《可再生能源》教学资源（阅读资料）02-Impacts of main factors on bioethanol fermentation

Available online at www.sciencedirect.com ScienceDirect BIORESOURCE TECHn○LOGY ELSEVIER Bioresource Technology99(2008)847-854 Impacts of main factors on bioethanol fermentation from stalk juice of sweet sorghum by immobilized Saccharomyces cerevisiae (CICC 1308) Ronghou Liu *Fei Shen Biomass Energy Engineering Research Centre.School of Agriculture and Biology.Shanghai JiaoTong University.2678 Qi Xin Road. Shanghai 201101,PR China Received 17 October 2006;received in revised form 17 January 2007;accepted 17 January 2007 Available online 13 March 2007 Abstract In order to attain a higher ethanol yield and faster ethanol fermentation rate,orthogonal experiments of ethanol fermentation with immobilized yeast from stalk juice of sweet sorghum were carried out in the shaking flasks to investigate the effect of main factors namely,fermentation temperature,agitation rate,particles stuffing rate and pH on ethanol yield and COz weight loss rate.The range analysis and analysis of variance(ANOVA)were applied for the results of orthogonal experiments.Results showed that the optimal con- dition for bioethanol fermentation should be A B,C,D,namely,fermentation temperature,agitation rate,particles stuffing rate and pH were 37C,200rpm,25%and 5.0,respectively.The verification experiments were carried out in shaking flasks and 5L bioreactor at the corresponding parameters.The results of verification experiments in the shaking flasks showed that ethanol yield and CO,weight loss rate were 98.07%and 1.020gh,respectively.The results of ethanol fermentation in the 5L bioreactor showed that ethanol yield and fer- mentation time were 93.24%and 11 h,respectively.As a result,it could be concluded that the determined optimal condition A BC3D was suitable and reasonable for the ethanol fermentation by immobilized Saccharomyces cerevisiae.The conclusion in the research would be beneficial for application of ethanol fermentation by immobilized S.cerevisiae from stalk juice of sweet sorghum. 2007 Elsevier Ltd.All rights reserved. Keywords:Suitable parameters;Sweet sorghum;Immobilized Saccharomyces cerevisiae;Bioethanol fermentation 1.Introduction malt or molds.Cellulose (from wood,agricultural residues, waste sulfite liquor from pulp,and paper mills)must like- Ethanol production as an alternative fuel energy wise be converted into sugars,generally by the action of resource has been a subject of great interest since the oil cri- mineral acids.Once simple sugars are formed,enzymes sis of the 1970s (Tao et al,2005).Therefore,a strong need from microorganisms can readily ferment them to ethanol exists for efficient ethanol production with low cost in raw (Lin and Tanaka,2006).As for materials,one of the prime material and production process.The varied raw materials sources being investigated for ethanol is sweet sorghum. used in the production of ethanol via fermentation are con- Sweet sorghum {Sorghim biocolor (L.)Moench is a high veniently classified into three main types of raw materials: biomass-and sugar-yielding crops (Bryan,1990),mean- sugars,starches,and cellulose materials.Sugars(from sug- time,the stalk of sweet sorghum contains quite a few quan- arcane,sugar beets,sweet sorghum.molasses,and fruits) tities of soluble (glucose and sucrose)and insoluble can be converted into ethanol directly.Starches(from corn, carbohydrates (cellulose and hemicellulose)(Jasberg et al., cassava,potatoes,and root crops)must firstly be hydro- 1983).Therefore,of many crops currently being investi- lyzed to fermentable sugars by the action of enzymes from gated for energy and industry,sweet sorghum is one of the most promising,particularly for ethanol production (Gnansounou et al.,2005). Corresponding author.Tel:+862164783844:fax:+8621 64193285. The advantages of immobilized cells over free cell sys- E-mail address:liurhou@sjtu.edu.cn (R.Liu). tems have been extensively reported (Plessas et al.,2007). 0960-8524/S-see front matter 2007 Elsevier Ltd.All rights reserved. doi:10.1016f.biortech.2007.01.009

Bioresource Technology 99 (2008) 847–854 0960-8524/$ - see front matter © 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.biortech.2007.01.009 Available online at www.sciencedirect.com Impacts of main factors on bioethanol fermentation from stalk juice of sweet sorghum by immobilized Saccharomyces cerevisiae (CICC 1308) Ronghou Liu ¤ , Fei Shen Biomass Energy Engineering Research Centre, School of Agriculture and Biology, Shanghai JiaoTong University, 2678 Qi Xin Road, Shanghai 201101, PR China Received 17 October 2006; received in revised form 17 January 2007; accepted 17 January 2007 Available online 13 March 2007 Abstract In order to attain a higher ethanol yield and faster ethanol fermentation rate, orthogonal experiments of ethanol fermentation with immobilized yeast from stalk juice of sweet sorghum were carried out in the shaking Xasks to investigate the eVect of main factors, namely, fermentation temperature, agitation rate, particles stuYng rate and pH on ethanol yield and CO2 weight loss rate. The range analysis and analysis of variance (ANOVA) were applied for the results of orthogonal experiments. Results showed that the optimal con￾dition for bioethanol fermentation should be A4B3C3D4, namely, fermentation temperature, agitation rate, particles stuYng rate and pH were 37 °C, 200 rpm, 25% and 5.0, respectively. The veriWcation experiments were carried out in shaking Xasks and 5 L bioreactor at the corresponding parameters. The results of veriWcation experiments in the shaking Xasks showed that ethanol yield and CO2 weight loss rate were 98.07% and 1.020 g h¡1 , respectively. The results of ethanol fermentation in the 5 L bioreactor showed that ethanol yield and fer￾mentation time were 93.24% and 11 h, respectively. As a result, it could be concluded that the determined optimal condition A4B3C3D4 was suitable and reasonable for the ethanol fermentation by immobilized Saccharomyces cerevisiae. The conclusion in the research would be beneWcial for application of ethanol fermentation by immobilized S. cerevisiae from stalk juice of sweet sorghum. © 2007 Elsevier Ltd. All rights reserved. Keywords: Suitable parameters; Sweet sorghum; Immobilized Saccharomyces cerevisiae; Bioethanol fermentation 1. Introduction Ethanol production as an alternative fuel energy resource has been a subject of great interest since the oil cri￾sis of the 1970s (Tao et al., 2005). Therefore, a strong need exists for eYcient ethanol production with low cost in raw material and production process. The varied raw materials used in the production of ethanol via fermentation are con￾veniently classiWed into three main types of raw materials: sugars, starches, and cellulose materials. Sugars (from sug￾arcane, sugar beets, sweet sorghum, molasses, and fruits) can be converted into ethanol directly. Starches (from corn, cassava, potatoes, and root crops) must Wrstly be hydro￾lyzed to fermentable sugars by the action of enzymes from malt or molds. Cellulose (from wood, agricultural residues, waste sulWte liquor from pulp, and paper mills) must like￾wise be converted into sugars, generally by the action of mineral acids. Once simple sugars are formed, enzymes from microorganisms can readily ferment them to ethanol (Lin and Tanaka, 2006). As for materials, one of the prime sources being investigated for ethanol is sweet sorghum. Sweet sorghum {Sorghum biocolor (L.) Moench} is a high biomass- and sugar-yielding crops (Bryan, 1990), mean￾time, the stalk of sweet sorghum contains quite a few quan￾tities of soluble (glucose and sucrose) and insoluble carbohydrates (cellulose and hemicellulose) (Jasberg et al., 1983). Therefore, of many crops currently being investi￾gated for energy and industry, sweet sorghum is one of the most promising, particularly for ethanol production (Gnansounou et al., 2005). The advantages of immobilized cells over free cell sys￾tems have been extensively reported (Plessas et al., 2007). * Corresponding author. Tel.: +86 21 64783844; fax: +86 21 64193285. E-mail address: liurhou@sjtu.edu.cn (R. Liu)

848 R.Liu,F.Shen Bioresource Technology 99 (2008)847-854 Cell immobilization can be more effective because cell Table 1 washout in continuous operation is prevented,and,hence. The composition of culture media cell separation and/or recycle are not required for main- Composition of Slant culture Liquid medium Fermentation taining high cell density in the bioreactor;thus,the biopro- culture media medium (g(100mL)- medium cesses can be operated more efficiently (Tzeng et al.,1991). (g(100mL)- (g(100mL)-) Many researches concerned with immobilized cells have Glucose or sucrose 5.0 5.0 10.0 been carried out throughout the world.Particularly,there is Yeast extract 0.5 0.5 0.5 Peptone 0.5 0.5 0.5 an increasing interest in the practical applications of immo- K,HPO 0.1 0.1 0.1 bilized cells in ethanol production(Kobayashi and Nakam- MgSO4·7H,O 0.1 0.1 0.1 ura,2004),and considerable researches have been Agar 2.0 performed over the last 20 years into the use of immobi- lized cell systems for the production of fuel and potable- grade ethanol (Bardi et al..1996). the yeast cells were up to 108 cellsmL-observed by micro- There are many factors that have influence upon the eth- scope,the yeast cells in liquid culture medium were mature; anol yield and fermentation rate in fermentation process, Fermentation culture was carried in a 500mL flask with a such as fermentation temperature,agitation rate,pH and working volume of 200-250 mL.The flask was inoculated particles stuffing rate that is defined as a ratio of immobi- with 20-25 mL mature yeast cells of liquid culture medium lized yeast particles weight to fermentation solution weight. and incubated at 30C and 150 rpm for about 48h.When The immobilization process changes the environmental, the yeast cells were up to 10 cellsmL-,the yeast cells in physiological and morphological characteristics of cells, fermentation culture medium were mature,and could be along with the catalytic activity (Prasad and Ishra,1995). used for the yeast cells immobilization. As a result,the fermentation conditions are different between free yeast and immobilized yeast fermentation. 2.3.Yeast cells immobilization and proliferation The aim of the current work was to investigate the effect of main variables (fermentation temperature,agitation rate, Na-alginate (A.R)powder was dissolved in distilled particles stuffing rate and pH)on ethanol yield and CO2 water in a 75-80C water bath for 30min at the ratio of weight loss rate and to determine optimal condition for eth- 2.5-3.0:100.The mature yeast cells of fermentation culture anol fermentation by immobilized yeast from stalk juice of medium were mixed with the Na-alginate solution at the sweet sorghum. ratio of 1:10.The mixture was injected into 0.05 M calcium chloride(A.R)solution to form Ca-alginate particles by the 2.Methods equipment that was designed by us.The produced Ca-algi- nate particles were of near-spherical shape about 3 mm in 2.1.Sweet sorghum and organisms diameter.After 12h gelation,the immobilized yeast parti- cles were stored at 4C. Sweet sorghum cultivar SN Tianza No.2 was cultivated Immobilized yeast particles were added into the fermen- in the farm of Shanghai Jiao Tong University.The fresh tation culture medium at the ratio of 1:10(W/W)in 500mL crops were harvested and leaves were stripped from the flasks.The flasks were incubated at 30C and 150rpm for fresh stalks by hand before the stalks were squeezed by a 48 h.To determine the quantity of yeast in particles,the three-roller mill to obtain fresh juice.The fresh juice was immobilized yeast particles were dissolved in 0.1 M PBS sterilized in autoclave sterilizer at 0.15 MPa and 121C for solution and then observed by microscope.When the quan- 15min and stored at 4C in a refrigerator before it was tity of yeast cell in immobilized yeast particles was up to used for fermentation. 108 cellsmL-1,the immobilized yeast could be used for fer- Laboratory strain of Saccharomyces cerevisiae CICC mentation. 1308 (obtained from Centre of Industrial Culture Collec- tion of China)was used.The strain was stored on slant cul- 2.4.Equipment ture medium at 4C in a refrigerator. The equipment used in this study was a 5L bioreactor 2.2.Culture media and microorganism culture (GUJS-5C,East Biotech.Co.Zhenjiang China).The sche- matic diagram of the 5L bioreactor is shown in Fig.1. The composition of culture media is shown in Table 1. The activated Saccharomyces cerevisiae CICC 1308 was 2.5.Experiment design inoculated into flesh slant culture medium in a test tube and cultivated in a thermotank at 30C for 1-3d until the white The L16(45)orthogonal table was designed to investigate lawn appeared on the slant culture and distributed evenly; the influence of main parameters,namely the fermentation Precultures were incubated for 20-24h at 30+0.5C with temperature,agitation rate,particles stuffing rate and pH orbital shaking at 150 rpm in a 25 mL-test tube containing a on bioethanol fermentation.According to the literature loopful of yeast cells in 10mL liquid culture medium.When (Zheng and Jiang,2004)and our practical experience of

848 R. Liu, F. Shen / Bioresource Technology 99 (2008) 847–854 Cell immobilization can be more eVective because cell washout in continuous operation is prevented, and, hence, cell separation and/or recycle are not required for main￾taining high cell density in the bioreactor; thus, the biopro￾cesses can be operated more eYciently (Tzeng et al., 1991). Many researches concerned with immobilized cells have been carried out throughout the world. Particularly, there is an increasing interest in the practical applications of immo￾bilized cells in ethanol production (Kobayashi and Nakam￾ura, 2004), and considerable researches have been performed over the last 20 years into the use of immobi￾lized cell systems for the production of fuel and potable￾grade ethanol (Bardi et al., 1996). There are many factors that have inXuence upon the eth￾anol yield and fermentation rate in fermentation process, such as fermentation temperature, agitation rate, pH and particles stuYng rate that is deWned as a ratio of immobi￾lized yeast particles weight to fermentation solution weight. The immobilization process changes the environmental, physiological and morphological characteristics of cells, along with the catalytic activity (Prasad and Ishra, 1995). As a result, the fermentation conditions are diVerent between free yeast and immobilized yeast fermentation. The aim of the current work was to investigate the eVect of main variables (fermentation temperature, agitation rate, particles stuYng rate and pH) on ethanol yield and CO2 weight loss rate and to determine optimal condition for eth￾anol fermentation by immobilized yeast from stalk juice of sweet sorghum. 2. Methods 2.1. Sweet sorghum and organisms Sweet sorghum cultivar SN Tianza No. 2 was cultivated in the farm of Shanghai Jiao Tong University. The fresh crops were harvested and leaves were stripped from the fresh stalks by hand before the stalks were squeezed by a three-roller mill to obtain fresh juice. The fresh juice was sterilized in autoclave sterilizer at 0.15 MPa and 121 °C for 15 min and stored at 4 °C in a refrigerator before it was used for fermentation. Laboratory strain of Saccharomyces cerevisiae CICC 1308 (obtained from Centre of Industrial Culture Collec￾tion of China) was used. The strain was stored on slant cul￾ture medium at 4 °C in a refrigerator. 2.2. Culture media and microorganism culture The composition of culture media is shown in Table 1. The activated Saccharomyces cerevisiae CICC 1308 was inoculated into Xesh slant culture medium in a test tube and cultivated in a thermotank at 30 °C for 1–3 d until the white lawn appeared on the slant culture and distributed evenly; Precultures were incubated for 20–24 h at 30§ 0.5 °C with orbital shaking at 150 rpm in a 25 mL-test tube containing a loopful of yeast cells in 10 mL liquid culture medium. When the yeast cells were up to 108 cells mL¡1 observed by micro￾scope, the yeast cells in liquid culture medium were mature; Fermentation culture was carried in a 500 mL Xask with a working volume of 200–250 mL. The Xask was inoculated with 20–25 mL mature yeast cells of liquid culture medium and incubated at 30 °C and 150 rpm for about 48 h. When the yeast cells were up to 108 cells mL¡1 , the yeast cells in fermentation culture medium were mature, and could be used for the yeast cells immobilization. 2.3. Yeast cells immobilization and proliferation Na-alginate (A.R) powder was dissolved in distilled water in a 75–80 °C water bath for 30 min at the ratio of 2.5–3.0:100. The mature yeast cells of fermentation culture medium were mixed with the Na-alginate solution at the ratio of 1:10. The mixture was injected into 0.05 M calcium chloride (A.R) solution to form Ca-alginate particles by the equipment that was designed by us. The produced Ca-algi￾nate particles were of near-spherical shape about 3 mm in diameter. After 12 h gelation, the immobilized yeast parti￾cles were stored at 4 °C. Immobilized yeast particles were added into the fermen￾tation culture medium at the ratio of 1:10 (W/W) in 500 mL Xasks. The Xasks were incubated at 30 °C and 150 rpm for 48 h. To determine the quantity of yeast in particles, the immobilized yeast particles were dissolved in 0.1 M PBS solution and then observed by microscope. When the quan￾tity of yeast cell in immobilized yeast particles was up to 108 cells mL¡1 , the immobilized yeast could be used for fer￾mentation. 2.4. Equipment The equipment used in this study was a 5 L bioreactor (GUJS-5C, East Biotech. Co. Zhenjiang China). The sche￾matic diagram of the 5 L bioreactor is shown in Fig. 1. 2.5. Experiment design The L16(45 ) orthogonal table was designed to investigate the inXuence of main parameters, namely the fermentation temperature, agitation rate, particles stuYng rate and pH on bioethanol fermentation. According to the literature (Zheng and Jiang, 2004) and our practical experience of Table 1 The composition of culture media Composition of culture media Slant culture medium (g (100 mL)¡1 ) Liquid medium (g (100 mL)¡1 ) Fermentation medium (g (100 mL)¡1 ) Glucose or sucrose 5.0 5.0 10.0 Yeast extract 0.5 0.5 0.5 Peptone 0.5 0.5 0.5 K2HPO4 0.1 0.1 0.1 MgSO4 · 7H2O 0.1 0.1 0.1 Agar 2.0 – –

R.Liu.F.Shen Bioresource Technology 99 (2008)847-854 849 Air filter Vapour filter According to the optimal fermentation condition obtained from the analysis results of orthogonal test.the verification experiments were performed in the shaking flasks and 5L bioreactor V6 2.7.Analysis Exhaust gas 来v3XV2v1 The sugar content (sucrose,glucose and fructose)of stalk juice was determined by HPLC with Evaporative Light Scattering Detector (Alltech 2000 ELSD). Vapour in Air in 100mgmL-I of sucrose (AR),glucose (AR)and fructose (AR)was respectively used as the standard solution for the standard curve.Chromatogram column was Waters uBondapak TM(3.9 x 300mm),and the column tempera- Heater ture was controlled at 25C.Software of the system was 10 Enpower offered by Waters Co.The mobile phase was the ode V12 acetonitrile and water (no ions)with the ratio of 75:25(V/ V11 Samping Exhausting V),and the flow rate was ImLmin-.The samples were centrifuged at 6000rpm and 4C for 10min,and supernate was filtered by 0.45 um filter membrane (Waters Sep-Park- Cis)before they were analyzed by HPLC.Samples of 20uL Cooling water were injected in the column by an auto-sample injection. Vapour out All the samples were performed in duplicate and the sugar concentrations were determined according to the standard Fig.1.The schematic diagram of the 5 L bioreactor curves of each sugar. ethanol production,the interaction between these four fac- Ethanol was determined by GC with a FID detector. tors had a little influence on the results and was ignored in The internal standard method was applied.Chromatogram the study.The L16(45)orthogonal test is shown in Table 2. column and samples injection were stainless steel column The blank factor is a dummy here and is used for error esti- and auto headspace samples injection.Ethanol (GR)and mation normal propyl alcohol(GR)were used for the standard curve and internal standard substance,respectively.The 2.6.Fermentation column temperature was controlled at 200C.N,was used as a carrier gas(40mL min).The flow rates of H,and O, (NH)SO of 0.2%,K2HPO of 0.125%and MgSO of were 40mL min and 500mLmin,respectively.Samples 0.05%were added into the sterilized stalk juice before the of 0.3uL were injected directly into the column.All deter- fermentations were carried out(Liu et al..2005) minations were done by means of standard curves,and results were the mean of two repetitions. Table 2 CO2 weight loss in fermentation was measured by an The L16(4)orthogonal test accurate balance (0.01 g).The total weight of the fermenta- Experiment Temperature Agitation Blank Particles pH tion solution and the shaking flask was determined in every (C) rate(rpm) stuffing hour rate (% Nos. A B C D 3.Results and discussion 100 2 150 0 3.1.The orthogonal experiment results ofethanol 3 3421 53050 fermentation 45 5.0 Batch fermentations in the shaking flasks for ethanol 6789 12345 382862868111144444444773777 功050505050 45 43 4 production were carried out in triplicate.Both the ethanol yield and the CO,weight loss rate were considered as opti- 341 5005050503 mization objective of bioethanol fermentation.The sugar 50 concentration of the juice was changed with different har- vesting time and juice storage time.Therefore,the ethanol 243 yield(%),which is defined as a percentage of the actual eth- anol output to the theoretic ethanol output,was introduced 16 208 3.5 in this study to judge the ethanol production in the fermen- 4.0 tation.The batch ethanol fermentation rate can be directly

R. Liu, F. Shen / Bioresource Technology 99 (2008) 847–854 849 ethanol production, the interaction between these four fac￾tors had a little inXuence on the results and was ignored in the study. The L16(45 ) orthogonal test is shown in Table 2. The blank factor is a dummy here and is used for error esti￾mation. 2.6. Fermentation (NH4)2SO4 of 0.2%, K2HPO4 of 0.125% and MgSO4 of 0.05% were added into the sterilized stalk juice before the fermentations were carried out (Liu et al., 2005). According to the optimal fermentation condition obtained from the analysis results of orthogonal test, the veriWcation experiments were performed in the shaking Xasks and 5 L bioreactor. 2.7. Analysis The sugar content (sucrose, glucose and fructose) of stalk juice was determined by HPLC with Evaporative Light Scattering Detector (Alltech 2000 ELSD). 100 mg mL¡1 of sucrose (AR), glucose (AR) and fructose (AR) was respectively used as the standard solution for the standard curve. Chromatogram column was Waters Bondapak TM (3.9 £300 mm), and the column tempera￾ture was controlled at 25 °C. Software of the system was Enpower oVered by Waters Co. The mobile phase was the acetonitrile and water (no ions) with the ratio of 75:25 (V/ V), and the Xow rate was 1 mL min¡1 . The samples were centrifuged at 6000 rpm and 4 °C for 10 min, and supernate was Wltered by 0.45 m Wlter membrane (Waters Sep-Park￾C18) before they were analyzed by HPLC. Samples of 20 L were injected in the column by an auto-sample injection. All the samples were performed in duplicate and the sugar concentrations were determined according to the standard curves of each sugar. Ethanol was determined by GC with a FID detector. The internal standard method was applied. Chromatogram column and samples injection were stainless steel column and auto headspace samples injection. Ethanol (GR) and normal propyl alcohol (GR) were used for the standard curve and internal standard substance, respectively. The column temperature was controlled at 200 °C. N2 was used as a carrier gas (40 mL min¡1 ). The Xow rates of H2 and O2 were 40 mL min¡1 and 500 mL min¡1 , respectively. Samples of 0.3 L were injected directly into the column. All deter￾minations were done by means of standard curves, and results were the mean of two repetitions. CO2 weight loss in fermentation was measured by an accurate balance (0.01 g). The total weight of the fermenta￾tion solution and the shaking Xask was determined in every hour. 3. Results and discussion 3.1. The orthogonal experiment results of ethanol fermentation Batch fermentations in the shaking Xasks for ethanol production were carried out in triplicate. Both the ethanol yield and the CO2 weight loss rate were considered as opti￾mization objective of bioethanol fermentation. The sugar concentration of the juice was changed with diVerent har￾vesting time and juice storage time. Therefore, the ethanol yield (%), which is deWned as a percentage of the actual eth￾anol output to the theoretic ethanol output, was introduced in this study to judge the ethanol production in the fermen￾tation. The batch ethanol fermentation rate can be directly Fig. 1. The schematic diagram of the 5 L bioreactor. Air filter Vapour filter Vapour in Air in Exhaust gas Electromagnetic valve water in Vapour in Cooling Vapour in Samping & Exhausting Cooling water & Vapour out Fermentor Heater Motor V3 V2 V1 V4 V5 V6 V7 V8 V9 V10 V11 V12 Table 2 The L16(45 ) orthogonal test Experiment Temperature (°C) Agitation rate (rpm) Blank Particles stuYng rate (%) pH Nos. A B CD 1 28 100 1 15 3.5 2 28 150 2 20 4.0 3 28 200 3 25 4.5 4 28 250 4 30 5.0 5 31 100 2 25 5.0 6 31 150 1 30 4.5 7 31 200 4 15 4.0 8 31 250 3 20 3.5 9 34 100 3 30 4.0 10 34 150 4 25 3.5 11 34 200 1 20 5.0 12 34 250 2 15 4.5 13 37 100 4 20 4.5 14 37 150 3 15 5.0 15 37 200 2 30 3.5 16 37 250 1 25 4.0

850 R.Liu,F.Shen Bioresource Technology 99 (2008)847-854 Table 3 For comparison of the significance of the above four Orthogonal experiment results of ethanol yield and CO,weight loss rate variables in influencing the increased ethanol yield. Experiment nos. Ethanol yield (% CO,weight loss rate (gh-) the analysis of variance (ANOVA)method was used.F 70.08±0.745e 0.373±0.0058k values of four factors were as follows:Ftemperature=152.28, 66.66±2.607e 0.463±0.0058j Fagitation rate=17.46,Fparticles stulling rate=3144 and FpH= 82.54±2.143cd 0.485±0.0046i 44.80,which were all greater than the critical F value=4.40 82.90±3.762cd 0.484±0.0040i 84.98±1.519bcd 0.488±0.0040i (ppH>particles stuffing rate>agitation rate. 90 -Temperature -Agitation rate The optimal condition for improving ethanol yield was -Particles stuffing rate 88 determined as AB:C:Da.Thus the optimal fermentation condition was determined as follows:temperature 37.C. 86- agitation rate 200 rpm,particles stuffing rate 25%,pH 5.0. 84- Table4 82 The range analysis of L(45)orthogonal experiments of ethanol yield Temperature Agitation Blank Particles pH 80 rate stuffing rate 78- A B C D K 909.55 978.24 983.51 968.82 1020.07 76 1014.88 991.81 982.44 987.28 948.03 1016.07 1028.81 1027.23 1034.23 1017.93 74十 1078.73 1020.37 1026.05 1028.90 1029.20 30 35 4.0 4.5 5.0 55 75.79 81.52 81.95 80.73 85.01 25 28 31 37 0 84.57 Temperature (C) 82.65 81.87 82.27 79.00 50 100 250 300 k3 84.67 85.73 85.60 86.18 85.16 gitation rate(rpm) 89.89 85.03 85.50 85.74 85.77 15 25 30 14.10 4.21 3.73 5.45 6.76 Particles stuffing rate (% A B C D Fig.2.The ethanol yields at different levels and factors

850 R. Liu, F. Shen / Bioresource Technology 99 (2008) 847–854 described by CO2 weight loss rate, which is deWned as a ratio between the total CO2 weight loss and total fermenta￾tion time. Table 3 shows the orthogonal experiment results of ethanol yield and CO2 weight loss rate. 3.2. Impacts of multi-factors on ethanol yield of ethanol fermentation The range analysis was applied to clarify the importance sequence of temperature, agitation rate, particles stuYng rate and pH in the orthogonal experiments. Table 4 shows the range analysis of L16(45 ) orthogonal experiments of eth￾anol yield. Results showed that range (R) of factor A was 14.10, which ranked the Wrst. Factor D was 6.76, which ranked second. Factor C was 5.45, which ranked third. The factor B was 4.21, which ranked the last. The bigger R value of a factor represents greater eVect on the Wnal ethanol yield. According to the range, the order of inXuence is temperature > pH > particles stuYng rate > agitation rate. The optimal condition for improving ethanol yield was determined as A4B3C3D4. Thus the optimal fermentation condition was determined as follows: temperature 37 °C, agitation rate 200 rpm, particles stuYng rate 25%, pH 5.0. For comparison of the signiWcance of the above four variables in inXuencing the increased ethanol yield, the analysis of variance (ANOVA) method was used. F values of four factors were as follows: FtemperatureD152.28, Fagitation rate D17.46, Fparticles stuYng rateD 31.44 and FpH D 44.80, which were all greater than the critical F value D4.40 (p< 0.01). The result of F-test meant that the four factors were all high signiWcant in inXuencing ethanol yield. The ethanol yields at diVerent levels and factors are described in Fig. 2. The ethanol yield increased from 75.79% to 89.89% while the fermentation temperature was increased from 28 °C to 37 °C. The highest yield of ethanol was 89.89% at a fermentation temperature of 37 °C. In some degree, ethanol formation is dependent on tempera￾ture, and an increase in temperature results in an increased concentration of total ethanol (Etievant, 1991; Mallouchos et al., 2003). In addition, the optimum temperature of free S. cerevisiae fermentation was always about 30 °C (Torija et al., 2003; Muenduen et al., 2006). The optimum tempera￾ture of immobilized S. cerevisiae ethanol fermentation was higher than that of free yeasts. This phenomenon may be due to the reason that the immobilized yeast in fermenta￾tion exists heat transfer process from the particle surface to its inside; The maximum yield of ethanol of 85.77% was obtained at pH 5.0. Generally speaking, H+ concentrations in the fermentation solution could change the charge quan￾tities of yeast cytoplasma membrane. As a result, the per￾meability of yeast cytoplasma membrane for some materials and ions including carbon source, nitrogen source and other inorganic salt ions would be changed with the change of H+ concentration (Jia et al., 2005). The optimum pH of free yeast ethanol fermentation was 4.8–5.0 (Zhang, 1995), while the optimum pH obtained in this experiment Table 3 Orthogonal experiment results of ethanol yield and CO2 weight loss rate Note: a–j indicated signiWcant diVerence at p < 0.05. Experiment nos. Ethanol yield (%) CO2 weight loss rate (g h¡1 ) 1 70.08 § 0.745e 0.373 § 0.0058k 2 66.66 § 2.607e 0.463 § 0.0058j 3 82.54 § 2.143cd 0.485 § 0.0046i 4 82.90 § 3.762cd 0.484 § 0.0040i 5 84.98 § 1.519bcd 0.488 § 0.0040i 6 85.15 § 2.668bcd 0.458 § 0.0029j 7 80.61 § 0.445d 0.733 § 0.0000e 8 87.55 § 1.299bc 0.770 § 0.0046d 9 81.60 § 0.748d 0.522 § 0.0000h 10 89.08 § 0.358ab 0.858 § 0.0046b 11 85.47 § 1.942bcd 0.769 § 0.0035d 12 82.53 § 0.552cd 0.587 § 0.0000g 13 89.42 § 1.957ab 0.662 § 0.0081f 14 89.71 § 1.162ab 0.772 § 0.0092d 15 93.31 § 0.000a 0.835 § 0.0040c 16 87.14 § 0.406bc 0.934 § 0.0098a Table 4 The range analysis of L16(45 ) orthogonal experiments of ethanol yield Temperature Agitation rate Blank Particles stuYng rate pH A B CD K1 909.55 978.24 983.51 968.82 1020.07 K2 1014.88 991.81 982.44 987.28 948.03 K3 1016.07 1028.81 1027.23 1034.23 1017.93 K4 1078.73 1020.37 1026.05 1028.90 1029.20 k1 75.79 81.52 81.95 80.73 85.01 k2 84.57 82.65 81.87 82.27 79.00 k3 84.67 85.73 85.60 86.18 85.16 k4 89.89 85.03 85.50 85.74 85.77 R 14.10 4.21 3.73 5.45 6.76 Q A4 B3 C3 D4 Fig. 2. The ethanol yields at diVerent levels and factors

R.Liu.F.Shen Bioresource Technology 99 (2008)847-854 851 was 5.0,which indicates that the pores on the matrix of Ca- the order of influence is temperature>agitation rate>pH> alginate are large enough to keep the nutrition materials particles stuffing rate.The optimal condition for accelerat- and inorganic salt ions go through from the outside of ing fermentation rate was determined as A B:CD, immobilized yeast particles to its inside freely;The results namely,temperature,agitation rate,particles stuffing rate showed that as the particles stuffing rate was increased the and pH were 37C,200rpm,25%and 3.5,respectively. ethanol yield was increasing up to 25%.After this particles The results of analysis of variance (ANOVA)showed stuffing rate the ethanol yield was decreasing.A particles that F values of four factors were as follows:Ftemperature= stuffing rate of 25%resulted the maximum ethanol yield of 7100.67,Fagitation rate=2542.64,Fparticles stuling rate920.53, 86.18%.The growth and metabolism of yeast cells would be FpH=1510.77,which were all greater than the critical F restricted when too high immobilized yeast particles value=4.40 (p<0.01).It could be concluded that the four stuffing rate is offered in the ethanol fermentation because factors were all high significant in influencing fermentation the nutrition material is not infinite.In addition.when the rate. particles stuffing rate is in a lower value,the quantity of The CO2 weight loss rates at different levels and factors immobilized yeast used for fermentation would be in fall. are displayed in Fig.3.The CO2 weight loss rate increased Hence,fermentation efficiency would be decreased at a with the increased temperature from 28C to 37C.The lower particles stuffing rate.The highest ethanol yield of highest CO,weight loss rate was 0.801 gh-at a fermenta- 85.73%was obtained at an agitation rate of 200 rpm.Both tion temperature of 37C.This could be attributed to many the permeation intensity of nutrition materials from the fer- kinds of enzymes in the yeast cells which work for the etha- mentation solution to the inside of yeast cells and that of nol formation in different biochemistry processes.In fact, ethanol from the inside of yeast cells to the fermentation the ethanol fermentation is the enzyme catalysis processes solution could be improved with the increased agitation Therefore,the fermentation rate would increase with the rate.These processes would enhance the sugar utilization increased fermentation temperature (Qi and Zhang,1999). and weaken the inhibition of ethanol to the yeast cells. The highest CO,weight loss rate of 0.706gh-was However,too high agitation rate would not work for obtained at agitation rate of 200rpm and kept almost improvement of ethanol yield any more because of the limi- invariable when the agitation rate was increased from tation of metabolism ability of the yeast cells.Therefore, 200rpm to 250rpm.This is probably due to the fact that the ethanol yield was not enhanced when the agitation rate the fluidity of substrate in fermentation has direct influence exceeded 200 rpm on the mass transfer between the substrate and the immobi- lized yeast during a certain range in the cells immobiliza- 3.3.Impacts of multi-factors on CO2 weight loss rate of tion system(Cheng,2000).When the fluidity of substrate ethanol fermentation solution exceeds a certain value.it is the membrane thick- ness of immobilized yeast particles to control the mass CO,is the other product of ethanol fermentation from transfer.As a result,fermentation rate would be enhanced sugar by S.cerevisiae.When batch fermentation rate was with the increased agitation rate from 100rpm to 200rpm, considered,the CO2 weight loss rate could represent fer- and kept almost invariable from 200rpm to 250 rpm. mentation rate instead of ethanol production rate.Table 5 shows the range analysis of L(4)orthogonal experiments 0.85 of CO2 weight loss rate.Range (R)of factor A was 0.353, -Temperature which ranked the first.Factor B was 0.194,which ranked 0.80- 一o一A接i准tion ra1e Partides stufling rate second.Factor D was 0.161,which ranked third.The factor C was 0.120,which ranked the last.According to the range, 0.75 0.70 Table 5 The range analysis of L(4)orthogonal experiments of CO,weight loss rate 0.60 Temperature Agitation Blank Particles pH rate stuffing rate 0.55 A B C 0.50- K1 5.381 6.136 7.605 7.397 8.509 7.346 7.656 7.120 7.993 7.958 0.45- K 8.208 8.468 7.648 8.296 6.579 9.613 8.288 8.175 6.862 7.502 0.40 30 40 0.448 0.511 0.634 0.616 0.709 pH 4.5 0.612 0.638 0.539 0.666 0.663 Temperature (C) k3 0.684 0.706 0.637 0.691 0.548 100 10 700 20 Agitation rate(rpwm） ka 0.801 0.691 0.681 0.571 0.625 15 R 0.353 0.194 0.142 0.120 0.161 Particles stuming rate ( Q As B3 C3 Fig.3.The CO,weight loss rate at different levels and factors

R. Liu, F. Shen / Bioresource Technology 99 (2008) 847–854 851 was 5.0, which indicates that the pores on the matrix of Ca￾alginate are large enough to keep the nutrition materials and inorganic salt ions go through from the outside of immobilized yeast particles to its inside freely; The results showed that as the particles stuYng rate was increased the ethanol yield was increasing up to 25%. After this particles stuYng rate the ethanol yield was decreasing. A particles stuYng rate of 25% resulted the maximum ethanol yield of 86.18%. The growth and metabolism of yeast cells would be restricted when too high immobilized yeast particles stuYng rate is oVered in the ethanol fermentation because the nutrition material is not inWnite. In addition, when the particles stuYng rate is in a lower value, the quantity of immobilized yeast used for fermentation would be in fall. Hence, fermentation eYciency would be decreased at a lower particles stuYng rate. The highest ethanol yield of 85.73% was obtained at an agitation rate of 200 rpm. Both the permeation intensity of nutrition materials from the fer￾mentation solution to the inside of yeast cells and that of ethanol from the inside of yeast cells to the fermentation solution could be improved with the increased agitation rate. These processes would enhance the sugar utilization and weaken the inhibition of ethanol to the yeast cells. However, too high agitation rate would not work for improvement of ethanol yield any more because of the limi￾tation of metabolism ability of the yeast cells. Therefore, the ethanol yield was not enhanced when the agitation rate exceeded 200 rpm. 3.3. Impacts of multi-factors on CO2 weight loss rate of ethanol fermentation CO2 is the other product of ethanol fermentation from sugar by S. cerevisiae. When batch fermentation rate was considered, the CO2 weight loss rate could represent fer￾mentation rate instead of ethanol production rate. Table 5 shows the range analysis of L16(45 ) orthogonal experiments of CO2 weight loss rate. Range (R) of factor A was 0.353, which ranked the Wrst. Factor B was 0.194, which ranked second. Factor D was 0.161, which ranked third. The factor C was 0.120, which ranked the last. According to the range, the order of inXuence is temperature > agitation rate > pH > particles stuYng rate. The optimal condition for accelerat￾ing fermentation rate was determined as A4B3C3D1, namely, temperature, agitation rate, particles stuYng rate and pH were 37 °C, 200 rpm, 25% and 3.5, respectively. The results of analysis of variance (ANOVA) showed that F values of four factors were as follows: FtemperatureD 7100.67, Fagitation rateD2542.64, Fparticles stuYng rateD 920.53, FpH D 1510.77, which were all greater than the critical F valueD 4.40 (p< 0.01). It could be concluded that the four factors were all high signiWcant in inXuencing fermentation rate. The CO2 weight loss rates at diVerent levels and factors are displayed in Fig. 3. The CO2 weight loss rate increased with the increased temperature from 28 °C to 37 °C. The highest CO2 weight loss rate was 0.801 g h¡1 at a fermenta￾tion temperature of 37 °C. This could be attributed to many kinds of enzymes in the yeast cells which work for the etha￾nol formation in diVerent biochemistry processes. In fact, the ethanol fermentation is the enzyme catalysis processes. Therefore, the fermentation rate would increase with the increased fermentation temperature (Qi and Zhang, 1999). The highest CO2 weight loss rate of 0.706 g h¡1 was obtained at agitation rate of 200 rpm and kept almost invariable when the agitation rate was increased from 200 rpm to 250 rpm. This is probably due to the fact that the Xuidity of substrate in fermentation has direct inXuence on the mass transfer between the substrate and the immobi￾lized yeast during a certain range in the cells immobiliza￾tion system (Cheng, 2000). When the Xuidity of substrate solution exceeds a certain value, it is the membrane thick￾ness of immobilized yeast particles to control the mass transfer. As a result, fermentation rate would be enhanced with the increased agitation rate from 100 rpm to 200 rpm, and kept almost invariable from 200 rpm to 250 rpm. Table 5 The range analysis of L16(45 ) orthogonal experiments of CO2 weight loss rate Temperature Agitation rate Blank Particles stuYng rate pH A B CD K1 5.381 6.136 7.605 7.397 8.509 K2 7.346 7.656 7.120 7.993 7.958 K3 8.208 8.468 7.648 8.296 6.579 K4 9.613 8.288 8.175 6.862 7.502 k1 0.448 0.511 0.634 0.616 0.709 k2 0.612 0.638 0.539 0.666 0.663 k3 0.684 0.706 0.637 0.691 0.548 k4 0.801 0.691 0.681 0.571 0.625 R 0.353 0.194 0.142 0.120 0.161 Q A4 B3 C3 D1 Fig. 3. The CO2 weight loss rate at diVerent levels and factors

852 R.Liu.F.Shen Bioresource Technology 99 (2008)847-854 The highest CO,weight loss rate of 0.663gh-was for CO,weight loss rate.So,the higher ethanol yield could obtained at pH 3.5.According to range analysis of pH,pH be obtained at pH 5.0. was the third influence factor on the fermentation rate.The As a result,the optimal condition for immobilized S. fermentation rate was seriously controlled by other factors. cerevisiae ethanol fermentation was determined as A4B3- Therefore,there was a decrease trend of CO,weight loss C.Da,but this optimal condition did not appear in the rate with increase of the pH. orthogonal test.Thus,verification experiments with the The maximum CO2 weight loss rate was 0.691gh-at corresponding parameters of the optimal condition A B3- particles stuffing rate of 25%.An increase of the particles C.D.should be carried out. stuffing rate could increase the yeast cells concentration in a certain volume fermentation solution,which could weaken 3.5.The verification experiments in shaking flasks and 5 L the inhibition of substrate and ethanol to the yeast cells. bioreactor Thus,the fermentation rate would be boosted with the increased particles stuffing rate from 15%to 25%,but the The verification experiments in 250mL shaking flasks too high particles stuffing rate in a certain concentration with the corresponding parameters of the optimal condi- fermentation solution would lead to yeast cells in"hungry tion A B3CD4 were carried out in triplicate.Results stage"and thus would decrease the fermentation rate. showed that the mean of ethanol yield and CO,weight loss rate were 98.07%and 1.020gh-,respectively,when the 3.4.Optimizing condition for fermentation total fermentation time was 5h,which were all higher than those of orthogonal experiments in Table 3.This demon- According to the analysis results of ethanol yield and strated that the optimal condition ABC Da was suitable CO2 weight loss rate,the optimal condition for improving and reliable for the ethanol fermentation by immobilized S. ethanol yield was determined as A B3C Da,while the opti- cerevisiae. mal condition for increasing CO2 weight loss rate was In order to further verify the reliability of the results determined as A BC,D.Therefore,optimal fermentation from the shaking flasks,additional experiments with the temperature,agitation rate and particles stuffing rate could corresponding parameters of the optimal condition be firstly determined as 37C,200rpm and 25%,respec- A B C Da were carried out in 5L bioreactor.The ethanol tively. fermentation results in 5L bioreactor is shown in Fig.5. In order to optimize pH in the ethanol fermentation pro- Fig.5 showed that the initial total sugar concentration was cess,the influence of pH on ethanol yield and CO,weight 68.85mgmL-.The final ethanol concentration and the loss rate is shown in Fig.4.Judged by Fig.4,pH 4.0 and 4.5 residual sugar concentration were 32.91 mgmL-and were firstly excluded to be as suitable values for ethanol fer- 1.33mgmL-at the end of fermentation of 11h,respec- mentation because the ethanol yield was the lowest at pH 4 tively.Through calculation,the final ethanol yield was and CO,weight loss rate was the lowest at pH 4.5.pH 5.0 93.24%.The ethanol yield of the optimal condition was decided as a suitable parameter,instead of pH 3.5,for A B3C3D in 5L bioreactor was lower than that of verifica- ethanol fermentation,because pH was the second impor- tion in the shaking flasks,but higher than those of the tant factor for ethanol yield and the least important factor orthogonal experiments,which further demonstrated that -Ethanol yield -0.72 86- -CO weight loss rate 0.70 0.68 84 0.66 0.64 豆 82 0.62 0.60 0.58 80 0.56 .0.54 3.0 3.5 4.0 4.5 5.0 5.5 pH Fig.4.Influence of pH on ethanol yield and CO,weight loss rate

852 R. Liu, F. Shen / Bioresource Technology 99 (2008) 847–854 The highest CO2 weight loss rate of 0.663 g h¡1 was obtained at pH 3.5. According to range analysis of pH, pH was the third inXuence factor on the fermentation rate. The fermentation rate was seriously controlled by other factors. Therefore, there was a decrease trend of CO2 weight loss rate with increase of the pH. The maximum CO2 weight loss rate was 0.691 g h¡1 at particles stuYng rate of 25%. An increase of the particles stuYng rate could increase the yeast cells concentration in a certain volume fermentation solution, which could weaken the inhibition of substrate and ethanol to the yeast cells. Thus, the fermentation rate would be boosted with the increased particles stuYng rate from 15% to 25%, but the too high particles stuYng rate in a certain concentration fermentation solution would lead to yeast cells in “hungry stage” and thus would decrease the fermentation rate. 3.4. Optimizing condition for fermentation According to the analysis results of ethanol yield and CO2 weight loss rate, the optimal condition for improving ethanol yield was determined as A4B3C3D4, while the opti￾mal condition for increasing CO2 weight loss rate was determined as A4B3C3D1. Therefore, optimal fermentation temperature, agitation rate and particles stuYng rate could be Wrstly determined as 37 °C, 200 rpm and 25%, respec￾tively. In order to optimize pH in the ethanol fermentation pro￾cess, the inXuence of pH on ethanol yield and CO2 weight loss rate is shown in Fig. 4. Judged by Fig. 4, pH 4.0 and 4.5 were Wrstly excluded to be as suitable values for ethanol fer￾mentation because the ethanol yield was the lowest at pH 4 and CO2 weight loss rate was the lowest at pH 4.5. pH 5.0 was decided as a suitable parameter, instead of pH 3.5, for ethanol fermentation, because pH was the second impor￾tant factor for ethanol yield and the least important factor for CO2 weight loss rate. So, the higher ethanol yield could be obtained at pH 5.0. As a result, the optimal condition for immobilized S. cerevisiae ethanol fermentation was determined as A4B3- C3D4, but this optimal condition did not appear in the orthogonal test. Thus, veriWcation experiments with the corresponding parameters of the optimal condition A4B3- C3D4 should be carried out. 3.5. The veriWcation experiments in shaking Xasks and 5 L bioreactor The veriWcation experiments in 250 mL shaking Xasks with the corresponding parameters of the optimal condi￾tion A4B3C3D4 were carried out in triplicate. Results showed that the mean of ethanol yield and CO2 weight loss rate were 98.07% and 1.020 g h¡1 , respectively, when the total fermentation time was 5 h ,which were all higher than those of orthogonal experiments in Table 3. This demon￾strated that the optimal condition A4B3C3D4 was suitable and reliable for the ethanol fermentation by immobilized S. cerevisiae. In order to further verify the reliability of the results from the shaking Xasks, additional experiments with the corresponding parameters of the optimal condition A4B3C3D4 were carried out in 5 L bioreactor. The ethanol fermentation results in 5 L bioreactor is shown in Fig. 5. Fig. 5 showed that the initial total sugar concentration was 68.85 mg mL¡1 . The Wnal ethanol concentration and the residual sugar concentration were 32.91 mg mL¡1 and 1.33 mg mL¡1 at the end of fermentation of 11 h, respec￾tively. Through calculation, the Wnal ethanol yield was 93.24%. The ethanol yield of the optimal condition A4B3C3D4 in 5 L bioreactor was lower than that of veriWca￾tion in the shaking Xasks, but higher than those of the orthogonal experiments, which further demonstrated that Fig. 4. InXuence of pH on ethanol yield and CO2 weight loss rate

R.Liu,F.Shen Bioresource Technology 99 (2008)847-854 853 36 70 32 --Ethanol concentration --Total sugar concentration 28 50 20- 0 16 12 20 10 .0 0 3456789012 10 23 Fermentation time (h) Fig.5.The results of the group AB:C:D ethanol fermentation in a 5 L-stirring bioreactor. the determined optimal fermentation condition A B3C,Da of Food Science and Engineering of Shanghai Jiao Tong was reasonable for improving the ethanol yield.In addition, University for the analysis of sugar concentrations with agitation was the least important factor for ethanol yield HPLC and the second important factor for fermentation rate in the four factors,thus,the same agitation rate of 200 rpm References used in two different reactor systems had only a little influ- ence on the ethanol yield and some influence on shortening Bardi,E.P.et al,1996.Room temperature and low temperature wine mak- fermentation time. ing using yeast immobilized on gluten pellets.Process Biochemistry 5, To sum up,the optimal condition for immobilized S. 425430. cerevisiae ethanol fermentation was finally determined as Bryan,W.L,1990.Solid state fermentation of sugars in sweet sorghum. A B:C,D namely,the fermentation temperature,agitation Enzyme Microbial Technology 12,437-442. Cheng.J.F 2000.Studied on Technology of continuous beer fermentation rate,particles stuffing rate and pH were 37C,200rpm, and the fermentation kinetic of yeast cells immobilized by new-style 25%and 5.0,respectively. ceramic carriers.Ph.D.Degree Thesis,Zhe Jiang University,Hangz- hou.pp.89-93. 4.Conclusions Etievant,X.P,1991.Wine.In:Maarse,H.(Ed.).Volatile compounds in foods and beverages.Culinary and Hospitality Industry Publications Services,New York,pp.490-507. Bioethanol production from sweet sorghum by immobi- Gnansounou,E.et al,2005.Refining sweet sorghum to ethanol and sugar: lization technology is promising as an alternative fuel.In economic trade-offs in the context of North China.Bioresource Tech- order to attain a higher yield of ethanol and fermentation nology96,885-1002. Jasberg.B.K.et al.,1983.Preservation of sweet sorghum biomass.Biot- rate in ethanol fermentation,main parameters of ethanol echnolgy and Bioengineering Symposium 13.113-120. fermentation by immobilized S.cerevisiae including tem- Jia.S.B.et al,2005.Alcohol Craft (New Edition).Chemical Industry perature,agitation rate,particles stuffing rate and pH were Press,Beijing. investigated.Based on the analysis of orthogonal experi- Kobayashi.F..Nakamura.Y.2004.Mathematical model of direct ethanol ments and verification experiments,the optimal fermenta- production from starch in immobilized recombinant yeast culture.Bio- tion condition was determined as follows:temperature chemical Engineering Journal 21,93-101. Lin,Y,Tanaka,S.,2006.Ethanol fermentation from biomass resources: 37C,agitation rate 200rpm,particles stuffing rate 25%, current state and prospects.Applied Microbiology and Biotechnology. pH 5.0.The results of the research would be beneficial for Available from:. the application of immobilized S.cerevisiae ethanol fer- Liu.R.H.et al,2005.Bioethanol fermentation from stalk juice of sweet mentation from stalk juice of sweet sorghum. sorghum by immobilized yeast.Journal of Shanghai Jiaotong Univer- sity S1,146-151. Mallouchos,A.et al.,2003.Wine fermentation by immobilized and free Acknowledgements cells at different temperature,effect of immobilization and temperature on volatile by-products.Food Chemistry 80,109-113. This research work is supported by EU Project (ICA4- Muenduen,P.et al.,2006.Mathematical modeling to investigate tempera- 2002-10023).We appreciate Mrs.Yumin Liu of Instrumen- ture effect on kinetic parameters of ethanol fermentation.Biochemical Engineering Journal 28,36-43. tal Analysis Centre of Shanghai Jiao Tong University for Plessas,B.et al.,2007.Use of Saccharomyces cerevisae cells immobilized the analysis of ethanol concentrations with GC.We also on orange peel as biocatalyst for alcoholic fermentation.Bioresource thank Associate Professor Dayun Zhao in the Department Technology 98,860-865

R. Liu, F. Shen / Bioresource Technology 99 (2008) 847–854 853 the determined optimal fermentation condition A4B3C3D4 was reasonable for improving the ethanol yield. In addition, agitation was the least important factor for ethanol yield and the second important factor for fermentation rate in the four factors, thus, the same agitation rate of 200 rpm used in two diVerent reactor systems had only a little inXu￾ence on the ethanol yield and some inXuence on shortening fermentation time. To sum up, the optimal condition for immobilized S. cerevisiae ethanol fermentation was Wnally determined as A4B3C3D4, namely, the fermentation temperature, agitation rate, particles stuYng rate and pH were 37 °C, 200 rpm, 25% and 5.0, respectively. 4. Conclusions Bioethanol production from sweet sorghum by immobi￾lization technology is promising as an alternative fuel. In order to attain a higher yield of ethanol and fermentation rate in ethanol fermentation, main parameters of ethanol fermentation by immobilized S. cerevisiae including tem￾perature, agitation rate, particles stuYng rate and pH were investigated. Based on the analysis of orthogonal experi￾ments and veriWcation experiments, the optimal fermenta￾tion condition was determined as follows: temperature 37 °C, agitation rate 200 rpm , particles stuYng rate 25%, pH 5.0. The results of the research would be beneWcial for the application of immobilized S. cerevisiae ethanol fer￾mentation from stalk juice of sweet sorghum. Acknowledgements This research work is supported by EU Project (ICA4- 2002-10023). We appreciate Mrs. Yumin Liu of Instrumen￾tal Analysis Centre of Shanghai Jiao Tong University for the analysis of ethanol concentrations with GC. We also thank Associate Professor Dayun Zhao in the Department of Food Science and Engineering of Shanghai Jiao Tong University for the analysis of sugar concentrations with HPLC. References Bardi, E.P. et al., 1996. Room temperature and low temperature wine mak￾ing using yeast immobilized on gluten pellets. Process Biochemistry 5, 425–430. Bryan, W.L., 1990. Solid state fermentation of sugars in sweet sorghum. Enzyme Microbial Technology 12, 437–442. Cheng, J.F., 2000. Studied on Technology of continuous beer fermentation and the fermentation kinetic of yeast cells immobilized by new-style ceramic carriers. Ph.D. Degree Thesis, Zhe Jiang University, Hangz￾hou. pp. 89–93. Etievant, X.P., 1991. Wine. In: Maarse, H. (Ed.), Volatile compounds in foods and beverages. Culinary and Hospitality Industry Publications Services, New York, pp. 490–507. Gnansounou, E. et al., 2005. ReWning sweet sorghum to ethanol and sugar: economic trade-oVs in the context of North China. Bioresource Tech￾nology 96, 885–1002. Jasberg, B.K. et al., 1983. Preservation of sweet sorghum biomass. Biot￾echnolgy and Bioengineering Symposium 13, 113–120. Jia, S.B. et al., 2005. Alcohol Craft (New Edition). Chemical Industry Press, Beijing. Kobayashi, F., Nakamura, Y., 2004. Mathematical model of direct ethanol production from starch in immobilized recombinant yeast culture. Bio￾chemical Engineering Journal 21, 93–101. Lin, Y., Tanaka, S., 2006. Ethanol fermentation from biomass resources: current state and prospects. Applied Microbiology and Biotechnology. Available from: . Liu, R.H. et al., 2005. Bioethanol fermentation from stalk juice of sweet sorghum by immobilized yeast. Journal of Shanghai Jiaotong Univer￾sity S1, 146–151. Mallouchos, A. et al., 2003. Wine fermentation by immobilized and free cells at diVerent temperature, eVect of immobilization and temperature on volatile by-products. Food Chemistry 80, 109–113. Muenduen, P. et al., 2006. Mathematical modeling to investigate tempera￾ture eVect on kinetic parameters of ethanol fermentation. Biochemical Engineering Journal 28, 36–43. Plessas, B. et al., 2007. Use of Saccharomyces cerevisae cells immobilized on orange peel as biocatalyst for alcoholic fermentation. Bioresource Technology 98, 860–865. Fig. 5. The results of the group A4B3C3D4 ethanol fermentation in a 5 L-stirring bioreactor

854 R.Liu,F.Shen Bioresource Technology 99 (2008)847-854 Prasad,B.Ishra,I.M.,1995.On the kinetic and effectiveness of immobi- Tzeng,J.W.et al.,1991.Ethanol fermentation using immobilized cells in a lized whole-cell batch cultures.Bioresource Technology 53,269-275. multistage fluidized bed bioreactor.Biotechnology and Bioengineering Qi,YZ.Zhang.S.X.,1999.Kinetics of Biochemistry Reaction and Biore. 10,1253-1258. actor.Chemical Industry Press,Beijing. Zhang.K.C.1995.Alcohol and Distilling Wine Craft.China Light Indus. Tao,F.J.et al,2005.Ethanol fermentation by an acid-tolerant Zymomonas try Press,Beijing. mobilis under non-sterilized condition.Process Biochemistry 40,183-187. Zheng,S.H.,Jiang,F.H,2004.Design of Experiments and Data Process- Torija,M.J.et al,2003.Effects of fermentation temperature on stain pop- ing.Industry and Building Materials Press of China,Beijing. ulation of Saccharomyces cerevisiae.International Journal Food Microbiology 80,47-53

854 R. Liu, F. Shen / Bioresource Technology 99 (2008) 847–854 Prasad, B., Ishra, I.M., 1995. On the kinetic and eVectiveness of immobi￾lized whole-cell batch cultures. Bioresource Technology 53, 269–275. Qi, Y.Z., Zhang, S.X., 1999. Kinetics of Biochemistry Reaction and Biore￾actor. Chemical Industry Press, Beijing. Tao, F.J. et al., 2005. Ethanol fermentation by an acid-tolerant Zymomonas mobilis under non-sterilized condition. Process Biochemistry 40, 183–187. Torija, M.J. et al., 2003. EVects of fermentation temperature on stain pop￾ulation of Saccharomyces cerevisiae. International Journal Food Microbiology 80, 47–53. Tzeng, J.W. et al., 1991. Ethanol fermentation using immobilized cells in a multistage Xuidized bed bioreactor. Biotechnology and Bioengineering 10, 1253–1258. Zhang, K.C., 1995. Alcohol and Distilling Wine Craft. China Light Indus￾try Press, Beijing. Zheng, S.H., Jiang, F.H., 2004. Design of Experiments and Data Process￾ing. Industry and Building Materials Press of China, Beijing