8/17/2019 Review Paper on Biohydrogen
1/22
Review
A critical literature review on biohydrogen production by pure
cultures
Omneya Elsharnouby a, Hisham Hafez b,*, George Nakhla a,c, M. Hesham El Naggar a
a Department of Civil and Environmental Engineering, The University of Western Ontario, London, Ontario N6A 5B9, CanadabGreenField Ethanol Inc., Chatham, Ontario N7M 5J4, Canadac
Department of Chemical and Biochemical Engineering, The University of Western Ontario, London, Ontario N6A 5B9, Canada
a r t i c l e i n f o
Article history:
Received 19 November 2012
Received in revised form
21 January 2013
Accepted 7 February 2013
Available online 13 March 2013
Keywords:
Biohydrogen
Fermentation
Pure cultures
Mesophilic
Thermophilic
Anaerobic
a b s t r a c t
Global research is moving forward in developing hydrogen as a renewable energy source in
order to alleviate concerns related to carbon dioxide emissions and depleting fossil fuels
resources. Biohydrogen has the potential to replace current hydrogen production tech-
nologies relying heavily on fossil fuels. Batch and continuous systems employing pure
mesophiles and thermophiles isolates and co-cultures of isolates have been investigated.
The co-cultures of the isolates achieved better results than mono-cultures of the isolates
with respect to different parameters. This paper presents a critical review of the literature
reporting on fermentative biohydrogen production by pure cultures of bacteria in different
systems. Synergies between different types of bacteria, i.e. strict and facultative, and a
comparison between mono- and co-cultures, types of feedstocks, and preferred feedstocks
for mono- and cultures are outlined.
Copyright ª 2013, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights
reserved.
1. Introduction
The challenges of environmental pollution and traditional
energy reserves depletion are focussing intensive research on
alternative energy production. Hydrogen is widely regarded as
one of the most promising energy carriers, because of its high
efficiency of conversion to usable power, non-polluting
oxidation products, and high gravimetric energy [1]. These
advantages render hydrogen as an attractive candidate to
reduce reliance on conventional fossil fuels.
Biological hydrogen production is suitable for a variety of
feedstocks including organic waste material, and is less
energy intensive compared to other hydrogen processes.
Biological methods include photosynthetic hydrogen pro-
duction and dark fermentative hydrogen production.
Photosynthetic hydrogen production involves transforming
solar energy into hydrogen via photosynthetic bacteria.
However, its application is challenged by the low transfer
efficiency of light, complexity in reactors design, and low
hydrogen production rates [2,3]. On the other hand,
fermentative hydrogen production facilitates high hydrogen
production rate through a simple operation, making the
process an increasingly popular option for hydrogen
production.
* Corresponding author. Tel.: þ1 519 784 6230; fax: þ1 519 352 9559.E-mail addresses: [email protected] (O. Elsharnouby), [email protected], [email protected] (H. Hafez),
[email protected] (G. Nakhla), [email protected] (M.H. El Naggar).
Available online at www.sciencedirect.com
j o u r n a l h o m e p a g e : w w w . e l s e v i er . c o m / l o c a t e / he
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 4 9 4 5 e4 9 6 6
0360-3199/$ e see front matter Copyright ª 2013, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.ijhydene.2013.02.032
mailto:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]://www.sciencedirect.com/science/journal/03603199http://www.elsevier.com/locate/hehttp://dx.doi.org/10.1016/j.ijhydene.2013.02.032http://dx.doi.org/10.1016/j.ijhydene.2013.02.032http://dx.doi.org/10.1016/j.ijhydene.2013.02.032http://dx.doi.org/10.1016/j.ijhydene.2013.02.032http://dx.doi.org/10.1016/j.ijhydene.2013.02.032http://dx.doi.org/10.1016/j.ijhydene.2013.02.032http://www.elsevier.com/locate/hehttp://www.sciencedirect.com/science/journal/03603199mailto:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]
8/17/2019 Review Paper on Biohydrogen
2/22
Several factors influence the fermentative hydrogen pro-
duction process, which have to be optimized for enhanced
performance. Chief among these factors are: inoculum, sub-
strate, reactor type, and temperature, which seem to impact
both hydrogen yield and hydrogen production rate, albeit with
varying importance [4]. Hydrogen yield is significantly influ-
enced by the inoculum type, as the fermentation end products
are influenced by the bacterial metabolism. The inoculumused for fermentative hydrogen production include: mixed
communities of anaerobic bacteria obtained from anaerobic
sludge digesters [5,6], compost piles [7], and pure cultures of
known species of hydrogen-producing bacteria. In pure cul-
ture systems, metabolic shifts are more easily detected due to
the reduced diversity of the biomass. Moreover, studies
employing pure cultures can reveal important information
regarding conditions that promote high hydrogen yield and
production rate [8].
Numerous pure bacterial cultures have been used in recent
studies to produce hydrogen from various substrates. Never-
theless, only a few review papers with limitedscope are found
in the literature addressing fermentative hydrogen produc-tion by pure cultures [9,10]. For example, the review article [9]
merely presented data on the challenges and prospective of
biohydrogen production by pure cultures, without any critical
analysis, and provided minimal insight on the potential
applications.
In this paper, a critical review of 195 studies employing
pure cultures was conducted considering the most important
parameters [1e103]. The relative effectiveness of co-cultures
of pure isolates and mono-cultures of these isolates is dis-
cussed. In addition, comparative studies between employing
thermophilic and mesophilic cultures, batch and continuous
systems, and the different types of feedstocks, are evaluated.
Table 1 summarizes the data of considered studies withrespect to operational and performance parameters. Sixteen
different types of pure cultures were employed in fermenta-
tive hydrogen production processes solely or co-cultured in
the studies listed in Table 1. It should be noted that for ease of
comparison, the hydrogen production rate and hydrogen yield
from these studies were normalized to L H2 /L/day and mol H2 /
mol hexose equivalent, respectively.
2. Effect of synergies between co-cultures ontechnical and economic efficiencies
Ten independent studies considered in this review havecompared the effectiveness of co-cultures of pure isolates
with their mono-cultures in fermentative hydrogen produc-
tion. Table 2 summarizes the operational and performance
parameters of the mono-cultures and co-cultures studied. In
allten studies, the co-cultures achievedbetterresults than the
mono-cultures with respect to different parameters. Exam-
ining the available literature, it is evident that the motivation
behind employing co-cultures, rather than mono-cultures,
was either economical or technical. From economy view
point, co-cultures can help maintain anaerobic conditions for
strict high hydrogen producers and eliminate the need for an
expensive reducing agent. From the technical view point, co-
cultures can improve the hydrolysis of complex sugars and
plant biomass, and can provide a wider range of pH for bac-
teria to fermenthydrogen.In moststudies, both economic and
technical reasons were important considerations and inter-
dependant. It was also found that there are primarily three
different types of co-culture of pure isolates. An explanation
of the synergistic effect in the co-culture processes for each
type is provided below.
The first type of co-cultures involves strict and facultativeanaerobes. Obligate anaerobes are extremely sensitive to O2,
and their H2-producing abilities are inhibited by a slight
amount of O2, which requires the addition of a reducing agent
such as L-cysteine to stabilize H2 production. In order to
eliminate the cost of the expensive reducing agent, facultative
anaerobes are used to consume O2 in a medium, so anaerobic
conditions are readily attained without the need for a
reducing agent. Therefore, a strict anaerobe such as Clos-
tridium sp., and a facultative anaerobe such as Enterobacter sp.
are co-cultured in the same reactor under optimum culture
conditions for H2 production, to achieve stable and high-yield
H2 production without a reducing agent.
Jenni et al. [11] investigated the applicability of mixedculture of Clostridium butyricum and Escherichia coli for stable H2production without any reducing agents. They utilized
glucose as substrate, at a temperature of 37 C, and an opti-
mum pH of 6.5 in a batch reactor. The authors noted that the
gas production pattern of batch fermentations by E. coli and C.
butyricum differed, i.e. E. coli continued gas production long
after the exponential growth. Mono-cultures of E. coli and C.
butyricum achieved H2 yields of 1.45 mol-H2 /mol-glucose
consumed, and 2.09 mol- H2 /mol-glucose consumed, while
the co-cultures achieved 1.65 mol-H2 /mol-glucose consumed.
It should be emphesized, however, that even though C.
butyricum achieved a higher molar hydrogen yield i.e. specific
hydrogen production, the co-culture facilitated a greaterglucose conversion efficiency resulting in a higher overall
volumetric hydrogen production. These findings indicated
that employing co-cultures was more economical by elimi-
nating the expensive reducing agent, and technically more
effective with higher hydrogen production.
Haruhiko et al. [12] examined the O2 tolerance and H2-
producing stability of the mono- and mixed cultures of C.
butyricum and Enterobacter aerogenes in batch and continuous
flowstudies using starch as a substrate at temperature of 37 C
and pHs of 6.5 and 5.5, respectively. In the batch study, E.
aerogenes hardly produced H2 from starch since it had no
ability to utilize starch. H2 production by C. butyricum without
a reducing agent occurred after a long lag time of 12 h. In caseof C. butyricum with 0.1% L-cysteine as reducing agent, H2 was
evolved after a short lag time of 5 h. On the other hand, H 2production by the mixed culture of C. butyricum and E. aero-
genes occurred after even a shorter lag time of less than 2 h,
and the amount of H2 evolved was the largest at 175% of that
produced by C. butyricum with a reducing agent. This confirms
that the mixed culture could produce H2 without a reducing
agent, since E. aerogenes consumed O2 rapidly, and thus
maintaining the anaerobic conditions conducive for bio-
hydrogen production. In the continuous-flow study, the case
of C. butyricum with a reducing agent, exhibited H2 production
after 15 h due to removal of O2 in the reactor by the reducing
agent. The hydrogen production by C. butyricum without the
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 4 9 4 5 e4 9 6 64946
http://dx.doi.org/10.1016/j.ijhydene.2013.02.032http://dx.doi.org/10.1016/j.ijhydene.2013.02.032http://dx.doi.org/10.1016/j.ijhydene.2013.02.032http://dx.doi.org/10.1016/j.ijhydene.2013.02.032
8/17/2019 Review Paper on Biohydrogen
3/22
Table 1 e Operational and performance parameters of the reviewed 194 experiments.
Culture(s) Reactortype
TC
Substratetype
Substrateconcentration
(g/L)
pH Hydrogen yie(mol H2 /mol glucose
equivalent)
1-Caldicellulosiruptor saccharolyticus
Batch 70 Pretreated wheat
straw
10 7.2 3.8
Batch 70 Pretreated barelystraw 20l
7 ND
Batch 70 Glucose 20 7 3.4
Batch 70 Carrot pulp
hydrolysate
10 7 2.8
Batch 72 Glucose 31 7 2.8
Batch 72 Glucose 10 7 3.4
Batch 72 PSPa 10 7 3.5
Batch 72 PSP-H2b 10 7 3.4
Batch 72 GXSc 10 6.8m 3.2
Batch 72 SSBd 20l 6.8m 2.8
Batch 70 Sucrose 10 7m 2.96
Batch 72 Glu/Xyle 10 7 3.4
Batch 72 Glu/Xyle 14 7 3.3
Batch 72 Glu/Xyle 28 7 2.4
Batch 72 Miscanthushydrolysate
10 7 3.4
Batch 72 Miscanthus
hydrolysate
14 7 3.3
Batch 72 Miscanthus
hydrolysate
28 7 2.4
2-Thermotoga neapolitana
DSM 4359 CSABR 75 Xylose 5 7 3.36
DSM 4359 Batch 75 Xylose 5 7.5 1.31
Batch 75 Glucose 27 7 3
Batch 75 Glucose 10 7 2.9
Batch 75 PSP-H2b 10 7 3.3
Batch 75 PSPa 10 7 3.8
Batch 75 Glucose 10 7 3.5
Batch 75 Glucose 20 7 3.4
Batch 75 Fructose 10 7 3.4 Batch 75 fructose 20 7 3.2
Batch 75 Glucose þ
Fructose
10 7 3.3
Batch 75 Glucose þ
Fructose
20 7 3
Batch 75 Carrot pulp
hydrolysate
10 7 2.7
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-
8/17/2019 Review Paper on Biohydrogen
4/22
Table 1 e ( continued )
Culture(s) Reactortype
TC
Substratetype
Substrateconcentration
(g/L)
pH Hydrogen yie(mol H2 /mol glucose
equivalent)
Batch 75 Carrot pulp
hydrolysate
20 7 2.4
DSM 4359 Fed
batch-CSABR
75 Xylose 5 7.5m 2.66
Fed batch-
CSABR
75 Glucose 5 7.5m 3.2
Fed batch-
CSABR
75 Sucrose 5 7.5m 2.5
Batch 85 Glucose 2.5 7.5 3.75
Batch 77 Glucose 2.5 7.5 3.85
Batch 70 Glucose 2.5 7.5 3.18
Batch 65 Glucose 2.5 7.5 3.09
Batch 60 Glucose 2.5 7.5 2.04
Batch 80 Glucose 7.5 7.5 1.84
Batch 80 Glu/Xyle 10 7 3.3
Batch 80 Glu/Xyle 14 7 3.2
Batch 80 Glu/Xyle 28 7 2.5
Batch 80 Miscanthushydrolysate
10 7 2.9
Batch 80 Miscanthus
hydrolysate
14 7 3.2
Batch 80 Miscanthus
hydrolysate
28 7 2
Batch-
Serum
bottels
80 Glucose 5 7.5 3.85
3-Clostridium DMHC-10
Batch 37 Glucose 10 5 3.35
4-Enterobacter Cloacae
IIT-BT08 Batch 36 Sucrose 10 6 3.014
Batch 36 Glucose 10 6 2.2
Batch 36 Cellobiose 10 6 1.42
F.P01 Batch 36 Maltose 10 5 0.729 DM11 Batch 37 Glucose 10 6.5 3.31
Batch 37 Glucose 10 6.5 2.2
Batch 37 Glucose 10 6.5 3.1
5-Clostridium beijerinckii
L9 Batch 35 Glucose 3 7.2 2.81
Fanp 3 Batch 35 Glucose 10 6.47e6.98 2.52
AM21B Batch 36 Glucose 10 6.5 2
AM21B Batch 36 Starch 10 6.5 1.8
RZF-1108 Batch 35 Glucose 9 7 1.97
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-
8/17/2019 Review Paper on Biohydrogen
5/22
Batch 30 Glucose 2.34 6.3 ND
Batch 36 Starch 10 6.8 ND
Batch 41 Starch 10 6.8 ND
Batch 36 Starch 10 6m ND
Batch 36 Starch 10 7m ND
L9 Batch 35 Glucose 6 6.4 1.72
RZF-1108 Batch 35 Glucose 10 6.5 1.96
6-Clostridium butyricum
ATCC19398 Batch 36 Glucose 3 7.2 2.29
CGS5 Batch 37 Sucrose 17.8 5.5 1.39 W5 Batch 39 Glucose 10 6.5 0.81
EB6 Batch 37 POMEf ND 5.5 0.22
EB6 Batch 37 Glucose 10 6 0.6
Batch 37 SCB hemicellulose
hydrolysateh20k 5.5 1.73g
Batch 37 Glucose (and 200e
400 mg/l phenol))
5 6.5 1.46
CGS5 Batch 37 Xylan hydrolysate 14.2 7.5 0.84
CGS5 Batch 37 Pretreated straw
hydrolysate
9.2 7.5 0.91
CWBI1009 Batch 30 Glucose 5 5.2m 1.7
EB6 Batch 37 Glucose 15.7 5.6 2.2
TISTR 1032 Batch 37 Sugarcane juice 22.3
(sucrose)
6.5 1.33
TISTR 1032 Batch 37 Sucrose 22.3 6.5 1.34 TISTR 1032
(immobilized)
Repeated
batch
37 Sugarcane juice 22.3
(sucrose)
6.5 1.52
CGS5 Batch 37 Chlorella vulgaris ESP6
(microalgal hydrolysate)
9 5.5m ND
W5 Batch 37 Molass 100 7 1.63
TM-9A Batch 37 Glucose 10 8 3.1
TM-9A Batch 37 Arabinose 10 8 0.06
TM-9A Batch 37 Raffinose 10 8 2.7
TM-9A Batch 37 Sucrose 10 8 1.49
TM-9A Batch 37 Trehalose 10 8 1.61
TM-9A Batch 37 Xylose 10 8 0.59
TM-9A Batch 37 Cellobiose 10 8 0.94
TM-9A Batch 37 Cellulose 10 8 0.06
TM-9A Batch 37 Ribose 10 8 0.84
TM-9A Batch 37 Galactose 10 8 0.86
TM-9A Batch 37 Fractose 10 8 0.84
TM-9A Batch 37 Mannose 10 8 0.67
W5 Batch 39 Molasses 100 6.5 1.85
Batch 37 Glucose 3 6.5 2.09
C. butyricum
and Escherichia
coli
Batch 37 Glucose 3 6.5 1.65
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-
8/17/2019 Review Paper on Biohydrogen
6/22
Table 1 e ( continued )
Culture(s) Reactortype
TC
Substratetype
Substrateconcentration
(g/L)
pH Hydrogen yie(mol H2 /mol glucose
equivalent)
C. butyricum and
Enterobacter
aerogenes HO-39
Repeated
batch
37 Sw eet potato
starch residue
ND 5.25m 2.7
7-Thermoanaerobactermathranii A3N
A3N Batch 70 Sucrose 10 8 2.69
A3N Batch 70 Glucose 10 8 2.64
A3N Batch 70 Xylose 5 8 2.5
A3N Batch 70 Starch 5 8 ND
8-Thermoanaerobacterium thermosaccharolyticum
W16 Batch 60 Xylose 10 6.5 2.62
PSU-2 Batch 60 Sucrose 20 6.25 2.53
W16 Batch 60 Xylose þ
Glucose
10 6.5 2.45
W16 Batch 60 Glucose 10 6.5 2.42
W16 Batch 60 Hydrolysed
corn stover
ND 7 2.24g
W16 Batch 60 Glucose 5 6.7 2.07
W16 Batch 60 Glucose 10 7 ND W16 Batch 60 Xylose 10 7 ND
W16 Batch 60 Glucose þ
Xylose þ
Arabinose
7.5,2.2,0.3 7 ND
W16 Batch 60 Hydrolysed
corn stover
ND 7 ND
PSU-2 Cont. UAi 60 Sucrose 20 5.5 1.11
PSU-2 Cont.UASB j 60 Sucrose 20 5.5 1.77
W16 Batch 60 Xylose 12.24 6.8 2.84
Thermoanaerobacterm
thermosaccharolyticum
GD17 and C. thermocellum JN4
Batch 60 Cellulose 5 4.4 1.8
9-Ethanoligenens harbinese
YUAN-3 Batch 35 Glucose 10 5 1.91
YUAN-3 CSTR 35 Glucose 10 5 1.93 B49 Batch 37 Glucose 9 7 1.83
B49 Batch 37 Glucose 12 7 1.71
B49 Batch 37 Glucose 6 7 1.36
B49 Batch 35 Glucose 10 6 1.67
B49 Batch 37 Glucose 10 7 2.2
B49 Batch 35 Glucose 14.5 6 2.2
B49 Batch 35 Glucose 10 6 2.26
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-
8/17/2019 Review Paper on Biohydrogen
7/22
Ethanoligenens
harbinese B49
and C. acetobutylicum X9
Batch 37 Microcrystalline
cellulose
10 5 1.32
10-Klebsiella pneumoniae
ECU-15 Batch 37 Glucose 10 6 2.07
DSM2026 Batch 37 Glycerol 20 6.5 0.53
11-Pantoea agglomerans
Batch 37 Glucose 10 7.2 3.8
Batch 37 Glucose 20 7.2 4.2
Batch 37 Glucose 20(saline
conditions)
7.2 3.3
12-Clostridium tyrobutyricum
JM1 Batch 37 Glucose 20 6.3 3.24
JM1 CSTR 37 Glucose 5 6.7 1.81
FYa102 Batch 35 Glucosen 3 7.2 1.47
FYa102 CSTR 35 Glucoseo 12 6 1.06
FYa102 CSTR 35 Glucosep 12 6 1.42
ATCC 25755 Fed batch 37 Glucose 50 5.7 2.33
13-Clostridium acetobutylicum
M121 Batch 37 Glucose 3 7.2 1.8
X9 Batch 37 Microcrystalline
cellulose
10 5 0.59
ATCC 824 Cont.
Trickling bed reactor
30 Glucose 10 6.2 0.9
ATCC 824 Batch 36 Cassava
wastewater
5k 7 2.41
14-Escherichia coli
Batch 37 Glucose 3 6.5 1.45
S3 Batch 30 Glucose 5 6.8 0.84
S6 Batch 30 Glucose 5 6.8 0.49
WDHL Batch 37 Glucose 15 6 0.3
WDHL Batch 37 Galactose 15 6 1.12
WDHL Batch 37 Lactose 15 6 1.02
WDHL Batch 37 Glucose þ
Galactose
7.5, 7.5 6 1.02
DJT135 Batch 35 Arabinose 10 6.5 1.2
DJT135 Batch 35 Fractose 10 6.5 1.27
DJT135 Batch 35 Galactose 10 6.5 0.69
DJT135 Batch 35 Glucose 10 6.5 1.51
DJT135 Batch 35 Lactose 10 6.5 0.37
DJT135 Batch 35 Maltose 10 6.5 0.2
DJT135 Batch 35 Mannitol 10 6.5 0.88
DJT135 Batch 35 Sorbitol 10 6.5 1.36
DJT135 Batch 35 Sucrose 10 6.5 0.35
DJT135 Batch 35 Terhalose 10 6.5 0.52
DJT135 Batch 35 Xylose 10 6.5 0.68
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-
8/17/2019 Review Paper on Biohydrogen
8/22
Table 1 e ( continued )
Culture(s) Reactortype
TC
Substratetype
Substrateconcentration
(g/L)
pH Hydrogen yie(mol H2 /mol glucose
equivalent)
SH5 Fed-Batch 37 Soduim
formate
16.1 6.5 ND
15-Clostridium thermocellum
JN4 Batch 60 Cellulose 5 4.4 0.8
7072 Batch 55 Cellulose 5 7.4 1.2
7072 CSTR, 100 L Cron stalk 30 7.4 0.45
7072 CSTR, 10 L 55 Cron stalk 30 7.4 0.43
7072 Batch 55 Cron stalk 5 7.4 ND
ATCC 27405 CSTR 60 Cellulose 4 7 1.29
ATCC 27405 CSTR 60 Cellulose 3 7 1.53
ATCC 27405 CSTR 60 Cellulose 2 7 1.65
ATCC 27405 CSTR 60 Cellulose 1.5 7 0.98
ATCC27405 Batch 55 Delignefied
wood fibres
0.1 6.5 1.6
ATCC 27405 Batch 60 Cellulose 1 6.8 1.9
C. thermocellum
and C. thermosaccharolyticm
Batch 55 Corn stalk
waste
10 7.2 ND
C. thermocellum and
C.thermosaccharolyticm
CSTR 55 Corn stalk
waste
10 7.2 ND
C. thermocellum DSM1237
and C.thermopalmarium
DSM 5974
Batch 55 Cellulose 9 7 1.36
16- Enterobacter aerogenes
HO-39 Batch 38 Glucose 10 6.5 1
HO-39 Fed batch 37 Glucose 10 6.5 0.8
HO-39 Batch 38 Maltose 10 6.5 0.7
ATCC29007 Batch 38 Glucose 21.25 6.13 ND
Batch 38 Glucose 0.2 7 ND
Batch 37 Glycerol 20 7 0.2
Batch 37 Glucose 10 5.8 0.89
E 82005 Batch 38 Glucose 10 5.8 1
IAM 1183 Batch 37 Xylose 5 6.3 2.64
IAM 1183 Batch 37 Galactose 10 6.3 2.82 IAM 1183 Batch 37 Mannose 25 6.3 0.96
IAM 1183 Batch 37 Rahamnose 5 6.3 0.48
IAM 1183 Batch 37 Arabinose 10 6.3 1.56
ATCC35029 Batch 37 Glycorel 21 ND 1.22
NCIMB 10102 Continuous
packed col.
40 Corm starch
hydrolysate
ND 5.5 2.55
NCIMB 10102 Batch 40 Starch
hydrolysate
20 6.5 1.09g
W23 Batch 35 Glucose 5 6.5 1.87
http://-/?-http://-/?-
8/17/2019 Review Paper on Biohydrogen
9/22
Enterobacter aerogenes
W23 and Candida
maltosa HY-35
Batch 35 Glucose 5 6.5 2.19
Enterobacter aerogenes
and C. butyricum
Batch 36 Starch ND 6.5 2
Enterobacter aerogenes
and C. butyricum
(immobiolized)
Repeated
batch
36 Starch ND 5.5 2.6
ND: Not defined.a Untreated potato steam peels, Molar yields were based on the amount of starch in untreated PSP assuming 100% starch consumption.
b The starch in the PSP was liquefied with alpha-amylase, and then the liquefied starch was further hydrolysed to glucose by amyloglucosidase.
c Mixture of glucose, xylose and sucrose.
d Sweet sorgham bagasse.
e Glucose: Xylose ¼ 7:3.
f Palm oil mill effluent.
g Mol H2 /mol total sugar.
h Sugarcane bagasse hemicellulose hydrolysate.
i Up flow carrier free anaerobic system.
j Up flow anaerobic sludge blanket.
k g COD/L.
l g sugars/L.
m Controlled pH.
n 2 g/L peptone was added with the substrate.
o 8 g/L peptone was added with the substrate.p 0.36 g/L, 1.4 g/L peptone and ammonium chloride respectively, were added with the substrate.
8/17/2019 Review Paper on Biohydrogen
10/22
reducing agent was inhibited by oxygen, and could not be
recovered at all even after 50 h, indicating complete damage
for the bacterial cells. In contrast, H2 production by the mixed
culture, which was initially inhibited by oxygen, was recov-
ered immediately within 0.5 h. These results confirm that the
mixed culture can remove O2 in the reactor and recover H2production immediately.
Similarly, Yokoi et al. [13] investigated the synergies be-tween strict and facultative anaerobes. A sustained high bio-
hydrogen yield of 2.7 mol H2 /mol glucose was attained by a
mixed culture of C. butyricum and E. aerogenes HO-39. The
mixed bacteria utilized starch waste consisting of sweet po-
tato starch residue asa carbon source, and corn steepliquor as
a nitrogen source. The experiment was conducted in a fed
batchculture, at a temperature of 37 C, and a controlled pH of
5.25. The results proved that a mixed culture of C. butyricum
and E. aerogenes could produce hydrogen from starch at a high
yield of more than 2 mol of hydrogen per 1 mol of glucose
without any reducing agents, since E. aerogenes, a facultative
anaerobe, removed oxygen and generated anaerobic condi-
tions in the reactor.
The second type of co-cultures reported in the literature
was between cellulose degrading anaerobes and high
hydrogen producers via fermenting simple sugars. The most
common dark fermentation procedure employed to generate
hydrogen from cellulose materials involved expensive pre-
treatment processes, such as delignification, and hydrolysis
[14]. Since pre-treatment processes are expensive, fermenta-
tive hydrogen production from cellulosic materials is desir-able. Therefore, many studies investigated employing co-
cultures of two bacterial strains: one with the capability of
hydrolysing cellulose, and the other is a high hydrogen pro-
ducer utilizing simple sugars.
For example, Yan et al. [15] investigated biohydrogen pro-
duction from cellulose using the thermophilic anaerobic
bacterium Clostridium thermocellum JN4, and the co-cultures of
the aforementioned bacterium with Thermoanaerobacterium
thermosaccharolyticum GD17. The C. thermocellum JN4 can
decompose cellulose but cannot completely utilize the cello-
biose and glucose produced by the cellulose degradation,
while the T. thermosaccharolyticum GD17 can utilize mono-
sugars. The experiment was conducted at pH of 4.4, and
Table 2 e Operational and performance parameters for studies employing mono and co-cultures.
Culture(s) Reactortype
TC
Substratetype
Substrateconcentration
(g/L)
pH Hydrogenyield (mol H2 /mol/glucose,
hexose equivalent)
H2productionrate (L/L/d)
Ref. no.
Clostridium butyricum Batch 37 Glucose 3 6.5 2.09 0.41 [11]
Escherichia coli Batch 37 Glucose 3 6.5 1.45 0.33 [11]C. butyricum and
Escherichia coli
Batch 37 Glucose 3 6.5 1.65 0.52 [11]
Enterobacter aerogenes
and C. butyricum
Batch 37 Starch ND 6.5 2 ND [12]
Enterobacter aerogenes
and C. butyricum
(immobiolized)
Repeated
batch
37 Starch ND 5.5 2.6 ND [12]
Enterobacter aerogenes
HO-39 and C. butyricum
Repeated
batch
37 Sweet
potato
starch
residue
ND 5.25a 2.7 0.977 [13]
C. thermocellum JN4 Batch 60 Cellulose 5 4.4 0.8 0.01 [15]
Thermoanaerobacterium
thermosaccharolyticum
GD17 and C. thermocellum JN4
Batch 60 Cellulose 5 4.4 1.8 0.33 [15]
C. acetobutylicum X9 Batch 37 Microcrystall ine
cellulose
10 5 0.59 21.33 [14]
C. acetobutylicum X9 and
Ethanoligenens harbinese
Batch 37 Microcrystall ine
cellulose
10 5 1.32 11.08 [14]
Clostridium thermocellum
and C. thermosaccharolyticum
Batch 55 Corn stalk
waste
10 7.2 ND 0.34 [16]
Clostridium thermocellum
and C. thermosaccharolyticum
CSTR 55 Corn stalk
waste
10 7.2 ND 0.44 [16]
Clostridium thermocellum
DSM1237 and C.
thermopalmarium
DSM5974
Batch 55 Cellulose 9 7 1.36 0.42 [17]
Enterobacter aerogenes W23 Batch 35 Glucose 5 6.5 1.87 5.8 [18]
Enterobacter aerogenesW23 and Candida
maltosa HY-35
Batch 35 Glucose 5 6.5 2.19 6.27 [18]
ND: Not defined.
a Controlled pH.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 4 9 4 5 e4 9 6 64954
http://-/?-http://dx.doi.org/10.1016/j.ijhydene.2013.02.032http://dx.doi.org/10.1016/j.ijhydene.2013.02.032http://dx.doi.org/10.1016/j.ijhydene.2013.02.032http://dx.doi.org/10.1016/j.ijhydene.2013.02.032http://-/?-
8/17/2019 Review Paper on Biohydrogen
11/22
temperature of 60 C in a batch reactor.The C. thermocellum JN4
resulted in hydrogen yield of about 0.8 mol H2 /mol glucose,
with lactate as the main product. When C. thermocellum JN4
was co-cultured with T. thermosaccharolyticum GD17, hydrogen
production was doubled and H2 yield increased to a high level
of 1.8 mol H2 /mol glucose. Butyrate was the most abundant
byproduct and lactate was not detected at the end of the co-
cultures process.Aijie et al. [14] employed dark fermentation of microcrys-
talline cellulose to produce biohydrogen using mono and co-
cultures. Clostridium acetobutylicum ATCC 824(X), a high
hydrogen producer from microcrystalline cellulose was uti-
lized to produce hydrogen at temperature of 37 C, and pH of
5.0. The mono-culture of X9 yielded hydrogen after a 5-h time
lag. The corresponding hydrogen yield, maximum hydrogen
production rate, and cellulose hydrolysis ratio reached
755 mL/L medium, 6.4 mmol H2 /h/g dry cell, and 68.3%,
respectively. The co-cultures of C. acetobutylicum X9 and strain
Ethanoligenens harbinense B49, which can produce hydrogen
efficiently from both monosaccharides and microcrystalline
cellulose, yielded hydrogen immediately following initiationof fermentation. The hydrogen yield, maximum hydrogen
production rate, and cellulose hydrolysis ratio of 1810 mL/L
medium, 55.4 mmol H2 /h/g dry cell, and 77.6%, respectively,
were achieved. The strain B49 rapidly removed reducedsugars
produced by cellulose hydrolysis by X9, hence improving
cellulose hydrolysis and subsequent hydrogen production.
Another example of technical and economical efficiencies
attained by the synergies between co-cultures in fermentative
hydrogen production process is provided by Qian et al. [16].
The authors utilized a combination of cellulose-hydrolysing
bacteria and highly efficient hydrogen producing bacteria to
optimize hydrogen production from cellulosic waste. They
employed a mix of C. thermocellum and Clostridium thermo-saccharolyticum utilizing corn stalk waste. C. thermocellum is a
cellulose-degrading bacterium, which has the potential for
direct hydrogen production from lignocellulosic waste, hence
eliminating the need for an extensive hydrolysing process, but
with low biohydrogen yield. The C. thermosaccharolyticum is a
non-cellulolytic high hydrogen-producing stain. The experi-
ments were conducted in both batch and continuous-flow
modes at temperature of 55 C, and pH of 7.2. At the end of
the C. thermocellum mono-culture experiment, cellobiose,
glucose, and xylose contents were found in the fermentation
broth. However, cellobiose and xylose were not detected at the
end of the C. thermosaccharolyticum and C. thermocellum co-
cultures experiments because C. thermosaccharolyticum cul-tures produced hydrogen, organic acids (acetate), and other
components. The hydrogen yield in the co-culture batch
fermentation reached 68.2 mL/g-cornstalk, which was 94.1%
higher than that in the mono-culture, and the rate of
hydrogen production reached 14.1 mL H2 /L/h. A hydrogen
yield of 74.9 mL/g cornstalk, as well as production rate of
18.5 mL H2 /L/h were achieved using the optimized co-culture
method in the scaled-up reactor.
Alei et al. [17] investigated employing a cellulolytic
hydrogen-producing bacterium and a non-cellulolytic high
hydrogen-producing bacterium. In their study, C. thermocellum
DSM 1237, a cellulolytic hydrogen-producing bacterium, was
co-cultured with Clostridium thermopalmarium DSM 5974, a
non-cellulolytic high hydrogen-producing bacterium. The
bacteria utilized cellulose as the sole substrate, at tempera-
ture of 55 C, and pH of 7.0. The co-culture produced nearly
double the amount of hydrogen produced in C. thermocellum
monocultures. Ethanol and acetate were the main metabolites
in C. thermocellum monocultures, whereas the co-cultures
produced butyrate as the main metabolite. These results
support the synergies between cellulolytic anaerobes andhigh-yield biohydrogen producers. It is thought that using co-
cultures of various types of complex sugars degrading anaer-
obes in general and high hydrogen producers would give the
same results as using co-cultures of cellulose degrading an-
aerobes and high hydrogen producers.
The third type of co-cultures reported in the literature in-
volves aciduric hydrogen producing microorganisms and high
hydrogen producers. Due to the inhibitory impact of organic
acids produced during fermentation on biohydrogen pro-
ducers, near neutral or weak acidic conditions are mandatory
to attain high hydrogen yields. However, pH control is un-
economical due to the large quantity of chemicals needed.
Aciduric microorganisms can produce hydrogen at low pHs,hence reducing or even eliminating buffering requirements.
Thus, using a hydrogen producing bacterium co-cultured with
aciduric microorganisms can achieve stable and high-
hydrogen production at low pHs [18]. In this study, a batch
experiment was carried out to measure the hydrogen-
producing ability of a mixed culture of Candida maltosa HY-
35, an aciduric microorganism, which can produce hydrogen
at pH as low as 1.3, and a facultative anaerobe E. aerogenes W-
23, which has aciduric hydrogen producing properties (can
produce hydrogen at pH of 4.0 ). These mixed cultures at 35 C
attained a hydrogen yield of 1735 mL/L, representing 17.15%
and 119.90% higher yield than the monocultures E. aerogenes
W-23 and C. maltosa HY-35, respectively. Meanwhile, theaverage hydrogen production rate of the mixed culture was
261.1 mL/h/L, which was 7.85% and 146.23% higher than those
of the monoculture of E. aerogenes W-23 and C. maltosa HY-35,
respectively. In this case, the co-cultures of hydrogen pro-
ducing aciduric microorganism and high hydrogen producing
anaerobe allowed for dispensing or reducing amount of buff-
ering agents by providing a wider range of pH for bacteria to
ferment within. In addition, Candida consumed the lactate,
succinic and citric acids produced by Enterobacter during
fermentation and slowed the shift in pH. These results
confirm the synergies between the mixed cultures.
The success of co-cultures of high hydrogen producing
bacteria and hydrogen producing aciduric bacteria empha-sizes the potential for employing co-cultures of multi species
high-hydrogen producing-bacteria with different optimum pH
ranges to naturally realize hydrogen production over a wide
pH range. For instance, it was found that the optimum pH for
hydrogen production and growth rate for Clostridium DMHV-10
is 5.0 [19]. Thebacterium fermented 10 g/L glucose with a yield
of 3.35 mol H2 /mol glucose, at temperature of 37 C, and an
initial pH of 5.0. The bacterium is capable of growth and
hydrogen production within a pH range of 5.0e7.0. On the
other hand, it was reported thatthe optimum pH for hydrogen
production and growth rate for the hydrogen producer Enter-
obacter Cloacae was found to be 6.5 [20]. The bacterium yielded
3.31 mol H2 /mol glucose from fermenting 10 g/L glucose at a
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 4 9 4 5 e4 9 6 6 4955
http://dx.doi.org/10.1016/j.ijhydene.2013.02.032http://dx.doi.org/10.1016/j.ijhydene.2013.02.032http://dx.doi.org/10.1016/j.ijhydene.2013.02.032http://dx.doi.org/10.1016/j.ijhydene.2013.02.032
8/17/2019 Review Paper on Biohydrogen
12/22
temperature of 37 C, and an initial pH of 6.5. This bacterium
was capable of growing and producing hydrogen within a pH
range of 4.5e8.0. Thus, these two strains could be co-cultured
at an initial pH of 6.5 eliminating the need for a buffer and a
reducing agent to maintain the anaerobic conditions for the
Clostridium bacteria.
3. Comparative study between thermophilesand mesophiles
Temperature is one of the most important operational pa-
rameters in fermentative H2 production. Temperature affects
the growth rate metabolic pathways of microorganism, sub-
strate hydrolysis rate and hydrogen production rate.
Fermentative reactions can be operated at mesophilic
(25e40 C), thermophilic (40e65 C) or hyperthermophilic
(>80 C) temperatures [21]. It has been demonstrated that
within a specific temperature range, increasing the tempera-
ture accelerates hydrogen production, with sharply dropping
activity of hydrogen producers outside the optimum temper-ature range [8]. The approximately 200 investigations
reviewed in this study can be classified into two groups: 116
studies were conducted within mesophilic range; and 78
studies were carried out within thermophilic range. Per-
forming experiments employing mesophilic cultures is
generally less expensive. However, it was reported that ther-
mophilic and hyperthermophilic cultures seem to exhibit su-
perior performance in hydrogen production. The highest
reported hydrogen yields in the literature, which were close to
the theoretical maximum of 4.0 mol-H2 /mol-glucose, were
achieved by using extreme thermophiles [22,23].
In general, thermophiles are thought to be robust micro-
organisms that produce stable enzymes. It is widely acceptedthat more hydrogen can be produced under thermophilic
conditions than under mesophilic conditions [24]. However,
the data available in the literature does not always support
this hypothesis, and seem to be substrate dependant. This is
because some mesophilic bacteria have better bacterial ki-
netics than thermophilic ones utilizing the same substrate,
despite operating at much lower temperatures. For instance,
the hyperthermophilic bacterium, Thermotoga neapolitana in a
batch experiment at a temperature of 77 C, and a pH of 7.5,
was capable of producing 3.85 mol H2 /mol glucose, from 2.5 g/
L glucose, with a hydrogen production rate of 0.56 L/L/d [22].
Giuliana et al. [23] reported that T. neapolitana achieved a
maximum hydrogen yield of 3.85 mol H2 /mol and a maximumhydrogen production rate of 1.2 L/L/d, utilizing 5 g/L of glucose
in serum bottles at a temperature of 80 C, pH of 7.5. On the
other hand, the maximum hydrogen yield of 3.8 mol H2 /mol
glucose and hydrogen production rate of 1.82 L/L/d were
attained by the mesophilic bacterium Pantoea agglomerans
utilizing 10 g/L glucose as substrate, at a temperature of 37 C,
and a pH of 7.2 [25]. Although the latter bacterium was oper-
ating at mesophilic tempratures, it produced hydrogen at a
higher rate than the thermophilic one from glucose (mono-
saccharide), and with almost the same yield.
The maximum hydrogen yield reported in the literature by
a thermophile utilizing fructose (another type of mono-
saccharides) was 3.4 mol H2 /mol hexose equivalent, with a
maximum hydrogen production rate of 2.4 L/L/d, by the bac-
teria T. neapolitana [26]. The bacteria utilized 10 g/L fructose at
a temperature of 75 C, and a pH of 7.0. Nevertheless, the
maximum hydrogen yield reported in the literature by a
mesophile utilizing 10 g/L fructose in a batch at a temperature
of 35 C, and a pH of 6.5 was 1.27 mol H 2 /mol hexose equiva-
lent by the bacteria E. coli [27].
The maximum hydrogen yield attained by a thermophileutilizing sucrose (i.e. di-saccharide) was 2.96 mol H2 /mol
hexose equivalent with a hydrogen production rate of 4.5 L/L/
d, which was achieved using Caldicellulosiruptor saccharolyticus,
at a pH of 7, a temperature of 70 C, and an initial concen-
tration of 10 g/L, in a batch reactor [28]. The maximum
hydrogen yield of a mesophile utilizing sucrose was reported
by Narendra et al. [29] for E. cloacae. They achieved a hydrogen
yield and a hydrogen production rate of 3.1 mol H2 /mol hexose
equivalent and 15.84 L/L/d, respectively, at a temperature of
36 C, a pH of 6.0, and initial sucrose concentration of 10 g/L.
It has beenrecently reported [30] that mesophilic anaerobic
bacteria cannot utilize cellulose (i.e. complex sugars) effec-
tively. The addition of exogenous cellulose enzymes isnecessary for hydrolysis of cellulose to generate H2 by meso-
philic anaerobic bacteria. On the other hand, thermophilic
anaerobic bacteria can effectively utilize cellulose [31], and
therefore, they have a great potential for H2 production from
cellulose without the addition of exogenous cellulose [15]. In
addition, the high operating temperature of the thermophiles
enhances the hydrolysis rate. For example, Rumana et al. [32]
reported that the bacterium C. thermocellum utilized 1 g/L cel-
lulose, and produced a hydrogen yield of 1.9 mol H2 /mol
hexose equivalent. On the other hand, the maximum attained
hydrogen yield reported in the literature from mesophiles (C.
acetobutylicum and E. harbinense) utilizing cellulose at a con-
centration of 10 g/L was only 1.32 mol H2 /mol hexose equiv-alent [14]. These results confirm that thermophiles can more
effectively utilize complex sugars for hydrogen production
than mesophiles.
4. Bioreactor configuration
Fermentative hydrogen production was applied in both batch
reactors and continuous systems. Batch-mode reactors are
easily and flexibly operated. This has resulted in the wide
utilization of batch reactors for determining the biohydrogen
potential of organic substrates. In industrial context, however,
continuous bioprocesses are recommended for practicalconsiderations such as waste stock management, economic
feasibility, and practical engineering design [33].
Seven cases of batch studies are reported in the literature
to be successfully scaled up or applied to continuous-flow
systems. The performance of scaled up continuous systems
varied from one study to another. In some cases, the same or
even better performance was achieved compared to their
batch counterparts. In other cases, however, the continuous
systems displayed lesser but stable performance. Table 3 il-
lustrates the differences in performance and operational pa-
rameters between the batches and continuous-flow systems
employing the same bacteria, and utilizing the same sub-
strates. Scrutinizing the results in Table 3, it may be concluded
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 4 9 4 5 e4 9 6 64956
http://dx.doi.org/10.1016/j.ijhydene.2013.02.032http://dx.doi.org/10.1016/j.ijhydene.2013.02.032http://dx.doi.org/10.1016/j.ijhydene.2013.02.032http://dx.doi.org/10.1016/j.ijhydene.2013.02.032
8/17/2019 Review Paper on Biohydrogen
13/22
Table 3 e Operational and performance parameters of the batch and their continuous counterparts reviewed studies.
Culture Reactortype
TC Substratetype
Substrateconcentration
(g/L)
pH Hydrogen yield(mol H2 /mol
glucose, hexoseequivalent)
1-Clostridium thermocellum
7072 Batch 55 Corn stalk 5 7.4 ND
7072 CSTR (10 L) 55 Corn stalk 30 7.4 0.43
7072 CSTR (100 L) 55 Corn stalk 30 7.4 0.45 2-Clostridium tyrobutyricum
FYa102 CSTR 35 Glucoseb 12, 8 6 1.06
FYa102 CSTR 35 Glucosec 12, 0.35, 1.4 6 1.42
FYa102 Batch 35 Glucosed 3, 2 7.2 1.47
JM1 CSTR 37 Glucose 5 6.7 1.81
JM1 Batch 37 Glucose 20 6.3 3.24
3-Ethanoligenens harbinese
YUAN-3 Batch 35 Glucose 10 5 1.91
YUAN-3 CSTR 35 Glucose 10 5 1.93
4-Thermotoga neapolitana
DSM 4359 Batch 75 Xylose 5 7.5 1.31
DSM 4359 CSTR 75 Xylose 5 7 3.36
5-Thermoanaerobacterium thermosaccharolyticum
PSU-2 Cont. UASBa 60 Sucrose 20 5.5 1.77
PSU-2 Batch 60 Sucrose 20 6.25 2.53
ND: Not defined.
a UASB ¼ up flow anaerobic sludge blanket.
b 2 g/L peptone was added with the substrate.
c 8 g/L peptone was added with the substrate.
d 0.36 g/L, 1.4 g/L peptone and ammonium chloride respectively, were added with the substrate.
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-
8/17/2019 Review Paper on Biohydrogen
14/22
that optimizing the operational parameters is the key for
achieving a successful continuous-flow system.
Continuous-flow system performance was superior to the
batch in thermophilic fermentation of cornstalk by C. ther-
mocellum 7072 [1]. In the batch test involving a substrate of 5 g/
L cornstalk at a temperature of 55 C and a pH of 7.4, hydrogen
yield and hydrogen production rate of 38.8 mL/g and 201.4mL/
L/h were achieved. The continuous stirred tank reactor (CSTR)was fed with 30 g/L of cornstalk, and operated at the same
temperature and pH as the batch test. The maximum
hydrogen production rate in the 10 L CSTR, and the 100 L CSTR
was 767.5, and 739.9 mL/L/h, respectively. The authors
attributed the higher hydrogen production rates and shorter
lag-phase period observed in the two CSTRs to the improved
mixing conditions in the two reactors, compared to the
anaerobic bottles. The hydrogen yield in the 10 and 100 L
CSTRs reached 58.3 and 61.4 mL/g of cornstalk. Acetate and
ethanol were the major end products of fermentation by C.
thermocellum for cornstalk, and the ratio of ethanol/acetate
was lower in both CSTRs than in the 125 mL anaerobic bottles.
The substrate concentration was the only different opera-tional parameter between the CSTRs and batch tests. Thus, it
was presumed that it also contributed to the higher produc-
tion rate and yield, as higher substrate concentration in-
creases fermentation rate. The shift from the ethanol to the
acetate pathway in the CSTRs explained the higher attained
hydrogen yields in the CSTRs.
Cheng and Liu [1] used C. thermocellum and achieved a
hydrogen yield of 1.2 mol H2 /mol hexose equivalent, at a pH of
7.0, and a temperature of 60 C. These results are consistent
with the 0.98e1.65 mol H2 /mol hexose equivalent observed by
Lauren et al. [34] employing a CSTR at neutral pH, and influent
of cellulose concentration in the range of 1.5e4 g/L.
In another study, Liang-Ming et al. [35] investigated thefermentative biohydrogen production in CSTRs using Clos-
tridium tyrobutyricum FYa102. Two CSTRs were employed in
this study: one, denoted (GP), was fed with 12 g/L of glucose
and 8 g/L of peptone, while the other, denoted (GA), was fed
with 12 g/L of glucose, 1.4 g/L of ammonium chloride, and
0.360 g/L of peptone. The experiments were carried out at a
temperature of 35 C, and a pH of 6.0. The hydrogen yield and
hydrogen production rate achieved for the (GA) and (GP) re-
actors were 1.42 mol H2 /mol glucose, and 3.1 L/L/d, and
1.06 mol H2 /mol glucose, and 10.3 L/L/d, respectively. On the
other hand, Pei-Ying et al. [31], conducted biochemical
hydrogen potential (BHP) tests to investigate the metabolism
of glucose fermentation and hydrogen production perfor-mance of C. tyrobutyricum FYa102 in batches. Glucose and
peptone were used in the fermentation medium at initial
concentrations of 3 g/L, and 2 g/L, respectively. The experi-
ment was conducted at a temperature of 35 C and a pH of 7.2.
The attained hydrogen yield and hydrogen production rates
were 1.47 mol H2 /mol glucose, and 1.6 L/L/d, respectively. In
glucose re-feeding experiments, the C. tyrobutyricum FYa102
fermented additional glucose during re-feeding tests, pro-
ducing a substantial quantity of hydrogen. The higher
hydrogen production rates attained in the CSTRs were
attributed to the higher rate of fermentation resulting from
the higher concentrations of glucose and peptone in the me-
dium, and the better mixing conditions in the CSTRs.
Although the CSTRs were fed with a much higher concentra-
tion of glucose, the hydrogen yields were almost the same or a
bit less than those attained in the batch studies. This differ-
ence in hydrogen yields was attributed to the different
ambient pHs in the batches and the CSTRs experiments. It is
believed that if the CSTRs experiments were conducted at a
pH of 7.2, higher hydrogen yields would have been attained. It
must be noted, however, that in the CSTRs study [35], thehydrogen production rate in the (GP) reactor was 3.5 times
higher than the (GA) due to a much higher organic loading
rate, but the yield of the (GP) was only 75% of that of the (GA)
due to a lower glucose fraction.
Ji et al. [36] immobilized the hydrogen producing anaerobe,
C. tyrobutyricum JM1 in a packed-bed reactor using poly-
urethane foam as support media. The hydraulic retention
time (HRT) condition for maximum hydrogen production rate
in this system was 2 h, where the main metabolite was
butyrate with low lactate concentration, andhydrogen yield of
1.81 mol H2 /mol glucose was attained at a pH of 6.7, temper-
ature of 37 C, and a feed glucose concentration of 5 g/L.
Therefore, the immobilized system was an effective and sta-ble approach for continuous hydrogen production for efficient
utilization of carbon substrates with good hydrogen-
producing performance. However, in a later study by the
same group [37], the effects of pH on hydrogen fermentation
of glucose by the same bacterium were investigated in batch
cultivations. The batcheswere conducted at different pHs (6.0,
6.3, 6.7), temperature of 37 C, and a glucose concentration of
20 g/L. The initial low glucose concentration (such as the 5 g/L
of glucose used in the previous study [36]) resulted in a low
fermentation rate, and consequently a low hydrogen yield. It
was proven that a pH of 6.3 was optimum for hydrogen pro-
duction with a high concentration of butyrate, and a hydrogen
yield of 3.24 mol H2 /mol glucose. The lower hydrogen yieldachieved in the continuous-flow system was attributed to the
un-optimized pH and substrate concentration.
Defeng et al. [38] studied hydrogen production of auto-
aggregative (self-flocculating granular) E. harbinense YUAN-3
in a batch reactor and a continuous stirred-tank reactor
(CSTR), with glucose as substrate under non-sterile condi-
tions. In the batch reactor, the optimized operational condi-
tions constituteda pH of 5.0, temperature of 35 C, and glucose
concentration of 10 g/L. The maximum hydrogen yield and
hydrogen production rate under the optimum operational
conditions were 1.91 mol H2 /mol glucose and 1.66 L/L/d,
respectively. In the CSTR, hydrogen gas yield reached a
maximum of 1.93 mol H2 /mol glucose, and H2 production ratereached a maximum of 19.6 L/L/d.The strain YUAN-3 was well
retained in thereactor. The overflow rate of cells was less than
0.1% in the continuous flow reactor, at a dilution rate of 0.5/h.
However, after 7 days of continuous operation some other
hydrogen-producing bacterial species appeared and formed a
stable community with YUAN-3. The hydrogen yield
decreased from 0.93 mol H2 /mol glucose to 1.5 mol H2 /mol
glucose and stabilized thereafter. The dominant populations
in the continuous-flow reactor were affiliated with M. hominis,
and M. sueciensis, and the majority of dominant populations
belonged to E. harbinense, which were enriched during opera-
tion of the reactor. These results indicate that a successful
continuous operation was achieved. It is evident from the
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 4 9 4 5 e4 9 6 64958
http://dx.doi.org/10.1016/j.ijhydene.2013.02.032http://dx.doi.org/10.1016/j.ijhydene.2013.02.032http://dx.doi.org/10.1016/j.ijhydene.2013.02.032http://dx.doi.org/10.1016/j.ijhydene.2013.02.032
8/17/2019 Review Paper on Biohydrogen
15/22
above results that optimizing the operational parameters for
the auto-aggregative bacteria achieved continuous stable
hydrogen production, despite the occurrence of microbial
shift.
Tien et al. [30] investigated biohydrogen production from
xylose by T. neapolitana in batch culture using serum bottles
and a continuously stirred anaerobic bioreactor (CSABR). A
maximum hydrogen production rate of 0.44 L H2 /L/d and amaximum hydrogen yield of 1.31 mol H2 /mol hexose equiva-
lent were obtained in the serum bottles test, at an initial
xylose concentration of 5.0 g/L, and a pH of 7.5. The CSABR
was run at uncontrolled and controlled pH conditions. In the
uncontrolled pH experiments, the fermentation process
ceased before the complete consumption of substrate due to
the drastic decrease in pH. In pH-controlled cultures, much
higher H2 production and xylose utilization rates were ach-
ieved as evident from the levels of acetic acid varying from 2.5
to 3.5 g/L compared to 2.5 g/L in the uncontrolled batches. In
contrast to acetic acid production, lactic acid production was
the lowest under pH-controlled conditions. Subsequently, the
H2 production rate increased exponentially reaching themaximum level. The maximum H2 yield, and hydrogen pro-
ductionrateof 2.8 molH2 /mol xylose consumed, and 2.66 L H2 /
L/d were measured while the pH was maintained at 7.0. It
appears that controlling pH at neutral limit instead of an
initial pH of 7.5 was the key for a stable continuous system.
Another example of the effect of un-optimized operational
parameters on the performance of continuous-flow system
was provided by Sompong et al. [39]. They investigated the
fermentation of 20 g/L sucrose by the bacterium T. thermo-
saccharolyticum strain PSU-2 in an UASB bioreactor. The system
was stable, and the hydrogen yield and production rate of
1.77 mol H2 /mol hexose and 5.9 L H2 /L/d were achieved at a
temperature of 60 C and a pH of 5.5. However, the same au-thors investigated in another study [40] the fermentation of
the same concentration of sucrose under the same tempera-
ture, and a wide range of pH (4.0e9.0) by the same bacterium
and observed maximum hydrogen yield and hydrogen pro-
duction rate of 2.53 mol H2 /mol hexose equivalent, and 6.5 L/L/
d at a pH of 6.25. Therefore, it is presumed that if the contin-
uous system was operated at the optimum pH, a higher
hydrogen yield and production rate would have been realized.
5. Feedstocks
Hydrogen can be produced from a wide spectrum of carbo-hydrates. Nevertheless, 80% of the studies reported in the
surveyed literature have investigated hydrogen production by
dark fermentation from pure sugars, such as glucose, or su-
crose as substrate. Only a few studies have focussed on sus-
tainable substrate conversion (Fig. 1). However, for real value
to the society and environment, biohydrogen should be pro-
duced from renewable feedstocks (real waste) [41]. The po-
tential feedstocks include: biomass, agricultural waste bi-
products, lignocellulosic products (wood and wood waste),
waste from food processing, aquatic plants, algae, agricul-
tural, and livestock effluents. If used under appropriate con-
trol, these resources would become the major source of
energy in the future. In this study, different types of
feedstocks are discussed in terms of their applicability and
operational challenges, as well as the motivation for their use
in fermentative hydrogen production.
5.1. Pure carbohydrates (synthetic waste)
Pure carbohydrate sources are expensive raw materials for
real scale hydrogen production (which can only be viablewhen based on renewable and low cost sources). Neverthe-
less, the majority of the reviewed studies utilized pure car-
bohydrates as substrate, including: monosaccharides
(glucose, xylose, fructose, arabinose, mannose and ribose);
disaccharides (sucrose, cellubiose, maltose, and lactose); or
polysaccharides (starch, cellulose, and xylan).
Simple sugars such as glucose, sucrose, and lactose are the
most commonly used pure substrates due to their ease of
biodegradability, relatively simple structures, and presence in
real industrial effluents [42,43]. Unlike starch and cellulose,
they require short fermentation times (i.e. process HRT),
which makes these substrates preferred model substrates for
hydrogen production studies. Model substrates are employedin fermentative hydrogen production processes to study bac-
terial kinetics, assess adequate nutrients, and identify the
optimized operational parameters for the process [19,44e48].
However, real life applications involve complex sugars, and
thus it is indispensable to employ these substrates in the dark
fermentation process in order to provide relevant insight into
system performance [1,49].
Hydrolysis of real waste comprising different sugars was
modelled by co-digestion of various pure substrates. The
control experiments assessed the fermentation preferences of
the bacteria among the different types of sugars. Pan-
agiotopoulos et al. [50] conducted four experiments to eval-
uate the hydrogen production and the main organic acids by
Pure monosaccharides
59%
Pure polysaccharides
11%
Sustainable
feedstocks
(current+future)
20%
Fig. 1 e Percentage of the usage of pure and real waste
substrates in the reviewed literature.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 4 9 4 5 e4 9 6 6 4959
http://dx.doi.org/10.1016/j.ijhydene.2013.02.032http://dx.doi.org/10.1016/j.ijhydene.2013.02.032http://dx.doi.org/10.1016/j.ijhydene.2013.02.032http://dx.doi.org/10.1016/j.ijhydene.2013.02.032
8/17/2019 Review Paper on Biohydrogen
16/22
C. saccharolyticus from the hydrolysed sweet sugar bagasse
(SSB) and from a mixture of pure sugars (glucose, sucrose and
xylose). They employed SSB in 2 experiments at sugar con-
centrations of 10 g/L and 20 g/L, and conducted two control
experiments employing mixtures of pure sugars of glucose,
xylose, and sucrose, at concentrations of 10 g/L and 20 g/L. At a
sugar concentration of 10 g/L, consumption of pure sugars and
sugars of SSB hydrolysate was complete within similarfermentation time. At the 20 g/L of pure sugars, consumption
was still incomplete at 72 h, while sugars of SSB hydrolysate
were completely consumed at 70 h. The consumption pattern
of 20 g/L sugars of SSB hydrolysate sugars differed markedly
from that of pure sugars. Lactate production only occurred in
fermentations on SSB hydrolysate. Therefore, hydrogen pro-
duction and hydrogen yields were higher in fermentations on
pure sugars than of SSB hydrolysate. The high rate of glucose
consumption in the fermentation of 20 g/L of SSB hydrolysate
sugars coincided with the high rate of lactate production in
this fermentation. Basically, at sugar concentration of 10 g/L,
the batch fermentations under controlled conditions
confirmed the results of the fermentability tests, but at highersubstrate concentrations, lactate production increased
dramatically at the expense of hydrogen production.
Trus de Vrije et al. [51] investigated thermophilic hydrogen
production using C. saccharolyticus and T. neapolitana on hy-
drolysate of the lignocellulosic feedstock Miscanthus (obtained
from enzymatic hydrolysis) in batch tests. Control experi-
ments were also conducted at different mixing ratios of xylose
and glucose to assess the utilization preference of the bacteria
for a sugar type over the other and the optimum substrate
concentration. The authors observed that T. neapolitana
showed a preference for glucose over xylose, which were the
main sugars in the hydrolysate, while C. saccarolyticus
consumed both at a similar rate. Lactate production by C.saccarolyticus was very low in fermentations on pure sugars, as
well as on hydrolysate. T. neapolitana produced more lactate
on the hydrolysate than on pure sugars. The optimum total
sugars concentration was 17 g/L, and C. saccharolyticus offered
the advantage of nearly 10% higher hydrogen yield during
growth on Miscanthus hydrolysates as compared to T. neapo-
litana, but the rates of substrate consumption and hydrogen
production by T. neapolitana were 5e18% higher.
Trus de Vrije et al. [26] investigated hydrogen production
from carrot pulp hydrolysate (obtained from enzymatic hy-
drolysis) by the same thermophilic bacteria C. saccharolyticus
and T. neapolitana. The main sugars in the hydrolysate were
glucose, fructose, and sucrose. Therefore, they initiallyinvestigated hydrogen production from different concentra-
tions of glucose, fructose, and mixtures of glucose and fruc-
tose as control experiments. in order to assess the adequate
degree of hydrolysation and optimized substrate concentra-
tion by determining the preferred sugar type for bacteria. They
observed that in fermentations of 10 g/L glucose and 10 g/L
fructose, C. saccharolyticus could virtually completely consume
all substrates with almost identical rates of consumptions. In
contrast, T. neapolitana consumption trend was different for
glucose than fructose, suggesting a preference for glucose. In
fermentations of 20 g/L of substrate, the consumption of
substrate was incomplete for both cultures even after 2 days.
Also, they found that the cultures productivities were
equivalent or higher than the productivities achieved with the
corresponding pure sugars (mixtures of glucose and fructose)
at 10 g/L sugars. Doubling the hydrolysate concentration had
adverseeffect on hydrogen production, with a severe decrease
in yield in C. saccharolyticus cultures and a decrease in pro-
ductivity with T. neapolitana.
Nan-Qi Ren et al. [52] investigated the utilization of an agro-
waste, corn stover, as a renewable lignocellulosic feedstockforthe fermentative H2 production by the moderate thermophile
T. thermosaccharolyticum W16. The corn stover was hydrolysed
by cellulase with supplementation of xylanase after delignifi-
cation with 2% NaOH, producing glucose, xylose, and arabi-
nose. To determine the fermentative behaviour of the
bacterium, a set of control experiments supplemented with
glucose, xylose, and a mixture of glucose, xylose, and arabi-
nose at a fixed total sugar quantity of 10 g/Lwere undertaken.
The concentrations of glucose, xylose, and arabinose in the
mixture were at the same levels as found in the corn stover
hydrolysate. It was observed that the bacterium showed pref-
erence for glucose over the other types of sugars. The bacte-
rium grew well on the hydrolysate and reached a similaroptical density and maximum hydrogen production rate as on
simulated medium, although hydrogen yield was slightly
higher on hydrolysate. Although the molar carbon balances in
the control experiments closed at 100%, carbon balances did
not close in the hydrolysate, most probably due to the inter-
ference of unidentified components in the hydrolysate.
5.2. Sustainable feedstocks (real waste)
For sustainable biohydrogen production, the feedstock has to
be cheap and would have to meet the following criteria: car-
bohydrate produced from sustainable resources; sufficient
concentration that fermentative conversion and energy re-covery is energetically favourable; and minimum pretreat-
ment [53].
Biomass is a viable renewable resource. It includes agri-
cultural residues, energy crops, and industrial wastes, which
can be used for the production of power, heat and biofuels
[50]. Producing hydrogen from biomass greatly enhances the
security of supply [26]. Therefore, most recent studies
employed biomass for biological conversion via fermentation
processes. As shown in Table 4, sugar-containing crops
(e.g.sweet sorghum and sugar beet), starch-based crops (e.g.
corn and wheat), ligno-cellulosics (e.g. fodder grass and mis-
canthus), and food industry by-products are all biomass types
used as substrates in the literature [1,13,16,26,49e
52,54e
64].In view of the increasingly negative public reaction to the
use of food for biofuel production, employing energy crops as
feedstocks for biofuels generation (e.g. wheat straw, barely
straw, corn stalk, miscanthus, and cassava) is widely accepted
in the scientific community [1,51,54,55,61]. They are
commonly referred to as second generation cellulosic
biomass. Additionally, utilizing industrial and agricultural
waste residues (e.g. delignified wood fibres, and corn stalk
waste) [16,35,49,52,60], or food industry waste (e.g. carrot pulp,
potato steam peels, sugarcane waste, sweet sorghum syrup,
corn starch, and sweet potato starch) [13,26,50,56e59,63,64]
addresses the concerns of skyrocketing food and energy
prices.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 4 9 4 5 e4 9 6 64960
http://dx.doi.org/10.1016/j.ijhydene.2013.02.032http://dx.doi.org/10.1016/j.ijhydene.2013.02.032http://dx.doi.org/10.1016/j.ijhydene.2013.02.032http://dx.doi.org/10.1016/j.ijhydene.2013.02.032
8/17/2019 Review Paper on Biohydrogen
17/22
Table 4 e Operational and performance parameters of the studies utilizing sustainable feedstocks.
Substrate types Culture(s) Reactor type TC Substrateconcentration
(g/L)
pH Hydrogen yi(mol H2 /mol glu
hexose equiva
1-Current sustainable feedstocks:
Pretreated wheat straw Caldicellulosiruptor saccharolyticus Batch 70 10 7.2 3.8
Pretreated barely straw Caldicellulosiruptor saccharolyticus Batch 70 20h 7 ND
Carrot pulp hydrolysate Caldicellulosiruptor saccharolyticus Batch 70 10 7 2.8
PSPa Caldicellulosiruptor saccharolyticus Batch 72 10 7 3.5 PSP-H2b Caldicellulosiruptor saccharolyticus Batch 72 10 7 3.4
SSBc Caldicellulosiruptor saccharolyticus Batch 72 20h 6.8i 2.8
Miscanthus hydrolysate Caldicellulosiruptor saccharolyticus Batch 72 10 7 3.4
Miscanthus hydrolysate Caldicellulosiruptor saccharolyticus Batch 72 14 7 3.3
Miscanthus hydrolysate Caldicellulosiruptor saccharolyticus Batch 72 28 7 2.4
PSP-H2b Thermotoga neapolitana Batch 75 10 7 3.3
PSPa Thermotoga neapolitana Batch 75 10 7 3.8
Carrot pulp hydrolysate Thermotoga neapolitana Batch 75 10 7 2.7
Carrot pulp hydrolysate Thermotoga neapolitana Batch 75 20 7 2.4
Miscanthus hydrolysate Thermotoga neapolitana Batch 80 10 7 2.9
Miscanthus hydrolysate Thermotoga neapolitana Batch 80 14 7 3.2
Miscanthus hydrolysate Thermotoga neapolitana Batch 80 28 7 2
SCB hemicellulose
hydrolysatef C. butyricum Batch 37 20g 5.5 1.73e
Pretreated straw
hydrolysate
C. butyricum CGS5 Batch 37 9.2 7.5 0.91
Sugarcane juice C. butyricum TISTR 1032 Batch 37 22.3
(sucrose)
6.5 1.33
Sugarcane juice C. butyricum TISTR 1032
( immobilized)
Repeated
batch
37 22.3
(sucrose)
6.5 1.52
Molass C. butyricum W5 Batch 37 100 7 1.63
Sweet potato starch
residiue
C. butyricum and Enterobacter
aerogenes HO-39
Repeated
batch
37 ND 5.25i 2.7
Hydrolysed corn stover Thermoanaerobacterium
thermosaccharolyticum W16
Batch 60 ND 7 2.24e
Hydrolyzed corn stover Thermoanaerobacterium
thermosaccharolyticum W16
Batch 60 ND 7 ND
Cassava wastewater Clostridium acetobutylicum
ATCC 824
Batch 36 5g 7 2.41
Cron stalk Clostridium thermocellum 7072 CSTR, 100 L 55 30 7.4 0.45
Cron stalk Clostridium thermocellum 7072 CSTR, 10 L 55 30 7.4 0.43
Cron stalk Clostridium thermocellum 7072 Batch 55 5 7.4 ND
Delignefied wood fibres Clostridium thermocellum
ATCC27405
Batch 55 0.1 6.5 1.6
Corn stalk waste Clostridium thermocellum and
C. thermosaccharolyticum
Batch 55 10 7.2 ND
Corn stalk waste Clostridium thermocellum and
C. thermosaccharolyticum
CSTR 55 10 7.2 ND
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-
8/17/2019 Review Paper on Biohydrogen
18/22
Table 4 e ( continued )
Substrate types Culture(s) Reactor type TC Substrateconcentration
(g/L)
pH Hydrogen yi(mol H2 /mol glu
hexose equiva
Corn starch hydrolysate Enterobacter aerogenes NCIMB 10102 Continuous
packed col.
40 ND 5.5 2.55
Starch hydrolysate Enterobacter aerogenes
NCIMB 10102
Batch 40 20 6.5 1.09e
2-Future sustainablefeedstocks:
POMEd C. butyricum EB6 Batch 37 ND 5.5 0.22
Glycerol Klebsiella pneumoniae DSM2026 Batch 37 20 6.5 0.53
Glycerol Enterobacter aerogenes Batch 37 20 7 0.2
Glycorel Enterobacter aerogenes
ATCC35029
Batch 37 21 ND 1.22
Chlorella vulgaris ESP6
(microalgal hydrolysate)
C. butyricum CGS5 Batch 37 9 5.5i ND
ND: Not defined.
a Untreated potato steam peels, Molar yields were based on the amount of starch in untreated PSP assuming 100% starch consumption.
b The starch in the PSP was liquefied with alpha-amylase, and then the liquefied starch was further hydrolyzed to glucose by amyloglucosidase.
c Sweet sorgham bagasse.
d Palm oil mill effluent.
e mol H2 /mol total sugar.
f Sugarcane bagasse hemicellulose hydrolysate.
g g COD/L.
h g sugars/L.
i Controlled pH.
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-
8/17/2019 Review Paper on Biohydrogen
19/22
The reported hydrogen yields from biomass used as sub-
strates varied greatly from approximately 20% to more than
90% of the theoretical 4 moles of H2 per mol of hexose. The
diversity of the applied feedstocks and pretreatment methods
hardly allow a comparison of hydrogen production efficiency.
5.3. Future feedstocks
Based on the reviewed literature, It is evident that the current
focus is primarily on food, agriculture, and industry-related
substances to provide sustainable feedstocks. The bio-
hydrogen technology would have a rather limited scope if the
feedstock range and sources cannot be expanded intensively
[41]. Furthermore, some wastewaters have promising poten-
tial for biohydrogen production process (Table 4), including:
oil industry wastewaters [65]; and biodiesel wastes containing
glycerol [66e68]. In addition, microalgal biomass, which is
produced by CO2 fixation through photosynthesis of micro-
algae, was proven to be a good sustainable feedstock [69].
Although these industrial by-products are considered prom-
ising approaches for sustainable biohydrogen production, theyields reported from utilizing these wastewaters were still low
compared to the traditional sustainable feed stocks. Further
research in utilizing these substrates via dark fermentation,
and the adequate pretreatments methods is required.
6. Concluding remarks
Based on the findings of this literature review, the following
remarks can be drawn:
Attaining technical and economic efficiencies is the main
drive behind employing co-cultures of pure bacteria infermentative hydrogen production.
There are three types of co-cultures of pure isolates
a) Co-cultures of strict high hydrogen producers and
facultative anaerobes, which is used to attain anaer-
obic conditions without the need to add expensive
reducing agents. These co-cultures yield better per-
formance parameters, especially for complex sugars
substrates.
b) Co-cultures of cellulose-degrading anaerobes and high
hydrogen producers capable of producing hydrogen
from simpler forms of sugars. These co-cultures offer
economical and technical advantages over cellulose
degrading anaerobe solely or enzymatically hydro-lysed cellulose. They produce hydrogen in two steps;
cellulose degrading anaerobe via the initial degrada-
tion followed by high hydrogen-producing anaerobe
from the degraded sugars.
c) Co-cultures of aciduric microorganisms and hydrogen
producers which reduces alkali consumption, hence
reducing or eliminating the need for a buffer to
maintain a neutral or weak acidic pH.
The perceived advantages of thermophiles over mesophiles
appear to be substrate-dependant:
a) For simple sugars (monosaccharides and di-
saccharides), either mesophiles or thermophiles could
produce more hydrogen from fermenting simple
sugars depending on bacterial kinetics, and the sub-
strate type.
b) For complex sugars, thermophiles outperform meso-
philes in terms of hydrogen production due to their
ability to degrade complex substrates, in addition to
the increased hydrolysis and fermentation rates
associated with the high operating temperature.
It is essential to first determine optimal operational condi-tions batch studies. Continuous systems can then be oper-
ated under these optimal operational conditions to achieve
a sustainable system with same or better performance pa-
rameters than its batch counterpart.
Biodiesel wastes, oil industry wastewaters, and microalgal
biomass have significant potential as sustainable feed
stocks in the near future.
Certain aspects of biohydrogen production merit further
research, including:
a) Employing co-cultures of high hydrogen producers of
facultative and strict anaerobes with different opti-
mum pH ranges.
b) Use of co-cultures of complex sugars degrading an-aerobes and high hydrogen producers.
c) Enhancement of the hydrogen yields and hydrogen
production rates from dark fermentation of emerging
sustainable feedstocks.
Acknowledgement
The authors acknowledge NSERC, GreenField Ethanol, Union
Gas, and Admira Energy for their financial support of the
project, as well as the Ontario Trillium Ph.D. Scholarship
Program awarded to Ms. Omneya Elsharnouby.
r e f e r e n c e s
[1] Cheng X, Liu C. Hydrogen production via thermophilicfermentation of cornstalk by Clostridium thermocellum.Energy Fuels 2011;25:1714e20.
[2] Das D, Veziroglu TN. Advances in biological hydrogenproduction processes. Int J Hydrog Energy 2008;33:6046e57.
[3] Nishio N, Nakashimada Y. High rate production of hydrogen/methane from various substrates and wastes.Adv Biochem Eng Biotechnol 2004;90:63e87.
[4] Wang J, Wan W. Factors influencing fermentative hydrogenproduction: a review. Int J Hydrog Energy 2009;34:799e811.
[5] Hafez H, Baghchehsaraee B, Nakhla G, Karamanev D,Margaritis A, El Naggar H. Comparative assessment of decoupling of biomass and hydraulic retention times inhydrogen production bioreactors. Int J Hydrog Energy 2009;34:7603e11.
[6] Hafez H, Nakhla G, El Naggar H. An integrated system forhydrogen and methane production during landfill leachatetreatment. Int J Hydrog Energy 2010;35:5010e4.
[7] Ginkel S, Sung S, Lay J. Biohydrogen production as afunction of pH and substrate concentration. Environ SciTechnol 2001;35:4726e30.
[8] Hafez H, Nakhla G, El Naggar H. Biological hydrogenproduction. In: Sherif SA, editor. Handbook of hydrogen
energy. Boca Raton: CRC Press, Taylor and Francis; 2012.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 4 9 4 5 e4 9 6 6 4963
http://dx.doi.org/10.1016/j.ijhydene.2013.02.032http://dx.doi.org/10.1016/j.ijhydene.2013.02.032http://dx.doi.org/10.1016/j.ijhydene.2013.02.032http://dx.doi.org/10.1016/j.ijhydene.2013.02.032
8/17/2019 Review Paper on Biohydrogen
20/22
[9] Lee D, Show K, Su A. Dark fermentation on biohydrogenproduction: pure culture. Bioresour Technol 2011;102:8393e402.
[10] Carere CR, Sparling R, Cicek N, Levin DB. Third generationbiofuels via direct cellulose fermentation. Int J Mol Sci 2008;9:1342e60.
[11] Seppa ¨ la ¨ JJ, Puhakka JA, Yli-Harja O, Karp MT, Santala V.Fermentative hydrogen production by Clostridium butyricum
and Escherichia coli in pure and cocultures. Int J Hydrog Energy 2011;36:10701e8.
[12] Yokoi H, Tokushige T, Hirose J, Hayashi S, Takasaki Y. H2production from starch by a mixed culture of Clostridiumbutyricum and Enterobacter aerogenes. Biotechnol Lett 1998;20:143e7.
[13] Yokoi H, Maki R, Hirose J, Hayashi S. Microbial production of hydrogen from starch-manufacturing wastes. BiomassBioenerg 2002;22:389e95.
[14] Wang A, Ren N, Shi Y, Lee D. Bioaugmented hydrogenproduction from microcrystalline cellulose using co-cultureedClostridium acetobutylicum and Ethanoigenensharbinense. Int J Hydrog Energy 2008;33:912e7.
[15] Liu Y, Yu P, Song X, Qu Y. Hydrogen production fromcellulose by co-culture of Clostridium thermocellum JN4 and
Thermoanaerobacterium thermosaccharolyticum GD17. Int JHydrog Energy 2008;33:2927e33.
[16] Li Q, Liu C. Co-culture of Clostridium thermocellum andClostridium thermosaccharolyticum for enhancing hydrogenproduction via thermophilic fermentation of cornstalkwaste. Int J Hydrog Energy 2012;37:10648e54.
[17] Geng A, He Y, Qian C, Yan X, Zhou Z. Effect of key factors onhydrogen production from cellulose in a co-culture of Clostridium thermocellum and Clostridium thermopalmarium.Bioresour Technol 2010;101:4029e33.
[18] LuW,WenJ,ChenY,SunB,JiaX,LiuM,etal.Synergisticeffectof Candida maltosa HY-35 and Enterobacter aerogenes W-23 onhydrogen production. Int J Hydrog Energy 2007;32:1059e66.
[19] Kamalaskar LB, Dhakephalkar PK, Meher KK, Ranade DR.High biohydrogen yielding Clostridium sp. DMHC-10
isolated from sludge of distillery waste treatment plant. Int JHydrog Energy 2010;35:10639e44.
[20] Nath K, Kumar A, Das D. Effect of some environmentalparameters on fermentative hydrogen production byEnterobacter cloacae DM11. Can J Microbiol 2006;56:525.
[21] Sinha P, Pandey A. An evaluative report and challenges forfermentative biohydrogen production. Int J Hydrog Energy2011;36:7460e78.
[22] Munro SA, Zinder SH, Walker LP. The fermentationstoichiometry of Thermotoga neapolitana and influence of temperature, oxygen, and pH on hydrogen production.Biotechnol Prog 2009;25:1035e42.
[23] d’Ippolito G, Dipasquale L, Vella FM, Romano I,Gambacorta A, Cutignano A, et al. Hydrogen metabolism inthe extreme thermophile Thermotoga neapolitana. Int J
Hydrog Energy 2010;35:2290e5.[24] Zhang T, Liu H, Fang HHP. Biohydrogen production from
starch in wastewater under thermophilic condition. J Environ Manage 2003;69:149e56.
[25] Zhu D, Wang G, Qiao H, Cai J. Fermentative hydrogenproduction by the new marine Pantoea agglomerans isolatedfrom the mangrove sludge. Int J Hydrog Energy 2008;33:6116e23.
[26] de Vrije T, Budde MAW, Lips SJ, Bakker RR, Mars AE,Claassen PAM. Hydrogen production from carrot pulp bythe extreme thermophiles Caldicellulosiruptor saccharolyticusand Thermotoga neapolitana. Int J Hydrog Energy 2010;35:13206e13.
[27] Ghosh D, Hallenbeck PC. Fermentative hydrogen yieldsfrom different sugars by batch cultures of metabolically
engineered Escherichia coli DJT135. Int J Hydrog Energy 2009;34:7979e82.
[28] Niela E, Buddeb M, Haasb G, Walb F, Claassenb P, Stamsa A.Distinctive properties of high hydrogen producing extremethermophiles, Caldicellulosiruptor saccharolyticus andThermotaga elfii. Int J Hydrog Energy 2002;27:1391e8.
[29] Kumar N, Das D. Enhancement of hydrogen production byEnterobacter cloacae IIT-BT 08. Process Biochem 2000;35:
589e93.[30] Ngo TA, Nguyen TH, Bui HTV. Thermophilic fermentative
hydrogen production from xylose by Thermotoga neapolitanaDSM 4359. Renew Energy 2012;37:174e9.
[31] Lin P, Whang L, Wu Y, Ren W, Hsiao C, Li S, et al. Biologicalhydrogen production of the genus Clostridium: metabolicstudy and mathematical model simulation. Int J Hydrog Energy 2007;32:1728e35.
[32] Islam R, Cicek N, Sparling R, Levin D. Influence of initialcellulose concentration on the carbon flow distributionduring batch fermentation by Clostridium thermocellum ATCC27405. Appl Microbiol Biotechnol 2009;82:141e8.
[33] Guo XM, Trably E, Latrille E, Carrère H, Steyer J. Hydrogenproduction from agricultural waste by dark fermentation: areview. Int J Hydrog Energy 2010;35:10660e73.
[34] Magnusson L, Cicek N, Sparling R, Levin D. Continuoushydrogen production during fermentation of a-cellulose bythe thermophillic bacterium Clostridium thermocellum.Biotechnol Bioeng 2009;102:759e66. 2008;102:759-66.
[35] Whang L, Lin C, Liu I, Wu C, Cheng H. Metabolic andenergetic aspects of biohydrogen production of Clostridium tyrobutyricum: the effects of hydraulicretention time and peptone addition. Bioresour Technol2011;102:8378e83.
[36] Jo JH, Lee DS, Park D, Park JM. Biological hydrogenproduction by immobilized cells of Clostridium tyrobutyricum
JM1 isolated from a food waste treatment process. BioresourTechnol 2008;99:6666e72.
[37] Jo JH, Lee DS, Park JM. The effects of pH on carbon materialand energy balances in hydrogen-pr