- Email: [email protected]
polyvinylidene fluoride composite membrane
Accepted Manuscript Immobilized ethanol fermentation coupled to pervaporation with silicalite-1/ polydimethylsiloxane/polyvinylidene fluoride composite membrane Di Cai, Song Hu, Changjing Chen, Yong Wang, Changwei Zhang, Qi Miao, Peiyong Qin, Tianwei Tan PII: DOI: Reference:
S0960-8524(16)31164-6 http://dx.doi.org/10.1016/j.biortech.2016.08.036 BITE 16939
To appear in:
Bioresource Technology
Received Date: Revised Date: Accepted Date:
19 June 2016 10 August 2016 11 August 2016
Please cite this article as: Cai, D., Hu, S., Chen, C., Wang, Y., Zhang, C., Miao, Q., Qin, P., Tan, T., Immobilized ethanol fermentation coupled to pervaporation with silicalite-1/polydimethylsiloxane/polyvinylidene fluoride composite membrane, Bioresource Technology (2016), doi: http://dx.doi.org/10.1016/j.biortech.2016.08.036
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Immobilized ethanol fermentation coupled to pervaporation with silicalite-1/polydimethylsiloxane/polyvinylidene fluoride composite membrane
Di Cai, Song Hu, Changjing Chen, Yong Wang, Changwei Zhang, Qi Miao, Peiyong Qin*, Tianwei Tan
*
Corresponding authors
Address: No.15 beisanhuan east road, chaoyang district, beijing, 100029. E-mail: [email protected]
National energy R&D center for biorefinery, Beijing University of Chemical Technology, Beijing 100029, PR China
1
Abstract
A novel silicalite-1/polydimethylsiloxane/ polyvinylidene fluoride hybrid membrane was used in ethanol fermentation-pervaporation integration process. The sweet sorghum bagasse was used as the immobilized carrier. Compared with the conventional suspend cells system, the immobilized fermentation system could provide higher ethanol productivity when coupled with pervaporation. In the long-term of operations, the ethanol productivity, separation factor, total flux and permeate ethanol concentration in the fed-batch fermentation-pervaporation integration scenario were 1.6 g/L h, 8.2-9.9, 319-416 g/m2 h and 426.9-597.2 g/L, respectively. Correspondingly, 1.6 g/L h, 7.8-9.8, 227.8-395 g/m2 h and 410.9-608.1 g/L were achieved in the continuous fermentation-pervaporation integration scenario, respectively. The results indicated that the integration process could greatly impove the ethanol production and separation perfromances.
Key words
Immobilized fermentation; Pervaporation; Silicalite-1/polydimethylsiloxane/polyvinylidene fluoride membrane; Ethanol
2
1. Introduction
Bio-ethanol, the liquid alcohol, is currently concerns as one of the most important direct substitutes for fossil fuel in transportation (Li et al., 2013; Mussatto et al., 2010). Although bio-ethanol has already been under commercialization, the traditional fermentation processes still calls for further improvement to meet the competition of fossil fuels (Fan et al., 2014a). Significantly, due to the end product toxicity, conventional free-cell batch fermentation suffered from low ethanol productivity and cell density, resulted in low ethanol yield, high equipment investment and high energy inputs, and further influence on the overall cost of the bio-ethanol production (Wei et al., 2014; Jin et al., 2011). In order to achieve high ethanol productivity, fed-batch and continuous fermentation are the extractive ones. The reactor volume and the capital cost of the process could be decreased with the relatively high productivity (Kumar and Gayen, 2011). Moreover, cell immobilization technique is another feasible way to overcome the drawbacks of low productivity of ethanol. By applying the immobilized technique, hyper cells are absorbed onto the surface of carrier. Besides that, the interspaces of immobilized carrier could provide a favorable extracellular micro-environment. The robustness of strains could be extensively improved (Yu et al., 2007; Chang et al., 2014; Hartono et al., 2010). Thus, there has the possibility that combined the cells immobilization with fed-batch or continuous fermentation. For the improvement of the ethanol separation process, in recent years, pervaporative separation of volatile organic compound based on the principal of the selective permeation is in the ascendant. This technique has been applied for in situ ethanol recovery from fermentation broth (Shao and Huang, 2007; Wei et al., 2014; Chen et al., 2014; Chovau et al., 2011; Fu et al., 2016; Fan et al., 2016). The application of pervaporation technique was proved to have a positive effect on the ethanol productivity. More importantly, it contributed to the reduction of the energy demand for ethanol separation (Vane, 2005; 2008; Dafoe and Daugulis, 2014). During the integration, ethanol is produced in the bioreactor and simultaneously separated by 3
pervaporation. Unfortunately, in the long-term integration systems, the accumulation of inhibitors includes cell debris and toxic by-product could foul the surface of pervaporation membrane, resulted in the decrease of the separation performances (Dubreuil et al., 2013; Liu et al., 2011). When applying the cells immobilization technique in the fermentation-pervaporation integration process, cells were able to immobilize on the surface of the carriers. Hence, the concentration of cell debris in broth could be decreased, and the tolerance of cells to the toxic by-products would be enhanced. However, up to now, only a few literatures focus on the immobilized bioreactor coupled with pervaporation using alginate as immobilized carrier (Ding et al., 2011; Shabtai et al., 1991). In these studies, cells were entrapping in gels. The alginate material loaded is costly and is a hindrance to the substrate diffusion (Maryse and Dravko, 1996; Yu et al., 2007). More importantly, there was only a little enhancement of fermentation performances when applying these immobilization carriers in the coupled system. In comparison with the alginate carriers, sweet sorghum bagasse (SSB) was proved to be an excellent carrier for yeast cells (Yu et al., 2010; 2012). In this study, SSB was used as the immobilize carrier in the ethanol fermentation-pervaporation integration process for the first time. A novel silicalite-1/polydimethylsiloxane/ polyvinylidene fluoride (silicalite-1/PDMS/PVDF) pervaporation membrane was practiced in the pervaporation unit for effective in situ separation ethanol product from fermentation broth. Performances of batch, fed-batch and continuous fermentation-pervaporation integration system were evaluated and compared. The integration process showed good properties in long-term ethanol production, which may contribute greatly to the efficient and energy saving bio-ethanol fermentation processes.
2. Material and methods
2.1. Culture and inoculum preparation 4
The mutant strain Saccharomyces cerevisiae 3013 was laboratory stored and was used in all experiments. The MY medium for maintain the yeast strain and the seed culture were described in Yu et al. study (2007; 2010). The fermentation media contained (in g/L): glucose 150, polypeptone 80, MgSO4 12, KH2PO4 0.6, followed by adjusting to pH 5.5 and then autoclaved at 121 oC for 20 min before use.
2.2. Pervaporation membrane and module
The pervaporation membrane used throughout the experiments was laboratory made. Silicalite-1 (20 % loading), polydimethylsiloxane (PDMS), and tetraethoxysilane (TEOS) were mixed in n-hexane under magnetic stirring (1000rpm) for 2h. When a homogeneous suspension was obtained, the dibutyltin dilaurate (DBTDL) was added into as catalyst. After stirring for 5 minutes, the solution was then degassed under vacuum and subsequently coated on the polyvinylidene fluoride (PVDF) membrane by using an automatic coating machine (Elcometer 4340, Elcometer Limited). Finally, the composite membrane was kept under room temperature for 24 h in order to assure complete cross-linking and evaporate the residual solvent. A stainless steel module with an effective area of 19.63 cm2 was used for mounting the membrane.
2.3. Immobilized fermentation-pervaporation integration process set up
The sweet sorghum stem was kindly provided by Chinese Academy of Agricultural Sciences (the experimental field in Shunyi, Beijing). The method to generate the SSB carrier has been fully described in the previous study (Chang et al., 2014; Yu et al., 2007). The integration process was set up on a homemade apparatus which was similar with the experimental procedure in Fu et al. (2016) study. When applied in batch operation, about 20 g of SSB was put into the 1 L bioreactor and sterilized. 500 ml sterilized fermentation medium was pumped into the bioreactor. After that, 10 % 5
highly motile cells were inoculated into the sterilized (121 oC for 20 min) fermentation medium, and the bioreactor practiced at 30 oC for ~20 h. Then, the sterilized membrane module was started and the in situ ethanol removal system was practiced. The liquid phase in the immobilized bioreactor was circulated through the membrane module at 0.5 L/min via peristaltic pump (Baoding Chuangrui Precision Pump Co., Ltd., China). At the same time, the pressure on the permeate side of pervaporation membrane was kept at below 100 Pa by a vacuum pump. Permeate samples were collected each 3 h by a cold trap in a liquid nitrogen bath. When operated under the fed-batch fermentation scenario, 4-5 folds of fermentation media was pumped into the bioreactor when the residual glucose concentration below 10 g/L, and brought the glucose concentration back to ~150 g/L level. As for the continuous fermentation, 4-5 folds of the fermentation media was continuously pumped into the bioreactor at a dilution rate of 0.016 /h
2.4. Calculations
Formula for the separation factor of pervaporation was defined as described in literature (Van Hecke, 2013): ߙ௧/௪௧ =
௬ೌ /௬ಹమ ೀ
(1)
௫ೌ /௫ಹమ ೀ
Where αୣ୲୦ୟ୬୭୪/୵ୟ୲ୣ୰ is the separation factor of ethanol/water mixture; x and y refers to the mass ratio of components in the fermentation broth and the permeate, respectively. The flux of component i is defined as: ௐ
ܬ = ∗∆௧
(2)
Where ∆t refers to the operating time; A stand for the membrane area of the module; Wi is the weight of i in the permeate. The pervaporation separation index (PSI) is frequently used to compare and reflect the separation performance of membranes, which is calculated by the following equation (Keawkannetra et al., 2014): 6
ܲܵܬ = ܫ௧௧ (ߙ௧/௪௧ − 1)
(3)
Where Jtotal refers to the total flux of membrane.
2.5. Analysis
As described in previous work (Cai et al., 2013), the ethanol concentration in the fermentation broth and in the permeate side of membrane were determined by a gas chromatography equipped with a flame ionization detector (Trace 1300, Thermofisher Scientic, USA). The residual glucose concentration was determined by a glucose biosensor (SBA 40C, Biological Institute of Shandong Academy of Science, China) according to the study of Xue et al. (2010). And the free cells concentration was measured by a spectrophotometer (UV1902, Shanghai AuCy Technology Instrument Co. Ltd., China) at 600 nm based on the method by Fan et al. (2014a). Simples were tested in duplication and the results were the average ones.
3. Results and discussion
3.1. The comparison of batch fermentation in free cells system and immobilized cells system integration
As mentioned above, the caves on the surface of SSB structure could provide a feasible micro-environment for the microbial cells metabolism. And the SSB based cells immobilization system could dramatically enhance the ethanol productivity (Yu et al., 2007). Similar with the result of Yu et al. (2007), the cells immobilization system using the SSB carrier showed higher productivity compared with conventional free cells system. Fig.1 showed batch fermentation performances in suspended cells system and the cells immobilization system coupled with pervaporation. The initial sugar concentrations were both measured at ~160 g/L. In the free cells coupled system, it took 45 h to consume all of the initial sugar and the average ethanol productivity 7
was 1.5 g/L h. While in the immobilized cells system, only 35 h after inoculation, all of the initial sugar was depleted. Because the fermentation period was obviously shortened (10 h was survived), the average ethanol productivity in the cells immobilization process was increased to 2.03 g/L h, which was 1.4 times higher than that of the suspended cells one. Besides, the overall ethanol concentration generated in the free cells coupled system was 67.7 g/L with a yield of 0.44 g/g. By contrast, about 70.4 g/L of ethanol was cumulated in the cells immobilization process, and the yield of ethanol was slight increased to 0.45 g/g. As for the ethanol concentration remained in the bioreactor, since the fermentation period of suspended cells system was longer than that of the cells immobilization one, the integration time in free cells scenario was also longer. As a result, more ethanol was separated in the free cells system. On the other hand, owing to the relatively low productivity of ethanol, it supplied lower amount of ethanol in unit time in the suspended cells system when compared with the cells immobilization system. Correspondingly, with the synergistic effect of ethanol production and in situ removal, a lower concentration of ethanol was achieved in the free cells system. By contrast, about 54.3 g/L of ethanol was generated in the immobilized bioreactor when the fermentation terminated. It was 23.8 g/L higher than that of free cells based system. Besides, thanks to the stable micro-environment in the immobilized bioreactor, the metabolism of yeast was not inhibited with the higher ethanol concentration remained. Additionally, since the higher concentration of ethanol in feed corresponded to higher total permeate flux and solvent concentration (Qin et al., 2014; Jee and Lee, 2014), the fermentation-pervaporation coupled system with cells immobilization provided the better pervaporation performance. Moreover, the suspended cells concentration at the end of the conventional suspended cells bioreactor and the cells immobilization bioreactor coupled system were 12.65 g/L and 8.32 g/L, respectively. That is, a large part of yeast cells was successfully immobilized on the SSB carrier. Since the suspended cells and the cell debris have a negative effect on the performance of pervaporation especially the ethanol permeability (Vane et al., 2010; Liu et al., 2011), the coupled system with the 8
SSB carrier might be of benefit to the ethanol separation by pervaporation, especially in long-terms of operation.
3.2 Fed-batch fermentation-pervaporation integration
Fig.2 shows the kinetics of fed-batch fermentation integrated with continuous pervaporation using SSB carrier within 190 h. As can be seen from Fig.2a, the fed-batch fermentation-pervaporation coupled system provided relatively stable ethanol concentration in the fermentation broth. Ethanol concentration was fluctuated between 82.4-108.2 g/L after 80 h of fermentation in the cells immobilized bioreactor. The suspended yeast cells concentration showed a clear increase in the first two cycles, then, it was maintained at 10.2-17.7 g/L. In comparison with the 17.1-19.8 g/L suspended cells obtained in the suspended cells fermentation-pervaporation process (Fan et al., 2014b), when applying the SSB as the carrier for immobilization, the suspended cells concentration was maintained at a low level, which helps to protect the pervaporation membrane from the contamination of cells (Liu et al., 2011; Vane et al., 2010). It should be note here that the suspended cells in the immobilization system have lower activity than the immobilized ones (Yu et al., 2007; 2010). As it was indicated in Liu et al. study (2011), the pervaporation membrane fouling only occurred in the live broth. Therefore, the membrane fouling caused by cells might be significantly limited by the inactive suspended cells in the cells immobilization system. Besides that, it showed that when integrated suspended cells fermentation processes with pervaporation, a dramatic decrease of yeast concentration was occurred in long terms of operation (Fan et al., 2014a; 2014b; Chen et al., 2012). It is likely attributed to the inhibition of several secondary metabolites to the cells metabolism. Since the toxic organic acids and glycerol cannot probably dissolve into the nonpolar PDMS layer of the pervaporation membrane, the accumulation of secondary metabolites was reported to inhibit the active cells in broth (Fan et al., 2014a). However, fortunately, the micro-environment provided by the SSB carrier could protect the yeast cells from the deteriorating culture (Yu et al., 2010). As a result, 9
the suspended cells concentration in the current work showed a general upward trend, though there was a considerably large fluctuation of the suspended cells concentration with time. Overall, at the end of fed-batch fermentation, a total 301.2 g/L of ethanol was produced from 682 g/L of glucose (Fig. 2b). The overall ethanol yield from glucose consumed was 0.44 g/g. As it is shown in Fig.2c, the ethanol productivity was gradually decreased with time until 23 h. After that, an obvious raise of ethanol productivity was occurred due to the integration of pervaporation. With the increase of ethanol concentration remained in the fermentation broth, the ethanol productivity was decreased with time and finally maintained at 1.6 g/L h level. A zigzag curve of glucose consumption rate with time was generated as well. It showed positive correlation property to the residual glucose concentration. Generally, the higher the glucose concentration in broth, the higher glucose consumption rate obtained. The relative low consumption rate under low glucose concentration might be caused by the lack of carbon source and nutritions. On the permeate side of pervaporation membrane, high ethanol concentration ranged from 426.9-597.2 g/L was obtained, suggesting that the ethanol product of fermentation was effectively separated and concentrated by the noval silicalite-1/PDMS/PVDF pervaporation membrane (Fig.3a). As it illustrated in Fig.3b, the ethanol flux was gradully increased in the first 80 h and then maintianed at 147.6-247.5 g/m2 h level. Therefore, the ethanol flux showed positive correlation to the ethanol titer in the fermentation broth. With the increase ethanol concentration on the feed side of pervaporation membrane, the free volume and the chain mobility of the silicalite-1/PDMS/PVDF layer were increased (Jee and Lee, 2014; Bettens et al., 2010). As for water flux, since water transport trough the pervaporation membrane is independent to the alcohol concentration in feed (Niemistö et al., 2013; Qin et al., 2014), the water flux was constant during the opration with values around 119.4-225.1 g/m2 h. It is also a great evidence to the long-term stability of both the novel pervaporation membrane and the integration process. In previous literatures, a drop of membrnae permeability was occurred when the actual fermentation broth 10
contaminated the surface of membrane (Vane et al., 2010). However, during the almost 200 h of operation in this work, the water flux was stable while ethanol flux showed co-relations to the retentate borth, which might be also appreciate to the cells immobilization. The total flux influenced by the water and ethanol flux showed similar trend with the ethanol flux. 319-416 g/m2 h of total flux was obtained in the stable stage of permeance. Additionally, the separation factor and PSI were waved from 8.2 and 2561.4 to 9.9 and 3759.2, respectively. And briefly, a stable integrtaion performance was obtained in the fed-batch fermentation-pervaporation integration process using SSB as immobilization carrier.
3.3 Continuous fermentation-pervaporation integration
Kinetics of long-term continuous fermentation-pervapration integrtaed system was further envaluated. Fig. 4 shows the time course profiles of ethanol production during the fermentation. The ethanol concentration in the continuous fermentation was gradually increased and then maintained at 90-110 g/L after 70 h of fermentation, which was higher than that of the fed-batch scenario (Fig.4a). Therefore, the metabolism of yeast in the continuous fermentation process was more vigorous than the strains in the fed-batch bioreactor. As the permeatability of the pervaporation membrane is the rate-determining step, ethanol in the feed side (bioreactor) was gradully cumulated and maintained a higher titer. At the same time, the productivity of strains was decreased due to the inhibition of the solvent product, while the ethanol separation rate was increased by the higher concentration in broth. Thus, a dynamic equilibrium between the ethanol production and separation was occurred, which was simialr with the phenomenon described in Xue et al. (2015) study. As for the time course of the free cells concentration in the fermentation broth, it was generally similar with the result of fed-batch one. The only different of the two scenarios is that the continuous fermentation porcess had a higher suspended cells concentration in the stable stage, suggesting that the metabolism of yeast increased in the continuous fermnetation process and further influenced the ethanol productivity. 11
Overall, after 190 h of fermentation, about 305.6 g/L of ethanol was produced from 719.7 g/L of glucose, while the ethanol productivity and yield were 1.61 g/L h and 0.42 g/g respectively, under a dilution rate of 0.016 /h (Fig.4b). The ethanol yield in the continuous fermentation system was a little shorter (0.03 g/g) than that of the fed-batch fermentation scenario, while the productivity in the continuous fermentation scenario was higher than that of the fed-batch one. It might be caused by the more vigorous bacteria metabolism in the continuous fermentation system. In that case, more carbon source was ultized as the substrate for the bio-synthesize of cell structure, which resulted in lower ethanol yield. As for the residual glucose in the fermentation broth, it was increased slightly with time, indicated that the envrionment for cells metabolism was more rigorous due to the acumulation of ethanol. At the end of fermentation, there were 54 g/L of glucose remained in the fermentation broth. The kinetics of ethanol productivity was first increased and then gradully decreased. In addition, the glucose consumption rate was dropped to 2.8 g/L h at 31 h, then fluctuated from 5.5 g/L h to7.5 g/L h until at the end of fermentation (Fig.4c). The drop of glucose consumption rate was likely due to the lack of carbon source at the beginning of continuous opreation. On the permeate side of membrane, 410.9-608.1 g/L of ethanol product was achieved after 62 h (Fig.5a), which was similar with the ethanol production obtained by the fed-batch fermentation-pervaporation integrated process. In consideration of the relatively higher ethanol concentration in the continuous bioreactor, the sepration factor of ethanol in the continuous scenario was little lower than that of the fed-batch scenario (see Fig.5c). This phenomenon could be explained by the contamination of the suspended cells. Higher suspended cells concentrtaion could clog the channel on the membrane surface. Fortunately, the higher concentrtaion of suspended cells showed little influence on the flux of membrane during the total 190 h of operation. Total flux and ethanol flux ranging from 227.9 g/m2 h to 395 g/m2 h and 86 g/m2 h to 208.2 g/m2 h were obtained, respectively. Similar with the phenomenon of the fed-batch fermentation, the water flux remained stable, ranging from 129.7 g/m2 h to 203.7 g/m2 h (Fig.5b). Additionally, similar with the trend of separation factor, the PSI 12
showed a zigzag curve, and maitained at a stable level of 2957.3±495.
3.4 Comparaion to other studies
Pervaporation is an attractive technique for in situ recovery of ethanol from fermentation broth (Vane, 2008; Fan et al., 2014a). However, one of the challenge for its industrail application is the ageing and fouling effect of membrane in long terms of operation (Vane et al., 2010). In the current work, SSB, the agriculture residual, was applied for cells immobilization in the fermentation-pervaporation separation processes. Although gels carrier have already been applied in the similar processes, to our best knowledge, it is the first time that using lignocellulosic material as carrier in the integration processes. The SSB carrier, with unique straucture on the surface, could hugely imporve the overall fermentation and pervapration performances. As can be seen from Table 1, compared with relative studies, the separation factor of ethanol in the current work is higher than others (on average). It demonstrated that higher feed concentration and larger separation factor correspond to lower ethanol recovery energy (Vane, 2005; 2008; Van Hecke, et al., 2012). Hence, the energy demand for ethanol recovery in the current work was lower than others. Besides that, the membrane performances including separation factor and the fluxes was stable in long-term integration. Therefore, the novel silicalite-1/PDMS/PVDF composite membrane could generally meet the demand of ethanol separation in long-term fermentation process. Previous works, however, showed a descent of membrane performances in the long-term of operations (Fan et al., 2014b). And the secondary metabolites showed more severe inhibition to the yeast cells in the suspended cells system (Fan et al., 2014a). By contrast, the current process with SSB as immobilized carrier showed long-term stability. More importantly, with little containmination of the fermentation broth, the pervaporation membrane might have a longer age for ethanol separation, then, further avioding the waste of memebrane and decreasing the membrane cost. 13
4. Conclusions
A system for ethanol fermentation was constracted by coupling the cells immobilization fermentation process with pervaporation unit using silicalite-1/PDMS/PVDF membrane. The SSB carrier led to a superior ethanol productivity and low concentrations of suspended cells, while the novel pervaporation membrane was effective in long-term operation. Ethanol sepration factor ranging from 8.2 to 9.9 and 7.8 to 9.8 were achieved under fed-batch and continuous fermentation, respectively. The integration processes provided high ethanol productivity and permenate ethanol concentration.
Acknowledgements
This work was supported in part by the National Basic Research Program of China (Grant No. 2013CB733600), the National Nature Science Foundation of China (Grant Nos. 21390202, 21476015), National High-Tech R&D Program of China (Grant Nos. 2014AA021904, 2014AA021903), the Fundamental Research Funds for the Central Universities (Grant No. YS1407), and the long-term subsidy mechanism from the Ministry of Finance and the Ministry of Education of PRC.
14
References
1. Bettens, B., Verhoef, A., van Veen, H.M., Vandecasteele, C., Degréve, J., Van der Bruggen, B., 2010. Pervaporation of binary water-alcohol and methanol-alcohol mixtures through microporous methylated silica membranes: Maxwell-Stefan modeling. Comput. Chem. Eng. 34, 1775-1788. 2. Cai, D., Chang, Z., Wang, C., Ren, W., Wang, Z., Qin, P., Tan, T., 2013. Impact of sweet sorghum cuticular waxes (SSCW) on acetone-butanol-ethanol fermentation using Clostridium acetobutylicum ABE 1201. Bioresour. Technol. 149, 470-473. 3. Chang, Z., Cai, D., Wang, C., Li, L., Han, J., Qin, P., Wang, Z., 2014. Sweet sorghum bagasse as an immobilized carrier for ABE fermentation by using Clostridium acetobutylicum ABE 1201. RSC Adv. 4, 21819-21825. 4. Chen, C., Tang, X., Xiao, Z., Zhou, Y., Jiang, Y., Fu, S., 2012. Ethanol fermentation kinetics in a continuous and closed-circulating fermentation system with a pervaporation membrane bioreactor. Bioresour. Technol. 114, 707-710. 5. Chen, J., Zhang, H., Wei, P., Zhang, L., Huang, H., 2014. Pervaporative behavior and integrated process for concentrating lignocellulosic ethanol through polydimethylsiloxane (PDMS) membrane. Bioprocess Biosyst. Eng. 37, 183-191. 6. Chovau, S., Gaykawad, S., Straathof, A.J.J., Van der Bruggen, B., 2011. Influence of fermentation by-products on the purification of ethanol from water using pervaporation. Bioresour. Technol. 102, 1669-1674. 7. Dagoe, J.T., Daugulis, A.J., 2014. In situ product removal in fermentation systems: improved process performance and rational extractant selection. Biotechnol. Lett. 36, 443-460. 8. Debreuil, M.F.S., Vandezande, P., Van Hecke, W.H.S., Porto-Carrero, W.J., Dotremont, C.T.E., 2013. Study on ageing/fouling phenomena and the effect of upstream nanofultration on in-situ product recovery of n-butanol through ploy[1-(trimethylsilyl)-1-propyne] pervaporation membranes. J. Membr. Sci. 447, 15
134-143. 9. Ding, W.W., Wu, Y.T., Tang, X.Y., Yuan, L., Xiao, Z.Y., 2011. Continuous ethanol fermentation in a closed-circulation system using an immobilized cell coupled with PDMS membrane pervaporation. J. Chem. Technol. Biotechnol. 86, 92-87. 10. Ding, W., Xiao, Z., Tang, X., Deng, K., Fu, S., Jiang, Y., Yuan, L., 2012. Evolutionary engineering of yeast for closed-circulating ethanol fermentation in PDMS membrane bioreactor. Biochem. Eng. J., 60, 56-61. 11. Fan, S., Xiao, Z., Tang, X., Chen, C., Zhang, C., Zhang, Y., Deng, Q., Yao, P., Li, W., 2014a. Inhibition effect of secondary metabolites accumulated in a pervaporation membrane bioreactor on ethanol fermentation of Saccharomyces cerevisiae. Bioresour. Thecnol. 162, 8-13. 12. Fan, S., Xiao, Z., Zhang, Y., Tang, X., Chen, C., Li, W., Deng, Q., Yao, P., 2014b. Enhance ethanol fermentation in a pervaporation membrane bioreactor with the convenient permeate vapor recovery. Bioresour. Technol. 155, 229-234. 13. Fan, S., Xiao, Z., Li, M., 2016. Energy efficient of ethanol recovery in pervaporation membrane bioreactor with mechanical vapor compression eliminating the cold traps. Bioresour. Technol. 211, 24-30. 14. Fu, C., Cai, D., Hu, S., Miao, Q., Wang, Y., Qin, P., Wang, Z., Tan, T., 2016. Ethanol fermentation integrated with PDMS composite membrane: An effective process. Bioresour. Technol. 200, 648-657. 15. Hartono, S.B., Qiao, S.Z., Liu, J., Jack, K., Ladewig, B.P., Hao, Z., Lu, G.Q.M., 2010. Functinaolized mesoporous silica with very large pores for cellulase immobilization. J. Phys. Chem. C. 114, 8353-8362. 16. Jee, K.Y., Lee, Y.T., 2014. Preparation and characterization of siloxane composite membranes for n-butanol concentration from ABE solution by pervaporation. J. Membr. Sci. 456, 1-10. 17. Jin, C., Yao, M., Liu, H., Lee, C.F., Ji, J., 2011. Progress in the production and application of n-butanol as a biofuel. Renew. Sustain. Energ. Rev. 15, 4080-4106. 18. Kaewkannetra, P., Chutinate, N., Moonamart, S., Kamsan, T., Chiu, T.Y., 2012. 16
Experimental study and cost evaluation for ethanol separation from fermentation broth using pervaporation. Desalin. Water Treat. 41, 88-94. 19. Kumar, M., Gayen, K., 2011. Developments in biobutanol production: New insights. Appl. Energ. 88, 1999-2012. 20. Li, S., Li, G., Zhang, L., Zhou, Z., Han, B., Hou, W., Wang, J., Li, T., 2013. A demonstration study of ethanol production from sweet sorghum stems with advanced solid state fermentation technology. Appl. Energ. 102, 260-265. 21. Liu, G., Wei, W., Wu, H., Dong, X., Jiang, M., Jin, W., 2011. Pervaporation performance of PDMS/ceramic composite membrane in acetone butanol ethanol (ABE) fermentation-PV coupled process. J. Membr. Sci. 373, 121-129. 22. Maryse, G., Dravko, D., 1996. Wood blocks as a carrier for Saccharomyces cerevisiae used in the production of ethanol and fructose. Chem. Eng. J. 61, 233-240. 23. Mussatto, S.I., Dragone, G., Guimarᾶes, P.M.R., Silva, J.P.A., Carneiro, L.M., Roberto, I.C., Vicente, A., Domingues, L., Teixeira, J.A., 2010. Technological trends, global market, and challenges of bio-ethanol production. Biotechnol. Adv. 28, 817-830. 24. Shabtai, Y., Chaimovitz, S., Freeman, A., Katchalski-Katzir, E., Linder, C., Nemas, M., Perry, M., Kedem, O., 1991. Continuous ethanol production by immobilized yeast reactor coupled with membrane pervaporation unit. Biotechnol. Bioeng. 38, 869-876. 25. Shao, P., Huang, R.Y.M., 2007. Polymeric membrane pervaporation. J. Membr. Sci. 287, 162-169. 26. Wei, P., Cheng, L.H., Zhang, L., Xu, X.H., Chen, H.L., Gao, C.J., 2014. A review of membrane technology for bioethanol production. Renew. Sustain. Energ. Rev. 30, 388-400. 27. Van Hecke, W., Vandezande, P., Claes, S., Vangeel, S., Beckers, H., Diels, L., De Wever, H., 2012. Integrated bioprocess for long-term continuous cultivation of Clostridium acetobutylicum coupled to pervaporation with PDMS composite membranes. Bioresour. Technol. 111, 368-377. 17
28. Vane, L.M., 2005. A review of pervaporation for product recovery from biomass fermentation processes. J. Chem. Technol. Biotechnol. 80, 603-629. 29. Vane, L.M., 2008. Separation technologies for the recovery and dehydration of alcohols from fermentation broth. Biofuels Bioprod. Bioref. 2, 553-588. 30. Vane, L.M., Namboodiri, V.V., Meier, R.G., 2010. Factors affecting alcohol-water pervaporation performance of hydrophobic zeolite-silicone rubber mixed matrix membranes. J. Membr. Sci. 364, 102-110. 31. Xue, F., Gao, B., Zhu, Y., Zhang, X., Feng, W., Tan, T., 2010. Pilot-scale production of microbial lipid using starch wastewater as raw material. Bioresour. Technol. 101, 6092-6095. 32. Xue, C., Yang, D., Du, G., Chen, L., Ren, J., Bai, F., 2015. Evaluation of hydrophobic micro-zeolite-mixed matrix membrane and integrated with acetone-butanol-ethanol fermentation for enhanced butanol production. Biotechnol. Biofuels 8, 105. 33. Yu, J., Yue, G., Zhong, J., Zhang, X., Tan, T., 2010. Immobilization of Saccharomyces cerevisiae to modified bagasse for ethanol production. Renew. Energ. 35, 1130-1134. 34. Yu, J., Zhang, T., Zhong, J., Zhang, X., Tan, T., 2012. Biorefinery of sweet sorghum stem. Biotechnol. Adv. 30, 811-816. 35. Yu, J., Zhang, X., Tan, T., 2007. An novel immobilization method of Saccharomyces cerevisiae to sorghum bagasse for ethanol production. J. Biotechnol.129, 415-420.
18
Figure captions
Fig.1 The comparison of the batch fermentation kinetics between immobilized cells and free cells system integrated with continuous pervaporation. Fig.2 Fermentation kinetics of fed-batch ethanol fermentation with continuous pervaporation for ethanol recovery.(a) Cumulated ethanol production and residual glucose concentration remained in the fermentation broth; (b) Ethanol concentration in the fermentation broth and the yeast cell concentration; (c) Ethanol productivity and glucose consumption rate in 6 feeding cycles Fig.3 Membrane performance during the fed-batch ethanol fermentation with continuous pervaporation.(a) Ethanol titer on the permeate side obtained in 6 feeding cycles; (b) Ethanol, water and total flux; (c) The separation factor of ethanol and the overall PSI. Fig.4 Fermentation kinetics of continuous ethanol fermentation with continuous pervaporation for ethanol recovery, the dilution rate of fermentation was 0.016/h. (a) Cumulated ethanol production and residual glucose concentration remained in the fermentation broth; (b) Ethanol concentration in the fermentation broth and the yeast cell concentration; (c) Ethanol productivity and glucose consumption rate in 190 h. Fig.5 Membrane performances during of the continuous ethanol fermentation with continuous pervaporation, the dilution rate of fermentation was 0.016/h. (a) Ethanol titer on the permeate side obtained in 190 h; (b) Ethanol, water and total flux; (c) The separation factor of ethanol and the overall PSI.
19
Fig.1
20
Fig.2
21
Fig.3
22
Fig.4
23
Fig.5
24
Table 1 Study
Current advance in ethanol fermentation integrated with pervaporation under different fermentation types. Chen et
Fu et al.,
Fu et al.,
Ding et al.,
Chen et al.,
Fan et al.,
Shabtai et
Ding et al.,
al., 2014
2016
2016
2012
2012
2014b
al., 1991
2011
Fermentation
Fed-batch
Fed-batch
Continuous
Continuous
Continuous
Continuous
Continuous
Operation
Suspended
Suspended
Suspended
Suspended
Suspended
Suspended
cells
cells
cells
cells
cells
PDMS
PDMS/PVDF
PDMS/PVDF
PDMS
PDMS
Membrane
This study
This study
Continuous
Fed-batch
Continuous
Immobilized
Immobilized
Immobilized cells
Immobilized cells
cells
cells
cells
PDMS
PS/PSF
PDMS/PA
silicalite-1/PDMS
silicalite-1/PDMS
/PVDF
/PVDF
type Total flux
1300-1400 396.2-663.7
332.4-548.1
-
774
370
1750
300–690
319-416
227.8-395
4.1-5 a
8.6-11.7
8-11
-
4.85
9.5
5b
5.0–7.2 b
8.2-9.9
7.8-9.8
-
70-80
~100
47.9-54.2
45.3-48.8
46
30 (g/g)
48 (g/g)
82.4-108.2
90-110
25.8-30.2
417.2
~446.3
-
216.2-221.1
320
17–20 (g/g)
190-310
426.9-597.2
410.9-608.1
(g/m2 h) Separation factor Ethanol conc. in broth (g/L) Ethanol conc. in
(g/g)
25
permeate (g/L) yield (g/g)
-
0.42
0.42
0.42-0.43
0.37
0.38
-
0.41
0.44
0.42
Productivity
13.4
1.58
1.78
1.5-1.6
1.51
2.31
30
9.6
~1.6
~1.6
(g/L h) a
Estimated by authors.
b
Selectivity.
1. Sweet sorghum bagasse was used as the immobilized carrier. 2. Silicalite-1/PDMS/PVDF membrane was used for in situ ethanol separation. 3. 410.9-608.1 g/L of ethanol was obtained in the integration process. 4. Approximate 10 of ethanol separation factor was obtained in the integration process.
26
S0960-8524(16)31164-6 http://dx.doi.org/10.1016/j.biortech.2016.08.036 BITE 16939
To appear in:
Bioresource Technology
Received Date: Revised Date: Accepted Date:
19 June 2016 10 August 2016 11 August 2016
Please cite this article as: Cai, D., Hu, S., Chen, C., Wang, Y., Zhang, C., Miao, Q., Qin, P., Tan, T., Immobilized ethanol fermentation coupled to pervaporation with silicalite-1/polydimethylsiloxane/polyvinylidene fluoride composite membrane, Bioresource Technology (2016), doi: http://dx.doi.org/10.1016/j.biortech.2016.08.036
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Immobilized ethanol fermentation coupled to pervaporation with silicalite-1/polydimethylsiloxane/polyvinylidene fluoride composite membrane
Di Cai, Song Hu, Changjing Chen, Yong Wang, Changwei Zhang, Qi Miao, Peiyong Qin*, Tianwei Tan
*
Corresponding authors
Address: No.15 beisanhuan east road, chaoyang district, beijing, 100029. E-mail: [email protected]
National energy R&D center for biorefinery, Beijing University of Chemical Technology, Beijing 100029, PR China
1
Abstract
A novel silicalite-1/polydimethylsiloxane/ polyvinylidene fluoride hybrid membrane was used in ethanol fermentation-pervaporation integration process. The sweet sorghum bagasse was used as the immobilized carrier. Compared with the conventional suspend cells system, the immobilized fermentation system could provide higher ethanol productivity when coupled with pervaporation. In the long-term of operations, the ethanol productivity, separation factor, total flux and permeate ethanol concentration in the fed-batch fermentation-pervaporation integration scenario were 1.6 g/L h, 8.2-9.9, 319-416 g/m2 h and 426.9-597.2 g/L, respectively. Correspondingly, 1.6 g/L h, 7.8-9.8, 227.8-395 g/m2 h and 410.9-608.1 g/L were achieved in the continuous fermentation-pervaporation integration scenario, respectively. The results indicated that the integration process could greatly impove the ethanol production and separation perfromances.
Key words
Immobilized fermentation; Pervaporation; Silicalite-1/polydimethylsiloxane/polyvinylidene fluoride membrane; Ethanol
2
1. Introduction
Bio-ethanol, the liquid alcohol, is currently concerns as one of the most important direct substitutes for fossil fuel in transportation (Li et al., 2013; Mussatto et al., 2010). Although bio-ethanol has already been under commercialization, the traditional fermentation processes still calls for further improvement to meet the competition of fossil fuels (Fan et al., 2014a). Significantly, due to the end product toxicity, conventional free-cell batch fermentation suffered from low ethanol productivity and cell density, resulted in low ethanol yield, high equipment investment and high energy inputs, and further influence on the overall cost of the bio-ethanol production (Wei et al., 2014; Jin et al., 2011). In order to achieve high ethanol productivity, fed-batch and continuous fermentation are the extractive ones. The reactor volume and the capital cost of the process could be decreased with the relatively high productivity (Kumar and Gayen, 2011). Moreover, cell immobilization technique is another feasible way to overcome the drawbacks of low productivity of ethanol. By applying the immobilized technique, hyper cells are absorbed onto the surface of carrier. Besides that, the interspaces of immobilized carrier could provide a favorable extracellular micro-environment. The robustness of strains could be extensively improved (Yu et al., 2007; Chang et al., 2014; Hartono et al., 2010). Thus, there has the possibility that combined the cells immobilization with fed-batch or continuous fermentation. For the improvement of the ethanol separation process, in recent years, pervaporative separation of volatile organic compound based on the principal of the selective permeation is in the ascendant. This technique has been applied for in situ ethanol recovery from fermentation broth (Shao and Huang, 2007; Wei et al., 2014; Chen et al., 2014; Chovau et al., 2011; Fu et al., 2016; Fan et al., 2016). The application of pervaporation technique was proved to have a positive effect on the ethanol productivity. More importantly, it contributed to the reduction of the energy demand for ethanol separation (Vane, 2005; 2008; Dafoe and Daugulis, 2014). During the integration, ethanol is produced in the bioreactor and simultaneously separated by 3
pervaporation. Unfortunately, in the long-term integration systems, the accumulation of inhibitors includes cell debris and toxic by-product could foul the surface of pervaporation membrane, resulted in the decrease of the separation performances (Dubreuil et al., 2013; Liu et al., 2011). When applying the cells immobilization technique in the fermentation-pervaporation integration process, cells were able to immobilize on the surface of the carriers. Hence, the concentration of cell debris in broth could be decreased, and the tolerance of cells to the toxic by-products would be enhanced. However, up to now, only a few literatures focus on the immobilized bioreactor coupled with pervaporation using alginate as immobilized carrier (Ding et al., 2011; Shabtai et al., 1991). In these studies, cells were entrapping in gels. The alginate material loaded is costly and is a hindrance to the substrate diffusion (Maryse and Dravko, 1996; Yu et al., 2007). More importantly, there was only a little enhancement of fermentation performances when applying these immobilization carriers in the coupled system. In comparison with the alginate carriers, sweet sorghum bagasse (SSB) was proved to be an excellent carrier for yeast cells (Yu et al., 2010; 2012). In this study, SSB was used as the immobilize carrier in the ethanol fermentation-pervaporation integration process for the first time. A novel silicalite-1/polydimethylsiloxane/ polyvinylidene fluoride (silicalite-1/PDMS/PVDF) pervaporation membrane was practiced in the pervaporation unit for effective in situ separation ethanol product from fermentation broth. Performances of batch, fed-batch and continuous fermentation-pervaporation integration system were evaluated and compared. The integration process showed good properties in long-term ethanol production, which may contribute greatly to the efficient and energy saving bio-ethanol fermentation processes.
2. Material and methods
2.1. Culture and inoculum preparation 4
The mutant strain Saccharomyces cerevisiae 3013 was laboratory stored and was used in all experiments. The MY medium for maintain the yeast strain and the seed culture were described in Yu et al. study (2007; 2010). The fermentation media contained (in g/L): glucose 150, polypeptone 80, MgSO4 12, KH2PO4 0.6, followed by adjusting to pH 5.5 and then autoclaved at 121 oC for 20 min before use.
2.2. Pervaporation membrane and module
The pervaporation membrane used throughout the experiments was laboratory made. Silicalite-1 (20 % loading), polydimethylsiloxane (PDMS), and tetraethoxysilane (TEOS) were mixed in n-hexane under magnetic stirring (1000rpm) for 2h. When a homogeneous suspension was obtained, the dibutyltin dilaurate (DBTDL) was added into as catalyst. After stirring for 5 minutes, the solution was then degassed under vacuum and subsequently coated on the polyvinylidene fluoride (PVDF) membrane by using an automatic coating machine (Elcometer 4340, Elcometer Limited). Finally, the composite membrane was kept under room temperature for 24 h in order to assure complete cross-linking and evaporate the residual solvent. A stainless steel module with an effective area of 19.63 cm2 was used for mounting the membrane.
2.3. Immobilized fermentation-pervaporation integration process set up
The sweet sorghum stem was kindly provided by Chinese Academy of Agricultural Sciences (the experimental field in Shunyi, Beijing). The method to generate the SSB carrier has been fully described in the previous study (Chang et al., 2014; Yu et al., 2007). The integration process was set up on a homemade apparatus which was similar with the experimental procedure in Fu et al. (2016) study. When applied in batch operation, about 20 g of SSB was put into the 1 L bioreactor and sterilized. 500 ml sterilized fermentation medium was pumped into the bioreactor. After that, 10 % 5
highly motile cells were inoculated into the sterilized (121 oC for 20 min) fermentation medium, and the bioreactor practiced at 30 oC for ~20 h. Then, the sterilized membrane module was started and the in situ ethanol removal system was practiced. The liquid phase in the immobilized bioreactor was circulated through the membrane module at 0.5 L/min via peristaltic pump (Baoding Chuangrui Precision Pump Co., Ltd., China). At the same time, the pressure on the permeate side of pervaporation membrane was kept at below 100 Pa by a vacuum pump. Permeate samples were collected each 3 h by a cold trap in a liquid nitrogen bath. When operated under the fed-batch fermentation scenario, 4-5 folds of fermentation media was pumped into the bioreactor when the residual glucose concentration below 10 g/L, and brought the glucose concentration back to ~150 g/L level. As for the continuous fermentation, 4-5 folds of the fermentation media was continuously pumped into the bioreactor at a dilution rate of 0.016 /h
2.4. Calculations
Formula for the separation factor of pervaporation was defined as described in literature (Van Hecke, 2013): ߙ௧/௪௧ =
௬ೌ /௬ಹమ ೀ
(1)
௫ೌ /௫ಹమ ೀ
Where αୣ୲୦ୟ୬୭୪/୵ୟ୲ୣ୰ is the separation factor of ethanol/water mixture; x and y refers to the mass ratio of components in the fermentation broth and the permeate, respectively. The flux of component i is defined as: ௐ
ܬ = ∗∆௧
(2)
Where ∆t refers to the operating time; A stand for the membrane area of the module; Wi is the weight of i in the permeate. The pervaporation separation index (PSI) is frequently used to compare and reflect the separation performance of membranes, which is calculated by the following equation (Keawkannetra et al., 2014): 6
ܲܵܬ = ܫ௧௧ (ߙ௧/௪௧ − 1)
(3)
Where Jtotal refers to the total flux of membrane.
2.5. Analysis
As described in previous work (Cai et al., 2013), the ethanol concentration in the fermentation broth and in the permeate side of membrane were determined by a gas chromatography equipped with a flame ionization detector (Trace 1300, Thermofisher Scientic, USA). The residual glucose concentration was determined by a glucose biosensor (SBA 40C, Biological Institute of Shandong Academy of Science, China) according to the study of Xue et al. (2010). And the free cells concentration was measured by a spectrophotometer (UV1902, Shanghai AuCy Technology Instrument Co. Ltd., China) at 600 nm based on the method by Fan et al. (2014a). Simples were tested in duplication and the results were the average ones.
3. Results and discussion
3.1. The comparison of batch fermentation in free cells system and immobilized cells system integration
As mentioned above, the caves on the surface of SSB structure could provide a feasible micro-environment for the microbial cells metabolism. And the SSB based cells immobilization system could dramatically enhance the ethanol productivity (Yu et al., 2007). Similar with the result of Yu et al. (2007), the cells immobilization system using the SSB carrier showed higher productivity compared with conventional free cells system. Fig.1 showed batch fermentation performances in suspended cells system and the cells immobilization system coupled with pervaporation. The initial sugar concentrations were both measured at ~160 g/L. In the free cells coupled system, it took 45 h to consume all of the initial sugar and the average ethanol productivity 7
was 1.5 g/L h. While in the immobilized cells system, only 35 h after inoculation, all of the initial sugar was depleted. Because the fermentation period was obviously shortened (10 h was survived), the average ethanol productivity in the cells immobilization process was increased to 2.03 g/L h, which was 1.4 times higher than that of the suspended cells one. Besides, the overall ethanol concentration generated in the free cells coupled system was 67.7 g/L with a yield of 0.44 g/g. By contrast, about 70.4 g/L of ethanol was cumulated in the cells immobilization process, and the yield of ethanol was slight increased to 0.45 g/g. As for the ethanol concentration remained in the bioreactor, since the fermentation period of suspended cells system was longer than that of the cells immobilization one, the integration time in free cells scenario was also longer. As a result, more ethanol was separated in the free cells system. On the other hand, owing to the relatively low productivity of ethanol, it supplied lower amount of ethanol in unit time in the suspended cells system when compared with the cells immobilization system. Correspondingly, with the synergistic effect of ethanol production and in situ removal, a lower concentration of ethanol was achieved in the free cells system. By contrast, about 54.3 g/L of ethanol was generated in the immobilized bioreactor when the fermentation terminated. It was 23.8 g/L higher than that of free cells based system. Besides, thanks to the stable micro-environment in the immobilized bioreactor, the metabolism of yeast was not inhibited with the higher ethanol concentration remained. Additionally, since the higher concentration of ethanol in feed corresponded to higher total permeate flux and solvent concentration (Qin et al., 2014; Jee and Lee, 2014), the fermentation-pervaporation coupled system with cells immobilization provided the better pervaporation performance. Moreover, the suspended cells concentration at the end of the conventional suspended cells bioreactor and the cells immobilization bioreactor coupled system were 12.65 g/L and 8.32 g/L, respectively. That is, a large part of yeast cells was successfully immobilized on the SSB carrier. Since the suspended cells and the cell debris have a negative effect on the performance of pervaporation especially the ethanol permeability (Vane et al., 2010; Liu et al., 2011), the coupled system with the 8
SSB carrier might be of benefit to the ethanol separation by pervaporation, especially in long-terms of operation.
3.2 Fed-batch fermentation-pervaporation integration
Fig.2 shows the kinetics of fed-batch fermentation integrated with continuous pervaporation using SSB carrier within 190 h. As can be seen from Fig.2a, the fed-batch fermentation-pervaporation coupled system provided relatively stable ethanol concentration in the fermentation broth. Ethanol concentration was fluctuated between 82.4-108.2 g/L after 80 h of fermentation in the cells immobilized bioreactor. The suspended yeast cells concentration showed a clear increase in the first two cycles, then, it was maintained at 10.2-17.7 g/L. In comparison with the 17.1-19.8 g/L suspended cells obtained in the suspended cells fermentation-pervaporation process (Fan et al., 2014b), when applying the SSB as the carrier for immobilization, the suspended cells concentration was maintained at a low level, which helps to protect the pervaporation membrane from the contamination of cells (Liu et al., 2011; Vane et al., 2010). It should be note here that the suspended cells in the immobilization system have lower activity than the immobilized ones (Yu et al., 2007; 2010). As it was indicated in Liu et al. study (2011), the pervaporation membrane fouling only occurred in the live broth. Therefore, the membrane fouling caused by cells might be significantly limited by the inactive suspended cells in the cells immobilization system. Besides that, it showed that when integrated suspended cells fermentation processes with pervaporation, a dramatic decrease of yeast concentration was occurred in long terms of operation (Fan et al., 2014a; 2014b; Chen et al., 2012). It is likely attributed to the inhibition of several secondary metabolites to the cells metabolism. Since the toxic organic acids and glycerol cannot probably dissolve into the nonpolar PDMS layer of the pervaporation membrane, the accumulation of secondary metabolites was reported to inhibit the active cells in broth (Fan et al., 2014a). However, fortunately, the micro-environment provided by the SSB carrier could protect the yeast cells from the deteriorating culture (Yu et al., 2010). As a result, 9
the suspended cells concentration in the current work showed a general upward trend, though there was a considerably large fluctuation of the suspended cells concentration with time. Overall, at the end of fed-batch fermentation, a total 301.2 g/L of ethanol was produced from 682 g/L of glucose (Fig. 2b). The overall ethanol yield from glucose consumed was 0.44 g/g. As it is shown in Fig.2c, the ethanol productivity was gradually decreased with time until 23 h. After that, an obvious raise of ethanol productivity was occurred due to the integration of pervaporation. With the increase of ethanol concentration remained in the fermentation broth, the ethanol productivity was decreased with time and finally maintained at 1.6 g/L h level. A zigzag curve of glucose consumption rate with time was generated as well. It showed positive correlation property to the residual glucose concentration. Generally, the higher the glucose concentration in broth, the higher glucose consumption rate obtained. The relative low consumption rate under low glucose concentration might be caused by the lack of carbon source and nutritions. On the permeate side of pervaporation membrane, high ethanol concentration ranged from 426.9-597.2 g/L was obtained, suggesting that the ethanol product of fermentation was effectively separated and concentrated by the noval silicalite-1/PDMS/PVDF pervaporation membrane (Fig.3a). As it illustrated in Fig.3b, the ethanol flux was gradully increased in the first 80 h and then maintianed at 147.6-247.5 g/m2 h level. Therefore, the ethanol flux showed positive correlation to the ethanol titer in the fermentation broth. With the increase ethanol concentration on the feed side of pervaporation membrane, the free volume and the chain mobility of the silicalite-1/PDMS/PVDF layer were increased (Jee and Lee, 2014; Bettens et al., 2010). As for water flux, since water transport trough the pervaporation membrane is independent to the alcohol concentration in feed (Niemistö et al., 2013; Qin et al., 2014), the water flux was constant during the opration with values around 119.4-225.1 g/m2 h. It is also a great evidence to the long-term stability of both the novel pervaporation membrane and the integration process. In previous literatures, a drop of membrnae permeability was occurred when the actual fermentation broth 10
contaminated the surface of membrane (Vane et al., 2010). However, during the almost 200 h of operation in this work, the water flux was stable while ethanol flux showed co-relations to the retentate borth, which might be also appreciate to the cells immobilization. The total flux influenced by the water and ethanol flux showed similar trend with the ethanol flux. 319-416 g/m2 h of total flux was obtained in the stable stage of permeance. Additionally, the separation factor and PSI were waved from 8.2 and 2561.4 to 9.9 and 3759.2, respectively. And briefly, a stable integrtaion performance was obtained in the fed-batch fermentation-pervaporation integration process using SSB as immobilization carrier.
3.3 Continuous fermentation-pervaporation integration
Kinetics of long-term continuous fermentation-pervapration integrtaed system was further envaluated. Fig. 4 shows the time course profiles of ethanol production during the fermentation. The ethanol concentration in the continuous fermentation was gradually increased and then maintained at 90-110 g/L after 70 h of fermentation, which was higher than that of the fed-batch scenario (Fig.4a). Therefore, the metabolism of yeast in the continuous fermentation process was more vigorous than the strains in the fed-batch bioreactor. As the permeatability of the pervaporation membrane is the rate-determining step, ethanol in the feed side (bioreactor) was gradully cumulated and maintained a higher titer. At the same time, the productivity of strains was decreased due to the inhibition of the solvent product, while the ethanol separation rate was increased by the higher concentration in broth. Thus, a dynamic equilibrium between the ethanol production and separation was occurred, which was simialr with the phenomenon described in Xue et al. (2015) study. As for the time course of the free cells concentration in the fermentation broth, it was generally similar with the result of fed-batch one. The only different of the two scenarios is that the continuous fermentation porcess had a higher suspended cells concentration in the stable stage, suggesting that the metabolism of yeast increased in the continuous fermnetation process and further influenced the ethanol productivity. 11
Overall, after 190 h of fermentation, about 305.6 g/L of ethanol was produced from 719.7 g/L of glucose, while the ethanol productivity and yield were 1.61 g/L h and 0.42 g/g respectively, under a dilution rate of 0.016 /h (Fig.4b). The ethanol yield in the continuous fermentation system was a little shorter (0.03 g/g) than that of the fed-batch fermentation scenario, while the productivity in the continuous fermentation scenario was higher than that of the fed-batch one. It might be caused by the more vigorous bacteria metabolism in the continuous fermentation system. In that case, more carbon source was ultized as the substrate for the bio-synthesize of cell structure, which resulted in lower ethanol yield. As for the residual glucose in the fermentation broth, it was increased slightly with time, indicated that the envrionment for cells metabolism was more rigorous due to the acumulation of ethanol. At the end of fermentation, there were 54 g/L of glucose remained in the fermentation broth. The kinetics of ethanol productivity was first increased and then gradully decreased. In addition, the glucose consumption rate was dropped to 2.8 g/L h at 31 h, then fluctuated from 5.5 g/L h to7.5 g/L h until at the end of fermentation (Fig.4c). The drop of glucose consumption rate was likely due to the lack of carbon source at the beginning of continuous opreation. On the permeate side of membrane, 410.9-608.1 g/L of ethanol product was achieved after 62 h (Fig.5a), which was similar with the ethanol production obtained by the fed-batch fermentation-pervaporation integrated process. In consideration of the relatively higher ethanol concentration in the continuous bioreactor, the sepration factor of ethanol in the continuous scenario was little lower than that of the fed-batch scenario (see Fig.5c). This phenomenon could be explained by the contamination of the suspended cells. Higher suspended cells concentrtaion could clog the channel on the membrane surface. Fortunately, the higher concentrtaion of suspended cells showed little influence on the flux of membrane during the total 190 h of operation. Total flux and ethanol flux ranging from 227.9 g/m2 h to 395 g/m2 h and 86 g/m2 h to 208.2 g/m2 h were obtained, respectively. Similar with the phenomenon of the fed-batch fermentation, the water flux remained stable, ranging from 129.7 g/m2 h to 203.7 g/m2 h (Fig.5b). Additionally, similar with the trend of separation factor, the PSI 12
showed a zigzag curve, and maitained at a stable level of 2957.3±495.
3.4 Comparaion to other studies
Pervaporation is an attractive technique for in situ recovery of ethanol from fermentation broth (Vane, 2008; Fan et al., 2014a). However, one of the challenge for its industrail application is the ageing and fouling effect of membrane in long terms of operation (Vane et al., 2010). In the current work, SSB, the agriculture residual, was applied for cells immobilization in the fermentation-pervaporation separation processes. Although gels carrier have already been applied in the similar processes, to our best knowledge, it is the first time that using lignocellulosic material as carrier in the integration processes. The SSB carrier, with unique straucture on the surface, could hugely imporve the overall fermentation and pervapration performances. As can be seen from Table 1, compared with relative studies, the separation factor of ethanol in the current work is higher than others (on average). It demonstrated that higher feed concentration and larger separation factor correspond to lower ethanol recovery energy (Vane, 2005; 2008; Van Hecke, et al., 2012). Hence, the energy demand for ethanol recovery in the current work was lower than others. Besides that, the membrane performances including separation factor and the fluxes was stable in long-term integration. Therefore, the novel silicalite-1/PDMS/PVDF composite membrane could generally meet the demand of ethanol separation in long-term fermentation process. Previous works, however, showed a descent of membrane performances in the long-term of operations (Fan et al., 2014b). And the secondary metabolites showed more severe inhibition to the yeast cells in the suspended cells system (Fan et al., 2014a). By contrast, the current process with SSB as immobilized carrier showed long-term stability. More importantly, with little containmination of the fermentation broth, the pervaporation membrane might have a longer age for ethanol separation, then, further avioding the waste of memebrane and decreasing the membrane cost. 13
4. Conclusions
A system for ethanol fermentation was constracted by coupling the cells immobilization fermentation process with pervaporation unit using silicalite-1/PDMS/PVDF membrane. The SSB carrier led to a superior ethanol productivity and low concentrations of suspended cells, while the novel pervaporation membrane was effective in long-term operation. Ethanol sepration factor ranging from 8.2 to 9.9 and 7.8 to 9.8 were achieved under fed-batch and continuous fermentation, respectively. The integration processes provided high ethanol productivity and permenate ethanol concentration.
Acknowledgements
This work was supported in part by the National Basic Research Program of China (Grant No. 2013CB733600), the National Nature Science Foundation of China (Grant Nos. 21390202, 21476015), National High-Tech R&D Program of China (Grant Nos. 2014AA021904, 2014AA021903), the Fundamental Research Funds for the Central Universities (Grant No. YS1407), and the long-term subsidy mechanism from the Ministry of Finance and the Ministry of Education of PRC.
14
References
1. Bettens, B., Verhoef, A., van Veen, H.M., Vandecasteele, C., Degréve, J., Van der Bruggen, B., 2010. Pervaporation of binary water-alcohol and methanol-alcohol mixtures through microporous methylated silica membranes: Maxwell-Stefan modeling. Comput. Chem. Eng. 34, 1775-1788. 2. Cai, D., Chang, Z., Wang, C., Ren, W., Wang, Z., Qin, P., Tan, T., 2013. Impact of sweet sorghum cuticular waxes (SSCW) on acetone-butanol-ethanol fermentation using Clostridium acetobutylicum ABE 1201. Bioresour. Technol. 149, 470-473. 3. Chang, Z., Cai, D., Wang, C., Li, L., Han, J., Qin, P., Wang, Z., 2014. Sweet sorghum bagasse as an immobilized carrier for ABE fermentation by using Clostridium acetobutylicum ABE 1201. RSC Adv. 4, 21819-21825. 4. Chen, C., Tang, X., Xiao, Z., Zhou, Y., Jiang, Y., Fu, S., 2012. Ethanol fermentation kinetics in a continuous and closed-circulating fermentation system with a pervaporation membrane bioreactor. Bioresour. Technol. 114, 707-710. 5. Chen, J., Zhang, H., Wei, P., Zhang, L., Huang, H., 2014. Pervaporative behavior and integrated process for concentrating lignocellulosic ethanol through polydimethylsiloxane (PDMS) membrane. Bioprocess Biosyst. Eng. 37, 183-191. 6. Chovau, S., Gaykawad, S., Straathof, A.J.J., Van der Bruggen, B., 2011. Influence of fermentation by-products on the purification of ethanol from water using pervaporation. Bioresour. Technol. 102, 1669-1674. 7. Dagoe, J.T., Daugulis, A.J., 2014. In situ product removal in fermentation systems: improved process performance and rational extractant selection. Biotechnol. Lett. 36, 443-460. 8. Debreuil, M.F.S., Vandezande, P., Van Hecke, W.H.S., Porto-Carrero, W.J., Dotremont, C.T.E., 2013. Study on ageing/fouling phenomena and the effect of upstream nanofultration on in-situ product recovery of n-butanol through ploy[1-(trimethylsilyl)-1-propyne] pervaporation membranes. J. Membr. Sci. 447, 15
134-143. 9. Ding, W.W., Wu, Y.T., Tang, X.Y., Yuan, L., Xiao, Z.Y., 2011. Continuous ethanol fermentation in a closed-circulation system using an immobilized cell coupled with PDMS membrane pervaporation. J. Chem. Technol. Biotechnol. 86, 92-87. 10. Ding, W., Xiao, Z., Tang, X., Deng, K., Fu, S., Jiang, Y., Yuan, L., 2012. Evolutionary engineering of yeast for closed-circulating ethanol fermentation in PDMS membrane bioreactor. Biochem. Eng. J., 60, 56-61. 11. Fan, S., Xiao, Z., Tang, X., Chen, C., Zhang, C., Zhang, Y., Deng, Q., Yao, P., Li, W., 2014a. Inhibition effect of secondary metabolites accumulated in a pervaporation membrane bioreactor on ethanol fermentation of Saccharomyces cerevisiae. Bioresour. Thecnol. 162, 8-13. 12. Fan, S., Xiao, Z., Zhang, Y., Tang, X., Chen, C., Li, W., Deng, Q., Yao, P., 2014b. Enhance ethanol fermentation in a pervaporation membrane bioreactor with the convenient permeate vapor recovery. Bioresour. Technol. 155, 229-234. 13. Fan, S., Xiao, Z., Li, M., 2016. Energy efficient of ethanol recovery in pervaporation membrane bioreactor with mechanical vapor compression eliminating the cold traps. Bioresour. Technol. 211, 24-30. 14. Fu, C., Cai, D., Hu, S., Miao, Q., Wang, Y., Qin, P., Wang, Z., Tan, T., 2016. Ethanol fermentation integrated with PDMS composite membrane: An effective process. Bioresour. Technol. 200, 648-657. 15. Hartono, S.B., Qiao, S.Z., Liu, J., Jack, K., Ladewig, B.P., Hao, Z., Lu, G.Q.M., 2010. Functinaolized mesoporous silica with very large pores for cellulase immobilization. J. Phys. Chem. C. 114, 8353-8362. 16. Jee, K.Y., Lee, Y.T., 2014. Preparation and characterization of siloxane composite membranes for n-butanol concentration from ABE solution by pervaporation. J. Membr. Sci. 456, 1-10. 17. Jin, C., Yao, M., Liu, H., Lee, C.F., Ji, J., 2011. Progress in the production and application of n-butanol as a biofuel. Renew. Sustain. Energ. Rev. 15, 4080-4106. 18. Kaewkannetra, P., Chutinate, N., Moonamart, S., Kamsan, T., Chiu, T.Y., 2012. 16
Experimental study and cost evaluation for ethanol separation from fermentation broth using pervaporation. Desalin. Water Treat. 41, 88-94. 19. Kumar, M., Gayen, K., 2011. Developments in biobutanol production: New insights. Appl. Energ. 88, 1999-2012. 20. Li, S., Li, G., Zhang, L., Zhou, Z., Han, B., Hou, W., Wang, J., Li, T., 2013. A demonstration study of ethanol production from sweet sorghum stems with advanced solid state fermentation technology. Appl. Energ. 102, 260-265. 21. Liu, G., Wei, W., Wu, H., Dong, X., Jiang, M., Jin, W., 2011. Pervaporation performance of PDMS/ceramic composite membrane in acetone butanol ethanol (ABE) fermentation-PV coupled process. J. Membr. Sci. 373, 121-129. 22. Maryse, G., Dravko, D., 1996. Wood blocks as a carrier for Saccharomyces cerevisiae used in the production of ethanol and fructose. Chem. Eng. J. 61, 233-240. 23. Mussatto, S.I., Dragone, G., Guimarᾶes, P.M.R., Silva, J.P.A., Carneiro, L.M., Roberto, I.C., Vicente, A., Domingues, L., Teixeira, J.A., 2010. Technological trends, global market, and challenges of bio-ethanol production. Biotechnol. Adv. 28, 817-830. 24. Shabtai, Y., Chaimovitz, S., Freeman, A., Katchalski-Katzir, E., Linder, C., Nemas, M., Perry, M., Kedem, O., 1991. Continuous ethanol production by immobilized yeast reactor coupled with membrane pervaporation unit. Biotechnol. Bioeng. 38, 869-876. 25. Shao, P., Huang, R.Y.M., 2007. Polymeric membrane pervaporation. J. Membr. Sci. 287, 162-169. 26. Wei, P., Cheng, L.H., Zhang, L., Xu, X.H., Chen, H.L., Gao, C.J., 2014. A review of membrane technology for bioethanol production. Renew. Sustain. Energ. Rev. 30, 388-400. 27. Van Hecke, W., Vandezande, P., Claes, S., Vangeel, S., Beckers, H., Diels, L., De Wever, H., 2012. Integrated bioprocess for long-term continuous cultivation of Clostridium acetobutylicum coupled to pervaporation with PDMS composite membranes. Bioresour. Technol. 111, 368-377. 17
28. Vane, L.M., 2005. A review of pervaporation for product recovery from biomass fermentation processes. J. Chem. Technol. Biotechnol. 80, 603-629. 29. Vane, L.M., 2008. Separation technologies for the recovery and dehydration of alcohols from fermentation broth. Biofuels Bioprod. Bioref. 2, 553-588. 30. Vane, L.M., Namboodiri, V.V., Meier, R.G., 2010. Factors affecting alcohol-water pervaporation performance of hydrophobic zeolite-silicone rubber mixed matrix membranes. J. Membr. Sci. 364, 102-110. 31. Xue, F., Gao, B., Zhu, Y., Zhang, X., Feng, W., Tan, T., 2010. Pilot-scale production of microbial lipid using starch wastewater as raw material. Bioresour. Technol. 101, 6092-6095. 32. Xue, C., Yang, D., Du, G., Chen, L., Ren, J., Bai, F., 2015. Evaluation of hydrophobic micro-zeolite-mixed matrix membrane and integrated with acetone-butanol-ethanol fermentation for enhanced butanol production. Biotechnol. Biofuels 8, 105. 33. Yu, J., Yue, G., Zhong, J., Zhang, X., Tan, T., 2010. Immobilization of Saccharomyces cerevisiae to modified bagasse for ethanol production. Renew. Energ. 35, 1130-1134. 34. Yu, J., Zhang, T., Zhong, J., Zhang, X., Tan, T., 2012. Biorefinery of sweet sorghum stem. Biotechnol. Adv. 30, 811-816. 35. Yu, J., Zhang, X., Tan, T., 2007. An novel immobilization method of Saccharomyces cerevisiae to sorghum bagasse for ethanol production. J. Biotechnol.129, 415-420.
18
Figure captions
Fig.1 The comparison of the batch fermentation kinetics between immobilized cells and free cells system integrated with continuous pervaporation. Fig.2 Fermentation kinetics of fed-batch ethanol fermentation with continuous pervaporation for ethanol recovery.(a) Cumulated ethanol production and residual glucose concentration remained in the fermentation broth; (b) Ethanol concentration in the fermentation broth and the yeast cell concentration; (c) Ethanol productivity and glucose consumption rate in 6 feeding cycles Fig.3 Membrane performance during the fed-batch ethanol fermentation with continuous pervaporation.(a) Ethanol titer on the permeate side obtained in 6 feeding cycles; (b) Ethanol, water and total flux; (c) The separation factor of ethanol and the overall PSI. Fig.4 Fermentation kinetics of continuous ethanol fermentation with continuous pervaporation for ethanol recovery, the dilution rate of fermentation was 0.016/h. (a) Cumulated ethanol production and residual glucose concentration remained in the fermentation broth; (b) Ethanol concentration in the fermentation broth and the yeast cell concentration; (c) Ethanol productivity and glucose consumption rate in 190 h. Fig.5 Membrane performances during of the continuous ethanol fermentation with continuous pervaporation, the dilution rate of fermentation was 0.016/h. (a) Ethanol titer on the permeate side obtained in 190 h; (b) Ethanol, water and total flux; (c) The separation factor of ethanol and the overall PSI.
19
Fig.1
20
Fig.2
21
Fig.3
22
Fig.4
23
Fig.5
24
Table 1 Study
Current advance in ethanol fermentation integrated with pervaporation under different fermentation types. Chen et
Fu et al.,
Fu et al.,
Ding et al.,
Chen et al.,
Fan et al.,
Shabtai et
Ding et al.,
al., 2014
2016
2016
2012
2012
2014b
al., 1991
2011
Fermentation
Fed-batch
Fed-batch
Continuous
Continuous
Continuous
Continuous
Continuous
Operation
Suspended
Suspended
Suspended
Suspended
Suspended
Suspended
cells
cells
cells
cells
cells
PDMS
PDMS/PVDF
PDMS/PVDF
PDMS
PDMS
Membrane
This study
This study
Continuous
Fed-batch
Continuous
Immobilized
Immobilized
Immobilized cells
Immobilized cells
cells
cells
cells
PDMS
PS/PSF
PDMS/PA
silicalite-1/PDMS
silicalite-1/PDMS
/PVDF
/PVDF
type Total flux
1300-1400 396.2-663.7
332.4-548.1
-
774
370
1750
300–690
319-416
227.8-395
4.1-5 a
8.6-11.7
8-11
-
4.85
9.5
5b
5.0–7.2 b
8.2-9.9
7.8-9.8
-
70-80
~100
47.9-54.2
45.3-48.8
46
30 (g/g)
48 (g/g)
82.4-108.2
90-110
25.8-30.2
417.2
~446.3
-
216.2-221.1
320
17–20 (g/g)
190-310
426.9-597.2
410.9-608.1
(g/m2 h) Separation factor Ethanol conc. in broth (g/L) Ethanol conc. in
(g/g)
25
permeate (g/L) yield (g/g)
-
0.42
0.42
0.42-0.43
0.37
0.38
-
0.41
0.44
0.42
Productivity
13.4
1.58
1.78
1.5-1.6
1.51
2.31
30
9.6
~1.6
~1.6
(g/L h) a
Estimated by authors.
b
Selectivity.
1. Sweet sorghum bagasse was used as the immobilized carrier. 2. Silicalite-1/PDMS/PVDF membrane was used for in situ ethanol separation. 3. 410.9-608.1 g/L of ethanol was obtained in the integration process. 4. Approximate 10 of ethanol separation factor was obtained in the integration process.
26