Biofuels generation from sweet sorghum: Fermentative hydrogen production and anaerobic digestion of the remaining biomass

Bioresource Technology 99 (2008) 110–119

Biofuels generation from sweet sorghum: Fermentative hydrogen production and anaerobic digestion of the remaining biomass Georgia Antonopoulou a,b, Hariklia N. Gavala a,b,*, Ioannis V. Skiadas K. Angelopoulos c, Gerasimos Lyberatos a,b a b

a,b

,

Department of Chemical Engineering, University of Patras, Karatheodori 1 st., 26500 Patras, Greece Institute of Chemical Engineering and High Temperature Chemical Processes, 26504 Patras, Greece c Department of Biology, University of Patras, 26500 Patras, Greece Received 10 July 2006; received in revised form 26 November 2006; accepted 27 November 2006 Available online 25 January 2007

Abstract The present study focuses on the exploitation of sweet sorghum biomass as a source for hydrogen and methane. Fermentative hydrogen production from the sugars of sweet sorghum extract was investigated at different hydraulic retention times (HRT). The subsequent methane production from the effluent of the hydrogenogenic process and the methane potential of the remaining solids after the extraction process were assessed as well. The highest hydrogen production rate (2550 ml H2/d) was obtained at the HRT of 6 h while the highest yield of hydrogen produced per kg of sorghum biomass was achieved at the HRT of 12 h (10.4 l H2/kg sweet sorghum). It has been proved that the effluent from the hydrogenogenic reactor is an ideal substrate for methane production with approximately 29 l CH4/kg of sweet sorghum. Anaerobic digestion of the solid residues after the extraction process yielded 78 l CH4/kg of sweet sorghum. This work demonstrated that biohydrogen production can be very efficiently coupled with a subsequent step of methane production and that sweet sorghum could be an ideal substrate for a combined gaseous biofuels production.  2006 Elsevier Ltd. All rights reserved. Keywords: Biofuels; Fermentation; Hydrogen; Methane; Sweet sorghum

1. Introduction Renewable energy sources have received great interest from the international community during the last decades. Biomass is one of the oldest and the most promising energy sources and includes organic and animal wastes, wastewater, energy crops, agricultural and industrial residues that can be used for the production of biofuels. Nowadays, biomass provides approximately 14% of the total worldwide energy needs (International Energy Agency, 1998) *

Corresponding author. Address: Department of Chemical Engineering, University of Patras, Karatheodori 1 st., 26500 Patras, Greece. Tel.: +30 2610997858; fax: +30 2610993070. E-mail addresses: [email protected], hng@biocentrum. dtu.dk (H.N. Gavala). 0960-8524/$ - see front matter  2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.biortech.2006.11.048

and represents an important contributor to the world economy (Chum and Overend, 2001; Parikka, 2004). There are several conversion technologies for the energy exploitation of biomass, that is, direct combustion (which is responsible for over 97% of the world’s bioenergy production), gasification, pyrolysis, and biological treatment (Demirbas, 2004). The latter has economical and environmental advantages compared to the other technologies. Biomass can be biologically converted to liquid or gaseous fuels (Claassen et al., 1999), such as ethanol, methanol, methane and hydrogen, which was recently characterized as the fuel of the future. Hydrogen is a clean and environmentally friendly fuel, which produces water instead of greenhouse gases when combusted. It can be produced by renewable raw materials, such as organic wastes, and possesses a high-energy yield

G. Antonopoulou et al. / Bioresource Technology 99 (2008) 110–119

(122 kJ/g) due to its light weight. Furthermore, hydrogen could be directly used to produce electricity through fuel cells (Lay et al., 1999; Benemann, 1996). One of the main drawbacks of hydrogen use is the difficulty of securing a safe storage, especially in automobiles. However, this problem could be overcome by the use of metal hydrides and carbon nanotubes, which reversibly adsorb hydrogen at room temperature and low pressures (Ramachandran and Menon, 1998; Noike and Mizuno, 2000; Ajayan and Zhou, 2001). There are still significant technical barriers concerning these technologies but overcoming them is an area of active research. Biological hydrogen production, one of the several ways to produce hydrogen, has received special attention during the last decade. Biohydrogen may be produced by cyanobacteria and algae through biophotolysis of water (Asada and Miyake, 1999) or by photosynthetic and chemosyntheticfermentative bacteria. Anaerobic fermentative bacteria produce hydrogen without photoenergy, and so the cost of hydrogen production is 340 times lower than the photosynthetic process (Morimoto, 2002; Atif et al., 2005). In addition, it is well known that carbohydrates are the main source of hydrogen during fermentative processes and therefore wastes/wastewater or agricultural residues rich in carbohydrates can be considered as potential sources of hydrogen (Kapdan and Kargi, 2006). Glucose and sucrose are the fermentation substrates most studied in the laboratory (Chen et al., 2001; Zoetemeyer et al., 1982a). Degradation of sugars is accompanied by the production of hydrogen and various metabolic products, mainly volatile fatty acids (acetic, propionic and butyric acids), lactic acid and ethanol, during the fermentation process. The hydrogen yield varies proportionally to the final metabolic products. Production of acetic and butyric acids favors production of hydrogen (Nandi and Sengupta, 1998; Hawkes et al., 2002), with the fermentation to acetic acid giving the highest theoretical yield of 4 mol H2/mol hexose, while low H2 yields are associated with more reduced end products, i.e. propionic and lactic acids and ethanol. However, mixed acid fermentation producing butyrate in excess of acetate occurs upon biological degradation of glucose by clostridia-type microflora (Chen et al., 2001; Mizuno et al., 2000; Fang and Liu, 2002). This genus produces hydrogen using the activities of pyruvate – ferrodoxin – oxidoreductase and hydrogenase enzymes. Sporeformer clostridia are selected from natural environments by heat treatment (Lin and Chang, 1999; Lay, 2000; Chen and Lin, 2001; Sung et al., 2002). Boiled anaerobically digested sludge has been shown to give successful start-up of continuous laboratory-scale bioreactors fed with glucose (Mu et al., 2006). Hydrogen production by fermentative bacteria is highly dependent on the conditions of the process, such as pH, hydraulic retention time (HRT) and gas partial pressure, which affect the microbial metabolic balance and subsequently the fermentation end-products. In general, the dominant metabolism in a mixed acidogenic culture depends strongly on the pH of the microbial culture

111

(Lay, 2000) and hydrogen production is suppressed by both low and high pH (Afschar et al., 1986; Ueno et al., 1996; Chen et al., 2002; Lee et al., 2002). It has been reported that maximum hydrogen yields are obtained when the pH of the culture medium is between 5 and 6 (Zoetemeyer et al., 1982b; Lay, 2000; Fang and Liu, 2002). On the other hand, Ren et al. (1997) stated that optimum hydrogen production would occur at pH less than 5. Thus, it is important to control the pH in order to maintain satisfactory hydrogen production. Apparently, fermentative hydrogen production (acidogenesis) process does not significantly reduce the organic content of the feed. Usually, chemical oxygen demand (COD) removal is below 20% during hydrogen production process, which corresponds to a mean hydrogen production of 2.5 mol/mol glucose. This can be removed in a subsequent anaerobic digestion step with the conversion of organic content to methane. The present study focuses on the exploitation of sweet sorghum biomass as a source for hydrogen and subsequent methane production and organic matter stabilization from the hydrogen reactor effluent. In general, biomass from energy crops, such as sweet sorghum, can be used as raw material for biohydrogen production. Sweet sorghum is an annual C4 plant characterized by high photosynthetic efficiency. It is a high biomass yielding and rich on carbohydrates crop. Its stalks mainly consist of sucrose that amounts up to 55% of dry matter and of glucose (3.2% of dry matter). They also contain cellulose (12.4%) and hemicellulose (10.2%). Sweet sorghum biomass is rich in readily fermentable sugars and thus it can be considered as an excellent raw material for fermentative hydrogen production. Overall, out of many ‘‘new crops’’ that are currently investigated as potential raw materials for energy and industry, sweet sorghum seems to be the most promising one (Dalianis et al., 1996; Gosse, 1996). To date, ethanol and methane are among the best-known microbial products produced from sweet sorghum (Jackman, 1987; Richards et al., 1991; Christakopoulos et al., 1993; Lezinou et al., 1994, 1995; Mamma et al., 1995, 1996). Specifically, the energy yield from ethanol obtained from the above referenced studies ranged between 6500–8900 kJ/kg dry and 1400–2700 kJ/ kg fresh sorghum biomass, respectively (assuming that the energy yield from ethanol is 26500 kJ/kg). The present study concerns: (a) the fermentative production of hydrogen from the sugars contained in a sorghum extract in a continuous stirred tank type bioreactor (H2-CSTR) at various hydraulic retention times using an indigenous mixed microbial culture, and (b) the subsequent anaerobic treatment of the effluent of the H2-CSTR with the simultaneous production of methane in a continuous stirred tank type reactor as well (CH4-CSTR). Furthermore, the methane potential of the solids remaining after the extraction process is determined and the overall potential of sweet sorghum biomass for hydrogen and methane production is assessed. The sweet sorghum biomass used in the present study was produced through biological cultivation.

112

G. Antonopoulou et al. / Bioresource Technology 99 (2008) 110–119

2. Experimental 2.1. Analytical methods Determinations of soluble (after centrifugation at 3000 rpm for 10 min and filtration of the supernatant liquid) chemical oxygen demand (COD), total (TSS) and volatile (VSS) suspended solids and Kjendahl nitrogen were carried out according to Standard Methods (APHA, 1975). For total P determination, the persulfate digestion method and the ascorbic acid method (APHA, 1975) were employed. For the quantification of volatile fatty acids (VFA) and ethanol (EtOH), 1 ml of sample acidified with 30 ll of 20% H2SO4 was analyzed on a gas chromatograph (VARIAN CP-30), equipped with a flame ionization detector. The oven was programmed from 105 C to 160 C at a rate of 15 C/min, and subsequently to 235 C (held for 3 min) at a rate of 20 C/min for VFA analysis and from 60 C (held for 1 min) to 230 C (held for 0.5 min) at a rate of 45 C/min for ethanol analysis. Helium was used as the carrier gas at 15 ml/min, the injector temperature was set at 175 C and the detector at 225 C and 200 C, for VFA and ethanol determinations, respectively. The concentration of lactic acid was measured on a liquid chromatograph (DIONEX DX300), equipped with an electron conductivity detector and a Dionex IonPac column (AS11-HC, 4 · 250 mm Analytical). The eluent (sodium hydroxide solution) flow rate was 1.5 ml/min and the analysis was carried out at 30 C. The produced gas composition in hydrogen and methane was quantified with a gas chromatograph (VARIAN STAR 3600) equipped with a thermal conductivity detector and a packed column with nitrogen as carrier gas. The injector, column and detector temperatures were set at 70 C, 80 C and 180 C, respectively. The measurement of the produced gas volume was based on the displacement of acidified water. For the determination of carbohydrates, a colored sugar derivative was produced through the addition of L-tryptophan and sulfuric and boric acids and subsequently measured colorimetrically at 520 nm (Joseffson, 1983). 2.2. Feedstock Fossil fuel equipments and agrochemicals (e.g. fertilizers and pesticides) are usually used when applying conventional farming methods and techniques for energy crops cultivation resulting to increased emissions of CO2 and nitrogen oxides (Cook and Beyea, 2000). Therefore, alternative cultivation methods, such as organic or biological farming, need to be used so that the cultivation of energy crops is emissions neutral. The sweet sorghum biomass (Sorghum bicolor L. Moench) used in the present study was produced in field experiments through biological farming techniques according to European Regulation EC 2092/91. The experiments were conducted at the University of Patras experimental station (latitude 3825 0 N, longitude 218 0 E). Sweet sorghum var. Keller seeds were sown at mid of May in plot rows

and the stalks were harvested at mid of October. The size of plots was 7 · 7 m2. The distance between rows was 0.70 m and between plants on the same row 0.20 m (about 7 plants per m2). An amount of 145 mm water for irrigation was applied by a drip irrigation system. For soil fertility green manure from vetch (Vicia sativa L.) biomass was incorporated in soil at mid of April. Several microorganisms preparations diluted in tap water (CompeteR Plus, MycorRTree Injectable, Plant health Care, Inc. Pittsburg, Austria) and potassium/magnesium fertilizer (Patentkali, Veterin) were also added in soil about a month before the seeds were sowed. Ten plants were sampled per 15 days and fresh and dry weights of leaf and stalks were measured for biomass productivity estimation. The sugar content of stalks was measured by a refractometer after juice extraction. 2.3. Sorghum extraction process After the harvesting of sorghum stalks, the fresh stems were stripped from the leaves, the stalks chopped to a size of 20 cm and were stored in the freezer at 20 C. Subsequently, these were milled by a laboratory grinder to an average particle size of 1–2 mm. Extraction of free sugars of the sorghum biomass was done in batches, by mixing 5 kg of milled sorghum stalks with 30 l of tap water for 1 h, at 30 C. After the extraction process a liquid fraction, rich in soluble carbohydrates and a solid fraction were obtained. The liquid fraction (sorghum extract) was preserved at 20 C and used as substrate for the hydrogen and subsequent methane production experiments in the CSTR-type bioreactors. The methane potential of the solid fraction was determined in batch experiments. 2.4. Continuous experiments for biohydrogen production A 500 ml active volume mesophilic (35 C) CSTR-type digester (H2-CSTR) was started-up and fed with sorghum extract. The reactor was cylindrical in shape, made of stainless steel and stirred periodically for 15 min, 2 times per hour. For start-up, the reactor was filled up with sorghum extract, operated anaerobically at a batch mode for 24 h in order to activate the indigenous microflora and was subsequently switched to the continuous mode at the designated HRT. The reactor was operated anaerobically at hydraulic retention times (HRT) of 24, 12, 8, 6 and 4 h and fed intermittently with sorghum extract maintained at a temperature below 4 C for 1, 2, 3, 4 and 6 min, respectively, every 3 h. Feeding was programmed always with the stirring on. Simultaneous flow of the effluent occurred during feeding by liquid overflow, in order to maintain constant reactor volume. The initial concentration of carbohydrates was calculated for every feeding cycle according to the following equation: Q

S ¼ S 0  ðS 0  S in Þ  eV t

ð1Þ

where S is the resulting concentration when feeding was completed, S0 is the influent concentration, Sin is the concen-

G. Antonopoulou et al. / Bioresource Technology 99 (2008) 110–119

2.6. Batch experiments for methane potential determination Batch experiments were carried out in duplicates at 35 C in 160 ml serum vials, in order to determine the methane potential of the solid fraction obtained after the extraction process. Solids were dried at 105 C, pulverized and diluted with tap water at a final concentration of 4 g total solids/l (solution A). 30 ml mixed liquor from the methanogenic reactor was used as inoculum and 20 ml of solution A were added as substrate. The content of the vials was gassed with a gas mixture of N2/CO2 (80/20) in order to secure anaerobic conditions. The vials were sealed with butyl rubber stoppers and aluminum crimps and methane production was monitored versus time. 3. Results and discussion 3.1. Sweet sorghum biomass Biomass and stalk sugar productivity during crop development are shown in Fig. 1. The results showed that stalks

The average characteristics of sorghum extract obtained from the different extraction batches are presented in Table 1. No ethanol or volatile fatty acids were detected. It is noticeable that the liquid fraction obtained from the extraction process consisted mainly of soluble carbohydrates, which are an ideal substrate for the fermentative hydrogen production. The concentration of nitrogen and phosphorus were too low to support microbial growth and thus the sorghum extract was supplemented with urea and KH2PO4 to make up for nutrient deficiency, as mentioned in the previous section. 3.3. Results of the continuous experiments for hydrogen production in the H2-CSTR No traces of methane were detected during the operation of the hydrogenogenic reactor at any time. In Table 2 the main characteristics of the reactor at each steady state are presented, while the corresponding biogas and hydrogen production rates and hydrogen yields are presented in Table 3. The glucose conversion efficiency was greater than 95% at all HRTs. The biogas and hydrogen production rate

2000

dry biomass productivity (experimental) dry biomass productivity (Boltzmann sigmoidal fit) sugar productivity (experimental) sugar productivity (Boltzmann sigmoidal fit)

200 -2

Sugar productivity (g m )

A 3 l active volume mesophilic (35 C) CSTR-type digester (CH4-CSTR) was started up using anaerobic sludge and fed with the effluent of the hydrogenogenic reactor. The digester had the same structural characteristics with H2CSTR and was operated anaerobically at a hydraulic retention time of 20 days. The reactor was fed for 1 min every 8 h. It must be noted, however, that the hydrogenogenic and the methanogenic reactor were not connected; the effluent of the H2-CSTR was collected, homogenized and preserved at 20 C before use. The reactor performance (biogas production and composition in CH4, pH, soluble COD and VFA concentration) was monitored 3–4 times a week and complete characterization of the reactor effluent was made when steady state was reached.

3.2. Characterization of sorghum extract

-2

2.5. Continuous experiments for methane production

are mature for cropping early in October about 110 days after plant emergence. In this period the fresh biomass yield was 91 ton/ha.

Dry biomass productivity (g m )

tration when feeding started, namely the concentration measured at the end of each cycle, Q is the volumetric feeding rate, V is the reactor volume and t is the duration of feeding. 2.24 g NaOH and 6.8025 g KH2PO4 per liter of sorghum extract were added in order to maintain the pH of the H2-CSTR at levels (4.7–5.5) allowing for hydrogen production. Also 2 g urea (NH2CONH2) per liter of sorghum extract was added to make up for N deficiency of the feed. Gas and liquid samples were taken 10–15 min before feeding started. The reactor performance (biogas production and composition in H2, pH, carbohydrates, soluble COD, ethanol, butanol, lactic acid and VFA concentration) was monitored daily throughout the experimental period. Gas samples were analyzed for methane daily, in order to monitor whether methane production was taking place. Complete characterization of the reactor effluent was made once a steady state was reached. Steady-state here meant constant conditions between successive feeding cycles.

113

1500 150 1000 100

500

50

0 0 160 170 180 190 200 210 220 230 240 250 260 270 280 290 Julian days

Fig. 1. Dry biomass and sugar productivity during crop development.

Table 1 The main characteristics of sorghum extract Characteristics

Value

pH TSS (g/l) VSS (g/l) Total COD (g/l) Soluble COD (g/l) Soluble carbohydrates (g/l) Total Kjendhal nitrogen (g/l) Total phosphorus (g/l)

7.5 ± 0.5 1.98 ± 0.27 1.87 ± 0.35 18.5 ± 2.5 17.5 ± 2.0 17.0 ± 2.0 0.025 0.035

114

G. Antonopoulou et al. / Bioresource Technology 99 (2008) 110–119

hydrogen yield is due to the competing reactions shown in Eqs. (4)–(10) and the obtained values are comparable to those reported in relevant studies with other feed materials, as shown in Table 4. At this point it should be emphasized that the start-up of the reactor was made with indigenous sweet sorghum microflora without heat treatment, which generally favors the spore-forming, hydrogen-producing clostridia. The fact that the sorghum microflora can be used for the production of hydrogen is important for the operation of a full-scale plant suggesting that no extra energy will be required neither for the start-up (heat treatment) of the reactor nor for the pasteurization or sterilization of the influent. Based on the above observations on the hydrogen yield and the production rate, the optimal HRT seems to be between 12 and 6 h. The concentrations of the main metabolic products measured at each steady state are presented in Fig. 2. The dominant metabolic products were butyric and acetic acids, with the production of butyric acid being 1.5–3 times higher than that of acetic acid. The concentration of butyric acid was the highest (5780 mg/l) at the HRT of 12 h while at the HRT of 8 h, the highest concentration of acetic acid was obtained (3480 mg/l). The concentration of propionic acid increased with the HRT, while ethanol production was favored at the HRT of 6 and 4 h, reaching the highest concentration of 634 mg/l at the HRT of 4 h. The concentration of lactic acid decreased with increased HRT until no lactic acid was detected at the HRT of 24 h. This could be attributed to either a different distribution of metabolic products at different HRT, or the accumulation of lactic acid due to kinetic limitation of its

Table 2 The characteristics of the H2-CSTR at steady state HRT (h)

pH

TSS (g/l)

VSS (g/l)

Glucose (mg/l)

Efficiency of glucose consumption (%)

24 12 8 6 4

5.5 5.3 4.8 4.9 4.8

8.5 5.5 10.7 12.8 22.0

6.4 4.1 8.8 10.2 16.5

110 ± 20 470 ± 70 680 ± 110 450 ± 20 290 ± 70

99.4 97.4 95.3 97.6 97.9

systematically increased, when the HRT was decreased from 24 to 6 h, where the higher hydrogen production rate was obtained (2550 ± 90 ml H2/d). Further decrease of the HRT to 4 h led to a reduction of the overall biogas and the hydrogen production rates. In general, the biogas composition in hydrogen always lied between 30% and 40% and was higher (40%) for the HRTs of 12, 8, and 6 h compared to those at 24 and 4 h (30% and 35%, respectively). Moreover, the yield of hydrogen produced per mole of glucose consumed and, consequently, that of hydrogen produced per kg of sweet sorghum as well (Table 3), reached the highest value at the HRT of 8 h, amounting up to 0.86 ± 0.04 mol H2/mol glucose consumed and 10.4 l H2/kg sweet sorghum, respectively. That corresponds to an energy yield of 104 kJ/kg fresh and 400 kJ/kg dry sorghum biomass, respectively (assuming that the energy yield from hydrogen is 122 000 kJ/kg). In general, the hydrogen yield was within the range of 0.37–0.86 mol H2/mol glucose consumed, significantly lower than the theoretical yield of 4 and 2 mol H2/mol glucose for acetic and butyric acid production, respectively. In general, the decreased

Table 3 Biogas and hydrogen production rates and hydrogen yields in the H2-CSTR at steady state accompanied by their standard deviation HRT (h)

Biogas composition in H2 (%)

Biogas production rate (ml/d)

Hydrogen production rate (ml/d)

Hydrogen yield (mol H2/mol glucose (consumed))

Hydrogen yield (l H2/kg sweet sorghum biomass)

24 12 8 6 4

30.4 ± 1.2 39.9 ± 1.2 40.5 ± 1.9 39.2 ± 0.5 35.0 ± 1.5

1490 ± 130 4260 ± 180 5160 ± 120 6510 ± 200 6180 ± 400

410 ± 40 1740 ± 100 2070 ± 120 2550 ± 90 2180 ± 150

0.37 ± 0.02 0.86 ± 0.04 0.75 ± 0.05 0.70 ± 0.02 0.41 ± 0.02

4.9 10.4 8.4 7.6 4.3

Table 4 Hydrogen productivity of different mesophilic mixed cultures fed with sugar-based medium in continuous suspended growth systems Study

Fang and Liu (2002) Lin and Chang (1999) Lin and Chang (2004) Ueno et al. (1996) Mizuno et al. (2000) (with sparging) Mizuno et al. (2000) (non-sparging) Gavala et al. (2006) Present

Operating conditions

Reactor characteristics and efficiency regarding hydrogen production

HRT (h)

Organic loading rate (mmol glucose/l/d)

pH

VSS (mg/l)

% H2

l H2/l/d

mmol H2/mmol glucose

6 6 6 12 8.5

155.5 416 416 109.45 150.11

5.5 5.7 6.2 6.8 6

780 1270 1670 1590 1060

64 43.1 42.6 64 22.9

4.6 15.9 8.0 4.4 4.8

2.1 1.7 1.4 2.6 1.4

8.5

150.11

6

1450

53.4

3.0

0.9

233 188.89

5.1 5.3

680 4140

42.2 39.9

4.4 3.6

1.6 0.9

6 12

G. Antonopoulou et al. / Bioresource Technology 99 (2008) 110–119 acetic acid propionic acid butyric acid ethanol lactic acid

6000 5500 5000

mg products /L

4500

115

In general, production of acetic and butyric acid favors the production of hydrogen, according to Eqs. (2) and (3) while production of propionic acid consumes hydrogen (Eq. (4)): Acetic acid production

4000

C6 H12 O6 + 2H2 O ! 2CH3 COOH + 2CO2 + 4H2

3500 3000

ð2Þ

Butyric acid production

2500

C6 H12 O6 ! CH3 CH2 CH2 COOH + 2CO2 + 2H2

2000 1500

ð3Þ

Propionic acid production

1000

C6 H12 O6 + 2H2 ! 2CH3 CH2 COOH + 2H2 O

500 0 24h

12h

8h

6h

4h

Fig. 2. The main metabolic products in the hydrogenogenic reactor, H2-CSTR, at the different steady states.

consumption to acetic and/or propionic acids. Butanol, valeric, isovaleric and isobutyric acids were not detected at all during fermentative hydrogen production from sweet sorghum extract in the CSTR at all steady states reached. In Fig. 3, the measured and calculated COD concentrations of the reactor content at each steady state are presented. Calculated COD concentration represents the sum of COD of different products (hydrogen, ethanol, lactic, butyric, propionic and acetic acids) as well as the COD of non-consumed carbohydrates measured as glucose units. The remaining COD value after subtracting the calculated COD from the measured COD concentration corresponds to non-identified metabolic products during glucose fermentation. The remaining COD represented only a very small percentage of the COD measured (Fig. 3) at HRTs in the range of 4–12 h. Therefore, it can be assumed that the main metabolic products of sweet sorghum sugars fermentation were the compounds already measured. On the other hand, at the HRT of 24 h, the remaining COD amounted up to 15% of the COD measured, suggesting that there was another metabolic product not detected at that HRT.

Lactic acid is produced from glucose via three metabolic pathways, the homofermentative (Eq. (5)), the heterofermentative (Eq. (6)) and the bifidum pathway (Eq. (7)). In all three pathways the hydrogen balance is zero, i.e. no hydrogen is consumed nor produced. The same is for ethanol production, where the hydrogen balance is zero as well (Eq. (8)). Homofermentative pathway C6 H12 O6 ! 2CH3 CHOHCOOH

ð6Þ Bifidum pathway 2C6 H12 O6 ! 3CH3 COOH + 2CH3 CHOHCOOH C6 H12 O6 ! 2CH3 CH2 OH + 2CO2

14000 COD (mg/L)

ð8Þ

A flow chart showing the different reaction pathways discussed in the present study is presented in Fig. 4. Based on the above equations and on the metabolic products measured the anticipated hydrogen production rate was calculated in the H2-CSTR, i.e. 2 mmol of hydrogen per 1 mmol of acetic acid produced, 2 mmol of hydrogen per 1 mmol of butyric acid produced and minus 1 mmol of hydrogen per 1 mmol of propionic acid produced. The

16000

12000 10000 8000 6000 4000 2000 0 8 HRT (h)

ð7Þ

Ethanol production

COD - products COD- calculated

12

ð5Þ

Heterofermentative pathway C6 H12 O6 ! CH3 CHOHCOOH + CH3 CH2 OH + CO2

18000

24

ð4Þ

6

4

Fig. 3. COD measured (COD-products) and calculated at different steady states of H2-CSTR.

116

G. Antonopoulou et al. / Bioresource Technology 99 (2008) 110–119

GLUCOSE

(eq. 4) C6H12O6 + 2H2 → 2CH3CH2COOH + 2H2O GLYCERALDEYDE PHOSPHATE

PROPIONIC ACID

[PYRUVIC ACID]

[ACETYL-CoA]

(eq. 8) C6H12O6 → 2CH3CH2OH + 2CO2 ETHANOL

(eqs. 5, 6, 7)

(eq. 3)

C6H12O6 → 2CH3CHOHCOOH C6H12O6 → CH3CHOHCOOH + CH3CH2OH + CO2 2C6H12O6 → 2CH3CHOHCOOH + 3CH3COOH

C6H12O6 → CH3CH2CH2COOH + 2CO2 + 2H2 BUTYRIC ACID

LACTIC ACID

H2 + CO2

(eq. 9) 3CH3CHOHCOOH → 2CH3CH2COOH + CH3COOH + CO2 + H2O

(eq. 10)

(eq. 2) C6H12O6 + 2H2O→ 2CH3COOH + 2CO2 + 4H2

PROPIONIC ACID + ACETIC ACID

4H2 + 2CO2 → CH3COOH + 2H2O

ACETIC ACID

Fig. 4. Flow chart showing the different reaction pathways of glucose.

production rates of acetic, propionic and butyric acids and the theoretically calculated hydrogen production rate, rHcalc, compared to the experimentally measured hydrogen production rate, rHexp, are shown in Table 7. It was observed that the rHexp is much lower than the rHcalc. Subsequently and assuming that the production of acetic and propionic acids proceeded through lactic acid fermentation according to Eq. (9), the anticipated hydrogen production rate was calculated taking into account that 2 mmol of hydrogen per 1 mmol of butyric acid are produced. Again, the rHexp is lower than the rHcalc. An explanation could be that microorganisms producing acetic acid with hydrogen consumption (Eq. (10)) have been established in the system and thus consuming a considerable amount of the produced hydrogen: 3CH3 CHOHCOOH ! 2CH3 CH2 COOH + CH3 COOH + CO2 + H2 O 4H2 + 2CO2 ! CH3 COOH + 2H2 O

ð9Þ ð10Þ

3.4. Results of continuous experiments for methane production in the CH4-CSTR The mean values of the main characteristics of the influent of the anaerobic digester after collection of the effluent of the H2-CSTR, homogenization and preservation at 20 C are presented in Table 5. The influent of the methanogenic reactor was rich in volatile fatty acids, as it was anticipated, with the concentration of soluble carbohydrates being almost negligible compared with the soluble COD concentration. The digester was operated at an

Table 5 The main characteristics of the influent of the CH4-CSTR Characteristic

Value

pH TSS (g/l) VSS (g/l) Soluble COD (g/l) Soluble carbohydrates (g COD/l) Volatile fatty acids (g COD/l)

4.7 ± 0.5 1.9 ± 0.1 1.5 ± 0.1 15.9 ± 1.5 0.9 ± 0.3 12.2 ± 2.0

HRT of 20 days for more than 120 days (which correspond to six hydraulic retention times), with an influent flow rate of 150 ml/d. The characteristics of the CH4-CSTR at steady state are presented in Table 6. Fig. 5 shows the evolution of biogas and methane throughout the experimental period. The reactor performance was very stable with acetic and butyric acids at low levels, while ethanol and propionic and lactic acids were not detected at all. The percentage of COD removal was approximately 97%, implying that the performance of the CSTR is not kinetically limited. This implies that it should be possible to reduce someTable 6 The characteristics of the CH4-CSTR at steady state accompanied by their standard deviation Characteristic

Value

pH Alkalinity, mg CaCO3/l Biogas production (l/d) % in CH4 Soluble COD (mg/l) Acetic acid (mg/l) Butyric acid (mg/l)

7.5 ± 0.1 6650 ± 50 1.14 ± 0.12 64 560 ± 170 50 ± 10 40 ± 5

G. Antonopoulou et al. / Bioresource Technology 99 (2008) 110–119 1600

biogas methane

Production rate, ml/d

1400 1200 1000 800 600 400 200 0 0

20

40

60

80

100

120

140

160

180

Time, d

Fig. 5. The evolution of biogas and methane in the CH4-CSTR throughout the experimental period.

what the HRT, without loss in performance. The methane production rate reached 0.73 l CH4 per day giving a yield of 4.9 l methane per l of influent. This latter corresponded to a yield of 29 l CH4/kg sweet sorghum, considering that 5 kg of sweet sorghum biomass were initially mixed with 30 l of water. The corresponding energy yield is 950 kJ/kg fresh and 3660 kJ/kg dry sorghum biomass, respectively (assuming that the energy yield from methane is 50120 kJ/kg). 3.5. Results of batch experiments for methane potential determination The methane production during the batch experiment for the determination of the methane potential of the remaining 45

with solids control without solids

40

CH4, ml

35 30 25 20 15 10 5 0 0

2

4

6

8

10

12

14

16

Time, d

Fig. 6. Methane production during the batch experiment for the determination of the methane potential of the solid fraction after the extraction process.

117

solids after the extraction process is shown in Fig. 6. The calculated methane potential of the solid residues, after subtracting the control values is 11 mmol CH4/g TS. This yield corresponds to the production of 78 l CH4/kg of sweet sorghum and to an energy yield of 2560 kJ/kg fresh and 9845 kJ/kg dry sorghum biomass, respectively. The methanogenic potential of the solid residues after the extraction process is much higher (more than double) than that of the H2-CSTR effluent. Moreover, a pre-treatment process aiming at the increase of the bioavailability of the solid organic matter could precede the anaerobic digestion of the solid fraction resulting at an even higher methane recovery. This issue is currently under investigation. 4. Conclusions The present study investigated the fermentative hydrogen production from the sugars of sweet sorghum extract as well as the subsequent methane production from the effluent of the hydrogenogenic process and the methane potential of the remaining solids after the extraction process. It has been proved that continuous fermentative hydrogen production from sweet sorghum extract is possible and stable using the indigenous microflora without a pre-heating step. The highest biogas and hydrogen production rate (6500 ml biogas/d and 2550 ml H2/d, respectively) was obtained at the HRT of 6 h while the highest yield of hydrogen produced per kg of sorghum biomass was achieved at the HRT of 12 h (10.4 l H2/kg sweet sorghum). In order to decide for the optimal HRT, for a full-scale plant, is necessary to take into account both economic (based on hydrogen production rate) and technical (based on hydrogen yield) aspects. Moreover, the present study showed that sweet sorghum extract could be used for hydrogen and methane production in a two-stage process. It has been proved that the effluent from the hydrogenogenic reactor is an ideal substrate for methane production with approximately 107 l CH4/kg sweet sorghum, 78 l of which come from the solid residues. Continuous methane production of the H2-CSTR effluent yielded 29 l CH4/kg of sweet sorghum, while the methanogenic potential of the solid residues after the extraction process yielded another 78 l CH4/kg sweet sorghum.

Table 7 Production rate of acetic, propionic and butyric acids and the theoretically calculated hydrogen production rate compared with the experimentally measured hydrogen production rate in the H2-CSTR HRT (h)

24 12 8 6 4

Production of acetic and butyric acids (mmol/d)

Hydrogen production (mmol/d)

Acetic acid

Propionic acid

Butyric acid

Theoretically calculated taking into account all acids production

Theoretically calculated taking into account only the butyric acid production

Experimentally measured

19.93 30.98 86.99 68.27 95.36

4.04 3.34 1.70 2.05 4.91

26.87 65.66 93.52 127.68 186.38

89.55 189.94 359.31 389.85 558.56

53.73 131.31 187.03 255.35 372.76

18.35 77.59 92.37 113.75 97.37

118

G. Antonopoulou et al. / Bioresource Technology 99 (2008) 110–119

To sum up, this work demonstrated that biohydrogen production can be very efficiently coupled with a subsequent step of methane production and that sweet sorghum could be an ideal substrate for a combined gaseous biofuels production. Acknowledgements The authors thank the General Secretariat for Research and Technology for the financial support of this work under ‘‘PENED 2001, 01ED390’’. References Afschar, A.S., Schaller, K., Schugerl, K., 1986. Continuous production of acetone and butanol with shear-activated Clostridium acetobutylicum. Appl. Microbiol. Biotechnol. 23 (5), 315–322. Ajayan, P.M., Zhou, O.Z., 2001. Applications of carbon nanotubes. Top. Appl. Phys. 80, 391–425. APHA, AWWA, WPCF, 1975. In: Franson, M.A. (Ed.). Standard Methods for the Examination of Water and Wastewater, American Public Health Association, Washington, DC. Asada, Y., Miyake, J., 1999. Photobiological hydrogen production. J. Biosci. Bioeng. 88 (1), 1–6. Atif, A.A.Y., Fakhru’l-Razi, A., Ngan, M.A., Morimoto, M., Iyuke, S.E., Veziroglou, N.T., 2005. Fed batch production of hydrogen from palm oil mill effluent using anaerobic microflora. Int. J. Hydrogen Energy 30, 1393–1397. Benemann, J., 1996. Hydrogen biotechnology: progress and prospects. Nat. Biotechnol. 14 (9), 1101–1103. Chen, C.C., Lin, C.Y., 2001. Start-up of anaerobic hydrogen producing reactors seeded with sewage sludge. Acta Biotechnol. 21 (4), 371–379. Chen, C.C., Lin, C.Y., Chang, J.S., 2001. Kinetics of hydrogen production with continuous anaerobic cultures utilizing sucrose as the limiting substrate. Appl. Microbiol. Biotechnol. 57 (1–2), 56–64. Chen, C.C., Lin, C.Y., Lin, M.C., 2002. Acid–base enrichment enhances anaerobic hydrogen production process. Appl. Microbiol. Biotechnol. 58 (2), 224–228. Christakopoulos, P., Li, L.W., Kekos, D., Makris, B.J., 1993. Direct conversion of sorghum carbohydrates to ethanol by a mixed microbial culture. Bioresource Technol. 45 (2), 89–92. Chum, H.L., Overend, R.P., 2001. Biomass and renewable fuels. Fuel Process. Technol. 71 (1–3), 187–195. Claassen, P.A.M., van Lier, J.B., Contreras, A.M.L., van Niel, E.W.J., Sijtsma, L., Stams, A.J.M., de Vries, S.S., Weusthuis, R.A., 1999. Utilisation of biomass for the supply of energy carriers. Appl. Microbiol. Biotechnol. 52 (6), 741–755. Cook, J., Beyea, J., 2000. Bioenergy in the United States: progress and possibilities. Biomass Bioenergy 18, 441–455. Dalianis, C., Panoutsou, C., Dercas, N., 1996. Sweet and fiber sorghum, two promising biomass crops. In: First European Seminar on Sorghum for Energy and Industry, Toulouse, France, 1–3 April, pp. 173–176. Demirbas, A., 2004. Combustion characteristics of different biomass fuels. Prog. Energy Combust. Sci. 30 (2), 219–230. Fang, H.H.P., Liu, H., 2002. Effect of pH on hydrogen production from glucose by a mixed culture. Bioresource Technol. 82 (1), 87–93. Gavala, H.N., Skiadas, I.V., Ahring, B.K., 2006. Biological hydrogen production in suspended and attached growth anaerobic reactor systems. Int. J. Hydrogen Energy 31 (9), 1164–1175. Gosse, G., 1996. Overview on the different routes for industrial utilization of sorghum. In: Abstracts Book, First European Seminar on Sorghum for Energy and Industry, Toulouse, France, 1–3 April, p. 2. Hawkes, F.R., Dinsdale, R., Hawkes, D.L., Hussy, I., 2002. Sustainable fermentative hydrogen production: challenges for process optimisation. Int. J. Hydrogen Energy 27 (11–12), 1339–1347.

International Energy Agency (IEA), 1998. World Energy Outlook, 1998 Edition, Available from: . Jackman, E.A., 1987. In: Bu’Lock, J., Kristiansen, B. (Eds.), Basic Biotechnology. Academic Press, New York, pp. 309–336. Joseffson, B., 1983. Rapid spectrophotometric determination of total carbohydrates. In: Grasshoff, K., Ehrhardt, M., Kremling, K. (Eds.), Methods of Seawater Analysis. Verlag Chemie GmbH, Berlin, pp. 340–342. Kapdan, I.K., Kargi, F., 2006. Bio-hydrogen production from waste materials. Enzyme Microbial. Technol. 38 (5), 569–582. Lay, J.J., 2000. Modelling and optimization of anaerobic digested sludge converting starch to hydrogen. Biotechnol. Bioeng. 68 (3), 269–278. Lay, J.J., Lee, Y.J., Noike, T., 1999. Feasibility of biological hydrogen production from organic fraction of municipal solid waste. Water Res. 33 (11), 2579–2586. Lee, Y.J., Miyahara, T., Noike, T., 2002. Effect of pH on microbial hydrogen fermentation. J. Chem. Technol. Biotechnol. 77 (6), 694– 698. Lezinou, V., Christakopoulos, P., Kekos, D., Macris, B.J., 1994. Simultaneous saccharification and fermentation of sweet sorghum carbohydrates to ethanol in a fed-batch process. Biotechnol. Lett. 16 (9), 983–988. Lezinou, V., Christakopoulos, P., Li, L.W., Kekos, D., Macris, B.J., 1995. Study of a single and mixed culture for the direct bio-conversion of sorghum carbohydrates to ethanol. Appl. Microbiol. Biotechnol. 43 (3), 412–415. Lin, C.Y., Chang, R.C., 1999. Hydrogen production during the anaerobic acidogenic conversion of glucose. J. Chem. Technol. Biotechnol. 74 (6), 498–500. Lin, C.Y., Chang, R.C., 2004. Fermentative hydrogen production at ambient temperature. Int. J. Hydrogen Energy 29 (7), 715–720. Mamma, D., Christakopoulos, P., Koullas, D., Kekos, D., Makris, B.J., Koukios, E., 1995. An alternative approach to the bioconversion of sweet sorghum carbohydrates to ethanol. Biomass Bioenergy 8 (2), 99– 103. Mamma, D., Koullas, D., Fountoukidis, G., Kekos, D., Makris, B.J., Koukios, E., 1996. Bioethanol from sweet sorghum: simultaneous saccharification and fermentation of carbohydrates by a mixed microbial culture. Process Biochem. 31 (4), 377–381. Mizuno, O., Dinsdale, R., Hawkes, F.R., Hawkes, D.L., Noike, T., 2000. Enhancement of hydrogen production from glucose by nitrogen gas sparging. Bioresource Technol. 73 (1), 59–65. Morimoto, M., 2002. Why is the anaerobic fermentation in the production of the biohydrogen attractive? In: The Proceedings of Conversion of Biomass into Bioenergy. Organized by New energy and Industrial Technology Development Organization (NEPO), Japan and Malaysian Palm oil Board (MPOP). Mu, Y., Zheng, X.J., Yu, H.Q., Zhu, R.F., 2006. Biological hydrogen production by anaerobic sludge at various temperatures. Int. J. Hydrogen Energy 31 (6), 780–785. Nandi, R., Sengupta, R., 1998. Microbial production of hydrogen: an overview. Crit. Rev. Microbiol. 24 (1), 61–84. Noike, T., Mizuno, O., 2000. Hydrogen fermentation of organic municipal wastes. Water Sci. Technol. 42 (12), 155–162. Parikka, M., 2004. Global biomass fuel resources. Biomass Bioenergy 27 (6), 613–620. Ramachandran, R., Menon, R.K., 1998. An overview of industrial uses of hydrogen. Int. J. Hydrogen Energy 23 (7), 593–598. Ren, N.Q., Wang, B.Z., Huang, J.C., 1997. Ethanol type fermentation from carbohydrate in high rate acidogenic reactor. Biotechnol. Bioeng. 54 (5), 428–433. Richards, B.K., Cummings, R.J., Jewell, W.J., 1991. High rate low solids methane fermentation of sorghum, corn and cellulose. Biomass Bioenergy 1 (5), 249–260. Sung, S., Raskin, L., Duangmanee, T., Padmasiri, S., Simmons, J.J., 2002. Hydrogen production by anaerobic microbial communities exposed to repeated heat treatments. In: Proceedings of the 2002 U.S. DOE Hydrogen Program Review.

G. Antonopoulou et al. / Bioresource Technology 99 (2008) 110–119 Ueno, Y., Otsuka, S., Morimoto, M., 1996. Hydrogen production from industrial wastewater by anaerobic microflora in chemostate culture. J. Ferment. Bioeng. 82 (2), 194–197. Zoetemeyer, R.J., Arnoldy, P., Cohen, A., Boelhouwer, C., 1982a. Influence of temperature on the anaerobic acidification of glucose on

119

a mixed culture forming part of a two-stage digestion process. Water Res. 16 (3), 313–321. Zoetemeyer, R.J., van den Heuvel, J.C., Cohen, A., 1982b. pH influence on acidogenic dissimilation of glucose in an anaerobic digester. Water Res. 16, 303–311.