0 évaluation0% ont trouvé ce document utile (0 vote)
13 vues8 pages
Ethanol production from syngas by Clostridium strain P11 using corn steep liquor as a nutrient replacement to yeast extract. About 32% more ethanol was produced with 20 g L A1 CSL media in 250-mL bottle fermentations compared to media with 1 g L A1 YE after 360 h. The use of CSL also enhanced butanol production by sevenfold compared to YE in bottle fermentations.
Ethanol production from syngas by Clostridium strain P11 using corn steep liquor as a nutrient replacement to yeast extract. About 32% more ethanol was produced with 20 g L A1 CSL media in 250-mL bottle fermentations compared to media with 1 g L A1 YE after 360 h. The use of CSL also enhanced butanol production by sevenfold compared to YE in bottle fermentations.
Ethanol production from syngas by Clostridium strain P11 using corn steep liquor as a nutrient replacement to yeast extract. About 32% more ethanol was produced with 20 g L A1 CSL media in 250-mL bottle fermentations compared to media with 1 g L A1 YE after 360 h. The use of CSL also enhanced butanol production by sevenfold compared to YE in bottle fermentations.
Ethanol production from syngas by Clostridium strain P11 using corn steep
liquor as a nutrient replacement to yeast extract
Prasanth Maddipati, Hasan K. Atiyeh
, Danielle D. Bellmer, Raymond L. Huhnke
Department of Biosystems and Agricultural Engineering, Oklahoma State University, Stillwater, OK 74078, USA a r t i c l e i n f o Article history: Received 20 December 2010 Received in revised form 15 March 2011 Accepted 16 March 2011 Available online 21 March 2011 Keywords: Clostridium Corn steep liquor Ethanol Syngas fermentation Yeast extract a b s t r a c t The feasibility of replacing yeast extract (YE) by corn steep liquor (CSL), a low cost nutrient source, for syngas fermentation to produce ethanol using Clostridium strain P11 was investigated. About 32% more ethanol (1.7 g L 1 ) was produced with 20 g L 1 CSL media in 250-mL bottle fermentations compared to media with 1 g L 1 YE after 360 h. Maximum ethanol concentrations after 360 h of fermentation in a 7.5-L fermentor with 10 and 20 g L 1 CSL media were 8.6 and 9.6 g L 1 , respectively, which represent 57% and 60% of the theoretical ethanol yields from CO. Only about 6.1 g L 1 of ethanol was obtained in the medium with 1 g L 1 YE after 360 h, which represents 53% of the theoretical ethanol yield from CO. The use of CSL also enhanced butanol production by sevenfold compared to YE in bottle fermenta- tions. These results demonstrate that CSL can replace YE as the primary medium component and signif- icantly enhance ethanol production by Clostridium strain P11. 2011 Elsevier Ltd. All rights reserved. 1. Introduction The dominant biofuel in the US is ethanol made via yeast-based fermentation of corn starch, which is a well-developed technology. Ethanol made via fermentation of synthesis gas (syngas) produced from gasication of non-edible feedstocks such as biomass and municipal solid wastes is relatively a new technology. Syngas, primarily containing CO, CO 2 , and H 2 , can be fermented to ethanol and acetic acid by acetogenic organisms such as Clostridium ljung- dahlii (Klasson et al., 1992; Phillips et al., 1994; Younesi et al., 2005), Clostridium autoethanogenum (Abrini et al., 1994), Clostrid- ium carboxidivorans P7 (Ahmed and Lewis, 2007; Rajagopalan et al., 2002), and Clostridium strain P11 (Huhnke et al., 2010; Saxena, 2008). These microorganisms utilize the reductive acetyl- CoA pathway, also known as the WoodLjungdahl pathway, for the synthesis of acetyl-CoA, conservation of energy and growth and production of acetic acid and ethanol (Wood et al., 1986). The overall reactions for ethanol and acetic acid production from H 2 and CO are (Vega et al., 1989): 6CO 3H 2 O !C 2 H 5 OH4CO 2 1 6H 2 2CO 2 !C 2 H 5 OH3H 2 O 2 4CO2H 2 O !CH 3 COOH2CO 2 3 2CO 2 4H 2 !CH 3 COOH2H 2 O 4 When equal moles of CO and H 2 are supplied, theoretically two- thirds of the carbon from CO can be converted to ethanol and the remaining carbon is accounted for in CO 2 production as seen by combining Eqs. (1) and (2). However, all the carbon supplied in the form of CO theoretically can be converted to acetic acid by combining Eqs. (3) and (4). Butanol production has also been reported from CO (Shen et al., 1999; Worden et al., 1991) or a mixture of CO, CO 2 and H 2 (Babu et al., 2010; Datar et al., 2004; Liou et al., 2005; Rajagopalan et al., 2002) by only three bacteria. Butyribacterium methylotrophi- cum was reported to produce between 0.3 and 1.4 g L 1 butanol from pure CO (Shen et al., 1999; Worden et al., 1991). However, Clostridium strain P11 produced 0.3 g L 1 butanol from a simulated syngas mixture (Babu et al., 2010) and C. carboxidivorans produced 1.3 g L 1 butanol from syngas derived from cellulosic feedstock (Rajagopalan et al., 2002). The overall reactions for butanol produc- tion from H 2 and CO are (Rajagopalan et al., 2002): 12CO 5H 2 O !C 4 H 9 OH8CO 2 5 12H 2 4CO 2 !C 4 H 9 OH7H 2 O 6 A recent literature review on biomass-derived syngas fermentation into biofuel describes the factors that affect syngas 0960-8524/$ - see front matter 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.biortech.2011.03.047
Corresponding author. Address: Department of Biosystems and Agricultural
Engineering, 227 Ag Hall, Oklahoma State University, Stillwater, OK 74078, USA. Tel.: +1 405 744 8397; fax: +1 405 744 6059. E-mail address: hasan.atiyeh@okstate.edu (H.K. Atiyeh). Bioresource Technology 102 (2011) 64946501 Contents lists available at ScienceDirect Bioresource Technology j our nal homepage: www. el sevi er . com/ l ocat e/ bi or t ech fermentation, which include type of biocatalyst and growth media, mass transfer, reactor conguration and operating conditions (Munasinghe and Khanal, 2010). Composition of the fermentation medium plays an important role in cell and product yields. The development of a low-cost fer- mentation medium alternative containing all essential nutrients required for cell growth and product formation would reduce the overall cost of the fermentation process. Standard mediumfor Clos- tridium strain P11 is composed of yeast extract, vitamins, minerals, trace metals and reducing agent (Saxena and Tanner, 2010). Apart from the reducing agent, yeast extract (YE) is the most expensive component. Some inexpensive nutrients that could replace YE are corn steep liquor (CSL), hydrolyzed cotton seed our, hydrolyzed soy our and ethanol stillage (Witjitra et al., 1996). CSL is a major by-product fromthe cornwet milling industry. It is rich in vitamins, minerals, amino acids and proteins (Azeredo et al., 2006; Kadamand Newman, 1997). It has been used as growth med- ium for mannitol production by Lactobacillus intermedius (Racine and Saha, 2007), and solvent productionby Saccharomyces cerevisiae (Kadam and Newman, 1997), Clostridium beijerinckii (Parekh et al., 1999) and Zymomonas mobilis (Lawford and Rousseau, 1997). The cost of bacto-yeast extract from Difco Laboratories (Detroit, MI) is $157.49 kg 1 (Racine and Saha, 2007), while the cost of CSL on an industrial scale is $0.07 kg 1 (Lawford and Rousseau, 1997). As per 2010, the industrial price of spray dried YE is $9.2 kg 1 , while the cost of industrial CSL is $0.18 kg 1 . The incorporation of CSL at 20 g L 1 in the fermentation medium resulted in elimination of a few growth factors from standard Clostridium strain P11 medium (Saxena, 2008). However, this study did not investigate growth and product kinetics of strain P11 in CSL media. In addition, tests were done in 160-mL serumbottles with 10 mL of mediumand only CO was fed to strain P11, which was different from the syngas composition that potentially will be used for ethanol production on a large scale. Corn steep liquor (CSL) is selected in the present study as an alternative to yeast extract (YE) for syngas fermentation because it is rich in nutrients and lower in cost compared to YE. Accord- ingly, the main objective of this study was to determine whether CSL is a potential alternative nutrient source to YE in syngas fer- mentation medium and to evaluate growth and product kinetics of Clostridium strain P11 using CSL as compared to YE. 2. Methods 2.1. Microorganism Clostridium strain P11 (ATCC PTA-7826) provided by Dr. Ralph Tanner, University of Oklahoma, was maintained on standard yeast extract medium. Clostridium strain P11 is highly sensitive to O 2 ; and hence, the fermentation was performed under strict anaerobic conditions. In order to reduce the lag phase and also to ensure that viable cells were inoculated into the medium, cells from the stock inoculum were passaged twice prior to inoculation. Cell passaging and syngas fermentation were performed in 250-L serum bottles with 100 mL working volume. After cell passaging, the subculture at 10% (v/v) was transferred to the fermentation serum bottles and the 7.5-L fermentor to follow syngas fermentation. 2.2. Fermentation media Three media formulations developed by Saxena (2008) were used (Table 1). The standard medium contained yeast extract, min- erals, trace metals, vitamins, morpholinoethanesulfonic acid (MES), resazurin and cysteine sulde. The second medium (10 g L 1 CSL) had the same components as the standard medium except yeast extract was replaced withCSL. The third medium(20 g L 1 CSL) only contained trace metals, MES, resazurin, ammonium chloride and cysteine sulde. Yeast extract (Difco Laboratories, Detroit, MI) and CSL (SigmaAldrich, St. Louis, MO) were used as complex nitrogen and nutrient sources in the fermentation media. The compositions of minerals, vitamins and trace metal stock solutions are shown in Table 2. Unless stated, all media components were purchased from SigmaAldrich (St. Louis, MO). In order to produce consistent results, CSL from the same batch was used throughout the study. CSL contained about 50% solids. Before the addition of CSL into the fermentation medium, the sol- ids from crude CSL were removed by centrifugation at 13,000 rpm for 10 min. Only the liquid portion of the crude CSL was used in the syngas fermentation experiments to avoid interference of solids with measurement of P11 growth. The presence or removal of solids from CSL did not affect ethanol and acetate formation by strain P11 (Saxena, 2008). In addition, removal of solids from CSL did not affect the nutritional requirements of Zymomonas ethanol Table 1 Composition of media formulations used for syngas fermentation (Saxena, 2008). Standard yeast extract CSL a CSL a 1 g L 1 10 g L 1 20 g L 1 Yeast extract (g) 1 Corn steep liquor (g) 10 20 Minerals (mL) 30 30 Trace metals (mL) 10 10 10 Vitamins (mL) 10 10 MES b (g) 10 10 10 0.1% Resazurin (mL) 1 1 1 Ammonium chloride (g) 1.25 4% Cysteine sulde (mL) 10 10 2.5 Total volume (L) 1 1 1 a CSL = corn steep liquor. b MES = morpholinoethanesulfonic acid. Table 2 Composition of trace metal, vitamin and mineral stock solutions (Huhnke et al., 2010). Trace metal stock solution g L 1 Cobalt chloride 0.20 Ferrous ammonium sulfate 0.80 Manganese sulfate 1.00 Nickel chloride 0.20 Nitrilotriacetic acid 2.00 Sodium molybdate 0.02 Sodium selenate 0.10 Sodium tungstate 0.20 Zinc sulfate 1.00 Mineral stock solution g L 1 Ammonium chloride 100 Calcium chloride 4 Magnesium sulfate 20 Potassium chloride 10 Potassium phosphate 10 Sodium chloride 80 Vitamin stock solution mg L 1 p-(4)-Aminobenzoic acid 5 d-Biotin 2 Calcium pantothenate 5 Folic acid 2 MESNA a 10 Nicotinic acid 5 Pyridoxine 10 Riboavin 5 Thiamine 5 Thioctic acid 5 Vitamin B12 5 a MESNA = mercaptoethanesulfonic acid. P. Maddipati et al. / Bioresource Technology 102 (2011) 64946501 6495 fermentations (Lawford and Rousseau, 1997). The liquid portion of crude CSL contained 75 g L 1 of total sugars (43 g L 1 maltose, 11 g L 1 glucose, 5 g L 1 xylose, 6 g L 1 galactose, 7 g L 1 arabinose and 3 g L 1 mannose). The initial total sugar concentrations in the 10 and 20 g L 1 CSL were 0.6 and 1.2 g L 1 , respectively. Resazurin solution (0.1%) was added into the medium as a redox indicator. Morpholinoethanesulfonic acid (MES) was added as a biological buffer to prevent excessive uctuations in the pH during fermenta- tion. The initial pH of the medium was adjusted to 6.1 using 2 N KOH before inoculation. 2.3. Syngas Commercial syngas with gas composition similar to producer gas generated from our gasication facility was used. The syngas was composed of 20% CO, 15% CO 2 , 5% H 2 and 60% N 2 by volume. 2.4. Fermentation in 250-mL serum bottles Syngas fermentations were conducted in 250-mL serum bottles (Wheaton, NJ) with 100 mL working volume containing the three different media described in Table 1. All media components except reducing agent were mixed in deionized water and pH was ad- justed to 6.1 using 2 N KOH. The media were boiled and then purged with N 2 for 3 min to remove O 2 . The serum bottles were sealed using gas impermeable butyl rubber septum-type stoppers and aluminum crimp seals. One milliliter of 4% cysteine sulde solution was added to each bottle to scavenge any remaining dis- solved O 2 . The fermentation media were autoclaved (Primus Ster- ilizer Co., Inc., Omaha, NE) at 121 C and 205 kPa for 20 min. The medium was inoculated with 10% (v/v) of Clostridium strain P11. Sterile 0.2 lm PTFE (polytetrauoroethylene) membrane l- ters (VWR International, West Chester, PA) were used in the inlet line that fed syngas to strain P11 at 239 kPa. The bottles were placed vertically on an orbital shaker (Innova 2100, New Bruns- wick Scientic, Edison, NJ) at 150 rpm and 37 C. Liquid samples were withdrawn every 24 h from the bottles under aseptic condi- tions to measure optical density at 660 nm, pH and acetic acid and solvent concentrations. Gas in the head space in each bottle was replaced every 24 h with fresh syngas at 239 kPa. All studies were performed in triplicate. 2.5. Fermentation in 7.5-L fermentor A 7.5-L Bioo 110 (New Brunswick Scientic Co., Edison, NJ) with a 3 L working volume was used. Agitation was provided by three six-blade Rushton turbine impellers of 59 mm diameter located at an equal distance on a top-driven impeller shaft. Four bafes were symmetrically arranged to avoid vortex formation of liquid media and improve mixing. A microsparger (New Brunswick Scientic Co., Edison, NJ) with a pore size of 1015 lm made of 316 SS material was used for gas sparging. Syngas owing out of the fermentor passed through a condenser cooled at 5 C using a refrig- erated circulator (1156 D, VWR International, West Chester, PA). Condensed vapor was returned to the fermentor. Uncondensed vapor from the condenser with the syngas was bubbled through a water trap lled with 150 mL deionized water maintained at 10 C using chilled water. The water trap captured ethanol that was not condensed in the condenser. The amount of ethanol col- lected in the water trap was included in the total ethanol produced after 360 days of syngas fermentation. BioCommand software was used for data acquisition, monitoring and controlling the fermentor. All media components in Table 1 except cysteine sulde were added into the fermentor to a total volume of 3 L. The pH of the media was adjusted to 6.1 before sterilization using 5 N KOH. The fermentor containing the medium was autoclaved at 121 C and 205 kPa for 20 min. The medium was allowed to cool to room temperature and was then purged with N 2 at 0.2 L min 1 for 24 h to remove any dissolved O 2 before the start of the fermentation. Then, syngas (20% CO, 15% CO 2 , 5% H 2 and 60% N 2 by volume) was piped to the fermentor at 0.15 L min 1 and cysteine sulde solution was added as per Table 1. The medium was then inocu- lated with 10% (v/v) of strain P11 cells. The pressure inside the fer- mentor was controlled at 143 kPa using a back pressure regulator. The temperature in the fermentor was maintained at 37 C and the agitation speed was set at 150 rpm. Antifoam B Emulsion (SigmaAldrich, Saint Louis, MO) at a concentration of 10% (v/v) was used to control foam formation inside the fermentor. About 0.2 mL of sterilized antifoam solution was added when the height of the foam in the fermentor was two inches. Gas samples were withdrawn from the head space periodically to determine gas com- position. Liquid samples were collected to measure optical density at 660 nm, pH and acetic acid and solvent concentrations. 2.6. Analytical procedures 2.6.1. Cell concentration Cell concentration was determined using a UVVis spectropho- tometer (Varian Inc., Palo Alto, CA) at 660 nm. Samples with OD values above 0.4 were diluted so that the OD was within the linear range of the calibration curve (cell mass, g L 1 = 0.396 OD 0.0521) as established using the dry weight method (Panneerselvam, 2009). 2.6.2. Solvent analysis Liquid samples were centrifuged at 13,000 rpm for 13 min. The supernatant was ltered through 0.45-lm nylon membrane lters (VWR International, West Chester, PA). Ethanol, acetic acid and butanol concentrations were analyzed using a gas chromatograph (GC), (Agilent Technologies, Wilmington, DE). A PoraPak QS 80/ 100 (Alltech, Deereld, IL) packed column connected to a ame ionization detector (FID) was used. The GC was operated under iso- thermal conditions with the oven temperature set at 210 C. He- lium was the carrier gas with a ow rate of 0.025 L min 1 . The resulting chromatograms were analyzed using ChemStation data analysis software. 2.6.3. Gas analysis A GC equipped with a thermal conductivity detector (TCD) was used for gas analysis. A capillary column, Carboxen 1010 PLOT (Supelco, Bellefonte, PA), was used to detect CO, CO 2 , H 2 and N 2 . Ar- gon was used as a carrier gas in the GC with an initial gas ow rate of 0.4 mL min 1 for the rst 12 min, and then it was increased at a rate of 0.1 mL min 1 until reaching 0.8 mL min 1 . Oven tempera- ture was set at 32 C for 12 min, after which the temperature was increased at a rate of 30 C per min until it reached 236 C. The temperatures of the column inlet and detector were set at 200 and 230 C, respectively. Resulting chromatograms were ana- lyzed using ChemStation data analysis software. The GC was cali- brated regularly by injecting samples with different volumes of known gas concentrations. Error in CO, CO 2 , H 2 and N 2 measure- ments was below 5%. 2.6.4. Oxidationreduction potential The oxidationreduction potential (ORP) was measured using a redox probe with an Ag/AgCl reference electrode (InPro3250/325/ PT100, Mettler Toledo, Columbus, OH) connected to a multi- parameter M400 transmitter (Mettler Toledo, Columbus, OH). The redox potential value from Ag/AgCl reference electrode was calculated vs. standard hydrogen electrode (SHE) by the addition of +222.4 mV to the value obtained by the Ag/AgCl electrode (Kakiuchi et al., 2007). 6496 P. Maddipati et al. / Bioresource Technology 102 (2011) 64946501 2.6.5. Sugar analysis The sugar content of the liquid portion of crude CSL was mea- sured using high pressure liquid chromatography (HPLC) (Agilent 1100 series, Wilmington, DE) with refractive index detector. The column used was Aminex HPX 87P (Bio-Rad, Sunnyvale, CA, USA), which was operated at 85 C with de-ionized water as the mobile phase pumped at 0.6 mL min 1 for 30 min per sample. 2.6.6. Statistical analysis Analysis of variance (ANOVA) was determined using the GLM procedure of SAS Release 9.2 (Cary, NC). A Dunnetts test (Dunnett, 1955) at 95% condence level was used to determine the statistical signicance of the fermentation parameters (pH, cell mass, etha- nol, acetic acid and butanol) between the three different media used in 250-mL fermentation bottles. Signicance level was tested at p = 0.05. 3. Results and discussion 3.1. Fermentation in 250 mL serum bottles 3.1.1. Cell growth and pH proles Cell mass concentrations of Clostridium strain P11 increased during the rst 48 h in the media that contained 1 g L 1 YE and 10 g L 1 CSL (Fig. 1). However, cells were in the exponential phase in the rst 24 h in the medium with 20 g L 1 CSL. No lag phase was observed in any of the media used. In addition, cells entered the stationary phase in all media after 48 h. Minor changes in cell mass concentrations were observed during the stationary phase until 360 h. The specic growth rates of strain P11 in 1 g L 1 YE, 10 g L 1 CSL and 20 g L 1 CSL media were 0.059, 0.049 and 0.076 h 1 , respectively. The specic growth rate in the medium with 20 g L 1 CSL was 29% and 55% higher than in 1 g L 1 YE and 10 g L 1 CSL media, respectively. Though the specic growth rate of strain P11 in the 1 g L 1 YE medium was 20% higher than in the 10 g L 1 CSL medium, about 84% more cells were produced in the 10 g L 1 CSL medium (Fig. 1). This could be due to the presence of additional nutrients such as sugars, water soluble vitamins, ami- no acids, minerals and trace metals in CSL (Liggett and Kofer, 1948). Statistical analysis indicated that the cell mass concentra- tions in 10 and 20 g L 1 CSL media between 24 and 360 h were sig- nicantly higher than in the 1 g L 1 YE medium, conrming that nutrients present in CSL signicantly enhanced cell growth (p < 0.05). Also, cell mass concentrations in 10 g L 1 CSL between 24 and 264 h were signicantly lower than in the 20 g L 1 CSL medium (p < 0.05). However, the difference in cell mass concentra- tion between the two CSL media after 264 h of fermentation was insignicant (p > 0.05). The pH of the YE and CSL media decreased during cell growth due to the production of acetic acid (Fig. 1). The pH in the 1 g L 1 YE medium decreased from 6.1 to 4.8 at 168 h, while the pH in the 10 g L 1 CSL medium decreased from 6.0 to 5.1 at 240 h, at which time the acetic acid concentrations in these media reached a maximum (Fig. 2). Subsequently, an increase in pH to 5.1 and 5.3 was observed after 360 h in the 1 g L 1 YE and 10 g L 1 CSL media, respectively. This correlated with a decrease in acetic acid concentration. The lowest pH value in the 20 g L 1 CSL medium was 5.4 after 48 of fermentation, after which the pH started to in- crease due to acetic acid consumption during syngas fermentation (Fig. 2). The pH in the 20 g L 1 CSL mediumwas signicantly higher than in 1 g L 1 YE and 10 g L 1 CSL media after 72 h (p < 0.05). Also, the pH in the 10 g L 1 CSL medium was signicantly higher com- pared to 1 g L 1 YE medium between 72 and 264 h (p < 0.05). 3.1.2. Product proles Acetic acid and ethanol were the major products formed during syngas fermentation in both YE and CSL media (Fig. 2). Acetic acid is a primary metabolite, and its production is associated with cell growth. The production of acetic acid was higher during the growth phase and gradually decreased when the cells entered the stationary phase. A maximum acetic acid concentration of 2.6 g L 1 was reached at 144 h in the 1 g L 1 YE medium (Fig. 2). Acetic acid concentration then decreased to 2.3 g L 1 at the end of the fermentation. Acetic acid concentrations were signicantly lower between 96 and 168 h and from 96 to 360 h in the 10 and 20 g L 1 CSL media, respectively, compared to 1 g L 1 YE medium (p < 0.05). The difference in acetic acid concentrations between 10 g L 1 CSL and 1 g L 1 YE media was insignicant after 192 h (p > 0.05). Consumption of acetic acid by strain P11 was observed at later stages of fermentation in the three media. More acetic acid was consumed in the 20 g L 1 CSL medium compared to the other two media. Acetic acid consumption was associated with produc- tion of ethanol. This was also observed in other syngas fermenta- tion studies using Clostridium strain P11 (Frankman, 2009; Panneerselvam et al., 2010; Saxena, 2008) and C. carboxidivorans P7 (Hurst and Lewis, 2010). Ethanol production started in the stationary phase (Fig. 2). The rate of ethanol production increased in the early stationary phase and decreased with the decline in cell concentration. The nal eth- anol concentration in the 20 g L 1 CSL medium was 1.7 g L 1 , which was 31% and 21% higher than in 1 g L 1 YE and 10 g L 1 CSL media, respectively. In addition, the amount of ethanol 0.0 0.1 1.0 4.5 5.0 5.5 6.0 6.5 7.0 0 48 96 144 192 240 288 336 384 C e l l
m a s s
( g
L - 1 ) Time (h) p H Fig. 1. Cell mass (solid symbols) and pH (open symbols) proles using Clostridium strain P11 in 250-mL serum bottles: (s) 1 g L 1 YE medium, (h) 10 g L 1 CSL medium, (D) 20 g L 1 CSL medium (n = 3). 0.0 1.0 2.0 3.0 4.0 0.0 1.0 2.0 3.0 0 48 96 144 192 240 288 336 384 E t h a n o l
( g / L ) Time (h) A c e t i c a c i d
( g / L ) Fig. 2. Acetic acid (solid symbols) and ethanol (open symbols) proles using Clostridium strain P11 in 250-mL serum bottles: (s) 1 g L 1 YE medium, (h) 10 g L 1 CSL medium, (D) 20 g L 1 CSL medium (n = 3). P. Maddipati et al. / Bioresource Technology 102 (2011) 64946501 6497 produced in the 20 g L 1 CSL medium was signicantly higher compared to 1 g L 1 YE and 10 g L 1 CSL media from the beginning of the fermentation until 240 h (p < 0.05). Higher ethanol produc- tion in 20 g L 1 CSL can also be due to the presence of more nutri- ents compared to 1 g L 1 YE and 10 g L 1 CSL media. The difference in ethanol concentrations between the 1 g L 1 YE and 10 g L 1 CSL media was statistically insignicant during the entire fermentation (p > 0.05). Ethanol productivities of Clostridium strain P11 were 3.6, 4.0 and 4.7 mg L 1 h 1 in 1 g L 1 YE, 10 g L 1 CSL and 20 g L 1 CSL media, respectively. Butanol was also produced during syngas fermentation in YE and CSL media (Fig. 3). Butanol production started after 24 and 72 h in the 10 and 20 g L 1 CSL media, respectively. After 96 h, butanol was observed in the 1 g L 1 YE medium. The nal concen- trations of butanol were below 0.1 and 0.6 g L 1 in the YE and CSL media, respectively. Statistical analysis indicated that CSL has sig- nicant effect on butanol production by strain P11. Over sevenfold more butanol was produced in the 10 and 20 g L 1 CSL media com- pared to the 1 g L 1 YE medium, which was statistically signicant (p < 0.05). The difference in butanol concentrations in the 10 and 20 g L 1 CSL media after 360 h was insignicant (p > 0.05). The amount of butanol produced by Clostridium strain P11 in 20 g L 1 CSL medium using 20% CO, 15% CO 2 , 5% H 2 and 60% N 2 gas mixture was two times lower than with C. carboxidivorans using 25% CO, 15% CO 2 and 60% N 2 gas mixture (Rajagopalan et al., 2002) and B. methylotrophicum using pure CO (Worden et al., 1991), respec- tively. The compositions of the gas mixture and media used have a large effect on the amount of solvents produced during fermentation. 3.2. Fermentation in 7.5-L fermentor After studying the growth and product kinetics of Clostridium strain P11 in 250-mL serum bottles, it was necessary to scale-up the process to better understand and evaluate the kinetics in a pro- totype bench top stirred tank reactor. Therefore, syngas fermenta- tions by Clostridium strain P11 with the three media compositions used in bottle fermentations were investigated in a 7.5-L stirred tank reactor. Fermentation runs were not conducted in triplicate due to limited time and available resources. 3.2.1. Cell growth and pH proles The growth and pH proles of strain P11 in 1 g L 1 YE, 10 g L 1 CSL and 20 g L 1 CSL media are shown in Fig. 4. Similar fermenta- tion proles were noticed in all three media. There was a short lag phase followed by an exponential phase. The cell concentrations in all three media increased exponentially during the rst 48 h of fermentation. The specic growth rates of strain P11 were 0.055, 0.061 and 0.075 h 1 in the 1 g L 1 YE, 10 g L 1 CSL and 20 g L 1 CSL media, respectively (Table 3) . The specic growth rates of C. ljungdahlii (Phillips et al., 1994), and Clostridium strain P11 (Frank- man, 2009) in yeast extract medium were reported as 0.06 and 0.0491 h 1 , respectively. These specic growth rates are lower than those with strain P11 in the 20 g L 1 CSL medium. Similar to the results obtained in bottle fermentations, more cells were produced in media that contained CSL. The maximum cell concentrations in 1 g L 1 YE, 10 g L 1 CSL and 20 g L 1 CSL media were 0.53, 0.55 and 0.74 g L 1 , respectively. About 39% and 35% more cells were produced in the 20 g L 1 CSL medium compared to 1 g L 1 YE and 10 g L 1 CSL media, respectively. This conrms the advantage of using CSL in the fermentation medium compared to YE. The drastic decrease in cell concentration in 10 g L 1 CSL medium after 264 h was due to foam formation, which caused cells to stick onto the wall of the fermentor. Foam forma- tion in the medium was due to the proteins present in the medium and proteins released from dead cells (Doran, 2006). In the present study, the cell mass yield (g cells produced per g CO consumed) calculated at the maximum cell mass concentrations in 1 g L 1 YE medium was about 15% higher than in both CSL media (Table 3). This is due to the consumption of more CO by strain P11 at the times at which the maximum cell mass concentrations were reached in 10 g L 1 CSL (192 h) and 20 g L 1 CSL media (240 h) compared to 1 g L 1 YE medium (168 h) as shown in Figs. 4 and 6a. The pH proles for the fermentation in the three media used were generally similar. The pH of the 1 g L 1 YE and 10 g L 1 CSL media decreased from 6 to 4.7 during cell growth in the rst 72 h due to the production of acetic acid (Fig. 4). However, the pH in the 20 g L 1 CSL medium decreased from 6 to 4.7 after 0.0 0.2 0.4 0.6 0.8 0 48 96 144 192 240 288 336 384 Time (h) B u t a n o l
( g / L ) Fig. 3. Butanol proles using Clostridium strain P11 in 250-mL serum bottles: (s) 1 g L 1 YE medium, (h) 10 g L 1 CSL medium, (D) 20 g L 1 CSL medium (n = 3). 0.0 0.1 1.0 4.5 5.0 5.5 6.0 6.5 7.0 0 48 96 144 192 240 288 336 384 C e l l
m a s s
( g
L - 1 ) Time (h) p H Fig. 4. Cell mass (solid symbols) and pH (open symbols) proles using Clostridium strain P11 in 7.5-L fermentor: (s) 1 g L 1 YE medium, (h) 10 g L 1 CSL medium, (D) 20 g L 1 CSL medium. Table 3 Fermentation characteristics of Clostridium strain P11 in yeast extract (YE) and corn steep liquor (CSL) media in 7.5-L fermentor. Medium YE CSL CSL 1 g L 1 10 g L 1 20 g L 1 Fermentation time (h) 360 360 360 l max (h 1 ) 0.055 0.061 0.075 Cell mass yield a (g cells g 1 CO consumed) 0.0055 0.0047 0.0048 Acetic acid (g L 1 ) 1.5 2.4 3.4 Ethanol (g L 1 ) 6.1 8.6 9.6 Ethanol productivity (mg L 1 h 1 ) 16.9 23.9 26.7 Ethanol yield from CO (% of theoretical value) 53 57 60 Conversion efciency of CO (%) 11 15 15 Conversion efciency of H 2 (%) 5 7 5 a Values were calculated at maximum cell mass concentrations. 6498 P. Maddipati et al. / Bioresource Technology 102 (2011) 64946501 120 h, during which cells were still growing. Then, the pH in all three media slightly increased due to consumption of acetic acid (Fig. 5). The difference in pH proles for 1 g L 1 YE and 10 g L 1 CSL media was smaller compared to the 20 g L 1 CSL medium after 144 h of fermentation. 3.2.2. Product proles Similar to bottle fermentations, acetic acid and ethanol produc- tion in all media with Clostridium strain P11 was observed during the acidogenic and solventogenic phases, respectively. However unlike bottle fermentations, no butanol was produced by strain P11 in the 7.5 L fermentor. The focus of the present study was not on butanol production. In addition, the results showed that the type of media and reactor conguration used has a large effect on strain P11 ability to produce butanol. Acetic acid concentration increased in the rst 72 h in both 1 g L 1 YE and 10 g L 1 CSL media and during the rst 120 h in the 20 g L 1 CSL medium (Fig. 5). Acetic acid formation by strain P11 was growth related, which resulted in a decrease in the pH of the fermentation medium (Fig. 4). The maximum acetic acid concentrations in 1 g L 1 YE, 10 g L 1 CSL and 20 g L 1 CSL media were 3.9, 4.2 and 4.0 g L 1 , respectively. After the concentration of acetic acid reached a maximum, strain P11 started to consume it. However, the consumption of acetic acid was the highest in the 1 g L 1 YE medium (Fig. 5). About 60% and 44% of the acetic acid was consumed between 72 and 360 h in 1 g L 1 YE and 10 g L 1 CSL media, respectively (Fig. 5). Only 16% of the acetic acid in the 20 g L 1 CSL medium was consumed between 120 and 360 h. In bottle fermentations, about 57% of the acetic acid was consumed in the 20 g L 1 CSL medium compared to 12% in the 1 g L 1 YE medium (Fig. 2). The switch to ethanol production was observed during deceler- ated growth and stationary phases when the pH was around 4.7 (Figs. 2 and 5). Ethanol formation by strain P11 in both bottle fer- mentation and 7.5-L fermentor was non-growth associated. The partial pressures of CO in the headspace in bottle fermentations and 7.5-L fermentor were 48 and 29 kPa, respectively. It was re- ported that ethanol production by C. carboxidivorans P7 was non- growth related when the partial pressure of CO in the headspace was below 106 kPa (Hurst and Lewis, 2010). However, ethanol for- mation was growth associated with C. carboxidivorans P7 when the partial pressure of CO in the headspace was above 106 kPa. In the present study, the rate of ethanol production in all media was high in the early stationary phase and decreased with time especially with the decline in cell mass concentration. After 360 h of fermen- tation, ethanol concentrations in 1 g L 1 YE, 10 g L 1 CSL and 20 g L 1 CSL media were 6.1, 8.6 and 9.6 g L 1 , respectively (Table 3). This includes ethanol in the 7.5-L fermentor and collected in the water trap after the condenser. The volumetric ethanol produc- tivities of Clostridium strain P11 in 10 g L 1 CSL and 20 g L 1 CSL media were 41% and 58% higher than in 1 g L 1 YE medium, respec- tively (Table 3). About 3 g L 1 of ethanol was formed after 504 h in a 5-L fermentor using strain P11 and cotton seed extract (CSE) as a medium component (Kundiyana et al., 2010). In the present study, ethanol productivities in 10 and 20 g L 1 CSL media (Table 3) were over fourfold higher than previously reported for CSE media. Ethanol formation was noticed when the ranges of redox poten- tial (ORP) in 1 g L 1 YE, 10 g L 1 CSL and 20 g L 1 CSL media were 206 to 118 mV SHE, 150 to 45 mV SHE and 121 to 75 mV SHE, respectively. The ORP range for ethanol production with strain P11 in cotton seed extract medium was from 240 to 130 mV SHE (Kundiyana et al., 2010). In the present study, the presence of more nutrients in the 20 g L 1 CSL medium could have contributed to the higher perfor- mance of strain P11 compared to 1 g L 1 YE medium. The initial to- tal sugar concentrations in the 10 and 20 g L 1 CSL media were 0.6 and 1.2 g L 1 , respectively. Among the sugars (maltose, glucose, xy- lose, galactose, arabinose and mannose) in CSL media, only glucose was utilized by strain P11 after 360 h of syngas fermentation. The initial glucose concentrations in the 10 and 20 g L 1 CSL media were 0.12 and 0.32 g L 1 , respectively. Only 10% and 76% of the ini- tial amounts of glucose in 10 and 20 g L 1 CSL media were con- sumed after 360 h, respectively. The maximum possible amount of ethanol production assuming 100% conversion efciency of glu- cose to ethanol in the 10 and 20 g L 1 CSL media are 0.06 and 0.16 g L 1 , respectively. These amounts of ethanol that can be pro- duced from the sugars present in CSL are negligible compared to 8.6 and 9.6 g L 1 of ethanol produced from syngas in 10 and 20 g L 1 CSL media, respectively (Table 3). This conrms that prac- tically all of the ethanol formed in the 7.5-L fermentor with 10 and 20 g L 1 CSL media was produced by strain P11 using syngas. 0.0 0.2 0.4 0.6 0.8 1.0 0.0 2.0 4.0 6.0 0 48 96 144 192 240 288 336 384 C O
( m o l ) Time (h) H 2 ( m o l ) -0.5 0.0 0.5 1.0 1.5 2.0 2.5 0 48 96 144 192 240 288 336 384 C O 2 ( m o l ) Time (h) b a Fig. 6. (a) Cumulative CO (solid symbols) and H 2 (open symbols) consumed; (b) cumulative CO 2 produced (open symbols) by Clostridium strain P11 in 7.5-L fermentor: (s) 1 g L 1 YE medium, (h) 10 g L 1 CSL medium, (D) 20 g L 1 CSL medium. 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 0.0 1.0 2.0 3.0 4.0 5.0 0 48 96 144 192 240 288 336 384 E t h a n o l
( g
L - 1 ) Time (h) A c e t i c a c i d
( g
L - 1 ) Fig. 5. Acetic acid (solid symbols) and ethanol (open symbols) proles using Clostridium strain P11 in 7.5-L fermentor: (s) 1 g L 1 YE medium, (h) 10 g L 1 CSL medium, (D) 20 g L 1 CSL medium. P. Maddipati et al. / Bioresource Technology 102 (2011) 64946501 6499 Ethanol yields in 1 g L 1 YE, 10 g L 1 CSL and 20 g L 1 CSL media were 53%, 57% and 60% of the theoretical values from CO, respec- tively (Table 3). Based on prices of CSL ($0.18 kg 1 ) and YE ($9.2 kg 1 ) in 2010, CSL contributed $0.30 L 1 ethanol produced in 20 g L 1 CSL med- ium at 9.6 g L 1 ethanol, which is fourfold lower than YE contribu- tion in 1 g L 1 YE medium at 6.1 g L 1 ethanol. The use of 20 g L 1 CSL is currently not cost-effective at the present ethanol titers and yields from syngas fermentation. However, higher ethanol concen- tration and yield will make the use of CSL more cost effective. 3.2.3. Gas consumption Both CO and H 2 were utilized by strain P11 for growth and ace- tic acid and ethanol production. The cumulative CO and H 2 con- sumed by strain P11 during syngas fermentation is shown in Fig. 6a. The total moles of CO consumed by strain P11 in 1 g L 1 YE, 10 g L 1 CSL and 20 g L 1 CSL were 4.3, 5.5 and 5.8 mol, respec- tively. The total moles of H 2 consumed by strain P11 in 1 g L 1 YE, 10 g L 1 CSL and 20 g L 1 CSL were 0.45, 0.63 and 0.48 mol, respec- tively. About 39%, 23% and 28% of the total CO was consumed in 1 g L 1 YE, 10 g L 1 CSL and 20 g L 1 CSL media, respectively, dur- ing the rst 72 h of fermentation for mostly growth and acetic acid production. During the same fermentation period, about 55%, 61% and 57% of the total H 2 was consumed in 1 g L 1 YE, 10 g L 1 CSL and 20 g L 1 CSL media, respectively. In the acetyl-CoA pathway, both H 2 and CO serve as an energy and electron source for cell growth and product formation (Rags- dale, 2004). Signicant amounts of carbon from CO can be con- verted to cell material and ethanol if H 2 is utilized as an electron source. The depletion of H 2 in the fermentation medium would re- duce the amount of carbon available for ethanol production be- cause a fraction of CO would be utilized for generating reducing equivalents required (Ahmed and Lewis, 2007). Therefore, increas- ing the concentration of H 2 in the gas phase would increase the contribution of CO to ethanol production. In contrast, high concen- trations of CO were found to inhibit hydrogenase enzyme and thus reduce the uptake of H 2 in non-CO fermenting organisms (Bennett et al., 2000; Kim et al., 1984). This could explain the decrease in H 2 consumption during the stationary phase with strain P11 (Fig. 6a). Trends in CO consumption proles (Fig. 6a) were similar to eth- anol proles in all media used (Fig. 5). Ethanol production started after 48 h of the fermentation in all media. For ethanol production and cell maintenance between 48 and 360 h, approximately 77%, 89% and 85% of the total moles of CO were consumed in 1 g L 1 YE, 10 g L 1 CSL and 20 g L 1 CSL media, respectively. This shows that most of the CO consumed by strain P11 was utilized for etha- nol production according to Eq. (1). More CO was consumed in the 20 g L 1 CSL medium compared to 1 g L 1 YE and 10 g L 1 CSL media (Fig. 6a), which resulted in higher ethanol production in the 20 g L 1 CSL medium (Fig. 5). The conversion efciencies of CO and H 2 in all media used were low due to the continuous supply of syngas at a rate that strain P11 could not utilize (Table 3). To en- hance the CO and H 2 conversion efciencies, lower syngas ow rates and/or higher initial cell concentrations in the fermentor should be used. The consumption of CO 2 was observed during the growth phase in the 1 g L 1 YE medium (Fig. 6b). Then, there was no net produc- tion of CO 2 in this medium. However, there was a net production of CO 2 in 10 and 20 g L 1 CSL media. Some CO 2 was consumed in 10 and 20 g L 1 CSL medium during late stationary and death phases. Acetic acid production was observed in late stationary phase and death phase in 10 and 20 g L 1 CSL media with the consumption of CO 2 and H 2 via Eq. (4). It is clear that strain P11 utilized more CO than H 2 for growth and product formation (Fig. 6a). Over 50% of the total H 2 consumed in the fermentation was utilized by strain P11 during growth and acetic acid production. Moreover, small amounts of CO 2 were con- sumed in the fermentation media with a net CO 2 production espe- cially in both CSL media (Fig. 6b). The CO 2 and H 2 consumption by strain P11 could have been used for ethanol or acetic acid produc- tion according to Eqs. (2) and (4). This is also supported by the cur- rent results because CO 2 utilization by strain P11 was mostly associated with acetic acid production (Figs. 5 and 6b). To make syngas fermentation more efcient, the medium used should be inexpensive and process parameters such as syngas composition and feeding rate should be optimized to increase syn- gas conversion efciency and ethanol productivity. 4. Conclusions CSL can be used as a lower cost nutrient source in syngas fer- mentation using Clostridium strain P11 to replace more expensive YE. High cell density and extended stationary phase were achieved in the 7.5-L fermentor with 10 and 20 g L 1 CSL media, which re- sulted in over 40% more ethanol production compared to the 1 g L 1 YE medium. The use of CSL as a nutrient enhanced butanol production in bottle fermentations. Over sevenfold more butanol was produced in both CSL media compared to the 1 g L 1 YE medium. Acknowledgements Support for this research was provided by USDA-CSREES Special Research Grant Award 2008-34417-19201, Oklahoma Agricultural Experiment Station, and the Oklahoma Bioenergy Center. References Abrini, J., Naveau, H., Nyns, E.-J., 1994. Clostridium autoethanogenum, sp. nov., an anaerobic bacterium that produces ethanol from carbon monoxide. Archives of Microbiology 161 (4), 345351. Ahmed, A., Lewis, R.S., 2007. Fermentation of biomass-generated synthesis gas: effects of nitric oxide. Biotechnology and Bioengineering 97 (5), 10801086. Azeredo, L.A.I.D., Lima, M.B.D., Coelho, R.R.R., Freire, D.M.G., 2006. A low-cost fermentation medium for thermophilic protease production by Streptomyces sp. 594 using feather meal and corn steep liquor. Current Microbiology 53 (4), 335 339. Babu, B.K., Atiyeh, H.K., Wilkins, M.R., Huhnke, R.L., 2010. Effect of the reducing agent dithiothreitol on ethanol and acetic acid production by Clostridium strain P11 using simulated biomass-based syngas. Biological Engineering Transactions 3 (1), 1935. Bennett, B., Lemon, B.J., Peters, J.W., 2000. Reversible carbon monoxide binding and inhibition at the active site of the Fe-only hydrogenase. Biochemistry 39 (25), 74557460. Datar, R.P., Shenkman, R.M., Cateni, B.G., Huhnke, R.L., Lewis, R.S., 2004. Fermentation of biomass-generated producer gas to ethanol. Biotechnology and Bioengineering 86 (5), 587594. Doran, P.M., 2006. Reactor engineering. In: Bioprocess Engineering Principles. Academic Press, London, pp. 333391. Dunnett, C.W., 1955. A multiple comparison procedure for comparing several treatments with a control. Journal of the American Statistical Association 50, 10961121. Frankman, A.W., 2009. Redox, Pressure and Mass Transfer Effects on Syngas Fermentation. M.S. Thesis, Department of Chemical Engineering, Brigham Young University, p. 106. Huhnke, R.L., Lewis, R.S., Tanner, R.S., 2010. Isolation and Characterization of Novel Clostridial Species. US Patent No. 7704,723. Hurst, K.M., Lewis, R.S., 2010. Carbon monoxide partial pressure effects on the metabolic process of syngas fermentation. Biochemical Engineering Journal 48 (2), 159165. Kadam, K.L., Newman, M.M., 1997. Development of a low-cost fermentation medium for ethanol production from biomass. Applied Microbiology and Biotechnology 47 (6), 625629. Kakiuchi, T., Yoshimatsu, T., Nishi, N., 2007. New class of Ag/AgCl electrodes based on hydrophobic ionic liquid saturated with AgCl. Analytical Chemistry 79 (18), 71877191. Kim, B.H., Bellows, P., Datta, R., Zeikus, J.G., 1984. Control of carbon and electron ow in Clostridium acetobutylicum fermentations: utilization of carbon monoxide to inhibit hydrogen production and to enhance butanol yields. Applied and Environmental Microbiology 48 (4), 764770. 6500 P. Maddipati et al. / Bioresource Technology 102 (2011) 64946501 Klasson, K.T., Ackerson, M.D., Clausen, E.C., Gaddy, J.L., 1992. Bioconversion of synthesis gas into liquid or gaseous fuels. Enzyme and Microbial Technology 14 (8), 602608. Kundiyana, D.K., Huhnke, R.L., Maddipati, P., Atiyeh, H.K., Wilkins, M.R., 2010. Feasibility of incorporating cotton seed extract in Clostridium strain P11 fermentation medium during synthesis gas fermentation. Bioresource Technology 101 (24), 96739680. Lawford, H.G., Rousseau, J.D., 1997. Corn steep liquor as a cost-effective nutrition adjunct in high-performance Zymomonas ethanol fermentations. Applied Biochemistry and Biotechnology 6365 (1), 287304. Liggett, R.W., Kofer, H., 1948. Corn steep liquor in microbiology. Microbiology and Molecular Biology Reviews 12 (4), 297311. Liou, J.S.-C., Balkwill, D.L., Drake, G.R., Tanner, R.S., 2005. Clostridium carboxidivorans sp. nov., a solvent-producing Clostridium isolated from an agricultural settling lagoon, and reclassication of the acetogen Clostridium scatologenes strain SL1 as Clostridium drakei sp. nov. International Journal of Systematic and Evolutionary Microbiology 55 (5), 20852091. Munasinghe, P.C., Khanal, S.K., 2010. Biomass-derived syngas fermentation into biofuels: opportunities and challenges. Bioresource Technology 101 (13), 5013 5022. Panneerselvam, A., 2009. Effect of Glucose and Reducing Agents on Syngas Fermentation by Clostridia Species P11. M.S. Thesis, Department of Biosystems and Agricultural Engineering, Oklahoma State University, p. 82. Panneerselvam, A., Wilkins, M.R., DeLorme, M.J.M., Atiyeh, H.K., Huhnke, R.L., 2010. Effects of various reducing agents on syngas fermentation by Clostridium ragsdalei. Biological Engineering 2 (3), 135144. Parekh, M., Formanek, J., Blaschek, H.P., 1999. Pilot-scale production of butanol by Clostridium beijerinckii BA101 using a low-cost fermentation medium based on corn steep water. Applied Microbiology and Biotechnology 51 (2), 152157. Phillips, J.R., Clausen, E.C., Gaddy, J.L., 1994. Synthesis gas as substrate for the biological production of fuels and chemicals. Applied Biochemistry and Biotechnology 4546, 145157. Racine, F.M., Saha, B.C., 2007. Production of mannitol by Lactobacillus intermedius NRRL B-3693 in fed-batch and continuous cellrecycle fermentations. Process Biochemistry 42 (12), 16091613. Ragsdale, S.W., 2004. Life with carbon monoxide. Critical Reviews in Biochemistry and Molecular Biology 39 (3), 165195. Rajagopalan, S., Datar, P., Lewis, R.S., 2002. Formation of ethanol from carbon monoxide via a new microbial catalyst. Biomass and Bioenergy 23 (6), 487493. Saxena, J., 2008. Development of an Optimized and Cost-effective Medium for Ethanol Production by Clostridium Strain P11. PhD, University of Oklahoma, Norman, OK, p. 110. Saxena, J., Tanner, R.S., 2010. Effect of trace metals on ethanol production from synthesis gas by the ethanologenic acetogen, Clostridium ragsdalei. Journal of Industrial Microbiology and Biotechnology. doi:10.1007/s10295-010-0794-6. Shen, G.J., Shieh, J.S., Grethlein, A.J., Jain, M.K., Zeikus, J.G., 1999. Biochemical basis for carbon monoxide tolerance and butanol production by Butyribacterium methylotrophicum. Applied Microbiology and Biotechnology 51 (6), 827832. Vega, J.L., Prieto, S., Elmore, B.B., Clausen, E.C., Gaddy, J.L., 1989. The biological production of ethanol from synthesis gas. Applied Biochemistry and Biotechnology 2021, 781797. Witjitra, K., Shah, M.M., Cheryan, M., 1996. Effect of nutrient sources on growth and acetate production by Clostridium thermoaceticum. Enzyme and Microbial Technology 19 (5), 322327. Wood, H.G., Ragsdale, S.W., Pezacka, E., 1986. The acetyl-CoA pathway of autotrophic growth. FEMS Microbiology Letters 39 (4), 345362. Worden, R.M., Grethlein, A.J., Jain, M.K., Datta, R., 1991. Production of butanol and ethanol from synthesis gas via fermentation. Fuel 70 (5), 615619. Younesi, H., Najafpour, G., Mohamed, A.R., 2005. Ethanol and acetate production from synthesis gas via fermentation processes using anaerobic bacterium Clostridium ljungdahlii. Biochemical Engineering Journal 27, 110119. P. Maddipati et al. / Bioresource Technology 102 (2011) 64946501 6501