- Email: [email protected]
Effect of the accuracy of pH control on hydrogen fermentation
Accepted Manuscript Effect of the Accuracy of pH Control on Hydrogen Fermentation Chungman Moon, Sujin Jang, Yeo-Myeong Yun, Mo-Kwon Lee, Dong-Hoon Kim, Mi-Sun Kim PII: DOI: Reference:
S0960-8524(14)01560-0 http://dx.doi.org/10.1016/j.biortech.2014.10.128 BITE 14171
To appear in:
Bioresource Technology
Received Date: Revised Date: Accepted Date:
25 August 2014 23 October 2014 26 October 2014
Please cite this article as: Moon, C., Jang, S., Yun, Y-M., Lee, M-K., Kim, D-H., Kim, M-S., Effect of the Accuracy of pH Control on Hydrogen Fermentation, Bioresource Technology (2014), doi: http://dx.doi.org/10.1016/j.biortech. 2014.10.128
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Effect of the Accuracy of pH Control on Hydrogen Fermentation
Chungman Moona, Sujin Janga,b, Yeo-Myeong Yunc, Mo-Kwon Leea, Dong-Hoon Kima, Mi-Sun Kima,b*
a
Biomass and Waste Energy Laboratory, Korea Institute of Energy Research, 152 Gajeong-ro,
Yuseong-gu, Daejeon 305-343, Republic of Korea b
Division of Renewable Energy Engineering, University of Science and Technology, 217
Gajeong-ro, Yuseong-gu, Daejeon 305-350, Republic of Korea c
Department of Civil and Environmental Engineering, KAIST, 373-1 Guseong-dong, Yuseong-
gu, Daejeon 305-701, Republic of Korea *
Corresponding author: Mi-Sun Kim (E-mail address: [email protected], Tel: +82-42-860-
3554; fax: +82-42-860-3739)
ABSTRACT pH, known as the most important parameter in H2 fermentation, cannot be precisely controlled in a scaled-up fermenter as in a lab fermenter. In the preset work, to assess the effect of pH control accuracy on H2 fermentation, the pH was controlled at 6.0±0.1, 6.0±0.3, 6.0±0.5, 6.0±0.7, and 6.0±0.9 during batch fermentation of food waste. Up to deviation of ±0.3, a high H2 yield of 1.67-1.73 mol H2/mol hexoseadded was attained with producing butyrate as a major metabolite (>70% of total organic acids produced). A huge drop of H2 production, however, was observed at deviation > ±0.5 with lowered substrate utilization and increased production of lactate. Next 1
generation sequencing results showed that Clostridium was found to be the dominant genus (76.4% of total number of sequences) at deviation of ±0.1, whereas the dominant genus was changed to lactic acid bacteria such as Streptococcus and Lactobacillus with increase of deviation value.
Keywords: Hydrogen fermentation; pH deviation; lactate; butyrate; next generation sequencing; food waste
1. INTRODUCTION Hydrogen (H2) is recognized as a promising energy carrier, since it produces only water when combusted and has the highest energy content per unit weight of any known fuel (142 kJ/g) (Das and Veziroglu, 2008). The current methods for producing H2, however, involve the use of fossil fuels, cracking natural gas, heavy oil, naphtha, and coal. In contrast, biological routes, whereby microorganisms convert solar energy or electrons contained in organics and inorganics to H2, are environmentally benign (Kim et al., 2011). Among various biological approaches, dark H2 fermentation (so called H2 fermentation) is considered the most practically applicable method owing to its high H2 production rate, simplicity in operation, and the utilization of organic waste as a feedstock (Levin et al., 2004; Hallenbeck et al., 2012). In H2 fermentation, operation parameters such as pH, temperature, substrate concentration, and hydraulic retention time (HRT) are known to have substantial effects on the performance. In particular, extensive research dealing with pH effects on H2 production has been carried out, as the pH directly affects the hydrogenase activity, metabolic pathway, and population dynamics (Dabrock et al., 1992; Khanal et al., 2004; Lay, 2000). H2 production is always accompanied by 2
acids generation, causing a pH drop in the fermenter. Thus, to maintain high H2 productivity and achieve high H2 yield, pH should be controlled by adding an alkaline solution such as KOH or NaOH, generally in a pH range of 5.0-6.5 (Khanal et al., 2004; Kawagoshi et al., 2005; Hawkes et al., 2007). In a lab-scale fermenter where homogeneity is guaranteed with sufficient agitation, pH can be precisely controlled, deviating only ±0.1 from the optimum set value. However, in the case of scale-up and practical implementation, it is difficult to ensure uniform pH throughout the fermenter (Amanullah et al., 2001). Dead zone might exist with insufficient agitation. Furthermore, when the pH sensor responds to a pH drop to add alkaline solution, the pH near the alkaline solution injection port would increase above the optimum value, whereas the pH far from the injection port would remain below the optimum value. This phenomenon might impede the accuracy of pH control, consequently resulting in a drop of H2 production. The aim of this study is to investigate the effect of pH control accuracy on H2 fermentation, by varying the deviation from ±0.1 to ±0.9 at a set pH value of 6.0. As an H2 fermentation method, an innovative batch process, producing H2 from food waste (FW) without inoculum addition, was used. When proper pretreatment such as acid-shock was applied to FW, it naturally decomposed along with H2 production (Kim et al., 2009). During the fermentation, the production of gas and organic acids was periodically monitored. In addition, the microbial community was analyzed using a next generation sequencing (NGS) tool, and the results were correlated with the fermentation performance.
2. METERIALS and METHODS
3
2.1. Feedstock In this study, FW collected from the cafeteria in Korea Institute of Energy Research in Daejeon, Korea was used as a feedstock for H2 production. Before fermentation, the feedstock was shredded by a grinder smaller than 5 mm in diameter and kept in a refrigerator at 1°C to block natural decomposition. The concentrations of total solids (TS), volatile solids (VS), chemical oxygen demand (COD), carbohydrate, and ammonia of FW were 126.9 g TS/L, 122.8 g VS/L, 149.0 g COD/L, 83.7 g Carbo. COD/L (g/L as carbohydrate COD), and 113.5 mg NH4-N/L, respectively.
2.2 Hydrogen fermentation method Acid-pretreatment of FW was conducted at pH 2 by adding a 6 N HCl solution, and was continued for 12 h with agitation at 100 rpm. Then, the acid-pretreated FW corresponding to 60 g Carbo. COD/L was added to the batch fermenter with an effective volume of 300 mL, while the remainder of the effective volume was filled with tap water. Neither an external inoculum nor a basal medium was added. Before fermentation, the pH was readjusted to 8.0±0.1, which was identified as the optimal initial pH value in our previous work, by 6 N KOH addition (Kim et al., 2011). Each batch fermenter was equipped with a pH sensor. After purging with N2 gas for 10 min to provide anaerobic conditions, the fermentation began. The pH was allowed to drop from 8.0 to 6.0 via fermentation, after which point pH of 6.0±0.1, 6.0±0.3, 6.0±0.5, 6.0±0.7, and 6.0±0.9 was automatically controlled by the addition of a 3 N KOH solution using pH controller (Cns, Model Marado-PDA). For example, at pH 6.0±0.3, the alkaline solution pump started to act at pH 5.7 and stopped when pH reached 6.3. The optimal operational pH in H2 fermentation was generally reported in the range of 5.0-6.5, in which the activity of hydrogenase intense and 4
the metabolic pathway for H2 production is thermodynamically favorable (Dabrock et al., 1992; Khanal et al., 2004). The fermenter was agitated at 100 rpm by a mechanical mixer and placed in a temperature controlled room at 35±1oC. The gas produced and its constituents were monitored at 1-5 h intervals. All tests were carried out in triplicate and the results were averaged.
2.3 Analytical methods Measured biogas production was adjusted to the standard conditions of temperature (0°C) and pressure (760 mmHg) (STP). The H2 content in the biogas was determined by a gas chromatography (GC, Gow Mac series 580) using a thermal conductivity detector and a 1.8 m × 3.2 mm stainless-steel column packed with molecular sieve 5A with N2 as a carrier gas. The content of CH4 was measured using a GC of the same model as noted previously with a 1.8 m × 3.2 mm stainless-steel column packed with Porapak Q (80/100 mesh) using N2 as a carrier gas. The temperatures of the injector, detector, and column were kept at 80, 90, and 50°C, respectively, in both GCs (Kim et al., 2011). Concentrations of TS, VS, COD, and ammonia were measured according to Standard Methods (Clescerl et al., 1998). The concentration of carbohydrate was determined by the colorimetric method (Dubois et al., 1956). Organic acids were analyzed by a high performance liquid chromatograph (HPLC) (Finnigan Spectra SYSTEM LC, Thermo Electron Co.) with an ultraviolet (210 nm) detector (UV1000, Thermo Electron) and an 100 mm × 7.8 mm Fast Acid Analysis column (Bio-Rad Lab.) using 0.005 M H2SO4 as a mobile phase. The liquid samples were pretreated with a 0.22 µm membrane filter before injection to the HPLC.
5
2.4 Microbial community analysis 2.4.1 Sampling, DNA extraction, and PCR Deoxyribonucleic acids (DNAs) in a fermented broth were extracted using an Ultraclean Soil DNA Kit (Cat #12800-50; Mo Bio Laboratory Inc., USA). The extracted DNA was subsequently purified with an UltraClean Microbial DNA Isolation Kit (Mo Bio Laboratories, CA, USA). A library was then prepared using polymerase chain reaction (PCR) products according to the GS FLX titanium library prep guide, and the libraries were quantified using Picogreen assay (Victor 3). The emPCR, corresponding to clonal amplification of the purified library, was carried out using a GS-FLX titanium emPCR Kit (454 Life Sciences). A 20 ng aliquot of each sample DNS was used for a 50 ul PCR reaction. The 16S universal primers 27F (5’ GAGTTTGATCMTGGCTCAG 3’) and 800R (5’ TACCAGGGTATCTAATCC 3’) were used to amplify the 16s rRNA genes (Øvreås et al., 1997). A Fast Start High Fidelity PCR System (Roche) was used for PCR under the following conditions: 94°C for 3 min followed by 35 cycles of 94°C for 15 sec; 55°C for 45 sec and 72°C for 1 min; and a final elongation step at 72°C for 8 min.
2.4.2 High throughput pyrosequencing and sequence analysis The PCR products were purified by using AMPure beads (Beckman coulter). After the products were purified and quantified, sequencing was performed using a 454 pyrosequencing Genome Sequencer FLX Titanium (Life Sciences, CT, USA), according to the manufacturer’s instructions, by a commercial sequencing facility (Macrogen, Seoul, Korea).The sequences generated from pyrosequencing were mainly analyzed with the software MOTHUR for preprocessing (quality-adjustment, barcode split), identification of operational taxonomic units 6
(OTUs), taxonomic assignment, community comparison, and statistical analysis (Schloss et al., 2009). Sequences were filtered to minimize the effects of poor sequence quality, and sequencing errors were minimized by removing sequences with more than one ambiguous base call and by retaining only sequences that were 300 nt or longer (Li et al., 2012). Sample-specific sequences were collected according to the barcode sequences tagged to each sample. The barcode, as well as forward and reverse primer sequences, was trimmed from the initial sequences. As a pyrosequence pre-processing step, sequences that were shorter than 300 nt, had one or more ambiguous base calls, or had multiple barcode or primer motifs were excluded from the analysis. The trimmed sequences from each barcode bin were aligned using Infernal, and associated covariance models were obtained from the Ribosomal Database Project Group (Cole et al., 2009). The sequences spanning the same region were then realigned with the NCBI BLAST database (www.ncbi.nlm.nih.gov). In database screening with the BLAST program, the threshold E-value for including a sequence in the next iteration was 0.001. A distance matrix was calculated from the aligned sequences, and operational taxonomical units (OTUs; 90–100% sequence similarity) were assigned using the furthest-neighbor clustering algorithm. The OTUs defined by a 3% distance level were phylogenetically classified with a modified bacterial RDP II database containing 164,517 almost full-length 16S rRNA gene sequences prepared using TaxCollector (http://www.microgator.org).
3. RESULTS AND DISCUSSION 3.1. pH change and H2 fermentation Fig. 1 shows the time courses of cumulative H2 production and pH change profile at various deviation ranges of ±0.1-±0.9. The cumulative H2 production curves were well fitted by the 7
modified Gompertz equation with high R2 values (>0.99) (Chen et al., 2006). The fermentation ended within 30 h, and CH4 was not detected during the entire experimental period. As fermentation began, pH gradually dropped and reached 6.0 after 5.5 h, and H2 started to evolve after 4 h, in all fermenters. At deviation of ±0.1, addition of an alkaline solution was commenced after 5.5 h, and the solution was continuously added every 0.3-0.7 h until 20 h of fermentation at which time H2 was vigorously produced. During the whole fermentation, the total number of times adding alkaline solution was 22 at deviation of ±0.1, while it drastically decreased to 7, 5, 3, and 1 at deviation of ±0.3, ±0.5, ±0.7, and ±0.9, respectively. It appeared that as the deviation increased, the fermentation time when pH was controlled below optimum pH of 6.0 increased. For example, pH was below 6.0 for 14.9 h at deviation of ±0.3, while it was below 6.0 for 19.8 h at deviation of ±0.7. After 5 h of fermentation, 27-29 mL of 3 N KOH was added to keep the pH 6.0±0.1 and ±0.3. At deviation of ±0.5-±0.9, there was a sharp reduction of alkaline solution added, ranging 6-21 mL. Although a 1.5 times higher H2 production rate was achieved at deviation of ±0.1 compared to at ±0.3, there was little apparent difference in the overall performance between the two cases (Table 1). A similar amount of H2 production (3.51-3.64 L) was observed at deviation ranges of ±0.1 and ±0.3, corresponding to a H2 yield of 1.67-1.73 mol H2/mol hexoseadded. On VS and COD basis, the H2 yield ranged 137.0-142.1 mL H2/g VSadded and 109.5-113.6 mL H2/g CODadded, respectively, which were in the range of previous works using organic solid waste as a feedstock (De Gioannis et al., 2013). This indicates that the acid-pretreatment applied in this study was effective for the successful H2 fermentation of FW. A huge drop of H2 production, however, was observed at ±0.5, ±0.7, and ±0.9 with lowered carbohydrate degradation and VS reduction. H2 yield gradually dropped as the deviation 8
increased, with the lowest yield of 0.69 mol H2/mol hexoseadded at ±0.9. Severe inhibition on substrate utilization was observed at deviation ±0.9. Carbohydrate degradation and VS reduction dropped to 59% and 26%, respectively, at ±0.9. This drop of substrate utilization may be one of the reasons for the low production of H2, but the main reason was found by monitoring the production of organic acids.
3.2. Organic acids production The production profiles of organic acids at various pH deviation values are shown in Fig. 2. The main organic acids were butyrate, acetate, and lactate, and their composition was varied depending on the deviation value. In addition, the total production of organic acids decreased from 48.1 to 28.9 g COD/L as the deviation range increased, likely due to a gradual drop of substrate utilization. Butyrate was found to be the dominant metabolite at deviation of ±0.1 and ±0.3, where high H2 yields were achieved, occupying 77.2% and 70.7% of total organic acids produced, respectively. However, at deviation > ±0.5, the production of lactate significantly increased while butyrate production was suppressed. At ±0.7, the highest lactate concentration of 18.5 g COD/L was observed, occupying 54.8% of total organic acids produced. This trend of different composition can be easily linked to the different H2 production performance. According to the following reactions (Eq. (1) and (2)), the production of butyrate has often been related with high H2 yield, while lactate is considered as a main metabolite that should be avoided in H2 fermentation (Angenent et al., 2004). In particular, due to the existence of a large amount of indigenous lactic acid bacteria (LAB), the minimization of lactate production has been raised as a key issue for the successful H2 production from food-processing waste (Noike et al., 2002; Kim et al., 2009). 9
Glucose → Butyrate + 2CO2 + 2H2 + 3ATP
(1)
Glucose → 2Lactate + 2ATP
(2)
From the above results, it was found that high pH deviation caused a drop of H2 production, by decreasing the substrate utilization and shifting the main metabolite from butyrate to lactate. The well-known main H2-producer, Clostridium sp., is also capable of producing lactate, but only under specific conditions such as high H2 partial pressure and nutrient limitation (phosphate and iron) (Bahl and Dürre, 2001). However, even in those conditions, the lactate composition was generally below 20%, which suggests that there has been a change of the microbial community by varying the deviation.
3.3. Microbial community analysis To assess the microbial community structure in response to pH deviation, collected mixed-liquor samples at the end of fermentation were used to pyro-sequence the former region of 16S rRNA gene using the 454 GS-FLX sequencer. A total of 52,531 reads were obtained from a single lane of an 8-lane pico-titer plate on a Genome Sequencer FLX titanium system. The decrease in the Shannon index indicates that species richness and evenness decreased with an increase of deviation (Table 2). The sequences of bacteria were assigned to the genus and species levels to evaluate the relative microbial abundance in each sample. It was observed that the dominant population remarkably changed from H2-producers to LAB as the deviation increased (Fig. 3). In terms of genus level, Clostridium, capable of producing H2, was found in the sample operating at deviation of ±0.1, 10
occupying 76.4% of the total number of sequences (Yun et al., 2014). Meanwhile, Lactococcus, Enterococcus, and Lactobacillus, which all are LAB, accounted for only 7.4%, 3.8%, and 0.4% of the total number of sequences, respectively (Kim et al., 2014). As the pH deviation increased to ±0.7, it was observed that Clostridium drastically dropped to 14.8%. Instead, Streptococcus newly emerged, and its abundance increased to 73.3% of the total number of sequences. This could be related to the characteristic defense mechanism of Streptococcus that allows it to endure pH fluctuations. According to Hamilton and Buckley (1991), Streptococcus cells can compensate for fluctuating external pH via control of intracellular pH to change the permeability of protons into the cell. However, when external pH was further decreased to 5.0, metabolic damage occurred in cells by breaking certain enzymes, and this might be linked to the observation of a noticeable decrease in Streptococcus (0.1%) in the sample with ±0.9 (Harvey, 1965). Meanwhile, the relative abundance of Lactobacillus which can survive at low pH, jumped to 80.9% of the total number of sequences (Kim et al., 2014). However, it was thought that despite the gradual increase in occupation of Lactobacillus in the sample at deviation of ±0.9, a decrease in richness of bacteria at ±0.9 (the Shannon index) limited substrate utilization and lactate production as described above (Fig. 2(e) and Table 2) (Torsvik and Øvreås, 2002). In order to observe the diversity of bacteria in greater detail, the sequences to species level was further assigned by choosing 10 representative microorganisms (sequences > 3%) as shown in Table 3. In total, two representative species belonged to Clostridium sp., Lactobacillus sp., and Streptococcus sp., respectively, and one representative species belonged to Citrobacter sp., Leuconostoc sp., Enterococcus sp., and Lactococcuss sp. The most dominant species in the sample ±0.1 was Clostridium butyricum, which often has been observed in H2 fermentation of organic wastes using anaerobic mixed cultures (Yun et al., 2014). However, an increase of pH 11
deviation resulted in a decreased abundance of Clostridium sp. while Streptococcus lutetiensis and Lactobacillus delbrueckii subsp. bulgaricus ATCC 11842 became dominant. Streptococcus lutetiensis, previously known as Streptococcus infantarius subsp. coli, can be found in feces of human infants (Almuzara et al., 2013; Poyart et al., 2002). Lactobacillus delbrueckii subsp. bulgaricus ATCC 11842 is known to survive in a wide pH range (Adimpong et al., 2013). From this study, it can be concluded that although H2-producers were initially dominated by acid-shock, the precise control of pH should be provided for the successful H2 fermentation. When the deviation was ≥ 0.5, the activity of LAB was revived, which significantly decreased the H2 yield. It is suggested that pH controlling devices such as probe, meter, and alkaline solution pump should be equipped and operated to minimize the deviation less than ±0.3.
4. CONCLUSIONS During the batch H2 fermentation, high H2 yields ranging 1.67-1.73 mol H2/mol hexoseadded were achieved when pH was controlled at 6.0±0.1 and 6.0±0.3. A significant drop of H2 production (0.69-0.98 mol H2/mol hexoseadded), however, was observed when deviation was ≥ 0.5, accompanied with a lowered substrate utilization and the shift of main metabolite from butyrate to lactate. The change of dominant population from Clostridium to LAB with the increase of deviation value was seen by NGS analysis. These findings indicate the importance of pH control accuracy for the successful H2 fermentation, even though the activity of non-H2-producers were initially suppressed.
ACKNOWLEDGEMENT This work was supported by Korea District Heating Corporation, and the Energy Efficiency & 12
Resources Core Technology Program of the Korea Institute of Energy Technology Evaluation and Planning (KETEP), granted financial resource from the Ministry of Trade, Industry & Energy, Republic of Korea (20132020000170)
REFERENCES 1.
Adimpong, D.B., Nielsen, D.S., Sørensen, K.I., Vogensen, F.K., Sawadogo-Lingani, H., Derkx, P.M., Jespersen, L., 2013. Lactobacillus delbrueckii subsp. jakobsenii subsp. nov., isolated from dolo wort, an alcoholic fermented beverage in Burkina Faso. Int. J. Syst. Evol. Microbiol. 63, 3720-3726.
2.
Almuzara, M., Bonofiglio, L., Cittadini, R., Ocampo, C.V., Montilla, A., Del Castillo, M., Ramirez, M.S., Mollerach, M., Vay, C., 2013. First Case of Streptococcus lutetiensis Bacteremia Involving a Clindamycin-Resistant Isolate Carrying the lnuB Gene. J. Clin. Microbiol. 51, 4259-4261.
3.
Amanullah A., McFarlane, C.M., Emery, A.N., Nienow, A.W., 2001. Scale-down model to simulate spartial pH variations in large-scale bioreactors. Biotechnol. Bioeng. 73, 390-399.
4.
Angenent, L.T., Karim, K., Al-Dahhan, M.H., Wrenn, B.A., Domíguez-Espinosa, R., 2004. Production of bioenergy and biochemical from industrial and agricultural wastewater. Trends Biotechnol. 22, 477-485.
5.
Bahl, H., Dürre, P., 2001. Clostridia: Biotechnology and medical applications, Wiley-VCH, Weinheim.
6.
Chen, W.H., Chen, S.Y., Khanal, S., Sung, S. 2006. Kinetic study of biological hydrogen production by anaerobic fermentation. Int. J. Hydrogen Energ. 31, 2170-2178.
7.
Clescerl, L.S., Greenberg, A.E., Eaton, A.D., 1998. Standard methods for the examination of 13
water and wastewater, 20th ed. American Public Health Association, Washington, D.C., pp. 2-57-9. 8.
Cole, J.R., Wang, Q., Cardenas, E., Fish, J., Chai, B., Farris, R.J., Kulam-Syed-Mohideen A.S., McGarrell, D.M., Marsh, T., Garrity, G.M., Tiedje, J.M., 2009. The Ribosomal Database Project: improved alignments and new tools for rRNA analysis. Nucleic Acids Res. 37, D141-145.
9.
Dabrock, B., Bahl, H., Gottschalk, G., 1992. Parameters affecting solvent production by Clostridium pasteurianum. Appl. Environ. Microbiol. 58, 1233-1239.
10. Das, D., Veziroglu, T.N., 2008. Advances in biological hydrogen production processes. Int, J. Hydrogen Energ. 33, 6046-6057. 11. De Gioannis, G., Muntoni, A., Polettini, A., Pomi, R., 2013. A review of dark fermentative hydrogen production from biodegradable municipal waste fractions. Waste Manage. 33, 1345-1361. 12. Dubois, M., Gilles, K.A., Hamilton, J.K., Rebers, P.A., Smith, F., 1956. Colormetric method for determination of sugars and related substances. Anal. Chem. 28, 350-356. 13. Hallenbeck, P.C., Abo-Hashesh, M., Ghosh, D. 2012. Strategies for improving biological hydrogen production. Bioresource Tchenol. 110, 1-9. 14. Hamilton, I.R., Buckley, N.D., 1991. Adaptation by Streptococcus mutans to acid tolerance. Oral Microbiol. Immunol. 6, 65-71. 15. Harvey, R. J., 1965. Damage to Streptococcus lactis resulting from growth at low pH. J. Bacteriol. 90, 1330-1336.
14
16. 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 Energ. 32, 172-184. 17. Kawagoshi, Y., Hino, N., Fujimoto, A., Nakao, M., Fujita, Y., Sugimura, S., Furukawa, K., 2005. Effect of inoculum conditioning on hydrogen fermentation and pH effect on bacterial community relevant to hydrogen production. J. Biosci. Bioeng. 100, 524-530. 18. Khanal, S.K., Chen, W.H., Li, L., Sung, S., 2004. Biological hydrogen production: effects of pH and intermediate products. Int. J. Hydrogen Energ. 29, 1123-1131. 19. Kim, D.H., Kim, S.H., Jung, K.W., Kim, M.S., Shin, H.S., 2011. Effect of initial pH independent of operational pH on hydrogen fermentation of food waste. Bioresource Technol. 102, 8646-8652. 20. Kim, D.H., Kim, S.H., Shin, H.S., 2009. Hydrogen fermentation of food waste without inoculum addition. Enzyme Microb. Tech. 45, 181-187. 21. Kim, D.H., Lee, M.K., Moon, C., Yun, Y.M., Lee, W., Oh, S.E., Kim, M.S., 2014. Effect of hydraulic retention time on lactic acid production and granulation in an up-flow anaerobic sludge blanket reactor. Bioresource Technol. 165, 158-161. 22. Lay, J.J., 2000. Modeling and optimization of anaerobic digested sludge converting starch to hydrogen. Biotechnol. Bioeng. 68, 269-278. 23. Levin, D.B., Pitt, L., Lov, M., 2004. Biohydrogen production: prospects and limitations to practical application. Int. J. Hydrogen Energ. 29, 173-185. 24. Li, W., Fu, L., Niu, B., Wu, S., Wooley, J., 2012. Ultrafast clustering algorithms for metagenomic sequence analysis. Brief. Bioinform. 13, 656-668. 25. Noike, T., Takabatake, H., Mizuno, O., Ohba, M., 2002. Inhibition of hydrogen fermentation 15
of organic wastes by lactic acid bacteria. Int. J. Hydrogen Energ. 27, 1367-1371. 26. Øvreås, L., Forney, L., Daae, F.L., Torsvik, V., 1997. Distribution of bacterioplankton in meromictic Lake Saelenvannet, as determined by denaturing gradient gel electrophoresis of PCR-amplified gene fragments coding for 16S rRNA. Appl. Environ. Microbiol. 63, 33673373. 27. Poyart, C., Quesne, G., Trieu-Cuot, P., 2002. Taxonomic dissection of the Streptococcus bovis group by analysis of manganese-dependent superoxide dismutase gene (sodA) sequences: reclassification of 'Streptococcus infantarius subsp. coli' as Streptococcus lutetiensis sp. nov. and of Streptococcus bovis biotype 11.2 as Streptococcus pasteurianus sp. nov. Int. J. Syst. Evol. Microbiol. 52, 1247-1255. 28. Schloss, P.D., Westcott, S.L., Ryabin, T., Hall, J.R., Hartmann, M., Hollister, E.B., Lesniewski, R.A., Oakley, B.B., Parks, D.H., Robinson, C.J., Sahl, J.W., Stres, B., Thallinger, G.G., Van Horn, D.J., Weber, C.F., 2009. Introducing mothur: open-source, platform independent, community-supported software for describing and comparing microbial communities. Appl. Environ. Microbiol. 75, 7537-7541. 29. Torsvik, V., Øvreås, L., 2002. Microbial diversity and function in soli: from genes to ecosystems. Curr. Opin. Microbiol. 5, 240-245. 30. Yun, Y.M., Kim, D.H., Oh, Y.K., Shin, H.S., Jung, K.W., 2014. Application of a novel enzymatic pretreatment using crude hydrolytic extracellular enzyme solution to microalgal biomass for dark fermentative hydrogen production. Bioresource Technol. 159, 365-372.
16
Fig.1 Cumulative H2 production and pH change profile during H2 fermentation of food waste at pH (a) 6±0.1, (b) 6±0.3, (c) 6±0.5, (d) 6±0.7, and (e) 6±0.9 Fig. 2 Organic acids production profile during H2 fermentation of food waste at pH (a) 6±0.1, (b) 6±0.3, (c) 6±0.5, (d) 6±0.7, and (e) 6±0.9 Fig. 3 Next generation sequencing results in genus level (The deviation was varied from ±0.1 to ±0.9 at an optimum set pH value of 6.0.)
17
Fig.1 Cumulative H2 production and pH change profile during H2 fermentation of food waste at pH (a) 6±0.1, (b) 6±0.3, (c) 6±0.5, (d) 6±0.7, and (e) 6±0.9
18
Fig. 2 Organic acids production profile during H2 fermentation of food waste at pH (a) 6±0.1, (b) 6±0.3, (c) 6±0.5, (d) 6±0.7, and (e) 6±0.9
19
20
Fig. 3 Next generation sequencing results in genus level (The deviation was varied from ±0.1 to ±0.9 at an optimum set pH value of 6.0)
21
Table 1 Effect of pH control accuracy on average H2 fermentation performance (The deviation was varied from ±0.1 to ±0.9 at an optimum set pH value of 6.0) H2 yield Sample
Carbohydrate
VS
H2 production
mol H2/mol
mL H2
mL H2
removal
reduction
rate
hexoseadded
/g VSadded
/g CODadded
(%)
(%)
(L H2/L/d)
6±0.1
1.73
142.1
113.6
89 ± 2
50 ± 4
26.7
6±0.3
1.67
137.0
109.5
88 ± 1
50 ± 3
18.1
6±0.5
0.98
80.5
64.4
81 ± 1
45 ± 2
15.1
6±0.7
0.79
64.4
51.5
64 ± 1
35 ± 3
11.2
6±0.9
0.69
56.9
45.5
59 ± 1
26 ± 2
15.4
22
Table 2 Analytical results of the microbial gene libraries obtained from next generation sequencing Sample
Read data
Shannon index
pH 6.0±0.1
10,148
1.9065±0.0363
pH 6.0±0.3
10,338
1.7475±0.0293
pH 6.0±0.5
9,775
1.5294±0.0349
pH 6.0±0.7
10,327
1.3171±0.0352
pH 6.0±0.9
11,943
1.0762±0.0316
23
Table 3 Species level identification of the dominant sequences from each sample (>3% of total sequences)
Microorganism
Clostridium butyricum Streptococcus lutetiensis Streptococcus gallolyticus
pH
pH
pH
pH
pH
6.0±0.
6.0±0.
6.0±0.
6.0±0.
6.0±0.
1 (%)
3 (%)
5 (%)
7 (%)
9 (%)
58.3
26.3
8.7
11.0
6.8
0.0
46.7
70.6
65.3
0.0
0.0
0.0
8.2
7.0
0.0
0.0
0.0
0.0
0.0
78.7
0.3
0.6
2.4
1.3
0.8
6.8
0.1
0.0
0.0
2.5
3.7
2.5
2.1
4.9
3.3
0.0
0.4
0.9
1.9
0.6
0.3
0.2
0.1
0.1
1.1
2.3
0.0
0.0
0.0
0.0
Accessio n (#) NR_0421 44.1 NR_0370 96.1 NR_0748 49.1
Simil arity (%) 99.0
100.0 100.0
Lactobacillus delbrueckii subsp. bulgaricus ATCC
NR_0750 19.1
99.0
11842 Clostridium baratii Lactococcus lactis subsp. lactis Il1403 Enterococcus hirae ATCC 9790 Lactobacillus reuteri DSM 20016 Leuconostoc lactis KCTC 3528 Citrobacter youngae
24
NR_0292 29.1 NR_1039 18.1 NR_0750 22.1 NR_0750 36.1 NR_0408 23.1 NR_0415 27.1
98.0 99.0
99.0 99.0
99.0 98.0
25
Highlights . A significant effect of pH control accuracy on H2 fermentation performance High H2 yield and substrate utilization at deviation of ±0.1 and ±0.3 Low H2 production with lowered substrate utilization at deviation > ±0.5 Lactic acid bacteria became dominant with increase of deviation value.
26
S0960-8524(14)01560-0 http://dx.doi.org/10.1016/j.biortech.2014.10.128 BITE 14171
To appear in:
Bioresource Technology
Received Date: Revised Date: Accepted Date:
25 August 2014 23 October 2014 26 October 2014
Please cite this article as: Moon, C., Jang, S., Yun, Y-M., Lee, M-K., Kim, D-H., Kim, M-S., Effect of the Accuracy of pH Control on Hydrogen Fermentation, Bioresource Technology (2014), doi: http://dx.doi.org/10.1016/j.biortech. 2014.10.128
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Effect of the Accuracy of pH Control on Hydrogen Fermentation
Chungman Moona, Sujin Janga,b, Yeo-Myeong Yunc, Mo-Kwon Leea, Dong-Hoon Kima, Mi-Sun Kima,b*
a
Biomass and Waste Energy Laboratory, Korea Institute of Energy Research, 152 Gajeong-ro,
Yuseong-gu, Daejeon 305-343, Republic of Korea b
Division of Renewable Energy Engineering, University of Science and Technology, 217
Gajeong-ro, Yuseong-gu, Daejeon 305-350, Republic of Korea c
Department of Civil and Environmental Engineering, KAIST, 373-1 Guseong-dong, Yuseong-
gu, Daejeon 305-701, Republic of Korea *
Corresponding author: Mi-Sun Kim (E-mail address: [email protected], Tel: +82-42-860-
3554; fax: +82-42-860-3739)
ABSTRACT pH, known as the most important parameter in H2 fermentation, cannot be precisely controlled in a scaled-up fermenter as in a lab fermenter. In the preset work, to assess the effect of pH control accuracy on H2 fermentation, the pH was controlled at 6.0±0.1, 6.0±0.3, 6.0±0.5, 6.0±0.7, and 6.0±0.9 during batch fermentation of food waste. Up to deviation of ±0.3, a high H2 yield of 1.67-1.73 mol H2/mol hexoseadded was attained with producing butyrate as a major metabolite (>70% of total organic acids produced). A huge drop of H2 production, however, was observed at deviation > ±0.5 with lowered substrate utilization and increased production of lactate. Next 1
generation sequencing results showed that Clostridium was found to be the dominant genus (76.4% of total number of sequences) at deviation of ±0.1, whereas the dominant genus was changed to lactic acid bacteria such as Streptococcus and Lactobacillus with increase of deviation value.
Keywords: Hydrogen fermentation; pH deviation; lactate; butyrate; next generation sequencing; food waste
1. INTRODUCTION Hydrogen (H2) is recognized as a promising energy carrier, since it produces only water when combusted and has the highest energy content per unit weight of any known fuel (142 kJ/g) (Das and Veziroglu, 2008). The current methods for producing H2, however, involve the use of fossil fuels, cracking natural gas, heavy oil, naphtha, and coal. In contrast, biological routes, whereby microorganisms convert solar energy or electrons contained in organics and inorganics to H2, are environmentally benign (Kim et al., 2011). Among various biological approaches, dark H2 fermentation (so called H2 fermentation) is considered the most practically applicable method owing to its high H2 production rate, simplicity in operation, and the utilization of organic waste as a feedstock (Levin et al., 2004; Hallenbeck et al., 2012). In H2 fermentation, operation parameters such as pH, temperature, substrate concentration, and hydraulic retention time (HRT) are known to have substantial effects on the performance. In particular, extensive research dealing with pH effects on H2 production has been carried out, as the pH directly affects the hydrogenase activity, metabolic pathway, and population dynamics (Dabrock et al., 1992; Khanal et al., 2004; Lay, 2000). H2 production is always accompanied by 2
acids generation, causing a pH drop in the fermenter. Thus, to maintain high H2 productivity and achieve high H2 yield, pH should be controlled by adding an alkaline solution such as KOH or NaOH, generally in a pH range of 5.0-6.5 (Khanal et al., 2004; Kawagoshi et al., 2005; Hawkes et al., 2007). In a lab-scale fermenter where homogeneity is guaranteed with sufficient agitation, pH can be precisely controlled, deviating only ±0.1 from the optimum set value. However, in the case of scale-up and practical implementation, it is difficult to ensure uniform pH throughout the fermenter (Amanullah et al., 2001). Dead zone might exist with insufficient agitation. Furthermore, when the pH sensor responds to a pH drop to add alkaline solution, the pH near the alkaline solution injection port would increase above the optimum value, whereas the pH far from the injection port would remain below the optimum value. This phenomenon might impede the accuracy of pH control, consequently resulting in a drop of H2 production. The aim of this study is to investigate the effect of pH control accuracy on H2 fermentation, by varying the deviation from ±0.1 to ±0.9 at a set pH value of 6.0. As an H2 fermentation method, an innovative batch process, producing H2 from food waste (FW) without inoculum addition, was used. When proper pretreatment such as acid-shock was applied to FW, it naturally decomposed along with H2 production (Kim et al., 2009). During the fermentation, the production of gas and organic acids was periodically monitored. In addition, the microbial community was analyzed using a next generation sequencing (NGS) tool, and the results were correlated with the fermentation performance.
2. METERIALS and METHODS
3
2.1. Feedstock In this study, FW collected from the cafeteria in Korea Institute of Energy Research in Daejeon, Korea was used as a feedstock for H2 production. Before fermentation, the feedstock was shredded by a grinder smaller than 5 mm in diameter and kept in a refrigerator at 1°C to block natural decomposition. The concentrations of total solids (TS), volatile solids (VS), chemical oxygen demand (COD), carbohydrate, and ammonia of FW were 126.9 g TS/L, 122.8 g VS/L, 149.0 g COD/L, 83.7 g Carbo. COD/L (g/L as carbohydrate COD), and 113.5 mg NH4-N/L, respectively.
2.2 Hydrogen fermentation method Acid-pretreatment of FW was conducted at pH 2 by adding a 6 N HCl solution, and was continued for 12 h with agitation at 100 rpm. Then, the acid-pretreated FW corresponding to 60 g Carbo. COD/L was added to the batch fermenter with an effective volume of 300 mL, while the remainder of the effective volume was filled with tap water. Neither an external inoculum nor a basal medium was added. Before fermentation, the pH was readjusted to 8.0±0.1, which was identified as the optimal initial pH value in our previous work, by 6 N KOH addition (Kim et al., 2011). Each batch fermenter was equipped with a pH sensor. After purging with N2 gas for 10 min to provide anaerobic conditions, the fermentation began. The pH was allowed to drop from 8.0 to 6.0 via fermentation, after which point pH of 6.0±0.1, 6.0±0.3, 6.0±0.5, 6.0±0.7, and 6.0±0.9 was automatically controlled by the addition of a 3 N KOH solution using pH controller (Cns, Model Marado-PDA). For example, at pH 6.0±0.3, the alkaline solution pump started to act at pH 5.7 and stopped when pH reached 6.3. The optimal operational pH in H2 fermentation was generally reported in the range of 5.0-6.5, in which the activity of hydrogenase intense and 4
the metabolic pathway for H2 production is thermodynamically favorable (Dabrock et al., 1992; Khanal et al., 2004). The fermenter was agitated at 100 rpm by a mechanical mixer and placed in a temperature controlled room at 35±1oC. The gas produced and its constituents were monitored at 1-5 h intervals. All tests were carried out in triplicate and the results were averaged.
2.3 Analytical methods Measured biogas production was adjusted to the standard conditions of temperature (0°C) and pressure (760 mmHg) (STP). The H2 content in the biogas was determined by a gas chromatography (GC, Gow Mac series 580) using a thermal conductivity detector and a 1.8 m × 3.2 mm stainless-steel column packed with molecular sieve 5A with N2 as a carrier gas. The content of CH4 was measured using a GC of the same model as noted previously with a 1.8 m × 3.2 mm stainless-steel column packed with Porapak Q (80/100 mesh) using N2 as a carrier gas. The temperatures of the injector, detector, and column were kept at 80, 90, and 50°C, respectively, in both GCs (Kim et al., 2011). Concentrations of TS, VS, COD, and ammonia were measured according to Standard Methods (Clescerl et al., 1998). The concentration of carbohydrate was determined by the colorimetric method (Dubois et al., 1956). Organic acids were analyzed by a high performance liquid chromatograph (HPLC) (Finnigan Spectra SYSTEM LC, Thermo Electron Co.) with an ultraviolet (210 nm) detector (UV1000, Thermo Electron) and an 100 mm × 7.8 mm Fast Acid Analysis column (Bio-Rad Lab.) using 0.005 M H2SO4 as a mobile phase. The liquid samples were pretreated with a 0.22 µm membrane filter before injection to the HPLC.
5
2.4 Microbial community analysis 2.4.1 Sampling, DNA extraction, and PCR Deoxyribonucleic acids (DNAs) in a fermented broth were extracted using an Ultraclean Soil DNA Kit (Cat #12800-50; Mo Bio Laboratory Inc., USA). The extracted DNA was subsequently purified with an UltraClean Microbial DNA Isolation Kit (Mo Bio Laboratories, CA, USA). A library was then prepared using polymerase chain reaction (PCR) products according to the GS FLX titanium library prep guide, and the libraries were quantified using Picogreen assay (Victor 3). The emPCR, corresponding to clonal amplification of the purified library, was carried out using a GS-FLX titanium emPCR Kit (454 Life Sciences). A 20 ng aliquot of each sample DNS was used for a 50 ul PCR reaction. The 16S universal primers 27F (5’ GAGTTTGATCMTGGCTCAG 3’) and 800R (5’ TACCAGGGTATCTAATCC 3’) were used to amplify the 16s rRNA genes (Øvreås et al., 1997). A Fast Start High Fidelity PCR System (Roche) was used for PCR under the following conditions: 94°C for 3 min followed by 35 cycles of 94°C for 15 sec; 55°C for 45 sec and 72°C for 1 min; and a final elongation step at 72°C for 8 min.
2.4.2 High throughput pyrosequencing and sequence analysis The PCR products were purified by using AMPure beads (Beckman coulter). After the products were purified and quantified, sequencing was performed using a 454 pyrosequencing Genome Sequencer FLX Titanium (Life Sciences, CT, USA), according to the manufacturer’s instructions, by a commercial sequencing facility (Macrogen, Seoul, Korea).The sequences generated from pyrosequencing were mainly analyzed with the software MOTHUR for preprocessing (quality-adjustment, barcode split), identification of operational taxonomic units 6
(OTUs), taxonomic assignment, community comparison, and statistical analysis (Schloss et al., 2009). Sequences were filtered to minimize the effects of poor sequence quality, and sequencing errors were minimized by removing sequences with more than one ambiguous base call and by retaining only sequences that were 300 nt or longer (Li et al., 2012). Sample-specific sequences were collected according to the barcode sequences tagged to each sample. The barcode, as well as forward and reverse primer sequences, was trimmed from the initial sequences. As a pyrosequence pre-processing step, sequences that were shorter than 300 nt, had one or more ambiguous base calls, or had multiple barcode or primer motifs were excluded from the analysis. The trimmed sequences from each barcode bin were aligned using Infernal, and associated covariance models were obtained from the Ribosomal Database Project Group (Cole et al., 2009). The sequences spanning the same region were then realigned with the NCBI BLAST database (www.ncbi.nlm.nih.gov). In database screening with the BLAST program, the threshold E-value for including a sequence in the next iteration was 0.001. A distance matrix was calculated from the aligned sequences, and operational taxonomical units (OTUs; 90–100% sequence similarity) were assigned using the furthest-neighbor clustering algorithm. The OTUs defined by a 3% distance level were phylogenetically classified with a modified bacterial RDP II database containing 164,517 almost full-length 16S rRNA gene sequences prepared using TaxCollector (http://www.microgator.org).
3. RESULTS AND DISCUSSION 3.1. pH change and H2 fermentation Fig. 1 shows the time courses of cumulative H2 production and pH change profile at various deviation ranges of ±0.1-±0.9. The cumulative H2 production curves were well fitted by the 7
modified Gompertz equation with high R2 values (>0.99) (Chen et al., 2006). The fermentation ended within 30 h, and CH4 was not detected during the entire experimental period. As fermentation began, pH gradually dropped and reached 6.0 after 5.5 h, and H2 started to evolve after 4 h, in all fermenters. At deviation of ±0.1, addition of an alkaline solution was commenced after 5.5 h, and the solution was continuously added every 0.3-0.7 h until 20 h of fermentation at which time H2 was vigorously produced. During the whole fermentation, the total number of times adding alkaline solution was 22 at deviation of ±0.1, while it drastically decreased to 7, 5, 3, and 1 at deviation of ±0.3, ±0.5, ±0.7, and ±0.9, respectively. It appeared that as the deviation increased, the fermentation time when pH was controlled below optimum pH of 6.0 increased. For example, pH was below 6.0 for 14.9 h at deviation of ±0.3, while it was below 6.0 for 19.8 h at deviation of ±0.7. After 5 h of fermentation, 27-29 mL of 3 N KOH was added to keep the pH 6.0±0.1 and ±0.3. At deviation of ±0.5-±0.9, there was a sharp reduction of alkaline solution added, ranging 6-21 mL. Although a 1.5 times higher H2 production rate was achieved at deviation of ±0.1 compared to at ±0.3, there was little apparent difference in the overall performance between the two cases (Table 1). A similar amount of H2 production (3.51-3.64 L) was observed at deviation ranges of ±0.1 and ±0.3, corresponding to a H2 yield of 1.67-1.73 mol H2/mol hexoseadded. On VS and COD basis, the H2 yield ranged 137.0-142.1 mL H2/g VSadded and 109.5-113.6 mL H2/g CODadded, respectively, which were in the range of previous works using organic solid waste as a feedstock (De Gioannis et al., 2013). This indicates that the acid-pretreatment applied in this study was effective for the successful H2 fermentation of FW. A huge drop of H2 production, however, was observed at ±0.5, ±0.7, and ±0.9 with lowered carbohydrate degradation and VS reduction. H2 yield gradually dropped as the deviation 8
increased, with the lowest yield of 0.69 mol H2/mol hexoseadded at ±0.9. Severe inhibition on substrate utilization was observed at deviation ±0.9. Carbohydrate degradation and VS reduction dropped to 59% and 26%, respectively, at ±0.9. This drop of substrate utilization may be one of the reasons for the low production of H2, but the main reason was found by monitoring the production of organic acids.
3.2. Organic acids production The production profiles of organic acids at various pH deviation values are shown in Fig. 2. The main organic acids were butyrate, acetate, and lactate, and their composition was varied depending on the deviation value. In addition, the total production of organic acids decreased from 48.1 to 28.9 g COD/L as the deviation range increased, likely due to a gradual drop of substrate utilization. Butyrate was found to be the dominant metabolite at deviation of ±0.1 and ±0.3, where high H2 yields were achieved, occupying 77.2% and 70.7% of total organic acids produced, respectively. However, at deviation > ±0.5, the production of lactate significantly increased while butyrate production was suppressed. At ±0.7, the highest lactate concentration of 18.5 g COD/L was observed, occupying 54.8% of total organic acids produced. This trend of different composition can be easily linked to the different H2 production performance. According to the following reactions (Eq. (1) and (2)), the production of butyrate has often been related with high H2 yield, while lactate is considered as a main metabolite that should be avoided in H2 fermentation (Angenent et al., 2004). In particular, due to the existence of a large amount of indigenous lactic acid bacteria (LAB), the minimization of lactate production has been raised as a key issue for the successful H2 production from food-processing waste (Noike et al., 2002; Kim et al., 2009). 9
Glucose → Butyrate + 2CO2 + 2H2 + 3ATP
(1)
Glucose → 2Lactate + 2ATP
(2)
From the above results, it was found that high pH deviation caused a drop of H2 production, by decreasing the substrate utilization and shifting the main metabolite from butyrate to lactate. The well-known main H2-producer, Clostridium sp., is also capable of producing lactate, but only under specific conditions such as high H2 partial pressure and nutrient limitation (phosphate and iron) (Bahl and Dürre, 2001). However, even in those conditions, the lactate composition was generally below 20%, which suggests that there has been a change of the microbial community by varying the deviation.
3.3. Microbial community analysis To assess the microbial community structure in response to pH deviation, collected mixed-liquor samples at the end of fermentation were used to pyro-sequence the former region of 16S rRNA gene using the 454 GS-FLX sequencer. A total of 52,531 reads were obtained from a single lane of an 8-lane pico-titer plate on a Genome Sequencer FLX titanium system. The decrease in the Shannon index indicates that species richness and evenness decreased with an increase of deviation (Table 2). The sequences of bacteria were assigned to the genus and species levels to evaluate the relative microbial abundance in each sample. It was observed that the dominant population remarkably changed from H2-producers to LAB as the deviation increased (Fig. 3). In terms of genus level, Clostridium, capable of producing H2, was found in the sample operating at deviation of ±0.1, 10
occupying 76.4% of the total number of sequences (Yun et al., 2014). Meanwhile, Lactococcus, Enterococcus, and Lactobacillus, which all are LAB, accounted for only 7.4%, 3.8%, and 0.4% of the total number of sequences, respectively (Kim et al., 2014). As the pH deviation increased to ±0.7, it was observed that Clostridium drastically dropped to 14.8%. Instead, Streptococcus newly emerged, and its abundance increased to 73.3% of the total number of sequences. This could be related to the characteristic defense mechanism of Streptococcus that allows it to endure pH fluctuations. According to Hamilton and Buckley (1991), Streptococcus cells can compensate for fluctuating external pH via control of intracellular pH to change the permeability of protons into the cell. However, when external pH was further decreased to 5.0, metabolic damage occurred in cells by breaking certain enzymes, and this might be linked to the observation of a noticeable decrease in Streptococcus (0.1%) in the sample with ±0.9 (Harvey, 1965). Meanwhile, the relative abundance of Lactobacillus which can survive at low pH, jumped to 80.9% of the total number of sequences (Kim et al., 2014). However, it was thought that despite the gradual increase in occupation of Lactobacillus in the sample at deviation of ±0.9, a decrease in richness of bacteria at ±0.9 (the Shannon index) limited substrate utilization and lactate production as described above (Fig. 2(e) and Table 2) (Torsvik and Øvreås, 2002). In order to observe the diversity of bacteria in greater detail, the sequences to species level was further assigned by choosing 10 representative microorganisms (sequences > 3%) as shown in Table 3. In total, two representative species belonged to Clostridium sp., Lactobacillus sp., and Streptococcus sp., respectively, and one representative species belonged to Citrobacter sp., Leuconostoc sp., Enterococcus sp., and Lactococcuss sp. The most dominant species in the sample ±0.1 was Clostridium butyricum, which often has been observed in H2 fermentation of organic wastes using anaerobic mixed cultures (Yun et al., 2014). However, an increase of pH 11
deviation resulted in a decreased abundance of Clostridium sp. while Streptococcus lutetiensis and Lactobacillus delbrueckii subsp. bulgaricus ATCC 11842 became dominant. Streptococcus lutetiensis, previously known as Streptococcus infantarius subsp. coli, can be found in feces of human infants (Almuzara et al., 2013; Poyart et al., 2002). Lactobacillus delbrueckii subsp. bulgaricus ATCC 11842 is known to survive in a wide pH range (Adimpong et al., 2013). From this study, it can be concluded that although H2-producers were initially dominated by acid-shock, the precise control of pH should be provided for the successful H2 fermentation. When the deviation was ≥ 0.5, the activity of LAB was revived, which significantly decreased the H2 yield. It is suggested that pH controlling devices such as probe, meter, and alkaline solution pump should be equipped and operated to minimize the deviation less than ±0.3.
4. CONCLUSIONS During the batch H2 fermentation, high H2 yields ranging 1.67-1.73 mol H2/mol hexoseadded were achieved when pH was controlled at 6.0±0.1 and 6.0±0.3. A significant drop of H2 production (0.69-0.98 mol H2/mol hexoseadded), however, was observed when deviation was ≥ 0.5, accompanied with a lowered substrate utilization and the shift of main metabolite from butyrate to lactate. The change of dominant population from Clostridium to LAB with the increase of deviation value was seen by NGS analysis. These findings indicate the importance of pH control accuracy for the successful H2 fermentation, even though the activity of non-H2-producers were initially suppressed.
ACKNOWLEDGEMENT This work was supported by Korea District Heating Corporation, and the Energy Efficiency & 12
Resources Core Technology Program of the Korea Institute of Energy Technology Evaluation and Planning (KETEP), granted financial resource from the Ministry of Trade, Industry & Energy, Republic of Korea (20132020000170)
REFERENCES 1.
Adimpong, D.B., Nielsen, D.S., Sørensen, K.I., Vogensen, F.K., Sawadogo-Lingani, H., Derkx, P.M., Jespersen, L., 2013. Lactobacillus delbrueckii subsp. jakobsenii subsp. nov., isolated from dolo wort, an alcoholic fermented beverage in Burkina Faso. Int. J. Syst. Evol. Microbiol. 63, 3720-3726.
2.
Almuzara, M., Bonofiglio, L., Cittadini, R., Ocampo, C.V., Montilla, A., Del Castillo, M., Ramirez, M.S., Mollerach, M., Vay, C., 2013. First Case of Streptococcus lutetiensis Bacteremia Involving a Clindamycin-Resistant Isolate Carrying the lnuB Gene. J. Clin. Microbiol. 51, 4259-4261.
3.
Amanullah A., McFarlane, C.M., Emery, A.N., Nienow, A.W., 2001. Scale-down model to simulate spartial pH variations in large-scale bioreactors. Biotechnol. Bioeng. 73, 390-399.
4.
Angenent, L.T., Karim, K., Al-Dahhan, M.H., Wrenn, B.A., Domíguez-Espinosa, R., 2004. Production of bioenergy and biochemical from industrial and agricultural wastewater. Trends Biotechnol. 22, 477-485.
5.
Bahl, H., Dürre, P., 2001. Clostridia: Biotechnology and medical applications, Wiley-VCH, Weinheim.
6.
Chen, W.H., Chen, S.Y., Khanal, S., Sung, S. 2006. Kinetic study of biological hydrogen production by anaerobic fermentation. Int. J. Hydrogen Energ. 31, 2170-2178.
7.
Clescerl, L.S., Greenberg, A.E., Eaton, A.D., 1998. Standard methods for the examination of 13
water and wastewater, 20th ed. American Public Health Association, Washington, D.C., pp. 2-57-9. 8.
Cole, J.R., Wang, Q., Cardenas, E., Fish, J., Chai, B., Farris, R.J., Kulam-Syed-Mohideen A.S., McGarrell, D.M., Marsh, T., Garrity, G.M., Tiedje, J.M., 2009. The Ribosomal Database Project: improved alignments and new tools for rRNA analysis. Nucleic Acids Res. 37, D141-145.
9.
Dabrock, B., Bahl, H., Gottschalk, G., 1992. Parameters affecting solvent production by Clostridium pasteurianum. Appl. Environ. Microbiol. 58, 1233-1239.
10. Das, D., Veziroglu, T.N., 2008. Advances in biological hydrogen production processes. Int, J. Hydrogen Energ. 33, 6046-6057. 11. De Gioannis, G., Muntoni, A., Polettini, A., Pomi, R., 2013. A review of dark fermentative hydrogen production from biodegradable municipal waste fractions. Waste Manage. 33, 1345-1361. 12. Dubois, M., Gilles, K.A., Hamilton, J.K., Rebers, P.A., Smith, F., 1956. Colormetric method for determination of sugars and related substances. Anal. Chem. 28, 350-356. 13. Hallenbeck, P.C., Abo-Hashesh, M., Ghosh, D. 2012. Strategies for improving biological hydrogen production. Bioresource Tchenol. 110, 1-9. 14. Hamilton, I.R., Buckley, N.D., 1991. Adaptation by Streptococcus mutans to acid tolerance. Oral Microbiol. Immunol. 6, 65-71. 15. Harvey, R. J., 1965. Damage to Streptococcus lactis resulting from growth at low pH. J. Bacteriol. 90, 1330-1336.
14
16. 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 Energ. 32, 172-184. 17. Kawagoshi, Y., Hino, N., Fujimoto, A., Nakao, M., Fujita, Y., Sugimura, S., Furukawa, K., 2005. Effect of inoculum conditioning on hydrogen fermentation and pH effect on bacterial community relevant to hydrogen production. J. Biosci. Bioeng. 100, 524-530. 18. Khanal, S.K., Chen, W.H., Li, L., Sung, S., 2004. Biological hydrogen production: effects of pH and intermediate products. Int. J. Hydrogen Energ. 29, 1123-1131. 19. Kim, D.H., Kim, S.H., Jung, K.W., Kim, M.S., Shin, H.S., 2011. Effect of initial pH independent of operational pH on hydrogen fermentation of food waste. Bioresource Technol. 102, 8646-8652. 20. Kim, D.H., Kim, S.H., Shin, H.S., 2009. Hydrogen fermentation of food waste without inoculum addition. Enzyme Microb. Tech. 45, 181-187. 21. Kim, D.H., Lee, M.K., Moon, C., Yun, Y.M., Lee, W., Oh, S.E., Kim, M.S., 2014. Effect of hydraulic retention time on lactic acid production and granulation in an up-flow anaerobic sludge blanket reactor. Bioresource Technol. 165, 158-161. 22. Lay, J.J., 2000. Modeling and optimization of anaerobic digested sludge converting starch to hydrogen. Biotechnol. Bioeng. 68, 269-278. 23. Levin, D.B., Pitt, L., Lov, M., 2004. Biohydrogen production: prospects and limitations to practical application. Int. J. Hydrogen Energ. 29, 173-185. 24. Li, W., Fu, L., Niu, B., Wu, S., Wooley, J., 2012. Ultrafast clustering algorithms for metagenomic sequence analysis. Brief. Bioinform. 13, 656-668. 25. Noike, T., Takabatake, H., Mizuno, O., Ohba, M., 2002. Inhibition of hydrogen fermentation 15
of organic wastes by lactic acid bacteria. Int. J. Hydrogen Energ. 27, 1367-1371. 26. Øvreås, L., Forney, L., Daae, F.L., Torsvik, V., 1997. Distribution of bacterioplankton in meromictic Lake Saelenvannet, as determined by denaturing gradient gel electrophoresis of PCR-amplified gene fragments coding for 16S rRNA. Appl. Environ. Microbiol. 63, 33673373. 27. Poyart, C., Quesne, G., Trieu-Cuot, P., 2002. Taxonomic dissection of the Streptococcus bovis group by analysis of manganese-dependent superoxide dismutase gene (sodA) sequences: reclassification of 'Streptococcus infantarius subsp. coli' as Streptococcus lutetiensis sp. nov. and of Streptococcus bovis biotype 11.2 as Streptococcus pasteurianus sp. nov. Int. J. Syst. Evol. Microbiol. 52, 1247-1255. 28. Schloss, P.D., Westcott, S.L., Ryabin, T., Hall, J.R., Hartmann, M., Hollister, E.B., Lesniewski, R.A., Oakley, B.B., Parks, D.H., Robinson, C.J., Sahl, J.W., Stres, B., Thallinger, G.G., Van Horn, D.J., Weber, C.F., 2009. Introducing mothur: open-source, platform independent, community-supported software for describing and comparing microbial communities. Appl. Environ. Microbiol. 75, 7537-7541. 29. Torsvik, V., Øvreås, L., 2002. Microbial diversity and function in soli: from genes to ecosystems. Curr. Opin. Microbiol. 5, 240-245. 30. Yun, Y.M., Kim, D.H., Oh, Y.K., Shin, H.S., Jung, K.W., 2014. Application of a novel enzymatic pretreatment using crude hydrolytic extracellular enzyme solution to microalgal biomass for dark fermentative hydrogen production. Bioresource Technol. 159, 365-372.
16
Fig.1 Cumulative H2 production and pH change profile during H2 fermentation of food waste at pH (a) 6±0.1, (b) 6±0.3, (c) 6±0.5, (d) 6±0.7, and (e) 6±0.9 Fig. 2 Organic acids production profile during H2 fermentation of food waste at pH (a) 6±0.1, (b) 6±0.3, (c) 6±0.5, (d) 6±0.7, and (e) 6±0.9 Fig. 3 Next generation sequencing results in genus level (The deviation was varied from ±0.1 to ±0.9 at an optimum set pH value of 6.0.)
17
Fig.1 Cumulative H2 production and pH change profile during H2 fermentation of food waste at pH (a) 6±0.1, (b) 6±0.3, (c) 6±0.5, (d) 6±0.7, and (e) 6±0.9
18
Fig. 2 Organic acids production profile during H2 fermentation of food waste at pH (a) 6±0.1, (b) 6±0.3, (c) 6±0.5, (d) 6±0.7, and (e) 6±0.9
19
20
Fig. 3 Next generation sequencing results in genus level (The deviation was varied from ±0.1 to ±0.9 at an optimum set pH value of 6.0)
21
Table 1 Effect of pH control accuracy on average H2 fermentation performance (The deviation was varied from ±0.1 to ±0.9 at an optimum set pH value of 6.0) H2 yield Sample
Carbohydrate
VS
H2 production
mol H2/mol
mL H2
mL H2
removal
reduction
rate
hexoseadded
/g VSadded
/g CODadded
(%)
(%)
(L H2/L/d)
6±0.1
1.73
142.1
113.6
89 ± 2
50 ± 4
26.7
6±0.3
1.67
137.0
109.5
88 ± 1
50 ± 3
18.1
6±0.5
0.98
80.5
64.4
81 ± 1
45 ± 2
15.1
6±0.7
0.79
64.4
51.5
64 ± 1
35 ± 3
11.2
6±0.9
0.69
56.9
45.5
59 ± 1
26 ± 2
15.4
22
Table 2 Analytical results of the microbial gene libraries obtained from next generation sequencing Sample
Read data
Shannon index
pH 6.0±0.1
10,148
1.9065±0.0363
pH 6.0±0.3
10,338
1.7475±0.0293
pH 6.0±0.5
9,775
1.5294±0.0349
pH 6.0±0.7
10,327
1.3171±0.0352
pH 6.0±0.9
11,943
1.0762±0.0316
23
Table 3 Species level identification of the dominant sequences from each sample (>3% of total sequences)
Microorganism
Clostridium butyricum Streptococcus lutetiensis Streptococcus gallolyticus
pH
pH
pH
pH
pH
6.0±0.
6.0±0.
6.0±0.
6.0±0.
6.0±0.
1 (%)
3 (%)
5 (%)
7 (%)
9 (%)
58.3
26.3
8.7
11.0
6.8
0.0
46.7
70.6
65.3
0.0
0.0
0.0
8.2
7.0
0.0
0.0
0.0
0.0
0.0
78.7
0.3
0.6
2.4
1.3
0.8
6.8
0.1
0.0
0.0
2.5
3.7
2.5
2.1
4.9
3.3
0.0
0.4
0.9
1.9
0.6
0.3
0.2
0.1
0.1
1.1
2.3
0.0
0.0
0.0
0.0
Accessio n (#) NR_0421 44.1 NR_0370 96.1 NR_0748 49.1
Simil arity (%) 99.0
100.0 100.0
Lactobacillus delbrueckii subsp. bulgaricus ATCC
NR_0750 19.1
99.0
11842 Clostridium baratii Lactococcus lactis subsp. lactis Il1403 Enterococcus hirae ATCC 9790 Lactobacillus reuteri DSM 20016 Leuconostoc lactis KCTC 3528 Citrobacter youngae
24
NR_0292 29.1 NR_1039 18.1 NR_0750 22.1 NR_0750 36.1 NR_0408 23.1 NR_0415 27.1
98.0 99.0
99.0 99.0
99.0 98.0
25
Highlights . A significant effect of pH control accuracy on H2 fermentation performance High H2 yield and substrate utilization at deviation of ±0.1 and ±0.3 Low H2 production with lowered substrate utilization at deviation > ±0.5 Lactic acid bacteria became dominant with increase of deviation value.
26