Butanol Production by a Novel Efﬁcient Method Using Mixed Cultures of Clostridium beijerinckii and Arthrobacter sp. in Stirred-Tank and Gas-Lift Bioreactors

: Arthrobacter sp. BCC 72131, an obligate aerobic bacterium, was used to create anaerobic conditions, and Clostridium beijerinckii TISTR 1461 was used as a butanol producer in an acetone-butanol–ethanol (ABE) fermentation. Sweet sorghum juice (SSJ) medium containing 60 g/L of total sugar supplemented with 1.27 g/L of (NH 4 ) 2 SO 4 was used as a butanol production (BP) medium. Arthrobacter sp. was inoculated into the BP medium in 1-L screw-capped bottles. After 2, 4, 6 and 12 h of Arthrobacter sp. cultivation at 30 ◦ C, C. beijerinckii was transferred into the BP medium to start butanol production at 37 ◦ C. The results showed that C. beijerinckii inoculation after 6 h of Arthrobacter sp. cultivation gave the highest butanol titer ( P B ) at 12.56 g/L, with a butanol yield ( Y B/S ) and volumetric butanol productivity ( Q B ) of 0.34 g/g and 0.23 g/L · h, respectively. These values are approximately 10–27% higher than those of the control experiment using a single culture of C. beijerinckii TISTR 1461 and oxygen-free nitrogen (OFN) gas ﬂushing to create anaerobic conditions. Field emission scanning electron microscopic (FE-SEM) images of Clostridium cells, as well as protein and free amino nitrogen concentrations in the broth during butanol fermentation were also studied to conﬁrm the results. The butanol fermentation was then carried out in a 5.6-L stirred-tank and a 1.2-L low-cost gas-lift bioreactor by the mixed cultures using the optimal time of Clostridium inoculation. The P B , Y B/S and Q B values obtained were not signiﬁcantly different from those in the 1-L screw-capped bottles. Hence, Arthrobacter sp. can be used as a novel method to create anaerobic conditions instead of a traditional method employing OFN gas ﬂushing. Using mixed cultures of Arthrobacter sp. BCC 72131 and C. beijerinckii TISTR 1461 is a practical method to produce butanol on a large-scale, both in complex and low-cost bioreactors.


Introduction
Butanol is widely used as a chemical starting material for making plastics, plasticizers, polymers, paints, lubricants, brake fluids and synthetic rubber [1,2]. Butanol has a high energy density of 29.2 MJ/L and can be substituted for gasoline (energy density of 32 MJ/L) with no modifications to the current internal combustion engines [3]. Presently, butanol is most often obtained from crude oil [4]. However, with the increasing fluctuations in petroleum oil availability and costs as well accumulating environmental dilemmas, butanol fermentations are alternatives of renewed interest. Additionally, it is an excellent renewable fuel that can be considered a better transportation fuel than bioethanol in future fuel systems [5], mainly due to its higher number of carbon atoms and consequently higher energy content, miscibility with gasoline and less corrosivity [6]. Additionally,

Raw Materials
Sweet sorghum juice (SSJ) (cv. KKU 40) extracted from sorghum stalks was obtained from the Faculty of Agriculture, Khon Kaen University, Thailand. It was concentrated from a total soluble solids content of 17 to 68 • Bx by heating at 80-85 • C and stored at −20 • C to protect it against bacterial growth [18]. The composition of concentrated SSJ in this study consisted of total soluble sugars (677.02 g/L), sucrose (236.63 g/L), glucose (201.61 g/L), and fructose (233.14 g/L).

Butanol Production (BP) Medium Preparation
Concentrated SSJ was diluted with distilled water to obtain a juice containing 60 g/L of total sugars. The juice was then supplemented with 1.27 g/L of (NH 4 ) 2 SO 4 (BDH, Poole, UK) and autoclaved before use as a butanol production (BP) medium [11].
Arthrobacter sp. BCC 72131 was purchased from the Thailand Bioresource Research Center (TBRC), Khlong Luang, Pathumthani, Thailand. Cells were grown in a sterile nutrient broth (NB) under shaking conditions at 200 rpm and 30 • C for 6-7 h. Afterwards, the culture was transferred to fresh NB under the same conditions to obtain an optical density at 600 nm of 0.5 (~0.8 g/L of cell dry weight) [11] before use as an inoculum to consume oxygen in a fermentation broth and thereby create anaerobic conditions. For a positive control experiment, OFN gas at a flow rate of 0.5 vvm was purged for 10 min into a sterile BP medium (Section 2.2) in 1-L screw-capped bottles with a working volume of 700 mL to create anaerobic conditions before fermentation [21]. Five percent (v/v) of the active growing cells (Section 2.3) was then transferred to the BP medium. The fermentation was operated at 37 • C with an agitation rate of 150 rpm until the end of fermentation. For a negative control experiment, the fermentation was carried out in the same manner as the positive control experiment, except there was no OFN gas flushing. First, 5% (v/v) of Arthrobacter sp. BCC 72131 in NB (Section 2.3) was inoculated in sterile BP medium in 1-L screw-capped bottles and incubated at 30 • C with 150 rpm for 2, 4, 6 and 12 h. After each time, the active growing cells of C. beijerinckii TISTR 1461 were transferred to a BP medium to start the ABE fermentation. The fermentation was controlled at 37 • C with agitation at 150 rpm.

Butanol Production by Mixed Cultures in Stirred-Tank and Gas-Lift Bioreactors
A 5.6-L stirred-tank bioreactor (BioFlo ® 320 Vessels, Eppendorf, Hamburg, Germany) and a 1.2-L low-cost gas-lift bioreactor, shown in Figure 1, were sterilized before adding a sterile BP medium with working volumes of 3.6 and 0.8 L, respectively. Arthrobacter sp. BCC 72131 (5% v/v) was then inoculated into the sterile BP medium in both bioreactors and incubated at 30 • C with agitation rate of 150 rpm using a six-blade dish turbine and four baffles in the stirred-tank bioreactor. Mixing in the gas-lift bioreactor was controlled with a draft tube and the circulation of gases in the head space was achieved using a peristaltic pump (Masterflex ® , Radnor, PA, USA) at a flow rate of 0.25 vvm (modified from [22]). After the optimum time of Arthrobacter cultivation (obtained as described in Section 2.4.2), active C. beijerinckii TISTR 1461 cells were added into the bioreactors to start ABE fermentations.
Butanol fermentations were performed at 37 • C. During fermentation, samples were taken at predetermined time intervals for analyses.
four baffles in the stirred-tank bioreactor. Mixing in the gas-lift bioreactor was controlled with a draft tube and the circulation of gases in the head space was achieved using a peristaltic pump (Masterflex ® , Radnor, PA, USA) at a flow rate of 0.25 vvm (modified from [22]). After the optimum time of Arthrobacter cultivation (obtained as described in Section 2.4.2), active C. beijerinckii TISTR 1461 cells were added into the bioreactors to start ABE fermentations. Butanol fermentations were performed at 37 °C. During fermentation, samples were taken at predetermined time intervals for analyses.

Analytical Methods
Samples were centrifuged at 12,000 rpm for 10 min to the separate cells and particles from the supernatant (Sigma 1-14, Sartorius, Osterode am Harz, Germany). The supernatant was filtered through a 0.45 µm nylon membrane (Sartorius, Goettingen, Germany) for analysis. Acetone, butanol, ethanol, acetic acid, and butyric acid in the supernatant were analyzed using a gas chromatograph (Shimadzu, GC-2014, Kyoto, Japan) employing a Porapak Q column (80/100 mesh, 3 m × 2 mm, Resteck, PA, USA), and nitrogen was used

Analytical Methods
Samples were centrifuged at 12,000 rpm for 10 min to the separate cells and particles from the supernatant (Sigma 1-14, Sartorius, Osterode am Harz, Germany). The supernatant was filtered through a 0.45 µm nylon membrane (Sartorius, Goettingen, Germany) for analysis. Acetone, butanol, ethanol, acetic acid, and butyric acid in the supernatant were analyzed using a gas chromatograph (Shimadzu, GC-2014, Kyoto, Japan) employing a Porapak Q column (80/100 mesh, 3 m × 2 mm, Resteck, PA, USA), and nitrogen was used as a carrier gas [11]. pH was measured using a pH meter. The total sugar concentration was determined using a phenol-sulfuric acid method [23]. A quantitative analysis of fructose, glucose and sucrose was performed with an HPLC-RI (Waters, MA, USA) using an Inertsil ® NH2 column (5 µm, 250 × 4.6 mm, GL Sciences, Tokyo, Japan). The analysis was performed at 35 • C with a flow rate of 0.8 mL/min using an isocratic system with a mixture ratio of 75 parts acetonitrile: 25 parts water (v/v) as a mobile phase [24,25]. Protein and free amino nitrogen (FAN) concentrations under optimal conditions and control experiments were also measured using the Bradford method [26] and a ninhydrin assay [27], respectively.
The morphology of C. beijerinckii TISTR 1461 during fermentation was observed using field emission scanning electron microscopic (FE-SEM) analysis. Samples were taken from the bioreactors, quickly filtered through a 0.2 µm cellulose acetate membrane (Whatman ® , Maidstone, England), and washed twice with distilled water. The filtrate and bacterial cells remaining on the membrane were then dried overnight at 50 • C in a hot air oven. Then, the dried material was sputter-coated with gold. An accelerating voltage of 10 kV under a vacuum was used to yield high-resolution images. Microscopic analysis was performed with a FEI Helios Nanolab G3 CX Dual Beam FIB/SEM, Thermo Fisher Scientific, Hillsboro, OR, USA.
All experiments were performed in triplicate, and the results are expressed as mean values ± SD. The butanol yield (Y B/S , g/g), ABE yield (Y ABE/S , g/g), volumetric butanol productivity (Q B , g/L·h) and sugar consumption (SC, %) were calculated as described in Daengbussadee et al. [11].

Butanol Production from SSJ Medium by Single and Mixed Cultures
Butanol fermentations by a single culture of C. beijerinckii TISTR 1461 with and without OFN gas flushing were carried out as positive and negative control treatments, respectively. The pH changes during the fermentation of both control treatments were similar ( Figure 2a). However, the P B (9.88 ± 0.38 g/L), ABE concentrations (P ABE , 17.61 ± 0.63 g/L) and sugar utilization (53.46 ± 0.46%) were clearly higher for the positive control with OFN gas flushing than those of the negative control with no OFN gas flushing (Figure 2b-d and Table 1). This confirms that C. beijerinckii TISTR 1461 produced higher butanol levels under strictly anaerobic conditions.
For mixed cultures, after 2, 4, 6 and 12 h of Arthrobacter sp. cultivation, C. beijerinckii TISTR 1461 was then inoculated to start the ABE fermentation. The results showed that the ABE fermentation profiles of all conditions of the mixed cultures were similar to those of the positive control treatment using OFN gas flushing ( Figure 2). The P B , sugar utilization and Q B of the conditions under which C. beijerinckii TISTR 1461 was inoculated after 2 and 4 h were not significantly different compared to the positive control treatment (Table 1). These results clearly indicated that an obligate aerobe, Arthrobacter sp., could be used to create anaerobic conditions. The inoculation of C. beijerinckii TISTR 1461 after 6 h gave the highest P B (12.56 ± 0.55 g/L) and P ABE (21.18 ± 0.44 g/L). Surprisingly, the P B and P ABE of the mixed cultures under this condition were approximately 20-27% higher than those of the positive control treatment. The P B when C. beijerinckii was inoculated after 12 of Arthrobacter sp. cultivation was not significantly different from inoculation after 6 h (Table 1). This suggests that 6 h of Arthrobacter sp. cultivation was appropriate for developing anaerobic conditions for butanol production by C. beijerinckii TISTR 1461. The P ABE value when C. beijerinckii was inoculated after 12 h of Arthrobacter sp. cultivation was lower than that after 6 h, but the sugar utilization of the former was higher than the latter. These observations might be due to the differences in the by-products produced by the pyruvate metabolism pathway that were not detected in this study such as hydrogen, isopropanol, lactic and propionic acids, carbon dioxide and lipids [28]. ture, the protein content in the broth was relatively constant and the FAN continuously decreased as the ABE fermentation progressed. In the fermentation by the mixed cultures, the increase in protein and FAN within 24 h after C. beijerinckii inoculation might have been due to the autolysis of Arthrobacter sp. This might promote the growth and butanol production of C. beijerinckii TISTR 1461 in a way that is not possible using a single culture of the Clostridium.   Table 1. Kinetic parameters of ABE fermentation by the mixed cultures of C. beijerinckii TISTR 1461 and Arthrobacter sp. BCC 72131 at different times of Arthrobacter sp. cultivation prior to C. beijerinckii inoculation and by a single culture of C. beijerinckii TISTR 1461 using OFN gas flushing. It was found that sugar consumption by Arthrobacter sp. before C. beijerinckii TISTR 1461 inoculation in all experiments ranged from 1.54 to 3.96 g/L. This implies that Arthrobacter sp. consumed some sugar and all the oxygen in the fermentation broth to grow or survive. During ABE fermentation, C. beijerinckii TISTR 1461 continuously consumed sugars in the fermentation broth. Under inoculating C. beijerinckii after 6 h of Arthrobacter sp. cultivation, the total remaining sugars at the end of fermentation were 24.42 g/L (~60% sugar consumption). The residual sugar fraction consisted of glucose (0.40 g/L), fructose (8.90 g/L), sucrose (9.57 g/L) and non-fermentable sugars (5.55 g/L). The fermentable sugars in the broth were not completely consumed, which might have been due to butanol toxicity (P B , 12.56 ± 0.55 g/L and P ABE , 21.18 ± 0.44 g/L). This was supported by Xu et al. [29] who reported that the cell growth stopped when the butanol concentration reached~10 g/L and the fermentation stopped at~12 g/L of butanol, although there was still plenty of glucose present in the medium.
After 12 h of fermentation by the mixed cultures, the cell debris of Arthrobacter sp. was observed under a light microscope. Therefore, the protein and free amino nitrogen (FAN) in the fermentation broth were determined during the ABE fermentation ( Figure 3). It was found that the initial values of protein and FAN in the SSJ medium using the mixed cultures and single cultures were not different. Protein and FAN increased within 24 h and decreased after that in the mixed cultures. In the fermentation by the single culture, the protein content in the broth was relatively constant and the FAN continuously decreased as the ABE fermentation progressed. In the fermentation by the mixed cultures, the increase in protein and FAN within 24 h after C. beijerinckii inoculation might have been due to the autolysis of Arthrobacter sp. This might promote the growth and butanol production of C. beijerinckii TISTR 1461 in a way that is not possible using a single culture of the Clostridium.

Morphology of C. beijerinckii TISTR 1461 during ABE Fermentation by a Single Culture and Mixed Cultures
The morphologies of C. beijerinckii TISTR 1461 during ABE fermentation by single and mixed cultures under optimal conditions were compared. The FE-SEM micrographs of a single culture using OFN gas flushing are shown in Figure 4. In the first phase-the acidogenesis phase (0-12 h) (Figure 2a)-the cells appeared as short and highly motile rods (Figure 4a,b). During 24-36 h of ABE fermentation, the cells were enlarged, swollen and moved sluggishly (Figure 4c,d). In this phase-the solventogenesis phase-high butanol levels were produced (Figure 2b). After that, forespores were observed (Figures 4e and 5a), and the maximal P B value was achieved at 48 h of fermentation (Figure 2b). The occurrence of forespores might have been due to butanol toxicity. Then, no butanol was produced until the end of fermentation. During this period, the cells had two survival mechanisms: solventogenesis and sporulation (Figure 4f) [30].
The morphologies of C. beijerinckii TISTR 1461 during ABE fermentation by single and mixed cultures under optimal conditions were compared. The FE-SEM micrographs of a single culture using OFN gas flushing are shown in Figure 4. In the first phase-the acidogenesis phase (0-12 h) (Figure 2a)-the cells appeared as short and highly motile rods (Figure 4a,b). During 24-36 h of ABE fermentation, the cells were enlarged, swollen and moved sluggishly (Figure 4c,d). In this phase-the solventogenesis phase-high butanol levels were produced (Figure 2b). After that, forespores were observed (Figures 4e  and 5a), and the maximal PB value was achieved at 48 h of fermentation (Figure 2b). The occurrence of forespores might have been due to butanol toxicity. Then, no butanol was produced until the end of fermentation. During this period, the cells had two survival mechanisms: solventogenesis and sporulation (Figure 4f) [30].
In ABE fermentations by the mixed cultures, the morphology of C. beijerinckii TISTR 1461 appeared perfect with strongly motile long rods (Figure 5b), and a higher PB (12.56 ± 0.55 g/L) was obtained at 48 h after C. beijerinckii TISTR 1461 inoculation (Table 1). In contrast, the morphology of single culture C. beijerinckii TISTR 1461 at 48 h was imperfect and forespores were observed (Figure 5a). A lower PB value (9.88 ± 0.38 g/L) was obtained in a single culture ( Table 1). The results of FE-SEM indicated that Arthrobacter sp. could not only be used to effectively create anaerobic conditions, but its use also promoted butanol production from the SSJ medium by C. beijerinckii TISTR 1461.   of a single culture using OFN gas flushing are shown in Figure 4. In the first phaseacidogenesis phase (0-12 h) (Figure 2a)-the cells appeared as short and highly m rods (Figure 4a,b). During 24-36 h of ABE fermentation, the cells were enlarged, sw and moved sluggishly (Figure 4c,d). In this phase-the solventogenesis phase-high tanol levels were produced (Figure 2b). After that, forespores were observed (Figur and 5a), and the maximal PB value was achieved at 48 h of fermentation (Figure 2b) occurrence of forespores might have been due to butanol toxicity. Then, no butanol produced until the end of fermentation. During this period, the cells had two sur mechanisms: solventogenesis and sporulation (Figure 4f) [30]. In ABE fermentations by the mixed cultures, the morphology of C. beijerinckii T 1461 appeared perfect with strongly motile long rods (Figure 5b), and a higher PB (12 0.55 g/L) was obtained at 48 h after C. beijerinckii TISTR 1461 inoculation (Table 1). In trast, the morphology of single culture C. beijerinckii TISTR 1461 at 48 h was imperfec forespores were observed (Figure 5a). A lower PB value (9.88 ± 0.38 g/L) was obtain a single culture ( Table 1). The results of FE-SEM indicated that Arthrobacter sp. could only be used to effectively create anaerobic conditions, but its use also promoted bu production from the SSJ medium by C. beijerinckii TISTR 1461.  In ABE fermentations by the mixed cultures, the morphology of C. beijerinckii TISTR 1461 appeared perfect with strongly motile long rods (Figure 5b), and a higher P B (12.56 ± 0.55 g/L) was obtained at 48 h after C. beijerinckii TISTR 1461 inoculation (Table 1). In contrast, the morphology of single culture C. beijerinckii TISTR 1461 at 48 h was imperfect and forespores were observed (Figure 5a). A lower P B value (9.88 ± 0.38 g/L) was obtained in a single culture ( Table 1). The results of FE-SEM indicated that Arthrobacter sp. could not only be used to effectively create anaerobic conditions, but its use also promoted butanol production from the SSJ medium by C. beijerinckii TISTR 1461.

Butanol Production by the Mixed Cultures in Stirred-Tank and Gas-Lift Bioreactors
To study the capability of butanol production by the mixed culture in a low-cost bioreactor, i.e., a gas-lift bioreactor, the fermentation profiles of butanol production from the SSJ medium by mixed cultures of C. beijerinckii TISTR 1461 and Arthrobacter sp. BCC 72131 (inoculating C. beijerinckii after 6 h of Arthrobacter cultivation) in the screw-capped bottle, stirred-tank and gas-lift bioreactors were compared in Figure 6. The pHs profiles in the three bioreactors were similar (Figure 6a). The pHs decrease corresponded to the production of acids (Figure 6b,d), indicating that bacterial cells were active. Moreover, the movement of cells during this period under the microscope was swift (data not shown). In a stirred-tank bioreactor, the sugar consumption and product formation during the first period of fermentation were slightly higher than those in the screw-capped bottle and gas-lift bioreactor (Figure 6e-i). These might be due to the positive effect of the mixing. The results suggested that mixing, mass transfer and heat transfer phenomena in the stirredtank bioreactor occurred faster than those in the screw-capped bottle and gas-lift bioreactor. However, the P B of each mixing method in screw-capped bottle (12.56 ± 0.55 g/L), stirredtank bioreactor (12.59 ± 0.45 g/L) and gas-lift bioreactor (12.00 ± 0.48 g/L) were not significantly different at the end of fermentation ( Table 2). The kinetic parameters of ABE fermentation in the various bioreactors tested were not significantly different, suggesting that the mass transfer phenomenon in the bioreactors successfully occurred via mixing. In addition, the ABE yield (Y ABE/S ) and Y B/S under all tested conditions was not significantly different, indicating that the types of bioreactor used did not affect the metabolic pathway of ABE fermentation by the mixed cultures (Table 2). With regard to the sizes of the bioreactors and the head space during ABE fermentation, the working volumes of the 1-L screw-capped bottle, 5.6-L stirred-tank bioreactor and 1.2-L gas-lift bioreactor were 0.7, 3.6 and 0.8 L, respectively. These correspond to head space volumes of 30, 36 and 33%, respectively. Similar ABE fermentation results in Table 2 imply that the different sizes of three bioreactors with a similar head space did not have any impact on ABE fermentation. Therefore, Arthrobacter sp., an obligate aerobe bacterium, is an effective microorganism for creating anaerobic conditions for butanol production by C. beijerinckii TISTR 1461 in all bioreactors, including the low-cost bioreactor, i.e., the gas-lift bioreactor.

Comparison of Butanol Production by Mixed Cultures with Other Studies
In this study, Arthrobacter sp. BCC 72131 was used to create anaerobic conditions for butanol production by C. beijerinckii TISTR 1461. Earlier research was performed using other microorganisms (e.g., Bacillus subtilis, B. cereus, B. cellolyticus and Saccharomyces cerevisiae) to create anaerobic conditions for butanol production by Clostridium. Butanol production levels by mixed cultures are summarized in Table 3. It was found that the P B values in all previous research ranged from 3.77 to 14.90 g/L, while Q B and Y B/S ranged from 0.02 to 0.23 g/L·h and from 0.21 to 0.34 g/g, respectively. The P B values in the current study were relatively high compared to earlier works. In our study, the P B was slightly lower than that reported by Abd-Alla and Elsadek El-Enany [15], which might have been due to the differences in initial substate concentration, the bacterial cells used and environmental conditions for ABE fermentation. However, Q B and Y B/S in our study were higher than those in other studies by approximately 1.1-11.5-fold. This implies that Arthrobacter sp. has high potential for use to create anaerobic conditions for butanol production. Butanol production by the mixed cultures of Arthrobacter sp. and C. beijerinckii TISTR 1461 has high potential for use in large-scale production.

Comparison of Butanol Production by Mixed Cultures with Other Studies
In this study, Arthrobacter sp. BCC 72131 was used to create anaerobic conditions for butanol production by C. beijerinckii TISTR 1461. Earlier research was performed using other microorganisms (e.g., Bacillus subtilis, B. cereus, B. cellolyticus and Saccharomyces cerevisiae) to create anaerobic conditions for butanol production by Clostridium. Butanol production levels by mixed cultures are summarized in Table 3. It was found that the PB values in all previous research ranged from 3.77 to 14.90 g/L, while QB and YB/S ranged from 0.02 to 0.23 g/L·h and from 0.21 to 0.34 g/g, respectively. The PB values in the current study were relatively high compared to earlier works. In our study, the PB was slightly lower than that reported by Abd-Alla and Elsadek El-Enany [15], which might have been due to the differences in initial substate concentration, the bacterial cells used and environmental  An alternative method to create anaerobic conditions for ABE fermentations is using chemicals that strongly react with oxygen such as sodium dithionite (STDN). It was reported that 0.25 mM SDTN could be used to create anaerobic conditions for C. beijerinckii to produce butanol (8.51 g/L) but very low butanol productivity (0.1 g/L·h) was obtained [11]. According to our results, Arthrobacter sp. is suitable to be used to create anaerobic conditions for butanol production. During anaerobic condition creation, sugar consumption by Arthrobacter sp. in various bioreactors was only 2.20-2.76 g/L, corresponding to 3.67-4.60% of the total sugar. However, the economics of using the mixed cultures (Arthrobacter sp. and C. beijerinckii TISTR 1461) and a single culture of C. beijerinckii TISTR 1461 with OFN gas flushing needs to be further evaluated [31]. Mixed-sugars (25 g/L of xylose and 25 g/L of glucose) C. acetobutylicum CH02 + S. cerevisiae 250-mL flask 8.33 0.09 0.18 [32] P B , butanol concentration; Q B , volumetric butanol productivity; and Y B/S , butanol yield.

Conclusions
Mixed cultures of Arthrobacter sp. BCC 72131 and C. beijerinckii TISTR 1461 were successfully used to produce butanol from a simple sweet sorghum juice medium. Arthrobacter sp. is a highly effective bacterium for creating anaerobic conditions that can be used in place of the traditional method employing OFN gas flushing. Inoculating C. beijerinckii 6 h after Arthrobacter sp. was the most appropriate method for butanol production from sweet sorghum juice. The mixed cultures yielded higher butanol production than a single culture of C. beijerinckii TISTR 1461. Butanol production by mixed cultures of Arthrobacter sp. BCC 72131 and C. beijerinckii TISTR 1461 can be applied in large-scale stirred-tank and low-cost gas-lift bioreactors.