Environ. Sci. Technol. 2010, 44, 3317–3323
Waste Activated Sludge Fermentation for Hydrogen Production Enhanced by Anaerobic Process Improvement and Acetobacteria Inhibition: The Role of Fermentation pH YUXIAO ZHAO, YINGUANG CHEN,* DONG ZHANG, AND XIAOYU ZHU State Key Laboratory of Pollution Control and Resources Reuse, School of Environmental Science and Engineering, Tongji University, 1239 Siping Road, Shanghai 200092, China
Received September 30, 2009. Revised manuscript received January 2, 2010. Accepted March 20, 2010.
In this study an efficient strategy, i.e., controlling the fermentation pH at constant pH 10, for significantly increasing hydrogen yield from waste activated sludge (WAS) via the improvement of anaerobic process (sludge solubilization, hydrolysis, and acidification) and inhibition of hydrogen consumption by acetobacteria was reported. Without addition of pure hydrogen producer and nutrient source, the effect of different constant pH in the range of pH 4-11 on hydrogen production from WAS was compared with that of different initial pH. The maximal hydrogen yield was observed respectively at constant pH 10 and initial pH 10, but the former was 47.8% higher than the latter (26.9 versus 18.2 mL per gram volatile suspended solids) and much greater than that reported in literature. Then, the mechanisms for constant pH 10 resulting in remarkably higher hydrogen production than initial pH 10 were investigated. It was observed that constant pH 10 fermentation showed much higher solubilization of sludge main particulate organic matters, hydrolysis of solubilized organic materials and acidification of hydrolyzed products, which were of benefit to the hydrogen production. Also, there was more acetic but less propionic acid in the constant pH 10 test, which was in correspondence with the theory of fermentation type affecting hydrogen production. Moreover, in the reactor of initial pH 10 the produced hydrogen was readily converted toaceticacid,butnoobvioushydrogenconsumptionwasobserved in constant pH 10 reactor. Further investigation of microorganisms with enzymes analysis and fluorescence in situ hybridization (FISH) indicated that the activity and growth of acetobacteria in the reactor of constant pH 10 was much lower than those in initial pH 10 reactor.
Introduction Recently, biological production of hydrogen from organic wastes has drawn much attention (1-4), by which human resource, such as fossil fuels, is saved, wastes is reused, and sustainable H2 supply can be attained. Waste activated sludge (WAS) is a byproduct of biological municipal wastewater * Corresponding author phone: 86-21-65981263; fax: 86-2165986313; e-mail:
[email protected]. 10.1021/es902958c
2010 American Chemical Society
Published on Web 04/08/2010
treatment plant. It contains a significant amount of protein and carbohydrate. Several publications have reported that hydrogen can be biologically produced by anaerobic fermentation of WAS (5-7). Usually, there are four stages involved in anaerobic fermentation of sludge: solubilization of sludge particulate organic matters, hydrolysis of solubilized sludge organic materials, acidification of hydrolyzed products, and methane generation. Hydrogen is produced in the acidification step. Obviously, improving the first three steps and inhibiting the last one would increase hydrogen production during sludge anaerobic fermentation. Nevertheless, most of the studies in literature focused on preventing methane generation, and pretreating sludge by ultrasonic, acidic, alkaline, heat-shocking, or freezing-thawing method was reported to be able to kill methanogens and increase hydrogen production (8-11). Although biohydrogen yield could be increased by pretreating sludge with the method reported in literature, the accumulated H2 was still observed to be consumed during sludge anaerobic fermentation even when there were no methanogens (9, 12). It seems that apart from methanogens other hydrogen consumers are still active after sludge pretreated by the documented methods. As acetobacteria have been reported to be able to bioconvert hydrogen to acetate (see eq 1) under anaerobic conditions, their presence has been proposed to be responsible for the consumption of hydrogen during biohydrogen production from organic wastewater or wastes in the case of no methanogens (13, 14). Thus, inhibiting the activity of acetobacteria would improve biohydrogen production. + 4H2 + 2HCO3 + H ) CH3COO + 4H2O
(1)
During biohydrogen production the main metabolism of organic compound in the step of acidification, taking glucose as an example may undergo different types of fermentation, i.e., acetic, propionic, butyric, and ethanol type-fermentation. To a mixed (such as activated sludge) microbial fermentation, usually more than one liquid product (acetic, propionic, etc.) can be observed (15). Two molecules of hydrogen are generated with ethanol, acetic acid, and butyric acid type fermentations, while no hydrogen is produced with propionic acid one (16). Nevertheless, there is no investigation made in literature on hydrogen production from waste activated sludge influenced by the change of fermentation type. In this study an efficient strategy, i.e., controlling the fermentation pH at constant pH 10, for significantly increasing hydrogen yield from waste activated sludge via the improvement of anaerobic process (sludge solubilization, hydrolysis, and acidification) and inhibition of hydrogen consumption by acetobacteria was reported. As some researchers have reported that a constant acidic pH or an alkaline initial pH can inhibit methane generation and therefore improve hydrogen production from organic wastewater or wastes (9, 12, 17, 18), the comparison of initial and constant pH (in the range of pH 4-11) affecting biohydrogen production from waste activated sludge without addition of pure hydrogen producer and nutrient source was made first. Then, the reasons for significantly higher hydrogen production achieved at constant pH10 than at initial pH 10 were investigated.
Materials and Methods Waste Activated Sludge. The WAS was withdrawn from the secondary sedimentation tank of a municipal wastewater treatment plant in Shanghai, China. The sludge was conVOL. 44, NO. 9, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3317
centrated by settling at 4 °C for about 24 h, and its main characteristics (average data and standard deviations of three tests) are as follows: pH 6.6 ( 0.2, total suspended solids (TSS) 17.34 ( 1.58 g/L, volatile suspended solids (VSS) 12.50 ( 1.09 g/L, soluble chemical oxygen demand (SCOD) 145 ( 23 mg/L, total chemical oxygen demand (TCOD) 17400 ( 260 mg/L, total carbohydrate 1632 ( 79 mg-COD/L, and total protein 6133 ( 169 mg-COD/L. Heat Pretreatment. As hydrogen produced under anaerobic conditions can be consumed by both methanogens and acetobacteria, in order to investigate the influence of different pH control strategy on the activity of acetobacteria, it is necessary to remove methanogens from all fermentation systems. Although at constant pH 10 methane generation could be inhibited (19), there was some methane in the reactor of initial pH 10. In order to eliminate the influence of methanogens, all waste activated sludge used in this study was pretreated at 102 °C for 30 min, which has been reported to be an efficient method to kill methanogens (12). After cooling down to room temperature, the waste activated sludge was used for the following investigations. All the following experiments were conducted in triplicate, and one way analysis of variance (ANOVA) at 0.05 level was used to analyze the data. Comparison of H2 Production at Different Initial and Constant pH. Sixteen serum bottles, with 400 mL of the heat pretreated sludge each, were divided into two groups with 8 in each one. The pH value of sludge in both groups 1 and 2 was adjusted to 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0, 11.0, respectively, by adding 4 M hydrochloric acid (HCl) or 4 M sodium hydroxide (NaOH). After flushed with nitrogen gas to remove oxygen, all bottles were capped with rubber stoppers, sealed, and placed in an air-bath shaker (120 rpm) at 37 ( 1 °C. During the entire fermentation process, the pH value in group 1 was not adjusted but recorded, whereas it was controlled to the initial value in group 2 by adding 4 M HCl or 4 M NaOH with an automatic titrator. The total gas volume was measured by releasing the pressure in the bottles using a glass syringe (100 mL) to equilibrate with the room pressure as in the Owen method (20), and the syringe was always in the bottle so that the accumulative volume was followed with time. The cumulative hydrogen gas volume was calculated by the following mass balance equation (13) VH,i ) VH,i-1 + CH,i(VG,i)- CH,i-1(VG,i-1)
(2)
where VH,i and VH,i-1 are respectively the cumulative hydrogen gas volumes in the current (i) and previous (i-1) time intervals, VG,i and VG,i-1 are respectively the total biogas volumes in the current and previous time intervals, CH,i and CH,i-1 are respectively the fractions of hydrogen gas in the headspace of the bottle measured using gas chromatography in the current and previous time intervals, and VH is the total volume of headspace in the reactor. Gas was sampled with a 5 mL syringe for composition analysis. By analysis of the changes of hydrogen, soluble protein and polysaccharide, the effects of constant and initial pH 10 on hydrogen and sludge main particulate organic matters solubilization were obtained. Experiments of Constant and Initial pH 10 Affecting the Hydrolysis of Solubilized Sludge Organic Matter. Soluble protein and polysaccharide were the main solublized sludge particulate organic matters, to investigate the effect of constant pH 10 and initial pH 10 on the hydrolysis of solubilized sludge particulate organic materials, the following fermentation tests with synthetic wastewater containing bovine serum albumin (BSA, average molecular weight Mw 67000, model protein compound used in this study) and dextran (Mw ∼23800, model polysaccharide compound) were conducted. The synthetic wastewater, which consisted of (mg/L of tap water) 1000 KH2PO4, 400 CaCl2, 600 MgCl2 · 6H2O, 3318
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 9, 2010
100 FeCl3, 0.5 ZnSO4 · 7H2O, 0.5 CuSO4 · 5H2O, 0.5 CoCl2 · 6H2O, 0.5 MnCl2 · 4H2O, and 1 NiCl2 · 6H2O, of 800 mL was divided equally into 2 serum bottles, and heat-pretreated sludge of 40 mL was inoculated in each bottle. 1.6 g BSA and 0.4 g dextran were added to each bottle and dissolved (the mass ratio of protein and carbohydrate was almost the same as that in the sludge). The pH in 2 bottles was adjusted to 10.0 by adding 4 M NaOH. During the experiments, the pH in one bottle was maintained at pH 10 but not controlled in another one. All other operations were the same as described above. Experiments of Constant and Initial pH 10 Affecting Acidification of Hydrolyzed Products. After hydrolysis protein and polysaccharide were converted to amino acid and monosaccharide, respectively. The comparison of the effect of constant and initial pH 10 on the acidification of hydrolyzed products was conducted with synthetic wastewater containing L-alanine (model amino acid compound used in this study) and glucose (model monosaccharide compound). The above synthetic wastewater of 800 mL was divided equally into 2 serum bottles, and the heat-pretreated sludge of 40 mL was inoculated in each bottle. 1.6 g L-alanine and 0.4 g glucose were added to each bottle and dissolved. The pH value in 2 bottles was adjusted to 10.0 by adding 4 M NaOH. During the experiments, the pH in one bottle was maintained at pH 10 but not controlled in another one. All other operations were the same as described above. Experiments of Effect of Different Constant pH on Fermentation Type. In order to study the effect of pH on fermentation type, i.e., on bioconversion of hydrolyzed products to volatile fatty acids (VFA) and hydrogen, the following experiments with identical synthetic wastewater were conducted. The above synthetic wastewater of 1600 mL was divided equally into 4 serum bottles, and the heatpretreated sludge of 40 mL was inoculated in each bottle. 1.6 g BSA and 0.4 g dextran were added to each bottle and dissolved (the mass ratio of protein and carbohydrate was almost the same as that in sludge and solution). The pH value in 4 bottles was adjusted respectively to 7.0, 8.0, 9.0, and 10.0 by adding 4 M HCl or 4 M NaOH with an automatic titrator. The pH in each bottle was maintained at its initial value. Sample was obtained with a 5 mL syringe. All other operations were the same as described above. Experiments of Influence of Constant pH 10 and Initial pH 10 on Hydrogen Consumption. The pH value of pretreated sludge of 1600 mL was adjusted to pH 10 and divided equally into four serum bottles (A1, A2, A3, and A4). During the entire fermentation process, the pH value in A1 and A2 was not adjusted, whereas it was controlled at pH 10 in A3 and A4. When the hydrogen production did not increase with time, the VFA concentration was measured. Then, the bottles of A1 and A3 were flushed with a combined gas (40% hydrogen, 10% carbon dioxide, and 50% nitrogen) for 5 min to ensure that they were filled with the synthetic hydrogencontaining gas, and the A2 and A4 bottles (served as the controls) were flushed with pure nitrogen (99.99%). All bottles were capped again with rubber stoppers, sealed, and placed in a air-bath shaker (120 rpm) at 37 ( 1 °C for 48 h. Hydrogen gas in the headspace and VFA in the liquid were measured after 48 h. Analytical Methods. Gas component was measured using a gastight syringe (0.1 mL injection volume) and a gas chromatograph (GC112A, China) equipped with a thermal conductivity detector with nitrogen as the carrier gas. The determinations of VFA, ethanol, lactic acid, protein, carbohydrate, TS, and VS were the same as described in our previous publications (19, 21). The pH value was measured by a pH meter. The COD (chemical oxygen demand) conversion factors are 1.5 g-COD/g protein, 1.06 g-COD/g carbohydrate, 1.07 g-COD/g acetic, 1.51 g-COD/g propionic, 1.82 g-COD/g butyric, and 2.04 g-COD/g valeric (22). The
FIGURE 1. Effect of different initial pH (a) and different constant pH (b) on average hydrogen production during sludge fermentation. Error bars represent standard deviations of triplicate tests. total VFA was calculated in this study as the sum of measured acetic, propionic, n-butyric, isobutyric, n-valeric, and isovaleric acid. The activities of pyruvate-ferredoxin oxidoreductase (POR), formate dehydrogenase (FDH), and formyltetrahydrofolate synthetase (FTHFS) were assayed in this study. For their determinations, 25 mL of the fermentation mixture was taken out of the anaerobic fermentation bottles and then washed and resuspended in 10 mL of 100 mM sodium phosphate buffer (pH 7.4). The suspension was sonicated at 20 kHz and 4 °C for 30 min to break down the cells of acidogenic bacteria and then centrifuged at 10000 rpm and 4 °C for 30 min to remove the waste debris. The extracts were kept cold on ice before they were used for enzyme activity assay. POR activity was assayed as Yakunin et al (23). FDH was assayed by measuring the increase of absorbency at 340 nm with nicotinamide adenine dinucleotide (NAD) as the electron acceptor. FTHFS was assayed as 5,10-methenyltetrahydrofolic acid forming by measuring the increase of absorbency at 350 nm (24). The FISH (fluorescence in situ hybridization) technique with 16S rRNA-targeted oligonucleotide probes was employed to monitor the difference of microbial community when sludge was fermented for hydrogen production at constant pH 10 and initial pH 10. In situ hybridization was performed according to the protocol proposed by Moter and Go¨bel (25). The oligonucleotide probes used in this study were specific for acetogenic bacteria (Chis150, 5′- TTA TGC GGT ATT AAT CTY CCT TT -3′) (26) and acetobacterium species (AW, 5′GGC TAT TCC TTT CCA TAG GG -3′) (27). Oligonucleotides were synthesized and fluorescently labeled with HEX for Chis150 and 6-FAM for AW at the 5′ end. At the time of maximal hydrogen production appeared, the anaerobic biomass was withdrawn from the reactors and fixed in 4% freshly prepared paraformaldehyde solution for 3-4 h at 4 °C. 10 µL of fixed sample was mounted on a glass slide using a micropipet. It was then prehybridized in 5 × SSC, 0.1% N-lauroylsarcosine, 0.02% SDS, 1% blocking reagent, and formamide for 0.5 h at 50 °C (28). After removal of the prehybridization solution, 20 µL of hybridization buffer (0.9 M NaCl, 20 mM Tris-HCl (pH 7.2), 0.01% SDS) with 5 ng probes were applied to the sample. Hybridizations were performed in a hybridization incubator (ThermoBrite, America) at 46 °C for 3 h. The hybridization stringency was adjusted by adding formamide to the hybridization buffer (30% for AW and 45% for Chis150). Hybridization was followed twice by a stringent washing buffer (20 mM TrisHCl (pH 7.2), 0.01% SDS). The washing buffer was removed by rinsing the slides with distilled water, and the slides were air-dried. The slides were mounted to avoid bleaching and examined with epifluorescence microscope (Nikon, Japan). Percentage distributions of the acetogenic bacteria and
acetobacteria were determined from fraction area occupied by the different bacteria with image analysis system (ImagePro Plus, V6.0, Media Cybernetics).
Results Comparison of H2 Production at Different Initial and Constant pH. As shown in Figure 1(a), the hydrogen production in the range of pH 4-11 first increased with time extension and then decreased with further increase of fermentation time. It should be noted that at any initial pH investigated, the hydrogen accumulated in all bottles was consumed by hydrogen consumers (no methane gas was detectable) after 120 h. It can also be seen from Figure 1(a) that when the fermentation initial pH changed, the maximal hydrogen production varied. At initial pH 4, the maximal hydrogen was only 3.2 mL/g-VSS, which was the lowest one in Figure 1(a). The maximal hydrogen production increased almost with the increase of initial pH. Nevertheless, as initial pH increased from 10 to 11, there was little increase in the maximal hydrogen production (from 18.2 to 18.8 mL/g-VSS, or 13.5 to 14.0 mL/g-TSS). The observations made in Figure 1(a) were almost the same as those in the literature in the range of initial pH 4-10 (9). Figure 1(b) illustrates the time curves of cumulative hydrogen production at different constant pH. Compared with Figure 1(a) it can be seen that the behavior of hydrogen production at different constant pH was almost the same as that at different initial pH except for pH 10 and pH 11. There was almost no hydrogen produced at constant pH 11, which might be due to the continuously toxic effect of strong alkalinity, and there was no hydrogen consumption observed at constant pH 10. During the initial 116 h the hydrogen production at constant pH 10 increased significantly with time (F ) 2.24, Fcrit ) 1.79, P ) 0.01 < 0.05), and no significant increase was observed after that time (F ) 0.05, Fcrit ) 2.42, P ) 0.99 > 0.05). At time of 116 h the hydrogen production at constant pH 10 was 26.9 mL/g-VSS (20.0 mLH2/g-TSS), which was 47.8% higher than the maximal hydrogen achieved with the initial pH 10.0 fermentation. It should be emphasized that hydrogen consumption occurred in all tests except for constant pH 10. As far as we know the hydrogen production at constant pH 10 was much greater than that reported in literature (7-11). In the following text the reasons for constant pH 10 fermentation showing significantly higher hydrogen production than initial pH 10 were addressed. Effect of Initial and Constant pH 10 on Solubilization of Sludge Main Particulate Organic Matters. Sludge organic matters are mainly present in particle form. During sludge fermentation for biohydrogen production, these organic materials usually undergo solubilization, hydrolysis, and acidification. The changes of soluble protein and carbohydrate were applied in the current study to express the VOL. 44, NO. 9, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3319
FIGURE 2. Effect of initial and constant pH on the concentrations of soluble protein (a) and carbohydrate (b) at fermentation time of 46 and 65 h. At these two fermentation times the average hydrogen productions with initial pH 10 were almost the same as those with constant pH 10. Error bars represent standard deviations of triplicate tests. solubilization of sludge main particulate organic matters. As shown in Figure 2, both the soluble protein and carbohydrate with the constant pH 10 fermentation were greater than with the initial pH 10, which suggested that constant pH 10 provided more soluble substrates for hydrogen production. It is well-known that sludge components are cemented together by extracellular polymeric substances (EPS), such as polysaccharide and protein (19). Also, there are large numbers of microorganisms, and protein and polysaccharide are cell composition. It was observed that no matter the initial pH was 4 or 10, in the experiments of initial pH affecting hydrogen production the pH changed with time, and the final pH was between 5.5-7.5 (Figure S1, Supporting Information). To the initial pH 10 test, its pH decreased remarkably from 10 to 7.3 during the initial 46 h and then decreased slightly with time (the final pH was 7.2). According to our previous publication, the solubilization of sludge protein and carbohydrate increased significantly with pH FIGURE 3. Effect of initial and constant pH on VFA and (19). It was due to the decrease of pH from initial 10 to final hydrogen at the time of maximal hydrogen production. Error 7.2 which caused lower solubilization of sludge particulate bars represent standard deviations of triplicate tests. protein and carbohydrate in the test of initial pH 10 than that of constant pH 10; lower concentrations of soluble constant pH 10 were significantly higher than that with initial protein and carbohydrate were therefore observed. pH 10. Thus, one reason for sludge fermentation at constant Effect of Initial and Constant pH 10 on Hydrolysis of pH 10 showing higher hydrogen production was due to the Soluble Protein and Polysaccharide. The effect of initial improvement of acidification of hydrolyzed products. and constant pH 10 on the hydrolysis of solubilized sludge Effect of Initial and Constant pH 10 on Fermentation main particulate organic matters was investigated with Type. Hydrogen production is also affected by the fermentasynthetic wastewaters of protein and polysaccharide. The tion type, the higher the acetic, butyric, and ethanol or the degradation of BSA and dextran was 54.5% and 84% with lower the propionic, the greater hydrogen production is constant pH 10, whereas it was respectively 40.1% and 67.1% (16, 18). In this study when sludge was fermented respectively with initial pH 10 at the time of maximal hydrogen production at constant and initial pH 10, the distributions of individual (84 h for constant pH 10 and 48 h for initial pH 10, see Figure VFA at the time of maximal hydrogen production are shown S2, Supporting Information). Although partial BSA and in Figure 4 (no ethanol and lactic acid were detectable in all dextran might be adsorbed by sludge, from the difference of tests). There were no significant differences in the concenconstant pH 10 and initial pH 10, it can be concluded that trations and percentages of n-butyric, isobutyric, n-valeric, constant pH 10 significantly enhanced the hydrolysis of both and isovaleric acid between two experiments. The statistical soluble protein and polysaccharide. The same observations analysis (Table S1, Supporting Information) indicated that could be made even after fermentation for 96 h (see Figure the concentration and percentage of acetic acid in the reactor S3, Supporting Information). More hydrolysis provided more of constant pH 10 were significantly greater than those in substrate for acidification, and greater hydrogen production the reactor of initial pH 10 (1193.4 against 879.4 mg-COD/L, was observed (see Figure S2b, Supporting Information). and 58.2% versus 45.1%), but the opposite observations were Effect of Initial and Constant pH 10 on Acidification of made with propionic acid (227.4 against 407.9 mg-COD/L Hydrolyzed Products. The hydrolyzed products of protein and 11.1% versus 20.9%). It was found in our previous study and polysaccharide were converted to VFA and hydrogen in that the activity of oxaloacetate transcarboxylase, a key acidification step. The results at the time of maximal hydrogen enzyme related to propionic acid formation, was lower at production (84 h for constant pH 10 and 48 h for initial pH pH 10 than that in the range of pH 6-9 (21), which might 10) are shown in Figure 3. It should be emphasized that apart be the reason for more propionic acid in the reactor of initial from acetic and propionic acids there was no other VFA and pH 10 (the pH decreased to around 7 in the initial pH 10 ethanol generated, and the concentration of propionic was experiment, Figure S1). greater than that of acetic when the current synthetic As discussed above the pH decreased from 10 to 7.2 in the wastewater was fermented at initial or constant pH 10. As initial pH 10 test, whereas in the constant pH 10 experiment seen from Figure 3, acetic, propionic and hydrogen with the pH maintained at 10. It might be that the lower acetic 3320
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 9, 2010
FIGURE 4. Effects of initial and constant pH on the concentration of individual VFA (a) and the percentage of individual VFA accounting for total VFA (b) at the time of maximal hydrogen production (acetic (A), propionic (P), n-butyric (n-B), isobutyric (iso-B), n-valeric (n-V), and isovaleric acid (iso-V)). Error bars represent standard deviations of triplicate tests.
FIGURE 5. Effect of different constant pH on VFA at the time of maximal hydrogen production during fermentation of synthetic wastewater of protein and carbohydrate. A (acetic), P (propionic), and B (the sum of n-butyric and isobutyric). n-Valeric and isovaleric acid, the lowest two VFA, were measured at the time of maximal hydrogen production, and their average total concentrations were respectively 338.4, 316.5, 380.2, and 394.0 mg-COD/L at pH 7, pH 8, pH 9, and pH 10, which accounted for 20.3%, 17.4%, 18.6%, and 18.3% of the total VFA.
TABLE 1. Comparison of VFA Concentration and Hydrogen Consumption after Filling the Hydrogen-Containing Synthetic Gas for 48 h VFA concentration (mg-COD/L)a
H2 filling control H2 filling constant pH10 test control initial pH10 test
A
P
n-B
iso-B
n-V
iso-V
H2 consumption (%)
1303.6 ( 42.4 1113.3 ( 32.5 1496.4 ( 38.0 1452.2 ( 40.9
482.3 ( 21.4 431.8 ( 18.8 370.4 ( 20.1 357.5 ( 27.8
227.1 ( 19.9 241.4 ( 20.7 266.5 ( 20.6 261.3 ( 16.0
339.2 ( 27.3 327.8 ( 20.8 250.6 ( 17.2 243.7 ( 15.3
94.6 ( 7.0 90.2 ( 5.9 84.5 ( 6.0 92.3 ( 8.1
275.7 ( 22.8 287.6 ( 14.3 327.8 ( 20.4 316.4 ( 22.1
97.5 ( 0.6 --b 1.3 ( 0.1 --
a Before hydrogen filling the average concentration of acetic, propionic, n-butyric, isobutyric, n-valeric, and isovaleric acid was respectively 952.0, 446.3, 184.2, 211.7, 64.8, and 246.9 mg-COD/L in the initial pH10 test, which was respectively 1286.4, 265.9, 158.7, 237.2, 53.2, and 271.1 mg-COD/L in the constant pH10 test. A (acetic), P (propionic), n-B (n-butyric), iso-B (isobutyric), n-V (n-valeric), and iso-V (isovaleric). b No hydrogen was filled in the control experiments.
and higher propionic acid observed in the initial pH 10 test was due to the decrease of fermentation pH. In the following text different constant pH in the range of pH 7-10 affecting VFA formation and hydrogen production were conducted in a series of reactors with identical synthetic wastewater containing soluble protein and polysaccharide, and the results are shown in Figure 5 (at pH 7 and pH 8 the maximal hydrogen production appeared at the fermentation time of 48 h, whereas it appeared at 60 and 84 h at pH 9 and pH 10). The statistical analysis indicated that pH significantly influenced VFA and hydrogen generation (Table S2, Supporting Information). As seen in Figure 5a, the percentages of acetic were the greatest at pH 10 and pH 9 (39.3% and 32.9%), indicating that the fermentation type was an acetic one. Nevertheless, the fermentation type became propionic
at pH 8 and pH 7. The data in Figure 5b showed that the sum of acetic and butyric (A+B) declined with the decrease of pH, which was consistent with the variation of hydrogen production. As discussed above, acetic and butyric typefermentations benefit hydrogen production. It can be easily understood therefore that sludge fermentation in constant pH 10 test had higher hydrogen production than in initial pH 10 experiment. Effect of Constant and Initial pH 10 on Hydrogen Consumption. As shown in Table 1, after fermentation for 48 h the average acetic was 1113.3 mg-COD/L in the control test of initial pH 10 (without hydrogen filling), whereas it was 1303.6 mg-COD/L in the hydrogen filling experiment. Obviously, the filling of hydrogen to the initial pH 10 sludge fermentation system caused significant increase of acetic VOL. 44, NO. 9, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3321
mg-protein), the statistical analysis revealed that this difference was insignificant (F ) 4.04, Fcrit ) 7.71, P ) 0.11 > 0.05). It seems that when sludge was fermented at constant pH 10 the activity of hydrogen production bacteria was almost the same as that at initial pH 10. The data in Figure 6b,c, however, indicated that during most of the fermentation time both the activities of FDH and FTHFS in constant pH 10 test were much lower than those in initial pH 10 experiment, which suggested that constant pH 10 showed lower hydrogen consumption and acetic acid synthesis. This was in correspondence with the observations made in the above hydrogen consumption and VFA synthesis experiments (Table 1). Effect of Constant and Initial pH 10 on Sludge FISH Analysis Results. At the time of maximal hydrogen production, the FISH assay of sludge from two reactors was conducted (Figure S5, Supporting Information), and the results were analyzed with image analysis system (ImagePro Plus, V6.0). It was found that the constant pH 10 reactor had a higher acetogenic bacteria (30.0% against 23.2%) and lower acetobacteria (0.6% versus 6.2%) than the initial pH 10 reactor. The ratio of acetobacteria to acetogenic bacteria was 1:50 in the constant pH 10 reactor, whereas it was 1:3.7 in the initial pH 10 system, which further proved that sludge fermentation at constant pH 10 inhibited hydrogen consumption by acetobacteria and thereby enhanced hydrogen yield.
Discussion FIGURE 6. Comparison of the activity of POR (a), FTHFS (b), and FDH (c) in the initial and constant pH 10 sludge fermentation experiments. Error bars represent standard deviations of triplicate tests. acid (F ) 37.97, Fcrit ) 7.71, P ) 0.0035 < 0.05). At the same time, 97.5% hydrogen was consumed in the hydrogen filling reactor. When the pH was controlled at constant pH 10, however, there was very low hydrogen consumption (only 1.3%), and there were no significant changes in VFA concentrations (compared to the control, Table 1). At other fermentation time the same observations were made (data not shown). As no methane was detectable in these experiments due to sludge pretreated by heat, it was more likely that significantly hydrogen consumption in the initial pH 10 test was due to the participation of acetobacteria, a group of bacteria producing acetate from a variety of energy (for example, H2) and carbon (for example, CO2, HCO3-) sources via Wood-Ljungdahl pathway (29) according to eq 1. Thus, one reason for constant pH 10 showing higher hydrogen production than initial pH 10 was due to its inhibitory effect on acetobacteria. Effect of Constant and Initial pH 10 on the Observed Activities of Key Enzymes during Sludge Fermentation. According to the supposed metabolic pathway for biohydrogen production from sludge (Figure S4, Supporting Information), pyruvate-ferredoxin oxidoreductase (POR) is the enzyme that catalyzes thiamine pyrophosphate (TPP)dependent oxidative decarboxylation of pyruvate to form acetyl-CoA, carbon dioxide, and hydrogen (23). Formate dehydrogenase (FDH) and formyltetrahydrofolate synthetase (FTHFS) are the main enzymes involved in the WoodLjungdahl pathway for hydrogen consumption and acetic acid synthesis (29). The time curve of POR activity (Figure 6a) in the initial pH 10.0 test was almost the same as that in the constant pH 10 except that the former took a shorter time to reach the maximal activity than the latter (30 versus 48 h). Although the maximal POR at initial pH 10 was slightly higher than that at constant pH 10 (1.68 against 1.59 units/ 3322
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 9, 2010
Hydrogen production is usually conducted at acidic pH (7, 8, 10, 11). Recently, Cai et al. (9) did the work of initial pH affecting hydrogen production, and they reported that the maximal hydrogen occurred at initial pH 11.0, and less hydrogen consumption occurred at high initial pH. As seen from the current study on initial pH and constant pH affecting hydrogen production, the maximal hydrogen accumulation was observed respectively at initial pH 10 and constant pH 10, and the latter was much higher than the former. However, the produced hydrogen was consumed in all initial pH experiments, but there was no hydrogen consumption at constant pH 10 experiment. Oh et al. (12) in their publication reported that heat treatment was sufficient to prevent methanogenesis, but hydrogen consumption was still observed in their studies, which was supposed to be the presence of acetobacteria. However, no effective method for inhibiting the activity of acetobacteria was reported in their studies. In this study it was found that by controlling the pH at constant pH 10 the activity of acetobacteria could be effectively inhibited. It has been believed that a high initial pH benefited CO2 absorption, which was supposed to be the reason for less hydrogen consumption at high initial pH (9). However it may not be suitable to say that because the consumers such as acetobacteria are in the liquid phase and are not in the gas one. In fact, some researchers have pointed out that HCO3- could be used as a carbon source for hydrogen consumers (12), and the consumption of hydrogen should occur in the liquid phase instead of the gas one. In this study, no CO2 was detected in gas phase at constant pH 8 and 9, but hydrogen consumption still happened. Thus the absence of gas phase CO2 may not be the reason for hydrogen consumption inhibited at constant pH 10 in this study. No hydrogen consumption at constant pH 10 in this study was thought to be constant pH inhibiting the activity of acetobacteria, which has been supported by the data of microorganisms and enzymes. Theoretically the H2 COD was only about 33.3% of glucose COD, if we take glucose for example and suppose it is completely converted to H2 and acetic (30). In fact the
reported conversion from glucose COD to H2 COD was 4%-8.2% with heat pretreatment sludge as hydrogen producer and glucose as substrate (12). It is well-known that waste activated sludge is a more difficult substrate than glucose to be degraded and converted to hydrogen. Thus, the conversion ratio of sludge COD to H2 COD was lower than that of glucose. According to the above studies it can be seen that the sludge COD to H2 COD conversion ratio observed by this study was 12.7% greater than that by Cai et al. Further studies are being directed toward the effect of different alkaline reagents on hydrogen production and the economics in pilot-scale experiments.
Acknowledgments This work was financially supported by the Foundation of State Key Laboratory of Pollution Control and Resource Reuse (PCRRK09002) and the program for NCET in university (060373).
Supporting Information Available Tables S1 and S2 and Figures S1-S5. This material is available free of charge via the Internet at http://pubs.acs.org.
Literature Cited (1) Kapdan, I. K.; Kargi, F. Bio-hydrogen production from waste materials. Enzyme. Microb. Technol. 2006, 38 (5), 569–582. (2) Ueno, Y.; Fukui, H.; Goto, M. Operation of a two-stage fermentation process producing hydrogen and methane from organic waste. Environ. Sci. Technol. 2007, 41 (4), 1413–1419. (3) Valdez-Vazquez, I.; Rı´os-Leal, E.; Carmona-Martı´nez, A.; Mun ˜ozPa´ez, K. M.; Poggi-Varaldo, H. M. Improvement of biohydrogen production from solid wastes by intermittent venting and gas flushing of batch reactors headspace. Environ. Sci. Technol. 2006, 40 (10), 3409–3415. (4) Zhu, H. G.; Wayne, P.; Robert, B. Biohydrogen production by anaerobic co-digestion of municipal food waste and sewage sludges. Int. J. Hydrogen Energy 2008, 33 (14), 3651–3659. (5) Jeong, T. Y.; Cha, G. C.; Yoo, I. K.; Kim, D. J. Hydrogen production from waste activated sludge by using separation membrane acid fermentation reactor and photosynthetic reactor. Int. J. Hydrogen Energy 2007, 32 (5), 525–530. (6) Wang, C. C.; Chang, C. W.; Chu, C. P.; Lee, D. J.; Chang, B. V.; Liao, C. S. Producing hydrogen from wastewater sludge by Clostridium bifermentans. J. Biotechnol. 2003, 102 (1), 83–92. (7) Wang, C. C.; Chang, C. W.; Chu, C. P.; Lee, D. J.; Chang, B. V.; Liao, C. S.; Tay, J. H. Using filtrate of waste biosolids to effectively produce bio-hydrogen by anaerobic fermentation. Water Res. 2003, 37 (11), 2789–2793. (8) Ting, C H.; Lee, D. J. Production of hydrogen and methane from wastewater sludge using anaerobic fermentation. Int. J. Hydrogen Energy 2007, 32 (6), 677–682. (9) Cai, M. L.; Liu, J. X.; Wei, Y. S. Enhanced biohydrogen production from sewage sludge with alkaline pretreatment. Environ. Sci. Technol. 2004, 38 (11), 3195–3202. (10) Xiao, B. Y.; Liu, J. X. Biological hydrogen production from sterilized sewage sludge by anaerobic self-fermentation. J. Hazard. Mater. 2009, 168 (1), 163–167. (11) Massanet-Nicolau, J.; Dinsdale, R.; Guwy, A. Hydrogen production from sewage sludge using mixed microflora inoculums: effect of pH and enzymatic pretreatment. Bioresour. Technol. 2008, 99 (14), 6325–6331.
(12) Oh, S. E.; Van Ginkel, S.; Logan, B. E. The relative effectiveness of pH control and heat treatment for enhancing biohydrogen gas production. Environ. Sci. Technol. 2003, 37 (22), 5186–5190. (13) Diekert, G. CO2 reduction to acetate in anaerobic bacteria. FEMS Microbiol. Lett. 1990, 87, 391–395. (14) Kotsopoulos, T. A.; Fotidis, Ioannis, A.; Tsolakis, N.; Martzopoulos, G. G. Biohydrogen production from pig slurry in a CSTR reactor system with mixed cultures under hyper-thermophilic temperature (70°C). Biomass Bioenergy 2009, 33 (9), 1168–1174. (15) Ren, N. Q.; Chua, H.; Chan, S. Y.; Tsang, Y. F.; Wang, Y. J.; Sin., N. Assessing optimal fermentation type for bio-hydrogen production in continuous-flow acidogenic reactors. Bioresour. Technol. 2007, 98 (9), 1774–1780. (16) Li, J. Z.; Zheng, G. C.; He, J. G.; Chang, S.; Qin, Z. Hydrogenproducing capability of anaerobic activated sludge in three types of fermentations in a continuous stirred-tank reactor. Biotechnol. Adv. 2009, 27 (5), 573–577. (17) Fang, H. H. P.; Liu, H. Effect of pH on hydrogen production from glucose by a mixed culture. Bioresour. Technol. 2002, 82 (1), 87–93. (18) Khanal, S. K.; Chen, W. H.; Li, L.; Sung, S. Biological hydrogen production: effects of pH and intermediate products. Int. J. Hydrogen Energy 2004, 29 (11), 1123–1131. (19) Yuan, H.; Chen, Y.; Zhang, H.; Jiang, S.; Zhou, Q.; Gu, G. Improved bioproduction of short-chain fatty acids (SCFAs) from excess sludge under alkaline conditions. Environ. Sci. Technol. 2006, 40 (6), 2025–2029. (20) Owen, W. F.; Stuckey, D. C.; Healy, J. B., Jr.; Young, L. Y.; McCarty, P. L. Bioassay for monitoring biochemical methane potential and anaerobic toxicity. Water Res. 1979, 13 (6), 485–492. (21) Feng, L. Y.; Chen, Y. G.; Zheng, X. Enhancement of waste activated sludge protein conversion and volatile fatty acids accumulation during waste activated sludge anaerobic fermentation by carbohydrate substrate addition: the effect of pH. Environ. Sci. Technol. 2009, 43 (12), 4373–4380. (22) Grady, C. P. L.; Daigger, G. T.; Lim, H. C. Biological Wastewater Treatment; Marcel Dekker Inc.: New York, 1999. (23) Yakunin, A. F.; Hallenbeck, P. C. Purification and characterization of pyruvate oxidoreductase from the photosynthetic bacterium Rhodobacter capsulatus. Biochim. Biophys. Acta 1998, 1409 (1), 39–49. (24) O’Brien, W. E.; Ljungdahl, L. G. Fermentation of fructose and synthesis of acetate from carbon dioxide by clostridium formicoaceticum. J. Bacteriol. 1972, 109 (2), 626–632. (25) Moter, A.; Go¨bel, U. B. Fluorescence in situ hybridization (FISH) for direct visualization of microorganisms. J. Microbiol. Methods 2000, 41 (2), 85–112. (26) Franks, A. H.; Harmsen, H. J. M.; Raangs, G. C.; Jansen, G. J.; Schut, F.; Welling, G. W. Variations of bacterial populations in human feces measured by fluorescent in situ hybridization with group-specific 16S rRNA-targeted oligonucleotide probes. Appl. Environ. Microbiol. 1998, 64 (9), 3336–3345. (27) Ku ¨ sel, K.; Pinkart, H. C.; Drake, H. L.; Devereux, R. Acetogenic and sulfate-reducing bacteria inhabiting the rhizoplane and deep cortex cells of the sea grass Halodule wrightii. Appl. Environ. Microbiol. 1999, 65 (11), 5117–5123. (28) Bach, H. J.; Hartmann, A.; Schloter, M.; Munch, J. C. PCR primers and functional probes for amplification and detection of bacterial genus for extra cellular peptidases in single strain and soil. J. Microbiol. Methods 2001, 44 (2), 173–182. (29) Mu ¨ ller, V. Energy conservation in acetogenic bacteria. Appl. Environ. Microbiol. 2003, 69 (11), 6345–6353. (30) Karube, I.; Urano, N.; Matsunaga, T.; Suzuki, S. Hydrogen production from glucose by immobilized growing cells of clostridium butyricum. Appl. Microbiol. Biotechnol. 1982, 16, 5–9.
ES902958C
VOL. 44, NO. 9, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3323
