Influence of organic loading rates on continuous

H2 production by a membrane bioreactor from food waste

Influence of organic loading rate[JH1]  on continuous H2 production from food waste in membrane bioreactor  






The possibility of [JH2]  long-term stability of stability of H2 fermentation in an MBR (HF-MBR) with food waste by membrane bioreactor (HF-MBR) from food waste was investigated. An HF-MBR was started up using a heat-pretreated anaerobic mixed sludge to inactive activate H2-consuming bacteria, and acclimated with 4.5% total solids (TS) food waste slurry. The HF-MBR was operated at the under various[JH3]  organic loading rates (OLRs) of 70.2, 89.4 and 125.4 kg-COD/m3/day, corresponding to the hydraulic retention times (HRTs) of 18.7, 14.0 and 10.5 hrs, respectively. The biogas production at for the H2 content range of 44 ~ 48% H2 content range increased from 22.38 to 32.82 and 62.49 l/day when with the OLRs increasesd. The maximum H2 yield and H2 production rate were 111.1 ml-H2/g-VS added and 10.7 l-H2/l/day at an OLR of 125.4kg-COD/m3/day. The major volatile fatty acids (VFAs) produced from the degradation of organic fractions were acetate and butyrate. The Ttotal carbohydrate degradation was beyond better than 96% throughout the experimental runs. Continuous H2 production by HF-MBR from food waste of TS 4.5% TS food waste absent CH4 production at 5.5±0.1 pH was successfully achieved sustained in the HF-MBR for 90 days. without CH4 production at pH control of 5.5±0.1 in the bioreactor. The copy number of acidogens 16S rRNA gene was around 3.25 ´ 107 copies/ml, and whereas that of archaea was not detected. The microbial community is was predominated by[JH4]  Clostridium sp. strain Z6. The H2 production was significantly improved by shortening the HRTs and increasing the OLRs. The HF-MBR had showed a higher degradation potential and H2 production capacity at the high OLRs due to its higher cell- retention. in the bioreactor.[JH5] 


Keywords: Membrane bioreactor; hydrogen production; organic loading rate; food waste



1.      Introduction

H2 has enormous potentials as a future clean fuel of the future, which  will create benefits for the economy promising significant economic and as well as make contribution to the protection of the global warming environmental benefits, has enormous potential. World H2 production is around 500 billion Nm3/year. Most of world H2 this is mostly recovered from fossil fuels, such as typically[JH6]  by steam reforming[JH7]  of natural gas and gasification of coal (Hart D 1997). The Water electrolysis of water is has been proposed as a alternative greener recovery methods alternative[JH8] , to recovery H2 but the process is costly. to run.[JH9]  Therefore Consequently, both H2 production approaches that are both environmentally friendly and cost effective approaches for its production have been more important coveted and pursued. Anaerobic digestion of organic wastes will be a the best option to produce for production of biogas and reduce reduction of environmental pollutions, during the gasification, especially COx, CnHm and SOx, respectively during gasification.

In Japan, industrial wastes were produced about 298 million tons of industrial wastes are produced every year, 6.7% of which, i.e. or 20 million tons, were from are food wastes. Most parts of food wastes are generated by food manufacturers, and restaurants, and hotels, including and include unsold food and lunch boxes at supermarkets and convenience stores, all of which are incinerated or go to landfills (BJCS, 2004). On the other hand, it is noticed that tThe high water content of food wastes, however, will be make them good choice suitable for bio-gasification, which is accomplished by by means of fermentation technology.

In the anaerobic fermentative pathway, H2 is an intermediate metabolite converted from the complex organic materials for  bi-products of the biochemical degradation process under oxygen-free conditions in the liquid phase. Theoretical H2 yields can be obtained with the maximal 4 moles together with 2 moles of acetic acid as the carbohydrate fermentation end products from the fermentation process of carbohydrate (Hawkes et a., 2007). Fang et al. (2006) reported that the achievement of H2 production from soluble wastewater including rice slurry has been achieved in a continuous stirred tank reactor (CSTR) using anaerobic mixed microflora. The Ppotentials of fermentative H2 production from organic wastes such as bean curd manufacturing waste, rice bran and wheat bran was were investigated using by batch experiments (Noike et al., 2000). Generally, H2 production by anaerobic microflora can be influenced by environmental conditions such as the organic loading rate (OLR), HRT, pH, mixing intensity, nutrients and temperature. It is has been pointed out determined that low H2 production in continuous cultures was is due to the cell wash-out and the unstable cell levels in the operational condition such as under shorten HRT and solids retention time (SRT) [JH10] conditions.[JH11] 

TheCell retention techniques of cells has been focused on  have been topics of research interest for their application to the continuous H2 fermentation process, which provides high H2 production in the acidogenic phase from the hydrolysis of the complex organic materials, compared with the continuous-flow systems like such as CSTRs. Various Ccells retention of various types techniques has have been applied to the wastewater treatment in fermentation processes such as granular sludge bed, the attached anaerobic reactor and MBR (Akutsu et al., 2009; Rachman et al., 1998; Lee et al., 2006; Gavala et al., 2006,; Kim et al., 2008). Among cell retentions,[JH12]  The membrane MBR process is an extremely useful technique[JH13]  which for selectively separates separation of the requested materials via membrane pores, and is widely used in various fields including waste treatment, purification, and enrichment (Nomura et al., 2002). In the fermentation process, the membrane provides for excellent organic acid/bacterial cell separation efficiency, between organic acids and bacterial cells, reflecting the MBR’s higher biogas recovery efficiency compared with the conventional continuous stirred tank reactors (CSTRs)[JH14]  (Keith et al., 1995). Our previous study investigated the characteristics of fermentative H2 production at various solids retention time (SRTs) at under the mesophilic condition using in a submerged membrane bioreactor MBR (Lee et al., 2010b). Sustainable CH4-free H2 production with CH4-free was observed at for long SRTs with the control of pH (5.5) in the a mixed culture.

However, fermentative H2 fermentation in an MBR (HF-MBR) from with high-solids-content food waste with high-solids content has not been yet to be reported. yet. As such, currently there is no information available on the feasibility of long-term continuous HF-MBR operation with food waste. The drop of the pPermeate flux drop by membrane clogging and the change of bacterial community alteration by solids retention [JH15] will be are serious obstacles for to continuous operation (Hawkes et al., 2007; Lee et al., 2008). No information is available on the possibility of long-term operation in a continuous H2 fermentation in MBR from food waste.[JH16]  Therefore, iIt is necessary, therefore, to find that the important[JH17]  key to achieve[JH18]  sustainable H2 production in a HF-MBR be found. To that end, Tthe objective of this the present study was, first, to confirm the possibility of long-term stability stability of continuous HF-MBR from TS 4.5% TS using fermentative anaerobic bacteria and, second, to investigate the pertinent H2 production capacity at various organic loading rates (OLRs).


2.      Methods

2.1 Inocula and feedstock

Seed sludge was taken at from a digester storage tank in in the a CH4 fermentation plant for the combination combined treatment of food waste and sewage sludge. Anaerobic sludge was acclimated with food waste slurry of TS 4.1% TS food waste slurry (FWS) in a CSTR which that was operated had been in operation[JH19]  beyond for more than two years. The pH, VS, and alkalinity of sludge the acclimated sludge were 5.5, 38,100 mg/l and 1,650 mg CaCO3/l, respectively. The microbial community is was dominated by Clostridium sp. strain Z6. 

Food wastes were collected from a cafeteria at the National Institute of Environmental Studies, which consisted consisting of a great of variety of grains, vegetables, meats and fishes. As shown in Fig. 1, the food wastes were inputted loaded into the bioreactor after only once-per-week mechanical pretreatment using a circulation pump with and a cutting apparatus. once a week. TS concentration in  4.5% TS food waste slurry ranged from 4.5% was prepared by the dilution of tap-water and crushed crushing with the a cutting pump at a vacuum liquid circulation of 100 L/min and dilution with tap water to maintain a particle size of 5mm (Zenoah, Model: KD50MS). The pH in the FWS was in range of maintained at 4.3 ± 0.1. The characteristics of the FWS used in this study[JH20]  are represented listed in Table 1. The VS/TS ratio was 0.96 ± 0.019.


2.2  Operation of HF-MRB

The laboratory-scale submerged MBR of a laboratory-scale is shown in Fig. 1. The 5 L volume rectangular bioreactor, which was made from an acrylic resin, with a total volume of 5 L, has been employed for install was designed for two membrane modules. Each module (module size: 240 ´ 340 ´ 10 ) consists of The a membrane was a plate-flame-type membrane and the situated in a membrane cartridge was produced by Kubota.. The membrane  Each membrane had has[JH21]  a pore size of 0.45mm and an effective surface area of 0.1. per membrane module (module size : 240 ´ 340 ´ 10 ). Two membrane The modules were submerged in the bioreactor at intervals of 10 mm apart between membrane modules into bioreactor to effectively brush away solidss fractions attached on the membrane surfaces. [JH22] 

As shown in Table 1, a semi-continuous operation was employed conducted for the reduction of organic factions under the thermophilic temperature of 55± 0.5, which temperature was maintained constant by means of circulating water using within a water jacket surrounding surrounding the bioreactor. to keep a constant temperature desired. The MBR was operated at various OLRs of 70.2, 89.4 and 125.4 kg-COD/m3/day, corresponding to HRTs of 18.7, 14.0 and 10.5 hrs, respectively. The HRT-reducing Ffeeding periods for reducing HRTs were increased stepwise incrementally from 24 and and 48 to 96 cycles per day, corresponding to fed and non-fed periods;: the 1 min-On and 59 min-Off, 2 min-On and 28 min-Off, and finally 3 min-On and 12 min-OFF, respectively. To minimize cake formation on the membrane, firstly, coarse bubbles using a biogas re[JH23] circulation pump (T.G.K: FP-15N) in the headspace of the bioreactor were supplied coarse bubbles under to the underside of each membrane module at the flow rate of 3.5 L/min; using a biogas circulation pump (T.G.K: FP-15N), and secondly, permeates were intermittently extracted by a vacuum pump controlled by a programmed timer (TERAOKA, timely-S). TThe Food waste slurry FWS, additionally, was also semi-continuously fed into the bioreactor in parallel with the extraction of the permeates at the regulated time using peristaltic pumps.

In order Tto avoid prevent the excessive biomass accumulation, of biomass in the bioreactor[JH24] , the HRT/SRT ratios were maintained to at an equal equivalent value of 0.25 for each experimental condition, and the wasting of suspended surfeit sludge was carried out eliminated by as overflowing when new feed was fed into the bioreactor. The suction pressure of the membranes was monitored using a vacuum gauge, and flux obtained through the permeate line was continuously measured in order to maintain stable HRTs for the different experimental runs. The pH in the mixed culture was controlled by by means of a pH controller (Mettler Toledo, pH 2050e) using incorporating 2.0 N NaOH to maintain the pH[JH25]  value in range of at[JH26]  5.5±0.1.


2.3 Microbial community analysis

The predominant bacterial type was investigated determined by the 16S rDNA sequence analysis. The DNA was extracted from the mixed liquid in the HF-MBR and purified using extrap solid DNA kit plus  by the QProbe-PCR (polymerase chain reaction) method (J-Bio 21, Japan) using an extrap[JH27]  solid DNA kit plus. The concentration in the extracted DNA solution was measured by using a Picogreen dsDNA assay kit (Invitrogen). The volume of the purified DNA solution was 50 mL, and the concentration of that[JH28]  was 32.1g ng/mL. The PCR amplification of 16S rDNA sequences was carried out using the primer sets of 27f (5’-AGAGTTTGATCMTGGCTCAG-3’) (Marchesi et al., 1998) and Bac1392R (5’-ACGGGCGGTGTGTAC-3’) primer sets[JH29]  (Kirsti et al., 2006). The cycles of PCR cycles, performed with an automatic thermal cycler iCyclerTM (Applied Biosystems), were determined according to results obtained from the monitor of a QPrimer-PCR: . PCR was  performed with an automatic thermal cycler iCyclerTM (Applied Biosystems) using the following thermal program: an initial denaturation at 95 oC for 30 sec, 20 cycles of denaturation at 95 oC for 15 sec, annealing at 50 oC for 20 sec, and extension at 72 oC for 50 sec, with a final hold at 40 oC for 30 sec.

The amplified-PCR products of 16S rDNA were cloned using a TOPO TA PCR cloning kit (Invitrogen), according to the manufacturer’s instructions. A total of 32 clones were isolated and classified. The PCR products were purified and sequenced using a BigDye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems) and a model ABI3730xl sequencer (Applied Biosystems). Comparative analyses of the full-length sequences were carried out by the BLAST search. The sequences of the[JH30]  closest sequences were retrieved from GenBank.


2.32.4 Quantitative analysis by real-time PCR 

The DNA for the quantitative analysis of acidogens and methanogenic archaea was extracted by the same procedure as described in sSection 2.2[JH31] . The copies copying of 16S rRNA gene originated originating from the acidogens and methanogens was performed with the LightCycler 1.0 (Roche Diagnostics). The primer sets for the dominated dominant acidogens were Bac105YF and Bac1392R (Ritalahti et al., 2006). The PCR program process[JH32]  for the quantification of acidogens consisted of a DNA denaturation step at 95 oC for the initial 120 sec, 50 cycles of repeated denaturation at 95oC for 15 sec, annealing at 61oC for 20 sec, and extension at 72oC for 25 sec. The primer sets for the dominated dominant archaea were ARC787F and ARC1059R (Yu et al., 2005). The PCR condition process was as follows: initial denaturing step for 120 sec at 95oC, followed by 60 cycles of denaturation at 95 oC for 10sec, and annealing 30sec at 60oC for 30 sec. The standard curves for the acidogens and archaea was were determined from the PCR product amplification of PCR products using the genomic DNA of the Escherichia coli K12 (ATCC 10798) and Methanobacterium bryantii M.o.H. (ATCC 33272), respectively, as the standard DNA (R2 = 0.9992).  


2.42.5 Analytical methods

The biogas contents, including H2, CH4, CO2 and N2, were analyzed using a gas chromatograph equipped with a thermal conductivity detector (GC-8A, Shimadzu) and a φ 3 ´ 2 m stainless column packed with ShinCarbon ST 50/80 mesh (Shimadzu GLC Ltd., Japan). A The flow rate of argon, as the carrier gas, was 50ml/min. Lactate assay kit (BioVision, USA) was measured at O.D. 570nm for colorimetric assay.[JH33]  The volatile fatty acids (VFAs) and the alcohol were analyzed using a gas chromatograph (GC-14B, Shimadzu Co.,) equipped with a flame ionization detector (FID) and a Stabilwax-DA capillary column (30m ´ 0.53 mm ID, Resteck, USA). The total CODcr was measured by spectrophotometry at 620 nm, according to the closed reflux colorimetric method (HACH, DR/2010, USA). The carbohydrate concentration was determined by the method of phenol sulfuric acid method (Dubois et al., 1956). The protein was analyzed using the Lowery method (Protein assay kit, Co.[JH34] ,). The total nitrogen (TN), total phosphate (TP), NH4-N and PO4-P were determined by using an auto-analyzer TrAACs 8000 using equipped with[JH35]  a colorimeter (Bran + Luebbe K.K., Japan). The TS, VS and alkalinity were analyzed according to standard methods (APHA, 1995).


3. Results and discussion

3.1 Behavior Characteristics[JH36]  of H2 productions

The OLRs gradually increased in three stages at under each experimental condition when the fermentation parameters, including biogas production, VFAs concentration and VS concentration, reached at the steady-state conditions. It was observed that, as the OLRs increased, there was continuous excessive foaming continuously occurred through the biogas diffuser under the membrane module, resulting from the physiological changes of the anaerobic microbial consortium. when OLRs increased. Biogas flow rate was temporarily regulated[JH37]  between 2 and 3 l/min to prevent the damage of to the biogas recirculation pump without the when there was no addition of anti-foam reagent.

Figure. 2 shows the variation of the biogas production and biogas contents. The fluctuation in biogas production was somewhat observed only slight, due to little change in permeate flux[JH38] . The margins of error in the permeate flux was were[JH39]  less than 1.3%. Biogas production was gradually increased incrementally with the increase of increasing OLRs, in steps, which was and stably continued during the given operational periods. At variable OLRs As the OLRs were increased from 70.2 to 125.42 kg COD/m3/day, the biogas productions, s at various specific OLRs, were 22.4 ± 1.5, 32.8 ± 1.8 and 62.5 ± 3.7 l/day, corresponding to 9.9 ± 0.6, 15.6 ± 0.9 and 31.1 ± 2.7 l-H2/day. when OLRs increased from 70.2 to 125.42 kg COD/m3/day.  At the OLR of 70.2 kg-COD/m3/day, the content of[JH40]  biogas produced was mainly composed mainly of H2 of (at 44.1% ± 1.3%) and CO2 of (at 46.1% ± 1.3%),. respectively. At OLRs between 89.37 and 125.42 kg-COD/m3/day, the H2 content slightly increased from its level at the 70.2 kg-COD/m3/day OLR, H2 content slightly increased in from the range of 47.5% - to 48.5%. as compared with that in the OLR of 70.2 kg-COD/m3/day. This result showed that H2 production with MBRs was is favorable for high OLRs and shortens HRTs.

Sustainable H2 productions with in the MBR under thermophilic conditions [JH41] was achieved at considerably higher OLRs and under lower HRTs compared to with previous studies reported reporting from on food wastes (e.g.[JH42]  Kim et al., 2008). The CSTR, indeed, has been widely applied at under various operating conditions due to its simple design to control system[JH43] . However, the cell retention in the CSTR is the CSTR’s limiting factor because is cell retention, in that HRTs are equal to SRTs. The washout of cells in the bioreactor is frequently occurred occurs at low HRTs, and the the result being the failure of the H2 production process. is resultingly fail. Particularly Specifically, H2 fermentation from organic solids is very poor in the CSTR, is very poor due to low degradation of organic fractions and the aforementioned washout of cells. Kim et al. (2008) suggested that continuous H2 fermentation using an anaerobic sequencing batch reactor (ASBR) provides for cells retention in the bioreactor by means of the control of the SRT, independent from that of the HRT. From the point of biotechnology-based viewpoint, MBRs will be new challenge to  represent the future of  develop the fermentation technology development for bio-energy recovery[JH44] . MBRs lead to allow for the enhancement of the completed retention in organic solid fractions and microorganisms, which could result in can increasing increase the H2 production from the hydrolysis of organic fractions. in the bioreactor[JH45] . The HRT/SRT in the MBR will be are one of the important factors to improve for improvement of the membrane permeability of membrane and reduce the reduction of membrane biofouling. of membrane.

In the present study, the permeate Fflux was increased from 0.6 and 0.8 to 1.0 l//day in order to maintain[JH46]  the HRT within range of the 18.67-10.5 hr range. The permeate flux was monitored to security the margin of error (±1%) in HRT several times a per day.[JH47]  A relatively flux was stably[JH48]  maintained at OLRs between 70.2 and 89.37 kg-COD//day; where an unstable flux was evident at at the OLR of 125.42 kg-COD//day, was observed due to the drop of the permeate flux by resulting from membrane bio-fouling. After 91 days of operation[JH49] , the permeate flux was rapidly decreased from 1.0 to 0.2 l/m2/hr (figure not shown) and, due to the pressure difference across the membrane surface resulting from the formation of a cake layer, the a biogas bubble in the permeate line was observed formed. due to the pressure difference across membrane surface by the formation of cake layer.  From this result Nonetheless, the feasibility of long-term (90 days) HF-MBR operation of 90 days using the HF-MBR was confirmed, which fact can bolsters the possibility on a of sustainable H2 production from organic fractions of municipal solid waste. The permeate extracted from MBRs could be further application applied in to bioenergy recovery processes, for the to photo-H2 production by photosynthetic bacteria, or for to CH4 production (using via wet- or dry fermentation) by methanogenic bacteria.  


3.2 Organic matter removal

Tables 2[JH50]  shows the changes of the COD, carbohydrate and protein concentrations in the permeate, for reflecting the HF-MBR’s performance from for food waste of TS 4.5% TS food waste. The Aaverage CODcr, carbohydrate, and protein concentrations in the influent was were 52.7 ± 2.8, 17.7 ± 0.4 and 11.35 ± 0.5 g/L, respectively. The carbohydrate removal efficiencies of carbohydrate in permeate[JH51]  were beyond 97%, ranging from 0.56 ± 0.08 to 0.66 ± 0.03g/l, for all of the different[JH52]  OLRs. The Pprotein concentrations in the permeate ranged from 0.48 ~ to 0.83g/L, representing nearly the maximal 95.7% of removal efficiency, throughout the experimental runs. The VS concentration of VS in the bioreactor was stably kept maintained at 5.96 ± 0.28%, which corresponding to a nearly 1.38 times increase over that in the influent. A The solids fraction in the feed was completely separated by with the membrane pore size of 0.4 mm. However, organic fractions derived from biofilm formation of a biofilm was were observed in the permeate line for the operational periods, as a result of owing to the long retention of soluble substances filtered inside the membrane module,, when an itself caused by an increase in pore size caused by resulting from the predominance of the thermodynamic effect. is predominant. Membrane permeability for operation has gotten better at shorten HRT of 10 hr without the discharge from any ones.[JH53]  Notably, it was observed that the a swell in a flat sheet of the membrane module was unavoidable for the during idle time of membrane operation,[JH54]  ranging from 29 (R 2) to 59 (R 1) min, to minimize membrane clogging by the attachment of solids fraction of membrane performance and membrane module for the idle time of 15min (R 3) was relatively stable.[JH55]  It can be thought considered, then, that it needs to develop a suitable membrane material and its together with its module configuration needs to be developed for the thermophilic fermentation process. The membrane module should be also provided offer high-permeability and high-temperature durability in order so as to maintain long-term process stability. in the process. The VS/TS ratios in the feed were within range of the 0.94 ~ 0.96 range, which indicated indicating that the food waste occupied had a high degradable degradability content. The VS/TS ratio in the HF-MBR was reduced from 8.6 to 9.2% as compared with that in the feed[JH56] . This result shows that the reduction of the VS was achieved at over long solids retention SRTs[JH57]  in the membrane bioreactor and could be more better converted into biogas via hydrolysis at under a the thermophilic condition of 55. Under thermophilic conditions, the hydrolysis rate is higher than that that in under mesophilic conditions. Therefore, the HF-MBR could tolerate a larger loading of organic solids. In general, hydrolysis of particulate organics in the fermentation process depends upon is a function of operational parameters including temperature, pH, HRT and substrate concentration. If the hydrolysis rate is the a first-order reaction about for biodegradable organic matters[JH58] , it could can be expressed as equation (1). in the HF-MBR[JH59] . Under steady-state conditions, the change of MLSS (P) is[JH60]  almost zero. Accordingly,[JH61]  Hhydrolysis constants (k) were observed at 0.9, 1.2 and 1.6 (1/d), respectively, when the OLRs were increased.





where k is the reaction rate constant., P is the VS concentration in the bioreactor., and Pi is the VS concentration in the influent.

In the present study, the MBR could easily improve the retention of biomass in the bioreactor[JH62]  independently of HRTs for over long-term operations,. which are This is the suggested as a means to of obtaining the efficient H2 production, by that is, via the degradation of organic solids fractions and the fermentative pathways of bacteria. from the degradation of organic solids fractions.


3.3  Metabolites products

In the anaerobic fermentation, H2 is the intermediate product which that is positively correlated with the pathway of liquid-phase byproduct butyric acid and acetic acid metabolic[JH63]  pathways. as byproducts in the liquid phase[JH64] . The Ttotal metabolites observed in the present experimentation were increased from 15,399 mg/l and 19,268 mg/l to 20,933 mg/l with the increase of the OLRs, as shown in tTable 3. The major metabolites produced from by the degradation of the organic fractions in the food waste were, in fact, acetic acid and butyric acid., Aacetic acid ranged ranging from 11.8 to 15.1% and butyric acids accounted for from 24.4 ~ 27.9% range of the total metabolites, based on CODcr[JH65] . Propionic acid accounted for around 1.8 ~ 2.1% range when the OLRs were between 70.2 and 89.4 kg-COD/m3/day, and was decreased to 0.8% when the OLR was increased to 125.4 kg-COD/m3/day. Among the soluble metabolites, Llactic acid occupied fell in the range of 24.2 ~ 25% range among the soluble metabolites for each all of the OLRs; and its concentration was little increased with increase of increments in the OLR. The H2 fermentation pathway was dominantly, regardless of being in the form of lactic acid at high OLR levels. of OLR. [JH66] Lee et al., (2010b) reported that their low H2 yield[JH67]  was due to the metabolic pathway shift to lactic acid under SRT[JH68] . In the present study, the Iincrease in H2 production was accomplished by the cell retention under effected by the regulation of the SRT. The deficiency of the iron concentration in food wastes will be is a significant factor determining the formation of lactic acid. From these According to the experimental results, the increased in H2 production was resulted in less lesser production of propionic acid as well as more greater production of acetic acid and butyric acid., It indicates indicating that optimized the OLR in a MBR needs to be optimized in order to improve H2 production and avoid producing the butyrate-propionate via the fermentation pathway.

The effect of H2 production on the metabolic pathway has been studied in previously. other studies. Many researchers have been used employed the butyric acid/acetic acid (B/A) ratio as the quantitative indicator for inspecting of the H2 production pathway of the anaerobic bacteria. In the present study, the B/A ratio, based[JH69]  on the molar basis, was in the range of 1.65 ~ 1.7 range, which was observed almost the and showed similar[JH70]  values at all each of the OLRs. The molar B/A ratios were relatively high compared with the values[JH71]  reported of for thermophilic H2 fermentation from food waste of TS 10% TS food waste in our previous study (Lee et al., 2010a). It These results indicated that the optimal B/A ratio in the H2 fermentation is different differs depending on according to the cultivation conditions including culture, temperature and substrate conditions of cultivation. Chen et al. (2001) demonstrated the relationship between the B/A ratio and rH2 on in H2 production from a sucrose substrate. as substrate[JH72] . Kim et al. (2006) reported that the B/A ratio was beyond over 4.0 when the sucrose concentration was beyond more than 20g-COD/l, which conditions provides for favorable H2-producing metabolism in a CSTR[JH73] . Han and Shin (2004) reported that H2 production deteriorated when the B/A ratio was low (1.3)[JH74] . In this the present study, the B/A ratio from food waste of TS 4.5% TS food waste was mainly observed mainly at in the stable range at under all of the experimental conditions. This result indicates Therefore it could be concluded that that the metabolic pathway from the carbohydrate-rich food waste in the thermophilic HF-MBR kept maintained the the optimal B/A ratio[JH75]  for effective H2 formation due to the security effectiveness of the active cell retention in the bioreactor[JH76]  at at high OLRss and shortens short HRTss.


3.4 H2 yield

The A good correlation between the H2 yield and the OLRs was found with the H2 conversion in range of in the 10.4 ~ 18.4% H2 conversion range, and the efficiency of H2 production which was a the result of the a change in the metabolic pathway during the biodegradation of food waste, as shown in tTable 3. OLR of The 125.4 kg-COD/m3/day OLR allowed for the large amount of high H2 production and as well as for a the high quality of the permeate. The H2 yields, with rising OLRs, increased from 1.24 to 2.2 mol-H2/mol-hexose added, with increase of OLRs, which is a higher value compared with than achieved in our previous studies study[JH77]  (Lee et al., 2010a). As mentioned in sSection 3.2, the increase in H2 increase was resulted from the result of the degradation of organic waste via hydrolysis, which was effected by maintaining a higher microbial cells count. The incubation temperature of 55oC could also could enhance the substrate utilization rate of substrate and the decrease of the dissolved H2 concentration in the liquid phase (Duran and Speece 1997; Mu et al., 2006). In tTable 5, shows that the H2 yield using an ASBR was in ranging of around 80.9 ml-H2/g-VS at the HRT of 33 hr using food waste of VS 4.4% VS food waste (Kim et al., 2008). The Ccarbohydrate-rich biomass, via nitrogen or protein replenishment in sewage sludge, could successfully increased the H2 yield by the replenishment of sewage sludge as nitrogen or protein source[JH78]  (Kim et al., 2004; Kim and Lee. 2010a). Therefore, cContinuous H2 and CH4 fermentation in a two-stage process using including recirculation of digestion sludge, was significant therefore, meaning for improving can effectively improve H2 yields. Zhu et al. (2008) also demonstrated that an enhanced H2 yield, 112ml-H2/g-VS, was enhanced by the co-digestion of food waste, primary sludge and waste-activated sludge, which gave the H2 yield of 112ml-H2/g-VS  based on specifically a combination of feed and 250 ml-H2/g-VS based on food,. respectively.[JH79]  This result indicates that the feed stocks replenished contained nitrogen, metals and carbonate alkalinity, which is have important roles in improving H2 yields. It is interesting to note that, H2 yield in the HF-MBR was similar value of 111.1 ml-H2/g-VS but high H2 production rate was achieved at 10.7 l-H2/l/d compared with the previous other study using CSTR.[JH80]  This result[JH81]  inflected is an important illustration of the fact that H2 yields using from MBRs can be significantly can increased at the high OLRs and short HRTs without the addition of adding waste-activated sludge, sewage sludge or primary sludge. HF-MBRs have the high recovery potentials on in H2 formation via enriched bacteria at the higher loading capacity from the complex organic waste[JH82] . The retention of bacterial cells will be a crucial key factor for improving H2 production from the high-organic solids. 

3.5 Microbial community analysis

The clones developed from the sample were classified into three operational taxonomic units (OTUs). Table 3 shows that the three OTUs were closely related with to Clostridium, Thermoanaerobacterium, and Lactobacillus, respectively. The microbial community under the thermophilic condition of 55oC was less diverse, even though a food waste as complex substrate food waste was fed., It reflects evidencing that acidogenic bacteria demand high maintenance energy as well as nitrogen and phosphorus sources at high cultivation temperatures (Ziner et al., 1986). The microbial community is was predominated by Clostridium sp. strain Z6, that which was affiliated related to to 93.8 % among of the 32 clones, which that, together with acetic acid and butyrate as intermediate products, play an important role for in the production of H2. together with acetic acid and butyrate as intermediate product. 3.13% of total clones were affiliated to the Thermoanaerobacterium thermosaccharolyticum, which is known as an H2-producing bacteria in the thermophilic acidogenic fermentation, was related to 3.13% of the clones. The genus Thermoanaerobacterium, as a gram-type positive[JH83] , is are thermophilic anaerobes. T. thermosaccharolyticum can produce large amounts of H2 via the acetate-butyrate fermentation pathway at pH range of in the 5 - 6 pH range and at the cultivation temperatures of between 55 - and 60oC (Ueno et al., 2001; Akutsu et al., 2008). It has been reported that the H2 yield by T. thermosaccharolyticum is almost equal that by Clostridium butyricum (Shin et al., 2005). The rest remaining 3.13% of the clones was were related to Lactobacillus helveticus, which is well known as the bacteria for lactic acid production. L. helveticus can attain a high conversion efficiency into for lactic acid at a temperature of 42oC and a pH of 5.7 (Roy et al., 1986; Chiriani et al., 1992). It might be concluded, then, that Lactobacillus  is not going to cannot contribute to increasing higher H2 production rates (Jo et al., 2007). In fact, H2 production via the formation of lactic acid may could be as difficult as that given indicated in equation (3). On the other hand However, Yang et al. (2007) reported that high H2 production from cheese-processing wastewater was observed at the presence of when the genus Lactobacillus  was predominant over over Clostridia in the bacterial community.


Glucose   à   2Lactic acid          (3)


In general, the bacterial population depends on the is a function of cultural culture conditions such as temperature, pH, substrate and reactor type. Shin and Youn (2005) reported that T. thermosaccharolyticum dominated, as in fact was the only species at pH the 5.5±0.1 pH and temperature of a 55±1oC temperature in a CSTR. Kim et al. (2006) demonstrated that microbial communities under sparging of CO2 sparging were simplified as three species of C. tyrobutyricum, C. proteolyiticum and C. acidisoli[JH84] ., And but the did not observe any change in the bacterial population by N2 and internal biogas sparging. was not observed. Akutsu et al. (2008) reported that an H2 yield of 2.32 mol-H2/mol-glucose was obtained in thermophilic H2 fermentation from starch using a batch feeding reactor and that the Thermoanaerobacterium was predominant over Clostridium from for different seed sludges including waste-activated sludge, cattle manure and acidified potato. The Clostridium was dominated from in seed sludge in the of co-digested of the night soil[JH85]  and municipal organic waste. In the present study, the thermophilic HF-MBR was commonly mostly was dominated by the genus Clostridium, compared with similarly to the result reported in for  the mesophilic condition (Oh et al. 2004), due to the fact that the cultivation conditions were operated at similar pH in the range of pH likewise was controlled within the 5.4 ~ 5.6 range[JH86] .  (Oh et al. 2004). Both Clostridium and Thermoanaerobacterium, as the H2 producers, contributed largely to improving the improvement of the H2 productivity from by the degradation of food wastes.

The concentrations of the 16S rRNA genes of acidogenic bacteria and methanogenic archaea, at the OLR of 125.4 kg-COD/m3/day, was were quantified by real-time PCR as shown in table 6. It was and are expressed by in Table 6 as the Nno. of copies per 16S rRNA and per sample of ml, respectively. Acidogenic bacteria accounted for around 1.09 ´ 108 copies/mg-VS (3.25 ´ 107 copies/ml-DNA), and whereas methanogenic archaea was below the detection limit (1.03 ´ 101 copies/ml-DNA). This result indicated showed that the population level growth of methanogenic archaea was completely inhibited by the control of a the pH range of at 5.5±0.1, regardless of the organic solid accumulation and retention in the bioreactor. In conclusion, the microbial community in the thermophilic HF-MBR was favorable effectual to for sustainable H2 production despite microbial[JH87]  contamination from the complex organic waste.


4.      Conclusions

Continuous H2 production by MBR from food waste of TS 4.5% TS food waste was successfully achieved for 90 days with CH4-free biogas at a pH control of 5.5±0.1. in the bioreactor.[JH88]  The maximal H2 yield and H2 production rate were 111.1 ml-H2/g-VS added[JH89]  and 10.7 l-H2/l/day, respectively, at an OLR of 125.4 kg-COD/m3/day. The aAcetate and butyrate, as the major metabolites, were produced from the degradation of the food waste. The Ttotal carbohydrate degradation efficiency was beyond 96% throughout the experimental runs. The copy number of acidogens 16S rRNA genes at the OLR of 125.4 kg-COD/m3/day was 3.25 ´ 107 copies/ml-DNA, and whereas[JH90]  that of archaea was below the detection limit. The microbial community is was predominated by Clostridium sp. strain Z6, which, together with acetic acid and butyrate as intermediate products, plays an important role for in the production of H2. together with acetic acid and butyrate as intermediate product.  The The H2 production using MBR from food waste as organic solids[JH91]  was significantly improved by shortening the HRTs and increasing the OLRs. It This indicated that the MBR, due to its higher cell retention, had a higher degradation potential and a better H2 production capacity at high OLRs. due to its higher cell-retention in the bioreactor[JH92] .


Akutsu, Y., Lee, D.-Y., Chi, Y.-Z., Li, Y.-Y., Darada, H., Yu, H.-Q., 2009. Thermophilic fermentative hydrogen production from starch-wastewater with bio-granules. Int J Hydrogen Energy 34, 5061-5071.

Akutsu, Y., Li, Y.-Y., Tandukar, M., Kubota, K., Harada, H., 2008. Effects of seed sludge on fermentative characteristics and microbial community structures in thermophilic hydrogen fermentation of starch. Int J Hydrogen Energy 33, 6541-6548.

Biomass Japan comprehensive strategy (BJCS), 2004. 29 (in Japanese).

Chiriani, L., Mara, L., Tabacchioni, S., 1992. Influence of growth supplements on lactic acid production in whey ultrafiltrate by Lactobacillus helveticus. Applied Microbiology and Biotechnology 36, 461-464.

Chu, C.-F., Li, Y.-Y., Xu, K.-Q., Ebie, Y., Inamori, Y., Kong, H.-N., 2008. A pH- and temperature –phase two-stage process for hydrogen and methane production from food waste. Int. J. Hydrogen Energy 33. 4739-2746.

Duran, M., Speece, R.E. 1997. Temperature-stage anaerobic process. Environ Technol 18, 747-754.

Fang, H.H.P., Li, C., Zhang, T., 2006. Acidophilic biohydrogen production from rice slurry, Int J. Hydrogen energy 31, 683-692.

Gavala, H.N., Skiadas, I.V., Ahring, B.K,. 2006. Biological hydrogen production in suspended and attached growth anaerobic reactor systems. Int J Hydrogen Energy 31, 1164-1175.

Han, S.-K., Shin, H.-S., 2004. Biohydrogen production by anaerobic fermentation of food waste. Int J. Hydrogen Energy 29, 569-577.

Hart, D., 1997. Hydrogen power: the commercial future of the “ultimate fuel”. Financial times energy publishing, Lodon.

Hawkes, F.R., Hussy, I., Kyazze, G., Dinsdale, R., Hawkes, D.L., 2007. Continuous dark fermentative hydrogen production by mesophilic microflora: principles and progress. Int J Hydrogen Energy 32, 172–184.

Jo, J.H., Jeon, C.O., Lee, D.S., Park, J.M., 2007. Process stability and microbial community structure in anaerobic hydrogen producing microflora from food waste containing kimchi. J Biotechnol 131, 300-308.

Keith, B., Tom, S., 1995. The application of membrane biological reactors for the treatment of wastewaters. Biotech Bioengin 49, 601-610.

Kim, D.-H., Han, S.-K., Kim, S.-H., Shin, H.-S., 2006. Effect of gas sparging on continuous fermentative hydrogen production. Int J. Hydrogen Energy 31, 2158-2169.

Kim, M.-S., Lee, D.-Y., 2010. Fermentative hydrogen production from tofu-processing waste and anaerobic digester sludge using microbial consortium. Biores Technol 101, S48-S52.

Kim, S.-H., Han, S.-K., Shin, H.-S., 2004. Feasibillity of biohydrogen production by anaerobic co-digestion of food waste and sewage sludge. Int J Hydrogen Energy 29, 1607-1616.

Kim, S.-H., Han, S.-K., Shin, H.-S., 2008. Optimization of continuous hydrogen fermentation of food waste as a function of solids retention time independent of hydraulic retention time. Process Biochemistry 43, 213-218.

Kim, S.-H., Hang, S.-K., Shin, H.-S., 2006. Effect of substrate concentration on hydrogen production and 16S rDNA-based analysis of the microbial community in a continuous fermenter. Process Biochem 41,199-207.

Lee, D.-Y., Ebie, Y., Xu, K.-Q., Li, Y.-Y., Inamori, Y., 2010a. Continuous H2 and CH4 production from high-solid food waste in the two-stage thermophilic fermentation process with the recirculation of digester sludge. Biores Technol 101, S42-S47.

Lee, D.-Y., Li, Y.-Y., Noike, T., 2009. Continuous H2 production by anaerobic mixed microflora in membrane bioreactor. Biores Technol 100 (2): 690-695.

Lee, D.-Y., Li, Y.-Y., Noike, T., 2010b. Influence of solids retention time on continuous H2 production using membrane bioreactor. Int J Hydrogen Energy 35, 52-60.

Lee, D.-Y., Li, Y.-Y., Noike, T., Cha, G.-C., 2008. Behavior of extracellular polymers and bio-fouling during hydrogen fermentation with a membrane bioreactor. J. Membr. Sci 32, 13-18.

Lee, K.S., Lo, Y.C., Lin, P.J., Chang, J.S., 2006. Improving biohydrogen production in a carrier-induced granular sludge bed by altering physical configuration and agitation pattern of the bioreactor. Int J Hydrogen Energy 31 (12), 1648-1657.

Marchesi, J.R., Sato, T., Seightman, A.J., Martin, T.A., Fry, J.C., Hiom, S.J., 1998. Design and evaluation of useful bactriu-specific PCR primers that amlilfy genes coding for bacterial 16s rRNA. Appl. Environ. Microbiol 64: 795-799, Published erratum 1998. Appl. Environ. Microbiol 64, 2333.

Mu. Y., Zheng, X.-J., Yu, H.-Q., Zhu, R.-F., 2006. Biological hydrogen production by anaerobic sludge at various temeratres. Int J Hydrogen Energy 31, 780-785.

Noike, T., Mizuno, O., 2000. Hydrogen fermentation of organic municipal wastes, Wat Sci Technol 42(12), 155-162.

Nomura, M., Bin, Tang., Nakao. SI., 2002. Selective ethanol extraction from fermentation broth using a silicalite membrane. Sep Puri Technol 27(1), 59-66.

Oh, S.-E., Lyer, P., Bruns, M.A., Logan, B.E., 2004. Biological hydrogen production using a membrane bioreactor. Biotechnol Bioeng 87(1), 199-127.

 Rachman, M.A., Nakashimada, Y., Kakizono, T., Nishio, N., 1998. Hydrogen production with high yield and high evolution rate by self-flocculated cells of Enterobacter aerogenes in a packed-bed reactor. App Micro Biotech 49, 450-454.

Ritalahti, K.M., Amos, B.K., Sung, Y., Wu, Q., Koenigsberg, S.S., Löffler, F.E., 2006. Quantitative PCR targeting 16Sr RNA and reductive dehalogenase genes simultaneously monitors multiple Dehalococcoides strains. Appl. Environ. Mcrobiol 72 (4), 2765-2774.

Roy, D., Goulet, J., LeDuy, A., 1986. Batch fermentation of whey ultrafiltrate by Lactobacillus helveticus for lactic acid production. Applied and Microbiology Biotechnology 24, 206-13.

Shin, H.-S., Youn, J.-H., 2005. Conversion of food waste into hydrogen by thermophilic acidogenesis. Biodegradation 16, 33-44.

Ueno, Y., Otsuka, S., Morimoto, M., 1996. Hydrogen production from industrial wastewater by anaerobic microflora in chemostat culture. J Ferment Bioeng 82 (2), 194-197.

Yang, P., Zhang, R., McGarvey, J.A., Benemann, J.R., 2007. Biohydrogen production from cheese processing wastewater by anaerobic fermentation using mixed microbial communities. Int J Hydrogen Energy 32 (18), 4761-4771.

Yu, Y.S., Lee C.S., Kim J.A., Hwang S.H., 2005. Group-specific primer and probe sets to detect methanogenic communities using quantitative real-time polymerase chain reaction. Biotechnol Bioeng 89. 670 - 679.

Zhu, H., Parker, W., Basnar, R., Proracki, A., Falletta, P., BéLand, M., Seto, P., 2008. Biohydrogen production by anaerobic co-digestion of municipal food waste and sewage sludge. Int J Hydrogen Energy 33, 3651-3659.

Zinder, S.H., 1986. Thermophilic waste treatment systems. In: Brock TD, editor. Thermopiles: general, Molecular and applied biology. New York, Wiley-Interscience, 257. 





















Figure legends


Figure 1 Schematic diagram of experimental apparatus for submerged HF-MBR

Figure 2 Profiles of (a) biogas production (a) and (b) biogas composition (b) in HF-MBR

Figure 3 The cChange of CODcr, carbohydrate and protein concentrations in the permeate



























Fig. 1



































Fig. 2










Fig. 3

















Table legends


Table 1 Characteristics of food waste slurry

Table 2 The oOperational conditions of the HF-MBR

Table 3 H2 production yields and metabolites at under different operational conditions

Table 4 The cClosest relatives of 16S rDNA sequences retrieved from the thermophilic HF-MBR

Table 5 Summary of comparable H2 yields from food wastes in different continuous systems

Table 6 The qQuantitation [JH94] of acidogenic bacteria and methanogenic archaea
























Table 1





Total solids


4.5 ± 0.3

Volatile solids


4.3 ± 0.2

Total COD


52.7 ± 2.8

Total carbohydrate


17.7 ± 0.4

Total protein


11.4 ± 0.5

Total nitrogen


2.22 ± 0.25

Total phosphorus


176 ± 16



14 ± 4.8



28 ± 7.4



37 ± 8.3



4.3 ± 0.1















Table 2



R 1

R 2

R 3



Plate & flame



Pore size



Total volume (l)


Temperature (oC)

55 ± 0.5


5.5 ±0.1



OLR (kg-COD/m3/d)




HRT (h)




Biogas bubbling rate (l/min)












Table 3



OLR (kg-COD/m3/d)

H2 production rate

H2 yield


Soluble CODcr










mg/l as CODcr

R 1











R 2











R 3












aHAc = Acetic acid; bHPr = Propionic acid; cn-HBu = n-Butyric acid; dn-HCar = n-Carporic acid; eHLa = Lactic acid; fEtOH = Ethanol










Table 4




Affiliation Relation

Accession no.

Similarity (%)

No. of clones

Abundance (%)



Clostridium sp. Z6






Thermoanaerobacterales Family III. Incertae Sedis

Thermoanaerobacterium thermosaccharolyticum







Lactobacillus helveticus



















Table 5


Reactor type

Operational conditions

H2 production rate

H2 yield


Temp. (oC)

HRT (h)

SRT (d)

OLR (kg/m3/d)








125.4 (as COD)




This study





39 (as COD)




Lee et al.,2010





22.7 (as VS)




Wang et al., 2009





8.0 (as VS)




Shin et al., 2005





32.1 (as COD)




Kim et al., 2008





64.4 (as COD)




Chu et al., 2008

N.G = not given

N.A = not applicable


a SCRD = semi-continuous rotating drum

b ASBR = anaerobic sequencing batch reactor


Table 6



No. of rRNA determined by real-time PCR

a copies/mg-DNA

b copies/ml-sample



6.08 ´109


c 1.03´101

c 1.92 ´103



a Average No. of rRNA based on the amount of rDNA

b Average No. of rRNA based on the amount of sample

c Below detection limit



 [JH1]The singular here is better, despite the common use of the plural in the text..

 [JH2]Implicit here


 [JH4]OR (same meaning): “was predominantly”


 [JH6]Delete this if it does not fit.

 [JH7](?) OR: reformation

 [JH8]OR: “an alternative recovery method”


 [JH10]Not yet id’d post-Abstract

 [JH11]Probably a reference is needed for this.


 [JH13]implicit (already established)

 [JH14]… already identified post-Abstract


--passim for all like instances


“under the high-SRT condition”  /  OR(2): “under high SRTs”

 [JH16](this sentence was moved up in the paragraph for better organization)

 [JH17]Implicit in “key”



“in continuous operation”


 [JH21]I think the present tense is more appropriate here.


 [JH23]… **according to Figure 1



 [JH26]… you can use “at” because of the “+/-”

 [JH27]*OR (alternative meaning): “using the QProbe-PCR … and an extrap … kit”




 [JH31]*Probably you mean “2.3.” (Also, technically, 2.3 [or 2.2] is a subsection, but “Section” probably is okay here as a generic term.)


Here and passim, use “program” if that is the terminology typically employed in this context within the literature of your field.



 [JH35]OR: “using an auto … 8000 and….”


 [JH37]OR (alternative, more specific meaning): “limited to”

 [JH38]*OR (alternative meaning): “There was some fluctuation in biogas production, owing to a small change in the permeate flux”


“margin … was”


 [JH41]OR (here and passim): “the ~ condition”

 [JH42]… because you wrote “previous studies” (plural)


—if there was only this one study, delete “e.g.” and change “previous studies” to

“a previous study”




 [JH46]OR: “the flux increased … /day, maintaining….”


 [JH48](?) OR: “A relatively stable flux was”

 [JH49]OR: “continuous operation”

 [JH50]*Is this the right number (the right table)?


 [JH52]Implicit here






“a long SRT”




**Remember that if you are talking about a result specific to your experimentation (i.e. if you are NOT talking about a general truth), you need the past tense (was) here, and passim for all like situations.

 [JH61]Delete this if it does not fit the meaning.







“low H2 yields”


 [JH69]redundant give n the following “basis”

 [JH70]*OR (alternative meaning):

“almost the same”


“ratio was … value”



 [JH74]OR (alternative meaning):

“as low as 1.3”

 [JH75](?) In the previous sentence, you said that the B/A ratio was within a range.


OR: “the B/A ratio within the optimal range”


 [JH77]OR (if 2010a is not the only pertinent study—i.e. there were others too): “one of our previous studies”












 [JH89]Here and passim, make this “added” consistent with the other instances.


{Undo this change.}

 [JH91]Implicit (already established in this immediate context)


 [JH93]Neither included in page count nor checked