Biohydrogen production from desugared molasses (DM) using thermophilic mixed cultures immobilized on heat treated anaerobic sludge granules

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 6 ( 2 0 1 1 ) 1 4 2 6 1 e1 4 2 6 9

Available at www.sciencedirect.com

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

Biohydrogen production from desugared molasses (DM) using thermophilic mixed cultures immobilized on heat treated anaerobic sludge granules Prawit Kongjan a,1, Sompong O-Thong a,b, Irini Angelidaki a,* a b

Department of Environmental Engineering, Technical University of Denmark, DK-2800 Lyngby, Denmark Department of Biology, Faculty of Science, Thaksin University, Phathalung 93110, Thailand

article info

abstract

Article history:

Hydrogen production from desugared molasses (DM) was investigated in both batch and

Received 21 February 2011

continuous reactors using thermophilic mixed cultures enriched from digested manure by

Received in revised form

load shock (loading with DM concentration of 50.1 g-sugar/L) to suppress methanogens. H2

18 June 2011

gas, free of methane, was produced during batch cultivations, at different (DM) concen-

Accepted 25 June 2011

trations ranging from 1.5 g-sugars/L to 50.1 g-sugars/L. The highest yield of 237 ml-H2/g-

Available online 27 July 2011

sugar was achieved during the DM batch fermentation at concentration of 2.1 g-sugars/L, whereafter the yield decreased with increasing DM concentration. The enriched hydrogen

Keywords:

producing mixed culture achieved from the 16.7 g-sugars/L DM batch cultivation was

Biohydrogen

immobilized on heat treated anaerobic sludge granules in an up-flow anaerobic sludge

Desugarized molasses

blanket (UASB) reactor. The UASB reactor, operated at a hydraulic retention time (HRT) of

UASB reactor

24 h fed with 16.7 g-sugars/L DM showed good performance with a satisfactory hydrogen

Load shock pretreatment

yield of 269.5 ml-H2/g-sugar and rate of 4500 ml H2/l,d. Fluorescent in situ hybridization (FISH) analysis of the microbial community of sludge from batch fermentation and the UASB-granules after 54 days of operation, was dominated by Thermoanaerobacterium spp., which are key players in fermentative hydrogen production of DM under thermophilic conditions. Furthermore, the granules in the UASB reactor were also significantly containing Thermoanaerobacterium spp. and phylum Firmecutes (most Clotridium, Bacillus and Desulfobacterium) and Thermoanaerobacterium thermosaccharolyticum with a relative abundance of 36%, 27%, and 10% of total microorganisms, respectively. This study shows that hydrogen production could be efficiently facilitated by using anaerobic granules as a carrier, where microbes from mixed culture enriched in the DM batch cultivation were immobilized on, in an UASB reactor. Crown Copyright ª 2011, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.

1.

Introduction

The world is facing both a shortage of fossil fuels and environmental problems caused by greenhouse gases emissions

arising from fossil fuel use. Therefore renewable fuels are being sought to replace and/or substitute fossil fuel. Hydrogen is considered to a sustainable energy carrier because it has a high energy yield (142 kJ/g) and produces only water during

* Corresponding author. Tel.: þ45 4525 1429; fax: þ45 4593 2850. E-mail address: [email protected] (I. Angelidaki). 1 Present address: Department of Science, Faculty of Science and Technology, Prince of Songkla University (Pattani), Muang, Pattani 94000, Thailand. 0360-3199/$ e see front matter Crown Copyright ª 2011, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ijhydene.2011.06.130

14262

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 6 ( 2 0 1 1 ) 1 4 2 6 1 e1 4 2 6 9

its combustion [1]. Hydrogen can be generated by different means of thermo-catalytic reformation of hydrogen-rich organic compounds, electrolysis of water, and biological process. Currently, most of hydrogen is produced mainly by steam reforming of methane and electrolysis of water. However these processes do not reduce waste and require power derived from fossil energy or use fossil fuel as a raw material [2]. Alternatively, hydrogen can be generated by mixed anaerobic bacteria, which are grown on carbohydrates contained mainly in agricultural wastes and industrial effluents thus, the hydrogen fermentation process is a promising way to produce energy and treat organic wastes simultaneously [3,4]. Hydrogen can be generated from organic wastes, by mixed cultures, in the acidogenesis (fermentation) step of anaerobic digestion process. However hydrogen is subsequently utilized by methanogens via the hydrogenotrophic methanogenesis and/or homoacetogenesis pathway to produce methane and/ or acetate [5]. Therefore, original mixed-culture collected from natural sources need to be pretreated in order to eliminate the methanogens. Although several methods such as base, acid, 2-bromoethanesulfonic acid (BESA), load shock, and heat shock pretreatments have been previously demonstrated to be effective for selection of hydrogen producing bacteria [6e8], the load shock pretreatment is considered to be the most simple and cost effective method [7]. Moreover, O-Thong et al. [7] found that mixed culture exposed to shock loads, resulted in higher microbial diversity of hydrogen producers and higher hydrogen production rate than inocula pretreated by other methods. It has been reported that pretreated mixed fermentative bacteria, enriched or immobilized in the continuous reactor fed with non-sterile substrate can give stable hydrogen production rate [4,9]. Thus, the continuous bio-hydrogen process, which is more economical than the batch or semi-batch processes, can be potentially developed for commercial applications [10,11]. Fermentative hydrogen process using enriched mixed culture can be operated in various temperature ranges, from mesophilic (at 25e40  C) up to even of hyperthermophilic (at >80  C) [2]. Although using mesophiles at temperature between 35 and 40  C for converting carbohydrate rich substrates to hydrogen has been conducted in many previous investigations, thermophiles normally operated at a temperature range of 55e60  C have higher hydrogen production rate and yield and to higher extend are following the acetate pathway compared to mesophiles [11]. This is mainly attributed to higher favorable thermodynamics, increasing chemical and biological reaction rates, and to dominance of the acetate pathway contrary to ethanol or lactate pathway and higher tolerance to H2 partial pressure [4,7,9,10]. Additionally, thermophilic temperature is resulting to higher degree for pathogen destruction. Indeed 70  C for 1 h or temperature time combinations achieving the same pasteurization effect (e.g. 55  C 6 h) are applied as sanitation temperatures for biogas reactors [12]. On the contrary, low cell densities achieved and energy required for heating are the main drawbacks of hydrogen dark fermentation in temperature range of the thermophiles, immobilized reactor systems and hot industrial effluents and/or waste heat coupled are suggested for the thermophilic process operation [4,11].

Desugared molasses (DM) is the residual liquid, generated after de-sugaring beet molasses. Large amounts of DM are produced every year that are presently not utilized for energy production [13]. Although most sugars from beet molasses are removed, significant amounts of total sugars (ca.16%) are still remaining that may be utilized for energy recovery. Therefore, DM is a potential feedstock for hydrogen through dark fermentation process. Furthermore, it is rich inorganic in nitrogen, vitamins and salt that could support bacterial growth. The effluent from a hydrogen production step can be utilized for methane production in a subsequent step for efficient utilization of all the organic content to hydrogen and methane. In this study, hydrogen-producing inoculum, prepared by exposing digested manure to high DM load (also called shock load) at thermophilic temperature (55  C), was used to investigate the feasibility of the fermentative hydrogen production in both batch and UASB reactors. Additionally, microbial communities in both batch and UASB reactors were identified by the fluorescence in-situ hybridization (FISH) and polymerase chain reaction-denatured gradient gel electorphoresis (PCReDGGE) for better understand the process.

2.

Materials and methods

2.1. Substrate, medium, and preparation of shock load pretreated sludge Desugared molasses (DM) was kindly supplied by DANISCO (Nakskov, Denmark), and stored at 20  C until further use. Its characterizations were presented in Table 1. Basis anaerobic (BA) medium amended with 1 g/L of yeast extract which was prepared as described by Angelidaki and Sanders [14]. Anaerobic digested manure was collected from a pilot-scale continuously stirred tank reactor (CSTR) operated at 55  C (Lyngby, Denmark). The load shock pretreatments were conducted in 340 ml glass bottles in triplicate. 70 ml of digested manure and 30 ml of DM (167.1 g-sugar/L) was added into the bottles to achieve high loading of DM 50.1 g-sugar/L. The head space of the bottles was flushed with pure N2 for 3e5 min to ensure anaerobic conditions. The bottles were

Table 1 e Characteristics of desugared molasses (DM). Characteristic

Value

pH Total solid (TS) (%) Volatile solid (VS) (%) Total sugar (g/l) Total Kjendhal Nitrogen (g/l) NHþ 4 -N (mg/l) Nitrate (NO 3 ) (g/l) Bromine (Br) (mg/l) Sulfate (SO 4 ) (g/l) Chloride (Cl) (g/l) Lactate (g/l) Acetate (g/l)

7.3 42.8 28.2 167.1 6.72 740 1.1 62.5 5.5 19.8 40.8 11.6

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 6 ( 2 0 1 1 ) 1 4 2 6 1 e1 4 2 6 9

then immediately sealed with rubber stopper and closed with aluminum crimps. Subsequently they were incubated at 55  C for 3 days.

2.2. Batch hydrogen production from DM using inoculum prepared by load shock Batch fermentation was carried out using inoculum obtained from digested manure that was exposed to high loads of DM. The following DM concentrations were used: 1.5, 2.1, 8.4, 16.7, 33.4, and 50.1 g-sugars/L to determine the potential hydrogen production. The batch experiments were carried out in 55 mL serum vials filled with 20 mL mixtures, comprising 16 mL of DM diluted at different concentrations with BA medium (0.9e30% v/v) and 4 mL of inoculum. The serum vials were purged with pure N2 for 3e5 min to create anaerobic conditions and suddenly sealed with butyl stoppers, secured with aluminum crimps. All vials were then put into a 55  C incubator. A control vial was also prepared by using only pure BA medium and inoculum in order to account for possible background hydrogen production which was to be subtracted from hydrogen produced in the vials with various DM concentrations. The experiment was run in triplicate for each DM concentration.

Biogas composition (H2, CH4, and CO2) was measured using a gas chromatograph (GC) (MicroLab, Arhus, Denmark) equipped with a thermal conductivity detector (TCD). For hydrogen detection in the biogas, a 4.5 m  3 mm s-m stainless steel column packed with Molsieve SA (10/80) was used in the GC-TCD, while for CH4 and CO2 analysis, a parallel column of 1.1m  3/16 in. Molsieve 137 and 0.7m  1/4 in. chromosorb 108 was used [9]. The VFAs and alcohols were determined using a GC (GC-2010 Shimadzu, Kyoto, Japan) equipped with a flame ionization detector (FID) and Zebron (ZB)-FFAP column (dimensions 30 m  0.53 mm). Lactate and formate were analyzed by suppressed ion exclusion chromatography equipped with a high performance liquid chromatography (HPLC) pump L2100 Hitachi [4]. Total sugar was analyzed by anthrone-sulfuric acid method [15], using a UV/ VIS spectrometer (Perkin Elmer Lammda2). Chemical oxygen demand (COD), total Kjeldahl nitrogen (TKN), total solid (TS), volatile solid (VS) and ammonium nitrogen (NHþ 4 eN) were measured according to standard methods for the examination of water and wastewater [16]. Chloride, bromide, nitrate and sulfate were measured using the method for determination of inorganic anions by ion chromatography according to Fang et al. [17].

2.5. 2.3. Immobilization of enriched inoculum in thermophilic UASB reactor A laboratory scale UASB reactor with 240 mL total volume and liquid volume of 220 mL was used. The methanogenic granular sludge was obtained from a potato factory (Kruiningen, Netherlands). The granules were sieved and those with a diameter of between 0.25 and 0.5 mm and were autoclaved at 121 C for 30 min to remove methanogenic activity as previously described [9]. Eighty milliliters of these granules was added in a UASB reactor as biomass carriers. Inoculum (seed culture) taken from batch reactors producing hydrogen at DM concentration of 16.7 g-sugars/L was added into the UASB reactor and later flushed with nitrogen gas for 10 min to create anaerobic condition. To initiate good growth conditions in reactors, DM solution 16.7 gsugars/L (DM dissolved in BA medium) was added until the liquid volume reached 220 mL. The reactor was subsequently re-circulated for 1 week with a flow rate of 18 mL/h, to allow bacterial cells to attach to the granules. The reactor was maintained at 55  C by circulating hot water inside a water jacket surrounding the UASB a reactors. The DM concentration of 33.4 g-sugars/L diluted with water and BA medium were mixed 1:1 (v/v) and then fed into the reactor with 1 day HRT, corresponding to the net DM concentration of 16.7 g-sugars/L.

2.4.

Analysis methods

Chemical oxygen demand (COD) balance was made for sugar and lactate degradation to products. The microbial biomass (assumed formula C5H7O2N) formation used in COD balance was assumed to account 15% of the sugars removed in terms of COD [9]. The hydrogen yield (ml-H2/g-sugars) was calculated by measuring the total volume of hydrogen produced divided by g-sugar added.

14263

FISH community analysis

The microbial composition of thermophilic mixed cultures in the samples taken from batch reactors cultivated with 2.1 g-sugars/L DM and the UASB reactor on day 54 of the operation time was determined by FISH with 2 domain specific oligonucleotide probes (EUB338 targeting Bacteria and ARC915 targeting Archaea) [18] one phylum specific oligonucleotide probes (LGC354 targeting Firmicutes (low G þ C) gram positive bacteria [19], three genus specific oligonucleotide probes (Ccs432 targeting Caldicellulosiruptor, Tbm1282 targeting Thermoanaerobacterium [20] and Ttoga660 targeting Thermotogales [21] and one species specific oligonucleotide probes (Tbmthsacc184 targeting Thermoanaerobacterium thermosaccharolyticum [20] covering the main phylogenetic groups of thermophilic and extreme thermophilic hydrogen producers. FISH was performed as described by Amann et al. [22] using paraformaldehyde as a fixative. Prior to hybridization, sludge samples were dispersed by mild ultrasonic treatment (0.5 min pulsed) for 2 min and suspended in PBS buffer ethanol to final concentration of 50% ethanol. For quantitative analyses, the percentage of area coverage of signals from the probes were calculated using the QuantimentQ500W (Lecia, Cambridge, England) image analysis system. DAPI was used for total microbial cell quantification. All hybridizations were performed as simultaneous dual color hybridizations where the EUB338 probe served the role of a general counter stain for all bacteria. The abundances of each bacteria group were calculated as the ratio of area covered by biomass stained simultaneously with probes and DAPI to the area covered by DAPI stained biomass alone. For each sample, 25 microscopic fields (92  92 mm) were analyzed. Signal intensity was quantified indirectly as the percentage of cells whose brightness exceeded the visual detection limit [20]. Images were acquired with an epifluorescence microscope (Nikon Corporation, Japan) equipped with a 100W mercury lamp, a 100/1.25 an oil

14264

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 6 ( 2 0 1 1 ) 1 4 2 6 1 e1 4 2 6 9

objective and appropriate filter sets for FITC (Croma Technology Corp. USA) and Cy3 (Nikon Corporation, Japan).

3.

Results and discussion

3.1. Hydrogen production from batch fermentation at different DM concentrations In our investigation, hydrogen free of methane, with a hydrogen content in the head space of 61.2  2.9% was achieved using digested manure as original inoculum which was exposed it to high load (shock load) of DM concentration of 50.1 g-sugar/L). During fermentation at these high loads, high concentrations of volatile fatty acids (VFAs) are accumulated resulting in acidification, and thus efficient inhibition of methanogens [23]. By subsequent, inoculations at the same conditions, the methanogens are gradually eliminated. The inoculum obtained from exposure of digested manure to high concentrations of organic matter, was effective to select hydrogen producing bacteria and suppress methanogens, as no methane was detected during the batch cultivation with a DM concentration range of 1.5e50.1 g-sugars/L. Furthermore, hydrogen was produced with short lag time (w10 h) and later reached stationary phase within 3 days, suggesting that the level of inhibitors such as acetate, lactate, and sulfate contained in that range of DM concentrations has insignificant impact on the fermentation inhibition. The cumulative hydrogen increased with increasing DM concentration from 1.5 g-sugars/L to 16.7 g-sugars/L, whereafter no further increase was observed. (Fig. 1A). However, as shown in Fig. 1B, increasing DM concentrations from 2.1 g-sugars/L to 33.4 g-sugars/L, resulted in decreased hydrogen yields, indicating that utilization of organic matter in DM was decreasing. This could be have been caused by improper food to microorganisms ratio (F/M), Improper F/M could consequently lead hydrogen yield reduction via accumulation of organic acids, and consequently reduction of pH [24,25]. Thus, the hydrogen yields peaked at 237 ml-H2/g-sugar during the DM batch fermentation at 2.1 g-sugar/L, indicating that most proper F/M could be achieved. Meanwhile, slightly lower yield of 211 ml-H2/g-sugar obtained during the batch fermentation with 1.5 g-sugar/L could be possibly due to main part of the organic substrate under low substrate loading conditions could go primarily to maintain the cell growth, resulting in lower hydrogen production per substrate amount [26]. Table 2 demonstrates the metabolic concentrations achieved at the end of the DM batch fermentation at 1.5 gsugar/L and 2.1 g-sugar/L using mixed culture initially exposed to high DM loads. The fermentation end products were dominated mainly with butyrate and followed by acetate. Additionally, it is noticeable that lactate contained significant amount in the DM was undetectable, indicating that it was degraded during the fermentation as well. Additionally, Baghchehsaraee et al. [27] have previously found that degradation of lactate contained in starch could generate more residual NADH for hydrogen production via butyrate pathway. Clostridium spp. and sulfate reducing bacteria (SRB) of Desulfobacterium spp. can convert lactate to acetate, propionate, and

butyrate [28,29]. Regarding to the main fermentation products as shown in Table 2 (hydrogen, butyrate, and acetate), the possible metabolic reaction of simultaneous degradation of hexose and lactate is demonstrated in the Eq. (1). 3C6 H12 O6 þ 2C3 H6 O3 þ H2 O /C2 H4 O2 þ 3:5C4 H8 O2 þ 9H2 þ8CO2 DG  ¼ 244:8kJ=mol

(1)

The experimental hydrogen yield of 237 mL-H2/g-sugar, which is corresponding to w13% of COD from sugar (Table 2) is however still lower than above mentioned theoretical yield of which is 373.5 mL-H2/g-hexose. This could be explained by the conversion of some of substrate into cell mass and other metabolic products than butyrate and acetate. Hydrogen is either not produced, or is consumed for the generation of other metabolic products than acetate and butyrate. Furthermore, consumption of H2 homoacetogenesis can also cause low hydrogen production. Microbial biomass during the thermophilic fermentation has normally COD based yield of w14%e16% [9], consistent with our finding in Table 2.

3.2.

Hydrogen production from DM in the UASB reactor

Inoculum enriched for hydrogen in the batch fermentation with DM concentration of 16.7 g-sugar/d$L was then immobilized on granules in a UASB reactor and subsequently fed with DM concentration of 16.7 g-sugar/d$L at a 24-h HRT. The performance of UASB reactors is presented in Fig. 2. For 60 days of operation, no methane was detected, indicating that methanogenic archaea were completely repressed by the operating conditions in the attached growth reactors, due to the low pH between 5.0 and 6.0 that was established in the reactor, which was self regulated by bicarbonate buffer added the soluble end products generated in the process. For more than 4 weeks after start feeding, fluctuation of soluble fermentation products was observed with increasing tendency of sugar degradation, and hydrogen production. Eventually, stability on hydrogen production was observed in the last 2e3 weeks of the operation period along with 86% sugar degradation and hydrogen production rate and yield of 4500 mL-H2/d$L and 269.5 mL-H2/g-sugar (159.6 mL-H2/VS), respectively. Hydrogen energy recovery is accounting to 54% of the theoretical potential of 498 mL-H2/g-sugar. Under thermophilic conditions, comparable hydrogen production yields of 114 and 288 mL-H2/g -sugars were previously achieved when food waste and palm oil mill effluent were fed in to a packed-bed reactor and an anaerobic sequencing batch reactor, respectively [7,30]. The stability of the UASB reactor with high hydrogen production rate and yield is mainly attributed to good adaptation of the H2 producing microorganisms, which are attached on granules, to the new environment of continuous mode of operation feeding with DM. It’s noticeable that high hydrogen yields were achieved with the UASB reactors with the soluble end products dominated mainly by butyrate and followed by acetate, indicating the fermentative hydrogen reaction was a butyrate-acetate fermentation type as shown in Eq. (1). However, lower hydrogen yields in our experiments compared to the theoretical yield (373.5 mL-H2/g-hexose) possibly generated in

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 6 ( 2 0 1 1 ) 1 4 2 6 1 e1 4 2 6 9

14265

A

B

Fig. 1 e Cumulative H2 during batch fermentation at various DM concentrations (A) and H2 yield and % H2 obtained from batch fermentation at various DM concentrations (B).

Eq. (1) could be explained with some conversion of sugar into biomass and other intermediate products of ethanol and propionate. Apart of the consumed hexose is used for microbial biomass growth (in this study assumed to account of approx. 15% of the influent COD) [9], while other products than butyrate and acetate are the competitive mechanisms [31]. Interestingly, lactate was detected in significantly low amount in the reactor effluent compared to its content in the influents during stable operation in the UASB reactor (Fig. 2), indicating that lactate fermentation was not the predominant pathway and mixed thermophiles can degrade lactate very well. Additionally, high removal of lactate was found consistently with the observed increased in hydrogen production via acetate and butyrate [32]. DM is therefore considered to be a potential substrate for continuous hydrogen production. In our experiments, the UASB reactor indicated good hydrogen production from DM continuously fed at rather high OLR of 16.7 g-sugar/d$L. This could be mainly attributed to high density of hydrogen producing microorganisms on granular sludge in the UASB reactor, leading to high activity for converting sugar to hydrogen [33,34]. The good performance of the UASB reactor, in which inoculum originally pretreated by shock loads of digested manure, could be evaluated as an effective and practical way to enrich hydrogen producing cultures, while excluding methanogens. However, only around 20% of sugar content in the DM contributed to hydrogen, while the rest was mainly remained in the soluble form of organic acids. Thus, sequent treatment of the effluent in the methanogenic reactor is needed to recover more energy

in the form of methane remove organic pollution of the effluent [35]. A mixture of hydrogen and methane mixture obtained by a two stage anaerobic process can be utilized in the internal combustion engine as the superior fuel compared to methane alone [36].

3.3. Microbial community composition in batch cultivation and attached growth reactors Fig. 3 shows microbial composition of sludge samples from batch fermentation and the UASB-granules after 54 days of operation. The fluorescent in situ hybridization (FISH) analysis of the inoculum obtained from batch fermentation with DM as substrate, at initial concentration of 2.1 g-sugars/L was dominated by the hydrogen producing bacteria Thermoanaerobacterium spp. The microbial composition of this inoculum was mostly comprised of Eubacteria with 88% of total DAPI binding cells. Hybridization using probes targeting the Archeae domain also showed few positive cells in all samples with 2% of total DAPI binding cells. The microbial community structure of thermophilic inoculum treated by load shock comprised of 49% of Thermoanaerobacterium spp., 16% of T. thermosaccharolyticum, 15% of phylum Firmicutes most Clotridium, Bacillus and Desulfobacterium, 2% of Archaea and no members of Caldicellulosiruptor spp. and Thermotoga spp., respectively. Meanwhile, the microbial community structure of thermophilic UASB reactor fed with DM concentration of 16.7 g-sugars/L comprised 36% of Thermoanaerobacterium spp., 10% of T. thermosaccharolyticum, 27% of phylum Firmicutes most Clotridium, Bacillus and

14266

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 6 ( 2 0 1 1 ) 1 4 2 6 1 e1 4 2 6 9

Table 2 e Metabolic product concentrations and COD balance during batch fermentation at DM concentration of 1.5 and 2.1 g-sugars/L. Concentration mM

% COD distribution mg-COD/l

Batch (g-sugar/l)

1.5

2.1

1.5

2.1

1.5

2.1

Consumed sugar Consumed lactate Acetate Butyrate Propionate Ethanol Butanol Hydrogen Cell mass (Balance)a

8.3 4.1 3.0 5.7 1.0 1.6 0.5 13.8 1.8

11.3 5.7 4.7 8.5 1.4 1.3 0.6 22.1 2.4

1593.6 390.61 193.3 908.8 116.48 151.2 101.8 220.27 292.4

2246.4 546.9 302.7 1352.3 152.1 127.7 115.2 353.4 389.8

80.3 19.7 9.7 45.8 5.9 7.6 5.13 11.1 14.7

80.4 19.6 10.8 48.4 5.5 4.6 4.1 12.7 14.0

a Balance value ¼ total consumed COD-total produced COD.

Desulfobacterium, 3% of Archaea. No members Caldicellulosiruptor spp. and Thermotoga spp., were detected. The FISH images (Fig. 4) show that Thermoanaerobacterium spp. accounted for 49% of relative abundance in inoculum from 2.1 g-sugars/L DM batch fermentation, while in the granules from UASB reactor sludge accounted only for 36% relative abundance. Thus, the majority of bacteria existed in thermophilic sludge samples from the batch and UASB reactors were Thermoanaerobacterium spp. which are able to convert carbohydrates, sucrose, starch, xylose, and glucose to butyrate/acetate along with H2 and CO2 [37,38], corresponding to the fermentation products obtained from our fermentation reactors.

Several species of the genus Thermoanaerobacterium are known for their H2 production characteristics, including T. thermosaccharolyticum, Thermoanaerobacterium polysaccharolyticum, Thermoanaerobacterium zeae, Thermoanaerobacterium lactoethylicum and Thermoanaerobacterium aotearoense. All these organisms have optimal growth conditions at 55e70  C and at pH 5.2e7.8 [39], which are accordance with operating conditions used in this investigation. The predominant phylogenetic groups of Thermoanaerobacterium spp. detected by FISH in our thermophilic hydrogen producing systems were found to be similar to those previously detected by DGGE [7,40]. Second dominant bacterial group found during our experiments in the

Fig. 2 e Performance of the UASB reactor operated at a 1-day HRT and fed with 16.7 g-sugars/L DM. LA: Lactate; AA: Acetate; BA: Butyrate; EtOH: Ethanol; PA: Propionate; FA: Formate.

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 6 ( 2 0 1 1 ) 1 4 2 6 1 e1 4 2 6 9

14267

Fig. 3 e Microbial compositions of sludge samples from 2.1 g-sugars/L DM batch cultivation and granules in thermophilic hydrogen producing UASB reactor. The error bars indicate the standard deviations from a triplicate sampling analysis.

Fig. 4 e Fluorescent image of thermophilic hydrogen producing sludge (A) and (C) are image of sample from batch cultivation at DM concentration of 2.1 g-sugars/L and UASB reactor, respectively stained with DAPI for total microorganisms, (B) and (D) are image of samples from batch cultivation and UASB reactor, respectively hybridized with Tbm1282 probe labeled with Cy3 for detected bacteria Thermoanaerobacterium spp. group.

batch and UASB reactors was genera Clostridium, Bacillus and Desulfobacterium, spore forming bacteria [41]. Clostridium and Desulfobacterium spp. are able to degrade lactate to acetate and/ or H2 [29]. Therefore, the significant removal of lactate found could be explained with the degradation activity of Clostridium and Desulfobacterium spp.

4.

Conclusions

This study demonstrates that load shock of anaerobic digested manure is potentially method to suppress methanogens. No methane could be detected in the mixed gas produced. The

14268

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 6 ( 2 0 1 1 ) 1 4 2 6 1 e1 4 2 6 9

highest yield of 237 mL-H2/g-sugar was obtained during the DM batch fermentation at 2.1 g-sugar/L The bacterial community analyzed by FISH method was dominated by Thermoanaerobacterium spp. Further immobilization of the enriched inoculum in the UASB reactor could achieve high and stable hydrogen production. The UASB reactor showed good sugar and lactate degradation, and hydrogen production. By continuous feeding DM with an OLR of 16.7 g-sugar/d$L at a 24-h HRT in the UASB reactor, a hydrogen production rate and yield of 4500 mL-H2/d$L and 269.5 mL-H2/g-sugar, respectively was achieved.

Acknowledgement This work was supported by the PhD funds provided by the Ministry of Science and Technology of Thailand. Hector Garcia and Jens S. Sørensen are great acknowledged for their technical assistance.

references

[1] Hallenbeck P, Ghosh D, Skonieczny M, Yargeau V. Microbiological and engineering aspects of biohydrogen production. Ind J Microbiol 2009;49(1):48e59. [2] Levin D, Pitt L, Love M. Biohydrogen production: prospects and limitations to practical application. Int J Hydrogen Energy 2004;29(13):1425e6. [3] Guo W, Ren N, Chen Z, Liu B, Wang X, Xiang W, et al. Simultaneous biohydrogen production and starch wastewater treatment in an acidogenic expanded granular sludge bed reactor by mixed culture for long-term operation. Int J Hydrogen Energy 2008;33(24):7397e404. [4] Kongjan P, O-Thong S, Kotay M, Min B, Angelidaki I. Biohydrogen production from wheat straw hydrolysate by dark fermentation using extreme thermophilic mixed culture. Biotechnol Bioeng 2010;105:899e908. [5] Batstone D, Keller J, Angelidaki I, Kalyuzhny S, Pavlostathis S, Rozzi A, et al. Anaerobic digestion model no. 1 (ADM1). London, UK: IWA Publishing; 2002. [6] Luo G, Xie L, Zou Z, Wang W, Zhou Q. Evaluation of pretreatment methods on mixed inoculum for both batch and continuous thermophilic biohydrogen production from cassava stillage. Bioresour Technol 2010;101(3):959e64. [7] O-Thong S, Prasertsan P, Birkeland N. Evaluation of methods for preparing hydrogen-producing seed inocula under thermophilic condition by process performance and microbial community analysis. Bioresour Technol 2009; 100(2):909e18. [8] Zhu H, Be´land M. Evaluation of alternative methods of preparing hydrogen producing seeds from digested wastewater sludge. Int J Hydrogen Energy 2006;31(14): 1980e8. [9] Kotsopoulos T, Zeng R, Angelidaki I. Biohydrogen production in granular up-flow anaerobic sludge blanket (UASB) reactors with mixed cultures under hyper-thermophilic temperature (70 C). Biotechnol Bioeng 2006;94(2):296e302. [10] van Groenestijn J, Hazewinkel J, Nienoord M, Bussmann P. Energy aspects of biological hydrogen production in high rate bioreactors operated in the thermophilic temperature range. Int J Hydrogen Energy 2002;27(11):1141e7.

[11] Hawkes F, Hussy I, Kyazze G, Dinsdale R, Hawkes D. Continuous dark fermentative hydrogen production by mesophilic microflora: principles and progress. Int J Hydrogen Energy 2007;32(2):172e84. [12] Angelidaki I, Ellegaard L, Ahring B. Applications of the anaerobic digestion process. Adv Biochem Eng Biotechnol 2003;82:1e33. [13] Western Sugar Cooperative Inc. Molasses Desugarized Solubles, http://www.westernsugar.com/MDS.aspx; 2006. [14] Angelidaki I, Sanders W. Assessment of the anaerobic biodegradability of macropollutants. Rev Environ Sci Biotechnol 2004;3(2):117e29. [15] Dubois M, Gilles K, Hamilton J, Rebers P, Smith F. Colorimetric method for determination of sugars and related substances. Analytical Chemistry 1956;28(3):350e6. [16] APHA. Standard Methods for the Examination of Water and Waste Water. 19th ed. Washington, DC: American Public Health Association; 1995. [17] Fang C, Boe K, Angelidaki I. Anaerobic co-digestion of desugared molasses with cow manure; focusing on sodium and potassium inhibition. Bioresour Technol 2011;102(2):1005e11. [18] Stahl DA, Amann R. Development and application of nucleic acid probes. In: Stackebrandt E, Goodfellow M, editors. Nucleic acid techniques in bacterial systematics. Chichester, UK: John Wiley & Sons Ltd; 1991. p. 207e48. [19] Meier H, Amann R, Ludwig W, Schleifer KH. Specific oligonucleotide probes for in situ detection of a major group of gram-positive bacteria with low DNA GþC content. Syst Appl Microbiol 1999;22:186e96. [20] O-Thong S, Prasertsan P, Karakashev D, Angelidaki I. Specific detection of Thermoanaerobacterium spp., Thermoanaerobacterium thermosaccharolyticum, and Caldicellulosiruptor spp. in thermophilic biohydrogen producing system by fluorescent in situ hybridization (FISH). Int J Hydrogen Energy 2008a;33:6082e91. [21] Harmsen H, Prieur D, Jeanthon C. Group-specific 16S rRNA targeted oligonucleotide probes to identify thermophilic bacteria in marine hydrothermal vents. Appl Environ Microbiol 1997;63:4061e8. [22] Amann RI, Ludwig W, Schleifer KH. Phylogenetic identification and in situ detection of individual microbial cells without cultivation. Microbiol Rev 1995;59:143e69. [23] Voolapalli R, Stuckey D. Hydrogen production in anaerobic reactors during shock loadsdinfluence of formate production and H2 kinetics. Wat Res 2001;35(7):1831e41. [24] Kraemer J, Bagley D. Improving the yield from fermentative hydrogen production. Biotechnol Lett 2007;29(5):685e95. ¨ rlygsson J, et al. [25] Koskinen P, Lay C, Puhakka J, Lin P, Wu S, O High-efficiency hydrogen production by an anaerobic, thermophilic enrichment culture from an Icelandic hot spring. Biotechnol Bioeng 2008;101(4):665e78. [26] Magnusson L, Cicek N, Sparling R, Levin D. Continuous hydrogen production during fermentation of-cellulose by the thermophillic bacterium Clostridium thermocellum. Biotechnol Bioeng 2009;102(3):759e66. [27] Baghchehsaraee B, Nakhla G, Karamanev D, Margaritis A. Effect of extrinsic lactic acid on fermentative hydrogen production. Int J Hydrogen Energy 2009;34(6):2573e9. [28] Matsumoto M, Nishimura Y. Hydrogen production by fermentation using acetic acid and lactic acid. J Biosci Bioeng 2007;103(3):236e41. [29] Zellner G, Neudo¨rfer F, Diekmann H. Degradation of lactate by an anaerobic mixed culture in a fluidized-bed reactor. Wat Res 1994;28(6):1337e40. [30] Lee D, Ebie Y, Xu K, Li Y, Inamori Y. Continuous H2 and CH4 production from high-solid food waste in the two-stage thermophilic fermentation process with the recirculation of digester sludge. Bioresour Technol 2010;101(1S):42e7.

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 6 ( 2 0 1 1 ) 1 4 2 6 1 e1 4 2 6 9

[31] Angenent L, Karim K, Al-Dahhan M, Wrenn B, Domı´guezEspinosa R. Production of bioenergy and biochemicals from industrial and agricultural wastewater. Trends Biotechnol 2004;22(9):477e85. [32] Castello´ E, Santos CG, Iglesias T, Paolino G, Wenzel J, Borzacconi L, et al. Feasibility of biohydrogen production from cheese whey using a UASB reactor: links between microbial community and reactor performance. Int J Hydrogen Energy 2009;34:5674e82. [33] Kim J, Kim Y, Ryu J, Song B, Kim I, Yeom S. Immobilization methods for continuous hydrogen gas production biofilm formation versus granulation. Proc Biochem 2005;40(3):1331e7. [34] Zhang Z, Show K, Tay J, Liang D, Lee D. Enhanced continuous biohydrogen production by immobilized anaerobic microflora. Energy Fuels 2008;22(1):87e92. [35] Hallenbeck P, Ghosh D. Advances in fermentative biohydrogen production: the way forward? Trends Biotechnol 2009;27(5):287e97. [36] Gattrell M, Gupta N, Co A. Electrochemical reduction of CO2 to hydrocarbons to store renewable electrical energy and upgrade biogas. Energy Conversion Manage 2007;48(4): 1255e65. [37] Lee YE, Jain MK, Lee CY, Lowe SE, Zeikus JG. Taxonomic distinction of saccharolytic thermophilic anaerobes description

[38]

[39]

[40]

[41]

14269

of Thermoanaerobacterium xylanolyticum gen nov, sp nov, and Thermoanaerobacterium saccharolyticum gen nov, sp nov reclassification of Thermoanaerobium brockii, Clostridium thermosulfurogenes, and Clostridium thermohydrosulfuricum E10069 as Thermoanaerobacter brockii comb nov, Thermoanaerobacterium thermosulfurigenes comb nov, and Thermoanaerobacter thermohydrosulfuricus comb nov, respectively and transfer of Clostridium thermohydrosulfuricum 39e to Thermoanaerobacter ethanolicus. Int J Syst Bacteriol 1993;43:41e51. Zhang T, Liu H, Fang H. Biohydrogen production from starch in wastewater under thermophilic condition. J Environ Manage 2003;69:149e56. O-Thong S, Prasertsan P, Karakashev D, Angelidaki I. Thermophilic fermentative hydrogen production by the newly isolated Thermoanaerobacterium thermosaccharolyticum PSU-2. Int J Hydrogen Energy 2008b;33(4):1204e14. Ueno Y, Sasaki D, Fukui H, Haruta S, Ishii M, Igarashi Y. Changes in bacterial community during fermentative hydrogen and acid production from organic waste by thermophilic anaerobic microflora. J Appl Microbiol 2006; 101(2):331e43. Onyenwoke R, Brill J, Farahi K, Wiegel J. Sporulation genes in members of the low Gþ C Gram-type-positive phylogenetic branch (Firmicutes). Arch Microbiol 2004;182(2):182e92.