Bioresource Technology 101 (2010) 86648670

Contents lists available at ScienceDirect

Bioresource Technology
journal homepage: www.elsevier.com/locate/biortech

Effects of pretreatment methods for hazelnut shell hydrolysate fermentation

with Pichia Stipitis to ethanol
Yesim Arslan *, Nurdan Eken-Saraoglu
Department of Chemical Engineering, Gazi University, Maltepe 06570, Ankara, Turkey

a r t i c l e

i n f o

Article history:
Received 19 November 2009
Received in revised form 18 May 2010
Accepted 26 May 2010

Keywords:
Pichia stipitis
Hazelnut shell hydrolysate
Fermentation
Ethanol
Pretreatment

a b s t r a c t
In this study, we investigated the use of hazelnut shell as a renewable and low cost lignocellulosic material for bioethanol production for the rst time. High lignin content of hazelnut shell is an important
obstacle for such a biotransformation. Biomass hydrolysis with acids yields reducing sugar with several
inhibitors which limit the fermentability of sugars. The various conditioning methods for biomass and
hydrolysate were performed to overcome the toxicity and their effects on the subsequent fermentation
of hazelnut shell hydrolysate by Pichia stipitis were evaluated with shaking asks experiments. Hazelnut
shells hydrolysis with 0.7 M H2SO4 yielded 49 g l1 total reducing sugars and fermentation inhibitors in
untreated hydrolysate. First, it was shown that several hydrolysate detoxication methods were solely
inefcient in achieving cell growth and ethanol production in the fermentation of hazelnut shell hydrolysates derived from non-delignied biomass. Next, different pretreatments of hazelnut shells were considered for delignication and employed before hydrolysis in conjunction with hydrolysate detoxication
to improve alcohol fermentation. Among six delignication methods, the most effective pretreatment
regarding to ethanol concentration includes the treatment of shells with 3% (w/v) NaOH at room temperature, which was integrated with sequential hydrolysate detoxication by overliming and then treatment
with charcoal twice at 60 C. This treatment brought about a total reduction of 97% in furans and 88.4% in
phenolics. Almost all trialed treatments caused signicant sugar loss. Under the best assayed conditions,
ethanol concentration of 16.79 g l1 was reached from a hazelnut shell hyrolysate containing initial 50 g
total reducing sugar l1 after partial synthetic xylose supplementation. This value is equal to 91.25% of
ethanol concentration that was obtained from synthetic D-xylose under same conditions. The present
study demonstrates that Pichia stipitis is able to grow and ferment sugars to ethanol in detoxied hazelnut hydrolysate derived from delignied biomass.
2010 Elsevier Ltd. All rights reserved.

1. Introduction
With the depletion of the worlds petroleum supply, there has
been an increasing worldwide interest in alternative, non-petroleum based sources of energy. Ethanol derived from biomass has
the potential to be renewable transportation fuel that can replace
gasoline (Kim and Holtzapple, 2005). The use of bioethanol as a
source of energy would be more than just complementing solar,
wind and other renewable energy sources in the long run (Lin
and Tanaka, 2006). Moreover bioethanol can play an important
role in reducing green house gas emission. Ethanol use will increase because of its biodegradable, renewable and performance
qualities (Kumar et al., 2009). Ethanol is a high performance fuel
in internal combustion engines and burns relatively cleanly, especially as the amount of gasoline with which it is blended decreases
(Lynd, 1996).
* Corresponding author.
E-mail address: ysarslan@yahoo.com (Y. Arslan).
0960-8524/$ - see front matter 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biortech.2010.05.085

The largest potential feedstock for ethanol is lignocellulosic

biomass, which includes materials such as agricultural residues,
forest residues, wood, grass, waste paper and municipal wastes
(Noureddini and Byun, 2010). Basically, the lignocellulosic biomass composed of cellulose, hemicellulose and lignin. Cellulose
and hemicellulose are composed of mixture of carbohydrate polymers. Lignocellulosic biomass is an attractive material for bioconversion to ethanol because they are renewable, widespread and
cheap. One of the advantages of bioconversion with lignocellulosic materials is the opportunity to create biorenery producing
value-added co-products plus fuel bioethanol (Balat et al., 2008).
The main processing challenge in the ethanol production from
this resource is the feedstock pretreatment (Snchez scar and
Cardona, 2008).
The biological process for converting the lignocellulose to fuel
ethanol requires: (1) delignication to liberate cellulose and hemicellulose from their complex with lignin (2) depolimerization of
the carbohydrate polymers to produce free sugars and (3) fermentation of mixed hexose and pentose sugars to produce ethanol. The

Y. Arslan, N. Eken-Saraoglu / Bioresource Technology 101 (2010) 86648670

hydrolysis of lignocellulosic materials is usually catalyzed by acids

and enzymes. In general acid hydrolysis of lignocellulose is conducted with mineral acids such as H2SO4 or HCl (Laopaiboon
et al., 2010). In acid hydrolysis, the hemicellulose fraction is depolymerized at lower temperature than the cellulosic fraction to
xylose and other sugars (Chandel et al., 2007). The use of xylosefermenting yeasts such as Pichia stipitis that are able to co-ferment
xylose, glucose, mannose, galactose and cellobiose offers an opportunity for efcient utilization of hemicellulose derived hydrolyzates (Hernandez-Salas et al., 2009). Acid hydrolysis has some
limitations. If higher temperatures or longer residence time are applied, the hemicellulosic derived monosaccharides will degrade
and give rise to fermentation inhibitors. The major toxic compounds include furfural and hydroxymethylfurfural (sugar degradation products), acetic acid (released from the hemicellulosic
structure), several aromatic and phenolic compounds (lignin degradation products) and metallic ions (Klinke et al., 2004). These
compounds affect microbial metabolism and fermentative process
(Carvalho et al., 2004). The type and concentration of these compounds in hemicellulose hydrolysates depend upon the type of
raw material and hydrolysis conditions employed and their toxicity is the major factor limiting the bioconversion of lignocelluose
material (Hendriks and Zeeman, 2009; Laopaiboon et al., 2010).
Physical, physico-chemical, chemical and biological processes
have been used for pretreatment of lignocellulosic materials (Sun
and Cheng, 2002). Several techniques have been reported to overcome the inhibitory effect of these compounds during fermentation by yeasts, including adaptation of microorganism to the
medium (Roberto et al., 1991), treatments with molecular sieves,
ion-exchange resins or charcoal (Gong et al., 1993), steam stripping
and over titration (Roberto et al., 1994; 1995). Pretreatment must
improve the formation of sugars, avoid degradation or loss of carbohydrates, avoid the formation of by-products inhibitory to the
subsequent hydrolysis and fermentation processes and be costeffective. The utility of various agricultural residues for the production of organic fuels and chemicals has enormous potential for the
commercial applications. Numerous studies for developing largescale production of ethanol from lignocellulosics have been carried
out (Snchez scar and Cardona, 2008). The possible utilization of
inexpensive new feedstock requires the extensive evaluation of the
pretreatment conditions which is related to the nature and composition of lignocellulosic material. Studies have shown that pretreatment is the most signicant step for the cost and the success of the
cellulosic bioethanol technology (Balat et al., 2008).
Hazelnut shell can be one of the most important types of biomass, as it is an abundant, important agricultural and commercial
material in Turkey. At present, two-thirds of the world production
capacity of hazelnuts is provided by Turkey, with around 250 thousand tons of hazelnut shells per year (equivalent to 4.63  109 MJ)
produced in the Black-Sea region of Turkey alone (Dogru et al.,
2002). Elemental analysis of hazelnut shells was reported as
45.59% C, 4.59% H, 38.14% O, 1.26% ash and 10.07% moisture
(Midilli et al., 2000). Proximate analysis indicates that hazelnut
shells have 43.1% lignin, 27.5% hemicellulose, 24.7% cellulose,
3.4% alcohol-benzene extractives and 1.4% ash. Today, its main utilization remains as boiler fuel. Burning agricultural residues may
causes air pollution, soil erosion and a decrease in soil biological
activity (pr et al., 2007). Therefore any possible usage of hazelnut shells will yield economic as well as environmental dividends.
The conversion of hazelnut shell to useful chemicals such as acetic
acid, methanol (Ask et al., 1977), ammonia (Corlett, 1975), furfural
(Demirbas, 2006) and hydrogen (Midilli et al., 2000) has also been
investigated. No known effort has been made to utilize hazelnut
shell as a renewable and low cost lignocellulosic material for bioethanol production. High lignin content of hazelnut shell is an
important obstacle for such a biotransformation.

8665

The main objective of this work was to study effects of pretreatment methods on hazelnut shell hydrolysate fermentation to ethanol by Pichia stipitis. First the effect of hydrolysate detoxication
methods on the fermentability of hazelnut shell hydrolysate derived from non-delignied biomass was investigated. Next different pretreatments of hazelnut shells for delignication were also
employed before hydrolysis in conjunction with hydrolysate
detoxication to improve alcohol fermentation.
2. Methods
2.1. Microorganism and growth media
Pichia stipitis NRRL Y-7124 was grown at 30 C on agar slants.
The medium described by Slininger et al., (1982) was used for
growth. Inocula were prepared by transferring organism by loop
from two days slants to 250 ml Erlenmeyer asks containing
100 ml of growth medium. The inoculum medium consisted of
(g l1) 6.4 urea, 1.2 KH2PO4, 0.18 Na2HPO4, 10 yeast extract and
50 D-xylose (pH = 4.5). CaO, ZnO, FeCl3.6H2O, MgO, CuSO4.5H2O,
CoCl2.6H2O, H3BO3 were also included in the growth medium as
trace metals. The yeast was incubated aerobically with a magnetic
stirrer (600 rpm) at 30 C for 27 h prior to use.
2.2. Raw material
Hazelnut shells used in experiments were obtained from a local
plant in Dzce province in Turkey. Hazelnut shells were milled into
ne particles and screened into fractions (1.40.63 mm) to easy
reaction with acid. To reduce water content hazelnut shells were
dried in an oven at 105 C then 5.24% moisture content was measured. The Standard TAPPI method (Tappi, 1978) was applied to
determine the pentosan, which was found on dry basis as 29.26%
(w/w).
2.3. Delignication of hazelnut shell
To minimize the concentrations of fermentation inhibitors,
hazelnut shells were submitted to one of six delignication methods before acid hydrolysis (Table 1).
2.4. Preparation of hydrolysate samples
Hazelnut shell hemicellulosic hydrolysate was prepared from
non-delignied and delignied hazelnut samples in a glass reactor
equipped with a mechanically driven glass stirrer. The acid hydrolysis was conducted for 220 min with 0.7 M H2SO4 and a solid/liquid ratio of 1/5 (w/v). The temperature of hydrolysis was 90 C
and agitation was 100 rpm. At the end of hydrolysis, two fractions
were obtained and separated by ltration: a solid residue and a

Table 1
Delignication methods.
Treatment

Delignication methods

I
II
III
IV
V
VI

1% NaOH pretreatment at room temperaturea

1% NaOH pretreatment in an autoclaveb
3% NaOH pretreatment at room temperaturea
3% NaOH pretreatment in an autoclaveb
10% Hypochlorite pretreatment at room temperaturea
10% Hypochlorite pretreatment in an autoclaveb

a
Hazelnut shells were mixed with 1%(w/v) NaOH or 3%(w/v) NaOH or 10%(w/v)
hypochlorite solution by stirring at 1000 rpm and 18 C for 2 h. Then ltrated
hazelnut shells were washed until the wash water was neutral.
b
Hazelnut shells were mixed with 1%(w/v) NaOH or 3%(w/v) NaOH or 10%(w/v)
hypochlorite solution and kept in an autoclave for 2 h at 129 C and 2.57 atm. Then
ltrated hazelnut shells were washed until the wash water was neutral.

Y. Arslan, N. Eken-Saraoglu / Bioresource Technology 101 (2010) 86648670

8666

hydrolysate. Recovery of sugars from lignocellulose solubilization

was measured as the reducing sugar which primarily may contain
xylose and arabinose from hemicellulose and also glucose and cellobiose from cellulose. The untreated hydrolysate resulting from
non-delignied biomass contains 49 g total reducing sugar l1 before any detoxication of hydrolysate. Delignication treatments
also cause degradation or decomposition of polysaccharides in biomass which lowers the sugar recovery in hydrolysis step (Hendriks
and Zeeman, 2009). Detoxication treatment of hydrolysates prior
to fermentation also results in different sugar loss. To see exact effect of the remaining inhibitory constituents and to produce a
hydrolysate with a suitable sugar concentration for fermentation,
initial reducing sugar level was adjusted to 50 g l1 with synthetic
xylose supplementation into hydrolysates before each fermentation. The treated hydrolysates were further supplemented with
additional nutrients to maintain the growth medium composition.
The hazelnut shell hydrolysate and the remaining nutrients were
separately autoclaved and combined after sterilization. Synthetic
media were prepared by dissolving known amounts of xylose in
water and complemented with the same nutrients as hemicellulosic hydrolysates to compare the fermentation parameters.

2.7. Analytical methods

Biomass concentration was determined turbidiometrically at
600 nm and converted to the biomass dry weight. Total reducing
sugars were determined colorimetrically using dinitrosalicylic acid
reagent (Miller, 1959). Ethanol was measured by the dichromate
oxidation method which is based on the complete oxidation of ethanol by dichromate in the presence of sulfuric acid (Horwitz,
1980). Total furans in hydrolysate samples were estimated by a
spectrophotometric method based on the difference in absorbance
at 284 and 320 nm using a Hach DR/4000 spectrophotometer
(Hach Company, PO Box 389, Loveland, CO 8059) (Martinez et al.,
2000) before and after any pretreatment. Absorbances of each
hydrolysates at 280 nm were also measured with a spectrophotometer and compared with the absorbance of the untreated
hydrolysate. From the results, R280 were calculated by Eq. (1) as
shown below. From this value the effect of any treatment on
removing the furan and phenolic compounds from the hydrolysate
in comparison to untreated hydrolysate can be approximately
evaluated (Miyafuji et al., 2002).

R280

Absorbance of the hydrolysate after any treatment

Absorbance of the untreated hydrolysate

2.5. Detoxication of hydrolyzates and integrated treatments

The hydrolysates prepared from non-delignied biomass were
subjected to many detoxication methods (Treatments A, B, C, D,
E, F and G) before fermentation. These seven treatments were summarized in Table 2. Two step treatment methods (Treatments IVI)
as cleaning of hazelnut shells before hydrolysis followed by detoxication of hydrolysate were given in Table 3. Six different delignication treatments were performed for conditioning of hazelnut
shells prior to acidic hydrolysis. For hydrolysates derived from delignied hazelnut shells, a single detoxication method was employed in sequential manner after delignication of biomass.
Each hydrolysate was treated with CaO (76.66 g l1) until pH 10
and ltered and acidied to pH 6 with concentrated sulfuric acid.
After that overlimed hydrolysate was mixed with charcoal
(20 g l1) and stirred during one hour at 60 C twice. The hydrolysate was recovered with ltration.

2.6. Shaking ask experiments

All treated hydrolysates containing additional nutrients were
fermented by Pichia stipitis at 30 C in 250 ml shake asks having
135 ml medium at 150 rpm and pH 6. Hydrolysate fermentation
medium was inoculated with 10% (v/v) growth culture.

3. Results and discussion

3.1. Effect of hydrolysate detoxication
Several by-products of sugar and lignin degradation formed
during the hydrolysis process are toxic to the microbial metabolism and have negative effect on the fermentation efciency. The
inhibitors such as furfural, 5-HMF (hydroxymethylfurfural) and total phenols were considered toxic to microorganism. Acid hydrolysis of non-delignied hazelnut shell caused a release of reducing
sugar of 49 g l1 along with fermentation inhibitors (Table 2) such
as furans 116 mg l1 in untreated hydrolysate. R280 of untreated
hydrolysate was assigned as unity to see the effect of any treatment on removing the furans and phenolic compounds. In order
to overcome adverse effects, in a rst series of experiments, various
detoxications of hydrolysate samples were performed following
acidic hydrolysis with non-delignied biomass. Recovered reducing sugar level, total furan left and R280 after each detoxication
were summarized in Table 2. It is well known that the effectiveness
of a detoxication method depends on the type of toxic constituents present in hydrolysate which varies according to the raw
and hydrolysis conditions (Carvalho et al., 2006).
It seems that overliming (Treatment B) led to a drastic change in
furans and phenolics comparing to neutralization (Treatment A).

Table 2
Detoxication methods of hydrolysates prepared from non-delignied hazelnut shells.

Untreated
Treatment
Treatment
Treatment
Treatment
Treatment
Treatment
Treatment
Treatment

A
B
C
D
E
F
G
H

Hydrolysate detoxication

Recovered reducing
sugar after hydrolysis (g l1)

Total furan (mg/l)

R280

None
Neutralization to pH 6 with CaO
Overliming with CaOa
10 g l1 charcoal treatment one hour at 30 C, pH 2 and overliming
20 g l1 charcoal treatment one hour at 30 C, pH 2 and overliming
30 g l1 charcoal treatment one hour at 30 C, pH 2 and overliming
Overliming and 20 g l1 charcoal treatment one hour at 30 C, pH 6
Overliming and ethyl acetate treatment one hour with a volume ratio of 1:1
Overliming and 20 g l1 charcoal treatment one hour at 60 C, pH 6 twice

49.00
43.00
38.00
35.40
35.00
35.00
32.52
33.57
26.30

116.0
97.0
57.0
29.0
7.8
9.7
5.0
25.8
1.1

1.000
0.896
0.567
0.302
0.134
0.142
0.056
0.288
0.021

a
CaO addition until pH 10 and ltered and acidied to pH 6 with concentrated sulfuric acid. Analysis were carried out in triplicate. The values are mean of three replicates.
Standard deviation was within 10%.

Y. Arslan, N. Eken-Saraoglu / Bioresource Technology 101 (2010) 86648670

8667

Table 3
Delignication methods of hazelnut shells and effects of different treatments.
Hazelnut shell treatment

Treatment I
Treatment II
Treatment IIIb
Treatment IV
Treatment V
Treatment VI

1 wt.% NaOH pretreatment

at room temperature
1 wt.% NaOH pretreatment
in an autoclave
3 wt.% NaOH pretreatment
at room temperature
3 wt.% NaOH pretreatment
in an autoclave
Hypochlorite pretreatment
at room temperature
Hypochlorite pretreatment
in autoclave

Recovered reducing
sugar after
hydrolysis (mg l1)

Total furan (mg l1)

R280

Recovered
reducing
sugar after
detoxication
(g l1)

Total furan in
hdrolysate after
detoxication
(mg l1)

R280

27.54

60.4

0.64

21.30

2.55

0.029

28.86

78.4

0.80

13.45

3.42

0.035

26.40

82.6

0.83

13.61

2.70

0.027

27.17

93.0

0.95

17.10

2.94

0.037

49.00

114.0

1.00

47.24

97.35

0.465

40.00

95.0

1.00

33.73

9.03

0.100

Overliming with CaO and charcoal treatment at 60 C twice.

Treatment III experiment was performed in duplicate. All analysis was carried out in triplicate. The values are mean of three replicates. Standard deviation was within
10%.
b

Combination of overliming method with other treatments (in

Treatments CH) seems to yield a markedly benecial effect in
the removal of inhibitors. Furans concentrations and R280 values
in hydrolysate were found to be decreased in different amounts
with the sequential use of charcoal treatment before and after
overliming. Table 2 shows that the efciency of charcoal treatment
in removal of furan and phenolic compounds (R280) is changing
with charcoal amount, order of treatments, treatment temperature
and treatment number (Treatments C, D, E, F, and H). It is clear that
ethyl acetate extraction (Treatment G) wasnt as effective as charcoal in removing of phenolics in hazelnut shell hydrolyzate. After
each treatment signicant sugar loss took place. Sugar removals
varied for each detoxication procedure employed. All lime additions (Treatments BH) caused a progressive decline in sugar content, which correlated with an increase in pH after overliming.
Above pH 9.0 the rate of sugar loss increased dramatically. Amartey and Jeffries (1996) found overliming resulted in the loss of
D-glucose (14%), D-xylose (4%), L-arabinose (8%) and acetic acid
(43%). Treatment H which is detoxication with overliming in
combination with twice charcoal treatment promotes the highest
inhibitor removal and sugar loss. Even though, at the beginning
of the fermentations of hydrolysates subjected to Treatments B,
C, D, E, F, and G the concentration of furan was below the thresholds of inhibition (Almeida et al., 2007) and phenolic compounds
content were reduced to 2.156.7% of untreated hydrolysate level
no signicant cell growth and ethanol production were observed
with very little sugar metabolization. Only, fermentation of hazelnut hydrolysate subjected to Treatment H gave 5 g l1 ethanol
while cell growth was nearly inhibited and 48% sugar consumption
occurred (Arslan, 2007).
Twenty-one other hydrolysate detoxication methods performed (Arslan, 2007) not mentioned here had no positive impact
on fermentability of hazelnut shell hydrolysate. Vogel-Lohmeier
et al. (1998) reported the high sensitivity of Pichia stipitis to inhibitors and the effect of toxic compounds on xylose metabolism of
yeasts is very complex. They found some evidence of synergistic
inhibition effect of toxic compounds to cell metabolism. Other
researchers also demonstrated synergistic effects of HMF and furfural (Taherzadeh et al., 2000). However recently it was reported
that Pichia stipitis reduces the aldehyde group in the furan ring of
HMF and furfural (Liu et al., 2005). The inhibition could probably
be attributed to phenolic compounds. Lignin derived phenolic
compounds such as vanillin and syringaldehyde (Tran and Chambers, 1986) were identied as some of the most toxic components.

Phenolic inhibitors may act on biological membranes, causing loss

of integrity (Heipieper et al., 1994) which can explain the absence
of growth in hydrolysates derived from hazelnut shells with very
high lignin content while furan levels were below 100 mg l1 in
our study. This result brought the delignication of hazelnut shells
before hydrolysis into consideration.
3.2. Effect of combination of delignication methods and detoxication
Lignin content of hazelnut shell is very high (43.1% w/w). Fermentation trials in previous section indicated that the lignin content in hazelnut shell is relatively more important than other
structural feature on fermentation performance. In order to decrease the inuence of toxins coming from lignin, hazelnut shells
were conditioned before acidic hydrolysis.
Six different delignication treatments of hazelnut shells were
performed (Table 1). Hydrolysis conditions (acid concentration,
temperature, solid/liquid ratio, time) of delignied biomass were
same with non-delignied biomass. Comparison of recovered total
reducing sugars in acidic hydrolysate (26.4028.86 g reducing sugar l1) derived from alkaline treated biomass (Treatments IIV)
and with total reducing sugars in untreated hydrolysate (49 g
reducing sugar l1) showed a strong evidence of polysaccharides
loss as well as lignin solubilization (Hendriks and Zeeman, 2009).
Xylans can be selectively removed by peeling and hydrolytic reactions in alkaline treatment of hazelnut shell which leads to losses
of fermentable sugar for ethanol production. Prior to fermentation,
all hydrolysate samples were detoxied with Treatment H which
promoted the best results in fermentations with non-delignied
biomass hydrolysates in previous section. Sequential application
of pretreatment methods caused severe reduction in total sugar
left in hydrolysates. Fig. 1 shows sugar loss as percent of reducing
sugar (49.0 g l1) in original untreated hydrolysate after each treatment stage. Total loss in reducing sugar concentration varied between 3.57% and 72.63%. It is noteworthy that the most reducing
sugar loss occurred in Treatments II and III while Treatment V
led to minimal substrate diminishment. It was clear that the loss
of fermentable sugars is the major drawback of delignication
and detoxication treatments.
Fig. 1 also show total furan and phenolic removals caused by
various treatments. All delignication methods were capable of
reducing notably the concentrations of furan except hypochlorite
pretreatment at room temperature in Treatment V. It can be seen
that as a means of total furan and phenolics removal the best

Y. Arslan, N. Eken-Saraoglu / Bioresource Technology 101 (2010) 86648670

Sugar Loss or Furan Removal or Removal of

phenolics ( % of untreated hydrolysate)

8668

100
Detoxification
De-lignification

90
80
70
60
50
40
30
20
10
0
I

II

III

IV

VI

Treatments
Fig. 1. Loss of reducing sugar concentration, removal of total furan and phenolics.

delignication treatment was found as 1% NaOH pretreatment at

room temperature in Treatment I. Delignication method in Treatment V was inefcient both for furan and phenolic removals. However, when hypochlorite used in autoclave furan removal was
observed (Treatment VI). This indicated that hypochlorite treatment was only effective with high temperatures for furan removal.
Integration of delignication and detoxication in Treatments I, II,
III, and IV provided the best results.
3.3. Fermentation of Hazelnut Hydrolysate
Before the fermentation all treated hydrolysates were supplemented with synthetic xylose to reach 50 g l1 reducing sugar concentrations. Composition of different fermentation media used in
this report were described in Table 4. Figs. 24 show time course
for biomass growth, total reducing sugar concentration and ethanol level in treated hydrolysates and synthetic medium with Pichia
stipitis. As expected, Treatment V and Treatment VI were inefcient
for biomass growth and ethanol production with very low sugar
consumption. This result emphasizes the importance of the effec-

tiveness of feedstock pretreatment before hydrolysis. Results show

that cell growth was close for hydrolysates when Treatments IIV
were employed and inferior compared to the synthetic medium
(Fig. 2). Independent of the treatments, similar patterns of reducing sugar consumption were observed for hydrolysates subjected
to Treatments IIV (with mixed sugars) and for control run (with
xylose only) (Fig. 3). It should be pointed out that delignication
of hazelnuts prior to hydrolysis and detoxication of hydrolysates
before fermentation in Treatments IIV results in a better fermentability but still less than synthetic medium (Fig. 4).
The fermentation performance of Pichia stipitis in treated
hydroysates were also compared in Table 4. In studying ethanol production by Pichia stipitis, the fermentation of hydrolysate prepared
from hazelnut shells delignied with 3% NaOH at room temperature
and detoxied with overliming and charcoal (Treatment III)
resulted maximum ethanol concentration as 16.79 g l1. This treatment brought about a total reduction of 88.4% in furans and 97% in
phenolics. Maximum alcohol level reached with hydrolysate is
equal to 91.25% of ethanol concentration that was obtained from
synthetic xylose under the same conditions (18.40 g l1) with this

Table 4
Fermentation parameters.
Run No

a
b
c

Treatment

II

IIIc

IV

VI

Control

Hydrolysatea Identity

A mixture of hazelnut hydrolysate

(21.30 g reducing sugar l1) and
externally added xylose (28.70 g l1)
A mixture of hazelnut hydrolysate
(13.45 g reducing sugar l1) and
externally added xylose (36.55 g l1)
A mixture of hazelnut hydrolysate
(13.61 g reducing sugar l1)and
externally added xylose (36.39 g l1)
A mixture of hazelnut hydrolysate
(17.10 g reducing sugar l1) and
externally added xylose (32.9 g l1)
A mixture of hazelnut hydrolysate
(47.24 g reducing sugar l1)and
externally added xylose (2.76 g l1)
A mixture of hazelnut hydrolysate
(33.73 g reducing sugar l1)and
externally added xylose (16.27 g l1)
Xylose (50 g l1)
1

Ratio of reducing
sugar originated
from hazelnut shell

Ethanol
produced

0.43

Yp/s
(max ethanol/consumed
reducing sugar)

Qp
(max ethanol
conc /time)

10.12

0.282

0.112

0.27

16.10

0.415

0.179

0.27

16.79

0.432

0.186

0.34

15.64

0.449

0.174

(g l1)

0.95

4.440

0.218

0.035

0.68

4.67

0.334

0.052

0.44

0.194

0.00

18.4

Total reducing sugar: 50 g l .

Cultivation conditions employed: 135 ml media in 250 ml asks, 150 rpm, and pH 6, 30 C.
Treatment III and Control run experiments were performed in duplicate. The values are mean of two replicates. Standard deviation was within 10%.

Y. Arslan, N. Eken-Saraoglu / Bioresource Technology 101 (2010) 86648670

II

III

IV

VI

8669

control

14
12

Biomass (g/l)

10
8
6
4
2
0
0

24

48

67

90

125

t (h)
Fig. 2. Effect of detoxication methods employed on growth.

II

III

IV

IV

control

Total reducing sugar (g/l)

60
50
40
30
20
10
0
0

24

48

67

90

t (h)
Fig. 3. Effect of detoxication methods employed on reducing sugar consumption.

II

III

IV

VI

control

25

Ethanol (g/l)

20
15
10
5
0
0

24

48

67

90

125

t (h)
Fig. 4. Effect of detoxication methods employed on ethanol formation.

method ethanol yield was 0.432 g g1. Fermentation results of

hydrolysates subjected to Treatment II and IV were comparable to
Treatment III. The fermentation parameters such as ethanol yields
(YP/S) and productivity (QP) are also given in Table 4. Treatment IV
gave best outcome regarding to ethanol yield which is 0.449 gg1.
The most efcient treatment for productivity Treatment III with this
method productivity was 0.186 g l1 h1. Same conditions with
synthetic medium, productivity was 0.194 g l1 h1. It appears in
Table 4 that ethanol levels as well as yields and productivities for

hydrolysate fermentations may depend on substrate composition

in hydrolysate. Even though initial reducing sugar concentration
was maintained at 50 g l1 by supplementation of xylose in each
hydrolysate prior to fermentation, carbohydrate constituents were
not same. Slininger et al. (2008) in their work with Pichia sitipitis
suggested carbon sources may change the ability of a culture to resist and adapt to inhibitor stress additional to other fermentation
conditions. It is not clear that beside very low residual inhibitory
compounds whether initial high concentration of xylose in the

8670

Y. Arslan, N. Eken-Saraoglu / Bioresource Technology 101 (2010) 86648670

hydrolysates treated with Treatments II, III, IV played a role in the

direction of carbon toward ethanol production.
The present study demonstrates that some pretreatments of
hazelnut shell before the hydrolysis result in a better growth of
microorganism and fermentability of hydrolysate. However, the
fermentation parameters for these treated hydrolysates were generally lower than those obtained with synthetic medium. The raw
material and process steps before fermentation will strongly inuence the composition and the fermentability of the hydrolysate. It
is difcult to make comparison but the result of this study is in the
range of those obtained by other researchers. In literature, ethanol
yields of 0.130.44 g g1 and maximum ethanol concentration of
3.118 g l1 were reported by different investigators (Wilson
et al., 1989; Nigam, 2002; Agbogbo and Wenger, 2007; Kumar
et al., 2009).
4. Conclusions
For the rst time in this study, hazelnut shells containing
high lignin content were evaluated as substrates for ethanol production. The results of fermentation of hazelnut shell hydrolysate
by Pichia stipitis were presented regarding to the effect of pretreatment methods of hazelnut shells and hazelnut shell hydrolysate. The present study demonstrates that Pichia stipitis is
able to grow and ferment sugars to ethanol in detoxied hazelnut shell hydrolysate derived from delignied biomass. Removing of inhibitors with pretreatments is essential and key factor
in the biological conversion of hazelnut shells to ethanol. A wide
range of pretreatments targeting for conditioning of hydrolysate
or hazelnut shells have been trialed. The results indicated that
hydrolysate detoxication such as overliming and charcoal shaking was seen an effective way to remove some toxins but not solely enough for the growth of yeast and alcohol production in
hazelnut shell hydrolysate. A good fermentability could be
achieved if lignin content of shells was removed with 3% NaOH
treatment before hydrolysis of biomass. These delignication
and detoxication treatments were carefully integrated to realize
better cellulosic ethanol production from hazelnut shells. Ethanol
yield obtained from this method was 0.084 g ethanol/g hazelnut
shell.
Acknowledgements
We would like to thank Dr.Mbeccel Ergun and Mjgan TelliOkur, Faculty of Engineering, Gazi University for their valuable efforts in the review of this paper and helpful suggestions.
References
Agbogbo, F.K., Wenger, K.S., 2007. Production of ethanol from corn stover
hemicellulose hydrolysate using Pichia stipitis. J. Ind. Microbiol. Biotechnol.
34, 723727.
Almeida, J.R.M., Modig, T., Petersson, A., Hhn-Hgerdal, B., Liden, G., GorwaGrauslund, M.F., 2007. Increased tolerance and conversion of inhibitors in
lignocelllulosic hydrolysates by Saccharomyces cerevisiae. J. Chem. Technol.
Biotechnol. 82, 340349.
Amartey, S., Jeffries, T., 1996. An improvement in Pichia stipitis fermentation of acid
hydrolyses hemicellulose achieved by overliming (calcium hydroxide
treatment) and strain adaptation. World J. Microbiol. Biotechnol. 12, 281283.
Arslan, Y., 2007. The utilization of hazelnut shell for ethanol production. Ph.D.
Thesis, Gazi University Insttue of Science and Technology.
Ask, M., Deymer, J., Glensoy, H., 1977. Utilization of hazelnut shell. Chim. Acta.
Turc. 5, 2742.
Balat, M., Balat, H., z, C., 2008. Progress in bioethanol processing. Progr. Energ.
Combust. Sci. 34, 551573.
Carvalho, W., Canilho, L., Mussatto, S.F., Dragone, G., Morates, M.L.V., Solenzal, A.I.N.,
2004. Detoxication of sugarcane bagasse hemicellulosic hydrolysate with ionexchange resins for xylitol production by calcium alginate-entrapped cells. J.
Chem. Technol. Biotechnol. 79, 863868.

Carvalho, G.B.M., Musasatto, S.I., Cndido, E.J., Silva, J.B.A., 2006. Comparison of
different procedures for the detoxication of eucalyptus hemicellulosic
hydrolysate for use in fermentative processes. J. Chem. Technol. Biotechnol.
81, 152157.
Chandel, A.K., Kapoor, R.K., Singh, A., Kuhad, R.C., 2007. Detoxication of sugarcane
bagasse hydrolysate improves ethanol production by Candida shehatae NCIM
3501. 2007. Bioresour. Technol. 98, 19471950.
Corlett, R.F., 1975. Conversion of seattles solid waste to methanol or ammonia. The
trend in engineering, University of Washington D.
pr, Y., Gler, C., Akgl, M., Tascoglu, C., 2007. Some chemical properties of
hazelnut husk and its suitability for particleboard production. Build. Environ.
42, 25682572.
Demirbas, A., 2006. Furfural production from fruit shells by acid catalyzed
hydrolysis. Energ. Sourc. 28, 157165.
Dogru, M., Howarth, C.R., Akay, G., Keskinler, B., Malik, A.A., 2002. Gasication of
hazelnut shells in a down draft gasier. Energy 27, 415427.
Gong, C.S., Chen, C.S., Chen, L.F., 1993. Pretreatment of sugar cane bagasse
hemicellulose hydrolyzate for ethanol production by yeast. Appl. Biochem.
Biotechnol. 39 (40), 8388.
Heipieper, H.J., Weber, F.J., Sikkema, J., Keweloh, H., De Bont, J.A.M., 1994.
Mechanisms of resistance of whole cells to toxic organic solvents. Trends
Biotechnol. 12, 409415.
Hendriks, A.T.W.M., Zeeman, G., 2009. Pretreatments to enhance the digestibility of
lignocellulosic biomass. Bioresour. Technol. 100, 1018.
Horwitz, W., 1980. Dicromate oxidation methods. J. of AOAC. 13, 2527.
Kim, S., Holtzapple, M.T., 2005. Lime pretreatment and enzymatic hydrolysis of corn
stover. Bioresour. Technol. 96, 19942006.
Klinke, H.B., Thomson, A.B., Ahring, B.K., 2004. Inhibition of ethanol-producing yeast
and bacteria by degradation products produced during pre-treatment of
biomass. Appl. Microbiol. Biotechnol. 66, 1026.
Kumar, A., Singh, L.K., Ghosh, S., 2009. Bioconversion of lignocellulosic fraction of
water- hyacinth (Eichhornia crassipes) hemicellulose acid hydrolysate to
ethanol by Pichia stipitis. Bioresour. Technol. 100, 32933297.
Laopaiboon, P., Thani, A., Leelavatcharamas, V., Laopaiboon, L., 2010. Acid hydrolysis
of sugarcane bagasse for lactic acid production. Bioresour.Technol. 101, 1036
1043.
Lin, Y., Tanaka, S., 2006. Ethanol fermentation from biomass researches: current
state and prospects. Appl. Microbiol. Biotechnol. 69, 627642.
Liu, Z.L., Slininger, P.J., Gorsick, S.W., 2005. Enhanced biotransformation of furfural
and hydroxymethylfurfural by newly developed ethanologenic yeast strains.
Appl. Biochem. Biotechnol. 121 (124), 451460.
Lynd, L.R., 1996. Overview and evaluation of fuel ethanol from cellulosic biomass:
technology, economics, the environment and policy. Annu. Rev. Energ. Environ.
21, 403465.
Martinez, A., Rodriguez, M.E., York, S.W., Preston, J.F., Ingram, L.O., 2000. Use of UV
absorbance to monitor furans in dilute acid hydrolysates of biomass.
Biotechnol. Progr. 16, 637641.
Midilli, A., Rzayev, P., Olgun, H., Ayhan, T., 2000. Solar hydrogen production from
hazelnut shells. Int. J. Hydrogen Energ. 25, 723732.
Miller, G.L., 1959. Use of dinitrosaliciyle acid reagent for reducing sugar. Anal.
Chem. 31, 426430.
Miyafuji, H., Danner, H., Neuretier, M., Thoasser, C., Bvochora, J., Szola, O., Braun, R.,
2002. Detoxication of wood charcoal for increasing the fermentability of
hydrolysates. Enzyme Microb. Technol. 6234, 15.
Nigam, J.N., 2002. Bioconversion of water-hyacinth (Eichhornia crapsipes)
hemicellulose acid hydrolysate to motor fuel ethanol by xylose-fermenting
yeast. J. Biotechnol. 97, 107116.
Noureddini, H., Byun, J., 2010. Dilute-acid pretreatment of distillers grains and corn
ber. Bioresour. Technol. 101, 10601067.
Roberto, I.C., Mancilha, I.M., Souza, C.A.D., Felipe, M.G.A., Sato, S., Castro, H.F.D.,
1994. Evaluation of rice straw hemicellulose hydrolysate in the production of
xylitol by Candida guilliermondii. Biotechnol. Lett. 16, 12111216.
Roberto, I.C., Felipe, M.G.A., de Mancilha, I.M., Vitola, M., Sato, S., Silva, S.S., 1995.
Xylitol production by Candida guilliermondii as an approach for the utilization of
agroindustrial residues. Bioresour. Technol. 51, 255257.
Snchez scar, J., Cardona, C.A., 2008. Trends in biotechnological production of fuel
ethanol from different feedstocks. Bioresour. Technol. 99, 52705295.
Slininger, P.J., Bothast, R.J., Cau Wenberge, J.E.V., Kurtzman, C.P., 1982. Conversion of
D- xylose to ethanol by the yeast Pachysolen tannophilus. Biotechnol. Bioeng. 24,
371384.
Slininger, P.J., Gorsich, S.W., Liu, Z.L., 2008. Culture nutrition and physiology impact
the inhibitor tolerance of the yeast Pichia stipitis NRRL Y-7124. Biotechnol.
Bioeng. 102, 778790.
Sun, Y., Cheng, J., 2002. Hydrolysis of lignocellulosic materials for ethanol
production a review. Bioresour. Technol. 83, 111.
Taherzadeh, M.J., Gustafsson, L., Niklasson, C., Liden, G., 2000. Physiological effects
of 5- hydroxymethylfurfural on Saccharomyces cerevisiae. Appl. Microbiol.
Biotechnol. 53, 701708.
Tappi, 1978. Pentosans in wood and pulp, Tappi standart.
Tran, A.V., Chambers, R.P., 1986. Ethanol fermentation of red oak acid
prehydrolysate by yeast Pichia stipitis NRRL Y-7124. Enzyme Microb. Technol.
8, 439444.
Wilson, J.J., Deschalelets, L., Nishikawa, N.K., 1989. Comparative fermentability of
steam- pretreated aspenwood hemicellulose by Pichia stipitis CBS 5776. Appl.
Microb. Biotechnol. 31, 592596.