- Email: [email protected]
Characteristics of anaerobic ammonia removal by a mixed culture of hydrogen producing photosynthetic bacteria
Bioresource Technology 95 (2004) 151–158
Characteristics of anaerobic ammonia removal by a mixed culture of hydrogen producing photosynthetic bacteria Hiroo Takabatake a
a,*
, Kiyohiko Suzuki b, In-Beom Ko a, Tatsuya Noike
a
Department of Civil Engineering, Tohoku University, Aramaki-Aza Aoba06, Aoba-ku, Sendai 980-8579, Japan b Water Supply Div., Health Service Bureau, Ministry of Health, Labor and Welfare, Kasumigaseki 1-2-2, Chiyoda-ku, Tokyo 100-8975, Japan Received 5 June 2003; received in revised form 4 December 2003; accepted 10 December 2003 Available online 2 April 2004
Abstract It is known that the presence of ammonia inhibits hydrogen production by photosynthetic bacteria. In order to avoid it, a twostep process containing ammonia removal and hydrogen production was investigated in this study. Firstly, the effects of carbonate presence on ammonia removal by photosynthetic bacteria were investigated by the vial tests because it is known that the uptake of volatile fatty acids (VFAs) sometimes requires carbonate. The results of them showed that the presence of carbonate promoted the uptake of VFAs and ammonia. Especially, the uptake of propionate and/or butyrate required the presence of carbonate. The results of the batch experiments of two-step hydrogen production showed that the depletion of ammonia triggered hydrogen evolution. Herein, the presence of albumin did not inhibit hydrogen evolution and preferably it increased the hydrogen production rate. And the VFA-C/NH4 -N ratio in substrate fed into two-step hydrogen production process should be more than 6.0. Ó 2004 Elsevier Ltd. All rights reserved. Keywords: Two-step hydrogen production process; Photosynthetic bacteria; Acidogenic wastewater; Ammonia removal; Carbonate; Albumin
1. Introduction Interest in hydrogen gas as a promising energy source has been increasing recently. Hydrogen gas can be produced by water electrolysis, coal gasification or biological conversion from organic matters. Among them, biological conversion from organic matter has been studied as a candidate of energy saving processes by a large number of researchers worldwide. Up to the present time, a large number of bacteria have been shown to be hydrogen producers, including obligate and facultative anaerobes, aerobes, cyanobacteria and photosynthetic bacteria (Nandi and Sengupta, 1998; Roychowdhury et al., 1988; Zajic et al., 1978). Among them, this research focused on the hydrogen production by
*
Corresponding author. Present address: Toray Industries Inc., Global Environment Research Laboratories, Nihonbashi-Muromachi 2-chome, Toray Building 2-1, Chuo-ku, Tokyo 103-8666, Japan. Tel.: +81-332455111; fax: +81-775338407. E-mail address: [email protected] (H. Takabatake). 0960-8524/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.biortech.2003.12.019
photosynthetic bacteria, which are able to convert the energy of light into hydrogen energy. The carbon source that is particularly available for hydrogen production by photosynthetic bacteria is volatile fatty acid (VFA), i.e. acetic, propionic and butyric acid. Therefore, the discharged water from acidogenesis of organic waste which is mainly composed of acetic, propionic and butyric acid is one of the candidates as the effective and inexpensive carbon sources for hydrogen production. The compositions of acidogenic wastewater reported were summarized in Table 1 (Ghosh et al., 1975; Henry et al., 1987; Eastman and Ferguson, 1981). This table shows that acidogenetic wastewater sometimes contains high concentration of ammonia as well as organic acids. It is known that ammonia inhibits hydrogen production by photosynthetic bacteria because it represses the synthesis of the key enzyme nitrogenase (Hillmer and Gest, 1977a,b; Ormerod et al., 1961; Miyake et al., 1982; Zhu et al., 2001). Therefore, the ammonia removal is inevitable for effective hydrogen production. Fascetti and Todini (1995) proposed the two-step hydrogen production process with ammonia removal. In this process, the step for ammonia removal
152
H. Takabatake et al. / Bioresource Technology 95 (2004) 151–158
Table 1 The compositions of acidogenic wastewater Substrate/operating condition Waste water sludge Primary sludge/HRT 2 days, pH 5–7 Primary sludge/HRT 1.3 days, pH 7 Primary sludge/HRT 1.5 days, pH 5 Primary sludge
TS (g/l)
NHþ 4 -N (mg/l)
VFAs (mg/l) Acetate
Propionate
Butyrate
Valerate
62
490–600 505–923
2193–2803 660–1460
716–1015 650–1220
645–1176 120–524
823–1115 210–1280
51
1008
1640
260
920
90
61
757
960
150
1040
340
2500
2900
1900
1300
23
was ahead of hydrogen production process. They tested the hydrogen production by Rhodobacter sphaeroides RV from lactic acid as a sole carbon source and succeeded the ammonia removal in the first step, where ammonia was assimilated to the cell component, followed by the hydrogen production in the second step. On the other hand, it is known that the uptake of VFAs by photosynthetic bacteria requires the presence of carbonate (van Niel, 1944; Ormerod, 1956; Stanier et al., 1959). It means that the presence of carbonate may affect the ammonia removal. In order to clarify this point, the characteristics of ammonia removal in the two-step hydrogen production by a mixed culture of photosynthetic bacteria from the synthetic acidogenetic wastewater were investigated in this study, especially on the effect of carbonate coexistence and the composition of organic acids. And it was investigated in this study whether the inclusion of protein inhibits the ammonia removal or not, because protein can be included in the wastewater and would be nitrogen sources as well as ammonia.
2. Methods
Others (mg/l)
Bicarbonate (mg/l)
Reference
Organic N 1638–2150 Organic N 1916 Organic N 2084
10–6960
Ghosh et al. (1975) Henry et al. (1987)
4120
Henry et al. (1987)
3870
Henry et al. (1987) Eastman and Ferguson (1981)
tosynthetic bacteria. Twenty milliliters of microbial broth precultured for 2 days were inoculated into 60 ml of media in a vial. After the replacement of the gas phase in a vial with argon gas, the cultivation was started in a reciprocal shaker at 30 °C with light irradiation by incandescent lamps. The media consisted of basal media and carbon source media. The composition of the basal media was 750 mg/l of K2 HPO4 , 850 mg/l of KH2 PO4 , 200 mg/l of MgSO4 Æ 7H2 O, 20 mg/l of EDTA2Na, 0.75 mg/l of Na2 MoO4 Æ 2H2 O, 10 mg/l of FeSO4 Æ 7H2 O, 0.24 mg/l of ZnSO4 Æ 7H2 O, 1.26 mg/l of MnCl2 Æ 4H2 O, 0.04 mg/l of Cu(NO3 )2 Æ 3H2 O, 1 mg/l of thiamine Æ HCl, 0.015 mg/l of p-aminobenzonic acid, 0.015 mg/l of nicotinamide, 0.015 mg/l of biotin and 1 mg/l of nicotinic acid. The composition of carbon source media was acetic acid, propionic acid, butyric acid, sodium carbonate and/or ammonium chloride. At the beginning of the experiments, pH was adjusted at 7.0. 2.3. The batch experiments for two-step hydrogen production with ammonia removal The experimental apparatus of the batch experiment was shown in Fig. 1. A thin transparent glass bottle (1.5 l) was employed as a reactor for the batch experiment. For
2.1. Microorganisms The suspension broth of mixed community of photosynthetic bacteria was purchased from Higashi-Nihon Z-lant, Japan. This microflora consisted of R. sphaeroides, Rb. capsulata, Rhodopseudomonas palustris, Rhodospirillum rubrum and Ectothiorhodospira shaposhikowii. Before inoculating it, it was anaerobically subcultured in the media; sodium pyruvate (6 mM), sodium acetate (4 mM), sodium propionate (2 mM), butyrate (4 mM) and ammonium chloride (8 mM).
thermosensor pH controller
NaOH HCl sampling
gas holder
reactor (1.5L)
2.2. The vial experiments for ammonia removal by photosynthetic bacteria
stirrer blue light
The glass vials (100 ml) were utilized for the experiments of ammonia removal by a mixed culture of pho-
Fig. 1. The experimental apparatus of the batch experiment for twostep hydrogen production.
H. Takabatake et al. / Bioresource Technology 95 (2004) 151–158
the establishment of anaerobic condition in a reactor, the gas phase of a reactor was replaced with argon gas before starting batch experiment. The gas evolved was collected into the gas holder. The microbial broth in a reactor was always agitated by a magnetic stirrer. Blue light (wavelength of 500–700 nm was cut off) was irradiated from the both sides of the glass bottle, because it is known that the blue light irradiation was more effective than incandescent light irradiation for the suppression of the algae growth inhibiting the hydrogen production (Ko and Noike, 2002). The schematic procedure of batch experiments was as follows. Firstly, seed culture described before was inoculated into a media composed of 3280 mg/l (40 mM) of sodium acetate, 960 mg/l (10 mM) of sodium propionate, 1760 mg/l (20 mM) of butyric acid, 963 mg/l (18 mM), and the basal media as described before. The tremendous increase in cell density of the broth in a reactor reduces the light transparency in general. Such a condition must reduce the activity of photosynthetic bacteria. In order to prevent such a condition, the microbial suspension was centrifuged at 3000 rpm for 20 min to reduce the cell density when the increase in cell density of the broth to some extent was observed. Three batch experiments (Run 1–3) were carried out in this study. The conditions of them were shown in Table 2. Run 1–3 were maintained in anaerobic condition with different media composition. The media of Run 2 was that of Run 1 with the addition of 2000 mg/l of sodium carbonate. And the media of Run 3 was that of Run 2 with the addition of 500 mg/l of albumin, which was used as the presentative of the proteins in this study. 2.4. Analytical method VFAs and ammonium concentration were measured by the capillary electrophoresis (Photal, CAPI-3200, OHTSUKA, Japan). Total organic carbon (TOC) and inorganic carbon (IC) were determined by TOC analyzer (TOC-5000, Shimadzu, Japan). The measurement of dry weight was carried out in accordance with the Standard Method (American Public Health Association, 1992). However, the dry weight was calculated from the value of the optimal density at 680 nm (OD680) since the volumetric limitation of the sampling made the frequent sampling for the dry weight measurement impossible. Table 2 The operational condition of batch experiments Additional component of media Run 1 Run 2 Run 3
Sodium carbonate
Albumin
None 2000 mg/l 2000 mg/l
None None 500 mg/l
153
OD680 was the absorbance at 680 nm of wavelength by the spectrophotometer (HACH, DR/4000U). The hydrogen gas was measured by the gas chromatograph (Shimadzu, GC-8A) with thermal conductivity detector (TCD).
3. Results 3.1. Effect of carbonate addition and composition of VFAs In the case that VFAs are employed as carbon sources for the growth of photosynthetic bacteria, it is known that carbonate is sometimes utilized as a final electron acceptor (van Niel, 1944; Ormerod, 1956; Stanier et al., 1959). Therefore, the effect of carbonate addition on ammonia removal was investigated under the condition that the several compositions of VFAs were fed. For it, 15 batch experiments were carried out, where the media composition was listed in Table 3. In the experiments of Nos. 1–6, a sole kind of VFA (acetic, propionic or butyric acid) was fed as a carbon source. In the experiments of Nos. 7–12, two kinds of VFAs (acetic, propionic and/or butyric acid) were fed. And in the experiments of Nos. 13 and 14, three kinds of VFAs (acetic, propionic and butyric acid) were fed. In No. 15, any VFAs were not included. Approximately 1000 mg/l of sodium carbonate was added in the experiments of the even numbers and No. 15, whereas no carbonate in the other experiments. Ammonium chloride was added at approximately one-fourth mol as nitrogen of mol as carbon (without sodium carbonate) fed. Fig. 2 shows the time course of cell density, ammonia, VFA and IC in the experiment of No. 2 as an example of the results of the tests. Growth rate, ammonia removal rate, VFA reduction rate, IC increasing rate and the ratio of reduced VFA-C to reduced ammonia-N were Table 3 The composition of the carbon source media in the vial tests Run
Acetate (mM)
Propionate (mM)
Butyrate (mM)
Sodium carbonate (mg/l)
NH4 Cl (mM)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
40 40 – – – – 20 20 – – 20 20 20 20 –
– – 40 40 – – 20 20 20 20 – – 20 20 –
– – – – 40 40 – – 20 20 20 20 20 20 –
– 1000 – 1000 – 1000 – 1000 – 1000 – 1000 – 1000 500
23 22 35 35 45 46 32 31 38 48 46 41 45 46 10
154
H. Takabatake et al. / Bioresource Technology 95 (2004) 151–158
40 VFA (mM)
cell density (g-dryweight/l)
3 2 1 0
0
20
40 time (hr)
60
10 0
20
40 time (hr)
60
80
0 0
20
40 time (hr)
60
80
150 IC (mg/l)
NH4-N (mM)
20
0
80
15 10
100
5 0
acetate propionate butyrate
30
0
20
40 time (hr)
60
50
80
Fig. 2. The time course of cell density, ammonia, VFAs and IC in the experiment of No. 2.
calculated by the method of least squares and summarized in Table 4. Throughout these experiments, any gases were not evolved. The VFAs and ammonia were simultaneously reduced with the significant increase in cell density. These results indicated that VFAs and ammonia were assimilated into the cell components of photosynthetic bacteria. When acetic acid was fed as a sole carbon source (Nos. 1 and 2), the significant microbial growth was observed in both the presence and the absence of sodium carbonate. On the contrary of it, the absence of sodium carbonate depressed microbial growth in the case that propionic acid or butyric acid was fed as a sole carbon source (Nos. 3 and 5) though the remarked microbial growth was observed in the presence of carbonate (Nos. 4 and 6). In the presence of carbonate, the increase in IC was observed during the consumption of acetic acid (No. 1) while the decrease during the consumption of propionic (No. 3) and butyric acid (No. 5).
In the experiments where multi kinds of VFAs were added without carbonate (Nos. 7, 9, 11 and 13), no microbial growth was observed except for the experiments where acetic acid and propionic acid were fed (No. 7). The difference between No. 7 and a group of Nos. 9, 11 and 13 was the presence of butyric acid in the carbon source. And, the significant microbial growth was observed in the experiments of Nos. 8, 10, 12 and 14. These results indicated that the assimilation of butyric acid into biomass requires carbonate. In the experiment of No. 15 where no VFAs but carbonate was added, a slight microbial growth was observed. It indicated that carbonate was not so useful as a carbon source for microbial growth. Compared with the results of No. 2n and 2n 1 ðn ¼ 1; 2; . . . ; 7Þ, the growth rate and the reduction rate of ammonia and VFAs of No. 2n were always higher than these of No. 2n 1 (Table 4). It turned out that the presence of carbonate promoted the uptake of carbon
Table 4 Growth rate, ammonia removal rate, VFA reduction rate, IC increasing rate and the ratio of reduced VFA-C to reduced ammonia-N in the vial tests Nos. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Growth rate (mg dw/l/h)
Ammonia reduction rate (mM/h)
Acetate
VFA consumption rate (mM/h) Propionate
Butyrate
C/N ratio (mM/mM)
IC increasing rate (mg C/l/h)
87.7 117.5 0.0 42.6 0.3 35.2 14.3 69.6 0.9 28.0 4.3 101.5 2.7 62.4 5.5
0.393 0.445 0.060 0.325 0.000 0.168 0.100 0.629 0.041 0.275 0.099 0.877 0.000 0.291 0.022
1.187 1.279 – – – – 0.054 0.547 – – 0.107 0.489 0.000 0.048 –
– – 0.206 0.859 – – 0.077 0.462 0.031 0.227 – – 0.007 0.158 –
– – – – 0.000 0.592 – – 0.019 0.172 0.102 0.395 0.002 0.187 –
6.04 5.75 10.30 7.93 – 14.10 3.39 3.94 4.12 4.98 6.28 2.92 – 4.53 –
5.236 6.093 0.012 )0.158 0.032 )1.056 0.200 1.850 0.040 )0.188 )0.030 1.535 )0.003 )0.897 )0.692
H. Takabatake et al. / Bioresource Technology 95 (2004) 151–158
To investigate the effect of carbonate concentration on microbial growth, six sets of experiments were conducted with the addition of sodium carbonate at the concentration of 0, 50, 100, 250, 500 or 750 mg/l. The carbon source media used in this study contained 20 mM of acetic acid, 5 mM of propionic acid, 10 mM of butyric acid and 25 mM of ammonium chloride. The ammonia removal and increase in cell density after the incubation of 72 h in each experiment were showed in Fig. 3. The tendency of cell density and ammonia removal was similar to each other, i.e. they were increased with the increase in the concentration of carbonate when the carbonate added was less than 100 mg/l and they were almost equivalent at more than 100 mg/l of carbonate. This result showed that the sufficient microbial growth required more than 100 mg/l of carbonate in this case. Additionally, sodium carbonate was added at 72 h in the experiment where no carbonate was added. Then, the significant microbial growth involving the consumption of VFAs and ammonia began (data not shown). This shows that the concentration of carbonate
3.3. The batch experiments for two-step hydrogen production with ammonia removal The batch experiments of Run 1, 2 and 3 were carried out to investigate the influence of the addition of carbonate and protein on the ammonia removal followed by hydrogen production. Albumin was chosen as a typical protein in this study. The time course of cell density, cumulative volume of hydrogen gas evolution
NH4-N (mM)
3.2. Effect of carbonate concentration
is the unique factor limiting the growth rate of photosynthetic bacteria.
20 18 16 14 12 10 8 6 4 2 0
Run1 Run2 Run3
0
cumulative volume of hydrogen gas evolution (g/l)
and nitrogen source resulting in the increase in a microbial growth. Focusing on the experiments of Nos. 1, 2, 4, 6, 8, 10, 12 and 14, where remarkable uptake of carbon source was observed, IC increasing rates in the experiments where significant uptake of acetate was observed (Nos. 1, 2, 8 and 12) were positive whereas those in the other experiments were negative. This phenomenon show the interaction between the composition of carbon source and the role of carbonate for substrate uptake.
50
100 time (hr)
150
200
120 Run1 Run2 Run3
100 80 60 40 20 0 0
50
3
20
155
100 time (hr)
150
200
18
2
12 1.5
10 8
1
6 ammonia removal the increase in cell density
4 2
0.5
2 1.5 1 0.5 0
0 0
200 400 600 carbonate concentration (mM)
0 800
Fig. 3. The ammonia removal and the increase in cell density during the 72 h cultivation under the addition of sodium carbonate; 0, 50, 100, 250, 500 and 750 mg/l.
Run1 Run2 Run3
2.5 cell density (g/l)
14
increase in cell density (g/l)
ammonia removal (mM)
3
2.5
16
0
50
100 time (hr)
150
200
Fig. 4. The time course of cell density, cumulative volume of hydrogen gas evolution and ammonia concentration in Run 1 (neither carbonate nor albumin addition), Run 2 (carbonate) and Run 3 (both carbonate and albumin).
156
H. Takabatake et al. / Bioresource Technology 95 (2004) 151–158
and ammonia concentration in Run 1, 2 and 3 were shown in Fig. 4. In Run 1, microbial growth was terminated though both the carbon and nitrogen sources was still remained. Throughout of the batch experiments, no hydrogen gas evolution was observed. These results show that the condition of Run 1 was not suitable for microbial growth and ammonia removal, meaning that the absence of carbonate repressed the ammonia removal. In Run 2, ammonia supplied initially was completely consumed during 96 h with remarkable microbial growth. Thereupon, the broth in a reactor was centrifuged to decrease the cell density. The beginning of hydrogen gas evolution was observed at 144 h, and totally 100 ml of hydrogen gas was evolved during 192 h of batch experiment. Remarkable microbial growth was visible until 130 h, whereas slight microbial growth after 130 h. These results indicate that the addition of carbonate promoted microbial growth involving ammonium removal resulting in a significant production of hydrogen gas. In Run 3, the stable microbial growth involving ammonium removal was observed until 72 h. The centrifugation of the broth in a reactor was done at 72 and 120 h to reduce the cell density. Ammonia concentration was gradually decreased with time and depleted at 144 h. Then, the rapid evolution of hydrogen gas was occurred and 100 ml of hydrogen gas was totally evolved during 192 h of cultivation. During the remarkable production of hydrogen gas, the stable microbial growth was observed though the depletion of ammonium. In the initial of this batch experiment, 800 mg/l of albumin were added. It decreased down to 560 and 350 mg/l at 120 and 192 h, respectively. These results indicated that the presence of albumin did not inhibit the assimilation of ammonia and the hydrogen production, and albumin was available as a nitrogen source of microbial growth.
4. Discussion 4.1. Effect of carbonate addition on ammonia removal The results obtained in this research showed that the microflora of photosynthetic bacteria in this research assimilate ammonia and VFAs, such as acetic, propionic and butyric acid, and that the absence of carbonate limits their assimilation. The increase in IC during the uptake of acetic acid and the decrease in IC during the uptake of propionic and butyric acid were observed. van Niel reported that carbonate fixation was occurred during the uptake of the substrate more reductive than cell components and carbonate release during the uptake of the substrate more oxidative (van Niel, 1944). Ormerod et al. studied the fixation and release of carbonate during the uptake of the various kinds of substrate by using the carbonate labeled by 14 C (Ormerod,
1956). The release of carbonate was only observed for acetate uptake. And the simultaneous release and uptake of carbonate was observed for the uptake of proionic and butyric acid. Because carbonate plays a role as an electron acceptor when the substrates such as propionic and butyric acid more reductive than the cell components were anaerobically taken up, the absence of carbonate prevent propionic and butyric acid from being taken up due to the unbalance of the oxidative– reductive potential. Ormerod reported that the carbonate release for the uptake of 1 mM of acetic, propionic and butyric acid were +0.2–0.3, )0.23 and )0.38 mM respectively (Ormerod, 1956). On the contrary of it, they were +0.37–0.40, )0.02 and )0.15 mM respectively according to the experimental results where a sole kind of carbon source was fed in this study. When the equivalent amount of acetic and butyric acid was fed as carbon source, the uptake of acetic acid would supply a sufficient amount of carbonate for the uptake of butyric acid according to the results in this study. In the experiments of Nos. 11 and 13 where acetic and butyric acid were added as carbon source, remarkable uptake of butyric acid was not observed. It means that carbonate supplied via the uptake of acetic acid was insufficient for the uptake of butyric acid. The comparison between Nos. 5 and 11 showed that the coexistence of acetic acid in the absence of carbonate slightly increased the uptake rate of butyric acid. It indicated that the presence of acetic acid played the role of keeping the oxidative–reductive balance in a cell, resulting in the increase in the uptake rate of butyric acid. As described above, the results of this study indicated that the uptake of reductive substrate, such as propionic and butyric acid, required carbonate as an electron acceptor. However, the phenomenon was observed which could not be explained only by it. Under the absence of carbonate, the uptake rate of acetic acid was lower in the presence of propionic and/or butyric acid than that in the absence of them. It is not considered as the inhibitory effect of propionic and/or butyric acid on the acetate uptake because remarkable uptake of acetic acid was observed in the presence of carbonate. It indicated that carbonate also played a role other than that maintaining the oxidative–reductive balance. Further studies are required for the clarification of it. 4.2. Two-step hydrogen production with ammonia removal Hydrogen gas production rate, ammonia removal rate, growth rate, specific growth rate, VFA uptake rate, the ratio of VFA-C/NH4 -N and the ratio of actual/ theoretical hydrogen gas production was calculated from the experimental results in Run 1, 2 and 3, and they were summarized in Table 5.
H. Takabatake et al. / Bioresource Technology 95 (2004) 151–158
157
Table 5 Summary of the results of batch experiments for two-step hydrogen production with ammonia removal Hydrogen gas production rate Ammonia removal rate Growth rate Specific growth rate VFA uptake rate VFA-C/NH4 -Na Actual/theoretical hydrogen gas productionb a b
Unit
Run 1
Run 2
Run 3
ml/g/h mM/h mg/l/h 1/d mM/h mM/mM mM/mM
0.00 0.036 9.0 1.09 0.35 9.81 0.00
1.76 0.199 26.4 1.14 1.18 5.91 0.27
5.18 0.120 28.7 1.34 0.72 5.98 0.22
The value of consumed carbon of VFA divided by consumed nitrogen of ammonia during the presence of ammonia. Theoretical amount of hydrogen production was calculated as the following stoichiometric equations; Acetic acid: CH3 COOH þ 2H2 O ¼ 2CO2 þ 4H2 : Propionic acid: C2 H5 COOH þ 4H2 O ¼ 3CO2 þ 7H2 : Butyric acid: C3 H7 COOH þ 2H2 O ¼ 4CO2 þ 10H2 :
In Run 1, only a slight ammonia removal was observed, resulting that ammonia was still remained and no hydrogen gas was evolved after 192 h. It must be because the absence of carbonate inhibited the ammonia removal. This result also supported that the presence of carbonate plays an important role of ammonia removal. A significant hydrogen production was observed in Run 2 and Run 3. And the hydrogen production rate in Run 3 was much higher than that of Run 2. This indicated that the addition of albumin did not inhibit the hydrogen production and preferably increased the activity of hydrogen production. However, the addition of albumin decreased ammonia removal rate and VFA uptake rate. The ratio of consumed carbon of VFA divided by consumed nitrogen of ammonia during the presence of ammonia (VFA-C/NH4 -N) in Run 2 and 3 were 5.91 and 5.98 respectively. It means that the removal of 1 mol of ammonia required approximately 6.0 mol of carbons of VFAs. This indicated that the VFA-C/NH4 -N ratio in substrate fed into two-step hydrogen production process should be more than 6.0. The VFA-C/NH4 -N ratio of acidogenic wastewater of primary sludge listed in Table 1 was around 1.5–7.0. This means that such a wastewater is infeasible for hydrogen production by photosynthetic bacteria. The primary sludge contains high concentration of proteins. Therefore the acidogenic wastewater of primary sludge included high concentration of ammonia produced via the hydrolysis of proteins during acidogenesis. However, the acidogenic wastewater of the resource with high C/N ratio, such as noodle manufacturing wastewater, would be feasible for twostep hydrogen production by photosynthetic bacteria.
5. Conclusion The influence of carbonate on ammonia removal by photosynthetic bacteria was studied by vial tests. Ammonia were removed involving the uptake of carbon source and growth. The requirement of carbonate de-
pended on the components of carbon source. The uptake of acetate which is more oxidative than a cell component did not require carbonate, whereas the uptake of propionic and butyric acids which are more reductive required. This probably means that carbonate plays an important role for keeping the oxidative– reductive balance in a cell. Additionally, the influence of carbonate and protein on ammonia removal and hydrogen production were studied by the batch tests. The absence of carbonate inhibited the ammonia removal, resulting in no hydrogen production. And, the presence of protein inhibited neither ammonia removal nor hydrogen production. The substrate where the VFA-C/NH4 -N ratio is more than 6.0 should be feasible for the operation of the twostep hydrogen production. Acknowledgements This work was supported by Core Research for Evolutional Science and Technology (CREST) of Japan Science and Technology (JST). The authors sincerely appreciate Mr. Kuninobu Sakurai, Tohoku University, Japan, for the grateful help of the experimental progression. References American Public Health Association, 1992. Standard methods for the examination of water and wastewater. American Public Health Association, Washington, DC. Eastman, J.A., Ferguson, J.F., 1981. Solubilization of particulate organic carbon during the acid phase of anaerobic digestion. J. Water Pollut. Control Fed. 53. Fascetti, E., Todini, O., 1995. Rhodobacter sphaeroides RV cultivation and hydrogen production in a one- and two-stage chemostat. Appl. Microbiol. 44, 300–305. Ghosh, S., Conrad, F.R., Klass, D.L., 1975. Anaerobic acidogenesis of wastewater sludge. J. Water Pollut. Control Fed. 47 (1), 30–45. Henry, M.P., Sajjad, A., Ghosh, S., 1987. The effects of environmental factors on acid-phase digestion of sewage sludge. In: 42nd Purdue
158
H. Takabatake et al. / Bioresource Technology 95 (2004) 151–158
University Industrial Waste Conference Proceedings, pp. 727– 737. Hillmer, P., Gest, H., 1977a. H2 metabolism in the photosynthetic bacterium Rhodopseudomonas capsulata: H2 production by growing cultures. J. Bateriol. 129, 724–731. Hillmer, P., Gest, H., 1977b. H2 metabolism in the photosynthetic bacterium Rhodopseudomonas capsulata: production and utilization of H2 by resting cells. J. Bateriol. 129, 732–739. Ko, I.B., Noike, T., 2002. Use of blue optical filters for suppression of growth of algae in hydrogen producing non-axenic cultures of Rhodobacter sphaeroides RV. Int. J. Hydrogen Energy 27, 1297– 1302. Miyake, J., Tomizuka N., Kawamura S., 1982. Effect of glutamate on the hydrogen production by Rhodobacter rubrum. Report of the Fermentation Institute, Japan, vol. 58, pp. 25–31. Nandi, R., Sengupta, S., 1998. Microbial production of hydrogen: An overview. Critical Rev. Microbiol. 24 (1), 61–84. Ormerod, J.G., 1956. The use of radioactive carbon dioxide in the measurement of carbon dioxide fixation in Rhodospirillum rubrum. Biochem. J. 64, 373–380.
Ormerod, J.G., Ormerod, K.S., Gest, H., 1961. Light dependent utilization of organic compounds and photoproduction of molecular hydrogen by photosynthetic bacteria: Relationships with nitrogen metabolism. Arch. Biochem. Biophys. 94, 449– 463. Roychowdhury, S., Cox, D., Levandowsky, M., 1988. Production of hydrogen by microbial fermentation. Int. J. Hydrogen Energy 13 (7), 407–410. Stanier, R.Y., Doudoroff, M., Kunisawa, R., Contopoulou, R., 1959. The role of organic substrate in bacterial photosynthesis. Biochemistry 45, 1246–1260. van Niel, C.B., 1944. The culture general physiology, morphology, and classification of the non-sulfur purple and brown bacteria. Bacterial Rev. 8, 1. Zajic, J.E., Kosaric, N., Brossean, J.D., 1978. Microbial production of hydrogen. Adv. Biochem. Eng. 9, 58–109. 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 NHþ 4 . Int. J. Hydrogen Energy 26, 1149–1154.
Characteristics of anaerobic ammonia removal by a mixed culture of hydrogen producing photosynthetic bacteria Hiroo Takabatake a
a,*
, Kiyohiko Suzuki b, In-Beom Ko a, Tatsuya Noike
a
Department of Civil Engineering, Tohoku University, Aramaki-Aza Aoba06, Aoba-ku, Sendai 980-8579, Japan b Water Supply Div., Health Service Bureau, Ministry of Health, Labor and Welfare, Kasumigaseki 1-2-2, Chiyoda-ku, Tokyo 100-8975, Japan Received 5 June 2003; received in revised form 4 December 2003; accepted 10 December 2003 Available online 2 April 2004
Abstract It is known that the presence of ammonia inhibits hydrogen production by photosynthetic bacteria. In order to avoid it, a twostep process containing ammonia removal and hydrogen production was investigated in this study. Firstly, the effects of carbonate presence on ammonia removal by photosynthetic bacteria were investigated by the vial tests because it is known that the uptake of volatile fatty acids (VFAs) sometimes requires carbonate. The results of them showed that the presence of carbonate promoted the uptake of VFAs and ammonia. Especially, the uptake of propionate and/or butyrate required the presence of carbonate. The results of the batch experiments of two-step hydrogen production showed that the depletion of ammonia triggered hydrogen evolution. Herein, the presence of albumin did not inhibit hydrogen evolution and preferably it increased the hydrogen production rate. And the VFA-C/NH4 -N ratio in substrate fed into two-step hydrogen production process should be more than 6.0. Ó 2004 Elsevier Ltd. All rights reserved. Keywords: Two-step hydrogen production process; Photosynthetic bacteria; Acidogenic wastewater; Ammonia removal; Carbonate; Albumin
1. Introduction Interest in hydrogen gas as a promising energy source has been increasing recently. Hydrogen gas can be produced by water electrolysis, coal gasification or biological conversion from organic matters. Among them, biological conversion from organic matter has been studied as a candidate of energy saving processes by a large number of researchers worldwide. Up to the present time, a large number of bacteria have been shown to be hydrogen producers, including obligate and facultative anaerobes, aerobes, cyanobacteria and photosynthetic bacteria (Nandi and Sengupta, 1998; Roychowdhury et al., 1988; Zajic et al., 1978). Among them, this research focused on the hydrogen production by
*
Corresponding author. Present address: Toray Industries Inc., Global Environment Research Laboratories, Nihonbashi-Muromachi 2-chome, Toray Building 2-1, Chuo-ku, Tokyo 103-8666, Japan. Tel.: +81-332455111; fax: +81-775338407. E-mail address: [email protected] (H. Takabatake). 0960-8524/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.biortech.2003.12.019
photosynthetic bacteria, which are able to convert the energy of light into hydrogen energy. The carbon source that is particularly available for hydrogen production by photosynthetic bacteria is volatile fatty acid (VFA), i.e. acetic, propionic and butyric acid. Therefore, the discharged water from acidogenesis of organic waste which is mainly composed of acetic, propionic and butyric acid is one of the candidates as the effective and inexpensive carbon sources for hydrogen production. The compositions of acidogenic wastewater reported were summarized in Table 1 (Ghosh et al., 1975; Henry et al., 1987; Eastman and Ferguson, 1981). This table shows that acidogenetic wastewater sometimes contains high concentration of ammonia as well as organic acids. It is known that ammonia inhibits hydrogen production by photosynthetic bacteria because it represses the synthesis of the key enzyme nitrogenase (Hillmer and Gest, 1977a,b; Ormerod et al., 1961; Miyake et al., 1982; Zhu et al., 2001). Therefore, the ammonia removal is inevitable for effective hydrogen production. Fascetti and Todini (1995) proposed the two-step hydrogen production process with ammonia removal. In this process, the step for ammonia removal
152
H. Takabatake et al. / Bioresource Technology 95 (2004) 151–158
Table 1 The compositions of acidogenic wastewater Substrate/operating condition Waste water sludge Primary sludge/HRT 2 days, pH 5–7 Primary sludge/HRT 1.3 days, pH 7 Primary sludge/HRT 1.5 days, pH 5 Primary sludge
TS (g/l)
NHþ 4 -N (mg/l)
VFAs (mg/l) Acetate
Propionate
Butyrate
Valerate
62
490–600 505–923
2193–2803 660–1460
716–1015 650–1220
645–1176 120–524
823–1115 210–1280
51
1008
1640
260
920
90
61
757
960
150
1040
340
2500
2900
1900
1300
23
was ahead of hydrogen production process. They tested the hydrogen production by Rhodobacter sphaeroides RV from lactic acid as a sole carbon source and succeeded the ammonia removal in the first step, where ammonia was assimilated to the cell component, followed by the hydrogen production in the second step. On the other hand, it is known that the uptake of VFAs by photosynthetic bacteria requires the presence of carbonate (van Niel, 1944; Ormerod, 1956; Stanier et al., 1959). It means that the presence of carbonate may affect the ammonia removal. In order to clarify this point, the characteristics of ammonia removal in the two-step hydrogen production by a mixed culture of photosynthetic bacteria from the synthetic acidogenetic wastewater were investigated in this study, especially on the effect of carbonate coexistence and the composition of organic acids. And it was investigated in this study whether the inclusion of protein inhibits the ammonia removal or not, because protein can be included in the wastewater and would be nitrogen sources as well as ammonia.
2. Methods
Others (mg/l)
Bicarbonate (mg/l)
Reference
Organic N 1638–2150 Organic N 1916 Organic N 2084
10–6960
Ghosh et al. (1975) Henry et al. (1987)
4120
Henry et al. (1987)
3870
Henry et al. (1987) Eastman and Ferguson (1981)
tosynthetic bacteria. Twenty milliliters of microbial broth precultured for 2 days were inoculated into 60 ml of media in a vial. After the replacement of the gas phase in a vial with argon gas, the cultivation was started in a reciprocal shaker at 30 °C with light irradiation by incandescent lamps. The media consisted of basal media and carbon source media. The composition of the basal media was 750 mg/l of K2 HPO4 , 850 mg/l of KH2 PO4 , 200 mg/l of MgSO4 Æ 7H2 O, 20 mg/l of EDTA2Na, 0.75 mg/l of Na2 MoO4 Æ 2H2 O, 10 mg/l of FeSO4 Æ 7H2 O, 0.24 mg/l of ZnSO4 Æ 7H2 O, 1.26 mg/l of MnCl2 Æ 4H2 O, 0.04 mg/l of Cu(NO3 )2 Æ 3H2 O, 1 mg/l of thiamine Æ HCl, 0.015 mg/l of p-aminobenzonic acid, 0.015 mg/l of nicotinamide, 0.015 mg/l of biotin and 1 mg/l of nicotinic acid. The composition of carbon source media was acetic acid, propionic acid, butyric acid, sodium carbonate and/or ammonium chloride. At the beginning of the experiments, pH was adjusted at 7.0. 2.3. The batch experiments for two-step hydrogen production with ammonia removal The experimental apparatus of the batch experiment was shown in Fig. 1. A thin transparent glass bottle (1.5 l) was employed as a reactor for the batch experiment. For
2.1. Microorganisms The suspension broth of mixed community of photosynthetic bacteria was purchased from Higashi-Nihon Z-lant, Japan. This microflora consisted of R. sphaeroides, Rb. capsulata, Rhodopseudomonas palustris, Rhodospirillum rubrum and Ectothiorhodospira shaposhikowii. Before inoculating it, it was anaerobically subcultured in the media; sodium pyruvate (6 mM), sodium acetate (4 mM), sodium propionate (2 mM), butyrate (4 mM) and ammonium chloride (8 mM).
thermosensor pH controller
NaOH HCl sampling
gas holder
reactor (1.5L)
2.2. The vial experiments for ammonia removal by photosynthetic bacteria
stirrer blue light
The glass vials (100 ml) were utilized for the experiments of ammonia removal by a mixed culture of pho-
Fig. 1. The experimental apparatus of the batch experiment for twostep hydrogen production.
H. Takabatake et al. / Bioresource Technology 95 (2004) 151–158
the establishment of anaerobic condition in a reactor, the gas phase of a reactor was replaced with argon gas before starting batch experiment. The gas evolved was collected into the gas holder. The microbial broth in a reactor was always agitated by a magnetic stirrer. Blue light (wavelength of 500–700 nm was cut off) was irradiated from the both sides of the glass bottle, because it is known that the blue light irradiation was more effective than incandescent light irradiation for the suppression of the algae growth inhibiting the hydrogen production (Ko and Noike, 2002). The schematic procedure of batch experiments was as follows. Firstly, seed culture described before was inoculated into a media composed of 3280 mg/l (40 mM) of sodium acetate, 960 mg/l (10 mM) of sodium propionate, 1760 mg/l (20 mM) of butyric acid, 963 mg/l (18 mM), and the basal media as described before. The tremendous increase in cell density of the broth in a reactor reduces the light transparency in general. Such a condition must reduce the activity of photosynthetic bacteria. In order to prevent such a condition, the microbial suspension was centrifuged at 3000 rpm for 20 min to reduce the cell density when the increase in cell density of the broth to some extent was observed. Three batch experiments (Run 1–3) were carried out in this study. The conditions of them were shown in Table 2. Run 1–3 were maintained in anaerobic condition with different media composition. The media of Run 2 was that of Run 1 with the addition of 2000 mg/l of sodium carbonate. And the media of Run 3 was that of Run 2 with the addition of 500 mg/l of albumin, which was used as the presentative of the proteins in this study. 2.4. Analytical method VFAs and ammonium concentration were measured by the capillary electrophoresis (Photal, CAPI-3200, OHTSUKA, Japan). Total organic carbon (TOC) and inorganic carbon (IC) were determined by TOC analyzer (TOC-5000, Shimadzu, Japan). The measurement of dry weight was carried out in accordance with the Standard Method (American Public Health Association, 1992). However, the dry weight was calculated from the value of the optimal density at 680 nm (OD680) since the volumetric limitation of the sampling made the frequent sampling for the dry weight measurement impossible. Table 2 The operational condition of batch experiments Additional component of media Run 1 Run 2 Run 3
Sodium carbonate
Albumin
None 2000 mg/l 2000 mg/l
None None 500 mg/l
153
OD680 was the absorbance at 680 nm of wavelength by the spectrophotometer (HACH, DR/4000U). The hydrogen gas was measured by the gas chromatograph (Shimadzu, GC-8A) with thermal conductivity detector (TCD).
3. Results 3.1. Effect of carbonate addition and composition of VFAs In the case that VFAs are employed as carbon sources for the growth of photosynthetic bacteria, it is known that carbonate is sometimes utilized as a final electron acceptor (van Niel, 1944; Ormerod, 1956; Stanier et al., 1959). Therefore, the effect of carbonate addition on ammonia removal was investigated under the condition that the several compositions of VFAs were fed. For it, 15 batch experiments were carried out, where the media composition was listed in Table 3. In the experiments of Nos. 1–6, a sole kind of VFA (acetic, propionic or butyric acid) was fed as a carbon source. In the experiments of Nos. 7–12, two kinds of VFAs (acetic, propionic and/or butyric acid) were fed. And in the experiments of Nos. 13 and 14, three kinds of VFAs (acetic, propionic and butyric acid) were fed. In No. 15, any VFAs were not included. Approximately 1000 mg/l of sodium carbonate was added in the experiments of the even numbers and No. 15, whereas no carbonate in the other experiments. Ammonium chloride was added at approximately one-fourth mol as nitrogen of mol as carbon (without sodium carbonate) fed. Fig. 2 shows the time course of cell density, ammonia, VFA and IC in the experiment of No. 2 as an example of the results of the tests. Growth rate, ammonia removal rate, VFA reduction rate, IC increasing rate and the ratio of reduced VFA-C to reduced ammonia-N were Table 3 The composition of the carbon source media in the vial tests Run
Acetate (mM)
Propionate (mM)
Butyrate (mM)
Sodium carbonate (mg/l)
NH4 Cl (mM)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
40 40 – – – – 20 20 – – 20 20 20 20 –
– – 40 40 – – 20 20 20 20 – – 20 20 –
– – – – 40 40 – – 20 20 20 20 20 20 –
– 1000 – 1000 – 1000 – 1000 – 1000 – 1000 – 1000 500
23 22 35 35 45 46 32 31 38 48 46 41 45 46 10
154
H. Takabatake et al. / Bioresource Technology 95 (2004) 151–158
40 VFA (mM)
cell density (g-dryweight/l)
3 2 1 0
0
20
40 time (hr)
60
10 0
20
40 time (hr)
60
80
0 0
20
40 time (hr)
60
80
150 IC (mg/l)
NH4-N (mM)
20
0
80
15 10
100
5 0
acetate propionate butyrate
30
0
20
40 time (hr)
60
50
80
Fig. 2. The time course of cell density, ammonia, VFAs and IC in the experiment of No. 2.
calculated by the method of least squares and summarized in Table 4. Throughout these experiments, any gases were not evolved. The VFAs and ammonia were simultaneously reduced with the significant increase in cell density. These results indicated that VFAs and ammonia were assimilated into the cell components of photosynthetic bacteria. When acetic acid was fed as a sole carbon source (Nos. 1 and 2), the significant microbial growth was observed in both the presence and the absence of sodium carbonate. On the contrary of it, the absence of sodium carbonate depressed microbial growth in the case that propionic acid or butyric acid was fed as a sole carbon source (Nos. 3 and 5) though the remarked microbial growth was observed in the presence of carbonate (Nos. 4 and 6). In the presence of carbonate, the increase in IC was observed during the consumption of acetic acid (No. 1) while the decrease during the consumption of propionic (No. 3) and butyric acid (No. 5).
In the experiments where multi kinds of VFAs were added without carbonate (Nos. 7, 9, 11 and 13), no microbial growth was observed except for the experiments where acetic acid and propionic acid were fed (No. 7). The difference between No. 7 and a group of Nos. 9, 11 and 13 was the presence of butyric acid in the carbon source. And, the significant microbial growth was observed in the experiments of Nos. 8, 10, 12 and 14. These results indicated that the assimilation of butyric acid into biomass requires carbonate. In the experiment of No. 15 where no VFAs but carbonate was added, a slight microbial growth was observed. It indicated that carbonate was not so useful as a carbon source for microbial growth. Compared with the results of No. 2n and 2n 1 ðn ¼ 1; 2; . . . ; 7Þ, the growth rate and the reduction rate of ammonia and VFAs of No. 2n were always higher than these of No. 2n 1 (Table 4). It turned out that the presence of carbonate promoted the uptake of carbon
Table 4 Growth rate, ammonia removal rate, VFA reduction rate, IC increasing rate and the ratio of reduced VFA-C to reduced ammonia-N in the vial tests Nos. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Growth rate (mg dw/l/h)
Ammonia reduction rate (mM/h)
Acetate
VFA consumption rate (mM/h) Propionate
Butyrate
C/N ratio (mM/mM)
IC increasing rate (mg C/l/h)
87.7 117.5 0.0 42.6 0.3 35.2 14.3 69.6 0.9 28.0 4.3 101.5 2.7 62.4 5.5
0.393 0.445 0.060 0.325 0.000 0.168 0.100 0.629 0.041 0.275 0.099 0.877 0.000 0.291 0.022
1.187 1.279 – – – – 0.054 0.547 – – 0.107 0.489 0.000 0.048 –
– – 0.206 0.859 – – 0.077 0.462 0.031 0.227 – – 0.007 0.158 –
– – – – 0.000 0.592 – – 0.019 0.172 0.102 0.395 0.002 0.187 –
6.04 5.75 10.30 7.93 – 14.10 3.39 3.94 4.12 4.98 6.28 2.92 – 4.53 –
5.236 6.093 0.012 )0.158 0.032 )1.056 0.200 1.850 0.040 )0.188 )0.030 1.535 )0.003 )0.897 )0.692
H. Takabatake et al. / Bioresource Technology 95 (2004) 151–158
To investigate the effect of carbonate concentration on microbial growth, six sets of experiments were conducted with the addition of sodium carbonate at the concentration of 0, 50, 100, 250, 500 or 750 mg/l. The carbon source media used in this study contained 20 mM of acetic acid, 5 mM of propionic acid, 10 mM of butyric acid and 25 mM of ammonium chloride. The ammonia removal and increase in cell density after the incubation of 72 h in each experiment were showed in Fig. 3. The tendency of cell density and ammonia removal was similar to each other, i.e. they were increased with the increase in the concentration of carbonate when the carbonate added was less than 100 mg/l and they were almost equivalent at more than 100 mg/l of carbonate. This result showed that the sufficient microbial growth required more than 100 mg/l of carbonate in this case. Additionally, sodium carbonate was added at 72 h in the experiment where no carbonate was added. Then, the significant microbial growth involving the consumption of VFAs and ammonia began (data not shown). This shows that the concentration of carbonate
3.3. The batch experiments for two-step hydrogen production with ammonia removal The batch experiments of Run 1, 2 and 3 were carried out to investigate the influence of the addition of carbonate and protein on the ammonia removal followed by hydrogen production. Albumin was chosen as a typical protein in this study. The time course of cell density, cumulative volume of hydrogen gas evolution
NH4-N (mM)
3.2. Effect of carbonate concentration
is the unique factor limiting the growth rate of photosynthetic bacteria.
20 18 16 14 12 10 8 6 4 2 0
Run1 Run2 Run3
0
cumulative volume of hydrogen gas evolution (g/l)
and nitrogen source resulting in the increase in a microbial growth. Focusing on the experiments of Nos. 1, 2, 4, 6, 8, 10, 12 and 14, where remarkable uptake of carbon source was observed, IC increasing rates in the experiments where significant uptake of acetate was observed (Nos. 1, 2, 8 and 12) were positive whereas those in the other experiments were negative. This phenomenon show the interaction between the composition of carbon source and the role of carbonate for substrate uptake.
50
100 time (hr)
150
200
120 Run1 Run2 Run3
100 80 60 40 20 0 0
50
3
20
155
100 time (hr)
150
200
18
2
12 1.5
10 8
1
6 ammonia removal the increase in cell density
4 2
0.5
2 1.5 1 0.5 0
0 0
200 400 600 carbonate concentration (mM)
0 800
Fig. 3. The ammonia removal and the increase in cell density during the 72 h cultivation under the addition of sodium carbonate; 0, 50, 100, 250, 500 and 750 mg/l.
Run1 Run2 Run3
2.5 cell density (g/l)
14
increase in cell density (g/l)
ammonia removal (mM)
3
2.5
16
0
50
100 time (hr)
150
200
Fig. 4. The time course of cell density, cumulative volume of hydrogen gas evolution and ammonia concentration in Run 1 (neither carbonate nor albumin addition), Run 2 (carbonate) and Run 3 (both carbonate and albumin).
156
H. Takabatake et al. / Bioresource Technology 95 (2004) 151–158
and ammonia concentration in Run 1, 2 and 3 were shown in Fig. 4. In Run 1, microbial growth was terminated though both the carbon and nitrogen sources was still remained. Throughout of the batch experiments, no hydrogen gas evolution was observed. These results show that the condition of Run 1 was not suitable for microbial growth and ammonia removal, meaning that the absence of carbonate repressed the ammonia removal. In Run 2, ammonia supplied initially was completely consumed during 96 h with remarkable microbial growth. Thereupon, the broth in a reactor was centrifuged to decrease the cell density. The beginning of hydrogen gas evolution was observed at 144 h, and totally 100 ml of hydrogen gas was evolved during 192 h of batch experiment. Remarkable microbial growth was visible until 130 h, whereas slight microbial growth after 130 h. These results indicate that the addition of carbonate promoted microbial growth involving ammonium removal resulting in a significant production of hydrogen gas. In Run 3, the stable microbial growth involving ammonium removal was observed until 72 h. The centrifugation of the broth in a reactor was done at 72 and 120 h to reduce the cell density. Ammonia concentration was gradually decreased with time and depleted at 144 h. Then, the rapid evolution of hydrogen gas was occurred and 100 ml of hydrogen gas was totally evolved during 192 h of cultivation. During the remarkable production of hydrogen gas, the stable microbial growth was observed though the depletion of ammonium. In the initial of this batch experiment, 800 mg/l of albumin were added. It decreased down to 560 and 350 mg/l at 120 and 192 h, respectively. These results indicated that the presence of albumin did not inhibit the assimilation of ammonia and the hydrogen production, and albumin was available as a nitrogen source of microbial growth.
4. Discussion 4.1. Effect of carbonate addition on ammonia removal The results obtained in this research showed that the microflora of photosynthetic bacteria in this research assimilate ammonia and VFAs, such as acetic, propionic and butyric acid, and that the absence of carbonate limits their assimilation. The increase in IC during the uptake of acetic acid and the decrease in IC during the uptake of propionic and butyric acid were observed. van Niel reported that carbonate fixation was occurred during the uptake of the substrate more reductive than cell components and carbonate release during the uptake of the substrate more oxidative (van Niel, 1944). Ormerod et al. studied the fixation and release of carbonate during the uptake of the various kinds of substrate by using the carbonate labeled by 14 C (Ormerod,
1956). The release of carbonate was only observed for acetate uptake. And the simultaneous release and uptake of carbonate was observed for the uptake of proionic and butyric acid. Because carbonate plays a role as an electron acceptor when the substrates such as propionic and butyric acid more reductive than the cell components were anaerobically taken up, the absence of carbonate prevent propionic and butyric acid from being taken up due to the unbalance of the oxidative– reductive potential. Ormerod reported that the carbonate release for the uptake of 1 mM of acetic, propionic and butyric acid were +0.2–0.3, )0.23 and )0.38 mM respectively (Ormerod, 1956). On the contrary of it, they were +0.37–0.40, )0.02 and )0.15 mM respectively according to the experimental results where a sole kind of carbon source was fed in this study. When the equivalent amount of acetic and butyric acid was fed as carbon source, the uptake of acetic acid would supply a sufficient amount of carbonate for the uptake of butyric acid according to the results in this study. In the experiments of Nos. 11 and 13 where acetic and butyric acid were added as carbon source, remarkable uptake of butyric acid was not observed. It means that carbonate supplied via the uptake of acetic acid was insufficient for the uptake of butyric acid. The comparison between Nos. 5 and 11 showed that the coexistence of acetic acid in the absence of carbonate slightly increased the uptake rate of butyric acid. It indicated that the presence of acetic acid played the role of keeping the oxidative–reductive balance in a cell, resulting in the increase in the uptake rate of butyric acid. As described above, the results of this study indicated that the uptake of reductive substrate, such as propionic and butyric acid, required carbonate as an electron acceptor. However, the phenomenon was observed which could not be explained only by it. Under the absence of carbonate, the uptake rate of acetic acid was lower in the presence of propionic and/or butyric acid than that in the absence of them. It is not considered as the inhibitory effect of propionic and/or butyric acid on the acetate uptake because remarkable uptake of acetic acid was observed in the presence of carbonate. It indicated that carbonate also played a role other than that maintaining the oxidative–reductive balance. Further studies are required for the clarification of it. 4.2. Two-step hydrogen production with ammonia removal Hydrogen gas production rate, ammonia removal rate, growth rate, specific growth rate, VFA uptake rate, the ratio of VFA-C/NH4 -N and the ratio of actual/ theoretical hydrogen gas production was calculated from the experimental results in Run 1, 2 and 3, and they were summarized in Table 5.
H. Takabatake et al. / Bioresource Technology 95 (2004) 151–158
157
Table 5 Summary of the results of batch experiments for two-step hydrogen production with ammonia removal Hydrogen gas production rate Ammonia removal rate Growth rate Specific growth rate VFA uptake rate VFA-C/NH4 -Na Actual/theoretical hydrogen gas productionb a b
Unit
Run 1
Run 2
Run 3
ml/g/h mM/h mg/l/h 1/d mM/h mM/mM mM/mM
0.00 0.036 9.0 1.09 0.35 9.81 0.00
1.76 0.199 26.4 1.14 1.18 5.91 0.27
5.18 0.120 28.7 1.34 0.72 5.98 0.22
The value of consumed carbon of VFA divided by consumed nitrogen of ammonia during the presence of ammonia. Theoretical amount of hydrogen production was calculated as the following stoichiometric equations; Acetic acid: CH3 COOH þ 2H2 O ¼ 2CO2 þ 4H2 : Propionic acid: C2 H5 COOH þ 4H2 O ¼ 3CO2 þ 7H2 : Butyric acid: C3 H7 COOH þ 2H2 O ¼ 4CO2 þ 10H2 :
In Run 1, only a slight ammonia removal was observed, resulting that ammonia was still remained and no hydrogen gas was evolved after 192 h. It must be because the absence of carbonate inhibited the ammonia removal. This result also supported that the presence of carbonate plays an important role of ammonia removal. A significant hydrogen production was observed in Run 2 and Run 3. And the hydrogen production rate in Run 3 was much higher than that of Run 2. This indicated that the addition of albumin did not inhibit the hydrogen production and preferably increased the activity of hydrogen production. However, the addition of albumin decreased ammonia removal rate and VFA uptake rate. The ratio of consumed carbon of VFA divided by consumed nitrogen of ammonia during the presence of ammonia (VFA-C/NH4 -N) in Run 2 and 3 were 5.91 and 5.98 respectively. It means that the removal of 1 mol of ammonia required approximately 6.0 mol of carbons of VFAs. This indicated that the VFA-C/NH4 -N ratio in substrate fed into two-step hydrogen production process should be more than 6.0. The VFA-C/NH4 -N ratio of acidogenic wastewater of primary sludge listed in Table 1 was around 1.5–7.0. This means that such a wastewater is infeasible for hydrogen production by photosynthetic bacteria. The primary sludge contains high concentration of proteins. Therefore the acidogenic wastewater of primary sludge included high concentration of ammonia produced via the hydrolysis of proteins during acidogenesis. However, the acidogenic wastewater of the resource with high C/N ratio, such as noodle manufacturing wastewater, would be feasible for twostep hydrogen production by photosynthetic bacteria.
5. Conclusion The influence of carbonate on ammonia removal by photosynthetic bacteria was studied by vial tests. Ammonia were removed involving the uptake of carbon source and growth. The requirement of carbonate de-
pended on the components of carbon source. The uptake of acetate which is more oxidative than a cell component did not require carbonate, whereas the uptake of propionic and butyric acids which are more reductive required. This probably means that carbonate plays an important role for keeping the oxidative– reductive balance in a cell. Additionally, the influence of carbonate and protein on ammonia removal and hydrogen production were studied by the batch tests. The absence of carbonate inhibited the ammonia removal, resulting in no hydrogen production. And, the presence of protein inhibited neither ammonia removal nor hydrogen production. The substrate where the VFA-C/NH4 -N ratio is more than 6.0 should be feasible for the operation of the twostep hydrogen production. Acknowledgements This work was supported by Core Research for Evolutional Science and Technology (CREST) of Japan Science and Technology (JST). The authors sincerely appreciate Mr. Kuninobu Sakurai, Tohoku University, Japan, for the grateful help of the experimental progression. References American Public Health Association, 1992. Standard methods for the examination of water and wastewater. American Public Health Association, Washington, DC. Eastman, J.A., Ferguson, J.F., 1981. Solubilization of particulate organic carbon during the acid phase of anaerobic digestion. J. Water Pollut. Control Fed. 53. Fascetti, E., Todini, O., 1995. Rhodobacter sphaeroides RV cultivation and hydrogen production in a one- and two-stage chemostat. Appl. Microbiol. 44, 300–305. Ghosh, S., Conrad, F.R., Klass, D.L., 1975. Anaerobic acidogenesis of wastewater sludge. J. Water Pollut. Control Fed. 47 (1), 30–45. Henry, M.P., Sajjad, A., Ghosh, S., 1987. The effects of environmental factors on acid-phase digestion of sewage sludge. In: 42nd Purdue
158
H. Takabatake et al. / Bioresource Technology 95 (2004) 151–158
University Industrial Waste Conference Proceedings, pp. 727– 737. Hillmer, P., Gest, H., 1977a. H2 metabolism in the photosynthetic bacterium Rhodopseudomonas capsulata: H2 production by growing cultures. J. Bateriol. 129, 724–731. Hillmer, P., Gest, H., 1977b. H2 metabolism in the photosynthetic bacterium Rhodopseudomonas capsulata: production and utilization of H2 by resting cells. J. Bateriol. 129, 732–739. Ko, I.B., Noike, T., 2002. Use of blue optical filters for suppression of growth of algae in hydrogen producing non-axenic cultures of Rhodobacter sphaeroides RV. Int. J. Hydrogen Energy 27, 1297– 1302. Miyake, J., Tomizuka N., Kawamura S., 1982. Effect of glutamate on the hydrogen production by Rhodobacter rubrum. Report of the Fermentation Institute, Japan, vol. 58, pp. 25–31. Nandi, R., Sengupta, S., 1998. Microbial production of hydrogen: An overview. Critical Rev. Microbiol. 24 (1), 61–84. Ormerod, J.G., 1956. The use of radioactive carbon dioxide in the measurement of carbon dioxide fixation in Rhodospirillum rubrum. Biochem. J. 64, 373–380.
Ormerod, J.G., Ormerod, K.S., Gest, H., 1961. Light dependent utilization of organic compounds and photoproduction of molecular hydrogen by photosynthetic bacteria: Relationships with nitrogen metabolism. Arch. Biochem. Biophys. 94, 449– 463. Roychowdhury, S., Cox, D., Levandowsky, M., 1988. Production of hydrogen by microbial fermentation. Int. J. Hydrogen Energy 13 (7), 407–410. Stanier, R.Y., Doudoroff, M., Kunisawa, R., Contopoulou, R., 1959. The role of organic substrate in bacterial photosynthesis. Biochemistry 45, 1246–1260. van Niel, C.B., 1944. The culture general physiology, morphology, and classification of the non-sulfur purple and brown bacteria. Bacterial Rev. 8, 1. Zajic, J.E., Kosaric, N., Brossean, J.D., 1978. Microbial production of hydrogen. Adv. Biochem. Eng. 9, 58–109. 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 NHþ 4 . Int. J. Hydrogen Energy 26, 1149–1154.