Doi:10.1016/s0141-0229(03)00087-5

Enzyme and Microbial Technology 33 (2003) 71–78 The effect of Tween-20 on simultaneous saccharification and Malek Alkasrawi , Torny Eriksson , Johan Börjesson , Anders Wingren , Mats Galbe , Folke Tjerneld , Guido Zacchi a Department of Chemical Engineering 1, Lund University P.O. Box 124, SE-221 00 Lund, Sweden b Department of Biochemistry, Lund University P.O. Box 124, SE-221 00 Lund, Sweden Received 18 December 2002; received in revised form 25 March 2003; accepted 28 March 2003 Abstract
Simultaneous saccharification and fermentation (SSF) of steam-pretreated wood constitutes an attractive process configuration for ethanol production from biomass. However, the high enzyme addition in SSF contributes to a high process cost. In this study we explorethe effect of the non-ionic surfactant Tween-20 as an additive in SSF. Tween-20 addition at 2.5 g/l had several positive effects on SSF: (i)the ethanol yield was increased by 8%; (ii) the amount of enzyme loading could be reduced by 50%, while maintaining a constant yield;(iii) the enzyme activity increased in the liquid fraction at the end of SSF, probably by preventing unproductive binding of the cellulases tolignin, which could facilitate enzyme recovery; (iv) the time required to attain maximum ethanol concentration was reduced. Surfactantsas an additive in SSF can significantly lower the operational cost of the process. However, less expensive surfactants must be investigated.
2003 Elsevier Science Inc. All rights reserved.
Keywords: Ethanol production; Softwood; SSF; Enzyme recovery; Tween-20 1. Introduction
and fermentation (SHF) in terms of higher final ethanol yieldand lower energy consumption. However, SSF still requires The production of fuel ethanol from biomass as a substi- high enzyme concentrations and operates at a non-optimal tute for fossil fuels is still expensive. Recent economical cal- hydrolysis temperature, which influences the hydrolysis ef- culations showed that the production cost would be higher than the price of gasoline Lignocellulosic materials are Enzyme loading constitutes a significant contribution considered to be the most promising renewable resource for to the overall ethanol production cost in processes based fuel ethanol production. Softwood, such as spruce, would on enzymatic conversion of cellulose It is therefore provide an abundant feedstock in countries such as Sweden.
important to find methods of reducing the enzyme loading The production of fuel ethanol based on steam pretreatment and increasing the conversion rate. Earlier and recent in- of spruce followed by enzymatic hydrolysis and fermenta- vestigations suggest that the addition of surfactants to the enzymatic hydrolysis of cellulose may increase the conver- Improvements resulting in lower operation costs in the sion rate of cellulose to glucose Surfactant addition has ethanol production process can be achieved by modification been proven to increase the enzymatic hydrolysis of several of process parameters. Optimising substrate enzyme different cellulose-containing substrates. Castanon et al.
will increase the final ethanol yield. Pro- wed that the hydrolysis of newspaper increased by cess integration such as process stream recirculation will 14% after 48 h with the addition of Tween-80. The positive decrease the energy consumption Simultaneous sac- effect of surfactant addition has also been observed using charification and fermentation (SSF) has been found to be Avicel, tissue paper delignified steam-exploded wood economically favourable compared with separate hydrolysis Non-ionic surfactants have been proven to be effective in increasing cellulose hydrolysis, whereas anionic and cationic Corresponding author. Tel.: +46-46-222-82-97; surfactants decrease cellulose hydrolysis The ad- E-mail address: guido.zacchi@kath.lth.se (G. Zacchi).
sorption of cellulases to the solid cellulose substrate has been 0141-0229/03/$ – see front matter 2003 Elsevier Science Inc. All rights reserved.
doi:10.1016/S0141-0229(03)00087-5 M. Alkasrawi et al. / Enzyme and Microbial Technology 33 (2003) 71–78 shown to decrease in the presence of non-ionic surfactants The mechanism underlying the enhancement of The composition of spruce and steam-pretreated material enzymatic cellulose hydrolysis has been the object of intense research and speculation. Various mechanisms have been proposed, which can be divided into three main categories: (i) the surfactant changes the ultra structure of the substrate, making the cellulose more available to enzymatic attack (ii) surfactants increase enzyme stability, i.e. by reducing thermal denaturation or denaturation by shear forces and (iii) surfactants affect enzyme–substrate interaction, e.g.
adsorbed enzymes are prevented from inactivation by sur- factant addition In a recent publication from our laboratory, we proposed the dominating effect of surfac- tant on spruce lignocellulose hydrolysis, is enzyme interac- tion with lignin surfaces. Surfactant adsorption onto lignin prevents unproductive binding of enzymes to lignin. In the same study we found that it was possible to reduce the en- zyme loading by 50% by adding Tween-20 at 2.5 g/l n the present study, we focused on the influence of Tween-20 on the performance of SSF, in terms of final ethanol yield and residence time. We also investigated the possibility of reducing the enzyme loading, while maintaining the sameethanol yield, through the addition of Tween-20. Addi- tionally, the influence of Tween-20 on the activity of freeenzymes recovered from the liquid fraction after SSF was The enzymes used were Novozym 188 and Cellu- 2. Materials and methods
clast 1.5L. Novozym 188 had a ␤-glucosidase activity of535 IU/g, measured according to the method described by Berghem and Pettersson Celluclast 1.5L had a cellu-lase activity of 104 FPU/g measured according to Mandel Fresh chipped softwood of spruce was kindly provided by et al. a ␤-glucosidase activity of 38 IU/g. Cellu- a sawmill in southern Sweden (Htöörsågen AB, Höör). The clast 1.5L and Novozym 188 were kindly provided by Novo softwood was grinded and sieved to obtain 2–10 mm chip size. The chips were stored in a plastic bag at 4 ◦C prior touse. The composition (see was analysed using the Hägglund method wed by HPLC analysis. The drymatter (DM) content was determined to be 52% (w/w).
Filter paper activity in the liquid fraction samples from SSF was measured according to the method described by Mandel et al. with the following modifications:Novozym (0.1 IU/ml) was added to ensure excess of The 3.6 kg spruce chips were impregnated with sulphur ␤-glucosidase activity in all measurements and bovine dioxide (3%, w/w, moisture) for 20 min at room temperature.
serum albumin (BSA; 0.1 mg/ml) was added in order to Impregnation was performed in a tightly sealed plastic bag avoid non-specific binding of enzymes.
to allow penetration of the gas into the wood tissues. Theamount of SO2 absorbed was determined by weighing the 2.5. Fermentation of steam pretreatment hydrolysate plastic bag before and after impregnation, and was found tobe having 2.5% (w/w) moisture. The impregnated material This experimental part was performed to assess the hy- was pretreated with saturated steam at 215 ◦C for 5 min in a drolysate fermentability and the influence of Tween-20 on steam pretreatment unit equipped with a 10-l reactor, which the yeast performance without the presence of the enzymes has been described previously Steam pretreatment of keeping in mind that the operating fermentation must be spruce was performed in six batches, by loading 600 g dry the same as in SSF. This was also done to give an idea how weight in the reactor for each batch. Slurry from all the the yeast would perform in SSF. In the hydrolysate fermen- batches were mixed together and stored at 4 ◦C for further tation experiments, the slurry obtained after the steam pre- treatment step was filtered, and the solid part was removed.
M. Alkasrawi et al. / Enzyme and Microbial Technology 33 (2003) 71–78 The liquid that contains free sugars was diluted to the same using 5 mM H2SO4 as eluent, at a flow rate of 0.5 ml/min.
concentration as that obtained when the slurry was diluted Since mannose, xylose and arabinose were eluted at the to 5% fibrous material as in SSF. The liquid was diluted same time, these sugars were determined using a PL Hi-Plex to alleviate the inhibitory effect of formed inhibitors dur- Pb column (Polymer Labs, Shropshire, UK) at 80◦ C, using ing steam pretreatment. Since the pretreated material was ultrapure water as eluent at a flow rate of 0.5 ml/min. The stored at 4 ◦C, the diluted liquid was sterilised at 120 ◦C for DM content of the pretreated material and the yeast were 20 min and fermented using the same operating parameters determined by drying the materials overnight at 105 ◦C.
and the same amount of nutrients as in SSF, (see the follow-ing). Fermentation was performed with and without 2.5 g/lTween-20. In the evaluation of the liquid fermentation ex- 3. Results and discussion
periments, the ethanol yield was expressed as a percentageof the theoretical yield based on the amount of fermentable Tween-20 has been recognised in several studies to be an sugars (glucose and mannose) present in the liquid.
enhancer of enzymatic cellulose hydrolysis but few report about its ability to enhance the fermentation of cellulosic hydrolysates to ethanol. According to Ballesteroset al. actant addition to SSF of steam-pretreated All SSF runs were performed with a working weight of poplar substrate increased the final ethanol yield without 750 g in 1-l fermentors (Belach AB, Stockholm, Sweden), affecting the yeast viability. Previous work on surfactant under semi-sterile conditions since it is not possible to ster- effects on SSF of steam-exploded poplar has shown that the ilize the enzyme. The slurry, consisting of the pretreated ethanol yield can be increased by 6% by Tween-80 addition material, was diluted to 5% DM (insoluble solids) using However, the possibility of reducing enzyme loading fresh water. The diluted slurry and the yeast nutrients were by surfactant addition was not studied. Ooshima et al. autoclaved at 120 ◦C for 20 min. The enzyme preparations showed that the rate of SSF of pure cellulose (Avicel) was were added directly to the fermentor vessel.
slightly enhanced by the addition of Tween-20.
All the runs were performed at 37 ◦C, at pH 5, for 72 h or longer up to 144 h. The initial pH value was adjusted to 4.9–5.1 using solid Ca(OH)2, and during fermentation thepH was maintained at pH 5 by the addition of 10% NaOH.
gives the composition of the raw material and The concentration of Novozym was 4% (w/w) fibrous ma- the steam-pretreated material. The DM content of fibrous terial, which corresponds to a ␤-glucosidase activity of material (insoluble matter) in the pretreated material was 41 IU/g cellulose. The amounts of Celluclast 1.5L used were 13%. The recovery of glucose and mannose in the liquid 24, 12 or 6% (w/w) fibrous material, which corresponds to fraction were 28.6 and 81.4% of the theoretical content of cellulase activities of 44, 22 or 11 FPU/g cellulose, respec- glucan and mannan in the raw material, respectively. The tively. Normal baker’s yeast, Saccharomyces cerevisiae (S. recovery of glucan in the solid fraction was 70% of the cerevisiae) was inoculated at 0.5% (dry, w/w) of the total working weight. It has not escaped our attention that baker’syeast contains lactic acid bacteria and therefore antibiotics were added; Penicillin (20,000 U/ml) and Streptomycin(20 mg/ml) This was done to ensure that changes in Fermentation of the liquid fraction after pretreatment was the ethanol yield were not due to competing microorgan- performed with and without surfactant addition. The analy- isms. The composition of the supplemented yeast nutrients sis showed that the ethanol yield was 85% of the theoretical without Tween-20. This increased to 95% of the theoretical (0.025 g/l) and yeast extract (1.0 g/l). The ethanol yield was after 24 h upon adding 2.5 g/l Tween-20. This increase cor- expressed as percentage of the theoretical yield based on responds to an 11.8% increase in the final ethanol yield as the amount of fermentable sugars (glucose and mannose) a result of Tween-20 addition (This may be due to facilitated substrate uptake by the yeast and easing the masstransport 3.3. SSF with different concentrations of Tween-20 Samples taken from the raw material hydrolysis, the liquid fraction of the pretreatment step, the liquid fermentation In all SSF experiments the sum of glucose and mannose and the SSF were analysed using HPLC (Shimadzu, Kyoto, concentrations was approximately 23 g/l at the start. Man- Japan) equipped with a refractive index detector (Shimadzu).
nose and glucose were rapidly depleted. Furfural and HMF Ethanol, glucose, furfural, hydroxymethylfurfural (HMF), had completely disappeared by the end of SSF. In the ref- acetic acid and glycerol were analysed using an Aminex erence run (base case) without surfactant, with a Celluclast HP-87H column (Bio-Rad, Hercules, CA, USA) at 65 ◦C, loading of 44 FPU/g cellulose (24% of the insoluble fibrous M. Alkasrawi et al. / Enzyme and Microbial Technology 33 (2003) 71–78 The influence of Tween-20 on hydrolysate fermentation and SSF The influence of enzyme loading with and without the addition of 2.5 g/lTween-20 on the theoretical ethanol yield in SSF, and on the residence time required to reach the maximum ethanol concentration material), the highest ethanol concentration was reached af- ter about 72 h. The final ethanol yield was 86% of the the-oretical yield and the initial productivity based on the first The residence time required to reach the maximum ethanol concentration did not change with the addition of 1.25 g/l The addition of 1.25 and 2.5 g/l Tween-20 increased the Tween-20, but decreased to 48 h with the addition of 2.5 g/l ethanol yield by 8% compared with the reference run (base Tween-20 (Therefore, 2.5 g/l Tween-20 was used case). The initial productivity remained at the same level as an optimal concentration for the subsequent SSF runs.
(1.4 g/l/h) with the addition of 1.25 g/l Tween-20, but in- The rate-limiting step in the later stages of SSF is cellulose creased to 1.6 g/l/h when adding 2.5 g/l Tween-20 conversion, since the concentration of fermentable sugars Fig. 1. The influence of 2.5 g/l Tween-20 addition on the ethanol production in SSF with an enzyme loading of 44 FPU/g cellulose. SSF runs withTween-20 (—), and without Tween (----).
M. Alkasrawi et al. / Enzyme and Microbial Technology 33 (2003) 71–78 Fig. 2. The influence of 2.5 g/l Tween-20 addition on the ethanol production in SSF with an enzyme loading of 22 FPU/g cellulose. SSF runs withTween-20 (—), and without Tween (----).
is negligible in the solution after 24 h of fermentation and runs. Similar SSF runs were performed but with the addition complete cellulose conversion is known to require long hydrolysis times Therefore, it is likely that Tween-20 addition improved cellulose conversion, which is in agree- without Tween-20. At high enzyme loading (44 FPU/g cel- ment with our earlier results on the effect of Tween-20 on lulose) the final ethanol yield increased by 8% following enzymatic conversion using the same substrate It is the addition of Tween-20. The maximum ethanol concentra- also possible that the increased yield of ethanol was partly tion was reached after 48 h when Tween-20 was added com- due to the improved performance of the yeast. This is sup- pared with 72 h without the addition of Tween-20 ( ported by the results of fermentation of the pretreatment’s At a lower enzyme loading of 22 FPU/g cellulose, the addi- hydrolysate, which were better when Tween-20 was added tion of Tween-20 (2.5 g/l) increased the final ethanol yield in terms of final ethanol yield and also in the production rate by 17% compared with SSF without Tween-20. The res- of ethanol for the first 24 h (A higher concentration idence time required to reach the final maximum ethanol of Tween-20 (5 g/l) decreased the yield by 60%, and this concentration was 72 h whereas 96 h were required without is probably because the Tween-20 becomes toxic at higher Tween-20 addition (At the lowest enzyme loading (11 FPU/g cellulose) Tween-20 increased the final ethanolyield by 22%. The final ethanol yield was reached after 144 h 3.4. SSF with different enzyme concentrations without Tween-20 and after 120 h with Tween-20 Tween-20 addition to SSF increased the final ethanol yield at Three SSF runs were performed using different enzyme all enzyme concentrations, The increased ethanol loadings: 24, 12 and 6% Celluclast of the fibrous material yield is probably due to improved fermentation by the yeast corresponding to cellulase activities of 44, 22 and 11 FPU/g in combination with increased cellulose conversion. The cellulose. The Novozym (␤-glucosidase) concentration was maximum ethanol yield was reached earlier when Tween-20 constant (4% of the insoluble fibrous material) in all SSF M. Alkasrawi et al. / Enzyme and Microbial Technology 33 (2003) 71–78 Fig. 3. The influence of 2.5 g/l Tween-20 addition on the ethanol pro-duction in SSF with an enzyme loading of 11 FPU/g cellulose. SSF runswith Tween-20 (—), and without Tween (----).
SSF with 22 FPU/g cellulose and 2.5 g/l Tween-20 reached the same final ethanol yield at the same resi-dence time as when 44 FPU/g cellulose was used withoutTween-20. This finding will have a significant economicaleffect on the final ethanol cost since the cost of enzymesconstitutes a significant part of the total process cost Another potential economical benefit of Tween-20 addi- tion could be the shortening of residence time for SSF. Themaximum ethanol concentrations without Tween-20 addi-tion using an enzyme loading of 44 and 22 FPU/g cellulosewere reached at 72 and 96 h, respectively. The addition of Fig. 4. The influence of 2.5 g/l Tween-20 addition on the filter paper Tween-20 resulted in the shortening of the residence times activity in the SSF liquid fraction at: (a) 48 h and (b) at the time required from 72 to 48 h and from 96 to about 72 h. At low en- zyme loading (11 FPU/g cellulose), the maximum ethanolconcentration without Tween-20 was not reached within paper activity was also measured at the time when the max- 144 h when Tween-20 was added, the maximum ethanol imum ethanol yield was reached (In all measured concentration was reached after 120 h. The final yield was samples, the remaining enzyme activity in the liquid frac- similar to when the enzyme loading was four times higher tion was higher when Tween-20 was added (It has (44 FPU/g cellulose), but without Tween-20. However, us- been suggested that the addition of surfactants can prevent ing this low enzyme loading might not be suitable due to enzymes from becoming unproductively bound to lignin Therefore, the addition of a surfactant could increasethe possibility of recycling the enzymes after completion 3.5. Remaining filter paper activity in the liquid of cellulose hydrolysis. The cellulase activity in the SSF samples taken at times when the maximum ethanol concen-tration was reached, i.e. when most of the cellulose has been In order to evaluate the influence of Tween-20 on the hydrolysed (supports the previous findings that sur- amount of remaining cellulase activity in the liquid fraction factants reduce the binding of enzymes to lignin The after SSF, filter paper activity unit (FPU) was determined surfactant effect on cellulase adsorption was higher at high for selected samples during SSF. FPU was measured at enzyme concentration. This could be explained by the high 48 h for all six SSF runs (The remaining filter degree of conversion obtained with high enzyme loading, M. Alkasrawi et al. / Enzyme and Microbial Technology 33 (2003) 71–78 resulting in a lower amount of residual cellulose. Thus, the enzymes against the cost of Tween-20 (or some other surfac- proportion of lignin in the substrate was increased, leading tant with the same effect on SSF). In the comparison, it was to the release of more enzymes by surfactant addition.
assumed that the overall yield of ethanol, as well as the rateof fermentation were the same, which offset the effect of all 3.6. The effect of using Tween-20 on the process cost other costs such as those of raw material, capital and utilities.
These costs will thus be the same per litre of ethanol pro- The addition of Tween-20 will add an extra cost to the duced. This is a reasonable assumption when comparing the process, i.e. the cost of the Tween-20 itself, but on the other case with an enzyme activity of 44 FPU/g cellulose (case 1) hand, other process costs will be reduced. The enzyme cost without Tween-20 and the case where Tween-20 was added will be reduced due to the fact that the enzyme concentration and the enzyme activity was lowered to 22 FPU/g cellulose can be lowered while maintaining the same ethanol yield in (case 2). Although the case with 11 FPU/g (case 3) cellu- SSF due to the presence of Tween-20. It has been shown that lose and the addition of Tween-20, exhibited a lower rate the presence of Tween-20 may increase the rate of fermenta- of fermentation this case was included assuming the same tion. Thus, another option could be to shorten the residence residence time and ethanol yield as the base case.
time in the SSF, while maintaining almost the same ethanol The purpose of this economic evaluation was to calculate yield. This will cause the total fermentation volume and thus the maximum permissible cost of Tween-20, for the total the capital cost for the SSF step to decrease. The SSF step cost of the Tween-20 and enzymes to be the same as the cost is one of the most expensive for this kind of process It of the enzymes in the case with 44 FPU/g cellulose without might also be possible to recycle more of the enzymes due Tween-20. Thus, a cost was assumed for the enzymes and to the fact that higher activities of cellulases were present in then the maximum cost of Tween-20 for the cases with a low- the liquid fraction after SSF with Tween-20 than in the case ered enzyme load were calculated. The cost of the enzymes without Tween-20 (see Enzyme recycling could be can be regarded as the cost of producing the enzymes at the combined with process stream recirculation before distilla- ethanol plant or the cost of purchasing them from an enzyme tion as we have suggested 14 previously which would company. The cost was chosen to be in line with other stud- reduce the production cost further. The most probable sce- ies calculations were performed for a case assuming nario will be a combination of the improvements discussed no recycling and for a case assuming that the enzymes could be recycled. In the latter case all the activity remaining in the A complete economic evaluation of the process including liquid phase after SSF according to recovery) the capital cost, must be made to determine the overall im- was assumed to be recycled, by adsorption onto fresh sub- pact of the addition of Tween-20 on the production cost of strate, at no cost, which reduces the need for fresh enzymes.
ethanol. At this stage a limited economic evaluation is suffi- The results are shown in If the enzyme loading is re- cient since the experiments were carried out on a small scale.
duced to 11 FPU/g cellulose, i.e. case 3, the maximum allow- Thus, we chose to study only the effect of the cost of the able price for Tween-20 is roughly 3.3 SEK/kg if the cost of Fig. 5. Acceptable cost of detergent as a function of the cost of the enzymes per kilogram of ethanol produced.
M. Alkasrawi et al. / Enzyme and Microbial Technology 33 (2003) 71–78 the enzymes is 0.5 SEK/kg ethanol produced. If it is possible [4] Stenberg K, Tengborg C, Galbe M, Zacchi G, Palmqvist E, to recover all the activity in the liquid phase, the maximum Hahn-Hagerdal B. Recycling of process streams in ethanol production cost of Tween-20 can be increased to roughly 4.1 SEK/kg.
from softwood based on enzymic hydrolysis. Appl BiochemBiotechnol 1998;70–72:697–708.
As mentioned above, this calculation did not take into ac- [5] Larsson M, Galbe M, Zacchi G. Recirculation of process water count the capital costs or the overall yield of ethanol. More in the production of ethanol from softwood. Bioresour Technol detailed studies are needed to determine the process condi- tions that would result in the lowest ethanol production cost.
[6] Alkasrawi M, Galbe M, Zacchi G. Recirculation of process streams in fuel ethanol production from softwood based onsimultaneous saccharification and fermentation. Appl BiochemBiotechnol 2002;98–100:849–61.
4. Conclusions
[7] Wooley R, Ruth M, Sheehan J, Ibsen K. Lignocellulosic biomass to ethanol process design and economics utilizing co-current dilute The SSF of steam-pretreated softwood was improved acid prehydrolysis and enzymatic hydrolysis current and futuristic by the addition of Tween-20 due to a combination of in- scenarios. Technical report, NREL/TP-580-26157; (1999).
[8] Gregg DJ, Boussaid A, Saddler JN. Techno-economic evaluation of a creased hydrolysis rate and improved yeast fermentation.
generic wood-to-ethanol process: effect of increased cellulose yields The exact mechanisms of Tween-20 as an enhancer for and enzyme recycle. Bioresour Technol 1998;63:7–12.
both enzyme hydrolysis and sugar fermentation are not [9] Helle SS, Duff SJB, Cooper DG. Effect of surfactants on cellulose known. Tween-20 probably plays an important role in pre- hydrolysis. Biotechnol Bioeng 1993;42(5):611–7.
venting the unproductive binding of the cellulases to the 80 on enzymatic hydrolysis of newspaper. Biotechnol Bioeng lignin residues, allowing more enzyme to be available for the conversion of cellulose, resulting in a higher conversion [11] Ooshima H, Sakata M, Harano Y. Enhancement of enzymatic-hy- rate. This hypothesis is supported by the higher activity of drolysis of cellulose by surfactant. Biotechnol Bioeng 1986;28:1727– the free cellulases detected in the liquid fraction of SSF.
Concerning sugar fermentation, 16 surfactants in general [12] Kurakake M, Ooshima H, Kato J, Harano Y. Pretreatment of begasse by nonionic surfactant for the enzymatic hydrolysis. Bioresour are well known for their positive effect on any fermentation [13] Kaar WE, Holtzapple M. Benefits from Tween during enzymic The enzyme loading could be reduced by as much as hydrolysis of corn stover. Biotechnol Bioeng 1998;59:419–27.
50% without a reduction in ethanol yield. This has potential [14] Eriksson T, Börjesson J, Tjerneld F. Mechanism of surfactant effect economical implications since the enzyme cost is a major in enzymatic hydrolysis of lignocellulose. Enzyme Microb Technol2002;31(3):353–64.
contributor to the process cost, taking into consideration the [15] Hägglund E. Chemistry of wood. New York: Academic Press; 1951.
fact that the price of Tween-20 is much lower than the price of enzymes. The Tween-20 addition decreases enzyme ad- [16] Berghem L, Pettersson G. The mechanism of enzymatic cellulose sorption onto the residual solids after SSF. This could make degradation isolation and some properties of a b-glucosidase from enzyme recovery from the liquid fraction after SSF possible.
Trichoderma viride. Eur J Biochem 1974;46:295–305.
[17] Mandels M, Andreotti R, Roche C. Measurement of saccharifying This work will be continued, with the aim of finding cheaper cellulase. Biotech Bioeng Symp 1976;6:21–33.
surfactant with the same positive effects as Tween-20.
[18] Stenberg K, Tengborg C, Galbe M, Zacchi G. Optimization of steam pretreatment of SO2-impregnated mixed softwoods for ethanolproduction. J Chem Technol Biotechnol 1998;71:299–308.
Acknowledgments
[19] Stenberg K, Galbe M, Zacchi G. The influence of lactic acid for- mation on the simultaneous saccharification and fermentation (SSF) The Swedish National Energy Administration is gratefully of softwood to ethanol. Enzyme Microb Technol 2000;26:71–9.
[20] Kim MH, Lee SB, Ryu DDY. Surface deactivation of cellulase and its prevention. Enzyme Microb Technol 1982;4:99–103.
[21] Park JW, Takahata Y, Kajiuchi T, Akehata T. Effects of nonionic surfactant on enzymatic-hydrolysis of used newspaper. Biotechnol References
[22] Ballesteros I, Olivia JM, Carrasco J, Cabanas A, Navarro AA, [1] von Sivers M, Zacchi G. Ethanol from lignocellulosics: a review of Ballesteros M. Effect of surfactant and zeolites on simultaneous the economy. Bioresour Technol 1996;56:131–40.
saccharification and fermentation of steam-exploded poplar biomass to ethanol. Appl Biochem Biotechnol 1998;70–72:369–81.
Effect of substrate and cellulase concentration on simultaneous [23] Lee WG, Lee JS, Lee JP, Shin CS, Kim MS, Park SC. Effect saccharification and fermentation of steam-pretreated softwood for of surfactants on ethanol fermentation using glucose and cellulosic ethanol production. Biotechnol Bioeng 2000;68(2):204–10.
hydrolyzates. Biotechnol Lett 1996;18(3):299–304.
[3] Tengborg C, Galbe M, Zacchi G. Influence of enzyme loading and [24] Hinman ND, Schell DJ, Riley CJ, Bergeron PW, Walter PJ.
physical parameters on the enzymatic hydrolysis of steam-pretreated Preliminary estimate of the cost of ethanol production for SSF softwood. Biotechnol Progr 2001;17:110–7.
technology. Appl Biochem Biotechnol 1992;34/35:639–49.

Source: http://www3.lut.fi/webhotel/teke/tktpl/2013/artsut/elomaa.pdf