Available online at www.sciencedirect.

com

Bioresource Technology 99 (2008) 2162–2169

Fermentation of sunflower seed hull hydrolysate to ethanol

by Pichia stipitis
Müjgan Telli-Okur, Nurdan Eken-Saraçoğlu *

Department of Chemical Engineering, Faculty of Architecture and Engineering, Gazi University, Maltepe 06570, Ankara, Turkey

Received 7 November 2006; received in revised form 18 May 2007; accepted 18 May 2007
Available online 20 July 2007

Abstract

Ethanol production from sunflower seed hull hydrolysate was evaluated using Pichia stipitis NRRL Y-7124. The hydrolysate pre-
pared with 0.7 M H2SO4 at 90 °C was fermented as substrate in shaking bath experiments at 30 °C. In a group of experiments, the influ-
ence of various detoxification methods on the fermentability of hydrolysate was investigated at pH 6. Even though the ability of all
employed pretreatments to enhance fermentation performance was close, the sequential application of overliming with sodium sulfite
addition was the best detoxification method. Additional experiments were performed with detoxified hydrolysate to investigate the effect
of shaking rate (70–130 rpm) and initial pH (5.5–7) on the fermentation. The highest ethanol level 11 g L 1 was achieved at initial pH of
6 and 100 rpm shaking rate from a hydrolysate containing 48 g L 1 total reducing sugar. The corresponding alcohol yield and volumetric
productivity were 0.32 g g 1 and 0.065 g L 1 h 1.
Ó 2007 Elsevier Ltd. All rights reserved.

Keywords: Sunflower seed hull; Acid hydrolysis; pH; Shaking rate

1. Introduction for hydrolyzing agricultural and wood wastes. During

hydrolysis, several compounds such as furfural, hydroxy-
The possibility of the decline in worldwide crude oil pro- methylfurfural, acetic acid, syringic acid, vanillin, p-
duction in future developed a great interest in exploring hydroxybenzoic acid are released as a result of sugar or lig-
alternative energy sources. Ethanol produced from biomass nin degradation processes (Mussatto and Roberto, 2004).
is one attractive alternative for partial replacements of fos- The lignocellulosic hydrolyzates provide a rich medium
sil fuels. Large scale application of bio-ethanol in fuel for fermentation of sugars into ethanol, xylitol or other
blends will contribute to reduction of CO2 and other emis- valuable products. However, some of the degradation
sions from the transport sector. Unlike fossil fuels, ethanol products inhibit the metabolism of fermentative microor-
is a renewable energy source and can be produced through ganisms and negatively effect the efficiency of fermentation
fermentation of sugars. Lignocellulosic residues are consid- (Clarck and Mackie, 1984; Tran and Chambers, 1986; Van
ered as attractive raw materials for the production of fuel Zyl et al., 1991). The kind of toxic compounds and their
ethanol because of their availability in large quantities at amounts in lignocellulosic hydrolysates depend on both
low cost (Parisi, 1989). The major components of lignocel- the raw material and the operational conditions employed
lulosic biomass are hemicellulose, cellulose and lignin. The for hydrolysis. Different detoxification methods prior the
hemicellulose fraction can be easily hydrolysed to mano- fermentation for improvement of cell growth and ethanol
meric sugars, xylose and glucose under mild conditions. production have been suggested including neutralization,
Dilute acid hydrolysis is the one of two major processes over-liming, evaporation, ion exchange resins and charcoal
adsorption (Mussatto and Roberto, 2004). The effective-
*
Corresponding author. ness of a detoxification method depends on the hemicellu-
E-mail address: [email protected] (N. Eken-Saraçoğlu). losic hydrolysate and the microorganism employed for

0960-8524/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biortech.2007.05.036
 M. Telli-Okur, N. Eken-Saraçoğlu / Bioresource Technology 99 (2008) 2162–2169 2163

fermentation. Due to large proportion of xylose in hemicel- pH is a crucial variable during fermentation (Palmqvist
lulosic hydrolysates and the inability of Saccharomyces and Hahn-Hagerdal, 2000) Undissociated weak acids in
cerevisiae to ferment xylose to ethanol, the use of xylose- fermentation medium can diffuse across the plasma mem-
fermenting yeasts to coferment pentose and hexose sugars brane and may inhibit microorganism growth.
in hydrolysate offers on opportunity for the efficient utiliza- In this study, potential use of hemicellulosic hydrolysate
tion of hemicellulose component of agricultural residues. derived from oil industry sunflower seed hull bagasse in
Among the xylose fermenting yeasts, Pichia stipitis pro- alcohol fermentation using P. stipitis was investigated.
duces high ethanol yields under low oxygen level (Du Preez The choice of the best detoxification method to improve
et al., 1986a). the efficiency of fermentation process and the investiga-
All current bioethanol production is based on corn glu- tion the effects of initial pH of fermentation medium and
cose in the US and sucrose in Brazil. It is expected that the shaking speed on the fermentability of sunflower seed hull
cost of lignocellulosic ethanol can undercut that of starch hydrolysate were the objectives of this study.
based ethanol because low value agricultural residues can
be used. Large scale technologies for lignocellulosic feed- 2. Methods
stocks will likely be a reality in near future with a biomass
refinery approach (Lin and Tanaka, 2006). Recently the 2.1. Microorganism and growth media
potential bioethanol production from wasted crops and
the state of hydrolysis-fermentation technologies to pro- P. stipitis NRRL Y-124 was grown at 30 °C on agar
duce ethanol from lignocellulosic biomass were evaluated slants composed of (g L 1): glucose (10), yeast extract
(Kim and Dale, 2004; Hamelinck et al., 2005). A wide vari- (3), malt extract (3), peptone (5), agar (20). The medium
ety of biomass such as wood and forest product residues, described by Slininger et al. (1982) was used for growth.
grasses, agricultural crops and their residues can be used Inocula were prepared by transferring organism by loop
as feedstocks for ethanol production (Tran and Chambers, from two days slants to 250-mL erlenmeyer flasks contain-
1986; Delgenes et al., 1990; Roberto et al., 1991; Ferrari ing 100 ml of growth medium. The inoculum medium con-
et al., 1992; Hahn-Hägerdal et al., 1994; Saraçoğlu-Eken sisted of (g L 1): (6.4) urea, (1.2) KH2PO4, (0.18)
and Arslan, 2000; Nigam, 2001, 2002; Sharma et al., Na2HPO4, (10) yeast extract and (50) D-xylose (pH 4.5).
2002, 2004). Ethanol conversion systems could use multiple CaO, ZnO, FeCl3 Æ 6H2O, MgO, CuSO4 Æ 5H2O,
biomass feedstocks rather than a single feedstock to CoCl2 Æ 6H2O, H3BO3 were also included in the growth
achieve the desired scale towards the future. Since raw medium as trace metals. The yeast was incubated aerobi-
material cost comprises more than 20% of the production cally with a magnetic stirring at 30 °C for 27 h prior to use.
cost (Kaylen et al., 2000), novel fermentation feedstocks
from cheap non-traditional crops and agricultural wastes 2.2. Preparation of hydrolysate samples
might become attractive alternatives. Each feedstock
requires different technological approaches and this needs Sunflower seed hull baggasse acquired from an oil-pro-
to be understood and dealt with appropriately. Residues cessing unit in Edirne region, Turkey were ground and
that have already been collected have the advantage of screened to the size of 0.71–1 mm and used as raw material.
low cost such as sunflower seed hulls obtained from an Xylan content was estimated as 21.5% (w/w) by transform-
oil factory. Sunflower (Helianthus annus L.) is an abundant ing xylans with HCl to furfural, which was collected in the
plant in Turkey and has an annual production of 950,000 distillate and determined colorimetrically with orcinol
tons. About 550,000 tons of this production is extraction FeCl3. Sunflower seed hulls hemicellulosic hydrolysate
residue or sunflower-pressed baggasse of some edible oil was prepared in a glass reactor equipped with a mechani-
industries (Gerçel, 2002). Mutlu (1990) reported the com- cally driven glass stirrer. The acid hydrolysis was conducted
position of sunflower seed hulls as 47% cellulose, 26% for 3.5 h with 0.7 M H2SO4 and a solid/liquid ratio of 1/5
hemicellulose and 16% lignin on dry basis. Today, its main (w/v) at 90 °C. The hydrolysate was recovered after separa-
utilization remains as feedstocks for boilers. tion of solid and liquid fractions by filtration. The liquid
Fermentation variables such as cultivation conditions, fraction had a pH of 0.7. Recovery of sugars from lignocel-
pH of medium and dissolved oxygen concentration (Taher- lulose solubilization was measured as the reducing sugar
zadeh et al., 2000) are also associated with toxicity effects. which primarily may contain xylose and arabinose from
Oxygen supply is the most important environmental factor hemicellulose and also glucose and cellubiose from cellu-
in xylose fermentation by yeast. The aeration rate deter- lose. On the other hand, non-sugar reducing compounds
mines the partitioning of the flux of carbon from substrate which dissolved in hydrolysate may also be measured as
between growth and product formation (Du Preez, 1994). reducing sugar (Sun and Cheng, 2005). After hydrolysis,
It is also reported that the high aeration rate can effect the composition of hydrolysate was 37 g total reducing
the yeast adaption to hydrolysate medium by overcoming sugar L 1. Each time, the required amount of synthetic
the inhibitory effect of some compounds (Diz et al., xylose was supplemented into hydrolysate to maintain a
2002). The concentration of undissociated acids in lignocel- total reducing sugar concentration around 48 g L 1 before
lulosic hydrolysates is very dependent on pH and therefore any detoxification application. This supplementation was
2164 M. Telli-Okur, N. Eken-Saraçoğlu / Bioresource Technology 99 (2008) 2162–2169

carried out to see the effects of process variables in more 30 °C. The initial pH was varied in the range of 5.5–7.0.
concentrated hydrolysate without changing original The shaking rate was changed between 70 and 130 rpm
amount of toxicity in medium. The treated hydrolysates in fermentations conducted in 250 mL flasks containing
were further supplemented with additional nutrients to 135 mL medium at pH 6.
maintain the growth medium composition given above sec-
tion. The sunflower seed hull hydrolysate and the remaining 2.5. Analytical methods
nutrients were separately autoclaved and combined after
sterilization. The fermentability of diluted sunflower seed Growth was measured turbidometrically at 600 nm. Dry
hull hydrolysates containing 24–30 g L 1 reducing sugar cell mass was determined by centrifuging an aliquot of cul-
were also evaluated. Synthetic media were prepared by ture, followed by washing and drying the cell pellet to con-
dissolving known amounts of xylose in water and supple- stant weight in 105 °C to constant mass. Total reducing
mented with the same nutrients with hemicellulosic hydrol- sugars were determined colorimetrically using dinitrosali-
ysates to compare the fermentation parameters. cylic acid reagent (Miller, 1959). Ethanol was measured
by the dichromate oxidation method which is based on
2.3. Detoxification methods the complete oxidation of ethanol by dichromate in the
presence of sulfuric acid (Hormitz, 1980). Total furans in
To minimize the concentrations of fermentation inhibi- hydrolysates samples were estimated by a spectrophoto-
tors, the hydrolysate samples were submitted to one of four metric method based on the difference in absorbance at
detoxification methods (Mussatto and Roberto, 2004). 284 and 320 nm using a Hach DR/4000 spectrophotometer
(Martinez et al., 2000a) before and after any pretreatment.
2.3.1. Neutralization The amount of total mineral acids and total organic acids
Hydrolysate was heated to 60 °C in a water bath during in hydrolysate was measured by titration with 2 M NaOH
30 minutes and mixed vigorously during rapid addition of using a pH-meter.
CaO until (44.4 g L 1) pH was adjusted to 6.0. The sample
was filtered. 3. Results and discussion

2.3.2. Overliming Oil factory sunflower seed hull bagasse was hydrolyzed
Heated hydrolysate was treated with CaO until with sulfuric acid and the resulting hydrolysates were used
(66.67 g L 1) pH 10 and filtered and acidified to pH 6 with for ethanol production with P. stipitis yeast in shaking
concentrated H2SO4. flasks after detoxification. Low hydrolysis temperature
(90 °C) gives more hemicellulose solubilization than cellu-
2.3.3. Overliming combined with charcoal treatment lose. It can also be suggested that as a result of low temper-
Powdered charcoal was treated in boiling water and ature employed, some solubilized oligomers may remain as
dried at room temperature. Overlimed hydrolysates non-fermentable sugars in sunflower seed hydrolysate. It is
(200 mL) was mixed with charcoal (1 g) and stirred during known that DNS procedure gives the amount of reducing
24 h at 30 °C. The hydroysate was recovered with filtration. sugars as glucose equivalents, measured through the free
reducing groups. In this procedure, manomers and the free
2.3.4. Overliming combined with sulfite supplementation reducing groups at the end of the polymer chain are
Overlimed hydrolysate was supplemented with Na2SO3 detected. The presence of reasonable reducing sugar in
(3 g L 1) and sample was filtered. hydrolysate propose that manomeric sugars are significant
in hydrolysate. Davis et al. (2005) also reported a recovery
2.4. Shaking flask experiments of sugars from hemicellulose of approximately 80% by
using a relatively low temperature (100 °C) and low acid
In the first group of experiments, the influence of vari- concentration (2% v/v). On the other hand, P. stipitis has
ous detoxification methods on the fermentability of hydrol- also an advantage over S. cerevisiae in its ability to ferment
ysates was investigated. All treated hydrolysates containing xylose, glucose and cellobiose which is probably present in
additional nutrients were fermented by P. stipitis at 30 °C sunflower seed hydrolysate. Untreated hydrolysate had
in 250 mL shake flasks having 135 mL medium at 70 rpm 0.53 M H2SO4 and 0.37 M CH3COOH. Table 1 summa-
and pH 6. Hydrolysate fermentation medium was inocu- rizes the treatment procedures of detoxification to achieve
lated with 20% (v/v) growth culture to give an initial cell efficient fermentation. Before any treatment, the reducing
concentration between 1.92 and 1.98 g L 1. In control fer- sugar concentration in hydrolysates was either 48 g L 1
mentations, initial cell concentration was 1.1 g L 1 when (concentrated hydrolysates, Run 1–5) or 30 g L 1 (diluted
synthetic medium was inoculated with 10% (v/v) growth hydrolysates, Run 6–10). After treatments, total reducing
culture. Furthermore, experimental studies with the treated sugar concentration in hydrolysates decreased in various
hydrolysate have also been done in shaking flasks to eval- amounts (1.5–20%). Figs. 1(Run 1–5) and 2(Run 6–10)
uate the effect of initial pH and shaking speed on the alco- show time course for biomass growth, sugar consumption
hol fermentation of sunflower seed hull hydrolysate at and ethanol formation in treated hydrolysates and control
 M. Telli-Okur, N. Eken-Saraçoğlu / Bioresource Technology 99 (2008) 2162–2169 2165

Table 1
Effect of detoxification methods on ethanol fermentation of sunflower seed hull hydrolysates by P. stipitis
Detoxification method Substratea So Maximum YP/S (g ethanol/g total QP (max ethanol conc./ Theoretical yield%
(g L 1) ethanol (g L 1) sugar consumed) time) (g (L h) 1) (g ethanol/g substrate)
1 Control I (synthetic medium) 48 16.33 0.35 0.120 68.6
2 Overliming 42.3 7.59 0.38 0.054 74.5
3 Overliming + Na2SO3 47.3 8.50 0.28 0.060 55.0
4 Overliming + Charcoal 43.5 7.70 0.26 0.046 51.0
5 Neutralization 43.9 7.93 0.23 0.042 45.1
6 Control II (synthetic medium) 30.0 11.00 0.37 0.115 72.6
7 Overliming 29.0 5.47 0.26 0.050 51.0
8 Overliming + Na2SO3 26.0 5.70 0.3 0.040 58.8
9 Overliming + Charcoal 24.0 5.10 0.23 0.023 45.1
10 Neutralization 24.6 4.79 0.26 0.050 51.0
a
Total reducing sugar after pretreatment or xylose.

runs with P. stipitis. Even though cell growth patterns in reducing sugar in hydrolysates mediums was not com-
various hydrolysates and in synthetic media remain close, pletely consumed by P. stipitis. The percentage of reducing
the profiles of sugar utilization and ethanol production sugar used remained between 80% and 92% which may be
did not exhibit such a behavior. In spite of treatments, most likely due to partial removal of toxic elements in raw
Total Reducing Sugar (g/L)

50 Total Reducing Sugar (g/L)

45 Control 35
Overliming Control
40 30 Overliming
Overliming+NaSulfite
35 Overliming+Charcoal Overliming+NaSulfite
25
30 Neutralization Overliming+Charcoal
25 20 Neutralization
20 15
15
10 10
5 5
0
0
0 50 100 150 200 250
0 50 100 150 200 250 300
Time (hours) Time (hours)

14 12
Control
12 10 Overliming
Overliming+NaSulfite
Biomass (g/L)

Biomass (g/L)

10
8 Overliming+Charcoal
8 Neutralization
Control 6
6
Overliming
4 4
Overliming+NaSulfite
2 Overliming+Charcoal
2
Neutralization
0
0
0 50 100 150 200 250 0 50 100 150 200 250 300
Time (hours) Time (hours)
18 12
Control Control
16 Overliming
10 Overliming
14 Overliming+NaSulfite Overliming+NaSulfite
Ethanol (g/L)

Ethanol (g/L)

Overliming+Charcoal Overliming+Charcoal
12 Neutralization 8
Neutralization
10
6
8
6 4
4
2
2
0 0
0 50 100 150 200 250 0 50 100 150 200 250 300
Time (hours) Time (hours)

Fig. 1. Effect of detoxification method employed on: (a) reducing sugar Fig. 2. Effect of detoxification method employed on: (a) reducing sugar
consumption, (b) biomass growth, (c) ethanol formation (Run 1–5 in consumption, (b) biomass growth, (c) ethanol formation (Run 6–10 in
Table 1). Table 1).
2166 M. Telli-Okur, N. Eken-Saraçoğlu / Bioresource Technology 99 (2008) 2162–2169

hydrolysates. The fermentation parameters such as ethanol 5.7 g L 1 (52% of Control II) from a hydrolysate having
yield (YP/S), productivity (QP) and theoretical yield% are 26 g L 1 initial reducing sugar after overliming and
also given in Table 1. When compared with synthetic media Na2SO3 addition employed as detoxification. The corre-
the fermentations of detoxified hydrolysates are character- sponding ethanol yield and productivity were 0.3 g g 1
ized by slow kinetics with lower yield and productivity. (81% of Control II) and 0.04 g L 1 h 1 (35% of Control
Table 2 shows the furan amounts in concentrated hydroly- II) respectively. This observation suggests that inhibition
sates before and after detoxification. Although furans in diluted hydrolysates could not be attributed solely to
(hydroxymethyl furfural plus furfural) represent only a furans; other toxic components could be responsible for
group of toxins present in hemicellulose hydrolysates, low fermentability. It is reported that lignin degradation
recent studies have shown that, they may serve as useful products are more toxic to microorganisms than furans
markers for toxicity in hydrolysates (Martinez et al., even when concentrations are low (Parajo et al., 1998).
2000b, 2001). It is evident that, total furan concentration According to the results of detoxification procedures
was affected by all detoxification methods applied. The employed, the sequential application of over-liming with
highest removal of furans (68% removal) was achieved by sulfite addition was used for further experiments. It is
overliming combined with charcoal treatment. All the known that the effect of toxic compounds extremely com-
remaining methods resulted in a 44% reduction in average plex and varies with the cultivation conditions present in
in total furans. Martinez et al. (2001) reported 51% the fermentation process.
removal of total furans after optimal overliming of dilute There is evidence that inhibition depends not only the
acid hydrolysates of sugarcane bagasse. It is interesting concentration of toxic compounds but also on the oxygen
to mention that neutralization method was cable of altering concentration (Parajo et al., 1997) and pH of the medium
furan content in hydrolysates same as overliming. (Nigam, 2001). It is also well known that xylose fermenta-
Significant ethanol production in all treated hydroly- tion needs a low level oxygen to satisfy cellular metabolic
sates indicated that the level of furan aldehydes remained requirements. To investigate the effect aeration, the fermen-
after treatments was in the permissible range for ethanol tation of treated hydrolysate was carried out at shaking
production. In general, furfural is inhibitory to yeast rates between 70 and 130 rpm by P. stipitis. The corre-
metabolism at a level of 1 g L 1 or higher (Roberto sponding kLa values were estimated with the correlation
et al., 1991). Table 1 results presented that there was little given by Nikakhtari and Hill (2006). The kinetic parame-
difference in the ability of all applied detoxification meth- ters and kLa values are summarized in Table 3. Growth
ods to enhance the fermentability of sunflower seed hulls of biomass, sugar utilization rates and ethanol production
hydrolysate. The favorable effect in ethanol formation during fermentation of treated hydrolysate are shown in
associated with detoxification was most marked after treat- Fig. 3a–c. As expected increasing shaking rate from 70 to
ment with overliming followed by Na2SO3 addition for 130 rpm resulted in an increase in biomass concentration
both concentrated and diluted hydrolysates. This observa- (Fig. 3b). In contrast, the highest ethanol concentration
tion is in accordance with previous results obtained by 11 g L 1 and ethanol yield 0.32 g g 1 were achieved with
Nigam (2002) and Converti et al. (2000). In contrast to shaking rate 100 rpm. It seems that insufficient aeration
furan removal, active carbon treatment after over-liming slows substrate utilization whereas too much aeration
did not improve the fermentability any further. This could reduces ethanol yield because of either a shift from ethanol
be explained by the sensitivity of active carbon treatment to to cell growth or product oxidation (Nigam, 2002). Du
pH. Mussatto and Roberto (2004) reported that low pH Preez (1994) stated that growth and ethanol yields are
during treatment favors the removal of lignin products, inversely related in fermentation of D-xylose to ethanol.
which is not the case in this study. For Runs 2–5, a maxi- On the other hand ethanol productivity improved steadily
mal ethanol concentration of 8.51 g L 1 (52% of Control I) up to 0.067–0.075 g L 1 h 1 with increasing shaking rate.
was obtained for the overlimed and sulfited hydrolysate Other studies also have shown that the volumetric produc-
containing 47.3 g L 1 reducing sugar. The fermentation tion rates increases with increased aeration (Jeffries, 1985).
parameters, ethanol yield and productivity were found to Relatively earlier studies (Du Preez et al., 1986b; Slinin-
be 0.28 g g 1 (80% of Control I) and 0.06 g L 1 h 1 (50% ger et al., 1990) showed that the optimum pH was about
of Control I), respectively, under that condition. The best
ethanol concentration with P. stipitis for Runs 7–10 was
Table 3
Effect of shaking rate on ethanol fermentation of (detoxified by overliming
Table 2 combined with Na2SO3) sunflower seed hull hydrolysate with P. stipitis at
Furan amounts in hydrolysates before and after detoxification pH 6
1
Detoxification method Total furan (mg L ) % Furan removal Total reducing Shaking Maximum YP/S QP kLa
Untreated hydrolysate 779 – sugar (g L 1) rate ethanol (g g 1) (g L 1
h 1) (h 1)
Overliming 460 41 (rpm) (g L 1)
Overliming + Charcoal 250 68 47.3 70 8.50 0.280 0.060 2.78
Overliming + Na2SO3 409 48 48.0 100 11.00 0.320 0.065 3.97
Neutralization 450 42 48.0 130 8.97 0.255 0.075 5.16
 M. Telli-Okur, N. Eken-Saraçoğlu / Bioresource Technology 99 (2008) 2162–2169 2167

60 60
Total Reducing Sugar (g/L)

Total Reducing Sugar (g/L)

70 rpm pH=5.5
50 50
100 rpm pH=6
130 rpm pH=7
40 40

30 30

20 20

10 10

0 0
0 50 100 150 200 250 0 50 100 150 200 250
Time (hours) Time (hours)

12
25
70 rpm 10
20 100 rpm

Biomass (g/L)
130 rpm 8
Biomass (g/L)

15 6

10 4 pH=5.5
pH=6
2
5 pH=7
0
0 0 50 100 150 200 250
0 50 100 150 200 250 Time (hours)
Time (hours)
12
12
70 rpm
10 10
100 rpm
130 rpm
Ethanol (g/L)

8
Ethanol (g/L)

6 6

4 4
pH=5.5
2 2
pH=6
pH=7
0
0 50 100 150 200 250 0
0 50 100 150 200 250
Time (hours) Time (hours)
Fig. 3. Effect of shaking rate on: (a) reducing sugar consumption, (b) Fig. 4. Effect of initial pH on: (a) reducing sugar consumption, (b)
biomass growth, (c) ethanol formation during fermentation of sunflower biomass growth, (c) ethanol formation during fermentation of sunflower
seed hull hydrolysate at pH 6. seed hull hydrolysate at 100 rpm.

4–5.5 for alcohol fermentation with P. stipitis in synthetic

Table 4
medium. It is known that pKa of acetic acid is 4.76 at Effect of initial pH on ethanol fermentation of (detoxified by overliming
25 °C and the high amount of undissociated acetic acid combined with Na2SO3) sunflower seed hull hydrolysate with P. stipitis at
molecules in the hydrolysate medium have inhibitory effect 100 rpm
on cell growth under low pH. The acetic acid would freely Total reducing pH Maximum ethanol YP/S QP
diffuse across the cell membrane and once in the cytoplasm sugar (g L 1) (g L 1) (g g 1) (g L 1
h 1)
would ionize, thereby affecting the intracellular pH. As a 48 5.5 9.7 0.238 0.067
consequence, the energy and anabolic/catabolic pathways 48 6.0 11.0 0.320 0.065
can be uncoupled, resulting in a reduction of cell growth. 48 7.0 10.8 0.330 0.075
The inhibitory effect of acetic acid in hydrolysate fermenta-
tions can be minimized by maintaining a higher fermenta-
tion pH above the optimum range (Nigam, 2001). fermentation of sunflower seed hulls hydrolysate was
The effect of initial culture pH on ethanol production conducted between 5.5 and 7.0 pH, the fermentability of
was examined within the pH range 5.5–7.0 with P. stipitis hydrolysate was stable and no pronounced effect was
at shaking rate of 100 rpm (Fig. 4a–c). The kinetic para- observed.
meters belonging to fermentation experiments are given The high acid concentration (0.7 M) used in hydrolysis
in Table 4. As can be seen from our results, when the step in this study can be compensed by using relatively high
2168 M. Telli-Okur, N. Eken-Saraçoğlu / Bioresource Technology 99 (2008) 2162–2169

reaction temperature instead of 90 °C. To minimize the Converti, A., Dominguez, J.M., Perego, P., da Silva, S.S., Zilli, M., 2000.
amount of acid used in hydrolysate preparation, the cost Wood hydrolysis and hydrolysate detoxification for subsequent xylitol
production. Chem. Eng. Technol. 23 (11), 1013–1020.
effective separation and recycling of acid can be suggested. Davis, L., Jeon, Y.J., Svenson, C., Rogers, P., Pearce, J., Peiris, P., 2005.
In literature, it was reported that acid recycling is necessary Evaluation of wheat stillage for ethanol production by recombinant
for good economic performance (Hamelinck et al., 2005). Zymomonas mobilis. Biomass Bioenerg. 29, 49–59.
Recycling of acid also may reduce gypsum formation Delgenes, J.P., Moletta, R., Navarro, J.M., 1990. Acid hydrolysis of wheat
which has some value as an agricultural soil conditioner. straw and process considerations for ethanol fermentation by Pichia
stipitis Y-7124. Process Biochem. 25, 132–135.
The results of this study are in the range of those Diz, J., Cruz, J.M., Dominguez, H., Parajo, J.C., 2002. Xylitol production
obtained with the culture media made from hydrolysates from eucalyptus wood hydrolysates in low-cost fermentation media.
of other lignocellulosic wastes with P. stipitis (Nigam, Food Technol. Biotechnol. 40, 191–197.
2001; Sanchez et al., 2004). In literature, productivities of Du Preez, J.C., 1994. Process parameters and environmental factors
0.03–0.57 g L 1 h 1 and ethanol yields of 0.24–0.46 g g 1 affecting d-xylose fermentation by yeasts. Enzyme Microb. Technol.
16, 944–956.
were reported by different investigators. Ethanol yield from Du Preez, J.C., Bosch, M., Prior, B.A., 1986a. The fermentation of hexose
lignocellulosic feedstocks can be estimated from the U.S. and pentose sugars by Candida shehatae and Pichia stipitis. Appl.
Department of Energy website, which provides ‘‘Theoreti- Microbiol. Biotechnol. 23, 228–233.
cal Ethanol Yield Calculator’’ (EERE, 2007). Based on Du Preez, J.C., Bosch, M., Prior, B.A., 1986b. Xylose fermentation by
xylan content, potential theoretical ethanol yield from sun- Candida shehatae and Pichia stipitis: Effects of pH, temperature and
substrate concentration. Enzyme Microb. Technol. 8, 360–364.
flower seed hull is 0.144 L bioethanol per dry kg of feed- EERE, 2007. Theoretical ethanol yield calculator. Biomass Program, US
stock. In this study, about 51% of potential bioethanol Department of Energy, Energy Efficiency and Renewable Energy.
was reached with best ethanol concentration (11 g L 1). <http://www1.eere.energy.gov/biomass/ethanol_yield_calculator.html>.
This corresponds to 19.55 gallon per dry ton feedstock. Ferrari, M.D., Neirotti, E., Albornoz, C., Saucedo, E., 1992. Ethanol
When the ethanol yield per ton of feedstock is compared, production from eucalyptus wood hemicellulose hydrolysate by Pichia
stipitis. Biotechnol. Bioeng. 40, 753–759.
a ton of US corn can yield approximately 100 gallons of Gerçel, H.F., 2002. Production and characterization of pyrolysis liquids
ethanol. One ton of French sugar beets would be expected from sunflower- pressed bagasse. Bioresour. Technol. 85, 113–
to yield 25 gallons of ethanol and one ton of Brazilian sug- 117.
arcane would be expected to yield 20 gallons of ethanol Hahn-Hägerdal, B., Jeppson, H., Olsson, L., Mohagheghi, A., 1994. An
(USDA, 2006). Therefore, sunflower seed hull as a cheap interlaboratory comparison of the performance of ethanol–producing
micro-organisms in a xylose-rich acid hydrolysate. Appl. Microbiol.
by-product of edible oil factory can be used for bioethanol Biotechnol. 41, 62–72.
production in Turkey. Hamelinck, C.N., van Hooijdonk, G., Faaij, A.P.C., 2005. Ethanol from
lignocellulosic biomass: techno-economic performance in short-, mid-
4. Conclusion dle-and long-term. Biomass Bioenerg. 28, 384–410.
Hormitz, W. (Ed.), 1980. Official Methods of Analysis of The Association
of Official Analytical Chemists, 12th ed. Association of Official
The results of the fermentation of sunflower seed hull Analytical Chemists, Washington, DC.
hydrolysate by P. stipitis were presented and discussed in Jeffries, T.W., 1985. Emerging technology for fermenting d-xylose. Trends
this work regarding to the effects of different detoxification Biotechnol. 3, 208–212.
methods and fermentation conditions. In the light of the Kaylen, M.L., van Dyne, D., Choi, Y.S., Blasé, M., 2000. Economic
results obtained, it was clear that the sunflower seed hull feasibility of producing ethanol from lignocellulosic feedstocks.
Bioresour. Technol. 72, 19–32.
hydrolysate, which was prepared by 0.7 M H2SO4 at Kim, S., Dale, B.E., 2004. Global potential bioethanol production
90 °C and detoxified by overliming and Na2SO3 addition from wasted crops and crop residues. Biomass Bioenergy. 26, 361–
was suitable for ethanol production. It provides a new 375.
alternative for fuel production using edible oil factory Lin, Y., Tanaka, S., 2006. Ethanol fermentation from biomass resources :
wastes. However, to improve the fermentability of hydroly- current state and prospects. Appl. Microbiol. Biotechnol. 69, 627–
642.
sate, the optimization of fermentation conditions is Martinez, A., Rodriquez, M.E., York, S.W., Preston, J.F., Ingram, L.O.,
necessary. 2000a. Use of UV absorbance to monitor furans in dilute acid
hydrolysates of biomass. Biotechnol. Prog. 16, 637–641.
Acknowledgements Martinez, A., Rodriquez, M.E., York, S.W., Preston, J.F., Ingram, L.O.,
2000b. Effects of Ca (OH)2 treatments (Overliming) on the composi-
tion and toxicity of bagasse hemicellulose hydrolysates. Bitechnol.
The financial support provided by The Scientific and Bioeng. 69, 526–536.
Technical Research Council of Turkey (TÜBITAK,_
Martinez, A., Rodriquez, M.E., Wells, M.L., Sean, W.Y., Preston, J.F.,
_
MISAG-152) is gratefully acknowledged. Ingram, L.O., 2001. Detoxification of dilute acid hydrolysates of
lignocellulose with lime. Biotechnol. Prog. 17, 287–293.
References Miller, G.L., 1959. Use of dinitrosalicylic acid reagent for reducing sugar.
Anal. Chem. 31, 426–430.
Mussatto, S.I., Roberto, I.C., 2004. Alternatives for detoxification of
Clarck, T.A., Mackie, K.L., 1984. Fermentation inhibitors in wood
diluted – acid lignocellulosic hydrolysates for use in fermentative
hydrolysates derived from wood Pinus radiata. J. Chem. Technol.
process: a review. Bioresour. Technol. 93, 1–10.
Biotechnol. 34b, 101–110.
 M. Telli-Okur, N. Eken-Saraçoğlu / Bioresource Technology 99 (2008) 2162–2169 2169

Mutlu, S.F., 1990. Enzymatic conversion to sugars of sunflower stems and lysate with Pichia stipitis and Candida shehatae. Biotechnol. Lett. 22
seed hulls. PhD. Thesis, Ankara Univ. Institute of Science and (10), 855–858.
Technology. Sharma, S.K., Kalra, K.L., Grewal, H.S., 2002. Fermentation of
Nigam, J.N., 2001. Ethanol production from wheat straw hemicellulose enzymatically saccharified sunflower stalks for ethanol production
hydrolysate by Pichia stipitis. J. Biotechnol. 87, 17–27. and its scale up. Bioresour. Technol. 85, 31–35.
Nigam, J.N., 2002. Bioconversion of water-hyacinth (Eichornia crassipes) Sharma, S.K., Kalra, K.L., Kocher, G., 2004. Fermentation of enzymatic
hemicellulose acid hydrolysate to motor fuel ethanol by xylose – hydrolysate of sunflower hulls for ethanol production and its scale up.
fermenting yeast. J. Biotechnol. 97, 107–116. Biomass Bioenerg. 27, 399–402.
Nikakhtari, H., Hill, G.A., 2006. Closure effects on oxygen transfer and Slininger, P.J., Bothast, R.J., Cauwenberge, J.E.V., Kurtzman, C.P., 1982.
aerobic growth in shake flasks. Biotechnol. Bioeng. 95, 15–21. Conversion of D-xylose to ethanol by yeast Pachysolen tannophilus.
Palmqvist, E., Hahn-Hagerdal, B., 2000. Fermentation of lignocellulosic Biotechnol. Bioeng. 24, 371–384.
hydrolysates. I: inhibition and detoxification. Bioresour. Technol. 74, Slininger, P.J., Bothast, R.J., Ladisch, M.R., Okos, M.R., 1990. Optimum
17–24. pH and temperature conditions for xylose fermentation by Pichia
Parajo, J.C., Dominguez, H., Dominguez, J.M., 1997. Improved xylitol stipitis. Biotechnol. Bioeng. 35, 727–731.
production with Deboryomyces hansenii Y-7426 from raw or detoxified Sun, Y., Cheng, J.J., 2005. Dilute acid pretreatment of rye straw and
wood hydrolysates. Enzyme Microb. Technol. 21, 18–24. bermudagrass for ethanol production. Bioresour. Technol. 96, 1599–
Parajo, J.C., Dominguez, H., Dominguez, J.M., 1998. Biotechnological 1606.
production of xylitol. Part 3: Operation in culture media made from Taherzadeh, M.J., Niklasson, C., Liden, G., 2000. On-line control of fed-
lignocellulose hydrolysates. Bioresour. Technol. 66, 25–40. batch fermentation of dilute-acid hydrolyzates. Biotechnol. Bioeng. 69,
Parisi, F., 1989. Advances in lignocellulosics hydrolysis and in the 330–338.
utilization of the hydrolysates. Adv. Biochem. Eng. Biotechnol. 38, Tran, A.V., Chambers, R.P., 1986. Ethanol fermentation of red oak acid
53–87. prehydrolysate by the yeast Pichia stipitis CBS 5776. Enzyme Microb.
Roberto, I.C., Lacis, L.S., Barbosa, M.F.S., de Mancilha, I.M., 1991. Technol. 8, 439–444.
Utilization of sugar cane baggasse hemicellulose hydrolysate by USDA, 2006. The economic feasibility of ethanol production from sugar
Pichia stipitis for the production of ethanol. Proc. Biochem. 26, in the United States. Office of the Chief Economist, United States
15–21. Department of Agriculture (USDA). <http://www.usda.gov/oce/
Sanchez, G., Pilcher, L., Roslander, C., Modig, T., Galbe, M., Linden, G., reports/energy/index.htm>.
2004. Dilute acid hydrolysis for fermentation of the Bolivian straw Van Zyl, C., Prior, B.A., du Preez, J.C., 1991. Acetic acid inhibition of D-
material Brava. Bioresour. Technol. 93, 249–256. xylose fermentation by Pichia stipitis. Enzyme Microb. Technol. 13,
Saraçoğlu-Eken, N., Arslan, Y., 2000. Comparison of different pretreat- 82–86.
ments in ethanol fermentation using corn cob hemicellulosic hydro-