Bioresource Technology 100 (2009) 19831991

Contents lists available at ScienceDirect

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

Biodiesel production from sunower, soybean, and waste cooking oils by

transesterication using lipase immobilized onto a novel microporous polymer
Nadir Dizge a, Coskun Aydiner a, Derya Y. Imer a, Mahmut Bayramoglu b, Aziz Tanriseven c,
Blent Keskinler a,*
a

Gebze Institute of Technology, Department of Environmental Engineering, Gebze 41400, Turkey

Gebze Institute of Technology, Department of Chemical Engineering, Gebze 41400, Turkey
c
Gebze Institute of Technology, Department of Biochemistry, Gebze 41400, Turkey
b

a r t i c l e

i n f o

Article history:
Received 22 July 2008
Received in revised form 10 October 2008
Accepted 12 October 2008
Available online 22 November 2008
Keywords:
Biodiesel
Transesterication
Microporous polymeric biocatalyst
Thermomyces lanuginosus lipase
Taguchi methodology

a b s t r a c t
This study aims at carrying out lipase-catalyzed synthesis of fatty acid methyl esters (biodiesel) from
various vegetable oils using lipase immobilized onto a novel microporous polymeric matrix (MPPM) as
a low-cost biocatalyst. The research is focused on three aspects of the process: (a) MPPM synthesis
(monolithic, bead, and powder forms), (b) microporous polymeric biocatalyst (MPPB) preparation by
immobilization of lipase onto MPPM, and (c) biodiesel production by MPPB. Experimental planning of
each step of the study was separately carried out in accordance with design of experiment (DoE) based
on Taguchi methodology.
Microporous polymeric matrix (MPPM) containing aldehyde functional group was synthesized by
polyHIPE technique using styrene, divinylbenzene, and polyglutaraldehyde. Thermomyces lanuginosus
lipase was covalently attached onto MPPM with 80%, 85%, and 89% immobilization efciencies using
bead, powder, and monolithic forms, respectively. Immobilized enzymes were successfully used for
the production of biodiesel using sunower, soybean, and waste cooking oils. It was shown that immobilized enzymes retain their activities during 10 repeated batch reactions at 25 C, each lasting 24 h. Since
the developed novel method is simple yet effective, it could have a potential to be used industrially for
the production of chemicals requiring immobilized lipases.
2008 Elsevier Ltd. All rights reserved.

1. Introduction
In recent years, demand for fatty acid methyl esters (biodiesel)
being used as fuel in diesel engines and heating systems increased
due to recent rises in petroleum prices, increments in earth population and energy necessity, and development of government measures such as The EU Directive 2003/30/EC and The US Energy
Policy Act (EPAct, 1992) (Vicente et al., 2007). Transesterication
of vegetable oils to obtain biodiesel consists in replacing the glycerol
of triglycerides with a short chain alcohol in the presence of a catalyst. Transesterication reaction can be catalyzed by both homogeneous (basic or acidic) and heterogeneous (basic, acidic or
enzymatic) catalysts (Mittelbach, 1990). Recent studies show that
biodiesel can be produced enzymatically by lipase-catalyzed transesterication which has become more attractive in biodiesel production since the glycerol can be recovered easily and the purication
process for biodiesel is simple (Vicente et al., 1998). In addition,
the use of lipase in biodiesel production tolerates the water content
of oil and increases biodiesel yield by avoiding the soap formation.
* Corresponding author. Tel.: +90 262 605 32 07; fax: +90 262 605 32 05.
E-mail address: bkeskinler@gyte.edu.tr (B. Keskinler).
0960-8524/$ - see front matter 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biortech.2008.10.008

Selection of an immobilization strategy greatly inuences the

properties of biocatalyst. The decrement levels in activity and diffusion limitations occurring with immobilization are mainly
dependent on the properties of support material and the immobilization method. Support materials playing an important role in the
usefulness of an immobilized enzyme should be low-cost, and provide adequate large surface area together with the least diffusion
limitation in the transport of substrate and product for enzymatic
reactions (Arica et al., 2000). Enzymatic biodiesel production was
studied using lipases immobilized onto different supports such as
macroporous resin by adsorption (Yang et al., 2006), hydrophobic
solgel support by entrapment (Noureddini et al., 2005), silica
aerogel by encapsulation (Oraire et al., 2006) and chitin by chemical binding (Gomes et al., 2004).
The present work is on immobilization of Thermomyces lanuginosus lipase onto a novel microporous polymer in different forms
(bead, powder, and monolithic) having functional aldehyde groups
and its usage for biodiesel synthesis using sunower, soybean, and
waste cooking oils. The novel polymer synthesis, lipase immobilization, and biodiesel production were planned using Taguchi 2n
designs since a great number of independent process variables
are involved.

1984

N. Dizge et al. / Bioresource Technology 100 (2009) 19831991

Table 1
Fatty acid compositions of the sunower, soybean, and waste cooking oil.
Fatty acid oil

Sunower oil

Soybean oil

Waste cooking oil

Lauric (12:0)
Myristic (14:0)
Palmitic (16:0)
Palmitoleic (16:1)
Stearic (18:0)
Oleic (18:1)
Linoleic (18:2)
Linolenic (18:3)
Arachidic (20:0)
Gadoleic (20:1)
Behenic (22:0)
Lignoceric (24:0)

0.06
5.68
0.14
3.61
34.27
54.79
0.07
0.25
0.13
0.69
0.23

0.07
10.87
0.10
3.66
23.59
53.86
6.49
0.37
0.22
0.45
0.18

0.05
0.19
8.90
0.22
3.85
30.71
54.35
0.27
0.29
0.18
0.61
0.24

2. Methods
2.1. Materials
2.1.1. Chemicals
Styrene (STY), divinylbenzene (DVB), and glutaraldehyde (GA)
(25% w/v) were obtained from Merck. Sorbitan monooleate (Span
80) and p-nitrophenyl palmitate (p-NPP) were from Fluka. Potassium persulphate was purchased from Riedel-de-Haen. Vegetable
oils (sunower, soybean, and waste cooking oil) were obtained
locally. The composition of the oils with regard to fatty acid content was determined by gas chromatography (GC) in accordance
with The American Oil Chemical Society ofcial methods (AOCS,
1998) and the results were presented in Table 1.
2.1.2. Enzyme
Commercial soluble lipase from T. lanuginosus was purchased
from Novo Nordisk (Denmark). It is produced by submerged fermentation of a genetically modied Aspergillus oryzae microorganism. The enzyme activity was 16.3 U/ml. One unit of activity (U) is
dened as the amount of enzyme forming 1 lmol p-nitrophenol
per minute from p-nitrophenyl palmitate using lipase assay
method. Protein determinations were carried out using Bradford
method (Bradford, 1976).
2.2. Experimental procedure
All statistical analysis in polymer synthesis was performed for
only monolithic forms. Chosen factor levels and values of all variables in Taguchi design are shown in Table 2. Lipases immobilized

onto matrixes (bead, powder or monolithic form) were used for

biodiesel synthesis and the results obtained were compared.
2.2.1. Polyglutaraldehyde (PGA) synthesis
PGA was synthesized by polymerizing glutaraldehyde solution
(25% w/v), pH of which was 10.5 by adding NaOH solution (1 M),
at room temperature for 30 min and neutralizing the mixture using
HCl solution (1 M).
2.2.2. Preparation of MPPM in monolithic form and immobilization of
lipase using Taguchi methodology
MPPM was prepared by polymerizing the continuous phase of a
high internal phase emulsion consisting of organic and water
phases. The compositions of these phases are shown in Table 2.
The organic phase (5 ml) was composed of styrene, divinylbenzene, and Span 80. The water phase (45 ml) contained potassium
persulphate and polyglutaraldehyde solution. Span 80 (1.75 g)
and potassium persulphate (0.7 g) were used in all samples prepared. The aqueous phase was added dropwise to the organic
phase while stirring with a mechanical stirrer based on the conditions in Table 2. The emulsion was transferred to a mold and cured
at 80 C for 24 h (Dizge et al., 2008). The polymer was rst washed
with distilled water (50 ml) and then with tert-butanol (50 ml) to
remove impurities.
Immobilization was carried out by circulating enzyme solution
in calcium acetate buffer (55 ml, 25 mM, pH 6) by a peristaltic
pump through microporous monolithic polymer in disk form
(m = 0.36 g, D = 2 cm, and h = 0.5 cm) in a holder made of stainless
steel at various temperatures and immobilization time (Fig. 1a).
The immobilized enzyme was then washed with acetate buffer
(50 ml, 25 mM, pH 6) to remove the unbound enzyme. The supernatant and washing solutions were assayed for unbound protein.
SEM micrographs of microporous polymeric material (MPPM)
and microporous polymeric biocatalyst (MPPB) for monolithic
form are shown in Fig. 1c.
2.2.3. Preparation of MPPM in powder and bead forms and their usage
in lipase immobilization
Powder form in a range of 48 mesh in size was prepared from
the monolithic polymer produced using medium values (Table 2).
Beads were synthesized by modication of the work of Desforges
et al., in which the initiator, potassium persulphate, was used in
polyglutaraldehyde solution (Desforges et al., 2002).
Immobilization of lipase was carried out by reaction of powder
or bead matrix (1 g) with enzyme (1 ml) in calcium acetate buffer

Table 2
Factor levels and values of all variables in Taguchi designs applied during each of three consecutive experimentations.
Process

Factors

Parameter

Levels and values

Low (1)

Medium (0)

High (+1)

MPPM synthesis

X1
X2
X3
X4
X5

STY/DVB ratio (vol./vol.%)a

PGA concentration (vol.%)b
Monomers dozing time (min)
Mixing time (min)
Mixing rate (rpm)

0
1
5
20
300

2.5
8
13
40
450

5
15
21
60
600

MPPB synthesis

X6
X7
X8
X9

Lipase concentration (vol.%)

Immobilization temperature (C)
Immobilization time (h)
Enzyme ow (ml/min)

1
4
2
1

5.5
15
25
11

10
26
48
21

Biodiesel production

X10
X11
X12
X13
X14

Substrate ow (ml/min)
Reaction temperature (C)
Alcohol:oil ratio (mol/mol)
Water amount (%)
Total reaction time (h)

1
35
4:1
1
1

11
50
5:1
8
3

21
65
6:1
15
5

a
b

Organic phase (5 ml) contains Span 80 (0.75 ml) and a mixture of STY and DVB (4.25 ml).
Water phase (45 ml) contains PGA (0.45, 3.6, and 6.75 ml) and potassium persulphate (0.7 g).

N. Dizge et al. / Bioresource Technology 100 (2009) 19831991

1985

Fig. 1. Experimental set up ((a) for the immobilization of lipase and (b) for the enzymatic production of biodiesel (c) SEM micrographs for microporous polymeric matrix
(MPPM) and microporous polymeric biocatalyst (MPPB)).

(19 ml, 25 mM, pH 6) at 26 C for 25 h with gentle shaking

(250 rpm). The immobilized enzyme was then washed with acetate buffer (50 ml) to remove the unbound enzyme. The supernatant and washing solutions were assayed for unbound protein.
2.2.4. Biodiesel production by microporous polymeric biocatalyst
(MPPB) in monolithic form using Taguchi methodology
Biodiesel production was studied using MPPB in monolithic
form which was prepared using medium values (Table 2). The production was carried out using the system shown in Fig. 1b in which
the biocatalyst, MPPB, was placed in a holder. The vessel was initially lled with the vegetable oil (92 g), then placed on the constant-temperature mechanic stirrer (250 rpm) with its associated
equipment and heated to a predetermined temperature. Methanol
was added to the mixture in three-steps to avoid strong methanol
inhibition. Then, the mixture was recirculated through the holder
for duration indicated in Table 2. Glycerol was separated from biodiesel in a funnel within 2 h. The biodiesel phase was washed with
hot water twice (50 ml, 60 C) to remove methanol and glycerol
residues.
2.2.5. Biodiesel production by MPPB in powder and bead matrix
Immobilized lipase (1 g, powder or bead form) was reacted with
a mixture of oil (92 g) and methanol (20 g) in a complete-mixing
batch reactor (250 rpm) at 65 C for 24 h. Methanol was added to
the mixture in three-steps to avoid strong methanol inhibition.
After the reaction, MPPB was removed from biodiesel by ltration
and washed with tert-butanol (30 ml).
2.2.6. Determination of lipase activity
Lipase assay was performed using transesterication reaction of
p-nitrophenyl palmitate (p-NPP) and butanol, resulting in p-nitrophenol (p-NP), the amount of which was determined by a colorimetric method (Soares et al., 2003). The stock solution of p-NPP

(36 mM) was prepared in hexane. The reaction mixture comprised

of p-NPP solution (10 ml), water saturated hexane (9 ml) and butanol (1 ml). The reaction mixture was recirculated for 10 min at
25 C using a peristaltic pump with a rate of 21 ml/min. Reaction
mixture (50 ll) was added to 3 ml of triethylamin-ethanol solution
(3 ll triethylamine per ml ethanol). p-NP formed by the enzymatic
transesterication reaction was determined at 410 nm. Each of the
assays was performed in duplicate and mean values were
presented.
2.3. Analytical procedures
Tensile properties of the MPPM were determined using testing
machine Instron/5569. The stressstrain curves of the MPPM
were obtained following the standard ISO 2062 (1993).
The biodiesel content in the reaction mixture was analyzed
using a gas chromatograph (Agilent Technologies 6890N model)
equipped with a DB-WAX (length 30 m, inner diameter 0.53 mm,
lm thickness 0.5 lm) capillary column and a ame-ionization
detector (FID) connected to an integraph (DIN EN 14103, 2003).
Mono-, di-, tri-glycerides, free and total glycerol determinations
were carried out according to European standard test method of
DIN EN 14105 (2003). All of the analytical assays were performed
in duplicate and mean values were presented.
2.4. Experimental planning
This study was planned in a frame comprising three consecutive
stages: (a) MPPM synthesis having mechanical stability, (b) microporous polymeric biocatalyst (MPPB) preparation by immobilization of lipase onto MPPM, and (c) biodiesel production by MPPB.
Experiments were carried out in accordance with design of experiments (DoE) for more sophisticated analyses of variables with a
smaller number of experiments. In experimentations, three exper-

1986

N. Dizge et al. / Bioresource Technology 100 (2009) 19831991

Table 3
Experimental plan by L16 Taguchi design applied for MPPM and MPPB synthesis and the results.
Run

MPPM synthesis

MPPB synthesis

Factors

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19

Results

Factors

Results

Results

X1

X2

X3

X4

X5

Mechanical strength (MPa)

X6

X7

X8

X9

Immobilization efciency (%)

Enzyme activity (U/g disc)

1
1
1
1
1
1
1
1
+1
+1
+1
+1
+1
+1
+1
+1
0
0
0

1
1
1
1
+1
+1
+1
+1
1
1
1
1
+1
+1
+1
+1
0
0
0

1
1
+1
+1
1
1
+1
+1
1
1
+1
+1
1
1
+1
+1
0
0
0

1
+1
1
+1
1
+1
1
+1
1
+1
1
+1
1
+1
1
+1
0
0
0

1
1
+1
+1
+1
+1
1
1
+1
+1
1
1
1
1
+1
+1
0
0
0

3.98
9.27
5.65
5.81
3.34
5.56
4.89
0.61
4.35
7.87
2.87
5.32
5.67
3.93
4.07
2.57
3.13
3.27
3.25

1
+1
1
+1
+1
1
+1
1
+1
1
+1
1
1
+1
1
+1
0
0
0

1
+1
+1
1
+1
1
1
+1
1
+1
+1
1
+1
1
1
+1
0
0
0

+1
1
1
+1
1
+1
+1
1
1
+1
+1
1
+1
1
1
+1
0
0
0

+1
+1
+1
+1
1
1
1
1
1
1
1
1
+1
+1
+1
+1
0
0
0

30.32
16.81
16.65
1.96
27.29
1.89
6.65
2.13
3.37
56.09
53.14
31.74
86.88
18.40
15.08
88.84
20.83
21.11
21.07

4.88
5.17
4.34
5.14
2.89
3.25
2.89
0.09
1.97
5.12
5.26
6.29
3.59
2.98
2.07
2.58
12.83
12.86
12.87

imental structures were separately applied and all experiments

were carried out using sunower oil. Experimental plans were selected according to Taguchi 2n designs because of their simplicity
and easiness in application. 25 and 29 partial fractionalfactorial
designs were used for MPPM synthesis and biodiesel production
by MPPB, and MPPB synthesis, respectively. Standard Taguchi L16
and L8 orthogonal arrays requiring 16 and eight experiments
(Tables 3 and 4) were employed in experimentations of MPPM
and MPPB synthesis, and biodiesel production by MPPB, respectively (Ross, 1996). In MPPM experiments, ve variables (STY/
DVB ratio, PGA concentration, monomers dozing time, mixing
time, and mixing rate) were taken into account for the evaluation
of mechanical strength. The inuences of nine variables were studied in the synthesis of MPPB. The inuences of lipase concentration, immobilization temperature, immobilization time, and
enzyme ow on the effectiveness of MPPB synthesis were explored
together with ve variables taking place in previous stage. In addition, the inuences of alcohol to oil ratio, reaction temperature,
substrate ow, water amount, and total reaction time were also
investigated for biodiesel yields. Experimental error was determined by conducting three replicate experiments for the detailed
statistical analyses of the results. The reliability of each experimental plan was also conrmed by the validation tests in order to

Table 4
Experimental plan by L8 Taguchi design applied for biodiesel production by MPPB and
the results.
Run

Biodiesel production
Factors

1
2
3
4
5
6
7
8
9
10
11

Results

X10

X11

X12

X13

X14

Biodiesel yield (%)

1
1
1
1
+1
+1
+1
+1
0
0
0

1
1
+1
+1
+1
1
1
+1
0
0
0

1
+1
1
+1
1
+1
1
+1
0
0
0

+1
+1
1
1
+1
1
1
+1
0
0
0

+1
1
1
+1
1
1
+1
+1
0
0
0

37.3
43.1
36.7
57.0
51.5
50.1
44.6
63.8
40.0
39.7
39.5

determine sensibilities of the related model equations. Factor levels and values of all variables applied for each of experimentations
were presented in Table 2.
2.5. Statistical analysis
The results obtained from three consecutive stages of the study
were analyzed by analysis of variance (ANOVA) using MINITAB (PA,
USA) statistical software, which makes possible to interpret both
singly and interactively the inuences of input variables on the
process performance. Furthermore, the percentage contribution
of each input variable on each response can be comparatively
determined in addition to the rst order empirical relationships
between the variables and the responses (Chen et al., 2004; Cvijovic et al., 2005; Ross, 1996)

y b0

ip
X
i1

bi  xi

ip
X

bij  xi xj

i<j1

where y is the response parameter; i and p are the variable number

and the latest variable number; xi and xj are the input variables; b0,
bi, and bij are intercept, linear, and interaction parameters, respectively. The values of i are 1, 6, and 10, while those of p are 5, 9,
and 14 for MPPM synthesis, MPPB synthesis and biodiesel production, respectively.
3. Results and discussion
This study was planned in a frame comprising three consecutive
stages: (a) MPPM synthesis (monolithic, bead, and powder forms),
(b) microporous polymeric biocatalyst (MPPB) preparation by
immobilization of lipase onto MPPM, and (c) biodiesel production
by MPPB. Lipase from T. lanuginosus was immobilized successfully
onto MPPM (monolithic, bead, and powder forms) with high
immobilization efciencies, dened as the ratio of activity of
immobilized enzyme to the activity of soluble enzyme used in
immobilization. The immobilization efciencies obtained using
bead and powder forms were 80% and 85%, respectively.
The main aim of this study was to produce MPPB having high
mechanical strength and maximum enzyme activity. The mechanical strength of MPPB changed in the range 0.619.27 MPa (Table
3). The immobilization efciencies for MPPBs were between 2%

1987

N. Dizge et al. / Bioresource Technology 100 (2009) 19831991

and 89%. Higher mechanical strength is crucial with respect to lifetime of MPPB and enzyme leakages. However, highest immobilization efciency and activity per g disk were not obtained using the
monolith with highest mechanical strength (9.27 MPa) (Table 3).
As known, enzyme activity is a major parameter for reaction rate.
The disk having highest activity (12.87 U/g disk) had a mechanical
strength of 3.27 MPa and this value is an acceptable one. No crushing was observed when the MPPB with 3.27 MPa was used in repeated batch reactions. Thus, we used MPPM with moderate
mechanical strength (3.27 MPa) with highest activity per gram
disk (12.87 U/g disk) in biodiesel production. The MPPM used
was prepared under medium conditions (Table 2). The immobilization efciency of the disk used in the study was 21%, indicating
that all of the enzyme is not used under the conditions studied.
MPPB was used for biodiesel production and the parameters (substrate ow, reaction temperature, alcohol/oil ratio, water amount,
and total reaction time) were evaluated for maximum biodiesel
yield.

mechanical strength. However, PGA concentration and dozing time

had a negative effect on mechanical strength. The two-way interactions affected the process highly. The most inuential ones were
PGA concentrationmixing time (X2X4) and dozing timemixing
time (X3X4). The strength was also affected by three-way interactions among STY/DVB ratio, dozing time, and mixing time variables
(X1X3X4). It is well known that during the preparation of high
internal phase emulsions for the production of polyHIPE (PHP),
the viscosity of emulsion increases as the mixing time and rate increases. It is seen from previous studies that dropletdroplet interactions may be due to reduced size of droplets through the
increasing mixing time (Bhumgara, 1995). These results showed
that a polymeric matrix having more mechanical strength could
be obtained by increasing the mixing time and mixing rate which
led to smaller porous matrix structure. The increasing PGA concentration may hinder cross-linking of DVB monomers with STY ones
which resulted in formation of a less strong structure.
3.2. Lipase immobilization onto MPPM

3.1. Synthesis of MPPM

Lipase immobilization onto MPPM was carried out successfully.
MPPM contains polyglutaraldehyde functional groups which react
with amino, thiol, imidazole, and phenol groups of proteins (Bolgar
and Gaskell, 1996; Kaga et al., 1994; Weber et al., 1995). Taguchi
methodology was used in nding of immobilization conditions.
Main effect plots of single variables and relative inuences of all
variables are depicted in Fig. 3.
Immobilization efciency was affected by STY/DVB ratio (X1),
immobilization temperature (X7), immobilization time (X8), and
enzyme ow (X9). Immobilization time (X8) was the most impor-

Mechanical Test (MPa )

The inuences of single and interactive variables on mechanical

strength of MPPM were investigated and the results of DoEs calculations are presented in Table 3. Main effect plots of single variables and relative inuences of all variables are shown in Fig. 2.
Determinative factors for mechanical strength of MPPM were
established as single and two-way interactive variables. Various
single variables (PGA concentration (X2), dozing time (X3), and
mixing time (X4)) noticeably inuenced the mechanical strength
of MPPM (Fig. 2a). An increase in mixing time increased the

9.5
8.0

6.5
5.0
3.5
2.0
0

5
X1
(%vol/vol)

15

X2
(%vol)

21

20

X3
(min)

60

300

X4
(min)

600
X5
(rpm)

30

b
Relative Influence (%)

25

20

15

10

Lack of fit

Exp. error

X1X4X5

X1X3X4

X1X2X4

X4X5

X3X4

X2X4

X1X5

X1X4

X1X3

X1X2

X5

X4

X3

X2

X1

Fig. 2. Inuences of variables on MPPM synthesis ((a) main effect plots and (b) relative inuences).

1988

N. Dizge et al. / Bioresource Technology 100 (2009) 19831991

Activity (U/g disc )

Efficiency (% )

75
60

45
30
15
0
5.25

4.50
3.75
3.00
2.25
1.50

15

21 20

60 300

10 4

600 1

26 2

48 1

21

X1

X2

X3

X4

X5

X6

X7

X8

X9

(%vol/vol)

(%vol)

(min)

(min)

(rpm)

(%vol)

( C)

(h)

(mL/min)

55
50

Efficiency
Activity

Relative Influence (%)

45
40
35
30
25
20
15
10
5

Lack of fit

Exp. error

X2X8

X2X7

X1X6

X1X5

X1X4

X1X3

X9

X8

X7

X6

X5

X4

X3

X2

X1

Fig. 3. Inuences of variables on lipase immobilization onto MPPM ((a) main effect plots and (b) relative inuences).

tant factor. However, enzyme activity was affected negatively by

high PGA concentrations (X2) (Fig. 3). The relative inuences were
observed in decreasing order as X1 > X7 > X8 > X9 for efciency.
The immobilization efciency increases as STY/DVB ratio increases.
It is possibly due to the formation of large pores formation. Twoway interactions affected the immobilization efciency more than
single variables. The most inuential two-way interactions were
PGA concentrationimmobilization time (X2X8) and STY/DVB ratiolipase concentration (X1X6).
Activity of MPPB was affected by single variables (immobilization time (X8) and enzyme ow (X9)) and two-way interactive
variables (STY/DVB ratiomixing rate (X1X5), STY/DVB ratiolipase
concentration (X1X6), and PGA concentrationimmobilization
time (X2X8)) (Fig. 3). The most important variable was the immobilization time.
3.3. Biodiesel production by MPPB
The inuences of single and interactive variables on biodiesel
production by MPPB were studied. The results of DoEs are presented in Table 4. Main effect plots of single variables and relative
inuences of all variables are shown in Fig. 4.

Biodiesel yield was increased signicantly by such single variables as alcohol:oil ratio (X12), substrate ow (X10), reaction temperature (X11), and total reaction time (X14) (Fig. 4). However,
increased amount of water (X13) had insignicant effect on the
biodiesel yield. The important two-way interactions affecting biodiesel yield were substrate owalcohol:oil (X10X12) and substrate owtotal reaction time (X10X14). Different methanol to
oil ratios (4:16:1) were studied. It was found that an increase in
alcohol to oil ratio had a positive effect on biodiesel yield. The
6:1 ratio gave rise to the best result, in agreement with previous
researches (Krisnangkura and Simamaharnoop, 1992).
The inuence of reaction temperature (X11) on biodiesel yield
was very signicant in the range studied (from 35 to 65 C). There
was no decrease in activity at 65 C due to thermophilic characteristic of T. lanuginosus.
Previous study showed that biodiesel conversion depends on
the rate of substrate ow (Nie et al., 2006). A similar result was obtained in our study. Flow rate of substrate and reaction time had
positive effects on biodiesel yield. This shows that the reaction
was less diffusion-controlled.
It is well known that water content is an important factor
affecting the activity of an enzyme in a non-aqueous medium.

1989

Methyl Esters (% )

N. Dizge et al. / Bioresource Technology 100 (2009) 19831991

60
55

50
45
40
35
1

21

35

X10
(mL/min)

Relative Influence (%)

50

65

X11

X12
(M/M)

( C)

15

X13
(%)

X14
(h)

40

30

20

10

Lack of fit

Exp. error

X10X14

X10X12

X14

X13

X12

X11

X10

Fig. 4. Inuences of variables on biodiesel production by MPPB ((a) main effect plots and (b) relative inuences).

Table 5
Regression models estimated for MPPM and MPPB synthesis and biodiesel production.
Terms

Coefcients
MPPM synthesis (mechanical strength)

constant
X1
X2
X3
X4
X5
X6
X7
X8
X9
X10
X11
X12
X13
X14
X1X2
X1X3
X1X4
X1X5
X1X6
X2X4
X2X7
X2X8
X3X4
X4X5
X10X12
X10X14
X1X2X4
X1X3X4
X1X4X5

4.750
0.164
0.894
0.772
0.393
0.178









0.373
0.102
0.052
0.044

1.030


0.789
0.178


0.117
0.686
0.014

MPPB synthesis

Biodiesel production (methyl esters yield)

Efciency

Activity

28.600
15.600
2.320
1.550
1.350
2.180
1.520
14.900
12.100
5.790






4.560
5.920
1.170
1.730

5.490
3.030








3.660
0.076
1.110
0.074
0.171
0.237
0.047
0.027
0.432
0.187






0.392
0.339
0.561
0.488

0.228
0.103








47.700









4.180
3.930
5.170
0.600
2.350










1.350
1.280




These terms do not take part in relevant model equations that determined for each of three consecutive experimentations, separately.

1990

N. Dizge et al. / Bioresource Technology 100 (2009) 19831991

Table 6
Levels of variables for validation of each DoE set and the estimated model results together with experimental ones.
Variables

MPPM synthesis
(Mechanical strength, MPa)
Levels

Results
Exp.

X1
X2
X3
X4
X5
X6
X7
X8
X9
X10
X11
X12
X13
X14

+1
1
1
+1
1

MPPB synthesis

7.12

Model

7.27

(Immobilization efciency, %)

(Lipase activity, U/g disk)

Levels

Levels

Absolute
Error (%)

2.03

Results
Exp.

+1
1
1
+1
1
1
+1
+1
+1

86.19

Model

80.33

Absolute
Error (%)
+1
1
1
+1
1
1
+1
+1
+1

6.80

Biodiesel production
(Methyl esters yield, %)

Results

Levels

Exp.

Model

Absolute
Error (%)

7.21

7.72

6.60

+1
+1
+1
1
+1

Results
Exp.

Model

Absolute
Error (%)

59.83

60.10

0.45

Each experiment was repeated in triplicate and the results were given as arithmetical averages of the measurement values belonging to each process.

Some lipases possess a unique activating feature at the interface

between an aqueous and an organic phase (Lu et al., 2007). Hence,
the inuence of water content on biodiesel conversion was studied
and it was found that the effect of water on biodiesel yield was
insignicant as compared to other variables.

Regression models were obtained separately for each response

from DoE results on each of the three consecutive stages (Table 5).
It was observed that there is a respectable correlation between
experimental and model results for each of three experimental
steps. Correlations of 0.9995, 0.9999, and 0.9946 were obtained
for MPPM synthesis, MPPB synthesis, and biodiesel production,
respectively. Root mean squared error (RMSE) values were
0.011 MPa, 0.013%, 0.001 U/g disk, and 0.16% for mechanical
strength, immobilization efciency, immobilized lipase activity,
and biodiesel production, respectively.
In order to validate DoE sets, an experiment for each DoE set
was conducted. The accuracy of each model was validated with
triplicate experiments under the maximized reaction conditions
calculated using the model equations obtained. The results were
given in Table 6.
As seen, model results were in good agreements with the experimental ones, especially for biodiesel production with an absolute
error of 0.45%. As a result, the model was accurate and reliable for
predicting the process yields.
3.5. Effects of vegetable oils on enzymatic biodiesel synthesis
It has been reported that all the triglyceride sources should be
considered as lipase substrates. Various oils (sunower, soybean,
restaurant grease, waste vegetable oils, and animal fats) have been
commonly used in enzymatic biodiesel synthesis. However, such
substances as phospholipids and waxes in oils may negatively affect enzyme activity (Salis et al., 2007).
Three different oils (sunower, soybean, and waste cooking oil)
were used for biodiesel synthesis by MPPB in order to compare the
effectiveness of the transesterication reaction. All reactions were
performed for 5 h at 65 C by using a stoichiometric excess amount
of methanol (Fig. 5). The highest yield was obtained from sunower oil (63.8%). Biodiesel yields from soybean and waste cooking
oil were 55.5% and 50.9%, respectively.

Sunflower oil
Soybean oil
Waste cooking oil

60

Biodiesel yield (%)

3.4. Validation of all DoE experiments

75

45

30

15

0
0

Time (h)
Fig. 5. Variation of biodiesel production efciency with time for sunower,
soybean, and waste cooking oil (Operation conditions were: substrate ow 21 ml/
min, reaction temperature 65 C, alcohol:oil ratio 6:1, water 15%, total reaction time
5 h).

3.6. Effects of MPPB congurations on enzymatic biodiesel synthesis

The inuences of MPPB congurations on biodiesel synthesis
from sunower oil were studied using MPPB in different forms
(monolithic, powdered, and bead) in 5 h and 24 h reactions. In 5 h
reaction, biodiesel yields were 63.8%, 81.1%, and 86.9% using monolithic, bead, and powdered MPPB, respectively. In 24 h reaction, biodiesel yields were 90.2%, 93.9%, and 97.0% using monolithic, bead,
and powdered MPPB, respectively. As can be seen, the most effective
biocatalyst was powdered MPPB for the production of biodiesel. This
could be due to efcient mixing of reactants during reaction.
4. Conclusion
This study deals with the production of biodiesel by enzymatic
transesterication using microporous polymeric biocatalyst
(MPPB). In the study, Taguchi methodology was used for MPPM
synthesis in monolithic form, lipase immobilization onto monolithic MPPM and nally biodiesel production using monolithic

N. Dizge et al. / Bioresource Technology 100 (2009) 19831991

MPPB. In addition, MPPM in powder and bead forms were obtained

and used in biodiesel production. A microporous polymeric matrix
(MPPM) was synthesized using styrene, divinylbenzene, and
polyglutaraldehyde. SEM micrographs and FTIR spectrum showed
that copolymer can be produced as a porous structure having aldehyde functional groups. The polymeric matrix could be prepared in
a short time in any desirable shapes and is a favorable immobilization support for lipase from T. lanuginosus. The enzyme was covalently attached onto MPPM with 80%, 85%, and 89% immobilization
efciencies using bead, powder, and monolithic forms, respectively. It was shown that all forms of MPPB (monolithic, bead,
and powder) were used successfully for the production of biodiesel
using sunower, soybean, and waste cooking oil. The best biodiesel
yields were obtained from sunower oil (97%). The biodiesel yield
from waste cooking oil was lower (90.2%). This could be due to
contaminants formed in waste cooking oil affecting the enzyme.
Bead or powdered MPPB catalyzed transesterication reactions resulted in higher biodiesel yields due to efcient mixing of reactants
during reaction. This study shows that lipase immobilization onto
newly synthesized microporous support is rather effective process
in enzymatic transesterication for biodiesel production. It was
shown that immobilized lipase retained its activity during 10 repeated batch reactions at 25 C, each lasting 24 h. Since the developed novel method is simple yet effective, it could have a potential
to be used industrially for the production of chemicals (fatty acid
production, removal of fatty acids from oil and production of modied oils) requiring immobilized lipases.
Acknowledgement
This study was supported by the TUBITAK, the Scientic and
Technological Research Council of Turkey (Project No. MAG-261).
References
AOCS, 1998. Methods Cd 125, Ce 5b89, Ce 162, Cc 325, Ofcial Methods and
Recommended Practices of the American Oil Chemists Society, fth ed. AOCS,
Champaign, IL.
Arica, M.Y., Yavuz, H., Patir, S., Denizli, A., 2000. Immobilization of glucoamylase
onto spacer-arm attached magnetic poly-methylmethacrylate/microspheres:
characterization and application to a continuous ow reactor. J. Mol. Catal. B:
Enzy. 11, 127138.
Bhumgara, Z., 1995. A Study of the Development of PHP Foam Materials for Use in
Separation Processes. Ph.D. Thesis. University of Exeter.
Bolgar, M.S., Gaskell, S.J., 1996. Tandem mass spectrometric characterization of
modied peptides and proteins. Anal. Chem. 68, 23252330.
Bradford, M.M., 1976. A rapid and sensitive method for quantitation of microgram
quantities of protein utilizing the principle of proteindye binding. Anal.
Biochem. 72, 242254.

1991

Chen, X., Zhang, Y., Pickrell, G., Antony, J., 2004. Experimental design in ber optic
sensor development. Int. J. Prod. Perf. Manag. 53 (8), 713725.
Cvijovic, Z., Radenkovic, G., Maksimovic, V., Dimcic, B., 2005. Application of ANOVA
method to precipitation behaviour studies. Mater. Sci. Eng. A Struct. 397, 195
203.
Desforges, A., Arpontet, M., Deleuze, H., Mondain-Monval, O., 2002. Synthesis
and functionalisation of polyHIPE beads. Reac. Funct. Polym. 53, 183
192.
DIN EN 14103, 20032010. Fat and Oil Derivates Fatty Acid Methyl Esters (FAME)
Determination of Ester and Linolenic Acid Methyl Ester Contents.
DIN EN 14105, 20032010. Fat and Oil Derivates Fatty Acid Methyl Esters (FAME)
Determination of Free Glycerol and Total Glycerol and mono-, di-, triGlyceride Content.
Dizge, N., Keskinler, B., Tanriseven, A., 2008. Covalent attachment of microbial
lipase onto microporous styrenedivinylbenzene copolymer by means of
polyglutaraldehyde. Colloids Surf. B: Biointerfaces 66, 3438.
Gomes, F.M., Pereira, E.B., Castro, H.F., 2004. Immobilization of lipase on
chitin and ts use in nonconventional biocatalysis. Biomacromolecules 5,
1723.
ISO 2062, 1993. Textiles Yarns from Packages Determination of Single-end
Breaking Force and Elongation at Break Using Constant Rate of Extension (CRE)
Tester.
Kaga, H., Siegmund, B., Neufellner, E., Faber, K., 1994. Stabilization of Candida
lipase against acetaldehyde by adsorption onto celite. Biotechnol. Tech. 8,
369374.
Krisnangkura, K., Simamaharnoop, R., 1992. J. Am. Oil Chem. Soc. 69, 166175.
Lu, J., Nie, K., Xie, F., Wang, F., Tan, T., 2007. Enzymatic synthesis of fatty acid methyl
esters from lard with immobilized Candida sp. 99125. Process Biochem. 42,
13621370.
Mittelbach, M., 1990. Lipase catalyzed alcoholysis of sunower oil. J. Am. Oil Chem.
Soc. 67 (3), 168170.
Nie, K., Xie, F., Wang, F., Tan, T., 2006. Lipase catalyzed methanolysis to produce
biodiesel: optimization of the biodiesel production. J. Mol. Catal. B: Enzym. 43,
142147.
Noureddini, H., Gao, X., Philkana, R.S., 2005. Immobilized Pseudomonas cepacia
lipase for biodiesel fuel production from soybean oil. Bioresource Technol. 96,
769777.
Oraire, O., Buisson, P., Pierre, A.C., 2006. Application of silica aerogel encapsulated
lipases in the synthesis of biodiesel by transesterication reactions. J. Mol. Catal.
B: Enzym. 42 (34), 106113.
Ross, P.J., 1996. Taguchi Techniques for Quality Engineering. McGraw-Hill, New
York, pp. 1117/102142.
Salis, A., Monduzzi, M., Solinas, V., Polaina, J., MacCabe, A.P. (Eds.), 2007. Industrial
Enzymes: Structure Function and Applications. Springer. p. 315.
Soares, A.M., Costa, L.R., Frana, S.C., Trevisan, H.C., Roberts, T.J.C., 2003.
Immobilization of lipases and assay in continuous xed bed reactor. Protein
Peptide Lett. 10, 619628.
Vicente, G., Coteron, A., Martnez, M., Aracil, J., 1998. Application of the factorial
design of experiments and response surface methodology to optimize biodiesel
production. Ind. Crops Prod. 8, 2935.
Vicente, G., Martinez, M., Aracil, J., 2007. Optimisation of integrated biodiesel
production Part II: A study of the material balance. Bioresource Technol. 98,
17541761.
Weber, H.K., Srecher, H., Faber, K., 1995. Sensitivity of microbial lipases to
acetaldehyde formed by acyl-transfer reactions from vinyl esters. Biotechnol.
Lett. 17, 803808.
Yang, G., Tian-Wei, T., Kai-Li, N., Fang, W., 2006. Immobilization of lipase on
macroporous resin and its application in synthesis of biodiesel in low aqueous
media. Chinese J. Biotechnol. 22 (1), 114118.