Media optimization for biohydrogen production from waste glycerol by anaerobic thermophilic mixed cultures

Media compositions affecting thermophilic biohydrogen production from waste glycerol were optimized using response surface methodology (RSM) with central composite design (CCD). Investigated parameters used were waste glycerol concentration, urea concentration, the amount of Endo-nutrient addition and disodium hydrogen phosphate (Na₂HPO₄) concentration. Waste glycerol concentration and the amount of Endo-nutrient addition had a significant individual effect on the cumulative hydrogen production (HP) (p ≤ 0.05). The interactive effect on HP was found between waste glycerol and urea concentration as well as waste glycerol concentration and the amount of Endo-nutrient addition (p ≤ 0.05). The optimal media compositions were 20.33 g/L of waste glycerol, 0.16 g/L of urea, 3.97 g/L of Na₂HPO₄ and 0.20 mL/L of the amount of Endo-nutrient addition which gave the maximum HP of 1470.19 mL H₂/L. The difference between observed HP (1502.84 mL H₂/L) and predicted HP was 2.22%. The metabolic products from the fermentation process were 1,3-propanediol (1,3-PD), ethanol, acetic, formic, lactic, butyric, and propionic acids. Results from polymerase chain reaction-denaturing gradient gel electrophoresis (PCR-DGGE) analysis indicated that the hydrogen producers present in the fermentation broth was Thermoanaerobacterium sp.     

Bio-hydrogen production by a new isolated strain Rhodopseudomonas sp. WR-17 using main metabolites of three typical dark fermentation type

In this work, a new strain WR-17 was isolated for photo-fermentative hydrogen production and its hydrogen production capacity was investigated by utilizing main liquid byproducts of three dark fermentation types in batch culture. Experimental results indicated that strain WR-17 was identified as genus Rhodopseudomonas and maximum hydrogen yield of 2.42 mol H₂/mol acetate was obtained when the acetate was used as sole carbon source. Strain WR-17 had an excellent ability of using mixed short chain acids of three typical fermentations such as acetate and ethanol, acetate and butyrate, acetate and propionate. Result demonstrated that the metabolites of butyric acid-type fermentation as substrate is fitting to produce hydrogen and maximum cumulative hydrogen volume of 2156 ml/L-medium was obtained when acetate of 30 mmol/L and butyrate of 15 mmol/L were used. Therefore, butyric acid-type fermentation has great potential for further obtaining high hydrogen yield by the combining photo-fermentation.     

Optimizing the production of hydrogen and 1,3-propanediol in anaerobic fermentation of biodiesel glycerol

The conversion of glycerol in biodiesel waste streams to valuable products (e.g. hydrogen and 1,3-propanediol (1,3-PD)) was studied through batch-mode anaerobic fermentation with organic soil as inoculum. The production of hydrogen in headspace and 1,3-PD in liquid phase was examined at different hydrogen retention times (HyRTs), which were controlled by gas-collection intervals (GCIs) and initial gas-collection time points (IGCTs). Two purification stages of biodiesel glycerol (P2 and P3) were tested at three concentrations (3, 5 and 7 g/L). Longer HyRT (longer GCI and longer IGCT) led to lower hydrogen yield but higher 1,3-PD yield. The P3 glycerol at the concentration of 7 g/L had the highest 1,3-PD yield (0.65 mol/mol glycerol_consumed) at the GCI/IGCT of 20 h/65 h and the highest hydrogen yield (0.75 mol/mol glycerol_consumed) at the GCI/IGCT of 2.5 h/20 h), respectively. A mixed-order kinetic model was developed to simulate the effects of GCI/IGCT on the production of hydrogen and 1,3-PD. The results showed that the production of hydrogen and 1,3-PD can be optimized by adjusting HyRT in anaerobic fermentation of glycerol.     

High hydrogen production from glycerol or glucose by electrohydrogenesis using microbial electrolysis cells

The use of glycerol for hydrogen gas production was examined via electrohydrogenesis using microbial electrolysis cells (MECs). A hydrogen yield of 3.9 mol-H₂/mol was obtained using glycerol, which is higher than that possible by fermentation, at relatively high rates of 2.0 ± 0.4 m³/m³ d (E_ap = 0.9 V). Under the same conditions, hydrogen was produced from glucose at a yield of 7.2 mol-H₂/mol and a rate of 1.9 ± 0.3 m³/m³ d. Glycerol was completely removed within 6 h, with 56% of the electrons in intermediates (primarily 1,3-propanediol), with the balance converted to current, intracellular storage products or biomass. Glucose was removed within 5 h, but intermediates (mainly propionate) accounted for only 19% of the electrons. Hydrogen was also produced using the glycerol byproduct of biodiesel fuel production at a rate of 0.41 ± 0.1 m³/m³ d. These results demonstrate that electrohydrogenesis is an effective method for producing hydrogen from either pure glycerol or glycerol byproducts of biodiesel fuel production.     

Effects of hydraulic retention time on anaerobic hydrogenation performance and microbial ecology of bioreactors fed with glucose–peptone and starch–peptone

This study evaluated anaerobic hydrogenation performance and microbial ecology in bioreactors operated at different hydraulic retention time (HRT) conditions and fed with glucose–peptone (GP) and starch–peptone (SP). The maximum hydrogen production rates for GP- and SP-fed bioreactors were found to be 1247 and 412 mmol-H₂/L/d at HRT of 2 and 3 h, respectively. At HRT > 8 h, hydrogen consumption due to peptone fermentation could occur and thus reduced hydrogen yield from carbohydrate fermentation. Results of cloning/sequencing and denaturant gradient gel electrophoresis (DGGE) indicated that Clostridium sporogenes and Clostridium celerecrescens were dominant hydrogen-producing bacteria in the GP-fed bioreactor, presumably due to their capability on protein hydrolysis. In the SP-fed bioreactor, Lactobacillus plantarum, Propionispira arboris, and Clostridium butyricum were found to be dominant populations, but the presence of P. arboris at HRT > 3 h might be responsible for a lower hydrogen yield from starch fermentation. As a result, optimizing HRT operation for bioreactors was considered an important asset in order to minimize hydrogen-consuming activities and thus maximize net hydrogen production. The limitation of simple parameters such as butyrate to acetate ratio (B/A ratio) in predicting hydrogen production was recognized in this study for bioreactors fed with multiple substrates. It is suggested that microbial ecology analysis, in addition to chemical analysis, should be performed when complex substrates and mixed cultures are used in hydrogen-producing bioreactors.     

Effect of operation parameters on anaerobic fermentation using cow dung as a source of microorganisms

Anaerobic fermentation by microorganisms is a promising method of hydrogen production for it can be conducted at mild conditions. In this paper, a series of tests were carried out to investigate the effect of pH, hydraulic retention time (HRT), temperature (T) and substrate concentration on anaerobic dark fermentation. Glucose was utilized as model substrate. The Taguchi orthogonal array was applied in the experimental design and a verification experiment was tested. The results showed the optimal parameters for hydrogen production were pH 5.0, HRT 8.34 h, T 33.5 °C and substrate concentration of 14 g/L, with hydrogen yield of 2.15 mol H₂/mol glucose. Butyric-type fermentation occurred in most tests. According to the analysis of effluent contents, at pH 5.5, 5.0, 4.0, the effluent contained mostly butyric acid (43.1–56.6%), followed by acetic acid (24.6–29.8%); at HRT 4.17, 6.26, 8.34 h, the effluent contained mostly butyric acid (43.0–53.6%). Increasing temperature from 29 to 39.5 °C resulted in the decrease of butyric acid percentage but increase of ethanol percentage. Substrate concentration had little effect on product constitution.     

Biohydrogen production from waste glycerol and sludge by anaerobic mixed cultures

Key factors affecting biohydrogen production from waste glycerol and sludge by anaerobic mixed cultures were optimized using response surface methodology (RSM) with central composite design (CCD). Investigated parameters were waste glycerol concentration, sludge concentration, and the amount of Endo–nutrient addition. Concentrations of waste glycerol and sludge had a significant individual effect on hydrogen production rate (HPR) (p ≤ 0.05). The interactive effect on HPR (p ≤ 0.05) was found between waste glycerol concentration and sludge concentration. The optimal conditions for the maximum HPR were: waste glycerol concentration 22.19 g/L, sludge concentration 7.16 g-total solid (TS/L), and the amount of Endo–nutrient addition 2.89 mL/L in which the maximum HPR of 1.37 mmol H₂/L h was achieved. Using the optimal conditions, HPR from a co-digestion of waste glycerol and sludge (1.37 mmol H₂/L h) was two times greater than the control (waste glycerol without addition of sludge) (0.76 mmol H₂/L h), indicating a significant enhancement of HPR by sludge. Major metabolites of the fermentation process were ethanol, 1,3-propanediol (1,3-PD), lactate, and formate.     

Closing the 1,3-propanediol route enhances hydrogen production from glycerol by Halanaerobium saccharolyticum subsp. saccharolyticum

The effect of vitamin B₁₂ on hydrogen (H₂) and 1,3-propanediol (1,3-PD) production in glycerol fermentation of Halanaerobium saccharolyticum subsp. saccharolyticum was studied in batch cultivations. In addition, the effect of vitamin B₁₂ on glucose fermentation of H. saccharolyticum subsp. saccharolyticum and on glycerol fermentation of H. saccharolyticum subsp. senegalensis, that is a non-1,3-PD producing bacterium, and of Clostridium butyricum, that is known for vitamin B₁₂-independent 1,3-PD production, were studied for comparison. 1,3-PD production of H. saccharolyticum subsp. saccharolyticum was observed to be vitamin B₁₂-dependent and the H₂ production in glycerol fermentation was remarkably enhanced when 1,3-PD production was blocked by the lack of vitamin B₁₂ supplementation. The highest H₂ yield obtained by H. saccharolyticum subsp. saccharolyticum was 2.16 mol/mol glycerol. No significant effects of vitamin B₁₂ on metabolite production in glucose fermentation of H. saccharolyticum subsp. saccharolyticum or in glycerol fermentation of H. saccharolyticum subsp. senegalensis or C. butyricum were observed.     

Fermentative production of hydrogen and soluble metabolites from crude glycerol of biodiesel plant by the newly isolated thermotolerant Klebsiella pneumoniae TR17

A thermotolerant fermentative hydrogen-producing strain was isolated from crude glycerol contaminated soil and identified as Klebsiella pneumoniae on the basis of the 16S rRNA gene analysis as well as physiological and biochemical characteristics. The selected strain, designated as K. pneumoniae TR17, gave good hydrogen production from crude glycerol. Culture conditions influencing the hydrogen production were investigated. The strain produced hydrogen within a wide range of temperature (30–50 °C), initial pH (4.0–9.0) and crude glycerol concentration (20–100 g/L) with yeast extract as a favorable nitrogen source. In batch cultivation, the optimal conditions for hydrogen production were: cultivation temperature at 40 °C, initial pH at 8.0, 20 g/L crude glycerol and 2 g/L yeast extract. This resulted in the maximum cumulative hydrogen production of 27.7 mmol H₂/L and hydrogen yield of 0.25 mol H₂/mol glycerol. In addition, the main soluble metabolites were 1,3-propanediol, 2,3-butanediol and ethanol corresponding to the production of 3.52, 2.06 and 3.95 g/L, respectively.     

温度、pH和负荷对果蔬垃圾厌氧酸化途径的影响

研究考察了果蔬垃圾在F/M=2∶1,不同温度(35℃,50℃)条件下,控制pH(4.0,5.0,6.0)时,发酵产物组分的变化规律;考察了35℃,果蔬垃圾在控制pH的完全混合反应器(CSTR)中不同有机负荷下的酸化类型。结果表明:50℃下酸化产物浓度较35℃低,且在相同的pH下,高温和中温酸化呈现不同的酸化类型,pH=6.0时,高温酸化类型为乙醇型,中温酸化类型为丁酸型。中温条件下,提高pH有助于缩短酸化时间,但对酸化产物总量影响不大,且发酵类型随pH变化明显,pH=4.0为乙醇型发酵,pH=5.0为丙酸型发酵,pH=6.0为丁酸型发酵。最优静态酸化条件为F/M=2∶1,35℃和pH=4.0,酸化率能达到65.1%,其中乙醇浓度为10 508 mg/L(相当于每克挥发性固体中含有438 mg乙醇)。反应器连续运行结果显示,温度为35℃,pH=4.0时,随着负荷的提高,酸化产物中乙醇含量增加,有机负荷OLR>7 g/(L.d),停留时间HRT<5 d,逐渐呈现乙醇型发酵。     

Towards the integration of dark- and photo-fermentative waste treatment. 2. Optimization of starch-dependent fermentative hydrogen production

Eight natural microbial consortia collected from different sites were tested for dark, hydrogen production during starch degradation. The most active consortium was from silo pit liquid under mesophilic (37 °C) conditions. The fermentation medium for this consortium was optimized (Fe, NH₄⁺, phosphates, peptone, and starch content) for both dark fermentation and for subsequent purple photosynthetic bacterial H₂ photoproduction [Laurinavichene TV, Tekucheva DN, Laurinavichius KS, Ghirardi ML, Seibert M, Tsygankov AA. Towards the integration of dark and photo fermentative waste treatment. 1. Hydrogen photoproduction by purple bacterium Rhodobacter capsulatus using potential products of starch fermentation. Int J Hydrogen Energy 2008;33(23):7020–26], in the presence of the spent dark, fermentation effluent. The addition of Zn (10 mg L⁻¹), as a methanogenesis inhibitor that does not inhibit purple bacteria at this concentration, also did not inhibit dark, fermentative H₂ production. The influence of various fermentation end products at different concentrations (up to 30 g L⁻¹) on dark, H₂ production was also examined. Added lactate stimulated, but added isobutyrate and butanol strongly inhibited gas production. Under optimal conditions the fermentation of starch (30 g L⁻¹) resulted in 5.7 L H₂ L⁻¹ of culture (1.6 mol H₂ per mole of hexose) with the co-production mainly of butyrate and acetate.     

Biological hydrogen production in continuous stirred tank reactor systems with suspended and attached microbial growth

Fermentative H₂ production in continuous stirred tank reactor (CSTR) system with bacteria attached onto granular activated carbon (GAC) was designed to produce H₂ continuously. The H₂ production performances of CSTR with suspended and attached-sludge from molasses were examined and compared at various organic loading rates (8–40 g COD/L/d) at hydraulic retention time of 6 h under mesophilic conditions (35 °C). Both reactor systems achieved ethanol-type fermentation in the pH ranges 4.5–4.8 and 3.8–4.4, respectively, while ORP ranges from −450 to −470 mV and from −330 to −350 mV, respectively. The hydrogen production rate in the attached system was higher compared to that of the suspended system (9.72 and 6.65 L/d/L, respectively) while specific hydrogen production rate of 5.13 L/g VSS/d was higher in the suspended system. The attached-sludge CSTR is more stable than the suspended-sludge CSTR with regard to hydrogen production, pH, substrate utilization efficiency and metabolic products (e.g., volatile fatty acids and ethanol) during the whole test.     

Improving hydrogen production from cassava starch by combination of dark and photo fermentation

The combination of dark and photo fermentation was studied with cassava starch as the substrate to increase the hydrogen yield and alleviate the environmental pollution. The different raw cassava starch concentrations of 10–25 g/l give different hydrogen yields in the dark fermentation inoculated with the mixed hydrogen-producing bacteria derived from the preheated activated sludge. The maximum hydrogen yield (HY) of 240.4 ml H₂/g starch is obtained at the starch concentration of 10 g/l and the maximum hydrogen production rate (HPR) of 84.4 ml H₂/l/h is obtained at the starch concentration of 25 g/l. When the cassava starch, which is gelatinized by heating or hydrolyzed with α-amylase and glucoamylase, is used as the substrate to produce hydrogen, the maximum HY respectively increases to 258.5 and 276.1 ml H₂/g starch, and the maximum HPR respectively increases to 172 and 262.4 ml H₂/l/h. Meanwhile, the lag time (λ) for hydrogen production decreases from 11 h to 8 h and 5 h respectively, and the fermentation duration decreases from 75–110 h to 44–68 h. The metabolite byproducts in the dark fermentation, which are mainly acetate and butyrate, are reused as the substrates in the photo fermentation inoculated with the Rhodopseudomonas palustris bacteria. The maximum HY and HPR are respectively 131.9 ml H₂/g starch and 16.4 ml H₂/l/h in the photo fermentation, and the highest utilization ratios of acetate and butyrate are respectively 89.3% and 98.5%. The maximum HY dramatically increases from 240.4 ml H₂/g starch only in the dark fermentation to 402.3 ml H₂/g starch in the combined dark and photo fermentation, while the energy conversion efficiency increases from 17.5–18.6% to 26.4–27.1% if only the heat value of cassava starch is considered as the input energy. When the input light energy in the photo fermentation is also taken into account, the whole energy conversion efficiency is 4.46–6.04%.     

Biohydrogen production from crude glycerol by immobilized Klebsiella sp. TR17 in a UASB reactor and bacterial quantification under non-sterile conditions

Biohydrogen production from crude glycerol by immobilized Klebsiella sp. TR17 was investigated in an up-flow anaerobic sludge blanket (UASB) reactor. The reactor was operated under non-sterile conditions at 40^○C and initial pH 8.0 at different hydraulic retention times (HRTs) (2–12 h) and glycerol concentrations (10–30 g/L). Decreasing the HRT led to an increase in hydrogen production rate (HPR) and hydrogen yield (HY). The highest HPR of 242.15 mmol H₂/L/d and HY of 44.27 mmol H₂/g glycerol consumed were achieved at 4 h HRT and glycerol concentrations of 30 and 10 g/L, respectively. The main soluble metabolite was 1,3-propanediol, which implies that Klebsiella sp. was dominant among other microorganisms. Fluorescence in situ hybridization (FISH) revealed that the microbial community was dominated by Klebsiella sp. with 56.96, 59.45, and 63.47% of total DAPI binding cells, at glycerol concentrations of 10, 20, and 30 g/L, respectively.     

Hydrogen production by combined dark and light fermentation of ground wheat solution

Ground waste wheat was subjected to combined dark and light batch fermentation for hydrogen production. The dark to light biomass ratio (D/L) was changed between 1/2 and 1/10 in order to determine the optimum D/L ratio yielding the highest hydrogen formation rate and the yield. Hydrogen production by only dark and light fermentation bacteria was also realized along with the combined fermentations. The highest cumulative hydrogen formation (CHF = 76 ml), hydrogen yield (176 ml H₂ g⁻¹ starch) and formation rate (12.2 ml H₂ g⁻¹ biomass h⁻¹) were obtained with the D/L ratio of 1/7 while the lowest CHF was obtained with the D/L ratio of 1/2. Dark–light combined fermentation with D/L ratio of 1/7 was faster as compared to the dark and light fermentations alone yielding high hydrogen productivity and reduced fermentation time. Dark and light fermentations alone also yielded considerable cumulative hydrogen, but slower than the combined fermentation.     

Bioenergy production from glycerol in hydrogen producing bioreactors (HPBs) and microbial fuel cells (MFCs)

The supply of glycerol has increased substantially in recent years as a by-product of biodiesel production. To explore the value of glycerol for further application, the conversion of glycerol to bioenergy (hydrogen and electricity) was investigated using Hydrogen Producing Bioreactors (HPBs) and Microbial Fuel Cells (MFCs). Pure-glycerol and the glycerol from biodiesel waste stream were compared as the substrates for bioenergy production. In terms of hydrogen production, the yields of hydrogen and 1,3-propanediol at a pure-glycerol concentration of 3 g/L were 0.20 mol/mol glycerol and 0.46 mol/glycerol, respectively. With glucose as the co-metabolism substrate at the ratio of 3:1 (glycerol:glucose), the yields of hydrogen and 1,3-propanediol from glycerol significantly increased to 0.37 mol/mol glycerol and 0.65 mol/glycerol, respectively. The glycerol from biodiesel waste stream had good hydrogen yields (0.17-0.18 mol H₂/mole glycerol), which was comparable with the pure-glycerol. In terms of power generation in MFCs, pure-glycerol was examined at concentrations of 0.5-5 g/L with the highest power density of 4579 mW/m³ obtained at a concentration of 2 g/L. The power densities from the biodiesel waste glycerol were 1614-2324 mW/m³, which were likely caused by the adverse effects of impurities on electrode materials. An economic analysis indicates that with the annual waste stream of 70 million gallons of glycerol, the expected values generated from HPBs and MFCs were $311 and $98 million, respectively.     

Characteristics of the process of biohydrogen production from simple and complex substrates with different biopolymer composition

The work investigated the characteristics of the dark fermentation (DF) process of a number of simple (starch, sunflower oil, peptone, both separately and mixed) and complex (dog food, pig feed, sewage sludge) substrates using a mixed culture of microorganisms, with a controlled pH (5.5), at 55 °C. Peptone and sunflower oil were characterized by the lowest production of H₂, namely 5.0 and 2.3 ml H₂/g COD, respectively. The specific hydrogen yield from starch was 1.55 mol H₂/mol hexose. The addition of peptone and sunflower oil to starch reduced the specific yield of hydrogen from starch by 23%. A large difference in hydrogen production was observed during DF of complex substrates. The specific hydrogen yield from dog food was 46.5 ml H₂/g COD or 143.4 ml H₂/g carbohydrates; from pig feed – 32.1 ml H₂/g COD or 91.6 ml H₂/g carbohydrates; and from sewage sludge – 9.3 ml H₂/g COD or 98.0 ml H₂/g carbohydrates. Possible relationships between the biopolymer composition of substrates and characteristics of the DF process were analyzed using Spearman's rank correlation coefficients. The concentration of carbohydrates, as well as the ratio of carbohydrates/proteins and carbohydrates/fats, were the main factors influencing the high specific yield of H₂, its content in biogas, as well as the ratio of H₂/soluble metabolites. The concentration of proteins had a statistically significant positive effect on the accumulation of acetate and succinate, and carbohydrates - on the accumulation of caproate.     

Modulation of crude glycerol fermentation by Clostridium pasteurianum DSM 525 towards the production of butanol

High production yields and productivities are requisites for the development of an industrial butanol production process based on biodiesel-derived crude glycerol. However, impurities present in this substrate and/or the concentration of glycerol itself can affect the microbial metabolism. In this work, the effect of crude glycerol concentration on the production of butanol and 1,3-propanediol (1,3-PDO) by Clostridium pasteurianum DSM 525 is studied. Also, the effect of acetate and butyrate supplementation to the culture medium and the culture medium composition are evaluated. The results showed a marked effect of crude glycerol concentration on the product yield. The competitive nature of butanol and 1,3-PDO pathways has been evident, and a shift to the butanol pathway once using higher substrate concentrations (up to 35 g l⁻¹) was clearly observed. Butyrate supplementation to the culture medium resulted in a 45% higher butanol titre, a lower production of 1,3-PDO and it decreased the fermentation time. Acetate supplementation also increased the butanol titre but the fermentation was longer. Even though glycerol consumption could not be increased over 32 g l⁻¹, when the concentrations of NH₄Cl and FeCl₂ were simultaneously increased, the results obtained were similar to those observed when butyrate was supplemented to the culture medium; a 35% higher butanol yield at the expense of 1,3-PDO and a shorter fermentation. The results herein gathered suggest that there are other factors besides butanol inhibition and nutrient limitation that affect the glycerol consumption.     

Conversion of residual glycerol from biodiesel synthesis into 1,3-propanediol by a new strain of Klebsiella pneumoniae

A stepwise strategy was devised to optimize the bioconversion of residual glycerol from biodiesel synthesis into 1,3-propanediol (1,3-PD) by a new strain Klebsiella pneumoniae. This strain is unique in its ability to convert glycerol simultaneously into 1,3-PD. The influence of glycerol concentration, pH, temperature, nitrogen and mineral sources were first investigated using the Plackett–Burman (P–B) statistical design to screen the variables that significantly affect the production of 1,3-PD. Seven variables were statistically significant at 95% significance and a 2^7-3 fractional factorial design (FFD) was applied to further refine the optimization of culture conditions. Results showed that the ideal medium composition and culture conditions for the syntheses of 1,3-PD are: glycerol 65 g/L, yeast extract 5 g/L, peptone 5 g/L, (NH₄)₂SO₄ 7 g/L, K₂HPO₄ 7 g/L, and temperature of 37 °C. Experiments in batch bioreactors under controlled pH produced up to 23.80 g/L of 1,3-PD and 12.30 g/L of ethanol, while in fed-batch cultivations a three-fold increase of 1,3-PD production (36.86 g/L) was obtained compared with the results of the FFD design.     

Batch and continuous biohydrogen production from starch hydrolysate by Clostridium species

In this study, hydrogen gas was produced from starch feedstock via combination of enzymatic hydrolysis of starch and dark hydrogen fermentation. Starch hydrolysis was conducted using batch culture of Caldimonas taiwanensis On1 able to hydrolyze starch completely under the optimal condition of 55 °C and pH 7.5, giving a yield of 0.46–0.53 g reducing sugar/g starch. Five H₂-producing pure strains and a mixed culture were used for hydrogen production from raw and hydrolyzed starch. All the cultures could produce H₂ from hydrolyzed starch, whereas only two pure strains (i.e., Clostridium butyricum CGS2 and CGS5) and the mixed culture were able to ferment raw starch. Nevertheless, all the cultures displayed higher hydrogen production efficiencies while using the starch hydrolysate, leading to a maximum specific H₂ production rate of 116 and 118 ml/g VSS/h, for Cl. butyricumCGS2 and Cl. pasteurianum CH5, respectively. Meanwhile, the H₂ yield obtained from strain CGS2 and strain CH5 was 1.23 and 1.28 mol H₂/mol glucose, respectively. The best starch-fermenting strain Cl. butyricum CGS2 was further used for continuous H₂ production using hydrolyzed starch as the carbon source under different hydraulic retention time (HRT). When the HRT was gradually shortened from 12 to 2 h, the specific H₂ production rate increased from 250 to 534 ml/g  VSS/h, whereas the H₂ yield decreased from 2.03 to 1.50  mol H₂/mol glucose. While operating at 2 h HRT, the volumetric H₂ production rate reached a high level of 1.5 l/h/l.