-Gas shifting. Steady gas production was not observed until day 19,
Chromatograph (Varian 4900, USA) equipped with a thermal conductivity at a HRT of 8 h. After steady-state conditions were reached, the
detector. Hydrogen was analyzed using a Molsieve 5A Plot column with argon
HRT was shortened to 6 h, and then increased in an increment
as carrier gas at 60 8C; methane and carbon dioxide were analyzed using a
Propac Q column with helium as carrier gas at 60 8C. of 2–12 h. Five steady-state conditions, marked as Runs 1–5,
Aqueous samples: Aqueous samples were filtered through a 0.45 mm were at least maintained for one week at each HRT before
membrane before analysis. Volatile fatty acids (VFAs) and alcohol were changing the HRT (Fig. 1 and Table 1). The biogas produced
determined by a gas chromatograph (Agilent 6890, USA) equipped with a consisted of hydrogen and carbon dioxide and was free of
mass selective detector (220 8C), injector (200 8C) and a 30 m  0.25 mm HP-
methane throughout the study. The result thus lends support for
FFAP fused-silica capillary column. The temperature of the oven was initially
maintained at 60 8C for 4 min, increased to 168 8C at a ramp of 6 8C/min and the effectiveness of pretreatment protocol on inactivating
lastly heated to 200 8C at 10 8C/min and maintained for 2 min. Helium was used methanogens. Hydrogen content increased from initially 49%
as the carrier gas. Glucose concentration was determined by the phenol–sulfuric to 67% at day 5 and thereafter leveled off (Fig. 1d). Hydrogen
acid method [25]. Measurements of VSS were performed according to the gas accounted for about 61% of total biogas evolved in Run 1
Standard Methods [26].
and increased to 64% in the following runs (Table 2). This value
Microbial species composition: The microbial populations in hydrogen-
producing community at each HRT condition were examined and compared by is much higher than that of other studies (34–53%) [6,24,30]
analyzing the denaturing gradient gel electrophoresis (DGGE) profiles of 16S with the exception of Ueno et al. [31] who obtained the same
rDNA fragments. DNA in the sludge samples was extracted, and then the 16S level.
rDNA fragments were amplified by polymerase chain reaction (PCR), and As summarized in Table 1, a hydrogen yield of 1.6 mol H2/
separated by DGGE. For DNA extraction, a 100–200 mg sludge sample (wet
mol glucose was obtained in Run 1, at which the reactor was
weight) was mixed with 50 mL 20% sodium dodecyl sulfate, 600 mL glass bead
with a diameter of 0.1 mm and 500 mL saturated phenol in a 2 mL tube and operated at HRT 8 h. Hydrogen yield increased by about
treated in a bead beater at 5000 oscillations/s for 3 min. Subsequently, DNA 0.3 mol H2/mol glucose when the HRT was shortened to 6 h in
extraction was carried out according to the protocols described previously Run 2. However, a relatively consistent hydrogen yield of
[27,28]. The primers 357FGC (50 -CGC CCG CCG CGC GCG GCG GGC GGG approximately 1.9 mol H2/mol glucose was obtained as the
GCG GGG GCA CGG GGG GCC TAC GGG AGG CAG CAG-30 ) and 517R
HRTs were increased to 8 h, 10 h and 12 h in Runs 3, 4 and 5,
(50 -ATT ACC GCG GCT GCT GG-30 ) were used for PCR amplification. A
touchdown thermal profile technique [27] using 2 min for activation and 10 min respectively. The hydrogen production rate reached the
for extension was performed. The DGGE was performed following the method maximum of 7.8 L/L day at a HRT of 6 h, and it decreased,
described by Muyzer et al. [29]. Electrophoresis was conducted in a 1 TAE but not in a linear trend, with increasing HRT, whereby glucose
buffer solution at 85 V and 60 8C for 12 h. conversion rate increased from 77.8% at 6 h HRT to 99.4% at
12 h HRT. It is interesting to note that an increase of 0.8 L/
3. Results and discussion L day in hydrogen production rate was obtained in Run 3 as
compared to Run 1, both at 8 h HRT, as a consequence of the
3.1. Hydrogen production increased hydrogen yield.
The biomass concentration retained in the reactor was in a
The effect of HRTon hydrogen production was investigated in range of 843–1013 mg/L, but not in correlation with the HRT
a completely mixed reactor at a consistent pH and temperature of (Table 1). As only 77.8% of glucose was utilized at a HRT of 6 h

Table 1
Reactor performance at steady-state stage
Run no. Operating time HRT (h) VSS (mg/L) Glucose conversion H2 yield H2 production
(day) rate (%)a (mol H2/mol glucose) rate (L/L day)
1 20–25 8 910.83  39.06b 91.63  0.20 1.62  0.03 6.08  0.19
2 26–38 6 847.50  26.86 77.82  0.38 1.88  0.03 7.77  0.13
3 39–49 8 897.64  25.17 90.46  0.30 1.93  0.02 6.92  0.03
4 50–58 10 843.33  65.60 96.24  0.52 1.95  0.03 6.11  0.07
5 59–68 12 1013.33  95.57 99.44  0.07 1.91  0.02 4.95  0.03
a
Ten grams per litre in the influent.
b
Mean value  standard deviation.
 Z.-P. Zhang et al. / Process Biochemistry 41 (2006) 2118–2123 2121

39.10  1.22
35.22  0.75
36.97  0.95
35.25  1.54
36.34  0.77
it is not meaningful to shorten HRT to less than 6 h to further

CO2 (%)
enhance the hydrogen production rate.

Gaseous metabolites
3.2. Soluble metabolites

60.90  1.22
64.78  0.75
63.03  0.95
64.75  1.54
63.66  0.77
H2 (%)
Fig. 1e reveals the production of main VFAs and alcohol
associated with the HRT. Butyric, acetic and propionic acids and
ethanol were the main metabolites at the start-up, but butyric and
acetic acids increased significantly and became the dominant
metabolites after reaching steady-state stage. Butyric and acetic
2.13  0.15
2.10  0.51
2.44  0.28
2.16  0.60
2.22  0.35 acids accounted for approximately 82–94% (molar percentage)
BA/AA

of total metabolites, while relatively consistent percentage of

ethanol (4.6–7.3%) was obtained at the steady-state stage,
regardless of the variation of HRT. Notably, considerable
AA + BA (%)

82.30  3.63
91.87  1.53
92.85  1.06
91.83  1.40
94.07  0.43

production of propionic acid was observed during the culture

acclimation and first 8 h HRT (Run 1) periods, but it decreased to
0–0.18 mmol/L immediately after the HRT was transited from
8 h to 6 h (Fig. 1e). It appears that these propionate producers are
highly sensitive to a low HRT and 6 h whereby might be a critical
7.30  3.30
6.82  1.19
6.10  1.01
6.91  1.16
4.64  0.36
EtOH (%)

HRT to remove propionate producers from the system, which is

a

supported by the durable absence of propionic acid during the

HRT of 8–12 h in Runs 3–5. Consequently, a higher hydrogen
yield would be expected as aforementioned, because propionate
0.82  0.13
0.90  0.30
0.78  0.05
0.76  0.10
0.73  0.08

production involves consumption of both organic substrate and

HA (%)

hydrogen which is produced in other pathways according to the

a

theoretical equation (Eq. (1)).

AA, acetic acid; PA, propionic acid; BA, butyric acid; VA, valeric acid; HA, hexanoic acid; EtOH, ethanol;

C6 H12 O6 þ 2H2 ! 2CH3 CH2 COOH þ 2H2 O (1)

0.66  0.14
0.07  0.19
0.00  0.00
0.26  0.25
0.41  0.02
VA (%)

The HRT could be used as a tool to select microbial

populations whose growth rates are able to catch up with the
a

mechanical dilution created by continuous volumetric flow [32].

That shorter retention time led to reduction of propionate
56.01  4.37
62.22  4.23
65.88  2.39
62.81  6.01
64.90  2.78

production in anaerobic hydrogen production by mixed cultures

BA (%)

has been reported [6,18,33]. Cha and Noike [33] noticed that
a

propionate production was present at HRTs of 24 h and 48 h but

absent at a HRT of 12 h in a chemostat test using starch as the
8.92  0.81
0.34  0.11
0.27  0.02
0.24  0.28
0.14  0.10

substrate. Hussy et al. [18] found that immediate reduction of

PA (%)

propionate production was observed in a continuous hydrogen

fermentation from wheat starch as HRT was shortened to 12 h.
a

The propionate level decreased significantly as the HRT was

Soluble and gaseous metabolites obtained at steady-state stage

shortened to 6 h in the study by Lin and Chang [6]. These results

26.30  5.77
29.65  5.18
26.98  2.22
29.02  5.39
29.17  2.62

seem to provide evidence that the amount of propionate

b
AA (%)

producers in the mixed cultures is influenced by the HRT.

a

However, it was also noted that the propionate concentration did

not vary markedly in a HRT range of 4–24 h [15,23], and even
Soluble metabolites

alcohol (mmol/L)

decreased with increasing HRT [34]. The hydraulic dilution

Total VFAs and

effects in those studies appear rather inconsistent along with the

51.98  6.23c
47.71  5.60
48.71  3.34
57.15  5.20
63.60  3.13

Mean value  standard deviation.

present study. This is presumably due to the influence of other

Total products (mol/100 mol);

parameters such as substrate, inoculum and pretreatment protocol

used. Nevertheless, the retention time did show an effect of
hydraulic selection on the mixed culture in the present study.
HRT (h)

Moreover, the ratio of butyric acid and acetic acid (BA/AA

ratio) has been considered as a crucial indicator for evaluating
10
12
8
6
8

the efficiency of hydrogen-producing cultures [22,35]. The

ratios of butyrate and acetate obtained at steady-state
Run no.
Table 2

conditions are summarized in Table 2, showing BA/AA ratios

a

c
b

of 2.1–2.4, which are very close to the values reported by other

1
2
3
4
5
2122 Z.-P. Zhang et al. / Process Biochemistry 41 (2006) 2118–2123

researchers [22,36]. This corresponds to a maximal theoretical Table 3

Information of sludge samples for DGGE
hydrogen yield of 2.4 mol H2/mol glucose according to the fact
that 4 mol and 2 mol of hydrogen are generated from 1 mol of Lane no. HRT (h) Sampling
glucose, when acetate and butyrate are produced, respectively. time (day)
Since part of the glucose is shunted to the formation of ethanol 1 Seed a 0
and some other acids, i.e., valeric and hexanoic acids, which do 2 50 2
not contribute to hydrogen production, a hydrogen yield of 3 24 6
4 12 10
1.95 mol H2/mol glucose obtained in the present study is
5 8 16
reasonable and consistent with that (1.70–2.45 mol H2/mol 6 8 24
glucose) obtained in other studies using mixed cultures as the 7 6 34
hydrogen producers [11,16,23,24]. 8 8 47
9 12 64
a
3.3. Microbial diversity Seed sludge pretreated by heat and acid incubation.

The diversity of microbial communities at different HRTs seed sludge pretreated. The dominant species (the brightest
was analyzed and compared by PCR–DGGE techniques, with band in Lanes 2–4), however, were still shifting until the HRT
the DGGE profiles of 16S rDNA gene fragments, as shown in was shortened to 12 h during the start-up stage, which could
Fig. 3. Each band on the DGGE profile corresponded to a gene have contributed to a varying increase of hydrogen production
fragment of unique 16S rDNA sequences and accordingly rate as observed. Thereafter, the dominant species (Band B) in
represented a specific species in the microbial community. The the microbial community did not vary much. Moreover, the
intensity of a band represents the relative abundance of the diversity of microbial communities decreased with the HRT
corresponding microbial species [16,37]. shortening to 6 h as Bands A and C disappeared or faded, but
The DGGE profiles clearly show a shift of microbial after 15 days of acclimation it stabilized in Runs 2–5, regardless
population with HRTs (Fig. 2). An apparently intense band, of the HRTs extended. This seems to be evidences with respect
representing the dominant species on each lane was observed at to microbiological aspect to support the washout effect of HRT
each HRT after the reactor was started up as compared to the on propionate producers. Comparing Lanes 5 and 6, whose
sludge samples were taken at days 16 and 24, respectively,
during the first 8 h HRT period no variation of microbial
population can be seen, indicating that the microbial population
did not change by the acclimation process itself but by the HRT
shifts applied during this period.
In summary, shortening of the HRT to 6 h was able to
reduce the diversity of microbial population associated with
an elimination of propionate production without affecting the
existence of dominant species, which was the presumable
reason for the observed increase in hydrogen yield. On the
other hand, steady microbial population and hydrogen yield
observed as the HRT increased from 6 h to 12 h (Runs 2–5)
indicate that hydrogen yield representing the capability of
microorganisms to convert carbohydrate into hydrogen gas is
independent of the HRT. As a result, the hydrogen yield
could be considered as a function of the microbial
populations in the culture, but the HRT affects the microbial
community to a certain extent, and in turn shows an impact
on hydrogen yield. This might explain the view of HRT-
dependent hydrogen yield obtained by some other research-
ers in the similar systems [6,17,24]. In Chen and Lin’s study
[23], for example, it is likely that the increased hydrogen
yield from 0.57 mol H2/mol glucose to 1.70 mol H2/mol
glucose with a shortened HRT from 2 days to 0.5 day
resulted from the contributions of HRT-selected microbial
population.

4. Conclusions

Fig. 2. Variation of DGGE profiles with HRTs (see Table 3 for the sampling Based on the experimental results obtained, it can be
times and HRTs associated with the lanes). concluded that the hydraulic retention time (HRT) appears to be
 Z.-P. Zhang et al. / Process Biochemistry 41 (2006) 2118–2123 2123

a significant hydrodynamic selection on the mixed-microbial [17] Fan KS, Kan NR, Lay JJ. Effect of hydraulic retention time on anaerobic
hydrogenesis in CSTR. Bioresour Technol 2006;97(1):84–9.
populations. A stable microbial population was established by
[18] Hussy I, Hawkes FR, Dinsdale R, Hawkes DL. Continuous fermentative
washing out propionate producing bacteria as the HRT is hydrogen production from a wheat starch co-product by mixed microflora.
shortened to 6 h. An increase in the hydrogen yield from Biotechnol Bioeng 2003;84(6):619–26.
1.6 mol H2/mol glucose to 1.9 mol H2/mol glucose was [19] Kim SH, Han SK, Shin HS. Effect of substrate concentration on hydrogen
measured when the HRT was shortened from 8 h to 6 h. production and 16S rDNA-based analysis of the microbial community in a
Arising from the stabilized microbial population, the hydrogen continuous fermenter. Process Biochem 2006;41(1):199–207.
[20] Hartmanis M, Gatenbeck S. Intermediary metabolism in Clostridium
composition and yield were maintained at about 64% and acetobutylicum: levels of enzymes involved in the formation of acetate
1.9 mol H2/mol glucose, respectively, in the subsequent and butyrate. Appl Environ Microbiol 1984;47(6):1277–83.
operation at 8–12 h HRT. [21] Kim JO, Kim YH, Yeom SH, Song BK, Kim IH. Enhancing continuous
hydrogen gas production by the addition of nitrate into an anaerobic
reactor. Process Biochem 2006;41(5):1208–12.
References [22] Chen CC, Lin CY, Chang JS. Kinetics of hydrogen production with
continuous anaerobic cultures utilizing sucrose as the limiting substrate.
[1] Levin DB, Pitt L, Love M. Biohydrogen production: prospects and Appl Microbiol Biotechnol 2001;57(1–2):56–64.
limitations to practical application. Int J Hydrogen Energy 2004;29(2): [23] Chen CC, Lin CY. Using sucrose as a substrate in an anaerobic hydrogen-
173–85. producing reactor. Adv Environ Res 2003;7(3):695–9.
[2] Das D, Veziroglu TN. Hydrogen production by biological processes: a [24] Lin CY, Chang RC. Hydrogen production during the anaerobic acidogenic
survey of literature. Int J Hydrogen Energy 2001;26(1):13–28. conversion of glucose. J Chem Technol Biotechnol 1999;74(6):498–500.
[3] Benemann J. Hydrogen biotechnology: progress and prospects. Nat [25] Dubois M, Gilles KA, Hamilton JK, Rebers PA, Smith F. Colorimetric
Biotechnol 1996;14(9):1101–3. method for determination of sugars and related substrates. Anal Chem
[4] Kumar N, Das D. Enhancement of hydrogen production by Enterobacter 1956;28(3):350–6.
cloacae IIT-BT 08. Process Biochem 2000;35(6):589–93. [26] APHA. Standard methods for the examination of water and wastewater,
[5] Oh Y-K, Seol E-H, Kim JR, Park S. Fermentative biohydrogen production 19th ed., Washington, DC, USA: American Public Health Association;
by a new chemoheterotrophic bacterium Citrobacter sp. Y19. Int J 1998.
Hydrogen Energy 2003;28(12):1353–9. [27] Watanabe K, Teramoto M, Futamata H, Harayama S. Molecular detection,
[6] Lin C-Y, Chang R-C. Fermentative hydrogen production at ambient isolation, and physiological characterization of functionally dominant
temperature. Int J Hydrogen Energy 2004;29(7):715–20. phenol-degrading bacteria in activated sludge. Appl Environ Microbiol
[7] Chittibabu G, Nath K, Das D. Feasibility studies on the fermentative 1998;64(11):4396–402.
hydrogen production by recombinant Escherichia coli BL-21. Process [28] Watanabe K, Yamamoto S, Hino S, Harayama S. Population dynamics of
Biochem 2006;41(3):682–8. phenol-degrading bacteria in activated sludge determined by gyrB-tar-
[8] Wang CC, Chang CW, Chu CP, Lee DJ, Chang B-V, Liao CS. Producing geted quantitative PCR. Appl Environ Microbiol 1998;64(4):1203–9.
hydrogen from wastewater sludge by Clostridium bifermentans. J Bio- [29] Muyzer G, de Waal E, Uitterlinden A. Profiling of complex microbial
technol 2003;102(1):83–92. populations by denaturing gradient gel electrophoresis analysis of poly-
[9] Yokoi H, Maki R, Hirose J, Hayashi S. Microbial production of hydrogen merase chain reaction-amplified genes coding for 16S rRNA. Appl
from starch-manufacturing wastes. Biomass Bioenergy 2002;22(5):389–95. Environ Microbiol 1993;59(3):695–700.
[10] Taguchi F, Yamada K, Hasegawa K, Taki-Saito T, Hara K. Continuous [30] Chang J-S, Lee K-S, Lin P-J. Biohydrogen production with fixed-bed
hydrogen production by Clostridium sp. strain no. 2 from cellulose bioreactors. Int J Hydrogen Energy 2002;27(11–12):1167–74.
hydrolysate in an aqueous two-phase system. J Ferment Bioeng 1996; [31] Ueno Y, Otsuka S, Morimoto M. Hydrogen production from industrial
82(1):80–3. wastewater by anaerobic microflora in chemostat culture. J Ferment
[11] Van Ginkel S, Sung SW, Lay JJ. Biohydrogen production as a function of pH Bioeng 1996;82(2):194–7.
and substrate concentration. Environ Sci Technol 2001;35(24):4726–30. [32] Tijhuis L, Vanloosdrecht MCM, Heijnen JJ. Formation and growth of
[12] Lin CY, Lee CY, Tseng IC, Shiao IZ. Biohydrogen production from heterotrophic aerobic biofilms on small suspended particles in airlift
sucrose using base-enriched anaerobic mixed microflora. Process Bio- reactors. Biotechnol Bioeng 1994;44(5):595–608.
chem 2006;41(4):915–9. [33] Cha GC, Noike T. Effect of rapid temperature change and HRT on
[13] Kim JO, Kim YH, Ryu JY, Song BK, Kim IH, Yeom SH. Immobilization anaerobic acidogenesis. Water Sci Technol 1997;36(6–7):247–53.
methods for continuous hydrogen gas production biofilm formation versus [34] Dinopoulou G, Rudd T, Lester NJ. Anaerobic acidogenesis of a complex
granulation. Process Biochem 2005;40(3–4):1331–7. wastewater: I. The influence of operational parameters on reactor perfor-
[14] Majizat A, Mitsunori Y, Mitsunori W, Michimasa N, Jun’ichiro M. mance. Biotechnol Bioeng 1988;31(9):958–68.
Hydrogen gas production from glucose and its microbial kinetics in [35] Nandi R, Sengupta S. Microbial production of hydrogen: an overview. Crit
anaerobic systems. Water Sci Technol 1997;36(6–7):279–86. Rev Microbiol 1998;24(1):61–84.
[15] Yu HQ, Hu ZH, Hong TQ. Hydrogen production from rice winery [36] Kim IS, Hwang MH, Jang NJ, Hyun SH, Lee ST. Effect of low pH on the
wastewater by using a continuously-stirred reactor. J Chem Eng Jpn activity of hydrogen utilizing methanogen in bio-hydrogen process. Int J
2003;36(10):1147–51. Hydrogen Energy 2004;29(11):1133–40.
[16] Fang HHP, Liu H. Effect of pH on hydrogen production from glucose by a [37] Zhang T, Fang HH. Phylogenetic diversity of a SRB-rich marine biofilm.
mixed culture. Bioresour Technol 2002;82(1):87–93. Appl Microbiol Biotechnol 2001;57(3):437–40.