Vous êtes sur la page 1sur 6

Process Biochemistry 42 (2007) 834–839

www.elsevier.com/locate/procbio

A comparison between simultaneous saccharification and fermentation and


separate hydrolysis and fermentation using steam-pretreated corn stover
Karin Öhgren a, Renata Bura b, Gary Lesnicki c, Jack Saddler b, Guido Zacchi a,*
a
Department of Chemical Engineering, Lund University, P.O. Box 124, SE-22100 Lund, Sweden
b
Forest Products Biotechnology, Department of Wood Science, 4th Floor, Forest Science Centre, 4042-2424 Main Mall,
University of British Columbia, Vancouver, BC, Canada
c
Michael Smith Laboratories, Fermentation Pilot Plant Facility, 2185 East Mall, University of British Columbia, Vancouver, BC, Canada
Received 10 July 2006; received in revised form 19 January 2007; accepted 9 February 2007

Abstract
Two different process configurations, simultaneous saccharification and fermentation (SSF) and separate hydrolysis and fermentation (SHF),
were compared, at 8% water-insoluble solids (WIS), regarding ethanol production from steam-pretreated corn stover. The enzymatic loading in
these experiments was 10 FPU/g WIS and the yeast concentration in SSF was 1 g/L (dry weight) of a Saccharomyces cerevisiae strain. When the
whole slurry from the pretreatment stage was used as it was, diluted to 8% WIS with water and pH adjusted, SSF gave a 13% higher overall ethanol
yield than SHF (72.4% versus 59.1% of the theoretical). The impact of the inhibitory compounds in the liquid fraction of the pretreated slurry was
shown to affect SSF and SHF in different ways. The overall ethanol yield (based on the untreated raw material) decreased when SSF was run in
absence on inhibitors compared to SSF with inhibitors present. On the contrary, the presence of inhibitors decreased the overall ethanol yield in the
case of SHF. However, the SHF yield achieves in the absence of inhibitors was still lower than the SSF yield achieves with inhibitors present.
# 2007 Elsevier Ltd. All rights reserved.

Keywords: Steam pretreatment; SSF; SHF; Ethanol production; SO2; Corn stover; Ethanol concentration

1. Introduction Bio-ethanol can be produced from any biomass, thus access to


raw material is virtually unlimited. For example, agricultural by-
Ethanol, produced from sugar, starch and lignocellulosic products (straw, sugar cane bagasse, stover) provide a readily
biomass, is a liquid bio-fuel with the potential to replace some available, vast source of cheap biomass [4]. However, the
of the liquid fossil fuels used in transportation. Bio-ethanol, production of ethanol from lignocellulosic raw materials is more
produced from corn grain (starch) and sugar cane (sucrose) is difficult than from sugar or starch. Lignocellulosic materials
currently the most common renewable fuel [1] and the consist primarily of three components, namely cellulose,
introduction of ethanol on to the fuel market has been hemicellulose and lignin, of which the first two can be
facilitated by the positive effects of low-blend ethanol–petrol hydrolysed to monomeric sugars [5], which can then be
mixtures [2]. However, it is clear that the large-scale use of bio- fermented to ethanol using a hexose- and pentose-fermenting
ethanol will require lignocellulosic biomass to be used as raw organism. The hydrolysis of lignocellulose to monomeric sugars
material [3,1]. Furthermore, current data suggest that only can be achieved in many different ways. One thoroughly
lignocellulosic ethanol (ethanol made from lignocellulosic investigated method is to first treat the material using steam
biomass) offers large reductions in greenhouse gas emissions pretreatment [6], with or without a catalyst. Several studies have
compared with fossil fuels [3]. Bio-fuels also provide the been carried out on steam explosion as a method of pretreating
opportunity for non-oil-producing countries to be self-sufficient corn stover, the raw material used in this study, using dilute
in fuel. sulphuric acid or SO2 as a catalyst, with the aim of solubilizing
hemicellulosic sugars, and rendering the remaining cellulose
accessible to subsequent enzymatic hydrolysis [7–11].
* Corresponding author. Tel.: +46 46 222 82 97; fax: +46 46 222 45 26. After pretreatment, enzymatic hydrolysis is used to convert
E-mail address: guido.zacchi@chemeng.lth.se (G. Zacchi). the residual cellulose and hemicellulose into monomeric sugars.
1359-5113/$ – see front matter # 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.procbio.2007.02.003
K. Öhgren et al. / Process Biochemistry 42 (2007) 834–839 835

The sugars are then fermented to ethanol using yeast. When


enzymatic hydrolysis and fermentation are performed sequen-
tially, it is referred to as separate hydrolysis and fermentation
(SHF). However, the two process steps can be performed
simultaneously, i.e. simultaneous saccharification and fermenta-
tion (SSF). This was first done by Takagi et al. in 1977 [12]. SSF
was shown to be superior to separate hydrolysis and fermentation
when the whole slurry from steam pretreatment of softwood was
used [13]. Combining the two process steps also results in a lower
capital cost, and the fact that the ethanol concentration is higher
during SSF than SHF reduces the risk of contamination [14].
However, mixing the lignin residues with the yeast (as in SSF)
makes yeast recirculation very difficult. In addition, the
temperature optima for the yeast and the enzymes used differ,
which means that the conditions used in SSF cannot be optimal
Fig. 1. The experimental procedure.
for both the enzymes and the yeast.
Pretreatment hydrolysate has an inhibitory effect on
cellulose conversion in the enzymatic step [15,16] but this 2.3. Steam pretreatment
can be overcome by fermentation of the pretreatment
hydrolysate prior to enzymatic hydrolysis [17]. SSF may thus The water content in the corn stover was determined and the corn stover was
exhibit lower inhibition of the enzymes due to the concomitant impregnated with 3% SO2 (w/w, based on the dry weight). Steam pretreatment
was carried out in 50 g batches in a 2-L steam gun (Stake Tech II batch reactor,
fermentation. The pretreatment hydrolysate also has an
Stake Tech-Norvall, Ontario, Canada). Several batches were pretreated, after
inhibitory effect on the yeast, however, at low concentrations which the material was collected for analysis and stored at ÿ20 8C before use.
some of the inhibitors can actually have a positive effect on the Solid fractions, generated by pretreatment, were analysed in the same way as the
ethanol productivity and the ethanol yield by stressing the yeast raw material, and the liquid fraction was analysed with respect to monomeric
[18]. and oligomeric sugars, acetic acid, 5-hydroxy methyl furfural (HMF) and
furfural.
In this study, bench-scale (5 L) SSF and SHF at 8% water
insoluble solids (WIS) were compared using steam-pretreated
2.4. SSF and EH
corn stover. A WIS content of 8% is high enough to obtain
reasonable high ethanol concentrations, but low enough to The SSF and EH experiments were performed in a 15-L bioreactor with a
ensure that complete cell death is avoided. Both the whole working weight of 5 kg (Applikon, Schiedam, The Netherlands). The slurry
slurry (with all the inhibitors present) and washed slurry were from the pretreatment stage was either used as it was or was thoroughly washed
used in order to distinguish between inhibition due to by- with tap water. The pH was then adjusted to 5.0 with concentrated NaOH and
the WIS concentration was adjusted by the addition of tap water. In one series of
products formed in the steam pretreatment stage and sugar- experiments, additional glucose and xylose were added to the washed and
inhibition of the enzymes. The enzymes used were commercial conditioned slurry to adjust the sugar concentrations in the liquid fraction to the
enzyme mixtures from Novozymes A/S, Denmark, and the same level as in the experiment with the whole slurry. Sugar addition was based
yeast was a Saccharomyces cerevisiae strain cultivated in the on the sugar content in the liquid fraction of the pretreated slurry.
steam-pretreatment hydrolysate and thus adapted to the harsh A commercial cellulase mixture, cellulase NS 50013 (69.5 Filter Paper Unit
(FPU)/mL), supplemented with the b-glucosidase preparation, beta-glucosidase
environment of SSF. NS 50010, both from Novozymes A/S, Bagsværd, Denmark, was used. The
enzymatic activity in the experiments was 10 FPU/g WIS and the b-glucosidase
supplementation constituted 25% of the volume of cellulase added. The
2. Materials and methods commercial xylanase mixture, Multifect xylanases, (Genencor International,
Rochester, NY, USA), was also added in one series of experiments. The dosage
2.1. Raw material of xylanases was based on the protein content in the xylanase mixture (43 g/mL)
and equalled 0.006 g xylanase protein/g WIS, which was equivalent to 20% of
All the investigations were performed using corn stover from North the protein content in the cellulase added.
America, kindly supplied by NREL, Golden, Colorado, USA. After collection, A spent sulphite liquor-adapted strain of S. cerevisiae, (Tembec I) provided
the corn stover was chopped, air-dried and then stored at room temperature. The by Tembec Ltd. (Témiscaming, QU, Canada), was used at a concentration of
sugar and lignin contents of the raw material were determined according to 1 g/L (dry yeast). This yeast ferments glucose but not xylose. The yeast was
NREL [19]. purified and then cultivated on the liquid obtained after pretreatment of corn
stover to adapt it to the conditions used in SSF (see Section 2.5 below).
2.2. Experimental set-up Nutrients were added in the SSF experiments so that the concentrations in
the fermentor were 0.5 g/L (NH4)2HPO4, 0.025 g/L MgSO47H2O and 1.0 g/L
After impregnation with SO2, the stover was steam-pretreated at 190 8C for yeast extract. The temperature in the fermenters was kept at 35 8C during SSF
5 min. These conditions have previously been determined to be optimal for and 45 8C during EH, and all experiments were run for 120 h with the pH being
SO2-impregnated, steam pretreatment of corn stover [11]. The resulting slurry maintained at 5.0 by manual addition of a 50% NaOH. Samples were withdrawn
was then analysed and used in SSF and enzymatic hydrolysis (EH). The WIS after 0, 2, 4, 8, 24, 28, 32, 48, 72, 96 and 120 h, and analysed regarding ethanol,
content was adjusted with water, after which enzymes or enzymes and yeast sugars, acetic acid, lactic acid and sugar degradation products.
were added. The initial WIS concentration in all the experiments was 8%. The dry matter of the liquid, DMliquid, and of the whole slurry, DMslurry,
An overview of the experimental set-up is shown schematically in Fig. 1. were measured by drying a sample of the liquid fraction and of the slurry,
836 K. Öhgren et al. / Process Biochemistry 42 (2007) 834–839

respectively. The WIS content in the pretreated slurry was calculated accord- 3. Results and discussion
ing to [20]:

WIS ¼ ðDMslurry ÿ ð1 ÿ DMslurry Þ  DMliquid Þ 3.1. Raw material and steam pretreatment

The composition of the raw material is presented in Table 1. It


2.5. Yeast cultivation
was found to be within the normal range [23–27], although,
2.5.1. Inoculum culture the normal range is quite wide since the content of corn stover
Pure yeast culture (50 mg) was added to a 300-mL Erlenmeyer flask, which tends to differ depending on region, year, crop maturity, etc.
contained 100 mL sterile medium at a pH of 5.0. The medium composition was (Thomas S. Golden, Colorado, USA: National Bioenergy
as follows: 16.5 g/L glucose, 7.5 g/L (NH4)2SO4, 3.5 g/L KH2PO4, 0.75 g/L Center, National Renewable Energy Laboratory; 2003, personal
MgSO4, 10 mL/L trace metal solution [21] and 5 g/L yeast extract. The culture
communication).
was incubated at 30 8C for 24 h.
Table 2 summarizes the composition of the pretreated slurry.
2.5.2. Aerobic batch cultivation The sugar concentration in the liquid fraction includes both
In order to produce cell mass before the fed-batch phase, batch cultivation oligomeric and monomeric sugars.
on glucose, with a volume of 0.750 L, was carried out at 30 8C under sterile The recovery of sugars in the pretreatment stage (in both the
conditions. A 2.5-L bioreactor (Applikon, Schiedam, The Netherlands) was solid and liquid fractions) was 92.6% for glucose and 73.0% for
used for the batch and fed-batch cultivation stages. The medium contained the
xylose.
following components: 30 g/L glucose, 23 g/L (NH4)2SO4, 11 g/L KH2PO4,
2.6 g/L MgSO4, 34 mL/L trace metal solution and 5 g/L yeast extract. The pH
was maintained at 5.0 by automatic addition of 10% NaOH. Cultivation was 3.2. Simultaneous saccharification and fermentation
started by adding 60 mL inoculum culture. The dissolved oxygen tension (DOT)
was measured with a DOT electrode and was not allowed to fall below 40%. When SSF was performed using the whole slurry the ethanol
This was ensured by regulating the stirrer speed and air and oxygen gas flow.
productivity during the first 24 h was 0.55 g/Lh. After 120 h,
2.5.3. Aerobic fed-batch cultivation 20.5 g/L ethanol had been produced, which corresponds to an
After 24 h of batch cultivation, feeding of the pretreatment liquid started. overall ethanol yield of 72.4% based on the theoretical ethanol
The pretreatment liquid was enriched with 60 g glucose/L, as it contained a very production from glucose (Fig. 2(a)). In order to evaluate the
low concentration of hexoses, and pH-adjusted to 5.0 with 50% NaOH. effect of the inhibitors present in the liquid fraction of the slurry,
Glucose-enriched pretreatment liquid (1 L) was added over a period of 19 h
the pretreated material was washed thoroughly with tap water,
at a feed rate of 0.88 mL/min, and the pH was maintained at 5.0 by automatic
addition of 10% NaOH. The DOT was, again, not allowed to fall below 40%. after which the WIS was adjusted with water and additional
sugars were added in order to compensate for the sugars in the
2.5.4. Cell harvest liquid fraction that had been lost during washing. Washing the
When feeding had ceased, the cultivation liquid containing the yeast was material and then adding sugars is not a feasible process
transferred from the fermenter to centrifugation vessels and centrifuged at alternative, but was done in this study to evaluate the inhibitory
3500 rpm for 5 min (Multifuge 3 L-R (HWS 6027), Heraeus, Hanau, Germany).
The supernatant was discarded and the amount of pellet corresponding to 1 g/L
effect in SSF.
dry yeast was transferred to a beaker. Tap water was added to 200 mL, The concentration of acetic acid was much lower in SSF
whereupon the cell suspension was added to the SSF. The time lapse between with washed material (Fig. 2(b)) since the acetic acid liberated
the end of the fed-batch phase and the addition of the harvested cells to the SSF when hemicellulose is hydrolysed during pretreatment is
fermentation phase was less than 1 h. removed during washing. A minor amount of hemicellulose
was still present in the solid fraction after pretreatment
2.6. Analysis
(Table 2). When this hemicellulose is hydrolysed in the
The sugar and the lignin contents in the raw materials were determined enzymatic hydrolysis, the concentration of acetic acid increases
using concentrated acid hydrolysis at room temperature, followed by dilute acid slightly. The difference in sugar concentration between washed
hydrolysis at 121 8C to hydrolyse the cellulose and hemicellulose, according to and non-washed slurry at 0 h is due to uncertainties in the
the analytical procedure recommended by NREL [19]. The sugars were then analysis procedure as the high viscosity which makes it difficult
analysed with HPLC using a Dionex DX-2500 with an ion-exchange (PA1,
to take a representative sample. After 4 h, the viscosity had
Dionex) column and a pulsed amperometric detector with a gold electrode. The
remaining acid-insoluble lignin was measured by weighing, and the amount of decreased significantly and the samples taken then and
acid-soluble lignin was determined from the adsorption at 205 nm against a 4% thereafter were considered to be representative. The ethanol
H2SO4 blank.
The amounts of monosaccharides and inhibitors in the liquid fraction after Table 1
pretreatment were determined by HPLC. Glucose, xylose, galactose and Composition of corn stover expressed as % of dry matter
mannose, lactic acid, glycerol, acetic acid, ethanol, HMF and furfural were
Glucan 36.1
separated on an Aminex HPX-87-H column at 65 8C using 5 mmol/L H2SO4 as
Xylan 21.4
eluent at a flow rate of 0.6 mL/min. All samples were filtered through a 0.20-mm
Arabinan 3.5
filter and diluted properly analysis. To measure the total amount of sugars
Galactan 2.5
(monomers and oligomers) in the liquid fraction after pretreatment, the samples
Mannan 1.8
were hydrolysed with 4% H2SO4 at 121 8C for 1 h. They were then analysed
Lignina 17.2
with HPLC as described above.
Ash 7.1
The FPU activity of the cellulase mixture used was determined according to
Acetyl 3.2
Adney and Baker [22] and the protein content of the xylanase mixture was
a
determined using a BCS Protein Assay (Pierce, Rockford, IL, USA). Acid-soluble lignin and lignin ash included.
K. Öhgren et al. / Process Biochemistry 42 (2007) 834–839 837

Table 2 productivity during the first 24 h of 0.95 g/L h and a final


The concentration of sugars, degradation products, DM and WIS in pretreated
glucose concentration of 36.7 g/L, which corresponds to 65.9%
material (solid and liquid)
overall glucose yield (Fig. 3(a)). In EH of washed slurry the
DMslurry (%) 16.7 values were a glucose productivity of 1.46 g/L h and a final
WIS in slurry (%) 10.6
glucose concentration of 42.4 g/L, which corresponds to 78.5%
Content in WIS (% of dry weight) overall glucose yield (Fig. 3(b)). Removing the inhibitors seems
Glucan 52.7 thus to have a positive effect on enzymatic hydrolysis. The
Xylan 8.8
Lignin 27.7a
inhibitory effect of steam pretreatment hydrolysate on
enzymatic hydrolysis observed here is in accordance with
Concentration in the liquid fraction (g/L)
earlier studies on steam-pretreated spruce [17] and on steam-
Glucose 6.6
Xylose 24.8 pretreated poplar [16].
Acetic acid 2.6 The enzymes used in enzymatic hydrolysis are not only
HMF 0.6 inhibited by the degradation products present in the steam-
Furfural 0.7 pretreatment hydrolysate. They are also end-product inhib-
a
Includes both acid-soluble and acid-insoluble lignin and lignin ash. ited. This means that the very product that they produce
inhibits them due to a competitive inhibiting effect.
Consequently, endo- and exoglucanases are inhibited by
productivity during the first 24 h of SSF with washed slurry was cellobiose, and b-glucosidase is inhibited by glucose. The
0.62 g/Lh, and the ethanol concentration after 120 h reached inhibiting effect of the sugars present in the pretreatment
18.2 g/L, corresponding to an overall ethanol yield of 64.1% hydrolysate was investigated by not replacing the sugars
(Fig. 2(b)). The higher overall ethanol yield when non-washed removed in the washing process. EH using washed material
slurry is presumably due to the fact that inhibitors present in the with no addition of additional sugars was performed with the
pretreatment liquid can increase the ethanol production if they same WIS content (8%) as in the other experiments resulting
are not present at too high concentrations. At low concentra- in the same glucose productivity during the first 24 h (1.41 g/
tions, up to 100 mmol/L, acetic acid, for example, has been L h, data not shown) as for the EH with washed material and
shown to increase the ethanol yield [18]. Removing the the addition of additional sugars. It was thus concluded that
inhibitors can thus decrease the ethanol yield. The slight the inhibitory effect of the pretreatment hydrolysate on the EH
increase in ethanol productivity during the first 24 h with could be attributed to sugar degradation products and other by-
washed solids can be explained by the lower concentration of products produced during the steam pretreatment and not the
fermentable inhibitors such as furfural and HMF. These sugars present in the liquid fraction. However, the fairly high
inhibitors are metabolized by the yeast, leading to a lag phase in sugar concentration at the end of the EH probably caused some
ethanol production [18]. The glucose concentration decreased end-product inhibition.
to zero much faster when the washed solids was used in SSF
than when the whole slurry was used (Fig. 2(a) and (b)) since 3.4. SSF versus SHF
the yeast does not need to metabolize any inhibitors in the
washed slurry. SSF was compared with SHF for three types of slurry: whole
slurry, washed slurry with additional sugars and whole slurry
3.3. Enzymatic hydrolysis with additional xylanases for improved enzymatic hydrolysis.
To be able to compare SHF with SSF, the SHF results were
The effect of inhibitors on EH was evaluated using the same based on the EH results assuming 90% fermentation yield in a
kind of study as described above for SSF. EH was performed on 24 h long separate fermentation (0.46 g ethanol per gram
the whole slurry as well as on washed slurry with additional glucose). SHF will thus take 144 h since the EH takes 120 h and
sugars added to compensate for the soluble sugar loss in the the fermentation is assumed to take 24 h. Table 3 summarizes
washing procedure. EH of the whole slurry resulted in a glucose the results of the comparison between the SSF and SHF.

Fig. 2. Ethanol, glucose and acetic acid concentrations during SSF of whole slurry (a) and washed slurry with additional sugars (b), both at 8% WIS.
838 K. Öhgren et al. / Process Biochemistry 42 (2007) 834–839

Fig. 3. Glucose and acetic acid concentrations during SHF of whole slurry (a) and washed slurry with additional sugars (b), both at 8% WIS.

Table 3 In EH of whole slurry with additional xylanases, the glucose


Comparison between ethanol yield from SSF (after 120 h) and SHF (after 120 h productivity during the first 24 h was 1.18 g/L h. This is higher
hydrolysis and 24 h fermentation)
than the glucose productivity for the whole slurry but lower than
Ethanol yield Overall ethanol Ethanol concentration for washed slurry. Since adding xylanases to the whole slurry
in SSF/SHF (%) yield (%) after 120/144 h (g/L)
can boost the ethanol productivity in SSF compared with
Whole slurry washed slurry but not in EH, fermentation clearly has an effect
SSF 78.2 72.4 20.5 on the inhibition of the enzymes. The fermentation of
SHF 64.1 59.3 16.8
pretreatment hydrolysate to overcome inhibition in enzymatic
Washed slurry with additional sugar hydrolysis has been demonstrated previously by Tengborg et al.
SSF 69.3 64.1 18.2 [17] using spruce. The glucose concentration after 120 h in the
SHF 76.2 70.6 19.4
EH of whole slurry with additional xylanases was 39.1 g/L
Whole slurry with additional xylanases (Fig. 4(b)), corresponding to an overall glucose yield of 70.2%,
SSF 81.5 75.4 21.4
which is lower than the yield for the washed slurry with
SHF 68.2 63.2 17.9
additional sugars.
The fermentation yield in the separate fermentation was assumed to be 90% SSF was shown to be the preferable process configuration
(0.46 g ethanol/g glucose).
when the whole slurry was used (Table 3), which is in
accordance with the results found by Söderström et al. [13] for
The results for the whole slurry and the washed slurry with softwood. This is due to three factors: (1) lower glucose
additional sugars are based on the results presented above for concentration and thus decreased glucose inhibition of the
SSF and EH. In SSF of whole slurry with additional xylanases, enzymes; (2) less inhibition of the enzymes by degradation
the ethanol productivity during the first 24 h was 0.69 g/L h, products present in the pretreatment hydrolysate since
which was slightly higher than those for the whole slurry and fermentation seems to afford some detoxification; (3) the
the washed slurry with standard enzyme addition. The positive effect of low concentrations of some of the inhibitors
increased enzymatic hydrolysis rate is also evidenced by the on the fermentation yield. When washed slurry was used, the
fact that the glucose concentration did not decrease to zero until ethanol yield in SHF was higher than in SSF (Table 3), probably
after 72 h (Fig. 4(a)). The ethanol concentration after 120 h was since the enzymes in the SHF were allowed to work at optimal
21.4 g/L (Table 3), corresponding to an overall ethanol yield of temperature. Alfani et al. [28] also found that washed steam-
75.4%. The difference in sugar concentration between SSF and pretreated wheat straw gave a higher ethanol yield when SHF
EH at 0 h is again due to the high viscosity, which makes it was used rather than SSF. Adding a washing step in the ethanol
difficult to take a representative sample. production process would, however, increase the production

Fig. 4. Glucose, acetic acid and ethanol concentrations during SSF (a) and SHF (b) of whole slurry with additional xylanases, both at 8% WIS.
K. Öhgren et al. / Process Biochemistry 42 (2007) 834–839 839

cost and cause a considerable loss of sugars, which cannot be [11] Öhgren K, Zacchi G, Galbe M. Optimization of steam pretreatment of
SO2-impregnated corn stover for fuel ethanol production. Appl Biochem
replaced in a large-scale process. Another way of decreasing
Biotechnol 2005;121–124:1055–67.
the concentration of inhibitors before hydrolysis is to add a [12] Takagi M, Abe S, Suzuki S, Emert GH, Yata N. A method for production
detoxification step [29]. However, this may constitute up to of alcohol direct from cellulose using cellulase and yeast. In: Process
22% of the total ethanol production cost [30] and is therefore bioconversion symposium; 1977.p. 551–71.
best avoided. [13] Söderström J, Galbe M, Zacchi G. Separate versus simultaneous sacchar-
Adding extra xylanases to the process increases the ethanol ification and fermentation of two-step steam pretreated softwood for
ethanol production. J Wood Chem Technol 2005;25(3):187–202.
yield slightly for both SSF and SHF of the whole slurry [14] Wyman CE, Spindler DD, Grohmann K. Simultaneous saccharification
(Table 3). The beneficial effect of xylanases in enzymatic and fermentation of several lignocellulosic feedstocks to fuel ethanol.
hydrolysis has been established previously [20]. However, the Biomass Bioenergy 1992;3(5):301–7.
increase in ethanol yield due to the addition of xylanase must be [15] Mes-Hartree M, Saddler JN. The nature of inhibitory material present in
pretreated lignocellulosic substrates which inhibits the enzymatic hydro-
weighted against the cost.
lysis of cellulose. Biotechnol Lett 1983;5(8):531–53.
[16] Cantarella M, Cantarella L, Gallifuoco A, Spera A, Alfani F. Effect of
4. Conclusions inhibitors released during steam-explosion treatment of poplar wood on
subsequent enzymatic hydrolysis and SSF. Biotechnol Prog 2004;20:
The inhibitors present in the liquid fraction of the pretreated 200–6.
[17] Tengborg C, Galbe M, Zacchi G. Reduces inhibition of enzymatic
slurry had a significant and negative impact on enzymatic
hydrolysis of steam-pretreated softwood. Enzyme Microbial Technol
hydrolysis. This negative impact was decreased when SSF was 2001;28:835–44.
used. Due to the reduction of glucose inhibition in the [18] Larsson S, Palmqvist E, Hahn-Hägerdal B, Tengborg C, Stenberg K,
enzymatic hydrolysis during SSF, the detoxifying effect of Zacchi G, et al. The generation of fermentation inhibitors during dilute
fermentation and the positive effect of inhibitors present in the acid hydrolysis of softwood. Enzyme Microbial Technol 1999;24:151–9.
[19] Sluiter A, Hames B, Ruiz R, Scarlata C, Sluiter J, Templeton D. Deter-
pretreatment hydrolysate (for example acetic acid) on the
mination of structural carbohydrates and lignin in biomass. Golden, CO:
fermentation, SSF was concluded to be a better process NREL; 2004.
configuration than SHF when the whole slurry was used. [20] Öhgren K, Vehmaanperä J, Siika-Aho M, Galbe M, Viikari L, Zacchi G.
Adding extra xylanases had a slightly positive effect on the High temperature enzymatic prehydrolysis prior to simultaneous sacchar-
ethanol yield in both SSF and SHF of the whole slurry. ification and fermentation of steam pretreated corn stover for ethanol
production. Enzyme Microbial Technol 2007;40(5):1100–7.
[21] Taherzadeh MJ, Liden G, Gustafsson L, Niklasson C. The effects of
References pantothenate deficiency and acetate addition on anaerobic batch fermen-
tation of glucose by Saccharomyces cerevisiae. Appl Biochem Biotechnol
[1] Gray KA, Zhao L, Emptage M. Bioethanol Curr Opin Chem Biol 1996;46:176–82.
2006;10(2):141–6. [22] Adney B, Baker J. LAP-006; measurement of cellulase activity. Golden,
[2] Lynd LR, Cushman JH, Nichols RJ, Wyman CE. Fuel ethanol from CO: NREL; 1996.
cellulosic biomass. Science 1991;251:1318–23. [23] Esteghlalian A, Hashimoto AG, Fenske JJ, Penner MH. Modeling and
[3] Farrell AE, Plevin RJ, Turner BT, Jones AD, O’Hare M, Kammen DM. optimization of the dilute-sulfuric-acid pretreatment of corn stover, poplar
Ethanol can contribute to energy and environmental goals. Science and switchgrass. Bioresour Technol 1997;59:129–36.
2006;311:506–8. [24] Kaar WE, Holtzapple MT. Using lime pretreatment to facilitate the
[4] Kim S, Dale BE. Global potential bioethanol production from wasted enzymic hydrolysis of corn stover. Biomass Bioenergy 2000;18:189–99.
crops and crop residues. Biomass Bioenergy 2004;26:361–75. [25] Kalman G, Varga E, Reczey K. Dilute sulphuric acid pretreatment of corn
[5] Dumitriu S. Polysaccharides. New York: Markker Deccer Inc.; 1998. stover at long residence times. Chem Biochem Eng Q 2002;16(4):151–7.
[6] Galbe M, Zacchi G. A review of the production of ethanol from softwood. [26] Schell DJ, Walter PJ, Johnson DK. Dilute sulphuric-acid pretreatment of
Appl Microbiol Biotechnol 2002;59:618–28. corn stover at high solids concentration—scientific note. Appl Biochem
[7] Parekh SR, Parekh RS, Wayman M. Ethanol and butanol production by Biotechnol 1992;34–35:659–65.
fermentation of enzymatically saccharified SO2-prehydrolysed lignocel- [27] Torget R, Walter P, Himmel M, Grohmann K. Dilute-acid pretreatment of
lulosics. Enzyme Microbial Technol 1988;10(11):660–8. corn residue and short-rotation woody crops. Appl Biochem Biotechnol
[8] Schell DJ, Farmer J, Newman M, Mcmillan JD. Dilute-sulfuric acid 1991;28–29:75–86.
pretreatment of corn stover in pilot-scale reactor—investigation of yields, [28] Alfani F, Gallifuoco A, Saporosi A, Spera A, Cantarella M. Comparison of
kinetics, and enzymatic digestibilities of solids. Appl Biochem Biotechnol SHF and SSF processes for bioconversion of steam-exploded wheat straw.
2003;105–108:69–85. J Ind Microbiol Biotechnol 2000;25:184–92.
[9] Tucker MP, Kim KH, Newman MM, Nguyen QA. Effects of temperature [29] Palmqvist E, Hahn-Hägerdal B. Fermentation of lignocellulosic hydro-
and moisture on dilute-acid steam explosion pretreatment of corn stover lysates. I. Inhibition and detoxification. Bioresour Technol 2000;74:
and cellulase enzyme digestibility. Appl Biochem Biotechnol 2003; 17–24.
165–77. [30] von Sivers M, Zacchi G, Olsson L, Hahn-Hägerdal B. Cost analysis of
[10] Wayman M, Parekh SR. SO2 prehydrolysis for high-yield ethanol-pro- ethanol production from willow using recombinant Escherichia coli.
duction from biomass. Appl Biochem Biotechnol 1988;17:33–43. Biotechnol Prog 1994;10:555–60.

Centres d'intérêt liés