Inhibition of biohydrogen production by ammonia - Penn State ...
ARTICLE IN PRESS WAT E R R E S E A R C H
40 (2006) 1167 – 1172
Available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/watres
Inhibition of biohydrogen production by ammonia Michael B. Salernoa, Wooshin Parkb, Yi Zuoc, Bruce E. Loganc, a
Department of Chemical Engineering, Penn State University, University Park, PA 16801, USA Department of Environmental Science and Engineering, Gwangju Institute of Science and Technology (GIST), 1 Oryong-dong, Buk-gu, Gwangju 500-712, Republic of Korea c Department of Civil and Environmental Engineering, Penn State University, University Park, PA 16801, USA b
art i cle info
A B S T R A C T
Article history:
Ammonia inhibition of biohydrogen production was investigated in batch and continuous
Received 20 September 2005
flow reactors with glucose as a substrate. In batch tests, biohydrogen production rate was
Received in revised form
highly dependent on pH and ammonia (defined as the sum of NH3 of NH+4 species)
19 December 2005
concentrations above 2 g N/L. At pH ¼ 6.2, the maximum production decreased from 56 mL/
Accepted 16 January 2006
h at 2 g N/L to 16 mL/h at 10 g N/L. At pH ¼ 5.2, production decreased from 49 mL/h (2 g N/L) to 7 mL/h (16 g N/L). Hydrogen yield remained relatively constant in batch tests, varying
Keywords:
from 0.96 to 1.17 mol-H2/mol-glucose. In continuous flow tests, both hydrogen production
Biohydrogen production
rates and yields were adversely affected by ammonia. When the reactor (2.0 L) was first
Animal wastewater
acclimated under batch conditions to a low nitrogen concentration (o0.8 g N/L), H2
Ammonia inhibition
production and yields under continuous flow mode conditions were 170 mL/h and 1.9 molH2/mol-glucose, but decreased with increased ammonia concentrations up to 7.8 g N/L to 105 mL/h and 1.1 mol-H2/mol-glucose. There was no hydrogen production under continuous flow conditions if the reactor was initially operated under batch flow conditions at ammonia concentrations above 0.8 g N/L. It is concluded that the hydrogen production is possible at high concentrations (up to 7.8 g N/L) of ammonia in continuous flow systems as long as the reactor is initially acclimated to a lower ammonia concentration (o0.8 g N/L). & 2006 Elsevier Ltd. All rights reserved.
1.
Introduction
Bacterial hydrogen production via fermentation has the potential to treat wastewater streams while producing a clean, renewable energy carrier in the process (Benemann, 1996; Logan, 2004). Different wastewaters that have been considered for biohydrogen production via ‘‘dark’’ fermentation (i.e. non-photobiological-based processes) include sewage sludge (Cai et al., 2004), food wastes (Kim et al., 2004; Zhu et al., 2002; Han and Shin, 2004), and animal (Angenent et al., 2004) wastewater streams. Hydrogen can be produced using a variety of reactor and operating conditions, including batch (Logan et al., 2002; Van Ginkel et al., 2001) and continuous flow stirred tanks (Fang and Liu, 2002; Iyer et al., 2004), Corresponding author. Tel.: +814 863 7908; fax: +814 863 7304.
E-mail address: [email protected] (B.E. Logan). 0043-1354/$ - see front matter & 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.watres.2006.01.024
membrane (Oh et al., 2004), and upflow anaerobic sludge blanket (UASB) reactors (Chang and Lin, 2004). High hydrogen yields are needed to make the process economical (Logan, 2004). Obligate anaerobes such as Clostridia species are mainly responsible for fermenting sugars to hydrogen at high yields (Iyer et al., 2004), producing acetate, butyrate, and other fermentation end products as waste products (Gottschalk, 1986). For these bacteria, the maximum theoretical yield of hydrogen by known biochemical pathways is 4 mol-H2/mol-glucose, when acetate is produced according to C6 H12 O6 þ 2H2 O ! 4H2 þ 2CO2 þ 2C2 H4 O2 :
(1)
However, the need to regenerate NAD+ often results in the production of other volatile acids and solvents. For example,
ARTICLE IN PRESS 1168
WAT E R R E S E A R C H
40 (2006) 1167– 1172
bacteria produce only 2 mol-H2/mol-glucose when butyrate produced (Levin et al., 2004) via C6 H12 O6 ! 2H2 þ 2CO2 þ C4 H8 O2 :
(2)
Overall yields are lowered from these maximum values due to the production of cell biomass. Several factors must be considered for maximizing hydrogen production in fermentation systems. It is important to avoid the loss of hydrogen to hydrogen-consuming anaerobes, such as methanogens (Lay, 2001). This can be done by heat-treating the inoculum to select for spore-formers, such as Clostridia, for glucose-fed reactors or even heat-treating the wastewater to kill methanogens (Levin et al., 2004; Lay, 2001; Oh et al., 2003; Ting et al., 2004). Low pH can also be used to minimize the growth of methanogens (Phelps and Zeikus, 1984). It has been shown that longer hydraulic retention times (HRT) and lower chemical oxygen demand (COD) levels (factored together as COD loading rate) contribute to greater overall efficiency of hydrogen production in continuously fed reactors (Dabrock et al., 1992; Van Ginkel and Logan, 2005). Other factors such as inoculum, substrate, temperature, nitrogen sparging, and initial start up have all been examined by various researchers in an effort to optimize the production of hydrogen (Dabrock et al., 1992; Hawkes et al., 2002; Iyer et al., 2004). While animal wastewaters have been considered as potential wastewater sources for biohydrogen production, these wastewaters contain high (2 g/L) concentrations of ammonia (Ceccherini et al., 1998; Cruz et al., 2000). Ammonia is known to inhibit methanogenesis (Lay et al., 1998; Borja et al., 1996; Bhattacharya and Parkin, 1989; Koster and Koomen, 1988), but the effect of ammonia on hydrogen production has not been well examined. Zhu et al. (2001) found that NH+4 concentrations of o0.14 g/L did not adversely affect hydrogen production by Clostridium butyricum. Lay et al. (1998) determined that it was the NH3 species, as opposed to the NH+4 ion, that was responsible for increased lag times in methanogenesis, but that NH+4 ions had a larger effect than NH3 on the overall methanogenic activity. To investigate the inhibition to hydrogen-producing bacteria, various concentrations of ammonium chloride were added to a feed solution containing glucose as the energy source under both batch and continuous flow conditions. The effect of ammonia was examined in terms of hydrogen production rates and overall hydrogen yields. Ammonia is defined here as the sum of both NH3+NH+4 forms, with specific chemical formulas used to denote specific ammonia species.
2.
Methods
2.1.
Batch tests
(3.76 g/L ¼ 4 g/L COD) in a mineral salt medium (MSM) (Oh et al., 2003), and buffered to a pH of 5.2 or 6.2, using 0.07 M of 2-(N-mopholino) ethanesulfonic acid monohydrate (MES) (J.T. Baker, Phillipsburg, NJ). The MSM consisted of the following nutrients (listed per liter of feed) for all experiments: 360 mg MgSO4 7H2O, 50 mg FeCl3, 48 mg NiCl2 6H2O, 66 mg CaCl2 2H2O, 23 mg ZnCl2, 21 mg CoCl 6H2O, 10 mg CuCl2 2H2O, 30 mg MnCl2 4H2O, 175 mg K2HPO4, and 175 mg KH2PO4. NH4Cl was separately added to each reactor at concentrations of 0.5, 2, 5, 7, and 10 g N/L. Batch tests were conducted in duplicate using 250 mL of liquid in 300 mL bottles stirred at 300 rpm in a constant temperature room (3071 1C). All bottles were initially flushed with nitrogen gas before being sealed with rubber stoppers. Gas production was measured continuously using a respirometer system (Challenge Environmental Systems AER-200, Fayetteville, Arkansas, USA), while the composition of headspace gas was periodically sampled (10–20 sample points per experiment). The biohydrogen production was calculated from the composition and the volume of biogas produced, as previously described (Logan et al., 2002).
2.2.
Continuous flow tests were conducted using a fermentor (New Brunswick Scientific BioFlo 110) filled to 2.0 L at 30 1C. The reactor was sparged with nitrogen gas, and inoculated with heat-treated agricultural soil (Logan et al., 2002). MSM and glucose (4 g/L COD) were added to the reactor. L-cysteine (0.5 g/L) was added to scavenge dissolved oxygen and lower the initial redox potential. Bottles of feed (10 L, autoclaved) were sparged with N2 for at least 30 min before use, and then continuously sparged and mixed with a magnetic stirrer during tests. NH4Cl was added in concentrations of 0.8, 1.6, 3.9, 5.3, and 7.8 g N/L. For each experiment, the reactor was initially operated for 3–4 days in batch mode (with L-cysteine), and then switched to continuous-flow conditions at a HRT of 10 h (no L-cysteine). Redox potential and pH were monitored during start up to ensure that the redox potential remained negative and that pH was constant at 5.5. Small (2 mL total per reactor) additions of 2 M KOH were needed to maintain the pH during the initial batch mode. During continuous flow operation the pH was automatically maintained at a pH of 5.5 using KOH (2 M). The reactor was determined to be operating under steady state conditions on the basis of gas production rate, which typically required 6–8 days of continuous flow mode operation. Total gas production was measured using a bubble trap system consisting of a graduated pipette with a side inlet. The time for 2–10 mL of gas production, depending on flow rate, was recorded in triplicate.
2.3. Dewatered and thickened sludge was obtained from the Pennsylvania State University Wastewater Treatment Plant in State College, PA, and heat treated as previously described (Oh et al., 2003). The sample was either used within one day or stored in a sealed container at room temperature (24 1C) for no more than 1 month. The reactors were inoculated with the heat-treated sludge (4 g/L) and grown on glucose
Continuous-flow reactor experiments
Analysis
Hydrogen gas was measured using a GasTightTM syringe (SGE, 0.5 mL injection volume for continuous tests, 0.25 mL for batch tests) and analyzed using a gas chromatograph (GC; Model 310, SRI Instruments, Torrence, CA) equipped with a thermal conductivity detector and a molecular sieve column (Alltech Molesieve 5A 80/100) with nitrogen as the carrier gas.
ARTICLE IN PRESS
Results and discussion
3.1.
Batch reactors
2
250 pH 5.2 pH 6.2 200
Yield at pH 5.2
1.5
Yield at pH 6.2 150 1 100 0.5 50
0
0 0.5
The rate of total biogas production in batch tests decreased, and the apparent lag increased, with increasing ammonia concentrations (Fig. 1). However, total hydrogen gas production was not significantly affected (p40:18, ANOVA), and averaged 12975 mL (795% confidence interval; Fig. 2) for both
Yield (mol/mol)
Nitrogen and carbon dioxide were sampled in the same manner and analyzed using a second GC of the same model but containing a different column (Alltech Porapak Q 80/100) with helium as the carrier gas. Pure gases were used as calibration standards. Organic acids, solvents, and alcohols were measured in the aqueous phase using a GC (Agilent GC-FID). Liquid samples were filtered using 0.2 mm pore-diameter filters (Whatman) and diluted before analysis with 50% formic acid. Samples were stored sealed in a refrigerator (4 1C) prior to analysis. Residual glucose was measured (in duplicate) with the phenol and sulfuric acid assay using standards in a range of 10–60 mg/L.
3.
1169
4 0 (200 6) 116 7 – 117 2
Hydrogen Production (mL)
WAT E R R E S E A R C H
2 5 7 Ammonia (g N/L)
10
Fig. 2 – Total hydrogen production (bars) and yield (symbols) as a function of ammonia concentration at pH ¼ 5.2 and 6.2 in batch experiments. Error bars show 95% confidence intervals (n ¼ 2).
50 pH = 5.2 Time for 100 mL (hr)
40
Gas Production (mL)
300 250 200 150
20
10
0
50
0
20
(A)
40 Time (hr)
60
80
300 Gas Production (mL)
30
100
0
0.5
2
5 7 Ammonia (g N/L)
10
Fig. 3 – Time for 100 mL of gas production at different ammonia concentrations at pH ¼ 5.2 or 6.2 in batch experiments. Error bars show 95% confidence intervals (n ¼ 2).
250 200 0.5 g N/L 150
2 g N/L
100
5 g N/L 7 g N/L
50
10 g N/L 0 0 (B)
pH = 6.2
20
40 Time (hr)
60
80
Fig. 1 – Total biogas production over time at (A) pH ¼ 5.2 and (B) pH ¼ 6.2 in batch experiments. Total gas production was not affected by high levels of ammonia, but the lag time increased and production rate decreased. Each data series shown in the figure is the average of two runs.
pH values tested (5.2 and 6.2). The change in the pH during a batch cycle was affected by organic loading, but decreased only by 3.6–9.6% and therefore was not expected to effect hydrogen production. The hydrogen yield was also not affected by ammonia concentration in batch tests (p40:32, ANOVA), and averaged 1.070.04 mol-H2/mol-glucose at ammonia concentrations of up to 10 g/L (Fig. 2). The bacteria continued to produce hydrogen until the all the glucose in the system was consumed, which resulted in approximately the same amount of total hydrogen being produced despite the different levels of ammonia. The lag time, defined arbitrarily here as the time for 100 mL of gas production, increased with ammonia concentration (Fig. 3) from 12 h (0.5 g N/L) to 35 h (10 g N/L), and the maximum gas production rate decreased (Fig. 4). The maximum production rate (the highest rate of gas
ARTICLE IN PRESS WAT E R R E S E A R C H
40 (2006) 1167– 1172
production, as determined by differentiating the gas production data) at pH 6.2 decreased from 56 mL/h (at 2 g N/L) to 16 mL/h (at 10 g N/L). At pH 5.2, it decreased from 49 to 7 mL/h, at 2 and 10 g N/L, respectively. As the specific form of the ammonia (NH3 or NH+4 ) is highly dependent on pH, it is important to note the difference in reactor performance under different pH values in terms of ammonia species. At the conditions tested here, most of the ammonia is present as NH+4 as the NH3/NH+4 ratio varied from 1.3 104 to 1.3 103 mol/mol. At a given total ammonia/ ammonium concentration, the concentration of NH3 molecules is an order of magnitude greater at a pH of 6.2 than at 5.2. For example, in tests at an ammonia concentration of 10 g/L, the NH3 concentration was 13 mg N/L at pH ¼ 6.2, and 1.3 mg/L at a pH of 5.2. The reactors showed longer lag times and slower production rates at a pH of 5.2, suggesting that the inhibitory mechanism is different than that expected on the basis of just NH3 concentrations.
3.2.
Continuous-flow reactor experiments
pH 5.2
60
350 Total Gas H2
300 250 200 150 100 50 0 0
2
4 6 Ammonia (g N/L)
8
Fig. 5 – Total gas and biohydrogen gas production in continuous-flow experiments as a function of total ammonia concentration (at pH 5.5). The changes in both total gas and H2 production were statistically significant (for H2: slope ¼ 9.05, R2 ¼ 0:97, po0:02). Gas production was completely inhibited when the reactor was started at or above 1.6 g N/L, but not at 3.8, 5.3, or 7.8 g N/L when initially started at 0.8 g N/L.
pH 6.2 2.5
40
2.0 Yield (mol/mol)
Max.Gas Prod. Rate (mL/hr)
Acclimation to a low nitrogen concentration prior to the introduction of a high concentration of NH3/NH+4 (40.8 g N/L) was critical to successful production of hydrogen. When the initial concentration of NH3/NH+4 in the reactor operated under batch conditions was 1.6 g N/L or higher, total gas production and hydrogen production were all negligible, and the reactor inoculum washed out when the reactor was switched to continuous mode operation (data not shown). Therefore, in subsequent tests the reactor was first run in batch mode at 0.8 g N/L, and then switched to a feed containing the same or higher concentration of NH3/NH+4 . Ammonia toxicity has been shown for methanogens in batch experiments (Lay et al., 1998), where concentrations above 4 g N/L showed reduced methane production and increased lag
times. Toxicity has also been shown in UASB reactors (Borja et al., 1996), where digestion was inhibited above 5 g N/L. Hydrogen production in a 2 L chemostat decreased by 15% (from 170 to 145 mL/h) when the ammonia concentration was increased from 0.8 to 3.9 g N/L in the feed after reactor acclimation to 0.8 g N/L (Fig. 5). The hydrogen yield decreased from 1.9 to 1.6 mol-H2/mol-glucose when the ammonia concentration was increased from 0.8 to 3.9 g N/L. During this feed transition, the total gas production rate decreased (259–230 mL/h) and hydrogen production decreased (170–145 mL/h). The glucose removed increased from 86% to 91% (this change was not statistically significant, p40:1). Hydrogen production was further inhibited, while yield was
Gas Production (mL / hr)
1170
20
0 0.5
2
5
7
10
Ammonia (g N/L)
1.5 1.0 0.5 0.0 0
Fig. 4 – Maximum gas production rate in batch experiments as a function of ammonia concentration. Lines show linear regression of data at the different pHs (the dashed line is for pH 5.2, and the solid line is for pH 6.2). Error bars show 95% confidence intervals.
2
6 4 Ammonia (g N/L)
8
Fig. 6 – Yield of hydrogen production (in mol H2 gas per mol of glucose consumed) in continuous-flow reactor experiments. The decrease is statistically significant (slope ¼ 0.12, R2 ¼ 0:98, po0:01).
ARTICLE IN PRESS WAT E R R E S E A R C H
1171
4 0 (200 6) 116 7 – 117 2
Table 1 – Summary of results from a 2 L continuous-flow reactor experiments, including hydrogen production and hydrogen yield at various levels of ammonia concentration Ammonia (g/L)
Total gas (mL/h)
H2 (mL/h)
H2
Yield (mol/mol)
Glucose removed (%)
0.8 3.9 5.3 7.8
259 230 216 156
170 145 138 105
66 63 64 67
1.9 1.6 1.5 1.1
86 91 92 92
Table 2 – Residual acids, solvents, and alcohols in solution in chemostat experiments under steady state conditions Ammonia (g/L) 0.8 3.9 5.3 7.8 1.6
Acetone (ppm)
Methanol (ppm)
Ethanol (ppm)
n-Butanol (ppm)
Acetate (ppm)
Propionate (ppm)
Butyrate (ppm)
— — — 11.471 8.771.7
3.270.3 — — — —
2376 2273 29.779.6 480734 4307110
2.170.3 6778 38710 — —
510730 550770 470710 52071 110710
4.370.3 1.770.2 — — —
1240790 8707110 12707200 37875 2571.0
Dashes indicate concentrations below the detection limit. n-Propanol was below detectable limits for all samples. (Errors represent 95% confidence intervals, with n ¼ 3 for each analysis).
decreased, when the reactor was transitioned to a feed containing 5.3 or 7.8 g N/L. The hydrogen yield consistently decreased with ammonia concentration from 1.9 mol-H2/mol-glucose at 0.8 g N/L, to 1.1 mol/mol at 7.8 g N/L (Fig. 6). The concentration of hydrogen gas remained constant at 6571.6% for the different ammonia feed concentrations (Table 1). Thus, the decrease in hydrogen yield was due to reduced gas production, with the hydrogen production rate decreasing from 170 mL/h at 0.8 g N/L, to 105 mL/h at 7.8 g N/L. The glucose removal rate increased slightly, from 86% to 92%, as the ammonia concentration increased (Table 1), suggesting that the shift to end products other than hydrogen resulted in increased substrate utilization. The liquid products in the reactor were analyzed to determine how the changes in the influent ammonia concentrations affected the liquid products of glucose fermentation. The main soluble products were acetic and butyric acids for conditions that resulted in successful hydrogen production (Table 2). Addition of ammonia also increased the production of n-butanol. The levels of acetone, methanol, and propionate were below detectable limits for all experiments. The 1.6 g N/L of ammonia run shown at the bottom of Table 2 is the data from when the reactor was started up at 1.6 g N/L, whereas for all other runs the reactor was started up at 0.8 g N/L and then switched to the appropriate value. When started at 1.6 g N/L, the reactor produced more than 400 ppm ethanol, but little acetate or butyrate (Table 2). The high production of ethanol in the reactor started up at an ammonia concentration of 1.6 g N/L indicates that either spore-forming hydrogen producers were not the dominant community, or that the bacteria began utilizing an ethanolproducing metabolic pathway. This pathway, via acetalde-
hyde as an intermediate, does not produce H2 like the acetate and butyrate pathways do. The low levels of acetate and butyrate show that these hydrogen-producing pathways were not being utilized. The high levels of acetate and butyrate in the reactor run at 0.8 g N/L are indicative of the high hydrogen production and yield observed. The ratio of acetate to butyrate increased when 3.9 g N/L was introduced to the reactor. However, given the values shown in Table 2, and assuming 4 mol-H2/mol-glucose via the acetate pathway versus 2 mol-H2/mol-glucose via the butyrate pathway, we would predict 13% less hydrogen production from the high ammonia reactor than the low ammonia reactor. This is very close to the measured decrease in hydrogen production rate of 15%. A similar analysis performed for the 7.8 g N/L conditions would predict a 37% decrease in hydrogen production, which is similar to the 38% observed decrease.
4.
Conclusions
These results show that hydrogen production can be inhibited by high concentrations of ammonia, such as those present in animal wastewaters. Under fed batch conditions, the overall yield of hydrogen is not affected but the rate of hydrogen production is decreased. When a continuous flow reactor was initially operated (in batch mode) at a high (X1.6 g N/L) concentration of ammonia, hydrogen production failed. However, if the reactor was first started up at a lower ammonia concentration (0.8 g N/L), hydrogen could be produced but the overall yield was reduced from 1.9 to 1.1 mol-H2/mol-glucose for ammonia concentrations of 0.8 to 7.8 g N/L. The concentration of hydrogen in the gas remained constant under all continuous flow conditions (6571.6%), and
ARTICLE IN PRESS 1172
WAT E R R E S E A R C H
40 (2006) 1167– 1172
thus the decrease in hydrogen yields was due to reduced gas production rates. It is therefore possible to produce hydrogen under high (up to 7.8 g N/L) concentrations of ammonia but the ammonia can reduce the overall rate or conversion of glucose to hydrogen.
Acknowledgements This research was supported by Grant 68-3A75-3-150 from the Natural Resources Conservation Service of the United States Department of Agriculture. R E F E R E N C E S
Angenent, L.T., Karim, K., Al-Dahhan, M.H., Wrenn, B.A., Domiguez-Espinosa, R., 2004. Production of bioenergy and biochemicals from industrial and agricultural wastewater. Trends Biotechnol. 22 (9), 477–485. Benemann, J., 1996. Hydrogen biotechnology: progress and prospects. Nat. Biotechnol. 14, 1101–1103. Bhattacharya, S.K., Parkin, G.F., 1989. The effect of ammonia on methane fermentation processes. J. Water Pollut. Control Fed. 61 (1), 55–59. Borja, R., Sanchez, E., Weiland, P., 1996. Influence of ammonia concentration on thermophilic anaerobic digestion of cattle manure in upflow anaerobic sludge blanket (UASB) reactors. Process Biochem. 31 (5), 477. Cai, M., Liu, J., Wei, Y., 2004. Enhanced biohydrogen production from sewage sludge with alkaline pretreatment. Environ. Sci. Technol. 38, 3195–3202. Ceccherini, M.T., Castaldini, M., Piovanelli, C., Hastings, R.C., McCarthy, A.J., Bazzicalupo, M., Miclaus, N., 1998. Effects of swine manure fertilization on autotrophic ammonia oxidizing bacteria in soil. Appl. Soil Ecolog. 7, 149–157. Chang, F.Y., Lin, C.Y., 2004. Biohydrogen production using an upflow anaerobic sludge blanket reactor. Int. J. Hydrogen Energy 29, 33–39. Cruz, H.J., Freitas, C.M., Alves, P.M., Moreira, J.L., Carrondo, M.J.T., 2000. Effects of ammonia and lactate on growth, metabolism, and productivity of BHK cells. Enzyme Microb. Technol. 27, 43–52. Dabrock, B., Bahl, H., Gottschalk, G., 1992. Parameters affecting solvent production by Clostridium pasteurianum. Appl. Environ. Microbiol. 58, 1233–1239. Fang, H.H.P., Liu, H., 2002. Effect of pH on hydrogen production from glucose by a mixed culture. Bioresource Technol. 82, 87–93. Gottschalk, G., 1986. Bacterial Metabolism, second ed. Springer, New York. Han, S.K., Shin, H.S., 2004. Biohydrogen production by anaerobic fermentation of food waste. Int. J. Hydrogen Energy 29, 569–577.
Hawkes, F.R., Dinsdale, R., Hawkes, D.L., Hussy, I., 2002. Sustainable fermentative hydrogen production: challenges for process optimization. Int. J. Hydrogen Energy 27, 1339–1347. Iyer, P., Bruns, M.A., Zhang, H., Van Ginkel, S., Logan, B.E., 2004. Hydrogen gas production in a continuous flow bioreactor using heat-treated soil inocula. Appl. Microbiol. Biotechnol. 89 (1), 119–127. Kim, S.H., Han, S.K., Shin, H.S., 2004. Feasibility of biohydrogen production by anaerobic co-digestion of food waste and sewage sludge. Int. J. Hydrogen Energy 29, 1607–1616. Koster, I.W., Koomen, E., 1988. Ammonia inhibition of the maximum growth rate (mm) of hydrogenotrophic methanogens at various pH-levels and temperatures. Appl. Microbiol. Biotechnol. 28, 500–505. Lay, J.J., 2001. Biohydrogen generation by mesophilic anaerobic fermentation of microcrystalline cellulose. Biotechnol. Bioeng. 74, 280–287. Lay, J.J., Li, Y.Y., Noike, T., 1998. Influence of pH and ammonia concentration on the methane production in high-solids digestion processes. Water Environ. Res. 70 (5), 1075. Levin, D.B., Pitt, L., Love, M., 2004. Biohydrogen production: prospects and limitations to practical application. Int. J. Hydrogen Energy 29, 173–185. Logan, B.E., 2004. Feature article: biologically extracting energy from wastewater: biohydrogen production and microbial fuel cells. Environ. Sci. Technol. 38 (9), 160A–167A. Logan, B.E., Oh, S.E., Kim, I.S., Van Ginkel, S., 2002. Biological hydrogen production measured in batch anaerobic respirometers. Environ. Sci. Technol. 36 (11), 2530–2535. Oh, S.E., Van Ginkel, S., Logan, B.E., 2003. The relative effectiveness of pH control and heat treatment for enhancing biohydrogen gas production. Environ. Sci. Technol. 37, 5186–5190. Oh, S.E., Iyer, P., Bruns, M.A., Logan, B.E., 2004. Biological hydrogen production using a membrane bioreactor. Biotechnol. Bioeng. 87 (1), 119–127. Phelps, T.J., Zeikus, J.G., 1984. Influence of pH on terminal carbon metabolism in anoxic sediments from a Midly Acidic Lake. Appl. Environ. Microbiol. 48 (6), 1088–1095. Ting, C.H., Lin, K.R., Lee, D.J., Tay, J.H., 2004. Production of hydrogen and methane from wastewater sludge using anaerobic fermentation. Water Sci. Technol. 50 (9), 223–228. Van Ginkel, S., Sung, S., Lay, J.J., 2001. Biohydrogen production as a function of pH and substrate concentration. Environ. Sci. Technol. 35, 4726–4730. Van Ginkel, S.W., Logan, B.E., 2005. Increased biological hydrogen production with reduced organic loading. Water Res. 39, 3819–3826. Zhu, H., Wakayama, T., Asada, Y., Miyake, J., 2001. Hydrogen production by four cultures with participation by anoxygenic phototrophic bacterium and anaerobic bacterium in the presence of NH+4 . Int. J. Hydrogen Energy 26, 1149–1154. Zhu, H., Ueda, S., Asada, Y., Miyake, J., 2002. Hydrogen production as a novel process of wastewater treatment—studies on tofu wastewater with entrapped R. sphaeroides and mutagenesis. Int. J. Hydrogen Energy 27, 1349–1357.
40 (2006) 1167 – 1172
Available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/watres
Inhibition of biohydrogen production by ammonia Michael B. Salernoa, Wooshin Parkb, Yi Zuoc, Bruce E. Loganc, a
Department of Chemical Engineering, Penn State University, University Park, PA 16801, USA Department of Environmental Science and Engineering, Gwangju Institute of Science and Technology (GIST), 1 Oryong-dong, Buk-gu, Gwangju 500-712, Republic of Korea c Department of Civil and Environmental Engineering, Penn State University, University Park, PA 16801, USA b
art i cle info
A B S T R A C T
Article history:
Ammonia inhibition of biohydrogen production was investigated in batch and continuous
Received 20 September 2005
flow reactors with glucose as a substrate. In batch tests, biohydrogen production rate was
Received in revised form
highly dependent on pH and ammonia (defined as the sum of NH3 of NH+4 species)
19 December 2005
concentrations above 2 g N/L. At pH ¼ 6.2, the maximum production decreased from 56 mL/
Accepted 16 January 2006
h at 2 g N/L to 16 mL/h at 10 g N/L. At pH ¼ 5.2, production decreased from 49 mL/h (2 g N/L) to 7 mL/h (16 g N/L). Hydrogen yield remained relatively constant in batch tests, varying
Keywords:
from 0.96 to 1.17 mol-H2/mol-glucose. In continuous flow tests, both hydrogen production
Biohydrogen production
rates and yields were adversely affected by ammonia. When the reactor (2.0 L) was first
Animal wastewater
acclimated under batch conditions to a low nitrogen concentration (o0.8 g N/L), H2
Ammonia inhibition
production and yields under continuous flow mode conditions were 170 mL/h and 1.9 molH2/mol-glucose, but decreased with increased ammonia concentrations up to 7.8 g N/L to 105 mL/h and 1.1 mol-H2/mol-glucose. There was no hydrogen production under continuous flow conditions if the reactor was initially operated under batch flow conditions at ammonia concentrations above 0.8 g N/L. It is concluded that the hydrogen production is possible at high concentrations (up to 7.8 g N/L) of ammonia in continuous flow systems as long as the reactor is initially acclimated to a lower ammonia concentration (o0.8 g N/L). & 2006 Elsevier Ltd. All rights reserved.
1.
Introduction
Bacterial hydrogen production via fermentation has the potential to treat wastewater streams while producing a clean, renewable energy carrier in the process (Benemann, 1996; Logan, 2004). Different wastewaters that have been considered for biohydrogen production via ‘‘dark’’ fermentation (i.e. non-photobiological-based processes) include sewage sludge (Cai et al., 2004), food wastes (Kim et al., 2004; Zhu et al., 2002; Han and Shin, 2004), and animal (Angenent et al., 2004) wastewater streams. Hydrogen can be produced using a variety of reactor and operating conditions, including batch (Logan et al., 2002; Van Ginkel et al., 2001) and continuous flow stirred tanks (Fang and Liu, 2002; Iyer et al., 2004), Corresponding author. Tel.: +814 863 7908; fax: +814 863 7304.
E-mail address: [email protected] (B.E. Logan). 0043-1354/$ - see front matter & 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.watres.2006.01.024
membrane (Oh et al., 2004), and upflow anaerobic sludge blanket (UASB) reactors (Chang and Lin, 2004). High hydrogen yields are needed to make the process economical (Logan, 2004). Obligate anaerobes such as Clostridia species are mainly responsible for fermenting sugars to hydrogen at high yields (Iyer et al., 2004), producing acetate, butyrate, and other fermentation end products as waste products (Gottschalk, 1986). For these bacteria, the maximum theoretical yield of hydrogen by known biochemical pathways is 4 mol-H2/mol-glucose, when acetate is produced according to C6 H12 O6 þ 2H2 O ! 4H2 þ 2CO2 þ 2C2 H4 O2 :
(1)
However, the need to regenerate NAD+ often results in the production of other volatile acids and solvents. For example,
ARTICLE IN PRESS 1168
WAT E R R E S E A R C H
40 (2006) 1167– 1172
bacteria produce only 2 mol-H2/mol-glucose when butyrate produced (Levin et al., 2004) via C6 H12 O6 ! 2H2 þ 2CO2 þ C4 H8 O2 :
(2)
Overall yields are lowered from these maximum values due to the production of cell biomass. Several factors must be considered for maximizing hydrogen production in fermentation systems. It is important to avoid the loss of hydrogen to hydrogen-consuming anaerobes, such as methanogens (Lay, 2001). This can be done by heat-treating the inoculum to select for spore-formers, such as Clostridia, for glucose-fed reactors or even heat-treating the wastewater to kill methanogens (Levin et al., 2004; Lay, 2001; Oh et al., 2003; Ting et al., 2004). Low pH can also be used to minimize the growth of methanogens (Phelps and Zeikus, 1984). It has been shown that longer hydraulic retention times (HRT) and lower chemical oxygen demand (COD) levels (factored together as COD loading rate) contribute to greater overall efficiency of hydrogen production in continuously fed reactors (Dabrock et al., 1992; Van Ginkel and Logan, 2005). Other factors such as inoculum, substrate, temperature, nitrogen sparging, and initial start up have all been examined by various researchers in an effort to optimize the production of hydrogen (Dabrock et al., 1992; Hawkes et al., 2002; Iyer et al., 2004). While animal wastewaters have been considered as potential wastewater sources for biohydrogen production, these wastewaters contain high (2 g/L) concentrations of ammonia (Ceccherini et al., 1998; Cruz et al., 2000). Ammonia is known to inhibit methanogenesis (Lay et al., 1998; Borja et al., 1996; Bhattacharya and Parkin, 1989; Koster and Koomen, 1988), but the effect of ammonia on hydrogen production has not been well examined. Zhu et al. (2001) found that NH+4 concentrations of o0.14 g/L did not adversely affect hydrogen production by Clostridium butyricum. Lay et al. (1998) determined that it was the NH3 species, as opposed to the NH+4 ion, that was responsible for increased lag times in methanogenesis, but that NH+4 ions had a larger effect than NH3 on the overall methanogenic activity. To investigate the inhibition to hydrogen-producing bacteria, various concentrations of ammonium chloride were added to a feed solution containing glucose as the energy source under both batch and continuous flow conditions. The effect of ammonia was examined in terms of hydrogen production rates and overall hydrogen yields. Ammonia is defined here as the sum of both NH3+NH+4 forms, with specific chemical formulas used to denote specific ammonia species.
2.
Methods
2.1.
Batch tests
(3.76 g/L ¼ 4 g/L COD) in a mineral salt medium (MSM) (Oh et al., 2003), and buffered to a pH of 5.2 or 6.2, using 0.07 M of 2-(N-mopholino) ethanesulfonic acid monohydrate (MES) (J.T. Baker, Phillipsburg, NJ). The MSM consisted of the following nutrients (listed per liter of feed) for all experiments: 360 mg MgSO4 7H2O, 50 mg FeCl3, 48 mg NiCl2 6H2O, 66 mg CaCl2 2H2O, 23 mg ZnCl2, 21 mg CoCl 6H2O, 10 mg CuCl2 2H2O, 30 mg MnCl2 4H2O, 175 mg K2HPO4, and 175 mg KH2PO4. NH4Cl was separately added to each reactor at concentrations of 0.5, 2, 5, 7, and 10 g N/L. Batch tests were conducted in duplicate using 250 mL of liquid in 300 mL bottles stirred at 300 rpm in a constant temperature room (3071 1C). All bottles were initially flushed with nitrogen gas before being sealed with rubber stoppers. Gas production was measured continuously using a respirometer system (Challenge Environmental Systems AER-200, Fayetteville, Arkansas, USA), while the composition of headspace gas was periodically sampled (10–20 sample points per experiment). The biohydrogen production was calculated from the composition and the volume of biogas produced, as previously described (Logan et al., 2002).
2.2.
Continuous flow tests were conducted using a fermentor (New Brunswick Scientific BioFlo 110) filled to 2.0 L at 30 1C. The reactor was sparged with nitrogen gas, and inoculated with heat-treated agricultural soil (Logan et al., 2002). MSM and glucose (4 g/L COD) were added to the reactor. L-cysteine (0.5 g/L) was added to scavenge dissolved oxygen and lower the initial redox potential. Bottles of feed (10 L, autoclaved) were sparged with N2 for at least 30 min before use, and then continuously sparged and mixed with a magnetic stirrer during tests. NH4Cl was added in concentrations of 0.8, 1.6, 3.9, 5.3, and 7.8 g N/L. For each experiment, the reactor was initially operated for 3–4 days in batch mode (with L-cysteine), and then switched to continuous-flow conditions at a HRT of 10 h (no L-cysteine). Redox potential and pH were monitored during start up to ensure that the redox potential remained negative and that pH was constant at 5.5. Small (2 mL total per reactor) additions of 2 M KOH were needed to maintain the pH during the initial batch mode. During continuous flow operation the pH was automatically maintained at a pH of 5.5 using KOH (2 M). The reactor was determined to be operating under steady state conditions on the basis of gas production rate, which typically required 6–8 days of continuous flow mode operation. Total gas production was measured using a bubble trap system consisting of a graduated pipette with a side inlet. The time for 2–10 mL of gas production, depending on flow rate, was recorded in triplicate.
2.3. Dewatered and thickened sludge was obtained from the Pennsylvania State University Wastewater Treatment Plant in State College, PA, and heat treated as previously described (Oh et al., 2003). The sample was either used within one day or stored in a sealed container at room temperature (24 1C) for no more than 1 month. The reactors were inoculated with the heat-treated sludge (4 g/L) and grown on glucose
Continuous-flow reactor experiments
Analysis
Hydrogen gas was measured using a GasTightTM syringe (SGE, 0.5 mL injection volume for continuous tests, 0.25 mL for batch tests) and analyzed using a gas chromatograph (GC; Model 310, SRI Instruments, Torrence, CA) equipped with a thermal conductivity detector and a molecular sieve column (Alltech Molesieve 5A 80/100) with nitrogen as the carrier gas.
ARTICLE IN PRESS
Results and discussion
3.1.
Batch reactors
2
250 pH 5.2 pH 6.2 200
Yield at pH 5.2
1.5
Yield at pH 6.2 150 1 100 0.5 50
0
0 0.5
The rate of total biogas production in batch tests decreased, and the apparent lag increased, with increasing ammonia concentrations (Fig. 1). However, total hydrogen gas production was not significantly affected (p40:18, ANOVA), and averaged 12975 mL (795% confidence interval; Fig. 2) for both
Yield (mol/mol)
Nitrogen and carbon dioxide were sampled in the same manner and analyzed using a second GC of the same model but containing a different column (Alltech Porapak Q 80/100) with helium as the carrier gas. Pure gases were used as calibration standards. Organic acids, solvents, and alcohols were measured in the aqueous phase using a GC (Agilent GC-FID). Liquid samples were filtered using 0.2 mm pore-diameter filters (Whatman) and diluted before analysis with 50% formic acid. Samples were stored sealed in a refrigerator (4 1C) prior to analysis. Residual glucose was measured (in duplicate) with the phenol and sulfuric acid assay using standards in a range of 10–60 mg/L.
3.
1169
4 0 (200 6) 116 7 – 117 2
Hydrogen Production (mL)
WAT E R R E S E A R C H
2 5 7 Ammonia (g N/L)
10
Fig. 2 – Total hydrogen production (bars) and yield (symbols) as a function of ammonia concentration at pH ¼ 5.2 and 6.2 in batch experiments. Error bars show 95% confidence intervals (n ¼ 2).
50 pH = 5.2 Time for 100 mL (hr)
40
Gas Production (mL)
300 250 200 150
20
10
0
50
0
20
(A)
40 Time (hr)
60
80
300 Gas Production (mL)
30
100
0
0.5
2
5 7 Ammonia (g N/L)
10
Fig. 3 – Time for 100 mL of gas production at different ammonia concentrations at pH ¼ 5.2 or 6.2 in batch experiments. Error bars show 95% confidence intervals (n ¼ 2).
250 200 0.5 g N/L 150
2 g N/L
100
5 g N/L 7 g N/L
50
10 g N/L 0 0 (B)
pH = 6.2
20
40 Time (hr)
60
80
Fig. 1 – Total biogas production over time at (A) pH ¼ 5.2 and (B) pH ¼ 6.2 in batch experiments. Total gas production was not affected by high levels of ammonia, but the lag time increased and production rate decreased. Each data series shown in the figure is the average of two runs.
pH values tested (5.2 and 6.2). The change in the pH during a batch cycle was affected by organic loading, but decreased only by 3.6–9.6% and therefore was not expected to effect hydrogen production. The hydrogen yield was also not affected by ammonia concentration in batch tests (p40:32, ANOVA), and averaged 1.070.04 mol-H2/mol-glucose at ammonia concentrations of up to 10 g/L (Fig. 2). The bacteria continued to produce hydrogen until the all the glucose in the system was consumed, which resulted in approximately the same amount of total hydrogen being produced despite the different levels of ammonia. The lag time, defined arbitrarily here as the time for 100 mL of gas production, increased with ammonia concentration (Fig. 3) from 12 h (0.5 g N/L) to 35 h (10 g N/L), and the maximum gas production rate decreased (Fig. 4). The maximum production rate (the highest rate of gas
ARTICLE IN PRESS WAT E R R E S E A R C H
40 (2006) 1167– 1172
production, as determined by differentiating the gas production data) at pH 6.2 decreased from 56 mL/h (at 2 g N/L) to 16 mL/h (at 10 g N/L). At pH 5.2, it decreased from 49 to 7 mL/h, at 2 and 10 g N/L, respectively. As the specific form of the ammonia (NH3 or NH+4 ) is highly dependent on pH, it is important to note the difference in reactor performance under different pH values in terms of ammonia species. At the conditions tested here, most of the ammonia is present as NH+4 as the NH3/NH+4 ratio varied from 1.3 104 to 1.3 103 mol/mol. At a given total ammonia/ ammonium concentration, the concentration of NH3 molecules is an order of magnitude greater at a pH of 6.2 than at 5.2. For example, in tests at an ammonia concentration of 10 g/L, the NH3 concentration was 13 mg N/L at pH ¼ 6.2, and 1.3 mg/L at a pH of 5.2. The reactors showed longer lag times and slower production rates at a pH of 5.2, suggesting that the inhibitory mechanism is different than that expected on the basis of just NH3 concentrations.
3.2.
Continuous-flow reactor experiments
pH 5.2
60
350 Total Gas H2
300 250 200 150 100 50 0 0
2
4 6 Ammonia (g N/L)
8
Fig. 5 – Total gas and biohydrogen gas production in continuous-flow experiments as a function of total ammonia concentration (at pH 5.5). The changes in both total gas and H2 production were statistically significant (for H2: slope ¼ 9.05, R2 ¼ 0:97, po0:02). Gas production was completely inhibited when the reactor was started at or above 1.6 g N/L, but not at 3.8, 5.3, or 7.8 g N/L when initially started at 0.8 g N/L.
pH 6.2 2.5
40
2.0 Yield (mol/mol)
Max.Gas Prod. Rate (mL/hr)
Acclimation to a low nitrogen concentration prior to the introduction of a high concentration of NH3/NH+4 (40.8 g N/L) was critical to successful production of hydrogen. When the initial concentration of NH3/NH+4 in the reactor operated under batch conditions was 1.6 g N/L or higher, total gas production and hydrogen production were all negligible, and the reactor inoculum washed out when the reactor was switched to continuous mode operation (data not shown). Therefore, in subsequent tests the reactor was first run in batch mode at 0.8 g N/L, and then switched to a feed containing the same or higher concentration of NH3/NH+4 . Ammonia toxicity has been shown for methanogens in batch experiments (Lay et al., 1998), where concentrations above 4 g N/L showed reduced methane production and increased lag
times. Toxicity has also been shown in UASB reactors (Borja et al., 1996), where digestion was inhibited above 5 g N/L. Hydrogen production in a 2 L chemostat decreased by 15% (from 170 to 145 mL/h) when the ammonia concentration was increased from 0.8 to 3.9 g N/L in the feed after reactor acclimation to 0.8 g N/L (Fig. 5). The hydrogen yield decreased from 1.9 to 1.6 mol-H2/mol-glucose when the ammonia concentration was increased from 0.8 to 3.9 g N/L. During this feed transition, the total gas production rate decreased (259–230 mL/h) and hydrogen production decreased (170–145 mL/h). The glucose removed increased from 86% to 91% (this change was not statistically significant, p40:1). Hydrogen production was further inhibited, while yield was
Gas Production (mL / hr)
1170
20
0 0.5
2
5
7
10
Ammonia (g N/L)
1.5 1.0 0.5 0.0 0
Fig. 4 – Maximum gas production rate in batch experiments as a function of ammonia concentration. Lines show linear regression of data at the different pHs (the dashed line is for pH 5.2, and the solid line is for pH 6.2). Error bars show 95% confidence intervals.
2
6 4 Ammonia (g N/L)
8
Fig. 6 – Yield of hydrogen production (in mol H2 gas per mol of glucose consumed) in continuous-flow reactor experiments. The decrease is statistically significant (slope ¼ 0.12, R2 ¼ 0:98, po0:01).
ARTICLE IN PRESS WAT E R R E S E A R C H
1171
4 0 (200 6) 116 7 – 117 2
Table 1 – Summary of results from a 2 L continuous-flow reactor experiments, including hydrogen production and hydrogen yield at various levels of ammonia concentration Ammonia (g/L)
Total gas (mL/h)
H2 (mL/h)
H2
Yield (mol/mol)
Glucose removed (%)
0.8 3.9 5.3 7.8
259 230 216 156
170 145 138 105
66 63 64 67
1.9 1.6 1.5 1.1
86 91 92 92
Table 2 – Residual acids, solvents, and alcohols in solution in chemostat experiments under steady state conditions Ammonia (g/L) 0.8 3.9 5.3 7.8 1.6
Acetone (ppm)
Methanol (ppm)
Ethanol (ppm)
n-Butanol (ppm)
Acetate (ppm)
Propionate (ppm)
Butyrate (ppm)
— — — 11.471 8.771.7
3.270.3 — — — —
2376 2273 29.779.6 480734 4307110
2.170.3 6778 38710 — —
510730 550770 470710 52071 110710
4.370.3 1.770.2 — — —
1240790 8707110 12707200 37875 2571.0
Dashes indicate concentrations below the detection limit. n-Propanol was below detectable limits for all samples. (Errors represent 95% confidence intervals, with n ¼ 3 for each analysis).
decreased, when the reactor was transitioned to a feed containing 5.3 or 7.8 g N/L. The hydrogen yield consistently decreased with ammonia concentration from 1.9 mol-H2/mol-glucose at 0.8 g N/L, to 1.1 mol/mol at 7.8 g N/L (Fig. 6). The concentration of hydrogen gas remained constant at 6571.6% for the different ammonia feed concentrations (Table 1). Thus, the decrease in hydrogen yield was due to reduced gas production, with the hydrogen production rate decreasing from 170 mL/h at 0.8 g N/L, to 105 mL/h at 7.8 g N/L. The glucose removal rate increased slightly, from 86% to 92%, as the ammonia concentration increased (Table 1), suggesting that the shift to end products other than hydrogen resulted in increased substrate utilization. The liquid products in the reactor were analyzed to determine how the changes in the influent ammonia concentrations affected the liquid products of glucose fermentation. The main soluble products were acetic and butyric acids for conditions that resulted in successful hydrogen production (Table 2). Addition of ammonia also increased the production of n-butanol. The levels of acetone, methanol, and propionate were below detectable limits for all experiments. The 1.6 g N/L of ammonia run shown at the bottom of Table 2 is the data from when the reactor was started up at 1.6 g N/L, whereas for all other runs the reactor was started up at 0.8 g N/L and then switched to the appropriate value. When started at 1.6 g N/L, the reactor produced more than 400 ppm ethanol, but little acetate or butyrate (Table 2). The high production of ethanol in the reactor started up at an ammonia concentration of 1.6 g N/L indicates that either spore-forming hydrogen producers were not the dominant community, or that the bacteria began utilizing an ethanolproducing metabolic pathway. This pathway, via acetalde-
hyde as an intermediate, does not produce H2 like the acetate and butyrate pathways do. The low levels of acetate and butyrate show that these hydrogen-producing pathways were not being utilized. The high levels of acetate and butyrate in the reactor run at 0.8 g N/L are indicative of the high hydrogen production and yield observed. The ratio of acetate to butyrate increased when 3.9 g N/L was introduced to the reactor. However, given the values shown in Table 2, and assuming 4 mol-H2/mol-glucose via the acetate pathway versus 2 mol-H2/mol-glucose via the butyrate pathway, we would predict 13% less hydrogen production from the high ammonia reactor than the low ammonia reactor. This is very close to the measured decrease in hydrogen production rate of 15%. A similar analysis performed for the 7.8 g N/L conditions would predict a 37% decrease in hydrogen production, which is similar to the 38% observed decrease.
4.
Conclusions
These results show that hydrogen production can be inhibited by high concentrations of ammonia, such as those present in animal wastewaters. Under fed batch conditions, the overall yield of hydrogen is not affected but the rate of hydrogen production is decreased. When a continuous flow reactor was initially operated (in batch mode) at a high (X1.6 g N/L) concentration of ammonia, hydrogen production failed. However, if the reactor was first started up at a lower ammonia concentration (0.8 g N/L), hydrogen could be produced but the overall yield was reduced from 1.9 to 1.1 mol-H2/mol-glucose for ammonia concentrations of 0.8 to 7.8 g N/L. The concentration of hydrogen in the gas remained constant under all continuous flow conditions (6571.6%), and
ARTICLE IN PRESS 1172
WAT E R R E S E A R C H
40 (2006) 1167– 1172
thus the decrease in hydrogen yields was due to reduced gas production rates. It is therefore possible to produce hydrogen under high (up to 7.8 g N/L) concentrations of ammonia but the ammonia can reduce the overall rate or conversion of glucose to hydrogen.
Acknowledgements This research was supported by Grant 68-3A75-3-150 from the Natural Resources Conservation Service of the United States Department of Agriculture. R E F E R E N C E S
Angenent, L.T., Karim, K., Al-Dahhan, M.H., Wrenn, B.A., Domiguez-Espinosa, R., 2004. Production of bioenergy and biochemicals from industrial and agricultural wastewater. Trends Biotechnol. 22 (9), 477–485. Benemann, J., 1996. Hydrogen biotechnology: progress and prospects. Nat. Biotechnol. 14, 1101–1103. Bhattacharya, S.K., Parkin, G.F., 1989. The effect of ammonia on methane fermentation processes. J. Water Pollut. Control Fed. 61 (1), 55–59. Borja, R., Sanchez, E., Weiland, P., 1996. Influence of ammonia concentration on thermophilic anaerobic digestion of cattle manure in upflow anaerobic sludge blanket (UASB) reactors. Process Biochem. 31 (5), 477. Cai, M., Liu, J., Wei, Y., 2004. Enhanced biohydrogen production from sewage sludge with alkaline pretreatment. Environ. Sci. Technol. 38, 3195–3202. Ceccherini, M.T., Castaldini, M., Piovanelli, C., Hastings, R.C., McCarthy, A.J., Bazzicalupo, M., Miclaus, N., 1998. Effects of swine manure fertilization on autotrophic ammonia oxidizing bacteria in soil. Appl. Soil Ecolog. 7, 149–157. Chang, F.Y., Lin, C.Y., 2004. Biohydrogen production using an upflow anaerobic sludge blanket reactor. Int. J. Hydrogen Energy 29, 33–39. Cruz, H.J., Freitas, C.M., Alves, P.M., Moreira, J.L., Carrondo, M.J.T., 2000. Effects of ammonia and lactate on growth, metabolism, and productivity of BHK cells. Enzyme Microb. Technol. 27, 43–52. Dabrock, B., Bahl, H., Gottschalk, G., 1992. Parameters affecting solvent production by Clostridium pasteurianum. Appl. Environ. Microbiol. 58, 1233–1239. Fang, H.H.P., Liu, H., 2002. Effect of pH on hydrogen production from glucose by a mixed culture. Bioresource Technol. 82, 87–93. Gottschalk, G., 1986. Bacterial Metabolism, second ed. Springer, New York. Han, S.K., Shin, H.S., 2004. Biohydrogen production by anaerobic fermentation of food waste. Int. J. Hydrogen Energy 29, 569–577.
Hawkes, F.R., Dinsdale, R., Hawkes, D.L., Hussy, I., 2002. Sustainable fermentative hydrogen production: challenges for process optimization. Int. J. Hydrogen Energy 27, 1339–1347. Iyer, P., Bruns, M.A., Zhang, H., Van Ginkel, S., Logan, B.E., 2004. Hydrogen gas production in a continuous flow bioreactor using heat-treated soil inocula. Appl. Microbiol. Biotechnol. 89 (1), 119–127. Kim, S.H., Han, S.K., Shin, H.S., 2004. Feasibility of biohydrogen production by anaerobic co-digestion of food waste and sewage sludge. Int. J. Hydrogen Energy 29, 1607–1616. Koster, I.W., Koomen, E., 1988. Ammonia inhibition of the maximum growth rate (mm) of hydrogenotrophic methanogens at various pH-levels and temperatures. Appl. Microbiol. Biotechnol. 28, 500–505. Lay, J.J., 2001. Biohydrogen generation by mesophilic anaerobic fermentation of microcrystalline cellulose. Biotechnol. Bioeng. 74, 280–287. Lay, J.J., Li, Y.Y., Noike, T., 1998. Influence of pH and ammonia concentration on the methane production in high-solids digestion processes. Water Environ. Res. 70 (5), 1075. Levin, D.B., Pitt, L., Love, M., 2004. Biohydrogen production: prospects and limitations to practical application. Int. J. Hydrogen Energy 29, 173–185. Logan, B.E., 2004. Feature article: biologically extracting energy from wastewater: biohydrogen production and microbial fuel cells. Environ. Sci. Technol. 38 (9), 160A–167A. Logan, B.E., Oh, S.E., Kim, I.S., Van Ginkel, S., 2002. Biological hydrogen production measured in batch anaerobic respirometers. Environ. Sci. Technol. 36 (11), 2530–2535. Oh, S.E., Van Ginkel, S., Logan, B.E., 2003. The relative effectiveness of pH control and heat treatment for enhancing biohydrogen gas production. Environ. Sci. Technol. 37, 5186–5190. Oh, S.E., Iyer, P., Bruns, M.A., Logan, B.E., 2004. Biological hydrogen production using a membrane bioreactor. Biotechnol. Bioeng. 87 (1), 119–127. Phelps, T.J., Zeikus, J.G., 1984. Influence of pH on terminal carbon metabolism in anoxic sediments from a Midly Acidic Lake. Appl. Environ. Microbiol. 48 (6), 1088–1095. Ting, C.H., Lin, K.R., Lee, D.J., Tay, J.H., 2004. Production of hydrogen and methane from wastewater sludge using anaerobic fermentation. Water Sci. Technol. 50 (9), 223–228. Van Ginkel, S., Sung, S., Lay, J.J., 2001. Biohydrogen production as a function of pH and substrate concentration. Environ. Sci. Technol. 35, 4726–4730. Van Ginkel, S.W., Logan, B.E., 2005. Increased biological hydrogen production with reduced organic loading. Water Res. 39, 3819–3826. Zhu, H., Wakayama, T., Asada, Y., Miyake, J., 2001. Hydrogen production by four cultures with participation by anoxygenic phototrophic bacterium and anaerobic bacterium in the presence of NH+4 . Int. J. Hydrogen Energy 26, 1149–1154. Zhu, H., Ueda, S., Asada, Y., Miyake, J., 2002. Hydrogen production as a novel process of wastewater treatment—studies on tofu wastewater with entrapped R. sphaeroides and mutagenesis. Int. J. Hydrogen Energy 27, 1349–1357.
