- Email: [email protected]
Integrated ABE fermentation-gas stripping process for enhanced butanol production from sugarcane-sweet sorghum juices
Biomass and Bioenergy 98 (2017) 153e160
Contents lists available at ScienceDirect
Biomass and Bioenergy journal homepage: http://www.elsevier.com/locate/biombioe
Research paper
Integrated ABE fermentation-gas stripping process for enhanced butanol production from sugarcane-sweet sorghum juices n, Mario Daniel Ferrari, Claudia Lareo* Eloísa Rocho Depto. Bioingeniería, Facultad de Ingeniería, Universidad de la República, J. Herrera y Reissig 565, CP 11300, Montevideo, Uruguay
a r t i c l e i n f o
a b s t r a c t
Article history: Received 18 August 2016 Received in revised form 26 November 2016 Accepted 4 January 2017
The biobutanol production by ABE (acetone-butanol-ethanol) fermentation with an integrated product recovery system by gas stripping was studied. A mixture of industrial sugarcane-sweet sorghum juices was fermented using Clostridium acetobutylicum DSM 792. Empirical models were developed to determine an operational condition for the integrated process that contributes both to mitigate the butanol inhibitory effect and to obtain a highly concentrated butanol condensate. A Monod model supplemented with a term describing product inhibition was employed to describe cell growth, butanol formation, and substrate consumption, respectively. Gas stripping was described by a first order model. The models showed satisfactory agreement with the experimental data in terms of cell growth, sugar consumption, and butanol production and extraction. A gas recycle flowrate in the range 0.3e0.6 vvm allowed to maintain butanol concentration below the inhibitory concentration (8 g/L) and to obtain a concentrated butanol condensate after phase separation, which could reduce energy consumption in the final product recovery. In a fed-batch fermentation coupled with in situ gas stripping, total sugar conversion and 18.6 g/L butanol distributed 42% in the fermentation broth and 58% in the condensate, were obtained. © 2017 Elsevier Ltd. All rights reserved.
Keywords: Biobutanol ABE fermentation Clostridium Gas stripping Sugarcane Sweet sorghum Kinetic model
1. Introduction During the last decades, there has been an increasing interest for the production of chemicals and fuels from renewable resources. Reasons for this include climate change, global warming, and energy security [1]. Butanol is considered an attractive biofuel as it has clearly superior properties to ethanol due to its higher energy density, less volatile and explosive and less hygroscopic [1e4], in addition to being an important chemical precursor for paints, polymers and plastics, and widely used solvent [5]. Biobutanol can be produced by the acetone-butanol-ethanol (ABE) fermentation by various Clostridium spp., in which a solvent mixture is produced, generally in the ratio 3:6:1 acetone-butanol-ethanol respectively. The economic viability of fuel biobutanol production largely depends on the costs of raw materials and energy consumed. Energy consumption can be high due to the low butanol titer and purity reached in the fermentation broth [1,3]. The raw material utilized in biobutanol production is one of the major cost factors in its production [3e5]. Both sugarcane and
* Corresponding author. E-mail address: clareo@fing.edu.uy (C. Lareo). http://dx.doi.org/10.1016/j.biombioe.2017.01.011 0961-9534/© 2017 Elsevier Ltd. All rights reserved.
sweet sorghum can offer more advantages than other crops since they produce a residue (bagasse) which can be burnt for steam production to satisfy the energy demand of the industrial processes [6,7]. Therefore, raw materials with high carbohydrate content, efficient transformation processes energetically optimized, and an accessible, low cost energy source are needed. Fuel bioethanol is currently produced in the north of Uruguay mainly from sugarcane. Sweet sorghum juice alone or mixed with sugarcane juice is also used to extend the plant facility working time. Since both biobutanol and bioethanol could be produced from the same feedstock, its production could take place in the same n, http://www. facility using similar equipment (Alur SA, Bella Unio alur.com.uy/agroindustrias/bella-union/, access on 11/21/2016). One of the major challenges in biobutanol production is the intensive energy consumption in product recovery caused by the low product concentrations reached in the fermentation broth [8]. This is attributed mainly to cell toxicity or fermentation inhibition by butanol [9e12]. Furthermore, the butanol yield and productivity are also low, due to the co-production of acetone and ethanol. In order to solve this problem, extensive research and development efforts have been made. Some approaches that have been proposed include improve butanol tolerance of strains by mutagenesis and metabolic engineering [1,5,13,14], to couple butanol production
154
n et al. / Biomass and Bioenergy 98 (2017) 153e160 E. Rocho
with an in situ extraction process. Different techniques for in situ butanol recovery process (adsorption, liquid-liquid extraction, pervaporation, vacuum flash fermentation and gas stripping) have been evaluated as promising techniques to improve productivity by mitigating butanol inhibition [15e19]. Even these techniques are more economically competitive comparing to conventional distillation for butanol recovery from the dilute fermentation broth, the deficiency in recovering a high titer of butanol is still the challenge. Gas stripping is a physical separation process where target compounds are removed from an aqueous solution by bubbling gas through it [20]. It has many advantages over the other methods: simple operation, no harm to culture, and does not requires any chemicals nor membranes. It also collaborates with the agitation and homogenization of the system [10,13,21e27]. However, its potential in butanol recovery and energy saving for biobutanol production has not yet been fully explored nor optimized [13,28]. Most of previous studies on integrated ABE fermentation with in situ gas stripping have been conducted at relatively low butanol concentrations (5 g/L or less) to minimize butanol toxicity [9]. Several works reported that the gas flowrate [24,28] and the butanol titer in the feed solution affect gas stripping process [20,28]. Xue et al. [28] reported that it is necessary to conduct gas stripping at a butanol titer higher than 8 g/L in the feed solution in order to obtain a condensate with a butanol titer higher than its solubility in water (~80 g/L at 20 C), which will result in phase separation and in a more energy-efficient butanol recovery process. On the other hand, the recycling of large amount of gas in industrial reactors would be uneconomical. The effect of gas recycle rate should be evaluated in order to optimize gas stripping conditions as an interest in commercialization this technology. The aim of this work was to evaluate butanol production from a mixture of sugarcane-sweet sorghum juices using C. acetobutylicum DSM 792 in an integrated process (fermentation with in situ gas stripping) that contributes both to mitigate the butanol inhibitory effect and to obtain a highly concentrated butanol condensate. Kinetics models describing butanol formation by fermentation and butanol extraction by gas stripping were developed to study the effect of gas recycle rate on the performance of the fermentation coupled with in situ gas stripping and to determine the most favorable operational condition. 2. Materials and methods 2.1. Experimental assays 2.1.1. Characterization of raw material A mixture of concentrated industrial juices of sugar cane (75%) n, and sweet sorghum (25%) was provided by Alur SA (Bella Unio Uruguay) and it was stored at 5 C. Table 1 shows its composition. 2.1.2. Microorganism, inoculum preparation and fermentation media Clostridium acetobutylicum DSM 792 was used in all fermentations. A stock culture was maintained in Reinforced Clostridial Medium (RCM) at 4 C. A pre-culture was prepared by inoculating 10 mL of the stock culture into a serum bottle containing 100 mL RCM medium and incubating at 37 C for 24e48 h until active growth was observed. Glucose-based and the industrial media were used, containing 50e60 g/L of total sugar concentration expressed as glucose equivalent and 1% (v/v) P2 solutions. The P2 solutions contains K2HPO4 50 g/L, KH2PO4 50 g/L, ammonium acetate 220 g/L, p-amino benzoic acid 0.1 g/L, thiamine 0.1 g/L, biotin 0.001 g/L, MgSO4.7H2O 20 g/L, MnSO4.H2O 1 g/L, FeSO4.7H2O 1 g/L, NaCl 1 g/L. The seed culture for the fermentation was prepared in 250-mL
Table 1 Composition of concentrated sugarcane-sweet sorghum juices. Component
Unit
Value
Sucrose Glucose Fructose Ethanol Acetic acid Butyric acid Succinic acid cis-Aconitic acid trans-Aconitic acid Furfural 5-Hidroximetilfurfural Phosphorous Nitrogen (Kjeldahl) Aluminum Calcium Cooper Iron Magnesium Manganese Potassium Sodium Zinc Ash Moisture Density (20 C)
% % % % % % % % % % % % % % % % % % % % % % % % kg/m3
75 5 3 0.69 0.11 0.34 0.15 0.13 1.05 nd nd <0.5 0.31 <0.002 0.18 <0.0005 0.0047 0.11 0.0012 0.76 0.04 <0.0005 2.5 32 1330
% percentage in dry weight except moisture. nd: no detected (detection limit: 0.003%).
bottles containing 100 mL of the same medium used in the fermentation but adjusted to reach half of the total sugar concentration. The pH was adjusted to 6.0 ± 0.1. The medium supplemented with 1 g/L of yeast extract was swept with O2-free N2 over the headspace of the bottles. It was sterilized at 121 C during 15 min. On cooling to 37 C, 1% (v/v) of filter-sterilized P2 stock solutions were added followed by inoculation with 10% (v/v) highly active cells that were grown in RCM. The culture was incubated at 37 C, 150 rpm for 20e24 h until the cell density reached an optical density (OD) value of ~3. 2.1.3. Batch fermentation Fermentation tests with glucose-based medium were performed in bottles of 250 mL with 100 mL of medium containing 50e60 g/L of glucose. The pH was adjusted initially to 6.0 ± 0.1. The medium was swept with O2-free N2 over the headspace of the bottles. It was sterilized at 121 C during 15 min. On cooling to 37 C, 1% (v/v) of filter-sterilized P2 stock solutions were added, followed by inoculation with 9% (v/v) highly motile cells. The bottles were incubated in an orbital shaker at 150 rpm and 37 C. Samples were withdrawn at intervals for analysis. Tests were conducted in duplicate. A batch fermentation was also performed in a 5 L-bioreactor (Infors HT, Switzerland) containing 2 L of the industrial juices diluted to reach 75 g/L of total sugar concentration. The pH was adjusted to 6.0 ± 0.1. The medium was swept with O2-free N2 over the headspace of the bioreactor. It was sterilized at 121 C during 15 min. On cooling to 37 C, 1% (v/v) of filter-sterilized P2 stock solutions were added followed by inoculation with 6% (v/v) highly active cells. Fermentation was performed at 37 C and 150 rpm. No antifoam was necessary during the fermentation. 2.1.4. Fed batch fermentation coupled with in-situ gas stripping A schematic diagram of the integrated reactor set up is shown in Fig. 1. Prior to the gas stripping experiment, the condenser and gascirculation line were flushed with O2-free N2 gas. A 5-L bioreactor (Infors HT, Switzerland) containing 2.5 L of the industrial medium
n et al. / Biomass and Bioenergy 98 (2017) 153e160 E. Rocho
Fig. 1. Schematic diagram of the integrated reactor set up.
(total sugar concentration 60 g/L) was inoculated with 6% (v/v) highly active cells of C. acetobutylicum DSM 792. The pH was adjusted initially to 6.0 ± 0.1. The temperature was controlled at 37 C and the agitation at 150 rpm. The fermentation was allowed to proceed in batch mode for 48 h (until the butanol concentration was approximately 8 g/L), after which gas stripping was applied at a flowrate of 1 L/min by recycling the off gasses (CO2 /H2) through the system using a twin-head Masterflex® peristaltic pump and 18 size Tygon pump tubing (Cole-Parmer). The ABE vapors were cooled in a condenser to 0 C, using a solution of glycerol-water (30% (v/v)) cooled in a refrigerated circulating bath (Polystat, Cole-Parmer). The condensate was collected in a flask immersed in the refrigerated circulating bath, which was pumped into a 250-mL flask which was used as a storage tank for phase separation. Samples were aseptically withdrawn at intervals for brix grades, sugars, products and optical density analysis. When residual sugars decreased to ~10 g/L (about 48 h of fermentation), concentrated juice solution (600 g/L total sugars) was added to reach a sugar concentration close to 60 g/L. No antifoam was necessary during the fermentation. 2.1.5. Gas stripping assays Gas stripping experiments were done in a 5-L bioreactor (Infors HT, Switzerland) (Fig. 1) with 2.5 L medium. Experiments with and without cells were performed. For the experiments without cells, two solutions were prepared by adding butanol to distilled water and to the industrial juices to reach a concentration of 6 g/L. Gas stripping was performed by recycling the headspace gasses through the solutions at a flow rate of 1 L/min (0.4 vvm). For the assays with cells, the fermented medium containing 1.6 g/L of cells and 6 g/L of butanol was utilized. Gas stripping was performed by recycling the headspace gasses through the solutions at a flow rate of 1 L/min (0.4 vvm) or 0.5 L/min (0.2 vvm). The condenser was maintained at 0 C and the bioreactor temperature and agitation were 37 C and 150 rpm, respectively. Samples were aseptically withdrawn at intervals (~1 h) for butanol analysis. A control test of the gas stripping system was also performed by recycling the headspace gasses in order to determine the butanol mass recovery. The model solution was prepared with the industrial juices and the addition of butanol at the same concentration as in the fermentation broth (6 g/L) and at a gas flow rate of 0.4 vvm. The condenser was maintained at 0 C. The bioreactor was operated at 37 C and 150 rpm. Samples were taken from the bioreactor and condensate storage flask periodically (after ~15e20 h) and then were analyzed to calculate the butanol mass recovery. 2.1.6. Analytical methods The fermentation evolution was monitored by brix grades using
155
a refractometer (Master, Atago) as an estimation of sugars content and by optical density (OD) in a spectrophotometer (Unico, UV2100) at 600 nm as a measured of cell density. Microscopic observation was periodically done to observe the morphology and motility of the cells. Dry cell weight (DCW) was determined by measuring OD at 600 nm: one unit of OD corresponded to 0.37 ± 0.01 g/L of DCW. After removing the bacterial cells by centrifugation at 6500 rpm for 10 min, the clear fermentation broth was analyzed for sugar and product concentrations. Sucrose, glucose and fructose concentrations were determined by HPLC (Shimadzu, Kyoto, Japan) using a Shodex SUGAR KS-801 column at 55 C, ultrapure water as eluent at a flow rate of 0.7 mL/min and a refractive index detector (RID-10A). Total sugars were expressed in glucose equivalents (glucose þ fructose þ 1.05 x sucrose). Acetone, ethanol and organic acids (acetic, butyric and succinic), were measured with a gas chromatograph (GC, Shimadzu GC-2010) equipped with a flame ionization detector and a fused silica column (RTX®-Wax, 30 m long, 0.5 mm film thickness and 0.32 mm ID, Restek). Hydroxymethylfurfural and furfural were determined by HPLC using a Phenomenex ROA-Organic Acid Hþ column and an UV detector at 210 nm cis-Aconitic acid and transaconitic acid were also determined by HPLC, but using a Shodex RSpack KC-811 column. Sodium, potassium, calcium, magnesium, manganese, aluminium, copper and zinc were determined by atomic emission spectrometry, iron by atomic absorption spectrometry, proteins by Kjeldahl, and phosphorus by the spectrophotometric method using the molybdenum blue reaction. The moisture and ash contents were determined at 60 C and 550 C, respectively. 2.2. Process models The Monod kinetic model is the most widely used empirical model that describes microbial growth under substrate-limiting conditions. The butanol accumulated in the culture broth tends to restrain the cell growth, and therefore, the Monod model was modified by introducing the term of product inhibition [29]. The model was also modified by introducing another term which includes a cell death constant parameter, which has been justified by several authors as a drop in biomass concentration once the solventogenic phase had started [30,31]. Therefore, the equations below were developed as follows:
dX dt
¼
Ks þ S
dS m X ¼ dt YX=S dP dt
¼
mm S
¼
m X YP=S YX=S
1
X
mm S
P Kp
kd X
Ks þ S ¼
a
1
mm S Ks þ S
P Kp
1
a
P Kp
(1)
X YX=S a X
(2)
YP=S YX=S
(3)
where X is the DCW (g/L), m is the specific growth rate (h1), mm is the maximum specific growth rate (h1), S is the growth-limiting substrate concentration (g/L), Ks is the substrate saturation constant (g/L), kd is the specific cell death rate (h1), P is the butanol concentration (g/L), Kp is the product concentration at which no cell growth occurs (g/L), a is the degree of product inhibition, YX/S (g/g) is the biomass yield coefficient and YP/S (g/g) is the butanol yield coefficient. Gas stripping of butanol and other organic solvents from an aqueous solution can be described, macroscopically, by a first order process reported by Truong and Blackburn [32] as shown by Eq. (4), which has been verified in many studies [20,21,24,32].
n et al. / Biomass and Bioenergy 98 (2017) 153e160 E. Rocho
156
3. Results and discussion
dP ¼ dt
rs
¼ ks a P
(4)
where rs represents the stripping rate of the butanol (g/Lh), P and ksa are butanol concentrations in the aqueous solution (g/L) and stripping rate coefficient (h1), respectively. The stripping rate coefficient has been modeled by the equation shown in Eq. (5) in which this parameter is linearly proportional to the gas flow rate [20,24,32].
Q ðHc Þm V
ks a ¼ b
(5)
where b and m are power function constants, V is the total liquid volume of the fermentation broth (L), Q is the gas flow rate (L/min) and Hc is the Henry's law constant (weight of butanol per mL air/ weight of butanol per mL water). The integrated process of fermentation coupled with in situ gas stripping can be modeled using Eq. (6).
dP dt
¼ ¼
m X YP=S YX=S
mm S Ks þ S
ks a P 1
P Kp
a X
YP=S ks a P YX=S
(6)
At steady state (dP/dt ¼ 0), and if ksa can be modeled as Eq. (5), the following relationship is obtained:
mm S
Ks þ S
1
P Kp
a X
YP=S YX=S
¼ b
Q ðHc Þm P V
(7)
MATLAB® software (MathWorks, Nattick, MA, USA) was used to estimate parameters values that produce the best fit between the experimental data obtained and the results of the model predictions. Fitting accuracy of the models was evaluated through analysis of coefficient of determination, R2. Differences between simulated (simul) and experimental (exp) results for butanol concentration for the n sample points, were evaluated by means of root mean square error (RMSE), calculated by Eq. (8) [33]. Goodness of the model fitting was determined in terms of lower values of the RMSE.
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi Pn 2 i¼1 ½ðexpÞi ðsimulÞi RMSE ¼ n
Batch fermentation studies of C. acetobutylicum DSM 792 were conducted using a glucose-based medium. Eqs. (1)e(3) fitted well to the experimental data. Table 2 shows the model parameters and the coefficients of determination. The parameter determination performed in this work allowed the proposed model to describe butanol production appropriately, granted by the coefficient of determination obtained (R2P ¼ 0.97) (Fig. 2), although it did not consider the production of acetone and ethanol. According to our knowledge, no kinetic parameters for Clostridium acetobutylicum have been reported in the literature. The maximum specific growth rate (mm) value determined by the model was similar to the experimental data obtained in this work (0.15e0.18 h1) and within the range of experimental values reported by Monot et al. [34,35] for Clostridium acetobutylicum in a glucose medium (0.08e0.35 h1). 3.2. Gas stripping model Gas stripping studies were conducted using a butanol solution and the industrial medium containing 6 g/L butanol. Table 3 shows the parameters determined using the value of Henry's constant for butanol partitioning in the air-water system of 1.4 104 weight of butanol per mL air/weight of butanol per mL water, value reported at 25 C by Buttery [36] and corrected at 37 C according to Sander [37]. ksa values for both the butanol solution and the industrial medium with and without cells at 0.4 vvm (Table 3), were similar to that obtained by Liao et al. [20]. However, they were lower than those found by de Vrije et al. [21] and Chen et al. [9] since their experiments were carried out at greater gas flow rates (1 vvm) using nitrogen as carrier gas (Table 4). Coefficient of determination obtained for the model tested showed accept able fit to the experimental data (R2 > 0.90). As expected, when the gas flow rate
(8)
Table 2 Kinetic model parameters. Parameter
Unit
Value
mm
h1 g/L g/g g/g g/L h1 e e e e
0.22 0.20 0.11 0.19 11 0.04 1.7 0.87 0.98 0.97
Ks YX/S YP/S Kp kd a R2X R2S R2P
3.1. Fermentation model
R2X, R2S, R2P are coefficient of determination for Eq. (1), Eq. (2) and Eq. (3), respectively.
Fig. 2. Glucose, biomass and butanol concentration profiles during a batch fermentation of the glucose-based medium. Experimental (symbols); simulated (lines).
Table 3 Model parameters for the gas stripping process. Parameter
Gas flowrate ksa b m R2
Unit
vvm h1 e e e
Butanol solution
Industrial juices
Without cells
Without cells
With cells
0.4 0.009 5.8 104 0.055 0.99
0.4 0.010 5.0 104 0.017 0.92
0.4 0.011 4.9 104 0.011 0.92
0.2 0.006 5.3 104 0.010 0.91
n et al. / Biomass and Bioenergy 98 (2017) 153e160 E. Rocho
157
Table 4 Comparison of parameters for gas stripping process. Gas flowrate (vvm)
Initial butanol concentration (g/L)
Cells
ksa (h1)
Reference
0.4 0.4 0.4 1 1 1 1 1 2.6 4.8 4.8
6 6 12 6 10 10 12 11.5 8 8 10
No Yes No Yes No No No Yes No No Yes
0.010 0.011 0.010 0.051a 0.042 0.053 0.055 0.051 0.023 0.059 0.027
This study This study [20] [9] [40] [21] [9] [21] [24] [24] [24]
Data reported at 36 or 37 C. a Estimated from the data reported by Chen et al., 2014.
decreased to 0.2 vvm the ksa value was lower. No difference was found in the ksa value for experiments with and without cells (Table 3). This was in agreement with the results found by Ezeji et al. [22] for a solution with a butanol concentration below 7.5 g/L. Higher values reported in the literature of gas flowrate than 1 vvm (Table 4) did not increase the ksa values. The control test performed in order to establish the butanol mass losses in the gas stripping system showed a recovery higher than 97%.
3.3. Batch and fed batch fermentation coupled with gas stripping A batch fermentation of the industrial juices with an initial sugar concentration of 76 g/L was performed. The fermentation profiles are shown in Fig. 3. The fermentation ceased after 100 h, when the butanol concentration was 10.5 g/L (17.6 g/L total ABE), showing that the culture was unable to utilize all sugars (78% sugar conversion) which could be attributed to the toxic effect of butanol (Table 5). OD in the fermentation broth gradually increased in
Fig. 3. Experimental biomass, pH, acids, sugars and solvents concentration profiles during the batch fermentation of industrial juices.
n et al. / Biomass and Bioenergy 98 (2017) 153e160 E. Rocho
158
Table 5 Comparison of fermentations of different media without and with in situ gas stripping. Parameter
Total initial sugars Fermentation time Sugar conversion Acetone Ethanol Butanol Total ABE Butanol yield Butanol productivity ABE productivity Acetic acid Butyric acid A:B:Ec Maximum biomass a b c
Unit
g/L h % g/L g/L g/L g/L g/g g/Lh g/Lh g/L g/L e g/L
Glucose-based medium
Industrial juices
Batch
Batch
Fed-batch with gas strippinga
49.2 ± 0.6 48 93 ± 2 5.3 ± 0.2 1.4 ± 0.1 9.4 ± 0.4 16.1 ± 0.7 0.20 ± 0.00 0.20 ± 0.01 0.34 ± 0.01 0.8 ± 0.1 0.8 ± 0.0 4:7:1 3.4
76 ± 1 102 78 ± 2 5.9 ± 0.6 1.2 ± 0.1 10.5 ± 1.0 17.6 ± 1.8 0.18 ± 0.01 0.10 ± 0.01 0.17 ± 0.02 1.0 ± 0.1 1.7 ± 0.3 5:9:1 2.8
59 (þ39b) ± 1 143 100 ± 2 10.0 ± 1.0 3.3 ± 0.4 18.6 ± 1.8 31.8 ± 3.2 0.16 ± 0.01 0.13 ± 0.01 0.22 ± 0.02 1.0 ± 0.1 0.8 ± 0.1 3:6:1 3.3
Values in g/L are calculated based on the bioreactor volume (2.5 L). Total sugar added during fed batch mode. Weight mass solvent relation.
~78 h, and then decreased at the end of the fermentation showing cell lysis under microscopic observation. In order to alleviate the toxic effect of butanol on cells, an integrated process of fermentation with gas stripping was performed. For this system in which butanol is both being produced and stripped, change in butanol concentration is given by Eq. (6). The stripping rate can be raised by increasing the gas flow rate (Eq. (5)), and therefore the butanol concentration in the bioreactor would decrease to very low values. This may conduce to a dilute butanol condensate without obtaining phase separation since more water would be removed. Thus, the gas flow rate has to be selected in order to have a butanol concentration lower than its cell toxic level and also to obtain a concentrated butanol condensate. Ezeji et al. [24] reported that a gas flowrate of 4.8 vvm was sufficient to
maintain butanol concentration below toxic levels. However, Xue et al. [28] reported that flow rates higher than 1.6 vvm resulted in lower butanol titers in the condensate. By solving Eq. (7) a gas flow rate of 0.3e0.6 vvm, equivalent to ksa of 0.010e0.017 h1, was obtained to maintain a butanol titer of 8 g/L in the broth, which to our knowledge was the lowest flow rate reported for similar solvent production rates (Table 4). For greater solvent production rates, higher values of ksa would be needed which might lead to higher flowrates. A fed-batch fermentation was performed. The sugar concentration during fermentation was kept below 60 g/L, and thus below sugar inhibitory concentration [38]. When the butanol concentration in the fermentation broth was ~8 g/L (48 h), gas stripping was started at gas flowrate fixed in 1 L/min to avoid inhibitory butanol
Fig. 4. Experimental biomass, pH, acids, sugars and solvents concentration profiles during the fed batch fermentation of industrial juices. Dashed lines indicate the total production including products collected in the gas stripping condensate and those remained in the fermentation broth. The arrow indicates when the gas stripping started.
n et al. / Biomass and Bioenergy 98 (2017) 153e160 E. Rocho
159
could reduce energy consumption in the final product recovery. In a fed-batch fermentation coupled with in situ gas stripping, total sugar conversion and 18.6 g/L accumulated butanol distributed 42% in the fermentation broth and 58% in the condensate, were obtained. Acknowledgements The financial support was provided by CSIC-ANCAP (Uruguay). The authors thank Alur SA for providing industrial sugary material n e Innovacio n (ANII) and to the Agencia Nacional de Investigacio n for the posgraduate scholarship of Eloísa Rocho (POS_NAC_2013_1_11825). Fig. 5. Total sugars and butanol concentration profiles during a fed batch fermentation of industrial juices with in situ gas stripping. Experimental (symbols); simulated (lines).
concentration. Sugar and products profiles are shown in Fig. 4 and the fermentation parameters in Table 5. A butanol yield of 0.16 g/g was obtained in the integrated process, which was similar to the butanol yield without gas stripping. This phenomenon disagrees with several researchers which showed increase of butanol yield in the integrated process for other raw materials [8,23] but was in accordance with the results showed by Cai et al. [39] for sweet sorghum juice. It was observed a steady state for the butanol concentration in the fermentation broth since it remained almost unchanged. A sugar conversion of 100% and 31.8 g/L ABE (10.0 g/L acetone, 18.6 g/L butanol, 3.3 g/L ethanol) were produced with 244 g sugar consumed in 143 h. When fresh concentrated industrial juice was added into the bioreactor (at 48 h) a reduction in the sugar consumption rate was observed. Several reasons may provide the cause for these, including lack of nutrients as they were not added with the concentrated juices addition. The butanol concentration in the condensate was high enough for effective phase separation, generating an organic phase containing 444.8 g/L butanol, 42.6 g/L acetone, 12.5 g/L ethanol and 0.2 g/L butyric acid and a lower aqueous phase contained 79.1 g/L butanol, 37.5 g/L acetone, 9.6 g/L ethanol and 0.2 g/L butyric acid. Acetic acid was not detected in either both phases. The data obtained showed that the gas flowrate used (0.4 vvm which was equivalent to a ksa of 0.011 h1) was adequate to remove the butanol produced in the fermentation and to obtain a butanol titer lower than the inhibitory one. Although the kinetic did not take into account the cofermentation of different sugars (glucose, fructose, sucrose), the predicted values by Eq. (6) with S expressed as glucose equivalents, showed a satisfactory agreement with the experimental data (Fig. 5). The RMSE values calculated by Eq. (8) for biomass, sugars and butanol curves 0.76, 4.7 and 0.79, respectively, were lower than or close to 4, value which indicates that the model prediction is in good agreement with the experimental data according to Ranjan et al. [33]. 4. Conclusions Empirical models for the butanol production by C. acetobutylicum DSM 792 fermentation and in situ gas stripping product extraction showed satisfactory agreement with the experimental data in terms of cell growth, sugar consumption, and butanol production and extraction. A gas recycle flowrate in the range 0.3e0.6 vvm allowed to maintain butanol concentration below the inhibitory concentration (8 g/L) and to obtain a concentrated butanol condensate after phase separation which
References €kkila €, H. Ojamo, E. Muurinen, R.L. Keiski, Challenges in bio[1] V. García, J. Pa butanol production: how to improve the efficiency? Renew. Sustain. Energy Rev. 15 (2) (2011) 964e980. [2] P. Durre, Food vs fuel, Biotechnol. J. 2 (2007) 1525e1534. [3] M. Kumar, K. Gayen, Developments in biobutanol production: new insights, Appl. Energy 88 (6) (2011) 1999e2012. [4] S.Y. Lee, J.H. Park, S.H. Jang, L.K. Nielsen, J. Kim, K.S. Jung, Fermentative butanol production by Clostridia, Biotechnol. Bioeng. 101 (1) (2008) 209e228. [5] E.M. Green, Fermentative production of butanol - the industrial perspective, Curr. Opin. Biotechnol. 22 (3) (2011) 337e343. [6] M.O.S. Dias, M.P. Cunha, C.D.F. Jesus, G.J.M. Rocha, J.G. Pradella, C.E.V. Rossell, R. Maciel Filho, A. Bonomi, Second generation ethanol in Brazil: can it compete with electricity production? Bioresour. Technol. 102 (19) (2011) 8964e8971. n, M.D. Ferrari, C. Lareo, Energy evaluation of fuel bio[7] V. Larnaudie, E. Rocho ethanol production from sweet sorghum using very high gravity (VHG) conditions, Renew. Energy 88 (2016) 280e287. [8] C. Xue, J. Zhao, F. Liu, C. Lu, S. Yang, F. Bai, Two-stage in situ gas stripping for enhanced butanol fermentation and energy-saving product recovery, Bioresour. Technol. 135 (2013) 396e402. [9] Y. Chen, H. Ren, D. Liu, T. Zhao, X. Shi, H. Cheng, N. Zhao, Z. Li, B. Li, H. Niu, W. Zhuang, J. Xie, X. Chen, J. Wu, H. Ying, Enhancement of n-butanol production by in situ butanol removal using permeating-heating-gas stripping in acetone-butanol-ethanol fermentation, Bioresour. Technol. 164 (2014) 276e284. [10] T.C. Ezeji, N. Qureshi, H.P. Blaschek, Continuous butanol fermentation and feed starch retrogradation: butanol fermentation sustainability using Clostridium beijerinckii BA101, J. Biotechnol. 115 (2) (2005) 179e187. [11] T.C. Ezeji, N. Qureshi, H.P. Blaschek, Microbial production of a biofuel (acetone-butanol-ethanol) in a continuous bioreactor: impact of bleed and simultaneous product removal, Bioprocess Biosyst. Eng. 36 (1) (2013) 109e116. [12] A.P. Mariano, R. Maciel Filho, T.C. Ezeji, Energy requirements during butanol production and in situ recovery by cyclic vacuum, Renew. Energy 47 (2012) 183e187. [13] H.-J. Huang, S. Ramaswamy, Y. Liu, Separation and purification of biobutanol during bioconversion of biomass, Sep. Purif. Technol. 132 (2014) 516e540. pez-Contreras, W. Kuit, N.A.J. Siemerink, S.W.M. Kengen, J. Springer, [14] A.M. Lo P.A.M. Claasen, Production of longer-chain alcohols from lignocellulosic biomass: butanol, isopropanol and 2,3-butanediol, in: K. Waldron (Ed.), Bioalcohol Production, Woodhead Publishing, Cambridge, UK, 2010, pp. 415e460. [15] D. Cai, H. Chen, C. Chen, S. Hu, Y. Wang, Z. Chang, Q. Miao, P. Qin, Z. Wang, J. Wang, T. Tan, Gas stripping-pervaporation hybrid process process for energy-saving produt recovery from acetone-butanol-ethanol (ABE) fermentation broth, Chem. Eng. J. 287 (2016) 1e10. [16] A. Kujawska, J. Kujawski, M. Bryjak, W. Kujawski, ABE fermentation products recovery methods - a review, Renew. Sustain. Energy Rev. 48 (2015) 648e661. [17] A.P. Mariano, C.B.B. Costa, D.D.F. de Angelis, F. Maugeri Filho, D.I.P. Atala, M.R. Wolf Maciel, R. Maciel Filho, Optimisation of a continuous flash fermentation for butanol production using the response surface methodology, Chem. Eng. Res. Des. 88 (5e6) (2010) 562e571. [18] C. Xue, F. Liu, M. Xu, I.-C. Tang, J. Zhao, F. Bai, S.-T. Yang, Butanol production in acetone-butanol-ethanol fermentation with in situ product recovery by adsorption, Bioresour. Technol. 219 (2016) 158e168. [19] C. Xue, F. Liu, M. Xu, J. Zhao, L. Chen, J. Ren, F. Bai, S.-T. Yang, A novel in situ gas stripping-pervaporation proces integrated with acetone-butanol-ethanol in acetone-butanol-ethanol fermentation for hyper n-butanol production, Biotechnol. Bioeng. 113 (1) (2016b) 120e129. [20] Y.-C. Liao, K.-M. Lu, S.-Y. Li, Process parameters for operating 1-butanol gas stripping in a fermentor, J. Biosci. Bioeng. 118 (5) (2014) 558e564. pez-Contreras, [21] T. de Vrije, M. Budde, H. van der Wal, P.A.M. Claassen, A.M. Lo “In situ” removal of isopropanol, butanol and ethanol from fermentation broth by gas stripping, Bioresour. Technol. 137 (2013) 153e159. [22] T.C. Ezeji, N. Qureshi, H.P. Blaschek, Production of acetone, butanol and
160
[23]
[24]
[25] [26]
[27]
[28]
[29] [30]
[31]
n et al. / Biomass and Bioenergy 98 (2017) 153e160 E. Rocho ethanol by Clostridium beijerinckii BA101 and in situ recovery by gas stripping, World J. Microbiol. Biotechnol. 19 (2003) 595e603. T.C. Ezeji, N. Qureshi, H.P. Blaschek, Acetone butanol ethanol (ABE) production from concentrated substrate: reduction in substrate inhibition by fed-batch technique and product inhibition by gas stripping, Appl. Microbiol. Biotechnol. 63 (6) (2004) 653e658. T.C. Ezeji, P.M. Karcher, N. Qureshi, H.P. Blaschek, Improving performance of a gas stripping-based recovery system to remove butanol from Clostridium beijerinckii fermentation, Bioprocess Biosyst. Eng. 27 (3) (2005) 207e214. T.C. Ezeji, N. Qureshi, H.P. Blaschek, Bioproduction of butanol from biomass: from genes to bioreactors, Curr. Op. Biotechnol. 18 (3) (2007) 220e227. N. Qureshi, B.C. Saha, M.A. Cotta, Butanol production from wheat straw by simultaneous saccharification and fermentation using Clostridium beijerinckii: Part IIdfed-batch fermentation, Biomass Bioenergy 32 (2) (2008) 176e183. N. Qureshi, B.C. Saha, R.E. Hector, M.A. Cotta, Removal of fermentation inhibitors from alkaline peroxide pretreated and enzymatically hydrolyzed wheat straw: production of butanol from hydrolysate using Clostridium beijerinckii in batch reactors, Biomass Bioenergy 32 (12) (2008) 1353e1358. C. Xue, G. Du, J. Sun, L. Chen, S. Gao, M. Yu, S.-T. Yang, F. Bai, Characterization of gas stripping and its integration with acetone e butanol e ethanol fermentation for high-efficient butanol production and recovery, Biochem. Eng. J. 83 (2014) 55e61. O. Levenspiel, The Monod equation: a Revisit and a Generalization to product inhibition situations, Biotechnol. Bioeng. 22 (1980) 1671e1687. W.-L. Chang, Acetone-butanol-ethanol Fermentation by Engineered Clostridium Beijerinckii and Clostridium Tyrobutyricum, The Ohio State University, 2010. Thesis. zquez-Sa nchez, M.C. Montes-Horcasitas, R.A. Lo pez, Development of a H.I. Vela
[32] [33]
[34]
[35]
[36] [37] [38]
[39]
[40]
phenomenological kinetic model for butanol production using Clostridium beijerinckii, Rev. Mex. Ing. Quím. 13 (3) (2014) 103e112. K. Truong, J. Blackburn, The stripping of organic chemicals in biological treatment process, Environ. Prog. 3 (3) (1984) 143e152. A. Ranjan, R. Mayank, V.S. Moholkar, Process optimization for butanol production from developed rice straw hydrolysate using Clostridium acetobutylicum MTCC 481 strain, Biomass Conv. Bioref 3 (2) (2012) 143e155. F. Monot, J.-M. Engasser, H. Petitdemange, Influence of pH and undissociated butyric acid on the production of acetone and butanol in batch cultures of Clostridium acetobutylicum, Appl. Microbiol. Biotechnol. 19 (1984) 422e426. F. Monot, J.-R. Martin, H. Petitdemange, R. Gay, Acetone and butanol production by Clostridium acetobutylicum in a synthetic medium, Appl. Environ. Microbiol. 44 (1982) 1318e1324. R. Buttery, L. Ling, D. Guadagni, Volatilities of aldehydes, ketones, and esters in dilute water solution, Food Volatiles 17 (2) (1969) 385e389. R. Sander, Compilation of Henry's law constants (version 4.0) for water as solvent, Atmos. Chem. Phys. 15 (2015) 4399e4981. K.W. Staggs, D.R. Nielsen, Improving n-butanol production in batch and semicontinuous processes through integrated product recovery, Process Biochem. 50 (10) (2015) 1487e1498. D. Cai, Y. Wang, C. Chen, P. Qin, Q. Miao, C. Zhang, P. Li, T. Tan, Acetonebutanol-ethanol from sweet sorghum juice by an immobilized fermentationgas stripping integration process, Bioresour. Technol. 211 (2016) 704e710. pez-Contreras, K.A. Pyrgakis, T. de Vrije, M.A.W. Budde, K. Kyriakou, A.M. Lo A.C. Kokossis, A process integration approach for the production of biological iso-propanol, butanol and ethanol using gas stripping and adsorption as recovery methods, Biochem. Eng. J. 116 (2016) 176e194.
Contents lists available at ScienceDirect
Biomass and Bioenergy journal homepage: http://www.elsevier.com/locate/biombioe
Research paper
Integrated ABE fermentation-gas stripping process for enhanced butanol production from sugarcane-sweet sorghum juices n, Mario Daniel Ferrari, Claudia Lareo* Eloísa Rocho Depto. Bioingeniería, Facultad de Ingeniería, Universidad de la República, J. Herrera y Reissig 565, CP 11300, Montevideo, Uruguay
a r t i c l e i n f o
a b s t r a c t
Article history: Received 18 August 2016 Received in revised form 26 November 2016 Accepted 4 January 2017
The biobutanol production by ABE (acetone-butanol-ethanol) fermentation with an integrated product recovery system by gas stripping was studied. A mixture of industrial sugarcane-sweet sorghum juices was fermented using Clostridium acetobutylicum DSM 792. Empirical models were developed to determine an operational condition for the integrated process that contributes both to mitigate the butanol inhibitory effect and to obtain a highly concentrated butanol condensate. A Monod model supplemented with a term describing product inhibition was employed to describe cell growth, butanol formation, and substrate consumption, respectively. Gas stripping was described by a first order model. The models showed satisfactory agreement with the experimental data in terms of cell growth, sugar consumption, and butanol production and extraction. A gas recycle flowrate in the range 0.3e0.6 vvm allowed to maintain butanol concentration below the inhibitory concentration (8 g/L) and to obtain a concentrated butanol condensate after phase separation, which could reduce energy consumption in the final product recovery. In a fed-batch fermentation coupled with in situ gas stripping, total sugar conversion and 18.6 g/L butanol distributed 42% in the fermentation broth and 58% in the condensate, were obtained. © 2017 Elsevier Ltd. All rights reserved.
Keywords: Biobutanol ABE fermentation Clostridium Gas stripping Sugarcane Sweet sorghum Kinetic model
1. Introduction During the last decades, there has been an increasing interest for the production of chemicals and fuels from renewable resources. Reasons for this include climate change, global warming, and energy security [1]. Butanol is considered an attractive biofuel as it has clearly superior properties to ethanol due to its higher energy density, less volatile and explosive and less hygroscopic [1e4], in addition to being an important chemical precursor for paints, polymers and plastics, and widely used solvent [5]. Biobutanol can be produced by the acetone-butanol-ethanol (ABE) fermentation by various Clostridium spp., in which a solvent mixture is produced, generally in the ratio 3:6:1 acetone-butanol-ethanol respectively. The economic viability of fuel biobutanol production largely depends on the costs of raw materials and energy consumed. Energy consumption can be high due to the low butanol titer and purity reached in the fermentation broth [1,3]. The raw material utilized in biobutanol production is one of the major cost factors in its production [3e5]. Both sugarcane and
* Corresponding author. E-mail address: clareo@fing.edu.uy (C. Lareo). http://dx.doi.org/10.1016/j.biombioe.2017.01.011 0961-9534/© 2017 Elsevier Ltd. All rights reserved.
sweet sorghum can offer more advantages than other crops since they produce a residue (bagasse) which can be burnt for steam production to satisfy the energy demand of the industrial processes [6,7]. Therefore, raw materials with high carbohydrate content, efficient transformation processes energetically optimized, and an accessible, low cost energy source are needed. Fuel bioethanol is currently produced in the north of Uruguay mainly from sugarcane. Sweet sorghum juice alone or mixed with sugarcane juice is also used to extend the plant facility working time. Since both biobutanol and bioethanol could be produced from the same feedstock, its production could take place in the same n, http://www. facility using similar equipment (Alur SA, Bella Unio alur.com.uy/agroindustrias/bella-union/, access on 11/21/2016). One of the major challenges in biobutanol production is the intensive energy consumption in product recovery caused by the low product concentrations reached in the fermentation broth [8]. This is attributed mainly to cell toxicity or fermentation inhibition by butanol [9e12]. Furthermore, the butanol yield and productivity are also low, due to the co-production of acetone and ethanol. In order to solve this problem, extensive research and development efforts have been made. Some approaches that have been proposed include improve butanol tolerance of strains by mutagenesis and metabolic engineering [1,5,13,14], to couple butanol production
154
n et al. / Biomass and Bioenergy 98 (2017) 153e160 E. Rocho
with an in situ extraction process. Different techniques for in situ butanol recovery process (adsorption, liquid-liquid extraction, pervaporation, vacuum flash fermentation and gas stripping) have been evaluated as promising techniques to improve productivity by mitigating butanol inhibition [15e19]. Even these techniques are more economically competitive comparing to conventional distillation for butanol recovery from the dilute fermentation broth, the deficiency in recovering a high titer of butanol is still the challenge. Gas stripping is a physical separation process where target compounds are removed from an aqueous solution by bubbling gas through it [20]. It has many advantages over the other methods: simple operation, no harm to culture, and does not requires any chemicals nor membranes. It also collaborates with the agitation and homogenization of the system [10,13,21e27]. However, its potential in butanol recovery and energy saving for biobutanol production has not yet been fully explored nor optimized [13,28]. Most of previous studies on integrated ABE fermentation with in situ gas stripping have been conducted at relatively low butanol concentrations (5 g/L or less) to minimize butanol toxicity [9]. Several works reported that the gas flowrate [24,28] and the butanol titer in the feed solution affect gas stripping process [20,28]. Xue et al. [28] reported that it is necessary to conduct gas stripping at a butanol titer higher than 8 g/L in the feed solution in order to obtain a condensate with a butanol titer higher than its solubility in water (~80 g/L at 20 C), which will result in phase separation and in a more energy-efficient butanol recovery process. On the other hand, the recycling of large amount of gas in industrial reactors would be uneconomical. The effect of gas recycle rate should be evaluated in order to optimize gas stripping conditions as an interest in commercialization this technology. The aim of this work was to evaluate butanol production from a mixture of sugarcane-sweet sorghum juices using C. acetobutylicum DSM 792 in an integrated process (fermentation with in situ gas stripping) that contributes both to mitigate the butanol inhibitory effect and to obtain a highly concentrated butanol condensate. Kinetics models describing butanol formation by fermentation and butanol extraction by gas stripping were developed to study the effect of gas recycle rate on the performance of the fermentation coupled with in situ gas stripping and to determine the most favorable operational condition. 2. Materials and methods 2.1. Experimental assays 2.1.1. Characterization of raw material A mixture of concentrated industrial juices of sugar cane (75%) n, and sweet sorghum (25%) was provided by Alur SA (Bella Unio Uruguay) and it was stored at 5 C. Table 1 shows its composition. 2.1.2. Microorganism, inoculum preparation and fermentation media Clostridium acetobutylicum DSM 792 was used in all fermentations. A stock culture was maintained in Reinforced Clostridial Medium (RCM) at 4 C. A pre-culture was prepared by inoculating 10 mL of the stock culture into a serum bottle containing 100 mL RCM medium and incubating at 37 C for 24e48 h until active growth was observed. Glucose-based and the industrial media were used, containing 50e60 g/L of total sugar concentration expressed as glucose equivalent and 1% (v/v) P2 solutions. The P2 solutions contains K2HPO4 50 g/L, KH2PO4 50 g/L, ammonium acetate 220 g/L, p-amino benzoic acid 0.1 g/L, thiamine 0.1 g/L, biotin 0.001 g/L, MgSO4.7H2O 20 g/L, MnSO4.H2O 1 g/L, FeSO4.7H2O 1 g/L, NaCl 1 g/L. The seed culture for the fermentation was prepared in 250-mL
Table 1 Composition of concentrated sugarcane-sweet sorghum juices. Component
Unit
Value
Sucrose Glucose Fructose Ethanol Acetic acid Butyric acid Succinic acid cis-Aconitic acid trans-Aconitic acid Furfural 5-Hidroximetilfurfural Phosphorous Nitrogen (Kjeldahl) Aluminum Calcium Cooper Iron Magnesium Manganese Potassium Sodium Zinc Ash Moisture Density (20 C)
% % % % % % % % % % % % % % % % % % % % % % % % kg/m3
75 5 3 0.69 0.11 0.34 0.15 0.13 1.05 nd nd <0.5 0.31 <0.002 0.18 <0.0005 0.0047 0.11 0.0012 0.76 0.04 <0.0005 2.5 32 1330
% percentage in dry weight except moisture. nd: no detected (detection limit: 0.003%).
bottles containing 100 mL of the same medium used in the fermentation but adjusted to reach half of the total sugar concentration. The pH was adjusted to 6.0 ± 0.1. The medium supplemented with 1 g/L of yeast extract was swept with O2-free N2 over the headspace of the bottles. It was sterilized at 121 C during 15 min. On cooling to 37 C, 1% (v/v) of filter-sterilized P2 stock solutions were added followed by inoculation with 10% (v/v) highly active cells that were grown in RCM. The culture was incubated at 37 C, 150 rpm for 20e24 h until the cell density reached an optical density (OD) value of ~3. 2.1.3. Batch fermentation Fermentation tests with glucose-based medium were performed in bottles of 250 mL with 100 mL of medium containing 50e60 g/L of glucose. The pH was adjusted initially to 6.0 ± 0.1. The medium was swept with O2-free N2 over the headspace of the bottles. It was sterilized at 121 C during 15 min. On cooling to 37 C, 1% (v/v) of filter-sterilized P2 stock solutions were added, followed by inoculation with 9% (v/v) highly motile cells. The bottles were incubated in an orbital shaker at 150 rpm and 37 C. Samples were withdrawn at intervals for analysis. Tests were conducted in duplicate. A batch fermentation was also performed in a 5 L-bioreactor (Infors HT, Switzerland) containing 2 L of the industrial juices diluted to reach 75 g/L of total sugar concentration. The pH was adjusted to 6.0 ± 0.1. The medium was swept with O2-free N2 over the headspace of the bioreactor. It was sterilized at 121 C during 15 min. On cooling to 37 C, 1% (v/v) of filter-sterilized P2 stock solutions were added followed by inoculation with 6% (v/v) highly active cells. Fermentation was performed at 37 C and 150 rpm. No antifoam was necessary during the fermentation. 2.1.4. Fed batch fermentation coupled with in-situ gas stripping A schematic diagram of the integrated reactor set up is shown in Fig. 1. Prior to the gas stripping experiment, the condenser and gascirculation line were flushed with O2-free N2 gas. A 5-L bioreactor (Infors HT, Switzerland) containing 2.5 L of the industrial medium
n et al. / Biomass and Bioenergy 98 (2017) 153e160 E. Rocho
Fig. 1. Schematic diagram of the integrated reactor set up.
(total sugar concentration 60 g/L) was inoculated with 6% (v/v) highly active cells of C. acetobutylicum DSM 792. The pH was adjusted initially to 6.0 ± 0.1. The temperature was controlled at 37 C and the agitation at 150 rpm. The fermentation was allowed to proceed in batch mode for 48 h (until the butanol concentration was approximately 8 g/L), after which gas stripping was applied at a flowrate of 1 L/min by recycling the off gasses (CO2 /H2) through the system using a twin-head Masterflex® peristaltic pump and 18 size Tygon pump tubing (Cole-Parmer). The ABE vapors were cooled in a condenser to 0 C, using a solution of glycerol-water (30% (v/v)) cooled in a refrigerated circulating bath (Polystat, Cole-Parmer). The condensate was collected in a flask immersed in the refrigerated circulating bath, which was pumped into a 250-mL flask which was used as a storage tank for phase separation. Samples were aseptically withdrawn at intervals for brix grades, sugars, products and optical density analysis. When residual sugars decreased to ~10 g/L (about 48 h of fermentation), concentrated juice solution (600 g/L total sugars) was added to reach a sugar concentration close to 60 g/L. No antifoam was necessary during the fermentation. 2.1.5. Gas stripping assays Gas stripping experiments were done in a 5-L bioreactor (Infors HT, Switzerland) (Fig. 1) with 2.5 L medium. Experiments with and without cells were performed. For the experiments without cells, two solutions were prepared by adding butanol to distilled water and to the industrial juices to reach a concentration of 6 g/L. Gas stripping was performed by recycling the headspace gasses through the solutions at a flow rate of 1 L/min (0.4 vvm). For the assays with cells, the fermented medium containing 1.6 g/L of cells and 6 g/L of butanol was utilized. Gas stripping was performed by recycling the headspace gasses through the solutions at a flow rate of 1 L/min (0.4 vvm) or 0.5 L/min (0.2 vvm). The condenser was maintained at 0 C and the bioreactor temperature and agitation were 37 C and 150 rpm, respectively. Samples were aseptically withdrawn at intervals (~1 h) for butanol analysis. A control test of the gas stripping system was also performed by recycling the headspace gasses in order to determine the butanol mass recovery. The model solution was prepared with the industrial juices and the addition of butanol at the same concentration as in the fermentation broth (6 g/L) and at a gas flow rate of 0.4 vvm. The condenser was maintained at 0 C. The bioreactor was operated at 37 C and 150 rpm. Samples were taken from the bioreactor and condensate storage flask periodically (after ~15e20 h) and then were analyzed to calculate the butanol mass recovery. 2.1.6. Analytical methods The fermentation evolution was monitored by brix grades using
155
a refractometer (Master, Atago) as an estimation of sugars content and by optical density (OD) in a spectrophotometer (Unico, UV2100) at 600 nm as a measured of cell density. Microscopic observation was periodically done to observe the morphology and motility of the cells. Dry cell weight (DCW) was determined by measuring OD at 600 nm: one unit of OD corresponded to 0.37 ± 0.01 g/L of DCW. After removing the bacterial cells by centrifugation at 6500 rpm for 10 min, the clear fermentation broth was analyzed for sugar and product concentrations. Sucrose, glucose and fructose concentrations were determined by HPLC (Shimadzu, Kyoto, Japan) using a Shodex SUGAR KS-801 column at 55 C, ultrapure water as eluent at a flow rate of 0.7 mL/min and a refractive index detector (RID-10A). Total sugars were expressed in glucose equivalents (glucose þ fructose þ 1.05 x sucrose). Acetone, ethanol and organic acids (acetic, butyric and succinic), were measured with a gas chromatograph (GC, Shimadzu GC-2010) equipped with a flame ionization detector and a fused silica column (RTX®-Wax, 30 m long, 0.5 mm film thickness and 0.32 mm ID, Restek). Hydroxymethylfurfural and furfural were determined by HPLC using a Phenomenex ROA-Organic Acid Hþ column and an UV detector at 210 nm cis-Aconitic acid and transaconitic acid were also determined by HPLC, but using a Shodex RSpack KC-811 column. Sodium, potassium, calcium, magnesium, manganese, aluminium, copper and zinc were determined by atomic emission spectrometry, iron by atomic absorption spectrometry, proteins by Kjeldahl, and phosphorus by the spectrophotometric method using the molybdenum blue reaction. The moisture and ash contents were determined at 60 C and 550 C, respectively. 2.2. Process models The Monod kinetic model is the most widely used empirical model that describes microbial growth under substrate-limiting conditions. The butanol accumulated in the culture broth tends to restrain the cell growth, and therefore, the Monod model was modified by introducing the term of product inhibition [29]. The model was also modified by introducing another term which includes a cell death constant parameter, which has been justified by several authors as a drop in biomass concentration once the solventogenic phase had started [30,31]. Therefore, the equations below were developed as follows:
dX dt
¼
Ks þ S
dS m X ¼ dt YX=S dP dt
¼
mm S
¼
m X YP=S YX=S
1
X
mm S
P Kp
kd X
Ks þ S ¼
a
1
mm S Ks þ S
P Kp
1
a
P Kp
(1)
X YX=S a X
(2)
YP=S YX=S
(3)
where X is the DCW (g/L), m is the specific growth rate (h1), mm is the maximum specific growth rate (h1), S is the growth-limiting substrate concentration (g/L), Ks is the substrate saturation constant (g/L), kd is the specific cell death rate (h1), P is the butanol concentration (g/L), Kp is the product concentration at which no cell growth occurs (g/L), a is the degree of product inhibition, YX/S (g/g) is the biomass yield coefficient and YP/S (g/g) is the butanol yield coefficient. Gas stripping of butanol and other organic solvents from an aqueous solution can be described, macroscopically, by a first order process reported by Truong and Blackburn [32] as shown by Eq. (4), which has been verified in many studies [20,21,24,32].
n et al. / Biomass and Bioenergy 98 (2017) 153e160 E. Rocho
156
3. Results and discussion
dP ¼ dt
rs
¼ ks a P
(4)
where rs represents the stripping rate of the butanol (g/Lh), P and ksa are butanol concentrations in the aqueous solution (g/L) and stripping rate coefficient (h1), respectively. The stripping rate coefficient has been modeled by the equation shown in Eq. (5) in which this parameter is linearly proportional to the gas flow rate [20,24,32].
Q ðHc Þm V
ks a ¼ b
(5)
where b and m are power function constants, V is the total liquid volume of the fermentation broth (L), Q is the gas flow rate (L/min) and Hc is the Henry's law constant (weight of butanol per mL air/ weight of butanol per mL water). The integrated process of fermentation coupled with in situ gas stripping can be modeled using Eq. (6).
dP dt
¼ ¼
m X YP=S YX=S
mm S Ks þ S
ks a P 1
P Kp
a X
YP=S ks a P YX=S
(6)
At steady state (dP/dt ¼ 0), and if ksa can be modeled as Eq. (5), the following relationship is obtained:
mm S
Ks þ S
1
P Kp
a X
YP=S YX=S
¼ b
Q ðHc Þm P V
(7)
MATLAB® software (MathWorks, Nattick, MA, USA) was used to estimate parameters values that produce the best fit between the experimental data obtained and the results of the model predictions. Fitting accuracy of the models was evaluated through analysis of coefficient of determination, R2. Differences between simulated (simul) and experimental (exp) results for butanol concentration for the n sample points, were evaluated by means of root mean square error (RMSE), calculated by Eq. (8) [33]. Goodness of the model fitting was determined in terms of lower values of the RMSE.
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi Pn 2 i¼1 ½ðexpÞi ðsimulÞi RMSE ¼ n
Batch fermentation studies of C. acetobutylicum DSM 792 were conducted using a glucose-based medium. Eqs. (1)e(3) fitted well to the experimental data. Table 2 shows the model parameters and the coefficients of determination. The parameter determination performed in this work allowed the proposed model to describe butanol production appropriately, granted by the coefficient of determination obtained (R2P ¼ 0.97) (Fig. 2), although it did not consider the production of acetone and ethanol. According to our knowledge, no kinetic parameters for Clostridium acetobutylicum have been reported in the literature. The maximum specific growth rate (mm) value determined by the model was similar to the experimental data obtained in this work (0.15e0.18 h1) and within the range of experimental values reported by Monot et al. [34,35] for Clostridium acetobutylicum in a glucose medium (0.08e0.35 h1). 3.2. Gas stripping model Gas stripping studies were conducted using a butanol solution and the industrial medium containing 6 g/L butanol. Table 3 shows the parameters determined using the value of Henry's constant for butanol partitioning in the air-water system of 1.4 104 weight of butanol per mL air/weight of butanol per mL water, value reported at 25 C by Buttery [36] and corrected at 37 C according to Sander [37]. ksa values for both the butanol solution and the industrial medium with and without cells at 0.4 vvm (Table 3), were similar to that obtained by Liao et al. [20]. However, they were lower than those found by de Vrije et al. [21] and Chen et al. [9] since their experiments were carried out at greater gas flow rates (1 vvm) using nitrogen as carrier gas (Table 4). Coefficient of determination obtained for the model tested showed accept able fit to the experimental data (R2 > 0.90). As expected, when the gas flow rate
(8)
Table 2 Kinetic model parameters. Parameter
Unit
Value
mm
h1 g/L g/g g/g g/L h1 e e e e
0.22 0.20 0.11 0.19 11 0.04 1.7 0.87 0.98 0.97
Ks YX/S YP/S Kp kd a R2X R2S R2P
3.1. Fermentation model
R2X, R2S, R2P are coefficient of determination for Eq. (1), Eq. (2) and Eq. (3), respectively.
Fig. 2. Glucose, biomass and butanol concentration profiles during a batch fermentation of the glucose-based medium. Experimental (symbols); simulated (lines).
Table 3 Model parameters for the gas stripping process. Parameter
Gas flowrate ksa b m R2
Unit
vvm h1 e e e
Butanol solution
Industrial juices
Without cells
Without cells
With cells
0.4 0.009 5.8 104 0.055 0.99
0.4 0.010 5.0 104 0.017 0.92
0.4 0.011 4.9 104 0.011 0.92
0.2 0.006 5.3 104 0.010 0.91
n et al. / Biomass and Bioenergy 98 (2017) 153e160 E. Rocho
157
Table 4 Comparison of parameters for gas stripping process. Gas flowrate (vvm)
Initial butanol concentration (g/L)
Cells
ksa (h1)
Reference
0.4 0.4 0.4 1 1 1 1 1 2.6 4.8 4.8
6 6 12 6 10 10 12 11.5 8 8 10
No Yes No Yes No No No Yes No No Yes
0.010 0.011 0.010 0.051a 0.042 0.053 0.055 0.051 0.023 0.059 0.027
This study This study [20] [9] [40] [21] [9] [21] [24] [24] [24]
Data reported at 36 or 37 C. a Estimated from the data reported by Chen et al., 2014.
decreased to 0.2 vvm the ksa value was lower. No difference was found in the ksa value for experiments with and without cells (Table 3). This was in agreement with the results found by Ezeji et al. [22] for a solution with a butanol concentration below 7.5 g/L. Higher values reported in the literature of gas flowrate than 1 vvm (Table 4) did not increase the ksa values. The control test performed in order to establish the butanol mass losses in the gas stripping system showed a recovery higher than 97%.
3.3. Batch and fed batch fermentation coupled with gas stripping A batch fermentation of the industrial juices with an initial sugar concentration of 76 g/L was performed. The fermentation profiles are shown in Fig. 3. The fermentation ceased after 100 h, when the butanol concentration was 10.5 g/L (17.6 g/L total ABE), showing that the culture was unable to utilize all sugars (78% sugar conversion) which could be attributed to the toxic effect of butanol (Table 5). OD in the fermentation broth gradually increased in
Fig. 3. Experimental biomass, pH, acids, sugars and solvents concentration profiles during the batch fermentation of industrial juices.
n et al. / Biomass and Bioenergy 98 (2017) 153e160 E. Rocho
158
Table 5 Comparison of fermentations of different media without and with in situ gas stripping. Parameter
Total initial sugars Fermentation time Sugar conversion Acetone Ethanol Butanol Total ABE Butanol yield Butanol productivity ABE productivity Acetic acid Butyric acid A:B:Ec Maximum biomass a b c
Unit
g/L h % g/L g/L g/L g/L g/g g/Lh g/Lh g/L g/L e g/L
Glucose-based medium
Industrial juices
Batch
Batch
Fed-batch with gas strippinga
49.2 ± 0.6 48 93 ± 2 5.3 ± 0.2 1.4 ± 0.1 9.4 ± 0.4 16.1 ± 0.7 0.20 ± 0.00 0.20 ± 0.01 0.34 ± 0.01 0.8 ± 0.1 0.8 ± 0.0 4:7:1 3.4
76 ± 1 102 78 ± 2 5.9 ± 0.6 1.2 ± 0.1 10.5 ± 1.0 17.6 ± 1.8 0.18 ± 0.01 0.10 ± 0.01 0.17 ± 0.02 1.0 ± 0.1 1.7 ± 0.3 5:9:1 2.8
59 (þ39b) ± 1 143 100 ± 2 10.0 ± 1.0 3.3 ± 0.4 18.6 ± 1.8 31.8 ± 3.2 0.16 ± 0.01 0.13 ± 0.01 0.22 ± 0.02 1.0 ± 0.1 0.8 ± 0.1 3:6:1 3.3
Values in g/L are calculated based on the bioreactor volume (2.5 L). Total sugar added during fed batch mode. Weight mass solvent relation.
~78 h, and then decreased at the end of the fermentation showing cell lysis under microscopic observation. In order to alleviate the toxic effect of butanol on cells, an integrated process of fermentation with gas stripping was performed. For this system in which butanol is both being produced and stripped, change in butanol concentration is given by Eq. (6). The stripping rate can be raised by increasing the gas flow rate (Eq. (5)), and therefore the butanol concentration in the bioreactor would decrease to very low values. This may conduce to a dilute butanol condensate without obtaining phase separation since more water would be removed. Thus, the gas flow rate has to be selected in order to have a butanol concentration lower than its cell toxic level and also to obtain a concentrated butanol condensate. Ezeji et al. [24] reported that a gas flowrate of 4.8 vvm was sufficient to
maintain butanol concentration below toxic levels. However, Xue et al. [28] reported that flow rates higher than 1.6 vvm resulted in lower butanol titers in the condensate. By solving Eq. (7) a gas flow rate of 0.3e0.6 vvm, equivalent to ksa of 0.010e0.017 h1, was obtained to maintain a butanol titer of 8 g/L in the broth, which to our knowledge was the lowest flow rate reported for similar solvent production rates (Table 4). For greater solvent production rates, higher values of ksa would be needed which might lead to higher flowrates. A fed-batch fermentation was performed. The sugar concentration during fermentation was kept below 60 g/L, and thus below sugar inhibitory concentration [38]. When the butanol concentration in the fermentation broth was ~8 g/L (48 h), gas stripping was started at gas flowrate fixed in 1 L/min to avoid inhibitory butanol
Fig. 4. Experimental biomass, pH, acids, sugars and solvents concentration profiles during the fed batch fermentation of industrial juices. Dashed lines indicate the total production including products collected in the gas stripping condensate and those remained in the fermentation broth. The arrow indicates when the gas stripping started.
n et al. / Biomass and Bioenergy 98 (2017) 153e160 E. Rocho
159
could reduce energy consumption in the final product recovery. In a fed-batch fermentation coupled with in situ gas stripping, total sugar conversion and 18.6 g/L accumulated butanol distributed 42% in the fermentation broth and 58% in the condensate, were obtained. Acknowledgements The financial support was provided by CSIC-ANCAP (Uruguay). The authors thank Alur SA for providing industrial sugary material n e Innovacio n (ANII) and to the Agencia Nacional de Investigacio n for the posgraduate scholarship of Eloísa Rocho (POS_NAC_2013_1_11825). Fig. 5. Total sugars and butanol concentration profiles during a fed batch fermentation of industrial juices with in situ gas stripping. Experimental (symbols); simulated (lines).
concentration. Sugar and products profiles are shown in Fig. 4 and the fermentation parameters in Table 5. A butanol yield of 0.16 g/g was obtained in the integrated process, which was similar to the butanol yield without gas stripping. This phenomenon disagrees with several researchers which showed increase of butanol yield in the integrated process for other raw materials [8,23] but was in accordance with the results showed by Cai et al. [39] for sweet sorghum juice. It was observed a steady state for the butanol concentration in the fermentation broth since it remained almost unchanged. A sugar conversion of 100% and 31.8 g/L ABE (10.0 g/L acetone, 18.6 g/L butanol, 3.3 g/L ethanol) were produced with 244 g sugar consumed in 143 h. When fresh concentrated industrial juice was added into the bioreactor (at 48 h) a reduction in the sugar consumption rate was observed. Several reasons may provide the cause for these, including lack of nutrients as they were not added with the concentrated juices addition. The butanol concentration in the condensate was high enough for effective phase separation, generating an organic phase containing 444.8 g/L butanol, 42.6 g/L acetone, 12.5 g/L ethanol and 0.2 g/L butyric acid and a lower aqueous phase contained 79.1 g/L butanol, 37.5 g/L acetone, 9.6 g/L ethanol and 0.2 g/L butyric acid. Acetic acid was not detected in either both phases. The data obtained showed that the gas flowrate used (0.4 vvm which was equivalent to a ksa of 0.011 h1) was adequate to remove the butanol produced in the fermentation and to obtain a butanol titer lower than the inhibitory one. Although the kinetic did not take into account the cofermentation of different sugars (glucose, fructose, sucrose), the predicted values by Eq. (6) with S expressed as glucose equivalents, showed a satisfactory agreement with the experimental data (Fig. 5). The RMSE values calculated by Eq. (8) for biomass, sugars and butanol curves 0.76, 4.7 and 0.79, respectively, were lower than or close to 4, value which indicates that the model prediction is in good agreement with the experimental data according to Ranjan et al. [33]. 4. Conclusions Empirical models for the butanol production by C. acetobutylicum DSM 792 fermentation and in situ gas stripping product extraction showed satisfactory agreement with the experimental data in terms of cell growth, sugar consumption, and butanol production and extraction. A gas recycle flowrate in the range 0.3e0.6 vvm allowed to maintain butanol concentration below the inhibitory concentration (8 g/L) and to obtain a concentrated butanol condensate after phase separation which
References €kkila €, H. Ojamo, E. Muurinen, R.L. Keiski, Challenges in bio[1] V. García, J. Pa butanol production: how to improve the efficiency? Renew. Sustain. Energy Rev. 15 (2) (2011) 964e980. [2] P. Durre, Food vs fuel, Biotechnol. J. 2 (2007) 1525e1534. [3] M. Kumar, K. Gayen, Developments in biobutanol production: new insights, Appl. Energy 88 (6) (2011) 1999e2012. [4] S.Y. Lee, J.H. Park, S.H. Jang, L.K. Nielsen, J. Kim, K.S. Jung, Fermentative butanol production by Clostridia, Biotechnol. Bioeng. 101 (1) (2008) 209e228. [5] E.M. Green, Fermentative production of butanol - the industrial perspective, Curr. Opin. Biotechnol. 22 (3) (2011) 337e343. [6] M.O.S. Dias, M.P. Cunha, C.D.F. Jesus, G.J.M. Rocha, J.G. Pradella, C.E.V. Rossell, R. Maciel Filho, A. Bonomi, Second generation ethanol in Brazil: can it compete with electricity production? Bioresour. Technol. 102 (19) (2011) 8964e8971. n, M.D. Ferrari, C. Lareo, Energy evaluation of fuel bio[7] V. Larnaudie, E. Rocho ethanol production from sweet sorghum using very high gravity (VHG) conditions, Renew. Energy 88 (2016) 280e287. [8] C. Xue, J. Zhao, F. Liu, C. Lu, S. Yang, F. Bai, Two-stage in situ gas stripping for enhanced butanol fermentation and energy-saving product recovery, Bioresour. Technol. 135 (2013) 396e402. [9] Y. Chen, H. Ren, D. Liu, T. Zhao, X. Shi, H. Cheng, N. Zhao, Z. Li, B. Li, H. Niu, W. Zhuang, J. Xie, X. Chen, J. Wu, H. Ying, Enhancement of n-butanol production by in situ butanol removal using permeating-heating-gas stripping in acetone-butanol-ethanol fermentation, Bioresour. Technol. 164 (2014) 276e284. [10] T.C. Ezeji, N. Qureshi, H.P. Blaschek, Continuous butanol fermentation and feed starch retrogradation: butanol fermentation sustainability using Clostridium beijerinckii BA101, J. Biotechnol. 115 (2) (2005) 179e187. [11] T.C. Ezeji, N. Qureshi, H.P. Blaschek, Microbial production of a biofuel (acetone-butanol-ethanol) in a continuous bioreactor: impact of bleed and simultaneous product removal, Bioprocess Biosyst. Eng. 36 (1) (2013) 109e116. [12] A.P. Mariano, R. Maciel Filho, T.C. Ezeji, Energy requirements during butanol production and in situ recovery by cyclic vacuum, Renew. Energy 47 (2012) 183e187. [13] H.-J. Huang, S. Ramaswamy, Y. Liu, Separation and purification of biobutanol during bioconversion of biomass, Sep. Purif. Technol. 132 (2014) 516e540. pez-Contreras, W. Kuit, N.A.J. Siemerink, S.W.M. Kengen, J. Springer, [14] A.M. Lo P.A.M. Claasen, Production of longer-chain alcohols from lignocellulosic biomass: butanol, isopropanol and 2,3-butanediol, in: K. Waldron (Ed.), Bioalcohol Production, Woodhead Publishing, Cambridge, UK, 2010, pp. 415e460. [15] D. Cai, H. Chen, C. Chen, S. Hu, Y. Wang, Z. Chang, Q. Miao, P. Qin, Z. Wang, J. Wang, T. Tan, Gas stripping-pervaporation hybrid process process for energy-saving produt recovery from acetone-butanol-ethanol (ABE) fermentation broth, Chem. Eng. J. 287 (2016) 1e10. [16] A. Kujawska, J. Kujawski, M. Bryjak, W. Kujawski, ABE fermentation products recovery methods - a review, Renew. Sustain. Energy Rev. 48 (2015) 648e661. [17] A.P. Mariano, C.B.B. Costa, D.D.F. de Angelis, F. Maugeri Filho, D.I.P. Atala, M.R. Wolf Maciel, R. Maciel Filho, Optimisation of a continuous flash fermentation for butanol production using the response surface methodology, Chem. Eng. Res. Des. 88 (5e6) (2010) 562e571. [18] C. Xue, F. Liu, M. Xu, I.-C. Tang, J. Zhao, F. Bai, S.-T. Yang, Butanol production in acetone-butanol-ethanol fermentation with in situ product recovery by adsorption, Bioresour. Technol. 219 (2016) 158e168. [19] C. Xue, F. Liu, M. Xu, J. Zhao, L. Chen, J. Ren, F. Bai, S.-T. Yang, A novel in situ gas stripping-pervaporation proces integrated with acetone-butanol-ethanol in acetone-butanol-ethanol fermentation for hyper n-butanol production, Biotechnol. Bioeng. 113 (1) (2016b) 120e129. [20] Y.-C. Liao, K.-M. Lu, S.-Y. Li, Process parameters for operating 1-butanol gas stripping in a fermentor, J. Biosci. Bioeng. 118 (5) (2014) 558e564. pez-Contreras, [21] T. de Vrije, M. Budde, H. van der Wal, P.A.M. Claassen, A.M. Lo “In situ” removal of isopropanol, butanol and ethanol from fermentation broth by gas stripping, Bioresour. Technol. 137 (2013) 153e159. [22] T.C. Ezeji, N. Qureshi, H.P. Blaschek, Production of acetone, butanol and
160
[23]
[24]
[25] [26]
[27]
[28]
[29] [30]
[31]
n et al. / Biomass and Bioenergy 98 (2017) 153e160 E. Rocho ethanol by Clostridium beijerinckii BA101 and in situ recovery by gas stripping, World J. Microbiol. Biotechnol. 19 (2003) 595e603. T.C. Ezeji, N. Qureshi, H.P. Blaschek, Acetone butanol ethanol (ABE) production from concentrated substrate: reduction in substrate inhibition by fed-batch technique and product inhibition by gas stripping, Appl. Microbiol. Biotechnol. 63 (6) (2004) 653e658. T.C. Ezeji, P.M. Karcher, N. Qureshi, H.P. Blaschek, Improving performance of a gas stripping-based recovery system to remove butanol from Clostridium beijerinckii fermentation, Bioprocess Biosyst. Eng. 27 (3) (2005) 207e214. T.C. Ezeji, N. Qureshi, H.P. Blaschek, Bioproduction of butanol from biomass: from genes to bioreactors, Curr. Op. Biotechnol. 18 (3) (2007) 220e227. N. Qureshi, B.C. Saha, M.A. Cotta, Butanol production from wheat straw by simultaneous saccharification and fermentation using Clostridium beijerinckii: Part IIdfed-batch fermentation, Biomass Bioenergy 32 (2) (2008) 176e183. N. Qureshi, B.C. Saha, R.E. Hector, M.A. Cotta, Removal of fermentation inhibitors from alkaline peroxide pretreated and enzymatically hydrolyzed wheat straw: production of butanol from hydrolysate using Clostridium beijerinckii in batch reactors, Biomass Bioenergy 32 (12) (2008) 1353e1358. C. Xue, G. Du, J. Sun, L. Chen, S. Gao, M. Yu, S.-T. Yang, F. Bai, Characterization of gas stripping and its integration with acetone e butanol e ethanol fermentation for high-efficient butanol production and recovery, Biochem. Eng. J. 83 (2014) 55e61. O. Levenspiel, The Monod equation: a Revisit and a Generalization to product inhibition situations, Biotechnol. Bioeng. 22 (1980) 1671e1687. W.-L. Chang, Acetone-butanol-ethanol Fermentation by Engineered Clostridium Beijerinckii and Clostridium Tyrobutyricum, The Ohio State University, 2010. Thesis. zquez-Sa nchez, M.C. Montes-Horcasitas, R.A. Lo pez, Development of a H.I. Vela
[32] [33]
[34]
[35]
[36] [37] [38]
[39]
[40]
phenomenological kinetic model for butanol production using Clostridium beijerinckii, Rev. Mex. Ing. Quím. 13 (3) (2014) 103e112. K. Truong, J. Blackburn, The stripping of organic chemicals in biological treatment process, Environ. Prog. 3 (3) (1984) 143e152. A. Ranjan, R. Mayank, V.S. Moholkar, Process optimization for butanol production from developed rice straw hydrolysate using Clostridium acetobutylicum MTCC 481 strain, Biomass Conv. Bioref 3 (2) (2012) 143e155. F. Monot, J.-M. Engasser, H. Petitdemange, Influence of pH and undissociated butyric acid on the production of acetone and butanol in batch cultures of Clostridium acetobutylicum, Appl. Microbiol. Biotechnol. 19 (1984) 422e426. F. Monot, J.-R. Martin, H. Petitdemange, R. Gay, Acetone and butanol production by Clostridium acetobutylicum in a synthetic medium, Appl. Environ. Microbiol. 44 (1982) 1318e1324. R. Buttery, L. Ling, D. Guadagni, Volatilities of aldehydes, ketones, and esters in dilute water solution, Food Volatiles 17 (2) (1969) 385e389. R. Sander, Compilation of Henry's law constants (version 4.0) for water as solvent, Atmos. Chem. Phys. 15 (2015) 4399e4981. K.W. Staggs, D.R. Nielsen, Improving n-butanol production in batch and semicontinuous processes through integrated product recovery, Process Biochem. 50 (10) (2015) 1487e1498. D. Cai, Y. Wang, C. Chen, P. Qin, Q. Miao, C. Zhang, P. Li, T. Tan, Acetonebutanol-ethanol from sweet sorghum juice by an immobilized fermentationgas stripping integration process, Bioresour. Technol. 211 (2016) 704e710. pez-Contreras, K.A. Pyrgakis, T. de Vrije, M.A.W. Budde, K. Kyriakou, A.M. Lo A.C. Kokossis, A process integration approach for the production of biological iso-propanol, butanol and ethanol using gas stripping and adsorption as recovery methods, Biochem. Eng. J. 116 (2016) 176e194.