Biological carbon monoxide conversion to acetate production by mixed culture

Bioresource Technology 211 (2016) 478–485

Contents lists available at ScienceDirect

Bioresource Technology journal homepage: www.elsevier.com/locate/biortech

Biological carbon monoxide conversion to acetate production by mixed culture Chul Woo Nam a, Kyung A Jung b, Jong Moon Park a,b,c,⇑ a

Department of Chemical Engineering, Pohang University of Science and Technology, 77 Cheongam-ro, Nam-gu, Pohang 37673, South Korea Bioenergy Research Center, Pohang University of Science and Technology, 77 Cheongam-ro, Nam-gu, Pohang 37673, South Korea c Division of Advanced Nuclear Engineering, Pohang University of Science and Technology, 77 Cheongam-ro, Nam-gu, Pohang 37673, South Korea b

h i g h l i g h t s
Undefined mixed cultures from wastewater treatment plant were used as inocula for CO conversion.
Fermentation conditions for HAc production were optimized using RSM with BBD.
Higher HAc production was achieved at 20.81% CO, 41.38% CO2 and 7.18 pH.
The 23.6 g/L of high HAc production was achieved at continuous gas condition.

a r t i c l e

i n f o

Article history: Received 1 January 2016 Received in revised form 18 March 2016 Accepted 19 March 2016 Available online 22 March 2016 Keywords: Gas fermentation Carbon monoxide Acetate Response surface method Box–Behnken design

a b s t r a c t To utilize waste CO for mixed culture gas fermentation, carbon sources (CO, CO2) and pH were optimized in the batch system to find out the center point and boundary of response surface method (RSM) for higher acetate (HAc) production (center points: 25% CO, 40% CO2, and pH 8). The concentrations of CO and CO2, and pH had significant effects on acetate production, but the pH was the most significant on the HAc production. The optimum condition for HAc production in the gas fermentation was 20.81% CO, 41.38% CO2, 37.81% N2, and pH 7.18. The continuous gas fermentation under the optimum condition obtained 1.66 g/L of cell DW, 23.6 g/L HAc, 3.11 g/L propionate, and 3.42 g/L ethanol. Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction Carbon monoxide (CO) is a toxic waste gas produced in large quantity from petrochemical industries, electric power plants, and iron smelters processes. The utilization of CO gas as a feedstock for organic chemicals production through gas conversion reaction could create a potential profit of the industry from reducing waste disposal cost (Sim et al., 2008; Vega et al., 1990). CO can be converted to various organic chemicals such as acetate (HAc), acetone, and ethanol by a microbiological CO conversion process, called gas fermentation (Mohammadi et al., 2011). The process has high specificity, mild operating conditions, ability of utilizing various types of gases, and resistance to toxic gases (Sim et al., 2008). Earlier gas fermentation studies were based on pure-culture fermentation systems (Guo et al., 2010; Kundiyana ⇑ Corresponding author at: Department of Chemical Engineering, Pohang University of Science and Technology, 77 Cheongam-ro, Nam-gu, Pohang 37673, South Korea. E-mail address: [email protected] (J.M. Park). http://dx.doi.org/10.1016/j.biortech.2016.03.100 0960-8524/Ó 2016 Elsevier Ltd. All rights reserved.

et al., 2010a,b; Younesi et al., 2005), however, the pure culture system requires a high operation cost to sustain sterilization conditions (Younesi et al., 2005). On the other hand, the mixed culture does not require the sterilization condition, and has high adaptation capacity to various substrates including toxic components. Therefore, the mixed culture can be more suitable for applying to large industrial systems (Kleerebezem and van Loosdrecht, 2007). In the gas fermentation of CO gas, HAc can be produced as the major component in the large amount without any additional energy source. HAc is utilized as an end-product by itself as well as an intermediate for further chemical processes to produce value-added compounds (Batlle-Vilanova et al., 2016). Thus, enhancing HAc productivity in the gas fermentation results in maximization of carbon recovery efficiency and reduction of CO2 emission from the fermentation process (Sim et al., 2008). In the gas fermentation system, the type of carbon sources (e.g., CO and CO2) and pH could significantly affect microbial growth and various metabolite production. As the energy and carbon source in the gas fermentation, the partial pressure of CO would influence the cell growth and formation of various products (Hurst and

479

C.W. Nam et al. / Bioresource Technology 211 (2016) 478–485

Lewis, 2010). Although CO2 is not an energy source in the system, it can act as a starting material in most carbon-fixation processes because the formation of acetyl coenzyme A needs CO2 (Heiskanen et al., 2007). Thus the partial pressure of CO2 can also affect the cell growth rate and final products. Variation of pH conditions in the system causes the final product yield, and the composition of metabolites (i.e., organic acids). At low pH (4–5.5), butyrate and HAc were the main products in the gas fermentation; however, at high pH (10–12), HAc was selectively formed with over 70% of the total organic acid in the system (Jankowska et al., 2015; Temudo et al., 2007). In this study, the concentration of CO ([CO]), [CO2], and pH were optimized to maximize HAc production in a mixed culture gas fermentation system using response surface methodology (RSM). The statistical approach using RSM was also implemented to quantify the relationships among three measurable responses with the one vital factors. Finally, the optimized condition was applied to a continuous gas fermentation system to design a high-efficiency CO conversion process model using the mixed culture system. 2. Materials and methods 2.1. Preparation of gas fermentation inoculums Raw sludge was obtained from a domestic wastewater treatment plant (Daegu, South Korea). To select organic-acidproducing microorganism, the sludge was heated at 90 °C for 30 min to kill methanogens. Inocula for gas fermentation were acclimatized in a 5 L fermenter (Biotron Inc., South Korea) under anaerobic conditions, then 3 L of the heat-treated inocula and 2 L of anaerobic modified M9 medium (CaCl2 0.015 g/L, Na2HPO4 6 g/ L, KH2PO4 3 g/L, NaCl 0.5 g/L, NH4Cl 1 g/L, MgSO4 0.5 g/L, and yeast extract (BD, USA) 0.5 g/L) were supplied to the fermenter. In the medium, the nitrogen source (e.g., yeast extract and NH4Cl) constitutes 12.5% of total nutrient composition because microorganisms use it to create energy to grown and reproduce (Sim et al., 2008). The pH was controlled between 6.0 and 6.5 by an automatic pH controller (Model KB-250, K&B, South Korea) using 5 M NaOH (Samchun, South Korea) and 1 M HCl (Deajung, South Korea) solutions. Temperature was maintained at 37 °C and stirring rate was maintained at 200 rpm. After acclimation, 12 mL/min of mixed gas (25% CO, 25% CO2, 50% N2) was supplied to the fermenter by a gas mixer (Automatic gas mixing system SHGM-4000, Sehwa Corp., South Korea). 2.2. Batch experimental preparation Batch experiments were performed in 280-mL serum bottles with 100-mL working volume. The modified M9 medium which was purged with nitrogen gas for 10 min to maintain anaerobic condition was used. Experiments were conducted with 0.22 g/L of the pretreated inocula. The initial pH of the medium was adjusted using potassium phosphate buffer. The batch experiment was conducted in three steps optimization of [CO], [CO2], and pH in serial order. First, volumetric percentage of [CO] was regulated at 0%, 25%, 50%, 75% or 100% with [N2]. Based on the best [CO], volumetric percentage of [CO2] was then adjusted at 0%, 25%, 50% or 75% with [N2]. Then pH values of 4, 6, 8 or 10 were applied at the best combination of [CO], [CO2] and [N2] (Heiskanen et al., 2007). All batch experiment was conducted in duplicate. The serum bottles sealed with butyl rubber stoppers and aluminum seal caps were purged by the mixed gas for 6 min. The bottles were put into operation in a shaking incubator at 200 rpm and 37 °C. Liquid (2 mL) and gas (100 lL) samples were collected every 12 h for further analysis.

2.3. Analytical methods The pH in the reactor was continuously monitored using a pH meter (405-DPAS-SC-K85, METTLER TOLLEDO, Switzerland). Organic acids were quantified using a high performance liquid chromatograph (HPLC, Agilent Technology 1100 series, Agilent Inc., USA) equipped with an organic acid and alcohol analysis column (Aminex HPX-87H, BIORAD Inc., USA), a refractive index detector, and a diode array detector. [CO], [N2], and [CO2] were analyzed using a gas chromatograph (GC, Model 6890N, Agilent INC, USA) equipped with a capillary column (Supelco CarboxenTM1010 PLOT capillary column, 30 m  0.32 mm, Sigma–Aldrich, USA) and a pulsed discharged detector. The carrier gas was He with flow rate of 1.9 mL/min; the temperatures of oven, inlet and detector were 120, 150, and 240 °C, respectively. CO consumption [mg/L] and CO consumption rate [mg/L/day] were calculated as follows:

CO consumption ¼

½COi  ½COa reactor working volume

CO consumption rate ¼

CO consumption t

ð1Þ

ð2Þ

where [CO]i is the initial volume of the [CO], [CO]a is the actual volume of the [CO] at sampling time, and t is the time until all CO was consumed. HAc production yield Y [g/g CO] and HAc production rate [g/L/h] was calculated as follows:

HAc production yield Y ¼ HAc production rate ¼

½HAca  ½HAcc A

½HAca t

ð3Þ ð4Þ

where [HAc]a [mg/L] is the HAc concentration, [HAc]c is HAc concentration at [CO] = 0% (control condition), A [mg/L] is inlet CO volume, and t is the time until achieve the highest HAc concentration. The cell dry weight (cell DW) was estimated by filtering a cell suspension sample through a pre-dried and pre-weighed 0.45lm nitrocellulose filter (Millipore, USA). The sample was dried at 110 °C and then weighed by the difference. Cell DW yield [g/g CO] was calculated as follows:

Cell DW yield ¼

cell DWa  cell DWi A

ð5Þ

where cell DWa [g/L] is the actual cell DW, cell DWi is the initial cell DW, and A [g/L] is inlet CO volume. 2.4. Response surface methodology (RSM) A 3K factorial Box–Behnken model was used as an experimental design to optimize the process parameters (pH, [CO], [CO2]) to maximize HAc yield. In the three-factor condition, the Box–Behnken design (BBD) has the advantage of requiring fewer runs than other RSM protocols (Jo et al., 2008; Sim et al., 2008). The BBD is efficient to estimate second degree quadratic polynomial and to combine values optimizing the response within the region of the three-dimensional observation space (Annadurai et al., 1999). To develop the regression equation, the relation between the codedand actual values are shown as follows:

xi ¼

ðX i  X i Þ DX i

ð6Þ

where xi is the coded value of the ith independent variable, Xi is the uncoded value of the ith independent variable, X i is the uncoded

480

C.W. Nam et al. / Bioresource Technology 211 (2016) 478–485

value of the ith independent variable at the center point and DXi is the step change value (Jo et al., 2008). To analyze effects of different reaction conditions on CO conversion, [CO], [CO2], and pH were chosen as the independent variables of the BBD. The relationship between variables and responses is described as follows: 3 X

i¼1

i¼1

ai X i þ

3 X 3 X

aii X 2i ð7Þ

3. Results and discussion

aij X i X j

i¼1 i
3.1. Effect of carbon monoxide on acetate production in batch condition

where Xi are the input variables, which influence the response variable Y, a0 the offset term, ai the ith linear coefficient, aii the quadratic coefficient and aij is the ijth interaction coefficient. The input values of X1, X2, and X3 corresponding to the maximum value of Y were solved by setting the partial derivatives of the functions to zero (Jo et al., 2008).

At all [CO], cell DW did not increase for the first 72–108 h, but increased exponentially until 240 h, then stabilized after 200 h (Fig. 1a). The initial lag time increased as [CO] increased (72 h at 25%, 96 h at 50% and 75%, and 108 h at 100%). The highest cell DW concentration was 0.56 g/L at [CO] = 75%, and the highest cell DW yield was 0.3275 g/g CO at [CO] = 25%. Cell DW yield decreased with as [CO] increased (Fig. 1d) because high partial pressure of CO can retard cell growth (Hurst and Lewis, 2010). The mass transfer limitation of CO in commercialized plants could affect cell growth and metabolite formation (Hurst and Lewis, 2010), however, the

2.5. Liquid batch, and continuous gas fermentation For operating liquid batch and continuous gas fermentation processes, stirred tank reactors (1 L, Biotron Inc., South Korea) were filled with 100 mL of acclimated and enriched inocula and 900 mL

(b)

0.7 25 % 50 % 75 % 100%

Cell DW (g/L)

0.6

2500 25 % 50 % 75 % 100%

2000

CO concentration (mg/L)

(a)

0.5

0.4

0.3

1500

1000

500

0.2

0 0

100

200

300

400

0

100

200

Time (h)

(d) 0.35

600 25 % 50 % 75 % 100%

500

300

200

100

0.8 0.20 0.6 0.15 0.4 0.10 0.2

0.00 100

200

Time (h)

300

400

1.0

0.25

0.05

0

250

1.2 Cell DW yield Acetate production yield Acetate production rate

0.30

400

0

400

Time (h)

Cell DW yield (g/g CO)

Acetate concentration (mg/L)

(c)

300

0.0 25%

50%

75%

200

150

100

50

Acetate production rate (mg/L/day)

þ

3 X

Acetate production yield (g/g CO)

Y ¼ a0 þ

of anaerobic modified M9 medium (CaCl2 0.015 g/L, Na2HPO4 6 g/L, KH2PO4 3 g/L, NaCl 0.5 g/L, NH4Cl 1 g/L, MgSO4 0.5 g/L, and yeast extract (BD, USA) 0.5 g/L). The mixed gas ratio was adjusted at the optimum condition obtained by the RSM analysis, and the flow rate was maintained at 15 mL/min by the gas mixer. An automatic pH controller was used to maintain pH at the optimum point obtained by the RSM analysis. Temperature was maintained at 37 °C and the stirring rate was maintained at 200 rpm.

0

100%

CO contents (%)

Fig. 1. Effect of the CO concentration on cell DW (a), CO removal (b), acetic acid production (c), and the profile of cell DW yield, acetic acid production yield, and acetic acid production rate (d).

481

C.W. Nam et al. / Bioresource Technology 211 (2016) 478–485

3.2. Effect of carbon dioxide on acetate production in batch condition

volumetric mass transfer coefficient under different [CO] conditions shows that the mass transfer limitation was miscible in the batch system (noted that D [cell DW]/[CO] was 0.25, 0.23, 0.17, and 0.12 g/g CO at [CO] = 25%, 50%, 75%, and 100%, respectively). Under all [CO] conditions, the lag phase of CO consumption continued until 72–108 h (Fig. 1b). After the lag phase, CO was rapidly consumed at [CO] = 25% and 50%, but CO was still almost constant at [CO] = 75% and 100%. The initial CO consumption may have been devoted to fulfilling the energy requirement for biomass accumulation; after the cell growth stabilized, the cells converted CO to metabolites such as HAc, propionate, and ethanol. The maximum CO consumption rate was 4.71 mg/L/h at [CO] = 25%. The HAc production increased over the reaction without the initial lag phase (ca. 100 mg/L of HAc at the first phase of fermentation) because microorganisms easily utilized the yeast extract rather than the CO source (Fig. 1c). After adaptation to the CO condition, actual HAc production from CO began. HAc concentration increased as [CO] increased. The highest total HAc concentration was 543.89 mg/L at [CO] = 75%, but the highest HAc production yield and rate were 0.663 g/g CO and 198.76 mg/L/day, respectively, at [CO] = 25%. Under high [CO] condition, ethanol production increased: 18.95 mg/L at 25%, 41.00 mg/L at 50%, 103.84 mg/ L at 70%, and 209.09 mg/L at 100%. Therefore, the condition at [CO] = 25% was chosen for the highest HAc production in the gas fermentation.

(b)

0.7 0% 25 % 50 % 75 %

0.6

0.5

0% 25 % 50 % 75 %

0.4

0.3

0.2

500

400

300

200

100

0.1

0 20

40

60

80

100

120

140

160

180

0

20

40

60

Time (h)

(d)

700

Acetate concentration (mg/L)

120

140

160

500

400

300

200

1.0

0.6 0.8

0.4

0.6

0.4 0.2 0.2

100

0

0.0 0

20

40

60

80

100

Time

180

1.2

0.8 Cell DW yield Acetate production yield Acetate production rate

0% 25 % 50 % 75 %

600

100

Time (h)

Cell DW yield (g/g CO)

(c)

80

120

140

160

180

0.0 0%

25%

50%

250

200

150

100

50

Acetic acid production rate (mg/L/day)

0

Acetate production yield (g/g CO)

Cell dw (g/L)

700

600

CO concentraion (mg/L)

(a)

Supplementation of CO2 to the gas fermentation system caused decreasing the lag phase and increasing the final cell DW (Fig. 2a). The lag phase was reduced from 60 h at [CO2] = 0% to 48 h and 36 h at [CO2] = 25% and 50%, respectively, because the Wood-Ljungdahl pathway commenced with reduction of CO2 to formate by formate dehydrogenase (FDH) (Mohammadi et al., 2011). Thus the presence of CO2 triggered the carbon fixation reaction and, consequently, reduced the lag phase for CO consumption in the gas fermentation (Fig. 2b). After the lag phase, CO was rapidly consumed and the CO consumption rate increased as [CO2] increased from 25% to 50%. When [CO2] exceeded the specific level (i.e., 50%), the highest CO consumption rate (0.696 mL/h) was maintained at both [CO2] = 50% and 75%. Supplementation of CO2 increased HAc production rate and HAc production yield (Fig. 2c). HAc production generally increased as [CO2] increased: 460.15 mg/L at [CO2] = 0%, 533.18 mg/L at [CO2] = 25%, and 581.92 at [CO2] = 50%. However, at [CO2] = 75%, HAc production (554.23 mg/L) was lower than that at [CO2] = 50%. At [CO2] = 50% condition, the batch culture obtained the highest HAc production yield and rate (0.929 g/g CO, and 225.11 mg/L/d, respectively). Even though [CO2] did not affect ethanol production, as [CO2] increased, propionate production was diminished (125.77 mg/L at 0%, 104.29 mg/L at 25%, 66.17 mg/L at 50%, and

0

75%

CO2 contents (%)

Fig. 2. Effect of the CO2 concentration on cell DW (a), CO removal (b), acetic acid production (c), and the profile of cell DW yield, acetic acid production yield, and acetic acid production rate (d).

482

C.W. Nam et al. / Bioresource Technology 211 (2016) 478–485

(b)

1.0 pH pH pH pH

Cell dw (g/L)

0.8

4 6 8 10

600 pH 4 pH 6 pH 8 pH 10

500

CO concentration (mg/L)

(a)

0.6

0.4

400

300

200

100

0.2

0 20

40

60

80

100

120

0

20

40

60

Time (h)

(d)

1000 pH pH pH pH

800

100

4 6 8 10

600

400

200

0

1.2

1.2 Cell DW yield Acetic acid production yield

1.0

1.0

0.8

0.8

0.6

0.6

0.4

0.4

0.2

0.2

0.0 0

20

40

60

120

Time (h)

Cell DW yield (g/g CO)

Acetate concentration (mg/L)

(c)

80

80

100

120

Acetic acid production yield (g/g CO)

0

0.0 4

Time (h)

6

8

10

pH

Fig. 3. Effect of the pH on cell DW (a), CO removal (b), acetic acid production (c), and the profile of cell DW yield, acetic acid production yield, and acetic acid production rate (d).

70.83 mg/L at 75%). Hence the highest relative content of HAc (97.7%) in total metabolite was achieved at at [CO2] = 50%. The Wood-Ljungdahl pathway generates no net ATP synthesis needed for cell growth, so the electron transport system is needed to obtain additional ATP for cell growth (Hurst and Lewis, 2010). In a proposed electron transport chain in acetogenic bacteria (Hurst and Lewis, 2010), electron exchange between CO dehydrogenase (CODH) and flavoprotein occurs for ATP generation. Since the rate of ATP generation was directly proportional to the ratio of [CO]/ [CO2], the supplementation of CO2 induced enhancing HAc production in the gas fermentation.

Table 1 Comparison of fermentation results under different CO, CO2 and pH. Parameters

3.3. Effect of pH on acetate production in batch condition As pH increased, cell DW also increased, but CO consumption rate decreased. Cell DW was almost twice as high at pH 4 and 6 (0.74 and 0.79 g/L, respectively) than at pH 8 (0.49 g/L) and pH 10 (0.38 g/L) (Fig. 3a). CO consumption rate was highest at pH 8 (11.788 mg/L/h), intermediate at pH 4 and 6 (4.715 mg/L/h), and

Cell DW (g/L)

Acetate (mg/L)

Propionate (mg/L)

Ethanol (mg/L)

CO (%) 25 50 75 100

4.71 4.53 4.00 3.34

0.42 0.53 0.55 0.52

374.89 439.20 535.87 521.73

87.67 0.00 117.71 54.96

18.85 41.00 103.84 209.09

CO2 (%) 0 25 50 75

6.21 7.25 8.70 8.70

0.30 0.33 0.38 0.39

447.17 533.18 577.43 554.24

125.77 104.29 66.17 70.83

19.92 15.56 14.67 15.52

4.715 4.715 11.788 3.188

0.74 0.79 0.49 0.38

582.87 509.29 671.63 331.34

54.43 29.55 0 25.84

31.55 25.68 0 16.73

pH 4 6 8 10* *

CO consumption rate (mg/L/h)

At pH 10, 19.02 mg/L of formate was produced additionally.

483

C.W. Nam et al. / Bioresource Technology 211 (2016) 478–485 Table 2 The Box–Behnken experimental design with three independent variables. Run

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

CO (%)

CO2 (%)

pH

A

Code

B

Code

C

Code

25 40 25 25 25 25 40 25 10 10 10 10 40 40 25

0 1 0 0 0 0 1 0 1 1 1 1 1 1 0

40 40 60 60 40 40 60 20 20 60 40 40 40 20 20

0 0 1 1 0 0 1 1 1 1 0 0 0 1 1

8 9 9 7 8 8 8 9 8 8 9 7 7 8 7

0 1 1 1 0 0 0 1 0 0 1 1 1 0 1

Acetate production yield Y (mg/g CO) 386.23 56.30 57.12* 255.11 406.34 417.18 166.93 93.42* 219.40 245.98 159.61* 395.04 155.51 148.20 287.73

* The acetic acid production yield was calculated by Eq. (1). The negative number results shows the lower yield than the control (0% of CO).

Table 3 ANOVA for response Surface quadratic model analysis of variance table. Source

Sum of squares

df

Mean square

F

Prob > F

Model A-CO B-CO2 C-pH AB AC BC A2 B2 C2 Residual Lack of fit Pure error Cor total

478,800 3778.75 300.06 226,900 15.37 51856.39 1187.20 34880.01 45434.98 139,300 2909.53 2416.52 493.01 481,700

9 1 1 1 1 1 1 1 1 1 5 3 2 14

53197.53 3778.75 300.06 226,900 15.37 51856.39 1187.20 34880.01 45434.98 139,300 581.91 805.51 246.51

91.42 6.49 0.52 389.89 0.026 89.11 2.04 59.94 78.08 239.41

<0.0001 0.0514 0.5048 <0.0001 0.8773 0.0002 0.2126 0.0006 0.0003 <0.0001

3.27

0.2431

Y ¼ 403:25  21:73A þ 6:12B  168:40C  1:96AB þ 113:86AC þ 17:23BC  97:19A2  110:93B2  194:25C 2 ; ð8Þ

lowest at pH 10 (3.188 mg/L/h) (Fig. 3b). HAc production started immediately under all pH conditions (Fig. 3c), however, the highest HAc production (671.63 mg/L) was obtained at pH 8. The HAc production at pH 4 and pH 6 was 532.86 and 509.28 mg/L, respectively, even though the initial HAc production was similar to that at pH 8. Final HAc production was the lowest (331.34 mg/L) at pH 10 because CO was partially utilized for HAc production and cell growth. The pH affects the enzymatic activity related to ATP production in the CO gas fermentation. Microorganisms and the gas fermentation system preferred to grow in pH 4–6, so most of CO gas was consumed for ATP generation by CODH. As pH increased to slightly basic pH 8, cell growth was partially inhibited, but HAc production was reinforced. However, at basic pH 10, cell growth, CO consumption, and HAc production were all diminished because CODH and FDH activities were strongly limited. In the Wood-Ljungdahl pathway, the first CO2 reduction by FDH was optimized under slightly basic conditions (pH 6–9) (Andreesen and Ljungdahl, 1974; Hatrongjit and Packdibamrung, 2010). The pH also influences subsequent reactions of phosphotransacetylase (PTA) and acetate kinase (AK) activities (Zhu and Yang, 2004). As pH increased, PTA and AK enzyme activities was enhanced, therefore the production of other metabolites under basic condition was lower than under acidic condition. Therefore, pH 8, which achieve highest HAc yield and selectivity, was chosen as the best pH for HAc production (Table 1). 3.4. Response surface method analysis To achieve the maximum HAc production, the optimum levels of [CO], [CO2] and pH, and the effect of their interactions on the gas fermentation system were determined by RSM with the Box– Behnken design. Each [CO], [CO2] and pH variable significantly affects HAc production yield in the gas fermentation, and the multiple relation of each factor also causes unpredictable results. Thus the ‘one-variable-at-a-time’ approach was applied to this study (Table 2). At the center point ([CO] = 25%, [CO2] = 40%, pH 8) chosen in the previous experiment, the response was dry cell weight = 213.42 mg/L, HAc production = 411.63 mg/L, and HAc production yield = 403.25 mg/g CO. A quadratic model was suggested to fit the data by least-squares, and all coded values regardless of their significance were included in the following equation:

where Y [mg/L] is predicted HAc production yield, and A, B and C are the coded variables of CO, CO2, and pH, respectively. Analysis of variance (ANOVA) was conducted to analyze the significance of the fit of the second-order polynomial equation to the data (Table 3). The ANOVA model had F = 91.42, with P (>F) < 0.0001, and was highly significant. C, AC, A2, and C2 were significant terms; i.e., pH, and [CO] were more influential than the other variables. The effect of [CO2] was not as important as pH and [CO]. The model had R2 = 0.994, which means that it explains 99.4% of the variability of the response. These results means that the predicted values and its model were reliable for HAc production. The optimum level of each condition and the effect of their interactions on HAc production were plotted as threedimensional response surface and two-dimensional contour plots (Fig. 4). The figures were based on Eq. (7) with one variable held at the optimum condition and the others varied. In Fig. 4a, the effect of [CO] (A) and [CO2] (B) on the HAc production yield was investigated holding pH (C = 7.92) constant. Y increased as [CO] increased from 10% to 25%, then decreased as [CO] increased further, because at [CO] > 25% it becomes toxic. The optimum [CO] and [CO2] concentrations were 21.39%, and 7.65%, respectively, but change in [CO2] caused less change than did change in [CO]. pH had the greatest influences on HAc production yield (Fig. 4b, c) because pH affects CO consumption and the enzyme that produces HAc. The optimum conditions for HAc production were [CO] = 20.81%, [CO2] = 41.38%, and pH = 7.18, at which the predicted HAc production yield was 434.26 mg/g CO, which is 81.1% the theoretical yield of 535.71 mg/g CO (1 mol HAc per noted 4 mol CO). Thus the response surface optimization could be successfully used to evaluate the HAc production efficiency. 3.5. Gas fermentation system for CO and CO2 utilization in liquid batch and gas continuous operation In the liquid batch and gas continuous operation system, the exponential increase phase lasted until 120 h, when the cell growth reached 1.6 g/L. Organic acid production began at 48 h; HAc was primarily produced, and other components such as propionate and ethanol were also formed in small quantity (Fig. 5). Since microorganisms do not prefer increasing acidity by HAc production that inhibits the cell growth, microbial metabolic reaction tends to shift toward production of nongrowth-associated metabolites such as ethanol, propionic acid, and butyric acid, in order to hold increasing acidity during

484

C.W. Nam et al. / Bioresource Technology 211 (2016) 478–485

Fig. 4. Three dimensional surface and two dimensional contour plots of acetic acid production yield against CO and CO2 concentrations (a), CO2 concentration and pH (b), and CO concentration and pH (c). Each constant level was indicated on the plot.

the gas fermentation (Robinson et al., 2001). Thus, as the gas fermentation progressed, production of ethanol and propionate were increased, whereas HAc production slightly decreased. When the total acid concentration reached 30 g/L, all production of acids and ethanol stopped and the cell growth also slightly decreased. The final cell growth was 1.66 g/L and the

concentrations of HAc, propionic acid, and ethanol were 23.6, 3.11, and 3.42 g/L, respectively. As higher HAc production (>21 g/L) leaded to product inhibition, all metabolic activity ceased after 528 h as well as the cell biomass began to dwindle. The relative content of HAc accounted for 78.3% of the total organic acid under the condition obtained by RSM analysis

485

C.W. Nam et al. / Bioresource Technology 211 (2016) 478–485 5

35 Biomass Lactic acid Formic acid Acetic acid Proponic acid Ethanol Total organic products

25

4

3

20

15

2

10

Biomass production (g/L)

Product concentration (g/L)

30

1 5

0

0 0

2 00

40 0

60 0

80 0

Time (h) Fig. 5. Gas fermentation system for CO conversion in liquid batch and gas continuous operation.

and its concentration (23.6 g/L) was higher than those of reported earlier (19.16 g/L) (Mohammadi et al., 2011).

4. Conclusions Gas fermentation has a significant opportunity to produce commodity chemicals from waste CO gas. The variables such as gas contents and pH influenced the HAc production yield in the mixed culture CO gas fermentation; however, the pH was the most significant variable in the system. Continuous gas fermentation under the optimum condition ([CO] = 20.81%, [CO2] = 41.38%, and pH 7.18) obtained higher HAc production (23.6 g/L) than the previously reported (19.16 g/L). Thus, gas fermentation using mixed culture can be a promising way to utilize CO gas and these results can be used to guide development of gas fermentation process. Acknowledgements This work was supported by the BK21+ program through the National Research Foundation of Korea funded by the Ministry of Education, Science and Technology, the Marine Biotechnology Program (Marine BioMaterials Research Center) funded by the Ministry of Oceans and Fisheries, Korea, POSCO and the Korea Institute of Energy Technology Evaluation and Planning (KETEP) grant funded by the Korea Government Ministry of Knowledge Economy (No. 2012K130), and the Research and Development Program of the Korea Institute of Energy Research (KIER) (B4-247402). References Andreesen, J.R., Ljungdahl, L.G., 1974. Nicotinamide adenine dinucleotide phosphate dependent formate dehydrogenase from Clostridium thermoaceticum: purification and properties. J. Bacteriol. 120, 6–14. Annadurai, G., Mathalai Balan, S., Murugesan, T., 1999. Box–Behnken design in the development of optimized complex medium for phenol degradation using Pseudomonas putida (NICM 2174). Bioprocess. Eng. 21, 415–421.

Batlle-Vilanova, P., Puig, S., Gonzalez-Olmos, R., Balaguer, M.D., Colprim, J., 2016. Continuous acetate production through microbial electrosynthesis from CO2 with microbial mixed culture. J. Chem. Technol. Biotechnol. 91, 921–927. Guo, Y., Xu, J., Zhang, Y., Xu, H., Yuan, Z., Li, D., 2010. Medium optimization for ethanol production with Clostridium autoethanogenum with carbon monoxide as sole carbon source. Bioresour. Technol. 101, 8784–8789. Hatrongjit, R., Packdibamrung, K., 2010. A novel NADP+-dependent formate dehydrogenase from Burkholderia stabilis 15516: screening, purification and characterization. Enzyme Microb. Technol. 46, 557–561. Heiskanen, H., Virkajärvi, I., Viikari, L., 2007. The effect of syngas composition on the growth and product formation of Butyribacterium methylotrophicum. Enzyme Microb. Technol. 41, 362–367. Hurst, K.M., Lewis, R.S., 2010. Carbon monoxide partial pressure effects on the metabolic process of syngas fermentation. Biochem. Eng. J. 48, 159–165. Jankowska, E., Chwiałkowska, J., Stodolny, M., Oleskowicz-Popiel, P., 2015. Effect of pH and retention time on volatile fatty acids production during mixed culture fermentation. Bioresour. Technol. 190, 274–280. Jo, J.H., Lee, D.S., Park, D., Park, J.M., 2008. Statistical optimization of key process variables for enhanced hydrogen production by newly isolated Clostridium tyrobutyricum JM1. Int. J. Hydrogen Energy 33, 5176–5183. Kleerebezem, R., van Loosdrecht, M.C., 2007. Mixed culture biotechnology for bioenergy production. Curr. Opin. Biotechnol. 18, 207–212. Kundiyana, D.K., Huhnke, R.L., Maddipati, P., Atiyeh, H.K., Wilkins, M.R., 2010a. Feasibility of incorporating cotton seed extract in Clostridium strain P11 fermentation medium during synthesis gas fermentation. Bioresour. Technol. 101, 9673–9680. Kundiyana, D.K., Huhnke, R.L., Wilkins, M.R., 2010b. Syngas fermentation in a 100-L pilot scale fermentor: design and process considerations. J. Biosci. Bioeng. 109, 492–498. Mohammadi, M., Najafpour, G.D., Younesi, H., Lahijani, P., Uzir, M.H., Mohamed, A. R., 2011. Bioconversion of synthesis gas to second generation biofuels: a review. Renewable Sustainable Energy Rev. 15, 4255–4273. Robinson, T., Singh, D., Nigam, P., 2001. Solid-state fermentation: a promising microbial technology for secondary metabolite production. Appl. Microbiol. Biotechnol. 55, 284–289. Sim, J.H., Kamaruddin, A.H., Long, W.S., 2008. Biocatalytic conversion of CO to acetic acid by Clostridium aceticum––medium optimization using response surface methodology (RSM). Biochem. Eng. J. 40, 337–347. Temudo, M.F., Kleerebezem, R., Van Loosdrecht, M., 2007. Influence of the pH on (Open) mixed culture fermentation of glucose: a chemostat study. Biotechnol. Bioeng. 98, 69–79. Vega, J.L., Clausen, E.C., Gaddy, J.L., 1990. Design of bioreactors for coal synthesis gas fermentations. Resour. Conserv. Recycl. 3, 149–160. Younesi, H., Najafpour, G., Mohamed, A.R., 2005. Ethanol and acetate production from synthesis gas via fermentation processes using anaerobic bacterium, Clostridium ljungdahlii. Biochem. Eng. J. 27, 110–119. Zhu, Y., Yang, S.T., 2004. Effect of pH on metabolic pathway shift in fermentation of xylose by Clostridium tyrobutyricum. J. Biotechnol. 110, 143–157.