IEA-HIA Annex21
BioHydrogen Production
Annex 21
Athen, Nov.05-07, 2008
OA: Jun Miyake, AIST/U. Tokyo
BioHydrogen: Concept
BioHydrogen In the World
U.K.
Canada
U.S.A.
Cuba
Denmark
Norway
Netherlands
Sweden
Hungary
Germany
China
Korea
Japan
Poland
Taiwan
Israel
France
Thailand
Switzerland
Portugal
Russia
Italy
Turkey
Bulgaria
India
Malaysia
Singapore
H2 Price Evaluation
Year Org. Source
Methods
1996 NHA Natural gas
Steam refoming
Crude oil
Pertial oxidation
Coal
Gasification
Biomass
Gasification
Hydroelectric Water electrolysis
Wind power Water electrolysis
2010 IPCC Biomass
Gasification
Coal
Gasification
Water power Water electrolysis
Natural gas
Steam refoming
Nuclear power Water electrolysis
Sunlight
Water electrolysis
Price ($/GJ)
5.32
9.57
10.65
12.73
26.62
31.95
17.59
20.45
24.69
26.01
31.33
33.18
120 Yen/$
Annex21 Members
Annex21
Members
Canada
Finland
France
Germany
Japan
Korea
Netherlands
Norway
Sweden
UK
USA
Potential
HIA
Members
Singapore
Turkey
Russia
Taiwan
Hungary
Latvia
Thailand
New Zealand
India
China
Portugal
Annex21 4 Subtasks

A. Biohydrogen Systems.
Goal: increase achievable H2 production from substrates above
currently achievable yields (e.g. 3 to 4 moles H2/ mole of glucose).
Activity Leader: Netherlands

B. Basic Studies for Photobiological Hydrogen Production.
Goal: demonstrate potentially practical processes for conversion of
water or organic substrates to H2 with solar energy.
Activity Leader: United States of America

C. Bio-inspired Systems.
Goal: identify promising applications of enzymes and biologicallyinspired processes for hydrogen production and fuel cells
Activity Leader: Sweden/France

D. Overall Analysis.
Goal: to provide independent evaluations and analyses for the
various technologies considered by the Annex.
Activity Leader: Japan
Subtask A
Dark fermentation
Continuous H2 Production from
soybean-curd (TOFU) processing waste
Using anaerobic mixed microflora at 60℃
Maximal H2 production rate 7 L H2/L/day
Maximal yield 1.6 mol H /mol glucose
Hollow-fiber microfiltration membrane
module improved H2 production rate
showing the ranges of 0.9-1.3 m3/day
in a 100 L scaled system
Subtask A
Genetic research for photosynthetic
bacteria
Genetic study of Rhodobacter sphaeroides
KD131, isolated from the bay area in Korea
Genetic approach to improve H2 production
Circle view of Rhodobactor sphaeroides KD 131
Mainly consists of an uptake-hydrogenase
and two nitrogenases gene clusters that
are identified by genome analysis of a biohydrogen producer, R. sphaeroides KD131.
Subtask A
Cellulolytic, anaerobic bacterium
Dr. Levin
C. thermocellum possesses the highest
rate of cellulose-degradation of all
known microorganisms
Clostridium termitidis
Clostridium thermocellum
Subtask A
Microbial electrolysis cell (MEC)
Membrane-less continuous flow microbial electrolysis cell (MEC) with a gasphase cathode. This MEC had the highest volumetric rate of hydrogen reported
or this type of device, 6.3 L/L/day.
Subtask A
Dr. Lalman’s group (University of Windsor)
Supplement with linoleic acid improved hydrogen production
from glucose using a mixed anaerobic culture toward 2.4 mol H2/ mol glucose
, while yields were lower at pH 7.6 (1.4 mol H2/ mol glucose).
Subtask A
Dr. Nakhla’s group (University of Western
Ontario)
The maximum yields of 2.3 and 1.6 mol H2/mol glucose by resulting microflora
were achieved for the 65 degrees C pretreated anaerobically digested and
activated sludges, respectively. DNA analysis of the microbial community showed
that the elevated pretreatment temperatures reduced the species diversity.
Subtask A

Dr. Beland’s group (Environment Canada) examined hydrogen production from food
waste (FW), primary sludge (PS) and waste activated sludge (WAS). FW was found to
have low pH buffering capacity while the values for PS and WAS were relatively higher.
All combinations of the feedstocks (FW+PS, FW+WAS and FW+PS+WAS) showed
enhanced hydrogen production potential as compared with the individual wastes. A
mixing ratio of 1:1 was found to be the best among the ratios tested for all three co digestion groups. A hydrogen yield of 112 mL/g volatile solid (VS) added was obtained
from a combination of FW, PS and WAS.
Subtask A
the Maness group at NREL has been studying the fermentation of dilute-acid,
pretreated corn stover lignocellulose by the thermophilic cellulose-degrader
Clostridium thermocellum 27405. Yields and rates of H2 production depend
on carbon loadings, with higher carbon substrate loading yielding faster
rates of H2 production but with lower molar yields of H2. Nevertheless, this is
the first demonstration that corn stover lignocellulose can serve as a
feedstock in support of H2 production. Using cellobiose as the substrate, the
group also observed an increase in total amount of H2 when various
metabolic pathways are blocked using chemical inhibitors. These findings
provide the proof of concept that metabolic pathway engineering is a
promising approach in directing more cellular flux toward H2 production
provided a genetic system can be developed for this microbe.
Subtask A
The fermentative consortium at NREL (Maness, Ghirardi, and
Seibert) was also shown to metabolize algal starch, lipid and
protein at a non-optimal yield of 0.45 mg H2/mg biomass; is
able to utilize algal biomass isolated from a variety of growth
conditions, both fresh and frozen (damaged), without the
need for pre-treatment; and in collaboration with the
Tsygankov laboratory (Institute of Basic Biological Problems,
Russian Academy of Sciences, Pushchino, Russia), the
integration of fermentative with photosynthetic bacterial H2
production was demonstrated using potato wastes as the
initial substrate for fermentation.
Subtask A
In Netherlands, the model hydrogen-producing extreme thermophilic
Caldicellulosiruptor saccharolyticus was studied in detail.
By using 13C-NMR the occurrence of the Embden-Meyerhof (EM)
pathway in the extreme thermophile C. saccharolyticus was shown.
Glucose fermentation via the EM pathway to acetate results in a
theoretical yield of 4 mol of hydrogen and 2 mol of acetate per mol of glucose.
The measured yields were dependent on the growth rate.
The highest hydrogen yields of 82 to 90% of theoretical maximum
(3.3 to 3.6 mol H2 per mol glucose) were obtained at low growth rates.
whole genome of C. saccharolyticus became available.
Subtask A
The formate metabolism of a Moorella strain and a Desulfovibrio strain
was studied. It was found that these strains were able to grow by the
hydrolysis of formate to hydrogen and bicarbonate. However, these
strains were only able to grow when the hydrogen was efficiently removed
by methanogens. This led to syntrophic growth.
Subtask A
In the UK, the global flour industry produces 96 million ton/year of wheatfeed, which is
mainly used for animal feed. This co-product is high in carbohydrates and potentially a
significant substrate for biohydrogen production. A 10 l bioreactor, inoculated with
sewage sludge, was operated on wheatfeed (10 g l−1) at pH 5.5 and 35 °C in batch
and semi-continuous mode (15 h hydraulic retention time (HRT)).
Wheatfeed hydrolysate was also investigated in continuous mode (15 h HRT).
NaOH–H2O2 treatment of 25 g l−1 wheatfeed resulted in hydrolysate containing
on average 8.1 g l−1 total sugar. Hydrogen yields of 64 and 56 m3 H2 per ton dry
weight were produced from wheatfeed in batch and 56 m3 H2 per ton dry weight
of wheatfeed in semi-continuous mode. Hydrogen yields from hydrolysate were only
22 and 31 m3 H2 per ton dry weight, (or 0.9 mol H2 per mol hexose degraded,
assuming all sugar is hexose). Fermentation of unhydrolysed wheatfeed is therefore
recommended. It is calculated that approximately 264 m3/ton of CH4 can be produced
from a subsequent anaerobic digestion stage. The biohydrogen produced
(diesel equivalents) from the 1.2 million ton/year of wheatfeed in the UK would be
more than twice that required for transportation by the UK flour industry.
(Hawkes et al. (2008) Bioresource Technol. 99 5020–5029. )
Subtask A
Biohydrogen from olive mill waste water:
Hydrogen production from olive mill waste
water by Rhodobacter sphaeroides was
achieved in a two-step process involving
clay pretreatment and photofermentation.
Compared to the photofermentation using
raw olive mill waste water (14 lH2/lOMW)
(Eroglu et al., 2008a), the amount of
photobiological hydrogen production was
doubled by using the effluent of a clay
pretreatment process (31.5 lH2/lOMW).
Subtask A
Biohydrogen from waste ground wheat:
The objective of this study was the
optimization of biohydrogen production
from ground waste wheat solution by dark
and photofermentation processes using
mesophilic dark fermentative bacteria
(heat-treated, anaerobic sludge) and
photosynthetic bacteria (Rhodobacter
species). In dark fermentations, maximum
hydrogen yield (1.53 molH2/mol glucose)
and the specific rate (98 mlH2/ (g
biomass.h)) were obtained when the C/N
and C/P ratios were 200 and 1000,
respectively. (Argun et al., 2008a; Argun et
al., 2008b; Oztekin et al., 2008).
Subtask A

Biohydrogen from dark fermenter effluents: Biohydrogen production by
Rhodobacter capsulatus on thermophilic dark fermenter effluent of
miscantus hydrolysate in indoor batch-operation was investigated (Uyar
et al., 2008). The pretreatment of dark fermenter effluent with iron and
buffer addition resulted in a drastic increase (from 0.3 l H2/lc) to 1.0
lH2/lc) in H2 production. Photobiological hydrogen production on dark
fermenter effluents of molasses and potato steam peels were
investigated with Rhodobacter capsulatus wild type and hup deleted
mutant strains. The maximum productivity achieved was 2.73 g H2/(m3.h)
on molasses effluent and 1.1 gH2/(m3.h) on potato steam peels effluent
adjusted with Fe, Mo and buffer in indoor batch-operations.
Subtask A

Genetics and Strain Improvement: Genetically improved
strains of Rhodobacter capsulatus and Rhodobacter
sphaeroides were carried out by deletion of the uptake
hydrogenase gene. A 70% increase with hup- strain of R.
capsulatus (Ozturk et al., 2006) and 20% increase with hupstrain of R. sphaeroides in hydrogen production were
observed (Kars et al., 2008).
Subtask A


Physiology and Biochemistry of PNS Bacteria
The effect of light intensity, wavelength and different darklight cycle protocols and different initial dark periods on
growth and hydrogen production of Rhodobacter
sphaeroides were examined in indoor batch operations
(Uyar et al., 2007). The rate of hydrogen production
increased with increased light intensity and reached
saturation at around 270 W/m2. The effect of fluctuating
temperature in outdoor conditions on photobiological
hydrogen production was investigated by Rhodobacter
capsulatus (Özgür et al., 2008). Hydrogen production
decreased by 50% if the outdoor temperature fluctuates
between 15°C and 40°C.
Subtask A


Process Development
Investigation of performance of the flat plate solar
bioreactor (8 l) operating in outdoor conditions with
different carbon sources and olive mill waste water was
studied by Rhodobacter sphaeroides O.U.001 (Eroglu et
al., 2008b). Hydrogen production rate was the highest
(0.01 lH2/(lc.h)) when malate was the carbon source.
Large scale panel (40 L) and tubular (80 L)
photobioreactors have been constructed and
continuous hydrogen production have been achieved by
different strains of R.capsulatus on acetate.
Subtask B

The unicellular green alga Chlamydomonas reinhardtii possesses a [FeFe]-hydrogenase
HydA1 (EC 1.12.7.2), which is coupled to the photosynthetic electron transport chain.
Large amounts of H2 are produced in a light-dependent reaction for several days when
C. reinhardtii cells are deprived of sulfur. The analysis of the H 2 metabolism of Sdepleted C. reinhardtii cultures showed that electrons for H 2-production are provided
both by PSII activity and by a non-photochemical plastoquinone reduction pathway,
which is dependent on previous PSII activity. A kinetic model has been developed to
relate culture evolution from standard photosynthetic growth to H 2 producing cells.
Subtask B

Chlamydomonas reinhardtii H2 production, narrowly linked to the
photosynthetic process, results from complex metabolic reactions highly
dependent on the environmental conditions of the cells. A kinetic model has
been developed to relate culture evolution from standard photosynthetic
growth to H2 producing cells (Fouchard et al (2009) Biotechnol. Bioeng. 102:
232-245). It represents transition in sulfur-deprived conditions and the two
main processes then induced which are an over-accumulation of intracellular
starch and a progressive reduction of PSII activity for anoxia achievement.
Because these phenomena are directly linked to the photosynthetic growth,
two kinetic models were associated, the first (one) introducing light
dependency (Haldane type model associated to a radiative light transfer
model), the second (one) making growth a function of available sulfur amount
under extracellular and intracellular forms (Droop formulation). The model
parameters identification was realized from experimental data obtained with
especially designed experiments and a sensitivity analysis of the model to its
parameters was also conducted. Model behavior was finally studied showing
interdependency between light transfer conditions, photosynthetic growth,
sulfate uptake, photosynthetic activity and O2 release, during transition from
oxygenic growth to anoxic H2 production conditions.
Subtask B

In the USA, the S-systems analysis formalism was used to
model the metabolic pathways involved in H2 production by
sulfur-deprived algae, and a sensitivity analysis was performed
(Jorquera et al., 2008, Int. J. Hydrogen Energy 33:2167-2177).
The mathematical model successfully described the literature
data with respect to O2 and H2 evolution rates in response to
sulfur deprivation and to the re-addition of sulfate to the
system. The sensitivity analyses identified targets for future
protein engineering aimed at increasing the rates of H2
production by sulfur-deprived algae.
Subtask B

Green Algea, seven strains found positive for H 2 production
under anaerobic conditions were also screened for the
ability to produce H2 under S deprivation.
Subtask B

Cyanobactera, in the unicellular cyanobacterium
Synechocystis PCC 6803 the pentameric bidirectional
NiFe-hydrogenase (HoxEFUYH, encoded by the hox genes)
is the sole enzyme involved in the H2 metabolism. Novel
and unique finding already published. This may indicate a
function related to the assembly of a functional uptake
hydrogenase, hypothetically in the assembly of the small
subunit of the enzyme. The level of analysis provided by
this investigation presents various tools and knowledge that
are important for future development of cyanobacterial
biohydrogen production.
Subtask C


Bio-inspired Systems:
The synthesis and characterisation of the pro-ligand complexes are interesting
model of the active site of [NiFe] hydrogenase, indicating the imidazole
functions has a strong effect on the electronic structure of the complexes.
Dissymetrically disubstituted di-iron azadithiolate complexes protonate
exclusively at the N atom of the bridge, like the hexacarbonyl precursor but in
contrast to symmetrically disubstituted analogues; substitution of dppe for two
CO groups noticeably increases the kinetics of the electrocatalytic proton
reduction process. Electrochemical investigations on a structural analogue of
the [2Fe](H) subsite of [FeFe]H(2)ases were conducted in MeCN/NBu(4)PF(6).
The most efficient molecular photocatalytic systems reported so far for
hydrogen production could compete with some platinum-based systems.
Quantum yield values up to 16% under visible irradiation.
Subtask C


Bio-inspired Systems:
The extreme thermophile Pyrococcus furiosus hydrogenase
was proposed to be a new type of H2 evolution
hydrogenase, solubule and membrane-bound ones.
Subtask C


In Canada, Dr. Hallenbeck’s group has been looking at the
heterologous expression of Desulfovibrio vulgaris
Hildenborough. The highest yields (at, or somewhat higher
than 2 mol H2/mol glucose) were obtained with cultures
limited for glucose.
Bio-hybrid/ biofuel cell efforts are investigating
hydrogenases as biocatalysts for creating hybrid chargetransfer complexes with nanomaterials, and as catalysts in
photoelectrochemical devices.
Subtask C

In the USA, bio-hybrid/ biofuel cell efforts are investigating
hydrogenases as biocatalysts for creating hybrid chargetransfer complexes with nanomaterials, and as catalysts in
photoelectrochemical devices.
Subtask C


Biofuel cells:
In France, modification of gold and graphite electrodes with
commercially available carbon nanotubes is a useful
technique for immobilization of Desulfovibrio fructosovorans
[NiFe] hydrogenase, for hydrogen evolution or consumption
(Lojou et al. (2008) J Biol Inorg Chem. 13:1157-67).
Multiwalled carbon nanotubes, single-walled carbon
nanotubes (SWCNs), and amine-modified and carboxylfunctionalized SWCNs were used and compared.
Subtask C

In korea, Bioelectrocatalytic hydrogen (H2) production was
studied using Thiocapsa roseopersicina hydrogenase in a
two-compartment proton-exchange-membrane (PEM) fuelcell system equipped with carbon-paper electrodes. Sodium
dithionite(SD), as an electron donor, and hydrogenase, as a
catalyst, were used in the anodic oxidation reaction and in
the cathodic reduction reaction, respectively.
Subtask C

In Netherlands, the interaction between bacteria and
electricity is being studied. Bacterial processes can be
used to produce electricity in biofuel cells. However,
microbial processes can also be stimulated by electricity,
e.g. for hydrogen production from acetate in the dark.
Subtask C

In the USA, the Maness group at NREL has collaborated
with Dr. Bruce Logan (Penn State Univ.) in testing a novel
concept, combining dark fermentation with a microbial
electrolysis cell (MEC) to improve H2 molar yield. In their
work, the organic acids and ethanol present in the
fermentation effluent from either cellobiose or corn stover
lignocellulose fermentors were successfully converted to H2
via the MEC. The integrated process resulted in a combined
H2 molar yield that substantially exceeded the biological
maximum, which is 4 mol H2 per mol hexose.
Subtask C

In the UK, a hydrogen-producing Bio-Fuel Cell was
developed. Ammonia losses during swine wastewater
treatment were examined using single- and two-chambered
microbial fuel cells. Ammonia removal was 60% over 5 days
for a single-chamber MFC with the cathode exposed to air
(air.cathode), versus 69% over 13 days from the anode
chamber in a two-chamber MFC with a ferricyanide
catholyte. In both types of systems, ammonia losses were
accelerated with electricity generation. The possibility of
biological ammonia oxidation with current generation was
investigated by intermittent dosing of reactors with
ammonia and by conducting cyclic voltametry.
Subtask C


Enzyme engineering:
Hydrogenases are among the metalloenzymes for which a gas-substrate tunnel has been
described by using crystallography and molecular dynamics. However, the correlation
between protein structure and gas-diffusion kinetics is unexplored. Two quantitative
methods for probing the rates of diffusion within hydrogenases were recently described
(Leroux et al. 2008 Proc Natl Acad Sci U S A. 2008 105:11188-93. One uses protein film
voltammetry to resolve the kinetics of binding and release of the competitive inhibitor
CO; the other is based on interpreting the yield in the isotope exchange assay. The
study of structurally characterized mutants of a NiFe hydrogenase revealed that two
mutations, which significantly narrow the tunnel near the entrance of the catalytic
centre, decrease the rates of diffusion of CO and H 2 toward and from the active site by
up to 2 orders of magnitude. This proves the existence of a functional channel, which
matches the hydrophobic cavity found in the crystal. However, the changes in diffusion
rates do not fully correlate with the obstruction induced by the mutation and deduced
from the x-ray structures. It was therefore demonstrated that diffusion rates cannot be
predicted and should be measured and that the role of side-chains are essential in the
dynamics of this mechanism.
Subtask C

A theoretical QM/MM study of the [NiFe] hydrogenase
from Desulfovibrio fructosovorans has been performed to
investigate possible routes of proton transfer between the
active site and the protein surface (Galvàn et al. 2008
Proteins. 73: 195-203). The minimum energy paths, were
obtained using a modified version of the nudged elastic
band method, for a set of proposed pathways. The
calculations were carried out for the crystallographic
structure and for several structures of the protein obtained
from a molecular dynamics simulation. The results show
one of the studied pathways to be preferred for transport
from the active site to the surface, but the preference is
not so strong when transport occurs in the opposite
direction.
Subtask C

Maturation of the [FeFe]-hydrogenase active site depends on at least the
expression of three gene products called HydE, HydF, and HydG. The structure
at high resolution of the recombinant, reconstituted S-adenosine-Lmethionine- dependent HydE from Thermotoga maritima was solved recently
(Nicolet et al. 2008 J. Biol. Chem. 283:18861-72). Besides the conserved
[Fe(4)S(4)] cluster involved in the radical-based reaction, this HydE was
reported to have a second [Fe(4)S(4)] cluster coordinated by three Cys
residues. However, depending on the reconstitution and soaking conditions, this
second cluster is either a [Fe(2)S(2)] centre, with water occupying the fourth
ligand site or is absent. We have carried out site-directed mutagenesis studies
on the related HydE from Clostridium acetobutylicum, along with in silico
docking and crystal soaking experiments, to define the active site region and
three anion-binding sites inside a large, positive cavity, one of which binds
SCN(-) with high affinity. Although the overall triose-phosphate isomerasebarrel structure of HydE is very similar to that of biotin synthase, the residues
that line the internal cavity are significantly different in the two enzymes.