Chapter 4

Hydrogen Production: Photofermentation

Alessandra Adessi and Roberto De Philippis

Keywords Photofermentation • Photosynthetic bacteria • Organic acids • Photobioreactors

• Metabolic versatility • Nitrogenase • Uptake hydrogenase • C/N

In this chapter, the production of hydrogen with purple nonsulfur (PNS) bacteria is
presented, describing the main physiological features of PNS bacteria, the enzymes
involved in H2 production and the technological aspects related to the process of
photofermentation. In particular, the conversion yields of the substrates, both synthetic
and derived from waste products, utilized for H2 production are presented, together
with some considerations on the efficiency of the conversion of the light energy into
hydrogen energy. The problems related with the scaling up of the process and with
the use of solar light are finally discussed, presenting some of the open problems
still to be solved in order to make this process economically feasible.

4.1 Introduction

One of the possible ways for biological hydrogen production is connected with a
process referred to as photofermentation. In this process, anoxygenic photosynthetic
bacteria, and in particular, purple nonsulfur (PNS) bacteria, are able to reduce H+
ions to gaseous H2, using both the reducing power deriving from the oxidation of
organic compounds, such as low-molecular weight fatty acids, and the energy deriv-
ing from light. This process is generally considered very promising, due to the high
substrate conversion yields that can be achieved, the possibility to use a wide spec-
trum of the sunlight, the absence of O2 evolving reactions (that would inhibit the

A. Adessi • R. De Philippis (*)

Department of Agricultural Biotechnology, University of Florence, Piazzale Delle Cascine 24,
I-50144, Florence, Italy
e-mail: [email protected]

P.C. Hallenbeck (ed.), Microbial Technologies in Advanced Biofuels Production, 53

DOI 10.1007/978-1-4614-1208-3_4, © Springer Science+Business Media, LLC 2012
54 A. Adessi and R. De Philippis

H2-producing enzymes), and the possibility to couple this kind of H2 production

processes with waste disposal (Basak and Das 2007).
As it has been said, the main microorganisms involved in this process are PNS
bacteria, a group of anoxygenic photosynthetic bacteria widely distributed in many
natural environments, in particular, in anoxic water environments. In this chapter,
the main features of PNS bacteria and of their application in hydrogen production
processes are discussed, also presenting the open problems still to be solved in order
to make this process economically feasible.

4.2 Purple Nonsulfur Bacteria: Systematics, Habitats,

and Main Metabolic Features

Purple nonsulfur (PNS) bacteria are anoxygenic phototrophic bacteria that contain
photosynthetic pigments and are able to perform anoxygenic photosynthesis under
anoxic conditions; they mainly belong to the taxonomic group Alphaproteobacteria,
even if many species belong to the Betaproteobacteria (respectively, 18 and 3 genera
have been recognized for each group, as described by Madigan and Jung 2009). The
Alphaproteobacteria are divided into three subgroups: a-1 for Rhodospirillum and
relatives, a-2 for Rhodopseudomonas and relatives, and a-3 for Rhodobacter and
relatives (Imhoff 2006). Indeed, it is a very diverse group as regards morphology,
internal membrane structure, carotenoid composition, utilization of carbon sources
and electron donors, cytochrome c structures, lipid composition, quinone composition,
lipopolysaccharide structure, and fatty acid composition (Imhoff 1995). However,
sequence analysis of the proteins composing the reaction center (Nagashima et al.
1997) showed a similarity that suggests a lateral gene transfer for the acquisition of
their phototrophic capacity, even corroborated by the close relation with strictly
chemotrophic relatives (Imhoff 2006).
PNS bacteria can be found in aquatic environments rich in organic soluble matter
as lakes, waste water ponds, and costal lagoons. Some representatives can also be
found in sediments and moist soils, and some even in marine and hyper saline envi-
ronments. They usually occur in temperate habitats, but some PNS bacteria reside
in thermal springs and in cold polar waters (Imhoff 2006). However, eutrophic
ponds are the most common habitat where, only occasionally, PNS bacteria can
form dense blooms; more frequently they inhabit the anoxic or low-oxygen tension
layers of water bodies. At the same time, a suitable light irradiation is preferred even
if not strictly necessary.
The publication of the first complete genome sequence of a PNS bacterium,
Rhodopseudomonas palustris (Larimer et al. 2004), pointed out the metabolic
versatility of these bacteria. Such a complexity of metabolic pathways requires
further discussion. The unique characteristic of purple bacteria is their ability to
form their energy carrier (ATP) in the absence of oxygen by using sunlight as a
source of energy. All PNS bacteria can grow photoheterotrophically using reduced
4 Hydrogen Production: Photofermentation 55

Fig. 4.1 Main processes related to hydrogen production, under photoheterotrophic growth in
nonnitrogen fixing conditions: anoxygenic photosynthesis, ATP synthesis, TCA cycle, hydroge-
nase, and nitrogenase activities. The straight black arrows indicate the electron flow. The lightning
symbol indicates light excitation. Abbreviations: Cyt bc1 = cytochrome bc1 complex; Cyt c2 = cyto-
chrome c2; Fd = ferredoxin; RC = Reaction Center; Succinate – DH = succinate dehydrogenase;
NADH-DH = NADH dehydrogenase

carbon compounds as electron donors and carbon source; some species can also
grow photolithoautotrophically using S2−, H2, or Fe2+ as electron donors and CO2 as
the sole carbon source (Larimer et al. 2004).
The cycle of anoxygenic photosynthesis is presented schematically in Fig. 4.1: a
photon stimulates the excitation of bacteriochlorophylls in the reaction center and
this energy is used for the release of an electron which reduces the quinone Q. Once
the quinone is doubly reduced (i.e., after a second photon is captured), it picks up
protons from the cytoplasmic space and translocates through the membrane to reach
the cytochrome bc1 complex: here electrons are addressed to the cytochrome c2 (Cyt c2)
while protons are released in the periplasmic space. Cyt c2 is then able to reduce the
oxidized primary electron donors in the RC, thus closing the cycle. The protons
accumulated in the periplasm form an electrochemical gradient which is used by the
ATP-synthase to generate ATP.
The reduced quinones can open the cycle making the NADH dehydrogenase
working in the “reversed” way to reduce NAD+ to NADH, and the succinate dehy-
drogenase can also work “backwards” reducing fumarate to succinate (processes not
shown in Fig. 4.1). The reversed NADH dehydrogenase reaction is also the way to
refurnish the cell with NADH reducing equivalents (Adessi and De Philippis 2012).
56 A. Adessi and R. De Philippis

In the presence of O2, anoxygenic photosynthesis in purple bacteria is inhibited

and ATP is synthesized through cellular respiration. Under dark anoxic conditions,
electron acceptors other than oxygen can be used for respiration: some conventional
substrates, such as sulfur and nitrogen compounds, and some “exotic” substrates like
DMSO (dimethylsulfoxide), TMAO (trimethylamine-N-oxide), and even arsenate
and halogenated aromatics (Zannoni et al. 2008).
Depending on the metabolic mode PNS bacteria carry out, carbon compounds
have different roles being not only a carbon source but also a source of reducing
power. In photoheterotrophy they cover both roles, but if some inorganic electron
donor is present, carbon is exclusively assimilated. During respiration, carbon
compounds are mainly oxidized, and only a small part is assimilated. It has to be
stressed that PNS bacteria are also able to fix CO2 in autotrophic conditions, using
RuBisCO to refurnish the cell with organic carbon, but some activity of this enzyme
has also been observed during heterotrophic growth in order to equilibrate the redox
state (Tabita 1995).
PNS bacteria are able to use a wide variety of organic carbon compounds,
namely, the intermediates of the tricarboxylic acid cycle, pyruvate and acetate,
organic acids, amino acids, alcohols, and carbohydrates; some of them are high-
lighted in frames in Fig. 4.2. Some species can also use one-carbon atom com-
pounds such as methanol and formate, while some other species grow using aromatic
organic compounds such as benzoate, cinnamate, chlorobenzoate, phenylacetate,
or phenol (Harwood 2008). This versatility is not surprising, considering the vari-
ety of natural environments in which PSN bacteria have been found, as above
reported.
Among PNS bacteria, only Alphaproteobacteria use inorganic sulfur compounds
as electron donors for reductive carbon dioxide fixation during photolithoauto-
trophic growth. In particular, they oxidize sulfide and/or thiosulfate to sulfate and/
or elemental sulfur (Sander and Dahl 2008). That bears witness to a spurious nomen-
clature, as purple “nonsulfur” bacteria actually use sulfur compounds; it has to be
said that any rate they generally have a lower tolerance to sulfide, resulting in toxic-
ity at lower concentrations than for purple sulfur bacteria.
Purple bacteria can use oxidized nitrogen compounds as electron acceptors,
carrying out denitrification. Many are complete denitrifiers, while some are so-
called partial denitrifiers, because of the four enzymes required for the nitrate
reduction to N2 not all are present (Shapleigh 2008). Nitrogen reduction can have
both an assimilatory or dissimilatory purpose. To be assimilated nitrate is reduced
to nitrite and then directly to ammonia (Richardson et al. 2001), but this ability is
not widespread across the group. The preferred way to assimilate nitrogen is fixa-
tion through nitrogenase that reduces nitrogen to ammonia. The enzyme produces
hydrogen as a byproduct, but also functions in the absence of molecular nitrogen
using protons as electron acceptors to dissipate the excess of reducing power in
the cell.
Hydrogen can also be an electron donor for purple bacteria, oxidized by a mem-
brane bound enzyme, named hydrogenase. The reaction can take place in both
directions, depending on the presence or absence of the substrates. A further discus-
sion of these two enzymes is in Sect. 4.3.
4 Hydrogen Production: Photofermentation 57

4.3 Enzymes Involved in Hydrogen Production

It is known that nitrogen fixation is related to hydrogen production, for example, it

was calculated that one million tons of H2 per year is produced by nodule bacteria
(Evans et al. 1987), well known as nitrogen fixing microorganisms. Nitrogenase is
the enzyme that, in all the N2-fixing prokaryotes, including PNS bacteria, is responsible
for hydrogen production, catalyzing the reaction (4.1) that leads to the production
of one H2 molecule per molecule of N2 fixed.

N 2 + 8H + + 8e − + 16ATP → 2NH 3 + H 2 + 16ADP (4.1)

Usually, nitrogen fixing microorganisms also possess a mechanism to uptake the

hydrogen produced in case of need of reducing power dividing it into electrons and
protons (4.2) through the activity to hydrogenase, a membrane bound enzyme, able
to catalyze the reaction in both directions. The uptake reaction is schematically
presented in Fig. 4.1.

H 2 ↔ 2H + + 2e − (4.2)

High hydrogenase activities have been observed in cells possessing an active nitrogenase;
the hydrogen produced by nitrogenase stimulates the synthesis of hydrogenase in
growing cells, even though the synthesis of hydrogenase is not closely linked geneti-
cally to the synthesis of nitrogenase (Coulbeau 1980).

4.3.1 Nitrogenase

Nitrogenase is a two-protein complex consisting of a dinitrogenase containing Fe

and Mo as cofactors and having a molecular weight of 250 kDa, and of a dinitroge-
nase reductase (containing Fe) of about 70 kDa. Some alternative nitrogenases have
been described by Larimer et al. (2004), namely, a Vanadium nitrogenase and a
Fe-only nitrogenase. The three isozymes produce different ratios of hydrogen and
ammonia (McKinlay and Harwood 2010) as shown in reactions (4.3–4.5).

Mo - nitrogenase : N 2 + 8H + + 8e − + 16ATP → 2NH 3 + H 2 + 16ADP (4.3)

V - nitrogenase : N 2 + 12H + + 12e − + 24ATP → 2NH 3 + 3H 2 + 24ADP (4.4)

Fe - nitrogenase : N 2 + 24H + + 24e − + 48ATP → 2NH 3 + 9H 2 + 48ADP (4.5)

Nitrogenase catalyzes a very expensive reaction in terms of energy, and thus it is

very strictly regulated by the presence of dissolved ammonium ions. The regulation
of nitrogenase has been studied in Rhodobacter capsulatus that only contains Mo
and Fe nitrogenases. In Rb. capsulatus, the regulation has been modeled as a three-level
58 A. Adessi and R. De Philippis

control mechanism, as described by Masephol et al. (2002), but the regulatory

cascade described for this microorganism might not be applicable to all PNS
bacteria, due to various differences, including a variable presence or absence of the
three isozymes.
As shown in Fig. 4.1, in the absence of molecular nitrogen, the enzyme, catalyzing
reaction (4.6), dissipates the excess of reducing equivalents deriving from other
metabolic processes.

8H + + 8e − + 16ATP → 4H 2 + 16ADP (4.6)

This is the reaction used for hydrogen production processes; as shown in Fig. 4.1,
nitrogenase under nonnitrogen fixing conditions uses ATP and electrons deriving
from the cyclic photosynthesis: the electrons are transferred to nitrogenase by ferre-
doxins that have been previously reduced in an ATP-consuming reaction.

4.3.2 Hydrogenase

The hydrogenases are iron–sulfur proteins distributed into two main phylogenetically
distinct classes, the [NiFe]-hydrogenases and the [FeFe]-hydrogenases, which contain,
respectively, a Ni and a Fe atom or two Fe atoms at their active site.
The [NiFe]-hydrogenases are the most studied and the kind most frequently
found in photosynthetic bacteria. The synthesis of these enzymes occurs under
anaerobic conditions, and is usually negatively regulated by O2. [NiFe]-hydrogenases
are divided into four groups, according to Vignais (2008), based on their function:
1. Uptake hydrogenases: Respiratory enzymes, which recover electrons from H2,
reducing the membrane-soluble quinones; they are involved in anaerobic
respiration.
2. Cytoplasmic H2 sensors: Regulatory enzymes, able to activate the cascade regu-
lating the respiratory hydrogenases in the presence of H2.
3. Bidirectional heteromultimeric cytoplasmic [NiFe]-hydrogenases: Enzymes able
to bind NAD and NADP and to work in both directions, either to generate reduced
nucleotides, or to dispose of exceeding electrons.
4. H2 evolving, energy-conserving, membrane-associated hydrogenases: These
multimeric enzymes appear to couple anaerobic oxidation of one-carbon-atom
organic compounds to the production of H2.
[FeFe]-Hydrogenases are very uncommon in PNS bacteria, yet its presence in
Rp. palustris is confirmed by the genome sequence (Larimer et al. 2004). This could
be a result of horizontal gene transfer, as it is an enzyme usually found in anaerobic
prokaryotes as clostridia and sulfate-reducing bacteria. In H2 production processes,
active uptake hydrogenases are undesirable, as they affect the gas production: in
particular, an inactivation of such enzymes usually leads to an enhanced hydrogen
production (Ooshima et al. 1998; Franchi et al. 2004; Kim et al. 2006; Öztürk et al.
2006; Kars et al. 2008).
4 Hydrogen Production: Photofermentation 59

Glucose Fructose

Membrane

Glucose 6 P Fructose 1 P

Entner Embden
Doudoroff Meyerhof
Pathway Pathway

3−PGAL 3−PGAL DHAP

(2X)

3−PGA CO2 + H2

Calvin cycle

PEP
CO2

Lactate

Pyruvate D−Malate

CO2

CO2 CO2
Acetyl−CoA Acetate

Butyrate
Okaloacetate
Citrate

L−Malate TCA cycle Isocitrate

CO2
Fumarate
2 − Oxoglutarate Glutamate

Succinate

CO2

Propionate

Fig. 4.2 Carbon metabolism in PNS bacteria. Frames highlight some of the most common
substrates metabolized by PNS bacteria. Abbreviations: 3-PGAL = glyceraldehydes-3-phosphate;
DHAP = di-hydroxy-aceton-phosphate; 3-PGA = 3-phospho-glyceric acid; PEP = phospho-
enol-pyruvate
60 A. Adessi and R. De Philippis

4.3.3 Conversion of Substrates to Hydrogen

As has been described, hydrogen production in purple bacteria is related to many

metabolic processes that deal with ATP generation (photosynthesis), carbon metab-
olism (TCA cycle and carbon fixation), and nitrogen fixation (see Fig. 4.1). Usually,
all the processes involved in energy generation, as photosynthesis and H2 oxidation,
and energy consumption, as N2 and CO2 fixation, are globally regulated by the two
component system RegB–RegA (Elsen et al. 2000). A deeper understanding of
the relationships occurring among these processes could help improving hydrogen
production by finding the right balance between them.
As mentioned above, the preferred substrates for hydrogen production are the
low-molecular weight organic acids that can easily enter the TCA cycle, which is
very active during anaerobic photosynthetic growth. The scheme (Fig. 4.2) repre-
sents carbon metabolism in PNS bacteria, even if not all species and genera follow
this scheme: for example, Rp. palustris does not have the Entner–Doudoroff path-
way (Larimer et al. 2004). Figure 4.2 shows also the role of the Calvin cycle in
carbon metabolism. Joshi and Tabita (1996) demonstrated that the absence of
the reductive pentose phosphate CO2 fixation pathway enhances the synthesis of
nitrogenase even in the presence of ammonium ions, as the reduction of CO2 is, in
photoheterotrophy, just another way to dissipate the reducing power deriving from
organic carbon compounds.
An important parameter in the evaluation of the yield of a hydrogen production
process is the substrate conversion efficiency, calculated as the ratio between the
moles of hydrogen produced and the moles theoretically obtainable if all the substrate
were converted to CO2 and H2. Thus, considering the most common organic acids
utilized in photofermentation processes (Barbosa et al. 2001), the conversion yields
can be calculated from the following reactions:

Lactate : C3 H 6 O3 + 3H 2 O → 6H 2 + 3CO2 (4.7)

Acetate : C2 H 4 O2 + 2H 2 O → 4H 2 + 2CO2 (4.8)

Malate : C4 H 6 O5 + 3H 2 O → 6H 2 + 4CO2 (4.9)

It has to be stressed that these reactions are theoretical, because they are neither
considering the utilization of the substrate for the growth nor the limiting factors
occurring in a culture. On the basis of these reactions, the gas should be expected to
be composed of 66.7% H2 and 33.3% CO2 when growing on lactate and acetate; 60%
H2 and 40% CO2 when growing on malate. Actually the gas phase above the culture
is much richer in H2 than in CO2, due to a partial solubilization of CO2 in the culture
medium and also to a partial fixation to CO2 for anabolic reactions. A 100% conver-
sion efficiency was reported by Sasikala et al. (1990), but in a limited culture volume
(2 ml); substrate conversion yields (reported in Table 4.1) mainly range for acetate
between 69 and 75%; for lactate between 50 and 85%; for malate from 25 to 88%.
 4

Table 4.1 Substrate conversion yields, light conversion efficiencies, mean H2 production rates, and maximum H2 production rates of PNS bacteria growing on
synthetic media
Substrate Light Mean rate Maximum rate
Carbon source Organism conversion (%)a conversion (%)b (ml l−1 h−1) (ml l−1 h−1) References
Acetate Rhodopseudomonas sp. 72.80 0.90 n.a. 25.20 Barbosa et al. (2001)
Rhodopseudomonas palustris 14.80 0.10 n.a. 2.20 Barbosa et al. (2001)
Rhodobacter capsulatus 75 0.68 16.78c n.a. Özgür et al. (2009)
Rhodobacter sphaeroides ZX-5 69 14.04c 90 Tao et al. (2008)
Lactate Rhodopseudomonas sp. 9.60 0.40 n.a. 10.70 Barbosa et al. (2001)
Rp. palustris 12.60 0.50 n.a. 9.10 Barbosa et al. (2001)
Rb. sphaeroides RV 50–80 n.a. n.a. 36.60 Fascetti and Todini (1995)
Rb. capsulatus JP91 52.70 n.a. 21.50 38.50 He et al. (2006)
Hydrogen Production: Photofermentation

Rb. capsulatus IR3 68.20 n.a. 33.20 52.50 He et al. (2006)

Rb. sphaeroides ZX-5 81.20 n.a. 34.60c 103 Tao et al. (2008)
Rubrivivax gelatinosus L31 50.50 n.a. n.a. 2.90 Li and Fang (2008)
Rb. capsulatus IR3 84.80 n.a. 34.40 58.20c He et al. (2005)
Malate Rhodopseudomonas sp. 6.60 0.08c n.a. 1.10 Barbosa et al. (2001)
Rp. palustris 36 0.30 n.a. 5.80 Barbosa et al. (2001)
Rb.sphaeroides O.U.001 36.00 n.a. n.a. 8.00 Koku et al. (2003)
Rubrivivax gelatinosus L31 24.60 n.a. 9.03c 2.70 Li and Fang (2008)
Rb. sphaeroides ZX-5 78.90 n.a. n.a. 92 Tao et al. (2008)
Rb. sphaeroides ZX-5 88.26 n.a. 69.78 165.90 Li et al. (2009)
Rb. sphaeroides O.U.001 n.a. n.a. 10.00 12 Eroğlu et al. (1999)
n.a. not available
a ,b
Calculated as indicated in the text
c
Calculated by authors
61
62 A. Adessi and R. De Philippis

The conversion efficiency is strongly affected by the C/N ratio in the culture.
Indeed, a high C/N ratio in the culture medium usually leads to higher hydrogen pro-
duction compared with a low C/N ratio, where a higher cell growth occurs (Redwood
et al. 2009). In the latter case, the conversion efficiency decreases due to the con-
sumption of the organic acids for cell growth instead of for hydrogen production.
This problem becomes a very relevant matter when wastewaters or liquors
derived from other fermentation processes are utilized for the production of H2 by
means of photofermentation. This matter will be treated in the following section of
this chapter.

4.5 Research on Hydrogen Production Processes

Carried Out with PNS Bacteria

Hydrogen production using PNS bacteria is an attractive process due to the oppor-
tunity of producing a nonpolluting energy vector through a nonpolluting biological
process that can also be combined with the disposal of various kinds of wastes. It is
also possible to couple the production of H2 with the production of other valuable
products, as vitamins or biological plastic materials (PHB), or with the use of the
spent biomass itself as a fertilizer. These opportunities are the lights that drive the research
along an uneven path. The hurdles along this path are constituted by the costs and
by the low efficiencies of conversion of substrate and of light into H2. Costs are
related to the type of substrate used for the photofermentation, to the light source
and, mainly, to the construction of photobioreactors. The complex issues regarding
photobioreactors are discussed in Sect. 4.6. There are also energetic costs related
the light sources as well as culture mixing.
The aim of the research in this field is to find the best balance between costs and
yields: if the yield is not very high the costs have to be very limited, but if a very
high yield is reached a higher cost can be tolerated. It is possible to individuate two
different lines in the current research: on the one hand there are groups working with
very low-cost systems, using complex low-cost substrates and natural light sources;
on the other hand there are groups working on the optimization of hydrogen produc-
tion, aiming at obtaining the best production rates with the best technology avail-
able. Regardless of the approach chosen, the goal is a cost-effective process.

4.5.1 Substrates for Hydrogen Production Using PNS Bacteria

4.5.1.1 Synthetic Substrates

The use of synthetic media for hydrogen production process has a very high relevance,
as it describes the behavior of the microorganisms in a controlled system, where the
culture medium is completely defined. Culture conditions are usually very homogeneous
4 Hydrogen Production: Photofermentation 63

with regard to temperature and pH, respectively, around 30°C and around 7.0. Based
on the two reviews recently published by Koku et al. (2002) and Kapdan and Kargi
(2006), it is possible to say that acetate, lactate, and malate are the most commonly
used organic acids for hydrogen production; only a little data is available for
butyrate, and some data refer to the use of sugars, such as glucose or fructose.
The C/N ratio is quite variable among the studies reported, as it has to be optimized
according to the microorganism and the carbon substrate used.
Considering the three most frequently utilized substrates, it appears that their
conversion yield is highly dependent on the PNS bacterial strain and on the condi-
tions utilized (Table 4.1), showing a high variability even within the data obtained
with the same substrate. In any case, the best conversion yields are in the range
75–88% (Table 4.1). In all the studies, the maximum rate of H2 production was
reported, but quite frequently the mean production rate, which is a parameter of
great interest for practical applications, is not available. The maximum rates reported
range from very low values (about 1–2 ml l−1 h−1) to very high values (103 and
165 ml l−1 h−1). In particular, the highest maximum production rate so far obtained
was achieved by using high light intensity with stacked cultures of Rhodobacter
sphaeroides ZX-5 (Li et al. 2009).
However, the little data that is available on light conversion efficiency shows
very low values, thus pointing out that one of the main critical points to be solved is
the efficiency of light utilization by the PNS bacteria. Indeed, it has been reported
that a light conversion efficiency of about 10% should be achieved in order to make
this process economically feasible (Basak and Das 2007; Akkerman et al. 2002).

4.5.1.2 Substrates Deriving from Wastes of Industrial or Agricultural

Processes

As mentioned above, one of the most interesting features of PNS bacteria is their
capability to use, for the production of H2, waste residues derived from industrial or
agricultural processes. This characteristic gives two potential economic advantages
to this process (1) the substrate is free or very cheap, being a waste derived from other
processes; (2) the use of these substrates for H2 production processes reduces or
eliminates the cost of their treatment and/or disposal. Moreover, these substrates are,
in many cases, available in large amounts. However, it has to be considered that the
H2 production plant must be located close to the site or area of production of wastes,
in order to maintain at the lowest level the expenses for transporting this material.
According to Koku et al. (2002), wastes deriving from sugar refineries (Yetis
et al. 2000), tofu factories (Zhu et al. 1999a), olive mills (Eroğlu et al. 2002), municipal
solid wastes (Fascetti et al. 1998), dairy plants (Türkaslan et al. 1998), and lactate
fermentation plants (Sasikala and Ramana 1991) have been used, at proper dilutions
and with proper additions, with Rb. sphaeroides. In particular, as they report, the best
waste used with this species was tofu wastewater, considering the high gas production
rates and the fact that the wastewater was not diluted and no addition of other nutrients
was necessary; they obtained a gas production rate of 15.9 ml l−1 h−1.
64 A. Adessi and R. De Philippis

Hydrogen production using a Rhodopseudomonas sp. strain was investigated by

Singh et al. (1994) using a substrate potato starch, sugarcane juice, and whey: the best
result was obtained with sugarcane juice, which showed a specific rate of hydrogen
production of 45 ml/g of dry weight per hour. De Philippis et al. (2007) reported the
use of a fermentation broth, derived from the spontaneous fermentation of vegetable
wastes, for the production of H2 with Rp. palustris; a mean production rate of
16 ml l−1 h−1 was obtained. Tao et al. (2008) tested three different wastewaters using
Rb. sphaeroides ZX5: a wastewater derived from a succinate producing factory
(mixed with a synthetic medium without carbon source), a wastewater of a fuel
ethanol manufacturer (with a threefold dilution), and a kitchen waste (diluted twofold)
obtaining, respectively, maximum rates of 55, 48, and 45 ml l−1 h−1.
Efforts have been made in order to overcome some of the problems arising from
the use of low-cost substrates (Redwood et al. 2009):
1. Control of the NH4+-dependent inactivation of nitrogenase.
A big concern about the use of these substrates is the low conversion of substrates
and nitrogenase “switch off” due to the presence of nitrogen sources in the sub-
strate. For solving this problem, various approaches have been followed:
• Development of NH4+ insensitive strains (Zinchenko et al. 1991; Yagi et al.
1994; Zinchenko et al. 1997).
• Electroseparation of NH4+ from the culture medium (Redwood and Macaskie 2007).
• Use of immobilized cultures in anion selective matrices (Zhu et al. 1999b, 2001).
2. A bioreactor design aimed at optimizing the process (discussed in Sect. 4.6).
3. The development of uptake hydrogenase deficient strains.
As discussed in Sect. 4.3.2, the inactivation of uptake hydrogenases leads to
enhanced hydrogen production (Ooshima et al. 1998; Franchi et al. 2004; Kim
et al. 2006; Öztürk et al. 2006; Kars et al. 2008).
4. The development of PHB-deficient strains.
The biosynthesis of storage material such as poly-beta-hydroxybutyrate competes
with hydrogen production, as it has the same function of dissipating excess reduc-
ing power (Vincenzini et al. 1997; Koku et al. 2002). The development of mutants
with inactivation of the PHB biosynthetic pathway resulted in improved hydrogen
production rates, but only when uptake-hydrogenase activity had also been abol-
ished (Franchi et al. 2004; Kim et al. 2006); the strain with only the PHB biosyn-
thetic pathway deleted did not exhibit increased rates (Franchi et al. 2004).
However, it has to be stated that most of the above-mentioned studies have
been done at a laboratory scale, and their real applicability needs to be verified at a
pilot scale.

4.5.2 Solar Irradiation and Related Issues

As mentioned above, light irradiance is very important factor when using photosynthetic
bacteria and it has to be stressed that in a cost-effective system the best solution
4 Hydrogen Production: Photofermentation 65

would appear to be the use of natural solar light. However, there are a number of
problems arising from the use of natural irradiation. They are presented below
together with some new possibilities for enhancing light conversion efficiency.
Even if purple bacteria are able to use a wide range of the solar light spectrum
(400–950 nm), in fact this PAR (photosynthetic active radiation) for purple bacteria
is only 65.8% of total solar radiation (Akkerman et al. 2002). Another problem
comes from the light saturation of PNS bacterial cultures: Miyake et al. (1999)
showed how, in an outdoor experiment, the maximum rates (3.4 l m−2 h−1) were
obtained 2–3 h after the maximum light intensity at noon, while during the period
of the day with the highest irradiation (about 1.0 kW m−2) hydrogen production rates
were significantly lower, thus indicating probable photoinhibition. The same study
points out how the intrinsic variability of solar light makes the rates vary along with
light intensity during the day: this means that the process is continuously varying
and the rates cannot be constant. It is anyway interesting to observe how after the
night period, when gas production ceases, photoevolution of gas starts again after a
lag period of 2–4 h.
Özgür et al. (2009) showed how when passing from indoor experiments to
outdoors, temperature fluctuation becomes a very relevant concern, as temperature
fluctuation decreases H2 production by 50%, and being subject to light/dark cycles
further decreases it. In another photobioreactor irradiated by solar light (Eroğlu
et al. 2008), but bigger in volume (8.0 l instead of the 0.550 l of the previous article
cited), an average production rate 10 ml l−1 h−1 was obtained when using malate as a
substrate.
In any case, it is evident that absent or insufficient light irradiation stops hydrogen
production, and this has an effect on the total gas production, that is surely lower
than the amount that can be produced by continuous illumination. In this regard, it
is worth considering integrated artificial and solar light systems; as an example a
system has been proposed (Ogbonna et al. 1999) to overcome the solar light variations
during the day, bad weather periods, and the night periods: solar light was collected
by Fresnel lenses equipped with a light-tracking sensor; the solar light collection
device was connected to optical fibers that brought light into light radiators which
homogeneously diffused light into the photobioreactor. This system was equipped
with a light intensity sensor that in the case of insufficient solar light intensity switched
on an artificial light to supply the culture’s light needs. This ingenious system may
have opened up a path aiming to create an homogeneous and continuous hydrogen
production process. However, a careful evaluation of the costs of the use of artificial
light and benefits in terms of increased hydrogen production must be done before
proposing these technologically complex systems.

4.5.3 Scaling Up of the H2-Producing Processes

McKinlay and Harwood (2010) stressed the relevance of the energetic yield of
hydrogen production rates compared with crop-based biofuels if only it was possible
66 A. Adessi and R. De Philippis

to linearly scale up the actual biohydrogen production processes: with a virtual

production process using Rp. palustris they calculated that it is possible to obtain
23–29 equivalent L of gasoline ha−1 day−1, that is much higher than the single equivalent
L of gasoline ha−1 day−1 obtainable with soybean-based biodiesel. This is actually a
wide margin which leaves many possibilities, since, even if the scaling up were not
linear, it can probably remain more convenient than crop-based biofuels. Some
recent studies concern this aspect which is tightly linked with natural irradiation and
with the use of low-cost substrates owing to the higher costs of a scaled-up process
in comparison with a low volume process.
The passage to a higher culture volume leads to higher amounts of gas production,
but the rates and the efficiencies can be significantly lower than with smaller culture
volumes. Eroğlu et al. (2008) carried out a H2-producing photofermentation in an
8 L photobioreactor using natural irradiation with Rb. sphaeroides. Various substrates
were tested (malate, lactate, acetate, and olive mill wastewater) and a 10 ml l−1 h−1
production rate was obtained with malic acid, while a 3 ml l−1 h−1 production rate
was obtained with olive mill wastewater (4% olive mill wastewater in distilled
water). De Philippis et al (2007) scaled up the hydrogen production process starting
from a 0.25 L bioreactor to an 11.0 L column photobioreactor using vegetable
wastewater (50% wastewater in distilled water) as a substrate, with Rp. palustris.
Using artificial light irradiation, rates decreased from 16 to 11 ml l−1 h−1 in the
scaled-up process.
When using much larger volumes, the processes should use a fed-batch processes
instead of a batch, as once the biomass is inoculated it is desirable to use it as long
as possible due to the complex operations needed to manage big volumes of culture.
An example of a fed batch, semipilot scale biomass production process was reported
by Carlozzi and Sacchi (2001): they used a temperature controlled tubular culture
system of 53 l volume containing growing Rp. palustris under solar irradiation with
an irradiated area of 1.52 m2 and a total ground footprint of 2.0 m2. Even though
the process was not aimed at hydrogen production, it highlights the importance of
operating at the right biomass concentration when growing phototrophic bacteria.
The authors indicated an optimal culture concentration of 1 g of dry weight l−1 to
obtain the best biomass productivity and also demonstrated that, in order to maintain
this cell concentration, a fed-batch process is necessary.
Boran et al. (2010) have investigated hydrogen production using Rb. capsulatus
in a solar tubular photobioreactor of 80l volume with an illuminated area of 2.0 m2
and a total footprint of 2.88 m2. They artificially illuminated the culture during the
exponential phase of growth, than they started feeding and natural irradiation.
At the end of 32 days, they had obtained 80 l of hydrogen, but gas production
actually started only after 6 days and after cells had reached a concentration of
0.8 g l−1. A mean rate of 0.31 and a maximum rate of 0.74 mol H2 m−3 h−1 were
obtained. They calculated a mean light intensity of 90 W m−2 during the light hours
out of a 13-day period, and on this basis they calculated a conversion efficiency of
1%. However, it has to be stressed that during such a long period of time the solar
light intensity might have undergone many significant variations, and thus calculating
a mean value would appear not to be the best way to evaluate this parameter. They
4 Hydrogen Production: Photofermentation 67

also observed that, when the light intensity was below 90 W m−2, hydrogen production
ceased and the substrate was only used for cell growth. However, also considering
that the producing system operated under not yet optimized conditions, the results
seem promising. Indeed, they have shown the feasibility of the scaling up of the
system, and the possibility to maintain a culture of PNS bacteria for a long time (about
1 month) under natural irradiation with a significant production of hydrogen.

4.6 The Issue of Light and of the Geometry of Photobioreactors

One of the most relevant issues to be considered for the optimization of the production
of H2 via photofermentation is the photochemical efficiency of the system.
Considering the absorption spectrum of purple bacteria (Fig. 4.3), Miyake (1998)
calculated that for the production of a single molecule of H2, 11 photons are required
at 860 nm; Akkerman et al. (2002) calculated that 14–15.8 photons are required for
a molecule of H2 at 522 nm. Even though there are no data available on the quantum
yields at the other wavelengths utilizable by purple bacteria, it has been estimated
that the overall theoretical photosynthetic efficiency (PE) (4.9) is at least 10%. The
details of these calculations can be found in Akkerman et al. (2002).

Free energy of the total amount of H 2 produced (4.10)

PE = × 100.
Total energy of the light incident on the culture
Moving from theory to real applications, the value of light conversion efficiency
dramatically drops down not only under natural sunlight, but also under artificial
irradiation. Indeed, high light conversion efficiencies have only been reached using
such low light intensities that the production rates are not high enough to be considered
interesting for a H2 production process. Barbosa et al. (2001) observed that higher
light intensities may decrease PE, but usually increase hydrogen productivity.
Miyake and Kawamura (1987) reported light conversion efficiencies of 7.9 and 6.2%
under illumination by a xenon lamp at 50 W m−2 and by a solar simulator at 75 Wm−2,
respectively: those are very low light intensities to reach a gas evolution significant
for a production process. Thus, it can be seen that light intensity, light quality and
sources, light distribution and photobioreactor design, are all very important issues
for the optimization of H2 production with PNS bacteria and are thoroughly treated
in the following paragraphs.

4.6.1 Light Intensity

One of the problems in comparing the data of light efficiency of different H2-
producing systems is due to the lack of homogeneity in the way light intensities are
measured and reported in different papers. Indeed, a first cause of unhomogeneity
68 A. Adessi and R. De Philippis

Fig. 4.3 Absorption spectrum typical for PNS bacteria. Absorption maxima at 805 and 875 are
due to bacteriochlorophyll a

comes from different ways of measuring light, based on different theoretical assump-
tions, which makes it difficult to convert one unit of measurement to another.
Anyway, to give an order of magnitude of the light intensities more frequently used,
they range in the thousands of lux, usually around 10 klux, and in hundreds of both
mmol (photons) m−2 s−1 and W m−2.
Another cause of unhomogeneity in the data reported in the literature is related
to the different light requirements of the organisms utilized, and with the character-
istics of the light source used (incandescent lamp, LEDs, solar light), which have
different spectra of emission as well as different intensities and angle of incidence.
It has also to be stressed that in many cases, if not all, the value of the light energy
used for calculating the light efficiency with (4.9) is that impinging on the photobio-
reactor, and not the one actually absorbed by the culture. As an example, Uyar et al.
(2007) indicate, for Rb. sphaeroides, a minimum light intensity of 270 W m−2 to obtain
high hydrogen production rates; they state that this value is equivalent to 4,000 lux
and 1,370 mmol (photons) m−2 s−1.

4.6.2 Light Quality and Sources

When using artificial light, the most used light sources are tungsten lamps as their
emission spectrum covers the absorption spectrum of PNS bacteria (see Figs. 4.3
and 4.4). Particularly important is the near infrared emission, the absorption maximum
of bacteriochlorophylls.
4 Hydrogen Production: Photofermentation 69

Fig. 4.4 Emission spectrum of a tungsten lamp at an equivalent color temperature of 3,053°K

As tungsten lamps are energy-expensive light sources, some alternatives can be

offered by light emitting diodes (LEDs). Kawagoshi et al. (2010) utilized long-wavelength
LEDs (LW-LEDs), with a maximum emission at 850 nm, to produce hydrogen by
means of a halo-tolerant photosynthetic bacterium. They state that LEDs have a life
time ranging between 20,000 and 30,000 h, while a tungsten lamp lasts for 1,000–
2,000 h, and they prefigure a reduction of energy cost by 98% using LEDs instead
of tungsten lamps.
At the present state of knowledge in technology and processes, it is possible to
predict that the best producing system may come from the optimization of a medium
scaled, naturally irradiated hydrogen production process (Boran et al. 2010), provided
with an artificial light supply switched on as soon as light intensity decreases under
a threshold value (Ogbonna et al. 1999), and where the artificial light is provided by
wavelength-selected LEDs (Kawagoshi et al. 2010).

4.6.3 Photobioreactor Design and Light Distribution

Photobioreactors for biohydrogen production with PNS bacteria are closed systems
that allow the maintenance of anaerobic conditions, and prevent H2 gas leakage;
they are characterized by a high illuminated surface to volume ratio, and need for a
mixing system to keep the cells as much uniformly illuminated as possible. A very
large part of literature regarding photobioreactors is dedicated to microalgal cultures,
and mainly for biomass production processes (Ugwu et al. 2007). Even if some light
distribution characteristics can be common, those processes differ significantly from
hydrogen production processes using purple bacteria.
Two main types of photobioreactors are suitable for hydrogen production
processes with PNS bacteria: flat panel bioreactors and tubular reactors; air lift pho-
tobioreactors are not appropriate for this kind of process as it would be necessary to
70 A. Adessi and R. De Philippis

bubble argon gas inside the reactor, instead of air, in order to maintain anaerobic and
not-nitrogen-fixing conditions, and this would drastically increase the costs.
Flat panel reactors are rectangular transparent boxes that can be either vertical or
inclined in sun direction. They are only a few centimeters thick (1–5 cm, according
to Akkerman et al. 2002) and this exposes cells to only very short mixing-induced
light/dark cycles, which otherwise decrease hydrogen productivity. The advantage
in using flat panel photobioreactors is the possibility to arrange a set of reactors one
behind the other, at a proper distance, to increase the ground area productivity
(Gebicki et al. 2009).
Tubular photobioreactors consist of transparent tubes placed either horizontally
or with an inclination of 10°–30°, south oriented (Gebicki et al. 2009); mixing is
mechanical and gas is collected at the top of the bioreactor. In these systems, the gas
is collected on one side of the tube and this imposes a limit in length; indeed, the
longer the tube, the longer the time the gas bubble stays inside the reactor where H2
could be taken up by cells, thus decreasing hydrogen productivity.
A relevant aspect when using outdoor cultures is the management of light inten-
sities, not only when they are not sufficient, but also when they are too high, as
mentioned above (Sect. 4.5.2). Wakayama and Miyake (2002) developed a light-shade
bands photobioreactor system and were able to obtain a light conversion efficiency
of 3.5% at 800 W m−2 of solar light intensity, while photoinhibition was observed at
400 W m−2 without the shading system.

4.7 Integrated Dark- and Light-Fermentation Processes

for Hydrogen Production

In Sect. 4.5.1, the possibility of producing hydrogen using low-cost starting

substrates in order to minimize the costs of the process was discussed. Another
possible solution is integration with a dark-fermentative H2 production processes: in
these kinds of two-stage H2 production processes, heterotrophic bacteria are utilized
to ferment sugars producing hydrogen, carbon dioxide, and organic acids which are
subsequently used for photofermentation with PNS bacteria. In this way, the light
independent bacteria and photosynthetic bacteria would provide an integrated system
for maximizing the hydrogen yield from glucose. As glucose fermentation in heterotrophic
bacteria is thermodynamically limited to partial sugar oxidation (4.10), a subsequent
fermentation of acetate by phototrophic bacteria (4.11) could lead to the theoretical
maximum conversion of 12 moles of H2 per mole of glucose (4.12).

C6 H12 O6 → 2C2 H 4 O2 + 4H 2 + 2CO2 (4.11)

2C2 H 4 O2 + 4H 2 O → 8H 2 + 4CO2 (4.12)

C6 H12 O6 → 12H 2 + 6CO2 (4.13)

4 Hydrogen Production: Photofermentation 71

Some very promising results have been obtained by Nath and Das (2005) and by
Asada et al. (2006) that reached, respectively, a global yield of 5.3 and 7.1 mol (H2)
mol (glucose)−1 with such integrated processes. The details of the two-stage processes
are discussed in Chap. 7.

4.8 Concluding Remarks

At the present time, increases in energy demands makes it necessary to find new
energy sources, which at the same time reduce the emission of greenhouse gases.
Photofermentative H2 production appears promising because of the possibility of
providing renewable and sustainable energy while using free solar light and managing
waste disposal. However, large-scale production is still far from being practical, as
some obstacles remain to be overcome. Several of these challenges were discussed in
specific topics in this chapter; the most pressing open issues are detailed below:
• Light delivery is a very complex problem, involving both light sources and the
design of efficient photobioreactors.
• There are several points concerning substrates. Substrates need to be found that
have the right C/N ratio and are readily available. Large-scale production requires
industrial, retail, and agricultural wastes suitable for photofermentation that are
produced throughout the year, and with a stable composition. Moreover, BioH2
plants will need to be placed near the waste production sites. Therefore, future
factories of this type should be built in locations with large surface areas available,
in order to provide sufficient area for photobioreactors.
• Multiple-organism systems appear to be a promising method for efficient energy
generation (Redwood et al. 2009; McKinlay and Harwood 2010), and, although
scaling up requires solving an number of economic problems, current laboratory
scale data are encouraging.

Acknowledgments The authors gratefully acknowledge the Italian Ministry of University and
Research (MIUR) (FISR, “IDROBIO” Project), Italian Ministry of Agricultural, Food and Forest
Politics (MIPAAF) (“IMERA” Project), and Ente Cassa di Risparmio di Firenze (“Firenze
Hydrolab” Project) that partially supported the research carried out in their laboratory and men-
tioned in this chapter.

References

Adessi A, De Philippis R (In press) Purple Bacteria: Electron Acceptors and Donors. In: Lennarz
WJ, Lane MD (eds.) Encyclopedia of Biological Chemistry, vol. x, pp. xxx–xxx,. Elsevier
Akkerman I, Janssen M, Rocha J, Wijffels RH (2002) Photobiological hydrogen production:
photochemical efficiency and bioreactor design. Int J Hydrogen Energy 27:1195–1208
Asada Y, Tokumoto M, Aihara Y, Oku M, Ishimi K, Wakayama T (2006) Hydrogen production by
co-cultures of Lactobacillus and a photosynthetic bacterium, Rhodobacter sphaeroides RV. Int
J hydrogen Energy 31:1509–1513
72 A. Adessi and R. De Philippis

Barbosa MJ, Rocha JMS, Tramper J, Wijffels RH (2001) Acetate as a carbon source for hydrogen
production by photosynthetic bacteria. J Biotechnol 85:25–33
Basak N, Das D (2007) The prospect of purple non-sulfur (PNS) photosynthetic bacteria for
hydrogen production: the present state of the art. World J Microbiol Biotechnol 23:31–42
Boran E, Özgür E, van der Burg J, Yücel M, Gündüz U, Eroğlu I (2010) Biological hydrogen
production by Rhodobacter capsulatus in a solar tubular photo bioreactor. J Clean Prod.
doi:10.1016/j.jclepro.2010.03.018
Carlozzi P, Sacchi A (2001) Biomass production and studies on Rhodopseudomonas palustris
grown in an outdoor, temperature controlled, underwater tubular photobioreactor. J Biotechnol
88:239–249
Coulbeau A, Kelley BC, Vignais PM (1980) Hydrogenase activity in Rhodopseudomonas capsu-
lata: relationship with nitrogenase activity. J Bacteriol 141–148
De Philippis R, Bianchi L, Colica G, Bianchini C, Peruzzini M, Vizza F (2007) From vegetable
residues to hydrogen and electric power: feasibility of a two step process operating with purple
non sulfur bacteria. J Biotechnol 131S:S122–S126
Elsen S, Dischert W, Coulbeau A, Bauer CE (2000) Expression of uptake hydrogenase and molyb-
denum nitrogenase in Rhodobacter capsulatus is coregulated by the RegB–RegA two-component
regulatory system. J Biotechnol 182(10):2831–2837
Eroğlu E, Eroğlu I, Gündüz U, Yücel M, Türker L (2002) Biological hydrogen production from
olive mill wastewater by Rhodobacter sphaeroides O.U. 001. In: Gokçekus H (ed) International
conference on environmental problems of Mediterranean region in Near East University,
Nicosa, Northern Cyprus, Book of Abstracts
Eroğlu I, Aslan K, Gündüz U, Yücel M, Türker L (1999) Substrate consumption rates for hydrogen
production by Rhodobacter sphaeroides in a column photobioreactor. J Biotechnol
70:103–113
Eroğlu I, Tabanoğlu A, Gündüz U, Eroğlu E, Yücel M (2008) Hydrogen production by
Rhodobacter sphaeroides O.U. 001 in a flat plate solar photobioreactor. Int J Hydrogen Energy
33:531–541
Evans HJ, Harker AR, Papen H, Russell SA, Hanus FJ, Zuber M (1987) Physiology, biochemistry,
and genetics of the uptake hydrogenase in Rhizobia. Annu Rev Microbiol 41:335–361
Fascetti E, Todini O (1995) Rb. sphaeroides RV cultivation and hydrogen production in a one- and
two-stage chemostat. Appl Microbiol Biotechnol 44:300–305
Fascetti E, D’Addario E, Todini O, Robertiello A (1998) Photosynthetic hydrogen evolution with
volatile organic acids derived from the fermentation of source selected municipal solid wastes.
Int J Hydrogen Energy 23(9):753–760
Franchi E, Tosi C, Scolla G, Penna DG, Rodriguez F, Pedroni MP (2004) Metabolically engineered
Rhodobacter sphaeroides RV strains for improved biohydrogen photoproduction combined
with disposal of food wastes. Mar Biotechnol 6:552–565
Gebicki J, Modigell M, Schumacher M, Van Der Burg J, Roebroeck E (2009) Development of
photobioreactors for anoxygenic production of hydrogen by purple bacteria. Chem Eng Trans
18:363–366. doi:10.3303/CET0918058
Harwood CS (2008) Degradation of aromatic compounds by purple nonsulfur bacteria. In: Hunter
CD, Daldal F, Thurnauer MC, Beatty JT (eds) Advances in photosynthesis and respiration, vol
28, The purple phototrophic bacteria. Springer, Dordrecht, pp 577–594
He D, Bultel Y, Magnin JP, Roux C, Willison JC (2005) Hydrogen photosynthesis by Rhodobacter
capsulatus and its coupling to a PEM fuel cell. J Power Sources 141:19–23
He D, Bultel Y, Magnin JP, Willison JC (2006) Kinetic analysis of photosynthetic growth and
photohydrogen production of two strains of Rhodobacter capsulatus. Enzyme Microb Technol
38:253–259
Imhoff JF (1995) The anoxygenic phototrophic purple bacteria. In: Boone DR, Castenholz RW,
Garrity GM (eds) Bergey’s manual of systematic bacteriology, 2nd edn. Springer, New York,
pp 631–637
Imhoff JF (2006) The phototrophic alpha-proteobacteria. In: Dworkin M, Falkow S, Rosenberg E,
Schleifer KH, Stackebrandt E (eds) The prokaryotes. Springer, New York, pp 41–64
4 Hydrogen Production: Photofermentation 73

Joshi HM, Tabita FR (1996) A global two component signal transduction system that integrates the
control of photosynthesis, carbon dioxide assimilation, and nitrogen fixation. Proc Natl Acad
Sci USA 93:14515–14520
Kapdan IK, Kargi F (2006) Bio-hydrogen production from waste material. Enzyme Microb
Technol 38:569–582
Kars G, Gündüz U, Rakhely G, Yücel M, Eroğlu I, Kovacs LK (2008) Improved hydrogen produc-
tion by hydrogenase deficient mutant strain of Rhodobacter sphaeroides O.U.001. Int J
Hydrogen Energy 33(12):3056–3060
Kawagoshi Y, Oki Y, Nakano I, Fujimoto A, Takahashi H (2010) Biohydrogen production by isolated
halotolerant photosynthetic bacteria using long-wavelength light-emitting diode (LW-LED).
Int J Hydrogen Energy. doi:10.1016/j.ijhydene.2009.11.121
Kim MS, Baek JS, Lee JK (2006) Comparison of H2 accumulation by Rhodobacter sphaeroides
KD131 and its uptake hydrogenase and PHB synthase deficient mutant. Int J Hydrogen Energy
31:121–127
Koku H, Eroğlu I, Gündüz U, Yücel M, Türker L (2002) Aspects of metabolism of hydrogen
production by Rhodobacter sphaeroides. Int J Hydrogen Energy 27:1315–1329
Koku H, Eroğlu I, Gündüz U, Yücel M, Türker L (2003) Kinetics of biological hydrogen produc-
tion by the photosynthetic bacterium Rhodobacter sphaeroides O.U. 001. Int J Hydrogen
Energy 28:381–388
Larimer FW, Chain P, Hauser L, Lamerdin J, Malfatti S, Do L, Land ML, Pelletier DA, Beatty JT,
Lang AS, Tabita FR, Gibson JL, Hanson TE, Bobst C, Torres y Torres JL, Peres C, Harrison
FH, Gibson J, Harwood CS (2004) Complete genome sequence of the metabolic versatile
photosynthetic bacterium Rhodopseudomonas palustris. Nat Biotechnol 22:55–61
Li RY, Fang HHP (2008) Hydrogen production characteristics of photoheterotrophic Rubrivivax
gelatinosus L31. Int J Hydrogen Energy 33:974–980
Li X, Wang TH, Zhang SL, Chu J, Zhang M, Huang MZ, Zhuang YP (2009) Enhancement of
phototrophic hydrogen production by Rhodobacter sphaeroides ZX-5 using a novel strategy-
shaking and extra-light supplementation approach. Int J Hydrogen Energy 34:9677–9685
Madigan MT, Jung DO (2009) An overview of purple bacteria: systematic, physiology and habi-
tats. In: Hunter CD, Daldal F, Thurnauer MC, Beatty JT (eds) Advances in photosynthesis and
respiration, vol 28, The purple phototrophic bacteria. Springer, Dordrecht
Masephol B, Drepper T, Paschen A, Groß S, Pawlowski A, Raabe K, Riedel KU, Klipp W (2002)
Regulation of nitrogen fixation in the photoautotrophic purple bacterium Rhodobacter capsulatus.
J Mol Microbiol Biotechnol 4(3):243–248
McKinlay JB, Harwood CS (2010) Photobiological production of hydrogen gas as a biofuel. Curr
Opin Biotech 21:244–251
Miyake J, Kawamura S (1987) Efficiency of the light conversion to hydrogen by the photosyn-
thetic bacterium Rhodobacter sphaeroides. Int J Hydrogen Energy 12(3):147–155
Miyake J (1998) The science of biohydrogen. In: Zaborsky OR (ed) Biohydrogen. Plenum,
London, pp 7–18
Miyake J, Wakayama T, Schnackenberg J, Arai T, Asada Y (1999) Simulation of the daily sunlight
illumination pattern for bacterial photo-hydrogen production. J Biosci Bioeng 88(6):659–663
Nagashima KVP, Hiraishi A, Shimada K, Matsuura K (1997) Horizontal transfer of genes coding
for the photosynthetic reaction centers of purple bacteria. J Mol Evol 45(2):121–136
Nath K, Das D (2005) Hydrogen production using Rhodobacter sphaeroides strain O.U. 001 using
spent media of Enterobacter cloacae strain DM11. Appl Microbiol Biotechnol 68:533–541
Ogbonna JC, Soejima T, Tanaka H (1999) An integrated solar and artificial light system for internal
illumination of photobioreactors. J Biotechnol 70:289–297
Ooshima H, Takakuwa S, Katsuda T, Okuda M, Shirasawa T, Azuma M, Kato J (1998) Production
of hydrogen by a hydrogenase deficient mutant of Rhodobacter capsulatus. J Ferment Bioeng
85:470–475
Özgür E, Uyar B, Öztürk Y, Yücel M, Gündüz U, Eroğlu I (2009) Biohydrogen production by
Rhodobacter capsulatus on acetate at fluctuating temperatures. Resour Conserv Recycl.
doi:10.1016/j.resconrec.2009.06.002
74 A. Adessi and R. De Philippis

Öztürk Y, Yücel M, Daldal F, Mandacı S, Gündüz U, Türker L, Eroğlu I (2006) Hydrogen produc-
tion by using Rhodobacter capsulatus mutants with genetically modified electron transfer
chains. Int J Hydrogen Energy 31:1545–1552
Redwood MD, Macaskie LE (2007) Method and apparatus for biohydrogen production. British
patent application no. 0705583.3
Redwood MD, Paterson-Beedle M, Macaskie LE (2009) Integrating dark and light bio-hydrogen
production strategies: towards the hydrogen economy. Rev Environ Sci Biotechnol 8:149–185
Richardson DJ, Berks BC, Russell DA, Spiro S, Taylor CJ (2001) Functional, biochemical and
genetic diversity of prokaryotic nitrate reductase. Cell Mol Life Sci 58(2):165–178
Sander J, Dahl C (2008) Metabolism of inorganic sulfur compounds in purple bacteria. In: Hunter
CD, Daldal F, Thurnauer MC, Beatty JT (eds) Advances in photosynthesis and respiration, vol
28, The purple phototrophic bacteria. Springer, Dordrecht, pp 595–622
Sasikala K, Ramana CV, Raghuveer Rao P, Subrahmanyam M (1990) Effect of gas phase on the
photoproduction of hydrogen and substrate conversion efficiency in the photosynthetic bacterium
Rhodobacter sphaeroides O.U. 001. Int J Hydrogen Energy 15(7):795–797
Sasikala K, Ramana CV (1991) Photoproduction of hydrogen from waste water of a lactic acid
fermentation plant by a purple non-sulfur photosynthetic bacterium Rhodobacter sphaeroides
O.U. 001. Indian J Exp Biol 29:74–75
Shapleigh JP (2008) Dissimilatory and assimilatory nitrate reduction in the purple photosynthetic
bacteria. In: Hunter CD, Daldal F, Thurnauer MC, Beatty JT (eds) Advances in photosynthesis
and respiration, vol 28, The purple phototrophic bacteria. Springer, Dordrecht, pp 623–642
Singh SP, Srivastava SC, Pandey KD (1994) Hydrogen production by Rhodopseudomonas at the
expense of vegetable starch, sugarcane juice and whey. Int J Hydrogen Energy 19(5):437–440
Tabita FR (1995) The biochemistry and metabolic regulation of carbon metabolism and CO2 fixa-
tion in purple bacteria. In: Blankenship RE, Madigan MT, Bauer CE (eds) Anoxygenic photo-
synthetic bacteria. Kluwer, Dordrecht, pp 885–914
Tao Y, He Y, Wu Y, Liu F, Li X, Zong W, Zhou Z (2008) Characteristics of a new photosynthetic
bacterial strain for hydrogen production and its application in wastewater treatment. Int J
Hydrogen Energy 33(3):963–973
Türkaslan S, Yigit DÖ, Aslan K, Eroğlu I, Gündüz U (1998) Photobiological hydrogen production
by Rhodobacter sphaeroides O.U. 001 by utilization of waste water from milk industry. In:
Zaborsku OR (ed) Biohydrogen. Plenum, New York
Ugwu CU, Aoyagi H, Uchiyama H (2007) Photobioreactors for mass cultivation of algae. Bioresour
Technol 99:4021–4028
Uyar B, Eroğlu I, Yücel M, Gündüz U, Türker L (2007) Effect of light intensity, wavelength and
illumination protocol on hydrogen production in photobioreactors. Int J Hydrogen Energy
32:4670–4677
Vignais PM (2008) Regulation of hydrogenase gene expression. In: Hunter CD, Daldal F, Thurnauer
MC, Beatty JT (eds) Advances in photosynthesis and respiration, vol 28, The purple phototrophic
bacteria. Springer, Dordrecht, pp 743–757
Vincenzini M, Marchini A, Ena A, De Philippis R (1997) H2 and poly-b-hydroxybutyrate, two
alternative chemicals from purple non sulfur bacteria. Biotechnol Lett 19(8):759–762
Wakayama T, Miyake J (2002) Light shade bands for the improvement of solar hydrogen produc-
tion efficiency by Rhodobacter sphaeroides RV. Int J Hydrogen Energy 27:1495–1500
Yagi K, Maeda I, Idehara K, Miura Y, Akano T, Fukatu K, Ikuta Y, Nakamura HK (1994) Removal
of inhibition by ammonium ion in nitrogenase-dependent hydrogen evolution in a marine
photosynthetic bacterium Rhodopseudomonas sp. strain W-1S. Biochem Biotech 45/46(1):
429–436
Yetis M, Gündüz U, Eroğlu I, Yücel M, Türker L (2000) Photoproduction of hydrogen from sugar
refinery wastewater by Asada Y, Miyake J O.U. 001. Int J Hydrogen Energy 25:1035–1041
Zannoni D, Schoepp-Cothenet B, Hosler J (2008) Respiration and respiratory complexes. In:
Hunter CD, Daldal F, Thurnauer MC, Beatty JT (eds) Advances in photosynthesis and respira-
tion, vol 28, The purple phototrophic bacteria. Springer, Dordrecht, pp 537–561
4 Hydrogen Production: Photofermentation 75

Zhu H, Suzuki T, Tsygankov AA, Asada Y, Miyake J (1999a) Hydrogen production from tofu
wastewater by Rhodobacter sphaeroides immobilized in agar gels. Int J Hydrogen Energy
24:305–310
Zhu H, Wakayama T, Suzuki T, Asada Y, Miyake J (1999b) Entrapment of Rhodobacter sphaeroides
in cationic polymer/agar gels for hydrogen production in the presence of NH4+. J Biosci
Bioeng 88(5):507–512
Zhu H, Wakayama T, Asada Y, Miyake J (2001) Hydrogen production by four cultures with participa-
tion by anoxygenic photosynthetic bacterium and anaerobic bacterium in the presence of
NH4+. Int J Hydrogen Energy 26(11):1149–1154
Zinchenko VV, Kopteva AV, Belavina NV, Mitronova TN, Frolova VD, Shestakov SV (1991) The
study of Rhodobacter sphaeroides mutants of different type with derepressed nitrogenase.
Genetika 27(6):991–999
Zinchenko VV, Babykin M, Glaser V, Mekhedov S, Shestakov SV (1997) Mutation in ntrC gene
leading to the derepression of nitrogenase synthesis in Rhodobacter sphaeroides. FEMS
Microbiol Lett 147:57–61