Simultaneous Saccharification and Fermentation of Several Lignocellulosic

Embed Size (px)

Citation preview

  • 7/28/2019 Simultaneous Saccharification and Fermentation of Several Lignocellulosic

    1/7

  • 7/28/2019 Simultaneous Saccharification and Fermentation of Several Lignocellulosic

    2/7

    30 2 C. E. W YMAN l al .

    Yeast

    Fig. 1. The simultaneous saccharification and fermentation(SSF) process combines cellulose hydrolysis to glucose andfermentation of glucose to ethanol in one vessel.

    In this paper, a summ ary will be provided ofthe evaluations of several potential feedstocksfor ethanol production via the SSF process.Because only selected data can be presentedhere, particular emph asis will be placed on thevariation in SSF performance with substrate,fermentative microbe, and enzyme augmenta-tion. It will be shown that a wide range offeedstocks is suitable for ethanol production.Recomm endations will then be made for futureresearch to achieve economic goals.

    2. MATERIALS AND METHOD S2.1. M a t er i a l s

    Four woody crops w ere employed in thisstudy: Popu l u s m ax imow i c z i i x n i g r a (HybridNE388), Popu l u s t r i choca rpa x del t o i des (Hy-brid Nil), t r emu l o i des (aspen), and a nativestrain of sweetgum ( L i qu i damba r s t y r a c l sua ) .The herbaceous crops for this study were weep-ing love grass (E rag ro t i s cu r vu l a ) , Sericea les-pedeza (Lespedeza cuneata) , and switchgrass( P a n i cu r n v i r g a t um ) . P . max im ow i c z i i , P . t r i -choca rpa , L . s ty ra@ua, and the herbaceouscrops were supplied to us through the coordi-nation of the Bioma ss Production Programmanag ed by Oak Ridge National Laboratory(ORN L), Oak Ridge, Tennessee for the U.S.Departmen t of Energy (DOE). The cornresidues used in this study w ere harvested fromthe eastern Colorado slope in the fall of 1984and selected for absence of rot. Wheat strawwas also obtained locally. The fermentationyeast used were Sacchromyces ce rev i s iue (D,A),a National Renewa ble Energy Laboratory(NREL) strain genetically derived from RedStar brewers yeast, and Bre t tanomyces c lausen i iY-1414 obtained from the Northern RegionalResearch Laboratory (NRR L), U .S. Depart-ment of Agriculture, Peoria, Illinois. Chemicalswere purchased from the Sigma Chemical C om-pany, St Louis, M issouri, and yeast extract a ndpeptone growth media were ordered from

    Difco, Detroit, Michigan. Cellulase enzymecame from Genencor Inc., San Francisco, Cali-fornia, and P-glucosidase (Novozyme-188) fromNO VO Laboratories, Inc., Wilton, Connecticut.Shaker flask, 250 ml Pyrex grad uated vessels,and Braun B iostat V fermentation vessels wereused for the fermentations.2.2. M et h o d s

    Shaker flask SSFs were carried out in 250 mlflasks outfitted with stoppers co nstructed tovent CO2 through a water trap. These flaskscontained 100 ml of fermentation broth andwere agitated at 1 50 rpm in a shaker incubatorat 37C. A 1% yeast extract and 2% peptone(w/v) media (YP) were used with a substrateloading of 7.5% (w /v) lignocellulose. A lipidmixture of ergosterol (5 mg 1-l) and oleic acid(30mg 1-l) was added to the media for im -proved ethanol yield. * Also, penicillin andstreptomycin at 10 mg 1-l were used to minimizebacterial contamination. The inocula weregrown in a shaker flask with YP media and 2%(w/v) glucose at 37C and l/l0 (v/v) yeastculture to total volume of media w as added tothe fermentation. The substrate was autoclavedin fermentation flasks, and sterile media, lipids,antibiotics, and enzyme were added before theinoculum.

    Ethanol concentrations in the supernatantwere measured by gas chromatography using aPorapak QSO /lOO column. The internal stan-dard was 4% isopropanol. Larger scale 3 1 SSFswere also run at selected conditions. For theseexperiments, residual sugars (glucose and cel-lobiose) were determined as glucose by incu-bation of the sample with 2 mgm l- almondextract /?-glucosidase from Sigma for 1 h at37C and total sug ars were measured on themodel 27 glucose analyzer from Yellow SpringsInstruments, Yellow Springs, Ohio. Viable celldensities were measured as colony forming units(CFU) by plating serial dilutions on YP mediumsupplemented with either glucose (YPD ) orcellobiose (YPC).Cellulase enzyme loadings of 7, 13, 19, and 2 6IU g- lignocellulose were used in the shakerflask screening experiments to span the range ofactivity previously shown to be important forSSFs. In this paper, IU stands for internationalunits of filter paper activity in ,umol glucosemin-I.** /3-glucosidase enzyme w as reported inthis study a t ratios of 1 and 8 parts to 1 part ofcellulase as measured by IU of /I-glucosidaseper IU of cellulase. The /?-glucosidase activity

  • 7/28/2019 Simultaneous Saccharification and Fermentation of Several Lignocellulosic

    3/7

    Simultaneous saccharification and fermentation 30 3was determined by p -nitrophenyl-fi -glucoside(pNpg) assays at a temperature of 37C becausethis is the temperature for the SSFs. The activityof cellulase increases as temperatures areincreased to an o ptimum at 45C,7*8 thetemperature selected for saccharification with-out fermentation studies. The IUPA C revisionsof measured cellulase activities indicate that thelevel of fl-glucosidase in an enzyme preparationmay affect the results of the cellulase assay infilter paper units.22

    Woody crops were completely d ebarked. Two500 mg batches of Wiley-milled (2 mm screen)wood, herbaceous plants, or crop residues werepretreated with dilute sulfuric acid (0.45-0.50%v/v) in a 2-gal Parr Reactor. All materials werepretreated at about 140C for 1 h with stirringat 1 85 rpm. After reaction, the slurries w erewashed several times with hot water in a largeBuchner funnel lined with a linen sheet to bringthe pH of 1.3 up to 4.5. These b atches werecombined, immediately placed in freezer storagebags, and stored at - 20C. Approximately70% of the pretreated woods dry weight wascellulose, 29% was lignin and acid-insolubleash, and 1% was xylan. Approximately 58%of the pretreated grasses dry weight wascellulose, 40% was lignin and acid-insolubleash, and 2% was xylan. The pretreated legumegave a lower dry weight of cellulose at 45%,with 51% lignin and acid-insoluble ash, and2.5% xylan.

    Shaker flask results are reported as a percent-age of maxim um theoretical ethanol yields anddo not account for the substrate used in cellgrowth. T hus, the maxim um expected ethanolyield is about 95%, assum ing tha t about 5% ofthe substrate is needed for cell growth. Thesecalculations are based on the measured ethanolconcentrations and a 56.7% theoretical ethanolyield for conversio n of cellulose to ethano l only.However, the saccharifications of cellulose arereported on the basis of the percentage of themaximum amount of sugars possible. Thus,comparison of the SSF and straight saccharifi-cation results must consider that subsequentfermentation of the sugars produced in the latterwill also result in about a 5% loss to cell growthand maintenance.A method was developed to estimate theresidual cellulose by poisoning the yeast cellswith NaF and allowing the excess enzyme tocomplete the saccharification from a giventime of SSF. Background measurem ents ofglucose were subtracted, which stem from the

    /I-glucosidase enzyme itself and from some re-sidual glucose not taken up by the cells at thetime of the sample. At lower enzyme levels,additional cellulase and /I glucosidase wereadded to complete the saccharification.

    The straight saccharification yields are calcu-lated as the amount of glucose producedcompared to the potential glucose in the cellu-lose feed. The substrate level was limited to7.5% cellulose for a small- and large-scale SSFsbecause mixing problems were encountered athigher cellulose levels during SSF evaluationsperformed with S. ce rev i s iae and pretreatedwheat straw at higher substrate levels.18

    3. RESULTSSupplemen tation of the Genencor 15OL en-

    zyme (batch II) used in this study with j?-glucosidase was used to increase ethanol yieldsand saccharification rates to values observedwith a previous batch of this enzyme. Previouswork indicated a n approximately 35% re-duction in pNpg activity g- of protein forbatch II when compared with batch I.***With-out b-glucosidase supplemen tation, a decreaseof approximately 12% in ethanol productionwas observed.

    Table 1 illustrates the final saccharificationyields at 45C for all woody and herbaceouscrops studied plus corn cobs, corn stover,and wheat straw at selected cellulase loadingsand supplementation with /.I-glucosidase. Forwoody crops, the best overall rates of hydrolysisand final conversions to simple sugars wereobserved for P . ma ximow i c zi i an d L . st y r a ct j l u a .The highest enzyme loadings of 26 IU cellulasewith an 8: 1 ratio of /I-glucosidase achieved amaxim um cellulose conversion of 82% with P .ma x imow i c zi i . The saccharifications took up to8 days to achieve the highest yields.

    For herbaceous crops, the best overall rates ofhydrolysis and final conversions to simple sug-ars were observed for weeping love grass andswitchgrass. The highest enzyme loading of 26IU cellulase gg cellulose, with an 8: 1 ratio of/I-glucosidase, was necessary to achieve 70-73%conversion for these grasses, and the saccharifi-cations took 5 days to achieve these yields. Only39% of the cellulose rem aining in the pretreatedlegume, S. lespedeza, was saccharified underthese conditions. It has been reported23 thatthe legume requires harsher conditions of pre-treatment (180C for 20 min using 0.5% H2S04v/v) to achieve the high conversion yields

  • 7/28/2019 Simultaneous Saccharification and Fermentation of Several Lignocellulosic

    4/7

    304 C. E. WYMAN t al.Table 1.Summ ary of linal(8-day) saccharification yields for acid pretreated woody andherbaceous crops with selected cellulase and /I-glucosidase loadings at 45CIU j-glucosidase:IU 0: l 8:1IU Cellulose/a lianocellulose 7 13 19 26 7 13 19 26_ _Woody cropsPopulus maximowiczii x ni gra*

    Populus tri chocarpa x akl toidesPopulus tremuloidesSweetgum Li quidambar styract&a *Herbaceous cropsWeeping love grass*Switchgrass*Sericea lespedezaCorn cob*Corn stoverWheat straw+

    24 48 66 72 23 50 82 8214 28 40 47 14 31 35 6517 30 48 65 16 30 43 6432 46 53 66 34 48 68 7430 4732 458 1255 6448 6434 48

    6053157877-

    69 47 6061 47 5619 9 1786 69 8384 64 8064 60 73

    6864249086

    737039908977

    *Saccharification yields are expressed in percentage of theoretical conversion.

    (approximately 70%) observed w ith the othertwo grasses.Table 2 compares the final (7-8 day) ethanolyields for SSF with different substrates atselected enzyme loadings and yeast combi-nations run in 100ml shaker flasks. The ratesand yields are consistently higher for P . mux i -m o w i cz i i compared to the other woody cropswith S. cereu is iue as the fermenting yeast at allenzyme loadings. However, the mixed culturegenerally achieves the best overall rates andhighest yields for L . s r y r a cz~uu at the lowerenzyme loadings without /3-glucosidase sup-

    plementation. With fi-glucosidase supplemen-tation, P. t remu lo ides g ives the best results withthe mixed culture at lower cellulase loadingswhile P . ma x imow i c zi i performs best withgreater cellulase use. Lower rates and yieldsresult with P . t r i choca rpa than with other sub-strates.

    Table 2 also compares the final ethanol yieldsfor SSF with herbaceous substrates at selectedenzyme loadings and yeast combinations run in250 ml flasks. Just a s for the straight saccharifi-cation, the rates and yields are consistentlyhigher for weeping love grass and switchgrass

    Table 2. Summary of final yields in SSFs for pretreated ligocellulosic biomass ru n at37C for selected cellulase and /I-glucosidase loadingsIU B-alucosidase:IU 0: l 8: lIU Cellulase/g lignocelluloseYeastWoody cropsPopulus maximowiczii x n igra*Populus tr ichocarpa x deltoidesPopulus tremuloides aspen)Sweetgum Li quidambar styractfiu a

    Herbaceous cropsWeeping love grassSwitchgrassSericea lespedezaCorn cobCorn stoverWheat strawYeastWoody cropsPopulus maximowiczii x nigraPopulus tr ichocarpa x deltoidesPopulus tremuloides aspen)Sweetgum L iqui dambar styraet#?uaHerbaceous cropsWeeping love grassSwitchgrassSericea lespedezaCorn cobCorn StoverWheat Straw

    7 13 19 26 7 13 19 26

    42 61 75 80 79 86 88 9039 47 55 62 56 61 77 8234 51 66 76 69 81 83 8441 61 70 77 77 83 86 8647 5741 5218 2258 6354 5945 57

    64 75 63 7159 61 56 7027 30 28 3380 87 87 9177 84 82 86- 76 67 77Mixed Culture*

    85 8981 8444 5294 9490 92- 90

    S. cerevisiae

    53 67 74 77 65 84 86 8639 43 47 52 52 66 73 8063 66 74 76 73 78 79 7968 72 77 78 67 79 82 8462 65 70 77 71 85 88 8649 57 63 67 62 73 83 8725 28 33 33 29 41 49 5076 85 89 92 92 93 96 9675 84 87 89 86 89 92 9266 73 - 76 79 83 - 85

    *Mixed culture of Br ettanomyces clauseni i and Saccharomyces cereoisiae.

  • 7/28/2019 Simultaneous Saccharification and Fermentation of Several Lignocellulosic

    5/7

    Simultaneous saccharification and fermentation

    100??26

    60% Yield

    40

    305

    02 81 0:l 6:l 0:1 6:l 0:l a:i 0:l 6:l 0:l a: iPopulus Populus Sweetgum Sweetgum Switchgrass Switchgrassmaximo+viczi i maximowiczi i Liquidambar Liquidambarx nigra x nigra styraciflua styracifluaSH F SS F SH F SS F SHF SSF

    Fig. 2. Comparisons of final yields as a percentage of the theoretical maximum product potential forsaccharification only (SHF) and SSF of selected hardwoods and herbaceous crops. Cellulase loadings of7, 13, 19, and 26 IU g- cellulose are employed without (0: 1) and with (8: I) /I-glucosidase supplemen-

    tation.

    than for the legume, S. Z espedeza, for all enzymelevels. It is also apparent that higher yields areobserved for weeping love grass and generallyfor switchgrass than for any of the woodycrops at lower cellulase loadings without p-glucosidase supplementation. The fermentationwith the mixed culture particularly excels atlow cellulase usage without /?-glucosidase ad-ditions.

    Similar trends are observed for corn cobs,corn stover, and wheat straw . However, it isevident that the yields for corn cobs are the bestof all of the substrates evaluated, followedclosely by corn stover. With high cellulase load-ings and fi-glucosidase supplemen tation, yieldsof up to 96% are realized with corn cobs andstover. However, even at low cellulase levels of

    7 IU g- without additional j-glucosidase,yields of about 75% are achieved with the mixedculture while values of 5458% are attainedwith S. ce rev i s iae alone. The performance forwheat stra w was generally slightly better thanor equivalent to the herbaceous crops butnot as good as for corn cobs or stover through-out.All of the substrates tested reached thehighest conversions of cellulose to ethanol withS. ce rev i s iae at higher cellulase enzyme loadingswith B-glucosidase added. The mixed culture ofS. ce rev i s iae an d B. c lausen i i was slightly lesssensitive to b-glucosidase supplemen tation.This observation can be explained by pro-duction of fl-glucosidase during SSFs by thecellobiose-fermenting yeast B. c luusen i i while

    100, I ??26??19w 13m-f

    % Yield

    Corn co b Corn cob Wheat straw Wheat strawSHF SSF SH F SS FFig. 3. Comparison of final yields as a percentage of the theoretical maximum product potential forsaccharification only (SHF) and SSF of corn cobs and wheat straw. Cellulose loadings of 7, 13, 19, and

    26 IU gg cellulose are used without (0: 1) and with (8: 1) j?-glucosidase supplementation.

  • 7/28/2019 Simultaneous Saccharification and Fermentation of Several Lignocellulosic

    6/7

    306 C. E. WYM AN et al .S. cerevis iue alone can achieve similar resultswith /I-glucosidase supplemen tation.

    It is apparent that the SSF process giveshigher yields than straight saccharification if wecompare the final yields of both Tables 1 and2. This comparison is shown in Figs 2 and 3.The increased yield is because of the low re-sidual sugars present in the SSF from yeastfermentation that relieve inhibition of the cellu-lase enzyme experienced in straight saccharifica-tion.

    4. CONCLUSIONS

    From our data, we can conclude that corncobs, corn stover, wheat straw, and weepinglove grass respond well, in descending order, todilute acid pretreatment; achieve fast enzymatichydrolysis rates; and give high ethanol yields bythe SSF process. Switchg rass performed some-what more poorly. All the pretreated woodycrops reached similar final conversions of cellu-lose to ethanol in the SSF process, bu t the finalyields and rates of hydrolysis for P. ma x i m o w -i c z i i were the highest for S. ce rev i s iue with orwithout fi-glucosidase supplemen tation. How-ever, the mixed culture of S. ce rev i s iae an d B.c luusen i i generally achieved higher final yields atlower cellulase loadings without /I-glucosidasesupplementation. Only the legume S. lespedezadid not respond as well to the dilute ac idpretreatment conditions used for the other twograsses, and the conversion of cellulose tosugars or ethanol was much less for this sub-strate. However, it may be possible to increasethe yields by alteration of the pretreatmentconditions. On the other hand , corn cobs andstover exhibited the best performance in theSSF process of any substrate we have evalu-ated.Without /I-glucosidase supplementation atlow cellulase loadings, the overall rates andyields were generally better for the mixed cul-ture than for S. ce rev i s iae because of the ad-ditional p-glucosidase activity produced by B.c luusen i i . However, S. ce revk iae performedabout the same with substantial /I-gIucosidaseaddition and/or high cellulase loadings. Theseresults point out the benefit of providing proper/.I-glucosidase levels to prevent accumulation ofthe powerful inhibitor cellobiose. Thus, the useof a mixed culture of a cellobiose-fermentingorganism w ith an ethanol-tolerant strain orsupplementation of the enzyme broth with b-

    glucosidase will continue to be desirable toobtain high yields until a cellulase enzyme thathas higher /I-glucosidase activity or is lessinhibited by cellobiose is developed or a moreethanol-tolerant cellobiose-fermenting microor-ganism is found.Acknow ledgements-This research was supported by theEthanol from Biomass Program of the DO E Biofuels Sys-tems Division.

    1. J. D. Wright, Ethanol from lignocellulose: an overview.Energy Prog. 8(2), 71 (1988).

    2. J. D. Wright, Ethanol from biomass of enzymatichydrolysis.Chem. Eng. Prog. 62, August (1988):3. J. D. Wrieht. C. E. Wvman and K. Grohmann. Simul-

    4.

    5.

    6.

    7.

    8.

    9.

    10 .

    11 .

    12 .

    13 .

    14 .

    15 .

    16 .

    taneous s&harification and fermentation of lig&cellu-lose: process evaluation. App. Bi ochem. Biotechnol. 18,75 (1988).W. F. Gauss, S. Suzuki and M. Takagi, M anufacture ofalcohol from cellulosic materials using plural ferments.U.S. Pat. No. 3,990,944, November 9,-1976.M. Takaai. S. Abe. S. Suzuki. G. H. Emert and N. Yata.A meth&l for production of alcohol directly fromcellulose using cellulasc and yeast. Proc. BioconwrsionSymp. IlT, Delhi, India, 55i (1977).P. J. Blotkamo. M. Takaai. M. S. Pemberton and G. H.Enert, Enzymatic hydr&sis of cellulose and simul-taneous saccharification to alcohol. AI ChE Symp . Ser.74, 85 (1978).M. Holtzapple, Inhibition of Trichoderma reesei cellu-lase by sugars and solvents, American Institute ofChemical E ngineers 1 989, Annual Meeting, San Fran-cisco, CA, November 8, 1989.M. Takag i, Inhibition of cellulase by fermentationproducts. Bi ot echnol . B io eng. 24, 1506 (1984).S. M. Lastick. D. D. Snindler and K. Grohmann.Fermentation of solubi cellooligosaccharides byyeasts. Wood Agri c. Residues (E. J. Soltes, Ed.), 239 pp.(1983).S. M. Lastick, D . D. Spindler, S. Terre1 and K.Grohman n, Simultaneous sacchatitication and fetmen-tation of cellulose. Biotechnology 84, 277 (1984).J. Szczodrak, The use of cellulases from a j-glucosidaschyperproducing mutant of Trichoderma reesei in simul-taneous saccharification and fermentation of wheatstraw. Biorechnol Bioeng. 9, 1112 (1989).J. Szczorak and Z . Targonski, Selection of thermotoler-ant yeast strains for simultaneous saccharification andfermentations of cellulose. Bi ot echnol . Bioeng. 31, 300(1988).J. N. Saddler, M . Mes-Hartree, E. K. C. Yu and H. H.Brownell, Enzymatic hydrolysis of various pretreatedlignocellulosic substrates and the fermentation of theliberated sugars to ethanol and butanediol. Biotechnol.Bi oeng. Symp. 13, 225 (1983).J. N. Saddler, Conversion of Cellul ose to Ethanol Usinga Tw o-Stage Process. Forintek Canada Corn., Ottawa,Ontario, &nada, 45, May (1982). _J. N. Saddler, C. Hoagen, M. K. Chan and G. Louis-Seize, Ethanol fermentations of enzymatically hy-drolyzed pretreated wood fractions using Trichodermacellulase, Zymomonas mobi l i s and Saccharomycea cere-oi siae. Can. J. Mcrobi ol . 28, 1311 (1982).D. B. Rivers, G. M. Zanin and G. H. Emert, Evaluationof sugar cane bagasse and rice straw as process sub-strates for production of ethyl alcohol. P roc. ArkansasAcad. Sci. 38, 95 (1984).

  • 7/28/2019 Simultaneous Saccharification and Fermentation of Several Lignocellulosic

    7/7

    Simultaneous saccharification and fermentation 30 717. D. D. Spindler, C. E. Wym an, A. Mohagheghi and K.Grohman n, Thermotolerant yeast for simultaneous sac-charification and fermentation of cellulose to ethanol.

    Appl . Biochem. Biot echnol. 17, 279 (1988).18. D. D. Spindler, C. E. Wyman, K. Grohmann and A.Mohagheghi, Simultaneous saccharification and fer-mentation of pretreated wheat straw to ethanol w ithselected yeast strains and fl-glucosidase supplem en-tation. Appl . Biochem. Biot echnol. 20/21, 529 (1989).19. D. D. Spindler, C. E. Wyman and K . Grohmann,Evaluation of thennotolerant yeasts in controlled simul-taneous saccharifications and fermentation of celluloseto ethanol. Bi ot echnol . Bi oeng. 34(2), 189 (1989).

    20. C. E. Wyman, D. D. Spindler, K. Grohmann and S. M.Lastik, Simultaneous saccharification and fermentationof cellulose with the yeast Brettanomyces clausenii.Bi ot echnol . Bi oeng. 17, 221 (1986).

    21. J. H. Janssen, N. Burris, A. Woodw ard and R. B.Bailey, Lipid enhanced ethanol production by Kluyvero-myces ragil is. Appl . Envir on. M icrobi al. 4!5,598 (1983).

    22. T. K. Ghose, Measurements of cellulase activities. PureAppl . Chem. 59, 257 (1987).23. R. Torget, P. Walter, M. Himmel and K. Grohmann,Dilute acid pretreatment of short rotation woody andherbaceous crops. Appl . Biochem. Bi otechnol. 24/S, I 15(1990).