Applied Energy 156 (2015) 174–184

Contents lists available at ScienceDirect

Applied Energy

journal homepage: www.elsevier .com/ locate/apenergy

Hydrogen production from glycerol by Escherichia coli and otherbacteria: An overview and perspectives

http://dx.doi.org/10.1016/j.apenergy.2015.07.0090306-2619/� 2015 Elsevier Ltd. All rights reserved.

⇑ Corresponding author. Tel.: +374 60710520.E-mail address: Trchounian@ysu.am (A. Trchounian).

Karen Trchounian a,b, Armen Trchounian a,⇑a Department of Microbiology, Plants and Microbes Biotechnology, Yerevan State University, 1 A. Manoukian Str., 0025 Yerevan, Armeniab Department of Biophysics, Faculty of Biology, Yerevan State University, 1 A. Manoukian Str., 0025 Yerevan, Armenia

h i g h l i g h t s

� Glycerol is a perspective carbon containing source to produce H2 by Escherichia coli.� Metabolic pathways, responsible hydrogenases and dependence of H2 production on external factors are summarized.� H2 production can be improved by glycerol added to glucose containing medium.� H2 production biotechnology would be further developed using glycerol as a feedstock.

a r t i c l e i n f o

Article history:Received 20 March 2015Received in revised form 25 June 2015Accepted 4 July 2015

Keywords:Biohydrogen productionGlycerolDark fermentationHydrogenasesE. coli and other bacteria

a b s t r a c t

Hydrogen (H2) is a clean, effective and renewable fuel which can be produced by different methodsincluding biological ones, namely fermentation and biophotolysis. To improve fermentative H2 produc-tion the strategies, implicating use of by-products, utilization of carbon containing organic wastes andoptimization of biotechnology process conditions, are developed. Glycerol, a biodiesel by-product, canserve as a cheap carbon containing source to produce H2 by Escherichia coli. Recent data on metabolicpathways, responsible hydrogenases and dependence of H2 production on external factors during glyc-erol fermentation are summarized. The strains are constructed to enhance H2 yield. The mixed carbonsources (glycerol and glucose) fermentation is a novel approach: glycerol added to glucose containingmedium increases H2 production; different carbon sources comprising wastes can be used. H2 productionfrom glycerol by different bacteria is overviewed; cultures types, new technologies and optimal condi-tions, purification of H2 and developing bioreactors are highlighted. All of these are significant for furtherdeveloping H2 production biotechnology from glycerol and perspective for applied energy systems.

� 2015 Elsevier Ltd. All rights reserved.

Contents

1. Hydrogen as a perspective fuel and estimation of its global market . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1752. Biological methods in global hydrogen production and dark fermentation by Escherichia coli . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1753. Novel discovery of glycerol fermentation by E. coli and biohydrogen production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1774. Glycerol fermentation by E. coli: pathways, different from sugar fermentation, and bioenergetic advantage . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1775. Hydrogenases for hydrogen production by E. coli during glycerol fermentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1776. Proton ATPase, proton-motive force and hydrogenases. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1787. Metabolic engineering of E. coli to improve hydrogen production from glycerol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1788. Hydrogen production by E. coli during mixed carbon sources (glycerol and glucose) fermentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1799. Glycerol fermentation and hydrogen production by different bacteria; cultures types, new technologies and optimal conditions . . . . . . . . . . 179

Fadh

K. Trchounian, A. Trchounian / Applied Energy 156 (2015) 174–184 175

10. Purification of biohydrogen produced from glycerol and developing bioreactors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18111. Concluding remarks and perspectives for developing effective systems for energy production. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181

Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181

1. Hydrogen as a perspective fuel and estimation of its globalmarket

Molecular hydrogen (H2) is a valuable gas as an effective, cleanand renewable energy source and as feedstock for chemical, food,pharmaceutical and some other industries, metallurgy, productionof electronic devices. H2 is also used in space exploration, espe-cially in space shuttles. Importantly, high energy (�142 kJ g�1;�3.5 time higher than oil) is released and only water (H2O) butnot carbon dioxide is formed during H2 combustion [1,2]:

2H2 þ O2 ! 2H2O:

H2 has low radiation level and relatively safe: the explosion limitsby volume for H2 in air of 18.3–59% are much higher than thosefor gasoline (<3.3%) and natural gas (<14%) [3].

Moreover, H2 is not available in nature but it can be producedfrom unlimited sources like water or renewable ones like biomassand carbon containing wastes. Therefore, demand on H2 produc-tion has increased considerably in recent years and H2 can ulti-mately replace oil and natural gas.

H2 is produced in both small-scale and large-scale volumes.World production of H2 is more than 50 million tons [4] andaccording to the International Energy Agency estimated to be of�65 million tons by production capacities in China, USA,European and other countries [5]. H2 production is increased overthe world rapidly (�6–10%); it will decrease costs of H2 productionwhich is competitive with oil and natural gas [3]. This can improvethe efficiency and reliability of energy systems. H2 is already usedin gas stations (>200 stations distributed in North and SouthAmerica, Europe, Asia and Australia and �1000 stations will beconstructed by 2025) for engines of different transport vehicles(cars, buses and ships are made by different companies likeDaimler, Ford, Nissan, Mercedes, BMW etc) and in fuel cells to gen-erate electric power [3,6]. As for growing economy, proposed glo-bal energy demand by 2050 will be tripled, and produced fossilfuels cannot satisfy energy market. Thus, very soon H2 and fuelcells can become one of the main components in global energyeconomy.

ig. 1. Comparison of chemical and biological methods for H2 production with theirvantages (high yield, low temperature, low cost) and disadvantages (low yield,

igh temperature, high cost).

2. Biological methods in global hydrogen production and darkfermentation by Escherichia coli

Different methods have been developed for H2 production.Electrolysis of water, steam reforming of methane or hydrocar-bons, gasification of coal, oxidation of oil and natural gas, andauto-thermal processes are well-known and largely used methodsfor global H2 production, but not cost-effective due to high temper-ature requirements (>700 �C), non-friendly method for environ-ment (greenhouse gas emission) and non-renewable one (exceptelectrolysis of water) (Fig. 1) [1,2]. Biological production of H2

(biohydrogen) includes dark and photo-fermentation of carboncontaining substrates, biophotolysis of water by bacteria andmicroalgae (see Fig. 1). The theoretical maximum yields for H2 pro-duction by these methods can be different according to thewell-known reactions for:

dark fermentation of sugars ðglucose; C6H12O6Þ � C6H12O6

þ 6H2O! 12H2 þ 6CO2;

photo-fermentation of organic acids ðacetic acid;C2H4O2Þ� C2H4O2 þ 2H2Oþ light! 4H2 þ 2CO2 or

ðsuccinic acid;C4H6O4Þ � C4H6O4 þ 4H2Oþ light! 7H2 þ 4CO2;

biophotolysis of water� 2H2Oþ light! 2H2 þ 6O2ðdirectÞ or

Fig. 2. Combined putative metabolic pathways of glycerol and glucose fermenta-tion in E. coli. The pathways are adapted from [2,18,20,23–25]. Linear arrowsindicate pathways only for glycerol fermentation, broken arrows indicate pathwaysonly for glucose fermentation, and solid broken arrows indicate pathways for bothglucose and glycerol fermentation. The end products are formatted as italics. 2PG,2-phosphoglycerate; 3PG, 3-phosphoglycerate; AcCoA, acetyl-Coenzyme A; ADP,adenosine diphosphate; ATP, adenosine triphosphate; DHA, dihydroxyacetone;DHAP, dihydroxyacetone phosphate; GAP glyceraldehyde-3-phosphate; NADH,dihydrodiphosphopyridine nucleotide; NAD+, diphosphopyridine nucleotide; PGP,1,3-diphosphate glycerate; G6P, glucose 6 phosphate; F1,6P, fructose-1,6 diphos-phate; AP, Acetyl -phosphate.

0

2000

4000

6000

8000

10000

12000

2010 2011 2012 2013 2014

Bio

dies

el, m

ilion

L

USA

EU

Brazil

Indonesia

China

Thailand

Fig. 4. Biodiesel production in developed countries (USA, EU, Brazil, Indonesia,China and Thailand) during 2010–2014. Data are from US Energy InformationAdministration, Form EIA-22 M ‘‘Monthly Biodiesel Production Report’’; USDAForeign Agricultural Service Global Agricultural Information Network Reports #NL4025, # ID1420, #BR13005, #CH14038, TH4057.

H2->2H++2e-GLYCEROL

pH 7.5

2H++2e- ->H2

Hyd-3

In

Out

Hyd-4

Hyd-1Hyd-2

GLYCEROL

A

FORMATE

GlpF

FORMATE

2H++2e- ->H2

Hyd-3

In

Out

Hyd-4

Hyd-1

Hyd-2

B

GLYCEROL

GlpF

176 K. Trchounian, A. Trchounian / Applied Energy 156 (2015) 174–184

12H2Oþ light! 12H2 þ 6O2 ðindirectÞ:

There is also the method of bioconversion of carbon oxide (CO)to H2 by bacteria when H2 is produced from water according to thereaction:

COþH2O! CO2 þH2:

Biohydrogen has significant advantages over chemical methodssince it could be performed at relatively low temperatures(25–37 �C), atmospheric pressure and with relatively high rates(see Fig. 1). This can reduce the costs for H2 production. There isone important argument: biohydrogen production creates anopportunity to develop decentralized energy systems when energyproduction plants can be located not so far from carbon containingresources. Moreover, generation of different contaminants withhigh level, including volatile fatty acids, can be avoided. Therefore,biohydrogen production will be increased in upcoming years.

Among biological methods dark fermentation with microbes,including bacteria, is the best promising one [2,7], andEscherichia coli is well-studied facultative anaerobic bacteria [8,9]performing mixed-acid fermentation of carbohydrates (sugars),which are common and largely present and stored in plant tissues,and of other organic carbon sources – different wastes. This bac-terium is easy for manipulation in lab and large-scale conditionsbecause of difficulty in maintaining strict anaerobic conditions.Fermentation pathways in E. coli are complex (Fig. 2) [10]. Butdue to known genome sequence and analysis (defined transcrip-tion and translation systems, broad spectrum of already con-structed mutant strains), metabolic flux analysis, as well as theability of fast growth on cheap substrates (wastes), they can begenetically and metabolically engineered for constructing highlyeffective strains with increased H2 production [8], compared tothe theoretical maximum yield of 12 mol H2 per mole of glucose(see reaction above) [9,11,12]. However, sugars are expensive,and this maximum is hard to be reached by bacteria even of exter-nal energy supply; 4 mol H2 per mole of glucose is considered asnowadays reality. Moreover, some mechanical, thermal (<150 �C),alkaline or acidic, ultrasonic or other pretreatment of differentresources and wastes is recommended; but this requires additionalcosts. Interestingly, dark fermentative production of H2 is alsomore efficient and rate is higher than H2 production by photosyn-thetic pathways; the cost of fermentative H2 production is�300-fold lower than of photosynthetic one [7]. There is also oneimportant property of E. coli – this is a non-pathogenic (condition-ally pathogenic) bacterium. And H2 production is stable for a rela-tively long period (several tens of days, depending on culture type).Thus, dark fermentation has advantages due to rapid H2 produc-tion, relatively high H2 yield, no solar light requirement, stable bac-teria and low cost.

Fig. 3. Strategies for main approaches to enhancing of H2 production by bacteria.

H2->2H++2e-GLYCEROL

pH 5.5

Fig. 5. H2 cycling and responsible Hyd-enzymes in E. coli cell membrane duringglycerol fermentation. These schemes are adapted from [2,13,25,34]. The GlpF isglycerol transporter [19]. The direction of enzyme operation to produce and/or tooxidize H2 is shown (arrows) at pH 7.5 (A) and 5.5 (B).

Different strategies are developed for the improvement of fer-mentative H2 production and lowering the costs of its production[2,9,11,12], including use of by-products, utilization of carbon con-taining organic wastes, construction of effective bacterial strainsand optimization of biotechnology process conditions (Fig. 3).The most important problem is to find out a cheap substrate for

K. Trchounian, A. Trchounian / Applied Energy 156 (2015) 174–184 177

fermentation and to construct effective strains with stable respon-sible enzymes with high activity of H2 production.

3. Novel discovery of glycerol fermentation by E. coli andbiohydrogen production

It has been recently discovered by Gonzalez’s group [13] thatglycerol (C3H8O3) can be fermented by E. coli; and H2 is detectedamong the end products during glycerol fermentation at slightlyacidic pH (pH 6.3). Furthermore, glycerol fermentation and H2 lib-eration have been also revealed by our group [14] at slightly alka-line pH (pH 7.5) in batch culture. This is very intriguingphenomenon applicable in clean and sustainable energy produc-tion because (1) glycerol, a biodiesel by-product (1 kg crudeglycerol per 10 kg biodiesel), or product of biomass, vegetable oilsand animal fats, is a very cheap (prices of crude glycerol is 3–10cents/lb and reformed glycerol is 40–50 cents/lb); (2) glycerol islargely available (hundreds of plants in different countries world-wide in North and South America, Europe and Asia withlarge-scale biodiesel production (Fig. 4); average annual growthof biodiesel market is estimated to be of 42%; crude glycerol couldbe reformed into pure one) and (3) glycerol is effective renewablecarbon containing source for obtaining biomass, bacterial fermen-tation and H2 production, compared to sugars and other organiccarbon sources [1,2,15].

Crude glycerol from biodiesel as well as pure glycerol hasrapidly become a competitive raw material to produce differentfuels and chemicals, compared with sugars (glucose) [11–13].The crude glycerol has an acidic (pH 2.4) to alkaline pH (pH �12)and contains 45–88% (v/v) glycerol; however, pure glycerol(>98%) has an acidic pH (pH �5.5). Glycerol is completely solublein water (>500 g L�1 at 20 �C) and stable. These properties of crudeand pure glycerol are important for application to produce H2. Thehigh degree of reduction of carbon atoms in glycerol confers theability to produce H2 at higher yields when compared with glucose[15,16]. On the other hand, the theoretical yield is 7 mol H2 forevery mole glycerol according to the reaction [17]:

C3H8O3 þ 3H2O! 7H2 þ 3CO2:

It is difficult to suggest so high yield in the application of glycerol toproduce H2 by bacteria, however it can be reached. But H2 producedby E. coli has negative impact on the cell growth and glycerol fer-mentation [18]. To avoid this effect, gaseous H2 removal and cumu-lating can be essential steps to improve H2 production and shouldbe used in continuous culture.

An updated overview of H2 production from glycerol by E. coliand other bacteria with perspectives for developing of effectivesystems of energy production is provided in the present paper.The glycerol fermentation pathways, the mechanisms of hydroge-nase (Hyd) enzymes biosynthesis, determination of their specific

Table 1H2 production by E. coli Hyd enzymes during glycerol fermentation in batch culture at dif

Characteristics or properties of Hyd enzymes

H2 production rate in wild type, mmol H2 min�1 (g dry weight)�1

Responsible Hyd enzymesRecycling H2

Inhibition by DCCDRequirement of the FOF1-ATPaseCoupled generation of a proton-motive forceIntracellular pHSensitivity to osmotic stress

a For references, see the text.b The signs ‘‘+’’ or ‘‘�’’ represent the presence or the absence of the characteristics, re

activity and operation mechanisms during glycerol fermentationat different environmental conditions, especially pH, cultures typesand new technologies based on glycerol would be of greatimportance.

4. Glycerol fermentation by E. coli: pathways, different fromsugar fermentation, and bioenergetic advantage

Glycerol can permeate into the cell of E. coli [19]. The glycerolfermentation pathways within the cell are relatively simpleinter-linked biochemical reactions known from glycolysis [2,20–23].But unlike glycolytic pathways they have a unique coupled reac-tion on the level of phosphoenolpyruvate (PEP) conversion intopyruvate (PYR) generating a cycle pathway (see Fig. 2). These path-ways of glycerol fermentation can be presented as shown, at thestage of PEP some intermediates may be used for succinate forma-tion, whereas all other end products, including formic acid(HCOOH), are formed from PYR [10,24,25] (see Fig. 2). The glycerolfermentation pathways are not clear yet although succinic, aceticand formic acids and ethanol as well as H2 are shown to be pro-duced but lower acetic and less lactic acids could be ended fromglycerol [18]. Especially, H2 can be produced by E. coli from formate[26–28]. Moreover, additional ATP could be obtained (see Fig. 2):glycerol fermentation has a bioenergetic advantage. These differ-ences between the fermentations of glycerol and sugars by E. coliare important enough and could reveal novel pathways to regulateH2 production. Importantly, H2 generated from formate could beevolved whereas CO2 (see Fig. 2) might be used for cell growth[13,29]. No other gaseous products can be ended from darkfermentation.

Interestingly, glycerol concentration of 10 g L�1 has been shownto be optimal for E. coli cell growth and H2 production: a higherconcentration of glycerol can decrease H2 production rate and yield[9,14,18,22]. This is likely to the effect of sugars (glucose) concen-tration on H2 production by E. coli when the highest yield wasobtained with culture limited for glucose [9].

5. Hydrogenases for hydrogen production by E. coli duringglycerol fermentation

E. coli processes multiple membrane-associated [Ni–Fe]-containing Hyd enzymes to produce H2 due to simple redoxreaction [2,16,25,30,31]:

2Hþ þ 2e� ! H2:

These enzymes are responsible for the stable H2 production.However, these enzymes not only produce but also oxidize H2

[11,16,25,32–35], and therefore H2 cycling is suggested (Fig. 5).The latter is a novel phenomenon in the bioenergetics ofmixed-acid fermentation and can play a role in maintaining a

ferent pHs.

H2 productiona

pH 7.5 pH 5.5

4.57 6.79Hyd-2 is major, and Hyd-1 is less Hyd-3+b +� ++ ++ +7.0 5.7Hyd-3 and Hyd-4 Hyd-1, Hyd-3 and Hyd-4

spectively.

178 K. Trchounian, A. Trchounian / Applied Energy 156 (2015) 174–184

proton-motive force [31]. Interestingly, the standard redox poten-tial of 2H/H2 couple is �414 mV: this value depends on H2 steadystate concentration and can be lowered to �270 mV [16]. Therange of redox potential is important in determining H2 productionand oxidation mode of Hyd enzymes. It is also important thatgenes coding these Hyd enzymes and proteins participating intheir maturation are known, mechanisms of their expression andregulation are complex and have not been understood well[10,32,36–42]. However, manipulating with these genes is aneffective tool to regulate and especially to increase H2 production.The problem is to determine Hyd enzymes, which are responsiblefor H2 production during glycerol fermentation, and to reveal theirdependence on external and other factors.

The principal findings on Hyd enzymes and H2 production areclearly shown with E. coli during glycerol fermentation and sum-marized in Table 1.

E. coli wild type and mutant strains with defects in differentgenes coding Hyd enzymes and their maturation have differentHyd enzymatic activity and H2 production rate depending on pH.Hyd-2 (hyb) mostly and Hyd-1 (hya) partially are responsible forH2 production at slightly alkaline pH (pH 7.5); Hyd-3 (hyc) andHyd-4 (hyf) can work in H2 oxidizing mode (see Fig. 5) [2,14,34].H2 producing Hyd-2 activity is confirmed by the other groups[43,44] and by the results with E. coli hya hyc or hyb hyc doublemutants [45]. In the stationary growth phase and in the absenceof Hyd-3, Hyd-2 can be forced to evolve H2 at pH 7.5 [45]. Thus,the major contribution of Hyd-2 to H2 production during glycerolfermentation at pH 7.5 resulted from changed metabolism and sur-prisingly influenced on proton reduction. At the same time, H2 pro-duction by E. coli using glycerol is �1.5-fold higher at pH 5.5 thanat pH 6.5 and Hyd-3 is a major H2 producing enzyme during glyc-erol fermentation at acidic pH (see Fig. 3) [34]. The latter seems tobe clear due to the result decade ago that in addition to pH 6.5Hyd-3 becomes a major H2 producing enzyme also at pH 7.5 duringglucose fermentation when formate (30 mM) is supplemented[26,27]. All these findings with Hyd enzymes, which are responsi-ble for H2 production at different pHs, are significant for manipula-tion to produce substantial amount of H2 depending on externalconditions. Otherwise it is possible to have incorrect conclusionon the absence of substantial H2 production by E. coli using differ-ent substrates when modified strain (hyaB hybC hycA) lacking Hydenzymes was used and aerobic conditions were applied [46].

It is very interesting that H2 production rate determined in liq-uid and gaseous phases was markedly different during glycerol butnot glucose fermentation by E. coli: a significantly (30-fold) moreH2 is observed in liquid [46]. This suggests an increased role ofH2 cycling upon glycerol fermentation.

6. Proton ATPase, proton-motive force and hydrogenases

E. coli wild type and atp mutant lacking the proton FOF1-ATPasealso have different Hyd enzymatic activity and H2 production ratedepending on pH [47]. Hyd activity was inferred by native PAGEelectrophoresis to be dependent on the active FOF1-ATPase duringglycerol as well as glucose fermentation, especially at extremepHs [48]. This might be resulted from link between Hyd enzymesand the FOF1-ATPase, physiological role of which is in maintaininga proton-motive force, key intermediate in energy conversion incells [31,49].

H2 production is inhibited by N,N0-dicyclohexylcarbodiimide(DCCD), inhibitor for the FOF1-ATPase [49]: DCCD inhibited H2 pro-duction at acidic (pH 5.5) but not slightly alkaline (pH 7.5) medium(see Table 1). It should be noted that DCCD inhibition was reversedduring glucose fermentation [34,49]. Moreover, under glycerol fer-mentation Hyd-2 activity could disappear by protonophore actionindicating the requirement of a proton-motive force [50,51].

Then, proton-motive force is generated at different pHs. Theintracellular pH and, hence, proton-motive force were lower atpH 7.5 compared with those during glucose fermentation [50].Proton motive force generation is changed in E. coli hypF mutantlacking all Hyd enzymes [50] due to the effects on theFOF1-ATPase too [52]. Therefore, Hyd impact on an overallproton-motive force generation is disclosed; H+ pumping by Hydis suggested.

H2 production is inhibited completely by 0.5 mMN-ethylmaleimide (NEM) [50,51], an inhibitor of glycerol kinase[53], which catalyzes the formation of GP and then DHAP (seeFig. 2). However, NEM had no effect on membrane potential [51]suggesting different mechanism for NEM action on glycerol meta-bolism and H2 production by E. coli. Moreover, H2 production isvery sensitive to diphenylene iodonium (Ph2I), an inhibitor of[Ni–Fe]-Hyd enzymes in the other bacteria [54]: even 1 nM Ph2Iinhibited completely H2 production [51]. These results indicatethe absolute role of known Hyd enzymes in H2 production duringglycerol fermentation.

Besides, H2 production from glycerol is sensitive to osmoticstress provided by sucrose (see Table 1): Hyd-3 and Hyd-4 are sen-sitive to osmotic stress but at different pHs, whereas Hyd-1 isosmosensitive at low pH [55].

Importantly, these data have been mainly obtained by usingelectrochemical determination of H2 with a pair of oxidation–reduction (redox) titanium-silicate and platinum electrodes, asdescribed elsewhere [14,22,30,35,47,56–58]. In contrast to plat-inum electrode, titanium-silicate electrode is not sensitive to H2

and oxygen; therefore this pair of redox electrodes allows detect-ing exclusively H2 in liquids during fermentation under anaerobicconditions (in the absence of oxygen). Various controls have ruledout interference by the other fermentation end products (seeFig. 2). This approach is close to the method with Clark-type elec-trode employed by different groups in worldwide labs [46,59–62]:a good correlation between redox potential measuring differentredox electrodes readings difference and H2 production was shownin liquids. These redox electrodes have a high sensitivity to H2, along life expectancy and can be applied to control H2 productionin different systems. Moreover, the results obtained with redoxelectrodes have been confirmed by chemical (with using perman-ganate in sulfur acid [63,64]) and other (for instance, withDurham tubes [65]) methods.

There is a problem with comparison of data on H2 yield sincethis parameter is given by different groups in moles of H2 per moleof substrate, dry weight of biomass or volume of culture.

7. Metabolic engineering of E. coli to improve hydrogenproduction from glycerol

The progress with glycerol fermentation pathways andenhanced H2 production from glycerol has been achieved by meta-bolic engineering of E. coli [2,9,14,18,21,43]. Potential strategies forincreasing H2 production by Hyd enzymes have been outlined andwhole-cell systems and cell-free systems were compared recentlyby Wood’s group [66]. It is possible to redirect metabolic pathways,to induce the DHAP production pathway, to block lactic, aceticacids and ethanol production (see Fig. 2). Indeed, Hu and Wood[67] obtained an improved strain with H2 production of 0.68 mmolH2 L�1 h�1 in glycerol medium; this was 20-fold increased H2 pro-duction compared with precursor one. A new strain with 5-foldenhanced H2 yield has been created since the old one for H2 pro-duction from glucose was not suitable [68]. Moreover, Wood withco-workers [43,69] have identified uncharacterized genes (54% ofE. coli genome is experimentally determined; the other part isuncharacterized or computationally predicted) whose inactivation

Table 2Comparison of H2 production by E. coli from crude and pure glycerol as substrates for dark fermentation.

Type of glycerol Culture medium composition and pH Characteristics of H2 production References

Crude (80% glycerol) MOPSa minimal media; 10 g L�1 tryptone, 5 g L�1 yeast extract (pH 6.3) Maximal yield of �13 mmol H2 L�1 at 10 g L�1 glycerol [72]Pure (>98%) MOPS minimal media; 10 g L�1 tryptone, 5 g L�1 yeast extract (pH 6.3) Maximal yield of �11.5 mmol H2 L�1 at 10 g L�1 glycerol [72]

a 3-(N-morpholino) propane sulfonic acid.

K. Trchounian, A. Trchounian / Applied Energy 156 (2015) 174–184 179

is beneficial for H2 production from glycerol. They found out thestrains with ability for �1.6-fold higher H2 production and�2.1-fold higher H2 yield. The yield of H2 reached was 95% of thetheoretical maximum but this was not perfection. However, theconditions, especially pH, are important in using these mutantstrains for H2 production to have maximal activity of responsibleHyd enzymes (see Table 1).

The results can be explained due to that the strains producemore formate and less ethanol and acetate than the wild type dur-ing glycerol fermentation (see Fig. 2), as shown [18,45], and thetranscription of genes for Hyd enzymes might be increased asdetermined for hycE during glucose fermentation [70].Interestingly, the mutant strains grow faster on glycerol than par-ent strains so they achieve a reasonable anaerobic growth rate[67,69] which is important for biomass synthesis in biotechnology.

Interestingly, a new model for metabolic engineering based onan in vitro synthetic enzymatic pathway has also been developedto enhance H2 production from biomass in a distributed, carbon–neutral and low-cost manner [71]. The latter resulted in anincrease of 67-fold in H2 production by E. coli. This approach mightbe applied for effective H2 production from glycerol.

This is a good basis to consider glycerol as a perspective sub-strate to produce H2 with high rate and yield due to effective exter-nal factors and metabolic engineering as well.

8. Hydrogen production by E. coli during mixed carbon sources(glycerol and glucose) fermentation

It is interesting that both crude and pure glycerol can result in ahigh H2 yield by E. coli; however, use of crude glycerol instead ofpure glycerol as a carbon source did not change H2 production(Table 2) [72]. Given this finding, the useful approach to increaseH2 production would be determining H2 production detected dur-ing the mixed carbon sources (pure glycerol and glucose)fermentation:

� Glycerol could be fermented in the presence of sugars (glucose)at slightly alkaline pH, and DCCD inhibited H2 production in apH-dependent manner; especially, total inhibition of H2 pro-duction was observed in fhlA (gene for transcriptional activatorof hyc [73]) mutant at pH 7.5 and pH 5.5 suggesting that FhlAmight directly relate to the FOF1-ATPase. Optimal concentrationwas 10 mM glycerol. In addition, Hyd-4 activity mainly and

Table 3Changes in H2 production rates by E. coli wild type strain (BW25113) upon glyceroladded into the peptone medium with glucose. H2 production by the cells grown onmixed carbon sources (glucose and glycerol) was assayed with either adding glucose(glucose assay) or glycerol (glycerol assay) in the same concentrations as during thegrowth [57]. For comparison, H2 production rates by E. coli grown on sole glucose orsole glycerol added into the peptone medium were determined. The standard errorswere not more than 3%.

pH H2 production rate (mmol H2 min�1 (g dry weight)-1)

0.1% glucose 0.5% glycerol 0.1% glucose + 0.5% glycerol

Glucose assay Glycerol assay

7.5 8.43 3.70 15.54 9.60

Hyd-2 activity to some extent to produce H2 have been revealedat low pH [42,74]. Moreover, Hyd-1 and Hyd-2 formed H2 oxi-dizing activity at pH 7.5.

� Glycerol added into the medium with glucose increased signif-icantly (2- and more fold) H2 production [57]; this was observedat slightly alkaline pH (Table 3). However, the decrease in H2

production was shown for acidic pH [57]. Thus, glucose andglycerol concentrations might have significant role in H2 pro-duction rate changing.

� Formate added into the medium with glycerol stimulated E. coligrowth and H2 production at different pHs [75,76].Interestingly, formate (10 mM) recovered H2 production ofHyd-2 (hybC) or Hyd-4 (hyfG) mutants depending on pH[75,76]. These results highlight the key role of Hyd-3 at bothpH 6.5 and pH 7.5, as well as the role of Hyd-2 and Hyd-4 atpH 7.5 for H2 production by E. coli during glycerol and formateco-fermentation.

The mixed carbon sources fermentation is novel phenomenon,its pathways are complex but particularly clear as represented(see Fig. 2); a detailed study is required. However, this approachhas been already employed with different fermentation substratesand different bacteria [2,11,12]. In addition to strategies alreadydesigned (see Fig. 3), the mixed carbon sources fermentation mightbe used in biotechnological applications for pre-cultivation of bac-terial cells, regulation of Hyd enzymes activity towards enhancingH2 production and utilizing glycerol and other carbon sources con-taining industrial, agricultural and food (kitchen) wastes as well aswastewater. There are both economic (cheap wastes) and environ-mental (wastes utilization) benefits.

9. Glycerol fermentation and hydrogen production by differentbacteria; cultures types, new technologies and optimalconditions

Glycerol fermentation has been established in different culturestypes - batch culture with changing conditions and continuoussteady-state culture using different approaches with several bacte-ria like Klebsiella pneumoniae [77], Halanaerobium saccharolyticum[78], Clostridium sporogenes [46], Enterobacter sp. and Citrobacterfreundii [79] and many others [46,80–89]; and H2 is among theend products of higher yield (Table 4). Probably, glycerol can befermented by different bacteria due to specific enzymes namelyglycerol dehydrogenase catalyzing the first steps of fermentationpathways to DHA (see Fig. 2). H2 production from glycerol was con-centration dependent and depended on crude or pure glycerol aswell as on medium composition, adaptation time and other condi-tions. But higher glycerol concentration has been shown to declineH2 yield enhancing by Bacillus thuringiensis [88]; this might berelated to H2 negative impact on glycerol fermentation, as sug-gested with E. coli (see above). It was reported that the productionof H2 decreased with an increase of the concentration of biodieselwastes [90]. Therefore, dilution of glycerol or continuous culturemight increase H2 production. Moreover, glycerol based on organicwastes has been evaluated for H2 production by Clostridium sp. orEnterobacter aerogenes using different – response surface method-ology [83,84,86] (see Table 4). Recently, immobilization of E.

Table 4H2 production from glycerol by different bacteria during dark fermentation; culture types used.

Bacteria Culture type; fermentation method pHa Characteristics and efficiency of H2 production fromglycerol

References

Bacillus thuringiensis Batch and continuous culture 7.0 Increased production yield of 0.646 and 0.748 molH2/mol glycerol by 20 g L�1 glycerol in minimal saltmedium (2% ammonium nitrate and 1% sodiumnitrate, respectively) and 2.2 mol H2/mol glycerol incomplex media in batch culture with small volume(250 mL); 0.393 mol H2/mol glycerol in theimmobilized culture on banana leaves (1% sodiumnitrate)

[88]

Clostridium pasteurianum Batch and continuous culture 7.0 Improved production rate of 256 mL H2 L�1 h�1 andyield of 1.11 mol H2/mol glycerol by 10 g L�1 crudeglycerol in batch culture; production rate of 103 mLH2 L�1 h�1 and yield of 0.5 mol H2/mol glycerol by10 g L�1 pure glycerol or 166 mL H2 L�1 h�1 andyield of 0.77 mol H2/mol glycerol by 10 g L�1 crudeglycerol in continuous culture

[86]

Clostridium sporogenes Batch culture 7.3 High production rate of 1.50 and 1.42 mmol H2

L�1 h�1 by 22 g L�1 crude glycerol in liquid and ingaseous phases in flasks (200 mL) containing Luria–Bertani medium with phosphate buffered salinetrace elements, respectively

[46,82]

Enterobacter aerogenes Batch and continuous aerobic culture, bioreactor Increased H2 production yield by optimization ofthe medium composition (15 g L�1 pure glyceroland different salt contents) in 500 mL bioreactor

[86,87]

Enterobacter aerogenes Response surface culture 5.0 High productivity from pure glycerol of 9 mmol H2

L�1 h�1 and from crude glycerol of 6.2 mmol H2

L�1 h�1; stability and reusability by theimmobilized cells

[83,84]

Halanaerobium saccharolyticum Batch culture 7.0–7.4 Effective production of 0.6 mol H2/mol glycerolunder 2.5 g L�1 pure glycerol

[78,90,91]

Klebsiella pneumoniae Batch culture, bioreactor Increased H2 production yield by optimization ofthe medium composition (11 g L�1 glycerol anddifferent salt contents)

[77]

Klebsiella sp. Batch culture, response surface culture 8.0 Increased H2 production yield by optimization ofthe medium composition (20 g L�1 crude glyceroland different salt contents for batch culture in36 mL medium in 60 mL serum bottle or 11 g L�1

crude glycerol and different salt contents forresponse surface culture)

[85]

a Temperature was 35–40 �C.

180 K. Trchounian, A. Trchounian / Applied Energy 156 (2015) 174–184

aerogenes has been optimized to develop biocatalyst for continuousH2 production from glycerol [83–85]. In addition, B. thuringiensis incontinuous culture by immobilization on cheap and available plantwaste has been shown to give H2 yield from glycerol to be suitablein large-scale H2 production [88]. The last methods are extensivelydeveloped for optimizing the biotechnology conditions and forlong period H2 production due to biomass retention.

Importantly, fermentation pathways in H. saccharolyticum havebeen reconstructed according to genome sequence analysis; multi-ple [Fe–Fe]-Hyd enzymes of different type were suggested; two ofthese Hyd enzymes were identified [89]. There are interesting find-ings for different bacteria on the dependence of H2 production

Fig. 6. Two stage technology for enhanced H2 production by bacteria using darkand photo-fermentation. Dashed lines present structure (selective membrane) tosplit cultures between stages.

from concentration of glycerol, composition of culture medium,pH, temperature and other factors [46,74,86,91,92]. All these con-ditions would be useful to apply for enhanced H2 production.Interestingly, the efficiency for photo-fermentative conversion ofcrude glycerol into H2 by Rhodopseudomonas palustris is nearly90% of the theoretical maximum [80,81]. This would be studiedin order to integrate dark and photo-fermentation into atwo-stage process and to use mixed cultures with E. coli to havea high H2 yield and more effective H2 production; light and strictconditions are required (Fig. 6). Fermentative H2 producing bacte-ria can be combined with photosynthetic producers to evolve addi-tional H2 at the second stage using organic acids of fermentation(see Fig. 2) from the first stage (see Fig. 6). For this technology itwould be important to employ light and dark alternations, whichcan change H2 production during photo-fermentation [93].

Moreover, Mangahil et al. [94] have reported about H2 produc-tion from crude glycerol (pH �12; 45% glycerol) using enrichedmicrobial community with dominated Clostridium sp.; optimalconditions (pH 6.5, 37 �C and 1 g L�1 glycerol in small volume ofbatch culture (50 mL in 120 mL serum bottle) were established.Interestingly, H2 was suggested to be end product together withacetate and less butyrate; and methanol contained in crude glyc-erol was not utilized [94]. H2 production rate from crude and pureglycerol was similar each with other [94], as in E. coli (see Table 2).New technology for H2 production from glycerol containing dairywastes has been evaluated with syntrophic consortium in singlechamber microbial electrolysis cell [95].

K. Trchounian, A. Trchounian / Applied Energy 156 (2015) 174–184 181

To have reliable and comparable data, a new index of maximumspecific H2 producing activity by bacteria has been proposedrecently for mixed cultures [96]. This will be beneficial for thedeveloping of systems for energy production.

It is important to use renewable resources for dark fermenta-tion, and carbons containing organic wastes of different kind forobtaining bacterial biomass are valuable sources. Glucose, sucrose,lactose or starch mixtures [97,98], rice and wheat straws, cornstalks [99,100], cheese whey [101], and many other sources[2,9,11,12,102,103] have been shown to be useful for H2 produc-tion by E. coli and other bacteria and their mixtures during fermen-tation. Recently Kumar et al. [104] have published updatedinformation about lignocellulose H2 production by dark fermenta-tion discussing pretreatment and hydrolysis methods to achieveefficient process. But glycerol added to these sources and differentbacterial mixtures would further enhance H2 production.Searching for cheap potential waste materials is being continued.The economic advantage is obvious; but biotechnology conditionsshould be optimized for different wastes and cultures.

Besides, different groups [105–107] have reported H2 produc-tion from glycerol in batch and continuous cultures of hyperther-mophilic bacteria Thermotoga neapolitana and T. maritima(>75 �C). Interestingly, H2 production from pre-treated glycerolwas higher than that from untreated one [103]; pure glycerolwas effective [106]. Again, H2 production decrease was with anincrease in glycerol concentration [107]. However, H2 productionby these bacteria at high temperatures has low rate, and expensivenitrogen substrates (yeast extract) are required for obtaining bio-mass. Probably H2 by these bacteria is produced from formatevia hydrogenase as suggested with hyperthermophile Pyrococcusfuriosus (>80 �C) [108]. This should be further studied.

Thus, H2 production from glycerol by different bacteria has beenreported with different culture types and wastes, under new tech-nologies and optimal conditions. A progress in H2 productionbiotechnology by different bacteria using glycerol as a feedstockduring dark and photo-fermentation can be achieved with compar-ison and critical analysis of the findings for E. coli. This would be ofsignificance to design large-scale energy producing systems.

10. Purification of biohydrogen produced from glycerol anddeveloping bioreactors

There is an important problem in H2 production biotechnology:it is its purification. The H2 gas purification operational systemsand different separation membranes have been reviewed recentlyby Bakonyi et al. [109]. For separating H2 gas from gaseous mixtureof CO2/H2 (see Fig. 3 and Section 3) in batch cultures differentmembranes (inorganic, organic and mixed-matrix, non-porousand porous) have been applied [110,111]. Particularly, amorphousor crystal metallic membranes mainly based on palladium areextremely selective to H2 but they have disadvantages (fragility,hard operation, high cost) so they are less useful [112,113]. It hasbeen shown that polymeric dimethyl siloxane membranes havehigh separation selectivity of CO2/H2 mixture [111]. Membranetechnologies are compact, effective and ecologically friendly, theyhave simple operation mode and the further technologicalimprovements of different materials such as modified polymerswill increase the separation selectivity of H2 and thus will beemployed in H2 economy; effects of fermentation technology pro-cess conditions should be addressed [109,114]. These membranesare key parts of different bioreactors for H2 production.

The other problem is to develop bioreactors suitable for H2 pro-duction from glycerol. Serum bottles of small volume or lab-scalefermentors are used for batch cultures; packed-bed or upflowand other bioreactors (see Table 4) are employed for continuouscultures [109,115]. In addition, large-scale bioreactors are already

used. However, it should be required to take into account H2 pro-duction dependence on substrates (wastes) concentration; usingof immobilized cells; protection of bacteria from impurities incrude glycerol and from contamination by other bacteria; agita-tion, maintaining strict anaerobic conditions and long-term stabil-ity of biotechnology processes. Interestingly, lactic acid bacteriahave been found in bioreactors and these bacteria as contaminantscan inhibit H2 production by the other bacteria [116,117].Therefore, it is required to protect pure or mixed cultures fromcontamination during production process. All these would beimportant for further optimization of technological process condi-tions for better H2 production.

11. Concluding remarks and perspectives for developingeffective systems for energy production

The important remark is in that [Ni–Fe]-Hyd enzymes in E. coliare reversible depending on bacterial growth phase, pH and fer-mentation substrates and possibly on metabolic pathways[2,14,33,34,43,69,74,118]. Therefore by optimizing these condi-tions it would be probable to enhance H2 production from cheapsubstrate – glycerol as well as from carbohydrates (sugars) andcarbon containing organic wastes [11,12]. The economic value isadvantaged. The other remark is with some link or cross-talkbetween Hyd enzymes forming H2 cycling suggested but the mech-anisms underlying how this is controlled are still not clearly under-stood [31,55].

There are many unclear problems with glycerol fermentation byE. coli, metabolic pathways and end products, Hyd enzymes activ-ity and reversibility and their dependence on pH, osmotic stressand other external factors. Besides, H+ cycle and H2 cycling areimportant to regulate H2 production by E. coli [31]. A little is knownwith thermodynamic and kinetic analysis of H2 production.

Additional efforts are required to develop large-scale technol-ogy to produce H2 by E. coli and other bacteria using glycerol. Inthis respect, it is of interest that lab-scale two-stage anaerobicdigestion should be more productive than one-stage process forH2 production (see Fig. 6) and will recover more energy from bio-mass [119].

In summary, H2 is stated to be a promising clean fuel, and bio-conversion of glycerol to H2 has additional environmental benefitreducing greenhouse gas emission [120].

All the findings together are of significance for further develop-ment of effective H2 production biotechnology using glycerol andits mixtures as a carbon source for fermentation, applied energyproduction.

Acknowledgements

This study was supported by Research Grants from the StateCommittee of Science, Ministry of Education and Science ofArmenia, to AT (13-1F002), Armenian National Science andEducation fund (ANSEF, USA) to KT (Biotech-3460), GermanAcademic Exchange Service (DAAD, Germany) scholarships to AT(A/13/03789) and KT (A/11/79636).

References

[1] Khanna S, Goyal A, Moholkar VS. Microbial conversion of glycerol: presentstatus and future prospects. Crit Rev Biotechnol 2012;32:232–65.

[2] Trchounian A. Mechanisms for hydrogen production by different bacteriaduring mixed-acid and photo-fermentation and perspectives of hydrogenproduction biotechnology. Crit Rev Biotechnol 2015;35:103–13.

[3] Alazemi J, Ansdrews J. Automotive hydrogen fuelling stations: aninternational review. Renew Sustain Energy Rev 2015;48:483–99.

[4] Winter CJ. Into the hydrogen energy economy – milestones. Int J HydrogenEnergy 2005;30:681–5.

182 K. Trchounian, A. Trchounian / Applied Energy 156 (2015) 174–184

[5] Shubber K. Clean hydrogen fuel created with sunlight and water in newmethod. <http://www.wired.co.uk/news/archive/2013-08/05/hydrogen-gas-solar-reactor> [accessed 05.08.13].

[6] Godula-Jopek A. Introduction. In: Godula-Jopek A editor. HydrogenProduction by Electrolysis. Wiley-C-VCH Verlag GmbH & Co. 2015. p. 1–33.

[7] Hallenbeck PC. Fermentative hydrogen production: principles, progress, andprognosis. Int J Hydrogen Energy 2009;34:7379–89.

[8] Maeda T, Sanchez-Torres V, Wood TK. Metabolic engineering to enhancebacterial hydrogen production. Microb Biotechnol 2008;1:30–9.

[9] Rosales-Colunga LM, Rodriguez ADL. Escherichia coli and its application tobiohydrogen production. Rev Environ Sci Biotechnol 2015;14:123–35.

[10] Bock A, Sawers G. Fermentation. In: Neidhardt, FG, editor-in-Chief. Escherichiacoli and Salmonella. Cellular and Molecular Biology. ASM Press: WashingtonDC, 2006. <http://www.ecosal.org>.

[11] Ghimire A, Frunzo L, Pirozzi F, Trably E, Escudie R, Lens PNL, et al. A review ondark fermentative biohydrogen production from organic biomass: processparameters and use of by-products. Appl Energy 2015;144:73–95.

[12] Arimi MM, Knodel J, Kiprop A, Namango SS, Zhang Y, Geiben S-U. Strategiesfor improvement of biohydrogen production from organic-rich wastewater: areview. Biomass Bioenergy 2015;75:101–18.

[13] Dharmadi Y, Murarka A, Gonzalez R. Anaerobic fermentation of glycerol byEscherichia coli: a new platform for metabolic engineering. BiotechnolBioengineer 2006;94:821–9.

[14] Trchounian K, Trchounian A. Hydrogenase 2 is most and hydrogenase 1 is lessresponsible for H2 production by Escherichia coli under glycerol fermentationat neutral and slightly alkaline pH. Int J Hydrogen Energy 2009;34:8839–45.

[15] Clomburg JM, Gonzalez R. Anaerobic fermentation of glycerol: a platform forrenewable fuels and chemicals. Trends Biotechnol 2013;31:20–8.

[16] Thauer RK, Kaster AK, Goenrich M. Energy conservation in chemotropicanaerobic bacteria. Bacteriol Rev 1977;41:100–80.

[17] Nahar G, Dupont V. Hydrogen via steam reforming of liquid biofeedstock.Biofuels 2012;3:167–91.

[18] Murarka A, Dharmadi Y, Yazdani SS, Gonzalez R. Fermentative utilization ofglycerol by Escherichia coli and its implications for the production of fuels andchemicals. Appl Environ Microbiol 2008;74:1124–35.

[19] Stroud RM, Miercke LJW, O’Connell J, Khademi S, Lee JK, Remis J, et al.Glycerol facilitator GlpF and the associated aquaporin family of channels.Curr Opin Struct Biol 2003;13:424–31.

[20] Cintolesi A, Comburg JM, Rigou V, Zygourakis K, Gonzalez R. Quantitativeanalysis of the fermentative metabolism of glycerol in Escherichia coli.Biotechnol Bioengineer 2011;109:187–98.

[21] Ganesh I, Ravikumar S, Hong SH. Metabolically engineered Escherichia coli asa tool for the production of bioenergy and biochemicals from glycerol.Biotechnol Bioproc Engineer 2012;17:671–8.

[22] Poladyan A, Avagyan A, Vassilian A, Trchounian A. Oxidative and reductiveroutes of glycerol and glucose fermentation by Escherichia coli batch culturesand their regulation by oxidizing and reducing reagents at different pHs. CurrMicrobiol 2013;66:49–55.

[23] Kim K, Kim SK, Park YC, Seo JH. Enhanced production of 3-hydroxy-propionicacid from glycerol by modulation of glycerol metabolism in recombinantEscherichia coli. Bioresour Technol 2014;156:170–5.

[24] Booth IR. Glycerol and methylglyoxal metabolism. In: Neidhardt, FG, editor-in-Chief, EcoSal – Escherichia coli and Salmonella. Cellular and MolecularBiology. ASM Press: Washington DC, 2006. <http://www.ecosal.org>.

[25] Trchounian K, Poladyan A, Vassilian A, Trchounian A. Multiple and reversiblehydrogenases for hydrogen production by Escherichia coli: dependence onfermentation substrate, pH and FOF1-ATPase. Crit Rev Biochem Mol Biol2012;47:236–49.

[26] Mnatsakanyan N, Vassilian A, Navasardyan L, Bagramyan K, Trchounian A.Regulation of Escherichia coli formate hydrogenlyase activity by formate atalkaline pH. Curr Microbiol 2002;45:281–6.

[27] Mnatsakanyan N, Bagramyan K, Trchounian A. Hydrogenase 3 but nothydrogenase 4 is major in hydrogen gas production by Escherichia coliformate hydrogenlyase at acidic pH and in the presence of external formate.Cell Biochem Biophys 2004;41:357–65.

[28] Bakonyi P, Nemestóthy N, Lövitusz É, Bélafi-Bakó K. Application of Plackett-Burman experimental design to optimize biohydrogen fermentation by E. coli(XL1-BLUE). Int J Hydrogen Energy 2011;36:13949–54.

[29] Futatsugi L, Saito H, Kakegawa T, Kobayashi H. Growth of an Escherichia colimutant deficient in respiration. FEMS Microbiol Lett 1997;156:141–5.

[30] Bagramyan K, Trchounian A. Structure and functioning of formate hydrogenlyase, key enzyme of mixed-acid fermentation. Biochemistry (Moscow)2003;68:1159–70.

[31] Trchounian A, Sawers RG. Novel insights into the bioenergetics of mixed-acidfermentation: can hydrogen and proton cycles combine to help maintain aproton motive force? IUBMB Life 2014;66:1–7.

[32] Redwood MD, Mikheenko IP, Sargent F, Macaskie LE. Dissecting the roles ofEscherichia coli hydrogenases in biohydrogen production. FEMS Microbiol Lett2008;278:48–55.

[33] Lukey MJ, Parkin A, Roessler MM, Murphy BJ, Harmer J, Palmer T, et al. HowEscherichia coli is equipped to oxidize hydrogen under different redoxconditions. J Biol Chem 2010;285:3928–38.

[34] Trchounian K, Sanchez-Torres V, Wood TK, Trchounian A. Escherichia colihydrogenase activity and H2 production under glycerol fermentation at a lowpH. Int J Hydrogen Energy 2011;36:4323–31.

[35] Poladyan A, Trchounian K, Sawers G, Trchounian A. Hydrogen-oxidizinghydrogenases 1 and 2 of Escherichia coli regulate the onset of hydrogenevolution and ATPase activity, respectively, during glucose fermentation atalkaline pH. FEMS Microbiol Lett 2013;348:143–8.

[36] Menon NK, Robbins J, Wendt JC, Shanmugan KT, Przybyla AE. Mutationalanalysis and characterization of the Escherichia coli hya operon, whichencodes [NiFe] hydrogenase 1. J Bacteriol 1991;173:4851–61.

[37] Menon NK, Chatelus CY, Dervartanian M, Wendt JC, Shanmugam KT, Peck HD,et al. Cloning, sequencing, and mutational analysis of the hyb operonencoding Escherichia coli hydrogenase 2. J Bacteriol 1994;176:4416–23.

[38] Sauter M, Bohm R, Bock A. Mutational analysis of the operon (hyc) determinghydrogenase 3 formation in Escherichia coli. Mol Microbiol 1992;6:1523–32.

[39] Andrews SC, Berks BC, Mcclay J, Ambler A, Quail MA, Golby P, et al. A 12-cistron Escherichia coli operon (hyf) encoding a putative proton-translocatingformate hydrogenlyase system. Microbiology 1997;143:3633–47.

[40] Richard DJ, Sawers G, Sargent F, McWalter L, Boxer DH. Transcriptionalregulation in response to oxygen and nitrate of the operons encoding the [Ni–Fe] hydrogenases 1 and 2 of Escherichia coli. Microbiology 1999;145:2903–12.

[41] Hube M, Blokesch M, Bock A. Network of hydrogenase maturation inEscherichia coli: role of accessory proteins HypA and HybF. J Bacteriol2002;184:3879–85.

[42] Trchounian K. Transcriptional control of hydrogen production during mixedcarbon fermentation by hydrogenases 4 (hyf) and 3 (hyc) in Escherichia coli.Gene 2012;506:156–60.

[43] Sanchez-Torres V, Yusoff MYM, Nakano C, Maeda M, Ogawa HI, Wood TK.Influence of Escherichia coli hydrogenases on hydrogen fermentation fromglycerol. Int J Hydrogen Energy 2013;38:3905–12.

[44] Pinske C, Jaroschinsky M, Linek S, Kelly CL, Sargent F, Sawers RG. Physiologyand bioenergetics of [NiFe]-hydrogenase 2-catalyzed H2-consuming and H2-producing reactions in Escherichia coli. J Bacteriol 2015;197:296–306.

[45] Trchounian K, Trchounian A. Escherichia coli hydrogenase 4 (hyf) andhydrogenase 2 (hyb) contribution in H2 production during mixed carbon(glucose and glycerol) fermentation at pH 7.5 and pH 5.5. Int J HydrogenEnergy 2013;38:3919–27.

[46] Dimanta I, Grunduls A, Nikolaeva V, Kleperis J, Muiznieks I. Crude glycerol asa perspective substrate for bio-hydrogen production in Latvia. Int Sci J AltEnergy Ecology 2012;9:28–31.

[47] Blbulyan S, Avagyan A, Poladyan A, Trchounian A. Role of Escherichia colidifferent hydrogenases in H+ efflux and the FOF1-ATPase activity duringglycerol fermentation at different pH. Biosci Rep 2011;31:179–84.

[48] Trchounian K, Pinske C, Sawers RG, Trchounian A. Dependence on the F0F1-ATP synthase for the activities of the hydrogen-oxidizing hydrogenases 1 and2 during glucose and glycerol fermentation at high and low pH in Escherichiacoli. J Bioenerg Biomembr 2011;43:645–50.

[49] Trchounian A. Escherichia coli proton-translocating F0F1-ATP synthase and itsassociation with solute secondary transporters and/or enzymes of anaerobicoxidation-reduction under fermentation. Biochem Biophys Res Comm2004;315:1051–7.

[50] Trchounian K, Blbulyan S, Trchounian A. Hydrogenase activity and proton-motive force generation by Escherichia coli during glycerol fermentation. JBioenerg Biomembr 2013;45:253–60.

[51] Trchounian K, Trchounian A. Clean energy technology development:hydrogen production by Escherichia coli during glycerol fermentation. In:Dincer I, Colpan CO, Kizilkan O, et al., editor. Proceedings of the 13thInternational Conference on Clean Energy, Istanbul (Turkey), 2014. p. 1322–8.

[52] Blbulyan S, Trchounian A. Impact of membrane-associated hydrogenases onthe FoF1-ATPase in Escherichia coli during glycerol and mixed carbonfermentation: atpase activity and its inhibition by N, N’-dicyclohexylcarbodiimide in the mutants lacking hydrogenases. ArchBiochem Biophys 2015;579:67–72.

[53] Pettigrew DW. Inactivation of Escherichia coli glycerol kinase by 5,5’-dithiobis(2-nitrobenzoic acid) and N-ethylmaleimide: evidence fornucleotide regulatory binding sites. Biochemistry 1986;25:4711–8.

[54] Hakobyan L, Gabrielyan L, Trchounian A. Relationship of proton motive forceand the FOF1-ATPase with bio-hydrogen production activity of Rhodobactersphaeroides: effects of diphenylene iodonium, hydrogenase inhibitor, and itssolvent dimethylsulphoxide. J Bioenerg Biomembr 2012;44:495–502.

[55] Trchounian K, Trchounian A. Escherichia coli multiple [Ni–Fe]-hydrogenasesare sensitive to osmotic stress during glycerol fermentation but at differentpHs. FEBS Lett 2013;587:3562–6.

[56] Bagramyan K, Mnatsakanyan N, Poladyan A, Vassilian A, Trchounian A. Theroles of hydrogenases 3 and 4, and the F0F1-ATPase, in H2 production byEscherichia coli at alkaline and acidic pH. FEBS Lett 2002;516:172–8.

[57] Trchounian K, Sargsyan H, Trchounian A. Hydrogen production by Escherichiacoli depends on glucose concentration and its combination with glycerol atdifferent pHs. Int J Hydrogen Energy 2014;39:6419–23.

[58] Gabrielyan L, Sargsyan H, Hakobyan L, Trchounian A. Regulation of hydrogenphotoproduction in Rhodobacter sphaeroides batch culture by externaloxidizers and reducers. Appl Energy 2014;131:20–5.

[59] Fernandez VM. An electrochemical cell for reduction of biochemical: itsapplication to the study of the effect pf pH and redox potential on the activityof hydrogenases. Anal Biochem 1983;130:54–9.

[60] Eltsova ZA, Vasilieva LG, Tsygankov AA. Hydrogen production by recombinantstrains of Rhodobacter sphaeroides using a modified photosynthetic apparatus.Appl Biochem Microbiol 2010;46:487–91.

K. Trchounian, A. Trchounian / Applied Energy 156 (2015) 174–184 183

[61] Noguchi K, Riggins DP, Eldahan KC, Kitko RD, Slonczewski JL. Hydrogenase-3contributes to anaerobic acid resistance of Escherichia coli. PLoS ONE2010;5:e10132.

[62] Piskarev IM, Ushkanov VA, Aristova NA, Likhachev PP, Myslivets TS.Establishment of the redox potential of water saturated with hydrogen.Biophysics 2010;55:13–7.

[63] Bagramyan KA, Martirosov SM. Formation of an ion transport supercomplexin Escherichia coli. An experimental model of direct transduction of energy.FEBS Lett 1989;249:149–52.

[64] Maeda T, Wood TK. Formate detection by potassium permanganate forenhanced hydrogen production in Escherichia coli. Int J Hydrogen Energy2008;33:2409–12.

[65] Barrett EL, Kwan HS, Macy J. Anaerobiosis, formate, nitrate, and pyrA areinvolved in the regulation of formate hydrogenlyase in Salmonellatyphimurium. J Bacteriol 1984;158:972–7.

[66] Maeda T, Sanchez-Torres V, Wood TK. Hydrogen production byrecombinantEscherichia coli strains. Microb Biotechnol 2012;5:214–25.

[67] Hu H, Wood TK. An evolved Escherichia coli strain for producing hydrogen andethanol from glycerol. Biochem Biophys Res Comm 2010;391:1033–8.

[68] Tran KT, Maeda M, Wood TK. Metabolic engineering of Escherichia coli toenhance hydrogen production from glycerol. Appl Microbiol Biotechnol2014;98:4757–70.

[69] Tran KT, Maeda T, Sanchez-Torres V, Wood TK. Beneficial knockouts inEscherichia coli for producing hydrogen from glycerol. Appl MicrobiolBiotechnol 2015;99:2573–81.

[70] Yoshida A, Nishimura T, Kawagushi H, Inui M, Yukawa H. Enhanced hydrogenproduction from formic acid by formate hydrogen lyase-overexpressingEscherichia coli. Appl Env Microbiol 2005;71:6762–8.

[71] Rollin JA, del Campo JM, Myung S, Sun F, You C, Bakovic A, et al. High-yieldhydrogen production from biomass by in vitro metabolic engineering: mixedsugars coutilization and kinetic modeling. Proc Nat Acad Sci USA2015;112:4964–9.

[72] Chaudhary N, Ngadi MO, Simpson B. Comparison of glucose, glycerol andcrude glycerol fermentation by Escherichia coli K12. J Bioprocess Biotecniq2012;S1:001.

[73] Schlensog V, Bock A. Identification and sequence analysis of the geneencoding the transcriptional activator of the formate hydrogenlyase systemof Escherichia coli. Mol Microbiol 1990;4:1319–27.

[74] Trchounian K, Soboh B, Sawers RG, Trchounian A. Contribution ofhydrogenase 2 to stationary phase H2 production by Escherichia coli duringfermentation of glycerol. Cell Biochem Biophys 2013;66:103–8.

[75] Trchounian K, Trchounian A. Escherichia coli hydrogen gas production fromglycerol: effects of external formate. Renew Energy 2015;83:345–51.

[76] Trchounian K, Abrahamyan V, Poladyan A, Trchounian A. Escherichia coligrowth and hydrogen production in batch culture upon formate alone andwith glycerol co-fermentation at different pHs. Int J Hydrogen Energy 2015.http://dx.doi.org/10.1016/j.ijhydene.2015.06.087.

[77] Liu F, Fang B. Optimization of bio-hydrogen production from biodiesel wastesby Klebsiella pneumoniae. Biotechnol J 2007;2:374–80.

[78] Kivisto A, Santala V, Karp M. Hydrogen production from glycerol usinghalophilic fermentative bacteria. Bioresour Technol 2010;101:8671–7.

[79] Maru BT, Constanti M, Stchigel AM, Medina F, Sueiras JE. Biohydrogenproduction by dark fermentation of glycerol using Enterobacter andCitrobacter sp. Biotechnol Prog 2013;29:31–8.

[80] Ghosh D, Tourigny A, Hallenbeck PC. Near stoichiometric reforming ofbiodiesel derived crude glycerol to hydrogen by photofermentation. Int JHydrogen Energy 2012;37:2273–7.

[81] Pott RWM, Howe CJ, Dennis JS. Photofermentation of crude glycerol frombiodiesel using Rhodopseudomonas palustris: comparison with organic acidsand the identification of inhibitory compounds. Bioresour Technol2013;130:725–30.

[82] Dimanta I, Nikolaeva Y, Grunduls A, Muiznieks I, Kleperis J. Assessment ofbio-hydrogen production from glycerol and glucose by fermentative bacteria.Energetika 2013;59:124–8.

[83] Han J, Lee D, Cho J, Lee J, Kim S. Hydrogen production from biodiesel by-product by immobilized Enterobacter aerogenes. Bioproc Biosyst Engineer2012;35:151–7.

[84] Reungsang A, Sittijunda S, O-thong S. Bio-hydrogen production from glycerolby immobilized Enterobacter aerogenes ATCC 13048 on heat-treated UASBgranules as affected by organic loading rate. Int J Hydrogen Energy2013;38:6970–9.

[85] Chookaew T, O-Thong S, Prasertsan P. Statistical optimization of mediumcomponents affecting simultaneous fermentative hydrogen and ethanolproduction from crude glycerol by thermotolerant Klebsiella sp. TR17. Int JHydrogen Energy 2014;39:751–60.

[86] Lo YC, Chen XJ, Yuan Y, Chang JS. Dark fermentative hydrogen productionwith crude glycerol from biodiesel industry using indigenous hydrogen-producing bacteria. Int J Hydrogen Energy 2013;38:15815–28.

[87] Sarma SJ, Brar SK, Le Bihan Y, Buelna G, Soccol SR. Hydrogen production frommeat processing and restaurant waste derived crude glycerol by anaerobicfermentation and utilization of the spent broth. J Chem Technol Biotechnol2013;88:2264–71.

[88] Kumar P, Sharma R, Ray S, Mehariya S, Patel SK, Lee JK, et al. Darkfermentative bioconversion of glycerol to hydrogen by Bacillus thuringiensis.Bioresour Technol 2015;182:383–8.

[89] Mangayil R, Aho T, Karp M, Santala V. Improved bioconversion of crudeglycerol to hydrogen by statistical optimization of media components. RenewEnergy 2015;75:583–9.

[90] Ito T, Nakashimada Y, Senba K, Matsui T, Nishio N. Hydrogen and ethanolproduction from glycerol-containing wastes discharged after biodieselmanufacturing process. J Biosci Bioeng 2005;100:260–5.

[91] Kivisto A, Largo A, Ciranna A, Sanrala V, Roos C, Karp M. Genome sequence ofHalanaerobium saccharolyticum subsp. saccharolyticum strain DSM 6643T, ahalophilic hydrogen-producing bacterium. Genome Announcements 2013;1.e00187-13.

[92] Jitrwung R, Verrett J, Yargeau V. Optimization of selected salts concentrationfor improved biohydrogen production from biodiesel-based glycerol usingEnterobacter aerogenes. Renew Energy 2013;50:222–36.

[93] Sargsyan H, Gabrielyan L, Hakobyan L, Trchounian A. Light-dark durationalternation effects on Rhodobacter sphaeroides growth, membrane propertiesand bio-hydrogen production in batch culture. Int J Hydrogen Energy2015;40:4084–91.

[94] Mangahil R, Karp M, Santala V. Bioconversion of crude glycerol from biodieselproduction to hydrogen. Int J Hydrogen Energy 2012;37:12198–204.

[95] Montpart N, Rago L, Baeza JA, Guisasola A. Hydrogen production in singlechamber microbial electrolysis cells with different complex substrates. WaterRes 2015;68:601–5.

[96] Mu Y, Yang H-Y, Wang Y-Z, He C-H, Zhao Q-B, Wang Y, et al. The maximumspecific hydrogen producing activity of anaerobic mixed cultures: definitionand determination. Sci Rep 2014;4:5239.

[97] Gupta M, Velayutham P, Elbeshbishy E, Hafez H, Khafipour E, Derakhshani H,et al. Co-fermentation of glucose, starch, and cellulose for mesophilicbiohydrogen production. Int J Hydrogen Energy 2014;39:20958–67.

[98] Vendruscolo F. Starch: a potential substrate for biohydrogen production. Int JEnergy Res 2014;39:293–302.

[99] Marone A, Izzo G, Mentuccia L, Massini G, Paganin P, Rosa S, et al. Vegetablewaste as substrate and source of suitable microflora for bio-hydrogenproduction. Renew Energy 2014;68:6–13.

[100] Song ZX, Li XH, Li WW, Bai YX, Fan YT, Hou HW. Direct bioconversion of rawcorn stalk to hydrogen by a new strain Clostridium sp. FS3. Bioresour Technol2014;157:91–7.

[101] Debowski M, Korzeniewska E, Filipkowska Z, Zielinski M, Kwiatkowski R.Possibility of hydrogen production during cheese whey fermentation processby different strains of psychrophilic bacteria. Int J Hydrogen Energy2014;39:1972–8.

[102] Kapdan IK, Kargi F. Biohydrogen production from waste materials. EnzymeMicrob Technol 2006;38:569–82.

[103] Choi J, Ahn Y. Characteristics of biohydrogen fermentation from varioussubstrates. Int J Hydrogen Energy 2013;39:3152–9.

[104] Kumar G, Bakonyi P, Periyasamy S, Kim SH, Nemestóthy N, Bélafi-Bakó K.Lignocellulose biohydrogen: Practical challenges and recent progress. RenewSustain Energy Rev 2015;44:728–37.

[105] Ngo TA, Kim MS, Sim SJ. High-yield biohydrogen production from biodieselmanufacturing waste by Thermotoga neapolitana. Int J Hydrogen Energy2011;36:5836–42.

[106] Ngo TA, Sim SJ. Dark fermentation of hydrogen from waste glycerol usinghyperthermophilic eubacterium Thermotoga neapolitana. Environ ProgSustain Energy 2012;31:466–73.

[107] Maru BT, Bielen AAM, Constanti M, Medina F, Kengen SWM. Glycerolfermentation to hydrogen by Thermotoga maritima: proposed pathway andbioenergetic considerations. Int J Hydrogen Energy 2013;38:5563–72.

[108] Sapra R, Bagramyan K, Adams MW. A simple energy-conserving system:proton reduction coupled to proton translocation. Proc Natl Acad Sci USA2003;100:7545–50.

[109] Bakonyi P, Nemestóthy N, Bélafi-Bakó K. Biohydrogen purification bymembranes: an overview on the operational conditions affecting theperformance of non-porous, polymeric and ionic liquid based gasseparation membranes. Int J Hydrogen Energy 2013;38:9673–87.

[110] Bakonyi P, Nemestóthy N, Lankó J, Rivera I, Buitrón G, Bélafi-Bakó K.Simultaneous biohydrogen production and purification in a double-membrane bioreactor system. Int J Hydrogen Energy 2015;40:1690–7.

[111] Ramírez-Morales JE, Tapia-Venegas E, Nemestóthy N, Bakonyi P, Bélafi-BakóK, Ruiz-Filippi G. Evaluation of two gas membrane modules for fermentativehydrogen separation. Int J Hydrogen Energy 2013;38:14042–52.

[112] Lai T, Yin H, Lind ML. The hydrogen permeability of Cu–Zr binary amorphousmetallic membranes and the importance of thermal stability. J Membr Sci2015;489:264–9.

[113] Al-Mufachi NA, Rees NV, Steinberger-Wilkens R. Hydrogen selectivemembranes: a review of palladium-based dense metal membranes. RenewSustain Energy Rev 2015;47:540–51.

[114] Shao L, Low BT, Chung TS, Greenberg AR. Polymeric membranes for thehydrogen economy: contemporary approaches and prospects for the future. JMembr Sci 2009;327:18–31.

[115] Barca C, Soric A, Ranava D, Giudici-Orticoni MT, Ferrasse JH. Anaerobicbiofilm reactors for dark fermentative hydrogen production fromwastewater: a review. Bioresour Technol 2015;185:386–98.

[116] Noike T, Takabatake H, Mizuno O, Ohba M. Inhibition of hydrogenfermentation of organicwastes by lacticacid bacteria. Int J Hydrogen Energy2002;27:1367–71.

184 K. Trchounian, A. Trchounian / Applied Energy 156 (2015) 174–184

[117] Baghchehsaraee B, Nakhla G, Karamanev D, Margaritis A. Revivability offermentative hydrogen producing bioreactor. Int J Hydrogen Energy2011;34:2573–9.

[118] Maeda T, Sanchez-Torres V, Wood TK. Escherichia coli hydrogenase 3 is areversible enzyme possessing hydrogen uptake and synthesis activities. ApplMicrobiol Biotechnol 2008;76:1036–42.

[119] Schievano A, Tenca A, Lonati S, Manzini E, Adani F. Can two-stage instead ofone-stage anaerobic digestion really increase energy recovery from biomass?Appl Energy 2014;124:335–42.

[120] Sarma SJ, Brar SK, Le Bihan Y, Buelna G. Bio-hydrogen production bybiodiesel-derived crude glycerol bioconversion: a techno-economicevaluation. Bioproc Biosyst Eng 2013;36:1–10.