Transcript
Industrial Crops and Products 36 (2012) 584–588 Contents lists available at SciVerse ScienceDirect Industrial Crops and Products journal homepage: www.elsevier.com/locate/indcrop Optimization of ethanol production by Saccharomyces cerevisiae UFPEDA 1238 insimultaneous saccharification and fermentation of delignified sugarcane bagasse J.R.A. Santos, M.S. Lucena, N.B. Gusmão, E.R. Gouveia ∗ Department of Antibiotics - Federal University of Pernambuco, Cidade Universitária - CEP 50670-901, Recife, PE, Brazil a r t i c l e i n f o Article history: Received 27 July 2011Received in revised form24 September 2011Accepted 1 October 2011Available online 13 December 2011 Keywords: Sugarcane bagasseEnzymatic hydrolysisSimultaneous saccharification andfermentation a b s t r a c t Ethanol production by Saccharomyces cerevisiae UFPEDA1238 was performed in simultaneous sacchar-ification and fermentation of delignified sugarcane bagasse. Temperature (32 ◦ C, 37 ◦ C), agitation (80;100rpm), enzymatic load (20 FPU/g cellulose and 10%, v/v -glucosidase or 10 FPU/g cellulose and5% -glucosidase) and composition of culture medium were evaluated. Ethanol concentration, enzy-maticconvertibilityofcelluloseandvolumetricproductivitywerehigherthan25g/L,72%and0.70g/Lh,respectively,after30h,whentheculturemedium1and20FPU/gcellulose/10%,v/v -glucosidaseortheculture medium 2 and 10 FPU/g cellulose/5% -glucosidase were used in SSF at 37 ◦ C and 80rpm. In theSSF with culture medium 2 (supplemented with ammonium, phosphate, potassium and magnesium),150L ethanol/t bagasse was achieved, with minimum enzyme loading (10 FPU/g cellulose and 5%, v/v -glucosidase) for 8%, w/v of solids, which is often an important requirement to provide cost-efficientsecond generation ethanol processes.© 2011 Elsevier B.V. All rights reserved. 1. Introduction Currently,thereisgrowinginterestintheuseoflignocellulosesbioresources,includingagro-industrialresidues,suchassugarcanebagasse, in different processes as production ethanol and enzymes(Carrilo et al., 2005). Bagasse and sugarcane straw are lignocellu- losic materials that have attracted interest from scientists in Brazilas potential sources for lignocellulosic ethanol production (Silvaet al., 2010).There are several technologies available for the conversion of lignocellulosic materials into simple monomeric sugars. The maindifference between those technologies is the catalyst used for thebreak-down of polysaccharides in the raw material (Kádár et al.,2004).Enzymatichydrolysiscanbeusedforobtainingfermentablesugars from polysaccharides contained in lignocellulosic biomass(Sun and Cheng, 2002).Simultaneous saccharification and fermentation (SSF) is a pro-cess scheme for integrating enzymatic hydrolysis into the overallcellulosetoethanolbioconversionprocess(Martínetal.,2002).The SSF is a more efficient process than separate hydrolysis and fer-mentation (SHF), since it reduces the accumulation of sugar andminimizes end-product inhibition (Brethauer and Wyman, 2010).Moreover, simultaneous saccharification and fermentation (SSF)techniqueprovidesthepossibilityofdecreasingtheproductioncost ∗ Corresponding author. E-mail address:
[email protected] (E.R. Gouveia). (Kádáretal.,2004)andtoreducetheriskofcontamination(Wyman et al., 1992).The main microorganisms used for industrial ethanol produc-tion are yeasts. Saccharomyces cerevisiae , the yeast traditionallyused for ethanol production, cannot metabolise xylose, thesecondmostabundantsugarinlignocellulosichydrolysates(Hahn-Hägerdal et al., 2001). Saccharomyces strains require temperature lower than 35 ◦ C(Kádár et al., 2004). However, S. cerevisiae UFPEDA 1238 (CultureCollection of Department of Antibiotics of the Federal Universityof Pernambuco, Brazil) performed higher ethanol production at37 ◦ C than at 30 ◦ C (Santos et al., 2010a). On the other hand, cel- lulases, which are frequently applied in the cellulose hydrolysis,have 50 ◦ C as the optimal temperature. At lower temperatures, thesubstantiallylowerhydrolysisrateswouldbeunfavorableintermsofincreasedprocessingtime(Kádáretal.,2004;Adsuletal.,2005).Theoptimaltemperaturefortheyeastandtheenzymesuseddiffer,whichmeansthattheconditionsusedinSSFcannotbeoptimalforboth the enzymes and the yeast (Öhgren et al., 2007).Thetaskofhydrolyzinglignocellulosetofermentablemonosac-charide is still technically problematic because the digestibility of cellulose is hindered by many physical–chemical, structural andcompositionalfactors.Thepretreatmentisanecessarysteptoaltersomestructuralcharacteristicsoflignocelluloses,increasingglucanandxylanaccessibilitytotheenzymaticattack.Thecombinationof the composition of the substrate, type of pretreatment, and loadand efficiency of the enzymes used for the hydrolysis have a greatinfluenceonbiomassdigestibility,althoughtheindividualimpacts 0926-6690/$ – see front matter © 2011 Elsevier B.V. All rights reserved.doi:10.1016/j.indcrop.2011.10.002 J.R.A. Santos et al. / Industrial Crops and Products 36 (2012) 584–588 585 ofthesefactorsontheenzymatichydrolysisarestillunclear(Alviraet al., 2010).The aim of this study was to evaluate the ethanol productionby S. cerevisiae UFPEDA 1238 in simultaneous saccharification andfermentationofdelignifiedsugarcanebagasse(8%,w/v).Theeffectsoftemperature(32 ◦ Cand37 ◦ C),agitation(80and100rpm),enzy-maticload(20FPU/gcelluloseand10%,v/v -glucosidase;10FPU/gcellulose and 5%, v/v -glucosidase) and composition of culturemedium were evaluated. 2. Material and methods 2.1. Raw material and delignification Sugarcane bagasse, pretreated by steam explosion at 200 ◦ C for7min on the pilot scale, was kindly provided by Department of Biotechnology of Engineering College of Lorena (University of SaoPaulo). A portion of the pretreated material was delignified with1% (w/v) NaOH. The delignification reaction was made in a reactorRegmed (AUE/20), fitted with mixing and heating systems, using aliquid–solid1:10(w/v).Theoperationwascarriedoutat100 ◦ Cfor1h.The content of polysaccharides and lignin in the raw materialwas determined by two-step analytical acid hydrolysis, accord-ing to the analytical procedure recommended Rocha et al. (1997)and validated for sugarcane bagasse by Gouveia et al. (2009).Polysaccharide content was calculated after the chromatographicquantification of sugars in the hydrolysates, and lignin was deter-mined as the hydrolysis residue. 2.2. Enzymes and activities A commercial preparation of Trichoderma reesei cellulases (Cel-luclast1.5L:42.40FPU/mLand21.10CBU/mL)anda -glycosidase(1340CBU/mL)preparation(Novozym188),bothfromNovozymesA/S (Bagsværd, Denmark), kindly donated by the Department of Biotechnology of Engineering College of Lorena (University of SãoPaulo)wereadded.Enzymeactivitiesexpressedinfilterpaperunits(FPU)/mL and in unit of cellobiose (CBU)/mL, were determinedaccording to the method described by Ghose (1987). 2.3. Microorganism Theindustrialstrain S. cerevisiae UFPEDA1238wasused,whichwas kindly provided by the Culture Collection of Department of Antibiotics of the Federal University of Pernambuco, Brazil. Thisculturewasmaintainedonthemediumculturecontaining(ing/L):glucose (20), yeast extract (5), peptone (3) and agar (15), at pH 7.0. 2.4. Inoculum culture Pure yeast culture growth in culture medium described in Sec-tion 2.3, was added to a 500mL Erlenmeyer flask, which contained 100mL of following medium: glucose (20g/L), yeast extract (5g/L)andpeptone(3g/L)atpH7.0.TheErlenmeyerflaskwasincubatedinarotaryshakerat30 ◦ Cand250rpm.After12h,thecellssuspensionwasfiltratethrougha0.45 mfilter.Thefiltratewasdiscardedandthe cells were re-suspended in 10mL sterile water and transferredto a 250mL Erlenmeyer flask, containing 90mL of fermentationmedium. 2.5. Simultaneous saccharification and fermentation Simultaneoussaccharificationandfermentationwasperformedin 250mL Erlenmeyer flasks. Each Erlenmeyer flask contained90mL of fermentation medium (with the nutrients dissolved ina sodium citrate buffer at 50mM and pH 4.8) and 8g of delig-nified bagasse. The Erlenmeyer flasks were incubated in a rotaryshaker at 50 ◦ C and 150rpm. After a 6h prehydrolysis, each Erlen-meyer flask was inoculated with yeast cells (described in Section2.4) and incubated at 37 ◦ C and 80rpm. Initial cell concentrationwas 1g/L. Nutrients added were: (NH 4 ) 2 SO 4 1g/L; K 2 HPO 4 0.5g/L;MgSO 4 · 7H 2 O 0.25g/L; yeast extract 2g/L; peptone 1g/L (culturemedium 1) and (NH 4 ) 2 SO 4 2g/L; KH 2 PO 4 2g/L; MgSO 4 · 7H 2 O0.75g/L; yeast extract 4g/L (culture medium 2).Enzyme loads of 10 or 20 FPU/g cellulose (Celluclast 1.5L)and 5 or 10%, v/v (of the volumetric Celluclast 1.5L addition) -glucosidase were used. Temperature and agitation were kept at37 ◦ C/80rpm, 32 ◦ C/80rpm and 37 ◦ C/100rpm. These three condi-tions were chosen according to our previous study (Santos et al.,2010b). The experiments were performed in duplicates. The enzy-matic convertibility of cellulose (ECC) was calculated based inethanol concentration (Martín et al., 2008).ECC = E f − E i C i × 0 . 57where E f is the final ethanol concentration (g/L); E i is the initialethanolconcentration(g/L); C i ,initialcelluloseconcentration(g/L).Thefactor0.57isthestoichiometricyieldofethanolfromcellu-lose. 2.6. Chromatographic analysis Sugars, carboxylic acids, ethanol and furan aldehydes werequantified by HPLC (Agilent HP 1100, Germany). All sampleswere filtered through a 0.45 m filter. Cellobiose, glucose, arabi-nose, xylose, acetic acid, formic acid and ethanol were separatedon an Aminex HPX-87H + (Bio-Rad, Hercules, CA, USA) columnat 50 ◦ C, using 5mM H 2 SO 4 at a flow rate of 0.6mL/min asmobile phase, and detected RI-detector (Agilent). Furfural and 5-hydroxymethylfurfural (HMF) were separated on a C-18 column(Beckman)at25 ◦ C,using11.2/88.8acetonitrile/1%(v/v)aceticacidmixture at a flow rate of 0.8mL/min as mobile phase, and detectedby their UV absorbance at 274nm (Agilent). 3. Results and discussion Sugarcane bagasse, pretreated by steam explosion contained49.89%cellulose,7.99%hemicelluloseand34%lignin(Gouveiaetal.,2009). Steam pretreated sugar cane bagasse was delignified foravoidingtheinfluenceoflignin,sincethiscompoundformsabarrierto enzymatic attack (Chang and Holtzapple, 2000).The yield of cellulosic pulp recovered after the alkaline delig-nification was 50%. The pulp contained 81.8% cellulose, 6.4%hemicellulosesand3.0%lignin.Thecontentofcelluloseinthesolidfractionincreasedasaresultofthesolubilisationoflignin.Inaddi-tion to the increase of cellulose content and decrease of lignincontent,anenhancementofitsenzymaticconvertibilityisexpectedsince it has previously been reported that NaOH increased hard-wood digestibility from 14 to 55% concomitantly with a reductionof lignin content from 24–55% to 20% (Kumar et al., 2009).Steam explosion is the most widely employedphysical–chemical pretreatment for lignocellulosic biomass.The auto-hydrolysis of acetyl groups present in hemicellulose(Alvira et al., 2010) is observed. The lignin is redistributed and to someextentremovedfromthematerial(Panetal.,2005).Removal of hemicelluloses and the redistribution of lignin probably mayhave exposed the material and increased the delignification(91.10%).Ethanol concentration, volumetric productivity and the enzy-matic convertibility of cellulose (ECC), were higher at 37 ◦ C and80rpm (Fig. 1), while in the other two SSF (32 ◦ C/80rpm and 586 J.R.A. Santos et al. / Industrial Crops and Products 36 (2012) 584–588 Fig. 1. Ethanol concentration, ECC and Q P , in each SSF run at 20 FPU/g cellulose,10%, v/v -glucosidase and culture medium 1: A (32 ◦ C, 80rpm); B (37 ◦ C, 80rpm);C (37 ◦ C, 100rpm). 37 ◦ C/100rpm), were found similar values. Ethanol concentration,ECC and volumetric productivity, after 30h (considering a 6h pre-hydrolysis), reached 26.03g/L, 74.30% and 0.88g/Lh, respectively,in SSF at 37 ◦ C and 80rpm. Decreasing temperature (37–32 ◦ C) orincreasing agitation (80–100rpm) decreased the ECC (21.62%), theethanol concentration (23.08%) and the volumetric productivity(22.70%).According to Öhgren et al. (2007), the temperature used in SSF cannot be optimal for enzymes and yeast. However, in our studies,37 ◦ C was optimal temperature found to SSF of sugar cane bagasseby S. cerevisiae UFPEDA 1238. This strain presented higher ethanolconcentration at 37 ◦ C than at 30 ◦ C or 45 ◦ C, when was performedfermentation with sucrose (Santos et al., 2010a).Analysis of variance by Origin 6.0 was performed with ethanolconcentrationsobtainedat28hintheSSFruns,whenthetempera-tureandagitationwerevaried(SSF:A,BandC).Theseresultsweresignificantly different ( F =247.93; ˛ =0.05). However, the analysisof variance between the A SSF (32 ◦ C, 80rpm) and C SSF (37 ◦ C,100rpm),showedthattheethanolconcentrationswerenotsignif-icantly different ( F =0.0969; ˛ =0.05).For evaluating decreasing enzyme load and of the composi-tion of culture medium, experiments were carried out at 50 ◦ Cand 150rpm with 10 or 20 FPU/g cellulose and 5 or 10%, v/v -glucosidase, respectively. After 6h, yeast suspension was inocu-latedandthetemperatureandagitationwerereducedto37 ◦ Cand80rpm, respectively. SSF were carried out with two culture mediaaccording to composition described in Section 2.5.As can be seen in Fig. 2, decreasing enzyme loads (20–10 FPU/g and10–5%,v/v -glucosidase)alsodecreased(8.40%)theECC,whenthe culture medium 1 was used in both SSF (B and E). However,whentheenzymeloadwasreducedandtheculturemedium2wasused, the ECC decreased 0.86% only (B SSF and D SSF).Decreasing enzyme loads also decreased the volumetric pro-ductivity about 20% (B SSF and E SSF) and 12% (B SSF and D SSF),respectively,whenculturemedium1ortheculturemedium2wasused.Ontheotherhand,theethanolconcentrationincreased6.45%and decreased 8.45%, when the enzyme loads were reduced andthe culture medium 2 (B SSF and D SSF) or 1 (B SSF and E SSF),respectively, were used.The enzymatic convertibility of cellulose (ECC) and the ethanolconcentration were higher than 72% and 25g/L, respectively, at28h, 37 ◦ C and 80rpm, when the culture medium 1 (20 FPU/g cel-lulose and 10% -glucosidase) or the culture medium 2 (10 FPU/gcelluloseand5% -glucosidase)wasused.Martínetal.(2008)f ound ECC higher than 80%, after 120h, in SSF with S. cerevisiae or Mucor indicus strains.However,theethanolconcentrationsachievedwere Fig.2. Ethanolconcentration,ECCand Q P ,ineachSSFrunat37 ◦ Cand80rpm:B(20FPU/gcellulose,10%,v/v -glucosidase,culturemedium1);D(10FPU/gcellulose,5%,v/v -glucosidase, culture medium 2); E (10 FPU/g cellulose, 5%, v/v -glucosidase,culture medium 1). not higher than 20g/L as a consequence of the low cellulose con-tent of the raw material, although these authors have utilized highwater insoluble solids content (10%).In ethanol production from lignocellulosic materials, ethanolconcentrationshouldbeashighaspossibleinordertominimizetheenergyconsumptioninevaporationanddistillation(Wingrenetal.,2003). Increasing water insoluble solids content, in SSF, increasesthe glucose and ethanol concentrations. A water insoluble solidcontent of 8% was high enough to obtain reasonable ethanol con-centration (higher than 25g/L).Analysis of variance was performed with the ethanol concen-tration obtained at 28h in the SSF runs, when the enzyme loadand culture medium were varied (SSF: B, D and E). These resultsweresignificantlydifferent( F =15.75; ˛ =0.05).However,theanal-ysis of variance between the B SSF (20 FPU/g cellulose, 10%, v/v -glucosidase,culturemedium1)andDSSF(10FPU/gcellulose,5%,v/v -glucosidase, culture medium 2), showed that the maximumethanol concentrations were not significantly different ( F =7.08; ˛ =0.05).Fast dissolution and ECC almost 40%, after a 6h pre-hydrolysisat50 ◦ Cand150rpmwasachievedforbothenzymeloads.InSSFof trebol (Martín et al., 2008), was also found rapid dissolution after 6h at 50 ◦ C. Santos et al. (2010b) observed that SSF of sugar cane bagasse without pre-hydrolysis is a slower process.Ethanolconcentrationsandvolumetricproductivities( Q P )inallconditions were higher than that found by Öhgren et al. (2007)at 35 ◦ C in isothermal SSF and Kádár et al. (2004) in isothermal SSF (40 ◦ C)ornon-isothermalSSF(50 ◦ Cduring24hpre-hydrolysisand30 ◦ C after inoculation of yeast). Martín et al. (2008) also obtained lower ethanol concentration and volumetric productivity in non-isothermal SSF (50 ◦ C during 6h pre-hydrolysis and 32 ◦ C afterinoculation of yeast). Table 1 shows a comparison between some ethanol concentrations and volumetric productivities found in lit-erature and in the present work.Supplementationofculturemediumwithhigherconcentrationsof ammonium, potassium, phosphorus and magnesium may havefavorably influenced the fermentation. S. cerevisiae uses nitrogenasammonium,amide(urea)oramine(aminoacids).Phosphorusisabsorbed as H 2 PO − 4 , and the sulfur can be assimilated as sulfate(Lima et al., 2000). Martín et al. (2008) reported that high con- tent of potassium and phosphorus in the culture medium was alsofavorably for the fermentation.In SSF, after 12h, glucose was completely consumed in allfive conditions (20 FPU/g cellulose and 10%, v/v -glucosidase: J.R.A. Santos et al. / Industrial Crops and Products 36 (2012) 584–588 587 Table 1 Ethanol concentration using various substrates, microorganisms and enzyme load in isothermal and non-isothermal SSF.Microorganism SSF a T ( ◦ C) N-SSF b T ( ◦ C) Pre-hydrolysis (h) Ethanol c (g/L) Q Pd (g/Lh) Reference K. marxianus 40 – – 17.8 0.25 Kádár et al. (2004)– 50 and 30 24 16.0 0.22 S. cerevisiae 40 – 16.6 0.23– 50 and 30 24 15.1 0.21 S. cerevisiae 35 – – 20.5 0.17 Öhgren et al. (2007) S. cerevisiae – 50 and 32 6–8 23.7 0.20 Tomás-Pejó et al. (2008) S. cerevisiae UFPEDA 1238 – 50 and 37 6 27.71 0.77 Present work (Fig. 2 – SSF D) a Isothermal. b Non-isothermal. c Ethanol concentration. d Volumetric productivity. A B C D E050100150200 Y i e l d ( L E t h a n o l / t o n b a g a s s e ) SSF Fig. 3. Yield (L ethanol/t bagasse) under five different conditions: A (32 ◦ C, 80rpm, 20 FPU/g cellulose, 5%, v/v -glucosidase, culture medium 1); B (37 ◦ C, 80rpm, 20 FPU/gcellulose, 5%, v/v -glucosidase, culture medium 1); C (37 ◦ C, 100rpm, 20 FPU/g cellulose, 5%, v/v -glucosidase, culture medium 1); D (37 ◦ C, 80rpm, 10 FPU/g cellulose, 5%,v/v -glucosidase, culture medium 2); E (37 ◦ C, 80rpm, 10 FPU/g cellulose, 5%, v/v -glucosidase, culture medium 1). 32 ◦ C/80rpm; 37 ◦ C/80rpm; 37 ◦ C/100rpm - 10 FPU/g cellu-lose and 5%, v/v -glucosidase: 37 ◦ C/80rpm/culture medium 1;37 ◦ C/80rpm/culturemedium2).Inisothermal(40 ◦ C)SSFofSolkaFloc 2000 the glucose concentration stayed at around 3–5g/L dur-ing 72h (Kádár et al., 2004), when S. cerevisiae or Kluyveromycesmarxianus were utilized. These authors reported that the cell mayhave been suffered with this temperature and that it was espe-ciallyunexpectedwiththethermotolerant K.marxianus ,whichwasthought to be performing much better at 40 ◦ C than S. cerevisiae .Cellobiose concentrations were lower than 0.2g/L, after 12h,utilizing10or5%,v/v -glucosidase.Öhgrenetal.(2007)performed SSF with 25%, v/v -glucosidase to hydrolysis all cellobiose. On theotherhand,Chenetal.(2007),whennotsupplementingthehydrol- ysis with -glucosidase, observed a severe inhibition of cellulasesactivityduetotheaccumulationofcellobiose(7.4g/L).Whentheseauthors supplemented with -glucosidase (6.5 CBU/g substrate),the concentration of cellobiose decreased to 0.6g/L.Fig. 3 shows a comparison of ethanol yield from raw mate-rial. Ethanol volume (in L) in relation to bagasse mass (in ton)was calculated considering the recovering of solids after the alka-line delignification (50%) and steam explosion (68%). In SSF withculture medium 2, enzyme load 10 FPU/g cellulose and 5%, v/v -glucosidase (D), yield higher than 150L EtOH/t bagasse wasachieved. This represents 100L EtOH/t sugarcane, since each tonof cane generates 2/3 of bagasse. 4. Conclusions Theadequatecombinationofconditionsofpre-hydrolysis(6hat50 ◦ Cand15rpm),temperature(37 ◦ C),agitation(80rpm),compo-sition of culture medium (with higher concentration of nutrients)andenzymeload(10FPU/gcelluloseand5%,v/v -glucosidase)wasa successful method to ethanol production by S. cerevisiae UFPE1238 in SSF of delignified sugarcane bagasse. High yield ethanolwith minimum enzyme loads and lower time (34h, consideringa 6h pre-hydrolysis) was achieved, which is often an importantrequirement to provide cost-efficient second generation ethanolprocesses. Acknowledgements The authors acknowledge the financial support from ConselhoNacional de Desenvolvimento Científico e Tecnológico, Brasilia DF,Brazil(CNPq)andfromFundac¸ãodeAmparoàCiênciaeTecnologiado Estado de Pernambuco (FACEPE). References Adsul,M.G.,Ghuleb,J.E.,Singhb,R.,Shaikhb,H.,2005.Polysaccharidesfrombagasse:applicationsincellulaseandxylanaseproduction.Carbohydr.Polym.57,67–72.
